provided for non-commercial research and education use. not for … · experiment 2 all the pigs...

The Canadian Journal of Veterinary Research is published by Canadian Veterinary Medical Association. The attached copy is

furnished to readers for personal, non-commercial research and education use. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party Web sites are prohibited. Those who require further

information regarding reprints or archiving and manuscript policies are encouraged to contact [email protected].

Provided for non-commercial research and education use. Not for reproduction, distribution or commercial use.

Canadian Journal of Veterinary Research Revue Canadienne de Recherche Vétérinaire

Canadian Journal of Veterinary Research Revue Canadienne de Recherche Vétérinaire

ArticlesSnatch-farrowed, porcine-colostrum-deprived (SF-pCD) pigs as a model for swine infectious disease research Yanyun Huang, Deborah M. Haines, John C.S. Harding. . . . . . . . . . . . . . . . . . . . . . . .81

Prediction of serum IgG concentration by indirect techniques with adjustment for age and clinical and laboratory covariates in critically ill newborn calvesGilles Fecteau, Julie Arsenault, Julie Paré, David C. Van Metre, Charles A. Holmberg, Bradford P. Smith . . . . . . . . . . . . . . . . . . . . . . . 89

A method to quantify infectious airborne pathogens at concentrations below the threshold of quantification by cultureTimothy D. Cutler, Chong Wang, Steven J. Hoff, Jeffrey J. Zimmerman . . . . . . . 95

Influence of temperature and organic load on chemical disinfection of Geobacillus steareothermophilus spores, a surrogate for Bacillus anthracisJiewen Guan, Maria Chan, Brian W. Brooks, Liz Rohonczy . . . . . . . . . . . . . . . . . . . . . . . . . 100

Effect of a straw-derived xylooligosaccharide on broiler growth performance, endocrine metabolism, and immune responseSun Zhenping, Lv Wenting, Yu Ruikui, Li Jia, Liu Honghong, Sun Wei, Wang Zhongmie, Li Jingpan, Shan Zhe, Qin Yuling. . . . . . . . . . 105

The rabbit as an infection model for equine proliferative enteropathy Francesca Sampieri, Andrew L. Allen, Nicola Pusterla, Fabio A. Vannucci, Aphroditi J. Antonopoulos, Katherine R. Ball, Julie Thompson, Patricia M. Dowling, Don L. Hamilton, Connie J. Gebhart . . . . . . . .110

Effects of vitamin E supplementation on cellular a-tocopherol concentrations of neutrophils in Holstein calvesHidetoshi Higuchi, Erina Ito, Hidetoma Iwano, Shin Oikawa, Hajime Nagahata. . . . . . . . . . . 120

Preliminary study on the factors influencing rabbit doe reproductive efficiency: Effect of parity, day of mating, and suckling on ovarian status and estrogen levels at day 6 of pregnancyMaria Laura Marongiu, Corrado Dimauro . . .126

A p r i l / Av r i l 2 0 1 3 , V o l . 7 7 N o . 2table of contents continued on back cover/la table des matières se poursuit sur la couverture arrière

Subscriptions/Abonnements — 2013Canada $150 $ ( applicable GST/HST;

TPS/TVH en vigueur) Foreign/Étranger $165 $ US/É.-U.

Published quarterly by the Canadian Veterinary Medical AssociationPublication trimestrielle de l’Association canadienne des médecins vétérinaires

C a n a d i a n J o u r n a l o fVe t e r i n a r y R e s e a r c h

Formerly CANADIAN JOURNAL OF COMPARATIVE MEDICINE

The journal publishes the results of original research in veterinary and comparative medicine. Manuscripts must be as concise as pos-sible, and the research described must represent a significant con-tribution to knowledge in veterinary medicine. Full-length papers, short communications, and review papers are welcome. All manu-scripts will be reviewed for scientific content and editorial accuracy.

Manuscripts must conform to the Instructions for Authors found on the CVMA Web site. Consultation of a recent issue of the journal to ensure that the manuscript conforms to current style is also recom-mended. Please use the Web site (www.canadianveterinarians.net) to submit articles. A publication charge of $90 per published page, plus a cost of $10 per table, $15 per illustration, and $150 per color illustration is payable by contributors. Instructions for ordering reprints are sent with the galley proofs. Color illustrations will only be published if the author(s) agree to pay the extra cost.

R e v u e c a n a d i e n n e d er e c h e r c h e v é t é r i n a i r e

Remplace la REVUE CANADIENNE DE MÉDECINE COMPARÉE

La revue publie les résultats de recherches originales en médecine vétérinaire et comparative. Les manuscrits doivent être aussi brefs que possible et la recherche décrite doit apporter une contribution impor-tante à la médecine vétérinaire. Les exposés complets, les communi-cations brèves et les comptes rendus sont bienvenus. Nous véri fierons le fonds scientifique et la forme de tous les manuscrits.

Les manuscrits doivent être conformes aux directives aux auteurs qui se trouvent sur le site Web de l’ACMV. La consultation d’un récent numéro de la revue est aussi profitable, afin de s’assurer qu’un manuscrit se conforme au style courant. Veuillez utiliser le site Web (www.veterinairesaucanada.net) pour soumettre les articles. Les auteurs doivent acquitter des frais de 90 $ par page publiée, 10 $ pour chaque tableau, 15 $ pour chaque illustration, et 150 $ pour chaque illustration en couleur. Un formulaire de commande de tirés à part accompagne l’en voi des épreuves d’un manuscrit. Les illustrations en couleurs ne seront publiées que si les auteurs en assument les frais supplémentaires.

Established in 1937 ISSN 0830-9000

Address all correspondence to/Adresser toute correspondance à :Canadian Journal of Veterinary Research Revue canadienne de recherche vétérinaire339, rue Booth Street, Ottawa, Ontario K1R 7K1Tel./Tél. : (613) 236-1162 — Fax/Télécopieur : (613) 236-9681e-mail/Courriel : [email protected]© Canadian Veterinary Medical Association 2013.

L’Association canadienne des médecins vétérinaires 2013

Canadian Journal of Veterinary ResearchRevue canadienne de recherche vétérinaireEditor — Rédacteur Éva Nagy, Guelph, OntarioAssociate Editor — Rédacteur adjoint Deborah M. Haines, Saskatoon, SaskatchewanAssistant Editors — Assistants à la rédaction Serge Messier, Saint-Hyacinthe (Québec) Jeffrey J. Wichtel, Charlottetown, Prince Edward IslandManaging Editor — Directrice de la rédaction Heather Broughton, Ottawa, OntarioAssistant Managing Editor — Directrice adjointe à la rédaction Stella Wheatley, Ottawa, OntarioAdvertising Manager — Gérante de la publicité Laima Laffitte, Wendover, OntarioEditorial Coordinator/Coordonnatrice de la rédaction Beverley Kelly, Ottawa, Ontario

Abstracted and/or indexed in:/Résumé et/ou catalogué dans :Biological Abstracts Index VeterinariusChemical Abstracts Nutrition AbstractsCurrent Contents (Agric. Section) Science Citation IndexEMBASE/Excerpta Medica Veterinary BulletinIndex Medicus Derwent Veterinary Drug File

Return undeliverable Canadian addresses to: 339 rue Booth Street, Ottawa, Ontario K1R 7K1, e-mail: [email protected]

Typesetting/Typographie AN Design Communications www.an-design.ca

FOR PERSONAL USE ONLY

Article

2013;77:81–88 The Canadian Journal of Veterinary Research 81

I n t r o d u c t i o nIn porcine research, especially that investigating infectious dis-

eases, obtaining pigs that are free of porcine pathogens is essential. Currently, 3 main methods are used to obtain such pigs: testing of conventional pigs for antigen and antibodies of certain pathogens,

the pigs testing negative being labeled specific pathogen free (SPF); the cesarean-derived colostrum-deprived (CDCD) method; and the gnotobiotic or germ-free technique. The advantage of the SPF method is its convenience, low technical requirement, and cost effi-ciency. However, when the research requires freedom of infection with pathogens that are highly prevalent in pig populations, such as

Snatch-farrowed, porcine-colostrum-deprived (SF-pCD) pigs as a model for swine infectious disease research

Yanyun Huang, Deborah M. Haines, John C.S. Harding

A b s t r a c tThe current study tested the benefit of commercially available spray-dried bovine colostrum (The Saskatoon Colostrum Company, Saskatoon, Saskatchewan) in raising snatch-farrowed, porcine-colostrum-deprived (SF-pCD) pigs. In experiment 1, 12 SF-pCD pigs received a liquid diet composed mainly of bovine colostrum from birth to day 10; 6 remained on the same liquid diet (COL), and the other 6 were fed a diet composed mainly of milk replacer (RPL) until weaning. In experiment 2, 12 SF-pCD pigs were fed mainly bovine colostrum before weaning; after weaning, 6 were fed a starter diet containing 20% (w/w) bovine colostrum powder (STARTER-COL), and the other 6 were fed a starter diet without any bovine colostrum (STARTER-CTRL) until termination (day 42 or day 49). In experiment 1 the COL pigs had significantly fewer fever-days than did the RPL pigs. In experiment 2 diarrhea, typhlocolitis, and pancreatic degeneration developed in 4 of the STARTER-COL pigs after weaning. In both experiments all the pigs fed mainly bovine colostrum before weaning survived until termination. All pigs tested free of swine influenza virus H1N1 and H3N2, Porcine reproductive and respiratory syndrome virus, and Porcine parvovirus. In experiment 2 all the pigs tested free of Porcine circovirus type 2 (PCV2), but some in both groups tested positive for Torque teno virus genogroups 1 and 2. In conclusion, with the use of snatch-farrowing and bovine colostrum, pigs can be raised in the absence of porcine maternal antibodies with 100% survival and freedom from most porcine pathogens of biologic relevance. This model is potentially suitable for animal disease research.

R é s u m éLa présente étude visait à tester l’avantage du colostrum bovin déshydraté disponible commercialement pour élever des porcs captés à la mise-bas et privés de colostrum porcin (SF-pCD). Dans l’expérience 1, 12 porcs SF-pCD ont reçu une diète liquide composée principalement de colostrum bovin de la naissance au jour 10; 6 sont demeurés sur la même diète liquide (COL), et les 6 autres étaient nourris avec une diète composée principalement de substitut de lait (RPL) jusqu’au sevrage. Dans l’expérience 2, 12 porcs SF-pCD étaient nourris principalement avec du colostrum bovin avant le sevrage; après le sevrage, 6 étaient nourris avec une diète de début contenant 20 % (poids/poids) de poudre de colostrum bovin (STARTER-COL), et les 6 autres étaient nourris avec une diète de début mais sans le colostrum bovin (STARTER-CTRL) jusqu’à la fin de l’expérience (jour 42 ou jour 49). Dans l’expérience 1, les porcs COL avaient significativement moins de jours avec fièvre que les porcs RPL. Dans l’expérience 2, de la diarrhée, une typhlocolite et une dégénération du pancréas s’est développée chez 4 des porcs STARTER-COL après le sevrage. Dans les 2 expériences tous les porcs nourris principalement avec du colostrum bovin avant le sevrage ont survécu jusqu’à la fin de l’expérimentation. Tous les porcs se sont avérés négatifs pour les virus H1N1 et H3N2 de l’influenza porcin, le virus du syndrome reproducteur et respiratoire porcin, et le parvovirus porcin. Dans l’expérience 2, tous les porcs ont testé négatif pour le circovirus porcin de type 2 (PCV2), mais quelques-uns dans les 2 groupes ont testé positif pour le virus Torque teno des génogroupes 1 et 2. En conclusion, avec l’utilisation de la mise-bas avec captation et de colostrum bovin, les porcs peuvent être élevés en absence d’anticorps maternels porcins avec un taux de survie de 100 % et l’absence des principaux agents pathogènes porcins d’importance biologique. Ce modèle est potentiellement approprié pour la recherche sur les maladies animales.

(Traduit par Docteur Serge Messier)

Department of Large Animal Clinical Sciences, Western College of Veterinary Medicine, University of Saskatchewan, 52 Campus Drive, Saskatoon, Saskatchewan S7N 5B4 (Huang and Harding); Department of Veterinary Microbiology, Western College of Veterinary Medicine, University of Saskatchewan, 52 Campus Drive, Saskatoon, Saskatchewan S7N 5B4, and The Saskatoon Colostrum Company, 30 Molaro Place, Saskatoon, Saskatchewan S7K 6A2 (Haines).

Address all correspondence to Dr. Yanyun Huang; telephone: 306-966-6480; e-mail: [email protected]

Received December 22, 2011. Accepted March 22, 2012.


82 The Canadian Journal of Veterinary Research 2000;64:0–00

Porcine circovirus type 2 (PCV2), this method may be inadequate, as most pigs have antibodies against these pathogens, either maternal or acquired, or are actively infected with the pathogen of interest. As a result, researchers may have to screen a large number of farms and pigs to obtain a reliable pig source and then select pigs after the level of maternally derived antibodies has waned.

The CDCD and gnotobiotic methods use cesarean section to obtain term piglets from pregnant sows. The CDCD pigs are raised in sterile compartments for several days and then in a clean room (1). Gnotobiotic pigs are raised entirely in sterile compartments. Although the CDCD and gnotobiotic methods are reliable for obtain-ing pathogen-free pigs, they have several disadvantages compared with the SPF method, including the need for surgery, specialized facilities, and sterile compartments, and the greater cost. In addition, gnotobiotic experiments are limited to only a few weeks, after which the pigs outgrow the sterile compartments. There is also a risk that cesarean-derived pigs, especially if delivered before term, do not fully experience the prenatal serum cortisol surge that occurs in vagi-nally delivered pigs (2). This surge plays an important role in tissue maturation, immunoglobulin absorption, surfactant production, and deposition of glycogen in muscle and liver (3). This helps to explain why morbidity and mortality rates are higher in cesarean-derived piglets than in naturally born cohorts. Further, there is debate as to whether gnotobiotic pigs have the same immunologic responses as naturally born pigs because they are not exposed to microbial programming, as are piglets reared in natural environments (4).

Recent attempts to raise snatch-farrowed, porcine-colostrum-deprived (SF-pCD) pigs with bovine colostrum have provided an alternative method for obtaining susceptible pigs for infectious dis-ease experiments (5,6). In this method, pigs are fed bovine colostrum

for 3 d and a porridge of milk replacer and dry feed from day 4 to 14 and then are weaned onto a dry diet on day 15. The advantages of this method are that the pigs experience natural vaginal delivery and bacterial colonization of their intestines and thus are more representative of conventional pigs than are CDCD or gnotobiotic pigs. However, the survival rates in these studies were at most 80%, mostly because of deaths from septicemia due to Escherichia coli or staphylococci (5,6). A further disadvantage was that the pigs were weaned much earlier than in the commercial industry and at an age when there may not be enough digestive enzymes to efficiently cope with solid feed.

In the present study a method of raising SF-pCD pigs is described that used commercially available bovine colostrum prod-ucts (HeadSTART and Calf’s Choice Total HiCal; The Saskatoon Colostrum Company, Saskatoon, Saskatchewan). With this method the pigs were weaned at 21 or 24 d of age, all survived, and they were free of major porcine pathogens and maternal antibodies, such that they were fully susceptible to challenge with pathogens and receptive to vaccination at an early age. This method is thus an alternative to cesarean-derived methods of producing porcine-colostrum-deprived pigs for infectious disease research.

M a t e r i a l s a n d m e t h o d s

Snatch-farrowing, animal care, and experimental design

This work was approved by the University of Saskatchewan’s Animal Research Ethics Board and adhered to the Canadian Council on Animal Care guidelines for humane animal use (7).

Table I. Ingredients in the liquid diets fed until weaning to snatch-farrowed, porcine-colostrum-deprived pigs in experiments 1 and 2

Experiment 1 Days 1 to 3 Days 4 to 10 Days 11 to 15 Days 16 to 23 Experiment 2Ingredienta RPL COL RPL COL RPL COL RPL COL Days 1 to 3 Days 4 to 21Colostrum A 20 20 0 0 0 0 0 0 20 0 (%, w/v)

Colostrum B 0 0 15 15 Reduce to 5 15 5 15 0 15 (%, w/v)

Milk replacer 0 0 0 0 Increase to 13 0 13 0 0 0 (%, w/v)

Iron (mg/kg 250 250 250 250 250 250 250 250 250 250 milk solids)

IgY K88 2 2 2 2 2 2 2 2 2 2 (g/pig/d)

Warm water 0.3–0.53 0.67–1.67 1.84–2.5 2.67–3.33 0.3–0.53 0.67–3.33 (L/pig/d)a Colostrum A — HeadSTART, The Saskatoon Colostrum Company, Saskatoon, Saskatchewan; Colostrum B — Calf’s Choice Total HiCal, The Saskatoon Colostrum Company; milk replacer — WetNurse, Prairie Micro-Tech, Regina, Saskatchewan; iron — Enfamil Fer-In-Sol syrup (30 mg of elemental iron per 5 mL), Mead Johnson & Company, Ottawa, Ontario; IgY K88 — Hyper-Egg K88; J.H. Hare & Associates, Winnipeg, Manitoba.RPL — Replacement: dietary colostrum was gradually replaced with milk replacer from day 11 until 5% colostrum remained in the diet; COL — Colostrum: the diet consisted mainly of bovine colostrum until day 20.



Pregnant sows at the Prairie Swine Center, Saskatoon, Saskatchewan, were used as the source of SF-pCD pigs. Parturition was induced by intramuscular injection of a commercial prosta-glandin analogue (Planate; Merck Animal Health Canada, Kirkland, Quebec) on day 115 of pregnancy. The perivulvar area of the sows and the pens in which the sows were housed were washed with warm water twice on the day before expected parturition and sprayed with an iodine disinfectant (Prepodyne; West Penetone, Ville d’Anjou, Quebec). Before parturition a clean drape was placed behind the sow to reduce the risk of environmental contamination of the piglets. During parturition, the piglets were snatched before contacting the floor, farrowing equipment, or barn facilities. The umbilical cords were clamped and disinfected with Prepodyne. Within 1 min of birth the piglets were placed in sealed plastic

containers fitted with a high-efficiency particulate air filter, which provided filtered fresh air during transportation to a Biosafety Level 2 animal care room with positive pressure ventilation at the University of Saskatchewan. The animal room was disinfected twice at 24-h intervals before pig entry: first with 7% hydrogen peroxide solution (Peroxigard; Bayer, Toronto, Ontario) and then with 1% Virkon solution (Antec International, Lavaltrie, Quebec). The piglets were raised in groups of 2 (for experiment 2) or 3 (for experiment 1) per pen on an elevated plastic floor. The concrete floor beneath the plastic flooring was washed 2 to 3 times each week in experiment 1 and daily in experiment 2. The room temperature was set at 30°C from day 1 to day 21 and was decreased by 1°C weekly thereafter. Before weaning, a heat lamp was provided for each pen and the height of the lamps adjusted according to the pigs’ comfort level

Table II. Ingredients and nutrient levels in the dry starter diet fed to the pigs after weaning

Experiment 2 Experiment 1 STARTER-COL STARTER-CTRLIngredients (%) (%) (%)Wheat 30.90 44.80 30.90Colostrum B 0 20.00 0White fish meal 8.61 0 8.61Oat groat 15.00 15.00 15.00Whey permeate 20.00 11.40 20.00Soybean meal 15.00 3.60 15.00NuProa 5.00 0 5.00Canola oil 3.00 2.00 3.00Limestone 0.43 0.90 0.43Salt 0.44 0.63 0.44Monocalcium phosphate 0 0.63 0Zinc oxide 0.40 0.40 0.40Lysine 0.47 0.36 0.47Starter microbial phytase 0.20 0.20 0.20Choline chloride 0.08 0.08 0.08Methionine 0.25 0.03 0.25Threonine 0.20 0 0.20L-tryptophan 0.03 0 0.03Nutrientsb

Crude protein 21.7 21.5 21.7Crude fat 1.5 8.4 1.5Crude fibre 5.6 1.6 5.6Digestible energy (Mcal/kg) 3.69 3.76 3.69Net energy (Mcal/kg) 2.62 2.76 2.62Calcium 0.9 0.9 0.9Available phosphorus 0.8 0.7 0.8Sodium 0.47 0.45 0.47Total lysine 1.58 1.64 1.58Total threonine:lysine 0.64 0.77 0.64Total methionine cystine:lysine 0.58 0.61 0.58Total tryptophan:lysine 0.17 0.20 0.17a Alltech Canada, Guelph, Ontario.b As-fed basis; estimated 90% dry matter.STARTER-COL — Diet devoid of all animal by-products except 20% (w/v) colostrum B; STARTER-CTRL — Same starter diet as used in experiment 1.



(pigs sleep comfortably on their sides under the heat lamp with their abdomens exposed).

After entry to the animal care room the piglets were fed a liquid diet (Table I) hourly for the first 6 h, every 2 h from 6 to 24 h of age, and 4 times per day (8 am, 12 am, 4 pm, and 10 pm) thereafter until weaning. The pigs were initially bottled-fed, and the liquid diet was also provided in a liquid feeder (product BPW4; Miller Manufacturing Company, Eagan, Minnesota, USA) in the pen to encourage drinking from the feeder. Once a pig was observed to be drinking from the feeder, bottle-feeding was discontinued for that pig. All pigs began to drink from the feeder within 48 h of age. The liquid diets contained combinations of the following ingredients: spray-dried bovine colostrum powder containing at least 25% bovine immunoglobulin G (bIgG) (HeadSTART), spray-dried bovine colostrum powder containing at least 14% bIgG (Calf’s Choice Total HiCal), commercial pig milk replacer (WetNurse; Prairie Micro-Tech, Regina, Saskatchewan), spray-dried whole egg powder containing polyclonal antibodies to E. coli K88 (Hyper-Egg K88; J.H. Hare & Associates, Winnipeg, Manitoba), and an oral iron supplement (Enfamil Fer-In-Sol syrup; Mead Johnson & Company, Ottawa, Ontario; 30 mg of elemental iron per 5 mL). The targeted dry matter fed to each pig in the liquid diet gradually increased from 66 g/d on day 1 to 500 g/d on day 20, and the feeding volume increased from 150 to 3300 mL/d over the same period. The exact feeding amount and volume varied somewhat according to appetite. The pigs were weaned on day 24 (in experiment 1) or day 21 (in experiment 2) to a custom starter diet without porcine by-products (Table II); the treat-ment group in experiment 2 was weaned to a diet that contained bovine colostrum powder.

No prophylactic parenteral antibiotic treatment was used before the pigs entered the animal care room or before weaning, and no antibiotics were added to the starter feeds. In experiment 1, when a pig’s temperature was greater than 40°C, oxytetracycline (Bio-Mycin 200; Boehringer Ingelheim, Burlington, Ontario) was administered intramuscularly (20 mg/mL) to relieve the symptoms of septicemia.

Experiment 1 aimed to compare the effects of 2 different liquid diet formulations on the health of SF-pCD pigs. The 12 SF-pCD pigs used in this experiment were conveniently placed, in the order they arrived at the facility, into 4 3 6 ft (1.23 3 1.85 m) pens (3 pigs per pen) equipped with a liquid feeder. Unlimited access to water was provided by a nipple drinker. All the pigs were housed in the same room throughout the experiment. The 6 pigs in 2 systematically selected pens (pens 1 and 2) were fed a liquid diet composed mainly of bovine colostrum for the first 10 d, and then the colostrum was gradually replaced with milk replacer (RPL) until 5% (w/v) colos-trum was left in the diet; they were designated the RPL group. The 6 pigs in the remaining 2 pens (pens 3 and 4) were fed a liquid diet composed mainly of bovine colostrum (COL) throughout days 1 to 20 and were designated the COL group. The starter diet was introduced to all the pigs on day 20, and the pigs were fully weaned on day 24. Blood samples were drawn from the cranial vena cava of half of the pigs in each group into tubes containing ethylene diamine tetraacetic acid (EDTA) 24 h after the initial colostrum feeding and weekly thereafter. The rectal temperature was measured daily from entry to the animal care room until day 33. The experiment was ter-

minated on day 35, when all the surviving animals were euthanized with intravenous barbiturate and underwent necropsy.

Experiment 2 aimed to evaluate the potential benefits of adding bovine colostrum powder to the dry starter diet fed after weaning. The 12 SF-pCD pigs used in this experiment were housed 2 per pen, conveniently selected. All were kept in the same room throughout the experiment. Before weaning on day 21, all the pigs were fed a liquid diet composed mainly of bovine colostrum. At weaning, the pigs were systematically allocated into 2 groups. The pigs in 3 pens (pens 2, 4, and 6) were fed a dry starter diet devoid of animal by-products except for 20% (w/w) colostrum (Calf’s Choice Total HiCal) and designated the STARTER-COL group. The pigs in the remaining 3 pens (pens 1, 3, and 5) served as the control group and were fed the same starter diet as was used in experiment 1 without additional colostrum; they were designated the STARTER-CTRL group. Blood samples were drawn from the cranial vena cava into EDTA tubes 24 h after the initial colostrum feeding and weekly thereafter. The rectal temperature was measured daily until day 30 and twice weekly thereafter. Half of the pigs in each group were euthanized by intravenous barbiturate on day 42 and the remainder on day 49. Necropsy was done immediately after euthanasia.

Packed cell volume (PCV) and plasma total protein concentration

In both experiments the PCV was determined in the whole blood samples by the capillary tube method, and the plasma total protein concentration was determined by refractometry.

Histopathological examination of tissuesInternal organs were collected at necropsy, preserved in 10%

formalin, and processed for histopathological examination. Tissues examined included brain, nasal turbinate, salivary gland, tonsil, thymus, lung, heart, bronchial lymph node, stomach (fundus and pylorus), duodenum, jejunum, ileum, spiral colon, cecum, mesen-teric lymph node, adrenal gland, kidney, and superficial inguinal lymph node.

Adjunct tests for porcine pathogensIn both experiments the plasma from blood collected immediately

before necropsy was examined for the presence of pathogen-specific antibodies by enzyme-linked immunosorbent assay with commercial kits used according to the manufacturer’s instructions by Prairie Diagnostic Services (PDS), Saskatoon, Saskatchewan. The samples were tested for antibodies to swine influenza virus (SIV) with Swine Influenza Virus Antibody Test Kit H1N1 (part 99-06731) and Swine Influenza Virus Antibody Test Kit H3N2 (part 99-09332) from IDEXX Laboratories, Westbrook, Maine, USA, and for Porcine reproductive and respiratory syndrome virus (PRRSV) with PPRS X3 HerdChek, Porcine Reproductive and Respiratory Virus Antibody Test Kit (part 99-18070), from IDEXX Laboratories. An in-house hemagglu-tination (HA) inhibition assay adapted from a previously described protocol (8) was used by PDS to test for Porcine parvovirus (PPV) antibody. Briefly, PPV propagated in the laboratory from a diagnostic isolate was used as the HA antigen and was incubated at a concen-tration of 6 HA units with 2-fold dilutions of serum starting at 1:20. Hemagglutinating activity was assessed with the use of chicken



erythrocytes. An antibody test for PCV2 was done with a previously described in-house immunoperoxidase monolayer assay (9). In both experiments polymerase chain reaction (PCR) was used to test for the presence of PCV2 DNA in plasma collected at all time points and in spleen, superficial inguinal lymph node, and tonsil collected at necropsy (10). Additionally, in experiment 2 PCR was used to test for the presence of DNA of Torque teno virus (TTV) genogroups 1 and 2 in bone marrow (11). For pigs that died or were euthanized for humane reasons before the end of the experiment, lung, spleen, and joint or abdominal fluid was cultured for bacteria under both aerobic and anaerobic conditions.

Plasma IgG concentrationsThe concentration of bIgG in plasma was determined by radial

immunodiffusion (RID) as previously described (12). The plasma half-life of bIgG was calculated with the following formula:

ConBT1/2 = (T*log2)/log ConE

where: T1/2 is the half-life, T the total time period, ConB the begin-ning bIgG concentration, and ConE the ending bIgG concentration. All plasma collected in experiment 2 was also tested for porcine (p)IgG by RID as previous described (13). The plasma pIgG concen-trations were not compared by statistical methods because half of the pigs in each group were euthanized on day 42 and the remain-der on day 49, which substantially reduced the statistical power (n = 3).

Statistical analysisDifferences between the groups in frequencies of lesions were

compared by Fisher’s exact test. The group difference in plasma bIgG concentration on day 1 was compared by the Mann–Whitney U test. The numbers of fever-days, defined as the total number of days that individual pigs had a body temperature of 39.5°C or greater, were compared between groups by generalized linear models with the use of a Poisson regression and SPSS Predictive Analytics SoftWare Statistics 18 (SPSS, Chicago, Illinois, USA). Probabilities of less than 0.05 were considered statistically significant.

Re s u l t s

Experiment 1All the pigs learned to drink from the liquid feeder within 48 h

of entry to the animal care room. Four of the six RPL pigs and all the COL pigs survived until the termination of the experiment (on day 35). One RPL pig was observed to have atresia ani on day 2, defecating by way of a rectovaginal fistula. Lethargy, anorexia, and temperature elevation (to 40.4°C) developed on day 16, and the ani-mal was euthanized by intravenous administration of barbiturate. Acute fibrinous polyserositis and renal petechiae were noted grossly. Histopathological changes were consistent with acute septicemia, and E. coli was cultured from lung, spleen, and joint fluid. Lethargy, anorexia, and temperature elevation (to 41.4°C) developed in another

Figure 1. Body temperatures of snatch-farrowed, porcine-colostrum-deprived (SF-pCD) pigs in experiment 1. Between days 11 and 20 of life the pigs fed bovine colostrum as the main component of their diet (left panel) had significantly fewer days of fever (P = 0.009) than the pigs for which the colostrum was gradually replaced with milk replacer from day 11 (right panel) before weaning to a dry feed.

Days Days



RPL pig, on day 22. Gross and histological lesions indicated acute septicemia, and E. coli was cultured from lung, spleen, and abdomi-nal fluid. Both E. coli isolates were not further characterized.

The RPL pigs had significantly more fever-days than the COL pigs from day 11 to day 20 (16 fever-days in 50 total pig-days for the RPL pigs versus 5 fever-days in 54 pig-days for the COL pigs; P = 0.009) but not from day 1 to day 10 or from day 21 to day 33 (P . 0.05) (Figure 1). Two of the 4 RPL pigs and 4 of the 6 COL pigs had mild neutrophilic infiltration in the mesenteric lymph nodes, but the dif-ference was not statistically significant (P . 0.05). No other patho-logical changes were observed in the pigs that survived until day 35.

The RPL pigs had marginally but significantly higher plasma concentrations of bIgG at days 1 and 7 (P = 0.0495) than the COL pigs even though the groups had the same diet during this time. The levels were highest 24 h after initial colostrum intake and then decreased rapidly in the first 2 wk of life to negligible levels (Figure 2). The half-life of bIgG in experiment 1 was 9.2 d. The PCV and plasma total protein concentration remained within normal limits throughout the experiment (data not shown).

At termination, all the pigs were negative for antibodies to SIV (H1N1 and H3N2), PRRSV, and PPV. However, 1 COL pig was posi-tive for PCV2 antibody, and PCV2 DNA was detected by PCR in the plasma from day 21 to day 35, as well as in spleen and superficial inguinal lymph node collected at necropsy. Another COL pig, reared in the same pen, remained negative for PCV2 antibody but had PCV2 DNA detected by PCR in spleen and superficial inguinal lymph node collected at necropsy.

Experiment 2All the pigs learned to drink from the liquid feeder by 48 h. All

pigs in both groups survived until the end of the experiment. The feces of all the STARTER-CTRL pigs were of normal consistency throughout the experiment. Four of the six STARTER-COL pigs had semiliquid diarrhea from day 27 (6 d after weaning) until the end

of the experiment, but they remained alert and otherwise healthy. Necropsy of these pigs revealed a dilated colon and cecum that contained unformed feces. The large intestinal mucosa was red-dened, and mesocolonic blood vessels were congested. Histologic examination showed typhlocolitis with combinations of the follow-ing changes: bacterial attachment on the mucosa, degeneration and necrosis of the epithelial cells, and congestion, edema, and mixed inflammatory cell infiltration of the lamina propria. No adjunct tests were carried out to determine the cause of the typhlocolitis. In addition, the pancreas was firm and had a nodular appearance on gross examination in all the STARTER-COL pigs; microscopic examination showed pancreatic glandular epithelial degeneration and necrosis with regeneration and fibrosis. In the STARTER-CTRL pigs the pancreas was normal macroscopically and microscopically except in 1 pig, which had mild microscopic pancreatic degeneration. Both groups had only 2 fever-days in total.

As in experiment 1, the plasma concentrations of bIgG were highest 1 d after initial colostrum intake, then decreased rapidly in the first 2 wk to negligible levels (Figure 2). The half-life of bIgG in experiment 2 was 5.5 d. Porcine IgG was not detectable in the plasma until day 21 and then gradually increased to 5 g/L by day 49 (Figure 2). The PCV and plasma total protein concentration remained within normal limits through the experiment (data not shown).

All the pigs were negative at the end of experiment 2 for antibod-ies to SIV (H1N1 and H3N2), PRRSV, PPV, and PCV2. The absence of PCV2 infection was confirmed by PCR in plasma and tonsil; however, PCR detected TTV1 DNA in 1 STARTER-COL pig and 4 STARTER-CTRL pigs, as well as TTV2 DNA in 1 of the latter 4 pigs.

D i s c u s s i o nIn the current study, neonatal SF-pCD pigs were successfully

raised on a bovine-colostrum-based liquid diet, all remaining serologically negative for SIV, PRRSV, and PPV. Moreover, all but 1 pig remained serologically negative for PCV2, and only 2 pigs (in experiment 1) had detectable PCV2 DNA in serum or tissues. In both experiments, all the pigs fed mainly bovine colostrum before weaning survived until termination of the experiment.

The bovine colostrum was likely the main factor contributing to the high survival rate. It is well-known that the epitheliochorial pla-centation in pigs and other farm mammals prevents macromolecule transportation from dam to fetus. Thus, pigs are born hypogam-maglobulinemic, with only trace amounts of immunoglobulin in the blood. As a result, neonatal pigs have a low survival rate when neither sow colostrum nor immunologic supplements derived from other sources are fed, in spite of intensive treatment with antibiot-ics (14). In other studies using bovine colostrum in neonatal pigs, colostrum was fed for only a few days (5,6,14), and the survival rate was at most 80% (6,14). In the present experiments, the use of com-mercially available spray-dried bovine colostrum powder allowed more protracted use of the colostrum and enables the SF-pCD tech-nique to be easily adopted by others. It is clear from our results that providing bovine colostrum to pigs after gut closure provides health benefits up to weaning at 3 wk of age. Although immunoglobulins cannot be absorbed in a substantial quantity after gut closure, and the plasma half-life of bIgG in these experiments was shorter than

Figure 2. Plasma concentrations of bovine and porcine immunoglobulin G (bIgG and pIgG) in the SF-pCD pigs in experiments 1 and 2. Squares indi-cate mean values for experiment 1. Triangles indicate mean values for bIgG and diamonds mean values for pIgG in experiment 2. The vertical lines represent standard deviations.



that of pIgG (at least 12 to 14 d) (15), they and the other antimicrobial substances present in colostrum may provide local immunity in the gastrointestinal tract.

It is noteworthy, however, that bovine colostrum fed after weaning to the SF-pCD pigs, in experiment 2, was associated with persistent diarrhea, typhlocolitis, and pancreatic degeneration. The mechanism for this phenomenon was not determined. The level of fat in the diet supplemented with colostrum was 8.4% compared with 1.5% in the diet without the supplementation (Table II). The additional colostrum powder in the diet after weaning, without consideration of the fat lev-els in the diet, may exceed the capacity of the pancreatic enzymes and cause maldigestion and diarrhea (likely osmotic). Irrespective of the reason, the finding of no benefit to supplementation with bovine colos-trum after weaning substantially reduces the cost of raising these pigs.

The normal PCV of the SF-pCD pigs indicates that oral iron supplementation was sufficient to prevent anemia. The industry-standard practice of administering iron dextran parenterally was not used in this study since there was a concern that any injection would increase the risk of septicemia. For the same reason, only half of the pigs were bled in experiment 1. However, in experiment 2, when all the pigs were bled weekly, no septicemia occurred in any pig. Thus, injections and blood collection can be done in SF-pCD pigs without health compromise, provided the procedures are conducted hygienically. Indeed, in a subsequent SF-pCD experiment in the same laboratory, iron dextran was administered intramuscularly in the neck without ill effects (data not shown).

The roles of polyclonal antibodies to E. coli K88, which reportedly prevented diarrhea due to K88 (F4) E. coli (16), were not character-ized in this study. This component was included in the diet to help avoid potential illness due to K88 E. coli because F4 colibacillosis is a common disease of young pigs in the commercial swine industry, and vaccination of dams before farrowing is routine.

Pyrexia was rare in experiment 2, in contrast to experiment 1. This may be due to the higher level of sanitation during experiment 2, as the concrete under the plastic floor was cleaned more frequently (daily) than in experiment 1 (2 to 3 times weekly), discouraging the accumulation of fecal microorganisms in the environment. The benefit of high sanitation is further emphasized by the fact that when SF-pCD pigs were raised on a concrete floor bedded with straw, diarrhea and septicemia due to E. coli developed in 4 of 9 pigs (5). Thus, good hygiene practices are strongly recommended for the health of SF-pCD pigs.

The results of this experiment demonstrate that it is possible to raise SF-pCD pigs that remain free of porcine pathogens, including SIV, PRRSV, PPV, and PCV2. The source farm was serologically free of PRRSV but positive for SIV, PPV, and PCV2. Snatch-farrowing, preventing contact with barn facilities, minimizing exposure to barn air, and disinfecting the sows and piglets are likely all critical in pre-venting horizontal transmission of these pathogens to SF-pCD pigs. Although the source of PCV2 infection in 2 pigs in experiment 1 was not definitively identified, retrospective evaluation of the biosecurity protocols guiding entry into the animal care room and the timing of the infections implicated a bottle of antibiotic previously used on a PCV2-positive farm. In experiment 1 all the pigs remained PCV2-negative before day 21, which indicates that vertical transmission of PCV2 was unlikely. After the biosecurity measures were improved,

all the pigs in experiment 2 remained PCV2-negative. Thus, it is possible, although not guaranteed, that PCV2-negative pigs can be obtained from a PCV2-positive farm using the SF-pCD method, provided the piglets are not infected prenatally. A recent study showed that 39.9% of the piglets on 5 North American farms were born PCV2-viremic (17). Although in that study most of the piglets were born of gilts, which may be more likely to vertically infect their progeny in utero, the putatively high rate of viremia at birth on some farms highlights that PCV2-free status is not guaranteed. We recom-mend excluding PCV2-viremic sows to increase the probability of obtaining PCV2-free piglets.

Similarly, the presence of TTV1 and TTV2 in some of the SF-pCD pigs in the present study was not unexpected since vertical transmis-sion of TTV had been reported (18,19). Even gnotobiotic derivation cannot guarantee freedom from infection with TTV or other verti-cally transmitted pathogens.

The SF-pCD method has advantages over the conventional SPF method in obtaining pathogen-free pigs and is less technically demanding than CDCD and gnotobiotic models. Since conventional SPF pigs have suckled their dams and remain on the farm before their inclusion in an experiment, there is a high risk of exposure and possibly infection with organisms circulating on the farm. Moreover, the presence of maternally derived antibodies against endemic pathogens means that SPF piglets cannot be used for experiments until passive immunity has decayed. Unlike the CDCD and gnotobi-otic methods, SF-pCD derivation requires neither surgery nor sterile compartments. Thus, the technical requirements are low and easily adapted to any laboratory. The natural birth of SF-pCD pigs allows them to be exposed to the vaginal flora and the potential for infection by the vaginal microbiota and any pathogens present. The technique also avoids the need to sacrifice the donor sow after surgery.

In conclusion, this study established a method to raise SF-pCD pigs with high health and survival that balances convenience, freedom from major pathogens, animal welfare, cost, and pathogen susceptibility. The SF-pCD model is a good alternative to current methods of producing pigs for infectious disease research. The main disadvantages are the intensive labor associated with bottle-feeding for the first 48 h, the risk of contamination by vaginal microbiota, and the risk of septicemia if the environment is not hygienic.

A c k n o w l e d g m e n t sThis research was made possible with a contribution from

Saskatchewan Agriculture and Food through the Agriculture Development Fund. The bovine colostrum powder and the bIgG assay kits were generously donated by The Saskatoon Colostrum Company, Saskatoon, Saskatchewan. The Hyper-Egg K88 was donated by J.H. Hare & Associates, Winnipeg, Manitoba. The authors appreciated the help of diagnosticians at Prairie Diagnostic Services, Saskatoon, and Dr. Malachy Young, of Gowan’s Feed Consulting, Wainwright, Alberta, for formulating the starter diets. Daniel Petri’s expertise in raising colostrum-deprived pigs and the assistance of Crissie Auckland, Shrijana Dhakal, Rayna Gundvalsen, Blake Balog, Atul Desai, and the staff at the Animal Care Unit of Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, was critical to the success of this research.



Re f e r e n c e s 1. Struve R. A source of CDCD swine for research. Proc Am Assoc

Swine Vet Conf 1999:117. 2. Sangild PT, Holtug K, Diernaes L, Schmidt M, Skadhauge E.

Birth and prematurity influence intestinal function in the new-born pig. Comp Biochem Physiol A Physiol 1997;118:359–361.

3. Fowden AL, Li J, Forhead AJ. Glucocorticoids and the prepara-tion for life after birth: Are there long-term consequences of the life insurance? Proc Nutr Soc 1998;57:113–122.

4. Butler JE, Weber P, Sinkora M, et al. Antibody repertoire develop-ment in fetal and neonatal piglets. VIII. Colonization is required for newborn piglets to make serum antibodies to T-dependent and type 2 T-independent antigens. J Immunol 2002;169:6822–6830.

5. Oliveira S, Galina L, Blanco I, Canals A, Pijoan C. Naturally-farrowed, artificially-reared pigs as an alternative model for experimental infection by Haemophilus parasuis. Can J Vet Res 2003;67:146–150.

6. Blanco I, Galina-Pantoja L, Oliveira S, Pijoan C, Sanchez C, Canals A. Comparison between Haemophilus parasuis infection in colostrum-deprived and sow-reared piglets. Vet Microbiol 2004;103:21–27.

7. Olfert ED, Cross BM, McWilliam AA, eds. Guide to the Care and Use of Experimental Animals. 2nd ed. Volume 1. Ottawa, Ontario: Canadian Council on Animal Care, 1993. Available at www.ccac.ca/Documents/Standards/Guidelines/Experimental_Animals_Vol1.pdf

8. Carmichael LE, Joubert JC, Pollock RV. Hemagglutination by canine parvovirus: Serologic studies and diagnostic applications. Am J Vet Res 1980;41:784–791.

9. McNair I, Marshall M, McNeilly F, et al. Interlaboratory testing of porcine sera for antibodies to porcine circovirus type 2. J Vet Diagn Invest 2004;16:164–166.

10. McIntosh KA, Tumber A, Harding JCS, Krakowka S, Ellis JA, Hill JE. Development and validation of a SYBR green real-time

PCR for the quantification of porcine circovirus type 2 in serum, buffy coat, feces, and multiple tissues. Vet Microbiol 2009; 133:23–33.

11. Kekarainen T, Sibila M, Segalés J. Prevalence of swine Torque teno virus in post-weaning multisystemic wasting syndrome (PMWS)-affected and non-PMWS-affected pigs in Spain. J Gen Virol 2006;87:833–837.

12. Chelack BJ, Morley PS, Haines DM. Evaluation of methods for dehydration of bovine colostrum for total replacement of normal colostrum in calves. Can Vet J 1993;34:407–412.

13. Olson GL, Robine L, Rosengren LB, et al. Parturition induction 2 days prior to term decreases birth weight, lactational growth but not piglet maturity, health or post-weaning growth. Can J Anim Sci 2009;89:219–228.

14. Gomez GG, Phillips O, Goforth RA. Effect of immunoglobulin source on survival, growth, and hematological and immunologi-cal variables in pigs. J Anim Sci 1998;76:1–7.

15. Rooke JA, Carranca C, Bland IM, et al. Relationships between passive absorption of immunoglobulin G by the piglet and plasma concentrations of immunoglobulin G at weaning. Livest Prod Sci 2003;81:223–234.

16. Marquardt RR, Jin LZ, Kim JW, Fang L, Frohlich AA, Baidoo SK. Passive protective effect of egg yolk antibodies against entero-toxigenic Escherichia coli K88 infection in neonatal and early weaned piglets. FEMS Immunol Med Microbiol 1999;23:283–288.

17. Shen H, Wang C, Madson DM, Opriessnig T. High prevalence of porcine circovirus viremia in newborn piglets in five clinically normal swine breeding herds in North America. Prev Vet Med 2010;97:228–236.

18. Martinez-Guino L, Kekarainen T, Segalés J. Evidence of Torque teno virus (TTV) vertical transmission in swine. Theriogenology 2009;71:1390–1395.

19. Pozzuto T, Mueller B, Meehan B, et al. In utero transmission of porcine torque teno viruses. Vet Microbiol 2009;137:375–379.


Article


Prediction of serum IgG concentration by indirect techniques with adjustment for age and clinical and laboratory covariates in critically

ill newborn calvesGilles Fecteau, Julie Arsenault, Julie Paré, David C. Van Metre, Charles A. Holmberg, Bradford P. Smith

A b s t r a c tThe objective of this study was to develop prediction models for the serum IgG concentration in critically ill calves based on indirect assays and to assess if the predictive ability of the models could be improved by inclusion of age, clinical covariates, and/or laboratory covariates. Seventy-eight critically ill calves between 1 and 13 days old were selected from 1 farm. Statistical models to predict IgG concentration from the results of the radial immunodiffusion test, the gold standard, were built as a function of indirect assays of serum and plasma protein concentrations, zinc sulfate (ZnSO4) turbidity and transmittance, and serum g-glutamyl transferase (GGT) activity. For each assay 4 models were built: without covariates, with age, with age and clinical covariates (infection and dehydration status), and with age and laboratory covariates (fibrinogen concentration and packed cell volume). For the protein models, dehydration status (clinical model) and fibrinogen concentration (laboratory model) were selected for inclusion owing to their statistical significance. These variables increased the coefficient of determination (R2) of the models by $ 7% but did not significantly improve the sensitivity or specificity of the models to predict passive transfer with a cutoff IgG concentration of 1000 mg/dL. For the GGT assay, including age as a covariate increased the R2 of the model by 3%. For the ZnSO4 turbidity test, none of the covariates were statistically significant. Overall, the R2 of the models ranged from 34% to 62%. This study has provided insight into the importance of adjusting for covariates when using indirect assays to predict IgG concentration in critically ill calves. Results also indicate that ZnSO4 transmittance and turbidity assays could be used advantageously in a field setting.

R é s u m éL’objectif de cette étude était de développer un modèle de prédiction de la concentration sérique des IgG chez des veaux malades à partir de techniques indirectes, et d’évaluer si la capacité de prédiction du modèle peut être améliorée par l’inclusion de l’âge et de certains paramètres clinique et de laboratoire. Soixante-dix-huit veaux gravement malades âgés entre 1 et 13 jours et élevés sur une même ferme ont été sélectionnés. Des modèles statistiques pour prédire la concentration sérique des IgG mesurée par immunodiffusion radiale (épreuve de référence) ont été construits à partir de mesures indirectes (concentration des protéines sériques, concentration des protéines plasmatiques, turbidité au sulfate de zinc, transmittance au sulfate de zinc, concentration sérique de la GGT (g-glutamyl transférase). Pour chacune des techniques indirectes, 4 modèles ont été construits : sans covariable, avec l’âge seulement, avec l’âge et des paramètres cliniques (présence d’un foyer d’infection et état d’hydratation), et avec l’âge et les paramètres de laboratoires (concentration plasmatique en fibrinogène, hématocrite). Pour les modèles incluant les protéines sériques et plasmatiques, l’état d’hydratation et la concentration plasmatique en fibrinogène ont été retenus (statistiquement significatif). L’inclusion de ces variables augmentait la valeur du R2 des deux modèles de $ 7 %, mais n’avait pas d’impact significatif sur la sensibilité ou spécificité des modèles pour prédire le transfert de l’immunité passive (utilisant une valeur seuil de 1000 mg/dL). Pour le modèle GGT, l’âge a été retenu comme covariable et son inclusion augmentait le R2 de 3 %. Pour les modèles de sulfate de zinc, aucune covariable n’était statistiquement significative. La valeur du R2 des différents modèles construits variait de 34 % à 62 %. Les résultats de cette étude soutiennent l’importance d’inclure certains autres paramètres pour prédire le succès du transfert de l’immunité passive à partir de techniques indirectes dans une population de veaux gravement malades. Ils indiquent également que les techniques de turbidité et de transmittance au sulfate de zinc sont appropriées pour une utilisation à la ferme.

(Traduit par les auteurs)

Faculté de médecine vétérinaire, Université de Montréal, CP 5000, Saint-Hyacinthe, Québec J2S 7C6 (Fecteau, Arsenault); Epidemiology and Surveillance Section, Canadian Food Inspection Agency, CP 5000, Saint-Hyacinthe, Québec J2S 7C6 (Paré); College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado 80523-1678, USA (Van Metre); School of Veterinary Medicine, University of California, Davis, California 95616, USA (Holmberg, Smith).

Address all correspondence to Dr. Gilles Fecteau; telephone: 450-773-8521, ext. 8337; fax: 450-778-8158; e-mail: [email protected]

Received November 11, 2011. Accepted February 16, 2012.



I n t r o d u c t i o nPartial or complete failure of passive transfer (FPT) of immuno-

globulins to suckling calves is a significant problem, increasing the risk of both illness and death (1–3). It is assessed by measuring the serum IgG concentration, ideally within 24 to 48 h after birth. Radial immunodiffusion (RID) is the only test that directly measures this concentration, and it is considered the gold standard.

In practice, evaluation of the serum IgG concentration is often done in older calves with neonatal disease. The level is rarely mea-sured directly. Instead, less expensive and more rapid tests measur-ing the concentrations of total globulins or other proteins known to be correlated with the IgG concentration (3,4) are generally used. Available tests include measurements of the serum concentration of total solids by refractometry, sodium sulfate turbidity, zinc sulfate (ZnSO4) turbidity, serum g-glutamyl transferase (GGT) activity, and whole blood glutaraldehyde gelation.

In previous studies the interpretation and validation of the results of various indirect assays for the serum IgG concentration were generally based on defining a cutoff value for predicting the failure or adequacy of passive transfer (5,6). However, in a clinical setting such an approach does not allow for critical appraisal of the level of IgG passive transfer, which may directly influence decision-making in regard to treatment and prognosis. Moreover, the impact of dif-ferent clinical parameters, such as dehydration and sepsis, on the association between a particular indirect assay and the gold standard was not taken into account in previous studies and warrants further investigation.

This study proposed prediction models based on various indirect assays for estimating the serum IgG concentration in ill calves and compared the predictive ability of the models with and without the inclusion of age, clinical covariates, and laboratory covariates. The null hypothesis was that these covariates do not have an impact on the association between various indirect assays and the serum IgG concentration.

M a t e r i a l s a n d m e t h o d sThe data used in this study originated from an observational study

based on a convenience sample. From a large calf-rearing farm in the San Joaquin Valley of California, we selected 78 critically ill male Holstein calves between 1 and 13 d of age with a total clinical score of 5.5 or higher. This score is based on evaluation of feces, hydration status, calf behavior, umbilicus, and scleral vessel characteristics (7). There was no information available on colostrum feeding since the animals arrived on the premises at approximately 1 d of age. At the time of selection the calves were clinically evaluated, age was noted, and blood samples were collected by a previously described technique (8).

Clinical evaluationThe presence of infection and hydration status were recorded

according to a previously described protocol (7). Infection was con-sidered present if at least 1 of the following was noted: a positive blood culture, visual detection of an infection site (e.g., septic joint, hypopyon, mucopurulent discharge, or abscess), or an umbilical

score $ 2. Calves presenting obvious signs of dehydration (such as sunken eyes) were considered dehydrated.

Laboratory evaluationWithin 8 h of collection, blood samples were centrifuged and

analyzed. Packed cell volume (PCV) was estimated in microcapil-lary tubes, fibrinogen concentration was evaluated by the heat precipitation method, and serum and plasma protein concentrations were determined with a refractometer. For all other analyses, serum was frozen at 220°C for a maximum of 6 mo. The concentration of IgG was evaluated with an RID kit (Immunoglobulins Bovine IgG Kit; ICN Biomedicals, Costa Mesa, California, USA) according to manufacturer specifications. The RID plates were read with an electronic plate reader (Binding Site, San Diego, California, USA). Serum GGT activity was determined with the Synchron CX5 device (Beckman Coulter, Brea, California, USA). Zinc sulfate turbidity and transmittance were evaluated with a ZnSO4 heptahydrate solution prepared by adding 250 mg of ZnSO4.7H2O to 1 L of distilled water. Approximately 4 mL of the solution was put in glass vacuum tubes 16 mm in diameter, with caution taken to maintain negative pressure, and 0.1 mL of serum was added. After gentle mixing, the tubes were incubated for 1 h at room temperature. Zinc sulfate turbidity was then evaluated semiquantitatively (for total failure, partial failure, or adequate protection) by placing newspaper behind the tubes: a tube through which letters could be easily distinguished was categorized as showing total failure, a tube through which letters were illegible but detectable as shadows was designated as showing partial failure, and a totally opaque tube was categorized as showing adequate

Table I. Descriptive statistics for the variables used in models to predict the serum IgG concentration in 78 critically ill calves

Continuous variables Unit Range MedianAssay Serum protein concentration g/L 36–81 46 Plasma protein concentration g/L 43–90 56 ZnSO4 transmittance % 0–64 32 Serum GGT activity IU/L 8–586 19

Demographic covariate Age Days 1–13 5

Laboratory covariate Packed cell volume % 21–54 35 Fibrinogen concentration g/L 3–16 7Categorical covariates Category nAssay ZnSO4 turbidity Total failure 39 Partial failure 26 Adequate protection 13

Clinical covariate Infection Yes 16 No 62 Dehydration Yes 19 No 59ZnSO4 — zinc sulfate; GGT — g-glutamyl transferase.



Table II. Estimated coefficient of regression (b) with standard error (SE) and P-value for various linear regression models to predict the log-transformed serum IgG concentration in the calvesa

Model and variable b SE P-valueSerum protein No covariate (R2 = 41%) Intercept 20.450 0.365 0.22 Serum protein level 0.054 0.007 , 0.001 Age and clinical covariates (R2 = 49%) Intercept 20.679 0.350 0.06 Serum protein level 0.061 0.007 , 0.001 Dehydration 20.448 0.133 , 0.01 Age and laboratory covariates (R2 = 48%) Intercept 20.138 0.360 0.70 Serum protein level 0.057 0.007 , 0.001 Fibrinogen concentration 20.062 0.020 , 0.01

Plasma protein No covariate (R2 = 34%) Intercept 20.502 0.428 0.24 Plasma protein level 0.046 0.007 , 0.001 Age and clinical covariates (R2 = 42%) Intercept 20.803 0.416 0.06 Plasma protein level 0.053 0.007 , 0.001 Dehydration 20.451 0.143 , 0.01 Age and laboratory covariates (R2 = 50%) Intercept 20.431 0.376 0.26 Plasma protein level 0.058 0.007 , 0.001 Fibrinogen concentration 20.100 0.020 , 0.001

GGT No covariate (R2 = 47%) Intercept 0.578 0.202 , 0.01 GGT activity (log-transformed) 1.103 0.135 , 0.001 Age (R2 = 50%) Intercept 0.313 0.233 0.18 GGT activity (log-transformed) 1.175 0.136 , 0.001 Age 0.031 0.014 0.04

ZnSO4 transmittance No covariate (R2 = 62%) Intercept 3.210 0.104 , 0.001 Transmittance 20.033 0.003 , 0.001ZnSO4 turbidity No covariate (R2 = 60%) Intercept 1.746 0.068 , 0.001 Adequate protection 1.433 0.136 , 0.001 Partial failure 0.557 0.108 , 0.001a The log (IgG) can be predicted using the following equations:

log (IgG) = bIntercept Sb3

Where 3 = the value of the variable for the individual and b is the estimated regression coefficient for this variable.



protection. Zinc sulfate transmittance was obtained by reading the optical density of each tube with a spectrophotometer (Coleman Spec 20; PerkinElmer, Downers Grove, Illinois, USA) at 485 nm. All ZnSO4 assays were performed the same day by the same investigator on the same ZnSO4.7H2O preparation.

Statistical analysisFor each of the 5 assays, 4 linear regression models were built

with the logarithmic transformation (base 10) of the serum IgG concentration measured by RID as the dependent variable (9). The assay results were put in the models in their original scales, apart from the GGT results, which were log-transformed (base 10). The log transformations were used to improve the normality and linear-ity of the residuals. Full main-effect models were built first. Of the 4 models, the 1st included only the assay, the 2nd also included age, and the 3rd and 4th models also included age and either clinical covariates (dehydration and infection) or laboratory covariates (PCV and fibrinogen concentration), respectively. A backward-elimination procedure based on the F-test with P . 0.05 as the criterion of elimi-nation was then used. Variables were removed only if removal did not change the estimated coefficient of the assay by more than 20%. First-order interactions between assay and covariates were tested 1 at a time with the resulting models and were selected for inclusion in the final model if their P-value was less than 0.05 (F-test). For each final model the coefficient of determination (R2) was computed. Normal distribution of the residuals was visually assessed on a normal-probability plot. Outliers were detected visually as well as by using studentized residuals and the difference in fitted values (DFFITs) after deletion of a case. Studentized residuals were plot-ted against the predicted value to evaluate homoscedasticity and against each continuous predictor variable to evaluate the linearity of the regression function. When an outlier was identified, the model was refitted to exclude the outlier, and changes in the coefficient or P-value were studied. All statistical analyses were performed with the PROC GLM procedure in SAS 9.1 (SAS Institute, Cary, North Carolina, USA).

The ability of each final model to predict passive transfer status was computed. The RID test result was used as the gold standard: calves were classified as having FPT if the serum IgG concentration was less than 1000 mg/dL and otherwise were considered to have adequate passive transfer (APT). The sensitivity of the various models was estimated as the percentage of calves with predicted FPT (gauged by model predictions transformed from a logarithmic scale to the original scale) among the calves with FPT. Specificity was estimated as the percentage of calves with predicted APT among the calves with APT. Confidence intervals (95%) for the sensitivity

and specificity estimates were calculated with the PROC FREQ pro-cedure in SAS 9.1 with the use of exact computations for binomial distributions. For each assay, McNemar tests with exact computa-tions were performed to determine if the sensitivity and specificity differed significantly (a = 0.05) between each 2 3 2 combination of the 4 models. The positive predictive value for each model was estimated as the probability (%) of FPT among the calves with pre-dicted FPT. Similarly, the negative predictive value for each model was estimated as the probability (%) of APT among the calves with predicted APT. These predictive values were calculated with the assumption of the same proportion of FPT in the population as observed in the study sample.

Re s u l t sIn the 78 calves included in the study, the serum IgG concentration

ranged from 5 to 6980 (median 135) mg/dL. With a cut-off value of 1000 mg/dL the proportion of calves with FPT was 86%. Descriptive statistics for the assay results and covariates are presented in Table I.

The final linear regression models are presented in Table II. Only models with covariates selected for inclusion are presented. For the serum protein, plasma protein, and GGT assays, at least 1 covari-ate was kept in the final model owing to its statistical significance (P , 0.05). For the 2 assays based on ZnSO4, none of the tested covariates was significantly associated with the log-transformed serum concentration of IgG.

During the model building, no evidence of confounding was detected, and no interactions were statistically significant (all P $ 0.10). Visual analysis of the residuals did not reveal any impor-tant departure from normality, homoscedasticity, or linearity of the regression function. No outlier was detected visually or according to studentized residuals (all # 2.8) or DFFITs (all # 0.64) except for a relatively high studentized residual value (3.15) in a single model in 1 calf. Because exclusion of this value did not influence model selec-tion and had only a minor impact on coefficient estimates (, 15%), it was kept for analysis.

The estimated coefficients of regression were used to predict the log IgG serum concentration for each calf with each model. As an example, the equation presented in Figure 1 was used to predict the log IgG concentration from the plasma protein model with age and laboratory covariates in a calf with a plasma protein concentration of 60 g/L and a fibrinogen concentration of 7 g/L. The predicted log IgG value was then back-transformed on its original scale (i.e., 102.349 = 223 mg/dL) (Figure 1).

The sensitivity, specificity, and predictive values of the various models are presented in Table III. No statistically significant differ-

Predicted log(IgG) = bIntercept [bPlasma proteins 3 Plasma proteins (g/L)] [bFibrinogen 3 Fibrinogen (g/L)]

Predicted log(IgG) = 20.431 [0.058 3 60] [20.100 3 7]

Predicted log(IgG) = 2.349

Figure 1. Equation used to predict the log serum IgG concentration from the plasma protein model developed in this study with age and laboratory covariates. This model included the intercept and the plasma protein and fibrinogen concentrations as variables according to the model selection building procedure.



ence in sensitivity or specificity was detected between the 4 models of each assay (all P . 0.25 by the McNemar exact test).

D i s c u s s i o nIn this study, FPT was common, affecting 86% of the study

population. Although this proportion is not representative of that in the normal population of dairy calves, it may reflect the occur-rence in sick calves presented to veterinarians. All the animals were male Holstein calves fed and housed in an identical manner and sampled over a short period (within 5 d), and all the blood samples were stored and analyzed in an identical manner. This protocol likely minimized the number of potential confounding covariates. The solution used for the evaluation of ZnSO4 turbidity and trans-mittance was prepared with a 250-mg/L solution rather than the typical 208-mg/L solution (4). In our clinical experience, adding a 3rd category (for partial failure) instead of relying on the traditional 2 categories makes readings easier; it is also likely to provide more information about the serum IgG concentration.

Among the clinical covariates, hydration status had a significant effect of similar magnitude in the serum and plasma protein models. Clinical dehydration was associated with a decrease in serum IgG

concentration, perhaps because of poor suckling as a common factor. However, this association was unexpected since PCV, considered an indicator of dehydration, was not selected for inclusion in the labora-tory models. Further exploration of the data showed that hydration status is a poor predictor of PCV in a simple linear regression model (R2 = 3%, P = 0.12). Thus, other unidentified covariates might play an important role in the observed relationship.

Among the laboratory covariates, fibrinogen concentration was a significant predictor of IgG concentration in the models based on serum and plasma protein concentrations estimated by refractometry. Inclusion of the fibrinogen concentration most likely provided an adjustment for the presence of nonimmunoglobulin proteins, such as acute-phase proteins, in the presence of infection. When fibrinogen concentration was included as a covariate the R2 increased 7% with the serum model and 16% with the plasma model. This observation supports the importance of such adjustment for IgG prediction. Age had a significant effect in the GGT model. An increase in serum IgG concentration was observed as age increased, as previously reported (10). However, the inclusion of age had a minor impact on the R2 of the model.

The coefficient of determination for the 10 models ranged from 34% to 62%. Although this indicates that about half of the changes

Table III. Sensitivity, specificity, positive and negative predictive values of the various models in predicting failure of passive transfer (FPT) of immunoglobulins to the calves, with a cut-off for the IgG concentration of , 1000 mg/dL and radial immunodiffusion (RID) as the gold standard

Sensitivity Specificity Positive Negative Model and (95% confidence (95% confidence predictive predictive covariates included interval)a interval)b valuec valued

Serum protein None 100 (95, 100) 18 (2, 52) 88 (67/76) 100 (2/2) Dehydration 100 (95, 100) 27 (6, 61) 89 (67/75) 100 (3/3) Fibrinogen level 100 (95, 100) 45 (17, 77) 92 (67/73) 100 (5/5)

Plasma protein None 100 (95, 100) 27 (6, 61) 89 (67/75) 100 (3/3) Dehydration 100 (95, 100) 27 (6, 61) 89 (67/75) 100 (3/3) Fibrinogen level 100 (95, 100) 36 (11, 69) 91 (67/74) 100 (4/4)

GGT None 97 (90, 100) 27 (6, 61) 89 (65/73) 60 (3/5) Age 99 (92, 100) 27 (6, 61) 89 (66/74) 75 (3/4)

ZnSO4 transmittance None 99 (92, 100) 73 (39, 94) 96 (66/69) 89 (8/9)

ZnSO4 turbidity None 94 (85, 98) 82 (48, 98) 97 (63/65) 69 (9/13)a Percentage of calves with FPT as predicted by the model among the 67 calves with FPT according to RID.b Percentage of calves with adequate passive transfer (APT) as predicted by the model among the 11 calves with APT according to RID.c Numbers in parenthesis represent the number of calves with FPT as predicted by the model over the number of calves with FPT according to RID, based on an 86% prevalence of FPT in the sample.d Numbers in parenthesis represent the number of calves with APT as predicted by the model over the number of calves with APT according to RID, based on an 86% prevalence of FPT in the sample.



in IgG concentration could be explained by a particular assay and by the covariates included in the model, it also reminds us that almost half of the variations remain unexplained. Interestingly, the highest R2 was observed in the model that included ZnSO4 transmittance with no covariate.

The use of mathematical equations to predict IgG concentration on the basis of indirect assays can be useful in a clinical setting, allowing adjustment for covariates and facilitating comparison of the results obtained from various assays. The equations give an estimate on a continuous scale, which is highly informative. The prediction could be done directly by a practitioner using a simple calculator or a program in computerized medical records for a direct result based on input values for various tests. The predicted IgG could then be directly interpreted or classified as adequate or not adequate. Based on a cut-off of 1000 mg/dL, the models developed in this study were very sensitive at predicting FPT, 94% to 100% of the calves with an IgG concentration , 1000 mg/dL being correctly classified. However, the specificity of the models to predict APT was highly variable, ranging from 18% to 45% for the serum protein, plasma protein, and GGT models; it was more satisfactory for ZnSO4 transmittance (73%) and turbidity (82%). Adjustment with covariates did not allow a significant increase in the ability of the models to correctly classify passive transfer. However, the evaluation of the models’ ability to predict APT was based on a very limited population of calves and thus should be interpreted as preliminary. The high sensitivity of the models combined with the high prevalence of FPT (86%) led to very good predictive values, although the negative predictive values were based on a very small sample. In a clinical setting, predictive values are of great interest because they can be interpreted as the probability that a calf with a predicted condition truly has the condi-tion according to the gold standard. These values are influenced by the prevalence of FPT in the population and should be extrapolated only to similar populations or be recomputed for populations with different expected prevalence rates. However, in a clinical setting the proportion of ill calves with FPT is likely to be high, as was observed in this study.

These results suggest that for some assays the inclusion of clinical or laboratory covariates improves the accuracy of IgG-concentration predictions, particularly when the results are interpreted on a con-

tinuous scale. In general, the ZnSO4 transmittance and turbidity assays performed well in predicting the IgG concentration and could be used advantageously in a field setting.

Re f e r e n c e s 1. Beam AL, Lombard JE, Kopral CA, et al. Prevalence of failure

of passive transfer of immunity in newborn heifer calves and associated management practices on US dairy operations. J Dairy Sci 2009;92:3973–3980.

2. Dewell RD, Hungerford LL, Keen JE, et al. Association of neo-natal serum immunoglobulin G1 concentration with health and performance in beef calves. J Am Vet Med Assoc 2006;228: 914–921.

3. Roy JHB. The Calf. 5th ed. Toronto, Ontario: Butterworths, 1990:39–42.

4. Weaver DM, Tyler JW, Van Metre DC, Hostetler DE, Barrington GM. Passive transfer of colostral immunoglobulins in calves. J Vet Intern Med 2000;14:569–577.

5. Tyler JW, Parish SM, Besser TE, Van Metre DC, Barrington GM, Middleton JR. Detection of low serum immunoglobulin concen-trations in clinically ill calves. J Vet Intern Med 1999;13:40–43.

6. Lee SH, Jaekal J, Bae CS, et al. Enzyme-linked immunosorbent assay, single radial immunodiffusion, and indirect methods for the detection of failure of transfer of passive immunity in dairy calves. J Vet Intern Med 2008;22:212–218.

7. Fecteau G, Paré J, Van Metre DC, et al. Use of a clinical sepsis score for predicting bacteremia in neonatal dairy calves on a calf rearing farm. Can Vet J 1997;38:101–104.

8. Fecteau G, Van Metre DC, Paré J, et al. Bacteriological culture of blood from critically ill neonatal calves. Can Vet J 1997;38: 95–100.

9. Dohoo I, Martin W, Stryhn H. Veterinary Epidemiologic Research. 2nd ed. Charlottetown, Prince Edward Island: AVC, 2009:323–360.

10. Wilson LK, Tyler JW, Besser TE, Parish SM, Gant R. Prediction of serum IgG1 concentration in beef calves based on age and serum gamma-glutamyl-transferase activity. J Vet Intern Med 1999;13:123–125.


Article


I n t r o d u c t i o nAirborne transmission poses a major challenge to the control

of human and animal pathogens. For humans, airborne transport has been linked to the transmission of Coccidioides immitis (1); Mycobacterium tuberculosis (2); Legionella spp. (3); smallpox virus (4); and a variety of other pathogenic fungi, bacteria, and viruses (5–8). For animals, some of the most economically significant pathogens are transmitted in bioaerosols, such as foot-and-mouth disease virus (9), classical swine fever virus (10), and porcine respiratory and reproductive syndrome virus (11). Of importance to both human and animal health are major zoonotic pathogens transmitted via aerosols, including influenza virus (12,13), severe acute respira-tory syndrome (SARS) coronavirus (14), Yersinia pestis (6), Bacillus anthracis (15), and others.

In aerobiology, dose-response curves are useful for describing the probability (y-axis) that a specific dose (x-axis) of an airborne pathogen will produce infection in a susceptible host (5,16). Under experimental conditions, dose-response curves can be derived by

individually exposing susceptible animal hosts to a known quantity of pathogen and then monitoring each animal for evidence of infec-tion under conditions that preclude the possibility of infection from all other sources (17,18). The proportion of individuals that become infected at each dose provides the raw data upon which the dose-response curve is based.

A variety of statistical techniques can be used to analyze the dose-response relationship, with ID50, the dose required to infect 50% of the population, being the most useful summary statistic of the dose-response for any defined pathogen-host system (19,20). A standard dose-response curve is defined by 4 parameters: the baseline (bottom), the maximum response (top), the slope of the curve, and the mid-point of the curve. But the exact parameters of a dose-response curve depend on the pathogen (21), the strain or isolate of the pathogen (19), the host species (22), and specific host factors, such as age and immune status (23).

In a research setting, host- and pathogen-specific factors that affect the dose-response curve can be accounted for by careful experi-mental design. A larger challenge is the requirement to determine

A method to quantify infectious airborne pathogens at concentrations below the threshold of quantification by culture

Timothy D. Cutler, Chong Wang, Steven J. Hoff, Jeffrey J. Zimmerman

A b s t r a c tIn aerobiology, dose-response studies are used to estimate the risk of infection to a susceptible host presented by exposure to a specific dose of an airborne pathogen. In the research setting, host- and pathogen-specific factors that affect the dose-response continuum can be accounted for by experimental design, but the requirement to precisely determine the dose of infectious pathogen to which the host was exposed is often challenging. By definition, quantification of viable airborne pathogens is based on the culture of micro-organisms, but some airborne pathogens are transmissible at concentrations below the threshold of quantification by culture. In this paper we present an approach to the calculation of exposure dose at microbiologically unquantifiable levels using an application of the “continuous-stirred tank reactor (CSTR) model” and the validation of this approach using rhodamine B dye as a surrogate for aerosolized microbial pathogens in a dynamic aerosol toroid (DAT).

R é s u m éEn aérobiologie, les études dose-réponse sont utilisées pour estimer le risque d’infection que représente pour un hôte susceptible l’exposition à une dose spécifique d’un agent pathogène en suspension dans l’air. Dans un environnement de recherche, les facteurs spécifiques à l’hôte et à l’agent qui affectent le continuum dose-réponse peuvent être tenus pour compte dans le design expérimental, mais l’obligation de déterminer précisément la dose d’agent pathogène à laquelle l’hôte a été exposée représente souvent un défi. Par définition, la quantification des agents pathogènes viables en suspension dans l’air est basée sur la culture des microorganismes, mais certains agents pathogènes aériens sont transmissibles à des concentrations inférieures au seuil de quantification par culture. Dans cet article nous présentons une approche pour le calcul de la dose d’exposition à des niveaux non-quantifiables microbiologiquement en utilisant une application du modèle de réaction en réservoir avec agitation continue (CSTR) et la validation de cette approche en utilisant le colorant rhodamine B comme substitut à des agents pathogènes microbiens mis en aérosol dans un tore dynamique (DAT).


Department of Veterinary Diagnostic and Production Animal Medicine, College of Veterinary Medicine, Iowa State University 50011-1250, USA (Cutler, Wang, Zimmerman); Department of Agricultural and Biosystems Engineering, College of Agriculture and Life Sciences, Iowa State University, Ames, Iowa 50011-1250, USA (Hoff); Department of Statistics, College of Liberal Arts and Sciences, Iowa State University, Ames, Iowa 50011-1210, USA (Wang).

Address all correspondence to Dr. Zimmerman; telephone: (515) 294-1073; fax: (515) 294-3564; e-mail: [email protected]

Received January 21, 2012. Accepted May 16, 2012.



the dose of infectious pathogen to which the host was exposed. Estimation of the exposure dose requires measurements of the total volume of air respired by the host and the concentration of viable airborne pathogen, for example, liters of air respired by the sus-ceptible host 3 pathogen concentration per liter of air = exposure dose. In domestic animals, total respired air can be measured using spirometric instrumentation (18).

Estimates of the concentrations of infectious airborne pathogens are often more difficult to achieve. By definition, quantification of viable airborne pathogens is based on techniques that require the micro-organism to replicate in culture. Culture-based meth-ods are less analytically sensitive than contemporary molecular techniques, e.g., polymerase chain reaction (PCR), but molecular assays are not a good substitute because they do not differentiate between infectious and non-infectious micro-organisms (5,19). If the pathogen is not highly transmissible, i.e., if transmission requires a large exposure dose, the dose-response curve may be determined despite the requirement to quantify infectious micro-organisms in culture. Commonly, airborne pathogens are transmissible at con-centrations below the threshold of quantification by culture. Under these circumstances, the exposure dose, and hence the probability of transmission, is incalculable (16,24). In this paper, we present an engineering approach to calculate the exposure dose at microbiologi-cally unquantifiable levels.

The continuous-stirred tank reactor (CSTR) is a vessel character-ized by steady-state and uniform internal conditions due to mixing. The reactions and processes occurring within the CSTR may be defined for the conditions of the vessel. For example, as shown in Table I, the concentration of a target within a CSTR can be predicted at any time (t) using the equation:

Ct = Cin (1 2 e2t(Q/V)) = C = Cin (1 2 e2x)

where:

Ct = target concentration at time = tCin = target input concentration at time = 0t = timeQ = flow rate (incoming rate = outgoing rate)V = volume of the CSTRe = the base of natural logarithm (Euler’s number)x = one complete exchange of the volume of the CSTR

Continuous-stirred tank reactors are widely used in a variety of industrial, chemical, and biological applications, including bioreac-tors, fermentation vessels, and wastewater treatment. Goldberg et al (25) introduced the use of a continuously rotating drum (dynamic aerosol toroid) to experimental aerobiology as a method to maintain and study infectious particles suspended in aerosols over time. A rotating dynamic aerosol toroid (DAT) housed in an environment that preserves the pathogen’s infectivity, e.g., held at temperatures below freezing, is a CSTR. As such, the concentrations of the air-borne components within the DAT can be predicted at time (t) if the exchange volumes are known. This paper provides experimental evidence to support this concept and discusses the application of this approach to the problem of estimating the concentration of airborne pathogens at microbiologically unquantifiable levels.

M a t e r i a l s a n d m e t h o d sThe objective of this experiment was to determine whether the

CSTR-derived calculations could accurately predict the concentra-tion of an airborne target in a DAT given known exchange volumes. As a surrogate for an airborne micro-organism, a fluorescent dye (rhodamine B) was aerosolized into a 400 L DAT held at 24°C. In 10 replicates, 12 air samples (200 L each) were collected and the fluorescence measured. These data were tested for: (i) a linear relationship between the concentration of rhodamine B removed (log10 Mc,out(t)) and volume of aerosol removed (Vextracted), and (ii) a significant difference between the theoretical and the observed rhodamine B regression lines.

A stainless steel 400 liter DAT was constructed based on the description provided by Goldberg et al (25). For temperature control, the DAT was housed in a custom-built refrigeration unit (Carroll Coolers Inc., Carroll, Iowa, USA) maintained at 24°C. During opera-tion, the DAT rotated at 4 RPM (Brother International Gearmotors, Bridgewater, New Jersey, USA). Three HEPA-filters (Fisher Scientific, Hampton, New Hampshire, USA) were fixed to ports on the periph-ery of the DAT to allow for pressure equilibration during nebuliza-tion and impingement. The entire system was disassembled and cleaned between each of the 10 replicates.

In each replicate, approximately 50 mL of a 13 phosphate buffered saline (PBS; Thermo Scientific, Rockford, Illinois, USA) solution containing 0.08% v/v; rhodamine B (Sigma Chemical Company, St. Louis, Missouri, USA) and 0.1% v/v Antifoam A Emulsion (Sigma Chemical Company) was nebulized into the DAT using a 24-jet Collison nebulizer (BGI, Waltham, Massachusetts, USA) operating at 40 PSI. According to the manufacturer’s specifications, these parameters aerosolized the solution at a rate of 1.1 mL per min and produced particles 1.9 mm in diameter. After nebulization and prior to sampling, the cloud was allowed to equilibrate within the DAT for 60 min. This allowed for complete mixing, sedimentation, and thermal equilibration of aerosolized rhodamine B.

Air samples were collected using sterile AGI-30 glass imping-ers (Ace Glass, Vineland, New Jersey, USA) containing 20 mL of sterile 13 PBS (Thermo Scientific) as collection fluid. Impingers were operated at a constant flow rate of 12.5 L per min using oil-less pumps (Fisher Scientific, Hampton, New Hampshire, USA). Pump performance was monitored using a vacuum pressure gauge

Table I. Target concentration as a function of the number of complete exchanges of a continuous-stirred tank reactor (CSTR)

Target retainedExchange (x)a Ct /Cin

b in the CSTR (%)1 1 2 e21 37.02 1 2 e22 14.03 1 2 e23 5.04 1 2 e24 1.85 1 2 e25 0.7a x = one complete exchange of the volume of the CSTR.b Mass balance of target concentration.Ct — target concentration at time = t; Cin — target input concentration time at time = 0.



(Cato Western, Tucson, Arizona, USA). Twelve 200-L air samples were taken in succession, i.e., 6 complete evacuations of the DAT over a period of approximately 3.5 h. All samples were maintained on ice until sampling was completed. Thereafter, a 1.5 mL aliquot of each sample was dispensed into a disposable ultraviolet trans-missible cuvet (Fisher Scientific, Pittsburgh, Pennsylvania, USA), allowed to warm to 20°C. The amount of rhodamine B dye in each sample was measured using a fluorometer (Turner BioSystems, Sunnyvale, California, USA) equipped with a green optical kit (Turner BioSystems). Results were expressed as raw fluorescence units. Prior to each replicate, the fluorometer was evaluated using a rhodamine B solid standard (Turner BioSystems).

To predict the concentration of an airborne target (e.g., rhoda-mine B) within a CSTR as samples are drawn from the drum and replaced with filtered inlet air, a mass balance equation of the target’s concentration in the drum can be written as:

d(Mc,drumma) = ma{Mc,in 2 Mc,out} [Equation 1]

dt

where:

Mc,drum = mass fraction of the target inside drum, kgtarget/kga

ma = mass of air inside drum, kga

t = time, sma = mass flow rate of air through drum, kga/sMc,in = mass fraction of target entering drum, kgtarget/kga

Mc,out = mass fraction of target leaving drum, kgtarget/kga

Based on the work of Goldberg et al (25), a DAT is a well-mixed ves-sel. Therefore, the mass fraction of target leaving the drum is repre-sentative of the mass fraction inside the drum and can be stated as:

Mc,drum = Mc,out [Equation 2]

and Equation 1 can be re-written as:

d(Mc,outMa) = ma{Mc,in 2 Mc,out} [Equation 3]

dt

Integrating Equation 3 results in the common form of a perfectly mixed, but dynamically changing, mass fraction starting from a known initial mass fraction as:

Mc,out(t) = Mc,out(t = 0)e2(t ma/ma) [Equation 4]

Equation 4 states that the mass fraction of target in a perfectly mixed drum at any time (t) (Mc,out(t)) is a function of the initial concentration inside the (Mc,out(t = 0)) drum and the exponential decay character-ized by the mass of air inside the drum (ma) and the mass flow rate of air through the drum (ma). Assuming constant air density, equation (4) can be further described by:

Mc,out(t) = Mc,out(t = 0)e2(Vextracted/Vdrum) [Equation 5]

The starting point for determining the concentration of the target at time (t) is the initial mass fraction within the drum (Mc,out(t = 0)). Determination of target concentration requires sampling the air and subsequent sample analysis. This extraction process results in an interruption of the initial mass fraction of the target.

Equation 5 can be used to back-calculate the initial target mass fraction (t = 0) within the drum from the initial extracted sample

(t = 1). Thus, after the first sample extraction, some known amount of drum air has been extracted (Vextracted) resulting in a mass fraction of (Mc,out(t1)) providing an estimation of the initial mass fraction determined as:

Mc,out(t1) Mc,out(t = 0) = e2(Vextracted,t1

/Vdrum) [Equation 6]

Equation 6 represents the initial mass fraction inside the drum. Since the drum behaves as a well-mixed vessel, all subsequent sample extractions and the resulting mass fractions will obey the mixing model as given in equation (5) using the estimate for the initial mass fraction given in equation (6). The final relationship becomes:

Mc,out(t1) Mc,out(t) =

e2(Vextracted,t1/Vdrum) e

2(Vextracted/Vdrum) [Equation 7]

where:

Mc,out(t1) = mass fraction from the first sampled extraction, kgtarget/kga

Vextracted,t1 = volume of drum air extracted for the first sample, liters

Vdrum = fixed volume of the drum, liters

Converting Equation 5 to log10 format, the mass fraction of the DAT can be mathematically represented as:

log10 Mc,out(t) = log10 Mc,out(t=0) 2 (log10 e/Vdrum)*Vextracted

where:

Mc,out(t=0) = concentration of rhodamine B at (t = 0)Mc,out(t) = rhodamine B concentration at current timeVextracted = the running total of the volume removedVdrum = the total volume in containere = the base of natural logarithm (Euler’s number).

This mathematical representation contained 2 assumptions:

1. There is a linear relationship between (log10 Mc,out(t)) and Vextracted;2. The slope of the rhodamine B regression line was equal to

2(log10 e/Vdrum).

If both of these assumptions are true, then, log10 Mc,out(t) is a lin-ear function of Vextracted with intercept log10 Mc,out(t=0) and slope 2(log10 e/Vdrum). Thus, the linear relationship between log10 Mc,out(t) and Vextracted may be used to estimate the concentration of rhodamine B at any time along the regression line. To validate assumption (1), the concentration of rhodamine B (log10) in sequential air samples collected in each of 9 replicates was analyzed using a simple linear regression model using the REG procedure (SAS, version 9.2; SAS Institute Inc, Cary, North Carolina, USA) and the coefficient of determination (R2) was calculated. To validate assumption (2), the hypothesis that the average slope was equal to the theoretical slope (log10 e/Vdrum) was tested using the Student’s t-test.

Re s u l t sA total of 10 replicates were attempted. One replicate (number 4)

failed because of technical problems that occurred during the procedure. For the remaining 9 runs, least square estimates of the intercept and slope, as well as the coefficient of determination (R2),



were calculated for each regression line (Table II). R2 described the proportion of response variation explained for by the linear model and ranges from 0 to 1, with a large R2 value indicative of a good fit of the linear model. The mean R2 for the 9 replicates was 0.93 with a standard deviation of 0.07. Overall, the linear regression line explained 93% of the variation in log10 transformed rhodamine B data. The average slope of the 9 runs was not significantly different (P = 0.1593) from the theoretical slope (20.0011). Thus, these data show that the linear relationship between log10 Mc,out(t) and Vextracted may be used to estimate the concentration of an airborne target in a DAT given known exchange volumes.

D i s c u s s i o nSuccessful airborne transmission occurs in 3 basic steps: (i) aero-

solization of the infectious agent; (ii) environmentally dependent movement, dilution, and inactivation of airborne infectious particles; and (iii) contact, entry, and replication within a susceptible host (26). From the perspective of prevention and control, the goal is to under-stand and model the transmission of airborne pathogens in order to design effective counter-measures. Both macro- and micro-level approaches are useful in meeting this objective. That is, field data collected over the course of an outbreak may be useful for modeling the airborne spread of a pathogen within a population (27). Likewise, the basic steps and their components may be evaluated indepen-dently under controlled conditions to understand the contribution of each to the process of transmission (18). That is: (i) quantify the rate at which the pathogen is excreted into the environment; (ii) measure the rate of inactivation of the airborne infectious pathogen under specific environmental conditions; and (iii) estimate the likelihood that exposure to a specific dose of the airborne infectious pathogen will produce a response (infection) in an individual host.

This study addressed the third step in this process and, in par-ticular, the specific problem of deriving dose-response curves under experimental conditions in which transmission occurs at concentrations below the threshold of quantification for culture-based methods. In this experiment, a tracer was used to model the behavior of an aerosolized pathogen in a rotating DAT. Tracers (e.g., uranine, rhodamine B, and Bacillus subtilis spores) have been used extensively in experimental aerobiology (28). Songer (29) aerosol-ized rhodamine B dye simultaneously with virus (Newcastle disease virus, infectious bovine rhinotracheitis virus, vesicular stomatitis virus, T3 bacteriophage) to track the physical loss of airborne virus within a DAT. In an experiment of similar design, Hermann et al (30) found no significant difference between the slopes of rhodamine B dye and porcine reproductive and respiratory syndrome virus RNA detected by quantitative PCR, i.e., the concentrations of rhodamine B and viral RNA declined in the DAT at the same rate. Under the con-ditions of this experiment, the fact that the theoretical line and the experimental line were not significantly different provided evidence that physical loss did affect the outcome of the tracer values. Thus, rhodamine B concentration has been shown to reflect target patho-gen concentrations under conditions similar to those reported here.

The physical parameters and experimental conditions of this study merit discussion. This experiment was conducted in a 400 L DAT rotated at 4 RPM. However, a variety of DAT sizes and rotation

speeds are reported in the literature, for example, 140 L (29), 1000 L (31), and 2500 L (32). A review of the literature found no evaluation of the effect of DAT dimensions, volume, and rate of rotation on the behavior of suspended particles. Therefore, it would be of value to confirm the results reported here using the described methodology.

In this experiment, the environmental conditions were designed to preserve target pathogen infectivity. In particular, the 24°C temperature at which the DAT was maintained would be expected to preserve the infectivity of a target pathogen indefinitely. At tem-peratures above freezing, the slope of the airborne pathogen would diverge from the slope of the rhodamine B. Therefore, the inactiva-tion of the target pathogen over time would need to be accounted for in the estimation of the airborne pathogen concentration. This is not an insignificant consideration because the rate of airborne pathogen concentration inactivation is affected by isolate (26,27), the suspension medium (32,33), temperature (29,34), and relative humidity (30,35). Therefore, it is preferable to avoid this complication by maintaining the DAT at temperatures below 0°C.

In dose-response studies, the CSTR model solves the problem of estimating the exposure dose when the concentrations of airborne pathogens are at microbiologically unquantifiable levels.

Specifically, the linear relationship between log10 Mc,out(t) and Vextracted may be used to estimate the concentration of a target at any point along the regression line. Thus, log10 Mc,out(t) is a linear function of Vextracted with intercept log10 Mc,out(t=0) and slope 2(log10 e/Vdrum). Therefore, under conditions similar to those reported here, the CSTR model solves the problem of estimating the exposure dose when the concentrations of airborne pathogens are at microbiologically unquantifiable levels.

A c k n o w l e d g m e n t sThe study was supported in part by Pork Checkoff funds distrib-

uted through the National Pork Board, Des Moines, Iowa, USA and the PRRS CAP USDA NIFA Award 2008-55620-19132.

Re f e r e n c e s 1. CDC (Centers for Disease Control and Prevention), 2009. Increase

in coccidioidomycosis — California, 2000–2007. MMWR Morb Mortal Wkly Rep 2009;58:105–109.

Table II. Parameters describing the linear relationship between the concentration of airborne rhodamine B and the volume of air extracted from a dynamic aerosol toroid

Replicate R2 Slope Intercept 1 0.96 20.0013 3.47 2 0.83 20.0021 3.40 3 0.96 20.0030 3.16 5 0.94 20.0007 3.20 6 0.95 20.0011 4.07 7 0.80 20.0007 2.56 8 0.97 20.0012 4.08 9 0.99 20.0013 4.7210 0.98 20.0016 4.70



2. de la Rua-Domenech R. Human Mycobacterium bovis infection in the United Kingdom: Incidence, risks, control measures and review of the zoonotic aspects of bovine tuberculosis. Tuberculosis (Edinb) 2006;86:77–109.

3. Diedern BMW. Legionella spp. and Legionnaries’ disease. J Infect 2007;56:1–12.

4. Feigel J, Clarke RC, Edwards DA. Airborne infectious disease and the suppression of pulmonary bioaerosols. Drug Discov Today 2006;11:51–57.

5. Douwes J, Thorne Pl, Pearce N, Heederik L. Bioaerosol health effects and exposure assessment: Progress and prospects. Br Occup Hyg Soc 2003;47:187–200.

6. Nicas M, Nazaroff WW, Hubbard A. Toward understanding the risk of secondary airborne infection: Emission of respirable pathogens. J Occup Environ Hyg 2005;2:134–154.

7. Farnsworth JE, Goyal SM, Kim SW, et al. Development of a method for bacteria and virus recovery from heating, ventila-tion, and air conditioning (HVAC) filters. J Environ Monit 2006; 8:1006–1013.

8. Tang JW, Eames I, Chan PK, Ridgway GL. Factors involved in the aerosol transmission of infection and control of ventilation in healthcare premises. J Hosp Infect 2006;64:100–114.

9. Alexandersen S, Brotherhood I, Donaldson AI. Natural aerosol transmission of foot-and-mouth disease virus to pigs: Minimal infectious dose for strain 01 Lausanne. Epidemiol Infect 2002;128: 301–312.

10. Weesendorp E, Landman WJM, Stegeman A, Loeffen WLA. Detection and quantification of classical swine fever virus in air samples originating from infected pigs and experimentally produce aerosols. Vet Microbiol 2008;127:50–62.

11. Dee SA, Deen J, Cano JP, Batista L, Pijoan C. Further evaluation of alternative air-filtration systems for reducing the transmis-sion of porcine reproductive and respiratory syndrome virus by aerosol. Can J Vet Res 2006;70:168–175.

12. Loosli C, Lemon H, Robertson O, Appel E. Experimental airborne influenza infection. 1. Influence of humidity on survival of virus in air. Proc Soc Exp Biol Med 1943;53:205–206.

13. Wong S, Yuen K. Avian influenza virus infections in humans. Chest 2006;129:156–168.

14. Booth TF, Kournikakis B, Bastien N, et al. Detection of airborne severe acute respiratory syndrome (SARS) coronavirus and environmental contamination in SARS outbreak units. J Infect Dis 2005;191:1472–1477.

15. Inglesby TV, Henderson DA, Bartlett JG, et al. Anthrax as a bio-logical weapon: Medical and public health management. J Am Med Assoc 1999;281:1735–1745.

16. Pillai SD, Ricke SC. Bioaerosols from municipal and animal wastes: Background and contemporary issues. Can J Microbiol 2002;48:681–696.

17. French NP, Kelly L, Jones R, Clancy D. Dose-response relation-ships for foot and mouth disease in cattle and sheep. Epidemiol Infect 2002;128:325–332.

18. Hermann JR, Muñoz-Zanzi CA, Zimmerman JJ. A method to provide improved dose-response estimates for airborne patho-gens in animals: An example using porcine reproductive and respiratory syndrome virus. Vet Microbiol 2009;133:297–302.

19. Ward RL, Akin EL. Minimum infective dose of animal viruses. Crit Rev Environ Control 1984;14:297–310.

20. Spouge JL. Statistical analysis of sparse infection data and its implications for retroviral treatment trials in primates. Proc Natl Acad Sci U S A 1992;89:7581–7585.

21. Liao CM, Chang CF, Liang HM. A probabilistic transmission dynamic model to assess indoor airborne infection risks. Risk Anal 2005;25:1097–1107.

22. Thurston-Enriquez JA, Haas CN, Jacangelo J, Riley K, Gerba CP. Inactivation of feline calicivirus and adenovirus type 40 by UV radiation. Appl Environ Microbiol 2003;69:577–582.

23. Jani JV, Holm-Hansen C, Mussa T, et al. Assessment of measles immunity among infants in Maputo City, Mozambique. BMC. Public Health 2008;8:386–396.

24. Gillespie RR, Hill MA, Kanitz CL. Infection of pigs by aerosols of Aujeszky’s disease virus and their shedding of the virus. Res Vet Sci 1996;60:228–233.

25. Goldberg LJ, Watkins HMS, Boerke EE, Chatigny MA. The use of a rotating drum for the study of aerosols over extended periods of time. Am J Hyg 1958;68:86–93.

26. Stärk KDC. The role of infectious aerosols in disease transmission in pigs. Vet J 1999;158:164–181.

27. Keeling MJ, Woolhouse ME, Shaw DJ, et al. Dynamics of the 2001 UK foot and mouth epidemic: Stochastic dispersal in a heterogeneous landscape. Science 2001;294:813–817.

28. Verreault D, Moineau S, Duchaine C. Methods for sampling of airborne viruses. Microbiol Mol Biol Rev 2008;72:413–444.

29. Songer JR. Influence of relative humidity on the survival of some airborne viruses. Appl Microbiol 1967;15:35–42.

30. Hermann JR, Hoff S, Muñoz-Zanzi C, et al. Effect of tempera-ture and relative humidity on the stability of infectious porcine reproductive and respiratory syndrome virus in aerosols. Vet Res 2007;38:81–83.

31. Adams DJ, Spendlove JC, Spendlove RS, Barnett BB. Aerosol sta-bility of infectious and potentially infectious reovirus particles. Appl Environ Microbiol 1982;44:903–908.

32. Ehrlich R, Miller S, Idoine LS. Effects of environmental fac-tors on the survival of airborne T-3 coliphage. Appl Microbiol 1964;12:479–482.

33. Benbough JS. Some factors affecting the survival of airborne viruses. J Gen Virol 1971;10:209–220.

34. Elazhary MA, Derbyshire JB. Effect of temperature, relative humidity and medium on the aerosol stability of infectious bovine rhinotracheitis virus. Can J Comp Med 1979;41:158–167.

35. Sattar SA, Ijaz MK, Johnson-Lussenburg CM, Springthorpe VS. Effect of relative humidity on the airborne survival of rotavirus SA11. Appl Environ Microbiol 1984;47:879–881.


Article


I n t r o d u c t i o nAnthrax is a highly lethal infectious disease caused by the

endospore- forming bacterium Bacillus anthracis, which is a Gram-positive soil organism commonly found in nature. Although anthrax can affect all mammals including humans, it is primarily a disease of herbivores, such as cattle, sheep, horses, pigs, goats, and camels, with hyperacute or acute symptoms and usually with a fatal out-come. Anthrax is not transmitted from sick to healthy animals, but is generally acquired by the ingestion of spores dispersed into the environment (1). Flooding, drought, and other natural or man-made disturbances can bring the spores up to the soil surface where graz-

ing animals are at risk of exposure to the organism. The disease is characterized by outbreaks, usually involving a small number of animals, but may at times turn into an epidemic with serious con-sequences. According to reports by the Canadian Food Inspection Agency, anthrax outbreaks occur sporadically in the Canadian prairie provinces, often during the summer and sometimes in the colder months, and can affect hundreds of animals (2).

It is expected that exposure to B. anthracis could be reduced by proper disposal of infected animal carcasses and wastes and by cleaning and disinfecting contaminated surfaces. Fumigation with formaldehyde or hydrogen peroxide has been used for emergency decontamination of indoor surfaces in buildings at temperatures

Influence of temperature and organic load on chemical disinfection of Geobacillus steareothermophilus spores, a surrogate

for Bacillus anthracisJiewen Guan, Maria Chan, Brian W. Brooks, Liz Rohonczy

A b s t r a c tThis study evaluated the influence of temperature and organic load on the effectiveness of domestic bleach (DB), Surface Decontamination Foam (SDF), and Virkon in inactivating Geobacillus stearothermophilus spores, which are a surrogate for Bacillus anthracis spores. The spores were suspended in light or heavy organic preparations and the suspension was applied to stainless steel carrier disks. The dried spore inoculum was covered with the disinfectants and the disks were then incubated at various temperatures. At 220°C, the 3 disinfectants caused less than a 2.0 log10 reduction of spores in both organic preparations during a 24-h test period. At 4°C, the DB caused a 4.4 log10 reduction of spores in light organic preparations within 2 h, which was about 3 log10 higher than what was achieved with SDF or Virkon. In heavy organic preparations, after 24 h at 4°C the SDF had reduced the spore count by 4.5 log10, which was about 2 log10 higher than for DB or Virkon. In general, the disinfectants were most effective at 23°C but a 24-h contact time was required for SDF and Virkon to reduce spore counts in both organic preparations by at least 5.5 log10. Comparable disinfecting activity with DB only occurred with the light organic load. In summary, at temperatures as low as 4°C, DB was the most effective disinfectant, inactivating spores within 2 h on surfaces with a light organic load, whereas SDF produced the greatest reduction of spores within 24 h on surfaces with a heavy organic load.

R é s u m éCette étude a permis d’évaluer l’influence de la température et de la charge organique sur l’efficacité de javellisant domestique (DB), de mousse de décontamination de surface (SDF) et de Virkon pour inactiver les spores de Geobacillus stearothermophilus, un substitut pour les spores de Bacillus anthracis. Les spores ont été suspendues dans des préparations organiques légères ou denses et la suspension étaient appliquées sur des disques d’acier inoxydable. L’inoculum séché de spores était recouvert avec les désinfectants et les disques étaient ensuite incubés à différentes températures. À 220 °C les trois désinfectants ont entrainé une réduction de moins de 2 log10 du nombre de spores dans les deux préparations organiques durant une période d’essai de 24 h. À 4 °C, le DB a causé, en dedans de 2 h, une réduction de 4,4 log10 de la quantité de spores dans les préparations organiques légères, à peu près 3 log10 plus élevé que ce qui a été atteint par la SDF ou le Virkon. Dans les préparations organiques denses, après 24 h à 4 °C la SDF avait réduit le dénombrement de spores par 4,5 log10, ce qui était à peu près 2 log10 plus élevé que ce qui a été obtenu avec le DB ou le Virkon. En général, les désinfectants étaient les plus efficaces à 23 °C mais un temps de contact de 24 h était requis pour la SDF et le Virkon pour réduire le nombre de spores dans les deux préparations organiques par au moins 5,5 log10. Une activité désinfectante comparable avec le DB n’a été observée qu’avec une charge organique légère. En résumé, à des températures aussi basse que 4 °C, le DB était le désinfectant le plus efficace inactivant les spores en moins de 2 heures sur des surfaces avec des charges organiques légères, alors que la SDF a causé la plus grande réduction de spores en-dedans de 24 h sur des surfaces avec une charge organique élevée.


Ottawa Laboratory (Fallowfield), Canadian Food Inspection Agency, Ottawa, Ontario K2H 8P9.

Address all correspondence to Jiewen Guan; telephone: (613) 228-6698; fax: (613) 228-6670; e-mail: [email protected]

Received May 18, 2012. Accepted June 26, 2012.



around 20°C (3,4). At similar temperatures, sodium hypochlorite, hydrogen peroxide, peracetic acid, and chlorine dioxide were found to be effective for routine surface decontamination in a hospital or a laboratory setting (5,6). In addition, most of these chemicals are also recommended for surface decontamination of fomites such as farm equipment, surgical instruments, and vehicles in an agricultural setting at temperatures above 10°C (7). However, as temperatures in Canada and many other countries are frequently below 10°C, more information is needed on the effectiveness of chemical disinfectants at colder temperatures.

Since the 2001 anthrax bioterrorist attacks in the United States, Geobacillus stearothermophilus spores have been used as a surrogate for B. anthracis spores to assess the effectiveness of disinfectants and decontamination processes (3,4,8). Since G. stearothermophilus spores are heat-resistant, they are also used as biological indicators of the effectiveness of heat sterilization processes (9).

The objective of this study was to evaluate the effectiveness of 3 chemical disinfectants in killing G. stearothermophilus spores at temperatures of 220°C, 4°C, 10°C, and 23°C. Domestic bleach has been recommended for inactivation of B. anthracis spores by the US Environmental Protection Agency and the US Centers for Disease Control (5). Virkon has been widely used in an effort to kill infectious agents in various environments (10–12). Surface Decontamination Foam (SDF) was recently developed for inactivating chemical and biological agents in a wide range of field environments (13,14).


Carrier disksThe second tier quantitative carrier test (15) was used for evalu-

ating the sporicidal activity of the 3 disinfectants. Disks (1 cm in diameter; 0.75 mm thick) of brushed stainless steel (AISI No. 430; Muzeen & Blythe, Winnipeg, Manitoba) were used as carriers. They were washed 3 times with distilled water and dried at 60°C for 1 h. The dried disks were sterilized at 121°C for 25 min before use.

Spore inoculumA suspension of G. stearothermophilus American Type Culture

Collection (ATCC) strain 7953 spores (Spordex) was obtained from a commercial source (Steris, Mentor, Ohio, USA). The suspension was heated at 100°C for 35 min to inactivate vegetative cells and sonicated for 5 min to break up clumps. The treated suspension was mixed with an organic preparation and the mixture containing 4.6 3 106 colony-forming units (CFUs) of spores was loaded onto a carrier disk as inoculum. There were 2 types of organic preparations: one was used to simulate a light organic challenge (equivalent to 5% to 10% serum (15) on relatively clean surfaces and the other to simulate a heavy organic challenge (light organic preparation plus 5% garden soil) on relatively dirty surfaces. The light organic prepa-ration was a peptide and protein mixture (16) that contained 0.35% weight/volume (w/v) tryptone (Sigma, Oakville, Ontario), 0.25% w/v bovine serum albumin (Sigma), and 0.04% w/v mucin (Sigma) in 0.01 M phosphate-buffered saline (PBS, pH 7.2). The heavy organic preparation was a mixture of the light organic preparation and 5% w/v garden soil (Premium Nature Mix; Modugno-Hortibec,

St. Isidore, Quebec). The garden soil was pre-sterilized at 121°C for 90 min to prevent the introduction of other live microorganisms into the tests.

DisinfectantsStandard hard water was used to prepare disinfectant solutions

to avoid variations in results that may derive from differences in tap water quality. The water was prepared in accordance with AOAC 960.09 (17) to a standard hardness of 400 ppm as calcium carbonate. Solutions of the 3 disinfectants were prepared as follows. Domestic bleach (Clorox; Oakland, California, USA) containing about 5.25% sodium hypochlorite (52 500 ppm available chlorine) was diluted to obtain a chlorine concentration of 5 250 ppm for testing. Surface Decontamination Foam (SDF) (Allen Vanguard, Ottawa, Ontario) includes 3 separate reagents: GPA-2100 decontaminant (A); GPB-2100 buffer (B); and GCE-2000 surfactant (C). The SDF solution was prepared according to the manufacturer’s instructions: i) 1.8 g B and 4.5 g C were dissolved in 150 mL of water; ii) 7.8 g A was dissolved in 50 mL of water; and iii) the 2 solutions were mixed immediately before use. The 2% Virkon solution (Antec, Suffolk, United Kingdom) was prepared by dissolving 4 tablets in 1.0 L of water. For tests at 220°C, propylene glycol [1,2-propanediol, Sigma, 40% volume/volume (v/v) final concentration] was added to the disinfectant solutions as an antifreeze agent.

NeutralizerA neutralizer was used to immediately stop the activity of the dis-

infectants at the end of a test period in order to provide an accurate contact time. The neutralizer solution contained 1.56 g of sodium thiosulfate (Na2S2O2 · 5H2O; Thermo Fisher Scientific, Ottawa, Ontario), 0.07 g of lecithin (Thermo Fisher Scientific), and 0.1 mL of Tween 80 (Thermo Fischer Scientific) in 100 mL of PBS and was sterilized at 121°C for 15 min before use.

Test procedureContact time course experiments were carried out to evaluate the

disinfectants. Six time points were included for each experiment: 5 min, 15 min, 30 min, 1 h, 2 h, and 24 h. Duplicate sample disks and duplicate control disks were prepared for each time point. Ten microliters of the spore inoculum was applied to the surface of each disk and the discs were air dried in a biosafety cabinet for 1 h. Each disk, with the inoculum side up, was then placed in a 30-mL Nalgene polypropylene straight-side vial (Thermo Fisher Scientific). Disinfectant solution (50 μL) was added to each test disk to cover the dried inoculumn and 50 μL of PBS was added to each control disk. Vials containing the disks were incubated at 220°C, 4°C, 10°C, or 23°C for specific periods up to 24 h to assess the activity of bleach and SDF and at 220°C, 4°C, and 23°C for periods up to 24 h for Virkon. At the end of each contact time, 9.95 mL of neutralizer solu-tion was immediately added to 2 vials with test disks and 2 vials with control disks to stop the activity of the disinfectants and the vials and contents were vortexed for 1 min. A 10-fold serial dilution was made of the suspension in each vial. Each of the serial dilutions was passed through a 0.2-μm membrane filter (Supor 200 membrane filter; Pall, Ann Arbor, Michigan, USA) in a magnetic filter holder (Pall). The filter unit was subsequently rinsed 3 times with 10 mL of



PBS to maximize spore recovery. The membrane filter was removed from the unit and placed on the surface of a tryptic soy agar (TSA; Voigt Global Distribution, Lawrence, Kansas, USA) plate and incu-bated at 56°C in a relative humidity of approximately 60%. After incubation for 2 d and 5 d, colonies were counted.

Statistical analysisDuplicate contact time course experiments were conducted to

evaluate each disinfectant and duplicate sets of test and control disks were used for each time point. The difference in recovery of

live spores from control and test disks was recorded as spore reduc-tion (log10) at each time point to indicate the sporicidal efficiency of disinfectants for a specific contact time. The data presented were the means of spore reduction from duplicate sets of control and test disks in duplicate experiments. Student’s t-test was used to determine statistical significance (P , 0.05) in difference in spore reduction among various temperatures with 1 disinfectant solution or among various disinfectant solutions at 1 tempera-ture. The comparison was done with either light or heavy organic preparation.

Figure 1. Effects of disinfectants on reduction of the spores of Geobacillus stearothermophilus at temperatures of 220°C ( and ), 4°C ( and ), 10°C ( and ), and 23°C ( and ). The spores in the light (equivalent to 5 ~ 10% serum, open symbols) and the heavy (equivalent to 5 ~ 10% serum plus 5% garden soil, solid symbols) organic prepa-rations were treated with domestic bleach (A and B), SDF (C and D), and Virkon (E and F) at contact times of 5 min, 15 min, 30 min, 1 h, 2 h, and 24 h. The data were the mean difference of the spore counts recovered from duplicate sets of the control and the sample disks in duplicate experiments and the standard deviations were less than 1.2 log10 colony-forming units (CFUs). The recovery of spores from untreated control disks ranged from 5.5 to 5.7 log10.



Re s u l t sThe recovery of G. stearothermophilus spores from control disks

ranged from 5.5 to 5.7 log10 (data not shown) and was not influenced by time, temperature, or exposure to organic matter. The neutral-izer solution had no effect on the survival of the spores (data not shown), but the effectiveness of all disinfectants was influenced by the treatment variables.

The propylene glycol that was added to the disinfectant solutions prevented freezing during the 24-h test period at 220°C, but under these conditions, the disinfectants produced less than a 2.0 log10 reduction of spores. With the light organic load, after a contact period of 2 h at 4°C or 10°C, the DB had reduced spore counts by 4.4 and 4.7 log10, respectively. These reductions were about 3 log10 higher than those produced by SDF or Virkon (4°C only) and the differences were significant (P , 0.05). After 24 h at 4°C or 10°C, both DB and SDF had reduced spore counts by more than 5.0 log10. Virkon was not tested at 10°C but spore reduction after 24 h at 4°C was about 2 log10 lower than for the other disinfectants (Figures 1, A, C, and E).

With the heavy organic load, the disinfectants reduced spore counts by only less than 2 log10 within 2 h at either 4°C or 10°C. After 24 h at these temperatures, however, the SDF had reduced spore counts by at least 4.5 log10. These reductions were significantly higher (P , 0.05) than those produced by DB or Virkon (Figures 1, B, D, and F).

With both the light and heavy organic load, the disinfectants were usually most effective at 23°C but contact for at least 24 h was required for SDF and Virkon to kill at least 5.5 log10 of spores. The DB was comparable in its effectiveness with the light organic load, but not with the heavy load. It was concluded that at temperatures as low as 4°C, DB was the most effective disinfectant for inactivating spores within 2 h on surfaces with a light organic load. In compari-son, SDF produced the greatest reduction of spores within 24 h on surfaces with a heavy organic load.

D i s c u s s i o nThe spores of G. stearothermophilus, Bacillus atrophaeus, Bacillus

cereus, and Bacillus subtilis have all been used as surrogates for spores of B. anthracis in previous studies (3,4,8,18). Geobacillus stearothermophilus was selected for use in this study because its spores have high heat resistance that aided in separating them from other microorganisms. It is also possible that the G. stearothermophilus spores are more resistant to oxidizing chemicals than are Bacillus surrogates (3,8). This may account for the fact that the disinfectants used in this study had to be in contact with the preparations of spores for up to 24 h in order to inactivate approximately 6 log10 of G. stearothermophilus.

The laboratory study described here was designed to simulate the disinfection that would be required to inactivate B. anthracis spores on the surfaces of farm equipment under field conditions. For this reason, both light and heavy organic challenges were included for evaluating the disinfectants. The light organic preparation, which is equivalent to 5% to 10% serum, has been widely used in evaluating disinfectants in medical settings and other environments (16). Such

organic loads may be found on pre-cleaned farm equipment and veterinary tools. The heavy organic preparation, which is equivalent to 5% to 10% serum plus 5% sterilized garden soil, was intended to provide more stringent challenges to disinfection, such as may be encountered on inadequately cleaned tractors, trucks, and other equipment. As discussed in other studies, organic matter may reduce the effectiveness of disinfectants by interacting with the active ingre-dients and by forming a physical barrier that could hinder contact between disinfectants and microorganisms (19).

In the present study, DB was the most effective disinfectant against spores in the light organic preparations, in which it killed at least 4 logs of spores within 2 h at 4°C, 10°C, or 23°C. These findings are in agreement with a report by Best, Springthorpe, and Sattar (20). However, DB was not effective against spores in the heavy organic preparation at any of these temperatures. The effectiveness of SDF in both light and heavy organic preparations may be attributed to its surfactant ingredients. In agreement with earlier reports (10–12), the sporicidal effect of Virkon was compromised by organic loads.

Environmental temperature varies widely in many countries and in this study, the disinfectants applied to solid surfaces at 220°C reduced G. stearothermophilus spore counts by only approximately 2 log10. These results differed substantially from those of other studies (21,22), which reported that disinfectants killed at least 6 log10 of B. subtilis spores in suspension tests at temperatures from 0 to ~ 240°C. The disinfectants used in those studies were solutions made of sodium hypochlorite, peracetic acid, or ß-propiolactone. Although tests on solid surfaces or suspensions have been officially accepted by international organiza-tions for evaluating sporicidal disinfectants (23), a much longer contact time was required to inactivate anthrax spores by alcoholic peracetic acid on a solid surface than in a suspension (24). It is therefore evident that test results could be influenced by many factors, including the type of spores used as surrogates, the type of disinfectants, tempera-ture, and the nature of the tests.

For this study, the tests were carried out on solid surfaces in order to simulate conditions that would exist on farms. It is evi-dent, however, that cleaning and disinfection could only reduce the levels of B. anthracis spores in the environment since farm equipment is often difficult to clean and spores washed from the equipment could survive in the soil and be picked up by boots and tires. Nevertheless, this study suggests that DB and SDF, applied at temperatures ranging from 4°C to 23°C, could substantially reduce the load of B. anthracis counts on prewashed farm equipment and thereby reduce the chances for spread of the disease.

A c k n o w l e d g m e n t sThis study is a part of the CRTI 08-0122TD project entitled

“Verification of Decontamination Processes in the AgriFood Context,” which is funded by the Canadian Chemical, Biological, Radiological, and Nuclear (CBRNE) Research and Technology Initiative (CRTI). The authors thank Dr. S.A. Sattar of the University of Ottawa in Ottawa, Ontario for introducing the second tier quanti-tative carrier test method to assess the sporicidal activity of chemical disinfectants and Dr. J.L. Spencer, emeritus research scientist at the Canadian Food Inspection Agency in Ottawa, Ontario, for discussing and reviewing the manuscript.



Re f e r e n c e s1. Fasanella A, Galante D, Garofolo G, Jones MH. Anthrax under-

valued zoonosis. Vet Microbiol 2010;140:318–331.2. Epp T, Argue C, Waldner C, Berke O. Spatial analysis of an

anthrax outbreak in Saskatchewan, 2006. Can Vet J 2010;51: 743–748.

3. Rogers JV, Sabourin CLK, Choi YW, et al. Decontamination assessment of Bacillus anthracis, Bacillus subtilis, and Geobacillus stearothermophilus spores on indoor surfaces using a hydrogen peroxide gas generator. J Appl Microbiol 2005;99:739–748.

4. Rogers JV, Choi YW, Richter WR, et al. Formaldehyde gas inac-tivation of Bacillus anthracis, Bacillus subtilis, and Geobacillus stearothermophilus spores on indoor surface materials. J Appl Microbiol 2007;103:1104–1112.

5. Heninger SJ, Anderson CA, Beltz G, Onderdonk AB. Decon-tamination of Bacillus anthracis spores: Evaluation of various disinfectants. Appl Biosaf 2009;14:7–10.

6. Chatuev BA, Peterson JW. Analysis of the sporicidal activity of chlorine dioxide disinfectant against Bacillus anthracis (Sterne strain). J Hosp Infect 2010;74:178–183.

7. Guidelines for the Surveillance and Control of Anthrax in Humans and Animals. World Health Organization (WHO) [homepage on the Internet]. 3rd edition 2003. Available from: http://www.who.int/csr/resources/publications/anthrax/whoemczdi986text.pdf Last accessed on December 5, 2012.

8. Sabbah S, Springthorpe S, Sattar SA. Use of a mixture of surro-gates for infectious bioagents in a standard approach to assessing disinfection of environmental surfaces. Appl Environ Microbiol 2010;76:6020–6022.

9. Watanabe T, Turukawa S, Hirata J, Koyama T, Ogihara H, Yamasaki M. Inactivation of Geobacillus stearothermophilus spores by high-pressure carbon dioxide treatment. Appl Environ Microbiol 2003;69:7124–7129.

10. Hernndez A, Martró E, Matas L, Martin M, Ausina V. Assessment of in-vitro efficacy of 1% Virkon against bacteria, fungi, viruses and spores by means of AFNOR guidelines. J Hosp Infect 2000;46:203–209.

11. Angelillo IF, Bianco A, Nobile CGA, Pavia M. Evaluation of the efficacy of glutaraldehyde and peroxygen for disinfection of dental instruments. Lett Appl Microbiol 1998;27:292–296.

12. Coates D. Sporicidal activity of sodium dichloroisocyanurate, peroxygen and glutaraldehyde disinfectants against Bacillus subtilis. J Hosp Infect 1996;32:283–294.

13. Love AH, Bailey CG, Hanna ML, et al. Efficacy of liquid and foam decontamination technologies for chemical warfare agents on indoor surfaces. J Hazard Mater 2011;196:115–122.

14. Technology Information Summary on Surface Decontamination Foam. U.S. EPA [homepage on the Internet]. EPA, 2009. Available from: http://www.epa.gov/nhsrc/pubs/TISSurfaceDecontami nationFoam.pdf Last accessed on December 5, 2012.

15. Sattar SA, Springthorpe VS, Adegbunrin O, Zafer AA, Busa M. A disc-based quantitative carrier test method to assess the viru-cidal activity of chemical germicides. J. Virol Methods 2003;112: 3–12.

16. ASTM International. Standard Quantitative Disk Carrier Test Method for Determining the Bactericidal, Virucidal, Fungicidal, Mycobactericidal and Sporicidal Activities of Liquid Chemical Germicides, E2197-02, West Conshohocken: ASTM, 2002.

17. Official Methods of Analysis. AOAC (Association of Official Analytical Chemists) International. Washington DC, AOAC 960.09. Available from: http://www.eoma.aoac.org Last accessed December 5, 2012.

18. Tomasino SF, Pines RM, Cottrill MP. Determining the efficacy of liquid sporicides against spores of Bacillus subtilis on a hard nonporous surface using the quantitative three step method: Collaborative study. J AOAC Int 2008;91:833–852.

19. Springthorpe VS, Sattar SA. Carrier tests to assess microbicidal activities of chemical disinfectants for use on medical devices and environmental surfaces. J AOAC Int 2005;88:182–201.

20. Best MV, Springthorpe S, Sattar SA. Feasibility of a combined carrier test for disinfectants: Studies with a mixture of five types of microorganisms. Am J Infect Control 1994;22:152–162.

21. Jones LA, Jr, Hoffman RK, Phillips CR. Sporicidal activity of peracetic acid and ß–propiolactone at subzero temperatures. Appl Microbiol 1967;15:357–362.

22. Jones LA, Jr, Hoffman RK, Phillips CR. Sporicidal activity of sodium hypochlorite at subzero temperatures. Appl Microbiol 1968;16:787–791.

23. Humphreys PN. Testing standards for sporicides. J Hosp Infect 2011;77:193–198.

24. Nattermann H, Becker S, Jacob D, Klee SR, Schwebke I, Appel B. Efficient killing of anthrax spores using aqueous and alcoholic peracetic acid solutions. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2005;48:939–950.


Article


I n t r o d u c t i o nAntibiotics as growth promoters have been used for decades

in poultry production to improve farm performance and control intestinal pathogens. With increasing interest in discontinuing the use of antibiotics as feed additives the search for alterna-tives has intensified. Ideally these alternatives should improve growth performance and maintain sound health for the chick-ens. Therefore, the search for new types of feed additives that are pollution-free has become the focus of current research (1). China is a large agricultural country, and substantial amounts of agricultural by-products, such as corn cob, cotton seed hull, and straw, that are rich in cellulose-type xylanase are produced every year (2). Xylooligosaccharides (XOS) can be produced from many

edible fungi through the hydrolysis of semicellulose by xylanase (3,4). Straw chaff, the substrate used for cultivating edible fungi, is regarded as XOS after the biologic degradation of fungi through 2 to 3 batches of fungus production. To explore the biologic char-acteristics of XOS and their application in livestock and poultry production, the authors studied the effects of XOS on growth performance, endocrine metabolism, and immune response in broiler chickens.

M a t e r i a l s a n d m e t h o d sThe animal research protocols conformed to those approved by the

Yangzhou University Animal Care and Use Committee, Yangzhou, China.

Effect of a straw-derived xylooligosaccharide on broiler growth performance, endocrine metabolism, and immune response

Sun Zhenping, Lv Wenting, Yu Ruikui, Li Jia, Liu Honghong, Sun Wei, Wang Zhongmie, Li Jingpan, Shan Zhe, Qin Yuling

A b s t r a c tThe aim of this work was to evaluate the effect of 3 levels of supplemental xylooligosaccharides (XOS) from straw on the growth performance, endocrine metabolism, and immune response of broiler chickens. Day-old, healthy Arbor Acres broilers (n = 192) received a basal diet of maize–soybean meal and, depending on the group to which they were allocated, no additive (control group) or the following experimental treatments for 59 d: treatment 1: 5 g XOS/kg; treatment 2: 10 g XOS/kg; and treatment 3: 20 g XOS/kg. By day 59 the body weight gain of the chickens receiving treatment 2 had increased by 9.44% (P , 0.01) over the gain of the control group. The levels of serum triiodothyronine, thyroxine, and insulin on day 44 were significantly higher in the treatment groups than in the control group. The titers of antibody to the avian influenza H5N1 virus on day 24 were also significantly higher in the treatment groups than in the control group, and on day 59 the titer of the chickens receiving treatment 2 were still significantly increased (P , 0.05). Thus, the addition of XOS to feed can increase growth performance, enhance endocrine metabolism, and improve immune function in broiler chickens.

R é s u m éL’objectif de ce travail était d’évaluer les effets de trois niveaux d’un supplément de xylooligosaccharides (XOS) provenant de la paille sur les performances de croissance, le métabolisme endocrinien, et la réponse immunitaire de poulets à griller. Des poussins à griller en santé âgés d’un jour (n = 192) de race Arbor Acres ont reçu une alimentation de base maïs-soya et, selon le groupe auquel ils ont été assignés, aucun additif (groupe témoin) ou pendant 59 j le traitement expérimental suivant : traitement 1 : 5 g XOS/kg; traitement 2 : 10 g XOS/kg; et traitement 3 : 20 g XOS/kg. Au jour 59, le gain de poids corporel des poulets recevant le traitement avait augmenté de 9,44 % de plus (P , 0,01) que le gain du groupe témoin. Les niveaux sériques de triiodothyronine, de thyroxine et d’insuline au jour 44 étaient significativement plus élevés dans les groupes de traitement que dans le groupe témoin. Les titres d’anticorps contre le virus de l’influenza aviaire H5N1 au jour 24 étaient également significativement plus élevés dans les groupes de traitement que dans le groupe témoin, et au jour 59 les titres des poulets recevant le traitement 2 étaient encore significativement augmentés (P , 0,05). Ainsi, l’ajout de XOS à l’alimentation peut augmenter les performances de croissance, augmenter le métabolisme endocrinien et améliorer la fonction immunitaire des poulets à griller.


College of Veterinary Medicine, Yangzhou University, Yangzhou 225009, China (Sun, Lv, Yu, Li, Liu, Sun, Wang, Li, Zhe); Jianhu Agriculture College, Jianhu 334700, China (Qin).

Address all correspondence to Professor Sun Zhenping; telephone: 86 514 87972245; fax: 86 514 87972218; e-mail: [email protected]

Received February 14, 2012. Accepted June 26, 2012.



Production of XOSEdible fungi were used by the Animal Physiology Laboratory

of the College of Veterinary Medicine at Yangzhou University to ferment and degrade straw chaff to make XOS. Briefly, the edible fungus Pleurotus ostreatus, obtained from the Yangzhou Academy of Agricultural Sciences, was cultivated in potato dextrose agar (200 g of potato, 15 g of agar, 20 g of glucose, 3 g of KH2PO4, 10 mg of VB1, and 1.5 g of MnSO4; Difco Laboratories, Detroit, Michigan, USA) at 30°C for 7 d, at which time mycelia completely covered the surface of the Petri dishes. Liquid medium was inoculated with 1 disk of agar plate mycelium 6 mm in diameter per 10 mL of medium. The liquid medium contained (per liter) 1.4 g of (NH4)2SO4, 2.0 g of KH2PO4, 0.1 g of urea, 0.3 g of MgSO4 · 7H2O, 0.3 g of CaCl2, 5.0 mg of FeSO4 · 7H2O, 1.56 mg of MnSO4 · H2O, 2.0 mg of CoCl2, and 1.4 mg of ZnSO4 · 7H2O. Solid fermentation was carried out in 500-mL plas-tic flasks containing 350 g of sterile solid medium (pH 5.5) plus 10% of liquid mycelium. The solid medium contained, per 100 g, 96 g of straw, 1 g of CaHPO4, 1.4 g of lime, 1.5 g of gypsum, and 0.6 mg of carbendazole. The flasks were incubated at 30°C for 40 d in a warm room without agitation. Paper chromatography and photoelectric colorimetry were used to determine that the XOS content of the degraded straw medium was 82.48 mg/g and that the other macro-nutrients consisted mainly of neutral detergent fiber (81.81 mg), acid detergent fiber (56.15 mg), and crude protein (8.30 mg).

Animals and experimental treatmentsDay-old Arbor Acres (AA) broiler chickens (n = 192) were pro-

vided by Nantong Haian Poultry Breeding Farm, Nantong, Jiangsu, China. Males and females were identified and housed separately in 6 wire cages per group, 8 birds per cage (3 cages for males and 3 for females). Each bird was numbered and weighed. The birds were ran-domly allocated to 1 of 4 experimental groups for 59 d. All received a basal diet of maize–soybean meal with no additive (control group) or XOS: 5 g/kg (treatment 1); 10 g/kg (treatment 2); or 20 g/kg (treatment 3). The basal diet was prepared by the authors accord-ing to the recommendations of the US National Research Council

(5). The details of the feed formula and nutritional levels are shown in Table I. The feed was made into pellets for use in 3 periods: the 1st period was from day 1 to day 21, the 2nd from day 22 to day 42, and the 3rd from day 43 to day 59.

Animal managementThe chickens were kept in cages in a chicken house during the

entire period of the experiments. Infrared lights and heaters were used to maintain warmth; natural ventilation and a sustained light-ing system were also implemented. The temperature was kept at 23.4°C to 29.0°C, and the relative humidity ranged from 50% to 80%. All the chickens had free access to feed and water.

Vaccination and sample collectionOn day 14 all the chickens were given an inactivated vaccine

against the H5N1 subtype of avian influenza (AI) virus via a single intramuscular injection. Starting on day 24 blood samples were collected every 5 d from the plantar vein of 6 chickens (3 male and 3 female) selected randomly from each group.

Measurement of parametersTo assess growth performance the weight of the chickens was

measured on days 1 and 59. Body weight gain, feed intake, and feed conversion ratio were determined.

The radioimmunoassay (RIA) technique was used to determine the serum concentrations of triiodothyronine (T3), thyroxine (T4), and insulin on day 44. The RIA assay kits were produced by Beijing Kemei Dongya Biotechnology Company, Beijing, China.

Serum antibody titers against the AI H5N1 vaccine virus were determined by hemagglutination inhibition (HI) (6) with use of a preparation of 1% chicken erythrocytes made by conventional technique. The antibody titers were expressed as the average of log2.

Statistical analysisA Microsoft Excel database was established for all the experi-

mental data. One-way analysis of variance was used in SPSS 11.5 statistical software (SPSS, Chicago, Illinois, USA) to determine the

Table I. Feed formula and nutritional levels of the broiler chickens’ basal diet

Composition Age (d) Age (d)(% of feed) 1–21 22–42 43–59 Nutritional level 1–21 22–42 43–59Corn 53.80 62.00 69.72 MJ (mCal/kg) 2.950 3.050 2.966Soybean meal 38.16 29.97 24.62 Crude protein (%) 21.92 19.06 17.72Fish meal 1.50 1.50 1.00 Linoleic acid (%) 1.52 1.73 1.73Limestone 1.10 0.90 1.00 Calcium (%) 1.00 0.85 0.81Calcium phosphate 1.70 1.60 1.60 Phosphorus (%) 0.47 0.42 0.42Table salt 0.30 0.30 0.30 Sodium chloride (%) 0.35 0.35 0.35Lysine 0.02 0.02 0.08 Lysine 1.14 0.96 0.96Methionine 0.25 0.21 0.18 Methionine (%) 0.54 0.47 0.47Vegetable oil 2.17 2.50 0.50 Methionine cystine 0.88 0.78 0.75Premixa 1.00 1.00 1.00a Supplied per kilogram of 1% premix: vitamin A, 1500 IU; vitamin D3, 200 IU; vitamin E, 10 IU; vitamin K, 0.5 mg; thiamine, 1.8 mg; riboflavin, 3.6 mg; pyridoxine, 3.5 mg; vitamin B12, 0.01 mg; pantothenic acid, 10 mg; niacin, 35 mg; choline, 1300 mg; biotin, 0.15 mg; folic acid, 0.55 mg; manganese, 60 mg; zinc, 40 mg; copper, 8 mg; iron, 80 mg.



differences between groups. Results for the multiple-range test vari-ant of the least significant difference method are shown as mean 6 standard error. Differences between means were considered signifi-cant at P , 0.01 and P , 0.05.

Re s u l t sAs shown in Table II, the chickens receiving treatment 2 had a

9.44% greater gain in body weight (P , 0.01) and a 4.18% lower feed conversion ratio (P , 0.05) than the control group. In addition, the chickens receiving treatment 3 had a 3.61% lower feed intake (P , 0.05) than the control group.

As shown in Table III, on day 44 the serum level of T3 in treatment groups 1 and 3 was greater than that in the control group by 63.64% (P , 0.01) and 90.51% (P , 0.01), respectively, and the serum level of T4 in treatment groups 1, 2, and 3 was greater than that in the control group by 35.11% (P , 0.01), 36.98% (P , 0.01), and 22.24% (P , 0.01), respectively. In addition, the serum insulin level in treat-ment groups 1 and 3 was greater than that in the control group by 12.53% (P , 0.05) and 16.28% (P , 0.01), respectively.

As shown in Table IV, on day 24 the serum HI antibody titer to the AI H5N1 vaccine virus was greater in treatment groups 1, 2, and 3 than in the control group by 31.77% (P , 0.01), 27.71% (P , 0.01), and 19.61% (P , 0.05), respectively. In addition, comparing with the

titers in the control group, the titer in treatment group 2 was 33.78% higher (P , 0.05) on day 59.

D i s c u s s i o nIn this study, broilers with XOS-supplemented diets had greater

body weight gain than those with a basal diet, along with decreased feed conversion, in agreement with the Food and Agriculture Organization of the United Nations (7), which suggested that XOS could be considered emerging prebiotics and defined a prebiotic as “a nonviable food component that confers a health benefit on the host associated with modulation of the microbiota”. Made up of xylose units, and approximately half as sweet as sucrose, XOS can be produced by enzymatic hydrolysis from xylan, the main component of plant hemicelluloses and therefore readily available in nature (8). The fungal genera Trichoderma, Aspergillus, Fusarium, and Pichia are considered great producers of xylanases (9–13). White-rot fungi have also been shown to produce extracellular xylanases that act on a wide range of hemicellulose materials and are also useful as food sources (14) and metabolites of interest to the pharmaceutical, cosmetic, and food industries (15,16). In this study, the edible fungus P. ostreatus was used to ferment and degrade straw chaff to make XOS. During the process, other nutritional agents were produced (e.g., neutral and acid detergent fiber, crude protein, and minerals)

Table II. Effects of xylooligosaccharides (XOS) added to the feed on the growth performance of the chickens in 59 d

Growth performance, mean 6 standard errora

Treatment group Body weight gain (kg) Feed intake (kg) Feed conversion ratioControl: basal diet 3.486 6 0.052 6.402 6 0.110 1.844 6 0.044Treatment 1: 5 g XOS/kg 3.694 6 0.085 6.475 6 0.057 1.767 6 0.031Treatment 2: 10 g XOS/kg 3.815 6 0.099** 6.479 6 0.063 1.716 6 0.045*Treatment 3: 20 g XOS/kg 3.510 6 0.050 6.171 6 0.048* 1.765 6 0.028a Significant difference from the mean for the control group at P-values of * , 0.05 and ** , 0.01.

Table III. Effects of XOS on serum hormone concentrations after 44 d

Hormone concentration, mean 6 standard errora

Triiodothyronine Thyroxine Insulin Treatment group (ng/mL) (ng/mL) (ng/mL)Control 1.422 6 0.159 4.654 6 0.317 7.095 6 0.333Treatment 1 2.327 6 0.189** 6.288 6 0.136** 7.984 6 0.357*Treatment 2 1.829 6 0.171 6.375 6 0.170** 7.762 6 0.190Treatment 3 2.709 6 0.272** 5.689 6 0.148** 8.250 6 0.275**a Significant difference from the mean for the control group at P-values of * , 0.05 and ** , 0.01.

Table IV. Changes over time in titer of hemagglutination inhibition antibody to the H5N1 subtype of avian influenza virus

Treatment Average log2 titer, mean 6 standard error;a age (d)group 24 29 34 39 44 49 54 59Control 4.1 6 0.2 5.8 6 0.3 4.6 6 0.2 5.4 6 0.3 5.0 6 0.3 6.5 6 0.4 4.8 6 0.5 5.1 6 0.5Treatment 1 5.4 6 0.3** 5.3 6 0.3 4.7 6 0.4 4.8 6 0.3 5.3 6 0.3 5.0 6 0.3 5.2 6 0.3 5.9 6 0.6Treatment 2 5.3 6 0.3** 5.5 6 0.5 4.8 6 0.3 5.1 6 0.5 5.6 6 0.5 5.6 6 0.2 5.7 6 0.2 6.8 6 0.4**Treatment 3 4.9 6 0.3* 4.3 6 0.3 4.6 6 0.2 5.1 6 0.3 4.3 6 0.3 5.0 6 0.4 4.9 6 0.2 5.1 6 0.5a Significant difference from the mean for the control group at P-values of * , 0.05 and ** , 0.01.



that could help to modify feed intake (17,18). The fact that XOS are relatively stable in acidic conditions may protect XOS from decom-position when passing through the stomach (19). Degradation in the intestine has been studied in vitro with an artificial model of diges-tive enzymes (a-amylase, pancreatin, gastric juice, and intestinal brush border enzymes), and no hydrolyzation of XOS (xylobiose) was observed (20). This suggests that XOS may be nondigestible or of low digestibility and would reach the colon intact. In addition, XOS can decompose harmful substances, produce organic acids and other beneficial substances, reduce fecal stench, and improve the farming environment (21).

The effects of T3 and T4 on avian metabolism and growth rates have been well-documented (22–24). The biologic activity of T3 is several times greater than that of T4, and T3 functions much faster than T4. Therefore, T3 is believed generally to be the main thyroid hormone with respect to physiological function (25). Kühn et al (26) found that the level of T3 in poultry is related positively to growth. Growth is slowed significantly in an animal with hypothyroidism. The relation between thyroid secretion rate and growth demon-strated in chickens (27) suggests that the poorer growth performance of the control birds in the present study could be partly attributed to decreased T3 activity. Comparison of the serum concentrations of T3 between the control and treatment groups indicated that XOS increased the T3 activity, though not always significantly. The level of T4 in the serum was significantly increased in the birds receiving supplemental XOS compared with the control group. These results indicate that XOS can improve thyroid function and that it helps the thyroid hormones participate in the growth and metabolism of poul-try. The main mechanism of action of T3 involves control of the gene expression and synthesis of growth hormone. It also increases insu-lin and the RNA content of muscle, which further promote protein synthesis (28). In the group receiving a supplement of 20 g XOS/kg, although the serum levels of T3 and T4 increased significantly, there was no notable difference in body weight gain compared with the control group. This may be explained by excessive supplementation; however, this question needs further exploration.

The main functions of insulin are to decrease the blood glucose level, to regulate growth, and to participate in a wide range of metabolic processes. One of the most important mechanisms of action of insulin is to facilitate the transport of glucose and amino acids into cells. This promotes protein synthesis and increases the concentration of glucose in the cells, which facilitates glycolysis and, further, improves the synthesis of fatty acids and the breakdown of triglycerides. In the meantime, insulin inhibits the degradation of glycogen, proteins, and triglycerides. An increased level of insulin assists glucose transport into cells, decreases the concentration of cyclic adenosine monophosphate, and enhances glycogen synthesis (29). This study showed that the insulin level in the groups treated with XOS was higher than that of the control group. The possible reasons are that XOS is rich in amino acids, mainly lysine (1.2%) and tryptophan (0.8%), and that XOS can facilitate intestinal and bowel movement and further increase digestibility. Other studies have shown that insulin can facilitate the transformation of T4 to T3 (30). Therefore, XOS supplementation can improve the metabolism of broilers.

Avian influenza is one of the most damaging diseases affecting the poultry industry. In the present experiment the serum levels of anti-body against the AI H5N1 vaccine virus were significantly greater in the treatment groups than in the control group. This suggests that XOS can strengthen humoral immunity in poultry. Because the level of maternal antibody may have had an influence on the experimental results, we will avoid the interference of maternal antibody and other external factors in future experiments.

A c k n o w l e d g m e n tThis work was supported by grant 08KJD180011 from the Jiangsu

Province Natural Science Foundation.

Re f e r e n c e s 1. Fasina YO, Thanissery RR. Comparative efficacy of a yeast

product and bacitracin methylene disalicylate in enhancing early growth and intestinal maturation in broiler chicks from breeder hens of different ages. Poult Sci 2011;90:1067–1073.

2. Gong Y, Jiang M. Biodiesel production with microalgae as feedstock: From strains to biodiesel. Biotechnol Lett 2011;33: 1269–1284.

3. Gobinath D, Madhu AN, Prashant G, Srinivasan K, Prapulla SG. Beneficial effect of xylo-oligosaccharides and fructo- oligosaccharides in streptozotocin-induced diabetic rats. Br J Nutr 2010;104:40–47.

4. Yoshida T, Tsubaki S, Teramoto Y, Azuma JI. Optimization of microwave-assisted extraction of carbohydrates from indus-trial waste of corn starch production using response surface methodology. Bioresour Technol 2010;101:7820–7826. Epub 2010 Jun 7.

5. NRC. Nutrient Requirements of Poultry. 9th ed. Washington, DC, National Academy Press, 1994.

6. Ducatez MF, Cai Z, Peiris M, et al. Extent of antigenic cross-reactivity among highly pathogenic H5N1 influenza viruses. J Clin Microbiol 2011;49:3531–3536.

7. Food Quality and Standards Service, Food and Agriculture Organization of the United Nations. FAO Technical Meeting on Prebiotics. 2007 Sep 15–16. Rome, Italy. Available from www.aat-taa.eu/index/en/company/download/1262610500.html Last accessed January 17, 2013.

8. Teng C, Yan Q, Jiang Z, Fan G, Shi B. Production of xylooligosac-charides from the steam explosion liquor of corncobs coupled with enzymatic hydrolysis using a thermostable xylanase. Bioresour Technol 2010;101:7679–7682.

9. Wong KKY, Saddler JN. Trichoderma xylanases, their properties and application. Crit Rev Biotechnol 1992;12:413–435.

10. Chistakopoulos P, Nerinckx W, Kekos D, Macris B, Claeyssens M. Purification and characterization of two low molecular mass alkaline xylanases from Fusarium oxysporum F3. J Biotechnol 1996;51:181–189.

11. De Vries RP, Kester HC, Poulsen CH, Benen JA, Visser J. Synergy between enzymes from Aspergillus involved in the degradation of plant cell wall polysaccharides. Carbohydr Res 2000;327: 401–410.



12. Den Haan R, Van Zyl WH. Enhanced xylan degradation and utilization by Pichia stipitis overproducing fungal xylanolytic enzymes. Enzyme Microb Technol 2003;33:620–628.

13. Adsul MG, Ghule JE, Shaikh H, et al. Enzymatic hydroly-sis of delignified bagasse polysaccharides. Carbohydr Polym 2005;62:6–10.

14. Buswell JA, Chang ST. Biomass and extracellular hydrolytic enzyme production by six mushroom species grown on soybean waste. Biotechnol Lett 1994;16:1317–1322.

15. Jong SC, Donovick R. Antitumoral and antiviral substances from fungi. Adv Appl Microbiol 1989;34:183–262.

16. Cai QE, Yue XY, Niu TQ, Ji C, Ma QG. The screening of cul-ture condition and properties of xylanase by white-rot fungus Pleurotus ostreatus. Process Biochem 2004;39:1561–1566.

17. Centeno C, Arija I, Viveros A, Brenes A. Effects of citric acid and microbial phytase on amino acid digestibility in broiler chickens. Br Poult Sci 2007;48:469–479.

18. Fébel H, Mézes M, Pálfy T, et al. Effect of dietary fatty acid pat-tern on growth, body fat composition and antioxidant param-eters in broilers. J Anim Physiol Anim Nutr 2008;92:369–376.

19. Imaizumi K, Nakatsu Y, Sato M, Sedarnawati Y, Sugano M. Effects of xylooligosaccharides on blood glucose, serum and liver lipids and cecum short-chain fatty acids in diabetic rats. Agric Biol Chem 1991;55:199–205.

20. Koga K, Fujikawa S. Xylo-oligosaccharides. In: Nakakuki T, ed. Oligosaccharides: Production, Properties and Applications. Japanese Technology Reviews. Philadelphia, Pennsylvania: Gordon and Breach Science Publishers, 1993:130–143.

21. Zhou J, Du WX, Yang TG. Effects of different xylooligosaccharide combinations on productive performance of ducks and laying hens. Anim Husb Vet 2006;9:21–24.

22. Scott T, van der Zijpp A, Glick B. Effect of thiouracil-induced hypothyroidism on the humoral immunity of New Hampshire chickens. Poult Sci 1985;64:2211–2217.

23. Donkoh A. Ambient temperature: A factor affecting performance and physiological response in broiler chickens. Int J Biometeorol 1989;33:259–265.

24. Haddad EE, Mashaly MM. Effect of thyroidectomy of immature male chickens on circulating thyroid hormones and on responses to thyroid-stimulating hormones and chronic cold exposure. Poult Sci 1989;68:365–376.

25. Proszkowiec-Weglarz M, Richards MP, Humphrey BD, Rosebrough RW, McMurtry JP. AMP-activated protein kinase and carbohydrate response element binding protein: A study of two potential regulatory factors in the hepatic lipogenic program of broiler chickens. Comp Biochem Physiol B Biochem Mol Biol 2009;54:68–79.

26. Kühn ER, Decuypere E, Rudas P. Hormonal and environmental interactions on thyroid function in the chick embryo and post-hatching chicken. J Exp Zool 1984;232:653–658.

27. Leung FC, Taylor JE, van Iderstine A. Thyrotrophin-releasing hormone stimulates body weight gain and increased thy-roid hormones and growth hormone in plasma of cockerels. Endocrinology 1984;115:736–740.

28. Fernández I, Martín del Río R, Orensanz LM. Surgical ablation of oviductal extrinsic innervation changes GABA levels in the rat fallopian tube. Life Sci 1985;36:1733–1737.

29. Larabee JL, Maldonado-Arocho FJ, Pacheco S, et al. Glycogen synthase kinase 3 activation is important for anthrax edema toxin-induced dendritic cell maturation and anthrax toxin receptor 2 expression in macrophages. Infect Immun 2011;79: 3302–3308. Epub 2011 May 16.

30. Mitsuma T, Nogimori T, Chaya M. Beta-casomorphin inhibits thyrotropin secretion in rats. Exp Clin Endocrinol 1984;84: 324–330.


Article


I n t r o d u c t i o nLawsonia intracellularis is a Gram-negative, obligate intracellular

bacterium that affects a wide range of domestic, wild, avian, and laboratory animal species (1–5). Principally, L. intracellularis is known

as the etiologic agent of porcine proliferative enteropathy (PPE) and equine proliferative enteropathy (EPE), with approximately 98% 16S recombinant DNA (DNAr) gene similarities reported between strains (2,4,6). Associated with L. intracellularis’s intra-cytoplasmic replica-tion is the patent proliferation induced in the host’s enterocytes of

The rabbit as an infection model for equine proliferative enteropathyFrancesca Sampieri, Andrew L. Allen, Nicola Pusterla, Fabio A. Vannucci, Aphroditi J. Antonopoulos,

Katherine R. Ball, Julie Thompson, Patricia M. Dowling, Don L. Hamilton, Connie J. Gebhart

A b s t r a c tThe objective of this study was to demonstrate the susceptibility of rabbits to Lawsonia intracellularis obtained from a case of clinical equine proliferative enteropathy (EPE). This is a preliminary step toward developing a rabbit infection model for studying pathogenesis and therapy of EPE in horses. Nine does were equally assigned to 3 groups. Animals in 2 groups (Group 1 and Group 2) were orally inoculated with different doses of cell-cultured L. intracellularis. Controls (Group 3) were sham-inoculated. Feces and blood were collected before the rabbits were infected and at 7, 14, and 21 days post-infection (DPI). Serum immunoglobulin G (IgG) titers were measured using an immunoperoxidase monolayer assay (IPMA) and fecal samples were analyzed with quantitative polymerase chain reaction (qPCR). A doe from each group was euthanized at 7, 14, and 21 DPI for collection and evaluation of intestinal samples. Tissues were stained by routine hematoxylin and eosin (H&E) method and immunohistochemistry (IHC) with L. intracellularis-specific mouse monoclonal antibody. At 14 DPI, serologic responses were detected in both infected groups, which maintained high titers through to 21 DPI. Lawsonia intracellularis DNA was detected in the feces of Group 2 on 7 DPI and in both infected groups on 14 DPI. Gross lesions were apparent in Group 1 and Group 2 on 14 DPI. Immunohistochemistry confirmed L. intracellularis antigen within cells of rabbits in Group 1 and Group 2 on 7, 14, and 21 DPI. No lesions, serologic response, shedding, or IHC labeling were found in Group 3 rabbits. This study describes an EPE rabbit model that simulates natural infection, as typical lesions, immune response, and fecal shedding were present.

R é s u m éCette étude visait à démontrer la susceptibilité des lapins à Lawsonia intracellularis obtenu d’un cas clinique d’entéropathie proliférative équine (EPE). Ceci est une étape préliminaire dans le développement d’un modèle d’infection chez le lapin pour étudier la pathogénie et le traitement de l’EPE chez les chevaux. Neuf lapines ont été assignées également à 3 groupes. Les animaux dans deux groupes (Groupe 1 et Groupe 2) ont été inoculés oralement avec différentes doses de L. intracellularis cultivés sur cellules. Les témoins (Groupe 3) étaient faussement inoculés. Des fèces et du sang ont été prélevés avant que les lapins soient infectés et aux jours 7, 14 et 21 post-infection (DPI). Les titres sériques d’immunoglobulines G (IgG) ont été mesurés par une épreuve d’immunoperoxydase en monocouche (IPMA) et les échantillons de fèces ont été analysés par réaction quantitative d’amplification en chaîne par la polymérase (qPCR). Une lapine de chaque groupe a été euthanasiée 7, 14 et 21 DPI pour prélèvement et évaluation d’échantillons intestinaux. Les tissus étaient colorés à l’aide d’hématoxyline et éosine (H&E) et en immunohistochime (IHC) avec un anticorps monoclonal de souris spécifique à L. intracellularis. Au jour 14 post-infection, une réponse sérologique a été détectée chez les animaux des deux groupes infectés, et des titres élevés ont été maintenus jusqu’à 21 DPI. De l’ADN de L. intracellularis fut détecté dans les fèces du Groupe 2 au jour 7 PI et dans les 2 groupes infectés au jour 14 PI. Des lésions macroscopiques étaient apparentes dans le Groupe 1 et le Groupe 2 au jour 14 PI. L’immunohistochime a confirmé la présence d’antigène de L. intracellularis à l’intérieur des cellules de lapins dans les Groupes 1 et 2 aux jours 7, 14 et 21 PI. Aucune lésion, réponse sérologique, excrétion, ou marquage en IHC n’ont été trouvés chez les lapins du Groupe 3. La présente étude décrit un modèle lapin d’EPE qui imite l’infection naturelle, étant donné la présence de lésions typiques, de réponse immunitaire et d’excrétion fécale.


Department of Veterinary Biomedical Sciences (Sampieri, Antonopoulos, Ball, Dowling, Hamilton), Department of Veterinary Pathology (Allen), Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, Saskatchewan; Department of Veterinary Medicine and Epidemiology, School of Veterinary Medicine, University of California, Davis, California, USA (Pusterla); Department of Veterinary and Biomedical Sciences, College of Veterinary Medicine, University of Minnesota, Saint Paul, Minnesota, USA (Vannucci, Gebhart); Canadian Light Source, Saskatoon, Saskatchewan (Thompson).

Address all correspondence to Dr. Francesca Sampieri; telephone: 502-727-3460 fax: 270-767-7450; e-mail: [email protected] or [email protected]

Received February 10, 2012. Accepted June 6, 2012.



either the small or large intestine, which reduces nutrient absorp-tion and results in malabsorption, diarrhea, depression, weight loss, abdominal pain, and even death (7–8). Typically, EPE affects the enterocytes of the distal portion of the jejunum and ileum of wean-ling foals. Several individual, isolated cases of EPE were reported before 1999 (9–12). The first large outbreaks in foals occurred in 1999 in Quebec and Ontario, Canada and clinical EPE has since become increasingly important worldwide (6,13–14).

Lawsonia intracellularis is a challenging organism to isolate and maintain in vitro, as its growth requires an intracellular environ-ment and a specific microaerophilic atmosphere (15). Many studies about susceptibility to L. intracellularis in pigs targeted not only bacterial culture in vitro, but also experimental reproduction of PPE in animal models. Hamsters are the main laboratory species that can be naturally affected by L. intracellularis and the disease in this species is commonly and non-specifically referred to as “wet tail” (2,8). Hamsters were the first model species in which information about lesion progression was obtained experimentally, as disease was induced through laborious filtration of scrapings of porcine-infected ileal tissue (16). Since the early 1990s, hamsters have been considered a potentially advantageous model for PPE because they cost less and provide larger statistical power of experiments than mice (17–21). Reproduction of the disease in laboratory species using intestinal mucosa homogenate or pure cell cultures has tradition-ally been limited to pigs or hamsters, either infected (from strains collected from animals of the same species) or cross-infected (from strains collected from different animal species) (17,22–23). Although a few investigators have successfully produced typical lesions of proliferative enteropathy (PE) in mice, hamsters continue to provide a reliable infection model for investigating PPE using porcine ileal scraping homogenates as inocula (8,18,24).

In foals, EPE has only been experimentally reproduced after oral inoculation of an equine strain of L. intracellularis, which originated from the intestine of an affected foal that succumbed to EPE (25). There are no published reports about EPE reproduced in a different species, however, which raises the question of whether other animals are susceptible to the equine L. intracellularis strain. Two preliminary infection trials in hamsters, which used a virulent pure cell culture equine L. intracellularis strain, did not yield EPE lesions. However, classical microscopic lesions were clearly observed in positive control hamsters infected with a pure cell culture porcine L. intracellularis strain (Sampieri F, Vannucci FA, Allen AL, et al. 2012, University of Saskatchewan, Western College of Veterinary Medicine).

Interestingly, natural L. intracellularis infection is also recognized in rabbits, although less frequently than in hamsters. Clinical signs are similar to those noted in other species, although the intestinal lesions are also often found in the cecum and large colon (18,26–28). Furthermore, fecal shedding has been observed through quantitative polymerase chain reaction (qPCR) of fecal material recovered from wild rabbits living on the grounds of equine stud farms where EPE was endemic (1).

The objective of this study was to reproduce EPE lesions in rab-bits, due to the anatomical and physiological similarities of their gastrointestinal systems to the horse (29). This is the first report of experimental infection in rabbits with a pure cell culture of L. intracellularis derived from a clinical case of EPE and it therefore

represents a valuable preliminary step toward developing a rabbit infection model for studying EPE in horses.


AnimalsNine, 7- to 8-wk-old female New Zealand white rabbits (Charles

River Canada, Pointe Claire, Quebec) were used (Oryctolagus cuniculus, spp. strain -052). The rabbits originated from colonies tested at regular intervals for absence [viral antibody free (VAF)] of Reovirus, Rotavirus, L. intracellularis, Helicobacter spp., cilia- associated respiratory (CAR) bacillus, Salmonella spp., Pasteurella spp., Pseudomonas aeruginosa, Treponema helminths spp., Clostridium piliforme, and Eimeria spp. among other viral, bacterial, and para-sitic diseases. On arrival at the Animal Care Unit at University of Saskatchewan’s Western College of Veterinary Medicine, the 9 does were mandatorily acclimated to the room for 1 wk. At the beginning of the experimental trial, the rabbits were weighed (range: 1.8 to 2.3 kg), examined, treated with a local anesthetic (EMLA Cream; Astra Zeneca Canada, Mississauga, Ontario), and tattooed on the right ear. Rabbits were randomly assigned to 3 groups of 3 animals each (Group 1, Group 2, and Group 3), group-housed in isolated pens in a Containment Level-2 room, fed rabbit pellets (Whole Earth Rabbit Ration; Federated Co-operatives, Saskatoon, Saskatchewan) ad libitum, and maintained using standard husbandry conditions (12/12 h light-dark cycle and 20°C 6 2°C room temperature).

All groups were housed in the same room, but were managed separately, with particular care to assess Group 3 (uninfected con-trols) first. Animals were bedded on ventilated shavings. The base of each pen was separated and isolated from other pens by PVC compartments (40 to 50 cm high) and a stainless steel mesh at the top of the pens prevented direct contact between rabbits in adjacent pens and the mixing of bedding. Cleaning and disinfection procedures were standardized for the facilities and a peroxide-based disinfec-tant (Peroxigard; Bayer, Toronto, Ontario) was used. This study was approved by the Animal Research Ethics Board of the University of Saskatchewan and conducted according to Canadian Council on Animal Care (CCAC) guidelines.

Inoculum preparationLawsonia intracellularis inocula were prepared as described in a

previous paper (25). The infectious material was harvested during necropsy of a foal diagnosed with EPE, for which clinical signs were confirmed by histopathology and immunohistochemistry (IHC) findings. The L. intracellularis harvested from the ileal mucosa were subsequently cultured on McCoy cells [American Type Culture Collection (ATCC) CRL-1696], as described in previous papers (15,30). Two different doses of inocula were prepared and each was suspended in 3 mL of buffered sucrose/phosphate/glutamate (SPG) medium to be administered to rabbits in Group 1 and Group 2 (31). Sham treatment (SPG medium only) was administered to Group 3 (controls). An SYBR green-based PCR quantitation assay was used to determine the bacterial concentration of the inocula (32). This assay targets the aspartate ammonia lyase gene (aspA gene), which is represented just once on the L. intracellularis chromosome, so that



each aspA gene copy corresponds to 1 organism (32). Since a lower culture passage appears related to higher L. intracellularis virulence, passage cultures lower than 20 (passages 9 and 10, for Groups 1 and 2, respectively) were used to infect the rabbits, but with differ-ent doses, to potentially compensate for the difference in passage. Thus, Group 1 received a dose equivalent to 1.3 3 108 bacteria/rabbit and Group 2 received a dose equivalent to 2.5 3 108 bacteria/rabbit. These infecting doses, obtained from the same equine-infected isolate previously used in the foal model, are approximately 150 to 200 times smaller than the doses used to infect foals and 6 to 10 times larger than the dose used to infect hamsters (25) (Sampieri F, Vannucci FA, Allen AL, et al. 2012, University of Saskatchewan, Western College of Veterinary Medicine).

Inoculation proceduresRabbits were inoculated with L. intracellularis on the first day

after the acclimation period (1 wk after arrival). Prior to the infec-tion, pooled fecal samples were collected from each separate pen, each rabbit was weighed and its general health was assessed, and a blood sample was collected from the central artery of the left ear, which had previously been prepped with a local anesthetic cream (EMLA Cream; Astra Zeneca Canada). A local lidocaine gel block (Xylocaine Gel 2%; Astra Zeneca Canada) was applied inside the right nostril and a 40-cm long feeding tube (Kendall Sovereign; Tyco Healthcare Group, Mansfield, Massachusetts, USA) was inserted through the nose into the stomach. Each rabbit received either a 3-mL L. intracellularis inoculum or a 3-mL SPG medium-only treat-ment naso-gastrically.

Collection of samplesChanges in body weight were monitored daily throughout the

21 d of the study. Demeanor, gross appearance, fecal consistency and quality, appetite, and self- and mutual grooming were monitored twice daily using a 5-step grading system, a score of 0 corresponding to normal and 4 to severely abnormal findings. All rabbits readily adapted to handling and their interest could be easily stimulated with carrots and apples (treats). Treats were offered twice daily to check appetite, as it is difficult to assess each individual’s food intake in the group-housing setting. Lack of interest in treats was considered a sign of decreased well-being (33).

Blood samples for serology were collected from the central artery of the left or right ear of each rabbit once weekly and at the time of euthanasia. Pooled fecal samples were also collected from each separate pen once weekly and individual fecal samples were col-lected at the time of euthanasia. One doe per week in each group was randomly selected and humanely euthanized with an intravenous overdose (720 mg/rabbit) of pentobarbital (Euthanyl; Bimeda-MTC, Cambridge, Ontario). Necropsies were performed to observe evidence of gross lesions and to collect multiple samples (1.25 to 2.5 cm long) from the intestinal tract for microscopic examination. Samples were immediately placed in phosphate-buffered 10% for-malin solution.

Sample analysisMacroscopic examination — Shortly after euthanasia, the abdomi-

nal cavity was opened, the mesentery and mesocolon were dissected,

and the organs were examined visually, from stomach to rectum. Changes in thickness, discoloration, or content of the intestinal tract were observed. The sample of duodenum was collected 2.5 cm distal to the pylorus; 1 jejunal sample was collected approximately mid- distance between the end of the duodenum and the beginning of ileum, whereas the terminal jejunal sample was collected proximal to the last jejunal Peyer’s patch. The ileal sample was collected proximal and adjacent to the ileal Peyer’s patch, approximately 5 cm proximal to the ileocecal valve. The ileocecal valve was also collected in its entirety, including the ampulla coli and sacculus rotundus. The sample of cecum was collected in the transitional area between cecum and appendix ceci and the terminal portion of the appendix ceci was also collected. The sample of large colon was collected adjacent and distal to the cecum, whereas the sample of terminal colon was collected at a mid-distance between the end of the large colon and the rectum.

Histology and immunohistochemistry — Two adjacent formalin-fixed, paraffin-embedded sections per sample were cut and stained by hematoxylin-eosin method (H&E) and streptavidin method, using anti-L. intracellularis-specific murine monoclonal antibody (IHC), to observe for typical proliferative lesions of the intestinal epithelium and the presence of the antigen within the cells (30,34). The L. intracellularis-specific antigen in the enterocytes was blindly evaluated with a 5-grade IHC scoring system as follows: grade 0 for no antigen labeling in the tissue; grade 1 for up to 25% of crypts labeled; grade 2 for 26% to 50% of crypts labeled; grade 3 for 51% to 75% of crypts labeled; and grade 4 for 76% to 100% of crypts labeled (30). For each rabbit, the negative control for each tissue sec-tion consisted of a correspondent tissue section IHC-labeled, except for the primary antibody. Furthermore, pig ileal tissues, known to be negative and positive for L. intracellularis infection, were labeled with the murine anti-L. intracellularis monoclonal antibody to confirm the antibody’s specificity and sensitivity, respectively.

Serology analysis — In serum, anti-L. intracellularis-specific IgG concentration was measured by an immunoperoxidase monolayer assay (IPMA), as reported in a previous paper (35). Positive serum samples were end-point titrated, starting with a dilution of 1:30 up to 1:1920. Control samples consisted of serum from a rabbit before (negative control) and after (positive control) hyperimmunization with L. intracellularis, purified from cell culture. Also, serum total protein concentration was measured with the refractometer method to investigate if changes in serum total protein concentration were comparable to the values reported in foals naturally and experimen-tally infected with EPE (25).

Quantitative PCR analysis — qPCR analysis was conducted on fecal samples, as described in a previous paper (25). The puri-fied DNA was analyzed by qPCR for presence of L. intracellularis aspA gene copies (1,36). For each target gene, 2 primers and an internal, fluorescent-labeled TaqMan probe [59end, reporter dye FAM (6-carboxyflourescein), 39end, quencher dye TAMRA (6-carboxytetramethylrhodamine)] was designed using Primer Express software (Applied Biosystems, Foster City, California, USA) (Table I). TaqMan PCR systems were validated using 2-fold dilutions of genomic DNA (gDNA), testing positive for the target genes. Dilutions were analyzed in triplicate and a standard curve was plotted against the dilutions. The slope of the standard curve was used to calculate amplification efficiencies using the formula



E = 10 1/-s21. Each system needed to achieve greater than 95% effi-ciency to be considered. The detection limit for “L.intra system” is from 5 to 10 copies/mL of DNA (determined with plasmid DNA). The “L.intra system” detects all L. intracellularis, regardless of the host species. Known positive controls and no template controls were run on every plate and met previously determined standardization values. DNA quality was determined by the “PanBacteria system,” with a control value under 30, to pass quality control (37).

RT-reaction and real-time TaqMan PCR — Each PCR reaction contained 203 primer and probes for the respective TaqMan system, with a final concentration of 400 nM for each primer and 80 nM for the TaqMan probe and commercially available PCR Mastermix (TaqMan Universal PCR Mastermix; Applied Biosystems) contain-ing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 5 mM MgCl2, 2.5 mM deoxynucleotide triphosphates, and 0.625 U AmpliTaq Gold DNA polymerase per reaction; 0.25 U AmpErase UNG per reaction; and 1 mL of the DNA sample in a final volume of 12 mL. The samples were placed in a 384-well plate and amplified in an automated fluorometer (ABI PRISM 7900 HTA FAST; Applied Biosystems). Applied Biosystem’s standard amplification conditions were used: 2 min at 50°C, 10 min at 95°C, 40 cycles of 15 s at 95°C, and 60 s

at 60°C. Fluorescent signals were collected during the annealing temperature and cycle threshold (Ct) values were extracted with a threshold of 0.1 and baseline values of 3 to 12 for all samples using SDS 2.2.1 (Applied Biosystems). Absolute quantitation was calcu-lated by a standard curve and expressed as copy numbers of the L. intracellularis aspA gene per gram of feces.

StatisticsIn general, only descriptive statistics were used in this study as

sample numbers were insufficient for statistical analysis. However, body weight and serum total protein concentration were analyzed with a Kruskal-Wallis non-parametric test and Dunn post-hoc testing for multiple comparisons among groups. A commercial software (GraphPad Prism 5.4 Software, La Jolla, California, USA) was used, with P value set at # 0.05.

Re s u l t s

Clinical signsDuring the entire duration of this study, no evidence of diarrhea,

overt signs of pain, e.g., bruxism, or prostration was detected in the rabbits. Minor details of incomplete grooming around the tail region were noted in rabbits of all groups, but these were observed sporadically and deemed not relevant. No remarkable changes were noted in any group during the first 7 DPI, except for a transient (1 d’s duration only) weight loss, or a weight gain smaller than half the individual’s average daily gain. This weight loss affected control and infected rabbits alike on 1 DPI and was ascribed to the stressful procedures on the day of infection (day 0). As the rabbits were still growing, their body weight gain should have ranged from 100 to 200 g/week (or 14.2 to 28.5 g/day), as per the provider’s indications (Charles River Laboratories). The averaged daily weight gains of all rabbits were within limits, with 2 rabbits well above that range (52 g and 49.1 g for Group 1 and Group 2, respectively). Negative weight changes were not statistically significant (P = 0.87) in any of the groups throughout the 21 DPI (Figure 1). However, weight losses or gains smaller than half the individual’s average daily gain were noted in infected rabbits in Group 1 and Group 2 from 12 to 17 DPI. The only animal that showed depression and dullness between 12 (rated 1) and 14 (rated 2) DPI was the Group 1 doe euthanized at 14 DPI, which showed marked progressive weight loss and lack of appetite. Additionally, although general self-grooming appeared

Table I. Specific forward and reverse primer sequences, along with probe, target gene, and accession numbers for the technique of polymerase chain reaction (PCR) used to diagnose L. intracellularis fecal shedding in rabbits after infectious challenge with pure cell bacteria. Note the rate of efficiency in the last column of the table

System Forward primer Reverse primer Probe Target Accession number EfficiencyL. intra bcL.intra-114f bcL.intra-263r bcL.intra-201p Aspartate EU127293.1 95.6% CACTTGCAAACAA CATTCATATTTGTACTTGT TCCTTGATCAATTTGTTGTGGAT ammonia TAAACTTGGTCTTC CCCTGCA TGTATTCAAGG lyase

Pan-bacteria PB.283f PB.352r PB.305 16S Multi-sequence 98.2% GGATGATCAGCCA CCAATATTCCTCACTGCT CCCGTAGGAGTCTGGACCGTG rRNA alignment CACTGGA GCC TCTCA

Figure 1. Weight gains throughout the experiment (21 DPI). In the 2 infected groups, a trend to reduced weight gain and marginal weight loss can be noted, although these were statistically not different than the control weight gains (P = 0.87).



normal, the normal residues of cecotrophes on the perineum of this doe were not cleaned, which is deemed an abnormal behavior.

Necropsy findingsGross lesions were not evident in any of the rabbits examined

on 7 DPI, but on 14 DPI, infected rabbits in Group 1 and Group 2 showed hyperemia and edematous corrugation of the serosal surface of the jejunum (such as a “cobblestone-like” appearance) (Figure 2a) compared to controls (Figures 2b and 2d). Additionally, typical pro-liferative enteropathy lesions were observed in the mucosal surface (Figure 2c). There was a similar “cobblestone-like” appearance in the ileum, particularly close to the ileocecal valve. These gross changes were consistent with proliferative enteropathy lesions described in rabbits, pigs, and horses (2,14,27). The cecum of the Group 1 infected rabbit appeared generally hyperemic, once opened, and it was deemed non-specific for EPE. Hyperemia was not noticeable in the cecum of does in Group 2 or Group 3.

By 14 DPI, the Group 1 rabbit that was euthanized had lost over 200 g in 3 d. In contrast, the rabbits in Group 2 and Group 3 gained marginal or no weight, but did not lose weight. Also, the terminal colon and rectum of the Group 1 rabbit were mostly empty, although cecum and colon content did not differ from the control doe.

By 21 DPI, only mild, non-specific changes were noted on gross pathology at necropsy. These changes included hyperemia and mild wall-thickening (edema) of the serosal surface of the jejunum of rabbits in both Group 1 and Group 2, both of which were more pronounced in Group 1.

Histology and immunohistochemistryTypical histological lesions, characterized by hyperplasia of

immature enterocytes and absence of goblet cells, were observed by H&E staining, but no inflammation was detected. These prolif-erative changes were consistently associated with the presence of intracellular bacteria identified by IHC (Figures 3a and 3b). Results of IHC analysis on intestinal sections are summarized in Table II. Immunohistochemistry analysis was conducted on intestinal sec-tions from duodenum to rectum at 7, 14, and 21 DPI. At 7 DPI, L. intracellularis antigen was found as multifocal (mild to diffuse) labeling in the apical cytoplasm of enterocytes in the jejunum, cecal tip, and colon of the infected rabbit in Group 1 and in the jejunum, ileum, cecum, and colon of the infected rabbit in Group 2 (Figure 3a). The labeled antigen in cecal epithelium is shown in Figure 3b. At 7 DPI, all stained antigen was in the enterocytes and no L. intracellularis antigen was detectable in the lamina propria of any of the intestinal sections.

At 14 DPI, the antigen was found in the epithelium and lamina propria as focal to multifocal and diffuse labeling and the extent of the microscopic lesions previously observed in the H&E staining was associated with the increased IHC grading (grades 2 or 3). In the infected rabbit in Group 2, L. intracellularis antigen was not found in the lamina propria of the terminal jejunum, appendix ceci, or colon samples, but the grading was higher due to the higher frequency of antigen detection in the epithelium.

At 21 DPI, IHC analysis showed no detectable antigen in any of the infected groups, except a focally labeled area noted in the lamina propria of the cecal tip of the infected rabbit in Group 1. Thus, IHC

Figure 2. Comparison of infected and uninfected jejunum in a rabbit model for L. intracellularis infection. Scale bar, 1 cm. a — Serosal surface of the mid-jejunum from an infected rabbit (Group 1 inoculum) at 14 DPI. Note the cobblestone-like appearance of the serosal surface. A round jejunal Peyer’s patch is visible on the left side (arrow). b — Serosal surface of a mid-jejunum section of a control rabbit. c — Luminal surface of the mid-jejunum from an infected rabbit (Group 1 inoculum) at 14 DPI. Note the rugae of the mucosa. d — Luminal surface of the mid-jejunum from an uninfected control rabbit. Note the smooth surface.

a

c

d

b



labeling and histologic analysis demonstrated that the rabbits had lesions from about 7 through 21 DPI, with highest severity noted at 14 DPI. No presence of labeled L. intracellularis antigen was detected in the duodenum, terminal colon, or rectum of any groups. In addi-tion, antigen was not detected in the intestinal tract of the control rabbits (Group 3) at any point in the experiment.

Serum analysesAs shown in Figure 4, serologic responses were detectable by

immunoperoxidase monolayer assay (IPMA) analyses in the infected groups at 14 DPI. Lawsonia intracellularis infection generated a high serologic response in both infected groups, coinciding with the severity of lesions. No significant differences (P = 0.72) were found among groups with regard to total protein concentrations in sera.

Quantitative PCR analysisThe results of the qPCR analysis are shown in Figure 5. Lawsonia

intracellularis was detected in the feces of infected rabbits in Group 2 at 7 DPI and in rabbits in both Groups 1 and 2 at 14 DPI, becoming undetectable at 21 DPI. In Group 3 rabbits (controls), L. intracellularis organisms were never shed in feces throughout the experiment.

D i s c u s s i o nThis study describes the initial development of a rabbit infection

model to investigate EPE in horses. Results show that there is merit in substituting rabbits for horses in infection model studies for EPE, as juvenile rabbits became infected after inoculation with a pure cell culture equine strain of L. intracellularis. Such findings provide a basis for pathogenesis studies, as this EPE infection model can be obtained within a relatively short incubation period and lesions can resolve quickly without subjecting animals to excessive discomfort. Although purified through growth in cell culture, the equine strain of L. intracellularis was pathogenic for rabbits. The infection did not cause marked clinical illness, but macroscopic and microscopic lesions were detected and confirmed by histopathology and IHC. Such findings were comparable to those observed in foals naturally and experimentally infected with EPE (6,25). Similarly, in the infected rabbits, fecal qPCR findings and IPMA serology showed fecal shed-ding consistent with bacterial active replication and marked immune response, respectively.

A trend of reduced growth performance was noted in rabbits in Group 1 and Group 2 at around 12 and 17 DPI, which approximately corresponded with the time of the infection’s peak (14 DPI) when lesions appeared more severe on gross pathology and IHC labeling. Moderate weight loss and dullness, along with decreased appetite, were the most remarkable clinical signs noted. Overall, the clinical signs observed were mild; the experimentally infected rabbits in this model could probably have survived through the acute phase of the infection. Diarrhea was never noted, contrary to what is reported in both naturally infected rabbits and naturally and experimentally infected foals (25,27).

The clinical condition of infected does in this study appeared normal on 21 DPI, although the surviving rabbit in Group 1 (the most clinically affected group) did not appear to gain weight at the rate exhibited before the peak of disease (or less than half the average daily weight gain). Even with no clinical support, recovery appeared to occur by 21 DPI, as evidenced by improved weight gain and the absence of major IHC labeling. Mild IHC labeling (grade 1) in the cecum of the Group 1 rabbit could be consistent with ongoing recovery. This differs from reports about rabbits exhibiting clinical disease after natural exposure when clinical signs consist of marked diarrhea, weight loss, and dehydration or from sub-clinically affected rabbits, as they are commonly reported to be exposed to the disease (1,27–28,38).

According to previous reports and personal communications about infection of hamsters with porcine strain L. intracellularis grown in cell culture, clinical signs are usually subtle or absent, yet characteristic histopathology lesions are present (8,24). In the rabbits of this study, gross lesions typical of EPE extended from the

Figure 3. Immunohistochemistry (IHC) stain. L. intracellularis antigen-specific staining with AEC substrate-chromogen and counterstained with Mayer’s hematoxylin. a — Note ileal crypts of an experimentally infected rabbit (14 DPI). Note the hyperplasia of immature enterocytes associated with bacterial infection (apical membrane). Scale bar, 10 mm. b — Note the cecal mucosa of an experimentally infected rabbit (14 DPI), showing labeled antigen in the superficial epithelium. Scale bar, 100 mm.

a

b



mid-jejunum to the terminal ileum in spite of mild clinical signs (6,11,14). In both horses and pigs, L. intracellularis is characteristi-cally found free within the apical cytoplasm of small intestinal crypt cells, particularly in the distal jejunum and ileum (2). At the disease’s peak, cecal hyperemia was noted in the Group 1 rabbit and deemed a non-specific change, although cecal and colonic proliferative lesions are typically found in naturally infected rabbits (26). In this cross-infection model, lesion location may be related to the bacterial isolate involved, rather than to the animal species affected. In other words, even in rabbits, it appears that EPE is located preferentially in those sites typical for horses, such as the jejunum and ileum, whereas mild, non-specific lesions are found only in sites observed in overtly diseased rabbits, such as the cecum and colon (27,38). However, further studies are warranted to confirm this suspicion.

Lesions were confirmed through IHC labeling starting at 7 DPI, with their severity increasing at 14 DPI, but decreasing dramati-cally by 21 DPI. Such a finding is quite novel, as the incubation period is longer in experimentally infected horses and pigs (2,25). Furthermore, the presence of antigen in the lamina propria was noted, particularly in the infected rabbit in Group 1, which suggests an infection at an advanced stage or resolution of the lesions later on (21 DPI), as antigen was observed in the macrophages in the lamina propria (39).

The dose of bacterial challenge, administered orally as this is the natural infection route, differed slightly in the 2 infected groups, but this did not seem to have a major impact on results, as infec-tion was achieved in both groups. For both inocula, the bacterial virulence was verified by challenge of natural host (4 foals) with the same lot of L. intracellularis in a different study, in which foals developed clinical EPE, with severe manifestations (3 out of 4) and demise (1 out of 4) (40).

On gross inspection, the passage 9 inocula (Group 1) caused marked lesions and temporary weight loss, whereas the passage 10 inocula (Group 2) caused milder lesions and even milder clinical signs. Although studies in pig models show that increased (10-fold) concentrations of inoculum are related to increased severity of clinical signs, different virulence did not appear related to a 2-fold increase in concentration in the rabbit model (31). Moreover, severity of clinical appearance in this model could be related to a low passage isolate, which somehow compensated for a lower concentration of the inoculum. Although further investigation is needed, it is clear at this stage that a low passage isolate can cause lesions in rabbits.

This L. intracellularis cross-infection study shows fecal shedding in infected rabbits starting at 7 DPI and increasing by 14 DPI. As the TaqMan PCR technique detects the presence of genetic material whether it is viable or inactivated, it is interesting to note that fecal shedding in rabbits appears limited to the presence of active infec-tion (based on IHC lesions). Shedding was not detectable at 21 DPI, when clinical signs had already subsided and IHC labeling was drastically reduced. In the foal infection model, fecal shedding is reported to last longer than 21 DPI, but a specific relation between duration of fecal shedding, presence of lesions, and IHC labeling in the equine model is not available, as the foals were enrolled in a sur-vival study (25). In this initial experiment, the information gathered about serologic conversion is limited due to the short experimental timeline. Rabbits manifested a high serologic response on 14 DPI and even higher responses (. 1920) in both diseased groups by the final week. Such a high immune response showed that this EPE infection model in rabbits could be considered a “time-compressed” course of infection, compared to the equine EPE infection model, as such high titers were achieved in foals only by 28 or 35 DPI (25). Contrary to what is found in foals, a clinically significant hypoproteinemia

Table II. Immunohistochemistry (IHC) results in rabbits orally inoculated with L. intracellularis. Results are separated by group and by days post-infection (DPI) (7, 14, and 21)

Group 3 rabbits (control)a Group 1 rabbitsa Group 2 rabbitsa

Section 7 d 14 d 21 d 7 d 14 d 21 d 7 d 14 d 21 dDuodenum 0 0 0 0 0 0 0 0 0Mid jejunum 0 0 0 0 3 0 1 1 0End jejunum 0 0 0 1 1 0 2 2 0Ileum 0 0 0 0 1 0 2 1 0Cecum 0 0 0 2 2 1 2 1 0Appendix ceci 0 0 0 0 1 0 1 1 0Ileocecal valve 0 0 0 0 1 0 2 1 0Colon 0 0 0 1 2 0 2 2 0Terminal colon 0 0 0 0 0 0 0 0 0Rectum 0 0 0 0 0 0 0 0 0Note: Numbers in table refer to a 5-grade IHC scoring system. Grade 0 is equivalent to no antigen labeling detected (no shading), grade 1 (light gray shading) equivalent to 25% antigen labeled, grade 2 (gray shading) equivalent to 50%, grade 3 (dark gray shading) equivalent to 75%, and grade 4 (very dark gray shading) equivalent to 100% antigen labeled. a The groups consisted of 3 rabbits each, which were euthanized once weekly, 1 per group. Group 1 was infected orally with 1.3 3 108 bacteria/rabbit and Group 2 with 2.5 3 108 bacteria/rabbit. Group 3 was inoculated with medium only. Bacterial virulence of inocula was verified by challenge of natural host (4 foals) with the same lot of bacteria in different studies (25,40).



was not found in the rabbits of our model, nor was edema of the limbs or abdomen ever detected, although these sequelae would normally occur after 21 DPI. Further studies are needed with larger numbers of rabbits and over a longer time period to investigate the onset of hypoproteinemia and hypoalbuminemia in this rabbit model.

On the basis of clinical, diagnostic, gross pathology, and IHC find-ings, it is concluded that the incubation period in rabbits infected with an equine L. intracellularis isolate appears to be shorter than it is for pigs and hamsters challenged with a pig isolate or for the EPE infection model in foals (2,8,25). Moreover, intestinal lesions appeared to resolve by 21 DPI. These time factors could constitute an experimental advantage for the model in rabbits. Furthermore, infection is reproducible with no discomfort or overt disease. In pharmacology research, such a finding would represent an optimal model for drug preclinical testing. In addition, rabbits, like horses, are hindgut fermenters, and unlike laboratory rodents, offer the advantages of a larger body weight and blood volume and accessible blood vessels for collecting multiple blood samples. Further studies with this infection model could be used for pathogenesis and epide-miology studies, particularly when associated with such a remark-able serologic response. In fact, the ability to overcome infection quickly and without overt disease can represent an epidemiologic risk, or may explain the limited success of prevention measures, if the equine L. intracellularis strain can be carried sub-clinically by rabbits living on the grounds of a horse farm (1).

Finally, the success of this cross-infection experiment and the failure to generate infection in hamsters, which were previously the animal model of choice for proliferative enteropathies, suggest

a selective, secondary host adaptation of a given L. intracellularis strain. Further studies are necessary in order to define the details of such a discovery and its multiple implications.

A c k n o w l e d g m e n t sThe authors thank Dr. Colette Wheler at the Animal Resources

Centre, the support staff of the Animal Care Unit at Western College of Veterinary Medicine, University of Saskatchewan, Dr. Samantha M. Mapes for her contribution to PCR use and description, and Angela T. Gebhart for editing. Francesca Sampieri and Katherine R. Ball are fellow trainees of the Canadian Institutes of Health Research — Training Grant in Health Research Using Synchrotron Techniques (CIHR-THRUST). Fabio A. Vannucci was supported by the Brazilian Government sponsoring agency “Conselho Nacional de Desenvolvimento Cientifico e Tecnologico” (CNPq).

Re f e r e n c e s 1. Pusterla N, Mapes S, Rejmanek D, Gebhart C. Detection of

Lawsonia intracellularis by real-time PCR in the feces of free-living animals from equine farms with documented occurrence of equine proliferative enteropathy. J Wildl Dis 2008;44:992–998.

Figure 4. Serologic response detected in experimentally infected rab-bits and uninfected control rabbits after oral challenge with an equine strain of L. intracellularis. The rabbits were infected with a strain proven virulent in foals and did not show any circulating antibody until 14 DPI, which indicates that the immune response requires a minimum time to show and does not manifest equally in every individual (note the different titers in different animals).

Figure 5. Quantitative polymerase chain reaction (qPCR) results (rep-resented by L. intracellularis DNA gene copies/g feces) in rabbits experimentally infected with an equine strain of L. intracellularis and in uninfected controls. Note that the represented fecal shedding was obtained for each group, not for each individual. From these preliminary results, the fecal shedding of L. intracellularis after cross-infection from horses to rabbits can be noted as early as 7 DPI, but apparently resolves by 21 DPI.



2. Lawson GH, Gebhart CJ. Proliferative enteropathy. J Comp Pathol 2000;122:77–100.

3. McOrist S, Gebhart CJ, Boid R, Barns SM. Characterization of Lawsonia intracellularis gen. nov., sp. nov., the obligately intracel-lular bacterium of porcine proliferative enteropathy. Int J Syst Bacteriol 1995;45:820–825.

4. Cooper DM, Swanson DL, Barns SM, Gebhart CJ. Comparison of the 16S ribosomal DNA sequences from the intracellular agents of proliferative enteritis in a hamster, deer, and ostrich with the sequence of a porcine isolate of Lawsonia intracellularis. Int J Syst Bacteriol 1997;47:635–639.

5. Hotchkiss CE, Shames B, Perkins SE, Fox JG. Proliferative enteropathy of rabbits: The intracellular Campylobacter-like organism is closely related to Lawsonia intracellularis. Lab Anim Sci 1996;46:623–627.

6. Pusterla N, Gebhart C. Equine proliferative enteropathy caused by Lawsonia intracellularis. Equine Vet Educ 2009;8:415–418.

7. Wong DM, Alcott CJ, Sponseller BA, Young JL, Sponseller BT. Impaired intestinal absorption of glucose in 4 foals with Lawsonia intracellularis infection. J Vet Intern Med 2009;23:940–944.

8. Vannucci FA, Borges EL, de Oliveira JS, Guedes RM. Intestinal absorption and histomorphometry of Syrian hamsters (Mesocricetus auratus) experimentally infected with Lawsonia intracellularis. Vet Microbiol 2010;145:286–291.

9. Brees DJ, Sondhoff AH, Kluge JP, Andreasen CB, Brown CM. Lawsonia intracellularis-like organism infection in a miniature foal. J Am Vet Med Assoc 1999;215:511–514.

10. Williams NM, Harrison LR, Gebhart CJ. Proliferative enteropathy in a foal caused by Lawsonia intracellularis-like bacterium. J Vet Diagn Invest 1996;8:254–256.

11. Schumacher J, Schumacher J, Rolsma M, Brock KV, Gebhart CJ. Surgical and medical treatment of an Arabian filly with prolifera-tive enteropathy caused by Lawsonia intracellularis. J Vet Intern Med 2000;14:630–632.

12. Duhamel GE, Wheeldon EB. Intestinal adenomatosis in a foal. Vet Pathol 1982;19:447–450.

13. Frazer ML. Lawsonia intracellularis infection in horses: 2005–2007. J Vet Intern Med 2008;22:1243–1248.

14. Lavoie JP, Drolet R, Parsons D, et al. Equine proliferative enter-opathy: A cause of weight loss, colic, diarrhoea and hypopro-teinaemia in foals on three breeding farms in Canada. Equine Vet J 2000;32:418–425.

15. Lawson GH, McOrist S, Jasni S, Mackie RA. Intracellular bacteria of porcine proliferative enteropathy: Cultivation and mainte-nance in vitro. J Clin Microbiol 1993;31:1136–1142.

16. Jasni S, McOrist S, Lawson GH. Experimentally induced prolif-erative enteritis in hamsters: An ultrastructural study. Res Vet Sci 1994;56:186–192.

17. Stills HF, Jr. Isolation of an intracellular bacterium from hamsters (Mesocricetus auratus) with proliferative ileitis and reproduc-tion of the disease with a pure culture. Infect Immun 1991;59: 3227–3236.

18. Murakata K, Sato A, Yoshiya M, et al. Infection of different strains of mice with Lawsonia intracellularis derived from rabbit or porcine proliferative enteropathy. J Comp Pathol 2008;139: 8–15.

19. Stills HF, Jr, Hook RR, Jr. Experimental production of prolifera-tive ileitis in Syrian hamsters (Mesocricetus auratus) by using an ileal homogenate free of Campylobacter jejuni. Infect Immun 1989;57:191–195.

20. Stills HF, Fox JG, Plaster BJ, Dewhirst FE. A ‘new’ Clamydia sp strain SFPD isolated from transmissible proliferative ileitis in hamsters. Microb Ecoln Health Dis 1991;4:S99.

21. Stills HF, Jr, Hook RR, Jr, Sprouse RF. Utilization of monoclonal antibodies to evaluate the involvement of Campylobacter jejuni in proliferative ileitis in Syrian hamsters (Mesocricetis auratus). Infect Immun 1987;55:2240–2246.

22. Boutrup TS, Schauser K, Agerholm JS, Jensen TK. Application of a pig ligated intestinal loop model for early Lawsonia intracel-lularis infection. Acta Vet Scand 2010;52:17.

23. Boutrup TS, Boesen HT, Boye M, Agerholm JS, Jensen TK. Early pathogenesis in porcine proliferative enteropathy caused by Lawsonia intracellularis. J Comp Pathol 2010;143:101–109.

24. Jasni S, McOrist S, Lawson GH. Reproduction of proliferative enteritis in hamsters with a pure culture of porcine ileal symbi-ont intracellularis. Vet Microbiol 1994;41:1–9.

25. Pusterla N, Wattanaphansak S, Mapes S, et al. Oral infection of weanling foals with an equine isolate of Lawsonia intracellularis, agent of equine proliferative enteropathy. J Vet Intern Med 2010; 24:622–627.

26. Schoeb TR, Fox JG. Enterocecocolitis associated with intraepithe-lial Campylobacter-like bacteria in rabbits (Oryctolagus cunicu-lus). Vet Pathol 1990;27:73–80.

27. Horiuchi N, Watarai M, Kobayashi Y, Omata Y, Furuoka H. Proliferative enteropathy involving Lawsonia intracellularis infection in rabbits (Oryctlagus cuniculus). J Vet Med Sci 2008; 70:389–392.

28. Watarai M, Yoshiya M, Sato A, Furuoka H. Cultivation and characterization of Lawsonia intracellularis isolated from rabbit and pig. J Vet Med Sci 2008;70:7731–7733.

29. Harcourt-Brown F. Biological characteristics of the domestic rab-bit (Oryctolagus cuniculi). In: Harcourt-Brown F, ed. Textbook of Rabbit Medicine. Woburn: Elsevier Health Sciences, 2002:1–18.

30. Guedes RM, Gebhart CJ. Comparison of intestinal mucosa homogenate and pure culture of the homologous Lawsonia intracellularis isolate in reproducing proliferative enteropathy in swine. Vet Microbiol 2003;93:159–166.

31. Guedes RM, Winkelman NL, Gebhart CJ. Relationship between the severity of porcine proliferative enteropathy and the infec-tious dose of Lawsonia intracellularis. Vet Rec 2003;153:432–433.

32. Wattanaphansak S, Gebhart CJ, Anderson JM, Singer RS. Development of a polymerase chain reaction assay for quan-tification of Lawsonia intracellularis. J Vet Diagn Invest 2010;22: 598–602.

33. Johnson-Delaney C. Anatomy and physiology of the rabbit and rodent gastrointestinal system. 2006 Proc Assoc Avian Vet 2006:9–17.

34. Guedes RM, Gebhart CJ. Preparation and characterization of polyclonal and monoclonal antibodies against Lawsonia intracel-lularis. J Vet Diagn Invest 2003;15:438–446.

35. Guedes RM, Gebhart CJ, Deen J, Winkelman NL. Validation of an immunoperoxidase monolayer assay as a serologic test for



porcine proliferative enteropathy. J Vet Diagn Invest 2002;14: 528–530.

36. Pusterla N, Jackson R, Wilson R, Collier J, Mapes S, Gebhart C. Temporal detection of Lawsonia intracellularis using serology and real-time PCR in Thoroughbred horses residing on a farm endemic for equine proliferative enteropathy. Vet Microbiol 2009;136:173–176.

37. Leutenegger CM, Mislin CN, Sigrist B, Ehrengruber MU, Hoffmann-Lehmann R, Lutz H. Quantitative real-time PCR for the measurement of feline cytokine mRNA. Vet Immun Immunopath 1999;71:291–305.

38. Duhamel GE, Klein EC, Elder RO, Gebhart CJ. Subclinical prolif-erative enteropathy in sentinel rabbits associated with Lawsonia intracellularis. Vet Pathol 1998;35:300–303.

39. Jensen TK, Møller K, Leser TD, Jorsal SE. Comparison of histol-ogy, immunohistochemistry and polymerase chain reaction for detection of Lawsonia intracellularis in natural porcine prolifera-tive enteropathy. Eur J Vet Pathol 1997;3:115–123.

40. Pusterla N, Vannucci FA, Mapes SM, et al. Efficacy of an aviru-lent live vaccine against Lawsonia intracellularis in the prevention of proliferative enteropathy in experimentally infected weanling foals. Am J Vet Res 2012;73:741–746.


Article


I n t r o d u c t i o nVitamin E is an integral component of all lipid membranes and

protects them from attack by reactive oxygen species (ROS) (1,2). Polyunsaturated fatty acids of membranes are vulnerable to attack by ROS, which can initiate a chain reaction of lipid destruction that destroys the membrane of the cell. Vitamin E can quench peroxida-tion reactions in membranes and is probably the most important antioxidant present in cell membranes (3). The newborn dairy calf has minimal reserves of lipid-soluble vitamins and fetal uptake is limited (4,5). Vitamin E supplementation has been shown to have positive effects on the immune system of young dairy calves (6,7). We have reported that providing vitamin E supplementation to

newborn calves activates killing ability and intracellular signal transduction, including protein kinase C, intracellular Ca2, and tyrosine kinase of neutrophils (8). It has been reported that cellular a-tocopherol concentrations of red blood cells were approximately 5 times higher in vitamin E-supplemented Holstein calves than in control calves (9). However, levels of alpha (a)-tocopherol concentra-tion in neutrophils from calves have not yet been examined.

The cellular uptake of a-tocopherol within high-density lipopro-tein (HDL) is reported to be mediated by scavenger receptor class B, type I (SR-BI) (10,11). Moreover, a previous study using SR-BI-deficient mice has shown a deficiency in the uptake of a-tocopherol from plasma lipoproteins to specific tissues (12). Rajapaksha et al (13) sequenced the bovine SR-BI and examined changes in expression

Effects of vitamin E supplementation on cellular a-tocopherol concentrations of neutrophils in Holstein calves

Hidetoshi Higuchi, Erina Ito, Hidetoma Iwano, Shin Oikawa, Hajime Nagahata

A b s t r a c tThe effects of vitamin E supplementation on cellular a-tocopherol concentrations of neutrophils from Holstein calves and the mechanism of scavenger receptor class B type I (SR-BI)-mediated uptake of a-tocopherol were examined. Cellular a-tocopherol concentrations in vitamin E-treated calves increased from 3.5 6 0.38 to 7.2 6 0.84 μg/107 cells, respectively, within 14 d after vitamin E supplementation; these concentrations were significantly higher than those of control calves (P , 0.01). The expression indices of SR-BI [a major receptor that recognizes high-density lipoprotein (HDL)] mRNA in neutrophils were two to five times higher (P , 0.01) in neutrophils obtained from vitamin E-supplemented calves compared with those from control calves, and anti-SR-B1 antibody, ranging from 0.1 to 1.0 μg/mL, significantly (P , 0.01) decreased cellular a-tocopherol concentrations of neutrophils. Cytochalasin D and latrunculin B, major inhibitors of actin polymerization of neutrophils, significantly decreased cellular a-tocopherol concentrations of neutrophils (P , 0.01). Our results demonstrated that in vitamin E-supplemented calves: 1) a-tocopherol is mainly distributed with HDL, 2) a-tocopherol within HDL is recognized by SR-BI on the surface of neutrophils, and 3) rearrangement of the actin cytoskeleton is a crucial step for the uptake of a-tocopherol by neutrophils.

R é s u m éOn a examiné les effets d’un supplément de vitamine E sur les concentrations cellulaires d’a-tocophérol des neutrophiles provenant de veaux Holstein et le mécanisme de prise d’a-tocophérol médié par les récepteurs ramasseurs de classe B type I (SR-BI). Les concentrations cellulaires d’a-tocophérol chez les veaux traités avec de la vitamine E ont augmenté de 3,5 6 0,38 à 7,2 6 0,84 mg/107 cellules, respectivement, à l’intérieur d’un délai de 14 j après une supplémentation en vitamine E; ces concentrations étaient significativement plus élevées que celles des veaux témoins (P , 0,01). Les taux d’expression d’ARNm de SR-BI [un récepteur majeur qui reconnaît les lipoprotéines de haute-densité (HDL)] dans les neutrophiles étaient 2 à 5 fois plus élevés (P , 0,01) dans les neutrophiles obtenus de veaux ayant reçu un supplément de vitamine E comparativement à ceux des veaux témoins, et des anticorps anti-SR-B1, allant de 0,1 à 1,0 mg/mL, ont réduit significativement (P , 0,01) les concentrations cellulaires d’a-tocophérol des neutrophiles. La cytochalasine D et la latrunculine B, des inhibiteurs majeurs de la polymérisation de l’actine des neutrophiles, ont diminué de manière significative les concentrations cellulaires d’a-tocophérol des neutrophiles (P , 0,01). Nos résultats ont démontré que chez les veaux recevant un supplément de vitamine E : 1) l’a-tocophérol est principalement distribué avec les HDL, 2) l’a-tocophérol dans les HDL est reconnu par les SR-BI sur la surface des neutrophiles, et 3) le réarrangement du cytosquelette d’actine est une étape cruciale pour la prise d’a-tocophérol par les neutrophiles.


Animal Health Laboratory (Higuchi, Ito, Nagahata), Department of Veterinary Herd Health (Oikawa), Department of Health and Environmental Sciences and Veterinary Biochemistry, and Department of Bioscience (Iwano), School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, Hokkaido 069-8501, Japan.

Address all correspondence to Dr. Hidetoshi Higuchi; telephone/fax: 81-11-388-4844; e-mail: [email protected]

Received October 31, 2011. Accepted May 13, 2012.



of messenger ribonucleic acid (mRNA) of SR-BI during corpus luteum development in vivo and granulosa cell luteinization in vitro. Sequencing of the bovine HDL-receptor showed that it codes for a protein of 509 amino acids with close identity to hamster, mouse, rat, and human sequences (13). Examination of the tissue distribu-tion of the HDL-receptor mRNA showed high levels in adrenal cortex and corpus luteum and lower levels in spleen and liver (13). However, expression levels of SR-BI mRNA and the mechanism for SR-BI-mediated uptake of a-tocopherol have not been clarified with bovine neutrophils. It has been reported that cytochalasin D, latrunculin B, and jasplakinolide are chemical reagents for actin polymerization of neutrophils and were effective in determining if the rearrangement of the neutrophil cytoskeleton is related to the uptake of a-tocopherol (14–16).

In this study, we examined the effects of vitamin E supplementa-tion on the cellular a-tocopherol concentrations of neutrophils from Holstein calves and the mechanism for SR-BI-mediated uptake of a-tocopherol.


CalvesFourteen newborn Holstein calves were randomly divided into

2 groups of 7 calves each. Each calf was given 4 L of the same colos-trum within 2 h of birth and then injected intramuscularly with physiological saline (control calves) or 25 IU/kg body weight (BW) of vitamin E (dl-a-tocopherol; Nihon Zenyaku, Fukushima, Japan) (vitamin E-supplemented calves) 48 h after birth. The control calves were also used for preparing neutrophils 7 d after physiological saline injection (9 d of age) to examine the effects of anti-SR-B1 anti-body, cytochalasin D, latrunculin B, and jasplakinolide on cellular a-tocopherol concentrations in vitro. All experimental procedures complied with the guidelines of the Committee for Animal Welfare

at Rakuno Gakuen University, which are based on the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, 1996).

Blood samplingThirty milliliters of blood was drawn from the jugular vein of

each calf into a tube containing heparin (20 IU/mL) for neutrophil preparation and into a plain tube for serum preparation. Blood samples were collected on the day of supplementation (0 d) and 1, 3, 7, 14, 21, and 28 d after supplementation.

Isolation of neutrophilsNeutrophils were isolated from heparinized blood by Ficoll-

Conray solution (specific gravity 1.078), followed by hypotonic red blood cell lysis, as described in a previous paper (17). Neutrophils were resuspended in Hanks’ balanced salt solution containing Ca2 and Mg2 (HBSS; Nissui Pharmaceutical, Tokyo, Japan) to a concen-tration of 1 3 107 cells/mL. The resulting cell population was made up of . 95% neutrophils, as determined by Wright-Giemsa staining, and . 99% of the cells were viable when assessed by trypan blue dye exclusion.

Lipoprotein separationSerum collected from both groups was used immediately

for lipoprotein separation without freezing. Lipoprotein sepa-ration was done based on the method described in a previous paper (18) using a fixed-angle rotor (TLA-110; Beckman Coulter, Fullerton, California, USA) in an ultracentrifuge (OPTIMA TLX; Beckman Coulter) at 16°C. Chylomicrons (CM; 117 000 3 g for 10 min, d , 0.95), very low-density lipoprotein (VLDL; 657 000 3 g for 2 h and 55 min, d , 1.006), low-density lipoprotein (LDL; 657 000 3 g for 4 h and 20 min, d , 1.063), and high-density lipoprotein (HDL; 657 000 3 g for 7 h and 15 min, d , 1.21) were prepared.

Figure 1. Serum and cellular a-tocopherol concentrations in control and vitamin E-supplemented calves. Values are expressed as mean 6 standard error of the mean (SEM). control calves; vitamin E-supplemented calves. *Significantly different (P , 0.01) from values in control calves.

Serum

Days after supplementation Days after supplementation

a-t

ocop

hero

l (m

g/10

7 cel

ls)

a-t

ocop

hero

l (m

g/dL

)Neutrophils



Concentrations of a-tocopherolConcentrations of a-tocopherol in serum, lipoproteins, and neu-

trophils were measured by high performance liquid chromatography (HPLC), as described by Ametaj et al (19) and Campbell et al (20).

SR-BI mRNA levels of neutrophilsNeutrophils isolated from both control and vitamin E-supplemented

calves were analyzed for SR-BI mRNA expression. Reverse transcription- polymerase chain reaction (RT-PCR) amplification of SR-BI and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA as a housekeeping gene were carried out as previously described by Rajapaksha et al (13). The expression index of SR-B1 was calculated by dividing the quantity of SR-B1 mRNA by the quantity of GAPDH mRNA.

Effect of anti-SR-BI antibody on a-tocopherol concentrations of neutrophils

Neutrophils isolated from 7 control calves (7 d after physiologi-cal saline supplementation, 9 d of age) were used to examine the effects of anti-SR-B1 antibody on cellular a-tocopherol concentra-tions of neutrophils. Neutrophils (2 3 107/mL) were pre-incubated with 0.01 to 10 mg/mL of anti-SR-BI antibody (Novus Biologicals, Littleton, Colorado, USA) at 37°C for 5 min, and treated with bovine serum containing 800 mg/mL of a-tocopherol at 37°C for 30 min. Cellular a-tocopherol concentrations of neutrophils were measured as previously described.

Effects of cytochalasin D, latrunculin B, and jasplakinolide on cellular a-tocopherol concentrations of neutrophils in vitro

Neutrophils isolated from 7 control calves (7 d after physiologi-cal saline supplementation, 9 d of age) were used to examine the effects of cytochalasin D, latrunculin B, and jasplakinolide on cel-lular a-tocopherol concentrations of neutrophils. Cytochalasin D, latrunculin B, and jasplakinolide were dissolved in HBSS at final concentrations of 0.01 to 1 mg/mL (14–16). Neutrophils (2 3 107/mL) were pre-incubated with 0.1 to 1 mg/mL each of cytochalasin D, latrunculin B, and jasplakinolide at 37°C for 5 min, and treated with bovine serum containing 800 mg/mL of a-tocopherol at 37°C for 30 min. Cellular a-tocopherol concentrations of neutrophils were measured as previously described.

StatisticsAll data were expressed as mean 6 standard error of the mean

(SEM). Groups were compared for statistical difference by using Student’s t-test with Bonferroni correction for changes in serum and cellular a-tocopherol concentrations, a-tocopherol concentrations of lipoprotein fractions, and gene expression of SR-B1 of neutrophils. One-way analysis of variance (ANOVA), followed by Tukey’s test, was used to study the effect of anti-SR-BI antibody, cytochalasin D, latrunculin B, and jasplakinolide treatment on cellular a-tocopherol concentrations. Values of P , 0.05 were regarded as significant.

Re s u l t s

Serum and cellular a-tocopherol concentrationsChanges in serum and cellular a-tocopherol concentrations

of neutrophils for both control calves and those given vitamin E are shown in Figure 1. Serum a-tocopherol concentrations in the vitamin E-supplemented calves increased from 352.2 6 29.7 to 805.6 6 30.0 mg/dL within 14 d after vitamin E supplementa-tion. These concentrations were significantly higher (P , 0.01) than those in control calves, which ranged from 186.2 6 26.2 to 235.6 6 23.9 mg/dL. Similarly, cellular a-tocopherol concentrations of neutrophils isolated from vitamin E-supplemented calves increased from 3.5 6 0.38 to 7.2 6 0.84 mg/107 cells within 14 d after vita-min E supplementation and these concentrations were significantly higher (P , 0.01) than those in control calves, which ranged from 1.8 6 0.42 to 2.6 6 0.46 mg/107 cells.

Figure 2. Concentrations of a-tocopherol in lipoprotein fractions obtained from control and vitamin E-supplemented calves at 0 h and 24 h after supplementation. Values are expressed as mean 6 SEM. control calves; vitamin E-supplemented calves. *Significantly different (P , 0.01) than values in the control calves.

Lipoprotein fraction

Lipoprotein fraction

a-t

ocop

hero

l (m

g/dL

)a

-toc

ophe

rol (

mg/

dL)



Concentrations of a-tocopherol in lipoprotein fractions

Changes in concentrations of a-tocopherol in each lipoprotein fraction at 0 h and 24 h after vitamin E supplementation are shown in Figure 2. Twenty-four hours after vitamin E supplementation, a-tocopherol concentrations of HDL fractions (603.2 6 27.2 mg/dL) were significantly higher (P , 0.01) than those of control calves. No significant differences of CM, VLDL, and LDL fractions were detected between control and vitamin E-supplemented calves.

Gene expression of SR-B1 of neutrophilsData on SR-BI mRNA expression of neutrophils obtained from

control and vitamin E-supplemented calves at 0 to 21 d after vita-min E supplementation are shown in Figure 3. The expression indi-ces of SR-B1 mRNA levels of vitamin E-supplemented calves were approximately 1.5 to 5 times higher (P , 0.01) than those of control calves at 1, 3, 7, and 14 d after vitamin E supplementation.

Effect of anti-SR-BI antibody on cellular a-tocopherol concentrations

Data on the effect of anti-SR-BI treatment on cellular a-tocopherol concentrations of neutrophils are shown in Figure 4. Cellular a-tocopherol concentrations of neutrophils were significantly (P , 0.01) decreased after anti-SR-BI treatment, with concentrations ranging from 0.1 to 10 mg/mL.

Effects of cytochalasin D, latrunculin B, and jasplakinolide treatment on cellular a-tocopherol concentrations of neutrophils

Data on the effects of cytochalasin D, latrunculin B, and jas-plakinolide treatment for cellular a-tocopherol concentrations of neutrophils in vitro are shown in Figure 5. Cellular a-tocopherol concentrations of neutrophils treated with 1 mg/mL of cytochala-

sin D were significantly decreased (P , 0.01). Similarly, latrunculin B at concentrations of 0.1 and 1 mg/mL also significantly decreased (P , 0.01) the cellular a-tocopherol concentrations of neutrophils. In contrast, jasplakinolide at concentrations of 0.1 and 1 mg/mL significantly enhanced (P , 0.01) cellular a-tocopherol concentra-tions of neutrophils.

D i s c u s s i o nVitamin E is a potent, lipid-soluble, chain-breaking antioxidant

that maintains membrane integrity by reducing oxidation of poly-unsaturated fatty acids in membranes (1–3). We have previously shown that vitamin E supplementation enhanced killing activity and modified intracellular signal transduction of neutrophils from Holstein calves (8). The effect of vitamin E supplementation on cel-lular a-tocopherol concentrations of neutrophils, however, has not been examined. The present study showed that cellular a-tocopherol concentrations of neutrophils obtained from vitamin E-supplemented calves were significantly higher than those in control calves, similar to the changes in serum a-tocopherol concentrations. Our results indicated that changes in cellular a-tocopherol concentrations of neutrophils were paralleled with that of serum a-tocopherol concen-trations. Roquet et al (9) reported that cellular a-tocopherol concen-trations of red blood cells from Holstein calves were approximately 5 times higher in vitamin E-supplemented calves than in control calves. However, cellular a-tocopherol concentrations of neutrophils in calves have not been reported. This is the first report to describe cellular a-tocopherol concentrations of neutrophils from Holstein calves and their changes with vitamin E supplementation.

We examined the distribution of supplemented a-tocopherol in each lipoprotein fraction after vitamin E supplementation and found that a-tocopherol concentrations of HDL fractions were approximately 2 times higher than those of the control calves.

Figure 3. Data on SR-BI mRNA expression of neutrophils obtained from control and vitamin E-supplemented calves at 0 to 21 d after vitamin E supplementation. Values are expressed as mean 6 SEM. control calves

vitamin E-supplemented calves. *Significantly different (P , 0.01) than values in the control calves.

Days after supplementation

Exp

ress

ion

inde

x

Figure 4. Effect of anti-SR-BI antibody on cellular a-tocopherol concen-trations of neutrophils in vitro. Values are expressed as mean 6 SEM. *Significantly different (P , 0.01) than values in mg/mL of anti-SR-BI antibody.

a-t

ocop

hero

l (m

g/10

7 cel

ls)

Anti-SR-BI concentration (mg/mL)



No obvious differences were observed in a-tocopherol concentra-tions in other lipoprotein fractions isolated from control and vita-min E-supplemented calves. Our results suggest that a- tocopherol is mainly distributed in HDL fractions isolated from vita-min E-supplemented calves. These results were similar to previous findings, which showed that 80% to 95% of total plasma a-tocopherol is associated with HDL among mouse plasma lipoprotein (21,22).

It has been reported that endothelial cells and enterocytes have a scavenger receptor class B, type I (SR-BI) to recognize HDL, which is important for the uptake of HDL-associated lipids (10,11). Rajapaksha et al (13) sequenced the bovine SR-BI (HDL receptor) and examined the changes in expression of its mRNA during corpus

luteum development in vivo and granulosa cell luteinization in vitro. Sequencing of the bovine SR-BI showed that it codes for a protein of 509 amino acids with close identity to hamster, mouse, rat, and human sequences (13). Examination of the tissue distribution of the SR-BI mRNA showed high levels in the adrenal cortex and corpus luteum and lower levels in the spleen and liver. Using a semi- quantitative RT-PCR, levels of SR-BI mRNA were measured in corpora lutea from cattle at known stages of the estrus cycle and in bovine granulosa cells luteinized in culture (13). This evidence prompted us to hypothesize that SR-BI plays a key role in the uptake of HDL-associated a-tocopherol with bovine neutrophils. In this study, we have clarified that the expression indices of the SR-BI mRNA levels of vitamin E-treated calves were 1.5 to 5 times higher than those in control calves. Eggermont (23) reported that a-tocopherol is not only an antioxidant, but also regulates gene expression by binding to nuclear receptors. The precise mechanism for regulating gene expression, however, is still unknown. Our results clarified that a-tocopherol is one of the factors regulating the expression of SR-BI mRNA in bovine neutrophils.

We investigated whether SR-BI was functionally involved in the uptake of HDL-associated a-tocopherol with neutrophils. Cellular a-tocopherol concentrations were significantly decreased after anti-SR-BI treatment, ranging from 0.1 to 10 mg/mL. These results were similar to previous findings that determined the effects of SR-BI on vitamin E uptake of the enterocyte (10). Our results suggest that SR-BI is a crucial surface receptor that allows bovine neutrophils to recognize HDL-associated a-tocopherol.

Cytochalasin D and latrunculin B are major inhibitors of actin polymerization and depress phagocytosis of monocytes and neutrophils (14,15), while jasplakinolide is an enhancer of actin polymerization of neutrophils (15,16). We investigated the effects of cytochalasin D, latrunculin B, and jasplakinolide on cellular a-tocopherol concentrations of neutrophils to determine if the rearrangement of the actin cytoskeleton of neutrophils was related to the uptake of a-tocopherol. Cellular a-tocopherol concentra-tions of neutrophils treated with cytochalasin D were significantly decreased at concentrations of 1 mg/mL. Similarly, latrunculin B at concentrations of 0.1 and 1 mg/mL also significantly decreased cellular a-tocopherol concentrations of neutrophils. In contrast, jasplakinolide at concentrations of 0.1 and 1 mg/mL significantly increased cellular a-tocopherol concentrations of neutrophils. The final concentrations of these chemicals that induced changes in cellular a-tocopherol concentrations of neutrophils were similar to previous findings (14–16). These results suggested that the actin polymerization of the cell membrane is an important process in the uptake of a-tocopherol by bovine neutrophils.

In summary, in vitamin E-supplemented calves: 1) a-tocopherol is mainly distributed with HDL; 2) a-tocopherol within HDL is recog-nized by SR-BI on the surface of neutrophils; 3) actin polymerization of cell membranes is a crucial step for the uptake of a-tocopherol by neutrophils; and 4) expression of SR-BI mRNA may be regulated by a-tocopherol. This is the first report to describe the mechanism of the uptake of a-tocopherol of neutrophils from Holstein calves. Our results should stimulate further study aimed at determining the mechanisms by which a-tocopherol supplementation enhance neutrophil functions.

Figure 5. Effect of cytochalasin D, latrunculin B, and jasplakinolide treatment on cellular a-tocopherol concentrations of neutrophils in vitro. Values are expressed as mean 6 SEM. *Significantly different (P , 0.01) than concentrations of mg/mL of each inhibitor.

Jasplakinolide (mg/mL)

a-t

ocop

hero

l (m

g/10

7 cel

ls)

a-t

ocop

hero

l (m

g/10

7 cel

ls)

a-t

ocop

hero

l (m

g/10

7 cel

ls)

Latrunclin B (mg/mL)

Cytochalasin D (mg/mL)

Cytochalasin D

Latrunculin B

Jasplakinolide



Re f e r e n c e s1. Rice DA, Kennedy S. Assessment of vitamin E, selenium and

polyunsaturated fatty acid interactions in the aetiology of dis-ease in the bovine. Proc Nutr Soc 1988;47:177–184.

2. Burton GW, Traber MG. Vitamin E: Antioxidant activity, bio-kinetics, and bioavailability. Annu Rev Nutr 1990;10:357–382.

3. Putnam ME, Comben N. Vitamin E. Vet Rec 1987;121:541–545.4. Malone JI. Vitamin passage across the placenta. Clin Perinatol

1975;2:295–307.5. Van Saun RJ, Herdt TH, Stowe HD. Maternal and fetal vitamin E

concentrations and selenium-vitamin E interrelationship in dairy cattle. J Nutr 1989;119:1156–1164.

6. Cipriano JE, Morrill JL, Anderson NV. Effect of dietary vitamin E on immune responses of calves. J Dairy Sci 1982;65:2357–2365.

7. Eicher-Pruiett SD, Morrill JL, Blecha F, Higgins JJ, Anderson NV, Reddy PG. Neutrophil and lymphocyte response to supple-mentation with vitamins C and E in young calves. J Dairy Sci 1992;75:1635–1642.

8. Higuchi H, Nagahata H. Effects of vitamins A and E on super-oxide production and intracellular signaling of neutrophils in Holstein calves. Can J Vet Res 2000;64:69–75.

9. Roquet J, Nockels CF, Papas AM. Cattle blood plasma and red blood cell a-tocopherol levels in response to different chemical forms and routes of administration of vitamin E. J Anim Sci 1992; 70:2542–2550.

10. Reboul E, Klein A, Bietrix F, et al. Scavenger receptor class B type I (SR-BI) is involved in vitamin E transport across the enterocyte. J Biol Chem 2006;281:4739–4745.

11. Tachikawa M, Okayasu S, Hosoya K. Functional involvement of scavenger receptor class B, type I, in the uptake of alpha-tocopherol using cultured rat retinal capillary endothelial cells. Mol Vis 2007;13:2041–2047.

12. Trigatti B, Rayburn H, Viñals M, et al. Influence of the high den-sity lipoprotein receptor SR-BI on reproductive and cardiovascu-lar pathophysiology. Proc Natl Acad Sci USA 1999;96:9322–9327.

13. Rajapaksha WR, McBride M, Robertson L, O’Shaughnessy PJ. Sequence of the bovine HDL-receptor (SR-BI) cDNA and changes

in receptor mRNA expression during granulosa cell luteinization in vivo and in vitro. Mol Cell Endocrinol 1997;134:59–67.

14. Gavrilin MA, Bouakl IJ, Knatz NL, et al. Internalization and phagosome escape required for Francisella to induce human monocyte IL-1beta processing and release. Proc Natl Acad Sci USA 2006;103:141–146.

15. Hwang I, Sprent J. Role of the actin cytoskeleton in T cell absorption and internalization of ligands from APC. J Immunol 2001;166:5099–5107.

16. Bengtsson T, Orselius K, Wetterö J. Role of the actin cytoskeleton during respiratory burst in chemoattractant-stimulated neutro-phils. Cell Biol Int 2006;30:154–163.

17. Higuchi H, Nagahata H. Comparison of superoxide production, protein kinase C and tyrosine kinase activities in neutrophils from neonatal calves and cows. Res Vet Sci 1998;65:139–143.

18. Oikawa S, Mizunuma Y, Iwasaki Y, Tharwat M. Changes of very low-density lipoprotein concentration in hepatic blood from cows with fasting-induced hepatic lipidosis. Can J Vet Res 2010;74:317–320.

19. Ametaj BN, Nonnecke BJ, Franklin ST, et al. Dietary vita-min A modulates the concentrations of RRR-alpha-tocopherol in plasma lipoproteins from calves fed milk replacer. J Nutr 2000;130:629–636.

20. Campbell SE, Stone WL, Lee S, et al. Comparative effects of RRR-alpha- and RRR-gamma-tocopherol on proliferation and apop-tosis in human colon cancer cell lines. BMC Cancer 2006;6:13.

21. Sattler W, Levak-Frank S, Radner H, Kostner GM, Zechner R. Muscle-specific overexpression of lipoprotein lipase in trans-genic mice results in increased alpha-tocopherol levels in skeletal muscle. Biochem J 1996;318:15–19.

22. Kostner GM, Oettl K, Jauhiainen M, Ehnholm C, Esterbauer H, Dieplinger H. Human plasma phospholipid transfer protein accelerates exchange/transfer of alpha-tocopherol between lipoproteins and cells. Biochem J 1995;305:659–667.

23. Eggermont E. Recent advances in vitamin E metabolism and deficiency. Eur J Pediatr 2006;165:429–434.


Article


I n t r o d u c t i o nControl of the interval from parturition to subsequent conception

is crucial to the optimal reproductive rate for a species. The remating program [interval from kindling (parturition) to mating] most widely used in the rabbit industry is the semi-intensive rhythm, which is 11 to 12 d postpartum. This should represent a compromise between the doe’s need for recovering energy after parturition and the eco-nomic demand for increased numbers of kits weaned per year. Since

a negative energy balance can be detrimental to the reproductive process (1), one of the main reasons for lengthening the remating interval is to prolong the dry period so that the doe, especially if primiparous, recovers body energy completely (2–4).

Ovulation does not occur spontaneously in rabbits: it is a neuro-endocrine reflex triggered by copulation. Even though rabbit does can be mated just after kindling and be concurrently pregnant and lactating, their reproductive efficiency varies considerably with par-ity (the number of times the doe has kindled), physiological state

Preliminary study on factors influencing rabbit doe reproductive efficiency: Effect of parity, day of mating, and suckling on ovarian status

and estrogen levels at day 6 of pregnancyMaria Laura Marongiu, Corrado Dimauro

A b s t r a c tThe rabbit corpus luteum becomes an estradiol-dependent tissue by day 6 of gestation, and adequate estrogen is critical to avoid pregnancy failure. The aim of this study was to investigate the effect of parity (primiparous or multiparous), day of mating (11 or 21 d postpartum), and suckling status (suckling or nonsuckling) on various reproductive traits in hybrid rabbit does (n = 96). Ovarian structures on day 6 after coitus were evaluated by means of ultrasonography. Blood samples were collected that day, and the serum was analyzed for estradiol-17b by radioimmunoassay (RIA). Parity and suckling had significant effects on mating rate (P , 0.01 and P , 0.05, respectively). More does accepted the male on day 11 than on day 21 (P , 0.05). Ovulation frequency was significantly affected by parity (P , 0.05), day of mating (P , 0.01), and suckling (P , 0.01). Fewer ovarian large follicles and lower estradiol-17b levels were detected in suckling compared with nonsuckling rabbits (P , 0.01). Since estrogen concentrations are commonly used to assess follicular growth and steroidogenic capacity, the lower hormonal levels in the suckling rabbits may reveal poorer ovarian activity, which could result in reduced reproductive efficiency. Our observations confirm the existence of a partial antagonism between lactation and reproduction in rabbits. Further research is needed to elucidate these phenomena, including when artificial insemination is done. Ultrasonography could represent a noninvasive and reliable method for studying several reproductive functions and dysfunctions in rabbits.

R é s u m éChez la lapine le corps jaune devient un tissu dépendant de l’œstradiol au 6e jour de la gestation, et une quantité adéquate d’estrogène est critique pour éviter l’interruption de la gestation. L’objectif de l’étude était d’examiner, chez des lapines hybrides (n = 96), l’effet de la parité (primipare ou multipare), le jour de l’accouplement (11 ou 21 jours post-partum), et si elles allaitaient ou non, sur différentes caractéristiques de reproduction. Les structures ovariennes au jour 6 après le coït ont été évaluées par échographie. Des échantillons de sang ont été prélevés la même journée et le sérum analysé par radio-immunoessai (RIA) pour l’œstradiol-17b. La parité et le fait d’allaiter avaient un effet significatif sur le taux de conception (respectivement P , 0,01 et P , 0,05). Plus de lapines étaient réceptives au mâle au jour 11 comparativement au jour 21 (P , 0,05). La fréquence d’ovulation était affectée de manière significative par la parité (P , 0,05), le jour de l’accouplement (P , 0,01) et le fait d’allaiter (P , 0,01). Moins de larges follicules ovariens et des niveaux plus faibles d’œstradiol-17b ont été trouvés chez les lapines qui allaitaient comparativement à celles qui ne le faisaient pas (P , 0,01). Étant donné que les concentrations d’œstrogène sont fréquemment utilisées pour évaluer la croissance folliculaire et la capacité de stéroidogénèse, les niveaux plus faibles d’hormones chez les lapines qui allaitent peuvent révéler une activité ovarienne plus pauvre, qui pourrait se traduire par une efficacité reproductrice moindre. Nos observations confirment l’existence d’un antagonisme partiel entre la lactation et la reproduction chez les lapins. De la recherche supplémentaire est nécessaire afin d’élucider ces phénomènes, incluant lorsque l’insémination artificielle est utilisée. Chez les lapins, l’échographie pourrait s’avérer une méthode non-invasive et fiable pour étudier plusieurs fonctions reproductrices ainsi que des dysfonctions.


Dipartimento di Agraria, Sezione di Scienze Zootecniche, Università di Sassari, Via De Nicola 9, 07100 Sassari, Italia.

Address all correspondence to Dr. Maria Laura Marongiu; telephone: 39 079 229304; fax: 39 079 229302; e-mail: [email protected]




(whether lactating or not and the stage of lactation), and sexual receptivity at mating (5). The current study aimed to investigate the effects on various aspects of rabbit reproduction of parity, suckling, and remating interval. The ovarian structures were examined by means of ultrasound scanning, which allowed the experiment to be conducted entirely in vivo and by a noninvasive procedure.

In previous studies of the relationship between different mana-gerial strategies and rabbit ovarian function (ovulation response, ovulation rate, fertilization rate, embryo survival before and after implantation, pregnancy rate), the females were slaughtered and their reproductive tracts removed and dissected (6) or, alternatively, the ovaries were recovered by laparotomy (7). Even though ultraso-nography has been used for pregnancy diagnosis (8), to characterize fetal growth (9), and to evaluate the maternal and fetal vessels (10), there is a paucity of information in the literature regarding specific study of the rabbit ovaries by this method.

In this work special focus was given to the ovarian population of large follicles (LF; diameter greater than 1.5 mm) present at day 6 of pregnancy because this follicular category, being steroidogenically active, should be crucial to survival of the developing corpora lutea (CL) and thus to reproductive efficiency. Blood estradiol levels in plasma were measured to evaluate the steroidogenic capability of the follicles through the first days of pregnancy.


Animals and experimental designThe 96 hybrid rabbit does used in the study were housed in indi-

vidual metal cages under controlled light conditions (14 h of light, 10 h of darkness), were fed ad libitum a commercial pelleted diet, and had free access to tap water. The protocols involving the care and use of the animals were approved by the Bioethics Committee of the University of Sassari, Sassari, Italy.

The 50 primiparous and 46 multiparous does were studied in a 2 3 2 3 2 factorially designed experiment. The variables considered were parity, day of attempted mating [day 11 (n = 46) or day 21 (n = 50)] after parturition, and whether the doe suckled the litter after the day of parturition [S (n = 46)] or did not [N (n = 50)]. The time that a doe was exposed to a given male (of proven fertility) ranged from 60 to 120 s. If copulation occurred, the female was removed immediately and classed as mating. A nonmating doe was defined as one that had been exposed to 6 males for a total of about 12 min. Mating performance was assessed in the morning and after controlled lactation (1 period of suckling per day for 15 min).

The CL and follicle populations of the mated females (S and N) were evaluated by means of transabdominal real-time B-mode ultrasound scanning on day 6 after coitus, which corresponded to postpartum days 17 (groups 11-N and 11-S) and 27 (groups 21-N and 21-S). Blood samples were obtained from the central ear artery (11) on the same day and were processed to yield serum, which was stored at 220°C until assayed. The concentration of estradiol-17b (E2) in the serum was determined by radioimmunoassay (RIA) with a commercial kit (Radim, Pomezia, Italy) based on competition between antigens labeled with iodine 125 (radioactive conjugate) and nonlabeled antigens (calibrator sample) for specific binding sites in

antiserum-coated tubes. After incubation, all unbound material was removed and radioactivity measured. Uncoated tubes were prepared for measurements of total activity (T) and nonspecific binding (NSB). Tubes coated with rabbit antibody against E2 were prepared for measurements in the zero calibrator (Bo), calibrators 1 to 6, control serum, and samples as follows. First, 100 μL of Bo was added to the NSB tube, and 100 μL of each additional calibrator as well as the control serum and samples was pipetted into the corresponding tube. Next, 500 μL of the radioactive conjugate was pipetted into all the tubes, whose contents were then mixed by vortex. After incubation for 3 h 6 5 min at 37°C 6 2°C the contents were carefully aspirated by pump from all tubes except the uncoated T tube. The radio-activity in the tubes was measured with a g-counter. The sensitivity of the assay was 2 pg. The intra-assay coefficient of variation (CV) was 3%.

Scanning procedureFor ultrasound scanning the nonsedated rabbit does were put in

the supine position. The abdominal skin was shaved at the beginning of the experiment, and before each scanning session the abdominal area was covered with scanning gel. The scans were done with a 10-MHz transducer (model UST-987.7.5; Aloka, Tokyo, Japan) with a real-time B-mode ultrasound scanner (SSD-900; Aloka). During each scanning session the scanner settings that affect image attributes (overall time gain, near-field and far-field gains, compensation, and beam focus) were kept at predetermined levels. The images were displayed at maximum magnification.

To scan the left and the right ovaries the rabbits were laid in a dorsolateral position. The ovaries were located by first identifying the kidney as a major landmark and then moving the transducer a few millimeters caudolaterally from the caudal edge of the kidney, remembering that the ovaries have a superficial location under the skin. The best technique was to start with the left ovary, which was in general easier to find. One or more on-screen images of both ovaries were studied to assess the follicular and luteal structures. To investigate the follicular populations, individual follicles on each ovary were identified, measured, and classified by diameter as small (# 1.5 mm) or large (. 1.5 mm). The total number of LF was calculated by counting those on both ovaries. The occurrence of ovulation was evaluated by the presence of CL, since rabbits are induced ovulators, and functional CL should not be present in the ovaries of unmated females. The ovulation rate was determined, as recommended by the International Rabbit Reproduction Group (12), by counting the number of CL on both ovaries.

Statistical methodStatistical analysis was done using SAS software, version 8.1 (SAS

Institute, Cary, North Carolina, USA). A mixed procedure was used according to an autoregressive model to analyze repeated measures, including the effects of parity, remating interval, suckling, and their interactions on mating rate (number of does accepting the male/number of observed does 3 100) (12), ovulation frequency (num-ber of ovulating does/number of mated does 3 100) (12), ovarian structures, and E2 concentrations. Means were compared with a protected t-test; differences were considered significant when the P-value was less than 0.05.



Re s u l t sResults from the statistical analyses are depicted in Table I. Parity

and suckling had significant effects on mating rate (P , 0.01 and P , 0.05, respectively). More does accepted a male on day 11 than on day 21 (P , 0.05). There were significant interactions between parity and day as well as between day and suckling with respect to mating rate. Ovulation frequency was depressed by suckling (P , 0.01) and was significantly affected by parity (P , 0.05), day of mating (P , 0.01), and the interactions parity 3 day, suckling 3 parity, and day 3 suckling.

The ovarian follicles generally appeared as anechoic wide circular areas on the ultrasound scans. However, when numerous within the same ovary they were sometimes flattened and packed together; in this situation there is a risk that the total number will be underesti-mated. There were fewer LF, accompanied by lower serum E2 levels, in suckling does than in nonsuckling does (P , 0.01). In contrast, the numbers of LF and the serum E2 levels were not influenced by parity or by day of mating. Data on LF number and E2 levels referred to the rabbit does that ovulated. The number of CL, visible as moderately hyperechoic structures, was not affected significantly by parity, day of mating, suckling, or their interactions. Luteal structures were in some cases not clearly discernible because they appeared almost isoechoic with the ovarian parenchyma rather than hyperechoic.

D i s c u s s i o nThe higher percentage of rabbit does mating on day 11 than on

day 21 postpartum is presumably consistent with greater sexual receptivity, and in fact the sexual behavior of does varies with the lactation stage. The percentage of does that accept a male is very high on the day of parturition, decreases at day 4 postpartum, increases

at day 11 postpartum, and returns to the highest level after weaning (5). The primiparous lactating rabbits in our study had very poor mating performance on day 21 postpartum, the day of maximum milk production and lowest live body weight for primiparous rabbits (1): an energy deficit could have affected sexual receptivity.

Mating was not followed by ovulation in some does, both suck-ling and nonsuckling, in this study. However, in the primiparous subjects this lack of ovulation occurred only in the suckling group, which may indicate different effects exerted by suckling upon the hypothalamic centers regulating sexual receptivity and the release of luteinizing hormone (13). A possible relationship with parity requires further investigation.

The small difference in follicular population was not unexpected, since the rabbit doe maintains a relatively constant supply of pre-ovulatory follicles. In domestic rabbits, in which there is no sponta-neous gonadotropin surge, follicles undergo waves of continuous maturation, such that ovulable follicles are present at nearly any given time (14). The lower population of LF in the suckling group might be responsible for the lower serum E2 levels of this group compared with the nonsuckling group. Since E2 concentrations have been commonly used to assess follicular growth and steroidogenic capacity (15,16), the lower hormonal levels in the suckling rabbits may reveal poorer ovarian activity, which could result in reduced reproductive efficiency.

Ovulation frequency was negatively affected by suckling, which confirms that in nursing rabbits, as in other species, sexual recep-tivity and fertility after natural mating and artificial insemination appear to be depressed during the period of lactation. Indeed, these does are less fertile as a direct consequence of the lack of ovula-tion, fertility failure, or embryo death (5). The existence of a partial antagonism between lactation and reproduction, reflecting the cor-responding hormonal antagonism between prolactin and the release

Table I. Effects of parity, day of mating, suckling, and their interactions on reproductive traits of rabbit does

Mean 6 standard error and level of statistical significancea

Reproductive trait Mating Ovulation Corpora Large Estradiol-17b

or interaction rate (%) frequency (%) lutea (n) folliclesb (n) level (pg/mL)Parity (P) Primiparous 66.0 6 4.9** 79.7 6 5.8* 8.9 6 0.2 15.0 6 1.5 14.5 6 1.3 Multiparous 95.6 6 6.7** 68.2 6 4.5* 10.3 6 0.8 16.2 6 1.7 15.3 6 1.5Day of mating (D; number of days postpartum) 11 86.8 6 6.5* 69.6 6 3.1** 9.0 6 0.2 15.9 6 1.3 14.9 6 1.9 21 74.0 6 5.1* 50.0 6 3.9** 10.0 6 0.7 15.3 6 1.8 14.8 6 1.6Suckling status (S) Suckling 76.0 6 5.9* 47.8 6 3.7** 9.4 6 0.9 12.1 6 1.1** 12.6 6 1.7** Not suckling 88.1 6 6.3* 68.1 6 4.2** 9.9 6 0.6 19.01 6 2.2** 17.5 6 2.1**P 3 D ** * NS NS NSS 3 P NS ** NS NS NSD 3 S * * NS NS NSa Asterisks indicate a significant difference between a pair of means or a significant interaction at P-values of , 0.05 (*) or , 0.01 (**). The remain-ing pairs of means did not differ significantly. The interaction P 3 D 3 S was not significant.b Those more than 1.5 mm in diameter.NS — not significant.



of gonadotropins in the rabbit, has been widely reported (17,18). In the current study prolactin was not assayed, but it is well known that secretion of this hormone is increased during lactation (19).

In this study, the serum E2 levels were lower in suckling than in nonsuckling rabbits on day 6 after coitus. Estrogens have been shown to play a key role in the maintenance of CL in the rab-bit, prompting Hilliard and Eaton (20) to refer to estradiol as the “ultimate luteotropin” in this species. The rabbit CL becomes an estradiol-dependent tissue by day 6 of pregnancy, as reported by Miller and Keyes (21), who demonstrated that adequate E2 at day 6 of pregnancy and pseudopregnancy is critical to continued develop-ment of the CL. The luteotropic E2 is supplied by ovarian LF, whose destruction leads to immediate failure of the CL and termination of pregnancy and pseudopregnancy (22–24).

Suckling had a depressive effect on both ovulation frequency and LF number in this study. Regarding the steroidogenic capability of the follicles present through the first days of pregnancy, early stud-ies indicated that LF generally have a higher steroid content than small follicles and that the appearance of new LF may be reflected in the rise in serum E2 concentration by day 4 to 6 after ovulation (20,25). Moreover, Mills et al (26) postulated a relationship between the estrogen secretion and the wave of follicle growth occurring in the first 6 d after mating. More recently Marongiu and Gulinati (27) used ultrasound scanning to inquire into this relationship after inducing ovulation by injection of human chorionic gonadotropin.

By performing transabdominal real-time ultrasonography, this trial demonstrated that rabbit ovaries can be evaluated in vivo. This noninvasive technique was found effective since the ultrasonic image allows immediate interpretation in most circumstances. Indeed, ultrasound scanning could represent a reliable method for studying several reproductive functions in the rabbit.

Since the ovulation response and fertility after artificial insemina-tion are high in rabbits that accept mating and significantly lower when they are not receptive to the male (5), the data from the cur-rent experiment appear suitable to support additional research aimed to elucidate the same postpartum reproductive phenomena when artificial insemination is done. Nevertheless, further studies are necessary to clarify follicular growth patterns, overall ovarian dynamics, and uterus changes during simultaneous pregnancy and lactation, with the aim of optimizing reproductive efficiency through breeding systems adapted to rabbit doe physiology.

A c k n o w l e d g m e n t sThis research was supported by F.A.R. (Fondo di Ateneo per la

Ricerca) of the University of Sassari, Sassari, Italy, and by the Bank of Sardinia Foundation. The authors thank Dr. Ana Helena Dias Francesconi for English revision.

Re f e r e n c e s 1. Rebollar PG, Pérez-Cabal MA, Pereda N, et al. Effects of parity

order and reproductive management on the efficiency of rabbit productive systems. Livest Sci 2009;121:227–233.

2. Partridge GG, Allan SJ, Findlay M, et al. The effects of reduc-ing the remating interval after parturition on the reproductive

performance of the commercial doe rabbit. Anim Prod 1984; 39:465–472.

3. Cervera CJ, Fernandez-Carmona J, Viudes P, et al. Effect of remating interval and diet on the performance of female rabbits and their litters. Anim Prod 1993;56:399–405.

4. Marongiu ML, Gulinati A, Bomboi G, et al. Lengthening of the remating interval improves body condition and reproduction efficiency of lactating rabbit does. Ital J Anim Sci 2009;8:822 (Abstract).

5. Theau-Clément M. Preparation of the rabbit doe to insemination: A review. World Rabbit Sci 2007;5:61–80.

6. Fortun L, Prunier A, Lebas F. Effects of lactation on fetal survival and development in rabbit does mated shortly after parturition. J Anim Sci 1993;71:1882–1886.

7. Garcia-Garcia RM, Arias-Alvarez M, Rebollar PG, et al. Influence of different reproductive rhythms on serum estradiol and tes-tosterone levels, features of follicular population and atresia rate, and oocyte maturation in controlled suckling rabbits. Anim Reprod Sci 2008;114:423–433.

8. Ypsilantis P, Saratsis PH. Early pregnancy diagnosis in the rabbit by real time ultrasonography. World Rabbit Sci 1999;7: 95–99.

9. Chavatte-Palmer P, Laigre P, Simonoff E, et al. In utero charac-terisation of fetal growth by ultrasound scanning in the rabbit. Theriogenology 2008;69:859–869.

10. Polisca A, Scotti L, Orlandi R, et al. Doppler evaluation of maternal and fetal vessels during normal gestation in rabbits. Theriogenology 2010;73:358–366.

11. Marongiu ML, Gulinati A, Floris B. A procedure for rabbit blood serial collection. Ital J Anim Sci 2007;6:773 (Abstract).

12. International Rabbit Reproduction Group. Recommendations and guidelines for applied reproduction trials with rabbit does. World Rabbit Sci 2005;13:147–164.

13. Pau K-Y, Orstead KM, Hess DL, et al. Feedback effects of ovarian steroids on the hypothalamic–hypophyseal axis in the rabbit. Biol Reprod 1986;35:1009–1023.

14. Fleming MW, Rhodes RC, Dailey RA. Compensatory responses after unilateral ovariectomy in rabbits. Biol Reprod 1984;30: 82–86.

15. Wallach EE, Noriega C. Effects of local steroids on follicular development and atresia in the rabbit. Fertil Steril 1970;21: 253–267.

16. Ubilla E, Rebollar PG. Influence of the postpartum day on plasma oestradiol concentrations, sexual behaviour, and con-ception rate, in artificially inseminated lactating rabbits. Anim Reprod Sci 1995;38:337–344.

17. Theau-Clément M, Roustan A. A study on relationships between receptivity and lactation in the doe and their influence on repro-ductive performances. J Appl Rabbit Res 1992;15:412–421.

18. Boiti C. Underlying physiological mechanisms controlling the reproductive axis of rabbit does. In: Proceedings of the 8th World Rabbit Congress, 2004 Sept 7–10, Puebla, Mexico; World Rabbit Science Assoc, Castanet-Tolosan, France:186–206.

19. Durand P, Djiane J. Lactogenic activity in the serum of rabbits during pregnancy and early lactation. J Endocrinol 1977;75: 33–42.



20. Hilliard J, Eaton LW. Estradiol-17b, progesterone and 20a hydroxypregn-4-en-3-one in rabbit ovarian venous plasma. II. From mating through implantation. Endocrinology 1971;89: 522–527.

21. Miller JB, Keyes PL. Transition of the rabbit corpus luteum to estrogen dependence during early luteal development. Endocrinology 1978;102:31–38.

22. Keyes PL, Nalbandov AV. Maintenance and function of corpora lutea in rabbits depend on estrogen. Endocrinology 1967;80: 938–946.

23. Rennie P. Luteal–hypophyseal interrelationship in the rabbit. Endocrinology 1968;83:323–328.

24. Niswender GD, Juengel JL, Silva PJ, et al. Mechanisms control-ling the function and life span of the corpus luteum. Physiol Rev 2000;80:1–29.

25. Challis JR, Davies IJ, Ryan KJ. The concentration of progesterone, estrone and estradiol-17b in the plasma of pregnant rabbits. Endocrinology 1973;93:971–976.

26. Mills T, Copland A, Osteen K. Factors affecting the postovulatory surge of FSH in the rabbit. Biol Reprod 1981;25:530–535.

27. Marongiu ML, Gulinati A. Ultrasound evaluation of ovarian follicular dynamics during early pseudopregnancy as a tool to inquire into the high progesterone (P) syndrome of rabbit does. In: Xicato G, Trocino A, Lukefahr SD, eds. Proceedings of the 9th World Rabbit Congress, 2008 June 10–13, Verona, Italy; World Rabbit Science Assoc, Castanet-Tolosan, France:393–397.


Article


I n t r o d u c t i o nThe minimum alveolar concentration (MAC) of an inhalational

anesthetic is defined as the alveolar concentration at sea level at which there is no purposeful movement in 50% of patients in response to a supra-maximal stimulus (1–3). Recent studies in dogs have investigated MAC derivatives, such as the MAC at which there is no motor movement (MACNM) (4) and the MAC at which the autonomic response to noxious stimuli is blocked (MACBAR) (5,6).

Nitrous oxide (N2O) is a colorless, non-flammable gas that is used in humans for its analgesic, immobilizing, and anxiolytic effects (7–9). Compared with other inhalational anesthetics, however, N2O is low in potency and is less potent in dogs than in humans. Reported MAC values for N2O in dogs vary from 188% (2) to 222% (10) and N2O is used primarily as an adjunct to volatile anesthetics for its MAC-decreasing properties. In a recent study, the authors found that 70% N2O decreased the MAC, MACNM, and MACBAR of sevoflurane in dogs by 24%, 25%, and 35%, respectively (5).

The purpose of this study was to evaluate the effects of 70% N2O on the MAC, MACNM, and MACBAR of isoflurane (ISO) in dogs. It was hypothesized that 70% N2O would significantly decrease the MAC, MACNM, and MACBAR of ISO.


AnimalsSix adult (2 to 3 y of age), purpose-bred, mixed-breed, intact male

dogs (14 6 1 kg) were determined to be healthy based on physical examination. Food was withheld for 12 h before anesthesia, but access to water was allowed. Each dog was anesthetized once. The MAC, MACNM, and MACBAR were determined in that order for ISO alone (baseline) and then for ISO with 70% N2O (treatment). This sequence of determination of MAC and its derivatives was used to expedite the process and is standard procedure in our laboratory.

The study protocol was approved by the Institutional Animal Care and Use Committee at the University of Tennessee and was

The effect of nitrous oxide on the minimum alveolar concentration (MAC) and MAC derivatives of isoflurane in dogs

Debra A. Voulgaris, Christine M. Egger, M. Reza Seddighi, Barton W. Rohrbach, Lydia C. Love, Thomas J. Doherty

A b s t r a c tThis study investigated the effects of 70% nitrous oxide (N2O) on the minimum alveolar concentration (MAC) of isoflurane (ISO) that prevents purposeful movement, the MAC of ISO at which there is no motor movement (MACNM), and the MAC of ISO at which autonomic responses are blocked (MACBAR) in dogs.

Six adult, healthy, mixed-breed, intact male dogs were anesthetized with ISO delivered via mask. Baseline MAC, MACNM, and MACBAR of ISO were determined for each dog using a supra-maximal electrical stimulus (50 V, 50 Hz, 10 ms). Nitrous oxide (70%) was then administered and MAC and its derivatives (N2O-MAC, N2O-MACNM, and N2O-MACBAR) were determined using the same methodology. The values for baseline MAC, MACNM, and MACBAR were 1.39 6 0.14, 1.59 6 0.10, and 1.72 6 0.16, respectively. The addition of 70% N2O decreased MAC, MACNM, and MACBAR by 32%, 15%, and 25%, respectively.

R é s u m éCette étude avait comme objectif d’évaluer chez des chiens les effets de 70 % d’oxyde nitreux (N2O) sur la concentration alvéolaire minimum (MAC) d’isoflurane (ISO) qui empêche les mouvements volontaires, la MAC d’ISO à laquelle il n’y a pas de mouvement moteur (MACNM), et la MAC d’ISO à laquelle les réponses autonomes sont bloquées (MACBAR).

Six chiens mâles intacts adultes de race mélangée ont été anesthésiés avec de l’ISO administré via un masque. Les valeurs de base de MAC, MACNM et de MACBAR d’ISO ont été déterminées pour chaque chien à l’aide d’un stimulus électrique supra-maximal (50 V, 50 Hz, 10 ms). De l’oxyde nitreux (70 %) fut ensuite administré et la MAC et ses dérivées (N2O-MAC, N2O-MACNM et N2O-MACBAR) déterminées à l’aide de la même méthodologie. Les valeurs des données de base de MAC, MACNM et MACBAR étaient respectivement 1,39 6 0,14, 1,59 6 0,10 et 1,72 6 0,16. L’ajout de 70 % de N2O a entrainé des diminutions de MAC, MACNM et MACBAR de 32 %, 15 % et 25 %, respectivement.


California Animal Rehabilitation, Santa Monica, California, USA (Voulgaris); Department of Small Animal Clinical Sciences (Egger), Department of Large Animal Clinical Sciences (Seddighi, Doherty), and Department of Biomedical and Diagnostic Sciences (Rohrbach), University of Tennessee, Knoxville, Tennessee, USA; Animal Emergency and Referral Associates, Fairfield, New Jersey, USA (Love).

Address all correspondence to Dr. Christine M. Egger; telephone: (865) 974-8387; fax: (865) 974-5554; e-mail: [email protected]




carried out in accordance with the Guide for the Care and Use of Experimental Animals.

AnesthesiaAnesthesia was induced with ISO (IsoFlo; Abbott Animal Health,

Abbott Park, Illinois, USA) in oxygen delivered via mask from a circle breathing system. After tracheal intubation, anesthesia was maintained with ISO in oxygen (2 L/min) using a small animal anesthetic machine (North American Drager, Telford, Pennsylvania, USA). Ventilation was controlled to maintain the end-tidal carbon dioxide partial pressure (Pe9CO2) at between 35 to 45 mmHg. Arterial blood samples were drawn from each subject at the time of each MAC, MACNM, and MACBAR determination to ensure that arterial carbon dioxide tension (PaCO2), arterial partial pressure of oxygen (PaO2), and acid-base status were within normal limits. Dogs were placed in lateral recumbency, a 20-ga cephalic catheter (MILA International, Erlanger, Kentucky, USA) was placed, and lactated Ringer’s solution (Abbott Animal Health) was infused (3 mL/kg body weight per h).

End-tidal ISO (E9ISO), end-tidal N2O (E9n2o), and Pe9CO2 were monitored continuously with an infrared gas analyzer (Criticare Systems, Waukesha, Wisconsin, USA). Samples were drawn from the proximal end of the endotracheal tube at a rate of 150 mL/min. At the beginning of the study, the monitor was calibrated with the cali-bration gases supplied by the manufacturer (1% ISO in 5% CO2 and 60% N2O; Criticare Systems). Body temperature was monitored using an esophageal probe (Criticare Systems). A circulating warm water blanket and a warm air blanket (Bair Hugger; Arizant Healthcare, Eden Prairie, Minnesota, USA) were used to maintain body tem-perature within the normal range (37.5°C to 38.5°C). Arterial blood pressure was monitored continuously from a 20-ga catheter placed in a dorsal pedal artery, using a monitor (Criticare Systems) and a disposable transducer (Baxter Healthcare Corporation, Deerfield, Illinois, USA). The middle of the sternum was taken as the zero point for blood pressure measurement. Heart rate and electrocardiogram (ECG) were monitored continuously using a 3-lead system and hemoglobin saturation (SpO2) was monitored continuously using a tongue probe (Criticare Systems).

Determination of baseline MACThe determination of baseline MAC began approximately 45 min

after induction of anesthesia and with the E9ISO held constant at 1.5% for at least 15 min. A supra-maximal stimulus (50 V, 50 Hz, 10 ms) was delivered (Grass Instrument Company, West Warwick, Rhode Island, USA) via two 25-ga electrode needles inserted sub-cutaneously 5 cm apart over the mid-ulnar area. Two single stimuli with a 5-s interval were delivered initially, followed 5 s later by a continuous stimulus of 5 s duration, which was repeated after 5 s (11). Purposeful movement was defined as gross movement of the head or extremities. Twitching of the stimulated limb, coughing, swallowing, rigidity, tail movement, or chewing were not considered purposeful movements. If purposeful movement occurred, the E9ISO was increased by 0.1% or 0.2% depending on the magnitude of the response; otherwise, it was decreased by 0.1% and the stimulus was reapplied after a 15-min equilibration period. The MAC was defined as the mean of the lowest E9ISO at which purposeful movement did

and did not occur. All MAC values were determined in duplicate and the mean value was taken as the baseline MAC for that animal. If the difference between these values was greater than 10%, a third value was determined and the mean of these 3 values was taken as the baseline MAC for that animal.

Determination of baseline MACNMAfter MAC was determined, the E9ISO was maintained at 1.5% for

at least 15 min before the baseline MACNM was determined using the same methodology as for MAC. The MACNM was defined as the lowest E9ISO at which there was no motor movement, purposeful or non-purposeful, in response to the noxious stimulus. Twitching of the stimulated limb was not considered a positive response.

Determination of baseline MACBARAfter baseline MACNM was determined, the E9ISO was maintained

at 1.5% for at least 15 min before initiating baseline MACBAR deter-mination. During each pre-stimulus period, heart rate (HR) and mean arterial pressure (MAP) values were recorded from the arterial line and were stable for at least 5 min, varying by less than 1%. The greatest HR and MAP values during this time period were taken as the baseline. The baseline MACBAR was determined using the same methodology as for MAC and MACNM. MACBAR was defined as the lowest E9ISO that prevented a $ 15% increase in baseline MAP and HR in response to the noxious stimulus during the 60-s period beginning at the time of the first stimulus.

Administration of N2OAfter baseline MAC, MACNM, and MACBAR were determined,

administration of 70% N2O began. After a 15-min equilibration period with the E9n2o maintained at 70% and the E9ISO at 1.5%, the treatment MAC endpoints (N2O-MAC, N2O-MACNM, and N2O-MACBAR) were determined using the same methods previously described for the baseline MAC and its derivatives.

Time recording began immediately after the initial equilibration period and time to determination of MAC, MACNM, and MACBAR was cumulative. The dogs were evaluated for tissue damage, lame-ness, and pain for 24 h after recovery.

Statistical analysisPercent change in MAC, MACNM, and MACBAR was calculated

according to the formula:

(treatment value 2 baseline value)/(baseline value) 3 100

A mixed-model analysis of variance (ANOVA) (PROC MIXED) was used to determine the effect of treatment on MAC, MACNM, and MACBAR. Dog was included as a random factor in the model. Dog, treatment, and endpoint were included as class variables. Independent variables included treatment, endpoint, time, and the 2-way interaction between endpoint and treatment. A second mixed-model ANOVA was used to compare the percent change in MAC among endpoints (MAC, MACNM, and MACBAR). Class variables included in the model were dog and endpoint. Endpoint was the independent variable and dog was included as a random factor in the model. A multiple range test according to the method of Tukey was used to adjust for multiple comparisons. Fit of the models was



evaluated using the -2 log likelihood ratio and the fit of residuals from the model to a normal distribution. Residuals were evaluated using the test statistic of Shapiro-Wilk. Effect of treatment on percent change in MAC at each endpoint was evaluated using a paired t-test [PROC UNIVARIATE]. Data are expressed as least squares means (LSM) and standard error of the mean (SEM). A P-value of # 0.05 was considered significant.

Re s u l t sThe mean baseline values for MAC, MACNM, and MACBAR were

1.39%, 1.59%, and 1.72%, respectively (Table I). Administering 70% N2O decreased these values by 32%, 15%, and 25%, respectively. Baseline MACNM was not significantly different than N2O-MACNM. While the percent change in MACBAR was not significantly different than the percent change in MAC, the percent change in MACNM was significantly different than the percent change in MAC and MACBAR (Table I). The estimated hemoglobin saturation was . 95% and PaO2, PaCO2, and acid-base status were normal at all times before and dur-ing administration of N2O. Recovery from anesthesia was uneventful and the dogs resumed normal activities within 2 to 3 h of recovery. The stimulated limbs appeared normal at all times.

D i s c u s s i o nIn this study, administering 70% N2O decreased MAC, MACNM,

and MACBAR (Table I). The baseline MAC value of 1.39% was compa-rable to the values reported for dogs in previous studies: 1.38% (12), 1.28% (13), and 1.34% (11). While interindividual variation in MAC values of 10% to 20% is typical (3), variation was minimized in this study by the use of only 1 observer. The MAC can also be affected by extremes of PaCO2, PaO2, body temperature, and arterial blood pressure. These variables were maintained within normal range in each patient throughout the experiment.

The addition of 70% N2O decreased the MAC by 32%, which is comparable with the MAC-sparing effects of N2O reported in pre-vious studies. In halothane-anesthetized dogs, 75% N2O decreased MAC by 34% (10) and in sevoflurane-anesthetized dogs, 70% N2O

decreased MAC by 24% (5). In a clinical study of dogs undergoing ovariohysterectomy, a 37% decrease in requirement for isoflurane was reported when 64% N2O was included in the anesthetic protocol (14). These results are also consistent with the effects of N2O in other species. For example, 70% N2O decreased the MAC of isoflurane in rats by 40% (15) and 75% N2O decreased the MAC in swine by 38% (16). In desflurane-anesthetized dogs, however, 70% N2O decreased the MAC by only 16% (17). Differences among studies are likely due to individual variation, sample size, inhalational anesthetic, and experimental design.

The MACNM in this study was 1.59% or 1.14 MAC (Table I). This ratio of MACNM/MAC is comparable to the reported ratio of 1.16 for sevoflurane MACNM/MAC in dogs (5). These data are also in general agreement with a study of human surgical patients, which reported that the E9ISO that prevented movement in 95% of the population was approximately 25% greater than the MAC (18). In contrast, a com-parable endpoint in halothane-anesthetized ponies was equivalent to 1.6 MAC (19), which may reflect differences among species and inhalational anesthetics. The addition of 70% N2O decreased MACNM by 15% (Table I), but there was wide variability among dogs. To the authors’ knowledge, there are no published reports on the effect of N2O on MACNM in dogs. In another study, the authors determined that 70% N2O decreased sevoflurane MACNM in dogs by 25% (5).

In this study, baseline MACBAR was 1.72% or 1.24 MAC (Table I). This endpoint is typically greater than the other MAC derivatives, as autonomic responses are activated at lower stimulus levels and are more resistant to blockade than movement responses (20). Suppression of this response may be clinically relevant because autonomic activation can have deleterious effects on the patient (20–22). There is limited information on MACBAR in dogs and other veterinary species, and most MACBAR studies in humans include N2O in the baseline anesthetic protocol, which makes it difficult to compare results. Recent studies by the authors reported MACBAR values of 1.27 MAC (5) and 1.4 MAC (6) for sevoflurane. Reported MACBAR values vary widely in other species. A study of isoflurane-anesthetized goats reported a MACBAR of 2.8 MAC (23), but in cats anesthetized with isoflurane, the MACBAR was only 1.1 MAC (24). In rats, the MACBAR for sevoflurane did not differ significantly from the MAC (25), and MACBAR values of 2.58 MAC (26) and 3.9 MAC (27) for sevoflurane have been reported in human female patients. Variations of such magnitude are likely due to the same factors as those discussed previously for MAC.

In the present study, the mean decrease in MACBAR with the addition of 70% N2O was 25%. To the authors’ knowledge, there are no published reports on the effect of 70% N2O on the MACBAR of isoflurane in dogs. In a previous study, however, the authors found that 70% N2O decreased the MACBAR of isoflurane by approximately 35% in dogs (5).

Decreases in MAC and its derivatives with N2O could be due to its analgesic and/or immobilizing effects. Although the mechanisms of action of N2O are not completely understood, its analgesic actions are likely separate from its immobilizing effects (9,28). Immobility during general anesthesia is mediated by motor neurons located in the ventral horn of the spinal cord (29,30). Interestingly, it has been shown that neurons in the ventral horn are more sensitive to the depressant effects of N2O than are neurons in the dorsal horn

Table I. Baseline and treatment MAC, MACNM, and MACBAR values of isoflurane and percent change in each value after adding 70% N2O in 6 male dogs

MAC endpoint Baseline Treatment % ChangeMAC 1.39 6 0.14a 0.98 6 0.14b 231.9 6 3.31

MACNM 1.59 6 0.10c 1.37 6 0.10c 214.9 6 3.32

MACBAR 1.72 6 0.16d 1.31 6 0.12e 224.9 6 3.31

MAC — minimum alveolar concentration; MACNM — minimum alveo-lar concentration at which there is no motor movement; MACBAR — minimum alveolar concentration at which autonomic response is blocked.a,b,c,d,e Values in the same row with different letters are significantly different.1,2 Values in the same column with different numbers are significantly different (P # 0.05). All values are presented as least squares mean 6 standard error of the mean.



(31). The immobilizing effects of N2O may be due to N-methyl-D-aspartate (NMDA) receptor blockade in the ventral horn (32,33), although mechanisms involving monoaminergic pathways have also been suggested (34).

Numerous mechanisms have been proposed to explain the analge-sic actions of N2O. Although still controversial, prevailing evidence supports the involvement of opioid and alpha-2 adrenergic receptors (28,35,36). Specifically, nitrous oxide induces analgesia by activating opioidergic neurons in the periaqueductal gray matter and norad-renergic neurons in the locus coeruleus. This results in modulation of nociceptive transmission at the level of the spinal cord (7,37–39).

In this study, the decrease in MAC and its variants with the addi-tion of 70% N2O ranged from 15% to 32%. It therefore appears that N2O provides a clinically important reduction in MAC. The differ-ence in the magnitude of the effect of N2O on MAC and MACNM is surprising because they are both presumably mediated at the level of the spinal cord. This difference may be due to the small sample size and variability among subjects.

In this study, the determination of MAC and its derivatives was not randomized. Determining MAC provides a starting point for the determination of MACNM or MACBAR, as previously published studies and experience have indicated that MACNM and MACBAR are usually higher than MAC (5,23,26,27). Determining MAC and its derivatives in this order expedites the process. The current study and a previous study (5) from our laboratory demonstrated that the time to determine MAC and its derivatives has no significant effect on outcome, although this does not completely rule out an effect of order of determination.

The benefits of administering N2O must be weighed against its potential adverse effects on patients, personnel, and the environment. The patient’s oxygenation must be monitored continuously through-out the perioperative period, as hypoxemia is more likely when using N2O. Long-term exposure to N2O can have adverse effects on per-sonnel, including bone marrow suppression, spontaneous abortion, teratogenicity, genotoxicity, and myelinopathies (39–41). In addition, N2O contributes to ozone depletion in the environment (40,41).

In conclusion, adding 70% N2O significantly decreased the MAC, MACNM, and MACBAR of isoflurane (ISO) in dogs by 32%, 15%, and 25%, respectively.

A c k n o w l e d g m e n tThe authors thank Abbott Animal Health for donating IsoFlo for

this study. Abbott Animal Health had no role, however, in the study design; collection, analysis, or interpretation of the data; writing of the manuscript; or in the decision to submit the manuscript for publication.

Re f e r e n c e s1. Merkel G, Eger EI 2nd. A comparative study of halothane and

halopropane anesthesia including a method for determining equipotency. Anesthesiology 1963;24:346–357.

2. Eger EI 2nd, Saidman LJ, Brandstater B. Minimum alveolar anesthetic concentration: A standard of anesthetic potency. Anesthesiology 1965;26:756–763.

3. Quasha AL, Eger EI 2nd, Tinker JH. Determination and applica-tions of MAC. Anesthesiology 1980;53:315–334.

4. Seddighi R, Egger C, Rohrbach B, Cox S, Doherty T. The effects of midazolam on the end-tidal concentration of isoflurane necessary to prevent movement in dogs. Vet Anaesth Analg 2011;38:195–202.

5. Seddighi R, Egger CM, Rohrbach BW, Hobbs M, Doherty TJ. The effect of nitrous oxide on sevoflurane MAC and MAC deriva-tives in dogs. Am J Vet Res 2012;73:341–345.

6. Love L, Egger C, Rohrbach B, Cox S, Hobbs M, Doherty T. The effect of ketamine on the MACBAR of sevoflurane in dogs. Vet Anaesth Analg 2011;38:292–300.

7. Maze M, Fujinaga M. Pharmacology of nitrous oxide. Best Pract Res Clin Anaesthesiol 2001;15:339–348.

8. Emmanouil DE, Quock RM. Advances in understanding the actions of nitrous oxide. Anesth Prog 2007;54:9–18.

9. Jinks SL, Carstens E, Antognini JF. Nitrous oxide-induced anal-gesia does not influence nitrous oxide’s immobilizing require-ments. Anesth Analg 2009;109:1111–1116.

10. Steffey EP, Gillespie JR, Berry JD, Eger EI 2nd, Munson ES. Anesthetic potency (MAC) of nitrous oxide in the dog, cat, and stump-tail monkey. J Appl Physiol 1974;36:530–532.

11. Valverde A, Morey TE, Hernandez J, Davies W. Validation of several types of noxious stimuli for use in determining the minimum alveolar concentration for inhalation anesthetics in dogs and rabbits. Am J Vet Res 2003;64:957–962.

12. Yang XL, Ma HX, Yang ZB, et al. Comparison of minimum alveolar concentration between intravenous isoflurane lipid emulsion and inhaled isoflurane in dogs. Anesthesiology 2006; 104:482–487.

13. Steffey EP, Howland D Jr. Isoflurane potency in the dog and cat. Am J Vet Res 1977;38:1833–1836.

14. Duke T, Caulkett NA, Tataryn JM. The effect of nitrous oxide on halothane, isoflurane and sevoflurane requirements in ven-tilated dogs undergoing ovariohysterectomy. Vet Anaesth Analg 2006;33:343–350.

15. Santos M, Kuncar V, Martinez-Taboada F, Tendillo FJ. Large concentrations of nitrous oxide decrease the isoflurane minimum alveolar concentration sparing effect of morphine in the rat. Anesth Analg 2005;100:404–408.

16. Tranquilli WJ, Thurmon JC, Benson GJ. Anesthetic potency of nitrous oxide in young swine (Sus scrofa). Am J Vet Res 1985; 46:58–60.

17. Nishimori CT, Nunes N, Paula DP, Rezende ML, Souza AP, Santos PSP. Effects of nitrous oxide on minimum alveolar con-centration of desflurane in dogs. Arquivo Brasileiro de Medicina Veterinária e Zootecnia 2007;59:97–104.

18. de Jong RH, Eger EI 2nd. MAC expanded AD50 and AD95 val-ues of common inhalation anesthetics in man. Anesthesiology 1975;42:384–389.

19. Doherty TJ, Geiser DR, Frazier DL. Comparison of halo-thane minimum alveolar concentration and minimum effec-tive concentration in ponies. J Vet Pharmacol Ther 1997;20: 408–410.

20. Desborough JP. The stress response to trauma and surgery. Br J Anaesth 2000;85:109–117.



21. Ben-Eliyahu S, Page GG, Schleifer SJ. Stress, NK cells and cancer: Still a promissory note. Brain Behav Immun 2007;21:881–887.

22. Bartal I, Melamed R, Greenfeld K, et al. Immune perturbations in patients along the perioperative period: Alterations in cell surface markers and leukocyte subtypes before and after surgery. Brain Behav Immun 2010;24:376–386.

23. Antognini JF, Berg K. Cardiovascular responses to noxious stimuli during isoflurane anesthesia are minimally affected by anesthetic action in the brain. Anesth Analg 1995;81:843–848.

24. March PA, Muir WW 3rd. Minimum alveolar concentration mea-sures of central nervous system activation in cats anesthetized with isoflurane. Am J Vet Res 2003;64:1528–1533.

25. Docquier MA, Lavand’homme P, Ledermann C, Collet V, De Kock M. Can determining the minimum alveolar anes-thetic concentration of volatile anesthetic be used as an objec-tive tool to assess antinociception in animals? Anesth Analg 2003;97:1033–1039.

26. Nakata Y, Goto T, Ishiguro Y, Terui K, Nimi Y, Morita S. Anesthetic doses of sevoflurane to block cardiovascular responses to incision when administered with xenon or nitrous oxide. Anesthesiology 1999;91:369–373.

27. Ura T, Higuchi H, Taoda M, Sato T. Minimum alveolar concen-tration of sevoflurane that blocks the adrenergic response to surgical incision in women: MACBAR. Eur J Anaesthesiol 1999; 16:176–181.

28. Koyama T, Fukuda K. Involvement of the kappa-opioid receptor in nitrous oxide-induced analgesia in mice. J Anesth 2010;24: 297–299.

29. Rampil IJ, King BS. Volatile anesthetics depress spinal motor neurons. Anesthesiology 1996;85:129–134.

30. Jinks S, Bravo M, Hayes SG. Volatile anesthetic effects on mid-brain-elicited locomotion suggest that the locomotor network in the ventral spinal cord is the primary site for immobility. Anesthesiology 2008;108:1016–1024.

31. Kim J, Yao A, Atherley R, Carstens E, Jinks SL, Antognini JF. Neurons in the ventral spinal cord are more depressed by iso-

flurane, halothane, and propofol than are neurons in the dorsal spinal cord. Anesth Analg 2007;105:1020–1026.

32. Sonner JM, Antognini JF, Dutton RC, et al. Inhaled anesthetics and immobility: Mechanisms, mysteries, and minimum alveolar anesthetic concentration. Anesth Analg 2003;97:718–740.

33. Antognini JF, Atherley RJ, Dutton R, Laster MJ, Eger EI 2nd, Carstens E. The excitatory and inhibitory effects of nitrous oxide on spinal neuronal responses to noxious stimulation. Anesth Analg 2007;104:829–835.

34. Petrenko AB, Yamakura T, Kohno T, Sakimura K, Baba H. Reduced immobilizing properties of isoflurane and nitrous oxide in mutant mice lacking the N-methyl-D-aspartate recep-tor GluR1 subunit are caused by the secondary effects of gene knockout. Anesth Analg 2010;110:461–465.

35. Guo TZ, Davies MF, Kingery WS, Patterson AJ, Limbird LE, Maza M. Nitrous oxide produces antinociceptive response via alpha 2B and/or alpha 2C adrenoceptor subtypes in mice. Anesthesiology 1999;90:470–476.

36. Sawamura S, Kingery WS, Davies MF, et al. Antinociceptive action of nitrous oxide is mediated by stimulation of norad-renergic neurons in the brainstem and activation of [alpha]2b adrenoceptors. J Neurosci 2000;20:9242–9251.

37. Zhang C, Davies MF, Guo TZ, Maze M. The analgesic action of nitrous oxide is dependent on the release of norepineph-rine in the dorsal horn of the spinal cord. Anesthesiology 1999;91:1401–1407.

38. Fujinaga M, Maze M. Neurobiology of nitrous oxide-induced antinociceptive effects. Mol Neurobiol 2002;25:167–189.

39. Sanders RD, Weimann J, Maze M. Biologic effects of nitrous oxide: A mechanistic and toxicologic review. Anesthesiology 2008;109:707–722.

40. Myles PS, Leslie K, Silbert B, Paech MJ, Peyton P. A review of the risks and benefits of nitrous oxide in current anaesthetic practice. Anaesth Intensive Care 2004;32:165–172.

41. Ryan SM, Nielsen CJ. Global warming potential of inhaled anes-thetics: Application to clinical use. Anesth Analg 2010;111:92–98.


Article


Oxidative stress, superoxide production, and apoptosis of neutrophils in dogs with chronic kidney disease

Adriana Carolina Rodrigues Almeida Silva, Breno Fernando Martins de Almeida, Carolina Soares Soeiro, Wagner Luis Ferreira, Valéria Marçal Félix de Lima, Paulo César Ciarlini

A b s t r a c tOxidative stress is a key component in the immunosuppression of chronic kidney disease (CKD), and neutrophil function may be impaired by oxidative stress. To test the hypothesis that in uremic dogs with CKD, oxidative stress is increased and neutrophils become less viable and functional, 18 adult dogs with CKD were compared with 15 healthy adult dogs. Blood count and urinalysis were done, and the serum biochemical profile and plasma lipid peroxidation (measurement of thiobarbituric acid reactive substances) were determined with the use of commercial reagents. Plasma total antioxidant capacity (TAC) was measured with a spectrophotometer and commercial reagents, superoxide production with a hydroethidine probe, and the viability and apoptosis of neutrophils with capillary flow cytometry and the annexin V-PE system. The plasma concentrations of cholesterol (P = 0.0415), creatinine (P , 0.0001), and urea (P , 0.0001) were significantly greater in the uremic dogs than in the control dogs. The hematocrit (P = 0.0004), urine specific gravity (P = 0.015), and plasma lipid peroxidation (P , 0.0001) were significantly lower in the dogs that were in late stages of CKD than in the control group. Compared with those isolated from the control group, neutrophils isolated from the CKD group showed a higher rate of spontaneous (0.10 6 0.05 versus 0.49 6 0.09; P = 0.0033; median 6 standard error of mean) and camptothecin-induced (18.53 6 4.06 versus 44.67 6 4.85; P = 0.0066) apoptosis and lower levels of superoxide production in the presence (1278.8 6 372.8 versus 75.65 6 86.6; P = 0.0022) and absence (135.29 6 51.74 versus 41.29 6 8.38; P = 0.0138) of phorbol-12-myristate-13-acetate stimulation. Thus, oxidative stress and acceleration of apoptosis occurs in dogs with CKD, the apoptosis diminishing the number of viable neutrophils and neutrophil superoxide production.

R é s u m éLe stress oxydatif est un élément clé dans l’immunosuppression de maladie rénale chronique (MRC), et la fonction des neutrophiles peut être affectée par le stress oxydatif. Afin de vérifier l’hypothèse que chez les chiens urémiques avec MRC le stress oxydatif est augmenté et les neutrophiles deviennent moins viables et fonctionnels, 18 chiens adultes avec MRC ont été comparés à 15 chiens adultes en santé. Des analyses sanguines et urinaires ont été effectuées, de même que le profil biochimique sérique et la peroxydation des lipides plasmatiques (mesures des substances réactives à l’acide thiobarbiturique) ont été déterminés au moyen de réactifs commerciaux. La capacité antioxydante plasmatique totale (CAT) a été mesurée à l’aide d’un spectrophotomètre et de réactifs commerciaux, la production de superoxyde avec une sonde hydroéthidine, et la viabilité et l’apoptose des neutrophiles avec un cytomètre à flux capillaire et le système d’annexine V-PE. Les concentrations plasmatiques de cholestérol (P = 0,0415), de créatinine (P , 0,0001) et d’urée (P , 0,0001) étaient significativement plus élevées chez les chiens urémiques comparativement aux chiens témoins. L’hématocrite (P = 0,0004), la gravité spécifique de l’urine (P = 0,015), et la peroxydation des lipides plasmatiques (P , 0,0001) étaient significativement plus faibles chez les chiens dans les stades avancés de MCR que chez les chiens témoins. Comparativement aux neutrophiles provenant des chiens témoins, ceux provenant des chiens avec MRC montraient un taux plus élevé d’apoptose spontanée (0,10 6 0,05 versus 0,49 6 0,09; P = 0,0033; médiane 6 écart-type) et d’apoptose induite par la camptothécine (18,53 6 4,06 versus 44,67 6 4,85; P = 0,0066) et des niveaux plus faibles de production de superoxyde en présence (1278,8 6 372,8 versus 75,65 6 86,6; P = 0,0022) et en absence (135,29 6 51,74 versus 41,29 6 8,38; P = 0,0138) de stimulation par l’acétate de phorbol-12-myristate-13. Ainsi, le stress oxydatif et l’accélération de l’apoptose surviennent chez des chiens avec MRC, l’apoptose diminuant le nombre de neutrophiles viables et la production de superoxyde par les neutrophiles.


Department of Clinics, Surgery, and Animal Reproduction, College of Veterinary Medicine of Araçatuba, São Paulo State University, 793 Clóvis Pestana Street, 16050-680, Araçatuba, São Paulo, Brazil.

Address all correspondence to Dr. Paulo César Ciarlini; e-mail: [email protected]

Received December 16, 2011. Accepted April 5, 2012.



I n t r o d u c t i o nIn humans, kidney disease is an important cause of immunosup-

pression (1) that increases the risk of death from bacterial infection owing to the neutrophil dysfunction (2) associated with oxidative stress (3).

Neutrophils produce reactive oxygen species (ROS) when nico-tinamide adenine dinucleotide phosphate oxidase is activated, generating the superoxide anion essential to the bactericidal func-tion of neutrophils (4). Although the ROS derived from superoxide are required for the defence mechanism of neutrophils, free radicals produced in excess can damage a number of cellular structures, thus inducing lipid peroxidation and accelerating apoptosis (5). The increase in superoxide anion in the neutrophils of humans with chronic kidney disease (CKD) alters the function of endothelial cells, mesangial cells, and podocytes and reduces renal sodium flow and excretion (6,7).

The immunosuppressive effect associated with CKD has rarely been investigated in veterinary medicine. Recently, Kralova, Leva, and Toman (8) verified that, unlike that which occurs in humans, CKD in dogs does not alter the phagocytic ability of neutrophils. In contrast, Barbosa, Mori, and Ciarlini (9) verified in vitro that, as in humans, the oxidative metabolism and apoptosis of neutrophils is affected in uremic dogs. Oxidative stress due to a decrease in plasma antioxidant capacity and an increase in neutrophil oxidative metabo-lism has also been observed in cats with CKD (10). It is accepted that the viability and function of human neutrophils are affected by oxidative stress and uremic toxins (11), and there is evidence in late stages of CKD of increased apoptosis and decreased superoxide production in human neutrophils (12). However, investigators who evaluated the total and endogenous antioxidant status, such as by measuring uric acid and albumin levels, in humans with CKD reported conflicting results (13,14).

Since the progression of CKD in humans (14) and cats (10) is associated with the harmful effects of oxidative stress, it is important to investigate whether such stress also occurs in dogs and whether it alters superoxide production and neutrophil survival. Therefore, the hypothesis that oxidative stress occurs in dogs with CKD and that in this condition neutrophils become less viable and functional was tested.


Animal selectionFor the CKD group, 18 uremic adult dogs were selected that were

being treated at the Veterinary Hospital of the Araçatuba College of Veterinary Medicine, São Paulo, Brazil, and had CKD equivalent to stage III or IV in the International Renal Interest Society (IRIS) stag-ing system (15). The control group consisted of 15 adult dogs with no history of chronic or systemic disease and normal results of physi-cal examination, blood count, urinalysis, and measurement of the plasma concentrations of albumin, cholesterol, creatinine, urea, and uric acid. The experimental groups were formed after all the dogs, which were of various breeds and both sexes, had been examined on at least 2 occasions; the control animals had to show no change

in physical and laboratory findings. Dogs with a history of recent treatment that could affect renal function or leukocytes were not included. The study was approved by the Animal Experimentation Ethics Committee of the Faculty of Veterinary Medicine of São Paulo State University, and samples were obtained with consent of the owners.

Sample preparationBlood was collected from the dogs into polypropylene heparin-

ized vacuum tubes (Vacutainer Plus Plastic Sodium Heparin; Becton, Dickinson and Company, Rutherford, New Jersey, USA) that were protected from light; immediately, 4 mL was used for neutrophil isolation, and 3 mL was centrifuged to obtain plasma, which was stored at −20°C, protected from light, for a maximum of 2 wk, pend-ing biochemical analyses, quantification of lipid peroxidation, and determination of total antioxidant capacity (TAC). Another 3 mL of whole blood was collected into plastic tubes containing potassium ethylene diamine tetraacetic acid (K2EDTA) (Vacutainer Plus Plastic K2EDTA; Becton Dickinson) for a complete blood cell count, which was done immediately.

Complete blood cell count, urinalysis, and biochemical analyses

The overall concentrations of leukocytes, erythrocytes, and hemo-globin were obtained by means of an electronic veterinary blood cell counter (model CC-530; CELM, São Paulo, Brazil), and the hematocrit was determined with the Strumia microcapillary method (centrifugation for 5 min at 12 700 3 g). The differential leukocyte count was done on blood smears stained by quick Panotic dye (Instant-Prov; Newprov Produtos para Laboratórios, Pinhais, Brazil) in accordance with the guidelines and criteria outlined by Jain (16). All the biochemical tests were performed with the use of commercial reagents (BioSystems, Barcelona, Spain) at 37°C in an automated analyzer (BTS-370 plus; BioSystems) previously calibrated with com-mercial calibrator and the reactions monitored with control levels I and II. The plasma concentrations of albumin were determined by the bromocresol green assay, cholesterol by the oxidase/peroxidase enzymatic assay, creatinine by the alkaline picrate kinetic assay, urea by the urease/glutamate dehydrogenase-coupled ultraviolet enzy-matic assay, and uric acid by the uricase/peroxidase enzymatic assay.

Measurement of oxidative stressBesides determining plasma levels of the antioxidants albumin

and uric acid, the plasma TAC was quantified in an automated analyzer by a method that inhibits 2.29-azino diethylbenzothiazo-line sulfonic acid cation formation (Randox Laboratories, London, England). The TAC analyses were monitored with a standard anti-oxidant specific for automation (Randox Laboratories).

Plasma lipid peroxidation was determined by quantifying plasma thiobarbituric acid reactive substances (TBARS) by means of a com-mercial reagent kit (TBARS Assay Kit; Cayman Chemical Company, Ann Arbor, Michigan, USA). Absorbance (at 530 nm) of the reactions was measured in a plate reader (Spectra Count Reader; Packard BioScience, Meriden, Connecticut, USA) in accordance with the manufacturer’s recommendations, starting with a standard com-mercial malondialdehyde (MDA) solution (500 mM) and using a



computer program (GraphPad Prism, version 4; GraphPad Software, San Diego, California, USA). A curve for final concentrations of 0, 0.625, 1.25, 2.5, 5, 10, 25, and 50 nmol/mL of MDA was developed. Each point on the curve was obtained from the mean value for 10 repetitions.

Neutrophil isolationTo isolate neutrophils, 4 mL of whole blood containing 10 IU of

heparin per milliliter of blood was transferred to sterile conical poly-propylene tubes containing a double-gradient separation composed of equal volumes (3 mL each) of Histopaque-1119 and 1077 (Sigma Chemical Company, St. Louis, Missouri, USA). After centrifugation at 340 3 g for 30 min, the layer of polymorphonuclear (PMN) cells was aspirated and washed twice with aqueous ammonium chloride (0.14 M) for complete lysis of residual erythrocytes. Next, the sample was centrifuged (at 100 3 g) for 5 min in Hanks’ balanced salt solu-tion (Sigma) without Ca2 or Mg2, and 1 mL of RPMI medium (Sigma) was added to the cell sediment. Cell concentration was determined in a hemocytometer, and cell viability was estimated by the trypan blue exclusion method (17). The sample of isolated PMN cells was diluted in RPMI to obtain a final cell concentration of 106/mL, with the purity and viability of the neutrophils 90% and 95% or greater, respectively.

Measurement of neutrophil superoxide production

A hydroethidine (HE) (dihydroethidium bromide; Polysciences, Warrington, Pennsylvania, USA) probe was used to measure super-oxide production from the neutrophils (18). Briefly, 180 mL of neutro-phils (106/mL) suspended in RPMI medium was added to 20 mL of HE buffer solution (0.1 mmol/L) with and without 20 mL of phorbol 12-myristate 13-acetate (PMA; Sigma), 3.2 μmol/L. After incubation at 37°C for 15 min, the sample was maintained on ice and protected from light until the reading. Mean fluorescence was measured in a capillary flow cytometer (Guava EasyCyte Mini Flow Cytometry System; Guava Technologies, Hayward, California, USA) adjusted

to the wavelength of maximum emission (593 nm) and excitation (473 nm) of ethidium bromide. To eliminate debris, the neutrophil population characterized by the largest cell size was selected. Ten thousand events were acquired and analyzed with a specific computer program (Guava Express CytoSoft Data Acquisition and Analysis Software, Personal Cell Analysis System, version 4.1, 2006; Guava Technologies). Superoxide production was quantified from the intensity of the fluorescence in spontaneous and stimulated oxidation of HE.

Measurement of the apoptotic index of the neutrophils

The percentages of apoptotic and viable neutrophils were deter-mined with use of the Guava Nexin® Reagent (Guava Technologies, Millipore, Hayward, California, USA). The procedure for each test and proper instrument performance were verified in accordance with the manufacturer’s recommendations. Briefly, 2 samples of 100 mL of neutrophils (106/mL) suspended in RPMI medium were incubated at 37°C for 1 h with 100 mL of camptothecin (CAM) (19.8 mmol/L; Sigma) and with 100 mL of RPMI medium (CAM absent). After incubation, 100 mL of each sample was transferred to a microtube with 100 mL of annexin V-PE reagent and stored at room temperature, protected from light, for 10 min. In a capillary flow cytometer 10 000 events were acquired, and the size of the viable and apoptotic populations of neutrophils were determined with a specific computer program.

Statistical analysisAfter studying the distributions of the variables for normality and

homoscedasticity, as recommended by Zar (19), statistical differences between groups were determined with use of the unpaired t-test, the unpaired t-test with the Welch correction, and the Mann–Whitney test. Statistical analyses were performed with statistical software [SAS System, release 9.2 (2008); SAS Institute, Cary, North Carolina, USA]. A P-value of less than 0.05 was considered significant for all tests.

Table I. Laboratory findings in healthy control dogs and dogs with chronic kidney disease (CKD)a

Mean and standard deviationMeasure Control CKD P-valueHematocrit (%) 48.27 6 2.9b 31.26 6 13.93c , 0.0004**Leukocyte count (3 103/mL) 13 6 3.37b 21.05 6 11.51c , 0.0203**Urine specific gravity 1.044 6 0.018b 1.016 6 0.002c , 0.0105**Plasma concentration Albumin (g/L) 30.36 6 3.29b 25.73 6 3.7c , 0.0033* Cholesterol (mg/dL) 208.0 6 76.74b 299 6 121.78c , 0.0415* Creatinine (mg/dL) 1.07 6 0.13b 4.55 6 3.02c , 0.0001** Urea (mg/dL) 46.45 6 9.85b 315.21 6 159.95c , 0.0001** Uric acid (mg/dL) 1.33 6 0.59b 1.05 6 0.82b , 0.3589* TBARS (nmol/mL) 43.4 6 8.39b 21.51 6 9.39c , 0.0001*Plasma TAC (mmol/L) 2.73 6 0.24b 2.08 6 0.43c , 0.0002**a Noncoincident letters in the same row represent significant differences in the unpaired t-test (*) and the unpaired t-test with Welch’s correction (**).TBARS — thiobarbituric acid reactive substances, representing plasma lipid peroxidation; TAC — total antioxidant capacity.



Re s u l t sThe hematologic profile and results of urinalysis and plasma

biochemical analysis are summarized in Table I. The hematocrit, urine specific gravity, plasma albumin concentration, and TBARS and TAC concentrations were significantly lower in the uremic dogs than in the control dogs. The plasma concentrations of cholesterol, creatinine, and urea in the dogs with CKD were significantly above baseline, but there was no difference between the groups for uric acid. The viability (Figure 1) and superoxide production (Figure 2) of neutrophils from the dogs with CKD were lower both at rest and after stimulation with CAM (Figure 1) or PMA (Figure 2). An increase in the rate of neutrophil apoptosis was also observed in the dogs with CKD (Figure 3).

D i s c u s s i o nThe hematologic and biochemical profiles of the healthy dogs

selected for the control group remained within the reference ranges (20), demonstrating their health status. The clinical signs, increased plasma concentrations of cholesterol, creatinine, and urea, decreased plasma albumin concentration, and nonregenerative normochromic normocytic anemia observed during selection of the CKD group per-sisted for at least 2 wk, demonstrating IRIS stage III or IV of kidney disease at the time of assessment (15).

Although the plasma concentration of the antioxidant uric acid was lower in the dogs with CKD than in the control group, the differ-ence was not statistically significant. The results for plasma uric acid concentration in human patients with kidney disease are conflicting, probably owing to the use of different methodologies (13,14).

The lower level of plasma albumin observed in the dogs with CKD could have contributed to their lower TAC. According to Terawaki et al (21), albumin exerts an important antioxidant role in CKD-related oxidative stress.

The TAC was significantly lower in the dogs with CKD than in the healthy control group, confirming that the oxidative stress previ-ously described in humans (13,14) and cats (10) also occurs in dogs. The observed difference supports the hypothesis that an antioxidant other than uric acid is decreased in dogs with CKD.

The lower antioxidant capacity did not promote the expected increase in TBARS. On the contrary, lipid peroxidation in the plasma was lower in the dogs with CKD than in the control group. The lower production of superoxide in the dogs with CKD may have contributed to reduced formation of TBARS. Although TBARS mea-surement is a commonly reported method of detecting MDA, it is insufficiently sensitive and is affected by interference from related species or overestimation due to stressing conditions during sample manipulation (22).

The lower superoxide production in the dogs with CKD fully supports the hypothesis that uremic toxins in dogs with IRIS

Figure 1. Viability of neutrophils from healthy dogs (Normal) and dogs with chronic kidney disease (CKD) not activated (A) or activated (B) by camptothecin (CAM). The values of upper quartile, median, and lower quartile are indicated in each box, and the bars outside the box indicate semiquartile ranges. Differences between the 2 groups of dogs signifi-cant at *P = 0.0026 and **P = 0.0022.

Figure 2. Superoxide production measured by the mean fluorescence of ethidium bromide in neutrophils from healthy dogs (Normal) and dogs with chronic kidney disease (CKD) not activated (A) or activated (B) by phorbol 12-myristate 13-acetate. The values of upper quartile, median, and lower quartile are indicated in each box, and the bars outside the box indicate semiquartile ranges. Differences between the 2 groups of dogs significant at *P = 0.0138 and **P = 0.0022.

Figure 3. Total apoptosis of neutrophils from healthy dogs (Normal) and dogs with chronic kidney disease (CKD) not activated (A) or activated (B) by camptothecin (CAM). The values of upper quartile, median, and lower quartile are indicated in each box, and the bars outside the box indicate semiquartile ranges. Differences between the 2 groups of dogs significant at *P = 0.0033 and **P = 0.0066.



stage III or IV kidney disease contribute to the diminished oxidative metabolism of the neutrophils, a potentially important mechanism that affects the innate immune response of dogs with renal failure, as previously reported for humans (23) and cats (10), as well as from in vitro experiments involving dog cells (9). However, this result is in disagreement with the observation by McLeish et al (24) and Rysz et al (3) of an increase in ROS production in human patients with uremia and the lack of observation of alterations in neutrophil oxidative metabolism in other studies of such patients (2,25,26). The results of investigations into the oxidative metabolism of human neutrophils in CKD remain highly contradictory; however, some of these differences are known to be due to the methods of neutrophil isolation and quantification of ROS (23).

As in humans (27), CKD in dogs causes a significantly higher rate of spontaneous and induced neutrophil apoptosis than in healthy controls. Our results contradict the report by Kralova, Leva, and Toman (8) that CKD does not alter the production of reactive oxygen radicals in neutrophils from dogs. Those investigators used chemi-luminescence, a method that is not specific for superoxide detection, as well as different procedures for isolation and incubation of the neutrophils. The results of the current study are consistent with the observations reported by Barbosa, Mori, and Ciarlini (9), who obtained in vitro evidence that uremic toxins in dogs promote the initial activation of the oxidative metabolism of neutrophils that subsequently induces the acceleration of apoptosis and diminished superoxide production. These findings support the hypothesis of Cendoroglo et al (12) that in the early stages of CKD an increase in the oxidative metabolism of neutrophils occurs, such that oxidants accumulate and eventually cause cell damage capable of accelerat-ing apoptosis and affecting superoxide production at a later stage.

How the mechanisms of oxidative stress in late stages of CKD accelerate apoptosis and diminish the oxidative metabolism of neutrophils in dogs remains to be determined.

In conclusion, oxidative stress occurs in dogs with CKD, as dem-onstrated by the lower plasma TAC in those dogs as compared with healthy controls. This change occurs concomitantly with inhibition of oxidative metabolism and acceleration of neutrophil apoptosis.

A c k n o w l e d g m e n tThis work was graciously funded by the São Paulo Research

Foundation.

Re f e r e n c e s1. Vanholder R, Van Laecke S, Glorieux G. What is new in uremic

toxicity? Pediatr Nephrol 2008;23:1211–1221.2. Anding K, Gross P, Rost JM, et al. The influence of uraemia and

haemodialysis on neutrophil phagocytosis and antimicrobial killing. Nephrol Dial Transplant 2003;18:2067–2073.

3. Rysz J, Kasielski M, Apanasiewicz J, et al. Increased hydrogen peroxide in the exhaled breath of uraemic patients unaffected by haemodialysis. Nephrol Dial Transplant 2004;19:158–163.

4. Huimin J, Qinjun X, Peijun Z, et al. Impaired GP-91PHOX gene expression and dysfunction of peripheral blood neutrophils in patients maintaining hemodialysis. Chin Med J 2000;113:120–123.

5. Kato S, Chmielewski M, Honda H, et al. Aspects of immune dysfunction in end-stage renal disease. Clin J Am Soc Nephrol 2008;3:1526–1533.

6. Vaziri ND, Dicus M, Ho ND, et al. Oxidative stress and dysregu-lation of superoxide dismutase and NADPH oxidase in renal insufficiency. Kidney Int 2003;63:179–185.

7. Nistala R, Whaley-Connell A, Sowers JR. Redox control of renal function and hypertension. Antioxid Redox Signal 2008;10: 2047–2089.

8. Kralova S, Leva L, Toman M. Polymorphonuclear function in naturally occurring renal failure in dogs. Vet Med (Praha) 2009;54:236–243.

9. Barbosa TS, Mori CK, Ciarlini PC. P.C. Efeito inibidor do soro urêmico sobre o metabolismo oxidativo dos neutrófilos de cães. Arq Bras Med Vet Zootec 2010;62:1352–1358.

10. Keegan RF, Webb CB. Oxidative stress and neutrophil function in cats with chronic renal failure. J Vet Intern Med 2010;24: 514–519.

11. Chonchol M. Neutrophil dysfunction and infection risk in end-stage renal disease. Semin Dial 2006;19:291–296.

12. Cendoroglo M, Bertrand LJ, Balakrishnan VS, Perianayagam M, King AJ, Pereira BJG. Neutrophil apoptosis and dysfunction in uremia. J Am Soc Nephrol 1999;10:93–100.

13. Clermont G, Lecour S, Lahet JJ, et al. Alteration in plasma anti-oxidant capacities in chronic renal failure and hemodialysis patients: A possible explanation for the increased cardiovascular risk in these patients. Cardiovasc Res 2000;47:618–623.

14. Erdogan C, Ünlücerci Y, Türkmen A, et al. The evaluation of oxidative stress in patients with chronic renal failure. Clin Chim Acta 2002;322:157–161.

15. International Renal Interest Society. IRIS Staging of CKD. 2006. Available at http://www.iris-kidney.com/guidelines/en/ staging_ckd.shtml (last accessed 2013 Jan 11).

16. Jain NC. Schalm’s Veterinary Hematology. 4th ed. Philadelphia: Lea & Febiger, 1986:5p.

17. Metcalf JA, Gallin JI, Nauseef WM, Root RK. Laboratory Manual of Neutrophil Function. New York: Raven Press, 1986:10p.

18. Walrand S, Valeix S, Rodriguez C, et al. Flow cytometry study of polymorphonuclear neutrophil oxidative burst: A comparison of 3 fluorescent probes. Clin Chim Acta 2003;331:103–110.

19. Zar JH. Biostatistical Analysis. 2nd ed. Englewood Cliffs, New Jersey: Prentice Hall, 1984:93–95.

20. Stockham SL, Scott MA. Fundamentals of Veterinary Clinical Pathology. Ames, Iowa: Iowa State University Press, 2002:384.

21. Terawaki H, Yoshimura K, Hasegawa T, et al. Oxidative stress is enhanced in correlation with renal dysfunction: Examination with the redox state of albumin. Kidney Int 2004;66:1988–1993.

22. Del Rio D, Stewart AJ, Pellegrini N. A review of recent studies on malondialdehyde as toxic molecule and biological marker of oxidative stress. Nutr Metab Cardiovasc Dis 2005;15:316–328.

23. Sardenberg C, Suassuna P, Andreoli MCC, et al. Effects of urae-mia and dialysis modality on polymorphonuclear cell apoptosis and function. Nephrol Dial Transplant 2006;21:160–165.

24. McLeish KR, Klein JB, Lenderer EL, et al. Azotemia, TNF alpha, and LPS prime the human neutrophil oxidative burst by distinct mechanisms. Kidney Int 1996;50:407–416.



25. Paul JL, Roch-Arveiller M, Man NK, Luong N, Moatti N, Raichvarg D. Influence of uremia on polymophonuclear leu-kocytes oxidative metabolism in end-stage renal disease and dialyzed patients. Nephron 1991;57:428–432.

26. Gastadello K, Husson C, Wens R, et al. Role of complement and platelet-activating factor in the stimulation of phagocytosis

and reactive oxygen species production during haemodialysis. Nephrol Dial Transplant 2000;15:1638–1646.

27. Majewska E, Baj Z, Sulowska Z, et al. Effects of uraemia and hae-modialysis on neutrophil apoptosis and expression of apoptosis-related proteins. Nephrol Dial Transplant 2003;18:2582–2588.


Article


Correlation of tumor-infiltrating lymphocytes to histopathological features and molecular phenotypes in canine mammary carcinoma:

A morphologic and immunohistochemical morphometric studyJong-Hyuk Kim, Seung-Ki Chon, Keum-Soon Im, Na-Hyun Kim, Jung-Hyang Sur

A b s t r a c tAbundant lymphocyte infiltration is frequently found in canine malignant mammary tumors, but the pathological features and immunophenotypes associated with the infiltration remain to be elucidated. The aim of the present study was to evaluate the relationship between lymphocyte infiltration, histopathological features, and molecular phenotype in canine mammary carcinoma (MC). The study was done with archived formalin-fixed, paraffin-embedded samples (n = 47) by histologic and immunohistochemical methods. The degree of lymphocyte infiltration was evaluated by morphologic analysis, and the T- and B-cell populations as well as the T/B-cell ratio were evaluated by morphometric analysis; results were compared with the histologic features and molecular phenotypes. The degree of lymphocyte infiltration was significantly higher in MCs with lymphatic invasion than in those without lymphatic invasion (P , 0.0001) and in tumors of high histologic grade compared with those of lower histologic grade (P = 0.045). Morphometric analysis showed a larger amount of T-cells and B-cells in MCs with a higher histologic grade and lymphatic invasion, but the T/B ratio did not change. Lymphocyte infiltration was not associated with histologic type or molecular phenotype, as assessed from the immunohistochemical expression of epidermal growth factor receptor 2, estrogen receptor, cytokeratin 14, and p63. Since intense lymphocyte infiltration was associated with aggressive histologic features, lymphocytes may be important for tumor aggressiveness and greater malignant behavior in the tumor microenvironment.

R é s u m éUne infiltration lymphocytaire abondante est fréquemment retrouvée dans les tumeurs mammaires malignes chez le chien, mais les caractéristiques pathologiques et les immunophénotypes associés avec l’infiltration restent à être élucidés. L’objectif de la présente étude était d’évaluer la relation entre l’infiltration lymphocytaire, les caractéristiques histopathologiques, et le phénotype moléculaire dans les carcinomes mammaires canins (CM). Cette étude a été réalisée en utilisant des méthodes histologiques et immunohistochimiques sur des échantillons archivés fixés à la formaline et enrobés de paraffine (n = 47). Le degré d’infiltration lymphocytaire a été évalué par analyse morphologique, et les populations de lymphocytes T et B ainsi que le ratio de cellules T/B ont été évalués par analyses morphométriques; les résultats ont été comparés avec les caractéristiques histologiques et les phénotypes moléculaires. Le degré d’infiltration lymphocytaire était significativement plus élevé dans les CM avec invasion lymphatique que dans ceux sans invasion lymphatique (P , 0,001) et dans les tumeurs de grade histologique élevé comparativement à ceux avec un grade histologique faible (P = 0,045). Les analyses morphométriques ont montré une quantité plus grande de cellules T et B dans les CM ayant un grade histologique élevé et invasion lymphatique, mais le ratio T/B n’a pas changé. L’infiltration lymphocytaire n’était pas associée avec le type histologique ou le phénotype moléculaire, tel qu’évalué par l’expression immunohistochimique du récepteur 2 du facteur de croissance épidermique, du récepteur d’estrogène, de cytokératine 14, et de p63. Étant donné que l’infiltration lymphocytaire marquée était associée avec des caractéristiques histologiques d’agressivité, les lymphocytes pourraient être importants pour l’agressivité des tumeurs et le comportement de malignité plus important dans le microenvironnement de la tumeur.


Small Animal Tumor Diagnostic Center, Department of Veterinary Pathology, College of Veterinary Medicine, Konkuk University, 1 Hwayang-dong, Kwangjin-gu, Seoul 143-701, Korea (J.-H. Kim, Im, N.-H. Kim, Sur); Hyu Animal Clinic, Iksan, Jeollabuk-do, Korea (Chon).

Address all correspondence to Dr. Jung-Hyang Sur; telephone: 82-2-450-4153; fax: 82-2-455-8124; e-mail: [email protected]

Dr. Seung-Ki Chon contributed equally to this work.

Dr. Jong-Hyuk Kim’s current address is the Department of Veterinary Clinical Sciences, College of Veterinary Medicine, University of Minnesota, St. Paul, Minnesota 55108, USA.

Received December 19, 2011. Accepted April 5, 2012.



I n t r o d u c t i o nThe relationship between inflammation and cancer has been

evaluated because the immune system is important in the tumor microenvironment (1). Large numbers of leukocytes infiltrate solid and metastasized tumors (2). Lymphocyte infiltration primar-ily stimulates the host immune response against tumors (3), and T-lymphocytes constitute most of the lymphocyte infiltration in human breast cancer (HBC) (2). Recently it has been suggested that a balance between the adaptive and innate immune responses mediated by lymphocytes may affect HBC progression and regres-sion (4). Cytotoxic T-lymphocytes (CTLs) kill tumor cells directly, and T helper 1 (Th1) cells, which secrete antitumor cytokines, exert mainly antitumor effects (4–6). Regulatory T (Treg-cells) and T helper 2 (Th2) cells promote tumor development by suppressing the antitumor immune response (4,7,8).

Abundant lymphocyte infiltration is frequently found not only in HBC but also in canine malignant mammary carcinoma (MC) (9). Canine MC may be regarded as a potent, spontaneous animal model of HBC (10,11) because canine MC and HBC have similar biologic and pathological features (11). Studies have shown that various fea-tures, including histologic features, clinical course, hormone levels, molecular markers, proliferation markers, and genetic mutations, are similar in canine MC and HBC (11). Hence, investigating the pathogenesis and biologic features of canine mammary tumors could improve our understanding of HBC and help identify new thera-peutics (11). Recently some articles reported the phenotypic features and prognostic implications of lymphocytes; the level of lymphocyte infiltration may be an important prognostic biomarker for canine MC (12–14). However, the identification of tumor-infiltrating lympho-cytes and the relationship between various pathological features of canine MC remain to be investigated.

Some authors have also reported on the identification and applica-tion of molecular phenotypes of canine MC because of the hetero-geneity of this disease according to human classifications (15,16). The molecular phenotypes of HBC correlate with prognosis and clinical outcome and are classified as follows: luminal A [estrogen receptor (ER) and HER (epidermal growth factor)-2–]; luminal B (ER/HER-2); HER-2-overexpressing (ER–/HER-2); basal (ER–/HER-2–/basal cell marker); and negative/null (ER–/HER-2–/basal cell marker–) (15–17). Luminal phenotypes are based on positive expression of ER, and basal phenotypes are determined by negative expression of ER and basal cell markers such as cytokeratin (CK) 5/6, CK14, p63, and P-cadherin (15–17). In human medicine, patients with HER-2-overexpressing cells and basal phenotypes have significantly shorter survival than those with luminal phenotypes (18,19). In canine MC, 2 previous reports correlated molecular phe-notypes with survival data; however, the results were inconsistent: 1 study showed that the basal phenotype was significantly associated with lower survival rates (15), whereas the other study found that patients with the basal phenotype had a better outcome than those with the luminal phenotypes, those with the luminal B phenotype having a worse outcome than those with the luminal A phenotype (16). A high degree of lymphocyte infiltration correlated with HER-2-positive HBC (20), and HER-2 may be an appropriate target for immunotherapy because T-lymphocytes respond to HER-2 peptides

(21,22). Therefore, lymphocyte infiltration is required to explore and investigate the relationship between molecular phenotypes in MC.

Because canine MC is a heterogeneous disease, it is challenging to explain the precise biologic features of MC, make a prognosis, and apply effective therapy; thus, specific tumor behavior and bio-logic features must be examined. In addition, the involvement of the immune response in the tumor microenvironment was recently highlighted in both human and animal cancer. Therefore, the aim of this study was to investigate whether lymphocytes infiltrate and affect a specific biologic condition or subtype of canine MC with heterogeneous biologic features by determining the total lymphocyte count, the T- and B-cell populations, and the T/B-cell ratio, correlat-ing lymphocyte infiltration with histologic variables, and correlating the results with the molecular phenotypes.


SamplesAll mammary samples were originally obtained from the

Veterinary Medical Teaching Hospital of Konkuk University, Seoul, Korea, or from private animal clinics. The samples had been submit-ted between 2006 and 2008 to the Small Animal Tumor Diagnostic Center, Department of Veterinary Pathology, Konkuk University. The 47 MC samples were obtained from purebred or mixed-breed female dogs aged 2 to 19 y [mean 10.6 6 3.9 (standard deviation) y; median 10.6 y]. Samples of normal mammary tissue from 3 clini-cally healthy dogs were used as negative controls for HER-2 and as positive controls for the CK and p63 immunostaining.

Histologic examination and classificationHistopathological examination was conducted on 4-mm-thick

sections of formalin-fixed, paraffin-embedded samples of canine mammary gland tissue that had been stained with hematoxylin and eosin (H&E). Canine MCs were classified according to the World Health Organization International Histological Classification of Tumors of Domestic Animals (23). Evidence of lymphatic invasion and the presence of tumor necrosis were recorded. The tumors were classified histologically as well-differentiated (grade I), moderately differentiated (grade II), or poorly differentiated (grade III) carcino-mas according to the Elston and Ellis grading system (24,25).

Immunohistochemical examination and scoringPrimary antibodies included anti-CD3, anti -CD79a

(DakoCytomation, Glostrup, Denmark), anti-HER-2, anti-ER (BioGenex, Fremont, California, USA), anti-CK14 (Abcam, Cambridge, England), and anti-p63 (Santa Cruz Biotechnology, Santa Cruz, California, USA). Sections were dewaxed in xylene and hydrated in graded ethanol. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide in phosphate-buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 2 mM KH2PO4), pH 7.4, for 20 min at room temperature (RT), and then 3 washes in PBS were done. Antigen retrieval was accomplished by boiling CD3, ER, CK14, and p63 in tris-ethylene diamine tetraacetic acid buffer (pH 9.0) and CD79a in citric acid buffer (pH 6.0) in a microwave oven (at 750 W). Slides with ER were incubated with



5% normal goat serum for 30 min at RT. Sections were subsequently incubated with the primary antibodies, and primary antibody bind-ing was detected with the 2-step EnVision horseradish peroxidase system (REAL EnVision Kit; DakoCytomation). The secondary polymer was applied to each slide for 40 min at RT, and the slides were washed 4 times with PBS before incubation with the appropri-ate substrates. The colorimetric reaction was stopped by 2 washes in distilled water, and the sections were counterstained with Harris hematoxylin. The primary antibodies, including isotype antibodies, are listed in Table I. The isotype negative controls were as follows: mouse IgG1 (eBioscience, San Diego, California, USA) for CD79a, HER, and ER; rabbit immunoglobulin fraction (DakoCytomation) for CD3; IgG3 (Abcam) for CK14; and IgG2a (BioLegend, San José, California, USA) for p63. The following positive controls were used: canine lymph node for CD3/CD79a, uterus for ER, myoepithelial cells from a normal mammary gland for CK14 and p63, and mam-mary carcinoma tissues for HER-2 overexpression, as in our previous study of HER-2 (26).

The immunohistochemical score was assigned through semi-quantitative analysis. The HER-2 immunoreactivity was scored on

a 4-point scale (0 to 3) according to the HercepTest scoring system (DakoCytomation) (27): 0 = no staining or membrane staining of less than 10% of the tumor cells; 1 = weak, incomplete membrane staining; 2 = weak or moderate, complete membrane staining of more than 10% of the tumor cells; and 3 = strong, complete mem-brane staining of more than 10% of the tumor cells. Scores of 0 and 1 were considered negative, whereas scores of 2 and 3 were consistent with HER-2 overexpression. The ER immunoreactivity was considered positive when more than 10% of the tumor cells had immunolabeled nuclei (15). Positivity for CK14 was defined as the detection of strong cytoplasmic staining in more than 1% of the invasive tumor cells (18), and positivity for p63 was defined as the detection of immunolabeling in the nuclei of more than 50% of the neoplastic cells (15). The percentage of positive cells was evaluated in 10 representative high-power fields (each with more than 1000 cells).

Morphologic and morphometric analysis of lymphocyte infiltration

In the H&E sections the degree of lymphocytic infiltration was morphologically classified by distribution and intensity in the

Table I. Primary antibodies used for immunohistochemical analysis of canine mammary carcinomas (MCs)

Target Antibodyantigen type Clone Sourcea Antigen retrieval Dilution Incubation IsotypeCD3 Pab IgG fraction DakoCytomation HIER for 10 min 1:150 2.5 h, RT Rabbit Ig fractionCD79a Mab HM57 DakoCytomation HIER for 10 min 1:150 Overnight, 4°C Mouse IgG1HER-2 Mab CB11 BioGenex None 1:100 3 h, RT Mouse IgG1ER Mab ER88 BioGenex HIER for 20 min 1:60 3 h, RT Mouse IgG1CK14 Mab LL002 Abcam HIER for 10 min 1:300 3 h, RT Mouse IgG3p63 Mab 4A4 Santa Cruz Biotechnology HIER for 15 min 1:100 Overnight, 4°C Mouse IgG2aa DakoCytomation, Glostrup, Denmark; BioGenex, Fremont, California, USA; Abcam, Cambridge, England; Santa Cruz Biotechnology, Santa Cruz, California, USA.HER-2 — epidermal growth factor receptor 2; ER — estrogen receptor; CK14 — cytokeratin 14; Pab — rabbit polyclonal antibody; Mab — mouse monoclonal antibody; Ig — immunoglobulin; HIER — heat-induced epitope retrieval; RT — room temperature.

Figure 1. Lymphocyte infiltration classified according to morphologic analysis. A — Inflammatory cells (asterisk) moderately infiltrate the stroma near the tumor cells (arrows). The degree of lymphocyte infiltration is 2. B — High-density lymphocyte infiltration (asterisk) around tumor nests (arrows). The degree of lymphocytic infiltration is 3. Hematoxylin and eosin. Bar = 70 mm.



intratumoral areas according to previous studies (12,14). Distribution was classified as follows: focal = presence of 1 to 3 inflammatory foci; multifocal = presence of more than 3 inflammatory foci; and diffuse = even distribution of inflammatory cells in the tumor section. Density was classified as follows: 1 = discrete; 2 = moderate; and 3 = intense. The scores for distribution and intensity were multiplied. In this classification system the infiltration degree varies between 1 and 9 and is classified as follows: degree 1 = score 1 to 3; degree 2 = score 4 to 6; and degree 3 = score 7 to 9 (Figures 1A and 1B).

Morphometric image analysis was done by a method described previously (28). Digital images were acquired with an Olympus microscope (model BX41; Olympus, Tokyo, Japan) and digi-

tal image transfer software (Leica Application Suite 2.7; Leica Microsystems, Heerbrugg, Germany). Morphometric analysis of T- and B-lymphocytes was accomplished with computerized image analysis software (Image Pro Plus 5.1; Media Cybernetics, Bethesda, Maryland, USA) according to a previous study (28). For each slide, 20 fields of immunolabeled images were acquired at a magnification of 4003 (403 objective, 103 ocular). The positive areas per 1.6 mm2 were calculated by converting the positive pixels into square milli-meters; the areas included the nuclear areas of the lymphocytes. The CD3-positive area divided by the CD79a-positive area represented the CD3/CD79a ratio. The total T/B-lymphocyte population was calculated by adding the CD3-positive and CD79a-positive areas.

Figure 2. Overexpression of epidermal growth factor 2 leads to strong, complete staining of the tumor cell membrane according to the HercepTest scoring system (DakoCytomation, Glostrup, Denmark). Intratumoral necrosis is also observed (asterisk). EnVision horse-radish peroxidase immunohistochemistry system (REAL EnVision Kit; DakoCytomation) with hematoxylin counterstain. Bar = 36 mm.

Figure 4. Tumor cells are diffusely immunolabeled with cytokeratin 14 in the cytoplasm. EnVision system with hematoxylin counterstain. Bar = 36 mm.

Figure 3. Strong immunoreaction for estrogen receptors in the nuclei of the tumor cells. EnVision system with hematoxylin counterstain. Bar = 36 mm.

Figure 5. Positive p63 immunostaining in the nuclei of the tumor cells. EnVision system with hematoxylin counterstain. Bar = 36 mm.



Molecular phenotypingThe MC molecular phenotypes were defined according to a

modified classification system (15,16) as follows: luminal A = ER, HER-2–, and any CK14 or p63; luminal B = ER, HER-2, and any CK14 or p63; HER-2-overexpressing = ER–, HER-2, and any CK14 or p63; basal = ER–, HER-2–, CK14, and/or p63); and negative/null = ER–, HER-2–, CK14–, and p63–.

Statistical analysisDifferences in the degree of lymphocyte infiltration, histologic

categorical variables, and molecular phenotypes were evaluated with Pearson’s chi-square test for categorical analysis. The normal-ity of the distribution was assessed by the Kolmogorov–Smirnov test, and the associations between the means for the 2 groups were examined by Student’s t-test or the Mann–Whitney U-test. Analysis of variance or the Kruskal–Wallis test was used to compare more than 2 groups. Statistical significance was established at P , 0.05. Statistical analysis was done with the use of SPSS, version 17.0 (SPSS, Chicago, Illinois, USA).

Re s u l t sThe canine MCs included simple carcinomas (n = 22), complex

carcinomas (n = 3), carcinomas in benign mixed tumors (n = 7), and mammary squamous cell carcinomas (n = 15). Lymphatic invasion of tumor cells was identified in 21 cases. Of the 47 tumors, 11 were grade I, 14 grade II, and 22 grade III. The HER-2 protein was detected

by immunohistochemical staining of the membrane of tumor cells (Figure 2) in 16 (34%) of the 47 MCs. Nuclear ER immunoreactivity (Figure 3) was positive in 21 (45%) of the 47 MCs, and positivity for CK14 or p63 was detected in 76%, as determined from immunola-beling in the cytoplasm and the nuclei of the tumor cells, respec-tively (Figures 4 and 5). From the immunohistochemical staining, the molecular phenotypes of the MCs were classified as follows: luminal A, 14; luminal B, 7; basal, 15; HER-2-overexpressing, 9; and negative/null, 2.

The degree of lymphocyte infiltration was not associated with his-tologic type of the MC (P = 0.042), presence of intratumoral necrosis (P = 0.178), or molecular phenotype (P = 0.052) according to mor-phologic analysis (Table II). However, a high degree of lymphocyte infiltration (3) was most frequent in the squamous cell subtype, whereas a low degree (1) was most frequent in the simple subtype. The degree of lymphocyte infiltration was significantly higher in MCs with lymphatic invasion than in those without lymphatic inva-sion (P , 0.0001) and in tumors of high histologic grade compared with those of low histologic grade (P = 0.045).

Morphometric image analysis showed significantly greater infil-tration of CD3 T-lymphocytes in MCs of high histologic grade (P = 0.035) and MCs undergoing lymphatic invasion (P = 0.008) and of CD79a B-lymphocytes in high-grade MCs (P = 0.011) and tumors undergoing lymphatic invasion (P = 0.001), compared with MCs of lower histologic grade and those not invading the lymphat-ics (Table III). Total lymphocytes were abundant in high-grade MCs (P = 0.020) and those with lymphatic invasion (P = 0.005). Again, no statistical associations were observed between lymphocyte

Table II. Evaluation of the degree of lymphocyte infiltration by morphologic analysis of histologic features and molecular phenotype

Degree of infiltration; number (and %) of MCsVariable (n = 47) 1 2 3 P-valuea

Histologic type NS Simple (n = 22) 10 (45) 4 (18) 8 (36) Complex (n = 3) 1 (33) 1 (33) 1 (33) Mixed (n = 7) 2 (29) 3 (43) 2 (29) Squamous cell (n = 15) 1 (7) 7 (47) 7 (47)Lymphatic invasion P , 0.0001 Presence (n = 21) 1 (5) 6 (28) 14 (67) Absence (n = 26) 13 (50) 9 (35) 4 (15)Histologic grade I (n = 11) 7 (64) 3 (27) 1 (9) P = 0.045 II (n = 14) 4 (28) 4 (28) 6 (43) III (n = 22) 3 (14) 8 (36) 11 (50)Necrosis NS Presence (n = 27) 5 (18) 10 (37) 12 (44) Absence (n = 20) 9 (45) 5 (25) 6 (30)Molecular phenotype NS Luminal A (n = 14) 7 (50) 5 (36) 2 (14) Luminal B (n = 7) 1 (14) 3 (43) 3 (43) Basal (n = 15) 3 (20) 4 (27) 8 (53) HER-2-overexpressing (n = 9) 2 (22) 3 (33) 4 (44)a With Pearson’s chi-square test. NS — not significant.



infiltration and histologic type, necrosis, and molecular phenotype. The histologic features and molecular phenotypes were not associ-ated with the CD3/CD79a ratio.

D i s c u s s i o nMammary gland tumors are the most common neoplasms in

female dogs and are a heterogeneous group with different prognoses (29). Although the classification of histologic features and biologic and molecular identification for making a prognosis have been explored, the immune response and the relationship between tumor behavior and lymphocyte infiltration are still unclear.

In this study, lymphocyte infiltration of canine MC tissues was assessed by morphologic and morphometric image analyses, and the results from these 2 methods consistently showed that lymphocyte infiltration was associated with an aggressive histologic grade and lymphatic invasion. Morphologic evaluation revealed that a high degree of lymphocyte infiltration was significantly correlated with lymphatic invasion and a high histologic grade. A previous mor-phologic analysis of inflammatory cells found that a high intensity of lymphocyte infiltration was associated with lymph node metas-tasis, advanced clinical stage, and low survival rate (12); our results are consistent with these results. Although we have no clinical or survival data, we suggest that an abundance of tumor-infiltrating lymphocytes in canine MC tissue is associated with aggressive tumor behavior and a higher degree of malignancy.

By morphometric image analysis, greater numbers of CD3 T-cells, CD79a B-cells, and total lymphocytes were observed in MCs of high histologic grade with lymphatic invasion. Recent

studies reported that an abundance of intratumoral CD3 T-cells correlated with tumor invasiveness and shorter overall survival in canines with malignant mammary tumors (13,14). One of those studies showed that the number of T-cells increased more in the malignant mammary tumors than in benign tumors (14), whereas the other study showed that the number of intratumoral T-cells was higher in benign tumors than in malignant tumors (13). The T-cell count is likely different in benign and malignant tumors; however, for malignant tumors, a high degree of T-cell infiltration was cor-related with a poor prognosis in both studies. In addition, the latter study showed that a greater abundance of intratumoral T-cells was associated with tumor invasiveness (13); our T-cell infiltration results are consistent with this observation. For B-cells, the infiltration likely increases in canine MCs compared with benign tumors, although this difference is not statistically significant (14). Another study showed that the percentage of B-cells in MCs is higher in dogs with lymph node metastasis (12). In HBC, B-cell infiltration is associated with progression of malignant tumors (30). Because our results revealed that more B-cells are observed in high-grade MCs with lymphatic invasion, we suggest that greater infiltration of B-cells is associated with aggressive tumor behavior in canine MC.

In addition, the data for total infiltration of T- and B-cells obtained by morphometric analysis were similar to those obtained by mor-phologic analysis: greater lymphocyte infiltration was associated with higher histologic grade and increased lymphatic invasion. This suggests that the infiltrating lymphocytes are crucial for tumor malignancy in canine MC and supports the HBC reports that abundant lymphocyte infiltration in high-grade and invasive carcinomas correlates with metastasis and poor prognosis (4,31,32).

Table III. Comparison of lymphocyte subpopulations by morphometric analysis of histologic features and molecular phenotype

Lymphocyte subpopulation, positive area/1.6 mm2 (mean 6 standard deviation)a

Variable CD3 T-cells CD79a B-cells Total T/B-cells CD3/CD79a ratioHistologic type NS* NS* NS* NS†

Simple 0.078 6 0.102 0.027 6 0.052 0.106 6 0.143 0.277 6 0.79 Complex 0.028 6 0.032 0.008 6 0.012 0.036 6 0.033 14.38 6 0.54 Mixed 0.043 6 0.055 0.011 6 0.012 0.055 6 0.066 2.73 6 0.58 Squamous cell 0.119 6 0.120 0.028 6 0.035 0.147 6 0.144 3.36 6 0.95Lymphatic invasion P = 0.008‡ P = 0.001‡ P = 0.005‡ NS§

Presence 0.118 6 0.112 0.038 6 0.049 0.157 6 0.148 2.70 6 0.72 Absence 0.055 6 0.085 0.013 6 0.031 0.067 6 0.105 3.92 6 1.02Histologic grade P = 0.035* P = 0.011* P = 0.020* NS†

I 0.025 6 0.047 0.003 6 0.004 0.029 6 0.047 3.05 6 1.04 II 0.072 6 0.087 0.023 6 0.040 0.096 6 0.119 4.31 6 0.99 III 0.118 6 0.119 0.035 6 0.049 0.153 6 0.152 2.72 6 0.73Necrosis NS‡ NS‡ NS‡ NS§

Presence 0.103 6 0.113 0.027 6 0.045 0.131 6 0.143 3.26 6 0.85 Absence 0.055 6 0.080 0.020 6 0.037 0.075 6 0.113 3.01 6 0.84Molecular phenotype NS* NS* NS* NS†

Luminal A 0.028 6 0.043 0.005 6 0.007 0.033 6 0.045 5.84 6 0.59 Luminal B 0.133 6 0.141 0.049 6 0.052 0.182 6 0.181 2.53 6 0.76 Basal 0.111 6 0.111 0.038 6 0.059 0.148 6 0.157 2.91 6 0.83 HER-2-overexpressing 0.076 6 0.095 0.016 6 0.015 0.092 6 0.097 2.09 6 0.76a Statistical significance determined by * the Kruskal–Wallis test, † analysis of variance, ‡ the Mann–Whitney U-test, and § Student’s t-test.



We investigated whether the relative amount of T- and B-cells in the infiltration is related to the immune response; however, no tumor features correlated with the relative amount of T- and B-cells. Additional insights will likely require the investigation of T-cell subpopulations, including CTLs and Treg- and Th-cells, rather than estimation of the ratio of T- and B-cells.

Furthermore, the lymphocyte infiltration data were statistically evaluated to investigate a possible association with molecular phe-notype because some reports have shown a correlation between molecular phenotype and tumor behavior or prognosis in both HBC and canine MC (15,16,33). Basal and HER-2-overexpressing pheno-types are aggressive and correlate with a poor prognosis for HBC patients (18,19,33). In canine MCs, although controversial results have been reported, the basal phenotype correlating with a high histologic grade (15) or a low grade (16), the molecular phenotype correlation might provide significant information on the biologic and pathological features of MCs. However, morphologic and morpho-metric analyses failed to show an association between lymphocyte infiltration and molecular phenotype in the present study. Further studies are thus required to clarify whether molecular phenotype correlates with prognosis and whether lymphocytes are involved in determining the molecular phenotype of canine MCs.

In conclusion, a large number of lymphocytes, including T- and B-cells, consistently infiltrate canine MCs and are associated with lymphatic invasion and a high histologic grade. The results of this study suggest that tumor-infiltrating lymphocytes are important for tumor progression and metastatic potential and not tumor killing or regression. This is the first report of the histopathological features of lymphocyte infiltration in different molecular phenotypes. Although the distinct functions of T- and B-cells remain to be determined — including by studies of the involvement of subpopulations of T-cells and the clinical implications of lymphocyte infiltration — this study has provided important data that support the relationship between lymphocytes and cancer in dogs.

A c k n o w l e d g m e n t sThis study was supported by Konkuk University in 2012. The

authors thank Mrs. Rae-Hwa Jang for excellent technical assistance. They also acknowledge the private veterinary clinics for providing the canine mammary samples.

Re f e r e n c e s 1. Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;

420:860–867. 2. O’Sullivan C, Lewis CE. Tumour-associated leucocytes: Friends

or foes in breast carcinoma. J Pathol 1994;172:229–235. 3. Ben-Baruch A. Host microenvironment in breast cancer develop-

ment: Inflammatory cells, cytokines and chemokines in breast cancer progression: Reciprocal tumor–microenvironment interac-tions. Breast Cancer Res 2003;5:31–36.

4. DeNardo DG, Coussens LM. Inflammation and breast cancer. Balancing immune response: Crosstalk between adaptive and innate immune cells during breast cancer progression. Breast Cancer Res 2007;9:212.

5. Moss RB, Moll T, El-Kalay M, et al. Th1/Th2 cells in inflamma-tory disease states: Therapeutic implications. Expert Opin Biol Ther 2004;4:1887–1896.

6. Munk ME, Emoto M. Functions of T-cell subsets and cytokines in mycobacterial infections. Eur Respir J Suppl 1995;20:668s–675s.

7. Tan TT, Coussens LM. Humoral immunity, inflammation and cancer. Curr Opin Immunol 2007;19:209–216.

8. Johansson M, Tan T, de Visser KE, Coussens LM. Immune cells as anti-cancer therapeutic targets and tools. J Cell Biochem 2007;101:918–926.

9. Rutten VP, Misdorp W, Gauthier A, et al. Immunological aspects of mammary tumors in dogs and cats: A survey including own studies and pertinent literature. Vet Immunol Immunopathol 1990;26:211–225.

10. Rivera P, von Euler H. Molecular biological aspects on canine and human mammary tumors. Vet Pathol 2011;48:132–146. Epub 2010 Dec 7.

11. Queiroga FL, Raposo T, Carvalho MI, Prada J, Pires I. Canine mammary tumours as a model to study human breast cancer: Most recent findings. In Vivo 2011;25:455–465.

12. Estrela-Lima A, Araujo MS, Costa-Neto JM, et al. Immuno-phenotypic features of tumor infiltrating lymphocytes from mammary carcinomas in female dogs associated with prognostic factors and survival rates. BMC Cancer 2010;10:256.

13. Carvalho MI, Pires I, Prada J, Queiroga FL. T-lymphocytic infiltrate in canine mammary tumours: Clinic and prognostic implications. In Vivo 2011;25:963–969.

14. Saeki K, Endo Y, Uchida K, Nishimura R, Sasaki N, Nakagawa T. Significance of tumor-infiltrating immune cells in spontane-ous canine mammary gland tumor: 140 cases. J Vet Med Sci 2012;74:227–230. Epub 2011 Sep 22.

15. Gama A, Alves A, Schmitt F. Identification of molecular phe-notypes in canine mammary carcinomas with clinical implica-tions: Application of the human classification. Virchows Arch 2008;453:123–132.

16. Sassi F, Benazzi C, Castellani G, Sarli G. Molecular-based tumour subtypes of canine mammary carcinomas assessed by immuno-histochemistry. BMC Vet Res 2010;6:5.

17. Matos I, Dufloth R, Alvarenga M, Zeferino LC, Schmitt F. p63, cytokeratin 5, and P-cadherin: Three molecular markers to dis-tinguish basal phenotype in breast carcinomas. Virchows Arch 2005;447:688–694. Epub 2005 Oct 19.

18. Kim MJ, Ro JY, Ahn SH, Kim HH, Kim SB, Gong G. Clinicopatho-logic significance of the basal-like subtype of breast cancer: A comparison with hormone receptor and Her2/neu- overexpressing phenotypes. Hum Pathol 2006;37:1217–1226.

19. Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A 2001;98:0869–10874.

20. Alexe G, Dalgin GS, Scanfeld D, et al. High expression of lymphocyte-associated genes in node-negative HER2 breast cancers correlates with lower recurrence rates. Cancer Res 2007; 67:10669–10676.

21. Baxevanis CN, Sotiriadou NN, Gritzapis AD, et al. Immunogenic HER-2/neu peptides as tumor vaccines. Cancer Immunol Immunother 2006;55:85–95.



22. Milano F, Guarriera M, Rygiel AM, Krishnadath KK. Trastuzumab mediated T-cell response against HER-2/neu overexpressing esophageal adenocarcinoma depends on intact antigen processing machinery. PLoS One 2010;5:e12424.

23. Misdorp W, Else RW, Hellmen E, Lipscomb TP. Histological classification of mammary tumors of the dog and the cat. In: International Histological Classification of Tumors of Domestic Animals. Washington DC: Armed Forces Institute of Pathology, 1999:18–25.

24. Elston CW, Ellis IO. Pathological prognostic factors in breast can-cer. I. The value of histological grade in breast cancer: Experience from a large study with long-term follow-up. Histopathology 2002;41:154–161.

25. Karayannopoulou M, Kaldrymidou E, Constantinidis TC, Dessiris A. Histological grading and prognosis in dogs with mammary carcinomas: Application of a human grading method. J Comp Pathol 2005;133:246–252.

26. Kim JH, Im KS, Kim NH, Yhee JY, Nho WG, Sur JH. Expression of HER-2 and nuclear localization of HER-3 protein in canine mammary tumors: Histopathological and immunohistochemical study. Vet J 2011;189:318–322.

27. Reis-Filho JS, Pinheiro C, Lambros MB, et al. EGFR amplification and lack of activating mutations in metaplastic breast carcino-mas. J Pathol 2006;209:445–453.

28. Kim JH, Yu CH, Yhee JY, Im KS, Kim NH, Sur JH. Canine classi-cal seminoma: A specific malignant type with human classifica-tions is highly correlated with tumor angiogenesis. BMC Cancer 2010;10:243.

29. Misdorp W. Tumors of the mammary gland. In: Meuten DJ, ed. Tumors in Domestic Animals. 4th ed. Ames, Iowa: Iowa State University Press, 2002:575–606.

30. Coronella-Wood JA, Hersh EM. Naturally occurring B-cell responses to breast cancer. Cancer Immunol Immunother 2003; 52:715–738.

31. Wong PY, Staren ED, Tereshkova N, Braun DP. Functional analy-sis of tumor-infiltrating leukocytes in breast cancer patients. J Surg Res 1998;76:95–103.

32. Chin Y, Janseens J, Vandepitte J, Vandenbrande J, Opdebeek L, Raus J. Phenotypic analysis of tumor-infiltrating lymphocytes from human breast cancer. Anticancer Res 1992;12:1463–1466.

33. van’t Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002;415:530–536.


Article


I n t r o d u c t i o nThe Arabian, or one-humped, camel (Camelus dromedarius) makes

up about 94% of the world’s camel population, at a total of 17.5 mil-lion (1). Due to its unique adaptation to drought and heat, its abil-ity to sustain thirst, and to subsist, produce, and reproduce under conditions untenable for the survival of other species of domestic mammals, the camel continues to be indispensable in some parts of the Arabian Peninsula and Africa as a source of milk and meat as well as a means of transportation.

The camel has many prompt and highly efficient complex mecha-nisms for maintaining the volume of body fluids and preventing

their loss, especially of blood. In a series of recent studies, we showed that camel platelet functions (agonist-induced aggrega-tion and PFA100 closure time) (2) and the ultrastructure of camel platelets are significantly different in many respects from that of human platelets (3,4). In the most detailed study yet on hemostasis in the camel (5), we also reported much higher levels of clotting fac-tor VIII activity (FVIII:C) in camel plasma than in humans. Whether of human or porcine origin, clotting factor VIII (FVIII) is still the most used clotting factor in replacement therapy. Further studies of this factor in the camel or in any other animal are of interest as they may lead to further understanding of the biology of this factor and to finding other sources with potential therapeutic value.

Clotting factor VIII (FVIII) and thrombin generation in camel plasma: A comparative study with humans

Abdel Galil M. Abdel Gader, Abdul Karim M. Al Momen, Abdulqader Alhaider, Marjory B. Brooks, James L. Catalfamo, Ahmed A. Al Haidary, Mansour F. Hussain

A b s t r a c tThe objective of this study was to characterize the highly elevated levels of clotting factor VIII (FVIII) in camel plasma. Whole blood was collected from healthy camels and factor VIII clotting activity (FVIII:C) assays were conducted using both the clotting and the chromogenic techniques. The anticoagulant citrate phosphate dextrose adenine (CPDA) produced the highest harvest of FVIII:C, the level of plasma factor VIII, compared to heparin:saline and heparin:CPDA anticoagulants. Camel FVIII can be concentrated 2 to 3 times in cryoprecipitate. There was a significant loss of camel FVIII when comparing levels of FVIII in camel plasma after 1 h of incubation at 37°C (533%), 40°C (364%), and 50°C (223%). Thrombin generation of camel plasma is comparable to that of human plasma. It was concluded that camel plasma contains very elevated levels of FVIII:C, approaching 8 times the levels in human plasma, and that these elevated levels could not be attributed to excessive thrombin generation. Unlike human FVIII:C, camel FVIII:C is remarkably heat stable. Taken together, these unique features of camel FVIII could be part of the physiological adaptation of hemostasis of the Arabian camel in order to survive in the hot desert environment.

R é s u m éL’objectif de la présente étude était de caractériser les niveaux très élevés du facteur de coagulation VIII (FVIII) dans le plasma de chameau. Du sang entier a été prélevé de chameaux en santé et des épreuves d’activité de coagulation du facteur VIII (FVIII:C) ont été effectuées en utilisant des techniques chromogéniques et de coagulation. L’anticoagulant citrate phosphate dextrose adénine (CPDA) a permis la récolte la plus élevée de FVIII:C, le niveau plasmatique de facteur VIII, comparativement aux anticoagulants héparine:saline et héparine CPDA. Le FVIII de chameau peut être concentré 2 à 3 fois dans des cryoprécipités. Il y avait une perte significative de FVIII de chameau lorsque l’on comparait les niveaux de FVIII dans le plasma de chameau après 1 h d’incubation à 37 °C (533 %), 40 °C (364 %), et 50 °C (223 %). La génération de thrombine dans le plasma de chameau est comparable à celle dans le plasma humain. Il a été conclu que le plasma de chameau contient des niveaux très élevés de FVIII:C, atteignant près de 8 fois le niveau dans le plasma humain, et que ces niveaux élevés ne pouvaient pas être attribué à une génération excessive de thrombine. Comparativement au FVIII:C humain, le FVIII:C de chameau est très stable à la chaleur. Prises dans leur ensemble, ces caractéristiques uniques du FVIII de chameau pourraient faire partie de l’adaptation physiologique de l’hémostase du chameau arabe afin de lui permettre de survivre dans l’environnement chaud du désert.


The Coagulation Research Laboratory, Physiology Department, College of Medicine and King Khalid University Hospital (Gader, Al Momen, Alhaider) and College of Agriculture and Food Sciences (Al Haidary, Hussain), King Saud University, Riyadh 11461, Saudi Arabia; Comparative Coagulation Section, Animal Health Diagnostic Center, Cornell University, Ithaca, New York, USA (Brooks, Catalfamo).

Address all correspondence to Professor A.M. Abdel Gader; telephone: (966) 1-467-1042; fax: (966) 1-467-2567; e-mail: [email protected]; [email protected]

Received May 1, 2011. Accepted March 7, 2012.



The present study aimed at characterizing the elevated levels of FVIII activity in camel plasma compared with human plasma. This includes selecting the most appropriate anticoagulant solution to ensure the highest recovery of FVIII, investigating the heat stability of FVIII, and determining whether the elevated activity of FVIII can be concentrated further in cryoprecipitate and accounted for by excessive thrombin generation.


CamelsSamples were obtained from the Najdi breed of Arabian one-

humped camels (Camelus dromedarius) at the College of Agriculture and Food Sciences farm, King Saud University, as well as from private camel farms in Riyadh, Saudi Arabia. These camels were disease-free, and under continuous veterinary care. Camels were provided with water ad libitum. The study was approved by the College of Medicine Research Centre Ethical Committee. The animals studied were all females aged 5 to 11 y [mean 6 standard deviation (SD): 7.4 6 2.4 y]. As on dairy farms, female camels are tradition-ally reared as a source of milk, while males are butchered for meat.

Blood collection and processingBlood was collected from a large vein in the leg directly into

250-mL blood bags (Single Whole Blood Bag; Terumu Corporation, Tokyo, Japan), which contained 35 mL of citrate phosphate dex-trose adenine (CPDA-1) anticoagulant. The blood was drawn for 5 to 7 min and collected in sample tubes containing sodium citrate (0.11 M) by cutting the blood bag tubing close to the needle and allowing blood to flow freely into the citrate tubes up to a prede-termined level, to give a blood:citrate ratio of 9:1. Blood in blood bags and sample tubes was mixed thoroughly with anticoagulant before being put in insulated containers (Electrolux, Luton, UK) for transport.

For laboratory assay purposes, platelet-poor plasma was sepa-rated by centrifuging blood samples in a refrigerated (4°C to 8°C) centrifuge (Jouan SA, Cedex, France), 3000 rpm, for 15 min. Aliquots of plasma were frozen at 280°C until assayed in batches at a later date. Assays in frozen plasma aliquots were undertaken within 1 mo of blood collection.

Laboratory procedures in RiyadhBags containing whole blood were centrifuged soon after collec-

tion in a refrigerated (4°C to 8°C) centrifuge (Jouan SA) at 4000 rpm/min for 15 min. The blood bag was then placed on a manual plasma expresser and plasma was expressed into a second satellite bag that was sealed off by a di-electric sealer. The plasma bag was stored flat at 280°C. Plasma was frozen less than 3 h after the phlebotomy.

Anticoagulant solutions usedCamel whole blood was collected in 3 blood collection bags contain-

ing 3 different anticoagulants: citrate, phosphate, dextrose, adenine (CPDA) (Terumu, Tokyo, Japan); heparin only (4 units/mL of blood — heparin sodium 25 000 IU/mL, Batch # 0482; HIKMA Pharmaceuticals, Amman, Jordon); and CPDA:heparin (equal volumes).

Heat stability of camel factor VIIIFrozen camel and human plasmas (n = 26) were thawed at 37°C

and then sampled further in 2 aliquots: 1 was incubated for 1 h in a water bath at 40°C and the other in a water bath at 50°C. Factor VIII chromogenic assay was undertaken on the 3 samples at 37°C, 40°C, and 50°C.

Laboratory assays (Riyadh)Fibrinogen was assayed as functional (clottable) fibrinogen by the

turbidimetric method of Ellis and Stransky (6).Clotting activity (FVIII:C): The 1-stage assay (7) was used with

the initial 1:10 dilution for camel plasma using human hemophilic plasma (Immuno-depleted plasma for factor VIII assay; Diagnostica Stago, Asnières sur Seine, France). The assay was carried out in an automated coagulometer (STA Compact; Diagnostica Stago). The necessary 1:10 dilution for the factor VIII assay was undertaken by the machine, which provides a printout of the final level of activity of the plasma sample.

Clotting factor VIII chromogenic activity was undertaken in an automated coagulometer (STA Compact; Diagnostica Stago) using commercial kits (Coamatic Factor VIII; Chromogenix, Lexington, Massachusetts, USA) according to the manufacturer’s instructions.

Normal human standard plasma was obtained by pooling fresh plasma from 20 healthy adults, most of whom were blood donors. Aliquots were stored at 280°C. Camel standard plasma was pooled from 10 camels, 5 of each gender.

Studies at the Animal Health Diagnostic Center, Cornell University, USA

Single units of fresh frozen plasma (FFP) (approximately 150 mL of FFP/unit) with paired 5-mL aliquots of citrate plasma were col-lected from 20 camels. An additional 10 camels (5 of each gender) were sampled to prepare 5-mL aliquots of citrate plasma alone. The samples remained frozen in transit from Riyadh and were stored at 270°C on arrival. To prepare camel cryoprecipitate, the FFP units were slowly thawed at 3°C for 24 h. The thawed FFP was then spun under refrigeration in centrifuge tubes for 10 min at 1400 3 g. Before assay, the plasma samples and cryoprecipitate pellets were thawed at 37°C for 15 min and then held on wet ice. The 5-mL aliquots from 10 individual camels were pooled after thawing for use as a same-species plasma standard.

Assay methodsFunctional (clottable) fibrinogen was measured by the Clauss

method using a 100 National Institutes of Health (NIH) U/mL human thrombin reagent and human fibrinogen standard (Fibrinogen; Diagnostica Stago, Parsippany, New Jersey, USA) in an automated, mechanical endpoint coagulation instrument (STA Compact; Diagnostica Stago).

Factor VIII clotting activity (FVIII:C) was measured in a 1-stage activated partial thromboplastin time (aPTT) assay using human factor VIII-deficient substrate plasma (George King Biomedical, Overland Park, Kansas, USA), a commercial rabbit brain cepha-lin and ellagic acid aPTT reagent (Actin; Dade Behring, Newark, Delaware, USA), and a mechanical endpoint detection instrument



(ST4; Diagnostica Stago). The assay was performed by incubat-ing 50 mL of the test sample or standard with 50 mL of deficient plasma and 50 mL of aPTT reagent for 3 min at 37°C, then triggering coagulation by adding 50 mL of 0.025 M calcium chloride (CaCl2). A standard curve was derived from serial dilutions (1:5; 1:10; 1:20) of a commercial calibrator plasma (FACT; George King Biomedical) containing 1 IU/mL (100%) human FVIII:C. In a preliminary experi-ment, the pooled camel plasma was assayed in serial dilutions (1:10 through 1:160) to obtain a clotting time comparable to the midpoint of the human standard curve. For subsequent assays, a 4-point standard curve was generated from serial dilutions of the pooled camel plasma (1:20; 1:40; 1:80; 1:160; r2 . 0.96). The test samples (camel plasma and cryoprecipitate) were assayed at a 1:160 dilu-tion. All samples and standards were assayed in duplicate. The mean clotting times of the test samples were calculated using log-log transformation, as percentage FVIII:C compared to the human and camel standards, containing 100% human or camel FVIII, respectively.

The concentration of von Willebrand factor antigen (vWF:Ag) in plasma was measured in a double-sandwich enzyme-linked immuno sorbent assay (ELISA) configured with polyclonal anti-human von Willebrand factor (vWF) antibodies, as described in a previous study (8). The vWF:Ag content of individual camel plasma and cryoprecipitate samples was reported as the percentage of the human reference plasma (FACT) and the pooled camel plasma.

Thrombin generation (TG) of the human FACT plasma and the pooled camel plasma was determined using an assay kit (Technothrombin TGA; Technoclone, Vienna, Austria) and the kit manufacturer ’s software application (Gen5 Software Protocol; Technothrombin TGA) on a microplate reader (Biotek Instruments, Winooski, Vermont, USA). In this assay, the amount of thrombin generated over time in the test sample was detected by cleavage of a fluorogenic thrombin substrate (9). In separate experiments, throm-bin generation was initiated using 1 of 3 reagents: low concentration of phospholipid and low tissue factor (2 pM human tissue factor; Technothrombin TGA RB), low concentration of phospholipid but high tissue factor (5 pM human tissue factor; Technothrombin TGA

RC-L), or high concentration of phospholipid with high tissue factor (Technothrombin TGA RC-H).

The data output included lag phase, peak thrombin, time to peak thrombin, area under curve (AUC), and qualitative thrombin genera-tion curves (thrombogram). These parameters were measured for the human FACT plasma, the human FVIII-deficient plasma, the pooled camel plasma, and 1:1 mixtures of human FVIII-deficient plasma with either human FACT plasma or pooled camel plasma.

Figure 1. Summary of factor VIII clotting activity (FVIII:C) in camel plasma (n = 29) using human plasma and camel plasma as standards. Results are presented as means and vertical bars represent standard deviation.

Figure 2. Factor VIII clotting activity (FVIII:C) in camel plasma (n = 21) collected in 3 different anticoagulants: CPDA (citrate, phosphate, dex-trose, adenine); CPDA:heparin; and saline:heparin. FVIII was assayed by the chromogenic technique. Results are presented as means and vertical bars represent standard deviation.

Figure 3. The effect of heat on levels of factor VIII clotting activity (FVIII:C) in camel and human plasma (n = 10) after 1 h of incubation at 37°C, 40°C, and 50°C. Results are presented as means and vertical bars represent standard deviation. The level of FVIII:C was measured by the chromogenic technique and the results are expressed as % of human pooled plasma, representing 100% activity.

FV

III

(%)



Statistical methodsThe Statistical Package for Social Sciences (SPSS), version 11, was

used for data analysis. The data were expressed as mean 6 SD. The 1-way analysis of variance (ANOVA) was used to compare all param-eters for different groups of data and the multiple comparisons were used to find the significance between every 2 groups. Where appro-priate, the correlation coefficient was worked out to find the relation between the different parameters for normal and abnormal groups.

Re s u l t s

Riyadh assaysFactor VIII clotting activity (FVIII:C) in camel plasma

Using the 1-stage clotting assay and human standard, the FVIII:C was found to be almost 8 times higher using the camel plasma as a standard (Figure 1).

Effect of different anticoagulant solutions on factor VIII clotting activity (FVIII:C)

Citrate phosphate dextrose adenine (CPDA) gave the highest yield of camel FVIII:C compared to CPDA:heparin and saline:heparin anticoagulants (Figure 2).

Heat stability of camel factor VIIIThere was a significant drop in FVIII activity after 1 h of incuba-

tion of camel plasma compared to activity at 37°C (mean 6 SD: 532.2 6 186.9%), 40°C (364.4 6 111.3%; a drop of 33%), and 50°C (223.3 6 66.1%; a drop of 58%) (Figure 3). It must be noted, however,

that even after 1 h of incubation at 50°C, the FVIII activity of the camel plasma remained quite high (223.3 6 66.1%). The differences in the mean values are statistically significant (P # 0.001; 1-way ANOVA).

The respective levels in human FVIII activity were 108.3 6 29.0%, 69.6 6 14.9%; (a drop of 28%), and 38.4 6 10.3% (a drop of 58%). The differences in the mean values are statistically significant (P # 0.001).

Cornell assaysPlasma fibrinogen, FVIII, and vWF:Ag

Camel plasma FVIII:C and vWF:Ag were assayed against both the pooled camel and human control plasmas, which represent 100% activity (Table I).

The fibrinogen content of camel plasma was found to be simi-lar to that of humans, with all camels having values within 150 to 400 mg/dL, which is the expected range for healthy subjects. The mean fibrinogen for the resultant pooled camel plasma was 239 mg/dL (Table I). There were differences in species, however, in results for FVIII:C and vWF:Ag analyses. Similar to the find-ings in the Riyadh assay system, clotting times in camel plasma in the 1-stage FVIII assay were much faster than in human plasma. Although the camel plasmas were assayed at an 8-fold higher dilution to “harmonize” the clotting times so that they fell on the human standard curve, the mean FVIII:C for these dilutions was approximately 300% that of the human calibrator (Table I). While the calculated values of FVIII:C derived from the same-species standard curve were lower, the relative FVIII:C for the individual camels was consistent between the human and camel assay standards (r2 = 0.99). The apparent vWF concentration of camel plasma was approximately

Table I. Summary of the assays for fibrinogen, factor VIII clotting activity (FVIII:C), and von Willebrand factor antigen (vWF:Ag) carried out at Cornell University (USA) in 10 camel plasma samples

Human standard Camel standardSample Serial Fibrinogen FVIII:C vWF:Ag FVIII:C vWF:Agnumber number mg/dL % % % % 1 126 167 81 15 13 51 2 127 169 309 12 73 57 3 128 227 484 14 131 77 4 129 216 215 13 46 68 5 130 180 289.5 12 67 95 6 131 243 549 12 154 93 7 132 306 421 18 109 138 8 133 264 452 23 120 177 9 134 216 299 17 70 10110 140 259 262 14 59 96Mean 224.7 336.2 15.0 84.2 95.3SD 45.0 140.4 6.8 43.1 36.1Camel pool 239 370 14 92 110Human control 312 93 95 4 660Note: FVIII:C was measured against both the human and camel pooled standards. The human calibrator contained 1 IU/mL (100%) of FVIII:C. Camel plasmas were assayed at an 8-fold higher dilution than the human calibrator and results are presented for this dilution with no correction factor.SD — standard deviation.



15% to 20% of human plasma, with little discriminatory value among individuals using the human standard (Table I). The affinity and specificity of the anti-vWF antibodies used in this assay to detect camel vWF are unknown. It is possible that variable antibody reactiv-ity, rather than relative vWF protein deficiency, caused the observed low concentration of vWF:Ag in camel plasma.

Camel cryoprecipitateCamel FFP was thawed for 32 h, which is longer than the usual

24 h, in order to enhance the harvest of cryoprecipitate. The cryopre-cipitate was recovered from 90 mL of starting plasma after 10 min of centrifugation at 1400 3 g. The results of the assays undertaken in 10 camels (3 females and 7 males) are shown in Table II.

The mean volume of recovered cryoprecipitate per 90 mL of camel FFP was 2.6 mL (SD: 0.65 mL; 2 to 3.5 mL range). In the experience of 2 of the authors (MB and JC), this volume was approximately half that of the expected volume obtained from canine plasma using this slow-thaw technique. Nevertheless, the harvested cryoprecipitate produced an approximately 2.5-fold mean increase in hemostatic proteins [fibrinogen, FVIII:C, and vWF:Ag content compared with the starting camel plasma (Table II)]. Similarly, combined results from all 10 camels (Table II) showed that fibrinogen had been con-centrated in the cryoprecipitate fraction to more than twice (approxi-mately 2.5-fold) its concentration in plasma (P , 0.001).

Thrombin generation (TG)Pronounced inter-species variability was also found in TG experi-

ments, with camel plasma consistently demonstrating lower peak thrombin and overall endogenous thrombin potential (AUC) than human plasma (Table III). The reagent composition influenced thrombin generation in the camel plasma more than in human plasma. In human plasma, an increase in tissue factor content from 2 pM to 5 pM with high phospholipid produced a less than 10% increment in AUC. In contrast, increasing the tissue factor and par-ticularly the phospholipid content in the trigger reagent resulted

in a pronounced, almost 6-fold, increase in camel AUC (Table III). While camel factor VIII supported thrombin generation when combined with human factor VIII-deficient plasma, thrombogram values approached that of the human plasma factor VIII-deficient mixture only in the high tissue factor and high phospholipid reagent (Reagent RC-H).

D i s c u s s i o nAn elevated level of factor VIII clotting activity (FVIII:C) in camel

plasma (. 3 to 5 times) compared to that of humans was a consistent finding in this study. The activity is best demonstrated when whole blood is collected in citrate anticoagulants rather than in heparin and can be further concentrated (2 to 3 times) in cryoprecipitate. We also found that camel factor VIII could support thrombin generation in human FVIII-deficient plasma.

The high FVIII:C in camel plasma was demonstrable in both 1-stage and 2-stage assays in spite of the marked differences in the configuration and endpoints of these assays. In the 1-stage assay, coagulation is initiated through contact activation and generation of factor XIIa with sequential activation of the intrinsic pathway factors. Thrombin is generated through the subsequent assembly of tenase and prothrombinase complexes, and then acts on fibrinogen to form the fibrin clot endpoint. The human substrate-deficient plasma in the assay mixture contains sufficient fibrinogen and coagulation factors to form a fibrin clot, providing that the test plasma can sup-ply FVIII:C to allow assembly of a functional tenase complex. In the absence of same-species factor VIII-deficient plasma, measurement of animal FVIII:C in 1-stage assays requires a heterologous system in which species differences in contact activation, protease activity, and endogenous inhibitors may affect clotting time independently of the FVIII:C in the assay mixture.

On the other hand, 2-stage assays do not require substrate- deficient plasmas and their configuration reduces the reagent/

Table II. Summary of results of assays of camel plasma fibrinogen, factor VIII clotting activity (FVIII:C), and von Willebrand factor antigen (vWF:Ag) in both camel plasma and cryoprecipitate (cryoppt) undertaken in 10 camels

Plasma Cryoppt Plasma Cryoppt Plasma CryopptSerial fibrinogen fibrinogen FVIII:C FVIII:C vWF:Ag vWF:Agnumber (mg/dL) (mg/dL) (%) (%) (%) (%)135 208 317 103 296 108 205142 191 409 66 315 76 101143 310 562 158 440 88 147144 266 424 83 341 84 245145 322 677 64.5 387 89 160150 268 1124 114 297 105 541155 214 316 133 190 106 152156 279 568 199 225 84 200158 541 1424 75 175 80 236160 326 750 245 267 162 262

Mean 292.5 657.1 124.05 293.3 98.2 224.9Median 273.5 565 108.5 296.5 88.5 202.5SD 99.5 362.1 60.6 83.7 25.1 121.9SD — standard deviation.



endpoint variability inherent in clotting time tests. In the first stage of the assay, an intrinsic tenase complex is assembled and in the second stage, the amount of factor Xa formed is determined by the cleavage of a factor Xa substrate. The assay used in this study con-tains a bovine factor reagent containing factor IX, factor X, thrombin, and phospholipid and the test sample provides FVIII:C. The results obtained in this assay are comparable across species, since all the components of the tenase complex, exclusive of factor VIII, are present in the assay mixture and the endpoint is cleavage of a small peptide substrate.

Our finding that camel FVIII consistently demonstrated relatively high cofactor activity compared to human FVIII in both 1-stage and 2-stage assays indicates an inherent difference in FVIII:C between species that cannot be attributed solely to assay conditions.

Previous studies have found that the use of the primary antico-agulant heparin for collecting human blood doubles FVIII recovery in FFP and cryoprecipitate compared to conventional ACD or CPD anticoagulants (10,11). In these 2 studies (10,11), the presence of heparin in plasma was controlled for by either neutralizing heparin with protamine or adding equivalent amounts of heparin to the plasma standard, as the researchers used the 1-stage FVIII:C clot-ting assay, which is not suited for heparinized plasma. In this study, we used the chromogenic method according to the manufacturer’s instructions. It was shown that the chromogenic method was more

suitable for measuring elevated levels of FVIII, demonstrating the best inter-assay variation compared to the 1-stage clotting and antigenic assays, in addition to being unaffected by the presence of heparin (12). All assays were also run simultaneously on the same day on all samples collected in different anticoagulant solutions. It is not clear why blood collected in CPDA-1 anticoagulant gave the highest yield of camel FVIII (Figure 2). We can only speculate on the basis of our early finding that camel platelets were remarkably activated when camel blood was collected in heparin (2). It is pos-sible that platelet activation would be accompanied by secondary activation of coagulation and some consumption of FVIII. Similar platelet activation could not be seen when human whole blood was collected in citrate anticoagulant (2).

As for the heat stability of camel FVIII (Figure 3), we noted with much interest the significant drop in levels of camel FVIII after 1 h of incubation at 37°C, 40°C, and 50°C. It must also be noted, however, that even after 1 h of incubation at 50°C, the levels of camel FVIII remained quite high (220.4%). Even after prolonged (1 h) incubation at 50°C, camel FVIII contained approximately twice the FVIII:C of our pooled human plasma standard when measured in the chromo-genic FVIII:C assay (Figure 3). This heat stability was not absolute, as raising the temperature incrementally from 40°C to 50°C resulted in a progressive loss in activity. Nevertheless, further investigation is warranted to define the property of camel FVIII that allows it to

Table III. Summary of the lag time, peak thrombin, time to peak, and AUC (area under curve) for thrombin generation (TG) test on both camel and human plasma

TG parameters Peak Time to TG Lag time thrombin peakSample reagent (min) (nm) (min) AUCHuman plasmaa RBb 13.6 208.3 21.6 3890Camel plasma RBb 20.6 11.9 26.1 184FVIII-deficient plasma RBb . 60 0 NA NAFVIII-deficient human plasma RBb 24.6 70.3 36.6 2416FVIII-deficient camel plasma RBb 24.1 13.9 34.1 415 Human plasma RC-Lc 14.6 233.3 22.1 4155Camel plasma RC-Lc 21.1 16.7 27.1 314FVIII-deficient plasma RC-Lc . 60 0 NA NAFVIII-deficient human plasma RC-Lc 20.1 97.8 32.1 3035FVIII-deficient camel plasma RC-Lc 31.1 9.2 40.6 218 Human plasma RC-He 11.1 377.4 16.6 4251Camel plasma RC-Hd 11.6 82.7 18.1 1095FVIII-deficient plasma RC-Hf . 60 0 NA NAFVIII-deficient human plasma RC-He 16.1 225.6 25.1 3587FVIII-deficient camel plasma RC-He 12.6 169.8 19.1 2670a FACT human reference plasma (George King Biomedical).b RB = Triggering solution with low phospholipid and low tissue factor.c RC-L = Triggering solution with low phospholipid.d RC-L = Triggering solution with low phospholipid and high tissue factor.e RC-H = Triggering solution with high phospholipid.f RC-H = Triggering solution with high phospholipid and high tissue factor.NA = Not available.



participate in the assembly of a highly active tenase complex after remaining at a temperature well above the physiologic body temper-ature of human beings. Camels live and survive in deserts where the ambient temperatures during the summer could be well above 50°C. Camels maintain their body temperature at 37°C when hydrated and this increases by 6°C to 7°C when they become dehydrated (13,14). In a previous study, we found that camel platelets showed minimal changes in ultrastructure when heated to 50°C, while at the same temperature, human platelets were destroyed, the cytoplasm was dissolved, and most granules were lost (4).

To explain the very elevated levels of FVIII:C in camel plasma, we probed the possibility that excessive thrombin generation (TG) could be a factor. While currently available thrombin generation assays are sensitive to FVIII activity, many other hemostatic proteins and their inhibitors influence TG. In the 3 TG assay configurations we fol-lowed, camel plasma generated lower amounts of thrombin than the human control plasma. The reaction conditions optimized for human plasma, however, may not support maximal TG in other species. The endogenous thrombin potential of human plasma (summarized as AUC) (Table III) increased only slightly in reaction mixtures contain-ing increasing amounts of tissue factor and phospholipid. In contrast, a high content of tissue factor and phospholipid reagent produced an approximately 10-fold increase in AUC in camel plasma compared to the low tissue factor/phospholipid reagent. High phospholipids in the reaction mixture particularly enhanced the rate of TG in camel plasma as evidenced by a shortening of the lag time and time to peak thrombin. The enhancement of TG we observed in a mixture of camel plasma and human factor VIII-deficient plasma may reflect the role of human factor VIIa in activating tissue factor, in addition to phos-pholipid supplied by residual human cell-derived microparticles and perhaps dilution of endogenous coagulation inhibitors in camel plasma. The TG assay system must be further modified in order to investigate these species differences.

Further analyses of camel FVIII cofactor activity are warranted to investigate these species differences. Regions within the C2 domain of human FVIII are involved in FVIII binding to phosphatidylserine (15), and therefore sequence analyses of the homologous region in the camel FVIII gene may complement functional analyses to further clarify how phospholipid interaction with camel FVIII influences its cofactor activity. Ultimately, identifying the properties of camel FVIII that produce a “superior” cofactor activity will provide a greater understanding of mechanisms underlying assembly and/or function of the intrinsic tenase complex.

The results of the present study raise the question of whether camel FVIII could have therapeutic use in a manner similar to por-cine FVIII that is widely used to manage bleeding in patients with hemophilia who develop inhibitors to human FVIII. Could camel FVIII be a potential alternative treatment for hemophiliacs? Certainly, camel clotting factor VIII (FVIII:C) is of interest in the context of hemophilia therapy due to its remarkably elevated clotting levels, its heat stability, and the recently emerging knowledge of the homol-ogy between camel antibodies and human heavy-chain variable domains (VHHs) and their low immunogenic potential (16,17). It is tempting to suggest an approach similar to that used in developing the currently available recombinant human and porcine factor VIII (18). This development is the result of early efforts in the prepara-

tion of FVIII concentrate from human and porcine plasma to the deployment of gene technology to produce recombinant FVIII. Basic preliminary work needs to be done, however, in order to characterize the molecular biology of camel FVIII before any steps are taken to develop a therapeutic product made of camel FVIII.

It was concluded that the remarkably elevated levels of FVIII in camel plasma found in this study support the idea of a more active state of the coagulation system in the camel; elevated fac-tor VIII levels are a known index of hypercoagulability (19) and present a risk of arterial and venous thrombosis (20). In addition to the heat stability of FVIII, this may be yet another physiological adaptation that protects the camel from excessive loss of blood or fluids in hot, arid desert conditions. Given the unique immune sys-tem of the camel, further studies may be carried out to investigate the possibility of using camel FVIII as a starting material for treating hemophilia.

A c k n o w l e d g m e n t sThe authors thank Mohamed Hamid and Luman Gasmelsid for

technical assistance, Abdul Rahman Garelnabi and Ali Al-Sheehi for camel blood collection, and Farah Chatila for secretarial help. This work is supported by a grant from the Deanship of Research (National Programs; No. AR-18), College of Medicine Research Centre King Saud University in Riyadh, Saudi Arabia.

Re f e r e n c e s1. Youssef OK, Babiker SA. The desert camel as meat animal. Meat

Sci 1989;26:245–254.2. Gader AMA, Ghumlas AK, Hussain MF, Al-Haidary AI. Platelet

aggregation and platelet function analyser 100 (PFA-100) closure time in camels — A comparative study with humans. Comp Clin Pathol 2006;15:31–37.

3. Gader AMA, Ghumlas AK, Hussain MF, Al-Haidary A, White JG. Ultrastructure of camel blood platelets: A comparative study with human, bovine, and equine cells. Platelets 2008;19:51–58.

4. Al-Ghumlas AK, Gader AMA, Hussain MF, Al-Haidary A, White JG. Effects of heat on camel platelet structure and func-tion — A comparative study with humans. Platelets 2008;19: 163–171.

5. Hussein MF, Al-Momen AK, Gader AMA. Haemostatic parame-ters in the camel (Camelus dromedarius) comparison with humans. Comp Haematol Int 1992;2:92–96.

6. Ellis BC, Stransky AA. A quick and accurate method for the determination of fibrinogen in plasma. J Lab Clin Med 1961;58: 477–488.

7. Sirridge MA, Shannon R. Laboratory Evaluation of Hemostasis and Thrombosis. 3rd ed. Philadelphia, Pennsylvania: Lea & Febiger, 1983:58–160.

8. Benson RE, Catalfamo JL, Dodds WJ. A multispecies enzyme-linked immunosorbent assay for von Willebrand’s factor. J Lab Clin Med 1992;119:420–427.

9. Hemker HC, Al Dieri R, DeSmedt E, Béguin S. Thrombin gen-eration, a function test of the haemostatic-thrombotic system. Thromb Haemost 2006;96:553–561.



10. Rock GA, Cruickshank WH, Tackaberry ES, Palmer DS. Improved yields of factor VIII from heparinized plasma. Vox Sang 1979;36:294–300.

11. Palmer DS, Rosborough D, Perkins H, Bolton T, Rock G, Ganz PR. Characterization of factors affecting the stability of frozen heparinized plasma. Vox Sang 1993;65:258–270.

12. Chandler WL, Ferrel C, Lee J, Tun T, Kha H. Comparison of three methods for measuring factor VIII levels in plasma. Am J Clin Pathol 2003;120:34–39.

13. Al-Faraj A, Al-Haidary A. Measurement and stimulation of camel core body temperature response to ambient temperature. Int J Agri Biol 2006;4:531–534.

14. Al-Haidary A. Effect of dehydration on core body temperature of young Arabian camels (Camelus dromedarius). J King Saud Univ Agri Sci 2005;18:1–7.

15. Foster PA, Fulcher CA, Houghten RA, Zimmerman TS. Synthetic factor VIII peptides with amino acid sequences contained within

the C2 domain of factor VIII inhibit factor VIII binding to phos-phatidylserine. Blood 1990;75:1999–2004.

16. Harmsen MM, De Haard HJ. Properties, production, and appli-cations of camelid single-domain antibody fragments. Appl Microbiol Biotechnol 2007;77:13–22.

17. Muyldermans S, Baral TN, Retamozzo VC, et al. Camelid immunoglobulins and nanobody technology. Vet Immunol Immunopathol 2009;128:178–183.

18. Gangadharan B, Parker ET, Ide LM, Spencer HT, Doering CB. High-level expression of porcine factor VIII from genetically modified bone marrow-derived stem cells. Blood 2006;107: 3859–3864.

19. Bobrow RS. Excess factor VIII: A common cause of hypercoagu-lability. J Am Board Fam Pract 2005;18:147–149.

20. Kamphuisen PW, Eikenboom CJ, Bertina RM. Elevated fac-tor VIII levels and the risk of thrombosis arteriosclerosis. Thromb Vasc Biol 2001;21:731–738.


Short Communication Communication brève


Atypical enteropathogenic Escherichia coli (EPEC) are a cause of diarrhea in ruminants and humans, although these bacteria can also be isolated from healthy ruminants and humans (1,2). These bacteria cause a characteristic attaching and effacing (AE) lesion in the gut mucosa because of the intimate bacterial adhesion to the enterocytes and effacement of the brush border microvilli (2). Formation of AE lesions is governed by the locus of enterocyte effacement (LEE). This locus contains the eae gene, which encodes an outer membrane protein called intimin, necessary for intimate attachment to epithelial cells (2). The LEE also encodes a type III secretion system; EPEC-secreted proteins (Esp), such as EspA, which is essential for the delivery of proteins into the host cell; and LEE-encoded regulator (Ler), which positively regulates virulence factors (3,4). Additional non-LEE-encoded virulence factors have been described, such as Efa1/LifA, which is encoded by the pathogenicity island PAI OI-122, and enterohemolysin, which is encoded by large plasmids, such as pO157 (5,6).

Molecular characterization of atypical EPEC strains isolated from humans and ruminants has revealed a tremendous diversity of virulence genes among different strains (1,6,7). Therefore, it is pos-sible that different atypical EPEC strains have different pathogenic potential and that only some of them can cause disease in humans and animals. It seems likely that only the atypical EPEC strains possessing certain virulence genes, such as efa1/lifA and ehxA, can cause disease (1,5,6). However, it is also possible that differences in pathogenicity among strains may be due, at least partially, to different expression patterns of some virulence genes. In support of this hypothesis, a previous study has found that expression of virulence genes differs among enterohemorrhagic E. coli O157:H7 strains depending on whether their genotype is predominant among human clinical cases or among bovines (8). Despite this evidence suggesting a link between expression profiles of virulence genes and pathogenicity, the authors are unaware of studies that compare

Differences in virulence gene expression between atypical enteropathogenic Escherichia coli strains isolated from diarrheic

and healthy ruminantsPilar Horcajo, Gustavo Domínguez-Bernal, Javier Carrión, Ricardo De La Fuente,

José A. Ruiz-Santa-Quiteria, José A. Orden

A b s t r a c tDifferences in the pathogenicity of atypical enteropathogenic Escherichia coli (EPEC) strains may be due, at least partially, to different expression patterns of some virulence genes. To investigate this hypothesis, the virulence gene expression patterns of 6 atypical EPEC strains isolated from healthy and diarrheic ruminants were compared using quantitative real-time reverse transcription polymerase chain reaction after growing the bacteria in culture medium alone or after binding it to HeLa epithelial cells. Some virulence genes in strains from diarrheic animals were upregulated relative to their expression in strains from healthy animals. When bacteria were cultured in the presence of HeLa cells, the ehxA and efa1/lifA genes, previously associated with the production of diarrhea, were expressed at higher levels in strains from diarrheic animals than in strains from healthy animals. Thus, the expression levels of some virulence genes may help determine which atypical EPEC strains cause diarrhea in ruminants.

R é s u m éDes différences dans la pathogénicité de souches atypiques d’Escherichia coli entéropathogènes (EPEC) peuvent être dues, au moins en partie, à différents patrons d’excrétion de quelques-uns des gènes de virulence. Pour étudier cette hypothèse, les patrons d’expression des gènes de virulence de six souches atypiques d’EPEC isolées de ruminants en santé et avec diarrhée ont été comparés par une épreuve quantitative en temps réel de réaction d’amplification en chaîne par la polymérase par transcription réverse après avoir fait croître la bactérie dans un milieu de culture seul ou après l’avoir liée à des cellules épithéliales HeLa. Quelques gènes de virulence dans des souches provenant d’animaux avec diarrhée étaient régulés à la hausse relativement à leur expression dans les souches provenant d’animaux en santé. Lorsque les bactéries étaient cultivées en présence de cellules HeLa, les gènes ehxA et efa1/lifA, précédemment associés avec la production de diarrhée, étaient exprimés à des niveaux plus élevés dans les souches provenant d’animaux diarrhéiques que dans les souches provenant d’animaux en santé. Ainsi les niveaux d’expression de certains gènes de virulence pourraient aider à déterminer quelles souches atypiques d’EPEC causent de la diarrhée chez les ruminants.


Departamento de Sanidad Animal, Facultad de Veterinaria, Universidad Complutense, 28040 Madrid, Spain.

Address all correspondence to Dr. José A. Orden; telephone: 34 91 394 3704; fax: 34 91 394 3795; e-mail: [email protected]

Received May 16, 2012. Accepted June 7, 2012.



levels of virulence gene expression among atypical EPEC strains isolated from healthy and diarrheic ruminants.

The aim of this study was to establish whether differences in viru-lence gene expression can be detected between atypical EPEC strains that share the same serotype and virulence gene profile, but differ in whether they were isolated from healthy or diarrheic ruminants.

Six strains were tested in this study in pairs, with one strain com-ing from a diarrheic ruminant and the paired strain coming from a normal ruminant. The strain pairs were as follows: O153:HNM strains (NM indicates non-motile) isolated from a diarrheic lamb and a healthy sheep, O26:HNM strains isolated from a diarrheic calf and a healthy cow, and O26:H11 strains isolated from a diarrheic lamb and a healthy goat. The genes analyzed were eae, espA, ler, efa1/lifA, and ehxA, which encode enterohemolysin. The strains used in this study have previously been characterized (1) and possess all 5 genes analyzed, except for the O26:HNM strains, which lack ehxA.

Virulence gene expression in enteropathogenic bacteria is regu-lated by environmental conditions (9), so expression was analyzed in the present study under 2 sets of conditions: after culturing the bacteria in Dulbecco’s modified Eagle medium (DMEM; Lonza, Verviers, Belgium) and after co-culturing them in the same medium together with HeLa cells. Dulbecco’s modified Eagle medium was chosen because it provides a controlled environment with minimal variation in growth conditions, and it has also been shown to induce

expression of LEE genes (8,10). The HeLa epithelial cells were used because growth in the presence of epithelial cells may better simu-late the gastrointestinal tract environment than growth in DMEM (8). In the first set of conditions, strains were grown to exponential phase in DMEM, the bacteria were sedimented by centrifugation (5000 3 g for 5 min) and the pellet was stored in RNA stabilization solution (RNA later; Invitrogen, Prat de Llobregat, Barcelona, Spain) for subsequent RNA purification. In parallel, under the second set of conditions, HeLa cells were infected with “preactivated” bacteria, which had been grown to early- or mid-log phase at 37°C in DMEM (9), at a multiplicity of infection of 100:1. After 5 h of incubation, the HeLa cells were washed vigorously to remove nonadhering bacteria, and cells with adhered bacteria were collected in RNA stabilization solution until RNA isolation. Culturing of the atypical EPEC strains alone in DMEM or in the presence of HeLa cells was carried out at 37°C in an atmosphere of 5% CO2.

Total bacterial RNA was extracted using a commercial RNA isolation kit (High Pure RNA Isolation kit; Roche Applied Science, Sant Cugat del Vallés, Barcelona, Spain) and cDNA was synthe-sized using a commercial kit (qScript cDNA SuperMix kit; Quanta Biosciences, Gaithersburg, Maryland, USA) according to the manu-facturer’s instructions. Quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR) was done using a RT-PCR thermocycler (7300 Real-Time PCR System; Applied Biosystems, Tres Cantos, Madrid, Spain) and a commercial dye (FastStart Universal SYBR Green Master Mix; Roche Applied Science). Samples were tested in duplicate. Previously published primers were used to amplify the eae, ler, and ehxA genes (11–13). In addition, prim-ers were designed in the present study to amplify the espA gene (espAq-F: 59-GATGCCTCATTCATATCAGC-39 and espAq-R: 59-TCGGTGTTTTTCAGGCT GC-39) and efa1/lifA gene (efaF-RT: 59-TGGTAGTCAGGTATACATCCGTATTTC-39 and efaR-RT: 59-GCTGAAAACCGGCACAAT-39). Data were analyzed using Gene Expression’s CT Difference method and individual PCR efficiencies were determined using LinRegPCR, as previously described (14). The 16S rRNA housekeeping gene was used as the internal control (15). Genes whose expression in strains from diarrheic animals relative to their expression in strains from healthy animals that changed 0.5- to 1.5-fold, were classified as unchanged.

When cultured in DMEM, there was no difference in gene expression except for eae, espA, and ler in the pair of O26:H11 strains (Figure 1A). These 3 genes were expressed at levels 3- to 6-fold higher in bacteria isolated from a diarrheic lamb than from a healthy goat.

When virulence gene expression was compared between strains from diarrheic and healthy hosts after being cultured in contact with HeLa cells, the eae and espA genes were expressed at similar levels between strains, whereas ler, ehxA, and efa1/lifA were expressed at levels 2- to 7-fold higher in O153:HNM and O26:H11 strains isolated from diarrheic animals (Figure 1B). The presence of the ehxA and efa1/lifA genes has been previously associated with diarrhea caused by atypical EPEC in children and neonatal ruminants (1,5,6). In addition, the results of this study suggest that overexpression of these genes may also determine whether atypical EPEC strains can cause diarrhea in neonatal ruminants. Our results on the expression of the eae and espA genes in the pair of O26:H11 strains grown in contact with epithelial cells are not in agreement with those results

Figure 1. Expression of virulence genes in strains of atypical enteropatho-genic Escherichia coli isolated from diarrheic ruminants relative to their expression in strains isolated from healthy ruminants. A — Strains grown in Dulbecco’s modified Eagle medium. B — Strains grown in contact with HeLa cells. NM — nonmotile.

Virulence gene

Virulence gene



obtained using the same strains grown in DMEM or with the results of Vanaja et al (8), who found an association between overexpres-sion of LEE genes in E. coli O157:H7 strains grown in DMEM and virulence. It is possible that these differences may be due to the influence of environmental conditions on bacterial gene expression. Thus explaining how Jandu et al (16) were able to find differences in gene expression when E. coli O157:H7 strains were cultured in the presence and absence of epithelial cells.

In summary, the results of this study show that some virulence genes are upregulated in strains isolated from diarrheic animals relative to their expression in strains isolated from healthy animals. Thus, the expression levels of some virulence genes may help determine which atypical EPEC strains cause diarrhea in ruminants. Since this is a preliminary study involving a small number of strains, further studies are required to confirm the role of virulence gene expression in the production of diarrhea by atypical EPEC strains in neonatal ruminants.

A c k n o w l e d g m e n tThis study was supported by a grant from Banco Santander-

Universidad Complutense (INBAVET 920338).

Re f e r e n c e s 1. Horcajo P, Domínguez-Bernal G, de la Fuente R, et al. Compari-

son of ruminant and human attaching and effacing Escherichia coli (AEEC) strains. Vet Microbiol 2012;155:341–348.

2. Nataro JP, Kaper JB. Diarrheagenic Escherichia coli. Clin Microbiol Rev 1998;11:142–201.

3. Elliott SJ, Sperandio V, Girón JA, et al. The locus of enterocyte effacement (LEE)-encoded regulator controls expression of both LEE- and non-LEE-encoded virulence factors in enteropatho-genic and enterohemorrhagic Escherichia coli. Infect Immun 2000;68:6115–6126.

4. Garmendia J, Frankel G, Crepin VF. Enteropathogenic and enterohemorrhagic E. coli infections: Translocation, translocation, translocation. Infect Immun 2005;73:2573–2585.

5. Afset JE, Bruant G, Brousseau R, et al. Identification of virulence genes linked with diarrhea due to atypical enteropathogenic Escherichia coli by DNA microarray analysis and PCR. J Clin Microbiol 2006;44:3703–3711.

6. Scaletsky ICA, Aranda KRS, Souza TB, Silva NP, Morais MB. Evidence of pathogenic subgroups among atypical enteropatho-genic Escherichia coli strains. J Clin Microbiol 2009;47:3756–3759.

7. Beutin L, Kaulfuss S, Herold S, Oswald E, Schmidt H. Genetic analysis of enteropathogenic and enterohemorrhagic Escherichia coli serogroup O103 strains by molecular typing of virulence and housekeeping genes and pulsed-field gel electrophoresis. J Clin Microbiol 2005;43:1552–1563.

8. Vanaja SK, Springman AC, Besser TE, Whittam TS, Manning SD. Differential expression of virulence and stress fitness genes between Escherichia coli O157:H7 strains with clinical or bovine-biased genotypes. Appl Environ Microbiol 2010;76:60–68.

9. Rosenshine I, Ruschkowski S, Finlay BB. Expression of attaching/ effacing activity by enteropathogenic Escherichia coli depends on growth phase, temperature, and protein synthe-sis upon contact with epithelial cells. Infect Immun 1996;64: 966–973.

10. Abe H, Tatsuno I, Tobe T, Okutani A, Sasakawa C. Bicarbonate ion stimulates the expression of locus of enterocyte effacement-encoded genes in enterohemorrhagic Escherichia coli O157:H7. Infect Immun 2002;70:3500–3509.

11. Chassagne L, Pradel N, Robin F, et al. Detection of stx1, stx2, and eae genes of enterohemorrhagic Escherichia coli using SYBR Green in a real-time polymerase chain reaction. Diagn Microbiol Infect Dis 2009;64:98–101.

12. Poirier K, Fauche SP, Béland M, et al. Escherichia coli O157:H7 survives within human macrophages: Global gene expression profile and involvement of the Shiga toxins. Infect Immun 2008;76:4814–4822.

13. Rashid RA, Tabata TA, Oatley MJ, et al. Expression of putative virulence factors of Escherichia coli O157:H7 differs in bovine and human infections. Infect Immun 2006;74:4142–4148.

14. Schefe JH, Lehmann KE, Buschmann IR, Unger T, Funke-Kaiser H. Quantitative real-time RT-PCR data analysis: Current concepts and the novel “gene expression’s CT differ-ence” formula. J Mol Med 2006;84:901–910.

15. Spano G, Beneduce L, Terzi V, Stanca AM, Massa S. Real-time PCR for the detection of Escherichia coli O157:H7 in dairy and cattle wastewater. Lett Appl Microbiol 2005;40:164–171.

16. Jandu N, Ho NKL, Donato KA, et al. Enterohemorrhagic Escherichia coli O157:H7 gene expression profiling in response to growth in the presence of host epithelia. PLoS One 2009;4:e4889.


Produits et services Supplies and services

RENOUVELEZ VOTRE ABONNEMENT

DÈS MAINTENANT!

RENEW YOUR SUBSCRIPTION

NOW!

Four issues of the Canadian Journal of Veterinary Research for 2013(online)

Quatre numéros de la Revue canadienne de recherche vétérinaire pour 2013

$150 $ ( applicable GST/HST; Canadian/Canadien TPS/TVH en vigueur)

$165 $ US/É.-U. Foreign/Étranger $50 $ ( applicable GST/HST) Back issues/Anciens numéros each/chacun

Contact/Communiquez avec :

Beverley Kelly Editorial Coordinator/Coordonnatrice de la rédactionCanadian Journal of Veterinary ResearchRevue canadienne de recherche vétérinaire339 rue Booth Street Ottawa, Ontario K1R 7K1Canada

Tel./Tél. : (613) 236-1162, x117Fax/Télécopieur : (613) 236-9681E-mail/Courriel : [email protected]

w w w . c a n a d i a n v e t e r i n a r i a n s . n e t

w w w . v e t e r i n a i r e s a u c a n a d a . n e t

Tables of contentsComing events

Article abstractsClassified advertising

Subscription informationInstructions for authors

Tables des matières Événements à venirRésumés d’articles Annonces classées Information sur l’abonnement Directives à l’intention des auteurs

Canadian Journal of Veterinary Research

available online now!

Articles are now easily accessible over the Internet at

PubMed Centralwww.pubmedcentral.com

Reaching Canada’s VeterinariansGet your message into

The Canadian Veterinary JournalFor more information contact:Laima LaffitteAdvertising ManagerTel.: (613) 673-2659Fax: (613) 673-2462e-mail: [email protected]


The effect of nitrous oxide on the minimum alveolar concentration (MAC) and MAC derivatives of isoflurane in dogsDebra A. Voulgaris, Christine M. Egger, M. Reza Seddighi, Barton W. Rohrbach, Lydia C. Love, Thomas J. Doherty . . . . . . . . . . 131

Oxidative stress, superoxide production, and apoptosis of neutrophils in dogs with chronic kidney diseaseAdriana Carolina Rodrigues Almeida Silva, Breno Fernando Martins de Almeida, Carolina Soares Soeiro, Wagner Luis Ferreira, Valéria Marçal Félix de Lima, Paulo César Ciarlini . . . . . . . . . . . . . . . . . . . . .136

Correlation of tumor-infiltrating lymphocytes to histopathological features and molecular phenotypes in canine mammary carcinoma: A morphologic and immunohistochemical morphometric studyJong-Hyuk Kim, Seung-Ki Chon, Keum-Soon Im, Na-Hyun Kim, Jung-Hyang Sur . . . . . . . . . . . . . . . . . . . . . . . 142

Clotting factor VIII (FVIII) and thrombin generation in camel plasma: A comparative study with humansAbdel Galil M. Abdel Gader, Abdul Karim M. Al Momen, Abdulqader Alhaider, Marjory B. Brooks, James L. Catalfamo, Ahmed A. Al Haidary, Mansour F. Hussain . . . . . . . . . . . . . . . . . . . . 150

Short Communication/ Communication brèveDifferences in virulence gene expression between atypical enteropathogenic Escherichia coli strains isolated from diarrheic and healthy ruminantsPilar Horcajo, Gustavo Domínguez-Bernal, Javier Carrión, Ricardo De La Fuente, José A. Ruiz-Santa-Quiteria, José A. Orden. . . . . . . . . . . . . . . . . . . . . . . . . .158