intramammary immunization with live-attenuated staphylococcus aureus: microbiological and...

EISEVIER FEMS Immunology and Medical Microbiology 14 (1996) 45-51

rpDMMUtf;OGY AND

MICROBIOLOGY

Intramammary immunization with live-attenuated Staphylococcus aureus: microbiological and immunological

studies in a mouse mastitis model

Ver6nica Garcia a, Marisa G6mez a, Mercedes Iglesias a, Norberto Sanjuan b, Magdalena Gherardi ‘, M. Cristina Cerquetti ‘, Daniel Sordelli a3*

a Laboratorio de Bacteriologia, Facultad de Medicina, Universidad de Buenos Aires, Paraguay 2155 P-12, 1121 Buenos Aires, Argentina b Laboratorio de Patologia Experimental, Departamento de Microbiologia, Facultad de Medicina, Uniuersidad de Buenos Aires,

Buenos Aires, Argentina ’ Centro de Estudios Fannacol6gicos y Botcinicos, CONICET, Buenos Aires, Argentina

Received 5 December 1995; revised 22 February 1996; accepted 22 February 1996

Abstract

Mammary infection was induced in lactating mice by intramammary injection of StuphyZococcus aureus. Histopathologi- cal analysis revealed in-filtration and lesions of varying magnitude that were still apparent 21 days after the challenge. Concomitantly, viable S. uureus was recovered from infected mammary glands. Mice were immunized by the intramammary route with 5 X lo6 colony forming units of a temperature-sensitive mutant of S. aureus and subsequently received a boosting injection seven days later. On day 14 mice were challenged by the intramammary route with the wild-type strain. Intramammary immunization induced a significant increase in milk IgA (P < 0.05), serum IgG (P < 0.05) and serum IgA (P < 0.05) on the day of the challenge, when compared with non-immunized mice. Immunization decreased significantly (P < 0.01) the number of S. aureus colony forming units recovered 96 h after intramammary challenge. In conclusion, the feasibility of immunizing locally with temperature-sensitive S. aureus to induce immunity in the mouse mammary gland was demonstrated. The mouse model of mastitis is proposed as a useful system for screening temperature-sensitive S. aureus

strains to be utilized in Ihe development of a vaccine.

Keywords: Mastitis; Live-attenuated; Staphylococcus aureus; Vaccine

1. Introduction

Staphylococcus aureus is a natural inhabitant of mammalian skin and mucous epithelia [l]. Although S. aureus is widely recognized as a human pathogen it also causes disease in animals. In domestic rumi-

* Corresponding author. Tel: + 54 (1) 961 2021 ext. 219; Fax:

+ 54 (1) 962 5404; E-mail: [email protected]

nants S. aureus is mainly involved in mammary infections of lactating females and, in cows, S. au-

reus accounts for 20-30% of total episodes of mastitis. It is estimated that milk losses due to mastitis range from lo-25% of the total yield, according to the severity of inflammation and stage of lactation when the infection occurs. In addition, S. aureus in raw milk used by dairy industries may potentially cause significant public health problems [l]. Much interest has been devoted to the development of a

092%8244/96/$15.00 0 1996 Federation of European Microbiological Societies. All rights reserved

PII SO928-8244(96)00020-X

46 V. Garcia et al./ FEMS Immunology and Medical Microbiology 14 (1996) 45-51

vaccine to prevent staphylococcal bovine mastitis ogy, School of Medicine, and were maintained under but, so far, little success has been achieved. Reasons standard conditions [4]. Female mice weighing 25-30 for this could be the disparity of antigens used, the g after their first pregnancy were used for i.ma. lack of identified virulence factors and/or the time inoculation during lactation. Fourteen days after par- schedule in relation to the lactation cycle [2]. More turition mice were anesthetized with pentobarbital important, no anti-staphylococcal vaccine has ful- and 50 ~1 of a wt-S. aureus suspension containing filled the most desirable goal of vaccination, i.e. 105-lo6 colony forming units (cfu) was injected prevention of infection and disease of the mammary into the fourth mammary gland on each side (L4 and

gland. R4) as described elsewhere [5].

We have obtained temperature-sensitive (tsl mutants of S. aureus and have shown that parenteral immunization with ts-S. aureus confers protection from intraperitoneal (i.p.> challenge with the parental wild-type (wt) S. aureus [3]. We hypothesize that stimulation of the mucosal immune system with a live attenuated vaccine would be a valid approach to prevention of the disease in lactating cows. This study was aimed at testing whether local immunization with ts-S. aureus induces protection from S. aureus intramammary &ma.) challenge. To this pur-

pose we have utilized a mouse model of mastitis.

2.3. Bacteriological and histopathological studies

2. Materials and methods

2.1. Bacteria and culture media

The S. aureus 8325-4 (phi-111 strain (the wt strain) used in this study was generously provided by Dr. John J. Iandolo (Department of Pathology, Col- lege of Veterinary Medicine, Kansas State Univer-

sity, Manhattan, KS). Temperature-sensitive derivatives were obtained from the wt strain as described earlier [3]. The mutant used in this study (A523) replicates well at low temperatures (below 32°C) but undergoes a limited number of divisions when trans- ferred to the mammalian body temperature. Bacteria were cultured on Tryptic Soy agar (Difco Laborato- ries, Detroit, MI) at 28°C (ts mutant) or 37°C (wt strain) for 14 h, and harvested with saline. Bacteria were washed by centrifugation at 5000 X g, 10 min, at 4°C and suspended to the appropriate density in saline.

At different time-points, which ranged from 24 h to 21 days, groups of mice were sacrificed and bacteriological studies were performed using the left-hand side (L,) mammary glands. Briefly, the glands were homogenized by means of a tissue grinder in 2 ml ice-cold distilled water and the homogenates were plated quantitatively on Mannitol Salt agar (Difco) to establish the number of cfu

remaining in each gland. Right-hand side (R4) mammary glands were excised for histological examination. Whole glands were fixed in Bouin’s fluid in order to examine the tissue in its entirety. Samples were dehydrated with 70- 100% ethanol, cleared with xylene and embedded in paraffin. Several slides per mouse were stained with Hematoxylin and Eosin, while some others were used for IgA detection using the peroxidase-antiperoxidase method (Dako, Carpin- teria, CA) [6]. Primary rabbit polyclonal antiserum to mouse IgA was used (Calbiochem-Novabiochem

Corporation, La Jolla, CA).

2.4. Immunization schedules

2.2. Mastitis mouse model

Six- to eight-week-old Swiss mice were obtained from our vivarium at the Department of Microbiol-

One day after parturition, the offspring were removed from lactating mothers 2 h before inoculation. Mothers were anesthetized with pentobarbital and immunized by i.ma. inoculation [5] of 50 ~1 containing 5 X lo6 cfu ts-S. aureus into L, and R, mammary glands. A boosting injection containing the same dose was administered 7 days later by the same route, whereas control mice received saline alone. Other groups of mice were immunized by the i.p. route according to two different immunization protocols. In the first immunization schedule, 5 X lo6 cfu ts-S. aureus was injected i.p. one day after parturition, and mice received a boosting inoculation 7 days

V. Garcia et al. / FEMS Immunology and Medical Microbiology 14 (1996) 45-51

later by the same route. In the other immunization

protocol, animals received three i.p. inoculations (1 X lo* cfu each one) of ts-S. aureus by the i.p. route on days 7-10 before parturition, on day one after parturition, and 7 days later.

2.5. Milking mouse model

Milk was collected using a milking device for small laboratory animals built according to a previ-

ously described design [7]. Pups were removed from lactating mice 2 h before milking to allow milk accumulation in the mammary glands. Mice were anesthetized with pentobarbital and received 3 U oxytocin (BIOL, Buenos Aires, Argentina) by the i.p. route. Fifteen minutes later milk was collected from L, and R, glands and cells were removed by centrifugation at 10000 X g for 10 min at 4°C.

Supematants were kept at -20°C until assayed. In other experiments, milk was cultured quantitatively on Mannitol Salt agar (Difco).

2.6. Antibody detection

Antibody studies were performed on immunized and control mice. Plasma was obtained from blood withdrawn immediately before the animals were challenged, on day 14 after immunization. Milk was obtained on the same day. Plasma and milk levels of IgA and IgG anti-S. aureus antibodies were determined by micro-ELISA [81. Heat-killed S. aureus was used as the ELISA antigen and conjugates were affinity-purified anti-mouse heavy chain goat IgG or IgA conjugated with alkaline phosphatase (Sigma). The enzyme substrate was p-nitrophenyl-phosphate (Sigma) and absorbance at 405 nm was measured with a micro-plate realder (Labsystems Lab., Helsinki, Finland). Results were expressed in ELISA units. Each ELISA unit was defined as Abos X dilution-’ x 100.

2.7. Statistical analysis

Data with non-normal distribution and non-homo-

geneous variances were compared with the Wilcoxon Rank Sum test. Data with normal distribution and homogeneous variances were compared with the Stu- dent’s t-test. P values lower than 0.05 for individual comparisons were considered significant.

47

3. Results

3.1. S. aureus intramammary infection mouse model

Because the characteristics of mammary infection in mice may depend upon the S. aureus strain utilized, we evaluated first the features of the disease caused by strain 8325-4 (phi-11). Mice were chal-

lenged by the i.ma. route with 2.7 X lo5 cfu wt-S. aureus. Twenty-four hours later the average cfu number recovered from the glands was 1 X lo6 cfu and remained almost constant until day 7 post-challenge. By day 10 a significant decrease (P < 0.001,

Student’s t-test) was observed and the cfu number decreased even further by day 21 (Fig. 1). Twenty- four and 48 h after bacterial challenge from lo3 to lo5 cfu ml-’ were detected in the milk of all mice, whereas by 72 h and 7 days after bacterial challenge S. aureus was recovered in only 75% and 50% of the mice, respectively.

Histological studies of mammary glands challenged with S. aureus revealed the development of a classical inflammatory process. One day after chal-

lenge polymorphonuclear leukocyte infiltration, vas- cular congestion and edema were apparent. A re- markable increase in the number of polymorphonuclear leukocytes was observed 4 days after the challenge (Fig. 2B). Lesions were more severe by day 14, when abscesses were observed in several animals (Fig. 2E). In several mice polymorphonuclear leukocyte infiltration was followed by mononuclear cell

33

Y

!$=2

Fig. 1. Isolation of S. auretu from mammary glands after experi-

mental i.ma. challenge. Mice were challenged by the i.ma. route

with 2.7 X lo5 wt-S. aureus cfu. Each point represent the arith-

metic mean + S.E.M. from 6-8 mammary glands.

48 V. Garcia et al./ FEMS Immunology and Medical Microbiology 14 C 1996) 45-51

infiltration (photograph not shown). By day 21 post- infection most mammary glands exhibited an inflammatory response composed of mononuclear cells (Fig. 2F). IgA immunoreactive plasma-cells were found in the mammary gland and draining lymph nodes by day 14 after challenge. By day 21 the number of IgA immunoreactive plasma-cells in the mammary gland

was high and appeared to correlate reciprocally with the number of cfu recovered from mammary gland homogenates. On the basis of bacteriological and histopathological findings, the effect of immunization on the development of murine experimental mastitis was evaluated 4 days after challenge. Nu- merical data, i.e. S. aureus cfu in mammary gland

homogenates and levels of anti-S. aureus antibodies in milk and serum were adopted for further studies.

3.2. The effect of immunization on S. aureus mammary infection

The number of S. aureus cfu recovered from the mammary glands of mice immunized by the i.ma. route was significantly lower (7 X lo2 cfu) than that

found in control mice (1.5 X IO5 cfu) (Fig. 3). Con- versely, the number of cfu recovered from mammary glands of mice immunized by any of the i.p. protocols was as high as that recovered from control mice glands (P > 0.5). Inoculation of ts-S. aureus did not

Fig. 2. Histopathological examination of mammary glands after i.ma. infection with S. aureus. (A) Mock-infected mammary gland of a

pregnant mouse (control). (B) Ninety-six hours post-infection: diffuse inflammatory infiltration in the mammary tissue and surrounding a

blood vessel. Cells were mainly polymorphonuclear leukocytes. (0 Six days and (D) 10 days post-infection: the infiltrate was composed of

mononuclear cells. Two types of lesions were observed by day 14 after i.ma. S. aureus challenge: abscess formation (E) and diffuse

mononuclear cell infiltration (photograph not shown). Finally, by 21 days after challenge the inflammatory pattern was characterized by

mononuclear infiltration of the mammary gland parenchyma (F). Hematoxylin-eosin stain, 330 X

V. Garcia et al. / FEMS Immunology and Medical Microbiology 14 (1996) 45-51 49

g CONTROL IMMUNIZED

Fig. 3. Effect of i.ma. immunization on wt-S. aureus clearance

from the mammary gland. Retrieval of bacteria from mammary

glands of control and immunized mice ( * * : P < 0.01, Wilcoxon

Rank Sum test).

decrease the milk yield, as measured 7 days after the last i.ma. administration. Intramammary immunization (5 X lo6 cfu, twice) increased significantly the

r

6W

100

0 CONTROL IMMUNlZiD CONTROL IMMUNlZED

3000 Serum IgG I I 12000 Serum IgG

*

0

-_II 6gw

2ocQ

CONTROL IMMUNl2ED 0

CONTROL IMMUNIZED

MO

Fig. 4. Effect of i.ma. immunization on anti-S. aureus milk and

serum antibody levels. Top-left: levels of specific anti-S. aurew IgA in milk after i.ma. immunization. Top-right: levels of specific

anti-S. aureus IgA in serum after i.ma. immunization. Bottom-left:

levels of specific anti-S. aureur IgG in serum after ima. immu-

nization. Bottom-right: levels of specific anti-S. aureus IgG in

serum after i.p. immunization (1 X 10s cfu, three times). The

significance level was (* ) P < 0.05, Wilcoxon Rank Sum test.

Each bar represent the metdian from lo-15 mice. Shaded bars:

i.ma. immunization. Solid bars: i.p. immunization.

levels of milk specific IgA (Fig. 4, top-left) and

serum specific IgA and IgG (Fig. 4, top-right and bottom-left, respectively). Conversely, i.p. administration of ts-S. aureus (same dose, same schedule, different route) failed to increase the levels of either milk or serum, IgG or IgA specific antibodies. In- traperitoneal immunization (1 X lo8 ts-S. aureus cfu, three times) significantly increased the levels of circulating specific IgG (Fig. 4, bottom-right), as i.ma. immunization did, but failed to increase the levels of circulating anti-S. aureus IgA. Interestingly, im- proved clearance of wt-5. aureus from the mammary gland after i.ma. immunization did not appear to be related to the levels of circulating IgG specific antibodies.

4. Discussion

It has long been an immunological axiom that mucosal responses are best stimulated by local appli- cation of antigen. It has also been reported that i.ma. vaccination increases the level of specific antibody in milk, and that this relative increase exceeded that

occurring in the blood [9]. Using a murine model of mastitis, we show in this study that immunization with a temperature-sensitive mutant of S. aweus by the i.ma. but not the i.p. route decreases the severity of mammary gland infection. This finding correlated with increased specific immunoglobulin levels in milk.

Live-attenuated mutants of many different pheno- types can be constructed from different parental

strains. Because many potentially useful derivatives need to be tested before choosing the vaccine strain, it is imperative to have an animal model that can be utilized for screening purposes. The logical approach is to use a model developed in a small-size mammal. The mouse mammary gland model utilized in this study offers the opportunity of using a large number of animals and, given the small size of the glands, studying bacterial growth and the histopathological modifications using whole glands in vivo [l]. In addition, the use of the milking model described in this report allowed us to evaluate the presence of specific milk immunoglobulins by a procedure that offers several advantages over other previously described methods [lO,ll]. We were able to collect

50 V. Garcia et al./ FEMS Immunology and Medical Microbiology 14 (I9961 45-51

multiple milk samples from the same mother and to recover a significant milk volume as soon as two hours after the pups were separated from their mother. We did not need to remove the nipples for milk collection as it was done by Parr et al. [l l] and, moreover, the pups were successfully reunited with

their mothers after milking. Although our findings by day 21 after S. UUY~US challenge might reflect the first stages of a chronic process in the mouse mammary gland, there is no doubt that natural disease of the cows and experimental infection in mice have significant differences. Therefore, once a suitable ts-S. aureus strain has been selected in screening studies conducted with the mouse model, definitive immunopathological evaluation will have to be conducted in dairy cows.

The purpose of vaccination is to increase the efficacy of mammary gland natural defenses against S. aureus infection [l]. To develop a vaccine, Yoshida et al. [ 121 and Watson [ 131 have considered

exopolysaccharide production by including encapsu- lated strains of S. aureus. Watson [14,15] also demonstrated that milk losses can be diminished by parenteral vaccination with a preparation containing killed bacteria, a toxoid derived from S. aureus

toxin-alpha and an adjuvant. The ultimate usefulness of the vaccine prepared by Watson remains to be determined. On a different approach, Fattom et al. [16] have developed a vaccine prepared with ex- opolysaccharides isolated from S. aureus, using protein carriers as immunopotentiators, as described in

other licensed conjugated vaccines. Amorena et al. [ 171 have described a staphylococcal mastitis vaccine which includes inactivated bacteria, S. aureus toxoid and S. aureus exopolysaccharide immunopotentiated with liposomes. Other authors, however, considered that control of immune responses and safety of the vaccine could be more easily reached by using selected, purified and well-characterized bacterial antigens instead of crude whole-cell preparations [l]. As an example of that, Mamo et al. [2] have demonstrated that a relatively small portion of fibronectin- binding protein can induce considerable protection from challenge with S. aureus in a mouse model. The low efficacy of most of these potential vaccines, however, does not support their practical use.

With regard to the administration route for a staphylococcal mastitis vaccine our results agree with

those of Finch et al. [18], who demonstrated the efficacy of local immunization of cattle against Streptococcus uberis experimental i.ma. challenge. The authors showed that the only antibody which appeared to be involved in protection was S. uberis-

specific IgM induced by i.ma. immunization. In this

study we demonstrated that protection of the murine mammary gland against infection with S. uureus can be achieved by i.ma. vaccination with a ts-S. aureus.

The reduction of cfu numbers in mammary glands after i.ma. immunization appears to correlate with increased levels of S. uureus-specific IgA in milk but not with the levels of milk or serum IgG. Indeed, mice immunized by the i.p. route were not protected against i.ma. challenge with S. aureus and no levels of specific IgA were detected in milk from these

animals. In previous studies using other bacterial strains we demonstrated that local immunization with

live-attenuated bacteria induces protection in the lower airways and that, similarly, this protection correlated with high levels of specific IgA but not with the levels of plasma IgG [19,20]. In this study, correlation between protection and milk specific IgA levels may not necessarily mean that milk IgA is the sole factor responsible for protection, because an- tibacterial defenses of the mammary gland are multi- factorial and other mechanisms may be responsible for enhanced clearance of S. aureus. Therefore, a specific role for cellular immunity cannot be ruled out and more detailed knowledge of this response merits further investigation.

In conclusion, the feasibility of using local immunization protocols with live-attenuated strains of S. uureus to induce immunity in the mouse mammary gland was established. The chronic mouse model of mastitis is proposed as a useful system for screening of ts-S. uureus strains to be utilized in the development of a vaccine to prevent staphylococcal mastitis in dairy ruminants.

Acknowledgements

This study was financed in part by grants from UBACYT, Universidad de Buenos Aires; IDB-CON- ICET, Buenos Aires; the International Foundation for Sciences (IFS), Stockholm, Sweden; the ‘Albert0 J. Roemmers’ Foundation, Buenos Aires; and the Antorchas Foundation, Buenos Aires, Argentina.

V. Garcia et al. / FEM.7 Immunology and Medical Microbiology 14 (1996) 45-51 51

References

[l] Sutra, L. and Poutrel, B. (1994) Virulence factors involved in

the pathogenesis of bovine intramammary infections due to

Staphylococcus aureus. J. Med. Microbial. 40, 78-89.

[2] Mamo, W., Jonsson, P., Flock, J., Lindberg, M., Muller, H.,

Wadstrom, T. and Nelson, L. (1994) Vaccination against

Staphylococcus aureus mastitis: immunological response of

mice vaccinated with fibronectin-binding protein (FnBP-A)

to challenge with S. aureus. Vaccine 12, 988-992.

[3] Sordelli, D.O., Iglesias. M.F., Cerquetti, M.C., Catalano, M.

and Morris-Hooke, A. ‘(1993) Temperature-sensitive mutants

of Staphylococcus aureus:isolation and preliminary charac-

terization. Curr. Microbial. 27, 125-129.

[4] U.S. Department of Health and Human Services. Guide for

the care and use of laboratory animals. Public Health Ser-

vice, National Institutes of Health, NIH Publication N”.

86-23, Bethesda, Maryland, 1985.

[5] Chandler, R.L. (1970) Experimental bacterial mastitis in the

mouse. J. Med. Microbial. 3, 273-282.

[6] Stemberg, L.A., Hardy, P.H., Cuculis, J.J. and Meyer, H.G.

(1970) The unlabeled .antibody enzyme method of immuno-

chemistry. Preparation and properties of soluble antigen anti-

body complex (horseradish peroxidase-anti-horseradish per-

oxidase) and its use in identification of Spirochetes. J. His-

tochem. Cytochem. 18, 315-333.

[7] Temple, P.L. and Kon, K.S. (1937) A simple apparatus for

milking small laboratory animals. Biochem. 31, 2197-2198.

[8] Sordelli, D.O., Rojas, R.A., Cerquetti, MC., Morris Hooke,

A., Degnan, P.J. and Bellanti, J.A. (1985) Enzyme-linked

immunosorbent assay antibody responses to a temperature-

sensitive mutant of Pseudomonas aeruginosa. Infect. Im- mun. 50, 324-327.

[9] Nonnecke B.J., Elsken L.A. and Kehrli M.E., Jr. (1986)

Local and systemic immune response in the cow after intra-

mammary vaccination during lactation. Vet. Immunol. Im-

munopathol. 11, 31-4,4.

[lo] McBumey, J.J., Meier.. H. and Hoag, W.G. (1964) Device for

milking mice. J. Lab. Clin. Med. 64, 485.

[ll] Parr, E.L., Bozzola, J.J. and Parr, M.B. (1995) Purification

and measurement of secretory IgA in mouse milk. J. Im-

munol. Methods 180, 147-157.

[12] Yoshida, K., Ichiman, Y. and Narikawa, S. (1984) Staphylo-

coccal capsular vaccine for preventing mastitis in two herds

in Georgia. J. Dairy Sci. 67, 620-627.

[13] Watson, D.L. (1988) Vaccination against experimental

staphylococcal mastitis in ewes. Res. Vet. Sci. 45, 16-21.

[14] Watson, D.L. (1992) Vaccination against experimental

staphylococcal mastitis in dairy heifers. Res. Vet. Sci. 53,

346-353.

[15] Watson, D.L. and Davies, H.I. (1993) Influence of adjuvants

on the immune response of sheep to a novel Staphylococcus aureus vaccine. Vet. Microbial. 34, 139-153.

[16] Fattom, A., Shiloach. J., Bryla, D., Fitzgerald, D., Pastan, I.,

Karakawa, W.W., Robbins, J.B. and Schneerson, R. (1992)

Comparative immunogenicity of conjugates composed of the

Staphylococcus aureus type-8 capsular polysaccharide bound

to carrier proteins by acid dihydrazide or N-succinimidyl-Z

(2-pyridyldithio) propionate. Infect. Immun. 60, 584-589.

[17] Amorena, B., Baselga, R. and Albizu, I. (1994) Use of

liposome-immunopotentiated exopolysaccharide as a compo-

nent of an ovine mastitis staphylococcal vaccine. Vaccine 12,

243-249.

[18] Finch, J.M., Hill, A.W., Field, T.R. and Leigh, J.A. (1994)

Local vaccination with killed Streptococcus uberis protects

the bovine mammary gland against experimental intramam-

mary challenge with the homologous strain. Infect. Immun.

62, 3599-3603.

1191 Sordelli, D.O., Garcia, V.E., Fontan, P.A., Cerquetti, M.C.

and Morris Hooke, A. (1992) Intranasal immunization with

temperature-sensitive mutants protects granulocytopenic mice

from lethal pulmonary challenge with Pseudomonas aeruginosa. Curr. Microbial. 24, 9-14.

[20] Garcia, V.E., Sordelli, D.O., Gherardi, M.M. and Morris

Hooke, A. (1991) Humoral immunity induced in the lower

respiratory tract by local immunization with temperature-sen-

sitive mutants of Pseudomonas aeruginosa. Curr. Microbial.

22, 265-269.

intramammary immunization with live-attenuated staphylococcus aureus: microbiological and...

Documents