Qr

STFPa

b

c

(d

Ce

f

a

ARR1A

KVAQQBIC

1

ilcimmit

h0

Comparative Immunology, Microbiology and Infectious Diseases 45 (2016) 1–8

Contents lists available at ScienceDirect

Comparative Immunology, Microbiologyand Infectious Diseases

jo u r n al homepage: www.elsev ier .com/ locate /c imid

uillaja brasiliensis saponins induce robust humoral and cellularesponses in a bovine viral diarrhea virus vaccine in mice

amuel Paulo Cibulskia,b,1, Fernando Silveirac,∗,1, Gustavo Mourglia-Ettlind,hais Fumaco Teixeiraa,b, Helton Fernandes dos Santosa,b, Anna Carolina Yendoe,ernanda de Costae, Arthur Germano Fett-Netoe, Grace Gosmannf,aulo Michel Roehea,b

FEPAGRO Saúde Animal, Instituto de Pesquisas Veterinárias Desidério Finamor, Laboratório de Virologia, Eldorado do Sul, RS, BrazilDepartamento de Microbiologia Imunologia e Parasitologia, Laboratório de Virologia, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, BrazilLaboratorio de Carbohidratos y Glicoconjugados, Departamento de Desarrollo Biotecnológico – Facultad de Medicina, Universidad de la RepúblicaUdelaR), Av. Alfredo Navarro 3051, CP 11600, Montevideo, UruguayCátedra de Inmunología, Departamento de Biociencias – Facultad de Ciencias/Química, Universidad de la República (UdelaR), Av. Alfredo Navarro 3051,P 11600, Montevideo, UruguayLaboratório de Fisiologia Vegetal, Centro de Biotecnologia e Departamento de Botânica, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, BrazilFaculdade de Farmácia, Universidade Federal do Rio Grande do Sul, Av. Ipiranga 2752, Porto Alegre 90610-000, RS, Brazil

r t i c l e i n f o

rticle history:eceived 17 September 2015eceived in revised form6 December 2015ccepted 19 January 2016

eywords:

a b s t r a c t

A saponin fraction extracted from Quillaja brasiliensis leaves (QB-90) and a semi-purified aqueous extract(AE) were evaluated as adjuvants in a bovine viral diarrhea virus (BVDV) vaccine in mice. Animalswere immunized on days 0 and 14 with antigen plus either QB-90 or AE or an oil-adjuvanted vaccine.Two-weeks after boosting, antibodies were measured by ELISA; cellular immunity was evaluated byDTH, lymphoproliferation, cytokine release and single cell IFN-� production. Serum anti-BVDV IgG, IgG1and IgG2b were significantly increased in QB-90- and AE-adjuvanted vaccines. A robust DTH response,

+

accinedjuvantuillaja brasiliensisB-90VDV

mmune response

increased splenocyte proliferation, Th1-type cytokines and enhanced production of IFN-� by CD4 andCD8+ T lymphocytes were detected in mice that received QB-90-adjuvanted vaccine. The AE-adjuvantedpreparation stimulated humoral responses but not cellular immune responses. These findings reveal thatQB-90 is capable of stimulating both cellular and humoral immune responses when used as adjuvant.

© 2016 Elsevier Ltd. All rights reserved.

ellular immunity

. Introduction

Bovine viral diarrhea (BVD) is a viral disease with major signif-cance in cattle, where it may be responsible for huge economicosses [1]. Acute bovine viral diarrhea virus (BVDV) infections areommon, usually resulting in mild disease characterized by fever,ncreased respiratory rate, diarrhea and leucopenia. Although ani-

als generally recover, the effect of BVDV on the immune systemay impair the host’s resistance to disease [2]. In addition, BVDV

nfections in pregnant cows can give rise to abortions, malforma-ions, reduced size at birth, decreased growth curve and birth of

∗ Corresponding author.E-mail address: fernandosilveiragonzalez@gmail.com (F. Silveira).

1 Contributed equally to this work.

ttp://dx.doi.org/10.1016/j.cimid.2016.01.004147-9571/© 2016 Elsevier Ltd. All rights reserved.

persistently infected calves; the latter is a major source for virusperpetuation and dissemination of the infection in herds [2].

Vaccination aims to induce the development of acquired immu-nity by inoculation of immunogenic components of a particularpathogen or closely related microorganisms [3]. Nonliving vaccineantigens, especially purified or recombinant subunit vaccines, areoften poorly immunogenic and require additional components tohelp stimulate protective immunity based on antibodies and effec-tor T cell functions [4,5]. Thus, adjuvants capable of enhancing bothhumoral and cellular immune responses are a subject of high inter-est to vaccinology as well as to the vaccine industry [6].

Triterpenoid saponins extracted from Quillaja saponaria Molinahave a long usage record as adjuvants in veterinary vaccines [6,7].

Saponins extracted from an alternative three, Quillaja brasiliensis,native to South America, have been evaluated as immune enhancer[8–10]. In particular, a saponin fraction extracted from Q. brasilien-sis named QB-90 and aqueous leaves extracts (AE) of this species

2 Micro

wrb

cpdp

2

2

rfdptcp(Lf

2

ohwtas1cei

2

Aiscem(Mens(td(t

2

FAm

S.P. Cibulski et al. / Comparative Immunology,

ere demonstrated to stimulate both humoral and cellular immuneesponses to viral antigens at levels comparable to those inducedy Q. saponaria saponins [8–10].

The aim of this study was to evaluate the potential adjuvantapacity of QB-90 and AE when added to experimental vaccinereparations formulated with a model viral antigen (bovine viraliarrhea virus; BVDV) and inoculated into mice. These were com-ared to classical adjuvants as Alum and IFA.

. Material and methods

.1. Vaccine adjuvants

Q. brasiliensis leaves were collected from adult plants natu-ally growing in the municipality of Canguc u, RS, Brazil. Extractsrom leaves of Q. brasiliensis were prepared from air-dried pow-ered leaves resuspended in distilled water (1:10, w/v), filtered,artitioned with ethyl acetate and lyophilized. The product washen submitted to further purification through reverse-phasehromatography to obtain fraction QB-90, as described in detailreviously [10]. Quil A® was purchased from Brenntag BiosectorDenmark). Alum was obtained from Omega Produtos Quimicostda. (Brazil). Incomplete Freund’s Adjuvant (IFA) was purchasedrom Sigma (USA).

.2. Cytotoxicity

The haemolytic activity of saponins is one of the main indicatorsf cytotoxicity. In order to make a comparative evaluation of theaemolytic activities of Quil A, QB-90 and AE, the three preparationsere tested for induction of haemolysis over a range of concen-

rations (10–2000 �g/mL) on 0.5% rabbit red blood cells (RBCs),s previously described [9]. Physiological saline solution and Q.aponaria saponins at 250 �g/mL were used as indicators of 0% and00% haemolysis, respectively. Samples were tested in quadrupli-ate in V-bottomed microtiter plates. The haemolytic activity wasxpressed as the end point dilution capable of inducing haemolysisn 50% of the RBCs (HD50).

.3. Antigen production

MDBK (Madin-Darby bovine kidney cells) were obtained fromTCC (originally CCL-22TM) and cultured at 37 ◦C in a 5% CO2

ncubator in Eagle’s minimal essential medium (E-MEM; Gibco)upplemented with 10% fetal bovine serum and antibiotics (peni-illin 100 IU/mL; streptomycin 100 �g/mL). Cells were subculturedvery 3–4 days following standard procedures [11]. For virusultiplication, an autochthonous cytopathogenic BVDV1 isolate

EVI001/94) was inoculated onto nearly confluent monolayers ofDBK cells at a multiplicity of infection of 0.01. When cytopathic

ffect was evident in about 90% of monolayers, cells and super-atants were harvested and frozen at −80 ◦C. Subsequently, theupernatant was thawed, clarified by low speed centrifugation1500 × g for 10 min) aliquoted and used as virus’ stocks. Infec-ious titers obtained were around 108.3 50% tissue culture infectiousoses per mL (TCID50) before inactivation with binary ethylenimineBEI) [12]. The inactivated virus suspension was used as antigen inhe vaccine formulations.

.4. Vaccination of mice

Female Swiss mice (6–7 weeks old) were purchased from theundac ão Estadual de Produc ão e Pesquisa em Saúde (FEPPS, Portolegre, Brazil). Prior to the beginning of the experiments, the ani-als were acclimatized for 1 week under controlled temperature

biology and Infectious Diseases 45 (2016) 1–8

(22 ± 2 ◦C) and humidity, with a cycle of 12/12 h day/night and fedwith standard pelleted food and tap water ad libitum.

Mice were then divided in ten groups of seven; each group wasimmunized subcutaneously on the hind neck with one of the fol-lowing: 150 �L of BVDV antigen in the presence of 50 �L of QB-90(containing 100, 50 or 10 �g per dose); AE (containing 400, 200 or100 �g per dose); Quil A (50 �g per dose); alum (100 �g per dose);incomplete Freund’s adjuvant (IFA, 150 �L of antigen emulsifiedwith 150 �L of oil) or without adjuvant (antigen only). The formu-lations of BVDV vaccines were filtered through 0.22 �m (Millipore)and kept at 4 ◦C until use. A booster injection with the same volumeof vaccine preparations (200 �L) was applied 2 weeks later (day14). Blood were collected via tail vein on days 0, 14 and 28 post-inoculation of the first dose of vaccine and kept frozen at −20 ◦Cuntil processed further.

All experiments were performed in compliance with the Euro-pean Convention for the Protection of Vertebrate Animals Usedfor Experimental and Other Scientific Purposes (European TreatySeries–No. 170 revised 2005) and the procedures of the Brazil-ian College of Animal Experimentation (COBEA). The project wasapproved by the Ethics Commission on Animal Experimentation(CEUA) of FEPAGRO Animal Health, Instituto de Pesquisas Veter-inárias Desidério Finamor (IPVDF).

2.5. Immunoassays for antibodies

Anti-BVDV IgGt (total), IgG1 and IgG2a were determined ineach serum sample in an indirect ELISA. ELISA plates (Nunc) werecoated with the same BVDV antigen used for preparation of thevaccine antigens above and diluted (1:100, v/v) in PBS (pH 7.2)for 18 h at 4 ◦C. After adsorption, plates were washed two timeswith 200 �L of PBS containing 0.05% Tween-20 (PBS-T). Wells thenreceived 180 �L of a blocking solution (PBS-T with 5% non-fatdry milk) and incubated for 60 min at 37 ◦C. Subsequently, theplates were washed twice with PBS-T. Sera were appropriatelydiluted in PBS-T (1:300 to IgGt and IgG1 and 1:100 to IgG2a) andadded to duplicate wells. After 1 h at 37 ◦C, plates were washedthree times with PBS-T and incubated with adequate dilutions ofperoxidase conjugated anti-mouse IgG, anti-mouse IgG1 or anti-mouse IgG2a (Sigma) for 1 h at 37 ◦C. After washing, 100 �L of OPD(ortho-phenylenediamine, Sigma) with 0.03% of H2O2 were addedto each well. After 30 min of incubation in the dark at room tem-perature, the reaction was stopped with the addition of 1 M HCl(25 �L/well). Optical density was determined in a microplate readerset to 492 nm. Data were expressed as the mean OD value of thesamples minus the mean OD recorded in control wells.

2.6. Delayed type hypersensitivity (DTH) assay

DTH responses were evaluated in three animals of each group onday 28 post inoculation of the first dose. The assay was performedby intradermal (ID) injections of 10 �L of the same BVDV antigenused for production of the vaccines, in the right hind footpad. Thethickness of the footpad was measured with a caliper 24 h beforeand 24 h after ID injections. The BVDV-specific response of eachanimal was calculated based on values of its injected footpad minusthe average of the basal swelling [13].

2.7. Splenocyte proliferation assay

Twenty eight days after the priming, spleens were collectedunder aseptic conditions, immersed in RPMI 1640 medium (Invi-

trogen) and mechanically dissociated to obtain a homogeneouscell suspension. Erythrocytes were lysed with ACK (Ammonium-Chloride-Potassium) lysis buffer. Subsequently, cells were pelletedby low speed centrifugation (380 × g at 4 ◦C for 10 min), washed

Micro

ts1aewopoh(tpfdwtwwc

cc2wwuctaM

2

Bafl(ot0

2

g(dPap4FMrctt

2

at

growing interest for safer and more effective vaccines for ani-mals and humans [15]. Here, semi-purified aqueous extracts of Q.brasiliensis, as well as QB-90 fraction of saponins increased anti-BVDV total IgG levels through elevation of IgG1, a Th2 antibody

S.P. Cibulski et al. / Comparative Immunology,

hree times in fresh RPMI 1640 and resuspended in theame medium supplemented with 0.05 mM 2-mercaptoethanol,00 IU/mL penicillin, 100 �g/mL streptomycin, 2 mM l-glutamine,nd 10% fetal bovine serum. Cell counting by trypan blue dyexclusion revealed >95% viability. Splenocytes were then seededith 100 �L of a cell suspension containing 2.5 × 106 cell/mL

f medium into each well of a 96-well flat-bottom microtiterlate (Nunc). Subsequently, 100 �L of BVDV antigen or mediumnly was added. Plates were then incubated at 37 ◦C in aumid atmosphere containing 5% CO2. After 68 h, 50 �L of MTT1-(4,5-Dimethylthiazol-2-yl)-3,5-diphenylformazan, Sigma) solu-ion (2 mg/mL) were added to each well and incubated for 4 h. Thelates were then centrifuged at 1400 × g for 5 min and the untrans-ormed MTT was removed carefully by pipetting. Subsequently, aimethyl sulfoxide (DMSO) solution (192 �L DMSO; 8 �L 1 N HCl)as added to wells in volumes of 100 �L. After 15 min incubation,

he absorbance was measured in an ELISA reader at 550 nm withavelength reference fixed at 620 nm. The stimulation index (SI)as calculated as the ratio of absorbance of mitogen-stimulated

ultures and the absorbance of non-mitogen-stimulated cultures.In order to evaluate CD8+ T-cell proliferation, splenocytes (106

ells) were labeled with 1 �M of CFSE (carboxyfluorescein suc-inimidyl ester, CellTraceTM, Life Technologies) and cultured in4-well plates for 3 days as above. After culturing, cells wereashed twice in PBS with 2% of fetal bovine sera and stainedith anti-CD8 (CD8a Rat Anti-Mouse mAb, PE conjugate, Molec-lar Probes). The cells were analyzed in a fluorescence activatedell sorter (FACScanto II; BD). CD8+ proliferation was expressed ashe ratio of divided daughter cells to total lymphocytes, expresseds a percentage, in analogy to calculation of a stimulation index inTT proliferation assays.

.8. Cytokine quantification by cytometric bead array (CBA)

Splenocytes (5 × 105 cells) were pulsed for three days withVDV antigen (as described above, as in splenocyte proliferationssay) and supernatants harvested. Cytokines were measured byow cytometry with aid of the mouse BD cytometric bead arrayCBA) mouse Th1/Th2/Th17 Cytokine Kit (BD Biosciences). The the-retical limit of detection of the analyzed cytokines was 0.1 pg/mLo IL-2, 0.03 pg/mL to IL-4, 1.4 pg/mL to IL-6, 0.5 pg/mL to IFN-�,.9 pg/mL to TNF, 0.8 pg/mL to IL-17 and 16.8 pg/mL to IL-10.

.9. Measurement of CD4+/IFN-�+ and CD8+/IFN-�+ cells

Splenocytes (1 × 106 cells) were pulsed for 18 h with BVDV anti-en (as described above). After 13 h of antigen pulse, brefeldin AGolgiPlugTM, Pharmingen) was added to the medium as previouslyescribed [14]. Following incubation, cells were washed twice inBS supplemented with 2% FBS and stained with anti-CD8 andnti-CD4 conjugates (FITC conjugates, Molecular Probes), in inde-endent assays. Subsequently, the cells were washed, fixed with% paraformaldehyde and permeabilized with saponin (Sigma).ixed cells were then stained intracellularly for IFN-� (PE conjugate,olecular Probes). FACS analysis was performed as above and the

esults were analyzed with the FlowJo (Tree Star) software. The per-entage of virus specific IFN-� synthesis was calculated accordingo the formula: (% of cells synthetizing IFN-� after virus stimula-ion) − (% of cells synthetizing IFN-� after culture in medium only).

.10. Statistical analysis

GraphPad Prism 5.0 was used for drawing graphs and statisticalnalyses. Significant differences between experimental and con-rol groups were analyzed by one-way ANOVA, using Dunnett’s

biology and Infectious Diseases 45 (2016) 1–8 3

post-test. Differences were considered statistically significantwhen P ≤ 0.05.

3. Results and discussion

3.1. QB-90 saponins and AE from Q. brasiliensis have lowerhaemolytic activity than Quil A

Fig. 1 shows the compared haemolytic activities of QB-90,AE and Quil A. The HD50 values were 90.89 �g/mL for QB-90,1214.04 �g/mL for AE and 27.99 �g/mL for Quil A. The AE and QB-90 saponins fraction were significantly less hemolytic than Quil A(Fig. 1).

Critical attributes to be examined in the development of vac-cine adjuvants are efficacy, safety and tolerability. In the resultsobtained here, no evidence of local or systemic toxicity was foundeven after multiple injections of the adjuvanted vaccine. This is con-sistent with data from our previous experiments that confirmedthat QB-90, as well as AE, were efficacious, safe and well toleratedin mice [8,9].

3.2. QB-90 and AE enhance anti-BVDV antibody responses in mice

To evaluate the humoral immune responses, serum levels ofBVDV-specific IgG and isotypes in all groups of vaccinated micewere assessed by indirect ELISA. Total specific IgG as well as IgG1and IgG2a were elevated in all groups inoculated with QB-90-adjuvanted preparations (Fig. 2). Groups that received vaccinepreparations with 400 and 200 �g of AE also induced a statisti-cally significant increase in antibody levels. However, IgG levels onthe group immunized with the preparation containing 100 �g ofAE as adjuvant were not significantly enhanced (Fig. 2). A dose-dependent antibody response was detected in mice immunizedwith vaccines containing QB-90 and AE as adjuvants. Mice immu-nized with vaccine preparations containing the largest amount ofQB-90 induced the highest antibodies levels; mice immunized withAE-adjuvanted vaccines displayed levels of enhancement similar tothose verified in the IFA-adjuvanted group.

The search of new immunoadjuvants based on plant-derivedsaponins has increased in recent years, particularly in view of the

Fig. 1. In vitro toxicity of QB-90, AE and Quil A. Haemolytic activity expressedas percent haemolysis referred to saline and Q. saponaria saponins (250 �g/mL),which were used as 0% and 100%, respectively. Results are presented as the meanvalue ± S.D. (n = 4).

4 S.P. Cibulski et al. / Comparative Immunology, Microbiology and Infectious Diseases 45 (2016) 1–8

Fig. 2. QB-90 saponins from Quillaja brasiliensis and aqueous extract (AE) enhances anti-BVDV antibody titers. Adult female CF1 mice were immunized twice s.c. at a 2-weekinterval with BVDV antigen alone (empty bar) or together with QB-90 (light gray bars) or AE (gray bars). Blood samples were collected 2 weeks after the second immunizationa re indB

iiw

3r

fT

nd anti-BVDV total IgG, IgG1, IgG2a measured by ELISA. Significant differences aVDV antigen and no adjuvant.

sotype, with contribution of IgG2a, a Th1 isotype. These results aren agreement with those previously obtained when viral antigens

ere co-administered with QB-90 saponins [8–10].

.3. QB-90 and AE increases anti-BVDV cellular immuneesponses when used as adjuvants

Delayed-type hypersensitivity (DTH) is a useful approachor evaluating cell-mediated immune responses associated withh1 reactivity. Immunized mice were subjected to detection of

icated: *(P ≤ 0.05), **(P ≤ 0.01) and ***(P ≤ 0.001), with the group immunized with

BVDV-specific DTH reaction using the footpad swelling response2 weeks after boosting. Fig. 3 summarizes the results of the DTHassay in the different groups of mice. A significant DTH responsewas observed in mice immunized with QB-90 (100 and 50 �g), AE(400 and 200 �g), Quil A and IFA-adjuvanted preparations. Hereagain dose-related DTH responses to QB-90-adjuvanted and AE-

adjuvanted preparations were evident. The vaccines containing100 �g QB-90 and 400 �g AE proved superior to Quil A and IFA onthe generation of DTH responses, whereas QB-90 50 �g and QuilA showed no significant differences. As positive DTH reaction is

S.P. Cibulski et al. / Comparative Immunology, Microbiology and Infectious Diseases 45 (2016) 1–8 5

Fig. 3. Delayed type hypersensitivity (DTH) in immunized mice. The assay was carried out 2 weeks after the second immunization of mice with the BVDV preparatione ols were mice injected only BVDV antigen (no adjuvant). The DTH responses expresseda re indicated: *(P ≤ 0.05), **(P ≤ 0.01) and ***(P ≤ 0.001), with the group immunized withu

act

a2gecFncwpB

iidriffb9a(wvt1ctUwt

p[arrp

Fig. 4. Splenocyte proliferation assays. Mice (n = 6/group) were s.c. immunized withBVDV plus saline, or adjuvanted with 100 �g of QB-90 (100 �g), 400 �g of AE(400 �g), IFA or unadjuvanted vaccine. Splenocytes were prepared 2 weeks afterboosting and cultured with BVDV antigen or RPMI 1640 medium during 68 h. Spleno-cyte proliferation was measured by the MTT method (refer to text for methods), andis shown as a stimulation index (SI) (A). CD8+ T-cell proliferation was measured byculturing CFSE-labeled splenocytes with BVDV antigen for 3 days (B). The valuesare presented as means ± S.D. Significant differences are indicated: *(P ≤ 0.05) and

ither with no adjuvant or formulated with QB-90, AE and others adjuvants. Contrs mean values of footpad thickness increase ± S.D. (n = 3); significant differences anadjuvanted vaccine.

ttributed to memory Th1 CD4+ T cells [16] and these results indi-ate that AE and QB-90 are capable of stimulating Th1 cells againsthe administered antigen.

To assess whether AE and QB-90 saponins adjuvant enhancesnti-BVDV T-cell responses splenocytes were isolated from mice8 days post-BVDV immunization and cultured with BVDV anti-en for 3 days in a standard MTT and CFSE proliferation assays. Theffects of AE (400 �g) and QB-90 (100 �g) as well as IFA on spleno-yte proliferation in response to BVDV stimulation are shown inig. 4. Mice that received the QB-90-adjuvanted vaccine had sig-ificantly higher BVDV-specific cell proliferation responses thanontrol mice (P ≤ 0.01) (Fig. 4A). Furthermore, mice receiving BVDVith QB-90 had significantly enhanced BVDV-specific CD8+ T-cellroliferation, when compared to those of mice immunized withVDV alone (P ≤ 0.001) (∼3 fold increase, Fig. 4B).

Th1 cells produce IFN-�, IL-2, and lymphotoxins, playing a crit-cal role in directing cell-mediated immune responses, which aremportant for clearance of intracellular pathogens. Th2 cells pro-uce IL-4, IL-5, IL-10, and IL-13, which are important for humoralesponses [17,18]. To better characterize the cytokine profilenduced by immunization of BVDV with AE and QB-90, splenocytesrom immunized animals were re-stimulated with BVDV in vitroor 3 days and supernatants measured using a BD Th1/Th2/Th17ead array (Fig. 5). The strong induction of Th1 antibodies by QB-0 saponins was also observed on the Th1 cytokine levels. QB-90djuvanted vaccine induces high levels of IFN-� (P ≤ 0.01), TNFP ≤ 0.001) and IL-2 (P ≤ 0.01) (Th1 induced cytokines). The resultsere in agreement with elevated IgG2a antibody levels in mice

accinated with QB-90. Moreover, QB-90 significantly enhanceshe production of IL-10 (suppressive cytokine) (P ≤ 0.05), and IL-7 (Th17 cytokine) (P ≤ 0.05). These data reveal that Th2-inducedytokines (IL-4 and IL-6) were not significantly increased in micehat received the QB-90-adjuvanted vaccine, suggesting a Th1 bias.nlike the group immunized with QB-90, the group immunizedith the AE-adjuvanted vaccine revealed no significant increase in

he production of any of the cytokines analyzed (P ≥ 0.05).Another important adjuvant attribute, particularly for thera-

eutic vaccines, is the ability to induce robust T-cell responses19]. The induction of memory CD8+ T-cells by exogenous

ntigen requires antigen cross-presentation, normally a featureestricted to dendritic cells [20]. The strong anti-BVDV CD8+ T-cellesponses detected in mice immunized with the QB-90-adjuvantedreparation perhaps indicates that QB-90 enhances antigen

**(P ≤ 0.01), with the group immunized with unadjuvanted vaccine.

6 S.P. Cibulski et al. / Comparative Immunology, Microbiology and Infectious Diseases 45 (2016) 1–8

Fig. 5. Th1, Th2 and Th17 cytokine profile. Spleens were collected from mice (n = 3) immunized with BVDV alone or with IFA, AE or QB-90, and cultured with BVDV antigen fort and pra

cd

trMrb

hree days. Cytokines in the supernatant were quantitated by cytometric bead array

nd ***(P ≤ 0.001), with the group immunized with unadjuvanted vaccine.

ross-presentation possibly by a chemotactic effect that recruitsendritic cells to the site of immunization.

An important natural mechanism for clearance of viral infec-ions in vivo is represented by cell-mediated immunity. CD8+ T cells

ecognize viral antigens in the form of short peptides presented byHC class I molecules on the surface of virus-infected cells. The

ecognition of these viral peptides in the context of MHC moleculesy CD8+ T cells can trigger the specific lysis of infected cells [21]

esent by mean ± S.E.M. Significant differences are indicated: *(P ≤ 0.05), **(P ≤ 0.01)

or noncytopathic intracellular inactivation of virus mediated bycytokines such IFN-� and TNF-� [22,23].

To investigate IFN-� production by T cells, splenocytes wererestimulated with BVDV and the frequencies of IFN-�+ CD4+

and IFN-�+ CD8+ T cells were analyzed by flow cytometry.Mice immunized with unadjuvanted vaccine were used as con-trols. Frequencies of IFN-�+ CD4+ T cells and IFN-�+ CD8+ Tcells were significantly elevated in mice immunized with the

S.P. Cibulski et al. / Comparative Immunology, Micro

Fig. 6. QB-90 induces antigen-specific IFN-� production in CD4+and CD8+ T cells.Mice (n = 3) were immunized with unadjuvanted antigen (BVDV) or antigen plus IFA,AE (400 �g) or antigen plus QB-90 (100 �g). Mice were euthanized on day 28 afterfirst immunization to collect spleens. Splenocytes (1 × 106 cells) were re-stimulatedwith BVDV for 18 h and intracellular production of IFN-� by CD4+ (A) and CD8+ Tcells (B) were analyzed by flow cytometry. Statistical analysis was done by one-way ANOVA and the differences between the treatments were compared by Dunet’stest. Significant differences are indicated: *(P ≤ 0.05) and **(P ≤ 0.01), with the groupi

QnvfrQa

sararQa

sornmiasi

[

[

[

[

[

[

[

[

[

[

[20] W.R. Heath, F.R. Carbone, Cytotoxic T lymphocyte activation by cross-priming,

mmunized with no adjuvant.

B-90-adjuvanted vaccine when compared to the group immu-ized with unadjuvanted vaccine (Fig. 6). The QB-90-adjuvantedaccine promoted a 14-fold increase in IFN-�+ CD4+ T cells and a 25-old increase in IFN-�+ CD8+ T cells when compared with mice thateceived unadjuvanted vaccine. These findings indicate that theB-90-adjuvanted vaccine is capable of inducing antigen-specificctivation of CD8+ and CD4+ T cells.

The findings reported here suggest that QB-90, in particular,eems to be an interesting alternative to the usage of Quil A®

s adjuvant for vaccines. Additional testing would unavoidablyequire evaluation of the QB-90-adjuvanted vaccine on cattle, thectual target host for BVDV prophylaxis. Considered globally, ouresults demonstrate that the BVDV vaccine adjuvanted with theB-90 fraction of Q. brasiliensis saponins induces strong humoralnd cellular immune responses in mice.

Another relevant issue is the fact that the large scale use of Q.aponaria bark saponins may compromise sustainable productionf the product. New strategies for rational and sustainable explo-ation of biological resources are necessary in order to preserveatural flora and to ensure extraction of raw materials with maxi-um saponin yields. Recently, attempts have been made aiming

n vitro clonal propagation of Q. brasiliensis. Studies with leaves

nd seedlings led to detailed knowledge of factors that affectaponin yields [24]. The “renewability” of the process of extract-ng bioactive saponins from leaves with strong adjuvant potential

[

biology and Infectious Diseases 45 (2016) 1–8 7

assumes an even more attractive role. As an interesting side effect,the availability of an alternative source of saponins such as Q.brasiliensis may decrease pressure on production of Q. saponaria andthus prove helpful to contribute to the sustainability of the latter[15,24,25].

Competing interests

None of the authors has any potential financial conflict of inter-est related to this manuscript.

Acknowledgements

This work was supported by the Programa de Desarrollo de lasCiencias Básicas (PEDECIBA) and Agencia Nacional de Innovacióne Investigación (ANII) from Uruguay and the Conselho Nacional deDesenvolvimento Científico e Tecnológico (CNPq), Financiadora deEstudos e Projetos (FINEP, Grant 01.10.0783.04) and Coordenac ãode Aperfeic oamento de Pessoal de Nível Superior (CAPES) fromBrazil to a CAPES/UDELAR Project. Work developed while PMR wason receipt of a CNPq 1A Research Grant.

References

[1] A.L. Lindberg, Bovine viral diarrhoea virus infections and its control. A review,Vet. Q. 25 (2003) 1–16.

[2] B.W. Brodersen, Bovine viral diarrhea virus infections: manifestations ofinfection and recent advances in understanding pathogenesis and control,Vet. Pathol. 51 (2014) 453–464.

[3] E.N. Meeusen, J. Walker, A. Peters, P.P. Pastoret, G. Jungersen, Current status ofveterinary vaccines, Clin. Microbiol. Rev. 20 (2007) 489–510, table of contents.

[4] D.T. O’Hagan, C.B. Fox, New generation adjuvants – from empiricism torational design, Vaccine 33 (Suppl 2) (2015) B14–B20.

[5] S.G. Reed, M.T. Orr, C.B. Fox, Key roles of adjuvants in modern vaccines, Nat.Med. 19 (2013) 1597–1608.

[6] H.X. Sun, Y. Xie, Y.P. Ye, Advances in saponin-based adjuvants, Vaccine 27(2009) 1787–1796.

[7] S.G. Sparg, M.E. Light, J. van Staden, Biological activities and distribution ofplant saponins, J. Ethnopharmacol. 94 (2004) 219–243.

[8] F. de Costa, A.C. Yendo, S.P. Cibulski, J.D. Fleck, P.M. Roehe, F.R. Spilki, et al.,Alternative inactivated poliovirus vaccines adjuvanted with Quillajabrasiliensis or Quil-A saponins are equally effective in inducing specificimmune responses, PLOS ONE 9 (2014) e105374.

[9] F. Silveira, S.P. Cibulski, A.P. Varela, J.M. Marques, A. Chabalgoity, F. de Costa,et al., Quillaja brasiliensis saponins are less toxic than Quil A and have similarproperties when used as an adjuvant for a viral antigen preparation, Vaccine29 (2011) 9177–9182.

10] J.D. Fleck, C. Kauffmann, F. Spilki, C.L. Lencina, P.M. Roehe, G. Gosmann,Adjuvant activity of Quillaja brasiliensis saponins on the immune responses tobovine herpesvirus type 1 in mice, Vaccine 24 (2006) 7129–7134.

11] R.I. Freshney, Animal Cell Culture: A Practical Approach, 2nd ed., IRL Press atOxford University Press, Oxford, England and New York, 1992.

12] F. Brown, An overview of the inactivation of FMDV and the implications whenresidual virus is present in vaccines, Dev. Biol. Stand. 75 (1991) 37–41.

13] F. Silveira, S. Rossi, C. Fernandez, G. Gosmann, E. Schenkel, F. Ferreira,Alum-type adjuvant effect of non-haemolytic saponins purified from Ilex andPassiflora spp, Phytother. Res.: PTR 25 (2011) 1783–1788.

14] A.P. de Souza, L.H. Haut, R. Silva, S.I. Ferreira, C.R. Zanetti, H.C. Ertl, et al.,Genital CD8+ T cell response to HIV-1 gag in mice immunized by mucosalroutes with a recombinant simian adenovirus, Vaccine 25 (2007) 109–116.

15] F. de Costa, A.C. Yendo, J.D. Fleck, G. Gosmann, A.G. Fett-Neto,Immunoadjuvant and anti-inflammatory plant saponins: characteristics andbiotechnological approaches towards sustainable production, Mini Rev. Med.Chem. 11 (2011) 857–880.

16] D.J. Cher, T.R. Mosmann, Two types of murine helper T cell clone. II.Delayed-type hypersensitivity is mediated by TH1 clones, J. Immunol. 138(1987) 3688–3694.

17] S. Crotty, A brief history of T cell help to B cells, Nat. Rev. Immunol. 15 (2015)185–189.

18] D.J. Marciani, New Th2 adjuvants for preventive and active immunotherapy ofneurodegenerative proteinopathies, Drug Discov. Today 19 (2014) 912–920.

19] A.O. Kamphorst, K. Araki, R. Ahmed, Beyond adjuvants: immunomodulationstrategies to enhance T cell immunity, Vaccine 33 (Suppl 2) (2015) B21–B28.

Curr. Opin. Immunol. 11 (1999) 314–318.21] D. Kagi, B. Ledermann, K. Burki, R.M. Zinkernagel, H. Hengartner,

Lymphocyte-mediated cytotoxicity in vitro and in vivo: mechanisms andsignificance, Immunol. Rev. 146 (1995) 95–115.

8 Micro

[

[

[

associated with abiotic and biotic stresses, Plant Physiol. Biochem. 66 (2013)

S.P. Cibulski et al. / Comparative Immunology,

22] T.R. Mosmann, L. Li, S. Sad, Functions of CD8 T-cell subsets secreting different

cytokine patterns, Semin. Immunol. 9 (1997) 87–92.

23] L.G. Guidotti, F.V. Chisari, Cytokine-mediated control of viral infections,Virology 273 (2000) 221–227.

24] F. de Costa, A.C. Yendo, J.D. Fleck, G. Gosmann, A.G. Fett-Neto, Accumulationof a bioactive triterpene saponin fraction of Quillaja brasiliensis leaves is

[

biology and Infectious Diseases 45 (2016) 1–8

56–62.25] A.C. Yendo, F. de Costa, G. Gosmann, A.G. Fett-Neto, Production of plant

bioactive triterpenoid saponins: elicitation strategies and target genes toimprove yields, Mol. Biotechnol. 46 (2010) 94–104.