827Vet. Res. 36 (2005) 827–838© INRA, EDP Sciences, 2005DOI: 10.1051/vetres:2005033

Original article

Immunogenicity in dogs of three recombinant antigens (TSA, LeIF and LmSTI1) potential vaccine

candidates for canine visceral leishmaniasis

Ricardo Toshio FUJIWARAa,e*, André Macedo VALEa, João Carlos FRANÇA DA SILVAa, Roberto Teodoro DA COSTAa,

Josiane da Silva QUETZa, Olindo Assis MARTINS FILHOb, Alexandre Barbosa REISc,d, Rodrigo CORRÊA OLIVEIRAc,

George Lins MACHADO-COELHOd, Lilian Lacerda BUENOe, Jeffrey Michael BETHONYc,e, Glen FRANKf, Evaldo NASCIMENTOa,

Odair GENAROa,Wilson MAYRINKa, Steven REEDg, Antonio CAMPOS-NETOh

a Laboratório de Leishmanioses e Vacinas, Departamento de Parasitologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais (UFMG), CEP 31270-901,

Belo Horizonte, MG, Brazilb Laboratório de Doença de Chagas, Centro de Pesquisas René Rachou – FIOCRUZ/MG,

CEP 30190-002, Belo Horizonte, MG, Brazilc Laboratório de Imunologia Celular e Molecular, Centro de Pesquisas René Rachou – FIOCRUZ/MG,

CEP 30190-002, Belo Horizonte, MG, Brazild Escola de Farmácia, Universidade Federal de Ouro Preto (UFOP), CEP 35400-000,

Ouro Preto, MG, Brazile The George Washington University, Washington, DC 20037, USA

f Heska Corporation, Fort Collins, CO, USAg The Infectious Disease Research Institute, Seattle, WA 98104, USA

h The Forsyth Institute, Boston, MA 02115, USA

(Received 9 September 2004; accepted 12 April 2005)

Abstract – Control of canine visceral leishmaniasis (VL) remains a difficult and serious problemmostly because there is no reliable and effective vaccine available to prevent this disease. A mixtureof three recombinant leishmanial antigens (TSA, LeIF and LmSTI1) encoded by three genes highlyconserved in the Leishmania genus have been shown to induce excellent protection against infectionin both murine and simian models of cutaneous leishmaniasis. A human clinical trial with theseantigens is currently underway. Because of the high degree of conservation, these antigens might beuseful vaccine candidates for VL as well. In the present study, using the dog model of the visceraldisease, we evaluated the immunogenicity of these three antigens formulated with two differentadjuvants, MPL-SE® and AdjuPrime®. The results were compared with a whole parasite vaccineformulated with BCG as the adjuvant. In order to investigate if sensitization with the recombinantantigens would result in recognition of the corresponding native parasite antigens upon infection,the animals were exposed for four weeks after the termination of the immunization protocol withthe recombinant antigens to a low number of L. chagasi promastigotes, an etiological agent of VL.Immune response was evaluated by quantitative ELISA in the animal sera before and after exposure

* Corresponding author: mtmrtf@gwumc.edu

Article published by EDP Sciences and available at http://www.edpsciences.org/vetres or http://dx.doi.org/10.1051/vetres:2005033

828 R.T. Fujiwara et al.

to the viable parasites. Both antigen specific IgG1 and IgG2 antibody levels were measured.Immunization of dogs with the recombinant antigens formulated in either MPL-SE® or AdjuPrime®

resulted in high antibody levels particularly to LmSTI1. In addition, this immunization although tolow levels, resulted in the development of antibody response to the whole parasite lysate. Importantly,experimental exposure with low numbers of culture forms of L. chagasi promastigotes caused a clearboost in the immune response to both the recombinant antigens and the corresponding nativemolecules. The boost response was predominantly of the IgG2 isotype in animals primed with therecombinant antigens plus MPL-SE®. In contrast, animals primed with the recombinant antigensformulated in AdjuPrime® as well as animals vaccinated with crude antigen preparation respondedwith mixed IgG1/IgG2 isotypes. These results point to the possible use of this antigen cocktailformulated with the adjuvant MPL-SE® in efficacy field trials against canine VL.

visceral leishmaniasis / dogs / vaccine / Leishmania chagasi

1. INTRODUCTION

Leishmaniasis are a complex group ofdiseases caused by several intracellular pro-tozoa of the genus Leishmania that infectmacrophages of a variety of mammalsincluding humans and dogs [13, 37]. Both,tegument (skin and external mucosa) andviscera (primarily liver, spleen, and bonemarrow) are the target organs of differentspecies of Leishmania. Visceral leishmani-asis (VL) or kala-azar is fatal without treat-ment. About 500 000 new human cases ofVL are registered annually [19]. Moreover,VL has been shown to be a serious oppor-tunistic disease associated to AIDS andother immunocompromised patients [19].VL is caused by the so-called Leishmaniadonovani complex [24]. Domesticated andwild dogs are the main reservoir of the dis-ease [16]. Because drug treatment of infecteddogs is expensive and poorly effective andbecause there is no effective vaccine for yetready use despite many trials [5, 12, 18],elimination of seropositive animals hasbeen used in some countries [16] where vis-ceral leishmaniasis is zoonotic as a meansto control the human disease.

Several laboratories have been dedicat-ing much effort to the development of a pro-phylactic anti-leishmania vaccine consti-tuted of killed organisms. Even thoughsome promising early results were achievedagainst human cutaneous leishmaniasis,these observations were not easily repro-duced [1, 2, 23, 25–27, 29, 32, 36]. Vacci-

nation of dogs with a killed preparation ofleishmania organisms mixed with the humanvaccine BCG (bacille Calmette – Guerin) asthe adjuvant results in protection againstexperimental VL in kennel dogs [28]. How-ever, field trials with this vaccine showedno protection whatsoever [14, 15, 33, 34].

Over the past years, several Leishmaniarecombinant antigens have been identifiedand demonstrated to have promising vac-cine potential to cutaneous leishmaniasis[39, 44–46]. Three highly conserved anti-gens among the Leishmania genus, TSA,LmSTI1 and LeIF, have been shown toinduce excellent protection in both the murineand non-human primate models of the humandisease [8, 41]. Moreover, vaccination withplasmid DNA encoding these antigens con-ferred protection against L. major infectionin BALB/c [9, 30, 31]. A single recombinantpolyprotein comprising the sequences of allthree recombinant proteins has been pro-duced and tested in BALB/c mice against L.major infection formulated with MPL-SE®

(Monophosphoryl Lipid A plus squalene)and the results showed excellent protectiveimmunity against the challenge with viru-lent parasites [41]. This polyprotein is cur-rently being tested in phase I/II human clin-ical trials against cutaneous leishmaniasis.

Because these three antigens are highlyconserved in the Leishmania genus it islikely that they may be useful as an anti-VLvaccine as well. In the present studies, webegan pre-clinical studies to address thispossibility. Prime/boost experiments were

Immunogenicity of LmSTI1, TSA and LeIF in dogs 829

performed in dogs primed with a mixture ofTSA, LmSTI1 and LeIF formulated withclinically approved adjuvants (MPL-SE®

and AdjuPrime®) followed by boost withintra-venous inoculation of low numbers ofviable Leishmania chagasi promastigotes(to mimic natural exposure to infection), theetiological agent of visceral leishmaniasisin humans and dogs in the New World.

2. MATERIALS AND METHODS

2.1. Animals and immunizations

Seven groups of dogs were used for theexperiments in this study. Thirty-five ani-mals, purpose-bred, parasite naïve, identi-fied by ear tattoo, were selected from a col-ony of beagles, bed and maintained underconditions designed to exclude any possi-ble contaminating Leishmania infections.The dogs were between 8 and 12 monthsold, well-fed animals under constant super-vision by a veterinarian and had all receivedtheir routine vaccinations against parvovi-rosis, distemper, adenovirosis-2, hepatitis,parainfluenza and leptospirosis (Recom-bitek®, Merial Inc., USA). All animals alsoreceived a single intranasal dose againstadenovirosis-2, parainfluenza and Borde-tella bronchiseptica (Bronch Shield III®,Fort Dodge, USA). The dogs were alsotreated with anti-helminthic drugs (EndalPlus®, Schering Plough, Brazil) and withanti-ectoparasites (Frontline®, Merial Inc.,USA), and were quarantined for approxi-mately four weeks before beginning the vac-cine trial. One group received a mixture ofTSA, LmSTI1 and LeIF (10 µg each) withAdjuPrime® (1 mg, Pierce Chemical, USA)as the adjuvant; a second group was vacci-nated with TSA, LmSTI1 and LeIF (10 µgeach) and monophosphoryl lipid A (MPL)plus squalene (MPL-SE®) (50 µg, CorixaCo., USA) [41] as an adjuvant; a third groupreceived a whole parasite vaccine (mix ofLeishmania amazonensis – IFLA/BR/1967/PH8 – and Leishmania braziliensis – MCAN/BR/1972/C348 – crude extracts) and non-

live lyophilized BCG (FAP, Rio de Janeiro,Brazil) as the adjuvant. The other four groupsreceived AdjuPrime® (1 mg), MPL-SE®

(50 µg), BCG (three decreasing doses–400 µg of BCG in the first dose, 300 µg inthe second and 200 µg in the last dose) aloneor 1 mL of sterile PBS (Phosphate BufferSaline, pH = 7.2) as a placebo. All vaccineswere administered subcutaneously. All ani-mals received three doses of their respec-tive vaccines at intervals of four weeks. Thedogs were experimentally boosted intrave-nously with a low inoculum of 106 culture pro-mastigotes of Leishmania chagasi (MHOM/BR/1972/BH46) four months after the thirddose of the vaccine. The animals were fol-lowed up for 15 months and were sacrificed9 months post-exposure with viable L. cha-gasi. The promastigotes used were obtainedfrom 14 day cultures in NNN/LIT mediumof macerated spleens from infected ham-sters. The sacrifice was carried out with anoverdose of barbiturates (Thionembutal®,Abbot, São Paulo, Brazil). All animalsincluded in this investigation were treatedfollowing the guidelines for animal exper-imentation of the USA National Institute ofHealth in order to keep animal suffering toa minimum. This work was approved by theEthical Committee of Animal Research ofthe Federal University of Minas Gerais,Belo Horizonte, Brazil (Protocol No. 008/02).

2.2. Immunological evaluation

Peripheral blood samples from the jug-ular vein of the animals were taken beforethe immunizations (Pre-bleed), after eachdose of immunization and every month afterexposure to viable L. chagasi. Parasite spe-cific antibodies were determined by con-ventional enzyme-linked immunosorbentassay (ELISA) and by indirect fluorescentassay (IFA). For ELISA the recombinantantigens TSA, LeIF, LmSTI1, rK26 [3] andrK39 [6] and a soluble lysate of L. chagasi(SLcA) were used. The antigens werecoated onto 96-well microplates (Maxi-SorpTM, Nalge Nunc Intl., USA) at a con-centration of 0.5 µg/well for recombinant

830 R.T. Fujiwara et al.

antigens and 10 µg/well for SLcA. The serawere added at a dilution of 1:80 followed bywashes and addition of peroxidase conju-gated goat anti-dog IgG1 or sheep anti-dogIgG and IgG2 (Bethyl Laboratories Inc.,Montgomery, TX, USA). The wells werethen washed and substrate and chromogen(O-Phenylenediamine, Sigma-Aldrich Co.,USA) were added and the absorbance wasread on an automatic ELISA microplatereader (Multiskan® MCC 340, Labsystems,Helsinki, Finland) at 492 nm. The conju-gate anti-IgG1 was used at a dilution of1:1 000 and the conjugates anti-IgG andIgG2 were used at 1:8 000 and 1:16 000dilutions, respectively. IFA was carried outas described [7] using promastigotes of Leish-mania amazonensis (MHOM/BR/1960/BH6) maintained by weekly passages inLIT medium. Serial dilutions of the serawere examined until a visual end-point wasreached. IFA titers of 1:40 or higher wereconsidered positive as established by TheDepartment of Health of Brazil.

3. RESULTS

3.1. Antibody response to recombinant antigens in vaccinated and control dogs

Serum samples were obtained at the daysindicated and they were assayed by ELISAusing the recombinant antigens used in thevaccines TSA, LeIF and LmSTI1, and to thesoluble L. chagasi antigen (SLcA). Theresults revealed that only dogs immunizedwith the recombinant antigens formulatedeither with MPL-SE® or AdjuPrime® pro-duced specific IgG against each of the indi-vidual components of the vaccines (Figs. 1Aand 1B). No specific IgG against the TSA,LeIF and LmSTI1 was found in the serafrom control animals immunized only withadjuvants (data not shown). The absorb-ance value at 492 nm to LeIF was found tobe the lowest of the three components. Anti-body titers, particularly to LmSTI1 andTSA, reached maximal values after the third

immunization, remained at a plateau forapproximately four weeks and then declinedsteadily for the next six weeks. Moreover,immunization with the recombinant anti-gens, regardless of the adjuvant used,resulted in the production of low levels ofanti-parasite antibodies (Figs. 1C and 1D).

3.2. Booster response to recombinant and native antigens after exposure of dogs to viable L. chagasi

One important requirement of successfulvaccines to infectious diseases is that expo-sure to the disease etiological agent booststhe host memory cells generated by the vac-cination leading to an effective amplifica-tion of effector immune cells. For most vac-cines composed of complex mixtures of themicrobe’s native antigens (e.g. viable atten-uated vaccines and killed organisms) suchrecognition is in general achieved becauseof the very complexity of the vaccine com-ponents. However, the memory cells gen-erated by immunization with highly puri-fied recombinant antigens may not necessarilybe re-stimulated in vivo after exposure tothe infectious agent because the correspond-ing native antigens may not be readily avail-able or produced by the microbe in thein vivo milieu.

To more precisely evaluate if challengewith viable leishmania organisms wouldboost the immune response of dogs primedwith the recombinant antigens it was impor-tant to determine in advance the kinetics ofsensitization caused by the exposure of theanimals with a low number of viable pro-mastigote parasites. The dogs were inocu-lated intravenously with 106 culture pro-mastigote of L. chagasi and their immuneresponse to parasite specific antigens wasmonitored over a period of 9 months.Whole parasite lysate antigen as well asK26 and K39 antigens, which are acceptedmarkers of parasite replication in vivo, wereused to kinetically follow sensitization. Thesera of dogs were obtained monthly afterparasite exposure and the antibody responseto these antigens was measured by ELISA.

Immunogenicity of LmSTI1, TSA and LeIF in dogs 831

The results are expressed in Figure 2 andindicate that this exposure of dogs to viableL. chagasi parasites requires at least threeto four months in order to generate a detect-able antibody response to crude parasitelysate as well as to the antigens K26 andK39.

Therefore, in order to ascertain that vac-cination with the recombinant antigensLmSTI1, TSA, and LeIF resulted in mem-ory cells that would be boosted in vivo byexposure of the animals to viable organ-isms, dogs previously sensitized with therecombinant antigens were inoculated intra-venously with 106 culture L. chagasi pro-mastigotes and the immune responses to

both recombinant antigens and parasitelysate was kinetically evaluated by ELISA.The priming of the dogs with the recom-binant antigens formulated with MPL-SE®

generates a strong immunological memoryto both recombinant antigens (Fig. 3) andparasite lysate (Fig. 4) that is clearly boostedby exposure to viable parasites. These resultscorroborate the findings from the IFA tech-nique which shows that the dogs immunizedwith recombinant antigens formulated withMPL-SE® rather than with AdjuPrime® dem-onstrated higher levels of parasite specificantibodies (Tab. I). Specific low titers forIFA were found positive in control groupsat 270 days after boost (data not shown).High levels of specific anti-recombinant

Figure 1. Immune response to recombinant antigens (LmSTI1, TSA and LeIF) and L. chagasi lysateafter immunization with recombinant antigens formulated with MPL-SE® (A and C) or AdjuPrime®

(B and D). The results are expressed as the mean average of absorbance in sera from each group(n = 5). The y-axis represents the ELISA absorbance values at 492 nm of the sera samples diluted1:80.

832 R.T. Fujiwara et al.

antigen antibodies were detected approxi-mately three months previously to the appear-ance of the antibody to the parasite anti-genic markers K26 and K29 and L. chagasilysate in dogs not primed with the recom-binant antigens (Fig. 2). In contrast, theboosting effect observed in dogs eitherprimed with the recombinant antigens for-mulated with AdjuPrime® or previouslyvaccinated with the whole parasite vaccinewas much less evident.

3.3. The phenotype of the humoral immune response to native parasite antigens generated in dogs immunized with recombinant antigens formulated in MPL-SE® and AdjuPrime®

Similarly to resistance to CL, severalevidences suggest that resistance to VL isalso correlated with the emergence of par-asite specific Th1 response and that theTh2 response is non-protective. Thereforeit became important to evaluate the anti-parasite phenotype (Th1 × Th2) response

induced by vaccination of the dogs with therecombinant antigens formulated with MPL-SE® and AdjuPrime®. Serum samples fromvaccinated animals collected before andafter exposure to viable L. chagasi were

Figure 2. Kinetics of sensitization by low par-asite inoculum. The results are expressed as themean average of absorbance in sera from 40 bea-gles experimentally inoculated intravenouslywith 106 culture promastigotes of L. chagasi.The y-axis represents the ELISA absorbancevalues at 492 nm. All sera were used in a 1:80dilution. No sign of infection or symptoms ofdisease were observed until the end of the study.

Figure 3. Boost effect in the specific antibodyresponses against the recombinant antigenscaused by exposure to viable parasites (totalIgG). The boost effect is represented by the anti-IgG absorbance values at 492 nm for LmSTI1,TSA and LeIF in dogs immunized with thoserecombinant antigens formulated with MPL-SE® (A) or AdjuPrime® (B) and experimentallyinoculated with 106 culture promastigotes of L.chagasi, one month after the immunizations. Allsera were used at a 1:80 dilution.

Immunogenicity of LmSTI1, TSA and LeIF in dogs 833

used in an ELISA format specially designedto detect both IgG1 and IgG2 anti-parasiteantibodies. In the murine model, IgG1 andIgG2 isotypes of immunoglobulins are

important surrogates of Th2 and Th1 phe-notypes of immune responses respectivelyand the ratio of IgG2/IgG1 during theimmune response to a particular antigen hasbeen used as a faithful readout of the phe-notype (Th1/Th2) response that is gener-ated to that antigen. Figure 5 shows theresults and clearly indicates that the ratioIgG2/IgG1 anti-parasite antibodies wasapproximately 40 times that in dogs immu-nized with the recombinant antigens formu-lated in MPL-SE®. In contrast, vaccinationof dogs either with the recombinant anti-gens formulated in AdjuPrime® or withwhole parasite vaccine generated low ratios(≤ 1–6). The dogs from the placebo, Adju-Prime® alone and MPL-SE® alone exposedto viable L. chagasi promastigote groupsproduced undetectable levels of both IgG1and IgG2 anti-parasite antibodies.

4. DISCUSSION

Extensive vaccination trials in Braziland Ecuador against cutaneous leishmani-asis in humans have demonstrated that acocktail of five killed Leishmania stocks ora single strain of L. amazonensis inducesprotection from natural infection [1, 13, 23,32]. However, to date, the protection effi-cacy of these or other vaccines in dogs

Figure 4. The boost effect caused by infectionin animals immunized with recombinant anti-gens formulated either with MPL-SE® or Adju-Prime® as well as animals immunized with awhole parasite vaccine formulated with BCG.The boost effect is represented by anti-IgGabsorbance values for soluble L. chagasi anti-gen at 492 nm in dogs immunized with recom-binant antigens (LmSTI1/TSA/LeIF) formu-lated with MPL-SE® or AdjuPrime®, a wholeparasite vaccine (mix of L. amazonensis and L.braziliensis) or PBS pH = 7.2 (Placebo). Theanimals from all groups (n = 5, each) were inoc-ulated with 106 culture promastigotes of L. cha-gasi. The sera were used at a 1:80 dilution.

Table I. Antibody titers obtained by IFAa.

Groups Animal 0 days after boost 90 days after boost 180 days after boost 270 days after boost

MPL-SE® + recombinant antigens

6 < 1:40 < 1:40 1:160 1:809 < 1:40 < 1:40 1:2560 1:1024015 < 1:40 < 1:40 1:80 1:16019 < 1:40 1:40 1:10240 1:2048024 < 1:40 < 1:40 < 1:40 < 1:40

AdjuPrime® + recombinant antigens

4 < 1:40 1:80 1:160 1:16033 < 1:40 < 1:40 < 1:40 1:4036 < 1:40 < 1:40 < 1:40 1:4038 < 1:40 < 1:40 < 1:40 1:4041 < 1:40 1:80 1:640 1:1280

a Positive titers (≥ 1:40) are presented in bold.

834 R.T. Fujiwara et al.

against visceral leishmaniasis is highly con-troversial [15, 28, 33, 34]. Therefore severalefforts have been dedicated to the develop-ment of such vaccines.

Here, we began the evaluation of a mix-ture of the recombinant antigens TSA,LmSTI1 and LeIF as candidate vaccine forVL. These antigens have been successfullyshown to induce an excellent protectionagainst cutaneous leishmaniasis in the murineand non-human primate models [9, 30, 41]and currently are been tested in Phase I/IIhuman vaccine clinical trials. Because theseantigens are highly conserved among theLeishmania species and are expressed inboth the amastigote and the promastigoteforms of the parasites [39, 40, 46], theycould be useful as a component of a pan-Leishmania vaccine [20]. Despite the pos-sibility that a single antigen could by itself

induce good protection, we chose to test amixture of these three proteins because acocktail of several antigens is conceivablya better vaccine for both prophylactic andtherapeutic applications because a vaccinecontaining a broader range of protectiveepitopes is unlikely to suffer from MHCrelated unresponsiveness in a heterogene-ous population. Moreover, the use of sev-eral antigens can potentially decrease theeffects of the selective pressure on the par-asite to modify multiple genes [41].

The primary aim of the current studieswas the evaluation of the immunogenicityin dogs of the recombinant antigens formu-lated in two commercially available adju-vants. These studies were primarily concen-trated in determining two key immunologicalaspects of an anti-leishmania vaccine: first,to investigate if the recombinant antigenscould prime an immune response that wasboosted by viable L. chagasi. Second, toevaluate the phenotype (Th1 × Th2) of theimmune response that is generated by thetwo vaccine formulations. These two aspectsof the immune response are critical because,as for any effective anti-microbial vaccine,the immune response needs to be boostedby infection or exposure to the infectiousagent. In view of the fact that the currentstudies are testing recombinant antigens,this crucial pre-requisite in vaccine devel-opment was imperative to be investigatedbecause an immune response to recom-binant antigens not necessarily recognizesthe corresponding native antigens. Moreo-ver, since immunity to VL is primarilymediated by Th1 cells [17, 35], the evalua-tion of the Th1/Th2 paradigm induced by avaccine candidate is also a critical elementto be investigated. The current studies werenot aimed at investigating any protectiveeffect of the two vaccine formulationsbecause protection studies against VL inkenneled dogs have been a controversialissue mostly because the results obtainedfrom this kind of experiment are not neces-sarily translated into the outcome of vac-cine trials with unconfined dogs [15, 28].

Figure 5. IgG2/IgG1 ratio as indicative of Th1/Th2 immune response after vaccination withrecombinant antigens formulated with MPL-SE® or AdjuPrime® as well as vaccination witha whole parasite vaccine formulated with BCG.The results are expressed as the ratio of anti-dogIgG1 and anti-dog IgG2 horseradish-peroxidaseconjugated absorbance values at 492 nm of eachdog serum (1:80) from animals immunized withrecombinant antigens (LmSTI1/TSA/LeIF) for-mulated with MPL-SE® or AdjuPrime® orimmunized with a whole parasite vaccine (mixof L. amazonensis and L. braziliensis). All ani-mals were experimentally infected with a lowinoculum (106 promastigotes) of viable para-sites of L. chagasi. Goat anti-IgG1 and anti-IgG2 heavy chain specific antibodies were usedat dilutions 1:1 000 and 1:16 000, respectively.

Immunogenicity of LmSTI1, TSA and LeIF in dogs 835

The choice of the adjuvants tested in thecurrent study was based on reports of the lit-erature indicating that they have been pre-viously shown to work in anti-leishmaniavaccine experiments and that they werecommercially available and licensed to beused in humans and/or animals. Both Adju-Prime® and MPL-SE® fulfill these criteria[10, 21, 22, 42, 43]. Immunogenicity of therecombinant antigens was measured prima-rily by ELISA specifically designed to detectcanine antibodies of IgG isotypes specificfor the recombinant antigens. Unfortunately,a direct evaluation of the T cell response toinvestigate the pattern of cytokine produc-tion upon stimulation with antigens couldnot be performed because canine reagentsnecessary for these assays were not com-mercially available. However, because IgG1and IgG2 responses are strictly T celldependent we used them as readouts to eval-uate the overall immunogenicity of therecombinant antigens in dogs. In addition,for humans and mice, IgG1 and IgG2 sub-types have been traditionally used as surro-gates of the Th2 and Th1 phenotypes ofimmune responses respectively. For dogs,although the association between IgG sub-types and the immune response phenotypewere not demonstrated yet, experimentalevidences indicate that this associationshould occur [4, 11, 38].

After the immunizations, the anti-recom-binant antigens antibody response wasreadily detected by both LmSTI1 and TSAbeing stronger in animals immunized withthe antigens formulated in MPL-SE® thanAdjuPrime®. In contrast the immune responseto LeIF, even after three immunizations wasonly borderline detected in animals immu-nized with the antigen formulated in MPL-SE® and undetected in animals immunizedwith the antigens formulated in Adju-Prime®. It is noteworthy that the antibodyresponse to LmSTI1 was much strongerthan to the responses to TSA and LeIF,regardless of the adjuvant. These resultswere consistent with previous observationsthat indicate that LmSTI1 is a stronger

immunogen than TSA and LeIF in bothmice and monkeys using MPL-SE® orAlum plus IL-12 as adjuvants [8, 41]. Moreimportantly though was the observationthat the immune response generated afterimmunization of the dogs with the recom-binant antigens (regardless of the adju-vants) although at low levels reacted withthe parasite lysate and was clearly boostedby exposure of the dogs with viable L. cha-gasi. This recognition is crucial in vaccinedevelopment particularly when recombinantproteins constitute the immunizing antigenof a vaccine. It is not rare that the immuneresponse to recombinant antigens, pro-duced for example in E. coli, does not rec-ognize the corresponding native moleculeproduced by the infectious agent. Differentpos-translation modifications can lead toaltered conformation of the recombinantmolecule. This modification may result inthe recognition of epitopes in the recom-binant antigens that are different from theones that are normally recognized in theoriginal native molecule. In the presentstudy no such restriction occurred becauseviable leishmanial organisms readily boostedthe immune response to the recombinantantigens. Therefore these results fulfill thisimportant requisite to validate a recom-binant antigen as a vaccine candidate.

The second pre-requisite in vaccine devel-opment against leishmaniasis i.e. the induc-tion of a predominantly Th1 response mayalso have been fulfilled with the adjuvantMPL-SE®. By using IgG2 and IgG1 sub-types of IgG response as surrogates of theTh1 and Th2 helper phenotype of immuneresponses respectively, it was evident thatimmunization with the recombinant anti-gens formulated with MPL-SE® resulted inan immune response preponderantly of theIgG2 sub-type (IgG2/IgG1 ratio ≥ 40). Thiswould likely point to a dominant Th1 responseto the recombinant antigens. Indeed, thisbias towards inducing a Th1 response hasalso been observed in mice immunized withLmSTI1 mixed with MPL-SE® [41]. Incontrast, immunization of the dogs with the

836 R.T. Fujiwara et al.

recombinant antigens formulated with Adju-Prime®, or vaccination with a whole para-site vaccine result in higher IgG1 produc-tion (IgG2/IgG1 of ≤ 1–6), which suggestsa Th2 response or at best, points to a mixedTh1/Th2 response.

In conclusion, given the facts that therecombinant antigens TSA/LmSTI1/LeIFhave already been shown to be highly pro-tective against CL in two different animalmodels of the disease and that the currentwork demonstrates that immunization of dogswith these antigens formulated with theadjuvant MPL-SE® induces almost exclu-sively a Th1 response (IgG2/IgG1 ≥ 40), itis reasonable to assume that a combinationof TSA/LmSTI1/LeIF with the adjuvantMPL-SE® constitutes a highly attractivevaccine formulation to be used in field trialsagainst canine VL.

REFERENCES

[1] Armijos R.X., Weigel M.M., Aviles H.,Maldonado R., Racines J., Field trial of a vac-cine against New World cutaneous leishma-niasis an at-risk child population: safety,immunogenicity, and efficacy during the first12 months of follow-up, J. Infect. Dis. 177(1998) 1352–1357.

[2] Armijos R.X., Weigel M.M., Calvopina M.,Hidalgo A., Cevallos W., Correa J., Safety,immunogenicity, and efficacy of an auto-claved Leishmania amazonensis vaccine plusBCG adjuvant against New World cutaneousleishmaniasis, Vaccine 22 (2004) 1320–1326.

[3] Bhatia A., Daifalla N.S., Jen S., Badaro R.,Reed S.G., Skeiky Y.A., Cloning, characteri-zation and serological evaluation of K9 andK26: two related hydrophilic antigens ofLeishmania chagasi, Mol. Biochem. Parasi-tol. 102 (1999) 249–261.

[4] Boag P.R., Parsons J.C., Presidente P.J.A.,Spithill T.W., Sexton J.L., Characterisation ofhumoral immune responses in dogs vacci-nated with irradiated Ancylostoma caninum,Vet. Immunol. Immunopathol. 92 (2003)87–94.

[5] Borja-Cabrera G.P., Correia Pontes N.N., DaSilva V.O., Paraguai de Souza E., SantosW.R., Gomes E.M., Luz K.G., Palatnik M.,Palatnik de Sousa C.B., Long lasting protec-tion against canine kala-azar using the FML-

QuilA saponin vaccine in an endemic area ofBrazil (Sao Goncalo do Amarante, RN), Vac-cine 20 (2002) 3277–3284.

[6] Burns J.M. Jr., Shreffler W.G., Benson D.R.,Ghalib H.W., Badaro R., Reed S.G., Molecu-lar characterization of a kinesin-related anti-gen of Leishmania chagasi that detects spe-cific antibody in African and Americanvisceral leishmaniasis, Proc. Natl. Acad. Sci.USA 90 (1993) 775–779.

[7] Camargo M.E., Fluorescent antibody test forthe serodiagnosis of american trypanosomia-sis. Technical modification employing pre-served cultures forms of Trypanosoma cruziin a slide test, Rev. Inst. Med. Trop. São Paulo8 (1966) 227–234.

[8] Campos-Neto A., Porrozzi R., Greeson K.,Coler R.N., Webb J.R., Skeiky Y.A., ReedS.G., Grimaldi G. Jr., Protection against cuta-neous leishmaniasis induced by recombinantantigens in murine and non-human primatemodels of the human disease, Infect. Immun.69 (2001) 4103–4108.

[9] Campos-Neto A., Webb J.R., Greeson K.,Coler R.N., Skeiky Y.A.W., Reed S.G., Vac-cination with plasmid DNA encoding TSA/LmSTI1 leishmanial fusion proteins confersprotection against Leishmania major infec-tion in susceptible BALB/c mice, Infect.Immun. 70 (2002) 2828–2836.

[10] Cook J.A., Holbrook T.W., Parker B.W., Vis-ceral leishmaniasis in mice: protective effectof glucan, J. Reticuloendothel. Soc. 27 (1980)567–573.

[11] Chabalgoity J.A., Moreno M., Carol H.,Dougan G., Hormaeche C.E., Salmonellatyphimurium as a basis for a live oral Echino-coccus granulosus vaccine, Vaccine 19 (2001)460–469.

[12] Da Silva V.O., Borja-Cabrera G.P., CorreiaPontes N.N., De Souza E.P., Luz K.G., PalatnikM., Palatnik de Sousa C.B., A phase III trialof efficacy of the FML-vaccine against caninekala-azar in an endemic area of Brazil (SaoGoncalo do Amaranto, RN), Vaccine 19(2000) 1082–1092.

[13] De Luca P.M., Mayrink W., Alves C.R.,Coutinho S.G., Oliveira M.P., Bertho A.L.,Toledo V.P., Costa C.A., Genaro O.,Mendonca S.C., Evaluation of the stabilityand immunogenicity of autoclaved and non-autoclaved preparations of a vaccine againstAmerican tegumentary leishmaniasis, Vac-cine 17 (1999) 1179–1185.

[14] Fallah E., Mohebali M., Zareh N.M., Prepa-ration and evaluation of Leishmania vaccinesin Iran for the control of kala-azar, Acta Par-asitol. (Warsawa) 45 (2000) 202–203.

Immunogenicity of LmSTI1, TSA and LeIF in dogs 837

[15] Genaro O., Pinto J.A., Costa C.A., Franca-Silva J.C., Silva J.C., Sanguinetti L.S.R.,Vieira E.P., Toledo V.P.C.P., Mayrink W.,Phase III randomized double blind clinicaltrial of the efficacy of a vaccine against caninevisceral leishmaniasis in urban area of MontesClaros, MG, Brazil, Mem. Inst. Oswaldo Cruz91 (1996) 166.

[16] Gavgani A.S.M., Mohite H., Edrissian G.H.,Mohebali M., Davies C.R., Domestic dogownership in Iran is a risk factor for humaninfection with Leishmania infantum, Am. J.Trop. Med. Hyg. 65 (2002) 511–515.

[17] Goto H., Lindoso J.A., Immunity and immu-nosuppression in experimental visceral leish-maniasis, Braz. J. Med. Biol. Res. 37 (2004)615–623.

[18] Gradoni L., An update on antileishmanial vac-cine candidates and prospects for a canineLeishmania vaccine, Vet. Parasitol. 100(2001) 87–103.

[19] Guerin P.J., Olliaro P., Sundar S., Boelaert M.,Croft S.L., Desjeux P., Wasunna M.K.,Bryceson A.D.M., Visceral leishmaniasis: cur-rent status of control, diagnosis, and treatment,and a proposed research and developmentagenda, Lancet Infect. Dis. 2 (2002) 494–501.

[20] Handman E., Leishmaniasis: current status ofvaccine development, Clin. Microbiol. Rev.14 (2001) 229–243.

[21] Heppner D.G., Gordon D.M., Gross M.,Wellde B., Leitner W., Kzych U., SchneiderI., Wirtz R.A., Richards R.L., Trofa A., HallT., Sadoff J.C., Boerger P., Alving C.R.,Sylvester D.R., Porter T.G., Ballou W.R.,Safety, immunogenicity and efficacy of Plas-modium falciparum repeatless circumsporo-zoite protein vaccine encapsulated in lipo-somes, J. Infect. Dis. 174 (1996) 361–366.

[22] Holbrook T.W., Cook J.A., Parker B.W.,Immunization against Leishmania donovani:glucan as an adjuvant with killed promastig-otes, Am. J. Trop. Med. Hyg. 30 (1981) 762–768.

[23] Marzochi K.B., Marzochi M.A., Silva A.F.,Grativol N., Duarte R., Confort E.M., ModabberF., Phase 1 study of an inactivated vaccineagainst American tegumentary leishmaniasisin normal volunteers in Brazil, Mem. Inst.Oswaldo Cruz 93 (1998) 205–212.

[24] Mauricio I.L., Howard M.K., Stothard J.R.,Miles M.A., Genomic diversity in the Leish-mania donovani complex, Parasitology 119(1999) 237–246.

[25] Mayrink W., Costa C.A., Magalhaes P.A.,Melo M.N., Dias M., Lima A.O., MichalickM.S., Williams P., A field trial of a vaccine

against American dermal leishmaniasis,Trans. R. Soc. Trop. Med. Hyg. 73 (1979)385–387.

[26] Mayrink W., Williams P., Costa C.A., DiasM., An experimental vaccine against Ameri-can dermal leishmaniasis: experience in thestate of Espirito Santo, Brazil, Ann. Trop.Med. Parasitol. 79 (1985) 259–269.

[27] Mayrink W., Antunes C.M.G., Costa C.A.,Melo M.N., Dias M., Michalick M.S.,Magalhaes P.A., De Oliveira Lima A.,Williams P., Further trials of a vaccine againstAmerican cutaneous leishmaniasis, Trans. R.Soc. Trop. Med. Hyg. 80 (1986) 1001.

[28] Mayrink W., Genaro O., Franca-Silva J.C., daCosta R.T., Tafuri W.F., Toledo V.P., da SilvaA.R., Reis A.B., Williams P., da Costa P.W.,Phase I and II open clinical trials of a vaccineagainst Leishmania chagasi infections indogs, Mem. Inst. Oswaldo Cruz 91 (1996)695–697.

[29] Mayrink W., Santos G.C., Toledo V.P.C.P.,Guimaraes T.M., Machado-Coelho G.L.,Genaro O., da Costa C.A., Vaccination ofC57BL/10 mice against cutaneous leishmani-asis using killed promastigotes of differentstrains and species of Leishmania, Rev. Soc.Bras. Med. Trop. 35 (2002) 125–132.

[30] Mendez S., Gurunathan S., Kamhawi S.,Belkaid Y., Moga M.A., Skeiky Y.A.,Campos-Neto A., Reed S., Seder R.A., SacksD., The potency and durability of DNA- andprotein-based vaccines against Leishmaniamajor evaluated using low-dose, intradermalchallenge, J. Immunol. 166 (2001) 5122–5128.

[31] Mendez S., Belkaid Y., Seder R.A., Sacks D.,Optimization of DNA vaccination againstcutaneous leishmaniasis, Vaccine 20 (2002)3702–3708.

[32] Modabber F., Vaccines against leishmaniasis,Ann. Trop. Med. Parasitol. 89 (1995) 83–88.

[33] Mohebali M., Fallah E., Jamshidi S., HajjaranH., Vaccine trial against canine visceral leish-maniasis in the Islamic Republic of Iran, Rev.Santé Méditerranée Orientale 4 (1998) 234–238.

[34] Mohebali M., Fallah E., Mobedi I., HajjaranH., Field trial of autoclaved Leishmania vac-cines for control of canine visceral leishmani-asis in Meshkin-Shahr, Iran, Arch. Razi Inst.50 (1999) 87–92.

[35] Murray H.W., Hariprashad J., Coffman R.L.,Behavior of visceral Leishmania donovani inan experimentally induced T helper cell 2(Th2)-associated response model, J. Exp.Med. 185 (1997) 867–874.

838 R.T. Fujiwara et al.

To access this journal online: www.edpsciences.org

[36] Nascimento E., Mayrink W., Costa C.A.,Michalick M.S., Melo M.N., Barros G.C.,Dias M., Antunes C.M., Lima M.S., TaboadaD.C., Liu T.Y., Vaccination of humansagainst cutaneous leishmaniasis: cellular andhumoral immune responses, Infect. Immun.58 (1990) 2198–2203.

[37] Nieto C.G., García-Alonso M., Requena J.M.,Mirón C., Soto M., Alonso C., Navarrete I.,Analysis of the humoral immune responseagainst total and recombinant antigens ofLeishmania infantum: correlation with dis-ease progression in canine experimental leish-maniasis, Vet. Immunol. Immunopathol. 67(1999) 117–130.

[38] Oliveira-Mendes C., Paraguai de Souza E.,Borja-Cabrera G.P., Maria Melo Batista L.,Aparecida dos Santos M., Ellner Parra L.,Menz I., Palatnik M., Palatnik de Sousa C.B.,IgG1/IgG2 antibody dichotomy in sera of vac-cinated or naturally infected dogs with vis-ceral leishmaniosis, Vaccine 21 (2003) 2589–2597.

[39] Skeiky Y.A.W., Guderian J.A., Benson D.R.,Bacelar O., Carvalho E.M., Kubin M., BadaroR., Trinchieri G., Reed S.G., A recombinantLeishmania antigen that stimulates humanperipheral blood mononuclear cells to expressa Th1-type cytokine profile and to produceinterleukin 12, J. Exp. Med. 181 (1995) 1527–1537.

[40] Skeiky Y.A.W., Kennedy M., Kaufman D.,Borges M.M., Guderian J.A., Scholler J.K.,Ovendale P.J., Picha K.S., Morrissey P.J.,Grabstein K.H., Campos-Neto A., Reed S.G.,LeIF: a recombinant Leishmania protein thatinduces an IL-12-mediated Th1 cytokine pro-file, J. Immunol. 161 (1998) 6171–6179.

[41] Skeiky Y.A.W., Coler R.N., Brannon M.,Stromberg E., Greeson K., Crane R.T., WebbJ.R., Campos-Neto A., Reed S.G., Protectiveefficacy of a tandemly linked, multi-subunitrecombinant leishmanial vaccine (Leish-111f) formulated in MPL® adjuvant, Vaccine20 (2002) 3292–3303.

[42] Stoute J.A., Kester K.E., Krzych U., WelldeB.T., Hall T., White K., Glenn G., OckenhouseC.F., Garcon N., Schwenk R., Lanar D.E., SunP., Momin P., Wirtz R.A., Golenda C., SlaouiM., Wortmann G., Holland C., Dowler M.,Cohen J., Ballou W.R., Long-term efficacyand immune responses following immuniza-tion with the RTS, S malaria vaccine, J. Infect.Dis. 178 (1998) 1139–1144.

[43] Ulrich J.T., Myers K.R., Monophosphoryllipid A as an adjuvant. Past experiences andnew directions, in: Powell M.F., NewmanM.J. (Eds.), Vaccine design: the subunit andadjuvant approach, Plenum Press, New York,1995, p. 495.

[44] Webb J.R., Kaufman D., Campos-Neto A.,Reed S.G., Molecular cloning of a novel pro-tein antigen of Leishmania major that elicitsa potent immune response in experimentalmurine leishmaniasis, J. Immunol. 157 (1996)5034–5041.

[45] Webb J.R., Campos-Neto A., Skeiky Y.A.W.,Reed S.G., Molecular characterization of theheat-inducible LmSTI1 protein of Leishmaniamajor, Mol. Biochem. Parasitol. 89 (1997)179–193.

[46] Webb J.R., Campos-Neto A., Ovendale P.,Martin T.I., Stromberg E.J., Badaro R., ReedS.G., Human and murine immune responsesto a novel Leishmania major recombinant pro-tein encoded by members of a multicopy genefamily, Infect. Immun. 66 (1998) 3279–3289.