TEXT

Top Abstract Text References

Leishmaniasis has recently been declared one of the world's most serious parasitic diseases by the World Health Organization (http://www.who.org). Approximately 350 million people are currently at risk of acquiring the various forms of the disease, and the annual incidence of new cases is about 2 million (1.5 million cases of cutaneous leishmaniasis and 0.5 million cases of visceral leishmaniasis). Dogs also develop leishmaniasis and may serve as an important domestic reservoir for the human diseases. No precise information on global mortality rates is available; however, the World Health Organization reports 100,000 deaths from visceral leishmaniasis over the past 5 years in a population of less than 1 million people in the western Upper Nile region of southern Sudan. These alarming figures are believed to be due primarily to the impracticality of controlling the vectors that transmit the diseases and to the lack of an efficacious vaccine. However, protective immunity has been achieved in some individuals after cure of the active disease or after vaccination with viable leishmanial organisms as well as with crude antigenic preparations of leishmanial organisms (5, 10, 20, 23). Therefore, the development of an efficacious anti-Leishmania subunit vaccine is, in principle, feasible. In recent years, several recombinant leishmanial antigens have been identified and tested as vaccine candidates (1, 18, 19). However, there have been no follow-up reports on the efficacy or utility of these antigens in humans or in other animal models beyond the murine model.

In this communication, we present two recombinant leishmanial antigens that are promising vaccine candidates against human leishmaniasis. This assertion is based on the protection that these antigens induced in both murine and nonhuman primate models of cutaneous leishmaniasis. The antigens LmSTI1 and TSA, which we characterized previously (28, 29), were tested as vaccine candidates because they elicit primarily a Th1-type response in BALB/c mice infected with Leishmania major. In the murine model, the Th1 response phenotype is associated with protection and the Th2 response phenotype is associated with susceptibility or aggravation of the disease (14, 17, 22). Considering that L. major-infected BALB/c mice mount predominantly a Th2 response to most of the parasite antigens, it became interesting to test the protective potential of LmSTI1 and TSA because, even under this strongly biased Th2 response, during infection these antigens stimulate preferentially immune responses of the protective phenotype. The rationale behind this hypothesis was that by stimulating the immune system with a vaccine that induces strongly biased antileishmanial Th1 responses in the absence of Th2 responses, protection could be achieved.

BALB/c mice (Charles River Laboratories, Wilmington, Mass.) were initially immunized with LmSTI1 and TSA mixed with interleukin 12 (IL-12) as an adjuvant, because this cytokine is an effective in vivo modulator of Th1 responses when administered mixed with several antigens (2, 8, 16, 26). Both antigens were expressed and purified as previously described (28, 29) and were virtually endotoxin free (<10 endotoxin units per mg of protein) as determined by the Limulus amoebocyte assay (BioWhittaker, Walkersville, Md.). To evaluate whether this protocol of immunization induces a Th1 response to both TSA and LmSTI1, mice were immunized in the footpad with 10 µg of the individual antigens or with a mixture of them, in the presence or absence of 1 µg of IL-12 (Genetics Institute, Cambridge, Mass.). Mice were boosted 3 weeks later with the same antigen formulations used in the primary immunization. Ten days after the second immunization, the mice were bled and sacrificed. Antibody responses to TSA and LmSTI1 were evaluated by standard enzyme-linked immunosorbent assay (ELISA). T-cell responses (antigen-induced proliferative responses and cytokine production) were measured in draining lymph node cells. This protocol of immunization confirmed previous observations indicating that a specific Th1 response in the absence of a Th2 response is induced when IL-12 is used as an adjuvant (2, 8, 16, 26). Thus, mice immunized with the antigens alone (in the absence of IL-12) developed immunoglobulin G1 (IgG1) but not IgG2a antibody responses to the immunizing antigens. In contrast, when IL-12 was used as an adjuvant, high titers of specific IgG1 and IgG2a antibody responses to both TSA and LmSTI1 were observed. Interestingly, immunization with the antigen mixture induced the same antibody titers (of both isotypes) to the individual antigens as those induced by immunization with the individual antigens (data not shown). These results suggest that this protocol of immunization induces a Th1-type response to both antigens and that no antigenic competition between TSA and LmSTI1 occurs.

To further investigate the phenotype (Th1 or Th2) of immune responses induced by these antigens, lymph node cells from immunized mice were cultured in the presence of the specific antigens and proliferative responses were measured in a 3-day assay by incorporation of [³H]thymidine. Cytokine production (gamma interferon [IFN-] and IL-4) was measured in culture supernatants by sandwich ELISA. IFN- and IL-4 capture and developing monoclonal antibodies (clones R4-6A2, XMGI.2, 11B11, and BVD6-24G2) were purchased from PharMingen (San Diego, Calif.). To increase the sensitivity of the IL-4 ELISA, 1 µg of anti-IL-4 receptor monoclonal antibody (Immunex Corp., Seattle, Wash.)/ml was added to the cultures (28). The results show that TSA and LmSTI1 induce production of high concentrations of IFN- (Fig. 1) and no IL-4 (data not shown). In addition, no antigenic competition (proliferative response or cytokine production) was observed when the mixture of antigens was used to immunize the mice. These experiments thus confirm that this protocol of immunization induces a typical Th1 response to both TSA and LmSTI1, either used as individual antigens or as a mixture. No T-cell assays were performed with lymph node cells of mice immunized with the antigens but without IL-12 (no draining lymph nodes could be found). In addition, spleen cells of these animals did not respond to stimulation with either TSA or LmSTI1 (not shown).

View larger version (24K): [in this window] [in a new window]   FIG. 1.   T-cell responses of BALB/c mice immunized with the recombinant leishmanial antigens TSA and LmSTI1 formulated with IL-12 as an adjuvant. Mice were immunized subcutaneously in the rear footpad with either 10 µg of TSA, 10 µg of LmSTI1, or a mixture of 10 µg of each of these antigens. Before immunization, the antigens were mixed with IL-12 to achieve 1 µg of this cytokine per injection. Mice were boosted 3 weeks later with the same antigenic formulation used in the primary immunization. Ten days after the boost, animals were sacrificed and lymph node (popliteal) cells were obtained and cultured for 3 days in the presence of various concentrations of either TSA or LmSTI1 or with medium alone. Proliferative responses were assayed by incorporation of [3H]thymidine. Cytokine production (IFN- and IL-4) was assayed in culture supernatants by sandwich ELISA. (A) Proliferative responses and IFN- production of mice immunized with TSA (plus IL-12) or with a mixture of TSA plus LmSTI1 (plus IL-12) in response to stimulation with TSA. (B) Proliferative responses and IFN- production of mice immunized with LmSTI1 (plus IL-12) or with a mixture of TSA plus LmSTI1 (plus IL-12) in response to stimulation with LmSTI1. No IL-4 was detected in any culture supernatants (data not shown).

For protection studies, BALB/c mice were similarly immunized subcutaneously in the left footpad twice (3 weeks apart) with either the individual antigens or both antigens, mixed with IL-12. As controls, separate groups of mice were immunized with IL-12 alone or with saline. Three weeks after the last immunization, the mice were infected in the right footpad with 10⁴ amastigote forms of L. major freshly isolated from infected BALB/c mice. Footpad swelling was then measured weekly. The results are expressed in Fig. 2 and clearly indicate that the mice immunized with either LmSTI1 or TSA mixed with IL-12 were protected against infection. In contrast, when mice were immunized with the antigens mixed with saline, no protection was observed. The level of protection induced by LmSTI1 was consistently strong. Conversely, TSA induced partial protection. Moreover, immunization of mice with both antigens plus IL-12 resulted in protection comparable to that induced by LmSTI1 mixed with IL-12. These results indicate that LmSTI1 as a single antigen induces excellent protection and may perhaps by itself constitute a vaccine against leishmaniasis. However, a cocktail composed of LmSTI1 and TSA is in theory a better vaccine, because by mixing the antigens an amplification of the parasite epitopes involved in induction of a protective anti-Leishmania immune response is achieved. This is a desirable condition, because a vaccine containing a broad range of different protective epitopes is unlikely to suffer from major histocompatibility complex-related unresponsiveness even in a heterogeneous population such as humans. In the mouse model, the actual participation of TSA in the protection induced by the cocktail of the two antigens was not ascertainable because the protection attained with LmSTI1 alone was comparable to that attained with the mixture of the two antigens. However, because immunogenicity studies revealed that no antigenic competition occurs between the two antigens and since TSA alone plus IL-12 induces protection, it is possible that TSA also participates, albeit redundantly in the BALB/c model of leishmaniasis, in the protection induced by the antigenic mixture. Therefore, inclusion of TSA should be helpful in achieving broader protective immune responses in outbred populations of animals such as dogs and humans. Finally, it is noteworthy that the protective properties of LmSTI1 and TSA are not simply a ubiquitous phenomenon consequent to the modulation of the mouse immune response to the Th1 phenotype by the adjuvant IL-12. We have tested this adjuvant with 14 different recombinant leishmanial antigens: LeIF (25); Ldp23 (9); hsp83 (24); K26 and K39 (7); IG6, 4A5, 2A10, IE6, IB11, 8G3, 4H6, and 2F11 (21); and Lmsp1a (unpublished). Protection was observed only when LmSTI1 and/or TSA was used as an antigen (data not shown). Therefore, in addition to inducing a Th1-type response, triggering of protective parasite epitopes per se is also a crucial ingredient for eliciting protection against leishmaniasis in the murine model.

View larger version (18K): [in this window] [in a new window]   FIG. 2.   Vaccination of BALB/c mice against L. major infection with the recombinant leishmanial antigens TSA and LmSTI1. Mice (five per group) were immunized subcutaneously in the left footpad twice (3 weeks apart) with 10 µg of either TSA or LmSTI1 or with a mixture of 10 µg of each of these antigens. Before immunization, the antigens were mixed with IL-12 (1 µg of this cytokine per injection). Three weeks after the last immunization, the mice were infected in the right footpad with 104 amastigote forms of L. major; footpad swelling was measured weekly thereafter.

In order to evaluate the efficacy of LmSTI1 and TSA in an animal model more relevant to humans, the nonhuman primate Macaca mulatta (rhesus monkey) was used. The monkey model of cutaneous leishmaniasis, though used to a much lesser extent than the mouse model, has been accepted as a system that more closely mirrors human immunity for vaccine development against several infectious diseases (3, 4, 6, 12, 13, 15, 27).

For these studies, adult rhesus monkeys were obtained from the Primate Research Center of the Oswaldo Cruz Foundation (Rio de Janeiro, Brazil). The Primate Research Center operates in accordance with the guidelines of the Committee on the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council, U.S. Department of Health and Human Services. The animals were laboratory-bred and -reared young adults (5 to 7 years old) of both sexes. Monkeys (six) were vaccinated twice, 1 month apart, with a vaccine preparation containing a mixture of 25 µg of LmSTI1, 25 µg of TSA, 2 µg of recombinant human IL-12 (Genetics Institute), and 200 µg of alum (Rehydragel HPA; Reheis, Inc., Berkeley Heights, N.J.) as described previously (15). The monkeys were boosted 1 month later with the antigens and alum alone (i.e., no IL-12 was included in these injections). Forty days after the last boost, the monkeys were infected in the eyelid with 10⁷ metacyclic promastigotes of L. major, and the development of lesions was monitored for the next 3 months. Lesion size (in square millimeters) was measured weekly using the average diameter of a circle approximately encompassing the lesion.

Vaccination of rhesus monkeys with heat-killed Leishmania promastigotes, using IL-12 and alum as adjuvants, has been shown previously by Kenney et al. to be safe and efficacious (15). In addition, neither IL-12 nor alum alone changed the course of the leishmanial infection in these animals. Moreover, IL-12 at a total of either 10 or 30 µg per vervet monkey did not alter the course of infection of these animals with L. major (11). Therefore, in our studies only saline-injected monkeys (six) were used as controls. Upon immunization with the recombinant antigens formulated in a mixture of IL-12 and alum, no systemic reactions in the monkeys were observed throughout the whole period of the experiment (~8 months after the first injection of the mixture). A small transient nodule developed at the site of injection and self-resolved in approximately 10 days. In the vaccination protocol used by Kenney et al., a nodule was also observed at the vaccination site. The duration of these nodules was in general longer in their studies than in our studies. These different results are apparently accounted for by the different antigen preparations used. In the former studies, whole heat-killed parasites were used (particulate antigens) and the duration and consistency of the nodules were dependent on the amount of antigen used in the vaccine. Monkeys that received 0.25 mg of antigen had nodules that were qualitatively softer and resolved more quickly than nodules in monkeys that were vaccinated with 1 mg of antigen. In contrast, in our studies the monkeys were vaccinated with soluble antigen and at much lower doses (50 µg).

To ascertain the immunogenicity of the individual recombinant antigens in the rhesus monkeys vaccinated with a mixture of LmSTI1 and TSA, sera from immunized and control animals were obtained before vaccination, at 3 weeks after the first immunization, and at 1 week after each boost. The anti-LmSTI1 and anti-TSA antibody responses (IgG isotype) were tested by ELISA using a specific horseradish peroxidase-labeled goat anti-rhesus monkey IgG antiserum (Accurate Chemical & Scientific Corporation, Westbury, N.Y.). Figure 3 illustrates the results of these experiments and indicates that although the antibody responses to both recombinant antigens reached a plateau after the second immunization, the anti-LmSTI1 antibody response was detected earlier and was consistently stronger than that observed for TSA. These results are in agreement with the antibody responses to these antigens in the mouse model and suggest a lack of competition between LmSTI1 and TSA in the monkey model as well.

View larger version (21K): [in this window] [in a new window]   FIG. 3.   Specific antibody response to the individual recombinant leishmanial proteins TSA and LmSTI1 in rhesus monkeys vaccinated with a mixture of these antigens formulated with recombinant human IL-12 and alum as adjuvants. Anti-LmSTI1 and anti-TSA antibody responses (IgG isotype) were tested by ELISA using a specific horseradish peroxidase-labeled goat anti-rhesus monkey IgG antiserum. Monkeys were immunized twice (approximately 1 month apart) with a vaccine containing, per dose, 25 µg of LmSTI1, 25 µg of TSA, 2 µg of IL-12, and 250 µg of alum in a final volume of 250 µl. The vaccine was administered subcutaneously in the deltoid area. One month after the second injection, the monkeys were boosted with a mixture containing 25 µg of each recombinant antigen suspended in 250 µl of saline-alum. All monkeys were anesthetized with ketamine before the injections. Results are from serum samples collected before immunization, at 3 weeks (w) after the first dose of vaccine, and at 1 week after each boost. All sera were diluted 1:20, and results are expressed as the optical density (OD) at 490 nm.

All nonvaccinated control monkeys developed, after the challenge with L. major, erythema and nodules of various sizes at the site of inoculation that eventually ulcerated at 2 to 3 weeks postinfection and lasted for at least 8 weeks. Lesion development peaked at 5 to 6 weeks postinfection (Fig. 4). Three weeks after challenge, lesion development was evident in all control animals and lesions were macroscopically similar to those occurring in human cutaneous leishmaniasis (Fig. 5A). Histological examination (hematoxylin and eosin) of the skin lesions at 8 weeks after infection showed a chronic infiltrate of mononuclear cells accompanied by neutrophils, apparently associated with lysis of a few parasitized macrophages and liberation of extracellular amastigotes and/or a tuberculoid-type granulomatous reaction of differentiated macrophages interspersed with more- or less-numerous lymphocytes and plasma cells (Fig. 5A). In contrast, no lesions developed in any of the vaccinated monkeys after challenge infection (Fig. 4 and 5B). Biopsy samples obtained from these animals at 8 weeks postinfection were not positive for parasites by direct microscopic evaluation. In addition, histopathological findings were characteristic of skin lesions in the scarring phase, showing a fibroblast response and nonspecific focal infiltration of mononuclear inflammatory cells in the dermis, clustered mainly around postcapillary venules of the vascular plexus (Fig. 5B). These findings illustrate that histological changes reflect host immune status in cutaneous leishmaniasis and that susceptibility to leishmanial infection can be artificially modified.

View larger version (15K): [in this window] [in a new window]   FIG. 4.   Protection of rhesus monkeys against cutaneous leishmaniasis by vaccination with the recombinant leishmanial antigens LmSTI1 and TSA formulated with recombinant human IL-12 and alum as adjuvants. Monkeys were immunized as described in the legend to Fig. 2. Control monkeys were injected with saline only. Forty days after the last immunization, control and vaccinated monkeys (six animals per group) were infected intradermally in the left upper eyelid with 107 metacyclic promastigotes of L. major suspended in 100 µl of RPMI medium. All monkeys were anesthetized with ketamine before the injections. Lesion development was inspected every 2 or 3 days, and lesion sizes, in square millimeters, were measured at 3-week intervals (bars show standard deviations).

View larger version (70K): [in this window] [in a new window]   FIG. 5.   Morphological characteristics of lesions in control and vaccinated rhesus monkeys following infection with L. major. The typical macroscopic appearance of cutaneous leishmaniasis is illustrated at 6 and 8 weeks in control monkey O19 and at 8 weeks in control monkeys N15 and O25 (A). Vaccinated monkeys did not develop any lesions, as illustrated in four monkeys at 8 weeks after challenge (B). Also shown are microscopic characteristics of the inflammatory reactions in a control animal (A) and in a vaccinated animal (B). Note the large infiltration of granulocytes associated with nonspecific inflammation in the nonvaccinated monkey as opposed to the predominant mononuclear infiltration, a hallmark of the specific repair process, associated with fibrosis in the vaccinated animal.

In order to evaluate the duration of the protection induced by the recombinant antigens, we rechallenged all monkeys with 10⁷ L. major promastigotes at approximately 4 months after the initial challenge. In the nonvaccinated control group, three animals developed lesions that were smaller and of shorter duration than those that developed after the primary challenge. Another monkey developed the same lesion pattern that developed in the first challenge, and two monkeys did not develop any lesions. In contrast, none of the monkeys in the vaccinated group developed lesions after the rechallenge (data not shown). These results clearly point to a long-lasting anti-L. major immunity induced in monkeys by the recombinant antigens LmSTI1 and TSA.

In summary, this is the first study to show that a subunit vaccine composed of two recombinant leishmanial antigens induces protection in both the mouse and monkey models of cutaneous leishmaniasis. The results clearly point to excellent protective effects of vaccination of rhesus monkeys with a combination of the recombinant antigens LmSTI1 and TSA. The possibility that protection can be achieved in this model with only one of the two antigens has not been addressed in the present studies. However, experiments are currently in progress to investigate this possibility and to test the effectiveness of LmSTI1 and TSA as vaccine candidates against visceral leishmaniasis in dogs. Despite the limitations of IL-12 as an adjuvant (as it is an expensive product), these studies are highly relevant for vaccine development against the human diseases because of the excellent protection that was achieved with purified recombinant antigens in an animal model that is evolutionarily close to humans. Moreover, we have preliminary data pointing to high protective efficacy of LmSTI1 and TSA formulated with the adjuvant MPL-SE in the mouse model. This new adjuvant is suitable for human use and apparently can replace IL-12 in the murine model. In conclusion, we present here a combination of two recombinant leishmanial antigens that has potential to be the first efficacious subunit vaccine against human leishmaniasis.