1. Background

The leishmaniases affect an estimated 12 million people worldwide, with another 350 million people at risk of infection.[1] Cutaneous leishmaniasis, the most common form of clinical disease, can cause considerable morbidity and disfigurement. Infection with Leishmania spp. occurs following intradermal inoculation of virulent metacyclic promastigotes into the mammalian host by the phlebotomine sandfly vector. The promastigotes are rapidly phagocytosed by mononuclear phagocytes where they replicate within the mammalian host.

The are a number of different etiologies of cutaneous leishmaniases that are distinct in their epidemiology, clinical features, and geographical distribution (summarized in table I). With only rare exceptions, the Leishmania spp. that primarily cause cutaneous disease do not visceralize. The visceralizing parasites, Leishmania chagasi and Leishmania infantum (now considered by some taxonomists to be the same organism), can occasionally cause localized cutaneous lesions without visceral involvement. Leishmania donovani, the cause of classical visceral leishmaniasis (kala-azar) in the Old World, can also cause post-kala-azar dermal leishmaniasis, a condition of diffuse hypopigmented and erythematous nodular lesions that commonly involve the face and torso. These cutaneous lesions may develop during, or up to several years after treatment for kala-azar. Localized cutaneous leishmaniasis (LCL) is usually limited to one or a few skin ulcers that develop at the site of the sandfly bite and increase in size and often ulcerate over the course of several weeks to months. Lesions caused by Leishmania major and Leishmania mexicana usually heal spontaneously after 3 to 6 months, leaving a depressed scar. In general, lesions caused by Leishmania (Viannia) spp. tend to be larger and follow a more chronic course. In rare instances parasites (organisms of the L. mexicana complex; and rarely the Viannia subgenus in the New World and Leishmania aethiopica in the Old World) disseminate to involve large areas of skin (diffuse cutaneous leishmaniasis). An immunological defect probably underlies this severe form of cutaneous leishmaniasis. Mucosal leishmaniasis is an uncommon but serious disease (usually caused by parasites of the Viannia subgenus) and is characterized by destructive lesions of the nasal or oropharyngeal mucosa that occur following dissemination of the parasite from a primary skin infection.

Table I
figure Tab1

Etiologies of cutaneous leishmaniasis

Full size image

Field studies in endemic areas reveal people who have no history of disease, but evidence of infection by a positive Leishmanin [delayed-type hypersensitivity (DTH)] skin test.[2,3] In most endemic areas subclinical infection occurs more frequently than active disease.[2]

2. Acquired Immunity and the Rationale for Vaccine Development

Within an endemic region, the prevalence of skin test—positivity increases and the incidence of clinical disease decreases with age indicating the acquisition of immunity in the population over time.[4,5] Retrospective epidemiological studies in both the Old and New World indicate that individuals who have been subclinically infected, or healed from previously active cutaneous leishmaniasis, are commonly immune to a subsequent clinical infection.[5,6] Exceptions to this have usually occurred in individuals whose primary infection was in the distant past, or following immunosuppressive therapy.[6] However, in a prospective study carried out in Colombia,[2] new active lesions due to Leishmania braziliensis most often developed in individuals who had a typical scar. Thus, within a population there may be individuals who do not acquire immunity and are susceptible to reinfection or relapse.[2,7]

The presence of a positive Montenegro skin test (MST)—reaction, with or without a history of previous active disease, also appears to be predictive of immunity to subsequent disease.[5] In an endemic region in Peru, the size of the MST—reaction correlated with the degree of protection against subsequent clinical infection, suggesting that the magnitude of the antigenic exposure and subsequent host response determines the degree of acquired immunity.[5] In this study population, there was a demonstrable loss in MST—positivity over time in individuals without a previous lesion, but a clinical lesion led to a relatively permanent MST—positivity. Thus, the maintenance of a DTH response and an immune status, appears to require an initial antigenic exposure above a certain threshold, and/or boosting of the initial response by repeated exposure.

The acquisition of protective immunity following natural infection was confirmed by experimental challenge studies performed with L. major more than 30 years ago.[6,8–10] Individuals with healed lesions were shown to be resistant to subsequent experimental challenge with L. major.[6,11–13] A local DTH reaction often occurred at the site of cutaneous challenge.[6,11,14] Immunity to re-infection appeared to be intact only after the lesion began to re-epithelialize,[6,8,12,14,15] and immunity did not develop when the initial lesion was excised at an early stage.[9,10] Taken together, these findings suggest that a certain threshold of antigenic sensitization (over time) must be reached for the development of complete immunity to ensue, and/or that the nature of the immune response evolves to a protective phenotype with initiation of the healing process.

3. Mechanisms of Protective Immunity and Requirements for Effective Vaccination

The nature of the host cellular immune response to the infecting parasite determines the evolution and outcome of infection. Studies of the immunopathogenesis and mechanisms of protective immunity in both human and experimental cutaneous leishmaniasis have defined a number of features that should be characteristic of an effective vaccine. The mechanisms of vaccine-induced protective immunity are summarized in figure 1 and figure 2.

Fig. 1
figure 1

Vaccine-induced immunity against cutaneous leishmaniasis. 1. The induction phase of vaccine-induced immunity is initiated by administration of the vaccine with adjuvant (shown here by the intradermal route). 2. The vaccine antigens are captured by immature dendritic cells (DCs). 3. Dendritic cells mature and migrate in response to antigen/adjuvant—induced autocrine or paracrine production of inflammatory mediators. 4. Upon arrival in the regional lymph node the mature DCs, having up-regulated co-stimulatory molecules and interleukin-12 (IL-12) expression, present the processed antigen to naïve T cells (T). 5. The T cells are activated in the presence of IL-12, resulting in the expansion of an antigen-specific type 1 T cell population. Depending on the type of antigen, the specific T cell population may be CD4+ (activated by protein antigen processed through the class II pathway) or CD8+ (activated by DNA encoded antigen processed through the class I pathway), or both. 6. A subset of activated antigen-specific T cells become long-lived memory T cells. The mechanisms by which memory T cells are maintained have not been fully defined, but persistent antigen and/or IL-12 probably play a role.

Full size image
Fig. 2
figure 2

Vaccine-induced immunity against cutaneous leishmaniasis. 1. The effector phase of vaccine immunity is initiated in the vaccinated individual in response to intradermal inoculation of Leishmania promastigotes by the sand fly vector. 2. In the skin the parasites are rapidly phagocytosed by resident macrophages and/or dendritic cells (DC). 3. The DCs are induced to mature and migrate, carrying live parasites and/or bearing parasite antigens, to the regional lymph node. 4. In the lymph node the antigen(s) are presented to the memory T cells. 5 & 6. Under the influence of interleukin-12 (IL-12), the memory T cells rapidly expand into an antigen-specific type 1 [interferon (IFN)-γ—producing] effector T cell population. The expression of IL-12 also inhibits the generation of a parasite-induced pathogenic type 2 effector T cells response that may occur in some infected individuals. 7. The type 1 effector T cells are recruited to the site of infection. 8 & 9. T cell-derived IFNγ activates the infected macrophages to kill the intracellular parasites. Heavily parasitized macrophages are probably ineffective at killing parasites because of down-regulation of macrophage effector function. Lightly infected or uninfected macrophages that are exposed to IFNγ are able to rapidly kill invading parasites before macrophage function is impaired.

Full size image

An effective vaccine should prime the host for a type 1, but not type 2, cellular immune response to antigens that are readily recognized in the context of an infectious challenge. The importance of a type 1 T cell response for antileishmanial immunity is well established from studies of experimental models, and these data are generally supported by studies in humans. Resistance in murine cutaneous leishmaniasis, caused by L. major, is associated with the capacity of CD4+ T cells [T helper (Th) cell type 1 subset] to generate interferon (IFN)-γ in response to Leishmania.[16–19] The generation of interleukin (IL)-12, which promotes Th1 but not Th2 subset expansion, is critical for the development of resistance.[20,21] Studies involving T cells isolated from patients with active, healed, or subclinical LCL, although descriptive in nature, also support the role of Th1 responses in control of infection and subsequent immunity.[22–25] Despite these data, vaccine-induced production of IFNγ is not always a reliable in vitro correlate of protection against parasite challenge,[26,27] possibly because the parasite epitopes targeted by vaccine-induced immunity are not appropriately displayed on the infected macrophage.[28] Therefore, although IFNγ production is necessary, it may not be sufficient to confer protective immunity.

On the other hand, for a vaccine to effectively induce antileishmanial immunity it should prevent the host from responding to infectious challenge with the generation of counter-protective cytokines such as IL-4, IL-10, and tumor growth factor-β. These cytokines are associated with progression of L. major infection in the murine model[17,19,29] and the expression of IL-4 and IL-10 seem to be associated with a more severe disease phenotype in humans.[30–32] Early parasite-induced production of IL-4 is especially detrimental because it inhibits IL-12 receptor expression and the development of a Th1 response.[33] For effective vaccination these pathogenic responses must be redirected or overcome by a vaccine-induced type 1 cytokine response.[34]

The optimal vaccine may need to elicit a broad-based immune response to include both parasite-specific CD4+ and IFNγ-producing and/or cytotoxic CD8+ T cells. The role of type 1 CD4+ T cells in natural and vaccine-induced immunity is well established, especially for immunity induced by protein vaccines.[19,35] Evidence that type 1 CD8+ T cells could play a significant role comes from several studies, and vaccines that prime for both responses may be advantageous. Natural infection, which is the most effective inducer of long-term protective immunity, primes for major histocompatibility complex (MHC) class I restricted CD8+ T cells that contribute to immunity.[36,37] The role of type I CD8+ T cells in vaccine-induced immunity is especially evident following vaccination with antigen encoding DNA.[34,38]

A protective vaccine-induced immune response should lead to macrophage effector activity early after exposure to the parasite. There is mounting evidence that several functions of macrophages, including antigen-presenting capacity, cytokine generation, intracellular signaling, and leishmanicidal mechanisms, are significantly impaired once the macrophages are heavily parasitized.[39–46] At least some of these detrimental effects can be abrogated by early priming of the macrophages with IFNγ,[42,47,48] suggesting that an early type 1 response to parasite challenge could prevent the subversive effect that infection has on antileishmanial macrophage effector function.

The requirements for the maintenance of vaccine-induced memory T cells are hotly debated. Data from humans and experimental models indicate that although memory T cells can persist in the absence of specific antigen, periodic re-exposure to antigen clearly enhances the level of protective immunity.[49] Although the data are not conclusive, the maintenance of adequate T cell memory for long-term protective immunity against cutaneous leishmaniasis is likely to require the persistence or repeated delivery of antigen. Durable immunity following natural infection may in part be related to the subclinical persistence of parasites following healing of the clinical lesion. By analogy then, durable vaccine-induced immunity may be favored by delivery of vaccine antigen(s) in a form (such as vaccine protein complexed with alum adjuvant, or as vaccine DNA) that is likely to persist.[50] Repeated natural exposure to Leishmania may also contribute to the maintenance of naturally-acquired immunity,[5] and vaccine-induced immunity may be similarly boosted by natural exposure to the parasite within the endemic region. Despite this role of persistent or repeated antigenic stimulation, there are data that suggest that once antigen specific memory/effector T cells are induced, the sustained expression of IL-12 alone is sufficient to maintain long-term protection.[51] Clearly, the factors that contribute to durable vaccine-induced immunity need to be better defined.

4. Vaccination with Sandfly Salivary Antigens

Recently, the role of the immune response to components of the sandfly saliva that are introduced with the parasites during natural infection has been investigated. Naïve mice that are bitten by a sandfly do not develop a clinically detectable reaction but mice that have been sensitized by previous exposure to sandfly bites develop a strong DTH reaction, with infiltration of the site by macrophages, dendritic cells, lymphocytes, neutrophils, and eosinophils.[52,53] Although sandfly salivary components delivered simultaneously with L. major to naïve mice promoted the development of lesions,[54,55] pre-exposure to sandfly saliva or non-infective bites resulted in resistance to infection.[52,54] The production of IL-12 and IFNγ was a prominent part of the local response of the pre-sensitized mouse to the sandfly inoculum and may contribute to the antileishmanial defense through a bystander effect by early activation of macrophages for parasite killing and/or by rapid generation of a Leishmania-specific Th1 response.[52] Recently, a Phlebotomus papatasi salivary protein, SP15, was identified as a target of the immune response following natural exposure to sandfly bites.[56] Vaccination of C57Bl/6 mice with this antigen, either in protein or DNA form, conferred protection against challenge with L. major when inoculated with salivary gland homogenates. These studies need to be extended to other experimental models and to humans, but they have potential implications for investigations of antileishmanial vaccines. First, in field trials of vaccine candidates, the sandfly exposure status of the study population must be considered since a sandfly saliva—initiated immune response (if it were protective) could influence the baseline incidence of clinical disease in the study population and thus make vaccine-induced immunity less appreciable. Second, the inclusion of defined salivary antigens in a multi-component antileishmanial vaccine could improve vaccine efficacy, albeit at the potential cost of sensitizing the vaccinees for hypersensitivity reactions to subsequent non-infective sandfly bites.

5. Vaccination with Live Virulent Parasites

The recognition that Old World LCL (oriental sore caused by L. major) was a self-limited disease that produced long-standing immunity, and the occasional case of severe or disfiguring disease, led to the centuries old practice of vaccination with virulent parasites (L. major), called leishmanization. Several authors describe the practice of exposing the buttocks of infants to the bite of sandflies, or the inoculation of material from a lesion of another individual, so that the ensuing lesion would confer protection against subsequent disfiguring lesions and scars on the face.[14,57,58] A number of subsequent studies (reviewed in Melby [59]), primarily from Central Asia and Israel, demonstrated that vaccination of humans with axenically cultured L. major promastigotes resulted in immunity to re-infection in most of the vaccinated individuals, as long as the vaccine strain produced a clinical lesion.[8,60–62] In the largest trial of leishmanization, more than 1.3 million soldiers and civilians were vaccinated with live L. major promastigotes during the Iran-Iraq war of the 1980s. The typical live vaccine-induced lesions were 5 to 10mm in diameter and persisted for 4 to 6 months before self-healing. However, 2 to 3% of vaccinees developed large, non-healing lesions that required treatment. Evaluation of a subset of the study population showed that the incidence of naturally-acquired infection in the vaccinated (leishmanization) group was 2.6% (14/530), but in unvaccinated volunteers it was 14.2% (250/1724).[63,64] These leishmanization studies were pivotal to confirm the feasibility of prevention of human cutaneous leishmaniasis through vaccination, but the logistical difficulty of the approach and the occasional problem of severe vaccine-strain lesions also offered a strong argument for the development of a killed or subunit vaccine.

6. Vaccination with Live Attenuated Parasites

The epidemiological association of previous subclinical infection with protective immunity, and experimental infection in humans suggested that the development of optimal immunity may require replication and/or persistence of the parasitic antigens over a period of time. A number of studies have examined the use of live, attenuated parasite strains as vaccines in experimental animal models, but studies involving human volunteers have not been performed. Initial work in mice demonstrated that inoculation with avirulent parasites could prevent disease caused by challenge with homologous virulent parasites.[65,66] Intravenous administration of parasites that had been γ-irradiated was also able to confer protection against challenge with homologous (L. major) or heterologous (L. mexicana, Leishmania amazonensis, Leishmania panamensis) strains.[67–70] To be protective the irradiated parasites had to be metabolically active and retain the capacity to transform from promastigotes to intracellular amastigotes.[71] The protective effect was mediated by CD4+ T cells.[68,69] Subcutaneous administration of the irradiated promastigote vaccine to BALB/c mice exacerbated disease following cutaneous challenge,[67,70] but resistant (CBA) mice were protected by this type of vaccination.[71] This discrepancy may be explained by the finding that intravenous immunization of BALB/c mice leads to protective tolerance [possibly directed against the Leishmania homolog of receptors for Activated C Kinase (LACK) antigen] so that the pathogenic Th2 response is subverted.[72]

Recent advances have enabled the use of genetically modified Leishmania that can undergo a limited number of replications in vivo and thus mimic a sub-clinical infection. Immunization of mice with L. major promastigotes that had the dihydrofolate reductase gene disrupted led to protection against challenge with wild type L. major or L. mexicana.[73,74] Similarly, inoculation of L. mexicana promastigotes rendered avirulent by disruption of genes encoding cysteine proteases a and b conferred immunity against virulent parasite challenge.[75] Although such attenuated vaccines have the advantage of providing persistent exposure to multiple antigenic epitopes, the logistics of their use in the field would be complicated.

7. Vaccination with Killed Parasites

The use of killed parasites as a skin-test antigen for diagnosis and epidemiological studies, and the presence of a DTH response against this antigen preparation in immune individuals prompted the investigation of killed parasites as vaccines to prevent human cutaneous leishmaniasis. Initial studies performed by Brazilian investigators in the 1930s and 1940s suggested that significant protection could be achieved by vaccination with phenol-saline suspensions of cultured parasites.[76,77] These initial observations were followed by studies by Mayrink et al.[78–80] Using a vaccine cocktail of five different cultured Brazilian parasite stocks delivered intramuscularly, 78% of 231 recipients of the killed vaccine converted to a positive DTH skin test, compared with none of 100 unvaccinated volunteers.[78] In two subsequent efficacy trials, a similar cocktail of killed promastigotes marginally reduced the incidence of infection among those vaccinated and who had converted to a positive leishmanin skin test as compared with the placebo group.[79,80] In several other trials, using different killed New World promastigote vaccines, vaccinated individuals had a higher rate of skin test conversion (38 to 86%) and antigen-specific Th1 responses than unvaccinated volunteers.[81–83] In studies in Ecuador and Venezuela, inclusion of BCG in an intradermal vaccine formulation appeared to enhance vaccine immunogenicity,[81,84] but in a Colombian study intramuscular administration of killed promastigotes alone was superior to intradermal delivery of promastigotes plus BCG.[83] In this study intradermal vaccine plus BCG resulted in an unacceptable frequency of painful cutaneous nodules and ulcers.[83]

Studies of a killed (autoclaved) L. major promastigote plus BCG vaccine delivered intradermally have been performed in Iran.[85,86] In a randomized trial of a single dose of vaccine there was a low rate of skin test conversion (16 and 3% at 80 days and 1 year after vaccination, respectively), and there was no reduction in the 2 year incidence of anthroponotic cutaneous leishmaniasis (caused by Leishmania tropica).[86] Interestingly, when the data was stratified by sex there was significant protection among the vaccinated boys, possibly because boys were more likely to receive a booster effect from exposure to infected sandflies. In a parallel study, performed in an area endemic for zoonotic cutaneous leishmaniasis (caused by L. major), post-vaccination skin test conversion was higher (36 and 33% at 80 days and 1 year after vaccination, respectively), but there was no decrease in the 2 year incidence of cutaneous leishmaniasis.[85] However, there was a significant decrease in the incidence of cutaneous leishmaniasis during the first year among those who converted the skin test. A similar decrease in disease incidence among skin test converters was observed in a field trial of autoclaved L. major vaccine against visceral leishmaniasis in Sudan.[87] This suggests that there is potential for improvement in the efficacy of a killed vaccine if its immunogenicity could be enhanced to achieve a higher rate of skin test conversion, possibly through use of an improved adjuvant or administration of multiple doses. Because of studies showing that IL-12 was an effective adjuvant in the vaccination of mice against L. major,[88] the autoclaved L. major vaccine plus recombinant IL-12 was tested in vervet monkeys. Indeed the addition of IL-12 to the killed promastigote vaccine did lead to an enhanced antileishmanial type 1 response, but the animals were not protected against experimental parasite challenge.[26] In contrast, vaccination of rhesus monkeys with killed L. amazonensis plus IL-12 and alum as adjuvants conferred protection against parasite challenge.[50] It may be that the alum-mediated persistence of the antigen and/or IL-12 plays a role in the maintenance of a protective immune response.[51]

8. Vaccination with Subunit Protein and DNA Vaccines

Several purified and recombinant antigens have been demonstrated to induce protective immunity in the murine model of cutaneous infection with L. major, L. mexicana, or L. amazonensis (summarized in table II). In most instances, a highly susceptible mouse strain (e.g. BALB/c) has been challenged with a large dose of highly virulent parasites (usually L. major) by subcutaneous inoculation. These mice develop progressive infection with a highly polarized Th2 response. Such vaccine studies must be interpreted with some caution because this model is not very representative of human LCL (human disease occurs after deposition of a few hundred metacyclic promastigotes into the dermis and is characterized by a dominant Th1 response). A recently developed challenge model involving intradermal inoculation of a resistant mouse strain with several hundred sandfly-derived promastigotes appears to be more representative of human LCL.[54,89,90] To date, all of the studies of subunit vaccines have been pre-clinical. It will be necessary to validate the results of animal experiments in clinical studies in humans.

Table II
figure Tab2

Defined subunit vaccine candidates for cutaneous leishmaniasis

Full size image

The first molecularly defined molecule to receive attention as a vaccine candidate against cutaneous leishmaniasis was the abundant surface antigen lipophosphoglycan (LPG). Immunization of susceptible mice with purified L. major LPG conferred significant protection against parasite challenge.[119,120] Although the LPG was initially thought to induce T cell responses, it was subsequently shown that T cells were responding to a protein complex tightly associated with LPG.[121–123] The dominant LPG—associated protein was found to be an 11 kDa protein subsequently called kinetoplastid membrane protein-11.[124] Vaccination studies with this isolated protein have not been reported.

The highly expressed surface protease (gp63) has received the most extensive study as a subunit vaccine candidate. In several experimental models using different parasite and mouse strains, vaccine forms, adjuvants and routes of immunization, the gp63 vaccine has shown variable efficacy. Immunization with purified native gp63 showed protection against parasite challenge[91,92] but immunization with the purified recombinant protein was unsuccessful.[93] Expression of gp63 in BCG was protective against cutaneous L. major challenge.[99,100] Both L. major and L. mexicana gp63 expressed by an orally-delivered attenuated Salmonella strain induced protection against cutaneous challenge with the homologous parasite.[95–98] Delivery of gp63 as the antigen-encoding DNA was also protective in the L. major challenge model.[105,106] However, T cells from mice immunized with gp63 were unable to induce leishmanicidal macrophage activation.[92] Furthermore, although recombinant gp63 was effective in eliciting T cell responses from patients with active or cured leishmanial infection, sensitization of naïve human T cells in vitro did not effectively induce T cell responses.[125] These latter findings dampen the enthusiasm held for the potential for gp63 as a vaccine for humans.

The Leishmania homolog for receptors for activated C kinase (LACK) antigen is of interest as a vaccine candidate because the prominent role it plays in the immunopathogenesis of experimental L. major infection. During infection of mice that express I-Ad major histocompatibility complex (MHC) class II molecules (e.g. BALB/c strain), expression of the immunodominant LACK antigen drives the expansion of IL-4 secreting, disease-promoting T cells that express a Vβ4/Vα8 T cell receptor.[126,127] Depletion of LACK-reactive T cells diminished the early IL-4 response allowing the development of a protective Th1 phenotype.[127,128] Furthermore, immunization of highly susceptible BALB/c mice with a truncated (24 kilodalton, kDa) version of the 36 kDa LACK antigen, delivered in either protein (with recombinant IL-12 adjuvant) or DNA form, conferred protection against cutaneous challenge with L. major.[34,111,112,129] Delivery of the vaccine antigen by expression in Listeria monocytogenes also conferred partial protection.[130] The protective effect of the LACK vaccine was mediated by IL-12—dependent IFNγ production, which was sufficient to re-direct the early pathogenic IL-4 response observed in this infection model.[34] Interestingly, CD8+ T cells and persistent IL-12 production were required for the sustained Th1 immunity and durable protection achieved by the LACK DNA vaccine.[38,51] The protection achieved by the LACK protein + IL-12 protein was not dependent on activation of CD8+ T cells and was not durable.[34] Thus, the LACK vaccine protects against cutaneous L. major infection by redirecting the early T cell response away from a pathogenic IL-4 response and toward a protective Type 1 response. Although the LACK antigen is highly conserved in both Old and New World Leishmania spp. the efficacy of this vaccine against strains other than L. major has not been determined. Importantly, it remains to be seen whether this vaccine would protect against leishmanial disease that is not highly IL-4 dependent (e.g. human LCL).

The glycosyl-phosphatidylinositol anchored membrane glycoproteins gp46/M2 and promastigote surface antigen (PSA)-2 belong to a gene family that is present in all Leishmania spp., except L. braziliensis.[131–133] Immunization of CBA mice with the purified 46 kDa glycoprotein gp46 and Corynebacterium parvum adjuvant resulted in protection against a challenge with L. amazonensis promastigotes; partial protection was observed in BALB/c and C57BL/6 mice. [107] BALB/c mice immunized with gp46/M-2—recombinant vaccinia virus were significantly protected against challenge with low dose L. amazonensis (sterile immunity in 45 to 76% of the animals).[108] gp46/M-2—Specific, class I-restricted CD8+ T cell lines were found to recognize macrophages infected with L. amazonensis,[134] but the recombinant antigen was recognized by T cells from only a small percentage of patients with active or healing New World LCL.[135]

Vaccination of C3H/He and BALB/c mice with purified L. major PSA-2 (Mr of 80 to 94 kDa) plus C. parvum as an adjuvant resulted in a antigen-specific Type 1 cytokine response and protection from lesion development after challenge infection with virulent L. major.[109] Vaccination with Escherichia coli—derived recombinant PSA-2 did not confer protection in spite of a vaccine-induced Th1 response,[27,109] but immunization with recombinant PSA-2 expressed by and purified from L. mexicana promastigotes did protect.[109] Thus, some forms of the recombinant antigen induce an antigen-specific Th1 response that is apparently not recalled by in vivo infection. Further study showed that a PSA-2 DNA vaccine induced an exclusive Th1 response and protected against parasite challenge, but immunization with E. coli—derived recombinant PSA-2 in immune stimulating complexes (ISCOMs) induced a mixed Th1/Th2 response that was not protective.[110] Peripheral blood mononuclear cells (PBMCs) from individuals with a past history of self-healing cutaneous leishmaniasis (putatively immune individuals) demonstrated lympho-proliferative and IFNγ responses to L. major PSA-2.[136] Thus, studies in experimental models and limited experience with human cells suggest that PSA-2, if delivered in the proper form, could have a role in vaccination of humans against LCL.

Two L. major antigens, stress-inducible 1 (LmSTI-1) and Thiol-specific-antioxidant (TSA) protein were found to elicit primarily a type 1 T cell response in infected BALB/c mice.[115,137] Immunization of mice with the individual proteins plus IL-12, or a combination of LmSTI1, TSA, and IL-12 protein, or LmSTI1, TSA, and LACK DNA were protective against L. major cutaneous challenge.[89,116] Furthermore, vaccination of a small number of rhesus monkeys with LmSTI1 plus TSA plus IL-12 (protein) protected against cutaneous L. major challenge.[116]

Immunization of BALB/c mice with two amastigote stage-specific membrane proteins (P4 and P8) purified from Leishmania pifanoi induced protective immunity against parasite challenge.[114] Immunization with either the P4 or P8 antigen, when combined with C. parvum, conferred protection against homologous challenge with L. pifanoi promastigotes, but only the P8 antigen induced protective immunity against L. amazonensis challenge. Both the P4 and P8 proteins elicited a Th1 recall response by PBMCs from patients with New World LCL.[135,138] The P4 antigen has been identified as a single strand-specific nuclease;[139] the identity of the P8 antigen is unknown.

Several members of the Leishmania cysteine proteinase family have been studied as vaccine antigens. Immunization with a purified L. major amastigote-specific cysteine proteinase induced an antigen-specific IFNγ response and conferred protection against parasite challenge.[117] Immunization of mice with recombinant L. mexicana cysteine proteinase 5 (the protein encoded by the lmcpb cDNA) plus IL-12 adjuvant also conferred partial protection against parasite challenge.[118] However, immunization with a purified cysteine proteinase (A2) from L. pifanoi provided only minimal protection against homologous parasite challenge.[114] This antigen was found to elicit a Th1 response by PBMCs from a majority of individuals who had active or healed New World LCL.[135]

Leishmania eukaryotic initiation factor (LeIF) is a gene homolog of the eukaryotic initiation factor 4A initially isolated from L. braziliensis that induced a Th1-type T cell response in PBMC and induced T cell-independent production of IL-12 by antigen presenting cells.[113,140,141] Immunization of BALB/c mice, with recombinant LeIF without adjuvant, preferentially induced a Th1 response, but provided only partial protection against L. major challenge when the LeIF was given repeatedly as immunotherapy.[113] This protein may have a role as an adjuvant in combination with other vaccine candidate antigens.

Recently, DNA expression library immunization has enabled the screening of an entire parasite genome for immuno-protective proteins. Studies in experimental models of both cutaneous and visceral leishmaniasis have demonstrated that this is a feasible and powerful approach for identification of protective antigens in a relatively unbiased way.[142,143] Further utilization of this approach will no doubt yield new (and unexpected) vaccine candidate antigens that may have relevance to prevention of human disease.

9. Vaccine Adjuvants and Delivery Systems

The nature of the T cell response that is induced by vaccination is influenced by the associated inflammatory response and cytokine milieu in which the T cell activation occurs. Traditionally, killed whole bacteria (e.g. Corynebacterium parvum) or subcellular bacterial constituents (e.g. Salmonella monophosphoryl lipid A, mycobacterial cell wall components) have been used as adjuvants to augment antileishmanial vaccine-induced immunity.[109,118,144] Live bacterial vectors (e.g. Salmonella, BCG) that deliver the vaccine antigen(s), as well as induce a non-specific adjuvant effect, have been effectively used with subunit vaccines.[96,98,100] In addition to the advantage of sustained delivery of the vaccine antigen, the inflammatory signals elicited in response to a bacterial adjuvant may support the generation of antigen-specific memory T cells.[145] Other adjuvants such as alum and poloxamer 407, which form a complex with the antigen to enable its slow release, have been successfully used in vaccination against experimental cutaneous leishmaniasis.[50,102]

Recombinant cytokines, or the DNA that encodes the active cytokine, have much to offer in terms of adjuvant activity. T cells that are activated in the presence of IL-4 are driven toward a Th2 phenotype, whereas T cells that are induced in the presence of IL-12 are driven toward a Th1 phenotype. Not surprisingly then, IL-12 has also been demonstrated to augment the induction of a Th1 response and improve the efficacy of antileishmanial protein vaccines[88,112] and to enhance the durability of a DNA vaccine against cutaneous leishmaniasis.[34]

As noted above, DNA immunization has been identified as a way to obtain sustained delivery of a protein encoded by an isolated gene. Additionally, the DNA plasmid itself has strong immunostimulatory adjuvant activity, promoting a type 1 immune response. This effect is mediated by specific bacterial DNA sequences containing unmethylated CpG dinucleotide motifs that induce the production of IL-12, thereby polarizing the cellular response toward a Th1 phenotype.[146,147] Additionally, oligonucleotides that incorporate these immunostimulatory sequences activate dendritic cells and induce IL-12 production, thereby making them an attractive candidate for a Th1-promoting adjuvant.[148–150] Immunostimulatory oligonucleotides were shown to be an effective adjuvant when co-administered with vaccines against experimental cutaneous leishmaniasis.[151,152]

10. Conclusions

Protective immunity is commonly acquired following cutaneous infection with dermatropic Leishmania spp. so prevention of disease through prophylactic immunization appears to be feasible. Since vaccination with live, virulent parasites is no longer acceptable, attention has turned to the use of killed or attenuated parasite vaccines and defined antigens. Whole parasite vaccines have the advantage of delivering multiple antigenic peptides, but the efficacy of killed promastigote vaccines in clinical trials has been largely disappointing. If a subunit vaccine approach is taken it is likely that the induction of protective immunity will require responses directed against several different antigens/epitopes. Critical to the development of any vaccine is the determination of the requirements for maintenance of vaccine-induced memory T cells. Durable immunity is likely to require the delivery of the vaccine/adjuvant in such a way that there is sustained antigen release or antigen persistence, and/or sustained IL-12 production. The generation of memory T cells that have a low threshold for activation upon exposure to parasite challenge will be necessary. This effector T cell response must be prompt so as to pre-empt the subversive effect the intracellular parasite has on its macrophage host, since it is the activated macrophage that ultimately kills the parasite. Currently, vaccine-induced IFNγ production is the best indicator of a protective immune response, but better in vitro correlates of protective immunity need to be established, especially for vaccine studies involving humans.

In preclinical studies a number of purified or recombinant antigens appear promising as vaccine candidates for cutaneous leishmaniasis. Systematic studies need to be performed of these vaccine antigens to compare their efficacy, optimize vaccine formulation and route of delivery, and to identify effective and safe vaccine adjuvants. In most cases, protective epitopes need to be characterized, and multi-epitope vaccines will need to be constructed and studied. Further study of the efficacy of these antigens in experimental models that more closely reflect the immunopathogenesis of human infection is essential. Experimental challenge studies that utilize sandfly-transmitted infections would be especially meaningful because they provide a route and intensity of challenge that is representative of human infection, and deliver sandfly salivary components that may have immunomodulatory effects on the host. Ultimately, studies will need to move from the laboratory to the field to determine if these vaccines can have an impact on the prevention of cutaneous leishmaniasis.