Recent progress in vaccine development against Leishmania species infections

The understanding of the immunobiology of infections caused by the protozoan parasite leishmania is now extensive and has pinpointed the importance ofT cell-mediated immunity. Several vaccination strategies using either killed parasites, subunit vaccines, DNA vaccines or live attenuated strains have been used successfully with and without adjuvants to induce cellular immunity and protect against leishmania infections. The most recent progress in leishmania vaccine development is described.

Progres recents dans le developpement d'un vaccin contre les infections aux especes de Leishmania RESUME : La comprehension de l'immunobiologie des infections causees par le protozoaire parasite leishmania est desormais substantielle et a precise !'importance de l'immunite mediee par Jes lymphocytes T. Plusieurs approches vaccinales faisant appel a des parasites tues, a des vaccins a sous-unites, a des vaccins a l'ADN ou a des souches vivantes attenuees ont ete utilisees avec succes avec et sans adjuvants pour induire une immunite cellulaire et une protection contre Jes infections a leishmania. On decrit Jes progres les plus recents survenus dans la mise au point d'un vaccin contre leishmania.
T he protozoan leishmania is the cause of a wide spectrum of diseases in humans and domestic animals. The clinical manifestations of leishmaniasis vary with the species, encompassing cutaneous leishmaniasis (oriental sore), mucocutaneous leishmaniasis (espundia) and visceral leishmaniasis (kala-azar), the most severe form of the disease, which is often fatal if untreated. The flagellated promastigote leishmania is transmitted to humans by the phlebotomine sand fly. After its ingestion and entry into the phagolysosome of the host macrophages, the promastigote differentiates into the aflagellated amastigote, which then replicates within the macrophage. Leishmania are distributed worldwide, and between 10 and 15 million people are estimated to be infected, with 400,000 new cases reported every year (1). More than one-third of the world's population lives in endemic areas and is at risk of contracting an infection (2). Increases in travel and intervention in regional conflicts such as 'Operation Desert Storm' (3) have increased the number of leishmania cases in nonendemic areas. The incidence of leishmaniasis is also rising because of the lack of vaccines, difficulty of vector control and increased resistance to chemotherapy (4). Leishmania has also emerged as a serious opportunistic pathogen in human immunodeficiency virus (HIV)-infected humans, and several cases have been reported around the Mediterranean littoral (5). plays a key role in the control of Leishmania mqjor infection. It has been shown that mice with disrupted genes for IFN -y or IFN-y receptor failed to resolve their lesions ( 14,15). IL-12 also favours Thl cell development through its ca pacity to stimulate IFN-y production by Th 1 cells ( 16) and is crucial for cure. Mature Th 1 cells could, thus, expand in number and release more IFN-y, which increases IL-12 production by the macrophages. It is possible that IL-12 is produced by macrophages upon leishmania infection and that it induces Th I ce lls to secrete IFN-y. Both IL-12 and IFN-y may drive CD4+ T cell differentiation towards the Th 1 lineage ( 1 7) and are required for effective resolution of leishmaniasis ( 18). The essential role of IL-12 in Th 1 development has been confirmed by results showing that neutralization of IL-12 or deletion of the IL-12 gene in resistant mice led to the development of a Th2 response after leishmania infection (18 , 19).
In contrast, susceptibility to leishmania is characterized by a Th2 response, which produces predominantly IL-4. This cytokine promotes high antibody titres directed towards the parasite but does not activate macrophages for parasite killing. Cytokines produced by Th2 cells exert a macrophage deactivating function. IL-4 has been shown to hamper the activation of macrophages induced by !FN-y and to suppress the up-regulation of the gene for interferon regulatory factor 1 (20). Several reports support the crucial role of cytokines in directing CD4 + T cell differentiation and, consequently, the outcome of the disease. Macrophages pre-incubated in vitro with cytokines before infection with leishmania acquired the capacity to kill the intracellular parasites (21). Moreover, cytokines such as IFN -y, tumour necrosis factor-alpha (TNF-a), IL-12 and granulocyte-macrophage colony-stimulating factor (GM-CSF) have been used as antileishmanial therapy in experimental models (22 -2 6) and in human Leishmania donovani infection (27).
Natural killer (NK) cells have also been suggested to play a role in the development of a Th 1 response by secreting IFN-y at early stages following leishmania infection, and it has been shown that depletion of NK cells in resistant mice favours parasite multiplication (28). Accessory molecules such as the CD40 cell surface molecule (CD40L) on activated T cells have also been shown to be necessary for the generation of a protective cell-mediated immune response to leishmania, presumably via its interaction with the CD40 receptor (CD40R) on primed monocytes/macrophages, which induces IL-12 secretion (29). Resistant mice deficient in either the CD40 or its ligand molecules failed to generate a Th 1 response and were unable to control infection (30,31). It is, therefore, possible that during infection of resistant mice strains, macrophages present leishmanial antigens to T cells, resulting in CD40L activation, which in conjunction with !FN-y induces !L-12 production by the macrophages (30). CD40L may also have a direct effect on the synthesis of nitric oxide (32). Several studies have reported that antigen-specific cos+ T cells are similarly important in the resolution of cutaneous and visceral infections (33,34) . However, it is unclear how these cells execute this function because results from several investigations attempting to demonstrate recognition of leishmania-infected  (35) have shown that cos+ T cells are not required for the long term control of a primary infection with L mqjor.
The humoral immunity induced by many viral and bacterial vaccines mediates protection that is maintained over a long period of time. In contrast, for other intracellular infections such as with L mqjor for which cell -mediated immunity is required for protection, the mechanisms for developing durable responses after vaccination have not been well defined. A critical question is whether prophylactic vaccination requires exclusively the induction of a Th 1 response to protect against leishmanial infections or whether the prevention of a Th2-type response is also needed. It has been recently suggested that generation of a protective immunity is dependent on the induction of an exclusive Th 1 response and, most importantly, that the Th2 cell development has to be blocked (36). Generation of long term protective immunity against leishmaniasis must induce memory T cells, which in the presence of the parasite are stimulated to secrete protective Th 1 cytokines.

VACCINATION STRATEGIES AGAINST LEISHMANIASIS
Several obsetvations suggest that vaccination may be feasible in the control of leishmaniasis caused by the protozoan leishmania. For example, some leishmania skin infections, such as Old World cutaneous leishmaniasis, have a propensity to be self-limiting. Cure is followed by long-lasting immunity against reinfection. Furthermore, protective immunity also appears to develop in successfully treated kala-azar patients, suggesting that sterile immunity may be established following first contact with the parasite. The traditional practice of leishmanization, performed either by exposing certain areas of the skin to sand fly bites or by scratching the skin area with infected material, has shown certain protection against reinfection but is not widely accepted because of disease complications that have recurred in some patients (2). Overall, the experience supports the notion that a vaccine against leishmani asis is a reasonable possibility.  Experimentally, avirulent, temperature-sensitive mutants obtained by chemica l mutagenesis, lethally irradiated or heat killed, or soluble extracts of promastigotes have been used successfully in immunizing mice against cutaneous leishmaniasis (37)(38)(39). First-generation vaccines based on killed parasites with or without adjuvant have reached various stages of phase I, 11 or Ill trials in humans (2) ( Table 1). In Venezuela, over 16,000 individuals have been vaccinated with killed Leishmania mexicana and/or Leishmania braziliensis with or without Bacille Calmette-Guerin (BCG) (2). In Brazil, efforts to develop a nonliving promastigote vaccine against American cutaneous leishmaniasis have been made by using killed polyvalent promastigotes derived from Leishmania mexicana, Leishmania amazonensis, Leishmaniagt{Yanensis and Leishmania braziliensis (40). Vaccinated individuals in the above studies showed an enhanced level of IFN-y production and cos+ T-specific cells. In a recent study in Ecuador, over 70% of vaccinated children with killed leishmania promastigotes of three strains (L braziliensis, Lguyanensis and L amazonensis) combined with BCG were protected from cutaneous leishmaniasis (41). Two recent vaccine trials in Iran with killed L major promastigotes plus BCG were encouraging, but showed lesser efficiency relative to the incidence of the disease (42) . Live parasites have also been used for vaccination , administered either in a low dose (43,44) or as recombinant parasites expressing a cytotoxic gene (45). Both studies have demonstrated a protective immune response in susceptible BALB/c mice against infectious challenge.
Fractionation is an improvement over using the whole parasite in terms of both standardization and reduction of unwanted side effects, and effective immunization against leishmaniasis. Leishmanial antigens, either coated with lipids or uncoated, or in the presence of adjuvants or not, as well as T cell antigens, have been used as vaccines mainly in murine experimental systems (Table 1). Second-generation vaccine candidates, including recombinant molecules that constitute major components of the parasite membrane such as glycoprotein 63 (gp63) or glycoprotein 46 (gp46), have been tested in an experimental murine system ( Table 2). An oral recombinant gp63 vaccine (46) was shown to confer protective immunity against cutaneous leishmaniasis (47), and an improved version of this vector conferred considerab le cross-resistance to L donovani (48). The leishmania-specific C04 + T cell LACK antigen administered as a recombinant protein has been shown to transfer protection against L mqjor in susceptible BALB/c mice (49). T cells generated during the cure of leishmanial infections in humans recognize a broad range of leishmanial antigens, suggesting that although single-mo lecul e  vaccines such as gp63 or lipophosphoglycan have success in inbred mice, the search for a single major protective antigen for vaccination against disease in humans may be fruitless (50). Polymorphism for MHC class I and Il molecules on the surface of infected cells in genetically diverse human populations also makes a single-antigen vaccine less attractive because some members of the population may fail to bind the antigen for presentation to T cells (51). However, as Soong et al (52) reported , this may be different with the use of a multiple amastigote antigen vaccine.
The protease gp63 has also been administered as a DNA vaccine , and a Th I response was found to be associated with protection in vaccinated mice (53) (Table 3). Immunization with plasmid DNA has been shown to induce protective immunity in a variety of experimental systems through both MHC class I-and class II -restricted T cell responses (54 ,55). Vaccination with DNA encoding the immunodominant LACK parasite antigen also confers protective immunity to mice infected with L mqjor (56), and protection seemed to be associated with IL-12 production. Similarly, vaccination with promastigotespecific antigen 2 DNA can protect mice against L mqjor infection by developing a Th 1 type response (36).
Live recombinant attenuated vectors are widely used for vaccination as vehicles to express immunogenic proteins from several pathogens. Genetically or naturally attenuated microbes such as salmonella , BCG and vaccinia-expressing leishmanial antigens , toxins or cytokines have been administered as vaccines in murine experimental systems (Table 4). Attempts to generate live attenuated leishmania strains for vaccination purposes have also been recently undertaken. Our ability to introduce stable new genes and disrupt or delete endogenous ones has provided the tools necessa,y to generate genetically less virulent or avirulent parasites that may be used as safe live vaccines for protozoa I diseases. Development of live parasites attenuated by molecular means has been a driving force in the recent years (Table 5). Null mutants by gene targeting have been generated in L mqjor (dllfr-ts [57) hsp!OO, [58], and L mexicana (cysteine proteinase gene 12C [lmcpb] [59]). Parasites in which a large cluster of cysteine proteinase genes were replaced by gene targeting were found to be less virulent and to confer protection against challenge (60). Using antisense RNA against the amastigote specific L dono vani A2 gene, parasites became less virulent and animals were protected against challenge (61 ). Disruption of the L mqjor hsp!OO gene has resulted in markedly delayed lesion development in mice (58). Although encouraging, protection was often not perfect and a minor population of the parasites was reverting to virulence, indicating that further work is required (60,61) . We have disrupted the gene encoding for trypanothione reductase (TR), an enzyme keeping trypanothione into its reduced form (62). The pivotal role of TR in oxidative stress management suggests that it might be an attractive target for the production of avirulent strains. Attempts to yield a null mutant for the TR gene obtained a TR trisomic mutant with two alleles successfully disrupted by the two selectable markers and a third wild-type allele as a result of genomic translocation (63). The resulting polyploidy strongly suggests that TR gene is essential for leishmania promastigotes. A similar conclusion has been drawn for the dfJ/r-ts gene (64) and for a cdc2-related kinase (59). Despite that one TR allele was left, susceptible BALB/c vaccinated with a L donovani TR disruption mutant were protected against reinfection with a wild-type L donovani strain. Tovar et al (65) reported similar results more recently.
Other putatively interesting targets for the development of live vaccines against leishmaniasis using gene targeting technologies may be the amastigote-specific genes because their inactivation should alter either the capacity of the parasite to efficiently differentiate or its capacity to survive once inside the phagolysosomes. New targets identified through the ongoing L mqjor sequence project may also be chosen for the generation of live attenuated mutants. Ideally, an attenuated live vaccine should cover a large spectrum of different species because the epidemiology of strains responsible for leishmaniasis is not the same in the New World and in the Old World , and, therefore, vaccine composition should change accordingly. Given that T cell -mediated immunity is required for the development of a protective immune response against leishmania infections, live attenuated vaccines should be ideal candidates for vaccination. Live attenuated vaccines have been used against a number of human and animal pathogens with high efficacy and safety. Several viral (66 ,67) and bacterial (68-70) pathologies can be prevented efficiently using live Can J Infect Dis Vol 10 Suppl C May 1999 attenuated pathogens as vaccines. BCG, a naturally live attenuated Mycobacterium bovis strain, is one of the most widely used vaccines in the world, being administered to approximately 100 million children each year. All BCG strains used so far as vaccines are safe (70). Auxotrophic strains of BCG were recently made to obviate potential adverse effects of BCG vaccine in HIV-positive severely immunocompromized individuals (71). Although side effects were not seen in several studies of HIV-sero positive children (71), the safety of live attenuated BCG strains and of other live attenuated vaccine candidates needs to be carefully addressed.
The elaboration of an efficient vaccine involves the generation of short lived effector cells, but also generation of long term protective memo1y cells. The nature of the cells that confer immune memory against parasitic infections and the mechanisms by which it is obtained are unknown. To assess the requirements for the development of a long term immunity following a leishmania infection, we have developed a suicide-type system based on the expression of the thymidine kinase gene of Herpes simplex virus-1 (tk) gene in leishmania that become hypersensitive to treatment with ganciclovir, and we have tested the potential of using these recombinant parasites as a vaccination approach (45). Mice infected with TKrecombinant leishmania and treated four days later with ganciclovir to clear the parasites that were protected against infective challenge (45).
Adjuvants in vaccines are thought to function in several ways, including targeting of antigens to macrophages, CD4+ T cell subset differentiation and macrophage activation, but the mechanism(s) by which they act is poorly understood. BCG has often been used in several combinations (Table 1)  been demonstrated. Vaccination of BALB/c mice with leishmanial antigens and IL-12 promoted the development of CD4 + T cell response (72), suggesting that IL-12 can substitute for bacterial adjuvants (Table l). In Kenya, IL-12 has been tested in vaccination experiments in primates (73). In other cases IL-2, lFN-y and TNF-a have been expressed by recombinant attenuated bacteria and tested as vaccines againstL mqjor infections (Table 4). we have also tested the potential of expressing cytokine genes by the parasite to induce macrophage activation. GM--CSF-expressing leishmania were used to infect BALB/c mice, and the overproduction of this cytokine seems to control the level of infection during the first weeks.

CONCLUSIONS
In recent years, considerable progress has been made concern ing the understanding of the immunobiology of leishmania infections, in the genetics of leishmania and in the isolation of parasite surface molecules. This knowledge has been used to develop ingenious and effective strategies for vaccination against leishmaniasis. With further basic and clinical research, it is possible that for the first time an effective vaccine will be available for an intracellular parasite.