WO1999051264A2 - Leishmania cysteine proteinases - Google Patents

Abstract

The present invention relates to a Leishmania vaccine, more particularly a Leishmania vaccine comprising active or inactive cysteine proteinase(s), particularly in recombinant form, for use in immunising mammals, such as dogs and/or humans.

Description

Leishmania Cysteine Proteinases

The present invention relates to a Leishmania vaccine, more particularly a Leishmania vaccine comprising active or inactive cysteine proteinase (s) for use in immunising mammals, such as dogs and/or humans.

The World Health Organisation estimates that Leishmania is prevalent in 88 countries of the world with a population at risk estimated at 367 million with 400,000 deaths a year. Leishmania species also infect other mammals. In many countries, leishmaniasis in dogs is an important problem. The close proximity of humans and dogs, and the evidence that dogs and other mammals act as reservoirs of human infection, have meant that there has been widespread killing of dogs as a disease control measure. This is clearly unsatisfactory, but also has not made a significant impact on the spread of human disease. Thus the development of a vaccine would be of immense benefit to both humans and dogs . Leishmaniasis encompasses a large spectrum of clinical diseases, which depending upon the parasite species and the host immune response, can have various outcomes. In humans these clinical syndromes include single cutaneous lesions, which may or may not spontaneously heal, and more severe cases associated with metastasis to other cutaneous or mucocutaneous sites, as well as visceralization of the parasites leading to a fatal infection if not properly treated. Present understanding of the factors that lead to this diversity of clinical symptoms has, in large part, come from studies using murine models (reviewed Alexander & Russell, 1992; Liew & O'Donnell, 1993). While the vast majority of mouse strains develop healing lesions when infected subcutaneously with L . major, virtually ail develop non-healing lesions full of parasites when infected subcutaneously with L . mexicana (Roberts et al . , 1989; Roberts et al . , 1990). Furthermore, as L . mexicana will visceralise from primary lesions in mice under the influence of genetic controls originally identified from studies of L . donovani , this parasite in mice offers an excellent model system for putative vaccine studies against disseminating disease in a variety of susceptible genotypes.

In areas such as Southern Europe and South America where the causative agents of visceral leishmaniasis are L . infantum and L . chagasi , respectively, the domestic dog represents the major animal reservoir of the disease. The incidence of canine leishmaniasis throughout Southern Europe can range from 10% to as much as 60% (Dye et al . , 1992, 1993). Once a dog develops symptoms the outcome is invariably fatal (Bray 1982; Slappendel 1988). A major question remaining to be addressed, however, is whether vaccination can effectively induce a cellular immune response and protection against canine leishmaniasis.

It is the general consensus of opinion that acquired protective immunity against murine leishmaniasis, both cutaneous and visceral, is dependent on the ability to mount an IL-12 driven CD4+ T Helper 1-type response (reviewed Bogdan et al . , 1996). This lymphocyte subset produces IFN-γ which mediates protection by up-regulating macrophage inducible nitric oxide synthase (iNOS) expression and NO production which is microbicidal for the parasites (Liew & O'Donnell, 1993). Consequently, neutralisation of IL-12 or IFN-y, or inhibition of NO production, results in disease exacerbation (reviewed Bogdan et al . , 1996).

The immunological pathways leading to the development of non-healing progressive disease are less well characterised and more contentious. Thus, although a large number of studies have indicated that susceptibility to L . maj or (reviewed Liew u O'Donnell 1993; Bogdan et al . , 1996), L . mexicana (Satoskar et al . , 1995) and . amazonensis (Afonso & Scott, 1993) is related to developing a Th2 response and IL-4 production with down-regulation of Thl-associated activities, further studies on these species, as well as L . donovani , suggest that the inability to mount a Thl response rather than the presence of Th2 response may determine susceptibility (Kaye et al . , 1991; Satoskar & Alexander, 1995; Noben-Trauth et al . , 1996). Nevertheless, recent studies using IL-4 gene-deficient mice from a number of genetic backgrounds have demonstrated a requirement for IL-4 in L . mexicana disease progression. In the absence of this cytokine, lesions did not develop at the site of subcutaneous inoculation (Satoskar et al . , 1995) .

Proteinases have been shown to play an important role in the pathogenicity of parasitic protozoa (see Robertson et al . , 1996; Coombs & Mottram, 1997). L . mexicana contains multiple, highly active cysteine proteinases (CPs) , many of which are stage-regulated. The present inventors have characterised biochemically a large number of CPs, many of which are stage-specific (reviewed Coombs & Mottram, 1997) , and have isolated three L . mexicana CP genes; cpa , a single copy gene encoding a non-abundant cathepsin L-like CP (Mottram et al . , 1992); cpb , a multicopy gene which encodes the major cathepsin L-like CPs of amastigotes (Souza et al . , 1992); and cpc , a single copy gene encoding a cathepsin B-like CP (Bart et al . , 1995).

In Leishmania , genes can be deleted by homologous recombination using gene-specific targeting DNA linked to an antibiotic resistance gene, such as hyg or neo , providing positive selection. cpa null mutants have been generated, but have no detectable phenotype (Souza et al . , 1994) . cpb null mutants, however, were found to have a virulence phenotype (Mottram et al . , 1996).

All life-cycle stages of the cpb null mutant can be cultured in vitro , demonstrating that the gene is not essential for growth or differentiation of the parasite under these conditions. The null mutant, however, exhibits a marked phenotype affecting virulence - its infectivity to macrophages is reduced 5-10 fold. Data suggest that the mutants can only survive in a subpopulation of macrophages, but the parasites that successfully infect these macrophages grow normally. cpa/ cpb double null mutants were also created using four antibiotic selectable markers, hyg, ble, sat and pur (Mottram et al . , 1996) . These had a similar phenotype to the cpb null mutant in terms of macrophage infectivity, showing that cpa does not compensate for the loss of cpb functions in this phenotypic test.

It was also observed that subcutaneous lesions in BALB/c mice resulting from inoculation of the cpb null mutant appeared considerably later than those due to the wild-type parasites, but nevertheless lesions were observed

(Mottram et al, 1996).

The development of a live Leishmania vaccine line by gene replacement has previously been studied. A conditional auxotroph of L . major in which the dihydrofolate reductase-thymidylate synthase (dhfr-ts) gene had been deleted, was evaluated for its usefulness as a potential vaccine. The dhfr-ts mutant was found to be an effective vaccine line for immunizing against cutaneous

Leishmania and was shown to be incapable of establishing a persistent infection or causing disease in the most susceptible strains of mice tested. However, mild infections were observed after subsequent parasite challenge, which is generally undesirable for a vaccine

(Titus et al. , 1995) .

It is among the objects of the present invention to provide an improved Leishmania vaccine.

In one aspect the present invention provides the use of at least one Leishmania cysteine proteinase in the preparation of a vaccine.

The at least one Leishmania cysteine proteinase may be an active or a substantially inactive cysteine proteinase.

It is understood that "an active cysteine proteinase" is one which displays an enzyme activity associated with cysteine proteinases and which is antigenic or immunogenic. The at least one "substantially inactive Leishmania cysteine proteinase" is understood to be substantially enzymatically inactive, but is antigenic or immunogenic. Thus the substantially inactive cysteine proteinases of the present invention have been modified such that the modified cysteine proteinase (s) is/are dysfunctional in terms of enzymatic ability, but are however antigenic or immunogenic, such that a host may elicit an immune response to the inactive cysteine proteinase (s) . Thus the term "substantially inactive Leishmania cysteine proteinase" also extends to polypeptide fragments of cysteine proteinase which display an antigenic and/or immunogenic function.

Generally speaking the active or inactive cysteine proteinase (s) is/are provided as a recombinant protein which has been expressed from a functional or defective Leishmania cysteine proteinase gene respectively. A defective cysteine proteinase gene may for example have er. erec. αeiacc-Vi o site-directed mutagenesis of a functional cysteine proteinase gene. A further aspect of the invention relates to the vaccine itself.

A "defective cysteine proteinase gene" is one which is substantially incapable of encoding for a native cysteine proteinase or a functional equivalent thereof. In line with common terminology cysteine proteinase is understood to relate to and include the proteolytic enzymes containing a nucleophilic cysteine as a member of the catalytic machinery. This is described in detail in Barrett and Rawlings 1996. Thus, a "defective cysteine proteinase gene" means that the cysteine proteinase gene has been modified for example by a deletion, insertion, substitution (or other change in the DNA sequence such as rearrangement) such that the cysteine proteinase gene is generally incapable of expressing a functionally competent cysteine proteinase from said gene. The "defective cysteine proteinase gene" is however capable of expressing a defective cysteine proteinase which is substantially inactive enzymically.

To produce an inactive cysteine proteinase the active site cysteine of the cp gene can be changed for example by site-directed mutagenesis to a glycine residue. This mutation results in the production of a full length cysteine proteinase enzyme that is functionally inactive. Similar dysfunctional cysteine proteinases can be produced by further mutagenesis as described for example in example 2. Alternatively a cysteine proteinase may be rendered inactive by chemical inactivation for example by using irreversible cysteine proteinase inhibitors such as iodoacetate, E64 and n-ethylmaleimide.

The active or inactive Leishmania cysteine proteinase may be selected from all species of Leishmania including L . braziliensis, L . peruviana, L . guyanensis . L . mexicana . L . maj or , L . amazonensis , L . infantum , L . c aσasi and L . donovani , Preferably the active or inactive Leishmania cysteine proteinase displays cross-protection to other Leishmania species. Thus, for example, a mammal immunised with an active or inactive L . mexicana cysteine proteinase as described herein may not only provide protection to infection from disease-causing L . mexicana but also to other disease-causing species, as listed above. If the active or inactive cysteine proteinase is for example a L . mexicana cysteine proteinase then said at least one active or inactive cysteine proteinase may be selected from, for example, CPA, obtainable from a single copy gene encoding a non-abundant cathepsin L-like CP; CPB, obtainable from a multicopy gene which encodes the major cathepsin L-like CPS of amastigotes; and CPC, obtainable from a single copy gene encoding a cathepsin B-like CP.

The present inventors have now also identified corresponding genes to cpa and cpb in L . infantum , such that active and inactive forms of L . infantum cysteine proteinase (s) may also be produced as described herein and used in the preparation of a vaccine. Therefore, the present invention further provides a vaccine formulation comprising at least one active or substantially inactive L . infantum cysteine proteinase. The present invention also provides L . infantum related genes for cpa and cpb (see Figures 5 and 7a, 7b respectively) and their corresponding polypeptides (see Figures 6 and 8a, 8b respectively) , as well as dysfunctional equivalents thereof.

The active or inactive cysteine proteinase may for example be obtained from the cpa , cpb and cpc genes of . mexicana as well as the corresponding cpa and cpb genes of

L . infantum as disclosed herein (see Figures 5 and 7a, 7b) .

The present invention therefore also relates to the use of L . infantum cpa and cpb genes (see Figures 5 and 7a, 7b) and to the corresponding proteins expressed therefrom (see Figures 6 and 8a, 8b) .

The dysfunctional cysteine proteinase (s) may be provided in the vaccine as a purified or semi-purified protein or expressed by other carriers such as bacteria, viruses and protozoa.

The active or inactive cysteine proteinases of the present invention retain a sufficient immunogenic function to elicit at least a cellular immune response (such as cytotoxic T-cell response more preferably a Thl cell response) in a host animal, such as a dog or human. If the prophylactic and/or therapeutic effect of an appropriate active or inactive cysteine proteinase of the present invention is to be augmented, an appropriate adjuvant protein or polypeptide, such as a cytokine, for example, γ interferon (γIFN) can also be employed as a component of a vaccine or pharmaceutical composition of the invention.

The Leishmania cysteine proteinases of the present invention may De administered in the form of a DNA vaccine as described for example by Tighe et al 1998 or by a protein particle delivery system for example Hepatitus B surface antigen particle or Hepatitis B core antigen particle or yeast TY particle as described for example by Gilbert and Hill 1997.

Leishmania cysteine proteinases of the present invention may be applied directly to the cells of an animal in vivo , or by in vitro infection of cells taken from the said animal, which cells are then introduced back into the animal. Active or inactive Leishmania cysteine proteinases may be delivered to various tissues of the animal body including muscle, skin or blood cells thereof. The Leishmania cysteine proteinases may be injected into for example, muscle or skin using a suitable syringe.

Leishmania cysteine proteinase for injection, may be prepared in unit dosage form in ampoules, or in multidose containers. The cysteine proteinases may be present in such forms as suspensions, solutions, or emulsions in oily or preferably aqueous vehicles. For any parenteral use, particularly if the formulation is to be administered intravenously, the total concentration of solutes should be controlled to make the preparation isotonic, hypotonic, or weakly hypertonic. Nonionic materials, such as sugars, are preferred. Any of these forms may further comprise suitable formulatory agents, such as starch or sugar, glycerol or saline. The compositions per unit dosage, whether liquid or solid, may contain from 0.1% to 99% of parasite material. In a further embodiment of the invention there is provided a vaccine against Leishmania comprising at least one active or substantially inactive Leishmania cysteine proteinase. The vaccine of the invention may optionally include a further compound having an immunogenic function such as a cytokine, for example, v interferon.

In a preferred presentation, the vaccine can also comprise an adjuvant. Adjuvants in general comprise substances that boost the immune response of the nost in a non-specific manner. A number of different adjuvants are known in the art. Examples of adjuvants may include Freund ' s Complete adjuvant, Freund ' s Incomplete adjuvant, liposomes, and niosomes as described in WO 90/11092, mineral and non-mineral oil-based water-in-oil emulsion adjuvants, cytokines, short immunostimulatory polynucleotide sequences for example in plasmid DNA containing CpG dinucleotides such as those described by Sato Y. et al. (1996) ; Kreig A.M. (1996) .

In addition, the vaccine may comprise one or more, suitable surface-active compounds or emulsifiers, e.g. Span or Tween.

In a further aspect of the invention there is provided the use of a cysteine proteinase (s) as described herein for the manufacture of a vaccine for the prophylaxis and/or treatment of Leishmaniasis. Most preferably, the use is in dogs or humans.

In a further aspect of the invention there is provided a method of treating animals which comprises administering thereto a vaccine composition comprising at least one active or substantially inactive cysteine proteinase as described herein to animals in need thereof. Preferably, the animals are dogs or humans. Naturally, the vaccine formulation may be formulated for administration by oral dosage, by parental injection or otherwise.

The invention also provides a process for preparing a Leishmania vaccine, which process comprises admixing at least one active or substantially inactive cysteine proteinase as herein described with a suitable carrier or adjuvant.

The mode of administration of the vaccine of the invention may be by any suitable route which delivers an immunoprotective amount of the protein of the invention to the subject. However, the vaccine is preferably administered parenterally via the intramuscular or deep subcutaneous routes. Other modes of administration may also be employed, where desired, such as oral administration or via other parental routes, i.e., intradermally, intranasally , or intravenously. Generally, the vaccine will usually be presented as a pharmaceutical formulation including a carrier or excipient, for example an injectable carrier such as saline or pyrogenic water. The formulation may be prepared by conventional means. It will be understood, however, that the specific dose level for any particular recipient animal will depend upon a variety of factors including age, general health, and sex; the time of administration; the route of administration; synergistic effects with any other drugs being administered; and the degree of protection being sought. Of course, the administration can be repeated at suitable intervals if necessary.

Embodiments of the invention will now be illustrated by way of the following Figures and Examples; wherein Figure 1 shows an SDS-PAGE and Western Blot analysis of the purification of CPB2.8ΔCTE from E . coli . Panel A- induction of CPB2.3ΔCTΞ expression by IPTG. SDS-PAGE gel stained with Coomassie Blue. Lane M_r, protein standards; lane 1, E. coli lysate pre-IPTG induction; lane 2, E . coli lysate post-IPTG induction for 5 hours at 37°C. (His)_β- tagged pro-CPB2.8ΔCTE is indicated by the arrow. Panel B- purification from inclusion bodies and activation of CPB2.8ΔCTE. SDS-PAGE gel stained with Coomassie Blue. Lane M_r, protein standards; lane 1, E . coli whole cell lysate; lane 2, soluble fraction of E . coli lysate; lane 3, inclusion body fraction; lane 4, inclusion body fraction washed and solubilised in 8 M urea; lane 5, post-dialysis; lane 6, pooled Mono Q fractions; lane 7, post-activation. Approximately lOμg of protein/ lane . Arrows indicate the major, detectable forms of CPB2.8ΔCTE. Panel C- (His)₆- tagged pro-CPB2.8ΔCTE purified by nickel agarose resin. SDS-PAGE analysis, Coomassie Bue staining, the protein is indicated by the arrow. Panel D- Western Blot analysis of the samples detailed for Panel B, using polyclonal antiserum raised against CPB2.8ΔCTE. 1 μg of protein/ lane. Panel E- silver-stained SDS-PAGE gel of CPB2.8ΔCTE eluted from the Sephadex G50 column. Figure 2 shows a Gelatin-SDS-PAGE analysis of CPB2.8ΔCTE during purification. Lane 1, pooled Mono Q fractions before activation; Lane 2, after activation via 8 h at 37°C in 0.1 M Na acetate buffer pH 5.5, 2 mM EDTA, 10 mM DTT, 0.9 M NaCl; lane 3, whole cell lysate of L . mexicana CPB null mutant expressing CPB2.8+CTE via an episome. The three groups of activities are labelled A, B and C.

Figure 3 is a schematic representation of the processing steps during CPB2.8ΔCTE maturation. The solid block represents the protein encoded by the complete gene encoding CPB2.8ΔCTE, with the three domains distinguished. The numbers 1-4 signify the N-terminus of the four proteins identified, the N-terminal amino acid sequences being given below. The amino acid sequence surrounding the (His)₆-tag is given as this deviates from the gene sequence of the native gene (Mottram et al . , 1997).

Figure 4 shows an analysis of the early stages of CPB2.8ΔCTE processing during the purification. SDS-PAGE analysis of the recombinant prot=_. —:_". incl s o bcdy extraction, solubilisation in urea and subsequent dialysis. The effects of inhibitors on the refolding, re-activation and the initial step in CPB2.8ΔCTE processing that occurrs during the dialysis step was also investigated. Lane M_r, protein standards; lanes 2-6, inclusion body extraction and solubilisation: after Triton buffer wash (lane 2) ; after 2 M urea buffer washes (lanes 3 and 4) ; after water wash (lane 5); after solubilisation in 8 M urea (lane 6); lanes 1, 7-9, after dialysis: control, no inhibitor (lane 7); with 10 μM E64 (lane 8); with 1 μM pepstatin (lane 9); with 1 mM PMSF (lane 1) .

Figure 5 shows the DNA sequence of the Leishmania infantum cpa gene.

Figure 6 shows the amino acid sequence of the L . infantum CPA protein.

Figures 7a and 7b show DNA sequence of the L . infantum cpb gene . Figures 8a and 8b show amino sequence of the L . infantum CPB protein.

Figure 9 shows the alignment of deduced amino acid sequence of Leishmania infantum cpa with the deduced amino acid sequence of L . mexicana (lmcpa , Genbank accession number X62163) . Dashes (-) represent amino acid identity. *** show the 3 amino acid insertion sequence characteristic of this class of cysteine proteinase (Mottram et al., 1992). The arrow shows the predicted cleavage site of pro/mature domains. The active site cysteine is indicated

(*)•

Figure 10 shows the alignment of deduced amino acid sequence of Leishmania infantum CPB with the deduced amino acid sequences obtained from GenBank. These include L . mexicana (CPB2.8, Z49962), L . pifanoi (lpcys2, M97695) , L . major (lmjl , U437006) and the arrows show the predicted site of pro-mature domain and mature/C- terminal extension processing. Dashes (-) represent amino acid identity, and the arrows show the predicted site of pro/mature domain and mature/C-terminal extension processing. The active site cysteine is indicated (*) .

Example 1

Expression of active and inactive forms of recombinant CPB

Details described below are for the generation, expression and purification of active CPBΔCTE cysteine proteinase (The CPB2.8 lacking the C-terminal extension). However, the same protocol may essentially be followed for the production of other CPB isoenzymes, including ones rendered inactive by site-directed mutagenesis or ones containing the C-terminal extension, either in full or in part, and also ones modified by site-directed mutagenesis.

The CPB isoenzymes of L . mexicana are expressed as inactive zymogens comprising from the N-terminus: a pre- region of 18 amino acids that is rapidly removed by a signal peptidase; a 106-107 amino acids pro-region; a 210- 220 amino acids mature domain including the active site; and, characteristic of just this class of cysteine proteinases, a 16-100 amino acids C-terminal domain (Souza et al., 1992; Mottram et al., 1997). The conversion of the zymogens to mature, active enzyme is thought to require processing of the pro-region and C-terminal domain resulting in mature enzymes of approximate M_r23300 (Robertson & Coombs, 1994) .

Cloning cpb2 . 8 into pQE-30

PCR Amplification of Leishmania mexicana cpb2 . 8

A PCR product was amplified from the lmcpb2 . 8 gene (Mottram et al . , 1996, Proc. Natl . Acad . Sci . USA, EMBL data base number Z49962) using primers JH9601 and JH9602:

Primer JH9601 GGATCCGCCTGCGCACCTGCGCGCGCGA

Primer JH9602 AAGCTTCTACCGCACATGCGCGGACACGG

PCR Conditions:

20μl reactions using 0.5μl (1 unit) of VENT Polymerase (New England Biolabs) with 11. lx PCR mix. After the PCR reaction was complete, A overhangs were added by incubating with Taq Polymerase at 72°C for 10 minutes before DNA was phenol/chloroform extracted and ethanol precipitated. Cloning of cpb2 . 8

The cpbg2 . 8 PCR product was cloned into the pTAG vector (R&D Systems) . It was then excised from pTAG with BamKl and HindiII and cloned into the pQE-30 vector (Qiagen) , using the same restriction enzymes, to give plasmid clone pGL180. pGL180 was transformed into M15[pREP4] Escherichia coli for expression studies.

Expression of recombinant CPB2.8ΔCTE

CPB2.8ΔCTE is the form of cysteine proteinase lacking the C-terminal domain, which is mentioned above.

A single colony of the E . coli M15pREP4 expression strain transformed with the pGL180 construct was inoculated into 8 ml of LB broth supplemented with ampicillin (100 g.ml^"1) and kanamycin (25 μg.ml^" ) and grown at 37°C. This culture was used to inoculate 400 ml of LB/Amp/Kan broth and the culture grown until an OD₆₀₀ of 0.7 was obtained. r.-.c cΛ rs53-.:n cf CP52.3ΔCTΞ was induced by the adci on of IPTG to a final concentration of 1 mM. After 3 hours, the bacterial cells were pelleted by centrifugation at 4000g for 10 min.

Isolation of active, recombinant CPB2.8ΔCTE from the soluble fraction

The bacterial pellet was resuspended in 10 ml of 50 mM Tris/HCl buffer, pH 8 containing 5 mM EDTA and 5% (w/v) sucrose. The suspension was subjected to two rounds of freeze-thaw and six 30 second bursts on a sonicator at 4° C. The bacterial lysate was centrifuged at 6000 g for 10 min to pellet insoluble material and leave a supernatant fraction containing recombinant enzyme. This soluble fraction was dialysed for 3 hours against 400 volumes of 0.1 M Tris/HCl buffer, pH 8, 0.5 M KC1, 1 mM β- mercaptoethanol (buffer A) . The sample was filtered through a 0.22 μm filter and loaded directly at 0.2 ml. min^"1 onto a Ni-agarose (Qiagen) column pre-equilibrated in buffer A. The chromatography was effected with a stepped gradient of 0-1M imidazole in 0.1 M Tris/HCl buffer, pH 8, 0.5 M KC1, 1 mM β- mercaptoethanol at a flow rate of 1 ml.min^"*. Contaminating proteins eluted from the column in 10 ml of buffer A. The recombinant CPB2.8ΔCTE enzyme was eluted with 10-20 mM imidazole in buffer A. Purity of the enzyme was assessed using silver stained SDS-PAGE. The yield from 200 ml of culture was approximately 1-2 mg of CPB2.8ΔCTE enzyme.

Isolation of active, recombinant CPB2.8ΔCTE from the inclusion bodies

Following bacterial cell lysis as descibed above, the pelleted inclusion bodies were washed (that is, resuspended and then subsequently pelleted again via 6000 g for 10 min) once in 10 ml of 50 mM Tris/HCl buffer, pH 8 containing 5 mM EDTA, 0.1% Triton X100, twice in 10 ml of 50 mM Tris/HCl buffer, pH 8 containing 5 mM EDTA, 2 M urea, and then finally once in 10 ml of distilled, deionised water. The washed pellet was solubilised at 37°C with vigorous shaking in 10 ml of 0.1 M Tris/HCl buffer, pH 8 containing 8 M urea, 10 mM DTT. The solubilised CPB2.8 was diluted to a final protein concentration of 0.01-0.5 mg.ml^"1 in 0.1 M Tris/HCl buffer, pH 8 containing 5 mM EDTA, 8 M urea. The CPB2.8ΔCTE enzyme was then allowed to refold to native conformation by the removal of the chaotroph by dialysis for 15 hours against 100 volume of 0. IM Tris/HCl buffer, pH 7 containing 5 mM EDTA, 5 mM cysteine. The reducing agent was then removed by dialysis for 2 hours against 100 volume of the same buffer minus cysteine. The active enzyme resulting from this procedure was then purified by ion exchange chromatography. Purification of active, recombinant CPB2.8

The refolded CPB2.8ΔCTE was filtered through a 0.22 μm filter and loaded immediately at 1 ml.min^"' onto a 1 ml Mono Q column pre-equilibrated in 20 mM Tris/HCl, pH 7 , 0.01 % Triton X100 (buffer B) . The chromatography was developed at a flow rate of 1 ml. min^"1 with a stepped gradient of 0-1 M NaCl in buffer B. Pro-enzyme eluted over 200-600 mM NaCl and mature enzyme with 400-600 mM NaCl. Samples were dialysed against 20 mM Tris/HCl, pH 7, 0.01% Triton X-100 to remove salt, and frozen until use. Purity was assessed using silver stained SDS-PAGE. Yield of active, pure CPB2.8ΔCTE was 4-5 mg from 200 ml of culture.

Conversion of pro-form CPB2.8ΔCTE to mature, fully active enzyme

Peak activity fractions from the Mono Q eluant were pooled and conversion to fully active, mature enzyme facilitated by acidification using a modification of a protocol devised for the activation of recombinant cruzain (personal communication, Prof. J. McKerrow) . Enzyme (0.1- 0.5 mg.ml^"1) was incubated at 37°C for 4-8 h in the presence of 0.1 M Na acetate buffer pH 5.5 , 0.9 M NaCl, 2 mM EDTA, 10 mM DTT. Full conversion to mature form was verified by analysis on Coomassie Blue-stained SDS-PAGE, gelatin SDS- PAGE and by the large increase in activity towards n- benzoyl-pro-phe-arg-p-nitroanilide hydrochloride (BzPFRNan) . Inhibition of zymogen processing

The identity of the enzyme mediating zymogen processing during dialysis was investigated by addition of 10 μM E64 , 1 μM pepstatin or 1 mM PMSF to the dialysis buffers. Similar analysis of the protein processing subsequent to the Mono Q step involved the addition of either 10 μM E64, 1 μM pepstatin, 1 mM PMSF or aprotinin to the acidification buffer. Incubation was for 4-8 hours at 37°C and the resultant samples analysed for activity and zymogen processing by SDS-PAGE and gelatin SDS-PAGE.

Purification of mature CPB2.8ΔCTE from contaminating peptides

The final purification step involved the removal of contaminating peptides derived from the pro-region removed and digested during the full activation of CPB2.8ΔCTE. The activated sample was resolved on a 10 ml Sephadex G50 (Amersham-Pharmacia Biotech) gel filtration column at room temperature. Resin was pre-swollen and the column equilibrated in 0.1 M Na acetate pH 5.5 , 2 mM EDTA, 10 mM DTT, 0.45 M NaCl, 0.01% Triton X-100. A sample (250 μl) of activated CPB2.8ΔCTE was developed at a flow rate of 0.75 ml. min^"1 over 12 ml and 0.5 ml fractions of the eluant collected. CPB2.8ΔCTE was detected by activity towards BzPFRNan and by gelatin SDS-PAGE.

Purification of CPB2.8ΔCTE by nickel agarose affinity chromatography

CPB2.8ΔCTE was isolated from solubilised E . coli cell lysates using the denaturing method of Qiagen (Qiagen Ltd. , Crawley, UK) and nickel agarose chromatography, according to the manufacturer's instructions. Protein quantification, gel electrophoresis, enzyme assays

The protein concentration of samples was determined using the BioRad dye-binding protein assay (Bio-Rad Laboratories Ltd. , Hemel Hempstead, UK) and Pierce BCA protein assay (Pierce Chemical Co., Rockford, IL, USA). Denaturing gel electrophoresis followed the standard method of Laemmli (1970). For M_r determinations, a Benchmark protein ladder (Life Technologies, Paisley, UK) was used. Gelatin SDS-PAGE was carried out as described (Robertson & Coombs, 1990) . Activity towards BzPFRNan was assayed continuously at 37°C for 30 s at 405 nm in 400 μl of 0.1 M NaP0 pH 6.0, 10 mM DTT, 250 μM substrate following pre- incubation of the buffer for 3 min at 37°C. For kinetic analyses, approximately 1-2 μg of enzyme was assayed, in triplicate, using twelve individual substrate concentrations encompassing K_m/4 to K_m x 10. The assay was also adapted to microtitre plate use from Robertson & Coombs (1990) . Approximately l-5μl of enzyme sample was tssayed in a final volume cf 1C0 -l of 0.1 M NaP0 pH 6.0, 10 mM DTT and the reaction started with the addition of 1 mM BzPFRNan. Absorbance at 405 nm was determined immediately following incubation for 10 min at 37°C with gentle agitation. Activity was expressed as μmoles of product produced per min per 1 ml volume (s = 9.5 x 10³) . Enzyme inhibition assays were carried out in quadruplicate using the microtitre assay. Inhibitors were used at the maximum recommended working concentration in the presence of excess (1 mM) substrate.

N-terminal amino acid sequencing

CPB2.8ΔCTE (1 μg) was resolved on sequencing grade

(Fluka) 10 % (w/v) acrylamide gels crosslinked with piperazine diacrylamide (PDA, BioRad) using the discontinuous denaturing Tris/HCl buffered system of

Laemmli (1970) . Gels were pre-run at 6 ml for 1 h with 50 μM reduced glutathione in the upper reservoir buffer and the buffer replenished. Samples were dissolved in sample buffer, heated for 2 min at 100°C and run with 0.1 mM sodium thioglycolate in the upper reservoir buffer. Prior to transfer, the gel was pre-equilibrated for 20 min in transfer buffer, 10 mM 3-[cyclohexylamino] -1-propane sulphonic acid (CAPS) , 10% (v/v) methanol pH 11 (Matsuderia, 1987) . Problott transfer membrane (Applied Biosystems, Weiterstadt, Germany) was pre-wetted in 100% (v/v) methanol for 3 min and then stored in transfer buffer until use. Proteins were transferred for 30 min at room temperature and the membrane stained with Coomassie Blue stain following the manufacturer's instructions. N-terminal amino acid sequencing analysis was carried out by Dr. B. Dunbar at the Proteomics Unit, Department of Molecular & Cell Biology, University of Aberdeen.

Antibody production

Polyclonal antiserum against CPB isoenzymes isolated from L . mexicana promastigotes was raised in New Zealand White rabbits as described previously (Robertson & Coombs, 1994) . A second polyclonal antiserum was produced by the same methodology and using the recombinant pro-mature CPB2.8ΔCTE purified in this study using denaturing conditions with nickel agarose affinity chromatography.

Western Blotting

Following SDS-PAGE, proteins were transferred to Hybond

ECL nitrocellulose (Amersham-Pharmacia biotech) for 45 min at 4°C and 100 V in 20 mM Tris, 150 mM glycine, 20% (v/v) methanol. Transfer was confirmed using reversible Ponceau S stain, and the membrane blocked for 4-6 h at room temperature in 20 M Tris/HCl pH 7.5 , 15 mM NaCl, 5% (w/v) powdered milk, 0.2% (v/v) Tween 20. The membrane was probed overnight at 4°C with a 1 in 2000 dilution of anti- CPB antisera in 20 mM Tris/HCl pH 7.5 , 15 mM NaCl, 1% (w/v) powdered milk, 0.1% (v/v) Tween 20. All the subsequent steps were at room temperature. Membrane was washed for 2 h with four changes of 20 mM Tris/HCl pH 7.5 , 15 mM NaCl, 1% (w/v) powdered milk and probed with a 1 in 2000 dilution of donkey anti-rabbit horseradish peroxidase-linked antiserum for 2 h in 20 mM Tris/HCl pH 7.5, 150 mM NaCl, 1%

(w/v) powdered milk. Unbound secondary antibody was washed for 2 h with three changes of 20 mM Tris/HCl pH 7.5 , 15 mM NaCl, 1% (w/v) powdered milk and then with 20 mM Tris/HCl pH 7.5, 150 mM NaCl prior to detection using Pierce

SuperSignal chemiluminescent substrate (according to the manufacturer's instructions).

RESULTS

Expression of CPB2.8ΔCTE in Ξ . coli

Expression of CPB2.8 lacking the C-terminal domain

(CPB2.8ΔCTE) in E . coli resulted in a distinct protein of ~ 38000-M, (Figure 1A, lane 2) . This corresponds well with the predicted M_r of 37100 for the zymogen form of CPB2.8ΔCTE with an additional 930-M_r contributed by the (Histidine) ₆- tag. No expression of the protein was apparent in cells before the addition of IPTG (Figure 1A, lane 1) .

The optimum conditions for CPB2.8ΔCTE expression were determined by varying IPTG concentration, length of induction and growth temperature. Production increased when induction time was extended up to 4 h, but there was no further increase by 5 h. IPTG concentration could be decreased from 2 mM to 0.5 mM without effect on CPB2.8ΔCTE expression, but lower expression was observed when <0.5 mM IPTG was used. Under these conditions at 37°C, 90% of the expressed protein was associated with inclusion bodies and 10% with the soluble phase (Figure IB, compare lanes 1-3) . Reduction of growth temperature to 18 °C pre and/or post- induction only increased the relative levels of soluble enzyme to 20% and reduced the overall yield, thus enforcing purification of CPB2.8ΔCTE from inclusion bodies. Purification of recombinant proteins from inclusion bodies has certain advantages including the relatively large quantities of starting material, the simplicity of removing inclusion bodies from other cell material and the minimising of problems from proteolysis due to E . coli proteinases.

Purification and refolding of CPB2.8ΔCTE from inclusion bodies

CPB2.8ΔCTE was expressed as a (His)₆-tag fusion with the intention of achieving rapid and efficient purification on nickel agarose resin using the methods of Qiagen. Accordingly, initial attempts to purify CPB2.8ΔCTE from inclusion bodies used this method under denaturing conditions. Cells were lysed in either 6 M guanidine hydrochloride or 8 M urea in 0.1 M NaP0 , 0.01 M Tris/HCl pH

8, stirred for 1 h at 18°C and the recovered supernatant loaded onto a 1 ml nickel agarose column. A series of 8 M urea buffers of decreasing pH were used to remove unbound contaminants and finally, to elute the (His) ₆-tagged CPB2.8ΔCTE. Pure (His) _β-tagged pro-CPB2.8ΔCTE eluted at pH 4.5 (Figure IC) , and this was employed in the production of specific antisera. However, even though the preparation had some activity towards gelatin it was inactive towards N- benzoyl-pro-phe-arg-p-nitroanilide hydrochloride and all attempts to convert this pro-form to mature, fully active enzyme were unsuccessful. Alternative lysis, solubilisation and elution techniques were investigated in conjunction with purification by nickel agarose, but with no increased success . Consequently, an alternative method of avoiding the procedures involved in the use of the nickel agarose chromatography step was adopted (in the belief that these adversely affected the enzyme) . The inclusion body phase was isolated, washed and solubilised in 8 M urea using a method adapted from Kuhelj et al . (1995). EDTA was present in all wash buffers to chelate any heavy metal contamination which could inactivate the enzyme. The Triton X-100 and 2 M urea washes were functional in the removal of contaminants, as previously reported (Babbitt et al . , 1990), and resulted in a significant increase in purity (Figure IB, compare lanes 3 and 4) . At this stage, the majority of solubilised CPB2.8ΔCTE was still in the pro-mature form as judged by its relative mobility on SDS- PAGE (Figure IB, lane 4). However, during dialysis approximately 60% of the protein was converted to a faster mobility form with a M_r of 27000 (Figure IB, lane 5) . Both the 38000 and 27000-M_r species reacted with antisera raised against anti-CPBs antisera (Figure ID, lane 5) . Analysis of this post-dialysis sample revealed that it had activity towards both gelatin, with two bands of activity post- electrophoresis of similar mobilities to those of the post Mono Q sample (Figure 2), and N-benzoyl-pro-phe-arg-p- nitroanilide hydrochloride (Table 1) . The estimated M_r of 27000 from SDS-PAGE, was higher than the predicted 23300-M_r for mature CPB2.8ΔCTE. Aberrant mobility on SDS-PAGE has been observed for pro- and mature cathepsin L and cruzain, the major cysteine proteinase of T . cruzi (Eakin et al . , 1992; Menard et al . , 1998) and can result from a preponderance of certain amino acids. Support for 27000-M_r form representing the mature enzyme was that the native, mature form of CPB2.8ΔCTE partially purified from L . mexicana promastigotes co-migrated on SDS-PAGE and zymograms with the 27000-M_r recombinant species (Figure 2) .

Table 1 : Purification summary

1: based on total protein

2: using continuous assay with N-benzoyl-pro-phe-arg-p- nitroanilide as substrate

The optimum conditions for the dialysis step were investigated. Maximum recovery of soluble enzyme was dependent upon low (i.e., 0.1-0.5 mg.ml^"1) protein concentrations. Smaller dilutions lead to an increase in particulate material and considerably reduced yields, probably resulting from incorrect pairing due to molecular crowding. Consequently, approximately 0.1 mg.ml^"1 was used routinely. The best pH of the dialysis buffer was investigated. In an attempt to mimic conditions in which the enzyme might be expected to have optimum catalytic activity, and so autohydrolysis should be maximal, dialysis was performed at both pH 6 (in 0.1 M Na acetate buffer, 5 mM EDTA, 5 mM cysteine) and pH 5.5 (in 20 mM L-histidine, 150 mM NaCl, 5 mM EDTA, 5 mM cysteine buffer) . However, substitution of the standard Tris/HCl buffer conditions with either of the above buffers lead to increased levels of precipitation but with no better conversion efficiency. Dialysis against Tris/HCl pH 7 , 5 mM EDTA, 5 mM cysteine overnight followed by dialysis for 2 h against either the L-histidine or Na acetate buffers also lead to precipitation, losses in yield and no increase in conversion. Dialysis at pH 8.5 resulted in no precipitation. However the majority of the CPB2.8ΔCTE remained as the 38,000-M_r pro-form and could not be converted to mature enzyme. Similarly, the optimum pH for solubilisation of the inclusion bodies was found to be 8.0. The use of acidic pHs resulted in poorer solubilisation and subsequent conversion to mature enzyme. The pH optima for CPB2.8ΔCTE solubilisation and refolding closely resemble those for human cathepsin B (Kuhelj et al . , 1995).

Following dialysis, the partially purified CPB2.8ΔCTE was further purified by anion exchange chromatography using a 1 ml Mono Q column and elution with a 0-1 M NaCl linear gradient. Both the pro-mature and mature proteins eluted between 250 mM and 600 mM NaCl, with the majority of enzyme activity eluting at 350-500 mM NaCl (results not shown) . Analysis of the pooled peak fractions by Coomassie stained SDS-PAGE (Figure IB, lane 6) revealed the presence of four major bands of 38000-M_r (designated pro-mature) , 34000-M_r (which was shown to represent a processing intermediate, see later) , 27000-M_r (designated mature) , and <10000-M_r (thought to be pro-region peptides) . The conversion of the 38000-M_r form to one of 34000 was also shown by Western blotting (Figure ID) . Two bands of activity were observed on gelatin gels (Figure 2, lane 1). The upper, lower mobility band on occasion appeared to comprise two closely migrating activities. The specific activity of this sample towards N-benzoyl-pro-phe-arg-p-nitroanilide hydrochloride was 36.2 nmoles. min^"' .mg^"', reflecting both a purification and some conversion resulting from the Mono Q step (Table 1). Activation of CPB2.8ΔCTE

The acidification protocol developed for the conversion of the 38000-M_r and 34000-M _r species to the M _r 27000 form proved highly effective, as confirmed by analysis of the samples post-acidification by SDS-PAGE (Figure IB, lane 7) and Western blotting (Figure ID, lane 7) . The conversion was also confirmed using gelatin gels which showed that there had been almost total conversion to an apparently single fast mobility form (Figure 2, lane 2) . Conversion was also accompanied by a significant increase in activity towards N-benzoyl-pro-phe-arg-p-nitroanilide hydrochloride (Table 1) and by the proteolysis of minor contaminants including further proteolysis of the released pro-region peptides (Figure IB, lane 7) .

Activation to the mature enzyme appeared to be affected by the concentration of the CPB2.8ΔCTE sample. The optimum range was 0.1-0.5 mg.ml^"', although conversion was observed with 5-fold lower protein concentrations. In contrast, protein concentrations >2 mg.ml^"" often resulted in protein precipitation. Conversion was also dependent on acidic conditions, which is consistent with autohydrolysis by the enzyme that occurs in the acidic lysosomal compartment of the parasite. Complete conversion occurred within 8 h at 37 °C and pH 5.5, it took longer at pH 5.0 and pH 6.0, whereas at pH 7.0 none was observed within 8 h. At pH 4.0, there was a very rapid initial activation but no more occurred after 10 min. A similar phenomenon was observed for mouse cathepsin L activation (Mason and Massey, 1992) and may reflect enzyme instability at this low pH. Conversion was considerably slower at temperatures below 37°C (results not shown) . Other important factors for successful conversion were the presence of reducing conditions and high ionic strength. Omission of one or both of NaCl and DTT resulted in slower conversion, it taking 8 h rather than the normal 2-4 h. NaCl was already present in the post-Mono Q fractions, and so it is not possible to determine whether activation is absolutely dependent on NaCl being present but clearly additional NaCl is beneficial.

The Mono Q step contributed to the continued purification of CPB2.8ΔCTE but it also appeared to be vital for complete conversion to active, mature enzyme. Attempts to activate the CPB2.8ΔCTE sample prior to resolution on Mono Q were largely unsuccessful, with there being little or no conversion for the pre-Mono Q sample over 4-8 h whereas the post-Mono Q sample was completely converted in this time.

Analysis of the N-terminal amino acid sequences of the various protein bands gave insight into the stages occurring during the processing (Figure 3). The 38000-M_r species was shown to indeed be the pro-mature protein, the 34000-M_r species which appeared post-Mono Q was found to have had part of the pro-region removed, and the active enzyme produced by activation of the post-Mono Q sample was shown to have the identical N-terminus as native CPB enzyme, confirming that the whole pro-domain had been removed. However, the apparent 27000-M_r protein in the pre- Mono Q sample was found to have an additional seven amino acid residues on its N-terminus compared with the truly mature enzyme. This difference correlates with the relatively lack of activity of this protein towards N- benzoyl-pro-phe-arg-p-nitroanilide hydrochloride (Table 1) .

Analysis of zymogen processing

Inhibitors were used in attempts to identify the enzymes responsible for the various processing steps and to determine whether or not they are mediated by CPB2.8ΔCTE itself. The initial conversion of approximately 60% of the population of (His) ₆-tagged pro-CPB2.8ΔCTE (38000-M_r species) to the 27000-M_r form that occurred during the dialysis step was not affected by inhibitors directed against cysteine proteinases (E64), aspartic proteinases (pepstatin) or serine proteinases (PMSF) (Figure 4) . However, a second Triton wash during inclusion body isolation completely abolished the conversion of CPB2.8ΔCTE to the 27000-M_r form during dialysis, suggesting that this removed the mediator of the initial processing. Zymogen processing during the acidification step was also analysed. Inhibition of activation and complete abolition of CPB2.8ΔCTE activity itself (as observed on gelatin gels and by assay) was effected by E64 , but not by pepstatin or PMSF. This result indicates that the final processing can be an auto-catalytic event.

Removal of pro-region peptides

The final step in the purification of CPB2.8ΔCTE involved the removal by gel filtration of peptides resulting from removal and partial proteolysis of the pro- region. This resulted in an increase in specific activity (Table 1) . Analysis by silver stained or Coomassie stained SDS-PAGE revealed a major band of approximate M_r 27000 (Figure IE) . Occasionally, traces of pro-region peptides co-eluted with the mature enzyme possible due to incomplete conversion prior to gel filtration or continued association of the clipped pro-region peptides with the mature enzyme.

Characterisation of CPB2.8ΔCTE activity

The activity of the fully-processed and therefore mature CPB2.8ΔCTE was analysed kinetically and with respect to inhibitor sensitivity. Whilst pepstatin and PMSF were ineffectual in inhibiting CPB2.8ΔCTE, compounds known to be specific inhibitors of cysteine proteinases were effective against CPB2.8ΔCTE (Table 2). The kinetic parameters of the mature enzyme were similar to those reported for cruzain, a homologous enzyme from T . cruzi (Table 3). Chemical inactivation of CPB2.8ΔCTE

2 mg of CPB2.8ΔCTE (specific activity 600nmoles/min/mg) was inactivated by incubation with lOμM of the irreversible cysteine proteinase inhibitor E64 (trans-epoxysuccinyl-L- leucylamido-(4-guanidino) butane) for 1 hr at 37°C. Unbound E64 was removed by dialysis against 100 volumes (2 times 12 hr) Phosphate Buffered Saline. No cysteine proteinase activity was detected in the preparation of E64-treated CPB2.8ΔCTE.

Expression of CPB2.8CTE

The full length cpb2 . 8 gene, including the C-trminal extension (CTE) , was cloned into the pGE30 vector as follows. Plasmid pGL393, which contains the full length cpb2 . 8 gene (Mottram et al., 1996), was digested with .Kp.nl and Hindi11 and the 2kb DNA fragment isolated. This was cloned into Kpnl /Hindlll digested pGL280 to give plasmid pGL394, pGL394 was used to transform E . coli strain M15[pREP4] and expression and purification of CPB2.8CTE carried out as described for CPB2.8ΔCTE.

Table 2: Effect of protease inhibitors on CPB2.8ΔCTE activity

The assays were performed in quadruplicate as described in Materials & Methods using ImM N-benzoyl-pro-phe-arg-p- nitro-anilide as substrate. Enzyme (^~2μg) was pre- incubated in buffer ± inhibitor for 10 mins at room temperature before assay initiated by addition of substrate. Inhibitors were used at recommended effective concentrations (Beynon & Salverson, 1990) .

Table 3 : Kinetic parameters for recombinant enzymes using substrate, N-benzoyl-pro-phe-arg-p-nitroanilide ^a data published in Eakin et al . (1992) J . Biol . Chem

267: 7411-7420 ^b assuming M_r 23300 for mature, active CPB2.8ΔCTE Example 2

Production of recombinant proteins of mutated cpj as vaccine candidates

The genes encoding different copies of CPB were mutated in a number of ways so that variants of the protein could be generated as candidate vaccines.

Mutations were incorporated into the pBluescript constructs of the cDNA (Souza et al , 1992) and cpb2 .8 (Mottram et al , 1996) using the QuikChange Site-Directed mutagenesis kit (Stratagene) and verified by sequence analysis. Primers used were as follows (only the sense strand primer is shown and the mutated sites are underlined) :

(a) CPB2.8 mut ASM, GGTGCGTGCGGGTCGGCTGGGCGTTCTCGG

(b) CPB2.8 mut glycos, GCCCGAGTGCTCGAGCAGCAGTGAGCTCG

(c) cDNA mut 18, GACGCCGGTGAAGAATCAGGGTGCGTG

(d) cDNA mut 84, CGAACGGGCACCTGTACACGGAGGACAGC (e) cDNAmutx3 , GCTGCGATGACATGAACGATGGTTGCGACGGCGGGCTGATGC

Mutation

(a) Residue 25 of mature form CPB2.8 from a cysteine to a glycine. This mutation removes the active site cysteine. (b) Residue 103 of mature form CPB2.8 from an asparagine to a serine. This mutation removes the glycosylation site.

(c) Residue 18 of mature form CPB cDNA (Souza, et al 1992) from an aspartic acid to an asparagine, alters activity of

CPB. (d) Residue 84 of mature form CPB cDNA from a histidine to a tyrosine. Alters activity of CPB.

(e) Residues 60, 61 and 64 of CPB cDNA from aspartic acid (60), asparagine (61) and serine (64) to asparagine, aspartic acid and aspartic acid respectively. Alters activity of CPB cDNA. All sequencing was performed with an ABI 373 DNA sequencer (Perkin-Elmer) and analysed using the Wisconsin GCG package.

Example 3

Isolation of Leishmania infantum cpa and cpb genes

In Mediterranean countries, L . infantum is responsible for both visceral and cutaneous leishmaniasis with the dog serving as the principal reservoir. In Spain, for example, there is a prevalence of 0.3 Visceral Leishmaniasis cases per 100,000 inhabitants and it is calculated that between 3% and 5% of all Spanish dogs are seropositive.

The L . infantum cpa and cpb genes were isolated by screening a L . infantum genomic library (Soto et ai . , 1993) with L . mexicana cpa (Mottram et al . , 1992) and cpb (Souza et al . , 1992) -specific gene probes. The genomic library was prepared in the EMBL3 lambda vector from total DNA partially digested with Sau3Al from a L . infantum strain from a case of nαπan Visceral Leishmaniasis (reference strain, MHOM/FR/78/LEM-75) . 3 lambda clones were isolated for L . infantum cpa and 4 lambda clones were isolated for L . infantum cpb .

The complete open reading frame of the L . infantum cpa gene, together with 5' and 3' flanking sequence, was subcloned on a 3.2 kb Pstl fragment into pBluescript vector for sequencing (partial sequence shown in Figure 5) . The L . infantum cpa gene has 89% nucleotide sequence identity with the L . mexicana cpa gene (Mottram et al . , 1992) and 92% identity with the L . chagasi cpa gene (O ara-Opyene and Gedamu, 1997) . The predicted L . infantum protein (Figure 6) has 86% amino acid sequence identity with L . mexicana CPA and 89% identity with the L . chagasi CPA (Omara-Opyene and Gedamu, 1997) . Outwith the open reading frame of the cpa genes, there is sequence variation in the 5' and 3' flanks that will allow the design of specific gene targeting fragments for deletion of the L . infantum cpa gene .

A PCR approach was used to amplify the complete ORF of a cpb gene from one of the L . infantum lambda clones containing cpj . Sense and antisense primers were designed based on the L . mexicana and L . chagasi cysteine proteinase sequences. The sequences of the two primers for cpb were: sense primer (CPBM1) 5' GTGCGAGCTGTGGCCTCTGCGT 3' and antisense primer (CPBM2), 5' GGCGCGCGCGCACCCAAGG 3'.

PCR reactions were carried out using DNA from lambda as template, 5% DMSO and Taq and KlenTaq LA polymerase (Sigma). Conditions used in PCR were 94° 5', 15 cycles (94° 1', 45° 2', 72° 2') and final extension 72° 5'. The temperature for the extension was 68° when the PCR used KlenTaq LA polymerase. 1.7 kb cpb PCR products were cloned into the pGEM-T vector. Sequence from the L . infantum cpb PCR products (Figures 7a and 7b) showed >85% identity with the L . mexicana cpb gene.

L . infantum CPB (Figures 8a and 8b) had 84% amino acid sequence identity with L . mexicana CPB2.8 (Figure 9). Figure 10 shows a pile-up of CPB sequences from L . infantum , L . mexicana , L . pifanoi and L . major .

References

Afonso, L.C.C. & Scott, P. (1993). Immune responses associates with suspectibility of C57BL10 mice to

Leishmania amazonensis . Infect. Immun. 61, 2952-2959.

Alexander, J & Russell, D.G. (1992) . The interaction of

Leishmania species with macrophages. Adv. Parasitol. 31,

175-254.

Babbitt, P.C., West, B.L., Buechter, D.D., Kuntz , I.D. &

Kenyon, G.L. (1990) Bio/Technology, 8, 945-949.

Barrett, A.J. and Rawlings, N.D. (1996) Families and clans of cysteine peptidases. Perspect. Drug Disc. Design 6, 1-

11.

Bart, C. , Coombs, G.H., & Mottram, J.C. (1995). Isolation of lmcpc, a gene encoding a Leishmania mexicana cathepsin

B- like cysteine proteinase. Mol. Biochem. Parasitol. 73,

271-274.

Beyer, B.M. & Dunn, B.M. (1996) J. Biol. Chem. 271,

15590-15596.

Bogden, C. , Geesner, A, Solbach, W. & Rollinghof, M.

(1996) Invasion, control and persistence of Leishmania parasites. Curr. Opinion Immunol. 8, 517-525.

Bray, R.S. (1982) The zoonotic potential of reservoirs of leishmaniasis in the Old World. Ecol. Disease 1, 257-267.

Coombs, G.H. & Mottram, J.C. (1997). Proteinases of trypanosomes and Leishmania. In trypanosomiasis and leishmaniasis: Biology and control. G. Hide. GC Mottram,

GH Coombs and PH Holmes, eds . (Oxford, UK: CAB

International). Pp 177-197.

Cigic, B., Krizaj , I., Kraij , B., Turk, V., Pain, R.H.

(1998) Biochem. Biophys Acta 1382, 143-150.

Chaudhuri, J.B. (1988) Annals of the New York Academy of

Sciences: 374-385.

Coulombe, R. , Grochulski, P., Sivaraman, J. , Menard, R. ,

Mort, J.S. & Cygler, M. (1996) EMBO J. 15, 5492-5503.

Cygler, M. , Sivaraman, J. , Grochulski, P., Coulombe, R. , Storer, A. & Mort, J.S. (1996) Structure 4, 405-416. De Souza Laeo, S., Lang, T. , Prina, E. , Hellio, R. & Antoine, J.C. (1995) Intracellular Leishmania amazonensis internalise and degrade MHC class II molecules of their host cells. Journal of Cell Science 108: 3219-3231.

Dye, C, Killick-Kendrick, R. , Vitutia M.M., Walton, R. , Killick-Kendrick, M. , Harith, A.E. Guy., M.W., Canavate, MC. & Hasibeder, G (1992) Epidemiology of canine leishmaniasis: prevalance, incidence and basic reproduction number calculated from a cross sectional serological survey on the island of Gozo, Malta. Parasitology 105, 35-41. Eakin, A.E., Mills, A. A., Harth, G. , McKerrow, J.H. & Craik, C.S. (1992) Journal of Biological Chemistry 267, 7411- 7420. Eakin, A.E., McGrath, M.E., McKerrow, J.H., Fletterick,

R.J. & Craik, C.S. (1993) Journal of Biological Chemistry

268, 6115-611.

Fancy, D. & Kodadek, T (1997) Tetrahedron 53, 11953-

11960. Fancy, D., Melcher, K. , Johnston, S.A. & Kodadek, T. (1996) Chemistry & Biology 3, 551-559.

Finkeleman, F.D., Holmes, J., Katona, I.M., Urban J.F., Beckman, M.P. Park, L.S., Schooley, K.A. , Coffman, R.L., Mossman, TR. , & Paul, WE. (1990). Lymphokine control of in vivo immunoglobulin isotype selection. Ann. Rev. Immunol. 8, 303-333.

Glew, R.H., Saha, A.K., Das, S. & Remaley, A.T. (1988) Microbiology Reviews 52, 412-432. Ishidoh, K. & Kominami, E. (1994) FEBS Lett. 352, 281- 284.

Jaenicke, R. & Rudolph, R. (1990) "Folding Proteins" in Protein Structure — a practical approach (ED. TE Creighton) , pp.191-223. IRL Press. Kaye, P.M., Curry, A.J., & Blackwell, J.M. (1991). Different production of Thl-and Th2-derived cytokines does not determine the genetically controlled or vaccine induced rate of cure in murine visceral leishmaniasis. J. Immuol. 146, 1763-2770.

Kreig A.M. (1996) Trends in Microbiol. 4 pp. 73-77.

Lebowitz, J.H. (1994) Transfection experiments with leishmania. Meth Cell Biol. 45, 65-78. Kuhelj, R. , Dolinar, M. , Pungercar, J. & Turk, V. (1995)

European Journal of Biochemistry 229, 535-539.

Laemmli, U.K. (1970) Nature 277, 680.

LeBowitz, J.H., Coburn, CM., McMahon-Pratt , D., and

Beverley, S.M. (1190). Development of a stable Leishmania expression vector and application to the study of parasite surface-antigen genes. Proc. Natl. Acad. Sci. USA 87,

9736-9740.

Liew, F.Y. & O'Donnel, CA (1993). Immunology of leishmanisis . Adv. Parasitol. 32, 161-258. Mach, L. , Mort, J.S. & Gloss, J. (1994) Journal of

Biological Chemistry 269, 13030-13035.

Matsudaira, P. (1987) Journal of Biological Chemistry

262, 10035-10038.

Mason, R.W. & Massey, S.D. (1992) Biochem. Biophys. Research Co m. 189, 1659-1666.

Menard, R. , Carmona, E., Takebe, S., Dufour, E., Plouffe,

C, Mason, R.W. & Mort, J.S. (1988) Journal of Biological

Chemistry 273, 4478-4484.

McDonald, J.K., Ellis, S. & Reilly, T.J. (1966) Journal of Biological Chemistry 241, 1494-1501.

McDonald, J.K. & Emerick, J.M.C. (1995) Arch. Biochem.

Biophys. 323, 409-422.

Mclntyre, G.F. & Erickson, A.H. (1993) Proceedings of the

National Academy of Science. USA 90, 10588-10592. Mottram, J.C, Frame, M.J., Brooks, D.R., Tetley, L. ,

Hutchison, J.E., Souza, A.E. & Coombs, G.H. (1997)

Journal of Biological Chemistry 272, 14285-14293.

Mottram, J.C, Robertson, CD., Coombs, G.H & Barry, JD

(1992). A developmentally regulated cysteine proteinase gene of Leishmania mexicana . Mol. Microbiol. 6, 1925-

1932.

Mottram J . C , Souza , A . E . , Hutchinson , J . E . , Carter , R, Frame M.J. & Coombs, G.H. (1996) . Evidence from disruption of the lmcpb gene array of Leishmania mexicana that cysteine proteinases are virulence factors. Proc Natl.

Acad. Sci. US 93,6008-6013. Noben-Trauth, N. , Kropf, P & Muller, I (1996).

Suspceptibility to Leishmania major infection in interleukin-4-deficient mice. Science 271, 987-990.

North, M.J. & Coombs, G.H. (1981) Mol. Biochem. Parasitol.

3, 293-300. North, M.J., Mottram, J.C. & Coombs, G.H. (1990)

Parasitology. 6, 270-275.

Omara-Opyene, A.L. and Gedamu, L. (1997) . Molecular cloning, characterization and overexpression of two distinct cysteine protease cDNAs from Leishmania donovani chagasi . Mol. Biochem. Parasitol. 90, 247-267.

Pamer, E.G., Davis, C.E. & So, N. (1991) Infect. Immun.

59, 1074-1078.

Roberts, M. , Alexander, J. & Blackwell, J.M. (1989).

Influence of Lsh, H-2 and an H-ll linked gene on visceralization and metastasis associated with Leishmania mexicana infection in mice. Infect Immun. 57, 875-881.

Roberts, M. , Alexander J. & Blackwell JM. (1990). Genetic analysis of Leishmania mexicana infection in mice: single gene (scl-2) controlled predisposition to cutaneous lesion development. J. Imunogen. 17, 89-100.

Robertson, CD, Coombs, G.H., North. M.J & Mottram, J.C.

(1996) . Parasite cysteine proteinases Perspect. Drug

Disc. Design 6, 99-118.

Robertson, CD. & Coombs, G.H.. (1994) Microbiology, 140, 417-424.

Robertson, CD. & Coombs, G.H.. (1990) Characterisation of three groups of cysteine proteinases in the amastigotes of

Leishmania mexicana mexicana . Molecular & Biochemical

Parasitology, 42, 269-276. Rowan, A.D., Mason, P., Mach, L. & Mort, J.S. (1992)

Journal of Biological Chemistry 267, 15993-15999.

Sato Y., et al. I munostimultary DNA sequences necessary for effective intradermal gene immunization. Science

(1996) 273, 352-354) .

Satoskar, A. & Alexander, J, (1995). Sex mediated suspectibility and differential TNF-∞ and IFN-δ mRNA expression in DBA/ 2 mice infected with Leishmania mexicana . Immunology 84, 1-4.

Satoskar, A., Bluethmann, H & Alexander, J. (1995).

Disruption of the murine interleukin-4 gene inhibits disease progression during Leishmania mexicana infection but does not increase control of L. donavani infection.

Infect. Immun. 63, 4894-4899.

Selzer, P.M., Chen, X., Chan, V.J., Cheng, M. , Kenyon,

G.L., Kunty, I.D., Sakanari, J.A., Cohen, F.E. & McKerrow,

J.H. (1997) Exp. Parasitology 87, 212-221. Smith, S.M. & Gottesman, M.M. (1989) Journal of Biological

Chemistry 264, 20487-20495.

Soto, M. , Requena, J.M., Garcia, M. , Gomez, L.C,

Navarrete, I., and Alonso, C (1993). Genomic organization and expression of two independent gene arrays coding for two antigenic acidic ribosomal proteins of Leishmania . J .

Biol. Chem. 268, 21835-21843.

Souza, A.E., Bates P. A., Coombs G.H. & Mottram JC (1994).

Null mutants for the impca cysteine proteinase gene in

Leishmania mexicana . Mol Biochem. Parasitol. 63, 213-220. Souza, A.E., Waugh, S., Coombs, G.H., & Mottram, J.C

(1992). Characterization of a multicopy gene for a major stage-specific cysteine proteinase of Leishmania mexicana .

FEBS Lett. 311, 124-127.

Titus, R.G., Gueriros-Filho, F.J., DeFreitas, LAR. , and Beverely, SM. (1995) Development of a safe line Leishmania vaccine line by gene replacement. Proc. Natt. Acad. Scie.

USA 92, 10267-10271.

Wittlin, S., Rosel, J. & Stover, D.R. (1998) Eur. J.

Biochem. 252, 530-536. Wolfram, M. , Ilg, T., Mottram, J.C. & Overath, P. (1995)

European Journal of Immunology 25, 1094-1100.

"Protein Particle vaccines against malaria", Gilbert, S.C. and Hill, A.V.S. Parasitology Today 13, 302 - 306, 1997. Gene Vaccination: Plasmid DNA is more than just a blueprint, Tighe, H. , Carr, M. , Roman, M. , Raz, E. Immunology Today 19, 89 - 97, 1998.

Claims

1. Use of at least one Leishmania cysteine proteinase in the preparation of a vaccine.

2. Use according to claim 1 wherein said cysteine proteinase(s) is/are active.

3. Use according to claim 1 wherein said cysteine proteinase(s) is/are substantially inactive.

4. Use according to any preceding claim wherein said cysteine proteinase (s) is/are provided as a recombinant protein.

5. A vaccine comprising at least one Leishmania cysteine proteinase.

6. A vaccine according to claim 5 wherein said cysteine proteinase (s) is/are active and/or substantially inactive.

7. A vaccine according to claim 6 wherein said active and/or inactive cysteine proteinase (s) is/are provided as a recombinant protein.

8. A vaccine according to claim 7 wherein the recombinant protein is obtainable from a defective cysteine proteinase gene.

9. A vaccine according to either of claims 6 or 7 wherein the cysteine proteinase is substantially inactive and has been rendered inactive by use of an irreversible cysteine proteinase inhibitor.

10. A vaccine according to any one of claims 6 - 8 wherein the cysteine proteinase is substantially inactive and has been rendered inactive by site directed mutagenesis.

11. A vaccine according to claim 10 wherein the site directed mutageneses replaces or removes the active site cysteine residue.

12. A vaccine according to any one of claims 6 - 11 wherein the cysteine proteinase lacks a C-terminal extension.

13. A vaccine according to any one of claims 6 - 11 wherein the cysteine proteinase includes a C-terminal extension.

14. A vaccine comprising nucleic acid capable of expressing at least one Leishmania cysteine proteinase.

15. A vaccine according to claim 14 wherein the cysteine proteinase (s) is/are active and/or substantially inactive.

16. A vaccine according to any one of claims 6 - 15 wherein the cysteine proteinase (s) is/are from L. braziliensis , L. peruviana, L. guyanensis , L. mexicana, L. major, L. amazonensis , L. infantum, L. chagasi or L. donovani .

17. A vaccine according to any one of claims 6 - 15 wherein the cysteine proteinase (s) is/are selected from CPA, CPB and CPC of L. mexicana.

18. A vaccine according to any one of claims 6 - 15 wherein the cysteine proteinase (s) is/are selected from CPA and CPB of L. infantum.

19. A vaccine according to claim 18 wherein the amino acid sequences of CPA and CPB are substantially as shown in Figures 6 and 8b respectively.

20. A vaccine according to claim 18 wherein the CPA and CPB are obtainable from the nucleic acid sequence shown in Figures 5 and 7b respectively.

21. A vaccine according to any one of claims 6 - 20 further comprising an adjuvant.

22. A vaccine according to claim 21 wherein the adjuvant is selected from Freund ' s complete adjuvant, Freund ' s incomplete adjuvant, liposomes, nisomes, mineral and non- mineral oil-based water-in-oil emulsion adjuvants, cytokines and short immunostimultaroy polynucleotide sequences .

23. CPA protein of L . infantum comprising the amino acid sequence substantially as shown in Figure 6.

24. CPB protein of L . infantum comprising the amino acid sequence substantially/ as shewn in Figure 8b.

25. cpa gene of L . infantum comprising the nucleic acid sequence substantially as shown in Figure 5.

26. cpjb gene of L . infantum comprising the nucleic acid sequence substantially as shown in Figure 7b.

27. CPA or CPB at least 90% identical to the sequences shown in Figure 6 and 8b respectively.

28. cpa or cpb at least 90% identical to the sequences shown in Figure 5 and 7b respectively.

29. Use of the CPA and/or CPB proteins or cpa and/or cpb nucleic acid sequences of any one of claims 23 - 28 in the preparation of a vaccine.