Heterologous protection in murine cutaneous leishmaniasis by G. F. Mitchell and E. Handman. (From the Walter and Eliza Hal! Institute of Medical Research, ...
Immunol. Cell Biol., 65 (Pt. 5) 387-392 (1987)
© Heterologous protection in murine cutaneous leishmaniasis by G. F. Mitchell and E. Handman (From the Walter and Eliza Hal! Institute of Medical Research, Melbourne, Victoria 3050, Australia.) Submitted June 18. 1987. Accepted for pubiication July 2, 1987.) Summary. Mice immunized with a glycolipid aniigen (GL) of Leishmania major plus adjuvant are relatively resistanl to subsequent infection with ihis protozoan parasite. The GL is affinity purified on the monoclonal anlibody WIC-79,3 which is L. major-specific and does not react with L. donovani. When another monoclonal, WIC-108.3, which cross-reacts with several Leishmania species, is used to affinity purify GL from L. donovani, the eluted materiai can partially protect genetically resistant mice against L. major. Thus, GL cross-reactions may in part underlie the known protective effects of crude L. donovani antigens against L. miT/or infection. Experiments with live parasites of the/.. ma/or isolate LRC-L119, that is nonpathogenic in mice, that does not survive long in macrophages in vitro, and that has not been shown to contain any WIC-79.3 reactive GL, indicated that this isolate wilt very effectively protect mice againsi subsequent infection. This raises the possibility that GL is only one of at least two different classes of vaccinating antigen capable of protectively immunizing mice in this cutaneous leishmaniasis model.
INTRODUCTION In recent years, several means of vaccinating mice against chronic infection with the intramacrophage protozoan parasite Leishmania major have been developed (1). These include injection of irradiated or disrupted promastigotes, living avirulent promastigotes, disrupted infected macrophages plus the adjuvant Corynebacterium parvum and, as demonstrated recently (2) an isolated glycolipid (GL) of promastigotes. With al! these vaccination approaches, success rates can be high, particularly in mice that are not at the extreme of genetic susceptibility represented by BALB/c. However, variability between experiments can also be high (3) and the severest constraint is imposed by route of injection, intravenous or intraperitoneal administration of the experimental vaccine being required. The reasons for this are obscure. In terms of human vaccine development, interest naturally focuses on a molecularly-defined vaccine that is invariably Abbreviations used in this paper: LmGL, glycolipid (GL) antigen of L. major isolated by the anti-Z.. major monoclonal antibody WIC-79.3; LdGL, a GL antigen of L. donovani isolated by the cross-reactive anti-/.,. major monoclonal antibody WIC-108.3; IP, intraperitoneally; SC. subcutaneously; PBS. mouse tonicity phosphatebuffered saline pH7 3.
effective when given via an acceptable injection route and with an acceptable adjuvant and vehicle. The GL antigen used in the studies of Handman and colleagues (2; 4-7) is purified from detergent lysates of promastigotes of the L. mtr/or cloned line V121 using immobilized monoclonal antibody, WIC-79.3. Whilst vaccination efficacy of this preparation (termed LmGL) when injected intraperitoneally with C. parvum can be impressive, a carbohydrate derivative of LmGL that still binds WIC-79.3 can increase the severity of disease in mice (4; 8). Related to this may be the well established observation that L3T4 "*" Ly2 ~ T cells can promote disease as well as resistance to disease in mice (1; 7; 9; 10). Regardless of the mechanisms of disease promotion in this infection, the available data from experiments in mice indicate that LmGL as presently isolated from L. major promastigotes may be unacceptable for use in humans, even though at present there is no evidence for a similar effect in man. Therefore, efforts have been directed to alternative methods of isolation (McConville et ai., unpublished data) and, more relevant to this paper, alternative forms of LmGL or other antigens that may or may not cross-react with LmGL. As a first step in this process, we have
388
G. F. MITCHELL .AND E . H A N D M A N
tested the vaccinating effect of a serologically cross-reactive GL of Leishmania donovani (i.e. LdGL) against infection with L major. The anti-L. major monoclonal WlC-108.3 which binds this aniigen was used for its affinity purification. In addition, a non-pathogenic L. major organism LRC-Ln9 that from several immunochemicai analyses appears to lack LmGL (5) has been used in vaccination studies in mice. The data indicate that LdGL preparations and, more particularly L119, contain antigens that can protect mice against subsequent challenge with L. major promastigotes.
the detergent-soluble preparation loaded on colutntis containing the anti-t. wo/or-specinc monoclonal antibody WlC-79.3 and the L. major and L. donovani cross-reactive monoclonal WIC-tO8.3 (13-15). The bound glycolipids were eluted from the columns with 6 M guanidine-HCl in the absence of detergent and immediately dialysed in detergent-free PBS. The carbohydrate content of the glycoiipids was measured using ihe phenol-sulphuric acid method of Dubois-Gillies (16). Glycotipids were stored at -20° until used. Killed Corynebacterium parvum was obtained from Wellcome, Beckenham, Kent, and diluted in PBS or normal saline (as were vaccinating parasites and antigens) prior to intraperitoneal injection in doses of 100 to 200 ng per mouse.
MATERIALS AND METHODS
Vaccination of mice against L. major using the non-pathogenic organism LU9. In two separate experiments, female BALB/K and C57BL/6 mice were injected intraperitoneally with living L119 on three occasions without adjuvant and challenged cutaneously with V12I. A high degree of resistance was demonstrated in C57BL/6 mice (more than half the mice in two experiments showing no lesions), with a lower though still significant degree of resistance being shown by the genetically susceptible BALB/K mice (Fig. 1). In one of these experiments the vaccinating effect of three injections of L1I9 was better than three injections of the nonpathogenic cloned line A12 derived from LRCL137 (3). In two other experiments, BALB/K or the more susceptible strain, BALB/B, were injected with either living or killed L119 with or without the adjuvant C parvum, by either intraperitoneal (IP) or subcutaneous (SC) routes. Results presented in Fig. 2 establish that living LU9 plus C. parvum injected IP was highly effective in achieving resistance in genetically susceptible mice. In contrast, injection of L1I9 plus C. parvum SC did not protect mice and, if anything, increased the severity of disease. Using the IP route, killed L119 plus C. parvum, and living L119 without adjuvant, provided some protection without being as effective as the living organism plus C. parvum. Further experiments have indicated that BALB/K mice are highly resistant when vaccinated with C. parvum and large numbers of either log phase or stationary
Mice Inbred female BALB/c.H-2^ (BALB/K), BALB/cH-2b (BALB/B) and C57BL/6 mice and male C3H/He were produced in a specific pathogen-free facility but mainiained conventionally, and were first used when at least 7-8 weeks old (11). After cutaneous injection of L. major promastigotes, C57BL/6 and C3H/He mice develop small lesions at the injection site that are usually resolved by two months; BALB/K mice develop larger lesions though they are not as susceptible as BALB/c or BALB/B and some resolution of lesions can occur after several months. Parasites All parasites used in this study originated at ihe WHO Reference Centre for Leishmaniases, Jerusalem, Israel. The virulent line VI2I (and the avirulent line AI2) was obtained by limit dilution cloning from the L. major isolate LRCLI37 (M HOM/IL/oo/Jericho 11) (12). The L. major LRC-LII9 was originally isolated in Kenya from a Tatera nigricaunda. thought to be the reservoir of L. major in that area. Although it does not cause lesions in mice, it was shown to be L. major by genomic characterization (5) and by isoenzyme analysis at 37 loci (Andrews, personal communication). The LRC-L52 strain of Leishmania donovani was isolated in India. Both LRC-LU9 and LRCL52 have been maintained exclusively in vitro in Schneider's Drosophila medium with 10; 10' living 1.119 promastigotes on days -21, -14 and - 7 . As in succeeding Figures, ihe number of mice in each group with lesions is indicated at one time point during the experiment together with ihe standard error of the arithmetic mean lesion score.
phase living L1I9 promastigotes. The data indicate that the LI 19 organism, that is apparently very deficient in GL (5), is nevertheless capable of inducing a substantial degree of protection against cutaneous leishmaniasis. Vaccinafion of mice against L. major using GL from L. donovani (LdGL) Groups of female C57BL/6 mice were injected IP with low doses of LmGL affinity purified from L major promastigotes using monoclonal WIC-79.3 or LdGL affinity purified from L. donovani promastigotes using monoclonal WIC-108.3. C parvum was used as adjuvant and injected simultaneously with 3-4 /ig antigen. Mice were boosted 3 weeks later with antigen only. They were challenged twice 14 days apart with V121, this being a particularly stringent test for vaccination efficacy. Data in Fig. 3 indicate that both LdGL and LmGL were partially
protective in C57BL/6 mice. A repeat experiment in male C3H/He mice that are also genetically resistant again provided evidence of heterologous protection against L. major using LdGL (Fig. 3). Experiments to be reported in a separate publication on antigen delivery and dosages indicate that the efficacy of antigens such as LmGL can be increased above that demonstrated in this series by incorporation into liposomes and/or increasing the dose of antigen. Despite relatively low levels of protection, the data pre.sented in Fig. 3 clearly establish that heterologous immunization (with LdGL) is as good as homologous immunization (with LmGL). DISCUSSION Cross-protection amongst Leishmania, often unidirectional and the exception rather than the rule (17), has been described in
390
G. F. MITCHELL AND E. H A N D M A N
A
BALB/c.H-2'*
BALB/c.H-2b
B
i -
3
3 -
\v •
1 -
/
1
2
f
"
T
/ ^ — — — ^
0
1
0
20
1
1
40 60 BO 100 120 20 40 60 Time after first challenge with V121 promostigotes (days)
80
100
fig. 2. Lesion scores in female BALB/c.H-2'' (panel A) and BALB/c.H-2i' mice (panel B) after cutaneous challenge with promastigoies of L. major V12L In panel A. mice were injected IP on day -14 with 20x10'' living LII9 promastigotcs (A A), or with C. parvum alone ( • • ) or with C /^arvum plus living L119(C'^—-O) or killed LI 19 ( • • } . Injections were repeated without C. parvum adjuvant at day - 7 and mice challenged with I y 10*' V12I on days 0 and 46. In panel B, the more susceptible BALB/c.H-2ti mice were immunized similarly except that the group given killed LI19 plus C. parvum was replaced by injection of living L n 9 plus C. parvum subcutaneously (• • ) . Injections were given to mice on day -21 (20x10* parasites with (O O) or without (A A) C parvum), day -14 (4 K 10'' parasites) and day - 7 (40x10* parasites with or without C parvum) with V121 challenge on day 0. A feature of this particular experiment is that, the BALB/c.H-2'' mice showed evidence of lale healing, an unusual occurrence with this mouse strain (12).
experimental animals and man (18; 19) with perhaps the most slriking data coming recently from the Wellcome laboratory (1). These investigators showed that intravenouslyinjected irradiated L. donovani promastigotes were as effective as irradiated L. major promastigotes at inducing resistance in highly susceptible BALB/c mice against L. major. The studies reported here indicate that the glycolipid antigens detected by monoclonal
antibody WIC-108.3 may contribute to protection by L. donovani against infection with L. major in mice. Definitive evidence that it is LdGL in WIC-108.3-affinity purified material that is responsible for protection against L. major must await the development of methods to purify this antigen other than through monoclonal antibody affinity purification or, better still, production by chemical synthesis.
VACCINATION AGAINST LEISHMANIA
10
20
391
30 Time after first challenge with V121 promastigotes (days)
Fig. 3. Lesion scores in female C57BL/6 (left hand panel) injected IP with 4 iig glycolipid from L. major Vt21 ( • •) or/,, donovani L52 {O O) together with C parvum or C purvum alone ([D H). on day 29 or not injected (A A), and challenged cutaneously on days 0 and 13 with I x 10^ /.. major VI21 promastigotes. Recipients of glycolipid received an additional IP injection of 3 ^g of either LmGL or ixlGL without adjuvant at day - 8. In a repeat experiment (right-hand panel) male C3H/He mice were injected IP with 10 ng of LmGL or LdGL together with C. parvum on day - 33 and again on day • 12 at a dose of 5 ng prior to 3 cutaneous challenges with 1 K 10^ V121 promastigotes on days 0, 21 and 35. ln a further experiment in repeatedly challenged CBH/HeJ female mice, recipients of LmGL or LxlGL plus C. parvum were clearly more resistant than recipients of C. parvum a\ot\c although a proportion of mice given LdGL (but not LmGL) showed a delayed appearance of lesions.
The L. ma/o/-glycoiipid (LmGL) recognized by WIC-79.3 is believed to be involved in macrophage recognition (6) and is a candidate vaccine molecule in cutaneous leishmaniasis provided that deleterious immune responses induced by the carbohydrate derivative can be averted (4; 8). LmGL is the only purified antigen so far reported to have a hostprotective effect against cutaneous leishmaniasis in genetically susceptible mice. It is not known whether bumans are bigb responders to the 'suppressive' carbohydrate antigens as is proposed to be the case for genetically susceptible BALB/c mice (4; 8). On the basis of the clinical course of disease in Old World cutaneous leishmaniasis, disease in the vast majority of individuals infected with L. major in an endemic area wili resemble that in genetically resistant C57BL/6 mice and C3H/He rather than genetically susceptible BALB/c mice and its H-2 congenic relatives. Analysis of the cellular immunology of cutaneous leishmaniasis in genetically resistant mice is hopefully more relevant to the human
situation than the events in BALB/c mice which, by contrast, may be more a model of human visceralizing disease caused by L. donovani. Whilst the carbohydrate suppressive antigens will exacerbate disease in C57BL/6 mice, such mice are presumably low responders because, in the few studies conducted, a convincing demonstration of disease exacerbation required multiple injections of high antigen doses and Freund's complete adjuvant (4; 8). Whether tbe carbohydrate derivative of LdGL is diseasepromoting in L. major infection bas yet to be resolved satisfactorily; one experiment in C3H/He, but not another in C57BL/6, has demonstrated that lesions can persist in recipients of L. donovani carbohydrate in FCA. Results of the other line of experimentation reported in this paper, namely, tbe vaccination efficacy of living L119, suggest strongly that other antigens besides LmGL are protective. However, this possibility relies on negative data obtained in the search for GL on L119
392
G. F. MITCHELL AND E . H A N D M A N
promastigotes using immunochemical approaches. Certainly, LI 19 promastigotes, either in log phase or stationary phase cultures, are negative for WIC-79.3-reactive material as determined by immunofluorescence, immunoprecipitation and sensitive radioimmunoassays. In vitro survival of LI 19 in macrophages can be promoted by preincubation of LI19 with LmGL and it is proposed that the avirulence of LI 19 in vivo is related to failure to survive in macrophages (5). The increased vaccinating potency of living as distinct from dead L119 is somewhat
puzzling if survival in macrophages is extremely short lived. Detailed studies on antigens of L119 should be rewarding in terms of the identification of vaccinating molecules, including other glycolipids. Acknowledgements: This work was supported by Ihe Nalional Healih and Medical Research Council of Australia, the Rockefeller FoundaEJon Great Neglec:[ed Diseases Network, ihe Leishmaniasis Componenl of the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases and Ihe National Institutes of Health Gram No. AII9347. We ihank Susan Wood, Nicki Cotierill. Maria Karvelas. Joan Curtis and Mandy Ludford for technical assistance.
REFERENCES 1. Howard, J. G. 1986. lnimunological regulation and control of experimental leishmaniasis. Ini. Rev. Exp. Paihoi 28:79-115. 2. Handman, E., and G, F. Mitchell. 1985, Immunization with Leishmania receptor for macrophages protects mice against cutaneous leishmaniasis. Proc. Nal. Acad. Sci. USA 82:5910-5914. 3. Mitchell. G. F., E. Handman and T. W. Spilhill, 1985. Examination of variables in ihe vaccination of mice againsl cutaneous leishmaniasis using living avirulcnt cloned lines and killed promastigoies of Leishmania major. Int. J. Paratilol. 15:677-684. 4. Handman, E., and G, F Mitchell. 1987. Lcishmaniamacrophage interaction: role of parasite molecules in infection and host protection. In Molecular Strategies of Parasite Invasion. (N. Agabian, H. Goodman, and N. Nogueira (eds.) Alan R. Liss, New York pp. 493-500, 5. Handman. E., L, Schnur, T. W. Spithill and G, F. Mitchell. 1986. Passive transfer of Leishmania lipopolysaccharide confers parasite survival in macrophages. J. Immunol. 137:3608-3614. 6. Handman, E., and J. W, Coding. 1985, The Leishmania receptor for macrophages is a lipidcontaining glycoconjugate. EMBO J. 3:2301-2306. 7. Mitchell. G, ¥.. and E. Handman. 1985. T lymphocytes recognize Leishmania giycoconjugaies. Parasitology Today 1:61-63, 8. Milchell, G. F., and E. Handman. 1986. The glycoconjugate derived from a Leishmania major receptor for macrophages is a suppressogenic, di.sease-promoting antigen in murine cutaneous leishmaniasis. Parasite Immunol. 8:255-263, 9. Liew, F. Y. 1986. Cell-mediated immunity in experimental cutaneous leishmaniasis. Parasitology Today 2:264-270. 10. Titus, R., G. Lima, H. Engers and J. Louis. 1984. Exacerbation of murine cutaneous leishmaniasis by adoptive transfer of parasite-specific helper T cell populations capable of mediating Leishmania major-
11.
12.
13.
14. 15.
16.
17.
18.
19.
specific delayed-type hypersensitivity. J. Immunol. 133:1594-1600.Mitchell, G. F. 1983. Murine cutaneous leishmaniasis: resistance in reconstituted nude mice and several Fl hybrids infected with Leishmania tropica major. J. Immunogenetics 10:395-412. Handman, E., R, E. Hocking, G. F. Mitchell and T. W. Spithill. 1983. Isolation and characlerizaiion of infective and non infective clones of Leishmania tropica. Moi. Biochem. Parasitol. 7:111-126. Delbarra, A., J. G, Howard and D. Snary. 1982. Monoclonal antibodies to Leishmania tropica major: Specificities and antigen location. Parasitology 85:523-531. Greenblali, C. L., G. M. Slutzky, A. A. Delbarra and D. Snary. 1983. Monoclonal antibodies for scrotyping Leishmania strains, J. Clin. Microbioi 18:191-193. Handman. E.. C. L, Greenblatt and J. W, Goding, 1984. An amphipathic sulphated glycoconjugate of Leishmania: Characterization with monoclonal antibodies. EMBO J. 3:2301-2306. Dubois, M., K. A. Gillies, J. K. Hamilton, P. A. Rebers and F. Smith. 1956. Colorimetric method for determination of sugars and related substances. Anal. C/iem. 28:3 50-356. Mauel, J. 1982. Cross-protection by non-pathogenic live and dead Leishmania. In Working Paper No. 5 lo the Scientific Working Group on the Imnuinology and Biochemistry of Leishmaniasis. World Health Organization, Geneva. Adier, S., and A. E, Gundcrs. 1964. Immunity to Leishmania mexicana following spontaneous recovery from Oriental sore. Trans. R. Soc. Trop. Med. Hvg. 58:274-277, deRosseil, R. A.. R. S. Bray and J. Alexander. 1987. The correlation between delayed hypersensitivity, lymphocyte activation and protective immunity in experimental murine leishmaniasis. Parasite Immunol. 9:105-115.
