Leishmania major and Leishmania donovani: Effect of LPG-containing and LPG-deficient strains on monocyte chemotaxis and chemiluminescence

EXPERIMENTALPARASITOLOGY 75,4424%g

(1992)

Leishmania major and Leishmania donovani: Effect of LPG-Containing and LPG-Deficient Strains on Monocyte Chemotaxis and Chemiluminescence SHOSHANA FRANKENBURG, Department

of Dermatology,

Hadassah

ANAT GROSS, AND VERA LEIBOVICI University

Hospital,

POB 12 000, Jerusalem,

Israel

FRANKENBURG, S., GROSS, A., AND LEIBOVICI, V. 1992. Leishmania mojor and Leishmania donovanir Effect of LPG-containing and LPG-deficient strains on monocyte chemotaxis and chemiluminescence. Experimental Parasitology 75, 442-448. Lipophosphoglycan (LPG) is a major glycolipid present on the membrane of Leishmania promastigotes and

amastigotes. We have previously shown that preincubation of peripheral blood monocytes with purified LPG inhibits IL-l production, chemotactic locomotion, and luminol-dependent chemiluminescence (LDCL). In the present study we tested the effect of LPG present on live parasites on monocyte activity. For this purpose, we used two mutant strains deficient in LPG and two LPG-containing strains. One pair was Leishmania major and the other Leishmania donovani. Monocytes in suspension were infected with the different parasite strains and tested for chemotactic locomotion and LDCL at different times between 1 and 72 hr after infection. In parallel, the percentage of infected monocytes was measured in stained cytospin preparations. The results obtained showed that at 1 hr of incubation only the LPG-containing strains inhibited chemotaxis, while the mutant strains showed a normal response. From 4 hr of incubation onwards, the mutant strains also inhibited monocyte chemotactic locomotion. LDCL was only slightly inhibited by the LPGcontaining strains after 1 hr, because of a high level of spontaneous stimulation, probably due to phagocytosis. At 24 and 72 hr all strains inhibited LDCL. These results suggest that LPG is responsible for early inhibition of macrophage activity, but that other factors are responsible for inhibition at later stages of in vitro infection. The model described here might represent a useful tool to further analyze the mechanisms involved in immune evasion of Leishmania paraSiteS.

D 1992 Academic

Press, Inc.

INDEX DESCRIPTORS AND ABBREVIATIONS: Leishmania

major: Leishmania

donovani;

Lipophosphoglycan; Monocyte inhibition; Chemotaxis; Chemiluminescence; Lipophosphoglycan (LPG); Luminol-dependent chemiluminescence (LDCL). INTRODUCTION

Lipophosphoglycan (LPG) is a major glycolipid present on the membrane of Leishmania parasites. LPG is present both in the promastigote and in the amastigote form of Leishmania strains, including Leishmania major and Leishmania donovani, the causative agents of Old World cutaneous and visceral leishmaniasis, respectively. LPG is a parasite ligand for macrophages (Handman and Goding 1985); experimental evidence suggests that LPG may play an important role in the intracellular survival of the parasite. We have recently shown that when purified LPG is added to peripheral

human monocytes, it inhibits monocyte IL-1 production, chemotactic locomotion, and luminol-dependent chemiluminescence (LDCL) (Frankenburg et al. 1990). Purified LPG has also been shown to inhibit oxidative burst (McNeely and Turco 1990), possibly through an inhibitory effect on protein kinase C activity (McNeely and Turco 1987). In vitro, LPG scavenges superoxide anions and free radicals (Chan et al. 1989). In addition, crude LPG (excreted factor) has been shown to inhibit lysosomal p-galactosidase in mouse macrophages (El-On et al. 1980). All the studies mentioned above have evaluated the effect of exogenously added LPG extracted from pro-

442 0014-4894192$5.00 Copyright All rights

0 1992 by Academic Press, Inc. of reproduction in any form reserved.

LPG-DEFICIENT

LEISHMANIA

AND HUMAN

mastigotes; recently it has been shown that LPG from L. major promastigotes differs antigenically from LPG from amastigotes of the same strain (Glaser et al. 1991). The purpose of the present work was to evaluate the effect of LPG, produced and presented by live promastigotes and amastigotes, on monocyte activity. Human monocytes were infected with promastigotes of L. major or L. donovani (LPG-containing), and with two LPG-deficient strains, and monocyte chemotactic locomotion and chemiluminescence were measured at different times after infection. MATERIALS

AND METHODS

Parasites. L. major strains used were LRC-L137 (137) and LRC-L119 (119). Strain 137 is an LPGcontaining strain isolated from the Jordan Valley. Strain 119 was shown to be a L. major by enzyme profile and does not produce LPG by standard procedures (Schnur 1982; Handman et al. 1986). L. donovani strains used were a parental strain of L. donovani and an LPG-deficient clone obtained from it, R2D2 (King and Turco 1988), kindly provided by S. Turco. The parasites were grown in Schneider’s medium (Hink 1979) containing 10% FCS (Biological Industries, Beith Haemek, Israel). ELBA for LPG measurement of L. major strains. One million promastigotes (137 or 119) per well were attached by centrifugation to ELISA plates (BectonDickinson, CA) and fixed by immersion in 0.25% glutaraldehyde. The wells were incubated with the monoclonal antibody WIC 79.3, specific for L. major LPG or with an unrelated control antibody. The ELISA was performed using an extravidin-biotin staining kit (Biomakor, Rehovot, Israel) according to the producer’s directions. Cell preparation and infections. Venous blood from normal human donors was separated on a FicollHypaque (Pharmacia, Uppsala, Sweden) gradient, as previously described (Boyum 1968). The mononuclear cells (MNC) were washed three times with RPM1 1640 (Ysum, Jerusalem, Israel); this white cell preparation contained 25-40% monocytes, as established by morphological examination and by nonspecific esterase staining. Two million MNC were exposed to 2 x lo6 stationary phase promastigotes in polypropylene tubes (1.5ml microtubes, Labotal, Jerusalem, Israel), in a final volume of 450 t.~lof RPM1 containing 10% FCS. The cells and promastigotes were incubated at 37°C in 5% CO, for l-24 hr and then washed three times by centrifugation at 800 rpm for 7 min to remove most extracellular parasites. The cells were further incu-

MONOCYTE ACTIVITY

443

bated for different time periods in the presence of 10% FCS. Only an occasional promastigote could be seen after 48 hr of incubation. For some experiments, LPGdeficient promastigotes were preincubated for 60 min with 100 l& LPG (purified LPG, devoid of the LPGassociated protein, kindly provided by S. Turco, Lexington, KY) and then added to the monocytes. Determination of infectivity. The proportion of monocytes containing intracellular parasites was determined by examination of May-Gruenwald and Giemsa-stained smears prepared by cytocentrifugation (Cytocentrifuge, Shandon Southern Products, England). Assay for chemotactic activity. Chemotactic activity was measured by the leading front technique, as previously described (Matzner et a/. 1985). Mononuclear cells, after incubation in the presence or absence of parasites for different times, were brought to a concentration of 3 x 106/0.2ml and placed in the upper compartment of a Boyden chamber. The chemoattractant, 2% zymosan-activated normal human serum, was placed in the lower compartment (0.2 ml); the compartments were separated by 5-urn pore size membranes (Sartorius, Gottingen, Germany). Incubation was carried out at 37°C and 5% COZ for 90 min. Assay for LDCL. MNC, after incubation in the presence or absence of parasites for different times, were brought to a concentration of 4 x IO6 cells/ml in Hanks’ balanced salt solution, pH 7.3, transferred to polyethylene tubes and 5 p,l luminol (5-amino-2,3dehydro-1,4-phtalo-2-indione and 2 rig/ml DMSO) were added. The monocytes were activated either with 20 ul streptococci (OD of 2 at 550 nm) and 10 ul histone (Sigma, 10 mg/ml) or with PMA (phorbol 12myristate 13-acetate, Sigma, 1 l&ml unless otherwise stated). LDCL was measured with a luminator (Lumak Bio-Counter 2010); light emission was recorded every 30 set for 5 min and maximal emission expressed as counts per minute (cpm). Statistical analysis. Statistical analysis was performed using Student’s t test. P < 0.05 was considered statistically significant. RESULTS

Two LPG-deficient Leishmaniu strains were used in this study. The L. donovuni mutant, R2D2, was shown to be deficient of LPG (King and Turco 1988). Since the L. major LPG-deficient strain, 119, has been shown to contain small amounts of LPG (D. Sacks, personal communication), we performed an ELISA test, in which a monoclonal antibody specific for L. major LPG was incubated with 119 or 137 promastigotes. The LPG-containing 137 strain gave

444

FRANKENBURG,

GROSS, AND LEIBOVICI

a positive reaction with dilutions of the antibody of l:lOO,OOO, while the 119 strain was negative at a 150 dilution (results not shown). The percentage monocyte infectivity was similar for L. major and L. donovani: At 1 hr the percentage infectivity of all the strains ranged from 3&100% of the infectivity at 24 hr. At 24 hr the LPG-deficient strains had the same levels of parasite infection as the LPG-containing strains, but then the level of infectivity of the LPGdeficient strains gradually decreased, and after 4 days of incubation the percentage infection of monocytes with the LPGdeficient strains decreased to lO-30%, whereas the infectivity of LPG-containing strains remained unchanged (Fig. 1). Effect of infection on chemotaxis (Fig. 2). Monocytes that were infected with L. major 119 (LPG-deficient) promastigotes for 1 hr, washed, and then tested for chemotactic locomotion showed activity that was not significantly (P < 0.06) lower than that of uninfected controls. In contrast, monocytes infected for 1 hr with L. major 137 (LPG-containing) parasites showed inhibition of chemotaxis and even inhibition of random migration (i.e., below the level of spontaneous migration, without the ad-

dition of the chemoattractant). Similar results were obtained at 2 hr of incubation; but at 4 hr the chemotactic locomotion of the LPG-deficient strain was also markedly inhibited. At 24 hr the 119-infected monocytes also showed significant inhibition of chemotaxis (P < 0.004), but at 72 hr the inhibition was not significant (P < 0.2) compared to the normal control of the same day. The 137-infected monocytes showed inhibition of both chemotaxis and random migration at 24 and 72 hr. Chemotactic locomotion of uninfected cells did not change during the 72-hr incubation period (P = 0.14). Monocytes infected with L. donovani R2D2 (LPG-deficient) promastigotes for 1 hr showed normal chemotactic locomotion, while the wild type (LPG-containing) promastigotes markedly reduced chemotactic locomotion. As for the L. major strains, reduction in chemotactic response of the LPG-deficient strain started between 2-4 hr after infection. Preincubation of R2D2 promastigotes with purified LPG before their addition to the monocytes induced marked decrease of chemotaxis (Fig. 2), confirming that the inhibition observed was caused by LPG. Effect of infection on LDCL. The effect

100

1

I

0’

1

24 hours

48 after infection

I

I

72

96

FIG. 1. Infection curves of human monocytes with L. major (full lines) and L. donovani (dotted lines). Full markers, LPG-containing strains; open markers, LPG-deficient strains. Values plotted are the means f SEM of 3-13 experiments.

LPG-DEFICIENT

LEISHMANIA

AND

HUMAN

MONOCYTE

ACTIVITY

445

60

a

T

60 60

0

lh

2h 4h monooyte-parasite

L. major

137

L. major

116

non-infected

24h 72h incubation

60

60 Wild

-20

type

m

R2D2

a

non-infected

-

R2D2+LPG

I

’ lh

2h 4h monocyte-parasite

24

h 72 incubation

h

FIG. 2. Chemotactic locomotion of monocytes infected with (a) L. major and (b) L. donovani. Each column represents the mean 2 SEM of two to five independent experiments.

of infection on chemiluminescence was similar for both strains of L. major and L. donovani. After 1 hr of infection the monocytes were activated even without the addition of a stimulant. Cytospin preparations of these MNC, after three washes, demonstrated the presence of many extracellular promastigotes, either attached to monocytes, or in clumps (and therefore were not discarded by the washings). Addition of PMA induced some enhancement of the response, which was slightly (but not significantly) higher in the presence of the LPGdeficient strains. Since the cells were already stimulated, we also measured LDCL after the addition of suboptimal concentra-

tions of PMA (l&50 rig/ml), but no improvement in the response was observed (results not shown). However, after 24 and 72 hr of incubation both the LPG-deficient and the LPGcontaining strains showed a significant inhibition of LDCL (Table I). The measurements at 24 and 72 hr were done after stimulation with either PMA or streptococci and histone, with similar results. DISCUSSION

The studies described here were performed to evaluate the role that LPG, the major glycolipid of leishmania parasites,

446

FRANKENBURG,

GROSS, AND LEIBOVICI

TABLE I Luminol-Dependent Chemiluminescence of Monocytes Infected with (a) L. major or (b) L. donovani a. L. major Incubation time (hr) 1 24 72

Unstimulated Stimulated Unstimulated Stimulated Unstimulated Stimulated

L. major 137 cpm k se 2901 2 3721 ? 91 f 245 2 68 2 127 *

407 487 (1.3)” 4 22 2 40

L. major 119 cpm + se 2856 k 5704 k 129 k 298 k 88 * 252 2

1075 936 (2.0) 13 41 19 90

Noninfected cpm f se 170 f 42 24,536 Ifr 12,673 (144) 158 + 36 1199 * 191 113 f 32 796 k 230

b. L. donovani Incubation time (hr) 1 24 72

Unstimulated Stimulated Unstimulated Stimulated Unstimulated Stimulated

Wild type cpm k se 4496 k 7458 k 204 k 495 * 66k 126 *

785 691 (1.6) 61 377 11 66

W, cpm k se 3991 2 9502 k 162 f 442 f 62 2 186 ”

1550 908 (2.4) 62 362 14 134

Noninfected cpm 2 se 181 + 30,136 + 300 f 8359 f 58 f 735 f

42 9666 (166) 199 2027 16 312

a Stimulation index: cpm stimulatedkpm unstimulated cells.

plays in the activation of infected human tion is inhibited by both strains, with and monocytes. For this purpose we used two without LPG. From these results it appears LPG-containing strains and two LPG- that LPG is mainly responsible for the inideficient strains. We first tested the ability tial inhibition of monocyte chemotaxis. of these strains to enter and maintain an This could relate to the fact that at l-2 hr infection in nonadherent human MNC. The promastigotes are present in the culture and results (Fig. 1) indicated that both LPG- are being actively phagocytized; on the deficient variants easily entered the phago- other hand, at 4 hr, almost all of the paracytic cells, but were not able to survive and sites have transformed into amastigotes. were mostly destroyed 3-4 days postinfec- McConville and Blackwell (1991) have been unable to detect LPG in L. donovani amastion, in contrast to the LPG-containing strains. Similar observations were made by tigotes; in addition, Glaser et al. (1991) Handman et al. (1986) for L. major and by have shown, using monoclonal antibodies, McNeely and Turco (1990) for L. donovani that the LPG from amastigotes of L. major strains. The growth curves of the parasites is antigenically different from the LPG of suggest that the LPG-deficient parasites, promastigotes. Therefore, it is possible that after successful entry into the monocytes, LPG from promastigotes and amastigotes became more vulnerable to killing, either has different functions, and that only LPG oxidative or nonoxidative . from promastigotes inhibits macrophage The results obtained in the chemotaxis activation. We are currently investigating experiments strongly suggest that LPG is this possibility. It should be noted that all responsible for the inhibition of chemotac- the studies of LPG function performed with tic activity of monocytes early in the course LPG extracts, whether purified or not, of infection. On the other hand, from 4 hr of were done with LPG from promastigotes. The reason for inhibition of chemotaxis incubation onwards, it seems that other factors affect chemotaxis, since the locomo- from 4 hr of infection onwards, even in

LPG-DEFICIENT

LEISHMANIA

AND

HUMAN

MONOCYTE

ACTIVITY

447

LPG-deficient strains, is not known. The Ginzburg for the use of the Luminator and S. Turco for the supply of LPG. membrane-bound acid phosphatases, which are present in both promastigotes REFERENCES and amastigotes (Glew et al. 1988), have been shown to reduce oxidative burst of BOYUM, A. 1968. Separation of leukocytes from blood and bone marrow. Scandinavian Journal of Clinical neutrophils (Remaley et al. 1985) and could and Laboratory Investigations 2 (Supp. 97), 77-79. also be a potential candidate for inhibition CHAN, .I., FUJIWARA,T., BRENNAN, P., MCNEIL, M., TURCO,S. J., SIBILLE, J., SNAPPER,M., AISEN, P., of monocyte activation. Other factors AND BLOOM, B. R. 1989.Microbial glycolipids: Poswhich affect leishmania virulence, such as sible virulence factors that scavenge oxygen radiposttranslational N-glycosilation of procals. Proceedings of the National Academy of Sciteins (Chang and Chaudhuri 1990) could ences USA 86, 2453-2457. also affect monocyte activity. These possi- CHANG, K., AND CHAUDHURI, G. 1990.Molecular determinants of Leishmania virulence. Annual Review bilities and others await further investigaof Microbiology 44, 4%529. tion. EL-ON, J., BRADLEY, D. J., AND FREEMAN, J. C. From the chemiluminescence experi1980. Leishmania donovani: Action of excreted facments it appears that monocytes, during the tor on hydrolytic enzyme activity of macrophages phagocytic stage (l-2 hr of infection), have from mice with genetically different resistance to infection. Experimental Parasitology 49, 167-174. an elevated oxidative burst even when inFRANKENBURG, S., LEIBOVICI, V., MANSBACH, N., fected with the LPG-containing strains. TURCO, S. J., AND ROSEN, G. 1990. Effect of glyThese results are in contrast with previous colipids of Leishmania parasites on human monofindings that LPG inhibits oxidative burst cyte activity: Inhibition by lipophosphoglycan. of monocytes (Frankenburg et al. 1990; Journal of Immunology 145, 4284289. McNeely and Turco 1990). This could be GALVAO-QUINTAO, L., ALFIERI, S. C., RYTER, A., AND RABINOVITCH, M. 1990. Intracellular differendue to the fact that most promastigotes are tiation of Leishmania amazonensis promastigotes to probably nonmetacyclic and therefore amastigotes: Presence of megasomes, cysteine proreadily ingested and killed. It is possible teinase activity and susceptibility to leucine-methyl that oxidative burst due to parasite phagoester. Parasitology 101, 7-13. cytosis is not inhibited by the LPG as it is GLASER,T. A., MOODY, S. F., HANDMAN, E., BACIC, A., AND SPITHILL, T. W. 1991.An antigenically dispresented by the live promastigotes. In adtinct lipophosphoglycan on amastigotes of Leishmadition, the fact that promastigotes are more nia major. Molecular and Biochemical Parasitology sensitive to oxidative metabolites than 45, 337-344. amastigotes has been well documented GLEW, R. H., SAHA, A. K., DAS, S., AND REMALEY, A. T. 1988. Biochemistry of the Leishmania spe(Murray 1981; Pearson et al. 1983). It reReviews 52, 412432. cies. Microbiological mains unclear why the oxidative burst, afHANDMAN, E., AND GODING, J. W. 1985.Leishmania ter stimulation with PMA, could be enreceptor for macrophages is a lipid-containing glyhanced only minimally, even with subopticoconjugate. EMBO Journal 4, 329-336. ma1 doses of PMA. One possible HANDMAN, E., SCHNUR, L. F., SPITHILL, T. W., AND MITCHELL, G. F. 1986. Passive transfer of explanation is that the monocytes are “exLeishmania lipopolysaccharide confers parasite surhausted,” and thus cannot be further stimvival in macrophages. Journal of Immunology 137, ulated for an oxidative burst response. It is 3608-3613. also possible that the oxidative burst is in- HINK, W. F. 1979. Cell lines from invertebrates. hibited by the same LPG-independent Methods in Enzymology 58, 45W66. mechanism that inhibits chemotaxis. KING, D. L., AND TURCO, S. J. 1988. A ricin agglutiACKNOWLEDGMENTS

Supported in part by Research Contract NOl-22668NIH-NIAID Epidemiology and Control of Arthropod Borne Diseases and the Nicholas Fund. We thank I.

nin-resistant clone of Leishmania donovani deticient in lipophosphoglycan. Molecular and Biochemical Parasitology

28, 285-294.

MATZNER, Y., PARTRIDGE,R. E. H., AND BABIOR, M. 1985.Chemotactic inhibitor in synovial fluid. Immunology 49, 131-138.

FRANKENBURG,

448

GROSS, AND

MCCONVILLE, M. J., AND BLACKWELL, J. M. 1991. Developmental changes in the glycosilated phosphatidylinositols of Leishmania donovani: Characterization of the promastigotes and amastigote glycolipids. The Journal of Biological Chemistry 266, 15170-15179. MCNEELY, T. B., AND TURCO, S. J. 1987. Inhibition of protein kinase C by the Leishmania donovani lipophosphoglycan. Biochemical and Biophysical Research Communication

148, 653-657.

MCNEELY, T. B., AND TURCO, S. J. 1990. Requirement of lipophosphoglycan for intracellular survival of Leishmania donovani within human monocytes. Journal

of Immunology

144, 2745-2748.

MURRAY, H. W. 1981. Susceptibility of Leishmania to oxygen intermediates and killing by normal macrophages. Journal of Experimental Medicine 153, 1302-1315.

LEIBOVICI

PEARSON,D. R., HARCUS, J. L., ROBERTS,D., AND DONOWITZ, G. R. 1983. Differential survival of Leishmania donovani amastigotes in human monocytes. Journal of Immunology 131, 19944999. REMALEY, A. T., GLEW, R. H., KUHNS, D. B., BASFORD,R. E., WAGGONER,A. S., et al. 1985. Leishmania donovani: Surface membrane acid phosphatase blocks neutrophil oxidative metabolite productions. Experimental Parasitology 60, 331-341. SCHNUR,L. F. 1982.The immunological identification and characterization of leishmanial stocks and strains, with special reference to excreted factor serotyping. In “Biochemical Characterization of Leishmania” (M. L. Chance and B. C. Walton, Eds.). UNDP/World Bank/WHO, Geneva. Received 9 April 1992; accepted 10 September 1992