0022-1767/87/1386-1926$02.00/0 THEJOURNAL OF IMMUNOLOGY Copyright 0 1987 by The American Associationof Immunologists

Vol. 138. 1926-1932. No. 6, March 15. 1987 Prfnted in U . S . A .

PARASITE-ACCESSORY CELLINTERACTIONS INMURINE

LEISHMANIASIS

11. Leishmania donouani Suppresses Macrophage Expression of Class I and Class I1 Major Histocompatibility ComplexGene Products'

In the present study, we examined the modulation of MHC class I1 and classI gene products on BALB/ c macrophages infected with the obligate intracellular protozoan Leishmania donouani. Our findings indicated that this organism suppressed macrophage expression of both classes of MHC antigens. These effects varied somewhat, depending on whether cells were in the basal state or were stimulated with interferon-y. Thus, class I1 density on interferon-y-treatedinfected macrophages was suppressed by as much as 90%,relative to lymphokinestimulated control cells. Induction ofH-2K and H2D by lymphokine treatment of infectedmacrophages was also markedly reduced. In the basal (non-lymphokine-treated)state, infected cells also showed reduced expression of H-2K and H-2D. but not I-A or I-E. The latter result was related to minimal levels of class I1 molecules on normal, in vitro cultured macrophages. Suppression of MHC gene products correlated with both the duration and intensity of leishmania infection and could not be overcomeby increasing doses of interferon-y. Culture of cells under conditions of cyclooxygenase inhibition completely abolished elevated synthesis of prostaglandin Ez by infected macrophages and augmented their responsiveness to lymphokine induction of class I1 antigens by 60 to 80%. These results indicate that L. donouani is capable of subverting a critical macrophage accessory function required for the induction of T lymphocyte immunity. This mechanism could account, at least in part, for defective parasite-specific cellmediated immunityseen during infections with this protozoan. Leishmania are obligate intracellular protozoa that residewithinmononuclearphagocytes.Parasite-specific cell-mediated immunity is believed to play a central role in both the murine and human immune responses to

these organisms. This concept is based on evidence indicating that theseT cell-dependent processes both contribute to the control of parasite proliferation within the host and also mediate acquired resistance to reinfection (1-4). Despite the importance of these responses, antileishmanial cell-mediated immunity may not appear for many months after infection, ormay fail to appear altogether (5, 6). This situation is exemplified best by persistentinfectionswith visceralizing strains of Leishmania such as Leishmania donovani and Leishmania chagasi. These syndromes have findings consistent with parasite-specificT cell nonresponsiveness,andthese persist unless curative chemotherapy is undertaken(710). To dissect the basic mechanisms responsible for these defective host responses, our studies with L. donouani have focused on infection-induced alterations in parasitized accessory cells. This approach has been based on two related considerations. First. accessory cells are of pivotal importance in the inductionof antigen-specific T lymphocyte activation. Thus, T cells recognize foreign antigens presented in the context of accessory cell major histocompatibility complex (MHC) class I1 gene products (11- 18). and when this occurs in the presence of interleukin 1 (IL 1; also a n accessory cell product),T cell activation may ensue (1 1,12, 19-21). Secondly, because Leishmania parasitize a major categoryof accessory cells (cells of the monocyte-macrophage series), they are strategically positioned tosubvert these and potentially other accessory cell functions.Sucheffects could clearly thwart the successful induction of anti-leishmanial T lymphocyte immunity. In the preceding companion paper (22). we reported findings which indicate that L. donovani both evades and suppresses the macrophage IL 1 response during the establishmentof intracellular parasitism. In the present investigation, we present evidence indicatingthatthisorganismalsosuppressesmacrophage expression of both class I1 and class I MHC gene products. Thus, L. donovani is capable of subverting discretemacrophageaccessoryfunctionsthatarerequired for the induction of T cell-dependent anti-parasite immunity.

Received for publication September8, 1986. Accepted for publication December 2. 1986. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. MATERIALS AND METHODS ' This work was supported by grants MA-6833 and MT-3399 from the Medical Research Council of Canada, and by grants 64(83-2) and 53(84Leishmania. Amastigotes of the Sudan strain 2s of L. donouani 2) from the BritishColumbia Health Care Research Foundation. Dr. Reiner is a Scholar of the BritishColumbia Health Care Research used in this study were isolated as described in the accompanying companion paper (22). Foundation. Please address requests for reprints to Dr. Reiner at the AnlmQls. Male Syrian hamsters (LaK:LVG) were purchased from Department of Medicine. University of British Columbia, 910 West 10th Charles River Lakeview, Newfield. N J . Female BALB/c mice, 4 to 8 Avenue, Vancouver, B.C. V5Z 1M9.

1926

1927

SUPPRESSION OF MACROPHAGE H-2 ANTIGENS BY LEISHMANIA wk old, were from Charles River Canada, St. Constant, Quebec, Canada. Chemicals and reagents. The preparation and sources of DulbecCO’S modified Eagle’s medium (DMEM:34500 mg/L glucose), divalent cation-free Hank’s balanced salt solution (HBSS),and fetal bovine serum (FBS)were as described (22). Recombinant murine interferon-? (IFN) was the generous gift of Dr. H. M. Shepard, Genentech,Inc., South San Francisco, CA. Purity was 99.96% and thespecific activity was 5.2 X lo7 U/mg (lot 320924). Working solutions were prepared inDMEM-S/lO%FBS. Sodium meclofenamate was a gift from Warner-Parke-Lambert (Indianapolis, IN). Dulbecco’s phosphate-buffered saline was from GIBCO (no. 310-4080: Burlington, Ontario, Canada) a t 1OX concentration and was diluted with sterile pyrogen-free water. Goat serum was also from GIBCO (no. 200-6210) and was heat-inactivated a t 56°C before use. l Z 5 I was purchasedfrom Amersham (IMS-30:Oakville, Ontario, Canada). Cell preparation and culture. Elicited peritoneal cells were obtained from mice that had been injected i.p. (3days previously] with 3 mlof sterile 2.4% thioglycollate (no. 0432-01: GIBCO). Peritoneal cavities were lavaged with 5 ml of DMEM-S with 10 U/ml heparin. The cells were washed once in medium, were resuspended. andwere adjusted to 4 x 1O6 cells/ml in DMEM-S/lO% FBS. Ten milliliters of the suspension were added to Extran 300 (B80002-82: BDH, Vancouver. BC)-washed glass petri dishes that had been previously heated overnight a t 185°C. The cells were incubated for 24 hr at 37°C in a 5% C0,/95% air humidified atmosphere and were then washed three times with warm (37°C) HBSS. After removal of nonadherent cells. cold HBSS with1 mMEDTA was added and the dishes were held on ice for 30 min. Adherent cells werethen removed by gentle scraping with a rubber policeman. These cells were 92% to 98% macrophages both by morphologic criteria in Wright-stained preparations and alsoby virtue of their phagocytic capability. After washing once in DMEM-S. the cells were resuspended a t 1 X 1 06/ml in sterile Teflon beakers (no. S70020: Scientific Specialties, Randallstown. MD). The cultureswere then either untreated orexposed to freshly isolated amastigotes ofL. donouani followed by gentle agitation for 30 min on a rotary shaker. After a n additional 4 hr incubation a t 37°C. the cultures were either untreated or were stimulated with IFN. Fresh IFN was added to the cultures daily: untreated cells received DMEM-S/lO%-FBS.Leishmania infection was quantitated by enumerating amastigotes in at least 200 cells per population under oil (10OOx)in Diff-Quik-stained cytocentrlfuge preparations. Binding assayfor cell surface antigens. Anti-I-Ad (MK-D6. no. 1360) waspurchased from Becton Dickinson (Sunnyvale, CA). The hybridoma 14-4-4s (23) secreting antibody specific forI-E of k,d,p,r haplotypes was obtained fromthe American Type Culture Collection, Rockville, MD, and ascites raised in BALB/c mice was purified by high-pressure liquid chromatography on a Protein Pak, DEAE 5 P W column (Waters Associates, Milford, MA). Lyophilized ascites from hybridomas 15-5-5s (anti-H-2Kd)(23) and 34-5-8s(anti-H-2Dd)(24) were generously provided by Dr. D. H. Sachs (National Cancer Institute, National Institutes of Health, Bethesda, MD). These were reconstituted inice-cold PBS. All mouse hybridoma antibodies were of the lgG2a subclass and were used at saturating concentrations.Before use, the monoclonals were screened for specificity against spleen cells from mice of correct and incorrect haplotypes. An irrelevant lgG2a mouse monoclonal, 6-90, with specificity for ferredoxin (25) was used at 25&/ml as a negative control in bindingassays. H-2K, H-2D. and class 11 antigens were quantitated in a n indirect lZ5Ibinding assay under saturating conditions. The method used was described previously (26) and was modified slightly for use with 96well, flexible, round-bottomed microtiter plates (no. 1-220-24, Dynatech. Fisher Scientific, Vancouver, BC).In brief, cells were recovered from Teflon beakers, werecentrifuged, and were resuspended in binding assay medium (medium) which consisted of PBS with 5 mM sodium azide and 1% heat-inactivated goat serum. After adjustment to the desired concentration, from 1 to 2.5 x lo5 cells (the number being held constant in each assay] were added per microtiter well in 0.2 ml. The plates were then centrifuged, the supernatants were aspirated, and the cells were resuspended in 0.05 ml of a saturating concentrationof monoclonal antibody in medium. After incubation for 1 hr at4°C. the cells werewashed four timesin medium and were resuspended in 0.05 ml of a saturating concentration (25 pg/ml) of ‘251-conjugatedanti-immunoglobulin (Ig).This latter consisted of ‘251-conjugatedF (ab’),fragments of goat anti-mouse 1g (261. After a 1-hr incubation with lz5I-anti-Ig, the wells were washed four timesas described above. The finalcell pellet was resuspended Abbreviations used in this paper: PGE2,prostaglandin Ea: DMEM-S. supplemented Dulbecco’s modified Eagle’s medium.

in 0.1 ml of medium and was transferred to 12 x 75-mm glass tubes for detection of bound radioactivity. Binding assay determinations were done in triplicate for each antibody with each cell population, and the data are expressed as the mean specific binding for each monoclonal antibody. These values were calculated by subtracting the meanbinding of the negative control antibody(6-90) to each cell population from each of the triplicate measurements of anti-H2 antibody binding to thecorresponding cell population. Cyclooxygenaseinhibition andprostaglandin E, (PCE,) radioimmunoassay. A fresh solution of sodium meclofenamate was prepared in DMEM-S at 200 p M . After sterile filtration (0.2 pm). aliquots of this working solution were added to cell cultures 1 hr before the addition of Leishmania. After 96 hr of culture with or without IFN, cells were recovered for measurement of I-A. and cellfree supernatantswere assayed for PGEz by using a rabbit antibody (Sigma Chemical Co., St. Louis. MO: P6765) a s described (27). Statistical analysis. Comparisons between experimental groups were performed by Student’s t-testfor independent means. RESULTS

Kinetics of induction of macrophage classII antigen expression. Stimulation of normal macrophages for 96 hr with 12.5U/ml of IFN was found to result in maximal induction of I-A and I-E. The results shown in Figure 1 represent a typical time course for class I1 induction. Control cells cultured with IFN showed a steady decline in I-E expression. Thus, by day 4, binding of anti-I-E to these cells wasessentially nil.Identicalresultswere found for I-A (data not shown). On the other hand, stimulation of cells with 12.5 U/ml of IFN for 96 h r resulted in a progressive increase in I-E expression. In contrast, leishmania-infected cells were essentially maximally induced at 24 hrof treatment with IFN. Relative to control cells, I-E induction on infected cells was suppressed by 67% (p < 0.001) and56% (p < 0.001) at 72 hr and 96 h r

T

1

0

24

72

40

Duration of culture

96

(h)

Figure I . Kinetics of class II antigen expression measured by binding of anti-I-Emonoclonal activity to normal (squares)or L. donouani-infected (circles)macrophagescultured in the absence (closed symbols) or presence (open symbols)of 12.5U/ml of IFN. Cells were exposed toLeishmania at a multiplicity of infection of 20: 1 andwere greater than 90% infected at all time points. The mean (kSD) numbers of amastigotes per cell were 7 f 0.8. 13.4 f 2.4. 19.5 k 1.9. and 26.9 f 1.7. respectively at 24 hr. 48 hr. 72 hr, and 96 hr. and did not differ between IFN- and non-IFN-treated groups. For anti-I-E binding, the data are the mean k S D of triplicate determinations at each point. Nonspecific binding of 6-90 to control and infected macrophages, respectively. ranged between3820 and 6458cpm and 3309 and 5138 cpm. In the IFN-stimulated state, the difference between infectedand control cells at 72 hr and 96 hr were bothsignificant at the 0.1% level.

1928

SUPPRESSION OF MACROPHAGE H-2ANTIGENS

of stimulation, respectively. When the kinetics of I-A induction on normal and infected cells were examined, results essentially parallel to those of I-E were found (data not shown). Of interest was the finding in some experiments that leishmania-infected, non-IFN-treated macrophages did not show a progressive decay in basal expressionof class I1 antigens over 96 hr of culture. Thus, the data shown in Figure 1 suggest the presence of a minimal stimulus acting to maintain basal levels of I-E expression by these cells. This stimulus was inconstant, however, because in the majority of experiments basal expression of class I1 antigens at 96 hr did not differ between normal and infected cells (see Figs. 2 and 4 and Table I). Dose response relationship f o r induction of macrophage class I1 antigen expression by IFN. To examine whether suppression of class I1 antigen expression by infected macrophages was related to a n alteration of the

-.-.5p0S0 '/ m

I

6.25 3.125

1

I

I

12.5

25

50

100

I

200

U/mL of gamma-interferon Figure 2. IFN dose-response relationship for induction of macrophage class 11 antigen expression measured by binding of anti-I-A monoclonal antibody to normal (squares) or L. donouani-infected [circles) macrophages after 96 hr of lymphokine stimulation. Cells were infected with Leishmania at a ratio of 20:l and were greater than 90%infected. The data are the mean f S D of triplicate measurements at each dose of IFN. Nonspecific binding of 6-90 to controland infected macrophages. respectively, ranged between 2657 and 3702 cpm and 3400 and 3891 cpm. The difference between normal and infected cells cultured with200 U/ml of IFN was significant a t the 0.01%level.

BY LEISHMANIA

IFN dose-response relationship,both normal and infected macrophages were cultured for 96 hr with IFN in doses ranging from 3 to 200 U/ml. The results of such a titration experiment, shown in Figure 2, demonstrated that increasing the optimal IFN concentration for normal cells by a s much a s eightfold failed to restore I-A expression by infected cells (IFN = 200 U/ml, I-A = 60% of control, p < 0.0001) to normal. Modulation of macrophage class4 and class-11 expression during infection with L. donovani. Parasiteinduced suppression of MHC antigen expression was not selective for class-I1 gene products. This was determined by culturing both normal and infected cells for 96 hr with or without IFN (12.5 U/ml) and assaying the binding of anti-class-I and anti-class-I1 monoclonal antibodies. A s shown in Figure 3, relative to normal cells, the binding of anti-I-A and anti-I-E by IFN-treated infected cells were < 0.01). suppressed by 70% (p < 0.001)and47%(p respectively. Similarly. binding of anti-H-2K was suppressed by approximately 50% in both the basal (p < 0.01) and IFN (p < 0.001)-treatedstates. In contrast, suppression of H-2D induction on IFN-treated infected cells was lesspronounced (15%< control, p< 0.005). Parasite dose-response relationship for suppression of macrophage classI and classI1 antigen expression. Examination of the relationship between intensity of infection and suppression of MHC antigen expression revealed that themacrophage parasite burdencorrelated with the degree of suppression of MHC antigens. A s shown in Figure 4,macrophages infected at aparasite:cell ratio of 2.5:l (38.2% of cells infected, 3.9 parasites/cell) and stimulated with 12.5 U/ml of IFN for 96 hr had a modest increase in anti-I-A binding over that of control cells. This result wasquickly reversed, however, as theintensity of infection was increased to5:l (55.2% infection, 4.5 parasites/cell).Binding of anti-I-A by these cells wassuppressed by 34% (p < 0.005). relative to control cells, and this suppressionincreased progressively with the intensity of infection. Thus, at 40 parasites:cell (85.5% infection, 22.3 parasites/cell), I-A density was decreased by 76% (p< 0.001). In contrast to the results of IFN stimulation, under basal conditions densities of I-A on normal and infected cells were low and did Binding of '251-goat anti-mouse Ig (CPM

Cell treatment Determinant

Ftgure 3. Simultaneous modulation of class I and class I1 MHC gene products on macrophages infected with L. donouanl (20:1) measured by binding of monoclonal antibodies directedat theindicated determinant. IFN was used a t 12.5 U/ml and binding was assayed at 96 hr. The data are the mean2 SD of triplicate measurements for each cell population. The arrows and numbers appearing in bars indicate the percent change in H-2 or Ia antigen expression by infected cells relative to the appropriate control. Nonspecific binding of 6-90 to control and infected macrophages. respectively. ranged between 1817 and 2379 cpm and 2142 and 2377 cpm. Significancevaluesfor differences between experimental groups are given in Results.

I-A

I-E

H-2K

H-2D

Leirhmonia

.

IFN-7 t

8

4 I

I

I

)

12 I

I

16 I

1

1

1929

SUPPRESSION OF MACROPHAGE H-2 ANTIGENS BY LEISHMANIA

-

H-2K

Figure 4. Parasite dose-response relatlonship for suppression of macrophage class I and class II antigen expression measured by binding of monoclonal antibodies directed atthe indicated determinant on normal (squares) or Infected (circles) cells cultured in the absence (closed symbols) or presence (open symbols] of to IFN (12.5 U/mI]. Cellswereexposed Lelshmanla at the indicated multiplicities of infection, and 4 hr later were either untreated or were stimulated with IFN (12.5U/ml). Antibody binding was assayed at 96 hr of culture. The data are the mean f SD of triplicate determinations for each cell population. Nonspecific binding of 6-90to control and infected macrophages, respectively, ranged between 3376and4281cpmand3700and4311 cpm. Data regarding intensities of infectlon and significance values for differences between experimental groups are given in Results.

x

5

CL

16-2j

12

-

3-

i

4

.-L

m

2.5

5

IO

20

2.5

40

5

IO

20

2.5

40

5

IO

20

40

Parasite infecting dose (amastigotes/macrophage) TABLE I IFN lnductlon of I - A expression on normal and L. donovani-infected macrophages cultured in the absence or presence of sodium

meclofenarnate" Expt. 1 Treatment of Macrophages I-A

-

+ + + +

+ +++

PGEl

Suppression of I-A

Expt. 2 I-A

Suppression PGEp

of I-A

~

-

51

+ ++ +

f

154 5082 1,172f452 727 f 2.003 75 f 138 14,984f 458 1.19Of 15.1 47 14 4,257f 1,062 2,188 f 1,548 72 66 446 f 167 75f238 18 403 47f 30 7.575f 315 92 & 35 6.966f 241 2,7665247' 462 9 26

199f235 1,046f 19 41 2 40 2,7142 93 f 283 937 f 161 f 435 2,541 f 908 f75 153 f 52 96f 166 2312 13 6.674f 687 172f 30 247f 58 44

90

a Macrophages werecultured with sodium meclofenamate for 1 hr at37°C before the addition of Leishmania at a multiplicity of infection of 20:1, Four hours later, the cells were either untreated or were stimulated with IFN (12.5U/ml). After 96 hr of culture, I-A expression was assayed by indlrect binding of monoclonal anti-I-A. and PGE. levels in cell supernatants were assayed by radioimmunoassay. b. The differences in I-A expression between IFN-stimulated Infectedcells cultured with and without meclofenamate were significant at the 5% level. Nonspecificbinding of 6-90to control and infected macrophages, respectively. ranged between5327 and 7768 cpm and 4382 to 7893 cpm. E

not differ from each other. Modulation of H-2K and H-2D in the basal state was found to differ from that of I-A. Thus, densitiesof class I antigens on cells infected at 2.5:l (followed by no IFN treatment) were suppressed by 5 0 % for both H-2D (p < 0.001) and H-2K [p < 0.05).Under these conditions, increasing parasite doses suppressed H-2K expression to nil without further altering H-2D. Macrophages infected at 2.5: 1 also had significantly suppressed responses for lymphokine inductionof H-2D (58%of control, p< 0.005) and H-2K (63%of control,p < 0.05). Increasing the parasiteinfecting dose to 5:l resultedin further decreases in responsiveness to IFN for induction of class I antigens. Maximal suppression was seen at this level where H-2D and H-2K were, respectively, 38% (p < 0.001) and 2 0 % (p < 0.001) of control. Effects of inhibition of macrophage cyclooxygenase activity on suppression of MHC class II antigens by L. donovani. To examine thepotential relationshipbetween parasite-induced increases in macrophage PGE2 synthesis and suppression of class I1 antigen expression, cells were incubated withthe cyclooxygenase inhibitor sodium meclofenamatebeginning 1 hr before infection. This treatment significantly augmentedI-A expression by IFNstimulated infected cells. The results of two representa-

tive experiments are shown in Table I. Consistent with our previous results (27), PGE2 synthesis by infected macrophages was markedly elevated (two to three times control). In response to IFN, expression of I-A by these cells was 2 8 % (Expt. 1) and 10% (Expt. 2) of control. Sodium meclofenamate (2 gM) effectively inhibited cyclooxygenase and reduced PGE2levels by morethan 90%. This was associated with 64% (Expt. 1, p < 0.05) and 80% (Expt. 2, p c 0.05) increases in I-A expression by IFN-treated infected cells. Inhibitortreatmenthadno significant effect on basal expression of I-A. Of interest was the finding thatIFN induction of I-A on meclofenamate-treated normal cells was only 44 to 5 0 % of that found on noninhibitor-treated cells. This result was not related to alterations incell viability. DISCUSSION

The findings of this study indicated that macrophages infected with L. donovani were suppressed in their ability to express class I1 MHC gene products in response IFN. to Parasite-inducedchanges in MHC antigens varied to some extent, depending on whether cells were in the basal state or were stimulated with IFN. Thus, in eight separate experimentsperformed under conditions of optimal IFN stimulation, the densities of I-A or I-E on in-

1930

SUPPRESSION OF MACROPHAGE H-2 ANTIGENS BY LEISHMANIA

fected cells were50 to 90%less than class I1 densities on acid (HETE)formation via lipoxygenationof arachidonate noninfected macrophages. IFN induction of class I anti- (27). This interpretationof the present result is consistgens was also impaired by Leishmania, and relative to ent with the findingsof Snyder et al. (30).who showed normal cells this effect was greater for H-2K than H-2D. that mono-HETE stimulate expressionof class I1 antigens For both classI and classI1 antigens, the response IFN to on murine macrophages. In this regard, in our previous correlated inversely with the intensity of infection, and studies we found that 0.2 pM meclofenamate inhibited parasite-induced suppression ofMHC antigens was not mono-HETE formation by 40 to 60%. Divergent effects overcome by increased concentrationsof lymphokine. of meclofenamate on normal (suppression of I-A) vs inUnlike the situation of IFN stimulation, basal expres- fected (stimulationof I-A)macrophages waslikely related sion of class I1 antigens on infected cellswas not reduced to the predominance of cyclooxygenase enzyme activity relative to normal macrophages. This was likely related in infected cells. Thus, inhibition of this pathway in to the nonconstitutive expressionof these gene products infected macrophages would have removed the suppres(28.29)which resulted innegligible levels on control cells sive influenceof PGE2 and haveprovided more substrate by 96 h r of culture. In contrast, class I antigens, which for conversion to I-A enhancing lipoxygenase products. are expressed constitutively, were suppressed by Leish- On the other hand, inhibitor treatment of normal cells mania in the basal state. Suppression of class I antigens may have resulted ina net reduction in basal lipoxygenunder basal conditions occurred with infecting inocula ase activities (and mono-HETE formation) and consea s low as 2.5 amastigotes per macrophage, and as the quent down-regulationof I-A. intensity of infection increased, the level of H-2K became With respect to the overall host-parasite interaction, several implications may be drawn from the findings of undetectable. Preliminary experiments using fluorescein-conjugated the present study. First, suppression of class I1 antigen anti-class I1 monoclonal antibodies and flow cytometry expression on Leishmania-infected macrophageswould demonstrated that the induction of class I1 antigens on likely contribute to defective induction of anti-leishmannormal cells by IFN was uniform across thecell popula- ial T cell responsiveness. This inference is based on a n obligatory tion (data not shown). Similarly, suppression of this re- exhaustive body of evidence demonstrating the sponse by leishmania infection was also uniform a n d role of accessory cell class I1 molecules in antigen-specific 1 7 , 1 8 3 4 ) .Furthermore, reports was not restricted to specific subpopulations of cells. T cell activation (1 1. 12. level of Furthermore, resultsof cell volume studies doneby FACS from several laboratories have indicated that the T lymphocyte stimulation obtained is proportional to the analyzer (data not shown) showed that normal and infected cells were equivalent in cell volume in the absence density of class I1 molecules on antigen-presenting cells and presenceof IFN treatment. These two determinations(34-37). In this regard, amplification of accessory cell T indicated that decreased expression of class I1 antigens, lymphocyte interactions is believed to occur through the as measured in the indirect binding assay, reflected a n induction byIFN of class I1 molecules on accessory cells. Thus. suppressionof this response to IFN by L. donovani actual decrease in antigen density on parasitized cells. of parasitized Expression of class I1 gene products has been shown to would be expected to impair the ability T cells. be highly regulated and is subject toa variety of negative accessory cells to activate leishmania-specific influences including oxygenated products of arachidonic Such a mechanism could account, at least in part, for obacid (30, 31) glucocorticoids, and a-fetoprotein (32, 33). deficientparasite-specificcell-mediatedimmunity Evidence for the role of arachidonic acid metabolites is served in both experimental and clinical forms of leishmaniasis (5- 10). clearest for E-series prostaglandins which down-regulate A second possible consequence of decreased class I1 class I1 antigens, although it also been has suggestedthat lipoxygenase metabolites of arachidonic acid also influ- density on cells infected with L. donovani is related to ence the expression of these MHC gene products (30). mechanisms of parasite killing. Elimination of LeishPrevious results from this laboratory showed that mac- mania from parasitized macrophages has been shown to and -independent mechrophages infected with L. donovani synthesized and re- occur by both oxygen-dependent leased markedly increased quantities of PGE2 (27). We anisms (38-41). Both of these processes were shown to by lymphokines and did not therefore examined the potential role of this system in be subject to modulation On the down-regulating IFN induction of class I1 gene products require direct lymphocyte-macrophage contact. on leishmania-infected cells. Consistent with our pre- other hand, other investigators have describedan alternative mechanism of macrophage activation for leishviousstudies,conditionedmediafrom96-hr-infected manial killing. This system required direct contact of Lytcellswerefoundtocontaintwotothreetimesmore I-A-reimmunoreactive PGEz t h a n did supernatants from nor- 1+2-lymphocyteswithmacrophages,andwas mal cells. Incubation of both infected and controls cells stricted, noncytotoxic, and independent of lymphokines of these studies, decreased with 2 pM meclofenamate resulted in greater than 90% (42-44). Thus, in the context I-A density on L. donovani-infected cells would be exinhibition of PGEz synthesis (both in the absence and of parasite killing. presence of IFN). This effect was accompaniedby signif- pected to result in decreased efficiency A third potential result of suppression of class I1 antiicant increases in I-A expression by IFN-stimulated ingen expression during L. donovani infection relates to fected cells. This finding provided evidence to support the hypothesis thatdown-regulation of MHC class I1 a n - modulation of macrophage IL 1 production and to the findings of the preceding companion paper (22). I t h a s tigens was related to parasite-induced alterations in macbeen proposed that macrophage class I1 antigen expresrophage arachidonic acid metabolism. sion is required forIL 1 production (45, 46). This concluTheobservationthatmeclofenamatetreatmentreduced IFN induction of I-A on normal cells may have beension is derived from studies in which treatmentof macrelatedtoinhibition of mono-hydroxyeicosatetraenoic rophages with anti-classI1 antibodies resulted in inhibi-

SUPPRESSION OF MACROPHAGE H-2 ANTIGENS BY LEISHMANIA

tion of IL 1 synthesis and, secondarily, to decreased T cell proliferation. The latter was reversed by the provision of exogenous IL 1. In the context of these results, it was proposed that contact of T cells with class I1 antigens on accessory cells represents a mechanism for induction of macrophage IL 1 synthesis (45).In the preceding companion paper (22), we reported that L. donouani is capable of infecting macrophages without eliciting a n IL 1 response. This observation was somewhat surprising in light of the knowledge that a wide variety of microbes, microbial products, and phagocytic stimuli are capable of directly eliciting macrophage IL 1 synthesis (47).This result suggested, therefore,that Leishmania had evolved a mechanism for evading direct stimulation of this fundamental host-defense mechanism.If, on the other hand, T cell recognition of macrophage class I1 antigens represents an alternative mechanism for IL 1 induction. it might be inferred that down-regulation of class I1 antigens by Leishmania represents a n additional means by which these organisms evade the macrophage IL 1 response. At present, inferencesregarding the importance of suppression of class I antigens on macrophages infected with L. donouani must remain tentative.It is of interest, however, to note that although Sypek and Wyler (44) found cell contact-dependent leishmanial killing to be principally I-A-dependent, less significant contributions for K and D loci-encoded antigens were also observed. Thus, it is possible thatthe markedsuppression of K and D molecules observed in the present study might also contribute to deficient parasite killing. Thepersistence of Leishmania withinmammalian hosts suggests that theseorganisms have evolved means of subverting the immune response. A logical point at which to invoke such mechanisms would be at theearly host-parasite interface. The resultsof this study and the preceding companion report (22) indicate that L. donouani is capable of both evading and subverting critical macrophage accessory functions. These findings may be related to the frequentlyineffectualhostimmune responses observed during infections withthis organism.

1931

eds., McGraw-Hill, New York. P. 270. 6 Pearson. R. D.. D. A. Wheeler. L. H. Harrison, and H. D. Kay. 1984.

The immunobiology of leishmanias. Rev. Infect. Dis. 5907. Iha, and A. D. M. Bryceson. 1983. Immunosuppression in Kenyan visceral leishmaniasis. Clin. Exp. Immunol. 51:207. 8 . Reiner, N. E. 1982. Host-parasite relationship in murine leishmaniasis: pathophysiological and immunological changes. Infect. I r n rnun. 38: 1223. 9. Carvalho, E. M.. R. S. Teixeira, and W. D. Johnson. 1981. Cellmediated immunity in American visceral leishmaniasis: reversible immunosuppression during acute infection. Infect. Immun. 33:498. 10 Carvalho, E. M., R. Badaro, S. G. Reed, T.C. Jones, and W. D. Johnson. 1985. Absence of gamma interferon and interleukin 2 productionduringactivevisceralleishmaniasis. J. Clin. Inuest. 76:2066. 1 1 Unanue, E. R. 1981. The regulatory role of macrophages in antigenic stimulation. 11. Symbiotic relationshipbetweenlymphocytesand macrophages. Adu. Immunol. 31:l. 12 Unanue. E.R.. D. L. Beller, C. Y.Lu,and P. M. Allen. 1984. Antigen presentation: comments on its regulation and mechanism. J. Immunol. 132: 1 . 13 Chestnut. R., S. Colon, and H. M. Grey. 1982. Antigen presentation by normal B cells, B cell tumors, and macrophages: functional and biochemical comparison. J. Irnmunol. 128: 1764. 14. Hirschberg, H.. L. R. Braathen, and E. Thorsby. 1982. Antigen presentation by vascular endothelial cells and epidermal Langerhans cells: the role of HLA-DR. Immunol. Rev. 66:57. 15. Steinman. R. M., and M. C. Nussenzweig. 1980. Dendritic cells: features and functions.Immunol. Rev. 53:127. 16. Sunshine, G. H., D. R. Katz, andM. Feldman. 1980. Dendritic cells induce Tcell proliferation to synthetic antigens under Irgene control. J. Exp. Med. 152: 1817. 17. Benacerraf, B. 1981. Role of MHC gene products in immuneregulation. Science212:1229. 18. Schwartz. R. H. 1985. T-lymphocyte recognition of antigen in association with gene productsof the major histocompatibility complex. Annu. Rev. Imrnunol. 3:237. 19. Mizel, S. B. 1982. Interleukin 1 and T cell activation. Immunol. Reu. 63:51. 20. Palacios. R. 1982. Mechanisms of T cell activation: role and functional relationship ofHLA-Dr antigens and interleukins. Immunol. Rev. 63: 73. 21. Durum, S. K., J. A. Schmidt, and J. J.Oppenheim. 1985. Interleukin 1: a n immunological perspective. Annu. Rev. Immunol. 3:263. 22. Reiner, N. E. 1987. Parasite-accessory cell interactions in murine leishmaniasis. I. Evasion and stimulus-dependent suppression of the macrophageInterleukin 1 response by Leishmania donouani. J. Immunol. 138:1919. 23. Ozato. K.. N. M. Mayer. and D. H. Sachs. 1980. Hybridoma cell lines H-2 and Ia antigens. J. Irnmunol. secreting antibodies to mouse 124:533. 24. Ozato. K., N. M. Mayer, and D. H. Sachs. 1982. Monoclonal antibodies to mouse major histocompatibility complex antigens. 11. A series of hybridoma clones producing anti-H-2d antibodies and examination of expression of H-2d antigens on the surface of these cells. Transplantation 34:113. 25. Weaver, M. S., L. Sikora, and J. G. Levy. 1982. The immune response to ferredoxin: characterization of a major idiotype in serum Acknowledgments. We are grateful to Dr.David H. using monoclonal antibody derived by cell fusion. Mol. Imrnunol. Sachs for his generous gift of monoclonal anti-class I 19:105. antibodies, and to Dr. H. Michael Shepard of Genentech 26. McMaster, W. R., and A.F. Williams. 1979. Identification ofIa glycoproteins in rat thymus and purification from rat spleen.J.Eur. Corp. for providing recombinant murineIFN-y. We thank Irnrnunol. 9:426. Mr. Randy Chiu for providing the anti-ferredoxin reagent, 27. Reiner, N. E., and C. J. Malemud. 1985. Arachidonic acid metabolism by Leishmania donouani infected murine peritoneal macroand Dr. Fumio Takei for provocative and helpful discusphages: in vitro evidence for parasite-induced alterations cyclooxin sions regarding the work. Ms. Cora Austria provided valygenase and lipoxygenase pathways. J. Immunol. 134:556. uable assistance in preparation of the manuscript. 28. Steinman. R.M., N. Nogueira. M. D. Witmer. J. D. Tydings, and I. S. Mellman. 1980. Lymphokine enhances the expression and synthesis of la antigens on cultured mouse peritoneal macrophages.J. REFERENCES Exp. Med. 152: 1248. 29. Beller. D. J., and E. R. Unanue. 1981. Regulation of macrophage 1. Skov, C.B., and D. W. Twohy. 1974. Cellular immunity to Leishof la antigensby peritoneal populations. 11. Synthesis and expression mania donouani. I. The effect of T cell depletion on resistance to L. exudate macrophages is a transient event.J. Immunol. 126:263. donouani in mice. J. Immunol. 113:2004. 30. Snyder. D. S.. D. I. Beller. and E. R. Unanue. 1982. Prostaglandins 2. Skov, C.B., and D. W. Twohy. 1974. Cellular immunity to Leishmodulate macrophage la expression. Nature299:163. mania donouani. 11. Evidence for synergy between thymocytes and 31. Steeg. P. S., H.M. Johnson. andJ. J. Oppenheim. 1982. Regulation lymph node cells in reconstitution of acquired resistance to L. donof immune macrophage Ia antigen expression by a n immune interouani in mice. J. Imrnunol. 113:2012. feron-like lymphokine: inhibitory effect of endotoxin. J. Immunol. 3. Mauel, J.. and R. Behin. 1976. Experimental cutaneous leishmaniasis. V. Protective immunity in subclinical and self-healing infection 129:2402. 32. Aberer, W.. L. Stingl, S. Pogantsch, andG. Stingl. 1984. Effects of in the mouse.Clin. Exp. Imrnunol. 23: 126. 4 . Preston, P. M., and D.C. Dumonde. 1982. Leishmaniasis: immunity, glucocorticosteroidson epidermal cell-induced immune response. J. Immunol. 133:792. immunopathology and immunodiagnosis. In Immunology of Parasitic Infections. s.Cohen and K. s. Warren, eds.Blackwell Scientific 33. Lu,C. Y.,P.S. Changelian. andE. R. Unanue. 1984. Alpha-fetoproPublications, Oxford. P. 299. tein inhibits macrophage expression of la antigens. J. Immunol. 5. Wyler, D. J., and P. D. Marsden. 1983. Leishmaniasis. In Tropical 132: 1 722. and GeographicalMedicine. K. S . Warren and A. A. F. Mahmoud, 34. Kurt-Jones, E. A., H. W. Virgin, .and E. R. Unanue. 1985. Relation 7. Ho, M. D., K. Koech, D.W.

1932

SUPPRESSION OF MACROPHAGE H-2 ANTIGENS BY LEISHMANIA

ship of macrophage la and membrane IL 1 expression to antigen presentation. J. Irnmunol. 135:3652. 35.Janeway. C. A,, K. Bottomly. J . Babich, P. Conrad, S . Conzen. B.Jones, J . Kaye. M. Katz, L.McVay. D. B. Murphy, and J. Tite. 1984. Quantitative variation in Ia antigen expression playsa central role in immune regulation. Imrnunol. Today 5:99 36. Nunez, G., E.J. Ball, L. K.Myers, and P. Stastny. 1985. The density of class I1 molecules influences allogeneic, autologous and antigen specific T cell responses in man. Fed.Proc. 44:561. 37. Beller. D. I. 1984. Functional significance of the regulation of macrophage Ia expression. Eur. J. Irnmunol. 14: 138. 38. Chang, K. P.. and J. W. Chiao. 1981. Cellular immunity of mice to Leishmania donouani in-vitro: lymphokine-mediated killing of intracellular parasites in macrophages. Proc. Natl. Acad. Sci. USA 78: 7083. 39. Nacy, C. A., M. S. Meltzer. E. J. Leonard, and D. J. Wyler. 1981. Intracellular replication and lymphokine-induced destruction of Leishmania trapica in C3H/Hen mouse macrophages. J. Imrnunol. 127:2381. 40.Murray, H. W., and D. M. Cartelli. 1983. Killing of intracellular Leishmania donouaniby human mononuclear phagocytes. Evidence for oxygen-dependent and independent leishmanicidal activity. J.

Clin. Inwst. 72:32. 41. Murray, H. W., B. Y. Rubin, and C. D. Rothemel. 1983. Killing of intracellular Leishmania donouanf by lymphokine-stimulated human mononuclear phagocytes. Evidence that interferon-gamma is the activating lymphokine. J. Clin. Inuest. 72:506. 42. Panosian. C. B., J. P. Sypek,and D. J. Wyler. 1984. Cell contactmediated activation of macrophages for anti-leishmanialdefense. I. Lymphocyte effector mechanism thatis contact dependent and noncytotoxic. J. Imrnunol. 133~3358. 43. Sypek,J. R.. C. B. Panosian. and D. J. Wyler. 1984. Cell contactmediated macrophageactivation forantileishmanial defense. 11. Identification of effector cell phenotype and genetic restriction. J. ~mrnunot.133~3351. 44. Sypek,J. P., and D. J. Wyler. 1985. Cell contact-mediated macrophage activationfor antileishmanial defense: mappingof the genetic restriction to the I region of the MHC. Clin. Exp. Irnrnunol. 62449. 45. Durum, S. K. and R. K. Gershon. 1982. Interleukin 1 can replace the requirement for I-A positive cells in the proliferation of antigen primed T cells. Proc. Natl. Acad. Sci. USA 79:4747. 46. Gilman, G. C., J. S. Rosenberg, and J. D. Feldman. 1983.Inhibition of interleukin synthesis and T cell proliferation by a monoclonal anti-la antibody. J. Imrnunol. 130: 1236. 47. Dinarello, C. A. 1984. Interleukin-1. Rev. Infect. Dis. 6:51.