Infection: Role of CD8+ T Cells

INFECrION AND IMMUNITY, Sept. 1993, p. 3730-3738 0019-9567/93/093730-09$02.00/0 Copyright X 1993, American Society for Microbiology Vol. 61, No. 9 Gamma Interferon Response in Secondary Leishmania major Infection: Role of CD8+ T Cells INGRID MULLER,1* PASCALE KROPF,' ROBERT J. ETGES,2 AND JACQUES A. LOUIS' World Health Organization Immunology Research and Training Centre' and Institute of Biochemistry, 2 University of Lausanne, CH-1066 Epalinges, Switzerland Received 9 April 1993/Returned for modification 21 May 1993/Accepted 14 June 1993 CD8+ T cells have been shown to contribute to the rapid resolution of secondary lesions developing in immune mice challenged with Leishmania major. In the present study, we assessed directly the participation of specific CD8+ T cells in the memory response induced in immune mice by reinfection. Lymphocyte populations from reinfected immune mice exhibit marked secondary gamma interferon (IFN-y) responses. The participation of IFN-'y-producing CD8+ T cells in the memory response elicited by secondary infectious challenge was demonstrated in both genetically resistant immune CBA mice and genetically susceptible immune BALB/c mice that were rendered resistant by administration of anti-cd4 monoclonal antibody in the early phase of the primary infection. The protective function of CD8+ T cells in experimental murine cutaneous leishmaniasis might thus be explained in part by their ability to secrete IFN-y. In this context, the neutralization of IFN-'y at the time of reinfection reduced the Leishmania-specific delayed-type hypersensitivity response, showing that this cytokine is involved in the recall of immunological memory to L. major in vivo. Leishmania spp. are obligate intracellular parasites of mammalian macrophages. Infection of mice with Leishmania major, one of the causes of human cutaneous leishmaniasis, results in clinical manifestations that mimic the spectrum of leishmaniases caused by different Leishmania species in humans. Mice from the majority of strains, such as CBA or C57BL/6, are resistant to infection in that they develop only small lesions that resolve spontaneously, leaving the animal immune to reinfection. In contrast, mice from a few strains, typified by BALB/c, are susceptible to infection and develop severe cutaneous lesions at the site of parasite inoculation, which is followed by visceralization of the infection and death (1, 10). The extreme susceptibility of BALB/c mice to infection with L. major can be overcome by a variety of experimental manipulations (6, 14, 38). For example, treatment of BALB/c mice with anti-cd4 monoclonal antibody (MAb) at the beginning of infection induces the expression of a resistant phenotype in these otherwise susceptible mice (21, 31, 39, 46). Although specific CD4+ T cells and the cytokines they produce make a significant contribution to the resolution of lesions induced by a primary infection with L. major (20, 22, 27, 41), indirect evidence indicates that CD8+ T cells can also play a role in immunity to a primary infection with both L. major and Leishmania donovani. CD8+ T cells from L. major-infected resistant CBA mice were shown to transfer Leishmania-specific delayed-type hypersensitivity (DTH) reactions to syngeneic recipients, providing evidence for the triggering of parasite-specific CD8+ T cells during infection with L. major (26). In contrast to normal immune resistant mice, which develop only small, rapidly healing secondary lesions, animals that receive anti-cd8 MAb at the time of reinfection develop severe, slowly healing secondary lesions (28). The anti-cd8 MAb treatment also reverses the immunity to reinfection induced in BALB/c mice made resistant either by vaccination with killed L. major promastigotes or * Corresponding author. Electronic mail address: imuller@ ulbiol.unil.ch. by intervention with anti-cd4 or with anti-interleukin 4 (IL-4) MAb (6, 30). Taken together, these data provide indirect evidence that Leishmania-specific CD8+ T cells contribute to the efficient control of secondary infection in immune mice, suggesting that CD8+ T cells participate in the secondary immune response triggered by reinfection of immune mice. Therefore, the purpose of this study was to assess directly the contribution of Leishmania-specific CD8+ T cells to the memory response triggered by the challenge of immune mice with L. major. It is indeed important to characterize carefully the cellular parameters of the memory response mounted by immune mice after reinfection since, ideally, vaccines should sensitize the host for the elicitation of protective memory responses upon infection. The observation that spleen and lymph node cells from reinfected, immune mice produce substantially more gamma interferon (IFN--y) in vitro in response to specific stimulation than immune cells obtained from unchallenged mice clearly indicates that a secondary response is triggered upon reinfection. The participation of CD8+ T cells in this secondary response elicited by reinfection was demonstrated by our findings that the CD8+ T cells present in the lymphoid tissues of immune reinfected mice mounted a much higher IFN-,y response upon specific stimulation in vitro than those from immune mice which had not been rechallenged. The production of IFN--y was used as a parameter for T-cell activation since IFN-y is essential for the efficient activation of parasitized macrophages, permitting them to kill intracellular L. major (35, 47). Finally, the contribution of IFN--y to the recall of immunological memory was also assessed by examining the effects of neutralization of this cytokine at the time of reinfection on the development of Leishmania-specific DTH. MATERIALS AND METHODS Mice. Adult CBA/J and BALB/c mice were purchased from IFFA-Credo, Saint-Germain-sur-l'Arbresle, France, and from HARLAN, OLAC, Zeist, Netherlands. 3730

VOL. 61, 1993 IFN-y RESPONSE IN SECONDARY L. MAJOR INFECTION 3731 Parasites. L. major LV 39 (MRHO/SU/59/P-strain) was maintained in virulent form by monthly passage in mice. Parasites isolated from skin lesions of infected mice were grown at 26 C in Dulbecco's modified Eagle's medium over rabbit blood agar (23). Infections and monitoring of lesions. Primary infection with 2 x 106 stationary-phase L. major promastigotes in 50,ul was performed by subcutaneous (s.c.) injection into the left hind footpad. Lesion development was monitored weekly by measuring the increase in thickness of the infected footpad compared with the uninfected contralateral footpad. Secondary infections were initiated only after the complete resolution of the primary lesions, with 2 x 106 stationary-phase L. major promastigotes in 50,u injected s.c. into the contralateral footpad. For secondary infections, the average footpad thickness at the time of reinfection was subtracted from the footpad thickness measured after reinfection. MAbs. Rat MAb GK 1.5 (immunoglobulin G2b [IgG2b], anti-cd4) (5) and anti-cd8 MAb 3.168.8.1(31M) (IgM) (40) were kindly provided by F. W. Fitch, University of Chicago, Chicago, Ill. Rat MAb H35.17.2 (IgG2b, anti-cd8) (36) was kindly supplied by M. Pierres, INSERM/CNRS, Marseille, France. Rat MAb AN-18.17.24 (anti-mouse IFN--y) was a gift from S. Landolfo, University of Torino, Torino, Italy (37). Rat MAb R4-6A2 (IgGl, anti-mouse IFN-y) (44) was obtained by G. Spitalny, and anti-mouse IFN--y MAb XMG1.2 (rat IgGl) (4) was kindly supplied by R. L. Coffman, DNAX, Palo Alto, Calif. Rat MAb RL 172.4 (IgM, anti-cd4) (3) was a gift from H. R. MacDonald, Ludwig Institute for Cancer Research, Epalinges, Switzerland. Induction of resistance to L. major infection in susceptible BALB/c mice by treatment with anti-cd4 MAb in vivo. Six hundred micrograms of anti-cd4 MAb GK1.5, precipitated by ammonium sulfate from ascites fluids, was injected intraperitoneally (i.p.) over a period of 24 h in 1 ml of phosphate-buffered saline into susceptible BALB/c mice during the first 12 days of infection (31). T-cell subset depletion. Single-cell suspensions (5 x 107 cells per ml) were incubated with a 1/10 dilution of hybridoma culture supernatant containing monoclonal IgM directed to CD4 (RL 172.4) or CD8 [3.168.8.1(31M)] T-cell surface antigens and a 1/10 dilution of a Low-Tox rabbit complement (Cedarlane, Hornby, Ontario, Canada). Control cell suspensions were treated with complement only. The T-cell-subset-depleted suspensions were washed three times before use, and the efficacy of killing was determined by fluorescence-activated cell sorter (FACS) analysis. Two-color fluorescence staining and FACS analysis. Samples of 2 x 106 cells were stained with a mixture of phycoerythrin-labeled anti-cd4 MAb (GK 1.5; Becton Dickinson, Mountain View, Calif.) and 5-(4,6-dichlorotriazinyl) aminofluorescein-conjugated anti-cd8 MAb (H35.17.2). Samples were analyzed on a flow microfluorometer (FACS II; Becton Dickinson) gated by a combination of narrowangle forward light scatter and perpendicular light scatter to exclude nonviable cells (24). Lymphocyte cultures. Unseparated or T-cell-subset-depleted spleen and lymph node cell suspensions were stimulated at the indicated cell numbers in the presence or absence of 5 x 106 live L. major promastigotes (inactivated by 2- to 5-min irradiation with UV light) per ml, in the presence or absence of 30 U of recombinant mouse IL-2 (rmil-2) per ml at 37 C under 7% CO2 in 24-well Costar plates in a final volume of 1 ml. rmil-2 expressed in X63Ag8-653 plasmacytoma cells was a gift from F. Melchers, Basel Institute for Immunology, Basel, Switzerland (16). Dulbecco's modified Eagle's medium (Seromed, Berlin, Germany) supplemented with 5% heat-inactivated fetal calf serum (Seromed), L-asparagine (36 mg/liter), L-flUtamine (216 mg/liter), L-arginine (200 mg/liter), 5 x 10- M 2-mercaptoethanol, 10 mm HEPES (N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid), 100 U of penicillin per ml, and 100,ug of streptomycin per ml was used as the culture medium. Twenty-four hours after initiation of the cultures, supernatants were removed and stored in small aliquots at -20 C until their IFN--y contents were measured. Detection of EFN-y. IFN-y was measured by enzymelinked immunosorbent assay as described previously (42) by using rat anti-mouse IFN-y MAb R4-6A2 and rat anti-mouse IFN--y MAb AN-18-17.24. Recombinant mouse IFN--y expressed in a B-lymphoma line (L1210), a gift from Y. Watanabe, Kyoto University, Kyoto, Japan (49), was used as a standard. DTH. The DTH responses were used to demonstrate the presence of sensitized T cells in immune mice in vivo. DTH was determined 24 h after reinfection by s.c. injection of 2 x 106 stationary-phase L. major promastigotes into the contralateral footpad of immune BALB/c and CBA mice by measuring the degree of footpad swelling with a dial-gauge caliper (Kr6plin, Schluchtem, Germany). Parasite burden. The numbers of living L. major in infected tissues were determined by using the parasite limiting dilution assay described by Titus et al. (48). Briefly, different dilutions of infected tissue homogenates were distributed in wells of microtiter plates containing blood agar slants. After 10 to 14 days of incubation at 26 C, the wells containing growing promastigotes were identified by microscopic examination. The frequency of L. major recovered from footpads, where the cutaneous lesions occurred, was determined by minimum x2 analysis applied to Poisson distribution. RESULTS Reinfection of L. major-immune mice triggers an enhanced IFN-y response by spleen and lymph node cells. The control and subsequent cure of the primary infection in either genetically resistant mice or in genetically susceptible mice rendered resistant by treatment with anti-cd4 MAb (here termed experimentally resistant mice) are mediated by the triggering of IFN--y-producing T cells (20, 30, 39). Although it is well known that these mice are immune to reinfection, their capacity to mount a secondary IFN--y response after challenge with L. major promastigotes has not yet been documented. Therefore, the capacity of immune resistant CBA mice and of immune, experimentally resistant BALB/c mice to mount a secondary IFN-y response after reinfection was investigated first. CBA mice and anti-cd4 MAb-treated BALB/c mice were infected in the footpad with L. major promastigotes, and the lesions were allowed to heal (for experimental protocol, see Fig. 1). After resolution of the primary lesion, mice from each group were reinfected in the contralateral footpad, and IFN--y production by spleen and lymph node cells, after 24 h of stimulation in vitro with live promastigotes, was measured and compared with that of similar cell populations taken from immune unchallenged mice. The results of a representative experiment are shown in Fig. 2. Clearly, lymphoid cells from reinfected immune mice of both strains secrete substantially more IFN--y upon antigenic stimulation in vitro. Compared with the unchallenged immune mice, reinfected immune mice produce 7 to 40 times more IFN-y as tested in several independent experiments. Further, the kinetics of

3732 MULLER ET AL. INFECTION " ~~~~~~~~~N 100-170 days 600 gg acd4 mab IREINFECTION I LDA _ / _ <KDTH H LDA HEALED /LDA Lz-- -_ 5-1 0 days \LZ~ K,DTH LDA INFECT. IMMUN. lymph node cells TOTAL CELLS FACS: %CD8+, %CD4+ STIMULATION IFN-y spleen cells KILLING OF CD4 FACS: %CD8+, %CD4+ STIMULATION IFN-y FIG. 1. Experimental protocol. Resistant CBA and susceptible BALB/c mice were infected s.c. on day 0 with 2 x 106 virulent L. major promastigotes. Susceptible BALB/c mice were rendered resistant experimentally by i.p. injection of 600,ug of anti-cd4 MAb during the first 12 days of infection. The development of the primary lesions was monitored weekly by measuring the footpad thickness. After complete resolution of the skin lesions (100 to 170 days postinfection), the parasite load in the footpads was determined by limiting dilution analysis (LDA) at the time lymphoid cells were stimulated for IFN-y production. Draining lymph node and spleen cells were depleted of CD4+ T cells by treatment with MAb and complement. The number of CD4' and CD8+ T cells in the total lymphoid cell populations as well as in the treated populations was determined by FACS analysis at the time of stimulation. The treated and control cell populations were stimulated for 24 h, and the culture supernatants were analyzed for their IFN-y content. Some of the immune animals from both strains of mice were reinfected s.c. with 2 x 106 L. major promastigotes in the contralateral footpad, and the DTH response was determined 24 h later. Five to six days after reinfection, cells from these mice were analyzed as described above. n = 10 to 15 mice per group. IFN--y production is more rapid in the secondary response than it is in the course of the primary infection (data not shown). Taken together, these results show that reinfection of immune mice induces the expression of a memory response manifested by the enhanced production of the macrophage-activating cytokine IFN--y. Specificity of the IFN-y production in vitro. IFN-y production by spleen cells in vitro was totally dependent upon the addition of specific antigen in culture (Fig. 3), i.e., live L. major promastigotes, while lymph node cells, although clearly stimulated by antigen, could produce some IFN--y without addition of exogenous antigen (Fig. 3). The apparently spontaneous release of IFN--y by lymph node cells was increased by the addition of rmil-2. The elevated background IFN--y release by lymph nodes cells, draining the site of infection, was clearly due to Leishmania antigens and a few live parasites in these organs (not shown). Exogenous rmil-2 (30 U/ml) did not increase the IFN--y production by spleen or lymph node cells in the presence of specific antigen. The amount of IFN-y produced was proportional to the number of cells in culture (Fig. 3). Under the culture conditions used, more than 1.5 x 106 cells per ml were required for clear IFN--y production. Leishmania-specific CD8+ T cells able to produce IFN'y are present in lymphoid organs of immune mice. Indirect evidence showed that CD8+ T cells contribute to the production of macrophage-activating lymphokines with demonstrable antileishmanial activity (30). Direct evidence that CD8+ T cells able to produce IFN--y in response to L. major are present in lymphoid tissues of immune mice was obtained by using CD8+ T cells positively sorted by FACS. CD8+ T cells (6 x 104) sorted from the spleens of immune CBA mice 5 days after reinfection were expanded in vitro for 4 weeks in the presence of syngeneic irradiated antigen-presenting cells (APC), IL-2, and living promastigotes as a source of antigen. Restimulation of 5 x 105 CD8+ T cells from these expansion cultures with L. major, in the presence of IL-2 and irradiated syngeneic APC, resulted in the production of 150 U of IFN--y per ml. The CD8+ phenotype of the cultured cells was confirmed by FACS analysis at the time of restimulation in vitro. Without added antigen, 78 U/ml was produced, probably because of a combination of transfer of antigen and the binding of exogenous IL-2 present in the control cultures to the IL-2 receptors expressed by the CD8+ T cells as a result of their culture and maintenance conditions. Contribution of CD8+ T cells to the memoxy IFN-y response triggered by reinfection of immune mice with L. major. Freshly isolated spleen and lymph node cells obtained from immune mice 5 days after reinfection were depleted of CD4+

VOL. 61, 1993 IFN-y RESPONSE IN SECONDARY L. MAJOR INFECTION 3733 200 BALB/c ( Spleen (D) l0 CBA Lymph nodes @) Spleen () Lymph nodes E U- 8 15C 10c I0-5C --IN I ~~N.D ~ N.D INDI healed reinfected healed reinfected healed reinfected healed reinfected FIG. 2. Comparison of the L. major-specific IFN--y production by lymphoid cells from healed versus reinfected CBA and BALB/c mice. Lymphoid cells from healed, immune CBA and BALB/c mice were stimulated (striped bars) or not (solid bars) with 5 x 106 viable, UV-inactivated L. major promastigotes per ml as the antigen. The amounts of IFN-y produced by 3 x 106 lymphoid cells per ml from healed, immune mice were determined after 24 h of stimulation and compared with those produced after 5 days of reinfection. The difference between replicate cultures was less than 10%. N.D., not detectable. T cells to analyze the capacity of the remaining CD8+ T cells to produce IFN-y (Fig. 1). Cell suspensions were treated with rat monoclonal IgM anti-mouse CD4 RL 172.4 and complement. Subsequent FACS analysis showed that the E SPLEEN 2500- _ > (D LN z~~~~a5 1,0000 0 3 6 12 24 0 3 6 12 Cells (106) FIG. 3. Specificity of IFN--y production. Five days after reinfection with L. major, unseparated spleen and lymph node cell populations from experimentally immune BALB/c mice were titrated and stimulated with 5 x 106 living L. major promastigotes per ml as the antigen (open squares) or with L. major promastigotes plus 30 U of rmil-2 per ml (solid circles), IL-2 alone (open triangles), or nothing (solid squares). Twenty-four hours later, culture supernatants were collected and tested for their IFN-y content. Unseparated lymph node cells (12 x 106) contained 4.1 x 106 CD4+ and 2.3 x 106 CD8+ T cells, and unseparated spleen cells (12 x 106) contained 2.5 x 106 CD4+ and 8 x 105 CD8+ T cells, as determined by FACS analysis at the time of stimulation. The difference between replicate cultures was less than 10%. CD4+ T cells represented only 0.1 to 1.0% of the population. The remaining cells (CD8+) produced substantial amounts of IFN-y (603 U/ml) only in response to antigenic stimulation with living promastigotes, provided that exogenous IL-2 was added. Similar CD4-depleted cell populations were prepared from the spleen and lymph node cells of immune and immune reinfected mice to compare the amounts of IFN-y produced by CD8+ T cells before and after reinfection. As shown in Fig. 4, CD8+ T cells from both spleen and lymph nodes, taken 5 days after reinfection of healed experimentally resistant BALB/c mice, released significantly more IFN-y in response to L. major than CD8+ T cells from control immune mice (i.e., not reinfected). The DTH response was measured 24 h after reinfection to confirm the presence of Leishmania-reactive T cells in immune mice in vivo (values are given in the figure legends). In addition, the parasite load of immune mice was determined before reinfection (Table 1), since lesion size alone is an insufficient criterion for the successful resolution of infection (13). The difference in IFN--y production between immune and immune reinfected mice was less pronounced when the parasite load of the former was high (data not shown). Similar results were obtained with CD8+ T cells from immune and immune reinfected CBA mice (Fig. 5), confirming that in both strains, reinfection elicits a secondary response of Leishmania-specific CD8+ T cells secreting IFN--y upon specific stimulation in vitro with live promastigotes. These results clearly show that CD8+ T cells contribute to the memory response triggered by reinfection with L. major. Immune lymphocyte populations, enriched in T cells by nylon wool purification, could not be utilized in these

3734 MULLER ET AL. INFECT. IMMUN. BALB/c SPLEEN i BALBc LYMPH NODES ( healed (8) reinfected I <) healed (D reinfected l~~~~~~~~~~~~~~ 3-2000 E 300- E U) 1500 v ND. I LD 7 14 28 2.6 N. 5.2 VA U0 10.5 N.D. 3.25 6.5 13 6.5 13 26 CD8+ T cells ( 105) FIG. 4. Comparison of IFN--y production by CD8+ T cells from healed and reinfected BALB/c mice. CD8+ T cells were obtained from spleens (A) or draining lymph nodes (C) of experimentally resistant immune BALB/c mice 130 days after s.c. infection or from spleens (B) and lymph nodes (D) 5 days after reinfection by treatment with anti-cd4 MAb and complement. Different cell concentrations were stimulated with 5 x 10' viable, UV-inactivated L. major promastigotes per ml and 30 U of rmil-2 per ml (striped bars) or with IL-2 alone (solid bars). Culture supernatants were collected 24 h later and tested for their IFN-y content. In the absence of both IL-2 and antigen, the CD8+ T cells did not produce detectable amounts of IFN-y. The numbers of CD8+ T cells and contaminating CD4+ T cells (<0.2%) were determined by FACS analysis at the time of stimulation. The difference between replicate cultures was less than 10%. The DTH response of the reinfected mice was 1.40 ± 0.1 mm 24 h after reinfection. N.D., not detectable. experiments because, compared with unfractionated cells, the eluted T cells released significantly less IFN--y upon stimulation with antigen and syngeneic APC from naive mice (Fig. 6), suggesting that activated responding T cells adhered to nylon wool. This reduced responsiveness could not be attributed to the use of APC from normal mice, since APC from either infected or normal mice are comparable in their TABLE 1. (a) co 200-4) z L loo Parasite load in the lesions of healed and reinfected micea Mouse No. of parasites/lesion strain ~Status of mice Left footpad Right footpad CBA Immune 36 NTb Immune and reinfected 18 7,428 BALB/c Immune 18 NT Immune and reinfected 18 59,014 a Genetically resistant CBA mice were infected s.c. in the left footpad with 2 x 106 virulent L. major promastigotes, and 103 days later, the number of viable parasites in the healed footpad was determined. A group of immune CBA mice was reinfected with 2 x 106 L. major promastigotes in the contralateral footpad, and 6 days after reinfection (105 days after primary infection), the numbers of viable parasites in both footpads were determined. Genetically susceptible BALB/c mice were also infected s.c. in the left footpad and were enabled to overcome their susceptibility by i.p. injection of 600 plg of anti-cd4 MAb. The parasite load in the footpads of immune BALB/c mice was determined 130 days after primary infection and 5 days after reinfection. b NT, not tested. capacity to trigger immune T cells separated by magnetic bead selection (data not shown). The number of CD8+ T cells obtained by fluorescence-actived cell sorting was not sufficient to study directly their capacity to release IFN--y; consequently, they would have required long-term expansion in vitro, which might have led to populations with properties different from those of freshly isolated CD8+ T cells. Therefore, this approach was not used to assess the CD8 memory response. Effect of IFN-y neutralization on the DTH response in immune resistant CBA mice. Immune resistant mice display a strong secondary immune response revealed in vivo by Leishmania-specific DTH. CD4+ T cells as well as CD8+ T cells contribute to DTH (28, 29). To assess the role of IFN-'y on the development of DTH, immune resistant mice received i.p. 500,ug of neutralizing anti-ifn--y MAb XMG 1.2 at the time of s.c. reinfection with L. major promastigotes. The results shown in Fig. 7 demonstrate that the neutralization of IFN--y in vivo reduced the DTH response by 50% 48 h after challenge. Thus, the IFN--y released by both CD4+ and CD8+ T cells upon reinfection is involved in the recall of immunological memory to L. major in vivo. DISCUSSION 1000 X c5 500_om In this study, we show that reinfection of L. majorimmune mice triggers a secondary or memory IFN--y response to which CD8+ T cells contribute. This participation of IFN--y-producing CD8+ T cells in the memory response IN

VOL. 61, 1993 IFN-y RESPONSE IN SECONDARY L. MAJOR INFECTION 3735 800-4000.--% 600 I 3000 z -. 0 c 400 z 200-0 8 16 32 2.9 5.7 11 8.3 16.5 33 3.4 6.8 14 CD8 T cells ( 105) S 2000! C 1000 _: FIG. 5. Comparison of IFN-y production by CD8+ T cells from healed and reinfected CBA-mice. CD8+ T cells were obtained from spleens (A) or draining lymph nodes (C) of healed CBA mice 126 days after s.c. infection or from spleens (B) and lymph nodes (D) 5 days after reinfection by treatment with anti-cd4 MAb and complement. Different cell concentrations were stimulated with 5 x 106 viable, UV-inactivated L. major promastigotes per ml and 30 U of rmil-2 per ml (striped bars) or with IL-2 alone (solid bars). Culture supernatants were collected 24 h later and tested for their IFN-y content. In the absence of both IL-2 and antigen, the CD8+ T cells did not produce detectable amounts of IFN-y. The numbers of CD8+ T cells and contaminating CD4+ T cells (<0.4%) were determined by FACS analysis at the time of stimulation. The difference between replicate cultures was less than 10%. The DTH response of the reinfected mice was 1.01 + 0.05 mm 24 h after reinfection. N.D., not detectable. elicited by secondary infectious challenge could be demonstrated in both genetically resistant immune CBA mice and experimentally resistant immune BALB/c mice. The IFN--y produced by lymphoid populations enriched in CD8+ T cells by treatment in vitro with anti-cd4 MAb and complement can be considered to result from CD8+ T cells for several reasons. First, direct proof of the ability of specific CD8+ T cells to produce IFN--y in vitro in response to stimulation with L. major was obtained by using fluorescence-activated-cell-sorted CD8+ T cells. Second, the IFN--y production by populations enriched in CD8+ T cells cannot be due to contaminating CD4+ T cells (<1%) since titration experiments, using unseparated lymphoid cells, revealed that no IFN-y could be detected at low cell densities (<3 x 105 cells per ml). Third, the addition of anti-major histocompatibility complex class I MAbs inhibited the IFN-y response by CD8+ T cells from both strains by 50 to 80% (data not shown). Fourth, the IFN--y response of cultures enriched in CD8+ T cells was both antigen and IL-2 dependent (Fig. 4), while that of the unseparated cells was only antigen dependent (Fig. 3). The IL-2 requirement for IFN--y production by CD8+ T cells in response to specific stimulation has been demonstrated in several antigenic systems (7, 17, 19). The strict antigen dependence of the response observed in the present study makes it unlikely that NK cells are involved to a significant degree, since although NK cells are triggered in Leishmania infection, their specificity for Leishmania antigens has not been shown. y/6 T cells from L. major-infected mice did not release detectable amounts of IFN-y under these conditions and therefore cannot account for the response (37a). In all experiments described here, CD8+ T cells obtained from immune mice 5 days after reinfection released substantially more IFN--y upon specific stimulation in vitro than those from immune unchallenged mice. The difference could be due either to an increase in the amount of IFN-y released by individual memory cells or to an increase in the frequency of responding memory cells after challenge in vivo. More recent results indicate that reinfection of immune mice leads to increased frequencies of responding CD8+ T cells (manuscript in preparation). CD8+ T cells have been shown to contribute to protective immune responses in Leishmania infection. Susceptible BALB/c mice, when thymectomized, lethally irradiated, and reconstituted with bone marrow cells, or when given anti- CD4 MAb, were able to control a primary infection with L. major (11, 12). The induced resistance was attributed to DTH-competent, radioresistant CD8+ T cells. In the murine model of generalized, visceral infections with L. donovani, an array of experimental results has provided evidence for a significant role of CD8+ T cells in immunity against this parasite. CD8+ T cells, in the presence of CD4+ T cells, were shown to be essential for both the resolution of primary infections and the resistance to reinfection of BALB/c mice with L. donovani (33). Indeed, both CD4+ and CD8+ T cells produce IFN-y and were shown to be required for hepatic 0

3736 MULLER ET AL. INFECT. IMMUN. E 200 0 0C - 0.5 E 100 Total cells Tcells FIG. 6. Effect of T-cell purification on the capacity of sensitized spleen cells to produce IFN-,y in response to L. major. Susceptible BALB/c mice were infected s.c. in one hind footpad with 2 x 106 virulent L. major promastigotes per ml. One hundred three days later, the spleens were removed. Unseparated spleen cells (4 x 106/ml) and nylon wool-purified T cells (1 x 106/ml) plus irradiated syngeneic APC (5 x 106/ml) were stimulated in the presence (striped bars) or absence (solid bars) of 106 living L. major promastigotes per ml as a source of antigen. The culture supernatants were tested for their IFN-y content 24 h later. One passage of the spleen cell suspension over nylon wool columns resulted in 70% Thyl+ cells, as determined by FACS analysis. The difference between replicate cultures was less than 10%. granuloma formation (45). Upon rechallenge with infectious parasites, the number of CD8+ T cells increased at the time when parasites were eliminated by the immune host (25). In mice expressing transgenic I-E molecules, a reduced ability to resist infection with L. donovani has been correlated with a reduced capacity to trigger IFN--y-producing CD8+ T cells (18). The activation of parasitized macrophages by IFN-,y in combination with other signals leads to destruction of intracellular L. major (2, 9, 15, 32, 34). The demonstration of an enhanced IFN--y production by memory CD8+ T cells in immune mice upon secondary infectious challenge could account for the role of these cells in conferring resistance to leishmaniasis (6, 28, 29, 30). Indeed, neutralization of endogenous IFN-y at the time of reinfection of immune mice with L. major results in a clear reduction of the DTH response (Fig. 7) and interferes with the resolution of secondary lesions (manuscript in preparation). A Plasmodium bergheispecific cytolytic CD8+ T-cell clone, which is able to lyse specific peptide-sensitized target cells, was shown to be able to elicit leishmanicidal activity in both syngeneic and allogeneic Leishmania mexicana-infected macrophages, provided that P. berghei peptide-sensitized macrophages were also present in the culture. This leishmanicidal activity was the result of an IFN--y-dependent bystander effect and was not due to the lysis of the infected macrophages (43). Indeed, direct cytolysis of parasitized macrophages by Leishmaniaspecific CD8+ T cells has not yet been demonstrated. Together, these results suggest that the Leishmania-specific CD8+ T cells generated during the course of infection exert 0.0 FIG. 7. Effect of neutralization of IFN--y on the development of the DTH response in immune resistant CBA mice. Resistant CBA mice were infected s.c. with 2 x 106 L. major promastigotes in the left footpad. Three months later, after complete resolution of the primary lesions, the DTH response was elicited by injection of 2 x 106 L. major promastigotes into the contralateral footpad (solid bars). One group of mice (striped bars) received at the time of challenge 500,ug of neutralizing anti-ifn-,y MAb (XMG1.2) i.p. Twenty-four and 48 h later, the DTH response was determined by subtracting the thickness of the right footpad before reinfection, which was used as a baseline, from the thickness of the challenged right footpad. Values are means ± standard errors of the mean (n = 5 mice per group). their host-protective role by activating infected macrophages to kill the obligate intracellular parasite. The expansion of Leishmania-specific CD8+ T cells implies that some parasite-derived peptides are able to gain access to the class I major histocompatibility complex pathway of antigen presentation, in spite of the phagolysosomal location of the amastigotes. Alternatively, the destruction of parasites by activated macrophages could lead to the release of parasite-derived antigenic peptides which could bind to major histocompatibility complex class I molecules expressed at the surface of noninfected or parasitized cells, where they could be recognized by CD8+ precursor cells. The latter mechanism is more consistent with the kinetics of appearance of the CD8+ T-cell-mediated antileishmanial effector function, which is more pronounced in the secondary responses in vivo. Accordingly, the destruction of parasites during the resolution of primary infections might be necessary to generate a sufficient amount of Leishmaniaderived peptides to sensitize parasite-specific CD8+ T cells. Since this process would require time, it could explain why CD8+ T cells appear to be more important for the host defense after reinfection. In addition, the well-characterized activation of Leishmania-specific CD4+ T cells at early stages of the infection could be required to provide a helper function necessary for the differentiation and maturation of CD8+ effector cells. Although the experiments reported here were designed to assess the participation of specific CD8+ T cells in response to secondary infection with L. major, it is clear that reinfection also leads to secondary responses of CD4+ T cells. Indeed, results of several experiments have shown that CD4+ T cells produce more IFN--y than CD8+ T cells 5 days after challenge of immune mice. However, recent data indicate that after reinfection, the expansion of specific CD8+ T cells is greater than that of CD4+ T cells (manuscript in preparation). Taken together, these results

VOL. 61, 1993 IFN-y RESPONSE IN SECONDARY L. AJOR INFECTION 3737 demonstrate the triggering of secondary CD8+ T-cell responses after infectious challenge of immune mice. These findings do not, of course, preclude an important role for Leishmania-specific CD4+ T cells in immunity to reinfection. The effector functions of Leishmania-specific CD8+ T cells might be synergized by the activity of CD4+ T cells. Although either T-cell subset alone can contribute to the host defense upon reinfection with L. major, the presence of both subpopulations is required for rapid and successful elimination of the parasites. Indeed, the depletion of both CD4+ and CD8+ T cells in immune resistant CBA mice completely abrogated their resistance to reinfection (28). Similarly, in mice infected with another intracellular protozoan parasite, Toxoplasma gondii, both CD4+ and CD8+ can be induced to release IFN--y in vitro. The IFN--y releasing CD8+ T cells, however, are the major effectors of immunity in vivo (8). The control of reactivation of chronic toxoplasmosis depends on the presence of both T-cell subsets as well as IFN-,y synthesis, which again appears to result from a synergy between CD4+ and CD8+ lymphocyte functions (7). In conclusion, we provide direct evidence for the elicitation of secondary CD8+ T-cell responses in immune mice after reinfection with L. major. Inasmuch as the memory CD8+T cells were able to secrete IFN--y, a key lymphokine activating macrophages to kill intracellular parasites, the results presented in this report offer a rationale for our previous observations demonstrating a role for CD8+T cells in the control of secondary infection with L. major. As yet, there exists no effective vaccine against leishmaniasis, nor is it known how the specific immune response should be targeted by a vaccine. To this end, the factors directing the immune response to host protection or disease progression have been the subject of many investigations. Our results indicate that CD8+ T cells, in addition to the CD4+ T-cell subsets, should be considered in the development of a vaccine against leishmaniasis. ACKNOWLEDGMENTS This work was supported by grants from the Swiss National Science Foundation (32-28872.90), the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Disease, and the Sandoz Research Foundation. We thank P. Zaech and C. Knabenhans for operating the FACS. REFERENCES 1. Behin, R., J. Mauel, and B. Sordat. 1979. Leishmania tropica: pathogenicity and in vitro macrophage function in strains of inbred mice. Exp. Parasitol. 48:81-91. 2. Bogdan, C., H. Moll, W. Solbach, and M. Rollinghoff. 1990. Tumor necrosis factor-a in combination with interferon-y, but not with interleukin 4 activates murine macrophages for elimination of Leishmania major amastigotes. Eur. J. Immunol. 20:1131-1135. 3. Ceredig, R, J. W. Lowenthal, M. Nabholz, and R. H. MacDonald. 1985. Expression of interleukin-2 receptors as a differentiation marker on intrathymic stem cells. Nature (London) 314: 98-100. 4. Cherwinski, H. M., J. H. Schumacher, K. D. Brown, and T. R Mosmann. 1987. Two types of mouse helper T cell clone. III. Further differences in lymphokine synthesis between Thl and Th2 clones revealed by RNA hybridization, functionally monospecific bioassays, and monoclonal antibodies. J. Exp. Med. 166:1229-1244. 5. Dialynas, D. P., Z. S. Quan, K. A. Wall, A. Pierres, J. Quintans, M. R. Loken, M. Pierres, and F. W. Fitch. 1983. Characterization of the murine T cell surface molecule, designated UT4, identified by monoclonal antibody GK1.5: similarity of L3T4 to the human Leu3/T4 molecule. J. Immunol. 131:2445-2451. 6. Farrell, J. P., I. Mulller, and J. A. Louis. 1989. A role for Lyt2' T cells in resistance to cutaneous leishmaniasis in immunized mice. J. Immunol. 142:2052-2056. 7. Gazzinelli, R., F. T. Hakim, S. Hieny, G. M. Shearer, and A. Sher. 1991. Synergistic role of CD4+ and CD8+ T lymphocytes in IFN--y production and protective immunity induced by an attenuated Toxoplasma gondii vaccine. J. Immunol. 146:286-292. 8. Gazzinelli, R, Y. Xu, S. Hieny, A. Cheever, and A. Sher. 1992. Simultaneous depletion of CD4+ and CD8+ T lymphocytes is required to reactivate chronic infection with Toxoplasma gondii. J. Immunol. 149:175-180. 9. Green, S. J., C. A. Nacy, and M. S. Meltzer. 1991. Cytokineinduced synthesis of nitrogen oxides in macrophages: a protective host response to Leishmania and other intracellular pathogens. J. Leukocyte Biol. 50:93-103. 10. Handman, E., R. Ceredig, and G. F. Mitchell. 1979. Murine cutaneous leishmaniasis: disease patterns in intact and nude mice of various genotypes and examination of some differences between normal and infected macrophages. Aust. J. Exp. Biol. Med. 57:9-29. 11. Hill, J. 0. 1991. Reduced numbers of CD4+ suppressor cells with subsequent expansion of CD8+ protective T cells as an explanation for the paradoxical state of enhanced resistance to Leishmania in T-cell deficient BALB/c mice. Immunology 72: 282-286. 12. Hill, J. O., M. Awwad, and R. J. North. 1989. Elimination of CD4+ suppressor T cells from susceptible BALB/c mice releases CD8+ T lymphocytes to mediate protective immunity against Leishmania. J. Exp. Med. 169:1819-1827. 13. Hill, J. 0., R. J. North, and F. M. Collins. 1983. Advantages of measuring changes in the number of viable parasites in murine models of experimental cutaneous leishmaniasis. Infect. Immun. 39:1087-1094. 14. Howard, J. C., G. Hale, and F. Y. Liew. 1981. Immunological regulation of experimental cutaneous leishmaniasis. IV. Prophylactic effect of sublethal irradiation as a result of abrogation of suppressor T cell generation in mice genetically susceptible to Leishmania tropica. J. Exp. Med. 153:557-568. 15. James, S. L. 1991. The effector function of nitrogen oxides in host defense against parasites. Exp. Parasitol. 73:223-226. 16. Karasuyama, H., and F. Melchers. 1988. Establishment of mouse cell lines which constitutively secrete large quantities of interleukin 2, 3, 4 or 5 using modified cdna expression vectors. Eur. J. Immunol. 18:97-104. 17. Kaufmann, S. H. E., E. Hug, and G. De Libero. 1986. Listeria monocytogenes-reactive T lymphocyte clones with cytolytic activity against infected target cells. J. Exp. Med. 164:363-368. 18. Kaye, P. M., A. Cooke, T. Lund, M. Wattle, and J. M. Blackwell. 1992. Altered course of visceral leishmaniasis in mice expressing transgenic I-E molecules. Eur. J. Immunol. 22:357-364. 19. Kelso, A., R. MacDonald, K. A. Smith, J.-C. Cerottini, and T. Brunner. 1984. Interleukin 2 enhancement of lymphokine secretion by T lymphocytes: analysis of established clones and primary limiting dilution microcultures. J. Immunol. 132:2932-2938. 20. Liew, F. Y. 1989. Functional heterogeneity of CD4+ T cells in leishmaniasis. Immunol. Today 10:40-45. 21. Liew, F. Y., S. Millott, R. Leichuck, S. Cobbold, and H. Waldmann. 1989. Effect of CD4 monoclonal antibody in vivo on lesion development, delayed-type hypersensitivity and interleukin 3 production in experimental murine cutaneous leishmaniasis. Clin. Exp. Immunol. 75:438-443. 22. Locksley, R. M., and P. Scott. 1991. Helper T-cell subsets in mouse leishmaniasis: induction, expansion and effector function. Immunoparasitol. Today 12/7:A58-61. 23. Louis, J. A., E. Moedder, R. Behin, and H. Engers. 1979. Recognition of protozoan parasite antigens by murine T lymphocytes. I. Induction of specific T lymphocyte dependent

3738 MULLER ET AL. proliferative response to Leishmania tropica. Eur. J. Immunol. 9:841-847. 24. MacDonald, H. R., R. K. Lees, B. Sordat, P. Zaech, J. L. Maryanski, and C. Bron. 1981. Age-associated increase in expression of the T cell surface markers Thy-1, Lyt-1, and Lyt2 in congenitally athymic (nu/nu) mice: analysis by flow microfluorometry. J. Immunol. 126:865-870. 25. McElrath, M. J., H. W. Murray, and Z. A. Cohn. 1988. The dynamics of granuloma formation in experimental visceral leishmaniasis. J. Exp. Med. 167:1927-1937. 26. Milon, G., R. G. Titus, J.-C. Cerottini, G. Marchal, and J. A. Louis. 1986. Higher frequency of Leishmania major-specific L3T4+ T cells in susceptible BALB/c as compared with resistant CBA mice. J. Immunol. 136:1467-1471. 27. Mitchell, G. F. 1984. Host protective immunity and its suppression in a parasitic disease: murine cutaneous leishmaniasis. Immunol. Today 5:224-226. 28. Muller, I. 1992. Role of T cell subsets during the recall of immunologic memory to Leishmania major. Eur. J. Immunol. 22:3063-3069. 29. Muller, I., J. Garcia-Sanz, R. Titus, R. Behin, and J. Louis. 1989. Analysis of the cellular parameters of the immune responses contributing to resistance and susceptibility of mice to infection with the intracellular parasite, Leishmania major. Immunol. Rev. 112:95-113. 30. Muller, I., T. Pedrazzini, P. Kropf, J. Louis, and G. Milon. 1991. Establishment of resistance to Leishmania major infection in susceptible BALB/c mice requires parasite-specific CD8+ T cells. Int. Immunol. 3:587-597. 31. Muller, I., T. Pedrazzini, and J. A. Louis. 1988. Experimentally induced cutaneous leishmaniasis: are L3T4+ T cells that promote parasite growth distinct from those mediating resistance? Immunol. Lett. 19:251-260. 32. Murray, H. W., G. L. Spitalny, and C. F. Nathan. 1985. Activation of mouse peritoneal macrophages in vitro and in vivo by interferon-y. J. Immunol. 134:1619-1622. 33. Murray, H. W., K. E. Squires, C. D. Miralles, M. Y. Stoeckle, A. M. Granger, A. Granelli-Piperno, and C. Bogdan. 1992. Acquired resistance and granuloma formation in experimental visceral leishmaniasis. Differential T cell and lymphokine roles in initial versus established immunity. J. Immunol. 148:1858-1863. 34. Nacy, C. A., B. J. Nelson, M. S. Meltzer, and S. J. Green. 1991. Cytokines that regulate macrophage production of nitrogen oxides and expression of antileishmanial activities. Res. Immunol. 142:573-576. 35. Nathan, C. F., H. W. Murray, M. E. Wiebe, and B. Y. Rubin. 1983. Identification of interferon-y as the lymphokine that activates human macrophage oxidative metabolism and antimicrobial activity. J. Exp. Med. 158:670-689. 36. Pierres, M., C. Goridis, and P. Golstein. 1982. Inhibition of murine T cell-mediated cytolysis and T cell proliferation by a rat INFECT. IMMUN. monoclonal antibody immunoprecipitating two lymphoid cell surface polypeptides of 94000 and 180000 molecular weight. Eur. J. Immunol. 12:60-69. 37. Prat, M., G. Gribaudo, P. M. Comoglio, G. Cavallo, and S. Landolfo. 1984. Monoclonal antibodies against murine -y interferon. Proc. Natl. Acad. Sci. USA 81:4515-4519. 37a.Rosat, J.-P., and J. A. Louis. Unpublished data. 38. Sadick, M. D., F. P. Heinzel, B. J. Holaday, R. T. Pu, R. S. Dawkins, and R. M. Locksley. 1990. Cure of murine leishmaniasis with anti-interleukin-4 monoclonal antibody. Evidence for a T-cell-dependent, interferon-y-independent mechanism. J. Exp. Med. 171:115-127. 39. Sadick, M. D., F. P. Heinzel, V. M. Shigekane, W. L. Fisher, and R. M. Locksley. 1987. Cellular and humoral immunity to Leishmania major in genetically susceptible mice after in vivo depletion of L3T4+ T cells. J. Immunol. 139:1303-1309. 40. Sarimento, M., A. L. Glasebrook, and F. W. Fitch. 1980. IgG or IgM monoclonal antibodies reactive with different determinants on the molecular complex bearing Lyt2 antigen block T cell mediated cytolysis in the absence of complement. J. Immunol. 125:2665-2672. 41. Scott, P., E. Pearce, A. W. Cheever, R. L. Coffman, and A. Sher. 1989. Role of cytokines and CD4+ T-cell subsets in the regulation of parasite immunity and disease. Immunol. Rev. 112:161-182. 42. Slade, S. S., and J. Langhorne. 1989. Production of interferon-y during infection of mice with Plasmodium chabaudi chabaudi. Immunobiology 179:353-365. 43. Smith, L. E., M. Rodrigues, and D. G. Russell. 1991. The interaction between CD8+ cytotoxic T cells and Leishmaniainfected macrophages. J. Exp. Med. 174:499-505. 44. Spitalny, G. L., and E. A. Havell. 1984. Monoclonal antibody to murine gamma interferon inhibits lymphokine-induced antiviral and macrophage tumoricidal activities. J. Exp. Med. 159:1560-1565. 45. Stern, J. J., M. J. Oca, B. Y. Rubin, S. L. Anderson, and H. W. Murray. 1988. Role of L3T4+ and Lyt2+ cells in experimental visceral leishmaniasis. J. Immunol. 140:3971-3977. 46. Titus, R. G., R. Ceredig, J.-C. Cerottini, and J. A. Louis. 1985. Therapeutic effect of anti-l3t4 monoclonal antibody GK1.5 on cutaneous leishmaniasis in genetically susceptible BALB/c mice. J. Immunol. 135:2108-2114. 47. Titus, R. G., A. Kelso, and J. A. Louis. 1984. Intracellular destruction of Leishmania tropica by macrophages activated with macrophage activating factor/interferon. Clin. Exp. Immunol. 55:157-165. 48. Titus, R. G., M. Marchand, T. Boon, and J. A. Louis. 1985. A limiting dilution assay for quantifying Leishmania major in tissues of infected mice. Parasite Immunol. 7:545-555. 49. Watanabe, Y., and T. Sakata. 1988. Highly efficient action of autocrine mouse interferon--y expressed via a retroviral vector. Eur. J. Immunol. 18:1627-1630.