Leishmania mexicana Infection Induces Impaired Lymph Node Expansion and Th1 Cell Differentiation Despite Normal T Cell Proliferation¹

Infections involving the protozoan parasite Leishmania can result in a diverse array of clinical outcomes that are determined by both the species or strain of Leishmania and the host’s immune status. In several mouse strains, Leishmania major induces a strong IL-12-driven Th1 response and a self-resolving infection (1, 2). In contrast, Leishmania mexicana infections are often associated with chronic, nonresolving lesions and poor production of Th1 cytokines (3, 4). Th2 cytokines have been associated with the chronic nature of this infection (5, 6), although the phenotype we observe is best characterized by the absence of a strong Th1 or Th2 response (7).

We hypothesized that defects in T cell proliferation might underlie the negligible immune response observed following L. mexicana infection. Consistent with this idea, IL-12 fails to promote healing when administered to mice infected with L. mexicana (4), which might be explained by a lack of sufficient proliferating T cells available to respond to IL-12. Alternatively, immunosuppressive mechanisms may interfere with Th1 cell differentiation. Previous studies have shown that Ab-opsonized parasites may induce increased IL-10 production and thereby limit immune responses (8). IL-10-deficient (IL-10^−/−) mice initially display a course of infection similar to that observed in control mice. However, after several months, IL-10^−/− mice are ultimately able to resolve their lesions, and this phenotype is observed in FcγR-deficient mice as well (7). These results suggest that one mechanism contributing to chronic L. mexicana infection may be the production of IL-10 by macrophages exposed to opsonized parasites. However, because IL-10^−/− mice exhibit the same early susceptibility as wild-type controls, it is possible that the IL-10-mediated suppression is secondary to a primary defect in the generation of an adequate T cell response.

To determine whether differences exist in T cell responses to L. major and L. mexicana during the first few weeks of infection, we tracked the behavior of CFSE-labeled naive T cells following adoptive transfer into congenic mice. For both naive and Ag-experienced CD4⁺ and CD8⁺ T cells, we found that L. major and L. mexicana elicit comparable levels of T cell proliferation in vivo. However, we discovered that L. mexicana fails to induce expansion of the lymph nodes (LNs)³ draining the site of infection. Therefore, although the percentages of responding T cells in the draining LNs were similar, the absolute number of responding T cells was dramatically reduced. The impaired expansion of draining LNs appeared to be related to a reduced recruitment of lymphocytes from the circulation. We also discovered that those T cells that did proliferate following infection with L. mexicana were impaired in their ability to differentiate into IFN-γ-producing cells. The reduced LN expansion was not recovered in the absence of IL-10, although IFN-γ production was improved.

The poor capacity for Ag-experienced T cells to produce IFN-γ in response to L. mexicana correlated with the failure of L. major-primed T cells to confer protection against secondary challenge with L. mexicana. However, robust Th1 differentiation could be recovered by the administration of immunostimulatory CpG-containing oligodeoxynucleotides (ODNs) at the time of infection. These studies demonstrate that the primary failure to control L. mexicana is associated with profound deficits in LN expansion, lymphocyte recruitment, and the differentiation of T cells into IFN-γ-producing cells.

C57BL/6J (B6), B6.PL-Thy1^a/Cy (B6 Thy 1.1), and B6.129P2-Il10^tm1Cgn/J (IL10^−/−) mice were obtained from The Jackson Laboratory. B6-Ly5.2/Cr (CD45.1) congenic and C57BL/6 NCr (CD45.2) control mice were obtained from the National Cancer Institute (Frederick, Maryland). Animals were maintained and experiments were conducted in a specific pathogen-free environment in accordance with guidelines established by the University of Pennsylvania Institutional Animal Care and Use Committee (Philadelphia, PA).

L. major (MHOM/IL/80/Friedlin), L. mexicana (MNYC/BZ/62/M379), or CPB-deficient (Δcpb) promastigotes from the same L. mexicana strain (9) were grown at 27°C in Grace’s insect medium (Invitrogen Life Technologies) supplemented with 20% heat-inactivated FBS and 2 mM glutamine. For live infections, stationary-phase promastigotes from day 7 cultures were washed twice in PBS and counted. Mice were infected by injecting 2 × 10⁶ parasites suspended in 50 μl of PBS into the hind footpad (s.c.). Mice receiving ODNs containing CpG motifs were coinjected with 50 μg of stimulatory ODNs (1826; 5′-TCC/ATG/ACG/TTC/CTG/ACT/TT-3′) synthesized on a phosphorothioate backbone (Coley Pharmaceuticals). Lesion development was monitored by subtracting the baseline footpad thickness of the uninjected foot from the footpad thickness of the infected foot. Footpad thickness was measured using digital calipers (Mitutoyo).

Donor cells for adoptive transfer were isolated from naive mice, mice previously infected with L. major and allowed to heal (L. major-immune), or mice infected with the L. mexicana deletion mutant (Δcpb). Briefly, single cell suspensions of pooled spleens and LNs were depleted of RBC using ammonium chloride potassium (ACK) lysing buffer (Cambrex). CD3⁺ T cells were purified using T cell columns (R&D Systems) according to the manufacturer’s protocol. To assess cellular proliferation, CD3⁺ T cells were stained with CFSE dye as previously described (10). Purified CD3⁺ T cells (10–20 × 10⁶) were transferred i.v. into naive congenic B6 recipient mice that were infected on the following day with 2 × 10⁶ L. major or L. mexicana parasites. To neutralize IL-10 in vivo, mice were injected with 1 mg of Ab against the IL-10 receptor (American Type Culture Collection, clone 1B1.3a), per mouse every 3 days.

Cells were isolated from spleens and LNs at various times after infection and stimulated with 50 ng/ml PMA, 500 ng/ml ionomycin, and 10 μg/ml brefeldin A (all from Sigma-Aldrich) for 6 h. Cells were harvested and washed in FACS staining buffer (PBS containing 0.1% BSA and 0.1% sodium azide) and incubated with Fc block (50 μg/ml 2.4G2 and 500 μg/ml rat Ig). Cells were then stained with fluorochrome-conjugated mAbs purchased from eBioscience or BD Pharmingen against appropriate cell surface markers, including CD4 (clone GK1.5 or RM4-5), Thy1.1 (clone OX-7), Thy1.2 (clone 53-2.1), B220 (clone RA3-6B2), CD8α (clone 53-6.7), CD11c (clone N418), CD45.1 (clone A20), and CD45.2 (clone 104). For intracellular cytokine detection, surface-stained cells were fixed in PBS with 2% paraformaldehyde and then permeabilized with 0.2% saponin (Sigma-Aldrich) in staining buffer before incubation with fluorochrome-conjugated mAbs against IFN-γ (clone XMG1.2), IL-4 (clone 11B11), or isotype-specific control Abs before data acquisition on a BD FACSCanto or FACSCalibur flow cytometer (BD Pharmingen).

Statistical significance was determined using paired, two-tailed Student’s t tests, and results were considered significant with a p < 0.05.

A striking difference exists in the ability of L. major vs L. mexicana parasites to promote expansion of the LNs draining the site of infection, with L. major infection inducing >7-fold greater cellularity than that observed in response to L. mexicana (Fig. 1,A, left). The failure of L. mexicana to drive LN growth does not appear to directly limit the expansion of any particular lymphocyte subset within the LN, as the relative proportions of lymphocytes were similar following infection with either parasite (Fig. 1 A, right).

We hypothesized that decreased T cell proliferation might contribute to both the poor LN expansion and the limited Th1 responses observed in mice infected with L. mexicana. To directly visualize L. mexicana-induced T cell proliferation during a primary infection, naive C57BL/6 T cells were labeled with CFSE and adoptively transferred into Thy1-disparate mice. These recipients were then infected with either L. major or L. mexicana for 4 wk, and the proliferative responses of donor T cells in the draining LNs were assessed by measuring CFSE dilution. Low levels of homeostatic CD4⁺ and CD8⁺ proliferation were observed in uninfected mice (Fig. 1,B, left). In contrast, substantial proliferation of both CD4⁺ and CD8⁺ donor T cells from the draining LNs (but not nondraining LNs) was observed by 4 wk following infection with either L. major or L. mexicana (Fig. 1,B). Surprisingly, there was no significant difference in the ability of T cells to proliferate in response to these two parasites, suggesting that no defect exists in the ability of L. mexicana to induce T cell proliferation during a primary infection (Fig. 1 C).

However, despite equivalent percentages of proliferating T cells, the absolute numbers of both CD4⁺ and CD8⁺ donor T cells responding to L. mexicana were significantly decreased relative to the numbers following L. major infection (Fig. 1 D). These results suggest that although L. mexicana induces ample T cell proliferation, the minimal LN expansion associated with this infection limits the overall magnitude of the Th1 response.

To address the apparent paradox between normal proliferation but substantially reduced LN expansion, we examined whether differential recruitment of naive cells into the draining LN might contribute to the decreased cellularity observed. Whole lymphocytes from spleens and LNs of naive CD45.1 mice were transferred into CD45.2 congenic mice that had been previously infected with either L. major or L. mexicana for 4 wk. Draining LNs were harvested from recipient mice 1 day later (to exclude the contribution of proliferation), and the numbers of donor cells that had trafficked into the LNs overnight were determined.

Similar to the dramatic reduction in the overall cellularity of draining LNs from L. mexicana-infected mice (Fig. 1,A), the number of total donor cells that had been recruited into the draining LNs after 1 day was drastically decreased in mice infected with L. mexicana compared with L. major (Fig. 2,A). Furthermore, the absolute numbers of all major lymphocyte subsets recruited into the draining LNs of L. mexicana-infected mice were significantly less than the numbers observed trafficking into the LNs of L. major-infected mice (Fig. 2 B).

Reduced Th1 cytokine responses during recall responses to L. mexicana have been previously described in vitro (7, 11), and a lower percentage of total CD4 T cells are observed to produce IFN-γ from L. mexicana-infected mice when stained ex vivo (Fig. 3,A). To better distinguish the Leishmania-specific differentiation of naive T cells into Th1 effectors, we used the same experimental approach described above to assess the frequency of responding cells differentiating into Th1 cells following L. mexicana infection. Consistent with endogenous host cell responses (Fig. 3,A), we found that only a small percentage of proliferating CD4⁺ T cells were capable of producing IFN-γ in L. mexicana-infected mice (Fig. 3,B, top). In contrast, high frequencies of IFN-γ⁺ responders were consistently observed in response to L. major infection. A similar trend in the ability of CD8⁺ T cells to produce IFN-γ was also observed, with nearly half of all proliferating CD8⁺ donor cells differentiating into IFN-γ-producers in response to L. major but not L. mexicana (Fig. 3,B, bottom). When absolute numbers of IFN-γ-producing donor T cells were calculated, the paucity of Th1-differentiated cells following L. mexicana infection was even more dramatic. L. major induced significantly more CD4⁺ and CD8⁺ IFN-γ producers (24- and 14-fold, respectively) than L. mexicana (Fig. 3 C). Therefore, these data reveal a profound impairment in Th1 cell differentiation of naive T cells following L. mexicana infection.

Although the percentage of IL-4-producing cells appeared slightly elevated in L. mexicana-infected mice, the frequency of IL-4-producers in either L. major- or L. mexicana-infected mice constituted only a small percentage of the total proliferating cells; furthermore, production of IL-13, another Th2-associated cytokine, was not significantly different following infection with either parasite (data not shown). Therefore, although roles for both of these cytokines have been demonstrated in the development or maintenance of chronic disease in other experimental models of L. mexicana infection (5, 6), our studies support previous observations (7, 12) that susceptibility to L. mexicana cannot simply be due to an overwhelming Th2 response.

We previously found that C57BL/6 mice resolve L. mexicana-induced lesions in the absence of IL-10 (7). Our results and those of others suggest that one primary source of IL-10 in mice infected with New World Leishmania parasites may be macrophages induced to produce IL-10 following the phagocytosis of Ab-opsonized parasites (7, 13). Therefore, we considered the possibility that poor Th1 cell development could be the result of increased or disproportionate IL-10 production by these innate immune cells following L. mexicana infection. To eliminate this factor, purified, naive CD3⁺ T cells were transferred into Thy1-disparate C57BL/6 or IL-10-deficient recipient mice, and IFN-γ production from donor cells was assessed. Interestingly, the absence of host-derived IL-10 in recipient mice did not facilitate the recovery of L. mexicana-induced Th1 differentiation relative to L. major. Instead, the percentages of both CD4⁺ and CD8⁺ responding T cells producing IFN-γ remained greatly diminished following L. mexicana infection (Fig. 3,D). The inability of responding T cells to differentiate into IFN-γ-producing cells following L. mexicana infection was accompanied by the same minimal expansion of draining LNs as was seen in wild-type mice (Fig. 3,E, left). Consequently, the absolute numbers of donor CD4⁺ and CD8⁺ T cells producing IFN-γ in response to L. mexicana were dramatically decreased as well (Fig. 3 E, right). These data therefore suggest that IL-10 production from innate cells is not responsible for the reduced Th1 cell responses seen early after L. mexicana infection.

Interestingly, when recipient mice were treated with mAbs targeting the IL-10-receptor (anti-IL-10R mAb) to neutralize the effect of IL-10 signaling from both host and donor cells, there was an observable increase in Th1 responses to L. mexicana infection. The percentage of proliferating donor T cells producing IFN-γ in response to L. mexicana was comparable to that observed following L. major in anti-IL10R mAb-treated mice (Fig. 3,F). However, when differences in overall LN expansion (Fig. 3,G) were taken into account, the magnitude of Th1 differentiation of donor T cells against L. mexicana remained significantly decreased (Fig. 3 H). These data indicate that T cell-derived IL-10 may play a role in limiting Th1 differentiation of naive T cells responding to L. mexicana infection. However, even in the absence of IL-10 signaling, L. mexicana elicits impaired LN expansion, which limits the magnitude of the Th1 response generated.

Ag-experienced T cells have a lower threshold for activation, and we next asked whether previously stimulated T cells would be better able to respond to L. mexicana infection. Because there is considerable cross-reactivity of T cells against different species of Leishmania, we first asked how well T cells from mice that had been previously infected with L. major would respond to L. mexicana. We previously demonstrated that Leishmania-responsive T cells from these mice include a population of central memory T cells that home to the LNs and require additional stimuli to differentiate into IFN-γ-producing Th1 cells (14). We chose to examine the in vivo responses of LN-homing cells by transferring T cells from L. major-immune mice into Thy1-disparate C57BL/6 recipients and comparing the proliferation and differentiation of donor T cells in the draining LNs in response to L. major or L. mexicana infection.

Consistent with the increased numbers of Leishmania-reactive T cells in L. major-immune mice, L. major-immune cells displayed greater responses (than naive cells) to infection with either parasite. In contrast to the moderate proliferation and minimal IFN-γ production observed after 4 wk in naive T cells, robust proliferation and IFN-γ production by primed CD4⁺ and CD8⁺ T cells could be readily observed by 2 wk of infection (Fig. 4,A). However, similar to our observations following naive T cell transfers, significantly higher percentages of IFN-γ-producing donor cells were present in L. major- vs L. mexicana-challenged mice. In addition, there continued to be a dramatic deficit in total LN expansion in response to L. mexicana infection, with similar reductions in absolute numbers of responding CD4⁺ and CD8⁺ T cells capable of producing IFN-γ (Fig. 4 B).

To ensure that differences in the reactivity of L. major-primed T cells to homologous vs heterologous parasites did not account for these results, we also examined the response of T cells isolated from mice previously immunized with a mutant L. mexicana parasite deficient in the virulence factor cysteine proteinase B (Δcpb) (9). Because of the reduced virulence of these parasites, mice infected with this mutant form of L. mexicana develop stronger Th1 responses, heal primary infections, and exhibit immunity against subsequent infections with wild-type L. mexicana parasites (9, 11, 15). Similar to primed cells from L. major immune mice, T cells from L. mexicana Δcpb-infected animals displayed poor differentiation into IFN-γ producers following challenge with L. mexicana as compared with L. major (Fig. 4,C). Furthermore, and consistent with all previously described naive and immune cell transfers, total cell numbers as well as absolute numbers of responding CD4⁺ and CD8⁺ donor cells producing IFN-γ in the draining LNs were significantly reduced following L. mexicana infection (Fig. 4 D). These results demonstrate that even previously activated T cells are deficient in their ability to differentiate into IFN-γ producers following L. mexicana infection, regardless of whether the cells were initially primed with L. mexicana or L. major.

In light of the dramatic impairment in the generation of IFN-γ-producing T cells from both naive and previously primed T cells in response to L. mexicana, we questioned whether L. major-immune mice would be resistant to challenge with L. mexicana. To test this, mice that had resolved an infection with L. major were challenged with L. mexicana, and the course of infection was monitored. Surprisingly, despite the poor capacity for T cells from these mice to differentiate into effector cells in response to L. mexicana, we found that L. major-immune mice were resistant to infection with L. mexicana (Fig. 5). However, these results are consistent with prior studies demonstrating cross-protection against Leishmania amazonensis gained by prior infection with L. major (16).

Thus, given the ability of L. major-immune mice to heal following infection with L. mexicana, we next asked whether T cells from L. major-immune mice would be sufficient to transfer protection against L. mexicana challenge. Consistent with previous results (14), the transfer of L. major-immune CD3⁺ T cells resulted in rapid control of L. major infection and lower parasite titers compared with the recipients of naive cells (Fig. 6,A). However, transfer of the same population of L. major-immune CD3⁺ T cells did not protect against L. mexicana infection. No reduction in lesion progression or parasite burden was observed in L. mexicana-infected mice receiving naive vs L. major-immune T cells (Fig. 6 B). These results suggest that further differentiation of primed T cells in the draining LN may be critical for the immunity of L. major-immune mice against L. mexicana, and that sufficient differentiation fails to take place in our adoptive transfer recipients. One factor that may contribute to the development or maintenance of protection in intact L. major-immune mice may be the presence of persistent L. major parasites to continuously expand an effector T cell pool from either naive or previously primed T cells.

Taken together, our results suggest a global defect in both LN expansion and priming for CD4 and CD8 T cells to become IFN-γ producers. TLR ligands, including ODNs containing CpG motifs, are commonly used to enhance immune responses, particularly with respect to Th1 polarization. Furthermore, CpG ODNs have been previously demonstrated to enhance resistance to various Leishmania species in other animal models (17, 18). Recent studies have additionally highlighted an integral role for dendritic cells (DCs) in the dramatic LN expansion that results from exposure to CpG (19). Therefore, we tested whether the coadministration of CpG ODNs at the time of L. mexicana infection would promote enhanced LN expansion, increase the number of IFN-γ producing cells, and lead to lesion resolution.

Endogenous T cells from mice treated with CpG at the time of infection displayed higher frequencies of IFN-γ than cells from either uninfected, L. mexicana-, or L. major-infected mice (Fig. 7,A). To determine whether CpG ODN might enhance in vivo the differentiation of IFN-γ producing T cells following L. mexicana infection, CD3⁺ T cells from mice previously infected with L. mexicana Δcpb were transferred into naive C57BL/6 recipients. Recipients were infected with L. mexicana and given CpG ODN at the time of infection. Donor T cells in recipients treated with CpG at the time of infection exhibited enhanced differentiation into IFN-γ producing T cells compared with mice infected with L. mexicana alone (Fig. 7,B). Similar patterns of Th1 differentiation were observed when CD3⁺ T cells from naive donor mice were transferred as well, albeit with delayed kinetics of proliferation and IFN-γ production (data not shown). Moreover, coadministration of CpG was associated with an increase in the size of draining LNs; a trend toward concomitant increases in absolute numbers of donor CD4⁺ and CD8⁺ responders producing IFN-γ was consistently observed as well, although this difference was not statistically significant (Fig. 7 C).

Finally, to determine whether the increased LN expansion and frequency of IFN-γ-producing cells was associated with resolution of L. mexicana infection, C57BL/6 mice were infected in the presence or absence of CpG ODN. Lesion progression and parasite burdens following infection were assessed over time. We found that mice receiving CpG ODN at the time of infection exhibited smaller lesion sizes and that these animals were better able to resolve these lesions than control mice (Fig. 8). Furthermore, lesion resolution was associated with lower parasite titers at the site of infection.

The tendency of South American species of Leishmania to trigger chronic, nonhealing lesions in mice has long been observed, but poorly understood (20). In this study, we provide the first analysis of the in vivo T cell response to L. mexicana. We discovered that L. mexicana infection is associated with reduced expansion, recruitment, and differentiation of T cells into IFN-γ producers in the draining LNs, although T cell proliferation following infection with L. mexicana is unimpaired. To determine whether Th1-polarizing adjuvants could overcome this deficit, we simultaneously injected mice with L. mexicana and CpG ODNs. CpG ODN treatment enhanced the expansion of the draining LNs, increased the percentage of T cells producing IFN-γ, and facilitated lesion resolution. Taken together, these results reveal a selective impairment in T cell priming that is apparent in both naive and previously primed T cells.

Several studies have shown that mice infected with New World species of Leishmania fail to develop a strong Th1 response, even in the absence of a strong Th2 response (7, 12, 21, 22, 23). Exogenous IL-12 was previously demonstrated to be insufficient to direct disease resolution in L. mexicana- or L. amazonensis-infected mice (4, 12). We found that the draining LNs are much smaller in L. mexicana-infected mice than those observed after L. major infection. Given the poor LN expansion and the inability of IL-12 to promote healing, we first hypothesized that impaired T cell proliferation might limit the accumulation of a pool of Leishmania-responsive cells and subsequent Th1 responses. However, our data demonstrate that no significant difference exists in the frequency of T cells proliferating in response to these two parasites.

Although equivalent T cell proliferation following infection with these parasites appears paradoxical in the face of the substantial difference in draining LN expansion, factors other than cell proliferation may play a larger role in determining the size of the draining LN (19, 24, 25). For example, experiments with the related parasite L. amazonensis suggest that this parasite fails to provoke the inflammatory response necessary for recruitment or migration of cells into the draining LN and the site of infection (21). Our results now demonstrate that decreased recruitment into the LN may also limit the expansion of draining LNs following L. mexicana infection. Additional experiments are underway to define the relative contributions of lymphocyte recruitment, proliferation, migration, and survival of specific LN cell subsets in determining LN size and cellularity following Leishmania infection.

IL-10 has the potential to influence Leishmania infections through the inhibition of various immune cell functions (2). Several studies have demonstrated that macrophages produce IL-10 following exposure to Ab-opsonized parasites (7, 8), which may then suppress the development of T cell responses in L. mexicana-infected mice. In the current study, reduced T cell responses were observed in the presence or absence of host-derived IL-10, suggesting that impaired early development of IFN-γ producing T cells in response to L. mexicana is not mediated by enhanced macrophage production of IL-10. Interestingly, treatment with blocking Abs against the IL-10 receptor reveals a role for T cell-derived IL-10 in regulating Th1 differentiation, although significant defects in LN expansion and the overall magnitude of the Th1 response generated remain obvious. Consistent with a minimal role for IL-10 in the early response to L. mexicana, we found that although mice are ultimately able to control L. mexicana infection in the absence of IL-10, healing is delayed until the mice are well into the chronic phase of infection (7). These results suggest that while IL-10 may contribute to immune suppression and parasite maintenance during chronic infection, there exists an early IL-10-independent defect in L. mexicana-induced LN expansion that limits the number of Leishmania-responsive T cells capable of producing IFN-γ.

The inability of adoptively transferred immune T cells from L. major-infected mice to protect against L. mexicana challenge was at first an unexpected result. This failure seems at odds with the strong immunity to L. mexicana observed in intact L. major-immune mice. Our interpretation of this result is that the preexisting frequency of Leishmania-responsive effector T cells isolated from L. major-immune mice is not sufficient to mediate protection without being constantly replenished from a pool of naive or central memory T cells. Even during secondary responses to L. major infection, we find that central memory T cells provide protection with slower kinetics than effector memory cells (14). Therefore, while the replenishment of Th1 effectors may take place in intact L. major-infected immune mice due to the presence of low numbers of L. major parasites, we hypothesize that L. mexicana is unable to efficiently drive the differentiation of these IFN-γ producing T cells from both naive and central memory T cells in otherwise naive recipients. In other model systems, the transfer of activated Th1 effector cells (generated in vitro) resulted in protection against L. amazonensis infection (21). Therefore, it is possible that increasing the dose of transferred cells and/or enriching the population for Leishmania-responsive effector or effector memory cells would transfer immunity to naive mice against L. mexicana infection.

What, then, is responsible for the poor immune responses seen following infection with these parasites? Although there are many possible cell types that could potentially contribute to poor Th1 priming, including B cells, NKT, γδ, and regulatory T cells, we hypothesize that the defects seen in the development of IFN-γ-producing T cells stem from reduced activation of DCs by L. mexicana vs L. major parasites. Although CpG has many effects on the immune system, our observation that CpG ODN coadministration can enhance the differentiation of T cells into IFN-γ producers is consistent with a DC defect that is overcome by activation via TLR9. Recent studies demonstrate that DCs are also necessary to initiate expansion of the LNs draining an inflammatory site (19, 25). Taken together, our results may therefore suggest that DCs are able to present Ag to induce T cell proliferation, but that these DCs are not sufficiently activated to promote LN expansion or T cell differentiation in response to L. mexicana. This partial activation status would be consistent with findings that indirect inflammatory signals can initiate DC maturation and Ag presentation but do not license the cells to promote full differentiation of effector T cells (26). Interestingly, the inflammatory stimulus provided by CpG administration not only promotes T cell differentiation but also induces LN expansion. Studies in our laboratory are currently underway to address the role of DCs in both LN expansion and T cell differentiation following Leishmania infection.

In summary, we have demonstrated for the first time that in vivo Th1 differentiation in response to L. mexicana infection is greatly impaired in both naive and previously primed T cells. Furthermore, T cell transfers demonstrate that Leishmania-specific T cells alone cannot transfer cross-protection against L. mexicana infection. However, despite drastic defects in both effector T cell development and overall LN expansion, the proliferative response of naive T cells to L. mexicana appears normal. These findings indicate that T cells fail to receive the necessary activating signals required for full T cell expansion and differentiation in response to L. mexicana. The ability of CpG to augment IFN-γ responses, increase LN expansion, and achieve control of infection may further support a role for DC impairment in susceptibility to L. mexicana and related species. We believe that our data call for more careful dissection of the initial interaction between these parasites and DCs or related cell types as well as further exploration of the immunogenic role of CpG or other TLR ligands to improve our understanding of the factors responsible for chronic cutaneous leishmaniasis.

We are grateful to Karen Joyce and Carlos Rodriguez for invaluable technical assistance, Jude Uzonna and Larry Buxbaum for insightful discussions and preliminary experiments that have contributed to a strong foundation for these and future L. mexicana studies, and members of the Department of Pathobiology for their useful feedback.

The authors have no financial conflict of interest.