CD8 cells of patients with diffuse cutaneous leishmaniasis display functional exhaustion: the latter is reversed, in vitro, by TLR2 agonists

Abstract

Leishmania mexicana (Lm) causes localized (LCL) and diffuse (DCL) cutaneous leishmaniasis. DCL patients have a poor cellular immune response leading to chronicity. It has been proposed that CD8 T lymphocytes (CD8) play a crucial role in infection clearance, although the role of CD8 cytotoxicity in disease control has not been elucidated. Lesions of DCL patients have been shown to harbor low numbers of CD8, as compared to patients with LCL, and leishmanicidal treatment restores CD8 numbers. The marked response of CD8 towards Leishmania parasites led us to analyze possible functional differences between CD8 from patients with LCL and DCL. We compared IFNγ production, antigen-specific proliferation, and cytotoxicity of CD8 purified from PBMC against autologous macrophages (MO) infected with Leishmania mexicana (MOi). Additionally, we analyzed tissue biopsies from both groups of patients for evidence of cytotoxicity associated with apoptotic cells in the lesions. We found that CD8 cell of DCL patients exhibited low cytotoxicity, low antigen-specific proliferation and low IFNγ production when stimulated with MOi, as compared to LCL patients. Additionally, DCL patients had significantly less TUNEL+ cells in their lesions. These characteristics are similar to cellular "exhaustion" described in chronic infections. We intended to restore the functional capacity of CD8 cells of DCL patients by preincubating them with TLR2 agonists: Lm lipophosphoglycan (LPG) or Pam3Cys. Cytotoxicity against MOi, antigen-specific proliferation and IFNγ production were restored with both stimuli, whereas PD-1 (a molecule associated with cellular exhaustion) expression, was reduced. Our work suggests that CD8 response is associated with control of Lm infection in LCL patients and that chronic infection in DCL patients leads to a state of CD8 functional exhaustion, which could facilitate disease spread. This is the first report that shows the presence of functionally exhausted CD8 T lymphocytes in DCL patients and, additionally, that pre-stimulation with TLR2 ligands can restore the effector mechanisms of CD8 T lymphocytes from DCL patients against Leishmania mexicana-infected macrophages.

Figure 1. Cytotoxicity assay of CD8 against autologous Leishmania-infected macrophages (MOi).

A) Culture dotplots (annexin V vs. 7AAD) from MO, MOi, CD8/MO, and CD8/MOi of LCL and DCL patients. In top, the R1 region corresponds to the CD14⁺ cells analyzed. The percentage of annexin V⁺ cells are shown. B) Analysis of flow cytometry assay. Cytotoxicity index (C.I.) was calculated from 10 LCL and 4 DCL patients. C) Analysis of Cr⁵¹ release assay. The % cytotoxicity was calculated from 9 LCL and 4 DCL patients. The bars show the mean of C.I. and % cytotoxicity. p was obtained by the Mann-Whitney U test.

Figure 2. Cytotoxic mechanism analysis.

A) LCL patient CD8 were labeled for granzyme B or FasL in CD8 alone or co-cultured with MOi (CD8+MOi) 1∶1 for 4 h. R1 was analyzed from the CD8 vs FSC dotplot (top). Data are representative of 10 LCL patients. B) Mean fluorescence intensity (MFI) for Granzyme B and FasL expression were analyzed in 10 LCL patients. T-student test was carried out between CD8 and CD8+MOi (p<0.01).

Figure 3. Lymphocyte proliferation in LCL and DCL patients.

A) CFDA-labeled PBMC were incubated with PBS (CNT), MOi or Con A [5 µM] for seven days. CFDA^low cell percentages of total CD8 are shown. Representative dotplots from 8 LCL and 4 DCL patients are shown. B). Proliferation analysis of CD8 from 8 LCL and 4 DCL patients, stimulated with PBS (black bars), MOi (white bars) and Con A (striped bars) are shown (p<0.01).

Figure 4. IFNγ production by CD8 from LCL and DCL patients.

A) PBMC were non-stimulated (CNT), stimulated with PMA-Ionomycin during 4 h or with MOi during 24 h. Density diagrams of IFNγ vs. CD8 from 10 LCL and 4 DCL patients are shown. B) CD8⁺IFNγ⁺ percentage analysis from 10 LCL and 4 DCL patients are shown. C) ELISA results of IFNγ production in supernatants of the co-incubation CD8-MOi from 9 LCL and 4 DCL patients are shown (p<0.01).

Figure 5. Apoptosis in LCL and DCL patient lesions.

Frozen tissue sections of LCL (A) and DCL (B) were stained with TUNEL. Two representative photographs are shown. C) Double staining of LCL patient tissue: TUNEL in peroxidase (brown), CD68 in phosphatase (red). Black arrows show TUNEL⁺ cells, red arrows show TUNEL⁺ CD68⁺ cells. D). Positive percentage analysis for TUNEL of 7 LCL and 5 DCL patients is shown. Three fields were counted, 200 cells per field (p<0.05). Bar  = 20 µm.

Figure 6. Effector functions of CD8 from DCL patients were restored by pre-stimulation with TLR2 ligands.

CD8 from DCL peripheral blood were purified and incubated in L. mexicana LPG [10 µg/ml] (CD8^L) or Pam3Cys [2 µg/ml] (CD8^P) for 24 h and washed twice. CD8 were analyzed for: A) Cytotoxicity, B) IFNγ production and C) Proliferation. One representative of 3 DCL patients is shown. D) Analysis of mean fluorescence intensity (MFI) for PD-1 expression on non-stimulated (CNT), LPG or Pam3Cys-stimulated CD8 from 3 DCL patients is shown (*p = 0.02, **p = 0.005).