CD4⁺ T Cells Which React to the Leishmania major LACK Antigen Rapidly Secrete Interleukin-4 and Are Detrimental to the Host in Resistant B10.D2 Mice

Abstract

Leishmania major induces the rapid production of interleukin-4 (IL-4) in both susceptible BALB/c and resistant B10.D2 mice. In both strains, IL-4 is produced by T cells which react to the parasite LACK (for Leishmania homolog of the receptor for activated C kinase) antigen. The rapid production of IL-4 in B10.D2 mice does not confer susceptibility but results in increased parasite burdens.

Susceptibility of BALB/c mice to the intracellular parasite Leishmania major is associated with a strong Th2 response (17). In contrast, resistance of other strains correlates with a Th1 response. Since interleukin-4 (IL-4) is required for the differentiation of naive T lymphocytes into Th2 effector cells (23), the role of this cytokine in susceptibility was studied extensively. Although early experiments have shown that BALB/c mice treated with neutralizing anti-IL-4 monoclonal antibodies (MAbs) at the time of infection developed a protective Th1 response and healed (21), recent results with mutant strains were more difficult to interpret. Thus, while IL-4-deficient BALB/c mice were resistant to L. major in one study (9), BALB/c mice carrying a similar mutation remained susceptible in another one (15).

Other investigators have undertaken a detailed analysis of the early production of IL-4 in resistant and susceptible strains. It was previously shown that the amount of IL-4 mRNA induced by L. major 4 days after infection does not predict susceptibility or resistance (19, 22). In contrast, an 8- to 10-fold increase in the amount of IL-4 transcripts was detected 16 h after infection in susceptible BALB/c mice but not in resistant C57BL/6, C3H/He, and CBA mice (12). This early peak of IL-4 was not detected in BALB/c mice made resistant to L. major by the administration of gamma interferon (IFN-γ) or IL-12 (12). In contrast, early IL-4 production was found in C57BL/6 mice made susceptible by treatment with anti-IFN-γ MAbs (12). To further investigate the causal relationship between the rapid accumulation of IL-4 transcripts induced by L. major and susceptibility, we compared the early production levels of IL-4 in BALB/c and B10.D2 mice, which have the same major histocompatibility complex (MHC) d haplotype but differ in their susceptibility to infection. We also performed similar studies with susceptible BALB.B and resistant C57BL/6 mice, which have the same MHC b haplotype.

Early accumulation of IL-4 mRNA in mice infected with L. major.

In the first experiment, we infected susceptible BALB/c and BALB.B mice and resistant B10.D2 and C57BL/6 mice with 10⁷ L. major (strain WHOM/IR/−/173) stationary-phase promastigotes. Draining lymph nodes (LN) were removed 20 h after infection and analyzed for the presence of IL-4 mRNA by reverse transcription-PCR as described previously (18). In agreement with previous reports (12), we found that L. major induced the early accumulation of IL-4 mRNA in BALB/c but not in C57BL/6 mice (Fig. 1A). An early burst of IL-4 mRNA was also observed in BALB.B mice. However, in comparison to the six- to sevenfold increase in the amount of IL-4 mRNA found in BALB/c mice, the burst exhibited by BALB.B mice was smaller. Unexpectedly, L. major also induced the accumulation of IL-4 mRNA in resistant B10.D2 mice. Thus, although the early production of IL-4 may be necessary for susceptibility, it is clearly not sufficient. Accordingly, resistance was found to be under the control of several genes (1, 20) and independently determined by both T-cell and non-T-cell compartments (24).

FIG. 1.

Early accumulation of IL-4 mRNA in susceptible and resistant mice infected with L. major. Mice of the indicated inbred strains (A) or IE-LACK transgenic mice and their negative littermates on the indicated background (B) were infected (solid bars) or not (empty bars) with 10⁷ L. major promastigotes. RNA was extracted from the draining LN 20 h after infection, and the relative levels of IL-4 mRNA were determined by semiquantitative reverse transcription-PCR. Data are means for two individual mice per group and are expressed in arbitrary units. Data are representative of three different experiments.

LACK-specific T cells are responsible for the rapid production of IL-4 induced by L. major in both BALB/c and B10.D2 mice.

Launois et al. have shown that the CD4⁺ T cells which rapidly secrete IL-4 in BALB/c mice express Vα8 and Vβ4 (11). Since CD4⁺ T cells which react to the parasite LACK (for Leishmania homolog of the receptor for activated C kinase) antigen also express Vα8 and Vβ4, it was proposed that these T cells were responsible for the early production of IL-4 in this strain (11). In order to further investigate this hypothesis, IE-LACK transgenic mice that were tolerant to LACK as the result of the transgenic expression of this antigen in the thymus (8) were backcrossed for 11 generations onto the BALB/c and B10.D2 backgrounds, respectively. IE-LACK transgenic mice and their negative littermates were infected with L. major and analyzed 20 h later for expression of IL-4 mRNA in their draining LN. As expected, L. major induced a five- to sevenfold increase in the amount of IL-4 mRNA in IE-LACK-negative littermates. As already observed for mice of inbred strains (Fig. 1A), the amount of IL-4 mRNA induced by the parasite was slightly higher in B10.D2- than in BALB/c-derived mice. In contrast, no IL-4 burst was found in IE-LACK transgenic mice, whether they were on the BALB/c or B10.D2 background (Fig. 1B). Thus, LACK-specific T cells were responsible for the very rapid production of IL-4 that was induced by L. major in both BALB/c and B10.D2 strains.

LACK-specific T cells express a Th2 phenotype in both BALB/c and B10.D2 mice.

We have previously shown that LACK-specific T cells express a Th2 phenotype in BALB/c mice (8). Since LACK-specific T cells rapidly secrete IL-4 in B10.D2 mice, we expected that these T cells would also express a Th2 phenotype in this latter strain. To investigate this hypothesis, B10.D2 and control BALB/c mice were infected with 2 × 10⁶ L. major promastigotes. Six days after infection, CD4⁺ T cells were purified from the draining LN by negative selection by using sheep anti-rat antibody-coated Dynabeads as described previously (16). A total of 5 × 10⁵ purified CD4⁺ T cells were incubated in vitro with 5 × 10⁵ mitomycin C-treated syngeneic splenocytes with or without an optimal concentration of LACK (amino acids 158 to 173; FSPSLEHPIVVSGSWD)- or GP63 (amino acids 364 to 378; GSCTQRASEAHASLL)-derived peptides (8, 25). Supernatants were collected 48 h later, and cytokine contents were measured by enzyme-linked immunosorbent assay as described previously (8). In contrast to T cells from naive mice which did not secrete any detectable amount of IL-4, IL-5, or IFN-γ when stimulated with LACK or GP63 peptides (data not shown), cells from both BALB/c and B10.D2 infected mice secreted high levels of IL-4 and IL-5 but almost no IFN-γ in response to the LACK peptide (Fig. 2). In contrast, T cells from both strains secreted IFN-γ but no IL-4 or IL-5 when stimulated with the GP63 peptide (Fig. 2). Thus, our data suggest that the nature of the early T-cell responses directed to individual parasite determinants depends mainly on the epitopes themselves, with non-MHC determinants playing a limited role. This is in contrast to the tendency of B10.D2 and BALB/c mice to mount polarized Th1 and Th2 responses, respectively. However, it was proposed that this latter phenomenon was linked to differences in the ability of BALB/c and B10.D2 mouse T cells to maintain responsiveness to IL-12, a phenotype which is controlled by a single locus on murine chromosome 11 (5–7). Since infection with L. major is characterized by the delayed induction of IL-12, the fact that the genetic background of the strain does not influence the development of these early-activated T cells could be explained.

FIG. 2.

Phenotype of LACK- and GP63-reactive T cells in infected mice. CD4⁺ T cells from the draining LN of BALB/c and B10.D2 mice were prepared 6 days after infection and incubated for 48 h without antigen (none) or with an optimal concentration of the indicated peptides. Cytokine contents were measured by enzyme-linked immunosorbent assay. Results from a representative experiment (of three) are shown. When cells from naive BALB/c or B10.D2 mice were incubated with LACK or GP63 peptides, the cytokine contents were below the limits of detection (1.4, 0.5, and 0.8 U of IFN-γ, IL-4, and IL-5, respectively, per ml).

It is not known why the T cells which react to this LACK peptide express a Th2 phenotype. In addition, it remains to be determined whether T cells which react to other LACK or GP63 epitopes express a Th2- or Th1-biased phenotype, respectively. Variables in the development of Th1 or Th2 cells include the amount of priming antigen (2), the affinity of the determinant for MHC class II molecules (10), and the nature of the antigen-presenting cells (4). Pertinent to this point may be the fact that GP63 is one of the most abundant proteins expressed in promastigotes, while LACK represents less than 0.1% of the parasite proteins.

B10.D2 mice made tolerant to LACK develop smaller lesions and exhibit decreased Th2 responses in comparison to their negative littermates.

Since IL-4 is critical for the development of counterprotective Th2 cells, we sought to determine whether the early production of IL-4 which was induced by L. major in B10.D2 mice would regulate the immune response directed toward the parasite. To this end, IE-LACK transgenic mice on the B10.D2 background and their negative littermates were infected with L. major and the thickness of the infected footpads was monitored with a metric caliper. We found that IE-LACK transgenic mice developed smaller lesions than their negative littermates (Fig. 3A). These differences correlated with reduced parasite burdens in the draining LN and in the footpads, as determined 5 weeks after infection by limiting dilution analysis (26) (Fig. 3B). The increased ability of IE-LACK transgenic mice to heal correlated with a 20-fold reduction in the number of IL-4-secreting parasite-specific T cells in their draining LN, as measured by an enzyme-linked immunosorbent spot (ELISPOT) assay following in vitro stimulation with soluble leishmania antigens (SLA) (Fig. 3C). In contrast, the numbers of IFN-γ- and IL-5-secreting cells were not significantly different in IE-LACK transgenic mice and in their negative littermates. Thus, the early production of IL-4 by LACK-specific T cells in B10.D2 mice was not sufficient to confer susceptibility, but it could impair their ability to eliminate the parasite. A similar phenomenon was observed in genetically resistant transgenic mice in which IL-4 was expressed by B cells and in B10.D2 mice infected with an adenovirus carrying the IL-4 gene (3, 14). The counterprotective role of IL-4 in these resistant strains could be explained by the ability of this cytokine to downregulate the antimicrobial activity of the macrophages or to reduce T-cell responsiveness to IL-12 (13).

FIG. 3.

LACK-reactive T cells express a Th2 phenotype and are detrimental to the host in both BALB/c and B10.D2 mice. IE-LACK transgenic mice and their negative littermates on a B10.D2 background were inoculated with 2 × 10⁶ stationary-phase L. major promastigotes. (A) Lesion development in IE-LACK transgenic mice (filled symbols) and their negative littermates (empty symbols). Each data point represents the mean lesion size for 10 mice ± standard deviation. (B) Parasite burden in the tissues of IE-LACK transgenic mice (solid bars) and of their negative littermates (empty bars) 5 weeks after infection. (C) Frequencies of cytokine-secreting cells in IE-LACK transgenic mice and in their negative littermates. CD4⁺ T cells were prepared from the draining LN of the indicated mice 10 days after infection. Cells were incubated for 72 h with syngeneic antigen-presenting cells and an optimal concentration of SLA. Live cells were recovered, and the numbers of cells secreting IL-4 (filled bars), IFN-γ (empty bars), and IL-5 (hatched bars) were measured by an ELISPOT assay. The number of cytokine-producing cells detected when cells from B10.D2 naive mice were incubated with SLA was below 20/10⁶ cells.