Chemokine receptors CCR6 and CXCR3 are necessary for CD4(+) T cell mediated ocular surface disease in experimental dry eye disease

Abstract

CD4(+) T cells are essential to pathogenesis of ocular surface disease in dry eye. Two subtypes of CD4(+) T cells, Th1 and Th17 cells, function concurrently in dry eye to mediate disease. This occurs in spite of the cross-regulation of IFN-γ and IL-17A, the prototypical cytokines Th1 and Th17 cells, respectively. Essential to an effective immune response are chemokines that direct and summon lymphocytes to specific tissues. T cell trafficking has been extensively studied in other models, but this is the first study to examine the role of chemokine receptors in ocular immune responses. Here, we demonstrate that the chemokine receptors, CCR6 and CXCR3, which are expressed on Th17 and Th1 cells, respectively, are required for the pathogenesis of dry eye disease, as CCR6KO and CXCR3KO mice do not develop disease under desiccating stress. CD4(+) T cells from CCR6KO and CXCR3KO mice exposed to desiccating stress (DS) do not migrate to the ocular surface, but remain in the superficial cervical lymph nodes. In agreement with this, CD4(+) T cells from CCR6 and CXCR3 deficient donors exposed to DS, when adoptively transferred to T cell deficient recipients manifest minimal signs of dry eye disease, including significantly less T cell infiltration, goblet cell loss, and expression of inflammatory cytokine and matrix metalloproteinase expression compared to wild-type donors. These findings highlight the important interaction of chemokine receptors on T cells and chemokine ligand expression on epithelial cells of the cornea and conjunctiva in dry eye pathogenesis and reveal potential new therapeutic targets for dry eye disease.

Figure 1. Increased expression of CCR6⁺ and CXCR3⁺ cells in the cervical lymph node and at the ocular surface in response the desiccating stress.

A. The percentage of CCR6⁺CD4⁺ T cells in the superficial cervical lymph node (CLN) in mice exposed to no stress (NS) or five (DS5) days of desiccating stress (DS) was determined by flow cytometry. The results represent mean±SD of five samples per time point for a total of ten samples (N = 10) in two independent experiments. B. The percentage of CCR6⁺CD4⁺ T cells in the ocular surface (OS). The results represent mean±SD of five samples per time point for a total of ten samples (N = 10) in two independent experiments. C. The percentage of CXCR3⁺CD4⁺ T cells at the CLN. The results represent mean±SD of five samples per time point for a total of ten samples (N = 10) in two independent experiments. D. The percentage of CXCR3⁺CD4⁺ T cells at the OS. The results represent mean±SD of five samples per time point for a total of ten samples (N = 10) in two independent experiments. E. A representative dot plot of the percentage of CD4⁺ cells that are CCR6⁺ or CXCR3⁺ in the CLN.

Figure 2. CCR6KO and CXCR3KO mice are resistant to dry eye disease.

A. Corneal permeability - Mean±SD of Oregon green dextran (OGD) permeability into the corneal epithelium of wild-type (WT), CCR6KO and CXCR3KO mice without (NS) or with 5 days of desiccating stress (DS5). OGD permeability was measured in five to seven mice per strain/time point/experiment. The graph represents the combined results of two independent experiments with a total of 10-14 mice per experimental group. B. Representative pictures of OGD staining of each strain. C. CD4⁺ T cell density – Mean±SD of the number of CD4⁺ cells per 100 µm of conjunctival epithelium of WT, CCR6KO, and CXCR3KO mice without (NS) or with 5 (DS5) or 10 (DS10) days of desiccating stress. D. Representative pictures of CD4⁺ T cell infiltration in each strain. * - indicates goblet cells; ↓ - indicates CD4⁺ T cells; CJ – conjunctiva; K – cornea. E. Goblet cell density - Mean±SD of the number of goblet cells per mm of conjunctival epithelium of WT, CCR6KO and CXCR3KO mice without (NS) or with 5 (DS5) or 10 (DS10) days of desiccating stress. Three sections from three animals were examined (N = 9) for each time point/mouse strain for both CD4⁺ T cell and goblet cell density. F. Representative pictures of PAS staining of each strain.

Figure 3. Th1 and Th17 associated cytokines are produced in the cervical lymph node but not the ocular surface.

A. Gene expression of IFN-γ, IL-17A, and IL-13 in the superficial cervical lymph node (CLN) of wild-type C57BL/6, CCR6KO, and CXCR3KO exposed to DS for 5 days (DS5) or not exposed (NS). B. Gene expression of IFN-γ, IL-17A, and IL-13 the ocular surface of wild-type C57BL/6, CCR6KO, and CXCR3KO exposed to DS for 5 days (DS5) or not exposed (NS). Results represent the mean±SD expression 2-3 animals per strain per time point in two independent experiments with a total of five mice. *, P<0.05, **, P<0.01.

Figure 4. CCR6KO and CXCR3KO CD4⁺ T cells cannot mediate dry eye disease.

Wild-type (WT), CCR6KO, or CXCR3KO CD4⁺ T cells from NS or DS5 mice were adoptively transferred to T cell deficient RAG1KO mice. A. CD4⁺ T cell density in the conjunctival epithelium of RAG1KO after adoptive transfer. B. Representative pictures of CD4⁺ T cell infiltration in each strain after adoptive transfer. C. Goblet cell (GC) density in the conjunctival epithelium of RAG1KO after adoptive transfer. Three sections from three animals were examined (N = 9) for each time point/mouse strain for both CD4⁺ T cell and goblet cell density. D. Representative pictures of PAS staining of each strain after adoptive transfer.

Figure 5. Inflammatory cytokines are not produced in response to DS in the cornea and conjunctiva of CCR6KO and CXCR3KO CD4⁺ T cell recipient mice.

Wild-type (WT), CCR6KO, or CXCR3KO CD4⁺ T cells from NS or DS5 mice were adoptively transferred to T cell deficient RAG1KO mice. mRNA analysis for inflammatory cytokines and matrix metalloproteinases (MMP) was performed on the corneal (A) and conjunctival (B) tissues of CD4⁺ T cell RAG1KO recipient mice. Mean±SD of transcript levels of corneal and conjunctival tissues. Results represent the mean±SD expression of 2-3 animals per strain/time point per experiment in two independent experiments for a total of 5-6 mice. *, P<0.05, **, P<0.001, ***, P<0.0001.