Histology of skin lesions before and after VACV infection. (A) Hematoxylin and eosin–stained skin tissues are shown for normal and eczematous mice before and 7 d after virus infection. Bar, 1 mm. (B) CD4⁺, CD8⁺, and Mac-1⁺ cells were stained by immunohistochemistry and mast cells were stained with toluidine blue. Data represent means and SEM values of cell numbers per high-power field (HPF; n = 8 each group). (C) Immunohistochemical staining of NK cells with anti-Ly49G2 (clone 4D11) mAb. Bar, 0.1 mm. (D) Ly49G2 (4D11)⁺ NK cells were enumerated with six mice per cohort. Shown are results representative of six independent experiments. Data represent means and SEM. *, P < 0.05; **, P < 0.01; and ***, P < 0.001. NS, not significant.

Role of IL-17A in reduced NK cell cytotoxicity in eczematous mice. (A) mRNA expression of IL-17A and Th17-related cytokines was analyzed by real-time PCR (spleen) or semiquantitative RT-PCR analysis (skin and draining lymph node). Values were normalized against those of normal mice. Shown are results representative of two independent experiments (n = 4–6 mice). *, P < 0.05; and **, P < 0.01 by the Student's t test. (B) CD3⁺CD4⁺IL-17⁺ Th17 cells were enumerated in draining lymph nodes. (C) The onset of skin lesion development was delayed (left) and the size of primary skin lesions was smaller (right) in mice treated with anti–IL-17 mAb. **, P < 0.01 versus control. (D) NK cells in spleens and lesional skin were enumerated by flow cytometry and immunohistochemistry, respectively. (E) Splenic NK cells expressing granzyme B (GzmB), perforin (Pfn), or IFN-γ were analyzed by flow cytometry in mice treated with anti–IL-17 or control mAb day 2 after infection. (F) Virus titers were measured on day 7. Shown are representative results from three independent experiments. *, P < 0.05; and **, P < 0.01 versus control. (G) Mice were NK-depleted by αAGM1 injection 1 d before VACV infection. Anti–IL-17 antibody was also intraperitoneally injected 2 h after αAGM1 injection. After VACV infection, anti–IL-17 was injected on days 1 and 3, and αAGM1 was injected on days 2 and 5. Skin lesion development was observed and lesion size was measured for 6 d. The result is a representative of two independent experiments. *, P < 0.05; **, P < 0.01; and ***, P < 0.001 versus rat IgG2a–injected mice. #, P < 0.05 versus anti–IL-17 treated, NK-nondepleted mice. Data represent means and SEM values.

Induction of erosive primary skin lesions in VACV-infected eczematous mice. (A) Eczematous skin lesions were induced by repeated Der f/SEB (D/B) treatments, and mice with a clinical score of ≥8 were infected intradermally with VACV (eczematous group). A cohort (normal group) of mice with healthy skin was also infected at the same anatomical site. (B) Typical eczematous (right) and normal (left) mice are shown on day 6 after infection. (C) The size of erosive skin lesions. Shown is a representative of at least 15 experiments using 4–10 mice in each group. (D) Virus titers in the infected skin (n = 7 mice per group). Shown are results representative of four independent experiments. Data represent means and SEM values. *, P < 0.05; **, P < 0.01; and ***, P < 0.001 versus normal mice. ND, not detected.

Reduced NK cell activity was critical for the development of VACV-induced erosive skin lesions in eczematous mice. (A) NK cell cytotoxic activity of splenocytes on day 2 after infection was measured using YAC-1 cells as target cells at the indicated effector-to-target ratios. Shown are results representative of 5 independent experiments. Data represent means and SEM values. *, P < 0.05; and ***, P < 0.001 versus uninfected mice. ###, P < 0.001 versus eczematous mice. (B) Flow cytometric analysis of CD3⁻ NK1.1⁺ NK cells expressing granzyme B, perforin, or IFN-γ in spleens. Shown are results representative of 8 independent experiments. *, P < 0.05; **, P < 0.01; and ***, P < 0.001. (C) Virus titers in spleens from αAGM1- or NRS-treated mice. (D) Sizes of primary skin lesions in αAGM1- or NRS-treated mice. *, comparison between αAGM1- and NRS-treated mice; #, comparison between αAGM1-treated normal and eczematous mice; and §, comparison between NRS-treated normal and eczematous mice (one, two, and three symbols indicate P < 0.05, 0.01, and 0.001, respectively. (E) Virus infection in αAGM1-treated normal mice induced erythematous papules at the inoculation site (indicated by the arrow; top) and satellite lesions. Satellite lesions were also induced in αAGM1-treated eczematous mice (indicated by arrowheads). Satellite lesions in both normal/αAGM1 and eczematous/αAGM1 mice were confirmed to contain live virus. Shown in C–E are representative results of three independent experiments (n = 4–8 mice per group). (F) Adoptive transfer of cultured NK cells. Sizes of primary skin lesions are shown. The inset shows the purity of transferred NK cells as analyzed by flow cytometry. Shown are results representative of three independent experiments. Data in A and C–F represent means and SEM values; data in B represent means and SD values.

IL-17A but not IL-17F reduces NK cell cytotoxicity in vitro. Splenic NK cells were incubated with the indicated cytokines for 48 h before flow cytometric analysis of NK cells expressing granzyme B, perforin, or IFN-γ. Shown are results representative of two independent experiments. Data represent means and SEM values. **, P < 0.01; and ***, P < 0.001 versus IL-4 by one-way analysis of variance.