Attenuation of Th1 effector cell responses and susceptibility to experimental allergic encephalomyelitis in histamine H2 receptor knockout mice is due to dysregulation of cytokine production by antigen-presenting cells

Abstract

Histamine, a biogenic amine with both neurotransmitter and vasoactive properties, is well recognized as an immunomodulatory agent in allergic and inflammatory reactions. It also plays a regulatory role in the development of antigen-specific immune responses. CD4+ T-cells from histamine H1 receptor (H1R)-deficient (H1RKO) mice produce significantly less interferon-gamma and more interleukin (IL)-4 in in vitro recall assays compared to wild-type controls. H1RKO mice are also less susceptible to acute early-phase experimental allergic encephalomyelitis indicating that H1R signaling in CD4+ T cells plays a central role in regulating pathogenic T-cell responses. In this study, we show that mice lacking histamine H2 receptor (H2RKO) are similar to H1RKO mice in that they develop encephalitogen-specific T-cell responses as assessed by proliferation and IL-2 production and present with less severe acute early-phase experimental allergic encephalomyelitis. However, unlike T cells from H1RKO mice, which exhibit a strong Th2 bias, T cells from H2RKO mice do not. Rather, they are uniquely characterized by a significant inhibition of Th1 effector cell responses. Given that both histamine and adjuvants such as pertussis toxin modulate antigen-presenting cell (APC) maturation and function, including T-cell-polarizing activity, we analyzed the cytokines/chemokines secreted by APCs from wild-type, H1RKO, and H2RKO mice. Significant differences in cytokine/chemokine production by APCs from unimmunized and immunized mice were delineated. APCs from H2RKO mice produce significantly less IL-12 and IL-6 and markedly greater amounts of MCP-1 compared to wild-type and H1RKO mice. Because MCP-1 is known to inhibit IL-12 production, the failure of H2RKO mice to generate encephalitogenic Th1 effector cell responses is consistent with inhibition of negative regulation of MCP-1 secretion by H2R signaling in APCs.

Figure 1

A: Mean clinical disease course for C57BL/6J (n = 15), H2RKO (n = 15), and (H1RKO × H2RKO) F₁ hybrid (n = 15) mice immunized with MOG_35-.55. The data represent the combined results of three independent experiments (n = 5). The Wilcoxon-Gehan test for time to event and the Friedman test for correlated observations indicated that both the onset (P < 0.01) and severity of clinical signs (P < 0.0001) seen in the H2RKO animals is significantly different from C57BL/6J mice. Data are plotted as the mean clinical score ± SD. B: Proliferative response of T lymphocytes from C57BL/6J and H2RKO mice to MOG_35-55 ex vivo. Mean cpm ± SD were calculated from triplicate wells.

Figure 2

IL-4 and IL-5 (A) and IFN-γ, TNF-α, and IL-2 (B) production by MOG_35-55 stimulated splenocytes from MOG_35-55-immunized C57BL/6J (□) and H2RKO (▪) mice. Cytokine production was determined by either ELISA or cytometric bead assay. Significance of differences between wild-type and H2RKO mice were determined using the Student’s t-test (*, P < 0.05).

Figure 3

Thymic (A) and splenic (B) T-cell composition of C57BL/6J, H1RKO, and H2RKO mice (n = 3).

Figure 4

Anti-MOG_35-55 IgG1 (A) and IgG2a (B) antibody response in C57BL/6J (□) and H2RKO (▪) mice immunized with MOG_35-55. Differences between strains at the respective dilutions were determined using the Student’s t-test (*, P < 0.05).

Figure 5

CNS lesions in C57BL/6J and H2RKO mice with EAE elicited by immunization with MOG_35-55. Brain sections from H2RKO (A) and C57BL/6J mice (B) and spinal cord sections from H2RKO (C) and C57BL/6J mice (D) have principally a mononuclear inflammatory response associated with demyelination. In addition, the H2RKO mice (A, C) have numerous eosinophils (arrows) in the inflammatory response when compared to C57BL/6J mice (B, D). Brain and spinal cord sections from complete Freund’s adjuvant-immunized H2RKO control mice (E, F) and complete Freund’s adjuvant-immunized C57BL/6J control mice (G, H) have rare mononuclear inflammatory cell infiltrates perivascularly (E, arrow) and in the meninges and submeningeal white matter (H, arrow). H&E stains. Scale bar, 50 μm (A–H).

Figure 6

Cytokine production by PE macrophages from unimmunized (A) and MOG_35-55-immunized (B) C57BL/6J, H1RKO, and H2RKO mice. Assays were performed using pooled cells from two to three mice per experiment. The results presented represent the average of four independent experiments and the data are expressed as the normalized average relative abundance ± SEM. Analysis of variance was used to evaluate the significance of differences in cytokine levels among the three strains followed by a Tukey’s multiple comparison test to determine the significance of differences between strains when warranted. For all analyses, significance was determined at P ≤ 0.05. Unimmunized: G-CSF (F = 1.80, P = 0.18); GM-CSF (F = 1.84, P = 0.18); IL-6 (F = 3.43, P = 0.05); IL-10 (F = 5.14, P = 0.02); IL-12 (F = 2.48, P = 0.11); IL-12p70 (F = 16.88, P < 0.001); IFN-γ (F = 6.71, P = 0.006); MCP-1 (F = 14.39, P < 0.001); MCP-5 (F = 83.6, P < 0.001); RANTES (F = 3.88, P = 0.03); SCF (F = 11.06, P = 0.001); sTNFR1 (F = 30.6, P < 0.001); TNF-α (F = 2.84, P = 0.08); TPO (F = 0.74, P = 0.49); VEGF (F = 0.82, P = 0.46). Immunized: G-CSF (F = 6.50, P = 0.006); GM-CSF (F = 4.31, P = 0.03); IL-6 (F = 8.81, P = 0.002); IL-10 (F = 3.52, P = 0.05); IL-12 (F = 4.64, P = 0.02); IL-12p70 (F = 0.05, P = 0.95); IFN-γ (F = 1.52, P = 0.24); MCP-1 (F = 8.17, P = 0.002); MCP-5 (F = 11.69, P < 0.001); RANTES (F = 3.12, P = 0.07); SCF (F = 6.45, P = 0.006); sTNFR1 (F = 7.26, P = 0.004); TNF-α (F = 4.05, P = 0.03); TPO (F = 5.39, P = 0.01); VEGF (F = 17.45, P < 0.001).

Figure 7

MCP-1 expression by PE macrophages from MOG_35-55-immunized C57BL/6J, H1RKO, and H2RKO mice. MCP-1 expression was assessed by TaqMan PCR in 3-hour adherent PE macrophages collected at 7 days after immunization by peritoneal lavage 3 days after the intraperitoneal injection of 1.5 ml of sterile thioglycolate (3%). Expression levels were normalized to H2RKO levels and expressed as a mean ± SD (n = 3). Analysis of variance was used to assess the significance of differences in MCP-1 levels followed by a Tukey’s multiple comparison test to determine the significance of differences between strains. For both tests, significance was determined at P ≤ 0.05.