The protective effect of 17beta-estradiol on experimental autoimmune encephalomyelitis is mediated through estrogen receptor-alpha

Abstract

Low-dose estrogen (E2) treatment significantly inhibits the clinical signs and histopathological lesions of experimental autoimmune encephalomyelitis (EAE), and is being used in clinical trials to treat multiple sclerosis. To assess the role of intracytoplasmic estrogen receptors in mediating suppression of EAE, we studied mice with disrupted estrogen receptor-alpha (Esr1) and -beta (Esr2) genes. We demonstrate that the protective effect of E2 is abrogated in B6.129-Esr1(tm1Unc) mice (Esr1-/-) but not in B6.129-Esr2(tm1Unc) mice (Esr2-/-). The loss of E2-mediated protection from EAE in Esr1-/- mice immunized with the encephalitogenic MOG-35-55 peptide was manifested phenotypically by the development of severe acute clinical signs and histopathological lesions even in the presence of moderately high serum E2 levels. This is in contrast to C57BL/6 wild-type (WT) mice and Esr2-/- mice in which E2 treatment resulted in comparable serum levels and markedly suppressed clinical signs of EAE and abolished inflammatory lesions in the CNS. This pattern showing a lack of E2-dependent inhibition of EAE in Esr1-/- mice was mirrored by an enhanced rather than a reduced secretion of TNF-alpha, IFN-gamma, and interleukin (IL)-6 in MOG-specific splenocytes and a lack of inhibition of message for inflammatory cytokines, chemokines and chemokine receptors in CNS tissue. These results indicate that the immunomodulatory effects of E2 in EAE are dependent on Esr1 and not Esr2 signaling.

Figure 1.

Wild-type C57BL/6, Esr1−/−, and Esr2−/− female mice were implanted with 2.5 mg 17β-estradiol (E2) or placebo pellets, respectively, 1 week before the induction of EAE. E2 treatment suppressed actively induced EAE in C57BL/6 and Esr2−/− mice, while in Esr1−/− mice this protective effect was lost. The data show mean clinical EAE scores for each day for C57BL/6 female mice treated with saline or E2 pellets, B6.129-Esr1−/− females treated with saline or E2 pellets, and B6.129-Esr2−/− females treated with saline or E2 pellets.

Figure 2.

Histopathological examination of spinal cord sections from C57BL/6, Esr1−/−, and Esr2−/− mice reveals the lack of protection against EAE in Esr1−/− mice treated with 2.5-mg E2 pellet. A: C57BL/6 mice treated with placebo pellet with EAE (magnification, ×50). B: C57BL/6 mice treated with 2.5-mg E2 pellet, fully protected from EAE. No visible signs of inflammation were present within spinal cord tissue. C: B6.129-Esr1−/− mice treated with placebo pellet with EAE. D: B6.129-Esr1−/− mice treated with 2.5-mg E2 pellet with EAE. Several foci of inflammation are clearly evident within the spinal cord tissue. E: B6.129-Esr2−/− mice treated with placebo pellet with EAE. F: B6.129-Esr2−/− mice treated with 2.5-mg E2 pellet, fully protected from EAE. No visible signs of inflammation were present within spinal cord tissue. Histopathological sections for placebo- and E2-treated Esr2−/− mice were essentially the same as in C57BL/6 mice.

Figure 3.

Cytokine expression in spleens of placebo-treated and E2-treated C57BL/6, Esr1−/−, and Esr2−/− mice. Spleens were recovered from immunized mice at the peak of disease (15 days post-immunization). A total of 4 × 10⁶ cells were stimulated with 10 μg MOG-35–55 peptide for 72 hours before removal of the supernatants for assessment of cytokine secretion by CBA for TNF-α, IFN-γ, IL-2, IL-4, and IL-5, or by ELISA for IL-6.

Figure 4.

RPA analysis of (A) cytokine, (B) chemokine, and (C) chemokine receptor mRNA in spinal cords of protected (E2-treated) and non-protected C57BL/6, Esr1−/−, and Esr2−/− mice, respectively. Expression of Ltn, TNF-α, IFN-γ, RANTES, MIP-1α, MIP-2, IP-10, and MCP-1, and CCR1, CCR2, and CCR5 was strongly upregulated in spinal cords of placebo-treated C57BL/6, Esr1−/−, and Esr2−/− mice, as well as E2-treated Esr1−/− mice at the peak of EAE. In contrast, E2 treatment greatly reduced expression of all chemokines and chemokine receptors in WT and Esr2−/− mice protected from EAE. Data presented are from one representative experiment of two or three total RPA evaluations that showed similar results. P ≤ 0.05, ^** P ≤ 0.001, ^*** P ≤ 0.0001 as compared to WT mice treated with the saline pellet.