Estrogen modulation of spinal microglia: a morphological analysis

Acute Intermittent Hypoxia (AIH) is a powerful therapeutic tool for promoting respiratory neuroplasticity including phrenic long-term facilitation (pLTF). Mechanistically, pLTF develops in the phrenic motor nucleus, located in the ventral cervical spinal cord (SC; C3-C5) and is subject to endogenous and exogenous influences such as gonadal sex hormones and inflammation. Loss of circulating gonadal hormones (testosterone in male rats; 17β-estradiol (E2) in female rats) undermines the efficacy of AIH and results in a loss of pLTF. Similarly, elevated systemic and central inflammation, regardless of source, results in loss of pLTF. Microglia, the primary innate immune cells of the central nervous system, adapt to inflammatory stimuli by changing their morphology and releasing local cytokines. Preliminary data from our lab indicates that spinal microglial density and spinal inflammatory cytokines increase 2 weeks post-ovariectomy (OVX) in female rats. Given the link between OVX-induced microglial changes and loss of pLTF post-OVX, we explored population and morphological changes in the microglia of female rats following loss of E2 via OVX. We hypothesized that spinal cord microglia from female rats would exhibit significant changes in morphology and population following OVX compared to intact female rats during proestrus (high circulating E2). Given that female rats do not exhibit pLTF during estrus, a period of low circulating E2, we further hypothesized that glial changes would be measurable during estrus compared to proestrus. Female rats (N = 24) were divided into OVX, OVX-sham, proestrus, and estrus groups (N=6 each). Rats were euthanized via isoflurane overdose 1-week post-OVX or after at least two full estrous cycles for intact rats. C3-C5 SC tissue was fixed in paraformaldehyde (PFA) and segmented into 40-micron (μm) slices. Tissue was stained with 4′,6-diamidino-2-phenylindole (DAPI) to identify nuclear structures, Choline Acetyltransferase (ChAT) for motor neurons, and ionized calcium-binding adaptor molecule 1 (IBA1) for microglia. Images were processed via NIS-Elements (Nikon Instruments; v 6), and structural analysis was completed via Imaris (Oxoford Instruments; v 11.0). Preliminary findings indicate IBA1 integrated density, microglia quantity and IBA1 density/microglia were increased in the ventral cervical SC post-OVX compared to shams (P< 0.0001, P=0.014, P< 0.0001, respectively; one-way ANOVA with Tukey post-hoc). Preliminary observation of microglia morphology suggest changes in soma size and branch characteristics between groups. In conclusion, our data support the hypothesis that reduced circulating E2 leads to localized changes in spinal microglia consistent with an elevated inflammatory state. Funding sources: NIH F31-HL175931, NIH R01-HL146477, and the UMN Rehabilitation Science Graduate Program. This abstract was presented at the American Physiology Summit 2026 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.