Perinatal inflammation results in decreased oligodendrocyte numbers in adulthood

Abstract

Aims: Maternal inflammation is a risk factor for preterm birth, and premature infants are often exposed to supplemental oxygen as a life-sustaining therapy. While more immature neonates are surviving, rates of neurodevelopmental impairment are not improving. We developed a novel mouse model with clinically relevant exposures to test the hypothesis that systemic maternal inflammation with transient neonatal hyperoxia exposure will induce a phenotype similar to diffuse periventricular leukomalacia (PVL) like that observed in premature human infants.

Main methods: Timed-pregnant C3H/HeN mice received intraperitoneal injections of lipopolysaccharide (LPS) or saline on embryonic day 16. Newborn pups were placed in room air (RA) or 85% oxygen (O2) for 14 days, followed by 14 days in RA recovery. Oligodendroglial and microglial populations were evaluated at 14 and 28 days.

Key findings: Brain weight to body weight ratios were lower in mice exposed to LPS. Oligodendrocyte numbers were decreased significantly in the cerebral cortex and hippocampus in groups exposed to LPS or LPS/O2 at 14 days, and persisted in the cerebral cortex at 28 days for LPS/O2 mice. At day 14, cleaved caspase 3 was increased and numbers of microglia were elevated in the cerebral cortex and hippocampus of LPS/O2 animals.

Significance: These data indicate that combining systemic maternal LPS and neonatal hyperoxic exposure impairs myelination, and suggests that this novel mouse model may represent a subtle, diffuse form of periventricular white matter injury that could provide a clinically relevant platform for further study of perinatal brain injury.

Keywords: Development; Hyperoxia; Inflammation; Lipopolysaccharide; Microglia; Oligodendrocyte; Periventricular leukomalacia; Premature brain injury; White matter injury.

Figure 1. Oligodendrocyte numbers are decreased in cerebral cortex by treatment with LPS and/or O₂

A and B, representative images of saline/RA, saline/O₂, LPS/RA and LPS/O₂ treament groups, labeled with CNPase as a marker for early myelinating oligodendrocytes, are shown (100×). C and D, labeled cortical oligodendrocyte cell bodies were counted per high-power field in the cerebral cortex between the cingulum and rhinal fissure. Significant effect of LPS and an effect of O₂ were indicated by two-way ANOVA at both day 14 and day 28, p<0.05, n=2–4 litters, n=1–2 pups per treatment group. LSD post hoc analyses indicated: * different than saline/RA; ^#different than LPS/RA; ^different than saline/O₂. Arrowheads provide orientation toward the white matter at the base of the cortex. Small arrows identify representative CNPase-positive cell bodies.

Figure 2. Oligodendrocyte numbers are decreased in dentate gyrus of hippocampus by treatment with LPS and/or O₂

A and B, representative images of saline/RA, saline/O₂, LPS/RA and LPS/O₂ treatment groups, labeled with CNPase as a marker for early myelinating oligodendrocytes, are shown (100×). C and D, labeled hippocampal oligodendrocyte cell bodies were counted per high-power field in the dentate gyrus. Significant effect of LPS and an effect of O₂ were indicated by two-way ANOVA at day 14, p<0.05, n=2–4 litters, n=1–2 pups per treatment group. LSD post hoc analyses indicated: * different than saline/RA; ^#different than LPS/RA; ^different than saline/O₂. No statistical differences were indicated at day 28.

Figure 3. Cleaved caspase 3 protein levels were elevated in the brains of animals exposed to LPS, with or without O₂

Proteins from whole brain homogenates obtained from saline/RA, saline/O₂, LPS/RA and LPS/O₂ treatment groups at day 14 were analyzed for cleaved caspase 3 by western blot. An effect of O₂ and an interaction between O₂ and LPS were indicated at both day 14 and 28 by two-way ANOVA, p<0.05, n=2–4 litters, n=1–2 pups per treatment group. LSD post hoc analyses indicated: *different than saline/RA; ^#different than LPS/RA; ^different than saline/O₂.

Figure 4. Immunohistochemical staining and quantification of cortical and hippocampal microglia

A, representative images of saline/RA, saline/O₂, LPS/RA and LPS/O₂, treatment groups of at day 14 are shown (100×). B, cortical and hippocampal microglial cell bodies were counted per high-power field in the the cingulum and rhinal fissure and the dentate gyrus, respectively. A significant effect of LPS was observed by two-way ANOVA, p<0.05, n=2–4 litters, n=3–4 pups per treatment group. LSD post hoc analyses indicated: *different than saline/RA.