Deficits in Seizure Threshold and Other Behaviors in Adult Mice without Gross Neuroanatomic Injury after Late Gestation Transient Prenatal Hypoxia

Abstract

Intrauterine hypoxia is a common cause of brain injury in children resulting in a broad spectrum of long-term neurodevelopmental sequela, including life-long disabilities that can occur even in the absence of severe neuroanatomic damage. Postnatal hypoxia-ischemia rodent models are commonly used to understand the effects of ischemia and transient hypoxia on the developing brain. Postnatal models, however, have some limitations. First, they do not test the impact of placental pathologies on outcomes from hypoxia. Second, they primarily recapitulate severe injury because they provoke substantial cell death, which is not seen in children with mild hypoxic injury. Lastly, they do not model preterm hypoxic injury. Prenatal models of hypoxia in mice may allow us to address some of these limitations to expand our understanding of developmental brain injury. The published rodent models of prenatal hypoxia employ multiple days of hypoxic exposure or complicated surgical procedures, making these models challenging to perform consistently in mice. Furthermore, large animal models suggest that transient prenatal hypoxia without ischemia is sufficient to lead to significant functional impairment to the developing brain. However, these large animal studies are resource-intensive and not readily amenable to mechanistic molecular studies. Therefore, here we characterized the effect of late gestation (embryonic day 17.5) transient prenatal hypoxia (5% inspired oxygen) on long-term anatomical and neurodevelopmental outcomes in mice. Late gestation transient prenatal hypoxia increased hypoxia-inducible factor 1 alpha protein levels (a marker of hypoxic exposure) in the fetal brain. Hypoxia exposure predisposed animals to decreased weight at postnatal day 2, which normalized by day 8. However, hypoxia did not affect gestational age at birth, litter size at birth, or pup survival. No differences in fetal brain cell death or long-term gray or white matter changes resulted from hypoxia. Animals exposed to prenatal hypoxia did have several long-term functional consequences, including sex-dichotomous changes. Hypoxia exposure was associated with a decreased seizure threshold and abnormalities in hindlimb strength and repetitive behaviors in males and females. Males exposed to hypoxia had increased anxiety-related deficits, whereas females had deficits in social interaction. Neither sex developed any motor or visual learning deficits. This study demonstrates that late gestation transient prenatal hypoxia in mice is a simple, clinically relevant paradigm for studying putative environmental and genetic modulators of the long-term effects of hypoxia on the developing brain.

Keywords: Animal model; Behavior; Developing brain; Imaging; Prenatal hypoxia.

Fig. 1:. Comparison of human and mouse neuroanatomic and milestone development with a schematic of prenatal hypoxia model.

(A) Timeline of human and mouse brain development. The human timeline spans the weeks of gestation (W) and months after delivery (M). The mouse timeline includes embryonic age (E) and postnatal days of life (P). Specific developmental milestones are indicated. (B) Schematic and timing of late gestation transient prenatal hypoxia exposure and performed experiments.

Fig. 2:. Prenatal hypoxia paradigm induces a canonical HIF1α response consistent with hypoxic insult in the fetal brains.

(A) Representative immunoblot if HIF1α (~120 kDa) and actin (~40 kDa) in fetal brains after the indicated duration of prenatal hypoxia. (B) Quantification of HIF1α protein level as normalized to actin, the loading control. (C) Vegfa mRNA levels in the fetal brain in normoxia and after the indicated duration of hypoxia. The points shown in the graphs are from individual fetal brains. Statistics are described in the methods section.

Fig. 3:. Maternal dams make typical nests after hypoxia exposure, and while there is an early decrease in weight, there are no other differences in litter health after hypoxia.

(A) The cumulative distance over time of pregnant dams during the first 90 minutes the mice were in the chamber. The area under the curve is shaded. The dashed line around each line represents the SEM of each condition. The first dashed black line is where mice reach 10% inspired O2 at about 20 minutes in the chamber, and the second dashed line is when they are at the goal of 5% inspired oxygen at 30 minutes. The statistics compare normoxia and hypoxia by the area under the curve for the first 20 minutes of chamber exposure and then for 1 hour after hypoxic mice reached goal oxygen level. (B) Scoring of maternal nestlets the morning after normoxia or hypoxia. Points represent individual dams that were assessed. (C) Litter size at birth and (D) gestational age at birth for indicated individual litters were evaluated. (E) Percent survival of offspring. Error bars represent standard error at that time point. The log-rank test was used for statistics between normoxia (solid black line) and hypoxia (dashed blue line). (F-H) Weights of pups and adult offspring at the indicated age. Points correspond to the individual mice weighed. Repeated measured mixed modeling statistics are described in the methods section.

Fig. 4:. Prenatal hypoxia does not increase cell death in the fetal brain.

(A) Representative image of apoptotic nuclei (arrows) within cortex, hematoxylin & eosin stain. Scale bar represents 50 μM (B-D) Number of apoptotic nuclei in the cortex, basal ganglia, and white matter reported as average seen in 10 separate high-powered fields. Statistics are described in the methods section.

Fig. 5:. Prenatal hypoxia does not result in long-term gross neuroanatomical damage

(A) Axial images of adult female brain MRI (averaged b0 map) of indicated conditions from individual mice. (B) Quantification of ventricular size in the setting of normoxia and hypoxia. Displayed symbols in (B) represent hypoxic animals from (A). (C) Mid-sagittal image demonstrating the areas that were quantified, the anterior cingulate cortex (light blue line), the genu of the corpus callosum (orange circle), and the splenium of the corpus callosum (pink circle). (D) Quantification of cortical thickness at the anterior cingulate. (E & F) Quantification of FA in the mid-sagittal plane of the corpus callosum’s genu and splenium, respectively. Statistics as outlined in the methods. Points represent individual mice. (G) Representative image of axial section comparing FA to axial, radial, and mean diffusivity (AD, RD, and MD, respectively)

Fig. 6:. Prenatal hypoxia leads to functional deficits in adult mice.

(A) Flurothyl seizure threshold study demonstrating time to first GTC. (B & C) Data from grip strength. (B) Forelimbs and (C) hindlimbs force. (D) The number of marbles buried. (E) Change in nestlet weight in short-term nestlet test. Statistics completed as outlined in the methods. Points represent individual mice.

Fig. 7:. Sex-dichotomous behavior differences after prenatal hypoxia.

(A) Time spent in open arms in the elevated zero maze. (B) Total time animals spent in the center of the open field. (C-F) Percent time in each mouse spent in each chamber during stages of social interaction study. (C-D) Data from habituation stage from males and females. (E-F) Data from novel mouse/object stage from males and females. Statistics as outlined in the methods. Points represent individual mice.

Fig. 8:. Prenatal hypoxia does not lead to deficits in learning or memory.

(A) The speed at fall on Rotarod. (B-C) Data from the Morris Water Maze. (B) Latency time to platform in place trials. (C) One-hour probe trial. (D) Twenty-four-hour probe trial. Statistics were performed as detailed in methods. Each point represents a single animal.