Elsevier

Neuroscience

Volume 384, 1 August 2018, Pages 300-313
Neuroscience

Research Article
Impaired Organization of GABAergic Neurons Following Prenatal Hypoxia

https://doi.org/10.1016/j.neuroscience.2018.05.021Get rights and content

Highlights

  • Prenatal hypoxia induced interneuron cell loss in the cortical plate as early as 4 days after the hypoxic event.

  • Interneurons that survive prenatal hypoxia fail to populate the proper layer in the developing cortex.

  • Origin and location on the anterior–posterior and medio-lateral axes affect GABAergic cell sensitivity to hypoxia.

  • Hypoxia induced changes in the intracellular distribution of Hif1a in neurons of the cortical plate.

Abstract

Several conditions related to the intrauterine environment are associated with neuropsychiatric conditions in later life. In humans, approximately 2% of infants are exposed to perinatal hypoxia–ischemia or prolonged anoxic insult, a condition to which very low birth weight preterm infants exhibit the highest susceptibility. Analyses of postmortem tissue link some presentations of these conditions to changes in GABA pathway functionality in the brains of affected subjects. Using animal models of early-life hypoxia–ischemia, losses of particular interneuron populations were reported. We hypothesize that the origin of GABAergic cell loss is in the mispositioning of neurons during the formation of the cerebral cortex. Here we report that in C57 black mice exposed to hypoxic conditions (9% O2; 3% CO2), 22–26% of cell loss was detected in the cortical plate as early as four days after the hypoxic event. Moreover, the surviving cells failed to populate the proper layers in the developing cortex. Differential sensitivities were observed in neurons that originated from different germinal zones. A significant effect of GABAergic cell location along the anterior–posterior and medio-lateral axes on neuron sensitivity to hypoxia was detected. Finally, changes in guidance molecules in the developing cortex, including increases in hypoxia-inducible factor 1-alpha levels and intracellular distribution, decreases in reelin levels in the cortical plate and altered organization of radial glia, were observed. These changes in the molecular landscape of the immediate environment of the immature neurons may contribute to the observed outcomes in neuronal migration to, and establishment in, the correct cortical layers. We suggest that the interneuron loss may be related to these early events.

Introduction

Several conditions related to the intrauterine environment have been proposed as risk factors for neurologic outcome, including pre- and perinatal hypoxia (Jones et al., 1998, Zornberg et al., 2000, Dalman et al., 2001, Cannon et al., 2002, Boog, 2004). These epidemiological findings are supported by a large number of animal studies that model the effects of early-life hypoxia on behavioral, histological and molecular variables relevant to learning disabilities, motor impairment and schizophrenia (Meyer et al., 1998, Boksa, 2004). In humans, approximately 1–2% of infants are exposed to perinatal hypoxia–ischemia or prolonged anoxic insult, and very low birth weight preterm infants are particularly vulnerable (Volpe, 2003). About 25–50% of these babies are at risk for significant cognitive, behavioral, attention and socialization impairments (Larroque et al., 2008, Marlow et al., 2005, Doyle and Anderson, 2010). The principal injury following hypoxia or hypoxia ischemia has been revealed through imaging methods, together with neurological and histological studies, to involve cerebral white and gray matter (Cannon et al., 2002, Boog, 2004, Volpe, 2003, Back et al., 2007, Jensen, 2006, Okereafor et al., 2008).

To increase our understanding of the mechanisms by which hypoxic insult affects the immature brain, animal models of pre- and neonatal hypoxia have been studied. Motor impairment and difficulties in learning and memory tasks were observed following prenatal hypoxia and hypoxia ischemia in mice and rats (McQuillen et al., 2003, Tashima et al., 2001, Zhuravin et al., 2004, Golan et al., 2004b, Golan et al., 2004a). In addition, deficits in tasks associated with schizophrenia, such as pre-pulse inhibition, were reported (Howell and Pillai, 2014). The search for the neuronal origin of the behavioral impairment revealed cell loss in various brain regions — among them, the cerebral cortex, hippocampus, striatum and cerebellum — in different animal models of prenatal and neonatal hypoxia (Golan and Huleihel, 2006, Van de Berg et al., 2002, Biran et al., 2010, Rees et al., 2011). Neuron populations that are particularly affected by hypoxic insult are the GABAergic interneurons. In response to perinatal anoxia, the density of GABA immunoreactive (IR) neurons in the cerebral cortex and hippocampus decreased seven days after insult, a condition that was maintained into adulthood (Dell'Anna et al., 1996). Reduced density of the sub-population of GABAergic neurons expressing the calcium-binding protein calbindin (CB) in the cortex was observed 20 days after hypoxic insult on embryonic day 17 (E17) (Louzoun-Kaplan et al., 2008). Similarly, hypoxic insult on E18 led to decreased densities both of cells expressing the GABA-synthesizing enzyme, GAD65, and of parvalbumin (PV) cells 12 days after insult (Mazur et al., 2010). Long-term decreases in the densities of PV and CB neurons in rat striatum were reported two months after perinatal asphyxia (Van de Berg et al., 2003) and in the cerebral cortex following prenatal hypoxia (Gerstein et al., 2005). Altered maturation of PV neurons following neonatal hypoxia ischemia has also been reported (Failor et al., 2010). In human infants, significant declines in GAD67-IR neurons in the sub-plate and white matter were observed following perinatal injury (Robinson et al., 2006). Functional analysis of inhibitory synaptic activity in newborn rats following hypoxia showed a transient effect on postsynaptic parameters and a persistent decrease in GABAergic neuron firing, which may reflect either changes in the internal properties of the neurons or a reduction in the population of interneurons (Sanchez et al., 2007). Collectively, these studies suggest that early-life hypoxia has long-lasting effects on the cortical interneuron population. In the search for the origin of the above-mentioned neuron loss, molecular pathways involved in angiogenesis and neuron migration in the developing hippocampus and cortical plate were studied. Enhancement of VEGF (24 h) and decreases in BDNF (24 h) and APP (2 h and 24 h), were reported (Golan et al., 2004b, Howell and Pillai, 2014, Golan et al., 2009). Cleaved reelin levels were sensitive to prenatal hypoxia in a region-specific manner, such that after hypoxia, they were elevated in the cortex (1 and 5 days) and reduced in the hippocampus (2 h and 24 h) (Golan et al., 2009). In the latter brain region, the suppression of their levels was shown to be long lasting (Howell and Pillai, 2014). In addition to the changes in these molecular cues, down regulation of the associated signaling molecules, BACE and Dab1, was also observed following prenatal hypoxia (Golan et al., 2009).

The impact of prenatal insult on cortical GABAergic interneurons, which originate in the ganglionic eminences, may vary depending on the stage of neurogenesis (Super et al., 1998, Soriano and Del Rio, 2005). From the ganglionic eminences, these cells migrate via a long, tangential course into the developing cortex where they complete their final differentiation to generate the various classes of inhibitory interneurons (Xu et al., 2004, Cobos et al., 2005). There is substantial evidence that the specific germinal origin controls the final fate of these cells and that broad subtypes of interneurons originate from each of the three divisions of the ganglionic eminences — the lateral, medial and caudal ganglionic eminences (LGE, MGE and CGE, respectively) — and that they follow somewhat distinct migratory routes to the developing cortex (Yozu et al., 2005, Nery et al., 2002, Wonders and Anderson, 2005, Miyoshi et al., 2010). In addition, cells originating in the embryonic preoptic area (POA) contribute an additional population of cortical interneurons with distinct neurochemical and morphological fates (Metin and Pedraza, 2014, Gelman et al., 2009, Gelman and Marin, 2010).

The tangential migration of interneurons from the MGE to the cortex forms two streams, a superficial stream in the marginal zone (MZ) and a deep stream in the subventricular zone (SVZ) and intermediate zone (IZ), and both streams are dynamically organized during embryonic and early postnatal life. In mice at E13.5, migrating cells course through the SVZ of the LGE and along the LGE/cortex border, at which point the two prominent cortical migratory streams can be readily distinguished. From E13.5 until E17.5, the MZ stream accumulates MGE-derived neurons, and from E17.5 to postnatal day 2 (P2), the number of these young neurons in the MZ stream decreases as they migrate radially to populate the cortex in a laminar-specific manner. In the SVZ stream, the number of MGE-derived interneurons increases from E13.5 to E15.5 and then decreases thereafter. These dynamic changes in the distributions of MGE-derived neurons — as shown by different labeling methods (Tanaka et al., 2003, Yokota et al., 2007, Li et al., 2008, Ang et al., 2003) — provide an anatomical and temporal framework for assessing the effects of prenatal hypoxia during this period of interneuron development. In humans, the process of interneuron migration to and within the cortex takes place from mid-gestation to the age of two years (Xu et al., 2011). Cortical GABAergic interneuron migration during this time is also the peak window of vulnerability for hypoxia ischemia in term and preterm infants (Xu et al., 2011).

The overall aim of the current study was to search for a link between prenatal hypoxia and a defect in GABAergic interneuron migration, which may increase the risk for neurologic and psychiatric conditions such as schizophrenia and autism. We used an animal model of hypoxia, in which mice that undergo 2 h of prenatal hypoxia (9% O2, 3% CO2) present a clear defect in cortical GABAergic neuron numbers and distributions (Gerstein et al., 2005), a recapitulation of the major phenotype in human subjects. We hypothesize that this defect is due to the failure of the interneurons to populate their correct positions in the cortex. To assess this possibility and to understand the developmental mechanism that excludes a portion of GABAergic interneurons from the developing cortex, we characterized the spatial and temporal patterns of the cortical distributions of inhibitory neurons induced by hypoxic insult.

Section snippets

Mice and hypoxia protocol

C57 black mice were ordered from Harlan, Israel. The mice were maintained in the animal facility of Ben-Gurion University of the Negev on a 12:12 h light/dark schedule with food and water provided ad libitum. Lhx6-GFP mice were obtained from the Gensat project and used as described in Li et al. (2008). All experiments were carried out in accordance with National Institutes of Health guidelines for the care and use of laboratory animals (NIH Publications No. 80-23, revised 1996) and with the

Effect of hypoxia on GABAergic cell numbers in the developing cortex

We began by investigating when interneuron loss was first evident in the developing cortex using two interneuron markers, GAD 65/67 (all interneurons) and Lhx6-GFP (MGE interneurons), in different sets of experiments.

Four days after hypoxic insult on E14.5, a significant drop in the total number of GAD+ cells in the developing cortex (22–26%) was first observed in lateral regions (F = 9.5; df 1,82; p < 0.001), particularly in the rostral sections (Fig. 1). Birth dating with BrdU indicated that

Discussion

The data collected during this study suggest that among the GABAergic cells, those that are the most vulnerable to maternal hypoxia comprise a population of interneurons that, under normoxic conditions, is already present in the cortical plate at E18.5. Considering that for the majority of the cortical neurons, peak proliferation took place around E12.5, our data suggest that interneurons that exited the ganglionic eminences before the hypoxic episode (before E14.5) were susceptible to the

Acknowledgments

This study was supported by a grant to H. Golan from the Israel Science Foundation (ISF) (grant number 309/13). The monoclonal antibody rat-401, deposited to the DSHB by Hockfield, S. from the Yale University School of Medicine (DSHB Hybridoma Product rat-401), was obtained from the Developmental Studies Hybridoma Bank, created by the NICHD of the NIH and maintained at The University of Iowa, Department of Biology, Iowa City, IA 52242.

Competing interests

The authors declare no competing financial interests.

References (67)

  • B. Larroque et al.

    Neurodevelopmental disabilities and special care of 5-year-old children born before 33 weeks of gestation (the EPIPAGE study): a longitudinal cohort study

    Lancet

    (2008)
  • V. Louzoun-Kaplan et al.

    Prenatal hypoxia down regulates the GABA pathway in newborn mice cerebral cortex; partial protection by MgSO4

    Int J Dev Neurosci

    (2008)
  • A. Madan et al.

    Developmental stage-specific expression of the alpha and beta subunits of the HIF-1 protein in the mouse and human fetus

    Mol Genet Metab

    (2002)
  • S. Rees et al.

    The biological basis of injury and neuroprotection in the fetal and neonatal brain

    Int J Dev Neurosci

    (2011)
  • E. Soriano et al.

    The cells of cajal-retzius: still a mystery one century after

    Neuron

    (2005)
  • H. Super et al.

    The functions of the preplate in development and evolution of the neocortex and hippocampus

    Brain Res Brain Res Rev

    (1998)
  • W.D. Van de Berg et al.

    Perinatal asphyxia induced neuronal loss by apoptosis in the neonatal rat striatum: a combined TUNEL and stereological study

    Exp Neurol

    (2002)
  • W.D. Van de Berg et al.

    Impact of perinatal asphyxia on the GABAergic and locomotor system

    Neuroscience

    (2003)
  • T. Acker et al.

    Cellular oxygen sensing need in CNS function: physiological and pathological implications

    J Exp Biol

    (2004)
  • E.S. Ang et al.

    Four-dimensional migratory coordinates of GABAergic interneurons in the developing mouse cortex

    J Neurosci

    (2003)
  • S.A. Back et al.

    Maturation-dependent vulnerability of perinatal white matter in premature birth

    Stroke

    (2007)
  • V. Biran et al.

    Cerebellar abnormalities following hypoxia alone compared to hypoxic-ischemic forebrain injury in the developing rat brain

    Neurobiol Dis

    (2010)
  • T.D. Cannon et al.

    Fetal hypoxia and structural brain abnormalities in schizophrenic patients, their siblings, and controls

    Arch Gen Psychiatry

    (2002)
  • D.J. Ceradini et al.

    Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1

    Nat Med

    (2004)
  • I. Cobos et al.

    Mice lacking Dlx1 show subtype-specific loss of interneurons, reduced inhibition and epilepsy

    Nat Neurosci

    (2005)
  • A.A. Curley et al.

    Cortical basket cell dysfunction in schizophrenia

    J Physiol

    (2012)
  • C. Dalman et al.

    Signs of asphyxia at birth and risk of schizophrenia. Population-based case-control study

    Br J Psychiatry

    (2001)
  • V.L. Dengler et al.

    Transcriptional regulation by hypoxia inducible factors

    Crit Rev Biochem Mol Biol

    (2014)
  • L.W. Doyle et al.

    Adult outcome of extremely preterm infants

    Pediatrics

    (2010)
  • R.D. D'Souza et al.

    A laminar organization for selective cortico-cortical communication

    Front Neuroanat

    (2017)
  • S. Failor et al.

    Neonatal cerebral hypoxia-ischemia impairs plasticity in rat visual cortex

    J Neurosci

    (2010)
  • D.M. Gelman et al.

    Generation of interneuron diversity in the mouse cerebral cortex

    Eur J Neurosci

    (2010)
  • D.M. Gelman et al.

    The embryonic preoptic area is a novel source of cortical GABAergic interneurons

    J Neurosci

    (2009)
  • Cited by (15)

    • Chronic Oxytocin treatment has long lasting therapeutic potential in a rat model of neonatal hypercapnic-hypoxia injury, through enhanced GABAergic signaling and by reducing hippocampal gliosis with its anti-inflammatory feature

      2021, Peptides
      Citation Excerpt :

      Studies that have investigated the influence of hypoxia on different types of neurons have revealed that GABAergic interneurons are particularly sensitive to hypoxia. It has been established that perinatal anoxia causes a decrease in the density of GABA immunoreactive neurons in the cerebral cortex and hippocampus within the first week following hypoxia [31]. These neurons never recover after this initial damage; thus, the loss is permanent.

    • Animal models for neonatal brain injury induced by hypoxic ischemic conditions in rodents

      2020, Experimental Neurology
      Citation Excerpt :

      Much research using this model has probed the effects of maternal hypoxia in conjunction with subsequent HI insult. These studies find that maternal hypoxia increases the vulnerability of neonates to subsequent HI insults (Van de Berg et al., 2002); some highlight specific epigenetic modulation that may explain the increased sensitivity (Ma and Zhang, 2015; Nisimov et al., 2018; Zhang et al., 2018). However, maternal hypoxia alone is not benign.

    • Brain oxidative damage in murine models of neonatal hypoxia/ischemia and reoxygenation

      2019, Free Radical Biology and Medicine
      Citation Excerpt :

      These events appeared to be associated with changes in: (a) guidance molecules in the developing cortex; (b) increased levels of the Hypoxia-Inducible Factor 1-alpha (Hif1α) as well as changes in its subcellular localization; (c) reduced amounts of Reelin; (d) altered organization of radial glia. Moreover, the cells in the rostral regions of P2 brains not only positioned abnormally, but also located at different distances from the blood capillaries, so that the distances from cell to capillary observed in the hypoxic group were larger than those found in the normoxic brain sections [159]. In an additional maternal hypoxic model designed in mice at gestational day 17, it was demonstrated that the impairment of cell migration in the hippocampus was mediated by the deterioration of Reelin/Dab1/APP signaling [160].

    View all citing articles on Scopus
    View full text