Relation of Mild Hypoxia in Premature Infants to Hippocampal Neuronal/Synaptic Dysmaturation and Subsequent Learning and Memory Deficits: Mechanisms and Potential for Later Correction

June 2026

Introduction

Learning, memory, and related cognitive impairments are common in survivors of preterm birth.¹ Although overt brain injuries, especially white matter injury, are known mediators of such deficits,² learning disturbances occur in as many as 50% of premature infants without evidence for overt brain injury.^1,3 The possibility that a proportion of this unexplained disability is related to hypoxemic episodes in the neonatal period is suggested by a variety of clinical studies.^4-6 However, such episodes are accompanied generally by pronounced hypoxemia, often with bradycardia, suggesting a hypoxic-ischemic insult. Of interest in the context of the report that stimulated this Commentary (see later) is the possible relation of subsequent specific disability to relatively prolonged exposure to mild hypoxia (oxygen saturation of 87%) without bradycardia or other evidence of ischemia. Such more prolonged, mild exposures occur not uncommonly in extremely or very premature infants with severe respiratory distress syndrome as well as a variety of other respiratory, cardiac or systemic disorders. Notably, although some large-scale studies^7-10 suggest that relatively mild decreases in oxygen saturation of varying durations in premature infants are not overtly harmful, careful long-term follow-up with measures of hippocampal-mediated memory and learning functions (see later) are lacking.

A recent experimental study, the focus of this Commentary, provides strong support for the notion that prolonged mild hypoxia (O₂ saturation of 87%) can lead to neurocognitive disability, identifies the hippocampus as a principal anatomical site involved, delineates the neurobiological basis for the hippocampal disturbance, and raises the possibility of correction of the disturbance well after the neonatal period (see later).¹¹ Particular involvement of hippocampal maldevelopment in later learning disturbances in survivors of premature birth has been suggested previously by neuropsychological data and by findings of disturbances of hippocampal development, in the absence of apparent destructive lesions of hippocampus.^1,12-17

Mild Neonatal Hypoxia, Synaptic Maturation and Adult Hippocampal Learning and Memory Deficits

Because the substance of the report to be highlighted in this Commentary concerns the developing hippocampus and its connections, I will begin by reviewing briefly relevant background concerning the developing hippocampus and its connections, with an emphasis on the fundamental circuitry of this brain region and the vulnerability of this rapidly evolving circuitry in the neonatal period

Relevant Background Concerning the Hippocampus and its Connections

The particular focus on the hippocampus by Riddle et al.¹¹ relates in considerable part to the critical functions of this region and its connections.^16,17 The intrinsic circuitry of the hippocampus and its cortical and subcortical connections are essential for such functions as memory, emotional regulation, learning and cognition. Many of these key functions of the hippocampus are driven by long-term potentiation (LTP), which involves the persistent strengthening of synapses between neurons (see later).

Overall, the hippocampus receives key inputs from entorhinal cortex to the dentate gyrus, the site of the initial neurons involved in the critical hippocampal circuit involved in memory and related functions.^16,17 Memory is processed via connections from axons of granule cells of the dentate gyrus to pyramidal neurons of Ammon’s horn (cornu ammonis), first to the dendrites of CA3 pyramidal cells, which then send axons to the dendrites of C1 pyramidal cells, with connections ultimately to the subiculum, which sends axons to multiple cortical and subcortical regions. Among this pathway of connections, those between the pyramidal cells of the CA3 and C1 regions of the cornu ammonis are especially important.

Preclinical studies have shown the vulnerability of these regions to overt hypoxic-ischemic insults,^16,17 particularly involving dendritic development and arborization.¹⁸ However, the possibility that mild hypoxia leads to such defects is unknown.

Thus, the current report¹¹ addresses the cytological and molecular effects of mild postnatal hypoxia in an animal model of the extremely/very preterm infant. The findings are of major importance.

Mild Neonatal Hypoxia in a Mouse Model of the Extremely/Very Premature Infant Alters Hippocampal Synaptic Maturation and Disrupts Adult Hippocampal Learning and Memory

To address the potential role of mild hypoxia in the premature infant in the genesis of impaired memory and learning frequently observed subsequently (see earlier), Riddle and coworkers utilized a mouse model.¹¹ Although correlates are not perfect, available data indicate that the hippocampus in the model utilized, the P2 (postnatal day 2) mouse, is most comparable developmentally to the extremely/very preterm human infant (Riddle A, personal communication).¹⁶

Mild Neonatal Hypoxia. The hypoxic insult was produced by a 30-minute exposure to 8% inspired oxygen, which resulted promptly in an oxygen saturation of 87%.¹¹ This exposure had been shown previously to produce no neuronal degeneration.¹⁹ Notably, the exposure was well tolerated by the neonatal mice, without the occurrence of bradycardia or such EEG abnormalities as background suppression or epileptic phenomena.¹¹

Hippocampal Gene Expression Responses in Neonatal (P2) Animals Exposed to Mild Hypoxia. Sophisticated RNA transcriptomic analyses on hippocampi 24 hours after mild hypoxia in the neonatal animals showed that multiple gene clusters were altered, especially enriched for neuronal and synaptic targets.¹¹ Particularly upregulated were numerous voltage and ligand-gated channel gene clusters that support synaptic function (see later).¹¹ Notably, more severe hypoxia was needed to lead to upregulation of genes involved in cell injury pathways, inflammation and glial development.

Hippocampal Learning Deficits and Memory Impairment Occur in Adult Mice Exposed to Mild Neonatal Hypoxia. To address the key issue of whether hippocampal learning and memory deficits occur in adult mice after exposure to mild neonatal hypoxia at P2, the investigators studied animals at 6-8 weeks (mouse adult equivalent age). Pronounced deficits in both hippocampal and amygdala-complex dependent learning were observed in the animals previously exposed to mild hypoxia in the neonatal period.¹¹ Similarly, experimental paradigms to assess memory impairment demonstrated such deficits in the adult animals previously exposed to mild hypoxia.

Hippocampal CA1 Dendritic Complexity is Impaired in Adult Mice Previously Exposed to Mild Neonatal Hypoxia. To determine whether the functional deficits just described have a structural correlate, the investigators performed sophisticated morphological analyses in the adult mice. Morphometric disturbances were clearly identifiable and were most apparent in CA1 basal dendritic arbors in adult animals previously exposed to mild neonatal hypoxia. Thus, a key portion of the hippocampal circuitry described earlier is impaired after neonatal hypoxia.

Early or Delayed White Matter Injury, Cell Death, and/or Inflammation Do Not Accompany Neonatal Hypoxia. Because cerebral white matter injury, cell death and inflammation are observed commonly with neonatal insults, Riddle et al.¹¹ asked whether these events could be mediating or modulating the effects on hippocampal structure and function just described in the adult mice after neonatal hypoxia. Assessment of oligodendroglial progenitors, myelination, axonal integrity, and neuroinflammation (via microglial and astrocytic inflammation) showed no evidence of acute or delayed cell death, neuroinflammation or demyelination that could contribute to the hippocampal neuronal dysmaturation or altered hippocampal learning in the young adult mice previously exposed to mild neonatal hypoxia.¹¹

Hippocampal CA1 Learning and Memory Mechanisms in Adult Mice are Disrupted by Neonatal Hypoxia. To determine the physiological basis of the hippocampal-mediated learning deficits and memory impairment noted in adult mice exposed to mild neonatal hypoxia, the investigators studied the key hippocampal synapses involved.¹¹ The principal cellular correlates of hippocampal learning and memory are CA3 synapses onto CA1 apical dendrites in Ammon’s horn, as described earlier. These key hippocampal synapses undergo lasting activity-dependent strengthening or so-called long-term potentiation (LTP), the critical cellular correlate of hippocampal learning and memory. Adult animals previously exposed to mild neonatal hypoxia exhibited persistently altered LTP, a deficit expected to lead to altered hippocampal learning and memory.¹¹

Neonatal Hypoxia was Followed by Altered Synaptic Strength at CA3-CA1 Synapses in Adult Animals. The neurophysiological findings in adult animals previously exposed as neonates to mild hypoxia, as just discussed, suggest that CA3-CA1 synaptic strength was impaired. To address this issue, Riddle et al.¹¹ studied presynaptic CA3 axon potentials and postsynaptic CA1 responses under conditions of excitatory amino acid neurotransmission, as occurs in vivo. The findings were consistent with reduced CA3-CA1 synaptic strength, and suggested a disturbance in regulatory channel activity.

Synaptic SK2 Channel Function in CA1 Neurons is Abolished in Adult Mice After Mild Neonatal Hypoxia. The key regulatory channel shown to be disturbed in the CA3-CA1 synapses is the SK2 channel. The latter normally causes potassium to leave the cell and thereby reduce neuronal firing.^20,21 These channels thereby act to provide feedback control of excitatory amino acid receptors (e.g., NMDA, AMPA) on hippocampal learning, memory and LTP. They normally become operative later in adolescence and adulthood (in experimental models) when memory and learning become prominent.¹¹ In the mouse, SK2 increases after the first week of life (Riddle A, personal communication).

In the adult animals previously exposed to mild neonatal hypoxia, this developmental shift to high SK2 channel activity does not occur.¹¹ This finding raises the next question as to why high SK2 channel function activity fails to develop in the mild hypoxia-exposed animal.

Loss of Synaptic SK2 Function is Mediated by Increased CK2 Phosphorylation of Synaptic Calmodulin in the Animals Previously Exposed to Neonatal Hypoxia. A critical regulator of SK2 function is CK2, which forms a complex with the channel and thereby confers a sensitivity to Ca⁺⁺.¹¹ (CK2 is a protein kinase that phosphorylates a variety of substrates involved in gene expression.) The findings suggest that SK2 function is persistently lost in the hypoxia-exposed animals by increased CK2 phosphorylation of synaptic calmodulin. Thus, aberrant developmental activation of CK2 seems key to the loss of SK2 activity after hypoxia.

Blockade of CK2 Restores Synaptic SK2 Channel Activity. With these data indicative of the importance of CK2 and its negative impact on SK2 channel function and thereby on development of hippocampal-mediated neurological functions, a key question is whether blockade of CK2 could restore SK2 channel function in the adult animals previously exposed to mild neonatal hypoxia. Remarkably, a specific CK2 inhibitor (Apamin, a toxin from bee venom) was shown to restore synaptic SK2 activity in hippocampal slices from animals previously exposed to hypoxia.¹¹ Whether such an agent could be utilized in vivo remains to be shown (see later), but if the observation in slices could be duplicated in the intact animal, correction of the hippocampal functional deficits, both neurophysiological and clinical, long after the neonatal exposure might be possible.

Summary / Conclusions

In a mouse model of the extremely/very premature infant, Riddle et al. have shown that a 30-minute exposure to mild hypoxia (O₂ saturation 87%) in neonatal mice resulted in a diminished expression of hippocampal synaptic components in the neonatal period.¹¹ These effects were observed without evidence of hypoxic-ischemic brain injury or inflammation.

Subsequently, the animals previously exposed to mild neonatal hypoxia exhibited impaired hippocampal dendritic and synaptic development.¹¹ Particular involvement of hippocampal CA1 neurons was apparent.

The subsequent impairment of hippocampal CA1 neuronal dendritic and synaptic development was accompanied by hippocampal learning and memory deficits and its neurophysiological correlate, LTP, in the adolescent/adult animal.¹¹ Such learning and memory deficits are common in the long-term neurocognitive studies of extremely/very premature infants.^1,17,22 These deficits, not unexpectedly, are generally undetected in short-term follow-up studies with such instruments as Bayley Scales.

A key synaptic channel later involved in memory and learning in hippocampus, the SK2 potassium channel, was found to be reduced in adolescent/adult mice previously exposed to mild hypoxia. The activity of this potassium channel provides critical modulatory feedback to AMPA- and NMDA-mediated neurotransmission. The SK2 channel is not expressed appreciably until adolescence/adulthood in the mouse, but its neonatal levels (and subsequently) are impacted by the exposure to mild hypoxia. The negative impact is mediated by the aberrant activation of a negative regulator of SK2, i.e., CK2.

Remarkably, synaptic SK2 activity in hippocampal slices from adult animals previously exposed to hypoxia was restored by exposure to a specific CK2 inhibitor.¹¹ This observation raises the extraordinary possibility that impaired hippocampal learning and memory in older children after premature birth could be corrected by inhibitors of CK2. Notably, re: feasibility, such drugs are currently under active development in clinical trials for treatment of a wide variety of cancers, and in preclinical studies for several neurodegenerative disorders.^23,24

A final comment concerning “relevance” is necessary. Indeed, is a 30-minute exposure to 87% O₂ saturation in a P2 mouse relevant to the human infant? Even if one accepts the likely conclusion that hippocampal development at P2 in the mouse bears strong similarities to that in the extremely/very premature infant, is a 30-minute exposure to 87% O₂ saturation unusually long? Because the authors have not yet reported studies with shorter exposures to such mild hypoxia, this concern is difficult to address directly. It is notable that no apparent effect on cerebral perfusion accompanied the insult (no bradycardia or EEG disturbance). My view is that the data are of great importance in demonstrating (a) the especial vulnerabilities of specific developing hippocampal neurons and synapses, (b) the role of these neurons and synapses in learning and memory disturbances, observed later in development (as in survivors of extremely/very premature birth), (c) the molecular bases of these disturbances, and (d) the extraordinary possibility of reversing such deficits by pharmacologic intervention long beyond the neonatal period.

Joseph J. Volpe, MD

Department of Neurology, Boston Children’s Hospital

Bronson Crothers Professor of Neurology, Emeritus, Harvard Medical School

Boston MA

References

1. Inder TE, Volpe JJ, Anderson PJ. Defining the Neurologic Consequences of Preterm Birth. N Engl J Med 389:441-53, 2023. DOI: 10.1056/NEJMra2303347
2. Volpe JJ. Dysmaturation of Premature Brain: Importance, Cellular Mechanisms, and Potential Interventions. Pediatr Neurol 95:42-66, 2019. DOI: 10.1016/j.pediatrneurol.2019.02.016
3. Hu M, Truong T, Zhu S, et al. Cognitive and Educational Outcomes After Preterm Birth or Low Birth Weight: An Umbrella Review and Meta-Analysis. JAMA Pediatr, 2026. DOI: 10.1001/jamapediatrics.2026.0533
4. Poets CF, Roberts RS, Schmidt B, et al. Association Between Intermittent Hypoxemia or Bradycardia and Late Death or Disability in Extremely Preterm Infants. JAMA 314:595-603, 2015. DOI: 10.1001/jama.2015.8841
5. Di Fiore JM, Raffay TM. The relationship between intermittent hypoxemia events and neural outcomes in neonates. Exp Neurol 342:113753, 2021. DOI: 10.1016/j.expneurol.2021.113753
6. Dassios T, Kaltsogianni O, Belani P, et al. Cumulative hypoxia, socioeconomic deprivation and neurodevelopmental outcomes in preterm infants. Respir Physiol Neurobiol 305:103942, 2022. DOI: 10.1016/j.resp.2022.103942
7. Boost II United Kingdom Collaborative Group, Boost II Australia Collaborative Group, Boost II New Zealand Collaborative Group, et al. Oxygen saturation and outcomes in preterm infants. N Engl J Med 368:2094-104, 2013. DOI: 10.1056/NEJMoa1302298
8. Askie LM, Darlow BA, Davis PG, et al. Effects of targeting lower versus higher arterial oxygen saturations on death or disability in preterm infants. Cochrane Database Syst Rev 4:CD011190, 2017. DOI: 10.1002/14651858.CD011190.pub2
9. Askie LM, Darlow BA, Finer N, et al. Association Between Oxygen Saturation Targeting and Death or Disability in Extremely Preterm Infants in the Neonatal Oxygenation Prospective Meta-analysis Collaboration. JAMA 319:2190-201, 2018. DOI: 10.1001/jama.2018.5725
10. Rasmussen MIS, Hansen ML, Pellicer A, et al. Cerebral Oximetry in Extremely Preterm Infants: 2-Year Follow-Up of the SafeBoosC-III Randomized Clinical Trial. JAMA Pediatr, 2026. DOI: 10.1001/jamapediatrics.2026.1066
11. Riddle A, Srivastava T, Wang K, et al. Mild Neonatal Hypoxia Targets Synaptic Maturation, Disrupts Adult Hippocampal Learning and Memory, and Is Associated with CK2-Mediated Loss of Synaptic Calcium-Activated Potassium Channel KCNN2 Activity. J Neurosci 46, 2026. DOI: 10.1523/JNEUROSCI.1643-25.2026
12. Gimenez M, Junque C, Narberhaus A, et al. Hippocampal gray matter reduction associates with memory deficits in adolescents with history of prematurity. Neuroimage 23:869-77, 2004. DOI: 10.1016/j.neuroimage.2004.07.029
13. Thompson DK, Wood SJ, Doyle LW, et al. Neonate hippocampal volumes: prematurity, perinatal predictors, and 2-year outcome. Ann Neurol 63:642-51, 2008. DOI: 10.1002/ana.21367
14. Thompson DK, Omizzolo C, Adamson C, et al. Longitudinal growth and morphology of the hippocampus through childhood: Impact of prematurity and implications for memory and learning. Hum Brain Mapp 35:4129-39, 2014. DOI: 10.1002/hbm.22464
15. Aanes S, Bjuland KJ, Skranes J, Lohaugen GC. Memory function and hippocampal volumes in preterm born very-low-birth-weight (VLBW) young adults. Neuroimage 105:76-83, 2015. DOI: 10.1016/j.neuroimage.2014.10.023
16. White TA, Miller SL, Sutherland AE, et al. Perinatal compromise affects development, form, and function of the hippocampus part two; preclinical studies. Pediatr Res 95:1709-19, 2024. DOI: 10.1038/s41390-024-03144-0
17. White TA, Miller SL, Sutherland AE, et al. Perinatal compromise affects development, form, and function of the hippocampus part one; clinical studies. Pediatr Res 95:1698-708, 2024. DOI: 10.1038/s41390-024-03105-7
18. McClendon E, Wang K, Degener-O’Brien K, et al. Transient Hypoxemia Disrupts Anatomical and Functional Maturation of Preterm Fetal Ovine CA1 Pyramidal Neurons. J Neurosci 39:7853-71, 2019. DOI: 10.1523/JNEUROSCI.1364-19.2019
19. Lodygensky GA, West T, Moravec MD, et al. Diffusion characteristics associated with neuronal injury and glial activation following hypoxia-ischemia in the immature brain. Magn Reson Med 66:839-45, 2011. DOI: 10.1002/mrm.22869
20. Sun J, Liu Y, Baudry M, Bi X. SK2 channel regulation of neuronal excitability, synaptic transmission, and brain rhythmic activity in health and diseases. Biochim Biophys Acta Mol Cell Res 1867:118834, 2020. DOI: 10.1016/j.bbamcr.2020.118834
21. Cassell SJ, Li W, Krautwald S, et al. Mechanism of SK2 channel gating and its modulation by the bee toxin apamin and small molecules. Elife 14, 2025. DOI: 10.7554/eLife.107733
22. Anderson PJ, Huppi PS. Neurodevelopmental Follow-Up of High-Risk Newborns. Chapter 14. In: Volpe JJ, Inder TE, Darras BT, De Vries LS, du Plessis AJ, Ferriero DM, Perlman JM, editors. Volpe’s Neurology of the Newborn. 7th ed. Philadelphia PA: Elsevier; 2025. p. 360-77.
23. Karmakar S, Chatterjee M, Basu M, Ghosh MK. CK2: The master regulator in tumor immune-microenvironment – A crucial target in oncotherapy. Eur J Pharmacol 994:177376, 2025. DOI: 10.1016/j.ejphar.2025.177376
24. Da Silva IIN, Ramirez D, Parylak SL, et al. CK2 inhibition suppresses glial inflammation in models of neuroinflammation and neurodegeneration. Nat Commun, 2026. DOI: 10.1038/s41467-026-71736-x