← Writing

Hypoxia-Ischemia, Seizures, and Hypoglycemia

neurosciencenecrosisexcitotoxicity

Necrotic neuron death is an acute form of cell death triggered by external neurological insults, such as hypoxia-ischemia, seizures, and hypoglycemia. It is characterized by a disruption of ATP metabolism. It is one of three modes of neuron death, along with apoptosis and neurodegeneration, though it is substantially more damaging. Apoptosis is a cell's "programmed death," characterized by cell shrinkage, chromosome condensation, and DNA degradation as the cell dismantles itself cleanly. Neurodegeneration involves the progressive loss of neurons over many years, eventually leading to apoptotic cell death. By contrast, necrotic neuron death is inflammatory and acute.

A central question in necrotic neuron death is twofold. First, it is surprising that upon analysis of the brain tissue of patients deceased from necrotic neuron death, only cells in the hippocampus are consistently found dead. Second, although the neurological insults are severe, there is a delay in neuron death. The answer lies in the similarities of the mechanisms caused by hypoxia-ischemia, seizures, and hypoglycemia. Although each insult arises from a different trigger, they converge on the same cascade.

Upstream Mechanistic Differences

Hypoxia-ischemia — the lack of oxygen (hypoxia) caused by a critical reduction of blood flow to tissue (ischemia) — creates an energy crisis. When oxygen cannot reach cells, they cannot perform aerobic respiration and instead rely on anaerobic respiration, which produces far fewer ATP (2 ATP as opposed to ~34 ATP in aerobic respiration). This causes the failure of energy-dependent cellular mechanisms, in particular the pumps that stabilize ion concentrations: Na⁺/K⁺-ATPase (NKA), the Na⁺/Ca²⁺ exchanger (NCX), and Ca²⁺-ATPase. In this scenario, Na⁺ floods in and K⁺ floods out, causing neurons to depolarize. This depolarization opens voltage-gated Ca²⁺ channels, so Ca²⁺ enters the cell and accumulates, since the pumps that normally eject calcium no longer work. This causes the downstream impairment of glutamate transporters, which rely on the ion gradient established by the pumps, such that the transporters begin pumping glutamate out into the synapse [Shen et al., 2022]. Interestingly, this mechanism is particularly prominent in areas of partial ischemia, called the penumbra, which exist where one of several blood vessels is blocked [Ermine et al., 2021]. In this situation, there is just enough oxygen for lactate fermentation, but not enough for oxidative phosphorylation, resulting in acidosis. The low pH further denatures proteins and disrupts enzyme function. In areas of complete ischemia, by contrast, cells die almost instantly.

The upstream mechanisms of seizures are different. Seizures occur when neurons fire in large, synchronized waves rather than at the controlled pace of normal brain activity. This can happen for three reasons. First, glutamate circuits themselves can become overactive. For example, gain-of-function mutations in K⁺ channel proteins shorten the refractory period by allowing repolarization to occur faster, as observed in mutations in KCNH2 (a potassium channel) causing shorter refractory periods [Hancox et al., 2023]. Additionally, gain-of-function mutations in AMPA receptors that remove the GluA2 subunit (which normally blocks calcium) produce calcium-permeable AMPA receptors, so calcium enters the cell whenever glutamate binds [Wright and Vissel, 2012]. This calcium influx can act as a secondary messenger that recruits more receptors and amplifies future excitation. Second, this can result from failures in inhibitory GABAergic circuits. In vitro experiments have shown that endocytosis of GABA-A receptors from the synaptic membrane (reducing inhibition) causes seizures [Blair et al., 2004]. This can result from fever in febrile seizures [Kang et al., 2006], traumatic brain injury [Guerriero et al., 2015], or neuroinflammation. Third, there are extraneous causes of increased neuron firing rate — for instance, in febrile seizures, hyperthermia lowers the threshold for action potential generation and increases neuron firing rate.

Hypoglycemia is a condition, common in diabetics, in which blood glucose drops below 70 mg/dL. This is a particular problem for neurons, which store virtually no glycogen and perform minimal fatty acid synthesis. When blood glucose falls, neurons lose their main substrate for oxidative phosphorylation; the electron transport chain cannot be sustained and ATP production collapses.

End Result Similarities

Once an energy crisis occurs, the pathway to necrotic cell death is extremely similar across the three neurological insults. The collapsed ion gradient reverses excitatory amino acid transporters (EAATs), whose role is to clear amino acids like glutamate into glial cells. This is particularly pronounced in hypoxia-ischemia and hypoglycemia, while in seizures the primary mechanism instead involves excessive vesicular release of glutamate from hyperactive neurons. This reversal relies on electrochemical gradients through secondary active transport that uses Na⁺, H⁺, and K⁺ movement as the driving force, causing excitatory amino acids to be pumped out of astrocytes into the extracellular space, as demonstrated by Attwell et al. [2000]. An abundance of these amino acids overactivates NMDA receptors — the "excitotoxicity" this essay refers to. Specifically, glutamate, glycine, and D-serine bind to NMDA receptors, where glutamate induces a conformational change in the GluN2 subunit and glycine/D-serine binds the GluN1 subunit [Yu and Lau, 2018]. Simultaneously, AMPA receptors are activated by glutamate, causing a depolarization of the membrane that spreads to NMDA receptors, expelling Mg²⁺ from the channel pore and allowing Ca²⁺ to pass through [Mayer et al., 1984]. This is additionally pertinent in seizures, where AMPA receptors themselves become more permeable to Ca²⁺. Ultimately, this leads to several pathologies associated with the buildup of Ca²⁺, including mitochondrial failure, cytoskeletal damage, generation of reactive oxygen species, and calcium-dependent phosphorylation of microtubule proteins — the proximate cause of necrotic neuron death.

Differences in Location of Damage

Hypoxia-ischemia, seizures, and hypoglycemia affect different parts of the brain. During hypoxia-ischemia, the CA1 region of the hippocampus is particularly vulnerable, for several reasons. First, CA1 is supplied by a single artery, compared to other brain regions that have multiple capillaries surrounding neurons, so a blood clot will be most damaging to this region. Second, the baseline energy demand in CA1 has been found to be higher, so neurons there are the first to exhaust their supplies. Third, Giovannini et al. [2020] observed that CA1 shows higher expression of NR2A/B (subunits of the NMDA receptor), which undergo increased phosphorylation during hypoxia-ischemia and become more permeable to calcium.

Damage caused by hypoglycemia is different, as Auer and Siesjö discovered in 1988. It affects many different parts of the brain, including the cerebellum, caudate nucleus, cerebral cortex, and spinal cord, along with the hippocampus (where the subiculum and dentate gyrus crest were the earliest affected, followed by CA1). This is because hypoglycemia is a restriction only on glucose, rather than oxygen as well, and therefore affects metabolically active regions more severely than hypoxia-ischemia does. Additionally, Fujioka et al. [1997] observed that in hypoglycemia-damaged brains there were no minor hemorrhages, since the lactic acid buildup from acidosis in hypoxia-ischemia — which damages blood vessels — does not occur.

Seizures are different still. Because they arise from excitatory neurons driving neighboring neurons to fire, the damage is particularly prominent in areas with dense excitatory connectivity, such as the hippocampus, amygdala, thalamus, and other limbic structures [Dudek and Staley, 2012]. The location of seizure damage also depends on whether the seizure is petit mal, grand mal, or focal. Petit mal seizures tend to affect the thalamus more strongly; grand mal seizures are more pertinent in the motor cortex; and focal seizures are localized mostly in the hippocampus and accumulate there, which is why the hippocampus is uniquely damaged over time, just as it is in hypoxia-ischemia and hypoglycemia.

Conclusion

Hypoxia-ischemia, seizures, and hypoglycemia are neurological insults that arise from completely different upstream mechanisms: lack of oxygen, abnormal synchronized neuronal firing, and glucose deprivation. However, by creating an energy crisis, they converge on the same shared downstream pathway that results in excitotoxic cell death. When energy is no longer present, ATP-dependent ion pumps stop working, leading to the accumulation of extracellular glutamate. Despite this similarity, the three insults also affect different regions of the brain, though the hippocampus is affected across all three. In hypoxia-ischemia, the CA1 region is disproportionately vulnerable; in hypoglycemia, metabolically active regions are targeted in particular; and in seizures, damage is concentrated in regions of dense neuronal connectivity.

This analysis of the similarities and differences among the three insults carries positive clinical implications. The shared downstream pathway can be a target for therapeutics, such as NMDA receptor inhibitors or calcium channel blockers — treatments that could be effective not just for the individual conditions, but for necrotic neuron death as a whole.

References

  1. Attwell, D., Barbour, B., & Szatkowski, M. (2000). Glutamate release in severe brain ischaemia is mainly by reversed uptake. Nature, 403(6767), 316–318.
  2. Auer, R. N., & Siesjö, B. K. (1988). Biological differences between ischemia, hypoglycemia, and epilepsy. Annals of Neurology, 24(6), 699–707.
  3. Blair, R. E., Sombati, S., Lawrence, D. C., McCay, B. D., & DeLorenzo, R. J. (2004). Epileptogenesis causes acute and chronic increases in GABA-A receptor endocytosis that contributes to the induction and maintenance of seizures in the hippocampal culture model of acquired epilepsy. Journal of Pharmacology and Experimental Therapeutics, 310(3), 871–880.
  4. Dudek, F. E., & Staley, K. J. (2012). When and how do seizures kill neurons, and is cell death relevant to epileptogenesis? In Jasper's Basic Mechanisms of the Epilepsies (4th ed.). Oxford University Press.
  5. Ermine, C. M., Bivard, A., Parsons, M. W., & Baron, J.-C. (2021). The ischemic penumbra: from concept to reality. International Journal of Stroke, 16(5), 497–509.
  6. Fujioka, M., Okuchi, K., Hiramatsu, K.-I., Sakaki, T., Sakaguchi, S., & Ishii, Y. (1997). Specific changes in human brain after hypoglycemic injury. Stroke, 28(3), 584–587.
  7. Giovannini, M. G., Lana, D., & Pepeu, G. (2020). CA1 and CA3 pyramidal neurons display different degrees of vulnerability to hypoxia-ischemia-induced damage. Frontiers in Cellular Neuroscience, 14, 585833.
  8. Guerriero, R. M., Giza, C. C., & Rotenberg, A. (2015). Glutamate and GABA imbalance following traumatic brain injury. Current Neurology and Neuroscience Reports, 15, 27.
  9. Hancox, J. C., Du, C.-Y., Butler, A., Zhang, Y., Dempsey, C. E., Harmer, S. C., & Zhang, H. (2023). Pro-arrhythmic effects of gain-of-function potassium channel mutations in the short QT syndrome. Philosophical Transactions of the Royal Society B, 378(1879), 20220165.
  10. Kang, J.-Q., Shen, W., & Macdonald, R. L. (2006). Why does fever trigger febrile seizures? GABA-A receptor γ2 subunit mutations associated with idiopathic generalized epilepsies have temperature-dependent trafficking deficiencies. Journal of Neuroscience, 26(9), 2590–2597.
  11. Mayer, M. L., Westbrook, G. L., & Guthrie, P. B. (1984). Voltage-dependent block by Mg²⁺ of NMDA responses in spinal cord neurones. Nature, 309(5965), 261–263.
  12. Shen, Z., Xiang, M., Chen, C., Ding, F., Wang, Y., Shang, C., Xin, L., Zhang, Y., & Cui, X. (2022). Glutamate excitotoxicity: potential therapeutic target for ischemic stroke. Biomedicine & Pharmacotherapy, 151, 113125.
  13. Wright, A., & Vissel, B. (2012). The essential role of AMPA receptor GluR2 subunit RNA editing in the normal and diseased brain. Frontiers in Molecular Neuroscience, 5, 34.
  14. Yu, A., & Lau, A. Y. (2018). Glutamate and glycine binding to the NMDA receptor. Structure, 26(7), 1035–1043.