In many species, reduced glutamatergic signaling can be interpreted as a biologically conserved mechanism of aging similar to sarcopenia or osteoporosis (Cox et al. 2022). Accordingly, this decline and in general, synaptic plasticity damage should accelerate with premature aging, including during AD progression. The main aim of this study was to evaluate the contribution of age-related changes in the balance between glutamate and GABA systems to the onset and progression of AD in the brain. Unexpectedly, we did not find pronounced significant differences in the functioning of these systems in the hippocampus between OXYS and Wistar rats, either at the level of mRNA expression or at the protein level. On the other hand, it was noted that the balance between the glutamate and GABA systems changes greatly with age in both rat strains. It should be pointed out that in the literature, information about alterations of the levels of glutamate is contradictory, as are data about enzymes of glutamate synthesis and degradation during aging and AD development. Some studies have shown a decrease in glutamate levels in the hippocampus, anterior cingulate gyrus, and other areas of the brain during aging (Schubert et al. 2004; Kaiser et al. 2005; Chang et al. 2009; Huang et al. 2017). Other authors state that there are no changes in glutamate and glutamine synthesis in the rat brain, and that glutamate turnover does not change with age. This claim is supported by a report on glutamine synthetase activity (Segovia et al. 2001); that study did not reveal its alterations with age.
Our results indicate that the levels of glutamate and of key enzymes of its synthesis (glutaminase) and degradation (glutamine synthetase) do not change with age and do not differ between Wistar and OXYS rats, thus pointing to the stability of glutamate synthesis in the hippocampus of rats throughout the life span. It should be mentioned that the absence of changes in the level of glutamate cannot be definitively considered an indicator of glutamatergic-system stability during aging and during the development of AD signs. Excitotoxicity of glutamate is mediated primarily by a disruption of its reuptake system; this problem leads to a high concentration of glutamate in the synaptic cleft and, as a consequence, to hyperactivation of NMDA receptors (Mira and Cerpa 2020).
Thus, an important determinant of the availability of glutamate for signaling processes is the system of its reuptake and recycling. Glutamate cannot penetrate the blood–brain barrier and is produced mainly by neurons and astrocytes. Nonetheless, neurons themselves are unable to synthesize glutamate from glucose through the tricarboxylic acid (TCA) cycle owing to the absence of pyruvate carboxylase (Magi et al. 2019). In this context, an important role is played by the formation of glutamate in astrocytes that occurs in two ways: via de novo synthesis in the TCA cycle (it accounts for ~ 15% of glutamate) or through the “recirculation” of glutamine from GABA and glutamate via reuptake (Hampe et al. 2018). Under physiological conditions, astrocytes remove ~ 90% of all CNS-released glutamate through excitatory-amino-acid transporters GLAST and GLT1, which are required for the maintenance of low, nontoxic concentrations of this neurotransmitter (Rodríguez-Giraldo et al. 2022). As our study showed, both in Wistar and OXYS rats, the hippocampal content of the GLAST protein increased by the age of 12 months and diminished by the age of 18 months. On the other hand, there were no interstrain differences. We failed to detect age-related changes of the GLT1 level in Wistar rats, whereas in OXYS rats, it significantly rose from the age of 3 to 12 months. Possibly, this finding is due to an age-specific increase in the level and accumulation of toxic forms of β-amyloid in the hippocampus of OXYS rats (Stefanova et al. 2015b). It is important to note that in the hippocampus of AD patients, subtle changes in the amount of GLT1 have been detected too: the spatial pattern of expression of this transporter is reported to be altered, and elevated immunostaining of GLT1 is observed in outgrowths of astrocytes in the neuropil and especially in CA1–3 zones of the hippocampus and the dentate gyrus (Yeung et al. 2021a).
NMDA and AMPA receptors are present in approximately 70% of mammalian brain synapses, predominantly in the cerebral cortex, amygdala, striatum, and hippocampus (Revett et al., 2013). The specific location of these receptors is of great importance because the glutamatergic system plays an important part not only in neuroplasticity but also in excitotoxicity (Babaei 2021). It has been demonstrated that as the brain ages, the glutamatergic system (which is mediated by NMDA receptors) becomes hypofunctional, and this deficiency can lead to cognitive dysfunction during normal aging and in pathological conditions (Cox et al. 2022). In addition, there is evidence that with aging, there is a decrease in the number of hippocampal NMDA receptors (Kumar 2015; Kumar and Foster 2018), and this decline significantly reduces the bioavailability of glutamate (Avila et al. 2017). NMDA receptors are heterotetramers composed of two obligatory NMDAR1 subunits and two regulatory subunits GluN2 (A–D) or GluN3 (A or B). In the hippocampus (because it is the brain region that regulates cognitive functions), the regulatory subunits are mainly represented by NMDA2A or NMDA2B (GluN2A and GluN2B, respectively) (Cercato et al. 2017).
In this work, we assessed age-related changes in amounts of subunits NMDAR1 and NMDA2B in the hippocampus of rats. There were no alterations of the protein level of NMDAR2B with age, and this level did not differ between Wistar and OXYS rats, whereas the level of subunit NMDAR1 in the hippocampus of Wistar rats diminished with age and did not change in OXYS rats. As a result, by the age of 18 months, the amount of NMDAR1 became significantly higher in OXYS rats than in Wistar rats. It is worth mentioning that similar changes of the NMDAR1 protein level were found in patients with AD (Yeung et al. 2021b). It is possible that upregulation of NMDAR1 in the hippocampus in AD is a compensatory mechanism because it has been reported that an increase in amounts of subunits NMDAR1 and NMDA2A, but not NMDAR2B, is associated with the consolidation and formation of spatial memory in the hippocampus (Yeung et al. 2021b).
In the hippocampus, AMPA receptors are a component of the majority of excitatory synapses, especially in the CA1 region (~ 80% of all receptors). At the same time, the GluA1 subunit is regarded by many authors as the main culprit of impaired synaptic plasticity in early stages of AD (Qu et al. 2021). Our analysis of the GluA1 level did not reveal significant interstrain and age-related differences; however, it should be noted that we assessed the GluA1 protein level in the whole hippocampus, whereas alterations of GluA1 expression can have dissimilar directions among different hippocampal regions.
Glutamate is also a direct precursor of GABA in the CNS. It used be thought that GABAergic neurons are more resistant to pathological effects of β-amyloid as compared to cholinergic or glutamatergic neurons (Li et al. 2016). A new hypothesis is that GABAergic dysfunction causes an excitatory/inhibitory imbalance and makes neurons more vulnerable to adverse external factors and pathological stress; this new hypothesis explains the deficit of functional connections in the brain during the development of AD (Bi et al. 2020). By contrast, our assays of GABA in the hippocampus of OXYS and Wistar rats did not reveal any interstrain difference in the content of the neurotransmitter.
GABA is a major inhibitory neurotransmitter in the brain and is synthesized from glutamate by glutamic acid decarboxylase (GAD). In the mammalian brain, GAD has two isoforms, GAD65 and GAD67 (Lee et al. 2019). GAD65 is predominantly localized to presynaptic terminals, while GAD67 is distributed throughout the cell. Of note, over 90% of GABA in the brain is synthesized by GAD67 (Chattopadhyaya et al. 2007; Lau and Murthy 2012). GAD67 knockout mice die within a week of birth, but mice deficient in GAD67 expression are viable, although they exhibit abnormal behavior (Sandhu et al. 2014). In contrast, GAD65 knockout mice survive but are susceptible to seizures (Kash et al. 1997). GAD67 dysfunction is associated with such brain disorders as schizophrenia (Toritsuka et al. 2021), bipolar disorder (Benes et al. 2007), and Parkinson's disease (Lanoue et al. 2010). With respect to AD, it is reported that GAD67 expression is not altered in postmortem brain tissue samples from AD patients, but whether GAD67 is involved in AD progression is largely unknown (Wang et al. 2017). Another paper showed that age and gender do not affect GAD67 expression (Ethiraj et al. 2021). Our study indicates that the level of GABA-T, i.e., the enzyme responsible for the degradation of GABA in the brain, is significantly lower, while the level of GAD67, which catalyzes the formation of GABA, is higher in the hippocampus of OXYS rats compared to Wistar rats at all tested ages. These results point to an increased demand for the formation of GABA in the hippocampus of OXYS rats; however, we did not detect significant interstrain differences in the amount of the GABA transporter protein GAT1, which removes GABA from the synaptic cleft.
In vitro experiments have shown that β-amyloid neurotoxicity impairs the activity of GABAergic neurons and attenuates inhibitory postsynaptic potentials by suppressing postsynaptic GABA receptors (Krantic et al. 2012; Ulrich 2015). By contrast, our quantitation of the GABA receptor protein GABAAR1 points to its elevated level in the hippocampus of OXYS rats (compared to Wistar rats) both in the period of manifestation and during active progression of AD signs. In the hippocampus, GABAAR1 expression proved to be highest in the CA1 region, and, in agreement with a number of studies, it did not change or increase with age (Rissman and Mobley 2011; Palpagama et al. 2019). We believe that this GABAAR1 overexpression in the hippocampus of OXYS rats at age 12 months is due to neurodegenerative alterations that begin as early as at the age of 3 months and progress with age (Stefanova et al. 2014). Additionally, by the age of 12 months, there is significant accumulation of β-amyloid in the hippocampus of OXYS rats (Stefanova et al. 2015b), and this is a likely reason for the upregulation of GABAAR1; however, this hypothesis requires further investigation.
To assess of the impact of changes in the glutamate/GABA system during the development of the AD-like pathology, we also compared between OXYS and Wistar rats expression levels of genes related to glutamate and GABA signaling pathways in the hippocampus. Our analysis uncovered their significant changes with age in the hippocampus of both rat strains, but we found no differences between Wistar and OXYS rats. Thus, mRNA levels of the genes encoding components of these signaling pathways do not differ between Wistar and OXYS rats at all stages of the development of AD-like pathology. An exception that we noted is genes coding for glutamate receptors (Grin3b and Grm6), glutamate decarboxylase 2 (Gad2), G protein subunit (Gng12), solute carrier family (Slc1a1 and Slc1a2), phospholipase A2 (Pla2g2d, Pla2g5, and Pla2g6), phospholipase C (Plcl1), protein phosphatase 3 (Ppp3r1), and trafficking kinesin protein 2 (Trak2). Nevertheless, changes in the expression of these genes at different stages of neurodegeneration in OXYS rats did not allow us to formulate any hypothesis about a contribution of these genes to the pathogenesis of AD. In contrast, we identified clear-cut aging-dependent downregulation of genes associated with glutamate/GABA signaling in both rat strains.