We previously found that rat CB expresses mRNAs of VGluT1-3 and EAAT1-3 [21]. In this study, we demonstrated that mRNAs of all these glutamate transporters are also expressed in human CB. We further found that among the glutamate transporters, the proteins of VGluT1 and 3, EAAT2 and 3, rather than VGluT1 and EAAT1 were detected with diverse levels in human and rat CB. The expression of VGluT3, EAAT2 and 3 were enriched in human and rat CB. In contrast, VGluT1 was expressed at quite a low level in the CB when compared with the brain. Moreover, we determined that VGluT3 is localized in type I cells, but not in type II cells in rat CB, while EAAT2 and EAAT3 were distributed in type I cells and type II cells of the CB. Furthermore, we found that CIH exposure elevated the protein level of EAAT3 as well as TH, but attenuated the level of VGluT3 and EAAT2 in the CB. Taken together, these results coupled with our previous studies indicate that glutamate secretion and uptake systems may occur in the CB, and glutamate transporters may contribute to glutamatergic signaling-dependent carotid chemoreflex to CIH.
The CB is an important peripheral chemoreceptor that senses the changes of oxygen, carbon dioxide and pH levels in the arterial blood. The CB is composed of type I cells, type II cells, capillaries, connective tissue and nerve fibers. Type I cells are considered as oxygen sensing cells and type II cells are GFAP-positive glia-like cells that envelop the clusters of type I cells. Reduction of PaO2 in arterial blood can cause type I cells to depolarize, then release a variety of neurotransmitters to act on the corresponding receptors on the sinus nerve; subsequently, transmitting signals to the nucleus tractus solitarius (NTS), by which the changes are relayed to other centers in the brain, resulting in increased respiration and a coordinated cardiovascular response. It is reported that intermittent hypoxia leads to CB chemoreceptor plasticity [25], however, the molecular mechanism of this plasticity remains uncertain. In the CB, there are wide spread synaptic contacts not only between afferent nerve endings and type I cells, but also between neighboring type I cells [26]. Thus, the changes in the weight or strength of these synapses, which exist at multiple sites in the CB, are implicated in the plasticity of CB response to hypoxia. Studies have shown that CB contains a number of neurotransmitters and neuromodulators [27], such as dopamine, acetylcholine, ATP, nitric oxide, angiotensin, 5-HT. ATP and acetylcholine play an excitatory role by acting on the P2 × 2/3 receptors or nicotine acetylcholine receptors, whereas dopamine plays an inhibitory role by stimulating postsynaptic dopamine D2 receptor. Interestingly, glutamate, as a major excitatory neurotransmitter in the central nervous system, has not been significantly studied in the CB. As early as 1990, Torrealba [28] first found that there was a large amount of glutamate present in cat CB type I cells by immunohistochemistry staining, and they [29] speculated that glutamate in cat CB might be acting as a metabolite rather than a neurotransmitter.
Glutamate is the most abundant amino acid in the central nervous system, which is involved in a variety of physiological synaptic transduction, and induces the changes in synaptic morphology and function. The essential factors for glutamatergic neurotransmitters to play a role include glutamate, glutamate receptors and downstream signal molecules, VGluTs and EAATs. In our previous study [20, 21] we found that a variety of NMDA and AMPA ionotropic glutamate receptors are expressed in rat CB. CIH upregulated NMDA receptor NMDAR1, NMDAR2A/2B and AMPA receptor GluR1, indicating that NMDA and AMPA receptors might be involved in the response of CB to hypoxia. Our RT-PCR results from this study, together with our previous report, revealed the presence of all VGluT1-3 and EAAT1-3 transcripts in human CB (Fig. 1) and rat CB. Among these glutamate transporters, the proteins for VGluT1, VGluT3, EAAT2 and EAAT3 were detected in human and rat CB (Fig. 2). Of note, we demonstrated an inability to detect protein for VGluT2 and EAAT1 in human and rat CB, even though both proteins were detected in brain of positive control; therefore, indicating that mRNA transcript of VGluT2 and EAAT1 might not be translated in human and rat CB.
In the nervous system, the basic condition for the release of glutamate as a neurotransmitter is that glutamate is transported into the transmitter vesicles through the VGluTs [22]. Therefore, it is believed that the VGluTs are indirect markers of glutamate functioning as a neurotransmitter. There are three subtypes of VGluTs. VGluT1 and VGluT2 are mainly distributed in glutamatergic neurons, while VGluT3 is not only distributed in glutamatergic neurons, but also in non-glutamic neurons, such as cholinergic neurons, GABA neurons and so on. Based on the results of western blotting in the present study, VGluT3 is the major type of vesicular glutamate transporter in the CB. Immunostaining showed that VGluT3 was mainly expressed in CB type I cells (Fig. 4A and 4B). Thus, we speculate that glutamate may be released from CB type I cells to synaptic space as a neurotransmitter to influence the electrical signal transmission of chemoreceptors. However, further study is needed to clarify whether glutamate act as a neurotransmitter in the CB.
EAATs are located on presynaptic membrane, synaptic vesicle and glial cell membrane. They are important for the recycling of excitatory amino acids, the termination of excitatory signals and the protection of nerve cells from excitotoxic damage. In order to maintain the physiological effect, glutamate, which is released into the synaptic gap, needs to be reuptake through EAATs located in the nerve terminals or glial cells to avoid glutamate excitotoxicity. In the central nervous system, about 80% of glutamate is removed by reuptake into nerve endings and 20% into glial cells. There are five subtypes of EAATs family [22], in which EAAT1/2/3 are mainly distributed in the central nervous system, while EAAT4 is limited to the cerebellum [30]. EAAT5 is mainly distributed in the retina [31]. EAAT1 and EAAT2 are mainly expressed in the cell membrane of astrocytes, and remove the glutamate. On the other hand, EAAT3, EAAT4 and EAAT5 are mainly expressed in neurons, and some researchers even think that EAAT3 is neuron-specific. In the present study, we found that both human and rat CBs express mRNAs of all EAATs, as well as proteins of EAAT2 and EAAT3. EAAT2 were distributed in both type I and type II cells in the CB (Fig. 4C and 4D). However, EAAT3 was mainly expressed in type I cells (Fig. 4E and 4F), and EAAT2 seemed to have no cell heterogeneity. In any case, the expression of EAATs in CB may play an important role in synaptic physiological signal transmission between type I cells and nerve endings, or between type I cells.
It was found that hypoxia could promote glutamate secretion, increase extracellular glutamate concentration and glutamate excitatory toxicity [32]. In this study, we also found that CIH decreased the expression level of VGluT3 and EAAT2, but increased the expression level of EAAT3 (Fig. 5). Accumulated evidences demonstrate that astrocytes can release glutamate [33–35], and glutamate reuptake induced by EAATs depends on the Na+ influx of Na+-K+ ATP enzyme. Under hypoxia, ATP synthesis was decreased, which leads to the reverse transport of EAATs, resulting in the release of glutamate from astrocytes to the synaptic space [34]. It was also found that GABA presynaptic neurons could reuptake glutamate into cells through EAAT3, providing glutamate for GABA synthesis [36]. Based on our findings, we speculate that hypoxia damages the integrity of the cell membrane after CIH treatment for 2 weeks, and a large amount of glutamate is released from type I cells into the synaptic space (Fig. 6). Similarly, hypoxia reverses the ability of EAAT2 to reuptake glutamate, resulting in the release of glutamate from type II glial cells. In order to prevent this toxicity, the expression levels of VGluT3 and EAAT2 are decreased as compensation. However, the expression of EAAT3 is increased compensatively, and subsequently causes reuptake of glutamate into type I cells. On the one hand, these changes protect type I cells from hypoxia injury by reducing release of glutamate and increasing reuptake of glutamate. On the other hand, reuptake of glutamate provides essential amino acids for synthesis of GABA, which as an important inhibitory neurotransmitter. GABA can reduce the excitatory neurotoxicity caused by excessive glutamate and protect cells from death. It is must be noted, however, the effect of glutamate transporters in the glutamatergic signaling-dependent carotid chemoreflex to CIH is needed to be further clarified.