Although both neurons and glial cells express glutamate transporters and participate in clearing extracellular glutamate, several lines of evidence suggest that the vast majority of synaptically released glutamate is transported into glia [
35,
37‐
39]. The high-affinity glutamate transporters expressed in astrocytes are essential for taking up the majority of glutamate released at excitatory synapses, maintaining low extracellular glutamate and preventing glutamate neurotoxicity [
35,
37‐
39]. Blocking glial glutamate transporters results in both exaggerated amplitude and longer duration excitatory postsynaptic currents (EPSCs) [
40‐
43], suggesting that glial glutamate transporters play a critical role in limiting the extrasynaptic diffusion of glutamate [
44]. The loss of glial glutamate transporter GLAST or GLT-1 produced elevated extracellular glutamate levels, neurodegeneration characteristic of excitotoxicity, and a progressive paralysis [
37]. Mice lacking GLT-1 show lethal spontaneous seizures and increased susceptibility to acute cortical injury due to increased level of residual glutamate in the brain [
38]. Although it remains unknown how neuronal and glial glutamate transporters function differently in spinal sensory processing and chronic pain, the glial glutamate transporters certainly draw more attention since more than 80% extracellular glutamate is uptaken by glial cells. There have been several studies showing that glutamate transporters may also play a similar role in spinal cord. Immunohistochemical studies reveal that GLT-1 and GLAST are abundantly expressed in astrocytes but not neurons or microglia of spinal cord dorsal horn [
14,
35,
36,
45]. The glutamatergic synaptic transmission in spinal lamina II neurons was altered by pharmacologically blocking GLT-1 [
9]. Pharmacological inhibition of glutamate transporters in the spinal cord elevates spinal extracellular glutamate concentrations and produces spontaneous nociceptive behaviors [
10], and also resulted in the hypersensitivity of spinal dorsal horn neurons to mechanical and thermal stimuli [
11]. Down-regulation of glutamate transporter protein expression and attenuation of glutamate uptake activity in the spinal dorsal horn is associated with neuropathic pain induced by peripheral nerve injury [
12,
13,
45] or by chemotherapy [
14]. Gene transfer of GLT-1 in to spinal astrocyes attenuated the induction but not maintenance of inflammatory and neuropathic pain in rats [
46].
However, not all studies have been in agreement on the role of glial glutamte transporters in pain signaling. For example, the induction of allodynia following intrathecal injection of prostaglandin E2, prostaglandin F2, NMDA or AMPA was diminished by spinal application of TBOA [
47]; and inhibition of GLT-1 activity or expression reduced formalin-induced nociceptive behavior [
48]. It remains unclear whether these data indicate that glutamate transporters show plasticity with the onset of pathological pain or whether in these latter studies an excess of glutamate resulted in activation of presynaptic metabotropic receptors that in turn resulted in an anti-nociceptive response profile.
It is well known that repetitive low-frequency noxious stimuli applied to a spinal dorsal root result in a cumulative depolarization and a progressive increase in action potential output from dorsal horn neurons ("Windup") [
49‐
51]. The cellular mechanisms responsible for the generation of wind-up are complex and still incompletely understood. Although the buildup of synaptically released glutamate was thought to be involved in cumulative depolarization of postsynaptic dorsal horn neurons with repeated stimuli, there has not been an efficient way to measure the concentration of glutamate in synaptic cleft in spinal cord. Since the synapses are ensheathed by glial processes [
34], measurement of glial glutamate uptake currents induced by activation of the transporters has been suggested as an independent method of monitoring the presynaptic release of glutamate [
17,
26]. Several studies have shown that synaptically evoked glutamate transporter currents could be used to monitor the presynaptic release of glutamate during long-term potentiation [
23,
24,
52]. The data shown here is consistent with this thesis as glutamate transporter currents in spinal astrocytes increase in amplitude and duration with increasing strength of single afferent volleys delivered to the dorsal root entry zone. The range of stimulus intensities delivered to the dorsal root entry zone, 0.5–5 mA in amplitude and 0.2 ms in duration, were adequate to encompass both low-threshold non-nociceptive as well as higher threshold, potentially nociceptive fibers [
53,
54]. The constancy of time course of glutamate transporter currents across stimulus intensities indicate that spinal astrocytes use similar transporter capacity to manage glutamate release from both sets of afferents and additionally that there is normally a reserve capacity for both types of inputs [
39]. Thus, the gradual reduction in transporter currents observed with repetitive stimulation at the dorsal root entry zone across frequencies raises intriguing interpretative possibilities. At least two explanations come to mind. One, the reduced currents could reflect reduced release of presynaptic glutamate or alternatively. Two, it could reflect the saturation of the glial transporters during repetitive stimulation. If saturated, it would have been expected that later evoked transporter currents would have a longer decay time resulting from the build-up of higher concentrations of glutamate in the extrasynaptic space, but this was not observed. Moreover, glutamate transporters have very rapid turn-over rate in the continuous presence of glutamate [
17,
39], and the total transport capacity of glial transporters were not overwhelmed during a brief high-frequency stimulation train at near-physiological temperature, even when a majority of the transporters are blocked by a competitive antagonist [
55]. Thus, it seems unlikely that the glial transporters were saturated during low-frequency repetitive dorsal root entry zone stimulation in this study. Altenratively, it might be a concern that the cotransport of 2 Na
+ :1Glu
- with countertransport of 1 K
+ and 1 OH
- involved in glutamate uptake through glutamate transporter would result in translocation of 1 net positive charge and reduce the net electric gradient. But the activation of a Cl
- conductance concomitant with transport would provide a potential mechanism to offset the depolarizing action of transmitter reuptake and dampen cell excitability [
56]. Hence, the decrease of glial glutamate transporter currents recorded here during repetitive stimulation is concluded to reflect reduced release of presynaptic (afferent-derived) glutamate, which argues against the build-up of glutamate in the synaptic cleft in the formation of "windup" following repetitive stimulation. Further studies need to be done to clarify this issue.