Optogenetic analyses of inhibitory neurotransmission in the mouse spinal dorsal horn

Abstract

In all higher organisms, the experience of pain is associated with an unpleasant feeling. Nevertheless, our ability to perceive pain warns us of harmful threats that originate either from the inside of our body or from the environment. Perceiving pain relies on a well-functioning nociceptive system, which starts with nociceptive fibers conveying noxious inputs from the periphery to the first relay platform in the central nervous system (CNS), namely the spinal dorsal horn. At this site, inputs are modulated and filtered by a network of excitatory and inhibitory interneurons. The final output is relayed via projection neurons to higher cortical areas in the brain where pain becomes a conscious experience. However, in some cases nociception becomes dysfunctional and eventually leads to persistent and chronic pain. Underlying symptoms of chronic pain are allodynia, where innocuous stimuli such as light-touch become painful and hyperalgesia defined as an enhanced sensitivity to noxious stimuli. Studies have shown that reduced GABAergic or glycinergic neurotransmission in the dorsal horn leads to increased nociceptive signaling to the brain. This indicates that a functional inhibitory network in the dorsal horn is important for avoiding uncontrolled transmission of nociceptive stimuli to higher levels of the central nervous system. Several studies in rodent models of neuropathic or inflammatory pain have demonstrated that GABAA receptors containing the α2 subunit (α2-GABAA receptors) in addition to certain glycine receptors play a key role in the inhibitory control of nociceptive signaling at the spinal cord level. In the three experimental studies of this thesis, I used optogenetics and whole-cell patch-clamp recordings to provide a detailed analysis of inhibitory neurotransmission in the spinal dorsal horn of mice. In the first chapter, I focused on a genetic mouse model of diminished GABAergic inhibition at the spinal cord level. The mice used in this study were hoxb8α2-/- mice which lack α2-GABAA receptors in the spinal cord up to the cervical segment C4. Although these mice lacked a GABAA receptor subtype (α2-GABAA receptors) that is critically involved in spinal nociceptive control and although GABAergic inhibitory postsynaptic currents (GABA-IPSCs) recorded from excitatory superficial dorsal horn neurons were reduced, these mice did not show exaggerated pain responses indicating that some compensatory mechanisms were active in these mice. Several potential compensatory mechanisms were investigated. No significant differences were found for the decay kinetics of GABA-IPSCs. No changes were found in the amplitude of glycinergic IPSCs or of membrane currents evoked by extracellular application of GABA. I also did not find significant changes in the amplitudes of sensory afferent evoked excitatory postsynaptic currents (EPSCs), which are known to be under control by presynaptic GABAA receptors. By contrast, I found increased bicuculline-sensitive (GABAergic) tonic membrane currents in hoxb8α2-/- mice. Furthermore, in immunohistochemical experiments, I found a significant increase in serotonin immunoreactivity and in the expression of the serotonin producing enzyme tryptophan hydroxylase (TPH2). Both the increases in tonic GABAergic currents and in serotonergic input to the spinal cord may on a behavioral level compensate for the loss of synaptic GABAergic inhibition and may explain why hoxb8α2-/- mice lack a pronociceptive phenotype. In the second chapter, Part A, I aimed at identifying the origin of inhibitory inputs onto excitatory lamina II (LII) neurons. To this end, I used laser scanning photostimulation (LSPS) in combination with whole-cell patch-clamp recordings in spinal cord slices of BAC transgenic vGAT::ChR2 mice. These mice express the blue light-activated ion channel channelrhodopsin 2 in all inhibitory neurons of the CNS including the spinal cord. About half (46%) of the sites triggering inhibitory input to the recorded neurons were located in the superficial dorsal horn, one third was located in the deep dorsal horn, and 21% were in the white matter. These results indicate that inhibitory input to LII neurons does not only originate from inhibitory neurons of the same lamina but also from deep dorsal horn interneurons and very likely from inhibitory fiber tracts descending from higher CNS areas. This interpretation is consistent with results described in Part B, where I characterized inhibitory neurotransmission in GlyT2::Cre;vGAT::ChR2 mice in which dorsal horn glycinergic neurons were locally ablated by virus-mediated cre-dependent diphtheria toxin expression. In these experiments, I found a reduction of inhibitory neurotransmission to excitatory LII neurons by 55%. Because most of the glycinergic neurons are located in the deep dorsal horn, these results also support a strong contribution of deep dorsal horn neurons to inhibition of neurons and circuits of LII. Taken together, this thesis provides new insights on the organization of inhibitory neurotransmission in the spinal dorsal horn and on potential mechanisms contributing to a compensation of compromised synaptic inhibition in spinal dorsal horn nociceptive circuits.