HCN3 has minimal involvement in the sensation of acute, inflammatory and neuropathic pain

Abstract

Hyperpolarization-activated cyclical nucleotide-gated (HCN) channels have recently garnered attention as drivers of chronic pain. These channels cause an inward current in response to membrane hyperpolarization, leading to the cyclical generation of action potentials. Of the four channels belonging to the HCN family (HCN1–4), HCN1 HCN2, and HCN3 are the main channels responsible for hyperpolarization-activated current in neurons. Previous research has shown that loss of HCN2 activity can eliminate thermal hyperalgesia in inflammatory pain, as well as both thermal and mechanical hyperalgesia in neuropathic pain. Additionally, HCN1 activity has been shown to regulate cold hypersensitivity and nerve injury-induced pain. Given the involvement of other HCN channels, Lainez et al. (2019) sought to determine the role of HCN3 in chronic pain. They first investigated the expression of HCN3 in sensory neurons by performing immunohistochemistry on cultured lumbar dorsal root ganglion (DRG) neurons. Staining with β3-tubulin and HCN3 antibodies showed that 60% of small (<20 μm diameter) and medium–large (>20 μm diameter) neurons expressed HCN3. Further staining was done to determine the co-localization of HCN3 with subpopulations of DRG neurons; about 35% of small unmyelinated neurons, 86% of large myelinated neurons, and 63% of small and medium neurons with nociceptive function were found to express HCN3. Together, these experiments reveal the abundant expression of HCN3 in DRG neurons of various functions and size classes. While these findings are consistent with a previous study showing widespread expression of HCN3 in rat DRG neurons (Chaplan et al. 2003), they contradict the low HCN3 expression found by a study in mice (Moosmang et al. 2001). Next, the authors investigated whether HCN3 contributes to the hyperpolarization-activated current (Ih). Whole-cell patch clamp experiments showed that 97 of 107 neurons with HCN2 deletion had a significant Ih, providing initial evidence that other isoforms contribute to the current. Previous research has identified HCN1 as predominantly involved in large neurons, whereas HCN3 probably contributes to the Ih in small and medium neurons (Momin et al. 2008). In combination with the low expression of HCN4 in these cells (Moosmang et al. 2001; Chaplan et al. 2003), HCN2 and HCN3 appear to be the major isoforms involved in small and medium neurons. Given this information, Lainez et al. (2019) then studied the relative contribution of HCN2 and HCN3 to Ih. To do so, they measured V1/2 following forskolin-induced cAMP elevation in wild type (WT), HCN2−/− and HCN3−/− neurons. The heightened levels of cAMP led to its binding to the various HCN isoforms; HCN1 and HCN3 are relatively insensitive to this interaction, but binding to HCN2 and HCN4 will cause an increase in V1/2. In small neurons a similar increase in V1/2 was seen in both WT and HCN3−/− groups after forskolin application, whereas no shift was seen in HCN2−/− neurons. This suggests that HCN2 is the main cAMP-sensitive isoform in small neurons, and that significant current is carried through HCN3. In medium neurons (20–30 μm diameter), an increase in V1/2 was again seen in both WT and HCN3−/− neurons, but the shift was significantly smaller in HCN3−/− neurons. Interestingly, application of forskolin increased V1/2 in HCN2−/− neurons, suggesting that HCN4 may play a greater role than previously thought. No significant change to V1/2 was seen in large neurons (>30 μm diameter) from WT, HCN2−/− or HCN3−/− mice following forskolin application, consistent with HCN1 being the main isoform present. Lainez et al. (2019) then studied how deletion of either isoform affected the maximum amplitude of the Ih current. A hyperpolarizing voltage step from −60 to −140 mV (from complete inactivity to maximal activation) was applied, and the Ih current was measured. In all neuronal sizes, the maximum Ih current was similar among WT and HCN3−/− neurons. While current density in small and medium HCN2−/− neurons was comparable to in the WT, large HCN2−/− neurons had a reduced current density. These results contradict previous findings that suggested only a minor role of HCN2 in the Ih of large neurons. To further test the contribution of the various HCN isoforms, Lainez et al. (2019) studied how deletion of HCN2 or HCN3 impacts the activation kinetics of DRG neurons. After a voltage step from −60 to −140 mV, current was recorded and fitted by a double exponential function. The results showed no change in the fast or slow time constants in both small and large HCN2−/− or HCN3−/− neurons. Conversely, both time constants were shortened in HCN3−/− but not HCN2−/− medium neurons. As such, HCN3 is probably the dominant HCN isoform in medium DRG neurons, and plays a role in shaping the properties of Ih. The excitability of smalland mediumdiameter DRG neurons were then tested by analysing their firing rate following injection of a depolarizing current. Application of current for 1 s in 10 pA steps from 0 to 60 pA revealed three different patterns of firing: silent (no firing), phasic (periods of firing) and tonic (continuous firing). The authors chose to analyse only neurons with a tonic firing pattern, because they are most likely to be relevant in chronic pain. They found no differences among small neurons taken from WT, HCN2−/− and HCN3−/− mice. However, medium HCN3−/− neurons displayed a higher firing frequency than WT and HCN2−/− neurons. All HCN3−/− neurons also displayed a higher frequency of phasic firing than was seen in WT and HCN2−/− neurons. The authors next evaluated whether HCN3 plays a role in acute, inflammatory and neuropathic pain. They first looked into acute pain by measuring responses to mechanical and thermal stimuli. Following von Frey stimulation to both WT and HCN3−/− mice, no difference