Chronic pain is associated with changes in expression of thousands of genes, making it hard to identify viable genetic targets for pain therapies. Pain researchers have thus turned to the much smaller number of epigenetic mechanisms, which scale gene expression up or down without editing DNA sequences, as possible targets for better treatments (see PRF related webinar and webinar summary). Now, researchers led by Hui-Lin Pan, University of Texas MD Anderson Cancer Center, Houston, US, report that the histone methylation enzyme G9a orchestrates large-scale changes in gene expression in injured rodent dorsal root ganglia (DRG) neurons, including silencing potassium channel genes important for neuronal excitability. In addition, they show that G9a activity is needed for the development and maintenance of chronic pain in the spinal nerve ligation (SNL) model of neuropathic pain in rodents.
“The authors have undoubtedly shown that G9a is key to the development and maintenance of long-term neuropathic pain states,” wrote Sandrine Geranton, University College London, UK, in an email to PRF, though she said it remains unclear which genes affected by G9a are responsible for pain relief. “G9a activity regulates the expression of many genes associated with neuropathic pain, and therefore it is currently impossible to attribute the analgesic effect of its inhibition to any specific gene.”
The findings were published November 9 in Nature Neuroscience.
The epigenetics of nerve injury-induced pain
SNL induces up- and downregulation of a large number of genes (Wang et al., 2002). Therefore, “it’s been really hard to get a global picture of how exactly different genes are involved in pain development,” said Pan. But since epigenetic processes, including post-translational modification of histones via acetylation and methylation, affect many genes, researchers have begun to explore how such alterations affect gene expression patterns in the development of chronic neuropathic pain (Denk et al., 2013); histones are the proteins around which DNA wraps.
In the new study, first author Geoffroy Laumet and colleagues examined the levels and activity of histone-modifying enzymes in rats that underwent ligation of the L5 and L6 spinal nerves or sham surgery. These histone modifiers included four histone deacetylases (HDACs) and two histone methyltransferases, including G9a and enhancer of zeste homolog 2 (EZH2), all of which have been linked to gene silencing. DRGs from injured animals had increased protein levels and messenger RNA (mRNA) expression of each histone modifier, excluding HDAC5, compared to cells from sham-operated controls. These changes were observed as early as five days following nerve injury and for at least four weeks. In addition, greater expression of these epigenetic regulators correlated with greater enzymatic activity in injured DRG, as suggested by decreased levels of acetylation and increased levels of methylation at particular substrates.
The researchers next asked how these epigenetic alterations in injured DRG neurons affected gene expression. Of particular interest were genes encoding potassium channels. “We knew from previous research that potassium channels are involved in maintaining resting membrane potential and regulating neuronal firing after nerve injury,” said Pan. Hence, they tracked the time course of DRG expression of four representative potassium channels responsible for different types of potassium currents, finding that SNL decreased levels of mRNA expression of each channel in the DRG, but not in the spinal cord, over four weeks, compared to sham controls. Like the epigenetic changes they observed earlier, the decreases in mRNA levels of all four potassium channels were detected at different time points after SNL, though whether this reflected an epigenetic mechanism remained unclear.
To address that issue, the authors examined histone methylation at promoter regions of the potassium channel genes. They found greater methylation mediated by G9a at promoter regions of each of the four potassium channel genes in injured DRG relative to controls. Next, they examined which of the histone modifiers was responsible for the silencing of potassium channel expression. Daily intrathecal administration of selective inhibitors of HDACs, G9a, or EZH2, or vehicle as control, over eight days after chronic pain was established revealed that inhibition of G9a normalized expression of all four potassium channel genes; inhibition of HDACs or EZH2 had variable effects on the expression of these genes. Inhibition of G9a also had clear effects on the electrophysiology of DRG neurons, as whole-cell recordings from a population of DRG neurons (expressing isolectin B4) showed that voltage-dependent potassium currents that were reduced in amplitude after nerve injury were restored to normal levels with the G9a inhibitor.
However, the four potassium channel genes were not the only genes regulated by G9a. Using RNA sequencing to analyze the gene expression profile in injured versus control DRG treated with either a G9a inhibitor or vehicle, the authors found that expression of over 2,000 genes was altered in injured DRG, with 42 of these being downregulated potassium channels. “What was really surprising was that the expression level of most (40) of the potassium channel genes was rescued by simply inhibiting G9a activity,” said Pan. In fact, G9a inhibition normalized expression of about 600 genes. Hence, G9a acted on a genomewide scale after nerve injury.
G9a drives pain behavior
To test whether G9a was necessary for chronic neuropathic pain, the authors intrathecally administered a G9a inhibitor over eight days to SNL rats and measured mechanical hyperalgesia with the Randall-Selitto paw pressure test, as well as tactile allodynia with von Frey filaments. Relative to vehicle treatment, G9a inhibition gradually restored baseline mechanical thresholds and alleviated tactile allodynia. However, consistent with previous findings (Denk et al., 2013), HDAC inhibition also eased pain sensitivity, but without affecting allodynia, pointing to the involvement of multiple histone modifiers in chronic pain. Indeed, combining G9a or HDAC inhibition with EZH2 inhibition produced analgesic effects greater than either treatment alone.
Since the earlier experiments indicated that G9a controls expression of hundreds of genes, “potassium channel genes are probably only part of the mechanism of how the G9a inhibitor reduced chronic pain—other signaling pathways are probably also responsible,” Pan said.
The analgesic effects seen with the G9a inhibitor were recapitulated with small interfering RNA (siRNA) specific to mRNA transcribed from the gene encoding G9a, Ehmt2. Since chronic pain was already well established before the investigators administered the siRNA or the G9a inhibitor, these results demonstrated that G9a is important for the maintenance of chronic pain after nerve injury.
Lastly, the authors used a conditional gene knockout approach to show that G9a was also necessary for the development of chronic pain. Here, mice with a deletion of Ehmt2 only in primary sensory DRG neurons displayed normal thermal and mechanical sensitivity, but failed to develop chronic pain symptoms after SNL, compared to control mice.
While the authors focused on injured DRG following SNL, Geranton said that G9a may also be important in uninjured DRG. “There is strong evidence to suggest that molecular changes in the uninjured afferents may underlie the development and/or maintenance of neuropathic pain as well,” she commented. “All drugs were delivered intrathecally and would have also impacted G9a expression/activity in the uninjured DRG. The same is possible with conditional knockout of G9a,” she added.
Pan is now studying how nerve injury recruits G9a, and whether the epigenetic mechanisms at play in the SNL model generalize to other chronic pain models.
Matthew Soleiman is a neuroscientist-turned-science writer currently residing in Nashville, Tennessee. Follow him on Twitter @MatthewSoleiman.
Image credit: Laumet et al., 2015