Targeting Epigenetic Mechanisms for Chronic Visceral Pain: A Valid Approach for the Development of Novel Therapeutics.

Abstract

Background:

Chronic visceral pain is persistent pain emanating from thoracic, pelvic or abdominal origin that is poorly localized with regard to the specific organ affected. The prevalence can range up to 25% in the adult population as chronic visceral pain is a common feature of many visceral disorders which may or may not be accompanied by distinct structural or histological abnormalities within the visceral organs. Mounting evidence suggest that changes in epigenetic mechanisms are involved in the top-down or bottom-up sensitization of pain pathways and the development of chronic pain. Epigenetic changes can lead to long-term alterations in gene expression profiles of neurons and consequently alter functionality of peripheral neurons, dorsal root ganglia (DRG), spinal cord, and brain neurons. However, epigenetic modifications are dynamic and thus detrimental changes may be reversible. Hence, external factors/therapeutic interventions may be capable of modulating the epigenome and restore normal gene expression for extended periods of time.

Purpose:

The goal of this review is to highlight the latest discoveries made towards understanding the epigenetic mechanisms that are involved in the development or maintenance of chronic visceral pain. Furthermore, this review will provide evidence supporting that targeting these epigenetic mechanisms may represent a novel approach to treat chronic visceral pain.

1. Introduction

Chronic pain is defined as pain lasting longer than three months after the resolution or in absence of an injury, and is experienced by patients as a marked reduction in the threshold required to induce pain. In patients suffering from chronic pain an innocuous stimulus can cause pain (allodynia) and/or noxious stimulus can trigger an amplified response (hyperalgesia).¹ Chronic pain is estimated to affect 100 million adults in the United States. Chronic pain significantly reduces a patient’s quality of life, but also increases the risk of developing mental health disorders such as anxiety and depression. The combined financial cost to society of chronic pain due to health care and loss of productivity is enormous and ranges up to a staggering 600 billion in the United States alone.² This review will focus on chronic visceral pain, which is persistent pain emanating from thoracic, pelvic or abdominal origin that is poorly localized with regard to the specific organ affected. The prevalence of visceral pain is higher in women than in men and can range up to 25% in the adult population.³ This high prevalence can be explained by the fact that chronic visceral pain is a common feature of many visceral disorders which may or may not be accompanied by distinct structural or histological abnormalities within the visceral organ. Pain disorders with concurrent mucosal inflammation describe inflammatory bowel disease (IBD), pancreatitis, bladder pain syndrome/interstitial cystitis (BPS/IC), and gynecological pain; whereas pain disorders without distinct structural or histological abnormalities include functional dyspepsia, functional chest pain, functional heartburn, functional dysphagia, irritable bowel syndrome (IBS), centrally mediated abdominal pain syndrome (CAPS), narcotic bowel syndrome (NBS), and functional anorectal pain (Figure 1).⁴ Reliable therapies that relieve chronic visceral pain represents a major unmet medical need, and such therapies would greatly improve the patient’s quality of life and reduce the burden to society of managing these patients. The efficacy of current therapeutics to treat visceral pain such as opioids and non-steroidal anti-inflammatory drugs is far from optimal and can produce unwanted and serious side-effects.⁵ Despite extensive research on the neurobiological mechanisms of chronic visceral pain, our understanding of the underlying molecular pathways is incomplete. This knowledge gap hampers the highly desirable and anticipated development of new visceral pain-relieving medications. Chronic pain develops when neuronal networks become ‘sensitized’. This process starts when the balance of neurotransmitters, receptors and other molecules in neurons is disturbed. In this way, the cellular functionality of peripheral neurons, dorsal root ganglia (DRG), spinal cord, and brain neurons can be altered.^1,6 Mounting evidence suggests that epigenetic modifications play an important role in sensitization and the development of chronic pain.^7–10 Epigenetic modifications are described as stable alterations in gene expression that arise during development and cell proliferation. However, epigenetic modifications are dynamic and can also be influenced by external environmental factors. In this way, external factors are capable of modulating gene expression for extended periods of time following a stimulus through changes in the epigenome. The importance of epigenetic modifications in neurons was first highlighted in synaptic plasticity during memory formation.^10–14 Interestingly, the epigenetic changes that occur during memory formation are similar to those that occur during the neuronal sensitization process. Epigenome alterations form an attractive molecular mechanism for sensitization, as these changes can lead to long-term changes in neuronal gene expression of the neurotransmitters, receptors, and other molecules involved in sensitization. Once established, these epigenetic changes that occur during sensitization may also explain chronic pain on a molecular level. In this review, we summarize the latest discoveries made towards understanding the epigenetic mechanisms that are involved in chronic visceral pain. Furthermore, we argue that targeting these epigenetic mechanisms may become a novel treatment for chronic visceral pain. As more and more drugs targeting the epigenome become available, we can start targeting the underlying causes of chronic visceral pain, instead of treating the symptoms with traditional analgesics.

Figure 1: Common Visceral Pain Disorders.

Left side: Visceral pain disorders associates with overt tissue inflammation, which may or may not be due to an infection. Right side: Visceral pain disorders with no identified pathological marker. Visceral pain is typically diffuse and may be referred to somatic structures (muscle, skin, joints) within the same or adjacent dermatomes or to other visceral organs. The location of selected organs is approximate. The name of the disorder has been illustrated near the affected organ. GERD, gastroesophageal reflux disease; IBD, inflammatory bowel disease; BPS, bladder pain syndrome; IC, interstitial cystitis; IBS, irritable bowel syndrome; NBS, narcotic bowel syndrome. This composite image was assembled from individual public domain images available from Wikimedia Commons.

2. The development of chronic visceral pain – neuronal sensitization from bottom up or top down

2.1. The acute pain pathway

Visceral organs are innervated by nociceptive neurons with cell bodies in the DRG. The free nerve endings of these neurons contain multiple receptor types to detect potential harmful changes in pH, stretch, temperature, or specific chemicals mediators such as those released in response to chronic stress.^15–17 The acute pain pathway is activated when noxious signals are registered in the periphery. These signals are transmitted to the dorsal horn of the spinal cord, where the first synapse in the pain pathway is located. Usually visceral afferents synapse at multiple spinal levels, which causes a diffuse localization of the initial noxious signal.¹⁸ The dorsal horn of the spinal cord is an important location in the pain pathway as here the pain signal can be modulated by local inhibitory interneurons or descending projections from the brainstem, before sending the signal to the brain.^6,19 The final synapse occurs in the thalamus, which acts as the primary hub in the central pain matrix.²⁰ From the thalamus, the signal is dispersed to several different brain regions. Distributing the signal to the amygdala, anterior cingulate cortex, hippocampus and nucleus accumbens, process the emotional component of pain.^21,22 The signal is sent to the somatosensory cortex for the localization of the pain and other regions such as the prefrontal cortex, cingulate and parietal cortex to define the quality and magnitude of the pain signal.²³ The underlying mechanism that drives the transition from acute to chronic pain is not well understood. However, evidence suggests that the acute to chronic pain transition requires certain adaptations in the aforementioned pain pathways.²⁴ In chronic visceral pain, the pain pathway can become detrimentally modulated when sensitization of primary (peripheral afferent), secondary (spinal) or tertiary (brainstem/thalamic) neurons occurs. Interestingly, sensitization can occur through a bottom-up or top down mechanism.

2.2. Sensitization from bottom-up mechanisms

Under normal circumstances, tissue injury or the release of local inflammatory mediators (chemokines, cytokines, histamine, proteases, prostaglandins, serotonin, corticotropin-releasing hormone (CRH)) trigger peripheral neurons to release neuromodulators (calcitonin gene related peptide (CGRP), nitric oxide (NO), substance P) that stimulate nerve activity.^25,26 Sensitization starts when there is prolonged exposure to the initial stimulus and release of neuromodulators that activate second messenger signaling cascades, the phosphorylation and/or altered expression of certain receptors (cation channels), which may affect the cellular pathways that ensure proper neuronal function. In this way, the action potential threshold of the affected neuron can be lowered or the number of action potentials fired upon reaching threshold can be increased.¹ In the bottom-up scenario, sensitization usually starts at the level of the primary peripheral neurons. As a result, the enhanced neuronal excitability of the primary nociceptive neuron will lead to increased neurotransmitter and neuromodulator release in the dorsal horn of the spinal cord after stimulation. On the secondary level, the increased and/or prolonged input from primary afferent neurons will increase the number of Ca²⁺ channels in dorsal horn neurons. Altering the Ca²⁺ permeability of dorsal horn neurons will lower the action potential threshold of these neurons. As a consequence, these neurons can fire action potentials with less input from the primary neurons. Another potential pathway leading to dorsal horn sensitization is through dysfunction of inhibitory interneurons. These neurons receive signaling input either directly from the sensitized primary neuron or indirectly via descending neurons from the brain.^27,28 Increased peripheral excitability and/or decreased inhibitory tone can lead to remodeling and persistent excitation of second order neurons and thus contribute to chronic pain.²⁹ Sensitization can also happen at the tertiary level. Increased neuronal afferent stimulation can cause remodeling in the brain regions of the central pain matrix. Enhanced signaling from the thalamus will subsequently sensitize the integration nuclei (amygdala, hippocampus, insula, or cingulate), which can also alter activation thresholds in these neurons. When sensitization has occurred in these regions, previously innocuous stimuli will not only be perceived as noxious, but also trigger the negative emotional response to chronic pain.

2.3. Sensitization from top-down mechanisms

It is known that stress and negative emotions can enhance the perception of nociception in the absence of an overt peripheral injury.^30–32 This happens through the direct sensitization of the central pain matrix. The importance of this top-down mechanism for visceral pain is illustrated by epidemiological evidence showing correlations between chronic visceral pain and stress, anxiety and depression.^33,34 Evidence from experimental models has shown that persistent stress can facilitate pain perception and sensitize pain pathways, promoting chronic visceral pain disorders.^35,36 An important mediator of this pathway is the hypothalamus-pituitary-adrenal (HPA) axis. Stress can initiate the HPA axis, by activating the paraventricular nucleus of the hypothalamus (PVN) to secrete CRH into to hypophyseal portal circulation. CRH binds to CRH₁ in the anterior pituitary gland, which stimulates the release of adrenocorticotropic hormone (ACTH) in the systemic circulation. In turn, ACTH binds in the adrenal cortex and cause the de novo synthesis and release of cortisol in humans and corticosterone in rats in the systemic circulation (CORT). Circulating CORT has a dual role, depending on the receptors and regions where it binds. When CORT binds to its high-affinity mineralocorticoid receptor (MR) and low-affinity glucocorticoid receptor (GR) in the hippocampus, PVN and cortical regions, it will trigger signaling cascades that induce a negative feedback loop in order to terminate the stress response. However, when CORT binds the same receptors in the amygdala, it increases CRH expression which opposes the feedback inhibition and facilitates the stress axis^37–41. The central nucleus of the amygdala (CeA) plays an important role in visceral pain as this region integrates viscerosensory signaling with neuroendocrine and autonomic responses to stressors through the expression of MR, GR and CRH.^42–44 The importance of CeA signaling has been illustrated by directly stereotaxically implanting CORT-containing micropellets bilaterally on the dorsal margin of the CeA of healthy rats. Elevated amygdala CORT induced visceral hypersensitivity, through CORT-mediated decreases in GR and increases in CRH within the CeA. Simultaneous infusions of GR agonists or CRH antagonists in the CeA reversed the CORT-induced visceral hypersensitivity. The same effects on visceral sensitivity and GR/CRH expression were found in rats after they underwent repeated water avoidance stress (WAS).⁴³ Through neuronal remodeling and unbalancing GR/CRH signaling in the CeA, WAS induces visceral hypersensitivity to colonic distension in rodents^45,46. HPA axis dysfunction can exacerbate visceral pain during stress.⁴⁷ Interestingly these effects persisted long after the CORT pellets were depleted or the repeated WAS procedure was discontinued.^45,48 Acute and chronic stress can alter the signaling in the central pain matrix, especially through HPA axis dysregulation and altered CeA activity/signaling. This can lead to exacerbation of pain perception and/or diminishment of anti-nociceptive and anti-stress signaling.⁴⁹ These mechanisms have been elucidated in patients with irritable bowel syndrome (IBS) where brain imaging studies revealed increased activity in the amygdala after colorectal stimulation.⁵⁰ Furthermore, these patients had an abnormal HPA axis reactivity to CRH challenge and an overall increase in CORT secretion, which is indicative of HPA axis dysregulation, and likely contributed to the chronic visceral pain in these patients.^51,52

3. Epigenetic changes

Epigenetic changes refer to stable alterations in (long-term) gene expression potential resulting from developmental or environmental signals. These alterations are the result of cellular mechanisms that integrate external signals by structurally adapting chromosomal regions to register, signal, or perpetuate altered activity states or by modulating transcript levels.⁵³ Epigenetic marks are essential for the structural and functional adaptation of chromosomal regions, making those regions more or less accessible for gene transcription. In this way, gene expression is modulated without interfering with the base pair sequence of the DNA itself. Structurally, epigenetic marks play an indispensable role in DNA condensation. By interacting with numerous other proteins, these marks allow the packing of an extremely long DNA strand into an extremely small nucleus. 140 base pairs of DNA, wrapped around a histone (H) octamer, forms a nucleosome. The central histone octamer is formed by two copies of each of the ‘core histones’ H2A, H2B, H3 and H4.⁵⁴ The N-terminal histone tail protrudes from the nucleosome and is available for post-translational modifications which can attract DNA-condensation or DNA-unwrapping enzymes. Nucleosomes are further condensed in chromatin, which can switch between two forms: the “open” euchromatin and the “closed” heterochromatin. In euchromatin, coding DNA sequences are poised for transcriptional initiation and enhancer activity when they are detached from the nucleosomes. In contrast heterochromatin is a highly compact state where genes are silenced.⁵⁵ Functionally, the precise interplay between transcription initiation factors and specific epigenetic marks allow the unwrapping of DNA at specific locations, where gene transcription needs to be initiated, without having to unwrap the whole chromatin structure. The multitude of these dynamic epigenetic “marks” are known as the epigenome. Interestingly, changing the epigenome does not always require the presence of chronic or prolonged stimuli. For instance, in the process of memory formation, short-lived external stimuli can also alter epigenetic marks causing persistent changes in gene expression.^56–58 Here we will briefly discuss the mechanism by which DNA methylation, histone (de)acetylation and non-coding RNA’s can induce changes in long-term gene expression.

3.1. Histone modifications

As stated above, DNA is wrapped around a central histone octamer, forming nucleosomes. The protruding N-terminal histone tail is available for post-translational modifications: lysine acetylation, serine/threonine phosphorylation, lysine/arginine methylation, ubiquitination and ADP-ribosylation (Figure 2A).⁵⁹ Of all these modifications, histone (de)acetylation and (de)methylation are the best studied. Histone acetyltransferases (HAT) can add acetyl groups to the lysine residues (K) of the histone tail. Acetylated histones are usually associated with increased gene expression.⁶⁰ In contrast, histone deacetylases (HDAC) can remove acetyl groups and stabilize the positive charge of the lysine side chains to enhance the binding of DNA to the histones. Hence, deacetylation usually prevents binding of transcription factors on the DNA and is associated with reduced gene expression.⁶¹ The activity of histone methyltransferases and demethylases determine the degree of histone side chain methylation. This modification also influences gene expression, but is not as straightforward as histone acetylation, because the effect on gene expression depends on the number of added methyl groups and the particular lysine residue methylated. For instance, mono-methylation of H3K9, H3K27 and H4K20 is associated with transcriptional activation whereas di- or tri-methylation are associated with transcriptional repression. Methylation of H3K36 and H3K79 are associated with active transcriptional elongation, but also repress transcriptional from cryptic promoters in gene bodies.^62–65

Figure 2: Epigenetic Modifications.

A) Nucleosomes are composed of an octamer of histones H2A, H2B, H3, and H4 (two dimers of 2A/2B and two dimers of H3/H4) with 146 base pairs of DNA wrapped around the complex (drawn as black double line). The H1 histone may also be attached as a linker to stabilize the DNA (not shown). Several epigenetically regulated modifications to the n-terminal tails of the histones influence DNA accessibility to transcription machinery including acetylation (red circles), methylation (gray circles), phosphorylation (purple circles), or ADP ribosylation (blue circles). The single letter abbreviation was used for each amino acid, with spaces after every ten, and subscript numbers indicating location. Multiple circles on a single amino acid represent alternative modifications. N-terminal sequences were based on consensus sequences for human histones at uniport.org: H2A - P0C0S8, H2B - P62807, H3 - P68431, H4 – P62805. B) DNA methyl transferases (DNMT) target cytosine residues immediately upstream of guanosine residues (CpG sites) within DNA to form 5-methylcytosine (5mC), which typically represses transcription. Ten-eleven-translocation (TET) proteins participate in demethylation by converting 5mC to 5-hydroxymethylcytosine (5hmC), which has also been demonstrated to participate in epigenetic regulation of gene expression. Further oxidation by TET proteins changes 5hmC to 5-formylcytosine (5fC) and subsequently 5-carboxylcytosine (5caC), which can then be converted back to cytosine (not pictured). The role of 5fC and 5caC in modulation of gene expression is uncertain due to the recent development of tools that can distinguish between 5mC, 5hmC, 5fC, and 5caC. For each structure, R is the sugar (deoxyribose)-phosphate group that forms the DNA nucleotide.

Histone modifications are dynamic, reversible and interconnected. The location, presence and combination of particular histone modifications form docking sites for other proteins. This “histone code” will recruit specific chromatin-associated and epigenetic proteins that will determine whether or not the DNA is accessible for transcription, repair, and replication.^54,62,66 Also, this code can be read by transcription factors that will ultimately initiate gene expression activity.⁶⁷ For example, histone deacetylation and methylation of certain lysine residues can recruit heterochromatin-associated proteins, which on their turn interact with DNA methyltransferases. As a result, histone modifications can result in DNA methylation and through interacting with the docked heterochromatin-associated proteins favor the condensation of methylated DNA into heterochromatin.^68,69 The opposite can happen when HATs are recruited and heterochromatin is transitioned into euchromatin to allow gene expression. Acetylation of certain histone tail residues forms a docking site for other proteins that are involved in chromatin unwrapping. Certain attracted chromatin adapters possess HAT abilities that will add more negatively charged acetyl group to the histone tails. In this way, the histone tail’s positive charge is neutralized, subsequently inducing the relaxation of the chromatin structure and facilitates the binding of the transcriptional machinery. The highly dynamic nature of histone modifications means that it is regarded as a more transient cellular modification that could quickly promote or silence gene expression in response to changes in environmental stimuli, in contrast to DNA methylation, which is regarded as more stable.^56,70

3.2. DNA methylation

DNA methylation is the process in which methyl groups are added to cytosine at the 5’ carbon position of the pyrimidine ring by DNA methyltransferases (DNMT). In mammals, methylation occurs predominantly on cytosines that are followed be a guanosine (CpG sites).⁷¹ DNMT3A and DNMT3B are responsible for de novo methylation, which occurs in response to environmental cues and is an important mechanism during development and cellular differentiation.⁷² These DNMTs are also responsible for the DNA methylation after the cell has fully differentiated. In order to maintain this cellular state, DNA methylation patterns are copied from the mother strand to the daughter strand during cell division by DNMT1. In most cases DNA methylation leads to gene silencing, as the added methyl groups structurally interfere with the binding capacity of transcription factors, thus preventing the initiation of gene transcription.⁷³ However, under certain conditions DNA methylation can lead to gene activation. In the case of the human telomerase reverse transcriptase gene, hypomethylation was associated with gene silencing. In addition to DNA methylation, DNA hydroxymethylation, wherein oxidation of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC) is catalyzed by ten-eleven-translocation (TET) proteins, has been identified in recent years as an important epigenetic process. In contrast to DNA methylation, hydroxymethylation often colocalizes with euchromatin and thus is associated with enhanced gene expression.^74–77 5hmC is also an intermediary step in the demethylation process and has been shown to be involved in a myriad of other processes such as cellular differentiation, neuronal development, aging, and DNA strand separation (Figure 2B).^78–82 Just like the ‘histone code’, the location and amount of methylated CpG’s can have profound effects on gene expression and chromatin structure. Methyl groups can serve as functional docking sites for proteins with a methyl-CpG binding domain. The ‘methylation code’ determines the outcome of these proteins docking. The proteins can recruit other transcriptional co-repressors such as HDACs (and initiate chromatin remodeling), but also facilitate binding of transcriptional co-activators.^73,83,84 For instance, methyl-CpG-binding protein 2 (MeCP2) acts as a transcriptional repressor, by recruiting histone methyltransferases that methylate histone lysine residues associated with transcriptional repression.⁸⁵ However, MeCP2 can also interact with transcriptional enhancer such as CREB1, promoting transcriptional initiation.⁸⁶ Through the recruitment of additional repressor or activator proteins, de novo DNA methylation can have a profound effect on long-term gene expression and chromatin structure. As mentioned above, histone modifications and DNA methylation are intertwined and can work together in gene silencing and heterochromatin formation by recruiting histone decacetylases.^87,88

3.3. Non-coding RNAs

Non-coding RNA plays an important role in cellular differentiation and development by regulating gene expression and chromatin structure.⁸⁹ There are two broad classes of non-coding RNAs (ncRNA): short and long ncRNA. Short ncRNAs are RNA strands shorter than 200 nucleotides and thus microRNA, short interference RNA and PIWI-interacting RNA are part of this group. On the other hand, long ncRNAs are longer than 200 nucleotides. ncRNA mediate histone modifications and can recruit DNA methyltransferases, inducing de novo DNA methylation and transcriptional silencing.⁹⁰

The best studied ncRNA are miRNAs, which are short (~19–25 nucleotides) non-coding RNA molecules that are generated from miRNA genes. The initial miRNA transcript is processed (via primary miRNA and precursor miRNA intermediates) into mature double stranded miRNA by the endonucleases ribonuclease 3 (Drosha) and Dicer. The guide strand from the double-stranded miRNA will be sequestered into the miRNA-induced silencing complex. This complex exerts control over the miRNA binding to its target mRNA. Perfect complementary binding between the miRNA and the target mRNA will lead to mRNA degradation, whereas incomplete binding inhibits further mRNA processing. Both mechanisms lead to the inhibition of target gene expression.⁹¹ Due to their short length, a miRNA can bind and regulate the expression of multiple target genes. In addition, each target gene can be regulated by a number of miRNAs. miRNAs are essential for the fine-tuning of gene expression and play important roles in developmental, physiological, and pathophysiological processes.⁹²

3.4. Techniques to Measure Epigenetic Modifications

Many standard biochemical and molecular biological techniques have been developed to measure the post-translational modifications to histones (acetylation, methylation, phosphorylation, ADP ribosylation) or the differential methylation of DNA (5mC, 5hmC, 5fC, 5caC) (Figure 2) that form the basis of epigenetic changes to the genome. While we refer the reader to recent reviews focused on the different methodologies for detecting epigenetic modifications to histones^93,94, DNA⁹⁵, or non-coding RNA⁹⁶, we will briefly provide a summary of the common techniques with a comparison of the strengths and weaknesses of each assay.

Targeted analysis of histone modifications, which will measure the amount or type of post-translational modification, can be accomplished with biochemical techniques, such as mass spectrometry or enzymatic assays, or through standard molecular biological techniques, such as western blot or immunofluorescence. The gold-standard for measuring histone-DNA interactions is chromatin immunoprecipitation (ChIP) that is used to provide insight about how the histone modification modifies the epigenetic regulation of a transcript. For well characterized histone modifications (acetylation, methylation), there are multiple vendors that provide kits and/or selective antibodies to analyze global expression within the cell or tissue of interest. A weakness of using the commercial kits is that nuclear protein extraction should be performed to enrich the sample for histones, rather than total protein extraction, which will limit the overall protein available for analysis from an individual sample and could require pooling samples across biologic replicates (when using cell culture or tissues from animal models) as typical commercial assays report a range of sensitivity from 1–500 ng for the target protein(s). A strength of the commercial kits is that the antibodies and reagents have been validated, and once a sufficient histone sample has been obtained, multiple modifications can be analyzed simultaneously using standard laboratory equipment. Alternatively, mass spectrometry may generally require a smaller amount of protein (depending on the purity of the sample) and can provide very specific information about the type of modification on the histone, but the cost of the specialized equipment typically requires a core-facility to provide the analysis service and there can be multiple purification steps to achieve a suitable sample for analysis.^97–99 There are also kits (typically colorimetric) that measure the activity of acetyltransferases, methyltransferases, deacetylases, or demethylases using 96-well plates and standard plate readers. These assays provide a relatively high throughput (24 or more samples depending on the number of technical replicates) for minimal cost (most assays are $300-$500). The same technical weaknesses occur as with other commercial kits with generating enough sample to test duplicate/triplicate replicates in the assay; however, by testing for activity of broad classes of histone modifying enzymes, these assays can be useful when evaluating novel therapeutics or for assessing potential epigenetic regulation in specific diseases, such as cancer.^100,101 Caution must also be used with interpretation of the results of the enzymatic assays as global changes in activity of the enzyme may or may not affect the expression of your gene of interest. ChIP is used to evaluate the association of histone modifications with specific DNA sequences. Typical protocols require several hundred-thousand to millions of cells, 5–50 mg of tissue, or 1–15 μg of purified chromatin, but some variants have been developed that use smaller samples. Standard ChIP protocols will cross-link proteins that are directly interacting with the DNA, and ChIP-ChIP assays can be performed to detect DNA regions with multiple bound proteins (i.e. a histone and a transcription factor); however, additional proteins that associate with the primary protein (i.e. a histone acetyltransferase that modifies the histone but does not interact directly with the DNA) may not be detected. Sequencing or PCR amplification identify the target DNA sequence that was cross-linked to the protein of interest. Bisulfite sequencing can also be used to evaluate the methylation status of the bound DNA.¹⁰²

Multiple techniques can be used to assess DNA methylation at either specific regions of interest or the whole genome, and recent reviews have provided detailed information on the strengths and weaknesses of those assays.^95,103 Briefly, to broadly analyze global methylation, there are methylation sensitive restriction enzymes (typically unable to cleave DNA when a methylated base is present) that can be used to digest DNA, and comparing sequencing information against restriction enzyme digestion that is insensitive to methylation can produce a map of the methylated sites within the sample region. An alternative strategy that can reduce costs is to enrich the sample for methylated regions by using methylcytosine antibodies or other methyl binding proteins to capture methylated DNA before analysis. For detecting single base differences in methylation, the gold standard is bisulfite sequencing that can be combined with methylation sensitive arrays, used with whole-genome sequencing, or with the more targeted approach of reduced representation bisulfite sequencing. In general, the major considerations with choosing an appropriate method is the amount of DNA within the sample (whole genome analysis methods require 100 ng – 10 μg of DNA) and the funds available to perform the assay and data interpretation (targeted assays can cost as little as $100 per sample, while whole genome analysis can exceed $5000 per sample). Additionally, specific assays are necessary to distinguish 5hmC from 5mC (such as oxidative bisulfite sequencing or TET-assisted bisulfite sequencing).¹⁰⁴

4. Epigenetic mechanisms in visceral pain

4.1. Evidence pointing towards epigenetic regulation

Neurons survive for decades and therefore exhibit a remarkably low regeneration rate. Epigenetic regulation is a mechanism through which neurons maintain their specific activities throughout their lifespan. However, neuronal epigenetic regulation can be influenced by environmental changes such as physical and/or psychological stress. Hence, considering this knowledge, environmentally induced epigenetic changes in neurons were hypothesized to be a valuable underlying mechanism in the context of pain sensitization and chronic visceral pain. This idea was supported by the following observations in animal models of adult stress. In our laboratory, we investigated the effects of (chronic) stress on colonic sensitivity. We observed that bilateral implantation of a CORT micropellet on the dorsal margin of the CeA in rats lead to the development of visceral hypersensitivity that persisted for a prolonged time either after the implanted CORT-micropellets had fully dissolved. We also observed a similar persistent visceral hypersensitivity in rats exposed to WAS for 1 hr /day for 7 days. Animal models of early life adversity also clearly illustrate this principle. In these models, rats exposed to prenatal or neonatal stress exhibit visceral hypersensitivity in adulthood.¹⁰⁵ Animal models of maternal separation, limited nesting and odor associated learning demonstrate visceral hypersensitivity in adulthood, long after the initial stressor in neonatal life has been removed.^106–108 Compared to undisturbed littermates, animals that received neonatal stress had long-lasting changes in gene expression, underlying and inducing visceral hypersensitivity. The longevity of these changes is best explained through stress-induced detrimental changes in epigenetic regulation. An overview of the epigenetic changes involved in visceral hypersensitivity is presented in Table 1. These epigenetic changes will be discussed in detail below.

Table 1.

Summary of epigenetic changes involved in visceral hypersensitivity and the effects of pharmacological treatment in rodent models.

Location	Epigenetic modification	Species	Sex	Model	Effect	Downstream gene expression	Pharmacology to reverse visceral hypersensitivity	Ref
Amygdala	DNA methylation	Rat	Male	Repeated WAS	↑GR methylation, ↓CRH methylation	↓GR, ↑CRH	HDAC inhibitor TSA i.c.v.	109
	Histone acetylation	Rat	Male	CORT implant	↓GR promotor acetylation	↓GR	HDAC inhibitor TSA intra-amygdala	110
Spinal cord	miRNA	Rat	Female	Neonatal cystitis	↑miR181a & miR181b	↓GABAAα−1	/	111
		Rat	Female	Neonatal cystitis	↑miR-92–3p	↓KCC2, ↓VGAT	miR-92 lentivirus i.t.	112
	Histone acetylation	Rat	Both	Prenatal + adult HeICS	↑H3 acetylation @BDNF promoter	↑BDNF	BDNF siRNA i.t. HAT inhibitor curcumin p.o. HAT inhibitor anacardic acid i.t.	105
		Rat	Male	Neonatal + adult TNBS	↑H3K9 & H3K12 acetylation @BDNF promoter	↑BDNF	HAT inhibitor garcinol i.t.	113
		Rat	Female	Forced swim test	↓Acetylation mGlu2/3 promoter	↓mGlu2/3	HDAC inhibitor SAHA i.t.	114
		Rat	Male	Maternal separation	↓Global H4K12 acetylation	/	HDAC inhibitor SAHA i.p.	119
DRG	miRNA	Rat	Male	TNBS	↓miR-199a	↑TRPV1	mir-199a lentivirus i.p.	120
	DNA methylation	Rat	Male	Repeated WAS	↑GR, CNR1 methylation	↓GR, ↓CNR1	DNMT1 siRNA i.t.	121
	Histone acetylation	Rat	Male	Repeated WAS	↑TRPV1 acetylation	↑TRPV1	/	121

Downward arrows indicate a decrease whereas upward arrows indicate an increase in epigenetic modifications or gene expression levels. / indicates not examined in the study. Pharmacological inhibitors were infused directly into the targeted area. DRG, dorsal root ganglia; miRNA, microRNA; WAS, water avoidance stress; CORT, corticosterone; HeICS, heterotypic intermittent chronic stress; TNBS, trinitrobenzenesulfonic acid; GR, glucocorticoid receptor; CRH, corticotropin-releasing hormone; H3, histone 3; BDNF, brain derived neurotrophic factor; H3K9, lysine 9 on histone 3; H3K12, lysine 12 on histone 3; mGlu2/3, group II metabotropic glutamate receptor; H4K12, lysine 12 on histone 4; CNR1, cannabinoid receptor 1; TRPV1, transient receptor potential cation channel subfamily V member 1; GABA_Aα−1, gamma-aminobutyric acid type A receptor alpha1 subunit; KCC2, Solute carrier family 12 member 5; VGAT, vesicular GABA transporter; HDAC, histone deacetylase; TSA, trichostatin A; HAT, histone acetyltransferase; SAHA, suberoylanilide hydroxamic acid; DNMT1, DNA methyltransferase 1; siRNA, small interfering RNA

4.2. Supraspinal epigenetic mechanisms responsible for visceral hypersensitivity

In our laboratory, we reported that DNA methylation patterns of key genes involved in the HPA axis were changed in the CeA of male Fisher rats exposed to 7 days of repeated WAS. Chronic stress increased the methylation of several CpG sites in the 1₇ promoter of GR, whereas the methylation at CpG dinucleotide 6 of CRH was decreased. These changes in DNA methylation were associated with concomitant decreases in GR and increases in CRH gene expression in the CeA of these animals, and resulted in heightened responses to visceral stimuli.¹⁰⁹ As GR signaling can directly inhibit CRH expression, the decreased GR expression might have been directly responsible for the increased CRH expression. Interestingly, direct intracerebroventricular infusion of the HDAC inhibitor, trichostatin A (TSA), directly preceding WAS, reversed the colonic hypersensitivity indicating that central changes in histone acetylation were involved in stress-induced colonic hypersensitivity.¹⁰⁹ In a subsequent study from our team, in rats with elevated amygdala CORT direct infusion of HDAC inhibitors into the CeA of male rats reversed colonic hypersensitivity. On a molecular level, CORT implantation reduced H3K9 acetylation at the GR promotor level. This deacetylation may have been caused by Sirtuin-6, a HDAC that was bound on the histones at the GR promoter. Sirtuin-6 may have decreased histone acetylation and consequently induced gene silencing. Not surprisingly, TSA infusion prevented Sirtuin-6 binding to the GR histones, restored GR expression and prevented visceral hypersensitivity.¹¹⁰ Both studies from our laboratory illustrate the importance of epigenetic mechanisms in the CeA-mediated induction of visceral pain. Interestingly, infusions of a HDAC inhibitor reversed both DNA methylation and histone deacetylation. This may imply that histone deacetylation precedes changes in DNA methylation in the CeA. The number of acetyl groups on the histone tails determines which other epigenetic regulators are attracted to the DNA. Hence, fewer acetylated residues may favor DNMTs. To date, no studies have been published that investigated miRNA-mediated regulation of GR and CRH in the CeA of stressed animals and their potential influence on visceral pain. These studies represent our future area of investigation.

4.3. Epigenetic mechanisms in the spinal cord of visceral hypersensitive animals

An example of bottom-up sensitization is the animal model for neonatal cystitis in which neonates receive an intravesicular injection of zymosan in the bladder. Often this neonatal challenge is combined with a second challenge in adulthood to trigger an increased response. Using this experimental model upregulation of miR-181a and miR-181b in the spinal cord has been observed.¹¹¹ Both miRNAs have multiple complementary binding sites in the 3’ UTR region of GABA_A receptors and induced long-term downregulation of GABA_Aa-1 in the adult animals. As GABA_A receptors are important inhibitors of the pain signaling pathways, spinal inhibition of GABAergic neurotransmission may have unmasked excitatory pathways that resulted in visceral hypersensitivity.¹¹¹ In a subsequent study of the same group, it was reported that other miRNAs, affecting different components of the GABA inhibitory pathway, were also upregulated after neonatally induced cystitis (and adult re-challenge). For instance, the upregulation of miR-92–3p caused the downregulation of KCC2 and VGAT in the lumbosacral spinal cord. Loss of these two ion transport channels interfered with GABA-mediated inhibition of pain signaling.¹¹² Administration of a miR-92–3p sponge (a miRNA inhibitor) to the spinal cord of these animals, reversed visceral hypersensitivity, highlighting the importance of miR-mediated mechanisms.¹¹² Detrimental effects of prenatal, neonatal and adult stress (top-down mechanism) on spinal cord neurons have been described by other investigators. When pregnant Sprague-Dawley rats underwent heterotypic intermittent chronic stress (HeICS), offspring were prone to visceral hypersensitivity after adult re-challenge. Visceral hypersensitivity in these animals was mediated by the increased pro-nociceptive BDNF expression in the spinal cord. The combination of prenatal HeICS and adult re-challenge, decreased HDAC1 binding to the histones at the BDNF promoter, while at the same time increasing H3 acetylation levels and RNA Pol II binding. Hence, BDNF expression in the spinal cord of these animals was upregulated, resulting in increased visceral nociception. Subsequent administration of BDNF antagonists or the HAT inhibitor curcumin suppressed BDNF upregulation in the spinal cord and decreased visceral hypersensitivity.¹⁰⁵ The importance of the pro-nociceptive BDNF pathway was further investigated in the neonatal induced cystitis/adult re-challenge model. In these animals, increases in BDNF were observed in the spinal cord and in the DRG projecting to the colon. On a molecular level, the HAT CREB binding protein was recruited to the BDNF promotor, increasing H3K9 and H3K12 acetylation levels, providing a docking opportunity for RNA Pol II and inducing subsequent BDNF expression. As in the previously discussed model, spinal administration of the HAT inhibitor garcinol blocked the upregulation of BDNF and prevented colonic hypersensitivity in these animals.¹¹³ Cao et al. showed that a forced swim test (a model of subchronic adult stress) can induce colonic hypersensitivity in adult rats. However, pretreating these rats with the HDAC inhibitor suberoylanilide hydroxamic acid (SAHA or vorinistat) increased the expression of mGlu2/3 receptors. These receptors have analgesic effects in the spinal cord and their upregulation attenuated colonic hypersensitivity.¹¹⁴ Interestingly, the mGlu2 receptor is also important in the estrogen-induced visceral hypersensitivity. 17b-estradiol-induced colonic hypersensitivity was reversed by SAHA administration, due to increases in histone acetylation at the mGlu2 promoter.¹¹⁵

Early life adversity is associated with the development of visceral hypersensitivity. Evidence stems from epidemiological studies in humans and animal models that early life adversity is a risk factor for the development of chronic visceral hypersensitivity.^116,117 A widely used animal model for early life adversity is maternal separation, as it can trigger the development of gastrointestinal disorders and psychological disorders.¹¹⁸ The profound influence of maternal separation on the epigenetic mechanisms in the spinal cord was illustrated by Moloney at al. (2015). Maternal separation of Sprague-Dawley rats lead to visceral hypersensitivity and a marked global decrease in H4K12 acetylation in the spinal cord with the administration of the HDAC inhibitor SAHA in adult animals normalizing acetylation levels and ameliorated colonic hypersensitivity.¹¹⁹

4.4. Epigenetic mechanisms in the dorsal root ganglia of visceral hypersensitive animals

Intracolonic infusion of trinitrobenzenesulfonic acid (TNBS) can be used for bottom-up sensitization and thus cause colonic hypersensitivity in adult rats both during and following recovery from the acute colonic inflammation. Zhou et al. showed that in TNBS-induced visceral hypersensitivity, TRPV1 was upregulated in the colon and DRG of these animals. TRPV1 is a non-selective ligand-gated cation channel that is expressed on peripheral primary afferent sensory neurons. TNBS induced a decrease in colonic miR-199a, a regulator of TRPV1 expression, thus upregulating TRPV1 and increasing colonic sensitivity. These effects were reversed by intraperitoneal injection of lentiviral miR-199a precursors in TNBS-treated animals. miR-199a decreased TRPV1 expression in the colon and DRG with a concomitant increase in colonic response thresholds.¹²⁰

Chronic adult stress is also known to induce epigenetic changes (contributing to sensitization) of DRG neurons. Exposure to repetitive WAS increased CpG methylation at the GR promoter in nociceptive neuronal cell bodies within L6-S2 DRG, consequently decreasing GR mRNA and protein expression.¹²¹ The decrease in GR had profound effects on the anti-nociceptive properties of the endocannabinoid system, through the concomitant decrease in CNR1 expression, as GR acts as one of its transcriptional activators.¹²¹ The epigenetic effects of chronic stress were not limited to increases in DNA methylation at the GR promoter. Chronic stress also increased methylation at the Cnr1 promoter and histone acetylation at the TRPV1 promoter.¹²¹ The latter mechanism was mediated by the HAT EP300, that was bound to the TRPV1 promoter causing increased H3 acetylation and consequently upregulating TRPV1 expression. The simultaneous decrease of GR/CNR1, which has TRPV1-inhibitory properties, and increase in TRPV1 contributed to colonic hypersensitivity.¹²¹ The intrathecal administration of DNMT1-siRNA prevented colonic hypersensitivity by reversing the increases in DNA methylation at the GR and CNR1 promoter levels, but had no effect on TRPV1 methylation or acetylation levels. However, as GR expression was restored, GR could prevent the upregulation of EP300 and thus prevent downstream TRPV1 promoter H3 acetylation. Interestingly, these changes in anti-/pro-nociceptive gene expression were only observed in L6-S2 DRG that innervate the pelvic organs, but were absent in the L4-L5 DRG that innervate the lower extremities.¹²¹

4.5. Evidence for epigenetic changes in patients suffering from visceral pain

The number of studies investigating epigenetic changes in patients with visceral pain is limited. However, some studies have indicated that these mechanisms may play a role in diseases that are characterized by visceral pain, such as irritable bowel syndrome or interstitial cystitis. When comparing the miRNA expression profiles between colon biopsies from patients suffering from IBS-D with visceral pain and healthy controls, Zhou et al. reported a decrease in miR-199 in these patients, which correlated with TRPV1 expression and visceral pain scores.¹²⁰ In a different study, it was found that miR-29a is upregulated in these patients.¹²² Comparing the genome-wide methylation patterns of peripheral blood mononuclear blood cells from IBS patients and healthy controls, revealed that 133 positions were differentially methylated. These affected genes were associated with the glutathione metabolism related to oxidative stress and neuropeptide hormone activity.¹²³ A genome-wide methylation study on voided urine from patients with interstitial cystitis revealed genes downstream of the MAPK pathways were differentially methylated when compared to healthy controls.¹²⁴ These studies highlight the epigenetic changes in the target tissues of patients. Changes in miRNA expression or DNA methylation pattern in these tissues can be used as valuable biomarkers for identifying patients at risk for functional visceral pain disorders that typically lack a single diagnostic test. Although an investigation of the epigenetic changes in the CNS of patients with chronic visceral pain is currently impossible we speculate that the epigenetic mechanisms elucidated in animal models may aid in our overall understanding of the underlying epigenetic reprogramming leading to chronic visceral pain in patients.

5. Epigenetic treatment approach for visceral pain

Although epigenetic marks are long-lasting and persistent in nature, they can be further modified or even reversed at a later time. These epigenetic marks are catalyzed and maintained by epigenetic modifier enzymes. In principle, this reversible nature of the epigenetic marks and the potential for the inhibition of these specific epigenetic enzymes, holds potential for the development of epigenetic therapeutics of the treatment of visceral pain.¹²⁵ From the preclinical models discussed above, we are starting to understand the basic epigenetic mechanisms that underlie chronic visceral pain, which is critical for the development of drugs acting to modulate epigenetic reprogramming.¹²⁶ In contrast, our knowledge of the therapeutic potential of the epigenetic effects of classic and new drugs on chronic visceral pain is still in its infancy. Classical analgesia such as non-steroidal anti-inflammatory drugs (NSAIDs) and opioids have epigenetic-altering properties, but these drugs are seldom used for this purpose. Aspirin, celecoxib, and sulindac are known to modulate miRNAs, suppress or reverse DNA methylation, and increase histone acetylation of several growth genes in human gastric mucosa, rat colon, and cultured human carcinoma cells.¹²⁷ As a consequence, long-term administration of these COX-inhibitors could have therapeutically relevant effects on the epigenome of patients with chronic visceral pain. However, these reports were contradicted by one study that did not find any changes in global DNA methylation in blood cells of women on long-term NSAID treatment.¹²⁸ To date, we know little about the effects if NSAIDs on selective DNA methylation or other epigenetic markers of specific pain-related genes. Recent data suggests that the mechanisms through which opioids induce opioid tolerance, addiction and hyperalgesia involve miRNA dysregulation and histone and DNA methylation of specific opioid receptor genes.^129–131 However, due to their undesirable side effects of opioids it is unlikely that the ability of opioids to affect epigenetic programming will be employed to treat pain.

As a proof of concept, we and others have shown the beneficial influence of HDAC inhibition in the central nervous system on visceral hypersensitivity. TSA and SAHA infusions in the brain or spinal cord reversed psychological stress-induced visceral hypersensitivity. In one of our studies, we even managed to identify an HDAC involved in chronic visceral pain. TSA administration inhibited the HDAC Sirtuin-6, among other HDACs, and prevented WAS-induced visceral hypersensitivity. As there exist 5 classes of HDACs, more research is needed to elucidate which other HDACs are involved in the epigenetic pathology of chronic visceral pain. In the aforementioned pre-clinical studies, pharmacological modulation of histone acetylation and DNA methylation had profound effects on nociceptive responses. HAT/HDAC and DNMT inhibitors have obtained FDA approval and are currently being used in the clinic for the treatment of various malignancies or have advanced sufficiently in clinical trials to be considered as potential treatment options.^132–135 In contrast, to date there are no miRNA modulating drug therapies that have obtained FDA approval. At this point miRNA therapies are entering the first stages of clinical trials.^136,137 Hence, we will focus on the potential use of clinically available epigenetic drugs. Valproic acid, vorinostat and entinostat are HDAC inhibitors approved for clinical treatments. To date, most clinical trials focus on the treatment potential in cancer, and no clinical studies on the efficacy of HDAC inhibitors for modulating visceral pain have been conducted. However, there are indications of their usefulness in pain modulation.^138,139 For instance, oral administration of the HDAC inhibitor givinostat reduced pain symptoms of systemic-onset juvenile idiopathic arthritis.¹⁴⁰ Moreover, the intravenous administration of the HDAC inhibitor romidepsin improved bone pain in certain patient populations.¹⁴¹ This evidence supports the case that the effects of HDAC inhibitors could be investigated in chronic visceral pain. Furthermore, valproic acid is currently available for treating neurological conditions such as epilepsy and bipolar disorder. 5-azacytidine and decitabine are clinically approved DNMT inhibitors for the treatment of certain cancers.¹⁴² Both drugs are analogs to the nucleoside cytidine and are incorporated into the DNA in order to exert their effects on DNMTs. Once incorporated in the DNA, 5-azacytidine and decitabine respectively reversible and irreversibly bind to DNMTs, leading to the depletion of these enzymes and inducing DNA hypomethylation.^143,144 Using these two DNMT has several drawbacks. As cell division is the most opportune and efficient time point for these drugs to get incorporated in the DNA, these drugs work well in (rapidly) diving cells.^145,146 5-azacytidine can also be incorporated in RNA while decitabine activity is S-phase specific. Moreover, decitabine is rapidly degraded by cytosine deaminase and does not bind to proteins and is thus rapidly excreted from the body.¹⁴⁷ Due to these problems, these drugs require high dosing to achieve therapeutic levels for malignancy treatment and even at these levels, the drugs induce unwanted side effects,¹⁴⁸ and questions remain whether a non-toxic dose is even possible.¹⁴⁹

The availability of epigenetic drugs may open up possibilities to use these drugs for the treatment of chronic visceral pain. However, targeting these pain-related pathways in the central nervous system faces a few unique challenges. First, little is known about the efficacy, especially from the DNMT inhibitors, in slowly or non-dividing cells such as neurons. New inhibitors are being developed specifically for non-dividing cells, which would render them interesting candidates for the modulation of chronic visceral pain. Second, the current generation of epigenetic drugs target the whole epigenome, don’t discriminate between particular HATs/HDACs/DNMTs and don’t distinguish between physiological and pathological epigenetic marks.¹⁵⁰ New drugs with higher cellular or protein specificity are being developed and/or tested. For instance, class I and class II HDAC inhibitors have been identified as valid anticancer targets and are being tested in clinical studies.¹⁵¹ However, the potential uncoordinated changes in a number of signaling cascades can induce severe, unpredictable and long-lasting epigenetic dysregulation. Questions remain whether lowering the administered dose could sufficiently reverse subtle pathological DNA methylation or histones acetylation levels, with only minor interference in physiologically relevant epigenetic processes. In this light, it is possible that the dose of the HDAC inhibitor givinostat required to treat systemic-onset juvenile idiopathic arthritis is lower than the dose required to treat chronic myeloproliferative neoplasms or multiple myeloma.^152,153 This might indicate that in certain cases non-transformed cells are more susceptible to epigenetic treatment. However, it is unclear whether these working concentrations in the periphery would also have an effect in the central nervous system, which brings us to the third hurdle to overcome. In order to modulate epigenetic pathways in the central nervous system these drugs will need to be able to cross the blood-brain or blood-spinal cord barrier. Certain viral vectors, nanoparticles, exosomes and various other techniques are being used to deliver drugs directly across the blood-brain and blood-spinal cord barrier or prolong their time at these barriers to facilitate their uptake in the central nervous system through other uptake methods, for instance receptor-mediated uptake or passive diffusion.^154–160 Just like other drugs, epigenetic drugs could be packaged in lipophilic substances or hydrophilic agents and potentially cross the blood-brain/blood-spinal cord barrier. Despite these improvements in delivery methods, high doses and/or long-term administration of the drug will be required for the drug to reach the central nervous system and exert its effects. Hence, it cannot be excluded that these drugs induce systemic (side) effects as the current generation of epigenetic drugs wasn’t designed to target specific cell types. Fourth, when these drugs have crossed the blood brain barrier, they’ll need to reach the specific areas involved in chronic visceral pain. It is unlikely that in the near future drugs will target only a specific region of the brain, without affecting other regions. However, improvements have been made in brain implants and brain surgery, so it would be possible to directly deliver epigenetic drugs to the area of interest, increasing treatment efficacy and diminishing the potential side effects. Epigenetic mechanisms are also involved in the peripheral pathways of chronic visceral pain. Targeting the central pathways with the current available drugs is extremely challenging. However, targeting the peripheral components of the pain pathway would also be beneficial for patients. Restoration of the epigenetic marks in the neurons at the level of the visceral organs could potentially ameliorate allodynia or pain perception by reducing the input signaling in the pain pathway.

In summary, while still in the early days of discovery, new therapeutics targeting epigenetic processes have the potential to revolutionize the treatment of chronic visceral pain disorders. Chronic visceral pain is a significant health care burden due to the poor efficacy of current therapeutics, in part due to the lack of knowledge of the underlying mechanisms. Epigenetic processes regulating gene expression could be responsible for the transition from acute to chronic visceral pain with differences in methylation, acetylation, or miRNA expression occurring from bottom-up or top-down mechanisms. Specific histone tail acetylation, DNA methylation and miRNA expression patterns can cause long-term changes in expression of ion channels and/or receptors that alter neuronal excitability, promoting chronic visceral pain. However, our understanding of the epigenetic component in visceral pain is still advancing at a rapid pace. There are numerous histone tail modifications, such as histone tail methylation, whose role in visceral pain have yet to be investigated. With ever improving research techniques, it’s only a matter of time before the role of other epigenetic mechanisms will be revealed. Environmental cues, such as childhood or adult stressors or colonic inflammation promote epigenetic modification of peripheral, spinal, and/or central neuronal pathways to induce hypersensitivity, as demonstrated by multiple preclinical models. A growing body of clinical evidence has also demonstrated changes in DNA methylation patterns and miRNA expression associated with chronic visceral pain. Interestingly, there is no uniform relationship between environmental factors and epigenetic outcomes, in fact here may be a dual role of histone acetylation and deacetylation in the development or reduction of visceral pain. For instance, early life stress increases global histone tail acetylation, whereas adult stress has the opposite effect. How these global changes in acetylation specifically affect genes relevant for visceral pain also remains to be studied. A context-specific mechanism would require two completely different treatment therapies. Therefore, personalized pharmacological treatment will be necessary. A person with visceral pain, originating from adult stress, could be helped with HDAC inhibitors, whereas a person with visceral pain, originating from pre- or neonatal stress could only be helped with HAT inhibitors. Although the current generation of drugs lacks sufficient tissue or pathway specificity, the development of new classes of selective compounds and/or targeted delivery systems will open the door to new therapies targeting epigenetic processes to treat chronic visceral pain.

Abbreviations

5caC

5-carboxylcytosine

5fC

5-formylcytosine

5hmC

5-hydroxymethylcytosine

5mC

5-methylcyosine

ACTH

adrenocorticotropic hormone

BDNF

brain derived neurotrophic factor

BPS

bladder pain syndrome

CAPS

centrally mediated abdominal pain

CeA

central nucleus of the amygdala

CGRP

calcitonin gene related peptide

ChIP

chromatin immunoprecipitation

CNR1

cannabinoid receptor 1

CNS

central nervous system

CORT

cortisol or corticosterone

CpG

cytosine-guanosine dinucleotides

CREB

cAMP response element binding protein

CRH

corticotropin-releasing hormone

CRH₁

CRH receptor type 1

DNMT

DNA methyltransferase

DRG

dorsal root ganglia

GABA_Aα−1

gamma-aminobutyric acid type A receptor alpha1 subunit

GERD

gastroesophageal reflux disease

GR

glucocorticoid receptor

H

histone

H3

histone 3

H3K12

histone 3, lysine 12

H3K9

histone 3, lysine 9

H4K12

histone 4, lysine 12

HAT

histone acetyltransferase

HDAC

histone deacetylase

HeICS

heterotypic intermittent chronic stress

HPA

hypothalamus-pituitary-adrenal

IBD

inflammatory bowel disease

IBS

irritable bowel syndrome

IC

interstitial cystitis

K

lysine

KCC2

Solute carrier family 12 member 5

MeCP2

methyl-CpG binding protein 2

mGlu2/3

group II metabotropic glutamate receptor

miRNA

microRNA

MR

mineralocorticoid receptor

mRNA

messenger RNA

NBS

narcotic bowel syndrome

ncRNA

non-coding RNA

NO

nitric oxide

NSAID

non-steroidal anti-inflammatory

Pol

polymerase

PVN

paraventricular nucleus of the hypothalamus

SAHA

suberoylanilide hydroxamic acid

siRNA

small interfering RNA

TET

ten-eleven translocation protein

TNBS

trinitrobenzenesulfonic acid

TRPV1

transient receptor potential cation channel subfamily V member 1

TSA

trichostatin A

VGAT

vesicular GABA transporter

WAS

water avoidance stress