DNA methylation

Gene regulatory elements (or promoters) contain regions concentrated with cytosines followed by guanines, which are collectively known as CpG sites. CpG sites are more frequently found at promotor regions compared to other sites of the human genome. These areas of increased CpG sites are collectively known as CpG islands or clusters. The cytosine bases of CpG sites are subject to methylation by enzymes known as DNA methyltransferases (DNMTs). The addition of methyl groups to promoter-region CpG islands typically decreases gene expression, while the removal of methyl groups from these regions tends to increase gene expression ( ). Conversely, intragenic (exonic or intronic) CpG methylation commonly results in genetic overexpression ( ).

Alterations to multiple CpG sites have been implicated in the pathogenesis of AD. The major methylation changes in AD are summarized in Table 17.1 . reported a role of differential CpG methylation patterns in AD by showing reduced expression of DNMT-1 in the peripheral blood mononuclear cells (PBMCs) of AD study participants compared to healthy controls. The frequency of these epigenetic findings across a wide range of study participants and ethnicities is unknown and will require additional subsequent studies to tease out whether these changes are truly causative or merely secondary changes in response to an activated, dysregulated immune response. Ongoing investigations and similar clinical studies are underway and may shed light on the impact of CpG methylation and demethylation on the onset and severity of AD.

microRNAs

miRNAs are short, single-stranded, noncoding RNA molecules that are 18 to 22 nucleotides in length and provide a novel mechanism of posttranscriptional control of genes within cells and tissues. miRNAs are transcribed by RNA polymerase II, undergo processing, and then become associated with protein complexes that bind to 3′ untranslated regions of target messenger RNA. The mechanism by which miRNAs disrupt translation or promote degradation of protein transcripts is well established, thereby confirming their importance in epigenetics ( ). miRNAs have also been shown to participate in RNA activation, whereby miRNAs directly bind promoter elements to positively regulate gene transcription ( ).

Approximately 2500 miRNAs have been characterized in humans and can influence the expression of 30% of all genes. miRNAs play a critical role in cellular development, proliferation, apoptosis, immune regulation, and the stress response to trauma or infections ( ). Dysregulated miRNAs have been associated with most human diseases and dermatologic conditions, including cutaneous malignancies, AD, SLE, and psoriasis ( ). Comparisons of miRNA expression profiles in skin and body fluids of AD participants versus healthy volunteers have implicated miRNAs in the pathogenesis of this inflammatory condition ( Table 17.2 ). Importantly, sequencing and gene expression algorithms can easily be used to predict and identify the gene target(s) of specific dysregulated miRNAs in AD as shown in Table 17.2 . These identified targets shed light on the immune perturbations driving the development of AD and represent an opportunity for novel therapeutic strategies via selective targeting of miRNAs as a means of restoring normal immune function. A number of clinical trials testing the efficacy of specific miRNAs for the treatment of a dermatologic condition are already underway in inflammatory conditions, such as cutaneous T-cell lymphoma (CTCL). Finally, miRNAs can be shared between cells and shuttled to distant body sites in the bloodstream via membrane-bound microvesicles called exosomes.

Posttranslational histone modifications

DNA associates with an octamer of proteins, known as histones, to form densely packed chromatin. Transcription of genes is determined by the degree of chromatin packing. When chromatin is open, transcriptional machinery can easily access DNA and initiate gene transcription. A greater degree of compaction, on the other hand, results in diminished gene expression. The degree of chromatic compaction is dictated by the posttranslational modification of the amino acids of the N-terminal tails of histones. Common modifications include methylation, acetylation, phosphorylation, and ubiquitination. Acetylation promotes chromatin opening. Histone acetyltransferase and histone deacetylases (HDAC) catalyze the addition or removal of acetyl groups from histones respectively. Histone methyltransferases catalyze histone methylation, which can either open or condense chromatin depending on the specific amino acid methylated and the number of methyl groups added ( ). To our knowledge, studies of histone modifications in allergic diseases have been limited to studies looking at the biologic role of HDAC and HDAC blockade in mice and in vitro studies. It is unclear if posttranslational histone modifications contribute to AD pathogenesis or whether HDAC blockade could result in the reversal of clinical symptoms affected by this condition.

Predisposing genetic mutations can act via epigenetic mechanisms

Genomewide association studies have already linked a number of genetic mutations and single nucleotide polymorphisms (SNPs) to AD pathogenesis. The direct link between disease-associated polymorphisms and complex, chronic diseases such as psoriasis and AD has been elusive. However, recent research in AD has demonstrated that these genetic alterations may have consequences on an epigenetic level via their contribution to differential DNA methylation patterns. For example, demonstrated that the T allele of SNP rs612529 on the promoter of the VSTM1 gene promotes demethylation of VSTM1 in monocytes. VSTM1 encodes a protein named SIRL-1, which acts as an inhibitor of innate immunity. As a result of the hypomethylation of VSTM1 , SIRL-1 expression increases and lowers the risk for AD in individuals who carry the rs612529 T allele. Interestingly, increased risk of AD was associated with the C allele, which is not associated with this demethylation event. also demonstrated that two SNPs (rs2299007, rs11740584) in the KIF3A gene generate new methylated CpG sites in nasal airway cells that result in decreased KIF3A expression. KIF3A is required for skin barrier homeostasis, and loss-of-function mutations in this gene in a murine model demonstrated increased transepidermal water loss, increased epidermal thickness, and dysregulated filaggrin and claudin-1 protein expression. This research provides a nice link between KIF3A polymorphisms and the clinical features of AD. Similar translational research that provides an epigenetic mechanism by which specific SNPs can contribute to skin barrier dysfunction and/or a dysregulated immune response would bridge a major gap in the study of AD and other inflammatory skin diseases.

Epigenetic alterations may disrupt epidermal integrity

Preservation of the skin’s allergen, microbial, and irritant barrier function requires adequate expression of epidermal structural and cell-cell adhesion proteins ( ). Filaggrin is an intermediate filament-associated protein expressed in the stratum corneum. Absent or decreased expression of filaggrin has been implicated in the pathogenesis of AD ( ). Filaggrin ( FLG ) gene loss-of-function mutations confer the strongest genetic risk of AD, though most patients with AD do not carry this mutation. Of the European population, 10% carry a null mutation in FLG , which results in mild ichthyosis vulgaris and a threefold increase in the risk of AD ( ).

Filaggrin has been the subject of many DNA methylation investigations. reported that hypermethylation of an intragenic CpG methylation site of FLG in whole blood was associated with a higher risk of AD in a birth cohort. However, the study did not measure or correlate FLG methylation with its expression level, and it is unclear whether the methylation status of FLG changes as this AD cohort ages. Both are important research questions to be answered. In contrast, reported hypomethylation of a FLG promoter CpG site in buccal cells of AD participants compared to healthy controls. Filaggrin expression, however, did not significantly differ between the two groups. did not discover methylation differences in the FLG promoter between AD participants and healthy controls in skin and blood samples. The variation in methylation status between these filaggrin studies may also be due in part to the heterogeneous participant cohorts and our current inability to tease out clinical subtypes of AD (e.g., early vs. late onset, mild vs. severe disease, flexural vs. nonflexural involvement).

CpG methylation differences have been observed in other genes involved in epidermal structural integrity. performed a large-scale screen of more than 450,000 CpG sites associated with 24,000 genes on the cord blood of an AD birth cohort. This team reported enriched methylation patterns of multiple genes involved in cell-to-cell adhesion, such as cadherins and tight junctions, in AD participants compared to healthy controls. Loss-of-function mutations in both cadherins and tight junctions have already been implicated in AD pathogenesis, so this finding is consistent with previous knowledge about the disease ( ). Downregulation of tight junction proteins via methylation results in increased paracellular permeability and enhanced keratinocyte proliferation, which are key features of AD. Diminished cadherin expression would also contribute to epidermal edema (spongiosis) as noted in histologic studies of AD lesional skin ( ). These early results suggest that the methylation status of AD-related genes may impact skin and/or alter epidermal structural integrity.

Epigenetic alterations regulate innate immune dysfunction in AD

Human β-defensins (hBD) make up a major class of antimicrobial peptides (AMP) that contribute to the innate immune defense of the skin. Three β-defensins (hBD-1, hBD-2, and hBD-3) are expressed in the skin. Inflammatory cytokines (e.g., interleukin-17 [IL17]), infections, and trauma/injury induce expression of hBD-2 and hBD-3, whereas hBD-1 is constitutively expressed ( ). Some studies suggest that hBD-1 is upregulated in AD skin lesions ( ), while others report that its expression is diminished ( ). This inconsistency has been shown for multiple AMP, though the vast majority of studies show a global reduction in AMP compared to global increases in psoriasis. This reduction in AD is largely mediated by interferon-gamma (IFN-γ) with synergistic effects from Th2-related cytokines, such as IL4 and IL13. Reduction in AMP, such as β-defensins, in response to Th2 and IFN-related genes may explain why Staphylococcus species colonize the skin in ~90% AD patients ( ).

Diminished AMP levels in AD results in an increased risk for bacterial and viral infections, such as gram-positive bacteria and herpes or vaccinia viruses ( ). reported hypermethylation of promoter CpG sites of DEFB1 , the gene encoding hBD1, in AD lesional skin samples compared to nonlesional and healthy control samples. Topical application of a DNMT inhibitor increased hBD-1 expression in normal human epidermal keratinocytes, which suggests that CpG hypermethylation could contribute to hBD-1 downregulation in skin and possibly AD lesions. The clinical testing of such inhibitors would be of interest in terms of its clinical efficacy on inflammatory skin lesions and/or the potential of reducing skin infections in this patient population.

IFN-γ and NF-κB intracellular signaling activity are two other hallmarks of the innate immune response. Their activation is observed in AD lesional skin ( ). miRNAs play an important regulatory role of IFN-γ and NF-κB intracellular signaling activity. miR-146a is a miRNA known for its general antiinflammatory activity, in contrast to the proinflammatory effects of miR-155 ( ). Though dependent on NF-κB for its expression, miR-146a inhibits the NF-κB pathway by targeting IL1 receptor-associated kinase 1 (IRAK1) and tumor necrosis factor receptor-associated factor 6 (TRAF6). miR-146a is overexpressed in AD keratinocytes and targets the caspase recruitment domain-containing protein 10 (CARD10), IRAK1, and CCL5, which are known mediators of the proinflammatory NF-κB and IFN-γ cascades. The likely role of miR-146a in AD is the loss of its antiinflammatory effects via inhibition of the innate immune response ( ). Accordingly, the role of miR-155 in AD is to sustain activation of the upregulated innate immune response in lesional AD skin. miR-155 also likely promotes the secretion of Th2 cytokines (discussed in more detail later). Bacterial antigens (e.g., lipopolysaccharide and staphylococcal superantigens) and multiple environmental allergens used in patch testing have also been shown to induce miR-155 expression in inflamed skin ( ). Interestingly, a number of miRNAs have also been found in significant concentrations in the breast milk and represent another understudied area in neonatal immunology that could lead to insights about the onset of AD ( ).

NLRP2 is a member of the NALP family of proteins that acts as an immunosuppressive signal by inhibiting NF-κB. NLRP2 is upregulated in IFN-γ–stimulated hair follicle-derived keratinocytes collected from AD participants ( ). discovered promoter hypermethylation and decreased expression of NLRP2 in the cord blood of a birth cohort with early-onset AD (i.e., diagnosis of AD before age 2 years) compared with healthy controls. The authors concluded that NLRP2 expression blunts excess NF-κB activity. They also suggest that NLRP2 gene hypermethylation and decreased protein expression lead to NF-κB overactivity resulting in an increased risk of AD. Selective blockade of NF-κB signaling could therefore have several indications for the treatment of multiple inflammatory conditions such as AD given the central role for NF-κB in the innate immune response.

Epigenetic alterations environmental allergen sensitivity

Active and passive exposure to tobacco smoke is associated with an increased prevalence of AD ( ). Several studies show that tobacco smoke may act via several independent epigenetic mechanisms. associated smoking with hypermethylation and decreased expression of the PYK2-binding protein, PITPNM2, in patients with atopy (seasonal allergies, asthma, and/or AD) compared to healthy controls. The exact role of PITPNM2 in inflammatory conditions is unclear, but early studies suggest that it may be involved in neutrophil function. Tobacco smoke has also been linked to the hypermethylation of NLRP2 , discussed earlier ( ). also quantified the number of regulatory T cells (Treg) via demethylation of the TSDR (Treg-specific demethylated region) of FOXP3 in cord blood. Maternal exposure to tobacco smoke during pregnancy was associated with lower Treg numbers, which was associated with an increased risk for AD during the first year of life. Decreased numbers of Treg cells in subsets of patients with AD could explain, in part, the hyperreactivity of the skin and the atopic march observed in patients.

Vitamin D deficiency is another environmental risk factor of AD ( ). discovered that severe 25-hydroxyvitamin D deficiency in cord blood was associated with a higher risk of AD. Severe vitamin D deficiency was also associated with hypomethylation and overexpression of the MICAL3 gene in leukocytes obtained from cord blood compared to healthy controls. MICAL3 has previously been implicated in reactive oxygen species (ROS) production and several proinflammatory signals ( ). The link between tobacco exposure, AD, and increased ROS was further supported by the finding of increased expression of OGG1 , a gene strongly associated with oxidative stress. Epigenetic mechanisms as a potential link between certain environmental exposures and the proinflammatory events in AD patients is an attractive area of study given our current inability to mitigate the injurious environment effects on the skin of so many AD patients.

Th2/Th22 axis predominance is influenced by epigenetic changes

All three major epigenetic modifications have been linked to the overactivation of the Th2 axis in AD. AD lesions share a Th2/Th22 predominance with smaller contributions from Th1 and Th17 immune axes ( ). CD4 ⁺ Th cell populations play a central role in the adaptive immune response that characterizes the immunologic dysfunction observed in AD ( ). miRNAs can further dysregulate the pathogenic T-lymphocyte populations in AD. suggest that miR-155 promotes a proliferation in pathogenic T lymphocytes in AD skin. Their research demonstrates that miR-155 is highly overexpressed in lesional CD4 ⁺ T cells and targets CTLA4, an important costimulatory receptor of the T-cell receptor, which counterbalances T-cell proliferation ( ). This suggested mechanism was further supported by the observation that T-lymphocyte counts increased in response to miR-155 cell transfection experiments.

A number of other dysregulated miRNAs in AD have been linked to T-lymphocyte genes that further regulate the adaptive immune response (see Table 17.2 ). miR-151a and miR-143 are two of these miRNAs that contribute to Th2 polarization in AD. reported that miR-151a is upregulated in AD skin and targets IL12RB2 , the gene encoding a subunit of the IL12 receptor. IL12-mediated signaling promotes Th1 differentiation and T-cell proliferation ( ). miR-151a therefore inhibits the Th1 immune response as a result of its overexpression in T cells by diminishing the expression of IL2, IL12, and IFN-γ—three canonic Th1 cytokines ( ). miR-151a–mediated downregulation of IL12R may also contribute to Th2 axis skewing since Th1/Th2 axes are known to counterregulate one another ( ). suggest miR-143 targets the IL13 receptor and is decreased in AD skin lesions. IL13-mediated inflammation induces IgE synthesis, eosinophilia ( ), and downregulates the expression of epidermal barrier proteins (filaggrin, loricrin, and involucrin) in keratinocytes ( ). In healthy individuals, miR-143 expression in epidermal keratinocytes inhibits IL13-mediated repression of the epidermal barrier proteins ( ). However, this protective feature is lost in AD due to decreased expression of miR-143, which leads to increased IL13 signaling in inflamed skin.

DNA methylation has been shown to alter the expression of thymic stromal lymphopoietin (TSLP) and subunits of the IgE receptor, which contribute to Th2 polarization. TSLP is a pivotal cytokine that is highly expressed in acute and chronic eczematous lesions and promotes Th2 axis polarization ( ). Its expression is correlated with AD severity scoring and measures of epidermal barrier dysfunction, such as transepidermal water loss ( ). demonstrated that the promoter of TSLP is hypomethylated and overexpressed in AD lesional skin compared to healthy controls. Application of a DNMT inhibitor increased TSLP expression, suggesting that demethylation events contribute to TSLP overexpression ( ). Genetic polymorphisms of the FCER1 gene, which encode specific subunits of the IgE receptor, have also been implicated in AD since Th2-mediated IgE signaling is an essential component of allergic sensitization ( ). reported demethylation and overexpression of the FCER1G promoter in monocytes of AD participants compared to healthy controls. Later studies attributed this demethylation event to TSLP-activated pSTAT5 recruitment of a DNA demethylase named TET2 ( ).

Histone deacetylation is another epigenetic mechanism that can enhance Th2 polarization. The application of HDAC inhibitors in AD murine models showed improvement in skin lesion severity and reduction of IL4 production in CD4 ⁺ T cells, and promoted the generation of Tregs in draining lymph nodes ( ). Whether this finding translates in humans will be of crucial importance given the nonspecific nature and controversy about the etiology of psoriasiform skin lesions observed in mice. Current research suggests that none of the frequently used murine models for the study of AD or psoriasis accurately recapitulate these human skin diseases but rather represent another inflammatory skin disease entity such as allergic or irritant contact dermatitis. Nevertheless, the clinical availability and use of HDAC inhibitors for the treatment of human disease in patients with both AD and CTCL will offer interesting “experiments in nature” and lead to relevant hypothesis-generating observations.

Th17 signaling and psoriasiform epidermal changes in chronic AD may act through epigenetic mechanisms

Th17 population skewing via IL17 production and STAT3 activation has been shown to induce epidermal hyperplasia ( ), a key feature of both psoriasiform and chronic eczematous lesions ( ). The pathogenic role of IL17 signaling in chronic AD lesional skin (or specific population subtypes) is just now starting to be elucidated. Aside from its role in T-lymphocyte cell proliferation, miR-155 promotes a shift toward the Th17 axis. correlated miR-155 overexpression with increased Th17 cell populations, as well as RORγt and IL17 expression in plasma, lesional skin, and perilesional skin obtained from AD participants. Adding a miR-155 mimic to AD-cultured keratinocytes increased the percentage of Th17 cells, whereas the addition of a miR-155 inhibitor decreased the overall percentage of Th17 cells. Further analysis implicated suppressor of cytokine signaling 1 (SOCS1) as the link between miR-155 and Th17 axis skewing since SOCS1 expression negatively correlated with Th17 markers.

miR-10a-5p, best known for its antiproliferative properties in many cancers ( ), may also acts as a protective cellular signal for limiting epidermal hyperplasia. miR-10a-5p has been shown to be overexpressed in AD lesional skin ( ) and targets genes involved in cell cycle progression and proliferation ( ). One such target is hyaluronan synthase 3 (HAS3), an enzyme involved in hyaluronic acid synthesis that is induced by proinflammatory cytokines such as IL4, IL13, and IFN-γ ( ). This suggests that miR-10a-5p plays a regulatory role in AD by limiting keratinocyte proliferation and keratinocyte-driven inflammation. This is further supported by evidence that downregulation of miR-10a-5p is also associated with delayed wound healing ( ). However, miR-10a-5p is insufficient alone to suppress keratinocyte proliferation in AD as shown by persistent elevations in Ki-67 and HAS3 (another miR-10a-5p target) despite significant overexpression of miR-10a-5p ( ). This suggests that miR-10a-5p may act in coordination with other miRNAs and/or other cellular signals such as IL19 and IL22, which promote epidermal hyperplasia in AD skin. Selective targeting of miRNAs for the treatment of a chronic disease like AD may therefore be beneficial when used in combination with other primary therapies.

Diagnostic epigenetic biomarkers of AD

As discussed, environmental factors play a focal role in AD and have an impact on DNA methylation patterns in AD-associated genes. DNA methylation is a complex phenomenon that may vary between cell type, age of participants, skin site, and chronicity of disease state. Though many studies have correlated DNA methylation patterns with gene expression and AD risk, establishment of causation is lacking. The lack of causation is particularly evident in studies that did not validate observations with DNA methylase inhibitors or follow specific methylation changes over time. Given recent technologic advances in translational genetics that have allowed for large-scale screening of methylation patterns, the longitudinal study of DNA methylation patterns in a given cell type or tissue over time in the same patient cohort may yield important insights into the onset or progression of disease activity and serve as a useful noninvasive biomarker of disease. Therefore additional studies evaluating DNA methylation are needed to establish the impact of these epigenetic modifications on AD pathogenesis at a molecular level. The ability to screen large numbers of miRNAs and predict putative miRNA targets will lead to more informed decisions as to which miRNA-gene target pairs to further study and evaluate for particular patient cohorts or disease states. Isolating miRNAs and collecting epigenetic data from serum or plasma samples via exosomes as opposed to biopsies of skin lesions offers an even less invasive approach to studying epigenetic modifications as potential disease biomarkers.

Epigenetics as a potential therapeutic approach in AD

DNMT inhibitors and TET inhibitors are actively being studied for the treatment of myelodysplastic and hematologic malignancies ( ). There are no human studies on therapeutic modulation of methylation-modifying enzymes in AD. However, since TET2 demethylation directly impacts expression of proinflammatory genes strongly associated with AD, it may be a desirable therapeutic target for future study. miRNA antagonists or agonists offer another therapeutic potential for further study in human disease and AD. The transfection of keratinocytes with these synthetic miRNAs could be used to restore normal expression for dysregulated genes associated with AD and offer personalized therapeutic options in the future ( ).

Histone deacetylase inhibitors (HDACi) are also the subject of study in inflammatory diseases due to their ability to lower levels of proinflammatory cytokines and inhibit angiogenesis. Vorinostat, the first US Food and Drug Administration-approved HDACi, is the focus of study in preclinical and clinical trials of many cancers ( ). The effectiveness of HDACi in AD and other allergic diseases has yet to be studied in humans. The impact on Th2 cytokines has, however, been studied in a number of preclinical models of allergic disease ( Table 17.3 ) suggesting their potential benefit for the treatment of atopic disease in humans. The principal concern with the use of HDACi or other medications that alter epigenetic modifications and the expression of multiple genes is serious adverse events or unwanted biologic effects during treatment. Vorinostat, for example, is associated with ventricular arrhythmias ( ) possibly due to changes in the expression of cardiac ion channels ( ). This is perhaps the reason why use is limited to advanced, treatment-resistant cancers and will likely limit the widespread use of this drug class unless more specific HDACi with a more favorable side effect profile are developed. Additional studies elucidating the specific mechanisms of regulating these epigenetic modifications may also lead to improved, more targeted therapies that will allow for their use in dermatology in the future.

Table 17.3

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Atopy Redefined Oxidative Stress, Environmental Factors, and Pollutants Humoral Factors Therapeutic Guideline Overview Wound Care Complementary and Alternative Approaches II

Stay updated, free articles. Join our Telegram channel

Join