DNA methylation effects on Van der Woude Syndrome phenotypic variability

Introduction

Orofacial clefts are common birth defects affecting approximately 1 in 1,000 individuals, with prevalence varying by ethnicity and gender. While most forms of orofacial clefting are non-syndromic, approximately 30% are associated with other clinical signs and symptoms and are denominated syndromic cases ¹. Van der Woude Syndrome (VWS), inherited through an autosomal dominant pattern, is among the most common syndromic forms of clefting, representing 2% of all orofacial clefts with a prevalence of 1 in 34,000 live births ².

Mutations in the gene Interferon Regulatory Factor 6 (IRF6) are causal for 70% of VWS cases, while 5% of cases are caused by mutations in Grainyhead-Like Transcription Factor 3 (GRHL3) and the remaining 25% are of unknown cause. Causal mutations in IRF6 commonly occur in the highly conserved DNA-binding domain (Exons 3 & 4) or the less conserved protein-binding domain (Exons 7-9) ^3,4.

While these various established mutations in IRF6 account for the presence of the syndrome, they do not fully explain the diverse phenotypic presentations of individuals with VWS. Lower lip pits (LP), occurring on the paramedian portion of the lower lip due to notching of the lips around day 36 of embryonic development, present in 85% of individuals with VWS ^5,6. Other possible phenotypes include cleft lip with or without cleft palate (CL/P), cleft palate only (CPO) and hypodontia ⁵. Despite the autosomal dominant with high penetrance inheritance pattern, phenotype presentations for any given IRF6-causal mutation vary in multiple ways, even amongst family members⁷. This marked phenotypic discordance amongst individuals with VWS is exemplified by a 2011 study on a pair of monozygotic twins with variable phenotypic expression ⁸. One twin presented with bilateral cleft lip with cleft palate and bilateral lower lip pits, while the other twin presented with bilateral lower lip pits only. The occurrence of phenotypic discordance in these identical twins suggests a potential role for epigenetic factors as phenotypic modifiers in VWS.

One epigenetic factor that has been hypothesized as implicated in phenotypic discordance in VWS is DNA methylation (DNAm). A 2023 study by Petrin et al. reported findings of DNAm differences in individuals with VWS⁴. This paper reported methylation profiles of two pairs of monozygotic twins. The first pair was previously reported in the study that discovered IRF6 as the first causal gene for VWS, while the second pair of twins is the pair reported in the aforementioned Jobling et al. paper where both individuals had VWS with discordant phenotypes ^4,7,8. Whole genome DNAm was generated for both pairs of twins. While there were also other regions with differential methylation, Tumor Protein 63 (TP63) was the most significant hit amongst cleft related genes, showing DNAm differences between twin members. In the Kondo et al. twin set, the twin with VWS had higher methylation levels in several cpg sites located in or near TP63 than the unaffected twin (delta beta from 5% to 9%); while in the Jobling et al. twin set, the twin with lip pits only had higher methylation levels in TP63 than the twin with cleft lip and palate and lip pits (delta beta from 5% to 14%). They also observed one CpG site (cg26035071) in the promoter region of IRF6 with higher absolute methylation levels in the twin with VWS in the Kondo et al. twin set (delta beta 9%) and in the twin with the more severe phenotype in the Jobling et al. twin pair⁴ (delta beta 10%).

Studies have shown a complex biological regulatory loop between TP63 and IRF6 in palatogenesis where normal levels of IRF6 are necessary to downregulate TP63. However, the presence of an IRF6 mutation can lead to aberrant DNAm of TP63 which can modify phenotypic expression. In the Jobling twins, the higher methylation levels of TP63 in the individual with lip pits only could have been enough to downregulate TP63 during palatal fusion, allowing for effective palatal closure. Another possibility is that the higher methylation level in the promoter region of the IRF6 gene, along with the causal mutation, contribute to a distinct phenotype. These hypotheses, along with the otherwise inexplicable phenotypic discordance observed in VWS, led to the development of this project.

In this article, we evaluated DNAm differences in three regions, TP63 promoter, IRF6 enhancer, and IRF6 promoter, among individuals with IRF6 causal mutations for VWS. We chose to analyze DNAm levels in TP63 because of the DNAm differences found in the twins with VWS from Petrin et al. 2023. Also, we wanted to investigate if differential methylation in IRF6 also leads to phenotypic heterogeneity. For IRF6 we selected the IRF6 immediate promoter and enhancer, MCS9.7. Of note a 2011 study showed that expression of IRF6 is repressed by promoter methylation in individuals with squamous cell carcinoma ⁹. This finding points to a pathogenic effect led by silencing IRF6 via DNA methylation of its promoter. The current study aims to explores the specific role of DNAm of IRF6 and TP63 on the phenotypic discordance observed in VWS.

Methods

Results

Discussion

Despite mutations in IRF6 representing a well-defined genetic cause for VWS, individuals with the syndrome display variable phenotypes. Epigenetic modifiers are thought to influence the variable phenotypic expression present in affected individuals.

We observed higher levels of DNAm in the IRF6 immediate promoter with more severe phenotypes. Individuals in Groups 1-4 of the primary analysis all had some type of clefting, with or without lip pits, which resulted in higher methylation levels than individuals with lip pits only. Individuals with any type of clefting with lip pits also had higher methylation levels than individuals with any cleft only. This is thought to be due to the critical role that IRF6 plays during lip and palate development, with normal levels of IRF6 expression needed for proper elevation and fusion of the palatal shelves.

IRF6 is part of a key biological regulatory loop with TP63 which functions to help coordinate epithelial proliferation and differentiation during formation of the orofacial complex⁹. Disruption of this loop due to mutations in either gene can result in orofacial clefting. Expression of IRF6 requires normal function of TP63, as TP63 regulates IRF6 expression by directly binding to two neighboring sites within MCS9.7: Motif1 and Motif2¹⁰. A study by Fakhouri et al. (2011) showed that abolishing both Motif1 and Motif2 significantly reduced and disrupted the MCS9.7 enhancer activity, whereas abolishing either motif separately reduced but did not disrupt enhancer activity ¹¹.

Normal IRF6 is expressed in high levels in the medial edge epithelia of the developing ectoderm, where it is an essential determinant of the keratinocyte proliferation-differentiation switch during lip and palate development^12-14. High levels of IRF6 also function indirectly to achieve palatal fusion by downregulating TP63 at the time of palatal fusion. Downregulation of TP63 at this point allows the pathway to enable disintegration of the median epithelial seam (MES) through apoptosis, allowing for mesenchymal continuity of the converging midfacial processes, resulting in successful palate closure ^10,15. This negative feedback mechanism of IRF6 on TP63 was demonstrated in the mouse model, where TP63 was spirited away at the time of palatal fusion in wild type mice while IRF6 reached the pinnacle of its expression. Conversely, in mutant mice that were IRF6 (-), the MES remained positive for TP63 expression, resulting in a cleft and further indicating that the downregulation of TP63 for palatal fusion necessitates normal IRF6 function¹⁰. Mutations in IRF6 causal for VWS can disrupt normal function of IRF6, hindering this important biological regulatory loop between IRF6 and TP63 and potentially resulting in orofacial clefting. IRF6 Enhancer, MCS9.7, is the site where TP63 binds and interacts with IRF6.

We hypothesize that, though the causal mutation acts to initially weaken the IRF6 protein, methylation of the IRF6 immediate promoter possibly compounds this by diminishing an already weakened protein, leading to a more severe phenotype, such as a cleft with lip pits. This follow-up study corroborates our previous findings that differential DNAm can act as modifiers influencing the severity of the phenotype.

Knowing that epigenetic changes can impact the process of craniofacial development, we now look at when these changes must occur. Epigenetic changes are impacted by several environmental factors, such as smoking, nutrition, stress, endocrine disruptors, etc. Given that the causal event for disruption of lip and palate formation occurs in-utero, these epigenetic changes likely occurred in either of the biological parents’ environments to alter the epigenetic makeup of their sperm and egg. Epigenetic mutations in the germline can be permanently programmed, potentially allowing for transgenerational transmission down the germline; however, extensive epigenetic reprogramming takes place upon fertilization^16,17. This extensive reprogramming makes it difficult to determine whether these epigenetic mutations in the biological parents’ germline contribute to the methylation levels of IRF6 observed in our study. Further studies are indicated to determine if factors in the parents’ environments contributing to epigenetic changes have a significant impact on craniofacial development.

There are some limitations in our study. To begin with, our groups were not evenly distributed in terms of the number of individuals, particularly in the case of Group 5, which exclusively consisted of individuals with lip pits. To better compare the methylation levels amongst phenotypes for VWS, we aim to increase our sample size in future studies. This will also be essential to determine if any true association exists between methylation of TP63 or the IRF6 enhancer, MCS9.7, and phenotype, as our study lacked the statistical power to draw this conclusion. Another limitation, common to many epigenetic studies, is that these samples were collected at a post-natal time point that does not correspond with the critical period of the disruption of lip or palate development, which occurred in-utero. Since epigenetic factors, such as DNAm, are influenced by environmental factors and can be altered throughout life, it is unclear whether the methylation levels observed in these individuals accurately corresponds with their methylation levels at the time of craniofacial development.

In the future, we aim to conduct a more extensive replication of this study with a larger sample size. This expanded research effort will enable us to robustly validate the assertions made in the original study and investigate the significance of MCS9.7 and TP63 epigenetic influences. We chose to study individuals with IRF6 mutations only, but we plan to extend the study to include individuals with GRHL3 mutations in the future.

Genetic counseling and traditional genome sequencing provide a helpful, but incomplete picture for the basis of phenotypic inheritance and phenotypic divergence in Van der Woude Syndrome. This study lays the framework for connecting the TP63-IRF6 signaling pathway and DNA methylation to the phenotypic divergence observed in VWS. Our study emphasizes the need for further understanding of the molecular mechanisms that regulate the epigenome and contribute to clinical presentation.

Funding

This work was supported by the NIH/NIDCR K01DE027995, R37DE08559 and NIH/NICHD P50 HD103556.