Laboratory for Oral Molecular Biology

Basic molecular and cellular research on human CLP is challenging since there are no proper human models available. Basically all of our knowledge on CLP is based on animal studies. Fortunately, the process of craniofacial morphogenesis is highly conserved across many species, and mice and humans share the majority of protein-coding genes. Yet, there are certain differences between humans and mice regarding face development, and therefore, not every finding in animals can be transferred to humans. Furthermore, there is a strong societal urge to minimize animal experimentation in research, which asks for alternative study models.

In line with this, we established a cell bank of CLP patient-derived fibroblasts and keratinocytes. During standard surgical closure of the lip at the age of 3-5 months, the marginal portion of the upper lip covering the cleft is in excess and needs to be removed. With the approval of the Kantonale Ethikkommission of Bern, Switzerland (protocol number: 2017-03194), we started to collect such discarded cleft lip tissues to isolate primary cells.

We developed a very robust and reproducible explant culture assay, allowing us the isolation of keratinocytes and fibroblasts from a single donor. We thoroughly characterize these cells by applying various assays before archiving them. We are well aware that the use of postnatal tissue to study abnormalities that arise early in development comes with some challenges and assumptions. But our own work with these cell cultures has convincingly demonstrated that most of the original tissue characteristics are maintained in the cells, allowing scientific discoveries, which might help to better understand CLP.

Currently, our unique cell bank consists of approximately 100 CLP patient-derived lip cell pairs (keratinocytes and fibroblasts), which is supplemented with tissue sections whenever the biopsy size allows it, enabling histological analyzes. Furthermore, we also regularly sample primary cells derived from control tissues (e.g., foreskin, cheek, eyebrow) and cells from the oral cavity (e.g., gingiva, periodontal ligament, dental pulp).

The human lip is a complex structure because it is a mucocutaneous junction, where the outer labial skin transitions into the inner oral mucosa. Between these two distinct tissues, there is a transition zone called vermilion.

Considering that we isolate cells from discarded lip tissue and that these biopsies cannot be standardized in their tissue (labial skin and mucosa) composition, we regularly isolate mixed cells derived from various proportions of labial skin and mucosa tissue. In particular cases, the pediatric surgeons performing the corrective cleft lip surgery can precisely separate labial skin from mucosa, which allows us to isolate pure lip skin and mucosa cells (keratinocytes and fibroblasts).

Using these cells, we aim to:

• Characterize the various cells that build the complex structure of the human lip; Identify specific markers that can distinguish the various tissue types.
• Check whether the pure cells maintain their original tissue identity when cultured in vitro.
• Specifically, immortalize skin and mucosa lip keratinocytes, which allows us to easily expand these cells for assays and readouts requiring a great number of cells.
• Establish 3D lip models using the skin and mucosa lip cells separately.
• Model the human vermillion in 3D using intraindividual lip skin and mucosa cell pairs.

In research, lip tissue biopsies are not easily accessible. We have gathered more than 100 lip biopsies, concentrating on CLP patients, which serve as the basis for these studies. Due mostly to a scarcity of lip tissue, a human vermillion model using lip skin and mucosa cells has not yet been produced. In order to create therapeutically relevant study tools, we might use our cells in 3D cultures and attempt to simulate the human vermilion.

A number of factors, including aging, infections, tumors, severe traumas, and craniofacial malformations, can induce lip abnormalities. In order to investigate the mechanisms behind these problems, appropriate and pertinent models are needed, as abnormalities in the lip structure can be quite disfiguring for those affected. We believe that our models can help with that. Furthermore, in an effort to reduce animal testing, such cell-based lip models align with the 3R (Replace, Reduce, Refine).   

Numerous genes have been identified, which, when mutated, can result in CLP. Among them are several transcription factors (TF) that regulate complex gene regulatory networks.

We aim to study CLP candidate gene function in healthy lip cells using the CRISPR/Cas9 technology. This approach allows us to either modulate protein expression or to correct pathological gene variants. We will follow this up with various molecular and cellular assays, including 3D cultures as well as OMICS data.

However, anticipating difficulties when using CRISPR/Cas9 and subsequent clonal selection of modified primary keratinocytes, we sought to first immortalize the patient-derived keratinocytes. We applied a one-step approach, with TERT overexpression and simultaneous shRNA-mediated p16^INK4A knockdown. The immortalized keratinocytes maintained all the original characteristics of the corresponding tissue and the primary cell cultures.

Data gained from these projects should strengthen our understanding about CLP in several ways:

• The general gene function using clinically relevant postnatal human cells derived from the anatomical site of the cleft, the lip.
• Which gene regulatory networks are regulated by distinct TFs and which by more than one?
• What can we learn from the gene activity in postnatal cells, translating it to the early embryogenesis?
• What is the minimal threshold required for a TF to fulfill its required activity?

In CLP patients, alveolar cleft, or the discontinuity of the maxillary arch, is one of the most common clinical manifestations. In these cases, grafting is an essential procedure in order to reestablish the integrity of the alveolar bone allowing proper tooth eruption.

During the closure of the cleft lip, which is performed at the age of 3-5 months, lip tissue fragments have to be excised by the surgeon in order to achieve proper lip reconstruction. These discarded fragments might represent an optimal source of cells with regenerative potential. As an alternative to bone grafts, we aim to use these autogenous cells as a new source for the regeneration of new bone and other tissues in CLP patients.

Our translational approach offers the possibility to develop a personalized therapeutic strategy for CLP patients with the potential to lower their burden of care by reducing the number of required invasive surgeries.

Fibroblasts are essential for tissue homeostasis, wound healing response, and growth and development. In order to accommodate these many traits, fibroblasts are remarkably diverse. Aside from phenotypic and functional differences between organs, fibroblasts also exhibit intra-tissue heterogeneity. The latter conclusion has been underlined by the identification and description of at least five subpopulations in skin that perform diverse and specific functions. The analysis of oral mucosa tissue showed similar findings.

Because we work with human lip tissue, a well-known mucocutaneous junction, we anticipate having highly variable fibroblast cell cultures. It is critical to identify the various subpopulations in our fibroblast cultures and determine which particular subpopulations are most successful in forming de novo bone in translational applications. This is clinically relevant since we are looking into harnessing the osteogenic capacity of CLP patient-derived lip fibroblasts for future regenerative treatment applications. 

As part of the Dental School of the University of Bern, our laboratory has access to many oral/dental-related tissues: Gingiva, Periodontal Ligament (PDL), dental pulp, palate, buccal mucosa. From these tissues, we isolate primary cells (see project “CLP patient-derived cell bank”), which we use to investigate the following research questions:

• To assess the osteogenic potential of PDL- and dental pulp-derived mesenchymal outgrowths from the same donor.
• PDL- and dental pulp-derived mesenchymal outgrowths: mesenchymal stem cells or fibroblasts?
• Can we model the human gingiva using primary cells?
• To examine whether and how fibroblasts dictate the keratinization pattern of the overlying epithelium.
• How specific titanium implant surface properties influence the behavior of clinically relevant soft tissue cells.

Genotype-phenotype correlations in Van der Woude Syndrome patients

Van der Woude syndrome (VWS, OMIM #119300) is the most common cleft syndrome accounting for 2% of all cleft cases with a prevalance of 1/35'000 to 1/100'000 newborns. Many VWS individuals (30-79%) present with very characteristic depression (pits) near the center of the lower lip in combination with CLP or cleft palate only. The condition is caused by gene mutations and is inherited in an autosomal dominant way.

Initially, the VWS locus was mapped to human chromosome 1q32-q41 and the gene encoding for the transcription factor Interferon Regulatory Factor 6 (IRF6) was found to be mutated in around 72% of VWS cases (Kondo et al., 2002). Later, dominant-negative mutations within the transcription factor grainyhead like transcription factor 3 (GRHL3), a direct target gene of IRF6, have been identified in VWS-affected families not having pathogenic mutations in IRF6 (Peyrard-Janvid et al., 2014). More recently, a third VWS-associated gene has been discovered: a rare missense mutation within NME1, encoding for an IRF6-interacting proteins, was found to cause VWS in individuals not having IRF6 and GRHL3 mutations (Parada-Sanchez et al., 2017).

Most literature is available on the function of IRF6 in VWS: Keratinocytes isolated from skin biopsies from the hip region of VWS individuals (harboring IRF6 mutations) exhibit an increased proliferation rate (Hixon et al., 2016) and VWS patients experience more wound healing complications following repair of their clefts (Jones et al., 2010) compared to non-syndromic CLP keratinocytes and patients, respectively. Up to now, more than 200 mutations within IRF6 have been found associated with either VWS, popliteal pterygium syndrome (PPS, OMIM #119500), which can mimic the allelic disorder VWS in mild cases, or with non-syndromic CLP. However, the underlying mechanistic link between gene mutation and phenotype remains elusive for most of them.

Using a combination of genetics, molecular and cellular biology we aim to establish genotype-phenotype correlations in VWS-patients, which should help to understand gene function in health and disease.

CLP and altered wound healing

Although a number of genes have been linked to cleft lip/palate (CLP), which is one of the most common congenital craniofacial malformations, the genetic cause is only known for a minority of the cases. A fraction of CLP patients undergoing surgical repair of the cleft suffers from excessive scar formation, which later can interfere with maxillary growth. The impaired wound healing procedure may be attributed to the differential gene expression of those patients who are prone to develop excessive scarring.

• To elucidate the contribution of the genetic background to the abnormal healing profile in vitro of CLP patients, a genotype-phenotype correlation of CLP patient-derived cells is attempted. The expression profiles of genes that are related to wound healing, extracellular matrix formation, and cleft lip/palate are being assessed in order to unravel a possible link between the aforementioned processes. Patient-derived cells can be categorized into different subgroups, based on their differential gene expression correlated to their behavior and motility.
• Since wound healing complications are observed in 30% of CLP patients undergoing primary cleft surgery, we hypothesize that certain CLP-associated transcription factors might be responsible for the impaired wound healing. We focus on the transcription factors IRF6 and GRHL3 that have been found mutated in CLP, are involved in wound healing and are also supposed to act as tumor suppressors. Since tumors are similar to non-healing wounds and misuse developmental signalling pathways for their manifestation, we speculate that mutations in these genes might represent common factors underlying the development of CLP, wound healing complications and cancer. The two genes of interest are knocked out with CRISPR-Cas9 in keratinocyte cell lines in order to elucidate their function in regard to proliferation, differentiation and migration potential.