Searching for New Genes Involved in Familial Colorectal Cancer Type X by Whole-Exome Sequencing

Introduction:

Hereditary Non-Polyposis Colorectal Cancer (HNPCC) is an autosomal dominant inherited condition characterized by an increased susceptibility to colorectal (CRC) and other associated cancers and defined by the Amsterdam I and II clinical criteria. An important fraction of these families is known as Lynch Syndrome and is caused by germline inactivating mutations in the mismatch repair (MMR) genes. This results in tumors that lack the corresponding proteins and fail to repair DNA through this pathway, causing microsatellite instability (MSI) and leading to an accumulation of somatic mutations. However, it is estimated that almost half of the families that fulfill the Amsterdam criteria do not present any defects in the MMR genes. For this reason, the term Familial Colorectal Cancer Type X (FCCTX) emerged to designate such group of HNPCC families that lack the MMR deficiency and resulting MSI that define Lynch Syndrome. Thus, FCCTX patients have microsatellite stable tumors with normal expression of the MMR genes, and the genetic basis underlying their predisposition to CRC and other related cancers remains to be elucidated.

Thanks to arrival of Next Generation Sequencing (NGS), some new CRC-predisposition genes have been recently identified; however, most FCCTX cases remain unexplained. FCCTX is thought to be a heterogeneous group of families, presumably including different yet-to-be-discovered genetic syndromes involving germline mutations in novel cancer-predisposing genes, but that could also result from a combination of low-penetrance mutations in different genes. For this reason, identifying the genetic cause of the increased cancer susceptibility in FCCTX families is still a challenge.

Objectives:

The ultimate aim of this study is the identification of new genes that could explain the increased cancer susceptibility in FCCTX families. Within this aim are enclosed the following secondary objectives: 1) to obtain a list of potentially pathogenic candidate variants for each family; 2) to classify and characterize the variants identified; 3) to perform functional and/or expression studies when possible and suitable. All this with the purpose of finding the cause of CRC heritability in the studied families.

Methodology:

The main method on which this work was based is whole-exome sequencing, one of the applications of NGS. The obtained data was strictly analyzed and those prioritized variants were validated by PCR followed by Sanger sequencing. In addition, the segregation and loss of heterozygosity (LOH) were studied. Finally, some of the candidate genes were characterized by different expression and functional essays, involving both recombinant proteins and cell cultures.

Results:

The thorough filtering and in silico analysis of the variants identified by whole-exome sequencing in the affected members of 13 FCCTX families allowed the prioritization of a list variants. After their validation, the segregation in other family members and LOH were studied when possible. A total of 44 candidate variants were selected, most of which affected genes involved in important cellular events such as DNA repair, tumor suppression or different cancer-related signaling pathways. On the other hand, based on the relevance of the candidate genes and the nature of the alterations, three of the variants were selected for a more in-depth evaluation.

One of the variants selected for further characterization was SETD6 c.791_792insA, p.(Met264IlefsTer3), carried by all the CRC-affected members of one of the families and absent in the healthy ones. SETD6 is a mono-methyltransferase known to regulate the NF-κB and Wnt signaling pathways, among other. The characterization of the truncated version of SETD6 proved that it lacks its enzymatic activity as a methyltransferase, while maintaining other properties such as its expression, localization and substrate-binding ability. In addition, the mutant allele is expressed in the tumor and the resulting mutant protein competes with the wild type for their substrates, pointing to a dominant negative nature. These findings suggest that this mutation impairs the normal function of SETD6, which may result in the deregulation of the different pathways in which it is involved, contributing to the increased cancer susceptibility in this family.

Another of the studied variants was PTPRT c.4099dup, p.(Asp1367GlyfsTer24), which showed a compatible segregation with the different types of cancer of the family (of the colon, breast and endometrium), without being present in two healthy elderly relatives. PTPRT is a tumor suppressor gene found to be frequently mutated at a somatic level in CRC and other cancers, having been proven that these mutations contribute to tumor development. The studied mutation results in the loss of a significant fraction of the second phosphatase domain of the protein, shown to be essential for its activity. In addition, the tumors of two of the carriers exhibit epigenetic inactivation of the wild-type allele and an altered expression of PTPRT’s downstream target genes, consistent with a causal role of this germline mutation in the cancer predisposition of the family.

Last but not least, the third variant was PYGO1 c.1084T>C, p.(Ser362Pro). The protein encoded by this gene plays an important role in the Wnt pathway, which is essential in CRC. Despite being a missense mutation, this variant is located in the zinc finger (PHD domain) of the protein. Even though the initial segregation study was favorable, a later analysis of the tumor of a third CRC-affected member showed that this patient did not carry the mutation. Interestingly, the study of mutant PHD proved that the mutation significantly decreases Pygo1’s histone-binding affinity, which would be expected to decrease the Wnt pathway contrary to what is usually seen in a CRC context. Therefore, these results do not support a causal role of this variant in the increased cancer predisposition of this family, although its implication in tumor progression cannot be ruled out.

Conclusions:

Among all the selected candidate variants, this work proves the causality of two germline truncating mutations in SETD6 and PTPRT, although more studies are necessary to determine the exact role that they play in cancer susceptibility. However, there is not enough evidence supporting a causal role for the variant identified in PYGO1. Additional studies will clarify the relevance of these and other candidate genes.