Molecular and genetic studies of DLG2 in neuroblastoma and colorectal cancer

Abstract

Neuroblastoma is one of the most common extra cranial solid tumors in children. It is one of the most common causes of death amongst the pediatric malignancies. There is however a large difference between the low risk neuroblastomas that are easier to treat compared to the high-risk neuroblastomas that are difficult to treat and account for a majority of the deaths. Two common groups of high-risk tumors have different genetic features, one with MYCN amplification and the other with loss of chromosome region 11q. Despite these two genetic alterations account for a high percentage of the total neuroblastoma cases they rarely occur together. Additionally, tumors with loss of 11q also tend to have decreased genome stability, resulting in increased DNA breakage. Currently, a number of candidate 11q tumor suppressor genes have been proposed, however, none of them have by themselves been able to explain the aggressive behavior of 11q-deleted neuroblastoma. For this reason, we continued the search and identified and characterized DLG2 a novel tumor suppressor gene residing in the 11q-deleted region. DLG2 resides on the reverse strand at the proximal edge of the 11q deleted region. Within this thesis we have characterized some of the functions of DLG2 using large scale genetic data such as microarray and next generation sequencing in combination with various molecular techniques and tumor models. In addition to its importance in neuroblastoma behavior, we have also been able to show that DLG2 is altered early, during chronic inflammation, within the tumorigenesis of colon cancer, a vastly different tumor compared to neuroblastoma, indicating that DLG2 may be a common target of various cancers.

Through a combination of bioinformatic analysis and molecular methods we have shown that DLG2 affects the cell cycle and DNA repair pathways. We also discovered that the expression of DLG2 was equally affected either by 11q-deletion or by MYCN-amplification, making it a target in all high-risk neuroblastomas. We continued to show that there are a number of different isoforms of DLG2 with isoform 2 and isoform 7/8 the major isoforms expressed in neuroblastoma. The expression of isoform 2 remained stable so the decrease in DLG2 expression in neuroblastoma could be attributed to the loss of isoform 7/8, which alters the interactive ability of DLG2. In order to further elucidate the impact of DLG2-loss on DNA repair pathways, we investigated the relationship of DLG2 and genome stability by inducing dsDNA breaks by UVC irradiation or by etoposide, a topoisomerase II poison. We showed that loss of DLG2 was sufficient to result in dsDNA breaks without additional stimulus and that DNA breakage was prevented when DLG2 was present, by the removal of cells after the induction of breaks. We finally showed that DLG2 was silenced by inflammation early in the development of colon cancer. We showed that DLG2 activated the inflammasome and resulted in a decrease in STAT3 phosphorylation in adjacent cells.

To conclude, DLG2 with more research, may provide a treatment target to increase survival of high-risk neuroblastoma patients.