Epigenetic studies in rhabdomyosarcoma

Division Of Basic Sciences - Nci

Investigators

Linked publications & trials

Abstract

In our previous studies of human rhabdomyosarcoma (RMS), we detected numerous DNA methylation differences between fusion-negative (FN) and fusion-positive (FP) RMS tumors. Furthermore, two methylation-defined subsets were detected in the FN RMS category and two subsets were detected in the FP RMS category; these subsets were associated with differences in the frequency of mutations that are commonly found in these categories. When considering the heterogeneity within the RMS family of tumors, we postulate that there are differences in the underlying cell of origin (lineage) as well as the genetic changes (driver mutations) that occur in these cells. Though it would be difficult to dissect the contribution of lineage and driver mutations to DNA methylation patterns in human RMS tumors, the availability of engineered mouse models of RMS provides a strategic direction to address these questions. In particular, my collaborator Dr. Charles Keller (Children's Cancer Therapy Development Institute) has used the Cre-LoxP system to generate a series of mouse models in which one or more driver mutations (Pax3-Foxo1 [P3F] fusion and deletions of Tp53, Rb1 or Ptch) is introduced into one of four myogenic lineages (as defined by activity of the Pax3, Pax7, Myf5 or Myf6 expression regulatory elements). Mice with an engineered Tp53 deletion +/- other driver mutations frequently developed soft tissue tumors resembling RMS or related sarcomas. After identifying tumors corresponding to a variety of lineages and driver mutations, DNA was subsequently isolated from these tumors for DNA methylation analysis on the Infinium Mouse Methylation BeadChip (Illumina). Unsupervised analysis of the DNA methylation data identified two main clusters of mouse tumors, one cluster in which P3F is expressed in Pax3, Myf5 or Myf6-associated lineages and a second cluster in which P3F is expressed in a Pax7-associated lineage or other driver mutations (without P3F) were expressed in any of the four lineages. Previous studies revealed very low P3F expression when P3F is directed to the Pax7 lineage (but not in the other three lineages) so that the two main clusters can be defined by high P3F expression vs low/no P3F expression. This finding provides evidence supporting a major contribution of the driver mutation P3F to the DNA methylation pattern. In addition to this finding, the cluster with low/no P3F expression consists of two distinct subclusters. Although both of the subclusters contain tumors in which a non-P3F driver mutation is expressed in the Pax3 or Myf6 lineage, one subcluster contains all tumors in which a driver mutation is expressed in the Pax7 lineage and the other subcluster contains all tumors in which a non-P3F driver mutation is expressed in the Myf5 lineage. It is particularly interesting to note that one subcluster contains all evaluable cases with Ptch1 and Tp53 deletions in the Myf5 lineage whereas the other subcluster contains all evaluable cases with Ptch1 and Tp53 deletions in the Pax7 lineage. This finding thus provides support that the lineage also makes a major contribution to the DNA methylation pattern in these tumors. Based on these findings, the DNA methylation pattern in these tumors appears to be influenced by both the driver mutations and the underlying lineage. Additional unsupervised studies to compare human and murine RMS tumors revealed substantial differences in DNA methylation pattern such that all human RMS tumors were more similar to each other and all mouse were more similar to each other than the FP RMS tumors or the FN RMS tumors in the two species. Subsequent supervised studies to determine genes that were differentially methylated between FP and FN RMS tumors in each species identified only a small number of genes that were differentially methylated in both species. In particular, for genes in which there was differential methylation of the 5' regulatory regions between FP and FN tumors, only 2% showed differential methylation in both human and mouse tumors. Similarly for genes in which there was differential methylation of the gene body regions between FP and FN tumors, only 4% showed differential methylation in both species. Therefore, despite the significance of DNA methylation changes in RMS, the specific genes involved appear to differ between human RMS tumors and the tumors developing in these engineered mouse model systems. This difference may reflect an intrinsic difference between human and mouse biology or may point to issues that are just not well recapitulated in these specific mouse models.