DNA Methylation Analysis Determines the High Frequency of Genic Hypomethylation and Low Frequency of Hypermethylation Events in Plasma Cell Tumors

B. Salhia, A. Baker, G. Ahmann, D. Auclair, R. Fonseca, J. Carpten
2010 Cancer Research  
Multiple myeloma (MM) is a plasma cell malignancy of the bone marrow, which evolves from a premalignant stage called monoclonal gammopathy of undetermined significance (MGUS). In some patients, an intermediate stage referred to as smoldering multiple myeloma (SMM) is clinically recognized, with the full-bore malignancy termed MM. We conducted a study to assess differential CpG methylation at 1,500 genic loci during MM progression and profiled CD138 + plasma cells from MGUS, SMM, and MM
more » ... ; human myeloma cell lines; and normal plasma cell (NPC) samples. We showed that the number of differentially methylated loci (DML) increased with tumor grade, and the vast majority were due to hypomethylation. Hierarchical clustering analysis revealed samples that coclustered tightly with NPC. These cases, referred to as "normal-like," contained significantly fewer DML when compared with their non-normal-like counterparts and displayed overall methylation levels resembling NPC. This study represents one of the first methylome interrogation studies in MM and points toward global hypomethylation at genic CpG loci as an important and early mechanism driving myelomagenesis. Determining the set of critical genes and pathways based on the myeloma methylome is expected to lead to an improved understanding of biological mechanisms involved in myelomagenesis. Cancer Res; 70(17); 6934-44. Figure 4. Percent methylation values for eight DML analyzed by GoldenGate and Pyrosequencing are displayed. Hypomethylation (A) and hypermethylation (B) is evident by GoldenGate or Pyrosequencing in 16 MM compared with 4 NPC samples. Salhia et al. NOTE: The column denoted by r is the number of network objects from the data set for a given protein class, whereas R is the number of network objects for a given protein class in the complete data set. N represents the total number of network objects in the complete data set. P values <0.05 are significant.
doi:10.1158/0008-5472.can-10-0282 pmid:20736376 fatcat:4mbtkau4svg7vaf4oi25xjval4