Gene silencing is a major mechanism of tumour suppressor gene inactivation in neoplasia, including multiple myeloma (MM). Prolyl hydroxylation is an important post-translational modification that affects the structure and function of collagen. The human genome contains 3 prolyl 3 hydroxylases (P3H) which are closely related in structure and function: Leprecan (P3H1), Leprecan Like 1 (Leprel1, P3H2) and Leprecan Like 2 (Leprel2, P3H3). Leprel2 is methylated in multiple solid tumour types, whereas methylation in Leprel1 appears specific for breast cancer. Leprecan is not detectably methylated in any solid tumour we have analysed. Here, we examine the CpG methylation status of the P3H genes in patients with MM.
Bone marrow aspirate samples from 38 MM patients (22 males, 16 females) were obtained at diagnosis. Methylation-specific PCR (MSP) was employed to study methylation in the Leprel 1, Leprel2 and Leprecan CpG islands. Genomic DNA was isolated and bisulphite modification performed using commercially available kits (QIAmp DNA mini kit, Qiagen and EZ DNA methylation kit, Zymo Research respectively). Control methylated and unmethylated genomic DNAs were included in each experiment. We used 2 sets of independent primers, which map to different areas of the leprel 1(P3H2) CpG island. Pyrosequencing is ongoing to confirm the sensitivity and specificity of MSP for analysis of P3H gene methylation in our study population. Ten bone marrow samples from patients with border line thrombocytopenia that were proved to have no neoplasia served as negative controls. Logistic regression analyses were used to measure the association between gene methylation and age, ISS stage, extramedullary disease, bone disease, anemia, renal failure, multiple treatments and relapsed/refractory disease.
None of the three genes under study was methylated in the control bone marrow samples, but methylation in the CpG island of Leprel 1 was detected in 20 cases (52%). No methylation was detected in the CpG islands of Leprel2 or Leprecan. Trends noted were that patients with methylated leprel 1 were more likely to have ISS stage 3 (OR 1.5, p=0.6), to have received multiple treatments (OR=1.7, p=0.4) and to have received radiotherapy for either bone lytic lesions or extramedullary disease (OR=5.6, p=0.1). Overall no significant association was found with any clinical parameter although the study might have been underpowered due to the small population size.
This is the first demonstration of methylation of genes encoding collagen, the prolyl hydroxylases in MM, suggesting that this may be a novel class of myeloma suppressors. Interestingly, Leprel1 was methylated at high frequency, whereas Leprecan and Leprel2 were unmethylated, implying a striking specificity in MM for methylation in Leprel1. These findings warrant further evaluation in a larger sample of patients in order to better define the prognostic and clinical utility of leprel1 methylation in MM.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.