The importance of distinguishing pseudogenes from parental genes

Abstract

No abstract

Dear Editor,

The July-August 2014 issue of Clinical Epigenetics featured a research article describing PTEN promoter hypermethylation in multiple myeloma by Piras et al. [1]. The importance of the PTEN gene has resulted in significant efforts to identify sequence, expression and methylation changes in cancer. Piras et al. concluded that PTEN hypermethylation occurred in a subset of multiple myeloma cases but that hypermethylation did not correlate with reduced gene expression or clinical parameters. The PTEN mRNA shares 97.8% sequence identity with a pseudogene known as PTENP1. A 921-bp region of the promoters of these genes is also 91% identical. Consequently, careful consideration of assay design is required to avoid amplification of PTENP1 rather than PTEN sequences. However, the method used by Piras et al. for measuring PTEN mRNA did not distinguish between these homologues, despite numerous studies showing that PTENP1 mRNA is ubiquitously expressed in both normal and cancer specimens [2]–[5]. Furthermore, previous studies have demonstrated that apparent methylation of the PTEN promoter is likely attributable to the non-specific amplification of the highly homologous PTENP1 gene [6],[7]. We have shown that the only reliable method for distinguishing between PTEN and PTENP1 promoter methylation is single-molecule bisulfite sequencing that utilizes sequence differences between the two genes to separately analyze individual promoter molecules [6],[8]. These methodological challenges make comparisons between methylation and expression impossible when using assays that do not reliably discriminate between PTEN and PTENP1, and also negate the value of correlating these features with clinicopathological characteristics.

The challenges posed by sequence homology with pseudogenes are by no means particular to the PTEN gene. For example, the DNA mismatch repair gene PMS2 shares >95.2% sequence identity with at least six other genes (PMS2CL, PMS2L2, PMS2P4, PMS2P5, PMS2P1 and PMS2P11) making analysis of the PMS2 CpG island promoter region particularly challenging.

In light of the recent manuscript by Piras et al., it is necessary to highlight the importance of rigorous methodology when investigating DNA methylation changes in cancer, especially concerning genes with homologues or pseudogenes such as PTEN.