The relative improvement in enrichment. Error bars represent SD, n = 3. M M-WGA, U

The relative improvement in enrichment. Error bars represent SD, n = 3. M M-WGA, U U-WGA, NoT no template controlWee et al. Clinical Epigenetics (2015) 7:Page 5 ofa methylation assay against GSTP1, a clinically important HDMR biomarker of prostate cancer [26, 27]. Currently, the methylation status of GSTP1 is the only FDA-approved methylation biomarker undergoing clinical trials [28] and is usually detected by head loop PCR [29]. While potentially useful, the approach has various limitations such as the need for bisulfite conversion and complicated primer design. Simpler assays GGTI298MedChemExpress GGTI298 27797473″ title=View Abstract(s)”>PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/27797473 are hence needed to facilitate widespread adoption. The MBD/HRP approach described herein may fulfil this need. As shown in Fig. 5, the assay could detect as little as a 5 methylated sample. As a control, we also tested the input DNA prior to MBD enrichment. Since the amount of input DNA can affect the resultant amount of amplification, we therefore propose that to improve accuracy, the ratio of the signal generated post-MBD to pre-MBD (initial DNA input) be used to generate a methylation score (Fig. 5c). This method also had good reproducibility with intra-assay CV (n = 8) of 7.8 and inter-assay CV (n = 4) of 9.1 (Additional file 1: Figure S4). To demonstrate feasibility on complex biological samples, we first assayed two cancer cell lines (Fig. 6a ). DuCap cells were found to be methylated but less so for HeLa cells. We then assay Jurkat cells before and after 5-Aza treatment (Fig. 6d ). As expected, GSTP1 was de-methylated significantly after 5-Aza treatment. Finally, the method was then applied to four DNA samples derived from the urine of metastatic, castration-resistant prostate cancer patients (Fig. 6g ). Three of the four patients were found to be highly methylated at the GSTP1 gene promoter. Together, these results indicated that the MBD/HRP colorimetric assay may be a useful alternative for detecting GSTP1 methylation in the clinic.The assay was further extended to the ESR1 gene promoter which is also aberrantly methylated in cancers [30]. To this end, we assayed Jurkat cells before and after 5-Aza treatment with M-WGA and U-WGA as specificity controls (Additional file 1: Figure S5). As expected, ESR1 was de-methylated in the 5-Aza-treated cells compared to na e Jurkat cells. The de-methylation of both GSTP1 and ESR1 in 5-Aza-treated cells were consistent with the genome-wide hypomethylation observed in Fig. 3. Based on the data shown here, a MBD/HRP-based approach may be useful for both total genomic and gene level methylation. However, a known limitation of all MBDbased approaches is the inability to provide methylation information at individual CpGs. In addition, MBD is also not able to provide 5hmC information [18]. While methods exist for 5hmC, there is no simple single-step method besides HPLC and/or MS able to simultaneously interrogate both 5mC and 5hmC. Nonetheless, this MBD/HRP approach could be adapted to an analogous assay for 5hmC using affinity reagents currently available for 5hmC [31].Adapting to an electrochemical assay using disposable screen-printed electrodesThe benefits of an electrochemical assay include miniaturization, improved performance and reduced cost [32?4]. Since TMB is electrochemically active [21], we had the potential to adapt our MBD/HRP assay onto commercially available screen-printed carbon electrodes (SPCE) as a convenient alternative readout platform (Fig. 7a). As a proof of concept, we applied the electroc.