DNA methylation results depend on DNA integrity – role of post mortem interval

Major questions of neurological and psychiatric mechanisms involve the brain functions on a molecular level and cannot be easily addressed due to limitations in access to tissue samples. Post mortem studies are able to partly bridge the gap between brain tissue research retrieved from animal trials and the information derived from peripheral analysis (e.g. measurements in blood cells) in patients. Here, we wanted to know how fast DNA degradation is progressing under controlled conditions in order to define thresholds for tissue quality to be used in respective trials. Our focus was on the applicability of partly degraded samples for bisulfite sequencing and the determination of simple means to define cut-off values.After opening the brain cavity, we kept two consecutive pig skulls at ambient temperature (19-21°C) and removed cortex tissue up to a post mortem interval (PMI) of 120h. We calculated the percentage of degradation on DNA gel electrophoresis of brain DNA to estimate quality and relate this estimation spectrum to the quality of human post-mortem control samples. Functional DNA quality was investigated by bisulfite sequencing of two functionally relevant genes for either the serotonin receptor 5 (SLC6A4) or aldehyde dehydrogenase 2 (ALDH2).Testing our approach in a heterogeneous collective of human blood and brain samples, we demonstrate integrity of measurement quality below the threshold of 72h PMI.While sequencing technically worked for all timepoints irrespective of conceivable DNA degradation, there is a good correlation between variance of methylation to degradation levels documented in the gel (R2=0.4311, p=0.0392) for advancing post mortem intervals (PMI). This otherwise elusive phenomenon is an important prerequisite for the interpretation and evaluation of samples prior to in-depth processing via an affordable and easy assay to estimate identical sample quality and thereby comparable methylation measurements.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.