Differentially methylated embryonal Fyn-associated substrate (EFS) gene as a blood-specific epigenetic marker and its potential application in forensic casework

Abstract

DNA methylation patterns have the ability to reveal the activities of genes within a certain tissue at a particular time point. Tissue-specific DNA methylation patterns have been previously investigated for their applicability in the identification of forensically relevant body fluids, however there is still a lack in robust markers. While following a genome-wide scale investigation has a great potential to reveal useful tissue-specific changes, a gene-targeted approach can also lead to significant outcomes, especially in genomic locations not included in the genome-wide experiments. In this study, the potential of the candidate embryonal Fyn-associated substrate (EFS) gene for the positive identification of whole blood was investigated. For this purpose, the methylation profile of a selected genomic region containing a total of 10 CpG sites was analysed in 124 individuals via bisulfite pyrosequencing. Volunteers donated various forensically relevant tissues, including whole blood, saliva, seminal fluid, vaginal fluid and menstrual secretion. Whole blood showed the highest levels of DNA methylation (mean=0.67), while semen samples were found to be very low methylated (mean=0.06). The remaining tissues demonstrated partial mean methylation levels; more specifically, saliva - 0.43, vaginal fluid - 0.22 and menstrual blood - 0.22. One out of the 10 analysed CpG sites, CpG4, showed to be more robust, resulting in not only the highest methylation difference between blood and the rest of the tissues, but also the lowest inter-individual methylation difference. The proposed pyrosequencing assay was found to be accurate, linear and reproducible. Lastly, the method's applicability to forensic casework was assessed via the analysis of very old bloodstains stored up to 18 years, blood DNA samples stored long-term up to 9 years, mixed stains as well as other 'forensic-like' samples. In the majority of cases the expected methylation ratios were obtained indicating a stable DNA methylation pattern, however caution is necessary when analysing low quantity and/or quality samples due to potential stochastic effects. Future validation experiments can shed more light into the usefulness of EFS locus as a promising blood-specific epigenetic marker.

Validated inference of smoking habits from blood with a finite DNA methylation marker set

“…Hence, the future development of a fast and cheap laboratory tool that allows the reliable targeted analysis of the 13 CpGs highlighted here by employing a technology that can handle low quality and/or quantity DNA would be valuable. Foreseeing the future development of such a lab tool, we only included CpGs with at least a β-value difference ≥ 10% in mean or median (depending on availability per EWAS) in at least one published EWAS, to ensure detectability of the DNA methylation differences with targeted analysis technologies currently available [77,78]. We view the positive results on epigenetic inference of smoking habits from blood presented here as a promising starting point for inferring more lifestyle factors using DNA methylation markers within the concept of epigenetic fingerprinting [17].…”

From forensic epigenetics to forensic epigenomics: broadening DNA investigative intelligence

“…Subsequently, several studies have been published using various DNA methylation loci and analysis methods for different forensically relevant tissues [ 30 – 33 ]. Reported genes include FOXO3 and EFS for blood [ 32 , 34 ], SLC12A8 and BCAS4 for saliva [ 30 , 34 ], DACT1 and C12orf12 for semen [ 31 , 35 ], LOC404266 and HOXD9 for vaginal secretion [ 34 ], and SLC26A10 and LTBP3 for menstrual blood [ 13 ]. The reliable epigenetic determination of more complex body fluids such as menstrual blood can be more challenging, mainly due to the combination of different cell types and smaller methylation effects of currently proposed markers [ 13 ].…”

Evaluation of massively parallel sequencing for forensic DNA methylation profiling

“…Currently, some laboratories are testing low template input material to determine if the quantity of input DNA required for accurate and reliable methylation analysis can be reduced . However, recent studies have highlighted the influence of stochastic effects on the methylation results of low template samples . The use of replicates and statistical models incorporating the expected level of variability are needed for reliable interpretation of the data in these circumstances.…”