NEW YORK (GenomeWeb) – A team from the University of Southern Denmark has used methylation arrays to identify a comprehensive list of candidate epigenetic biomarkers that could be useful in non-invasive prenatal diagnosis.
In the study, published late last month in PLoS One, the team led by USD researcher Lotte Hatt set out to distinguish fetal-specific methylated sites through a comparison of chorionic tissue and maternal blood cells using Illumina's Infinium HumanMethylation450 Beadchip kit.
The investigators were able to identify a number of CpG sites that were differentially methylated in chorionic tissue versus maternal blood — including markers on chromosomes 13, 18, and 21, as well as a list of alterations in regions associated with sub-microscopic deletion or duplication syndromes. Then, they narrowed their list down to a set of markers that would have particular utility in the development of PCR-based NIPT tests using methyl-sensitive restriction enzymes.
Hatt told GenomeWeb in an email that she and her collaborators had performed previous studies in the NIPT field looking at the isolation of fetal cells from maternal blood, as well as PCR-based cell-free DNA-based testing for gender determination.
Though next-gen sequencing-based NIPT has been successfully commercialized by a number of companies, Hatt wrote that NGS-based NIPT strategies still have some issues, namely that they can fail to deliver a result at low cell-free DNA fractions, and that they have lower specificity for trisomy 13 and trisomy 18 than they do for trisomy 21.
"Methylation markers hold promise for an improved specificity for trisomy detection and can possibly also be used in the diagnosis of smaller microdeletion or duplication syndromes," Hatt wrote.
Moreover, a major advantage in advancing such markers would be that they could allow NIPT using PCR rather than NGS, thus reducing cost and complexity, she said.
Hatt and her team's study is not the first to identify differentially methylated sites with potential for non-invasive fetal diagnosis, but thus far only a few sites have been reported as potential diagnostic markers.
In their study, Hatt and her colleagues conducted a broad search for methylation markers with potential utility for NIPT by comparing a set of 10 maternal blood samples with 12 corresponding first trimester chorionic samples from normal placentas using the Illumina beadchip technology. The CVS and maternal samples were not from the same women, but were matched for gestational age and the gender of the fetus.
This initial analysis yielded a starting list of about 144,000 differentially methylated CpGs (DMCs). The group then decided to narrow this list further by defining a more stringent threshold for hypo- or hypermethylation, since if these sites were to be useful as diagnostic markers, the difference in methylation would need to be significant. By limiting hypomethylation to below a b-value of 0.25 and hypermethylation to above a b-value of 0.75, the team was able to take the list down to about 6,000 DMCs.
Finally, to narrow the list even more in terms of markers with practical suitability for PCR-based testing using methyl-sensitive restriction enzymes, the investigators restrained the list to only those DMCs that contained a restriction site for one of 16 available MSREs. They also limited the list to sites with consistent methylation across at least 10 of the 12 study samples.
The end result was a list of 958 CpG sites, which the authors said were dispersed throughout the genome with possible markers located on all chromosomes.
Looking specifically for markers that could be useful for fetal aneuploidy detection, the team found 12 DMCs on chromosome 21, six on chromosome 18, and 26 on chromosome 13. The team also identified DMCs in genetic regions of interest other than those associate with trisomies 13, 18, and 21, including areas where smaller deletions cause known syndromes such as Prader-Willi, Smith-Magenis, and cri-du-chat, the authors wrote.
As a validation step, the group compared their data to DMC's identified in a previous study by researchers from the University of Pittsburgh. This study also used Illumina bead chip arrays, but in a smaller scale format, covering only 27,000 CpG sites, compared to the current kit's 450,000, and limited its analysis to chromosomes 13, 18, 21, X, and Y.
Of 47 total DMCs identified by the Pittsburgh group, the Danish researchers also picked out 44. In addition, the Danish researchers found that methylation levels for these 44 sites were to those determined in the previous study.
Beyond the recently published results, Hatt said she and her team have additional DNA methylation data that they expect to publish soon, but which she said she could not discuss in detail.
She also did not provide any details about whether her group has any plans to develop specific MSRE-based assays using any of the markers they identified.
In the study, the authors wrote that further investigation would be needed to validate any specific markers for use in fetal diagnosis.