6. PGT-A METHOD ALLOWING COMBINED PGT-M FOR DETECTION OF HETEROPLASMY AND ESTIMATION OF MTDNA MUTATION LOAD IN EMBRYO BIOPSIES. (August 2019)

• Record Type: Journal Article
• Authors: Myers, S.; Jasper, M.
• Is Part Of: Reproductive biomedicine online. Volume 39(2019)Supplement 1
• Page Start: e17
• Publication Date: 2019-08
• Subjects: PGT-M -- mitochondria -- mtDNA -- heteroplasmy -- WGA; Human reproductive technology -- Periodicals; Human embryo -- Periodicals; Reproduction -- Periodicals; 616.692
• DOI: 10.1016/j.rbmo.2019.04.041
• Languages: English
• ISSNs: 1472-6483

Abstract

Abstract : Introduction: A large number of pathogenic mitochondrial DNA (mtDNA) mutations have been identified and are implicated in a variety of disorders. As mtDNA is maternally inherited, women with pathogenic mtDNA mutations are likely to have affected children, the severity of the phenotype depending on the heteroplasmy proportion. A systematic meta-analysis showed that there is a ≥ 95% chance of being unaffected at a mutant level of ≤ 18% (Hellebrekers et al, 2012). PGT-M can be used to identify embryos with mutation loads below this phenotypic threshold. The PG-Seq™ kit yields superior coverage of mtDNA and may allow combined PGT-A and PGT-M for mtDNA diseases from single biopsies. Material and methods: Two cell lines, a reference and "mutant", with mitochondrial single nucleotide variants (SNV) were selected. Six 5-cell samples from each cell line were whole genome amplified (WGA). The replicates were pooled before the cell lines were mixed in proportions ranging from 0% to 100% (mutant), in increments of 10%. Libraries were prepared for each of the pools prior to sequencing using the MiSeq instrument (Illumina). SNV sites with < 10x depth across all samples were removed, leaving 31 SNV sites. The proportion of mutant SNV at each site was calculated and analysed. The data were corrected for variance introduced by pipetting error in pooling. These corrections were only applied to samples with proportions of 40%, 50%, and 60%, and subsequent analyses were restricted to these data. To model data at different read depths we combined data across sites. Results: The average (± sd) mitochondrial read count for cell line samples was 5037 (± 1083). After filtering, the average (± sd) depth of coverage for the cell line samples at the SNV sites was 25.8x (± 12.5). Combining SNV data in silico to achieve a read depth typical of a standard PG-Seq™ kit 48 sample run using a single embryo biopsy (500, 000 reads total), the observed SNV proportion was within 25% of the expected proportion in all cases. Combining SNV data to achieve a read depth of 5x a standard PG-Seq™ kit 48 sample run, the observed SNV proportion was within 11% of the expected proportion in all cases. This means that in any case where a heteroplasmy is detected at 7% or lower, including no detection, the mutant load should be below the phenotype threshold. Conclusions: The superior amplification of mtDNA achieved by the PG-Seq™ kit may allow combined PGT-A and PGT-M for mtDNA analysis. Combining PGT-A and PGT-M in this novel way would provide a streamlined workflow over currently available workflows. While accuracy is dependent on read depth, read depth can be increased by modulating sequencing throughput or by alternate means such as using PerkinElmer's Target Sequence Enrichment protocol. Due to variable but consistent depth of coverage across the mtDNA, some sites will be more suited for analysis by this method than others.