Abstract Study question Can metabolic profiling predict embryos at risk of single gene defects and chromosomal abnormalities and how are these reflected in their ultrustructure? Summary answer Different metabolic profiles are observed between embryos with chromosomal abnormalities and monogenic disorders and normal embryos, linked to altered mitochondrial and other organelles’ structure/function What is known already Down’s syndrome embryos and Monosomy 21 embryos have previously been shown to have differential expression of metabolites compared to normal embryos, but limited studies, have investigated in detail the metabolic profiling of embryos with other abnormalities or those with monogenic disorders in comparison to chromosomally normal embryos or their reflection in their ultrustructure. Study design, size, duration 520 embryos were biopsied on day5 for PGT-A (n = 178-47 cycles) or on day3 for PGT-A (n = 178-33 cycles) and PGT-M (n = 164-24 cycles). Following transfer of normal embryos, spare embryos, rejected for transfer following day5 or day3 biopsy were processed for TEM (n = 60 day3 biopsied, n = 60 day5 biopsied/vitrified), 60 unbiopsied/vitrified embryos from egg donation cycles used as control. Culture media were collected from the embryos and analysed by hydrophilic interaction liquid chromatography tandem mass spectrometry (HILIC-MS/MS). Participants/materials, setting, methods PGT-A and PGT-M were performed in 2 IVF Units. Metabolic profiling was conducted in a Forensic Toxicology Laboratory by HILIC-MS/MS (100 metabolites). Ultrastructure analysis by TEM was carried out in an Academic Hospital and Histology/Embryology Laboratory following embryo fixation in 3% glutaraldehyde, 1% osmium tetroxide, washes in PBS and staining with 1% aqueous uranyl acetate. Main results and the role of chance The implantation rates (+ve hCG/ET) for the PGT-M cycles was 68.2% and the for the PGT-A (blastocyst and cleavage biopsy cycles) 64.5% and 65% respectively. Characteristic patient specific metabolic profiles after screening for >100 primary metabolites were observed which differed between normal embryos that had resulted in a viable pregnancy and aneuploid and chaotic embryos although it was more difficult to find clear patterns in embryos with monogenic disorders. Logistic regression analysis revealed a number of metabolites with high predictive value which in combination with embryo score could serve in the future as non-invasive markers for the detection of chromosomal abnormalities before embryo transfer. TEM analysis revealed differences in the quality of cells and organelle activity which were reflected in the embryo metabolic profiles. Abnormal but well developed hatching blastocysts had mainly cells with good mitochondrial morphology/ activity, nice Golgi apparatus and well developed rough and smooth endoplasmic reticulum but depending on the aneuploidy or gene mutation involved, inner cell mass cells with limited organelles and occasionally lipofuscin droplets in the trophectoderm were evident. Chaotic poor quality embryos showed a lower number of mitochondria, often with no cisternae, increased number of vacuoles, and frequently problems in junctions between cells. Limitations, reasons for caution Although metabolic profiles were compared between normal and abnormal embryos, all the normal embryos were transferred to the uterus or remain vitrified for clinical purposes. Therefore the ultrastructure analysis is based only on biopsied abnormal embryos and as control unbiopsied embryos from egg-donation cycles with high chance of being normal. Wider implications of the findings This study shows high implantation rates after PGT—M and PGT-A and identified distinct differences in the metabolic profiles of normal and abnormal embryos providing unique metabolites which in the future could serve as non-invasive biomarkers for the detection of abnormalities before embryo transfer. Trial registration number Not applicable
Read full abstract