The first mitotic division of human embryos is highly error prone.

Donation of human embryos to research The NHS Research Ethics

Committee approved both the research project (Indicators of Oocyte and Embryo Development, 04/Q2802/26) and egg-sharing-to-research programme (19/WM/0003). All work was conducted under a Research Licence from the Human Fertilisation and Embryology Authority (HFEA; R0155; Indicators of Oocyte and Embryo Development). Informed consent for donation of eggs, embryos and sperm to research was provided voluntarily, at an appointment with a research nurse, in advance of their treatment by patients undergoing in vitro fertilisation (IVF) or intracytoplasmic sperm injection (ICSI) at the Centre for Reproductive Medicine (CRM), University Hospitals Coventry and Warwickshire (UHCW) NHS Trust or at the Edinburgh Fertility and Reproductive Endocrine Centre (EFREC). Details of the clinic success rates are published by the Human Fertilisation and Embryology Authority ( https://www.hfea.gov.uk/choose-a-clinic/clinic-search/results/13/ ). Participation in research was optional and did not affect the treatment that the patient received. Patients were aware of the purpose of the research. Standard clinical protocols were used including ovarian stimulation with FSH preparations according to either GnRH agonist or antagonist regimens, and the Origio suite of culture media. None of the embryos was subject to preimplantation genetic testing or sex determination. The deselected material used for research was unsuitable for the patients’ treatment, due to delay (0PN) or mis-fertilisation (1PN/3PN), and would otherwise have been disposed of. 1PN embryos may be haploid, but if diploid, may arise from fusion of the parental pronuclei and can lead to live birth 53 . 3PN embryos usually form by fertilisation with two spermatozoa, or polar body extrusion failure 54 . They are associated with spontaneous abortions, however a euploid live baby has been born from a tripronuclear embryo 55 . Patients providing material unsuitable for use in their treatment received no compensation. Egg-share material was collected from volunteer female patients aged ≤33 on the egg sharing to research programme at the CRM only. These patients voluntarily elect to share half of their eggs with the research programme, and in return receive their treatment for a reduced cost (£300 for IVF or £770 for ICSI, rather than the standard cost of £2800 (IVF) or £3250 (ICSI), excluding medication), subsidised by the research funder. This is in keeping with UK law and approved by the NHS Research Ethics Committee and the Human Fertilisation and Embryology Authority, referenced above. Patients recruited into the egg share programme progressed through their IVF cycle in the normal way, but immediately after egg collection, their cumulus-enclosed eggs were allocated randomly to either their own treatment or to research. In the event of an odd number of eggs being collected, the patient received one more egg than the research project. The patient’s own treatment proceeded routinely using half the original number of eggs. The eggs allocated to research were denuded of cumulus cells. Mature (metaphase II) eggs were inseminated with fertile donor sperm via ICSI using standard clinical protocols, in order to create embryos for research use in this project. One sperm donor was used for all research inseminations. Following sperm injection, eggs were incubated in an Embryoscope TM time lapse incubator, in an identical manner to embryos used clinically. Immature eggs collected at the same time from egg sharers were used for research not included in this publication. Chromosome imaging of human embryos Embryos arising from the egg-share programme were collected from the clinic 1–2 h after the fertilisation check had been completed by a clinical embryologist, ~17 h after insemination. Deselected embryos were collected from the clinic around 3–5 h after the fertilisation check. Thus, none of the embryos that we used were cryopreserved. For SiR-DNA treatment, the zona pellucida of individual embryos was removed by brief incubation and pipetting in prewarmed acid Tyrode’s solution (Sigma). Embryos were transferred to a Fluorodish (WPI) containing 2 µM SiR-DNA (Spirochrome) diluted in prewarmed Cleav medium (Origio) under mineral oil. Embryos were transported ~14 km from UHCW CRM to Warwick Medical School (WMS) in a portable incubator (K Systems) held at 37 °C, and transferred to the microscope immediately upon arrival. Image stacks (60 × 1.5 μm optical sections; 1 × 1 binning) were acquired every 10 or 15 min for a 24–36 h period. The embryos were imaged with a ×40 oil-immersion 1.3 NA objective (Olympus) using a DeltaVision Elite microscope (Applied Precision, LLC) equipped with a CoolSNAP HQ camera (Roper Scientific). Fluorescent images were acquired using an InsightSSI solid state illuminator (Applied Precision, LLC) attenuated to 32% and a Cy5 filter set with an exposure time of 0.05 s. Brightfield images were acquired attenuated to 10% with an exposure time of 0.1 s. A stage-top incubator (INU; Tokai Hit) maintained embryos at 37 °C and 5% CO 2 with further stabilisation from a microscope enclosure (Weather station; Precision Control) held at 37 °C. The temperature was confirmed with a calibrated probe (Fluke 52). Image sequences were inspected and analysed by hand using OMERO (Open Microscopy Environment). Embryos from egg-sharer patient 3371 and mRNA injected embryos were imaged using Marianas spinning disk confocal microscope from 3i (Intelligent Imaging Innovations) equipped with 2× Photometrics 95B Prime sCMOS cameras, using a ×40 oil- immersion 1.46, alphaPlnApo (Zeiss) objective. Fluorescent images were acquired every 15 min (60 × 1.5 µm optical sections) with a 640 nm laser attenuated to 5% with an exposure time of 50 ms and brightfield images were acquired attenuated to 10% with an exposure time of 40 ms. A stage-top incubator (Okolab) maintained embryos at 37 °C and 5% CO 2 . Temperature and CO 2 were verified using calibrated probes. mRNA synthesis and injection For mRNA injections, the T7-H2B-mCherry plasmid (Addgene 20972) was linearised with SmaI. Capped mRNA was synthesised using mMessage mMachine™ T7 transcription kit (Invitrogen) and diluted to a concentration of 1 µg/µl. Zygotes were microinjected using Eppendorf FemtoJet® 4i and Femtotip II microinjection capillaries with the following parameters; Injection pressure 150 hPa, compensation pressure 35 hPa, injection time 0.6 s, and incubated at 37 °C and 5% CO 2 for 3 h before start of imaging. Fluorescent time lapse movies were acquired using a 3i Marianas spinning disk confocal microscope as described above, except using the 561 nm laser attenuated to 3% with an exposure time of 60 ms and 48 × 1.5 µm optical sections.

Immunofluorescence

Zygotes were monitored for pronuclei break down, then washed through warmed PHEM buffer (60 mM PIPES, 25 mM HEPES, 10 mM EGTA, 4 mM MgSO 4 .7H 2 O; pH 6.9) with 0.25% Triton X-100 at 37 °C, fixed with 3.7% paraformaldehyde in PHEM buffer with 0.25% Triton X-100 for 30 min and permeabilised in PBS with 0.25% Triton X-100 for 15 min at room temperature. Embryos were stored in PBS with 0.05% Tween-20 (PBST) until immunofluorescence was performed. For immunofluorescence, embryos were blocked in 3% BSA in PBST at 4 °C overnight, then incubated with antibodies against α-tubulin (Mouse 1:200; T6074, Sigma) and CENPC (guinea pig 1:200; MBL, PB030) in 3% BSA PBST at 4 °C overnight. Embryos were then washed for 3 × 20 min in 1% BSA PBST and further incubated with secondary antibodies; goat anti-mouse Alexa Fluor 555 (1:500; A-21422, ThermoFisher Scientific) and goat anti-guinea pig Alexa Fluor 647 (1:500; A-21450, ThermoFisher Scientific). Embryos were mounted in ProLong™ Gold Antifade Mountant with DAPI ( P36931 , ThermoFisher Scientific) on a FluoroDish (FD35-100, WPI). Samples were imaged using LSM980 laser scanning confocal equipped with an Airyscan detector (Zeiss UK, Cambridge) using a Plan-APO (×63/1.4 NA) oil objective (Zeiss). A 0.14 µm optical section spacing was used to encompass the area of interest. Also, 405, 561 and 639 nm lasers were used to detect DAPI staining and Alexa Fluor 555 and Alexa Fluor 647, respectively. Images were prepared using Fiji (National Institutes of Health). Clinical imaging and analysis of human embryos A cohort of 304 clinical human embryos were analysed in this study. These embryos were imaged on an EmbryoScope TM as part of patient ART treatment. Hoffman modulation contrast images are collected every 10 min for up to 6 days (blastocyst formation). These embryos fall into 3 categories: 80 singly transferred embryos which gave rise to clinical pregnancy, 76 singly transferred embryos which did not give rise to clinical pregnancy and 148 embryos (from the 80 patients who had single embryo transfers and became pregnant) which were not transferred or cryopreserved due to abnormal morphology (80 which had a normal first cytokinesis, dividing into 2 cells and 68 which had an abnormal cytokinesis, dividing into 3 or more cells). Clinical pregnancy is defined as foetal heartbeat detection at 7 weeks. The timing of events (PNBD, the start of cytokinetic furrow ingression and the appearance of 2 distinct cells) during the first cell division were determined and presented in Fig. 4b . Nuclear phenotype was further assessed in blastomeres of 80 embryos as they progressed from the 2-cell to the 4-cell stage and presented in Fig. 6 . These are the same 80 embryos that gave rise to clinical pregnancy in Fig. 4b . If nuclei were not visible in all blastomeres, the embryos were excluded from analysis in Fig. 6d . Each blastomere was visually examined for nuclear variants using multiple-focal planes from time lapse sequences. A FIJI macro was used to compile embryoscope data files ( https://github.com/Laura190/folders2s ). The diameter of all visible nuclei was measured using FIJI line tool and converted from pixels to µm in Fig. 6c . Data and statistical analysis Mann–Whitney U -tests for Fig. 5b , Fisher’s exact tests for Figs. 2b, c , 5d, e , and 6b, e , and Kruskal–Wallis test with post hoc Dunn’s test for Fig. 4b were performed using MATLAB R2020A (Mathworks) inbuilt functions. Gaussian distribution in Fig. 6c was plotted using EzyFit 2.44 toolbox. Equation: y ( x ) = a _1*exp(-( x - m _1) 2 /(2* s _1 2 ))+ a _2*exp(-(x- m _2) 2 /(2* s _2 2 )) 56 . Reporting summary Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Supplementary information Supplementary Information Peer Review File Description of Additional Supplementary Files Supplementary Movie 1 Supplementary Movie 2 Supplementary Movie 3 Supplementary Movie 4 Supplementary Movie 5 Supplementary Movie 6 Supplementary Movie 7 Supplementary Movie 8 Supplementary Movie 9 Supplementary Movie 10 Supplementary Movie 11 Supplementary Movie 12 Reporting Summary