Investigating metabolic activity during oocyte and early embryo development through label-free metabolic imaging: a systematic approach for timelapse applications.

study questionIs it possible to assess label-free live cell metabolic imaging during early oocyte and embryo development? SUMMARY ANSWER: Label-free metabolic imaging can be systematically used during early development, showing no differences between controls and illuminated oocytes and embryos in terms of early development, blastocyst formation, and embryo outgrowth. WHAT IS KNOWN ALREADY: Non-invasive methods that are reliable to assess oocyte and embryo quality are a significant aim for ARTs. Changes in metabolic activity could lead to cell death or altered early development and low implantation potential. This could potentially be predicted by incorporating non-invasive measurements of metabolism. Metabolic imaging has been investigated through complex methodologies; however, scientific evidence for its utility during early oocyte and embryo development requires further investigation to assess potential translation in clinical settings. Measurements of metabolic activity could be a useful tool, as the autofluorescence of molecules such as nicotinamide adenine dinucleotide phosphate hydrogen (NAD(P)H) and flavin adenine dinucleotide (FAD) are a straightforward representation of mitochondrial function. STUDY DESIGN, SIZE, DURATION: Female mice (n = 15) and super-ovulated female mice (n = 30) were used to produce oocytes and embryos, respectively. Oocytes and in-vivo produced embryos were divided into the control group, sham control group, and illuminated group. Illuminated samples were assessed for both NAD(P)H and FAD levels in oocytes and NAD(P)H levels during early embryo development every 3 h using arbitrary units of autofluorescence (AU). Produced blastocysts were assessed for total cell and inner-cell-mass (ICM) number (by immunostaining for Oct4) and embryo outgrowth assays. Furthermore, safety live birth studies were also conducted. PARTICIPANTS/MATERIALS, SETTING,

methodsF1 (C57BL6/CBA) mouse strain was used. NAD(P)H and FAD autofluorescence levels were measured during oocyte and embryo development using confocal microscopy (Olympus FV1200). A confocal Z-stacking function was used to record 15 focal planes, using a 20×/0.95 NA air objective of the entire oocytes and embryos and opening the confocal pinhole system completely. Images were then collected and analysed using FIJI software (version: 2.0.0-rc-69/1.52n; ImageJ). Developmental rates, blastocyst cell numbers, outgrowth rates (for 4 days post blastocyst formation), and live birth rates were assessed. MAIN RESULTS AND THE ROLE OF CHANCE: Oocyte IVM and embryo culture experiments showed no significant differences in developmental rates between study groups (P > 0.05). Similarly, the total number of cells from blastocysts (control: 82.9 ± 5.6; sham: 76.5 ± 3.3; Illuminated: 77.1 ± 4.2; ± SEM) and ICM cells (control: 10.8 ± 1.3; sham: 9.4 ± 0.7; Illuminated: 11.9 ± 0.8; ± SEM) did not differ between groups (P > 0.05). Outgrowth assays of the study groups presented similar outgrowth areas during Days 5-8 (post) blastocyst development (P > 0.05). Illumination of oocytes demonstrated a significant increase in metabolic activity during IVM, measured by the optical redox ratio (ORR: FAD/NAD(P)H + FAD; P < 0.001). Illumination of embryos demonstrated significantly different NAD(P)H activity levels during embryo development, particularly between the two-cell stage (987.1 ± 36.2 AU), morula stage (1226.0 ± 31.5 AU) and blastocyst stage (649 ± 42.9 AU; ± SEM; P < 0.05). Additionally, embryos that did not form blastocysts also presented significantly decreased NAD(P)H activity levels at the two-cell stage (normal development: 987.1 ± 36.2; no blastocyst: 726.9 ± 121.7 AU; P < 0.05) to the morula stage (normal development: 1226.0 ± 31.5; no blastocyst: 886.0 ± 150.4 AU; P < 0.05) when compared with normally developing embryos. Our study indicated that metabolic imaging during early oocyte and embryo development presents no negative effects on developmental rates, blastocyst quality, and embryo outgrowths. Subsequently, live birth rates and offspring health showed no differences between controls and illuminated embryos at the blastocyst stage. Current results provide significant useful information about metabolic activity during live cell imaging as a potential method for timelapse metabolic imaging. LARGE SCALE DATA: N/A. LIMITATIONS, REASONS FOR CAUTION: The study was conducted using a mouse model and focused on early oocyte and embryo development, embryo outgrowths, live birth, and early offspring health. Thus, further studies of long-term offspring health are required to fully assess safety and to further validate potential wider applications. Validation in ageing models is also required to assess potential applications for embryo selection. WIDER IMPLICATIONS OF THE

findingsMeasurements of metabolic activity could be applied to determine oocyte and embryo metabolic activity using a variety of microscopy technology with low energy doses as described in this study. Further applications could link the use of metabolic imaging with timelapse technology and artificial intelligence applications to monitor culture conditions. STUDY FUNDING/COMPETING INTEREST(S): This study was funded in part by a research/educational grant from Ferring Pharmaceuticals, awarded from the Fertility Society of Australia and New Zealand (FSANZ). Funding was also provided in part by the Education Program in Reproduction and Development (EPRD), Department of Obstetrics and Gynaecology, Monash University. F.H. and M.H.-T. have applied for a patent in the topic of metabolic imaging. R.B.G. declares speakers' fees from Gedeon Richter and Ferring. The other authors have nothing to declare.