Study Question: Can we separate embryos cultured under either 7% or 20% oxygen atmospheres by measuring their metabolic heterogeneity?, Summary Answer: Metabolic heterogeneity and changes in metabolic profiles in morula exposed to two different oxygen concentrations were not detectable using traditional fluorophore and two-channel autofluorescence but were detectable using hyperspectral microscopy., What Is Known Already: Increased genetic and morphological blastomere heterogeneity is associated with compromised developmental competence of embryos and currently forms the basis for embryo scoring within the clinic. However, there remains uncertainty over the accuracy of current techniques, such as PGS and time-lapse microscopy, to predict subsequent pregnancy establishment., Study Design, Size, Duration: The impact of two oxygen concentrations (7% = optimal and 20% = stressed) during post-fertilisation embryo culture was assessed. Cattle embryos were exposed to the different oxygen concentrations for 8 days (D8; embryo developmental competence) or 5 days (D5; metabolism measurements). Between 3 and 4 experimental replicates were performed, with 40-50 embryos per replicate used for the developmental competency experiment, 10-20 embryos per replicate for the fluorophore and two-channel autofluorescence experiments and a total of 21-22 embryos used for the hyperspectral microscopy study., Participants/materials, Setting, Methods: In-vitro produced (IVP) cattle embryos were utilised for this study. Post-fertilisation, embryos were exposed to 7% or 20% oxygen. To determine impact of oxygen concentrations on embryo viability, blastocyst development was assessed on D8. On D5, metabolic heterogeneity was assessed in morula (on-time) embryos using fluorophores probes (active mitochondria, hydrogen peroxide and reduced glutathione), two-channel autofluorescence (FAD and NAD(P)H) and 18-channel hyperspectral microscopy., Main Results and the Role of Chance: Exposure to 20% oxygen following fertilisation significantly reduced total blastocyst, expanded and hatched blastocyst rates by 1.4-, 1.9- and 2.8-fold, respectively, compared to 7% oxygen (P < 0.05), demonstrating that atmospheric oxygen was a viable model for studying mild metabolic stress. The metabolic profiles of D5 embryos was determined and although metabolic heterogeneity was evident within the cleavage stage (i.e. arrested) embryos exposed to fluorophores, there were no detectable difference in fluorescence intensity and pattern localisation in morula exposed to the two different oxygen concentrations (P > 0.05). While there were no significant differences in two-channel autofluorescent profiles of morula exposed to 7% and 20% oxygen (main effect, P > 0.05), morula that subsequently progressed to the blastocyst stage had significantly higher levels of FAD and NAD(P)H fluorescence compared to arrested morula (P < 0.05), with no change in the redox ratio. Hyperspectral autofluorescence imaging (in 18-spectral channels) of the D5 morula revealed highly significant differences in four features of the metabolic profiles of morula exposed to the two different oxygen concentrations (P < 0.001). These four features were weighted and their linear combination revealed clear discrimination between the two treatment groups., Limitations, Reasons for Caution: Metabolic profiles were assessed at a single time point (morula), and as such further investigation is required to determine if differences in hyperspectral signatures can be detected in pre-compaction embryos and oocytes, using both cattle and subsequently human models. Furthermore, embryo transfers should be performed to determine the relationship between metabolic profiles and pregnancy success., Wider Implications of the Findings: Advanced autofluorescence imaging techniques, such as hyperspectral microscopy, may provide clinics with additional tools to improve the assessment of embryos prior to transfer., Study Funding/competing Interest(s): This study was funded by the Australian Research Council Centre of Excellence for Nanoscale BioPhotonics (CE140100003). The Fluoview FV10i confocal microscope was purchased as part of the Sensing Technologies for Advanced Reproductive Research (STARR) facility, funded by the South Australian Premier's Science and Research Fund. The authors declare there are no conflict of interest., (© The Author 2017. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com)