This Small Business Innovation Research Phase I project is focused on developing an innovative algorithm for non-invasive bovine embryo viability assessment using near-infrared spectroscopy (NIR). Assisted reproductive technologies have been developed by the livestock industry to obtain large numbers of offspring from genetically superior animals. Embryos produced by in vitro fertilization (IVF) and somatic cell nuclear transfer (SCNT or cloning) could play a central role in cattle production systems by impacting the rates of genetic improvement and efficient dissemination of superior genotypes. Broad-based implementation of these techniques is limited by the inability to select high quality embryos using current subjective visual assessment. The availability of a quantitative embryo selection method able to distinguish the subtle differences that separate a healthy pre-implantation embryo from a poorly performing one would greatly increase the overall efficiency of IVF and SCNT technologies. The research objectives of this project are: identify the differences in NIR spectral characteristics from spent media resulting from embryo culture of viable and non-viable embryos; develop an algorithmic model using the NIR spectral characteristics; and demonstrate that this algorithm can improve embryo selection accuracy, increasing pregnancy rate by at least 50% compared to the current morphological assessment.
The broader impact/commercial potential of this project would be greater utilization of in vitro reproductive technologies by cattle breeders, which would have enormous impact on genetic improvement and reproductive management. Currently, a quantitative embryo assessment method in livestock is not available. If successful, the proposed innovative approach will increase pregnancy and calving rates by 50% compared to the present level and reduce the cost of a SCNT calf by $7,500. Reduced cost could substantially increase the technology utilization with a potential financial value of over $350 million. Additionally, the annual economic impact could exceed $100 million in dairy IVF in the U.S. alone. A successful result also will have a significant impact on herd health and help the livestock industry produce more offspring using fewer recipients, which will reduce pressure on the environment, decrease per unit climate emissions and land use, and meet future global nutritional needs. Such efforts will contribute to food security and improve the social and economic well-being of rural communities. The NIR spectral data from the project could lead to discoveries that enhance the scientific understanding of early embryogenesis and drive further progress in reproductive technology.
NON-INVASIVE METHOD FOR BOVINE EMBRYO VIABILITY ASSESSMENT USING SURFACE-ENHANCED RAMAN SPECTROSCOPY National Science Foundation Proposal # IP-1045319 This SBIR phase I project is focused on proving the feasibility of developing an innovative algorithm for non-invasive bovine embryo viability assessment using surface-enhanced Raman spectroscopy (SERS). The objectives are to identify the differences in SERS spectral characteristics from spent media resulting from embryo culture of viable and non-viable embryos, to develop an algorithmic model using the spectral characteristics and to demonstrate that the model can improve embryo selection accuracy. In the study, a total of 208 embryo spent medium samples were collected, representing 133 somatic cell nuclear transfer (SCNT) and 75 in vitro fertilized (IVF) embryos. The embryos were transferred after the spent medium collection and the pregnancy outcomes were used as phenotypic markers for embryo viability. SERS was performed on each spent medium sample using a glass capillary sample holder coated with gold-doped sol-gel matrix. Replicate experiments and multiple acquisitions of the spectra at different spots of the capillary resulted in over 3000 sets of SERS data collected which were used for chemometric analysis and viability model development. Distinct SERS spectral characteristics were found between viable and non-viable embryos as well as SCNT and IVF embryos and male and female embryos. Preliminary embryo viability models were developed from partial least squares discriminant analysis (PLSDA) using the binary (viable/non-viable) embryonic data. For IVF samples, the pregnancy status at day 60 was used as a phenotypic marker for embryo viability. The accuracy of the preliminary model to discriminate viable and non-viable IVF embryos was 0.65. For SCNT samples, when the pregnancy status at day 60 was used as a phenotypic marker for embryo viability, there was no significant discrimination found yet between viable and non-viable embryos. In contrast, when birth status was used as a phenotypic marker, the preliminary PLSDA model resulted in a discrimination of viable and non-viable SCNT embryos with an accuracy of 0.73. Our results prove that SERS is a promising and invaluable tool to achieve our goal of non-invasive assessment of bovine embryo viability. The study also identified the major challenges and the future efforts for the phase II study in which we aim to build a robust viability model with a higher accuracy by increasing the sample size, identifying and removing the spectral artifact and reducing the noise level. Successful selection of viable embryos will result in lower production cost for cattle producers and greater utilization of advanced reproductive technologies such as IVF and SCNT, making the cattle industry more productive and sustainable.
