Rapid noninvasive characterization of ovarian follicular atresia in cultured white sturgeon (Acipenser transmontanus) by near infrared spectroscopy

We report a rapid and noninvasive method based on short wavelength near infrared (SW-NIR) spectroscopy to detect onset of atresia in farmed white sturgeon (Acipenser transmontanus) (N = 10). The only current means to assess follicular atresia is by direct oocyte examination which requires a surgical biopsy of the ovary. In this study, abdominal scans were collected noninvasively by SW-NIR on anesthetized females using a diffuse reflectance fiber optic probe. In addition, to further verify the ability to detect spectroscopical changes related to onset of atresia, during each sampling roughly 30 cm3 of ovarian follicles was surgically removed from each female and transferred to a Teflon holder for spectral acquisition. Comparison of spectra collected on normal or atretic fish or ovarian follicles was conducted using Principal Component Analysis (PCA). Principal components suggest that the best indicator of atresia onset is a decrease in the intensity of the lipid bands at 930 nm. Prediction models for atresia were constructed using Soft Independent Modeling of Class Analogy (SIMCA) with leave-one-out cross validation. Seventy one percent of all atretic spectra and 65% of normal spectra collected on ovarian follicles were correctly classified. Exclusion of spectra from two potential outlier fish improved the predictability of normal ovarian follicles to 76%. Similarly, 72% of all atretic spectra and 59% of all normal spectra collected noninvasively in whole fish were correctly classified. Exclusion from the model of spectra from the same two outlier fish improved prediction of atresia from 72% to 75% as well as improving the prediction of normal spectra from 59% to 62%. This study represents the first example of using a noninvasive approach based on SW-NIR to detect onset of atresia in female sturgeon. Upon further development, this approach may potentially replace the need for surgical biopsy to detect ovarian regression. The availability of the proposed spectroscopic approach would grant sturgeon growers a powerful tool to follow more closely the maturation cycle with the goal of producing a consistently uniform product, standardize processing conditions, and maximize caviar yield by harvesting fish when the ovarian follicles have the appropriate firmness and a larger size.