Abstract #19

# 19
L. A. Agcanas*1, K. R. Counsell1, N. Shappell2, S. Bowers3, P. L. Ryan3,4, S. T. Willard1,3, C. K. Vance1, 1Biochemistry, Molecular Biology, Plant Pathology and Entomology, Mississippi State University, Mississippi State, MS, USA;, 2Biosciences Laboratory, USDA-ARS, Fargo, ND, USA;, 3Animal and Dairy Science, Mississippi State University, Mississippi State, MS, USA;, 4Pathobiology and Population Medicine, Mississippi State University, Mississippi State, MS, USA.

The capability of near-infrared spectroscopy (NIR) to detect biomolecules in aqueous solutions, a sub-field of NIR called aquaphotomics, has yet to be fully explored. Aquaphotomics references water absorbance patterns and wavelength shifts in the 1st overtone of the water spectrum as they change patterns with solute composition and concentrations. Recently, NIR was used as a rapid method for detecting oestrus in Holstein raw milk and for monitoring reproductive stages in urine of the Bornean Orangutan (Pongo pygmaeus) and the giant panda (Ailuropoda melanoleuca). Near-infrared spectroscopy detects bond vibrations from organic molecules and water to create unique absorbance patterns, which are used to profile complex mixtures of biomolecules. The objectives of this study are to (1) characterise serum NIR spectral profiles for oestrus, metestrus, and diestrus in mares, and (2) determine if NIR can accurately decipher these reproductive phases from serum due to the biochemical effects of reproductive hormones. Mare oestrus cycles were assessed every other day by ultrasound and serum hormone analysis. Serum was collected via jugular venipuncture on day 0, 2, 4, 8, and 16 of each cycle for 3 consecutive cycles from each mare. Radioimmunoassay analysis of reproductive hormones E2 and P4 was used to validate and characterise the NIR spectra. Spectra were collected from triplicate samples of 200 μL of serum in a 1-mm-path-length quartz cuvette with an ADS FieldSpec®3 spectrophotometer, at room temperature (22°C). Chemometric analysis (Unscrambler® X v.10.4; CAMO Software) included pretreatment with a Savitsky-Golay second-derivative function for visual inspection of spectral features and principal component analysis (PCA), after mean centering, for distinction of reproductive status. Spectral peaks at 1347, 1367, and 1465 nm were unique to serum collected from mares in oestrus exhibiting high E2 (11.87 to 16.88 pg/mL). Early metestrus is characterised by prominent spectral peaks at 1383 and 1437 nm, corresponding to E2 levels (0.02 to 5.29 pg/mL) and P4 levels (0.02 to 3.61 ng/mL). Diestrus peaks were found at 1342, 1426, and 1473 nm, when P4 levels ranged from 5.14 to 9.60 ng/mL. In our PCA models, 98% of the total variance in serum spectra between any pair of reproductive phases was described in only 3 spectral PCs, (PC1 = 83–86%, PC2 = 9–10%, PC3 = 2–5% of total variance). The PCA scores separated into distinct groups, indicating clear spectral profiles describing each reproductive phase uniquely. In conclusion, PCA analysis and spectral variances indicate that NIR, specifically aquaphotomics, has the theoretical capability to discriminate complex mixtures of biomolecules present in mare serum during oestrus, metestrus, and diestrus reproductive states.