Abstract #T435

# T435
Comparison of two methods to quantify free amino acids in cow plasma.
Nelson Lobos*1, Glen Broderick2, Nancy Whitehouse5, Daniel Luchini3, Michel Wattiaux1, Peter Crump4, 1Department of Dairy Science, University of Wisconsin-Madison, Madison, WI, 2Broderick Nutrition & Research LLC, Madison, WI, 3Adisseo S.A.S, Alpharetta, GA, 4Department of Computing and Biometry, University of Wisconsin-Madison, Madison, WI, 5Department of Biological Sciences, University of New Hampshire, Durham, NH.

The objective was to determine how measurements of plasma free AA by gas chromatography (GC) after chloroformate derivatization compared with values obtained by ion-exchange chromatography (HPLC) followed by ninhydrin derivatization. Plasma was obtained from lactating dairy cows (mean DIM = 148) part of a trial were Met was abomasally infused (12 or 24 g Met/d) or fed in rumen-protected form (15 or 30 g Met/d). Blood samples (n = 89) were collected from coccygeal vessels into evacuated tubes. After centrifugation, plasma was split into 2 aliquots and stored frozen at −80°C. Each set was analyzed either by HPLC in a commercial laboratory, or in-house by GC using a kit (EZ:faast, Phenomenex). This method involved deproteinization with TCA, AA collection using solid phase extraction, derivatization with chloroformate, GC separation in a capillary column and flame ionization detection. Quantitation was based on area under the curve, using the internal standard ratio method (Norvaline). Data were analyzed by Pearson correlation (r; Table 1). However Bland-Altman posited that correlation is misleading to compare methods and instead proposed the inspection of differrence plots and limits of agreement. Calculation of concordance correlation coefficients provided a quantitative alternative (CCC; Table 1). Overall correlation between methods was 0.95 (P < 0.001); nevertheless, correlations varied greatly for individual AA from 0.97 for Met to 0.15 for Glu. For Met, most data were within the limits of agreement established by Bland-Altman approach. Furthermore among all AA, CCC for Met was the closest to 1 (indicating almost total agreement between methods; Table 1). For research focusing on Met, GC offers a cheaper and faster alternative to HPLC. Table 1. Free AA (except Arg and Trp) in plasma measured by two methods
Amino acidGC (µM)HPLC (µM)rCCC
His50.262.80.420.18
Ile107.6101.60.960.94
Leu121.2145.20.920.36
Lys72.480.80.350.67
Met32.833.90.970.99
Phe40.046.10.650.37
Thr186.9117.90.960.25
Val348.0266.10.890.25
Ala351.7267.90.880.21
Asn44.057.10.840.26
Asp9.43.90.510.07
Cys18.921.00.350.56
Glu56.139.70.150.08
Gln110.6200.60.780.10
Gly463.3348.80.920.36
Pro102.281.40.840.25
Ser127.195.30.890.32
Tyr30.542.10.600.31

Key Words: amino acid, methionine, chromatography