On-line sample preparation using restricted-access media in the analysis of the soy isoflavones, genistein and daidzein, in rat serum using liquid chromatography electrospray mass spectrometry
Dr. Doerge et al., On-line sample preparation using restricted-access media in the analysis of the soy isoflavones, genistein and daidzein, in rat serum using liquid chromatography electrospray mass spectrometry, RAP C MASS, 14(8), 2000, pp. 673-678
Soy isoflavones are the subject of many investigations in experimental anim
als and humans regarding possible modulation of endocrine activity and chem
oprevention of carcinogenesis. Genistein and daidzein, the principal biolog
ically active isoflavones in soy, were measured using on-line solid-phase e
xtraction (SPE) and liquid chromatography electrospray mass spectrometry (L
C/ES-MS) detection in serum of rats consuming a common open-formula (NIH 31
) chow that contained approximately 30 mu g each of genistein and daidzein
per gram of feed and a specially designed 'soy-free' chow that contained ap
proximately 60-fold lower isoflavones. The use of a restricted-access/rever
se phase trap cartridge and automated column switching permitted rapid and
robust analytical performance with many injections of plasma onto a reverse
phase LC column, Enzymatic deconjugation and a single centrifugation step
were the only sample preparation steps required, The limit of detection for
the isoflavones, based on the MS responses observed in serum from male and
female rats consuming the soy-free chow, was 0.020 mu M The method, which
uses deuterated isoflavones as internal standards, was determined to be acc
urate using spiked control serum (102-110% of added amounts) and precise us
ing spiked control serum and incurred serum (<6% relative standard deviatio
n), The average genistein and daidzein levels were determined in female (0.
62 and 0.25 mu M, respectively) and male rats (0.35 and 0.20 mu M, respecti
vely) consuming the standard diet. The sex difference observed for serum ge
nistein concentrations was statistically significant (p < 0.0001). These re
sults underscore the potential impact of standard open-formula diets on the
results from rodent bioassays of biological activity, Copyright (C) 2000 J
ohn Wiley & Sons, Ltd.