Consumption of Roasted Coffee Leads to Conjugated Metabolites of Atractyligenin in Human Plasma

Roasted coffee contains atractyligenin-2-O-β-d-glucoside and 3′-O-β-d-glucosyl-2′-O-isovaleryl-2-O-β-d-glucosylatractyligenin, which are ingested with the brew. Known metabolites are atractyligenin, atractyligenin-19-O-β-d-glucuronide (M1), 2β-hydroxy-15-oxoatractylan-4α-carboxy-19-O-β-d-glucuronide (M2), and 2β-hydroxy-15-oxoatractylan-4α-carboxylic acid-2-O-β-d-glucuronide (M3), but the appearance and pharmacokinetic properties are unknown. Therefore, first time-resolved quantitative data of atractyligenin glycosides and their metabolites in plasma samples from a pilot human intervention study (n = 10) were acquired. None of the compounds were found in the control samples and before coffee consumption (t = 0 h). After coffee, neither of the atractyligenin glycosides appeared in the plasma, but the aglycone atractyligenin and the conjugated metabolite M1 reached an estimated cmax of 41.9 ± 12.5 and 25.1 ± 4.9 nM, respectively, after 1 h. M2 and M3 were not quantifiable until their concentration enormously increased ≥4 h after coffee consumption, reaching an estimated cmax of 2.5 ± 1.9 and 55.0 ± 57.7 nM at t = 10 h. The data suggest that metabolites of atractyligenin could be exploited to indicate coffee consumption.


Supplemental information (SI)
The SI file contains the synthesis of 2,15-diketoatractyligenin with MS and NMR data; determination of the concentration of isolated metabolites M1 -M3 (Supporting Figure S1, Supporting Table S1, Supporting Table S2); Ion path parameters of the MS/MS system (Supporting Table S3); MRM traces of the analytes in plasma (Supporting Figure S2); back-calculated concentrations, precision and accuracy of matrix calibration curves for 1 -3 and M1 -M3 in plasma (Supporting Figure S7, Supporting Table S4); tabulated quantitative data in plasma samples from the coffee intervention study (Supporting Table S5); fragment spectra of M1 -M3 (Supporting Figure S4, Supporting Figure S5, Supporting Figure S6).
NMR and MS data were in accordance with the literature (Lang et al., 2013 Quantitative 1 H NMR to determine the concentration.The white residue of the evaporated 2,15diketoatractylgenin was dissolved in d4-methanol (2000 µl) to determine the concentration by quantitative 1 H NMR (qNMR).Signals used to calculate the concentration were the exocyclic protons of C17 at δ 5.92 and 5.34 ppm.The concentration was determined to be 2.66 µmol/ml.

Figure S3 .
Figure S3.Individual concentration-time plots of the study participants after coffee intervention (concentration <LloQ were substituted with zero).

Table S2 .
Results of the determination of the concentration of M1 -M3.

Table S3 .
ion path parameters for MS/MS detection.Entrance potential (EP) was -10 V for every analyte.Dwell time was 10 msec for each analyte.
Figure S2.Multiple Reaction Monitoring (MRM) traces of the coffee compounds 1 -3, compound 5, conjugated metabolites M1 -M3 isolated from coffee drinkers´ urine, and the internal standard 4 in blank plasma.Quantifier traces are solid; qualifiers are dotted and dashed (Table1).

Table S5 .
The concentration of 1 and M1 -M3 in human plasma after one dose of roasted coffee brew.Coffee compounds 2 and 3 were not detected in the samples.Note that none of the compounds were detected in any of the plasma samples from the control trial (data not shown).
a empty cells: no compound detected; b concentration below LloQ; c concentrations above LloQ