Many plant secondary metabolites exhibit some degree of biological activity in humans. It is a common observation that individual plant-derived compounds in vivo are present in the nanomolar concentration range at which they usually fail to display measurable activity in vitro. While it is debatable that compounds detected in plasma are not the key effectors of bioactivity, an alternative hypothesis may take into consideration that measurable concentrations also reside in compartments other than plasma. We analysed the binding of constituents and the metabolite δ-(3,4-dihydroxy-phenyl)-γ-valerolactone (M1), that had been previously detected in plasma samples of human consumers of pine bark extract Pycnogenol, to human erythrocytes. We found that caffeic acid, taxifolin, and ferulic acid passively bind to red blood cells, but only the bioactive metabolite M1 revealed pronounced accumulation. The partitioning of M1 into erythrocytes was significantly diminished at higher concentrations of M1 and in the presence of glucose, suggesting a facilitated transport of M1 via GLUT-1 transporter. This concept was further supported by structural similarities between the natural substrate α-D-glucose and the S-isomer of M1. After cellular uptake, M1 underwent further metabolism by conjugation with glutathione. We present strong indication for a transporter-mediated accumulation of a flavonoid metabolite in human erythrocytes and subsequent formation of a novel glutathione adduct. The physiologic role of the adduct remains to be elucidated.

The present randomized controlled study aimed to investigate the in vivo distribution of constituents or metabolites of the standardized maritime pine bark extract Pycnogenol®. Thirty-three patients with severe osteoarthritis scheduled for a knee arthroplasty were randomized to receive either 200 mg per day Pycnogenol® (P+) or no treatment (Co) over three weeks before surgery. Serum, blood cells, and synovial fluid samples were analyzed using liquid chromatography coupled to tandem mass spectrometry with electrospray ionization (LC-ESI/MS/MS). Considerable interindividual differences were observed indicating pronounced variability of the polyphenol pharmacokinetics. Notably, the highest polyphenol concentrations were not detected in serum. Catechin and taxifolin primarily resided within the blood cells while the microbial catechin metabolite δ-(3,4-dihydroxy-phenyl)-γ-valerolactone, ferulic, and caffeic acid were mainly present in synovial fluid samples. Taxifolin was detected in serum and synovial fluid exclusively in the P+ group. Likewise, no ferulic acid was found in serum samples of the Co group. Calculating ratios of analyte distribution in individual patients revealed a simultaneous presence of some polyphenols in serum, blood cells, and/or synovial fluid only in the P+ group. This is the first evidence that polyphenols distribute into the synovial fluid of patients with osteoarthritis which supports rationalizing the results of clinical efficacy studies.

Blood cells, particularly erythrocytes, present a significant compartment for distribution of drugs and endogenous compounds and have been suggested to be factored in pharmacokinetic and pharmacodynamic evaluations. We previously detected the binding of polyphenols to red blood cells and found indications for a facilitated uptake of the bioactive procyanidin metabolite δ-(3,4-dihydroxy-phenyl)-γ-valerolactone (M1) into human erythrocytes. The purpose of the present investigation was to develop an effective, sensitive and robust liquid chromatography tandem mass spectrometry (LC-MS/MS) method to quantify low concentrations of polyphenols in human blood cells. Various sample preparation methods including classic sample clean-up techniques and variations of the QuEChERS (quick, easy, cheap, effective, rugged and safe) approach were compared regarding compound recovery, matrix effects and overall process efficiency. The QuEChERS technique which involves a liquid-liquid extraction and clean-up by dispersive solid-phase extraction yielded best results. The method was fully validated for the six analytes: (+)-catechin, ferulic acid, M1, taxifolin, caffeic acid and δ-3-methoxy-4-hydroxy-phenyl- γ-valerolactone (M2) in human blood cells with an optimised QuEChERS sample preparation and prior enzymatic hydrolysis of analyte conjugates. The lower limits of quantification for the analytes ranged from 0.12 ng/mL for M1, M2 and taxifolin to 48.40 ng/mL for caffeic acid. The application of the method to a blood cell sample of a volunteer ingesting 100 mg/day of the standardised pine bark extract Pycnogenol(®) over the course of 3 weeks revealed measurable steady-state concentrations of catechin, M1, taxifolin, ferulic acid and M2. To our knowledge, this is the first report of using the QuEChERS approach for detection and quantification of plant-derived compounds in human blood cells. The method can be applied in pharmacokinetic studies to determine the distribution of polyphenols and their metabolites in human whole blood, blood cells or erythrocytes. This might contribute in gaining deeper insights into the in vivo distribution of polyphenols and their metabolites.

No abstract