In human studies Pycnogenol revealed diverse anti-inflammatory actions [1]. Double-blind, placebo-controlled studies in asthma patients showed reduced plasma [3] or urine [4] leukotriene concentrations after Pycnogenol supplementation, while asthma symptom scores and pulmonary function improved. Symptoms of osteoarthritis as pain and immobility of joints decreased in a double-blind, placebo-controlled study [5]. Oral [6] and topical [7] application of Pycnogenol reduced inflammation and delayed skin-cancer formation following UV-radiation in controlled experiments in mice.

The anti-inflammatory mechanisms of maritime pine bark extract have been elucidated in a variety of in vitro and cell culture studies [8,9]. Additionally to its radical scavenging activity an inhibition of NF-κB-dependent gene expression and decrease of the activity of various pro-inflammatory mediators and adhesion molecules was observed after incubation of cells with the Pycnogenol extract [8,9]. This experimental in vitro design that pursues to uncover pharmacological effects by addition of plant extracts to cell cultures and subsequent measurement of cellular responses is widely employed. However, this methodology might inherit a couple of pitfalls.

Plant extracts often comprise of high molecular weight components that cannot be absorbed in the gastrointestinal tract and thus will never reach a target cell in vivo. Furthermore, there are examples of metabolites that are not present in the original extract, but are formed in vivo as a result of intestinal bacterial and/or hepatic metabolism. After ingestion of Pycnogenol, for example, two metabolites derived from catechin were detected in human urine, δ-(3,4-dihydroxy-phenyl)-γ-valerolactone and δ-(3-methoxy-4-hydroxy-phenyl)-γ-valerolactone [10]. Valerolactone derivatives were also found after ingestion of green tea [11]. These newly formed metabolites may display significant efficacy and contribute to the observed in vivo effects. We recently elucidated the cellular effects of δ-(3,4-dihydroxy-phenyl)-γ-valerolactone and δ-(3-methoxy-4-hydroxy-phenyl)-γ-valerolactone and uncovered an antioxidant activity as well as the potential to inhibit release and enzymatic activity of matrix metalloproteinase 9 (MMP-9) [12].

Thus, pharmacokinetic issues of absorption and metabolism should be considered for valid identification of molecular pharmacological effects of plant extracts. A methodological approach that considers both the absorption and possible metabolism of plant extract components would involve laboratory animals or human volunteers who donate blood samples. These blood samples should contain all bioavailable active principles of the extract and allow an ex vivo analysis in all kind of molecular pharmacological effects in cell culture assays (Figure ​(Figure1).1). There are only few examples of experimental settings described in literature that use this approach. Effects of nettle herb [13] or willow bark extract [14] on cytokine release and effect of Harpagophytum extract on eicosanoid biosynthesis [15] were elucidated in whole-blood assays of human volunteers after ingestion of the extract. Recently, a potent ex vivo anti-HIV activity was detected in sera of volunteers after administration of Phyllanthus amarus plant material [16].

A statistically significant decrease of MMP-9 concentration in cell culture supernatant was induced by the plasma samples obtained after intake of Pycnogenol compared to basal values (Figure ​(Figure2).2). The mean MMP-9 concentration after LPS challenge of monocytes incubated with volunteers' plasma samples obtained before Pycnogenol ingestion was 17.06 ± 2.17 ng/mL per 2.5 × 10⁵ viable human monocytes. This concentration was reduced to 12.70 ± 1.24 ng/mL when monocytes were incubated with plasma obtained after 5 days Pycnogenol intake. This corresponded to a mean decrease in MMP-9 concentration of 25 %. The plasma of all study participants exhibited an inhibitory effect on LPS-induced MMP-9 secretion, but interindividual variations were obvious. The inhibitory effect ranged from 4.6 % to 39 % inhibition.

A statistically significant reduction of nuclear p65 concentration was observed when human monocytes were exposed to plasma (dilution 1:1 with cell culture medium) obtained after intake of Pycnogenol compared to basal values (Figure ​(Figure3).3). For this experiment, sufficient volumes of plasma were only available from five volunteers; two plasma samples had been used up for repetition experiments after cell culture contamination. The mean nuclear p65 concentration after LPS challenge was 2.98 ± 0.48 ng per 1.5 × 10⁶ viable human monocytes. This nuclear concentration was reduced to 2.51 ± 0.26 ng per 1.5 × 10⁶ viable cells when monocytes were incubated with plasma obtained after 5 days Pycnogenol intake. The plasma of all study five participants exhibited an inhibitory effect on LPS-induced NF-κB activation, but interindividual variations were obvious. The inhibitory effect ranged from 6 % to 25 % with a mean of 15.5 % inhibition.

For five volunteers whose plasma samples were used for both the MMP-9 secretion and NF-κB experiments a correlation of their plasmas' inhibitory activity on MMP-9 secretion and NF-κB nuclear translocation was calculated (Figure ​(Figure4).4). The correlation (Spearman rank correlation coefficient) was positive (r = 0.6) though not statistically significant due to limited number of samples.

The standardized maritime pine bark extract Pycnogenol has documented clinical anti-inflammatory activities [1,9]. In earlier studies we determined that the extract's metabolites δ-(3,4-dihydroxy-phenyl)-γ-valerolactone and δ-(3-methoxy-4-hydroxy-phenyl)-γ-valerolactone exhibited inhibitory activity on LPS-induced secretion of matrix metalloproteinase 9 (MMP-9) from human monocytes [12]. However, so far it remained elusive whether sufficiently high in vivo concentrations of any active extract components would be achieved after peroral intake of Pycnogenol. In the present study we applied an ex vivo methodology that takes absorption and metabolism of the extract into account. The plasma samples obtained from volunteers after ingestion of Pycnogenol were expected to contain active extract components that should attenuate inflammatory processes.

Indeed, we observed a statistically significant mean decrease of about 25 % in MMP-9 release when LPS-activated human monocytes were exposed to plasma of volunteers after repeated intake of Pycnogenol. The matrix degrading enzyme MMP-9 is highly expressed at sites of inflammation and contributes to the pathogenesis of various chronic inflammatory diseases. In asthma MMP-9 is upregulated and involved in remodeling processes [18-20]. MMP-9 also facilitates recruitment of inflammatory cells such as eosinophils and neutrophils across basement membranes [18]. Expression of MMP-9 was negatively correlated with pulmonary function in asthmatic patients [20]. As Pycnogenol has been reported to attenuate signs of inflammation in asthma patients [3,4] we now provide first evidence that this anti-inflammatory in vivo effect might be at least partially attributed to reduced MMP-9 secretion on a molecular level.

NF-κB is a molecule with a master function in inflammatory cytokine induction. It is also involved in regulation of immune functions, cell cycle control and apoptosis [21]. Upon nuclear translocation in response to an inflammatory stimulus it regulates various genes coding for proinflammatory mediators. We found that plasma of Pycnogenol treated volunteers statistically significantly inhibited NF-κB activation in LPS-stimulated monocytes by about 15 %. Though this effect is rather moderate it might well contribute to the anti-inflammatory effects of Pycnogenol in patients. Interestingly, NF-κB is also involved in MMP-9 expression [22,23]. Consistent with these reports we observed a positive correlation between inhibitory activity of matched plasma samples on MMP-9 and NF-κB. In vitro inhibition of NF-κB activation by plant extracts or constituents has been reported repeatedly [24]. Blocking IκB kinase activity has been reported as the underlying mechanism for restricting NF-κB activation by green tea polyphenols [25]. The mechanism of ex vivo inhibition of NF-κB nuclear translocation by plasma containing bioavailable active principles after Pycnogenol ingestion has yet to be identified.

Z.C. recruited the volunteers and organized the study, prepared the technical documentation for blood sampling.

J.M. and K.S. took care of the volunteers and performed blood sampling and processed samples according to the protocol.

A.L. prepared the project and processed blood samples.

Z.D. contributed to planning of the design and execution of the project and wrote the ethic's committee application.

P.H. conceived of and designed the study and wrote the manuscript.