Facilitated cellular uptake and suppression of inducible nitric oxide synthase by a metabolite of maritime pine bark extract (Pycnogenol)

Free Radical Biology and Medicine 53 (2012) 305–313

Contents lists available at SciVerse ScienceDirect

Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed

Original Contribution

Facilitated cellular uptake and suppression of inducible nitric oxide synthase by a metabolite of maritime pine bark extract (Pycnogenol) n ¨ Klaus Uhlenhut, Petra Hogger

Institut f¨ ur Pharmazie und Lebensmittelchemie, Universit¨ at W¨ urzburg, 97074 W¨ urzburg, Germany

a r t i c l e i n f o

abstract

Article history: Received 9 November 2011 Received in revised form 1 April 2012 Accepted 13 April 2012 Available online 23 April 2012

Many natural products exhibit anti-inflammatory activity by suppressing excessive nitric oxide (NO) production by inducible NO synthase (iNOS). The maritime pine bark extract Pycnogenol has been formerly shown to decrease nitrite generation, taken as an index for NO, but so far it was not clear which constituent of the complex flavonoid mixture mediated this effect. The purpose of this study was to elucidate whether the in vivo generated Pycnogenol metabolite M1 (d-(3,4-dihydroxyphenyl)-gvalerolactone) displayed any activity in the context of induction of iNOS expression and excessive NO production. For the first time we show that M1 inhibited nitrite production (IC50 1.3 mg/ml, 95% CI 0.96–1.70) and iNOS expression (IC50 3.8 mg/ml, 95% CI 0.99–14.35) in a concentration-dependent fashion. This exemplifies bioactivation by metabolism because the M1 precursor molecule catechin is only weakly active. However, these effects required application of M1 in the low-micromolar range, which was not consistent with concentrations previously detected in human plasma samples after ingestion of maritime pine bark extract. Thus, we investigated a possible accumulation of M1 in cells and indeed observed high-capacity binding of this flavonoid metabolite to macrophages, monocytes, and endothelial cells. This binding was distinctly decreased in the presence of the influx inhibitor phloretin, suggesting the contribution of a facilitated M1 transport into cells. In fact, intracellular accumulation of M1 could explain why in vivo bioactivity can be observed with nanomolar plasma concentrations that typically fail to exhibit measurable activity in vitro. & 2012 Elsevier Inc. All rights reserved.

Keywords: RAW 264.7 macrophages Human monocytes EA.hy 926 Inducible nitric oxide synthase Nitric oxide Phloretin Free radicals

Introduction Many plant secondary metabolites exhibit some degree of biological activity in humans [1]. Various classes of phytochemical compounds exhibit anti-inflammatory effects by targeting key mediators of the complex inflammatory pathways [2,3]. Uncontrolled inflammatory processes often play a pivotal role in chronic or degenerative diseases such as arthritis, vascular disorders, or neurodegeneration. In the inflammatory network nitric oxide (NO) is an established effector mediating inflammatory cell damage. Whereas a low level of NO production is essential for maintaining homeostasis, excess generation of NO after de novo synthesis of inducible nitric oxide synthase (iNOS) is the hallmark of inflammatory disorders. Numerous plant extracts or extract constituents have been examined for their potential to inhibit NO production or suppression of iNOS expression. In this context, the most popular test system is the murine cell line of RAW 264.7 macrophages, which reliably express high levels of iNOS after activation with inflammatory agents [4]. Typically, the extent of nitrite production,

n

Corresponding author. Fax: þ 49 931 318 5494. ¨ E-mail address: [email protected] (P. Hogger).

0891-5849/$ - see front matter & 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.freeradbiomed.2012.04.013

taken as an index for NO, is determined in the cell culture supernatants. Indeed various phytochemicals decrease the nitrite generation in this test system. Regarding the concentrations for inhibiting nitrite production by 50% (IC50) whole-plant extracts often require higher concentrations (Z50 mg/ml [5–7]) for the effect compared with defined single constituents. Among those defined single compounds many exhibit activity in the concentration range 10–30 mM, whereas only few have reported IC50 values below 10 mM [8,9]. Another natural product being widely used for inflammatory conditions is French maritime pine bark extract [10,11], which is monographed as a food supplement in the United States Pharmacopoeia [12]. A standardized bark extract that complies with this monograph is derived from of Pinus pinaster Ait. (Pycnogenol; Horphag Research Ltd., UK). Pycnogenol was formerly investigated for effects on nitrite production and iNOS expression and found to decrease nitrite generation in murine RAW 264.7 macrophages by 40% at a concentration of 100 mg/ml [13]. In a test system with renal tubular cells even lower concentrations of Pycnogenol (10 mg/ml) exhibited distinct effects on nitrite production and iNOS expression [14]. Recently it was reported that Pycnogenol attenuated induced iNOS gene expression and NO production in chondrocytes [15]. Maritime pine bark extract is

306

K. Uhlenhut, P. H¨ ogger / Free Radical Biology and Medicine 53 (2012) 305–313

of dysfunctional induction of iNOS expression and excessive NO production and might thus present an active principle contributing to the extract’s anti-inflammatory effects. In this context we also sought to consider whether M1 concentrations detected in vivo would be consistent with M1 concentrations necessary to induce any effect in vitro. Therefore, we determined the disposition of M1 in the presence of cells that are present in the blood such as monocytes/macrophages and endothelial cells. Furthermore, we investigated the possibility of a facilitated uptake into cells by co-incubation of M1 with the influx inhibitor phloretin (Fig. 2). Phloretin, or dihydronaringenin (3-(4-hydroxyphenyl)-1-(2,4,6-trihydroxyphenyl)propan-1-one), is the aglycon of phlorizin, which is a dietary constituent found in various fruit trees [25]. Phloretin inhibits various membrane transporters such as GLUT-1 glucose transporter [26], red blood cell urea [27], or monocarboxylate transporters [28].

a complex mixture comprising 65%–75% procyanidins that consist of catechin and epicatechin subunits of varying chain lengths. Other constituents are flavonoids, phenolic or cinnamic acids, and their glycosides. One important factor that needs consideration when the bioactivity of plant extracts is investigated is the bioavailability of their constituents [16]. Clearly, highly condensed procyanidins cannot be absorbed and thus cannot interact with target cells in vivo. We recently analyzed plasma samples of volunteers after oral intake of single and multiple doses of Pycnogenol and found catechin, ferulic acid, caffeic acid, and taxifolin in addition to 10 other yet unidentified compounds [17]. We also detected a pine bark extract metabolite in plasma samples, d-(3,4-dihydroxyphenyl)-g-valerolactone, or 5-[(3,4-dihydroxyphenyl)methyl]oxolan-2-one according to IUPAC nomenclature, which we named M1. This metabolite was also found in urine samples after ingestion of Pycnogenol [18]. Earlier we showed that M1 is a bioeffective compound displaying anti-inflammatory and antioxidant activity [19]. Notably, M1 is not a constituent of the extract, but it is generated in vivo from catechin units by gastrointestinal microbial activity [20–24] (Fig. 1). This metabolite, M1, has also been detected in human urine after ingestion of green tea [24]. The purpose of this study was to elucidate whether the Pycnogenol metabolite M1 displays any activity in the context

HO

OH

OH

Phloretin

OH

O

Fig. 2. Structural formula of phloretin (dihydronaringenin).

OH HO

O

OH

Procyanidin

OH

OH

OH HO

O

OH OH

Interflavan-cleavage

OH

4,8'-methylenebis(2-(3,4-dihydroxyphenyl)chroman-3,5,7-triol)

OH

Catechin

B

O

HO A

OH

C OH

OH

2-(3,4-dihydroxyphenyl)chroman-3,5,7-triol

Opening of the C-Ring OH OH

HO

OH

Breakdown of the A-Ring

OH 2-(3-(3,4-dihydroxyphenyl)-2-hydroxypropyl)benzene-1,3,5-triol

OH

Lactone-formation

Valerolactone = M1

OH O O

OH

5-(3,4-dihydroxybenzyl)-dihydrofuran-2(3H)-one

Beta-oxidation OH

Benzoic acids HO 3,4-dihydroxybenzoic acid

OH O

Fig. 1. Proposed pathway for formation of metabolite M1 from flavan-3-ol structures in vivo by metabolizing microbiota in the human colon (based on [23,24]).

K. Uhlenhut, P. H¨ ogger / Free Radical Biology and Medicine 53 (2012) 305–313

Materials and methods Chemicals and reagents L-N

6

-(1-Iminoethyl)lysine  HCl (L-NIL) was obtained from Enzo Life Science (Lausen, Germany). Polyclonal rabbit iNOS antibody, type II, was bought from Becton–Dickinson (Rutherford, NJ, USA). Rabbit b-tubulin III antibody; hydrocortisone; lipopolysaccharide (rough strains), Salmonella Minnesota Re 595 (LPS); sulfanilic acid; N-(1-naphthyl)ethylenediamine hydrochloride; 3-[(3-cholamidopropyl)dimethylammonio]-1-propane sulfonate; DL-dithiothreitol; phloretin; and neutral red were all purchased from Sigma–Aldrich (Taufkirchen, Germany). Metabolite M1 was synthesized by Matthias Rappold in the course of his diploma thesis [29]. Fraction I of maritime pine bark extract (Pycnogenol; Horphag Research, UK) was prepared by A. Schulze Elfringhoff (Institute of Pharmaceutical and Medicinal Chemistry, University ¨ of Munster) according to the method described by Blazso´ et al. [30]. Acetonitrile (gradient grade) was purchased from VWR (Darmstadt, Germany). Ultrapure Milli-Q water was used for all preparations. Buffer salts and other reagents were obtained from Sigma–Aldrich if not stated otherwise. Cell culture The murine macrophage cell line RAW 264.7 (TIB-71) was obtained from the American Type Culture Collection (Rockville, MD, USA). The EA.hy 926 human endothelial cell line was a kind gift from Dr. Cora Jean Edgell (University of North Carolina, Chapel Hill, NC, USA). Cells were kept in Dulbecco’s modified Eagle medium (DMEM) containing 4.5 g/L glucose without phenol red, supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 mg/ml streptomycin, 2 mM L-glutamine, 1 mM nonessential amino acids, all purchased from Biochrom AG (Berlin, Germany). Cells were grown in 75-cm2 tissue culture flasks (Nunc, Roskilde, Denmark) at 37 1C in a humidified atmosphere of 5% CO2. Medium was changed every second day. At confluence macrophages were removed from the flasks using a rubber policeman, seeded in six-well plates (Nunc) at initial cell counts of 3  106 per well, and allowed to adhere. Endothelial cells were removed from flasks by trypsinization and seeded in six-well plates (Nunc) at initial cell counts of 0.4  106 per well, allowed to adhere, and grown to confluence. Cell counting and routine vitality testing were carried out using trypan blue solution (Biochrom AG) and an inverted microscope (TS 100; Nikon ¨ GmbH, Dusseldorf, Germany). Fresh human monocytes were obtained from a local blood ¨ bank (Transfusionsmedizin, Wurzburg, Germany). Cells were washed and frozen immediately with 40% FBS and 10% dimethyl sulfoxide (DMSO) in freezing medium (DMEM; Biochrom AG) at a cooling rate of 1 1C/min and kept 1 day at  80 1C. After that the cells were transferred to a nitrogen-cooled tank at  196 1C for further storage. With one freeze/thaw cycle under such conditions the viability of the cells after rapid unfreezing in a water bath at 37 1C was greater than or equal to 50% and vital cells were used for the following experiments. Neutral red assay Detection of potential cytotoxic impact on cells and their membrane systems was carried out using the neutral red assay as previously described [31,32]. Briefly, RAW 264.7 cells were plated at densities of 0.3  106 per well in 96-well cell culture plates (Nunc) and incubated with the designated compounds for 14 h at 37 1C. DMSO (10% v/v) and 50 mg/ml ZnCl2 served as positive control. After being washed with fresh cell culture

307

medium the cells were incubated with 100 ml of neutral red solution (40 mg/ml in culture medium) for 3 h at 37 1C. Neutral red solution must be kept overnight at 37 1C and centrifuged at 10,000 g for 10 min before use to avoid interference due to suspended dye crystals. After another washing step the dye was extracted with 150 ml of extraction solution consisting of 50% ethanol (96%), 49% Milli-Q water, and 1% glacial acetic acid and incubated for 10 min on a horizontal shaker (Unimax 1010; Heidolph, Schwabach, Germany). Absorption measurements of the extracted dye were performed at 490-nm wavelength employing a microplate reader (Multiskan Ascent, Thermo Fisher Scientific, Waltham, MA, USA). Determination of nitrite with a modified Griess reaction RAW 264.7 cells were seeded in six-well plates at initial cell counts of 3  106 cells per well, allowed to adhere, grown to confluence for one additional day, and then preincubated with 0.1–50 mg/ml M1, 1 mg/ml Fraction I, 36.25 mg/ml hydrocortisone, and 1 mM L-NIL for 10 min. Afterward 1 mg/ml LPS was added for a concomitant incubation period of 14 h. Nitrite determination in the supernatants was carried out as described previously [33] with modifications. NaNO2 was used for calibration. Briefly, 200 ml of the supernatants was harvested and mixed with 50 ml of 6.5 M HCl and 50 ml of 37.5 mM sulfanilic acid solution in Milli-Q water. Thereafter the autoabsorption values (which were subtracted from determined values after final addition of N-(1-naphthyl)ethylenediamine hydrochloride) of the resulting solution were monitored at 600 nm (Multiskan Ascent; Thermo Scientific), as many flavonoids are unstable in aqueous solutions and show dynamic autoabsorption and autofluorescence. Then 50 ml of 12.5 mM N-(1-naphthyl)ethylenediamine hydrochloride solution in Milli-Q water was added and after incubation in the dark for 10 min absorption was monitored at 600 nm (Multiskan Ascent; Thermo Scientific). All results as net absorptions were normalized on total protein content of the cells after the incubation period determined by the Bradford method [34]. Western blotting Western blotting experiments were based on methods for detecting nitric oxide synthases that were reported previously [35]. For the determination of iNOS and b-tubulin protein expression in murine macrophages (incubated with M1 1–50 mg/ml, Fraction I (FI) 1 mg/ml, hydrocortisone (HC) 36.25 mg/ml for 14 h) total protein was isolated. Lysis buffer (pH 7.5) consisted of Tris–HCl 50 mM, EDTA 0.5 mM, EGTA 0.5 mM, DL-dithiothreitol 2.0 mM, glutathione (reduced) 7.0 mM, 3-[(3-cholamidopropyl)dimethylammonio]-1-propane sulfonate 20 mM, glycerol 10% (v/v). Shortly before the lysis buffer was used, the protease inhibitor mix Complete (Roche, Basel, Switzerland) was added. Immediately after a wash with ice-cold PBS lysis buffer the cells were scraped off the plate using a rubber policeman. The cell suspensions were incubated for 15 min at 40 1C followed directly by sonication in an ultrasonic bath (VWR, Darmstadt, Germany) for 35 s. Homogenates were centrifuged at 16,000 g for 20 min at 4 1C. Supernatants were harvested and protein content was determined using the Bradford method as described earlier [34]. Samples were mixed with reducing electrophoresis loading buffer (Carl Roth, Karlsruhe, Germany) and heated to 70 1C for 8 min. Thirty micrograms of each protein sample were separated on SDS-7.5% polyacrylamide gels. To visualize separation and scaling of protein size a prestained protein marker, PageRuler Plus, was applied (Fermentas, Thermo Scientific). Proteins were transferred to nitrocellulose membranes (Biosciences Whatman Filtration, Schleicher & Schuell, Maidstone, UK) using a semidry-blotting

K. Uhlenhut, P. H¨ ogger / Free Radical Biology and Medicine 53 (2012) 305–313

1.25 1.00 0.75 0.50 0.25

L

L 50

µg

µg

/m

/m

L 25

µg

/m

/m 10

µg 1

10 0

ng

m

/m

L

L

M

0.00

1

The HPLC system consisted of a binary pump Model 1525, an autosampler Model 717plus, and a UV/vis detector Model 2487 from Waters (Milford, MA, USA). Separations were carried out on a C18 Sunfire column (150  4.6-mm i.d., 5 mm) from Waters. The mobile phase consisted of 10 mM phosphate-buffered saline of pH 2.0 with 14% acetonitrile. An isocratic flow rate of 1 ml/min was used. M1 absorption was monitored at 279-nm wavelength and the data acquisition and analysis were performed with Waters’ Breeze software version 3.3. Before analysis cell culture media containing different concentrations of M1 were mixed with 10% solution of trichloroacetic acid and centrifuged for 10 min at 10,000 g at 4 1C (Mikrofuge 22 R; Beckmann Colter, Krefeld, Germany). Typically, 200 ml of the supernatant was immediately

The inhibition of NO production by M1 from LPS-stimulated murine RAW 264.7 macrophages was analyzed by quantifying the nitrite concentration in the cell culture supernatants (Fig. 3). The inhibition was dose-dependent and ranged from 15.45 70.17 (0.1 mg/ml M1) to 0.6870.85 mmol nitrite per gram protein (50 mg/ml M1) in relation to uninhibited control values of 16.1079.06 mmol nitrite per gram protein. The specific iNOS inhibitor L-NIL reduced nitrite production to 1.38 70.72 mmol per gram protein. The concentration of M1 that inhibited nitrite release by 50% (IC50) was 1.3 mg/ml M1 (95% CI 0.96–1.70) as determined by fitting results with sigmoidal dose–response curves. Because the absolute nitrite values revealed a considerable degree of variability between single experimental runs, the data have been normalized and expressed as fold of control with respect to the control value of the individual assay. Thus, M1 inhibited nitrite production from 0.9970.02 (0.1 mg/ml) to 0.0370.02 (50 mg/ml) with reference to control values (1.0). To control for cytotoxicity of M1 (0.1–100 mg/ml) toward murine RAW 264.7 macrophages cell viability was determined over 14 h using the neutral red assay. DMSO (10% v/v) and ZnCl2 (50 mg/ml) were used as positive controls. Untreated cells subjected to the same work flow served as negative controls (Fig. 4).

IL

HPLC method for rapid quantification of M1 in cell culture supernatants

Inhibition of NO production by the pine bark extract metabolite M1

C o

Human monocytes and murine RAW 264.7 macrophages were suspended in complete cell culture medium containing 10% FBS at a density of 5  106 vital cells per milliliter. Thereafter the cells were incubated either with 200 ng/ml M1 alone or with 200 ng/ ml M1 after preincubation with 55 mg/ml phloretin for 10 min. In parallel, control experiments were carried out accordingly for each variable without cells. Cell suspensions of 1 ml were aliquotted in closed reaction tubes (Eppendorf, Hamburg, Germany) and incubated at 37 1C on a horizontal shaker run at 70 rpm (Heidolph). At distinct time points (t ¼0, 5, 15, 30, 60, 120 min) cell suspensions and matched controls were centrifuged for 2 min at 2000 g and the supernatants were harvested and immediately analyzed by high-performance liquid chromatography (HPLC). Human EA.hy 926 endothelial cells were seeded out in six-well plates (Nunc), allowed to adhere, and grown to confluence as described above. Thereafter the cells were washed with culture medium and the plates were put on a horizontal shaker run at 70 rpm (Heidolph) while cells were incubated at 37 1C either with 200 ng/ml M1 alone or with 200 ng/ml M1 after preincubation with 55 mg/ml phloretin for 10 min in a final volume of 1 ml culture medium with 10% FBS. Controls without cells were treated accordingly. At distinct time points (t¼0, 5, 15, 30, 60, 120 min) samples of the supernatants were drawn, centrifuged for 2 min at 2000 g, and immediately analyzed by HPLC. Results were expressed as the ratio of M1 values from experiments with cells as numerator and control experiments without cells as denominator.

Results

N

Measurements of M1 binding/uptake into macrophages, monocytes, and endothelial cells

subjected to HPLC analysis. Retention time for M1 was tR ¼11.62 70.06 min. Linearity was proven between 0.031 and 2 mg/ml M1 in culture medium by analyzing seven equidistant concentration levels (r2 ¼0.999; slope¼1.92670.008; y-intercept  0.01770.007 from best-fit values). The lower limit of quantification for M1 was 0.031 mg/ml M1 with VK (coefficient of variation) values for accuracy of 100.1% and precision of 3.5%. Intraday accuracy and precision VK values for M1 ranged from 94.4 to 102.4 and 1.6 to 2.5%, respectively. Accordingly, interday accuracy and precision VK values comprised 94.8%–99.9 and 4.2%–5.2%.

L-

technique (Peqlab, Erlangen, Germany). Blots were blocked with 3% (m/v) bovine serum albumin (BSA) and 0.05% (v/v) Tween 20 in TBS-T (Tris 0.025 M, NaCl 0.19 M, pH 7.4) overnight at 4 1C. The membrane was cut in half at the 70-kDa protein size threshold. The upper membrane part was incubated with iNOS antibody diluted 1:1250 in TBS-T with 3% BSA and 0.02% sodium azide and the lower with b-tubulin antibody diluted 1:3000 for 6 h at room temperature on a horizontal shaker (Heidolph). Extensive washing of the blots with TBS-T was followed by incubation with secondary goat anti-rabbit antibody (SuperSignal West Pico rabbit IgG detection kit; Thermo Scientific) diluted 1:30,000 in TBS-T with 3% BSA for 1 h at room temperature. Blots with immunocomplexes were developed with peroxidase/luminol enhanced chemiluminescence (ECL) reagent (Thermo Scientific) and the proceeding ECL reaction was monitored using the ChemiDoc XRSþ system (Bio-Rad, Hercules, CA, USA). Densitometric semiquantitative analysis of protein bands was carried out using the open source NIH program ImageJ.

Inhibition of nitrite formation (x -fold control)

308

Metabolite M1 Fig. 3. Dose-dependent inhibition of nitrite formation in cell culture supernatants from M1-treated murine macrophages (RAW 264.7). Cells were preincubated with M1 (0.1–50 mg/ml) and 1 mM L-NIL as control for specific iNOS inhibition for 10 min before 1 mg/ml LPS was added. After 14 h the nitrite concentration in the supernatants was determined using the Griess reaction and the resulting values were normalized with respect to the protein concentration. Untreated cells subjected to the same work flow served as control (Co). Columns represent mean and standard deviation of quadruplicate samples.

K. Uhlenhut, P. H¨ ogger / Free Radical Biology and Medicine 53 (2012) 305–313

1.25

Inhibiton of nitrite formation (x -fold control)

1.25 1.00 0.75 0.50 0.25 0.00

1.00 0.75 0.50 0.25

To explore whether the decrease in NO production by M1 was related to an inhibitory effect on iNOS expression cell lysates were subjected to Western blotting and probed for iNOS protein and the housekeeping protein b-tubulin (Fig. 6). Protein bands of iNOS were immunodetected around 130 kDa and they revealed decreasing intensity under increasing concentrations of M1 (1–50 mg/ml). Densitometric analysis of the blots confirmed a dose-dependent inhibition of iNOS expression by M1. The iNOS expression was decreased by about 24% with 1 mg/ml M1 (0.7670.19 versus control 1.00). Higher M1 concentrations of 10, 25, and 50 mg/ml

/m L

Fr

ac tio n

M 1

H

C

Downregulation of iNOS by the pine bark extract metabolite M1

1.50 1.25 1.00 0.75 0.50 0.25

L

L µg /m

µg /m 50

L 25

/m µg 10

1

µg /m

L

+

-

0.00 C o

The cell viability was slightly improved at M1 concentrations of 0.1 (1.2170.04 versus control 1.0070.01) and 1 mg/ml (1.1170.03) while being comparable to control values at a concentration of 10 mg/ml (1.0170.07). M1 at a concentration of 25 mg/ml was still not toxic (0.9670.07), whereas higher concentrations of 50 (0.85 70.02) and 100 mg/ml (0.70 70.07) reduced the macrophage viability. For comparison, ZnCl2 and DMSO both exhibited strong toxicity (0.33 70.01 and 0.177 0.004, respectively). From these experiments it is obvious that M1 concentrations up to 25 mg/ml did not induce any cytotoxicity above control values. The effect of M1 was analyzed in comparison with the antiinflammatory corticosteroid hormone HC and FI of the maritime pine bark extract Pycnogenol (Fig. 5). Fraction I contains monomeric flavonoids and polyphenols featuring a sufficiently low molecular weight to be absorbed in vivo and be present in blood after ingestion of Pycnogenol. Hydrocortisone at a concentration of 35.25 mg/ml inhibited nitrite production by 43% to 10.5371.16 mmol per gram protein, Fraction I (1 mg/ml) reduced nitrite by 23% to 14.16 71.11 mmol per gram protein. The most pronounced effect was seen with M1 (0.1 mg/ml), which reduced nitrite production by approximately 50% to 9.2171.45 mmol per gram protein. With the positive control (1 mM L-NIL) 1.637 0.48 mmol nitrite per gram protein was determined, for untreated cells (negative control) 18.37 70.80 mmol nitrite per gram protein was measured. Data were normalized and expressed as fold of control with respect to the control value of the individual assay.

Fig. 5. Inhibition of nitrite formation in cell culture supernatants of murine RAW 264.7 macrophages pretreated with M1 (36.25 mg/ml), hydrocortisone (HC; 1 mg/ ml), Pycnogenol Fraction I (1 mg/ml), or L-NIL (1 mM) for 10 min before 1 mg/ml LPS was added. After 14 h the nitrite concentration in the supernatants was determined using the Griess reaction and the resulting values were normalized with respect to the protein concentration. Untreated cells subjected to the same work flow served as control (Co). Columns represent mean and mean deviation of the mean of triplicate samples.

iNOS expression (x -fold control)

Fig. 4. Cytotoxicity of M1 toward murine RAW 264.7 macrophages over a period of 14 h. Cells were treated with M1 (0.1–100 mg/ml) and cell viability was determined using the neutral red assay. DMSO (10% v/v) and ZnCl2 (50 mg/ml) served as positive controls. Untreated cells subjected to the same work flow served as negative controls (Co). Columns represent mean and mean deviation of the mean of triplicate samples.

I1

1

5 36 .2

N L-

Metabolite M1

µg

/m L µg

/m L µg

m 1 IL

L

/m

µg 50

M

C o

C o Zn D C M SO 2l 1 10 0 % 0 ng /m L 1 µg /m 10 L µg /m L 25 µg /m L 50 µg / 1 0 mL 0 µg /m L

0.00

C o

Cytotoxicity (x -fold control)

309

Metabolite M1 Fig. 6. Representative Western blot from lysates of murine RAW 264.7 macrophages left untreated or treated with M1 (1–50 mg/ml) followed by a co-incubation with LPS (1 mg/ml) for 14 h. Immunodetected iNOS displayed protein bands around 130 kDa, the housekeeping gene b-tubulin around 50 kDa. The densitometric analysis of iNOS protein bands with reference to b-tubulin protein expression is shown at the bottom. Data are reported as ratios referred to maximal stimulated control values (Co þ). Untreated cells subjected to the same work flow served as negative control (Co  ). Columns represent mean and mean deviation of the mean of triplicate samples.

further reduced the detectable iNOS protein (0.2970.17, 0.1570.10, and 0.0570.02, respectively). The IC50 for inhibition of iNOS expression was 3.8 mg/ml M1 (95% CI 0.99–14.35) as

K. Uhlenhut, P. H¨ ogger / Free Radical Biology and Medicine 53 (2012) 305–313

The binding and uptake of M1 were determined in murine RAW 264.7 macrophages (Fig. 8A) and primary human monocytes (Fig. 8B) in the presence and absence of the membrane transporter inhibitor phloretin. Compared to control incubations without cells M1 was bound or taken up, respectively, over the experimental period of 120 min. In the presence of murine RAW 264.7 macrophages the M1 concentrations decreased by 57% from a ratio (sample/control) of 1.0370.01 at t ¼0 to 0.4470.09 after 120 min (Fig. 8A). In the presence of phloretin the M1 concentration decreased comparably by 48% from a ratio (sample/control) of 0.9670.01 at t ¼0 to 0.5070.04 after 120 min. However, the co-incubation with phloretin decreased the M1 concentration considerably more slowly between 15 and 60 min, and after 60 min ratios of 0.5570.11 were determined in the absence and 0.7570.01 in the presence of phloretin. The difference between binding and uptake of M1 into human monocytes was even more distinct with and without the influence of phloretin (Fig. 8B). M1 concentrations decreased by 92% from a ratio (sample/control) of 0.9270.01 at t ¼0 to 0.07 70.01 after 120 min. In the presence of phloretin the M1 concentration decreased by 61% from a ratio (sample/control) of 0.9770.07 at t ¼0 to 0.3870.06 after 120 min. Between t ¼0 and t ¼15 min phloretin almost completely inhibited the decrease in M1 concentration. When the M1 value ratios of the control experiments from uptake of M1 in human monocytes and murine macrophages in the absence of cells were plotted they revealed no difference over the duration of the assay (Fig. 8C). This indicated that the M1 concentration in the presence or absence of phloretin was not different in both experimental series and thus suggests well comparable experimental conditions. The binding and uptake of M1 was also determined in human EA.hy 926 endothelial cells in the presence and absence of phloretin (Fig. 9). The M1 concentrations decreased by 76% from a ratio (sample/control) of 1.01 70.01 at t¼0 to 0.2470.05 after 120 min. In the presence of phloretin the M1 concentration decreased by 62% from a ratio (sample/control) of 1.00 70.02 at t ¼0 to 0.3870.14 after 120 min. In these experimental series the variability of the mean data was higher than seen in the assays with macrophages and monocytes.

Fig. 7. Representative Western blot of lysates of murine RAW 264.7 macrophages left untreated or treated with M1, FI (both 1 mg/ml), or HC (36.25 mg/ml) followed by co-incubation with LPS (1 mg/ml) for 14 h. Immunodetected iNOS displayed protein bands around 130 kDa, the housekeeping gene b-tubulin around 50 kDa.

Binding to RAW 264.7 cells (ratio sample/Co)

M1 binding to macrophages, monocytes, and endothelial cells

1.25 M1 + Phloretin M1

1.00 0.75 0.50 0.25 0.00 0

30

60 time [min]

90

120

1.25 Binding to human monocytes (ratio sample/Co)

determined by fitting results with sigmoidal dose–response curves. The inhibitory effect of M1 on iNOS expression was confirmed when cells were incubated with M1 in comparison with hydrocortisone and Fraction I (Fig. 7). M1 (1 mg/ml) decreased the iNOS expression more pronouncedly compared to Fraction I at the same concentration. Hydrocortisone (36.25 mg/ml) was slightly more potent in decreasing detectable iNOS protein compared to 1 mg/ml M1.

M1 + Phloretin M1

1.00 0.75 0.50 0.25 0.00 0

30

60 time [min]

90

120

1.25 1.00 ratio controls

310

0.75 0.50 0.25

Co M1/Co M1+ Phloretin (Monos) Co M1/Co M1+ Phloretin (RAWs)

0.00 0

30

60 time [min]

90

120

Fig. 8. (A) Uptake of M1 into murine RAW 264.7 macrophages. Cells were suspended in culture medium containing 10% FBS and incubated at 37 1C with 200 ng/ml M1 without or with 55 mg/ml phloretin, which was preincubated with the cells for 10 min. In parallel, control (Co) experiments were performed without cells. Remaining concentration of M1 in the supernatants was determined by HPLC. Data points represent mean and standard deviation of triplicate samples. (B) Uptake of M1 into primary human monocytes. Experiments were performed as described for (A). Data points represent mean and standard deviation of triplicate samples. (C) M1 value ratios of control experiments from uptake of M1 in human monocytes (Monos) and murine macrophages (RAWs). Data represent values in reference to the initial time point (t ¼ 0) expressed as ratio of controls with M1 (200 ng/ml) and of controls with M1 (200 ng/ml) plus phloretin (55 mg/ml). Data points represent mean and standard deviation of triplicate samples.

Discussion In this study we show for the first time that nitrite production, taken as an index for NO generation, and iNOS expression were

K. Uhlenhut, P. H¨ ogger / Free Radical Biology and Medicine 53 (2012) 305–313

Binding to EA.hy 926 cells (ratio sample/Co)

1.25

M1 + Phloretin M1

1.00 0.75 0.50 0.25 0.00 0

30

60 time [min]

90

120

Fig. 9. Uptake of M1 into human EA.hy 926 endothelial cells. Cells were seeded in six-well plates and were incubated at 37 1C with 200 ng/ml M1 without or with 55 mg/ml phloretin, which was preincubated with the cells for 10 min. In parallel, control (Co) experiments were performed without cells. Remaining concentration of M1 in the supernatants was determined by HPLC. Data points represent mean and standard deviation of triplicate samples.

inhibited in a concentration-dependent fashion by the flavonoid metabolite M1. However, these effects required application of M1 in the low-micromolar range, which was not consistent with concentrations previously detected in human plasma samples after ingestion of maritime pine bark extract. Thus, we investigated a possible accumulation of M1 in cells and observed strong binding of this flavonoid metabolite to macrophages, monocytes, and endothelial cells. This binding was distinctly decreased in the presence of phloretin, suggesting the contribution of a facilitated M1 transport into the cells. Using murine RAW 264.7 macrophages we analyzed the nitrite concentration in the cell culture supernatants after exposing the cells to bacterial LPS. The inflammatory cell response was significantly diminished by M1, with 50 mM (10 mg/ml) being comparable in effectiveness to the specific iNOS inhibitor L-NIL at 1 mM. The IC50 of M1 was 1.3 mg/ml (  5 mM) and thus at the lower end of a concentration range described for other bioactive plant extracts or extract constituents in the same assay system [5–9]. Interestingly, M1 was markedly more effective on a mg/ml basis than the maritime pine bark extract (Pycnogenol) itself. Earlier a 40% inhibition of nitrite release from murine RAW 264.7 macrophages was observed at a concentration of 100 mg/ml Pycnogenol [13]. A molecule with a structure similar to that of M1, featuring an additional vicinal hydroxyl group at the benzene ring, 5-[(3,4,5-trihydroxyphenyl)methyl]oxolan-2-one, revealed a 50% inhibition of nitrite release from LPS-stimulated RAW 264.7 macrophages at a concentration of 20 mM ( 5 mg/ml) [36]. Thus, it seems that the third hydroxyl function did not increase the inhibitory activity toward nitrite production. Assuming that the effect is at least partially based on direct scavenging of NO molecules, this is consistent with observations that (þ )gallocatechin with its trihydroxylated B-ring displayed no higher NO-scavenging activity compared to the dihydroxylated ( þ)catechin molecule [37]. Notably, the M1 precursor molecule catechin displays a rather weak activity toward induced nitrite generation, as high catechin concentrations of 100, 224, or 500 mM (29, 65, or 145 mg/ml) did not demonstrate significant inhibition [6,38,39]. This would contribute to explaining the considerably lower nitrite inhibition by the catechin-containing Fraction I compared to M1 in our study. The anti-inflammatory corticosteroid hormone hydrocortisone exhibited surprisingly weak attenuation of nitrite production compared to M1 and also in relation to Fraction I. An activity of

311

M1 about twice as high over that of hydrocortisone was earlier shown for inhibition of MMP-9 secretion [19]. However, in the present assay system M1 (1 mg/ml, 5 mM) was about 20-fold more potent in decreasing nitrite formation than hydrocortisone (36.25 mg/ml, 100 mM). Because hydrocortisone displays no direct NO-scavenging activity [40] this also supports the view that the attenuation of nitrite generation is at least partially due to NO quenching. The effects of M1 on murine RAW 264.7 macrophages were not associated with cytotoxicity as we demonstrated that M1 concentrations up to 25 mg/ml did not compromise cell viability. The neutral red uptake assay we used is well suited to the determination of toxic effects of flavonoids or polyphenols because it is highly sensitive for most types of cytotoxic mechanisms [41]. Moreover, no enzymes are used, as e.g., in the MTT or LDH cytotoxicity tests, and thus interference due to nonspecific enzyme inhibition or a compound’s redox activity can be ruled out. Interestingly, low M1 concentrations of 0.1 and 1 mg/ml even increased cell survival compared to untreated control cells. This might be ascribed to a nonspecific stimulation of mRNA expression, which has been reported for low concentrations of flavonoids [42]. In addition to the inhibition of nitrite production in activated RAW 264.7 macrophages we also found a concentration-dependent inhibition of iNOS expression by M1. M1 decreased the iNOS protein synthesis by 50% at a concentration of 3.8 mg/ml. The fact that slightly higher concentrations were required to inhibit iNOS expression compared to a decrease in NO biosynthesis is consistent with other reports [5,43] and it is indicative of a dual action in quenching NO and inhibiting iNOS overexpression. The iNOS protein is mainly regulated on the transcriptional level and activation of the transcription factors NF-kB and STAT-1a has been identified as an essential step for the iNOS induction [44–46]. We recently detected the metabolite M1 in plasma samples of human volunteers after intake of Pycnogenol [17] and these plasma samples inhibited NF-kB activation in an ex vivo assay [47], thus suggesting that M1 might prevent iNOS upregulation by interaction with the NF-kB signaling pathway. The modulation of this NF-kB pathway and inhibition of iNOS expression have been also shown for other plant extract constituents [6,48]. Though the Pycnogenol metabolite displayed highly effective inhibition of both nitrite generation and iNOS upregulation compared with many other plant extract constituents, the required effective concentrations of M1 still significantly exceeded previously detected plasma concentrations of M1 [17]. We thus explored the degree of cell disposition of M1 in relation to an incubation setup free of cells. In murine RAW 264.7 macrophages, human primary monocytes, and human EA.hy 926 endothelial cells we detected a highly pronounced cell binding of M1. Within 1 h of incubation approximately 50% of the metabolite was associated with the respective cells. Thus, the incubation concentration of 200 ng/ml M1 decreased to 100 ng/ml and this binding trend continued until the end of the experimental period at 120 min. Though it could not be determined whether M1 primarily bound to the outer cell membrane or was taken up into the cells, it can be deduced that we observed a high-capacity binding to the cells. After 120 min M1 concentrations were lower outside of human monocytes and endothelial cells compared to the cell-bound fraction. It has been previously pointed out that blood cells play a role in binding quercetin and resveratrol and that analysis of plasma instead of whole-blood samples underestimates the total concentrations of the compounds being present in the system [49]. It has been also shown that polyphenols are taken up into cells at highly variable extents depending on the cell line and the individual compound [50]. Our results revealed a

312

K. Uhlenhut, P. H¨ ogger / Free Radical Biology and Medicine 53 (2012) 305–313

marked association of M1 with macrophages, monocytes, and endothelial cells, which depicts a novel observation that has not been described before. Because the binding of M1 to those cells was rapid and pronounced we wondered whether diffusion processes would be the only driving force for cell association. Recently, we analyzed the plasma protein binding of various polyphenols and found that M1 bound to a lower extent to plasma proteins compared to most other compounds. In contrast to the flavonoids catechin and taxifolin, which exhibited a plasma protein binding of 495%, hydroxycinnamic acids such as caffeic and ferulic acid were bound to a lower extent (60%–80%) and only about 35% of the metabolite M1 was bound to plasma proteins [51]. To uncover any enhanced uptake of M1 we then co-incubated the cells with phloretin. Phloretin has been described as an inhibitor of various transporters; it potently inhibits the GLUT-1-type glucose transporter with a Ki of 0.3770.03 mM at pH 6.5 [26]. However, phloretin is not a specific inhibitor of GLUT-1 because interactions with other transporters have been reported as well, such as volume-sensitive chloride channels (IC50 30 mM [52]), aquaporin water channels [53], red blood cell urea transporter (IC50 75 mM [27]), and monocarboxylate transporter [28]. Moreover, inhibition of other GLUT isoforms [54] and the sodium glucose cotransporter SGLT-1 [55] has been described. Indeed, in the presence of phloretin the binding of M1 to all cell types that we investigated was markedly reduced. Within the initial 15 min of incubation phloretin almost completely abolished M1 binding compared to the incubations without phloretin. Thereafter, the cell binding of the metabolite was consistently lower in the presence of phloretin. This suggests that in addition to diffusion processes and binding to outer cell membranes a facilitated uptake of M1 via transporters occurred. Though phloretin lacks specificity in transporter inhibition it is possible that GLUT-1 actually played a role in the uptake of M1 because various GLUT isoforms are expressed in human monocytes and macrophages [56,57] and GLUT-1 is highly expressed at endothelial cells [58]. Interestingly, recently computer docking studies revealed that quercetin can slide through GLUT-1 transporters [59]. Experimental evidence supports a cell membrane permeation of quercetin via glucose transport proteins [60,61]. A transporter-facilitated uptake and accumulation of M1 in macrophages, monocytes, and endothelial cells would contribute to understanding why micromolar concentrations have to be used in cell culture assays, whereas only nanomolar concentrations are recovered from plasma samples in vivo. Our study has a number of limitations. The nitrite and iNOS inhibition experiments were performed with murine RAW 264.7 macrophages, which raises the question of transferability to human systems [62]. However, our recent results are consistent with reported anti-inflammatory effects of Pycnogenol in human studies [63] and the cell binding of M1 was also seen with human monocytes and a human endothelial cell line. Our data indicate a facilitated uptake of M1 although we did not differentiate how much of the compound is actually taken up into the cell or bound to outer cell membranes; future investigations need to identify the precise cellular localization of M1 and also the responsible phloretin-sensitive transporter. Moreover, we did not determine what happens to the metabolite after being taken up into the cells. It is possible that M1 is subjected to further intracellular metabolism reactions. To summarize, our results show that the Pycnogenol metabolite M1 mediated marked anti-inflammatory effects by inhibiting nitrite production in LPS-stimulated macrophages most possibly by a dual mechanism of direct NO quenching and inhibition of overexpression of iNOS. These effects were pronounced in comparison to reported data of other plant extract constituents and

exemplify bioactivation by metabolism because the M1 precursor molecule catechin is only weakly active. We also present novel data of binding and facilitated transporter-mediated uptake of M1 into macrophages, monocytes, and endothelial cells. In fact, intracellular accumulation of M1 could contribute to explaining why in vivo bioactivity can be observed with nanomolar plasma concentrations that typically fail to exhibit measurable activity in vitro.

Acknowledgments This work was supported by an educational research grant from Horphag Research, the company that markets Pycnogenols (Horphag Research Ltd., UK). P.H. received research grants from Horphag Research within 3 years before beginning the submitted work. References [1] Williamson, G.; Manach, C. Bioavailability and bioefficacy of polyphenols in humans. II. Review of 93 intervention studies. Am J Clin Nutr 81:243S–255S; 2005. [2] Chen, S. Natural products triggering biological targets—a review of the antiinflammatory phytochemicals targeting the arachidonic acid pathway in allergy asthma and rheumatoid arthritis. Curr Drug Targets 12:288–301; 2011. [3] Gosslau, A.; Li, S.; Ho, C. T.; Chen, K. Y.; Rawson, N. E. The importance of natural product characterization in studies of their anti-inflammatory activity. Mol Nutr Food Res 55:74–82; 2011. [4] Lorsbach, R. B.; Murphy, W. J.; Lowenstein, C. J.; Snyder, S. H.; Russell, S. W. Expression of the nitric oxide synthase gene in mouse macrophages activated for tumor cell killing: molecular basis for the synergy between interferon-g and lipopolysaccharide. J Biol Chem 268:1908–1913; 1993. [5] Kobuchi, H.; Droy-Lefaix, M. T.; Christen, Y.; Packer, L. Ginkgo biloba extract (EGb 761): inhibitory effect on nitric oxide production in the macrophage cell line RAW 264.7. Biochem Pharmacol 53:897–903; 1997. [6] Terra, X.; Valls, J.; Vitrac, X.; Merrillon, J. M.; Arola, L.; Ardevol, A.; Blade, C.; Fernandez-Larrea, J.; Pujadas, G.; Salvado, J.; Blay, M. Grape-seed procyanidins act as antiinflammatory agents in endotoxin-stimulated RAW 264.7 macrophages by inhibiting NFkB signaling pathway. J Agric Food Chem 55:4357–4365; 2007. [7] Kim, Y. H.; Kim, K. H.; Han, C. S.; Park, S. H.; Yang, H. C.; Lee, B. Y.; Eom, S. Y.; Kim, Y. S.; Kim, J. H.; Lee, N. H. Anti-inflammatory activity of Crinum asiaticum Linne var. japonicum extract and its application as a cosmeceutical ingredient. J Cosmet Sci 59:419–430; 2008. [8] Dirsch, V. M.; Kiemer, A. K.; Wagner, H.; Vollmar, A. M. Effect of allicin and ajoene, two compounds of garlic, on inducible nitric oxide synthase. Atherosclerosis 139:333–339; 1998. [9] Lee, S. K.; Hong, C. H.; Huh, S. K.; Kim, S. S.; Oh, O. J.; Min, H. Y.; Park, K. K.; Chung, W. Y.; Hwang, J. K. Suppressive effect of natural sesquiterpenoids on inducible cyclooxygenase (COX-2) and nitric oxide synthase (iNOS) activity in mouse macrophage cells. J Environ Pathol Toxicol Oncol 21:141–148; 2002. [10] Packer, L.; Rimbach, G.; Virgili, F. Antioxidant activity and biologic properties of a procyanidin-rich extract from pine (Pinus maritima) bark pycnogenol. Free Radic Biol Med 27:704–724; 1999. [11] Rohdewald, P. A review of the French maritime pine bark extract (Pycnogenol), a herbal medication with a diverse clinical pharmacology. Int J Clin Pharmacol Ther 40:158–168; 2002. [12] Maritime pine extract. In: United States Pharmacopeia. Rockville, MD: US Pharmacopeial Convention: 1196–1197; 2011. [13] Virgili, F.; Kobuchi, H.; Packer, L. Procyanidins extracted from Pinus maritima (Pycnogenol): scavengers of free radical species and modulators of nitrogen monoxide metabolism in activated murine RAW 264.7 macrophages. Free Radic Biol Med 24:1120–1129; 1998. [14] Kim, Y. J.; Kim, Y. A.; Yokozawa, T. Pycnogenol modulates apoptosis by suppressing oxidative stress and inflammation in high glucose-treated renal tubular cells. Food Chem Toxicol 49:2196–2201; 2011. [15] Peng, Y. J.; Lee, C. H.; Wang, C. C.; Salter, D. M.; Lee, H. S. Pycnogenol attenuates the inflammatory and nitrosative stress on joint inflammation induced by urate crystals. Free Radic Biol Med 52:765–774; 2012. [16] Manach, C.; Williamson, G.; Morand, C.; Scalbert, A.; Remesy, C. Bioavailability and bioefficacy of polyphenols in humans. I. Review of 97 bioavailability studies. Am J Clin Nutr 81:230S–242S; 2005. [17] Grimm, T.; Skrabala, R.; Chovanova, Z.; Muchova, J.; Sumegova, K.; Liptakova, ¨ A.; Durackova, Z.; Hogger, P. Single and multiple dose pharmacokinetics of maritime pine bark extract (Pycnogenol) after oral administration to healthy volunteers. BMC Clin. Pharmacol 6:4; 2006. ¨ [18] Duweler, K. G.; Rohdewald, P. Urinary metabolites of French maritime pine bark extract in humans. Pharmazie 55:364–368; 2000.

K. Uhlenhut, P. H¨ ogger / Free Radical Biology and Medicine 53 (2012) 305–313

¨ ¨ [19] Grimm, T.; Schafer, A.; Hogger, P. Antioxidant activity and inhibition of matrix metalloproteinases by metabolites of maritime pine bark extract (pycnogenol). Free Radic Biol Med 36:811–822; 2004. [20] Das, N. P. Studies on flavonoid metabolism: absorption and metabolism of (þ )-catechin in man. Biochem Pharmacol 20:3435–3445; 1971. [21] Crozier, A.; Del Rio, D.; Clifford, M. N. Bioavailability of dietary flavonoids and phenolic compounds. Mol Aspects Med 31:446–467; 2010. [22] Monagas, M.; Urpi-Sarda, M.; Sanchez-Patan, F.; Llorach, R.; Garrido, I.; Gomez-Cordoves, C.; Andres-Lacueva, C.; Bartolome, B. Insights into the metabolism and microbial biotransformation of dietary flavan-3-ols and the bioactivity of their metabolites. Food Funct 1:233–253; 2010. [23] Stoupi, S.; Williamson, G.; Drynan, J. W.; Barron, D.; Clifford, M. N A comparison of the in vitro biotransformation of (  )-epicatechin and procyanidin B2 by human faecal microbiota. Mol Nutr Food Res 54:747–759; 2010. [24] Roowi, S.; Stalmach, A.; Mullen, W.; Lean, M. E.; Edwards, C. A.; Crozier, A. Green tea flavan-3-ols: colonic degradation and urinary excretion of catabolites by humans. J Agric Food Chem 58:1296–1304; 2010. [25] Ehrenkranz, J. R.; Lewis, N. G.; Kahn, C. R.; Roth, J. Phlorizin: a review. Diabetes Metab Res Rev 21:31–38; 2005. [26] Martin, H. J.; Kornmann, F.; Fuhrmann, G. F. The inhibitory effects of flavonoids and antiestrogens on the Glut1 glucose transporter in human erythrocytes. Chem Biol Interact 146:225–235; 2003. [27] Martial, S.; Olives, B.; Abrami, L.; Couriaud, C.; Bailly, P.; You, G.; Hediger, M. A.; Cartron, J. P.; Ripoche, P.; Rousselet, G Functional differentiation of the human red blood cell and kidney urea transporters. Am J Physiol 271: F1264–F1268; 1996. [28] Garcia, C. K.; Brown, M. S.; Pathak, R. K.; Goldstein, J. L. cDNA cloning of MCT2, a second monocarboxylate transporter expressed in different cells than MCT1. J Biol Chem 270:1843–1849; 1995. [29] Rappold M. Synthese von d-(30 ,40 -Dihydroxyphenyl)-g-valerolacton und Derivaten sowie die Enantiomerentrennung der Zwischenstufen an chiraler Phase und Synthese des Azidoanilin-8-azidoadenosintriphosphat-Lithium¨ ¨ Pharmazie und Lebensmittelchemie, Unisalzes. Wurzburg: Institut fur ¨ Wurzburg, ¨ versitat 2010. [Diploma thesis]. ´ [30] Blazso, G.; Gabor, M.; Sibbel, R.; Rohdewald, P. Anti-inflammatory and superoxide radical scavenging activities of procyaninidins-containing extract from the bark of Pinus pinaster sol and its fractions. Pharm Pharmacol Lett 3:217–220; 1994. [31] Babich, H.; Borenfreund, E. Cytotoxic effects of food additives and pharmaceuticals on cells in culture as determined with the neutral red assay. J Pharm Sci 79:592–594; 1990. [32] Repetto, G.; del Peso, A.; Zurita, J. L. Neutral red uptake assay for the estimation of cell viability/cytotoxicity. Nat Protoc 3:1125–1131; 2008. [33] Guevara, I.; Iwanejko, J.; Dembinska-Kiec, A.; Pankiewicz, J.; Wanat, A.; Anna, P.; Golabek, I.; Bartus, S.; Malczewska-Malec, M.; Szczudlik, A. Determination of nitrite/nitrate in human biological material by the simple Griess reaction. Clin Chim Acta 274:177–188; 1998. [34] Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254; 1976. [35] Li, H.; Oehrlein, S. A.; Wallerath, T.; Ihrig-Biedert, I.; Wohlfart, P.; Ulshofer, T.; Jessen, T.; Herget, T.; Forstermann, U.; Kleinert, H. Activation of protein kinase Ca and/or e enhances transcription of the human endothelial nitric oxide synthase gene. Mol Pharmacol 53:630–637; 1998. [36] Lambert, J. D.; Rice, J. E.; Hong, J.; Hou, Z.; Yang, C. S. Synthesis and biological activity of the tea catechin metabolites, M4 and M6 and their methoxyderivatives. Bioorg Med Chem Lett 15:873–876; 2005. [37] Nakagawa, T.; Yokozawa, T. Direct scavenging of nitric oxide and superoxide by green tea. Food Chem Toxicol 40:1745–1750; 2002. [38] Kim, H. K.; Cheon, B. S.; Kim, Y. H.; Kim, S. Y.; Kim, H. P. Effects of naturally occurring flavonoids on nitric oxide production in the macrophage cell line RAW 264.7 and their structure–activity relationships. Biochem Pharmacol 58:759–765; 1999. [39] Wang, J.; Mazza, G. Inhibitory effects of anthocyanins and other phenolic compounds on nitric oxide production in LPS/IFN-g-activated RAW 264.7 macrophages. J Agric Food Chem 50:850–857; 2002. [40] Asanuma, M.; Nishibayashi-Asanuma, S.; Miyazaki, I.; Kohno, M.; Ogawa, N. Neuroprotective effects of non-steroidal anti-inflammatory drugs by direct scavenging of nitric oxide radicals. J Neurochem 76:1895–1904; 2001.

313

[41] Weyermann, J.; Lochmann, D.; Zimmer, A. A practical note on the use of cytotoxicity assays. Int J Pharm 288:369–376; 2005. [42] Sehm, J.; Polster, J.; Pfaffl, M. W. Effects of varied EGCG and (þ )-catechin concentrations on proinflammatory cytokines mRNA expression in cona-stimulated primary white blood cell cultures. J Agric Food Chem 53:6907–6911; 2005. [43] Wang, G. J.; Chen, Y. M.; Wang, T. M.; Lee, C. K.; Chen, K. J.; Lee, T. H. Flavonoids with iNOS inhibitory activity from Pogonatherum crinitum. J Ethnopharmacol 118:71–78; 2008. [44] Surh, Y. J.; Chun, K. S.; Cha, H. H.; Han, S. S.; Keum, Y. S.; Park, K. K.; Lee, S. S. Molecular mechanisms underlying chemopreventive activities of antiinflammatory phytochemicals: down-regulation of COX-2 and iNOS through suppression of NF-kB activation. Mutat Res 480–481:243–268; 2001. [45] Murakami, A. Chemoprevention with phytochemicals targeting inducible nitric oxide synthase. Forum Nutr 61:193–203; 2009. [46] Pautz, A.; Art, J.; Hahn, S.; Nowag, S.; Voss, C.; Kleinert, H. Regulation of the expression of inducible nitric oxide synthase. Nitric Oxide 23:75–93; 2010. [47] Grimm, T.; Chovanova, Z.; Muchova, J.; Sumegova, K.; Liptakova, A.; Durack¨ ova, Z.; Hogger, P. Inhibition of NF-kB activation and MMP-9 secretion by plasma of human volunteers after ingestion of maritime pine bark extract (Pycnogenol). J Inflammation (London) 3:1; 2006. [48] Puangpraphant, S.; de Mejia, E. G. Saponins in yerba mate tea (Ilex paraguariensis A. St.-Hil) and quercetin synergistically inhibit iNOS and COX-2 in lipopolysaccharide-induced macrophages through NFkB pathways. J Agric Food Chem 57:8873–8883; 2009. [49] Biasutto, L.; Marotta, E.; Garbisa, S.; Zoratti, M.; Paradisi, C. Determination of quercetin and resveratrol in whole blood—implications for bioavailability studies. Molecules 15:6570–6579; 2010. [50] Schaffer, S.; Halliwell, B. Do polyphenols enter the brain and does it matter? Some theoretical and practical considerations Genes Nutr 7:99–109; 2012. ¨ [51] Kurlbaum, M.; Hogger, P. Plasma protein binding of polyphenols from maritime pine bark extract (USP). J. Pharm Biomed Anal 54:127–132; 2011. [52] Fan, H. T.; Morishima, S.; Kida, H.; Okada, Y. Phloretin differentially inhibits volume-sensitive and cyclic AMP-activated, but not Ca-activated, Cl(  ) channels. Br. J Pharmacol 133:1096–1106; 2001. [53] Yang, B.; Ma, T.; Verkman, A. S. Erythrocyte water permeability and renal function in double knockout mice lacking aquaporin-1 and aquaporin-3. J Biol Chem 276:624–628; 2001. [54] Kalsi, K. K.; Baker, E. H.; Medina, R. A.; Rice, S.; Wood, D. M.; Ratoff, J. C.; Philips, B. J.; Baines, D. L. Apical and basolateral localisation of GLUT2 transporters in human lung epithelial cells. Pflugers Arch 456:991–1003; 2008. [55] Raja, M. M.; Tyagi, N. K.; Kinne, R. K. Phlorizin recognition in a C-terminal fragment of SGLT1 studied by tryptophan scanning and affinity labeling. J Biol Chem 278:49154–49163; 2003. [56] Malide, D.; Davies-Hill, T. M.; Levine, M.; Simpson, I. A. Distinct localization of GLUT-1, -3, and -5 in human monocyte-derived macrophages: effects of cell activation. Am J Physiol 274:E516–E526; 1998. [57] Fu, Y.; Maianu, L.; Melbert, B. R.; Garvey, W. T. Facilitative glucose transporter gene expression in human lymphocytes, monocytes, and macrophages: a role for GLUT isoforms 1, 3, and 5 in the immune response and foam cell formation. Blood Cells Mol Dis 32:182–190; 2004. [58] Olson, A. L.; Pessin, J. E. Structure, function, and regulation of the mammalian facilitative glucose transporter gene family. Annu Rev Nutr 16:235–256; 1996. [59] Cunningham, P.; Afzal-Ahmed, I.; Naftalin, R. J. Docking studies show that D-glucose and quercetin slide through the transporter GLUT1. J Biol Chem 281:5797–5803; 2006. [60] Strobel, P.; Allard, C.; Perez-Acle, T.; Calderon, R.; Aldunate, R.; Myricetin, Leighton F. quercetin and catechin-gallate inhibit glucose uptake in isolated rat adipocytes. Biochem J 386:471–478; 2005. [61] Vlachodimitropoulou, E.; Sharp, P. A.; Naftalin, R. J. Quercetin–iron chelates are transported via glucose transporters. Free Radic Biol Med 50:934–944; 2011. [62] Schneemann, M.; Schoeden, G. Macrophage biology and immunology: man is not a mouse. J Leukocyte Biol 81(579); 2007. discussion 580. [63] Maimoona, A.; Naeem, I.; Saddiqe, Z.; Jameel, K. A review on biological, nutraceutical and clinical aspects of French maritime pine bark extract. J Ethnopharmacol 133:261–277; 2011.