Inhibitory effect of plant phenolics on fMLP-induced intracellular calcium rise and chemiluminescence of human polymorphonuclear leukocytes and their chemotactic activity in vitro

Abstract

Context: Polymorphonuclear leukocytes (PMNs) produce oxidants, contributing to systemic oxidative stress. Diets rich in plant polyphenols seem to decrease the risk of oxidative stress-induced disorders including cardiovascular disease.

Objective: The objective of this study was to examine the in vitro effect of each of the 14 polyphenols on PMNs chemotaxis, intracellular calcium response, oxidants production.

Materials and methods: Blood samples and PMNs suspensions were obtained from 60 healthy non-smoking donors and incubated with a selected polyphenol (0.5–10 µM) or a control solvent. We assessed resting and fMLP-dependent changes of intracellular calcium concentration ([Ca²⁺]_i) in PMNs with the Fura-2AM method and measured fMLP-induced luminol enhanced whole blood chemiluminescence (fMLP-LBCL). Polyphenol chemoattractant activity for PMNs was tested with Boyden chambers.

Results: Polyphenols had no effect on resting [Ca²⁺]_i. Unaffected by other compounds, fMLP-dependent increase of [Ca²⁺]_i was inhibited by quercetin and catechol (5 µM) by 32 ± 14 and 12 ± 10% (p < 0.04), respectively. Seven of the 14 tested substances (5 µM) influenced fMLP-LBCL by decreasing it. Catechol, quercetin, and gallic acid acted most potently reducing fMLP-LBCL by 49 ± 5, 42 ± 15, and 28 ± 18% (p < 0.05), respectively. 3,4-Dihydroxyhydrocinnamic, 3,4-dihydroxyphenylacetic, 4-hydroxybenzoic acid, and catechin (5 µM) revealed distinct (p < 0.02) chemoattractant activity with a chemotactic index of 1.9 ± 0.8, 1.8 ± 0.7, 1.6 ± 0.6, 1.4 ± 0.2, respectively.

Conclusion and discussion: Catechol, quercetin, and gallic acid at concentrations commensurate in human plasma strongly suppressed the oxidative response of PMNs. Regarding quercetin and catechol, this could result from an inhibition of [Ca²⁺]_i response.

• Antioxidants
• chemiluminescence
• intracellular calcium
• chemotaxis
• oxidative burst
• polymorphonuclear leukocytes
• polyphenols

Introduction

Plant polyphenols are phytochemical compounds containing phenol units in their backbone structure. This large group of substances includes phenolic acids, flavonoids, stilbenes, and lignans. Once ingested in a meal and before absorption into the blood, these polyphenols are transformed into less complex compounds (e.g. phenolic acids) by the intestinal microflora (Aura, 2008). Thus, apart from their presence in fresh plants, phenolic acids may be formed in the digestive tract from other polyphenols, which increase their bioavailability making them important phytochemicals that can affect various processes in the human body. Numerous studies have shown a significant beneficial effect of a fruit- and vegetable-rich diet in relation to the risk of cardiovascular disease and cancer in humans. Repeated fruit and vegetable consumption reduced the frequency of myocardial infarction and stroke (Gillman et al., 1995; Ness & Powles, 1997; Tyrovolas & Panagiotakos, 2010), the latter was particularly reduced by intake of apples or green tea (Knekt et al., 2000; Tanabe et al., 2008). A diet rich in plants was protective against certain types of neoplasms, reducing the risk of colorectal adenomas, colorectal cancer, and lung cancer (Tyrovolas & Panagiotakos, 2010; Van Duijnhoven et al., 2009).

Although there are many possible mechanisms concerning the activity of polyphenols, their antioxidant and anti-inflammatory properties seem to contribute to both an anticancer and an anti-atherosclerotic effect (Michalska et al., 2010; Pandey & Rizvi, 2009). Oxidative stress that can facilitate the progression of mutagenesis or atherosclerosis (Delaney et al., 2012; Peluso et al., 2012) may be counteracted by plant polyphenols. Red onion extracts containing ferulic acid, gallic acid, 3,4-dihydroxybenzoic acid, quercetin, and kaempferol inhibited in vitro β-carotene and linoleic acid oxidation, impeded ferric ion-induced lipid peroxidation and protein fragmentation; furthermore they inhibited tobacco-induced mutagenicity in strains of Salmonella typhimurium (Singh et al., 2009). Regular consumption of dark chocolate (rich in polyphenols) by healthy humans reduced the intensity of DNA damage in peripheral blood mononuclear cells as compared with the consumption of white chocolate (Spadafranca et al., 2010). Ingestion of procyanidin-rich chocolate (flavonoid) increased the plasma antioxidant capacity and decreased the levels of circulating end products of lipid peroxidation in healthy humans (Wang et al., 2000).

Apart from the ability to inhibit low-density lipoprotein oxidation and the suppression of circulating monocyte chemoattractant protein-1, consumption of olive oil with a high polyphenol content for 3 weeks resulted in the reduced expression of proatherogenic and proinflammatory genes in the blood monocytes of healthy subjects (Castaner et al., 2012). Besides direct radical scavenging, this indicates that anti-inflammatory properties of polyphenols may contribute to their antioxidant effect. An increased activity of circulating inflammatory cells such as monocytes and polymorphonuclear leukocytes (PMNs) can predispose to oxidative stress. PMNs can produce large amounts of reactive oxygen species (ROS) when activated. Their stimulation leads to an increase in intracellular calcium concentration [Ca²⁺]_i involved in signal transmission to NADPH oxidase. Activation of this enzyme results in intense superoxide production followed by hydrogen peroxide (H₂O₂) and hypochlorous acid (HOCl) formation (Truett et al., 1988). Some studies have indicated that polyphenols can decrease oxidant production by PMNs in vitro (Ribeiro et al., 2013). However, little is known about this mechanism of action. Therefore, we decided to investigate the effect of 14 polyphenols and their metabolites (commonly found in plant foods) on the ability of human PMNs to produce ROS in a whole blood system and upon [Ca²⁺]_i changes after stimulation with N-formyl-methionyl-leucyl-phenylalanine (fMLP a synthetic analogue of bacterial chemotactic peptide). Furthermore, the chemoattractant activity of polyphenols for PMNs was investigated in vitro.

Materials and methods

Reagents, buffers, and solutions

Phenolics: 4-hydroxybenzoic acid, gallic acid, chlorogenic acid, 3,4-dihydroxycinnamic acid, 3,4-dihydroxyhydrocinnamic acid, ferulic acid, 3,4-dihydroxyphenylacetic acid, 3,4-dihydroxybenzoic acid, quinic acid, catechol, catechin, phloroglucinol, phlorizin, quercetin; FURA-2AM (acetoxymethyl ester), ethylene glycol bis (2-aminoethyl-ether)-tetraacetic acid (EGTA), Triton X-100, dimethyl sulfoxide (DMSO), N-formyl-methionyl-leucyl-phenylalanine (fMLP), 5-amino-2,3-dihydro-1,4-phthalazinedione (luminol), HEPES, Histopaque 1077, dextran (200 kDa), and human albumin were acquired from Sigma-Aldrich (St. Louis, MO). CaCl₂, KH₂PO₄, MgCl₂, KCl, NaCl (powder and 0.9% sterile aqueous solution), and d-glucose were purchased from POCH (Gliwice, Poland). Phosphate buffered saline (PBS, pH = 7.4, osmolarity 270–300 mOsm/L) was obtained from Biomed-Lublin SA (Lublin, Poland). All other reagents were of analytical grade. fMLP was dissolved in DMSO to a final concentration of 6 mM for intracellular calcium and chemotaxis assessment and 20 mM for luminescence assay. FURA-2AM and polyphenols were dissolved in DMSO to a final concentration of 2 and 60 mM, respectively. All stock solutions were stored at −80 °C until assay and were dissolved with 0.9% NaCl just before use (working solutions), polyphenols were at a concentration of 300 µM, fMLP at 60 µM (intracellular calcium assay), 200 µM (whole blood chemiluminescence), and 0.1 µM (in PBS, chemotaxis experiments). Aqueous solutions of 1 M NaOH, 1 M HCl, 1 M CaCl₂, 1 M MgCl₂, 1.8% NaCl, and 5% dextran in 0.9% NaCl were stored at 4 °C for no longer than 4 d. Aqueous solutions of 0.1 M EGTA (pH 9.3), 10% Triton X-100, and 1 g/L albumin in PBS were prepared freshly before use. Sterile, deionized water (resistance 18 Ω cm, from water purification system USF ELGA, Germany) was used throughout the study. HEPES buffer was prepared by adding 2 mmol HEPES, 13.2 mmol NaCl, 0.6 mmol KCl, 0.12 mmol KH₂PO₄, and water up to 100 mL and was stored in the dark at −10 °C for no longer than 2 weeks, after thawing and immediately before use, albumin and d-glucose were added to a final concentrations of 1 g/L and 5 mM, respectively. Luminol solution was prepared by dissolving 25 mg luminol in 90 mL of 0.1 M Na₂HPO₄ then pH was adjusted to 7.4 with 1 M HCl and volume filled to 100 mL with deionized water. Afterwards, the solution was filtered (0.2 µm Millipore filter) and stored at −10 °C in the dark for no longer than 2 weeks. The mixture solution was prepared freshly before the assay by adding 1 mL Ringer solution, 5 mL luminol solution, 0.2 mL of a 277.5 mM glucose solution to 3.6 mL of deionized water.

Blood donors

The group of blood donors involved 60 non-smoking apparently healthy subjects (members of our medical university staff, mean age 32 ± 8 years, 25 men). They did not receive any medications and food supplements that may affect PMNs functions within 3 months prior to the study. Samples of venous blood (27 mL for [Ca²⁺]_i and chemotaxis assay, 4 mL for chemiluminescence assay) were collected into a vacutainer tube with EDTA (Becton Dickinson, San Jose, CA) between 8:30 and 9:30 a.m. The Medical University of Lodz Ethics Committee approved the study protocol (approval RNN/62/09/KE) and all participants gave written informed consent.

Cell preparation

Human PMNs were isolated from venous blood according to Boyüm’s method (1968). Human albumin was used in isolation procedure. Dextran sedimentation was followed by centrifugation over Histopaque. The number of cells was determined with a hemocytometer. The purity of PMNs suspensions was 98% as estimated by the analysis of smears from each sample using Giemsa stain and the viability was above 98% as assessed by the trypan blue exclusion test (Blue & Janoff, 1978).

Measurement of cytosolic-free calcium

PMNs were saturated with Fura-2AM as described in our previous study (Bialasiewicz et al., 2004) and then suspended in 20 mM HEPES buffer containing 1 mM MgCl₂, 1 mM CaCl₂, (pH = 7.4) to a final concentration of 2 × 10⁶ cells/mL. In order to analyze the effect of polyphenols on fMLP-induced PMNs [Ca²⁺]_i response, part of the cells was incubated (30 min at 37 °C in darkness) with selected polyphenols at a concentration of 5 µM, or control solvent (DMSO dissolved in 0.9% NaCl). Then PMNs were transferred into a microcuvette of a fluorescence spectrometer. After 30 s measurement of baseline [Ca²⁺]_i-dependent fluorescence, [Ca²⁺]_i response was induced by the addition of a 5 µL fMLP solution (a final concentration of 0.5 µM), fluorescence measurement was continued for an additional 10 min.

To study the direct effect of polyphenols on PMNs [Ca²⁺]_i, part of the samples was first left for 15 min in darkness at 37 °C. After 30 s of fluorescence measurement, 10 µL of a tested polyphenol solution (a final concentration of 5 µM) or DMSO in 0.9% saline – a negative control or 5 µL of fMLP solution – a positive control (a final concentration of 0.5 µM) was added to the cells and measurement was continued for an additional 6 min. In all studied and control samples, the final concentration of DMSO was 1 mM. The [Ca²⁺]_i response was determined with dual wavelength measurements at an emission wavelength of 510 nm and excitation wavelengths of 340 and 380 nm on a Perkin Elmer Luminescence Spectrometer LS-50B (Perkin-Elmer, Norwalk, CT). Slit widths were set at 5.0 nm for emission and at 10.0 nm for excitation. All measurements were performed at 37 °C under constant stirring of 600 µL PMNs suspension (1.2 × 10⁶ cells) kept in a microcuvette (PE 5200-4339). Regarding PMNs from each donor, fluorescence signals were calibrated after cell lysis with 0.1% Triton X-100 (maximal signal) and subsequent addition of EGTA to a final concentration of 10 mM (minimal signal). All calculations of [Ca²⁺]_i were done using The Intracellular Biochemistry Application software (Perkin Elmer, Beaconsfield, Bucks, England 1990). Time curves of [Ca²⁺]_i were smoothed with a coefficient of 88 to reduce signal noise and then analyzed. The following parameters were calculated: (1) increment in the intracellular Ca²⁺ concentration (Δ[Ca²⁺]_i) – a difference between peak and basal [Ca²⁺]_i, (2) the ratio of Δ[Ca²⁺]_i in PMNs incubated with polyphenol to Δ[Ca²⁺]_i in cells incubated with control solvent (Δ[Ca²⁺]_i P/Δ[Ca²⁺]_i C). This ratio was calculated only for cells obtained from the same donor, being a specific value for each polyphenol.

Whole blood chemiluminescence assay

The effect of polyphenols on fMLP-stimulated luminal-enhanced whole blood chemiluminescence (fMLP-LBCL) was measured according to Kukovetz et al. (1997) with some modifications. Blood samples were prediluted with a mixture solution (30 µL of blood mixed with 1 mL of a mixture solution). Then 103 µL of diluted blood was added to 797 µL of a mixture solution (final blood dilution 300 times). Directly afterwards, 17 µL of a polyphenol solution (a final concentration of 5 µM) or DMSO in 0.9% saline (control solvent) was added to the corresponding blood samples (from the same blood donor) that were gently mixed and then placed in a multi-tube luminometer (AutoLumat Plus LB 953, Berthold, Germany, equipped with a Peltier-cooled detector to ensure a high sensitivity and a low and stable background signal noise) and incubated for 15 min at 37 °C in the dark. Then 100 µL of fMLP solution (final agonist concentration in the assayed sample 20 µM) or control solvent (DMSO in 0.9% NaCl final concentration 14 mM) was added by automatic dispensers and 7 s later the total light emission automatically began measuring for an additional 120 s. Individual results were displayed as a total sum of photostimuli generated from the sample during a 2 min measurement and were expressed in arbitrary units, read light units (RLU). Difference between fMLP-LBCL of control samples (without polyphenol) and those containing a polyphenol was calculated and expressed as a percent of control value (I%) being a specific measure of fMLP-LBCL inhibition caused with a given polyphenol. This was calculated only for samples from the same blood donor. Three polyphenols that revealed the highest inhibitory activity on fMLP-LBCL (I% ≥25% arbitrarily recognized as of biological importance) were further examined at concentrations of 0.5, 1, 2.5, 7.5, and 10 µM.

Chemotaxis assay

Polyphenol chemoattractant activity for human PMNs was assessed according to the procedure described in our previous study (Bialasiewicz et al., 2004), in 48-well microchemotaxis plates (Neuro Probe, Gaithersburg, MD) with wells separated by a polycarbonate membrane with a pore size of 5 µm (Harvath et al., 1980). About 29 µL samples of a 5 µM polyphenol solution, a 0.1 µM fMLP solution in PBS (known chemoattractant–positive control) and PBS alone (negative control) were placed in the bottom chambers of wells where appropriate. All solutions had the same content of DMSO (1 mM). 30 µL of PMNs suspension (10⁶ cells/ml in PBS containing 1 mM MgCl₂ and CaCl₂) were placed in the upper chambers. The ratio of the number of cells attracted by the tested substance to the number that migrated to PBS was a value specific for each compound and termed as a chemotactic index (C_i). Compounds that revealed a C_i ≥ 1.4 at a concentration of 5 µM were arbitrarily recognized as having a chemoattractant activity and were further examined at lower concentrations of 1 and 0.1 µM. This was based on the following data: preliminary control experiments showed a C_i = 2.4 ± 0.6 (n = 10) for fMLP at a concentration 50-times lower (0.1 µM), additionally human plasma levels of the vast majority of phenolics tested in our experiments did not exceed 5 µM (Guy et al., 2009; Lafay & Gil-Izquierdo, 2008; Manach et al., 2005; Vitaglione et al., 2007).

Statistical analysis

Results were expressed as a mean (SD) and median. Data distribution was tested with the Kolmogorov–Smirnov–Lilliefors test. According to the data distribution, the following test was used: Wilcoxon test or paired t-test for the dependent variables (polyphenol at a given concentration versus the positive and the negative controls), the Mann–Whitney U test or the unpaired t-test for the independent variables (comparison between two different concentrations of the same polyphenol and the comparison between two polyphenols at the same concentration). The Brown–Forsythe test for analysis of the equality of the group variances was used prior to the application of the unpaired t-test and if variances were unequal, Welch’s t-test was used instead of the standard t-test. A p value of <0.05 was considered significant.

Results

Effect of polyphenols on intracellular calcium concentration of resting PMNs

Polyphenols added to the PMNs suspension exerted no effect on [Ca²⁺]_i during a 6 min observation. We found no increase of [Ca²⁺]_i and observed a nearly flat shape in the time curves of [Ca²⁺]_i as well as in the control probes where DMSO was added (Figure 1, curves a and b). In control experiments, PMNs that did not respond to the addition of a polyphenol were able to increase their [Ca²⁺]_i in response to stimulation with fMLP, Δ[Ca²⁺]_i = 129 ± 40 nM, n = 18.

Figure 1. Typical changes of [Ca²⁺]_i in human PMNs after (a) direct stimulation with 5 μM ferulic acid; (b) after exposure to control solvent (DMSO in 0.9% saline, final DMSO concentration 1 mM); (c) cells preincubated for 30 min with DMSO in 0.9% saline and then stimulated with 0.5 μM fMLP; (d) cells preincubated for 30 min with 5 μM quercetin and then stimulated with 0.5 μM fMLP.

A 30-min preincubation of PMNs with quercetin and catechol significantly (p < 0.006 and p < 0.04, respectively) inhibited the fMLP stimulated rise of [Ca²⁺]_i by 32 ± 14% and 12 ± 10%, respectively (n = 6) (Table 1 and Figure 1, curves c and d). Although 3,4-dihydroxycinnamic acid gave mean Δ[Ca²⁺]_i P/Δ[Ca²⁺]_i C ratio similar to catechol, paired statistics did not reveal statistical significance. Incubation with the other 12 compounds did not affect the fMLP-evoked [Ca²⁺]_i response of human PMNs (Table 1).

Table 1. Effect of polyphenols on fMLP-induced [Ca²⁺]_i response of human PMNs in vitro.

	fMLP-stimulated increase of [Ca^2+]i in human PMNs [nM]
Tested polyphenol (5 µM)	Preincubated with control solvent (Δ[Ca^2+]i C)	Preincubated with polyphenol (Δ[Ca^2+]i P)	Mean ratio of Δ[Ca^2+]i P/Δ[Ca^2+]i C
Quercetin	148.7 ± 54.5 (154.4)	100.5* ± 33.3 (96.3)	0.68 ± 0.14 (0.62)
Catechol	108.9 ± 25.0 (100.2)	97.5# ± 29.0 (89.1)	0.88 ± 0.10 (0.88)
3,4-Dihydroxycinnamic acid	108.9 ± 25.0 (100.2)	96.8 ± 27.4 (90.2)	0.89 ± 0.17 (0.96)
Phlorizin	148.7 ± 54.5 (154.4)	128.4 ± 37.1 (137.6)	0.90 ± 0.32 (0.83)
Chlorogenic acid	108.9 ± 25.0 (100.2)	97.0 ± 21.9 (98.0)	0.91 ± 0.21 (0.85)
Quinic acid	108.9 ± 25.0 (100.2)	108.1 ± 29.7 (99.8)	0.99 ± 0.13 (0.97)
Phloroglucinol	148.7 ± 54.5 (154.4)	142.1 ± 41.9 (144.1)	0.99 ± 0.33 (0.89)
Ferulic acid	108.9 ± 25.0 (100.2)	107.8 ± 19.5 (107.7)	1.00 ± 0.10 (0.98)
3,4-Dihydroxyhydrocinnamic acid	129.9 ± 30.4 (117.3)	133.6 ± 29.8 (139.8)	1.02 ± 0.13 (1.04)
3,4-Dihydroxybenzoic acid	148.7 ± 54.5 (154.4)	153.2 ± 50.0 (155.3)	1.04 ± 0.15 (1.04)
Catechin	129.9 ± 30.4 (117.3)	141.3 ± 24.4 (145.2)	1.09 ± 0.09 (1.10)
4-Hydroxybenzoic acid	129.9 ± 30.4 (117.3)	144.8 ± 32.3 (137.3)	1.12 ± 0.19 (1.04)
3,4-Dihydroxyphenylacetic acid	129.9 ± 30.4 (117.3)	149.6 ± 41.7 (156.5)	1.14 ± 0.19 (1.10)
Gallic acid	129.9 ± 30.4 (117.3)	155.4 ± 53.0 (155.8)	1.17 ± 0.28 (1.10)

Δ[Ca²⁺]_i P/Δ[Ca²⁺]_i C ratio reflects the compound’s ability to suppress increase of [Ca²⁺]_i after PMNs activation with fMLP, values less than 1 indicate the suppression. Results obtained from six series of experiments with cells from nine donors are expressed as a mean ± SD (median).

*p < 0.006 versus control, #p < 0.04 versus control.

Effect of polyphenols on fMLP-induced luminol enhanced whole blood chemiluminescence

Catechol, quercetin, gallic acid, 3,4-dihydroxyhydrocinnamic acid, quinic acid, 3,4-dihydroxyphenylacetic acid, and 3,4-dihydroxycinnamic acid significantly decreased fMLP-induced LBCL at a concentration of 5 µM (Table 2). Among these compounds, catechol, quercetin, and gallic acid revealed the highest LBCL inhibition which reached 49 ± 5, 42 ± 15, and 28 ± 18%, respectively.

Table 2. Effect of polyphenols on fMLP-induced luminol enhanced whole blood chemiluminescence in vitro (fMLP-LBCL).

	fMLP-induced whole blood chemiluminescence [RLU]
Tested polyphenol (5 µm)	Preincubated with control solvent	Preincubated with polyphenol	Percent inhibition (I%)
Catechol	12 161 ± 8442 (9431)	5656* ± 4282 (3884)	49 ± 5 (50)
Quercetin	12 745 ± 6715 (12 017)	6930* ± 4006 (6162)	42 ± 15 (44)
Gallic acid	10 920 ± 8957 (7373)	6413* ± 4663 (4259)	28 ± 18 (30)
3,4-Dihydroxyhydrocinnamic acid	10 920 ± 8957 (7373)	7581* ± 5936 (5275)	19 ± 14 (20)
Quinic acid	12 161 ± 8442 (9431)	9395* ± 7700 (6578)	17 ± 14 (14)
3,4-Dihydroxyphenylacetic acid	10 920 ± 8957 (7373)	7642* ± 6041 (4763)	16 ± 15 (13)
3,4-Dihydroxycinnamic acid	12 161 ± 8442 (9431)	9321* ± 7070 (5960)	15 ± 9 (15)
Chlorogenic acid	12 161 ± 8442 (9431)	9920 ± 8350 (6963)	14 ± 15 (17)
Catechin	10 920 ± 8957 (7373)	9047 ± 8380 (5476)	12 ± 15 (13)
Phloroglucinol	12 745 ± 6715 (12 017)	9829 ± 5279 (9516)	9 ± 26 (13)
3,4-Dihydroxybenzoic acid	12 745 ± 6715 (12 017)	10 455 ± 6200 (11 198)	9 ± 16 (6)
Phlorizin	12 745 ± 6715 (12 017)	10 759 ± 5132 (10 203)	6 ± 13 (4)
Ferulic acid	12 161 ± 8442 (9431)	10 736 ± 8905 (8472)	5 ± 14 (6)
4-Hydroxybenzoic acid	10 920 ± 8957 (7373)	10 472 ± 9349 (6368)	1 ± 17 (3)

Results expressed in arbitrary read light units (RLU) were obtained from eight series of experiments with blood from seven donors. Values are given as a mean ± SD (median).

*p < 0.05 versus control.

Other tested phenolics (4-hydroxybenzoic acid, gallic acid, chlorogenic acid, 3,4-dihydroxycinnamic acid, 3,4-dihydroxyhydrocinnamic acid, quinic acid, ferulic acid, 3,4-dihydroxyphenylacetic acid, 3,4-dihydroxybenzoic acid, catechin, phloroglucinol, and phlorizin) exerted no significant effect on fMLP-LBCL (Table 2). The most active compounds (catechol, quercetin, and gallic acid) were further examined at concentrations of 0.5, 1, 2.5, 7.5, and 10 µM. At a concentration of 0.5 µM, none of the compounds induced a significant inhibition of fMLP-LBCL (Table 3). Catechol revealed significant inhibitory activities within the range of studied concentrations (1–10 µM), reaching I% of 67 ± 9% at 10 µM. With quercetin and gallic acid, the onset of a significant inhibition of fMLP-LBCL was noted at a concentration of 2.5 and 5 µM, respectively. Within a concentration range of 1–10 µM, catechol exerted the strongest inhibitory effect on fMLP-LBCL while the gallic acid was seen as the weakest chemiluminescence inhibitor among these three compounds (p < 0.05) (Table 3).

Table 3. Inhibitory effect of various concentrations of catechol, quercetin, and gallic acid on fMLP-induced luminol enhanced whole blood chemiluminescence in vitro.

	% inhibition of fMLP-induced whole blood chemiluminescence at given concentration (I%)
Tested polyphenol	0.5 µM	1 µM	2.5 µM	5 µM	7.5 µM	10 µM
Catechol	6 ± 9 (8)	13* ± 7 (13)	34* ± 9 (33)	49* ± 5 (50)	58* ± 7 (59)	67* ± 9 (68)
Quercetin	0 ± 12 (2)	11 ± 12 (12)	22* ± 10 (24)	42* ± 15 (44)	40* ± 24 (37)	53* ± 13 (58)
Gallic acid	1 ± 10 (0)	6 ± 10 (4)	10 ± 17 (11)	28* ± 18 (30)	39* ± 7 (41)	38* ± 14 (38)

Percent inhibition of chemiluminescent response (I%) by polyphenols, related to the control was calculated. Results obtained from eight series of experiments with blood from seven donors are expressed as a mean ± SD (median).

*Significant inhibition (p < 0.03 versus control).

Chemoattractant activity of polyphenols for human granulocytes

In all series of experiments, PMNs were able to migrate to the fMLP solution (a strong standard chemoattractant). C_i for 0.1 µM fMLP was 2.2 ± 0.7 (n = 35). Out of the 14 tested polyphenols at a concentration of 5 µM, four compounds (3,4-dihydroxyhydrocinnamic acid, 3,4-dihydroxyphenylacetic acid, 4-hydroxybenzoic acid, and catechin) revealed a mean C_i ≥ 1.4, which significantly differed from the C_i of PBS (a negative control solution) (Table 4). All phenolics consistently had a lower chemoattractant activity than fMLP at tested concentrations (p < 0.01). At a concentration of 1 µM, only 3,4-dihydroxyphenylacetic acid significantly attracted human PMNs, while at a concentration of 0.1 µM, no significant activity against PBS was found (Table 4).

Table 4. Chemoattractant properties of four plant phenolics for human PMNs: 4-dihydroxyhydrocinnamic acid, 3,4-dihydroxyphenylacetic acid, 4-hydroxybenzoic acid, and catechin.

	Tested polyphenol
Polyphenol concentration (µM)	3,4-Dihydroxy hydrocinnamic acid	3,4-Dihydroxy phenylacetic acid	4-Hydroxy benzoic acid	Catechin	Positive control fMLP (0.1 µM)	Negative control PBS
0.1	1.3 ± 0.8 (1.0)	1.2 ± 0.5 (1.2)	1.3 ± 0.6 (1.2)	1.2 ± 0.6 (1.1)	2.2*# ± 0.7 (2.1)	1.0 ± 0.0
1	1.2 ± 0.3 (1.1)	1.3* ± 0.2 (1.3)	1.3 ± 0.6 (1.2)	1.1 ± 0.4 (1.0)
5	1.9* ± 0.8 (1.6)	1.8* ± 0.7 (1.6)	1.6* ± 0.6 (1.6)	1.4* ± 0.2 (1.4)

Results expressed as a chemotactic index (C_i) were obtained from 12 series of experiments with cells from 44 donors shown as mean ± SD (median).

*p < 0.02 versus negative control (PBS with 1 mM DMSO), #p < 0.01 versus all tested polyphenols at all concentrations.

Discussion

fMLP is widely used in a variety of experiments regarding activated human PMNs. It binds to heterotrimeric G protein-coupled receptors that result in the activation of phoshoinositide-3-kinase and phospholipase C. Activation of phospholipase C in turn leads to the generation of intracellular second messengers: inositol triphosphate and diacylglycerol. They mobilize Ca²⁺ from intracellular stores that together with an extracellular Ca²⁺ influx result in a substantial elevation of [Ca²⁺]_i and the activation of protein kinase C (Ciz et al., 2012; Futosi et al., 2013). These events lead to the activation of a multisubunit enzyme, NADPH oxidase, which rapidly produces large amounts of superoxide radicals, a hallmark of the respiratory burst, which can then be monitored in blood using a luminol-dependent chemiluminescence technique (LBCL). fMLP is also a strong chemoattractant, inducing actin polymerization, membrane ruffling, and polarization that results in PMNs migration toward a concentration gradient of this peptide. We found that some plant phenolics can inhibit fMLP-induced [Ca²⁺]_i rise (catechol and quercetin) as well as fMLP-induced LBCL (catechol, quercetin, gallic acid, 3,4-dihydroxyhydrocinnamic acid, quinic acid, 3,4-dihydroxyphenylacetic acid, and 3,4-dihydroxycinnamic acid). Furthermore, some phenolics (3,4-dihydroxyhydrocinnamic acid, 3,4-dihydroxyphenylacetic acid, 4-hydroxybenzoic acid, and catechin) revealed a chemoattractant activity for human PMNs.

Effect of polyphenols on the intracellular calcium concentration of PMNs

Of the 14 tested compounds, only quercetin and catechol suppressed an fMLP-induced increase of [Ca²⁺]_i in PMNs. However, they had no impact on [Ca²⁺]_i in resting cells. A possible mechanism of action could include an interaction with a formyl peptide receptor or an intracellular transmission of an excitatory signal (inhibition of G protein-coupled receptors or phospholipase C). However, it cannot be excluded that these phenolics can form non-active complexes with IP3 or inhibit IP3-induced Ca²⁺ release from intracellular stores that cause Ca²⁺ influx from an extracellular space (Futosi et al., 2013). While little is known about the effects of these compounds on intracellular signaling in PMNs or other white blood cells, results of some experiments support our aforementioned hypotheses. In vitro quercetin induced the relaxation of murine airway smooth muscle cells by the direct inhibition of phospholipase C and the suppression of inositol phosphate synthesis. Moreover, it was accompanied by a decreased intracellular Ca²⁺ response to stimulation with histamine and bradykinin (Townsend & Emala, 2013). Quercetin also induced smooth muscle relaxation with simultaneous inhibition of extracellular Ca²⁺ influx through voltage-dependent Ca²⁺ channels in a model of an isolated mouse stomach (Rotondo et al., 2009). Preincubation of human platelets with quercetin resulted in both a decrease in aggregation and a release of 5-hydroxytryptamine after stimulation with collagen in vitro. This inhibitory effect involved among other things, a diminished tyrosine phosphorylation of phospholipase CG2 (Wright et al., 2010). Quercetin can also bind to calmodulin (Nishino et al., 1984), a multifunctional calcium-dependent signal transducer that in human granulocytes activates various kinases and is responsible for the modulation of their functions (Verploegen et al., 2005). Moreover, quercetin, in the same manner as wortmannin (a specific inhibitor of phosphoinositide-3-kinase), modulated the function of human basophils (Chirumbolo et al., 2010). It is possible that quercetin-induced inhibition of phosphoinositide-3-kinase can cause loss of activation of downstream kinases, which activates phospholipase C and can thereby be responsible for the insufficient production of inositol triphosphate and diacylglycerol after basophil stimulation with various agonists (Chirumbolo et al., 2010).

Although (to the best of our knowledge) there is no data on the effect of catechol on the function of human PMNs, it cannot be excluded that this compound caused a specific inhibition of intracellular signal transduction under the conditions of our experiments. Since quercetin and other phenolics can form a complex with divalent cations (e.g. Cu²⁺ and Fe²⁺) (Kim et al., 2013; Pekal et al., 2011) one may speculate that quercetin and catechol can bind Ca²⁺ and thus diminish the Ca²⁺-dependent signal from Fura 2. Nevertheless, a supposed chelation of extracellular and intracellular Ca²⁺ by both these compounds would be negligible considering the 200-fold higher Ca²⁺ level to polyphenol concentration in the buffer used for the preparation of experimental cell suspensions. In addition, a lack of any significant effect of phenolics on [Ca²⁺]_i in resting PMNs also excludes this possibility.

It seems that phospholipase C activity is a target for quercetin (and possibly catechol) responsible for suppression of [Ca²⁺]_i response of human PMNs after stimulation with fMLP. However, further studies with the use of specific signal transduction inhibitors, blockers of Ca²⁺ channels and measurement of phospholipase C activity are necessary for the characterization of the activity of quercetin and catechol on the Ca²⁺ response of human PMNs.

Effect of polyphenols on fMLP-induced luminol enhanced whole blood chemiluminescence (fMLP-LBCL)

Seven phenolics (catechol, quercetin, gallic acid, 3,4-dihydroxyhydrocinnamic acid, quinic acid, 3,4-dihydroxyphenylacetic acid, and 3,4-dihydroxycinnamic acid), significantly decreased fMLP-induced LBCL, which is related to PMNs respiratory burst. Among them, catechol, quercetin, and gallic acid revealed the highest inhibitory effect. One plausible mechanism of action of the first two most active compounds can include the inhibition of PMNs [Ca²⁺]_i response. The rise in [Ca²⁺]_i is necessary for NADPH oxidase activation and the production of ROS. Quercetin was also reported to inhibit cyclic nucleotide phosphodiesterase IV (Townsend & Emala, 2013). In human PMNs, inhibition of this enzyme resulted in the rise of intracellular cyclic 3′,5′-adenosine monophosphate and decreased respiratory burst (Nielson et al., 1990). Inhibition of phospholipase D activation could be another mechanism responsible for the suppression of fMLP-LBCL by quercetin (Takemura et al., 1997). Phospholipase D catalyzes the hydrolysis of phosphatidylcholine with the generation of phosphatidic acid, a signal molecule rapidly converted to diacylglycerol, an activator of protein kinase C (Ciz et al., 2012). Although this effect with quercetin was observed in fMLP-stimulated rabbit PMNs (Takemura et al., 1997), it cannot be excluded that this also occurred under the conditions of our experiments with human blood. Gallic acid can inhibit degranulation of PMNs and interfere with the assembly of active NADPH-oxidase (Kroes et al., 1992). Accordingly, its inhibitory effect on fMLP-LBCL can also involve specific inhibition of signal transduction in human PMNs. The polyphenols studied in our experiments exhibit antioxidant properties (deGraft-Johnson et al., 2007; Goupy et al., 2003; Hotta et al., 2002) and can react in vitro with ROS involved in fMLP-LBCL (Nakano et al., 2012). Thus, a direct unspecific antioxidant effect could contribute to fMLP-LBCL suppression in blood specimens preincubated with phenolics. This could be a leading mechanism, especially in the case of the inhibitory effect of 3,4-dihydroxyhydrocinnamic acid, quinic acid, 3,4-dihydroxyphenylacetic acid, and 3,4-dihydroxycinnamic acid, since these compounds were not reported to affect fMLP-activated signal transduction pathways in human PMNs (Ciz et al., 2012) and did not affect calcium response. In our previous study, we reported the highest antioxidant activity expressed as the ability to reduce Fe³⁺ ions (the FRAP method) for the following phenolics (in descending order at a concentration of 5 µM): 3,4-dihydroxyphenylacetic acid, quercetin, gallic acid, 3,4-dihydroxyhydrocinnamic acid, catechin, 3,4-dihydroxycinnamic acid, ferulic acid, 3,4-dihydroxybenzoic acid, chlorogenic acid, catechol, phloroglucinol, and phlorizin. Quinic acid and 4-hydroxybenzoic acid had no FRAP at this concentration (deGraft-Johnson et al., 2007). Assuming that compounds with a higher FRAP effectively scavenged ROS, the inhibitory effect of quinic acid and catechol on fMLP-LBCL can perhaps be explained in part by their interference in signal transduction pathways triggered by fMLP. In contrast, catechin presenting relatively high FRAP had no significant effect on fMLP-LBCL. Plant polyphenols and their metabolites can bind to erythrocytes and possibly to other circulating cells to form active antioxidant complexes in the very surface of these cells (Koren et al., 2010). This may explain to some extent the aforementioned discrepancies between phenolic FRAP and their inhibition of fMLP-LBCL. Another mechanism that may contribute to phenolic-induced suppression of fMLP-LBCL is the inhibition of myeloperoxidase activity in PMNs. Agonist-induced LBCL depends to a great extent on the myeloperoxidase-catalyzed conversion of H₂O₂ into HOCl which subsequently reacts with luminol to emit light (Bednar et al., 1996). Some phenolics (e.g. quercetin and kaempferol) can inhibit myeloperoxidase activity at micromolar concentrations (Pincemail et al., 1988; Shiba et al., 2008). Moreover, phenolics containing methylated catechol groups can serve as NADPH oxidase inhibitors after intracellular modification of their chemical structure by peroxidase (Steffen et al., 2007; Stolk et al., 1994; Vejrazka et al., 2005). However, this mechanism seems less likely to be responsible for fMLP-LBCL inhibition due to the relatively short preincubation time of blood samples with the studied compounds under the conditions of our experiments. In summary, half of the tested plant phenolics inhibited fMLP-LBCL. Among them, two most active compounds (catechol and quercetin) also suppressed Ca²⁺ response of human PMNs. Inhibition of fMLP-LBCL may result from both mechanisms: blockade of some steps of the intracellular signal transduction (e.g. increase in [Ca²⁺]_i) and/or unspecific scavenging of ROS generated by blood phagocytes in response to fMLP stimulation. In a study on rat PMNs, certain phenolics were unequally potent in suppressing chemiluminsecence of stimulated cells and scavenging radicals in cell-free models, which also suggests more than one mechanism of action (Ren et al., 2014).

Chemoattractant activity of polyphenols for human granulocytes

Four compounds (3,4-dihydroxyhydrocinnamic acid, 3,4-dihydroxyphenylacetic acid, 4-hydroxybenzoic acid, and catechin) at a concentration of 5 µM revealed a distinct chemoattractant activity for human PMNs. However, this activity at a concentration of 0.1 µM was many times lower in comparison with that of fMLP at an equimolar concentration. This is the first report (to the best of our knowledge) describing the chemoattractant properties of plant phenolics for human PMNs. Recently, inhibition of rat neutrophil chemotaxis toward a known chemoattractants (fMLP, lipopolysaccharide) by certain flavonoids including quercetin has been reported, but the direct chemoattractant activity of these phytochemicals was not investigated (Suyenaga et al., 2011). In contrast, the improvement of leukocyte functions including the chemotaxis of macrophages and lymphocytes was observed in mice after a long-term diet supplementation with polyphenol-rich biscuits containing different amounts of catechin, p-hydroxybenzoic acid, vanillic acid, p-coumaric acid, ferulic acid, rutin, and oryzanol (De la Fuente et al., 2011).

fMLP is a strong chemoattractant for human PMNs and like other certain chemoattractants beside inducing cells motility, it causes an increase in [Ca²⁺]_i, respiratory burst, and degranulation, augmenting LBCL. These effects are similar to that revealed by other inflammatory mediators such as leukotriene B4 and interleukin-8 (Futosi et al., 2013). However, none of the four chemoattractant phenolics induced an increase of [Ca²⁺]_i in resting PMNs or stimulated LBCL. Nevertheless, two of them (3,4-dihydroxyhydrocinnamic acid and 3,4-dihydroxyphenylacetic acid) inhibited fMLP-LBCL. Parallel attraction of PMNs and inhibition of their respiratory burst indicates the complexity of action. Induction of directional movement of PMNs without any stimulatory effect on other cell functions is not a unique property among the chemoattractant substances. For instance, bradykinin and innate defense-regulator peptide (IDR-1) both attracted human PMNs, while the first compound simply had no effect on [Ca²⁺]_i, the second compound had no effect on both [Ca²⁺]_i and superoxide formation (Ehrenfeld et al., 2006; Lee & Bae, 2008). Likewise, some phenolics can selectively switch on the transmission of signals leading to intracellular actin reorganization and induction of PMNs migration; however, elucidation of this activity requires further studies.

Relevance of in vitro results to in vivo conditions

The fundamental question concerning our in vitro experiments is whether the polyphenol concentrations that affected the function of PMNs can occur in vivo in human body fluids. Results of numerous interventional and observational clinical studies suggest that these concentrations can be found in human plasma after consumption of meals containing fruits and vegetables or drinking tea, coffee, and other beverages rich in polyphenols. Maximal plasma concentrations of phenolics according to available data were as follows: 3,4-dihydroxycinnamic acid 0.5 µM, 3,4-dihydroxyhydrocinnamic 0.3 µM, 3,4-dihydroxybenzoic 0.5 µM, ferulic acid 0.2 µM, chlorogenic acids 8 µM, gallic acid 2.1 µM, quercetin 4 µM, and catechin 1.6 µM (Guy et al., 2009; Lafay & Gil-Izquierdo, 2008; Manach et al., 2005; Vitaglione et al., 2007). Higher levels can be expected in the portal circulation, intestinal wall, and the liver, where phenolic compounds are absorbed and metabolized. Moreover, in the case of repeated ingestion of polyphenol-rich foods, the exposure of PMNs to these compounds would last much longer in vivo than in our experiments, which indubitably augments their effects. This situation can be of particular importance in the case of quercetin, which undergoes enterohepatic circulation (Moon et al., 2008) with a plasma half-life that extends from 11 to 28 h (Manach et al., 2005). Even supposing that this compound undergoes glucuronidation in the liver, it can influence PMNs in this form (Suri et al., 2008); furthermore, these cells have beta glucuronidase activity, which may deconjugate quercetin (Shimoi & Nakayama, 2005).

Our results are in accordance with some previous studies on the effect of phenolics on luminol and lucigenin-enhanced chemiluminescence of isolated human PMNs after stimulation with various agonists including fMLP. Limasset et al. (1999) described the inhibitory activity of phenolics at a concentration of 10 µM; however, this concentration is unequivocally higher than the circulating levels of these compounds in humans. In addition, they did not find any inhibitory effects of phenolics at lower concentrations. The differences in our study may result from the use of a more sensitive and stable chemiluminometer together with the measurement of the oxidative response of PMNs in the diluted whole blood system, which avoids an unspecific activation of PMNs during the isolation procedure (Kukovetz et al., 1997). While Ribeiro et al. (2013) studied the effects of polyphenols on isolated PMNs testing compounds at concentrations close to the ones used in our study, their study only investigated properties of flavonoids (e.g. quercetin), not phenolic acids. Furthermore, in their approach, PMNs were stimulated with phorbol 12-myristate 13-acetate (PMA), a relatively non-physiological activator of PMNs which directly activates protein kinase C, omitting the G-protein-coupled receptor (e.g. formyl peptide receptor) and phospholipase C, which normally transmits a signal after stimulation with fMLP (Castagna et al., 1982; Futosi et al., 2013).

It is possible that the most active compounds as inhibitors of fMLP-LBCL (e.g. catechol, quercetin, and gallic acid) can decrease in vivo, the oxidative stress related to excessive ROS production by PMNs. The known biological effects regarding the consumption of quercetin and gallic acid demonstrated in various clinical studies are in accordance with this assumption. Increased quercetin intake reduced circulating levels of various markers of oxidative stress: oxidized LDL, malondialdehyde, and advanced oxidation protein products (Boots et al., 2011; Egert et al., 2009; Valentova et al., 2007). Gallic acid used topically on the skin showed evidence of anti-aging activities when measured by skin elasticity (Manosroi et al., 2011). Oral consumption of green tea catechins resulted in increased concentrations of gallic acid in skin effusions and a reduced skin susceptibility to UV radiation (Rhodes et al., 2013). There is a lack of sufficient data regarding the antioxidant effect attributed to the consumption of catechol and catechin, in addition to phenolic acids such as 3,4-dihydroxyphenylacetic, 3,4-dihydroxyhydrocinnamic, and 4-hydroxybenzoic acid. The main dietary sources of these compounds include coffee, cocoa, and olives, which have often proved beneficial for human health post-consumption; however, considering the multiple content of phytochemicals in these products, the effects cannot be attributed to one compound.

Conclusions

Certain phenolics can suppress the fMLP-stimulated oxidative response of human PMNs in vitro. This inhibitory effect was observed at concentrations that can occur in human plasma, a mechanism of two of the compounds (quercetin and catechol) could result from inhibition of increased [Ca²⁺]_i preceding the respiratory burst of PMNs. Therefore, this is suggestive of a feasible decrease in the risk of a systemic imbalance between oxidants and antioxidants; it is also likely that this is perhaps one of the mechanisms underlying the health-promoting effect of fruits and vegetables.

Declaration of interest

The authors report that there are no declarations of interest. This study was supported by a research grant from the European Regional Development Fund through the Polish Innovative Economy Operational Program, contract N. UDAPOIG. 01.03.01-10-109/08-00 and grants from the Medical University of Lodz no. 503/0-079-01/503-01 and 503/1-151-02/503-01. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.