Food & Function Accepted Manuscript This is an Accepted Manuscript, which has been through the Royal Society of Chemistry peer review process and has been accepted for publication. Accepted Manuscripts are published online shortly after acceptance, before technical editing, formatting and proof reading. Using this free service, authors can make their results available to the community, in citable form, before we publish the edited article. We will replace this Accepted Manuscript with the edited and formatted Advance Article as soon as it is available. You can find more information about Accepted Manuscripts in the Information for Authors. Please note that technical editing may introduce minor changes to the text and/or graphics, which may alter content. The journal’s standard Terms & Conditions and the Ethical guidelines still apply. In no event shall the Royal Society of Chemistry be held responsible for any errors or omissions in this Accepted Manuscript or any consequences arising from the use of any information it contains. www.rsc.org/foodfunction Page 1 of 27 Food & Function 1 Polyphenols from artichoke heads (Cynara cardunculus (L.) subsp. scolymus Hayek): in vitro bio- 2 accessibility, intestinal uptake and bioavailability. 3 4 Isabella D’Antuonoa, Antonella Garbettaa, Vito Linsalataa, Fiorenza Minervinia, Angela Cardinalia*. 5 a Institute of Sciences of Food Production (ISPA), CNR, Via G. Amendola, 122/O, 70126 Bari, Italy t p 6 i r c 7 Corresponding author at: Institute of Sciences of Food Production, Via Amendola 122/O, 70126 Bari, s u 8 Italy. Telephone: +39 080 5929303, Fax: +39 080 5929374. n a 9 E-mail address: [email protected] (A. Cardinali) M 10 d 11 e t p e c c A n o i t c n u F & d o o F 1 Food & Function Page 2 of 27 12 Abstract 13 Artichoke is a rich source of health promoting compounds such as polyphenols, important for their 14 pharmaceutical and nutritional properties. In this study, the potential for bioavailability of the 15 artichoke polyphenols was estimated by using both in vitro digestion and Caco-2 human intestinal 16 cell models. In vitro digestive recoveries (bio-accessibility) were found to be 55.8% for total t p 17 artichoke phenolics and in particular, 70.0% for chlorogenic acid, 41.3% for 3,5-O-dicaffeoylquinic i r c 18 acid, and 50.3% for 1,5-O-dicaffeoylquinic acid, highlighting potential sensitivity of these s u 19 compounds to gastric and small intestinal digestive conditions. Uptake of artichoke polyphenols was n a 20 rapid with peak accumulation occurring after 30 min with an efficiency of 0.16%, according to the M 21 poor uptake of dietary polyphenols. Some compounds, such as coumaric acid, caffeic acid and caffeic d 22 acid derivatives, were also detected in the basolateral side assuming an extra and intracellular e t p 23 esterases activities on chlorogenic acid. Only apigenin-7-O-glucoside was absorbed and transported e c 24 through the Caco-2 monolayer demonstrating its bioavailability in the extent of 1.15% at 60 min. In c 25 addition, permeability coefficient (P =2.29 x 10-5 cm/sec), involving apical to basolateral transport A app n 26 of apigenin 7-O-glucoside, was calculated to facilitate estimation of absorption and transport through o 27 Caco-2 monolayer. Finally, the mono and dicaffeoylquinic acids present in artichoke heads, exert an i t c 28 antioxidant activity on human low density lipoprotein system correlated to their chemical structure. n u 29 In conclusion, the utilized in vitro models, although not fully responding to the morphological and F 30 physiological features of human in vivo conditions, could be a useful tool for investigating & d 31 mechanistic effects of polyphenols released from food matrix. o 32 o F 33 Keywords: Artichoke Polyphenols, in vitro digestion, Bioavailability, Permeability Coefficient 34 35 36 2 Page 3 of 27 Food & Function 37 Introduction 38 Artichoke (Cynara cardunculus (L.) subsp. scolymus Hayek) represents an important component of 39 Mediterranean diet and a good source of health-promoting compounds, such as phenolics, inulin, 40 fibres and minerals.1,2 The main compounds present in artichoke heads are caffeic acid derivatives, in 41 particular a wide range of caffeoylquinic acids with chlorogenic acid (CGA) as the most abundant of t p 42 them. In addition, other compounds present in small amount, such as glycosides of apigenin and i r c 43 luteolin and different cyanidin caffeoylglucoside, have been identified in artichoke tissues.1,3,4 s u 44 The pharmaceutical properties of artichoke polyphenols are widely studied and attributed to n a 45 many activities, such as hepatoprotective, anticarcinogenic, antioxidative, antibacterial, anti-HIV, M 46 bile-expelling, and diuretic.1 Others studies, already performed, have found evidence for d 47 antioxidative properties of artichoke leaf extracts against hydroperoxide-induced oxidative stress in e t p 48 cultured rat hepatocytes.5 In addition, the artichoke extract has shown to inhibit cholesterol e 49 biosynthesis and to protect low density lypoprotein (LDL) from in vitro oxidation.6-8 Nevertheless, to c c A 50 achieve any health properties, the polyphenols must be bioavailable, effectively absorbed from the n 51 gut into the circulation, and delivered to the target tissues where can exert their beneficial effects.9,10 o 52 The bioavailability (the fraction of a nutrient or compound ingested that, through the systemic i t c 53 circulation, reaches specific sites) is dependent upon the digestive stability of compound, its release n u 54 from the food matrix (referred as bio-accessibility), and the efficiency of its transepithelial passage.11 F 55 After the release from the food matrix, the bio-accessible polyphenols must be presented to the & d 56 brush-border of the small intestine in such a state that they can be absorbed into the enterocyte by o 57 passive diffusion or by active transport systems.12 Passive paracellular diffusion, overcoming the o F 58 tight junctions, may also occur, but this is not normally a major route of ingress, probably because 59 most polyphenols are too hydrophilic to penetrate the gut wall.12,13 The polyphenols bioavailability 60 differs greatly from a one molecule to another, and the most abundant in our diet is not necessarily 61 the better bioavailable.14, Moreover, the polyphenols bioavailability is considered to be low, not 3 Food & Function Page 4 of 27 62 exceeding the plasma concentrations of 10 µM, and their low absorption can be attributed to 63 glucuronidation and sulphation of free hydroxyl groups present in the chemical structures of the 64 different compounds.14,15 However, a part of their low bioavailability, dietary polyphenols, after a 65 meal rich in vegetables and fruit, may be present in the gastro-intestinal (GI) lumen, at much greater 66 concentrations where they can play an important role in protecting from oxidative damage and in t p 67 delaying the development of stomach, colon and rectal cancer.16 i r c 68 Although little is known about the in vivo bioavailability and digestive modification of artichoke s u 69 polyphenols,17,18 the in vitro digestive models were used to predict, in a simplified manner, the n a 70 polyphenols behavior in simulated digestive processes of GI tract. As reported by some authors, these M 71 models can provide important information on the stability and putative modifications of interest d 72 compounds, under GI conditions.19 In fact, Falè et al. (2013)19, have investigated on the composition, e t p 73 antioxidant activity, and stability of polyphenols present in artichoke infusion after GI digestion, e c 74 reporting the high stability of the identified flavonoids.19 Furthermore, a recent study performed by c A 75 our group, have evaluated the influence of gastro-intestinal digestion on antioxidant effect of n 76 artichoke polyphenols showing that, in vitro digestion did not modify the antioxidant activity of o 77 artichoke polyphenols, except for 1,5-O-dicaffeoylquinic acid (1,5 diCQA)20. On the other hand, ti c 78 many studies are already performed on bioavailability of pure standards, such as CGA, showing its n u 79 high stability also to the extreme gastric conditions. The CGA could be absorbed even in the F 80 stomach, in fact it was identified in both the gastric vein and aorta in its intact form.21 Instead, its & 81 bioavailability and metabolism were mainly dependent by gut microflora.22,23 d o 82 Conversely to previously reported studies performed mainly on standard compounds, or o F 83 extracts, or infusions, this paper aims to simulate a physiological digestion process and intestinal 84 absorption of bioactive compounds, such as polyphenols, from a fresh vegetable matrix. In particular 85 the digestive stability and bio-accessibility of the major classes of polyphenols present in artichoke 86 heads, using the in vitro digestion model were assessed. The influence of some chemical (pH, 4 Page 5 of 27 Food & Function 87 temperature and bile salts) and biological (gastric and pancreatic enzymes) GI conditions on the 88 artichoke polyphenols modifications, were investigated. Moreover, the intestinal absorption (as 89 predictors of bioavailability) was performed using Caco-2 cell line model. Permeability coefficient, 90 involving apical to basolateral transport of polyphenols, was calculated to facilitate estimation of 91 absorption and transport through Caco-2 monolayer. t p 92 i r c 93 Material and Methods s u 94 Materials n 95 Artichoke heads were supplied from a local market and stored at 4 °C until used. Extraction and a M 96 chromatography solvents, methanol (MeOH), glacial acetic acid (AcOH), ethanol (EtOH), ethyl d 97 acetate (EtOAc), were HPLC certified. Dulbecco’s modified Eagle’s medium (DMEM), Dulbecco’s e t p 98 phosphate-buffered saline (PBS), L-glutamine 200 mM, antibiotic and antimycotic solution, non e 99 essential amino acid solution, bovine serum albumin Cohn V fraction fatty acid depleted (BSA) were c c 100 purchased from Sigma Aldrich (Milan, Italy). Caco-2 (HTB-37) cell line was purchased from A 101 IZSLER (Brescia, Italy). Foetal bovine serum (FBS) was purchased from Gibco (Milan, Italy). The n o 102 enzymes:, α-amylase (from Bacillus species; catalogue n. A-6814), pepsin (from pig; catalogue n. P- i t c 103 7000), pancreatin (from pig; catalogue n. P-1750), mucin (from pig; catalogue n. M-2378), lipase n u 104 (from pig; catalogue n. L-3126), bovine bile extract (catalogue n. B-3883) used in vitro digestion F 105 were obtained from Sigma Aldrich (St. Louis, Mo., U.S.A.). The mono and dicaffeoylquinic acids & 106 used in this study were supplied by PhytoLab GmbH & Co. KG (Dutendorfer Str. 5-7, 91487 d o 107 Vestenbergsgreuth Germany). CGA, CAA, and LDL from human plasma, were purchased from o F 108 Sigma Aldrich, Milan, Italy. 109 110 Artichoke polyphenols extractions 5 Food & Function Page 6 of 27 111 The polyphenolic fraction presents in artichoke was extracted by using water, the solvent that better 112 simulate the extraction process in the digestive system. In particular, 4.5 g of blanched artichoke 113 heads (5 min, 100°C, ascorbic acid 0.5% in H O) were extracted by refluxing for 60 min at 100 °C 2 114 with 50 mL of H O containing 0.5% of ascorbic acid. Then, the aqueous solution was recovered and 2 115 further extracted for additional 30 min with 50 mL of the same solution. The extracts were filtered t p 116 through a Whatman 1 paper, pooled, filtered at 0.45 µm and utilized for HPLC analysis. i r c 117 In order to avoid the influence of ascorbic acid on biological activity and on polyphenols absorption, s u 118 24,25,26,27 another extract (hydroalcoholic extract) was obtained and utilized for the uptake experiments n a 119 and for the antioxidant activity assay (LDL). In particular, 4.5 g of blanched artichoke heads were M 120 extracted by refluxing with 50 mL of methanol/water (50:50, v/v), for twice (1 h and 30 min at d 121 100°C). The extracts were filtered through a Whatman 1 paper, pooled, filtered at 0.45 µm and stored e t p 122 at -20°C until analysis. e c 123 c A 124 n 125 HPLC Analysis o 126 Analytical-scale HPLC analyses of the artichoke extracts were performed employed Thermo i t c 127 Scientific HPLC spectra System equipped with a P2000 gradient pump, a SCM 1000 membrane n u 128 degasser, an UV6000LP UV/Vis DAD, an AS3000 autosampler, and ChromQuest 4.1 software. The F 129 UV–Vis absorption chromatogram was detected at 325 nm. Separation was performed by gradient & d 130 elution on a 4.6 × 250 mm reverse phase Luna C-18 (5 µm) column (Phenomenex Torrance, o 131 California, USA). The elution was performed using methanol (eluent A) and water/acetic acid 95:5 o F 132 (eluent B) following the method of Lattanzio.28 The gradient profile was: 85–60% B (0–25 min), 133 60% B (25–30 min), 60–37% B (30–45 min), 37% B (45–47 min), 37–0% B (47–52 min). The flow 134 rate was 1 mL/min. Samples were applied to the column by means of a 25 µL loop valve. 135 Polyphenols compounds were identified by retention time and spectra of the pure standard, apart the 6 Page 7 of 27 Food & Function 136 identification of 1-O-caffeoylquinic and 1,4-O-di caffeoylquinic acids, that was performed by 137 spectrum analysis and following the classification of Lattanzio et al. 20091. The polyphenols 138 concentrations were expressed as µg/mL calculated using their reference standards. Instead, in the 139 absence of reference standards, 1-O-caffeoylquinic was quantified as µg/mL of chlorogenic acid 140 equivalent, and 1,4-O-dicaffeoylquinic acid as µg/mL of 1,5-O-dicaffeoylquinic acid equivalent. t p i 141 r c 142 LDL oxidation in vitro assay s u 143 LDL oxidation was measured by monitoring the formation of hexanal, which is the major end n a 144 product of lipid peroxidation. LDL in PBS dispersion was diluted to a concentration of 1 mg of M 145 protein/mL. The production of hexanal was monitored by headspace, following the method of d e 146 Teissedre et al.29 with some modifications. Briefly, in 10 mL vial were added 50 µL of LDL samples, t p 147 CuSO (80 µM) solution and PBS to reach the volume of 4 mL, finally the vial was sealed and 4 e c 148 incubated for 2 h (propagation phase) at 37 °C, in order to determine the production of hexanal c A 149 formed in the control. At the same time, various concentrations of polyphenols from artichoke n 150 hydroalcoholic extract (2-20 µg/mL), caffeic acid (CAA) (0.2-1.5 µg/mL), and CGA (0.3-7.0 µg/mL) o i 151 were tested for their antioxidant activity. The hexanal formation was determined using Gas t c n 152 Chromatography (Varian CP3800) equipped with a flame ionization detector. Hexanal was separated u F 153 by a ZB-Wax-Plus fused silica Capillary column (30m x 0.32 mm i.d., 0.5 µm film thickness, Zebron & 154 Phenomenex Inc. Torrance, CA U.S.A) and helium was the carrier gas. GC conditions were as d 155 follows: injector temperature, 180 °C; detector temperature, 200°C; oven program, held at 40 °C for o o 156 2 min, increased at 20 °C/min to 140 °C, and then held for 1 min. The results, obtained after replicate F 157 analyses, were expressed as percent of relative inhibition: 158 (% In) = [(C - S)/C] x 100 159 where C was the amount of hexanal formed in the control and S was the amount of hexanal formed in 160 the sample. 7 Food & Function Page 8 of 27 161 162 In vitro gastro-intestinal digestion 163 Artichoke heads were subjected to gastric and pancreatic digestion, following the method of 164 Versantvoort et al. 30 Before to start, all the simulated digestive juices are heated to 37 °C for 2 h. 165 Artichoke head was blanched for 5 min at 100 °C in H O containing 0.5% of ascorbic acid, and 2 t p 166 homogenized in a laboratory blender for 1 min to simulate mastication. Homogenized samples (4.5 g) i r c 167 were transferred to a centrifuge tube and 6 mL of simulated saliva fluid, containing 0.0145% α- s u 168 amylase (w/v), 0.005% mucin (w/v) and, as reported in Table 1, several organic and inorganic salts n a 169 were added. Finally, the pH was adjusted, if necessary, at pH 6.8 ± 0.2, then the solution was M 170 incubated at 37 °C and rotated head-over-heels (55 rpm at 37 °C) (Rotator Type L2, Labinco BV, d e 171 Netherlands) for 5 min. Then, 12 mL of simulated gastric juice, containing 0.1% of pepsin (w/v), t p 172 0.3% of mucin (w/v) and, as reported in Table 1, several organic and inorganic salts, were added and e c 173 finally, the pH was adjusted, if necessary, at pH 1.5 ± 0.5. The mixture was rotated head-over-heels c A 174 for 2 h. .Finally, 12 mL of duodenal juice, containing 0.3% of pancreatin (w/v), 0.05% of lipase (w/v) n 175 and 6 mL of bile, containing 0.6% of bile (w/v) and, as reported in Table 1, several organic and o i 176 inorganic salts, were added. Finally, the pH was adjusted, if necessary, at pH. 6.5 ± 0.5, and the t c 177 mixture was rotated for another 2 h. Using the head-over-heels rotation in each steps of digestion, a n u 178 gentle but thorough mixing of the matrix with the digestive juices was achieved, simulating the F & 179 peristaltic movement. At the end of the in vitro digestion process, the samples were centrifuged for d 180 10 min at 2,900 xg and an aliquot of the supernatant (chyme) was recovered for the assessment of the o 181 bio-accessibility. During all the digestive process, different aliquots of samples in the different steps o F 182 (salivary, gastric and duodenal) were recovered in order to determine the polyphenols stability. Three 183 independent experiments were performed in duplicate. 184 The bio-accessibility of polyphenols, defined as the fraction of external dose released from its matrix 185 in the GI tract, was calculated as follows: 8 Page 9 of 27 Food & Function 186 Bioaccessibility (%) = (CF/CI) x 100 187 Where CF is the amount of polyphenols present in the digesta (chyme) and CI is the initial amount of 188 polyphenols present in artichoke head. 189 190 Intestinal bioavailability of polyphenols by using Caco-2 human cell line t p 191 To assess the potential intestinal absorption of artichoke polyphenols, experiments were carried out i r c 192 using the Caco-2 human intestinal cell line, following the method described by Failla et al. 31 with s u 193 some modifications. Briefly, Caco-2 cells were seeded at 1.2x105 cells/mL in cell culture inserts for 6 n a 194 well plates with polyethylene terephtalate (PET) track-etched membranes (pore size 0.4 µm, growth M 195 area 4.2 cm2, Falcon, BD), pretreated with poly-L-lysine (50 µg/mL), in complete DMEM, with 4.5 d 196 g/L glucose supplemented with 10% foetal bovine serum, 1% L-glutamine, 1% antibiotic and e t p 197 antimycotic solution, 1% non essential amino acid solution at 37 °C in a humidified atmosphere e c 198 containing 5% CO . The basolateral compartment was filled with 3 mL of complete DMEM. Cells 2 c A 199 monolayers were cultured for 21 days in order to obtain a full differentiated cells and media from n 200 apical and basolateral compartment were replaced twice a week. The integrity of the cells monolayer o 201 was evaluated by transepithelial electrical resistance (TEER) measurements using a volt-ohm meter i t c 202 (Millicel ERS-2, Millipore, Italy). TEER values were expressed as Ω/cm2. Only Caco-2 monolayers n u 203 showing TEER values higher than 700 Ω/cm2 were used for in vitro experiments. The absorption F 204 experiments were performed following protocol described by Neilson et al.32 with some & d 205 modifications. Briefly, monolayers were first washed with 2 mL PBS (pH 5.5), and then 2 mL of o 206 DMEM phenol red free, containing the hydroalcoholic artichoke extract at final polyphenols o F 207 concentration of 100 µg/mL, was applied to each well. Cells were then incubated at 37 °C for 30, 60, 208 90 and 120 min. Following incubation, media of apical and basolateral compartments were aspirated 209 and stored at -80 °C before the HPLC analysis. 210 9