1 Despite the economic importance and the diffusion of grapevine cultivation worldwide, little is known 2 about leaf chemical composition. We characterized the phenolic composition of Nebbiolo, Barbera, 3 Pinot noir, Cabernet Sauvignon, Grenache and Shiraz (Vitis vinifera L.) healthy leaves (separating blades 4 and veins) during the season. Quantitative and qualitative differences were found between leaf sectors 5 and among genotypes. In healthy grapevine leaves, anthocyanins,

more »

... hamnoside, 6 hexosides of dihydroquercetin and dihydrokaempferol exclusively accumulated in veins. Astilbin was 7 the only flavanonol detected in blades and the prevalent flavanonol in veins. Barbera distinguished for 8 the lowest proanthocyanidin and the highest hydroxycinnamate content; Pinot noir for the absence of 9 acylated-anthocyanins. Nebbiolo, Pinot noir and Cabernet Sauvignon displayed high concentration of 10 epigallocatechin gallate. Nebbiolo leaves showed the highest concentrations of flavanonols and the 11 widest profile differentiation. Knowledge derived from the present work is a contribution to find out leaf 12 polyphenol potential as a part of grapevine defense mechanisms and to dissect genotype-related 13 susceptibility to pathogens; moreover, it represents a starting point for future deepening about grapevine 14 and vineyard by-products as a source of bioactive phenolic compounds. 15 DAD-UV-MS/MS. 18 Grapevine (Vitis vinifera L.), one of the most widely cultivated plant species worldwide, 19 comprises 5000 to 10000 varieties 1 and it plays important role in the economy of many countries due to 20 wine, and fresh and dry grape production. Grapevine vegetative organs (shoots, stems and leaves) are 21 used in traditional plant-based medicine as a source of bioactive compounds. 2,3 According to recent 22 studies, grapevine leaves have beneficial effect on human health due to their anti-inflammatory, 23 antibacterial, anticancerogenic, antiviral, antioxidant properties. 4 In the Middle East and Mediterranean 24 regions, grapevine leaves are commonly used as food both in fresh and brined forms. 5 25 Grapevines produce large amount of secondary metabolites, including chemically heterogeneous 26 phenolic compounds. Due to this huge diversity, each group of phenolic compounds displays various 27 roles in grapevine biology and ecology, conferring them a key role in grapevine adaptation to the 28 environment. Polyphenols are part of the plant-defence mechanisms relying on molecular 29 communication among plants and pathogens, involving signals for the establishment of infection, the 30 activation of plant disease-resistance genes, the formation of elicitors, the activation of elicitor receptors 31 and, finally gene regulation. In many of these steps phenylalanine ammonia lyase (PAL) and the 32 chalcone synthase genes (CHs) are suppressed or over-expressed, resulting in the modulation of the 33 accumulation of main classes of polyphenols. 6 Accumulation of polyphenols varies among plant organs, 34 tissues and phenological stages. Many traits of the phenolic compound biosynthesis in grapevine berries 35 are well detailed and it is well-known that they are under genetic control, even though external abiotic 36 or biotic factors can influence polyphenolic concentrations and, sometimes, profiles. At the berry level 37 the wide differences in the polyphenolic composition of Vitis vinifera varieties and clones have been 38 investigated. 7-9 Polyphenol accumulation and profiles are influenced by seasonal climatic conditions, 39 biotic and abiotic stressors, soil and cultural practices. Nevertheless, some traits are genetically 40 determined, thus specific quantitative and qualitative chemical patterns characterize Vitis vinifera 41 varieties. In berries, the ratio between tri-hydroxylated and di-hydroxylated anthocyanins and the ratio 42 4 between caftaric acid and coutaric acid are stable and they have long time been proposed as tools to 43 classify Vitis vinifera varieties and clones. 7-9 Much less is known about vegetative organ polyphenolic 44 composition, even though specific molecules or groups of molecules could be responsible of the inner 45 and constitutive biochemical protection of the vine against various abiotic and biotic stressors. 10 46 Increasing knowledge about constitutive leaf polyphenols could be pivotal to explain the different level 47 of susceptibility to pathogens displayed by Vitis vinifera genotypes. Different compositional traits and 48 changes during the season in leaf compartments (blades and veins) can provide new insights about the 49 interpretation of plant interaction with pathogens specifically accumulating in these two different leaf 50 sectors. The present work investigates the polyphenolic concentration and profiles of grapevine leaves 51 during the vegetative season to individuate characteristic chemical patterns in some Vitis vinifera 52 varieties and to explore their constitutive accumulation as a part of grapevine defense potential 53 mechanism. To provide new insights about concentrations, profiles and trends of main polyphenols in 54 different leaf tissue, we analyzed blades and veins separately to spread light, in particular, on the vein 55 constitutive polyphenols that could help to understand the different susceptibility of Vitis vinifera 56 varieties to pathogens with vascular localization. To our knowledge, little is known about the 57 polyphenolic characterization of Vitis vinifera leaf blades and veins analyzed separately and about their 58 evolution during the vegetative season. Leaf polyphenols were analyzed spectrophotometrically and by 59 targeted analytical approach using HPLC-DAD for quantitative or semi-quantitative purposes and 60 HPLC-ESI-MS/MS for molecular identification. 61 MATERIALS AND METHODS 62 Plant material 63 The leaves of two major Italian varieties (Barbera -BR and Nebbiolo -NE) and of four 64 international varieties (Pinot noir -PN, Cabernet Sauvignon -CS, Grenache -GR and Shiraz -SH) were 65 sampled in the collection vineyard of DISAFA, University of Turin located at Grugliasco (45°03'N, 66 7°35'E; in Piedmont, Italy), in 2015. Vine density was 4400 vines/ha (0.90 m x 2.50 m), vines were 67 5 planted in 2008, vertical shoot positioned and trained to the Guyot pruning system. The vineyard is 68 located at 293 m above s.l., in a plain area. A detailed soil description is reported in Catoni et al. 11 69 Briefly, the A horizon pH was 7.9, organic C was 14.8 g kg -1 , sand was 882 g kg -1 , silt was 101 g kg -1 70 and clay 17 g kg -1 . The vineyard was organized in randomized blocks of maximum twelve vines each. 71 Leaf samples were collected at five different time points: 1 = 22 th of May (142 day of the year, DOY 72 142), 2 = 2 nd of July (DOY 183), 3 = 16 th of July (DOY 197), 4 = 29 th of July (DOY 210), 5 = 26 th of 73 August (DOY 238) in 2015. The general meteorological parameters of the vineyard are reported in 74 Supplementary Table 1. Three adult healthy leaves between the fourth and the seventh node of main 75 shoots per each block were collected from the west side of the row and immediately transported to the 76 laboratory where leaves were rinsed, dried with a paper before blades and veins separation and 77 extraction. 78 Dry matter content 79 Leaf tissue dry matter was measured gravimetrically by drying inside an oven at 110 ºC for 72 80 hours. 81 Sample extraction 82 Notwithstanding the well-known effects of water content on polyphenol final concentrations, we 83 decided to work on fresh leaves, immerging blades and veins in an appropriate and specifically chosen 84 extraction solvent (see below) soon after picking as freeze drying, including lyophilisation, can 85 imperfectly preserve plant secondary metabolites, particularly polyphenols, as previously reviewed. 12 86 To ascertain the most adequate extraction solvent for leaf polyphenol analyses, we extracted three 87 biological replicates of Nebbiolo blades and veins in seven different solvents: CH3OH 80%; CH3OH 88 80%/HCl 0.1%; acetone 50%; acetone 50%/HCl 0.1%; phosphate-citrate buffer (pH 3.6); hydroalcoholic 89 buffer (ethanol 12%, pH 3.2) and hydroalcoholic buffer (ethanol 40%, pH 3.9). This last gave the best 90 results (see Results) thus two grams of leaf blades and two grams of leaf veins were extracted in 25 mL 91 of this pH 3.9 hydroalcoholic buffer (40% ethanol, 2 g/L of Na2S2O5, 5g/L of tartaric acid, 22 mL/L of 92 6 1 N NaOH). The samples were homogenized with an Ultraturrax dispersing machine (IKA, Staufen, 93 Germany), centrifuged for 10 min at 4000 rpm. The supernatant was separated and kept in the dark. The 94 pellet was re-suspended in 20 mL of the same buffer; the resuspension was macerated for 30 minutes at 95 room temperature in the dark and then centrifuged again. The two extracts were combined and brought 96 to a final volume of 50 mL. Extracts were stored at -20 ˚C until further analysis. 97 Reagents and Standards 98 Bovine serum albumin (BSA), sodium hydroxide, triethanolamine (TEA), and urea were 99 purchased from Sigma-Aldrich S.r.l. (Milan, Italy). Folin-Ciocalteu reagent and tartaric acid were 100 purchased from Merck (Darmstadt, Germany). Sodium sulfate and sodium metabisulfite were purchased 101 from BDH Laboratory Supplies (Poole, England). Quercetin 3-O-glucoside, quercetin 3-O-glucuronide, 102 kaempferol 3-O-glucoside, kaempferol 3-O-glucuronide, myricetin 3-O-glucoside, isorhamnetin 3-O-103 glucoside, malvidin 3-O-glucoside, (+)-catechin, (-)-epicatechin, (-)-epicatechin gallate, (-)-104 epigallocatechin gallate, proanthocyanidin B1 and proanthocyanidin B2 were purchased from 105 Extrasynthèse (Genay, France). Astilbin and trans-caftaric acid were purchased from Sigma-Aldrich 106 S.r.l. (Milan, Italy); trans-fertaric acid and trans-coutaric acid were purchased from Phytolab 107 (Vestenbergsgreuth, Germany). 108 Spectrophotometric analyses 109 Total polyphenols (TP) in grapevine leaves were measured with the Folin-Ciocalteu reagent. 110 Absorbance was read at 760 nm in a UV/Vis spectrophotometer (Perkin Elmer, Lambda 25, 111 Beaconsfield, Bucks, U.K.) and TP were expressed as grams of (+)-catechin equivalents (CE) per kg of 112 leaf blade/vein fresh weight (FW). 113 Measurement of total proanthocyanidins (PA) in leaves was performed spectrophotometrically 114 by the improved protein precipitation method of Harbertson et al. 13 Briefly, 1 mL of BSA protein 115 solution was added to 500 µL of sample extract for PA-protein precipitation. Buffer containing 5% of 116 triethanol amine (TEA, v/v) and 5% of urea (w/v) was used for dissolving PA-protein pellet after 117 7 centrifugation and to support the colorimetric reaction with ferric chloride. Background and final 118 absorbances were measured at 510 nm and sample absorbance was determined by subtracting the 119 background absorbance from the final reading. The results were expressed as grams of (+)-catechin 120 equivalents (CE) per kg of leaf blades/veins FW.