AJEV Papers in Press. Published online August 17, 2016.

American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123

AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

1 Research Article
2 Biochemical Markers for Oenological Potentiality in a
3 Grapevine Aromatic Variety under Different Soil Types
4 Diana De Santis,1,3* Maria Teresa Frangipane,1,3 Elena Brunori,1,3
5 Pasquale Cirigliano,2 and Rita Biasi1
1
6 Department for Innovation in Biological, Agro-Food and Forestry Systems, University of Tuscia, via San
7 Camillo De Lellis snc, 01100 Viterbo, Italy; 2 Consiglio per la Ricerca in Agricoltura e l'analisi
8 dell'economia agraria (CREA), Research Unit on Viticulture (CREA-VIC), viale S.ta Margherita, 80 -
9 52100 Arezzo, Italy; 3these authors contributed equally to this work.
10 *Corresponding author ([email protected]; tel: +39 0761 357371; fax: +39 3487034809)
11 Acknowledgments: This research was supported by the regional funding PRAL (Programma Ricerca
12 Agroambientale, agroalimentare e agroindustriale Lazio) Cod 2003/22- 2006.
13 Manuscript submitted Dec 2015, revised Mar 2016, Jul 2016, accepted Jul 2016
14 Copyright © 2016 by the American Society for Enology and Viticulture. All rights reserved.
15

16 Abstract: Pedo-climatic conditions affect grape and wine quality and in particular the relation

17 soil-grape quality is at the core of the terroir definition. The study focuses on an aromatic and

18 autochthonous grapevine cultivar grown in the north of the Latium region (Centre of Italy), i.e.

19 cv Aleatico, in a heterogeneous environment (PDO ‘Aleatico di Gradoli’), where five sub-areas

20 have been selected as representative of the environmental variability. The ripening grape

21 parameters, volatile and phenolic compounds in wines derived from the grapes of these growing

22 areas, were analysed to assess the relationship among soil traits, biochemical grape and wine

23 parameters. Pedo-climate analysis was carried out following official protocols for soil texture

24 determination and bioclimatic Thermal Index of Winkler computation. The volatile wines

25 composition and phenolic compound were determinate using the SPME technique gas-

26 chromatographic method and standard method, respectively. Data were evaluated through

27 descriptive statistical methods (ANOVA and Pearson’s Coefficient) and multivariate statistical

1
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

28 analysis (PCA and HCA). The results proved that there is a ‘soil effect’ on the grapes and wines’

29 biochemical composition. The effect of soil on grape ripening parameters was found to be highly

30 significant with regard to total soluble solids and phenols concentration; both exhibited a high

31 correlation to soil’s sand content (%). In particular, soils with a -sandy-loam texture, moderate

32 skeleton content, offer the best wine performance in terms of aroma and phenolic content. The

33 study highlight the importance of the microzonation even in small wine-grape growing areas for

34 better diversified, and therefore more competitive, wine productions. The study enhances the

35 knowledge about the relationship between soil and grapevine aromatic varieties. Data points at

36 identifying biochemical parameters as markers of oenological potential according to

37 geographical origins.

38 Key words: cv Aleatico, berry and wine quality, bioclimatic indexes, HS-GC, polyphenols,

39 SPME

40 Introduction

41 Current market trends tend to wines with better sensorial features, and high quality

42 standards, like high quality grapes allow to produce. The number of academic papers related to

43 the study of the oenological potential of aromatic grape varieties in the last decade also

44 increased, proving the growing interest of winemakers and grape growers in the aromatic

45 grapevine varieties such as ‘Muscat’ aroma cvs (Crespan and Milani 2001).

46 The aromatic grapevine varieties are characterized by high monoterpenol’s

47 concentrations; in fact these compounds are identified as the main responsible of this aroma.

48 The more abundant monoterpenols are linalool and geraniol, commonly associated with floral,

49 rose-like flavours (Riberau-Gayon et al. 1998). Numerous studies on the volatile components of

2
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

50 the Vitis vinifera have helped the understanding of the complex metabolic pathway for the

51 formation of aromas. Nonetheless, the relationship between individual chemical components and

52 the defining aroma of wines is far from being fully understood, due to the complexity of wine

53 matrices, to the low concentration of some of the volatile compound partially responsible for the

54 wine aroma and flavour, and to complex physical and chemical processes that occur during the

55 grape ripening and wine fermentation.

56 Wine’s aroma profile derives from a balance of grape’s chemical compounds, in

57 particular volatile fraction and phenolic compounds - precursors of volatile phenols and wine

58 colour (Sagratini et al. 2012). The wine aroma is conditioned by intrinsic factors such as the

59 grape genetic specificity and other extrinsic factors such as the environmental conditions

60 (climate, soil, location) (Perestrelo et al. 2014 ), vine growing practices (Sala et al. 2004), and

61 winemaking (like fermentation conditions and aging) (Swiegers et al. 2005, Comuzzo et al.

62 2006, Piñeiro et al. 2006). In this sense, some studies have tried to establish links among

63 viticultural parameters and grape or/and wine composition. Several researches found that soil

64 impacts the overall quality of the grape and thus resultant sensory wine quality (Sabon et al.

65 2002, Van Leeuwen et al. 2004, Wang et al. 2015, Zerihun et al. 2015). This suggests that

66 although macroclimate, mesoclimate, and fruit microclimate all have major impacts on wine

67 quality and varietal peculiarity, soil may also have an independent effect. Micro and macro

68 elements of soil could be used to fingerprint wines from different grape-growing areas.

69 Furthermore, intrinsic geochemical parameters as the strontium isotope analysis would allow to

70 link each wine to its production zone (Bollati et al. 2015). Therefore, if in general the analysis of

71 volatile compounds is used to characterize different cultivars (Arvanitoyannis et al. 1999), these

3
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

72 biochemical compounds could also represent an important biochemical marker for the

73 differentiation of wines based on geographical origin (Perestrelo et al. 2014), particularly when

74 testing wine authenticity, pivotal issue after the introduction of European regulations for the

75 protection of the denomination of origin products. Therefore, fingerprinting techniques, based on

76 chemical composition can be used to characterize or classify wines according to origin, quality,

77 variety, type, etc. Furthermore, an interpretation of wine differences based on multivariate data

78 analysis such as the Principal Component Analysis (PCA) complemented with information

79 obtained by the Hierarchical Cluster Analysis (HCA) has been efficiently employed in wine

80 differentiation and classification to geographical origin (Perestrelo et al. 2014).

81 In the Latium region (Centre of Italy) the red grape variety ‘Aleatico’ is an example of

82 aromatic autochthonous grapevine germplasm under moderate erosion risk (according to the

83 Regional Developmental Programme (PSR) Latium 2007/2013 Reg (CE) N. 1698/05). It has

84 been objected to several viticultural and oenological studies in order to enhance the oenological

85 potential and to preserve an important biological and economic resource (Bellincontro et al.

86 2006, Biasi et al. 2007). Probably the Greeks were the first to introduce this variety in ancient

87 times, as it can be considered a mutation of the ‘Moscato’ grape. The corresponding Greek wine,

88 from Crete, is ‘non aromatic’. The name may derive from "July" (iouliatico in Greek), the month

89 in which veraison occurs. In Italy the Protected Denomination of Origin (PDO) ‘Aleatico di

90 Gradoli’ (D.P.R. 1972 modified with D.M. 2011) is a mono-varietal wine produced in a small

91 and highly heterogeneous area – mostly in the Gradoli municipality in the northern part of

92 Latium (Italy). The alcoholic content of this wine ranges between 12 and 17.5%; the wine is

4
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

93 commercially available in different typologies: traditional, liqueur wine, liqueur reserve and

94 passito wine.

95 The main purpose of this study was to identify and quantify the biochemical traits of

96 grape at ripening time and the volatile fraction and polyphenolic content of an aromatic

97 autochthonous cultivar of Latium region (Italy), i.e. the cv ‘Aleatico’ and to define through

98 Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA) the effect of

99 environmental factors, such as the soil, on the wine produced from grapes belonging to different

100 growing areas.

101 Materials and Methods

102 Plant material

103 The investigation focuses on the aromatic red grapevine variety ‘Aleatico’, an autochthonous

104 genotype of increasing importance due to its suitability to produce a wide range of oenological

105 products, including sweet wines from dehydrated grapes (Bellincontro et al. 2006). This cultivar

106 has been characterized through conventional phenotyping and genotyping procedures (Biasi et al.

107 2007), however unable to distinguish the high cultivar polymorphism. All the grape samples

108 were from the main typology, i.e. a medium-large winkled bunch (170 ± 66 g) with late ripening.

109 Study area

110 The study area is located in central Italy (northern of Rome) (Figure 1A) and is part of the large

111 territory where different PDOs (Protected Denomination of Origin) overlap (i.e. ‘Aleatico di

112 Gradoli’ (Figure 1B), ‘Colli Etruschi Viterbesi’, ‘EST!EST!!EST!!!). In particular, the

113 investigation was carried out within the historical and classic wine-grape growing area for the cv

114 ‘Aleatico’ that includes many localities, namely S. Antonio (SA), La Fratta (LF), Macchia del

5
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

115 Prete (MP), S. Magno (SM), Montemaggiore (MM) (Figure 1B and C). In order to determine

116 the physical characters of the studied DOP area, the following cartographic supports were used:

117 i) the regional geological map 1:50000 scale, ii) the regional topographic map (CTR) 1:10000

118 scale, iii) the elevation map (DEM) at the resolution of 40 m and iv) the aspect map, resulting

119 from a previous zonation study (Biasi et al. 2007). Aleatico’s vineyards are mostly present

120 between 300 and 450 m a.s.l., on the south-east exposed slope in the 1500 ha wide Landscape

121 System of the Bolsena lake basin. In particular, vineyards for grape sampling were located in the

122 Land Unit ‘Medium and low warmly exposed sides’, following the common zonation

123 methodology for hierarchically organized spaces.

124 Climate and bioclimatic indexes

125 The mean precipitations and the average monthly temperatures were calculated from data

126 recorded by a meteorological station (Integrated Service Agrometeorological Lazio Region -

127 SIARL) from 2004 to 2015 based in the study Land Unit (Bolsena, Viterbo, Italy) for a general

128 climatic classification (Fig.2). Furthermore, the Thermal Index of Winkler (IW) (Winkler 1962),

129 also called growing degree-days (GDD) has been calculated to evaluate suitability of grapevine

130 cultivation and classification of grape-wine growing areas (Fregoni and Pezzuto 2000). IW was

131 derived by the formula (1) as the sum of the average daily useful temperatures (Tmean-10°C)

132 during the grapevine growing season (from April 1st to October 31).

133 31/10
134 WI = ∑ (T mean - 10)
135 01/04 (1)
136

137

6
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

138 Pedological investigation and vineyard soil characterization

139 One model vineyard was selected for each locality in the PDO study area (Land Unit ‘Medium

140 and low warmly exposed sides’). The soil pedological profile and typology were classified. The

141 soil texture was determined following the standard methodology for deepness, mean texture, i.e.

142 sand-silt-clay ratio and skeleton percentage measurement (Soil Survey Staff 1998). Soil

143 classification was done according the international soil classification system (USDA-NRCS

144 1999).

145 Berry volatile compounds:

146 The determination of volatile compounds of grapes has been carried out on Grape berry juice (5

147 g) mixed with 5 mL of saturated CaCl2 (1:1 w/v) or wine alone (10 mL) were homogenized with

148 200 μL of standard solution (1-penten-3-one). The homogenate was collected in a 25 mL glass

149 miniflask (Supelco, Sigma-Aldrich Co, St Louis, MO, USA) and sealed with a Teflon silicone

150 septum. The sample, grapes or wine, was exposed to a solid phase micro-extraction fiber,

151 respectively, for 30 and 15 min, in a Thermo Haake DL30-V15B water bath (ENCO Spinea, Ve,

152 Italy) maintained at a temperature of 20±2°C. The SPME fiber (PDMS 100 μm) was then

153 inserted into the headspace. During the sampling time (30 min), the sample was stirred at

154 constant speed. After completion of sampling, The fiber, Polydimethylsiloxane (PDMS 100 μm)

155 (Sigma-Aldrich Co), was conditioned in the GC injection port at 250 °C for 2 h prior to use. For

156 as regards the chromatographic conditions they were the same as used for wines.

157 These method can be useful for monitoring grape maturation and to determine grape composition

158 before wine- making. The aromatic profile of Aleatico’s berry is chiefly characterized by

159 monoterpens, that represent the 85% of total berry aroma. We focus on five of the most relevant

7
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

160 monoterpenes for Aleatico cultivar: linalool, terpineol, citronellol, nerol and geraniol

161 (Mencarelli and Tonutti 2013).

162 Samples for microvinifications

163 Grapes: The experiment was conducted using grapes from the five sub-areas: La Fratta

164 (LF), S. Antonio (SA), Macchia del Prete (MP), S.Magno (SM), Montemaggiore (MM).

165 Aleatico’s grapes for all tested areas were harvested at technological maturity by

166 monitoring the sugar berry accumulation increment up to it remains constant. Then samples are

167 separately microvinificated. Three vines for sub-area are selected and three replicates of 33

168 berries each one are randomly collected on three bunches for vines. These replicates have been

169 used to analyze the chemical traits (sugars, pH and total acidity) of berries before the

170 microvinification.

171 Wine: A total of 85 kg of grapes for each sub-area was destemmed, crushed, divided into

172 five batches one for each sub-area and 10g·hL-1 of K2S2O5 were added to mash prior the

173 fermentation. All batches were inoculated with yeast (Saccharomyces cerevisiae, 10g dry wt·hL-
1
174 ) and punched down three time a day during active fermentation for 5 days, the duration of

175 maceration or contact between solids and liquid was 15 days at 25°C. At the end of maceration

176 the free run of each batches was drained and the pomace was pressed in a tank membrane press

177 in order to obtain the press run At last the free and press run for each batches have been

178 combined and the wine was stored in 100L stainess stell tanks at room temperature for five

179 mounths. After 5 months wines were bottled in 750 mL bottles sealed with corks.

8
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

180 Three replicates of wine samples obtained for each sub-area (five batches) has been

181 analyzed to assess chemical traits as titrable acidity, pH, alcohol content, polyphenols, total

182 anthocyanins and aromatic compounds.

183 Chemical parameters: The titrable acidity of wines were changed between 5.8 and 6.2

184 g·L-1. The pH values were ranged between 3.5 and 3.6 and alcohol content between 13.2 and

185 14.4 % vol. The analysis were determined according to the Regulation Official Methods by EU.

186 Spectrophotometric methods: quantitative estimate of the presence of phenolic

187 compounds was easily carried out involving spectrophotometric detection. A spectrophotometer

188 (Perkin Elmer, mod. Lambda 2) was employed. The methods are described in detail by Di

189 Stefano and Cravero in 1991 and were easily employed.

190 Total anthocyanins content of skins of 10 berries were extracted by a 25 ml

191 ethanol/water/HCL (70:30:1 v/v/v) solution for one day in the dark. The solution was diluted in

192 accordance with methods reported by Di Stefano at al. (1991) to register the absorbance

193 spectrum bettween 230 and 700 nm for estimation of total anthocyanin and flavonoid contents.

194 The Folin–Ciocalteu method proposed by Di Stefano et al. (1989) was used for the determination

195 of the total phenolics in wines. In brief, an aliquot (1 mL) of the appropriate diluted extracts was

196 added to a 10 mL volumetric flask, containing 5 mL of distilled water. Then, 0.5 mL of Folin-

197 Ciocalteu reagent was added and the contents mixed. After 3 min, 1.5 mL Na2CO3 solution of

198 concentration 5 g·L-1 was added and made up to a total volume of 10 mL distilled water. After

199 keeping the samples at 50 °C (water bath) for 16 min in sealed flasks and subsequent cooling,

200 their absorbances were read at 765 nm against distilled water as the blank. A calibration curve

201 was constructed using gallic acid standard solutions (0–100 mg·L-1). The concentration of total

9
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

202 phenolics is expressed as the gallic acid equivalent (GAE) per 1 g of fresh sample. All samples

203 were prepared in triplicate;

204 The determination of the total anthocyanins in wines was realized by the method

205 proposed by Di Stefano et al. (1989). The samples were diluted with a solution consisting of

206 70/30/1(v/v/v)ethanol/water/HCl (concentrated) and the absorbance was measured at 540 nm.

207 Due to the lack of a malvidin-3-glucoside standard, the total anthocyanic contents are expressed

208 as malvidin-3-glucoside equivalents and calculated using the following equation:

209 TA540 nm(mg·L-1) = A540 nm16.7d

210 where A540 nm is the absorbance at 540 nm and d is the dilution.

211 Volatile compounds characterization: the analysis of volatile compounds is normally

212 carried out by gas chromatography (GC). Yet even today, the extraction and concentration of

213 flavour components constitute a problem that has still not been satisfactorily resolved. In recent

214 years, different extraction methods based on solid-phase microextraction (SPME) have been

215 applied to analyse certain types of volatile compounds in wines (Vas et al. 1998, Haggerty et al.

216 2015) and in particular Vas et al. (1998) reported the use of SPME for fast screening of different

217 wine types, morover SPME is a solvent-free method presenting major advantages, such as small

218 sample volume and higher sensitivity and simplicity.

219 The analysis of volatile compounds was carried out applying the SPME technique

220 gaschromatographic method and the selected SPME parameters are shown below. Five mL of

221 wine was pipetted and placed into a 25ml glass mini flask (Supelco, Sigma-Aldrich Co, St Louis,

222 MO, USA). Each sample was spiked with 200 μl of a standard solution of 1-penten-3-one (CAS

223 Registry Number: 1629-58-9) (5gL-1 in Milli-Q water). A small magnetic stirring bar was also

10
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

224 added. The vial was tightly capped with a PTFE-faced silicone septum and placed in a

225 thermostatted block on a stirrer. The sample was equilibrated for 15 min at sampling temperature

226 in a Thermo Haake DL30-V15B water bath (ENCO Spinea, Ve, Italy) maintained at a

227 temperature of 20 ± 2 °C. The SPME fiber (PDMS 100 μm) was then inserted into the

228 headspace. During the sampling time (30 min), the sample was stirred at constant speed. After

229 completion of sampling, The fiber, Polydimethylsiloxane (PDMS 100 μm) (Sigma-Aldrich Co),

230 was conditioned in the GC injection port at 250 °C for 2 h prior to use..

231 Chromatography: after the selected extraction time the SPME fiber was insert into the

232 GC injection port and thermally desorbed at 230 °C for 7min. The splitless injector was mounted

233 on a model 5300 Mega Series gas chromatograph (Carlo Erba Instruments, Milan, Italy). It was

234 equipped with a fused silica capillary column impregnated with a polar phase of Carbowax 20M

235 (Alltech Assoc, Inc, Deerfield, IL,USA), 60m long × 0.25mm id and 0.25 μm film thickness.

236 Helium was used as carrier gas (27cm·s−1). The temperature was maintained at 40 ◦C for 7min,

237 then programmed to reach 230 °C at a rate of 3 °C min−1, with a final isotherm of 30 min. A high

238 sensitivity flame ionization detector (FID) at 260 °C was used. The signal was recorded and

239 integrated by a Mega Series integrator. Compounds identification was achieved by using a

240 Shimadzu 17A GC/MS and a Shimadzu QP 5050A MS and matching against the NIST 107 and

241 NIST 21 libraries. Volatile and semivolatile constituents were identified using mass spectral

242 matching against library standards RI and RI* (values for a Carbowax 20M column after running

243 a carbon alkane standard, also known as Kovats index KI) (Table 1).

244 Statistical analysis

11
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

245 All determinations were carried out in three replicates and all analytical results were subjected to

246 a statistical study performed at the p<0.05 significance level and carried out by DSAASTAT ver.

247 1.1 (Onofri 2007), using one-way ANOVA and Fisher’s LSD test.

248 Correlation analysis (Pearson’s coefficient) were used to identify the aspect of soil (physical)

249 that are associated with berry composition at harvest (sugar, acidity, anthocyanins, polyphenols).

250 A Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA) has been

251 carried out using all the major chemical and aromatic compounds of berries for the five sub-areas

252 at harvest time, in order to detect how the soil type could influence the berry quality.

253 Finally, in order to reveal the relationship among biochemical parameters of the Gradoli

254 Aleatico’s wines and soil traits of the grapevine growing areas, and to identify the wine

255 components that better concur to characterize wines according to geographical origin, the

256 composition data matrix of five samples (27 variables x 5 wine samples = 135 data) was

257 analysed using the PCA and HCA, based on Euclidian distances. The analyses were performed

258 using R software package

259 Results

260 The physical environment of the vineyards

261 The climate of the studied grape growing area, i.e. the PDO ‘Aleatico di Gradoli’ was

262 identifyied as Mediterranean whit mean annual precipitations, over the last 10 years, of 806 ±

263 230 mm of rain, mainly concentrated in Autumn (November –December). The drought events

264 are concentrated in summer (June – August) where the mean distribution of rainfall was of two-

265 four days per month.

12
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

266 The Winkler index reached in the last 6 years the mean value of 1827.5 ± 130.8 GG

267 (Table 2).

268 The whole area was characterized by high eterogenity of steep downslope directions

269 (Figure 1C) although the five areas tested were located on the south-east slopes.

270 The soil of the study area has evolved from the preminent lithology, i.e. ignimbrite, a

271 vulcanic material of different texture (ashes and lapillus) and poorly consistent. In the Land Unit

272 of vineyards (Land Unit ‘Medium and low warmly exposed sides’) were found soils at different

273 evolution stages: from the poor and moderately evolved soils (Entisols and Inceptisols), the most

274 represented, to the highly evolved types (Alfisols) with higher stability (USDA-NRCS 1999).

275 Based on pedological investigations and soil texture analysis three types of soils proved

276 to characterize the tested vineyards, i.e.: i) a sandy-loam or sandy soil - type ‘lapillo’ (local

277 denomination) , ii) a sandy skeletal and shallow soil profile – type ‘pianca’ (local name) and iii)

278 clay loam soil – type ‘volpaio’ (local name).

279 The characteristics of each soil type and their classification are resumed in Table 3. Three

280 vineyards, i.e. those located in LF, SM and SA sites were characterized by the presence of the

281 ‘lapillo’ soil. The vineyard located in MM had the ‘pianca’ soil and the vineyard located in MP

282 the ‘volpaio’ one.

283 Berry biochemical traits

284 The berry biochemical characters such as sugar content, acidity, as well as the phenolic fraction

285 concentration (total anthocyanins and polyphenols) are shown in Table 4A. At harvest time the

286 grapes from SM, MP, LF sub-areas exhibited a higher sugar concentrations than those from SA

287 and MM. Significant differences for sugar concentration seems to be related to the soil typology

13
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

288 and to depth (deep versus shallow soil). Berry grown in SA, SM and LF sub-areas, characterized

289 by ‘lapillo’ soil but different for skeletan and sandy content, shown significant difference for

290 total acidity, and didn’t exibith difference in antocyanins concentration.

291 The highest total anthocyanins and total poliphenols concentration was associated to

292 berries of MP grape growing sub-areas. In particular total poliphenols concentrations exhibithed

293 significant differences according to the soil type. The statistical correlation (pearson’s

294 coefficient) (Figure 3A and 3B) shown as sugars and poliphenols exibith polynomial trends

295 related to sand soil content. In particular soluble solid and phenols were highter in the clay –

296 loam soil (‘volpaio’ type – MP grape growing sub-area), while the lowest values were obtainted

297 in shallow soil whit high sand content (‘pianca’ type - MM grape growing sub-area).

298 Berry volatile compounds

299 The berry aromatic profile is showed in Table 4B, all the volatile compound shown significant

300 difference according to soil type. Berry chemical and aromatic parameters are used to carry out

301 the statistical PCA and HCA, has reported in Figure 4. The I PC ( Principal Component) and II

302 PC explain about 83% of the total variation, and the first component are related to linalool,

303 terpineol, citronellol, anthocyanins and polyphenols, while the second to nerol, geraniol, sugar

304 and acidity of berries at harvest time. All the five sub-areas are scattered on Cartesian plane

305 drew by I and II PC, in particular the HCA showed as the berry of this areas could be gather

306 together into two main group. Berries from sub-areas ‘Macchia del Prete’ (MP), the only

307 vineyards characterized by a clay loam soil, are more different from the other ones (sandy-soil).

308 Wine volatile compounds

14
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

309 In order to distinguish a possible classification on the basis of volatile profile, the wines

310 were assigned into three groups, corresponding to different sub-areas with homogeneous soil

311 traits (Figure 5): SA, SM and LF – ‘lapillo’ soil; MM – ‘pianca’ soil and MP – ‘volpaio’ soil.

312 Acetaldehyde (Figure 5A) is an important secondary aroma, it reached concentration of

313 0.15% and showing significance different among Aleatico di Gradoli wines related to the five

314 growing sub-areas. The composition of aldehyde in the studied Aleatico di Gradoli wine

315 presented uneven values compared to other components of the headspace examined. Despite the

316 fact that total aldehydes amounted to very low values, the total volatile compounds of samples

317 were highest for the decanal and the nonanal in MM, SA and SM, Figure 5A.

318 The most important flavor and aroma compounds formed from amino acids are higher

319 alcohols (Figure 5C and 5D) and their associated esters (Figure 5B). The esters compound in

320 the Aleatico wine showed the predominance of ethyl caprylate and ethyl decanoate and minor

321 amounts of ethyl acetate (Figure 5B).

322 Isoamyl alcohol gave the greater contribution to flavor for all Aleatico di Gradoli wines

323 (Figure 5C), in particular, significant differences among the growing sub-areas resulted from the

324 statistic. The ethyl esters for the five wines were the main chemical group followed by isoamyl

325 alcohol. Also for the ethyl esters contribution to total volatile wine profiles it has shown a

326 significance differences related sub-areas origin. The monoterpenes compounds (Figure 5E),

327 such as linalool, citronellol, and limonene, were the compounds most important for the Aleatico

328 wines discrimination and all data shown significance differences according to sub-areas wine

329 origins. In particular, citronellol content was from two to four times higher than the respectively

330 geraniol content for each sub-areas. The linalool and limonene contents showed similar evolution

15
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

331 related on soil type, the higher values for wines originate from Lapillo soil (sandy and deep soil)

332 than from Pianca and Volpaio soils.

333 Total phenols and total anthocyanins

334 Total phenols and anthocyanins are shown in Figure 5F. All data exhibited significant

335 differences among them. In particular, it was noted that wines produced with grapes from MM,

336 SA and SM presented a concentration of total anthocyanins and polyphenols concentrations

337 significantly higher than in the other samples. On the other hand MP amounted always to the

338 lower concentration of phenols and anthocyanins.

339 Statistical analysis - PCA

340 In Table 5 the set of the eigenvalues of PCA is reported along with the amount of inertia

341 explained by each corresponding axis and the cumulate inertia. We selected two principal axes of

342 interest, that explain about 76% of the total variation (measured by the inertia), yet 60% is the

343 minimum cumulate quality of representation of all the parameters and is sufficient for our

344 exploratory purposes (Camiz et al. 2008). In Table 6 the results of PCA are shown. For each

345 biochemical parameter and each selected component (axes) (I and II) are reported: (i) its

346 contribution to the axis and (ii) its correlation with the principal components. The data helps to

347 interpret the graphical representation, limited here to the principal plane delimited by axes 1 and

348 2, where all the items are represented.

349 The contribution of the chosen biochemical markers to the first principal axis (also shown

350 on the principal plane spanned by axes 1 and 2 in Figure 6) was mostly due most to the

351 monoterpene and ester fractions, while the contribution to the second principal axis was due to

352 mainly alcohols and aldehydes. The position of the wines derived from the grapes of the five

16
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

353 tested vineyards, is separated into 3 groups: i) LF, SM and SA ii) MM and iii) MP. In particular

354 LF and SM are opposite to MM on the first axis, while MP is opposite to SA.

355 The samples of the vineyards LF, SM and SA are characterized by the monoterpenes, MP

356 samples are characterized mainly by the alcohol and aldehyde components, while the MM

357 samples by the phenolic components.

358 Discussion

359 Based on the Thermal Index of Winkler (IW) values (Table 2), the PDO area proved to be

360 higly vocated for the production of quality wine and in particular for the cultivation of red berry

361 variety, following the classification reported in Fregoni and Pezzuto (2000).

362 Grape volatile compounds are responsible of aroma of wine and depending on the grape

363 variety, cultural practices, and climatic or biological factors. The concentration of aromatic

364 substances changes according to an optimal sugar/acid ratio (Ribéreau-Gayon et al. 1998). The

365 berry quality parameters underlighted the great potential of this cultivar as aromatic grape for

366 liqueur and passito wines (De Santis et al. 1999, Bellincontro et al. 2006).

367 Nowadays it’s known that the wine volatile compound is derived from multiple sources

368 and processes but one of the main direct contribution has been given to grape-derived aroma

369 compounds, including monoterpenes, norisoprenoids, aliphatics, phenylpropanoids,

370 methoxypyrazines, and volatile sulfur compounds (Robinson et al. 2014, Gonzalez-Barreiro et al.

371 2015).

372 In particular, even it is difficult to assess the impact of a single parameter of terroir - such

373 as the soil – on grape and wine quality, several researches have highlighted that grape and wine

17
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

374 composition are significantly affected by soil type, which influences the taste of the final product

375 (van Leeuwen et al. 2004, Zerihun et al. 2014, Wang et al. 2015).

376 Findings underlined, according to van Leeuwen et al. 2004, that different soil types

377 resulted in significant differences in total sugars and phenols concentration of berries.

378 Correlation analysis among berry ripening parameters (sugar and poliphenols content) and sand

379 soil content shown as this soil physical trait impact the overall quality of the grape and thus the

380 resultant sensory quality of the wines, according to Gomez-Miguez et al. 2007.

381 Aleatico grapes and wines, produced from vines growing on several soil types but in the

382 similar climatic condition, as the five sub-areas, could be discriminated from each other.

383 Multivariate data analysis is confirmed to be a discriminating method of the origin of grape and

384 wines. PCA and HCA carried out on berry at harvest time help to analyse the soil effect on berry

385 quality.

386 In particular, clay loam soils with low skeletan content - MP sub-areas - are beneficial for

387 improving grape quality, inasmuch it seams to be optimize (best values for many berry quality

388 parameters – Table 4) in this sub-areas as resulted also from cluster analysis (Figure 4). It might

389 be influenced by the soil texture: deep and shallows sandy soil, characterized by a moderate-high

390 amount of skeleton (LF, SA,SM and MM) and with lower clay seem to promote higher vigour

391 (Fraga et al. 2014), that in these grape-growing area could be undermine the Aleatico berry

392 quality.

393 All ripening parameters are pivotal factors for wine composition, but total soluble solids

394 (sugars) and poliphenols being the most important and discriminating substances.

18
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

395 Since it is difficult to ‘unhook’ a soil effect on grape composition from possible

396 contributions of many other factors (climate and cultivars), or established if it has a indirect

397 effect of vine attribute (Zerihun et al. 2014), this does not preclude an effect of soil factors on

398 technological and phenolic components of berries and wines. Moreover the microbial

399 communities on grapes may be affected by a large number of factors such as the geographic

400 location and its traits (soil and climate) (Barata et al. 2012), in addition to genetic traits of the

401 berry itself. The study of wine flavors is a critical issue because it is the result of a complex

402 mixture of volatile compounds. Among the 800 volatile molecules identified in wine, a relatively

403 limited number of them called varietal (or primary) aromas, play a crucial role in wine flavor and

404 typicality.

405 Findings underlined a consistent presence of esters and a low participation of aldehydes,

406 debt levels of higher alcohols and monoterpenes. These flavor profiles are consistent with the

407 average composition of red wines (Bonino et al. 2003). The values of the individual constituents,

408 expressed in percentage terms, highlights the complex composition of the volatile fraction for the

409 sub-areas samples observed. In particular for the monoterpene alcohols (Figure 4 E), that

410 represents the most common varietal aromas in the ‘Muscat’ varieties ( Bonino et al. 2003).

411 Acetaldehyde (Figure 4A) amounted to good concentration, contributing to diversify the

412 aroma complexity of wines and producing stable pigments in red wine, enhancing the wines

413 quality (Bakker and Clarke 2011). Acetaldehyde is normally present in large quantities in wine

414 as a metabolite of fermentation, because many of them are generated by fermentation, maturation

415 and aging in wood. As the fermentation process ends, acetaldehyde produced is introduced into

19
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

416 the yeast cell and reduced to ethanol, so that the content falls to lower levels at the end of

417 vinification.

418 The esters profile (Figure 4B) in the Aleatico wines show high level of esters able to

419 develop fruity notes to wines, such as ethyl octanoate, erthyl decanoate and dodecanoate gave a

420 positive contribution to wines quality. Most of them, in fact, have mature fruit flavor nuances

421 that are responsible for the ‘fruity’ and ‘floral’ sensory properties of wine (Sagratini et al.

422 2012). While acetate esters - isoamylacetate and hexylacetate- are responsible for tropical fruit

423 and banana-like aromas.

424 Esters are the product of the condensation between carboxyl groups of organic acids and

425 hydroxyl groups of alcohols or phenols. These, in small quantities and some with a rather high

426 threshold of perception, make a slight contribution to the aromatic composition of wine.

427 The production of esters during fermentation is influenced by many factors: the grape

428 ripeness, the esterase activity of strains of yeast and fermentation temperatures are the elements

429 of greatest influence. Low temperatures may encourage the formation of fruity "ester" as

430 isoamyl, isobutyl and ehyl acetate, while high temperatures favor the formation of esters with

431 higher molecular weight as ethyl octanoate, ethyl decanoate and phenylacetate. High

432 temperatures can also depress the formation of esters favoring hydrolysis. Low levels of SO2 and

433 clarification of the wort can both promote the synthesis and accumulation of foreign as well as

434 intracellular fermentation (carbonic maceration) and the lack of oxygen during fermentation.

435 Although esters in wine are mainly produced by yeast metabolism their production can be

436 influenced also by the grape variety (Swiegers et al. 2005).

20
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

437 The higher alcohols (Figure 4C and D) are generally found to be responsible for aroma

438 due to the fact that they are found in quantities above perception threshold, rather it is the

439 minimal concentration in which the component can be detected. At low levels (< 300 mg·L-1),

440 straight chain higher alcohols generally add complexity to the bouquet of a wine (Bakker and

441 Clarke 2011), such as for the Aleatico di Gradoli wines, where the greater contribution is given

442 by Isoamyl alcohol (Figure 4C), that generally accounts for more than 50% of all alcohol

443 fractions - and it is found nascent in the grape, but the majority is formed by yeast during

444 alcoholic fermentation and help to determine identity and complexity of wine (Bakker and

445 Clarke 2011).

446 The formation of higher alcohols during fermentation is markedly influenced by the

447 wine-making practices: the presence of oxygen, high temperature fermentation and materials

448 suspended in the fermenting juice promoting the proliferation of alcohols through the reduction

449 of aldehydes grape, reductive de-nitrification synthesis of amino acids or sugars.

450 Aleatico wines with a similar higher alcohol content produced in the five sub-areas, have

451 a distribution that differs only slightly from higher values in heptanol in the samples from the

452 SM and MP areas and an increase in hexanol in the LF area (Figure 4C and 4D).

453 The mono-terpenes compounds were the most important compounds for the

454 differentiation according to sub-areas classification. The aromatic grape varieties, such as

455 Muscat, Riesling and Gewürztraminer, contain large amounts of the monoterpenes geraniol and

456 nerol.

457 Geraniol has aromas described as rose-like and linalool aromas described as rose,

458 whereas linalool oxides are described as camphorous and nerol oxides as vegetative.In general,

21
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

459 more bound glycosides are found than the free terpenoids, and the ratios of bound to free

460 terpenoids can also vary amongst different grape cultivars. Muscat of Alexandria grapes, for

461 example, have a ratio of 5:1, whereas some non-Muscat varieties have a ratio of 1:1. Varietal

462 wine aroma from muscat-related grapes, for example, is mainly due to the presence of various

463 isoprenoid monoterpenes in the grapes, with the most important being linalool, geraniol, nerol,

464 and citronellol

465 The de novo biosynthesis of flavor and aroma compounds is probably the most important,

466 because, in general, fermentation- derived volatiles make up the largest percentage of total aroma

467 composition of wine in terms of numbers. All of these compounds have been proved to be odor-

468 active compounds in musts and wines (García-Muños et al. 2011).

469 Among the tested sub-areas, SM, SA and LF optimized the grape and wine whole. An

470 explanation might be the fact that these sub-areas, which are managed with the same agronomic

471 techniques, have the same soil type –‘lapillo’. As confirmed by other authors (Morlat and Bodin

472 2006), the total content of anthocyanins is strongly related to environmental factors, as also

473 influenced by vineyard management. This might be due to the synthesis of the enzymes involved

474 in the various stages of anthocyanin’s synthesis closely linked to environmental and

475 physiological factors. This concept implies that there is a strong relationship between the

476 composition of the grape, the characteristics of the wine and the territory of production.

477 The PCA distribution of the biochemical characters of the wine coming from different

478 physical environments for soil typology (Figure 5) shows a ‘soil effect’ from sandy-loam soil

479 type (‘lapillo’) (LF, SM and SA), to sandy skeletal and shallows soil (MM) (‘pianca’), to a clay-

22
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

480 loam soil (MP) type (‘volpaio’). The soil texture seems to be therefore strictly correlated to the

481 biochemical characters of the oenological products.

482 The phenolic components related to the grater skeletal and sandy content as in the

483 vineyards of MM and SA, SM, probably owing to the positive effect of a light water deficit

484 (stress) on berry phenolic composition, as found in previous work (Biasi et al. 2010). Soluble

485 phenolic compounds are crucial to determine some aspects of berry quality and influence the

486 volatile metabolites that are responsible for wine varietal aroma (Koundouras et al. 2006).

487 In particular when the Aleatico grapevine grows and undergoes water stress, like in

488 sandy soil, the highest values for phenolics and mono-terpenes components are reached and they

489 concur to improve the quality and variety of traits of the grape and wine production.

490 On a final note, the aromatic complexity of Aleatico’s wine is a result of the presence in

491 the headspace of different volatile chemicals.

492 Not all substances are perceived in the same way. The perception, in fact, depends on the

493 complexity of our olfactory system and is the result of the presence of a volatile substance in

494 space, the "smell" of a food, and our ability to appreciate.

495 Several studies proved that the terpenoid compounds express the link between the variety

496 and the aroma of wine (Bakker and Clarck 2011). This class is complex and largely glycosidic

497 bound in an odorless form. The aroma of a wine varietal is the result of the combination of

498 various quantitative and qualitative free monoterpenes present in the must. However, the terpene

499 fraction undergoes slight transformations by enzymatic hydrolysis and acid, as well as

500 isomerization and cyclization. These changes are slow and influenced by several factors

23
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

501 including light, temperature and microbial flora. Nonetheless this does not change the impression

502 that this class of components gives to the various types of vines.

503 The wine’s aroma is the result of labile and complex processes that build on several

504 elements such as grape variety, the winemaking process, and the grape maturation and aging

505 stages.

506 The monoterpenes, the minor constituents of the plants, follow the common biosynthetic

507 way and generally free and bound fraction increases during the ripening of berries. During the

508 winemaking process the monoterpenes may undergo various transformations: acid or enzymatic

509 hydrolysis, isomerization and cyclization.

510 Aging and storage provide time for their slow transformation. It is, however, well known that the

511 hydrolysis of glycosides in aromatic monoterpenes is faster than the isomerization and

512 rearrangement. After a long time, related terpenes have slowly converted into aromatic terpenes,

513 which rearrange much more slowly into new compounds. Some of them may also be less

514 odorous molecules contrarily to those from which they originate. Other factors that could

515 influence this important fraction are the fermentation process, the storage temperature, the pH

516 value and composition of the wine. All these elements of variability are still considered by many

517 authors, influential on the sparsely-quantitative composition terpenes. The description of this

518 class of aromatic components is extremely interesting because it reveals the real differences

519 between aromatic grape wines, indisputably linked to the variety. Wanting to decipher this

520 component in the Aleatico aromatic wine, one recognizes a strong fruity note, tempered by a

521 combination of smells: roses and a hint of citrus and balsamic. This verbal description-olfactory

522 attempts to translate the terpene composition and could be used to describe the aroma variety of

24
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

523 other important wines such as Muscat or Gewurztraminer for which the qualifying note is rose,

524 geraniol and nerol or Malvasia, where linalool predominates with its pronounced floral notes,

525 tempered by a citrus smell. Without wishing to invest this kind of description as highly scientific,

526 we can still conclude that the varietal mark set by terpenes is a strong element of distinction and

527 characterization. This parameter, also for its stability in wine, is potentially capable of being

528 effectively used to discriminate varietal wines made in analytical purity.

529 Conclusions

530 This study allowed us to determine the effect of soil on Aleatico’s berry ripening

531 parameters and wines. Among the berry biochemical traits at harvest time sugar and phenols

532 concentration were discriminant variables able to have a direct influence on wine quality. Grape

533 quality potential (high grape sugar and phenols) is high on the soil that induce water deficit,

534 especially on clay-loam (MP) and sandy-loam (SM) soil where water deficit occur early in the

535 season but are moderate.

536 Aleatico wines, produced in the PDO sub-district maintain unique and comparable

537 biochemical traits, therefore preserving the identity of a territorial production. The exception of

538 the wine produced from grapes from the sub-area, La Fratta (LF), which deviates from many

539 values observed, in respect to the total sample values recorded, may be attributed to a slight

540 "immaturity" compared to the others. Nonetheless, the terpene compounds, i.e. linalool,

541 citronellol, and limonene, were the most important compounds for the Aleatico wines

542 discrimination and all data shown significant differences according to sub-area grape origins.

543 Taken together our results proved that berry biochemical traits, volatile and phenolic

544 compounds may be considered as biochemical markers for high quality wine from highly

25
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

545 vocated environments. Based on the strong relationship among berry ripening traits, volatile and

546 phenolic compounds and structural soil characters, they provide not only an indication for

547 oenological potential but also they could represent pivotal indexes suitable to understand the

548 ‘soil effect’ on grape and wine quality, and to identify the best terroir conditions (following

549 microzonation) also within small although highly heterogenic grape-wine producing areas, like

550 the ones we investigated. The knowledge of the environment physical characters, mainly the soil

551 in our study, which better affect the quality of productions could address farmer’s efforts to

552 exploit the variability of the viticultural environment and to assure a sustainable use of the

553 grapevine genetic resources.

554 Literature Cited

555 Arvanitoyannis, I. S., Katsota , M. N., Psarra, E. P, ., Soufleros, E . H., and Kallithraka, S. 1999
556 Application of quality control methods for assessing wine authenticity: use of multivariate analysis
557 (chemometrics). Trends Food Sci. Tech.10: 321-336.

558 Bakker, J. and Clark, R.J. 2011. Wine Flavour Chemistry (Second edition). Eds. Wiley-Blackwell A John
559 Wiley & Sons, Ltd; Pubblication..

560 Barata, A., Malfeito-Ferreira, M., and Loureiro, V. 2012. The microbial ecology of wine grape berries.
561 Int. J. Food Microbiol. 153: 243-259.

562 Bellincontro, A., Fardelli, A., De Santis, D., Botondi, R. and Mencarelli, F. 2006. Postharvest ethylene
563 and 1-MCP treatments both affect phenols, anthocyanins, and aromatic quality of Aleatico grapes and
564 wine. Aust. J. Grape Wine R. 12: 141–149.

565 Biasi R., Cirigliano P., Di Francesco G., Brunori E., Muleo R. 2007. Attuazione dei protocolli di
566 selezione clonale su biotipi di Aleatico individuate nell’area ‘tipica’ della Doc Aleatico di Gradoli e
567 caratterizzazione-selezione del ‘Grechetto Rosso’ e di altri vitigni autoctoni dell’Alto Lazio: risultati
568 della ricerca. Rivista di Viticoltura ed Enologia 4: 127-143.

569 Biasi, R., Brunori, E., Ceccariglia, I., Botti, F. 2010. Il sistema vigneto del lago di Bolsena :
570 caratterizzazione della produzione di Cannaiola di Marta. Proceedings VIII International Terroir
571 Congress, June 2010, Soave (VR), Italy, 2: 165-170.

26
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

572 Bollati, A., Molin, P., Cifelli, F., Petrangeli, A.B., Parotto, M. and Mattei M. 2015. An integrated
573 methodology of viticultural zoning to evaluate terrains suitable for viticulture: the test area of
574 Cesanese DOC (Latium, central Italy). J. Wine Res. 26:1-17.

575 Bonino, M., Schellino, R., Rizzi, C., Aigotti, R., Delfini, C., & Baiocchi, C. (2003). Aroma compounds of
576 an Italian wine (Ruché) by HS–SPME analysis coupled with GC–ITMS. Food Chemistry, 80(1), 125-
577 133.

578 Camiz, S., Altieri, A., Manes, F. 2008. Pollution bioindicators statistical analysis of a case study. Water
579 Air Soil Pollut.1: 111-139.

580 Comuzzo, P., Tat, L., Tonizzo, A., Battistutta, F. 2006. Yeast derivates (extracts and autolysates) in
581 winemaking: release of volatile compounds and effects on wine aroma volatility. Food Chem. 99:217–
582 230.

583 Crespan, M. and Milani, N. 2001. The Muscats: A molecular analysis of synonyms, homonyms and
584 genetic relationships within a large family of grapevine cultivars. Vitis 40: 23- 30.

585 De Santis, D., Frangipane M.T., Anelli G. 1999. Volatile phenols in ‘Aleatico di Gradoli Riserva’ liqueur
586 wine. Ital. J. Food Sci. 11: 371-378.

587 Di Stefano, R. and Cravero, M.C. 1991. Metodi per lo studio dei polifenoli dell’uva. Rivista di
588 Viticoltutra ed Enologia 44: 37-45.

589 Di Stefano, R. Cravero, M.C., Gentilini N. 1989. Metodi per lo studio dei polifenoli dei vini.
590 L’Enotecnico 5: 83-89.

591 Fraga, H., Malheiro, A. C., Moutinho-Pereira, J., Cardoso, R. M., Soares, P. M., Cancela, J. J. and Santos,
592 J. A. 2014. Integrated analysis of climate, soil, topography and vegetative growth in Iberian
593 viticultural regions. PloS one 9:

594 Fregoni, C. and Pezzuto, S. 2000. Principes et premières approches de l’indice biocli-matique de qualité
595 de Fregoni. Progrès Agricole et Viticole 117: 390-396.

596 Garcıa-Muñoz, G., Asproudi, A., Cabello, F., Borsa D. 2011. Aromatic characterization and enological
597 potential of 21 minor varieties (Vitis vinifera L.). Eur. Food Res. Technol. 233: 473–481.

598 Gómez-Míguez, M.J., Gómez-Míguez, M., Vicario, I.M., Heredia, F.J., 2007. Assessment of colour and
599 aroma in white wines vinifications: Effects of grape maturity and soil type. J. Food Eng. 79: 758-764.

600 González-Barreiro, C., Rial-Otero, R., Cancho-Grande, B., & Simal-Gándara, J. 2015. Wine aroma
601 compounds in grapes: A critical review. Crit. Rev. Food Sci. Nutr. 55: 202-218.

602 Haggerty, J., Bowyer, P.K., Jiranek, V., Taylor, D.K. 2015. Optimisation and validation of a high-
603 throughput semi-quantitative solid-phase microextraction method for analysis of fermentation aroma
604 compounds in metabolomic screening studies of wines. Aust. J. Grape Wine R. 10

27
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

605 Koundouras, S., Marinos, V., Gkoulioti, A., Kotseridis, Y., and van Leeuwen, C. 2006. Influence of
606 vineyard location and vine water status on fruit maturation of nonirrigated cv. Agiorgitiko (Vitis
607 vinifera L.). Effects on wine phenolic and aroma components. J. Agric. Food Chem. 54: 5077-5086.

608 Mencarelli, F., and Tonutti, P. 2013. Sweet, reinforced and fortified wines: Grape biochemistry,
609 technology and vinification. John Wiley & Sons.

610 Morlat, R. and Bodin, F. 2006. Characterization of viticultural terroirs using a simple field model based
611 on soil depth – II. Validation of the grape yield and berry quality in the Anjou vineyard (France).
612 Plant Soil 281:55–69.

613 Onofri, A. 2007. Routine statistical analyses of field experiments by using an Excel extension.
614 Proceedings 6th National Conference Italian Biometric Society: "La statistica nelle scienze della vita e
615 dell’ambiente", Pisa, 20-22 June 2007: 93-96.

616 Perestrelo, R., Silva, C., Câmara J.S. 2014. A useful approach for the differentiation of wines according to
617 geographical origin based on global volatile patterns. J. Sep. Sci. 37:1974–1981.

618 Piñeiro, Z., Natera, R., Castro, R., Palma, M., Puertas, B., Barroso, C.G. 2006. Characterisation of
619 volatile fraction of monovarietal wines: influence of winemaking practices. Anal. Chim. Acta 563:
620 165–172.

621 Ribéreau-Gayon,P., Glories,Y., Maujean,A. and Dubourdieu,D. 1998. L’arôme des cépages. In:Traité
622 d’oenologie. Tome 2.Chimie du vin, stabilisation et traitements. (Dunod:Paris) pp.239–245.

623 Robinson, A. L., Boss, P. K., Solomon, P. S., Trengove, R. D., Heymann, H., and Ebeler, S. E. 2014.
624 Origins of grape and wine aroma. Part 1. Chemical components and viticultural impacts. Am. J. Enol.
625 Vitic. 65:1-24.

626 Sabon, I., de Revel, G., Kotseridis, Y., and Bertrand, A. 2002. Determination of Volatile Compounds in
627 Grenache Wines in Relation with Different Terroirs in the Rhone Valley. J. Agric. Food Chem. 50,:
628 6341-6345 Sagratini, G., Maggi, F., Caprioli, G., Cristalli, G., Ricciutelli, M., Torreggiani, E., Vittori,
629 S. 2012. Comparative study of aroma profile and phenolic content of Montepulciano monovarietal red
630 wines from the Marches and Abruzzo regions of Italy using HS-SPME–GC–MS and HPLC–MS. Food
631 Chem. 132: 1592–1599

632 Sala, C., Busto, O., Guasch, J., Zamora, F. 2004. Influence of vine training and sunlight exposure on the
633 3-alkyl-2-methoxypyrazines content in musts and wines from the Vitis vinifera variety Cabernet
634 Sauvignon. J. Agric. Food Chem. 52:3492–3497.

635 Soil Survey Staff, 1998. Keys to Soil Taxonomy. USDA-NRCS. Eighth Edition.

636 Swiegers, J.H., Bartowsky, E.J., Henschke, P.A., Pretorius, I.S. 2005. Yeast and bacterial modulation of
637 wine aroma and flavour. Aust. J. Grape Wine R. 11: 139–173.

638 USDA-NRCS, (1999) Soil taxonomy, 2nd Edition. Agricultural Handbook n. 436.Unitted States
639 Department of Agriculture (USDA), U.S. Government Printing Office, Washington DC, USA.Van

28
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

640 Leeuwen, C., Friant, P., Chone, X., Tregoat, O., Koundouras, S., & Dubourdieu, D. 2004. Influence of
641 climate, soil, and cultivar on terroir. Am. J. Enol. Vitic. 55: 207-217.

642 Van Leeuwen, C., Friant, P., Chone, X., Tregoat, O., Koundouras, S., & Dubourdieu, D. (2004). Influence
643 of climate, soil, and cultivar on terroir. Am. J. Enol. Vitic. 55: 207-217.

644 Vas, G. Y., Ga´l, J., Harangi, J., Dobo´, A., & Ve´key, K. (1998). Determination of volatile aroma
645 compounds of Blaufrankisch wines extracted by solid-phase microextraction. J. Chrom. Sci. 36: 505–
646 510.

647 Wang R., Sun Q., Chang Q. (2015) Soil Types Effect on Grape and Wine Composition in Helan
648 Mountain Area of Ningxia. PLoS ONE 10

649 Winkler, A. J. 1962 General Viticulture. University of California, USA.

650 Zerihun, A., McClymont, L., Lanyon, D., Goodwin, I., & Gibberd, M. (2015). Deconvoluting effects of
651 vine and soil properties on grape berry composition. J. Sci .Food Agric. 95: 193-203.
652

29
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

Table 1 Volatile compounds identified in the analysis of red wine made from Aleatico grapes.

Area values
LF MM MP SA SM RI RI*
1 Ethyl acetate 3.15E+07 2.76E+07 2.41E+07 2.51E+07 3.39E+07 900 907

2 Isoamyl acetate 4.20E+06 8.32E+06 1.08E+07 2.85E+06 6.07E+06 1136 1132

3 Ethyl caproate 1.81E+07 1.50E+07 1.61E+07 1.55E+07 1.61E+07 1243 1244

4 Hexyl acetate 2.55E+06 1.35E+06 1.99E+06 2.63E+06 1.93E+06 1327 1330

5 Ethyl lactate 4.98E+06 4.06E+06 4.13E+06 5.20E+06 5.91E+06 1343 1341

6 Ethyl caprylate 1.43E+08 1.13E+08 1.25E+08 1.09E+08 1.26E+08 1454 1446

7 Ethyl decanoate 7.06E+07 8.35E+07 8.32E+07 5.04E+07 7.23E+07 1658 1660

8 Ethyl laurate 4.79E+06 5.03E+06 4.55E+06 2.05E+06 4.08E+06 1858 1842

9 Acetaldehyde 2.07E+06 2.01E+06 2.25E+06 1.54E+06 1.72E+06 714 750

10 Heptanal 4.81E+05 2.40E+05 1.55E+06 3.41E+05 2.56E+05 1174 1172

11 Nonanal 7.22E+05 4.00E+06 8.23E+05 1.97E+06 1.30E+06 1416 1398

12 Decanal 9.27E+05 4.68E+06 2.70E+06 4.29E+06 2.98E+06 1524 1520

13 α -terpinene 1.39E+06 8.31E+05 9.07E+05 1.66E+06 1.11E+06 1178 1182

14 Linalool 3.18E+06 1.94E+06 2.06E+06 3.05E+06 2.59E+06 1553 1558

15 Terpineol 1.84E+06 1.09E+06 1.11E+06 1.70E+06 1.19E+06 1646 1636

16 Citronellol 6.21E+06 6.54E+06 3.94E+06 6.39E+06 6.26E+06 1750 1765

17 Geraniol 3.87E+05 2.73E+05 2.11E+05 9.48E+05 6.26E+05 1847 1885

18 Limonene 2.98E+06 1.85E+06 2.01E+06 3.42E+06 2.44E+06 1201 1205

19 Ethanol 1.92E+09 2.20E+09 1.85E+09 1.80E+09 1.77E+09 954 929

20 Isoamyl alcohol 4.71E+07 5.73E+07 7.09E+07 5.94E+07 7.17E+07 1216 1215

21 1-Hexanol 3.67E+06 1.42E+06 1.66E+06 1.93E+06 2.18E+06 1354 1360

22 Heptanol 2.12E+05 2.99E+05 5.07E+06 2.18E+05 5.22E+06 1313 1320

23 Octanol 1.03E+06 2.13E+05 4.66E+05 6.01E+05 5.92E+05 1570 1565

24 Nonanol 3.65E+05 1.88E+05 7.51E+05 3.71E+05 2.92E+05 1527 1525

25 Phenylethanol 1.64E+07 2.69E+07 3.75E+07 1.05E+07 2.03E+07 1956 1925

26 Total 2.29E+09 2.57E+09 2.25E+09 2.11E+09 2.15E+09 900 907

RI (mass spectral matching against library standards) and RI* (values for a Carbowax 20M column after running a
carbon alkane standard. also known as Kovats index KI).

30
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

Table 2 Winkler index values for the cv Aleatico native growing area.

2007 2008 2009 2010 2011 2012 Mean SD

1718.9 1696.1 1991.3 1746.9 1979.7 1832.0 1827.5 130.8

Table 3 Main soil physical characteristics for the three typology identifyed in the study area and in each sample
vineyard of the five sub-areas. SA, Sant’Antonio; LF, La Fratta; MP, Macchia del Prete; SM, San Magno; MM,
Montemaggiore.

Soil local Skeleton Vineyards’

Soil depth Texture Soil classification (†)
denomination content locality
Vitrandic Xerorthent
High Sandy SA
sandy skeletal, mesic
Lapillo Deep
Sandy Typic Xeropsamment,
Moderate SM, LF
loam mixed, mesic
Typic Hapludalf fine
Volpaio Deep Low Clay loam MP
loamy, mesic
Vitrandic Xerorthent
Pianca Shallow High Sandy MM
sandy skeletal, mesic
(†)Soil Survey Staff, 1998, USDA-NRCS, 1999.

31
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

Table 4 (A-B) Main soil physical characteristics for the three typology identifyed in the in the five sub-areas
related to berry biochemical traits [A] and volatile compounds [B] at harvest time (Mean, Standard deviation and
significance among sub-areas (p<0.05)). SA, Sant’Antonio; LF, La Fratta; MP, Macchia del Prete; SM, San
Magno; MM, Montemaggiore.
[A]

Vineyards' Acidity (g·L-1 Antocyanins Poliphenols

Texture Sugar (°Brix)
locality tartaric acid) (mg·L-1) (mg·L-1)
SA Sandy 20.9 ± 2.7 b 8.8 ± 2.1 a 605.4 ± 212.6 a 1564.0 ± 198.0 b
SM Sandy-loam 25.40 ± 3.8 a 4.2 ± 1.1 b 739.4 ± 174.2 a 1514.3 ± 145.8 b
LF Sandy-loam 22.0 ± 2.8 ab 4.8 ± 1.1 b 692.7 ± 207.9 a 1566.0 ± 171.1 b
MP Clay-loam 25.4 ± 1.8 a 4.1 ± 0.9 b 820.0 ± 278.9 a 1824.1 ± 221.8 a
MM Sandy 21.5 ± 3.4 b 7.1 ± 1.7 ab 742.4 ± 354.9 a 1374.5 ± 106.8 c

[B]

Linalool Terpineol Citronellol Nerol Geraniol

SA 17.9% e 3.2% b 67.7% a 10.2% bc 1.0% b
SM 30.9% b 3.9% ab 52.0% c 11.1% b 2.2% b
LF 44.9% a 5.7% a 5.5% d 39.5% a 4.4% a
MP 24.8% d 5.3% a 57.3% b 8.6% c 4.0% a
MM 28.5% c 5.2% a 53.7% c 11.5% b 1.1% b

Table 5 – PCA : eigenvalues. explained and cumulate inertia.

Explained
N° Eigenvalue Cumulate
Inertia
Inertia 0 2 4 6 8 10

1 11.9 44.0% 44.0%

2 8.8 32.4% 76.4%
3 3.3 12.0% 88.4%
4 3.1 11.6% 100.0%

32
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

Table 6 – PCA: contribution and correlation of the indicators in respect to the two principal axes selected.

I Axis II Axis
Contribution Correlation Contribution Correlation
1 Ethyl acetate 262 0.56 87 0.28
2 Isoamyl acetate 704 -0.91 117 0.32
3 Ethyl caproate 351 0.76 660 0.65
4 Hexyl acetate 545 0.80 251 0.47
5 Ethyl lactate 559 0.81 28 0.16
6 Ethyl caprylate 168 0.45 827 0.85
7 Ethyl decanoate 534 -0.80 259 0.48
8 Ethyl laurate 298 -0.59 173 0.39
9 Acetaldehyde 286 -0.58 684 0.77
10 Heptanal 199 -0.49 474 0.64
11 Nonanal 75 -0.30 1040 -0.95
12 Decanal 5 -0.08 671 -0.77
13 α -terpinene 719 0.92 5 0.06
14 Linalool 796 0.97 57 0.22
15 Terpineol 735 0.93 81 0.27
16 Citronellol 569 0.82 238 -0.46
17 Geraniol 656 0.88 28 -0.16
18 Limonene 741 0.94 15 0.12
19 Ethanol 29 0.19 752 -0.81
20 Isoamyl alcohol 5 -0.08 99 0.30
21 1-Hexanol 363 0.48 263 0.66
22 Heptanol 114 -0.37 1071 0.59
23 Octanol 571 0.53 323 0.82
24 Nonanol 13 -0.13 511 0.67
25 Phenylethanol 662 -0.89 199 0.42
26 Total anthocyanins 2 0.04 715 -0.79
27 Total polyphenols 38 0.21 1090 -0.98

33
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

Figure 1 Map of the study area (A). localization of the sub-areas within the PDO ‘Aleatico di Gradoli’ (B) and
spatial distribution of tested vineyards in respect to steepest downslope direction (C). SA. Sant’Antonio; LF. La
Fratta; MP. Macchia del Prete; SM. San Magno; MM. Montemaggiore.

34
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

Figure 2 Ombrothermic diagram based on meteorological data from a weather station close to Gradoli. The rainfall
(dotted grey line) and temperature (continue black line) represent the mean value of the last 11 years (2004-2015).

Figure 3 The correlation analysis (Pearson’s coefficient) between biochemical berry traits (sugar [A] and
poliphenols [B]) and soil’s sand content in the tested vineyards. SA. Sant’Antonio; LF. La Fratta; MP. Macchia del
Prete; SM. San Magno; MM. Montemaggiore.

35
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

Figure 4 PCA: representation of all berry quality parameters (chemical and volatile compounds) at harvest time. In
the box the cluster dendrogram.

36
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

Figure 5 Volatile and phenolic compounds determined in Aleatico di Gradoli (PDO) wines produced in the five
sub-areas (SA. Sant’Antonio; LF. La Fratta; MP. Macchia del Prete; SM. San Magno; MM. Montemaggiore) and
three soils type : Lapillo (fill color: white); Volpaio (fill color: black); Pianca (fill color :dark gray). Different letters
indicate significantly differences at p < 0.05. [A] Aldehydes. [B] Esters. [C] and [D] Higher alcohols. [E]
Monoterpenes. [F] Anthocyanins and Polyphenols.

37
 American Journal of Enology and Viticulture (AJEV). doi: 10.5344/ajev.2016.15123
AJEV Papers in Press are peer-reviewed, accepted articles that have not yet been published in a print issue of the journal
or edited or formatted, but may be cited by DOI. The final version may contain substantive or nonsubstantive changes.

Figure 6 - PCA: representation of all indicators on the plane spanned by the principal axes 1 and 2. HCA (dotted
line): representation of cluster analysis.

38