PaPer 310 Ital. J. Food Sci., vol. 27 - 2015 - Keywords: anthocyanins, biogenic amines, co-inoculation, procyanidin, sequential inoculation, wine - MANAGEMENT OF MALOLACTIC FERMENTATION AND INFLUENCE ON CHEMICAL COMPOSITION OF AGLIANICO RED WINES S. SURIANO*, M. SAVINO, T. BASILE, L. TARRICONE and D. DI GENNARO CRA Consiglio per la Ricerca e la Sperimentazione in Agricoltura, Unità di ricerca per l’uva da tavola e la vitivinicoltura in ambiente mediterraneo, Cantina Sperimentale di Barletta, Via Vittorio Veneto 26, 76121 Barletta, Italy *Corresponding author: Tel.+39 088 352 1346, Fax +39 088 352 8955, email: serafino.suriano@entecra.it AbstrAct A study has been carried out to determine the effects of lactic acid bacteria inoculation time on the fundamental components, procyanidins and biogenic amines content of Aglianico wines pro- duced in Apulia region. three different malolactic fermentation (MLF) techniques were compared: the co-inoculation, the sequential inoculation, and the traditional technique (spontaneous MLF). In the co-inoculation technique there was a delayed start and a late finish of the alcoholic fer- mentation. the colour intensity of the wine obtained with a spontaneous MLF was higher both at racking and after 12 months. significant changes in content of flavan-3-ols were found in wines made with different MLF managements. the levels of catechin monomers ((+)-catechin, (-)-epicat- echin, (-)-epicatechin-O-gallate) and procyanidin oligomers (b1-b4, and trimer c1) were lower in the co-inoculation wine. In the wine produced with a spontaneous MLF, the content of biogenic amines was significantly higher compared to the other two wines. Ital. J. Food Sci., vol. 27 - 2015 311 INtrODUctION the most important microbial activity that is responsible for the conversion of must into wine is the alcoholic fermentation (AF), which is carried out by saccharomyces yeasts. the malolactic fermentation (MLF) typically follows alcoholic fermentation and is carried out by in- digenous lactic acid bacteria (LAb) or induced by inoculation with selected bacterial starters. It is a decarboxylation of L-malic acid, a dicar- boxylic acid, with formation of a monocarbox- ylic acid, the L-lactic acid and carbon diox- ide, which is catalyzed by malolactic enzymes which are NADP dependent and require diva- lent cations such as manganese or magnesium ions (VINceNzINI et al., 2005). the MLF caus- es a significant evolution of wine and produc- es remarkable changes in its phenolic compo- sition and sensorial characteristics (cOsteLLO et al., 2012; cAbrItA et al., 2008; LòPez et al., 2011; sUrIANO et al., 2012). In addition to the reduction of the acidity of the wine, MLF in- creases the aromatic complexity and smooth- ness (VersArI et al., 1999; cOstANtINI et al., 2009; LOPez et al., 2011). Generally the MLF is favoured in red wines, in novello wines, in white wines aged in barrique, or in some spar- kling base wines (cAVAzzA et al., 2003). On the other hand, this fermentation produces a small amount of acetic acid and sometimes may also generate unpleasant odours, bitter-tasting com- pounds or substances that may be dangerous to consumers’ health, such as biogenic amines or precursors of ethyl carbamate (LONVAUD-FU- NeL, 1999). It has been verified by analysis that the concentration of biogenic amines in wine at the end AF is always quite low, while increas- es after MLF (GAFNer, 2005). Moreover, it was found that wines which undergo spontaneous MLF often have higher biogenic amine concen- tration than those in which the MLF is conduct- ed by select malolactic bacteria (cerrUtI et al., 1987; MAsqUè et al., 2008). biogenic amines are synthesized by microorganism through de- carboxylation of amino acids. between the main biogenic amines in wines there are tyramine, histamine, putrescine, cadaverine and pheny- lethylamine, synthesized by the decarboxyla- tion of the amino acids tyrosine, histidine, or- nithine, lysine and phenylalanine. these com- pound can cause adverse physiological reac- tions in susceptible individuals. Histamine can cause headaches, allergies, diarrhoea, palpita- tions and vomiting (stOckLey, 2004; bODMer et al., 1999), while tyramine is strongly vaso- constrictive (sILLA-sANtOs, 1996). these ef- fects may be enhanced by alcohol, which pre- vents the organism’s detoxifying mechanisms from working properly and by the presence of other amines such as putrescine and cadaver- ine (LANDete et al., 2005), both associated with poor sanitary quality of grapes (LeItAO et al., 2005) and responsible for major sensory defects in wines (LeHtONeN, 1996). Usually, the LAb used for MLF belong to the Oenococcus oeni species, anyway, it is possible to also find oth- er bacteria of the Lactobacillus, Leuconostoc and Pediococcus species (DIcks et al., 1995). However, even for the most resistant bacteria the conditions found in wine are close to the limits of survival, so that the transformation of 4-5 g/L of malic acid may requires even 15- 20 days (cAVAzzA et al., 2003). several times, this process may take several months, may oc- cur in some barrels and tanks but not in oth- ers and may be responsible for the occurrence of problems related to indigenous LAb species carrying out the MLF (LONVAUD-FUNeL, 2001) which may cause a range of undesirable chang- es to wine sensory properties, altered wine col- our, and may even lead to the generation of bio- genic amines (DAVIs et al., 1985). such a long time can be especially critical for those wines (such as novello wines) that must be processed and placed on the market in a short time, and moreover could be a risk since in the season in which the MLF takes place there may be sud- den temperature drops which may determine an arrest of the process until the next spring. there are advantages subsequent to an early and fast MLF such as: a more efficient utiliza- tion of fermentation tank in the busy posthar- vest period, thus a decrease of energetic costs resulting in optimization of the winemaking process; moreover it is possible a decrease of the microbiological risks reducing the growth of undesired microorganism and also allows an early commercialization of wines (JUssIer et al. 2006). It is therefore of fundamental im- portance a correct management of MLF. In this paper the influence of inoculation of lactic bac- teria on changes occurring on the polyphenol- ic characteristics, colour, biogenic amines and proanthocyanidin in Aglianico red wines was investigated by comparing the techniques of co-inoculation and sequential inoculation to a spontaneous MLF. MAterIALs AND MetHODs Experimental design and winemaking this research was conducted during the 2012 harvest on Aglianico grape variety, grown in a vineyard trained on espalier training system with Guyot pruning and cultivated according to the principles of organic viticulture. concern- ing the different possibilities of MLF manage- ment, in the Le.Vin.sud company of cerignola (Foggia, southern Italy) were carried out three experimental tests in order to evaluate the in- fluence of the timing of lactic bacteria inocula- tion, comparing the technique of co-inoculation (inoculation of bacteria 24 hours after the yeast 312 Ital. J. Food Sci., vol. 27 - 2015 inoculation), sequential inoculation (at the end of the AF) and the traditional technique without inoculation of any LAb, i.e. a spontaneous MLF, which was favoured by acting on certain oeno- logical practices, as further explained. the Agli- anico grapes were first destemmed and crushed, subsequently the mass of must and pomace was mixed, homogeneized and introduced in three different steel tank. From each steel tank, 3 x 100 kg (in triplicate) of must with pomace was utilized for each of the three winemaking tech- niques adopted, with the aim to determine the repeatability of the differences among the com- pared treatments. the different batches of must and pomace were subjected to the following win- emaking protocols: - co-inoculation or simultaneous inoculation of LAb (sIM). After crushing and destemming of grapes 40 mg/L of sO 2 was added. After two hours, Saccharomyces cerevisiae strain Lalvin r7 (Lallemand Inc, castel D’Avezzano-Verona – Italia) previously hydrated in water for 15 min at 38 °c was inoculated in the must (20 g/hL, about 6 x 106 cfu/mL.). After 24 hours a lac- tic bacterial culture of Lactobacillus plantarum V22tM (Lallemand Inc, Verona-Italy) was inocu- lated. the inoculation rate was 1g/hL (2 x 107 cfu/mL) must/wine prior re-hydrated in chlo- rine free water at 20°c for 15 min. the alcohol- ic fermentation took place under controlled tem- perature by cooling the mass if the temperature exceeded the threshold of 26°c. - sequential inoculation post alcoholic fer- mentation of LAb (PAF). the only difference from the previous protocol was the time of addition of the bacteria. the lactic bacteria were added at racking, which was performed at the end of the alcoholic fermentation (10 day pomace con- tact). the doses of yeast and bacteria employed were the same. After the inoculation of bacteria, at a dosage of 20 g/hL Opti’Malo Plus bacteri- al nutrient (Lallemand Inc, Verona –Italy) were added at wine in according to the manufactur- ers instructions. - spontaneous MLF (control). this MLF pro- cess was used as a comparison test for the oth- ers processes. After crushing and destemming of grapes were added about 40 mg/L of sO2, and then 20 g/L of previously hydrated Saccharomy- ces cerevisiae (Lalvin r7) yeast were inoculated. Also for this thesis, at racking/post alcoholic fer- mentation were added 20 g/hL of Opti’Malo Plus bacterial nutrient. All the vinification were carried out at 26°c ± 1. During the fermentative pomace contact period (10 days in all vinifications) the cap was pumping over three times a day and the temper- ature and must density were recorded. At the end of this period, all wines were pressed at 2 bars, racked with no added sulphur dioxide for encourage MLF (in control and PAF) and stored at 25°c. After MLF, the wines were racked again and 20 mg/L sulphur dioxide was added. the wines were cold stabilised (-4°c) for 1 month and then bottled without filtration. All analyses were made in triplicate at racking and after 6 months in the bottle (12 months after racking). AF was monitored by ethanol production and sugar depletion. MLF was monitored by l-mal- ic acid degradation and l-lactic acid production. AF and MLF were considered complete when re- sidual sugars were less than 2.5 g/L and l-mal- ic acid was less than 0.12 g/L. Wine composition total acidity, volatile acidity, reducing sug- ars, pH, total sO 2 , alcohol and total dry extract were all determined on wine according to eec regulation 2676/90. Chemicals and reference compounds standards, including trans-caffeoyl-tartar - ic acid, trans-p-coumaroyl-tartaric acid, caffeic acid, ferulic acid, p-coumaric acid, quercetin, myricetin, kaempferol, were supplied by sigma Aldrich. While standard of (+)-catechin, (-)-epi- catechin, procyanidin b1, procyanidin b2, ep- igallocatechin, epigallocatechin gallate were supplied by extrasynthese. the purities of the standards were all over 95%. All the solvents (methanol, acetonitrile, ethyl acetate) were HPLc grade. All the solutions were obtained with dis- tilled deionised water using carlo erba reagents. Spectrophotometric analysis Phenolic compounds were determined by spectrophotometric methods (DI steFANO et al., 1989; Di stefano et al., 1997) using a UV/ VIs Mod Lambda 25 double beam spectropho- tometer (Perkin elmer s.p.A.). the total antho- cyanins index was expressed as malvidin 3-glu- coside and calculate by the following expression: e maxvis x 16.17 x d (d=dilutions). the monomeric anthocyanins after separation and absorption on a c18 set Pak cartridge were eluted with 5 ml of acetonitrile and then diluted with hydrochlo- ric ethanol and calculated by: e maxvis x 16.17 x d (d=dilutions). the total polyphenols index ex- pressed as (+)-catechin was measured by: e1cm, 75 0nm x 186.5 x d (d=dilutions). the total fla- vonoids index was expressed as (+)-catechin and calculate with the graphic method of DI steFA- NO (1989). the flavanols reactive to vanillin (fla- vonols vanillin assay) were expressed as (+)-cat- echin = Δe x 290.8 x d (Δe=absorbance differ- ence between tests with and without vanillin; d=dilution). the proanthocyanidin content was determined after hot acid hydrolysis (bate-smith reaction) using a ferrous salt (FesO 4 ) as catalyst and expressed as cyanidin chloride. colour in- tensity and hue were estimated by measuring absorbance at 420, 520 and 620 nm according to eU regulation 1990. Ital. J. Food Sci., vol. 27 - 2015 313 HPLC analysis the fixed acids of wine (tartaric, malic, lac- tic, citric and shikimic acids) were determined by an HPLc isocratic elution (HPLc 1100 series Agilent technologies) with a Phenomenex syner- gi 4u Hydro-rP 80A (250x4.60 mm, 4 micron) with guard column, a mobile phase of phos- phoric acid 10-3 M, 0.7 mL/min flow rate, 25°c and a UV detector set at 210 nm (cANe, 1990). For flavans determination, the wine was sep- arated into two fractions containing, respective- ly, individual catechins and oligomeric proan- thocyanidins, using a c18 1g sep-Pak cartridge as described by sUN et al. (1999). About 5 ml of wine was adjusted to pH 7 and then filtered through a sep-Pak cartridge preconditioned with H 2 O. elution was carried on with 10 mL of H 2 O to eliminate phenolic acids. After drying the cartridges with N 2 , elution was carried out with 15mL of ethyl acetate to elute catechins and oligomeric proanthocyanidins (F I + F II). each fraction was evaporated to dryness and dissolved in methanol, followed by HPLc anal- ysis. A thermo ODs rP-c18 Hypersil 200x2.1 (5 μm) column with a guard column was used for flavans analysis. two ml of each extract- ed fraction were filtered on a 0.45 μm nylon membrane and immediately inject according to squadrito’s method (2007). separation was carried out at 30°c, the flow rate was 0.25 mL/ min and the injection volume 10 μL. the detec- tion was set at 280 nm, using phosphoric acid 10-3 M (solvent A) and acetonitrile (solvent b). the gradient elution program was: from 91 to 86% A in ten minutes; from 86 to 82% A in ten minutes; from 82 to 60% A in ten minutes; from 60 to 40% A in five minutes; from 40 to 91% A in five minutes; equilibration time of five min- utes. the peaks identification was performed comparing the retention times and absorption spectra of pure compounds (supplied from ex- trasynthese) and were found analogues to val- ues reported in the literature (bAOsHAN et al., 1998; rIcArDO et al., 1991). the determination of biogenic amines (bA) in wine was carried out by HPLc/FLD. A Hewl- ett-Packard (Agilent technologies Palo Alto, cA, UsA) 1100 series HPLc instrument was used, with a fluorescence detector set at excitation and emission wavelengths of 340 and 450 nm, respectively. the samples were subjected to an automatic pre-column derivatization pro- cedure using o-phthalaldehyde (OPA reagent, Agilent technologies, Palo Alto, cA, UsA). All separations were performed on a 200 x 4.6 mm, 5-µm Alltima c18 column (Alltech, Deerfield, IL, UsA), protected by a 7.5x4.6 mm guard car- tridge of the same type. samples were injected into the column after being filtered through a 0.2 mm rc filter (schleicher and schuell, keen, NH, UsA). the two eluents used as mobile phas- es were sodium acetate 50 mM (pH 7.2)/tHF (96:4) v/v (eluent A) and methanol (eluent b). the elution gradient programme followed the method described by NIcOLINI (2003). From a stock solution of 200 mg/L containing ag- matine, cadaverine, phenylethylamine, hista- mine, putrescine, and tyramine (standards pur- chased by sigma-Aldrich) in methanol, four di- luted solution were prepared and injected: 2.5, 5.0, 10.0, 20.0 mg/L. quantification of the bA was performed with an internal standard of 10mM of norvaline solution. Statistical analysis Multivariate statistical analysis was per - formed using r statistical software (r core team (2013), r Foundation for statistical comput- ing, Vienna, Austria). chemical analyses were repeated three times for each sample and the data are presented as mean ± sD. the one way analysis of variance (ANOVA), and Duncan mul- tiple comparison test to measure variation be- tween treatments at a probability level of p<0.05 were performed. resULts AND DIscUssION Wines composition the musts collected from the steel tanks had the following chemical/physical charac- teristics: control (spontaneous MLF) 210 g/L of reducing sugars, pH 3.30 and total acid- ity 6.40 g/L; sIM: 205 g/L of reducing sug- ars, pH 3.27 and total acidity 6.52 g/L; PAF: 214 g/L of reducing sugars, pH 3.35 and to- tal acidity 6.24 g/L. the winemaking process began on the 12th of October with the crush- ing and destemming of grapes and the yeasts inoculation for all the three experimental pro- cesses. the kinetics of AF and malic acid deg- radation are reported in Figs. 1 and 2 respec- tively. In table 1 it is reported the time re- quired for the AF and the MFL for each wine- making. the duration of the fermentation pro- cess was identical for the PAF and the con- trol (both 8 days), while it was longer for the sIM (about 10 days). However, all alcohol- ic fermentations were regular and complete. LAb in the sIM were able to perform MLF in 23-24 days from the beginning of winemak- ing. the wine obtained by sequential inoc- ulation (PAF) carried out the degradation of malic acid in 40-41 days from the beginning of the winemaking. Instead, the wine under - went a spontaneous MLF, despite the absence of added LAb, has finished the MLF after 57 days from the beginning of the vinification. therefore, the wine obtained by the sIM tech- nique has finished the MLF 33-34 days before of control wine. this data is important since time is a key factor from an economic, techni- 314 Ital. J. Food Sci., vol. 27 - 2015 Fig. 1 - kinetics of alcoholic fermentation. Fig. 2 - time course of malic acid degradation from the start of alcoholic fermentation. table 1 - time required to complete AF and MLF. Treatment Time for AF Time for malic acid degradation Vinification time (days) (days after bacterial inoculation) † AF + MLF (days) ‡ Co-inoculation (SIM) 10±0 23±1 24±1 Sequential inoculation (PAF) 8±0 33±4 41±4 Spontaneous MLF (Control) 8±0 NA 57±4 † MLF was considered complete when malic acid concentration was below 0.12 g/L. ‡Vinification time is the time from destemming/crushing to completion of AF and MLF. cal and practical point of view, for a good or - ganizational management of the winery. ta- ble 2 shows the results of the chemical/phys- ical analysis of wines at the end of alcohol- ic fermentation (racking) and 12 months af- ter racking. Differences were observed in the alcohol content, acidic profile, pH and total acidity. the sIM wine after alcoholic fermen- tation showed obvious signs of the beginning of MLF. Indeed, the malic acid content (1.44 Ital. J. Food Sci., vol. 27 - 2015 315 g/L) was less than control (1.61 g/L) and PAF (1.64 g/L) wines. In sIM wine the partial transformation of malic acid has produced a certain amount of lactic acid already at rack- ing. Moreover, sIM wine showed a lower to- tal acidity and a higher pH compared to con- trol and PAF wines. this was mainly due to the transformation of a diprotic acid (malic acid) with two acidic functional groups into a monoprotic acid (lactic acid) with only one acidic functional group, with a correspond- ing decrease in acidity and an increase of pH. It was observed a difference in the alcohol content of wines, in particular the sIM wine showed the lowest alcohol content. Probably, since the sugar content of the must subjected to the sIM process was slightly smaller thus less alcohol was produced. Another possible explanation is linked to the lactic acid bacte- ria that could have used part of the reducing sugars, in addition to malic acid, as nutrients for their metabolism. this not only may have influenced the alcohol content of wine, but furthermore had furnished an increased en- ergy for the cellular development of the bac- teria resulting in the production of more vol- atile compounds and greater amounts of ace- tic acid, as it was found in sIM wine. Indeed, the volatile acidity expressed as acetic acid was slightly higher in the sIM wine than in the other two wines. both possibilities may have contributed to the lower alcoholic con- table 2 - Wines composition after AF (racking) and after MLF (12 months after racking). At racking 12 months after racking Control SIM PAF Control SIM PAF Alcohol (Vol. %) x 12.39 ab 12.06 b 12.48 a 12.40 a 12.10 b 12.42 a s 0.08 0.07 0.08 0.12 0.08 0.09 Residual sugars (g/L) x 2.40 a 2.40 a 2.40 a 2.35 b 2.42 a 2.45 a s 0.18 0.15 0.20 0.22 0.25 0.20 Total dry extract (g/L) x 30.50 a 29.40 b 30.50 a 30.20 a 29.80 b 30.40 a s 2.50 1.80 2.10 2.08 2.18 2.10 pH x 3.35 b 3.53 a 3.36 b 3.45 b 3.61 a 3.44 b s 0.02 0.03 0.02 0.03 0.02 0.02 Total acidity (g/L) x 7.65 a 6.15 b 7.50 ab 5.63 a 5.03 c 5.10 b s 0.38 0.42 0.45 0.35 0.40 0.39 Volatile acidity (mg/L) x 0.54 b 0.60 a 0.56 b 0.55 b 0.60 a 0.50 c s 0.03 0.02 0.03 0.07 0.08 0.06 Total SO2 (mg/L x 24.05 a 22.10 b 22.04 b 32.10 b 58.10 a 30.10 c s 4.10 3.80 3.50 2.90 3.50 2.80 Tartaric acid x 3.04 b 2.90 c 3.20 a 2.91 a 2.86 ab 2.62 b s 0.28 0.22 0.18 0.4 0.38 0.25 L-malic acid (g/L) x 1.61 ab 1.44 b 1.64 a 0.12 a 0.05 b 0.06 b s 0.09 0.10 0.12 0.01 0.00 0.00 L-lactic acid (g/L) x 0.10 b 0.60 a 0.12 b 1.45 b 1.52 a 1.41 b s 0.00 0.01 0.00 0.08 0.06 0.05 Citric acid (g/L) x 0.25 b 0.27 ab 0.28 a 0.23 b 0.30 a 0.25 ab s 0.02 0.02 0.04 0.01 0.02 0.02 x, mean of three replicates; s, standard deviation. Mean values followed by the same letter in a row are not significantly different at the 0.05 level of significance. tent, as it is confirmed by the findings of some other authors which have observed a delay of alcoholic fermentation and a use of the sug- ars of must by LAb (LAFON-LAFOUrcADe et al., 1983). After one year of storage, the sIM wine still has a lower acid strength, represent- ed by a higher pH and a lower total acidity compared to the other wines. the content of sulfur dioxide, in order to favour the MLF es- pecially in the control, has been deliberately kept low. there were no significant differenc- es in respect of tartaric acid and citric acid. Polyphenolic composition and wine colour table 3 shows the polyphenolic composi- tion and chromatic characteristics of wines af- ter alcoholic fermentation and 12 months af- ter racking. the effects of different MLF starts showed a marked change in the content of polyphenols in sIM wine already at the end of the AF. Indeed, the index of total polyphe- nols, the total flavonoids, the total and mon- omeric anthocyanins contents are higher in the sIM wine, with variations ranging from 5 to 17%, than in the other wines. the differenc- es in tannins (proanthocyanidins and flavans reacting with vanillin) content between wines were not significant. After 12 months from racking, all the wines had finished the MLF thus a natural reduction of the polyphenolic compounds (total flavonoids, flavans, and an- 316 Ital. J. Food Sci., vol. 27 - 2015 thocyanins) was observed. Anyway, the total anthocyanins were still slightly higher in the sIM wine compared to control and consider- ably higher compared to PAF wine. From the other side, the content of proanthocyanidins (high molecular weight tannins) was signifi- cantly higher in the control than in the oth- er wines. therefore, an early MLF as for the sIM and PAF processes, causes greater losses over time of higher molecular weight tannins (proanthocyanidins). the increase in pH ob- served at racking for sIM wine has promoted the polymerization processes of high molecu- lar weight tannins leading to a partial precip- itation and thus their removal by pouring op- erations. some authors (cArOLINe and eVe- LINe, 2011; cOsteLLO et al., 2012) in tests of management of MLF have observed similar ef- fects in respect of these compounds. It is well known that MLF can reduce the colour inten- sity in red wines due to numerous factors as- sociated with the MLF (bUrNs  et al., 2013). Indeed, the Aglianico wines obtained with the sIM process, already at the end of alcoholic fermentation (table 3), showed a colour in- tensity of 8.84 which was significantly lower than PAF and control wines. Usually, this in- dex is positively correlated to the anthocyanin content, in this case, considered that in sIM wine there was an increase in pH mostly due to the partial MLF, despite a slightly higher an- thocyanin content respect to the other wines, table 3 - Phenolic composition and chromatic characteristics of Aglianico wines after AF (racking) and after MLF (12 months after racking). At racking 12 months after racking Control SIM PAF Control SIM PAF Total phenols (mg/L) x 1812 b 2062 a 1831 b 1674 b 1793 a 1652 ab s 66 66 37 49 52 41 Total flavonoids (mg/L) x 1982 c 2388 a 2008 b 1466 b 1585 a 1583 a s 73 74 45 42 44 36 Vanillin index (mg/L) V x 1026 a 1013 b 1017 b 829 b 935 a 659 c s 38 39 23 19 33 21 Proanthocyanidins (mg/L) L x 2489 a 2490 a 2398 b 2135 a 1693 b 1533 c s 71 77 59 67 55 38 Total anthocyanins (mg/L) x 321 ab 338 a 315 b 223 b 242 a 169 c s 15 16 19 12 12 15 Monomeric anthocyanins (mg/L) x 232 ab 251 a 227 b 90 b 110 a 92 ab s 12 9 12 4 6 8 D.O. 420 nm (P.O. 1cm) x 3.49 a 2.82 c 3.06 b 2.84 a 2.54 b 2.47 ab s 0.07 0.07 0.06 0.06 0.05 0.05 D.O. 520 nm (P.O. 1cm) x 7.16 a 5.17 c 6.23 b 4.03 a 3.69 b 3.39 c s 0.20 0.16 0.18 0.10 0.01 0.01 D.O. 620 nm (P.O. 1cm) x 1.06 a 0.85 c 0.94 b 0.75 c 0.78 b 0.99 a s 0.03 0.02 0.03 0.02 0.03 0.04 Colour intensity (P.O. 1cm) x 11.71 a 8.84 c 10.23 b 7.62 a 7.01 b 6.85 c s 0.20 0.10 0.14 0.10 0.12 0.08 Tint (E420/E520) x 0.49 0.54 0.49 0.70 0.69 0.73 V/L index x 0.41 0.41 0.44 0.39 0.55 0.43 x, mean of three replicates; s, standard deviation. Mean values followed by the same letter in a row are not significantly different at the 0.05 level of significance. a change in the balance of the pH-dependent anthocyanin pigments has determined a cer- tain loss of the red colour. the colour reduc- tion may also be due to the precipitation of the free anthocyanins molecules with polysaccha- rides and potassium bitartrate. twelve months after racking, the differences in colour inten- sity, although reduced, remained quite sig- nificant and the control wine still showed the highest value. Also the 420 nm and 520 nm absorbances were higher in the control wine, while the hue did not show significant differ- ences between wines. the composition of monomeric catechins and oligomeric procyanidins is shown in table 4. A common feature to all wines is the predomi- nance of the (+)-catechin among all monomeric flavanols. Among dimeric procyanidins b2 and b4 are present in greater quantities. the trim- eric procyanidins were detected in all wines but in small quantities. At racking, the sIM wine dif- fered from the other two wines because of the lowest content of almost all the flavan-3-ols, with the exception of a few gallic acid esters such as epicatechin gallate, epigallocatechin gallate and procyanidin b2 gallate. Less difference were found between the control and PAF. After one year from racking, all wines showed a reduc- tion in the content of (+)-catechin, (-)-epicate- chin, epigallocatechin, procyanidin b1, b3 and b4, while there was a general increase of procy- anidin b2, gallic acid esters and trimeric procy- Ital. J. Food Sci., vol. 27 - 2015 317 anidins c1 and t2. the sIM wine confirmed a lower content of almost all forms of monomeric catechins and oligomeric procyanidins, while the PAF showed concentrations that were even high- er than control wine. Also in this case, a fast- er MLF in sIM and PAF from the early stages of racking had caused a lower acidic strength, re- sulting in a loss of these compounds. table 5 - concentration (mg/L) of biogenic amines in Aglianico wines. At racking 12 months after racking Control SIM PAF Control SIM PAF Histamine x 2.78 a 2.44 c 2.65 b 3.53 a 0.24 ab 0.20 b s 0.74 0.22 0.39 0.92 0.02 0.02 Agmatine x 1.45 b 1.54 a 1.57 a 1.56 c 2.41 b 2.93 a s 0.46 0.38 0.40 0.55 0.63 0.74 Putrescine x 3.74 a 3.32 b 3.66 ab 10.51 a 8.48 c 9.54 b s 0.82 0.67 0.74 1.86 1.59 1.48 Tyramine x 0.70 b 0.72 ab 0.74 a 0.76 b 1.40 a 0.60 c s 0.08 0.09 0.06 0.09 0.10 0.03 Cadaverine x 1.42 ab 1.35 b 1.44 a 1.60 a 1.59 a 1.16 b s 0.36 0.42 0.40 0.39 0.28 0.29 Phenylethylamine x 0.29 ab 0.36 a 0.27 b 0.42 b 0.60 a 0.58 ab s 0.05 0.06 0.06 0.06 0.07 0.06 Total biogenic amines x 10.38 9.73 10.33 18.38 14.72 15.01 x, mean of three replicates; s, standard deviation. Mean values followed by the same letter in a row are not significantly different at the 0.05 level of significance. table 4 - concentration (mg/L) of monomeric catechins and oligomeric procyanidins in Aglianico wines. At racking 12 months after racking Control SIM PAF Control SIM PAF (+)-Catechin x 40.36 ab 36.7 c 41.34 a 35.06 b 28.73 c 39.68 a s 1.74 1.83 1.88 1.25 1.44 1.32 (−)-Epicatechin x 25.61 b 19.24 c 27.22 a 21.98 b 16.96 c 24.59 a s 1.21 0.71 1.29 1.12 1.09 1.15 Procyanidin B1 x 17.60 a 13.30 b 17.42 ab 12.09 a 7.76 c 10.39 b s 0.45 0.80 0.73 0.71 0.74 0.83 Procyanidin B2 x 37.40 a 29.60 b 36.40 ab 39.06 b 33.18 c 40.95 a s 1.75 1.67 1.98 1.44 1.56 1.22 Procyanidin B3 x 10.80 a 5.81 b 10.30 ab 7.52 b 4.95 c 8.84 a s 0.78 0.46 1.04 0.34 0.46 0.27 Procyanidin B4 x 30.73 b 27.55 c 32.60 a 28.26 b 24.77 c 32.59 a s 1.75 1.85 1.78 1.88 1.36 1.33 Procyanidin B2 gallate x 22.28 c 23.10 b 25.20 a 32.36 a 26.33 c 27.39 b s 0.97 0.85 1.04 0.98 1.44 1.65 Epicatechin gallate x 3.74 b 4.30 a 3.31 c 5.73 b 4.55 c 7.12 a s 0.24 0.88 0.37 0.54 0.74 0.67 Gallocatechin x 5.21 a 3.40 c 4.30 b 5.62 a 2.48 c 5.04 b s 0.24 0.23 0.27 0.38 0.55 0.29 Epigallocatechin x 4.33 ab 2.41 b 4.73 a 3.99 a 3.08 b 3.48 ab s 0.37 0.46 0.63 0.74 0.74 0.74 Epigallocatechin gallate x 1.35 c 3.30 a 2.37 b 1.24 c 4.45 a 4.01 b s 0.08 0.08 0.04 0.04 0.03 0.04 Trimer T2 x 6.53 c 6.95 b 8.64 a 7.67 b 8.44 a 8.22 ab s 0.72 0.83 6.78 0.34 0.46 0.48 Trimer C1 x 7.33 a 5.39 b 7.11 a 8.17 b 6.70 c 9.28 a s 0.77 0.71 0.84 0.66 0.53 0.79 x, mean of three replicates; s, standard deviation. Mean values followed by the same letter in a row are not significantly different at the 0.05 level of significance. Biogenic amines composition table 5 shows the concentrations of biogen- ic amines in Aglianico wines. the average con- centration of total amines at racking differs slightly between thesis submitted at different management of MLF, ranging from 9.73 mg/L in sIM wine to 10.38 mg/L in control wine. Af- 318 Ital. J. Food Sci., vol. 27 - 2015 tion in almost all the amines investigated with respect to the control (spontaneous MLF). After 12 months from racking, the average total con- tent of biogenic amines was lower in the wine underwent sequential inoculation compared to the co-inoculation. AckNOWLeDGMeNts the authors thank Apulia region for the financial support in the regional Development Program 2007/2013, Axis I Improvement of competitiveness in agricultural and forestry sectors, Integrated Projects of the Production chain - Meas- ure 124. Authors also thank Le.Vin.sud company for grapes and for support in vinification. reFereNces Abrahamse c.e. and bartowsky e.J. 2012. timing of malo- lactic fermentation inoculation in shiraz grape must and wine: influence on chemical composition. World J. Micro- biol. biotechnol (2012) 28:255-265. bodmer s., Imark c. and kneubuhl M. 1999. biogenic amines in foods: histamine and food processing. Inflamm. res. 48:296-300. burns  t.r., and Osborne J.P. 2013. Impact of Malolactic Fermentation on the color and color stability of Pinot noir and Merlot Wine. Am. J. enol. Vitic. 64(3):370-377. cabrita M.J., torres M., Palma V., Alves e., Patão r. and costa Freitas A.M. 2008. Impact of malolactic fermenta- tion on low molecular weight phenolic compounds. ta- lanta 74:1281-1286. cane P. 1990. Il controllo della qualità dei vini mediante HPLc: Determinazione degli acidi organici. L’enotecni- co, 26(1-3):69-72. capozzi V., russo P., Ladero V., Fernández M., Fiocco F., Alvarez M., Grieco F. and spano G. 2012. biogenicami- nes degradation by Lactobacillus plantarum: toward a potential application in wine. Frontiers in Microbiology - Food Microbiology. April 2012, Volume 3, Article 122, 6 cavazza A., Poznanski e., chiodini A. and zini c. 2003. bat- teri per la fermentazione Malolattica: una selezione da vini trentini. terra trentina, 24-28. cerutti G., Margheri G. and bongini F. 1987. sulla correlazi- one tra fermentazione malolattica ed ammine biogene vas- oattive. Vigne vini n. 14(11):39-42. costantini A., Garcìa Moruno e. and Moreno-Arribas M.V. 2009. biochemical transformations produced by malol- actic fermentation. In: Wine chemistry and biochemis- try. eds. M.V. Moreno-Arribas and M.c. Polo (springer sience business Media: New york), pp. 25-57. costello P.J., Francis I.L. and bartowsky e.J. 2012. Varia- tions in the effect of malolactic fermentation on the chem- ical and sensory properties of cabernet sauvignon wine: interactive influences of Oenococcusoeni strain and wine matrix composition. Aust. J. Grape Wine res. 18:287-301. coton, M., romano, A., spano, G., ziegler, k., Vetrana, c., Desmarais, c., Lonvaud-Funel, A., Lucas P. and coton e., 2010. Occurrence of biogenic amine-forming lactic acid bacteria in wine and cider. Food Microbiol. 27, 1078-1085. Gazzetta Ufficiale ce L 272 del 03/10/1990 (eec regula- tion 2676/90). Davis c.r.D., Wibowo D., eschenbruch r., Lee t.H. and Fleet G.H. 1985. Practical implications of malolactic fermenta- tion: A review. Am. J. enol. Vitic. 36:290-301. Jussier D., Morneau A.D. and De Ordunal r. M. 2006. ef- fect of simultaneous Inoculation with yeast and bacte- ria on Fermentation kinetics and key Wine Parameters of cool-climate chardonnay. Applied and enviromental Microbiology, p. 221-227 Vol. 72, No. 1. Dicks L.M.t., Dellaglio F. and collins M.D. 1995. Proposal to ter 12 months of storage, when wines had al- ready reached a chemical/physical and biolog- ical stabilization, it was observed an increase in the bA content. the differences found be- tween the different winemaking protocols were significant. the average content of total biogen- ic amines in the wine obtained from spontane- ous MLF (18.38 mg/L) was higher than co-in- oculation (14.72 mg/L) and sequential inocula- tion (15.01 mg/L) wines. In percentage the in- crease was 77.0% in control, 51.3% in sIM and 45.3% in PAF wine. this higher percentage in the control wine could be ascribed to a release of amino acids as a consequence of yeast lysis during AF and to the proliferation of LAbs with carboxylase activity during spontaneous MLF. Putrescine was the most abundant amine, with control showing the highest amount at racking (3.74 mg/L before MFL) and also 12 months af- ter racking (10.51 mg/L). Histamine, which is thought to be the cause of various adverse reac- tions to wines (tAyLOr et al., 1989; WANtke et al., 1996), after 12 months increased its concen- tration in the wine produced with spontaneous MLF (control) wine and vice versa decreased in sIM and PAF wines. Indeed, after 12 months his- tamine was almost absent in wines subjected to inoculation of LAb (0.24 mg/L in sIM and 0.20 mg/L in PAF), while the control wine showed an higher value (3.53 mg/L) which was anyway very similar to values reported by other authors (IzqUIerDO-cAñAs et al., 2008; PrAMAteF- tAkI et al., 2006). Probably the histidine decar- boxylase activity was present and active in wine produced with spontaneous MLF for the whole period of aging, instead was absent or not active in wines subjected to inoculation of LAb. cONcLUsION the simultaneous inoculation with yeast and bacteria (sIM) has reduced the duration of MLF of about 33 days compared to the wine obtained without the addition of any LAb (control). the simultaneous inoculation already at the end of the AF showed evident signs of the onset of MLF. After 12 months from racking, there was a weakening of differences in phenolic compounds content, but the wine underwent a spontaneous MLF (control) remained more colourful and more rich in high molecular weight tannins. concern- ing the content of monomeric catechins and ol- igomeric procyanidins it was observed the pre- dominance of (+)-catechin among the monomer- ic flavanols and the procyanidin b2 among the dimeric procyanidins. these compounds were lower in the co-inoculation wine compared to the other two wines. the different MLF manage- ments led to a different evolution in the content of biogenic amines. the simultaneous (sIM) and the sequential (PAF) inoculation of lactic acid bacteria for the MLF led to a significant reduc- Ital. J. Food Sci., vol. 27 - 2015 319 reclassify Leuconostoc oenos as Oenococcus oeni [corrig.] gen. nov., comb. nov.. Int. J. syst. bacteriol. 45:395-397. Di stefano r., cravero M.c. and Gentilini N. 1989. Metodi per lo studio dei polifenoli dei vini. L’enotecnico, 25(5):83-89. Di stefano r., Ummarino I. and Gentilini N. 1997. Alcuni aspetti del controllo di qualità nel campo enologico. Lo stato di combinazione degli antociani. Ann. Istit. sperim. enol. Asti: 105-121. Gafner J. 2005. stabilità biologica e amine biogene. Vinidea. net, rivista Internet tecnica del Vino. N. 2/1. Izquierdo-cañas P.M., García-romero e., Gómez-Alonso s., González M.F. and Palop-Herreros M.L. 2008. Amino ac- ids and biogenic amines during spontaneous malolactic fermentation in tempranillo red wines. J. Food compos. Anal. 21(8):731-735. Lafon-Lafourcade s., carre e. and ribéreau-Gayon P. 1983. Occurrence of lactic acid bacteria during the different stages of vinification and conservation of wines. Appl. environ. Microbiol. 46:874-880. Landete J.M., Ferrer s., Polo L. and Pardo I. 2005. biogen- ic amines in wines from three spanish regions. J.  Ag- ric. Food Chem. 53:1119-1124. Leitao M., Marques A. P. and san romao M. V. 2005. A sur- vey of biogenic amines in commercial Portuguese wines. Food control 16:199-204. Lonvaud-Funel A. 1999. Lactic acid bacteria in the quali- ty improvement and depreciation of wine. Anton, Leeuw, 67:317-331. Lonvaud-Funel A. 2001. biogenic amines in wines: role of lactic acid bacteria. FeMs Microbiol. Lett. 199:9-13. Lòpez r., Lòpez-Alfaro I., Gutièrrez A. r., tenorio c., cari- jo P., Gonzàles-Arenzana L. and santamarìa P. 2011. Malolactic fermentation of tempranillo wine: contribu- tion of the lactic acid bacteria inoculation to sensory quality and chemical composition. Int. J. Food science tech. 46:166-174. Masqué M.c., romero s.V., rico s., elórduy X., Puig A., cap- devila F., suárez c., Heras J.M. and Palacios A.t. 2008. coinoculo di lieviti e batteri: effetti sulla qualità e sulle amine biogene. www.infowine.com, rivista internet di vit- icoltura ed enologia, N. 8/2. Nicolini G., Larcher r., and bertoldi D. 2003. Indagine sul tenore di ammine libere in mosti d’uve di varietà autoc- tone. riv.Viti. enol. 1:15-27. Pramateftaki P.V., Metafa M., kallithraka s. and Lanaridis P. 2006. evolution of malolactic bacteria and biogenic amines during spontaneous malolactic fermentations in a Greek winery. Lett. in Appl. Microbiol. 43:155-160. ricardo da silva J.M., bourzeix M., cheynie V. and Mout- ounet M. 1991. Procyanidin composition of chardon- nay, Mauzac and Grenache blanc grapes. Vitis 30:245- 252. silla-santos M.H. 1996. biogenic amines: their importance in foods. Int. J. Food Microbiol. 29:213-231. squadrito M., corona O., Ansaldi G. and Di stefano r. 2007. relazione fra i percorsi biosintetici degli HctA, dei fla- vonoli e degli antociani nella buccia dell’uva. riv. Vit. enol. 3:59-70. sun b.s., Pinto t., Leandro M.c., ricardo-Da-silva J.M. and spranger I. 1999. transfer of catechins and proanthocy- anidins from solid parts of the grape cluster into wine. Am. J. enol. Vitic. 50(2):179-183. suriano s., ceci G. and tamborra P. 2012. Impact of differ- ent winemaking techniques on polyphenolic compounds of Nero di troia wine. Italian Food & beverage technol- ogy 70:5-15. stockley c.s. 2004. can histamine in wine cause adverse re- actions for consume. Australian and New zealand Grape- grower and Winemaker 485(77):79-82. taylor s.L., stratton J.e. and Nordlee J.A. 1989. Histamine Poisoning (scombroid Fish Poisoning): An Allergy-Like In- toxication. clinical Intoxication. 27(4-5):225-240. Versari A., Parpinello G.P. and cattaneo M. 1999. Leuconos- toc Oenos and malolactic fermentation in wine: a review. J. Ind. Microbiol. biotechnol. 23:447-455. Wantke F., Hemmer W., Haglmüller t., Götz M.,·and Jari- sch r. 1996. Histamine in Wine. Int. Arch. Allergy Im- munol. 110:397-400. zùniga M., Pardo I. and Ferrer s. 1993. An improved medi- um for distinguishing between homofermentative and het- erofermentative lactic acid bacteria. International Jour- nal of Food Microbiology, 18: 37-42. Paper Received July 11,2014, Accepted September 23, 2014