PRODUCTION OF FRUIT WINES USING NOVEL ENZYME PREPARATIONS

Methods and results : This paper covers the main characteristics of novel multi-enzyme complexes and the results of in-lab fruit-wine production with addition of enzymatic preparations, which could be used on an industrial scale. The juice yield and the content of suspended materials in the enzymatically treated samples were compared. Experiments included the sensory analysis of produced juices and fruit wines.


INTRODUCTION
This paper addresses an important technological problem of the fruit-wine industry (Codex Alimentarius : wine made from fruit other than grapes) : how to increase juice yield from raw material without compromising the quality of the final product.Fruit-wine consumption is significantly lower compared to traditional grape wines; however, in countries such as Great Britain, Poland or Russia, this type of beverage is well known and appreciated (Noller and Wilson, 2009 ;Kiselev et al., 2013).A vast range of raw materials can be used for fruit-wine production : apple, pear, pineapple, guava, kiwi (Soufleros et al., 2001), Chinese lychee, orange, cherry, cranberry, mango, passion fruit, papaya, peach, etc. Selection of the raw material is mainly determined by traditional recipes existing in the country of origin.
The fruit-wine technology is characterized by the specificity of the raw materials, which vary in their chemical content and requirements for different processing conditions.The production of such types of wine is often confronted with numerous problems such as low juice yield, difficulties with pressing, slow juice clarification, clouding and color changes in the final product (Volchok et al., 2013).
Currently, preprocessing of fruits and berries with various enzymes prior to pressing and filtration is considered to be the most effective solution to these technological problems (Jayani et al., 2005 ;Liew Abdullah et al., 2007), ensuring better fiber maceration and juice clarification, prevention of colloidal hazes, and achievement of balanced and diverse flavors (Ageeva and Markosov, 2013).Selection of enzymes is based on their activities required for a particular fruit or berry.
The Enzyme Biotechnology Laboratory of the Bach Institute for Biochemistry, Russ.Acad.Sci.(INBI RAS) is developing new superior enzymes and enzymatic complexes with several activities at ratios allowing the efficient processing of various raw materials.This article describes the processing of several fruit substrates, containing cellulose and hemicellulose, with the new multi-enzymatic complexes , followed by the lab-scale production of fruit wines.Both enzyme preparations were derived from recombinant strains of Penicillium verruculosum.These complexes were selected in relation to earlier experiments processing fresh viburnum and strawberry juice yield from the pulp and larger content of reducing sugarspreliminary processing experiments.The main criteria were yield increasing of juice and larger content of reducing sugars (Volchok et al., 2013).The results of the organoleptic analysis and the comparisons of enzyme-processed juices and wines against the nonprocessed samples are presented to acknowledge the efficiency of the proposed method and multienzymatic complexes.

Substrates
Ash berry, plum and black currant provided by the Russian State Agrarian University named after K.A. Timiryazev were used as raw fruit and berry material.Table 1 shows the dates of harvest and sampling.

Enzyme preparations
Multi-enzyme preparations were obtained by cotransformation of the auxotrophic host strain P. verruculosum 537.Expression plasmid encoding P. canescens pectin lyase (PELA) and Aspergillus niger β-glucosidase (BG), and transforming plasmid pSTA 10 were used in transformation experiments.Details of the developed process for the recombinant strains and enzyme preparations are described in Bushina et al. (2012).Preparations are in the form of a light brown powder (easily soluble in water) obtained by lyophilization of culture filtrates (micro-filtrated and concentrated by ultrafiltration method) after fermentation of recombinant P. verruculosum strains, and they show stable, high enzymatic activity in the range of 25-50ºС for temperature and 4.0-5.0 for pH.
Enzyme complexes contained pectin lyase A, cellobiohydrolase, endo-1,4-glucanase and βglycosidase.Earlier activity of enzymes was tested on apple, citrus and beet pectin (Morozova et al., 2010;Bushina et al., 2012).Their main enzymatic activities are presented in Table 2. Enzymatic activities towards polysaccharide substrates were determined from the initial rates of formation of reducing sugars by the Somogyi-Nelson method (Nelson, 1944 ;Somogyi, 1952).Activities against p-NP-derived substrates were determined at pH 5.0 and 40ºC by measuring p-nitrophenol release, as described elsewhere (Gusakov et al., 2005).All activities are expressed as international units per mg protein (U/mg) (one unit corresponds to the hydrolysis of 1 μmol of glycoside bonds from the substrate per minute).Methods of determination of enzymatic activities are described in detail in Bushina et al. (2012).

Scheme for producing wines
Fruit-wine materials were produced in the lab using the methodology presented in Table 3.

Analytical methods
During the experiments, juice yield, viscosity and suspension content in fermented samples were compared.

Characteristics of fruit juices
Suspension content of fruit must was evaluated gravimetrically by centrifugation.10-cm3 samples were put in pre-measured sedimentation tubes and centrifuged for 10 min at 3,000 rpm.Supernatant was removed, leaving the tubes upside down for 1 min.Sediment content was calculated by the equation: С= (m 2 -m 1 )*100/V, where m 2 -mass of sedimentation tube with sediment, g; m 1 -mass of empty sedimentation tube, g; and V -sample volume, cm 3 .
For determination of relative viscosity, samples were centrifuged for 10 min at 8,000 rpm.Then 5 cm 3 of liquid was incubated in an Ostwald viscometer for 5 min at 20ºC (Ashapkin et al., 2005).Relative viscosity was calculated by the equation: η=Т i /Т 0 , where Т i -flow time of selected sample, sec; and Т 0 -flow time of water, sec.

Sensory analysis of fruit juices and wines
Ten people were recruited for participating in the sensory analysis of produced juices and fruit wines.
The range of descriptors and reference terms allowing the complete organoleptic description of the juices and wines was selected previously (Baxter et al., 2005).10-scale evaluation maps were developed using the following key terms : "weak", "little" or "absent" for the left anchors and "strong" or "much" for the right anchorskey words.During the week (5 days), panel members participated in training sessions to ensure an homogeneous interpretation of the terms and correct filling of the score cards (Laboissiere et al., 2007).unsalted biscuits were provided for clearing the palate.Spider web plots were made for graphical representation of the tasting session results (Duarte et al., 2010).

Data analyses
Physical characteristics of the juices were measured in triplicate for each parameter.Identified Rrelative standard deviation (RSD %) and confidence interval were identified.
Student t-test capability on Microsoft Excel 2003 was used to determine the significance of the differences in between attributes (Doerffel, 1990).For data processing, a statistical significance level of P = 0.05 was used.

RESULTS AND DISCUSSION
The yield of free-run juice in the course of in-lab production of fruit wine is presented in Figure 1 (data obtained were recalculated for 1 ton of pulp -raw weight).It is important to note that due to high acidity the substrate pulp used for the experiment was diluted with water.The must yield from untreated pulp was used as a control.
It is clearly seen from the figure data that the use of enzymatic preparations significantly increases the yield of high quality free-run juice.
Table 4 shows the results of relative viscosity for the must produced from different raw materials and its sediment content.
The advantages of the enzymatic treatment of raw plant materials compared to non-processed samples are lower viscosity (lower biopolymer contentcellulose, hemicellulose, pectin -due to enzymatic destruction) and lower concentration of sediment in the fermented must.
Thus, enzyme preparation BI_3-227.4was chosen for treatment of black currant and plum due to its βglucosidase and pectin lyase activities leading to the rapid rarefaction bioconversion of pectin substrates.
Another enzymatic preparation, BI_3-227.7,was chosen for the ash berry treatment due to its cellulase and hemicellulose activities.These results were expected as the preparation formula, either BI_3-227.4or BI_3-227.7,correlates to the component composition of the cell wall of these plants.
Table 5 shows the results of the organoleptic analysis of the juices obtained with enzymes and the juices obtained by pressing after maceration.Participants in the sensory analysis especially noted more attractive color characteristics in the case of the enzymeprocessed plum juice compared to the reference sample.
Figure 2 shows significant differences in aroma and appearance characteristics between the enzymeprocessed samples and control.
Juices that have undergone enzymatic treatment are characterized by lower amount of suspended particles and lower turbidity, which facilitates the subsequent clarification and filtration processes.The organoleptic study of the wines produced from analyzed juice samples included the determination of fruity and floral notes in aroma and main flavor parameters (Table 6).
Provided data show that the use of multi-enzyme complexes for fruit-wine production has a positive effect on appearance and aroma characteristics (especially noticeable in the case of plum juice) without affecting other organoleptic attributes.
Figure 4 depicts the results of the sensory analysis of produced fruit wines.In the case of yellow plum, significant positive difference can be observed.The sample produced with the developed enzyme complex is characterized by noticeably lower cooked odor and better clarity compared to the reference.
Comparing flavor parameters of enzyme-processed wine and control showed no significant differences (Figure 5).
-    In earlier research held in the Enzyme Biotechnology Laboratory INBI RAS, the organoleptic parameters of the red wine processed with the help of monitored enzymatic complexes were compared to the wine produced according to traditional technology.Cabernet Sauvignon and the local grape variety Tsimlyansky Black (provided by the « SARKEL » department of the « Tsimlyansky wines » company), a local grape variety, served as raw material for the wine.Results obtained during the tasting session are described in Volchok et al. (2014).These data show that the developed multi-enzymes applied to the grape-wine industry allow to achieve well balanced wines with rich fruity aromas.

CONCLUSIONS
Samples of fruit wines were produced in-lab, including a maceration stage with the use of new enzymatic preparations consisting of the target activities of cellulase, β-glucosidase and pectin lyase.Different ratios of the target activities allowed to apply specific enzymatic agents to a particular type of raw material.The juice yield was higher compared to the reference sample.And produced fruit-wines material was were not inferior to the reference sample in quality attributes, showing lower viscosity, lower sediment content and higher color intensity.Lower sediment content was observed in the must treated by enzymatic preparations.Organoleptic analyses of juices and fruit wines showed a positive effect of multi-enzyme complexes on the sensory characteristics of products.
Obtained data clearly show the high efficiency of the new-generation enzymatic preparations in the fruitwine industry.

Figure 3 -Figure 4 -
Figure 3 -Flavor and consistency attributes for simples of juices

Figure 5 -
Figure 5 -Flavor and consistency attributes for simples of fruit-wines

Table 2 -Characteristics of multi-enzymatic preparations
* -Sum of cellulase activities is equated to 1.

Table 4 -Physical attributes of fruit must Figure 2 -Appearance and aroma attributes for simples of juices ©Vigne
et Vin Publications Internationales(Bordeaux, France)

Table 5 -Fruit juices sensory attributes mean values
paired parameters ash berry juices (** for black currant, 3* for yellow plum) with t-test above the significance level P *

Table 6 -Fruit wines sensory attributes mean values
paired parameters ash berry wines (** for black currant, 3* for yellow plum) with t-test above the significance level P *