Chitooligosaccharide as A Possible Replacement for Sulfur Dioxide in Winemaking

Featured Application: Chitooligosaccharide (COS) can be easily obtained by hydrolyzing chitosan oligomers produced from crustacean shell waste. COS possesses a wide range of biological activities, including antibacterial, antifungal, antiviral, antitumor, antioxidant, immunoregulatory, blood pressure control, and hypocholesterolemic e ﬀ ects. In contrast, SO 2 might cause health problems, such as allergic reactions. Furthermore, excessive use of COS doesnot negatively a ﬀ ect wine quality. Therefore, this study focuses on investigating the antimicrobial activity of COS during winemaking in comparison with the action of SO 2 . An economical amount (500 mg / L) of COS used as an additive during winemaking shows a comparable antimicrobial e ﬀ ect to 100 mg / L SO 2 and does not a ﬀ ect the fermentation process. Therefore, COS is a potential replacement for SO 2 in winemaking. Abstract: Sulfur dioxide (SO 2 ) has been used for centuries as a preservative in winemaking. However, the addition of SO 2 is associated with allergic reactions and can negatively a ﬀ ect wine quality. In our work, chitooligosaccharide (COS) was applied as an alternative to SO 2 in winemaking, and its antimicrobial activity during winemaking was investigated in comparison with the action of SO 2 . The optimal concentration of COS was identiﬁed as 500 mg / L. The antimicrobial e ﬀ ect of COS was evaluated using known and our own separated wine spoilage organisms. The antimicrobial e ﬀ ect of 500 mg / L COS was found to be comparable with that of 100 mg / L SO 2 . Furthermore, using 500 mg / L COS as an additive during winemaking did notinﬂuence the cell growth of Saccharomyces cerevisiae . Therefore, COS can be used as an additive in winemaking. wine spoilage organisms in comparison with the action of SO 2 . This is the ﬁrst report of COS being applied as an alternative to SO 2 in winemaking. The results show that COS is a potential replacement for SO 2 in winemaking.


Introduction
Sulfur dioxide (SO 2 ) has been used in winemaking for centuries as a versatile and efficient additive owing to its antimicrobial and antioxidant properties. SO 2 is often added to a grape crusher prior to fermentation to control unwanted microorganisms and polyphenol oxidase activity during winemaking, and it is added to wine after malolactic fermentation and prior to bottling as a preservative agent. The total amount of SO 2 added throughout the winemaking process can vary considerably, but it is normally between 160 and 250 mg/L (ppm). A sufficient concentration of SO 2 should be supplied to ensure wine has adequate stability against excessive oxidation or microbial development, which can compromise wine quality. Du and Pretorius [1] confirmed that prohibiting SO 2 as an antimicrobial agent without a suitable alternative would increase the risk of wine spoiled by yeasts and bacteria. Nevertheless, owing to potential related health problems, the use of SO 2 inwine has

Must Preparation
To simulate winemaking, a must was prepared on a small scale using a glass fermentation tank rather than a stainless-steel fermentation tank. Three kilograms of Kyoho grapes purchased from a local market were used for must preparation. Grapes were destemmed, washed, squeezed with a household juicer and crushed. The squeezing time was set at only 2 s to ensure that the seeds were not broken. The resulting juice was filtered using a sterile canvas to eliminate solid impurities that remained in the juice. A filtrate was added into a sterile glass fermentation tank and pasteurized by boiling in a 72 • C waterbath for 30 min. After pasteurization, the resulting sample was cooled down by ice-water bath right away to circumvent the aging of flavor compounds. Finally, the samples were retained at 4 • C until used after being divided into several groups (200 mL each).

Isolation of Wine Spoilage Organisms and Culture Conditions
As microorganisms existing on the grape skin are more likely to resist ethanol and survive in rotten grapes compared with microorganism contaminants from the environment, fresh grapes were not washed and exposed to air to allow for decay. Once the fresh grapes exposed to air began to rot and the smell of ethanol was detected, they were picked out to prepare a must. The contaminated must was transferred to a fermentation flask and fermented at 25 • C for 10 days with shaking every 12 h.
After fermentation, bacteria were isolated by inoculating the fermentation broth in a beef extract-peptone agar medium with 0.0125% filter-sterilized nystatin (pH: 7.0-7.2). Yeasts were isolated by inoculating the fermentation broth on a yeast extract peptone dextrose (YEPD) agar medium with 32,000 U/L filter-sterilized penicillin (pH: 6.0). Fungi were isolated by inoculating the fermentation broth on the Czapek-Dox agar medium with 40,000 U/L streptomycin (pH: 7.2). A separated single colony was submitted to Comate Bioscience Company, Ltd. (Nanjing, China) for further strain identification by 16S rRNA gene sequencing or 18S rRNA gene sequencing.

Vinification
Food-grade COS was provided as a gift by Dalian GlycoBio Company, Ltd. (China). The COS average molecular weight was less than 2000 Da, and the degree of polymerization was in a range of 2-10. Before adding to a pasteurized must, various concentrations of SO 2 (50, 100, and 200 mg/L) in the form of potassium metabisulfite or those of COS (0, 100, 500, and 1000 mg/L) were filtered through a 0.22µm filter membrane. A must without any additive was used as a blank group. All samples were then inoculated with 0.25 g/L Saccharomyces cerevisiae (Angel, Yichang, China)and 2 mL/L wine spoilage organism (specific type or mixture). During fermentation, the temperature was kept at 25 • C, and the samples were shaken every 12 h for 10 days.

Colony-Counting Assay
After fermentation, appropriate dilutions of bacterial/fungal suspension were prepared. Each suspension (100 µL) was uniformly spread onto overnight-dried beef extract-peptone agar plates or Czapek-Dox agar plates with a sterile spatula. After incubation at 30 • C for 24 h, all colonies were enumerated, and the mean values and the maximal scatter in CFUs were determined.

Application of Known Spoilage Microorganisms to Vinification
Pediococcus damnosus, Acetobacter aceti, Lactobacillus plantarum, and Lactobacillus brevis, which are known to spoil musts and wines, were selected for further application. Pediococcus damnosus and Acetobacter aceti were purchased from Biobw (Beijing, China). Lactobacillus plantarum and Lactobacillus brevis were stored in our laboratory. A pasteurized must supplemented with SO 2 or COS was inoculated with Saccharomyces cerevisiae (0.25 g/L) and one known wine spoilage organism (2 mL/L).
During fermentation, the temperature was kept at 25 • C, and the samples were shaken every 12 h for 10 days.
As the spoilage abilities of the known spoilage microorganisms were different, a colony-counting assay was conducted every 24 h for 5 days. At each time point, the appropriately diluted bacterial suspension was uniformly spread onto overnight-dried Lactobacilli MRS broth agar plates or Acetobacter medium agar plates containing natamycin (0.13 g/L) with a sterile spatula.

Growing Curves of Saccharomyces Cerevisiae and PichiaPastoris
Pichiapastoris (stored in our laboratory) is a type of wild yeast that can cause wine spoilage. Therefore, the effects of COS on the cell growths of Pichiapastoris and Saccharomyces cerevisiae were investigated using growth curves. A pasteurized must supplemented with 500 mg/L COS was inoculated with 0.2 g/L Saccharomyces cerevisiae or 2 mL/L Pichiapastoris. A pasteurized must without added COS, but inoculated with the same amount of Saccharomyces cerevisiae or Pichiapastoris, was set as a blank group. During fermentation, the temperature was kept at 25 • C, and the samples were shaken at 100 rpm for 18 h. The bacterial concentration was determined every 2 h using the optical density (OD) value at 600 nm (OD 600 ).

Statistical Analysis
All colony-counting assays were performed in triplicate. Analysis of means and standard deviations were tested using SPSS software (version 17.0; IBM Corp., New York, USA).

Results and Discussion
As shown in Figure 1, wine spoilage organisms were first isolated from rotten grapes using a beef extract-peptone medium or Czapek-Dox medium. The supplemented concentration of COS was then optimized by investigating the antimicrobial effects of various concentrations of COS (0, 100, 500, and 1000 mg/L) on each type of wine spoilage organism. Furthermore, the antimicrobial effects of the optimized concentrations of COS and SO 2 were compared. Finally, the antimicrobial effects of COS on the known wine spoilage microorganisms were estimated, and the effects of COS on the cell growths of Saccharomyces cerevisiae and Pichiapastoris, a type of wild yeast that can cause winespoilage, were evaluated.

Growing Curves of Saccharomyces Cerevisiae and PichiaPastoris
Pichiapastoris (stored in our laboratory) is a type of wild yeast that can cause wine spoilage. Therefore, the effects of COS on the cell growths of Pichiapastoris and Saccharomyces cerevisiaewere investigated using growth curves. A pasteurized must supplemented with 500 mg/L COS was inoculated with 0.2 g/L Saccharomyces cerevisiae or 2 mL/L Pichiapastoris. A pasteurized must without added COS, but inoculated with the same amount of Saccharomyces cerevisiae or Pichiapastoris, was set as a blank group. During fermentation, the temperature was kept at 25 °C, and the samples were shaken at 100 rpm for 18 h. The bacterial concentration was determined every 2 h using the optical density (OD) value at 600 nm (OD600).

Statistical Analysis
All colony-counting assays were performed in triplicate. Analysis of means and standard deviations were tested using SPSS software (version 17.0; IBM Corp., New York, USA).

Results and Discussion
As shown in Figure 1, wine spoilage organisms were first isolated from rotten grapes using a beef extract-peptone medium or Czapek-Dox medium. The supplemented concentration of COS was then optimized by investigating the antimicrobial effects of various concentrations of COS (0, 100, 500, and 1000 mg/L) on each type of wine spoilage organism. Furthermore, the antimicrobial effects of the optimized concentrations of COS and SO2 were compared. Finally, the antimicrobial effects of COS on the known wine spoilage microorganisms were estimated, and the effects of COS on the cell growths of Saccharomyces cerevisiae andPichiapastoris, a type of wild yeast that can cause winespoilage, were evaluated.

Optimization of Supplemented COS Concentration
Nine types of wine spoilage organism were isolated from the contaminated must, comprising six types of bacteria (Enterococcus, Bacillus cereus, Paenibacillus, Alkaliphilus, Pantoea sp., and Lactococcus) and three types of fungi (Aspergillus flavus, Aureobasidium pullulans, and Penicillium). Various concentrations of COS

Optimization of Supplemented COS Concentration
Nine types of wine spoilage organism were isolated from the contaminated must, comprising six types of bacteria (Enterococcus, Bacillus cereus, Paenibacillus, Alkaliphilus, Pantoea sp., and Lactococcus) and three types of fungi (Aspergillus flavus, Aureobasidium pullulans, and Penicillium). Various concentrations of COS (0, 100, 500, and 1000 mg/L) were added into the beef extract-peptone medium or Czapek-Dox medium without agar and then inoculated with a specific type of wine spoilage organism to investigate the antimicrobial effect. The antimicrobial effect of COS was evaluated by measuring the colony counts of the nine types of spoilage organism when applying COS at 100, 500, and 1000 mg/L to the must. Many previous reports have described the antibacterial and antifungal activities of COS against many types of bacteria and fungi [28][29][30][31]. COS can alter the cell membrane permeability, which protects the release of cell components and controls the entry of materials into the cell from the environment. Positively charged COS can then bind to and be absorbed into microbe cell walls, causing DNA penetration and blocking RNA transcription and finally resulting in microbial cell death [30]. However, the antimicrobial activities of COS change depending on various factors, such as the degree of polymerization and microorganism type. Therefore, to efficiently exert the antibacterial activity of COS during winemaking, the COS concentration supplemented into the must should be optimized.
As shown in Table 1, when 100 mg/L COS was applied to the must, most wine spoilage organisms were inhibited to a certain extent, and the colony countswere decreased, which were ina range of 12%-95% of those without COS. The inhibition rate was calculated according to the following formula: inhibition rate = (colony count (with COS)-colony count (without COS))/colony count (without COS). Enterococcus and Lactococcus were only inhibited by 21% and 12%, respectively. When the COS concentration was increased to 500 mg/L, the inhibition rates of bacterium or fungus reached almost 100%. When the COS concentration was increased to 1000 mg/L, the colony count was further decreased by one or two orders of magnitude, but the inhibition rate was not markedly changed. By considering the economic benefits and the antimicrobial effect, 500 mg/L COS was selected as the optimized concentration. Values are expressed as mean ± SD.

Comparative Analysis of COS and SO 2 Antimicrobial Effects
SO 2 is a traditional food additive that has been permitted for use in wine. The total limited concentration of SO 2 in wine can vary, such as 200 mg/L or 300 mg/L, because laws restricting SO 2 use in wine vary among different countries. However, winemakers tend to supplement wine with as little SO 2 as possible owing to negative health effects on consumers with special sensitivity. To simulate the real environment of contamination during winemaking, nine types of wine spoilage organism were mixed together and inoculated into a pasteurized must. Free SO 2 , rather than its bound form, can inhibit microbes, but the amount of free SO 2 is closely related to the temperature and pH of a must. Therefore, the fermentation temperature was kept at 25 • C, and the pH values of all musts before fermentation were determined to be around 3.5. Finally, SO 2 (up to 200 mg/L) was added to investigate its antimicrobial effect. Surviving colonies were only reduced by 15% when supplemented with 50 mg/L SO 2 compared to that without additives. Therefore, 50 mg/L SO 2 was clearly insufficient for large-scale antimicrobial action. Increasing the SO 2 concentration to 100 or 200 mg/L significantly reduced the amount of surviving spoilage organisms, which finally reached less than 0.01%. The amount of surviving colonies with asupplementation of 500 mg/L COS was close to that achieved by adding 100 mg/L SO 2 ( Table 2). Table 2. Comparative analysis of antimicrobial effects on the mixture of different wine spoilage organisms among using 500 mg/L COS and 50, 100, and 200 mg/L SO 2 as additives.

Additive
Concentration ( Values are expressed as mean ± SD. Furthermore, comparative analysis of the antimicrobial effects of COS (500 mg/L) and SO 2 (100 mg/L) as individual additives on nine different types of wine spoilage organism was conducted. As shown in Table 3, 500 mg/L COS showed better antibacterial effects than 100 mg/L SO 2 on all the nine wine spoilage organisms, except Enterococcus and Alkaliphilus. For three types of fungi, COSshowed greater antifungal effects than SO 2 . This might be due to COS originating from chitin, which composes the fungal cell wall. Among the six types of bacteria, COS showed greater antibacterial effects on Bacillus cereus, Paenibacillus, Pantoea sp., and Lactococcus compared with SO 2 , while SO 2 showed greater antibacterial effects on Enterococcus and Alkaliphilus. This implied that antimicrobial effects of COS or SO 2 varied among different microorganisms, which might be due to physicochemical characteristics of each microorganism type, including hydrophilicity and negative charge distributions on the cell surface [32]. Overall, 500 mg/L COS showed a greater antifungal effect than 100 mg/L SO 2 , but both of the additives had a comparable antibacterial effect. Table 3. Comparative analysis of antimicrobial effects on nine different wine spoilage organisms between using 500 mg/L COS and 100 mg/L SO 2 as additives. Values are expressed as mean ± SD.

Estimating Antimicrobial Effectsof COS on Known Wine Spoilage Microorganisms
As COS showed greater antimicrobial effects than SO 2 on our own separated wine spoilage microorganisms, including six types of bacteria and three types of fungi, its antimicrobial effects onknown wine spoilage microorganisms were also evaluated. Among known wine spoilage microorganisms [33], Pediococcus damnosus, Acetobacter aceti, Lactobacillus plantarum, and Lactobacillus brevis were selected for evaluation. As shown in Table 4, the colony counts of spoilage microorganisms grew during the first few days and then dropped rapidly owing to the increasing ethanol content. Neither SO 2 nor COS completely killed the microbes, but an inhibited microbe growth was observed. As ethanol resistance was different among these four wine spoilage microorganisms, their survival times during fermentation were different. The growth curve of each type of bacteria with 500 mg/L COS was always slightly below those with 100 mg/L SO 2 and far below those with no additive, as determined by the colony count from each time point (Table 4). Therefore, 500 mg/L COS showed stronger antibacterial effects comparable to 100 mg/L SO 2 . Table 4. The colony counts of four known wine spoilage organisms at each time point (24,48,72,96, and 120h) after using 500 mg/L COS or 100 mg/L SO 2 as an additive.

Effects of COS on Saccharomyces Cerevisiae and PichiaPastoris Cell Growths
To investigate the effects of COS on the cell growths of Saccharomyces cerevisiae and Pichiapastoris, a pasteurized must with 500 mg/L COS as an additive was inoculated with only Saccharomyces cerevisiae or Pichiapastoris, without any wine spoilage organism. A pasteurized must without any additive and inoculated with only Saccharomyces cerevisiae or Pichiapastoris was used as a control group. The bacterial concentration was evaluated using the OD 600 value. The logarithmic phase of the cell growth of Saccharomyces cerevisiae was slightly extended, when 500 mg/L COS was added, because bacterial growth was slightly slowed down. When the stationary phase was reached, the bacterial concentrations (OD 600 values) with or without COS were not significantly different. However, the growth of Pichiapastoris, a type of wild yeast that can cause wine spoilage [1], was significantly inhibited by 500 mg/L COS ( Figure 2). Therefore, 500 mg/L COS supplemented into the must inhibited Pichiapastoris cell growth but did not influence Saccharomyces cerevisiae cell growth. Therefore, using 500 mg/L COS as an additive during winemaking would not affect the fermentation process. As summarized in Table 5, compared with previously reported multiple additions and physical methods, COS showed a more universal antimicrobial effect without any adverse effects. Using 500 mg/L COS as an additive during winemaking did not influence Saccharomyces cerevisiae cell growth, meaning that COS did not influence the fermentation process. Furthermore, the radical scavenging activity of wine could be increased by adding COS, as confirmed by analysis of the DPPH and hydroxyl radical scavenging activities (data not shown). Therefore, COS shows potential for applications in the winemaking industry. As summarized in Table 5, compared with previously reported multiple additions and physical methods, COS showed a more universal antimicrobial effect without any adverse effects. Using 500 mg/L COS as an additive during winemaking did not influence Saccharomyces cerevisiae cell growth, meaning that COS did not influence the fermentation process. Furthermore, the radical scavenging activity of wine could be increased by adding COS, as confirmed by analysis of the DPPH and hydroxyl radical scavenging activities (data not shown). Therefore, COS shows potential for applications in the winemaking industry.

Conclusions
COS with an optimal concentration of 500 mg/L exerted antimicrobial activity against nine of our own separated wine spoilage organisms and five types of known wine spoilage organism. The antimicrobial effect of 500 mg/L COS was comparable to that of 100 mg/L SO 2 . Furthermore, using 500 mg/L COS as an additive during winemaking did notinfluence Saccharomyces cerevisiae cell growth. Therefore, COS is a potential replacement for SO 2 in winemaking.