Read jfsv64n6ms19990215 text version


Food Chemistry and Toxicology

Milk Protein Coatings Prevent Oxidative Browning of Apples and Potatoes

C. LE TIEN, C. VACHON, M.-A. MATEESCU, AND M. LACROIX ABSTRACT: Color analysis on apple and potato slices coated with calcium caseinate or whey protein solutions showed that the 2 coatings efficiently delayed browning by acting as oxygen barriers. The antioxidant properties of the films were realized using a model allowing the release of oxidative species by electrolysis of saline buffer. Whey proteins were a better antioxidant capacity than calcium caseinate. Furthermore, addition of carboxymethyl cellulose (CMC) to the formulations significantly improved their antioxidative power. Best scavenging of oxygen free radicals and reactive oxygen species was found for films based on whey proteins and CMC which inhibited by 75% the formation of colored compounds produced by the reaction of the oxidative species with N,N-diethyl-p-phenylenediamine. Keywords : antioxidant; oxidative browning; edible coating; whey proteins; caseinate; color analysis.

Food Chemistry and Toxicology

tion on sliced mushrooms significantly reduced enzymatic browning (Nisperos-Carriedo and others 1991). This work reports color measurements, performed by for the food industry. The normal approach to inhibit both enzymatic and nonenzymatic oxidative browning in trivariance analysis on sliced apples and potatoes coated foods has been the application of sulfites. However, health with milk protein formulations, in order to determine effecconcerns have limited their application (Sapers 1993). Other tiveness in postponing enzymatic browning. Furthermore, techniques, including modified atmosphere packaging (MAP) antioxidant properties of films cast from whey and calcium and vacuum packaging have been considered. While this ap- caseinate solutions were tested using the N,N-diethyl-p-pheproach can delay browning, excessive reduction of oxygen nylenediamine (DPD) colorimetric method. will damage the product by inducing anaerobic metabolism, Materials and Methods leading to breakdown and off-flavor formation. Furthermore, the removal of oxygen also entails a risk that anaerobic conditions in the product might become favorable for Materials the growth of Clostridium botulinum (Sapers 1993). Another Calcium caseinate (Alanate 380) was provided by New approach is the use of antibrowning agents based on citric Zealand Milk Products (Santa Rosa, Calif., U.S.A.). Concenacid or ascorbic acid. Although ascorbic acid can reduce en- trated whey protein powder was obtained from Saputo zymatic browning, it can increase nonenzymatic browning Cheese Ltd. (St-Hyacinthe, Quebec, Canada). Glycerol due to the its own oxidation into dehydroascorbic acid (99.5%, reagent grade) from American Chemicals Ltd. (DHAA) which then reacts with amino acids to yield brown (Montreal, Quebec, Canada), carboxymethyl cellulose sodicolors by the Maillard reaction or other nonenzymatic pro- um salt (CMC, low viscosity) and N,N-diethyl-p-phenylenedicesses (Kacem and others 1987). Furthermore, high concen- amine (DPD) from Sigma Chemical Co. (St. Louis, Mo., trations of acid or other chemical agents could significantly U.S.A.) were used as received, without further purification. alter food flavor and odor. Recently, Sapers and others (1997) showed that the use of harsh chemical treatments (heated Coating procedure and film formation acid solutions) can induce severe textural damage in prea) Preparation of coating solution: Coating solutions were peeled potatoes resulting in surface firming (case hardening) prepared with 5% protein (calcium caseinate or whey proand separation of the superficial tissues that affect texture tein powder), 2.5% glycerol, 0.25% CMC, and 0.125% CaCl 2 after mashing and slicing following cooking. Such defects according to a method developed in our laboratory (Ressouwould greatly limit the utilization of pre-peeled potatoes any and others 1998; Brault and others 1997). The compowhich have received the antibrowning treatment. nents were mixed in distilled water to obtain homogeneous Nussinovick and Kampf (1993), Hershko and Nussinovitch solution and heated at 80 C for 30 min. The solutions were (1998) proposed and used alginate-based coating solution to cooled at room temperature (20 1 C) and the final soluconserve the mushroom for a long period with a better ap- tion pH was 6.5. Solutions were prepared immediately before pearance and color. Tong and others (2000) showed that use and the color measurement was realized when the coatwhey proteins are effective antioxidants in salmon oil-in-wa- ing solutions was applied on slices. ter emulsions. The use of milk protein-based coatings, which McIntosh apples (Rougemont, Quebec, Canada) and Rusare flavorless, odorless, and edible, could also be beneficial set potatoes (Canada #1 from Prince Edward Island, Canada) for controlling enzymatic browning of cut fruits and vegeta- were used. Five slices (each about 1 cm thick) were cut from bles without inducing tissue damage. Edible coatings based 3 potatoes and 3 apples, dipped and held for 1 min in the on cellulose gums have already been used to effectively delay coating solutions and laid on a flat surface for drying at ripening in some climacteric fruits like mangoes, papayas, room temperature (20 1 C). Control potatoes and apples and bananas. Furthermore, application of the same formula- were cut and laid without dipping in the dishes and exposed




512 JOURNAL OF FOOD SCIENCE--Vol. 66, No. 4, 2001

© 2001 Institute of Food Technologists

Oxidative Browning of Apples and Potatoes . . .

to atmospheric air. The experiments were repeated 3 times. (b) Preparation of film-forming solution: Films used for measuring antioxidant properties were based on milk protein, glycerol and CMC only. Solutions containing 5% protein (calcium caseinate or whey protein powder), 2.5% glycerol and with or non 0.25% CMC were heated at 80 C during 30 min. Heat was essential for the formation of the intermolecular disulfide bonds, this process is necessary to obtain a flexible film with good mechanical properties (Vachon and others 2000; Le Tien and others 2000). The reason for CMC addition resides in its role as a matrix (protein stabilizing), whereas glycerol can play the role of a plasticizer agent, which prevents breaking film (Ressouany and others 1998). The solutions were cooled at room temperature (20 1 C) and the final solution pH was 6.5. Films were cast by applying 5 mL of the solution onto a 8.5 cm diameter Petri dishes (Fisher Scientific, Montreal, Quebec, Canada) and allowed to dry overnight at room temperature (20 1 C). Transparent dried films were peeled intact from the casting surface and reconditioned in a desiccator containing a saturated NaBr solution, ensuring 56% relative humidity (RH) at room temperature (20 1 C), for at least 48 h (Gontard and others 1992). MgSO4 0.86 mM, CaCl2 1.25 mM, glucose 11.0 mM, and EDTA 0.06 mM). Preliminary results showed that similar oxidative effect can be obtained by electrolysis of 0.15 M NaCl, in the same conditions (10 mA, 400 V for 1 min). The antioxidative capacity describes the film's capacity to inhibit the accumulation of oxidative species (able to oxidize DPD) and the red coloration at 515 nm. The reaction is calibrated using the nonelectrolyzed KH buffer solution (no oxidative species, ascribed to 100% scavenging) and the electrolyzed KH buffer solution (0% scavenging, in the absence of any antioxidants). The scavenging percentage is calculated following this equation: scavenging (%) = 100 ­ [ (OD sample /OD control) 100] (1)

where OD control represents the OD of electrolyzed solution in the absence of film. In fact, OD is directly related to the degree of oxidation of DPD reagent by the oxidative species. Thus a film able to reduce completely the level of reactive oxidative species will have a 100% scavenging capacity.

Results and Discussion

ties which are important for the film formation (hydrogen, hydrophobic and ionic interactions). However, the heating was necessary for the formation of the intermolecular disulfide bonds which improve the mechanical properties of film, particularly, the oxygen barrier (McHugh and Krochta 1994). Addition of CMC in the formulation resides in its role as a matrix, whereas glycerol can play the role of a plasticizing agent (Le Tien and others 2000). All milk protein-based films used in this experiment were transparent after complete well drying.

Color analysis and lightness evaluation

Color measurements were made at 5 min intervals for a total experimental period of 130 min. The color was evaluated by trivariance analysis using a Colormet reflectance spectrocolorimeter (Instrumar Engineering Ltd., St. John's, NF, Canada) using the standard CIELAB (1976) color system. Lightness is reported as L* and the hue angle value is given by tan-1 (b*/a*). The lightness value for perfect white is 100, while L* = 0 corresponds to black. As the hue decreases, red pigmentation increases. The a* axis (red) corresponds to a hue angle of 0°. Color measurements were done on 15 slices (potato or apple, n = 15).



Coating and the oxidative browning

The variation of the lightness parameter (L*) as a function of time for coated potato slices is presented in Figure 1. For Antioxidative capacities of film were evaluated following a the uncoated (control) slices, an increase in lightness was obmodified procedure of the DPD (N,N-diethyl-p-phenylenedi- served for the 1st 15 min. Then, L* value of the uncoated poamine) colorimetric method (APHA 1989), as reported by tato slices starts to progressively decrease with time for the remaining experimental period. Over the entire duration of the Dumoulin and others (1996). (a) Film thickness measurement: Film thickness was measured using a Mitutoyo Digimatic Indicator (Mitutoyo, Tokyo, Japan) at 5 random positions around the film. The average film thickness was in the range of 50 5 m and depended upon the formulation. (b) Antioxidative capacity determination: Films were cut in pieces of equal thickness (50 4 m), all measuring 0.8 2.5 cm (approximately 200 mg protein). They are then put in a cell containing 3 mL of Krebs-Henseleit (KH) buffer and submitted to electrolysis for 1 min (continuous current, 400 Volts; 10 mA) using a generator (Bio-Rad, model 1000/500). After electrolysis, a volume of 200 L of solution is sampled and added to 2 mL of DPD solution (25 mg/mL). The oxidative species react instantly with DPD producing a red coloration that can be measured at 515 nm. The colorimetric reaction was calibrated with potassium permanganate (KMnO 4) solution. The oxidative capacity of 1.00 mg/L of free chlorine solution (generating hypochlorous acid) corresponds to that of 5.63 mol/L KMnO4 solution. Use of electrolysis as a method to generate oxidative stress was first introduced by Jackson and others (1986) for physiological studies on perfused isolated organs. Oxidative Figure 1--Time course of the lightness parameter (L*) for damage was realized by electrolysis of KH buffer (NaCl 118.0 uncoated (control) and coated (calcium caseinate or whey 1.22) mM, NaHCO3 25.4 mM, KCl 4.8 mM, KH 2PO4 1.2 mM, protein) potato slices (n = 14; SD

Evaluation of antioxidant properties

Vol. 66, No. 4, 2001--JOURNAL OF FOOD SCIENCE


Food Chemistry and Toxicology

Oxidative Browning of Apples and Potatoes . . .

experiment (130 min), the L* of the control slices varied from 70.64 (at t = 0) to 66.15 at (t = 130 min). The loss of lightness can be estimated at 4.5% for the experimental period (130 min). Under the same conditions, the L* values observed for coated potato slices did not show any evidence of loss of lightness. A slight increase in lightness was even noticed for all types of coated potato slices between the 1st 45 min and 70 min respectively for calcium caseinate and whey proteins. After 130 min, the lightness of the calcium caseinate-coated slices was L* = 71.39 and a similar value (L* = 71.72) was noted for the whey-coated slices. The increase of L* value in coated and uncoated samples is probably due to the exudation of natural liquid present in potato or the coating solution that contribute to increase L* value. Afterward, L* value remained stable until the end of the experiment. Figure 2 shows the variation of hue angle for uncoated and coated potato slices. As the hue angle decreases, red pigmentation becomes more pronounced. The control (uncoated) slices undergo rapid appearance of red pigmentation as seen by the sharp decrease of the hue over the first 45 min. Then, the hue was stable for the remainder of the experimental period. For the coated slices (calcium caseinate and whey proteins), only a slight variation of the hue was noticed for the entire period of 130 min (Figure 1). Figure 3 and 4 show the lightness (L*) and hue results obtained for apple slices. Similar to the changes observed for potato slices, L* rapidly decreased with time for the uncoated apple slices (Figure 3). After 130 min, the average L* of the uncoated apples was 66.11 compared to 74.77 at t = 0. This represents an overall lightness loss of more than 8% for the entire period. For the coated apple slices, L* remained rather constant showing that the protein coatings effectively protected the fruit from oxygen. As for the hue (Figure 4), results show that, for all apple slices, the angle decreased slightly with time. The hue decrease was faster for uncoated slices and calcium caseinate-treated slices. Best prevention was found with the whey coating. Although the hue decreased, our data (Figure 3) show that moderate color fluctuations were not associated with darkening. Overall results clearly suggest that proteinbased edible coatings were successful in delaying oxidative browning in sliced apples and potatoes. Nisperos-Carriedo and others (1991) reported color measurements done on sliced mushrooms coated with a formulation containing vegetable oils, cellulose gums, emulsifiers, surfactants, and fatty acids. They showed that the coating reduced enzymatic browning. After 2 h, the coated mushrooms were lighter than the uncoated ones. Still, the coating did not completely inhibit darkening as the coated mushrooms were slightly darker after 2 h than the fresh-cut controls. Milk protein-based formulations appeared more effective in controlling oxidative browning of sliced potatoes and apples since color fluctuations were not enhanced by a lower L* (Figure 1 and 3), even after 2 h. A previous report showed that the addition of whey powder improved the oxidative stability of soybean oil (Browdy and Harris 1997). Protein coatings probably delay browning by preventing the oxidative process. An important factor implied in the inhibition of the oxidative browning is that the coating by protein solution represents an efficient barrier to oxygen (decreased oxygen penetration). However, this barrier is not total. Previous studies (McHugh and Krochta 1994) demonstrated that these coatings are not completely impervious to oxygen. Indeed, the oxygen permeability of milk protein-based films were varied between 19 and 43 cm 3.ìm/ m2.d.kPa (25 C). This feature would, consequently, lower the risks of creating undesirable anaerobic conditions and retarded oxidative browning. Other agents could also inhibit enzymatic browning by different mechanisms. Many proteins can exert certain antioxidative effects. The presence of several side residues of amino acids, in particular, the cysteine in the milk proteins can directly or indirectly inhibit the polyphenol oxidases. Dudley and Hotchkiss (1989) showed that cysteine inhibits polyphenol oxidase via its SH groups, acting as an agent coupling quinones and forming stable, colorless compounds. The same phenomenon has been observed by Berlett and others 1997. Kohen and others (1988) demonstrated that the histidine residues and its derivatives (having an imidazole compound) possess an antioxidant activity that was due to hydrogen donation. Indeed, the hydrogen on the ring nitrogen and on the methylene carbon next to imidazol ring are likely donors. Furthermore, the enzyme was also supposed to be competitively inhibited by the pres-

Food Chemistry and Toxicology

Figure 2--Time course of the hue for uncoated (control) and coated (calcium caseinate or whey protein) potato slices (n = 15; SD 1.58)

Figure 3--Time course of the lightness parameter (L*) for uncoated (control) and coated (calcium caseinate or whey protein) apple slices (n = 14; SD 1.42)

514 JOURNAL OF FOOD SCIENCE--Vol. 66, No. 4, 2001

Oxidative Browning of Apples and Potatoes . . .

ence of aromatic residues such as tyrosine, phenylalanine, and tryptophan (Berlett and others 1997). Probably milk protein coatings exert prevention through both effects as barrier limiting the access of oxygen and via a possible inhibitory action at the level of polyphenol oxidases located in the tissues cells. Additional effects could be related to the presence of other components in the film formulation. For instance CMC, as carbohydrate, can exert a nonspecific oxidative species scavenging activity (Wehmeier and Mooradian 1994). Another possible effect can be related to the carboxylic groups of CMC which, in certain conditions as a chelating agent, can interact with the copper binding site for the oxygen and decrease polyphenol oxidases activity (Sapers 1993). Hershko and Nussinovitch (1998) showed that alginate-based coating solution, a carbohydrate having a carboxylic group, allowed to conserve the mushroom for a long period with a better appearance and color. In addition, our results indicated that the best prevention was found with the whey coating. This could be due to the presence of fatty acids (5.4% in the whey protein) that significantly improved the moisture barrier properties and significantly reduced browning in fresh-cut apples (McHugh and Senesi 2000). In order to further evaluate the antioxidant capacities of these films, measurements were carried out using a model allowing the release of reactive oxidative species by electrolysis of saline buffer.

Antioxidant properties of protein films

The antioxidative power of protein films is presented in Figure 5. Milk protein films containing CMC had better antioxidative capacities than those based only on protein (caseinate or whey) and glycerol. The antioxidant capacities of milk proteinglycerol films were 37.63% for the calcium caseinate and 60.21% for whey protein films. When CMC was added to the film formulations, the antioxidant capacities increased to 66.14 % for calcium caseinate (Alanate) and to 75.17% for whey. It can be seen that for both types of formulations, whey proteins exhibit a higher antioxidative power than calcium caseinate. Furthermore, the addition of CMC increased the antioxidative power of these films by 43% for the calcium caseinate formulations and by 20% for the whey protein formulations. Many factors could account for the antioxidative properties of milk protein films. In addition to cysteine, aromatic amino acids like tyrosine, tryptophan and histidine are potent free radical targets (Berlett and Stadtman 1997). Furthermore, Colbert and Decker (1991) reported that the antioxidative activity of acid whey was related to a low-molecular-weight fraction (500-5000 Da). This observation is possibly related to smaller peptides. Several peptides such as carnosine and anserine have shown a great antioxidative effect (Kohen and others 1988). The higher antioxidative potential of whey proteins compared to calcium caseinate could be due to the presence of lactose. Indeed, the commercial whey protein used contained 14% lactose, which is known for its free radical scavenging effects (Wehmeier and Mooradian 1994). In addition, Xue and others (1998) showed that alginate and carboxymethyl chitosan retarded the hydroperoxide accumulation of methyl linoleate by effectively trapping peroxide radicals. Similarly, the addition of a polysaccharide like CMC can probably also increase the antioxidative activity of these formulations.

Figure 4--Time course of the hue for uncoated (control) and coated (calcium caseinate or whey protein) apple slices (n = 15; SD 1.29)


edible coatings on the browning reaction of sliced apples and potatoes. Results confirm that the formulations were effective in delaying browning reactions by acting as oxygen barriers and/or reactive oxidative species scavengers. Furthermore, an electrolysis model generating oxidative species was used to assess the antioxidative potential of these films. Whey was shown to be a better antioxidant than calcium caseinate. Such differences can be explained by the variations in amino acid composition in the 2 types of protein. Furthermore, lactose, which is present in commercial whey, can also account for the increased antioxidative activity; simple sugars are known for their free radical quenching effect. Similarly, the addition of a polysaccharide, like CMC, can further increase the antioxidative potential of our film formulations. The use of milk proteins as natural antioxidants is a promising development in related fields of food science. The proteins are flavorless, odorless, and edible, and could be beneficial for controlling enzymatic browning of cut fruits and vegetables without inducing tissue damage. Milk proteins represent interesting agents to be used in some processed foods to prevent the formation of hydroperoxides and lipid peroxidation.

Vol. 66, No. 4, 2001--JOURNAL OF FOOD SCIENCE



Figure 5--Antioxidant capacity (%) for film formulations based on calcium caseinate or whey proteins with or without CMC.


Food Chemistry and Toxicology

Oxidative Browning of Apples and Potatoes . . .


APHA. 1989. DPD colorimetric method. Standard method for the examination of water and wastewater. 17 th edition, Washington DC: American Public Health Association. 4-(62-64). Berlett BS, Stadtman ER. 1997 Protein oxidation in aging, disease and oxidative stress. J Biol Chem 272 (33): 20313-20316. Brault D, D'Aprano G, Lacroix M. 1997. Formation of free-standing sterilized edible films from irradiated caseinates. J Agric Food Chem 45 (8): 2964-2969. Browdy AA, Harris ND. 1997 Whey improves oxidative stability of soybean oil. J Food Sci 62 (2): 348-376. CIELAB. 1976. Color space adapted by CIE in 1974 with recommendations for associated psychometric color terms at the International Commissions on Illuminations (CIE, 1974). Method of measuring and specifying color rendering properties of light sources, Publication CIE N° 13-2 (TC.3.2) Paris: Bureau central de la CIE, 1976. Colbert LB, Decker EA. 1991. Antioxidant activity of an ultrafiltration permeate from acid whey. J Food Sci 56 (5): 1248-1250. Dudley ED, Hotchkiss JH. 1989. Cysteine as an inhibitor of polyphenol oxidase. J Food Biochem 13 (1): 65-75. Duckwork HW, Coleman JE. 1970. Physicochemical and kinetic properties of mushroom tyrosinase. J Biol Chem 245 (7): 1613-1625. Dumoulin M J, Chahine R, Atanasiu R, Nadeau R, Mateescu MA. 1996. Comparative antioxidant and cardioprotective effects of ceruloplasmin, superoxide dismutase and albumin. Arzneim.-Forsch./Drug Res 46 (9): 855-861. Gontard, N, Guilbert, S, CuQ JL. 1992. Edible Wheat gluten films: influence of the main process variables on film properties using response surface methodology. J Food Sci 57 (1): 190-199. Hershko V, Nussinovitch A. 1998. Relationships between hydrocolloid coating and mushroom structure. J Agric Food Chem 46 (8): 2988-2997 Jackson CV, Mickelson JK, Stringer K, Rao PS, Lucchesi BR. 1986. Electrolysis ­ induced myocardial dysfunction. A novel method for the study of free radical mediated tissue injury. J Pharmacol Methods 15 (4): 305-320. Kacem B, Cornell JA, Marshall M.R, Shireman RB, Matthews RF. 1987. Nonenzymatic browning in aseptically packaged orange drinks: Effect of ascorbic acid, amino acids and oxygen. J Food Sci 52 (6): 1668-1672. Kohen R, Yamamoto Y, Cundry KC, Ames BN. 1988. Antioxidant activity of carnosine, homocarnosine and anserine present in muscle and brain. Proc Natl Acad Sci 85 (9): 3175-3179. Le Tien C, Letendre M, Ispas-Szabo P, Mateescu MA, Delmas-Patterson G, Yu HL, Lacroix M. 2000. Development of biodegradable films from whey proteins by cross-linking and entrapment in cellulose. J Agric Food Chem 48 (11): 55665575. McHugh TH, Krochta JM. 1994. Milk-protein-based edible films and coatings. Food Technol 48 (1): 97-103. McHugh TH, Senesi E. 2000. Apple Wraps: A novel method to improve the quality and extend the shelf life of fresh-cut apples. J Food Sci 65 (3): 480-485. Nisperos-Carriedo MO, Baldwin EA, Shaw PE. 1991. Development of an edible coating for extending postharvest life of selected fruits and vegetables. Proc Florida State Hort Soc, No. 104, 122-125. Nussinovitch A, Kampf N. 1993. Shelf life extension and conserved texture of alginated coated mushrooms (Agaricus bisporus). J Food Technol 26: 469-475. Ressouany M, Vachon C, Lacroix M. 1998. Irradiation dose and calcium effect of the chemical properties of cross-linked caseinate films. J Agric Food Chem 46 (4): 1618-1623. Sapers GM. 1993. Browning of foods: Control by sulfites, antioxidants and other means. Food Technol 47 (10): 75-84. Sapers GM, Cooke PH, Heidel AE, Martin ST, Miller RL. 1997. Structural changes related to texture of pre-peeled potatoes. J Food Sci 62 (4): 797-803. Tong L M, Sasakis S, Mc Clements DJ and Decker EA. 2000. Antioxydant activity of whey in a salmon oil emulsion. J Food Sci 65 (8): 1325-1329. Vachon C, Yu H L, Yefsah R, Alain R, St-Gelais D and Lacroix M. 2000. Mechanical and structural properties of milk protein edible films cross-linked by heating and irradiation. J Agric Food Chem 48 (8): 559-566. Wehmeier K.R, Mooradian A.D. 1994. Autoxidative and antioxidative potential of simple carbohydrates. Free Radic Biol Med 17 (1): 83-86. Xue C, YU G, Hirata T, Terao J, LIN H. 1998. Antioxidative activities of several marine polysaccharides evaluated in a phosphatidylcholine-liposomal suspension and organic solvents. Biosci Biotechnol Biochem 62 (2): 206-209 MS 20000228

This work was funded by the FCAR-Novalait and CQVB program. A NSERC graduate studentship granted to C. Le Tien and an Institut Armand-Frappier postdoctoral fellowship granted to C. Vachon are gratefully acknowledged. The authors also thank Michèle Jobin for her collaboration.

Food Chemistry and Toxicology

Authors Le Tein, Vachon, and Lacroix are associated with INRS-Institut Armand Frappier, Laval, Canada, and author Mateescu is associated with the University of Quebec. Please address correspondence to Author Lacroix (E-mail: [email protected])

516 JOURNAL OF FOOD SCIENCE--Vol. 66, No. 4, 2001



5 pages

Find more like this

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

Microsoft Word - White Winemaking Paper - FINAL