g E, f ; , E: g {:1}: f, i . ; . , A ,f . , , "53324 {EN THE EGDY (”1.5. . ‘ ' gaiKuJ‘K a I kl“- (‘l ' \n - ”‘7‘; PM. Chm f‘dl "Q Ffi‘J-CPR {I}: I 6%8 1| _ 19m I 5‘ . '1 A #1 .. fl; ‘3 -.. f" ”49ng a? 3 :l‘ s €62? {he 0 a I $593 ‘33:? QZ‘GEVERSETY up if uHifix‘AN S 9-H“ In ,‘3 A? (an d LIBRARY Michigan State University ABSTRACT EFFECT OF LACTOSE FORTIFICATION ON THE BODY AND FLAVOR 0F CULTURED BUTTERMILK by Ronald D. Marshall Cultured buttermilk was prepared from raw skimmilk containing several concentrations of added lactose. In addition, some of the sam- ples were fortified with sodium citrate or non—fat dry milk. The skim- milk was heated in flowing steam for 30 min., cooled to 70772°F., inoculated with one per cent of an active starter culture and incubated for varying lengths of time at 72°F. Analyses were performed on each sample for pH, titratable acidity, biacetyl content, and biacetyl plus acetoin content. Some samples were subjected to gas chromatographic analysis. The buttermilks were scored by a panel of 3 competent judges to obtain a relative indication of consumer preference. The addition of lactose resulted in a decrease in the biacetyl con- tent of the cultured buttermilk. In samples fortified with sodium citrate, lactose fortification resulted in an increase in the total biacetyl- acetoin content. There was an improvement in body and texture of the lactose fortified samples as compared to controls or samples fortified with non-fat dry milk. Lactose fortification at a two per cent level was found to be optimum for the production of a desirable product. Limited gas chromatographic analyses, under the conditions described in this re- search, detected no change in the volatile constituents of cultured buttermilk as a result of lactose fortification. EFFECT OF LACTOSE FORTIFICATION ON THE BODY AND FLAVOR.OF CULTURED BUTTERMILK By RONALD D. MARSHALL A THESIS Submitted to the College of Agriculture Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER.OF SCIENCE Department of Food Science 1961 ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to Dr. C. M. Stine for his guidance and counsel in directing this research. My deep gratitude is extended to Dr. J. R. Brunner for his advice and encouragement throughout this study. Grateful acknowledgment is due to the Western Condensing Company of Appleton, Wisconsin, for providing funds to make this study possible. ii TABLE OF CONTENTS INTROD mTION O O O O O O O O O O O O 0 REVIEW OF LITERATURE . . . . . . . . . Butter starter cultures . . . . . Body and texture of cultured buttermilk Flavor and aroma of cultured buttermilk Precursors and formation of biacetyl and acetoin Breakdown and interconversion of biacetyl and acetoin EXPERIMENTAL PROCEDURE . . . . . . . . Sources of cultures . . . . . . . Preparation of cultured buttermilk ANALYTICAL PROCEDURES . . . . . . . . Biacetyl determinations . . . . . Acetoin-biacetyl determinations . Acidity measurements . . . . . . Gas chromatography . . . . . . . EXPERIMENTAL RESULTS . . . . . . . . . Analytical determinations . . . . Consumer preference . . . . . . . Gas chromatography . . . . . . . DISCUSSION . . . . . . . . . . . . . . Analytical determinations . . . . Consumer preference . . . . . . . Gas chromatography . . . . . . . SUMMARY AND CONCLUSIONS . . . . . . . LITERATURE CITED . . . . . . . . . . . iii Page 10 10 10 10 10 11 11 ll 14 14 22 23 24 24 27 28 29 30 II. III. IV. TABLES Page Effect of lactose fortification, varying incubation time, and 48 hours of storage at 4°C. on the characteristics of cultured buttermilk . . . . . . . . . . . . . . . . . . . . . 15 Effect of lactose fortification, with and without addition of 0.2 per cent sodium citrate, and varying incubation periods on the characteristics of cultured buttermilk . . . . . . . . 17 Effect of lactose fortification, with and without addition of 0.2 per cent sodium citrate, and varying incubation periods on the characteristics of cultured buttermilk . . . . . . . . 20 Comparison of fortification with non-fat dry milk (NFDM) and lactose on the characteristics of cultured buttermilk incubated 14 hours . . . . . . . . O O O O O O O O O O C O O O O O O O 2]- iv FIGURES Page Scheme for the production of pyruvic acid from lactose and Citric aCid O O O O O O O O O O O O O O O O O O O O O O O O O O 6 Scheme for the formation and interconversion of biacetyl, acetoin, and 2,3-butylene glycol . . . . . . . . . . . . . . . 7 INTRODUCTION Cultured buttermilk is a dairy product prepared by inoculating true buttermilk or, more commonly, skimmilk with a starter culture that pro- duces a desirable flavor and aroma. The resulting product contains little or no fat but is a good source of protein made readily digestible by the bacterial fermentation. Although cultured buttermilk possesses the basic attributes to be an important food product, lack of a uniformly desirable body and flavor have precluded widespread consumer acceptance. Research studies have provided methods for improving the quality of cultured buttermilk, but continued improvement is necessary to make cultured buttermilk a more saleable product. This study was made to determine the effect of lactose fortification on the body and flavor of cultured buttermilk. REVIEW OF LITERATURE Butter Starter Cultures A butter starter culture contains two distinct types of microorgan- isms. §, lactis and S, cremoris, commonly referred to as the lactic streptococci, produce most of the lactic acid in the manufacture of cultured buttermilk. The formation of volatile flavor and aroma sub- stances is due to the action of L, citrovorum, L, dextranicum, or S. diacetilactis. The growth of the flavor and aroma producing types, usually referred to as citrate fermenters or associate organisms, is dependent upon the acidic medium provided by the metabolic processes' of the lactic acid streptococci. Body and Texture of Cultured Buttermilk Hales (13) reported that most consumers prefer a cultured butter- milk that is not too viscous, but has a smooth consistency with no whey separation. According to Hammer and Babel (15), a desirable texture should be smooth and creamy, without lumpiness; the body should be of such character that it will stand up well when poured on a surface and have a velvety appearance. Foster g; 31. (10) suggested that the body of a good culture should resemble a firm custard. Flavor and Aroma of Cultured Buttermilk The flavor of cultured buttermilk should be delicate and pleasing (15). Foster'ggflgl. (10), proposed that such a flavor was associated with a clean acid taste and the aroma of biacetyl in a dilute solution of acetic and propionic acids. Many substances have been implicated in the desired flavor of cul- tured buttermilk. Hammer and Babel (15) classified the flavor and aroma compounds as lactic acid, volatile acids, carbon dioxide, and the 4-carbon neutral molecules (biacetyl, acetoin and 2,3-butylene glycol). vTraces of alcohols, aldehydes, and esters have also been prOposed as contributing to the flavor and aroma of cultured buttermilk (10). In a gas chroma- tographic study of starter distillate, Jennings (18) reported the identi- fication of ethyl alcohol, biacetyl, acetic acid and methyl acetate. The importance of acetoin and biacetyl to the flavor and aroma in lactic cultures was shown by Michaelian ggugl. (26), who found that butter cultures with a desirable flavor and aroma contained relatively large amounts of these compounds while cultures lacking in flavor were quite low in acetoin and biacetyl. Pien.g£hg1. (29) questioned the importance of biacetyl and acetoin as flavor components of butter cultures by observing that there was no close correlation between the content of biacetyl plus acetoin and a pleasing flavor and aroma. Agreement with this concept led Hoecker and Hammer (16) to suggest the contribution of compounds other than biacetyl and acetoin to the desirable flavor and aroma of butter cultures. However, data from most of the literature supports the belief that biacetyl and acetoin are the main contributants to the desired flavor and aroma char- acteristics of lactic cultures (1, 2, 3, 10, 14, 15, 24, 28). Precursors and Formation of Biacetyl and Acetoin Although the importance of biacetyl plus acetoin to the flavor of cultures has been generally accepted, the precursor(s) and mechanism of its formation in milk have not been clearly established (1). The early literature adamantly maintained that citrate, metabolized by the asso- ciate organisms, was the sole source of biacetyl and acetoin in cultured buttermilk (14). According to van Beynum and Pette (2), acetoin and biacetyl resulted from a condensation of two molecules of acetaldehyde in an acidic medium. They postulated that the acetaldehyde was formed from citrate with pyruvate serving as an intermediate in the reaction. Virtanen (33) believed that biacetyl and acetoin were derived from the lactose of milk and that citrate was utilized as a hydrogen acceptor, since it could be replaced by methylene blue or quinone. This theory was accepted by Cappens (4). Krenn and Valik (23) proposed that acetoin resulted from pyruvate formed in the dissimilation of lactose and that citrate acted as a hydrogen acceptor. They admitted that a small frac- tion of the acetoin could arise from citrate. Pette agreed that acetoin was formed from pyruvate; however, he stated that pyruvate originated from citrate by reversal activity of the condensing enzyme in the tricarboxylic acid cycle, with subsequent decar- boxylation of the oxaloacetate formed. Further work on the dissimilation of citrate disclosed that enzymes of A. aerogenes (5, 6) and S. faecalis (ll) cleaved citrate into oxaloacetate and acetate. Because free acetate was formed instead of acetyl Coenzyme A, they believed that the condensing enzyme of the tricarboxylic acid cycle was not involved. Dolin and Gunsalus (9) found different acetoin forming systems for animals, yeasts, and bacteria. The bacterial system had enzymes which catalyze the condensation of two pyruvate molecules to form a molecule of o<-acetolactate, with subsequent decarboxylation of the o<-acetolactate to acetoin. Because of the similarity of the bacterial acetoin forming systems studied, they suggested that the same mechanism is true for all bacteria that form acetoin. These findings were substantiated by Kobay- ashi and Kalnitsky (22) in a study using pyruvate-1-C14. However, they added that the decarboxylation of o<-acetolactate to acetoin took place only in an acidic medium. Alpha-acetolactate was detected as a fermenta- tion intermediate in the formation of acetoin by L, citrovorum (7). Juni (19) resolved the acetoin forming system of A,aerogenes into two fractions. One fraction condensed two molecules of pyruvate into o<9acetolactate via the route of a diphosphothiamine intermediate. The second fraction decarboxylated the o<$acetolactate to form acetoin. In an attempt to conclusively identify the precursor(s) of acetoin and biacetyl formed by the associate organisms of lactic cultures, Ander- sen (1) studied the fermentations of L, citrovorum and §, diacetilactis on uniformly labeled lactose-l-C14, lactose-6-Cl4, and sodium citrate-l, 5-C14. He concluded that both lactose and sodium citrate served as pre- cursors for the formation of acetoin and biacetyl by the associate organisms of lactic cultures and theorized that both microorganisms use the same pathways; a) lactose via the Embden-Meyerhoff glycolytic scheme and b) citrate through cleavage by citridesmolase into acetate and oxa- loacetate, with subsequent decarboxylation of the oxaloacetate into pyruvate. These proposals were based on the observations that lactose- l-Cl4 and 1actose-6-C14 gave essentially the same activity in the acetoin formed, thus ruling out the use of the pentose phosphate pathway. The failure of malonate to inhibit the acetoin formation from citrate sug- gested that the enzymes of the tricarboxylic acid cycle were absent. Andersen maintained that the controversial reports on the precursor(s) of acetoin and biacetyl stems from the fact that the breakdown of lactose Figure 1. Scheme for the conversion of lactose and citric acid to pyruvic acid Lactose lactase Glucose Galactose 8313010531359 ——>—- Galactose-l- HIOSphate hexolcmase waldenase Glucose-6- —-<—— phOSPhoglucomutase“ Glucose-l- PhOSphate BIOSphate phOSphO hexo- isomerase ATP ADP Fructose-é- Fructose-l-é- aldolase Glyceraldehyde- PhOSphate 4—4- DiphOSphate —— ->— 3-Ph03phate DPN'l" q phOSphate H+DPNH 4 3-Phospho- ATP-phOSphog lyceric 1, 3-Dipho Spho- Glyceric Acid tranSphOSphorylase Glyceric Acid phOSphoglycero- mutase 2-Phospho- enolase —————)-— PhOSphoenolpyruvic Glyceric Acid Acid Pyruvic ATP-phOSphOpyruvic . —<—— ACid tranSphOSphorylase oxaloacetic decarboxylase Citric citrides- Oxaloacetic Acetic Acid_ "101356 + Acid + Acid 6 Figure 2. Scheme for the formation and interconversion of biacetyl, acetoin, and 2,3-butylene glycol 2 Pyruvic Acid DPT OK-Acetolactic Acid cx-acetolactic decarboxylase ‘r—Acetoin ‘W DPNH +H DPN+ butylene glycol biacetyl reductase dehydrogenase DPN+ DPNH + H 2,3-Buty1ene Glycol Biacetyl H20 DPT Acetylbutanediol Acetyl-DPT DPN+ biacetylmethylcarbinol ‘\.Biacetyl reductase DPNH+H m Biacetylmethylcarbinol_;4, to pyruvate, via the Embden - Meyerhoff scheme, resulted in a concomitant production of the reduced form of diphosphopyridine nucleotide (DPNH). Whereas, the formation of pyruvate from citrate occurs without the pro- duction of DPNH. Breakdown and Interconversion of Biacetyl and Acetoin Biacetyl, acetoin, and 2,3-butylene glycol are related through oxi- dation-reduction mechanisms. This relationship has practical aspects regarding the flavor and aroma of cultured buttermilk. The amount of oxidized or reduced substances in the medium determines the proportion of these compounds. The presence of oxygen or highly oxidized substances . favor the formation of biacetyl, the important flavor constituent. H0w- ever, in the presence of a strongly reducing potential, the non-volatile 2,3-buty1ene glycol predominates. Lactic streptococci exhibit a strongly reducing potential, thus favoring the formation of 2,3-butylene glycol, rather than biacetyl, from acetoin. An early report by van Beynum and Pette (2) suggested that biacetyl was reduced by a hydrogen donor formed as an intermediate of lactose dissimilation. Evidence that the hydrogen donor produced in the dissimilation of carbohydrates was the reduced form of diphOSphopyridine nucleotide (DPNH) was suggested by reactions of the Embden - Meyerhoff glycolytic scheme. DeMoss‘ggflgl. (8) found that the enzyme catalyzing the interconver- sion of acetoin and 2,3-buty1ene glycol was a DPN linked dehydrogenase. Support of this concept was presented by Strecker and Harary (31), who named the enzyme butylene glycol dehydrogenase. In addition, they reported the existence of another diphOSphopyridine nucleotide linked dehydrogenase which catalyzed the irreversible reduction of biacetyl to acetoin. A different type of biacetyl breakdown was suggested by Green 25 31. (12) in a study of a diphOSphothiamine (DPT) dependent enzyme which cata- lyzed the conversion of two molecules of biacetyl into two molecules of acetic acid and one molecule of acetoin. They called this enzyme, bia- cetyl mutase. Recent studies of Juni and Heym (20, 21) have revealed a new pathway for the reduction of biacetyl and acetoin to 2,3-butylene glycol, which proceeds through the intermediate compounds, diacetyLmethylcarbinol and acetylbutanediol, and is dependent on the presence of diphosphothimmine and diphOSphopyridine nucleotide. These mechanisms would serve to explain the increase in 2,3-buty1ene glycol content which parallels the decrease in biacetyl and acetoin con- tent (3), and causes a deterioration in the flavor of cultured dairy products. EXPERIMENTAL PROCEDURE Sources of Cultures Two different commercial starter cultures were employed in this study. They were supplied as lyOphilized cultures from the manufacturer and are designated in this report as Culture I and II. Preparation of cultured Buttermilk Raw skimmilk, obtained from the Michigan State University Dairy Plant was combined with the appropriate amount of added substance(s) in a liter Erlenmeyer flask to a combined weight of 594 gms. The samples were covered with a metal foil cap and heated in flowing steam for 30 min. At the end of this time the samples were rapidly cooled to 70-72°F., inoculated with 6 ml. of active starter culture and incubated for varying lengths of time at 72°F. Samples which were not analyzed immediately after incubation were cooled and stored, without agitation, in a water bath at 4°C. ANALYTICAL PROCEDURES Biacetyl Determinations Biacetyl was determined spectrophotometrically using essentially the method described by Prill and Hammer (30). I A 25 gm. sample of cultured buttermilk was weighed into a 500 ml. distillation flask. The flask was connected to the distillation appara- tus and flooded with carbon dioxide to prevent the oxidation of acetoin to biacetyl by atmospheric oxygen. Steam was introduced slowly below the surface of the sample and 5.0-5.2 m1. of distillate was collected in 25- 30 min. The collection of the distillate in 1 m1. of hydroxylamine acetate 10 11 solution facilitated the immediate conversion of biacetyl to dimethylgly- oxime. The concentration of biacetyl, measured as ammono-ferrous dimethyl- glyoxime in a Model 6A Coleman Spectrophotometer at 530 mu (24), was obtained by referring to a standard curve prepared for the Spectrophoto- meter used in this study. Acetoin plus Biacetyl Determinations For the acetoin plus biacetyl determinations, 15 ml. of a 40 per cent ferric chloride solution was added to the 25 gm. sample of cultured buttermilk. The mixture was refluxed for 10 min. to facilitate the oxi- dation of acetoin to biacetyl. Because of the high concentration of biacetyl, 10.0-10.5 m1. of distillate was collected in 2 m1. of hydroxy- lamine acetate solution in 25-30 min. The development and measurement of color were the same as in the biacetyl determinations. Acidity Measurements Titratable acidity was determined by titrating a nine gram sample of cultured buttermilk with 0.1 N sodium hydroxide to the phenolphthalein endpoint. The results were expressed to the nearest one hundredth of one per cent as lactic acid. The pH measurements were made with a Beckman Zeromatic pH meter, using a calomel half-cell and a glass electrode, standardized to read accurately in the range of pH 4.2-5.0. The results were expressed to the nearest one tenth of a pH unit. Gas Chromatography Samples subjected to gas-liquid partition chromatography were ob- tained by a recycling collection method similar to that described by Nawar and Fagerson (27). 12 A 4000 gm. portion of cultured buttermilk was weighed into a 5 1. round-bottom flask, equipped with a thermometer to indicate the tempera- ture of the cycling vapors. The flask was placed in the recycling system which consists of, in order: the flask of cultured buttermilk,a drying tube packed with anhydrous potassium carbonate, an air cooled collection tube, a cold trap at 0°C., two collection traps in a dry ice—ethanol mix- ture at -60°C., and a pulsating pump (Model OV—20, Sigmamotor, Inc., N. Y.). The cycle was completed by extending the pump outlet to the bottom of the flask containing the cultured buttermilk. A water bath at room temperature was placed around the flask of cultured buttermilk and the pump was started. The water bath was slowly heated so that the temperature of the cycling vapors would be 85°C. at the end of 5 hrs., at which time the collection was terminated. The liquid collected in the four collection tubes was combined and kept at 0°C. Fifteen grams of sodium chloride were added to the sample to salt out the volatile organic compounds. One milliliter of diethyl ether, at a temperature of 0°C., was added to extract the organic compounds. The ether layer was removed in a 50 ml. separatory funnel by means of a sy- ringe and dried over sodium sulfate for 24 hrs. at ca. -20°C. in a stoppered vial. A Perkin-Elmer Model 154-B Vapor Fractometer, with a thermistor-type thermal conductivity detector, was employed for the gas chromatographic analysis. The column and column conditions were those advocated by Jennings (18). The column, 11 ft. of 0.25 inch c0pper tubing packed with 20 per cent polyoxyethylene polypropylene glycol ether (Pluronic F-68, Wyandotte) 13 on 40-60 mesh Celite was heated at a temperature 0f 130°C. The flow of the carrier gas, helium, was regulated at 90 cc./min. A six microliter sample was injected for analysis. EXPERIMENTAL RES ULTS Analytical Determinations The effect of fortification with two per cent lactose at 12, 14 and 16 hours incubation, before and after 48 hours storage at 4°C. using Culture I, is shown by the data presented in Table I. The lactose fortified sample had a lower biacetyl content that the control at all comparable incubation periods, before and after 48 hours storage at 4°C. Both samples possessed the greatest amount of biacetyl after 12 hours incubation and extended incubation resulted in a decrease in the biacetyl content which was much more pronounced with the lactose fortified sample. Storage for 48 hours at 4°C. resulted in a decrease in the biacetyl content of all samples. The lactose fortified sample had more biacetyl plus acetoin than the control after 12 hours incubation. However, at the end of 16 hours incu- bation the control had the greatest content of biacetyl plus acetoin. In the first trial, the biacetyl plus acetoin content after 14 hours in- cubation was greater in the control, but in the second trial the lactose fortified sample had the most biacetyl plus acetoin after 14 hours of in- cubation. The pH values of the control lactose fortified sample were essent- ially the same, with the exception of the 12 and 14 hour incubation results in the first trial where the lactose fortified sample had a slightly lower pH than the control. The lactose fortified samples had a slightly higher titratable acidity than the control after 12 and 14 hours incubation. This difference in titratable acidity is not evident when incubation was increased to 16 hours. 14 Table I. Effect of lactose fortification, varying incubation time, and 48 hours of storage at 4°C. on the characteristics of cultured buttermilk fl L Analytical Determinations Titratableb Biacetyl- Descriptiona Incubation acidity Biacetyl acetoin of sample (hr.) pH (%) (p.p.ggl_ (p.p.m.) TRIAL ONE End of incubation Control 12 4.8 0.68 3.10 186 14 4.7 0.78 1.70 232 16 4.5 0.82 1.49 238 2% Lactose 12 4.7 0.73 1.69 238 14 4.6 0.80 1.25 213 16 4.5 0.83 0.59 219 End of storage Control 12 4.8 0.71 2.66 224 14 4.6 0.76 2.24 252 16 4.5 0.80 1.14 242 2% Lactose 12 4.6 0.73 1.30 234 14 4.5 0.78 0.87 243 16 4.5 0.80 0.80 236 TRIAL TWO End of incubation Control 12 4.7 0.74 2.59 236 14 4.5 0.81 2.31 267 16 4.4 0.87 1.27 267 2% Lactose 12 4.7 0.76 2.52 247 14 4.5 0.84 1.17 271 16 4.4 0.87. 0.89 260 End of storage Control 12 4.7 0.75 2.24 257 14 4.6 0.82 1.33 271 16 4.5 0.85 1.04 258 2% Lactose 12 4.7 0.77 2.21 264 14 4.6 0.82 0.95 271 16 4.5 0.85 0.78 254 Culture I employed Expressed as lactic acid 16 The effect of fortification with one, two and three per cent lactose after 14 and 16 hours incubation, with and without the addition of 0.2 per cent sodium citrate and using Culture I, is shown by the data in Table II. Results of the biacetyl determinations indicate that the addition of lactose, under all conditions employed in this study, decreased the biacetyl content by an amount approximately proportional to the amount of lactose added. The addition of 0.2 per cent sodium citrate increased the biacetyl content of all samples, but this increase was quite small. The biacetyl content of the samples was higher after 14 hours incubation than after 16 hours incubation. The biacetyl plus acetoin content was much greater in the samples with 0.2 per cent added sodium citrate than in comparable samples without the addition of sodium citrate. When 0.2 per cent sodium citrate was added, the lactose fortified samples had a greater content of biacetyl plus acetoin than the controls after 14 and 16 hours incubation. In the samples without added sodium citrate, lactose fortification caused a decrease in the biacetyl plus acetoin content after 16 hours incubation; whereas, the biacetyl plus acetoin content of the lactose fortified sam- ples after 14 hours incubation, was greater in the second trial and only slightly less than the control in the first trial. The effect of added lactose on the pH was negligible, The addition of 0.2 per cent sodium citrate resulted in a slightly higher pH. Lactose fortification of one, two and three per cent showed no significant differ- ence in the titratable acidity after 14 and 16 hours incubation. In comparison, the addition of 0.2 per cent sodium citrate contributed a significant increase in the titratable acidity after both 14 and 16 hours incubation. 17 Table II. Effect of lactose fortification, with and without addition of p 0.2 per cent sodium citrate, at varying incubation periods Analytical determinations Titratablefii Biacetyl+ Descriptiona Incubation acidity Biacetyl acetoin of sample (hr.) pH (%) (p.p.m.), (p.p.m.) TRIAL ONE No sodium citrate added Control 14 4.4 0.89 1.44 273 1% Lactose 14 4.5 0.89 0.95 273 2% Lactose 14 4.4 0.89 0.70 258 3% Lactose 14 4.4 0.89 0.65 252 0.2% sodium citrate added Control 14 4.7 0.92 1.44 463 1% Lactose 14 4.6 0.92 1.20 503 2% Lactose 14 4.6 0.94 1.12 501 3% Lactose 14 4.6 0.94 1.02 520 No sodium citrate added Control 16 4.3 0.93 0.78 289 1% Lactose 16 4.4 0.94 0.69 272 2% Lactose 16 4.3 0.93 0.64 267 3% Lactose 16 4.4 0.93 0.54 256 0.2% sodium citrate added Control 16 4.6 0.95 1.20 489 1% Lactose 16 4.6 0.95 1.11 527 2% Lactose 16 4.6 0.96 1.01 520 3% Lactose 16 4.5 0.96 0.95 526 aCulture I employed bExpressed as lactic acid 18 Table II. Effect of lactose fortification, with and without addition of 0.2 per cent sodium citrate, at varying incubation periods (continued) ===§ : r : Analytical determinations Titratableb Biacetyl+ Descriptiona Incubation acidity Biacetyl acetoin of sample (hr.), pH (%) (p.p.m.) (p.p.m.) TRIAL TWO No sodium citrate added Control 14 4.4 0.84 1.72 256 1% Lactose 14 4.5 0.87 1.59 279 2% Lactose 14 4.4 0.85 1.29 267 3% Lactose 14 4.6 0.86 1.01 275 0.2% sodium citrate added Control 14 4.7 0.93 2.00 468 1% Lactose 14 4.6 0.92 1.70 475 2% Lactose 14 4.7 0.93 1.66 512 3% Lactose 14 4.7 0.94 1.56 508 No sodium citrate added Control 16 4.4 0.88 1.18 272 1% Lactose 16 4.4 0.90 1.11 257 2% Lactose 16 4.4 0.86 1.01 257 3% Lactose 16 4.3 0.87 0.99 254 0.2% sodium citrate added Control 16 4.6 0.94 1.78 468 1% Lactose 16 4.5 0.94 1.63 486 2% Lactose 16 4.5 0.95 1.58 480 3% Lactose 16 4.5 0.93 1.35 473 aCulture I employed bExpressed as lactic acid 19 The data in Table III shows the effect of lactose fortification after 14 and 16 hours of incubation, with and without the addition of 0.2 per cent sodium citrate and using Culture 11. Lactose fortification decreased the biacetyl content after 14 and 16 hours incubation. Following 16 hours incubation, the decrease in bia- cetyl content was approximately proportional to the amount of lactose added. The biacetyl content of the samples after 14 hours incubation was not indicative of the amount of lactose fortification. The two and three per cent lactose fortified samples had a biacetyl content of 2.06 and 2.11 ppm., reSpectively. After 14 hours incubation, without citrate addition, the lactose fortified samples contained less biacetyl plus acetoin than the control, but after 16 hours incubation the biacetyl plus acetoin content was greater in the lactose fortified samples. The one per cent lactose for- tified sample excepted. With the addition of 0.2 per cent sodium citrate, lactose fortification resulted in a significant increase in biacetyl plus acetoin after 14 and 16 hours incubation. The final pH values of the cultures were increased by the addition of sodium citrate, but lactose addition apparently caused no shift in pH. The titratable acidities of the lactose fortified samples without citrate addition were slightly higher than the control after 14 hours incubation, but not after the incubation was extended to 16 hours. Results showing a comparison of the effects of fortification with lactose and non-fat dry milk (NFDM) after 14 hours of incubation with Culture I are reported in Table IV. Fortification with NFDM resulted in higher titratable acidity than either the control or lactose fortified samples. The biacetyl content 20 Table 111. Effect of lactose fortification, with and without addition of 0.2 per cent sodium citrate, at varying incubation periods Analytical determinations Titratableb Biacetyl- Descriptiona Incubation acidity Biacetyl acetoin of sample (hr.) pH (%) (p.p.m.) (Dapqg.) No sodium citrate added Control 14 4 5 0.87 3.12 250 1% Lactose 14 4.6 0.89 3.00 232 2% Lactose 14 4.6 0.90 2.06 226 3% Lactose 14 4.5 0.92 2.11 243 0.2% sodium citrate added Control 14 4.8 0.93 3.60 376 1% Lactose 14 4.7 0.92 2.96 389 2% Lactose 14 4.6 0.93 2.48 410 3% Lactose 14 4.8 0.94 2.25 387 No sodium citrate added Control 16 4.5 0.92 2.78 258 1% Lactose 16 4.5 0.93 2.48 257 2% Lactose 16 4.6 0.91 2.36 264 3% Lactose 16 4.5 0.93 2.12 271 0.2% sodium citrate added Control 16 4.7 0.97 3.68 406 1% Lactose 16 4.7 0.95 3.42 435 2% Lactose 16 4.6 0.97 3.38 456 3% Lactose 16 4.7 0.95 3.28 463 aCulture II employed bExpressed as lactic acid 21 Table IV. Comparison of fortification with non-fat dry milk (NFDM) and lactose on the characteristics of cultured buttenmilk incubated 14 hours Analytical determinations Titratableb Descriptiona acidity Biacetyl Biacetyl+Acetoin Of sample in (701 (ID-Dom.) (popomo) Control 4.5 0.91 1.42 280 1% Lactose 4.6 0.91 1.32 282 2% Lactose 4.8 0.91 1.23 283 1% NFDM 4.6 0.94 1.31 307 2% NFDM 4.7 0.96 1.08 328 1% Lactose + 1% NFDM 4.5 0.94 1.24 306 aCulture I employed bExpressed as lactic acid 22 of the NFDM fortified samples was less than the biacetyl content of the control or the lactose fortified sample. The biacetyl-acetoin content was highest in the NFDM.sample. Consumer Preference A comparison of cultured buttermilk containing zero, one, two and three per cent addition of lactose (data not shown) disclosed that two per cent lactose fortification was most acceptable from the standpoint of flavor, body and texture of the cultured product. Lactose added to the extent of three per cent imparted excessive sweetness, even when the sample was cultured to a titratable acidity above 0.9 per cent. The flavor of the control was generally considered flat and the body was- criticized as being thin and watery. Although the samples fortified with one and two per cent lactose were very similar in flavor, the sample for- tified with two per cent lactose was preferred because of a superior body and texture. Flavor evaluations of the samples listed in Table 11 indicated that samples with two per cent lactose and no added sodium citrate, and the samples with one per cent lactose plus 0.2 per cent sodium citrate were preferred equally. Both samples were considered extremely smooth and flavorful. The samples fortified with 0.2 per cent sodium citrate only, were criticized for a weak body and a "vinegary" taste. The flavor of samples fortified with NFDM and lactose (Table IV) was very similar. However, the body of the samples fortified with NFDM was excessively viscous and texture lacked the smoothness exhibited by the lactose fortified samples. 23 Gas Chromatography The gas chromatographic study failed to disclose any differences between a control and a sample fortified with two per cent lactose. Nevertheless, the presence of six volatile components of cultured butter- 'milk, as compared to four volatile constituents reported by Jennings (18) in a gas chromatographic separation of starter distillate, was de- tected in the limited number of trials made. Two of the volatile sub- stances were tentatively identified as ethyl alcohol and methyl acetate. DISCUSSION Analytical Determinations The addition of citrate and non-fat dry milk (NFDM) to skimmilk has been a common practice for many years in the manufacture of cultured buttermilk. These materials have been used to increase the production of volatile flavor compounds, to increase viscosity, and to improve whey retention prOperties of the buttermilk. Several compounds have been proposed as contributing to the desir- able flavor and aroma of cultured buttermilk (10, 15). Jennings (18) reported the identification of four volatile compounds in starter culture distillate. However, biacetyl has been considered the most important flavor and aroma constituent of cultured buttermilk. Extensive research has not clearly established the precursor(s) or formation mechanisms of biacetyl and acetoin (the compound from which biacetyl arises) by lactic cultures (14). Recent studies by Andersen (1), showing that lactose, as well as citrate, served as a precurSor of biacetyl and acetoin have stim- ulated interest in the use of pure lactose as an additive to skimmilk being utilized for the manufacture of cultured buttermilk. The biacetyl content of lactic cultures is thought to be dependent on the amount of oxidized or reduced substances in the medium. ‘Michaelian 25 £1. (26) found that the biacetyl content of lactic cultures could be significantly increased by bubbling oxygen through the culture and could be substantially decreased by bubbling hydrogen or inert gases through the culture medium. Van Beynum and Pette (2) concluded that the strongly 24 25 reducing medium produced by the lactic streptococci was due to a reduced substance formed in the dissimilation of lactose and was instrumental in the reduction of biacetyl to acetoin and 2,3-buty1ene glycol. The pre- sentation of the Embden-Meyerhoff scheme disclosed that the reduced substance mentioned by van Beynum and Pette (2) was the reduced form of diphOSphOpyridine nucleotide (DPNH). Evidence that DPNH was the compound that caused the reduction of biacetyl to acetoin and 2,3-buty1ene glycol was given by Strecker and Harary (31). Andersen found that conversion of lactose into acetoin or biacetyl involved the concomitant production of DPNH, whereas the formation of biacetyl and acetoin from citrate occurred without the production of DPNH. It seems evident that DPNH has a definite effect on the biacetyl. content of cultured buttermilk, possibly through a decrease in the oxi- dation-reduction potential, which, according to Jenness and Patton (17) determines the amount of biacetyl in cultured buttermilk. An examination of the data in this study shows that the addition of lactose to cultured buttermilk caused a decrease in the biacetyl content that was approximately proportional to the amount of lactose added. How- ever, the samples fortified with lactose had a greater biacetyl plus acetoin content than the control samples after all incubation periods employed, when 0.2 per cent sodium citrate was added. Without the addi- tion of sodium citrate, the lactose fortified samples had a greater amount of biacetyl plus acetoin than the control after 12 hours incubation, but not after 14 or 16 hour incubation periods. A possible explanation of these results could be based on the increase in total biacetyl-acetoin content resulting from added lactose. A concomi- tant increase of the reduced form of diphOSphopyridine nucleotide (DPNH) 26 would tend to lower the oxidation-reduction potential, resulting in a decreased biacetyl content. The large increase observed in total biacetyl-acetoin content ob- tained in cultures with the addition of 0.2 per cent sodium citrate, agrees with the results 3r many workers (2, 14, 15, 16, 25, 26, 29) who concluded that citrate was the source of biacetyl and acetoin. However, these results do not exclude the possibility that citrate is acting as a hydrogen acceptor (4, 23, 33), thereby causing an increase in the oxida- tion-reduction potential. The results of this study showed that the addition of 0.2 per cent sodium citrate caused slight increase in the pH values. This observation concurs with the observations of Michaelian g2 g1. (25) and is presumably due to the buffering effect of sodium citrate. In comparison, the addi- tion of lactose had a negligible effect on the pH values. The data comparing the effects of NFDM and lactose fortification on the production of biacetyl and acetoin were not conclusive. The lac- tose fortified samples showed a higher biacetyl content that the samples fortified with NFDM, but the total biacetyl-acetoin content was signifi- cantly greater in the NFDM fortified samples. The titratable acidity of the NFDM fortified samples was greater than the titratable acidity of the lactose-fortified samples and this may exert an influence on the biacetyl and biacetyl-acetoin content. Additionally, the NFDM might lower the oxidation-reduction potential of the medium by providing more free sulf- hydryl groups during heat treatment of the milk. The two cultures employed in this study showed variation in the amount of biacetyl and acetoin produced and in the time of incubation 27 required for maximum production of these compounds; but the effect of fortification with lactose on the biacetyl and biacetyl-acetoin content I was very similar for both cultures. Consumer Preference The extreme smoothness of the lactose fortified samples could possi- bly be due to the formation of smaller casein micelles. Tumerman.g£”al. (32) reported that lactose exerts a protective influence on colloidal casein and inhibits particle aggregation by a mechanism yet to be elu- cidated. They suggested that lactose may sequester the calcium salts, thereby suppressing their destabilizing effect on the casein particles. The consumer preference studies indicate that there was very little agreement between a desirable flavor and the biacetyl or biacetyl-ace- toin content. Hoecker and Hammer (16) and Pien 25.31, (29) noted a simi- lar lack of agreement between the flavor of butter and the biacetyl or biacetyl-acetoin content. These results suggest the importance of com- pounds other than biacetyl and acetoin as flavor constituents. Whether the addition of lactose had a beneficial effect on the production or integration of these compounds remains to be determined. It is quite possible that the noticeable preference for samples containing lactose was due to a masking of the sour acid taste. It is not possible at this time to state the exact nature of the effects of adding lactose to cultured buttermilk but lactose fortifica- tion, in the pr0per amounts, had a beneficial influence on the body and flavor of the cultured buttermilks prepared in this study. 28 Gas Chromatography The gas chromatographic study could not show any differences in volatile components between a control culture and a two per cent lactose fortified sample, possibly because of an insufficiently sensitive detection system. Conceivably the minor components, not detectable by the chromatographic technique employed here, contribute significantly to the desirable flavor improvement in the lactose fortified samples. SUMMARY AND CONCLUSIONS Lactose fortification of cultured buttermilk, with or without 0.2 per cent sodium citrate, resulted in a decrease in biacetyl content that was approximately proportional to the amount of lactose added. In the samples with 0.2 per cent sodium citrate, lactose fortification increased the total biacetyl-acetoin content. Fortification of the cultures with lactose resulted in an improve- ment in the body and texture as compared to the control cultures or samples fortified with non-fat dry milk. Possibly, this improvement was due to the formation of smaller casein micelles. Fortification with two per cent lactose, in the absence of added sodium citrate, resulted in the best product. When 0.2 per cent sodium citrate was added, a one per cent level of added lactose seemed to be ‘most Optimum. Gas chromatographic studies revealed the presence of six volatile components in cultured buttermilk, two of which were tentatively identi- fied. No differences were detected between a control culture and a sample fortified with two per cent lactose. 29 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) LITERATURE CITED .Andersen, V. B. 1959. Formation of diacetyl and acetoin by certain starter organisms. Milchwissenschaft 14:429-432. Beynum, J. Van, and J. W. Pette. 1939. The decomposition of citric acid by Betacoccus cremoris. Jour. Dairy Res. 19:250-266. Camus, A., P. Laniesse, and J. Burdin. 1951. La formation du diacetyle dans les levains de beurrerie. Le Lait 31;225-233. (As translated by Stine, C. M., and J. J. Ritchie). Coppens, R. 1954. Quelques observations sur la biochimie des ferments d'aroma. Neth. Milk and Dairy Jour. 8:14-18. (In French with English summary). Dagley, S., and E. A. Dawes. 1953. Dissimilation of citric acid by bacterial extracts. Nature (London) 172: 345. Dagley, S., and E. A. Dawes. 1953. Dissimilation of citric acid by extracts of Aerobacter aerogenes. (Abst.) Biochem. Jour. §§:XVI. DeMan, J. C. 1956. Over de wijze waar0p diacetyl in culturenen van Betacoccus cremoris ontstaat. Neth. Milk and Dairy Jour. 19:38-52. (In Dutch with English summary). DeMoss, R. D., B. C. Bard, and I. C. Gunsalus. 1951. The mechanism of the heterolactic fermentation: A new route of ethanol formation. Jour. Bact..§2:499-501. Dolin,'M. 1., and I. C. Gunsalus. 1951. Pyruvic acid metabolism. 11. An acetoin forming enzyme system in Streptococcus faecalis. Jour. Bact. 62:199-214. Foster, E. M., F. E. Nelson, M. C. Speck, R. N. Doetsch, and J. C. Olson, Jr. 1957. Dairpricrobiology. Englewood Cliffs, N. J. Prentice-Hall, Inc. 492 pp. plus xvi. (11) Gillespie, D. C., and I. C. Gunsalus. 1953. An adaptive citric acid desmolase in Streptococcus faecalis. (Abst.) Bact. Proc. 53:80. 30 (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) 31 Green, D. E., P. K. Stumpf, and K. Zarudnaya. 1947. Diacetyl mutase. Jour. Biol. Chem. 167:811-816. Hales,‘M.'W. 1945. Cultures and Starters. 2nd ed. Chr. Hansen's Laboratory, Inc., Milwaukee, Wis. 76 pp. Hammer, B. W., and F. J. Babel. 1943. Bacteriology of butter cultures: A review. Jour. Dairy Sci. 26:83-168. Hammer, B. W., and F. J. Babel. 1957. Dairy Bacteriology. 4th ed. John Wiley and Sons, Inc., New York. 614 pp. plus ix. Hoecker, W. H., and B. W. Hammer. 1941. Flavor development in salted butter by pure cultures of bacteria. Preliminary results. Iowa Agr. Expt. Sta. Res. Bul. 290. Jenness, R., and S. Patton. . 1959. Principles gquairy Chemistry. John Wiley and Sons, Inc., New York. 446 pp. plus viii. Jennings, W. G. 1957. Application of gas-liquid partition chromatography to the study of volatile flavor compounds. Jour. Dairy Sci. 39; 271-2790 Juni, E. 1952. Mechanism of formation of acetoin by bacteria. Jour. Biol. Chem. 195:715-726. Juni, E., and G. Am Heym. 1956. A cyclic pathway for the bacterial dissimilation of 2,3- butanediol, acetoin, and diacetyl. I. General aspects of the 2,3-butanediol cycle. Jour. Bact. 11:425-432. JLmi, E., and G. A. Heym. 1956. A cyclic pathway for the bacterial dissimilation of 2,3- butanediol, acetoin, and diacetyl. II. The synthesis of diacetylmethyl-carbinol from diacetyl, a new diphOSphoth- immin catalyzed reaction. Jour. Bact._12:746-753. Kobayashi, Y., and G. Kalnitsky. 1954. Bacterial synthesis of o<3acetolactate. Jour. Biol. Chem. 211:473-477. (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) (33) 32 Krenn, J., and D. Valik. 1949. Untersuchungen uber die Art und Weise der Bildung der Aromastaffe der Butter. 12th Internatl. Dairy Cong., Stockholm. .2:516-520. (In German: Original not seen. Cited by Andersen, V. B. Milchwissenschaft 14:429-432). Krishnaswamy, M. A., and F. J. Babel. {1951. Biacetyl production by cultures of lactic acid producing streptococci. Jour. Dairy Sci. 34:374-385. Michaelian, M. B., B. W. Hammer, and W. H. Hoecker. 1938. Effect of pH on the production of acetoin plus diacetyl in milk by the citric acid fermenting streptococci. Jour. Dairy Sci. 21:213-218. Michaelian, M. B., W. H. Hoecker, and B. W. Hammer. 1933. Nawar, W. 1960. Pette, J. 1949. Pien, J., 1937. Prill, E. 1938. Strecker, 1954. Tumerman, 1954. Virtanen, 1937. Relationship of acetylmethylcarbinol and biacetyl to butter cultures. Iowa Agr. Expt. Sta. Res. Bul. 155. W., and I. S. Fagerson. Technique for collection of food volatiles for gas chroma- tographic analysis. Analyt. Chem.'§2:1534. W. Some aspects of the butter aroma problem. 12th Internatl. Dairy Cong., Stockholm. 2:572-578. J. Baesse, and R. Martin. Le dosage du diacetyle dans les beurres. Le Lait 11:675-698. (In French: Original not seen. Cited by Hoecker, W. H., and B. W. Hammer. Iowa Agr. Expt. Res. Bul. 290). A., and B. W. Hammer. A colorimetric method for the microdetermination of diacetyl. Iowa State Col. Jour. Sci..12:385-394. H. J., and I. Harary. Bacterial butylene glycol dehydrogenase and diacetyl reduc- tase. Jour. Biol. Chem.,2£l:263-270. L., H. Fram, and K. W. Cornely. The effect of lactose crystallization on protein stability in frozen concentrated milk. Jour. Dairy Sci. 21:830-839. A. I. The influence of oxygen on the formation of butter aroma. 11th Internatl. Dairy Cong., Berlin. 25121-123. r. I“. pm. as .. u. g... ‘. .,, ”flu- ‘. 3* 15;. .fi. h l I | 31451531 I I.|l l | I III. l I | l 3 1293 ElHllllHlllllllllllHllH