EL 3. ..g..' 11‘xt'l-isl'fl-¢‘MV ‘2 i EBR 1?; 7?" M Ichigar: "- ‘ ‘40 Universzty 13“?) ' c 13 THESIS This is to certify that the thesis entitled DETERMINATION OF VOLATILE FLAVOR CONSTITUENTS AND RESIIIIAL CARBOHYDRATES DURING THE FERMENTATION OF YOGURI' presented by Sebastiao C. C. Brandao has been accepted towards fulfillment of the requirements for Ph.D. degree in Food Science 4 W Major professor Datewo 0-7839 4w “ (“A a DETERMINATION OF VOLATILE FLAVOR CONSTITUENTS AND RESIDUAL CARBOHYDRATES DURING THE FERMENTATION OF YOGURT BY Sebastiao C. C. Brandao A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science and Human Nutrition 1980 e r! M. 34/ ABSTRACT DETERMINATION OF VOLATILE FLAVOR CONSTITUENTS AND RESIDUAL CARBOHYDRATES DURING THE FERMENTATION OF YOGURT BY Sebastiao C. C. Brandao The lactose, glucose, and galactose contents of yogurt during processing, fermentation, and storage were evaluated by high performance liquid chromatography using a Waters uBondapak/carbohydrate column. Since galactose and glucose cannot be resolved by this column, a method was developed to eliminate the glucose by glucose oxidase, thus allowing the quantitation of all three sugars. The volatile components of yogurt produced during processing, fermentation, and storage were identified and quantified using a modified head—space technique in which 5 grams of yogurt is mixed with 5 ml of 20% (w/w) sulfuric acid to produce a solution with a pH of 0.87. The advantages of this modification of previous head- space techniques include the pH control of the diluted sample, the prevention of any microbial growth during equilibration of the head-space gas (60°C and 204 rpm for Sebastiao C. C. Brandao 30 minutes) and the elimination of polymerization of acetaldehyde or other aroma compounds.- Acetaldehyde, acetone, ethanol and diacetyl were identified by gas chromatography—Mass spectrometry. Quantitative analyses were performed using standards added to yogurt mix and analyzed by gas chromatography. To evaluate the effect of the Streptococcus thermophilus to Lactobacillus bulgaricus ratio on the chemical components, a balanced culture (1:1 ratio) and an unbalanced culture (2:1 ratio) of cocci to rods were used to produce yogurt. The culture used throughout this work was a Chr. Hansen frozen concentrated culture. Results indicate that lactase activity was observed mainly during fermentation with very low activity during storage for 28 days. While the hydrolysis of lactose was faster for the balanced culture (30% lactose hydrolysis), the percentage of lactose hydrolyzed was higher for the unbalanced culture (34%) due probably to its longer incubation time to achieve the pH 4.25. The balanced culture metabolized 86% of the glucose and 13% of the galactose produced by the hydrolysis of lactose while the unbalanced culture metabolized 87% and 7% of the glucose and the galactose, respectively. The concentration of all volatile components increased during fermentation. Acetaldehyde began to be Sebastiao C. C. Brandao produced appreciably when the pH of the inoculated yogurt mix was about 6.0 for both culture ratios. From there on a steady increase was observed until the end of the fermentation. The acetaldehyde content of the yogurt produced by the balanced culture was higher (25.5 ppm) than the unbalanced culture (19.0 ppm) due to a steeper production curve. Lactobacillus bulgaricus, the organism which was deliberately out of balance in the latter culture, is the main producer of acetaldehyde, especially in the later phase of the fermentation. Therefore, the effect of unbalance of this microorganism on the yogurt culture results in a lag period in the second phase of the fermentation which allows the Streptococcus thermo— philus to keep reproducing and lowering the pH without increasing the acetaldehyde content appreciably. The ethanol, diacetyl, and acetone content also increased during fermentation but to a lesser extent than acetaldehyde. Yogurt stored for 14 days showed a slight decrease of acetaldehyde content and a slight increase in the ethanol content for both cultures. However, after 14 days a sharp increase in the ethanol content was observed with a concomitant decrease in the acetaldehyde content. These transformations suggest that ethanol dehydrogenase activity is very important during storage of yogurt, especially after approximately 14 days of storage. Sebastiao C. C. Brandao However, the activity of other enzymes may also play an important role on the stability of yogurt flavor. ACKNOWLEDGMENTS The author wishes to express his sincere gratitude and appreciation to Dr. C. M. Stine for his encouragement, guidance, and help during the course of this study. Thanks are also expressed to Dr. P. Markakis, Dr. H. A. Lillevik, Dr. B. R. Harte for their assistance and aid in preparing this manuscript. Grateful acknowledgment is also due to Dr. I. J. Gray for his advice and effort in reading this manuscript. The author feels grateful to the Departamento de Tecnologia de Alimentos da Universaidade Federal de Vicosa (Brazil) and to the Programa de Ensino Agricola Superior (PEAS-Brazil) for the opportunity and financial assistance granted to him. Very special gratitude is also due to the author's wife, Livia, and daughter, Ive, for their devotion, encouragement, patience, and understanding. ii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . V LIST OF FIGURES . . . . . . . . . . . Vi INTRODUCTION . . . . . . . . . . . . 1 LITERATURE REVIEW . . . . . . . . . . . 3 Definitions . . . . . . . . . . . 4 The Yogurt Mix . . . . . . . . 7 The Milk Fat Content . . . . . 7 The Solids Not Fat Content . . . . . 8 Stabilizers . . . . . . . . . . 9 Heat Treatment . . . . . . . . . 10 Homogenization . . . . . . . . . l4 Yogurt Fermentation . . . . . . . 17 The Microorganisms Used for Yogurt Pro- duction . . . . . . . . . . . 17 Culture Preparation . . . . . . . 21 Yogurt Cultures . . . 23 Metabolic Products of Yogurt Fermentation . 26 Lactose Fermentation . . . . . . . 26 Protein Fermentation . . . . . . . 31 Lipid Metabolism . . . . . . . . 33 Vitamin Metabolism . . . . . . . . 35 Components of Yogurt Flavor . . . . . 36 Background Flavor of Yogurt . . . . . 36 Yogurt Flavor . ,. . . . . . . 38 Nutritional Properties of Yogurt . . . . 44 Yogurt as a Food . . . . . . . . 45 Lactose Intolerance . . . . . . . 45 Therapeutical Properties of Yogurt . . 47 MATERIALS AND METHODS . . . . . . . . . 50 Yogurt Manufacture . . . . . . . . 50 Culture . . . . . . . . . . . 50 Culture Medium . . . . . . . . . 50 Bulk Starter Preparation . . . . . . 50 iii Flavor Analysis . . . . . . . Gas Chromatography . . . . . . . Head-Space Analysis . . . . Equilibrium Analysis . . . . . . Flavor Survey . . . . Sample Preparation for the Gas Chromatographic-Mass Spectrometric Analysis . . . . . Gas Chromatographic-Mass Spectrometric Analysis . . . . Analysis of Standard Compounds . . . Carbohydrate Analysis by High Performance Liquid Chromatography . . . . . . Apparatus . . . . . . . . . Mobile Phase . . . . . . . . . Standards . . . . . . . . . . Sample Preparation . . . . . Samples for Lactose Analysis . . Samples for Glucose and Galactose Analysis . . . Glucose Oxidation by Glucose Oxidase . RESULTS AND DISCUSSION . . . . . . . . Analysis of Carbohydrates in Yogurt by High Performance Liquid Chromatography (HPLC) The HPLC Method . . . . . . Analysis of Sugar Standards . . . . Oxidation of Glucose by Glucose Oxidase Changes in the Carbohydrates Content of Yogurt During Processing, Fermenta- tion and Storage . . Analysis of the Aroma Components of Yogurt . . Gas Chromatography of the Yogurt Aroma Equilibrium Analysis . . . Identification of the Volatile Compon— ents . . . . Survey of the Aroma Components of Plain Yogurt . . . . . Analysis of Standard Compounds . . Changes in the Aroma Components During Processing and Incubation . . . Changes in the Aroma Components during Storage . . . . . . Importance of the Flavor and. Aroma on Quality of Yogurt SUMMARY AND CONCLUSIONS . . . . . . . . REFERENCES iv Page 53 53 54 55 55 56 57 57 58 58 58 59 60 60 62 64 67 67 68 71 72 74 80 80 84 87 87 89 89 95 99 105 110 LIST OF TABLES Linear Regression Curves for Glucose, Galactose and Lactose Standards . Quantitative Analysis of Some Commercial Plain Low—Fat Yogurts . . . . . . Linear Regression Equations for Acetalde- hyde Acetone, Ethanol, and Diacetyl Page 71 88 89 Figure l. 10. ll. 12. LIST OF FIGURES Enzymatic transformation of galactose to glucose-l-Phosphate . . . . . . . . Overall metabolism of lactose to produce lactic acid . . . . . . . . . . Mechanisms for the degradation of pro— teins . . . . . . . . . . . . Mechanisms for the metabolism of lipids Compounds formed by thermal degradation of yogurt mix during heat treatment . . . . Flow diagram of yogurt making . . . . . Flow diagram to extract carbohydrates from yogurt samples . . . ., . . . . . Flow diagram of the preparation of samples for the analysis of glucose, galactose and_ lactose . . . . . . . . . . . HPLC chromatogram of an yogurt sample con— taining glucose, galactose and lactose . . Concentration of residual glucose after oxidation by glucose oxidase . . . . . The pH and the content of glucose, galac— tose, and lactose of yogurt during proces- sing, incubation and storage, using the balanced culture . . . . . . . . . The pH and the content of glucose, galac- tose and lactose during processing, incuba- tion, and storage of yogurt, using the unbalanced culture . . . . . vi Page 29 30 32 34 39 52 61 63 7O 73 75 78 Figure 13. 14. 15. 16. 17. 18. 19. 20. 21. Page Gas chromatogram of head-space gas of inoculated yogurt mix after 1.5 hours of incubation . . . . . . . . . . 81 Equilibrium curves for ethanol, acetone, and acetaldehyde . . . . . . . . . 85 Equilibrium curves for diacetyl and the unknown which occurred between acetone and ethanol . . . . . . . . . . 86 The pH and the concentration of acetalde- hyde, ethanol, acetone, and diacetyl of yogurt during processing and incubation using a balanced culture (cocci/rods ~ l/l) . . . . . . . . . . . . . 90 The pH and the concentration of acetalde- hyde, ethanol, acetone, and diacetyl of yogurt during unbalanced culture (cocci/ rods ~ 2/1) . . . . . . . . . . 93 The concentration of acetaldehyde, ethanol, acetone and diacetyl of yogurt (pH 4.25) produced by the balanced culture, during storage . . . . . . 96 The concentration of acetaldehyde, ethanol, acetone and diacetyl of yogurt (pH 4.25) produced by the unbalanced culture, during storage . . . . . . 97 The concentration of acetaldehyde, ethanol, acetone and diacetyle of yogurt (pH 4.50) produced by the unbalanced cul- ture during storage . . . . . . . . 98 Simplified diagram of the metabolic path— way to produce the flavor components of yogurt . . . . . . . . . . . . 103 vii INTRODUCTION Various forms of cultured milk are consumed in many nations of the world. A particular product may be preferred in one country, whereas in another country the product may notlxaavailable because of limited demand. Commercial yogurt production was introduced into the United States in 1942 by Dannon, but little attention was shown for it until the mid to late 19605. Since then, however, yogurt has experienced a fantastic growth rate. The trend into more varied yogurt products, such as soft serve and hard frozen yogurt, liquid yogurt, and yogurt drink will possibly expand this growth even more in the coming years. The introduction of frozen yogurt culture in the last decade has contributed enormously to successful commercial production of yogurt. This type of culture requires less transferring and handling thereby reducing the possibility of contamination or infection with bac- teriophage. Labor and equipment needs are also reduced and the use of this type of culture enables the manufac- ture of a product with standardized flavor and texture, thus diminishing inconsistent quality in the finished yogurt. Until now the fruit yogurt specialties have the biggest part of the market but plain yogurt has received most of the attention of researchers due to its basic function in all types of yogurt products. This project was undertaken to study the forma- tion of volatile components during the fermentation process, the stability of these volatile components during storage and the enzymatic hydrolysis of lactose to its constituent monosaccharides. The fermentation of the resulting glucose and galactose, during processing and storage of yogurt pro- duced from a frozen concentrated yogurt culture was also studied. The effect of culture handling, or more spe- cifically, the ratio of Streptococcus thermophilus to Lactobacillus bulgaricus during inoculation of yogurt mix on the chemical changes was also evaluated. LITERATURE REVIEW The exact origin of fermented milk is difficult to determine, but apparently evolved in Mesopatemia about 5,000 B.C., when the goat was first domesticated. The milk stored warm in gourds in that hot climate may have been fermented by microorganisms and formed a curd. Then it was observed that inoculating the milk with some of the fermented milk of the previous day would reduce the chance of failures and would produce the product more rapidly and uniformly Urniby natural contamination (Kosikowski, 1977). The first Turkish name for fermented milk appeared in the Eighth Century as "Yogurut" and subsequently was changed in the Eleventh Century to its present form (Rasic and Kurmann, 1978). Robinson and Tamine (1975) noted that, in spite of its obvious p0pularity, no precise definition of yogurt has ever been formulated, and even the spelling appears to be a matter of personal preference (i.e., yogurt, yoghurt, yaourti, yaourt, yahourth, yogur, etc.). Yogurt was introduced commercially in the United States by Dannon in 1942 in the New York Metropolitan area (Soskin, 1967). Definitions Yogurt is a product which results from culturing pasteurized milk with or without added nonfat milk solids, partly skim milk or skim milk with a characteriz- ing bacterial culture that contains the lactic acid producing bacteria Lactobacillus bulgaricus and Strepto— coccus thermophilus. Yogurt contains not less than 3.25 g percent milk fat; low-fat yogurt contains not less than 0.5 percent nor more than 2 percent milk fat; non-fat yogurt contains less than 0.5 percent milkfat. All the yogurt types mentioned must have not less than 8.25 per— cent milk solids not fat and they must have a titratable acidity of not less than 0.5 percent, expressed as lactic acid. The added milk—solids-not—fat must not change the ratio of protein to total nonfat solids and the protein efficiency ratio of all protein present must not be decreased as a result of adding such ingredients. Other optional ingredients include nutritive carbohydrate sweeteners, characterizing flavoring ingredients, color- ing, and stabilizers. All dairy ingredients used must be pasteurized or ultra-pasteurized and may be homogenized. To extend the shelf-life of the food, yogurt may be heat- treated after culturing is completed, to destroy viable microorganisms, provided that the parenthetical phrase "(heat-treated after culturing)" follows the name of the food on the label (Anonymous, 1977). A survey on household purchases of yogurt in the United States was made in 1970 by the American Dairy Association (Anonymous, 1970) and another in 1972 by the United Dairy Association (Anonymous, 1974). They con- cluded that yogurt consumption increases with income, yogurt is more popular with whites than with nonwhites, and the greater the degree of urbanization, the higher the usage of yogurt. At all ages, yogurt consumption is higher among females than males. Yogurt is more popular in the west and least popular in the south. The 1970 survey reported that the most frequent ways of eating yogurt are as a between-meal snack, as a dessert, and as a complete noon meal. Strawberry yogurt was by far the yogurt most preferred (61 percent) followed by Blueberry, Raspberry, Plain, Cherry, Orange/pineapple, Vanilla, Boysenberry, Appricot, Apple, Prune, and Coffee. Kroger and Fram (1975) reported that in a survey made in Pennsylvania, 74 percent of the surveyed households preferred fruit—flavored yogurt, while one-half preferred the Swiss style and 40 percent the Sundae style. Knutson et a1. (1978) classified yogurt into three distinct categories: fresh yogurt, frozen yogurt mix (soft serve), and hard frozen yogurt including novelties, while most of the yogurt production (about 98 percent) is fresh yogurt. Jachumsen (1978) reported that the interest of both soft and hard frozen yogurt has increased and presents a new market. Steinitz (1971) reviewed the standards for frozen yogurt, including yogurt sherbet and frozen yogurt dessert. The so-called fresh yogurt may be classified in many ways: Sundae Style-~Fruit is deposited on the bottom of ' the cup and plain yogurt over the fruit. 1 Sundae Style (Western)--Fruit is deposited in the cup, same as above, a lightly flavored and colored yogurt is deposited over the fruit. Sundae Style (Special)--Part of the fruit is placed into the cup, and the other portion of the fruit and flavored syrup is blended with the yogurt. This blend is deposited over the fruit. Swiss Style-—Fruit is blended into the yogurt and the cups are filled with this mixture. Swiss Type (Special)-—Yogurt is deposited into the cup and the fruit is then deposited over the yogurt. As the fruit settles, some flavor and color adheres to the yogurt (giving a slight marbled effect). Liquid Yogurt-~Fruit made with puree is blended into a liquid type product. This yogurt sets to a soft drinkable product. The Yogurt Mix Yogurt mix is a blend of all ingredients which will be inoculated with the yogurt starter culture to produce plain yogurt. The Milk Fat Content The milk fat content depends on the type of yogurt being manufactured (Nonfat yogurt, Low fat yogurt, and yogurt). Kosikowski (1977) suggested that in the United States the fat content of (Low fat) yogurt is 1.7 percent, but a considerable amount is still being made from whole milk of about 3.3 percent fat. However, Kroger and Weaver (1973) reported a market survey of yogurt sold in Central Pennsylvania in which the fat content ranged from 0.82 to 2.04 percent with an average of 1.18 percent and variance of 0.14. In another survey made in Canada by Duitschaever et a1. (1972) the fat content ranged from 0.9 to 3.6 per- cent with a mean of 1.98 percent. Richmond et a1. (1979) reported that there is still much variation in yogurt composition not only between brands but within the same brand. They commented that there has been little effort to standardize yogurt during the past seven years despite the fact that better uniformity in composition and quality would be beneficial to both consumer and processor. In general, the milk used for yogurt manufacture should be standardized to a fixed fat content in order to assure uniformity of the end product. By standardization manufacturers can influence to Some extent: (a) flavor and aroma; (b) consistency, viscosity, and stability; and (c) nutritive value (Cottenie, 1978). The Solids-Not-Fat Content Most yogurt is made from standardized milk with 1 to 6 percent milk solids not fat added (Czulak, 1962). Eet&a(1964) reported that the addition of milk solids not fat is unnecessary if a good symbiotic culture is used. However, Rasic and Kurmann(1978) noted that increased milk solids not fat leads to the following effects: (a) an increase in nutritive and caloric value, (b) an improvement of the organoleptic properties such as con- sistency and taste, and (c) an increase of the buffer capacity of the yogurt mix. Humphreys and Plunkett (1969) noted that higher than normal solids not fat in the mix increase acidity development, reducing coagulation time. However, the quality of the milk solids added should be carefully checked to ensure that it will have no deleteri— ous effect on the flavor and on the culture (Barber, 1962). Cottenie (1978) suggested the concentration of milk solids through evaporation, because it would build up a optimum casein micelle structure which reduces or eliminates the. tendency of wheying-off (synerysis). Kosikowski (1977) suggested the concentration of milk solids by ultrafil- tration and reverse osmosis. Stabilizers If properly made, plain yogurt requires no stabili- zer because a stiff, smooth gel with high viscosity is attained naturally (Meiklejohn, 1977). However, for the production of swiss-style yogurt stabilizers are usually added in order to avoid syneresis, to improve consistency and viscosity, and to establish a proper mouthfeel (Niel- sen, 1975). Stabilizers used in the manufacture of yogurt must fulfill the following requirements: (a) to be a safe, harmless food additive, (b) to have a neutral taste, (c) to present no masking of the natural yogurt flavor, (d) to be easily incorporated into the mix, (e) to hydrate under normal processing conditions, (f) to be stable against physical, chemical, and biological effects, (g) to give the desired texture properties at low concentration, (h) to be efficient at low temperatures, (1) to impart no color to the product, and (j) to be effective at low pH (Rasic and Kurmann, 1978). Radema and Dijk (1973) noted that some stabili— zers may retard acid production and may even cause whey separation. Hall (1975) stressed the fact that the com— position as well as the concentration of the stabilizer depend largely on the total solids content of the product. 10 Stabilizers such as modified starches, alginates, carageenan and gelatin are used to stabilize yogurt. With the exception of gelatin, the stabilizers cause observable changes in the microstructure of yogurt (Kalab et al., 1975). Kalab (1979) reported that transmission electron microscopy failed to distinguish between the short fibers and sheets formed by either pre-gelatinized waxy corn starch (2 percent) and carageenan (0.4 percent). Kalab et a1. (1976) observed differences in the microstructure of set and stirred yogurt by scanning electron microscopy. Set yogurt has an uninterrupted three-dimensional network composed of chains and clusters of casein micelles. Stirred yogurt has fewer chains and more clusters of micelles joined together by thin fibers. Meiklejohn (1977) pointed out that several factors may lead to a satisfactory texture in yogurt, such as increased solids, stabilizers, the pasteurization process, lower incubation temperatures and the technique used for cooling, fruiting, and filling. Heat Treatment The heat treatment of yogurt mix has several important effects (Rasic and Kurmann,1978): a. pasteurization (to kill pathogenic micro- organisms which may be present) 11 to kill the majority of other microorganisms, including phages, in order to increase keeping quality To inactivate naturally occurring enzymes to denaturate the whey proteins so a better body is achieved to improve starter activity to lessen the coagulation time to produce antioxidative properties, thus inhibiting fat oxidation, and to improve protein digestibility and to produce a softer curd The primary reason for the heat treatment of yogurt mix is the pasteurization required by the U. S. Department of Health, Education, and Welfare (Anonymous, 1978). However, the temperature-time combinations used for the heat treatment of yogurt mix results in a much higher treatment than that required for simple pasteuriza— tion (Yeager, 1975). A variety of heat treatments for yogurt mix have been suggested. Some of them are: a. 80-88°C for 15 to 30 minutes (Grigorov, 1965) 90°C for 2 to 3 minutes (Storgards and Aule, 1958; Schulz, 1953) 12 c. 88°C for 30 minutes (Kosikowski, 1977) d. 90°C for 38 seconds (Kosikowski, 1977) e. 140-150°C for 2 to 4 seconds (Rasic and Kurmann, 1978) These heat treatments would denaturate between 70 and 95 percent of the whey proteins (Larson and Rolleri, 1955; Pavey, 1974), and they would denature most of the enzymes present (Hall and Hedrick, 1966). They would also kill more than 99 percent of the microorganisms present (Hall and Trout, 1968) and would produce a product with good consistency (Storgards, 1964). There is a great controversy in the literature over the degree of heat treatment of yogurt mix which provides the best starter activity (Feldstein and West— hoff, 1979). Jago (1954) reported that milk autoclaved at 121°C for 15 minutes and milk boiled for 30 minutes improved acid production compared to cultured raw milk. He suggested this was due to the destruction of "the inhibitory property" which is found in raw milk. Greene and Jezeski (1957a and 1957b) reported several distinct heat treatments zones which affect starter activity. A primary zone of culture stimulation occurred in milks heated from 62°C for 30 minutes through 72°C for 40 min— utes, followed by a zone of culture inhibition of milk heated from 72°C for 45 minutes through 80°C for 10 to 120 minutes. Another stimulation zone occurred in milks 13 heated at 90°C for 60 to 180 minutes or autoclaved at 120°C for 15 to 30 minutes. A final inhibition of culture activity was noted in milk autoclaved for more than 30 minutes at 120°C. On the other hand, Olson (1966) noted that inconsistent results may be obtained if the activity of cultures is determined in milks heated at temperatures considerably higher than pasteurization. Feldstein and Westhoff (1979) indicated that several factors may affect the studies of the heat treatment of milk as it affects starter activity, such as solids not fat content, total solids content, milk quality and the variability among starter cultures. Believed to be responsible for the initial stimulation were the following: oxygen expulsion, destruction of inhibitors, partial protein hydrolysis, serum protein denaturation, formation of free amino acids and formation of formic acid (Stodklin, 1969; Greene and Jezeski, 1957a). Inhibition was associated with formation of toxic volatile sulfides and the succeeding stimulation to heat induced disappearance of sulfides (Greene and Jezeski, 1957b). Grigorov (1966) recognized the importance of preheating of milk to a minimum temperature of 85°C to obtain high quality yogurt. Yogurt made from heated milk is firmer and more resistant to syneresis than yogurt made from unheated milk (Kalab et a1., 1976). ,The 14 microstructure of yogurt is finer, being composed of smaller particles, than the network of yogurt of the same composition made from unheated milk. Davies et a1. (1978) observed filamentous appendages composed of denaturated B-lactoglobulin on casein micelles in milk heated at 95°C for 10 minutes or autoclaved at 121.1°C for 15 minutes. These appendages were retained to a late stage of fermen- tation. Creamer et a1. (1978) reported that the complex of denaturated B—lactoglobulin and k—casein on the micelle inhibited micelle fusion. The absence of this complex in ' micellar strands in yogurt from unheated milk leads to additional coalescence causing syneresis (Kalab, 1979). The heat treatment usually accepted for yogurt mix are those summarized by Kosikowski (1977), which are: Low Temperature Long Time - 88°C (190°F) for 30 minutes, High Temperature Short Time — 95°C (203°F) for 38 seconds. Homogenization Yogurt mix is homogenized to reduce the size of the fat globules and to create ahigh degree of fat dispersion. The newly formed fat globules are no longer completely covered with the original membrane material, but instead are covered with high concentrations of casein and lesser quantities of serum proteins (Brunner et al., 1953a and 1953b). Most investigators have observed an increase 15 in the viscositycxfhomogenized milk,which can be attribu- ted to newly formed protein-lipid complexes (Brunner,1974). Homogenization reduces the curd tension of yogurt mix, which would improve digestibility. Trout (1950) reviewed the effects of homogenization on the lowered curd tension of homogenized milk and concluded it was due to the increase in the number of globules serving as point of weakness in the coagulum and to the casein adsorption on the increased fat surface area, causing lower concen— tration in the serum. Galesloot (1958) studied the effect of homogeniza— tion of milk on the rheological properties of yogurt. He reported an increase in viscosity (after the product was stirred). Mulder and Walstra (1974) noted that agitating yogurt at high shear rates considerably impairs Viscosity. However, Richter and Hartman (1977) reported that pene- trometer values did not present any correlation with body and texture of yogurt. They noted that the major diffi- culty in relating penetrometer values to body and texture escores is that the penetrometer does not measure texture, and defects such.asgraininess,syneresis and sliminess are not detected bythe penetrometer; Thisis the result of body and texture being two distinct measurements. Bodycan gen— erally be considered a function of viscosity while texture is a function of the physical integrity of the body. 16 Homogenization affects the color of the milk rendering it more densely white than that of nonhomogen- ized milk (Trout et a1., 1935). Apparently this color change is due to the increase in the number and total sur- face area of fat globules capable of reflecting and scattering the rays of light (Sommer, 1946). Milk must be heated before homogenization to obtain efficient break-up of the fat globules. Homogenization of raw milk generally is not advised because of the activa- tion of lipase, yielding rancidity (Weinstein and Trout, 1950). The lipase is inactivated by the minimum heat treatment required for pasteurization (Hetric and Tracy, 1948). If inadequately heat treated raw yogurt mix is homogenized, it must be pasteurized as quickly as possible after homogenization to prevent lipolysis. Pressures from 2000 UDZSOO psi are normally used for homogenization and iftasecond stage isused, it usually would be 500 psi. These pressures will meet the U.S. Public Health Standards for homogenized milk (Anonymous, 1978). Temperatures of 57°C (135°F) or higher are used with an increase in efficiency resulting as the temperature of homogenization is increased (Hall and Trout, 1968). Mulder and Walstra (1974) reported a sharp influ- ence of temperature during homogenization on the fat globule size. Temperatures below 30°C had the sharpest influence while temperatures above 45°C had a small 17 influence. They also reported that precooling of milk for 24 hours at 4°C had a tremendous influence on the fat globule size when homogenization is made at temperatures below 30°C. Goulden and Phipps (1964) reported that the effect of temperature during homogenization on the fat globule size depends on the fat content; as the fat con- tent increases homogenization efficiency decreases, par— ticularly at high pressures and low temperatures. Yogurt Fermentation Fermentation may be defined as the metabolic process in which chemical changes are brought about on an organic substrate, whether protein, carbohydrate, or fat, through the action of enzymes elaborated by specific liv- ing microorganisms. In yogurt fermentation, the selection of the active bacterial cultures is essential for success- ful processing and product quality. The Microorganisms Used for Yogurt Production Two microorganisms, Lactobacillus bulgaricus and Streptococcus thermophilus, growing together symbioti- cally are responsible for the lactic fermentation of yogurt (Chambers, 1979). Both microorganisms belong to the homofermentative group of lactic acid bacteria (Pelczar and Reid, 1972). 18 Pette and Lolkema (1950a and 1950b) reported that the more proteolytic Lactobacillus bulgaricus stimulates the growth of Streptococcus thermophilus by liberating a number of amino acids from the casein, including valine (the most important). In spring the streptococci required the following amino acids: lysine, leucine, histidine, valine, cysteine, and aspartic acid. They also noted that in autumn the amino acids glutamic acid, isoleucine, glycine, tyrosine and methionine are required in addition to the six other amino acids required in spring. On the other hand, Bautista et a1. (1966) reported that histidine and glycine are the main stimulants of Streptococcus thermophilus. Galesloot et a1. (1968) proved that Streptococcus thermophilus also stimulates Lactobacillus bulgaricus by producing a factor that is equal to or can be replaced by formic acid. However, this stimulation was observed in moderately heated milk only, while in intensively heated milk this stimulation cannot be observed on account of the formic acid produced by the heat treatment. Veringa et a1. (1968) then furnished evidence that the substance which stimulates Lactobacillus bulgaricus, which is pro- duced by Streptococcus thermophilus is really formic acid. 19 Pette (1957) found that this symbiotic growth produces a shortened generation time of Streptococci cells and increased number due to the stimulation by the lactobacilli. Therefore, the streptococci grow faster during the early stage of fermentation, outnumbering the lactobacilli by three to four after the first hour (Pette and Lolkema, 1950a). Then the growth of streptococci is slowed down due to the adverse effect of lactic acid and the lactobacilli take over the fermentation and develop the acidity further. The production of formic acid by the streptococci stimulates the lactobacilli which will also have a shortened generation time. Consequently, the pro- duction of lactic acid is enhanced by the mutual symbiosis and the acidity of the product (yogurt) is always higher than in single culture growth (Pette, 1957; Tramer, 1973). From the technological point of view, this bac- terial symbiosis may produce undesirable effects if not properly controlled. For example, the risk of over- acidification of the yogurt is greater (i.e., during cool- ing). The use of a mild acid producing yogurt culture is recommended in order to reduce over-acidification (Rasic and Kurmann, 1978). Pette (1957) reported that yogurt with a 1:1 ratio of Streptococcus thermophilus and Lactobacillus bulgaricus had a count of 5.6 x 108 Streptococcus thermophilus (and 20 Lactobacillus bulgaricus) per m1 and an acidity of approxi— mately 1 percent lactic acid. If a higher acidity were allowed to develop the number of Streptococcus thermophilus would decrease while the number of Lactobacillus bulgaricus would increase, changing the ratio in favor of the bacillus. White (1966) recommended a final acidity, after cooling, of about 0.9 percent lactic acid; however, control of pH would be more recommended than control of the final acidity. He also suggested that cooling of the yogurt should be as fast as possible in order to avoid over- acidification. The final temperature of the yogurt should be below 4.4°C until consumed. Another very important technological factor is the incubation temperature-time relationship which is most con- venient and least expensive to the commercial manufactur- ing of yogurt. Lundsted (1971) suggested the temperature of 30°C (86°F) for 14 to 16 hours, depending on the cul- ture. Advantages of this method are: energy economy for heating and cooling, easier control of acidity due to slow acid development and formation of a better texture. Cham- bers (1979), on the other hand, suggested incubation temperatures between 41.1 and 42.4°C for shorter times (i.e., 3 to 5 hours) because it would tend to favor the lactobacilli and streptococci cultures equally and yield the desirable 1:1 ratio with some degree of reliability. 21 He also mentioned that if the temperature is increased in the range above 42.4 up to 46.1°C, thelactobacilli culture is favored and will predominate, while at temperatures below 41.1°C the streptococci culture is favored. Sellars (1973) referred to incubation temperature between 40 and 45°C, while Powell (1970) suggested between 42.4 and 45°C. Mayer and Powell (1966) patented a method using a temperature range from 43.3 to 46.1°C. In another patent Donay and Tahan (1964) mentioned a temperature range of 35 to 37.8°C. From this wide range of incubation temperatures it becomes clear that one should use the temperature rec— ommended by the starter manufacturer or standardize a temperature which would be better for the conditions used, depending on the solids not fat content, heat treatment used and the presence of other yogurt components. However, one should keep in mind that the propor- tion of Lactobacillus bulgaricus to Streptococcus thermophilus may be influenced not only by the incubation temperature, but also by the incubation time, the amount of inoculum, and the final acidity (Pette and Lolkema, 1951a and 1951b). Culture Preparation Cultures of yogurt bacteria are made in special asseptic flaks in which they have been cultivated by 22 periodic transfer to fresh media. The time intervals at which the transfers are made vary with the necessity of the manufacturer, but it should be less than one week. Davis (1956) suggested that reconstituted spray— dried nonfat dry milk be used as the transfer media for the propagation of the cultures. The reconstituted milk should be well mixed and sterilized in an autoclave for 15 minutes at 121.1°C to produce the desirable changes necessary for improved growth. These changes include the sterilization of the milk, denaturation of whey proteins, expelling dissolved oxygen and destruction of heat—labile inhibitors which might be present (Sellars, 1973). The inoculatechulture should be incubated at the same temperature which will be used for the bulk starter, and the microorganisms should be allowed to grow until the desirable ratio of 1:1 is achieved (Nielsen, 1975). The control of this ratio should be done by experts, since it is extremely difficult to open the flasks and determine the ratio without contaminating the culture and/or disturbing the culture medium. A control culture could, however, be run concomitant with the main culture to check the coccus to rod ratio. Usually dairy plants allow the culture to develop to a pH of 4.0 to 4.3 and then refrigerate at 15.5 to 21.1°C if another transfer is going to be made shortly or refrigerate at temperatures below 5°C for storage (Lundstedt, 1971). 23 Yogurt Cultures Yogurt culture, or yogurt starter, is a selected strain of yogurt bacteria which is intentionally added to yogurt mix to bring about specific changes in the appear- ance, body, texture, and flavor, characteristic of the desired final product. Yogurt is made from a starter culture of Streptococcus thermophilus and Lactobacilus bulgaricus either as a mixture which has been propagated together, or which are grown separately and then added to the yogurt mix. The two microorganisms have different optimal growth temperature and acidity. Therefore, handling of yogurt cultures requires that they be allowed to develop symbioti- cally and properly to produce a culture which will fulfill the requirements necessary for a product with good quali- ties. Tramer (1973) recommended the following desiderate of an ideal yogurt culture: a. Purity--free of contaminants b. Vigorous growth c. Production of good consistency in the product d. Production of good flavor e. Stability——microbial balance should be easily maintained f. No tendency to induce syneresis 24. g. Should not develop excessive acidity on storage h. Should be resistant to penicillin and other antibiotics 1. Ease of maintenance j. Phage resistant Acid production during and after incubation (cool- ing and storage) is also emphasized by Rasic and Kurmann (1978) as important qualities in yogurt culture. Different forms of culture are available commer- cially, such as liquid cultures, freeze-dried cultures, concentrated cultures, deep-frozen cultures, and concen- trated deep-frozen cultures. Yogurt cultures are best obtained from well-known commercial culture producers, from research institutes, or other reputable sources, and should be renewed at frequent intervals if phage is a problem. Yogurt plants are now handling yogurt cultures by two different methods. The first method is the conven- tional propagation culture while the second is the frozen intermediate culture. The conventional culture must be transferred to a mother culture which, in turn, is trans- ferred to an intermediate culture and then transferred to the bulk starter. Frozen cultures, on the other hand, are transferred directly to the bulk starter. One can 25 conclude that for the frozen intermediate culture only one transfer of the culture is required by the plant personnel and a minimum amount of labor and equipment is required, which will also reduce culture handling. Less handling of the culture reduces the chance of contamina- tion by either bacteria or phage, and transferring directly into the bulk starter tank (while keeping other portions of the canned frozen culture under refrigeration) will guarantee cultures of greater uniformity (Williamson, 1971). Frozen concentrated cultures have passed the experimental stage and have been used successfully in dairy plants (Lloyd, 1971). The storage of frozen cul- tures has advanced enormously in the last decade. Stad- houders et a1. (1971) reported that frozen cultures could be stored at -20 to -40°C, instead of in liquid Nitrogen at -l96°C, as long as it has cryoprotective agents such as glycerol or lactose, and that the pH of the concentrated is near neutrality at the time of freezing. Ziembia (1970) pointed out that dairy plants may now standardize flavor and texture, and reduce inconsis— tent quality in the finished product by using concentrated frozen cultures. While most researchers (Tramer, 1973; Mocquot, 1970; Robinson and Tamine, 1976) report a satisfactory 26 results with mixed culture, Ashton (1963) reported that incubation of Streptococcus thermophilus at 98-100°F and Lactobacillus bulgaricus at 108-110°F individually and then inoculation of different proportions of the two cul- tures would produce a more constant product. This tech- nique would also facilitate replacement of strains in case of problems. Metabolic Products of Yogurt Fermentation The main product of yogurt fermentation is lactic acid with small quantities of by-products (Zurakynski et a1., 1975). However, acid formation is but one of many changes produced by fermentation of yogurt during the growth of the culture in milk. Proteins are altered, distinctive flavor is produced, by-products are formed that are antagonistic to unwanted bacteria, and the starters contain components that are beneficial to the consumer (Speck, 1980). Lactose Fermentation Lactose is the carbohydrate which is fermented to yield lactic acid and many of the other by—products in a yogurt fermentation. Vakil and Shahani (1962) reported that lactose may be utilized through two routes. The first way is the hydrolytic breakdown into glucose and galactose by the enzyme lactase (Beta-galactosidase) while 27 the second way is the oxidation of the disaccharide by lactose dehydrogenase to form lactobionate, followed then by enzymatic degradation to gluconate and galactose. They further suggested that the same enzyme (lactase) might be responsible for both hydrolyses. Lactase functions both as Beta-galactosidase and as transgalactosidase; that is, lactase not only brings about hydrolytic cleavage of lactose, but also transfers galactosidyl residues to various acceptors including water, glucose, galactose, lactose and oligosaccharides. In solutions containing high concentrations of lactose, the enzymatic reactions may favor formation of oligosa- ccharides at the expense of monosaccharides, glucose and galactose (Aronson, 1952; Roberts and McFarren, 1953; Pazur and Gordon, 1953). Pazur (1953) postulated the following mechanism for lactase—induced hydrolysis of lactose: Glu—1.4 -Gal + Lactase %Gal-—Lactase + Glu Gal-Lactase + Glu tGlu—lfi-Gal + Lactase Gal-Lactase + Gal r-‘rtGal-l.6-Ga1 + Lactase Gal—Lactase + Glu-1.4 —Gal —-—-"'Glu—l.4—Ga1-l.6—Ga1 + Lactase Gal-Lactase + Glu-1.6 — Gal—”Glu—l.6~Gal—l.6-Gal + Lactase Gal-Lactase + Water > Gal + Lactase Roberts and McFarren (1953) and Roberts and Pettinati (1957) reported at least 11 oligosaccharides 28 among the products of the reactions. They also found that as the lactose content of the medium increased, so did the yield of the oligosaccharides. The metabolism of glucose to lactic acid proceeds according to the Embden-Meyerhof pathway (Lehninger, 1977). The lactic acid fermentation follows the glycoly- tic pathway up to the point at which pyruvic acid is produced. Then the pyruvic acid is converted to lactic acid by the enzyme lactic dehydrogenase. Galactose has to be transformed to glucose—l-phosphate to enter the glycolytic pathway (Figure 1). Then, galactose after being transformed to glucose-l-phosphate enters the glyco- lytic pathway and is metabolized to pyruvic acid. Details of the glycolityc pathway may be found in Lehninger (1977). The overall reaction for the production of lactic acid from lactose may be simplified as in Figure 2. Lactic acid bacteria may produce L(+)-lactic acid, D(—)-1actic acid and LD—lactic acid (racemic or inactive) depending on their enzymatic constitution (Rasic and Kurmann, 1978). The lactic acid content of yogurt may vary from 0.85 to 0.95 percent in a mild yogurt to 0.95 to 1.30 per- cent in a more acid yogurt. This acidity is responsible for the coagulation of yogurt, which is initiated at pH 5.2-5.3 and completed at pH 4.6-4.7. The calcium and phosphate associated with casein in the original 29 .Ahhmav Hmmcficsmq can Avmmav ocflmuo Ucm common Scum popmmpma «.oumnmmosmaalomoosam 0D omODOMHcm mo coapmfiuommcmuu UHDMESNGMI:.H ousmflm mmouomamoumos + mumsdmOBEIHsmmoosao Hmm Galacto- waldenase ommnmmmccue asasenns mumammosm HummODomewuomOOSkumQD 4 mmooSHUImQD\\\\V. mBD\\\ mos. + wumzdmonmuaummouomamo Lummmmwmwwwmwmmw a: + 8.06638 30 .Uflom UHDOMH cospoum Op mmouoma Mo Emwaonmuoe Haouo>oll.m ousmflm oumcoflnouocq opmnmmonmlm 1 mozzumm omoucom lTlllllr. -mwoasnam Noo . 1 oumcocsao omouomfimwlilll. mpcnmmonmlm \\ ImDMcOUSHm .1 oymnmmozmlw wxooolmnoquIN ll IoDMGOOSHw mmcflmmaoommomflao mumnnmonmum . . opmuomqnl. mu95531....opmnopamumomHEDmsmmDamianomOOSHw 1T omoosfioll ommcomoupwzoo mmouomq / omouomq ommuocq 31 particles progressively dissolve as the pH is lowered until, at the isoelectric point of pH 4.6 to 4.7, the casein is essentially free of bound salts (Jenness and Patton, 1976). Protein Fermentation Protein breakdown to amino acids is low in the manufacture of yogurt. Glutamic acid and proline are the two amino acids found in the highest concentration in yogurt and methionine and phenylalanine are found in the lowest amount (Rasic et a1., 1971). Ghadini and Pecosa (1963) pointed out that a considerable variation in the amino acids content of yogurt occur and that storage may influence the amino acids content. As in Figure 3, proteins are degraded by micro- bial enzymes to yield peptides and amino acids. These amino acids can undergo a variety of changes, such as side chain alteration, decarboxylation, transamination and oxidative deamination to alpha-keto acids. They may also react with sugars, through a nonenzymatic browning reaction to produce flavor compounds (Kilara and Shahani, 1978). Groux (1976) analyzed the amino acids content of yogurt and reported that the quantities of each of these 32 A. Proteins—————————-—~—*—Peptides + Amino acids B. B-l Side chain alteration Tryptophan = Indole B-2 Decarboxylation Lysine 4>»Cadaverine B—3 Transformation to alpha keto acids Asparato =:Oxaloacetate B-4 Oxidative deamination to alpha keto acids Glutamate__________..Alpha keto glutarate C. Amino Acids non enzymatic brownlng-»Flavor compounds Figure 3.~-Mechanisms for the degradation of proteins. amino acids were not in relation with their abundance in the milk protein. In general, the amount of each free amino acid increased with the fermentation, whereas two exceptions were observed: glycine and valine. He also suggested that these two amino acids could be important growth factor for the yogurt culture. He could not detect measurable cystine and arginine in the milk or in the yogurt. He concluded that the absence of these two free amino acids remains obscure. Lactobacillus bulgaricus is much more proteolytic than Streptococcus thermophilus; however Streptococcus thermophilus is capable of splitting urea while Lacto- bacillus bulgaricus cannot (Rasic and Kurmann, 1978). 33 The amino acid content of yogurt then depends on several factors, such as mix composition, heat treatment, incubation time and temperature, rate of cocci to rods, age and storage conditions and the culture itself. Lipid Metabolism Lipase activity of yogurt cultures is low. Chandan et a1. (1969) compared the lipase activity of several lactic cultures and found that it varied among the differ- ent microorganisms studied. The lipase activity may be more important for strains differences than for the species differences. Rasic et a1. (1973) reported that some fatty acids increases whereas others decrease, according to the initial yogurt mix and conditions used. Saturated fatty acid concentration increased as compared to the initial milk used, except stearic acid which decreases. Oleic, linoleic and palmitoleic acid concentration decrease slightly during fermentation (Tennous and Merat, 1969). The free fatty acids may be converted to Beta-keto acids and acetoacetate. The former may be metabolized further to methyl ketones whereas the latter may be con- verted to Acetyl-CoA and amino acids or to acetone. Some free fatty acids may also be esterified (Harper and Kristorffersen, 1956). Figure 4 represents some of the reactions of the metabolism of lipids. .Awhaav Hamsmnm can MHMHHM Eoumx *.mcflmfla mo EmHHODMDoE can now msmflcmcoozll.w ousmflm mumpmm moEODoM ahcpoz occpoom : oaoummoouoml lopomI lllIIIIY outpoo owoo<.4|||llll|¢oo proo< mpflom muumm comm UAHSHQoU OHOHQMO oaumusm mpflom ocflfifl . . . . mopHM®OWHmHHB msflmpoum 35 Vitamin Metabolism Vitamins are required for growth and multiplica— tion of bacteria. Some vitamins are consumed during yogurt fermentation while others are biosynthesized, depending on the species and strains characteristics of the bacteria as well as on the manufacturing conditions. For instance, severe heat treatment decreases the content of all heat labile vitamins. Acott and Labuza (1972) reported that yogurt is not as rich in B vitamins as the original milk because yogurt bacteria are users and not producers of B vitamins. Vitamin C is 70 to 80 percent lower in plain yogurt because of the prolonged heating of yogurt during pasteuri- zation. They also found a decrease of as much as 95 per- cent of choline while Gulko and Kruglova (1966) reported an increase of choline content from 15 to 161 percent. Blanc (1966) reported that niacin was stable during the heat treatment, decreased during the first stage of incubation and then increased to a concentration higher than the original yogurt mix. Reddy et a1. (1976) found that the biotin content progressively decreases during yogurt manufacture. They also reported an increase of more than 1,000 percent of folic acid during incubation. 36 Pantothenic acid is heat stable but it suffers a slight decrease in concentration during yogurt preparation (Gulco and Kryglova, 1966). Vitamin A constantly decreased in yogurt, even during storage. After 5 days of storage the decrease may be from 60 to 90 percent (Randoin and Causeret, 1956). In summary, during yogurt manufacture, vitamin C, vitamin A, thiamin, riboflavin, vitamins B and B and 12 6' pantothenic acid contents decrease, while folic acid and choline contents increase. During storage of yogurt, vitamin A, vitamin B folic acid, and vitamin C pro- 12’ gressively decrease. Components of Yogurt Flavor Yogurt flavor may be defined as the sensory qual- ity' of yogurt that results from the components contribut- ing to taste and odor. Background Flavor of Yogurt Although the predominant flavors of yogurt are the result of metabolism of substrates by the bacterial culture, an additional significant contribution to the overall flavor is made by the normal milk components and the effect of heat treatment on these components. Extended high temperature treatment of milk produces a substantial number of compounds including furfuryl alcohol, 37 furfural, hydroxymethylfurfural, maltol, acetol, methyl— glyoxal and acetaldehyde; butyric, propionic, acetic, formic, lactic, and pyruvic acids, ammonia, hydrogen sulfide, and carbon dioxide (Jenness and Patton, 1976). The milk flavor cannot be attributed to one com— pound, but rather to a mixture of compounds; whereas one compound will show different flavor characteristics at various concentrations, mixtures of compounds should accentuate this condition. Interestingly, a compound which has been implicated in the characteristic flavor of milk has also been observed in higher concentrations in many of the off-flavors which arise in milk and milk products (Patton et a1., 1956). The common practice of adding nonfat dry milk to milk to make the yogurt mix will also introduce the flavor components of the nonfat dry milk. The presence of C5 to C15 odd numbered carbon methyl ketones and a series of lactones have been demonstrated in various concentrations during studies concerned with the heat treatment and storage stability of dry milk products (Patton, 1961; Keeney and Patton, 1956; Parks and Patton, 1961; Muck et a1., 1963). Parlement and Nawar (1965) extended the number of lactones identified during heat treatment, while Park et a1. (1965) identified o-aminoacetophenone. In a more recent study Viani and Horman (1976) identified some compounds in heated milk which have never 38 been reported in the literature. Figure 5 shows the results of their work, using a combined gas chromatography- mass spectrometry (GC—MS) technique. They extracted the aroma constituents of heated milk (BO—90°C for 30 minutes) by stripping with argon at reduced pressure and low temperature, and then injected the extract in the GC- MS. This sampling condition, however, does change the balance of the milk aroma and quantitative work is not possible. Goerner et a1. (1968) identified acetone, ethanol, butanone-Z, and traces of diacetyl, as the major compon— ents of the head-space gas above skim milk "cooked" for 5 minutes. Yogurt Flavor Lactic acid, the major product of the metabolism of lactose by yogurt bacteria, is nonvolatile and odor— less, and therefore, it does not contribute to yogurt aroma. However, the acid tasts of yogurt is largely due to the presence of lactic acid (Lindsay, 1967). Several authors have accepted that the character- istic flavor of yogurt comes from lactic acid, acetaldehyde, diacetyl, acetone and acetic acid (Casalis, 1975; Kroger, 1973; Sandine et a1., 1972). Pette and Lolkema (1950c) were the first research- ers to recognize that acetaldehyde is the main component 39 Compound ‘ Precursor From Fat Acetone Butanone 3—Pentene—2—one 2—Hexanone Ketoacids 2-Heptanone 2-Nonanone 2—Undecanone Gamma Valerolactone Delta Caprolactone Delta Caprilactone Delta Tridecalactone Hydroxy Acids Pentane Methyl Cyclopentane Acetylpropionyl 2-Hydroxy—3—pentanone 3-Hydroxy—2-pentanone From Fat Benzaldehyde or Lactose Benzyl Alcohol Methyl Benzoate From Lactose Furfural Furfuryl alcohol 5—Methylfurfural Furyl methylketone 2—5-Dimethy1 furane 2—Furyl-3—propional Furylethylketone 2—Pentylfurane From Proteins Dimethyl sulfide Methionine Dimethyl sulfone Isobutyraldehyde Valine Phenyl acetaldehyde Phenylalanine Figure 5.—-Compounds formed by thermal degradation of yogurt mix during heat treatment.* *Adapted from Viani and Horman (1976). 40 of yogurt flavor, other than lactic acid. Hamdan et a1. (1971) reported that acetaldehyde is produced mainly in the second stage of fermentation by the Lactobacillus bulgaricus. Yogurt made exclusively with Streptococcus thermophilus had low flavor score and low acetaldehyde content. The rate of acetaldehyde production is dependent on the acidity. Bottazzi et a1. (1973) reported that acetaldehyde begins to be produced appreciably only when the pH is about 5.0 and then increases rapidly as the pH decreases to 4.4 - 4.3, thereafter increase little and finally stabilizing at a pH around 4.0. Bottazzi and Dellaglio (1967) compared several strains of Streptococcus thermophilus and found that most of them produced small quantities of acetaldehyde while one strain produced intermediate quantities, as compared with the acetaldehyde content of yogurt. Keenan and Bills (1968) suggested that relatively high concentrations of acetaldehyde are necessary to produce a desirable flavor in yogurt. Gorner, et a1., (1968) found that ripening of yogurt was characterized by the appearance of acetaldehyde during the first and second hour of incubation, rising to between 23 to 55 ppm in ripe yogurt. 41 Bottazzi and Vescovo (1969) compared the acetalde- hyde content with flavor of yogurt made from several Lactobacillus bulgaricus strains and concluded that yogurt with a weak flavor has less than 4.0 ppm of acetaldehyde, while good yogurt flavor was present when the acetalde— hyde content was greater than 8 ppm. They further reported that the normal acetaldehyde content of good flavored yogurt was 15 to 20 ppm. Hamdan et a1. (1971) reported levels of 22 to 26 ppm of acetaldehyde in natural yogurt; however, the con- centration decreased with time for two yogurt cultures used, while a third one showed no change. Rasic and Kur- mann(l978) also suggested a decrease of the acetaldehyde content during storage of yogurt, beginning five hours after manufacture is completed. However, they also men- tioned that other authors did not find any change after five days. They then concluded that there are consider- lable differences among cultures in their ability to reduce acetaldehyde during storage. Kroger (1976) claimed that 90 percent of the carbonyl compounds present in yogurt was acetaldehyde. Diacetyl and acetoin have been suggested as being part of yogurt flavor (Bottazzi and Dellaglio, 1967; Groux, 1973). However, some others have reported that the contribution of diacetyl and acetoin is minor (Gorner et a1., 1968; 42 Casalis, 1975; Gorner et a1., 1975). Lindsay and Day (1965) reported an important relationship between the concentration of acetaldehyde and diacetyl (l to 4 respectively) to produce a good flavored yogurt. However, data from Hild (1979) do not support this theory. El—Sadec et a1. (1972) reported that diacetyl has a negli— gible role on zabady (an Egyptian fermented milk similar to yogurt) flavor. Other carbonyl compounds have been reported in yogurt aroma, such as acetone, 2-butanone and others; they are considered of limited importance to the yogurt flavor (Rasic and Kurman, 1978). Gorner et a1. (1968) reported that acetone, 2-butanone and ethanol are produced during heat treatment and their concentration did not change significantly during fermentation while acetaldehyde suffers a very high increase in concentration. Bottazzi and Vescovo (1969) suggested that yogurt bacteria are capable of producing acetone in variable amounts. They also reported that the acetaldehyde/acetone ratio is important to yogurt flavor. A ratio of 0.4 to 1.0 pro- duced a slightly atypical flavor, while a ratio of 2.8 produced a good flavored yogurt. Groux (1976) reported that typical yogurt flavor was obtained after the removal of most of the acetaldehyde. They suggested that the remaining diacetyl and perhaps 43 acetoin were sufficient for maintainingaitypical flavor when the acetaldehyde content was very low. Moinas (1976) developed an aroma volatiles extrac- tion method in which compounds were extracted under vacuum and at low temperatures. Then the volatiles were concen- trated and identified by a GC-MS. He identified several components of yogurt aroma never reported before such as tridecanal, octadecane, propylphenyl acetate, butyl benzoate, gamma octalactone, and trans dodecene-Z-al. However, no correlation with flavor was made, but since the concentration of these compounds is very low, it would not be expected that they would contribute to the yogurt flavor. Groux (1976) reported that free amino acids did not contribute to yogurt flavor, however, they suggested that amino acids might be precursors for some important volatiles present in yogurt. Some fatty acids increase during fermentation as compared with the original yogurt mix. Acetic acid is produced in the highest amount followed by formic, caproic, caprylic, capric, butyric, propionic and isovaleric acids (Rasic et a1., 1973; Schormuller and Langner, 1960). Other organic acids which increase significantly but contribute little to yogurt flavor are: succinic, fumaric and pyruvic acids (Rasic and Kurmann, 1978). 44 Thus, in addition to controversy about the secon— dary components of yogurt flavor most researchers seem to agree that acetaldehyde and lactic acid are indeed the main components of yogurt flavor. Nutritional Properties of Yogurt Milk has been accepted universally as an almost complete food for the feeding of normal infants. The need for many of the nutrients in milk continues through adult— hood and old age. Qualitatively, milk contains practi- cally all the nutrients for growth and maintenances of life, particularly for the young infants. While some nutrients such as iron, copper, manganese, and ascorbic acid are not present in significant amounts or in pro- portions adequate to meet the needs of older children and adults, milk has been an integral component for the feed- ing of infants starting at age four to six months (Lam- pert, 1975). Transforming milk into various dairy foods usually results in a change in the composition of the resulting product, depending on what components has been either removed, concentrated, or alterated. In the United States, while the composition of most dairy foods is regulated by federal and state standards of identity (Anonymous, 1974a),the sanitary 45 quality of milk is regulated by the U. S. Public Health Service contained in the "Grade 'A' Pasteurized Milk" Ordinance (Anonymous, 1978). Yogurt as a Food Dairy products other than butter are known for their important contributions of protein, riboflavin, vitamin B niacin equivalents, and minerals, particu- 12’ larly calcium, phosphorous and magnesium, to the human diet (Anonymous, 1972). Acott and Labuza (1972) reported that with the exception of nicotinic acid, the levels of the B vitamins, ascorbic acid and vitamin A decrease dur- ing manufacture. The decrease was attributed to the micro- flora which utilized the vitamins for their own metabolic requirements. Although the vitamin content of yogurt is inferior to fresh whole milk, the protein and calcium content are greater due to the addition of solids not fat at the formulation stage. It has been assumed that yogurt has a better acceptance than milk by lactose intolerant individuals (Hurt, 1972; Hersh, 1972). Lactose Intolerance Lactose intolerance occurs in some people, espe- cially blacks, orientals, and some caucasians, due to a low lactase activity (less than 2 units of enzyme activity per gram of intestinal mucosa) (Anonymous, 1971). The 46 deficiency of lactase activity slow down the hydrolysis of lactose, which will pass into the lower small intes— tine and the colon where it provides a medium for micro— bial fermentation. Anaerobic microbial growth results in excess lactic acid which irritates the mucosa and accu— mulates large quantities of fluid to maintain isotonicity of the undigested lactose. The combination of fluid and acidic irritation and gas, results in abdominal cramps and pain, hypermotility, bloating, and diarrhea. These sym— ptoms occur after 1 to 4 hours after ingestion of lactose (Huang and Bayless, 1968). Some investigators have recommended yogurt as an alternative means of obtaining milk nutrients without incurring problems with lactose intolerance, due to a partial hydrolysis during manufacture and mainly due to the remaining lactase in yogurt (Gallagher et a1., 1974). The greatest lactase activity was observed in freshly prepared yogurt (Kilara and Shahani, 1974). Rand (1973) reported that only 10 to 20 percent of the lactose content of yogurt is metabolized. The actual residual content depends to a great extent upon how much extra solids not fat is added to the basic mix. The final concentration of lactose may range from 4.3 percent (Shanley, 1973) to 7.8 percent (Acott and Labuzza, 1972). 47 However, Kilara and Shahani (1974) reported that microbial lactase remains active after culturing and may help during human metabolism of the remaining lactose. Gray (1967) reported that certain lactases exhibit different affinities for lactose. Goodenough (1975) found that the lactase activity of intestinal contents and mucosa exhibited the greater activity in yitrg, when rats are fed natural yogurt, while rats fed pasteurized yogurt with nonviable microflora exhibited the lowest lactase activity. In addition, he demonstrated that at least transient survival of the yogurt microflora occurs in the upper gastro intestinal tract of rats after 3 hours after consumption of yogurt. Therapeutical Properties of Yogurt It is apparent that to derive any healthful bene- fits from yogurt bacteria in the diet, the microorganisms must survive the digestive process, implant in the intes- tinal tract, and produce metabolic products and/or anti- microbial agents active against enteropathogenic, putre- factive, and proteolytic microorganisms (Mikolajcik and Hamdan, 1975a and 1975b). Hawley et a1. (1959) reviewed various factors necessary for successful implantation of lactobacilli in the human intestinal tract. The two more important requirements were: (a) ingestion of a large 48 number of viable cells and (b) the availability of a read- ily fermentable carbohydrate for cell growth. Sandine et a1. (1972) reviewed the therapeutic benefits of lactobacilli and concluded that Lactobacillus acidophilus rather than Lacbocillus bulgaricus is implanted in the human intestinal tract. Lactobacillus bulgaricus exhibited a lower tolerance for bile salts, a higher growth temperature, less resistance to acid pH, and a more selective requirement of sugar than Lactobacillus acido- philus (Acott and Labuza, 1972). Lactobacillus bifidicus is also known to be naturally present in the large intes- tine and can be established in it (Damrau, 1954). However, studies have shown that yogurt, as a result of its ability to restore the normal lactic flora of the intestine while inhibiting undesirable proteolytic microorganisms, was found to be superior to an antibiotic (Neomycin-Kao-pectate) preparation in the treatment of children with infantile diarrhea (Yaziciogly and Yilmaz, 1966). While many investigators (Reddy and Shahani, 1971; Muralidhara et a1., 1972; Ehrlich, 1963; Singh, et a1., 1979) have found that isolated organisms used in cultured dairy products are active against various gram—positive and gram—negative microorganisms, others (Donaldson, 1964; Haenel, 1970; Paul and Haskins, 1972) question if they would indeed be able to change the microflora. 49 Humphrey and Plunkett (1969) and Acott and Labuza (1972) reported that it is not likely that Streptococcus thermophilus would implant in the intestinal tract because it is not tolerant to very acidic conditions. Lactobacillus bulgaricus, however, may reach the small intestine and could grow. Although a positive therapeutic value of yogurt has been suggested, clinical research should be undertaken to clarify and resolve the therapeutic properties of yogurt. MATERIALS AND METHOD S Yogurt Manufacture Culture Concentrated deep-frozen culture (Chr. Hansen AY3) cans were obtained and stored at -30°C until used. 1 Culture Medium Spray-dried nonfat dry milk was reconstituted to 10.0 percent (w/w), dispensed in 300 ml glass bottles and sterilized at 15 psig (121.1°C) for 10 minutes. The sterilized culture medium was then kept refrigerated (l-3°C) until used. Bulk Starter Preparation The concentrated deep-frozen culture was allowed to thaw and a portion (0.22 percent w/w) was aseptically transferred to the cold sterilized culture medium. The inoculated milk was then held in a constant temperature water bath at 43.9°C (lll.0°F) until the desired pH was attained. A second inoculated culture medium was used to check the pH so that the bulk starter was not disturbed. The bulk starter culture so obtained was then ready for the final transfer to the yogurt mix and was used imme- diately, without cooling. 50 51 Yogurt mix.-—Milk obtained from the Michigan State University Dairy Barn was standardized with spray dried nonfat dry milk and water to obtain yogurt mix with 3.25 percent fat and 12 percent (w/w) nonfat milk solids. The mixture was batch pasteurized at 87.8°C (190°F) for 30 minutes in 10 gallon stainless steel milk cans, cooled rapidly to 65.5°C (150°F) in an ice water bath and homog- enized in a two-stage Gaulin BW—40 Homogenizer at pres- 2 (500 sures of 140.6 kg/cm2 (2000 psig) and 35.2 kg/cm psig) on the first and second stages respectively. The homogenized, pasteurized product was then termed "yogurt mix." Yogurt preparation.—-The yogurt mix was cooled to 43.9°C (lll.O°F) and inoculated with 2.0 percent (w/w) of: the bulk starter culture previously prepared. The inocu- lated yogurt mix was well mixed and then transferred to glass beakers (1500 m1) covered with two layers of poly— ethylene film and kept in the 43.9°C (111°F) water bath. After the desired pH was attained, the yogurt was placed in a refrigerator at 2°C (Figure 6). Sampling.--Samples were withdrawn every half hour for flavor analysis and every hour for carbohydrate analy- sis. The disturbed inoculated yogurt mix, after sampling was discarded and a new, undisturbed inoculated yogurt Nonfat dry milk Milk 52 Water 7“\\\\\\‘~£:::”/,,,,/”’ Blending 3.25% Fat Mix 12% MSNF l Pasteurization 87.8°C (190°F) 30 minutes ) Cooling 65.6°C (150°F) l Nonfat dry milk W ter a (NFDM) Frozen Culture Homogenization lst stage--l40.6 kg/cm2(2000 psi) 2nd stage--35.2 kg/cm2(500 psi) Cooling to 43.9°C (111°F) Inoculation 2% (w/w) Incubation 43.9°C 1) Cooling (2°C) V Cold Storage 2°C 28 Days Bulk /(/// Starter Culture \/ Blending 10% MSNF V Sterilization 121.1°C (250°F) for 10 minutes 1 Refrigeration 2°C \ Inoculation 0.22% (W/V) Incubation 43.9°C (111°F) / Figure 6.-—Flow diagram of yogurt making. 53 mix was used for each sampling. Samples were also obtained from the unpasteurized yogurt mix, of the yogurt mix, and of the culture. Every time one sample was taken, the pH of the remaining product was measured. Samples of the yogurt were also obtained after storage for 1, 7, 14, 21 and 28 days for flavor and carbohydrate analysis. Flavor Analysis Gas Chromatography The aroma volatiles of yogurt samples were analyzed using a Hewlet Packard gas chromatograph 584A, equipped with electronic integrator and microprocessor. Three columns in tandem were necessary for proper resolu— tion. These columns were: V 1. A 7 ft long 15% DEGS (on Chromosorb AW) 2. A 7 ft long 10% DEGS (on Chromosorb AW) 3. An 8 ft long 10% CarbowaX 20M (on Chromosorb AW) The separated compounds were then detected by a flame ionization detector (FID). A reference column was required due to the tem— perature programming and the low attenuation used. The reference column was an 8 ft long 10 percent Carbowax 20M on Chronosorb AW. The flow rate of the reference column was optimized so that a straight base line was obtained during the analysis. 54 The conditions used for the gas chromatographic analysis were optimized and are listed as follows: Initial temperature (T1) = 80°C Time at T1 = 1.0 minute Rate = 3°C per minute Final temperature (T2) = 95°C Time at T2 = 1 minute Injection port temperature = 110°C Flame Ionization Detector temperature = 250°C Chart speed = 1 cm/min Attenuation = 3 Slope sensitivity = 0.02 (or modified as necessary) Flow in the column = 20.0 ml/minute Flow in the reference column = 8.9 ml/minute Nitrogen pressure = 50 psig Air pressure = 24 psig Hydrogen pressure = 16 psig Head-Space Analysis From each of the yogurt samples (obtained either during incubation or during storage) 5.0 9 portions were removed using a g1ass;tube. The samples were then placed in 50 m1 Hypo-VialTM (Pierce) flasks, which contained 5.0 ml of a 20 percent (w/w) solution of sulfuric acid. The flasks were purged briefly under a stream of helium 55 covered with tuf—Bond DiscsTM (Pierce: Teflon bonded to silicone) and aluminum sealed using a hand crimper. Then the flasks were placed in a gyratory water bath with a constant temperature of 60°C, moving at 204 rpm. After 10 minutes and 30 minutes, 2.0 ml of the head-space of the warm yogurt samples were removed with a 2.5 m1 Hamil— ton gas tight syringe and injected immediately onto the column. The syringes used were placed in a convection oven at 100°C for at least 20 minutes before being reused. The 50 m1 Hypo-vialTM flasks were kept in a convection oven at 100°C for at least 24 hours before being used and then they were cooled under a stream of helium. All analyses were in duplicate. Equilibrium Analysis Samples of head-space gas of yogurt were obtained after being in the gyratory water bath for 2, 3, 5 and every 5 minutes thereafter until 40 minutes. An equi- librium curve was then constructed for each component of the head-space gas. Flavor Survey Additional samples of plain yogurt were purchased from three local supermarkets. They were submitted to the treatment already described for head-space sampling and then anayzed by the gas chromatograph. This survey was 56 done to analyze qualitatively and quantitatively the com- ponents of the yogurts. Sample Preparation for the Gas Chromatographic-Mass spec— trometric Analysis Samples (2.0 to 4.0 m1) obtained from the head space gas, after being submitted to the conditions described, were injected into the gas chromotographic- mass spectrometric (GC-MS) system. However, some com- pounds were at low concentration and consequently, a liquid nitrogen trap of the aroma of yogurt was made to have a better response from the MS. A helium stream (5 ml/min) was used as carrier of the aroma from a 3Kg sample of yogurt in a sealed vessel, which was kept in a water bath at 60°C, the the nitrogen trap. 1‘]. The trap consisted of a 125 m1 Hypo—viall with the bottom inlet of porous porcelain and the outlet at the top of the flask. After collecting the aroma volatiles for 3 hours, the flask was sealed with the Tuf—BondTM disc, removed from the liquid nitrogen, placed in an acetone dry ice bath for 15 minutes, for transportation to the GC-MS, warmed to 60°C for 30 minutes and then 2.0 m1 of the head space gas injected into the GC—MS system. 57 Gas Chromatographic—Mass Spectrometry Analysis The GC—MS system used was a Hewlett Packard (HP) 5840 A gas chromatograph connected to a HP 5985 mass spectrometer by a jet separator. The same column system used for the head-space analysis was used in the GC-MS. Helium was used as the carrier gas, instead of nitrogen, due to the requirements ofthe GC-MS system for the jet separator. The operational conditions had to be modified in order to have satisfactory resolution of the aroma components. Preliminary work indicated that the initial temperature should be 70°C, and the final temperature 95°C, at a rate of 3°C per minute. The initial temperature was held for 1 minute. The injection port temperature was 110°C. The Ion source of the MS was set to 200°C and the electron multiplier voltage was 2600 volts. Approximate scan time was 400 scans per minute. The mass window used was 40 to 300, unless otherwise noted. Analysis of Standard Compounds From preliminary observations, the range of area for each component of the yogurt aroma was determined. Analytical grade standards of acetaldehyde, acetone, ethanol and diacetyl were added to a portion of yogurt mix separated just before inoculation. Each standard com- pound was added in a way that it would increase the area 58 in the range observed during preliminary work. Appro— priate dilutions were made in the yogurt mix and a volume of 2.0 ml of the head-space gas was injected onto the column after being in the gyratory water bath (204 rpm, 60°C) for 30 minutes. Therefore, standard curves were made over the whole range of variation of each component excluding the initial concentration (area) of the com- ponents which were present in the yogurt mix. Carbohydrate Analysis by High Performance Liquid Chromatography Apparatus The HPLC system was assembled from a Waters model 6000 pump, and a Waters Model RI-40l differential refrac- tometer detector and a Waters prepacked uBondapak/ CarbohydrateR column (4.2 mm i.d, x 30 cm). Samples were loaded onto the column with a Waters U6K septumless injector containing a 2 m1 sample loading loop. Volumes injected ranged from 1-10 ul. Dectector attenuation was held constant at 8 X and a Linear Instruments Model 232 recorder was used for peak measurement. Mobile Phase Isocratic separations were achieved with the mobile phase acetonitrile/water (75/25 v/v), non-spectro grade acetonitrile (Burdick and Jackson, Muskegon, MI); 59 reverse—osmosis, ion—exchanged, deaminated water. The flow rate was adjusted to deliver 2.0 ml/minute. Deionized degassed water was passed through the reference valve of the pump every morning or whenever necessary to remove any gas bubbles present so that a steady baseline resulted. Standards Carbohydrates (analytical reagent grade) were dried in yagug at 65°C for 24 hours, and standard solu— tions [1 percent (w/w) glucose, 1 percent (w/w) galactose and 5 percent w/w lactose] prepared. To 2.0 ml of the glucose and galactose solutions was added enough anhydrous trichloroacetic acid (TCA) to produce a 6 percent (w/v) solution of the acid. Before injection all solutions were filtered through a 0.45 um Metricel membrane filter (Gelman Filtration Products, Ann Arbor, MI.). Peak height measurements were used in quantifying these compounds. A linear regression equation between peak height and concentration was established for each compound. Recovery experiments were conducted by adding known quantities of standards to an yogurt sample and assaying the sample before and after the addition. All quantitative determinations were made in duplicate. 60 Sample Preparation Samples weighing 50.00 g were obtained from - unpasteurized yogurt mix, yogurt mix, inoculated yogurt mix (0 hour) and thereafter every hour of incubation until the pH 4.25 was achieved. They were covered with suffi- cient anhydrous ethanol (100 percent A.R.) to make the final concentration of ethanol 50 percent. The samples were then well mixed and refrigerated (1°C) until used. They were thawed, warmed to about 50°C and blended in a Waring blender for 2 to 3 minutes. The extract was fil— tered through Whatman No. 54 paper, the residue washed with 50.0 ml of a 50 percent ethanol solution. The pre- ciptate was placed in 100 ml of 50 percent ethanol and refiltered. The extract and washings were reduced to a volume of about 4—5 ml using a rotary vacuum evaporator. Portions of warm distilled water were used to remove the sugars from the evaporator flask. The final volume of the evaporated extract was 25.0 ml. The extract was then filtered through Whatman 42 paper. The extract thus obtained will be named "sugar extract" from hereon (Figure 7). Samples for Lactose Analysis To 1.0 ml of the sugar extract was added 1.0 m1 of distilled water. Then the solution was passed through a Waters C~18 Sep—PakR to remove the last traces of 61 SAMPLE (50.00g) Add Anhydrous Ethanol 50% Ethanol i (stored refrigerated) Blend in waring blender for 2.3 minutes 1 Filter (Whatman 54) 1 Wash with 50% Ethanol 1 Remove precipitate and extract again with 50% ethanol 1 Filter (Whatman 54) l Concentrate (rotary vacuum evaporation at temp below 40°C) 1 Dilute to 25.0 ml with distilled water 1 Filter (Whatman 42) 1 Sugar extract Figure 7.-—Flow diagram to extract carbohydrates from yogurt samples. 62 lipids. Just before injection onto the column it was filtered through a 0.45 um Metricel membrane filter. Volumes ranging from 1 to 5 ul of the prepared sample were injected onto the column (Figure 8). All quanti- tative determinations were performed in duplicate (two aliquots from the same sample). Samples for Glucose and Galactose Analysis First part.—-The sugar extract (2.0 ml) was passed through a Waters C-18 Sep—PakR. Enough TCA was added to produce a 6 percent (w/v) solution. Then it was filtered through a 0.45 um Metricel membrane filter just before injection onto the column. Volumes ranging from 1 to 10 ul were injected onto the column (Figure 8). Height of the peak consisting of glucose plus galactose was recorded. Second part.—-The sugar extract was passed through a Waters C-lS-Sep—PakR. Then a 1.0 ml aliquot was mixed with 1.0 m1 of acetate buffer pH 5.1 (0.05M) at 35°C containing 100 units of glucose oxidase and 500 units of catalase. The mixture was placed in a gyratory water bath at 100 rpm and 35°C for 8 hours. Then was added enough anhydrous TCA to produce a solution of 6 percent (w/v) the acid. This was filtered through a 0.45 um Metricel membrane filter just before injection. 63 .omouoma pun omowomamm .omoosam mo mflmwamcm osu Mow memEMm mo coflpmummmum may mo Emummap 30am I:.m ousmflh AH: OHIHV qum A; AocmuDEoE E: mv.ov umpaflm A. A>\3v unmoumm m OD pace OHDQOMOHOHSUHHD cp< + undo: m How 00mm um QDMQ Hopes muonmuhw A, wufl>fluom ommHMDmo mo open: com pan >DH>HDOM omcpflxo omOOSHm mo hues: ooa mansnmncoo comm pm is: mnav H.m mm Hmmmsn oucuooc oqmm z mo.o HE o.H ppm a nosaflam as o.H m on mcmnnfiwz E: A ma.o umuafla A, xmmlmom mHIO xmmlmom mHO n nmsounu mmmm m Smsoucu mmmm 1 omODOMme mafia omoosam mo mflmhamcm Domuwxm omODOMHmm mo mflmhfimcm AH: mnav oqmm L AocmuQEOE E1 mv.ov umuaflm A. amalgam mac n nmsounu mmmm + and on noun: LDHB opsaflo omouoma mo mflmmamc< Hmmsm 64 Height of the galactose peak was recorded. Volumes rang- ing from 1 to 10 ul were injected onto the column (Figure 8). Glucose Oxidation by Glucose Oxidase The activity of the glucose oxidase, B—D—glucose: oxygen l-oxidoreductase EC NO. 1.1.3.4, was determined using the procedure suggested by Sigma Chemicals Co. (1979). Unity definition.--One unit of glucose oxidase will oxidize 1.0 u mole of B-D—glucose to D-gluconic acid and hydrogen peroxide per minute at pH 5.1 at 35°C. Glucose oxidase VII was purchased from Sigma and stored dessicated and below 0°C until used. The procedure for the activity determination was as follows: Into a silica cuvette (1 cm light path) was pipetted the following: 2.40 ml of o-Dianisidine solution (0.0021M in 0.05 acetate buffer pH 5.1 at 35°C 0.50 ml of B—D-glucose solution (10 percent w/v) 0.10 ml of Peroxidase (> 60 Purpurogallin units/m1 The above reagents were mixed and equilibrated to 35°C. The absorption at 500 nm versus air was monitored 65 by a Beckman DU-2 spectrophotometer until steady. Then at zero time the following reagent was added: 0.10 ml of glucose oxidase containing approximately 0.54 unit/m1 in 0.05 acetate buffer pH 5.1 at 35°C The solutions were quickly mixed and the increase in absorption at 500 nm versus air was recorded at 500 nm for 4.5 minutes. A plot of absorption at 500 nm versus time was made and the glucose oxidase activity was cal— culated from the maximum linear rate using the formula u Molar units = (A 500nm/minute) X 3'1 gram solids 7.5 x mg sample/reaction mix After the activity was determined, several test tubes containing 1.0 ml of a 1.8 percent (w/v) glucose plus 1.0 m1 of 0.05 M acetate buffer, pH 5.1 at 35°C, containing 100 units of glucose oxidase and 400 units of catalase were placed in the gyratory water bath at 100 rpm at 35°C. Two test tubes were removed after 1, 2.5, 5,6.5 and 8 hours and enough anhydrous TCA was added to produce a 6 percent solution. Following filtration through a 0.45 um Metricel membrane filter the samples were injected onto the column. An equilibrium curve was constructed to ascertain the time required to oxidize all of the glucose to gluconic acid. 66 The same procedure was used to analyze the recovery of glucose and galactose in a mixture. RESULTS AND DISCUSSION Analysis of Carbohydrates in Yogurt by High Performance Liquid Chromatography (HPLC) High performance liquid chromatography (HPLC) has been proposed as a rapid and accurate procedure for the determination of carbohydrates in foods (Linden and Lawhead, 1975). Chemically bonded stationary phase carbohydrate-type columns (on microparticulate support) have been extensively used for the determination of mono and oligosaccharides, and some substituted carbohydrates (Conrad and Palmer, 1976; Brandao et al., 1980a, and 1980b; Dunmire and Otto, 1979). However, the analysis of certain dairy foods, specifically those containing glucose and galactose, has proved difficult. Warthesen and Kramer (1979) using a Waters u-Bondapak/carbohydrate column reported the analysis of glucose in ice cream mix; however, they did not differ- entiate between glucose and galactose, which are eluted together. Hurst, et a1. (1979) also failed to differen- tiate glucose from galactose in HPLC analysis of some dairy products such as ice cream and flavored yogurt. Fluid milk has about 11.7 mg/100 ml of galactose and about 13.8 mg/100 ml of glucose (Reineccius, et a1., 1979). 67 68 However, dairy products in which milk solids not fat, is added or which were fermented by lactase-producing bacteria are expected to have a higher concentration of these two sugars. For the analysis of glucose and galactose in yogurt, a method was developed in which glucose was oxidized by glucose oxidase and the galactose analyzed by HPLC. Knowing the galactose concentration, the height of the peak containing glucose and galactose together and the equations relating peak height to concen- tration of each sugar, the glucose content may be calcu— lated by simple mathematics. The HPLC Method The analysis of carbohydrates in milk by HPLC poses some problems, i.e., the presence of organic and inorganic salts which lead to a peak appearing just after the solvent front (water) and another small one after a few minutes. The second one may coincide with the glucose peak or may be eliminated depending on the mobile phase used. Another important factor observed is the condition of the column; older columns tend to show somewhat differ— ent peak(s) for salts. Glucose and galactose are practically nonresolv- able by Bonded-phase carbohydrate-type columns. Mobile phases with extremely low polarities, i.e., acetonitrile/ water 95/5 (v/v) were tried without success. The 69 connection of two columns in tandem also proved unsatis- factory for the resolution of these two sugars. There— fore, it was found that the only solution for the analysis of both glucose and galactose, by this type of carbohy— drate column is the elimination of one of the sugars by some type of reaction. The commercial availability of glucose oxidase with high activities, and the low concen— tration of glucose in yogurt made this sugar the one selected to be modified, or specifically, oxidized to gluconic acid, thus eliminating its contribution to the galactose peak. An ion exchange column, packed with Aminex A-5 Ca2+ form, has been used to resolve glucose and galactose (Scobell et a1., 1977). However, the analy- sis takes much longer than Bonded-Phase columns, and the column may be irreversibly "poisoned," requiring dis- mantling and unpacking of the column for regeneration. Another disadvantage of ion exchange column is the high temperature at which it must be used (85°C). The data in Figure 9 show thechromatogram devel— oped for the analysis of glucose, galactose, and lactose. The second peak after the solvent front, which would have the same retention time as glucose, was eliminated when using the conditions specified. Acetonitrile/water (80/20) presented the second peak when calcinated milk salts solubilized in distilled water, or a 0.5 percent 70 COLUMN‘WATERS uBONDAPfiK/CARBOHYDRATE MOBILE PHASE'ACETONITRILE/WATER(75/25) FLOW RATE'2.0 ML/MIN. DETECTOR‘R.L LACTOSE fl "fl? GLUCOSE AND/0R GALACTOSE W—T— 0 l 2 3 4 5 TIME, MIN. Figure 9.--HPLC chromatogram of an yogurt sample contain— ing glucose, galactose and lactose. 71 solution of sodium chloride was injected onto the column. Another very important advantage of this high-polarity mobile phase (higher water contents) is the quick elution of the sugars. Analysis of Sugar Standards Results of linear regression calculations of the standards are presented in Table 1. TABLE 1.—-Linear Regression Curves for Glucose, Galactose and Lactose Standards . L ' Correlat1on 1near range Sugar Equation* Coefficient of peak ht. (mm) Glucose y = 0.5363x -0.0173 0.9998 15-160 Galactose y = 0.8263X -O.4008 0.9998 15—160 Lactose y = 0.6112X ~1.5513 0.9992 20—165 *y = ug of the sugar x peak height in mm The samples analyzed for galactose contained 6 percent trichloroacetic acid (TCA) added to precipitate the glucose oxidase. Euber and Brunner (1979) reported that the presence of TCA in the injected sample affected peak heights measurements. Therefore, TCA was added also to the samples analyzed for galactose and glucose 72 together and to the glucose and galactose standards so the same linear regression equations could be used for all analyses. Glucose was calculated using the following equa— tion which is obtained from the subtraction of the height of the galactose peak from the combined glucose and galactose peak: glu(ug) =[H—(1.2088 xgal(ug)) +0.4008] 0.5363 -0.0173 where: Glu (ug) = micrograms of glucose in the volume , of the sample injected ' H = Total peak height of glucose plus galactose Gal (ug) = micrograms of galactose in the volume of the sample injected Recovery of added glucose and galactose (0.5 percent) of each sugar was 93 percent and 95 percent respectively, while the recovery of the addition of l per— cent lactose was 94 percent. Oxidation of Glucose by Glucose Oxidase Results of the oxidation of 1.8 percent (w/v) solution of glucose by 100 units of glucose oxidase is presented in Figure 10. Eight hours were necessary for the complete oxidation of glucose to gluconic acid. 73 $1.8 ‘5 10- -5.o 9. O E . . #91 ,_ 05 ~45 .. 2 S 8 - - _ E s W 7" O W 1 W 0 - v . . - Lfig'v . . . to N PO I 2 3 4i L I 7 I4 2: 28 INCUBATION TIME.HOURS STORAGE TIME . DAYS = Unpasteurized yogurt mix. = Pasteurized yogurt mix = Zero time oruz Figure 11.--The pH and the content of glucose, galactose, and lactose of yogurt during processing, incubation and storage, using the balanced culture. 76 pasteurization process, which may be attributed to brown— ing and/or reactions other than simple hydrolysis to free glucose and galactose. The slight increase of these monosaccharides did not correspond to the observed decrease in lactose con- tent. A constant decrease of the lactose content was observed during fermentation which closely followed the pH. The galactose content increase observed during fermentation does not account for all the galactose pro- duced by the hydrolysis of lactose (30 percent) suggest- ing thet while some galactose is metabolized (13 percent) the rest is found in cultured yogurt (87 percent) as free galactose. A slight increase in the glucose content of yogurt during fermentation was observed. Whereas 86 per— cent of the glucose produced by the hydrolysis of lactose was metabolized, 14 percent was found as free glucose in the finished yogurt. During storage of yogurt for the first day, it was observed a slight change of the carbohydrate content, particularly a decrease of the lactose content and an increase of the galactose content. This change may be attributed to the growth of the yogurt bacteria during cooling as well as a residual lactase activity. Yogurt 77 stored for extended periods still showed residual but low lactase activity. The increase in galactose paralleled the decline in lactose content and glucose decreased slightly during storage. The decrease of lactase activity during storage is probably due to the combination of low pH and low storage temperature (2°C). Engel (1973) reported that the activ— ity of a commercial lactase, isolated from Saccharomyces yeast, added to milk was strongly dependent on the pH and on the temperature. He suggested the use of lactase to hydrolyse lactose to glucose and galactose prior to yogurt culturing in order to sweeten yogurt without increasing calories. Figure 12 shows the changes which occurred during fermentation and storage of yogurt made with the unbal- anced culture. The lactose showed a slower change in concentration, but the total fermentation time was longer and the total amount of lactose metabolized was higher (34 percent) than that of the balanCed culture (30 per- cent). Only 7 percent of the released galactose was metabolized, as compared to 13 percent of the balanced culture while glucose was metabolized to approximately the same extent for both cultures, 87 percent and 86 percent for the unbalanced and the balanced culture, respectively. 78 ’ v a— I 3 70 o 3 3° 2 V O " 2 “I H\_ . :0 8 25‘ tr LACTOSE 55 -I .— g ' GALACTOSE g “ o GLUCOSE , ‘9 zO< 6.08 : H 2 5 P g: 2 w 4 8 I5- ~55 :‘3 o 4 a 5 6’ Vi..._._. z 8 IO -5.o$, ,_ I- < - O 5 MI \ W 45%: Z .- 3 . \ ._ G . 1...: a! 0 ' I I T ' ' r 40\ ON P O I 2 3 4 5 67 E l 7' I4' 2'I 28 S INCUBATION TIME, HOURS STORAGE TIME. ous N = Unpasteurized yogurt mix P = Pasteurized yogurt mix 0 — Zero time Figure 12.--The pH and the content of glucose, galactose and lactose during processing, incubation, and storage of yogurt, using the unbalanced culture. 79 The unbalanced culture also showed lactase activity during cooling and to a lesser extent, during storage. Galactose increased slightly during storage while glucose decreased slightly. These changes during storage may be the result of action of enzymes in the yogurt, since the yogurt bacteria do not grow at tempera— tures below 15°C.(Buchanan and Gibbons, 1974). Kilara and Shahani (1976) reported that lactase activity in yogurt increased with time of incubation, reaching a maximum at a pH around 5.0. They also found that Streptococcus thermophilus contained approximately three times more lactase than did Lactobacillus bulgarius. They observed that sonication had a strong influence in lactase activity, indicating that most of the lactase molecules are bound to the cell wall of the bacteria. Goodenough and Kleyn (1975) reported 32.3 percent reduction of lactose during laboratory controlled fermen- tation of yogurt. The galactose content increased from traces to about 1.2 percent, while the glucose content remained at trace levels. Commercial samples had lactose contents ranging from 3.4 percent to 4.7 percent, galactose from 1.5 percent to 2.5 percent and glucose from 0 percent to 0.12 percent. Euber and Brunner (1979) reported that about 35 percent of lactose was metabolized during the 5.7 hours of incubation necessary to produce a pH 4.5. 80 Davis and McLachland (1974) using a method to cal- culate the original lactose content based on the protein value, reported that lactose hydrolysis of samples obtained in London supermarkets ranged from 25 percent to 60 percent. The HPLC method as presented would have to be modified, by changing the polarity of the mobile phase, in order to anlayze for the presence of other carbohy- drates added during processing, i.e., sucrose and/or fructose and other sugar from fruits added to produce fruit yogurts. The conditions presented for the analysis of glucose, galactose and lactose might be insufficient to completely resolve sucrose and they are definitely incapable of resolving fructose without modification of the procedure. Analysis of the Aroma Components of Yogurt Gas Chromatography of the Yogurt Aroma A typical gas chromatogram depicting the separa— tion of the yogurt aroma volatiles is shown in Figure 13. A peak for the "solvent" also obtained when injecting air into the gas chromatograph (GC), appear as the first peak on the chromatogram. This peak may occur due to a dif— ference in the flow rate of nitrogen when 2.0 ml of gas is rapidly injected onto the column, or it could be also 81 m ACETALDEHYDE "SOLVENT" ACETONE ETHANOL DIACETYL UNKNOWN c)- hD~ g“- 45 (fl 1”“ _q- Figure]3.--Gas chromatogram of head—space gas of inoculated yogurt mix after 1.5 hours of incubation. 82 due to the momentary change of gases going through the flame ionization detector (FID). Air, itself, should not be detected by the FID. Injection of 2.0 m1 of pure nitrogen collected from a nitrogen purged 5.0 liters flask also presented the same peak, but with a lower FID response. Gorner et a1. (1968) suggested that 5 g of yogurt should be added to a 25 m1 flask containing 6 g of anhydrous sodium sulfate, for GC analysis. The sample was then placed in a 60°C moving water bath for 5 minutes. However, Hild (1979) considered the addition of 1 ml of water to 1 g of yogurt was essential, especially with samples containing stabilizers, before heating in the water bath. No equilibrium curves were presented for the components. When duplicating their procedures a peak was observed with a retention time of about 17 minutes, which was analyzed by the GC-MS as a possible polymerization product of the acetaldehyde. Another criticism of their methods in the change of pH, and even change of the con— sistency of the samples during the study of the fermenta— tion of yogurt, i.e., yogurt mix would have a pH around 6.5 and is liquid, while yogurt has a pH around 4.2 and is coagulated. To use their methods for quantitative analysis during the fermentation of yogurt, a series of 83 standards at several different pHs should be used which would make the method extremely impractical. The use of buffers to stabilize the pH was tried without much success due to the difference in pH observed for the samples. However, it was noted that the pH had an influence on the equilibrium of the aroma components. Another factor which might influence the results is the possible growth of microorganisms during the head space equilibrium at 60°C or during storage of the samples for analysis. The present method developed for the analysis of the head—space gas requires the addition of 5 g of yogurt to 5 ml of 20 percent (w/w) solution of sulfuric acid and then equilibration of the mixture in a gyratory water bath at 204 rpm and 60°C for 30 minutes. This method produces a solution containing about 10 percent sulfuric acid, which is the major acid to con- trol the pH of the mixture. Besides that, the peak which appeared at about 17 minutes and which could be related to polymerization of acetaldehyde, was eliminated. "Ghost" peaks sometimes found in the chromatogram were related to volatile chemicals in the air and were elim- inated by flushing the flask with helium and boiling the distilled water used to dilute the sulfuric acid. 84 Equilibrium Analysis Results of the equilibrium analysis are presented in Figures 14 and 15. Acetaldehyde reached the equilibirum in 5 minutes, acetone and ethanol in 10 minutes, while diacetyl required 30 minutes to reach equilibrium. An unidentified com- pound which eluted between acetone and ethanol, seemed to be decomposed during the equilibrium since its area decreased constantly. This compound had the same reten- tion time of ethyl acetate standard. However, its decomposition did not increase the ethanol area, nor any other area. After the equilibrium was reached, acetaldehyde and acetone had a slight tendency to decrease in concen— tration in the head—space gas, due probably to a slight bleeding (less than 0.5 ppm). Therefore, a sample was also removed from each yogurt sample at 10 minutes of incubation in the gyratory water bath for the analysis of ethanol, acetone and acetaldehyde. If the result of the other sample equilibrated for 30 minutes did not agree with the sample removed at 10 minutes to 0.5 ppm, for these compounds, the analysis was repeated. However, the occurrence of severe bleeding was observed very few times. 85 63‘ 6.0< 53‘ 5.6‘ $4- 52‘ 50* OETHANOL 2.6 2.4 2.2 o ACETONE CONCENTRATION . ppm N O ( u 9 o 29.0 OACETALDEHYDE 28.0 . T , 1 , , 1 , O 5 IO Is 20 25 30 35 4o TIMEJAINUTES Figure l4.--Equilibrium curves for ethanol, and acetaldehyde. acetone, 4.0 - 9 (I I CONCENTRATION OF DIACETYL.ppm 86 D DIACETYL 0 UNKNOWN >10 3.0- ~ »z.o )- -I.O 2'5 J 1 I I l r T In} 0 O 6 I0 I!) 20 25 30 35 40 TIME, MINUTES Figure 15.-—Equilibrium curves for diacetyl and the unknown which occurred between acetone and ethanol. (OOO'IX) NMONXNI'I :IO VBNV AUWIIBHV 87 Identification of the Volatile Components Identification of the yogurt aroma components by the columns used were based on the coincidence of reten— tion time and comparison of fast-scan mass spectral fragmentation patterns of unknowns with those for authen- tic compounds. In some instances, such as those for diacetyl and ethanol, the mass spectral data were suffi- cient for positive identifications. The results indicate a positive identification of acetaldehyde, acetone, ethanol, and diacetyl. Survey of the Aroma Com— ponents of Plain Yogurt The number and variety of yogurt cultures used by dairy plants is decreasing due to the increasing avail— ability and advantages of frozen intermediate cultures. The conventional yogurt culture method requires extreme attention to produce a standardized product. Moreover, these cultures usually develop according to the conditions under which they are grown and it is observed that small alterations of the metabolic process may occur depending on factors such as specific culture medium, heat treat— ment, incubation time—temperature relationship, amount of inoculum, final pH, and the final ratio of rods to cocci. The yogurt cultures used in the past by dairy plants were then naturally modified due to handling. 88 Consequently, yogurt produced by different dairy plants are expected to have different flavor balance and possi- ble different flavor components. Therefore, a flavor survey using the head—space technique was made on plain yogurts purchased from local supermarkets. The results of the survey are presented in Table 2. TABLE 2.—-Quantitative Analysis of Some Commercial Plain Low—Fat Yogurts “drink “$25239 ”@222? first Dannon 23.28 0.98 1.33 0.35 Yoplait 19.11 1.43 1.70 ' 0.26 Espirit 18.86 1.61 6.93 0.27 Yubi 13.46 0.50 4.22 0.25 Food Club 6.04 0.86 4.38 0.44 MSU* 35.32 1.35 2.37 1.09 *Culture being propagated in our laboratory. The data obtained in the survey indicate that the same compounds were present in the aroma of all the yogurts; however, a large variation was found in the con- centration of the components of the head—space gas. Acetaldehyde occurred in the highest concentration while diacetyl occurred in the lowest concentration and acetone showed the least variation. Analysis of Standard Compounds 89 Results of the linear regression analysis of the standard compounds are presented in Table 3. TABLE 3. --Linear Regression Equations for Acetaldehyde Acetone, Ethanol, and Diacetyl Linear Regression Corre. Conc. Range Standard Equation* Coef. (ppm) Acetalde- y = 6.4 x 10—40: - 0.0189 0.9982 0.1 60 hyde Acetone y = 5.1 x 10-4m + 0.0873 0.9978 0.1 5 Ethanol y = 1.90 x 10_3m + 0.1998 0.9959 0.2 60 Diacetyl y — 1.18 x 10_3m + 0.1138 0.9948 0.2 6 *y = concentration in ppm m = area of the peak (arbitrary units) Changes in the Aroma Com— ponents During Processing and Incubation The compounds identified in the head space of the yogurt mix before (N) and after (P) pasteurization are shown in Figure 16 i each component is plotte and incubation time. Th n which the concentration of d as a function of processing e unpasteurized mix contained ethanol and acetone in appreciable concentrations, which may have been introduced in it by the nonfat dry milk 90 1 E 3 _I §IO< 350 E I 8 III 2 S 3 550‘ -30.0 E, u i g d _ O 950- NH ’250 z I.” ' o E FACETALDEHYDE q, ‘2; . ETHANOL g R O- . u OACETONE 2°-° B 0 r g uDlACETYL S F 5 < . 95 30 we 5’. z 0 8 8 2 ° 20. a O ' a ' I0.0 g 0 IE 0 , I. < . 8 50 O )- IL 0 I A C J l . . . T . . . . NPO I 2 3 4 INCUBKHON NME.HOURS Unpasteurized yogurt mix Pasteurized yogurt mix zero time culture OO'UZ Figure l6.--The pH and the concentration of acetalde— hyde,ethanol,acetone,and diacetyl of yogurt during processing and incubation using a balanced culture (cocci/rods ~ 1/1). 91 added to the milk. Acetaldehyde was not detected in the unpasteurized mix but was present in the pasteurized yogurt mix. The pasteurization also increased the con— tents of acetone and ethanol. Diacetyl, on the other hand, was not detected in either pasteurized or unpas— teurized mix. To study the effect of the proportion of the cocci to rods in the yogurt aroma, two experiments were run using the same frozen concentrated culture. This culture has a tendency of a slower set when transferred to the sterilized milk to produce the bulk culture than with the latter is transferred to the yogurt mix. The former had a setting time of about 5.5 hours, to produce a bulk culture with a pH of approximately 4.25 and a coccus to rods ratio of 1:1. Transferral of the bulk culture to the yogurt mix before the desired 1:1 ratio has occurred is a practical reality. This will affect the fermentation pattern of compounds produced during a given time of incubation. Therefore, a study was made using a bulk culture with a pH 4.25 and a cocci to rods ratio of about 1:1 and another bulk culture with a pH 4.50 and a cocci to rods ratio of about 2 to l. The head-space gas of the cultures (c) was also analyzed and results are presented in Figure 16 and 92 Figure 17. The culture with a higher pH during fermenta- tion has a significantly lower concentration of acetal- dehyde while slightly higher concentrations of acetone, ethanol, and diacetyl were observed. The increased con- tent of acetaldehyde is due to its production by the Lactobacillus bulgaricus, which will produce acetaldehyde mainly at low pHs. Figure 16 and Figure 17 also show the effect of fermentation on the aroma volatiles of the inoculated mix and on the pH for the two cultures. The yogurt made with a culture to a pH 4.25 had a much faster fermenta- tion time (3.5 hours) to reach a pH 4.25. Incubation for 4.5 hours produced a yogurt with a pH 4.0. The yogurt mix inoculated with the unbalanced culture had much slower acid production, which started to accelerate after 1 hour of incubation, as compared with the balanced culture. The production of aroma components was also quite different. Acetaldehyde began to be produced appre- ciably after 1 hour of fermentation when using the balanced culture. Thereafter, its production was steady and reached 25.5 ppm after 3.5 hours of incubation (pH 4.25) and 29.5 ppm after 4.5 hours (pH 4.00). Pro— duction of all the other aroma components also commenced after 1 hour, when using the balanced culture. However, 93 E % g 7.0— pao < J 8 ’2- 2 w R 5 so- -30.o§ i 'k PH 3 g.- 3 w ACETALDEHYDE ? < . g c ACETONE 4 g I- C‘ "I g 4.0 DIACETYL _2Q° ;. ,_ < o is . E .< S 4 c z 3.0- 45.0 I" g 3 ’- D 3 z 8 O O O K O E LO‘ -s.o D 8 >. 3 0 s v . v w - v 1 1 . . 1 TJ O i N P o I 2 3 4 5 6J INCUBAHON'HME.HOUR$ = Unpasteurized yogurt mix = Pasteurized yogurt mix = Zero time = culture ()O'UZ Figure l7.--The pH and the concentration of acetaldehyde, ethanol, acetone, and diacetyl of yogurt during unbalanced culture (cocci/rods ~2/l)- 94 the increase of diacetyl, acetone, and ethanol was not as pronounced as acetaldehyde, compared to the initial concentrations. The yogurt mix inoculated with the unbalanced culture had acetaldehyde production initiated after 1.5 hours. However, production was not improved appreciably during the middle period of the fermentation as it was for the balanced culture. After 5 hours of incubation, the rate of production was significantly improved, but the pH 4.25 was achieved 0.5 hours later and the acetalde- hyde content never reached the same concentration as it did with the balanced culture. Diacetyl and acetone followed approximately the same pattern, but at a lower proportion. Ethanol had a different production rate between 0.5 hour and 2.5 hours of incubation. The yogurt mix inoculated with the unbalanced mix had its ethanol produced mainly between 0.5 and 2.0 hours, while the balanced mix produced ethanol mainly between 1.0 and 2.5 hours. This difference may be attributed to the meta- bolic pattern since concomitant with a higher ethanol production were observed at lower levels of lactic acid, diacetyl,acetone and acetaldehyde. The final ethanol production was approximately the same for both cultures. 95 Changes in the Aroma Com— ponents during Storage The storage of the yogurt had much influence in the concentration of acetaldehyde and ethanol while little change was observed in the acetone and diacetyl content. Figure 18 shows the change of the aroma components of yogurt (pH 4.25) produced by the balanced culture and Figure 19 of the unbalanced culture. The ethanol content increased constantly while the acetaldehyde decreased dur— ing storage. This change was much more pronounced after 14 days of storage. However, the ethanol content of the yogurt made from the unbalanced culture was about twice that of the yogurt made by the balanced culture, due probably to a higher ethanol dehydrogenase activity of the streptococci. Yogurt produced by the unbalanced culture was also refrigerated at a pH 4.5. As for the other yogurts, an increase of acetaldehyde is noted (Figure 20) due to the continued growth of the microorganisms during the time required to cool the yogurt down to temperatures which would inhibit their growth. Basically, the same pattern of ethanol production was observed, however, with higher rate during the first 14 days and lower rate after 14 days, as compared with the other yogurts. The acetalde- hyde ramained almost constant up to 21 days when its concentration began to decrease appreciably. 96 30.0- 25.0« 2 0.0‘ w ACETALDEHYDE o ETHANOL O ACETONE ‘3 DIACETYL £3 9 IOD‘ CONCENTRATION, ppm :34 21 28 O "i .34 STORAGE TIME. DAYS Figure 18.--The concentration of acetaldehyde, ethanol, acetone and diacetyl of yogurt (pH 4.25) produced by the balanced culture, during storage. 97 400‘ 350‘ 30.0- if ACETALDEHYDE 250‘ . ETHANOL E oACETONE 3 ‘t c DIACETYL .6. 20.0 .— ‘ K 5 Lulfifl‘ O 2 O O I0.0- 5.0- 4 W: a; ., A. in 4—4 0 i I! (4 2': 28 STORAGE TIME, DAYS Figure l9.--The concentration of acetaldehyde, ethanol, acetone and diacetyl of yogurt (pH 4.25) produced by the unbalanced culture, during storage. 98 30.0 25.0- a a: ACETALDEHYDE 3: . ETHANOL 2 20.0- 9 OACETONE 5 sauce-m. '2' Iss- El 3 o 4 0 10.0‘ s .o- 4 ”\——--,3—\— E : o i ’7 1'4 2'! 2'8 STORAGE T1ME,DAYS Figure 20.--The concentration or acetaldehyde, ethanol, acetone and diacetyl of yogurt (pH 4.50) produced by the unbalanced culture during storage. 99 Importance of the Flavor and Aroma on Quality of Yogurt Yogurt is a fermented product in which the flavor is characterized primarily by the acetaldehyde content and by the lactic acid content. The yogurt mix contributes a background flavor which is important to the overall yogurt flavor. The presence of secondary flavor compounds, such as diacetyl, contribute to the delicate flavor and aroma. However, the role of diacetyl may become more prominent when the acetaldehyde content is low (Groux, 1973). Hamdan et al. (1971) using a wet-chemical method determined the acetaldehyde content of yogurt during processing and storage. They reported that the cultures used produced maximum acetaldehyde of 22 to 26 ppm at the fifth hour, with a decline thereafter. However, no reference to other parameters, such as pH, acidity, and the rod to coccus ratio were made. They also reported that the acetaldehyde content of two cultures decreased during storage for 4 weeks at 4°C while in another it remained constant. The volatile components identified in the yogurt aroma (see Figure 16 and Figure 17) can be assigned to two main sources, namely, the yogurt mix itself and the metabolism of the mix constituents (mainly lactose) during incubation. lOO Yogurt, however, is a product of bacterial symbio- sis. Therefore, a close control of the ratio of the bacteria present in yogurt is not only recdmmended but necessary. It is generally accepted that the Strepto— coccus:Lactobacillus ratio in the final product should be 1:1, however, even more important than that is the Streptococcus:Lactobacillus ratio in the culture. Pette and Lolkema (1950a and 1950b) reported that the proportion of streptococci to lactobacilli in a yogurt culture is not constant during incubation. The rapid acid production is caused by the stimulation of the cocci by the rods, mainly due to its proteolytic action. Therefore, in the unbalanced culture (cocci/rods 2/1) not only the lactobacilli occur in lower concentra-. tion, as compared with a balanced culture, but also the streptococci have their growth limited by the lower amino acids content. With this impaired growth both bacteria will have their metabolism slowed down and the production of the necessary components of the yogurt flavor will be also impaired. A comparison of Figure 16 and Figure 17 shows that this is exactly the case. Furthermore, the final concentration of each aroma compound is also lower with the exception of ethanol in the case of the unbalanced culture, for the same final pH 4.25. Interestingly, the turning point for the acetaldehyde production was around lOl pH 6.0 for both cultures, although the unbalanced culture had a much slower increase in the rate of acetaldehyde production. During storage of yogurt several changes may occur which include microbial, enzymatic and abiotic or chemi- cal changes. The microbial spoilage of yogurt usually occurs in the form of colonies or film on the surface of yogurt. Samples of all yogurt batches produced were kept under refrigeration for at least 2 months and no visible colon— ies were observed on the surface of any yogurt. The abiotic spoilage of yogurt occurs due to changes mainly attributed to air and light oxidation. The enzymatic activity of stored yogurt is due mainly to the enzymes produced by the yogurt bacteria, since almost all the enzymes of the mix are inactivated by the heat treatment of the mix. The action of enzymes may cause defects such as gas formation, whey separation, rancidity, cheesy and bitter taste and over-acidification depending on the "post-acidification" properties of the yogurt culture. Other factors such as temperature, presence of air, light, packaging material and storage time may also influence the keeping quality of yogurt. The present research has shown an interesting change occurring during storage of yogurt. The increase 102 of ethanol content and the decrease of acetaldehyde con- tent, specially after 14 days of storage. The production of the metabolic flavor products of the yogurt bacteria follows the diagram shown in Figure 21 (Dumont and Adda, 1979). During the manufacture of yogurt some ethanol is produced from acetaldehyde by the enzyme alcohol dehydro- genase. This enzymatic action continues during storage. Bills and Day (1966) demonstrated that dehydrogenase activity occurs at low storage temperature by some lactic streptococci. They also reported that these organisms reduced acetaldehyde but not acetone. Keenan and Lindsay (1967) reported dehydrogenase activity by some lacto- bacilli. The storage of yogurt produced by the balanced culture presented a molecular equilibrium between acetalde— hyde and ethanol (Figure 18) which may be attributed to alcohol dehydrogenase activity. The storage of the yogurt made from the unbalanced culture presented the same indirect proportions between acetaldehyde and ethanol as the balanced yogurt, but the molecular equilibrium was not observed, which indicates the possible presence of other active enzymes. The lowest ethanol production during manufacture and incubation was observed by the balanced culture, while the unbalanced culture produced yogurt (stored at pH 4.25) with the highest ethanol production during storage. 103 CITRIC ACID GLUCOSE OXALOACETIC ACID + ACETIC ACID / -C02 PYRUVIC ACID Acetyl—P H2 +TPP --CO2 LACTIC ACID Acetaldehyde—TPP \\<:::::FP 1y -TPP Acetyl-CoA ! ACETOLACTIC ACID ACETALDEHYDE CoA ~TPP +H2 -CO2 DIACETYL +H2 ‘ ETHANOL ACETOIN +H V 2,3-BUTANEDIOL Figure 21.--Simp1ified diagram of the metabolic pathway to produce the flavor components of yogurt. 104 However, one should keep in mind that changes in the commercial manufacturing process, such as amount of inoculum, temperature of incubation and/or storage, and the yogurt mix itself might also have influence on the fermentation and storage changes of yogurt. SUMMARY AND CONCLUSIONS The lactose, glucose, and galactose content of yogurt during processing, fermentation, and storage were evaluated by high performance liquid chromatography using a chemically bonded stationary phase carbohydrate—type column (on microparticulate support). Since galactose and glucose cannot be resolved by this column, a method was developed to eliminate the glucose by glucose oxidase thus allowing the quantitation of all these sugars. The volatile components of yogurt produced during processing, fermentation, and storage were identified and quantified using a modified head—space technique. The head—space techniques reported in the literature for the analysis of yogurt did not control the pH of the liquid phase which was found to affect the equilibrium of the vapor phase. During yogurt fermentation, the pH gradu- ally decreases from 6.6 to about 4.2; consequently, a simple dilution with water affects the vapor phase equilibrium of samples of different pH. To control the pH, 5 g of yogurt were mixed with 5 ml of 20 percent (w/w) sulfuric acid to produce a solution with a pH of 0.87. The presence of lactic acid did not change the pH 105 106 appreciably. Other advantages of the sulfuric acid method is the prevention of any microbial growth and also elimination of polymerization of acetaldehyde or other aroma compounds during heating in a gyratory water bath (60°C and 204 rpm for 30 minutes) to equilibrate the vapor phase. Acetaldehyde, acetone, ethanol, and diacetyl were identified by relative retention in a gas chromatographic system and mass spectrometric data. Quantatitive analyses were performed using standards added to yogurt mix and analyzed by gas chromatography- Equilibrium curves were constructed for all volatile components. To evaluate the effect of culture handling, or more specifically the Streptococcus thermophilus to Lactobacillus bulgaricus ratio, on the chemical compon- ents, a balanced culture (1:1) and an unbalanced culture (2:1) of cocci to rods were used to produce yogurt. The culture used throughout this work was a commercial frozen concentrated culture. Results indicate that lactase activity was observed mainly during fermentation with very low activ- ity during storage for 28 days. While the hydrolysis of lactose was faster for the balanced culture (30 percent) the percentage of lactose hydrolyzed was higher for the unbalanced culture (34 percent) due probably to its 107 longer incubation time to achieve the pH 4.25. The balanced culture metabolyzed 86 percent of the glucose and 13 percent of the galactose produced by the hydroly- sis of lactose, while the unbalanced culture metabolized 87 percent and 7 percent of the glucose and the galactose respectively. This difference in the metabolism of galactose may occur due to the inability of Streptococcus thermophilus to utilize this sugar or due to a longer fermentation time, or slower growth rate, which would permit the bacteria to metabolyze the glucose molecules preferentially, as compared to the balanced culture. The lactose concentration curve closely followed the pH curve; as the lactose content of the mix decreases, so did the pH. The concentration of all the volatile components increased during fermentation. Acetaldehyde began to be produced appreciably when the pH of the inoculated yogurt mix was about 6.0 for both culture ratios. From there on a steady increase was observed until the end of the fer- mentation. The acetaldehyde content of the yogurt pro~ duced by the balanced culture was higher (25.5 ppm) than the unbalanced culture (19.0 ppm) due to a steeper pro- duction curve. Lactobacillus bulgaricus, the organism which was deliberately out of balance in the latter cul- ture, is the main producer of acetaldehyde, especially 108 in the later phase of the fermentation. Therefore, the effect of unbalance of this microorganism on the yogurt culture results in a lag period in the second phase of the fermentation which allows the Streptococcus thermo- philus to keep reproducing and lowering the pH without increasing the acetaldehyde content appreciably. The ethanol and diacetyl content also increased during fermentation, but to a lesser extent than acetaldehyde. The main production of these two components was observed during the transition of growth of Strepto- coccus thermophilus to Lactobacillus bulgaricus. Acetone also showed an increase during the transition of cocci to rods for the balanced culture, while for the unbal- anced culture, it did not change. Yogurt stored for 14 days showed a slight decrease of acetaldehyde content and a slight increase in the ethanol content for both cultures. However, after 14 days a sharp increase in the ethanol content was observed with a concomitant decrease in the acetaldehyde content. An approximate molecular equilibrium between acetaldehyde and ethanol was observed for the balanced culture, while the unbalanced culture did not show this equilibrium. Whereas yogurt produced with the unbalanced culture with a higher pH (4.5) and stored for 21 days showed a slightly steeper ethanol curve with practically 109 no change in the acetaldehyde content, a sharp increase of ethanol was observed after 21 days, concomitant with a decrease of acetaldehyde. These transformations sug— gest that ethanol dehydrogenase activity is very impor- tant during storage of yogurt, especially after approxi- mately 14 days of storage. However, the activity of other enzymes may also play an important role on the stability of yogurt flavor since the unbalanced culture did not show molecular relationship between acetaldehyde and ethanol. REFERENCES 110 REFERENCES Acott, K., and T. P. Labuza. 1972. Yogurt: Is it truly Adelle's BVitamin factory? Food Prod. Develop. 6:50. Aggarwal, M. L. 1974. Manufacturing yogurt from goat milk. Cultured Dairy Prod. J. 9 (3):11. Anonymous. 1970. "The household market for yogurt, uses and user profile." American Dairy Association. 13 pp. Chicago, Ill. Anonymous. 1971. Lactose intolerance. Dairy Council Dig. 42:31. Anonymous. 1972. The nutrient composition of cultured dairy foods. Dairy Council Dig. 43 (4):l6. Anonymous. 1974a. U.S.D.A. AGric. Handbook No. 51. Federal and state standards for the composition of milk products. Washington, D.C. Anonymous. 1974b. "The household yogurt market." United Dairy Ind. Assoc. 16 pp. Rosemont, Ill. Anonymous. 1977. Federal Register, Vol. 42, No. 112, Friday, June 10, 1977, 29923—29925. Anonymous. 1978. Grade A pasteurized milk ordinance. U. 8. Public Health Service, Publication No. 229, U. S. Department of Health, Education and Welfare, Washington, D.C. Aronson, J. 1952. Transgalactosidation during lactose hydrolysis. Arch. Bioch. Biophys. 39:370. Ashton, T. R. 1963. Cultured milk products. J. Soc. Dairy Tech. 16 (2):68. Association of Official Analytical Chemists. 1975. Official Method of Analysis 12th ed. 111 112 Bautista, E. S., R. S. Dahiya and M. L. Speck. 1966. Identification of compounds causing symbiotic growth of S. thermophilus and L. bulgaricus in milk. J. Dairy Res. 33:299. Bills, D. D. and E. A. Day. 1966. Dehydrogenase activity of lactic streptococci. J. Fairy Sci. 49:1473. Blanc, B. 1966. Influence du Stockage et des traite- ments industriels sur la teneur en vitamines du lait de vache. Le Lait, 46:141. Borber, F. W. 1962. Hygienic aspects of the processing of butter, cheese, flavored and fermented milks, and ice cream. In "Milk Hygiene," ed. World Health Organization, Monograph series No. 48. Geneva. Bottazzi, V. and F. Dellaglio. 1967. Acetaldehyde and diacetyl production by Streptococcus thermophillus and other lactic streptococci. J. Dairy Res. 34: 109. Bottazzi, V. and M. Vescovo. 1969. Carbonyl compounds produced by yogurt bacteria. Neth. Milk Dairy J. 23:71. Bottazzi, V., B. Battistotti and G. Montescani. 1973. Influence des souches seules et associus de E. Bulgaricus et S. thermophilus ainsi que des traitements du lait sur la production d'aldehyde acetique dans le yaourt. Le Lait 53:295. Brandao, S. C. C., M. L. Richmond, J. I. Gray, I. D. Morton and C. M. Stine. 1980. Separation and estimation of mono and disaccharides and sorbitol by high performance liquid chromatography. J. Food Sci. Submitted to publication. Brandao, S. C. C., M. L. Richmond, J. I. Gray, and C. M. Stine. 1980. High performance liquid chromato— graphic analysis of monosaccharides and some of their derivatives. J. Food Sci. Submitted to publication. Brunner, J. R., C. W. Duncan and G. M. Trout. 1953a. The fat globule membrane of nonhomogenized and homogenized milk. I. The isolation and amino— acid composition of the fat-membrane proteins. Food Res. 18:454. 113 Brunner, J. R., C. W. Duncan, G. M. Trout and M. Mackenzie. 1953b. The fat-globule membrane of nonhomogenized and homogenized milk. III. Differences in the sedimentation diagrams ofthe fat-membrane pro- teins. Food Res. 18:469. 0 Brunner, J. R. 1974. "Physical equilibria in milk. The lipid phase in 'Fundamentals of Dairy Chemis- try.'" Ed.: B. H. Webb, A. H. Johnson, and J. A. Alford. AVI, Westport, Conn. Buchman, R. E. and N. E. Gibbons. 1974. Bergeys Manual of Determinative Bacteriology. Baltimore, Williams and Wilkins. Casalis, J. 1975. Facteurs technologiques influencant la consistance, la texture, l'arome ou le gout du yaourt. Ind. Alim. Agr. 92:1253. ‘ Chambers, J. V. 1979. Culture and processing techniques important to the manufacture of good quality yogurt. Cultured Dairy Prod. J. 14 (2):28. Chandan, R. C., M. A. Searles and J. J. Finch. 1969. Lipase activity of lactic cultures. J. Dairy Sci. 52:894. Conrad, E. C. and J. K. Palmer. 1976. Rapid analysis of carbohydrates by high—pressure liquid chromatog— raphy. Food Tech. 30 (10):84. Cottenie, J. 1978. Yogurt processing in Europe. Cult. Dairy Prod. J. 13 (4):6. . Cramer, L. K., G. P. Berry, and A. R. Matheson. 1978. The effect of pH on protein aggregation in heated skim milk. New Zealand J. Dairy Sci. Technol. 13:9. Czulak, J. 1962. Commercial manufacture of yogurt. Aust. J. Dairy Technol. 17:52. Damrau, F. 1954. Therapeutical uses of yogurt, a review of medical literature. International yogurt Foundation LTD. New York. » Davies, F. L., P. A. Shankar, B. E. Brooker, and D. G. Hobbs. 1978. A heat-induced change in untra- structure of milk and its effect. J. Dairy Res. 45:53. 114 Davis, J. G. 1956. Yogurt and other cultured milks. J. Soc. Dairy Technol. 9:69. Davis, J. G. and T. McLachland. 1974. Yogurt in the United Kingdom: Chemical and microbiological analysis. Dairy Ind. 39:149. Donaldson, R. M. 1964. Normal bacterial populations of the intestine and their relation to intestinal function. New England J. Med. 270:1050. Donay, J. K. and V. G. Tahan. 1964. U. S. Patent 3.128. 190. April 7. Duitschaever, C. L., D. R. Arnott, and D. H. Bullock. 1972. Quality evaluation of yogurt produced commercially in Ontario. J. Milk Food Technol. 35:173. Dumont, J. P. and J. Adda. 1979. Flavour Formation in Dairy Products. In "Progress in Flavour Research." Ed.: Land, L. D. and H. E. Nursten. Applied Sci. Pub. LTD., London. Dunmire, D. L. and S. E. Otto. 1979. High pressure liquid chromatbgraphic determination of sugars in various food products. J.A.O.A.C. 62:176. Ehrlich, R. 1963. Treatment of enteric staphylococci infectiOns with L. acidophilus. Amer.~J.Proctol- ogy 14:53. El-Sadek, G. M., K. Nagiub, and A. Negm. 1972. Acetalde- hyde and diacetyl production in Zabady and Zabady cultures. Milchwiss. 27:756. Engel, W. G. 1973. The use of lactase to sweeten yogurt without increasing calories. Cultured Dairy Prod. J. 8 (3):6. - Euber, J. R.-and J. R. Brunner. 1979. Determination of lactose in milk products by high performance liquid chromatography. J. Dairy Sci. 62:685. Feldstein, F. J. and D. C. Westhoff. 1979. The influence of heat treatment of milk on starter activity: What about UHT? Cult. Dairy Prod. J. 14 (2):ll. Galesloot, T. E. 1958. Investigations concerning the consistency of yogurt. Neth. Milk Dairy J. 12:130. 115 Galesloot, T. E., F. Hassing, and H. A. Veringa. 1968. Symbiosis in yoghurt (I). Stimulation of Lactobacillus bulgaricus by a factor produced by Streptococcus thermophilus. Neth. Milk Dairy J. 22:50. Gallagher, C. R., A. L. Mooleson, and J. H. Caldwell. 1974. Lactose intolerance and fermented dairy products. J. Amer. Diet. Assoc. 65:418. Ghadimi, H. and P. Pecosa. 1963. Free amino acids of different kinds of milk. J. Clin. Nut. 13:75. Gorner, F., V. Palo, and M. Bertran. 1968. Changes of the content of volatile substances during ripen- ing of yogurt (German) Milchwiss. 23:94. Gorner, F., V. Palo, and M. Seginova. 1975. Changes in volatile compounds during yogurt ripening. Polnohospodarstvo. 21:173. Goodenough, E. R. 1975. Carbohydrate content of yogurt and effects of viable microflora upon digestion. Ph.D. Thesis. Rutgers University. Goodenough, E. R., and D. H. Kleyn. 1976. Influence of viable yogurt microflora and digestion of lactose by the rat. J. Dairy Sci. 59:601. Goulden, J. D. S. and L. W. Phipps. 1964. Factors affecting the fat globule sizes during the homog- enization of milk and cream. J. Dairy Res. 31: 195. Gray, G. M. 1967. Malabsorption of carbohydrate. Fed. Proc. 26:1415. Greene, V. W. and J. J. Jexeski. 1957a. Studies on starter metabolism II. The influence of heat- ing milk on the subsequent response of starter cultures. J. Dairy Sci. 40:1053. Greene, V. W. and J. J. Jexeski. 1957b. Studies on starter metabolism III. Studies on cysteine- inducted stimulation and inhibition of starter cultures in milk. J. Dairy Sci. 40:1062. Grigorov, T. A. 1965. Syneresis in yoghurt (GE) Milchwiss. 20:685. 116 Grigovov, H. 1966. Effect of heat treatment on cow's Groux, Groux, Gulko, Haenel, milk on the hydrophilic properties of the protein in Bulgarian yoghurt. Proc. 17th Int. Dairy Congr. Section F, 649. M. “1973. Etude des composants de la Haveur du yoghourt. Le Lait 53:146. M. 1976. Study on the components of yogurt flavor. Nestle Research News 1974/75 Switzerland. and A. Kurglova. 1966. Content of vitamins of the "B" complex in fermented milks. XVII Int. Dairy Congr. E/F, 689. H. 1970. Human normal and abnormal gastro- intestinal flora. Amer. J. Clin. Nut. 23:1433. Hall, C. W. and T. I. Hedrick. 1966. Drying of Milk and Milk Products. AVI, Westport, Conn. Hall, C. W. and G. M. Trout. 1968. Milk Pasteurization. AVI, Westport, Conn. Hall, T. A. 1975. Yogurt formulations with attention to Hamdan, Hamdan, Hamdan, Hansen, Harper, Harvey, stabilizers system. Cultured Dairy Prod. J. 10 (3):12. . I. Y., J. E. Kunsman, and D. D. Deane. 1971. Acetaldehyde production by combined yogurt cul- tures. J. Dairy Sci. 54:1080. I. Y., J. E. Kunsman, and D. D. Deane. 1971. Acetaldehyde production by combined yogurt cul- tures. J. Dairy Sci. 54:1080. L. Y., et al. 1971. Cited from "Yogurt" 1978. Eds. J. L. Rasic and Y. A. Kurmann. Tech. Dairy Publishing House. Copenhagen, Denmark: J. Dairy Sci. 54:7. R. G. and E. M. Craine. 1954. The reversible conversion of galactose-l-phosphate to glucose—1— phosphate. J. Biol. Chem. 208:293. W. J. and T. Kristoffersen. 1956. Biochemical aspects of cheese ripening. J. Dairy Sci. 39:1773. R. J. 1960. Production of acetone and acetal- dehyde by lactic streptococci. J. Dairy Res. 27: 41. 117 Hawley, H. B., P. A. Shepard and D. M. Wheater. 1959. J. Appl. Bacteriol. 22:360. Hersh, T. 1972. Lactose intolerance in health and disease. Nutrition News. 35:5. Hetrick, J. H. and P. H. Tracy. 1948. Effect of HTST treatments on some properties of milk. II. Inactivation of the lipase enzyme. J. Dairy Sci. 31:881. Hild, V. J. 1979. Quantitative determination of some important flavour components in dairy products by means of the head space technique (Ge). Milchwiss. 34:281. Huang, C. and T. M. Bayless. 1968. Milk and lactose intolerance in healthy orientals. Sci. 160:83. Humphreys, C. L. and M. Plunkett. 1969. Yogurt: a review of its manufacture. Dairy Sci. Abstr. 31 (ll):607. Hurt, D. H. 1972. Lactose intolerance. Presented at the Am. Cult. Dairy Inst. meeting. Atlantic City, N.J. October. Hurst, W. J., R. A. Martin and B. L. Zoumas. 1979. Applications of HPLC to characterization of individual carbohydrates in foods. J. Food Sci. 44:892. Jachumsen, A. 1978. Industrial production of frozen yoghurt. Cultured Dairy Prod. J. 13 (l):26. Jago, G. R. 1954. Factors influencing the lactic acid- producing properties of streptococci used in the manufacture of cheddar cheese. J. Dairy Res. 21: 111. Jenness, R. and S. Patton. 1976. Principles of Dairy Chemistry. 2nd ed. John Wiley and Sons, Inc. Kalab, M., D. B. Emmons, and A. G. Sargant. 1975. Milk gel structure. IV. Microstructure of yoghurt in relation to the presence of thickening agents. J. Dairy Res. 42:453. Kalab, M., D. B. Emmons, and A. G. Sargant. 1976. Milk get structure. V. Microstructure of yoghurt as related to the heating of milk. Milchwiss. 31:402. 118 Kalab, M. 1979. Microstructure of dairy foods. I. Milk products based on protein. J. Dairy Sci. 62:1352. Keenan, T. W., and D. D. Bills. 1968. Metabolism of volatile compounds by lactic starter culture microorganisms. A Review. J. Dairy Sci. 51: 1561. Keenan, T. W. and R. C. Lindsay. 1967. Dehydrogenase activity of Lactobacillus species. J. Dairy Sci. 50:1585. Keeney, P. G. and S. Patton. 1956. The coconut-like flavor defect of milk fat. II. Demonstration of deltalactone in dried cream, dry whole and evaporated milk. J. Dairy Sci. 39:1114. Kilara, A. and K. M. Shahani. 1974. B-Galactosidase activity of cultured and acidified dairy products. J. Dairy Sci. 57:592. Kilara, A. and K. M. Shahani. 1978. Lactic fermenta- tions of dairy foods and their biological signifi— cance. J. Dairy Sci. 61:1793. Knutson, R. D., K. Kepner, and J. Nichols. .1978. Yogurt and its products: a market profile. Cultured Dairy Prod. J. 13 (4):ll-l4. Kosikowski, F. 1977. Cheese and Fermented Milk Foods. Edwards Brothers, Inc., Ann Arbor, MI. Kroger, M. 1973. Controlling the quality of yogurt. Dairy and Ice Cream Field. 156:38. Kroger, M. and J. C. Weaver, 1973. Confusion about yogurt--compositional and otherwise. J. Milk Food Technol. 36:388. Kroger, M. and S. R. Fram. 1975. Consumer attitudes toward yogurt. Food Tech. 29 (ll):52. Kroger, M. 1976. Quality of yogurt. J. Dairy Sci. 59:344. Lampert, L. M. 1975. Modern Dairy Products. 3rd ed. Chemical Publishing Co., Inc., N.Y. 119 Larson, B. L. and G. D. Rolleri. 1955. Heat denaturation of the specific serum proteins in milk. J. Dairy Sci. 38:351. Larue, A. J. 1972. Cited from "Yoghurt" 1978. Eds. Rasic, J. L. and J. A- Kurmann. Techn. Dairy Publishing House. Copenhagen, Denmark. Lehninger, A. L. 1976. Biochemistry. Worth Publishers, Inc., N.Y. Linden, J.C. and C. L. Lawhead. 1975. Liquid chromatog- raphy of saccaharides. J. Chromatog. 105:125. Lindsay, R. C. and E. A. Day. 1965. Rapid quantitative method for determination of acetaldehyde in lactic starter cultures. J. Dairy Sci. 48:665. Lindsay, R. C. 1967. Cultured dairy products, in "Symposium on Foods: the Chemistry and Physiology of Flavors." AVI Pub. Co., Westport, Conn. Lloyd, G. T. 1971. New developments in starter tech— nology. Dairy Sci. Abst. 33(6):4ll. Lundstedt, E. 1971. Yogurt. Cultured Dairy Prod. J. 6(3):10. Lundstedt, E. 1973. Improved methods for the manufac- turing of yogurt and problems related to yogurt manufacture. Cultured Dairy Prod. J. 8(4):14. Mayer, B. M. and M. E. Powell. 1966. U. S. patent 3, 269, 842. Aug. 30. Mieklejohn, P. G. 1977. How to achieve better yoghurt viscosity. Cultured Dairy Prod. J. 12(1):6. Mikolajcik, E. M. and I. Y. Hamdan, 1975a. Lactobacillus acidophillus, I: growth characteristics and metabolic products. Cultured Dairy Prod. J. 10(1):10. Mikolajeik, E. M. and I. Y. Hamdan. 1975b. Lactobacillus acidophilus II: Antimicrobial agents. Cultured Dairy Prod. J. 10(1):18. Macquot, G. 1970. The selection and use of some micro- organisms for the manufacture of fermented and acidified milk products. Dairy Ind. 35:234. 120 Moinas, M. 1976. Volatiles of milk products, extraction and identification. Nestle Research News 1974/75. Switzerland. Muck, G. A., J. Tobias, and R. M. Whitney. 1963. Flavor of evaporated milk. I. Identification of some compounds obtained by the petroleum ether solvent partitioning technique from aged evaporated milk, J. Dairy Sci. 46:774. Mulder, H. and P. Walstra. The Milk Fat Globule. 1974. The University Press, Belfast, Northern Ireland. Muralidhara, K. S., W. E. Sandime, P. R. Elliker and D. C. England. 1972. Effect of feeding Lacto- bacillus concentrates on coliform excretion and securing in swine. J. Dairy Sci. 55:661. Nielsen, V. H. 1975. Factors which control the body and texture of commercial yogurt. Amer. Dairy Rev. 37:36. Olson, H.C. 1966. Effect of the inherent properties of milk on production of acid of selected lactic cultures. J. Dairy Sci. 49:254. Parks, 0. W. and S. Patton. 1961. Volatile carbonyl compounds in stored dry whole milk. J. Dairy Sci. 44:1. Parks, 0. W., D. P. Schwarts and M. Keeney. 1964. Identification of o-amino acetophenone as a flavor compound in stale dry milk. Nature 202:185. Parlement, T. H. and W. W. Nawar. 1965. Delta-capro- lactone in heated milk fat. J. Dairy Sci. 48:615. Pavey, R. M. 1974. Milk coagulation and protein denaturation. In "Fundamental of Dairy Chemistry." Ed. B. H. Webb, A. H. Johnson and J. A. Alford. AVI, Westport, Conn. Patton, S., D. A. Forss, and E. A. Day. 1956. Methyl sulfide and the flavor of milk. J. Dairy Sci. 39:1469. Patton, S. 1961. Gas chromatographic analysis of flavor in processed milks. J. Dairy Sci. 44:207. 121 Paul, D., and L. C. Hoskins. 1972. Effect of oral lactobacillus feedings on fecal lactobacillus counts. Amer. J. Clin. Nut. 25:763. Pazur, J. H. 1953. The mechanism of enzymatic synthe- sis of galactosyl oligosaccharides. J. Biol. Chem. 208:439. ' Pazur, J. H. and A. L. Gordon. 1953. Studies on inulin. The preparation and properteis of inulosiose. J. Am. Chem. Soc. 75:3458. Pelczar, M. J. and R. D. Reid. 1972. Microbiology. McGraw-Hill Book Company, N.Y. Pette, J. W. and H. Lolkema. 1950a. Symbiosis and antibiosis in mixed cultured of Lactobacillus bulgaricus and Streptococcus thermophilus. Neth. Milk Dairy J. 4:197. Pette, J. W. and H. Lolkema. 1950b. Growth stimulating factors for Streptococcus thermophilus. Neth. milk Dairy J. 4:209. Pette, J. R. and H. Lolkema. 1950c. Acid production and aroma formation in yogurt. Neth. Milk Dairy J. 4:261. Pette, J. W. and H. Lolkema. 1951a. Factor influencing the proportion of streptococci and lactobacilli in a yogurt culture. Neth. Milk Dairy J. 5:14. Pette, J. W. and H. Lolkema. 1951b. Firmness and whey separation of milk yogurt. Neth. Milk Dairy J. 5:27. Pette, J. W. 1957. Yoghurt and its problems. Killer Milchw. For schungsberichte 9:177. Pette, J. W. 1964. Cited from "Yoghurt" 1978. Eds. Rasic, J. L. and J. A. Kurmann. Tech. Dairy Publishign House. Copenhagen, Denmark. Powell, M. E. 1970. Yogurt, acidophilus and bulgarian milks. Cultured Dairy Prod. J. 5(1):4. Phuan, Z. 1974. Critical evaluation of the Harland—— Ahswrth test for the determination of the undenaturated whey proteins. Milchwiss, 29:459. 122 Phuan, Z. 1979. Heat treatment of cultured dairy products. J. Food Prot. 42:890. Radema, L. and R. V. Dijk. 1973. Thickening agents for yogurt. Mededelingen Nederlands Instituut voor zuivelonolerzoek. 7:51. Rand, A. G. 1973. The use of enzymes for the reduction of lactose levels in milk products. Aust. J. Dairy Tech. 28:68. Randoin, L. and J. Causeret. 1956. Teneur du yoghourt en quelques vitamines. Le Lait, 36:129. Rasic, J., T. Stojsavljevic and T. Curcic. 1971. A study of the aminoacids of yoghurt. II. Amino acids content and biolgoical value of the pro- teins of different kinds of yoghurt. Milchwiss 26:219. ‘ Rasic, J., N. Vucurovic, and B. Obradovic. 1973. Studies on free fatty acids in yoghurt made from cow;s, ewe's and goat's milk. Milchwiss 28:220. Rasic, J. L. and J. A. Kurmann. 1978. Yogurt. Tech. Dairy Publishing House. Copenhagen, Denmark. Reddy, G. V. and K. M. Shahani. 1971. Isolation of an antibiotic from Lactobacillus bulgaricus. J. Dairy Sci. 54:748. Reddy, K. P., K. M. Shahani, and S. M. Kulkarni. 1976. B—complex vitamins in cultured and acidified yogurt. J. Dairy Sci. 59:191. Reinecius, G. A., T. E. Kavanaugh and P. G. Kenney. 1970. Identification and quantitation of free neutral carbohydrates in milk products by gas liquid chromatography and mass spectrometry. J. Dairy Sci. 53:1018. Richmond, M. L., R. C. Chandan, and C. M. Stine. 1979. Yogurt--A compositional survey in the greater Lansing area. J. Food Prot. 42:424. Richter, R. L. and G. H. Hartman. 1977. Penetrometer evaluation of sour cream and yogurt at 1977 kultures and kurds clinic. Cultured Dairy Prod. J. 12 (4):l4. 123 Roberts, H. R. and E. F. McFarren. 1953. The formation of oligosaccharides during the lactase hydrolysis of lactose. Arch. Biochem. Biophys. 43:233. Roberts, H. R., and J. D. Pettinati. 1957. Cited from Fermentations in "Fundamentals of Dairy Chemistry." 1974. Eds. B. H. Webb, A. H. Jonson and J. A. Alford. AVI, Westport, Conn. Robinson, R. K. and A. Y. Tamine. 1976. Quality appraisal of yoghurt. J. Soc. Dairy Tech. 29: 148. ' Sandine, W. E., C. Daly, P. R. Elliker, and E. R. Vedamuthu. 1972. Causes and control of culture- related flavor defects in cultured dairy products. J. Dairy Sci. 55:1031. Sandine, W. E., K. S. Muralidhara, P. R. Elliker, and D. C. England. 1972. Lactic acid bacteria in food and health: A review. J. Milk Food Technol. 35:691. Schormuller, J. and H. Langner. 1960. Z. Lebens. Untersuch. Forsch. 113:197. Cited from "Yoghurt" 1978. Rasic, J. L. and J. A. Kurmann. Tech. Dairy Publishing House. Copenhagen, Denmark. Schulz, M. E. 1953. XIII. Int. Dairy Congr. II: 455 Milchkonsum in form von Nebenprodukten die Zubereitung dieser produkte. Scobell, H. D., K. M. Brobst and E. M. Steele. 1977. Automated liquid chromatographic system for analysis of carbohydrates mixtures. Cereal Chem. 54:905. Shanley, R. M. 1973. Free sugars in yogurt. Aust. J. Dairy Technol. 28:58. Sellars, R. L. 1973. Profitable manufacture of yogurt and propagation of yogurt cultures. Modern Dairy 52:19. Sigma Chemical Company. 1979. Personal Communication. St. Louis, Mo. Singh, J., A. Khanna, and H. Chander. 1979. Anti- bacterial activity of yogurt starter in cow and buffalo milk. J. Food Prod. 42:664. 124 Sommer, H. H. 1946. Market Milk and Related Products. 2nd ed. Madison, Wis. Soskin, D. H. 1967. Yogurt or yoghurt or yoghourt ou yohourt. Dairy and Ice Cream Field 123 (10):38-39. Speck, M. L. 1900. Potential benefits of natural lactic acid fermentations. Dairy and Ice Cream Field, 136(1):58. Stadhouders, J., G. Hup and L. A. Jansen. 1971. A study of the optimum conditions of freezing and storing concentrated mesophilic starters. Neth. Milk Dairy J. 25:229. Steinitz, W. S. 1971. Frozen yogurt--new product for the ice cream manufacturer. Cultured Dairy Prod. J. 6(3):17. Stocklin, P. 1969. The production and handling of yogurt on a commercial scale. Cultured Dairy Prod. J. 4:6. Storgards, T. and O. Aule. 1958. The influence of the treatment of milk on the quality of cultured milk. Dairy Sci. Abst. 14:507. Storgards, T. 1964. IDG, Ann. Bull. Part III:65. Cited from "Yoghurt" 1978. Ed. Rasic, J. L. and J. A. Kurmann. Tech. Dairy Publishing House. Copen- hagen, Denmark. Tennous, R. J. and A. J. Merat. 1969. Analysis of samples made from milk and from yogurt: chemical physical and biologic properties of butter and ghee. J. Am. Diet. Assoc. 55:267. Tramer, J. 1973. Yogurt cultures. J. Soc. Dairy Technol. 26(1):16. Trout, G. M., C. P. Halloran, and I. A. Gould. 1935. The effect of homogenization on some of the physical and chemical properties of milk. Mich. Agr. Expt. Sta. Tech. Bul. 145, 34 pp. Trout, G. M. 1950. Homogenized Milk. Michigan State College Press, MI. 125 Vakil, J. R. and K. M. Shahani. 1962. Different path- ways of lactose metabolism of Streptococcus lactis and their sensitivity to antibiotics. Presented at the 15th Annual Meeting of the Am. Dairy Sci. Assoc. June 19, 1962. Univ. of Maryland. Veringa, H. A., T. E. Galesloot, and H. Davelaar. 1968. Symbiosis in yogurt (II). Isolation and identifi- cation of a growth factor for Lactobacillus bulgaricus produced by Streptococcus thermophilus. Neth. Milk Dairy J. 22:114. Viani, R. and I. Herman. 1976. Composition of yoghurt aroma. Nestle Research News. 1974/75. Switzer- land. Warthesen, J. J. and P. L. Kramer. 1979. Analysis of sugars in milk and ice cream by high pressure liquid chromatograhy. J. Food Sci. 44:626. Weinstein, B. R. and G. M. Trout. 1950. Susceptibility of milk from various dairy breeds to homogeniza- tion—inducted rancidity. Mich. Agr. Exp. Sta. Quat. Bull. 32(3):31. White, J. C. 1966. Yogurt processing. Am. Dairy Review 28(12):50. Williamson, W. T. 1971. Handling of starters cultures by the plant. Paper presented at the 8th Annual Marschall Invitation Italian cheese seminar Wi., May 9. Yaziciogly, A., and M. Yilmaz. 1966. The combined effect of temperature and acidity on the keeping quality of yogurt. Milchwiss. 21:87. Yeager, C. 1975. Yogurt processing methods. Cult. Dairy Prod. J. 10(4):10. Ziembia, J. V. 1970. Top-quality cultures made in unique plant. Food Eng. 42:68. Zurakynski, A., V. A. Dinesen, and S. E. Rakshy. 1975. Productos lacteos fermentados y sus propiedades dieteticos. Yoghurt, Byoghurt y Biogard. Ind. Lechera 641:10. IIIIIIIIIIIIIZIIIIEIJIIIIIIIIIIIIIIIIIIIIIIIIiIIIIIIIIIIIIIII