UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Lactulose preparation using food-safe reagents Layton, Anne Alexandra 1997

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1997-0182.pdf [ 5.13MB ]
Metadata
JSON: 831-1.0099196.json
JSON-LD: 831-1.0099196-ld.json
RDF/XML (Pretty): 831-1.0099196-rdf.xml
RDF/JSON: 831-1.0099196-rdf.json
Turtle: 831-1.0099196-turtle.txt
N-Triples: 831-1.0099196-rdf-ntriples.txt
Original Record: 831-1.0099196-source.json
Full Text
831-1.0099196-fulltext.txt
Citation
831-1.0099196.ris

Full Text

LACTULOSE PREPARATION USING FOOD-SAFE REAGENTS by ANNE ALEXANDRA LAYTON B.Sc. (Agriculture), The University of B r i t i s h Columbia, 1992 THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Food Science We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA APRIL 1997 ©ANNE ALEXANDRA LAYTON, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT Lact u l o s e i s e f f i c i e n t l y synthesized from l a c t o s e using c a t a l y s t s such as b o r i c a c i d and t r i e t h y l a m i n e . However, since n e i t h e r c a t a l y s t i s food-safe, both must be removed a f t e r p r o c e s s i n g . Lactulose i s a l s o produced i n a d v e r t e n t l y during heat treatment of d a i r y products, although i n small q u a n t i t y . Studies have i n d i c a t e d t h a t a l t e r i n g the heat processing c o n d i t i o n s can improve l a c t u l o s e y i e l d . A high l a c t u l o s e , mixed carbohydrate p r e p a r a t i o n was produced without the use of t o x i c c a t a l y s t s . Using two Taguchi's f r a c t i o n a l f a c t o r i a l designs, e i g h t f a c t o r s were t e s t e d as t o t h e i r i n f l u e n c e on l a c t u l o s e y i e l d : pH, l a c t o s e , NaOH, c i t r a t e and phosphate co n c e n t r a t i o n s , heating temperature and d u r a t i o n , and p u r i f i c a t i o n of the l a c t o s e substrate. In the f i r s t d esign, l a c t o s e c o n c e n t r a t i o n (at l e v e l s of 40, 79, and 155 mg/mL) , pH (9.0, 10.5, and 12.0), heating temperature (90, 110, and 130°C), c i t r i c a c i d c o n c e n t r a t i o n (40, 70, 100 mM) and i n the second design, NaOH con c e n t r a t i o n (18, 50, and 100 mM) , was shown t o s i g n i f i c a n t l y i n f l u e n c e l a c t u l o s e y i e l d . A l l other f a c t o r s d i d not s i g n i f i c a n t l y i n f l u e n c e l a c t u l o s e y i e l d at the s e l e c t e d l e v e l s . The i n t e r a c t i o n s of l a c t o s e , c i t r a t e , and phosphate concentrations of the f i r s t design a l s o s i g n i f i c a n t l y i n f l u e n c e d l a c t u l o s e y i e l d . The c o n d i t i o n s s e l e c t e d f o r the conversion of l a c t o s e t o l a c t u l o s e was d e c a l c i f i e d whey permeate at > 70 mg/mL l a c t o s e , a pH of 10.5-11.0, w i t h an added 50 mM sodium c i t r a t e , was heat t r e a t e d a t 110°C f o r 10 minutes. Approximately 30% of i n i t i a l l a c t o s e was converted t o l a c t u l o s e v i a p r i m a r i l y the Lobry de Bruyn and Alberda van I l l Ekenstein transformation. Again using a Taguchi design, four f a c t o r s were t e s t e d t o i f they s i g n i f i c a n t l y i n f l u e n c e d the p r e f e r e n t i a l p r e c i p i t a t i o n of l a c t o s e over l a c t u l o s e i n a cooled aqueous s o l u t i o n : pH, sugar c o n c e n t r a t i o n , temperature decrease, and f i n a l temperature. The pH of the mixed carbohydrate s o l u t i o n (at l e v e l s of 7.0, 9.0, and 10.7) and sugar concentration (29, 39, and 52%) both s i g n i f i c a n t l y i n f l u e n c e d e i t h e r the l a c t u l o s e y i e l d of p r e c i p i t a t i o n or the sugar r a t i o i n the decant. For f u r t h e r study, the l a c t u l o s e p r e p a r a t i o n was concentrated to approximately 50% s o l i d s and pH 10.5, cooled from 65° C t o 20° C at 5C°/hour, and h e l d f o r 24 hours, p r e f e r e n t i a l l y p r e c i p i t a t i n g l a c t o s e over l a c t u l o s e . A f t e r one c o o l i n g c y c l e , there was a l a c t u l o s e y i e l d of approximately 82% and a 1:1 l a c t u l o s e : l a c t o s e r a t i o . A f t e r a second p r e c i p i t a t i o n of the decanted p o r t i o n , there was a 78% l a c t u l o s e y i e l d and a 3.4:1 l a c t u l o s e : l a c t o s e r a t i o . There was a t o t a l l o s s of about 4 0% of l a c t u l o s e through the two p r e c i p i t a t i o n c y c l e s . Ion-exchange columns removed the m a j o r i t y of the n a t u r a l and added s a l t s from the l a c t u l o s e preparations. A c t i v a t e d c h a r c o a l removed most of the brown c o l o u r of the p r e p a r a t i o n but a l s o 3 0% of the s o l i d s . The f i n a l syrup contained 59% l a c t u l o s e , 2 6% l a c t o s e , 5.0% g a l a c t o s e , 1.0% glucose, and 0.81% f r u c t o s e , based on t o t a l s o l i d s . Carbohydrates were assayed using an enzymatic spectro-photometric method. An u n i d e n t i f i e d substance was detected u s i n g t h i n - l a y e r chromatography of carbohydrates. i v TABLE OF CONTENTS Abstract i i Table of Contents i v L i s t of Tables i x L i s t of Figures v i i Acknowledgement x i 1. INTRODUCTION 1 2. LITERATURE REVIEW . . 3 2.1 LACTULOSE 3 2.la Physical properties of lactulose 3 2.lb Lactulose i n digestion 3 2.Ic Lactulose i n medical treatment 5 2.Id Commercial products 7 2.2 WHEY 8 2.3 LACTULOSE AS AN INTERMEDIATE IN THE DEGRADATION OF LACTOSE 11 2.3a Lobry de Bruyn and Alberda van Ekenstein transformation 11 2.3b Mailla r d reaction and Amadori rearrangement 15 2.4 LACTULOSE CONVERSION 17 2.4a Overview of conversion methods 17 2.4b Borate and triethylamine c a t a l y s t s 17 2.5 FACTORS AFFECTING LACTULOSE CONVERSION IN HEAT PROCESSING 18 2.5a The influence of temperature and heating time 18 2.5b The influence of pH 19 2.5c The influence of c i t r a t e and phosphate . 2 0 2.5d The influence of lactose concentration . 20 2.5e The influence of demineralization 21 2.5f Other influences on lactulose y i e l d .... 22 2.6 FRACTIONAL FACTORIAL EXPERIMENTAL DESIGN 22 2.7 PURIFICATION OF LACTULOSE 25 2.7a Overview of lactulose p u r i f i c a t i o n methods 25 2.7b Cold temperature p r e c i p i t a t i o n of lactose 25 V 3. MATERIALS AND METHODS 28 3.1 RAW MATERIALS 2 8 3.2 THE CONVERSION OF LACTOSE TO LACTULOSE . 28 3.2a Sample preparation 28 3.2b Heat processing 33 3.2c Calculation of lactulose y i e l d 35 3.2d S t a t i s t i c s for f r a c t i o n a l f a c t o r i a l designs 36 3.3 CONTINUOUS FLOW CONVERSION OF LACTOSE TO LACTULOSE 3 6 3.3a Design of heat exchanger 3 6 3.3b Determining heating times 37 3.4 PURIFICATION 37 3.4a Materials 37 3.4b Fractional f a c t o r i a l design for cold p r e c i p i t a t i o n 39 3.4bl Sample preparation 39 3.4b2 Calculation of lactulose y i e l d and sugar r a t i o 43 3.4c The second cycle of cold p r e c i p i t a t i o n . 44 3.4d Calcium phosphate removal 45 3.4e Demineralization 45 3.4f Decolourization 47 3.5 PROXIMATE ANALYSIS 47 3.5a Sampling throughout the process 48 3.5b Enzymatic carbohydrate assays 48 3.5bl Quantitative assays for lactulose, lactose, glucose, and fructose 48 3.5b2 A c t i v i t y of beta-galactosidase 51 3.5b3 Quantitative assay for galactose 54 3.5b4 Standard curves for carbohydrate assays 54 3.5c Thin layer chromatography assay of carbohydrates 65 3.5cl Qualitative assay of carbohydrates 65 3.5c2 Tagatose i d e n t i f i c a t i o n by TLC and spectrophotometry 66 3.5d Determination of nitrogen 70 3.5dl Determination of t o t a l nitrogen 7 0 3.5d2 Determination of protein 70 3.5e Determination of t o t a l s o l i d s and ash .. 71 3.5f Determination of pH and t i t r a t a b l e a c i d i t y 71 v i 4. RESULTS AND DISCUSSION 74 4.1 OVERVIEW OF PROCESS 74 4.2 THE CONVERSION OF LACTOSE TO LACTULOSE 7 6 4.2a The i n f l u e n c e of pH and NaOH co n c e n t r a t i o n . 79 4.2b The i n f l u e n c e of sodium phosphate concentration 82 4.2c The i n f l u e n c e of c i t r a t e c o n c e n t r a t i o n 84 4.2d The i n f l u e n c e of l a c t o s e c o n c e n t r a t i o n and p u r i f i c a t i o n 87 4.2e The i n f l u e n c e of temperature and time of heat treatment 90 4.2f Continuous flow heat exchanger 90 4.3 PURIFICATION 92 4.3a F r a c t i o n a l f a c t o r i a l design f o r c o l d p r e c i p i t a t i o n 92 4.3b The second c y c l e of c o l d p r e c i p i t a t i o n ..... 97 4.3c D e m i n e r a l i z a t i o n and d e c o l o u r i z a t i o n 99 4.4 PROXIMATE ANALYSIS 101 4.4a Carbohydrate standard curves using enzymatic assays 101 4.4b A c t i v i t y of beta-galactosidase 102 4.4c T h i n - l a y e r chromatography q u a l i t a t i v e assay of carbohydrates 103 4.4d Tagatose i d e n t i f i c a t i o n by TLC and spectrophotometer 106 4.4e Determination of t o t a l n i t r o g e n and p r o t e i n . 107 5. CONCLUSIONS 109 6. ABBREVIATIONS 111 7. REFERENCES 112 v i i LIST OF FIGURES Figure 1. The beta-furanose isomer of free lactulose .... 4 Figure 2. Possible degradative reaction routes of lactose i n milk; gal refers to galactosyl or galactose 13 Figure 3. Amadori rearrangement of lactosyl-amino product of the Maillard reaction i n milk to form a lactulosyl-amino compound 16 Figure 4. Overall process for the conversion of lactose to lactulose and p a r t i a l p u r i f i c a t i o n of lactulose 29 Figure 5. Schemes used i n the L 2 7(3 1 3) design for the conversion of lactose to lactulose during heat treatment of whey permeate 31 Figure 6. The heat exchanger with thermocouples determined come up and cool down times: top) the pressure t i g h t system, bottom) the heating or cooling c o i l with thermocouples i n sequence 38 Figure 7. Schemes used i n the Taguchi design L^ 7(3 1 3) for the p r e f e r e n t i a l cold p r e c i p i t a t i o n of lactose, following the design on Table 4 42 Figure 8. The standard curve of the absorbance of O-nitrophenol at 42 0 nm to determine the a c t i v i t y of the beta-galactosidase "Lactase 100,000" using ONPG hydrolysis 53 Figure 9. The standard curve of glucose using an enzymatic spectrophotometric assay 57 Figure 10. The standard curve of galactose using an enzymatic spectrophotometric assay 59 Figure 11. The standard curve of lactose using an enzymatic spectrophotometric assay 60 Figure 12. The standard curve of lactulose using an enzymatic spectrophotometric assay 61 Figure 13. The standard curve of fructose using an enzymatic spectrophotometric assay 63 Figure 14. The standard curve of the absorbance of tagatose concentrations at 256 nm 69 v i i i Figure 15 Figure 16 a, b Figure 17 a, b Figure 18, Figure 19, a, b Figure 20, Figure 21. a,b Figure 22 The standard curve of the absorbance of bovine gamma globulin protein at 595 nm for the Bio-Rad Protein Assay 72 The influence of pH and NaOH concentration on lactulose y i e l d during heat treatment of whey permeate using an L 2 7(3 1 3) f r a c t i o n a l f a c t o r i a l design The influence of sodium phosphate concentration interacting with lactose and c i t r i c a cid concentrations on lactulose y i e l d of heat treated whey permeate using an L 2 7(3 1 3) design The influence of c i t r i c acid / sodium c i t r a t e concentration on lactulose y i e l d during heat treatment of whey permeate using combined re s u l t s of L 2 7(3 1 3) designs #1 and #2 The influence of c i t r i c acid concentration interacting with lactose and sodium phosphate concentrations on lactulose y i e l d during heat 81 83 85 treatment of whey permeate using an L 2 7(3 design 13 The influence of lactose concentration on lactulose y i e l d during heat treatment of whey permeate using L 2 7(3 1 3) design #1 86 88 The influence of lactose concentration interacting with sodium phosphate and c i t r i c acid concentrations on lactulose y i e l d during heat treatment of whey permeate using an L ? 7(3 1 3) design J27 The influence of temperature on lactulose y i e l d during heat treatment of whey permeate using the combined res u l t s of L 2 7(3 1 3) designs #1 and #2 89 91 Figure 23 a, b Figure 24 The influence of pH and sugar concentration on the cold p r e c i p i t a t i o n of a lactulose: lactose solution using an L 2 7(3 1 3) design 97 Lactulose preparation at three stages of processing, from l e f t to r i g h t - d e c a l c i f i e d UF whey permeate (24.7% TS), a f t e r p r e c i p i t a t i o n (10.4% TS), and a f t e r deionization and decolourization (23.3% TS) .. 100 Figure 25. TLC on s i l i c a plates shows standard sugar solutions at varying concentrations and whey at d i f f e r e n t stages of process 104 ix LIST OF TABLES Table 1. Composition of a t y p i c a l lactulose syrup 6 Table 2. Lactulose contents of some commercial products 9 Table 3. Composition of whey and u l t r a f i l t r a t i o n whey permeate from Cheddar cheese 10 Table 4. One example of Taguchi's experimental design, orthogonal array L 2 7(3 1 3) 24 Table 5. S o l u b i l i t i e s of lactose and lactulose i n water at various temperatures 27 Table 6. Condition factors and t h e i r assigned l e v e l s of heat processed deproteinized whey investigated for influence on lactulose y i e l d using an L 2 7(3 1 3) design 3 0 Table 7. Heat processing of the two L 2 7(3 1 3) experimental designs for the conversion of lactose to lactulose 34 Table 8. Heat processing conditions for lactulose production from whey selected using Taguchi's f r a c t i o n a l f a c t o r i a l designs 40 Table 9. Condition factors and t h e i r assigned l e v e l s of a cold precipitated lactulose: lactose solution using an L 2 7(3 1 3) design 41 Table 10. Standard solutions containing varying r a t i o s of f i v e carbohydrates prepared for standard curves of a l l enzymatic assays 55 Table 11. Analysis of variance t e s t i n g the s i g n i f i c a n c e of slope and l i n e a r i t y of the regressional curve calculated for the glucose standard curve 58 Table 12. Analysis of variance te s t i n g the s i g n i f i c a n c e of slope and l i n e a r i t y of the regression curve calculated for the lactulose standard curves with a t - t e s t comparison of the slope and elevation of the curves 62 Table 13. Analysis of variance t e s t i n g the s i g n i f i c a n c e of slope and l i n e a r i t y of the regression curve calculated for the fructose standard curves with a t - t e s t comparison of the slope and elevation of the curves 64 X Table 14. The absorbance of aqueous carbohydrate solutions and the TLC solvent by spectrophotometer between 200 and 400 nm 68 Table 15. Proximate analysis, based on % t o t a l s o l i d s , of whey permeate at d i f f e r e n t stages of lactulose manufacture and p u r i f i c a t i o n 75 Table 16. Design factor combinations and r e s u l t s summary of the L 2 7(3 1 3) design #1 77 Table 17. Design factor combinations and r e s u l t s summary of the L 2 7(3 1 3) design #2 78 Table 18. Analysis of variance [Taguchi L 2 7(3 1 3) ] design #1 and #2, obtained from 27 experiments of heat processed demineralized whey permeate 80 Table 19. Results of the L 2 7(3 1 3) design, assaying lactose and lactulose i n both decant and pre c i p i t a t e portions aft e r cold p r e c i p i t a t i o n of a 28:72 lactulose:lactose s o l u t i o n 93 Table 20. Analysis of variance of the L 2 7(3 1 3) design of cold p r e c i p i t a t i o n i n a lactulose:lactose solution 95 Table 21. Results of 5 repl i c a t e s of a second cycle of cold p r e c i p i t a t i o n i n a lactulose: lactose solution 98 Table 22. Rf values of carbohydrate standards and various samples using thin-layer chromatography on a derivatized s i l i c a plate . 105 x i ACKNOWLEDGEMENTS The author would l i k e to thank Dr. Durance for his encouragement, advice and support throughout t h i s t h e s i s . Also to Canadian Inovatech Inc., Abbotsford, BC for t h e i r i n t e r e s t i n the study and f i n a n c i a l support. She also extends sincere appreciation to the other members of the supervisory committee, Dr. Vanderstoep, Dr. Li-Chan, and Dr. Nakai, for t h e i r input, and to Mr. S. Yee for te c h n i c a l assistance during the course of t h i s study. 1 1. INTRODUCTION Lactulose, a synthetic isomer of lactose, usually sold as the major constituent i n a mixed carbohydrate syrup, has strong worldwide markets. As a pharmaceutical i t i s used i n the treatment of chronic constipation and portal systemic encephalopathy. As an ingredient i n health foods and infant formulae i t i s a digestive aid, acting as an energy source for b e n e f i c i a l bacteria i n the colon. P u r i f i e d lactose i s e f f i c i e n t l y converted to lactulose by a l k a l i n e enolization using sodium borate and triethylamine as c a t a l y s t s . Lactulose yi e l d s of 86% have been reported using these c a t a l y s t s (Hicks et a l . , 1984). Since neither sodium borate nor triethylamine are food-safe, both must be removed a f t e r processing. Lactulose i s also produced in small quantities during thermal processing of evaporated and UHT milks. Its y i e l d reaches only approximately 4% i n these products. S t i l l , studies have indicated that a l t e r i n g thermal processing conditions and milk composition can influence lactulose production (Andrews, 1989; Martinez-Castro & Olano, 1980). Some studies have further suggested that using p u r i f i e d lactose as the substrate for lactulose production i s not necessary (Hicks et a l . , 1984). Whey, a waste product of cheese manufacture, may prove a more economical substrate for lactulose production. There was no published research found i n which the conditions of thermal processing were manipulated to maximize lactulose y i e l d i n whey using only food-safe reagents. In the present study, a process for lactulose production was investigated using food-safe reagents and using the lactose i n deproteinized whey as the substrate. The goal was to produce a colourless, high-lactulose, low-lactose mixed carbohydrate preparation. The main objectives of t h i s thesis were to: 1. Evaluate eight thermal processing factors i n influencing lactulose y i e l d using a f r a c t i o n a l f a c t o r i a l experimental design -Temperature of heat treatment -Duration of heat treatment -Whey permeate demineralization -pH -NaOH concentration -Lactose concentration -Added sodium c i t r a t e / c i t r i c acid concentration -Added sodium phosphate concentration 2. Evaluate four cold p r e c i p i t a t i o n factors i n influencing the p r e f e r e n t i a l p r e c i p i t a t i o n of lactose over lactulose, again u t i l i z i n g a f r a c t i o n a l f a c t o r i a l experimental design -Cooling rate - F i n a l temperature -pH -Carbohydrate concentration 3. Investigate further p u r i f i c a t i o n processes of the lactulose sugar mixture, including demineralization by ion exchange and decolourization by activated charcoal. 4 . Identify the changes occurring i n sample composition as the whey permeate was heat-treated and p u r i f i e d . Numerous components of the whey permeate were measured at various stages of processing. 3 2. LITERATURE REVIEW 2.1 LACTULOSE 2.la Physical properties of Lactulose Lactulose (4-0-Beta-D-galactopyranosyl-D-fructose) i s a synthetic disaccharide of molar mass 342.30 g/mol and a melting point of 169°C. I t i s a white, c r y s t a l l i n e , sweet t a s t i n g powder. The disaccharide i s composed of a galactose moiety linked to a fructose moiety by a 1 to 4 beta-glycosidic linkage. There are f i v e possible isomers of lactulose: the fructose moiety i n the form of an alpha or beta pyranose, an alpha or beta furanose, or lactulose i n an a c y c l i c form. Using current methods of lactulose production three isomers predominate: the beta furanose (Figure 1) , the beta pyranose, and the alpha furanose at a r a t i o of 0.74 5, 0.155, and 0.100 respectively (Jeffrey et a l . , 1983). 2.lb Lactulose i n digestion Lactulose was f i r s t synthesized from lactose i n 1929 by Montgomery and Hudson (1930), but i t was not u n t i l 1957 that the si g n i f i c a n c e of lactulose as a potential aid i n digestion was recognized (Petuely, 1957). Unlike many other carbohydrates such as lactose, lactulose i s not hydrolyzed by digestive enzymes i n the small i n t e s t i n e to be absorbed as monosaccharides into the bloodstream. Instead, i t i s transported to the large in t e s t i n e where i t i s hydrolyzed by resident saccharolytic microflora. Using lactulose as an energy source, some gram p o s i t i v e bacteria such as Bi f i d o b a c t e r i a species produce l a c t i c and a c e t i c acids, lowering pH and r a i s i n g osmotic pressure. These i n turn soften i n t e s t i n a l contents and may prevent the p r o l i f e r a t i o n of some gram negative Figure 1 The beta-furanose isomer of free lactulose . Of 5 possible isomers of free lactulose beta-furanose isomer (74.5% of t o t a l ) beta-pyranose isomer (15.5% of t o t a l ) alpha-furanose isomer (10.0% of t o t a l ) alpha-pyranose isomer a c y c l i c isomer (Jeffrey et a l . , 1983) 5 putrefactive and pathogenic species such as Clostridium, Bacteroides. Salmonella. Shigella. and Escherichia c o l i (Haenel, 1970; Mata et a l . , 1969a; Mata et a l . , 1969b; Rose & Gyorgy, 1955). 2.1c Lactulose i n medical treatment Lactulose i s used i n the treatment of chronic constipation and i n the prevention and treatment of p o r t a l systemic encephalopathy. I t i s available as an impure syrup containing about 50% (w/w) lactulose with lesser amounts of lactose, other sugars and acid. Commercial syrups such as "Duphalac" and "Laevolac" are s i m i l a r to the product described i n Table 1. In the treatment of chronic constipation, the primary actions of lactulose are: 1. the a c i d i f i c a t i o n of the colon's contents by organic acid production during b a c t e r i a l hydrolysis of lactulose, 2. the increase of osmotic pressure, and 3. a laxative e f f e c t . The dosage range for chronic constipation i s 15 to 45 mL of lactulose syrup d a i l y depending on the severity of the i l l n e s s (Avery et a l . , 1972). Portal systemic encephalopathy (PSE), a complication of advanced hepatic c i r r h o s i s , i s caused by ammonia and other nitrogenous substances acting as toxins i n the brain. Although i t s action i s not yet completely understood, lactulose may decrease the production, decrease the absorption, and/or increase the excretion of ammonia i n the digestive t r a c t (Kosman, 1976). The dosage range for PSE i s 3 0 to 50 mL of lactulose syrup three times d a i l y , or Table 1 Composition of a t y p i c a l lactulose syrup. 1 Contents g / 100 mL Lactulose 63 - 70 Lactose 4 - 8 Galactose 8 - 13 Other sugars 4 - 8 Orange flavours 2 C i t r i c acid 0.16 General dosage: Chronic constipation-15 - 45 mL/day Portal systemic encephalopathy-90 - 150 mL/day (Avery et a l . , 1972) i 7 enough to lower faeces pH from a normal 7 to a pH of 5.0 to 5.5 (Avery et a l . , 1972). The t o x i c i t y of lactulose taken o r a l l y was reported to be s i m i l a r to sucrose when tested using rat s (Okumura & Gomi, 1973). When acute t o x i c i t y of lactulose was tested on adult baboons, the animals survived a t o t a l dose of 37 mL lactulose syrup/kg body weight, close to the volume of t h e i r stomachs. Subacute l e v e l s of 20 mL/kg body weight d a i l y for six months showed no i l l e f f e c t s in young baboons apart from a decreased growth rate (Avery et a l . , 1972) . S t i l l , an overdose of t h i s potent laxative can lead to diarrhoea and severe dehydration. 2.Id Commercial products Lactulose i s a component of various dairy products, either added as an ingredient or occurring during the heat treatment of the food. I t i s often added to those dairy products s p e c i f i c a l l y marketed as therapeutic digestive aids. The popularity of these s p e c i a l t y products containing lactulose (and/or Bifidobacteria) i s currently centred i n Japan and Europe, and nominally i n the United States. Infant formulae have contained lactulose, at lea s t i n one brand, since 1960. I t was included to ensure that bottle-fed infants had a sim i l a r i n t e s t i n a l microflora, with a high B i f i d o b a c t e r i a faecal count, as breast-fed infants (Mizota & Tamura, 1987). For most infant formulae, the heat treatment during processing produces the lactulose i n the product. Spray dried and ultra-high temperature (UHT) formulae contain very l i t t l e l actulose, while in-container s t e r i l i z e d formulae can contain anywhere from 55 to 469 mg/100 mL of lactulose (Beach & Menzies, 8 1983) . Some other dairy foods, e s p e c i a l l y s t e r i l i z e d milks, can contain lactulose at r e l a t i v e l y high l e v e l s . Studies of Beach and Menzies (1983) and Hendrickse et a l . (1977) suggested the high end of the lactulose range found i n some in-container s t e r i l i z e d milks and infant formulae approached a l e v e l high enough to have a laxative e f f e c t . Table 2 d e t a i l s the lactulose content of various dairy foods. 2.2 WHEY Whey i s the f l u i d obtained by separating the coagulum from milk or cream. I t i s generally produced either as a sweet or an acid whey. Sweet whey, with a pH above 5.5, i s produced during the manufacture of cheese or rennet casein. Acid whey i s produced during the manufacture of cottage cheese, l a c t i c casein, or mineral acid casein (Sienkiewicz & Riedel, 1990a) . The disaccharide beta-lactose (4-0-FJ-D-galactopyranose) , i s a major constituent of whey along with proteins, minerals, acid, and water. Table 3 provides a breakdown of whey's t y p i c a l components. Whey i s a waste product of the cheese industry produced at a rate ranging between 7.5 kg whey/kg soft cheese and 11.3 kg whey/kg hard cheese. It i s used i n the production of a wide range of products including lactose, whey powder, lactalbumin, animal feeds, and fermentation media (Sienkiewicz & Riedel, 1990a,b). Often, whey protein i s isolated from whey to be sold as a commercial ingredient. Through a process of u l t r a f i l t r a t i o n (UF) , whey can be separated by membrane f i l t r a t i o n into two portions. The permeate consists mainly of lactose and s a l t s (Table 3). The portion remaining on the membrane contains the majority of protein, Table 2 Lactulose contents of some commercial products. Product L a c t u l o s e (g/100 g) (mg/100 mL) "BF-T" I n f a n t formula 1 "Hounyu" m i l k powder1 "Sawayaka" sour m i l k 1 P a s t e u r i z e d m i l k 2 S t e r i l i z e d m i l k 2 Powdered mi l k r e c o n s t i t u t e d 2 UHT m i l k 2 0.5 0.3 (>3xl0 7) 3 4 ( > l x l 0 8 ) 3 4 - 1 5 80 -200 2.5-30 10 - 30 1 (Mizota et a l . , 1987) 2 (Andrews, 1989) 3 B i f i d o b a c t e r i u m a d d i t i o n (/g) Table 3 Composition of whey and u l t r a f i l t r a t i o n whey permeate from cheddar cheese. Constituent Permeate Whey Total s o l i d s (%) 5. 7 6. 7 Total nitrogen (mg/g) 0. 26 1. 30 Protein (%) 0. 01 0. 60 Protein removal (%) 98. 4 Non protein nitrogen (mg/g) 0. 24 0. 34 Ash (%) 0. 50 0. 52 Lactose (%) 4. 9 5. 0 La c t i c acid (%) 0. 14 0. 14 PH 6. 1 6. 1 (Hargrove et a l . , 1976) 11 including lactalbumin and lactoglobulin (Hargrove et a l . , 1976). S t i l l , the supply of whey far exceeds the demand. Excess whey i s commonly disposed of by spreading on land or treated as waste water. The high b i o l o g i c a l oxygen demand (BOD) value of whey (approximately 40 g 02/L) makes i t s disposal an environmental dilemma (Sienkiewicz & Riedel, 1990c). 2.3 LACTULOSE AS AN INTERMEDIATE IN THE DEGRADATION OF LACTOSE Lactulose occurs as an intermediate i n the degradation of lactose. There are two methods by which lactose i n milk or whey i s degraded v i a free or bound lactulose: 1. Lobry de Bruyn and Alberda van Ekenstein transformation, and 2. the in t e r a c t i o n of amino acids with lactose i n the Ma i l l a r d reaction and subsequent Amadori rearrangement. One reported study heated milk for 2 0 minutes at a temperature of 110°C-150°C. Eighty percent of the degraded lactose followed the LA transformation and 20% followed the M a i l l a r d reaction (Berg & van Boekel, 1994). 2.3a Lobry de Bruyn and Alberda van Ekenstein transformation The Lobry de Bruyn and Alberda van Ekenstein (LA) transformation occurs mostly under a l k a l i n e conditions and i s the dominant reaction for t h i s study. This reaction i s not s p e c i f i c to lactose but describes more than 50 si m i l a r reactions, the common feature being an enediol intermediate (Davidson, 1967). Most occur i n the presence of a base catalyst. Common to nearly a l l recent descriptions of the LA transformation of lactose i s the degradation of lactose to 12 lactulose, (compounds 1 to 4a) i n Figure 2. Under the influence of a base cat a l y s t , the glucose moiety of lactose i s isomerized to fructose v i a an intermediate enolization. Beyond these i n i t i a l reactions, studies have not always agreed upon pathways or reaction products. Figure 2 further i l l u s t r a t e s the possible degradative reaction routes of lactulose i n milk according to Berg and van Boekel (1994). In t h e i r study, the main reaction products at t r i b u t e d to the LA transformation of milk or model milk solutions - heated to between 100°C and 150°C and assayed by HPLC - were lactulose (compound 4a), galactose (compounds 6a,b), formic acid (compounds 9a,b), and various C5 and C6 compounds. The formic acid produced was suggested as an explanation for much of the decrease i n pH during heating. The C5 compounds were believed to be formed along with the formic acid. Possible C5 compounds included 2-deoxyribose (compound 10), 3-deoxypentulose (compound 16) and/or f u r f u r o l (compound 17) . Small amounts of f u r f u r a l and hydroxymethylfurfural (compound 8) were also measured. Reaction intermediate C6 compounds such as deoxyosones (compounds 7,13) were thought to be formed but were not i d e n t i f i e d . It was suggested that the main route for lactose degradation during LA transformation was through 4-deoxyosone (compound 13) which was not considered an intermediate during Maillard/Amadori reactions (Ames, 1992). Berg and van Boekel also stressed that these were not simple reaction routes; reactions were occurring simultaneously and products could be formed through d i f f e r e n t routes. In Berg and van Boekel's study (1994), epilactose, tagatose, and isosaccharinic acids were not detected. The small amount of 13 HC=0 I HC-OH I HO-CH -I * HC-O-Gal I HC-OH I CH20H 3 HC-OH II C-OH I HO-CH —3 I ^ HC-O-Gal I HC-OH I CH20H 5 HC=0 I C-OH II CH 7 HC=0 I c=o I CH2 CH20H >HC=0 " O H HC-O-Gal G a l H ? " O H I 6a HC-OH Hydroxymethylfurfural It It ^> 4a I CH20H 4b HC=0 HC-OH I CH20H HC-OH Formic acid 9a + 1 HC-OH I HC-OH I HO-CH I HO-C-Gal I CH I CH20H Lactose CH20H I c=o I HO-CH I HC-O-Gal I HC-OH I CH20H Lactu lose HO-CH I HO-CH I Gal-C-OH I HC-OH I CH20H Epi lactose 14b 0 II HC-OH 1 HO-C-CH20H I CH2 I HC-OH I CH20H H C - O I CH2 I HC-OH I HC-OH I CH20H 10 2-Deoxyribose Isosaccharinic Ac id 9b Formic acid It 11 CH20H I C-OH II HO-C I HC-O-Gal I HC-OH I CH20H Gal 6b 12 CH20H I c=o I C-OH II CH I HC-OH I CH20H 13 CH20H I c=o I c=o I CH2 I HC-OH I CH20H CH20H Furfurol \ 14 15 HC-OH HC=0 II I C-OH I HC-OH 1 c=o > 1 c=o I ^ — 1 CH2 CH2 | HC-OH 1 HC-OH / 16 CH20H I c=o I CH2 I HC-OH I CH20H c = 0 3-Deoxypentulose i CH20H I CH20H Figure 2 Possible degradative reaction routes of lactose i n milk; gal re f e r s to galactosyl or galactose. (Berg and van Boekel, 1994) 14 epilactose thought possibly produced may have had i t s HPLC peak obscured by another carbohydrate. Tagatose was detected only i n model solutions which contained additional galactose. Epilactose (compound 4b) has been described as a reaction product of lactose degradation i n other studies. In one, i t was detected by GLC after degradation of lactulose (Olano & Martinez-Castro, 1981). Berg and van Boekel (1994) considered 1,2-enediol as the intermediate (compound 3) . In a study by Olano et a l . (1989), using a buffered 5% lactose solution (pH 6.6) heated at 120°C for 20 minutes and a GLC assay, 77.2 mg/100 mL of epilactose were formed (in conjunction with 474.2 mg/100 mL of produced l a c t u l o s e ) . In another study, epilactose was detected by GLC, but not at temperatures below 120°C (Martinez-Castro et a l . , 1986). D-Tagatose i s a product of D-galactose degradation along with formic acid, whether the lactose reaction follows the LA transformation or the Maillard reaction / Amadori rearrangement route. Tagatose can then be further degraded to t r i o s e s , then oxidized to acids. While Berg and van Boekel (1994) did not detect tagatose using HPLC, Troyano et a l . (1992) found 13 mg/L i n milk heated at 120°C for 20 minutes using GLC. Isosaccharinic acids have been described as lactose degradation products and often as the major organic acids produced (Olano & Martinez-Castro, 1981). Corbett and Kenner (1953) found beta-isosaccharinic acid and tagatose as degradative products of a l k a l i treated lactose; t h i s acid was also detected by Caru b e l l i (1966). Isosaccharinic acids are thought to occur through the beta-elimination degradation of lactulose (compounds 4a to 14b, in Figure 2. Apart from formic and isosaccharinic acids, other acids, 15 such as l a c t i c and acetic acids have also been included as reaction products (Gould, 1945; Keeney et a l . , 1950). 2.3b MaiHard reaction and Amadori rearrangement The M a i l l a r d reaction occurs in milk by the elimination of a water molecule, joining the carbonyl group of lactose to an amino acid. The amino acid i s often protein-bound and i s usually lysine. This p a i r i n g forms an amino glycoside (again often protein-bound), named l a c t o s y l - l y s i n e . Amadori rearrangement transforms l a c t o s y l -l y s i n e to i t s isomer l a c t u l o s y l - l y s i n e , Figure 3 (Andrews, 1986). While early reports suggested that t h i s reaction could produce free lactulose upon further degradation, i t i s now confirmed that t h i s does not occur (Berg and van Boekel, 1994). Free lactulose has been produced i n the laboratory from Amadori rearrangement reactions but these have only occurred under s p e c i f i c conditions such as t r e a t i n g lactose with p-toluene and hydrochloric acid (Adachi & Patton, 1961) . The main reaction products of heated milk and model milk solutions following these reactions, according to the Berg and van Boekel study (1994), were l a c t u l o s y l - l y s i n e , galactose, formic acid, hydroxymethylfurfural, and other M a i l l a r d products. They ^suggested that the formation of galactose; formic acid, and C5 and C6 compounds from lactulose degradation occurred whether i n i t i a t e d by the LA transformation or the Maillard reaction (Figure 2). 16 R N H I - C H I H C - O H + H -H O - C H H O - C - G a l I C H I C H 2 0 H RNH2+ I - C H H C - O H I H O - C H I H O - C - G a l I C H C H 2 0 H RNH + II HC I H C - O H I HO-CH H O - C - G a l I H C - O H I C H 2 0 H R N H I HC Ii C - O H H O - C H I H O - C - G a l H C - O H I C H 2 0 H R N H I C H 2 I c=o I H O - C H I H O - C - G a l H C - O H I C H 2 0 H Lactosyl-amino Lactulosyl-amino compound compound Figure 3 Amadori rearrangement of lactosyl-amino product of the Mailla r d reaction i n milk to form a lactulosyl-amino compound. (Andrews, 1986) 17 2.4 LACTULOSE CONVERSION 2 . 4a Overview of conversion methods Lactulose was f i r s t synthesized from lactose i n 1929 i n an a l k a l i n e solution of lime. Less than 20% of the lactose was converted to lactulose (Hicks & Parrish, 1980). Since then, lactulose conversion from lactose has been catalyzed using aluminates, magnesium oxide, ion exchange resins, and beta-galactosidases, among others, with varying degrees of success. Using calcium hydroxide as a catalyst produced a preparation with lactulose comprising 15% of t o t a l sugar; using sodium aluminate produced a 73% lactulose mixture (Parrish et a l . , 1980). Using beta-galactosidase as a catalyst, only 8% of lactose was converted to lactulose (Vaheri & Kauppinen, 1978). The combination of borate and triethylamine has one of the highest conversion rates. 2.4b Borate and triethylamine cata l y s t s Borate i s an e f f i c i e n t catalyst because i t complexes with lactulose, s h i f t i n g the equilibrium established during isomerization i n favour of lactulose and prevents degradative side-reactions. Using borate, a lactulose y i e l d of 70-80% has been achieved. However, i t i s impractical to use borate alone as borate: sugar r a t i o s of 50:1 are needed to reach these high y i e l d s (Mendicino, 1960). Triethylamine i s a t e r t i a r y amine added to r a i s e pH. Alone, i t can produce lactulose y i e l d s of 32% but with high a l k a l i n e degradation (Parrish, 1970). Combining the two and using a boric acid lactose r a t i o of 1:1, high lactulose y i e l d s can be achieved with low monosaccharide production and few degradative side-reactions. The rate of lactulose conversion increases with 18 b a s i c i t y from the addition of triethylamine (Hicks & Parrish, 1980). Both boric acid and triethylamine are t o x i c and must be removed, but can be recovered. Boric acid can be recovered using methanol; triethylamine can be removed by heating the soluti o n and cold trapping the evaporation (Hicks & Parrish, 1980). A study of isomerization of lactose using boric acid and triethylamine at 70°C for several hours at pH 11 produced a lactulose y i e l d of 86%. The f i n a l product contained the following sugars: lactulose at 86.8% of t o t a l carbohydrates, 8.1% lactose, 4.1% galactose, and 1.0% tagatose (Hicks et a l . , 1984). 2.5 F A C T O R S A F F E C T I N G L A C T U L O S E CONVERSION I N H E A T P R O C E S S I N G 2.5a T h e influence o f temperature and heating time Temperature and heating time have been tested i n combination and have been shown in several studies to influence the production of lactulose. Geir and Klostermeyer (1983) measured lactulose using an enzymatic assay after heating skimmed milks at various temperatures - 60, 70, 80, and 90° C - at f i v e time inte r v a l s between 5 and 3 0 minutes. There was a s i g n i f i c a n t increase i n y i e l d at higher temperatures, though no sample exceeded 14 mg lactulose/100 mL. Increasing heating time also increased y i e l d at higher temperatures. At 90°C, lactulose content rose from approximately 3 mg/100 mL after 5 minutes to 13 mg/100 mL af t e r 25 minutes. In t e s t i n g the influence of milk d i l u t i o n , sodium c i t r a t e concentration, and sodium phosphate concentration, f i v e d i f f e r e n t time/temperature treatment combinations were tested (Andrews & Prasad, 1987). These included 130°C for 2 and 4 minutes, 110°C for 19 3 and 15 minutes, and 90° C for 120 minutes. Although the manipulation of c i t r a t e and phosphate addition and milk d i l u t i o n affected the r e s u l t s , temperature and heating time did appear to influence lactulose production. Heat treatments of 130°C for 2 minutes and 110° C for 3 minutes consistently produced the least lactulose of the combinations tested. The difference among the other three time/temperature combinations were dependent on the other variables. Generally increasing lactulose occurred with 90° C for 120 min, 110° C for 15 min, then 130° for 4 min (Andrews & Prasad, 1987) . These results suggest that the increase from 2 to 4 minutes at 130° C and from 3 to 15 minutes at •110° C had a p o s i t i v e influence on lactulose production. As well, i n l i g h t of the displayed influence of heating time, one could argue that the increase i n lactulose at 110°C for 3 minutes to 130°C for 2 minutes was due to the influence of the higher temperature. 2.5b The influence of pH The LA transformation occurs most commonly i n a l k a l i n e solutions. The pH of a solution has been shown to be an i n f l u e n t i a l factor i n lactulose production. In commercial production of lactulose using lactose solutions, pH i s raised s u b s t a n t i a l l y . A pH of 11 was described by Hicks et a l . (1984) as an e f f e c t i v e pH for lactulose production using boric acid and triethylamine. Several studies have detailed the influence of pH. Although r e s t r i c t e d to a pH range near that of milk, i t was found that an increase i n pH from 6.6 to 7.0 i n casein-lactose solutions, heated at 12 0°C for 2 hours, corresponded to an increase i n lactulose 20 production (Adachi, 1959) . A study by Martinez-Castro and Olano (1980) found si m i l a r r e s u l t s as the pH of milk was raised from 6.0 to 7.5 and heated at 120°C for 15 minutes. At pH 6.6 there was a 1.6% lactulose y i e l d , while at 7.5 there was a 13.7% y i e l d . Lactose solutions heated at 40° C for 96 hours at pHs of 9,10,10.5, and 11 using b o r i c acid and triethylamine as ca t a l y s t s , gave increasing lactulose y i e l d s of 6%, 32%, 74%, and 83% respectively (Hicks & Parrish, 1980). • . 2.5c The influence of c i t r a t e and phosphate Both phosphate and c i t r a t e are naturally present i n milk, each at approximately 0.01 M (Florence et a l , 1985; Jenness & Patton, 1959). The addition of c i t r a t e or phosphate buffers to milk has been associated with an increase i n lactulose y i e l d upon heat treatment. In one study, milk was tested a f t e r various time/temperature treatments (Andrews & Prasad, 1987). The most e f f e c t i v e time/temperature combination (130°C for 4 minutes) produced lactulose at approximately 45 mg/100 mL without any buffer addition. After including 10 mL of 0.4 M of c i t r a t e buffer to 100 mL milk, lactulose reached approximately 200 mg/100 mL. Adding 10 mL of 0.4 M phosphate buffer increased lactulose to 125 mg/100 mL. These authors proposed that both phosphate and c i t r a t e acted as ampholytes and catalyzed the LA reaction. 2.5d The influence of lactose concentration In a 1989 study, lactulose production was measured i n heat treated milk at normal and double lactose concentrations. The re s u l t s suggested that lactulose production was a f i r s t order 21 reaction with respect to i n i t i a l lactose concentration (Andrews, 1989). While the lactulose content rose i n response to the increase i n lactose content, the lactulose y i e l d - the r a t i o of lactulose produced to i n i t i a l lactose - remained the same. Similar r e s u l t s were found using whey UF permeate and a boric acid / triethylamine catalyzed reaction. Lactulose y i e l d s i n whey permeates with lactose concentrations of 8.8 and 17.7 g/100 mL were equal at the end of the reaction (3 hours) . The more concentrated sample, however, had a somewhat slower reaction rate (Hicks et a l . , 1984) . 2.5e The influence of demineralization The presence of normal whey UF permeate components, including s a l t s , did not i n t e r f e r e with lactulose y i e l d , according to Hicks et a l . (1984) . In t h e i r study, whey and a p u r i f i e d lactose solution (whey u l t r a f i l t r a t e which had been deionized and decolourized) were isomerized with a boric acid c a t a l y s t and compared. I t was argued by the authors that using whey UF permeate instead of p u r i f i e d lactose would eliminate an energy-intensive r e f i n i n g step. However, i n a study by Andrews (1989), i t was found that 50 ppm calcium as calcium chloride added to milk, depressed lactulose production in heated milk. Here, Andrews proposed that the added calcium was complexing with the c i t r a t e and associating with the phosphate found naturally i n the milk. This reduced the amount of c i t r a t e and phosphate available to catalyze the LA reaction. 22 2.5f Other influences on lactulose y i e l d Other components of milk or processing conditions have been studied as to t h e i r e f f e c t on lactulose y i e l d . A few of the more important findings - f at content, protein content and milk storage - are summarized below. Geir and Klostermeyer (198 3) found that a f a t content of between 0% and 3.5% had no influence on lactulose y i e l d , when te s t i n g milk samples at various temperatures and heating times. Protein concentration was observed to be inversely related to lactulose production in milk by Andrews (1989). A f t e r studying lactulose y i e l d i n concentrated milks, he suggested that protein increased the formation rate of M a i l l a r d browning, forming condensation products of available lactose and lactulose. The e f f e c t of storage was observed i n UHT and s t e r i l i z e d milks, i n the same study (Andrews, 1989). Storage f o r 6 months i n darkness caused the lactulose content of some milks to r i s e and others to f a l l , with a mean increase of 12%. Since the lactulose contents of some samples declined, i t appeared to the author that lactulose was produced and degraded during storage. Storage at d i f f e r e n t temperatures was studied recently by Jimenez-Perez et a l . (1992) . UHT milk samples were stored for 90 days at d i f f e r e n t temperatures. The study showed a s i g n i f i c a n t increase i n lactulose during storage, es p e c i a l l y between 4 0-50°C. 2.6 FRACTIONAL FACTORIAL EXPERIMENTAL DESIGN In t e s t i n g the influence of a single factor on any given r e s u l t , i n t h i s case lactulose y i e l d , the number of experiments equals the number of levels at which the factor i s tested. The 23 d i f f i c u l t y occurs when more than one factor i s tested, since i t i s well known that one factor can influence another. With two or more factors, the number of experiments increases exponentially with the number of factors, as each factor l e v e l must be tested at each l e v e l of a l l other factors. F r a c t i o n a l f a c t o r i a l designs reduce the number of experiments by s e l e c t i n g only portions of a f u l l f a c t o r i a l design, yet s t i l l providing s i m i l a r information. . Interactions among factors not i d e n t i f i e d i n the design were assumed to not b e , s i g n i f i c a n t . In the 1970s and 1980s, Taguchi devised approximately 20 orthogonal arrays. These arrays are multifactor experimental designs constructed from Graeco-Latin square designs. Orthogonal arrays can be used as long as factor responses show l i n e a r behaviour. Arrays commonly occur i n d i f f e r e n t combinations of from 3 to 40 factors, 2 to 5 le v e l s , with various factor interactions and up to 169 experimental t r i a l s (Taguchi, 1987) . One example of an orthogonal array i s the L 2 73 1 3 (L for l a t i n , 27 t r i a l s , 3 l e v e l s and 13 f a c t o r s ) , i n Table 4. Interactions of two factors are shown by a connecting l i n e . The columns of Table 4 represent factor levels and the rows are the experimental t r i a l s . Often, orthogonal arrays are used to t e s t the significance of many factors at a few levels as a screening for subsequent optimization. An array with more le v e l s and interactions and fewer factors i s useful on i t s own to show factor and interaction trends. However, these designs are often not as useful for detailed optimization as are other more f l e x i b l e computerized designs due i n part to the r e s t r i c t i v e nature of assigned l e v e l s . 24 Table 4 One example of Taguchi's experimental design, orthogonal array L 2 7(3 1 3) . Factor Exp't No. A B C D E F G H I J K L M 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 2 2 2 2 2 2 2 2 2 3 1 1 1 1 3 3 3 3 3 3 3 3 3 4 1 2 2 2 1 1 1 2 2 2 3 3 3 5 1 2 2 2 2 2 2 3 3 3 1 1 1 6 1 2 2 2 3 3 3 1 1 1 2 2 2 7 1 3 3 3 1 1 1 3 3 3 2 2 2 8 1 3 3 3 2 2 2 1 1 1 3 3 3 9 1 3 3 3 3 3 3 2 2 2 1 1 1 10 2 1 2 3 1 2 3 1 2 3 1 2 3 11 2 1 2 3 2 3 1 2 3 1 2 3 1 12 2 1 2 3 3 1 2 3 1 2 3 1 2 13 2 2 3 1 1 2 3 2 3 1 3 1 2 14 2 2 3 1 2 3 1 3 1 2 1 2 3 15 2 2 3 1 3 1 2 1 2 3 2 3 1 16 2 3 1 2 1 2 3 3 1 2 2 3 1 17 2 3 1 2 2 3 1 1 2 3 3 1 2 18 2 3 1 2 3 1 2 2 3 1 1 2 3 19 3 1 3 2 1 3 2 1 3 2 1 3 2 20 3 1 3 2 2 1 3 2 1 3 2 1 3 21 3 1 3 2 3 2 1 3 2 1 3 2 1 22 3 2 1 3 1 3 2 2 1 3 3 2 1 23 3 2 1 3 2 1 3 3 2 1 1 3 2 24 3 2 1 3 3 2 1 1 3 2 2 1 3 25 3 3 2 1 1 3 2 3 2 1 2 1 3 26 3 3 2 1 2 1 3 1 3 2 3 2 1 27 3 3 2 1 3 2 1 2 1 3 1 3 2 A 25 2.7 PURIFICATION OF LACTULOSE 2.7a Overview of lactulose p u r i f i c a t i o n methods The separation of lactulose from lactose has proved to be a d i f f i c u l t task. Lactose and lactulose are s i m i l a r sugars with the same molecular weight. Many methods have been used i n the p u r i f i c a t i o n of lactulose: ion exchange to remove aldoses (Adachi & Patton, 1961), oxidation of lactose to lactobionic acid and i t s p r e c i p i t a t i o n as a calcium s a l t (Visser & van den Bos, 1988) , oxidation of aldoses with bromine, then p r e c i p i t a t i o n using s i l v e r sulphate and hydrogen sulphide (Montgomery, 1962), and the r e v e r s i b l e binding of lactulose to calcium hydroxide (Adachi, 1969), to name a few. 2.7b Cold temperature p r e c i p i t a t i o n of lactose Lactose consists of glucose and galactose moieties connected by a 1 to 4 beta-galactosidic linkage. Lactose i s present as two isomers (alpha and beta) which describe the conformation of the glucose moiety of the disaccharide. The r a t i o of beta to alpha lactose i n an aqueous solution increases as temperature increases, the beta isomer being the only stable isomer above 93.5°C (Short, 1978). During cold temperature p r e c i p i t a t i o n , the alpha isomer w i l l p r e c i p i t a t e f i r s t . The beta form w i l l then convert to the alpha form as p r e c i p i t a t i o n continues. Therefore, lactose c r y s t a l s occur generally i n the alpha monohydrate form. There are two rate determining steps i n t h i s p r e c i p i t a t i o n , the c r y s t a l l i z a t i o n of the alpha isomer and the conversion of the beta to the alpha isomer. Which i s the l i m i t i n g step i n a given p r e c i p i t a t i o n depends on temperature and other conditions (Zadow, 1984). Table 5 d e t a i l s the s o l u b i l i t i e s lactose and lactulose at various temperatures. At temperatures, lactulose i s more soluble i n water than lactose. 27 Table 5 S o l u b i l i t i e s of lactose and lactulose i n water at various temperatures. Temperature Lactose 1 Lactulose 2 (°C) (% w/w) (%w/w) 15 14.43 76.4 30 19.88 81 60 36.87 80 51.12 90 > 86 (Visser, 1982) (Oosten, 1967) 28 3. MATERIALS AND METHODS Figure 4 i l l u s t r a t e s the f i n a l process of lactulose conversion and p a r t i a l p u r i f i c a t i o n . I t provides an overview to the methods described here. 3.1 RAW MATERIALS A combination of cheddar and mozzarella wheys were supplied courtesy of Canadian Inovatech Inc., Abbotsford, BC. The whey mixture was previously deproteinized by u l t r a f i l t r a t i o n at that f a c i l i t y , concentrated to a t o t a l s o l i d s of approximately 5%, then frozen. Sugar solutions used i n t h i s study included alpha-lactose monohydrate, lactulose, D-glucose, D-fructose, D-galactose, and D-tagatose (Sigma Chemical Co., Mississauga, ON). 3.2 THE CONVERSION OF LACTOSE TO LACTULOSE 3.2a Sample preparation The influences of processing and compositional factors on lactulose y i e l d i n deproteinized whey permeate were evaluated using two L 2 7(3 1 3) f r a c t i o n a l f a c t o r i a l experimental designs. The factors of each design, together with t h e i r assigned l e v e l s are shown i n Table 6, the experimental schemes i n Figure 5. The 27 samples for the f i r s t experimental design were prepared as follows. 1. Eight l i t r e s of frozen u l t r a f i l t r a t e whey permeate were thawed and the calcium content reduced, as described i n Section 3.4d. The d e c a l c i f i e d whey permeate was then demineralized by ion-exchange, Section 3.4e, and concentrated using a roto-evaporator at 40°C. 29 WHEY P E R M E A T E Heat NaOH Calcium B D E C A L C I F I E D WHEY Heat Citrate NaOH \ ( MODIFIED WHEY ) Water R E T O R T E D WHEY NaOH. 1 s t PPT D E C A N T Water Water Heat Heat Discard Precipitate PRECIPITATED P R E P A R A T I O N -exchange (lon  DEIONIZED P R E P A R A T I O N ^Charcoa l Q D E C O L O U R I Z E D P R E P A R A T I O N Figure 4 Overall process for the conversion of lactose to lactulose and p a r t i a l p u r i f i c a t i o n of lactulose. 30 Table 6 Condition factors and t h e i r assigned l e v e l s of heat processed deproteinized whey investigated for influence on lactulose y i e l d using an L ? 7(3 1 3) design. J27 Experimental design #1 Level Factor pH 9.0 10, Lactose Concentration 40 79 (mg/mL) Heating Temperature (°C) 90 110 Heating Duration (min) 5 20 C i t r i c Acid (mM) 40 70 Disodium Phosphate (mM) 40 70 12, 115 130 90 100 100 Experimental design #2 Level Factor NaOH Concentration (mM) 18 P u r i f i c a t i o n 1 DPW Heating Temperature (°C) 105 Heating Duration (min) 10 Sodium C i t r a t e (mM) 50 Disodium Phosphate (mM) 0 50 DCW 115 40 70 50 100 LS 125 80 90 100 P u r i f i c a t i o n of the whey: DPW = Deproteinized whey. DCW = Deproteinized whey, calcium removed. LS = Lactose monohydrate solution. 31 L a c t o s e 1 / Pur i f icat ion 2 C , D / \ F , G Phosphate Citrate Time Temp. p H 1 / N a O H 2 M • Error Experimental design #1 Experimental design #2 Figure 5 Schemes used i n the L ? 7(3 1 3) design for the '27 '• conversion of lactose to lactulose during heat treatment of whey permeate. 2. C i t r i c acid and disodium phosphate were added to 15 mL graduated t e s t tubes. Amounts were calculated to provide the molar l e v e l selected i n the experimental design i n a 10 mL f i n a l sample volume. 3. A portion of the concentrated demineralized permeate was d i l u t e d to three 100 mL volumes of 5, 10, and 15% t o t a l s o l i d s . The lactose concentrations were assayed for use i n the c a l c u l a t i o n of lactulose y i e l d . Volumes of 8.0 mL of these d i l u t e d permeates were added to the te s t tubes containing c i t r i c acid and disodium phosphate according to the design. 4. The pH of the samples was raised with 5.0 M sodium hydroxide (NaOH) according to the design and then d i l u t e d with d i s t i l l e d deionized water (ddH20) to a f i n a l sample volume of 10 mL. 5. Heat processing u t i l i z e d an agitating water bath for the 90°C treatments and a steam re t o r t for the 110°C and 130°C treatments. Section 3.2b d e t a i l s the heating process. 6. Each of the 27 samples was diluted i f necessary and assayed for fructose and lactulose in duplicate using the enzymatic assays described i n Section 3.5bl. The samples of the second experimental design were prepared somewhat d i f f e r e n t l y . 1. Two l i t r e s of whey permeate were thawed and the calcium reduced i n one of the l i t r e s . Both l i t r e volumes, the whey permeate and d e c a l c i f i e d permeate, were concentrated by vacuum roto-evaporation and then diluted to both equal 89 mg/mL lactose. An aqueous solution of lactose monohydrate was prepared of equal lactose content. Volumes of 8.2 mL of each solution were added to nine 15 mL graduated test tubes. 2. Stock aqueous solutions of 0.91 M disodium phosphate and 2.3 M sodium c i t r a t e were prepared. Amounts of 0, 0.55, and 1.1 mL of the disodium phosphate solution (0, 50, and 100 mM i n 10 mL whey, res p e c t i v e l y ) , and 0.22, 0.31, and 0.40 mL of the sodium c i t r a t e s o l u t i o n (50, 70, and 90 mM in 10 mL whey respectively) were added to the appropriate te s t tubes. 3. Selected volumes of 5.0 M NaOH were added to the te s t tubes according to the design. To each test tube was then added 1.8 mL ddH20 less the combined amount of disodium phosphate, sodium c i t r a t e and NaOH. This brought each sample to a f i n a l volume of 10 mL and a lactose concentration of 73 mg/mL. 4. Retorting the samples i s described i n Section 3.2b. 5. Each of the 27 samples was diluted i f necessary and assayed i n duplicate for lactulose. 3.2b Heat processing Samples (4.5 mL) were placed i n 10 mL serum bottles ("400" brand, 25 X 54 mm diameter X height, 13 mm i . d . mouth, Wheaton Inc., M i l l v i l l e NJ) with aluminum seals crimped over rubber flange-s t y l e stoppers. Heating time using a waterbath (Magniwhirl constant temperature bath, Blue M E l e c t r i c Co., Blue Island IL) was defined as the time between the samples entering the waterbath and ex i t i n g to an ice bath for cooling. For the higher temperature treatments using a r e t o r t (Model 500W, FMC Corp., Central Engineering Laboratory, Santa Clara CA), heating time was defined as the time between the shutting of the vent and the opening of the cold water i n l e t . Come up and cool down times of the heating vessels and samples are shown i n Table 7. Three samples were prepared with a copper and constantan 34 Table 7 Heat processing of the two L 2 7(3 1 3) experimental designs for the conversion of lactose to lactulose. Temperature 1 Pressure 2 Come Up Time Cool Down Time Vessel 3 Sample of Sample to 80°C (°C) (lbs/inch 2) (sec) (sec) 90 0.0 128 8 105 2.9 50 85 20 110 6.1 65 105 35 115 9.8 70 135 40 125 18.9 85 160 55 130 24.6 90 185 75 90° C samples were heated by water bath, higher temperature samples by re t o r t . Gauge pressure was based on a sea l e v e l a l t i t u d e . Retort come up time measured from the clos i n g of the re t o r t vent to the re t o r t i n t e r i o r reaching the set temperature. 35 Teflon coated wire thermocouple, (Omega Tech. Co. Stamford CT) , inside serum bottles f i l l e d with 4.5 mL of ddH20. The thermocouple's Teflon coating was stripped, and the wire t i p s soldered together. The wire t i p was bent inward to keep the soldered t i p i n the centre of the l i q u i d , and the separated wires were folded over the mouth and flattened against the outside of the f l a s k . The rubber stopper was placed over the wires and the cover crimped overtop. Two exposed 7 cm needle thermocouples (O.F. Ecklund Inc., Fort Myers, FL) were placed i n the waterbath or r e t o r t to monitor the temperature of the heating medium. The thermocouples were connected to a F i e l d Logger (DT 100F Data Taker, Data Elect r o n i c s [AUST] Ltd., Rowville Australia) and personal computer which tracked temperatures at set 5 or 10 second in t e r v a l s using the "Decipher" program (Data E l e c t r o n i c s ) . 3.2c Calculation of lactulose y i e l d Experimental samples were assayed for sugars as described i n Section 3.5b. Percent lactulose y i e l d was then calculated from lactulose concentration: % lactulose y i e l d = (lactulose concentration mg/mL)(100%) ( i n i t i a l lactose concentration mg/mL) (Equation 1) This assay measured lactulose i n d i r e c t l y as fructose. Therefore, free fructose i n solution was also included i n the lactulose measurement. In experiment design #1, fructose was assayed separately to be subtracted from the measured lactulose amount. The following c a l c u l a t i o n would then provide a more 36 accurate lactulose measurement. lactulose mg/mL = (lactulose mg/mL) (fructose mcf/mL) X (342.32 q/mol) (180.16 g/mol) where 342.32 g/mol 180.16 g/mol molecular mass of lactulose molecular mass of fructose (Equation 2) 3.2d S t a t i s t i c s f o r f r a c t i o n a l f a c t o r i a l designs For each f r a c t i o n a l f a c t o r i a l experimental design, the mean was calculated from duplicate assays for each of the 27 experiments. These data were then analyzed by analysis of variance (ANOVA) to determine i f the factors and/or factor interactions tested had s i g n i f i c a n t influence on lactulose y i e l d . A s t a t i s t i c a l computer program "PTaguchi.BAS" designed by Nakai et a l . (1994), for the f r a c t i o n a l f a c t o r i a l designs of Taguchi, was used to simpl i f y c a l c u l a t i o n s . Graphing factors and factor interactions used the mean of the 9 experiment samples for each factor l e v e l . Confidence l i m i t s followed the equation: X ± (t. 0.05(2),DFe / n 1 / 2) X (sse/DFe) 1 / 2 where t, sse = DFe = n = 0.05(2),DFe two-tailed t value at the degrees of freedom of the error at P=0.05. number of samples. sums of squares of the error. degrees of freedom of the error. (Equation 3) 3.3 CONTINUOUS FLOW CONVERSION OF LACTOSE TO LACTULOSE 3.3a Design of heat exchanger A continuous flow heat exchanger was developed to produce the 37 heat treated whey permeate i n larger amounts than was achieved using samples of 4.5 mL i n the batch process. A heat exchanger was constructed using 1/4 inch outside diameter copper piping. The 13.4 m pipe was wound into a 13 cm diameter c o i l and placed i n an o i l bath and ice bath i n sequence. A p e r i s t a l t i c pump (Minipuls 3, Mandel, Guelph, ON) was used to c i r c u l a t e the whey through the pipe i n a closed-loop system. The copper pipe had a volume of 167 mL of water at 110°C. 3 . 3 b Determining heating times A shorter copper c o i l heat exchanger was prepared to estimate the come up and cool down times of the whey permeate (Figure 6). Needle thermocouples (described i n Section 3.2b) were inserted along the copper piping i n the o i l bath at up to 5 points: 54, 110, 167, 224, and 280 cm (an i n i t i a l 14 cm of piping remained outside the bath). The thermocouples were held i n the piping by 5 cm long brass compression tee j o i n t s (ULN 365, Master Plumber, Vancouver, BC) (Figure 6). Three thermocouples were rotated among the f i v e tee j o i n t s ; the tee jo i n t s were t i g h t l y capped i f empty. One other tee j o i n t connected the pipe to the i n l e t tubing. Using t h i s construction, the thermocouples tracked the heating of the whey permeate i n the o i l bath or, when the piping was reversed, tracked the cooling of the permeate i n the ice bath. 3.4 PURIFICATION 3.4a Materials Two solutions were used i n the p u r i f i c a t i o n studies: 1. lactose monohydrate:lactulose aqueous solutions of various 38 The heat exchanger with thermocouples determined come up and cool down times: top) the pressure t i g h t system, bottom) the heating or cooling c o i l with thermocouples i n sequence. 39 r a t i o s , and 2. d e c a l c i f i e d UF whey permeate, heat treated using the conditions d e t a i l e d i n Table 8. 3 . 4 b F r a c t i o n a l f a c t o r i a l design f o r cold p r e c i p i t a t i o n 3 . 4 b l Sample preparation An L 2 7(3 1 3) f r a c t i o n a l f a c t o r i a l design was used to evaluate four factors on the pr e f e r e n t i a l p r e c i p i t a t i o n of lactose over lactulose i n cooled, concentrated sugar solutions (Table 9, the experimental scheme i n Figure 7) . The 27 samples were prepared as follows. 1. A 200 mL aqueous solution containing 45.48 g lactose monohydrate and 16.80 g lactulose (0.39 lactulose/lactose) was prepared to simulate heat treated whey permeate. The soluti o n was concentrated, removing 54 mL water through a roto-evaporator. Of the remaining solution, 28.5 mL were removed for the low sugar concentration samples. The remaining solution was concentrated to a mid sugar concentration solution by removing 3 0 mL water. Again, 28.5 mL were removed for samples. The soluti o n was f i n a l l y concentrated, removing 11.5 mL of water for the high sugar concentration solution. Each solution was then assayed for lactose and lactulose content. 2. Each of the three 28.5 mL solutions of d i f f e r i n g sugar concentrations was separated into three 25 mL screw cap t e s t tubes, 9.5 mL volumes each. To these was added enough 5.0 M NaOH to reach the pH assigned i n the experimental design. The samples were d i l u t e d with ddH20 so that a l l samples t o t a l l e d 9.5 mL plus a combined 0.2 0 mL of NaOH and ddH20. At t h i s stage there were nine 40 Table 8 Heat processing c o n d i t i o n s f o r l a c t u l o s e production from whey s e l e c t e d u s i n g Taguchi's f r a c t i o n a l f a c t o r i a l designs. Conditions L e v e l pH Phosphate concentration Sodium c i t r a t e concentration Lactose conc e n t r a t i o n P u r i f i c a t i o n Heating temperature Heating time 10.5 - 11.0 0 mM 50 mM >70 mg/mL d e c a l c i f i e d UF whey permeate 110 °C 10 min 41 Table 9 Condition factors and t h e i r assigned l e v e l s of a cold precipitated lactulose: lactose solution using a L ? 7(3 1 3) design. 27' Factor Level Sugar concentration (%) Temperature decrease (C°/hr) F i n a l temperature (°C) 7.0 9.0 10.7 29 39 52 3 5 7 7 12 20 Sugar concentration measured afte r separation. 29 ± 2 39 ± 2 52 ± 3 42 Figure 7 Schemes used i n the Taguchi L 2 7(3 1 3) design for the p r e f e r e n t i a l cold p r e c i p i t a t i o n of lactose, following the design of Table 4. 43 samples, each with a unique pH and sugar concentration combination. 3. Each of the nine solutions was further divided into three 3 mL volumes. Each sample was placed i n an ag i t a t i n g water bath set at 60° C and 100 rpm. Each sample was then cooled following the factor l e v e l s of the experimental design. At 35°C, a l l samples were seeded with 2 mg lactose monohydrate. Samples were further cooled to t h e i r assigned f i n a l temperature by dropping waterbath temperature every 15 minutes, at which time they were moved to a re f r i g e r a t e d incubator and held for 24 hours. A l l the samples were cooled one at a time using three similar a g i t a t i n g waterbaths and 3 r e f r i g e r a t e d incubators set at 7, 12, and 20°C (Forma-Scientific Inc., Marietta, OH). 4. The decant layer was removed by pipette and d i l u t e d to 25 mL i n a volumetric fl a s k . The pr e c i p i t a t e was r e s o l u b i l i z e d with warm ddH20 and agitation, and then di l u t e d to 2 5 mL. 3. 4b2 Calcu l a t i o n of lactulose y i e l d and sugar r a t i o Measured lactulose and lactose concentrations (mg/mL) for the decant and pre c i p i t a t e portions of each sample were adjusted based on the standard curves of Section 3.5b4 and then adjusted for volume: lactulose/lactose (mg/mL) = [carbohydrate (mg/mL)1(25 mL) (3.0 mL) where: carbohydrate = sugar adjusted for standard (mg/mL) curve. 25 mL = volume after d i l u t i o n 3.0 mL = i n i t i a l sample volume (Equation 4) Percent lactulose y i e l d and sugar r a t i o were calculated using the following formulae: 44 %lactulose y i e l d = _ (decant lactulose mg/mL) X 100% {(decant lactulose mg/mL) + (precipitate lactulose mg/mL)} (Equation 5) %lactose y i e l d = _ (decant lactose mg/mL) X 100% ._ {(decant lactose mg/mL) + (precipitate lactose mg/mL)} (Equation 6) sugar r a t i o = decant lactulose (mg/mL) decant lactose (mg/mL) (Equation 7) 3.4c The second cycle of cold p r e c i p i t a t i o n Decant and pre c i p i t a t e portions from the f i r s t cycle were c o l l e c t e d and adjusted to the r a t i o s 50:50 and 14:86 lactulose:lactose respectively to simulate the sugar r a t i o of decanted and precipitated portions a f t e r a p r e c i p i t a t i o n cycle. A f t e r the pH was raised to 10.5 with 5.0 M NaOH, each portion was concentrated using a roto-evaporator to a point at which flakes began to form at a temperature of 60° C. This occurred at approximately 500 mg/mL i n the decant portion and 3 50 mg/mL i n the pr e c i p i t a t e portion. Volumes of 2.0 mL of the adjusted decant were added to each of f i v e screw cap 5 mL graduated t e s t tubes. To f i v e more t e s t tubes were added 2.0 mL of the adjusted p r e c i p i t a t e . A l l samples were cooled from 60° C at a rate of 5C° per hour. There was no lactose seeding of these samples. A l l samples were held at 20° C for 48 hours, at which time each was decanted, and the decant and pr e c i p i t a t e portions r e s o l u b i l i z e d i n separate 10 mL volumetric cyl i n d e r s . Samples were assayed for lactose and lactulose i n the same manner as af t e r the f i r s t p r e c i p i t a t i o n cycle. 45 3.4d Calcium phosphate removal After receiving the frozen, concentrated whey permeate from the supplier, i t was thawed and heat treated to p r e c i p i t a t e calcium s a l t s . The pH of the whey permeate was raised to 6.8 with 5.0 M NaOH and heated i n a half jacket steam k e t t l e . The permeate was held at a temperature of 85°C for 30 minutes. After i t had cooled and the calcium phosphate settled, the permeate was decanted and the p r e c i p i t a t e discarded (Durance Se Cross, 1992) . 3.4e Demineralization Demineralizing by ion-exchange was used on two occasions i n t h i s study. Amberlite IR12 0, a strong acid cation exchange resin, and Duolite A368, a strong base anion exchange r e s i n , were used i n sequence for demineralization (Rohm and Haas Inc., Philadelphia, PA) (Durance Se Cross, 1992) . 1. For most of t h i s study, whey permeate was f i r s t d e c a l c i f i e d , then heat-processed and cold- p r e c i p i t a t e d to synthesize and purify lactulose. As a f i n a l treatment, the lactulose mixture was demineralized by ion-exchange to remove natural and added s a l t s as well as colour. The lactulose mixture was pumped downward through 4 mL of Amberlite IR120 then upwards through 3.8 mL of Duolite A368 at a flowrate of 0.3 mL/min using a p e r i s t a l t i c pump (Econo-column pump, Bio-Rad Inc., Mississauga, ON). Flex-columns (Mandel) held the re s i n , each having a 20 micron porous polyethylene bed support and a 1.0 cm inside diameter. Tygon tubing (R-3 606, Cole-Parmer Instruments Co., Vernon H i l l s , IL) size 14, inside diameter of 1.6 mm was used throughout. Various volumes of samples at d i f f e r i n g 46 s o l i d s were demineralized on separate occasions with regeneration a f t e r each use. The pH of the mixture a f t e r flowing through each column was checked regularly for r e s i n exhaustion. Regeneration for each res i n column was required a f t e r use and before the f i r s t use. The cation column was regenerated upward with 6 mL of 1.0 M HC1 and 3 0 mL ddH20 at a flowrate of 0.9 mL/min. The anion column was regenerated downward with 6 mL of 1.0 M NaOH and 3 0 mL ddH20 at the same flowrate. For the newly purchased anion exchange r e s i n Duolite A368, an i n i t i a l exhaustion/regeneration treatment consisted of three cycles of 1.1 M HC1 and 1.0 M NaOH, then complete r i n s i n g with ddH20. A previous study, however, suggested further treatment to minimize the odour and taste of the re s i n which can li n g e r i n the demineralized permeate. The re s i n was soaked i n a 50% aqueous soluti o n of methanol for one hour, then one hour i n pure methanol, then a 50% solution again for an hour. After thorough r i n s i n g i n ddH20, the r e s i n was heated and held at 95°C u n t i l no r e s i n odour was detected, approximately 35 hours. 2. The whey permeate i n the f i r s t experimental f r a c t i o n a l f a c t o r i a l design was i n i t i a l l y demineralized by d e c a l c i f i c a t i o n and ion-exchange to remove natural s a l t s before any other treatment. This p r a c t i c e was not continued. The process was similar to the post-treatment demineralization but on a larger scale. Eight l i t r e s of thawed deproteinized whey at approximately 5% t o t a l s o l i d s were demineralized, using larger columns (jacketed Moduline medium-pressure 44 cm diameter columns, Amicon D i v i s i o n W.R. Grace & Co., Danver, MA) packed with 400 mL of Amberlite IR120 and 340 mL Duolite A368, a stronger pump 47 (Masterflex 7523-10, Cole-Parmer) and a flow rate of 35 mL/min. Again the pH of the permeate afte r using each column was checked r e g u l a r l y for r e s i n exhaustion. One cation r e s i n bed volume could e f f e c t i v e l y decationize 16 bed volumes of 5% d e c a l c i f i e d whey permeate before exhaustion, and one bed volume of anion r e s i n could deanionize 15 bed volumes of 5% whey permeate. The cation bed was regenerated upward with 750 mL of a 1.0 M HC1 soluti o n at a flowrate of 2 5 mL/min, and then rinsed with 4 L of ddH20 at a flowrate of 50 mL/min. The anion exchange bed was regenerated downward with 600 mL of 1.0 M NaOH at a flowrate of 25 mL/min. The bed was then rinsed with 3.25 L of ddH20 at a flowrate of 50 mL/min. 3 . 4 f Decolourization While the Duolite A368 p a r t i a l l y decolourized the heat treated p u r i f i e d whey permeate, activated charcoal was used to further remove the golden to amber colour. The same Flex-column, tubing, and Bio-rad pump as was used for the post-treatment demineralization was employed with the charcoal. A 9.5 mL bed volume of activated charcoal was used for the small amounts of high s o l i d s whey permeate with a downward flowrate of approximately 0.2 mL/min. Large volumes of ddH20 rinsed the sample through the charcoal. Fresh activated charcoal was used for each decolourization sample. 3.5 PROXIMATE ANALYSIS Various assays were performed on the whey permeate at d i f f e r e n t stages of processing to tes t for major compositional 48 changes taking place. These tests included measurements of t o t a l s o l i d s , ash, nitrogen, protein, pH, t i t r a t a b l e acid, lactose, lactulose, glucose, fructose, and galactose. The whey permeate was continually d i l u t e d and concentrated during processing so most te s t r e s u l t s were based on the t o t a l s o l i d s of each sample. 3.5a Sampling throughout the process The proximate analysis samples were a l l taken from one processing batch, tracking d e c a l c i f i e d whey permeate through r e t o r t i n g and p r e c i p i t a t i o n to a demineralized and decolourized mixed carbohydrate solution. In t h i s way, the same sample was tested at each processing stage and volumes were co n t r o l l e d and recorded. 3.5b Enzymatic Carbohydrate assays 3.5bl Quantitative assays f o r lactulose, lactose, glucose, and fructose. The enzymatic assays used for the q u a n t i f i c a t i o n of lactulose, lactose, glucose, and fructose i n a mixed carbohydrate solution followed c l o s e l y the lactulose assay method of Andrews (1984), but with the following modifications. 1. The beta-galactosidase from Aspergillus oryzae was used i n the assay. Consequently, the pH of Buffer I was reduced from 7.5 to 5.0, the optimum pH for "Fungal Lactase 100,000" for hydrolysis as outlined by the supplier. 2. Due to the higher concentration of carbohydrate i n samples i n the present study, the amount of sample added to the cuvette for measurement was di l u t e d ten-fold. 49 3. The method was adapted to include the measurement of lactose, fructose and glucose. 4. Fructose and lactulose were assayed i n conjunction with glucose and lactose respectively. The carbohydrates were assayed using the following buffers and solutions: Buffer I -0.33 mL Buffer II -1. 0 mL Buffer III 0.4 0 mL Enzyme 0.10 mL 0.10 mL Oxidation Solution 0.05 mL 0.05 mL ATP -0.10 mL NADP -0.10 mL HK/G6PD -0.02 mL PI -0.004 mL A 0.13 M c i t r i c acid - 0.31 M Na2HP04 (Mcllvaine) buffer of pH 5.0, containing 0.0041 M MgS04.7H20 and 0.046 M sodium azide as preservative. A 0.75 M triethanolamine hydrochloride (Sigma) and 0.010 M magnesium sulphate buffer adjusted to pH 7.6 with 5.0 M NaOH. Buffer II d i l u t e d f i v e - f o l d Lactose and lactulose assays: a suspension of "Fungal Lactase 100,000" beta-galactosidase from Aspergillus oryzae (Solvay Enzymes Inc., Elkhart, IN) of a c t i v i t y 230 units/mg (Section 3.5b2). Glucose and fructose assays: ddH20 Lactulose and fructose assays: an aqueous solution of 0.003 g glucose oxidase from A s p e r g i l l u s niger grade II and 0.15 mL catalase (Boehringer Mannheim International Ltd., Laval, PQ) Lactose and glucose assays Lactose/ lactulose assay Glucose/ fructose assay: ddH20 An aqueous solution of 0.082 6 M disodium Adenosine 5'-tetrahydrogen triphosphate (Boehringer Mannheim) and 0.47 M sodium carbonate An aqueous solution of 0.013 M Beta-nicotinamide adenine dinucleotide phosphate (Sigma) A hexokinase and glucose-6-phosphate-dehydrogenase solution (Boehringer Mannheim) A phosphoglucose isomerase solution (Boehringer Mannheim) 50 1. To 0.50 mL of mixed carbohydrate sample was added Buffer I, Enzyme, and ddH20 to 5 mL t o t a l volume and held 15 - 17 hours at 4 5°C i n a gently agitating waterbath. Lactose and lactulose were hydrolyzed to glucose/ galactose and fructose/ galactose respectively. 2. To the mixture was added Oxidation Solution, Buffer I I I , and 0.2 0 mL of 0.3 3 M NaOH. Two drops of s i l i c o n e antifoam reduced foaming during two hours of continued strong aeration at 45° C. Aeration was accomplished by pumping a i r , driven by an a i r compressor v i a t h i n tubing, to the bottom of the l i q u i d - f i l l e d t e s t tubes. Then, 0.10 mL of the NaOH was added and the solution was further incubated at 45° C for 15 minutes. Immersing for 20 minutes i n a b o i l i n g water bath denatured a l l added enzymes. 3. The solution was dil u t e d to 10 mL and f i l t e r e d (#42, Whatman Ltd. , Maidstone, UK) . Into a cuvette (UV grade methacrylate, Fisher S c i e n t i f i c , Ottawa, ON) were added 0.10 mL f i l t r a t e , 1.9 mL ddH20, Buffer II, ATP, and NADP. The sample absorbance was measured (Al, i n Equation 8) at 340 nm on a UV v i s i b l e spectrophotometer (Shimadzu S c i e n t i f i c Instruments, Columbia, MD). To t h i s was added HK/G6PD, and incubated at room temperature for 15 minutes. Fructose and glucose were phosphorylated to fructose-6-phosphate and glucose-6-phosphate respectively. The oxidation of the l a t t e r was coupled with a reduction of NADP which correlated to an increase i n ext i n c t i o n at the wavelength 340 nm (A2, i n Equation 8). PI was added with a further 2 0-minute incubation. The fructose-6-phosphate isomerized to glucose-6-phosphate, and the l a t t e r oxidized with residual G6PD, again causing a reduction i n NADP (A3, i n Equation 8). 51 The difference i n extinctions at 340 nm was proportional to the amount of carbohydrate present i n the mixture using the following equation. Carbohydrate mg/mL = (AE)(V)(MM)(D) 1000(1)(v)(e) where: AE = absorbance difference (A2 - Al) lactose and glucose assay (A3 - A2) lactulose and fructose assay V = t o t a l volume i n cuvette, 3.22 mL MM = molecular mass of lactulose and lactose, 342.31 g/mole. molecular mass of glucose and fructose, 180.16 g/mole. D = d i l u t i o n of sample, 20 1 = o p t i c a l l i g h t path of cuvette, 1 cm v = volume of sample in cuvette = 0.1 mL e = molar extinction c o e f f i c i e n t of reduced NADP at 340 nm, 6.31/(mmol)(cm) (Equation 8) In t h i s assay lactulose was hydrolyzed to fructose and galactose. Lactulose content was calculated s t o i c h i o m e t r i c a l l y from the measurement of fructose. Free fructose i n the sample was assayed separately and then subtracted (as lactulose) from the measured lactulose. A similar c a l c u l a t i o n was done fo r the lactose assay. Lactose content was calculated s t o i c h i o m e t r i c a l l y from glucose, and then free glucose was subtracted. 3.5b2 A c t i v i t y of Beta-galactosidase The a c t i v i t y of the beta-galactosidase from As p e r g i l l u s oryzae was estimated using an a r t i f i c i a l substrate, o-nitrophenyl-beta-D-galactopyranoside (ONPG), instead of lactose (Dobrogosz, 1981). Beta-galactosidase catalyses the hydrolysis of ONPG i n the following reaction: ONPG + H20 —> galactose + o-nitrophenol 52 While ONPG i s colourless, o-nitrophenol i s yellow i n alk a l i n e s o l u t i o n and can be measured spectrophotometrically. A volume of 4.1 mL of 0.050 M sodium phosphate buffer, pH 7.5, was mixed with 0.2 0 mL of 0.03 2 M reduced glutathione (Sigma Chemical Co.), and the mixture brought to 30°C. Enzyme was added i n a 0.20 mL volume, and 0.50 mL of pre-incubated 0.010 M ONPG was then added. After exactly 15 minutes the reaction was stopped by adding 1.0 mL of Na2C03 and the absorbance at 42 0 nm was measured. A standard curve was prepared r e l a t i n g absorbance at 42 0 nm to increasing o-nitrophenol concentrations, Figure 8. Conditions were i d e n t i c a l to the samples' assay except for the sub s t i t u t i o n of ddH20 for ONPG (0.05 mL ddH20 less the amount of added o-nitro-phenol) Linear regression was calculated and tested using ANOVA, i n the manner detailed i n Section 35b4. After the absorbances of tes t samples had been adjusted against the standard curve, the following equation estimated enzyme a c t i v i t y . 1 unit of enzyme a c t i v i t y 1 = The amount of enzyme which produced 1 micromole of o-nitrophenol/hour 1 Enzyme a c t i v i t y i s defined under the conditions s p e c i f i c to t h i s assay. (Equation 9) The beta-galactosidase used i n t h i s study was tested on two occasions, at the beginning and end of two years of use. For each t e s t the enzyme was assayed i n t r i p l i c a t e . The two enzyme sample res u l t s were then tested using a two-t a i l e d variance r a t i o t e s t to determine whether the samples came from the same sample population using the following c a l c u l a t i o n : 53 Mean from 3 r e p l i c a t e samples. Error bars = ± 1 standard deviation. Regression curve: y = (7.6)x + 0.047 r 2 = 0.992 n = 18 Figure 8 The standard curve of the absorbance of o-nitrophenol at 420 nm to determine the a c t i v i t y of the beta-galactosidase "Lactase 100,000" using ONPG hydrolysis. 54 F = s2, / s 2 2 where s 2 = sample variance (Equation 10) 3.5b3 Quantitative assay for galactose Galactose was measured i n mixed carbohydrate solutions using the Lactose/D-Galactose UV-method tes t k i t (Boehringer Mannheim, cat no. 176303). D-galactose was oxidized by nicotinamide adenine dinucleotide (NAD) to galactonic acid i n the presence of beta-galactose dehydrogenase. The spectrophotometer was used at wavelength 340 nm, measuring the sample before (Al) and a f t e r (A2, i n Equation 11) the addition of the beta-galactose dehydrogenase. The amount of NADH formed was stoichiometric with the amount of galactose present, using the formula: galactose mg/mL = (AE)(V)(MM)(D) 1000(1)(v)(e) where: AE = absorbance difference, (A2-A1) V = t o t a l volume i n cuvette, 3.30 mL MM = Molecular mass of galactose, 180.16 g/mole = d i l u t i o n of sample, 10 = o p t i c a l l i g h t path of cuvette, 1 cm = volume of sample in cuvette, 0.1 mL = molar extinction c o e f f i c i e n t of reduced NAD at 340 nm, 6.31 L/(mmole)(cm) (Equation 11) 3.5b4 Standard curves for carbohydrate assays Linear regression standard curves were prepared for a l l carbohydrates using the described enzymatic assays. Aqueous solutions of carbohydrate mixtures containing varying l e v e l s of the f i v e sugars were prepared to predict sugar concentrations i n the lactulose preparation at d i f f e r e n t stages of the study. Table 10 D 1 v e 55 Table 10 Standard solutions containing varying r a t i o s of f i v e carbohydrates prepared for standard curves of a l l enzymatic assays. Carbohydrate Standard Aqueous Solutions B C D E g/ 2 5 mL (% of sugar 1) lactose monohydrate2 l a c t u l o s e 2 galactose 2 glucose fructose 1.8 (50) 1.8 (50) 3.4 (93) 0.11 (3.0) 0.018 (0.50) 0.052 (1.5) 0.070 (2.0) 1.0 (27) 2.4 (68) 0.11 (3.0) 0. 035 (1.0) 0.035 (1.0) 2.9 (80) 0. 63 (18) 0. 035 (1.0) 0. 018 (0.50) 0. 018 (0.50) 2.4 (66) 1.0 (29) 0.070 (2.0) 0. 070 (2.0) 0. 052 (1.5) Lactose monohydrate calculated as lactose Assayed carbohydrates at a 10/1 d i l u t i o n 56 l i s t s the f i v e mixed carbohydrate solutions used i n each standard curve. A l l f i v e solutions along with ddH20 were assayed i n t r i p l i c a t e and sampled undiluted for glucose and fructose, and at a 10 f o l d d i l u t i o n for lactose, lactulose, and galactose. The standard curve of glucose i s shown i n Figure 9. The means were calculated from r e p l i c a t e samples and error was defined as ± 1 standard deviation. The c o e f f i c i e n t of determination, r 2 , measured the strength of the s t r a i g h t - l i n e r e l a t i o n s h i p by providing the proportion of the t o t a l v a r i a t i o n i n Y accounted for by the f i t t e d regression. r 2 = regression sums of squares t o t a l sums of squares (Equation 12) ANOVA was used to t e s t the significance of the l i n e a r i t y and slope of the regressional curve for each carbohydrate assay, Table 11 for glucose. The n u l l hypotheses were defined as the curves being l i n e a r and the slopes being equal to zero. The standard curves for galactose and lactose, along with summaries of ANOVA re s u l t s are i l l u s t r a t e d in Figures 10 and 11. The standard curve of lactulose i s shown i n Figure 12, the ANOVA r e s u l t s i n Table 12. Regression curves for lactulose assayed alone and i n conjunction with lactose were tested for the s i g n i f i c a n c e of l i n e a r i t y and slope. Then the two lactulose regression curves were compared for s i g n i f i c a n t differences in slope and elevation using the following equations. The two fructose regression curves were tested and compared i n the same manner, Figure 13 and Table 13. 57 0.00 0.50 1.00 1.50 2.00 2.50 3.00 Added glucose (mg/mL) V Exper imen ta l D a t a 1 Reg ress i on C u r v e 2 Mean from 3 r e p l i c a t e samples. Error bars = ± 1 standard deviation. Regression curve: y = (1.1)x - 0.0070 r 2 = 0.998 Figure 9 The standard curve of glucose using an enzymatic spectrophotometric assay. 58 Table 11 Analysis of variance te s t i n g the s i g n i f i c a n c e of slope and l i n e a r i t y of the regression curve calculated for the glucose standard curve. Source of Variation Degrees of Freedom Mean Square F-value 3 (x 10"3) Total Linear Regression 1 Residual Deviations from L i n e a r i t y 2 Within Groups 14 1 13 3 10 1180 16500 2 . 1. 16 33 2.41 7600 0.55n.s, 1 Testing the hypothesis that the slope of the regression curve i s zero. 2 Testing the hypothesis that the regression curve i s l i n e a r . 3 F = Q 07 r 0 .01 (1 )1 ,13 F = 3 71 r 0 .05 (1 )3 ,10 S i g n i f i c a n t at p 0.01. n.s. not s i g n i f i c a n t at p > 0.05. 0.50 —i—i—i—i—i—i—i—i—i—i—i i i i i i i i i i i i i i i * i 0-00 0.10 0.20 0.30 0.40 0.50 Added galactose (mg/mL) V Experimental Data 1 Regression Curve 2 Mean from 3 r e p l i c a t e samples. Error bars = ± 1 standard deviation. Regression curve: y = (0.96)x + 0.00090 r 2 = 0.997 n = 15 The regression curve was l i n e a r (p>0.05). The slope of the regression curve was not zero (p^O.Ol). Figure 10 The standard curve of galactose using an enzymatic spectrophotometric assay. 60 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Added lactose (mg/mL) v Experimental Data1 Regression Curve2 Lactose calculated from lactose monohydrate added. Measured lactose less added glucose. Mean from 3 r e p l i c a t e samples. Error bars = ± 1 standard deviation. Regression curve: y = (1.0)x - 0.059 r 2 = 0.998 n = 18 The regression curve was l i n e a r (p>0.05). The slope of the regression curve was not zero (p<0.01). Figure 11 The standard curve of lactose using an enzymatic spectrophotometric assay. 61 Measured lactulose less added fructose. Mean from 3 r e p l i c a t e samples. Error bars = ± 1 standard deviation. Regression curve: y = (0.92)x + 0.034 r 2 = 0.997 Figure 12 The standard curve of lactulose using an enzymatic spectrophotometric assay. t 62 Table 12 Analysis of varianc-e te s t i n g the s i g n i f i c a n c e of slope and l i n e a r i t y of the regression curve calculated for the lactulose standard curves with a t - t e s t comparison of the slope and elevation of the curves. Source of Var i a t i o n Degrees of Mean Square F-value Freedom (x 10"3) A Lactulose assayed alone: Total 17 10700 Linear Regression 1 1 182000 22000** Residual 16 8.27 Deviations from 4 4 . 50 0.47n.s. L i n e a r i t y 2 Within Groups 12 9.52 B Lactulose assayed i n conjunction with lactose: Total 17 10400 Linear Regression 1 1 175000 2900** Residual 16 60. 6 Deviations from 4 69 . 8 1.2n.s. L i n e a r i t y 2 Within Groups 12 57 . 6 Slope comparison 3: t = 0.96 n.s. Elevation comparison 4: t = 0.69 n.s. C Lactulose assays combined5: 10200 357000 11000** 33 . 8 54.5 1.8n.s. 31.1 Total 35 Linear Regression 1 1 Residual 3 4 Deviations from 4 L i n e a r i t y 2 Within Groups 3 0 1 Testing the hypothesis that the slope of the regression curve i s zero. 2 Testing the hypothesis that the regression curve i s l i n e a r . 3 Testing the hypothesis that the slopes of both regression curves are equal. 4 Testing the hypothesis that the elevations of both regression curves are equal. 5 Combined r e s u l t s of assays A and B. S i g n i f i c a n t at p ^ 0.01. n.s. not s i g n i f i c a n t at p > 0.05. 63 Mean from 3 r e p l i c a t e samples. Error bars = ± 1 standard deviation. Regression curve: y = (1.0)x + 0.025 r 2 = 0.997 Figure 13 The standard curve of fructose using an enzymatic spectrophotometric assay. 64 Table 13 Analysis of variance te s t i n g the slope and l i n e a r i t y of the regression curve calculated for the fructose standard curves and a t - t e s t comparison of the slope and elevation of the curves. Source of Variation Degrees of Mean Square F-value Freedom (xlO"3) A Fructose assayed alone: Total 14 1110 ** Linear Regression 1 1 15600 9300 Residual 13 1. 68 Deviations from 3 0.70 0.35n.s. L i n e a r i t y 2 Within Groups 10 1. 98 B Fructose assayed i n conjunction with glucose: Total 14 1080 Linear Regression 1 1 15100 3000** Residual 13 5. 11 Deviations from 3 7 . 63 1.8n.s. L i n e a r i t y 2 Within Groups 10 4.35 Slope comparison 3: t = 0.82 n.s. Elevation comparison 4: t = 0.94 n.s. C Fructose assays combined : Total 29 1060 Linear Regression 1 1 30600 9400** Residual 28 3.27 Deviations from 3 5.24 1.7n L i n e a r i t y 2 Within Groups 25 3 . 04 1 Testing the hypothesis that the slope of the regression curve i s zero. 2 Testing the hypothesis that the regression curve i s l i n e a r . 3 Testing the hypothesis that the slopes of both regression curves A and B are equal. 4 Testing the hypothesis that the elevations of both regression curves A and B are equal. 5 Combined r e s u l t s of assays A and B. S i g n i f i c a n t at p 0.01. n.s. not s i g n i f i c a n t at p > 0.05. 65 Slopes: t = . (b1 - bJ ^ ^ [ ( S 2 Y : X ) p / ( 2 x 2 ) 1 + ( S ^ y : X ) p / ( 2 x 2 ) 2 ] 1 / 2 where: b = 2 x y / £ x 2 (S 2 y. x) p = { f X Y 2 , ~ ( s x y ) 2 1 / ^ x 2 1 l + f y 2 2 - (£xy) 2 2 / g x 2 2 H [(DF, - 2) + (DF2 - 2)] Accept that the slopes are equal i f t > t Q 0 5 ( 2 ) v where: v = [(DF, - 2) + (DF2 - 2)] (Equation 13) Elevations: t = LLi,^i2) - b ( x 1 _ J ^ g 2 ) i , H S ^ . x J J l / n , + l / n 2 + (X, - X 2) 2/A c} 1 / 2 where: (S 2Y:X) c = {f ( ^ y 2 ) 1 + ( £ , y 2 ) 2 l - r f^xy) ,+ fcSxy) 2 1 2 / [ (£x2),+ (Sx 2) 21 n, + n 2 - 3 Ac = (fix 2), + (£x 2) 2 Accept that the elevations are equal i f t > t 0 5 ( 2 ) v (Equation 14) 3.5c Thin layer chromatography assay of carbohydrates 3.5cl Qualitative assay of carbohydrates Thin layer chromatography (TLC) using aminopropyl-bonded s i l i c a plates was used to separate and i d e n t i f y sugars following the method of Doner et a l . (1984). S i l i c a plates (Fisher S c i e n t i f i c Redi Plate S i l i c a Gel G, 250 microns, calcium bonded) were f i r s t d erivatized. 2 0 cm x 2 0 cm plates were immersed i n a TLC developing tank with a 1.0% 3-amino-propyl triethoxysilane (3APTS) dry hexane mixture f o r 15 minutes. The hexane was i n i t i a l l y dried overnight over calcium chloride i n a desiccator. The plates were then placed h o r i z o n t a l l y , covered i n 66 dry hexane for 15 minutes to rinse off excess 3APTS. Plates were dried at 60° C (> 30 i n . Hg) i n a vacuum oven (Model #5850-5 National Appliance Co., Portland OR) for a further 15 minutes. Aqueous solutions of 2 0% lactose, lactulose, glucose, galactose, fructose, and tagatose and a combination of 8% of a l l sugars were used as references to i d e n t i f y sugars of experimental samples. Standards (0.05 or 0.10 uL) were spotted at the plate's baseline. Concentrated whey permeate samples of 0.10 or 0.15 uL were spotted on the same plate. Plates were i r r i g a t e d with a 70:30 acetonitrile:ddH 20 mixture i n a TLC developing tank. Spots were made v i s i b l e by spraying with a 6:4:3 tert-butanol:ethanol:sulphuric acid mixture and heated at 180°C for 5 minutes. Rf factors were calculated and compared to Rf factors of the standard sugars for i d e n t i f i c a t i o n . The Rf factor i s defined as the rate of movement of the leading edge of the solute zone divided by the rate of movement of the leading edge of the solvent zone. In any given run, however, the distance t r a v e l l e d i s proportional to the rate of movement. The Rf factor for t h i s study then, was calculated as follows. Rf = distance the solute moved distance the solvent moved where: The distance the solute moved i s measured as the distance between the o r i g i n and the leading edge of the v i s i b l e spot. The distance the solvent moved i s measured as the distance between the o r i g i n and the mean of the solvent front l i n e measured at several points. (Equation 15) 3.5c2 Tagatose i d e n t i f i c a t i o n by TLC and spectrophotometry A 70:30 acetonitrile:ddH 20 solution and aqueous solutions of 67 lactose, lactulose, fructose, glucose, galactose, and tagatose were scanned on a spectrophotometer within the range of 200 to 400 nm, to determine the maximum absorbance wavelength of each (Table 14). S i l i c a plates were derivatized with 3APTS. A 20% tagatose aqueous solution and a reference mixture of 8% aqueous solutions of tagatose, glucose, and fructose were spotted at the baseline of one side of the 20 cm x 20 cm plate for spot v i s u a l i z a t i o n . Then, tagatose aqueous solutions of 0, 5.0, 10, 30, 50, 80, 100, and 150 mg/mL and concentrated process samples were spotted along the remaining baseline, a l l at 0.10 uL. Plates were i r r i g a t e d with 70:30 a c e t o n i t r i l e : ddH20 for 3 0 minutes. The plates were then protected, exposing only the two selected spots to the detection spray. Keeping the remaining spots protected, a hot a i r d r i e r heated the sprayed area u n t i l the spots were v i s i b l e . Using a scalpel and the v i s i b l e spots as a guide, tagatose standards and samples were scraped from the plate. As tagatose, glucose, and fructose a l l had absorbance peaks at 256 nm, the combination standard was used as a guide to ensure that only tagatose was removed. The samples were then mixed well with 1.0 mL ddH20, f i l t e r e d (Whatman #541) , and rinsed with 3 mL ddH20 into a visible/UV cuvette. A standard curve r e l a t i n g tagatose concentration and absorbance at 256.0 nm was prepared (Figure 14). The tagatose contents of process samples were then estimated based on absorbance at 2 56.0 and the standard curve. 68 Table 14 The absorbance of aqueous carbohydrate solutions and the TLC solvent by spectrophotometer between 2 00 and 4 00 nm. Carbohydrate Peak Absorbance Wavelength (nm) Solution Acetonitrile:ddH 20 2 3 3.0 Lactose 255.8 Lactulose 255.8 Glucose 256.0 Fructose 256.0 Tagatose 256.0 Galactose 256.4 69 Regression Curve: y = (.00072)x + 0.018 r 2 = 0.986 n = 10 Figure 14 The standard curve of the absorbance of tagatose concentrations at 256 nm. 70 3.5d Determination of nitrogen 3.5dl Determination of t o t a l nitrogen A micro-Kjeldahl method following the study of Concon and Soltess (1973) was used to determine the t o t a l nitrogen content at various steps i n the process. As described i n that study, samples (from 0.50 g to 5.00 g) were mixed with a s a l t - c a t a l y s t mixture of potassium sulphate and mercuric oxide. Then the sample was digested with sulphuric acid and hydrogen peroxide u n t i l the protein was converted to ammonium sulphate. A chemical colourimetric method was used to determine ammonia content i n the sample, using the "Technicon Autoanalyzer I I " , (Technicon Instruments Co., New York, NY). The ammonia was adjusted for an i n t e r n a l standard, then the nitrogen was calculated and converted to a percentage (w/w). Due to excessive foaming of the high carbohydrate sample in the current study, one change was made to the 19 73 procedure. Digestion f l a s k s of 100 mL were used instead of the standard 3 0 mL f l a s k s for some samples. As well, the process took approximately double the 10 minute digestion time suggested i n the method. 3.5d2 Determination of protein An estimate of t o t a l protein was determined using the Bio-Rad Protein Assay, micro-assay procedure. This dye-binding assay measured the colour change of Coomassie B r i l l i a n t Blue G-250 at a wavelength of 595 nm on spectrophotometer i n response to protein concentration. Samples were diluted i f necessary, dye was added, and a f t e r one hour the samples' absorbances were measured at 595 nm. Aqueous sugar solutions of similar concentration were used as 71 sample blanks. Blanks and a standard curve were prepared fresh for each t e s t i n g batch. The commercially prepared standard f o r the standard curve was a bovine gamma globulin protein (Bio-Rad Chem.) and was prepared and tested at f i v e concentrations (Figure 15). 3.5e Determination of t o t a l s o l i d s and ash Total s o l i d s and ash were determined i n sequence. Samples of from 1 to 5 grams of whey (depending on sample concentration) were weighed into pre-ashed, pre-weighed c r u c i b l e s . The samples were f i r s t concentrated at 60°C for 6 hours and then held i n a vacuum oven at 60°C for a further 15 hours, under > 3 0 i n . Hg of pressure. The portion of the i n i t i a l sample which remained a f t e r vacuum heating was calculated to be t o t a l s o l i d s . These same samples were then transferred to a muffle furnace (Box Type Muffle furnace, Blue M E l e c t r i c Co.) f o r 15 hours at a temperature of 425°C to 450°C. The remaining white residue after heating was considered ash. Dark spots i n the residue were dampened with a small amount of ddH20 and ashed for a further 15 hours. 3.5f Determination of pH and t i t r a t a b l e a c i d i t y The pH of samples was determined using a Corning pH meter, model 220, cal i b r a t e d against standard pH solutions. T i t r a t a b l e a c i d i t y (AOAC, 1990) was tested i n both the deproteinized whey and the retorted whey. L a c t i c acid i n the whey permeate was estimated using 0.10 M NaOH and phenolthalein as an indic a t o r . A conversion of 1.0 mL 0.10 M NaOH equalling 0.0090 g l a c t i c acid was used for c a l c u l a t i o n . 72 Regression Curve: y = (0.0088)x + 0.072 r 2 = 0.963 n = 19 Figure 15 The standard curve of the absorbance of bovine gamma globu l i n protein at 595 nm for the Bio-Rad Protein Assay. 73 To provide an estimation of the amount of acid produced during the heat treatment of lactose conversion, a d i f f e r e n t indicator and ca l c u l a t i o n were used. The pH of the permeate sample before heat treatment was 10.5; after heat treatment the sample pH dropped to 7.15. Berg and van Boekel (1994) suggested that the majority of acid produced during lactose degradation was formic acid; i t was assumed for c a l c u l a t i o n that t h i s was the only acid produced. A volume of 1.0 M NaOH was added to the post-heated permeate sample to bring the pH back up to 10.5. The NaOH used to r a i s e the pH would be proportional to an estimate of the amount of formic acid produced. Using the stoichiometry of the reaction of formic acid and NaOH to produce sodium formate, the amount of formic acid reacting with the added NaOH was calculated. 74 4. RESULTS AND DISCUSSION 4.1 OVERVIEW OF PROCESS This study describes a sequence of steps producing a high lactulose, mixed carbohydrate syrup from deproteinized whey. Figure 4 (p.29) i l l u s t r a t e s the ov e r a l l process i n i t s f i n a l stage. In summary, the deproteinized whey was f i r s t d e c a l c i f i e d , then heat treated under s p e c i f i c conditions to convert some of the lactose to lactulose, p a r t i a l l y p u r i f i e d by cold p r e c i p i t a t i o n , and f i n a l l y deionized, decolourized, and concentrated. Analyses of the lactulose preparation at the various steps i n the process tracked the major compositional changes which occurred. The r e s u l t s recorded i n Table 15 were gathered a f t e r the f r a c t i o n a l f a c t o r i a l experiments of conversion and p u r i f i c a t i o n , using the selected process conditions and levels of each study. Table 15 provides the best r e s u l t s of the lactulose preparation process i n i t s f i n a l stage. Results were based on the t o t a l s o l i d s of each sample as t h e i r d i l u t i o n s varied greatly. While t h i s adjustment made comparisons possible, some comparisons were s t i l l d i f f i c u l t . Minerals and sugars were added and removed from the preparation at various times, a l t e r i n g the amount of the s o l i d s . Whey UF permeate (sample A i n Table 15) i s s i m i l a r to the UF cheddar permeate of Table 3 (p.10) when the concentrations are adjusted to r e f l e c t concentrations based on the t o t a l s o l i d s . The l a c t i c acid content of the UF permeate used i n t h i s study i s somewhat lower than the sample in Table 3, and the nitrogen as well as protein content i s substantially higher. The difference i n protein i s probably due to a difference i n u l t r a f i l t r a t i o n effectiveness between the two samples. 75 Table 15 Proximate analysis, based on % t o t a l s o l i d s , of whey permeate at d i f f e r e n t stages of lactulose manufacture and p u r i f i c a t i o n . A Stages B Of C the process 1 D E F G Solids (%) (w/v) 5. 20 11.8 9 .77 10.2 9. 55 4. 16 2 . 58 Relative Solids Weight 2 100 104 . 4 48 . 9 41. 3 34. 1 Ash (%) 3 8. 04 7.68 12 .9 12.5 11. 6 1. 67 1. 82 Nitrogen (%) (w/w)3 0. 61 0. 61 0. 17 Protein (%) 3 0. 48 0.44 0.32 0. 094 PH 6. 1 6.5 10 .5 7.2 5. 0 T i t r a t a b l e Acid 4 (%) 0. 089 1.6 Lactose (% w/v)3 83 78 72 54 22 24 26 Lactulose (% w/v)3 0. 55 0.89 5 .4 19 55 58 59 Glucose (% w/v)3 0. 68 0.75 0 .70 0.53 0. 58 0. 77 1. 0 Fructose (% w/v)3 0. 092 0.13 0 .38 0.44 0. 40 0. 64 0. 81 Galactose (% w/v)3 2 . 2 2.4 2 . 3 5.2 5. 1 4. 7 5. 0 Sample A: Deproteinized whey Figure 4 B: Decalcified whey C: Modified whey D: Retorted whey E: P u r i f i e d lactulose preparation F: Deionized lactulose preparation G: Decolourized lactulose preparation Relative s o l i d s weights based on sample volumes. Calculated based on t o t a l s o l i d s of the sample, section 3.5e. T i t r a t a b l e acid calculated as l a c t i c acid for sample A and formic acid for sample D, section 3.5f. 2 3 4 76 The f i n a l product, sample G - decolourized lactulose preparation, i s quite d i f f e r e n t in composition from the lactulose syrups detailed i n Table 1 (p.6). In that t y p i c a l commercial lactulose syrup, lactulose i s the dominant sugar, up to 70% (w/v). Lactose, galactose and other sugars are present i n much lesser amounts, less than 13% (w/v) each. The f i n a l lactulose preparation of the current study was not concentrated to a syrup so straight comparison i s impossible. However, of the t o t a l s o l i d s of sample G, j u s t under 60% was lactulose and 2 6% was lactose. Sample G had much less lactulose and substantially more lactose than commercial syrups. 4.2 THE CONVERSION OF LACTOSE TO LACTULOSE The influence of eight factors - pH, NaOH, concentrations of sodium phosphate, c i t r a t e , and lactose, substrate p u r i f i c a t i o n , heating temperature, and heating time - on lactulose y i e l d during heat treatment of whey permeate was investigated using two Taguchi designs (L 2 ?3 1 3) (Figure 5) . The 27 t r i a l s for each design are described along with results i n Tables 16 and 17. The enzymatic assay for lactulose described i n Section 3.5bl includes fructose i n i t s measurement of lactulose. For t h i s reason fructose was assayed separately i n the f i r s t experimental design, to be subtracted from the measured lactulose. This provided a more accurate measurement of lactulose. As the fructose levels were very low i n the high end lactulose y i e l d samples, (less than 0.2% of measured lactulose was fructose) and te s t i n g was costly, fructose t e s t i n g was not continued for design #2. To eliminate the discrepancy between the two designs, reported re s u l t s of both include free fructose i n the 77 Table 16 Design factor combinations and r e s u l t s summary of the L 2 7(3 1 3) design #1. T r i a l no. Temp.' (°C) time (min) lactose (mg/mL) phosphate (mM) c i t r i c a c i d (mM) PH lactulose y i e l d (%) 1 90 5 40 40 40 9.0 4. 0 3.8 2 110 20 40 40 70 10. 5 28 28 3 130 90 40 40 100 12.0 4. 1 4.1 4 110 20 40 70 40 12.0 7. 8 7.6 5 130 90 40 70 70 9.0 15 15 6 90 5 40 70 100 10. 5 18 18 7 130 90 40 100 40 10. 5 12 12 8 90 5 40 100 70 12.0 8. 9 9.5 9 110 20 40 100 100 9.0 27 28 10 130 20 79 40 40 10.5 29 29 11 90 90 79 40 70 12.0 16 16 12 110 5 79 40 100 9.0 18 18 13 90 90 79 70 40 9.0 23 21 14 110 5 79 70 70 10.5 26 25 15 130 20 79 70 100 12.0 8. 7 9.9 16 110 5 79 100 40 12.0 13 13 17 130 20 79 100 70 9.0 23 25 18 90 90 79 100 100 10.5 25 26 19 110 90 115 40 40 12.0 18 17 20 130 5 115 40 70 9.0 27 27 21 90 20 115 40 100 10.5 20 20 22 130 5 115 70 40 10.5 30 30 23 90 20 115 70 70 12.0 18 18 24 110 90 115 70 100 9.0 27 26 25 90 20 115 100 40 9.0 11 11 26 110 90 115 100 70 10.5 28 26 27 130 5 115 100 100 12.0 12 12 Each sample experiment was performed i n duplicate, A and B. % lactulose was based on lactulose formed from o r i g i n a l lactose. 78 Table 17 Design factor combinations and r e s u l t s summary of the L 2 7(3 1 3) design #2. T r i a l no. Temp. (°C) Time (min) P u r i f i c a t i o n 1 Phosphate (mM) C i t r a t e (mM) NaOH (mM) l a c t u l o s (%) 1 105 10 DPW 00 50 50 26 27 2 115 40 DPW 00 70 100 20 19 3 125 80 DPW 00 90 18 26 26 4 115 40 DPW 50 50 18 25 25 5 125 80 DPW 50 70 50 18 17 6 105 10 DPW 50 90 100 19 18 7 125 80 DPW 100 50 100 10 9 8 105 10 DPW 100 70 18 25 25 9 115 40 DCW 100 90 50 21 21 10 125 40 DCW 00 50 100 20 20 11 105 80 DCW 00 70 18 24 25 12 115 10 DCW 00 90 50 28 29 13 105 80 DCW 50 50 50 27 28 14 115 10 DCW 50 70 100 23 22 15 125 40 DCW 50 90 18 26 26 16 115 10 DCW 100 50 18 25 25 17 125 40 DCW 100 70 50 21 20 18 105 80 DCW 100 90 100 19 18 19 115 80 LS 00 50 18 24 24 20 125 10 LS 00 70 50 29 30 21 105 40 LS 00 90 100 24 23 22 125 10 LS 50 50 100 24 25 23 105 40 LS 50 70 18 18 17 24 115 80 LS 50 90 50 25 25 25 105 40 LS 100 50 50 27 27 26 115 80 LS 100 70 100 17 17 27 125 10 LS 100 90 18 28 28 DPW = deproteinized whey permeate DCW = d e c a l c i f i e d DPW LS = lactose solution Each sample experiment was performed i n duplicate, A and B. % Lactulose based on 73 mg/mL i n i t i a l lactose. 79 lactulose concentration. Heat processing caused major changes to the adjusted whey permeate, as can be seen i n Table 15. Lactulose and galactose contents increased while lactose content decreased. Lactose was converted to lactulose which i n turn was degraded to galactose as det a i l e d i n Figure 2 (p. 13). A substantial drop i n pH from 10.5 to 7.15 was caused by the production of acids, possibly formic and/or isosaccharinic acid. The maximum lactulose y i e l d was approximately 3 0% of the i n i t i a l lactose content, and the treatment produced a l i q u i d of golden amber colour. This conversion far exceeded reported studies manipulating lactulose conversion in milk using heat processing and pH adjustment. As described in Section 2.5, Andrews (1989) found a maximum of approximately 2 00 mg/100 mL lactulose i n heated milk by i n d i v i d u a l l y adjusting lactose concentration, pH, addition of c i t r a t e and phosphate, heating temperature, or heating duration. This corresponds to approximately a 4% lactulose y i e l d . The larger y i e l d i n the present study was c e r t a i n l y due i n part to the high pH of whey treatment, a pH that i s not p r a c t i c a l with milk. S t i l l , the lactulose y i e l d achieved i n the present study c l e a r l y did not reach the 87% conversion l e v e l of one borate and triethylamine method (Hicks & Parrish, 1980) . 4.2a The influence of pH and NaOH concentration In the f r a c t i o n a l f a c t o r i a l design #1, pH at three l e v e l s -9.0, 10.5, and 12.0 - were shown to s i g n i f i c a n t l y influence lactulose y i e l d (p<0.01) using ANOVA (Table 18). This i s shown in Figure 16a, with higher y i e l d occurring at the mid pH l e v e l of 80 Table 18 Analysis of variance [Taguchi L 2 7(3 1 3) ] , design #1 and #2, obtained from 27 experiments of heat processed demineralized whey permeate. Source of v a r i a t i o n DF1 Mean Square F - r a t i o 2 Design #1 Lactose concentration (Lac) 2 127 45 ** Phosphate concentration (Phos) 2 3 .97 0 .56 n.s.3 C i t r i c acid concentration (Cit) 2 54 . 8 19 ** PH 2 336 120 ** Heating temperature 2 62.2 22 ** Heating duration 2 9.60 3 .4 n.s. Lac X Phos 4 30.5 11 * Phos X C i t 4 51.2 18 ** C i t X Lac 4 32 . 9 12 * Error 4 2.84 Design #2 P u r i f i c a t i o n (Pur) 2 22 . 3 2 .4 n.s. Phosphate concentration (Phos) 2 27.6 3 .0 n.s. Sodium c i t r a t e concentration (SCit) 2 14.3 1 .6 n.s. NaOH concentration 2 88 . 2 9 .6* Heating duration 2 43.7 4 .7 n.s. Pur X Phos 4 9.24 1 .0 n.s. Phos X SCit 4 12.3 1 .3 n.s. SCit X Pur 4 3.51 0 .4 n.s. Error 4 9.23 Degrees of freedom _0.05(1)2,4 0^.01(1)2,4 0^.05(1)4,4 ?0.01(1)4,4 6.94 18 . 0 6.39 16.0 * ** n. s 3 s i g n i f i c a n t at p ^  0.05 s i g n i f i c a n t at p ^ 0.01 not s i g n i f i c a n t at p > 0.05 The mean square for phosphate concentration was <1.0 and was incorporated into the error sums of squares. Figure 16a T3 CD > CD CO O o CO 81 Figure 16b 30 5 CD >-CD CO O «—• O CO 2 5 h 20 h 15 h 10 5 h PH J I L I I I I I I I I I I I I I I 1_ 0 10 20 30 4 0 50 60 7 0 8 0 90 1 0 0 1 1 0 1 2 0 NaOH Concentration (mM) — V— • —A — Design #2 samples Design #2 pH's Design #1 samples Figure 16 a,b The influence of pH and NaOH concentration on lactulose y i e l d during heat treatment of whey permeate using an L 2 7(3 1 3) f r a c t i o n a l f a c t o r i a l design. 82 10.5. In design #2, the NaOH concentration factor was substituted for pH. The 2 7 t r i a l r e sults of design #2 were incorporated into the pH figure by using the pHs measured for each sample. These data points, and those i n design #1, suggested that a pH of 10.5 to 11.0 led to the highest lactulose y i e l d . A pH of 11.0 was also found by Martinez-Castro and Olano (1980) and Hicks et a l . , (1984) to correspond to a higher lactulose y i e l d when u t i l i z i n g boric acid and triethylamine. NaOH concentration tested by ANOVA i n design #2 was found to be s i g n i f i c a n t (p<0.05) at the levels 18, 50, and 100 mM, Table 18. The lower NaOH concentrations corresponded to a higher lactulose y i e l d , Figure 16b. The set of conditions used i n further study included a pH at between 10.5 and 11.0, set by the addition of NaOH, (Table 8, p.40) . 4.2b The influence of sodium phosphate concentration Sodium phosphate concentration showed no s i g n i f i c a n t influence (p>0.05), at the level s of 40, 70 and 100 mM i n design #1, nor at the l e v e l s of 0, 50 and 100 mM i n design #2, Table 18. Sodium phosphate concentration was shown to be s i g n i f i c a n t i n influencing lactulose y i e l d i n design #1 as i t interacted with lactose (p^0.05) and c i t r i c acid concentration (p^O.01), Table 18. The l e v e l of phosphate influenced the e f f e c t that lactose concentration had on lactulose y i e l d . At the important higher l e v e l s of lactose concentration, the low and mid-phosphate concentrations corresponded to a higher lactulose y i e l d , Figure 17a. In Figure 17b, sodium phosphate concentration influenced the Figure 17a 30 i 83 25 h 20 (D CO o u CO 15 h 10 h 5 h T ? T -A 1 T -v 1 'l I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 'I I I l' 30 40 50 60 70 80 90 100 110 120 Lactose Concentration (mg/mL) Figure 17b 30 40 50 60 70 80 90 100 110 120 Citric Acid Concentration (mM) - A — - • - - V -40 mM phosphate 70 mM phosphate 100 mM phosphate Figure 17 a,b The influence of sodium phosphate concentration i n t e r a c t i n g with lactose and c i t r i c acid concentrations on lactulose y i e l d of heat treated whey permeate using an L 2 7(3 1 3) design. 84 a b i l i t y of c i t r i c acid to increase lactulose y i e l d . At the important mid l e v e l of c i t r i c acid, i t was the lowest phosphate l e v e l which corresponded to the highest lactulose y i e l d . In design #2, with modified le v e l s , no interactions between sodium phosphate and sodium c i t r a t e or lactose were found. The study of Andrews and Prasad d e t a i l e d i n Section 2.5c determined that sodium phosphate was i n f l u e n t i a l i n increasing lactulose y i e l d i n heat treated milk and i t was suggested that i t acted as a catalyst i n the LA reaction. In that study, sodium phosphate was tested along with (but independent of) c i t r i c acid and the l a t t e r proved a better catalyst (Andrews & Prasad, 1987). In the current study, sodium phosphate addition was not included i n further study due to i t s questionable effectiveness and higher cost. 4.2c The influence of c i t r a t e concentration In design #1, c i t r i c acid at 40, 70, and 100 mM l e v e l s showed a s i g n i f i c a n t influence (p<0.01) on lactulose y i e l d , (Table 18). These r e s u l t s are i n l i n e with those of Andrews and Prasad (1987), who found that c i t r i c acid influenced lactulose y i e l d i n heat treated milk. In design #2, c i t r i c acid was replaced with sodium c i t r a t e , more suited for higher pH solutions. The c i t r a t e levels were narrowed to 50, 70, and 90 mM and there was no s i g n i f i c a n t influence (p>0.05). Figure 18 i l l u s t r a t e s the r e s u l t s of both designs. C i t r i c acid displayed a s i g n i f i c a n t i n t e r a c t i o n with lactose and phosphate concentrations in Table 18. In Figure 19a, the mid l e v e l of c i t r i c acid and the highest lactose concentration led to 85 CD > CD CO o o CO 30 40 50 60 70 80 90 100 110 120 Design #1 samples Citrate Concentration (mM) - • -Design #2 samples Figure 18 The influence of c i t r i c acid / sodium c i t r a t e concentration on lactulose y i e l d during heat treatment of whey permeate using combined re s u l t s of L 2 7(3 1 3) designs #1 and #2. Figure 19a 86 2 g> >-CD CO o o CO 30 40 50 60 70 80 90 100 110 120 Lactose Concentration (mg/mL) Figure 19b 2 <D >• (D CO o u CO 30 40 50 60 70 80 90 100 110 120 — A — 40 mM citric acid Phosphate Concentration (mM) • •• • - v -70 mM citric acid 100 mM citric acid Figure 19 a,b The influence of c i t r i c a c i d concentration interacting with lactose and sodium phosphate concentrations on lactulose y i e l d during heat treatment of whey permeate using an L 2 7(3 1 3) design. 87 the highest lactulose y i e l d . In Figure 19b, the highest lactulose y i e l d resulted from the combination of no phosphate and again the mid-level of c i t r i c acid. There was no s i g n i f i c a n t i n t e r a c t i o n (p>0.05) between sodium c i t r a t e and sodium phosphate concentration or lactose concentration i n design #2, Table 18. Sodium c i t r a t e was added at a l e v e l of 50 mM i n further study. While there was no s i g n i f i c a n t difference shown among 50, 70, and 90 mM, the lowest was selected as the most economical. 4 .2d The influence of lactose concentration and p u r i f i c a t i o n In experimental design #1, lactose concentration was determined to s i g n i f i c a n t l y influence lactulose formation (p$0.05) (Table 18 and Figure 20) . The lower l e v e l of 40 mg/mL lactose corresponded to a substantially lower lactulose y i e l d than the higher l e v e l s of 79 and 115 mg/mL which produced s i m i l a r r e s u l t s i n y i e l d . Andrews (198 9) found that at le v e l s of normal lactose concentrations (approximately 50 mg/mL) and higher, there was a f i r s t order reaction i n regard to lactulose y i e l d . Higher concentrations of lactose also increased lactulose y i e l d as they interacted with phosphate and c i t r i c acid concentrations as shown in Figures 21a,b. The rate of p u r i f i c a t i o n of the substrate was studied i n experimental design #2. When deproteinized whey, deproteinized d e c a l c i f i e d whey, and a lactose solution were compared, there was no s i g n i f i c a n t difference (Table 18). The set of conditions selected for further study, Table 8, included d e c a l c i f i e d UF whey permeate of lactose concentration greater than 7 0 mg/mL. Further p u r i f i c a t i o n of the whey through 88 Figure 20 The influence of lactose concentration on lactulose y i e l d during heat treatment of whey permeate using L 2 7(3 1 3) design #1. Figure 21a 2 CD > CD co O O CO 30 25 30 40 50 60 70 80 90 100 110 Phosphate Concentration (mM) Figure 21b 120 2 CD > (D CO o o CO 20 h 15 10 i i i i I i • I i i i i I i i i i I i i i i I i i i i I i i i i I i i i i I i i i i 89 30 40 50 60 70 80 90 100 110 120 Citric Acid Concentration (mM) —A— • • • - V -40 mg/mL lactose 79 mg/mL lactose 115 mg/mL lactose Figure 21 a,b The influence of lactose concentration inter a c t i n g with sodium phosphate and c i t r i c acid concentrations on lactulose y i e l d during heat treatment of whey permeate using an L 2 7 ( 3 1 3 ) design. 90 demineralization p r i o r to heat treatment was not considered advantageous based on the re s u l t s of t h i s study and on minimizing cost. The demineralization step a f t e r heat treatment and p u r i f i c a t i o n removed added as well as natural s a l t s . 4.2e The influence of temperature and time of heat treatment Temperature and time of heat treatment at the chosen levels played a lesser r o l e i n the production of lactulose than was anticipated. While there was a s i g n i f i c a n t difference (p^O.Ol) i n the three widespread levels of temperature 90, 110, and 130°C i n design #1, there was no difference when the l e v e l s were narrowed to 105, 115, and 125°C, (Table 18 and Figure 22). There may have been less conversion to lactulose at 90° C and a higher amount of lactulose degradation at 130°C. I t i s also possible that a temperature as high as 13 0°C might have shown a higher lactulose y i e l d at a time shorter than 5 minutes. Heating time showed no s i g n i f i c a n t difference (p>0.05) i n design #1 nor i n #2. At a range of 5 to 9 0 minutes, there was s u r p r i s i n g l y l i t t l e response i n lactulose y i e l d . 4.2f Continuous flow heat exchanger Using the continuous flow heat exchanger described i n Section 3.3, acceptable lactulose conversion rates were noted at a flow rate of 6.8 mL/min. At t h i s flowrate, a 2.3:1 r a t i o lactose:lactulose solution was produced. At flow rates below 6 mL/min, some of the sugar solution burned onto the i n t e r i o r surface of the piping. A thermocouple was inserted into the copper piping 177 cm 91 CD >-CD oo o o CO 30 25 h 20 h 15 h 10 h 'i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i t i i i i i i i r 85 90 95 100 105 110 115 120 125 130 135 Design #1 samples Temperature (°C) - CD-Design #2 samples Figure 22 The influence of temperature on la c t u l o s e y i e l d during heat treatment of whey permeate using the combined res u l t s of L 2 7 ( 3 1 3 ) designs #1 and #2. 92 af t e r entering the o i l , roughly 15% of the distance the whey t r a v e l l e d . The whey temperature was within 2C° of the surrounding o i l , even when the flowrate was more than doubled, to 15 mL/min. Another thermocouple, placed at the mid-point of the cooling section of the pipe, monitored the whey permeate i n the ice bath. Here, the permeate temperature was always less than 6°C, even when the flowrate increased to over 30 mL/min. 4.3 PURIFICATION 4.3a Fr a c t i o n a l f a c t o r i a l design for cold p r e c i p i t a t i o n A f r a c t i o n a l f a c t o r i a l design L 2 7(3 1 3) was used to t e s t the influence of 4 factors - pH, concentration, the rate of temperature decrease, and the f i n a l temperature - on the p r e c i p i t a t i o n of lactose and lactulose i n a concentrated and cooled sugar solution. The substrate for t h i s p u r i f i c a t i o n study was a sugar solution which simulated the sugar r a t i o of the deproteinized d e c a l c i f i e d whey a f t e r heat treatment outlined i n Table 8. The 27 sample solutions of the f a c t o r i a l design were p r e c i p i t a t e d according to the experimental plan. The samples were then decanted and the p r e c i p i t a t e and decant portions were assayed for lactose and lactulose (Table 19). Percent lactulose y i e l d (the percentage of t o t a l lactulose which was not precipitated), percent lactose y i e l d (the percentage of t o t a l lactose which was not precipitated) and sugar r a t i o (the lactulose to lactose r a t i o i n the decant portion) were calculated using equations 5, 6, and 7 respectively. D i f f e r e n t calculations of the data provided a d i f f e r e n t perspective on the p r e c i p i t a t i o n : while a high lactulose y i e l d i s desirable, a 100% y i e l d i s achieved with no p r e c i p i t a t i o n whatsoever. Some 93 Table 19 Results of the L 2 7(3 1 3) design, assaying lactose and lactulose i n both decant and p r e c i p i t a t e portions af t e r cold p r e c i p i t a t i o n of a 28:72 lactulose: lactose solution. Sample Lactulose Y i e l d 1 Lactose Y i e l d 2 Lactulose/Lactose 3 (%) (%) unprecipitated 0.39 1 96 89 0.42 2 90 61 0. 62 3 83 39 0.95 4 92 59 0. 63 5 90 39 0.91 6 92 88 0.47 7 87 40 0.93 8 96 94 0.53 9 96 87 0.53 10 83 64 0.57 11 77 30 1.1 12 94 92 0.44 13 76 32 0.97 14 98 85 0. 49 15 91 69 0.56 16 100 96 0. 50 17 94 82 0.54 18 87 51 0.80 19 72 27 1.1 20 98 91 0.43 21 92 66 0.59 22 97 93 0.44 23 97 67 0. 57 24 80 41 0.81 25 96 77 0.57 26 87 43 0.92 27 95 97 0. 50 Lactulose (mg/mL) of decant . Lactulose (mg/mL) of decant + Lactulose (mg/mL) of p r e c i p i t a t e Lactose (mg/mL) of decant ._ Lactose (mg/mL) of decant + Lactose (mg/mL) of p r e c i p i t a t e Lactulose (mg/mL) of decant Lactose (mg/mL) of decant 94 p r e c i p i t a t i o n of lactulose inevitably occurs as lactose i s p r e c i p i t a t e d . On the other hand, sugar r a t i o alone cannot provide information on the lactulose l o s t i n the process. The r e s u l t s of the experiment included a wide v a r i e t y of y i e l d and r a t i o combinations. Lactulose y i e l d s ranged from 77% to 100%. This process did not aggressively p r e c i p i t a t e lactulose and i n a l l cases p r e f e r e n t i a l l y precipitated lactose over lactulose. The sugar r a t i o s of the decant portion ranged from 0.42 to 1.1, showing a marked increase i n the lactulose to lactose r a t i o from the i n i t i a l 0.3 9 r a t i o of the heat treated permeate. The r e s u l t s of t h i s study c e r t a i n l y i l l u s t r a t e the differences i n s o l u b i l i t i e s between lactulose and lactose at low temperatures, as d e t a i l e d i n Table 5. Samples 3, 5, 7, 11, 13, 19, and 26, deemed by t h i s author to have the most promising r e s u l t s , had on average a lactulose y i e l d of 82% and a 1:1 lactulose: lactose r a t i o . This r a t i o i s not as desirable as those found i n commercial lactulose mixtures, such as those described i n Table 1 with a 63-70:4-8 lactulose:lactose r a t i o . There was also a substantial loss of lactulose as well. Using the same seven samples as above, there was on average an 18% loss of lactulose through p u r i f i c a t i o n . This corresponded to a 64% loss of lactose. An ANOVA was used to test the s i g n i f i c a n c e of the factors and factor interactions on lactulose y i e l d and sugar r a t i o (Table 20). The pH and sugar concentration of the solution were found to be s i g n i f i c a n t (p^0.05). The f i n a l temperature of cooling and the rate at which the solutions were cooled did not s i g n i f i c a n t l y influence p r e c i p i t a t i o n at the chosen l e v e l s (p>0.05). The influence of pH and sugar concentration on p r e c i p i t a t i o n i s shown 95 Table 20 Analysis of variance of the L ? 7(3 1 3) design of cold p r e c i p i t a t i o n i n a lactulose:lactose solution. Source of Variation DF Mean Square F-value 1' 2 ANOVA % lactulose y i e l d Temperature drop 2 12.6 1.4 n.s. pH 2 81.9 8.8 ** F i n a l temperature 2 24.3 2.6 n.s. Concentration 2 469 51 ** Error 10 9 .28 ANOVA lactulose:lactose r a t i o Temperature drop 2 0.00320 0.01 n.s. 2 pH 2 0.00460 0.77 n.s. F i n a l temperature 2 0.00770 1.9 n.s. Concentration 2 0.546 140 ** Error 22 0.00397 n0.05(1)2,10 ^0.01(1)2,10 ;V 05(1)4,10 ^0.01(1)4,10 V05(1)2,22 ^0.01(1)2,22 4. 7, 3 , 5, 3 . 5, 10 56 48 99 44 72 * s i g n i f i c a n t at p < 0.05 ** s i g n i f i c a n t at p ^  0.01 n.s. not s i g n i f i c a n t at p > 0.05 2 The F-values of indicated factors and a l l interactions were below 1.0 and were incorporated into the error sums of squares. 96 in Figures 23a,b. Both lactulose y i e l d and lactulose:lactose r a t i o are shown. While a pH of 7, 9, or 11 had no s i g n i f i c a n t e f f e c t on the lactulose:lactose r a t i o , higher pH values increased lactulose y i e l d . Increasing sugar concentration had a dual e f f e c t , increasing the lactulose:lactose r a t i o but also decreasing lactulose y i e l d . 4 . 3 b The second cycle of cold p r e c i p i t a t i o n The p r e c i p i t a t e and decant portions of the f i r s t p r e c i p i t a t i o n cycle were adjusted to either 14:86 or 50:50 lactulose:lactose r a t i o , raised i n pH to 10.5, reconcentrated and cold-precipitated a second time. Five 2.0 mL re p l i c a t e samples of each portion were pr e c i p i t a t e d a second time, separated, and assayed, (Table 21). Aft e r the second cycle of p r e c i p i t a t i o n , when the decant from the f i r s t cycle was used, a lactulose y i e l d of approximately 78% was achieved. The lactulose:lactose r a t i o was 3.4:1. This means that a f t e r a second p r e c i p i t a t i o n cycle, the majority of the sugar i s lactulose. After the f i r s t p r e c i p i t a t i o n cycle, lactulose and lactose contents were about equal. When the pr e c i p i t a t e portion of the f i r s t cycle was used for a second cycle, a much lower lactulose y i e l d of approximately 60% occurred. The lactulose:lactose r a t i o was approximately 0.66:1, s i m i l a r to many of the 27 samples from the f i r s t cycle. The e f f i c i e n c y of two pr e c i p i t a t i o n s can be calculated t h e o r e t i c a l l y , for example using sample #11 and the mean r e s u l t of the second cycle. After two p r e c i p i t a t i o n cycles, a substantial 40% of the i n i t i a l lactulose would be l o s t during processing. However, with t h i s process i t i s possible that the lactulose l o s t Figure 23a 97 TJ cc CD CO o o CO 100 1.20 1.00 H 0.80 H 0.60 H 0.40 H 0.20 0.00 o CO k_ CD CO O ' o CO CD co o o CO Figure 23b 32 CD CD CO o o CO 100 1.20 H 1.00 H 0.80 H 0.60 H 0.40 H 0.20 0.00 o CO V— CD CO o o CO CD CO o o CO 25 30 35 40 45 50 Sugar concentration (%) - A — - • -Lactulose yield (%) Lactulose/lactose ratio Figure 23 a,b The influence of pH and sugar concentration on the cold p r e c i p i t a t i o n of a lactulose: lactose solution using an L 2 7(3 1 3) design. 98 Table 21 Results of 5 repl i c a t e s of a second cycle of cold p r e c i p i t a t i o n i n a lactulose: lactose solution. Sample Lactulose Y i e l d Lactulose/ (%) Lactose r a t i o Mean ± SD (range) Decanted portion of the f i r s t cycle 50:50 lactulose: lactose 78 ± 8.0 (67 - 88) 3.4 ± 0.37 (3.0 - 3.9) Prec i p i t a t e d portion of f i r s t cycle 14:86 lactulose: lactose 60 ± 8.9 (51 - 72) 0.66 ± 0.18 (0.49 - 0.90) 99 could be recovered through re-cycling of the p r e c i p i t a t e portions. In Table 15, the lactulose: lactose r a t i o of Sample E - the p u r i f i e d lactulose preparation, was 55:22 (or 2.5:1). This i s lower than the 3.4:1 r a t i o achieved using a lactulose/lactose sol u t i o n . This may be due to the natural and added s a l t s i n the whey permeate i n t e r f e r i n g with p r e c i p i t a t i o n . 4.3c Demineralization and decolourization Ion-exchange (Section 3.4e) was used to remove natural and added s a l t s . I t can be seen i n Table 15 that the ion-exchange was successful i n reducing the ash content of the permeate from an i n i t i a l 8.04% (of t o t a l solids) and post-purified l e v e l of 11.6% to a l e v e l of 1.67% after demineralizing. The ion-exchange did adsorb some of the carbohydrates as was calculated from the volumes before and a f t e r ion-exchange. There was 6.1% of the i n i t i a l t o t a l s o l i d s l o s t during ion-exchange. There did not appear to be any p r e f e r e n t i a l adsorption as the r a t i o of carbohydrates are similar before and a f t e r demineralization. The amber colour of the lactulose preparation caused by nonenzymatic browning during heating was s i g n i f i c a n t l y reduced by activated charcoal (Section 3.4f) and to a small degree by anion exchange. When the preparation was concentrated, however, a yellow tinge became apparent (Figure 24), which would c e r t a i n l y become more d i s t i n c t i f the preparation was further concentrated to a syrup. More importantly, when the volumes of the whey permeate before and a f t e r decolourization were taken into account, there was a 30% reduction i n t o t a l s o l i d s . Again, the r a t i o of the carbohydrates remained unchanged. 100 Figure 24 Lactulose preparation at three stages of processing, from l e f t to r i g h t - d e c a l c i f i e d UF whey permeate (24.7% TS), a f t e r p r e c i p i t a t i o n (10.4% TS), and a f t e r deionization and decolourization (23.3% TS). 101 4.4 PROXIMATE ANALYSIS 4.4a Carbohydrate standard curves using enzymatic assays Standard curves were developed for glucose, galactose, lactose, lactulose, and fructose using the f i v e standard solutions d e t a i l e d i n Table 10. In the standard curve for glucose (Figure 9) , the sig n i f i c a n c e s of the l i n e a r i t y and slope of the regressional curve were tested using ANOVA (Table 11) . Similar r e s u l t s are shown for galactose and lactose, Figures 10 and 11. For a l l three carbohydrate standard curves, the r 2 value was 0.997 or greater, the regression curve was linear, and i n a l l , s i g n i f i c a n t p o s i t i v e slopes existed. Fructose and lactulose were assayed i n a somewhat d i f f e r e n t manner: they were either assayed alone or i n conjunction with glucose and lactose respectively (Section 3.5b5). The assay used for these carbohydrates was o r i g i n a l l y used for quantifying lactulose only. In the method of Hicks et a l . (1984), the authors stated that glucose would inte r f e r e with the enzymatic assay of fructose. In the current study, lactulose and fructose were assayed with and without removing glucose. If there proved no difference i n measuring lactulose or fructose with or without the removal of glucose, then fructose could be assayed with glucose and lactulose could be assayed with lactose with a considerable reduction i n cost. Each regressional curve was f i r s t tested separately for slope and l i n e a r i t y . The slope and elevation of the two regression assay curves for each sugar were then compared by a t - t e s t for lactulose and fructose (Tables 12 and 13) . There was no s i g n i f i c a n t 102 difference r e s u l t i n g from glucose removal for eithe r the fructose or lactulose assay. This means that there was no influence on fructose measurement from the glucose present i n the samples tested i n t h i s study. The data for each sugar were combined and t h e i r regressional curves tested by ANOVA. Standard curves are depicted i n Figures 12 and 13, ANOVA resu l t s of the combined curves are summarized i n Tables 12 and 13, for lactulose and fructose respectively. 4.4b A c t i v i t y of beta-galactosidase The a c t i v i t y of a galactosidase enzyme, i t s a b i l i t y to hydrolyze lactose or lactulose, may lessen during conditions of storage. I t was important to test the a c t i v i t y of the fungal enzyme at the s t a r t and the end of experimentation. The a c t i v i t y of the beta-galactosidase enzyme from the Asp e r g i l l u s oryzae fungus was estimated to be 23 0 a c t i v i t y units/mg enzyme. One unit of enzyme a c t i v i t y i s defined as the amount that produces 1 micromole of o-nitrophenol per hour under the s p e c i f i c conditions described i n Section 3.5b2. Using ONPG and following the method i n that section, a standard curve was prepared measuring the absorbance of increasing concentrations of the hydrolysed product, o-nitrophenol at 420 nm (Figure 8). The regression curve of the enzyme, tested using ANOVA, was lin e a r at a significance of p .£0.01, and had a s i g n i f i c a n t p o s i t i v e slope. The corresponding r 2 value was 0.992. The beta-galactosidase was used i n experimentation for 22 months; the enzyme's a c t i v i t y was assayed i n t r i p l i c a t e at the beginning and end of experimentation. Using a two-tailed variance r a t i o t e s t (Section 3.5b2), there was no s i g n i f i c a n t difference 103 between the variance of the sample sets, concluding they were sampled from the same population. I t i s suggested then that the a c t i v i t y of the beta-galactosidase did not s i g n i f i c a n t l y change during the study. 4.4c Thin-layer chromatography q u a l i t a t i v e assay of carbohydrates Thin-layer chromatography (TLC) was used to determine i f s i g n i f i c a n t amounts of any unexpected carbohydrates were present at any stage of the process. Derivatized s i l i c a plates, following the method i n Section 3.5cl, were spotted with carbohydrate standards, standard mixtures, and process samples. An example of a plate spotted with standard carbohydrates and samples i s shown i n Figure 25 with corresponding Rf values i n Table 22. Doner et a l . (1984), used the same TLC method and tested many of the same sugars. Their Rf findings of lactose (0.187) and lactulose (0.215), were very s i m i l a r to the values obtained in t h i s study, (0.18 and 0.21-0.22) for lactose and lactulose. The Rf values of the monosaccharides obtained i n t h i s study were not consistent with those of Doner et a l . , although they did occur i n the same order. Doner et a l . , found tagatose to have an Rf of 0.425, fructose 0.406, and glucose 0.352. In t h i s study, tagatose was 0.40, fructose 0.36-0.37, and glucose 0.30. The deproteinized whey sample, assayed undiluted at 24.9% t o t a l s o l i d s , had a single strong Rf value of 0.18, s i m i l a r to the lactose standard. In Table 15, enzyme assays showed the deproteinized whey sample contained lev e l s of lactulose, glucose, and fructose below 0.7% (w/v); none were p o s i t i v e l y detected by t h i s TLC method. Galactose was not detected either, although shown 104 Standards: Samples; #1 2 3 4 5 6 7 8 #9 10 11 1 1 1 1 1 1 1 1 uL. uL. uL. uL. UL. uL. UL. uL. 0.15 uL. 0.15 uL. 0.15 uL. 20% glucose 0.05 uL 20% galactose 0.05 uL 20% f r u c t o s e 0.05 uL 20% l a c t o s e 0.05 uL 20% l a c t u l o s e 0.05 uL 20% tagatose 0.05 uL .8% a l l sugars 0.05 uL ,10% a l l sugars 0.05 uL no galactose D e c a l c i f i e d whey....0.1 (24.9% TS) P r e c i p i t a t e d 0.1 prepar a t i o n (10.4%) F i n a l preparation...0.1 (23.3% TS) uL -uL -uL -#12 13 14 15 16 17 18 19 #20 21 22 Figure 25 T h i n - l a y e r chromatography on s i l i c a p l a t e s shows standard sugar s o l u t i o n s at v a r y i n g concentrations and whey at d i f f e r e n t stages of process. 1 105 Table 22 Rf values of carbohydrate standards and various samples using thin-layer chromatography on a derivatized s i l i c a plate. Sample Sample Sample Description Rf 1 (0.10 uL) (0.05 uL) (0.15 uL) 1 12 glucose 0.28 0.32 2 13 galactose streaked from origin 3 14 fructose 0.36 0.37 4 15 lactose 0.18 0.18 5 16 lactulose 0.21 0.22 6 17 tagatose 0.40 0.40 7 18 a l l sugars 0.17-0.36 0.17-0.36 8 19 a l l less gal 0.17-0.22& 0.30-0.38 0.17-0.21& 0.30-0.38 20 9 deprot. whey 0.18 0.18 21 retorted 0.14-0.19 10 p u r i f i e d whey 0.12-0.19 22 decolourized 0.12-0.19 11 p u r i f i e d whey 0.06-0.13& 0.18-0.19 The mean measurement of the solvent front was 14.5 cm. 106 by the enzyme assay to be present i n the sample at 2.2% (w/v). Using t h i s TLC method, galactose i s d i f f i c u l t to i d e n t i f y because i t does not move up the solvent front, as described i n Section 3.5cl. The lactose preparation aft e r p r e c i p i t a t i o n (10.4% t o t a l solids) tested undiluted had overlapping spots from approximately 0.12 to 0.19 Rf value. The dominant carbohydrates according to the enzymatic assay (Table 15) were lactose and lactulose. In the TLC sample, there was no d i s t i n c t spot at the lactulose Rf of 0.21. Similar spotting occurred i n the decolourized lactulose preparation, at 23.3% t o t a l s o l i d s , with an Rf range of 0.12 to 0.19 for the lower volume sample. The low end of the spots did not correspond with an i d e n t i f i e d carbohydrate. Doner et a l . (1984), provided the Rf values for many sugars using t h i s TLC method. Few tested sugars possessed an Rf value lower than lactose, but included melibiose, gentiobiose, r a f f i n o s e , melezitose, maltotriose, and stachyose, none of which have been mentioned i n l i t e r a t u r e involving lactose reactions i n milk or whey. 4.4d Tagatose i d e n t i f i c a t i o n b y TLC and spectrophotometer No sample tested showed any spot formation near the tagatose standard's Rf value of 0.40. If tagatose was formed by the heat treatment, the amount was too low to be detected by t h i s method. A spectrophotometric method (Section3.5c2) was employed i n an e f f o r t to i d e n t i f y tagatose i n a process sample. The standard curve of tagatose using an absorbance of 256 nm, Figure 14, was not s i g n i f i c a n t i n i t s l i n e a r i t y . Dropping the three lowest tagatose concentrations (0, 5, and 10 mg/mL) produced a regression curve 107 which was s i g n i f i c a n t i n i t s l i n e a r i t y and slope. The three process samples tested, d e c a l c i f i e d whey permeate at 24.9% t o t a l s o l i d s , p u r i f i e d whey permeate at 10.4% t o t a l s o l i d s , and decolourized whey permeate at 23.3% t o t a l s o l i d s , d i d not reach 30 mg/mL even though highly concentrated. I t i s therefore s t i l l unknown whether tagatose was produced by the heat treatment and i n what amount. I t can be said that the amount produced, i f any, was l i k e l y to be under 3 0 mg/mL in these concentrated samples. This coincides with the findings of Berg and van Boekel (1994) who found no tagatose production i n heated milk using HPLC and Troyano et a l . (1992) who found only 13 mg/L using GLC. 4.4e Determination of t o t a l nitrogen and protein A micro-Kjeldahl method was used to determine t o t a l nitrogen present i n the whey permeate at d i f f e r e n t stages of the treatment process, (Section 3.5dl). Protein was determined using the Bio-Rad Protein assay described i n Section 3.5d2. The regression curve of the standard curve, Figure 15, was tested using ANOVA. Total nitrogen and protein, as l i s t e d i n Table 15, were found to be 0.61% and 0.48% (w/w of t o t a l solids) r e s p e c t i v e l y i n the deproteinized whey permeate Sample A. In Table 3 (Hargrove et a l . , 1976), whey from Cheddar cheese without deproteinization was described as having 1.9% (calculated as w/w of t o t a l solids) t o t a l nitrogen and 9.0% protein. After u l t r a f i l t r a t i o n , there was 0.4 6% t o t a l nitrogen and 0.18% protein. The UF whey permeate used i n the present study contained somewhat higher l e v e l s of nitrogen and protein than the example of Hargrove. In the d e c a l c i f i e d whey permeate Sample B, there was 0.61% 108 t o t a l nitrogen and 0.44% protein (w/w of t o t a l s o l i d s ) . The process of d e c a l c i f y i n g the whey permeate did not a l t e r the nitrogen or protein content. In the f i n a l product, decolourized p u r i f i e d heat treated lactulose preparation - Sample G, there was 0.17% nitrogen and 0.094% protein (w/w of t o t a l solids) . I t i s presumed that much of the t o t a l nitrogen and protein was adsorbed onto the ion-exchange and activated charcoal columns. 109 5 . CONCLUSIONS A high lactulose, mixed carbohydrate preparation was produced without the use of toxic catalysts. The f i n a l syrup contained 57% lactulose, 26% lactose, 5.0% galactose, 1.0% glucose, and 0.81% fructose, based on t o t a l s o l i d s . Using TLC, there was no tagatose detected, and there may have been an un i d e n t i f i e d carbohydrate. The f i n a l syrup was not as pure as commercial products, where lactulose content can be ten-fold greater than lactose. Manipulating conditions of thermal processing increased conversion of lactose to lactulose. Taguchi's f r a c t i o n a l f a c t o r i a l experimental design aided i n selection of conditions such that approximately 3 0% of i n i t i a l lactose was converted to free lactulose v i a the Lobry de Bruyn and Alberda van Ekenstein transformation. From eight i n i t i a l factors - pH, NaOH concentration, lactose concentration, p u r i f i c a t i o n , heating temperature, heating time, c i t r a t e concentration, and sodium c i t r a t e concentration - and two experimental designs, a set of conditions was chosen for the heat treatment of whey permeate. In the preferred method, d e c a l c i f i e d UF whey permeate at a pH of between 10.5 and 11.0 including 50 mM sodium c i t r a t e at a lactose concentration of >70 mg/mL was heat treated at 110° C for 10 minutes. P a r t i a l p u r i f i c a t i o n of lactulose was accomplished by concentrating a sugar solution s i m i l a r to heat treated whey permeate and cooling, p r e f e r e n t i a l l y p r e c i p i t a t i n g lactose over lactulose. After one cooling cycle, there was on average a lactulose y i e l d of 82% and a 1:1 lactulose: lactose r a t i o . Taguchi's f r a c t i o n a l f a c t o r i a l design was again used determining 110 that pH and sugar concentration each had a s i g n i f i c a n t e f f e c t on lactulose y i e l d and the lactulose:lactose r a t i o i n the p r e c i p i t a t i o n decant. The selected conditions for cold p r e c i p i t a t i o n based on these results was to maintain a pH of 10.5, concentrate to 50% lactose and lower the temperature by 5C°/hour from 65°C to 2 0°C, holding for 2 4 hours. A f t e r a second p r e c i p i t a t i o n of the decanted portion, there was a 78% lactulose y i e l d and a 3.4:1 lactulose:lactose r a t i o . There was a t o t a l loss of about 40% of the i n i t i a l lactulose through two p r e c i p i t a t i o n cycles. Ion-exchange columns removed the majority of the natural and added s a l t s from the lactulose preparations. Activated charcoal was not recommended as i t removed most of the brown colour of the permeate but also removed 30% of the t o t a l s o l i d s . This heat treatment process did not prove as e f f i c i e n t as current methods of lactose conversion to lactulose, such as using borate and triethylamine. However, the substrate, whey permeate, i s a waste-product and a disposal dilemma for producers. E f f i c i e n c y i n conversion may not be as c r u c i a l to a manufacturer as the cost of the process, adherence to l o c a l regulations, and customer acceptance. Conversion e f f i c i e n c y may also be overcome with a more e f f e c t i v e p u r i f i c a t i o n process to remove or hydrolyze excess lactose, thus improving the sugar r a t i o of the f i n a l syrup. I l l 6. ABBREVIATIONS ANOVA, analysis of variance 3-APTS, 3-aminopropyl triethoxysilane ATP, adenosine 5'-tetrahydrogen triphosphate CUT, come-up time ddH20, deionized d i s t i l l e d water GLC, gas-l i q u i d chromatography G6PD, glucose-6-phosphate-dehydrogenase HK, hexokinase HPLC, high-pressure l i q u i d chromatography LA, Lobry de Bruyn van Ekenstein MW, molecular weight NAD, nicotinamide-adenine dinucleotide NADP, nicotinamide-adenine dinucleotide phosphate ONPG, o-nitrophenyl-beta-D-galactopyranoside PI, phosphoglucose isomerase PSE, por t a l systemic encephalopathy UF, u l t r a f i l t r a t i o n UHT, ultra-high temperature 112 7 , REFERENCES Adachi, S. 1959. 15th Int. Dairy Congress, London 3: 1686-1691. Cited i n Andrews, G.R. 1986. Formation and occurrence of lactulose i n heated milk. J. Dairy Res. 53: 665-680. Adachi, S., 1969. Formation of a d i f f i c u l t l y soluble saccharide of lactulose with calcium hydroxide. Carb. Res. 9:242-246. Adachi, S., and Patton, S. 1961. Presence and s i g n i f i c a n c e of lactulose i n milk products - a review. J. Dairy S c i . 44: 1375-1393. Ames, J.M. 1992. The Maillard Reaction. In Biochemistry of food proteins. Hudson, B.J.(Ed). Els e v i e r Applied Science, London GB. p.99-153 In Berg, H.E., and van Boekel, M.A.J.A. 1994. Neth. Milk Dairy J. 48:157-175. Andrews, G.R. 1984. Distinguishing pasteurized, UHT and s t e r i l i z e d milks by lactulose content. J . Soc. Dairy Technol. 37: 92-95. Andrews, G.R. 1986. Formation and occurrence of lactulose i n heated milk. J. Dairy Res. 53(4): 665-680. Andrews, G.R. 1989. Lactulose i n heated milk. B u l l . Int. Dairy Fed. 238: 45-52. Andrews, G.R., and Prasad, S. 1987. E f f e c t of the protein, c i t r a t e , and phosphate content of milk on formation of lactulose during heat treatment. J. Dairy Res. 54(2): 207-218. AOAC. 1990. O f f i c i a l Methods of Analysis. 15th ed. Association of O f f i c i a l A n a l y t i c a l Chemists, Arlingthon, VA. 947.05 Avery, G.S., Davies, E.F., and Brogden, R.N. 1972. Lactulose: A review of i t s therapeutic and pharmacological properties with p a r t i c u l a r reference to ammonia metabolism and i t s mode of action i n portal systemic encephalopathy. Drugs 4: 7-48. Beach, R.C., and Menzies, I.S. 1983. Lactulose and other non-absorbable sugars i n infant milk feeds. Lancet 425-426. Berg, H.E., and van Boekel, M.A.J.A. 1994. Degradation of lactose during heating of milk. 1. Reaction pathways. Neth. Milk Dairy J. 48: 157-175. C a r u b e l l i , R. 1966. Transformation of disaccharides during borate ion-exchange chromatography, isomerization of lactose into lactulose. Carb. Res. 2:480. Concon, J.M., and Soltess, D. 1973. Rapid micro k j e l d a h l digestion of cereal grains and other b i o l o g i c a l materials. Anal. Biochem. 53: 35-41. 113 Corbett, W.M., and Kenner, J. 1953. The degradation of carbohydrates by a l k a l i . J. Chem. Soc. 2245-2247. Davidson, E.A. 19 67. Thioglycosides: Lobry de Bruyn transformation and isomerase reactions. Ch. 6.7 i n Carbohydrate Chemistry, p.185-194. Hott, Rinehart and Winston Inc., New York. Dobrogosz, W.J. 1981. Hydrolases (B-galactosidase). Ch 18.2.8 i n Manual of Methods for General Bacteriology, P. Gerhardt (Ed), p.375-376. Am. Soc. Microbiology, Washington, DC. Doner, L.W., Fogel, C.L., and B i l l e r , L.M. 1984. 3-Aminopropyl-bonded-phase s i l i c a , thin-layer chromatographic plates: t h e i r preparation, and application to sugar r e s o l u t i o n . Carb. Res. 125:1-11. Durance, T.C., and Cross, M. 1992. Production of high fructose dairy sweetener from cheese whey. Technical report to Canadian Lysozyme Inc., Abbotsford, B.C., Canada. Florence, E., Knight, D.J., Owen, J.A., Milner, D.F., and Harris, W.M. 1985. Nutrient content of l i q u i d milk as r e t a i l e d i n the UK. J Soc. Dairy Tech. 38:121-127. Geir, H. and Klostermeyer, H. 1983. Formation of lactulose during heat treatment of milk. Milch. 38(8): 475-477. Gould, I.A. 1945. Lacti c acid i n dairy products. The e f f e c t of heat on t o t a l acid and l a c t i c acid production and on lactose destruction. J. Dairy S c i . 28: 367-377. Haenel, H. 1970. Human normal and abnormal g a s t r o i n t e s t i n a l f l o r a . Am. J . C l i n . Nut. 23(11):1433-1439. Hargrove, R.E., McDonough, F.E., LaCroix, D.E., and Alfo r d , J.A. 1976. Production and properties of deproteinized whey powders. J. Dairy S c i . 59(1): 25-33. Hendrickse, R.G., Woodridge, M.A., and Russell, A. 1977. Lactulose i n baby milks causing diarrhoea simulating lactose intolerance. B r i t . Med. J. 1(6070): 1194-1195. Hicks, K. and Parrish, H. 1980. A new method fo r the preparation of lactulose from lactose. Carb. Res. 82: 393-397. Hicks, K., Raupp, D., and Smith, P. 1984. Preparation and p u r i f i c a t i o n of lactulose from sweet cheese whey u l t r a f i l t r a t e . J. Agr. Food Chem. 32(2): 288-292. Jef f r e y , G.A., Wood, R.A., Pfeffer, P.E., and Hicks, K.B. 1983. J. Am. Chem. Soc. 105: 2128-2133. Jenness, R. , and Patton S. 1959. Milk s a l t s . Ch 5 i n P r i n c i p l e s of Dairy Chemistry, p. 158-176. John Wiley & Sons, New York, NY. 114 Jiminez-Perez, S. , Corzo, N. , Morales, F.J., Delgado, T. , arid Olano, A. 1992. Effe c t of storage temperature on lactulose and 5-hydroxymethyl-furfural formation i n UHT milk. J. Food Prot. 55(4):304-306. Keeney, D.G., Patton, S., and Josephson, D.V. 1950. Studies of heated milk. Acetol and related compounds. J. Dairy S c i . 33: 526-530. Kosman, M.E., 1976. Lactulose (cephulac) i n portosystemic encephalopathy. J. Am. Med. Assoc. 236(21): 2444-2445. Martinez-Castro, I., and Olano, A. 1980. Influence of thermal processing on carbohydrate composition of milk: formation of epilactose. Milch. 35: 5-8. Martinez-Castro, I., Olano, A., and Corzo, N. 1986. Modifications and interactions of lactose with mineral components of milk during heating process. Food Chem. 21: 211-221. Mata, L.J., C a r r i l l o , C. , and V i l l a t o r o , E. 1969a. Fecal microflora i n healthy persons i n a pr e i n d u s t r i a l region. App. Micro. 17(4): 596-602. Mata, L.J., Urrutia, J.J., Garcia, B., Fernandez, R., and Behar, M. 1969b. Shigella i n f e c t i o n i n breast-fed Guatemalan Indian neonates. Amer. J. Dis. Child 117: 142-146. Mendicino, J.F. 1960. Effe c t of borate on the a l k a l i - c a t a l y z e d isomerization of sugars. J. Am. Chem. Soc. 82:4975-4979. Mizota, T., Tamura, Y., Tomita, T., andOkonagi, S. 1987. Lactulose as a sugar with physiological s i g n i f i c a n c e . B u l l . IDF. 212: 69-76. Montgomery, E. 1962. Alpha-lactulose: rearrangement of lactose to lactulose i n alkaline solution. Methods Carb. Chem. 1: 325-331. Montgomery, E.M., and Hudson, C.S. 193 0. J.Am. Chem. Soc. 52: 2101. Cited in Mizota, T., Tamura, Y., Tomita, T., and Okonagi, S. 1987. Lactulose as a sugar with p h y s i o l o g i c a l s i g n i f i c a n c e . B u l l . IDF. 212:69-76. Nakai, S., Arteaga, G.E., and Dou, J. 1994. PTaguchi BAS Food Science 401 Laboratory computer program, Dept. Food Science, University of B r i t i s h Columbia. Okumura, H. , and Gomi, T. 1973. Lab C l i n . 7: 3517. Cited i n Mizota, T. and Tamura, Y. 1987. Lactulose as a sugar with phy s i o l o g i c a l significance. B u l l . IDF. 212: 69-76. Olano, A., Calvo, M. and Corzo, N. 1989. Changes i n the carbohydrate f r a c t i o n of milk during heating processes. Food Chem. 31(4): 259-265. 115 Olano, A. and Martinez-Castro, I. 1981. Formation of lactulose and epilactose from lactose in basic media. A quantitative study. Milch. 36(9): 533-536. Oosten, B.J. 1967. Rec. Trav. Chim. Pays-Bas. 86: 673. Cited i n Mizota, T., Tamura, Y., Tomita, M., & Okonogi, S. 1987. Lactulose as a sugar with physiological s i g n i f i c a n c e . B u l l . Int. Dairy Fed. 212:69-76. Parrish, F.W. 1970. US Patent 3,514,327 i n Hicks, K.B., Raupp, D., and Smith, P. 1984. J. Agr. Food Chem. 32(2): 288-292. Parrish, F.W., Hicks, K. , and Doner, L. 1980. Analysis of lactulose preparations by spectrophotometric and high performance l i q u i d chromatographic methods. J. Dairy S c i . 63:1809-1814. Petuely, F. 1957. Z. Kinderheilkunde 79: 174. Cited i n Andrews, G.R. 1986. Formation and occurrence of lactulose i n heated milk. J. Dairy Res. 53: 665-680. Rose, C.S., and Gyorgy, P. 1955. Observations on b a c t e r i a l i n t e r a c t i o n . Proceedings of the Society for Experimental Biology and Medicine. 89:23-26. Short, J.L. 1978. Prospects for the u t i l i z a t i o n of deproteinized whey i n New Zealand - a review. New Zealand J. Dairy S c i . Tech. 13: 181-194. Sienkiewicz, T., and Riedel, C L . 1990a. Whey types and t h e i r composition. Ch. 3 i n Whey and whey u t i l i z a t i o n , p. 26-43. Sienkiewicz, T., and Riedel, C L . 1990b. U t i l i z a t i o n of whey. Ch. 4 i n Whey and whey u t i l i z a t i o n , p. 44-281. Verlag Th. Mann, Gelsenkirchen-Buer, Germany. Sienkiewicz, T., and Riedel, C L . 1990c. Other p o s s i b i l i t i e s for the u t i l i z a t i o n of whey and the outlook. Ch. 6 i n Whey and whey u t i l i z a t i o n , p. 320-332. Verlag Th. Mann, Gelsenkirchen-Buer, Germany. Taguchi, G. 1987. Linear graphs and t h e i r applications. Ch. 7 i n System of experimental design: Engineering methods to optimize guality and minimize costs. Vol. one. D. Clausing (Ed.), p. 150-207. Kraus Int. Pub. White Plains, N.Y. Troyano, E., Olano, A., Sanz, M.L., and Martinez-Castro, I. 1992. Is o l a t i o n and characterization of 3-deoxypentulose and i t s determination in heated milk. J. Dairy Res. 59: 507-515. Vaheri, M. and Kauppinen, K. 1978. The formation of lactulose by b-galactosidase. Acta Pharm. Fenn. 87: 75-82. 116 Visser, R.A. 1982. Growth of non-ionic lactose at various temperatures and super-saturations. Neth. Milk Dairy J. 36: 167-193. Visser, R. and van den Bos, M. 1988. Lactose and i t s chemical derivatives. B u l l . IDF. 233: 33-44. Zadow, J.G. 1984. Lactose: properties and uses. J . Dairy S c i . 67: 2654-2679. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099196/manifest

Comment

Related Items