Louisiana State University LSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1988 The exD tranase of Lipomyces Starkeyi and Its Use in Sugar Cane Processing. David William Koenig Louisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

Recommended Citation Koenig, David William, "The exD tranase of Lipomyces Starkeyi and Its Use in Sugar Cane Processing." (1988). LSU Historical Dissertations and Theses. 4576. https://digitalcommons.lsu.edu/gradschool_disstheses/4576

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS The most advanced technology has been used to photo­ graph and reproduce this manuscript from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are re­ produced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. These are also available as one exposure on a standard 35mm slide or as a 17" x 23" black and white photographic print for an additional charge. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

University Microfilms International A Bell & Howell Information Company 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 313/761-4700 800/521-0600

Order Number 8904545

The dextranase of Lipomyces starkeyi and its use in sugar cane processing

Koenig, David William, Ph.D.

The Louisiana State University and Agricultural and Mechanical Col., 1988

UMI 300 N. ZeebRd. Ann Arbor, MI 48106

The Dextranase of Lipomyces starkeyi and Its Use in Sugar Cane Processing

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy

in

The Department of Microbiology

by David William Koenig B.S., Bowling Green State University, 1979 M.S., University of Mississippi, 1982 August 1988 ACKNOWLEDGEMENTS

I would, like to thank Dr. F. Paul (BioEurope,

Ltd. , Toulouse, France) for the generous gift of the

isomaltooligosaccharides; Dr. Ramu Rao (Dept. Biochemistry,

LSU) for performing the amino acid analysis; Dr. John Larkin

(Dept. Microbiology, LSU) for purchasing the HPLC column;

Dr. Martin Hjortso (Dept. Chemical Engineering, LSU) for use of his HPLC; Durriya Sakar (Audubon Sugar Inst., LSU) for performing the dextran analyses on the sugar samples; Mickey

Brou and Ronnie Champton (Audubon Sugar Inst., LSU) for their technical help setting up and running the 500 liter fermentor; and Joe Roos for his discussions. Appreciation is extended to The American Sugar Cane League and Audubon Sugar

Institute for providing the funds and facilities to complete this research. I would also like to thank Dr. Donal

F. Day whose support, both moral and intellectual, was greatly appreciated. Finally I dedicate this work to my wife

Tracey. Her patience and understanding made this work possible.

ii TABLE OF CONTENTS

Page

ACKNOWLEDGEMENT i i

TABLE OF CONTENTS iii

LIST OF TABLES vii

LIST OF FIGURES viii

ABSTRACT x

INTRODUCTION 1

REVIEW OF LITERATURE 3

I. Dextran 3

II. Microbiology of Dextran Formation 4

III. General Characteristics of Dextranase 5

IV. Microbial Sources of Dextranase 6

V. Characteristics of the Penicillium Dextranase. 7

VI. Characteristics of the Chaetomiuro Dextranase.. 9

VII. Characteristics of the Lipomyces Dextranase... 11

VIII. Production and Regulation of Dextranase 12

IX. Description and Physiology of Lipomyces 15

X. Application of Dextranase in Sugar Processing. 18

XI. Application of Dextranase in Dental Hygiene... 22

METHODS AND MATERIALS 23

I. Cultures 23

II. Growth Medium and Culture Maintenance 23

III. Optimization of Fermentation Parameters 24

IV. Induction of Derepressed Mutant (ATCC 20825)... 25

iii TABLE OF CONTENTS (cont'd.)

Page

V. Scale-Up 26

VI. Enzyme and Protein Assay 26

VII. Enzyme Purification 27

VIII. Molecular Weight Determination by Gel Filtration 28

IX. Polyacrylamide Slab Electrophoresis 28

X. Isoelectric Focusing 31

XI. Determination of Carbohydrate Content 31

XII. Lectin Reactivity 32

XIII. Amino Acid Composition 32

XIV. Dextranase: pH Optimum and Stability 32

XV. Dextranase: Temperature Optimum and Stability.. 33

XVI. Determination of Michaelis-Menton Constants.... 34

XVII. Inhibitory Effect of Carbohydrates 34

XVIII. Effect of Salts and Inhibitors on Activity 34

XIX. Protease Inhibition 34

XX. Analysis of Reaction Products 35

XXI. Application of Dextranase in Sugar Processing.. 36

A. Laboratory tests 36

B. Pilot trial 36

C. Dextran analysis 37

RESULTS 38

I. Optimization of Fermentation Parameters 38

A. Effect of pH 38

iv TABLE OF CONTENTS (cont'd. )

Page

B. Effect of aeration 40

C. Effect of temperature 40

II. Dextranase Production by a Derepressed Mutant.. 41

A. Strain comparisons 41

B. Determination of inducer 42

C. Effect of exposure time to inducer 43

D. Saturation level of inducer 43

E. Effect of cell age on induction potential... 44

III. Scale-Up 53

IV. Enzymology 54

A. Purification of dextranase 54

B. Molecular weight determinations 66

C. Isoelectric focusing 66

D. Kinetic comparisons... 71

E. Carbohydrate and amino acid composition 72

F. Effect of pH and temperature 72

G. Reaction products 82

H. Effect of proteases on activity 82

I. Effect of salts on activity 85

J. Effect of enzyme inhibitors on activity 85

K. Reactivation of mercuric chloride treated enzyme 87

L. Effect of carbohydrates on activity 88

M. Effect of solvents on activity 88

v TABLE OF CONTENTS (cont'd)

Page

V. Application of Dextranase to Sugar Processing.. 93

A. Laboratory trials 93

B. Pilot trial 98

DISCUSSION 102

CONCLUSION 118

LITERATURE CITED 120

VITA 129

vi LIST OF TABLES

Page

Table 1. Effect of Medium pH on Lag Period, Growth Rate, and Dextranase Production 39

Table 2. pH Stability of Crude Versus Purified Dextranase 39

Table 3. Effect of Aeration on Growth Rate and Dextranase Yield 40

Table 4. Effect of Incubation Temperature on Growth Rate and Dextranase Yield 41

Table 5. Comparison of Wild Type (ATCC 12659) and the Derepressed Mutant (ATCC 20825) 42

Table 6. Compounds that Induce Dextranase Production 43

Table 7. Scale-Up 53

Table 8. Purification of Lipomyces starkeyi Dextranase 54

Table 9. Apparent Km Values for Isoelectric Forms.... 71

Table 10. Apparent Km Values for Native, Aggregate, and IGC 4047 Dextranase 71

Table 11. Amino Acid Compositions of Dextranase 73

Table 12. End Product Analysis 82

Table 13. Effect of Exposure to Proteases on Dextranase Activity 85

Table 14. Effect of Salts on Dextranase Activity 86

Table 15. Effect of Enzyme Inhibitors on Dextranase Activity 87

Table 16. Removal of Dextran from Mixed Juice: Effect of Incubation Temperature 93

Table 17. Removal of Dextran: Mill Trial Results 98

vii LIST OF FIGURES

Page Figure 1. Time Course of Dextranase Induction 45

Figure 2. Effect of Inducer Concentration on the Amount of Dextranase Produced 47

Figure 3. Effect of Cell Age on Induction Potential.. 49

Figure 4. Enzyme Activity Versus Time of Induction and Cell Concentration 51

Figure 5. Gel Filtration Profile for Crude Dextranase 56

Figure 6. Gel Filtration Profile for CM-Sepharose Purified Dextranase 58

Figure 7. Gel Filtration Profile for CM-Sepharose Purified then Dialyzed Dextranase 60

Figure 8. Gel Filtration Profile for A-0.5m Purified Dextranase 62

Figure 9. Gel Filtration Profile for Isopropanol Purified Dextranase 64

Figure 10. Densitometer Scan of 7.6 % SDS-PAGE Fractionation of Purified Dextranase 67

Figure 11. Densitometer Scan of Isoelectric Focusing Gel Fractionation of Purified Dextranase... 69

Figure 12. Effect of pH on Dextranase Activity 74

Figure 13. Effect of pH on Dextranase Stability 76

Figure 14. Effect of Temperature on Dextranase Activity 78

Figure 15. Effect of Temperature on Dextranase Stability 80

Figure 16. Time Course of Reaction Product Formation.. 83

Figure 17. Effect of Solvents on Dextranase Activity.. 89

Figure 18. Effect of Solvents on Dextranase Activity.. 91

viii LIST OF FIGURES (cont'd.)

Page

Figure 19. Effect of Enzyme Concentration Upon Dextran Removal From Cane Juice 94

Figure 20. A Time Course of Dextran Removal From Cane Juice 96

Figure 21. Comparison of Treated Versus Untreated Sugar 100

ix ABSTRACT

Dextranase is an enzyme which destroys dextran, a contaminant in sugar and a main constituent of dental plaque. Dextranase utilazation can solve a major economic problem in the sugar industry, and in dental preparations it can keep teeth essentially free of plaque. The dextranase from the yeast Lipomyces starkeyi was investigated because this organism has been used in food related processes. This attribute should allow for Food and Drug Administration

(FDA) approval for use of a Lipomyces dextranase. None of the existing commercial dextranase preparations have FDA approval. Lipomyces was found to produce maximum amounts of dextranase when grown in a minimal medium at pH 3.0. A derepressed mutant was selected and the optimum conditions for induction of dextranase production were determined.

B-Methyl-glucopyranoside was found to be a gratuitous inducer of dextranase synthesis. A fermentation system was scaled up to pilot scale and the enzyme was purified and characterized with respect to both its physical and kinetic parameters. The dextranase produced by Lipomyces was found to be catalytically similar to the dextranases from

Penicillium and Chaetomium. Dextranase from Lipomyces was used to treat stale cane juice. A 76 % reduction in dextran content, by the Haze method, was seen in the final sugar product.

x INTRODUCTION

Dextran is a polymer of glucose commonly found in stale

sugar cane. When dextran is present in cane juice, it can

increase syrup viscosities making crystallization difficult.

Some dextran is incorporated into the sugar crystals,

affecting both their shape and marketability. Sugar

refiners, who are the customers for raw sugar, levy a price

penalty on lextran-containing raw sugar. It is therefore

critical for the U. S. sugar industry that methods are made

available to remove dextran from raw sugar products.

Dextranase (EC 3.2.1.11) is an enzyme which selectively

hydrolyses dextran into small carbohydrate molecules.

Dextranase offers the possibility of a practical treatment

for dextran in the sugar industry. Application of dextranase in sugar cane processing is used overseas

(23,25,31,32,33,84).

Dextran is also a component of dental plaque

(20,21,43). The Japanese have shown (1,37,68,69,85) that inclusion of dextranase in toothpaste can greatly reduce or eliminate dental plaque. Dextranase-containing toothpastes are commercially available in Japan.

Since both sugar processing and toothpaste applications for dextranase have been demonstrated abroad, the question may be asked, why is it not used in the U.S.? The principle reason is that the current commercial dextranase

1 2 preparations have not obtained United States Food and Drug

Administration (FDA) approval. The commercial sources of dextranases are species of the fungi Penicillium (57) or

Chaetomium (51). These fungi produce various antibiotics and toxic metabolites as well as dextranase, hence the difficulty in obtaining FDA approval.

Lipomyces starkeyi, an ascosporogenous yeast, also produces dextranase (92). This yeast has been used in food related application (13,24,27,54,61,74) and is not known to produce antibiotics or toxic metabolites (60). These factors enhance the chance for FDA approval for a Lipomyces dextranase. Unfortunately, little was known about the production of dextranase by this yeast.

The objectives of this investigation were four fold:

(a) to optimize dextranase production in Lipomyces, (b) to select and characterize a strain of Lipomyces which would produce a dextranase when grown on a carbon source other than dextran and then scale-up production of this enzyme,

(c) purify and characterize the dextranase produced by this mutant Lipomyces, and (d) demonstrate the effectiveness of this dextranase in a pilot sugar mill trial. REVIEW OF LITERATURE

I. Dextran

Dextran is a biopolymer composed of primarily a 1-6 linked glucose monomers. This compound can have a 1-3 and a 1-4 linked glucan branches (84). The more a 1-3 linkages in the polymer, the less soluble dextran is in water. The molecular weight of dextran can range from 5 X 103 to

4 X 10"7. The dextran which is a problem in sugar processing is mainly the soluble type (high a 1-6). Dextran causes an increase in viscosities of sugar process streams and the deformation of the final products. These and associated problems cause large economic loss to the U. S. sugar industry.

Dextran which is associated with dental plaque is mainly of the insoluble type (high a 1-3) and is often termed mutan. The semantic difference between mutan and dextran is that mutan is made by Streptococcus mutans and is highly branched and insoluble, while dextran is not. The problem with this definition is that other microbes can produce highly branched, insoluble dextrans that are physically no different than mutan, but are called dextran in the literature.

3 4 II. Microbiology of Dextran Formation

The dextran in sugar processing is a result of the

growth of Leuconostoc mesenteroides on sucrose (84).

Leuconostoc metabolizes sucrose using the extracellular

enzyme dextransucrase (EC 2.4.1.5). Dextransucrase

hydrolyses sucrose and concurrently transglycosylates the

glucose into a polymer chain. The bacterium uses fructose as

a carbon source. Dextransucrase, concurrently with sucrose

hydrolysis, polymerizes the glucose monomers into dextran.

Dextransucrase is an inducible enzyme that is only produced when the organism is grown on sucrose (56). This enzyme is

not thought to be subject to catabolite repression (56).

Streptococcus mutans uses a similar enzyme system to produce the dextran associated with dental plaque. The glycosyl transferase complex (GTF) is the enzyme complex associated with polysaccharide formation in the oral streptococci (43,59). Clinical and biochemical studies have thus far shown that water-insoluble glucans produced by oral streptococci contribute to the induction of dental carries and plaque formation (43). Almost all streptococci produce dextranases concurrently with the production of GTF

(43,76,77) and there are several lines of evidence implicating the involvement of dextranases in the process of dextran formation.

GTF requires small isomaltooligosaccharides for the initiation of glucan synthesis. Dextranases can produce 5 these primer saccharides (6,20,21,43). Dextranase has also been shown to degrade partially the water-insoluble dextran associated with plaque (3,6,20), limit the production of water-insoluble glucan (6), and therefore inhibit the adherence of oral streptococci to dental enamel. The true role for dextranase in the oral streptococci is unclear.

III. General Characteristics of Dextranase

Dextranase (a-1,6-glucan hydrolase, EC 3.2.1.11) is an enzyme which hydrolyses the a 1-6 linkages found in dextran

(66). This enzyme does not cleave the a 1-4 or a 1-3 linkages. The enzyme can be classified into two catalytic types; exo-acting or endo-acting. Both types exhibit pseudo-first-order kinetics. Some dextranases have the ability to carry out transglycosylation reactions (34).

An exo-acting dextranase will preferentially attack the dextran polymer at the non-reducing ends, moving down the chain liberating glucose and isomaltose. The hydrolytic action of the exo-enzymes are terminated when they reach an a 1-3 or an a 1-4 branch point. These enzymes usually stop one glucose residue from the branch point (91). This enzyme will produce a large limit dextran. Because of this the exo-acting enzyme has limited commercial importance.

Exodextranases show the same apparent Km for dextran regardless of the size of the molecule (64).

Endodextranases act by randomly binding to the 6 substrate attacking the dextran molecules at internal a 1-6 linkages. The a 1-6 linkages found in the vicinity of the branches are recalcitrant to attack by endodextranases

(66,91). The products of dextranase action are branched oligosaccharides and oligosaccharides of the isomaltose series (66,91). The apparent Km values are inversely proportional to the molecular weights of the dextran used as substrate where the affinities for a high molecular weight dextrans are much larger than those for small molecular weight dextrans.

IV. Microbial Sources of Dextranase

Dextranases are produced by a variety of microbes.

Bacterial sources include: Pseudomonas (66), Flavobacterium

(81), Streptococcus (6,76), Arthrobacter (71),Actinomadura

(71), Cytophaga (34), and Bifidobacterium (34) species. Most of the bacterial dextranases are of the exo-type (35). The interest in dextranase from bacterial sources is due to their involvement in the formation of dental plaque.

Fungi of the genera Penicillium (9), Chaetomium (29),

Aspergillus (30), Fusarium (23), and Lipomyces (64,92) are known to produce dextranases. Most of the dextranases produced by the fungi are endodextranases, although

Lipomyces lipofer has been reported to produce an exodextranase (64). The major impetus for studying dextranase enzymes from fungi has been the commercial 7 application of these dextranases to food processing

(4,25,41,31,45,51,57,72) and to dental hygiene (37,68,69).

Commercially available dextranases are from Penicillium,

(Novo; DN 25L) (57) and Chaetomium (Miles; DEXTRANEX) (51).

The enzymes from both these fungal sources have been fully characterized with respect to their physical and enzymological properties. The Chaetomium enzyme has also been characterized as far as its usefulness in plaque reduction when combined with dental preparations (68,69,85).

V. Characteristics of the Penicillium Dextranase

A Penicillium (NRRL 1768) dextranase was purified by a method which included acetone partition, ammonium sulfate fractionation, gel filtration, iron precipitation and decantation, and diethylaminoethyl (DEAE)-cellulose chromatography (9). This procedure gave a 1,000-fold increase in specific activity with a direct yield of 2.6 %, and produced a preparation that was shown to be homogenous by gel electrophoresis and electrofocusing (9). The final specific activity was 27,000 IU/mg solids (9). Gel filtration chromatography gave a molecular weight of 41,000 for the purified enzyme (9).

The Penicillium enzyme was stable between pH 5 and 7, and had a activity optimum at pH 5.0. This enzyme had a temperature optimum for activity between 50 and 60 °C, but rapidly inactivated above 55 °C (9). This enzyme was very 8 stable in the presence of solvents (30 % methanol,

30 % acetone), exhibiting a 9 % loss of enzyme activity

after 24 hours at 4 °C. Aqueous solutions of pure dextranase

at concentrations on the order of 12 mg/ml did not exhibit

any loss of activity after storage for one year at 4 °C (9).

Lyophilized dextranase was not completely soluble in water

upon reconstitution, exhibiting a 30 % loss of enzyme activity (9).

Two isoelectric forms of the enzyme were found, one at

pH 4.1 and the other at pH 4.7 (9). The enzyme produced by

Penicillium is a glycoprotein (14), which may explain the

differences in isoelectric points. An analysis of the

attached carbohydrate showed that it was composed of

residues of mannose, galactose, and fucose (14). The amino

acid analysis of the purified enzyme is shown in the results section, Table 11.

The Penicillium enzyme was inactivated by mercuric salts, but addition of 5 % 2-mercaptoethanol restored activity. This is reasonable evidence that at least one sulfhydryl group is present and essential for enzyme activity (9).

The application data for the Penicillium dextranase comes from Novo Laboratories for the product DN 25L (57).

The DN 25L enzyme has a temperature optimum between 50 to

60 °C and a pH optimum of 5. The enzyme inactivates rapidly above 60 °C (activity measured after 30 min at pH 5.0) and 9 is not stable below pH 5.0 (30 min incubation at 50 °C)

(57). These characteristics are identical to those of the purified Penicillium (NRRL 1768) dextranase (9). The inclusion of twenty percent sucrose in the assay mixture did not change the reported reaction characteristics of DN 25L.

NOVO recommends the addition of this enzyme at the raw juice for removal of contaminating dextrans in sugar processing (57).

VI. Characteristics of the Chaetomium Dextranase

The dextranase produced by Chaetomium gracile

(SANK 18067) was purified by sequential chromatography on carboxymethyl (CM)- and then DEAE-cellulose columns. Two active fractions were isolated and shown to be electrophoretically pure (29). Their molecular weights were estimated by ultracentrifugation as 77,000 and 71,000. The isoelectric points were found to be 6.2 and 5.7, respectively (29). Both enzyme forms showed similar response to pH and temperature. The hydrolysis curves for dextran and the catalytic end products produced by the two forms were identical. The amino acid composition was also similar.

Both Chaetomium enzymes were glycoproteins, containing approximately 4.5 % sugar (29). This may explain the physical differences between the two forms. Sulfur containing amino acids were few, as was true with the

Penicillium enzyme. A summary of the amino acid composition 10 for the combined form is given in the results section,

Table 11. Both enzymes were found to interact with polyacrylamide, therefore an accurate estimation of molecular weight by polyacrylamide gel filtration was not available (29).

Summarizing the enzymatic properties of this dextranase, the pH optimum was at 5.5, the enzyme was stable between pH 5.5 and 11.0 and at temperatures lower than 55 °C. The temperature optimum for activity was between

55 and 65 °C (29). Mercuric, ferric, and cupric salts as well as sodium dodecyl sulfate (SDS) inhibited the enzyme activity. MgCl2, CaCl2, CoCl2, MnCl2, ZnCl2, NiCl2 and NaF had no effect upon activity (29).

The application data for the Chaetomium dextranase comes from Miles Laboratories and is for their product

DEXTRANEX. The enzymatic properties of DEXTRANEX are similar to those presented for the purified enzyme. In 20

Brix (Brix (Bx) = % solids) sugar juices the enzyme behaved the same as the purified enzyme. If the Brix was increased to 65 (Brix in syrup), the enzyme exhibited increased thermal stability (85 °C maximum). However, the rate of dextran hydrolysis decreased with increasing Brix. Miles therefore recommended the treatment of raw juice as the primary application point, although the company states that this enzyme can be used on syrups after the evaporators if the loss of activity due to Brix is factored into the dose 11 rate (51).

VII. Characteristics of the Lipomyces Dextranase

Yeast of the genus Lipomyces also produce dextranases.

A comprehensive survey of the physiology of Lipomyces starkeyi is given in the literature review section IX.

Interestingly, L. starkeyi produces an endodextranase (92) while L. lipofer produces a exodextranase (64).

The L. lipofer (IGC 4042) dextranase was purified by ultrafiltration and agarose gel filtration (64).

Purification by this method produced an enzyme with a specific activity of 50 IU/mg protein which corresponded to a 8 fold purification factor with a 25 % overall yield (64).

This preparation was considered pure since it gave a single band upon disc gel electrophoresis. Glucose was the only reaction product produced by this enzyme. The enzyme molecular weight was 29,000 by gel filtration and showed an isoelectric point of pH 7. The apparent Km (45 °C, pH 5) for dextran T-40 was 1.2 X 10~5 M. Glucose competitively inhibited the enzyme. The enzyme was stable over a pH range of 4.5 to 6.0, and after 2 hours at 50 °C 60 % of original activity remained. Optimum pH and temperature for activity of this enzyme were pH 4.5-5.0 and 45 °C, respectively. It was not determined if this enzyme was a glycoprotein.

The L. starkeyi (IGC 4047) dextranase was purified to homogeneity by a single isopropanol precipitation step. This 12 procedure produced a five fold increase in specific activity with a total recovery of 69 % (92). The purity was confirmed by gel electrophoresis. The enzyme had the following properties: molecular weight, 23,000; optimum temperature and pH for activity around 50 °C and pH 5.0, respectively; pH stability, pH 3.5 - 7.5. This dextranase exhibited a 30 % reduction of activity after 2 hours incubation at 50 °C

(pH 5.0). The isoelectric point of this protein was determined to be pH 5.4. The final products of dextran hydrolysis were glucose and isomaltooligosaccharides from isomaltose through isomaltohexaose, with isomaltose and isomaltotriose in the majority. The apparent Km values reported by the authors are typical for an endodextranase and are listed in results section Table 9. Webb and

Spencer-Martins (92) stated that there appeared to be product inhibition, but did not give any supporting data. It was not determined if the enzyme was a glycoprotein.

VIII. Production and Regulation of Dextranase

Dextranases in fungi are extracellular enzymes, which in most cases are under catabolite repression control and require inducers to trigger production from a derepressed state. Little has been reported concerning the induction and regulation of dextranases in fungi. The understanding of protein secretion and the regulation of catabolite repression in fungi is in its infancy as compared to the 13 prokaryotic systems.

The prokaryotic system of regulation is typified by the Lac operon. Briefly, it was found in E. coli that glucose produces a decrease in the levels of cAMP, and that cAMP is necessary for the transcription of genes sensitive to catabolite repression (50,58). Therefore in the presence of glucose, transcription of these genes is impaired and the enzymes coded by them are not synthesized.

Evidence has accumulated that cAMP does not play the same role in catabolite repression in yeast that it does in

1* c°li (26). Unfortunately, vital information is lacking to explain the exact role of cAMP and other metabolites in catabolite repression of yeast (26). Most of the descriptions of fungal catabolite repression involve the characterization of derepressed mutants.

Before taking a more detailed look at the area of catabolite repression, some consideration must be given to nomenclature. Repression is the state in which addition of a specific carbon source, for example glucose, to a culture inhibits the synthesis of certain proteins. When glucose is exhausted, the synthesis of those proteins may start. This process is called derepression. However some proteins also require a specific inducer for their synthesis. These proteins are therefore both repressible and inducible.

Dextranase in Lipomyces is an example of a repressible-inducible system (38). Repression by glucose 14 and inducibility are controlled by different mechanisms

(10,26,58). It is therefore possible for an enzyme to be

constitutive (synthesis independent of inducer) while

remaining sensitive to glucose repression.

A derepressed mutant is normally selected with a

gratuitous repressor. The most common method is the

selection of mutants on 2-deoxyglucose (3,42,49,89,94). This

compound is an analog of glucose and most yeast are unable

to utilize this as a carbon source. When 2-deoxyglucose is

combined with a second carbon source, such as starch, the yeast must overcome the glucose effect to producfe the

normaly repressed amylase to survive. Derepression by

2-deoxyglucose selection affects hexokinase activity

(42,46,48). At least in S. cerevisiae the loss of hexokinase P-II is necessary to overcome glucose repression

(18). P-II appears to have a dual role, acting both as a

phosphorylating enzyme (48) and as a regulatory protein

(48). A mutant yeast has been described where the P-II has been isolated having only phosphorylating activity and no regulatory role (48).

A preliminary model was proposed by Gancedo and

Gancedo (26) to describe catabolite repression in yeasts.

The expression of a gene is controlled by two types of cis acting sequences, silencers or enhancers. Different sets of genes acting in trans determine the activity of the enhancers and silencers. These genes may act 15 pleiotropically, or specifically. The expression of the genes acting in trans and the interaction of their products with genes acting in cis may in turn be controlled by metabolites whose levels vary between repressed and derepressed conditions. The metabolites which may be involved are thought to be glucose 6-phosphate, fructose 6-phosphate, fructose 1, 6-bisphosphate, and/or cAMP (26).

There are reports of compounds which enhance the production of dextranase in Penicillium. It has been found that the hydrolysis products of dextran will act as better inducers for dextranases than unhydrolyzed dextran (65).

Also it has been shown that oxidized dextrans will increase dextranase yields (65). 3-0-Methyl-D-glucose has been shown to increase both the production of dextranase and the growth rate of Penicillium (7,8). This was attributed to an increased uptake of amino acids, allowing for an increase in protein synthesis. Glucose inhibited the uptake of some amino acids into yeast cells (8), therefore it was postulated that the derepression caused by a gratuitous repressor also affected amino acid uptake.

IX. Description and Physiology of Lipomyces

Lipomyces starkeyi is a strictly aerobic soil yeast discovered by Starkey (79) in 1946. This ascosporogenous yeast exhibits multilateral budding and will produce 4 to 8 16 asci per ascus. The cells are usually ellipsoid or globose and are surrounded by a thick polysaccharide capsule.

Cells in older cultures usually include a large fat globule.

The type strain, Starkey's strain 72, is synonymous to

ATCC 12827.

Fermentative ability is absent in this strain, but it will grow aerobically on galactose, maltose, raffinose, starch, citric acid, mannitol, and glucose. The type strain was not tested for growth on dextran. The strain reported to grow on dextran was IGC 4047 (Gulbenkian Institute of

Science Culture Collection, Lisbon, Portugal) (92).

Lipomyces uses the pentose phosphate pathway to assimilate carbohydrate sources.

The genus Lipomyces is well studied with regard to starch utilization (49,53,74,89), lipid production

(13,54,55), and single cell protein (28,49,89). When grown under optimum conditions Lipomyces can produce up to 45 % of its total cell mass as neutral lipids (13,27). This aspect has been pursued commercially for the production of coca butter substitutes (24) and fuels (13,27,61). The yeast has also been used as a fodder (28,61,74,89) and a food additive (13,28,74). Furthermore, it is not known to produce antibiotics or toxic metabolites (60), as do

Penicillium and Chaetomium. These attributes contribute to the commercial acceptability of products produced by

Lipomyces. 17 Only two reports in the literature were found

pertaining to endodextranase production in Lipomyces

(17,92). These did not define optimum conditions for

production nor give an accurate description of the enzyme.

The production of the dextranase enzyme in L. starkeyi is

under catabolite repression control and requires induction

after derepression (38). The enzyme is an extracellular product (92). Derepressed mutants with regard to dextranase production have been isolated and some show the ability to hyperproduce this enzyme (38).

There is a wealth of information on medium optimization with regard to lipid production in Lipomyces spp.

(13,27,54,55,86). The yeast uses both organic and inorganic nitrogen sources (2) and has a vitamin requirement for biotin when grown above pH 5.5 (87,88). When grown below pH

5.5 Lipomyces is able to grow independently of any externally provided vitamins (87,88).

The optimum pH for growth L. starkeyi on glucose was reported to be pH 4.8 (88). Most studies have been conducted on yeast grown between pH 4.5 and 6.0 (2,19,36,88,92). Under these conditions L. starkeyi showed a long lag period

(88,92). L. starkeyi is a mesophilic organism showing optimum growth rates between 25 °C and 30 °C (63). In L. kononenkoae growth temperature was shown to be critical in the optimization of amylase production, with the most enzyme produced near the upper growth temperatures (19,75). 18

Dissolved oxygen levels have been shown to be critical in the production of lipids in L. starkeyi (54,86).

X. Application of Dextranase in Sugar Processing

The Louisiana sugar processing season is usually from

October to January. During this period of time there can be wide climatic fluctuations that effect harvest practices.

In Louisiana, sugar cane is mechanically cut and laid in rows, after which the leaves burned from the stalk. The cane stalk is transported to the raw sugar mill.

The initial infection with Leuconostoc usually occurs during harvesting. However, dextran levels are time dependent. Severe rain will cause delays in cane retrieval, also severe wind conditions, such as those encountered during a hurricane, can not only damage the cane but also delay delivery. Leuconostoc infection may also occur when a freeze is followed by a thaw.

Once the cane arrives at the raw sugar mill it is normally processed immediately, although mills will stock pile cane during the day for use during night operation.

Dextran build up can occur during this period if the cane is allowed to sit for 48 hours or more (84), and the time period can be less if the cane has already been damaged.

Under normal processing, the cane is washed and then is crushed by a series of mills which liberate the cane juice. This juice usually has a pH of 5.0 - 6.0, 35 °C, and 19

15 brix (of which 80 - 90 % is sucrose). It is at this point

in the process that the Australian sugar industry recommends

the application of dextranase (33). The juice is then

clarified by heating following pH adjustment with lime. This

process stream is filtered and evaporated to make syrup. The

Moroccan sugar industry applies dextranase at this point

(Personal communication, Youssef Oubrahim, Audubon Sugar

Inst., LSU). Syrup is usually pH 6.0-6.5, 65 °C, and

65 Brix.

Sugar is crystallized from the concentrated syrup to

make raw sugar and molasses. The sugar crystals are

separated from the molasses by centrifugation. Dextran

contamination may cause an elongation of the sugar crystals

increasing losses during centrifugation and raising the

sugar content of the molasses.

Dextran is associated with a variety of problems such

as difficulty in clarification and filtration of juices and syrups, increased scaling of evaporators, slowing of the overall processes and crystal deformation. Dextran contamination is also a problem in the beet sugar industry causing problems analogous to those seen in the cane sugar

industry (4,72). Some dextran is always found in commercial sugar, raw or refined. This engenders a host of complaints from large customers of refined sugar. These and other

problems arising from dextrans in raw sugar are well documented (84,90). 20

Tilbury (84) in Jamaica in the late 1960's conducted the first comprehensive study of dextran formation and showed that dextranase provided a potential solution for this problem. Covacevich and Richards (12), showed that the a 1-6 dextran was the predominant polysaccharide in stale cane. Fulcher and Inkerman (25,32) made laboratory and plant trials, eventually concluding that dextranase added to mixed juice could effectively eliminate dextran from raw sugar.

Studies were done to test the various commercially available dextranases for their effectiveness in dextran removal during sugar production. The conclusion was that the

Miles DEXTRANEX (Chaetomium) product was the best commercial product (32). This was based on the ability of the enzyme to withstand higher processing temperatures and its ability to work efficiently at low dextran concentrations (51). The

Australian industry has used dextranase from Chaetomium for the last ten years.

The Moroccan sugar industry routinely treats dextran contaminated syrup (Personal communication, Youssef

Oubrahim, Audubon Sugar Inst., LSU). This mode of treatment has also been tried in the Australian sugar industry (32), although the Australian's prefer treatment at the mixed juice stage. The Moroccan industry uses the Novo enzyme

(Penicillium) and have had no problems with regard to temperature effects. 21

As mentioned earlier, the commercial use of dextranase in the U.S. has been prevented by lack of FDA approval.

Nonetheless, tests at Audubon Sugar Institute have confirmed the effectiveness of commercial (Penicillium) dextranase on Louisiana cane syrup. In 1979 Polack and

Birkett (62), treated stale cane during a pilot scale test.

Cane was allowed to age after a freeze and then processed in the Audubon mill in the hope of making sugar. However, the extremely high dextran content made the syrup so gummy that crystallization was difficult and the crystals produced were badly elongated (needle grain). The syrup was treated with dextranase, whereupon crystallization proceeded almost normally, producing square crystals.

There have been reports in the Indian sugar industry that dextranase was inactivated by metals (45). This work was done with Penicillium enzyme and the metals (copper and iron) involved were thought to have originated from the soil in which the cane was grown. There was some indication that contamination was due to poor mill upkeep, with the metals being derived from the processing equipment (Personal communication, Dr. M. Saska, Audubon Sugar Inst., LSU). In either case it may be a significate problem for application of dextranase to sugar processing.

The institution of dextran penalties by refiners has intensified the interest in dextranase on the part of U. S. raw sugar industry. While field and factory managements can 22

minimize the occurrence of dextran through suitable

operating practices, they cannot eliminate it. Consequently

commercialization of an FDA approved dextranase is a much

desired goal for the U.S. sugar industry.

XI. Application of Dextranase in Dental Hygiene

Dextranase has application not related to sugar cane

processing. The dextranase from Chaetomium has been

incorporated in toothpastes for the removal of dental

plaque. For well over twenty years it has been known that

dextran formed by Streptococcus mutants is a major component

of dental plaque (43). The Japanese have produced a toothpaste, "Clinca Lion" (85), that incorporates

dextranase in its formulation. Studies from Japan (68,69,85)

and elsewhere (22) have shown the effectiveness of this

enzyme in removal and prevention of dental plaque. This

toothpaste is not available in the U.S. due to the lack of

FDA approval for the fungi used in the production of the dextranase. However, toxicology studies have shown no adverse effects of this toothpaste on laboratory animals or clinical subjects (69,85). METHODS AND MATERIALS

I. Cultures

Lipomyces starkeyi (ATCC 12659) was the parent strain used in this study. A derepressed mutant (ATCC 20825) was derived by UV mutagenesis of ATCC 12659, with selection on

2-deoxy-D-glucose (Sigma Chemical) following the procedure of Laires et al. (38).

II. Growth Medium and Culture Maintenance

A modification of the medium described by Workman and

Day (93) was used as the basal mixture. This medium will be referred to as WW. All chemicals were reagent grade. WW medium had the following composition in unit weight per liter of deionized water:

Salts (g) Trace Elements (ug)

(NH4)2S04 5.0 Fe(NH4)3(S04)z 14000 KH2P04 1.5 ZnS04.2H20 3000 MgS04.7H20 0.1 MnS04.4H20 2000 NaCl 0.1 CuS04.4H20 300 CaCl2.2H20 0.1 (NH4)MO7024.4H20 90 H3BO3 570

The medium differs from that of Workman and Day (93) in that vitamins were omitted from the formulation. This strain of

Lipomyces did not require vitamins for growth.

Stock cultures of ATCC 12659 were maintained at room temperature on agar slants of WW medium plus 10 g/1 (w/v) dextran (Sigma Chemical; 35,000 mol. wt.) as a carbon

23 24 source, with a pH of 4.0 after sterilization. Stock cultures of ATCC 20825 were maintained at room temperature on

WW medium plus 10 g/1 (w/v) dextran (35,000 mol.wt.) and

0.5 g/1 (w/v) 2-deoxy-D-glucose, pH 4.5 after sterilization.

Working cultures were maintained in shake flasks of WW medium plus 10 g/1 (w/v) of appropriate growth carbohydrate.

Symbols for media are as follows; WW-Dex = WW medium plus

10 g/1 dextran, WW-Glu = WW medium plus 10 g/1 (w/v) glucose, WW-DD = WW medium plus 10 g/1 (w/v) dextran combined with 0.5 g/1 (w/v) 2-deoxy-glucose. All were adjusted to pH 4.5 before sterilization.

III. Optimization of Fermentation Parameters

Optimization of growth and dextranase production with respect to pH , temperature, and aeration was conducted in

0.5 liter batch cultures grown in a 1.0 liter MicroFerm fermentor (New Brunswick Scientific Co.) with WW-Dex as the growth medium. The culture used was ATCC 12659. The pH was maintained by addition of 3.0 N NaOH using a Model pH-40

Automatic pH Controller (New Brunswick Scientific Co.). The pH was normally maintained at 3.0. The aeration rate was normally set at 1.0 vvm. Temperature was normally maintained at 27 °C. The agitation rate for all experiments was 200 rpm. Inocula were 0.1 % (v/v) of a 72 h WW-Dex grown culture. Growth was monitored by measuring absorbance (OD) at 660 nm and converted to total cell count and dry weight 25 by using standard curves.

IV. Induction of the Derepressed Mutant (ATCC 20825)

Experiments to determine the inducer for dextranase were done with resting cell cultures. The derepressed mutant

(ATCC 20825) was grown to mid log phase in shake flasks containing 60 0 ml WW-Glu. Cells were harvested by centrifugation (3000 X g) and washed three times with WW medium, concentrated 10 X and resuspended in 0.1 % (w/v) of the test inducer plus 0.1 % (w/v) glucose, and adjusted to a final volume of 5.0 ml. These cultures were incubated in

250 mm X 50 mm cotton stoppered tubes which were sparged by pumping air though a pasteur pipette. Cultures were normally allowed to incubate for 12 hours at 27 °C, after which dextranase activity and cell concentration were measured.

Determinations of the effect of cell age on induction potential were performed in resting cell cultures.

L. starkeyi (ATCC 20825) was grown in a 30 liter New

Brunswick fermentor containing 20 1 of WW-Glu , pH 3.0,

27 °C, agitation 200 rpm, with a sparge rate of 20 1pm.

Cells were harvested at various time points during growth and treated in the same manner as with previous induction trials. Resting cell cultures were challenged with 0.1 %

(w/v) inducer plus 0.1 % (w/v) glucose and treated as described previously.

A series of 500 ml fermentation experiments were 26 conducted to determine the optimum induction point for

dextranase with respect to enzyme yields of ATCC 20825 grown

on 1.0 % (w/v) glucose (WW-Glu), pH 3.0, 27 °C. Two

concentrations of dextran (35,000 raol. wt.), either 0.1 %

(w/v) or 0.2 % (w/v), were added to the culture at various

stages of growth. The cultures were allow to reach

stationary phase (84 h) and dextranase activity measured.

V. Scale-Up

The fermentation system developed for the depressed

mutant (ATCC 20825) was scaled-up to 500 1 using a fermentor built by NASA (NSTL, Bay St. Louis, MS). All

parameters of operation are described in the results section. Dextran (Sigma Chemical; 5,000,000-40,000,000 mol. wt.) was used as the inducer of dextranase; 0.1 % (w/v) was added. Inocula were 72 h WW-DD grown culture.

VI. Enzyme and Protein Assay

Dextranase was assayed by a modification of the method of Webb and Spencer-Martins (92). Enzyme preparations were incubated with 2.0 % (w/v) Dextran T-2000 (Pharmacia) in

0.05 M Citrate-Phosphate pH 5.5 at 50 °C for 10 - 30 min.

Activity was determined from the rate of increase in reducing sugar as measured by the 3,5-dinitrosalcylic acid method (80). One unit (IU) of dextranase is defined as the amount of enzyme which liberates 1 umole of glucose 27 equivalents in one minute under the described conditions.

Sodium acetate buffer 0.05 M, pH 5.5 replaced Citrate-

Phosphate buffer in determining the effect of salts and inhibitors on activity. No difference was observed in activity between the two assay buffers. Protein was determined by the method of Lowry et al. (44).

VII. Enzyme Purification

Culture supernatant was concentrated from 20 1 to 500 ml using a 10,000 mol. wt. cut off stacked membrane

(Pelicon, Millipore, Co.). Protein and enzyme activity was monitored throughout the course of the purification.

A CM-Sepharose (Pharamacia) column (75 cm X 1.5 cm) was prepared and equilibrated with 0.02 M potassium phosphate

(pH 6.0). The crude concentrate (9 mg protein/ml) was applied to the column. The enzyme was deabsorbed by elution with 0.5 M NaCl, and active fractions pooled. The pooled fractions were concentrated on a 10,000 molecular weight cut off membrane in an ultrafiltration apparatus (Amicon, Co.).

The final volume was 2.0 ml.

A descending flow (10.44 cm/h) agarose A-0.5m (BioRad) column (75 cm X 1.5 cm) was prepared and equilibrated with

0.05 M Citrate-Phosphate (pH 5.5) plus 0.15 M NaCl. The concentrated CM-Sepharose fraction was applied to the column

(1.6 mg protein/ml).The fractions with activity were pooled and concentrated. 28

VIII. Molecular Weight Determination by Gel Filtration

A down flow gel filtration column (75 cm X 1.5 cm) was prepared of agarose A-0.5m (BioRad) containing a total bed volume (Vt) of 115 ml. The gel was equilibrated with 0.02 M potassium phosphate buffer (pH 7.0) plus 0.17 M NaCl. The flow rate was 0.38 ml/minute. The column was standardized with the use of Biorad gel filtration standard (total protein 18 mg). Using the absorption at 280 nm to identify the protein peaks, a plot of A2ao versus the fraction number was used to calculate the elution volume (Ve) of the different protein species. The dextranase peak was identified by using the previously described enzyme assay.

The Kav was determined for each protein as follows:

= V» - V„ (40) Vt - VQ

A plot of Kav versus log molecular weight was used to obtain an approximate molecular weight.

IX. Polyacrylamide Slab Electrophoresis

Polyacrylamide slab electrophoresis was performed using a modification of the technique described by Laemmli (39). A series of reagents were prepared as follows: 29

Solutions

1. Sealing agar: 1.5 % (w/v) Noble agar 0.1 % (w/v) Sodium dodecyl sulfate (SDS)

2. 10 % (w/v) Sodium dodecyl sulfate

3. Standard acrylamide stock: 30 % (w/v) acrylamide 0.2 % (w/v) bis acrylamide

4. Separating gel buffer: 1 M Tris-HCl pH 8.7

5. Stacking gel buffer: 1 M Tris-Hcl pH 6.8

6. Tetraethylmethylethylenediamine (TEMED): commercial solution (BioRad)

7. Ammonium persulfate (APS): 10 % (w/v)

8. n-Butanol saturated with distilled water

9. 5X Running buffer: Tris base 15 g/1 Glycine 72 g/1 SDS 5 g/1 pH 8.3

10. Stacking gel stock: Solution 3 16.7 ml Solution 5 12.5 ml Solution 2 1.0 ml Water to 100 ml

11. Sample buffer: Tris base 1.21 g/1 2-Mercaptoethanol 0.1 % (v/v) SDS 0.1 % (w/v) EDTA 0.001 M Glycerol 40 % (v/v) Bromothymol blue 0.05 % (w/v) pH 8.3

12. Fixative solution: 25 % (v/v) Isopropanol 10 % (v/v) Acetic acid

13. Stain solution: 0.1 % (w/v) Coomassie blue R-250 25.0 % (v/v) Methyl alcohol 10.0 % (v/v) Acetic acid

14. Destain solution: 25 % (v/v) Methyl alcohol 10 % (v/v) Acetic acid

Solutions one through nine were degassed and then used to make the final gel solutions. The separating gel (7.6 % 30 acrylamide) was prepared by combining:

5.6 ml solution 3 8.4 ml solution 4 0.25 ml solution 2 9.4 ml distilled water 30 ul solution 6 120 ul solution 7

This mixture was poured between glass plates sealed with solution 1, covered with about 2.0 ml of water-saturated n-butanol and allowed to polymerize for 45 min. The stacking gel was made by combining:

10.0 ml solution 10 10.0 y.1 solution 6 40.0 i±l solution 7

After the separating gel polymerized, the n-butanol was poured off, the top of the gel rinsed with distilled water, and dried. The comb was inserted and the stacking gel solution poured and allowed to polymerize.

The sample was prepared by mixing it one to one with solution 11 and boiling for 5 minutes. There was at least

10 ug of protein added to each well. A standard mixture of know molecular weights (Sigma Chemical; SDS-6H) was processed in the same manner and loaded on to the gel as a reference for molecular weight determination.

The gels were subject to electrophoresis at 60 volts until the tracking dye migrated to the bottom of the stacking gel, then the voltage was raised to 120 volts and the electrophoresis allowed to proceed until the tracking dye was within 3 mm of the bottom of the gel.

The gels were fixed in solution 12 for two hours then 31 stained with solution 13 for 18 hours. The gels were destained electrophoretically with solution 14 and scanned with a Ephotec/Joyce Loebl densitometer at 530 nm with an aperture setting of 0.05 mm X 3.0 mm to detect protein.

X. Isoelectric Focusing

Isoelectric Focusing was performed using precast IsoGel agarose IEF plates, pH 5.0 - 8.5 (FMC Bioproducts). Plates were prepared and processed as recommended by the manufacturer (FMC BioProducts, Rockland, Maine). A standard mixture of proteins (Serva Chemical; Test Mix 9) was applied in the lane next to each sample. Protein profile was ascertained by densitometer scans as previously described.

The enzyme activity of the gel was determined by slicing an unstained gel into 0.9 mm sections. Each section was placed in a test tube with 1.0 ml 0.05 M Citrate-Phosphate (pH 5.5) allowed to elute overnight at 4 °C and assayed for enzyme activity.

XI. Determination of Carbohydrate Content

Enzyme (1 ml)was dialyzed overnight against 500 ml of

6 M urea followed by 3 changes of deionized water. The carbohydrate was determined by the phenol-sulfuric acid method (16) and the protein by the method of Lowry et al.

(44). 32

XII. Lectin Reactivity

The enzyme was tested for retention on a lectin support

as described by Montreuil et al. (52). Concanavalin-A

(Con-A) Sepharose (Sigma Chemical), Lens culinaris (Lentil)

Sepharose (Sigma Chemical), and Wheat germ Sepharose (Sigma

Chemical) were all tested. Each support was prepared and

equilibrated as previously described (52). Pure enzyme (500

IU) was applied to each column and the activities in the

void volumes determined. Various elution conditions were

used to remove the enzyme from the support. The ability of

the enzyme to bind and the conditions required to deabsorb

were compared for the three supports and relative

reactivity to the lectins determined.

XIII. Amino Acid Composition

Purified dextranase was dialyzed against distilled

water overnight. The protein sample was hydrolyzed in

6 N HC1 at 110 °C for 22 h. The amino acid composition of

this preparation was determined using a Pico-Tag Amino Acid

Analyzer. Tryptophan was not determined by this method.

XIV. Dextranase; pH Optimum and Stability

The pH optimum was determined using the standard

dextranase assay at 50 °C with a 10 min incubation time. The

pH of the substrate solutions covered a range from pH 2.5 to

7.0 in intervals of 0.5 pH units. The pH of the reaction 33 mixture was tested, before and after analysis and found not to change.

To determine the stability of dextranase, stock enzyme solution was diluted with 0.05 M Citrate-Phosphate and adjusted to the desired pH (between pH 2.5 and 8.0). The solutions were incubated at 25 °C for 72 hours. Every 24 hours the activity of each reaction mixture was assayed.

The pH was monitored and found not to change during the course of the experiment. A plot of the relative activity at each pH versus the time of incubation was used to determine the pH at which the enzyme was most stable.

XV. Dextranase: Temperature Optimum and Stability

The optimum temperature was determined using the standard dextranase assay run at various temperatures. A plot of the relative activity versus incubation temperature was used to determine the temperature optimum.

The temperature stability was determined by incubating the enzyme at specific temperatures for specific time intervals (0, 5, 10, and 20 minutes respectively) then measuring dextranase activity. A plot of the relative activity at each temperature versus time of incubation was used to determine the temperature at which the enzyme was the most stable. 34

XVI. Determination of Michaelis-Menten Constants

The Michaelis-Menton constants were determined for the

T-series dextrans (Pharmacia) by the method of Lineweaver and Burk (41).

XVII. Inhibitory Effect of Carbohydrates

Amounts up to 50 mM of various carbohydrates (Sigma

Chemical) were included in the standard assay buffer to test their affect upon activity.

XVIII. Effect of Salts and Inhibitors on Activity

Stock solutions of the compounds to be tested were made in 0.05 M sodium acetate (pH 5.5). A 100 jjlI aliquot of suitably diluted test solution was added to 5 ul of enzyme containing 850 ul of 0.05 M sodium acetate (pH 5.5) at 50 °C and incubated for 5 minutes, at which time 1.0 ml of 4.0 % dextran T-2000 in 0.05 M sodium acetate (pH 5.5), preincubated at 50 °C, was added. Activity was determined as previously described. All salts were of reagent grade.

XIX. Protease Inhibition

The effect of various proteases (Sigma Chemical) on dextranase activity were tested. Stock solutions and incubation parameters were as follows: 35

Incubation parameters Stock pH Buffer Temperature Protease mg/ml (°C)

Ficin 1.25 7.0 0.1 M K/P 37 Pepsin 1.25 6.0 0.1 M C/P 37 Bromelain 1.25 5.5 0.1 M C/P 37 Protease XIII 1.25 5.5 0.1 M C/P 37 Protease XXI 1.25 7.6 0.1 M K/P 37 Papain 1.14 6.5 0.1 M K/P 25 Trypsin 1.25 7.6 0.1 M C/P 25

Equal volumes of protease stock solutions and enzyme solutions were mixed and incubated under the conditions listed above; C/P denotes citrate-phosphate buffer and K/P signifies potassium phosphate buffer. A 100 ul sample was withdrawn from each reaction at 0, 2.0, 4.0, 6.0, and 8.0 hours and assayed for dextranase activity.

XX. Analysis of Reaction Products

Enzyme (200 IU) was incubated with 400 ml of 2 % dextran T-2000 (Pharmacia) at 40 °C, pH 5.5. Samples (10 ml) where taken at various time intervals and boiled for 5 min to stop the reaction. Samples were then filtered through a

0.22 pm membrane and assayed by HPLC for the oligosaccharide content.

The High Performance Liquid Chromatograph (HPLC) used was a Varian Vista 5500 with a refractive index detector linked to a Varian 4270 integrator for data acquisition.

The column employed for separations was a Rainin NH2-Dynamax 60A. The HPLC conditions were: solvent; acetonitrile:water 36

(1:1), column temperature; 25 °C, flow rate; 1.00 ml/min, column pressure 52 atm, with a 20 y.1 sample injected. The maltooligosaccharide series DP1-DP7 (Sigma Chemical) were used as external calibration standards. A comparison of the malto-series to the isomalto-series was performed. No differences in detector responses were observed. The isomaltooligosaccharides were a kind gift of Dr. F. Paul,

BioEurope Ltd., Toulouse, France.

XXI. Application of Dextranase in Sugar Processing

A. Laboratory tests

Fresh cane juice was spiked with dextran (Sigma

Chemical; 40,000,000 mol. wt. ) such that the final concentration of dextran was at least 25,000 ppm/Brix

(Brix (Bx) = % solids). Dextranase, 0.7 IU/ml, was added to each test. The pH of the final mixture was 5.6 and the incubation temperature was 25 °C Samples were taken every

15 minutes for one hour. The reaction was stopped by alcohol precipitation. Dextran was measured by the enzymatic ASI-II method (70).

B. Pilot trial

Two slings of sugar cane (2.9 metric tons) were obtained from the LSU Agricultural Research Station at St.

Gabriel, LA. From this cane, 2000 1 of mixed juice was produced. Sufficient dextran (Sigma Chemical; 40,000,000 mol. wt.) was added to this juice to make it approximately 37

10,000 ppm/Brix. Dextranase was added to one half of this juice such that the final enzyme concentration equaled 0.28

IU/ml. The dextranase treated juice was allowed to stand for one hour at 27 °C, pH 5.6, before clarification. Both the dextranase treated and untreated juices were processed using standard procedures to produce an "A" strike sugar. Dextran analysis was by either the ASI-II method (70) or the Haze method (11).

C. Dextran analysis

Dextran was measured in cane juice, syrup and sugar crystals by the enzymatic ASI-II method (70). Briefly, the samples were treated with ethyl alcohol (ETOH) to precipitate all polysaccharides present. The dextran in the reconstituted precipitate, was hydrolyzed with the use of dextranase. A standard curve was used to convert the amount of reducing sugars produced to the amount of dextran present.

Dextran in sugar crystals was also measured by the Haze method (11). This is the method that is employed by the sugar industry. Briefly, the samples are treated with ETOH and a turbidimetric measurement of the resulting precipitate is used to quantify dextran content. RESULTS

I. Optimization of Fermentation Parameters

A. Effect of pH

Lipomyces starkeyi (ATCC 12659) was found to grow best

over a pH range of 2.5-4.0 (Table 1). Other Lipomyces

strains have been reported to grow best at pH 4.8 (54,88).

The pH fell dramatically when L. starkeyi (ATCC 12659) was

grown in phosphate buffered medium. When WW-Dex was adjusted

to an initial pH of 5.0, the final pH was 2.4 at the end of

growth. Growth at pH values above 4.0, resulted in

increased lag time and decreased growth rates (Table 1).

Dextranase production was highest when growth was

between pH 3.0 and 5.0, with a maximum yield of 4.4 IU/ml

produced at pH 3.0. Specific production (IU/10a cells) was

also highest at pH 3.0, although values obtained between pH

3.0 and 5.0 were not appreciably different (Table 1). The

time required to reach peak enzyme yield was about 1.5 times

greater at pH 5.0 than at pH 3.0. After 200 hours growth at

pH 6.0, only trace amounts of enzyme were detected.

Dextranase yields and specific productions at pH 2.5 were over 5 times lower than at pH 3.0 (Table I). The decrease of enzyme was not reflected in a parallel decrease

in yeast growth. When Lipomyces was grown between a pH of

2.5 and 4.0, both cell yields and growth rates remained the same (Table 1).

38 39

Table 1

Effect of Medium pH on Lag Period, Growth Rate, and Dextranase Production

Time to Time tox maximum Final Specific reach Doubling enzyme enzyme dextranase Medium OD=0.8 time yield yield production PH (h) (h) (h) (IU/ml) (IU/108 cells)

2.5 48 6.3 72 0.7 1.09

3.0 48 6.0 83 4.4 6.35

4.0 58 5.7 84 3.3 5.88

5.0 98 13.3 124 3.2 5.31

6.0 200 48.0 xLag period.

The stability of the crude and the purified enzyme at pH 2.5 and pH 3.0 is summarized in Table 2. The crude enzyme preparation (concentrated culture broth) was completely inactivated at pH 2.5, wheras the purified still retained activity at that pH.

Table 2

pH Stability of Crude Versus Purified Dextranase K1 % Activity remaining after 24 h incubation at 30 °C pH Crude enzyme Purified enzyme

2.5 0 95

3.0 79 100 40

B. Effect of aeration

Aeration levels were found to alter enzyme levels but not growth yields (Table 3). Maximum enzyme production, 5.2

IU/ml, occurred with an aeration rate of 1 vvm. Aeration rates above 1.5 and below 1.0 vvm decreased dextranase yields.

Table 3

Effect of Aeration on Growth Rate and Dextranase Yield

Final Specific Aeration Doubling enzyme dextranase rate time yield production (vvm) (h) (IU/ml) (IU/108 cells) 0.24 5.7 2.4 2.79

0.50 5.8 2.9 3.14

1.00 6.0 5.2 4.29

1.50 5.9 4.0 4.24

2.00 6.0 3.1 3.05

2.50 5.8 2.5 2.15

C. Effect of temperature

The temperature range for growth (Table 4) was comparable to those reported for other Lipomyces spp.

(2,19,54). This strain (ATCC 12659) was not able to grow at or above 33 °C. Final enzyme yields and specific production values did not differ for the temperatures tested (Table 4). 41 Table 4

Effect of Incubation Temperature on Growth Rate and Dextranase Yield

Time to maximum Final Specific Incubation Doubling enzyme enzyme dextranase temperature time yield yield production (°C) (h) (h) (IU/ml) (IU/10a cells)

__ 33 no growth --

30 5.4 69 14.5 17.0

27 5.7 88 11.4 12.7

25 6.0 99 12.9 15.6

20 10.0 167 10.4 11.5

15 24.0 — —

II. Dextranase Production by a Derepressed Mutant

The derepressed mutant (ATCC 20825) was isolated by UV mutagenesis from ATCC 12659 coupled with 2-deoxy-D-glucose selective pressure (38). The mutation frequency of the surivors of irradiation was 4.7 X 10-4 for the derepressed trait.

A. Strain comparisons

The mutant strain (ATCC 20825) was found to produce increased levels of dextranase compared to the wild type

(hyperproduction) when grown on dextran as the sole carbon source (Table 5). As is characteristic of a derepressed mutant, this strain was able to produce dextranase when glucose and dextran were the combined carbon source

(Table 5). 42

Table 5 Comparison of Wild Type (ATCC 12659) and the Derepressed Mutant (ATCC 20825)

Dextranase production (IU/10a cells)3- Growth substrate ATCC 12659 ATCC 20825

1.0 % Glucose 0 0

1.0 % Dextran 0.80 2.24

1.0 % Glucose + 0.1% dextran 0 0.67 xDextranase measured after 72 h of growth on listed carbohydrate.

B. Determination of inducer

A number of potential inducers were screened to determine the most efficient inducer of dextranase production. A listing of those which tested positive is given in Table 6. The following compounds did not induce dextranase production in ATCC 20825; citrate, dextrin, amylose, inulin, melibiose, sorbose, cellobiose, starch, panose, maltoheptaose, melezitose, maltohexaose, ma1topentaose, maltose, stachyose, maltotriose, maltotetraose, 2-deoxyglucose, a-methyl-mannopyranoside,

1-B-methyl-galactopyranoside, 1-a-methyl-galactopyranoside,

3-O-methyl-glucopyranoside, and glucose.

It is apparent from Table 6, that

1-B-methyl-glucopyranoside (BMG) is the best dextranase inducer. Isoma 11opentaose, isomaltotriose,

1-a-methyl-glucopyranoside (AMG), and dextran all induce dextranase production to approximately the same levels

(Table 6). Isomaltotetraose and isomaltose were poor 43 inducers of dextranase.

Table 6

Compounds that Induce Dextranase Production

Compound Induction index1

Dextran 100

Isomaltopentaose 57

Isomaltotetraose 39

Isomaltotriose 73

Isomaltose 21

1-a-Methy1-glucopyranoside 68

1-fi-Methyl-glucopyranoside 219

Glucose 0 xDextran induced cells equal 100; all others compared to dextran induced cells. Determined after 12 h incubation with 0.1 % (w/v) test compound plus 0.1 % (w/v) glucose.

C. Effect of exposure time to inducer

The length of incubation time with the inducer affected the amount of dextranase produced. After 8 h all three inducers produced the same levels of enzyme (Figure 1).

After 8 h of exposure to AMG or to dextran enzyme production leveled off but the BMG induced cells continued producing enzyme until 12 h of incubation (Figure 1).

D. Saturation level of inducer

The inducer saturation level for maximum dextranase production was the same for AMG, BMG, and dextran, approximately 1 mg/ml of inducer for every 2.0 X 10a cells 44 (Figure 2). Although the final enzyme yields were different.

E. Effect of cell age on induction potential

The age of the culture had a direct influence on the amount of enzyme produced by the yeast (Figure 3). The younger cells produced more enzyme produced per cell then older cells. Although it is apparent from Figure 3 that young cells have the highest capacity for dextranase production, it is not apparent where the required point of induction is for maximum dextranase yields. A series of batch fermentation experiments were conducted to determine the point for optimal induction of ATCC 20825 grown on

WW-Glu. Two concentrations of dextran were used, either

0.1 % (w/v) or 0.2 % (w/v), added to the culture at various points in the growth curve. The cultures were allowed to reach stationary phase (84 h) and the dextranase activity was measured. The time of inducer addition during the fermentation was critical for optimization of dextranase production. Cells induced in the late exponential phase of growth (8 X 107 cells/ml) produced the most dextranase

(Figure 4). There were only minor differences in the maximum enzyme yields on doubling the amount of inducer at this point. Optimum production of dextranase in batch culture for the derepressed mutant (ATCC 20825) grown on a minimal medium plus 1.0 % glucose was when enzyme synthesis was triggered by 0.1 % dextran added at a cell concentration of

8 X 10s cells/ml. 45

Figure 1. Time course of dextranase induction. Procedures as described in text. Inducers tested: dextran (open boxes), B-methyl-glucopyranoside (closed boxes), and a-methyl-glucopyranoside (closed diamonds). o Dextranase activity (IU/10 cells)

© — ro w 4 , i , 1 1 47

Figure 2. Effect of inducer concentration on the amount of dextranase produced. Procedures as described in text. Incubation period 12 h. Inducers tested: dextran (open boxes), B-methyl-glucopyranoside (closed boxes), and a-methyl-glucopyranoside (closed diamonds). Inducer concentration (mg/ml) 49

Figure 3. Effect of culture age upon induction potential. Cells were harvested at various time points during growth (growth curve: open boxes) then used in resting cell culture for determination of dextranase activity after 12 h incubation with 0.1 % glucose plus the following inducers: dextran (closed boxes), B-methyl-glucopyranoside (open diamonds), and a-methyl-glucopyranoside (closed diamonds). Cell concentration (10^/ml) Q Dextranase activity (IU/10 cells)

O S3 A ©N C5 ON) W ©

otn

•si o

vO o

OS 51

Figure 4. Enzyme activity versus time of induction and cell concentration (closed circles). (A) inducer concentration equals 0.1 % (w/v) dextran. (B) Inducer concentration equals 0.2 % (w/v)dextran. Vertical bars are at the location of inducer addition. Their values show the dextranase activity after 84 h of incubation in batch fermentors. Cell concentration (10''/ml)

IO an ls»

H 3 (l>

(Im/fll) ^JATIOB aseuBJtrixaQ 53

III. Scale-Up

The ability to scale this system up was tested by comparing the dextranase production of the mutant yeast in

0.5, 20, and 500 liter batch fermentations. Unfortunately, the mutant that was used for these experiments, although still derepressed, lost the ability for hyperproduction.

Hence the level of enzyme produced was not as high as would be expected from the previous trials. Table 7 summarizes the scale-up data.

Table 7

Scale-Up

Fermentor size Parameter 0.5 liter 20 liter 500 liter

PH 3.0 3.0 2.5-3.0 Temperature (°C) 28 28 28

Sparge rate (vvm) 1.0 1.0 0.4

Inoculum volume (1) 0.02 1.0 25

Time to induction (h) 52.5 52.5 46.0

Induction cell concentration (g/1) 3.5 3.3 3.2

Time to harvest (h) 69.5 68.0 64.5

Harvest cell concentration (g/1) 5.3 4.9 4.7

Final dextranase activity (IU/ml) 4340 3917 3688

Dextranase yield (IU/g cells final) 820 793 786 54

IV. Enzymoloqy

A/Purification of dextranase

Enzyme for kinetic studies was produced by the derepressed mutant of Lipomyces starkeyi (ATCC 20825) grown under the previously described conditions. A five step purification protocol (Table 8) was required to purify dextranase from Lipomyces starkeyi (ATCC 20825).

Table 8

Purification of Lipomyces starkeyi Dextranase

Total Total Activity Specific Fold Procedure protein activity yield activity purif. (Step) (mg) (IU) (_%) (IU/mg)

1. Clarified broth 1740 55600 •- 32 • -

2. 10K concentrate 1718 55375 99.6 32 1.0

3. CM-Sepharose 148 55044 99.0 372 11.6

4. 10K concentrate 148 55044 99.0 372 11.6

5. A-0.5m agarose 28 38767 70.0 1385 43. 3

Dialysis of CM-Sepharose step 148 29524 53.1 199 6.2 IPA precipitation of clarified broth 378 55025 99.0 146 4. 5

The concentration Steps 2 and 4, were done for ease of handling. Two other steps (not numbered) are shown in

Table 8, one which involves the dialysis of the CM-Sepharose fraction (Step 3) and another which is the uses isopropanol precipitation to purify crude enzyme from Step 2, a technique described by Webb and Spencer-Martins (92). 55

Comparisons of each purification step with regard to protein and activity profile, as determined by agarose gel filtration (BioRad; A-0.5 m), are shown in

Figures 5 through 9.

The crude enzyme fraction (Step 1) showed a single activity peak, as determined by A-0.5m, with an estimated molecular weight of 68,000 (Figure 5). Five protein peaks were observed. The initial purification step, cation exchange chromatography (Step 3), gave a 12 fold purification factor with no appreciable loss of enzyme.

Figure 6 depicts the A-0.5m profile for the

CM-Sepharose fraction. Subsequent dialysis of the

CM-Sepharose fraction against 0.02 M potassium phosphate buffer produced a shift of both protein and activity

(Figure 7) as separated by the A-0.5m column. The activity was found in peaks ranging from about 68,000 to the void volume (>700,000). The enzyme found in the void volume

(>700,000) appeared to be an aggregate. It was only possible to dissociate the multimer with SDS. SDS dissociated the multimer to the 68,000 form, but irreversibly inactivated the enzyme.

A single protein and activity peak were obtained by fractionation of the CM-Sepharose preparation on a A-0.5m gel filtration column (Figure 8). This was the enzyme fraction used for all the kinetic and physical protein determinations. 56

Figure 5. Gel filtration profile for crude dextranase. Protein (open boxes) and enzyme activity (closed diamonds) are plotted against fraction number. ao

60

40

-20

60 70

Fraction number 58

Figure 6. Gel filtration profile for CM-Sepharose purified dextranase. Protein (open boxes) and enzyme activity (closed diamonds) are plotted against fraction number. 150 o CD X rt H Pi 3 Pi 100 (0CO pi n rt H- < H* 50

CJ a* IBfSUS&tf1' 30 70

Fraction number

Lnv£> 60

Figure 7. Gel filtration profile for CM-Sepharose purified then dialyzed dextranase. Protein (open boxes) and enzyme activity (closed diamonds) are plotted against fraction number. 150 0 (D X rt •i pi 3 01 0) 100 (D (U O rt H- < H' 50

a

70

Fraction number 62

Figure 8. Gel filtration profile for A-0.5m purified dextranase. Protein (open boxes) and enzyme activity (closed diamonds) are plotted against fraction number. ~ 0.2 80 rH O n> x rt Pi 60 3 P> COn> pi o 40 rt H* < H-

-20

cS ET

10 20 30 40 50 60 70 Fraction number

ON u> 64

Figure 9. Gel filtration profile for isopropanol purified dextranase. Protein (open boxes) and enzyme activity (closed diamonds) are plotted against fraction number. 80 a n> x rt •i P) 60 3 Pi n>CO PJ o 40 rt H- < rt 20 CJ 3*

70 Fraction number

UtON 66

Isopropanol precipitation by the method of Webb and

Spencer-Martins (92) was used to purify dextranase from ATCC

20825 but produced only a 5 fold increase in specific activity. This preparation was not pure as is apparent from

Figure 9. The molecular weight of the active fraction was the same as seen before (68,000).

B. Molecular weight determinations

The molecular weight of purified dextranase, by gel chromatography (A-0.5m), was calculated to be 68,000. The dextranase on SDS-PAGE (7.6 %) fractionated into four protein bands with the molecular weights of 74,000, 71,000,

68,000, and 65,000 respectively (Figure 10). Lipomyces dextranase in the native, nondenatured state, interacted with polyacrylamide making normal acrylamide electrophoresis unreliable. Reactivation of the SDS treated enzyme was not possible.

C. Isoelectric focusing

To demonstrate that the four bands appearing on

SDS-PAGE were dextranase, the purified enzyme was fractionated by agarose isoelectric focusing (IEF). This method separated the protein mixture into five isoelectric bands (Figure 11). All five forms were found to have dextranase activity and exhibited the same apparent Km values (Table 9). 67

Figure 10. Densitometer scan of 7.6 % SDS-PAGE fractionation of purified dextranase. The molecular weights are noted on figure. Relative density

CO rr ft) P o n> Hi O B rt O •d o Hi 09

o> 69

Figure 11. Densitometer scan of an isoelectric focusing gel fractionation of purified dextranase. The isoelectric values are given. 250 J

200 fs •M •H 01 a 0) ISO • *o Q) > •H •U

50

0

Distance from cathode (ram) 71

Table 9

Apparent Km Values of Isoelectric Forms pi Km (mM)

Combined 0.0030

6.06 0.0023

5.95 0.0030

5.80 0.0028

5.73 0.0025

5.61 0.0024 Standard enzyme assay conditions with T-2000 as substrate.

D. Kinetic comparisons

The apparent Km values for the native (68,000), aggregate

(>700,000), and the IGC 4047 dextranase (92) are given in

Table 10. No difference in kinetic values were seen between the three preparations.

Table 10

Apparent Km Values for Native, Aggregate, and IGC 4047 Dextranase

Native Aggregate IGC 40471 Dextran type (mM) (mM) (mM)

T-2000 0.004 0.003 0.003

T-500 0.018 0.017 0.013

T-70 0.094 0.096 0.143

T-40 0.190 0.173 0.171

T-10 0.651 0.449 0.456 xData from Webb and Spencer-Martins. 1983. Can. J. Microbiol. 29:1092-1095. Standard enzyme assay conditions. 72

E. Carbohydrate and amino acid composition

The purified dextranase was found to contain 8 % carbohydrate by weight. This glycoprotein was tested for its reactivity to lectins and was found to bind very strongly to

Concanavalin-A (Con-A), weakly to Lens culinaris agglutinin

(LCA), and not at all to Wheat germ agglutinin (WGA).

This implied that the primary component of the attached carbohydrate was rich in mannose.

The amino acid composition of the Lipomyces enzyme is shown in Table 11 as well as the reported amino acid composition of both the Chaetomium (29) and the Penicillium

(9) dextranases.

F. Effect of pH and temperature

Dextranase was found to exhibit its highest activity at pH 5.0 (Figure 12). The enzyme was very stable between pH

2.5 and 6.0. There was a dramatic loss of activity after 30 hours of exposure to a pH above 7.0 (Figure 13).

This enzyme was found to exhibit optimum activity at

50 °C (Figure 14). Dextranase was very stable between 30 and

40 °C. All activity was lost after 5 min exposure to 60 °C

(Figure 15). 73

Table 11

Amino Acid Compositions of Dextranase

Number of amino acid residues in protein Amino acid Chaetomium1 PeniciIlium2 Lipomvces

Lys 18 12 6

His 17 8 16

Arg 29 8 22

Asp 80 48 22

Thr 48 24 48

Ser 61 40 88

Glu 39 20 22

Pro 39 20 56

Gly 61 31 146

Ala 32 19 52

Cys 10 2 4

Val 48 24 27

Met 15 7 14 lie 44 24 24

Leu 29 14 21

Try 35 16 27

Phe 22 15 22

Trp 21 5 — — 5

Mol. Wt. 74 K3 41 K 73 K4 xData from Hattori, et al. 1981. Agric. Biol. 45:2409-2416. 2Data from Chaiet, et al. 1970. Appl. Microbiol. 20:421-426. 3Mol. wt. average of two isoenzymes. 4Mol. wt. average of multiple isoelectric forms, tryptophan is not determined by the Pico-Tag system. 74

Figure 12. Effect of pH on dextranase activity. Relative activity

hj A 0* CO o © © © © © ©

fs)

oo

£L 76

Figure 13. Effect of pH on dextranase stability. pH: 2.5 (open boxes), 3.0 (closed diamonds), 4.0 (closed squares), 5.0 (open diamonds), 6.0 (closed rectangles), 7.0 (open boxes), 8.0 (closed triangles). Relative activity NJ Ji 0> CO © © © © © © ©

IO o

©

CO ©

LL 78

Figure 14. Effect of temperature on dextranase activity. Relative activity 80

Figure 15. Effect of temperature on dextranase stability. Temperature: 30 °C (open boxes with dot), 40 °C (closed diamonds), 50 °C (closed boxes), 60 °C (open diamonds). 100

u •rl •S 4J O « 0) > •H •U i-lid at Pi

100 200 300

Time (min)

oo 82

G. Reaction products

The end products of the reaction were found to be typical for those of an endodextranase. The reaction products observed after 26.5 h of enzyme incubation with dextran are given in Table 12. The progression of reaction products are shown in Figure 16. Isomaltotetraose was more recalcitrant to enzymatic hydrolysis than was isomaltopentaose. The final products of the reaction were isomaltotriose and isomaltose. Very little glucose was formed.

Table 12

End Product Analysis

Carbohydrate Percent of total

Glucose 0.8

Isomaltose 41.5

Isomaltotriose 34.9

Isomaltotetraose 2.7

Isomaltopentaose 0.0

Higher isomaltooligosaccharides 20.1 Carbohydrates quantified by HPLC, after 26.5 h incubation of 2.0 % T-2000 with 0.7 IU/ml, 40 °C, pH 5.5.

H. Effect of proteases on activity

The proteases pepsin, protease XIII, papain, and trypsin had no affect on enzyme activity. The recoveries of activity for enzyme incubated with proteases is shown in

Table 13. S3

Figure 16. Time course of reaction product formation. Carbohydrate: isomaltose (open boxes), isomaltotriose (closed diamonds), isomaltotetraose (closed boxes), isomaltopentaose (open diamonds). Carbohydrate concentration (mg/ml)

o to a a* 09 o

•*L

O o

to o o

o o 85

Table 13

Effect of Exposure to Proteases on Dextranase Activity1

Exposure time (h ) Ficin Bromelain Protease XXI

0 100 100 100

2 61 97 0

4 34 90 0

6 23 85 0

8 14 82 0 ^Reported as percent activity recovered after incubation of the enzyme with the protease for the indicated length of time.

I. Effect of salts on activity

Various salts were included in the assay mixture to determine their effects on activity (Table 14). The citrate-phosphate buffer was replaced by sodium acetate to prevent chelation of the ions, otherwise the assays were done as described in the materials and methods. The following salts showed no inhibition at the 15 mM concentration; NH4C1, KC1, NaCl, and NaF.

J. Effect of enzyme inhibitors on activity

Enzyme inhibitors were tested at different concentrations and their effects are listed in Table 15.

Sodium dodecyl sulfate (SDS) inhibited the enzyme. The combination of Polyethylene ether (Triton X-100) and SDS overcame the inactivation. Addition of Triton X-100 after

SDS exposure did not reactivate the enzyme. 86

Table 14

Effect of Salts on Dextranase Activity3-

Salt Concentration (mM) Relative activity

Bads 11.40 51 0.11 101

CaCl2 13.90 48 0.14 83

CdCl2 1.39 20 0.14 78

CoCl2 1.10 60 0.11 89

CuS04 0.13 7

FeCl2 1.45 19 0.15 70

FeS04 1.36 0 0.14 30

HgCl 1.00 0

MnCl2 11.50 24 0.12 103

MgCl2 15.00 66 0.15 96

MgS04 15.00 64 0.15 95

NiCl2 11.10 14 0.11 89

ZnS04 1.24 8 0.12 52 xReported as percent activity recovered after incubation of the enzyme with the salt at 50 °C for 5.0 minutes. 87 K. Reactivation of mercuric chloride treated enzyme

Mercuric chloride (1 mM) completely inactivated dextranase (Table 14). Addition of 2-Mercaptoethanol (5 %) to mercuric chloride inactivated dextranase restored enzyme activity (Table 15).

Table 15

Effect of Enzyme Inhibitors on Dextranase Activityx

Inhibitor Concentration Relative Activity (%)

Control 100

EDTA2 5 mM 103

Dithiothreitol 5 mM 100 2-Mercaptoethanol (2ME) 5 % 103 HgCl + 2ME2 1. 0 mM + 5 % 131 Glutathione 125 mM 100

Guanidine-HCl 547 mM 13

274 mM 61 50 mM 103

Urea 547 mM 36 274 mM 77 50 mM 99

Sodium m-periodate 5 mM 15

SDS 0.5 % 0 0.05 % 46

Triton X-100 0.25 % 99 0.025 % 100

SDS + Triton X-1002 0.5 % + 0.25 % 80 0. 05 % + 0.025 % 122 '-Reported as percent activity recovered after incubation of the enzyme with inhibitor at 50 "C for 5.0 minutes. Combination of both chemicals. 88

L. Effect of carbohydrates on activity

Various carbohydrates were tested at a concentration of 10 mM, in the assay mixture as to their effect on enzyme activity. Glucose, 2-deoxyglucose, a-methyl-glucopyranoside,

B-methyl-glucopyranoside , maltose, maltotriose, maltotetraose, isomaltose, isomaltotriose, mannose, a-methyl-D-mannoside, and stachyose did not alter enzyme activity.

M. Effect of solvents on activity

Several solvents were tested for their effect on enzyme activity. Figures 17 and 18 depict these results.

Dextranase from Lipomyces exhibited no decrease in activity in a 7.5 % methanol (MEOH), 7.5 % Acetone (ACE), and 10 % ethanol (ETOH) solutions. Above these concentrations enzyme inactivation occurred. Isopropanol (IPA) and n-propanol

(NPA) inhibited the activity above the 2.5 % level

(Figure 17). NPA had a greater inhibitory effect than IPA.

Acetone (ACE) and ETOH enhanced enzyme activity at

5.0 % (v/v) (Figure 18). 89

Figure 17. Effect of solvents on dextranase activity. Solvent: methanol (open boxes), isopropanol (closed diamonds), n-propanol (closed boxes). 120 -i

100

•u •H > •rH 4J U cd •rl 4-1

~l ' T ' i 10 20 30 AO

Solvent concentration (Z) V0 O 91

Figure 18. Effect of solvents on dextranase activity. Solvent: ethanol (open boxes), acetone (closed diamonds). 120-1

too

80 -

Q> 60 - > •H 4J .Hcd

20 -

0 10 20 30 Solvent concentration (Z) 93

V. Application of Dextranase in Sugar Processing

A. Laboratory trials

The optimum temperature for Lipomyces dextranase

activity in cane juice was between 30 and 40 °C (Table 16),

although there was a large amount of dextran removed, 74 %, at 50 °C.

Figure 19 shows a typical enzyme response curve, with

the rate of dextran removal decreasing in cane juice as the

dextranase concentration decreases. As expected, the enzyme

worked at a constant rate (Figure 20). An estimated dosing

rate of 2 ppm enzyme on juice is required to remove 1000

ppm/Brix dextran in mixed juice (15 Brix, 25 °C holding temperature for 20 min). This value was estimated on

purified dextranase (1400 IU/mg protein).

Table 16

Removal of Dextran from Mixed Juice: Effect of Incubation Temperature

Incubation Incubation Dextranase Dextran concentration temperature period concentration Start Final (°C) (min) (IU/ml) (ppm/Bx) (ppm/Bx)

20 40 5.4 28263 6760

30 40 5.4 28263 1025

40 40 5.4 28263 1030

50 40 5.4 28263 7440 Dextran analysis by the ASI-II method. Values are averages of triplicates. 94

Figure 19. Effect of enzyme concentration upon dextran removal from cane juice (40 minute incubation at pH 5.4, 25 °C). Dextran analysis by ASI-II method. Values averages of triplicates. Enzyme concentration (IU/ml) 96

Figure 20. A time course of dextran removal from cane juice (0.7 IU/ml, pH 5.4, 25 °C). Dextran analysis by ASI-II method. Values averages of triplicates. Dextran removed (Z)

NJ w A en 9 O o O o o

to o

B ID A O BH* 3

0> o

CD o

L6 98

B.Pilot trial

Sufficient naturally contaminated cane was not available during the 1987 season for a pilot test, so mixed juice was spiked with dextran. Additions were made to a level higher than what normally might be encountered in a commercial situation. The enzyme dose was lower than would be recommended under normal applications.

In the pilot trial, the clarified juice showed a 45 % decrease in dextran after exposure to dextranase for one hour (Table 17). Extrapolating from the dose calibration curve in Figure 20, a 22 % reduction in dextran was to be expected. Approximately twice as much dextran was removed than was anticipated.

Table 17

Removal of Dextran: Mill Trial Results

Dextran concentration Untreated Treated Sugar product

Clarified mixed juice (ASI-II) 10870 342 6028 365

Syrup (ASI-II) 10540 140 5714 512

Sugar (ASI-II) 3625 95 1155 74 (Haze) 2240 540 Dextran analysis method as noted in table. Values are averages of triplicates, except for the Haze test.

The untreated mixed juice was almost impossible to process and the sugar produced could not be adequately purged. The small amount of sugar that could be recovered 99 was dark in color and needle grained (Figure 2IB). The dextranase treated material processed normally and the sugar produced was both square and light in color (Figure 21A).

Dextran analysis of both the syrup and the sugar produced demonstrated the effectiveness of this enzyme in removing dextran from cane juice (Table 17). A 76 % reduction in dextran content, by Haze test, was seen between treated and untreated sugar. 100

Figure 21. A comparison of treated (A) versus untreated sugar. Scale = 30 pi. 101 DISCUSSION

Optimum dextranase yields were obtained when

Lipomyces starkeyi (ATCC 12659) was grown on a minimal medium with dextran as the sole carbon source (WW-Dex) at pH

3.0. The low pH makes this strain of L. starkeyi an attractive vector for the commercial production of dextranase. This low pH for both growth and enzyme production minimizes the contamination problems usually associated with large scale fermentation (78). L. starkeyi fermentation, unlike other fungal dextranase fermentations, offers the convenience of submerged culture and the ease of yeast cell separation from an extracellular product. Below pH 5.5 L. starkeyi does not require a complex mixture of nutrients and vitamins for dextranase production, hence the development of industrial feedstocks should be simplified.

When the pH fell below 3.0, dextranase yields dropped dramatically. The loss of enzyme was not reflected in a parallel decrease in yeast growth. When L. starkeyi was grown between a pH of 2.5 and 4.0, both cell yields and growth rates remained the same (Table 1). The purified enzyme did not exhibit a major loss of activity at low pH values (below pH 3.0), whereas the crude enzyme preparation was completely inactivated. This may indicate the presence of an inhibitory factor (i.e., protease) in the crude preparation. There is probably a significant difference

102 103 between the enzyme tertiary structure at pH 2.5 compared to

3.0, which allows for inactivation. Very low pH, below 3.0, is known to unwind proteins (61).

Amylase production is directly related to the temperature of growth in Lipomyces kononenkoae (19).

Temperature had no effect on dextranase production in

L. starkeyi, but did have an effect on growth rate.

L. starkeyi was not able to grow at or above 33 °C. This may be of concern for large scale fermentations with regard to the cost of cooling the reaction vessel.

Oxygen availability was found to affect the amount of enzyme produced by L. starkeyi. Aeration rates above 1.5 and below 1.0 vvm decreased dextranase yields. Lipomyces is an obligate aerobe. Decrease in enzyme production with low sparge rates may be due to a reduction in energy availability, sufficient to reduce enzyme production but not to affect final growth yields. This would account for the slightly lower levels of enzyme produced in during the growth in the 500 1 fermentor. Higher aeration rates may result in oxidation of the dextranase, hence decreasing yields. There is evidence that this enzyme contains at least one sulfhydryl group, which is required for enzyme activity.

Another possibility is that Lipomyces, when grown in a highly aerated condition, could produce extracellular polysaccharides which would change the rheology of the fermentation broth, decreasing the oxygen transfer to the 104 cell. This might not be severe enough to affect the growth

rate but could decrease the enzyme produced. L. starkeyi is

known to produce an extracellular polysaccharide (60) but

further studies will be required to confirm if there is an

effect on fermentation rheology.

A drawback of the Lipomyces dextranase fermentation

system is the requirement for dextran as a growth substrate.

Raw materials are 60 % to 80 % of the production cost in

fermentations, therefore the selling price of the product

will be largely determined by the cost of the carbon source

(78). Therefore,the most useful L. starkeyi strain would be

a constitutive and derepressed mutant, producing dextranase

when grown on a cheap carbon source. A strain that was both

derepressed and constitutive with regards to dextranase

production was not observed during this study. However, a

derepressed hyperproducing mutant was selected.

Induction of extracellular carbohydrases are slightly

more complicated than of intracellular enzymes, due mainly

to the size of the primary substrate. Dextran is too large

to enter the cell, therefore Lipomyces must have a means of

detecting the macromolecule in the environment. This is

usually accomplished by a low constitutive rate of enzyme secretion. In the presence of substrate, low molecular weight products (i. e. the isomaltose series) are formed which accumulate and enter the cell causing the induction of dextranase. 105

A number of potential low molecular weight inducers were screened for activity. The isomalto-series induced dextranase while the malto-series did not. This implied that the inducer must be a 1-6 linked compound. Stachyose, a tetramer consisting of two a 1-6 linked galactopyranosyl units and one glucosyl and fructosyl unit, did not induce dextranase. It appears that the native inducer must be an a 1-6 glucan. The hierarchy of induction of dextranase by the isomalto-series is; isomaltotriose > isomaltopentaose > isomaltotetraose > isomaltose. Usually most low molecular weight compounds are poor inducers of extracellular polysaccharidases. This is thought to be because these compounds are rapidly degraded to glucose, producing catabolite repression (63). Therefore it appears that isomaltotriose could be the natural inducer of dextranase in

L. starkeyi. This was supported by the observation that isomaltopentaose exhibited the second highest induction potential.

The hydrolysis products of the action of dextranase on isomaltopentaose are isomaltotriose and isomaltose. The formation of isomaltotriose correlates to the increase in induction efficiency. In contrast isomaltotetraose is cleaved by dextranase into two molecules of isomaltose, which does not induce dextranase to the same level as isomaltotriose. In fact the level of induction by isomaltotetraose is closer to that of isomaltose than to 106

isomaltotriose.

For most extracellular enzymes, non-metabolizable

(gratuitous) inducers have not been found (63). A gratuitous inducer, B-methyl-glucopyranoside (BMG), was found for dextranase. a-Methyl-glucopyranoside (AMG) also induced dextranase, but it was found to be metabolized by

L. starkeyi. and hence not a true gratuitous inducer. BMG induced dextranase production to three times the level of

AMG. This appeared to be a function of inducer clearance rather than of hyper-induction.

At least two enzymes are required to metabolize dextran, a dextranase and an a-glucosidase. Dextranase will cleave dextran polymer primarily into isomaltotriose (IMT) and isomaltose (IM). Both IMT and IM are probably then transported into the cell and have been show to be inducers of dextranase. These oligosaccharides are also cleaved by a-glucosidase which hydrolyze these sugars to glucose, clearing the inducer. Lipomyces starkeyi produces an a-glucosidase which is cell associated (36).

The effect of inducer clearance can be seen in

Figure 1. AMG, BMG, and dextran mimic each other with respect to induction profile, until 8 hours after the initial exposure to inducer. After this point AMG and dextran induced cells stop producing enzyme. Dextran induced cells produce more enzyme than AMG induced cells, which is probably due to a slight increase in exposure time to the 107 inducer. AMG will enter the cell quickly, and hence be cleared by the a-glucosidase, faster than the induction saccharide formed by the action of dextranase upon dextran.

The dextranase involvement in the production of induction saccharides allows for the distribution of inducer molecules over an extended period of time, increasing the length of exposure of the cells to the inducer, as compared to AMG.

BMG induced cells continued to produce enzyme for 12 hours after addition of inducer. Production probably stops at this point due to energy starvation, i„ e. the exhaustion of the

0.1 % (w/v) glucose carbon source. This is supported by the fact that there were no measurable reducing sugars in the culture broth after 12 hours of incubation in resting cell culture.

All three inducers tested, dextran, AMG, and BMG appear to act in the same fashion. This implied that the limiting factor for induction with all three inducers was the same.

The decrease in final amounts of enzyme produced by dextran and AMG induced cells was due to inducer clearance.

Even though BMG is a gratuitous inducer, it is not an economically viable compound for use in large scale production of dextranase. Therefore dextran was the inducer of choice for the development of the fermentation protocol, for production of dextranase from the derepressed strain of

L. starkeyi (ATCC 20825).

The time of inducer addition during the fermentation 108 was critical for maximum dextranase production. Cells induced in the late exponential phase of growth (cell concentration 8 X 10"7 cells/ml) produced the most dextranase. There were only minor differences in maximum enzyme yields upon doubling the amount of inducer added at this point. This correlated well with the inducer saturation data shown in Figure 2. Optimum production of dextranase, for the derepressed mutant grown on a minimal medium (WW) plus 1.0 % glucose, was when enzyme synthesis was triggered by 0.1 % dextran (inducer) added at a cell concentration of 8 X 107 cells/ml.

Young cells have a higher capacity for enzyme production then older cells. In fact the increase in specific production of dextranase plateaus during the early exponential phase, corresponding with the dramatic decrease in enzyme yields seen at the same point in culture age. The increase in enzyme yields after this point is due to the increase in the number of individuals, not an increase in specific production values. The inability of older cells to produce as much enzyme (specific production) as younger cells may be due to respiratory repression, which is common phenomena in the fungi (5).

It may be possible to utilize induction of young yeast in continuous culture to produce large amounts of enzyme.

By keeping the culture in the young state and adding a continuous influx of inducer, it should be possible to 109 dramatically increase the enzyme yields. The low pH for growth and enzyme production make this type of fermentation an attractive alternative to the batch system.

The batch culture system using the derepressed mutant and induction with dextran was scaled-up from 0.5 1 to

500 1. Very little difference was observed in yeast growth or in the amount of dextranase produced in the 0.5, 20, or

500 liter fermentors. The specific enzyme yield was similar for all trials. Unfortunately it is not possible to compare the production values from Lipomyces to the other fungi that are used to commercially produce dextranase, because they are solid state fermentations (1). Dextranase produced by Lipomyces starkeyi (ATCC 20825) was purified in a five step procedure that gave a 43.3 fold increase in specific activity. The molecular weight of the purified enzyme was found to be 68,000 by gel filtration.

The molecular weight of the protein was also determined by

SDS-PAGE. This technique separated four proteins with the molecular weight ranging from 74,000 to 65,000. These values differ considerably from the value of 23,000 reported by

Webb and Spencer-Martins (92) for their Lipomyces dextranase. Webb and Spencer-Martins (92) also stated that the dextranase they purified showed a single band by

SDS-PAGE and they were able to separate their dextranase on a native nondenaturing PAGE to prove purity (92). The dextranase from Lipomyces (ATCC 20825) did not migrate 110 uniformly in a native PAGE. Separation by the technique of

Webb and Spencer-Martins (92) gave an active protein smear throughout the gel. This sort of interaction has also been reported for the Chaetomium dextranase (92).

To prove purity the protein mixture was separated on an agarose isoelectric focusing gel. This separation technique resulted in the resolution of 5 protein bands, all of which had dextranase activity. All the isoelectric forms were kinetically similar, exhibiting the same apparent

Km. Therefore the difference in physical structure seems to have no significant effect on the catalytic characteristics.

All carbohydrates tested had no effect on enzyme activity. This means that this enzyme is not regulated by product inhibition. Webb and Spencer-Martins had reported the presence of product inhibition with their dextranase (92).

The temperature and pH profiles of Lipomyces (ATCC

20825) dextranase mirrored the profiles reported by Webb and

Spencer-Martins (92). Although the stability of the ATCC

20825 dextranase at 60 °C was poorer than the IGC 4047 enzyme (92). The characteristics of the Lipomyces (ATCC

20825) dextranase with regard to the effects of pH and temperature on activity and stability are very similar to that of the Chaetomium enzyme (29) and the Penicillium enzyme (9).

The purified enzyme showed the same apparent Km values Ill as reported for the IGC 4047 dextranase (92). Both exhibited catalytic constants consistent with an endodextranase.

Although there is a significant difference between the enzymes' molecular weights there does not appear to be catalytic difference between the two.

If the dextranase eluted from a CM-Sepharose column is desalted by dialysis, an active enzyme aggregate is formed.

This aggregate is tightly bound , exhibiting a molecular weight of over 700,000. The aggregate will not dissociate upon the addition of salt. It will dissociate upon the addition of SDS, although the dissociated enzyme is not active. The aggregate form does not show any difference of catalytic characteristics (Km) compared to the native

(68,000) form.

The ATCC 20825 dextranase was a glycoprotein containing

8 % sugar. This sugar was determined to contained a large amount of mannose by lectin reactivity. Differing substitutions of the carbohydrate portion of this enzyme may be an explanation for the multiple molecular weights and isoelectric forms observed. Penicillium (14) and Chaetomium

(29) dextranases are both glycoproteins. The Penicillium enzyme has also been shown to contain a large amount of mannose in its carbohydrate moiety (14).

The amino acid composition of Lipomyces (ATCC 20825) dextranase is compared to other fungal dextranases in Table

11. Overall the dextranase from Lipomyces is more similar to 112 Chaetomium than to Penicillium. Lipomyces dextranase is very high in glycine residues. This means that this enzyme has a large area of relative neutral residues. The function of this region is unknown. The ratio of

(asx + glx) to (arg + lys + his), which is a measure of the overall charge of the protein, was 0.96. This value is typical of a basically charged protein. This enzyme does not appear to have any disulfide links which are important for activity, represented by the fact that glutathione has no effect on activity. Sulfur containing amino acid residues are few, but they are required for activity. This is also a characteristic of both the Chaetomium (29) and Penicillium

(9) dextranases.

Purified Lipomyces dextranase was completely inactivated by exposure to mercuric chloride. It was possible to reactivate the enzyme after exposure to the mercuric salt by addition of 2-mercaptoethanol. The amino acid composition shows four cystine residues. The

Penicillium enzyme also has a requirement of a sulfhydryl group for activity (9). Mercuric chloride had the same effect on the dextranase from Chaetomium (29).

Only three of the seven proteases tested had any effect on enzyme activity. The possible explanation for this was protection of the effected residues by the carbohydrate moiety. Another explanation might involve the tertiary structure of this protein. Folding of this basic protein may 113 allow for protection of the effected amino acids. In support of this view are the stability characteristics of the pure versus crude enzyme at pH 2.5. Of proteases that did effect activity, ficin and protease XXI had the most deleterious effect. Ficin attacks amines, esters, and has an action similar to papain. Papain had no effect on activity. This possible conflict could be due to differences in the sizes of the proteases and their ability to reach their substrate. Protease XXI has an action that mimics trypsin.

Protease XXI inhibited dextranase activity while trypsin did not. Again this may be due to the differences between the proteases with regard to their ability to contact their respective substrate.

The specificity of the endodextranase of L. starkeyi seems very similar to those of Penicillium (9), Aspergillus

(30), Chaetomium (29) and the Lipomyces enzyme reported by

Webb and Spencer-Martins (92). Following the progression of reaction products with time, it is apparent that the minimum glucan chain length that will be hydrolyzed by the enzyme is four, although isomaltotetraose (DP4) was not hydrolyzed as quickly as was isomaltopentaose (DP5). This observation is in agreement with the results given for other endodextranase (9,29,30,92).

Salts of importance to sugar processing which inhibit enzyme activity are CaCl2, CuS04, FeCl2, and MgCl2. All these salts are found in sugar juice or will be present in 114 the process stream (Personal communication, Dr. M. Saska,

Audubon Sugar Inst., LSU). Sodium fluoride did not have any effect upon enzyme activity. Most toothpastes have this salt as an active ingredient. Dextranase has been shown to remove dental plaque when incorporated into toothpaste (68,85).

SDS, a common ingredient in toothpaste, inactivates dextranase. It may be possible to stabilize this dextranase against SDS by addition of some detergent

(i. e. Triton X-100). There are reports of the stabilization of the Penicillium dextranase against SDS with the use of bentonite (47).

Dextranase from Lipomyces showed an increase in activity in 5 % ETOH solutions and retained 100 % activity up to and including 10 % ETOH solutions. This is also of importance for the application of dextranase in dental hygiene. One of the problems associated with the use of dextranase in toothpastes is the length of time the enzyme would be exposed to the dental plaque. Incorporation of the enzyme in a mouthwash would allow for an increased exposure time to substrate (dental plaque). The incorporation of enzymes in aqueous-organic cosolvent mixtures has been shown to have a positive influence on catalytic pH and temperature profiles (15,82).

The pH of the last aqueous solution from which the enzyme was recovered affects the pH of the micro-aqueous environment of the enzyme in the cosolvent system and hence 115 the charge balance of the protein (82). In fact there have been reports of the broadening of the pH rate profile with the use of cosolvent systems (82). This may occur with dextranase in such solvent systems but there is no hard data for confirmation. Incorporation of dextranase in an

ETOH-water mixture then might increase the effective pH range allowing for a more desirable product for the removal of dental plaque.

The temperature optimum of the Lipomyces dextranase in cane juice was 10 °C lower than the reported values for both the Miles DEXTRANEX (51) and the Novo DN 25L (47). Lipomyces dextranase had a temperature optimum of 50-60 °C when assayed in a defined reaction mixture, hence there would seem that there was a deleterious influence on enzyme stability associated with the components of fresh cane juice. There have been reports of enzyme inhibitors associated with cane juice (45), which may explain the decreased temperature optimum observed.

The low temperature optimum found with Lipomyces dextranase will not be a problem for treating cane juice prior to clarification since the temperature of the juice is rarely above 40 °C. But it could be a problem for the treatment in syrup, where the temperatures are usually above

70 °C.

In the mill trial, the clarified juice showed a 45 % decrease in dextran after exposure to dextranase for one 116 hour. Extrapolating from the dose calibration curve shown in

Figure 20 a 22 % reduction in dextran was expected.

Approximately twice as much dextran was removed than expected in the mill trial. This could be due to differences in the lots of cane, since the laboratory and mill trials were harvested separately. In fact the laboratory test cane was contaminated with dextran before milling. The mill trial cane was clean (no detectable dextran). Furthermore the mud content was very high in the laboratory cane juice as compared to juice used in the mill trial. This observation may support a role for reported enzyme inhibitors associated with soiled cane (45). It might be construed therefore that the dosing estimations are conservative, although it is more likely that the laboratory results more closely parallel a

"real-world" situation found with stale cane. Further mill trials utilizing naturally contaminated cane will be required to confirm this.

The untreated juice was impossible to boil and could not be adequately purged. The small amount of sugar that could be recovered was dark in color and needle grained

(Figure 21B). The dextranase treated material boiled normally and the sugar produced was both square and light in color (Figure 21A). Dextran analysis of the syrup and the sugar produced demonstrated the effectiveness of the enzyme in removing dextran from cane juice. A 76 % reduction in dextran content, by Haze, was seen between the treated and 117 untreated sugars. By increasing the dextranase dose or holding time it should be possible to make sugar with a dextran content below the penalty level using Lipomyces dextranase, making the use of dextran contaminated cane supply economically feasible. CONCLUSION

Dextranase is an enzyme which destroys dextran, a contaminant in sugar and a main constituent of dental

plaque. Using dextranase can solve a major economic problem

in the sugar industry, and in dental preparations it can keep teeth essentially free of plaque. The enzyme has been used successfully outside the U. S. In the U. S. the FDA has not approved the use of dextranase due to the toxic attributes of the fungi which are used to commercially produce dextranase. L. starkeyi which has been used in food related process, also produces a dextranase. The probability of FDA approval of dextranase produced by L. starkeyi is higher than for other fungal sources. In fact the FDA has already given verbal approval for large scale test trials of

Lipomyces dextranase in a sugar mill.

The program objectives of this study were to optimize fermentation parameters for production of dextranase by

L. starkeyi, purify and characterize the enzyme, and test dextranase in a pilot trial. Lipomyces has proven itself with regard to the production of enzyme as an easily maintained system that allows for the large scale production of dextranase. Mutants of this yeast were selected which were derepressed with regard to dextranase production, allowing for the growth of L. starkeyi on a carbon source other than dextran. These strains still require an inducer

118 119

for the production of dextranase. The ideal strain will be a

constitutive derepressed mutant.

Studies on the physical and enzymological

characteristics of dextranase from L. starkeyi show that it

is very similar to both commercially available dextranases.

This enzyme is inhibited by salts which are commonly found

in sugar mill streams. This could be a major drawback of

this enzyme and more application trials are required.

Application of dextranase to dextran contaminated cane

juice showed the effectiveness of this enzyme in sugar

processing. The institution of dextran penalties by refiners

has intensified the interest in dextranase on the part of

Louisiana cane processors and others. While field and factory managements can minimize the occurrence of dextran through suitable operating practices, they can not eliminate it. The commercialization of a dextranase for the U. S. sugar industry is a much desired goal. L. starkeyi dextranase can meet this need. LITERATURE CITED

1. Akabori, T., M. Hotta and K. Obayashi. 1976. Dextranase. Japan. Kokai 78,470,587. Int. CL C12 D1 3/10. 4 p.

2. Anderson, J. R. and E. Drew. 1972. Growth Characteristics of a Species of Lipomyces and its Degradation of Paraquat. J. Gen. Microbiol. 70:43-58.

3. Bailey, R. B. and A. Woodword. 1984. Isolation and Characterization of a Pleiotropic Glucose Repression Resistant Mutant of Saccharomyces cerevisiae. Mol. Gen. Genet. 193:507-512.

4. Barford, J. P. 1985. Control of Fermentation and Respiration in Schizosaccharomyces pombe. J. Ferment. Technol. 63(3):495-500.

5. Barfoed, S. and A. Mollgaard. 1987. Dextranase Solved Dextran Problems in DDS' Beet Sugar Factory. Zuckerindustrie (Berlin). 112(5):391-395.

6. Barrett, J F., T. A. Barrett and R. Curtiss III. 1987. Purification and Partial Characterization of the Multicomponent Dextranase Complex of Streptococcus sorbis and the Cloning of the Dextranase Gene. Infect. Immun. 55(3):729-802.

7. Bhanot, P. and R. G. Brown. 1980. Effect of 3-O-methyl-D-glucose on the Production of Glycosides by Cryptococcus laurentii and Saccharomyces cerevisiae. Can. J. Microbiol. 26:1289-1295.

8. Brown, R. G. and J. M. Embril. 1985. Effect of 3-O-Substituted Analogues of D-glucose and D-allose on Amino Acid Uptake by Penicillium sp. Can. J. Microbiol. 31:513-518.

9. Chaiet, L., A. J. Kempf, R. Harman, E. Kaczka, R. Weston, K. Nollstadt and F. J. Wolf. 1970. Isolation of a Pure Dextranase from Penicillium funiculosum. Appl. Microbiol. 20(3):421-426.

10. Ciriacy, M. 1977. Isolation and Characterization of Yeast Mutants Defective in Intermediary Carbon Metabolism and in Carbon Catabolite Derepression. Mol. Gen. Genet. 154:213-220.

120 121

11. Clarke, M. A. and F. G. Carpenter. 1977. Special Techniques for Analysis of Sugars and Nonsugars. In: Cane Sugar Handbook. Meade and Chen, eds. Wiley-Interscience, New York. 734-757.

12. Covacenich, M. T. and G. N. Richards. 1977. Studies on Dextran Isolated from Raw Sugar Manufactured from Deteriorated Cane. Proc. ISSCT. 10:3-9.

13. Cullimore, D. R. and M. Woobine. 1961. A Superior Fat-synthesizing Single-spore Strain of Lipomyces. Nature. 190:1022-1024.

14. Danilova, T. I., V. I. Maksimov, A. I. Chukhrova and G. A. Molodova. 1983. Carbohydrate Component of Dextranase from Penicillium funiculosium. Prikl. Biokhim. Mikrobiol. 19(1):104-110.

15. Deetz, J. S. and J. D. Rozzell. 1988. Enzyme-Catalysed Reactions in Non-Aqueous Media. Trends Biotechnol. 6:15-19.

16. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers and F. Smith. 1956. Colorimetric Method for the Determination of Sugars and Related Substances. Anal. Chem. 28:350-356.

17. Emeis, C., inventor. 1970. Verfehren zur Gewinnung von Poly- und Oligosaccharide spaltenden Enzymen unter Verwendung von Hefen. German (Fed. Rep.) Offenlegungsschirift 1,908,225. 17 p.

18. Entian, K. D. 1981. A Carbon Catabolite Repression Mutant of Saccharomyces cerevisiae with Elevated Hexokinase Activity: Evidence for Regulatory Control of Hexokinase PII Synthesis. Mol. Gen. Genet. 184:278-282.

19. Estrela, A. I., M. Lemos and I. Spencer-Martins. 1982. A Note on the Effect of Growth Temperature on the Pre"action of Amylases by the Yeast Lipomyces kononenkoae. J. Appl. Bacteriol. 52:465-467.

20. Felgenhauer, B. and K. Trautner. 1982. A Comparative Study of Extracellular Glucanhydrolase and Glucosyltransf erase Enzyme Activities of 5 Different Serotypes of Oral Streptococcus mutans. Arch. Oral Biol. 27:455-461. 122

21. Felgenhauer, B. and K. Trautner. 1983. The Reciprocal Action of Extracellular Streptococcus mutans Glucotranserase and Glucohydrolase. J. Dent. Res. 62:459.

22. Fitzgerald, R. J. , P. H. Keys, T. H. Stoudt and D. M. Spinell. 1968. The Effects of a Dextranase Preparation on Plaque and Caries in Hamsters, a Preliminary Report. JADA. 76:301-304.

23. Foster, D. H. , P. A. Inkerman and K. E. Mc Neil. 1980. Studies on Cane Deterioration in Australia. Proc. ISSCT. 13:2204-2213.

24. Fuji Oil Co. Ltd., assignee. 1975. Cocoa Butter Substitute by Fermentation. Netherlands. NL 76,041,177. 26 p.

25. Fulcher, R. P. and P. A. Inkerman. 1976. Dextranase. I. Characterization of the Enzyme for Use in Sugar Mills. Proc. Qd. Soc. Sug. Cane Technol. 43rd Conf. 295-305.

26. Gancedo, J.M. and C. Gancedo. 1986. Catabolite Repression Mutants of Yeast. FEMS Microbiol. Rev. 32:179-187.

27. Gierhart, D., inventor; CPC International, Inc., assignee. 1982. Multistage Process for the Preparation of Fats and Oils. UK GB 2,091,285. Int. CL3 c 12P 7/64, C5C 6C2. 9p.

28. Guerzoni, M. E., P. Lambertini, G. Lercker and R. Marchetti. 1985. Technological Potential of Some Starch Degrading Yeasts. Starch/Staerke. 37:52-57.

29. Hattori, A., K. Ishibashi and S. Minato. 1981. The Purification and Characterization of the Dextranase of Chaetomium gracile. Agric. Biol. Chem. 45(11)-.2409-2416.

30. Hiraoka, N., H. Tisuji, J. Funkumoto, T. Yamamoto and D. Tsuru. 1973. Purification and Characterization of the Dextranase of Aspergillus carneus. J. Peptide Protein Res. 5:161-170.

31. Inkerman, P. A. and L. Riddell. 1977. Dextranase. III. Refinements to the Enzymatic Process for the Treatment of Deteriorated Cane. Proc. Qd. Soc. Sug. Cane Technol. 44rd Conf. 215-223. 123 32. Inkerman, P. A. 1980. An Appraisal of the Use of Dextranase. Proc. ISSCT. 13:2411-2427.

33. Inkerman, P. A. and G. P. James. 1976. Dextranase II. Practical Application of the Enzyme to Sugar Mills. Proc. Qd. Soc. Sug. Cane Technol. 43rd Conf. 307-315.

34. Janson, J. and J. Porath. A Bacterial Dextranase. Meth. Enz. 8:615-621.

35. Kaster, A. G. and L. R. Brown. 1983. Extracellular Dextranase Activity Produced by Human Oral Strains of the Genus Bifidobacterium. Infect. Immun. 42(2):716-720.

36. Kelly, C. T., M. E. Moriarty and W. M. Fogarty. 1985. Thermostable Extracellular a-Amylase and a-Glucosidase of Lipomyces starkeyi. Appl. Microbiol. Biotechnol. 22:352-358.

37. Koboki, Y., D. Fujimoto, H. Aoki and K. Fujita, inventors. 1987. Toothpaste Containing Dextranase Bound Hydroxylapatite. Japan. Kokai 62 67,013 [87 67,013] (CL.A61 K7/20) 26Mar 1987, Appl. 85/206,678 20. 4 p.

38. Laires, A., I. Spencer-Martins and N. van Uden. 1983. Use of D-Glucosamine and 2-Deoxyglucose in the Selective Isolation of Mutants of the Yeast Lipomyces starkeyi Derepressed for the Production of Extracellular Endodextranase. Z. Allg. Mikrobiol. 23(9):601-603.

39. Laemmli, U. K. 1970. Cleavage of Structural Proteins During the Assembly of the Head of Bacteriophage T4. Nature. 227:680-685.

40. Laurent, T. C. and J. Killander. 1964. A Theory of Gel Filtration and its Experimental Verification. J. Chromatog. 14:317-330.

41. Lineweaver, H. and D. Burk. 1934. The Determination of Enzyme Dissociation Constants. J. Am. Chem. Soc. 56:658-666.

42. Lobo, Z. and P. K. Maitra. 1977. Resistance to 2-Deoxyglucose in Yeast: A direct Selection of Mutants Lacking Glucose-Phosphorylating Enzymes. Molec. Gen. Genet. 157:297-300. 124 43. Loesche, W. J. 1986. Role of Streptococcus mutans in Human Dental Decay. Microbiol. Rev. 50(4):353-380.

44. Lowry, 0. H., J. Rosenbrough, A. L. Farr and R. J. Randall. 1951. Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem. 193:265-275.

45. Madhu, G., L. Shukla and K. A. Prabhu. 1984. Application of Dextranase in the Removal of Dextran from Cane Juice. Int. Sugar J. 86:136-138.

46. Mahlberg, D., M. Hofer and A. Tauber. 1985. Sugar Transport and Hexose-ATP-Kinase Activity in a 2-Deoxy-D-glucose Tolerant Mutant of the Yeast Rhodotorula qlutinis. J. Gen. Microbiol. 131:479-485.

47. Maksimov, V. I., E. I. Belen Kaya, N. N. Sazonova and G. A. Molodova. 1982. Stabilization of a Dextranase of Penicillium funiculosum. Otkrytiya Izobret. Prom. Obraztsy. Tovarnye Znaki. 39:136-140.

48. McCann, A. K., F. Hiberg, P. Kenworthy and J. A. Barnett. 1987. An Unusual Hexose-ATP-Kinase with Two Catalytic Sites and a Role in Carbon Catabolite Repression in the Yeast Schwanniomyces occidentalis. J. Gen. Microbiol. 133:381-389.

49. McCann, A. K. and J. A. Barnett. 1984. Starch Utilization by Yeasts: Mutants Resistant of Carbon Catabolite Repression. Curr. Genet. 8:525-530.

50. McGinnis, J. F. and K. Paigen. 1969. Catabolite Inhibition: A General Phenomenon in the Control of Carbohydrate Utilization. J. Bacteriol. 100:902-913.

51. Miles. 1983. Dextranex. Fungal Dextranase for the Sugar Cane Industry. Technical information L-1475. Miles Laboratories, Elkhart, Indiana, U. S. A. lip.

52. Montreuil, J., S. Bouquelet, H. Y. Debray, B. Fournet, G. Spik and G. Strecker. Glycoproteins. Chapter 5. In: Carbohydrate Analysis: A Practical Approach. Chaplin and Kennedy, eds. IRL Press, Washington, D. C. 143-204.

53. Moulin, G. and P. Galzy. 1979. Study of an Amylase and Its Regulation in Lipomyces starkeyi. Agric. Biol. Chem. 43(6):1165-1171. 125

54. Naganuma, T., Y. Uzuka and K. Tanaka. Physiological Factors Affecting Total Cell Number and Lipid Content of the Yeast, Lipomyces starkeyi. J. Gen. Appl. Microbiol. 31:29-37.

55. Naganuma, T. , Y. Uzuka, K. Tanaka and H. Iizuka. 1987. Differences in Enzyme Activities of Lipomyces starkeyi Between Cells Accumulating Lipid and Proliferating Cells. J. Basic Microbiol. 27(1):35-42.

56. Neely, W. B. and J. Nott. 1962. Dextransucrase, an Induced Enzyme from Leuconostoc mesenteroides. Biochemistry, 1:1136-1140.

57. Novo. 1977. Dextranase Novo 25L - A Dextranase Decomposing Enzyme for the Sugar Industry. Product Data Information 112b - GB. Novo Enzyme Division, Hagsvaerd, Denmark. 7 p.

58. Panek, A. D. , A. L. Sampaio, G. C. Braz and J. R. Mattoon. 1978. Regulation of Energy Metabolism in Yeast. Relationships Between Carbohydrate Reserves, Catabolite Repression and Maltose Utilization. In; Biochemistry and Genetics of Yeasts. Skinner, Passmore, and Davenport, eds. Academic Press, New York. 145-159.

59. Parker, R. B. and H. R. Creamer. 1971. Contribution of Plaque Polysaccharides to Growth of Cariogenic Microorganisms. Arch. Oral Biol. 16:855-862.

60. Phaff, H. J. and C. P. Kortzman. 1984. Genus 14. Lipomyces Lodder et Kreger-van Rij. In: The Yeast a Taxonomic Study. Kreger-van Riji, ed. Elsevier, Amsterdam. 252-262.

61. Picatagglo, S. and R. Smittle, inventors; CPC International, Inc., assignee. 1979. Process for the Microbiological Production of Oil. European 0,005,277. Int. CI2 C 12 D 13/08, C 07 C 67/00, C 12 B 3/12. 21 p.

62. Polack, J. A. and H. S. Birkett. 1981. The Use of Dextranase in Processing Sour Cane. ASSCT. 8:127-133.

63. Priest, F. G. 1984. Extracellular Enzymes. Aspects of Microbiology 9. American Society for Microbiology, Washington, D. C. 79 p. 126

64. Ramos, A. and I. Spencer-Martins. 1983. Extracellular Glucose-Producing Exodextranase of the Yeast Lipomyces lipofer. Antonie van Leeuwenhoek. 49:183-190.

65. Reese, E. T. , J. E. Lola and F. W. Parrish. 1969. Modified Substrates and Modified Products as Inducers of Carbohydrase. J. Bacteriol. 100(3):1151-1154.

66. Richards, G. N. and M. Streamer. Studies on Dextranases. Part IV. Mode of Action of Dextranase Di on Oligosaccharides. Carbohydr. Res. 32:251-260.

67. Rosmann, M. G. and P. Argos. 1981. Protein Folding. Ann. Rev. Biochem. 50:497-532.

68. Saito, K. 1982. Application of Dextranase to Dentistry. Fureguransu Janaru. 10(2):101-107.

69. Saito, K. 1982. Inhibitory Effects of Dextranase Produced by Chaetomium gracile on the Development of Dental Plaque and Carries Occurrence in Rats. Koku Eisei Gakai Zasshi. 32(2):147-166.

70. Sarkar, D. and D. F. Day. 1986. Dextran Analysis: A Modified Method. ASSCT. 6:102-107.

71. Sawai, T., S. Ohara, Y. Ichimi, S. Okaji, K. Hisada and N. Fukaya. 1981. Purification and Some Properties of the Isomaltodextranase of Actinomadura Strain R10 and Comparison with that of Arthrobacter globiformis T6. Carbohydr. Res. 89:289-299.

72. Sayama, K., T. Kamada, T. Muratsubaki and K. Honjo. 1985. Decreased in Filtration Velocity by Polysaccharides and it's Increment by Dextranase Treatment in Beet Sugar Manufacturing Process. Nippon Nogei Kagaku Kaishi. 59(2):143-151.

73. Simonson, L. G., A. E. Liberta and A. Richardson. 1975. Characterization of an Extracellular Dextranase from Fusarium moniliforme. Appl. Microbiol. 30(5):855-861.

74. Spencer-Martins, I. and N. van Uden. 1977. Yields of Yeast Growth on Starch. European J. Appl. Microbiol. 4:29-35. 127

75. Spencer-Martins, I. 1982. Extracellular Isoamylase Produced by the Yeast Lipomyces kononenkoae. Appl. Environ. Microbiol. 44(6):1253-1257.

76. Staat, R. H. and C. F. Schachtele. 1974. Evaluation of Dextranase Production by the Cariogenic Bacterium Streptococcus mutans. Infect. Immun. 9:467-469.

77. Staat, R. H., T. H. Gawronski and C. F. Schachtele. 1973. Detection and Preliminary Studies on Dextranase-Producing Microorganisms From Human Dental Plaque. Infect. Immun. 8:1009-1016.

78. Stanbury, P. F. and A. Whitaker. 1984. Principles of Fermentation Technology. Peramon Press, New York. 255 p.

79. Starkey, R. 1946. Lipid Production by a Soil Yeast. J. Bacteriol. 51:33-50.

80. Sumner, J. B. 1924. A Method for the Measure of Reducing Sugars in Urine. J. Biol. Chem. 62:287-290.

81. Takagi, S., M. Shiota, Y. Mitsuishi, M. Kobayashi and K. Matsuda. 1984. An Isomaltotriose-Producing Dextranase From Flavobacterium sp. M-73: Action Pattern of the Enzyme. Carbohydr. Res. 129:167-177.

82. Tan, K. H. and R. Lovien. 1972. Enzymology in Aqueous-Organic Cosolvent Binary Mixtures. J. Biol. Chem. 247(10):3278-3285.

83. Tilbury, R. H. and S. M. French. 1974. Further Studies on Enzymatic Hydrolysis of Dextran in Mill Juices by Dextranases and Fungal a-Amylases. Proc. ISSCT. 15:1-10.

84. Tilbury, R. H. 1972. Dextrans and Dextranase. Proc. ISSCT. 14:1444-1458.

85. Unpublished Notes. 1982. Clinca Lion. Sankyo Co., Ldt., Tokyo, Japan. 30 p.

86. Uzuka, Y., T. Naganuma, K. Tanaka and K. Suzuki. 1985. Relation Between Neutral Lipid Accumulation and the Growth Phase in the Yeast, Lipomyces starkeyi, a Fat Producing Yeast. Agric. Biol. Chem. 49(3):851-852. 128

87. Uzuka, Y., T. Naganuma and K. Tanaka. 1974. Influence of Culture pH on the Initiation of Biotin Biosynthesis in a Strain of Lipomyces starkeyi. J. Gen. Appl. Microbiol. 20:277-287.

88. Uzuka, Y., T. Naganuma, K. Tanaka and Y. Odagiri. 1974. Effect of Culture pH on the Growth and Biotin Requirement in a Strain of Lipomyces starkeyi. J. Gen. Appl. Microbiol. 20:197-206.

89. van Uden, N. , C. Cabeca-Silva, A. Madeira-Lopes and I. Spencer-Martins. 1980. Selective Isolation of Derepressed Mutants of an a-Amylase Yeast by the Use of 2-Deoxyglucose. Biotechnol. Bioeng. 22:651-654.

90. Vane, G. W. 1981.The Effect of Dextran on the Distortion of Hard Candy. Sugar Ind. Technol. 40:95-102.

91. Walker, G. J. and A. Pulkownik. 1974. Action of a-l,6-Glucan Glucohydrolase on Oligosaccharides Derived from Dextran. Carbohydr. Res. 36:53-66.

92. Webb, E. and I. Spencer-Martins. 1983. Extracellular Endodextranase from the Yeast Lipomyces starkeyi. Can. J. Microbiol. 29:1029-1095.

93. Workman, W. and D. Day. 1984. The Cell Wall Associated Inulinase of Kluyveromyces fragilis. AAntonie van Leeuwenhoek. 50:349-353.

94. Zimmermann, F. K. and I. Scheel. 1977. Mutants of Saccharomyces cerevisiae Resistant to Carbon Catabolite Repression. Molec. Gen. Genet. 154:75-82. VITA

David W. Koenig was born in Lakewood, Ohio on September

25, 1957. He was educated in both the Ohio and Michigan school systems, and graduated from high school in 1975. He attended Bowling Green State University in Bowling Green,

Ohio, and recieved a B. S. in Biology in 1979. He entered the University of Mississippi in the fall of 1979 and recieved a M. S. degree in Biology in 1982. In August 1984 he began his graduate training at Louisiana State

University, and is a candidate for the degree of Doctor of

Philosophy in Microbiology.

129 DOCTORAL EXAMINATION AND DISSERTATION REPORT

Candidate: David William Koenig

Major Field: Microbiology

Title of Dissertation: The Dextranase of Lipomyces starkeyi and Its Use in Sugar Cane Processing.

Approved:

Major Professor an Chairman

Dean of the Graduate Sc!

EXAMINING COMMITTEE:

lUD

-UL2—U

Date of Examination:

Thursday, July 14, 1988