Micrcffilms International 300 N

Home , Glucosyltransferase, Invertase, Nigerose

INFORMATION TO USERS

This reproduction was made from a copy of a document sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this document, the quality of the reproduction is heavily dependent upon the quality of the material submitted.

The following explanation of techniques is provided to help clarify markings or notations which may appear on this reproduction.

1.The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity.

2. When an image on the film is obliterated with a round black mark, it is an indication of either blurred copy because of movement during exposure, duplicate copy, or copyrighted materials that should not have been filmed. For blurred pages, a good image of the page can be found in the adjacent frame. If copyrighted materials were deleted, a target note will appear listing the pages in the adjacent frame.

3. When a map, drawing or chart, etc., is part of the material being photographed, a definite method of "sectioning" the material has been followed. It is customary to begin filming at the upper ):;ft hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete.

4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services Department.

5. Some pages in any document may have indistinct print. In all cases the best available copy has been filmed.

University Micrcffilms International 300 N. Zeeb Road Ann Arbor, Ml 48106

8526145

Bhattacharjee, Mrinal Kanti

THE STUDY OF DONOR AND ACCEPTOR SUBSTRATE ANALOGS OF DEXTRANSUCRASE

The Ohio State University Ph.D. 1985

University Microfilms International 300 N. Zeeb Road, Ann Arbor, Ml 48106

PLEASE NOTE:

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .

1. Glossy photographs or pages

2. Colored illustrations, paper or print

3. Photographs with dark background ^

4. Illustrations are poor copy

5. Pages with black marks, not original copy

6. Print shows through as there is text on both sides of page

7. Indistinct, broken or small print on several pages S

8. Print exceeds margin requirements

9. Tightly bound copy with print lost in spine

10. Computer printout pages with indistinct print

11. Page(s) lacking when material received, and not available from school or author.

12. Page(s) seem to be missing in numbering only as text follows.

13. Two pages numbered . Text follows.

14. Curling and wrinkled pages

15. Other

University Microfilms International

THE STUDY OF DONOR AND ACCEPTOR SUBSTRATE

ANALOGS OF DEXTRANSUCRASE

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the

Degree Doctor of Philosophy in the Graduate School of

The Ohio State University

Mrinal K. Bhattacharjee, B.Sc., M. Sc.

The Ohio State University

1985

Reading Committee : jroved^by

Prof. M. M. King

Prof. R. M. Mayer Advi ser Prof. M-D. Tsai Department of Cheml DEDICATION

To my parents

ii ACKNOWLEDGEMENT

I wish to express my deepest gratitude to my adviser, Dr. Robert M.

Mayer, for his guidance and support throughout this study. I also like to thank my colleagues in the research group for their helpful suggestions.

I am grateful to Dr.Irwin Goldstein, Dr. Derek Horton and

Dr. Howard Sloan for providing some of the sugars and Dr. David A.

Grahame for providing highly purified dextransucrase.

Thanks are due to Dr. C. R. Weisenberger for obtaining the mass spectra. I wish to acknowledge Robert Daiber, Rebecca Jezerinac,

Vincent Leber, Stella Sung and Robert Wheatley for their assistance in certain technical aspects of the research.

i i i VITA

June 04, 1958 Born - Dhanbad, India

1978 B.Sc., Indian Institute of Technology, Kharagpur

1980 M.Sc., Indian Institute of Technology, Kharagpur

1980-1982 Teaching Associate, Chemistry Department, The Ohio State University, Columbus, Ohio

1982-1985 Research Associate, Chemistry Department, The Ohio State University, Columbus, Ohio

PUBLICATION

"Interaction of Deoxy Halo Derivatives of Sucrose with Dextransucrase", Bhattacharjee, M. K. and Mayer, R. M., Carbohydr. Res., in press.

FIELD OF STUDY

Major Field : Biochemistry

i v TABLE OF CONTENTS

Page

DEDICATION ii

ACKNOWLEDGEMENTS iii

VITA iv

LIST OF TABLES viii

LIST OF FIGURES ' ix

LIST OF ABBREVIATIONS x

CHAPTERS

I. INTRODUCTION 1

A. Structure of Dextran 3 B. Purification and Properties of Dextransucrase ... 7 C. Mechanism of Dextransucrase Reaction 9 1. Chain Initiation 9 2. Chain Propagation 12 a. Mode of Elongation 12 b. Direction of Chain Growth 13 c. Formation of Branches 20 d. Mechanism of Catalysis 21 i. Carbonium Ion Mechanism 22 ii. Covalent Ion Mechanism 24 iii. Hydrolysis 26 3. Chain Termination 27 D. Reaction Pathways of Dextransucrase 28 E. Substrate Specificity 29 1. Donor Substrate Specificity 31 2. Acceptor Substrate Specificity 33 F. Purpose of Investigation 34

II. MATERIALS AND METHODS 35

A. Materials 35 1. Organisms 35 2. Enzymes 35

v CONTENTS (continued) Page

3. Saccharides and Derivatives 36 4. Chromatography Materials 37 5. Chemicals and Reagents 37 B. Methods 38 1. Measurement of Dextransucrase Activity .... 38 a. Coupled Enzyme Assay 38 b. Radioactive Assay 39 2. Reactivation of SDS Inhibited Enzyme .... 40 3. Assay of Acceptor Activity 41 4. Inactivation of Enzyme . 42 a. Chemical Inactivation 42 b. Photochemical Inactivation 42 5. Chromatographic Methods 42 a. Thin Layer Chromatography 43 b. Paper Chromatography 43 c. Column Chromatography 45 i. Silica Gel 45 ii. Ion Exchange 45 iii. Gel Filtration 46 d. High Performance Liquid Chromatography . . 47 6. Radioactive Analysis 47 7. Spectroscopic Methods 48 a. *-H NMR Spectroscopy 48 b. C NMR Spectroscopy 48 c. UV/VIS Spaectroscopy 48 d. IR Spectroscopy 49 e. Mass Spectroscopy 49 8. Saccharide Analysis 49 9. SDS El ectrophoresis 49 10. General Synthetic Procedures 51 a. Preparation of Dry Solvents and Reagents . . 51 b. Halo Derivatives of Sugars 52 c. Methyl Glycosides 54 d. Acetyl ation Reactions 56 e. Deacetyl ation Reactions 56 11. Chemical Synthetic Procedures 56 a. 6,6'-dichloro-6,6'-dideoxy-sucrose .... 56 b. 6,6'-dibromo-6,6'-dideoxy-sucrose .... 58 c. Monobromosucroses 59 d. 6,1',6'-tribronio-6,r,6,-trideoxy-sucrose . 59 e. Methyl,6-chloro-6-deoxy-a-(D)-glucopyranoside 62 f. Methyl,6-bromo-6-deoxy-a-(D)-glucopyranoside 65 g. Methyl,6-iodo-6-deoxy-a-(D)-gl ucopyranoside 65 h. Methyl,6-bromo-6-deoxy-a-(D)-galactopyranoside 65 i. Methyl.6-bromo-6-deoxy-a-(D)-mannopyranoside 66 j. Methyl ,6-deoxy-a-(D)-gal actosides .... 69 k. Methyl ,6-deoxy-a-(L)-mannosides .... 72 1. Methyl-(D)-al 1 osides 72

vi CONTENTS (Continued) Page

m. Methyl,6-azido-6-deoxy-a-(D)-gl ucopyranoside 73 n. Methyl,3,6-anhydro-a-(D)-glucopyranoside . . 76 o. Methyl,2-deoxy-a-(D)-gl ucopyranoside ... 77 p. (D)-glucal 78 q. 6-deoxy-(L)-gl ucal 78 r. Methyl,2,3,6-trideoxy-a-(L)-erythro- hex-2-enopyranoside ... 79 s. 6'-bromo-6'-deoxy-maltose 79 12. Enzymatic Synthetic Procedures 87

III. RESULTS AND DISCUSSIONS 91

A. Interaction of Halosucrose Derivatives with Dextransucrase 91 B. Acceptor Substrate Specificity : Analogs of Methyl,a-(D)-glucopyranoside . . . 115 1. Acceptor Reactions 115 2. Unsaturated Derivatives 138 3. Photochemical Inactivation 139 C. Acceptor Substrate Specificity : Oligosaccharides . . 146 D. Acceptor Substrate Specificity : Branch Formation . . 167 1. Preparative Synthesis of Products with Methyl,a-(D)-gl ucopyranoside as Acceptor . . . 167 2. Preparative Synthesis of Products with Maltose as Acceptor .... 175 3. Preparative Synthesis of Products with 6'-bromo-6'-deoxy-maltose as Acceptor .... 178 4. Methyl,6-bromo-6-deoxy-a-(D)-glucopyranoside as Acceptor . . 183 5. Preparative Synthesis of Products with Methyl, 6-bromo-6-deoxy-a-(D)-glucopyranoside as Acceptor 191 6. Preparative Synthesis of Products with Nigerose as Acceptor .... 207

IV. CONCLUSION 219

V. BIBLIOGRAPHY 221

vii LIST OF TABLES

Table Page 1. Chemical shifts of synthesized sugars Ill

2. Comparison of acceptor activities of analogs of a-methyl glucoside 120

3. Effect of unsaturated derivatives of a-methyl glucoside on dextransucrase 141

4. Comparison of acceptor activities of maltooligosaccharides 152

5. Comparison of acceptor activities of disaccharides 166

6. Chemical shifts of acceptor products 218

vii i LIST OF FIGURES

Figure Page

1. Postulated structure of dextran synthesized by Streptococcus sanguis ..... 5

2. Model for chain growth from the non-reducing end as proposed by Neely 15

3. Model for chain growth from the reducing end as proposed by Ebert and Schenk 15

4. Model for chain growth at the reducing end for Dextransucrase as proposed by Robyt et.al. .... 18

5. Proposed pathway for reactions catalyzed by dextransucrase 29

1 O 6. C NMR spectra of halo derivatives of sucrose 60

7. Mass spectrum of 5,6'-dichloro-6,6'-dideoxy- sucrose peracetate 63 1 ^ 8. C NMR spectra of halo derivatives of monosaccharides . . 67

9. 13C NMR spectra of methyl glycosides of (D)-fucose ... 70

10. 13C NMR of methyl glycosides of (L)-rhamnose and allose . . 74

11. 13C NMR of other analogs of a-methyl glucoside .... 80

12. HPLC elution profiles 82

13. IR spectrum of methyl,6-azido-6-deoxy-a-(D)- glucopyranoside 84

14. 13C NMR of bromo maltose 88

15. Inactivation of dextransucrase as a function of concentration of halosucrose derivatives .... 93

16. Inactivation of dextransucrase by halosucrose derivatives as a function of preincubation time .... 95

17. ^ NMR spectrum of (chloromethyl)-triphenyl- phosphonium-chloride 99

ix CONTENTS (continued) Figure Page

18. Mass spectrum of (chloromethyl )-tripheny1- phosphonium chloride 101 19. Inactivation of dextransucrase as a function of concentration of (chloromethyl)- triphenyl-phosphoniurn chloride 104

20. Inactivation of dextransucrase by (chloromethyl)-triphenyl-phosphonium chloride as a function of preincubation time 106

21. Kinetics of dextransucrase inhibition by halosucrose derivatives 109

22. Formation of acceptor products : some typical acceptors . . 116

23. Methyl,2-deoxy-a-(D)-gl ucopyranoside as acceptor .... 125

24. Proposed model for binding of acceptor to dextransucrase . 129

25. Structural comparisons of acceptor analogs 136

26. Photochemical inactivation of dextransucrase with methyl,6-azido-6-deoxy-a-(D)-gl ucopyranoside . . . 144

27. Chromatographic separation of isomaltose oligosaccharides . 147

28. Maltose as acceptor 154

29. Isomaltose as acceptor 156

30. Nigerose as acceptor 158

31. 6'-bromo-6'-deoxy-maltose as acceptor 160

32. Methyl,6-bromo-6-deoxy-a-(D)-nigeroside as acceptor . . . 162

33. Preparative synthesis of acceptor products with a-methyl glucoside as acceptor .... 169

34. 13C NMR spectra of acceptor products with a-methyl glucoside as acceptor 171

35. *H NMR spectra of acceptor products with a-methyl glucoside as acceptor 173

36. Preparative synthesis of acceptor products with maltose as acceptor .... 176

x CONTENTS (continued) Figure Page 1 ^ 37. C NMR spectra of acceptor products with maltose as acceptor 179

38. NMR spectra of acceptor products with maltose as acceptor 181

39. Preparative synthesis of acceptor products with 6'-bromo-ma1tose as acceptor .... 184

40. 13C NMR spectrum of acceptor product with 6'-bromo-maltose as acceptor 186

41. Amounts of polymer and acceptor products formed as a function of time of reaction .... 189

42. Amounts of polymer and acceptor products formed as a function of acceptor concentration .... 192

43. Amounts of polymer and acceptor products formed under varying concentrations of donor and acceptor substrates . 194

44. Preparative synthesis of acceptor products with methyl,6-bromo-6-deoxy-a-(D)-glucopyranoside as acceptor . 197 1 "3 45. C NMR spectra of acceptor products with methyl, 6-bromo-6-deoxy-a-(D)-glucopyranoside as acceptor . . . 199

46. *H NMR spectra of acceptor products with methyl, 6-brorno-6-deoxy-oi-(D)-glucopyranoside as acceptor . . . 201

47. Preparative synthesis of acceptor products with nigerose as acceptor ...... 210 1 O 48. C NMR spectrum of mixture of oligosaccharide products with nigerose as acceptor .... 212

49. Possible mechanism of branch formation by termination of autopolymerization 216

xi LIST OF ABBREVIATIONS

dpm Disintegrations per minute

GTF Glucosyl transferase

HPLC High performance liquid chromatography

IU International unit

NBS N-Bromo-succinimide ppm Parts per mil 1 ion

SDS Sodium dodecyl sulfate

TLC Thin layer chromatography

TMS Tetramethyl silane INTRODUCTION

Dextransucrase (E.C. 2.4.1.5) is a glucosyl transferase that

catalyzes the formation of dextran by transfering glucosyl portion of

sucrose to a growing glucose homopolymer. Monomeric fructose is

released in the process. The overall reaction may be represented by the

following equation:

N V /

Dextransucrase is produced extracel1ularly by bacterial species of

Leuconostoc (1-3), Lactobacil 1 us (4) and Streptococcus (5-7). Some of

these bacteria are cariogenic and the cariogenicity has been attributed

to their ability to form insoluble dextran which is a major component of dental plaque (8). It has been shown that oral bacteria are aggluti nated by dextran (9). Thus, the dental plaque anchors the bacteria close to the tooth surface and also serves as a stable matrix for trapping nutrients needed by the bacteria for their metabolism.

Fructose, which is a byproduct of the dextransucrase reaction, is meta bolized by the bacteria to produce organic acids such as pyruvic acid, lactic acid and acetic acid (10). Plaque serves as a diffusional

1 2 barrier for these acids which concentrate on the tooth surface where they are insulated from the cleansing action of saliva (12, 13). These acids then degrade tooth enamel by localized progressive deminaraliza- tion, thus producing cavities.

The cariogenic importance of a group of anaerobic streptococci designated as Streptococcus mutans has been well substantiated (14,

15). S. mutans constitute a major part of dental plaque (16).

Streptococcus sanguis, however is not a major component of dental plaque. Its cariogenicity is also less compared to that of S. mutans.

In contrast to S. mutans, the relative abundance of S.sanguis in dental plaque has not been correlated with smooth surface caries in children

(17). However, recently it has been observed that S. sanguis is one of the first organisms to adhere to tooth surface in the process of tooth decay (18). Hence it is believed that S. sanguis plays an important role in initiating dental caries by making the environment ready for S. mutans. Throughout this investigation, enzyme from S.sanguis has been utilized because it offers several advantages. In our laboratory we have found that the organism produces only one type of dextransucrase which synthesizes a water-insoluble dextran. Mutans, on the other hand, are known to produce two types of glucosyltransferases; one type synthesizes a water-soluble dextran whereas the other produces an insoluble dextran (19). Also, unlike mutans, S. sanguis does not produce dextranase; consequently, studies on dextransucrase are not complicated by degradation of the product. Besides dextransucrase,

S. mutans are known to produce other sucrose utilizing enzymes such as invertase and fructosyl transferase. S. sanguis does not produce these 3

enzymes. Thus, compared to S. mutans, S. sanguis offers a simplified

system which is easier to work with.

A. Structure of dextran.

Dextran is a glucan (homopolymer of glucose) in which the pre- dominent linkage is a(l+6) (20). The term "mutan" is used for glucans composed primarily of a(l->-3) linkages (21, 22). Dextrans and mutans are major exocellular polysaccharides produced by certain Streptococcus species. However, mutan is not a major product of Streptococcus sanguis or of the Leuconostoc varieties.

Dextrans are compounds of diverse structure. In 1954, Jeanes and coworkers (23) characterized dextrans produced by a large number of bacterial strains. All the dextrans examined contained predominantly a(1+6) linkages. The number and type of secondary linkages varied among the species and were chiefly comprised of a(l->-4) ,a(l-»-3) and ct(l->-2) links. The dextran produced by S. sanguis has been characterized by

Arnett and Mayer (24) in our laboratory. Methylated alditol acetates derived from dextran were analyzed by gas chromatography and mass spectroscopy. The following results were obtained : 53% of glucose residues were substituted at C6, 15% at C3, 16% were terminal non- reducing end residues and 16% were located at branch points. These results show the average linkage composition and do not indicate if more than one type of dextran is being produced by S. sanguis. Certain organism are known to produce more than one type of dextran. Tung and coworkers (25) have suggested that the glucans produced by S. mutans consists of two types of polysaccharides : a predominantly a(l-»-3) linked 4

glucan and a dextran-like polysaccharide with single a(l+3) linked

glucose side chains. Kobayashi and Matsuda (26) have isolated two forms

of intracellular dextransucrase from L. mesenteroides : one enzyme syn

thesized a dextran containing a(l+2) and a(l+3) linkages in addition to

a(l-»-6) linkages; the polymer synthesized by the second enzyme contained

lesser amounts of a(l->-2) and a(l>3) linkages. In this laboratory Chen

(27) examined this jssue by treating S. sanguis dextran with dextranase

to produce oligosaccharides. The oligosaccharides were methylated by

the Hakamori method (28) and analyzed by gas-liquid chromatography and

mass spectrometry. The methylated alditol acetate derivative of a

tetrasaccharide showed the presence of a(l^-6) and a(l>3) linkages at

internal positions in the molecule. These data indicate that a single

type of dextran was being produced by S. sanguis. Further studies by

Lee (29) using proton NMR confirmed the presence of a(l->-6) and a(l->-3) linkages in this streptococcal polysaccharide. Based on all these

observations a hypothetical structure of dextran is shown in Figure 1.

Lower water solubility has been observed for heterogeneous dextrans containing a high degree of a(l->-3) linkages (23) and it has been suggested that, in general, water solubility is governed by the content of a(l+3) linkages (30). Specifically, those glucans which are rich in contiguous, or linear, a(l->-3) sequences rather than 1,3,6-branched residues appear to be less water soluble than those which do not contain a(1+3) linkages as part of their main chains (21, 31). Figure 1. Postulated Structure of Dextran Synthesized by Streprococcus

sangui s.

The schematic representation of the structure of dextran synthesized by S. sanguis ATCC 100558 is based on the results of Arnett and Mayer (24). The numbers in parenthesis refer to the percent of total glucosyl residues having the mentioned type of linkage.

A. Terminal glucosyl residues (16.2%)

B. Linear a-(]>6) linkage (54.0%)

C. Linear a-(l->-3) linkage (13.6%)

D. Branched glucosyl residues (16.2%)

5 0

0 0 0 w1\ Figure 1 7 B. Purification and properties of dextransucrase.

Varying degrees of success has been attained in producing pure

dextransucrase. S.mutan is known to produce two major GTF's. One

synthesizes a water-soluble glucan and is designated as GTF-S and the

other produces a water-insoluble glucan and is designated as GTF-I.

Studies of GTF's from mutans have been difficult due the presence of

other sucrose and dextran utilizing enzymes such as invertase,

fructosyltransferase, dextranase, and other uncharacterized proteins

(32). Comparisions of yields and specific activities of GTF's purified

in different laboratories have been further complicated by the

differences in bacterial growth conditions, purification methods, assay

methods and the requirement for primer dextran (33).

Growth conditions are known to affect the yield of enzyme

significantly. Some bacteria such as the Leuconostoc species require

the presence of sucrose in the culture for significant production of.

dextransucrase (34). The enzyme thus obtained is associated with a

large amount of dextran which is very difficult to remove (35). The

extracellular GTF's of S. mutans on the other hand, are constitutive

enzymes (36) which can be obtained relatively free of glucan from

cultures grown without sucrose. The percentages of the enzyme bound to

the cell and that found in the extracellular fluid also depend on the

amount of sucrose in the media (37, 38). The production of dextransucrase is also known to be affected by Tween 80, fructose and pH

of the medium (32).

Different laboratories have reported different purification

schemes. Most of these schemes involve one or more of the following 8

techniques : ammonium sulfate precipitation, gel filtration,

hydroxyapatite and DEAE chromatography, isoelectric focusing, dextran

affinity chromatography, ethanol precipitation, ultrafiltration using membranes, hydrophobic interaction chromatography, chromatofocusing, and

preparative PAGE (32). Specific activities of the purified enzyme pre

parations ranged from 0.3 to 35 IU/mg. However, it should be noted that

some laboratories use dextran T-10 in their assay, while others do not.

Many laboratories have observed the presence of carbohydrates in purified GTF preparations (6, 39, 40). Ciardi et. al. (41) have

reported that the carbohydrate moiety of the glucosyl transferase from

S. mutans could not be removed by boiling in SDS or by extensive treatment with dextranase, suggesting that the sugar is covalently bound to the enzyme. In this laboratory Grahame and Mayer (42) have prepared highly purified enzyme in which they could not detect any carbohydrate.

The limit of detectibility in this case was reported to be less than one molecule of sugar per molecule of enzyme.

Physical characteristics of the enzyme, as reported by various laboratories agree within certain limits. The 1^ for sucrose was usually found in the range of 1-10 mM. In most cases pH optima were within 5-7 and isoelectric points in the range of 4-6. Higher isoelectric points have also been reported (43). The most frequently reported molecular weights were between 100,000 and 200,000 daltons.

Electrophoretic analysis of native enzyme has been difficult because of its tendency to form aggregates of very high molecular weights (44, 45,

46). Multiple forms of dextransucrase has been commonly encountered and aggregation of monomeric enzyme units has been shown to account for the 9 appearance of these multiple forms. Grahame and Mayer(42) have also

provided evidence that low molecular weight forms of the enzyme arise by

proteolysis of the parent enzyme of molecular weight 174,000.

C. Mechanism of dextransucrase reaction.

The polymerization catalyzed by dextransucrase can be represented

by two different reactions :

(i). An acceptor dependent reaction

nG-F + Acc o- Acc-(G)n +nF

(ii). An acceptor-independent reaction

nG-F ts- (G)n +nF

where G-F stands for sucrose, G for glucose, F for fructose and Acc is

an acceptor molecule. The reactions are believed to be irreversible

since the enzyme cannot produce sucrose when supplied with dextran and

fructose (47, 48). The process of polymer formation can be separated. into three discrete steps : initiation, elongation and termination.

1. Chain Initiation

The process by which the polymerization of D-glucosyl residues by dextransucrase is initiated is not well understood. As shown above, the enzyme can catalyze the transfer of a glucosyl residue to an acceptor molecule. Howevers even in absence of added acceptor molecules the enzyme can catalyze de novo synthesis of dextran as shown in equation

(ii). The transfer of glucosyl residues from sucrose to endogenous acceptors bound to the enzyme does not truely represent initiation.

However, the presence of such endogenous primers would be difficult to demonstrate because even very small amounts of these molecules could 10

serve as acceptor substrates. De novo synthesi s of dextran has been

demonstrated by Mayer et.al. (49). They utilized highly purified enzyme

preparations in which no carbohydrate was detectable. In addition, the

enzyme was treated with dextranase and a-glucosidase. The sucrose was

also treated with dextranase to remove any possible oligosaccharide or

polysaccharide contaminants. All these steps had no effect on the"

catalytic rate of dextran formation. Recently, Grahame and Mayer (42)

carried out a carbohydrate analysis on highly purified dextransucrase

which had a specific activity of 110 IU/ing protein, the highest obtained

in this laboratory. Employing the method of Racusen (50) they could not

detect any carbohydrate in the enzyme preparation. From this they

concluded that there was less than one molecule of a neutral mono

saccharide residue per molecule of enzyme which represents the lower

limit of detectibility of carbohydrate in proteins by this method. This'

enzyme was also able to utilize sucrose in the absence of any added

acceptor. Thus, it appears reasonable that dextransucrase can catalyze

two polymerizations, one in the absence of acceptors by de novo

synthesis and the other by transferring glucosyl residues to acceptors.

In 1959 Hehre (51) had isolated a low molecular dextran (55,000-

60,000 daltons) that contained fructose at its reducing end. Based on this result they proposed that in de novo synthesis sucrose serves as an acceptor and thus initiates polymerization. In further support of this idea, Bailey et .a!. (52) demonstrated that the trisaccharide glucose-

a(l-»-6)-glucose-a(l->-2)-fructose served as an effective acceptor for dextransucrase. 11

However, there are several valid arguments against the idea of

sucrose acting as an acceptor : (i). In the reaction between

[^C]sucrose and dextransucrase, dextran is the only product observed, whereas with normal acceptors a series of oligosaccharide products are observed (49). (ii). Walker was unable to detect any oligosaccharide in the early stages of dextran synthesis from sucrose by the Leuconostoc enzyme (53). (ii i). Trehalose, which is an analog of sucrose, does not exhibit acceptor activity (49). (iv). When dextransucrase was reacted with [^C]suc.rose, labelled only in the fructose portion, no radio activity was incorporated into the dextran produced (54). This indicated the absence of a terminal "sucrose" linkage which would be expected if sucrose served as an acceptor, (v). It is known that in the presence of acceptors, addition of glucosyl residues to the growing chain is at the non-reducing end (53). However, it has been shown that in the reaction between C^C]sucrose and dextransucrase, the addition of new glucosyl residues take place at the reducing end of the growing chain (53,55,56).

One possible mechanism of initiation may involve transfer of a glucosyl moiety to a glucosyl enzyme intermediate at an adjacent active site as proposed by Robyt (56) and also by Ebert and Schenk (57). Both models require the participation of two equivalent nucleophilic groups at the active site. Starting with nascent enzyme, both groups attack sucrose to give glucosyl complexes. The C-6 hydroxyl group of one of these glucose units makes a nucleophilic attack onto the C-l of the other glucose unit thereby initiating the polymerization reaction. One piece of evidence in support of this mechanism is the isolation of a 12

glucosylated enzyme intermediate after short reaction time with

[•^Cjsucrose (58,59,60).

2. Chain Propagation.

(i). Mode of Elongation

Chain elongation involve' the transfer of new glucosyl residues to

nascent chains. This can take place by two general mechanisms :

a. A single chain mechanism, according to which the enzyme remains

associated with the growing polymer chain for a large number of

propagative steps after which chain termination and dissociation of the

complex occur.

b. A multi-chain mechanism in which a complex is initially formed

between enzyme and the polymer. The enzyme catalyzes the transfer of a

monomer unit to the polymer chain. Subsequently, the complex

dissociates and the enzyme can now associate with any available polymer molecule at random to continue the process. The choice of polymer molecule is probably not completely random but dependent on the size of

the chain that will provide the most favorable binding interactions for the enzyme.

It has been suggested that the acceptor-dependent polymerization proceeds via a multi-chain mecahnism in which products are released and then reassociate with the enzyme to function as acceptors (61). On the other hand it is generally thought that de novo synthesis occurs by a single chain mechanism (62). A single chain mechanism should give high molecular weight products. Bovey (63) has shown by light scattering measurements that the molecular weights of nascent dextran chains are 13

very high (50 X 10 ). The fact that no oligosaccharides are produced in

de novo synthesis is also consistent with the sinyle chain mechanism.

(ii) Direction of Chain Growth.

New glucosyl residues may be added at the reducing end or non-

reducing end of a growing chain. Neely (61) has suggested that in the

presence of acceptor, a glucosyl residue from sucrose is transferred to the non-reducing end of the acceptor molecule. Subsequently glucosyl

units are transferred to the non-reducing end of the terminal glucose residue of the growing chain. This model is depicted in Figure 2.

Walker (53) isolated labelled oligosaccharides from a reaction mixture containing dextransucrase, [14C]sucrose and isomaltose, in four fold excess over the [^C]sucrose. The [^C] isomaltotriose produced was treated with borohydride, which reduced the glucose residue at the reducing end. It was hydrolyzed with acid to'yield the monosaccharides, which were separated by paper chromatography. The results indicated that 0.15% of the radioactivity migrated as sorbitol, while the bulk of the label was at the non-reducing end as glucose. This indicated that the labelled glucosyl moiety of [^C]sucrose was being added to the non- reducing of the isomaltose.

Jung (64) has shown that the S. sanguis enzyme can transfer a glucosyl residue from [^C]-a-fluoro-glucose to maltose. Analysis of the trisaccharide product by borohydride reduction and hydrolysis, followed by paper chromatography demonstrated that the labelled glucose residue was at the non-reducing end. Ebert and Schenk (57) have proposed an insertion type mechanism for

growth at the reducing end as shown in Figure 3. According to this

model, a sucrose molecule enters the donor site of the enzyme and is

hydrolyzed. The glucosyl group so released is transferred to the

receptor site of dextransucrase. A second sucrose molecule enters the

vacant donor site, is hydrolyzed, and its glucosyl group is transferred

to the anomeric position of the first glucose residue at the receptor

site. The chain is propagated by a series of identical steps until an

acceptor molecule reacts with the growiny polymer to terminate elonga

tion. The acceptor molecule may be water in which case it will be

termination by hydrolysis. One objection to such a model is that hydro

lysis of the sucrose molecule would result in a loss of its energy of

activation. So the glucose has to be activated by the enzyme for poly

merization to take place, which would be energetically unfavorable.

Robyt et.a!. (56) have proposed a model for growth at the reducing

end (Figure 4). According to this model, the active site of the enzyme

has two equivalent nucleophilic groups : XI and X2. After binding two

molecules of sucrose, both become glucosylated releasing two molecules

of fructose. The C-6 hydroxyl group of one of the glucosyl units (X2)

can then make a nucleophilic attack at C-l of the other glucosyl unit

(XI). This releases one of the nucleophilic units (XI) and forms an

isomaltosyl moiety covalently bound to the enzyme. The free nucleophile

(XI) is again glucosylated and new glucosyl unit attacks the isomaltosyl

enzyme linkage forming an isomaltotriosyl unit and releasing the other nucleophile (X2). This is repeated to form a polymer. The chain might be transferred to water or an acceptor resulting in chain termination. Figure 2. Model for Chain Growth from the Non-reducing End as Proposed

by Neely (61).

Figure 3. Model for Chain Growth from the Reducing End as Proposed by

Ebert and Schenk (57).

d = Donor site

r = Receptor site

Glc = Glucose

Fru = Fructose

Glc Fru = Sucrose

A = Acceptor

15 16

GlcFru

+ GlcFru + A jl + Fru

r * GlcA

+ GlcFru + Glc, ,.A + Fru (n+1) G!c A n (n+1) A

Figure 2.

GlcFru E + GlcFru + Fru

d d c—— GlcFru

+ GlcFru + Fru - f- Glc ' Glc GlcGlc

•CslcFru) •ClcFru) cj + A (GlcFru) — + Glc A n r " Glc r Glc A n Figure 3. 17

According to this model new glucosyl units are added at the

reducing end of the growing chain. Robyt et.al. has provided evidence

for this. Dextransucrase from Leuconostoc mesenteroides immobilized on

Bio-gel P-2 beads, or bound to the exterior of bacterial cells, was

initially charged with unlabelled sucrose, followed by a long pulse of

low concentration of [^Cjsucrose and then a short chase of unlabelled

sucrose. The label was released by adjusting the reaction mixture to pH

2 and heating at 95°C for 10 minutes. The [^C]dextran was then

subjected to borohydride reduction, acid hydrolysis and paper chromato

graphy. The ratio of the labelled sorbitol to labelled glucose in the

pulse was 1:2 for the cells and 1:1 for the gel. In the pulse-chase the

ratio dropped to 1:10 for the cells and 1:100 for the gel. Although the

ratio of sorbitol to glucose was very high for the pulse, suggesting

that some degradation of dextran may have occurred, it is clear that new

glucosyl units are being added to the reducing end of the labelled

dextran.

Similar experiments were done by Ditson and Mayer (55) using highly

pure enzyme from S. sanguis and method of immobilization that retained

85% of the enzyme activity. Their data also support Robyt's model for

addition at the reducing end.

From the above mentioned experimental results it can be concluded that in the presence of acceptors addition takes place at the non-

reducing end whereas in de novo synthesis addition of new glucosyl

residues is at the reducing end. Figure 4. Model for Chain Growth at the Reducing End Dextr^

as Proposed by Robyt et.al (56).

18 iy ®f<3 ®^

r rK,~(3) r*i —*i ~dMD-(D •• CH OH 5 OH I * ate. I OH I CH2 CH? I I e> X "V© •*i©-© V® ©Kl whoro: Xj ,Xg 851 Two seporoto nucleophilic groups ot the octivo sits of doatran sucrose

« glucose CXI =» sucroso

/\ •«=» fructose oo ®* two glucose units linfcod by on (2-1,6 glucosidie bond

Figure 4. ^ (iii) Formation of Branches.

Several mechanisms for branch formation have been proposed (62).

Bovey (63) has demonstrated that light scattering of a reaction mixture

of dextransucrase and sucrose increased even after all the sucrose had

been exhausted. This indicated that branching and complexity of the dextran was increasing. Based on this observation he argued for a two

enzyme reaction which required a branching enzyme in addition to dextransucrase. Branch linkages are formed by scission of a(l+6) bonds

followed by their rearrangement to form a(l->-3) branch points. This branching enzyme was believed to be thermally more stable than dextransucrase and to be active only in the presence of magnesium ions. However, subsequent workers (65, 66) failed to see a metal ion requirement for branched dextran synthesis or to isolate a separate branching enzyme by means of refined separatory techniques. Thus, it appears that dextransucrase is responsible for synthesis of both a(l->-6) skeletal chains and the a(l->-3) branch linkages in dextrans. In addition

Ebert et.al. observed that the high molecular weights of 2-5 X 10® reported by Bovey (63) are due to association of dextrans having molecular weights of 3-5 X 10^ (67).

Ebert and Brosche (68) have suggested that branches are formed when a free dextran molecule acts as an acceptor to release dextran from a dextran-enzyme complex. They tested their hypothesis by adding a low O molecular weight [ H]dextran to an actively synthesizing system O containing sucrose and found that the [ H] label was incorporated into higher molecular weight dextran. However, this would also occur if o glucosyl units were being added to the [ H]dextran in the primer

dependent reaction.

Robyt and Taniguchi (69) have addressed the question of branch

formation. They immobilized dextransucrase on Bio-gel, labelled it with

[^C]sucrose and then washed away the unreacted sucrose. The labelled

enzyme was then reacted with a low molecular weight dextran (average

molecular weight 7,000) and the released [^C]dextran was subjected to

acetolysis. The products contained 92.7% [^C]glucose and 7.3%

[^C]nigerose, which was labelled only at the non-reduciny end. The

nigerose indicates the presence of a(l->-3) linkage which may take place

by transfer of labelled dextran from the immobilized enzyme to non-

labelled acceptor dextran. This probably occurs by the nucleophilie

attack of the C-3 hydroxyl group of the acceptor dextran onto the C-1 of the dextran-enzyme complex.

Another mechanism has been proposed by Hehre (70). According to this mechanism, branches are produced by transfers of glucosyl units

from sucrose to the non-reducing ends of growing branch chains. The author argued that the rapid synthesis of high molecular weight dextrans can take place if there is simultaneous propagation of several branch chains.

(iv) Mechanism of Catalysis.

A general model for glucosyl transferase and glucosyl hydrolase has been proposed by Hehre (71). According to this model, glucosyl transfer reactions can be represented as

Glycosyl-X + H-X1 Glycosyl-X' + H-X where X is typically a group such as hydroxide, fluoride, phosphate or

nucleotide diphosphate, and H-X' can be l-^O, sugars, R-NI^ etc. The

glycosyl-X is called the donor substrate which is represented by sucrose

in the dextransucrase reaction. The H-X1 is known as the acceptor

substrate, which in case of dextransucrase can be mono-, oligo- or

polysaccharides. The glycosidic iinkage of sucrose has a much larger

free energy of hydrolysis (AG0 = -6,600 cal/mole ) than the new linkage

formed (dextran, AG0 = -2,000 cal/mole) (72). This makes the reaction thermodynamically favorable. This activation of the donor substrate can also be provided by non-glycosidic bonds such as carbon-fluorine at C-l of glucose as in case of a-l-(D)-glucopyranosyl fluoride which has been

shown to be an effective donor substrate for dextransucrase (73-77).

Since the linkages in the glucosyl donor and the newly formed linkage in dextran are both a-, the overall reactfon occurs with retention of configuration. This can take place by either a carbonium ion mechanism, or covalent mechanism.

a. Carbonium Ion Mechanism.

Acid catalyzed protonation of the oxygen in the glycosidic bond produces a carbonium ion at the anomeric carbon of the glycosidic residue being transferred. Depending on the direction of attack on the ion, the product may show retention or inversion of configuration. In the carbonium ion species, the C-l is sp^ hybridized and is trigonal planar at C-l. It has been proposed that molecules which mimic transition state species at the active sites of enzymes can be potent inhibitors of those enzymes (78). One example of a carbonium ion 23

mechanism is that of lysozyme (79) in which the transition state may be

depicted as follows:

_ft f CH2OH H....-°-C.(01u 35)

(GlcNAC) NH "02C-(AsP 5D

The authors have shown that the 6-lactone derived from tetra-N-acetyl-

chitotetraose binds to lysozyme about 32 times more strongly than the

corresponding substrate, tetra-N-acetylchitotetraose. The 6-lactone is

also an effective inhibitor of the enzyme.

CH.OH

(GlcNAc) 3-

3 '-lactone of tetra-N-acotylchitotetrooae

Similar observations have been made for glycogen phosphorylase (80) and other glycosidases as well (81). In our laboratory, in the case of dextransucrase, Parnaik (54) showed that the 6-lactone of glucose did not inhibit the enzyme even at concentrations ten times greater than the

Km for sucrose which is 5 mM. This suggests that a carbonium ion is not involved in the reaction mechanism. b. Covalent Mechanism.

In this process, a nucleophi1ic group on the enzyme attacks the O anomeric carbon of the donor substrate in an SN reaction forming a

glucosyl-enzyme intermediate with inversion of configuration at the

anomeric carbon. The two inversions result in an overall retention of

configuration.

Some group transfer enzymes that form covalent intermediates may

also catalyze an isotope exchange reaction as shown below :

^ Enzyme # R-X + X ^ ^ R-X + X • where X is radioactive. In the case of a covalent intermediate, the isotope exchange is really the reversal of the first half-reaction, i.e.,

R-X + Enzyme ^ Enzyme-R + X

* • Enzyme-R + X ^ ^ Enzyme + R-X

the observation of such an exchange reaction supports a covalent mechanism although it does not rule out a carbonium ion mechanism. Many examples of such exchange reactions have been seen. Voet and Abeles

(82) have shown that sucrose phosphorylase catalyzes the isotope 32 exchanges between inorganic phosphate and [ P]-a-(D)-glucose-l- phosphate and between [^C]fructose and sucrose. Ping-pong kinetics observed for this enzyme also argue for a covalent intermediate (83).

Isolation of the covalent intermediate is a definite proof of its existence. Silverstein et.al. (83) have isolated the intermediate in 25

the case of sucrose phosphorylase by gel filtration. Dedonder (84) also

used gel filtration to isolate the intermediate in the case of levan-

sucrase, an enzyme which catalyzes the formation of a fructosyl polymer

from sucrose, and catalyzes isotope exchange reaction between

[^C]gl ucose and sucrose.

Dextransucrase is known to catalyze isotope exchange reaction

between [^C]fructose and sucrose. This has been shown in case of

dextransucrase from L. mesenteroides (3, 85), S. bovis (85) and

S. sanguis (47, 59, 76). Bourne et.al .(85) observed four [^C]

labelled products after incubating sucrose and [^C]fructose with

dextransucrase from L. mesenteroides. Two of these products they

identified as [^C]sucrose and [^C]leucrose (which is 5-0-a-(D)-

glucopyranosyl-(D)-fructose). The other two products were thought to be

formed by successive transfer of glucose to sucrose which may be

functioning as an acceptor.

In our laboratory (47, 59), using dextransucrase from S. sangui s,

only one product was observed and that was sucrose. Since only small

amounts of [^C]sucrose were observed, it is believed that the isotope

exchange observed was due to reversal of the first half-reaction rather

than reversal of the overall transferase reaction. Reaction of the enzyme with [^C]f ructose, unlabel led sucrose and unlabel led dextran as

an acceptor showed diminished rate of [^Cjsucrose production relative

to the reaction in the absence of acceptor. This suggests that dextran and the fructose compete for the same form of the enzyme. In addition,

Jung (76), in our laboratory, observed that dextransucrase could

catalyze isotope exchange between a-fluoroglucose and [^C]fructose. The covalent intermediate that was formed in the reaction between

dextransucrase and sucrose has been isolated and partially characterized

in our laboratory (59). It has been observed that the glucosyl-enzyme

linkage is labile under conditions sufficient to cleave an ester type 0- glycosidic linkage. Iwaoka and Mooser (60) were able to produce a glucosylated form of dextransucrase which was stable when exposed to 4M guanidine hydrochloride or 1% SDS. However, the radioactivity was rapidly and completely released as monomeric glucose when the complex was subjected to alkaline conditions (pH 8).

c. Hydrolysis Reaction.

The covalent intermediate that was formed in the reaction between sucrose and dextransucrase may undergo nucleophilic attack by a water molecule, thus releasing monomeric glucose. The overall reaction will then be hydrolysis of sucrose. In 1955 Goodman and coworkers (86) demonstrated sucrose hydrolysis activity associated with dextransucrase from L. mesenteroides. It is unclear if some of this hydrolytic activity may be attributed to the presence of invertase in the dextran sucrase preparation. Invertase activity has been observed in other dextransucrase preparations (87,88). In 1974 Fukui et.al. (7) isolated a dextransucrase and an invertase from S. mutans. The isolated dextran sucrase was shown to posses hydrolytic activity which was decreased by the addition of the acceptor substrate, dextran. Based on this competition the authors claimed that the dextransucrase had the ability to hydrolyze sucrose. In this laboratory it was shown that the hydrolytic activity comigrated with dextransucrase activity on disc polyacrylamide gel electrophoresis (89). At high concentrations of

sucrose the rate of hydrolysis decreased, and dextran formation was the

predominant reaction.

3. Chain Termination.

In the presence of acceptors such as maltose, isomaltose, and a-

methyl glucoside (90, 49), dextransucrase produces a series of

oligosaccharides, whereas, in the absence of acceptors only high

molecular weight dextran is produced. It appears that termination of

polymerization takes place by transfer of the growing chain to an

acceptor molecule to give a series of oligosaccharides. Robyt and

Walseth (91) have argued that large molecular weight dextrans can also

be released from the enzyme by low molecular weight acceptors. They

reacted enzyme with sucrose for 30 minutes. The reaction mixture was

passed over a Bio-Gel P6 column to separate the reacted or "charged"

enzyme from low molecular weight compounds such as unreacted sucrose.

In a reaction between enzyme charged with dextran and [^C] maltose,

5.6% of the radioactivity migrated as high molecular weight dextran.

From this the authors concluded that enzyme bound dextran was

transferred to [^Cjmaltose to form a [^Cjdextranosyl maltose product,

thereby releasing free enzyme. However, these results have been

strongly criticized by Parnaik (54) because the acceptor products

observed were in 60-fold excess of that expected. This leads to doubts

about the nature of the [^C]dextranosyl maltose product. Thus, the mechanism of chain termination is little known at present. D. Reaction Pathways of Dextransucrase.

The various partial reactions of dextransucrase which have been

discussed so far have been formulated by Luzio et.al. (59) and are shown

in Figure 5. The scheme shows that first sucrose and enzyme react reversibly (step 1) to form a glucosylated enzyme complex. Isotope exchange reactions prove the reversibility of the reaction. Which amino acid of the enzyme participates in the glucosyl enzyme bond is not yet known. This intermediate can undergo one of several reactions. It can be hydrolyzed (step 2) to give glucose and free enzyme. Step 3 shows that the glucosyl moiety of the intermediate can be transferred to the non-reducing end of an acceptor. It can react with more sucrose to form a dextranosyl enzyme complex (step 4) which is also described as de novo synthesis. Polymerization is terminated by hydrolysis (step 5) of the dextranosyl enzyme complex to give dextran and'free enzyme or by trans ferring the dextran to the non-reducing end of an acceptor. Hydrolysis

(steps 2, 5), transfer to acceptor (steps 3, 6) and isotope exchange reaction (step 1) can be seen as nucleophilic attack of water, a hydroxyl group of an acceptor or the C2 hydroxyl group of fructose at the anomeric center of the glucosyl or dextranosyl enzyme complex, respectively.

E. Substrate specificity.

The synthesis of dextran from sucrose may be viewed as a two substrate reaction, involving a glucosyl donor and an acceptor. Figure 5. Proposed Pathway for Reactions Catalyzed by Dextransucrase.

1. Reaction with sucrose.

2. Hydrolysis reaction.

3. Transfer to acceptor.

4. De novo synthesis.

5. Hydrolysis of dextranyl intermediate ?

6. Transfer of dextranyl intermediate ?

? indicates reactions which have not been established for dextransucrase.

29 CMjOM £§- OH • CM HjO CH^OH

CHjOH C„20h »0V-'c~/0m CHJOH Acceptor Cni ' A U°x ' = 0" Cm CM^W HO*—( OH CHjOH A -accopio' ChjOH CHJON XT'' OH n k°) '°' ho^ 'a j~ <>j°H MO ®M

CH,OH H0£ CHjOH

HOo*—( O4H \ MO CHJOH y HO*—( S5 ?? OH IU C nt Acceptor 6

Eni' CHjOH

Ml Acceptor OH

Figure 5. 1. Donor Substrate Specificity.

Donor substrate specificity of dextransucrase is rather narrow

since sucrose is the only naturally occurring donor substrate known.

Investigations with regard to donor substrates have been limited due to

the difficulty in preparing appropriate sucrose analogs. Lactulosucrose

(O-e-(D)-gal actopyranosyl-(l->-4)-CM-(D)-fructofuranosyl-(2-»-l)-a-(D)-

gl ucopyranoside), obtained by transfer of glucosyl residue from sucrose

to lactulose (4-0-3-(D)-galactopyranosyl fructose) by Leuconostoc

dextransucrase, was found to be an effective donor substrate for the

enzyme (3, 92). However, the rate of conversion to dextran was about

one third of that measured with sucrose. More recently, glucosyl

sorboside (a-(D)-glucopyranosyl-a-(L)-sorbofuranoside), made by reaction

with (L)-sorbose, a-(D)-glucopyranosyl phosphate and sucrose phosphorylase, has been shown to be a glucosyl donor substrate in dextran synthesis by S. mutans dextransucrase (93). Gegenhoff and Hehre

(94) demonstrated that a-(D)-l-fluoroglucose can function as a donor.

The fluorine atom is slightly smaller than the C-1 oxygen in sucrose and the carbon-fluorine bond contains about the same free energy of hydrolysis as the glycosidic bond in sucrose. Jung and Mayer (76) have demonstrated that the S. sanguis enzyme can utilize a-(D)-l- fluoroglucose in the absence of acceptor to form dextran. It also participated in the glucosyl transfer reaction to maltose to give a series of oligosaccharides. It is important to note that the Km and

Vmax values for sucrose and a-fluoroglucose are very similar indicating that a-fluoroglucose is as effective a donor substrate as sucrose. Thus the fructosyl residue of sucrose is not specifically recognized by the enzyme. In contrast, Grier and Mayer (77) have shown that a high degree

of specificity exists for the a-(D)-glucopyranosyl moiety. They

synthesized different analogs of a-fluoroglucose by modifying different

positions of the glucose moiety. However, none of these analogs could

function as donors though they all showed inhibition of the reaction.

The e-l-fluoroglucose was not a donor and was a poor inhibitor.

Similarly modification at position 4 of the glucose residue of sucrose

resulted in total loss of ability to serve as donor substrate (95, 96).

Recently it has been reported by Robyt that a number of (D)-gluco- oligosaccharides serve as glucosyl donors for dextransucrases from L. mesenteroides and S. mutans (97). They observed disproportionation reactions when donor substrate also acted as acceptor. For example, in the reaction with isomaltotriose, equal amounts of isomaltose and isomaltotetraose was observed initially followed by the appearance of higher degree of polymerization of isomalto-oligosaccharides. Robyt has also shown that p-nitrophenyl a-(D)-glucopyranoside exhibits donor substrate properties (98). This was, however, a very poor substrate and reacted at a rate which is 0.6% of that with sucrose. The enzyme was obtained from two Leuconostoc strains. With the S. mutan enzyme the rate is still lower by about 31 fold. These results show that absolute donor specificity may vary among enzymes from different bacterial sources. In this laboratory it has been found that p-nitrophenyl a-(D)- glucopyranoside does not show any ability to serve as donor for the enzyme from S. sangui s(99). 33

2. Acceptor Substrate Specificity.

In direct contrast to the narrow specificity of donor substrate,

the specificity toward acceptor is very broad (49, 53, 70, 91, 100).

Though the best known acceptor is dextran, several low molecular weight

acceptors such as monosaccharides, disaccharides and trisaccharides have

also been reported (49, 91). Next to dextran, the best known acceptors

are the disaccharides maltose and isomaltose which also compete with

dextran to produce oligosaccharides of low molecular weights (90). Among

monosaccharides, (D)-glucose and (D)-galactose are rather poor acceptors

(90) whereas a-methyl (D) glucoside is an efficient acceptor (101).

Cellobiose and lactose do not serve as acceptors suggesting that <*-

glucosyl transfer to 3-(D)-glycopyranosyl residues may not be feasible

(62). (D)-fructose acts as an acceptor to give sucrose (49) or leucrose

(90, 104). This explains the ability of dextransucrase to -catalyze an

isotope exchange reaction. Leuconostoc dextransucrase transfers

glucosyl residue from sucrose to lactulose to form lactulosucrose (3).

Iriki and Hehre demonstrated the the transfer of a-(D)-glucopyranose

from sucrose to the anomeric position of mannopyranose and galacto- furanose to form new non-reducing disaccharides (102).

Though no detailed investigation has been done regarding acceptor specificity, one observation made by many researchers is that the most effective acceptors appear to be molecules that contain a-(D)-glucopyranosyl residues at the non-reducing end (61, 62). For example, a-methyl-(D)-glucoside is a better acceptor than (D)-glucose.

It appears that the efficiency of an acceptor depends on the source of the enzyme. For example, fructose was found to be a better acceptor than a-methyl-(D)-gl ucoside for the S. sanguis enzyme (49) whereas the

opposite was found to be true for the enzyme from L. mesenteroides

(104). In this laboratory it has been observed that with the S. sanguis

enzyme sucrose is the only oligosaccharide produced when fructose is the

acceptor (49). However, other authors have reported that (D)-fructose

can also accept a glucosyl residue at position 5 to form the reducing

disaccharide leucrose (90, 104).

F. Purpose of Investigation.

Halo- derivatives of sucrose were synthesized. Kinetic studies have

been performed to examine their abilities to serve as reversible or

irreversible inhibitors of dextransucrase.

The acceptor substrate specificity of the enzyme was analyzed.

Analogs of a-methyl-(D)-glucopyranoside were synthesized to study the

importance of enzyme-acceptor interaction at individual positions of the

sugar ring. In particular, epimeric analogs of a-methyl-(D)-gluco-

pyranoside have been investigated as well as analogs containing changes

at C6.

Since dextran, a polysaccharide, is a better acceptor than mono- or disaccharides, one of the objectives of this investigation was to determine the acceptor properties of the intermediate oligosaccharides.

Compounds of the maltooligosaccharide series were employed as acceptors.

The structures of some of the acceptor products formed with mono- and disaccharides as acceptors were determined for the purpose of understanding the mechanism of branching. MATERIALS and METHODS

A. Materi als

1. Organisms.

Streptococcus sanguis ATCC 10558 was obtained from the American

Type Culture Collection, Washington D.C., and stored as a lyophilized

powder at -20°C.

2. Enzymes.

The following enzyme preparation were purchased from Sigma Chemical

Company (St. Luois, MO.) : glucose-6-phosphate dehydrogenase (type XV

frome Baker's yeast), hexokinase (from Baker's yeast), phosphoglucose

isomerase (from yeast), a-glucosidase (from yeast), invertase (from

Baker's yeast) and dextranase (grade 1 from Penicillium species).

Highly purified dextransucrase was provided by David A. Grahame who

purified it in this laboratory by either of the two methods : (i) the

procedure developed by Huang (106) and modified by Parnaik (54) or (ii) the procedure developed by Grahame and Mayer (42). The earlier method gave enzyme that had low specific activity (20-50 IU/mg protein, 2-4

IU/ml). The latter procedure gave enzyme of high purity and high

35 36

specific activity (110 IU/ing protein, 9-10 ID/ml) but was obtained in

inactive form and had to be reactivated by non-ionic detergents as will

be described in this chapter. The dextransucrase used in most of these

studies was purified by this method. Enzyme that was purified by the

earlier method was employed in experiments that required enzyme free of

detergents.

3. Saccharides and Derivatives.

Saccharides were obtained from the following sources : (D)-fucose;

(L)-rhamnose; isomaltose; isomaltotriose; maltose; maltotriose; maltopentaose; maltoheptaose; a-0-methyl glycosides of (D)-glucopyranose

(recrystal1ized from ethanol and dried); (D)-mannopyranose; (D)-galacto- pyranose; (D)-altropyranose; (D)-xylopyranose; 6-deoxy-(D)-gluco pyranose; 6-deoxy-(L)-gal actopyranose; and 3-0-methyl glycosides of (D)- glucopyranose and 6-deoxy-(D)-glucopyranose were purchased from Sigma

Chemical Company (St. Luois, MO). S-(D)-al1opyranose; 3,4,6-tri-O- acetyl glucal and 2,3-di-0-acetyl-6-deoxy-(L)-glucal were obtained from

Pfanstiehl Laboratories Waukeegan, IL). Dextran T-10 was from Pharmacia

Fine Chemicals (Piscataway, NJ). Dextran T-10 was dialyzed extensively to remove low molecular weight oligosaccharides and then lyophilized.

Maltotetraose and maltohexaose were kindly provided by Dr. Howard

Sloan of Children Hospital, Columbus, OH. Methyl,2,3,6-trideoxy-4-0- acetyl-erythro-hex-2-enopyranoside and the mixture methyl glycosides of

2-deoxy-3,4,6-tri-0-acetyl-glucose were kindly provided by Professor D.

Horton of The Ohio State University Columbus and nigerose was a gift from Professor I. Goldstein of University of Michigan, Ann Arbor. Radioactive saccharides were obtained from the following sources :

[U-^Cjsucrose (673mCi/mmol e) and [^C-glucosejsucrose (201mCi/mmole) from New England Nuclear Corp. (Boston, MA) and [^C]-a-0-methyl glucoside (291rnCi/mmole) from Amersham (Arlington Heights, IL).

4. Chromatography Materials.

Chromatographic supports were obtained from the following sources : silica gel 60; silica gel 60-G; plastic backed TLC plates (silica gel 60 and silica gel 60 F254) from E. Merck (Gibbstown, NJ); anion exchange resin AG1-X8, 20-50 mesh, hydroxide form and AG1-X2, 200-400 mesh, chloride form; cation exchange resin AG50W-X8, 20-50 mesh, hydrogen form; mixed bed resin AG501-X8, 20-50 mesh and Bio-Gel P2, minus 400 mesh were obtained from Bio Rad (Richmond, CA). Whatman 1MM and 3MM chromatography paper; Partisil M9,10/25 PAC; CSK guard column and

CO:PELL PAC for guard column were from Whatman Inc. (Clifton, NJ).

5. Chemicals.

Chemicals and reagents were obtained from the following sources : anthrone and triphenyl phosphine from MCB (Gibbstown, NJ); acrylamide; adenosine triphosphate (disodium salt, vanadium free, 99-100%); B- nicotinamide adenine dinucleotide phosphate (sodium salt, 98-100%);

1auryl sulfate (sodium salt 95%); Trizma base (tris[hydroxymethyl]~ aminomethane, 99.9%); Coomassie Brilliant Blue G; 2,5-diphenyl oxazole from Sigma Chemical Company (St. Louis, M0); carbon tetrabromide; deuterium oxide (99.8% and 99.999%); guanidine hydrochloride and

N,N,N' jN'-tetramethylenediamine from Aldrich Chemical Company (Milwaukee, WI); N-bromo succinimide (recrystal1ized from water and

dried) from Fisher Scientific Company (Fair Lawn, NJ); Triton X-100 from

Rohm and Haas (Philadelphia, PA); HPLC grade acetonitrile and water from

Baker Chemical Company (Phi 11ipsburg, NJ); and p-bis-[2-(5-phenyl-

oxazolyl)]-benzene from Research Products Int. Corp. (Elk Grove,II).

B. Methods.

1. Measurement of Dextransucrase Activity.

Dextransucrase was assayed by either one of the following methods,

a. Coupled Enzyme Assay

The amount of fructose released is measured by a modification (46)

of the procedure described by Carlson, Newbrun and Krasse (106). The

reactions involved in the assay are:

dextransucrase sucrose + (glucose)n > (glucose)p+^ + fructose

hexokinase fructose + ATP ^ fructose-6-P + ADP Mg++

phosphoglucose isomerase ^ frnrtn^p-R-P ^ glucose-6-P

glucose-6-phosphate dehydrogenase glucose-6-P + NADP+ ^ 6-phosphogluconate NADPH + H+

The formation of NADPH was measured spectrophotoinetrically at 340 nm.

The assay procedure was carried out in two steps. _3 Step 1. A reaction mixture containing 5 X 10 M Dextran T-10, 0.1M sucrose, 0.05M sodium phosphate buffer pH 6.0 and enzyme (usually 0.05 units) in a total volume of 0.2 ml was incubated at 37°C for 15 minutes

and then boiled at 100°C for 3 minutes to stop the reaction. A control

was run in a similar manner except that the enzyme had been heated at

100°C for 3 minutes prior to the reaction.

Step 2. An aliquot (10PL) from step 1 was withdrawn and added to IOOmL

of a reaction mixture containing 0.2 units each of hexokinase, phospho- -3 glucose isomerase and glucose-6-phosphate dehydrogenase, 0.5 X 10 M

ATP, 5.0 X 10"3M MgCl2> 5.0 X 10"3M NADP, 0.1M Tris.HCl, at pH 7.0 and

incubated at 37°C for 20 minutes during which the reaction goes to

completion. This was then diluted with water to a final volume of 1.0

mL and the absorbance was measured at 340 nm. Enzyme activity was

calculated according to the following equation : mole 0.2 mL Activity (umts/mL) = AA34Q X 6>2 A mL X x mL

0.1 mL 1 X y mL ^ 15 min

where AAg^Q = Difference between absorbance at 340 nm of experimental

and control.

x = volume of enzyme solution in step 1. y = aliquot of step 1 transferred to step 2.

One unit of enzyme was defined as amount of enzyme catalyzing the

release of 1 pmole of fructose per minute under the above mentioned

conditions b. Radioactive Assay.

The coupled enzyme assay method measures not only the fructose released due to polymerization but also the glucose and fructose released by hydrolysis (89). The method is also inappropriate for some kinetic studies. In particular, those sugars and sugar derivatives that are substrates or inhibitors of the enzymes in the assay mixture in step

2 will interfere in the measurement of fructose and cause erroneous

results. To circumvent these problems an alternate assay method was followed. In this method the direct incorporation of radioactivity from

[^C]sucrose into polymeric form was measured. The transfer products,' which were immobile on paper chromatograns were measured after development in a suitable solvent. This method is more reliable than the coupled enzyme assay because very low amounts of radioactivity can be detected and less experimental variables are involved. (T A typical reaction mixture contained 5 X 10 M Dextran T-10, 0.1M

[^C]sucrose (0.05pCi), 0.05M sodium phosphate buffer pH 6.0 and enzyme

(usually 0.05 units) in a total volume of 30pL. Reaction was initiated by addition of enzyme, incubated at 37°C for a predetermined time

(generally 2-5 minutes). The reaction was stopped by heating at 100°C for 2 minutes. Aliquots (10yL) were spotted on a Whatman 1MM paper chromatogram. After developement in solvent system (iii) the chromatogram was cut into 1 cm strips and counted as described in

Methods. Rate of reaction was determined by calculating the percentage of [^Cjglucose from sucrose that resided at the origin, or in some cases at the origin and in the oligosaccharide products.

2. Reactivation of SDS Inhibited Enzyme.

The enzyme preparation used in most of the investigations had SDS added during purification. The enzyme was inactive in the presence of high levels of SDS, but could be reactivated as follows : The inactive 41

brought to room temperature (they both are usually stored frozen).

Equal volumes of enzyme and Triton X-100 were mixed and kept at room

temperature for about 20 minutes during which the enzyme was reactivated

and could be assayed as described above. The presence of detergents at

the concentrations employed had no apparent effect on either method of

enzyme assay.

3. Assay of Acceptor Activity.

Acceptor activity was determined by reacting reactivated dextran-

sucrase (0.05 units i.e. IOmL) with a mixture of [gl ucose-^C-(ll)]-

sucrose (20mM, 0.12yCi) and acceptor at indicated concentrations in 0.1M phosphate buffer, pH 6.0, final volume 60pL at 37°C for specified lengths of time. Following reaction, 20pL aliquots were spotted on

Whatman 1MM paper and developed as will be explained later. Following development, the chromatograms were cut into strips (2.54cm X 1.0cm) which were then counted for radioactivity. In some cases repetitive chromatography was performed as will be described under Chromatographic

Methods. Radioactivity cornigrating with sucrose was eluted from the strips with water, evaporated to dryness, dissolved in 0.1M phosphate buffer, pH 4.5 and reacted with 4 units of invertase in a total volume of 35uL at 55°C for 10 minutes. After reaction, the reaction mixture was spotted on Whatman 1MM paper, developed, cut into strips and radio activity counted as before.

Invertase converts all the residual sucrose to glucose and fructose. Thus, any radioactivity not migrating with glucose was counted as due to acceptor product. 42

4. Inactivation of Enzyme

a. Chemical Inactivation.

The ability of some donor and acceptor substrate analogs to

inactivate dextransucrase was evaluated in two types of experiments.

Under one protocol the enzyme was preincubated with varying concen

trations of compounds for 60 minutes, and then assayed for activity.

Control preincubations were run under each set of conditions, whereby

the enzyme was allowed to stand in buffer for the prescribed period.

The second protocol involved preincubations for varying times, at fixed concentrations of reagents. The details for reaction conditions for

each individual reagent will be described in the appropriate figure and table legends under Results. b. Photochemical Inactivation

Samples to be irradiated with ultraviolet light were placed in quartz cuvettes (1cm path length) at a distance of 2 cm from a source of

I)V light (Mineral ight, UVS.12, 0.17 amps, Ultraviolet Products Inc., San

Gabriel, CA). For control reaction the cuvette with was wrapped in aluminum foil and placed 2cm from the source of UV light. At specified times aliquots were removed and assayed.

5. Chromatographic Methods.

The following solvent systems have been employed :

(I) Ethanol/ethyl acetate/water : 45/5/3

(II) Benzene/ethyl acetate : 4/1

(III) n-Propanol/ethyl acetate/water : 6/1/3

(IV) Butanone/acetic acid/boric acid saturated water : 9/1/1 (V) Acetonitril e/water/ammoni urn hydroxide : 75/25/1 (HPLC grade)

a. Thin Layer Chromatography

Reactions and columns were monitored by thin layer chromatography.

Plastic backed TLC sheets precoated with 0.2 mm layer thickness silica

gel 60 with or without fluorescent indicator were used. Development was

done in solvent system (I), unless otherwise stated. Aromatic compounds

were visualized under UV light and sugars were detected by spraying with

25% sulfuric acid in methanol and charring at 100°C.

Preparative TLC plates with layer thickness up to 2mm were prepared

in the laboratory. A slurry of silica gel 60 G (150g), calcium sulfate

(40g) in water (about 400mL) was spread on glass plates (20cm X 20cm) using an adjustable Thin Layer Spreader. The plates were dried at room temperature overnight and then in the oven at 100°C for one hour. After development in solvent system (I) in a glass tank (30cm X 10cm X 25cm), lcm blocks of silica gel were scraped off and the sugar was eluted from the gel with acetone. The purity was checked by TLC.

b. Paper Chromatography

Paper chromatography was carried out on Whatman 1MM paper in a descending manner. All chromatograms were developed at room temperature in solvent systems (III) or (IV). Chromatograms were visualized by the silver nitrate dip procedure (107) which is briefly discussed here. The chromatogram was dipped in a mixture of 5mL 0.1M periodic acid in water and lOOmL acetone, dried, dipped in 0.6% (w/v) AgNO^ in acetone, dried, dipped in 2% (w/v) NaOH in ethanol, dipped in 5% (w/v) ^5203 in water and finally washed with cold running water. Sugars show up as black

spots.

For preparative paper chromatography Whatman 3MM paper was

employed. A small strip of it was cut to detect sugar by the silver

nitrate dip procedure as mentioned above or by radioactive analysis as

will be discussed later. The region where the desired sugar migrated

was cut into pieces and the sugar eluted with water, concentrated and

analyzed by NMR spectroscopy.

Sugars which migrated close to the origin as in case of oligo

saccharides with high degree of polymerization were made to migrate

farther from the origin by either of the following two methods

Continuous chromatography, in which the chrotogram was developed

for several days, depending on the Rf value of the sugar. In this method the slow moving sugars moved farther from the origin , however, the fast moving ones migrated off the paper.

Repetitive chromatography, in which the chromatogram was developed till the solvent front moved one length of the paper, after which it was taken out, dried, and developed again. This was repeated several times depending on the Rf value of the sugar. In this method the slow moving sugars moved far from the origin and at the same time the fast moving sugars remained on the paper.

In both methods a small strip of the chromatogram was cut to detect sugar by the silver nitrate dip procedure as described before or by radioactive analysis as will be discussed later. C. Column Chromatography.

(i) Si 1ica Gel .

Silica gel 60, 70-230 mesh was employed for column chromatographic

separation of compounds after synthetic reactions. Depending on the amounts of compounds to be separated and their Rf values, either of the

following two sizes of columns was used : 2cm X 30cm or 3cm X 80cm. The

flow rate was typically about lmL/minute and 10-15mL fractions were collected. The solvent system employed will be mentioned separately for each synthetic procedure. Fractions were analyzed by TLC; appropriate fractions were pooled, concentrated and analyzed.

(ii) Ion Exchange.

Some methyl glycosides such as methyl allosides, were separated on a column (2.4cm X 43cm) of Bio Rad AG1-X2 resin (OH" form, 200-400 mesh) by eluting with water. Since the hydroxide form was not available commercially the chloride form was converted to the hydroxide form with

1M NaOH. A slurry of the chloride form of the resin in water was poured into the column. After the resin settled it was equilibrated with about

20 bed volumes of 1M NaOH solution. Completion of the conversion was tested by the pH of the effluent and absence of precipitate in acidic silver nitrate solution. During separation of sugars a typical flow rate was lmL/minute and about 20mL fractions were collected.

Halo methyl triphenylphosphoniurn halides were separated from sugars by passing a solution of the mixture in water through a column (2cm X

20cm) of Bio Rad AG1-X8 resin (OH" form, 20-50 mesh) at a flow rate of about lmL/minute. The halides were retained by the column and pure sugars were eluted. The same effect could be obtained by stirring the

resin with a solution of the mixture in water and then filtering to give

pure sugar in the filtrate.

(iii) Gel Filtration.

Separation of a mixture of oligosaccharides into its component w.as

achieved by gel filtration through Bio Gel P2 (minus 400 mesh).

Depending on the difference in degree of polymerization of the oligo

saccharides to be separated, two column sizes were used : 2cm X 50cm and

2cm X 120cm. The columns were prepared by following the procedure

recommended by Bio Rad. Dry Bio Gel P2, minus 400 (3.5mL bed volume/g)

was hydrated in 7 volumes of deionized water for four hours at room

temperature. Half of the supernatant was decanted. The rest was

deaerated by aspirating at reduced pressure. 20% of the column was

filled with water and the slurry poured into the column. When about 2cm

bed had formed it was allowed to flow. The void volume was determined

by using blue dextran, detected spectrophotmetrically or dextran T-10,

detected by the anthrone method. For all separations water was the

solvent. The flow rate was maintained constant at 6mL/hr with the help

of a peristaltic pump (LKB) and lmL fractions were collected. All

separations were done at room temperature (higher temperature gives better resolution). The column was calibrated with glucose and malto-

oligosaccharides (DP 2-7), which were detected by the anthrone method

(p49). Following chromatography on a preparative scale, sugars were detected by the anthrone method or by radioactivity measurements when

using radioactive sugars. 47

d. High Performance Liquid Chromatography.

Compounds which were difficult to separate by conventional column

chromatography were separated by HPLC using a Varian 5000 Liquid

Chromatograph. The column was semi-preparative Partisil M-9, 10/25 PAC

from Whatman. Sugars were detected using a refractive index detector

from Varian (series RI-3). The reference cell had the same solvent

system used as eluant. The solvent system (acetonitrile/water/ammonia :

25/75/1) was degassed by sonication at low pressure before use. The

sample to be separated was dissolved in the same solvent system and

filtered under vacuum through a fritted disc (10-15M). Depending on the

concentration, a 100-400pL sample (about 50mg sugar) was injected in a

single run. The flow rate was 1.5mL/minute and fractions were collected

at 30 second intervals.

6. Radioactive Analysis

Following development, radioactive paper chromatograms were dried and cut into strips 1cm X 2.54cm, beginning 0.5cm below the origin.

Strips were placed in vials, each containing lOmL of scintillation fluid, and counted in a Packard 460-C Tri-Carb Liquid Scintillation

Spectrometer. The scintillation fluid had the following composition :

4.02g of 2,5-diphenyloxazole (PP0) and 0.10g of l,4-bis-[2-(5-phenyl- oxazolyl)]-benzene (P0P0P) per liter of spectral grade toluene. 48

7. Spectroscopic Methods.

a. -^H NMR Spectroscopy

nuclear magnetic resonance spectroscopy was performed on a

Bruker WP-200 Fourier Transform Spectrometer. Water insoluble compounds

were dissolved in deuterated chloroform (99.8 atom% D) with tetramethyl-

silane (TMS) as internal reference. Water soluble compounds were

dissolved in D^O (99.8 atom% D). The HDO peak was assigned at 4.63 ppm. All chemical shifts were expressed in ppm downfield from either

TMS or sodium 2,2-dimethyl-2-si1apentane-5-sulfonate (DSS). For some compounds the HDO peak overlapped with the peak of interest. In such cases the compound was first dissolved in D2O (99.8 atom% D), evaporated to dryness to remove all HDO molecules and then dissolved in D£0 (99.996 atom% D). In this way the water peak was reduced to give better resolution.

b. 13C NMR Spectroscopy.

«C nuclear magnetic resonance spectroscopy was performed on a

Bruker WP-200 Fourier Transform Spectrometer. Water insoluble compounds were dissolved in deuterated chloroform (99.8 atom% D) with TMS as internal reference. Water soluble compounds were dissolved in D2O (50-

100 atom% D) with acetone as internal or external reference.

c. UV/VIS Spectroscopy.

Absorbance values in the visible or ultraviolet regions were recorded on a Zeiss PM4QIII or Beckman 25 spectrophotometer. In all cases 1cm path length cuvettes were used. 49

d. IR Spectroscopy.

Infrared spectra were obtained on a Perkin Elmer 457 grating

spectrometer. Samples were taken in Perkin Elmer liquid cells with

0.1mm teflon spacer and KBr windows. For solid sample, the compound

(ling) was mixed with lOOmg KBr and mechanically mixed to homogeneity. A

pellet was made in a Wabash hydraulic press. In either case, the peak

at 1603 cm"* for a polystyrene film was used as a reference.

e. Mass Spectroscopy.

Mass spectra were obtained by Dr. R. C. Weisenberger in a Kratos MS

30 Mass Spectrometer at the Campus Chemical Instrument Center, The Ohio

State University.

8. Saccharide Analyses.

Sugars were detected and quantitated by the anthrone method (108,

109). 1.0 ml of 0.2% (w/v) anthrone in sulfuric acid was added to 0.5mL

of sugar sample in water (O.Ol-O.lmM) in an ice bath. The reactants

were mixed in the ice bath taking care not to increase the temperature.

From 0°C (in ice) the tubes were directly transferred to 100°C (heating block), heated for 10 minutes and immediately cooled in ice. Absorbance was determined at 620 nm. Glucose or sucrose was used as the standard.

9. SDS Gel Electrophoresis.

Fairbanks' method (110) was applied for SDS electrophoresis. 6.0% polyacrylamide gels were prepared by mixing the following solutions in the order 1isted : 2.4mL 10% w/v sodium lauryl sulfate.

2.4mL 0.4M Tris acetate, 0.02M ethylene diamine tetraacetic acid

(EDTA), 0.2M sodium acetate, pH 7.4.

4.8mL 30% w/v acrylamide, 0.80% w/v N, N methylene bisacrylamide.

6.0yL N, N, N', N' tetramethylethylenediamine (TEMED).

7.2mL water.

7.2mL 0.5% w/v ammonium persulfate (freshly made).

The solution was poured into glass tubes (0.5cm I.D. X 9.5cm) which had been capped with Parafilin, and the solutions (0.5cm X 7.5cm) were over- layed with a solution (about 50pL) of 0.01% w/v SDS, 0.15% w/v ammonium persulfate, 0.05% v/v TEMED and allowed to polymerize. An hour or two after polymerization was evident, the overlay solution was removed and replaced with electrode buffer (about 0.5 mL) which consisted of 1.0%

SDS, 0.04M Tris.acetate, 0.002M EDTA, 0.02M sodium acetate, pH 7.4, and the gels were left overnight at room temperature. Samples were evaporated to dryness and redissolved in 50pL of a solution containing

1.0% SDS, 10% glycerol, 40mM dithiothreitol, l.OmM EDTA, 5|jg/mL pyronin

Y, and lOmM Tris.HCl, at pH 7.8. The samples were carefully layered on top of gels. Electrophoresis was conducted at 25V for 3-4 hours or until the pyronin dye front had migrated about 6cm. The final position of the pyronin Y was recorded by cutting the gel at the dye front. SDS was removed from the gels by soaking them in 25% isopropanol, 10% acetic acid (200mL per gel) for 18 to 20 hours. Protein was stained with

Coomassie Brilliant Blue G-250, 0.04% in 3.5% HCIO^ (111) for about three hours. The background was destained in 7.5% acetic acid overnight. The destaining solution was changed at least three times. 51

10. General Synthetic Procedures.

a. Preparation of Dry Solvents and Reagents.

(i) Pyridine.

300mL pyridine was refluxed over about 3g Cal^ for 2 hours. It was

then distilled at atmospheric pressure. The fraction boiling at 113-

114°C was collected and stored over 4A molecular sieves.

(ii) Methanol.

300mL methanol, 2.5g magnesium turnings, 0.4g iodine and 0.2mL

carbon tetrachloride were refluxed till all the magnesium had

dissolved. An additional 200mL methanol was added, refluxed for one

hour and distilled at atmospheric pressure. The fraction boiling at

65°C was collected and stored over 4A molecular sieves (112).

(iii) Sodium Methoxide.

l.Og of sodium was taken in an Erlenmeyer flask and covered with

Xylene. The flask was heated until the metal melted but stayed in the

cage of surface oxide. The heating source was removed, and the flask

was swirled gently to cause the sodium to flow out of the shells and

form several globules. The flask was then cooled under nitrogen without

agitation so that sodium globules would not unite. When the metal had

solidified, the clean globules were removed, dried rapidly with filter paper, and added to 25 mL of dry methanol in a flask containing 4A molecular sieves. The vessel was flushed with dry nitrogen until all the

sodium had reacted. The reagent was standardized by titration with standard HC1 and stored air tight in a dessicator. 52

(iv) N,N-Dimethyl Formamide.

DMF was refluxed over calcium oxide, distilled and stored over 4A

molecular sieves (112).

(v) Carbon Tetrachloride and 1,1,2,2-tetrachloroethane.

These were dried by storing over activated-5A molecular sieves.

Molecular sieves were activated by heating over a burner for one hour

and cooling to room temperature.

b. Halo Derivatives of Sugars.

Sugars were halogenated at primary carbons by a modification of the

method of Anisuzzaman and Whistler (113). The advantage of this method

is that the halogenation is selective for primary hydroxyl groups. Thus

blocking and deblocking steps are not necessary. The-reaction with

sucrose is shown in the following equation :

Pyridine

H OH

The sugar was dissolved in pyridine by stirring at room temperature or by heating when needed. The solution was cooled and triphenyl phosphine was added, followed by addition at 0°C of carbon tetrahalide in several portions. The resulting mixture was protected from moisture. After 53

keeping the mixture at the desired temperature for the required length

of time, methanol (lOmL/lg of hydroxyl compound) was added to decompose

any excess reagent. Pyridine was removed by evaporation as an azeotrope

with toluene. Non-carbohydrate material was removed by either of the

following two ways :

(i). Water was added to the residue. Carbohydrates dissolved in

the water whereas PhgPO, unreacted PhgP and CSr^ were left as a

precipitate. This was filtered.

(ii). The residue was dissolved in chloroform and bound to

activated silicic acid and filtered. The residue was washed with

chloroform until no more aromatic compound was present in the washing.

The halogenated sugars were then eluted with solvent system (I). The

silicic acid retained most of the unreacted sugar. This could be eluted with 95% ethanol.

The halogenated sugars so obtained were found to be contaminated with an aromatic compound. Several aromatic side products have been previously reported (114-117) for this reaction. The contaminant was separated from the sugar by silica gel column chromatography or by

HPLC. Later, this aromatic compound was found to be halomethyl- triphenylphosphonium halide. Being an ionic compound this could be easily separated from the sugar by passing through an ion exchange column (AG1-X8, 0H"form, 20-50 mesh).

The sugars were separated on silica gel column with solvent system

(I). In some cases HPLC or preparative TLC was necessary. Halo derivatives of monosaccharides usually could be crystallized from ethanol/hexane. Sugars which could not be brominated by the above method were

brominated by the method of Hanessian (118, 119). This involves the

reaction between N-bromosuccinimide (NBS) and the benzylidene acetal of

the sugar which was obtained by reacting benzaldehyde dimethyl acetal

with the sugar. In a typical reaction, the benzylidene acetal

derivative of the sugar was reacted with NBS in presence of barium carbonate (1:1.2:1.4 molar ratio respectively) in carbon tetrachloride

under reflux for 2-3 hours under normal room illumination. The mixture originally milky, became light yellowish red. In those cases where the

starting acetal s were insoluble in hot carbon tetrachloride, complete solution could be effected by addition of dry 1,1,2,2-tetrachloroethane. Barium carbonate was used to neutralize the hydrogen bromide that was formed in the reaction. After reaction, the solvent was removed by evaporation and the residue extracted with methylene chloride and filtered. The filtrate was washed with a saturated solution of sodiun bicarbonate, dried over MgSO^, evaporated and then crystallized. For best results, the NBS was crystallized from water and dried before use.

c. Methyl Glycosides.

Methyl glycosides were synthesized by standard glycosidation techniques (120, 121) in 1% methanolic HC1. The HC1 was prepared in situ by reaction between a calculated amount of acetyl chloride and methanol (122) according to the equation 55

For example, for 1% HC1 in 25 mL CHgOH, 0.49mL of CHgCOCl was needed

(considering the density of acetyl chloride to be l.lg/mL). In a

typical reaction, about lg of free sugar was heated under refluxed in

30mL of dry 1% methanolic HC1 for 24 hours. Subsequently, the solvent

was evaporated and products separated on silica gel column or AG1-XB,

50-100 mesh ion exchange column, or by fractional crystallization.

In the glycosidation reaction the relative proportions of a-

pyranoside, 0-pyranoside, ct-furanoside and 3-furanoside depend on the

relative proportions of the corresponding forms of the free sugar in

solution. (D)-Allose in aqueous solution exist mainly as the 0-

pyranose, due to destabilization of the a-pyranose by 1,3-diaxial

interaction between H0-1, HO-2 and HO-3. Alkaline earth metal ions are

known to complex with three hydroxyl groups in an axial-equatorial-axial

arrangement on successive carbon atoms of a cyclohexane or

tetrahydropyran ring, or in a cis-cis relationship on successive carbons

of a tetrahydrofuran ring (123). This complexation can stabilize the a-

pyranose form of allose. For example, it has been shown (124) that when

allose is dissolved in 0.85M calcium chloride solution, the proportion

of a-pyranose increases from 14% to 37%. Allose can form the complex in

both a-pyranoid and a-furanoid forms. Fisher glycosidation is also

influenced by the presence of alkaline earth metal ions. These ions

slow the rate of reaction considerably and it is necessary to increase

the acid concentration to ten times that used in the absence of salt in order to obtain similar reaction rates. 56

d. Acetylation Reactions.

Unsubstituted sugars can be peracetyl ated with acetic anhydride in

the presence of either acid or base (125). A base catalyzed reaction

was performed in most cases. For example, 5g of the anhydrous sugar was dissolved in a cooled mixtutre of 25mL acetic anhydride and 35mL dry

pyridine. After 18 hours at room temperature, the reaction mixture was poured with stirring into lOOmL of ice-water. The peracetylated sugar precipitated and was recrystal1ized from ethanol.

e. Deacetylation Reactions.

Deacetylation was carried out by the method of Zemplen and Pacsu

(126). By this method the acetyl groups are converted catalytically to methyl acetate. In general, lg of the acetate was suspended or dissolved in lOmL of dry methanol to which 2-4 drops of 1M methanolic sodium methoxide were added. The reaction mixture was kept at room temperature for about an hour or until the reaction was complete. The solution was then neutralized with cation exchange resin AG 50W-X8 hydrogen form, 20-50 mesh. The resin was filtered off and the solution evaporated to a sirup which was then crystallized from a suitable solvent, usually methanol.

11. Chemical Synthetic Procedures. a. 6,61-diehioro-6,61-dideoxysucrose

Sucrose (2.lg) was dissolved in dry pyridine (125mL) by heating the mixture up to 80°C for 10 minutes. Triphenyl phosphine (lOg) was dissolved in it; the solution was cooled in ice and carbon tetrachloride 57

(lOinL) was added. Following reaction at 60°C for 45 minutes, methanol

(20mL) was added and kept at room temperature for an additional two

hours. Pyridine was removed by evaporation as an azeotropic mixture

with toluene. Triphenylphosphine oxide, was removed by precipitation

from water. According to TLC, 6,6'-dichloro-6,6'-dideoxy sucrose (R^

0.28) was the major product with minor amounts of unidentified

contaminants with values of 0.09 and 0.43. The major product was

separated from the others on a silicic acid column with solvent

system(I). However, the sugar was still contaminated with a water

soluble aromatic compound which was removed by rechromatography. This

water soluble aromatic compound was later identified to be (chloro-

methyl)-triphenyl-phosphoniurn chloride which could be easily removed by

passing through a column of AG1-X8 as explained before (p 45). The 1 O dichlorosucrose was characterized by its C NMR and by mass spectral

analysis. The yield was 1.4g (61%).

In the 13C NMR spectrum of sucrose (Figure 6a) the chemical shifts

for the CI1 and C61 are close. Hence it is difficult to differentiate

between 1',6'-dichloro- and 6,6'-dichlorosucrose. The major product of

the reaction has been reported to be 6,6'-dichlorosucrose (113). For

further evidence, mass spectral analysis was performed. The compound

(0.5g) was dissolved in lOmL pyridine and mixed with acetic anhydride

(5mL). The reaction mixture was kept overnight at room temperature and then evaporated. Trace amounts of pyridine were removed by co- evaporation with toluene. The residue was dissolved in chloroform, and the solution washed with water, dried over ^SO^ and evaporated to yield a sirup. The yield was 0.75g (90%). The mass spectrum of the 58

peracetylated product is shown in Figure 7. No molecular ion peak was

observed as it breaks into monosaccharides. If both the chlorines were

in the fructosyl moiety, then a peak should be observed for dichloro-

diacetyl fructose m/e 325. However, no such peak was observed. On the

contrary, a peak with m/e = 307 was observed which was due to

monochloro-triacetyl glucose or monochloro-triacetyl fructose. Thus it

is clear that the major product obtained in the reaction was 6,6'- dichlorosucrose. The rest of the peaks in the mass spectrum were due to

fragmentation of the monochloro-triacetyl monosaccharide as explained by

Kochetkov and Chizhov (127).

b. 6,6'-dibromo-6,6'-dideoxy sucrose.

Sucrose (2.1g) was dissolved in dry pyridine (lOOmL) by heating.

It was cooled in ice and trfphenyl phosphine (lOg) was added. Carbon tetrabromide (12.7g) was added gradually and mixed while the flask remained in ice. Reaction at 50°C for five hours produced 6,61-dibromo-

6,6'-dideoxy sucrose (Rf 0.29) as the major product along with minor amounts of 6-bromo-6-deoxy sucrose (Rf 0.09), 61-bromo-61-deoxy sucrose

(R^ 0.09), 6,1',6'-tribronio-6,l',6'-trideoxy sucrose (Rf 0.58) as well as other unidentified sugars (Rf 0.42 and 0.80). The dibromo sucrose was separated from the others by silicic acid column chromatography.

Occassionally rechromatography was required to remove traces of

(bromomethyl)-triphenyl-phosphonium bromide (Rf 0.09) which was detected by its UV absorbance on TLC with fluorescent indicator. The product was 1 ^ characterized by its C NMR spectrum (Figure 6c). The yield was 1.5g

(52%). 59

c. Monobromosucroses.

Keeping the molar ratios of reagents the same as in case of dibromo

sucrose, the time and temperature of reactions were varied in order to

achieve an alteration in the product ratios. Reaction at 40°C for one

hour produced 6-bromo- and 6'-bromo-sucroses as the only products (both

have Rf value 0.09); however, more than 50% of the sucrose remained

unreacted (R^ 0.0). Column chromatography did not resolve 6-bromo and

,6'-bromo sucroses, they remained contaminated with some sucrose and

(bromomethyl)-triphenyl-phosphonium bromide. They were separated by

HPLC, which did not completely resolve 6-bromo and 6'-bromo-sucroses as

shown in Figure 12a. For complete separation, the leading edge of the

peak (61-bromo-sucrose) and the trailing edge of the peak (6-bromo-

sucrose) were rechromatographed. By repeating this several times, 6-

bromo-6-deoxy-sucrose was obtained in pure form (final yield about 10%);

however, the 61-bromo-sucrose remained as a mixture with 6-bromo-sucrose

in about 1:1 ratio (yield about 15%). The NMR spectra of 6-bromo-

sucrose and of the mixture of 6-bromo and 61-bromo-sucroses are shown in

Figures 6e and 6f respectively.

d. 6,1',6'-tribromo-6,1',6'-trideoxy-sucrose.

Keeping the molar ratios of the reagents same as in case of dibromosucrose, when the reaction was conducted at 85°C for 45 minutes,

the proportion of 6,1',6'-tribromo-sucrose increased substantially, although the 6,6'-dibromo-analog was still the predominant product.

However, due to the short time and the high temperature of reaction,

some of the sucrose was left unreacted and some new products were 1 o Figure 6. C NMR Spectra of Halosucrose Derivatives.

Chemical shift values are from TMS.

a. Sucrose b. 6,6'-diehioro-6,61-dideoxy-sue rose

c. 6,61-dibromo-6,6'-dideoxy-sucrose d. 6,1' ,6'-tribromo-6,r ,6'-trideoxy-sucrose e. 6-bromo-6-deoxy-sucrose f. Mixture of 6-bromo-6-deoxy and 6'-bromo-6'-deoxy- sucroses

60 C' c.< C3C5 cr C3" C6 Q,' C2'

acetone

yWJj U

MJ!

u L

4/V|mv N. •*

i_JbJ 'uL—JAl Jl L

JL ft. :± 1A1 JL 1 ... i I iJ i i—»— / liitil.il '' I ' • '' j ' • ' • I 100 80 60 40

Figure 6. observed (Rf 0.52 and 0.80). Due to the high temperature of the

reaction the mixture was deep brown in color. After evaporation of the

pyridine the mixture was dissolved in chloroform and bound to silicic

acid (about 25g) which was filtered, washed with chloroform to remove

most of the PhgP, PhgPO and CBr^. It was then washed with solvent

system (I) which eluted only the brominated sugars. Most of the

coloring matter and unreacted sucrose remained bound to the silica. The

sugars were then separated by column chromatography with solvent system

(I) after which the 6,1' ,6'-tribromo-6,l',6'-trideoxy-sucrose was

further purified by preparative thin layer chromatography as explained

before. The final yield was about 5%. The NMR spectrum of 6,1',61 -

tribromosucrose is shown in Figure 6d.

e. Methyl,6-chloro-6-deoxy-a-(D)-gl ucopyranoside.

lOg of a-methyl glucoside were dissolved in 150mL dry pyridine.

27g of PhjP were dissolved in it. The solution was cooled in ice and

27mL of CCl^ were added. The reaction temperature was maintained at 60-

65°C for one hour. The product had a Rf value 0.52. After addition of

30mL of methanol, the reaction mixture was left at room temperature

overnight. After evaporating the solvent, non-carbohydrate materials

were precipitated with water. The sugar was purified on a silica gel

column and crystallized twice from ethanol/hexane. The yield of the 13 product after recrystal1ization was 2.7g (25%). The C NMR spectrum of the product is shown in Figure 8a. Figures 7. Mass Spectrum of 6,6'-dichloro-6,6'-dideoxy-sucrose

peracetate.

63 1

T"T T'TTTTnTTT' •50 400

CT 30?

87 218

00 :eo Figure 7 CTl 65

f. Methyl,6-bromo-6-deoxy-a-(D)-glucopyranoside.

lOg of a-methyl glucoside were dissolved in 150raL of dry

pyridine. 26g of Ph^P were added and the solution cooled in ice. 17g

of CBr^ were added and the mixture left at room temperature overnight.

After evaporating the solvent, non-carbohydrate materials were

precipitated with water. The sugar was purified on a silica gel column

and crystallized twice from ethanol/hexane. The yield of the product

after recrystal1ization was 3.3g (25%). The NMR spectrum of the

product is shown in Figure 8b.

g. Methyl,6-iodo-6-deoxy-a-(D)-glucopyranoside.

Methyl,6-bromo-6-deoxy-a-(D)-glucopyranoside (l.Og) and KI (2.8g)

were dissolved in lOmL of N,N-dimethyl formamide and reacted at 55°C for

16 hours. Both the starting material and the product had the same

value (0.52). However, the NMR of the product (Figure 8c) later

confirmed that the reaction was completed in 16 hours. A yellow

precipitate was formed during the reaction. This was filtered, the

filtrate was evaporated and chloroform was added to the residue. The

sugar dissolved in the chloroform whereas the inorganic compounds

remained insoluble and was filtered and the filtrate evaporated. The

product was crystallized twice from ethanol/hexane. The yield of the

recrystallized product was 0.75g (64%).

h. Methyl ,6-bromo-6-deoxy-a-(D)-galactopyranoside.

Methyl ,a-(D)-galactopyranoside (2g) was dissolved in 30mL of pyridine. PhgP (5.2g) was dissolved in it, the solution was cooled in ice and 3.4g of CESr^ were added. They were reacted at 60°C. The

reaction was slow. Even after 90 minutes of reaction only about 40% of

the starting material was converted to products (visual estimation by

TLC). Another unusual behaviour was the formation of two products (Rf

0.50, major and 0.47, minor). After 90 minutes 30mL of methanol were

added and the mixture was left overnight at room temperature. After

evaporating the solvent, non-carbohydrate materials were removed by

precipitating from water. The products were separated on a silica gel

column. Complete separation was not achieved. The major product was

characterized to be methyl,6-bromo-6-deoxy-a-(D)-galactopyranoside. It 11 was crystallized from ethanol/hexane. The yield was 50mg (2%). The C

NMR spectrum of the product is shown in Figure 8d. The minor product was not identified.

1. Methyl ,6-bromo-6-deoxy-a-(D)-mannopyranoside.

Methyl ,a-(D)-mannopyranside (5g) in dry pyridine (lOOmL) was mixed with 14g of Pf^P. The mixture was cooled in ice and after addition of

9.5g CBr^ it was kept at 60°C for four hours. 75 mL methanol were added to it at this point and the mixture was left at room temperature overnight. Only one product (R^ 0.62) was observed on TLC. After evaporating the solvent, the non-carbohydrate materials were removed by precipitating from water. The product was purified on silica gel column with solvent system (I). The sugar could not be crystallized. It was contaminated with water soluble (bromomethyl)-triphenyl phosphonium bromide which was removed by passing the sample through a column of anion exchange resin AG1X-8, 0H"form, 50-100 mesh. The effluant was Figure 8. 13C NMR Spectra of Halo Derivatives of Monosaccharides.

Chemical shift values are from TMS.

a. Methyl ,6-chloro-6-deoxy-a-(D)-glucopyranoside b. Methyl,6-bromo-6-deoxy-a-(D)-gl ucopyranoside c. Methyl,6-iodo-6-deoxy-a-(D)-glucopyranoside d. Methyl,6-bromo-6-deoxy-a-(D)-galactopyranoside e. Methyl ,6-bromo-6-deoxy-a-(D)-mannopyranoside

67 ft'fi

MAI _ _ A —

i • i ' ' i i • i i i • • i • i i • i i i i i i i i_ ' • • i i i • • i i • i i i i • ' ' i i i i i i i ' i ' ' i 100 80 60 40 20 0

Figure 8. 69

evaporated to a sirup which was dried in a dessicator. The yield was

3.0g (45%). The *3C NMR spectrum of the compound is shown in Figure 8e .

j. Methyl,6-deoxy-(D)-gal actosides.

700mg of (D)-fucose (i.e. 6-deoxy-(D)-galactose) were dissolved in

30mL of dry methanol containing 0.6mL of acetyl chloride. The reaction

mixture was refluxed for 30 hours after which the solvent was removed by

evaporation. There were four products formed with Rf values of 0.56

(3-methyl furanoside), 0.44 (a-methyl furanoside) and 0.23-0.31 (mixture

of a-and 3-methyl pyranosides). From the NMR spectrum of the

mixture (not shown) it appeared that the ratio of amounts of products

was a-furanoside/3-furanoside/a-pyranoside/3-pyranoside : 1/3/12/6. The

sugars were separated on a 2cm X 50cm column of anion exchange resin

AG1X-8, 50-100 mesh, 0H"form with water as solvent at a flow rate of

2.2mL/niinute with a 3cm hydrostatic head The separation of sugars by

ion exchange column has been reported previously by Evans and Angyal

(129). They had employed AG1X-2, 200-400 mesh for separating methyl

allosides. This experiment shows that methyl ,6-deoxy-(D)-galactosides

can be successfully separated on a column of the anion exchange resin

AG1X8, 50-100 mesh. Fractions were collected at 8 minutes intervals.

Fractions 4-7 contained a mixture of a-and 3-pyranosides, fractions 32-

40 contained pure methyl,6-deoxy-3-(D)-galactofuranoside (60mg, yield

8%) and fractions 8-31 contained a mixture of all four products.

Fractions 8-31 were pooled, concentrated and rechromatographed to give

pure methyl,6-deoxy-a-(D)-galactofuranoside (lOmg, yield 1.3%) and a mixture of a-and 3-pyranosides which were separated from their mixture by fractional crystallization from dry p-dioxane. Figure 9. NMR Spectra of Methyl Glycosides (D)-Fucose.

Chemical shifts are from TMS.

Figure 8a. Methyl,6-deoxy-a-(D)-galactopyranoside

b. Methyl,6-deoxy-B-(D)-gal actopyranoside

c. Methyl,6-deoxy-a-(D)-galactofuranoside

d. Methyl,6-deoxy-S-(D)-galactofuranoside

70 71

Figure 9. Methyl ,6-deoxy-a-(D)-galactopyranoside crystallized first (260mg, yield

34%) and then methyl,6-deoxy-0-(D)-galactopyranoside crystallized

(lOOmg, yield 13%). The compounds were characterized by comparing their

13C NMR spectra with published ones (128). The spectra are

shown in Figures 9a-d.

k. Methyl, 6-deoxy-(L)-mannosides.

3.3g of (L)-Rhamnose.H20 (i.e. 6-deoxy-(L)-mannose) were dissolved

in 30mL of dry methanol; 0.47mL of acetyl chloride were added dropwise.

The mixture was refluxed overnight. Two products were seen on TLC : Rf

0.54 (methyl,6-deoxy-3-(L)-mannopyranoside) and 0.27 (methyl,6-deoxy-a- 1 ^ (L)-mannopyranoside). From the C NMR spectrum of the mixture (not

shown) it appeared that the ratio of a- to 0- anomer was 1:2. A very

small amount of another product (Rf 0.66) was also seen on TLC. This

was not identified. The compounds were separated on silica gel to give methyl ,6-deoxy-a-(L)-mannopyranoside (700mg, yield 20%) and methyl,6-

deoxy-3-(L)-mannopyranoside (l.lg, yield 31%). The nmr spectra of

the compounds are shown in Figures 10a and 10b. These were compared

with the published spectra (130).

1. Methyl,(D)-al1osides.

Methyl allosides were synthesized according to the method of Evans

and Angyal(140). Strontium chloride (1.78g, dried in an oven at 90°C

for several days) and methyl orthoformate (0.61mL) were dissolved in dry methanol (20mL) in which acetyl chloride (0.72mL) had been dissolved.

3-(D)-allose (lg) was added and the mixture refluxed for 72 hours. 73

Sodium acetate (2g) was added to the cooled solution, and methanol was evaporated at reduced pressure. Acetic anhydride (15mL) was added, and the mixture was boiled gently until the bumpy solid became a fine powder. The cooled mixture was diluted with water (30mL), stirred for

30 minutes and extracted with chloroform (lOmL and 2 X 5mL). The extracts were combined, washed by stirring with water (2 X 30mL) and saturated aqueous sodium hydrogen carbonate (30mL), dried with MgSO^ and evaporated. A solution of the residue in 20mM methanolic sodium methoxide was left for 16 hours at room temperature and then evaporated. The product was eluted with water from a column (2.4cm X

43cm) of Bio Rad AG1X-2 resin, 0H"form 200-400 mesh, collecting 20mL fractions which were monitored by TLC. The hydroxide form had been obtained from the chloride form by equilibration with sodium hydroxide as explained before. Fractions 10-13 had Rf value of 0.08 and contained methyl,a-(D)-allopyranoside (50mg, yield 4.6%). Fractions 14-21 had R^ value of 0.17 and contained a mixture of methyl a- and 3-(D)- allofuranoside (14mg, yield 1.4%). The yields were low compared to values reported in the literature probably because the allose was not dried prior to reaction. The compounds were characterized by comparing their

13C NMR spectra (Figure 10c, lOd) with published spectra (128).

m. Methyl,6-azido-6-deoxy-a-(D)-gl ucopyranoside.

Methyl ,6-bromo-6-deoxy-a-(D)-glucopyranoside (5g) was added to an ice cold mixture of acetic anhydride (25mL) and dry pyridine (35mL).

The solution was allowed to stand at room temperature for 18 hours and Figure 10. NMR Spectra of (L)-Rhamnose and (D)-AHose.

Chemical Shifts are from TMS.

a„ Methyl,6-deoxy-a-(L)-mannopyranoside b. Methyl,6-deoxy-3-(L)-mannopyranoside c. Methyl ,a-(D)~al 1opyranoside d. Methyl,a-(D)-allofuranoside Figure 10. 7 6

then poured with stirring into 200mL of ice and water. There was an

immediate crystallization of methyl ,2,3,4-tri-0-acetyl-6-bromo-6-deoxy-

oc-(D)-glucopyranoside (6.9g, yield 91%).

The acetylated product (6.9g) was dissolved in 50mL dry DMF

containing 7.5g sodium azide and stirred at 50°C for 18 hours. The

reaction mixture was then poured with stirring into 150mL of ice and

water whereupon the product, methyl,2,3,4-tri-0-acetyl-6-azido-6-deoxy-

a-(D)-glucopyranoside, crystallized. It was recrystal1ized from absolute ethanol. The yield of the recrystal!ized product was 5.6g

(90%).

The product was deacetyl ated as follows : The above product (5.6g) was dissolved in 50mL dry methanol. Ten drops of 1M sodium methoxide in methanol were added to it and allowed to react at room temperature for two hours. The solution was neutralized wjith cation exchange resin AG

50W-X8, hydrogen form, 20-50 mesh. The resin was filtered and the solution was evaporated to dryness to give an amorphous powder (2.0g, yield 55%). The NMR spectrum of the compound is shown in Figure

11a. The compound was also characterized by the infrared strong absorbance peak (131) of the azido- group at 2100cm"* (Figure 13).

n. Methyl,3,6-anhydro-a-(D)-gl ucopyranoside.

Fischer and Zach (132) described the synthesis of Methyl,3,6- anhydro-3-(D)-glucopyranoside starting from methyl ,6-bromo-6-deoxy-3-

(D)-glucopyranoside. Haworth, Owen and Smith (133) synthesized methyl,3,6-anhydro-a-(D)-glucopyranoside starting from methyl,6-0-p- tolyl sulfonyl-a-(D)-glucopyranoside. In this investigation the method of Fischer and Zach was applied to synthesis of the a-anomer. 77

Methyl,6-bromo-6-deoxy-a-(D)-glucopyranoside (0.9g) was dissolved

in 50mL 1M NaOH and heated at 85-90°C for two hours. The product had an

Rf value of 0.48 while that of the starting material 0.52. The

closeness of the values made it difficult to differentiate between them 1 ^ on TLC. However, the C NMR of the product, taken at a later time

(Figure lib), showed that the reaction had gone to completion. The mixture was neutralized with solid carbon dioxide and evaporated to

dryness. The residue was extracted with acetone and filtered.

Evaporation af the acetone gave methyl,3,6-anhydro-a-(D)-glucopyranoside

as a glass (0.6g, yield 72%).

o. Methyl ,2-deoxy-a-(D)-glucopyranoside.

A mixture of methyl glycosides of 2-deoxy glucose in their peracetate forms was provided by Dr. Derek Horton. 1.5g of it were dissolved in lOmL of dry methanol. 10 drops of 1.4M methanolic sodium methoxide were added and left to react overnight at room temperature. A

TLC in solvent system (I) showed one major product (Rf 0.40, about 60%), one minor product (R^ 0.62, about 30%) and several other products (R^

0.50, 0.68 and 0.77, about 10%). The products were separated on a silica gel column (3cm X 80cm) to obtain the product with R^ 0.40. The

13c NMR spectrum of this compound revealed that it is a mixture of methyl,2-deoxy-a- and 3-(D)-glucopyranosides (about 70:30). The other products were not characterized. Since the two compounds had the same

Rf value, their separation was attempted by HPLC on a Parti si1 M9 column. The elution profile is shown in Figure 12b. The small peak at fractions 17-22 represent the solvent and other impurities . The large 78

peak following it is a mixture of the a-and 3-pyranosides. So the

desired resolution was not achieved. However, the peak has a shoulder

at fraction 27. This is due to the p-anomer. The first three fractions

(22-24) of the peak were pooled together. The NMR spectrum (Figure

11c) of this shows that it is highly enriched in the a-anomer. The

mixture is estimated to contain about 90% of the a-anomer and 10% of the

g-anomer. The yield of the enriched a-anomer was 160mg (18.2%). This

mixture was employed in the acceptor reaction.

p. (D)-glucal.

3,4,6-tri-0-acetyl-(D)-glucal (2g) was dissolved in lOmL of dry methanol and 10 drops of 1.4M sodium methoxide in methanol were added.

The mixture was left at room temperature overnight. Two products were observed with values of 0.43 (major) and 0.69 (minor). After neutralization of the reaction mixture and evaporation of the solvent, the sugars were separated on silica gel column with solvent system (I) to give (D)-glucal, Rf 0.43 (0.75g, yield 20%) as a sirup. The minor product was not identified. (D)-glucal was characterized by its NMR spectrum (Figure lid) which agreed with the published spectrun (128).

q. 6-deoxy-(l.)-gl ucal (also known as (L)-rhamnal).

2,3-di-0-acetyl-6-deoxy-(L)-glucal (2g) was dissolved in lOmL of dry methanol and 10 drops of 1.4M sodium methoxide in methanol were added. The mixture was left at room temperature overnight. There was one product (R^ 0.71) seen on TLC. The solution was neutralized with AG

50V/ X8, 20-50 mesh and filtered. The filtrate was evaporated to dryness and the product crystallized twice from ethyl acetate/hexane to give 6- 1 ^ deoxy-(L)-glucal (lg, yield 30%). The C NMR spectrum of the compound

is shown in Figure lie.

r. Methyl ,2,3,6-trideoxy-a-(L)-erythro-hex-2-enopyranoside.

Methyl ,4-0-acetyl-2,3,6-trideoxy-a-(L)-erythro-hex-2-enopyranoside

(300mg), which was kindly provided by Dr. D. Horton, was dissolved in

5mL of dry methanol. 5 drops of 1.4M sodium methoxide in methanol were

added to it. The mixture was left at room temperature overnight. The deacetylated product had a Rf value of 0.85. The mixture was neutralized with AG 50W X8, filtered and evaporated to give the title compound as a sirup (120mg, yield 52%). The NMR spectrum of the compound is shown in Figure llf.

s. 6'-bromo-6'-deoxy-maltose.

The synthesis was first attempted by the halogenation method of

Anisuzzaman and Whistler (113). When maltose, Ph^P and CBr^ were reacted in pyridine at 60°C no sugar product was observed on TLC even after 24 hours. However, the reaction mixture had turned black immediately after adding the CBr4. The maltose used had one mole of water per mole of maltose. This water was thought to be responsible for the lack of reaction. The attempted removal of the water by lyophil1ization or by drying of the compound in a vacuum oven was not succesful. When the reaction was done at 45°C in the presence of anhydrous MgSO^ there were two products formed (R^ 0.14 and 0.49). Figure 11. NMR Spectra of Other Analogs of g-methyl Glucoside.

a. Methyl,6-azido-6-deoxy-a-(D)-glucopyranoside b. Methyl ,3,6-anhydro-a-(D)-glucopyranoside c. Methyl,2-deoxy-a-(D)-glucopyranoside d. (D)-glucal e. 6-deoxy-(L)-glucal f. Methyl,2,3,6-trideoxy-a-(L)-erythro-hex-2-enopyranoside

80 I.... I.... I... I.... I.... I.... I .... I .... I 100 80 60 40 20

• • I * * • • I • • • • I • • • • I • > * • ^ • ' •' t •11' I •1 i1I * * * * I * * •' I • • • • I • * * * 11 •1 • 11 •1111 1 •11' iso 120 90 60 30

Figure 11. Figure 12. HPLC Elution Profiles.

HPLC was performed as described in Methods. In a single run about

50mg sugar in .2mL was injected. Solvent system (V) was employed as the

eluant at a flow rate of 1.5mL per minute and fractions were collected

at 30 seconds intervals.

a. Mixture of 6-bromo- and 6'-bromo- sucroses. b. Mixture of a- and P- methyl glycosides of 2-deoxy-glucopyranose.

82 T T T T T o 4 8 12 16 20 24 28 0 4 Time (min) Time (min)

Figure 12. CO CO Figure 13. IR Spectrum of methyl,6-azido-6-deoxy-q-(D)-g1ucopyranoside.

84 MICRONS

5I 9

2500 ' 1800 1400 1000 WAVENUMBER (CM-1)

Figure 13.

COen 86

However, even after 6 hours of reaction, only about 25% of the maltose

was converted to products. These were separated on a silica gel column

with solvent system (I) and finally purified on a column of AG1X-8, 50-

100 mesh, 0H"form to remove the (bromomethyl )-triphenyl-phosphoniurn 1 ^ bromide. They were characterized by their C NMR spectra and found to

be methyl ,a-(D)-glucopyranoside (Rf 0.14) and methyl,3,6-anhydro-a-(D)- glucopyranoside (R^ 0.49). The slower moving product was crystallized

from ethanol and had a melting point of 166°C; the mixed melting point with methyl,a-(D)-glupyranoside was also 166°C. Though the reaction was not successful, it is interesting to note that a-methyl glucoside was

formed in the reaction. How the methyl glycoside was formed in the absence of any mathanol is not understood.

6'-bromo-6'-deoxy maltose was then synthesized by the method of

Takeo and Shinmitsu (134). A mixture of maltose monohydrate (5.0g), a,a-dimethoxy toluene (8mL) and p-toluenesulfonic acid monohydrate

(0.25g) in dry N,N-dimethyl formamide (38mL) was stirred at 50°C under a pressure of about 30mm Hg (created by water suction). After 5 hours of reaction, the TLC in chloroform/methanol : 3/1 showed one major product

(Rf 0.32) and several faster moving products, in addition to small a amount of unreacted maltose (Rf 0.01). Acetic anhydride (40mL) and sodium acetate (5g) were added and the reaction mixture was refluxed for

20 minutes. A TLC of the mixture in chloroform showed a major product

(Rf 0.25) and two minor products (Rf 0.16 and 0.09). The mixture was cooled and poured into crushed ice with stirring. The precipitate was filtered and washed with water. The residue was dissolved in chloroform and successively washed with aqueous sodium hydrogen carbonate and water. It was then dried over MgS04, filtered and evaporated. It was

crystallized twice from ethanol to give 4',6'-benzylidene maltose

hexaacetate (Rf 0.25 in CHCI3). The yield was 4.8g (51%).

A mixture of 4',6'-benzylidene maltose hexaacetate (3.0g), N-bromo-

succinimide (0.94g) and barium carbonate (previously dried in oven, 6g) in anhydrous CC14 (60mL) and 1,1,2,2-tetrachloroethane (60mL) was boiled

with stirring for 90 minutes under normal room illumination. A TLC in benzene/ethyl acetate : 3/1 showed a major product (Rf 0.38) and two minor products (Rf 0.22 and 0.11). The mixture was filtered through a layer of celite filter aid. The inorganic residue was washed with chloroform. The filtrate and washings were combined and evaporated.

The compounds were separated on a silica gel column with solvent system

(II) to give 4'-0-benzoyl-6'-bromo-maltose hexaacetate as a white sticky mass. The Yield was 2.4g(72%).

4'_0-benzoyl-6'-bromo-maltose hexaacetate (2g) was dissolved in

20mL of dry methanol. 3mL of 1.4M sodium methoxide in methanol were added and the mixture was left at room temperature overnight. A TLC in solvent system (I) showed a major product (Rf 0.05) and two minor products (Rf 0.1 and 0.0). It was separated on silica gel column to give pure 6'-bromo-maltose (Rf 0.05). The yield was0.6g, (56%). The 1 O C NMR spectra of maltose and 6'-bromo-maltose are shown in Figures 14a and 14b, respectively.

12. Enzymatic Synthetic Procedures.

Isomaltose oligosaccharides were prepared by partial hydrolysis of dextran with dextranase which yields isomaltose as the final product Figure 14. NMR Spectra Bromo-Maltose.

a. Maltose b. 6'-bromo-61-deoxy-maltose

88 89 (135). Due to partial hydrolysis, a mixture of oligosaccharides were

produced. Since isomaltose and isomaltotriose are available

commercially the hydrolysis conditions were designed such as to obtain a

maximum yield of oligosaccharides with a degree of polymerization from 4

to 10 and at the same time, a minimum possible yield of 2 and 3.

Dextran T-10 was chosen as the starting material because it consists mostly of a-( 1+6) linked glucose with only 3% a-(l->-3) links as seen by

*H NMR spectroscopy (not shown).

First a test reaction was undertaken to determine the reaction time for the best yield of the desired sugars. lOOmg of dextran T-10 was dissolved in lmL of 0.1M phosphate buffer pH 6.0 and IOmL dextranase (10 units in 0.1M phosphate buffer, pH 6.0) were added. The reaction mixture was kept at 37°C. At different times 5jjL aliquots were removed and spotted on Whatman 1MM paper. It was developed seven times by repetitive chromatography in solvent system (III) as described before.

The products were detected by the silver nitrate dip procedure. Based on the results of this experiment a 15 minutes reaction time was found to be suitable for the preparation of isomaltose oligosaccharides.

200mg dextran T-10 were dissolved in .1M phosphate buffer, pH 6.0,

20uL dextranase (20 units) were added and the mixture was reacted at

37°C for 15 minutes. It was then heated at 100°C for 10 minutes, cooled and loaded onto a P2 (minus 400 mesh) column, 2cm X 120cm and eluted with water as discussed before. RESULTS AND DISCUSSIONS

A. Interaction of Halosucrose Derivatives with Dextransucrase.

The characteristics of the active site of the enzyme are poorly

understood, although a number of investigators have attempted to

identify catalytically important functional groups. Some studies have

suggested that protonation of a functionality is important for activity

(136-138). Inoue and Smith (136) reported that partially oxidized

dextran is a potent inhibitor of dextransucrase and proposed that the

inhibition results from the interaction of dialdehydes with reactive

groups that are close to the dextran binding site. Photochemical

oxidation, in presence of methylene, blue or rose bengal, has been shown

to inactivate the enzyme (137). This was attributed to the modification

of histidyl residue at the active site.

Several groups (75-77) have shown that a-(D)-fluoroglucose is an

ideal analog of sucrose since it serves as a glucosyl donor, and has

and Vmax values comparable to those of sucrose. Sucrose analogs, such

as 6-azido-sucrose, have been reported to be inactivators (139). Mono-

and di-amino sucrose have been reported to be inhibitors and the

suggestion has been made that their inhibition is due to the perturbation of basic groups at the active site (140). Another class of inhibitory sucrose analogs is the halo-derivatives. In a patent report (141) it was indicated that 6,61-dichloro- and 6,6'-dibromo-sucrose could inactivate the enzyme. The halo-derivatives appeared to have the requisite characteristics of a compound that might cause modifications at the active site. It was therefore decided to examine these more carefully in this investigation.

The following halo-derivatives of sucrose have been synthesized as described under Methods : 6-bromo-6-deoxy-sucrose; 6,6'-dibromo-6,6'- dideoxy-sucrose; 6,1',6'-tribromo-6,r,6'-trideoxy-sucrose and 6,6'- dichloro-6,6'-dideoxy-sucrose. The ability of the sucrose analogs to inactivate dextransucrase was evaluated in two types of experiments.

Under one protocol the enzyme was preincubated with varying concen trations of the compounds for 60 minutes and then assayed for activity.-

Control preincubations were run under each set of conditions, whereby the enzyme was allowed to stand in buffer for the prescribed period.

The second protocol involved preincubation for varying times, at fixed concentrations of the reagents. The results of the studies using the first protocol are shown in Figure 15. Since the concentrations required for 50% inactivation ranged between 180mM for the 6-bromo- derivative and 482mM for the 6,6'-dichloro compound, it did not appear that any of these were exceptionally effective inhibitors. This was substantiated in studies carried out under the second protocol (Figure

16), in which only small differences between the control and experimental reactions are seen in six hour reactions. Furthermore, where differences are observed, it appears that the shapes of the curves are very similar, and the zero time points reflect the changes. 6'- Figure 15. Inactivation of Dextransucrase as a Function of

Concentration of Halosucrose Derivatives.

Dextransucrase (0.05 units) was incubated with the indicated 6- bromo-sucrose (©—-©), 6,6'-dibromosucrose (0 0), 6,1',6'-

tribromosucrose (O—O), and 6,6'-dichlorosucrose (d • ) at the indicated concentrations in the presence of 0.1M phosphate buffer, pH

6.0, in a final volume of 40pL at 37°C for 60 min. It was then reacted with [^Cjsucrose (final concentration 50mM, 1.5 X 10^ dpm) in the presence of 25mM Dextran T-10 and 0.1M phosphate buffer, pH 6.0, at 37°C for 5 min. Reaction was stopped by heating at 100°C for 3 min. 10uL from each was spotted on Whatman 1MM paper and Chromatographed and counted as described in Methods.

93 100

60 120 180 260 Figure 15. Concentration (mM) Figure 16. Inactivation of Dextransucrase by Halosucrose Derivatives

as a Function of Preincubation Time.

Dextransucrase (0.5 units) was incubated with A. 6-bromosucrose

(170mM); B. 6,61-dibromosucrose (160mM); C. 6,1' ,61-tribromosucrose

(90mM); and D. 6,6'-dichlorosucrose (190mM). Incubations were carried

out in the presence of 0.1M phosphate buffer, pH 6.0 at 37°C in a total

volume of 0.2mL. Aliquots (20pL) were removed at the indicated times

and reacted with [^C]sucrose (final concentration, 67mM, 1.5 X 10^ dpm)

in the presence of 33 mM Dextran T-10 and 0.1M phosphate buffer, pH 6.0,

in a total volume of 30pL at 37°C for 5 min. Reactions were stopped by

heating at 100°C for 3 min. 10uL from each was spotted on Whatman 1MM

paper and chromatographed and counted as described in Methods.

Reactions were carried out in presence (A—A) and absence (O—O) of the halosucrose derivatives.

95 -o

0.0

120 ffl to !«0 <50 120 160 200 ?«0 520 Tina ol prOIBCato'' OQ ImtA| Ti rs»® ot protacuboiion (mm1

I.S

0.0

40 •20 so 200 200 320 eo 120 160 200 ?«0 200 Tiiao ot preincubation (rami Figure 16 Timfl o( or a mc*ibol101 97

bromo-6'-deoxy-sucrose was obtained as a 1:1 mixture with the

corresponding 6-bromo-derivative. The mixture was also employed in

inactivation studies in both of the above ways. Again, only a small

change in activity was observed, indicating that 6'-bromo-61-deoxy-

sucrose is also not an effective inactivator. If these halo-derivatives

were covalently modifying the enzyme in such a way as to cause

inactivation, much greater differences would have been expected. The degrees of inactivation at 240mM concentrations are shown in Table 1.

6-bromo-sucrose produced 62% inactivation and was the most effective of the four compounds tested. The results in contrast to the patent report by Robyt and Zikopoulos (141), who indicated that 6,6'-dibromo-sucrose, and 6,61-dichl oro-sucrose form dead end complexes with the enzyme at much lower concentrations. For example, 150mM concentrations of 6,6'- dibromo-sucrose or 6,6'-dichloro-sucrose were reported to yield 100% and

80% inactivation respectively, in 10 minutes (pH not mentioned). The evaluation of the effectiveness of the analogs involved preincubation of the enzyme with the test compounds and was therefore similar to the method employed in the present investigation. A similar degree of inactivation was also observed in our laboratory in preliminary investigation with partially purified halo-sucrose derivatives. For example, preincubation of the enzyme with 30mM 6,6'-dichloro-sucrose for

90 minutes gave 50% inactivation. However, the sugar contained an 1 10 impurity which was aromatic as seen by UV absorbance and by H and C

NMR. This impurity was removed from the sugar as explained in Methods

(p 45). The purified sugar no longer exhibited the same degree of inactivation. The impurity was crystallized from acetone and characterized as (chloromethyl) triphenyl phosphonium chloride as

described below.

The different side products that have been reported in the reaction

of PhgP and carbon tetrahalide (114-117) are the following :

+ Ph3P0; Ph3PX2; Ph3P=CHX; Ph3P=CX2; [Ph3P-CH2X] X";

+ + + [Ph3P-CHX2] X"; [Ph3P-CX3] X~; [Ph3P=CH-PPh3] X~; and

2+ [Ph3P-CHX-PPh3] 2X".

Where X represents a halogen atom. Depending on the conditions of the

reactions one or more of these are produced. Mass spectal analysis was

performed on the impurity that was isolated from the reaction (Figure

18). The peaks with m/e values below 262 represent the mass spectrum of

Ph3P (m/e 262). Thus only peaks with mass greater than 262 will be

useful for characterization. However, since the compound decomposed at

high temperature (about 220°C), these peaks are not very intense. The

highest mass observed is in the region m/e 343-346. Among those

mentioned above the compounds which have molecular weight in this region

+ are (1) [Ph3P-CH2Cl] Cl" (m/e 346) and (2) Ph3P==CCl2 (m/e 344). The

impurity isolated from the reaction gave a white precipitate with AgN03

in nitric acid. This property corresponds to that of (chloromethyl)-

triphenyl-phosphonium chloride(l). The NMR spectrum of the compound

(Figure 17) shows a doublet at 5.28 ppm with a coupling constant of

6Hz. This agrees with the coupling constant of "^P-C-H. The integral of this doublet relative to the protons in the aromatic region corresponds to compound 1. Compound 2 does not have any P-C-H bond. An authentic sample of 1 was obtained from Aldrich Chemical Co. This exhibited a NMR spectrum identical to that in Figure 14. It was Figure 17. *H NMR Spectrum of (chloromethyl)-triphenyl-phosphonium-

chloride.

Chemical shifts are from TMS. Solvent is D2O.

99 K

Figure 17. Figure 18. Mass Spectrum of (Chloromethyl)-tripheny1-phosphonium-

chloride.

101 nt n 11m11(i n im hi vvwt 20 ;60 130 40M

IS3

40. 103 30. 20.

99 219 rrrrf 180 00 103

therefore concluded that the impurity in the 6,6'-dichloro-sucrose that

was responsible for inactivation of the enzyme was (chloromethyl)-

tri phenyl-phosphoni um chloride.

Inactivation of the enzyme by (chloromethyl)-triphenyl-phosphonium

chloride was measured using the same procedure employed with halosucrose

derivatives. Inactivation as a function of concentration and time are

shown in Figure 19 and 20 respectively. From Figure 19 it can be seen

that there is 50% inhibition at 8mM concentration of (chloromethyl)-

triphenyl-phosphonium chloride. From Figure 20 it is evident that there

is very little inactivation of the enzyme with increasing time of

preincubation with 5mM concentration of the inhibitor. Thus it appears

that the compound is an inhibitor and not an inactivator. The type of

inhibition has not been investigated. However, it was found that

inhibition was not specific for dextransucrase. For example, at 33mM concentration it inhibited the activity of assay mixture used in the

second step of dextransucrase assay to an extent of 67% The assay mixture contains the enzymes hexokinase, phosphoglucose isomerase and glucose-6-phosphate dehydrogenase.

Inhibition of dextransucrase by this compound explains why inhibition was observed with an impure preparation of 6,6'-dichloro-

6,6'-dideoxy-sucrose. This may also be the reason for the difference in results obtained in this investigation and the results obtained by Robyt and Zikopoulos (141). Figure 19. Inactivation of Dextransucrase as a Function of

Concentration of (chloromethyl)-triphenyl-phosphonium-

chloride.

Dextransucrase (0.05 units) was incubated with the compound at the indicated concentration in the presence of 0.1M phosphate buffer, pH

6.0, in a final volume of 30pL, at 37°C for 60 min. It was then reacted Id ^ with [ C]sucrose (final concentration 50mM, 1.5 X 10 dpm) in the presence of 25mM Dextran T-10 and 0.1M phosphate buffer, pH 6.0, in a total volume of 40uL at 37°C for 5 min. Reaction was stopped by heating at 100°C for 3 min. 10uL from each were spotted on Whatman 1MM paper and chromatographed and counted as described in Methods.

104 dpm(%) at origin

o so . I..

o o 3 O O -n £ 3 1 (D Q)

5' 3 to 3 *

OJ (O

SOI Figure 20. Inactivation of Dextransucrase by (chloromethyl)-triphenyl-

phosphonium-chloride as a Function of Preincubation Time.

Dextransucrase (0.5 units) was incubated with the 5mM concentration of the compound in the presence of 0.1M phosphate buffer, pH 6.0 at 37°C in a total volume of 0.2mL. Aliquots (20uL) were removed at the indicated times and reacted with [^Cjsucrose (final concentration,

67mM, 1.5 X 10^ dpm) in the presence of 33mM Dextran T-10 and 0.1M phosphate buffer, pH 6.0, in a total volume of 30uL at 37°C for 5 min.

Reaction was stopped by heating at 100°C for 3 min. 10uL from each were spotted on Whatman 1MM paper and chromatographed and counted as described in Methods. Reactions were carried out in the presence

(A A) and absence (O—O) of the phosphonium compound.

106 24 36 48 Time (min)

Figure 20. O 108

The apparent lack of active site modification by the halo-sucrose derivatives left the question of the mode by which these compounds

served to inhibit dextransucrase. It was therfore decided to determine if they acted as reversible inhibitors. A study of the kinetics of the reaction, as a function of sucrose concentration in the presence of fixed amounts of the analogs, was carried out. Since insufficient amount of 6,1' ,6'-tribromo-6,l',61-trideoxy- sucrose was available to carry out this type of analysis, only the other three substances were examined. Double reciprocal plots of the results, shown in Figure 21 indicate that 6,6'-dichloro-sucrose, and 6-bromo-sucrose are competitive inhibitors, while the 6,6'-dibromo-derivative displays mixture of competitive and non-competitive inhibition. Inhibition constants for these compounds were determined, and are given in the figure. It can be seen that none of the analogs are effective inhibitors, since their 1s vary between 47mM and 160mM. This must be compared with the Km for sucrose (5mM) under the reaction conditions used (49).

Previous studies by Grier and Mayer(77) attempted to evaluate substrate specificity by exploiting the observation that a-fluoroglucose serves as an effective donor substrate, and a competitive inhibitor. A series of a-fluoro-sugars, which were glucose epimers or derivatives were studied. Most of these also proved to be weak competitive inhibitors. Of specific interest to the present study were a-fluoroglucose, 6-deoxy-a-fluoroglucose, and a-fluoroxylose, which have

1 K.j s of 9.3mM, 5.3mM and 4.2mM respectively (77). Thus it appears that Figure 21. Kinetics of Dextransucrase Inhibition by Halosucrose

Derivatives.

Dextransucrase (0.05 units) in case of 6,6'-dich1orosucrose and

0.025 units in case of 6-bromosucrose) was reacted with the indicated concentrations of [^C]sucrose (1.5 X 10^ dpm in each reaction) in the presence of the indicated concentrations of the inhibitors and 25mM

Dextran T-10 in 0.1M phosphate buffer pH 6.0 in a total volume of 40yL, at 37°C for 2 min. Reactions were stopped by heating at 100°C for 3 min. and aliquots (10pL) were spotted on Whatman 1MM paper and chromatographed and counted as described in Methods. The percentage of the applied counts that remained at the origin was determined and plotted as the reciprocal of the rate against the reciprocal of the sucrose concentration.

109 8.0- 6-bromosucrose mM

6.0 84

4.0

2.0 K, = 47 mM

6,6'-dibromosucrose mM 105 63

2.0

K. = 160 mM

2.4 - 6,6-dtchlorosucrose , 97 2 0 - 1 y 58 © o 29 £ 1.6 - =1 // 0 e 1.2 I

- 0.8 K, = 154 mM

' ' i i i 0 0 2 0 4 0.6 08 1.0 Figure 21. (SucroseJ mM Ill

TABLE 1

List of chemical shifts of synthesized sugars.

Figure no. Spectrin Chemical Shifts (ppm from TMS)

6 a 104. 93.1 82.4, 77.7 75.2, 73.7, 73.4, 72. 70.4 63.4, 62.6 61.3

6 104. 93.3 81.7, 77.6 76.8, 73.3, 72.7, 72. 71.3 62.4, 46.2 45.5

6 104. 93.3 81.5, 77.8 77.8, 73.2, 72.6, 72. 72.1 62.4, 34.6 34.6

6 103. 93.5 81.2, 78.0 77.4, 74.1, 65.9, 65. 34.6 34.5, 33.8 » O - O r L • 104. 93.1 82.3, 77.5 o 73. 1, 72.3

6 c • 72. 71.7 63.5, 62.6

6 104. 93.3 81.6, 78.0 77.7, 73. 6, 73.6 72. 70.5 62.1, 61.5 34.2

a 100. 73.9 72.2, 71.6 71.3, 56.1, 45.3

b 100. 73.7 72.5, 72.1 71.2, 56. 2, 34.1

c 100. 74.5 73.5, 72.2 71.2, 56. 4, 7.8

d 100. 71.8 70.7, 70.3 68.9, 56. 2, 31.7

e 102. 72.5 71.5, 71.0 70.0, 56. 0, 34.4

9 a 100. 72.8 70.6, 68.9 67.4, 56.1, 16.2

9 b 104. 73.9 72.3, 71.8 71.4, 58.1, 16.3

9 c 102. 86.2 77.6, 76.0 70.4, 56. 0, 18.5

9 d 109. 88.2 81.9, 78.2 68.4, 55. 7, 19.2

10 a 101. 73.0 71.3, 71.0 69.3, 55. 6, 17.6

10 b 102. 73.6 73.1, 73.0 71.3, 57. 7, 17.6 112

TABLE 1 continued

Figure no. Spectrun Chemical Shifts (ppm from TMS)

LO c 100.3, 72.1, 68.3, 67.9, 67.1, 61.7, 56.4

0 d 103.7, 85.7, 72.5, 72.0, 69.7, 63.2, 56.3,

a 100.3, 73.9, 72.2, 71.4, 71.4, 56.2, 51.9,

b 99.4, 76.3, 71.9, 71.7, 70.2, 69.7, 58.3,

c 99.3, 73.1, 72.1, 69.2, 61.8, 55.4, 37.6,

d 144.7, 103.8, 79.1, 69.9, 69.3, 61.2,

e 144.7, 104.5, 75.8, 75.0, 69.6, 17.3

f 134.8, 126.2, 96.1, 69.4, 68.9, 56.2, 17.9

a 100.6, 100.5, 96.7, 92.8, 78.1, 77.9, 77.1, 75.5, 75.0, 74.2, 73.9, 73.6, 72.7, 72.6, 72.3, 70.9, 70.3, 61.7, 61.6, 61.5

14 b 100.7, 100.6, 96.7, 92.8, 78.4, 78.2, 77.2, 75.5, 75.0, 74.2, 73.5, 73.4, 72.7, 72.6, 72.4, 72.2, 71.1, 70.9, 62.0, 61.9, 34.2 113

when the hydroxyl group on position 6 is replaced by a less bulky group,

or when position 6 is removed as in case of a-fluoroxylose, the binding

of the analog increases. Based on these observations, it may be

expected that if the hydroxyl group at position 6 is replaced by a more

bulky group such as a halogen atom, it will result in lower binding. So

the increase in the 1s for sucrose derivatives with halogen at

position 6 is not surprising.

Since a-fluoroglucose acts as a donor substrate with a similar

to that of sucrose, it may be speculated that the fructosyl moiety may

not play a significant role in the binding to the enzyme. The present

investigation indicates that a bulky substitution on the 6' position

reduces binding even further (compare 6-bromo-sue rose with 6,6'-dibromo-

sucrose). This may be an indication that while the enzyme does not bind

the fructosyl group tightly, there is only limited spatial tolerance

around it.

In conclusion it can be said that none of the four halogenated

derivatives of sucrose tested could irreversibly inactivate the enzyme.

Part of the inactivation observed in the preincubation studies (Figure

15) may be a reflection of the fact that these compounds are weak

reversible inhibitors and were present during the measurement of enzyme activity. However, other factors may also be involved. The inhibition may also explain the fact that several chlorosucrose derivatives were

found to decrease acid production and adherence of S.mutans (142). . 114

B. Acceptor Substrate Specificity: Analogs of a-Methyl Glucoside.

A variety of compounds have been shown to serve as acceptor

substrates for dextransucrase. Most of these are simple glycosides and

a-(D)-glucopyranosyl oligosaccharides (49, 91). Dextran itself has been

shown to serve as an acceptor (49, 143). Some acceptors such as

fructose yield a single product, whereas others such as maltose,

isomaltose and a-methyl glucoside produce a homologous series of oligo

saccharides with an increasing number of newly added glucosyl residues.

The specificity of dextransucrase for acceptors has not been widely

investigated and there is an overall lack of information about this

issue. This question has been addressed in the following studies by

using the halogenated derivatives of a-methyl glucoside as well as other

analogs such as epimers, deoxysugars and several 3-anomers in their

pyranoside as well as furanoside forms.

Earlier studies on the effectiveness of compounds as acceptor

substrates for dextransucrase have utilized radiolabeled acceptors (49,

91) or labelled sucrose (103) and involved the measurement of the

incorporation of labels into products. Utilization of labelled acceptors gives direct evidence of those products that are derived by

transfer of glucose residues. However, such labelled compounds are not always available and the use of radioactive sucrose provides a reasonable alternative. In the present studies radioactive sucrose has been used in reactions catalyzed by the enzyme with the compounds of interest and the formation of oligomeric and polymeric products were measured by paper chromatography. The sum of radioactivity in these two classes of products is a measure of total activity of the enzyme. This 115

analysis also permitted one to determined whether the reactions yielded

a single or multiple products. Typical results are illustrated in

Figure 22 for methyl,a-(D)-glucopyranoside (compound 2 in Table 2),

methyl,6-deoxy-a-(D)-glucopyranoside (compound 8) and methyl ,6-bromo-6-

deoxy-a-(D)-glucopyranoside (compound 14). Some of the products were

not resolved from the residual radioactive sucrose on paper

chromatograins develpoed in solvent system (III). In those cases the

comigrating materials were eluted from the chromatograins with water, the

eluant evaporated to dryness, dissolved in phosphate buffer and treated

with invertase to degrade the residual sucrose. Upon rechromatography

in solvent system (IV) the invertase resistant products were easily

separated from the sucrose components.

The focus of the investigation was to evaluate the effect of

systematic changes in the structure of a simple but reasonably good

acceptor, a-methyl glucoside was selected because it is the simplest a-

(D)-glycopyranoside known as an acceptor and has a Km of 80mM (49).

When it is employed as an acceptor substrate, the Vmax is about a third

of that observed with maltose (49). The analogs examined for acceptor

activity fall into several groups. The first of these are epimers of methyl,a-(D)-glucopyranoside in which the configuration of the hydroxyl

group at each ring position has been inverted. These are compounds 3-6 in Table 2. This series is augmented with the corresponding 6-deoxy- derivatives (compounds 7-12), which includes one a-methyl furanoside

(compound 12) and a pentoside (compound 7). The 6-halo-derivatives of several of these epimers have been examined (compounds 13-17), as well as several hexosides (compounds 18-23). Figure 22. Formation of Acceptor Products. : Some Typical Acceptors.

Reactions were carried out as described in Methods using : (A) methyl,a-(D)-glucopyranoside (2); (B) methyl,6-deoxy-a-(D)-

glucopyranoside (8); and (C) methyl,6-bromo-6-deoxy-a-(D)-

glucopyranoside (14) as acceptors at 350mM concentrations. Reactions were carried out for 4 hours at which time aliquots (20uL) were spotted on Whatman 1MM paper and developed and counted as described in

Methods. The data are presented as the percentage of the total dpm spotted on the chromatogram as a function of migration distance.

116 117

E Q. *o *

33.3%

E o, •o

a o. XJ

Figure 22. 118

The results of these investigations are summarized in Table 2. For

each compound the concentration utilized and the reaction times are

shown, along with the percentage of the total isotope that is non-mobile

and is therefore attributed to the polymerization reaction. The table

also provides a schematic representation of the paper chromatographic

analysis of the low molecular weight products, formed by glucosyl

transfer to the acceptors. These data are summarized with the total of

the percentage of the dpm that appear in the products, the sum of the

isotope in all reaction products and the ratio of isotope in the acceptor products to the polymerization product.

Several interesting points can be drawn from the data presented in

Table 2. All the analogs served as acceptors to some degree. A comparison of the acceptor products observed after four hours of reaction shows that between 2.5 and 80% of the total radioactivity were in these products. The most effective analogs were the halogenated derivatives of methyl,a-(D)-glucopyranoside (compounds 13-15). On the other hand, methyl,ct-(D)-mannopyranosyl derivatives were the poorest acceptors (compounds 3 and 16).

Using the reaction with methyl,a-(D)-glucopyranoside as a reference, the majority of compounds examined appear to inhibit the overall rate of reaction. This can be seen by comparison of the sums of oligomeric and polymeric products formed. It is of interest to note that the only compounds that do not inhibit significantly are analogs of methyl,a-(D)-gl ucopyranoside that are modified at position 6 : methyl ,a-

(D)-xylopyranoside (compound 7), in which carbon 6 is missing , methyl,6-deoxy-a-(D)-glucopyranoside (compound 8), and methyl,6-chloro- Table 2. Comparison of Acceptor Activities of Analogs of g-methyl

glucoside.

Reactions were carried out as described in Methods using the

analogs at the concentrations shown in the table. At the indicated

times aliquots were spotted on Whatman 1MM paper and developed in

solvent system(II) and the chromatograms were analyzed for

radioactivity as described in Methods. The percent of the total dpm on

the chromatogram that was located at the origin was taken as a measure

of the polymer formed and is shown in column 3. The percentage of the

total dpm found in each oligosaccharide peak are schematically

represented in the center column of the table. The Rs value represent the ratio of the distance migrated by the compound to the distance migrated by sucrose. The sum of the percent dpm in the oligosaccharide peaks is taken as the total acceptor product formed, while the sun of the acceptor products and the polymerized products represents the total products. The last column is the ratio of the acceptor products to the polymer formed.

The structures of some of the acceptors are shown in Figure 24.

119 Table 2. ACCEPTOR ACT1WT Iff ALPHA KETHYLS.UCOSIDE AflALOSS

Acceptor Cone. Time Polymer Acc-Prod total Acc/ formed %DPM in Acceptor Products formed Prod Polymer r5 „ .4 .6 .8 1.0 1.2 1.4 1.6 l.E formed Prod nf! Hrs Sdpm 1 , . , Idpra Idpm

I. Hone ... 1 64.1 ... 64.1 4 77.1 — 77.1 —

2. a-O-methyl-(D)- 350 I 19.3 as 1.8 M as 13.0 38.0 as 65.2 84.5 3.33 glucopyranoside 4 19.4 04.L4&S && 14.2 37. * 66.8 86.2 3.41

3. a-O-methyl-(D)- 350 4 11.6 AS. °i9 0,4 2.5 14.1 0.22 mannopyranoslde 10 12.2 cu _ JLfl. . 1-ft 2-S 2.7 14.9 0.22

4. a-O-methyl-(O)- 350 1 23.4 1.B -UL. , A9. 10.4 33.8 0.44 allopyranoslde M 4 62.6 S.I ajj 13.7 1.0 22.8 85.4 0.36

5. a-O-methyl-(O)- 350 4 40.7 0.0 i.fi . .0,?, -2.7 6.2 46.9 0.15 qalactopyranoslde M 10 40.8 O.B 2U1 -2^ 3.0 6.5 47.3 0.16

6. a-O-methyl-(D)- 350 4 61.4 3.0 -2JL ..3,7 . i.3 ;.3' 13.0 74.4 0.21 altropyranoslde H 10 62.3 -JU „2JL 1.3 Jl± 9.9 72.2 0.16

7. a-O-methyl-(D)- 350 4 52.4 O.B JA. _A2_ Zi.O 1.8 29.8 82.2 0.57 xylopyranoslde * 10 53.1 •J|6. -1.B- -9.4 . XXiS-.U 37.2 90.3 0.70 0.9 44.7 8. 6-deoxy-a-0-methyl- 350 1 31.6 -b2- 45.5 76.9 1.44 N (D)-glucopyranostde 4 33.3 Jxfi. $JhB— 55.3 8fl.6 1.66 • 3.0 9. 6-deoxy-a-0-methyl- 300 1 16.1 . 1.3 7.9 12.4 28.5 0.77 H (L)-mannopyranoslde 4 45.0 JUS. 30.8 75.8 0.68

10. 6-deoxy-a-0-methy 1- 350 1 18.4 -lifl. 50.7 51.7 70.1 2.82 (D)-galactopyranoside H 4 20.7 1.9 . 61.P . 63.7 84.4 3.08

U. 6-deoxy-a-0-methyl - 350 1 27.1 • °l? 1.4 7.3 9.2 36.3 0.34 • (l)-galactopyranostde 4 54.9 1.4 5u3 9.9 64.8 0.18 Table 2 continued.

Acceptor Cone. Time Polymer Acc-Prod Total Acc/ formed %DPM in Acceptor Products formed Prod Polymer r „ .4 .6 .8 1.0 1.2 1.4 1.6 l.t 5 formed Prod • i . -| -ip 1— n#4 Hrs Sdpra —. 1 • Sdpra Sdpm M- -U... _®d_ 12. 6-deosy-o-0-methy\- 110 1 27.7 L2_ 7.4 35.1 0.27 a (O)-galctofuranoslde 4 68.2 1.6 PL* «.0 1J «_ 19.6 87.8 0.29

o.n 9?-a 13. 6-chloro-a -0-methy 1- 350 1 21.4 58.8 80.2 2.75 a (D)-glucopyranostde 4 21.5 as.a- 69.1 90.6 3.21

14. 6-bros» -a -0-methy 1 - 350 1 11.6 57.1 68.7 4.92 a ,0 (O)-glucopyranoside 4 13.7 '° 72.1 85.8 5.26

15. 6-1odo-a-0-rosthy1- 350 1 5.7 54.7 60.4 9.60 (D)-glucopyranoslde • 4 9.2 TPnff 79.6 88.8 8.65

16. 6-broiao-a-0-methy 1- 350 1 12.7 1.7 14.4 0.13 (D)-mannopyranos(de a 4 30.2 -&JI— 4.9 35.1 0.16 cy 3i«o 17. 6-brono-a -0-osethy 1 - 200 1 14.3 31.2 45.5 2.18 (O)-galactopyranoslde a 4 29.6 38.8 68.4 1.31

4.0 1S.S 18. B-O-nsethyl-tO)- 350 1 43.7 _LS_ 21.5 65.2 0.40 glucopyranoside a 4 53.4 •• luff. 11.9 65.3 0.20 a.? o.a ».» 19. B-O-iaathyl-(D)- 80 1 28.3 5.8 34.1 0.20 a allofuranoslde 4 58.1 11.5 69.6 0.20 9.4 20 6-deoiy -8 -0-methy 1 - 350 1 46.7 5.4 52.1 0.10 (D)-glucopyranoslde a 4 66.1 7.n 7.6 73.7 0.10

1.* S.Q 21. 6-deoxy-fl-0-methy1- 200 1 25.2 7.0 32.2 0.28 a (l)-mannopyranos(de 4 64.2 .il5_ 14.8 79.0 0.23 _ii2_ .IS.*- 22. 6-deoxy-fl-0-methy1- 350 1 6.6 15.3 21.9 2.32 a , 9 (D)-galactopyranostde 4 9.0 - J.4 27.4 36.4 3.04

23. 6-deoxy-S -0-methy 1 - 350 1 22.0 , 8.3 30.3 0.38 a io.a (D)-galactofuranostde 4 58.9 19.6 78.5 0.33 122

6-deoxy-a-(D)-glucopyranoside (compound 13). The 6-bromo- and 6-iodo-

derivatives of methyl,a-(D)-glucopyranoside (compounds 14 and 15) are

only slightly inhibitory, which can be attributed entirely to the

inhibition of polymerization since their relative abilities to serve as

acceptors are very similar to that of the 6-chloro-derivative. Thus it

would appear that changes at position 6 have little effect on the

ability of a compound to serve as inhibitor. On the other hand,

epimerization at position 2 resulted in compounds that exhibited the

greatest inhibitory activity : methyl,a-(D)-mannopyranoside (compound 3)

and methyl,6-bromo-6-deoxy-a-(D)-mannopyranoside (compound 16). This

inhibition can be attributed to both a decrease in transfer to acceptors

and in polymerization. It must be noted that these are also the poorest

acceptors. It can be concluded that the enzyme is unable to tolerate an

axial hydroxyl group at this position and that binding of a sugar with this configuration results in diminished activity.

Another interesting characteristic of analogs that are modified at position 6 is that they essentially yield a single acceptor product.

All the others produce what appears to be homologous series of products.

The 6-modified derivatives cannot serve as typical acceptors, since the expected mode of transfer would be to position 6 (53).

A consideration of relative acceptor reactivity of methyl ,a-(D)- glucopyranoside and its epimers (compounds 2-6) may provide some insight into the steric requirements of substrates and the mode of binding to the enzyme. It has already been pointed out that epimerization at position 2 is an alteration that results in a poor acceptor and an effective inhibitor of the enzyme. The C-3 epimer, methyl ,a-(D)-al1o- 123

pyranoside (compound 4) is a moderate acceptor and inhibits the total

activity by about 60%. Epimerization at C-4 results in a large decrease

in the ability to serve as an acceptor and inhibits total activity by

about 50%. Methyl ,ot-(D)-altropyranoside (compound 6), the 2,3-di-epiiner

of methyl,a-(D)-glucopyranoside is a better acceptor than the 2-epimer

and is only slightly inhibitory. A possible explanation of this will be

discussed later in this section. The C-l anomer, methyl,B-(D)-gluco

pyranoside (compound 18) is a moderate acceptor and the overall rate of

reaction is approximately 75% of the uninhibited rate. Thus, the loss

of equatorially oriented hydroxyl group at C-2 and C-4 appears to have a

profound effect on the enzyme activity.

The reason for the poor acceptor and strong inhibitor properties of

methyl,ot-(D)-mannopyranoside compared to the corresponding glucose

analog can be one of the following :

(i) The hydroxyl group at position 2 of the glucoside forms a hydrogen

bond with some group on the enzyme thereby resulting in good binding to

the acceptor site. The mannoside fails to participate in this hydrogen

bonding since its hydroxyl group at position 2 is axial. So the

mannoside is not a good acceptor.

(ii) There is steric interference of the hydroxyl group at position 2

of the mannoside with some group of the enzyme. This makes it difficult

for the mannoside to bind at the acceptor site and so it is not a good

acceptor. On the other hand, the glucoside does not experience this

steric interaction and so is a good acceptor.

The above two possibilities can be differentiated by employing methyl,2-deoxy a-(D)-glucopyranoside as acceptor. If it is a poor 124

acceptor, it would mean that the first possibility is correct. If it is

a good acceptor like a-methyl glucoside, it would imply that the second

possibility is correct. The result of the acceptor reaction is shown in

Figure 23. Since invertase treatment was not done, it is not known how

much of the acceptor products comigrate with sucrose. However it is

clear that in four hours of reaction all the .sucrose was consumed. This

is understood from the fact that 9% more counts were incorporated into

dextran between 1 hour and 4 hours of reaction. After one hour of

reaction there was only 12.8% sucrose left. Between one hour and four

hours of reaction the product migrating at 26-31 cm decreased from 52%

to 43%. If this decrease is due to conversion of the disaccharide to

trisaccharide, it has to react with equivalent amount of sucrose and so

the trisaccharide should show at least 18% of the total counts. But it

shows only 12% counts. Also, the sucrose peak after one hour reaction

should have had 18% counts in order to account for the increase in dextran and decrease in the disaccharide between 1 hour and 4 hours of reaction. The reason for this mathematical discrepancy is not understood. This might be explained if there is some type of dispro portionation of the acceptor products. For example, two molecules of the disaccharide may react to give a trisaccharide with the release of a non-labelled acceptor molecule. An example of such a disproportionate has been reported previously (97).

Assuming that all the sucrose was consumed in four hours of reaction, it can be said that 61% of the total dprn appeared in the acceptor products at 350mM concentration of methyl,2-deoxy-a-(D)- glucopyranoside. Thus its acceptor efficiency is comparable to that of Figure 23. Methyl ,2-deoxy-ct-(D)-glucopyranoside as Acceptor.

Dextransucrase (0.05 unit) was incubated with [^C-(glucose)]- sucrose (20mM, 0.12yCi) and the acceptor at 350mM concentration in O.IM phosphate buffer, pH 6.0, final volume 60yL at 37°C. After one hour

(O—O) and four hours (A—a) of reactions 20uL aliquots were spotted on Whatman 1MM paper, developed and analyzed by counting radioactivity as described in Methods.

125 30.1% 39.1%

sucrose 10 acc

, , 5 10 15 20 25 30 35 Figure 23. Distance of migration (cm) 127

a-methyl ylucoside (see Table 2). This supports the idea that an axial

hydroxyl group at position 2 of the glucopyranoside is not tolerated by

the enzyme and that hydrogen bonding of the enzyme with equatorial

hydroxyl group at position 2 may not be important for acceptor activity.

Bulky groups at C-6, such as halogens, have little effect, which

would imply that there is little contact at this position. These

derivatives as well as the 6-deoxy derivative and the xylopyranosyl

derivative are good acceptors. This was an unexpected result since the

position of substitution in methyl,a-(D)-glucopyranoside has been

established to be at C-6 (53). It is possible that sugars blocked at

C-6 have a different orientation when they bind to the enzyme and

therefore an alternate position becomes substituted. The enzyme is

known to produce a-(l->-6) linked dextran with a-(]>3) branches (24,

62). It is interesting to note that when methyl ,a-(D)-glucopyranoside

is rotated by 180° around the axis C1-C4, then C-3 occupies the position

previously occupied by carbon 6 and vice versa. Thus, assuming that both a-(l+6) and a-(l->-3)-l inks are formed at the same active site of the enzyme, this rotation may be a possible mechanism of branch formation.

The mechanism of branch formation will be discussed later. This flipping of the molecule does not drastically alter the orientation of the other hydroxyl groups. This can be seen by superimposing the rings of the flipped and unflipped molecule. In a similar way, the ability of the different compounds in Table 2 to serve as acceptors can be qualitatively explained if it is assumed that whenever the hydroxyl group at position 6 is missing, the acceptor binds to the enzyme in the flipped form. This is schematically shown in Figure 24. 128

For acceptor reactions of compounds 2-6 in Table 2 it is seen that

there is a strong steric hindrance due to hydroxyl groups oriented

axially above the plane of the ring at positions 2 and 4 (compounds 3

and 5). Also there is a weak steric hindrance due to a hydroxyl group

oriented axially below the plane of the ring at position 3 (compound 4)

and due to a hydroxyl group oriented equatorially at position 1

(compound 18). The fact that methyl,6-bromo a-(D)~mannopyranoside is a

poor acceptor also suggests that there is a strong steric interference

due to the hydroxyl group oriented axially below the plane of the ring

at the position occupied by the ring oxygen. Any other orientations do

not appear to affect the acceptor property of a compound. It is also to

be noted that methyl,6-deoxy a-and 0-(D)-galactopyranosides (compounds

10 and 22) are better acceptors than the corresponding glucose analogs

(compounds 8 and 20). It is however, not understood why methyl ,6-bromo-

a-(D)-galactopyranoside (compound 17) is a poorer acceptor than the

corresponding glucose analog (compound 14).

It is remarkable that all compounds examined served as acceptors to

some degree. The fact that sugars with (L)-configuration (compounds

9,11 and 21) methyl,3-(D)-hexopyranosides (compounds 20-22) and hexo-

furanosides (compounds 12,19 and 23) yielded labelled products suggests that the enzyme specificity is rather broad. This may be a reflection of the ability of the enzyme to bind sugars in a somewhat flexible manner within limits. As mentioned earlier, sugars with axially oriented hydroxyl groups at positions 2 and 4 (compounds 3 and 5) are poor acceptors. However, methyl,a-(D)-altropyranoside, the 2,3-di- epimer of methyl,a-(D)-glucopyranoside is a better acceptor than the 2- Figure 24. Proposed Model for Binding of Acceptor to Dextransucrase.

Structures are schematically drawn in such a way as to maintain the positions of C-l and C-4 fixed. Compounds which have C-6 blocked have been rotated by 180° around the C1-C4 axis thereby placing C-3 at the position previously occupied by C-5. Compounds have been drawn in their most favored conformations. The name of each compound is preceded by a number refering to the order in which it appears in Table 2 and followed by the total of all acceptor products formed in four hours of reaction as shown in Table 2. The numbers in parenthesis refer to the amount of polymer formed as shown in Table 2.

( ) weak steric hindrance

( ) strong steric hindrance

129 I

CH OH 2 OH HO HO HO OCH3 OCH3

a-O-methyl-(D)- 66 3 a-O-methyl-(D)- 2 5 glucopyranoside (19"4j mannopyranoside ^^

HC ch2oh ? ch2oh

ch3

a-O-methyl-(D)- 22.8 a-O-methyl-(D)- , 7 allopyranoslde (62.6) qalactopyranoside 7)

CH2OH

HO<* 18. HO 0-O-usthyl-(D)- j, Q glucopyranoside - (bo.4) Figure 24. 131

OCH3 •

s 3 a-O-methyl-(D)- 13>0 alt ropyranoside ^

CH2X HO

-O OCH-

15. 13. OCH3 6-iodo-a-0-methyl- 79.6 6-ch loro-a-0-methy1 - (D)-glucopyranoslde (9.2) (D)-glucopyranoslde ®9.1 8. 14. (21.4) 6-deoxy-a-0-methyl- 6-bromo-a-0-methy1- (D)-glucopyranoslde 55»3 (D)-glucopyranoslde 72.1 (33.3) (13.7)

OCH:

OCH3 a-0-methyl-(D)- 29.8 xylopyranoside (52*4)

17. 6-deoxy-

•O-

6-deoxy-a-0-methyl- ^ fl HO (L)-mannopyranoslde ^45"qj

^ 6-deoxy-a-0-methyl- CH (L)-galactopyranoslde (54.9)

HO 6-deoxy-fl-0-methyl- (L)-aannopyranoslde 14.8 (64.2)

16. 6-bromo-a-0-methyl- ^ g (D)-mannopyronoslde * . ^ JU•C )

Figure 24 continued. 133

HO ^-och3; •/ 20. 6-deoxy-e-O-methyl- (D)-glucopyranoside (66.1)

6-deoxy-e -0-methy1-•methyl ?7 « (D)-galactopyranoslde (go)

Figure 24 continued. 134

epimer. This may be due to the fact that methyl,a-(D)-altropyranoside

exists in solution as a 1:1 mixture of and conformations(154).

In the ^ conformation it has equatorially oriented hydroxyl groups at

C-2 and C-3. The superimposition of models for methyl, a-(D)-gluco-

pyranoside in the conformation and methyl,a-(D)-al tropyranoside in

the ^ conformation is shown in Figure 25. This was accomplished by

alligning the C1-C2 bonds and the C4-C5 bonds in the two structures. It

can be seen that hydroxyl groups are distributed in a very similar

arrangements. The hydroxyl group at position 4 of the altroside is

oriented axially below the plane of the ring and so does not have a

severe effect on the ability to serve as acceptor as also seen in the

case of methyl,6-bromo- and methyl,6-deoxy-a-(D)-galactosides (compounds

17 and 10). Carbon 6 is axially oriented and may have some effect.

Also, the methoxyl group at C-l is equatorially oriented which explains

why this is as good an acceptor as methyl,0-(D)-glucopyranoside

(compound 18).

Similar superimpositions of methyl,a-(D)-glucopyranoside with methyl,6-deoxy-a-(D)-galactopyranoside (compound 10), methyl ,6-deoxy-a-

(L)-galactopyranoside and methyl,6-deoxy-a-(L)-mannopyranoside are shown

in Figure 25. The (D)-galactose derivative has the hydroxyl group at

position 4 oriented axially below the plane of the ring when it binds to

the enzyme and hence is a good acceptor. The (L)-galactose derivative binds to the enzyme with its hydroxyl group at position 4 oriented

axially above the plane of the ring and so is a comparatively poor acceptor. 135

Similar superimposition of models, including those of the furano-

sides, have been done and it has been observed that the ability of a

compound to serve as an acceptor can be qualitatively explained by how

well it can be superimposed on a reference compound such as methyl ,ct-

(D)-glucopyranoside.

As discussed above, when the position 6 of an acceptor is blocked

the molecule binds in a flipped orientation such that addition of new

glucose unit can take place at position 3. It was of interest to see

how the enzyme interacts with the acceptor when both positions 3 and 6 are blocked. Methyl,3,6-anhydro-a-(D)-glucopyranoside was used for that purpose. At a 300mM concentration of acceptor and after one hour of reaction time, 21.2% of the total dpm were found at the origin and 7.6% in the acceptor products while after four hours 56.5% were at the origin and 12.8% in the acceptor products. The results indicate that the 3,6- anhydro- analog is a poor aceptor molecule and support the hypothesis that addition of the new glucose units occur preferentially to positions

6 or 3; however, the poor acceptor activity may also be due to poor binding of the analog.

By using models of acceptor molecules to consider the formation of specific acceptor products, Robyt and Eklund (145) have developed a hypothesis for the mechanism of the acceptor reaction. They postulated that in view of the broad specificity of the acceptor reactions, the acceptor molecules do not bind to a specific acceptor binding site on the enzyme. Rather, the acceptor molecules form hydrogen bonded complexes with glucosyl- or dextranosyl- enzyme complexes. The resulting ternary complexes lead to the specific products that are Figure 25. Structural Comparison of Representative Acceptor Analogs.

Photographs of Dreiding models of several analogs superimposed on methyl ,a-(D)-glucopyranoside are shown in the figure. Each structure

has C-l located in the lower left hand corner of the pyranose ring, and

superimpositions are achieved by alignment of the C1-C2 and the C4-C5 axes of both sugars. In each figure, the 3 and 5 positions of a-methyl glucoside are designated as 3 and 5, while the corresponding positions of the analogs are designated with 3' and 5'. (A) methyl,a-(D)- altropyranoside (®), a fair acceptor, is displayed in the ^ conformation with direct superimposition on the glucoside. (B) the conformer of methyl,6-deoxy-a-(D)-galactopyranoside (10), a good acceptor, is displayed with 180° flip of the ring around C1-C4 axis, thereby placing C-3' in juxtaposition to C-5 of a-methyl glucoside. (C) methyl,6-deoxy-a-(L)-mannopyranoside (9), an intermediate acceptor, is displayed in the ^ conformation with a 180° flip as described in

(B). (D) methyl,6-deoxy-a-(L)-galactopyranoside (11), a poor acceptor, is displayed in the ^ conformation with a 180° ring flip as described in (B).

136 137

/ /\ r

-r^r -0, \ V V A / /\ \

i' /

\ // \ / ' f * Q 4r=T » • - CO CO M \ \ r \

Figure 25. 138

observed. They proposed that monosaccharides fonn three hydrogen bonds,

two with the glucosyl unit and one with a catalytic amino group of the

enzyme. (D)-Fructose is an exception; it forms four hydrogen bonds,

three with the glucosyl unit and one with the catalytic amino group. No

hydrogen bonded complexes could be formed for the two non-acceptors,

a,a-trehalose and (D)-xylopyranose.

The results mentioned in this section do not agree with their

model. They have explained that methyl ,a-(D)-glucopyranoside is an

acceptor because hydrogen bonds are formed between hydroxyl groups at

C-1 and C-4 of the acceptor to those at C-6 and C-2 respectively of the

glucosyl-enzyme complex. However, it has been shown in the present

studies that methyl ,a-(D)-mannopyranoside and methyl,a-(D)-al lopyrano-

side are not good acceptors though they can still form the hydrogen

bonds. Table 2 also shows that methyl,a-(D)-xylopyranoside is a good

acceptor though, according to the proposed model of Robyt and Eklund, no

hydrogen bonded complex can be formed. Thus it seems more logical to assume an acceptor site with broad specificity rather than no acceptor

site at al1.

2. Unsaturated Derivatives.

Some unsaturated sugars were tested as irreversible inactivators of

HOH,C

(III) (I) (II) 139

the enzyme. These included the compounds (D)-glucal (I), 6-deoxy-(L)-

glucal also known as Rhamnal (II), and methyl ,2,3,6-trideoxy-a-(L)-

erythro-hex-2-enopyranoside (III)

The enzyme (0.05 unit) was preincubated with the compounds at 350

mM concentration in a total volume of 20nL for 60 minutes. The control

reaction was not preincubated. lOpL substrate containing 200mM

[•^Cjsucrose (6.7yCi/mL) and 10% dextran T-10 was added to each and

reacted for 5 minutes at 37°C. The reaction was stopped by boiling at

100°C for 2 minutes after which it was spotted on Whatman 1MM paper.

Following development of the chromatogram it was cut into strips (1cm X

2.54cm)and their radioactivities counted. The results are shown in

Table 3. The table shows that (D)-glucal does not significantly

inactivate the enzyme. 6-deoxy-(L)-glucal produces 60% inactivation in

60 minutes of preincubation. Compound (III) did not show any

inactivation; on the contrary, the dpm at the origin increased due to

the preincubation. It also produced some acceptor product. Why an

increase in the time of preincubation produced more acceptor product is

not clear. More detailed experiments with these compounds need to be

done before one can arrive at a conclusion.

3. Photochemical Inactivation of the Enzyme.

It has been observed that modification at position 6 had little effect on the acceptor property of a compound; but it is also known that the enzyme produces dextran with a-(l->-6) bonds. If position 6 is modified to contain a reactive group it may react at the acceptor site Table 3. Effect of Unsaturated Derivatives of g-methyl glucoside on

Dextransucrase

The enzyme (0,05 unit) was preincubated with the indicated compounds at 350mM concentration in a total volume of 20uL for 60 minutes. The control reactions were not preincubated. 10 pL substrate containing 200mM [^C-U]sucrose (6.7pCi/mL) and dextran T-10 were added and reacted for 5 minutes at 37°C. The reaction was stopped by boiling at 100°C for 2 minutes after which it was spotted on Whatman 1MM paper. Following development in solvent system (III) the chromatogram was cut into strips (1cm X 2.54cm) and their radioactivities counted.

140 141

TABLE 3.

Compound Preincubation Polymer Acceptor Total Activity (% dpm) Products {% dpm) (X dpm)

None no 6.4 - 6.4

I no 5.4 - 5.4

I yes 4.9 - 4.9

II no 9.8 - 9.8

II yes 3.9 - 3.9

III no 10.1 0.67 (Rs 1.3) 10.8

III yes 13.0 ' 6.4 (Rs 1.3) 19.4 142

to give irreversible inactivation. Methyl ,6-azido-6-deoxy-a-(D)-gluco-

pyranoside was used for the purpose. Azides are known to decompose in

UV light to give highly reactive nitrenes (147). The nitrene can then

react with the enzyme at the acceptor site to cause inactivation.

In order to determine if the compound binds to the enzyme, it was

first employed as an acceptor. The conditions of the reaction were the same as with other analogs shown in Table 2 except that the reaction was carried out in the dark by wrapping the test tube with aluminum foil.

At a 300mM concentration of the compound and after one hour of reaction time 8.7% of the total dpm were at the origin and 36.2% in the acceptor products (35.4% at Rs 1.9 and 0.8% at Rs 1.7) while after four hours of reaction time 15.3% of the total dpm were at the origin and 68.1% in the acceptor products (64.8% at Rs 1.9 and 3.3% at Rs 1.7). Thus, methyl,6- azido-6-deoxy-a-(D)-glucopyranoside was a good acceptor and similar to the 6-halo- derivatives of methyl,a-(D)-glucopyranoside. Since the compound binds well, it was employed in photochemical reactions. Three sets of experiments were performed.

(i). 0.5 units of the enzyme were mixed with a 25mM solution of the azido compound in a total volume of 200yL in a quartz cuvette and irradiated with UV light as described under methods.

(ii). The above preincubation was done in the dark by covering the tube with aluminum foil and irradiating the covered cuvette with UV

1 ight.

(iii). 0.5 units of the enzyme in a total volume of 200uL in a quartz cuvette were irradiated with UV light as in case of (i).

At different time intervals 20jjL aliquots were removed from each of the above tubes and reacted separately with 40mM [14C]sucrose (0.033yCi) in 143

a total volume of 25mL for 5 minutes in the dark at 37°C after which the mixture was boiled at 100°C for 2 minutes. Reaction mixtures were spotted on Whatman 1MM paper and developed in solvent system (III). The chromatograms were cut into strips (1cm X 2.54cm) and their radio activities counted. The results are shown in Figure 26. The activity of the enzyme preincubated with the azido compound in the dark remained nearly constant for 60 minutes. The activity in the other two sets decreased almost at the same rate. This indicates that the enzyme was being inactivated by the UV light and that the presence of methyl ,6- azido-6-deoxy-a-(D)-glucopyranoside did not enhance the rate of inactivation by any significant amount. It may be possible that the source of UV light liberated radiation of high energy which inactivated the enzyme. SDS polyacrylamide gel electrophoretic analysis of the enzyme after irradiating with UV light for 60 minutes showed it to contain at least seven bands of protein with molecular weights ranging from 20,000 to 178,000. Thus, it appears that the polypeptide was cleaved. This cleavage of protein was not specific for dextransucrase since other enzymes such as hexokinase, glucose-6-phosphate dehydrogenase, phosphoglucose isomerase, dextranase and glucose oxidase were all found to be cleaved by UV light into several fragments as detected by similar analysis.

From the above experiments it is clear that the azido compound does not show significant inactivation of the enzyme at 25mM concentration.

However, proper control of wavelength of irradiation needs to be exer cised in order to observe inactivation only due to the azido compound. Figure 26. Photochemical Inactivation of Dextransucrase with methyl,6-

azido-6-deoxy-g-(D)-glucopyranoside.

Dextransucrase (0.05 units) with (• •) or without (A—A) 25mM

concentration of the azido compound in a total volume of 200 jjL in a

quartz cuvette was irradiated with UV light as described in Methods. In

a control reaction the enzyme in the presence of the azido compound was

taken in a quartz cuvette which was covered with aluminum foil and

irradiated with UV light (O—O). At indicated time intervals 20pL

aliquots were removed from each and reacted separately with 40mM

[^C-U]sucrose (0.033pCi) in a total volume of 25uL for 5 minutes in the

dark at 37°C after which the reaction mixture was boiled at 100°C for 2 minutes. The reaction mixtures were spotted on Whatman 1MM paper, developed in solvent system (III) after which the chromatograms were cut

into strips (1cm X 2.54cm) and their radioactivities were counted.

144 20 30 40 50 60 figure 26. Time (min) en 146

C. Acceptor Substrate Reaction : Oligosaccharides.

Next to dextran, the best known acceptor substrate of dextran-

sucrase is maltose (49, 103). Several other oligosaccharides, namely

isomaltose, nigerose, lactose, cellobiose, melibiose, turanose and

raffinose, have been tested as acceptor of dextransucrase. Among those

mentioned above the only good acceptors are maltose, isomaltose and

nigerose. From this small list it is clear that there is an overall

lack of information on the acceptor specificity with respect to

oligosaccharides. Since dextran is a much better acceptor than a-methyl

glucoside, it was of interest to examine compounds of intermediate

molecular weight as acceptors. Since dextran contains mostly a-(l->-6)

linkage, the obvious choice of oligosaccharides for the purpose are the

isomaltose oligosaccharides which also contain a-(1+6) linkages.

However, since these sugars are not available commercially, they have to

be prepared. Experimental conditions were designed for their

purification in optimum yield.

As described under Methods, Dextran T-10 was treated with dextranase for a short time and the products were separated on a Bio Gel

P2 column. The elution profile of one such separation is shown in

Figure 27. It shows a good resolution of the products. However, when the experiment was carried out on a larger scale the separation was not as good (data not shown). There was a large peak at fractions 85-90 which represent the void (not shown in the figure). The first peak after the void may be a mixture of oligosaccharides. The last two peaks represent isomaltotriose and isomaltose. Major amount of the sugar is present in these three peaks. Since isomaltose and isomaltotriose are Figure 27. Chromatographic Separation of Isomaltose Oligosaccharides.

Dextran T-10 (200mg) was reacted with 20 units of dextranase in

0.1M phosphate buffer pH 6.0 (final volume 2mL) at 37°C for 15 minutes

following which it was heated at 100°C for 10 minutes. The

oligosaccharides were separated on a Bio-Gel P2 (minus 400 mesh) column,

2cm X 120cm using water as the eluant at a flow rate of 6mL/hr and ImL

fractions were collected. Sugars were detected by the Anthrone method as described in Methods. The figure shows the amount of sugar in each

fraction.

147 Fraction number

Figure 27. 1'49

available commercially, the aim of this separation was to obtain the

others in good yield. Experiments are underway in this laboratory to

obtain the oligosaccharides from DP4 to DP10 in better yields so that

acceptor reactions can be carried out.

As mentioned earlier, next to dextran, maltose is the best acceptor

known for the dextransucrase reaction. However, no information is

available about the acceptor activity of the higher homologs of the

maltooligosaccharide series. In this study these sugars, which included

maltose to maltoheptaose were employed as acceptors. Since all the

acceptor products migrate slower than sucrose or fructose, no invertase

treatment of the sucrose peak was necessary as in the case of mono

saccharide acceptor products. Also [^C-U]sucrose was used rather than

one labelled in glucose only because the fructose peak does not overlap

with the acceptor product peaks.

The results are shown in Table 4. The table shows the polymerized

product as the percentage of total isotope that is non-mobile, the sum

of the percentage of dpm that appear in the acceptor products and the

sum of the isotope in all reaction products except fructose and glucose.

It also shows the paper chromatographic analysis of the oligosaccharide

acceptor products. It is important to realize that if [^C-gl ucose]-

sucrose was used instead of [^C-U]sucrose, all the percentages would

have been twice what is shown in the table-.- From Table 3 it is clear that among the maltooligosaccharides tested only maltose is a good acceptor. The acceptor activity decreases from maltose to maltopentaose and then increases slightly to maltoheptaose. All these compounds inhibited autopolymerization to a great extent. In case of other acceptors that have been reported previously it was found that good 150

acceptors inhibited the rate of autopolymerization whereas poor

acceptors did not. Based on this, it has been proposed (103) that

addition of a low molecular weight acceptor diverts the (D)-glucosyl

group of sucrose away from making dextran and the amount of dextran and

its molecular weight are decreased. However, the results in Table 4 and

Table 2 show that even though some compounds are not good acceptors and

not donor substrate analogs, they still can inhibit the

autopolymerization reaction. The last column represents the total

activity of the enzyme. Maltose is actually an activator of the

enzyme. Inhibition of the enzyme by the rest of the series increases

from maltotriose to maltotetraose after which there is a slight decrease

to maltoheptaose. Though not directly related to this system, it may be

appropriate to mention here that studies on the inhibition of malto-

oligosaccharides of the cyclization reaction catalyzed by the cyclo-

dextrin glycosyltransferase from Klebsiella pneumoniae gave similar

results (147). The author observed that the degree of inhibition

increased from maltose to maltotetraose and then decreased with the

1 arger saccharides. In the previous section it has been reported that

some of the monosaccharide acceptor analogs also inhibited total activity of the enzyme. Most significant of these were methyl,a-(D)- mannopyranoside and methyl,6-bromo-6-deoxy-a-(D)-mannopyranoside shown in Table 2. It is not clearly understood how these compounds inhibit the enzyme or what the relation is between the acceptor substrate binding site and the donor substrate binding site. The mode of inhibition by the maltooligosaccharides is presently being studied in this laboratory. Table 4. Comparison of Acceptor Activities of Hal tool igosaccharides

Reaction mixtures contained 50mM acceptor, 50mM [^C-U]sucrose

(0.067pCi) and 0.04 units of dextransucrase (activated with triton X-100 as described under Methods), in a final volume of 40 pL. Reactions were conducted at 37°C for 60 minutes following which lOtiL aliquots were spotted on Whatman 1MM paper and developed in solvent system (III) in a repetitive manner. The chromatogram with DP2 was developed twice (each time 16hrs or until the solvent front moved 50cm). The product with DP3 was developed three times while, the chromatograms with DP4, DP5 and DP6 were developed four times and DP7 was developed five times. Standard solutions of the acceptors were spotted separately and developed in the same way. After development the chromatogram was cut into strips (1 X

2.54cm) and counted for radioactivity.

151 TABLE 4

Compound Polymer % DPM in Acceptor Products Acceptor Total Formed Product Product (%) Formed Formed Rs-t- 0.2 0.4 0.6 0.8 1.0 (%) (%)

Maitose 1.3 1.4 5.5 8.8 7.4 2.2 25.3 26.6

Maitotriose 2.1 .37 .53 1.8 2.6 1.7 7.0 9.1

Maitotetraose 2.4 .17 .19 .51 .57 1.1 2.5 4.9

Maitopentaose 3.3 .8 .45 .53 1.8 5.1

Maitohexaose 3.2 1.0 1.0 .69 2.7 5.9

Maitoheptaose 3.5 2.5 .73 3.2 6.7

VI r\> 153

Other disaccharides were also tested as acceptors. These included

isomaltose, nigerose, 6'-bromo-6'-deoxy-maltose and methyl,6'-bromo-6'-

deoxy-a-(D)-nigeroside (obtained by employing methyl ,6-bromo-6-deoxy-a/

(D)-glucopyranoside as acceptor as will be discussed in the next

section). Maltose was also tested for comparison. Reaction mixtures

contained acceptor (50mM or higher as specified), [^C(glucose)]-sucrose

(67mM, 0.04pCi) and 0.05 units of enzyme in a total volume of 30pL.

Reactions were carried out at 37°C for 60 minutes after which the mixtures were spotted on Whatman 1MM paper, developed in solvent system

(III). The chromatograms were dried,cut into strips (1cm X 2.54cm) and

their radioactivities counted. The acceptor product produced with 6'-

bromo-maltose comigrated with sucrose in solvent system (III). Thus,

the reaction mixture was treated with invertase and chromatographic development was done in solvent system (IV) which separated the product

from fructose and glucose. The results are shown in Figure 28-32.

Figure 32 suggests that there may be some product comigrating with sucrose. The sucrose peak was eluted with water, treated with 4 units of invertase in a total volume of 35pL, spotted on paper and rechromatographed in solvent system (IV). However, no invertase resistant product was found in this peak (data not shown). The results are also summarized in Table 5 which shows the percentage of the total dpm at the origin and in the acceptor products followed by the sum of the two which represents the total activity. Since these experiments were performed on different days, the actual numbers may vary slightly depending on the degree of reactivation of the enzyme achieved by Triton

X-100. Figure 28. Maltose as Acceptor.

Reaction mixture contained 0.05 unit enzyme, 67mM [^C(glucose)]- sucrose (0.04wCi) and acceptor (50mM) in a total volume of 30nL.

Reaction was carried out at 37°C for 60 minutes after which all of it was spotted on Whatman 1MM paper, developed in solvent system (III) and analyzed by counting the radioactivity as described in Methods.

154 14 7 m

101

0 5 10 15 20 25 30 35 Distance of migration fcm) Figure 28. •—» cn en Figure 29. Isomaltose as Acceptor.

Reaction conditions were the same as those described for Figure 28, however, isomaltose (50mM) was employed as the acceptor.

156 14t—

12 sucros©

10 15 20 25 30 35 Distance of migration (cm) Figure 29. Figure 30. Nigerose as Acceptor.

Reaction conditions were the same as those described for Figure 28, however, 50mM (O—O) and 200mM (A—A) nigerose were employed as the acceptor.

Reaction conditions were the same as those described for Figure 28, however, e'-bromo-e'-deoxy-maltose was employed as the acceptor. After reaction the mixture was treated with invertase (4 units) and development was done in solvent system (IV).

160 22.3%

Figure 31. Figure 32. Methyl ,6-bromo-6-deoxy-ct-(D)-nigeroside as Acceptor.

Reaction conditions were the same as those described for Figure 28, however, methyl ,6-bromo-6-deoxy-a-(D)-nigeroside was employed as the acceptor.

162 36.8%

10 acc

E a -o

10 15 20 25 30 35 Figure 32. Distance of migration (cm) 40 164

Table 5 shows that 6'-bromo maltose is a poor acceptor compared to

maltose. This may be due to poor binding of the bromo derivative. A

similar difference was also observed between a-methyl glucoside and

6-bromo-a-methyl glucoside at low concentrations. With maltose as

acceptor, the new glucosyl residue is added at the C61 position (148).

Since that position is blocked in 6'-bromo-maltose it was of interest to

determine the new position to which glucose is added. This will be

discussed in the next section.

Nigerose is a reasonably good acceptor for dextransucrase. It is

different from other acceptors, such as maltose and isomaltose, in that

it produces oligosaccharides of higher degrees of polymerization. These

did not migrate far nor were they well resolved on paper chromatograms.

There is uncertainty as to how much of the radioactivity at the origin

was due to acceptor products. It is clear that the acceptor products

are also good acceptors which explains the fact that high molecular

weight products were formed easily. 6-bromo-a-methyl nigeroside is a

good acceptor compared to nigerose. Comparing the relative acceptor

efficiencies of 61-bromo-maltose and maltose with those of 6-bromo-ct-

methyl-nigeroside and nigerose, it appears that bromine at position 6'

affects the acceptor activity more than bromine at position 6. However,

the glycosidic a-methyl group may also have its influence.

Table 5 shows that isomaltose is a better acceptor than maltose.

This is understandable because dextran, the natural acceptor for the enzyme has a-(l->-6) linkages and isomaltose is also a-(l->-6) linked whereas maltose is a-(l->-4) linked. This result is different from what Table 5. Comparison of Acceptor Activities of Pi saccharides.

Reaction mixtures contained 0.05 units of enzyme, 67mM

[^C(glucose)]-sucrose (0.04pCi) and 50mM acceptor in a total volume of

30uL. Reactions were carried out at 37°C for 60 minutes after which all of the reaction mixtures were spotted on Whatman 1MM paper. Following development in solvent system (III) the chromatograms were dried and cut into strips (1cm X 2.54cm) and their radioactivity counted.

165 166

TABLE 5

Compound Polymer Acceptor Total Formed Products Activity {% dpm) (% dpm) (% dpm)

Maltose 2.8 71.8 74.6

Isomaltose 7.5 89.1 96.6

Nigerose 25.6 41.3 66.9

6'-bromo-maltose 22.3 15.0 37.3

Methyl,6-bromo- a-(D)-nigeroside 36.8 29.2 66.0 167

was reported by Mayer et al (49) who observed that maltose was a better

acceptor than isomaltose. However, the differences in experimental

conditions may account for the differences in the results. In the

previous study lOOmM acceptor was reacted with lOOmM sucrose and 0.2

units of enzyme for 30 minutes, while in this case 50mM acceptor was

reacted with 67mM sucrose and 0.05 units of enzyme for 60 minutes.

Furthermore, Mayer et .al. (49) used labelled acceptors whereas in this

case labelled sucrose was used. If labelled acceptor is used then each

product molecule will contain radioactivity due to one molecule of

acceptor whereas, if labelled sucrose is used, the radioactivity will be

due to several glucose units. For example, a tetrasaccharide formed

from maltose will contain radioactivity derived from one acceptor unit and two glucose units.

D. Acceptor Substrate Reaction : Branch Formation.

Arnett and Mayer (24) have shown that dextran synthesized by dextransucrase contains principally a-(l->-6) linkages with some a-(l+3) links. Experiments were designed to determine the type of linkages that are formed when new glucosyl residues are added by the enzyme to small acceptor molecu!es.

1. Preparative Synthesis of Products with Methyl,a-(D)-glucopyranoside as Acceptor.

[14C] labelled methyl,a-(D0-glucopyranoside was employed as an acceptor substrate for synthesizing the oligosaccharide acceptor- products in sufficient amounts for analysis by 13C NMR spectroscopy.

The paper chromatographic separation of the products are shown in 168

Figure 33. The results are similar to those reported by Mayer et.al.

(49). The small peak between the disaccharide acceptor product and the

methyl ,a-(D)-glucoside is due to glucose. The disaccharide peak

overlaps with sucrose as seen by the AgNOg-NaOH test (107). The peak

was eluted with water which was the(i evaporated to dryness. The residue

was dissolved in O.IM phosphate buffer, pH 4.5 and after the addition of

invertase (20 units), reacted at 55°C for 4 hours. The products were

separated on a Bio Gel P2 column (2cm X 50cm, minus 400 mesh) using

water as the eluant at a flow rate of 5.5mL/hour. The fractions

containing the disaccharide were pooled, evaporated to dryness and

analyzed by NMR spectroscopy (Figure 34a). The chemical shift of an

unsubstituted C6 is 61 ppm. The spectrum of an a-(l->-6) link should have

a peak at 67 ppm and of an a-(l+3) link at about 80.5 ppm. The spectrum

shows a peak at 67 ppm which suggests that it is a-(l->-6) linked sugar.

The absence of any peak at 80-81 ppm indicates that there is no a-(l+3)

1inkage.

Since the higher homologs, DP>5 were not well separated and not in

sufficient quantities, a mixture of the oligosaccharides with Rs 0.05 to 1^ 0.33 were el uted from the paper and analyzed by C NMR spectroscopy

(Figure 34b). The spectrum shows that the oligosaccharides are composed of only a-(l+6) linkages and no a-(l-»-3) linkages. The chemical shift of

C-l1 can also give an indication of the type of linkage : a-(1+2) at

97.5ppm, a-(l+3) at 99.8ppm, ot-(l->-4) at lOl.Oppm and a-(l+6) at

99.4ppm. The NMR spectra of the disaccharide and the mixture of higher homologs are shown in Figure 35. The small doublet at 5.3 ppm in the spectrum of the disaccharide suggests that there is about 2% a-(l->-3) Figure 33. Preparative Synthesis of Acceptor Products with Methyl,

g-(D)-glucopyranoside as Acceptor.

The reaction mixture contained 17 units of enzyme, sucrose (200mM) and [^C-U]methyl ,a-(D)-glucopyranoside (6.4uCi, 140mM) in a total volume of 9mL. It was allowed to react at 37°C overnight and then spotted on ten 21cm lengths of Whatman 3MM paper and developed in solvent system (III). A portion (2.54cm) of the paper was cut into strips (1cm X 2.54cm) and radioactivity of each strip was counted.

169 6- sucrose acc

aE •o 3-

10 15 20 25 30 35 Distance of migration (cm) Figure 33. o Figure 34. NMR Spectra of Acceptor Products with Methyl,g-(D)-

G1ucopyranoside as Acceptor.

A. Disaccharide Product.

B. Mixture of oligosaccharide products with DP>5.

171

173

Figure 35. NMR Spectra of Acceptor Products with Methyl

q-(D)-glucopyranoside as Acceptor.

A. Disaccharide Product.

B. Mixture of oligosaccharide products with DP>5.

173 Figure 35. 175 1 ? linkage in the compound. This is different from the C NMR spectrum

(Figure 34a and b) which gave no indication of an a-(l->-3) linkage. This

doublet is not seen in the mixture of the higher homologs. Thus, it

seems that all the products formed have essentially only a-(l->-6)

linkages with a small amount of a-(l*3) linkage at the limits of detectibility. These results agree with those of Walker (149) who analyzed the linkages by enzymatic methods and reported that products

from isomaltooligosaccharide acceptors had very little o-(l>3) branches which may not be detected unless radioisotopes of high specific activity are used.

2. Preparative Synthesis of Products with Maltose as Acceptor.

The reaction mixture contained 7 units of enzyme, maltose (175mM) and sucrose (lOOmM) in a total volume of 5.6mL. In another reaction the concentrations of reactants were the same except that [*^C-U]sucrose

(0.13yCi) was used in a total volume of 40pL. Both mixtures were allowed to react at 37°C overnight. They were spotted separately on

Whatman 3MM paper and developed in solvent system (III). The radio active chromatogram was cut into strips (1cm X 2.54cm) and each strip was counted. The result is shown in Figure 36. This shows more of the trisaccharide product than that reported earlier by Mayer et.al (49) in which they employed a lower concentration of maltose (lOOmM) and reacted the mixture for only 30 minutes. Due to the large scale of the reaction, the higher homologs (DP>6) are not well resolved. The three peaks were eluted with water which was then evaporated. The products 1 o 1 1 o were then analyzed by C and AH NMR spectroscopy. The C NMR spectra Figure 36. Preparative Synthesis of Acceptor Products with Maltose as

Acceptor.

0.05 units of the enzyme were reacted overnight with a mixture of maltose (175mM) and [^C-U]sucrose (lOOmM, 0.13pCi) in a total volume of

40pL at 37°C. The reaction mixture was spotted on Whatman 3MM paper and developed in solvent system (III). The chromatogram was cut into strips

(lcm X 2.54cm) and radioactivity in each strip was counted.

176 12 maltose sucrose 10 fructose

,u 15 20 25 Jo- Distance of migration (cm) 35 40 Figure 36.

V4 178

of the three products along with that of maltose are shown in Figure

37. Chemical shift assignments for maltose were based on the published

spectrum (150). The spectra show that all three products had only a-

(1+6) linkages (peaks at about 99 ppm and 67 ppm) and no a-(l-*-3)

linkage (absence of peak at 80-81 ppm). The NMR spectra of the

compounds are shown in Figure 38. They show that all the three

compounds have only a-(l+6) linkages (peaks at 4.9 ppm) and no a-(l-»-3)

linkage (absence of peak at 5.3 ppm). The ratio of integrals of

anomeric proton peaks to the ring proton peaks confirm that the three

products are tri-, tetra- and pentasaccharides, respectively.

3. Praparative Synthesis of Products with 61-bromo-61-deoxy-maltose as

Acceptor.

The reaction mixture contained dextransucrase (5 units), 6'-bromo- maltose (200mM) in a total volume of 1.5mL at 37°C. Reaction was initiated by addition of 0.5mL of 200mM sucrose. 0.5mL of the sucrose solution was added every hour for three hours. The reaction was then continued for another four hours. The reaction mixture was treated with

60 units of invertase in a total volume of 2.6mL at 55°C for 30 minutes.

In another experiment the above reaction was repeated with 0.05 units of enzyme and proportional amounts of the other reactants except that the substrate was [^C-U]sucrose (6.7pCi/mL).

The labelled reaction mixture was spotted on a 2cm strip of Whatman

3MM paper and the unlabelled reaction mixture on four 20cm strips of

Whatman 3MM paper and developed in solvent system (IV) for 36 hours. The strips were dried and developed again for another 36 hours. The Figure 37. NMR Spectra of Acceptor Products with Maltose

as Acceptor.

A. Maltose.

B. Trisaccharide product.

C. Tetrasaccharide product.

D. Pentasaccharide product.

179 JL- 1 1

.l l—.—1^1

JUV- J L u

.ill _i i i i—i—i—i—*—i—i— 100 80

Figure 37. Figure 38. NMR Spectra of Acceptor Products with Maltose

as Acceptor.

A. Trisaccharide product.

B. Tetrasaccharide product.

C. Pentasaccharide product.

181 vl

Ul 183

radioactive chromatogram was counted as described in Methods. The

result is shown in Figure 39. It shows that the reaction with invertase

was not complete. Sugars migrating from 2.5cm to 4.5cm from the origin 1 ^ were eluted with water, the pH was adjusted to 7.0 and the C NMR 1 O spectrum taken (Figure 40b). Based on the C NMR spectra of similar

compounds (150, 151, 152) it was concluded that the new glucose unit was

added to carbon 2 of the glucose at the reducing end.

4. Methyl,6-bromo-6-deoxy-a-(D)-glucopyranoside as acceptor.

In the previous section it was mentioned that methyl,6-bromo-6- deoxy-a-(D)-glucopyranoside is a good acceptor substrate of dextran-

sucrase. At 350mM concentration of the acceptor and 20mM [^C]sucrose

(labelled in glucose only), about 72% of the radioactivity was found in the single acceptor product after four .hours of reaction. The product had high mobility on paper chromatogram (Rs 1.7). This acceptor product is different from those seen to date in our or other laboratories, because of its high Rs value and because it is the only major product produced. In addition, this was the first compound with position 6 blocked to be tested as acceptor. With typical acceptors the expected mode of transfer is at position 6 (53). Since large amounts of the product were observed, even though position 6 was blocked, it was of interest to characterize the product.

Strips from the paper chromatogram containing the acceptor product

(Figure 22) were washed with water. The washings (about 50mL) was concentrated to 3mL by evaporation. Concentrated HC1 (2mL) was added to it and heated at 60°C for about 24 hours. The HC1 was removed by Figure 39. Preparative Synthesis of Acceptor Product with 6'-bromo-

Maltose as Acceptor.

The reaction mixture contained dextransucrase (0.05 units), 6'-

bromo-maltose (200mM) in a total volume of 15uL at 37°C. Reaction was

initiated by addition of 5uL of 200mM [^C-U]sue rose (6.7uCi/mL). 5yL

of the sucrose solution was added every hour for three hours. The

reaction was then continued for another four hours. The reaction mixture was treated with 0.6 units of invertase in a total volume of

2.6mL at 55°C for 30 minutes. The reaction mixture was spotted on a 2cm

strip of Whatman 3MM paper and developed in solvent system (IV) for 36 hours. The strip was dried and developed again for another 36 hours.

The chromatogram was cut into strips (1cm X 2.54cm) and radioactivtiy in each strip counted.

184 12

•o-oo Q 10 15 20 25 30 35 40 Distance of migration (cm)

Figure 39. 00 1 3 Figure 40. C NMR Spectrum of Acceptor Product with

6'-bromo-maltose as Acceptor.

A. 6'-bromo-6'-deoxy-ma1tose.

B. Trisaccharide product.

186 Figure 40. 188

evaporation followed by storage over KOH overnight. The residue was

dissolved in a small volume of water, spotted on Whatman 1MM paper and

developed in solvent system (III). Standard sugars were also spotted

and detected by the AgN03-Na0H method (107) after development. The

radioactivity comigrated with the glucose standard (data not shown),

indicating that the product contained a glucose moiety. Treatment of

the product with a-glucosidase also generated a radioactive compound

that comigrated with glucose (data not shown) which suggests that the

product had an a-1inked glucose unit. These acceptor reactions were

carried out with SDS inactivated enzyme (42) which had to be reactivated

with Triton X-100. In order to rule out any participation by the

detergents, the experiment was repeated with enzyme which did not

contain detergent (105). The same product was formed, indicating that

the methyl,6-bromo-6-deoxy-a-(D)-glucopyranoside was the acceptor and

not SDS or Triton X-100. In another experiment, the product was

purified on a Bio Gel P2 column (2cm X 50cm, minus 400 mesh). The

radioactivity eluted at the usual position of a disaccharide as determined by calibrating the column with standard sugars. Thus, the new product was a disaccharide with an a-1 inked glucose unit.

Experiments were conducted to determine the effect of changes in the reaction time and concentrations of acceptor on the yields of high molecular weight and oligosaccharide acceptor products. At 350 mM acceptor and 20mM sucrose, the amount of dextran synthesized plateaued after one hour of reaction whereas the amount of acceptor product continued to increase (Figure 41). Thus, the ratio of disaccharide acceptor product and dextran remained nearly constant for one hour and then increased. Figures 41. Amounts of Polymer and Acceptor Products Formed as

a Function Time of Reaction.

The reaction mixture contained 0.15 units of enzyme, 300mM methyl ,6-bromo-6-deoxy-a-(D)-glucopyranoside and 20mM [^C-gl ucose]-

sucrose (6.6 X 10^ dpm) in a total volume of 150yL. The reaction was

initiated by addition of the enzyme. At indicated times, 10|iL aliquots

were spotted on Whatman 1MM paper. Following development in solvent system (III) the chramatograms were cut into strips (1cm X 2.54cm) and the radioactivity in each strip was counted. (O—O) % of total dpm in the polymer, (A—A) % of total dpm in acceptor products and (.•—o) ratio of acceptor product to polymer.

189 dpm (%) to

to o Acceptor Product Pdlymer 191

Figure 42 shows the changes in the amounts of products produced in

one hour as the acceptor concentration was varied. It shows that

increasing concentrations of acceptor inhibit the synthesis of dextran.

In this experiment the concentration of sucrose was kept constant at

20mM. In another series of reactions, the concentrations of the

acceptor as well as sucrose were increased to maintain their ratio

constant (Figure 43). It can be seen that with increasing amounts of

both substrates, the amount of dextran increased first, reached a

maximum and then decreased. The amount of acceptor product increased in

a sigmoidal manner.

4. Preparative synthesis of products with methyl,6-bromo-6-deoxy-a-(D)- glucopyranoside as acceptor.

This experiment was designed to produce sufficient amounts of the product for analysis by NMR spectroscopy. It has been mentioned earlier that at a 350mM concentration of the acceptor and 20mM [^Cjsucrose

(labelled only in glucose), about 72% of the radioactivity appeared in the single acceptor product after four hours of reaction. However, this represents only 4.1% of the total acceptor. In order to utilize more of the acceptor, fresh sucrose was added at regular time intervals to maintain the acceptor/donor ratio high because, the higher this ratio, the higher will be the ratio of the acceptor products to the polymer formed (148). The results in Figure 44 show that significant amounts of trisaccharide product was also produced. 25% of the counts were in dextran, 11% in the disaccharide product and 6% in the trisaccharide Figure 42. Amounts of Polymer and Acceptor Products Formed as a

Function of the Acceptor Concentration.

The reaction mixture contained 0.05 units of enzyme, 6.6mM [^C- glucose]-sucrose (44,000 dpm) and methyl, 6-bromo-6-deoxy-a-(D)- glucopyranoside at the indicated concentrations in a total volume of

30uL. After one hour of reaction, 20uL aliquot was spotted on Whatman

1MM paper. Following development in solvent system (III), the chromatograms were cut into strips (1cm X 2.54cm) and each strip was counted for radioactivity. (O—O) % of total dpm at the origin,

(A—A) % of total dpm in acceptor products and (•—• ) ratio of acceptor products to polymer.

192 AJ mole glucose transferred

09. •>> O o (0 XJF»S» o

to Acceptor Product Polymer vo co Figure 43. Amounts of Polymer arid Acceptor Products Formed Under

Varying Concentrations of Donor and Acceptor Substrates.

All conditions were the same as in Figure 24 except that concentrations of both the acceptor and sucrose were varied keeping their ratio constant at 7 : 2. (O—O) % of total dpm in the polymer,

(A—A) % of total dpm in the acceptor products and (•—• ) ratio of polymer to acceptor products.

194 mole glucose transferred N> U1

to "O

K) OJ in Acceptor Product Polymer »—» •>o ui 196

product. It is important to realize that if the reaction had been

carried out with [-^C-glucose]-sucrose, the percentages of the total dpm

in each of the products would have been doubled. These results indicate

that about 31umole of the disaccharide and 16ymole of the trisaccharide

were formed. The experiment was repeated to obtain more of the two

products. The peak for the disaccharide on the paper chromatogram

overlaps with t.he acceptor, methyl ,6-bromo-6-deoxy-a-(D)-

glucopyranoside, as seen by the AgNOg-NaOH test (107). The two were

separated on a Bio Gel P2 column (2cm X 50cm, minus 400 mesh) using

demineralized water as the eluant. 1 ? The products were characterized by their C NMR spectra which are

shown in Figure 45. The assignments of chemical shifts were based on

the published spectra of similar compounds (150, 151, 152). The peak at

61 ppm in the spectrum of the disaccharide is for unsubstituted carbon

6. The spectrum of an a-(l->-6) link should have a peak at 67 ppm. The

absence of any peak in this region indicates that the product is devoid

of o-(l->6) linkages. The peak for C-3 has been shifted by 80.4 ppm from its usual position at 75 ppm suggesting that it is a a-(l>3) linked disaccharide. The nmr spectrum of the disaccharide closely resembles that of a-nigerose with the peak for C-6 shifted by -28 ppm and that of C-l by 7 ppm.

Based on similar analysis it can be said that the trisaccharide has two a-(1+3) linkages and no a-(l+6) link. The absence of a a-(l->-6) linkage and the presence of a-(l+3) links can also be seen in the NMR spectra of the compounds (Figure 46). The anomeric proton for a-(l-»-6) link appears at 4.9 ppm whereas linkages at secondary positions of the Figures 44. Preparative Synthesis of Acceptor Products with Methyl,

6-bromo-6-deoxy-a-(D)-glucopyranoside as Acceptor.

400uL enzyme (1.3 units) were mixed with 200pL acceptor

(700mM). The reaction was initiated by addition of 25uL [^C-U]sucrose

(200mM, 0.167yCi) and maintained at 37°C. For the first three hours

25)jL aliquots of the labelled sucrose were added to it at 10 minute intervals. Sucrose additions were continued every 10 minutes for five hours; however, 10pL aliquots were added for three hours and 5yL aliquots were used for the next two hours. The reaction was continued overnight. It was then spotted on two 21cm lengths of Whatman 3MM paper and developed in solvent system (III). A portion (2.54cm) of it was cut into 2.54cm X 1cm strips and the radioactivity in each strip was counted.

197 22.3*t

12 sucrose fructose acc

10 15 20 25 30 35 Distance of migration (cm) 40 Figure 44. 1 o Figure 45. C NHR Spectra of Acceptor Products with Methyl,6-bromo-

6-deoxy-a-(D)-g1 ucopyranoside as Acceptor.

A. Disaccharide product.

B. Trisaccharide product.

199

Figure 46. *H NMR Spectra of Acceptor Products with 6-bromo-

6-deoxy-ct-(D)-gTucopyranoside as Acceptor.

A. Disaccharide product.

B. Trisaccharide product.

201 I I I i I I i i I I I I I I I I I I I I I 1 1 1 1 1 1 L 5 4 3 ro o Figure 46. ro 203

ring show a chemical shift of 5.3 ppm (153). Thus, the disaccharide was

characterized as methyl.glucopyranosyl a-(l-)-3)-6-bromo-6-deoxy-a-(D)-

glucopyranoside and the trisaccharide as methyl ,glucopyranosyl a-(l->-3)-

glucopyranosyl a-(l->-3)-6-bromo-6-deoxy-a-(D)-glucopyranoside.

The enzyme dextransucrase is known to produce dextran containing

primarily a-(l->-6) linkages and some branches with a-(l+3) linkages.

Thus compounds which have carbon 6 blocked were thought to be unable to

serve as typical acceptors. However, since it has been shown here that

methyl ,6-bromo-6-deoxy-a-(D)-glucopyranoside is an acceptor and the

disaccharide thus synthesized has a-(l-v3) linkages, these results

provide the first example for the formation of 100% a-(l*3) linkage

instead of a-(l->-6) linkage.

In 1980, Walker (149) analyzed the products obtained when

isomaltose oligosaccharides were employed as acceptors for S. mutans

0MZ176 (D)-glucosyltransferase and concluded that isomaltose oligo

saccharides must contain 7 or 8 glucose residues to elicit branching

activity. Mc Cabe and Smith (143) have observed that a S. mutans Kl-R

(D)-glucosyltransferase system modified primer dextrans so that they became both more efficient as acceptors and more highly branched. From this study they inferred that one of the transferases recognized internal (D)-glucose residues and catalyzed the formation of branch points. In this context,it may be mentioned that isomaltooctaose (IMg) was reported to be the smallest effective primer for glucan synthesis by

S. mutans 6715 (D)-gl ucosyl transferase (154). In contrast to all these reports, the present investigation shows that 100% a-(l+3) linkages in high yield can be synthesized by the enzyme even with a monosaccharide acceptor, provided carbon 6 is blocked. 204

In the previous section it was proposed that if the 6 position of

an acceptor is blocked, the molecule binds to the acceptor site in an

alternate orientation that is obtained by rotating the molecule by 180°

around the C1-C4 axis. In this orientation the C-3 occupies the same

position occupied by C-6 of the unflipped molecule. So the new glucosyl

unit is added to the C-3 of the acceptor. The pure a-(l->-3) structure of

the disacharide, obtained with methyl,6-bromo-6-deoxy-a-(D)-gluco-

pyranoside as the acceptor, supports this proposal.

Flipping of the molecule can take place either before or after it

binds to the acceptor site. Such rotation at the active site has been

shown for UDP-galactose-4-epimerase (155, 156). As discussed in the

previous section, the acceptor specificity of dextransucrase is very

broad. This may make such rotation at the acceptor site sterically

feasible.

The molecule may enter the acceptor site in the flipped orien

tation. It has been observed that with a-methyl glucoside as acceptor

the products show very little or no a-(l-»-3) branches. This suggests

that acceptor molecules usually enter the acceptor site in the unflipped

orientation. Only when carbon 6 is blocked, as in the case of methyl,6-

bromo-6-deoxy-a-(D)-glucopyranoside,is there productive binding in the

flipped orientation. However, a high concentration of the acceptor is

needed to observe any significant acceptor product (Figure 42).

Binding of acceptor molecules to the enzyme in the flipped

orientation may be a possible mechanism for branch formation. The

flipped molecule may bind to the enzyme in the same way as the unflipped molecule since the hydroxy! groups are similarly oriented. However, the 205

concept of flipping of molecules becomes more complicated with oligo-

and poly-saccharides. If the change in orientation takes place before

the acceptor enters the acceptor site, then the reaction has to follow

multi-chain mechanism, i.e., polymers of intermediate chain lengths

should dissociate from the enzyme and then bind in a flipped

orientation. For a single chain mechanism the change in orientation has

to take place at the acceptor site since the polymer remains bound to

the enzyme throughout the process. In acceptor dependent reactions a

homologous series of acceptor products are observed. Therefore, it is

believed that the acceptor-dependent reaction follows a multi-chain mechanism (49, 61). However, this does not necessarily mean that the

intermediates should dissociate from the enzyme after each addition of a

glucosyl unit. Polymerization may occur as a mixture of the single

chain mechanism with frequent random release of products of intermediate chain lengths. Thus, for an acceptor-dependent reaction it is possible

for the change in orientation of the acceptor molecule to take place either before or after it enters the acceptor site.

The trisaccharide acceptor product synthesized by the enzyme with

6'-bromo-maltose as acceptor has the new glucose unit added to position

2 of the glucose at the reducing end. Rotation of the 6'-bromo-maltose molecule by 180° around the axis C4'-C1 places the C-3' at the position previously occupied by C-5', as in the case of monosaccharides.

However, these two orientations of the molecule do not superimpose well with each other and so no linkage at C-3' is observed. If 6'-bromo- maltose is rotated by 180° around an axis passing through the anomeric oxygen of the non-reducing end and perpendicular to the plane C3-C5-C51- 206

C31 then the C2 occupies the position previously occupied by C-5 as

shown below.

Superimposition of the two orientations is not as good as in the case of

monosaccharides, discussed in section B of this chapter, but it is the

best superimposition possible of the different orientations of 6'-bromo-

maltose. This is probably the orientation in which it binds to the

enzyme for linkages at C-2 to be formed. However, this mode of binding

is not preferred for maltose since no linkages at secondary positions

are observed in the products when maltose is the acceptor.

In the case of oligosaccharides it may be expected that instead of

flipping the complete molecule, only the non-reducing end may flip to

position the C-3 of the non-reducing end to the position previously

occupied by C-5 of that unit. In other words, there has to be a

conformational change by rotating the non-reducing unit by 180° around

C4-0 bond. Such conformers do not superimpose well with the original

conformers. Also, such rotations of oligosaccharides at the acceptor

site may be more difficult than in the case of monosaccharides due to larger space requirements for the rotation. This is understood from the

fact that the oligosaccharides, produced with either maltose or a-methyl glucoside as acceptor, contain only ct-(l-»-6) bonds. There was some a- 207

(1+3) linkage (about 2%) detected by 1H NMR analysis of the disaccharide

product with a-methyl-glucoside as the acceptor. This may be due to the

fact that flipped and unflipped orientations of monosaccharides can be

made to superimpose with each other better than the corresponding

di saccharides.

From the preceeding discussions it can be seen that acceptor-

dependent reactions can, under special conditions, produce up tp 100% a-

(1+3) bonds even though with natural acceptors much less a-(l+3) bonding

occurs.

It is surprising to see that in the reaction with methyl,6-bromo-6-

deoxy-a-(D)-glucopyranoside as the acceptor, the resulting trisaccharide

has only a-(l+3) linkages. The non-reducing end of the disaccharide,

formed by the addition of the first glucose, has an unsubstituted C-6.

If the trisaccharide is formed by utilizing the disaccharide as an acceptor, it may be expected that the transfer would take place at carbon 6 of the non-reducing end. However, the results show that the transfer takes place at carbon 3 of the non-reducing end. This raises the question of whether a-(l+3) linked acceptors yield products with only a-(1+3) linkages. To address this question, nigerose, which is a-

(1+3) linked disaccharide of glucose, was employed as the acceptor and the resulting products were characterized.

6. Preparative Synthesis of Products with Nigerose as Acceptor.

250pL of nigerose (600mM) were added to 750pL of enzyme (3.8 units). The reaction was initiated by addition of 100nL of sucrose

(200mM). Then lmL of sucrose (200mM) was added in 100pL batches at 15 208 minute intervals. The reaction was continued overnight at 37°C and then

spotted on two 21cm long Whatman 3MM papers.

In another experiment, 10pL of nigerose (600mM) were added to 30uL

of enzyme (0.15 units). The reaction was initiated by addition of 4yL

[^C-U]sucrose (200mM, 0.027yCi). Then 40pL of the labelled sucrose

were added in 10 batches at 15 minute intervals. The reaction was continued overnight at 37°C and was spotted on Whatman 3MM paper. All the chromatograms were developed in solvent system (III) for 24 hours.

The chromatogram from the second experiment was cut into strips (1cm X

2.54cm) and radioactivity of each strip was counted. The results are shown in Figure 47. The large peak at 32-37cm was due to fructose.

Significant amounts of low molecular weight products were not observed. The high molecular weight products were not well resolved.

The mixture of sugars migrating at 1 to 4cm was eluted with water and 1 ? analyzed by C NMR (Figure 48). The spectrum shows that glucose units were added to the non-reducing ends to form a-(l*6) linkages. The following peak assignments are based on published spectra of similar compounds (150). C-l of the second glucose unit which is linked to the first as a-(l->-3) (100 ppm), a- and 3- forms of C-l of the first glucose unit (93 and 97 ppm), C-l of the rest of the glucose units with a-(l+6) linkages (99 ppm), C-6 of the first glucose unit (62 ppm), C-6 of the rest of the glucose units, all of which are a-(l+6) linked (67 ppm) and

C-3 of the first glucose unit in a-and g-forms (81 and 83 ppm). The absence of any other peak in the region 81-83 ppm suggests that this is the only a-(l+3) linkage present in the compound. Figure 47. Preparative Synthesis of Acceptor Products with Nigerose as

Acceptor.

1OuL nigerose (600mM) were added to 30yL of enzyme (0.15 units).

The reaction was initiated by addition of 4jjL [^C-lOsucrose (200mM,

0.027 uCi). Then 40pL of the labelled sucrose were added in 10 batches at 15 minute intervals. The reaction was continued overnight at 37°C and was spotted on Whatman 3MM paper. The chromatogram was developed in solvent system (III) for 24 hours and cut into strips (1cm X 2.54cm) and radioactivity of each strip was counted.

209 nigerose

sucrose

10 15 20 25 30 35 40 Distance of migration (cm) ro Figure 47. i—» o Figure 48. NMR Spectrum of Mixture of Oligosaccharide Acceptor

Products with Nigerose as Acceptor.

This contains the mixture of sugars migrating at 1 to 4cm from the origin on paper chromatogram (see figure 47).

211

Thus, it is clear that nigerose, which is an a-(l-v3) linked

disaccharide of glucose, adds new glucose units at position 6 of the

non-reducing end. This suggests that with methyl,6-bromo-6-deoxy-a-(D)-

glucopyranoside as acceptor, the trisaccharide product is synthesized

not by employing the intermediate disaccharide as an acceptor but by a

single chain mechanism without the release of the intermediate

disaccharide. This would imply that once a branch point is formed, the

next glucose units will be added in a single chain fashion to form a-

(1+3) links. Any of the intermediates that are released and later serve

as acceptors add the new glucose unit at position 6. When methyl,6-

bromo-6-deoxy-a-(D)-gl ucopyranoside was employed as acceptor very little

of the trisaccharide was produced in a one hour reaction (see Figure

22). This means that the formation of consecutive a-(l-v3) links by the

single chain mechanism is not a favored process and the product is

released from the enzyme after the formation of a branch point. As an

acceptor, the disaccharide cannot compete with the overwhelmingly large

amount of the monosaccharide which also is a better acceptor. After one

hour of reaction time the ratio of the monosaccharide to the disaccharide was 30:1. Hence very little trisaccharide was observed in

Figure 22. Even in the preparative synthesis, after complete reaction, the ratio of the monosaccharide to the disaccharide was 7:2. Thus, the trisaccharide had no detectable a-(l->-6) linakges. These results indicate that dextransucrase is capable of producing consecutive a-(]>3) linkages though no evidence of such linkages in natural dextran has as yet been reported. 214

Robyt and Taniguchi (69) have proposed that branching in dextran

takes place when a C-3 hydroxyl group on an acceptor dextran acts as a

nucleophile on C-l of the reducing end of a dextranosyl-enzyme complex,

thereby displacing dextran from the enzyme and forming an a-(l->-3) branch linkage. This model does not explain how branching can occur in the

absence of acceptor. This mechanism of branch formation by rotation at the acceptor site can be considered for the autopolymerization reaction in which chain initiation and elongation takes place at the reducing end and does not require an acceptor. It is believed that de novo synthesis occurs by a single chain mechanism (62). Thus, for branches to form during the autopolymerization reaction, the change in the mode of binding has to take place at the reaction site. It is not possible for the complete polymer to flip since that would require a large space.

Only the terminal glucose unit at the reducing end needs to rotate. „The terminal glucose unit at the reducing end can rotate by 180° around its

C5-C6 bond which is the only rotation possible for the rest of the polymer to remain fixed. This brings the C3 at the position previously occupied by CI as shown in Figure 48 (step 1). According to the model for growth at the reducing end, the active site of the enzyme has two equivalent nucleophilic groups. At any given time during catalysis one of them is bonded to a glucose and the other to the dextran being synthesized (Figure 4). For chain propagation, the nucleophilic C-6 hydroxyl group of the glucosyl enzyme complex attacks at C-l of the dextranosyl-enzyme. For branch formation the C-3 hydroxyl group acts as a nucleophile to displace the new glucose unit from the glucosyl-enzyme complex (step 2). It has been proposed (56) that the polymer is Figure 49. Possible Mechanism of Branch Formation by Termination of

Autopolymeri zation.

215 Figure 49. ro •—» 217 covalently bound to the enzyme. Thus, for rotation of the glucose unit at the reducing end this covalent bond has to be broken, possibly by hydrolysis. Luzio and Mayer (89) have provided evidence for hydrolysis of glucosyl-enzyme complex. Following hydrolysis, the glucose unit at the reducing end is no longer activated resulting in termination of the reaction. This model is similar to that proposed by Robyt and Taniguchi in that branch formation results in termination. Branch formation and termination of autopolymerization is shown schematically in Figure 49.

It involves the following steps :

(1). The dextranosyl-enzyme bond hydrolyses

(2). The reducing terminus glucose unit (B) rotates by 180° around the

C5-C6 bond (step 1).

(3). The new linkage formed is between C3 of the reducing terminus and

CI of the new glucose unit (C) added (step 2).

(4). The polymer is released from the enzyme (step 3).

(5). This polymer can rebind to the enzyme and continue chain

propagation by adding new glucosyl units (D) at the non-reducing

end of ring C as in acceptor-dependent reactions (step 4). 218

TABLE 6

List of chemical shifts of acceptor products (from TMS)

Fig. no. Spectrum Chemical Shifts

34 a 100.3, 98.8, 74.3, 74.0, 72.8, 72.4, 72.1, 71.0, 70.5, 70.4, 66.5, 61.4, 56.1

34 100.3 98.7, 74.4, 74.1, 72.8, 72.5, 72.4, 72.2, 71.2 71.0, 70.6, 66.7, 61.5, 61.5, 56.2

37 100.6 100.5 96.7 92.8 78.1, 77.9, 77.1, 75.5, 75.0 74.2 73.9 73.6 72.7, 72.6, 72.3, 70.9, 70.3 61.7 61.6 61.5

37 100.8 100.7 99.1 96.7 92.9, 78.6, 78.3, 77.1, 75.6 75.0 74.1 72.8 72.7, 72.6, 72.4, 72.3, 71.0 70.6 70.4 67.0 61.8, 61.7, 61.5

37 100.8 100.7 99.0 98.7 96.7, 92.8, 78.6, 78.3, 77.1 75.5 74.9 74.3 74.1, 72.8, 72.7, 72.6, 72.4 72.3 72.2 71.2 71.0, 70.5, 70.5, 67.0, 66.6 61.8 61.7 61.5

37 100.8 100.7 98.9 98.7 98.6, 96.7, 92.8, 78.3, 77.1 75.5 74.9 74.3 74.1, 72.8, 72.7, 72.6, 72.4 72.3 72.2 71.2 71.1, 70.9, 70.5, 67.0, 66.6 61.8 61.6 61.4

40 100.7 100.6 96.7 92.8 78.4, 78.2, 77.2, 75.5, 75.0 74.2 73.5 73.4 72.7, 72.6, 72.4, 72.2, 71.1 70.9 62.0 61.9 34.2

40 100.6 100.4 98.7 97.3 97.0, 90.2, 79.7, 77.8, 77.7 76.8 76.0 75.4 74.0, 73.8, 73.6, 73.5, 73.0 72.9 72.8 72.7 72.6, 72.4, 72.2, 71.2, 70.8 70.5 62.1 61.9 61.7, 61.5, 34.2

45 100.5 100.0 80.4 73.9 72.9, 72.8, 72.7, 71.0, 70.8 70.5 61.4 56.2 34.1

45 100.5 100.2 100.1, 100 0, 81.1, 80.4, 73.9, 72.9 72.7 72.6, 71 3, 71.0, 70.8, 70.6, 70.5 61.6 61.4, 61 2, 56.2, 34.1

4« 98.7, 74.4, 74.2, 72.8, 72.4, 71.2, 71.1, 70.6, 66.6, 61.6 CONCLUSION

It was observed that halo-derivatives of sucrose were poor inhibitors of dextransucrase. Concentrations required for 50% inactivation ranged between 180mM for the 6-bromo-derivative and 482mM for the 6,61-dichioro-derivative. A halogen at position 6' decreased binding suggesting that there is limited spatial tolerance around the fructosyl moiety of sucrose though the fructosyl group is not required for binding as has been reported earlier (77).

The acceptor specificity of the enzyme has been examined with a series of a-(D)-glucopyranosyl analogs and determined to be rather broad. The results also indicated that the enzyme cannot tolerate axial hydroxyl groups at position 2 and 4 of the glucopyranose ring. Further more, it seems that the intolerance at position 2 was not due to lack of appropriate hydrogen bonding but due to steric interaction with the enzyme. Derivatives of methyl,a-(D)-glucopyranoside modified at position 6 were good acceptors and produced one major product in each case. It has been proposed that this product is formed by binding of the acceptor in a flipped orientation whereby C-3 occupies the position occupied by C-6 of the unflipped molecule. Acceptor efficiencies of compounds have been correlated with their ability to superimpose on a reference compound such as methyl ,a-(D)-glucopyranoside.

219 220

The specificity of the enzyme for oligosaccharide acceptors was

studied with the maltooligosaccharides. Maltose was found to be the

best acceptor. The higher molecular weight oligosaccharides were not

good acceptors. However, all of them were strong inhibitors of the

autopolymerization reaction.

A mechanism for formation of branches in dextran was proposed,

based on characterization of the products synthesized when the following

compounds were employed as acceptors : methyl ,a-(D)-glucopyranoside; methyl ,6-bromo-6-deoxy-a-(D)-gl ucopyranoside; nigerose; maltose and 6'- bromo-61-deoxy-maltose. The results showed that during acceptor- dependent reactions the enzyme synthesizes the usual a-(l->-6) linkages but that it can synthesize other types of linkages if the normal position of reaction is blocked. According the proposed mechanism the acceptor molecule undergoes a rotation due to which the C3 of the non- reducing end occupies the position previously occupied by C-6. Thus, the new glucose is added to make a-(l-»-3) branches. After formation of the branch point, if the product does not dissociate from the enzyme, consecutive a-(l->-3) linkages are formed. If, however, it dissociates and rebinds to serve as acceptor, a-(l+6) linkages are formed. The mechanism has also been considered for branch formation by rotation of the glucose at the reducing end after hydrolytic termination of autopolymeri zation. BIBLIOGRAPHY

1. Hehre, E.J. and Sugg, J., Exptl. Med. 75 (1942) 339

2. Ebert, K.H. and Schenk, G., Z. Naturforsch. 17B (1962) 732

3. Hehre, E.J. and Suzuki, H., Arch. Biochem. Biophys. 113 (1966) 675

4. Hammond, B.F., Arch. Oral Biol. 14 (1969) 879

5. Carlsson, J., Newbrun, E. and Krasse, B., Arch. Oral Biol.

(1964) 469

6. Chludzinksi, A.M., Germaine, G.R. and Schachtele, C.F.

J. Bact. 118 (1974) 1

7. Fukui, K., Fukui, Y. and Moriyama, T., J. Bact. 118 (1974) 796

8. Gibbons, R.J. and Banghart, S.B., Arch. Oral Biol. 11 (1966) 533

9. Gibbons, R.J. and Fitzgerald, R.J., J. Bact. 98 (1969) 341

10. Berman, K.S. and Gibbons, R.J., Arch. Oral Biol. 11 (1966) 533

11. Houte, J. von, Wimkler, K.C. and Jansen, H.M., Arch. Oral Biol.

14 (1969) 45

12. Critchely, P., Wood, J.M., Saxton, C.A. and Leach, S.A.,

Caries Res. 1 (1964) 112

13. Si 1verstone,L.M., Brit. Dent. J. 120 (1966) 461

14. Guggenheim, B., Caries Res. 2 (1968) 147

15. Krasse, B. and Carlsson, J., Arch. Oral Biol. 15 (1970) 25

221 222

16. Jordan, H.V., Englander H.R. and Lim, S.» J. Amer. Dent. Assoc.

78 (1969) 1331

17. Littleton, N.W., Kakehashi, S. and Fitzgerald, R.J.,

Arch. Oral Biol. 15 (1970) 461

18. De Stoppelar, J.D., Van Houte, J. and Backer, O.D., Caries Res.

3 (1969) 190

19. Robyt, J.F. and Martin, P.J., Carbohydr. Res. 113 (1983) 301

20. Evans, T.H. and Hibbert, H., Adv. Carbohydr. Chem. 2 (1946) 203

21. Guggenheim, B., Helv. Odont. Acta 14 (1970) 89

22. Newnan, B.M., White, P., Mohan, S.B. and Cole, J.A.,

J. Gen. Microbiol. 118 (1980) 203

23. Jeanes, A., Haynes, W.C., Wilhan, C.A., Rankin, J.C.,

Melvin, E.H., Austin, M.J., Cluskey, J.E., Fisher, B.E.,

Tsuchiya, H.M. and Rist, C.E., J. Amer. Chem. Soc. 76 (1954) 5041

24. Arnett, A.T. and Mayer, R.M., Carbohydr. Res. 42 (1975) 339

25. Tung, K.K., Kovach, K.J., Pierre, K.J. and Davis, R.S.,

Fed. Proc. 33 (1974) 1452

26. Kobayashi, M. and Matsuda, K., Biochem. Biophys. Acta 397 (1975) 69

27. Chen, A., M.S. Thesis, The Ohio State University (1976)

28. Hakamori, S., J. Biochem. (Tokyo) 55 (1964) 205

29. Lee, H.C., M.S. Thesis, The Ohio State University (1976)

30. Suzuki, H. and Hehre, E.J., Arch. Biochem. Biophys. 104 (1964) 305

31. Hare, M.D., Svensson, S. and Walker G,J., Carbohydr. Res.

66 (1978) 245 223

32. Ciardi, J.E., Proceedings "G1ucosyltransferases, Glucans,

Sucrose and Dental Caries" Ed. Doyle, R.J., and Ciardi, J.E.

Sp. Supp. Chemical Senses (1983) pp 51

33. Montville, T.J., Cooney, C.L. and Sinskey, A.J.,

Adv. Appl. Microbiol. 24 (1978) 55

34. Kobayashi, M. and Matsuda, K., Biochem. Biophys. Acta

370 (1974) 441

35. Kobayashi, M. and Matsuda K., Biochem. Biophys. Acta

614 (1980) 46

36. Guggenheim B. and Newbrun E., He!v. Odontol. Acta 13 (1969) 84

37. Spinnel, D.M. and Gibbons, R.J., Infect. Immun. 10 (1974) 1448

38. Mc. Cabe, M.M. and Smith, E.E., Infect. Immun. 7 (1973) 829

39. Smith, D.J., Taubman, M.A. and Ebersole J.L., Infect. Immun.

23 (1979) 446

40. Scales, W.R., Long, L.W. and Edwards J. R., Carbohydr. Res.

42 (1975) 325

41. Ciardi, J.E., Hageage, G.J. Jr. and Wittenberger, C.L.

J. Dent. Res. 55 (1976) C-87

42. Grahame, D.A., and Mayer, R.M., Carbohydr. Res. In Press. (1985)

43. Mukasha, H., Shimamura A. Tsumori, H., Biochem. Biophys. Acta

719 (1982) 81

44. Figures, W.R. and Edwards, J.R., Biochem. Biophys. Acta

577 (1979) 142

45. Mc. Cabe, M.M. and Smith, E.E., Infect. Immun. 16 (1977) 760

46. Luzio, G.A., Grahame D.A., and Mayer, R.M., Arch. Biochem. Biophys.

216 (1982) 751 224

47. Huang, S., Ph.D. Dissertation, The Ohio State University (1976)

48. Neely, W.B., Adv. Carb. Chem. 15 (1960) 366

49. Mayer, R.M., Matthews, M.,Futterman, C., Parnaik, V.K. and

Jung, S.M., Arch. Biochem. Biophys. 208 (1981) 278

50. Racusen, D., Anal. Biochem. 99 (1979) 474

51. Hehre, E.J., J. Biol. Chem. 222 (1956) 739

52. Bailey, R.W., Barker, S.A., Bourne, E.J., Stacey, M. and

Theander, 0., Nature 179 (1957) 310

53. Walker, G.J., Carbohydr. Res. 30 (1973) 1

54. Parnaik, V.K., Ph.D. Dissertation, The Ohio State University (1979)

55. Ditson, S.L., and Mayer, R.M., Carbohydr. Res. 126 (1984) 170

56. Robyt, J.F., Kimble, B.K. and Walseth, T.F.,

Arch. Biochem. Biophys., 165 (1974) 634

57. Ebert, K.H. and Schenk, G., Adv. Enzymol. 30 (1968) 185

58. Parnaik, V.K., Luzio G.A., Grahame, D.A., Ditson, S.L., and

Mayer, R.M., Carbohydr. Res. 121 (1983) 257

59. Luzio, G.A., Parnaik, V.K., and Mayer, R.M., Carbohydr. Res.

121 (1983) 269

60. Iwaoka, K.R. and Mooser, G., Proceedings "G1 ucosyltransferase,

Glucans, Sucrose and Dental Caries." Ed. Doyle, R.J. and

Ciardi, J.E., Sp. Supp. Chemical Senses (1983) pp 113

61. Neely, W.B., Adv. Carb. Chem., 15 (1960) 341

62. Sidebotham R.L., Adv. Carb. Chem. 30 (1974) 371

63. Bovey, F.A., J. Polymer Sci. 35 (1959) 191

64. Jung, S.M., Ph.D. Dissertation, The Ohio State University (1976) 225

65. Ceska, M., Granath, K.A., Norrman, B. and Guggenheim, B.,

Acta. Chem. Scand. 26 (1972) 2223

66. Hill, M., M.S. Thesis, The Ohio State University (1978)

67. Ebert, K., Schenk, G., Rupprecht, G. and Brosche, M.,

Mackromol. Chem. 96 (1966) 206

68. Ebert, K.H. and Brosche, M., Biopolymers 5 (1967) 423

69. Robyt, J.F. and Taniguchi, H., Arch. Biochem. Biophys.

174 (1976) 129

70. Behre, E.J., J. Polymer Sci. Part C, 23 (1968) 239

71. Hehre, E.J., Bull. Soc. Chim. Biol. 42 (1960) 1713

72. Hassid, W.Z. and Neufeld, E.," The Enzymes", Ed. Boyer, P.D.

6 (1962) 278

73. Okada, G. and Hehre, E.J., Carbohydr. Res. 26 (1973) 240

74. Gold, A.M. and Osber, M.P., B.iochem. Biophys. Res. Commun.

42 (1971) 469

75. Figures, W.R. and Edwards, J.R., Carbohydr. Res. 48 (1976) 245

76. Jung, S.M. and Mayer, R.M., Arch. Biochem. Biophys. 208 (1981) 288

77. Grier, T.J. and Mayer, R.M., Arch. Biochem. Biophys. 212 (1981) 651

78. Wolfenden, R., Nature, 223 (1969) 704

79. Secernski, I.I., Lehrer, S.S. and Lienhard, G.E., J. Biol. Chem.

247 (1972) 4740

80. Tu, J.I., Jacobsen, G.R. and Graves, D.J., Biochem. 10 (1971) 1229

81. Levy, G.A. and Snaith, S.M., Adv. Enzymol. 8 (1966) 500

82. Voet, J.G. and Abeles, R.H., J.Biol. Chem. 245 (1970) 1020

83. Silverstein, R., Voet, J.G., Reed, D. and Abeles, R.H.,

J.Biol. Chem. 242 (1967) 1338 226

84. Dedonder, R., Meth. Enzymol. 8 (1966) 500

85. Bourne, E.J., Peters, J. and Weigel, H., J. Chem. Soc. (1964) 4605

86. Goodman, A., Weil, R.M. amd Stern, K.G., J. Biol. Chem.

217 (1955) 977

87. Chassy, B., Beall, J., Bielanski, R., Porter, E. and

Donkersloot, J., Infect, and Immun. 14 (1976) 408

88. Maynard, M. and Kuramitsu, H., Infect, and Immun. 23 (1979) 873

89. Luzio, G.A. and Mayer, R.M., Carbohydr. Res. Ill (1983) 311

90. Koepsell, H.J., Tsuchiya, H.M., Hellman, N.N., Kazenko, A.,

Hoffman, C-A., Sharpe, E.S. and Jackson, R.W., J. Biol. Chem.

200 (1953) 793

91. Robyt, J.F. and Walseth, T.F., Carbohydr. Res. 61 (1978) 433

92. Suzuki, H. and Hehre, E.J., Arch. Biochem. Biophys. 105 (1964) 339

93. Mazza, J.C., Akgerman, A. and Edwards,^ J.R., Carbohydr. Res.

40 (1975) 402

94. Genghoff, D.S. and Hehre, E.J., Proc. Soc. Exp. Biol. Med.

140 (1972) 1298

95. Feingold, D.S., Avigad, G. and Hestrin, S., J. Biol. Chem.

224 (1957) 295

96. Ebert, K.H., Ruprecht, G. and Schenk, G., Z. Naturforschg.

18B (1963) 442

97. Binder, T.P., Cote, G.L. and Robyt, J.F., Carbohydr. Res.

124 (1983) 287

98. Binder, T.P. and Robyt, J.F., Carbohydr. Res. 124 (1983) 287

99. Mayer, R.M. Unpublished results

100. Walker, G.J., J. Dent. Res. 51 (1972) 409 227

101. Jones, R.W., Jeanes, A. Stringer, C.S. and Tsuchiya, H.M.,

J. Amer. Chem. Soc. 78 (1956) 2499

102. Iriki, Y. and Hehre E.J., Arch. Biochem. Biophys. 134 (1969) 130

103. Robyt, J.F. and Eklund, S.H., Carbohydr. Res. 121 (1983) 279

104. Stodala, F.H., Sharpe, E.S. and Koepsell, H.J.,

J. Amer. Chem. Soc. 78 (1956) 2514

105. Huang, S., Lee, H.C. and Mayer, R.M., Carbohydr. Res.

74 (1979) 287

106. Carlson, J., Newbrun, E. and Krasse, B., Arch. Oral Biol.

14 (1964) 469

107. Trevelyan, W.E., Proctor, D.P. and Harrison, J.S., Nature (London)

166 (1950) 444

108. Seifter, S., Dayton, S., Novic, B. and Muntwyler, E.,

Arch. Biochem. 25 (1950) 191

109. Tong, H.K., Lee, K.H. and Wong, H.A., Anal. Biochem.51 (1973) 390

110. Fairbanks, G., Steck, T.L. and Wallach, D.F.H., Biochem.

19 (1971) 2606

111. Russell, R.R.B., FEMS Microbiol. Lett. 11 (1981) 279

112. Gordon, A.J. and Ford, R.A., "The Chemists' Companion : A Handbook

Practical Data, Techniques and References" Wiley Interscience,

New York, 1972 p 434

113. Anisuzzaman, A.K.M. and Whistler R.L., Carbohydr. Res.

61 (1978) 511

114. Ramirez, F. and Mc Kelvie, N., J. Amer. Chem. Soc. 79 (1957) 5829

115. Ramirez, F., Desai, N.B., Hansen, B. and Mc. Kelvie, N.,

J. Amer. Chem Soc. 83 (1961) 3539 228

116. Ramirez, F., Desai, N.B. and Mc. Kelvie, N., J. Amer. Chem. Soc.

84 (1962) 1745

117. Johnson, R.A. and Herr, M.E., J. Org. Chem. 37 (1972) 310

118. Hanessian, S., Carbohydr. Res. 2 (1966) 86

119. Hanessian, S. and Plessas, N.R., J. Org. Chem., 34 (1969) 1035

120. Fischer, E., Ber. 28 (1895) 1151

121. Paterson, T.S. and Robertson, J., J. Chem. Soc. (1929) 300

122. Feiser L. and Feiser M., "Reagents for Organic Synthesis"

Volume 1, pll

123. Angyal, S.J. and Davies, K.P., Chem. Commun. (1971) 500

124. Angyal, S.J. and Pickles, V.A., Aust. J. Chem. 25 (1972) 1695

125. Wolfrom, M.L. and Thompson, A., Meth. Carb. Chem. 2 (1963) 211

126. Zemplen, G. and Pacsu, E., Ber. 62 (1929) 1613

127. Kochetkov, N.K. and Chizhov, O.S., Adv. Carb. Chem. 21 (1966) 39

128. Bock, K. and Pedersen, C., Adv. Carb. Chem. and Biochem.

41 (1983) 45

129. Evans, M.E. and Angyal, S.J., Carbohydr. Res. 25 (1972) 43

130. Seo, S., Tomita, Y., Tori, K. and Yoshimura, Y.,

J. Amer. Chem. Soc. 100 (1978) 3331

131. Pigman, W and Horton, D., "The Carbohydrate, Chemistry and

Biochemistry", 2nd. edition, Vol 1 p 1400

132. Fischer, E. and Zach, K., Ber. 45 (1912) 456

133. Haworth, W.N., Owen, L.N. and Smith, F., J. Chem. Soc. (1941) 88

134. Takeo, K. and Shinmitsu, K., Carbohydr. Res. 133 (1984) 135

135. Takagi, S., Shiota, M., Mitsuishi, Y., Kobayashi, M. and

Matsuda, K., Carbohydr. Res. 129 (1984) 167 229

136. Inoue, M. and Smith, E.E., Carbohydr. Res. 80 (1980) 163

137. Koga, T. and Inoue, M., Carbohydr. Res. 93 (1981) 125

138. Ono, K., Nuessle, D.W., and Smith, E.E., Carbohydr. Res.

88 (1981) 119

139. Singh, S., Maynard, C.M., Lin, F-T., Taylor, K.G. and Doyle, R.J.,

"G1ucosyl transferases, Glucans, Sucrose and Dental Caries",

Ed. Doyle, R.J. and Ciardi, J.E. 1983, p 155

140. Thaniyavarn, S., Singh, S., Taylor, K.G., and Doyle, R.J.

same as #149, p 161

141. Robyt, J.F., and Zikopoulos, J.N., U.S. Patent 4,335,100

142. Drucker, D.B., Arch. Oral. Biol. 28 (1983) 833

143. Mc. Cabe, M.M. and Smith, E.E., Carbohydr. Res. 65 (1978) 223

144. Stoddart, J.F. "Stereochemistry of Carbohydrates",

Wiley Interscience, N.Y. 1971, p 89

145. Robyt, J.F. and Eklund, S.H., Bioorg. Chem. 11 (9182) 115

146. Calvert, J.G. and Pitts, J.N.Jr., "Photochemistry", Wiley and Sons

Inc. 1967 p 474

147. Bender H., Carbohydr. Res. 135 (1985) 291

148. Killey, M., Dimler, R.J. and Cluskey J.E., J. flmer. Chem. Soc.

77 (1955) 3315

149. Walker, G.J., Carbohydr. Res. 82 (1980) 404

150. Usui, T., Yamaoka, N., Matsuda, K., Tuzimura, K., Sugiyama, H. and & Seto, S., J. Chem. Soc. Perkin I (1973) 2425

151. Bock, K., Pedersen, C. and Pedersen, H., Adv. Carbohydr. Chem. and

Biochem. 42 (1984) 193 230

152. Colsori, P., Jennings, H.J. and Smith, I.C.P., J. Amer. Chem. Soc.

96 (1974) 8081

153. Bradbury, J.H. and Collins J.G. Carbohydr. Res. 71 (1979) 16

154. Germaine, G.R., Chludzinski, A.M. and Schachtele, C.F.,

J. Bacterid. 120 (1974) 287

155. Wong, S.S. and Frey, P.A., Biochemistry, 16 (1977) 298

156. Wong, S.S., Cassim, J.Y. and Frey, P.A., Biochemistry,

17 (1978) 516