Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The purification and characterization of two cellulose-binding, glycosylated cellulases from the bacterium… Langsford, Maureen Lynn 1988

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

Item Metadata

Download

Media
831-UBC_1988_A1 L36.pdf [ 13.65MB ]
Metadata
JSON: 831-1.0098066.json
JSON-LD: 831-1.0098066-ld.json
RDF/XML (Pretty): 831-1.0098066-rdf.xml
RDF/JSON: 831-1.0098066-rdf.json
Turtle: 831-1.0098066-turtle.txt
N-Triples: 831-1.0098066-rdf-ntriples.txt
Original Record: 831-1.0098066-source.json
Full Text
831-1.0098066-fulltext.txt
Citation
831-1.0098066.ris

Full Text

THE PURIFICATION AND CHARACTERIZATION OF TWO CELLULOSE-BINDING, GLYCOSYLATED CELLULASES FROM THE BACTERIUM CELLULOMONAS FIMI By Maureen Lynn Langsford M . S c , The Univers i ty of B r i t i s h Columbia, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Microbiology) We accept th i s thes is as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Apr i l 1988 Maureen Lynn Langsford, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, 1 agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of M i c r o b i o l o g y The University of British Columbia Vancouver, Canada Date A p r i l 22, 1988  DE-6 (2/88) i i ABSTRACT Cellulomonas f imi secretes several c e l l u l a s e a c t i v i t i e s as well as protease a c t i v i t y into the cul ture medium. In cont rast , few a c t i v i t i e s are bound to the c e l l u l o s e in the c u l t u r e . To character ize the c e l l u l a s e system and to Identify cloned gene products, i t was necessary to pur i f y nat i ve , intact c e l l u l a s e s . We hypothesized that the ce l lu lose -bound c e l l u l a s e s would be protected from p r o t e o l y s i s , and therefore represent the intact enzymes. Two c e l l u l a s e s were p u r i f i e d from Cellulomonas f l m i . Av ice l was recovered from cul tures and the proteins were eluted from i t with guanidine-HCl (Gdn-HCl). The Gdn-HCl extract was f ract ionated by Concanavalin A-Sepharose a f f i n i t y column chromatography and by Mono Q anion exchange column chromatography. The c e l l u l a s e s p u r i f i e d by th is procedure were an endoglucanase, EngA, and an exoglucanase, Exg. The p u r i f i e d enzymes were character i zed . EngA has Mr 57,000, pi 8 .2 , and is 10 % mannose by weight. Exg has Mv 56,000, p i 5 .8 , and is 8 % mannose by weight. Two recombinant DNA plasmids were i d e n t i f i e d as encoding EngA and Exg. The recombinant gene products were not g l ycosy la ted . The ro le of g l ycosy la t ion was studied by comparing some propert ies of the recombinant EngA and Exg with the native EngA and Exg. Both g lycosy lated and unglycosylated forms bound to A v i c e l . S e n s i t i v i t y to the C.f i m i protease was also compared. The g lycosy lated enzymes were protected from pro teo lys i s when bound to c e l l u l o s e . In cont rast , the unglycosylated forms were processed to y i e l d a c t i v e , truncated products with great ly reduced a f f i n i t y for c e l l u l o s e . The cleavage s i te was predicted based on s ize of the products and r e a c t i v i t y with ant i -PT serum. The N-terminal region of EngA and the C-terminal of Exg show 50 % conservation of sequence (Warren et a l . , 1986). This region appears to be the c e l l u l o s e - b i n d i n g domain and is not required for the i i i hydro lys is of soluble substrates . The C . f lml protease can p a r t i a l l y degrade g lycosy lated EngA when i t is not bound to c e l l u l o s e . Some of the mult iple CMCase a c t i v i t i e s in cul ture supernatants are derived from EngA by p a r t i a l p r o t e o l y s i s . i v TABLE OF CONTENTS PAGE ABSTRACT i i TABLE OF CONTENTS iv LIST OF TABLES v i i LIST OF FIGURES v i i i LIST OF ABBREVIATIONS x ACKNOWLEDGEMENTS x i i INTRODUCTION I. Background 1 II. Ce l lu lase Systems of Bacter ia 5 A. Cellulomonas 6 B. C lostr id ium 14 C. Thermonospora fusca 16 D. B a c i l l u s 17 E. Acet i v ibr1o ce11ulolvtleum 18 F. Pseudomonas 18 G. ErwInia chrvsanthemi 20 H. Bactero ides succinogenes 20 I. Ruminococcus 21 J . Fungal Ce l lu lases 22 III. P o s t - t r a n s l a t i o n a l Modi f icat ions 23 A. P ro teo lys is 23 B. G lycosy lat ion of Proteins 25 C. G lycosy lat ion of Ce l lu lases 31 IV. Objectives of This Research 33 MATERIALS AND METHODS I. Reagents and Solut ions 34 II. Bac te r ia l S t ra ins and Growth Media 34 III. Culture Condit ions 35 IV. Preparation of Culture Supernatants 35 V CONTENTS continued PAGE MATERIALS AND METHODS V. Enzyme and Protein Assays 35 VI. Polyacrylamide Gel E lect rophores is (PAGE) 36 VII. Immunoadsorption Chromatography 38 VIII. Immunoprecipitation 38 IX. Maxicel l Labe l l i ng 38 X. Ol igonucleot ide P u r i f i c a t i o n and Dot Blot Hybr id izat ion 39 XI. Native Ce l lu lase P u r i f i c a t i o n 40 XII. Amino Terminal Amino Acid Sequence Determinations 42 XIII . Amino Acid Composition Analys is 42 XIV. Carbohydrate Analys is 42 XV. Substrate Binding Assays 43 XVI. Preparat ion of Recombinant Ce l lu lases for Protease Digests 44 XVII. Preparat ion of Native Ce l lu lases for Protease Digests 45 XVIII. Preparat ion of Protease 45 XIX. Protease Digests of So lub le , Native Ce l lu lases 45 XX. Protease Digests of Substrate-bound Ce l lu lases 46 RESULTS I. The M u l t i p l i c i t y of A c t i v i t i e s in the Cel lu lase System 47 II. P u r i f i c a t i o n and Character i zat ion of Ce l lu lose -bound, Native Ce l lu lases 54 A. P u r i f i c a t i o n of Ce l lu lose -bound, Native Ce l lu lases 54 B. Character i zat ion of Cel lu lose -bound, Native Ce l lu lases 57 III. I d e n t i f i c a t i o n of the Recombinant Plasmids Expressing the Corresponding Ce l lu lases 71 IV. Function of the Glycosyl Groups of the Cel lu lose-bound Ce l lu lases 76 V. Character i zat ion of the C.f i m i Protease 79 VI. The Role of Proteo lys is in Creating Mult ip le Ce l lu lase A c t i v i t i e s 85 v i CONTENTS continued PAGE RESULTS VII. G lycosy la t ion Prevents the P r o t e o l y t i c Cleavage of Cel lu lose-bound Ce l lu lases 90 DISCUSSION I. P u r i f i c a t i o n and Character i zat ion of EngA and Exg 97 II. G lycosy lat ion 99 III. P ro teo lys i s 101 IV. Future Prospects 108 LITERATURE CITED 109 Appendix A E . c o l i s t ra ins and plasmids. 126 Appendix B SDS-PAGE M K c a l i b r a t i o n . 127 Appendix C Determination of serine and threonine 128 from the amino ac id a n a l y s i s . Appendix D GLC chromatograms of a l d i t o l acetates . 129 Mass spect ra . 130 Appendix E Western blot analys is contro l developed 131 with normal rabbi t serum. Appendix F Western blot ana lys is contro l with 132 ant i -PT serum. v i i LIST OF TABLES PAGE I. Substrates Used in the Measurement of Ce l lu lase A c t i v i t y 4 II. Ce l lu lases of Bacter ia 7 III. B a c t e r i a l Glycoproteins 28 IV. A c t i v i t y in Stored Supernatants 50 V. E lu t ion of Ce l lu lase A c t i v i t y From Avice l With Guanidine-HCl 55 VI. P u r i f i c a t i o n of the Native Endoglucanase From C . f imi 59 VII. P u r i f i c a t i o n of the Native Exoglucanase From C . f imi 60 VIII . Amino Acid Composition of EngA 68 IX. Amino Acid Composition of Exg 69 X. Carbohydrate Compositions of EngA and Exg 70 XI. Recombinant Ce l lu lase A c t i v i t y Recovered From Sepharose 4B and Con A-Sepharose 4B 77 XII. The E f fec t of Carbon Source on Protease Induction 80 XIII . Ce l lu lase A c t i v i t y in Reaction Supernatants Af ter Incubation With Protease 86 v i i i LIST OF FIGURES PAGE 1. Proposed mechanism for hydro lys is of c e l l u l o s e at the macromolecular leve l by complete fungal c e l l u l o l y t i c enzyme systems. 2 2. E lect ron micrograph of Cellulomonas f imi with A v i c e l . 12 3. E lec t rophoret i c p r o f i l e of c e l l u l a s e a c t i v i t y in cul ture supernatants. 13 4. Schematic presentat ion of the procedure for pu r i f y ing ce l lu lose -bound EngA and Exg. 41 5. Two dimensional PAGE of a 6 day cul ture supernatant. 48 6. F ract ionat ion of cul ture supernatant and ce l lu lose -bound proteins by Con A-Sepharose column chromatography. 49 7. Non-denaturing PAGE analys is of supernatants stored with or without PMSF. 51 8. Immunoadsorption of cul ture supernatant polypept ides . 53 9. Preparative f rac t ionat ion of ce l lu lose -bound c e l l u l a s e s by Con A-Sepharose chromatography. 56 10. FPLC anion exchange column chromatography of a c t i v i t y peaks I and II recovered from Con A-Sepharose. 58 11. SDS-PAGE ana lys i s of p u r i f i e d c e l l u l a s e s . 61 12. Non-denaturing PAGE analys is of p u r i f i e d c e l l u l a s e s . 62 13. I s o e l e c t r i c focussing ana lys is of p u r i f i e d c e l l u l a s e s in non-urea g e l s . 64 14. I s o e l e c t r i c focussing ana lys is of p u r i f i e d c e l l u l a s e s in urea g e l s . 65 15. The amino terminal amino ac id sequence of Exg. 66 16. Amino terminal amino ac id sequences of EngAl and EngA2, and the sequence of the o l igonucleot ide probe. 67 17. Polypeptides encoded by cloned DNA fragments. 72 i x LIST OF FIGURES continued PAGE 18. Immunoprecipitation of the pECl polypept ide. 74 19. Dot b lot hybr id i za t ion with the EngA o l igonucleot ide probe. 75 20. Comparison of Av ice l binding for native and recombinant EngA and Exg. 78 21. Induction of protease a c t i v i t y during growth on n o n - c e l l u l o s i c subst rates . 82 22. IEF ana lys is of protease a c t i v i t y induced during growth on various carbon sources. 83 23. Hide Pouder Azure a c t i v i t y p r o f i l e from an IEF g e l . 84 24. SDS-PAGE analys is of soluble EngAl a f t e r incubation with protease. 87 25. SDS-PAGE analys is of soluble Exg a f t e r incubation with protease. 89 26. SDS-PAGE analys is of the native and recombinant Avicel -bound EngA a f t e r incubation with protease. 92 27. SDS-PAGE analys is of the native and recombinant Avicel -bound Exg a f t e r incubation with protease. 94 28. Western blot analys is of the recombinant c e l l u l a s e s a f t e r incubation with protease. 96 29. Proposed b i func t iona l organizat ion of EngA and Exg 105 X ABBREVIATIONS BSA bovine serum albumin CBH T . reese i eel lobiohvdrolase cen A C . f imi cjene encodina EncrA cex C . f imi qene encoding Exct CMC carboxymethylce11ulose CMCase . enzyme which hydrolyses CMC Con A concanavalin A, a l e c t i n DNS d i n i t r o s a l i c y l i c ac id EG endo-ft -1,4-glucanase Endo H endo-f5-N-acetylglucosaminidase H Eng endo-f i -1,4-glucanase ER endoplasmic ret iculum Exg exo-|3- 1,4-glucanase F P L C ™ Fast Protein/Peptide L iqu id Chromatography GalNAc N-acetylgalactosamine Gdn-HCl guanidine-HCl GDP guanosine 5 ' -diphosphate GLC gas l i q u i d chromatography GlcNAc N-acetylglucosam ine HBAH hydroxybenzoic ac id hydrazine IEF i s o e l e c t r i c focussing ig immunoglobulin (type A or G) IPTG i sopropy l - f l -D - th i ogalactopyranos ide kb k i l o base (pai rs ) Me-Glc <x-me thyl -D -g lucos ide Mr r e l a t i v e molecular mass mRNA messenger RNA MUC me thyl umbel 1 i f eryl - (S-D-ce 11 ob i os ide PAGE polyacrylamide gel e lect rophores is PAS periodate S c h i f f ' s reagent PBS phosphate buffered sa l ine PMSF phenyl methy lsu l fony l f luor ide pNPC p-n i t ropheny- f l -D -ce1lobios ide x i p N P C a s e e n z y m e w h i c h h y d r o l y s e s p N P C P T b o x a s t r e t c h o f 2 0 - 2 2 a m i n o a c i d s c o n s i s t i n g s o l e l y o f p r o l i n e a n d t h r e o n i n e r e s i d u e s P T H p h e n y l t h i o h y d a n t i o n Rt m o b i l i t y r e l a t i v e t o t h e f r o n t S D S s o d i u m d o d e c y l s u l f a t e s p n t s u p e r n a t a n t S S C s t a n d a r d s o d i u m c i t r a t e T F A t r i f 1 u o r o a c e t i c a c i d T N P - C M C t r i n i t r o p h e n y l - c a r b o x y m e t h y l c e l l u l o s e UDP U r i d i n e 5 ' - d i p h o s p h a t e ACKNOWLEDGEMENTS I am indebted to Drs. D.G. K i l b u r n , R.C. M i l l e r , J r . and R . A . J . Warren for the i r superv is ion and support throughout th is work. I g r a t e f u l l y acknowledge Ms. K ie l land and Dr. R.Olefson for performing the amino terminal sequence analyses , Drs. R. MacGi l l i v ray and R. Cut ler for he lpfu l d iscuss ions and assistance with the amino ac id composition analyses , Dr. T. Atkinson for synthesiz ing the o l igonuc leot ides , Dr. G .S .S . Dutton for generously al lowing me to analyse the carbohydrates on his GLC, Dr. P. Reid and Ms. C. Park for he lpfu l d iscuss ions regarding g lycoprote in analyses, and A. Wong for the e lect ron microscopy work. I am gratefu l to W. Wakarchuk for photographic s e r v i c e s . I thank a l l the members of the Ce l lu lase Group for the i r r e l i a b l e c r i t i c i s m , and e s p e c i a l l y the women of Room 206: J . B e t t s , B.Gerhard, E.Kwan, and H.Smith, for the i r f r iendsh ip and technica l ass i s tance . I am gratefu l to Dr. J . K o h l i for h is encouragement to complete the t h e s i s . This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada. I dedicate th is thes is to my parents . 1 INTRODUCTION I Background Ce l lu lose is a component of plant c e l l wa l ls . The enzymatic breakdown of c e l l u l o s e to glucose has been widely studied with the intent ion of e s t a b l i s h i n g an economic means of recyc l ing waste plant matter to produce a var iety of valuable products (Mandels, 1983). However, several features of c e l l u l o s e structure make i t res i s tant to enzymatic hydro l ys i s . At the primary l e v e l , c e l l u l o s e may be considered a polymer of ce l lob iose molecules. The rather f l a t , r i b b o n - l i k e structure permits dense packing of polymers, and extensive hydrogen bonding within and between polymers creates highly c r y s t a l l i n e , r i g i d f i b e r s . Amorphous or disordered regions, comprising approximately 15% (Detroy and S t . J u l i a n , 1983) a l ternate with c r y s t a l l i n e reg ions . The degree of c r y s t a l 1 i n i t y may vary within a s ing le f iber and between f ibe rs from d i f f e r e n t sources. Another leve l of complexity is imposed on c e l l u l o s e structure by i t s assoc ia t ion with hemicel lulose and l i g n i n . Cotton is an example of a highly c r y s t a l l i n e c e l l u l o s e occurr ing in a pure form. The a b i l i t y of a c e l l u l a s e system to hydrolyse cotton has been presented as the test for true c e l l u l a s e a c t i v i t y (Johnson et a l . , 1982). The current model for the enzymatic hydro lys is of c e l l u l o s e is diagrammed in F ig 1. Hydrolysis is e f fec ted by three c lasses of enzymes, each of which hydrolyses the fl-1,4 g l y c o s i d i c l inkage between glucose units in the c e l l u l o s e cha in . Exoglucanases (Exg) ( E . C . 3 . 2 . 1 . 9 1 and E . C . 3 . 2 . 1 . 7 4 ) remove ce l lob iose or glucose un i t s , r e s p e c t i v e l y , from chain ends; endoglucanases (Eng) ( E . C . 3 . 2 . 1 . 4 ) hydrolyse bonds randomly within the cha in ; fl-glucosidases ( E . C . 3 . 2 . 1 . 2 1 ) hydrolyse ce l lob iose to g lucose. The f i r s t step in the enzymatic breakdown of c e l l u l o s e probably is the loca l reduct ion of the c r y s t a l l i n i t y to generate regions of 2 Amorphogenesls 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Endogluconose 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Endogluconose Exocelloblohydrolose \1/ Endogluconose Exocsllobiohydrolose ^ (Exoglucohydrolose) O-O O o O O O 0 0 O-O O O O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0-0 OO O O O £-GlucosIdase F ig 1. Proposed mechanism for hydro lys is of c e l l u l o s e at the macromolecular leve l by complete fungal c e l l u l o l y t i c enzyme systems (from Coughlan, 1985). Amorphogenesis is the s p l i t t i n g of hydrogen bonds to d is rupt c e l l u l o s e c r y s t a l s . The amorphous regions become suscept ib le to h y d r o l y t i c a t tack , f i r s t by endoglucanases then by exoglucanases. Exoglucanases are of two types, those that re lease c e l l o b l o s e uni ts (exoce l lob io -hydrolases) or glucose un i ts (exoglucohydrolases) from the nonreducing ends of c e l l u l o s e cha ins . fl-Glucosidases hydrolyse ce l lob lose to g lucose. Glucose uni ts are represented as c i r c l e s . 3 amorphous s t ruc ture . The un ident i f ied component responsible has been refered to as "hydrogen bondase" (Enari and Niku-Paavola, 1987), " m i c r o f i b r i l generating factor" ( G r i f f i n et a l . , 1984), and C i swel l ing factor (Montenecourt, 1983). Endoglucanases r e a d i l y attack the amorphous regions generating free chain ends, the substrate for exoglucanases. The l i b e r a t i o n of ce l lob iose provides the substrate for <5-glucos idases. The discovery that the major c e l l u l o l y t i c component of the fungus, Trichoderma reese i . is an exocel lobiohydrolase capable of degrading c r y s t a l l i n e c e l l u l o s e to c e l l o b i o s e , necess i tates some modif icat ion of the current model. There appears to be overlap in the substrate s p e c i f i c i t i e s of the three c lasses of c e l l u l a s e s . The hydro lys is of native c e l l u l o s e is d i f f i c u l t to monitor. Consequently, several der ived or subst i tuted forms of c e l l u l o s e have been used In the laboratory in order to study the a c t i v i t i e s of ind iv idual enzymes and to d i f f e r e n t i a t e between the three c lasses of c e l l u l a s e s . The usefulness of several substrates has been reviewed (Mul l ings , 1985). The substrates that w i l l be mentioned throughout th is work are summarized in Table I. Carboxymethylcellulose (CMC) is frequently used in the study of endoglucanases. It is a soluble c e l l u l o s e and i t s hydro lys is is measured in two ways: by the increase in reducing groups measured as equivalents of reducing sugar, usual ly g lucose, or by a decrease in v i s c o s i t y of the so lu t ion ind icat ing shorter chain lengths. The a c t i v i t y of endoglucanases may be described in terms of the randomness of attack when hydrolysing CMC: a rap id decrease in v i s c o s i t y with a slow increase in the number of reducing ends produced indicates a randomly act ing enzyme (Wood, 1985). Exoglucanase a c t i v i t y has been measured on predominantly c r y s t a l l i n e c e l l u l o s i c s , such as Av ice l and f i l t e r paper, by assaying for an increase In s o l u b i l i z e d reducing sugars or for a decrease in t u r b i d i t y . More recent l y , exoglucanase a c t i v i t y has been measured by hydro lys is of subst i tuted ee l lob ios ides 4 Table I. Substrates Used In The Measurement Of Ce l lu lase Act i v i t y* Substrate Descr ipt i on Assay products Cotton Alpha c r y s t a l l i n e , DP B : approx. 10,000, insoluble decrease in DP, decrease in t e n s i l e strength Av i ce l m i c r o c r y s t a l l i n e , DP: approx. 200, insoluble reducing groups, decrease in t u r b i d i t y F i l t e r Paper H 3 P0« swollen c e l l u l o s e c r y s t a l 1ine, insoluble amorphous, insoluble reducing groups, decrease in t u r b i d i t y reducing groups Carboxymethyl ee l lu lose amorphous, soluble reducing groups, decrease in DP chromophore r e l e a s e 0 p-N i t r o -phenyl-ft -D-ce11ob ios ide ce11otr iose analogue, soluble chromophore release (p-Nitrophenyl) Methyl -umbe11i feryl-fi-D-ce l lob ios ide e e l l o t r i o s e analogue, soluble chromophore release (methylumbel 1i fe r y l ) A References: Wood, 1985; Knowles et a l . , 1987. B DP: degree of polymerizat ion of glucose un i t s . c Tr initrophenyl -CMC (TNP-CMC) is an inso lub le , noncrys ta l l ine substrate . Hydrolysis re leases shor ter , soluble ce11ooligosaccharides with the attached chromophore, thus imparting a yellow co lor to the assay buf fer (Robson & Chambliss, 1984). 5 such as p -n i t ropheny l - f l -D - ce l lob ios ide (pNPC). The c e l l u l a s e systems of several microorganisms are composed of mult iple a c t i v i t i e s . In p a r t i c u l a r , most c e l l u l o l y t i c organisms secrete several endoglucanases. In the remainder of the INTRODUCTION, l i t e r a t u r e is reviewed descr ib ing the c e l l u l a s e systems of several c e l l u l o l y t i c b a c t e r i a , focussing on two c lasses of c e l l u l a s e s , the endo-and exoglucanases. Fungal c e l l u l a s e s , in p a r t i c u l a r the c e l l u l a s e system of T.reese i » are b r i e f l y descr ibed . Fol lowing t h i s , p o s t - t r a n s l a t i o n a l modif icat ions of c e l l u l a s e s are d iscussed, s p e c i f i c a l l y p ro teo lys i s and g l y c o s y l a t i o n . The theme throughout is towards understanding the o r i g i n of the m u l t i p l i c i t y of c e l l u l a s e a c t i v i t i e s . II Ce l lu lase Systems of Bacter ia C e l l u l o l y t i c bacter ia may be c l a s s i f i e d into three broad categor ies : s o i l composting organisms (aerobic or anaerobic , mesophil ic or thermophi l i c ) , plant pathogenic organisms (aerob ic ) , and rumen organisms (anaerobic) . Symbionts l i v i n g in marine shipworms (Waterbury et a l . , 1983; G r i f f i n et a l . , 1987) and in termites have also been character ized (McCarthy, 1987). The a b i l i t y to hydrolyze c e l l u l o s e is frequently associated with c e l l surface protruberances and the a b i l i t y to adhere to c e l l u l o s e (Lamed et a l . , 1987; Morris and Cole, 1987). Ce l lu lase a c t i v i t y may be detected in cul ture supernatants or i t may form complexes, e i ther c e l l - a s s o c i a t e d , as for Ruminococcus albus and CIostridiurn thermoceHum, or subst rate -assoc iated as in Ce11ulomonas sp. (Lamed et a l . , 1987). Among the many species of c e l l u l o l y t i c b a c t e r i a , the c e l l u l a s e complexes of only a few are well understood. However, a common theme is apparent: c e l l u l o l y t i c organisms elaborate mult iple a c t i v i t i e s with s i m i l a r substrate s p e c i f i c i t i e s . The biogenesis of the mult iple a c t i v i t i e s is not c l e a r l y understood. A combination of mult ip le genes and 6 d i f f e r e n t i a l p o s t - t r a n s l a t i o n a l modif icat ions appears to play a r o l e . To date, the best character i zed systems are those of C lostr id ium thermocellum and Cellulomonas f i m i . Technical d i f f i c u l t i e s , a t t r ibu ted to the hydrophobic nature of the p ro te ins , have hampered the charac te r i za t ion of ind iv idual components of the c e l l u l a s e systems of Bactero ides  succ inogenes (Groleau and Forsberg, 1983) and Acet iv ib r io  c e l l u l o l v t i c u m (Mackenzie et a l . , 1987). The c e l l u l a s e s of the aerob ic , thermophi l ic actinomycetes Thermomonospora fusca and Thermomonospora curvata as well as of Streptomvces spp. have been examined and were recent ly reviewed (McCarthy, 1987). Some of the propert ies of ind iv idual c e l l u l a s e s are summarized in Table II. A. Cellulomonas Cellulomonas species have been iso lated world wide from various habi tats such as s o i l s , urban wastes, sugar cane, and brewery sludge. At least one iso late has been reported to adhere to c e l l u l o s e (Lamed et a l . , 1987). Ce l lu lase a c t i v i t y in these organisms is inducible and regulated by ca tabo l i te repress ion (Stewart and Leatherwood, 1976; Beguin et a l . , 1977; Stoppok et a l . , 1982). Mutants have been iso lated with improved a b i l i t i e s to degrade cotton (Haggett et a l . , 1979), as well as mutants which are re leased from catabo l i te repress ion (Stewart and Leatherwood, 1976) or which are res i s tan t to end product i n h i b i t i o n (Choudhury et a l . , 1980). Upon induction by su i tab le carbon sources, mult iple c e l l u l a s e a c t i v i t i e s appear in cul ture supernatants, and some bind to insoluble c e l l u l o s e (Beguin and E isen , 1978; Langsford et a l . , 1984; Bagnara et a l . , 1986). fi-glucosidases and some endoglucanases are ce l l -bound (Beguin et a l . , 1977; Stoppok et a l . , 1982). Several c e l l u l a s e s have been p u r i f i e d from Ce11ulomonas spec ies . Three endoglucanases, one soluble and two bound to 7 Table II. Ce l lu lases of Bacter ia Organ ism & Enzyme Mr X IO"3 Pi CHO % Opt i mai pH Temp °C Spec i f ic Act i v i ty U mg-1 Ref A c e t i v i b r i o c e i l u l o i v t i c u s ExgCl 38.0 ND ND ND ND 0. 25 A 1 EngC2 33.0 ND ND ND ND 1 . 0 1 A 1 EngC3 Bac i11 us subt i l l s 10.4 ND ND ND ND 1 . 6 A 1 Eng 35.2 51 . 2 1 ND ND 4.8 45 (58) 5 5 0 T C 2,3 Bac i11 us sp. N4 El ( ) E2 54.0 ( ) 44. 2 ND ND 10.0 10.0 40 (60) 40 (80) 1. 2 C 2. 3 C 4,5 4,5 Bac i11 us sp. 1139 Eng 92.0 3. 1 ND 9.0 40 (50) 5 9 . 0 C 6 Bactero ides succinocrenes Eng 43.0 ND ND 6.0 50 4 3 . 0 C 7,8 Ce1\ulomonas r i m i EngAB 57.0 8.2 10 7.0 30 370.0° 9,10 Exg B 56.0 5.8 8 7.0 37 8 5 . 0 C 9. 3 N 9, 10 1 1 EngB 1 10. 0 R ND ND 7.0 37 ND 12 EngCl C2 130.0 120.0 ND ND ND ND ND 13 Cellulomonas s t r a i n l l b c EngAB 49-52 ND ConA* ND ND 12.6 C 14 EngBB 53.5 ND ConA* ND ND 0 . 9 C 14 EngC 118.0 ND PAS" ND ND 14.9 C 14 8 Table II continued. Organism & Enzyme Mv p i CHO X I O ' 3 % Opt imal pH Temp °C Spec i f ic Act i v i ty U mg-1 Ref Cellulomonas  termentans EngCFA EngCFB Cellulomonas  uda CB4 Eng Engl Eng2 Eng 3 Eng4 Eng5 Eng6 Clostr id ium  thermoce11um ML'IB — EG A EG B EG C EG D 40.0 58.0 57. 0 R 37. 0 R Eng Clostr id ium s p . B W Eng ND ND ND ND ND ND 7.0 7.0 37 37 6-8 55-60 15.8 C 4 . 6 N 17.6 C 1.5 N ND 15 15 16 63.0 ND ND 6-7 50 9 9 4 c 17 44.4 ND ND 6-7 40 2078 c 17 62.0 ND ND 6-7 50 2212 c 17 77.0 ND ND 6-7 50 255 c 17 143.0 ND ND 6-7 50 1776° 17 121.0 ND ND 6-7 50 4623° 17 56.0 6.2 ND 6.0 60 (80) 140.0 C 18 55. 0 R 5 3 . 0 R 66.0 ND ND ND ND 60 .0° 19,20 38. 0 R 40.0 6.2 ND 6 . 0 ° 6 . 6 N 60 (65) 27. 0 C 270.0" 21 65. 0 R 74.0 6 0 . 0 L m ND ND 6.0 60 428.O c 2 . 9 N 20,22 23 90.0 6.7 1 1 5.2 62 65 .0° 0. 17A 24 95.0 3.85 2 6.4 60 (80) 45. l c 0 . 8 A 0. 4 F ' c t 25 9 Table II. continued Organism & Enzyme Mr X 10~3 Pi CHO % Opt i mal pH Temp °C S p e c i f i c Act i v i ty U mg"1 Ref CvtoDhaaa Eng p 6.3 ND ND ND ND 9 .5° 26 Eng1 8.7 ND ND ND ND 3 . 2 C 26 Erwinia chrvsanthem i EngYE 35.0 55. 0 M 8.2 ND 5.5 37 33. 0 C 0. 16A 27,28 EngZ 45.0 4.3 ND 7.0 37 200.0° 0 . 6 A 27,28 Pseudomonas f1uorescens v a r . c e l l u l o s a EngA 40.0 ND 3.3 8.0 30 1150*° 29,30 EngB 85.0 ND 11.3 8.0 30 90* c 29,30 EngC p ND ND 8.5 7.0 30 1860*c 30 Pseudomonas solanacearum Eng 43.0 ND ND 7.5 50 105° 31 Ruminococcus aibus Exg 100.0° 5.3 ND 6.8 37 2 I*N 32 Ruminococcus tlave lac lens Exg 118.0° ND PAS" 5.0 39-45 0. 1 1 N 0.02"-0 .025 8 W 33 StreDtomvces f lavoqr i seus Exg 46.0 4.0 ConA~ ND ND 0 . 0 1 A 34 ThermomonosDora fusca XV. E l 94.0 3.6 < 1 6.5 56 (74) 768° 8 7 s w 0 . 0 1 R 35 E2 46.0 4.5 25 6.5 56 (58) 77° 4 4 S W <0.01 r 35 E5 45.0 ND ND ND ND ND 36 10 Table II. Ce l lu lases of B a c t e r i a . A l l c e l l u l a s e s were p u r i f i e d from cul ture supernatant except where noted: E=cell ex t rac t , P=periplasm, I=intrace11ular, B= ce l lu lose -bound ; L=cellulosome, R=recombinant gene product p u r i f i e d from E . c o l i . D=subunit s ize of dimer, M=immunological1y re la ted minor band. Other abrev iat ions a re : ND=not determined, CHO=carbohydrate covalent ly associated with the prote in expressed as weight % where known, or as PAS*' - 5 =does (not) s ta in with periodate S c h i f f ' s , ConA** -' =does (not) bind Concanavalin A. Temperatures given in brackets indicate the highest temperature at which the enzyme retained a c t i v i t y . S p e c i f i c A c t i v i t y is expressed as e i t h e r : 1. pmoles reducing sugar re leased per min per mg prote in fo r : C=CMC, A=Avicel , F = f i l t e r paper, SW=phosphoric ac id swollen c e l l u l o s e , Ct=cotton; 2. pmoles p-ni t rophenyl re leased per min per mg protein for N=pNPC; 3. pmoles t r in i t ropheny l re leased per min per mg prote in for TC = TNP-CMC; or 4. absorbance units at A e S o for *C, A40S for *N. See Table I for further desc r ip t ion of these subst rates . Ref=references: 1. Saddler & Khan, 1981; 2. Robson & Chambliss, 1984; 3. Robson & Chambliss, 1986; 4. Horikoshi et a l . , 1984; 5. Fukumori et a l . , 1986; 6. Fukumori et a l . , 1985; 7. Groleau & Forsberg, 1983; 8. Schel lhorn & Forsberg, 1984; 9. G i lkes et a l . , 1984a; 10. th i s work; 11. Langsford et a l . , 1987; 12. Owalabi et a l . , in press ; 13. B.Moser & N .G i lkes , unpublished observat ions; 14. Beguin & E i s e n , 1978; 15. Bagnara et a l . , 1986; 16. Nakamura et a l . , 1986; 17. Prasertsan & Doe l le , 1986; 18. Petre et a l . , 1981; 19. Beguin et a l . , 1983; 20. Beguin et a l . , 1987; 21. Petre et a l . , 1986; 22. J o l i f f et a l . , 1986a & 1986b; 23. Lamed & Bayer, in press ; 24. Ng & Zeikus, 1981; 25. Creuzet & Fr i xon , 1983; 26. Chang & Thayer, 1977; 27. Boyer et a l . , 1987; 28. Boyer et a l . , 1984b; 29. Yamane et a l . , 1970; 30. Yoshikawa et a l . , 1974; 31. Sche11, 1987; 32. Ohmiya et a l . , 1982; 33. Gardner et a l . , 1987; 34. MacKenzie et a l . , 1984; 35. Calza et a l . , 1985; 36. Ghangas & Wilson, 1987. 11 c e l l u l o s e , were p u r i f i e d from s t r a i n I lbc (Beguin and E isen , 1978). The enzyme p u r i f i e d from cul ture supernatant is la rge , with Mr 118,000, and not g l ycosy la ted , while the ce l lu lose -bound enzymes are g lycosy lated and smal ler , with MrS 53,500 and 49,000-52,000. Cellulomonas fermentans produces several enzymes act ive against CMC and pNPC when grown on ce l lob lose or c e l l u l o s e . Some components bind to the c e l l u l o s e . Two endoglucanases, M r S 40,000 and 58,000, were p u r i f i e d from cul ture supernatant (Bagnara et a l . , 1986). Six endoglucanases were detected in the cul ture supernatant of a ce l lu lose -grown s t r a i n UQM 2903 (Prasertsan and Doel le , 1986). These enzymes appear to be d i s t i n c t with respect to optimal hydro lys is cond i t ions , a f f i n i t y for CMC, and s u s c e p t i b i l i t y to i n h i b i t o r s . The MrS range from 44,000 to 140,000. C . f imi (F ig 2) elaborates several c e l l u l a s e s , some of which bind t i g h t l y to c e l l u l o s e (Langsford et a l . , 1984; G i lkes et a l . , 1984b). During growth on low concentrat ions of c e l l u l o s e , 10-12 e lec t rophoret ica l1y d i s t i n c t c e l l u l o l y t i c components were observed in cul ture supernatants (F ig 3) . The genes encoding 4 c e l l u l a s e s have been cloned and character ized (O 'Ne i l l et a l . , 1986a; Wong et a l . , 1986; Owalabi et a l . , in press ; B.Moser, unpublished observat ions) . The t r a n s c r i p t i o n a l regu lat ion of the genes cen A, cen B, and qex has been studied (Greenberg et a l . , 1987a; 1987b). The corresponding recombinant gene products, EngA, EngB and Exg, have also been character ized (Gi lkes et a l . , 1984a; 1984b) and p u r i f i e d from Escher ich ia c o l i recombinant DNA clones (Guo et a l . , in press ; Owalabi et a l . , in p ress ) . EngB, Mr 110,000, was p u r i f i e d by c e l l u l o s e a f f i n i t y from the periplasm of an E . c o l i recombinant DNA clone contain ing the cen B gene (Owalabi et a l . , in p ress ) . EngC has been p u r i f i e d from C. f imi cul ture supernatants. It was resolved into two components, EngCl, Mr 130,000, and EngC2, Mr 120,000 (N.Gi lkes and B.Moser, unpublished observat ions) . Both components have the same amino terminal amino ac id sequence (B.Moser, personal communication). 12 Fig 2. E lect ron micrograph of Cellulomonas f i m i with A v i c e l . Av icel was i so la ted from a C . f imi cul ture grown on 3 % Av ice l for 6 days. The sample was prepared by standard procedures for examination in an S100 Cambridge scanning e lec t ron microscope. Magnif icat ion is X 3,930. 13 10 20 30 40 SLICE No. F ig 3. E lec t rophoret ic p r o f i l e of c e l l u l a s e a c t i v i t y in cul ture supernatants. C . f imi was grown for 4 days on 0.2 % A v i c e l . The c e l l - f r e e cul ture supernatants obtained by cent r i fugat ion were concentrated by u l t r a f i l t r a t i o n . The supernatant was f ract ionated by non-denaturing PAGE. Following e lec t rophores i s , one lane was stained with Coomassie blue and scanned for prote in (upper panel ) . Another lane was excised from the gel and s l i c e d ho r i zon ta l l y into 2mm s l i c e s . Each s l i c e was eluted in P bu f fe r . The eluates were assayed for CMCase a c t i v i t y (lower panel ) . The arrows in the upper panel indicate prote in peaks which had a c t i v i t y . This f igure is provided by N .G i lkes . 14 B . C lostr id ium The c e l l u l a s e s of C lostr id ium thermoceHum are thermostabile and form a large g lobular protein complex termed the cellulosome (Lamed, et a l . , 1983a). Cellulosomes are synthesized during growth on ce l lob iose or c e l l u l o s e (Bayer et a l . , 1983). In a d d i t i o n , the c e l l u l a s e complex has true c e l l u l a s e a c t i v i t y , having the a b i l i t y to s o l u b i l i z e native and derived forms of c e l l u l o s e at a rate and extent comparable to T . reese i c e l l u l a s e (Johnson et a l . , 1982). P u r i f i e d cel lulosomes degrade insoluble c e l l u l o s e to ce l lob iose as the main hydro lys is product (Lamed, et a l . , 1983a; Johnson et a l . , 1982; Hon-nami et a l . , 1986b). In young cu l tures of the YS s t r a i n , cel lulosomes are associated with the outer c e l l wall and appear to mediate the attachment of c e l l s to substrate . As cu l tures age, the c e l l s are released from the cel lulosomes but the cel lulosomes remain bound to c e l l u l o s e un t i l the substrate is s o l u b i l i z e d (Bayer and Lamed, 1986). A low molecular weight yellow a f f i n i t y substance secreted by the ATCC 31449 s t r a i n coats the c e l l u l o s e and may f a c i l i t a t e the binding of the cellulosome to c e l l u l o s e (Ljungdahl et a l . , 1983; Coughlan et a l . , 1985). Cellulosomes p u r i f i e d from cul ture supernatant or from c e l l u l o s e have a diameter of 18-21 nm and a molecular weight of 2.1 X 10 s , estimated by sedimentation v e l o c i t i e s , (Lamed et a l . , 1983b) or 5 X 10 s , estimated from gel f i l t r a t i o n and sucrose density gradients (Hon-nami et a l . , 1986a). Larger cel lulosomes p u r i f i e d from c e l l u l o s e appear to be polycel lulosomes composed of 13-24 unit s ize cel lulosomes (Coughlan et a l . , 1985). Cellulosomes contain at least 14-20 d i f fe ren t polypeptides with Mrs from 45,000 to 210,000. Most of the polypeptides have CMCase a c t i v i t y , the strongest a c t i v i t i e s have MvS of 50,000-60,000 (Hon-nami et a l . , 1986a) or 75,000 and 170,000 (Lamed et a l . , 1983b). In further studies on the s t r a i n ATCC 31449, cel lulosomes in cul ture supernatants were compared with the two s ize c lasses 15 of substrate-bound ce1lulosomes. Cel lu lose-bound cel lulosomes were eluted from yellow c e l l u l o s e with water. Then the yellow c e l l u l o s e was used as an a f f i n i t y matrix to pur i f y the cel lulosomes in cul ture supernatants. SDS-PAGE polypeptide patterns of a l l three cellulosome types are very s i m i l a r but d i f f e r in the r e l a t i v e amounts of each polypept ide. The CMCase s p e c i f i c a c t i v i t i e s are a lso very s i m i l a r . However, the i r a c t i v i t i e s toward c r y s t a l l i n e c e l l u l o s e d i f f e r : the large substrate-bound cellulosome has the most while the soluble cellulosome has the least a c t i v i t y towards c r y s t a l l i n e c e l l u l o s e . Ce l lu lase a c t i v i t y in cul ture supernatants which was not associated with a cellulosome complex has a 10-100 fo ld lower s p e c i f i c a c t i v i t y towards CMC and very low a c t i v i t y towards c r y s t a l l i n e c e l l u l o s e (Hon-nami et a l . , 1986a and 1986b). The large polypeptide of Mr 210,000 is not c e l l u l o l y t i c but i t is the major immunoreactive component of the cellulosome (Lamed et a l . , 1983a). It is sens i t i ve to papain (Lamed et a l . , 1983a), g lycosy lated (Lamed and Bayer, in press ; Wu and Demain, in press) and present in the cel lulosomes of s t ra ins LQR1, ATCC 27405 (Bayer et a l . , 1983) and ATCC 31449 (Hon-nami et a l . , 1986a). The large prote in appears to be essent ia l for prote in complex form- at ion on c e l l u l o s e and synergizes with other enzymes in the hydro lys is of c r y s t a l l i n e c e l l u l o s e (Wu and Demain,in p ress ) . Two endo-f i -1,4-glucanases have been p u r i f i e d d i r e c t l y from C.thermocellum cul ture supernatants: a c e l l u l a s e , EGA, with Mr 56,000, p u r i f i e d from C.thermocellum s t r a i n NCIB 10682 (Petre et a l . , 1981), and a c e l l u l a s e with Mr 83,000-94,000, from C.thermocellum s t r a i n LQRI. The l a t t e r appears to be 11% carbohydrate (Ng and Zeikus , 1981). Three add i t iona l endoglucanases, EGB, EGC, and EGD, of C. thermocellum s t r a i n NCIB 10682 have been p u r i f i e d from E . c o l i s t ra ins car ry ing recombinant DNA plasmids contain ing the corresponding C.thermocellum genes. The recombinant gene 16 products were used to obtain s p e c i f i c a n t i s e r a , which then were used to ident i f y the corresponding c e l l u l a s e s , Mr 66,000, 40,000 and 74,000, respec t i ve l y , in C.thermoce11um culture supernatants (Beguin et a l . , 1983; Petre et a l . , 1986; J o l i f f et a l . , 1986a). Monoclonal antiserum ra ised against EGD reacts with a Mr 60,000 prote in in the cellulosome and const i tu tes 2-3% of the cel lu losomal prote in (Lamed and Bayer, in p ress ) . A l together , 21 d i s t i n c t fragments of C.thermocellum DNA ( s t ra in NCIB 10682) have been cloned into E . c o l i and express c e l l u l a s e a c t i v i t y (M i l l e t et a l . , 1985; Romaniec et a l . , 1987). This number c l o s e l y approximates the number of polypeptides detected in the cellulosome (Coughlan et a l . , 1985). Ce l lu lases have been character i zed from other C lostr id ium spec ies . Culture supernatant from the organism C.stercorar ium. was f ract ionated into 3 a c t i v i t i e s : xylanase, CMCase, and a weak c e l l u l a s e a c t i v i t y . This th i rd a c t i v i t y may be an exoglucanase because i t does not reduce the v i s c o s i t y of a CMC so lu t ion but i t does synergize with the CMCase a c t i v i t y in the hydro lys is of c r y s t a l l i n e c e l l u l o s e . The whole c e l l u l a s e complex (a sonicate of a 40h cu l ture) has only weak a c t i v i t y towards cotton (Creuzet et a l . , 1983). A CIostr id ium-1ike thermophi l ic organism, s t r a i n BW, e f f i c i e n t l y degrades c r y s t a l l i n e c e l l u l o s e . A c e l l u l a s e p u r i f i e d from th is s t r a i n has Mr 90,000-99,000 and is 2% carbohydrate (Creuzet and F r i xon , 1983). C. Thermomonospora fusca T . fusca YX is a gram v a r i a b l e , aerob ic , thermophi l ic , fi lamentous s o i l bacterium. It secretes an induc ib le , thermostable c e l l u l a s e system capable of e f f i c i e n t degradation of Av ice l (Hagerdahl et a l . , 1978). Ce l lu lase a c t i v i t y is found in cul ture supernatant and, to a lesser extent , bound to A v i c e l . The bound a c t i v i t y could be eluted 17 with water. The e lec t rophoret i c pattern of the bound proteins is s i m i l a r but simpler than the supernatant pattern (Hagerdahl et a l . , 1978). Two endoglucanases comprising approximately 90% of the supernatant a c t i v i t y have been extensively character i zed . One of these is 25% carbohydrate and both undergo p r o t e o l y t i c processing to smaller forms (Calza et a l . , 1985). A th i rd endoglucanase which makes up 5-10% of the CMCase a c t i v i t y in cul ture supernatants, has been cloned and expressed in E . c o l i (Ghangas and Wilson, 1987). The component(s) in cu l ture supernatant which f a c i l i t a t e s the hydro lys is of c r y s t a l l i n e c e l l u l o s e has not been i s o l a t e d . D. Bac i11 us General ly , B a c i l l u s species have been observed to produce low leve ls of c e l l u l a s e a c t i v i t y . B.subt i1 is DLG was iso lated from s o i l and se lected for CMC hydro lys is during growth on c e l l o b i o s e . The c e l l u l a s e of th is organism is not regulated by ca tabo l i te repress ion . In f a c t , growth on glucose st imulates c e 1 l u l a s e - s p e c i f i c mRNA production (Robson and Chambliss, 1987). An endoglucanase p u r i f i e d from th i s organism occurs in two forms: an i n t r a c e l l u l a r form, Mr 51,500, and an e x t r a c e l l u l a r form, Mr 35,200. P r o t e o l y t i c processing apparently occurs from the carboxyl terminus. The enzyme has a higher s p e c i f i c a c t i v i t y on TNP-CMC than does EG1 from the fungus T.reese i QM9414 (Robson and Chambliss, 1987). This spec ies , however, apparently lacks a complete c e l l u l a s e system since no a c t i v i t y was detectable on c r y s t a l l i n e c e l l u l o s e or on ce l lob iose (Robson and Chambliss, 1984). Two a l k a l o p h i l i c Baci11 us s t r a i n s , N4 and 1139, were iso lated from s o i l . S t ra in N4 secretes endoglucanase a c t i v i t y without induction by CMC or c e l l o b i o s e . Two very s i m i l a r CMCases p u r i f i e d from s t r a i n N4 have Mr approximately 50,000, and are stable to high pH and 18 temperature (Horikoshi et a l . , 1984). The genes encoding these two enzymes are very homologous (Fukumori et a l . , 1986) . A th i rd endoglucanase, Mr 92,000, was p u r i f i e d from s t r a i n 1139. It is inducible only during growth on CMC and in lesser amounts during growth on c e l l o b i o s e . During growth on 1% A v i c e l , no a c t i v i t y was found in cul ture supernatants. It has a higher a c t i v i t y towards CMC than the endoglucanases from s t r a i n N4 and optimal a c t i v i t y at pH 9 (Fukumori et a l . , 1985). The gene encoding th is c e l l u l a s e shares extensive homology with the genes encoding the other 2 a l k a l o p h i l i c CMCases (Fukumori et a l . , 1986). E. A c e t i v i b r i o c e l l u l o l v t i c u m A . c e l l u l o l v t icum is an obl igate anaerobe capable of u t i l i z i n g only c e l l u l o s e , ce l lob iose or s a l i c i n as carbon sources (Saddler et a l . , 1980). It contains a complete c e l l u l a s e system which w i l l hydrolyse c r y s t a l l i n e Av ice l (MacKenzie et a l . , 1985) at rates comparable to the T.reese i c e l l u l a s e complex (MacKenzie et a l . , 1987). Cel lobiose i n h i b i t s Avicelase but not CMCase production (Saddler et a l . , 1980). ft-glucosidase, Mr 85,000, exoglucanase CI, Mr 38,000, and endoglucanase C2 and C3, Mr 33,000 and 10,400, r e s p e c t i v e l y , were detected in cul ture supernatants (Saddler and Khan, 1981). C2 appears to be a tr imer of C3. Up to 6 endoglucanases were detected a f t e r separation by i s o e l e c t r i c focussing (MacKenzie et a l . , 1987). Avicelase was adsorbed by Av ice l (MacKenzie et a l . , 1985) and c e l l u l a s e a c t i v i t y could be eluted from c e l l u l o s e with water (MacKenzie et a l . , 1987) . Avicel -bound proteins include a large p ro te in , Mr 210,000, which is ant igenica l1y re la ted to the SI subunit of the C.thermocellum cellulosome (Lamed et a l . , 1987). F. Pseudomonas P. f luorescens var ce11u losa is an aerob ic , mesophi l ic , gram negative rod which was iso lated from s o i l . It secretes 19 inducible c e l l u l a s e complex when grown on c e l l u l o s e , sophorose or low leve ls (<0.01%) of c e l l o b i o s e . Two c e l l u l a s e s from cul ture supernatant and one c e l l u l a s e from osmotic shock f l u i d were p u r i f i e d . A l l three enzymes are endoglucanases and capable of hydrolyz ing to some degree a var iety of c e l l o d e x t r i n s , amorphous c e l l u l o s e and A v i c e l . The three c e l l u l a s e s are g l ycosy la ted , containing d i f f e r e n t r a t i o s of neutral and amino sugars (Yamane et a l . , 1970). A time course ana lys is of the c e l l u l a s e s secreted into the cul ture medium indicated that one of the enzymes was derived from the other. I n i t i a l l y , component B, Mr approximately 85,000 and e l e c t r o p h o r e t i c a l l y slow, comprised 70% of the supernatant a c t i v i t y . After 8 days of cul ture growth, component B was reduced to 20% of the supernatant CMCase a c t i v i t y . The to ta l a c t i v i t y d id not decrease with cul ture age. Instead, there was an increase in the amount of component A, an e lec t rophoret ica l1y fast a c t i v i t y with Mr 40,000. Intermediate s ize a c t i v i t i e s i so lated by gel f i l t r a t i o n were resolved e l e c t r o p h o r e t i c a l l y into d i f f e r e n t amounts of A and B, the r a t i o of A to B increasing with decreasing s ize (Yoshikawa et a l . , 1974). One way to expla in these rather confusing r e s u l t s would be to assume that component B is ac tua l l y a tetramer, with a subunit Mr 21,000. S p e c i f i c p ro teo lys is would remove hal f of each subunit r e s u l t i n g in a tetramer of component A with a subunit Mr 10,000. The compositions of the various gel f i l t r a t i o n f rac t ions then would be: 4B, 3B:1A, 2B:2A, 1B:3A, and 4A. It is of note that component B has 3 fo ld more carbohydrate and 10 fo ld less a c t i v i t y towards CMC than component A (Yamane et a l . , 1970). A fragment of P.f1uorescens var c e l l u l o s a DNA, 0.6kb long, enough to encode a Mr 20,000 p ro te in , was s u f f i c i e n t to encode c e l l u l a s e a c t i v i t y when cloned into E . c o l i (Wolff et a l . , 1986). Ce l lu lases have been character i zed in other Pseudomonads. P.solanacearum. the causative agent of le tha l b a c t e r i a l wi l t d isease , secretes a CMCase, Mr 43,000 ( S c h e l l , 1987). A 20 Pseudomonas sp. Isolated from act ivated sludge elaborated several c e l l u l a s e s during growth on c e l l u l o s e . Two endoglucanases from cul ture supernatant and one membrane-bound CMCase as well as several fl-glucosidases were p a r t i a l l y p u r i f i e d . No a c t i v i t y towards c r y s t a l l i n e c e l l u l o s e was detected (Ramasamy and Verachtert , 1980). G. Erwinia chrvsanthemi E.chrvsanthemi is a pathogenic enterobacterium which causes s o f t - r o t disease in p lan ts . S t ra in 3665 secretes a const i tu t i ve low leve l of CMCase which can be increased by d i f f e r e n t condit ions which reduce the growth ra te , suggesting regulat ion by ca tabo l i te repress ion (Boyer et a l . , 1984a). Two endoglucanases have been character ized from s t r a i n 3665. The f i r s t , designated EngZ, is a secreted enzyme induced during growth on g l y c e r o l . The enzyme is unstable: in cul ture supernatants i t tends to aggregate when concentrated or s tored , and i t is suscept ib le to p ro teo lys i s (Boyer et a l . , 1984b). The second enzyme, designated Eng Y, is found in c e l l extracts and poss ib ly in cul ture supernatants. Ant isera ra ised against the p u r i f i e d recombinant DNA-encoded enzyme, EngY, recognized two d i f f e r e n t s ized proteins in E.chrvsanthemi c e l l ex t rac ts ; therefore , th i s enzyme may also be processed (Boyer et a l . , 1987). Both enzymes hydrolyze CMC r e a d i l y , and to lesser extents, phosphoric ac id swollen c e l l u l o s e and A v i c e l . H. Bactero ides succinogenes Bacteroides succinogenes is a s t r i c t l y anaerobic, gram negative rumen organism. The c e l l u l a s e system is cons t i tu t i ve but more a c t i v i t y is produced during growth on c e l l u l o s e . The organism degrades c r y s t a l l i n e c e l l u l o s i c s such as cotton and straw, s lowly, suggesting that i t has a complete but i n e f f i c i e n t c e l l u l a s e system. C e l l s and outer membrane ves ic les adhere t i g h t l y to c e l l u l o s e (Forsberg et a l . , 1981). Over 50% of the CMCase a c t i v i t y in cul ture supernatants is 21 membrane-associated and sedimentable (Groleau and Forsberg, 1983). The membrane-bound f r a c t i o n was s o l u b i l i z e d with Tr i ton X-100 and f ract ionated into 4 components. A l l 4 components exhib i t very random a c t i v i t y on CMC. Each in turn were resolved into mult iple components by non-denaturing PAGE (Schel lhorn and Forsberg, 1984). The non-sedimentable CMCase a c t i v i t y in cul ture supernatants was f ract ionated into 3 components: a very large aggregate, Mr greater than 4 X 10 s ; a second component, Mr 45,000; and a t h i r d , smaller component. Each of these in turn were further resolved into mult ip le components by non-denaturing PAGE. The large aggregate and the smallest component have s i m i l a r behavior on CMC, hydrolyz ing i t in a less random manner. The smallest component a lso hydrolyzes pNPC. The large aggregate binds t i g h t l y to phosphoric ac id -swol len Av ice l (Schel lhorn and Forsberg, 1984). Since the 7 c e l l u l o l y t i c f rac t ions are comprised of mult iple polypept ides, the actual number of components in the c e l l u l a s e system of B. succ inoctenes is not known. Their hydrophobic nature and tendency to aggregate has impeded p u r i f i c a t i o n attempts. While i t is tempting to speculate that the membrane ves ic les are analogous to the Clostr id ium  thermoce11 urn cel lu losomes, a recent report concludes that the membrane ves ic les are produced as a resu l t of aging and are not essent ia l for c e l l u l o s e degradation (Gaudet and G a i l l a r d , 1987) . I. Rum inococcus Like B.succinogenes. R.albus is an anaerobic rumen organism. It u t i l i z e s more disordered forms of c e l l u l o s e , not having the c a p a b i l i t y to hydrolyse c r y s t a l l i n e c e l l u l o s e as B.succinogenes does. Ce l lu lase a c t i v i t y (CMCase), present in cu l tures grown on c e l l u l o s e or c e l l o b i o s e , is found loosely associated with the bac te r ia l c e l l wall and in the cul ture supernatant. CMCase a c t i v i t y in the c e l l wall is cor re la ted with the presence of a capsule (Stack and Hungate, 1984). 22 Based on gel f i l t r a t i o n experiments (Wood et a l . , 1982), the wall bound enzyme is la rge , Mv greater than 1.5 X 10 s , but not sedimentable. The s ize of the e x t r a c e l l u l a r enzyme var ies with the carbon source ranging from Hr 30,000 to greater than 1.5 X 10 6 . The high molecular weight a c t i v i t y apparently d i ssoc ia tes into smaller a c t i v i t i e s . An exoglucanase, Mr 100,000, was p u r i f i e d from a bovine rumen Isolate (Ohmiya et a l . , 1982). Another exoglucanase was character ized from R . f l a v e f a c i e n s . It is a dimer, each subunit is Mr 118,000 (Gardner et a l . , 1987). J . Fungal Ce l lu lases The c e l l u l a s e systems of fungi are general ly complex having several c e l l u l o l y t i c components, e s p e c i a l l y mult iple endoglucanases (Beldman et a l . , 1985? Bodenmann et a l . , 1985; Er iksson and Pettersson, 1975; Maloney and Coughlan, 1981). The most widely studied c e l l u l o l y t i c organism is Trichoderma  reese i . The c e l l u l a s e s of T.reese i have been extensively character ized and reviewed (Enari and Niku-Paavola, 1987; T e e r i , 1987; Knowles et a l . , 1987; Coughlan, 1985; Wood, 1985; Montenecourt, 1983). The T . reese i c e l l u l a s e system e f f i c i e n t l y hydrolyzes c r y s t a l l i n e c e l l u l o s e to glucose by the synerg i s t i c act ion of eel1obiohydrolases, endoglucanases, and (3-gl ucos idases. Character i zat ion of the indiv idual components has been hampered by the occurrence of isoenzymes. S t r a i n d i f fe rences and cul ture condit ions have a s i g n i f i c a n t e f f e c t on the number and propert ies of c e l l u l a s e s occurr ing in cu l ture supernatants. M u l t i p l i c i t y is thought to resu l t from p r o t e o l y t i c processing (Nakayama et a l . , 1976), d i f f e r e n t i a l g l ycosy la t ion (Gum and Brown, 1977), and in teract ion with other components in the cul ture medium (Niku-Paavola et a l . , 1985). In another c e l l u l o l y t i c fungus, Schizophvl1um commune. heterogeneous t ransc r ip ts from indiv idual genes also contr ibute to the observed m u l t i p l i c i t y (Wi l l i ck and S e l i g y , 1985). The propert ies of several fungal c e l l u l a s e s have been summarized (Coughlan, 1985). Fungal c e l l u l a s e s are t y p i c a l l y a c i d i c , 23 being i s o e l e c t r i c at pH 3 . 5 - 5 . 0 , g l ycosy la ted , and monomeric with an average Mr- 42,000. In T.reese i , two ce11obiohydro lases , CBH I and CBH II, have been i d e n t i f i e d by prote in p u r i f i c a t i o n (Fagerstam and Pettersson, 1980) and gene charac te r i za t ion (Shoemaker et a l . , 1983; Teeri et a l . , 1983; Chen et a l . , 1987; Teeri et a l . , 1987). These two enzymes act s y n e r g i s t i c a l l y and are s u f f i c i e n t to degrade c r y s t a l l i n e c e l l u l o s e to c e l l o b i o s e . CBH I has 3 isoforms which d i f f e r by pi but are immunologically r e l a t e d . CBH I is the major c e l l u l a s e , c o n s t i t u t i n g 60 % of the 20 grams to ta l secreted prote in per l i t r e of cu l ture (Knowles et a l . , 1987). As many as 6 endoglucanases have been i d e n t i f i e d by prote in p u r i f i c a t i o n (Beldman et a l . , 1985) and two genes have been character ized encoding EG I and EG III ( P e n t t i l a et a l . , 1986; Knowles et a l . , 1987). EG I represents 5-10 % of the to ta l secreted prote in and synergizes with CBH II (Henrissat et a l . , 1985). Amino ac id and DNA sequencing has revealed in teres t ing s t ruc tura l s i m i l a r i t i e s between the 4 T . reese i enzymes character ized thus far (Knowles et a l . , 1987). III. P o s t - t r a n s l a t i o n a l Modi f icat ions P o s t - t r a n s l a t i o n a l modif icat ions are par t l y responsible for generating the mult iple c e l l u l o l y t i c components of many organisms. Modi f icat ions to the t rans lated gene products include p ro teo lys i s and g l y c o s y l a t i o n . A. P ro teo lys i s In fung i , proteases appear to have several e f f e c t s on c e l l u l a s e s . An endoglucanase p u r i f i e d from T.reese i by ion exchange chromatography gave r i s e to several a c t i v i t i e s a f te r incubation with protease (Nakayama et a l . , 1976). The new a c t i v i t i e s were s l i g h t l y a l te red in the degree of randomness with which they attacked CMC. A protease-negative mutant of Humicola g r i sea var . thermoidea produced a 128,000 dalton c e l l u l a s e capable of hydrolyz ing and binding to Av ice l while a 24 protease over-producing mutant secreted a c e l l u l a s e which was hal f the s ize and f a i l e d to hydrolyze or bind to Av ice l (Hayashida and Mo, 1986). In addi t ion to a f f e c t i n g substrate s p e c i f i c i t y , proteases may be e f f e c t i v e in re leas ing c e l l u l a s e s from the hyphal c e l l wall (Kubicek, 1981; She i r -Ne iss and Montenecourt, 1984; Rao et a l . , 1986). Ac t i va t ion phenomena have been observed in two d i f f e r e n t fung i , Sporotrichum  pulverulentum (Er iksson and Pettersson, 1982) and P e n i c i l l i u m  ianthinel lum (Deshpande et a l . , 1984) where incubation of the c e l l u l a s e preparat ion with protease resu l ted in an increased c e l l u l a s e a c t i v i t y . These observations led to the postu lat ion that some c e l l u l a s e s may ex is t as zymogens or as p r o c e l l u l a s e s , respect i ve1y. The ro le of p ro teo lys i s in the biogenesis of mult ip le c e l l u l a s e s is c o n t r o v e r s i a l . Two reports (Labudova and Farkas, 1983; Kammel and Kubicek, 1985) concluded that the mult iple endoglucanases and fl-glucosidases of T.reese i were unique gene products since they were observed ear ly in cul ture growth. However, two e x t r a c e l l u l a r endoglucanases of Schizophvllum  commune. EG 1 and EG 2, d i f f e r in s ize but share the same amino terminal sequence. The p o s s i b i l i t y that EG 2 was derived p ro teo l y t i ca l1y from EG 1 was discussed (Paice et a l . , 1984). A Baci1lus subt i1 is DLG endoglucanase is thought to be processed at the carboxyl terminus p r io r to secret ion (Robson and Chambliss, 1987). Protease a c t i v i t y is secreted concurrent ly with c e l l u l a s e a c t i v i t y during growth of E.chrvsanthem i (Boyer et a l . , 1984b), T . fusca (Calza et a l . , 1985), C . f imi (Langsford et a l . , 1984), and T.reese i (She i r -Ne iss and Montenecourt, 1984). Immunological c r o s s - r e a c t i v i t y between several c e l l u l o l y t i c components of one organism (Enger and Sleeper , 1965; Fagerstam and Pettersson, 1979; Langsford et a l . , 1984; Niku-Paavola et a l . , 1985) suggests a p r o t e o l y t i c mechanism. However, p ro teo lys is has been d i r e c t l y implicated in a f f e c t i n g the m u l t i p l i c i t y of c e l l u l a s e s in only a few cases where a p u r i f i e d 25 or p a r t i a l l y p u r i f i e d c e l l u l a s e was treated with a co- induced protease secreted by the same host . In the actinomycete T . f u s c a , two fami l ies of immunologically re la ted endoglucanases were p u r i f i e d (Calza et a l . , 1985). The EG 1 family consisted of three enzymes. The largest enzyme, when incubated with protease, gave r i s e to the other two components. 81% of the i n i t i a l a c t i v i t y was recovered in the smallest product which was missing a fragment of Mr 24,000. B. G lycosy lat ion of Proteins Eukaryotic proteins other than cytoplasmic proteins are usual ly g lycoproteins (Sharon and L i s , 1982). Glycoproteins are proteins containing o l igosacchar ides covalent ly attached to se lected amino ac id residues (Hughes, 1983). Ol igosaccharides l inked v ia the C( l ) of N-acetylglucosamine (GlcNAc) to the amide group of asparagine are described as N- l inked sugars or N-glycans. 0-glycans are sugars joined through the C( l ) by g l y c o s i d i c bonds to the hydroxylated side chains of serine (Ser ) , threonine (Thr) , hydroxylysine, or hydroxyprol ine. O- l inkages to hydroxy1ysine and hydroxyproline are more common in col lagens and plant g lycoprote ins . Mannose (Man) 0-1 inked to Ser and Thr is common in fungi while N-acetylgalactosamine (GalNAc) 0-1 inked to Ser and Thr is common in higher organisms. One g lycoprote in may contain a mixture of N- and 0-1 inked glyeans. The structure and biosynthesis of g lycoproteins are widely studied and have been reviewed (Kornfeld and Kornfe ld , 1980 and 1985; Kukuruzinska et a l . , 1987). B r i e f l y , the techniques employed to e luc idate glycan st ructures include chemical , enzymic and phys ical approaches. A t yp i ca l chemical method is methylat ion. The glycopeptide is methylated p r io r to ac id hydro l ys i s . Analys is of the products by g a s - l i q u i d chromatography indicates which hydroxyl groups of the sugar were protected by methylation and which ones were involved in l inkages . Ol igosacchar ides may be sequenced with s p e c i f i c exo-26 and endoglycosidases. S t i l l further information about structure may be obtained from mass spectrometry and nuclear magnetic resonance analyses. N and 0 glycopeptide l inkages are frequently d is t ingu ished by the i r s e n s i t i v i t y to mild a l k a l i and i n h i b i t o r s . 0-1 inked sugars are removed in a ft-elimination react ion with mild a l k a l i while N- l inkages are r e s i s t a n t . Synthesis of N- l inked sugars is inh ib i ted by the a n t i b i o t i c tunicamycin and the sugar analogue 2-deoxy-D-glucose (2-DG). Tunicamycin i n h i b i t s the t ransfer of GlcNAc to do l i cho l pyrophosphate ( D o l - P - P ) , while 2-DG competes for do l i cho l phosphate <Dol-P) to prevent synthesis of l i p i d - 1 i n k e d o l igosacchar ides (Hubbart and Ivat t , 1981). The prote in sequence and conformation s p e c i f i e s which of the appropriate residues w i l l be g lycosy la ted . N- l inkages require the amino ac id sequence Asn-X-Thr(Ser) where X can be any amino ac id except Asp or Pro. In yeast there is a bias towards g l ycosy la t ing Asn-X-Thr sequences (Ernst et a l . , 1987). Not a l l such sequences are g l ycosy la ted . Most N-glycosylated asparagines are located in peptide sequences which favor the formation of ff-turns (Kornfeld and Kornfe ld , 1985). Recognition sequences for 0-1 inked sugars have not been i d e n t i f i e d . However, pro l ine near a serine or threonine s i g n i f i c a n t l y enhances the l i k e l i h o o d of those residues being g lycosy lated (Ernst et a l . , 1987; Lehle and Bause, 1984; Kornfeld and Kornfe ld , 1980) suggesting that a c c e s s i b i l i t y of the residue is more important. Biosynthesis of the glycans occurs v ia a complex pathway and is unique for 0 - and N- l inked un i t s . For N- l inked glycans, the o l igosacchar ide precursor is assembled at the endoplasmic ret iculum (ER) on the l i p i d c a r r i e r D o l - P - P . It is then t ransfered to a properly or iented and access ib le Asn-X-Thr(Ser) sequence as the prote in is t rans located across the ER membrane. Inside the ER, the precurser o l igosacchar ide moiety undergoes processing before the g lycoprote in is transported to the g o l g i . In the g o l g i , the ol igosacchar ide is further modif ied: chains 27 may be elongated and terminal addi t ions are made before the g lycoprotein reaches i t s mature form (Kornfeld and Kornfe ld , 1985). In yeast , the 0-1 inked sugars are t ransfered one sugar at a time to the prote in in the ER. The f i r s t mannose is t ransfered from the l i p i d c a r r i e r Dol -P while subsequent addi t ions are made v ia GDP nucleotide c a r r i e r s (Kukuruzinska et a l . , 1987). In higher organisms, GalNAc is t ransfered to the prote in from the nucleotide c a r r i e r UDP. Subsequent sugars are added one at a time v ia UDP and GDP c a r r i e r s (Schachter and Roseman, 1980). Unt i l recent l y , bacter ia were considered not to have g lycoprote ins . However, several g lycoproteins in bacter ia have now been i d e n t i f i e d , usual ly by s ta in ing pos i t i ve with p e r i o d a t e - S c h i f f ' s reagent in polyacrylamide g e l s . Some examples of b a c t e r i a l g lycoproteins have been summarized in Table III. T y p i c a l l y , these proteins are e x t r a c e l l u l a r , e i ther on the surface of the bacterium or secreted to the environment. Surface g lycoproteins may have s t ruc tu ra l functions or have adherance funct ions f a c i l i t a t i n g pathogenic i ty , while secreted glycoproteins are enzymes, e i ther carbohydrate depolymerases or proteases. With the exception of the Halobacterium surface g l ycoprote in , few have been extensively character i zed . Halobacterium sal inar ium lacks an outer membrane and peptidoglyean t yp ica l of gram negative c e l l s . Instead, i t contains a surface layer of a s ing le g lycoprote in species which is t i g h t l y associated with the cytoplasmic membrane. The structure of the glycans has been determined and recent ly reviewed <Sumper, 1987). The g lycoprote in contains 20 c lustered neutral d isacchar ides of the structure G l c - 1 , 3 - G a l - O - T h r , 10 su l fa ted o l igosacchar ides with the linkage G lc -N -Asn , and one la rge , su l fa ted glycan with the linkage GalNAc-N-Asn. Moreover, the bacterium possesses g lycosy lated f l a g e l l i n s which contain su l fa ted o l igosacchar ides with the linkage G lc -N -Asn . The g lycosy lated asparagine 28 Table III. Bac te r ia l Glycoproteins Source and Glycoprote in Locat ion Mr X 1 0 - 3 CHO' Linkage Ref Bac i11 us  subt i 1 is" 168 Bac i11o-peptidase F B . sub t i l is NA-64 a-amylase Baci11 us stear- othermoph i ius S layer prote in B .s tearo -thermoph11 us a-amylase B .s tearo -thermoph i1 us Glucoamylase Cellulomonas  ii m i EngA Exg Clostr id ium  thermoce Hum Eng Clostr id ium BW Eng C.thermocellum SI subunit cu l ture spnt cu l ture spnt c e l l wall cu l ture spnt 50.2 55.0 93.0 120.0 147.0 170.0 47.0 cul ture ND spnt cu l ture spnt, c e l l u l o s e cul ture spnt, ce11ulose cul ture spnt cul ture spnt 57.0 56.0 90.0 95.0 c e l l u -losome 210.0 250.0 GalN 3% Rha trace Man G le , G a l , GlcNAc 1 1-15% 27% 10% Man 8% Man 11% 2% 25-40% (75% G a l , 25% hexN) ND1 2% neutral ND 7% GlcN N- l inked ND ND O-Thr (Ser) O-Thr (Ser) ND ND 5,6 7 5,6 7 0-1 inked 10 29 Table III. continued Source and Glycoprote in Locat ion X I O - 3 CHO* Linkage Ref Corvn ibacter i um  sepedon i um Phytotoxin Halobacter i um sal inar ium Ce l l surface glycoprote in Mvcobacter i um  bov is BL'G Tubercul in act i ve prote i n Mvxococcus  xanthus" Surface prote in Pseudomonas r1uorescens v a r . c e l l u l o s a EngA EngB EngC Streptococcus  s a l i v a r i u s HB~ F i b r i l AgC Thermomono-spora fusca XY E 2 cul ture 21.4 34% Man spnt 18.5% Glc 5% Gal 1% Rha c e l l wall crude c e l l extract c e l l surface 200.0 10-12% (Glc ,GlcNAc, Gal ,GalNAc, HexUA-OSCT3) 26.0 21.5% Man 1% Glc 74.0 13% neutral sugars 1% GalNAc 1.4% UA cul ture 40.0 3.3% spnt (Gal ,Glc ,Man, Fuc, GlcNAc) cul ture 85.0 11.3% spnt (Glc ,Gal ,Man, Fuc, GlcNAc) periplasm ND c e l l 220-surface 280 8.5% (Glc,Man,Fuc, Gal,GalNAc 29% hexose (Glc ,Gal ,Rha) 12% GlcN cul ture spnt 46.0 25% Man-O-Thr 11 Gal -O-Thr Glc -N-Asn GalNAc-N-Asn ND 12 ND ND ND ND ND ND 13 14 15 15 15 16 17 18 30 Table III. continued Source and Location Mr CH0A Linkage Ref Glycoprotein X 1 0 - 3 Treponema  palidum Surface c e l l 30-35 GlcN ND 19 proteins surface 59.0 A CHO = carbohydrate: Gal=galactose, Glc=glucose, Man=mannose, Fuc=fucose, Rha=rhamnose, GlcN=glucosamine, GalN=galactosamine, GalNAc=N-acetylgalactosamine, GlcNAc= N-acetylglucosamine, hexN=hexosamine, UA=uronic a c i d . B ND = not determined. Ref: 1. Roitsch and Hageman, 1983; 2. Yamane et a l . , 1973; 3. Kupcu et a l . , 1984; 4. S r i vas tava , 1984; 5. th is work; 6. N .G i lkes , unpublished observat ion; 7. Gi lkes et a l . , 1984b; 8. Ng and Zeikus, 1981; 9. Creuzet and Fr ixon , 1983; 10. Lamed and Bayer, in press ; 11. Strobel et a l . , 1972; 12. Sumper, 1987; 13. Yano et a l . , 1984; 14. Maeba, 1986; 15. Yamane et a l . , 1970; 16. Weerkamp et a l . , 1986; 17. Weerkamp and Jacobs, 1982; 18. Calza et a l . , 1985; 19. Moskophidis and Mul ler , 1984. 31 The structure of the glycans has been determined and recent ly reviewed (Sumper, 1987). The g lycoprote in contains 20 c lus tered neutral d isacchar ides of the structure G l c - 1 , 3 - G a l - O - T h r , 10 su l fa ted o l igosacchar ides with the linkage G lc -N -Asn , and one la rge , su l fa ted glycan with the linkage GalNAc-N-Asn. Moreover, the bacterium possesses g lycosy lated f l a g e l l i n s which contain su l fa ted o l igosacchar ides with the linkage G lc -N -Asn . The g lycosy lated asparagine residues a l l occur in the acceptor sequence Asn-X -Thr (Ser ) . Biosynthesis of the su l fa ted o l igosacchar ides occurs on the cytoplasmic membrane and t ransfer of the completed carbohydrate moieties is mediated by l i p i d c a r r i e r s of the eukaryot ic type. C. G lycosy lat ion of Ce l lu lases Most fungal c e l l u l a s e s are g lycoprote ins . The carbohydrate moieties are la rge ly composed of neutral sugars with mannose predominant and glucosamine in trace amounts (Berghem et a l . , 1975; Gum and Brown, 1977; Rudick and E l b e i n , 1975; Shoemaker and Brown, 1978). In a d d i t i o n , arabinose (Er iksson and Pettersson, 1975) and fucose (Wi l l i ck and S e l i g y , 1985) have been detected. T y p i c a l l y , no s i a l i c ac id residues are found. Experiments to determine the glycopeptide l inkage and other s t ruc tura l features have been done for only a few of the c e l l u l a s e s . The carbohydrate attached to ft-glucosidase of Asoergi11 us fumigatus is res i s tant to mild a l k a l i treatment and is presumed to be N- l inked to Asn v ia GlcNAc (Rudick and E l b e i n , 1975). The production of the c e l l u l a s e s of Schizophvl1um commune is sens i t i ve to tunicamycin, suggesting that they contain N- l inked carbohydrate (Wi l l i ck and S e l i g y , 1985). fl-elimination and methylation studies of CBH I of T.reese i demonstrated that most of the neutral carbohydrate occurs in short , unbranched chains of 1 to 4 sugars 0-1 inked through mannose to approximately 16-17 Thr or 32 Ser residues (Gum and Brown, 1976). Peptide and amino ac id ana lys is of CBH I revealed that 3 of the 5 putative Asn-X-Thr(Ser) s i t e s are g lycosy lated (Fagerstam et a l . , 1984) and that these 3 Asn residues occur in regions of the molecule predicted to be exposed regions (Bhikhabhai et a l . , 1985) . The 0-1 inked sugars appear to be c lustered in a region r i c h in hydroxyl amino acids near the C terminus of the molecule (Fagerstam et a l . , 1984). A recent study of the carbohydrate of CBH I from T.reese i s t r a i n VTT-D-80133, has confirmed a s i m i l a r arrangement of 8 to 10 0-1 inked o l igosacchar ides of 1-4 units long, and 3 N- l inked o l igosacchar ides of which 78% are the (Man=GlcNAc2) type and 22% are of the (MansGlcNAca) type (Salovuori et a l . , 1987). Studies with the inh ib i to rs tunicamycin and 2-deoxyglucose suggest that N- l inked sugars are not required for c e l l u l a s e secret ion but do seem to provide some s t a b i l i z i n g e f f e c t s (Merivuori et a l . , 1985a, 1985b), while 0 - l i n k e d sugars seem to be necessary for secret ion (Merivuori et a l . , 1985b; Kubicek et a l . , 1987). Few b a c t e r i a l c e l l u l a s e s have been i d e n t i f i e d as g lycoprote ins . Those c e l l u l a s e s which are g lycosy lated range from as l i t t l e as 2% carbohydrate (Creuzet and Fr i xon , 1983) to as much as 25% (Calza et a l . , 1985). The non-enzymatic SI subunit , Mr 210,000, of the Clostr id ium thermocellum cellulosome is 25-40% carbohydrate. Prel iminary charac te r i za t ion (Lamed and Bayer, in press) revealed that 75% of the sugar is ga lactose . The remainder is N-acetylhexosamine. The carbohydrate is 0 - l i n k e d and occurs as complexes with M rs 500 and 9000. In another s t r a i n , the SI subunit was la rger , Mr 250,000, and the carbohydrate released by mild a l k a l i had M rs 9000 and 20,000. The carbohydrate compositions of the Pseudomonas f1uorescens var. ce1lu losa c e l l u l a s e s A, B, and C were determined by paper chromatography (Yamane et a l . , 1970). The neutral sugars g lucose, ga lactose , mannose and fucose were detected in a l l 33 three enzymes. In a d d i t i o n , GlcNAc was detected in c e l l u l a s e C and GalNac was found in c e l l u l a s e s A and B. IV. Objectives of This Research A combined biochemical and molecular c loning approach has improved our understanding of the c e l l u l a s e system of the gram pos i t i ve bacterium, Ce11ulomonas f imi• The work presented in th is thes is concerns the two major c e l l u l a s e s of C . f i m i , an endoglucanase, EngA, and an exoglucanase, Exg. Both enzymes are g lycosy lated and they bind t i g h t l y to c e l l u l o s e (Langsford et a l . , 1984; Gi lkes et a l . , 1984b). S imi la r a c t i v i t i e s are detected in cul ture supernatants during growth on low concentrat ions of c e l l u l o s e (Langsford et a l . , 1984). To cor re late the substrate-bound a c t i v i t i e s with the supernatant a c t i v i t i e s , and to cor re late the native enzymes with recombinant gene products, the p u r i f i c a t i o n and charac te r i za t ion of the substrate-bound EngA and Exg was undertaken, the resu l t s of which are presented here. The r e l a t i o n s h i p between these enzymes and two recombinant gene products is es tab l i shed . In a d d i t i o n , evidence is given which suggests a ro le for the g l ycosy la t ion of these two enzymes. 34 MATERIALS AND METHODS I Reagents and Solut ions Medium components were from D i f c o . Av ice l PH101 was purchased from FMC In ternat iona l , I re land. Chemicals were reagent grade or better and most were purchased from Sigma, F i sher , or BDH. E lect rophores is reagents were from Biorad except acrylamide which was from Serva. Radionuclides were from NEN. Guanidine-HCl was enzyme grade purchased from BRL. Chromatography grade reagents were used for g a s - l i q u i d chromatography. Buffers for FPLC were made up in HPLC grade water (BDH) and f i l t e r e d through 0.22 pm f i l t e r s ( M i l l i p o r e ) . Buffer compositions were as fo l lows: P bu f fe r : 50 mM sodium and potassium phosphate, pH 7 .0 ; C bu f fe r : 50 mM sodium c i t r a t e , pH 6 . 8 , 0.02 % bovine serum albumin (BSA); phosphate buffered sa l ine (PBS): 137 mM NaCl, 2.7 mM KC1, 1.7 mM K H 2 P 0 « , 8 mM Na*HP0«, pH 7 .4 ; TN bu f fe r : 20 mM T r i s - H C l , pH 7 .4 , 50 mM NaCl; T bu f fe r : 10 mM T r i s - H C l , pH 7 .4 ; standard sodium c i t r a t e (SSC): 0.15 M NaCl, 0.015 M sodium c i t r a t e , 0.5 mM EDTA, pH 7 .2 ; s o l u b i l i z a t i o n mix: 2 % sodium dodecyl su l fa te (SDS), 20 % g l y c e r o l ; 5 % 2-mercaptoethanol, 0.07 M T r i s - H C l , pH 6 . 8 , 0.01 % bromophenylblue. II Bac te r ia l S t ra ins and Growth Media Cellulomonas f imi ATCC 484 was grown on basal s a l t s medium (Stewart and Leatherwood, 1976) supplemented with glucose, carboxymethylcel lulose (CMC), A v i c e l , g l y c e r o l , c e l l o b i o s e , or xylan as necessary for the experiment. Escher ich ia c o l i s t ra ins C600 and CSR603 c a r r i e d recombinant pBR322 plasmids expressing c e l l u l a s e . E . c o l i s t ra ins JM101 and PM191 car r ied recombinant pUC plasmids expressing c e l l u l a s e . A l l E . c o l i s t ra ins except CSR603 were grown in L broth ( M i l l e r , 1972) without g lucose. CSR603 was grown in M9 medium (Zo l le r and Smith, 1983) for l a b e l l i n g of plasm id-coded proteins by the max l - ce l l method (Sancar et a l . , 1979). E .col1 s t ra ins and 35 plasmids are descr ibed in Appendix A. Stocks were maintained in 40 % g lycero l at -20°C . For s o l i d media, 1.2% agar was incorporated. III Culture Condit ions C . f imi precul tures were made from g lycero l stocks d i l u t e d 50 fo ld into basal s a l t s medium supplemented with glucose and casamino ac ids (CAA) and grown to late log phase. Precul tures were d i l u t e d 100 fo ld into fresh medium contain ing no CAA. For large scale enzyme preparat ion , 20 l i t r e cu l tures were grown in an aerated, s t i r r e d fermentor. For t h i s , an intermediary inoculum was grown on basal medium supplemented with 0.2 % Avice l for 2 days, then d i l u t e d 40 fo ld into the fermentor. A l l cu l tures were grown with vigorous aerat ion at 30°C. IV Preparation of Culture Supernatants C e l l s and res idual c e l l u l o s e were removed by cent r i fugat ion for 20 minutes at 10,000 X g and 4°C. The supernatants were decanted and made 0.02 % in NaN3 and 0.3 mM in phenylmethylsul fonyl f luor ide (PMSF). Supernatants were f i l t e r e d twice with suct ion through 2 Whatman glass f i b e r f i l t e r s , then concentrated 50-100 fo ld by u l t r a f i l t r a t i o n through an Amicon PM10 membrane in an Amicon pressure c e l l . V Enzyme and Prote in Assays Ce l lu lase a c t i v i t y was determined by c o l o r i m e t r i c methods measuring the release of reducing groups from CMC. For detect ion with the d i n i t r o s a l i c y l i c ac id (DNS) reagent (Mi 1 Ie r , 1959), 0.25 ml enzyme d i l u t e d appropr iate ly in P bu f fe r , was incubated with 0.5 ml 4% CMC (low v i s c o s i t y ) at 37°C. The react ions were stopped with the addi t ion of 0.8 ml DNS and 50 pi glucose (1 mg ml - 1 ) and steamed for 15 min to develop the c o l o r . The absorbance values at 550 nm were read against reagent blanks containing no enzyme. Glucose equivalents were determined by reference to a standard curve. 36 One unit of enzyme released reducing groups equivalent to 1 pmole glucose m i n - 1 . For detect ion with hydroxybenzoic ac id hydrazine (HBAH) reagent (Lever, 1972), enzyme was incubated at 30°C in C buf fer containing 0.2% CMC. The react ion was stopped by removing 0.5 ml incubation mixture to 1.0 ml HBAH. The react ions were steamed for 12 min and the absorbance at 400 nm was determined. Reducing groups were standardized against g lucose. Exoglucanase a c t i v i t y was measured by the hydro lys is of p -n i t ropheny l - f t -D -ce l lob ios ide (pNPC) (Ohmiya et a l . , 1982). The incubation mixture was 0.5 mM pNPC in P buf fe r . Incubations were c a r r i e d out at 37°C and stopped by the addi t ion of 0.5 ml of 1.0 M Na 2 C0 3 . Absorbance was measured at 400 nm. One unit of a c t i v i t y re leased 1 pmole of pNP min - 1 using E= 18.8 cm2 pmole - 1 . Protease a c t i v i t y was measured on Hide Powder Azure in TN buffer (Rinderknect et a l . , 1968) and standardized against co l lagenase. Af ter incubation at 37°C, unsolubi1ized material was centr i fuged at 12,000 X g and the amount of dye released into the buffer was measured at 595 nm. A c t i v i t y was expressed as pg equivalents of col lagenase or as unit equivalents of co l lagenase; 1 unit of col lagenase re leases 1 pmole L - leuc ine in 5h at pH 7.4 and 37°C. Prote in concentrat ion was measured by absorbance at 280 nm or c o l o r i m e t r i c a l l y according to Bradford (1976) or Lowry et a l . (1951) with BSA as standard. VI Polyacrylamide Gel E lect rophores is (PAGE) SDS-containing denaturing g e l s , 0.75 mm th i ck , were run according to Laemmli (1970). Relat ive mobi l i ty ( M r ) was estimated by comparison with Sigma SDS-6H Mr standards run on the same g e l . Non-denaturing gels containing no SDS, were 6 % acrylamide and 1.5 mm ( a n a l y t i c a l ) or 3 mm (preparative) thick (Jovin et a l . , 1964). Two dimensional gels were obtained by e lectrophoresing samples in non-denaturing gels for the f i r s t 37 dimension, exc is ing one lane and b o i l i n g i t in s o l u b i l i z a t i o n mix, then sea l ing i t with molten agarose to the top of the stacking gel of a SDS-denaturing gel for e lect rophores is in the second dimension. I s o e l e c t r i c focussing (IEF) was performed in v e r t i c a l s lab gels according to published procedures for denaturing gels (Winter et a l . , 1977) and native gels ( G u i l i a n , 1986). IEF gels were 0.75 mm th ick , 7 % acrylamide, 3-5 % ampholines (LKB), 10 % g l y c e r o l , and 4 M urea for denaturing g e l s . pH gradients in denaturing gels were determined by exc is ing one lane, s l i c i n g i t into 4mm s l i c e s , and e l u t i n g each s l i c e in d i s t i l l e d water. Af ter one or more hours, the pH of the water was determined with a pH meter. In native IEF g e l s , pH gradients were estimated from Pharmacia standards pH 3-10. Gels were stained with Coomassie Blue, fuchsin su lphi te (Zacharius et a l . , 1969), or with s i l v e r (Morrissey et a l . , 1981; Wray et a l . , 1981). Gels were preserved by dry ing onto Whatman f i l t e r paper or between layers of cel lophane. A l t e r n a t i v e l y , proteins in gels were t ransfered e lec t rophoret i ca l1y to n i t r o c e l l u l o s e (Towbin et a l . , 1979) and detected immunologically. N i t r o c e l l u l o s e b lots were incubated with s p e c i f i c rabbi t ant i -ce11ulase antiserum, then with goat ant i rabb i t IgG-a lkal ine phosphatase. A lka l ine phosphatase was detected by hydro lys is of 5 -b romo-4 -ch lo ro -3 - indo l y l phosphate (Blake et a l . , 1984). Enzyme a c t i v i t y was detected in gels by one of 3 methods: 1) Gels were soaked in 1.0 mM 4-methylumbe11ifery1-ft-D-c e l l o b i o s i d e (MUC) in P buf fe r . Bands of f luorescence were observed under U.V. l i gh t (302 nm) and photographed. 2) Gels were s l i c e d ho r i zon ta l l y into 2 mm s l i c e s , each s l i c e e luted with 0.5 ml P bu f fe r , and the eluate was assayed as descr ibed above. 3) A c t i v i t y was detected by an overlay technique. After e lec t rophores i s , the acrylamide gel was e q u i l i b r a t e d in P buffer by washing 2 times for 15 min, then sandwiched against a 1 % agarose gel in P buffer contain ing 1 % CMC (high v i scos i t y ) for detect ion of endoglucanase (Beguin, 1983) or 1 % skim milk 38 powder for detect ion of protease (Foltman et a l . , 1985). CMC overlays were stained with Congo Red to observe zones of c l e a r i n g . Milk overlays showed zones of c l o t t i n g where casein was hydrolyzed and were photographed d i r e c t l y . VII Immunoadsorption Chromatography Monoclonal antibody A2/36.17.2 was c r o s s - l i n k e d with dimethyladipimidate to Protein A-Sepharose 4B according to Schneider et a l . (1982) and packed into a column as described (Langsford et a l . 1984). Concentrated cul ture supernatant was appl ied to the column in PBS. Bound proteins were e luted with 0.1 M NI-UOH in 1 ml f r a c t i o n s . Fract ions were immediately l y o p h i l i z e d and the residue resuspended in P buf fer for ana lys is by SDS-PAGE. VIII Immunoprecipitation Proteins l a b e l l e d in maxicel ls were precleared by incubation with normal rabbi t serum (containing an estimated 50 pg IgG) 30 min on i ce . Heat k i l l e d , formalin f ixed Staphvlococcus aureus c e l l s were added and incubated a further 30 min on i c e . The mixture was c leared by cent r i fugat ion at 12,000 X g for 5 min. Precleared l a b e l l e d proteins were incubated 4 hours on ice with s p e c i f i c antiserum. S.aureus c e l l s were added and the mixture incubated 30 min more on i ce . The c e l l s were pe l le ted at 12,000 X g and washed repeatedly with 1 % Tr i ton X-100, 0.5 % BSA in PBS. F i n a l l y , the c e l l s were resuspended in s o l u b i l i z a t i o n mix and bo i led for 2 min. Insoluble matter was c leared by cent r i fugat ion and the supernatant was analyzed by SDS-PAGE. Label led proteins were detected by autoradiography. IX Maxicel l Labe l l i ng Plasm id-encoded proteins were labe l led with [ 3 S S1-methionine in E . c o l i CSR603 (Sancar et a l . , 1979). Ce l l s were grown in M9 medium with a m p i c i l l i n (50 pg ml - 1 ) to 10 a 39 c e l l s m l - 1 . 10 ml were i r rad ia ted for 15 sec in a p e t r i d ish from a distance of 43 cm with a model UVS-54 mineral ight lamp (short wave, 254 nm). C e l l s were treated with 200 pg m l - 1 cyc loser ine and l a b e l l e d with 25 pCi m l - 1 [ 3 = S]-methionine O800 Ci mmole- 1 ) . C e l l s were pe l le ted at 12,000 X g , resuspended in s o l u b i l i z a t i o n mix, and analyzed by SDS-PAGE. Label led proteins were detected by autoradiography. X Ol igonucleot ide P u r i f i c a t i o n and Dot Blot Hybr id izat ion The ol igodeoxyr ibonucleot ides 5'-TGGAAGCTC/GGACTGGACNTAC-3' were chemical ly synthesized as a mixture of 8 on an Appl ied Biosystems 380A DNA synthesizer by T. Atkinson using phosphite t r i e s t e r chemistry (Adams et a l . , 1983; Atkinson and Smith, 1984). The mixture was p u r i f i e d by PAGE on a 20 % acrylamide, 7 M urea sequencing g e l , followed by reversed phase chromatography on a C18 SEP-PAK car t r idge (Atkinson and Smith, 1984). 20 pmoles of o l igonucleot ide were end - labe l led with T - [ 3 2 P ] A T P (1000-3000 Ci mmole-1) using 5.5 units of T4 polynucleot ide kinase (Zo l le r and Smith, 1983). Plasmid DNA (described in Appendix A) was a lka l i - denatu red (Maniatus et a l . , 1983) and 0.1 pmole was spotted onto n i t r o c e l l u l o s e . The blot was incubated with 106 cpm (Cerenkov) of l abe l led o l igonucleot ide for 1 h at 24°C. Fol lowing h y b r i d i z a t i o n , the blot was washed in 6X SSC at increasing temperatures and autoradiographed between washes (Zo l le r and Smith, 1983). 40 XI Native Ce l lu lase P u r i f i c a t i o n The p u r i f i c a t i o n of substrate-bound native c e l l u l a s e s is shown schematical ly in F ig 4. 20 l i t r e cu l tures were grown on 3% Av ice l for 6 days. Cel lu lose-bound c e l l u l a s e s were eluted from Avicel as descr ibed (Gi lkes et a l . , 1984b) with 1.5 l i t r e s of 8 M guanidine-HCl (Gdn-HCl) in T buf fe r . The eluate was d i l u t e d to 0.2 M Gdn-HCl and concentrated to approximately 200 ml by u l t r a f i l t r a t i o n through Amicon PM10 membranes. The ent i re eluate was pumped onto a Concanavalin A-Sepharose 4B column (20 cm X 1.5 cm) at 1 ml m i n - 1 . Unbound proteins were washed from the column with TN buffer contain ing 100 pM each of MgCl 2 , C a C l s : , and M n C l a as recommended by the manufacturer. The s a l t concentrat ion was reduced by washing the column with T buffer contain ing 100 pM each of M g C l z , C a C l a , and M n C l a . Bound proteins were eluted with a 2 step gradient , f i r s t with 50 mM then with 100 mM a-methyl -D-g lucoside (Me-Glc) in T bu f fe r . The 50 mM Me-Glc f rac t ions containing exoglucanase (Exg) a c t i v i t y were pumped onto a Pharmacia Mono Q HR 5/5 column at 2 ml min - 1 using an FPLC system. Bound proteins were eluted at 1 ml min - 1 with a 0-1 M NaCl gradient in 20 mM T r i s , pH 8 . 3 . The 100 mM Me-Glc column f rac t ions contain ing endoglucanase (EngA) a c t i v i t y were pooled and the pH was adjusted to 9.8 with 1 M piperazine p r io r to loading onto the Mono Q column ( f i n a l piperazine concentrat ion was approximately 20 mM). Bound proteins were eluted with a 0-1 M NaCl gradient in 20 mM p iperaz ine . Mono Q f rac t ions containing c e l l u l a s e a c t i v i t y were pooled and desalted into HPLC grade H20 over a Biogel DG-P6 column (30 cm X 1.0 cm). Samples were then buffered by the add i t ion of the appropriate concentrated buffer and stored at 4°C, or were l y o p h i l i z e d to dryness and stored at -20^0. EngA was further resolved into 2 components, EngAl and EngA2, by preparative non-denaturing PAGE. EngAl and EngA2 were eluted from the g e l , desa l ted , and l y o p h i l i z e d . U l t r a f i l t r a t i o n , Con A chromatography, and gel band e lu t ions were performed at 4°C. A l l other p u r i f i c a t i o n steps were done at 22°C. 41 Air . Exhaust transfer culture \ Centrifuge Avicel 5 pin ^ 200 X g Culture on 31 Avicel, 30*C,6 days 1n stirred fermentor. Wash Avicel 6X by resuspendlng In T buffer (10 1), let Avicel settle, decant off buffer. Wash pellets 3X with 8H Gdn-HCl (100 •!). Keep Gdn-HCl eluate on Ice. Filter Gdn-HCl eluate through glass fiber f i lter " by suction FPIC HonoQ chromatography: Elutlon conditions for PNPCase fractions; A-20 mH Tris pH8.3; B-l H NaCl In A; CMCase fractions: A«20 mH Ptperazlne pH9.8; B-l H NaCl In A; elutc at 1 ml m1n Pool fractions with activity. Filter. ConA-Sepharose chromatography! Pump Antcon retentate onto column at 1 ml mln-1. Collect 10 ml fractions. Assay for protein, PNPCase, and CMCase. Ultrafiltration In stirred Antcon cell through PH10 membrane. Concentrate. Dilute 10X. Concentrate. Dilute 4X. Concentrate to 200 ml. F ig 4. Schematic presentat ion of the procedure for p u r i f y i n g ce l lu lose -bound EngA and Exg. 1. p e r i s t a l t i c pump; 2. f r a c t i o n c o l l e c t o r ; 3. buf fer mixing chamber; 4. Mono Q column; 5. U.V. monitor (280 nm). 42 XII Amino Terminal Amino Acid Sequence Determinations Sequencing by automated Edman degradation was performed at the T r i p a r t i t e Microsequencing F a c i l i t y , V i c t o r i a , Canada. Lyophi l i zed prote in was d isso lved in water and 2 nmoles were appl ied to an Applied Biosystems gas phase sequenator model 470A. Exoglucanase, FPLC p u r i f i e d and desa l ted , was sequenced twice for 24 cyc les with a 96% r e p e t i t i v e y i e l d (R. O lefson, personal communication). Endoglucanase f r a c t i o n EngAl was sequenced once for 45 cyc les and f r a c t i o n EngA2 was sequenced twice for 36 cyc les with a 96 % r e p e t i t i v e y i e l d . Amino acids were detected as phenylthiohydantion der i va t i ves as described (Hunkapi l lar , et a l . , 1983). XIII Amino Acid Composition Analys is Lyophi l i zed prote in was d isso lved in 0.3% TFA at 1 mg m l - 1 . 50 pg were a l iquoted into 4 tubes and l y o p h i l i z e d . Three samples were red isso lved in 6 N constant b o i l i n g HC1 (Pierce) and hydrolyzed for 24, 48, and 96 hours at 100°C in evacuated tubes. The fourth sample was oxidized with performic ac id p r i o r to hydro lys is in HC1 for 24 hours. Amino acids were analyzed by standard procedures on a Dionex D500 automatic amino ac id analyser in the Department of Biochemistry, UBC. XIV Carbohydrate Analys is P u r i f i e d and l y o p h i l i z e d c e l l u l a s e was hydrolyzed in 4M TFA (Pierce) (Neeser and Schweizer, 1984) in 0.3 ml s t i r r e d r e a c t i v i a l s (Pierce) at 110°C. Myoinositol was added as the internal standard. Hydrolysates were t ransfered to 15 ml corex con ica l centr i fuge tubes and l ibe ra ted monosaccharides were subsequently der i va t i zed to the i r a l d i t o l acetates (Park, 1984; Niedermeier, 1971* Lehnhardt and Winzler , 1968). Drying steps were done by rotarevaporat ion at 37°C. After a c e t y l a t i o n , the f i n a l residue was dr ied j_n. vacuo over f resh NaOH p e l l e t s and s u l f u r i c ac id for a minimum of 16 hours. A l l react ions were c a r r i e d out in chromic ac id washed glassware. Samples were 43 d isso lved in r e d i s t i l l e d chloroform and analysed by a Hewlett Packard 5890 GLC on a 15m Durabond 17 c a p i l l a r y column. Chromatography was c a r r i e d out with a thermal gradient beginning at 180°C for 2 min, increasing 5°C per min to 220°C, then holding at 220°C for 20 min. Nitrogen was the c a r r i e r gas. Peak areas were recorded with a Hewlett Packard 3392A integrater . Sugars were i d e n t i f i e d by comparing re tent ion times with known standards. I den t i f i ca t ions were corroborated by mass spectroscopy ana lys is at the mass spectroscopy f a c i l i t y , Department of Chemistry, UBC. XV Substrate Binding Assays Native enzymes p u r i f i e d from substrate were tested for the i r a b i l i t y to bind back to A v i c e l . Recombinant enzymes p u r i f i e d from E . c o l i (Arfman, 1986) by immunoadsorption column chromatography followed by Mono Q anion exchange chromatography, were tested for the i r a b i l i t y to bind to A v i c e l . For each time po int , 20 mg of Av ice l were weighed into 1.5 ml eppendorf tubes. 0.5 ml of d i s t i l l e d water was added and the tubes were autoclaved. The Av ice l was p e l l e t e d by cent r i fugat ion for 2 min at 12,000 X g . Soluble reducing groups were removed from the Av ice l p e l l e t by washing 3 times with d i s t i l l e d H 20, followed by an e q u i l i b r a t i o n wash in C bu f fe r . The Av ice l p e l l e t s were c h i l l e d on i c e . 200 pi of enzyme d i l u t e d in ice co ld C buf fer were added to each of the Av ice l p e l l e t s , vortexed b r i e f l y , then incubated on i c e . Samples incubated longer than 10 min were mixed p e r i o d i c a l l y to resuspend the A v i c e l . Reactions were terminated by cent r i fugat ion for 2 min at 12,000 X g . Supernatants were t ransfered to c lean 1.5 ml eppendorf tubes and kept on i ce . The Av ice l p e l l e t s were washed once with 1 ml ice cold C buffer and the wash was t ransfered to a c lean tube and kept on i ce . The p e l l e t s were saved on i ce . Af ter a l l time points were completed, the react ion supernatants, the Av ice l washes, and the Av ice l p e l l e t s were assayed for c e l l u l a s e a c t i v i t y . To 44 assay the p e l l e t s for pNPCase, 0.9 ml of P buf fer and 50 pi of 10 mM pNPC were added to the Av ice l and incubated at 37°C. The react ion was stopped by the addi t ion of 0.5 ml of 1 M sodium carbonate. Av ice l was removed by c e n t r i f u g a t i o n . The c l a r i f i e d supernatant was transfered to a c lean tube and the absorbance was measured at 400 nm. P e l l e t s were assayed for CMCase a c t i v i t y by resuspending the Av ice l in 0.2 % CMC in C buffer and incubating at 30°C. The react ion was stopped by p e l l e t i n g the Av ice l and t rans fe r r ing 0.5 ml to 1.0 ml HBAH reagent. The reagent blank was obtained from the Av ice l binding contro l where Av ice l was incubated with react ion buffer in the absence of enzyme. The a c t i v i t y was expressed as a percentage of the contro l where c e l l u l a s e was incubated on ice in the absence of A v i c e l . XVI Preparat ion of Recombinant Ce l lu lases for Protease Digests E . c o l i JM101 containing the plasmid pUCEC2 and E . c o l i PM191 containing the plasmid pUC12: :PTIS: :Sty or pUC12 were grown in L broth without glucose at 37°C. The cu l tures were made 2 mM IPTG when they reached an op t i ca l density of 0.35 at 600 nm. Af ter 24 hours of growth, the supernatants were c leared by cent r i fugat ion for 15 min at 12,000 X g , followed by f i l t r a t i o n through 0.2 pm Gelman polysulfone ac rod iscs . NaNa was added to 0.02 %. The supernatants were then added to autoclaved and washed Av ice l to give a f i n a l Av ice l concentrat ion of 10 % <w/v). The s lu r r y was incubated on ice for 1 hour with per iod ic mixing. Av ice l was pe l le ted by cent r i fugat ion and washed with TN bu f fe r , followed by 1 M NaCl in 20 mM T r i s - H C l , pH 7 .4 . No c e l l u l a s e a c t i v i t y was detected in the washes. Enzyme a c t i v i t y in the cul ture supernatant was measured before and a f te r Av ice l treatment and the bound a c t i v i t y was determined by subt rac t ion . 45 XVII Preparation of Native Ce l lu lases for Protease Digests FPLC p u r i f i e d substrate-bound c e l l u l a s e s were further f ract ionated by preparative nondenaturing PAGE to el iminate the p o s s i b i l i t y of cu l tu re -der i ved p r o t e o l y t i c products c o - p u r i f y i n g with the intact enzymes. Bands of a c t i v i t y were excised from the gel and e luted with P bu f fe r . Cel lu lose-bound c e l l u l a s e was obtained from a C . f imi cul ture grown for 6 days on 3 % A v i c e l . The Av ice l was washed 6 times with TN buffer to remove res idual c e l l s and unbound pro te ins . Exoglucanase a c t i v i t y bound to the Av ice l was estimated by the pNPCase assay descr ibed in XIV. XVIII Preparat ion of Protease C . f imi was grown for 3 days on basal medium supplemented with 0.1 % g l y c e r o l . The cul ture supernatant was c leared of c e l l s by c e n t r i f u g a t i o n , followed by f i l t r a t i o n through 0.2 pm polysulfone ac rod iscs . NaNa was added to 0.02 %. PMSF was added to a port ion to 50 pg m l - 1 . The r e s u l t i n g protease preparat ion was refered to as g lycero l cul ture supernatant. Protease a c t i v i t y was measured on Hide Powder Azure and was stable for several weeks at 4°C. XIX Protease Digests of So lub le , Native Ce l lu lases C . f imi EngAl and Exg a c t i v i t i e s e luted from a preparative non-denaturing polyacrylamide gel were each a l iquoted into 5, 1.5 ml eppendorf microfuge tubes and l a b e l l e d A to E. Glycerol cu l ture supernatant was added to tube B to y i e l d a ee l lu lase :p ro tease r a t i o of 1.5 HBAH units CMCase or 10 units pNPCase to 16 units collagenase equiva lents . An equivalent volume of the fo l lowing was added to the remaining tubes: to C: g lycero l supernatant contain ing PMSF; to D: TN buf fe r ; to E: TN buffer contain ing PMSF. Tube A was stored at 4°C for the incubation per iod , then an equivalent volume of g lycero l cul ture supernatant containing PMSF was added. Tubes B-E were 46 sealed and incubated at 30°C on a tube r o l l e r for 3 days. A l l react ions were stopped by c h i l l i n g on ice and adding PMSF to B and D. The react ions were assayed for c e l l u l a s e a c t i v i t y as described in V. Samples were prepared for SDS-PAGE ana lys is by adding a o n e - f i f t h volume of 2X s o l u b i l i z a t i o n mix and b o i l i n g for 2 min. XX Protease Digests of Substrate-bound Ce l lu lases Av ice l suspensions contain ing equivalent amounts of c e l l u l a s e a c t i v i t y were a l iquoted into 5, 1.5 ml eppendorf tubes, l a b e l l e d A to E. The Av ice l in tubes B-E was pe l le ted by cent r i fugat ion at 12,000 X g for 2 min. The p e l l e t s were resuspended in 1 ml of B: g lycero l cul ture supernatant; C: g lycero l cu l ture supernatant contain ing PMSF; D: TN buf fe r ; E: TN buffer contain ing PMSF. Tubes B to E were sealed and incubated at 30°C on a tube r o l l e r for 3 days. Tube A was stored at 4°C. The react ions were stopped by the addi t ion of PMSF to tubes B and D, and c h i l l i n g a l l tubes on i ce . Tube A was centr i fuged and the Av ice l was resuspended in 1 ml g lycero l cul ture supernatant containing PMSF. A l l tubes were centr i fuged again and the supernatants were t ransfered to clean tubes and assayed for c e l l u l a s e a c t i v i t y . A l l samples were prepared for SDS-PAGE a n a l y s i s : Av ice l p e l l e t s were washed once with TN bu f fe r , resuspended in s o l u b i l i z a t i o n mix, bo i led for 2 min then centr i fuged to p e l l e t the insoluble mater ia l . Reaction supernatants were mixed with a o n e - f i f t h volume of 2X s o l u b i l i z a t i o n mix and bo i led 2 min. 47 RESULTS I The M u l t i p l i c i t y of A c t i v i t i e s in the Ce l lu lase System I n i t i a l attempts were made to separate the c e l l u l a s e s in C . f imi cul ture supernatants by non-denaturing polyacrylamide gel e lect rophores is (PAGE). Proteins that could be stained with Coomassie Blue were cor re la ted with enzymatic a c t i v i t y by e l u t i n g gel s l i c e s and then assaying the eluates for CMCase (F ig 3) . This technique demonstrated that C . f imi cul ture supernatants contained a m u l t i p l i c i t y of c e l l u l a s e a c t i v i t i e s . An add i t iona l leve l of complexity was observed by two-dimensional PAGE. By th is technique, i t was shown that Coomassie s ta in ing bands in non-denaturing gels were themselves composed of several proteins (F ig 5 ) . Moreover, i t was determined that some of the c e l l u l a s e s were g l ycosy la ted . They could be stained with the p e r i o d a t e - S c h i f f ' s carbohydrate s t a i n , and they could be reta ined on a Concanavalin A Sepharose 4B column (Langsford et a l . , 1984 and F ig 6, panels A -C ) . Subsequent experiments provided support for the hypothesis that many of the c e l l u l a s e a c t i v i t i e s were re la ted and not a l l represented unique gene products. The to ta l amount of c e l l u l a s e a c t i v i t y in cul ture supernatants increased with cul ture age and the a c t i v i t y p r o f i l e s (F ig 3) changed with the age of the c u l t u r e . CMCase a c t i v i t i e s in older cu l tures had increased m o b i l i t i e s on non-denaturing PAGE (Langsford et a l . , 1984). A s i m i l a r observation was made with cul ture supernatants which were stored. The to ta l a c t i v i t y d id not change (Table IV) but with prolonged storage at 4°C, proteins developed a faster mobi l i t y . The addi t ion of the protease inh ib i to r phenylmethylsulfonyl f luor ide (PMSF) prevented some of these changes (Fig 7) . Indiv idual monoclonal ant ibodies ra ised against C . f imi c e l l u l a s e s in cul ture supernatant reacted with the proteins eluted from several d i f f e r e n t gel bands (Langsford, et a l . , 1984). Several proteins in supernatants from cul tures grown for 3 and 9 days were bound by an 48 Fig 5 . Two dimensional PAGE of a 6 day cu l ture supernatant. Supernatant obtained from a culture of C.f i m i grown on 0 . 1 % Avicel for 6 days was f ract ionated in the f i r s t dimension by non-denaturing PAGE. A lane was exc ised , bo i led for 5 minutes in s o l u b i l i z a t i o n mix, then electrophoresed into a 10 %, SDS-containing acrylamide gel for the second dimension. Numbers indicate Mr standards X 1 0 - 3 . 49 F ig 6 . F rac t ionat ion of cul ture supernatant and c e l l u l o s e -bound proteins by Con A-Sepharose column chromatography. Supernatant was prepared from a 0 . 1 % Av ice l cu l ture grown for 3 days. Av ice l from a culture grown for 6 days on 3 % Av ice l was e luted with Gdn-HCl to provide ce l lu lose -bound c e l l u l a s e s (Gdn-HCl e l u a t e ) . Samples were appl ied to a Con A-Sepharose column in TN bu f fe r . The column was washed with TN b u f f e r . Lectin-bound proteins were eluted with TN buffer conta in ing 1 0 0 mM oc -methyl-D-glucos ide (Me-g lc ) . TN and Me-glc column f rac t ions were analysed in dupl icate by non-denaturing PAGE. Lane 1 was stained with Coomassie Blue and lane 2 was stained with fuchsin s u l p h i t e . Panel A: to ta l supernatant p r o t e i n ; B: TN f r a c t i o n of supernatant p ro te in ; C: Me-glc f r a c t i o n of supernatant p r o t e i n ; D: to ta l Gdn-HCl e lua te ; E: TN f rac t ion of Gdn-HCl e luate ; F: Me-glc f rac t ion of Gdn-HCl e l u a t e . 50 Table IV. A c t i v i t y in stored supernatants Cu1ture Supernatant PMSF Day 3 Day 3 Day 6 Day 6 Day 9 Day 9 Act i v i ty <U ml - 1 ) 0.28 0.24 0. 28 0.29 0. 34 0. 30 Supernatants from cul tures grown on 0.1 % Avice l for 3, 6, and 9 days were a l iquoted into 2 tubes. To one a l iquot (+), PMSF was added to 50 pg m l - 1 . The other a l iquot contained no PMSF <-). A l l samples were stored at 4°C and analyzed several weeks l a t e r for CMCase a c t i v i t y . A c t i v i t y was expressed as pmoles glucose equivalents produced min - 1 m l - 1 . 51 A B C D E F Fig 7. Non-denaturing PAGE analys is of supernatants stored with or without PMSF. Samples described in the legend to Table 1 were analysed by non-denaturing PAGE. Lanes A, B, and C: 3, 6, and 9 day supernatant, respec t i ve l y , stored with PMSF. Lanes D, E, and F: 3, 6, and 9 day supernatants, r e s p e c t i v e l y , stored without PMSF. 52 immune-absorbent column which was made by c r o s s - l i n k i n g monoclonal antibody A2/36.17.2 to Prote in A-Sepharose 4B (F ig 8) . When C . f imi cu l tures were 0.5 % Avice l or more, the a c t i v i t y detectable in cul ture supernatants decreased. In cu l tures made 3% in A v i c e l , no c e l l u l a s e a c t i v i t y was found in the supernatant (Langsford et a l . , 1984). Ce l lu lase a c t i v i t y could be eluted from the Av ice l recovered from c u l t u r e s . The a c t i v i t y p r o f i l e of ce l lu lose -bound c e l l u l a s e was much simpler in comparison to the supernatant p r o f i l e s , having only two major zones of a c t i v i t y (Langsford et a l . , 1984). The ce l lu lose -bound c e l l u l a s e s were named EngA, an endoglucanase, and Exg, an exoglucanase (formerly CBI and CBII, respec t i ve l y , in Gi lkes et a l . , 1984). They corresponded in mobi l i ty to the major a c t i v i t i e s present in 3 day cul ture supernatants and were g lycoproteins (F ig 6, panels D-F) . F i n a l l y , a c t i v i t y in supernatants of cu l tures grown for 3 days, and to a much lesser degree of cu l tures grown for 9 days, could be absorbed by newly added Av ice l (Langsford et a l . , 1984). The fo l lowing hypothesis was formulated: the c e l l u l a s e s secreted by C . f imi are the products of only a few, for example, 3 or 4, genes. A l l of the a c t i v i t y secreted binds to c e l l u l o s e u n t i l the substrate is saturated. Then a c t i v i t i e s appear in the cul ture supernatant. Ce l lu lases not bound to c e l l u l o s e are suscept ib le to p r o t e o l y t i c degradation. This p a r t i a l p ro teo lys is reduces the a b i l i t y of the c e l l u l a s e s to bind to substrate . Proteo lys is and d i f f e r e n t i a l g l ycosy la t ion contr ibute to the apparent m u l t i p l i c i t y of c e l l u l a s e a c t i v i t i e s in C . f imi cul ture supernatants. 53 F ig 8. Immunoadsorption of culture supernatant polypept ides . Polypeptides in supernatants prepared from cu l tures grown on 0.1% Avice l were bound by an immunoadsorbent column prepared with monoclonal antibody A2/36.17.2 (see MATERIALS AND METHODS). The f rac t ions were analyzed by SDS-PAGE. Panel 1: 3 day cul ture supernatant; panel 2: 9 day cul ture supernatant. Lane A: Mr standards indicated by the numbers on the l e f t ( X 1 0 - 3 ) ; lane B: to ta l supernatant; lanes C and D: PBS f r a c t i o n s ; lanes E and F: NH-»OH eluted f r a c t i o n s . 54 II P u r i f i c a t i o n and Character i zat ion of Ce l lu lose -bound, Native Ce l lu lases A. P u r i f i c a t i o n of Ce l lu lose -bound, Native Ce l lu lases It was necessary to pur i f y nat ive , intact c e l l u l a s e s from C. f imi to character ize the enzyme system and to ident i f y cloned gene products. The a c t i v i t y p r o f i l e of the substrate-bound enzymes was r e l a t i v e l y s imple. In a d d i t i o n , i f our hypothesis regarding p r o t e o l y t i c degradation is co r rec t , the bound enzymes should represent the in tac t , secreted forms of the c e l l u l a s e s . The p u r i f i c a t i o n of the bound c e l l u l a s e s was undertaken. An out l ine of the p u r i f i c a t i o n scheme is given in F ig 4; further d e t a i l s of the p u r i f i c a t i o n steps are given in Mater ia ls and Methods. The f i r s t step in the enzyme p u r i f i c a t i o n was to recover Av ice l from cul tures and elute the enzymes from i t . It was reported prev iously (Beguin and E i s e n , 1978) that 8 M guanidine-HCl (Gdn-HCl) re leased ce l lu lose -bound c e l l u l a s e s . C . f imi c e l l u l a s e s could a lso be extracted from Avice l with 8 M Gdn-HCl (Gi lkes et a l . , 1984b). This was confirmed by washing Av ice l with increasing Gdn-HCl concentrat ion from 0-8 M (Table V). Residual a c t i v i t y was determined by assaying the c e l l u l o s e for exoglucanase a c t i v i t y . E lu t ion with the non - ion ic detergents T r i ton X-100, Tween 20, Tween 80, and NP-40 as well as t r iethylamine buffers was also tested (data not shown) but none proved as e f f e c t i v e as 8 M Gdn-HCl. Subsequently, 8 M Gdn-HCl was used for the recovery of c e l l u l a s e s from Avice l in a l l ce l lu lose -bound enzyme preparat i ons. In the second step of the p u r i f i c a t i o n , the a f f i n i t y of the c e l l u l a s e s for Con A was u t i l i z e d . By e lu t ing with a step gradient of oc-methyl-D-gl ucos ide in low ionic strength buf fe r , i t was possib le to separate exoglucanase from endoglucanase a c t i v i t y (F ig 9) . The prote in which d id not bind to Con A 55 Table V. E lu t ion of c e l l u l a s e a c t i v i t y from Avice l with Guan id ine -HC l . Guanidine-HCl A c t i v i t y Remaining Concentration on Av ice l CM) (%) 0 100 2 70 4 1 1 6 2 8 1 Av ice l was iso lated from a cul ture grown for 6 days on 3 % A v i c e l . The Av ice l was washed 4 times with T buffer containing Gdn-HCl at the s p e c i f i e d concentrat ion , 2 times with T buffer a lone, then assayed for pNPCase a c t i v i t y as descr ibed in MATERIALS AND METHODS, sect ion XV. A c t i v i t y remaining on the Av ice l was expressed as a percent of the a c t i v i t y when the Av ice l was washed only with T bu f fe r . 56 F i g 9. Preparative f r a c t i o n a t i o n of ce l lu lose -bound c e l l u l a s e s by Con A-Sepharose chromatography. Av ice l was i so la ted f roa 20 l i t r e s of culture a f t e r 6 days of growth on 3 % A v i c e l . The prote ins which were bound to the Av ice l were eluted with 8 M Gdn-HCl . The Gdn-HCl e l u a t e , d i l u t e d to 0.2M, was passed over a Con A-Sepharose 4B column. 10 ml f rac t ions were c o l l e c t e d and assayed. So l id l i n e : p r o t e i n ; dashed l i n e : CMCase a c t i v i t y expressed as pmoles glucose equivalents produced min - 1 f r a c t i o n - 1 ; dotted l i n e : pNPCase a c t i v i t y expressed as pmoles pNP produced min- 1 f r a c t i o n - 1 . Arrows indicate where the buf fe r condit ions changed: A: TN b u f f e r ; B: T bu f fe r ; C: T buffer contain ing 50 mM Me-Glc ; D: T buf fer conta in ing 100 mM Me-Glc. Peak I: exoglucanase a c t i v i t y ; peak II: endoglucanase a c t i v i t y . 57 consisted mainly of a yellow pigmented substance which did not enter non-denaturing gels and migrated as a high molecular weight substance on SDS-containing gels (F ig 11, lane B) . Anion exchange chromatography using a quaternary amine exchange res in (Mono Q) was the th i rd p u r i f i c a t i o n step. This step served to concentrate the enzymes into a small volume, usual ly 1.5-2 ml, and to remove minor contaminants (F ig 10). The p u r i f i c a t i o n data are summarized in Tables VI and VII. The data in these tables summarize the best of 6 EngA p u r i f i c a t i o n s and 10 Exg p u r i f i c a t i o n s . 31% of the CMCase a c t i v i t y in the Gdn-HCl eluate was recovered in the EngA f rac t ion with a 18- fo ld increase in s p e c i f i c a c t i v i t y . 50% of the pNPCase a c t i v i t y was recovered in the Exg f r a c t i o n with a 23 - fo ld increase in s p e c i f i c a c t i v i t y . B. Character i zat ion of Ce l lu lose -bound, Native Ce l lu lases The p u r i f i e d endoglucanase (EngA) and exoglucanase (Exg) were analysed by various types of PAGE. SDS-PAGE indicated that Exg and EngA were p u r i f i e d to apparent homogeneity as determined by Coomassie s ta in ing (Fig 11). The M s^ were 57,000 for EngA and 56,000 for Exg, as estimated by the i r e lec t rophoret i c mob i l i t i es r e l a t i v e to proteins of known molecular weight (See Appendix B for c a l i b r a t i o n curve) . The M r s could not be determined p rec i se l y with the standards used to c a l i b r a t e the g e l s . The M K values reported prev iously were 58,000 and 56,000 for EngA and Exg, respect i ve ly (Gi lkes et a l . , 1984b). EngA has since been shown to comigrate with a 53,000 marker and Exg to migrate marginally faster (N .G i lkes , unpublished observat ion) . Non-denaturing PAGE resolved EngA into 2 components, EngAl and EngA2 (F ig 12). EngA2 migrated s l i g h t l y more slowly than Exg, which migrated as a s ingle band. I soe lec t r i c focuss ing (IEF) gels gave further evidence of heterogeneity . In native 58 F i g 10. FPLC anion exchange column chromatography of a c t i v i t y peaks I and II recovered from Con A-Sepharose. Me-Glc f rac t ions containing exoglucanase a c t i v i t y and endoglucanase a c t i v i t y (F ig 9, peaks I and II, r espec t i ve l y ) were pooled for anion exchange chromatography by FPLC over a Mono Q column. Peak I was loaded onto the Mono Q column d i r e c t l y and e lu ted with a gradient of 0 to 1 M NaCl in 20 mM T r i s , pH 8.3 (Panel A) . Peak II was adjusted to pH 9.8 with 1 M piperazine base then loaded onto the Mono Q column and eluted with a gradient of 0 to 1 M NaCl in 20 mM p i p e r a z i n e , pH 9.8 (Panel B) . The s o l i d l i nes represent the absorbance at 280 nm; the dotted l i n e s indicate Buffer B concent ra t ion . The bars indicate f rac t ions which had c e l l u l a s e a c t i v i t y . 59 Table VI. P u r i f i c a t i o n of the native endoglucanase from C . f i m i . Sample Speci f ic Act i v i ty (U mg- 1) Total Units CMCase Y ie ld P u r i f i c a t i o n (%) Degree Guan id ine Eluate 1 1 2000 100 Con A Seph, pooled f ract i ons 90 1238 62 Mono Q peak f rac t i ons 200 614 31 18 EngA was p u r i f i e d a to ta l are from the best overa l l expressed as units CMCase, prote in measured by Lowry in step 3. of 6 t imes. The data in th i s table p u r i f i c a t i o n . S p e c i f i c a c t i v i t y is using the DNS reagent, per mg in steps 1 and 2, and by dry weight 60 Table VII. P u r i f i c a t i o n of the native exoglucanase from C . f i m i . Sample S p e c i f i c Total Y ie ld P u r i f i c a t i o n A c t i v i t y Units <%) Degree <U mg- 1) pNPCase Guan idine Eluate 0.18 26 100 1 Con A Seph. pooled f rac t ions 1.2 15 58 6 Mono Q peak f ract ions 4.1 13 50 23 Exg was p u r i f i e d a to ta l of 10 t imes. The data in th i s table are from the best overa l l p u r i f i c a t i o n . S p e c i f i c a c t i v i t y is expressed as uni ts pNPCase per mg prote in as measured by Lowry. 61 A B C D E F G Fig 11. SDS-PAGE ana lys is of p u r i f i e d c e l l u l a s e s . Samples from a l l steps in the p u r i f i c a t i o n procedure were electrophoresed in an 8-12 % gradient mini gel (approximate dimensions are shown in the f i g u r e ) . Lane A: to ta l Gdn-HCl e luate ; lane B: Con A column unbound f r a c t i o n ; lane C: Con A column peak I; Lane D: Mono Q column peak Exg; lane E: Con A column peak II; lane F: Mono Q column peak EngA; lane G: Mr standards (numbers indicate Mr X 1 0 - 3 ) . 62 A B [: tHt mm F ig 12. Non-denaturing PAGE analys is of p u r i f i e d c e l l u l a s e s . Mono Q peak c e l l u l a s e f ract ions were analysed by non-denaturing PAGE. Lane A: EngA; lane B: Exg. Arrows 1 and 2 indicate EngAl and EngA2. 6 3 IEF g e l s , EngA focussed, though incompletely, as a s ingle band at pH 8 . 2 , while Exg focussed sharply into 2 bands at pH 5 . 8 - 5 . 9 (F ig 13, lanes A -C ) . Hydrolysis zones in an overlay of a native IEF gel were coincident with the Coomassie s ta in r e s u l t s but indicated an add i t iona l exoglucanase a c t i v i t y band (F ig 13, lanes D-F) . In denaturing IEF g e l s , EngA again resolved into 2 bands between pH 7 -7 .3 and Exg showed a strong band at pH 6.2 with minor bands above and below (F ig 14). P u r i f i e d EngAl, EngA2 and Exg were subjected to amino terminal amino ac id sequence analys is (F ig 15 and 16). Both EngAl and EngA2 gave ident ica l sequences except at residues 32 and 33 where threonine was re leased from EngAl while serine was released from EngA2. From the same preparat ion , an a l iquot of EngAl and EngA2 were each hydrolysed for 24 hours for an amino ac id composition a n a l y s i s . Both gave ident ica l compositions within the er rors of the ana lys is (data not shown). It was concluded that these were the same enzyme, perhaps with very minor d i f fe rences derived p o s t - t r a n s l a t i o n a l 1 y or during the p u r i f i c a t i o n . Therefore, a l l subsequent analyses were performed with FPLC -pur i f ied endoglucanase (EngA) without further f rac t ionat ion by non-denaturing PAGE. A comprehensive amino ac id composition ana lys is of EngA and Exg is shown in Tables VIII and IX. A carbohydrate ana lys is was performed on EngA and Exg to determine the monosaccharide compositions of the g lycosy l moieties of these enzymes (Table X). Mannose was the major monosaccharide detected, while traces of arabinose, xy lose, galactose and glucose were also observed (see Appendix D for chromatography traces and mass spect ra ) . However, the amounts of these trace sugars var ied from one analys is to another and in each case were too small to ca lcu la te accurate ly . The trace sugars may have been contaminants from a i r -borne dust . EngA was found to be approximately 10% mannose by weight. Exg had s l i g h t l y l e s s ; i t was about 8% mannose. 6 4 A B C D E F Fig 13. I s o e l e c t r i c focussing analys is of p u r i f i e d c e l l u l a s e s in non-urea g e l s . Samples were focussed in a s lab gel for 80 minutes at 20 watts constant power. The IEF gel was stained for prote in with Coomassie blue (Lanes A -C ) . Endoglucanase a c t i v i t y was located by incubating the IEF gel with a CMC-agarose over lay . Zones of c l e a r i n g were detected a f t e r s ta in ing the agarose overlay with Congo Red (Lanes D and E ) . Exoglucanase a c t i v i t y was located by incubating the IEF gel in 1 mM MUC in P bu f fe r . Bands of f luorescence were detected by UV l i g h t and photographed. The pH gradient was determined from standard markers electrophoresed in the g e l . Lanes A and D: total Gdn-HCl e lua te ; lanes B and E: EngA; lanes C and F: Exg. 65 A B C Fig 14. I s o e l e c t r i c focussing analys is of p u r i f i e d c e l l u l a s e s in urea g e l s . Samples were focussed in a s lab gel contain ing 4 M urea for 8200 vh. The pH gradient was determined by s l i c i n g an empty lane into 4 mm s l i c e s , e lu t ing the s l i c e s in d i s t i l l e d H 20, and measuring the pH of the water with a micro e lec t rode . Lane A: to ta l Gdn-HCl e luate ; lane B: EngA; lane C: Exg. The numbers on the r ight indicate pH. 66 1 2 3 4 5 6 7 8 Ala Thr Thr Leu Lys Glu Ala Ala 9 10 11 12 13 14 15 16 Asp Gly Ala Gly Arg Asp Phe Gly 17 18 19 20 21 22 23 24 Phe Ala Leu Asp Pro Asn Arg Leu Figure 15. The amino terminal amino ac id sequence of Exg. Exoglucanase p u r i f i e d by anion exchange was desalted and l y o p h i l i z e d . Lyophi l i zed protein was resuspended in H20 and 2 nmoles were sequenced by automated Edman degradation. Amino acids were detected as PTH der i vat i ves and analyzed by HPLC. Exg was sequenced twice for 24 cyc les with 96% r e p e t i t i v e y i e l d <R. Olefson, personal communication). 67 1 2 3 4 5 6 7 8 9 10 EngAl Ala Pro Gly Arg Val Asp Tyr Ala Val EngA2 Ala Pro Gly Arg Val Asp Tyr Ala Val 11 12 13 14 15 16 17 18 19 20 EngAl Thr Asn Gin Trp Pro Gly Gly Phe Gly Ala EngA2 Thr Asn Gin Trp Pro Gly Gly Phe Ala 21 22 23 24 25 26 27 28 29 30 EngAl Asn Val Thr l i e Thr Asn Leu Gly Asp Pro EngA2 Asn Val Thr H e Thr Asn Leu Gly Asp Pro 31 32 33 34 35 36 37 38 39 40 EngAl Val Thr Thr Trp Lys Leu Asp Trp Thr Tyr EngA2 Val Ser Ser Trp Lys Leu 41 42 43 44 45 EngAl Thr Ala Gly Gin H e 01iqonucleot ide Probe Amino Acid 34 35 36 37 38 39 40 Sequence: Trp Lys Leu Asp Trp Thr Tyr Probe Sequence: TGG AAG CTC/G GAC TGG ACN TAC Correspond ing DNA Sequence: TGG AAG CTC GAC TGG ACC TAC F ig 16. Amino terminal amino ac id sequences of EngAl and EngA2, and the sequence of the o l igonucleot ide probe. Endoglucanase p u r i f i e d by anion exchange and non-denaturing PAGE was desalted and l y o p h i l i z e d . 2 nmoles of l y o p h i l i z e d prote in were resuspended in H20 and sequenced by automated Edman degradation. EngAl was sequenced once, EngA2 was sequenced twice; repe t i t i ve y ie lds were 96% (R. Olefson, personal communication). No PTH was detected in cycle 4. A deoxyribonucleotide was synthesized as a mixture of 8 which corresponded to the codons of residues 34 to 40. The 2 Inter was used as a hybr id i za t ion probe to f ind the recombinant plasmid encoding the amino terminus of EngA. The DNA sequence shown below the probe sequence is the actual sequence in the gene that the probe was equivalent to . 68 Table VIII. Amino ac id composition of EngA. Residue Hydrolysis Time (h) 24 48 96 Asx 43 .91 40 .08 41 .00 Thr 43 .64 43 .03 42 .21 Ser 27 .24 26 .01 23 .78 Glx 32 . 18 32 .76 33 . 34 Pro 33 .14 33 .12 32 .69 Gly 53 .59 54 . 42 55 . 25 Ala 48 .02 48 .36 49 .44 Cys B 6 .00 V a l c 24 .40 25 .06 25 . 15 Met 1 .86 2 .64 1 .74 I l e c 9 .66 9 .76 8 .98 Leu c 24 .85 23 .76 20 .94 Tyr 12 . 28 12 .69 10 .41 Phe 8 .93 8 .57 7 .95 Trp D His 2 .84 2 .87 2 .96 Lys 1 1 .92 12 . 47 12 .99 Arg 17 .39 16 .82 17 .30 3 point Assumed DNA Average Integral Sequence 47.7 42 41 44. 0* 44 44 28. 5 A 29 29 32.76 33 31 33.0 33 32 54.4 54 53 48.6 49 45 6 6 25. 1 1 25 28 2.08 2 2 9.71 10 12 24.31 24 25 11.8 12 13 8.5 9 9 15 2.9 3 3 12.5 13 13 17.2 17 16 FPLC -pur i f ied EngA was hydrolyzed as described in MATERIALS AND METHODS for an amino ac id composition a n a l y s i s . Except where ind icated , the assumed integra ls were averaged from the resu l ts of the hydrolyses performed for 24, 48, and 96 hours. The amino ac id composition deduced from the DNA sequence (Wong et a l . , 1986) is shown for comparison. Thr and Ser were extrapolated to zero time, see Appendix C. Cys was determined as c y s t e i c ac id a f t e r performic ac id ox idat i on. V a l , H e , and Leu averages were determined from the highest two values (they should increase with t ime). Trp was not determined. 69 Table IX. Amino ac id composition of Exg. Res idue Hydrolysis Time (h) 3 point Assumed DNA 24 48 96 Average Integral Sequence Asx 53. 30 48.37 48.65 50. 1 50 51 Thr 42.66 42.60 42.36 42. 7* 43 42 Ser 28.73 26.79 23.94 30. 1 A 30 30 Glx 36.78 37.60 37.98 37.45 37 35 Pro 28.36 28.28 28.04 28.2 28 28 Gly 41 .26 41.61 41 .84 41.6 42 39 Ala 59.96 59.77 59.81 59.8 60 57 Cys B 6.24 6 6 V a l c 31 .04 33.75 34.40 34. 1 34 37 Met 5. 25 5.26 4.70 5. 1 5 5 I l e c 7.43 7.77 7.77 7.77 8 9 Leu c 22. 15 21 .98 20.57 22.07 22 22 Tyr 10.12 10.07 9.64 9.9 10 10 Phe 20.87 20.88 20. 28 20.7 21 22 Trp D 1 2 His 5.04 5. 19 5.40 5.2 5 6 Lys 16.35 16.91 17.07 16.8 17 19 Arg 15.92 15 .98 16. 19 16.0 16 13 FPLC -pur i f ied Exg was hydrolyzed as described in MATERIALS AND METHODS for an amino ac id composition a n a l y s i s . Except where ind icated , the assumed integrals were averaged from the resu l t s of the hydrolyses performed for 24, 48, and 96 hours. The amino ac id composition deduced from the DNA sequence (O 'Ne i l l et a l . , 1986a) is shown for comparison. Thr and Ser were extrapolated to zero time, see Appendix C. Cys was determined as c y s t e i c ac id a f te r performic ac id ox idat i on. V a l , H e , and Leu averages were determined from the highest two values (they should increase with t ime). Trp was not determined. 70 Table X. Carbohydrate compositions of EngA and Exg. Ce l lu lase Sugar umoles suqar A % mg (dry wt) (w/w X 100) EngA mannose 0 .541 B 9.7 Exg mannose 0 .425 c 7.6 FPLC -pur i f ied EngA and Exg were hydrolysed as descr ibed in MATERIALS AND METHODS for a carbohydrate composition a n a l y s i s . A hydro lys is time course was done to e s t a b l i s h the condit ions for optimal release and recovery of monosaccharides. The optimal hydro lys is time was 6 hours for EngA and 4 hours for Exg. A Calculated according to Lehnhardt and Winzler , 1968. The mannose response factor = 1 when standard mannose was hydrolyzed in p a r a l l e l with the c e l l u l a s e samples. B Mannose values for Eng were averaged from 6 independent hydrolyses of 0 . 3 - 0 . 4 mg each. Standard dev iat ion = 0.054. c Mannose values for Exg were averaged from 3 independent hydrolyses of 0 . 4 - 0 . 6 mg each. Standard dev iat ion = 0 .07. 71 III I d e n t i f i c a t i o n of the Recombinant Plasmids Expressing the Corresponding Ce l lu lases Several C. f i m1 genes expressing c e l l u l a s e a c t i v i t y in E . c o l i were obtained by shotgun c lon ing Bam HI fragments of C . f imi DNA into the E . c o l i vector pBR322 (Whittle et a l . , 1982). I n i t i a l attempts were made to ident i f y these recombinant gene products by c o r r e l a t i n g the i r Mv-s (Gi lkes et a l . , 1984a) and the i r substrate s p e c i f i c i t i e s (Gi lkes et a l . , 1984b) with C.f i m i enzymes. The M^s of the plasm id-coded proteins were estimated from SDS-PAGE a f te r jji vivo l a b e l l i n g in max i - ce l l s (F ig 17). A unique polypeptide was d i sce rn ib le from each of the plasmids, pECl and pEC2. These unique polypeptides could be immunoprecipitated with ant i -ce11ulase antiserum (F ig 18 and Gi lkes et a l . , 1984a). The pECl and pEC2 gene products were estimated to have MrS of 51,000 and 53,000, respec t i ve l y . The a c t i v i t i e s encoded by these plasmids were exoglucanase (pECl) and endoglucanase (pEC2) (Gi lkes et a l . , 1984b). It was possib le that pECl c a r r i e d the gene encoding the ce l lu lose -bound Exg a c t i v i t y while pEC2 contained the gene encoding the ce l lu lose -bound Eng. Pos i t i ve i d e n t i f i c a t i o n was made from DNA and amino ac id sequence comparisons. The s t ruc tu ra l gene encoding exoglucanase a c t i v i t y on pECl was character ized and sequenced (O 'Ne i l l et a l . , 1986a). Di rect inspect ion of the t rans lated DNA sequence showed that i t encoded the amino terminal amino ac id sequence of the ce l lu lose -bound Exg. The 24 residues obtained from Exg by Edman degradation (F ig 16) were in complete agreement with the DNA sequence. The gene encoding Exg was c a l l e d cex (O 'Ne i l l et a l . , 1986a). The DNA sequence of the gene encoding endoglucanase a c t i v i t y on pEC2 was incomplete at the time that the amino terminal amino ac id sequence of EngA was a v a i l a b l e . An o l igonucleot ide of 21 nucleot ides was synthesized to ident i f y a recombinant DNA plasmid which contained the fragment of DNA encoding the EngA amino terminal sequence. The o l igonucleot ide 72 1 2 3 4 5 Fig 17. Polypeptides encoded by cloned DNA fragments. E . c o l i s t r a i n CSR603 was transformed with plasmids pBR322, pEC l , p E C l . l , and pEC2. The proteins encoded by the plasmids were labe l led with [ 3 S -S]methionine according to the procedure of Sancar et a l . (1979). The labe l led proteins were analysed by 7.5 % SDS-PAGE. Lane 1: CSR603; lane 2: CSR603(pBR322); lane 3: CSR603(pECl>; lane 4: CSR603(pEC1.1)J lane 5: CSR603(pEC2). Arrows indicate the pEC2 gene product (Mr 53,000), the pECl gene product (Mr 51,000), and the pBR322 tet and amp gene products. 73 Fig 18. Immunoprecipitation of the pECl polypept ide. The proteins encoded by pECl were labe l led as descr ibed in F ig 17. The l a b e l l e d proteins then were p rec ip i ta ted with various ant isera as descr ibed in MATERIALS AND METHODS, and the p rec ip i ta tes were analysed by 7.5-14 % gradient SDS-PAGE. Lane 1: the l a b e l l e d proteins before treatment with a n t i s e r a ; lane 2: prote in p rec ip i ta ted with ant isera ra ised against CMC-induced C . f imi e x t r a c e l l u l a r enzymes; lane 3: prote in p rec ip i ta ted by ant i se ra ra ised against Av ice l - induced C. f imi ex t race l lu la r enzymes. Numbered arrows on the l e f t re fe r to Mr markers X 1 0 - 3 . The large arrow on the r ight marked 51 indicates the pEC1 -spec i f i c immunoprecipitated polypept ide, Mr 51,000. 74 corresponded to amino acids 34-40 and was synthesized as a mixture of 8 d i s t i n c t sequences (F ig 15). The complexity of the o l igonucleot ide mixture was reduced by employing the codon bias demonstrated by the cex gene (O 'Ne i l l et a l . , 1986a). The o l igonucleot ide was radiolabe1 led and used to probe a dot blot on which various plasmid DNAs had been spotted. The probe hybr id ized with the plasmid pcEC2 (F ig 19). This plasmid c a r r i e d an add i t iona l 0.8 ki lobase Bam HI fragment of DNA immediately 5' to the Bam HI fragment cloned in pEC2. These two Bam HI fragments were shown to be contiguous in the C . f imi genome (Wong et a l . , 1986). It was thus found that the pEC2 gene product lacked the amino terminal 45 amino acids and was, In f a c t , a fusion prote in cons is t ing of a port ion of the te t racyc l ine res is tance gene product of pBR322 and the C terminal port ion of the endoglucanase EngA prote in (Wong, et a l . , 1986). The amino ac id compositions deduced from the completed DNA sequences of pECl and pcEC2 compared well with the experimental ly der ived amino ac id compositions of the C . f imi Exg and EngA, (Tables VIII and IX). In a d d i t i o n , the Mr-s estimated from the amino ac id compositions of the pECl and pcEC2 gene products (49,000 and 48,000, respect i ve ly ) were s i m i l a r to the M r S expected from the native Exg and EngA less 8 and 10 % for carbohydrate (51,500 and 51,300, r e s p e c t i v e l y ) . From these data and the sequence comparisons, i t was concluded that the plasmids pECl and pcEC2 contained the C . f imi genes encoding the c e l l u l o s e - b i n d i n g Exg and EngA. 75 F ig 19. Dot b lot hybr id i za t ion with the EngA o l igonuc leot ide probe. Plasmid DNA was a lka l i -denatured and 0.1 pmole was spotted onto n i t r o c e l l u l o s e . The blot was incubated with 106 cpm of the e n d - l a b e l l e d ol igodeoxyr ibonucleotide mixture for 1 h at 24°C. Fol lowing hybr id i za t ion , the blot was washed in 6X SSC at increasing temperatures and autoradiographed between washes. Panel A: autoradiography a f t e r washing at 25°C; Panel B: 35°C; Panel C: 50°C. Dot l a , pBR322; 2a, pDWl; 3a, pUC12::C16.6J 4a, pUC12::EBS737; lb , pcEC2; 2b, pEC2; 3b, pEC2.3; 4b, pEC2.1; l c , pEC3» 2c, pUC19::C35.6; 3c, pEC4; 4c, pEC5; lane d , C . f imi DNA in decreasing amounts: Id, 5 pg; 2d, 1 pg; 3d, 0.1 pg; 4d, 0.01 pg. Plasmid DNAs are explained in Appendix A. 76 IV Function of the Glycosyl Groups of the Cel lu lose-bound Ce l lu lases The genes encoding EngA and Exg a c t i v i t y were engineered for high level expression in E . c o l i (O 'Ne i l l et a l . , 1986b; Wong et a l . , 1986; Guo et a l . , in p ress ) . With the plasmid const ructs , pUC l2 : :PT IS : : s t y and pUCEC2 (Appendix A) , under ce r ta in growth condit ions (described in MATERIALS AND METHODS), substant ia l quant i t ies of recombinant c e l l u l a s e a c t i v i t y leaked into the cul ture medium (Guo et a l . , in p ress ) . Recombinant c e l l u l a s e a c t i v i t y was subsequently p u r i f i e d from cul ture medium using immunoadsorbent chromatography and Mono Q anion exchange column chromatography (Arfman, 1986). Amino terminal sequence analys is revealed that the recombinant Exg was accurate ly processed in E . c o l i while EngA was processed equal ly at two s i t e s , the correct one and an add i t iona l s i t e 3 amino acids upstream (Guo et a l . , in p ress ) . The recombinant c e l l u l a s e s appear not to be g l ycosy la ted : they d id not bind to Con A-Sepharose (Table XI) , they migrated on SDS gels true to the i r molecular weights as predicted from the DNA sequences (N .G i lkes , unpublished observat ion) , and no reports have been published to date demonstrating a g lycoprote in in E . c o l i . It therefore seemed feas ib le to use the recombinant c e l l u l a s e s as a source of unglycosylated c e l l u l a s e and compare the i r propert ies to the nat ive , g lycosy lated c e l l u l a s e s . S i g n i f i c a n t d i f fe rences in propert ies might suggest a funct ion for g l y c o s y l a t i o n . Both the native and the recombinant c e l l u l a s e s bound read i l y to Av ice l (F ig 20). Greater than 85% of the enzyme a c t i v i t y of the native and recombinant Exg and EngA adsorbed immediately to autoclaved Av ice l and remained bound over a 90 min period on i ce . V i r t u a l l y no a c t i v i t y could be detected in the buffer washes of the A v i c e l . The Av ice l p e l l e t s were assayed to ve r i f y that the enzymes were not inact i va ted . The sum of the a c t i v i t i e s detected in the three f rac t ions accounted 77 Table XI. Recombinant c e l l u l a s e a c t i v i t y recovered from Sepharose 4B and Con A-Sepharose 4B. Recombinant Sepharose 4B Con A-Sepharose 4B Enzyme Fract ion Fract ion 1 2 1 2 P r o t e i n A 96 4 95 5 Exg Act i v i t y 1 8 95 5 92 8 P r o t e i n A 99 1 88 12 EngA A c t i v i t y 0 79 21 82 18 E . c o l i (pUC12: :PTIS: :sty) and E . c o l i (pUCEC2) s t ra ins were grown under inducing condit ions at 37°C for 24 h to f a c i l i t a t e leakage of c e l l u l a s e a c t i v i t y into the cul ture medium. Adsorption to Con A-Sepharose 4B or to Sepharose 4B was done batchwise. C e l l - f r e e cul ture supernatant was incubated with e i ther Con A-Sepharose or Sepharose in a 30 ml centr i fuge tube on ice for 1 h. The supernatant was removed a f te r cent r i fugat ion and the matrix was washed 3 X with TN buffer ( f rac t ion 1), then with TN buffer containing 100 mM Me-Glc ( f rac t ion 2) . A l l f rac t ions were assayed for prote in and a c t i v i t y . Recovery was expressed as a % of the to ta l prote in or a c t i v i t y recovered from the matrix. A Prote in was determined by Biorad assay. B Exoglucanase a c t i v i t y was determined as pNPCase a c t i v i t y . c Endoglucanase a c t i v i t y was determined as CMCase a c t i v i t y . 78 100 A B 8 0 a • • 6 0 • o i § 4 0 • "o ^ 2 0 • . Activi • c D 8 0 i g 6 0 V • • 1 i 1 UJ 4 0 • • • • • • I 2C • 1 i * i * , • ^--^—r . ^_ 2 0 4 0 6 0 8 0 2 0 4 0 6 0 8 0 Incubation Time (min) Fig 20. Comparison of Av ice l binding for native and recombinant EngA and Exg. In 1.5 ml microfuge tubes, Av ice l was autoclaved in dH20 then washed and equ i l i b ra ted with ice co ld C bu f fe r . The Av ice l was then resuspended in C buf fe r contain ing recombinant Exg (panel A) , recombinant EngA (panel B) , native Exg (panel C ) , or native EngA (panel D) and incubated on ice for 0-90 min. After incubat ion, the Av ice l was centr i fuged for 2 min at 12,000 X g. The react ion supernatant was t ransfered to a c lean microfuge tube. The Av ice l was washed 1 X with C buf fer and kept as a p e l l e t on i c e . A l l f ract ions were kept on ice u n t i l a l l time points were taken and then a l l were assayed for a c t i v i t y as descr ibed in MATERIALS AND METHODS. Exg a c t i v i t y was measured as pNPCase; EngA a c t i v i t y was measured as CMCase. A c t i v i t y was expressed as % of the to ta l a c t i v i t y in the contro l react ion which contained c e l l u l a s e but no A v i c e l . 79 for less than 100% of the to ta l a c t i v i t y . This probably r e f l e c t s the i n a b i l i t y of soluble substrates to d i f fuse into the act ive s i t e when the enzymes are bound to insoluble c e l l u l o s e , and not the loss of enzymatic a c t i v i t y . Other propert ies of the c e l l u l a s e enzymes had been invest igated : s t a b i l i t y to pH and temperature var ia t ions (Arfman, 1986) and k i n e t i c parameters (Langsford et a l . , 1987). No s i g n i f i c a n t d i f fe rences were observed between the recombinant and the native c e l l u l a s e s . If the only physical d i f fe rence between the native and recombinant c e l l u l a s e s is the presence or absence of g l y c o s y l a t i o n , then these resu l t s indicate that g l ycosy la t ion plays no apparent ro le in pH or thermal s t a b i l i t y , hydro lys is of soluble c e l l u l o s i c substrates , or in binding to c r y s t a l l i n e c e l l u l o s e . V Character i zat ion of the C . f imi Protease Ear ly r e s u l t s suggested that a protease secreted by C . f imi had a ro le in generating the mult iple c e l l u l a s e a c t i v i t i e s observed in cul ture supernatants (Langsford et a l . , 1984). Protease a c t i v i t y was inact ivated by the inh ib i to r PMSF suggesting that the protease was a serine protease. Protease a c t i v i t y appeared to be induced under ce11ulase- inducing growth condit ions and regulated in part by catabo l i te repress ion (Table XII) . More protease a c t i v i t y was detected in the supernatants of cu l tures grown on CMC than on A v i c e l . Increasing the concentrat ion of Av ice l decreased the leve l of protease detected. No protease a c t i v i t y was detected on the Av ice l p e l l e t s or in the Gdn-HCl eluate of Av ice l (data not shown). To determine i f protease was induced by growth on c e l l u l o s e , further induction experiments were performed. C . f imi was grown on g lycero l with p a r a l l e l cu l tures grown on Avicel and xylan. Xylan was se lected because i t was a n o n - c e l l u l o s i c carbohydrate polymer. It was observed that 80 Table XII. The e f f e c t of carbon source on protease induct i on. Carbon Protease Source A c t i v i t y 0.1% Avice l 2.2 0.1% Avicel + 1% Glucose 0.2 0.1% CMC 12.0 0.1% CMC + 1% Glucose 0.3 1% Glucose 0.2 1% Cel lobiose 0.3 3% Av ice l 0.5 3% Av ice l <6d> 0.7 0.1% Av ice l (6d> 2.4 Supernatants were prepared as descr ibed in MATERIALS AND METHODS from cul tures grown for 3 or 6 days on the indicated carbon source. Supernatants were assayed for protease a c t i v i t y on Hide Powder Azure. A c t i v i t y is expressed as pg collagenase equivalents m l - 1 where 1 unit of collagenase re leases 1 pmole L - leuc ine in 5 h at pH 7.4 and 37°C. 81 xylan was a more potent inducer of protease than A v i c e l , but more importantly, equivalent leve ls of protease were induced by growth on g lycero l and Avicel (F ig 21, panel C) . This suggested that protease production was not regulated by induction by c e l l u l o s i c subst rates , but rather by catabo l i te repression during growth on glucose. On the other hand, c e l l u l a s e a c t i v i t y was induced by both c e l l u l o s i c and xylan subst rates . pNPCase and CMCase a c t i v i t i e s were induced by growth on Av ice l and xylan but not on g lycero l (F ig 21, panels B and D). A low leve l of c e l l u l a s e a c t i v i t y was detected in g lycero l supernatants. This CMCase a c t i v i t y may represent a basal cons t i tu t i ve level of enzyme expression since a low level of cen A messenger RNA was synthesized by C . f imi during growth on g l y c e r o l ; no cex message was detected to expla in the low level of pNPCase a c t i v i t y (Greenberg et a l . , 1987a). To determine i f the proteases induced by d i f f e r e n t carbon sources were the same, cul ture supernatants were analysed by native IEF PAGE. Protease a c t i v i t y in the gel was detected by casein hydro lys is in an agarose overlay (F ig 22). P r o t e o l y t i c a c t i v i t y appeared in cul ture supernatants in mult iple forms, as detected by IEF a n a l y s i s , during growth on A v i c e l , g l ycero l and xylan. The IEF patterns did not vary s i g n i f i c a n t l y with carbon source. Two major bands of a c t i v i t y were apparent between pH 5 . 7 - 5 . 8 as well as several minor bands above and below. One lane of the IEF g e l , which had received Av ice l cul ture supernatant, was s l i c e d hor i zon ta l l y and the gel s l i c e s were eluted with bu f fe r . The gel band eluates were assayed for Hide Powder Azure a c t i v i t y (F ig 23). The major Hide Powder a c t i v i t y was detected at pH 5 . 7 - 5 . 8 , corresponding to the major casein p r e c i p i t i n bands. Therefore, the major casein hydrolys ing proteases, ident ica l in A v i c e l , g l ycero l and xylan cul ture supernatants, were also ident ica l with the Hide Powder Azure act i v i ty . 82 20 * 10 e 40 Q_ 20 A 10 3 e < o CL E CD 0-5 £ <L> O 20 40 60 80 20 40 Culture Growth Time (Hr) 60 eo F ig 21. Induction of protease a c t i v i t y during growth on n o n - c e l l u l o s i c subst rates . C. f imi was grown on 0.1 % Av ice l ( O , 0.1 % g lycero l (•), or 0.1 % xylan (•), for 3 days. At various times during growth, samples were removed and supernatants were prepared as described (MATERIALS AND METHODS). The f i l t e r e d supernatants were assayed for prote in (by the method of Bradford, 1976; panel A) , pNPCase (panel B), Hide Powder Azure a c t i v i t y (expressed as unit equivalents of co l lagenase; panel C) , and CMCase (using the HBAH reagent; panel D). 83 A B C D E F G H I J K Fig 22. IEF ana lys is of protease a c t i v i t y induced during growth on various carbon sources. Supernatants descr ibed in the legend to F i g . 21 were analysed by i s o e l e c t r i c focussing in a non-urea s lab g e l . The samples were focussed for 90 min at 20 watts, constant power. After e lec t rophores i s , the gel was overlayed with an agarose gel containing d isso lved milk powder and incubated at 37°C for 6h. A po laro id photograph was made with type 57 f i l m . Lanes A - C : supernatants from cu l tu res grown for 25h; lanes E-G: for 49 h; lanes H-J : for 72 h. Lanes A, E, and H: supernatant from the culture grown on A v i c e l ; lanes B, F, and I: on g l y c e r o l ; lanes C , G, and J : on xy lan. Lanes D and K: pH standards. Numbers indicate pH estimated from standard markers. 84 E c m <j> m Q o F i g 23. Hide powder azure a c t i v i t y p r o f i l e from an IEF g e l . Supernatant from a cul ture grown on 0.1 % Avicel for 3 days was analyzed by IEF in a non-urea s lab g e l . A lane was excised and s l i c e d hor i zon ta l l y into 2 mm s l i c e s . The s l i c e s were eluted in TN buf fe r . The s l i c e e luates were assayed for Hide Powder Azure a c t i v i t y . S o l u b i l i z a t i o n of the blue dye was detected spectrophotometr ical ly at 595 nm. The pH gradient was estimated from pH standards electrophoresed in an adjacent lane . 85 VI The Role of Proteo lys is in Creating Mult ip le Ce l lu lase Act i v i t ies Because the ce l lu lose -bound enzymes could be p u r i f i e d to homogeneity, i t was possible to test the hypothesis that the m u l t i p l i c i t y of c e l l u l a s e s in the C . f imi system was generated in part by p r o t e o l y s i s . The protease induction data indicated that g lycero l cu l ture supernatants could serve as a source of ( e s s e n t i a l l y ) c e l l u l a s e - f r e e protease. I n i t i a l attempts to pur i f y the protease in Avicel -grown cul ture supernatants suggested that the pure protease was unstable ( J . B e t t s , unpublished observat ions) . The use of g l ycero l cu l ture supernatants would a l l e v i a t e prote in p u r i f i c a t i o n problems. The object of the experiment was to set up the react ion condit ions to mimic a 3 day, 0.1% Av ice l cul ture supernatant, incubate for 3 days,then examine the products to see i f a 6 day supernatant pattern was generated. FPLC -pur i f ied EngA and Exg were further f ract ionated by preparative non-denaturing PAGE to el iminate any cul ture derived p r o t e o l y t i c products which may have c o - p u r i f i e d with the c e l l u l a s e s . Mater ial e luted from the gel bands (corresponding to EngAl and Exg in F ig 12) was incubated with g lycero l cul ture supernatant. The react ions were stopped by the add i t ion of the protease i n h i b i t o r , PMSF, and assayed for c e l l u l a s e a c t i v i t y . Protease treatment caused no s i g n i f i c a n t change in a c t i v i t y (Table XII I ) . The react ion products were analyzed by SDS-PAGE. The resu l t s of the EngAl d igests are shown in F ig 24: prote in bands were detected by Western b l o t t i n g ana lys is using po lyc lonal rabbit antiserum ra ised against p u r i f i e d recombinant EngA (lanes A - I ) ; enzyme a c t i v i t y in the gel was detected using a CMC-agarose overlay (lanes J - R ) . Also included for comparison was unfract ionated FPLC -pur i f ied EngA. This showed a s ingle band by Western b l o t t i n g ana lys is (lane C) and a s ingle zone of enzyme a c t i v i t y (lane L ) . EngAl was indeed cleaved by the protease to y i e l d enzymatical1y act ive products with faster m o b i l i t i e s , MvS 45,000 to 48,000 (lane N) . In the absence of 86 Table XIII. Ce l lu lase a c t i v i t y in react ion supernatants a f t e r incubation with protease. Enzyme and Incubation Conditions i n i t i a l state A B C D E F nEngA1, s o l u b l e A 0.555 0.533 0. 555 0. 533 0. 519 -nExg, soluble* 0.096 0.098 0. 099 0. 095 0. 10 -nEng, bound A 0 0.01 0. 01 0 0 ND rEng, boundA 0 0.544 0. 103 0 0 0. 67 nExg, boundB 0 0.002 0. 002 0 0 0. 012 rExg, bound 3 0 0.0144 0. 004 0. 001 0. 001 0. 012 C e l l u l a s e s , e i ther free in so lu t ion (soluble) or bound to Av ice l (bound), were incubated for 3 days under various cond i t ions : A, at 4°C; B-E at 30°C with B, g lycero l cu l ture supernatant; C, g lycero l cu l ture supernatant containing PMSF; D, TN buf fe r ; E, TN buffer contain ing PMSF. Af ter incubat ion, Av ice l was removed by cent r i fugat ion and react ion supernatants were assayed for c e l l u l a s e a c t i v i t y . The procedure for obtaining Avicel -bound native and recombinant c e l l u l a s e s is described in MATERIALS AND METHODS. Av ice l with bound native or recombinant c e l l u l a s e was resuspended in g lycero l cu l ture supernatant to y i e l d a r a t i o of c e l l u l a s e to protease of 10 units pNPCase or 1.5 units CMCase (using DNS reagent) to 16 units of protease (expressed as units of collagenase equ iva lents ) . So lub le , p u r i f i e d native c e l l u l a s e s , e luted from a non-denaturing polyacrylamide g e l , were d i l u t e d into g lycero l cu l ture supernatant to y ie ld the above ra t ios of enzyme a c t i v i t y . Column F shows the c e l l u l a s e a c t i v i t y bound to Av ice l before incubat ion. n = nat i ve ; r = recombinant. A A c t i v i t y is expressed as pmoles glucose equivalents released min - 1 m l - 1 , using DNS reagent. B A c t i v i t y is expressed as pmoles pNP released min - 1 m l - 1 . 87 A B C D E F G H I J K L M N O P Q R 116-9 7 4 -66 -45 -29-F ig 24. SDS-PAGE analys is of soluble EngAl a f t e r incubation with protease. FPLC -pur i f ied native EngA was electrophoresed in a non-denaturing polyacrylamide g e l . The major a c t i v i t y (corresponding to EngAl in F ig 12) was e luted from the gel and incubated for 3 days under various condi t ions descr ibed in the legend to Table XIII . Samples were mixed 4 parts incubation mixture to 1 part double strength s o l u b i l i z a t i o n mix and bo i led for 2 min p r io r to e lec t rophores is . Samples were run on dupl icate 10 % acrylamide g e l s . After e l e c t r o p h o r e s i s , one gel was e l e c t r o b l o t t e d onto n i t r o c e l l u l o s e at 500 mA for 90 min and developed with antiserum ra ised against p u r i f i e d recombinant EngA (lanes A - I ) . The second gel was overlayed with a CMC-agarose gel and incubated at 37°C for 12 hours. The overlay was sta ined with Congo Red (lanes J - R ) . Lanes A and J : supernatant from a cul ture grown on 0.1 % Avice l for 3 days; lanes B and K: supernatant from a cul ture grown on 0.1 % Av ice l for 6 days; lanes C and L: p u r i f i e d EngA p r io r to f r a c t i o n a t i o n by non-denaturing PAGE, incubated with buffer at 4°C; lanes D and M: EngAl incubated at 4°C; lanes E and N: EngAl incubated at 30°C with g l ycero l cul ture supernatant; lanes F and 0: EngAl incubated at 30°C with g lycero l cul ture supernatant and PMSF; lanes G and P: EngAl incubated at 30°C with bu f fe r ; lanes H and Q: EngAl incubated at 30°C with buffer and PMSF; lanes I and R: g lycerol cu l ture supernatant incubated at 4°C with PMSF. The numbers are Mr markers X 1 0 - 3 . 88 protease, th is conversion d id not occur (lanes P and Q) . In the presence of protease and PMSF, a fa in t a c t i v i t y band can be seen (lane 0) but th is band also occurs with the protease alone (lane R) . This band may represent the basal leve l of cen A expression as the processed EngA. The p ro teo lys i s products appeared to be less ant igenic but more enzymatical1y act ive than the i r precursor (lane E) . Since the to ta l a c t i v i t y remained unchanged (Table XI I I ) , and since equal amounts of CMCase units were loaded in each lane, the larger hydro lys is zones made by the smaller enzymes may r e f l e c t an improved a b i l i t y to d i f fuse into the agarose over lay . Supernatants from cu l tures grown on 0.1% Avice l for 3 and 6 days were included for comparison. A c t i v i t y zones were present at Hr 57,000 and 48,000 by 3 days (lane J) but by 6 days the major a c t i v i t y was the Mr 48,000 a c t i v i t y and very l i t t l e of the Mr 57,000 a c t i v i t y was detectable (lane K) . The Western b l o t t i n g ana lys is showed that both proteins reacted with the anti -EngA antiserum (lanes A and B) suggesting that one was derived from the other. In. v i tro p ro teo lys i s of p u r i f i e d EngA e f f e c t i v e l y reproduced the in  v i vo s i t u a t i o n , although i t d id not proceed to the same extent. This demonstrates that in. v ivo , during growth on low concentrat ions of A v i c e l , EngA is hydrolysed in the cul ture supernatant, and that hydro lys is is already i n i t i a t e d by 3 days of growth. The observation that a c t i v i t y was not lost by protease treatment agreed also with the a c t i v i t y changes observed during growth on low concentrations of A v i c e l : a f te r growth for 3 days, the to ta l CMCase a c t i v i t y continued to increase s l i g h t l y over the next 6 days (Langsford et a l . , 1984). These r e s u l t s show conc lus ive ly that some of the c e l l u l o l y t i c a c t i v i t i e s observed in older cu l tures are derived p r o t e o l y t i c a l l y from EngA. The products of Exg a f t e r treatment with protease are shown in F ig 25: prote in bands were detected by s i l v e r s ta in ing (lanes A-H) and by Western b l o t t i n g ana lys is with 89 A B C D E F G 116. 97-4" 66-45-29-H I J K L M N O P Fig 25. SDS-PAGE analys is of soluble Exg a f te r incubation with protease. FPLC -pur i f ied Exg was electrophoresed in a non-denaturing polyacrylamide g e l . The a c t i v i t y was eluted from the gel and incubated for 3 days under various condi t ions described in the legend to Table XIII. Samples were mixed 4 parts incubation mixture to 1 part double strength s o l u b i l i z a t i o n mix and bo i led for 2 min p r i o r to e l e c t r o -phores is . Samples were run on dupl icate 10 % acrylamide g e l s . After e lec t rophores i s , one gel was stained with s i l v e r ( lanes A-H) . A second gel was e lec t rob lo t ted onto n i t r o c e l l u l o s e at 500 mA for 90 min and developed as a Western b lot with antiserum ra i sed against p u r i f i e d recombinant Exg ( lanes I -P) . Lanes A and I: g l ycero l cul ture supernatant incubated at 4°C with PMSF; lanes B and J : Exg incubated at 30°C with TN buffer and PMSF; lanes C and K: Exg incubated at 30°C with TN buf fe r ; lanes D and L: Exg incubated at 30°C with g l ycero l cu l ture supernatant and PMSF; lanes E and M: Exg incubated at 30°C with g lycero l cul ture supernatant; lanes F and N: Exg incubated at 4°C then d i l u t e d with g lycero l supernatant contain ing PMSF; lanes G and 0: supernatant from a cul ture grown on 0.1 % Avicel for 6 days; lanes H and P: supernatant from a cu l ture grown on 0.1 % Avicel for 3 days . Numbers are Mr markers X 1 0 - 3 . 90 antiserum ra ised against recombinant Exg (lanes I -P). The prote in v i s u a l i z e d in the s i l v e r stained gel suggested that no pro teo lys is occurred: the faster migrating protein at approximately 43,000 (lanes D-F) was present a lso in the g lycero l cul ture supernatant (lane A) . Lane F shows the sum of lanes A and B since i t represents Exg stored at 4°C then d i l u t e d into g lycero l cul ture supernatant with PMSF just p r io r to a n a l y s i s . The Western blot ana lys is revealed a poss ib le p ro teo lys i s product in samples incubated with g lycero l supernatant with and without PMSF (lanes L and M). The ant ibody -b ind ing , lower Mr product appeared to co-migrate with the s i l v e r s ta in ing band at Hr 43,000 but was not v i s i b l e in g lycero l supernatant (lanes I and N; lane N is the sum of lanes I and J ) . In the Av ice l cul ture supernatant samples, the ant ibody-b inding prote in appeared to be stable and did not s h i f t to a lower Hr product in the 6 day sample (lanes 0 and P) . The p o s s i b i l i t y of in. v i vo p ro teo lys i s can not be ru led out since the in tens i ty of the s ta in ing of the f u l l s ize band is not very st rong; therefore minor bands may be undetectable. However, the soluble Exg was c l e a r l y more res i s tant to p ro teo lys is in. v i tro than the soluble EngA. VII G lycosy lat ion Prevents the P r o t e o l y t i c Cleavage of Cel lulose-Bound Ce l lu lases Our hypothesis concerning the generation of the m u l t i p l i c i t y of the c e l l u l a s e s of C . f imi held that c e l l u l a s e s bound to c e l l u l o s e should be protected from p r o t e o l y s i s . To test t h i s , native c e l l u l a s e s bound to Av ice l were incubated with g lycero l cul ture supernatant. The recombinant c e l l u l a s e s , which had been shown to bind to Avicel (F ig 20), were adsorbed to Av ice l and treated with g lycero l cu l ture supernatant in p a r a l l e l experiments. Av ice l from a 6 day cul ture of C . f imi was used as the source 91 of substrate-bound native c e l l u l a s e . The amount of a c t i v i t y bound was estimated by measuring pNPCase. The amount of CMCase was assumed to be equal or l e s s , from enzyme p u r i f i c a t i o n y i e l d s . Av ice l was added to E.co1i cul ture supernatants to absorb out secreted recombinant EngA and Exg. The amount of a c t i v i t y bound was determined by measuring a c t i v i t y before and a f t e r adsorption to A v i c e l . As before , the protease react ions were set up to mimic the enzyme r a t i o s in a 3 day, low Avice l c u l t u r e , but in th is case the c e l l u l a s e a c t i v i t y was bound to A v i c e l ; the protease remained so lub le . The react ions were incubated for 3 days, then PMSF was added where necessary and the react ion supernatants were assayed for the s o l u b i l i z a t i o n of c e l l u l a s e a c t i v i t y (Table XII I ) . A very low leve l of native c e l l u l a s e a c t i v i t y was detectable in the supernatant of the react ion treated with act ive protease. In cont ras t , greater than 80% of the recombinant c e l l u l a s e a c t i v i t i e s , i n i t i a l l y bound to A v i c e l , was present in the supernatants of react ions t reated with act ive protease. The react ion products were analyzed by SDS-PAGE. Endoglucanase a c t i v i t y in the gel was detected with a CMC-agarose over lay . The native EngA appeared to be res i s tant to p r o t e o l y t i c degradation while the recombinant EngA was completely converted to a new a c t i v i t y with a Mr near 31,000 (F ig 26). The zymogram agreed with the a c t i v i t y data in Table XIII . In every r e a c t i o n , f u l l - s i z e , native CMCase a c t i v i t y was associated with Av ice l (lanes A-E) and was not detectable in react ion supernatants (lanes F - I ) . Intact recombinant CMCase a c t i v i t y was associated with Av ice l only in the contro l d igests (lanes K, M-0) and was not present in those react ion supernatants (lanes Q-S) . In the react ion contain ing act ive protease, no f u l l - s i z e recombinant EngA was detected. A c e l l u l o l y t i c , p ro teo lys i s fragment appeared in the react ion supernatant (lane P ) . Fol lowing p r o t e o l y s i s , less than 20% of the pro teo lys i s 92 Fig 26. SDS-PAGE analys is of the native and recombinant Avicel -bound EngA a f te r incubation with protease. Av ice l with native or recombinant EngA bound to i t was incubated with protease and cont ro ls in a 1 ml suspension as descr ibed in the legend to Table XIII . Reactions were stopped by the addi t ion of PMSF, then centr i fuged to y i e l d an Av ice l p e l l e t and the react ion supernatant. Samples were prepared for e l e c t r o -phoresis in a SDS-containing 10 % acrylamide gel as fo l lows : the Av ice l p e l l e t was resuspended in 250 ul s o l u b i l i z a t i o n mix, bo i led for 2 min, then cent r i fuged . The s o l u b i l i z a t i o n mix was t ransfered to a c lean tube. 15 u l , or 6%, of the Avicel -bound proteins were loaded onto the g e l , except in lane L where 70 u l , or 28 %, were loaded on the g e l . The protease react ion supernatants were mixed 4:1 with double strength s o l u b i l i z a t i o n mix, bo i led for 2 min, and 70 u l , or 5.5%, were loaded on the g e l . Fol lowing e lec t rophores i s , the gels were overlayed with a CMC-agarose gel and incubated at 37°C for 15 h. The overlay was then stained with Congo Red. Lanes A - I : native EngA; lanes K-S: recombinant EngA. Lanes A-E and K-0 : a c t i v i t y bound to A v i c e l ; lanes F-I and P-S: a c t i v i t y in react ion supernatant. Lanes A and K: incubated at 4°C; lanes B, F, L, and P: incubated at 30°C with g lycero l cul ture supernatant; lanes C, G, M, and Q: incubated at 30°C with g lycero l supernatant and PMSF; lanes D, H, N, and R: incubated at 30°C with TN bu f fe r ; lanes E, I, 0, and S : incubated at 30°C with TN buf fer and PMSF; lane J : g lycero l cul ture supernatant and PMSF stored at 4°C; lane T: E . c o l i <pUC12) culture supernatant proteins which bound to Av ice 1. 93 product of recombinant EngA remained associated with the Av ice l (lane L ) . Five times more sample was loaded in lane L than in lanes K and M-0 in order to see the hydro lys is zone made by the Avice 1-associated p ro teo lys i s product. The sample was d e l i b e r a t e l y overloaded to reveal the presence of intermediary cleavage products and the s ize of the products. No intermediate s ize a c t i v i t i e s were detected. A c t i v i t y associated with the Av ice l p e l l e t could be removed completely with further washing (Langsford et a l . , 1987). Substrate-bound Exg products were detected by Western b l o t t i n g ana lys is with antiserum ra ised against the recombinant Exg. Again the native Exg appeared r e s i s t a n t to p ro teo lys i s but the recombinant Exg was hydrolysed to y i e l d faster migrat ing, ant ibody-b inding proteins with M rs near 43,000 (F ig 27). The in tac t , ant ibody -b ind ing , native Exg was associated with Av ice l (lanes K-0) . No ant ibody-b inding prote in was observed in the react ion supernatants ( lanes R-U). In the react ions without protease, the f u l l - s i z e recombinant Exg was associated with Av ice l ( lanes F,G,H and J) and was not observed in the react ion supernatant ( lanes A and B) . After protease treatment, the f u l l s ize recombinant Exg was completely missing from the Av ice l p e l l e t but fa in t bands with M r s near 43,000 were observed when 5 - f o l d more sample was loaded (lane I) . Faint bands with Mr 43,000 were also observed in the contro l react ion containing g lycero l cu l ture supernatant and PMSF (lanes C and H). Since the g lycero l supernatant alone d id not contain a 43,000 prote in react ing with the antiserum (lane V) , i t is concluded that the res idual protease a c t i v i t y , less than 10 %, not inact ivated by PMSF, is responsible for the p a r t i a l degradation of recombinant Exg in the c o n t r o l . No f u l l s ize recombinant Exg was detected in any of the react ion supernatants. Since most of the pNPCase a c t i v i t y was recovered in the react ion supernatant a f te r protease 9 4 A B C D E F G H I J K L M N O P Q R S T U V - 2 9 Fig 27. SDS-PAGE analys is of the native and recombinant Avicel -bound Exg a f t e r incubation with protease. Av ice l with recombinant or native Exg bound to i t was incubated with protease and cont ro ls in a 1 ml suspension as descr ibed in the legend to Table XIII . Reactions were stopped and prepared for e lect rophores is as described in the legend to F i g . 26. 5-6 % of each sample was loaded on the gel except in lane I where 28 % was loaded. Fol lowing e lect rophores is in 10 % acrylamide SDS g e l s , the gels were e lec t rob lo t ted onto n i t r o c e l l u l o s e at 500 mA for 90 min. The b lots were developed with antiserum ra i sed against p u r i f i e d recombinant Exg. Lanes A - J : recombinant Exg; lanes K-U: native Exg. Lanes A-D and R-U: proteins in the protease react ion supernatants; lanes E-J and K-0: proteins bound to A v i c e l . Lanes A, F, 0, and U: incubated at 30°C with TN buffer and PMSF; lanes B, G, N, and T: incubated at 30°C with TN bu f fe r ; lanes C, H, M, and S: incubated at 30°C with g lycero l cul ture supernatant and PMSF; lanes D, I, L, and R: incubated at 30°C with g lycero l cul ture supernatant; lanes J and K: incubated at 4°C with buf fe r ; lane E: E . c o l i <pUC12> culture supernatant proteins which bound to A v i c e l ; lane P: supernatant from a C . f imi culture grown for 3 days on 0.1 % A v i c e l ; lane Q: supernatant from a C . f imi cul ture grown for 6 days on 0.1% A v i c e l . Lane V: g lycero l cu l ture supernatant and PMSF stored at 4°C. Numbers on the side are Mr standards X 1 0 - 3 . 95 treatment (Table XII I ) , i t was concluded that the ant ibody -b ind ing , p ro teo lys is product with Mr 43,000 retained c e l l u l a s e a c t i v i t y . A further contro l was done: a Western b lot of a l l the protease react ion products was probed with normal rabbit serum. No bands were detected (Appendix E ) . EngA and Exg share two regions of extensive amino ac id homology (Warren et a l . , 1986). One region named the PT box cons is ts of 20-22 residues of a l te rna t ing pro l ine and threonine. A synthet ic peptide (PTPTPTTPTPTPTTPTPTPTSG) was made (B. Singh, Univers i ty of Alberta) corresponding to the PT box of Exg and subsequently used to ra ise immune serum. The ant i -PT serum (courtesy of B.Moser) was used to probe a Western blot of the recombinant c e l l u l a s e p ro teo lys i s products in order to define the region where processing occurred (Fig 28). The ant i -PT serum did not react with the intact recombinant c e l l u l a s e s , nor did i t react with the s o l u b i l i z e d 31,000 EngA pro teo lys i s product, but i t d id react with a smaller prote in with Mr 18,000 to 20,000 and associated with Av ice l (lanes A-D, band II) . It is probable therefore , that the enyzmic fragment of the recombinant EngA did not contain the PT reg ion . The soluble Exg p ro teo l ys i s product, Mr 43,000, was recognized by the ant i -PT antiserum (lanes E-H, band I) ind icat ing that i t d id contain the PT box. The two bands v i s i b l e above 60,000 in a l l lanes were due to n o n - s p e c i f i c antibody b ind ing . An SDS gel was run with only sample bu f fe r , e lec t rob lo t ted to n i t r o c e l l u l o s e and probed with the ant i -PT antiserum. Two bands with Mr between 60,000 and 70,000 were observed (Appendix F ) . 96 A B C D E F G H 974-575-53-41-36 " • 9 •• Pig 28, Western b lot analys is of the recombinant c e l l u l a s e s a f te r incubation with protease. Samples of the protease digests of Avicel -bound recombinant Exg and EngA descr ibed in Table XIII and prepared for SDS-PAGE as descr ibed in Figs 26 and 27 were electrophoresed in a second 10 % acrylamide SDS gel and e lec t rob lo t ted onto n i t r o c e l l u l o s e . The b lot was developed with an ant isera ra ised against the PT box of the pECl gene product. Lanes A-D: recombinant EngA; lanes E-H: recombinant Exg. Lanes A and E: react ion supernatants a f t e r incubation with TN buf fe r ; lanes B and F: react ion supernatants a f t e r incubation with g lycero l cul ture supernatant; lanes C and G: protein bound to Av ice l a f te r incubation with TN b u f f e r ; lanes D and H: proteins bound to Avicel a f t e r incubation with g lycero l cu l ture supernatant; lane I: E . c o l i (pUC12) cul ture supernatant prote ins which bound to A v i c e l ; lane J : g lycero l culture supernatant. The volumes loaded on the gel were as indicated in Figs 26 and 27. The numbers on the l e f t indicate Mr markers X 1 0 - 3 . 97 DISCUSSION I P u r i f i c a t i o n and Character izat ion of EngA and Exg Two c e l l u l a s e s have been p u r i f i e d from the guanidine-HCl (Gdn-HCl) extract of A v i c e l . EngA, an endoglucanase, has an apparent Mr 57,000 as determined by SDS-PAGE. The p i is 8.2 under non-denaturing cond i t ions , and 7 . 0 - 7 . 3 when denatured by urea. EngA is a g lycoprote in and appears to be 10 % mannose. Exg, an exoglucanase, has an apparent My 56,000 as determined by SDS-PAGE. The p i is 5 . 8 - 5 . 9 under non-denaturing condit ions and 6.2 when denatured by urea. Exg is a lso a g lycoprote in and appears to be 8 % mannose. Several b a c t e r i a l c e l l u l a s e s have been character ized and some of the i r propert ies are summarized in Table II. EngA and Exg are t y p i c a l with regard to Mr and p i under denaturing cond i t ions . However, under native cond i t ions , the C . f imi EngA has a high p i s i m i l a r only to the EngY of E.chrvsanthemi (Boyer et a l . , 1987). EngA and Exg, p u r i f i e d from A v i c e l , appeared to represent a l l the renatured a c t i v i t i e s e luted from c e l l u l o s e with Gdn-HCl. This is indicated by the fact that no a c t i v i t y was recovered in the flow through f rac t ions of the Con A-Sepharose and Mono Q columns but only in those f rac t ions indicated in the e l u t i o n p r o f i l e s . The to ta l recovery of CMCase from Con A was greater than 85%, when the amount of CMCase a c t i v i t y contr ibuted by Exg is considered. Exg const i tu tes approximately 25% of the to ta l CMCase a c t i v i t y in the Gdn-HCl extract ( F i g . 9 ) . Incomplete recovery of CMCase from Con A was also observed by Beguin and Eisen (1978) due to substant ia l t r a i l i n g of the a c t i v i t y . The recovery of Exg from anion exchange chromatography was greater than 86% while the recovery of EngA was 50%. The major loss of EngA a c t i v i t y may have been the resu l t of p a r t i a l inac t i va t ion by d i l u t i o n and high pH p r io r to and during anion exchange chromatography. These p o s s i b i l i t i e s are supported by the fo l lowing : the addi t ion of BSA to enzyme react ions was observed to a id in s t a b i l i z i n g the 98 enzymes <N.Gilkes, unpublished observat ion) , and the enzymes were inact ivated at extreme pH (Arfman, 1986). The s p e c i f i c a c t i v i t y of EngA prev iously reported was 370 U mg - 1 (Gi lkes et a l . , 1984b), nearly 2 fo ld higher than that shown here. The d i f fe rence may r e f l e c t the absence of denaturation by high pH since the enzyme in the o r i g i n a l study was p u r i f i e d by non-denaturing PAGE. Other c e l l u l a s e s , for example EngB, may also have been bound to the c e l l u l o s e during cul ture growth but were not recovered.by th i s p u r i f i c a t i o n procedure. They may have washed off the Av ice l with T buffer during the f i r s t washing steps , or may have been denatured i r r e v e r s i b l y by Gdn-HCl. Recombinant EngB, the gene product of the cen B gene cloned in E . c o l i , was p u r i f i e d by i t s a f f i n i t y for A v i c e l . It e lutes from Av ice l at low ionic strength (Owalabi et a l . , in p ress ) . Non-denaturing PAGE and IEF revealed a degree of heterogeneity not observed by SDS-PAGE. The heterogeneity due to small d i f fe rences in charge may be the resu l t of l imi ted p r o t e o l y t i c processing at the carboxy terminal of the c e l l u l a s e s , incomplete renaturat ion of some of the prote in a f t e r denaturation by Gdn-HCl, or to microheterogeneity at the carbohydrate l e v e l . Processing at the amino terminus is un l ike ly since the amino terminal amino ac id sequence analyses gave only one sequence for each enzyme. The y ie lds during amino ac id sequencing were in agreement with the amount of prote in appl ied to the sequenator (R.Olefson, personal communication), thus e l iminat ing the p o s s i b i l i t y of s i g n i f i c a n t contaminants. The mult iple bands observed by IEF ana lys is may be caused by the assoc ia t ion of the c e l l u l a s e s with ampholyte components. Such interact ions can cause minor mult iple bands because the complexed proteins acquire a combined pi d i f f e r e n t from that of the free proteins (Scopes, 1982). A less l i k e l y p o s s i b i l i t y is that the mult iple proteins are isoenzymes which have ident ica l amino te rmin i , but which are products of other s i m i l a r genes. 99 Amino ac id composition ana lys is showed that Exg and EngA have a high pro l ine content r e l a t i v e to other b a c t e r i a l c e l l u l a s e s . Pro l ine content has been cor re la ted with the thermal s t a b i l i t y of o l i g o - 1 , 6 - g l u c o s i d a s e s of Baci11 us spp. (Suzuki et a l . , ' 1987). However, such a c o r r e l a t i o n can not be made for b a c t e r i a l c e l l u l a s e s . The pro l ine content of the C . f imi c e l l u l a s e s (EngA: 7.7 moles hi Exg: 6.7 moles %) was higher than the pro l ine content of the more thermally stable C.thermoce11um c e l l u l a s e s (EGA: 4.7 moles %, Beguin et a l . , 1985; EGD: 3.9 moles %, J o l i f f et a l . , 1986a; Eng: 5.1 moles %, Ng and Zeikus, 1981), and was comparable to the pro l ine content of the thermostable c e l l u l a s e s of T . fusca (Calza et a l . , 1985). The recombinant DNA clones encoding EngA and Exg were i d e n t i f i e d by amino ac id sequence comparisons and o l i g o -deoxyribonucleotide h y b r i d i z a t i o n . The plasmid pECl encodes Exg and pcEC2 encodes EngA. II. G lycosy lat ion EngA and Exg are g lycoprote ins . They s ta in with PAS in polyacrylamide g e l s , they bind to the l e c t i n Con A, and they release mannose as the major monosaccharide upon ac id hydro l ys i s . The separat ion of Exg from EngA by Con A a f f i n i t y chromatography suggests d i f fe rences in structure at the carbohydrate l e v e l . The a b i l i t y of one g lycoprote in to d isp lace another on a l e c t i n a f f i n i t y matrix, or the a b i l i t y of a gradient e l u t i o n to separate g lycoprote ins , is ind icat ive of d i f f e r e n t a f f i n i t i e s for the l e c t i n (Kornberg and Kornberg, 1978). Di f ferences in a f f i n i t y are a t t r ibu ted to d i f fe rences in carbohydrate s t ruc ture , e i ther in the number or the type of receptor s i t e s (B ie ly et a l . , 1976; Baenziger and F i e t e , 1979; Sharon and L i s , 1982). EngA, which had a higher a f f i n i t y for Con A, also has a higher mannose content than Exg. Other b a c t e r i a l c e l l u l a s e s are g lycoproteins (Calza et a l . , 1985; Hon-nami et a l . , 1986a; Ng and Zeikus, 1981; Yamane 100 et a l . , 1970; see Table III) but the carbohydrates of only the Pseudomonas c e l l u l a s e s have been character ized (Yamane et a l . , 1970). The c e l l u l a s e s of Pseudomonas f1uorescens var. c e l l u l o s a contained la rge ly glucose or ga lactose , with lesser amounts of mannose, fucose and amino sugars. The glycopeptide linkage was not reported. The carbohydrate moieties of the C . f imi enzymes were res i s tant to cleavage by Endo H and were sens i t i ve to a l k a l i (N .G i lkes , unpublished observations) suggesting that the sugars are 0-1 inked to the polypeptides v ia the hydroxyl groups of serine or threonine. Other b a c t e r i a l g lycoproteins contained 0-1 inked glycans (Lamed and Bayer, in press ; Strobel et a l . , 1972; Sumper, 1987; see Table I I I ) . The l inkages were i d e n t i f i e d as mannose-O-threonine (Strobel et a l . , 1972) and as galactose-O- threonine (Sumper, 1987). The 0-1 inked glycans occurred as large complexes (Lamed and Bayer, in press ; Strobel et a l . , 1972) and as small o l igosacchar ides of 2 to 4 sugar residues (Sumper, 1987). The number and s ize of o l igosacchar ides l inked to EngA and Exg was not determined. Prel iminary ft-elimination studies suggested that several hydroxyl amino acids were involved (N .G i lkes , unpublished observat ions) . It therefore seems l i k e l y that the carbohydrate occurs as several small o l igosacchar ides as is c h a r a c t e r i s t i c of the fungal c e l l u l a s e s (Gum and Brown, 1976; Salovuori et a l . , 1987). Compared with eukaryotes, g lycoproteins in prokaryotes are ra re . The r a r i t y of b a c t e r i a l g lycoproteins and the a v a i l a b i l i t y of unglycosylated EngA and Exg in the form of recombinant gene products expressed in E . c o l i . prompted an invest igat ion into the ro le of g l ycosy la t ion of the C . f imi c e l l u l a s e s . The propert ies of nat i ve , g lycosy lated c e l l u l a s e s were compared with the propert ies of the recombinant, unglycosylated EngA and Exg. The propert ies tested were: thermal and pH s t a b i l i t y (Arfman, 1986), k i n e t i c parameters (Langsford et a l . , 1987), and the a b i l i t y to adsorb to A v i c e l . No s i g n i f i c a n t d i f fe rences were observed. G lycosy lat ion 101 appears not to be important for the hydro lys is of synthet ic , soluble subst rates . G lycosy lat ion may however, a f f e c t the binding in teract ion between enzyme and c r y s t a l l i n e , insoluble substrates since i t has been shown that recombinant EngA and Exg, but not native EngA or Exg, may be eluted from Avjcel with water (N .G i lkes , unpublished observat ions) . E lu t ion of the native enzymes from c e l l u l o s e required strong denaturing cond i t ions . It may be that the glycan moiety interacts with c e l l u l o s e to influence binding or , the glycan may a f f e c t the conformation of the subst rate -b ind ing region to influence b inding . Obviously, the addi t ion of mannose residues to the proteins s i g n i f i c a n t l y 'enhances the potent ia l for hydrogen bonding to c e l l u l o s e . The influence of g l ycosy la t ion on the m u l t i p l i c i t y of the C . f imi c e l l u l a s e s was not invest igated . It would be in teres t ing to know i f the two EngA enzymes d i f f e r at the carbohydrate l e v e l , e i ther in amount or in s t ruc tu re . III. Proteolys i s C . f imi secretes one or more proteases which are regulated by ca tabo l i te repress ion and inh ib i ted by the serine protease i n h i b i t o r , PMSF. S o i l micro-organisms may secrete proteases to scavenge for nitrogen in a r e l a t i v e l y nut r ient -poor environment. Ce l lu lose is not r i c h in n i t rogen, consequently, i t is not su rp r i s ing that c e l l u l o l y t i c organisms secrete proteases in add i t ion to c e l l u l a s e s as has been observed for E.chrvsanthemi (Andro et a l . , 1984), P.solanacearum ( S c h e l l , 1987), T . fusca (Calza et a l . , 1985), S.pulverulentum (Er iksson and Pettersson, 1982) and T.reese i (Shei r -Neiss and Montenecourt, 1984). The proteases of E.chrvsanthemi and P.solanacearum are considered to be v i rulence factors (Wandersman et a l . , 1986; S c h e l l , 1987). The protease of E.chrvsanthemi was not induced during growth on g lycero l (Wandersman et a l . , 1986) in contrast to the protease a c t i v i t y of C . f i m i . 102 The proteases induced during growth of C . f imi on g lycero l appeared to be ident i ca l e lec t rophoret ica l1y with those induced during growth on A v i c e l . The induction of protease a c t i v i t y during growth on g lycero l provided a convenient means of obtaining a e e l l u l a s e - f r e e protease preparat ion with which to test our hypothesis concerning the o r ig in of the mult iple c e l l u l a s e s . The hypothesis held that the mult iple c e l l u l a s e s observed in the supernatants of cu l tures of C . f imi grown on low concentrations of Av ice l were derived from only a few gene products. The major CMCase present in supernatants of cu l tures grown for 3 days on 0.1% Avice l had, by 6 days of growth, developed a fas ter mobi l i ty on SDS-PAGE. Both the 3 day and the 6 day major CMCases reacted with s p e c i f i c antiserum ra ised against the recombinant EngA. Incubation of p u r i f i e d , native EngA with g lycero l cul ture supernatant e f f e c t i v e l y reproduced, in v i t ro . t h i s s h i f t in mobi l i ty observed jji v ivo . In a d d i t i o n , some intermediate a c t i v i t i e s were produced. Despite the p r o t e o l y t i c degradation, the net a c t i v i t y of EngA towards CMC remained unchanged. These resu l t s showed conc lus ive ly that some of the c e l l u l o l y t i c a c t i v i t i e s observed in cul ture supernatants are derived p ro teo l y t i ca l1y from EngA and are not due to secret ion of new and d i f f e r e n t gene products. A s i m i l a r experiment had been done with the T . fusca endoglucanase El (Calza et a l . , 1985). The c e l l u l a s e was p u r i f i e d as a family of three enzymes, each d i f f e r i n g in molecular weight, but a l l were ant igenica l1y r e l a t e d . The largest one was incubated with a protease f rac t ion and was processed into the smallest one. The processed form reta ined more than 80% of the s t a r t i n g a c t i v i t y even though it was missing a Mr 24,000 fragment. El is not g l ycosy la ted . A second aspect of our hypothesis held that the c e l l u l a s e s , once bound to A v i c e l , would be protected from p r o t e o l y s i s . This proved to be the case. No p ro teo l y t i c degradation was observed when native EngA and Exg, while bound to A v i c e l , were incubated with g lycero l cul ture supernatant. 103 In cont rast , when recombinant EngA and Exg were bound to Avicel and incubated with g lycero l cul ture supernatant, both c e l l u l a s e s were processed to smaller forms. The processed forms of recombinant EngA and Exg reta ined a c t i v i t y towards soluble substrates but had lost the a b i l i t y to bind to A v i c e l . These experiments with the C . f imi protease provide important information regarding the ro le of g l y c o s y l a t i o n , and funct ional and s t ruc tu ra l aspects of the two c e l l u l a s e s , EngA and Exg. Moreover, we can extend our hypothesis regarding the o r i g i n of c e l l u l a s e m u l t i p l i c i t y to include regu lat ion of substrate s p e c i f i c i t y by p r o t e o l y s i s . F i r s t l y , g l ycosy la t ion is c l e a r l y important in l i m i t i n g the s e n s i t i v i t y of the c e l l u l a s e s to p r o t e o l y s i s . When the native EngA and Exg were bound to c e l l u l o s e , they were completely protected from proteo lys is while the recombinant EngA and Exg were completely processed, apparently in a s p e c i f i c manner. Secondly, the new enzymatical1y act ive processed forms of native EngA migrated more slowly in SDS-PAGE than the enzymatical1y act ive processed recombinant EngA. This may indicate that d i f f e r e n t s i t e s on the native EngA were recognized for cleavage or that the processed forms were the g lycosy lated por t ions , causing them to migrate more slowly than the unglycosylated fragments. Further charac te r i za t ion of the processed forms is needed to determine which of these explanations is c o r r e c t . Amino terminal sequencing of the processed fragments should indicate the cleavage s i t e of the protease. Th i rd l y , nei ther binding to Av ice l nor g l ycosy la t ion alone are s u f f i c i e n t to protect against p r o t e o l y s i s . The observation that the recombinant EngA and Exg are processed completely while bound to A v i c e l , indicates that the protease cleavage s i t e is exposed and is not blocked by the A v i c e l . That the native EngA and Exg are completely protected while bound to A v i c e l , but not while free in s o l u t i o n , suggests two possib le means for g l ycosy la t ion to have an e f f e c t . The f i r s t is that a 104 conformation change in the c e l l u l a s e is induced upon binding to A v i c e l . The conformation change is mediated or enhanced by the glycan moiet ies , perhaps through hydrogen bonding to c e l l u l o s e . The conformation change protects the suscept ib le peptide bonds. The second p o s s i b i l i t y is that the carbohydrate is l inked to residues near the protease cleavage s i t e and s t e r i c l y hinders the protease to protect the suscept ib le peptide bonds. In th is second case, one may have to assume that there are two kinds of cleavage s i t e s recognized to expla in the protect ive e f f e c t of binding to A v i c e l . One cleavage s i t e is blocked when c e l l u l a s e s are bound to A v i c e l . The other s i t e is exposed when c e l l u l a s e s are bound to A v i c e l , but is g lycosy lated in the native c e l l u l a s e s and is therefore protected. F i n a l l y , the data permit the assignment of funct ional ro les to the s t ruc tu ra l regions descr ibed by Warren et a l . (1986). Three d i s t i n c t regions were descr ibed: an i r r e g u l a r , hydroxyl r i c h region of low charge dens i ty , separated by a region of a l t e r n a t i n g pro l ine and threonine residues (the PT box) from an ordered, charged region conta in ing , by analogy with lysozyme, a putative act ive s i t e (Warren et a l . , 1986). The region contain ing the putative act ive s i t e occurs at the carboxy terminus of EngA and at the amino terminus of Exg. The r e s u l t s of the p ro teo lys i s experiments support the proposal that the ordered, charged region is the hydro ly t i c domain and suggest that the i r r e g u l a r , hydroxyl r i c h region is the subst rate -b ind ing domain (F ig 29). The reasoning goes as fo i lows. Delet ion ana lys is of the cenA gene encoding EngA showed that the amino terminal region was not required for a c t i v i t y but that the carboxy terminal was essent ia l (Wong et a l . , 1986; Guo et a l . , in p ress ) . S imi la r ana lys is of the cex gene demonstrated that the amino terminus was required for a c t i v i t y while the carboxy terminus was not ( O ' N e i l l , 1986). Recombinant EngA was processed into two fragments by the C . f imi protease: an enzymatical1y act ive fragment no longer capable of 105 1 1 ' ' \ % t • PT ; AS; N * * x, x-0- ' ! i 1 112 134 418 b 1 w JAS; PT -^'^•.•.<-»».->M-y.\-.'v,".-.%-.-.-. -.-.s-.-.-. J s 1 316 335 F i g 29. Proposed b i f u n c t l o n a l organization, of EngA (a) and Exg (b) (from Langsford et a l . , 1987 and Warren et a l . , 1986). The shaded region represents the c e l l u l o s e - b i n d i n g domain, a low charge, hydroxyl r i c h reg ion . The unshaded region represents the c a t a l y t i c domain, an ordered, charged reg ion . PT Is the conserved prol Ine/threonine hinge reg ion ; AS Is the putat ive act ive s i t e ; arrows indicate the deduced C . f imi protease cleavage s i t e s of the non-glycosylated enzymes; numbers re fer to amino a c i d res idues , beginning at the mature N-termin1. 106 binding c e l l u l o s e and a smal ler , enzymatical1y inact ive fragment associated with Av ice l and containing the PT box. Recombinant Exg was also processed into an enzymatical1y act ive fragment no longer capable of binding to c e l l u l o s e but contain ing the PT box. The remaining fragment of the prote in was not detected, probably because i t had migrated of f the end of the gel due to i t s small s ize (Mr less than 7000). Delet ion ana lys is indicates that the enzymatic fragment of EngA must be the C terminal fragment and the enzymatic fragment of Exg must be the amino terminal fragment. The s ize and nature of the fragments react ing with the ant i -PT antibody suggests that processing occurs on the carboxy side of the PT box (F ig 29» Langsford et a l . , 1987). Cleavage here l i be ra tes the ordered, charged, hydro ly t i c reg ion , the a c t i v i t y observed in the protease react ion supernatants, and leaves behind the i r r e g u l a r , hydroxyl r i c h reg ion , the small peptide observed in the Av ice l p e l l e t . The i r regu lar regions of EngA and Exg share 50% homology at the amino ac id l e v e l , yet are dispensable for hydro ly t i c a c t i v i t y . Loss of th is region means a reduced a f f i n i t y for insoluble subst rates , ind icat ing that t h i s region is the substrate binding domain. These regions are polar and r i c h in hydroxy amino acids (Warren et a l . , 1986), and are therefore su i tab le for forming many hydrogen bonds with c e l l u l o s e . The C . f imi protease e f f e c t i v e l y d iv ides the c e l l u l a s e s into two funct ional domains which are phys i ca l l y separated from one another by the PT box. These funct ional assignments to s t ruc tura l regions have been tested by making a fusion prote in (Warren et a l . , in p ress ) . A gene fusion was constructed from the DNA fragments encoding only the proposed hydro ly t i c domains and the PT box of EngA and Exg. The fusion product produced in E . c o l i lacks the proposed c e l l u l o s e - b i n d i n g domain: i t has both endo- and exoglucanase a c t i v i t i e s but does not bind to A v i c e l . The s t ruc tu ra l and funct ional organizat ion proposed for the C . f imi EngA and Exg is s i m i l a r to that of the c e l l u l a s e s of 107 T.reese i . EG I, EG III, CBH I and CBH II have in common a region of approximately 30 amino acids with 70% sequence homology located at e i ther the amino or carboxy terminus of the prote in (Knowles et a l . , 1987). This homologous region is not essent ia l for the hydro lys is of soluble substrates but is required for binding to and hydro lys is of Av ice l (Van Ti lbeurgh et a l . , 1986; Knowles et a l . , 1987). A hypothet ical model of a c e l l u l a s e enzyme was proposed (Schmuck et a l . , 1986) based on amino ac id sequence data and on low angle X-ray d i f f r a c t i o n s tud ies , showing a t a d p o l e - l i k e s t ruc ture . The subst rate -binding domain (the t a i l ) was phys i ca l l y separated from the hydro ly t i c domain (the head) by an 0 -g lycosy lated hinge region r i c h in hydroxyl amino a c i d s . The PT box of C . f imi has been proposed to act l i k e a hinge region based on i t s sequence s i m i l a r i t y to the hinge region of IgA (Plaut et a l . , 1975). However, nei ther EngA nor Exg are cleaved by the Ne i s s e r i a  gonorrhoeae IgAl protease, (N .G i lkes , unpublished observations) which cleaves within the hinge region of IgAl . How the conserved domains of c e l l u l a s e enzymes s p e c i f i c a l l y a f f e c t the binding to insoluble c e l l u l o s e and i t s subsequent hydro l ys i s , remains to be def ined . Klyosov et a l . (1986) have proposed that the C i f a c t o r , described by others as the component which creates amorphous regions in c r y s t a l l i n e c e l l u l o s e , is not a unique component but rather i t is the capacity of a c e l l u l a s e enzyme to adsorb onto the surface of insoluble c e l l u l o s e . In the Clostr id ium thermocellum cel lu losome, there is a unique component, the SI subunit , which f a c i l i t a t e s the binding of the Avicelase component and the subsequent hydro lys is of Av icel (Wu and Demain, in p ress ) . In the absence of the non-enzymatic SI subunit , the Avicelase does not bind to A v i c e l , nor does i t hydrolyse A v i c e l . It has not yet been shown i f the subst rate -b ind ing domain of a c e l l u l a s e enzyme, or the SI subunit a lone, can influence the structure of c r y s t a l l i n e c e l l u l o s e , perhaps act ing as the postulated hydrogen bondase (Enari and Niku-Paavola, 1987). 108 P r o t e o l y t i c processing of the terminal c e l l u l o s e - b i n d i n g domains has been proposed as a means of regulat ing and adjust ing the substrate s p e c i f i c i t i e s of c e l l u l a s e s during the hydro lys is of complex carbohydrates (Knowles et a l . , 1987). G lycosy lat ion of c e l l u l a s e s may have evolved as a means of regu lat ing p r o t e o l y s i s . While there is insoluble substrate to hydrolyse, c e l l u l a s e s bind to i t and the g lycosy lated enzymes are protected from p r o t e o l y s i s . Once the substrate has been s o l u b i l i z e d , the enzymes become free in so lu t ion and suscept ib le to l imi ted p r o t e o l y s i s . Limited p ro teo lys i s to remove c e l l u l o s e - b i n d i n g regions may provide enzymes with improved a f f i n i t i e s for shorter , soluble e e l l o d e x t r i n s . S i t e - s p e c i f i c g l ycosy la t ion has been shown to regulate the p r o t e o l y t i c processing of an H5N2 inf luenza v i rus hemagglutinin, thus a f f e c t i n g the v irulence of the v irus (Deshpande et a l . , 1987). IV. Future Prospects The c e l l u l a s e s of C . f imi bind t i g h t l y to c r y s t a l l i n e c e l l u l o s e but they do not hydrolyse i t e f f i c i e n t l y . Af ter several days of cul ture growth, substant ia l quant i t ies of Av ice l remain in the cul ture medium. The indus t r ia l app l i ca t ion of the C . f imi ce11ulases in the production of glucose from waste c e l l u l o s i c s w i l l require improvements to the a b i l i t y to hydrolyse insoluble subst rates . The use of act ive s i t e inh ib i to rs and X-ray crysta l lography w i l l provide valuable information about the mode of act ion and the three-dimensional structure of the c e l l u l a s e s . Comparisons of such data to the data obtained for more e f f i c i e n t degraders of c e l l u l o s e should provide d i r e c t i o n for the s i t e s p e c i f i c mutagenesis of key amino acids with the aim of c reat ing an improved c e l l u l a s e . Prote in engineering has been successful in improving the a c t i v i t y and s t a b i l i t y of other enzymes (Lau and Fersht , 1987; Pantoliano et a l . , 1987) and should provide in teres t ing p o s s i b i l i t i e s for the C . f imi c e l l u l a s e s . 109 REFERENCES Adams,S.P. , Kavka ,K .S . , Wykes ,E . J . , H o l d e r , S . B . , and G a l l u p p i , G.R. (1983) Hindered dialkylamino nucleoside phosphite reagents in the synthesis of 2 DNA 5.1-mers. J.Amer.Chem. Soc. 105, 661-663. Andro ,T . , Chambost ,J .P. , Kotoujansky,A. , C a t t a n e o , J . , Ber thau,Y . , B a r r a s , F . , van Gi jsegem,F. and Coleno,A. (1984) Mutations of pectate lyase and c e l l u l a s e . J . B a c t e r i o l . 160, 1199-1203. Appleyard,R.K. (1954) Segregation of new lysogenic types during growth of a doubly lysogenic s t r a i n derived from Escher ich ia c o l i K12. Genetics 39, 440-452. Arfman,N. (1986) P u r i f i c a t i o n and charac te r i za t ion of two recombinant c e l l u l a s e s from E . c o l i and the generation of monoclonal ant ibodies against a native exoglucanase. M.Sc. Thes is , Un ivers i ty of Groningen, Groningen, The Netherlands. A tk inson ,T . , and Smith,M. (1984) S o l i d phase synthesis of o l igodeoxyr ibonucleot ides by the phosphite t r i e s t e r method. In Ol igonucleot ide synthes is : A p r a c t i c a l approach. G a i t , M . J . , e d . , pp. 35-81, IRL Press , Washington, D.C. Baenz iger , J .U . and F i e t e , D . (1979) St ructura l determinants of Concanavalin A s p e c i f i c i t y for o l igosacchar ides . J .B io l .Chem. 254, 2400-2407. B a g n a r a , C , Gaudin.C. and B e l a i c h , J . P . (1986) Phys io log ica l propert ies of Cellulomonas fermentans, a mesophil ic c e l l u -l o l y t i c bacterium. A p p l . M i c r o b i o l . B i o t e c h n o l . 26, 170-176. B a y e r , E . A . , Kenig,R. and Lamed,R. (1983). Adherence of C lostr id ium thermoce11 urn to c e l l u l o s e . J . B a c t e r i o l . 156, 818-827. Bayer ,E .A . and Lamed,R. (1986) U l t rast ructure of the c e l l -surface cellulosome of C lostr id ium thermocellum and i ts in te ract ion with c e l l u l o s e . J . B a c t e r i o l . 167, 828-836. Beguin,P. (1983) Detection of c e l l u l a s e a c t i v i t y in polyacrylamide gels using Congo-Red stained agar r e p l i c a s . Anal.Biochem. 131, 333-336. Begu in ,P . , Cornet ,P. and A u b e r t , J . - P . (1985) Sequence of a c e l l u l a s e gene of the thermophil ic bacterium Clostr id ium  thermoce Hum. J . B a c t e r i o l . 162, 102-105. 110 Begu in ,P . , Cornet ,P. and M i l l e t , J . (1983). I d e n t i f i c a t i o n of the endoglucanase encoded by the celB gene of Clostr id ium  thermoce Hum. Biochimie 65, 495-500. Beguin,P and E isen ,H . (1978) P u r i f i c a t i o n and p a r t i a l character -i za t ion of three e x t r a c e l l u l a r c e l l u l a s e s from Ce11ulomonas sp. Eur .J .Biochem. 87, 525-531. Begu in ,P . , E i sen ,H . and Roupas,A. (1977) Free and c e l l u l o s e -bound c e l l u l a s e s in a Cellulomonas spec ies . J .Gen. M i c r o b i o l . 101, 191-196. Begu in ,P . , G i l k e s , N . R . , K i l b u r n , D . G . , M i l l e r , R . C . , J r . , O ' N e i l l , G.P. and Warren,R.A.J . (1987) Cloning of c e l l u l a s e genes. CRC C r i t i c a l Reviews in Biotechnology 6, 129-162. Beldman,G., Searle-VanLeeuwen,M.F. , Rombouts,F.M. and Voragen, F . G . J . (1985) The c e l l u l a s e of Trichoderma v i r i d e . E u r . J . Biochem. 146, 301-308. Berghem,L.E.R. , Pe t te rsson ,L .G . and Ax io -F redr iksson ,U . (1975) The mechanism of enzymatic c e l l u l o s e degradation: character -i za t ion and enzymatic propert ies of a fl-1,4-glucan c e l l o b i o -hydrolase from Trichoderma v i r i d e . Eur .J .Biochem. 53, 55-62. Bhikhabhai ,R. , Johansson,G. and Pet tersson,G . (1985) C e l l o b i o -hydrolase from Trichoderma reese i : Internal homology and pred ic t ion of secondary s t ruc ture . In t . J .Pept ide Protein Res. 25, 368-374. B i e l y , P . Kratky ,Z . and Bauer,S. (1976) Interact ion of Concanava-l i n A with external mannan-proteins of Saccharomvces  cerev i s iae . Eur .J .Biochem. 70, 75-81. B l a k e , M . S . , Johnston ,K .H . , Russe11-Jones ,G . J . , G o t s c h l i c h , E . C . (1984) A r a p i d , sens i t i ve method for detect ion of a lka l ine phosphatase-conjugated ant i -ant ibody on western b l o t s . Anal.Biochem. 136, 175-179. Bodenmann,J., He in iger ,U . and Hohl,H.R. (1985) E x t r a c e l l u l a r enzymes of Phvtophthora in festans : endoce11ulase, ft-gluco-s idases , and 1,3-ft -glucanases. C a n . J . M i c r o b i o l . 31, 75-82. Boye r ,M . -H . , Cami ,B . , Chambost , J . -P . , Magnan,M. and Cattaneo,J . (1987) Character i zat ion of a new endoglucanase from Erwinia  chrvsanthemi. Eur .J .Biochem. 162, 311-316. B o y e r , M . - H . , Chambost ,J .P. , Magnan,M. and Cat taneo,J . (1984a) Carboxymethyl-ce11ulase from Erwinia chrvsanthemi: Produc-t ion and regulat ion of e x t r a c e l l u l a r carboxymethyl-c e l l u l a s e . J . B i o t e c h n o l . 1, 229-240. 111 (1984b) Carboxymethyl-ce11ulase from Erw in la chrvsanthem i : P u r i f i c a t i o n and p a r t i a l charac te r i za t ion of an endo- f l -1 ,4 -glucanase. J . B i o t e c h n o l . 1, 241-252. Bradford,M. (1976) A rapid and sens i t i ve method for the quant i ta t ion of microgram quant i t ies of prote in u t i l i z i n g the p r i n c i p l e s of prote in dye b ind ing . Anal.Biochem. 72, 248-254. C a l z a , R . E . , I rwin,D.C. and Wi lson,D.B. (1985) P u r i f i c a t i o n and charac te r i za t ion of two fl-1,4-endoglucanases from Thermomono-spora fusca . Biochemistry 24, 7797-7804. Chang,W.T.H. and Thayer,D.W. (1977) The c e l l u l a s e system of a Cytophaga spec ies . C a n . J . M i c r o b i o l . 23, 1285-1292. C h e n , C M . , G r i t z a l i , M . and Stafford ,D.W. (1987) Nucleotide sequence and deduced primary structure of ce l lob iohydro lase II from Tr ichoderma reese i . Bio/Technology 5, 274-278. Choudhury,N., Gray ,P .P . and Dunn,N.W. (1980) S a c c h a r i f i c a t i o n of sugar cane bagasse by an enzyme preparat ion from C e l l u l o - monas ; res istance to product i n h i b i t i o n . B i o t e c h n o l . L e t t . 2 , 427-428. Coughlan,M.P. (1985) The propert ies of fungal and b a c t e r i a l c e l l u l a s e s with comment on the i r production and a p p l i c a t i o n . Biotechnol .Genet .Engineer .Rev. 3, 39-109. Coughlan,M.P. , Hon-nami,K. , Hon-nami,H. , L jungdah l , L .G . , Paul in , J . J . and Rigsby,W.E. (1985) The c e l l u l o l y t i c enzyme complex of C lostr id ium thermoce11 urn is very la rge . Biochem.Biophys. Res. Comm. 130, 904-909. Creuzet ,N. and F r i xon ,C . (1983) P u r i f i c a t i o n and charac te r i za t ion of an endoglucanase from a newly i so lated thermophi l ic anaerobic bacterium. Biochimie 65, 149-156. C reuzet ,N . , B e r e n g e r , J . - F . and F r i xon ,C . (1983) Character izat ion of exoglucanase and synerg i s t i c hydro lys is of c e l l u l o s e in C lostr id ium stercorar ium. FEMS M i c r o b i o l . L e t t . 20, 347-350. Deshpande,K.L. , F r i e d , V . A . , Ando,M. and Webster,R.G. (1987) G lycosy lat ion a f f e c t s cleavage of art H5N2 inf luenza v irus hemagglutinin and regulates v i ru lence . P roc .Nat l .Acad . S c i . , USA 84, 36-40. Deshpande,V. Rao,M. , Keskar ,S . and Mishra.C . (1984) Occurrence of a p roce l lu lase in the cul ture f i l t r a t e s of Pen ic i l l i um  iahth inel lum. Enzyme Microb.Technol . 6, 371-374. Detroy,R.W. and S t . J u l i a n , G . (1983) Biomass conversion: Fermentation chemicals and f u e l s . CRC C r i t i c a l Rev. M i c r o b i o l . 10, 203-228. 112 E n a r i , T . - M . and N iku -Paavo la ,M. -L . (1987) Enzymatic hydro lys is of c e l l u l o s e : Is the current theory of the mechanisms of hydro-l y s i s va l id? CRC C r i t i c a l Rev. B iotechnol . 5, 67-87. Enger,M.D. and S l e e p e r , B . P . (1965) Mult ip le c e l l u l a s e system from Streptomvces ant ib iot icus . J . B a c t e r i d . 89, 23-27. E r i k s s o n , K . - E . and Pettersson,B. (1975) E x t r a c e l l u l a r enzyme system u t i l i z e d by the fungus Sporotrichum pulverulentum for the breakdown of c e l l u l o s e . I. Preparat ion, p u r i f i c a t i o n and physico-chemical charac te r i za t ion of f ive endo- l ,4 - (3 -glucanases. Eur .J .Biochem. 51, 193-206. (1982) P u r i f i c a t i o n and p a r t i a l charac te r i za t ion of two a c i d i c proteases from the white - rot fungus Sporotr ichum  pulverulentum. Eur .J .Biochem. 124, 635-642. E r n s t , J . F . , Mermod , J . - J . , DeLamarter , J .F . , M a t t a l i a n o , R . J . and Moonen,P. (1987) 0 -g l ycosy la t ion and novel processing events during secret ion of alpha-factor/GM-CSF fusions by Saccharomvces c e r e v i s i a e . Bio/Technology 5, 831-834. Fagerstam,L.G. and Pet te rsson ,L .G . (1979) The c e l l u l o l y t i c complex of Tr ichoderma reese i QM9414. An immunochemical approach. FEBS L e t t . 98, 363-367. ( 1980) The 1,4-<5-gl ucan eel lobiohydrolases of Tr ichoderma reese i QM9414. A new type of c e l l u l o l y t i c synergism. FEBS L e t t . 119, 97-100. Fagerstam,L .G . , Pe t te rsson ,L .G . and Engstrom,J.A. (1984) The primary structure of a 1,4-f5-gl ucan eel lobiohydrolase from the fungus Trichoderma reesei QM9414. FEBS L e t t . 167, 309-315. Fol tmann,B. , S z e c s i , P . B . and Tarasova,N. I . (1985) Detect ion of proteases by c l o t t i n g of casein a f t e r gel e lec t rophores i s . Anal.Biochem. 146, 353-360. Forsberg,C.W. , Bever idge ,T . J . and Hel ls t rom,A. (1981) Ce l lu lase and xylanase release from Bactero ides succinogenes and i t s importance in the rumen environment. Appl . Environ. M i c r o b i o l . 42, 886-896. Fukumori,F. Kudo,T. and Hor ikoshi ,K . (1985) P u r i f i c a t i o n and propert ies of a c e l l u l a s e from a l k a l o p h i l i c B a c i l l u s sp. No. 1139. J . G e n . M i c r o b i o l . 131, 3339-3345. Fukumori , . F . , Sash ihara ,N . , Kudo,T. and Hor ikoshi ,K . ( 1986) Nucleotide sequences of two c e l l u l a s e genes from a l k a l o -p h i l i c Baci11 us sp s t r a i n N-4 and the i r strong homology. J . B a c t e r i o l . 168, 479-485. 113 Gardner ,R.M. , Doerner,K.C. and White,B.A. (1987) P u r i f i c a t i o n and charac te r i za t ion of an exo- f l -1 ,4-g lucanase from Ruminococcus f lavefac iens FD-1 . J . B a c t e r i o l . 169, 4581-4588. Gaudet,G. and G a i l l a r d , B . (1987) Ves ic le formation and c e l l u l o s e degradation in Bactero ides succ inogenes c u l t u r e s : u l t r a -s t ruc tura l aspects . Arch. M i c r o b i o l . 148, 150-154. Ghangas,G.S. and Wi lson,D.B. (1987) Expression of a Thermomono- spora fusca c e l l u l a s e gene in Streptomvces 1i v idans and Bac i11 us subt i1 i s . App l .Env i ron .M ic rob io l . 53, 1470-1475. G i l k e s , N . R . , K i l b u r n , D . G . , Langsford ,M.L . , M i l l e r , R . C . , J r . , Wakarchuk,W.W., War ren ,R .A . J . , W h i t t l e , D . J . and Wong,W.K.R. (1984a) I so la t ion and charac te r i za t ion of Escher ich ia c o l i clones expressing c e l l u l a s e genes from Ce11ulomonas f i m i . J . G e n . M i c r o b i o l . 130, 1377-1384. G i l k e s , N . R . , Langsford ,M.L . , K i l b u r n , D . G . , M i 1 l e r , R . C . , J r . , and Warren,R.A.J . (1984b) Mode of act ion and substrate s p e c i f i c i t i e s of c e l l u l a s e s from cloned b a c t e r i a l genes. J . Biol .Chem. 259, 10455-10459. Greenberg,N.M., War ren ,R .A . J . , K i lbu rn ,D .G . and M i 1 l e r , R . C . , J r . (1987a) Regulat ion, i n i t i a t i o n , and termination of the cenA and ce x t ransc r ip ts of Ce11uIomonas f i m i . J . B a c t e r i o l . 169, 646-653. (1987b) Regulation and i n i t i a t i o n of cenB t ransc r ip t s of Cellulomonas f i m i . J . B a c t e r i o l . 169, 4674-4677. G r i f f i n , H . , D i n t z i s , F . R . , K r u l l , L . , and B a k e r , F . L . (1984) A m i c r o f i b r i l generating factor from the enzyme complex of Trichoderma r e e s e i . B iotechnol .B ioeng. 26, 296-300. G r i f f i n , H . L . , F r e e r , S . N . and Greene,R.V. (1987) E x t r a c e l l u l a r endoglucanase a c t i v i t y by a novel bacterium iso lated from marine shipworm. Biochem.Biophys.Res.Comm. 144, 143-151. Groleau,D. and Forsberg,C.W. (1981) C e l l u l o l y t i c a c t i v i t y of the rumen bacterium Bactero ides succinogenes. Can.J . M i c r o b i o l . 27, 517-530. Groleau,D. and Forsberg,C.W. (1983) P a r t i a l charac te r i za t ion of the e x t r a c e l l u l a r carboxymethylce11ulase a c t i v i t y produced by the rumen bacterium Bactero ides succ inogenes. Can.J . M i c r o b i o l . 29, 504-517. G u i l i a n , G . I s o e l e c t r i c focusing of mouse asc i tes f l u i d and myosin in the SE 600 v e r t i c a l s lab un i t . Hoefer 1986 Catalogue, pp 108-111. 114 Gum,E .K . , J r . and Brown,R.D. , J r . (1976) St ructura l c h a r a c t e r i z a -t ion of a g lycoprote in c e l l u l a s e , 1,4-ft -D-glucan c e l l o b i o -hydrolase C from Tr i choderma v i r ide . Biochim.Biophys.Acta 446, 371-386. (1977) Comparison of four p u r i f i e d e x t r a c e l l u l a r 1,4- f l -D-glucan eel lobiohydrolase enzymes from Tr ichoderma v i r ide. Biochim.Biophys.Acta 492, 225-231. Guo ,Z .M. , Arfman,N. , Ong ,E . , G i Ikes ,N .R . , K i l b u r n , D . G . , Warren, R . A . J . , and M i 1 I e r , R . C . , J r . ( in press) Leakage of Cellulomonas f i m i c e l l u l a s e s from Escher ich ia c o 1 i . FEMS L e t t e r s . Hagerdahl ,B .G.R. , Ferchak,J .D . and Pye ,E .K . (1978) C e l l u l o l y t i c enzyme system of Thermoact inomvees sp. grown on micro-c r y s t a l l i n e c e l l u l o s e . App l .Env i ron .M ic rob io l . 36, 606-612. Haggett ,K .D. , Gray ,P .P . and Dunn,N.W. (1979) C r y s t a l l i n e c e l l u l o s e degradation by a s t r a i n of Cellulomonas and i ts mutant d e r i v a t i v e s . E u r . J . A p p l . M i c r o b i o l . B i o t e c h n o l . 8, 183-190. Hayashida,S. and Mo,K. (1986) Production and c h a r a c t e r i s t i c s of Av i ce1 -d i s in teg ra t ing endoglucanase from a protease-negative Humicola qr i sea var. thermo idea mutant. Appl . Env i ron.Microbio1. 51, 1041-1046. H e n r i s s a t , B . , D r iguez ,H . , V i e t , C . and Schule in ,M. (1985) Synergism of c e l l u l a s e s from Tr ichoderma reese i in the degradation of c e l l u l o s e . Bio/Technology 3, 83-87. Hon-nami,K. , Coughlan,M.P. ,Hon-nami,H. and L jungdahl ,L .G . (1986a) Separation and charac te r i za t ion of the complexes cons t i tu t ing the c e l l u l o l y t i c enzyme system of C lostr id ium  thermoce Hum. A r c h . M i c r o b i o l . 145, 13-19. Hon-nami,K. , Coughlan,M.P. ,Hon-nami,H. , C a r r e i r a , L . H . and L jungdahl ,L .G . (1986b) Propert ies of the c e l l u l o l y t i c enzyme system of C lostr id ium thermoce Hum. Biotechnol.Bioeng.Symp. 15, 191-205. H o r i k o s h i , K . , Nakao,M., Kurono,Y. , and Sashihara,N. (1984) Ce l lu lases of an a l k a l o p h i l i c Baci11 us s t r a i n i so la ted from s o i l . C a n . J . M i c r o b i o l . 30, 774-779. Hubbard,S.C. , and I v a t t , R . J . (1981) Synthesis and processing of asparagine-1inked o l igosacchar ides . Ann.Rev.Biochem. 50, 555-583. Hughes,R.C. ( 1983) Glvcoprote ins . Chapman and Hal l L t d . , London 115 Hunkapi l lar ,M.W. , Hewick,R.M., Dreyer ,W.J . , and Hood,L .E . (1983) High s e n s i t i v i t y sequencing with a gas phase sequenator. Methods Enzymol. 91, 399-413. Jansson, P . - E . , Kenne,L . , L. iedgren, H . , L indberg, B. , and Lonngren,J . (1976) A p r a c t i c a l guide to the methylation ana lys is of carbohydrates. Chem.Commun. 8. Johnson,E .A . , Sakajoh,M. , Hal1iwe 1 1 , G . , Madia ,A . , and Demain, A . L . (1982) S a c c h a r i f i c a t i o n of complex c e l l u l o s i c substrates by the c e l l u l a s e system from Clostr id ium  thermoce Hum. Appl .Env i ron .Microb io1. 43, 1 125-1 132. J o l i f f , G . , Beguin,P. and A u b e r t , J . - P . (1986a) Nucleotide sequence of the c e l l u l a s e gene ceID encoding endoglucanase D of C lost r id ium thermoce11 urn. Nuc l .A .Res . 14, 8605-8613. J o l i f f , G . , Begu in ,P . , Juy ,M. , M i l l e t , J . , R y t e r , A . , Po l jak ,R . and Aubert, J . - P . (1986b) I s o l a t i o n , c r y s t a l l i s a t i o n and propert ies of a new c e l l u l a s e of C lostr id ium thermoce11 urn overproduced in Escher ich ia c o l i . Bio/Technol. 4, 896-900. J o v i n , T . , Chrambach,A. and Naughton,M.A. (1964) An apparatus for temperature-regulated polyacrylamide gel e lec t rophores i s . Anal.Biochem. 9, 351-369. Kammel,W.P. and Kubicek,C .P . (1985) Absence of postsecret ional modif icat ion in e x t r a c e l l u l a r proteins of Trichoderma  reesei during growth on c e l l u l o s e . J .Appl .Biochem. 7, 138-144. K l yosov ,A .A . , M i t k e v i c h , 0 . V . , and S i n i t s y n , A . P . (1986) Role of the a c t i v i t y and adsorption of c e l l u l a s e s in the e f f i c i e n c y of the enzymatic hydro lys is of amorphous and c r y s t a l l i n e c e l l u l o s e . Biochemistry. 25, 540-542. Knowles , J . , Lehtovaara,P. and T e e r i , T . (1987) Ce l lu lase fami l ies and the i r genes. Trends B iotechnol . 5, 255-261. Kornfeld ,R. and Kornfe ld ,S . (1978) Use of l e c t i n s in the study of mammalian g lycoprote ins , pp 437-449 in The Glvco- coni ugates 11: Mammalian Glycoprote ins , G l y c o l i p i d s and  Proteoglyeans. Horowitz, M.I. and Pigman,W., eds. Academic Press , Inc . , London. Kornfe ld ,R. and Korn fe ld ,S . (1980) 'Structure of g lycoproteins and the i r o l igosacchar ide u n i t s , in The Biochemistry of  Glycoproteins and Proteoglycans. W.J.Lennarz, ed. Plenum Press, N.Y. and London. Kornfe ld ,R. and Korn fe ld ,S . (1985) Assembly of asparagine-l inked o l igosacchar ides . Ann.Rev.Biochem. 54, 631-664. 116 K u b i c e k , C P . (1981) Release of carboxymethyl-ce11ulase and f3-gl ucos idase from c e l l walls of Tr ichoderma reese i . E u r . J . A p p l . M i c r o b i o l . B i o t e c h n o l . 13, 226-231. K u b i c e k , C . P . , Panda,T . , S c h r e f e r l - K u n a r , G . , Gruber ,F . and Messner,R. (1987) O- l inked but not N- l inked g l ycosy la t ion is necessary for the secret ion of endoglucanases I and II by Trichoderma r e e s e i . C a n . J . M i c r o b i o l . 33, 698-703. Kukuruzinska,M.A. , Bergh,M.L .E . and Jackson ,B . J . (1987) Protein g l ycosy la t ion in yeast. Ann.Rev.Biochem. 56, 915-944. Kupcu,Z . , Marz , L . ,Messner ,P . , and S l e y t r , U . B . (1984) Evidence for the g lycoprote in nature of the c r y s t a l l i n e c e l l wall surface layer of Bac i11 us stearothermophi1 us s t r a i n NRS2004/3a. FEBS L e t t . 173, 185-190. Labudova,I. and Farkas,V . (1983) Mult ip le enzyme forms in the c e l l u l a s e system of Trichoderma r e e s e i . Biochim.Biophys. Acta 744, 135-140. Laemmli,U.K. (1970) Cleavage of s t ruc tura l proteins during assembly of the head of bacteriophage T4. Nature, London 277, 680-685. Lamed,R. and Bayer ,E .A . ( in press) The cellulosome concept: e x o c e l l u l a r / e x t r a c e l l u l a r enzyme reactor centers for e f f i c i e n t binding and c e 1 1 u l o l y s i s . in Biochemistry and  Genetics of Ce l lu lose Degradation. Proceedings of the FEMS Symposium, P a r i s , September, 1987. A u b e r t , J . - P . , Beguin,P. and M i l l e t , J . , e d s . , Academic Press , Inc . , London. Lamed,R., Na imark , J . , Morgenstern,E. and Bayer ,E .A . (1987) Spec ia l i zed c e l l surface st ructures in c e l l u l o l y t i c b a c t e r i a . J . B a c t e r i d . 169, 3792-3800. Lamed,R., S e t t e r , E . , and Bayer ,E .A . (1983a) Character i zat ion of a c e l l u l o s e - b i n d i n g , ce11ulase-contain ing complex in CI ostr id i um thermoce Hum. J . B a c t e r i d . 156, 828-836. Lamed,R., S e t t e r , E . , Ken ig ,R . , and Bayer ,E .A . (1983b) The ce l lu losome- -A d iscrete c e l l surface organelle of CIostr i d i um thermoce Hum which exh ib i ts separate ant igen ic , ce11ulose-binding and various c e l l u l o l y t i c a c t i v i t i e s . Biotechnol.Bioeng.Symp. 13, 163-181. Langsford ,M.L . , G i l k e s , N . R . , S i n g h , B . , Moser ,B . , M i l l e r , R . C J r . , Warren,R.A.J . , and K i lbu rn ,D .G . ( 1987) G lycosy lat ion of b a c t e r i a l c e l l u l a s e s prevents p ro teo l y t i c cleavage between funct ional domains. FEBS L e t t . 225, 163-167. 117 Langsford ,M.L . , G i l k e s , N . R . , Wakarchuk,W.W., K i l b u r n , D . G . , M i l l e r , R . C . , J r . , and Warren,R.A.J . (1984) The c e l l u l a s e system of Cellulomonas f imi . J . Gen .Mic rob io l . 130, 1367-1376. L a u , F . T . - K . , a n d Fersht ,A .R . (1987) Conversion of a l l o s t e r i c i n h i b i t i o n to a c t i v a t i o n in phosphofructokinase by prote in engineer ing. Nature 326, 811-812. L e h l e , L . and Bause,E. (1984) Primary s t ruc tura l requirements for N- and 0 -g l ycosy la t ion of yeast mannoproteins. Biochim. Biophys.Acta 799, 246-251. Lehnhardt ,W.F. , and Winz le r ,R . J . (1968) Determination of neutral sugars in g lycoproteins by g a s - l i q u i d chromatography. J . Chromatography 34, 471-479. Lever,M. (1972) A new react ion for co lo r imet r i c determination of carbohydrates. Anal.Biochem. 47, 273-279. Lowry ,0 .H. , Rosebrough,N.J . , F a r r , A . L . , and R a n d a l l , R . J . (1951) Protein measurement with the F o l i n phenol reagent. J . Biol .Chem. 193, 265-275. L jungdah l , L .G . , P e t t e r s s o n , B . , E r i k s s o n , K . - E . and Wiege l , J . (1983) A yellow a f f i n i t y substance involved in the c e l l u l o l y t i c system of C lostr id ium thermoce11um. Curr. M i c r o b i o l . 9, 195-200. Mackenzie ,C .R. , B i l o u s , D . and Johnson,K.G. (1984) P u r i f i c a t i o n and charac te r i za t ion of an exoglucanase from Streotomvces  f lavogr i seus . C a n . J . M i c r o b i o l . 30, 1171-1178. MacKenzie,C.R. , B i l o u s , D . and P a t e l , G . B . (1985) Studies on c e l l u l o s e hydro lys is by A c e t i v i b r i o ce11u lo l v t i cus . App l .Env i ron .M ic rob io l . 50, 243-248. MacKenzie,C.R. , P a t e l , G . B . and B i l o u s , D . (1987) Factors involved in hydro lys is of microcrystal1ine c e l l u l o s e by Ace t i v i br i o  ce11ulolvt i cus . A p p l . E n v i r o n . M i c r o b i o l . 53, 304-308. Maeba,P.Y. (1986) I so lat ion of a surface g lycoprote in from Mvxococcus xanthus. J . B a c t e r i o l . 166, 644-650. Mandels,M. (1983) Appl icat ions of c e l l u l a s e s . Biochem.Soc. Trans. 13, 414-416. Maniatus ,T . , F r i t s c h , E . F . , and Sambrook,J. (1982) Molecular c lon ing : A laboratory manual. Cold Spring Harbour Laboratory, Cold Spring Harbour, New York. McCarthy,A.J . (1987) Lignoce11u1ose-degrading actinomycetes. FEMS Microb io l .Rev . 46, 145-163. 118 Meacock,P.A. and Cohen,S.N. (1980) P a r t i t i o n i n g of b a c t e r i a l plasmids during c e l l d i v i s i o n : a c i s - a c t ing locus that accomplishes stable plasmid inher i tance. Ce l l 20, 529-542. M e r i v u o r i , H . , Sands , J .A . and Montenecourt,B.S. (1985a) E f fec ts of tunicamycin on secret ion and enzymatic a c t i v i t i e s of c e l l u l a s e from Tr ichoderma reese i . A p p l . M i c r o b i o l . B iotechnol . 23, 60-66. M e r i v u o r i , H . , S i e g l e r , K . M . , Sands , J .A . and Montenecourt,B.S. (1985b) Regulation of c e l l u l a s e biosynthesis and secret ion in fung i . Biochemical Soc .Trans . 13, 411-414. M i l l e r , G . L . (1959). Use of d i n i t r o s a l i c y l i c ac id reagent for determination of reducing sugars. Anal.Chem. 31, 426-428. M i l l e r , J . H . (1972) Experiments in Molecular Genet ics . Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. M i l l e t , J . , P e t r e , D . , Begu in ,P . , Raynaud,0. and A u b e r t , J . - P . (1985) Cloning of ten d i s t i n c t DNA fragments of Clostr id ium  thermocellum coding for c e l l u l a s e s . FEMS M i c r o b i o l . L e t t . 29, 145-149. Moloney,A. and Coughlan,M.P. (1981) Sorpt ion of Talaromvces emersonii c e l l u l a s e on c e l l u l o s i c subst rates . B iotechnol . Bioeng. 25, 271-280. Montenecourt,B.S. ( 1983) Tr ichoderma reese i c e l l u l a s e s . Trends B iotechnol . 1, 156-160. M o r r i s , E . J . and C o l e , O . J . (1987) Relat ionship between c e l l u l o l y t i c a c t i v i t y and adhesion to c e l l u l o s e in Ruminococcus a lbus . J . G e n . M i c r o b i o l . 133, 1023-1032. Mor r i ssey , J .H . (1981) S i l v e r Sta in for proteins in polyacrylamide g e l s : A modified procedure with enhanced uniform s e n s i t i v i t y . Anal.Biochem. 117, 307-310. Moskophidis,M, and M u l l e r , F . (1984) Molecular charac te r i za t ion of g lycoprote in antigens on surface of Treponema pal 1idum: comparison with nonpathogenic Treponema phagedenis biotype Re i te r . Infect.Immun. 46, 867-869. Mul l ings ,R . (1985) Measurement of s a c c h a r i f i c a t i o n by c e l l u l a s e s . Enzyme Microb.Technol . 7, 586-591. Nakamura,K., Misawa,N., and Kitamura,K. (1986) Ce l lu lase genes of Cellulomonas uda CB4. II. Cloning and expression of a CM-cel lu lose hydrolyzing enzyme (endoglucanase) gene in Escher ich ia c o l i . J .B iotechnol . 3, 247-253. 119 Nakayama,M., Tomita ,Y . , S u z u k i , H . , and Nisizawa,K. (1976) P a r t i a l p ro teo lys is of some c e l l u l a s e components from Tr ichoderma v i r ide and the substrate s p e c i f i c i t y of the modified products. J .Biochem. 79, 955-966. Neeser , J . -R . and Schweizer ,T .F . (1984) A quant i tat ive determin-at ion by c a p i l l a r y g a s - l i q u i d chromatography of neutral and amino sugars (as O-methyloxime aceta tes ) , and a study on hydro l y t i c condit ions for g lycoproteins and po ly -saccharides in order to increase sugar recover ies . Anal.Biochem. 142, 58-67. Ng,T.K. and Z e i k u s , J . G . (1981) P u r i f i c a t i o n and charac te r i za t ion of an endoglucanase (1, 4-(5-D-glucan glucanohydrolase) from Clostr id ium thermoce11 urn. Biochem.J. 199, 341-350. Niedermeier,W. (1971) Gas chromatography of neutral and amino sugars in g lycoprote ins . Anal.Biochem. 40, 465-475. N i k u - P a a v o l a , M . - L . , Lappala inen ,A . , E n a r i , T . - M . and Nummi,M. (1985) A new appra isa l of the endoglucanases of the fungus Trichoderma r e e s e i . Biochem.J. 231, 75-81. Ohmiya,K. Shimizu ,M. , Taya,M. , and Shimizu,S. (1982) P u r i f i c a t i o n and propert ies of ce11obiosidase from Ruminococcus a lbus . J . B a c t e r i o l . 150, 407-409. O ' N e i l l , G . P . (1986) Structure of the gene encoding the exoglucanase of Ce11u1omonas f i m i and i t s expression and secret ion in Escher ich ia c o l i . Ph.D. Thes is , Un ivers i ty of B r i t i s h Columbia, Vancouver, Canada. O ' N e i l l , G . P . , Goh .S .H . , War ren ,R .A . J . , K i l b u r n , D . G . , and M i l l e r , R . C . , J r . (1986a) Structure of the gene encoding the exoglucanase of Ce11ulomonas f i m i . Gene 44, 325-330. O 'Ne i11 ,G .P . , K i l b u r n , D . G . , Warren,R.A.J . and M i 1 l e r , R . C . , J r . (1986b) Overproduction of a c e l l u l a s e gene with a high guanosine-p lus -cytos ine content in Escher ich ia c o l i . Appl . Env i ron .Mic rob io l . 52, 743-757. Owa lab i , J .B . , Begu in ,P . , K i l b u r n , D . G . , M i 1 l e r , R . C . , J r . and Warren,R.A.J . ( in press) The s t ruc tura l gene for endo-glucanase B of Ce11u1omonas f i m i and i t s expression in Escher ich ia c o l i . App l .Env i ron .M ic rob io l . P a i c e , M . G . , Desrochers ,M. , Rho,D. , J u r a s e k , L . , R o y , C , R o l l i n , C . F . , De M i g u e l , E . , and Yaguchi,M. (1984) Two forms of endoglucanase from the basidiomycete Schizophvl1 urn commune and the i r r e l a t i o n s h i p to other R-1,4-glycoside hydrolases. Bio/Technology 1, 535-539. 120 Pantol iano,M.W., L a d n e r . R . C , B ryan ,P .N . , R o l l e n c e , M . L . , Wood, J . F . and P o u l o s , T . L . (1987) Prote in engineering of s u b t i l i s i n BPN': Enhanced s t a b i l i z a t i o n through the introduct ion of two cyste ines to form a d i s u l f i d e bond. Biochemistry. 26, 2077-2081. P a r k , C M . (1984) Comparison of the carbohydrate composition of alphas-macroglobul in from cont ro ls and pat ients with Cyst ic F i b r o s i s . M.Sc. Thesis , Un ivers i ty of B r i t i s h Columbia, Vancouver, Canada. P e n t t i l a , M . , Lehtovaara,P. Nevala inen,H. , Bhikhabhai,R. and Knowles,J . (1986) Homology between c e l l u l a s e genes of Tr ichoderma reese i : complete nucleotide sequence of the endoglucanase I gene. Gene 45, 253-263. P e t r e , D . , M i l l e t , J . , Long in ,B . , Begu in ,P . , G i ra rd ,H . and Aubert, J . - P . (1986). P u r i f i c a t i o n and propert ies of the endoglucanase C of C lostr id ium thermoce Hum produced in Escher ich ia c o l i . Biochimie 68, 687-695. P e t r e , J . , Long in ,R . , and M i l l e t , J . (1981) P u r i f i c a t i o n and propert ies of an endo-ft -1,4-glucanase from Clostr id ium  thermoce Hum. Biochimie 63, 629-639. P l a u t , A . G . , G i b e r t , J . V . , A r tenste in ,M.S . and Capra , J .D . (1975) Ne i s s e r i a gonorrhoeae and Ne i sser ia men ing i t idus e x t r a c e l l u l a r enzyme cleaves human Immunoglobulin A. Science 190, 1103-1105. Praser tsan ,P . and Doelle,H.W. (1986) Separation and charac te r i za t ion of endoglucanases from cul ture f i l t r a t e s of Cellulomonas sp. A p p l . M i c r o b i o l . B i o t e c h n o l . 24, 326-333. Ramasamy,K. and Verachter t ,H . (1980) L o c a l i z a t i o n of c e l l u l a s e components in Pseudomonas sp. i so lated from act ivated sludge. J . G e n . M i c r o b i o l . 117, 181-191. Rao,M. , M i s h r a , C , Gaikwad,S. , and Sr i n i vasan, M . C . ( 1986) Release of c e l l u l a s e s from Pen ic i I l ium i anth ine Hum. Biotechnol .B ioeng. 28, 129-132. Rinderknecht ,H. , G e o k a s , M . C , Si 1verman,P., and Haverback,B.J . (1968) A new u l t r a s e n s i t i v e method for the determination of p r o t e o l y t i c a c t i v i t y . CI in .Chim.Acta 21, 197-203. Robson,L.M. and Chambliss,G.H. (1984) Character i zat ion of the c e l l u l o l y t i c a c t i v i t y of a Baci11 us i s o l a t e . Appl .Env i ron . M i c r o b i o l . 47, 1039-1046. ( 1986) Cloning of the Baci11 us subt i1 i s DLG ft-1,4-glucanase gene and i ts expression in Escher ich ia c o l i and B.subt i l l s . J . B a c t e r i o l . 165, 612-619. 121 ( 1987) Endo-jS-1, 4-glucanase gene of Baci 11 us subt i l l s DLG. J . B a c t e r i o l . 169, 2017-2025. Ro i t sch ,C .A . and Hageman,J.H. (1983) Baci11opeptidase F: Two forms of a g lycoprote in serine protease from Bac i11 us  s u b t i l is 168. J . B a c t e r i o l . 155, 145-152. RomaniecM.P.M. , C la rke ,N .G . and Hazlewood,G.P. (1987) Molecular c lon ing of C lostr id ium thermoce11 urn DNA and the expression of further novel endo-ft -1,4-glucanase genes in Escher ich ia  c o l i . J . G e n . M i c r o b i o l . 133, 1297-1307. Rudick ,M.J . and Elbe in ,A .D . ( 1975) Glycoprotein enzymes secreted by Aspergi11 us furnigatus: p u r i f i c a t i o n and propert ies of a second ft-glucosidase. J . B a c t e r i o l . 124, 534-541. S a d d l e r , J . N . and Khan,A.W. (1981) C e l l u l o l y t i c enzyme system of A c e t i v i b r i o c e l l u l o l v t i c u s . C a n . J . M i c r o b i o l . 27, 288-294. S a d d l e r , J . N . , Khan,A.W., and Mart in ,S .M. (1980) Regulation of c e l l u l a s e synthesis in A c e t i v i b r i o ce11u lo l v t i cus . Microbios 28, 97-106. S a l o v u o r i , I . , Makarow,M., Rauvala ,H . , Knowles , J . , and Kaar ia inen ,L . (1987) Low molecular weight high-mannose type glycans in a secreted prote in of the fi lamentous fungus Trichoderma r e e s e i . Bio/Technology 5, 152-156. Sancar ,A . , Hack,A. and Rupp,W.D. (1979) A simple method for i d e n t i f i c a t i o n of plasmid coded p ro te ins . J . B a c t e r i o l . 137, 692-693. Schachter ,H. and Roseman,S. (1980) Mammalian g l y c o s y l -t ransferases : Their role in the synthesis and funct ion of complex carbohydrates and g l y c o l i p i d s . in The Biochemistry  of Glycoproteins and Proteoglycans, pp 85-160. Lennarz, W. J . , ed. Plenum Press, New York and London. S c h e l l , M . A . (1987) P u r i f i c a t i o n and charac te r i za t ion of an endoglucanase from Pseudomonas solanacearum. Appl . Env i ron .M ic rob io l . 53, 2237-2241. Sche l lho rn ,H .E . and Forsberg,C.W. (1984) M u l t i p l i c i t y of ex t ra -c e l l u l a r <!-( 1, 4)-endoglucanases of Bacteroides succinogenes S85. C a n . J . M i c r o b i o l . 30, 930-937. Schmuck,M., P i l z , I . , Hayn,M. and Esterbauer ,.H. ( 1986) Invest igat ion of ce11obiohydrolase from Trichoderma reesei by small angle X-ray s c a t t e r i n g . B iotechno1.Lett . 8, 397-402. 122 Schne ide r ,C . , Newman,R.A., Suther land,D.R . , A s s e r , U . , and Greaves,M.F. (1982) A one-step p u r i f i c a t i o n of membrane proteins using a high e f f i c i e n c y immunomatrix. J .B io l .Chem. 257, 10766-10769. Scopes,R.K. (1982) Protein P u r i f i c a t i o n : P r i n c i p l e s and P r a c t i c e . Spr inger -Ver lag New York Inc. Sharon,N. and L i s , H . (1982) Glycoprote ins . in The Proteins 5, 1-144. Neurath,H. and H i l l , R . L . , eds. Academic Press , Inc . S h e i r - N e i s s , G . and Montenecourt,B.S. (1984) Character i zat ion of the secreted c e l l u l a s e s of Tr ichoderma reese i wi ld type and mutants during cont ro l l ed fermentations. Appl . M i c r o b i o l . B i o t e c h n o l . 20, 46-53. Shoemaker,S. and Brown,J r . ,R .D. (1978) Character izat ion of endo-1,4- f l -D-g lucanases p u r i f i e d from Tr ichoderma v i r i d e . Biochim.Biophys. Acta 253, 147-161. Shoemaker,S., Schweickart ,V. Ladner ,M. , Ge l fand ,D . , Kwok,S., Myambo,K., and Innis ,M. (1983) Molecular c lon ing of exo-ce11 obiohydrolase I derived from Tr ichoderma reese i s t r a i n L27. Bio/Technology 1, 691-696. S r i vas tava ,R .A .K : (1984) Studies on e x t r a c e l l u l a r and i n t r a -c e l l u l a r p u r i f i e d amylases from a thermophil ic B a c i l l u s  stearothermophilus. Enzyme Microb.Technol . 6, 422-426. S t a c k , R . J . and Hungate,R.E. (1984) E f fec t of 3-Phenylpropanoic ac id on capsule and c e l l u l a s e s of Ruminococcus albus 8. A p p l . E n v i r o n . M i c r o b i o l . 48, 218-223. S t e w a r t , B . J . , and Leatherwood,J.M. (1976) Derepressed synthesis of c e l l u l a s e by Cel lulomonas. J . B a c t e r i d . 128, 609-615. Stoppok,W., Rapp,P . , and Wagner,F. (1982) Formation, l o c a t i o n , and regulat ion of endo-1,4- f l -g lucanases and fl-glucosidases from Cellulomonas uda. App l .Env i ron .M ic rob io l . 44, 44-53. S t r o b e l , G . A . , Tamadge,K.W. and Albersheim,P. (1972) Observations on the structure of the phytotoxic g l yco -peptide of Corvnebacter i um sepedon icum. Biochim.Biophys. Acta 261, 365-374. Sumper,M. (1987) Halobacter ia l g lycoprote in b iosynthes is . Biochim.Biophys. Acta 906, 69-79. Suzuk i ,Y . , O i s h i , K . , Nakano,H. and Nagayama,T. (1987) A strong c o r r e l a t i o n between the increase in number of pro l ine residues and the r i s e in thermostabi l i t y of f ive Baci11 us o l i g o - 1 , 6 - g l u c o s i d a s e s . App l .Mic rob io l .B io techno1. 26, 546-551. 123 T e e r i , T . T . (1987) The c e l l u l o l y t i c enzyme system of Trichoderma  reese i : Molecular c lon ing , charac te r i za t ion and expression of the ce l lob iohydro lase genes. Technical Research Centre of F in land , Publ icat ions 38. T e e r i , T . T . , Lehtovaara ,P . , Kauppinen,S. , S a l o v u o r i , I . and Knowles,J . (1987) Homologous domains in Trichoderma reesei c e l l u l o l y t i c enzymes: gene sequence and expression of ce l lob iohydro lase II. Gene 51, 43-52. T e e r i , T . , S a l o v u o r i , I . and Knowles,J. (1983) The molecular c loning of the major c e l l u l a s e gene from Tr ichoderma reese i . Bio/Technology 1, 696-699. Towbin,H., S t a e h e l i n , T . , and Gordon,J . (1979) E lec t rophoret i c t ransfer of proteins from polyacrylamide gels to n i t r o -c e l l u l o s e sheets: procedure and some a p p l i c a t i o n s . Proc. N a t l . Acad. S c i . USA 76, 4350-4354. van T i l b e u r g h , H . , Tomme,P., C laeyssens ,M. , Bhikhabhai,R. and Pet tersson,G. (1986) Limited p ro teo lys i s of the c e l l o -biohydrolase I from Tr ichoderma reese i . Separation of funct ional domains. FEBS L e t t . 204, 223-227. Wandersman,C., Andro,T. and Bertheau,Y. (1986) E x t r a c e l l u l a r proteases in Erwinia chrvsanthemi. J . G e n . M i c r o b i o l . 132, 899-906. War ren ,R .A . J . , B e c k , C . F . , G i l k e s , N . R . , K i l b u r n , D . G . , Langsford, M.L . , M i l l e r , R . C . J r . , 0 ' N e i 1 1 , G . P . , Scheufens,M. and Wong, W.K.R. (1986) Sequence conservation and region s h u f f l i n g in an endoglucanase and an exoglucanase from Ce11ulomonas  f imi . P ro te ins : St ructure , Function and Genetics 1, 335-341. War ren ,R .A . J . , Gerhard,B. G i l k e s , N . R . , Owo lab i , J .B . , K i lburn , D.G. and M i 1 l e r , R . C . , J r . ( in press) A b i func t iona l exoglucanase-endoglucanase fusion p ro te in . Gene. Waterbury , J .B . , C a l l o w a y , C B . and Turner ,R.D. (1983) A c e l l u l o l y t i c n i t r o g e n - f i x i n g bacterium cul tured from the Gland of Deshayes in shipworms ( B i v a l v i a : Teredin idae) . Science 221, 1401-1403. Weerkamp,A.H. and Jacobs,T . (1982) Ce l l wa l l -assoc ia ted prote in antigens of Streptococcus s a l i v a r i u s : P u r i f i c a t i o n , p roper t ies , and funct ion in adherence. Infect.Immun. 38, 233-242. Weerkamp,A.H., van der Mei ,H.C . and L iem,R .S .B . (1986) St ructura l propert ies of f i b r i l l a r proteins i so lated from the c e l l surface and cytoplasm of Streptococcus s a l i var i us (K+) c e l l s and nonadhesive mutants. J . B a c t e r i o l . 165, 756-762. 124 W h i t t l e , D . J . , K i l b u r n , D . G . , Warren,R.A.J . and M i 1 I e r , R . C . , J r . ( 1982) Molecular c lon ing of a Cellulomonas f imi c e l l u l a s e gene in Escher ich ia c o l i . Gene 17, 139-145. W i l l i c k , G . E . and S e l i g y , V . L . (1985) M u l t i p l i c i t y in c e l l u l a s e s of Schizophvllum commune: Der ivat ion par t l y from heterogeneity in t r a n s c r i p t i o n and g l y c o s y l a t i o n . Eur .J .Biochem. 151, 89-96. Winter ,A . , E k , K . , and Andersson, U-B. (1977) Ana l y t i ca l e lec t ro focus ing in th in layers of polyacrylamide g e l s . LKB App l i cat ion Mote 250. W o l f f , B . R . , Mudry ,T .A. , G l i c k , B . R . and P a s t e r n a k , J . J . (1986) I so la t ion of endoglucanase genes from Pseudomonas  f luorescens subsp ce11u1osa and a Pseudomonas sp. Appl . Env i ron .Mic rob io l . 51, 1367-1369. Wong,W.K.R., Gerhard ,B . , Guo ,Z .M. , K i l b u r n , D . G . , War ren ,R .A . J . , and M i l l e r , R . C . , J r . (1986) Character i zat ion and structure of an endoglucanase gene cenA of Ce11ulomonas f imi . Gene 44, 315-324. Wood,T.M. (1985) Propert ies of c e l l u l o l y t i c enzyme systems. Biochem.Soc.Trans. 13, 407-410. Wood,T.M., Wi lson ,C .A . and Stewart ,C .S . (1982) Preparat ion of the c e l l u l a s e from the c e l l u l o l y t i c anaerobic rumen bacterium Ruminococcus albus and i t s release from the b a c t e r i a l c e l l wa l l . Biochem.J. 205, 129-137. Wray,W., B o u l i k a s , T . , Wray,V.P. and Hancock,R. (1981) S i l v e r s ta in ing of proteins in polyacrylamide g e l s . Anal . Biochem. 118, 197-203. Wu,J.H.D. and Demain,A.L. ( in press) Proteins of the CIostr idium  thermocellum c e l l u l a s e complex responsible for degradation of c r y s t a l l i n e c e l l u l o s e , in Biochemsitrv and Genetics of  Ce l lu lose Degradation, Proceedings of the FEMS symposium, P a r i s , September 1987. A u b e r t , J . - P . , Beguin,P. and M i l l e t , J . , e d s . , Academic Press , London. Yamane,K., Suzuki ,H . and Nisizawa,K. (1970) P u r i f i c a t i o n and propert ies of e x t r a c e l l u l a r and ce l l -bound c e l l u l a s e components of Pseudomonas f luorescens var. e e l l u l o s a . J .Biochem. 67, 19-35. Yan isch -Per ron ,C . , V i e i r a , J . and Mess ing , J . (1985) Improved M13 phage c lon ing vectors and host s t r a i n s : nucleot ide sequences of the M13mpl8 and pUC19 vectors . Gene 33, 103-119. Yano>0., Fukuda.T. , Abe,H. and Sudo,T. (1984) P u r i f i c a t i o n and charac te r i za t ion of a tubercu l in -ac t i ve substance from Mvcobacter i um bov i s BCG. J .Biochem. 96, 523-531. 125 Yoshikawa,T., Suzuki ,H. and Nisizawa,K. (1974) Biogenesis of mult iple c e l l u l a s e components of Pseudomonas f luorescens var. c e l l u l o s a . I. E f fec ts of cu l ture condit ions on the m u l t i p l i c i t y of c e l l u l a s e . J .Biochem. 75, 531-540. Zachar ius .R .M. , Z e l l , T . E . , M o r r i s o n , J . H . , and Woodlock,J . J . (1969) Glycoprotein s ta in ing fo l lowing e lect rophores is on acrylamide g e l s . Anal.Biochem. 30, 148-152. Z o l l e r , M . J . and Smith,M. (1983) 01 igonuc leot ide -d i rec ted mutagenesis of DNA fragments cloned in M13-derived vectors . Methods in Enzymology 100 Part B, 468-500 126 Appendix A E . co l i i i t ra in Genet ic characters Ref C600 t h i - 1 leuB6 l a c Y l tonA supE44 PM191 dra drm thr leu th i lacY recA56 supE JM101 S U P E th i ( lac -proAB) [F'traD36 proAB l a c l " Z CSR603 recAl uvrA6 phr1 M151 1 2 3 4 Recombinant Enzyme C . f imi DNA Fragment S ize Ref Plasmid Encoded (kb) pDWl Exg Exg B ih* B 20.2 5 pECl B S 6 . 6 6 pECl .1 Exg B B 2 . 6 7 pUC12::C16.6 Exg St S 6.6 8 pUC12::EBS7 37 Exg St S 1 .9 9 pUC12: :PT IS : : s t y Exg 1.9 9 pcEC2 EngA B B 10 B - B 6.0 pEC2 EngA B P 5.2 10 pEC2.3 EngA B A 3.5 10 pEC2.1 EngA Ss Ss 2.4 10 pUCEC2 EngA B B 1.6 1 1 pEC3 EngB B B 5.6 6 pUC19::C35.6 EngB B B 5.6 12 pEC4 ? B B 5.5 13 pEC5 ? 5.5 13 Ref: 1. Appleyard, 1954; 2. Meacock and Cohen, 1980; 3. Yanisch-Perron et a l . , 1985; 4. Sancar et a l . , 1979; 5. Whitt le et a l . , 1982; 6. G i lkes et a l . , 1984a; 7. O ' N e i l l et a l . , 1986a; 8. O ' N e i l l , personal communication; 9. O ' N e i l l et a l . , 1986b; 10. Wong et a l . , 1986; 11. Guo et a l . , in press ; 12. Owalabi, personal communication; 13. Paradis , personal communication. Notes: pDWl and pEC plasmids are pBR322 d e r i v a t i v e s . pEC plasmids car ry unique Bam HI r e s t r i c t i o n fragments of C . f imi DNA. These r e s t r i c t i o n fragments were recloned or subclonedTnto pUC plasmids as ind i ca ted . pUC12::EBS737. pUC12: :PTIS: :Sty , and pUCEC2 are gene fus ions . R e s t r i c t i o n s i t e s : B = Bam HI, P = Pvu I I , S = S a i l , St = S t y l , Ss = S s t l , A = Ava l . 127 Appendix B S igma Standard Mr X IO- 3 Log Mr Distance From Or ig in (mm) Rr Myos in 205 5. 31 6.0 0. 992 ft-galactos idase 116 5. 06 11.5 0. 190 Phosphorylase b 97. 4 4. 99 14.5 0. 240 Albumin, Bovine 66. 0 4. 82 21.0 0. 347 Albumin, Egg 45. 0 4. 65 29.0 0. 479 Carbon ic Anhydrase 29. 0 4. 46 43.0 0. 711 Dye Front 60.5 EngA 25.5 0. 425 Exg 26.0 0. 433 Dye Front 60.0 Hflllf4IIIT1IHtlf*llltllI11t-IIIHtllllHtltllflIlltTt<lllllt1fMltllllltttttttllltlltt1tltttltltl111lttllltHII*ttttHlllllllttlllllflt1tt-| 200 CO i o X 100-0.2 0 4 0.6 Calcu lat ions using L inear Regression where y = mx + b, y = log Mr, x = R*, m = - 1 . 1 6 7 4 , b =5.256, and omitt ing the myosin standard: Then for EngA, y = 4.7599 or Mr 57,500 Exg, y = 4.7505 or Mr 56,300 128 Appendix C 50 40 30 oo 20 CO •o •r— CO <D Q; 4-O $-CD JD 50 40 30 — • A — • • — 1 1 1 1 A— — B A 1 1 1 1 20 40 60 80 H y d r o l y s i s Time (Hour s ) Determination of serine and threonine from the amino ac id a n a l y s i s . Ser and Thr are determined by ext rapolat ing to zero time from the values obtained a f t e r hydrolys is for 24, 48, and 96 hours. Panel A: EngA. Panel B: Exg. (•) Thr; ( a ) se r . 129 Appendix D 2 1 u 3 1 4 A [ A 5 B 5 C { 1 D rhrJ r+ LJ ^ 5 10 15 Retention Time (min) GLC chromatographs of a l d i t o l acetates . EngA and Exg were hydrolysed in 4M TFA to release monosaccharides. Inos i to l was added as the internal standard a f t e r the hydrolysate cooled to room temperature. The monosaccharides were der i vat i zed to a l d i t o l acetates for ana lys is by gas l i q u i d chromatography. Panel A: standards purchased from Sigma and subsequently d e r i v a t i z e d . Peak 1. Fucose, 2. Arabinose, 3. Xy lose, 4. I n o s i t o l ; 5. Mannose; 6. Glucose; 7. Galactose. The arrow indicates where amino sugars e l u t e . Panel B: Standards, Peak 4. I n o s i t o l ; 5 . Mannose. Panel C: Mannose was hydrolysed in 4M TFA for 6 hours p r io r to d e r i v a t i z a t i o n . Peak 4. I n o s i t o l ; 5. Mannose. Panel D: EngA 6 hour hydro lysate . Peak 4. I n o s i t o l . Panel E: Exg 4 hour hydro lysate . Peak 4. I n o s i t o l . 130 Appendix D Mass Spectra . EngA and Exg were hydrolysed in 4M TFA for a carbohydrate composition a n a l y s i s . The hydrolysates were reduced and acety lated for ana l ys i s by GLC. An a l d i t o l acetate was observed in both EngA and Exg hydrolysates which had a retent ion time s i m i l a r to Mannose. The unknown a l d i t o l acetate f rac t ion was c o l l e c t e d and analysed by Mass Spectrometry. Panel A: hex i to l hexaacetate reference spectrum (from Jansson, et a l . , 1976). Panel B: Unknown peak from EngA hydro lysate . Panel C: Unknown peak from Exg hydrolysate . 131 Appendix E A B C D E F G H I J K L M N 0 116-57-5-53" 4 i - ! ; 36-Western ana lys is of the protease digests of Avicel -bound c e l l u l a s e s . Avicel -bound native and recombinant EngA and Exg were incubated with protease and contro ls as descr ibed in Table XIII. The d igests were stopped and separated by cent r i fugat ion into an Avicel -bound f rac t ion and an A v i c e l - free f r a c t i o n . Samples were prepared for SDS-PAGE as described in F igs 26 and 27. Fol lowing e lec t rophores i s , the gel was e l e c t r o b l o t t e d onto n i t r o c e l l u l o s e . The blot was developed with normal rabbit serum. Lanes B -E : Native EngA and Exg; lanes F - I : recombinant EngA; lanes J - M : recombinant Exg. Lane A: Mr standards; lanes B, F, and J : TN buffer c o n t r o l , A v i c e l - f r e e f r a c t i o n ; lanes D, H, and L: TN buf fer c o n t r o l , Avicel -bound f r a c t i o n ; lanes C,G, and K: incubated with g lycero l cul ture supernatant, A v i c e l - f r e e f r a c t i o n ; lanes E, I , and M : incubated with g lycero l cul ture supernatant, Avicel -bound f r a c t i o n ; lane N: E . c o l i (pUC12) cu l ture supernatant proteins which bound to A v i c e l ; lane 0: g lycero l cul ture supernatant. The numbers indicate Mr markers X i o - 3 . 132 Appendix F Western ana lys is control with ant i -PT serum. S o l u b i l i z a t i o n mix alone was appl ied to 7 wells of a 10 % acrylamide SDS g e l . Following e lec t rophores i s , the gel was e l e c t r o b l o t t e d onto n i t r o c e l l u l o s e . The blot was developed with a n t i - P T serum. The arrow indicates two bands of n o n - s p e c i f i c antibody binding in the Mr range 55-65 X 1 0 - 3 . PUBLICATIONS Langsford, M.L., Gilkes, N.R., Wakarchuk, W.W., Kilburn, D.G., M i l ler , R.C., J r . and Warren, R.A.J. 1984. The cel lulase system of Cellulomonas fimi• J . Gen. Microbiol. 130:1367-1376. Gilkes, N.R., Kilburn, D.G., Langsford, M.L., Mi l ler , R.C., J r . , Wakarchuk, W.W., Warren, R.A.J., Whittle, D.J. and Wong, W.K.R. 1984. Isolation and -characterization of Escherichia col i clones expressing cellulase genes from Cellulomonas f imi. J . Gen. Microbiol. 132:1377-1384. Gilkes, N.R., Langsford, M.L., Kilburn, D.G., Mi l ler, R.C., J r . and Warren, R.A.J. 1984. Mode of action and substrate spec i f i c i t ies of cellulases from cloned bacterial genes. J . B io l . Chem. 259:10455-10459. Warren, R.A.J., Beck, C.F., Gilkes, N.R., Kilburn, D.G., Langsford, M.L., Mi l ler , R.C., J r . , O 'Ne i l l , G.P., Scheufens, M. and Wong, W.K.R. 1987. Sequence conservation and region shuffling in an endoglucanase and exoglucanase from Cellulomonas f imi. Proteins. .1:355-341. Langsford, M.L., Gilkes, N.R., Singh, B., Moser, B., Mi l ler, R.C. J r . , Warren, R.A.J, and Kilburn, D.G. 1987. Glycosylation of bacterial cellulases prevents proteolytic cleavage between functional domains. FEBS letters. 225:163-167. Mi l ler, R.C., J r . , N.R. Gilkes, N.M., Greenberg, D.G., Kilburn, M.L. Langsford and R.A.J. Warren. 1987. Cellulomonas fimi cellulases and their genes. In: "Biochemistry and genetics of cellulose degradation". FEMS Symposium. Academic Press, London. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098066/manifest

Comment

Related Items