UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Structure of the gene encoding the exoglucanase of Cellulomonas fimi and its expression and secretion… O’Neill, Gary Paul 1986

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

Item Metadata


831-UBC_1986_A6_7 O53.pdf [ 10.72MB ]
JSON: 831-1.0096771.json
JSON-LD: 831-1.0096771-ld.json
RDF/XML (Pretty): 831-1.0096771-rdf.xml
RDF/JSON: 831-1.0096771-rdf.json
Turtle: 831-1.0096771-turtle.txt
N-Triples: 831-1.0096771-rdf-ntriples.txt
Original Record: 831-1.0096771-source.json
Full Text

Full Text

STRUCTURE OF THE GENE ENCODING THE EX08LUCANASE OF CELLULOMONAS FIMI AND ITS EXPRESSION AND SECRETION IN ESCHERICHIA COLI By GARY PAUL O'NEILL M.Sa.McGill University, 1982 B.Sa.McGill University, 1979  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 this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA JUNE 1986 © Gary Paul O'Neill  In presenting  this  thesis i n partial  f u l f i l m e n t of the  r e q u i r e m e n t s f o r an a d v a n c e d d e g r e e a t t h e of B r i t i s h Columbia, I agree that it  freely  t h e L i b r a r y s h a l l make  a v a i l a b l e f o r r e f e r e n c e and s t u d y .  agree t h a t permission f o r extensive for  University  s c h o l a r l y p u r p o s e s may  for  financial  _ of  Microbiology j " _  The U n i v e r s i t y o f B r i t i s h 1956 Main M a l l V a n c o u v e r , Canada V6T 1Y3 Date  October 14, 1986  Columbia  my  It is thesis  s h a l l n o t be a l l o w e d w i t h o u t my  permission.  _ Department  thesis  be g r a n t e d by t h e h e a d o f  copying or p u b l i c a t i o n of t h i s  gain  further  copying of t h i s  d e p a r t m e n t o r by h i s o r h e r r e p r e s e n t a t i v e s . understood that  I  written  ii  ABSTRACT In the Gram-positive bacterium Cellulomonas fimi. the pex gene encodes a secreted exoglucanase (Exg) involved in the degradation of cellulose. I n order to study the relationship between the structure and function of this cellulase, the nucleotide sequence of the pex gene was determined, and its overproduction in Escherichia coli was obtained to facilitate purification of large amounts of this enzyme. The pex gene was localized to a 2.58-kb BamHI-Sall fragment contained within a 6.6-kb BamHI fragment of C.fimiDNA, and its nucleotide sequence was determined. The pex coding region of 1452-bp (484 codons) was identified by comparison of the DNA sequence to the amino terminal amino acid sequence of the Exg produced by C. fimi. In the gene, the region encoding the amino terminus of the mature Exg is preceded by a sequence encoding a putative signal peptide of 41 amino acids, a translational initiation codon and a ribosome binding site. The nucleotide sequence immediately following the translational stop codon contained four inverted repeats, two of which overlap. A dramatic (9B.S%) bias occurs for guanosineor cytosine in the third position of the 35 codons utilized in the pex, The ability of the proposed 41 amino acid signal peptide to direct secretion from E. coli was shown by gene fusion experiments. A hybrid leader sequence containing the 6 amino-terminal amino acids of B-galactosidase (J36al) followed by the 37 amino acids of the carboxyl end of the leader peptide directed export of Exg into the periplasm of Lpoli In contrast, hybrid BGal-Exg proteins in which the leader sequence was not present were located in the cytoplasm. In an attempt to optimize the secretion of the Exg from E. poll, the Exg signal peptide W8S replaced with the signal peptide from the OmpA protein to yield an OmpA-Exg hybrid protein. The OmpA-Exg hybrid protein retained biological activity and was secreted into the periplasmic space of E. poll • A series of operon and gene fusions between various E. coli operons and the Exg coding sequence was carried out in order to optimize the expression of Exg in E. coli. The highest level of Exg production exceeded 20* of the total cellular protein. To obtain this level of overproduction, the Exg coding sequence was fused to a synthetic ribosome binding site, with an initiating ATG, and was placed under the control of the leftward promoter of phage lambda contained on a runaway replication plasmid vector.  iii  TABLE OF CONTENTS Page ABSTRACT  11  TABLE OF CONTENTS  Ill  LIST OF TABLES  viii  LIST OF FIGURES  ix  ABBREVIATIONS, NOMENCLATURE AND SYMBOLS  '...xi  ACKNOWLEDGEMENTS  xiii  INTRODUCTION  1  I Background  1  II Biodegradation of cellulosic materials  .....3  (A) Cellulose structure  3  (B) Microorganisms  3  (C) Enzymatic and nonenzymatic components of eellulase systems and the mechanism of enzymatic hydrolysis (D) Applications of cellulases III  5 10  The application of recombinant DNA techniques to the study and industrial exploitation of cellulose biodegradation (A) Cloning and characterization of cellulase genes (1) Molecular cloning of cellulase genes (a) Source of genes (b) Cloning vectors and hosts (c) Method of screening and selection (2) Characterization and properties of cloned cellulases  12 12 12 12 15 16 18  (a) Physical and biochemical characterization  18  (b) Physiological characterization  22  (i) Heterologous expression  22  iv  (11) Heterologous secretion of recombinant cellulases......24 (3) Genetic characterization (B) Exploitation of recombinant cellulases.. (1) Overproduction (a) Characterization of the gene  25 27 28 28  (b) Host organism for the expression of recombinant cellulases (c) Yectors  29 30  (d) Regulatory signals and elements required for gene expression  32  (I) Transcriptional elements and factors  32  (II)  Translatlonal signals and factors  (2) Secretion and export of recombinant cellulases  34 36  (a) Host organism  36  (b) Signals and Information required for secretion  36  MATERIALS AND METHODS  39  I  Bacterial strains, phages and plasmlds  II  Media  39  III  Buffers  39  IV  Biological screening for exoglucanase activity  39  Y  Growth and Induction of bacteria  41  Yl  DNA manipulations and recombinant DNA techniques  42  (A)  Enzymes, reagents and techniques  (B)  Preparation and purification of plasmld and phage DNA  (C) Synthetic oligodeoxyribonucleotides  39  42 42 43  (D) Primer annealing reaction and Klenow extension  43  (E)  44  Mung bean nuclease reaction  V  VII  Plasmld and phage vector constructions (A) Construction of Ml 3 Clones  44 44  (B) pEC-1.1, pEC-1.1 s, pEC-1.1 tac , pUC 12-1.1 cex., pUC 131.1cex,pUC12-1.lcex(748) and pUC 12-1.1 cex (859) (C) pCP3-cex, pUC 12-1.1 (737) and pUCI 2-1.1 (PTIS)  44 45  VIII Gel electrophoresis of DNA  46  IX  DNA sequence determination  46  X  Protein purification  47  (A) Cell fractionation  47  (B)  Immunoadsorbent chromatography purification  48  (1) Preparation of the immunoadsorbent  48  (2) Immunoadsorbent chromatography  49  ( 0 XI  Isolation of intracellular insoluble Exg aggregates  Electrophoresis of proteins  XII Assays of enzyme activities and protein concentrations  50 51  (A) p-nltrophenylcelloblosldase assay  51  (B) carboxymethylcellulaseassay  52  ( 0  fi-galactosldase assay  52  (D) B-lactamase assay  52  (E) Assay of protein concentration  53  RESULTS AND DISCUSSION I  50  54  Structure of the gene encoding the exoglucanase of C. flml. (A) Subcloning of the cex gene  54 54  (1) Detection of Exg activity coded by subcloned DNA fragments (2) Localization and direction of transcription of the cex gene  54 55  vi  (B) DNA sequence of the cjx gene  61  (1) DNA sequence determination  61  (2) Nucleic acid sequence.  63  (C) Structure of the cex gene and Its product  63  (1) Identification of a signal sequence, Initiating codon and rlbosome binding site (2) Transcriptional control elements  63 68  (a) Promoter  68  (b) Terminator  68  (3) Codon usage  69  (4) A tandemly repeating heptapeptide In the carboxyl terminus of the Exg  69  (5) Open reading frames In the 5' and 3' flanking sequences of the cex gene (6) Hydrophiliclty analysis of Exg and EngA peptides  74 74  (7) Dot matrix analysis of the cex gene with Itself and with the cenAgene  77  II  Immunoadsorbent purification of the recombinant Exg from E. coil  82  III  Expression of the Exg in E. coll.  87  (A) Introduction  87  (B) Construction of Exg-expression plasmids- Transcriptional fusions  88  (C) Construction of Exg-expression plasmids- Translational fusions  90  (1) lacZ fusions  90  (2) Expression of the Exg fused to a portable translational initiation site in plasmid pUC12  90  (D) Plasmids with thermoinducible control of runaway replication and transcription  94  vii  (1) Plasmld pCP3cex construction  94  (2) Plasmld pCP3cex-directed Exg expression  94  (3) Effect of induction temperature and length of Induction on Exg production by pCP3cex  97  (4) Cellular localization of Exg to insoluble aggregates and identification of the Exg (E) Conclusions IV  Secretion of Exg from E. coli (A) Functional analysis of the Exg leader sequence  102 1  04  104  (1) Construction of hybrid B6al-Exg leader sequences  104  (2) Cellular localization of BGal-Exg hybrid proteins  105  (3) Conclusions  106  (B) Use of the OmpA signal peptide for secretion of Exg from E. poll  LITERATURE CITED  99  110  (1) Construction of plasmlds expressing hybrid OmpA-Exg  111  (2) Expression of OmpA-Exg hybrid proteins in E. coli  114  (3) Conclusions  115 120  viii  LIST OF TABLES  Table  Page  I.  Summary of cloned prokaryotic and eukaryotlc cellulase genes  II.  Characterization of recombinant cellulases  lit.  19  Apparent molecular weights and cellular locations of prokaryotic cellulases  IV.  13  ....21  Length of coding sequences of 7 cellulase genes deduced from their DNA sequence  26  V.  Examples of specialized E. coli expression vectors  31  VI.  Bacterial strains, phage and plasmids  40  VII.  Base composition of each position in the codons used in the pex gene compared to other genes of high 6+C content and E. poll genes  VIII. IX.  Codon utilization for the pre-exoglucanese  72  Comparison of the codon usage in the pex gene to the average codon usage in 62 E. coli genes with tRNA levels In E. coll.  X.  71  Localization of Exg, D-lactamase,fiGaland the hybrid J36al-Exg In E. coli.  73 109  ix  LIST OF FIGURES  Figure  Page  1.  Structure of cellulose  4  2.  Schematic representation of cellulase activity In a cellulose fibril  7  3.  Synthetic substrates used In assaying Exg activity  56  4.  Functional detection of Exg on indicator plates and SDSpolyacrylamfde gels  57  5.  Subcloning and direction of transcription of the pex gene  59  6.  Localization of the pjx gene by in vitro deletion analysis  60  7.  Restriction map and sequencing strategy of the pex gene and its flanking regions  8.  Nucleic acid sequence and deduced amino acid sequence of the pex gene  9.  Schematic summary of the pex gene deduced from the nucleic acid sequence  10.  64  67  Hypothetical secondary structure which can form 1n the 3" nontranslated region following the Exg coding region  11.  62  70  Open reading frame analysis of the DNA sequence for the pex coding region and its 5' and 3' flanking regions  75  12.  Hydrophilicity analysis of the precursor forms for the C. fimi Exg and EngA  78  13.  Dot matrix comparisons of the Exg DNA coding sequence with itself and the EngA DNA coding sequence  79  14.  Scheme for the immunoadsorbent purification of the recombinant Exg  85  15.  SDS-PAGE analysis of the Exg purified by immunoadsorbent chromatography  16.  86  Summary of operon and gene fusions constructed between the Exg and E. coli genes for the purpose of the overproduction of the Exg in E. coM  89  X  17.  The DNA sequences of the RBS, translation initiation site, and aminotermini of fusion junctions of BGal-Exg expression plasmids and the level of Exg activity in cell extracts  18.  Comparison of the nucleotide sequence of the translational initiation regions of the PTIS and the cex gene  19.  20.  Diagrammatic representation of the A P|_-Exg expression system  21.  SDS-PAGE analysis of recombinant pCP3-cex-produced Exg and the natural secreted Exg  100  101  103  Scheme for the construction of BGal-Exg expression plasmids pUC12-l.lcex(748)and pUC12-1.1cex(859)  26.  98  SDS-PAGE analysis of the Exg overproduced by pCP3-cex and isolated by low speed centrifugatton  25.  96  SDS-PAGE analysis of proteins synthesized by E. coilC600( pel857) (pCP3-cex) at 30,34,37,40, and 43°C  24.  95  Timecourse of production following induction of Exg as determined by enzyme activity and SDS-PAGE analysis  23.  93  Scheme for the construction of the Exg-expression plasm ids pUC 12-1.1 (737) and pCP3-cex.  22.  92  108  Protocol for the directed access to specific DNA sequences using oligonucleotide primers  27.  Monitoring of the primer extension method by agarose gel electrophoresis  28.  Scheme for the construction of a plasmid vector in which the E. coll ompA signal peptide is fused to the mature terminus of the Exg  117 118  119  xi  ABBREVIATIONS, NOMENCLATURE AND SYMBOLS aa,  amino acld(s)  Ap,  amplcillin  bp,  basepalr(s)  BGal,  0-oal8Ctos1d8se  CCC,  covalently closed circular  pex,  gene coding for the exo-1 ,-4-D-glucanase of Cellulomonasflml  CMC,  carboxymethyl cellulose  CMCase, carboxymethyl cellulase dNTP, deoxyrlbonucleotldetriphosphates(dATP,dTTP,dGTP,dCTP) ds,  double-stranded  Exg,  exo-1,4-13-glucanase  HS-PBS, high salt (0.5 M NaCl)-phosphate buffered saline IPTG, lsopropyl-8-D-thiogalactopyranoside kb,  1000 bp  kda,  kllodalton(s)  lacZpo. lac promoter-operator lac, X,  lactose operon E. poll phage lambda  LB,  Lurla broth  MCS,  multiple cloning site  M ,  relative molecular mass  MUC,  methylumbelliferylcellobioside  nt,  nucleotlde(s)  r  ONP0, ortho-nitrophenyl-B-D-galactopyranoside ORF,  open reading frame  xii P ,  leftward promoter of A  PR,  rightward promoter of X  L  Pollk, Klenow (large) fragment of E. coliDNA polymerase I pNPC, para-nitrophenyl-B-D-cellobioside pNPCase,para-nitrophenyl-13-D-cellobiosidase PTIS,  portable translation initiation site  RBS,  ribosome binding site  ss,  single-stranded  tec ,  a hybrid E. coli transcriptional promoter containing the "-35" region of the trp_and the "-10" region of the Jac promoters  trp.  tryptophan operon  xiii  ACKNOWLEDGEMENTS  I am indebted to Drs. D.6. Kilburn, R.A.J. Warren, and R.C. Miller Jr. for their supervision, ideas, enthusiasm and patience throughout this work. I thank Drs. Michael Smith and J.T. Beatty for valuable discussions and advice. I am grateful to Drs. M. Inouye, E. Remaut and W. Fiers for their gifts of various expression plasmids and bacterial strains used In this study and to M. Langsford for providing unpublished amino acid sequence data. This work was supported by Strategic Grant 67-0941 and Grant 67-6608 from the Natural Sciences and Engineering Research Council of Canada to D.G. Kilburn, R.A.J. Warren, and R.C. Miller Jr.. I am grateful to the University of British Columbia Graduate Awards Committee for awarding to me a Wesbrook Fellowship for the years 1982-1986.1 dedicate this thesis to Margot.  1  INTRODUCTION (I) Background Cellulose, a linear polymer of hundreds to thousands of glucose residues, is the most abundant carbohydrate in existence. It is the major structural component of plants 8nd as such about 4 X 10 tons are produced per year by plants through photosynthesis (Coughlan, 1985b). 10  The renewability and the availability of cellulosic biomass makes this substance ideal as a source material for the production of food, fuels and chemicals. The long range objective of several international research groups has been the production of fuels and other products by fermentation of cellulose. This first requires the conversion of cellulose to glucose. Enzymatic and chemical methods of hydrolysis are available for the conversion of cellulose to glucose (Ladisch et al., 1983). A variety of bacterial and fungal species synthesize cellulolytic enzymes, broadly classed as cellulases, which degrade cellulose to glucose. However, these microorganisms often do not produce adequate levels of the enzymes for industrial use. The overall aim of this project has been the Investigation and the optimization of biodegradative cellulase systems employing recombinant DNA techniques. The hydrolysis of cellulose to glucose by microorganisms requires a multicomponent cellulase system. Three major types of hydrolytic activities are involved: endo-1,4-J3-glucanases (EC and exo-1,4-B-glucanases (EC acting synergistically in the breakdown of the cellulose chain to cellobiose, and B-1,4-glucosidase (EC, which hydrolyses the cellobiose to glucose. Our research group has undertaken a combined biochemical and genetic approach to define the cellulases of the Gram-positive prokaryoteCellulomonasfimi (Whittle et al., 1982; Gilkes et al., 1984a,b; Langsford et al., 1984; Wakarchuk et al., 1984; O'Neill et al., 1986a,b,c; Wong etal., 1986; Warren et al., 1986). Genes encoding exo-and endo-g1ucan8ses from C. fimi. have been isolated by molecular cloning and these recombinant plasm id-encoded enzymes have been correlated with the native enzymes ( Gilkes et al., 19848 and b; O'Neill et al., 1986a; Wong et al., 1986; M. Langsford unpublished observations). The gene encoding the exoglucanase (Exg) of C. Jjmi has been the focus of the research presented in this thesis. This Exg  2  removes cellobiose residues from the non-reducing end of the cellulose molecule by hydrolysis of the 13-1,4 glycosldlc linkages (Gllkes et al., 1984b). The native enzyme is secreted to the extracellular environment in a glycosylated form (Langsford et al., 1984). In my thesis work recombinant DNA techniques were used to characterize and manipulate the gene (cex) encoding Exg. The nucleotide sequence of the pex. gene was determined because of interest In the primary structure of the Exg, the mechanism of action of this cellulase and the role of glycosylation in its function. Because these studies and the development of commercial processes for the bloconversion of cellulose to glucose require large extracellular amounts of the enzymes, the high level expression of Exg in E. cjjli was established and secretion of Exg was studied In this organism. The Introduction Is divided Into three sections. The first section briefly reviews the structure of celluloslc materials, the microorganisms that contain cellulase systems and the composition of these cellulase systems. There are many reviews on these topics and the reader is referred to them for more detailed background (Beguln et al., 1986a; Blsarla and Ghose, 1981; Coughlan, 1985a and b; Detroy and St, Julian, 1980; Enarl, 1983; Enar 1 and Markannen ,1977; Fanetal., 1980a; Fan etal., 1980b; Flickinger, 1980; Gardner and Blackwell, 1974;Ghose, 1977; Gong and Tsao, 1979; Ladisch etal., 1983; Llnko etal., 1982;Mandels, 1982; Mandels, 1975; Rees etal., 1982; Reese etal., 1972; Reese, 1977; Ryu and Mandels, 1980; Wood, 1985a and b; Wood and McCrae, 1979). The second part of the Introduction reviews the molecular cloning of cellulase genes and the characterization of their gene structure and products. The third section deals with the aspects of molecular biology that have bearing on the industrial exploitation of cellulases.  3  (II) Biodegradation of cellulosic materials  (A) Cellulose structure Cellulose Is a linear homopolymer of glucose residues linked by 13-1,4 glycosldlc bonds (Gardner and Blackwell, 1974). However, the basic structural unit of the cellulose chain Is cellobiose, which is formed from two molecules of glucose. This Is due to the stereochemistry of the 6-1,4 linkage, which orients successive glucose residues at 180* to each other, orients the hydrogen and hydroxyl groups on adjacent glucose molecules In different positions, and places the 13-1,4 linkages In different planes (see Fig. 1). In nature, cellulose does not exist 8s simple chains of glucose and Is rarely found In a pure form. It Is commonly found as an Insoluble fibrous, semi-crystalline macromolecule embedded In or surrounded with proteins, minerals and other polysaccharides such as hemlcellulose, or by pectin and llgnln (Gardner and Blackwell, 1974; Rees et al., 1982). The degree of crystalllnlty, the chain length and the amount and type of other macromolecules associated with the cellulose are dependent on the source of the cellulose and its method of isolation or pretreatment. The simple cellulose polymer Is organized into several levels of higher order structure (see Fig. 1). At the first level a number of cellulose chains are packed together In parallel and held together by Intra- and Inter-molecular hydrogen-bonding to form a fibril. A bundle of fibrils Is termed a fiber. The fiber Is not uniform in composition and It displays crystalline regions In which the chains of cellulose are highly ordered and amorphous regions In which the cellulose chains are In relative disarray. These amorphous regions are thought to be crucial for the biodegradation of cellulose as they provide access for the initial enzymatic attack by cellulases.  (B) Microorganisms It Is Important to note, as discussed below, that the cellulase systems produced by microorganisms differ widely in the number and type of enzymes involved, the specific activities of the enzymes and the extent to which the cellulase system can degrade native cellulose (see  4  Figure 1. Structure of cellulose. (a) Stereochemical representation of the cellulose molecule. Arrows A and B represent the B-1,4-linkages lying In different planes within the cellulose fibril. Cleavage at these linkages will generate two different end group configurations. (b) Organisation of cellulose molecules in elementary fibrils. Regions of the fibril in which the polymers are highly ordered or are in relative disarray have been termed the crystalline and amorphous regions, respectively, (c) Cross-section of a wood fibre. Cellulose elementary fibrils are embedded in a matrix of hemicellulose and lignin, reducing their accessibility to enzymatic digestion (adapted from Fan et al., 1980a).  5  reviews by Blsarla and 6hose, 1981; Coughlan, 1985b; Enari and Markkanen, 1977; Oong and Tsao, 1979; Llnko etal., 1982; Mandels, 1975; Mandels, 1982; Reese, 1977; Ryu and Mandels, 1980; Wood, 1985a). Many different prokaryotic and eukaryotlc microbes synthesize cellulases but the nature of the particular cellulase system enables the microbe to either (1) partially degrade cellulose, or (11) to grow on cellulose as a sole carbon source, or (111) to effect the extensive hydrolysis of natural cellulose. The products of growth on cellulose are usually microbial cells, CO2 and methane. Effective producers of cellulases include fungal species such as Fusarium. Mvrothecium. Phanerochaete and Trlchoderma , Actlnomycetes such as Streotomvcetes arlseus and Thermomonosoora fusca and bacterial species such as Bacteroldes. Cellulomonas. and Clostridium. The most potent cellulolytlc systems come from mutants of the fungus Trichoderma reesel which can produce up to 20 mg/ml of extracellular protein, most of which 1s composed of cellulases (Henoy etal., 1984). Cellulolytlc organisms can be aerobic or anaerobic, free-living or In symbiotic association with other microbes. There is a limited amount of research on the Induction, repression and secretion of cellulases (rev. In Coughlan, 1985b). Most cellulases are extracellular except for the bacterial J3-glucos1dases which are often intracellular. The secretion of cellulases Is discussed In more detail below. Many of the cellulase systems produced by these organisms are inducible by growth on cellulose and repressed In the presence of simple sugars such as glucose. The actual Inducer molecule is not known, but, celloblose and lactose can induce cellulases In several species (Coughlan, 1985b).  (C) Enzymatic and nonenzymatic components of cellulase systems and the mechanism of enzymatic hydrolysis The complete hydrolysis of native cellulose to glucose by microbes Involves a complex system of several types of enzymes and other components. The enzymes that are directly Involved in cellulose degradation are classified as hydrolytic, phosphorylltic and oxidative. Other classes of enzymes, such as proteases and glycosylases, maybe Involved In the formation of active cellulases. The complexity of cellulase systems can be bewildering as specific enzyme functions  6  may be carried out by several genetically distinct enzymes and each of these enzymes may exist In a number of forms arising through differential glycosylatlon and proteolysis (Coughlan, 1985b; Langsford et al., 1984). Three major types of hydrolytlc activities are Involved In cellulase systems: endo- and exo-glucanases which act synerglstlcally In the breakdown of the cellulose chain to cellobiose, and 13-glucos1dase which hydrolyses the cellobiose to glucose (Fig. 2). Endoglucanases characteristically cleave 13-1,4 linkages at Internal sites within the cellulose polymer. Exoglucanases release cellobiose units from the non-reducing ends of the cellulose chains. The role of fj-glucosldase Is In the splitting of cellobiose to yield glucose. In addition to the 13-1,4 glucanolytlc enzymes, oxidative enzymes can also be Involved In cellulose degradation. These enzymes were first detected 1n the culture supernatants of Basldlomvcete species and are not part of all cellulase systems (Eriksson, 1978; Coughlan, 1985b). Two oxidative enzymes, cellobiose dehydrogenase and cellobiose oxidase, have been characterized. Both enzymes are Involved In the oxidation of cellobiose and short cellodextrtns to the corresponding lactones. The electron acceptors In the oxidation reactions Include llgnln and its degradation product, qulnone, for cellobiose dehydrogenase and O2 for cellobiose oxidase. The roles of these oxidation reactions may Include the oxidative catabollsm of cellobiose and In diminishing the inhibitory effect of cellobiose on cellulase activity (Coughlan, 1985b). Non-hydrolytlc components of the cellulase system have been proposed and several factors have been detected. Reese et al. (1950) originally postulated that non-hydrolytic components, termed C1 components, rendered the crystalline cellulose susceptible to attack by hydrolytie enzymes (Cy components). Several C^ -type components have been described (Coughlan > 1985b) which can either generate microfibrils from the cellulose structure without hydrolyslng 13-1,4 bonds or attach to the cellulose fibers and promote the binding of cellulases to the substrate (Ljungdahletal., 1983). There Is much speculation and several models on how all of these enzymes and components act to degrade cellulose. The hydrolytlc enzymes do work synerglstlcally, in that the extent of  7  Figure 2. Schematic representation of cellulase activity In a cellulose fibril. Individual glucose residues of the cellulose chains 8re represented by hexagons. The nonreducing end of 8 cellulose polymer is denoted by 8 filled hexagon.  7a  I  I Crystalline region Amorphous region  adsorption of cellulase enzymes  8 substrate hydrolysis obtained by combining the appropriate purified cellulases 1s greater than the sum of the extents of hydrolysis of the Individual preparations (Wood and McCrae, 1979; Wood, 1985b). This enzymatic synergism Is most pronounced when the cellulose has a high degree of crystal Unity, low with amorphous cellulose and absent with soluble cellulose derivatives (Wood and McCrae, 1979). The synergistic effect Is clearly observed with certain fungal endoglucanases that hydrolyze crystalline cellulose only In the presence of an appropriate exoglucanase (Coughlan, 1985b). The fungal cellulases are thought to be extracellular, but, some enzymes In the complex may be cell-associated. In the bacterium Clostridium thermocellum. a large (2 X 10 dalton) 6  structure called the cellulosome 1s found on the cell surface (Lamed et al., 1983a and b). The cellulosome Is hypothesized to position and orient the various enzymes and components of the cellulase complex with respect to the celluloslc substrate for efficient hydrolysis. The first model of the mechanism of cellulose degradation, which was proposed by Reese, Siu and Levlnson (1950) has been extensively modified ( rev. 1n Coughlan, 1985b). The original model involved a two step process. In the first step the Cj components of the cellulase system cause the swelling, modification and disruption of the hydrogen bonds in the crystalline cellulose. This is followed by the synergistic attack by Cx components (endo- and exoglucanases). This model has been modified to accommodate the fact that covalent linkages In the crystalline cellulose are broken In the Initial steps of cellulose degradation. The most recent model for the mechanism of cellulose degradation has been presented by Coughlan (1985b) and Is summarized below (Fig. 2). In the first set of reactions, termed amorphogenesis, the hydrogen-bonded crystalline structure of cellulose is disrupted and made accessible to the hydrolytlc enzymes. The amorphogenesis is followed by the synergistic attack of the cellulases at regions of low crystalllnlty with the random cleavage of Internal 6-1,4 linkages by endoglucanases. These random internal cleavages create ends for the removal of cellobiose from the non-reducing end by exoglucanases. The cellobiose is then split Into glucose by a cell-associated 13-glucosldase. This minimal model was developed for fungal cellulases but Is essentially the same for bacteria.  9  In most cellulolytic organisms there are multiple exo- and endo-glucanases. The number of enzymes Is further increased by the differential glycosylation and proteolytic processing of each of these different cellulases. The need for the multiple forms of the endo- and exo-glucanases may be due in part to the stereochemistry of the substrate (Wood, 1985b). As discussed earlier, the glycosldlc linkages between successive glucose residues are in different planes and the array of hydrogen and hydroxyl groups on each successive glucose Is different. Thus, the different enzymes may recognize the two different conformations of glucose presented In the cellulose polymer. The reasons for the proteolytic modifications and glycosylation of each enzyme are not known. Proteolytic processing has been shown to be required for activation of a procellulase from Penlcilllum lanthlnellum (Deshpande et al., 1984b). Glycosylation of fungal cellulases has been reported to be essential for the secretion, activity and the pH and thermal stability of the enzymes (Merlvuori et al., 1985) but this conclusion has been questioned (Murphy-Holland and Eveleigh, 1985). The study of the interaction of cellulases with cellulose Is a complicated research area due to the nature of the substrate, enzyme assays, and enzymes. Many different substrates and enzyme assays are employed in such studies. Commonly used substrates include insoluble derivatives such as filter paper, cotton, and commercially prepared cellulose (Avlcel) and soluble derivatives of cellulose such as carboxymethylcellulose (CMC) and hydroxyethyl cellulose (HEC). Since cellulose is an insoluble, fibrous, semi-crystalline polymer, its enzymatic breakdown differs from the hydrolysis of a soluble substrate and the features of the cellulosic substrate which should be considered are (1) purity (Ii) degree of polymerization; (ill) degree of crystallinity; (iv) particle size and surface area; (v) enzyme to substrate ratio; (vi) the topographical changes in the substrate as digestion proceeds; (vii) effect of other substances (Coughlan, 1985b). Measurement of enzyme activities of specific cellulases in culture filtrates is difficult due to the multiple cellulolytic activities present. Unfortunately, since many pure cellulases do not have a readily determinable effect on crystalline cellulose, determination of their enzymatic activities in a pure form is impossible. There is an assortment of detection and assay methods for  10  cellulases and specific assays for endo- and exoglucanases and J3-glucos1dases (Ghose et al., 1983; Rosenberg and Oberkotter, 1977; Mulllngs, 1985; Coughlan, 1985b). The most popular methods measure the release of reducing sugars from insoluble (e.g. filter paper) or soluble (e.g. CMC) substrates using a reaction with dlnltrosallcylate (Miller, 1959) or the Nelson-Somogyl method (Nelson, 1952). There are many pitfalls to these two methods; these h8ve been reviewed by Breuil and Saddler (1985). For example, release of reducing sugars from CMC is easily followed but this substrate cannot be attacked by all cellulases. Synthetic derivatives of glucose and cellobiose modified with para-nltrophenol and methylumbelllferol are useful for measuring figlucosldases and exoglucanases (Deshpande et al., 1984a; van Tilbeurgh et al., 1982). Another assay which Is sensitive and can distinguish between endo- and exo-glucanases is the measurement of the decrease In viscosity or Increase in fluidity of a solution of a soluble cellulose derivative. Adsorption and desorptlon of cellulases with the substrate were first reported by HalHwell (1961). In convincing studies Chanzy et al. (1984 and 1985) followed the adsorption of Trichoderma reesei celloblohydrolase I onto cellulose by gold labelling techniques. Their results clearly revealed that the enzyme adsorbed onto amorphous regions of the cellulose fibers. This was followed by the directional hydrolysis of the cellulose. Other substances may facilitate the attachment of cellulases to cellulose (Ljungdahl etal, 1983).  (D) Applications of cellulases Mandels (1985) has recently catalogued the current and potential applications of cellulases and several reviews have dealt with the economic feasibility of the industrial applications of these enzymes (Blsarla and Ghose, 1981). There are many current uses of cellulases in the food processing industry (Coughlan, 1985b; Mandels, 1985). The focus of many research efforts h8s been on the degradation of cellulosic wastes and Its conversion into fuels and edible products. The first problem with this conversion is the inaccessibility of cellulosic materials to cellulases. Cellulose found in plant materials Is Impregnated and coated with Hgnin which makes this material refractory to attack by cellulases. Various physical, biological and  11  chemical pretreatments to remove the lignln or expose the cellulose are used but these treatments raise the cost of processing the cellulosic material. Other problems Include the high cost of enzyme production, the large amounts of cellulases required, catabolite repression of the production of cellulases and the end-product inhibition of the enzymes.  12  111 The application of recombinant DNA techniques to the study and industrial exploitation of cellulose biodegradation  Recombinant DNA techniques have proven to be powerful tools in the characterization of cellulases. The first molecular cloning of a cellulase gene was reported in 1982 (Whittle et al., 1982). In the following four years fifteen other research groups reported the molecular cloning of more than thirty-five cellulase genes from 14 different organisms (see Table 1). From 1982 to 1986 the research focus has been on the characterization and the heterologous expression of these cloned cellulase genes. The most important achievements have included the determination of the nucleotide sequence for eight cellulase genes, and the overproduction in E. coM of a cellulase which led to the first crystallization of a cellulase ( Joliff et al., 1986a and b). The cloning and the molecular characterization of the recombinant cellulases h8s formed the basis for current experiments examining the potential use of these genes for applied experiments 8s discussed below. The following two sections provide (i) a summary of the achievements in the molecular cloning and characterization of cellulase genes and (ii) an overview of the exploitation of recombinant cellulases using molecular biological approaches. A recent review article by Beguin et al. (1986a) should be referred to for a detailed account of the cloning and characterization of recombinant cellulases.  (A) Cloning and characterization of cellulase genes  (1) Molecular cloning of cellulase genes  (a) Source of genes Table I summarizes the organismal source of the cloned cellulase genes, the number and type of genes cloned, the cloning vector and the method of screening of recombinant clones. The source of the cellulase genes has been prokaryotic in 14 studies and eukaryotic in 5 studies. The  Table I. Summary of cloned prokaryotic and eukaryotic cellulase genes  Organism  Genes cloned * type  cloning vector  Screening method  Reference  PROKARYOTIC AarobacteriumATCC21400  1 O-glucosidase  plasmld pUC 13  DNA probe  Wakarchuk et al.. 1986  Bacillus subtilis  1 endoglucanase  plasmid pBD64  Indicator plates  Koide et al... 1986  Bacillus subtilis DL6  1 endoglucanase  plasmid pPL1202  Immunological  Robson and Chambliss, 1986a  Bacillus SD. strain N-4  2 endoglucanases  plasmid pBR322  Indicator plates  Sashihara e t a l . . 1984  Bacteroides succinoaenes  1 endoglucanase  plasmid pUC8  indicator plates  Collier e t a l . , 1984  Cellulomonas fimi  2 endoglucanase  plasmld pBR322  immunological  Whittle e t a l . . 1982;  1 exoglucanase Clostridium thermocellum  Clostridium thermocellum  Gllkes et al., 1984a and b  7 endoglucanases  cosmid pHC79;  indicator plates;  Cornet e t a l . . 1983b;  iZ exoglucanases  plasmid pACYC184  enzyme assays  Millet e t a l . . 1985  2 6-glucosldases;  lambda phage 1059  Indicator plates  Schwarz e t a l . . 1985  plasmld pBR322  selection on cellobiose  Armentrout and Brown. 1981  2 6-glucanases Escherichia adecarboxylata  1 O-glucanase  Erwinia chrysanthemi  (not available)  Roberts e t a l . . 1985  Table I. (continued)  Organism  Genes cloned  cloning vector  * type  Screening  Reference  method  Erwinia chrvsanthemi  i\ endoglucanase  RP4:: miniMu  indicator plates  van Gijsegem et al., 1985  Erwinea chrvsanthemi  1 endoglucanase  cosmid pMMB4  indicator plates  B a r r a s e t a l . , 1984  Erwinia chrvsanthemi  1 endoglucanase  lambda phage 147.  indicator plates  Kotoujansky e t a l . . 1985  Thermomonospora YX  2 endoglucanases  plasmld pBR322  enzyme assay  Collmer and Wilson, 1983  Aspergillus niqer  1  6-glucosidase  cosmid p3030  indicator plates  Penttila e t a l . , 1984  Candida pelliculosa  1  6-glucosidase  not available  not available  Kohchl and Toh-e, 1985  Kluvveromyces fraailis  1 13-glucosldase  cosmid pHCG3  enzyme assay  Raynal and Guerineau, 1984  Trichoderma reesei  i exoglucanase  lambda phage L47.1  differential hybridization  Shoemaker et al.. 1983  1 endoglucanase  and plasmld pBR322  of cDNA probes; hybrid  Shoemaker et al., 1984  EUKARYOTIC  selectionof mRNA Trichoderma reegel  1 exoglucanase  lambda phage 1059  differential hybridization of cDNA probes  Teeri et al., 1983  15  greater number of cloned prokaryotic cellulase genes may reflect the comparative ease of cloning bacterial genes. The fact that there are few barriers to the expression of heterologous bacterial genes In E. coli has been capitllized on by nearly all research groups cloning bacterial cellulase genes. Thus, in most of these cases the method of screening has depended on the expression of the heterologous gene to produce an enzymatlcally active or an Immunologically reactive cellulase In £. opJi • In contrast, the structure of eukaryotlc genes precludes the direct expression of genomic clones containing Introns In £. coli. Hence, more complex cloning and screening procedures were required In the isolation of eukaryotlc cellulases (Shoemaker et al., 1983; Teerl et al., 1983). The properties of the cellulases from different organisms Is a major consideration In choosing the source of cellulase genes. On the one hand, the fungal cellulases generally have a higher specific activity. On the other hand, cellulases from the thermophilic prokaryote Clostridium thermocellum are enzymatlcally active at temperatures over 65*C and, therefore, may be more suitable for certain Industrial applications.  (b) Cloning vectors and hosts The cloning of cellulase genes, as with any other gene, has required a biological cloning and gene amplification system and a method to select or screen for the cloned gene. The cloning host, the size of the DNA fragments to be cloned and the method of detection of positive clones all have had bearing on the choice of the cloning vector. Specialized E. opJi plasmids, or cosmlds or phage lambda cloning vectors have been used in conjunction with E. coli as a host ( see Table I) except in one report in which B. subtilis was employed as the primary cloning host (Kolde et al., 1986). In this one exception, a genomic library of B. subtilis was constructed in the plasmid pBD64 and screened in B. subtilis (Kolde et al., 1986). Phage lambda vectors (Schwarz et al., 1985; Shoemaker et al., 1983; Teerl et al., 1983) andthecosmidspHC79 (Cornetetal., 1983), p3030 (Penttllaetal., 1984), pMMB34(Barras et al., 1984) and pHCG3 (Raynal and Oierineau, 1984) have been employed for cloning of cellulase genes because of their capacity to carry large DNA inserts and in the case of the lambda  16  vectors the ease with which large numbers of recombinant clones can be screened (Shoemaker et al, 1983; Teerl etal., 1983). TheE.coVL plasmids pBR322 (Armentroutand Brown, 1981; Collmer and Wilson, 1983; Sashlhara etal., 1984; Whittle etal., 1982), pACYCI 84 (Millet et al., 1985),pPL1202(RobsonandChambl1ss, 1986a) and pUC8 and 13 (Collier etal., 1984; Wakarchuk et al., 1986) have been successfully used to clone bacterial cellulase genes. These plasmids have the following features In common: (1) selectable antibiotic markers; (11) a variety of unique restriction sites for cloning; (111) the possibility of screening for recombinant clones carrying inserts by enzymatic complementation or antibiotic resistance inactlvatlon; (Iv) a high copy number; and (v) an origin of replication for autogenous replication. The correct choice of the restriction site on the plasmld for cloning has been crucial to the success of cloning several cellulases because the expression and detection of the cloned genes has relied on the insertion of the gene at positions in the plasmld vector which places the genes under the contol of plasmld-borne transcriptional and translational signals (Whittle etal., 1982; O'Neill etal., 1986c; Wong etal, 1986).  (c) Method of screening and selection The screening of genomic libraries of cellulolytic organisms In £. CJJU has been based on (1) the hybridization of synthetic oltgodeoxyrlbonucleotldes and differential hybridization of cDNA probes and (11) the expression in E. coVLor Bacillus subtllls of an Immunologically reactive or a enzymatically active product. In one case (Armentrout and Brown, 1981) a recombinant E. co_m clone containing a cloned 13-glucosldase gene was selected by growth of the transformants on plates containing cellobiose as the sole carbon source. Two Independent research groups have isolated the gene encoding the exocelloblohydrolase from the fungus Trlchoderma reesel bv screening a lambda genomic library using a differential hybridization strategy (Shoemaker et al., 1983; Teerl et al., 1983). In this procedure, I. reesel cultures were Induced specifically for cellulase production and poly-A RNA was isolated from the induced and unlnduced cells. The two poly-A RNA populations were then used for cDNA synthesis. Most of the cDNA species would be shared by both the induced  17 and unlnduced cDNA populations except for a small subpopulation of cellulase-specific mRNAs. The induced and uninduced cDNA populations were differentially hybridized to a T. reesei genomic library carried In a phage lambda cloning vector. A number of clones which hybridized only to the Induced cDNA species were picked for further characterization. From this population of positive clones Shoemaker et al. (1983) Isolated a genomic clone encoding the gene for the exocellobiohydrolase I of I. reesei by a mRNA hybrid selection technique in which poly-A mRNA from Induced I. reesei cells was hybridized to the positive clones, the mRNA that hybridized was eluted and translated In vitro. Immunopreclpltation employing an antiserum specific for the exocellobiohydrolase I was then used to identify In vitro translation reactions which produced exocellobiohydrolase I. Screening of gene banks In E. coli for cellulases which are enzymatlcally active or Immunologically reactive Is simple. Biological screening Is accomplished with Indicator plates (Teather and Wood, 1982; Wakarchuk etal., 1986; 6. O'Neill unpublished observations). Indicator plates for endoglucanases and exoglucanases are based on the hydrolysis of carboxymethyl cellulose and methylumbelllferylcelloblose, respectively. Immmunologlcal Identification of cellulase expressing colonies requires the preparation of an antiserum to the purified native cellulase (Robson and Chambllss, 1986a) or to the crude culture supernatant fluid of the cellulolytlc bacteria (Whittle et al., 1982). Although the Immunological screening technique Is more complicated than the Indicator plates, It has proven useful In studies In which the cloned cellulase activity is not detected on an indicator plate or if the cloned cellulase is weakly expressed (Whittle etal., 1982; Oilkes etal., 1984b). Furthermore, the immunological screening method may be essential In the identification of clones expressing noncellulolytlc but essential components of the exocellular cellulase system such as proteases (Deshpande et al., 1984b) and cellulose binding factors (Ljungdahl etal., 1983).  18  (2) Characterization and properties of cloned cellulases To date all of the cloned prokaryotic cellulases and eukaryotic cellulases have been expressed in either E. co]i, B. subtilis or S. cerevisiae (see Table II, part II). Thus, much work has been done on the characterization of the cellulase genes and their products. For the purpose of this Introduction, the study of the cloned genes and their cellulases has been divided into three sections: (1) biochemical characterization of the recombinant products; (11) physiological characterization of the heterologous expression of the cellulases; and (111) genetic characterization of the cellulase genes. A summary of the properties and the techniques used to analyze the cloned cellulases is given In Table 11.  (a) Physical and biochemical characterization The apparent molecular weights of the various cellulases coded for by cloned genes in E. coli or B. subtilis has been determined most often by SDS-PAGE analysis. The positions of recombinant cellulases on SDS-PAGE profiles of cell extracts has been determined by Western blotting and zymogram techniques (Beguin, 1983; Collmer and Wilson, 1983; O'Neill etal., manuscript submitted). The highest and lowest M/s for recombinant cellulases are 33,000 for a B. subtilis endoglucanase (Robson and Chambliss, 1986a) and 70,000 for a Thermomonospora endoglucanase (Collmer and Wilson, 1983). In making comparisons between the M 's of the r  recombinant cellulases with those of their native counterparts several observations that have bearing on the size of the enzymes should be considered. First, native cellulases are usually extracellular secreted proteins and as such they are synthesized as precursors with a signal peptide which Is removed during secretion. Second, the native eukaryotic and several bacterial cellulases are glycosylated. Third, post-secretlonal proteolytic processing of the native cellulases does occur. Thus, in order to compare the M 's of the native cellulases with those of their r  recombinant counterparts, signal peptide processing, proteolytic processing and glycosylation of the recombinant cellulases by E. coli or B. subtilis must be assumed to be the same as in the original host. Unfortunately, this is not the case. E. cgji and B. subtilis do not  Table ll. Characterization of recombinant cellulases and their genes.  P a r t I. B i o c h e m i c a l c h a r a c t e r i z a t i o n (A) Physical characteristics (1) Molecular weight (a) zymogram, activity PAG (b) Western blotting (c) minicell ormaxicell system (d) In vitro transcription/translation (e) sedimentation equilibrium (f) molecular sieve chromatography (2) Isoelectric polnt/electrofocusing (3) Amino acid sequence (4) Post-translatlonal modifications (5) X-ray crystallography (B) Immunological characteristics (1) Antigenic relatedness to recombinant and native cellulases (2) Inhibition of enzyme activity by antisera (C) Enzymological characteristics (1) Substrate specificity (2) Mode of action (3) Effect of pH on enzyme stability and activity (4) Effect of temperature on enzyme stability and activity (5) KM and V (6) Inhibitors M A X  Reference  3  4,11,20,30,41 7 8,11,43 10 16 33,37 16,37,40 21 21,30,36,37 16,17  P a r t II. P h y s i o l o g i c a l c h a r a c t e r i z a t i o n (A) Heterologous expression (1) Host organism (a) Saccharomvces cerevlslae (b) Bacillus gubtllts (c) Escherichia coll (2) Expression In heterologous hosts (a) Effect of host strain (b) Effect of growth phase (c) Overproduction (3) Heterologous secretion and export  Reference  3  18, 28, 32, 36-38 19,30 1,2, 5,6, 7,8, 10, 11, 16,19, 20, 22, 24, 25, 28, 30, 40, 41-43 7, 30, 37 7, 13, 16,19, 30,37 16, 25,41 1,6-8, 10, 11, 13, 18, 19,24, 30, 33, 36-38, 43  P a r t III. G e n e t i c c h a r a c t e r i z a t i o n 4,8,9,30,33,42 7,9,30,33 II, 12, 16,22,34,41 1,12, 16, 19 16,19,33 16, 19, 33 21 I  References: I) Armentrout and Brown, 19BI; 2) Barrasetal., 1984; 3) Begulnetal., 1985; 4) Begulnetal, 1983; 5) Beguln et al.,1986b; 6) Collier et al., 1984; 7) Collmer and Wllson,l983, 8) Cornet et al., 1983a; 9)Cornet et al., i983o,lO) Crosby etal, 1984; ll) Gilkes etal., 1984a; 12) Gilkes et al., 19846,13) Gtlkes etal., l9B4c;l4)Greenberget al., 1986 15) Greplnet and Beguln, 1986, 16) Jollff et al., 1986a; 17) Jollff et al., l986b;l8)Kohchl and Toh-e, 1985; l9)Koldeetal., 1986; 20) Kotoujansky et al., 1985; 21)Langsford,ML, Arfman, N, Gilkes, NR., Kllburn, D.G., Miller, R.C. and Warren, RAJ. (manuscript In preparation); 22) Millet et al., 1985; 23>0Nelllet al., 1986a,24)O'Neill et al ,1986b; 25)0Nelll et al., 1986c; 27) Petre et a l , 1985; 28) Raynal et a l , 1984; 29) Roberts et al, 1985; 30) Robson andChambllss, 1986a; 3l)RobsonandCnambllss, 1986b; 32)Sacco et a l , 1984.33) Sashlhara et al, 1984; 34) Schwarz et al, 1985; 35) Shoemaker et al, 1983, 36) Shoemaker et at, 1984; 37) Skipper er a l , 1986; 38) Skipper et al, 1985. 39) Teerl et al, 1983, 40) van Gljsegem et a l , 1985; 41 (Wakarchuk e t a l , 1986; 42) Whittle et al, 1982; 43) Wong et al, 1986.  (A) Homology to chomosomal cellulase genes 6-8. 10.27.28,33, and chromosomal organization 40,43 (B) Gene structure (Gross) (1) Localization of gene by subcloning 2, 7, 8, 16, 18, 23,28, 30,41,43 (2) Restriction analysis I -3, 6, 8, 15, 16,18-20, 22, 28, 30, 33-35, 39,41,43 (3) Dtrectton of transcription 5,7,8, 15, 16, 18,31,41,43 (C) Gene structure (Fine) (1) Nucleotide sequence 3, 15, 18, 23, 31, 35, 36, 43 (2) In vitro deletion analysis 18,24,41,43 (3) S |-nuclease mapping of mRNA 5,14 (D) Recombinant vector stability (E) Ability of transcriptional regulatory elements to function in foreign host  16 5, 8, 10, 30, 43  20  glycosylate proteins and while there are no published data on the processing of the signal peptides of the recombinant cellulases, Robson and Chambllss (1986a) and Cornet et al. (1983a) have reported multiple products resulting from proteolytic processing of the recombinant cellulases In E. coli that are not found In the original host (see Table III). For example, the mature native exocellular endoglucanase of B. subtilis DL6 has an M of 35,200 (Robson and Chambllss, r  1986a). The cloned gene directs the synthesis of products with Mp's of 51,000 and 39,000 in E. coli and products with PL's of 50,500,38,500,36,000,35,200 and 33,500 when expressed In an endoglucanase-deflcient strain of B.§ybJil1s( Robson and Chambllss, 1986a). Probably the most exciting achievement In the physical characterization of the recombinant cellulases is the crystallization and preliminary X-ray crystallographlc studies of the endoglucanase D of Clostridium thermocellum (Jollffe et al., 1986a and b). Antlsera prepared against the native cellulases have 1n all cases cross-reacted with their recombinant counterparts and vice-versa and antlsera reactive to cellulases can Inhibit enzymatic activity (Beguln etal., 1983; Collmer and Wilson, 1983; Cornet etal., 1983; Robson and Chambllss, 1986a; Sashlhara etal., 1984). Different cellulases from the same organism may show antigenic cross-reactivity (Collmer and Wilson, 1983; Cornet et al., 1983; Robson and Chambllss, 1986a; B. Moser unpublished results). Various enzymatic characteristics of recombinant cellulases have been studied (see Table II), but, comparison of these results Is difficult since the assays, substrates, reaction buffers and temperatures used in the studies are all different. All of the bacterial recombinant endoglucanases can hydrolyze CMC. Several recombinant cellulases from Clostridium thermocellum (Millet et al., 1985) and Cellulomonas fimi (Gilkes et al., 1984b) hydrolyze the agluconic bond of pNPC and MUC which suggests that these enzymes are exoglucanases. Only the exoglucanase of Cellulomonas fimi has been shown to be a true exoglucanase (Gilkes et al., 1984b). All other cloned bacterial cellulases are endoglucanases except for the celloblases from Aorobacterium (Wakarchuk et al., 1986) and Escherichia adecarboxvlata (Armentrout and Brown, 1981).  Table III. Apparent molecular weights and cellular locations of cloned prokaryotic cellulases  Source of gene  a subtilis B, 5UWI11S DLG Bacillus N-4 B. succinoaenes dflml £.. fimi "  C. thermocellum  Gene and/ or vector designation Host  3  r  -35.2  pBC5 PBRC5 pl_G4001b  a subtilis  pLG4001a pNKI pNK2 BC14 , cex: pEC-1 cenA:DCEC2  £ £ £ £ £  cenA: DK2.4 cenB;pEC3  S. cerevisiae t£Oji  "  tCQll S. cerevisiae £C0H £ coli ECQil  56  cenA; p6  ceiA  ceiA celB celD ThermomonosDora D3I6  a  Molecular size ( M ) x i 0 native recombinant  £ coli  a subtilis coli coli coil coli COli  £ coli a cerevislae  -  45 56 58  66  n.d. n.d. 50, 38,36, 35, 33 51,39 58 50 43 51  Cellular location  E E.P.C E,C P.C.M E , P, C E , P, C M  p,c  ad.  P, C E E P,C  56,55,50 n.d. 55,53 65 70  P,C C P.C  150,97,44 n.d. 1 90  E , P, C  Abbreviatlons: E , extracellular; P, periplasmlc; C, cytoplasmic; n.d. not done  8  Reference  Koide et al., 1986 Kolde et al., 1986 Robson and Chambliss, 1986a Robson and Chambllss, 1986a Sashlhara et al., 1984 Sashihara et al., 1984 •Crosby-et al., 1984 Gilkes et al., 1984a Wong et al., 1986 Skipper et al., 1986 Skipper et al., 1985 Gilkes et al., 1984a and J. Owalabi, unpublished observation Cornet et al., 1983a Sacco et al., 1984 Beguin et al., 1983a Jollff et al., 1986a Collmer and Wilson, 1983  22  Since the recombinant cellulases are not glycosylated In E. con In contrast to the native enzymes, the effect of glycosylate on the biological activity of these enzymes can be examined. Such a study has been carried out for two cellulases from Cellulomonas fimi (N. Gilkes, personal communication). The nonglycosylated recombinant enzymes and the their native counterparts did not differ significantly with regards to K , V m  max  , thermal stability, pH stability and specific  enzyme activity. These results are In agreement with those of Murphy-Holland and Evelelgh (1985) who reported that the lack of O-llnked glycosylatlon of Trlchoderma reesei cellulases has a minimal effect on the specific enzyme activity and that N-llnked glycosylatlon Is not essential for activity, pH stability or for thermal stability. In contrast, Merlvuori et al. (1985) found that glycosylatlon was essential to the secretion, activity and stability of the cellulases from Trlchoderma reesei.  (b) Physiological characterization (1) Heterologous expression All of the recombinant cellulase genes have been expressed In at least one foreign host (see Table 11). An exoglucanase gene and an endoglucanase gene from Trlchoderma reesei have been expressed In Saccharomvces cerevislae (Shoemaker et al., 1984). The expression In yeast of these genes required the removal of Introns and the placing of the genes under the transcriptional and translational control of yeast regulatory elements (Shoemaker et al., 1984). There have been no reports of the expression of these fungal cellulase genes in a bacterial host. B-glucosldase genes from Aspergillus nioer (Penttlla et al.. 1984) Candida pelliculosa (Kohchi and Toh-e. I985)and Kluweromvces fraallis (Raynal and Guerlneau, 1984) have also been expressed In S. cerevislae. All cloned bacterial cellulase genes have been expressed in E. poll, and two endoglucanase genes from Bacillus species have been expressed in Bacillus subtilis (see Table 11). An endoglucanase gene and an exoglucanase gene from Cellulomonas fimi have been expressed in Rhodobacter capsulata (Johnson et al., 1986), Pseudomon8s acidovorans and Klebsiella pneumoniae (R.A.J. Warren unpublished observations), and Saccharomvces cerevislae (Skipper et al., 1986; N.  23  Skipper, personal communication). An endoglucanase from Clostridium thermocellum also has been expressed in Saccharomvces cerevislae (Sacco etal., 1984). In several cases heterologous expression of the recombinant cellulase gene required its fusion to transcriptional and translational regulatory signals derived from E. coli (O'Neill et al., 1986c; Wong et al., 1986) or Saccharomvces cerevislae (Sktooer et al.. 1985). In general, the level of expression In the heterologous host, defined here as the specific cellulase activity In whole cell extracts, of the recombinant cellulase gene has been significantly lower than In the natural host (Collmer and Wilson, 1983; Jollff et al., 1986a; Skipper et al., 1985). One exception 1s the celD gene of Clostridium thermocellum (Jollffe et al., 1986a). The recombinant product of this gene can account for 1555 of the total cellular protein In E. coli, greatly exceeding the level obtained in the original host. Although the methods, substrates and conditions used by various researchers for assaying cellulase activity are different It Is clear that a given cellulase gene is expressed at different levels In a given host species. It has been established that the levels of cellulase expression In E. coli can vary according to the the strain (Collmer and Wilson, 1983; Kolde etal., 1986; Robson and Chambliss, 1986a), the growth phase of the host (Collmer and Wilson, 1983; Kolde etal., 1986; Robson and Chambllss, 1986a), the Influence of plasmld-borne regulatory signals (Jollff et al., 1986a; O'Neill et al., 1986c; Wong et al., 1986), the ability of gene expression signals to function In E. coJl (Jollff et al., 1986a; O'Neill et al., 1986c; Wong et al., 1986), the type and copy number of the vector containing the cloned cellulase gene, the growth medium, vector stability (Joliff et al., 1986a), and the effect of catabolite repression (Crosby et al., 1984). Other factors which may influence the heterologous expression of cloned cellulase genes include; (i) mRNA stability; (ii) susceptibility of the recombinant cellulase to host proteases; (iii) codon usage; (Iv) toxicity of the product to host cells; (v) the rate at which the precursor translation product can be processed and secreted; and (vi) the ability of a protein to fold into a normal secondary and tertiary structure when made in a foreign environment (Broach etal., 1983;Dubnau, 1983; Masui etal., 1983; Roggenkamp etal., 1985; Shatzman et al., 1983). The operator control region of the Bacteroides succinoqenes  24  endoglucanase gene functions in £.railas expression of this gene is subject to catabollte repression In this host.  (11) Heterologous secretion and export of recombinant cellulases  Most cellulases, with the exception of cellobiases, are extracellular (Coughlan, 1985b). The sequencing of seven cellulase genes has revealed that all the enzymes are synthesized as typical precursor secreted proteins with a signal peptide (see Table IV and Beguln et al., 1985; Greplnet and Beguln, 1986; Kohchi and Toh-e, 1985; O'Neill etal., 1986a; Shoemaker etal., 1983; Shoemaker et al., 1984; Wong et al., 1986;). Since the mechanisms of protein secretion are highly conserved among eukaryotes and prokaryotes (see rev. by Inouye and Halegoua, 1980; Mlchafills and Beckwlth, 1982; Pugsley and Schwartz, 1985; Randall and Hardy, 1984; Silhavy et al., 1982), it might be expected that the cloned cellulases would be secreted from the heterologous host. Indeed, the cellulases from several Bacillus species (Kolde et al., 1986; Robson andChambllss, 1986a; Sashlharaetal., 1984). Bacteroidessuccinooenes (Crosby et al.. 1984), Cellulomonasfimi (Gilkes etal. 1984a). Clostridium thermocellum (Cornetet al. 1983),and Thermomonosoora YX (Collmer and Wilson, 1983) are secreted by E. coli Into the periplasms space (see Table III). In addition, up to Z6% of the total enzyme activity of the Thermomonospora YX (Collmer and Wilson, 1983)andtwo Bacillus (Kolde etal, 1986; Sashihara etal, 1986) cellulases are found in the supernatant fluids of E. poll cultures. It appears that the export of the cellulases 1s specific and not due to cell lysis because only low levels of other cytoplasmic and periplasms marker enzymes are detected In the supernatant fluid (Collmer and Wilson, 1983; Kolde et al, 1986; Sashlhara et al., 1986). Since the outer membane of Gram-negative bacteria normally prevents the export of proteins it is unusual that these cellulases are exported into the medium. This may be due to the alteration of the outer membrane of E. coli by the recombinant cellulase resulting in a leaky membrane (Pugsley and Shwartz, 1985).  25  (c) Genetic characterization The third major area of the study of cloned cellulase genes is the genetic investigation of their structure and organization. The genetic organization and regulation of cellulolytlc systems Is expected to be complex (Millet et al., 1985). For example, ten distinct DNA fragments encoding cellulases have been Isolated from Clostridium thermocellum and It has been estimated that these DNA fragments represent a 70 kb region of the 2500 kb genome Involved in cellulose degradation (Millet etal., 1985). The molecular genetic characterization of cloned cellulase has generally Involved the following (see Table 11 for references): (1) subclonlng of a biologically active cellulase gene on a DNA fragment of minimal size; (Ii) constructing a detailed restriction endonuclease map of the subcloned gene; (ill) determination of the direction of transcription; (Iv) determination of the nucleic acid sequence of the cellulase gene (see Table IV); and (v) definition of the transcriptional start and stop sites. There Is extensive protein and DNA sequence homology between cellulase genes cloned from the same organism but convincing homology between cellulases or their genes from different species has not been found (Beguln et al., 1986a; M8cKay et al., 1985; Warren et al., 1986). Short peptide sequences appear to be highly conserved among cellulases from the same organism. The seven amino acid peptide P-T-P-T-P-T-T Is tandemly repeated three times in the C-terminus and the N-terminus of an exoglucanase and an endoglucanase from C. fimi, respectively (O'Neill et al., 1986; Wong et al., 1986). A 23 amino acid peptide direct repeat is also found in the pelA and ce]B gene of Clostridium thermocellum (Beguin et al., 1985; Grepinet and Beguin, 1986). The function of these sequences is not known but they may be involved in substrate binding.  26  Table IV. Coding sequences of 7 cloned cellulase genes deduced from their DNA sequences  Organism  Gene  No. of  No. of codons in  introns  Precursor Signal peptide  Reference  peptide  C. pelliculosa  fl-glucosidase 0  825  20  Kohchi andToh-e, 1985  C.fjmi  cex  0  484  41  O'Neill etal., 1986a  C. fimi  cenA  0  449  31  Wong etal., 1986  C. thermocellum  celA  0  488  32  Beguinetal., 1985  C. thermocellum  celB  0  563  27 or 32  T. reesei  CBHI  2  497  17  Shoemaker et al., 1983  T. reesei  EGI  2  459  22  Shoemaker etal., 1984  Grepinet 8nd Beguin, 1986  27  (B) Exploitation of recombinant cellulases  The potential for the industrial exploitation of cellulases is, in my opinion, the major driving force in cellulase research. Production of fuels or feedstocks by the fermentation of cellulose is only one facet of the industrial exploitation of cellulases. Mandels (1985) h8s summarized a variety of current and potential applications of cellulases. Most of the applications are in the food industry (i.e. the removal of cell walls and crude fiber from plant material to facilitate food processing). The annual Japanese production of cellulases in the 1970's for use in the food and beverage industry was 45 tons (Yamada, 1977). Thus, the successful industrial exploitation of cellulases involves their production in large quantities. Several classical ways which have been used to enhance cellulase production by microorganisms include: (i) the optimization of growth media; (ii) the identification of new cellulase producing species; and (iii) the generation of hyperproducing mutants of existing strains by random mutagenesis and screening. In addition, recent developments in recombinant DNA technology have provided a major opportunity to greatly increase the expression of most cloned gene products. These developments include the ability to isolate and then reintroduce the genetic sequences back into the original host or to introduce them into an unrelated host. This general approach h8s been used to obtain the high level expression of many other scientifically and commercially valuable proteins. Examples include human growth hormone.insulin, and interferons (rev. in Bollon, 1984). Yields of the product often exceed 5$ and can reach up to 50* of the total cellular protein. In addition to the overproduction of recombinant cellulases, recombinant DNA techniques are currently being applied in order to (i) optimize the secretion of cellulases from the heterologous host (Skipper et al., 1985); (ii) enhance the biological properties of enzymes through enzyme engineering; and (iii) introduce cellulase genes into hosts with unique metabolic capabilities such as nitrogen fixation (R. A. J. Warren, personal communication) and photosynthesis (Johnson et al., 1986) and ethanol production (Skipper et al., 1985). Much of the original research in this thesis deals with the overproduction and the secretion by E. cpjiof the C. fimi exoglucanase. The following  28  section of the Introduction provides an overview of the various aspects of obtaining and optimizing heterologous gene expression and secretion by E. coli  (1) Overproduction  The factors governing gene expression at the molecular level are Incompletely understood In prokaryotes and eukaryotes. Problems In the heterologous expression of cloned genes have been encountered at the transcriptional, translational and post-translational levels. For cloned genes studied thus far, obtaining high level expression has often required unique solutions (rev. In Bollon, 1984;Burnett, 1983; Derynck etal., 1983; Masut etal., 1983 ;Shatzman etal., 1983). In general, previous approaches to the elevation of heterologous gene expression by use of recombinant DNA techniques has Involved the following areas: (1) characterizing the structure of the gene; (2) increasing its copy number; (3) optimizing Its transcriptional efficiency; (4) optimizing Its translational efficiency; (5) overcoming toxicity of the protein product to the host; (6) stabilizing the protein product to host proteases; and (7) routing of the protein to the preferred cellular compartment. Studies in these areas as concerned with the expression of a gene in a heterologous prokaryotic host, particularly £. coli, will now be discussed In detail.  (a) Characterization of the gene In general, characterization of gene structure Is necessary because recombinant DNA procedures that are used to obtain increased expression of cloned genes require the precise modification, deletion or substitution of DNA sequences determined to be Important for gene expression. Data that should be available about the gene include: (1) the transcriptional orientation of the gene on a DNA fragment; (2) the Identification of transcriptional and translational regulatory signals on the DNA fragment; (3) the nucleic acid sequence; (4) the translational start and stop sites and the reading frame of the coding sequence; (5) the ability of the transcriptional 8nd translational gene expression signals to function in the heterologous host;  29  and (6) features of the primary amino acid sequence that may be Involved In post-translational modifications such as signal peptides for protein secretion, cysteine residues for disulfide bond formation, sites of glycosylation and substrate binding and active sites of the enzyme. Determination of the nucleic acid sequence of a gene and Its analysis often can provide sufficient information with respect to gene structure. However, independent approaches may be required to confirm conclusions deduced from nucleic acid sequence data. For example, the Identity of putative regulatory sequences can be verified by analyzing the effects of the MvllCQ deletion of these sequences upon expression and partial protein sequence may be required to Identify the coding sequence.  (b) Host organism for the expression of recombinant cellulases  At present the choice of hosts for the overproduction of recombinant cellulases is mainly limited to £. coli, Bacillus subtilis and Saccharomvces cerevisi8e. Cellulases have been expressed in other hosts Including the Dhotosvnthetic Rhodobacter capsulata (Johnson et al., 1986), the nitrogen fixing Klebsiella pneumoniae (RAJ. Warren, personal communication) and the Industrially important Brevibacterium lactofermentum (F. Par8d1s,personal communication). However the genetic engineering techniques and vectors for these organisms are rudimentary. A logical choice for the expression of cloned cellulases would be the original host. Vectors and protocols for the reintroduction of genes into cellulolytic organisms are available for Thermomonosoora YX . Erwinia chrvsanthemi and Bacillus subtilis (Robson and Chambliss. 1986a). Biologically active cellulases have been expressed in E. coli, Bacillus subtilis and Saccharomvces cerevlslae (e.g. Whittle et al., 1982; Robson and Chambliss, 1986a; Skipper et al., 1985; see Table 11). The main advantage of these host organisms Is that there has been extensive research into their genetics and physiology which has led to their extensive use for heterologous gene expression (for reviews see Broach et al., 1983; Dubnau, 1983; Kingsman et al., 1985; Masui et al., 1983). Since the degradation of cellulose requires the association of the  30  cellulases with the substrate, those hosts which readily export the enzymes are a natural choice for Industrial use. Bacillus subtilis and Saccharomvces cerevislae have the advantage .that they both secrete proteins to the extracellular environment. In contrast, the outer membrane of E. coli precludes extracellular protein secretion except In rare circumstances (Pugsley and Schwartz, 1985). However, E. coli remains a useful host for overproduction and Isolation of limited quantities of recombinant cellulases for scientific purposes. Another consideration In the selection of a host for the heterologous expression of cellulases Is the ability of the host to transform the cellulose-derived glucose Into a useful product. For example, yeast have the ability to ferment glucose to ethanol.  (c) Vectors Once a cellulase gene has been sufficiently characterized and a host for Its expression chosen, a vector must be used for the Introduction of the gene Into a host. Foreign genes can be Inserted into the host chromosome but the use of autonomously replicating vectors for the expression of foreign genes offers more opportunities to optimize the expression of a gene. Table Y lists a selection of expression vectors designed specifically for heterologous gene expression. Both bacteriophage and plasm ids have been successfully used for overproduction of products (Bollon, 1984). When choosing a vector, consideration should be given to its copy number, stability, regulatory signals available for gene expression, and selectable markers. Because of the multicopy nature of plasmid and phage vectors, insertion of a gene into these vectors results in an increase 1n the copy number of that gene and, usually, a corresponding increase in the yield of the gene product. This effect Is exploited with runaway replication plasmids which contain temperature-sensitive mutations in their replication control regions that permit unregulated plasmid replication and very high copy numbers at the non-permissive growth temperature. For example, following thermoinduction of replication of the runaway replication plasmid pCP3 for 2 hours, over 505B of total cellular DNA is plasmld DNA (Remaut et al., 1983).  Table V. Examples of specialized £. eo_li expression vectors  Expression Vector  Vector  Gene expression regulatory siqnals Transcription  type  Translation  Secretion  Reference  signals  Promoter Regulation pUC8 to 19  plasmid  lac  chemical  lacZ r.b.s.-ATG  none  Messing, 1983  M13mp2-19  phage  lac  chemical  lacZ r.b.s.-AT6  none  Messing, 1983  pPLc236  plasmid  XP  thermal  none  none  Remautetal., 1981  pCQV2  plasmld  XP  thermal  cro r.b.s.-ATG  none  Queen, 1983  pBN  plasmid  trp  chemical  trp r.b.s.-ATG  none  Nichols and Yanofsky, 1983  PDR540  plasmid  tac  chemical  none  none  Russell and Bennett, 1982  pINIII-ompA  plasmid  lop-lac  chemical  ompA r.b.s.-ATG ompA siqnal peptide Ghrayebetal., 1984  PYK283  plasmid  PhoA  chemical  PJ30Ar.b.s.-ATG pJ}pA.signal peptide  L  R  Miyakeetal., 1985  32  (d) Regulatory signals and elements required for gene expression  (1) Transcriptional elements and factors  The ability of £. coli to recognize transcriptional and translational signals In heterologous DNA Is variable. Often the efficiency of heterologous gene expression signals In £. coli is low compared to the most efficient E. coli signals (Hager and Rablnowltz, 1985; Kozak ,1983; McLaughlin et al., 1981). Thus, following host and vector selection, the next step for obtaining high- level heterologous cellulase expression In £. coli Is the placement of the gene under the control of optimal transcriptional and translational elements. Transcription Initiation and elongation In E. ooliare complex and Incompletely understood processes (for reviews see Gold et al., 1981; Kozak, 1983; Rosenberg and Court, 1979). For transcription Initiation In the bacterial species studied thus far a transcriptional promoter Is necessary. The promoter contains at least two conserved regions of DNA sequence, termed the"-35" site and the"-10" site or Pribnow box, which are positioned upstream of the site of transcriptional initiation (Rosenberg and Court, 1979). These DNA sequences, the spacing between them, and the less conserved sequences surrounding them modulate the 'strength' of the promoter. Operator regions, which are DNA sequences Involved in the regulation of transcriptional Initiation, may be located within the promoter sequence. Operator regions associate with factors, usually proteins, that can prevent or stimulate initiation at the promoter. Operator regions are very useful for heterologous gene expression because constitutive expression of a gene product may be deleterious to the host cell and controlled expression may minimize degradation of the protein product. At the most basic level then, transcription requires that an efficient, regulatable promoter for the binding of RNA polymerase be positioned to the 5' side of the cellulase gene. Thus, the transcriptional efficiency of cloned genes can often be increased by substituting 'strong' promoters for less efficient or non-functional promoters. A number of efficient E. coli Phage and bacterial promoters have been isolated (de Boer and Shepard, 1983c; Table V). In addition,  33  chemically synthesized promoters have been shown to function in vivo (Rossi et al., 1983). Some examples of natural and synthetic promoters Include those from phage lambda (Remaut et al., 1981; Rosenberg et al., 1983); the promoters from the tryptophan (Nichols and Yanofsky, 1983) and lactose (Messing, 1983) operons of E. coli., and the synthetic tac (Russell and Bennett, 1982; de Boer et al., 1983a) promoters. When fused to the gene of interest these promoters usually direct high levels of transcription. As mentioned above, It Is desirable that the promoter be regulatable, and that the initiation of transcription be balanced by transcriptional termination signals (Gentz et al., 1981). All of the promoters listed prevlosly contain operator control sequences and are either chemically or thermally inducible (Table V). This allows the regulated expression of a product whose overproduction may be lethal to the cell. Gentz et al. (1981) observed that strong, constitutive promoters cloned In plasmld vectors often yielded highly unstable plasmlds. These Investigators postulated that 'read-through' transcription, Initiating at the strong promoter sequences, Interefered with the origin of plasmld replication and led to plasmld instability and curing. Plasmld stability could be Increased by positioning transcription terminator signals Immediately downstream of the promoter sequences (Gentz et al., 1981). In bacteria transcriptional terminators often contain GC-rlch sequences that can form stable hairpin structures which are followed byAT-rich sequences (Rosenberg and Court, 1979). If not present on a cloned DNA fragment, terminator sequences contained on synthetic double stranded oligodeoxyribonucleotides can be fused to the 3" end of a cloned gene (Gentz et al., 1981). Obtaining efficient transcription Initiation and termination of a foreign gene does not ensure high yields or even production of the protein as has been documented by several recent studies (Looman et al., 1985 ;Stanssens etal., 1985 and 1986). Transcription and translation are coupled in prokaryotes. Thus, the ability or Inability of ribosomes to initiate and efficiently translate a mRNA can affect the final level of mRNA (Adhya and Gottesman, 1978; Looman et al., 1985; Stanssens et al., 1985 and 1986). The factors that Influence the translatabillty of a mRNA are discussed next.  34  (11) Translational signals and factors  Minimal requirements for initiation of translation in prokaryotes are a ribosome binding site or Shine-Dalgarno sequence (SD; Shine and Dalgarno, 1974), an Initiating codon and the proper configuration of these regulatory signals with respect to each other (Gold et al., 1981; Kozak, 1983). The SD sequence 1s a conserved 4 -bp purlne-rich stretch located between 3 and 12 nucleotides upstream of the translational start site. The consensus SD sequence is derived from part of 5'-GGAGG-3' and is complementarity to the 3' end of the 16S rlbosomal RNA. The SD sequence is more highly conserved in Gram-positive than in Gram-negative prokaryotes (McLaughlin et al., 1981; Murray and Rablnowitz, 1982). The translational start codon is most often ATG but GTG and TTG start sites are known (Gold et al., 1981; Kozak, 1983). Prokaryotic translational initiation sequences from a single bacterial species or from different species of bacteria do vary 1n their 'strength' (Gold etal., 1981; Kozak, 1983; Ray and Pearson, 1974 and 1975). Systematic studies of the factors regulating translational initiation have been carried out by analysing the level of expression obtained using synthetic oltgodeoxyrlbonucleotides coding for SD-ATG sequences (de Boer et al., 1983a and b)and based on these findings synthetic oligodeoxyrlbonucleotides encoding optimal SD-ATG sequences have been made commercially available (Pharmacia Ltd, Dorval, Quebec). These sequences can be fused to a gene of interest, but, as discussed below this does not always ensure high level translation. In addition to the SD sequence and initiation codon, many studies have shown that the mRNA sequences surrounding the SD-ATG region can greatly Influence translational initiation (Ganoza et al., 1982;Gheysenetal., 1982;Hall etal., 1982;Hu1 etal., 1984; Kasteleinetal., 1983; Looman et al., 1985; McCarthy et al., 1986; Schoner et al., 1984; Stanssens et al., 1985 and 1986; Whitehorn et al., 1985). These studies have demonstrated that the 5' untranslated sequences and coding information downstream from the ATG greatly influences the level of transcription and translation of an mRNA. Although it is not clear what prevents efficient  35  translation of abundant messages, the most recent studies point to the inability of rlbosomes to Initiate translation on the mRNA (Stanssens et al., 1986). Based on conclusions derived from secondary structure models of the 5' ends of mRNAs and the ability of these sequences to Initiate translation, It appears that the start codon and the rlbosome binding site must be 'accessible' to the rlbosome and not be burled In secondary structure (Gheysen et al 1982; Hall et al., 1982; Kasteleln et al., 1983; Stanssens et al., 1985). Stanssens et al.( 1986) recently demonstrated that minor variations In the untranslated leader region of the ]acZ gene led to a differential rate of translation initiation. These 5' mRNA leader variations also dramatically affected the steady-state amount of full-length lacl mRNA. These Investigators suggested that the translation Initiation regions that Initiated translation poorly led to an extended spacing between the RNA polymerase and the elongating rlbosome. This spacing caused transcriptional polarity (Adhya and Gottesman, 1978) by Increasing the extent of premature termination. In conclusion, If the heterologous gene to be expressed requires either a start codon or SD sequence It can be fused to DNA fragments carrying such sequences but the translatablllty of a chimeric SD-ATG sequence Is highly variable from one construction to another. Several groups have suggested, based on theoretical ( Grantham et al., 1981; Grosjean and Flers, 1982; Ikemura, 1981) and experimental findings (Bonekamp etal., 1985; Robinson et al., 1984; Varenne et al., 1984), that codon usage may play an Important role in regulation of gene expression and in heterologous gene expression. The basis for this argument Is that for efficient translation, the codons used In the foreign gene should be consistent with the levels of the cognate tRNAs in the host cell. Highly expressed E. coli genes show a marked bias In codon usage and this bias 1s directly reflected In the cellular tRNA concentrations.That 1s, highly expressed genes most often use codons for which the concentration of the cognate tRNA Is very high (Grantham et al., 1981; Grosjean and Flers, 1982). Often heterologous coding sequences do not show the same codon usage bias as the host organism. The argument follows that the cognate tRNAs for the codons used In the foreign gene may be so rare In the host cell that translation Is slowed. The extent of Influence of codon usage on gene expression Is not clear (Grantham et al., 1981; Grosjean and  36  Flers, 1982). It Is common, when engineering high level heterologous expression, to synthesize an entire gene based on optimal codon usage patterns of the host (Bollon, 1984).  (2) Secretion and export of recombinant cellulases  (a) Host organism The blodegradatlon of cellulose by cellulases requires the extracellular synergistic action of the different components of the cellulase system (Coughlan, 1985b). Thus, for the Industrial application of cellulases, the recombinant enzymes must be secreted to the extracellular environment. In contrast to E. coli, yeast and 6ram-pos1t1ve bacteria such as Bacillus subtilis are excellent host organisms for expression and secretion of cellulases, because they actively secrete a variety of proteins Into the culture medium. However It should be mentlonned here that E. coli does export some proteins such as collcins (Pugsley and Schwartz, 1985) and that several cellulases appear to passively leak through the outer membrane of E. coli (Collmer and Wilson, 1983; Kolde et al., 1986; Sashihara et al., 1984). In yeast, a second advantage for the expression of cellulase Is the ability to glycosylate secreted proteins (Ballou, 1982). This 1s Important as most fungal cellulases are glycosylated In their parental organism and their glycosylatlon is apparently essential to their stability and activity (Merlvuorl etal., 1985; rev. In Coughlan, 1985b). As mentlonned previously (page 22) this does not appear to be the case for the C. fimi enzymes.  (b) Signals and information required for secretion Secreted and exported proteins in prokaryotes are synthesized in a precursor form possessing a typical prokaryotic signal sequence which is absolutely required for secretion (rev. In Inouye and Halegoua, 1980; Mlchaelis and Beckwlth, 1982; Pugsley and Schwartz, 1985; Randall and Hardy, 1984; Silhavy et al., 1982). Other less clearly defined information contained within the mature secreted peptide is probably also necessary (Benson et al., 1984; Mlchaelis and Beckwith, 1982; Silhavy et al., 1982). Although amino acid sequences of signal peptides are  37  seldom Identical they consist of a positively charged hydrophillc amino terminus, a hydrophobic central region and a length of 15 to 30 amino acids for 6ram-negat1ve secreted proteins and up to 48 amino 8c1ds for Gram-positive secreted proteins. The signal peptides are removed during protein translocation by a signal peptidase (Pugsley and Schwartz, 1985; Randall and Hardy, 1984). Other information, termed export and sorting signals, for export and localization of the secreted protein Is apparently present In the mature protein (Benson and Sllhavy, 1983; Benson etal., 1984; Palva and Sllhavy, 1984). The nature of the secretion mechanism In bacteria Is unclear. According to the signal hypothesis of Blobel and Dobbersteln (1975), exported proteins are made In a precursor form with an amlno-termlnal signal sequence. The translation of the mRNA for such a peptide Initiates In the cytoplasm and may be associated with the membrane. As translation proceeds, the polypeptide Is translocated across the inner membrane (Pugsley and Schwartz, 1985; Randall and Hardy, 1984). The signal peptidase, which Is associated with the outer surface of the Inner membrane, recognizes a precise cleavage site on the secreted precursor protein and cleaves the signal peptide from the precursor at this site. The signal sequences appear to be somewhat Interchangeable between Gram-negative and Gram-positive prokaryotes and between eukaryotes and prokaryotes. This conclusion Is based on the observations that In some cases heterologous prokaryotlc signal peptides are secreted and processed by both prokaryotlc (Pugsley and Schwartz, 1985) and eukaryotic cells (Roggenkamp etal., 1981 ;Baty etal., 1981 ;W1edmann etal., 1984) and vice versa (Ta1m8dge etal., 1980; Palvaetal., 1983; Nagahari etal., 1985 ;Shiroza etal., 1985; Gray etal., 1985). Heterologous signal peptides may function In bacteria but the processing of the signal peptide may not always be Identical to the wild type situation. Plasmld vectors have been developed for the secretion of heterologous proteins In E. coli (TableVandGhrayebetal., 1984; Miyakeetal., 1985; Nagahari etal., 1985), B. subtilis (Fahnestock and Fisher, 1986; Palva et al., 1983; Vasantha and Thompson, 1986), and S. cerevisiae (Singh et al., 1984; Skipper et al., 1986).The systems all employ plasmids carrying  38  cloned signal peptides for secreted and exported proteins. Those employed have Included the ompA (Ghrayebetal., 1984),pmnF(Nagaharl etal., 1985)andphpA(Okaetal., 1985 )signal peptides of E. coli, the alpha-amylase and protease signal peptides from B_. subtilis (Palva et al., 1983; Vasantha and Thompson, 1986), and the killer toxin and mellbiase signal peptides (Skipper et al., 1985 and 1986) from S. cerevislae. The secretion vectors for Bacillus subtilis (Fahnestock and Fisher, 1986; Palva et al., 1983; Vasantha and Thompson, 1986) and S. cerevislae (Slnghetal., 1984; Skipper etal., 1986) and one vector for E. coli (Nagaharl et al., 1985) result In the selective secretion to the culture medium of heterologous proteins whereas all other E. coli secretion vectors lead to the periplasms localization of the foreign protein (Ghrayeb et al., 1984; Mlyake et al., 1985).  39  MATERIALS AND METHODS I  Bacterial strains, phages and pl8smids  A list of the bacterial strains, phages and plasmids used in these studies is given in Table VI. E. COVLC600 (thr-1 leu-6 thi-1 SUDE44 lacvY 1 tonA21) and the plasmids pel857 and pCP3 were gifts from Erik Remaut and Walter Fiers (Remaut et al., 1983). The plasmid pDR540 (Russell and Bennett, 1982) was purchased from Pharmacia Inc. (Dorval, Quebec, Canada). Stock cultures of bacteria were maintained at -20°C or -80°C in LB medium containing 15% glycerol.  II Media The preparation of media and supplements was as described by Maniatls et al. (1982), Miller (1972) and Messing (1983). LB medium contained per liter; 10 g of B8cto-tryptone, 5 g of Bacto-yeast extract and 5 g of NaCl. 2XYT medium contained per liter; 16 g of B8cto-tryptone, 10 g of Bacto-yeast extract and 5 g of NaCl. If necessary the pH of the media were adjusted to 7.2 with NaOH. M9 medium contained per liter; 6 g Na2HP0 ,3 g KH P0 ,0.5 g NaCl, 1 g NH C1,5 g 4  2  4  4  Casamlno acids, 1 mgof thiamine with the addition, after autoclaving, of 10 ml of 0.01 M CaCl2,2 ml of 1 M MgS0 and 10 ml of 20* glucose. Agar plates contained 1.558 agar. 4  III Buffers The composition and preparation of the buffers and solutions employed in these studies are described by Maniatis et al. (1982). Phosphate buffered saline (PBS), pH 7.2, contained per liter; 8 g NaCL, 0.2 g KC1,0.2 g KH P0 and 2.17 g Na2P0 (7 H 0). For immunoadsorbent 2  4  4  2  chromatography the NaCl concentration of the PBS was 0.5 M (high salt-phosphate buffered saline; HS-PBS)  IV Biological screening for exoglucanase activity E. coli clones carrying recombinant plasmids expressing Exg were picked onto both a master LB-agar plate containing the appropriate antibiotic and an LB-agar plate containing 100  40  Table VI. Bacterial strains, phage and plasmids  Bacterial strain  Genetic characters  L f i S l i C600  M-1  Reference  )euB6 lacYI tonA SUPE44  Appleyard, 1954  E.coJtPM191  dra drm thr leu thi lacY recA56 supE  Meacock and Cohen, 1980  £.£2!i JM83  ara A(lac-oroAB) rosL(=strA) 080 lacZAM15  Yanisch-Perron et al., 1985  Lcpl JM101  suoE thi A(lac-oroAB)  Phaae  Genetic characters  Reference  M13mp10  lac  Messing. 1983  M13mp11  lac  Messing, 1983  Plasmid  Genetic characters  Reference  pEC-1  ApR  Whittle etal.. 1982  PBR322  ApR TcR  Bolivar et al.. 1977  pUC12  ApR  Messing, 1983  DUC13  ApR  Messing. 1983  pDR540  ApR galK  pATH3  ApR  pINIII omoA-1  ApR. lflci . Plpp -lac  pCP3  ApR, ts runaway replication. O L ^ L *  Remautetal., 1983  pc!857  Km . cl857  Remaut etal., 1983  Russell and Bennett. 1962  +  Spindler etal.. 1984  P^p* +  R  IF' traD36 oroAB lacl<lZAM15] Yanisch-Perron et al.. 1985  +  +  +  Ghrayeb etal.. 1984  41  uM methylumbelllferyl cellobloslde (MUC; Sigma Chemical Co.). The use of methylumbelllferyl derivatives of glucopyranosldes for the quantitative assay of cellulases has been previously described by van Tllbeurgh et al. (1982). After incubation of the plates at 30 to 37°C for 4 to 12 hours the cells producing Exg activity were detected visually; under UV Illumination (300-354 nm) a purple halo surrounded positive colonies. E. coli does not contain those enzymes which hydrolyze MUC. The activity plates were photographed using a Polaroid Land camera loaded with Polaroid Type 57 film. Maximum photographic contrast was obtained by using a yellow filter (code * YK2), placing the plates on a black background and using a hand-held UV lamp. The protocol for the immunological screening of E. coli clones expressing C. Ami cellulases has been described by Whittle et al. (1982).  V Growth and Induction of bacteria Bacteria were grown In LB medium with the addition, after autoclaving, of 0.4* glucose. Selective antibiotic media contained one or more of the following; 50 p.g kanamycln/ml; 75 u.g ampicillin/ml; 15 ug tetracycline/ml. The X P and P promoters were induced thermally L  R  in the following way. After growth at 30°C to O.D.5oo=0.3 the cultures were divided and parallel samples were grown at 30X (nonlnduced) and at 41 *C (induced). The chemlcally-lnducible promoters tac, lac , the hybrid lfip_-lac and trp were Induced in the following way. Cells harboring plasmids containing these promoters were grown at 30 to 37°C in M9 media, supplemented with the appropriate antibiotics and 0.5* glycerol, to an O.D. oo=0.3. The cultures 6  were divided and to one half of each culture either 3-fi-indole acrylic acid (Sigma *l 1625) was 8dded to a final concentration of 5u.g/ml for the induction of the trp. promoter or Isopropyl-B-Dthiogalactopyranoslde (IPTG) was added to a final concentration of 2 mM for the induction of the lac , tac and lpp-lac promoters.  42  VI  DNA manipulations and recombinant DNA techniques  (A) Enzymes, reagents and techniques In general, DNA preparations, restriction enzyme reactions and recombinant DNA techniques were performed as described by Manfatls et al. (1982). Restriction endonucleases, DNA polymerase 1 (Klenow fragment), T4 DNA llgase, mung bean nuclease, deoxyrlbonucleotldes and the portable translation Initiation site (PTIS) were purchased from Pharmacia Inc. (Dorval, Quebec). Calf Intestinal phosphatase was obtained from Boehringer-Mannheim (Dorval, Quebec). All enzymes were used under the conditions suggested by the supplier. All other chemicals and reagents were from BDH Chemicals (Vancouver, Canada) or Sigma Chemical Company (St. Louis, Mo.). Bacterial transformations were carried out according to Manlatls et al. (1982) except for the following modification for strains receiving plasmids containing the leftward promoter (Pj_ ) of the phage X. The competent cells were incubated with DNA, heat-shocked at 34°C for 2 minutes, Incubated In LB medium for 1 hour at 30°C, plated on selective medium and then Incubated at 30°C.  (B) Preparation and purification of plasmld and phage DNA The procedures for the growth of phage M13 and purification of its ds and ss DNA form have been described In detail by Messing (1983). E. poji JM101 (Messing, 1983) was used routinely for the propagation of M13 phage. For the Isolation of plasmld DNA, E. coli containing the appropriate plasmid was grown to stationary phase In LB medium ( 0.2 to 1.01 for large scale preparations and 2 to 10 ml for small scale preparations). The cells were collected by centrifugation and the plasmid DNA was isolated by the alkaline lysis method of Birnboim and Doly (1979) as modified by Manlatls et al. (1982). Large scale preparations of plasmid DNA were further purified by one or two rounds of ultracentrifugation through CsCl-ethidium bromide gradients. The small scale alkaline method for the preparation of plasmld DNA yielded DNA that was sufficiently pure for restriction analysis. However, this DNA required further purification for  43  recombinant DNA manipulations and for DNA sequencing. Thus, the RNA was removed by digestion with T RNAse and precipitation with 2.5 M NH^OAc at 4°C for 15 minutes. The DNA in the 1  supernatant fluid was then precipitated with 2 volumes of EtOH, resuspended in TE buffer and purified on NACS-Prepac columns as described by the supplier (BRL/Glbco Canada).  ( 0 Synthetic ollgodeoxyrlbonucleotldes The deoxyollgorlbonucleotlde 5'-C6A(XACGCT(>\A66A6Gam^CGGCG-3' (30-mer) was synthesized chemically on an Applied Blosystems 380A DNA synthesizer by T. Atkinson using phosphite trlester chemistry, essentially as described (Adams et al., 1983; Atkinson and Smith, 1984). The oligodeoxyrlbonucleotlde was separated from Incomplete products by electrophoresis In a 16* polyacrylamide-7 M urea sequencing gel, located by UV-shadowIng, and extracted from the gel by the crush and soak method (Atkinson and Smith, 1984). The primers were further purified by binding to a Sep-Pak 0\Q reverse phase cartridge (Millipore/ Waters Assoc., Milford, MA) in water, followed by washing with water, and elution with 20* acetonltrile80* water. The DNA sequence of the oligonucleotide was confirmed by a modified (Zoller and Smith, 1983) chemical sequencing method (Maxam and Gilbert ,1980).  (D)  Primer Annealing Reaction and Klenow Extension  Twenty pmoles of M13mp 11 -cex DNA was mixed with a 10 fold molar excess of oligonucleotide primer in 250 ul of buffer containing 50 mM NaCl, 50 mM Tris-Cl (pH 8.0), and 10 mM MgCl2. The DNA was denatured by heating the solution in a 1.5 ml microfuge tube at 1 OO'C for 5 minutes and then allowed to anneal by cooling to 23"C over a period of 2 hours. The reaction mixture was then supplemented to give final concentrations of 500 uM for each dNTP and 1 mM dlthlothreltol. Fifteen units of DNA polymerase I (Klenow fragment) were added to the reaction (final reaction volume of 350 ul). The reaction mixture was incubated at 23°C for 2 hours. The DNA was purified by phenol extraction, precipitated twice with ethanol and then dissolved in 50 ul of 10 mM Tris-Cl, pH 8.0.  44  (E) Hung bean nuclease reaction DNA products from the elongation reactions of M13mp 11 -cex DNA primed with an ollgodeoxyribonucleotlde were treated for 20 minutes at 30°C with 10 units of mung bean nuclease per pmole of M13mp 11 -pex DNA In a reaction volume of 200 ul containing 30 mM NaOAc, pH 4.6,250 mM NaCl, 1 mM ZnCl2 and S% glycerol (Rosenberg et al., 1983). One unit of Pharmacia Inc. mung bean nuclease Is that amount of enzyme which produces one u.g of acidsoluble material per minute at 37 C. This unit definition Is equivalent to 0.004 units as e  described by Laskowskl (1980). The mung bean nuclease was inactivated by the addition of SDS to 0.255 and phenol extraction. The DNA was then precipitated with ethanol and resuspended in an appropriate buffer for restriction endonuclease digestion. In the experiments reported here, the DNA pellet was resuspended 1n 250 ill of buffer containing 150 mM NaCl, 10 mM Trls, pH 7.6, and 6 mM MgCl2, and 20 units of Sail restriction endonuclease and incubated for 16 hours at 37°C. The DNA was deprotelnized by phenol extraction and recovered by ethanol precipitation.  VII  Plasmld and phage vector constructions (A) Construction of M13 Clones  For the deletion of the pex signal peptide by a primer mutagenesis technique, I subcloned a 1.96-kb BamHI-Sall cex-codingfragment into Ml3mp11 andM13mp 10 (Messing, 1983) by standard molecular cloning techniques. The Exg recombinant designated M13mp 11 -cex contained a 1956-bp BamHI-Sall DNA fragment encoding the pex.gene Inserted between the BamHI and Sal' sites of M13mpl 1.  (B) pEC-1.1, pEC-l.ls, pEC-l.ltac , pUC12-1.1cex , pUC13-1.1pex, pUC 12-1.1 cex (748) and pUC 12-1.1 cex (859) The cex gene, on a 6.6-kb BamHI fragment of C. fjml DNA, was llgated into the BamHI site of  45  pBR.322, giving pEC-l (Whittle etal., 1982; Fig. 5). The cex gene was localized toa2.56-kb BamHI-Sall DNA fragment, by deletion analysis, yielding pEC-1.1 (Fig. 5 and 6). The plasmids pUC 12-1. Icex and pUCI 3-1. Icex (Fig. 5 and 6) contained the 2.56-kb fragment from pEC-1.1 positioned In opposite orientations downstream from the promoter-operator region of the E. coli lactose operon (lacZ P/O) In the plasmids pUCI 2 and pUCI 3 (Yielra and Messing, 1982). The plasmld pEC-1.1 tac was a derivative of pEC-1.11n which the tet promoter had been excised as an EcoRI-BamHl fragment and replaced by the lac.promoter of pDR540 (Russell and Bennett, 1982) (Fig. 5). The construction of plasmids pUC 12-1.1(748) and pUC 12-1.1(859) are described In the legend to Fig. 25.  (C) pCP3-cex, pUCI 2-1.1 (737) and pUCI 2-1.1 (PTIS) The plasmld pCP3 was used for the construction of pCP3-cex. It Is a derivative of the runaway replication vector pKN402, which contains the X P[_ promoter adjacent to a multiple cloning site (Remaut etal., 1983). For the construction of pCP3-cex (Fig. 19), the 5" untranslated sequences, the rlbosome binding site (RBS), and the Initiating codon of the cex gene were first removed and replaced with the promoter-operator region, the RBS, and the aminotermlnus of J3-galactos1dase (fiGal) from the E. coliJac operon and then with the RBS-ATG sequences of the PTIS. In the first step, pUC 12-1.1 cjx. W8S cut with SJyJ and BamHI, the staggered ends were repaired with DNA polymerase I (Klenow fragment), and the plasmid DNA was Itgated under dilute conditions to give pUC12-1.1 (737). This manipulation resulted in (i)the inframe fusion between codon 2 of the Exg leader sequence and codon 11 of the alpha-fragment of fiGal encoded by pUC 12 ; (11) the regeneration of the S&l cleavage site; and (ill) the replacement of the cex Initiating codon with a BamHI cleavage site. The nucleotide sequence and deduced amino acid sequence of thefiGal-Exgfusion region of pUCI 2-1.1 (737) are shown in Fig. 12. Plasmld pCP3-pex was obtained by ligating the following fragments: (i) a 1.7-kb cex-containing BamHI-Hindlll fragment which W8S excised from pUC 12-1.1(737); (ii)the 17-bp synthetic PTIS with EcoRI and BamHI cohesive ends; and (iii) pCP3 plasmid DNA which had been cleaved  46  with EcoRI and Hindi 11. In the resulting plasmld, termed pCP3-pex, PL Initiates transcription across the cex gene which is fused in-frame to the initiating codon of the PTIS. To obtain pUCI 2-1.1 (PTIS), pUCI 2-1.1 (737) was cut with EcoRI and Bam.HI and the 17-bp PTIS with EcoRI and BamHI cohesive end (Fig. 19) was inserted. This procedure resulted in the in-frame fusion of the second codon of the cex. leader sequence to the initiator ATG of the PTIS (Fig. 17).  VIII  Gel electrophoresis of DNA Agarose gels (0.6-2.0%) and polyacrylamidegels (5-8SS) were used for  the analysis and preparation of large (0.5-12 kb) and small (30-800 bp) DNA fragments, respectively. For preparative purposes, large DNA fragments were separated on 1.0-1.551 low melting temperature agarose gels (Bio-Rad). The bands of DNA were located by staining with ethidium bromide and excised. The DNA was recovered by heating the gel slices at 65°C for 5 min and extracting the agarose solution with phenol and phenol-chloroform-isoamyl alcohol (25:24:1, v/v/v). The DNA in the aqueous phase remaining after the organic extractions was precipitated with ethanol and resuspended in TE. DNA fragments in polyacrylamide gels were electrceluted from gel slices using the dialysis bag technique described by Maniatis et al. (1982). A 50 mM Trls-OH/ 50 mM boric acid/ 0.002 M Na^EDTA buffer system was employed for analytical purposes and a 0.04 M Trls-acetate/ 0.001 M Na^EDTA buffer system was employed for preparative work (Maniatis et al., 1982).  IX  DNA sequence determination  The DNA sequence of the pex gene was determined by both the Maxam and Gilbert (1980) and the Sanger et al. (1977) procedures. For chemical sequencing, DNA fragments were 3' endlabelled using the Klenow fragment of DNA polymerase I with one labelled [c*- P]-dNTP 32  (approx. 3000 Ci/mmol; New England Nuclear, Lachine, Quebec). As suggested by Ruther et al. (1981), fragments cloned into the pUC series plasmids were sequenced directly by the chemical  47  method after cleavage with a second restriction enzyme. Otherwise DNA fragments labelled at only one end were prepared by secondary restriction enzyme digestion followed by separation of the DNA fragments by polyacrylamlde or agarose gel electrophoresis and electroelution of the DNA fragments as described by Maniatis et al. (1982). For Sanger's dideoxy sequencing method, fragments were cloned into either M13mp 10 or M13mp 11, or the plasmids pUCI 2 and 13. For dideoxy sequencing of pUC vectors, plasmid DNA was extracted from 1.5 ml of stationary phase culture (Blrnbolm and Doly, 1979) and further purified by chromatography on NACS-Prepac columns (Glbco/BRL). Samples from the chemical and enzymatic sequencing methods were analyzed on 6 or 8SS denaturing (8 M urea) polyacrylamlde gels. The sequences obtained were recorded, aligned and analyzed using either the SEQNCE software program developed by A. Delaney (Delaney Software Ltd., Vancouver, B.C.) run on an AHMDAHL V8 mainframe computer or the DNA Inspector II software program (Textco, West Lebanon, New Hampshire) run on an Apple Macintosh 512K microcomputer.  X  Protein purification (A) Cell fractionation  The cellular locations of I30al, 13-lactam8se, Exg, and 66a1-Exg were determined as follows. Cells harboring an appropriate plasmld were grown by inoculating 0.05 ml of a fresh overnight culture into 50 ml of LB supplemented with 0.5* glycerol and 50 ug amplcillin/ml. The cultures were grown to a final Klett reading of 180, harvested by centrlfugatlon, washed and resuspended In 100 mM potassium phosphate buffer, pH 7.0. These cell suspensions were divided Into two parts, and an aliquot of each was used for the fractionation of the cytoplasmic and periplasm^ compartments by 8 cold osmotic shock procedure (Nossal and Heppel, 1966). In this procedure, the cells were collected by centrlfugation, exposed to 0.5 M sucrose-1 X 10~ M 4  EDTA, sedimented and then rapidly dispersed in cold 5 X 10" M MgCl2. The material released into 4  the fluid was termed the periplasmic fraction. The cytoplasmic fraction refers to the soluble material obtained from the osmotically shocked cells after breaking the spheroplasts by passage  48  through a French pressure cell at 10,000 psi. The cell extract fractions were prepared by passing the remaining aliquot of the Initial cell suspensions through a French pressure cell at 10,000 psi. Cell debris and unbroken cells were removed from the extracts by centrlfugation at 50,000 xg for 1 h.  (B)  Immunoadsorbent chromatography purification (1)  Preparation of the Immunoadsorbent  The preparation of rabbit antiserum raised against the native C_. fimi Exg has been described (Whittle etal., 1982). The Immunoglobulin 6 fraction of this serum was partially purified by two sequential precipitations with 40* ( N H ^ ^ (Garvey et al., 1977). The immunoglobulin 6 fraction then was linked covalently to cyanogen bromide-activated Sepharose 4B (Pharmacia) by the method of Cuatrecasas et al. (1968) but with some modifications. All procedures using CNBr were performed In a vertical flow fume hood. The CNBr (15 g CNBr: 30 g Sepharose 4B) was dissolved In distilled water by gentle stirring. The pH was then adjusted to and maintained at pH 11 by titration of 50* (w/v) NaOH into the gel slurry. The temperature of the slurry was held constant at 20*C by the addition of crushed Ice directly to the gel slurry. The reaction was completed when the pH stabilized at pH 11. When the reaction had ended the activated S4B (S4B*) was washed with 3 L of cold water and 21 of 0.1M NaHCC^. IgG preparations of rabbit antisera prepared(i) against supernatant fluids of C.fimicultures grown on CMC and Avlcel (anticellulase) or (ii) against whole cell extracts of E. coli (anti-E. coli) were diluted to a final protein concentration of 5mg/ml in 0.1M NaHC0 -0.5M NaCl. The antiserum preparations were 3  added to the S4B* at a ratio of 1 g protein per 30g of S4B* and then mixed gently at 4"C for 24 h. The gel was then washed with 21 of phosphate buffered saline containing 0.5M NaCl (HS-PBS). The excess active groups were blocked by alternate washings with 0.2 M glycine-HCI, pH 2.8, and 1M ethanloamine-0.1M N8HCO3-O.5 M N8C1. The antibody-agarose gel was then washed extensively with HS-PBS, packed into chromatography columns and stored 8t 4°C. The S4B-anticellulase antibody affinity column and the S4B-anti-E. coli antibody affinity column were  49  prepared using 3 g of rabbit anti-ceilulase IgG (bed volume of approx. 90 mis) and 160 mg of rabbit anti- E. coli IgG (bed volume of approx. 12 mis).  (2)  Immunoadsorbent chromatography  For the large scale purification of Exg, the cells were harvested from a 201 culture of E. coli C600 (pEC-1.1). Cells were grown to a cell density of 5-10 X 10 cells/ml In LB media 8  containing 0.2* glucose and 50 ug amplclllln/ml. The cells were collected by continuous flow centrlfugatlon and resuspended in PBS. The cells were then subjected to osmotic shock (Nossal and Heppel, 1966) to release periplasm 1c proteins or disrupted by passage through a French pressure cell at 10,000 psf to give a total cell lysate. At this point phenylmethyl sulfonyl fluoride (20 mg/ml in 95* EtOH), a serine protease inhibitor, and sodium azide (2* solution in H 0) 2  were added to final concentrations of 20 u.g/ml and 0.02*, respectively. The cell extracts were clarified by centrlfugatlon of the extracts at 30,000 X g for 30 min at 4°C. Approximately 6065* of the total cellular Exg activity W8s obtained in the periplasmic fraction and 20-25* of the total original Exg activity remained in the spheroplasts. Although there was an overall loss of 1020* of total Exg activity with the osmotic shock procedure, a six fold Increase in Exg specific activity was obtained In the shockate fluid due to the low protein content of the periplasmic fraction. Immunoadsorbent chromatography purification consisted of sequential passage of source material over a S4B-anti-cellulase antibody affinity column and then a S4B-anti-E. coli antibody affinity column. Whole cell extracts or osmotic shock fluids prepared from E. coll cells expressing Exg were loaded directly onto the S4B-anti-cellulase antibody affinity column. Elution of unbound proteins was performed by gravity flow with 1 to 21 of HS-PBS. Bound material was eluted by application of 3 M NaSCN, pH 7.0. On the basis of pNPCase assays, performed immediately following elution of the column, fractions containing Exg activity were pooled, concentrated by ultrafiltration on Amicon PM-30 membranes and then dialyzed against PBS-0.2* NaN . In the eluted fractions containing Exg activity, some contaminants which bound 3  50 nonspecifically to the S4B-anti-cellulase antibody affinity column were coeluted. These contaminants were removed by passage of the material over the S4B-anti-E. coli antibody affinity column. Exg assays revealed that the Exg was contained in the void volume of this column. These fractions were pooled, concentrated, dialyzed against PBS-0.2* NaN3 and stored at -20°C.  (C) Isolation of intracellular insoluble Exg aggregates A 50 ml culture of C600( pcl857)( pCP3-cex) was grown at 30°C In LB medium containing the appropriate antibiotics and induced at 41 °C for 90 minutes. The bacteria were pelleted by centrlfugatlon, washed twice In 50 mM Tris, pH7.5,0.1 mM EDTA, and disrupted by sonlcatlon. The Exg was separated from the soluble cellular material by low-speed centrifugation (3000 X g for 5 minutes). The insoluble material from this centrlfugatlon contained the Exg aggregates. The precipitate was washed three times with 50 mM Tris, pH 7.5,0.1 mM EDTA or until no Exg activity could be detected In the supernatant.  XI Electrophoresis of proteins Proteins were separated by electrophoresis in 10* polyacrylamlde gels in the presence of 0.1 * SDS by a modification of the method of Laemmli (1970). For the analysis of proteins in whole cell extracts, the cells from 1 ml of culture were resuspended In 250 pi of SDS-PA6E loading buffer (62.5 mM Tris-Cl, pH 6.8, 2* SDS, 10* glycerol, 5* 2-mercaptoethanol, 0.002* bromphenol blue), boiled for 5 minutes, and then centrifuged for 5 minutes in an Eppendorf microcentrifuge. A total of 12.5 to 25 pi of the solubilized material, equivalent to 50 to 100 pi of the culture, were loaded per slot. After electrophoresis, proteins were stained with Coomassie brilliant blue R-250. To estimate the percentage of Exg in a sample, the stained gel was dried on a dialysis membrane and scanned with a Helena Laboratories densitometer.  51  XII Assays of enzyme activities and protein concentrations (A) p-nltrophenylcelloblosldase assay Exg activity In crude cell extracts was measured by following the hydrolysis of pnitrophenyl-fi-D-cellobioslde ( pNPC; *N5759, Sigma Chemical Co., St. Louis, Mo.) as previously described (Gilkes et al., 1984b). Briefly, the hydrolysis of the agluconlc bond (see Fig.3) of pNPC was measured spectrophotometrlcally by following the release of p-nitrophenol (pNP). The pNPCase reaction conditions were as follows: 50 u.1 of enzyme extract were mixed with 250 ul of pNPC (2 mg/ml) in 50 mM KP0 , pH 7.6 and 700 ul of 50 mM KP0 , and then 4  4  Incubated at 37"C until a yellow color developed. The reaction was developed by the addition of 1 ml of 2% Na2C03 and the absorbance at 410 nm was measured. pNPCase activity was calculated by the following formula:  pNPCase=  [0.D.410] [volume of stopped reaction (ml)]  [volume of enzyme (ml)] [time of reaction (min)] [molar extinction coefficient]  The molar extinction coefficient of pNP at 410 nm is 18.5 ml/umol/cm (Deshpande et al., 1984a). One unit of pNPCase was that amount of Exg activity which liberated 1 umol of pNP per min at 37°C with the assay conditions described above. pNPCase activity in whole cells was determined as described forfi-galactosidaseby Miller (1972). In the whole cell pNPCase assay, the cells from 2 to 10 ml of culture were collected by centrlfugation, washed with 50 mM KP0 pH 7.0, resuspended in 250 ul of 50 mM KP0 pH 4>  4>  7.0, and rendered permeable by the addition of 10 ul of CHC1. The cell suspension was then 3  mixed with 250 ul of 2mM pNPC and incubated at 37°C. To quantitate the extent of the hydrolysis of pNPC, the cell suspension was centrifuged briefly in 8 microcentrifuge, and the supernatant fluid was decanted. The reaction was developed by the addition of an equal volume of 2% N82CO3 to the supernatant fluid and the O.D.^o of the clarified supernatant fluid was determined. Miller  52  (1972) gives a detailed explanation of the factors that should be considered when this assay Is used for quantitative measurements.  (B) Carboxymethylcellulaseassay The carboxymethylcellulase (CMCase) assay measures reducing sugars released from CMC by reaction of the reducing groups with dlnltrosallcyllc acid (DNS; Miller, 1959). The DNS reagent was prepared by dissolving 10 g dlnltrosallcyllc acid, 2 g phenol, 0.5 g sodium sulfite and 200 g sodium potassium tartate In 500 mis of 2% N80H and then diluting this solution with 500 ml of water.The assay was as follows: 300pl of cell extract In 100 mM KP0 , pH 7.0 and 500 pi 4  of 4* CMC in 100 mM KP0 , pH 7.0 were mixed and incubated at 37°C for 30 tol 20 min. This 4  mixture was then mixed with 800 pi of DNS reagent, steamed at 100°C for 15 m1n, allowed to cool and the absorbance at 550 nm W8s measured. A standard curve was determined for each assay using a 1 mg/ml (5.56 mmol/1) stock solution of glucose.  (C)fi-gal8ctos1daseassay fi-galactosldase activity was measured according to Miller (1972) by the hydrolysis of o-nitrophenyl-fi-D-gal8ctos1de (0NP6). The details of the reaction conditions and the calculation of enzyme activity Is essentially as described for pNPCase except that the absorbance was measured at 420 nm. One nmol/ml of ONP has an O.D.^onnr  00045  -  0 n e u n 1 t o f  ^  1 S  defined  as that amount of enzyme that releases 1 pmole of ONP/min at 37"C.  (D)fi-lactamase8ssay fi-lactamase was assayed by following the release of nltrocefoic 8cld from nitrocefin (0'Call8ghan  et al., 1972). Nitrocefin was a gift from Glaxo Group Res. Ltd., Greenford, Middx.,  England. A stock solution of nitrocefin was prepared by dissolving 0.5 mg of nitrocefin In 0.5 ml of DMSO and then diluting this solution into 9.5 ml of 100 mM KP0 , pH 6.9. A working solution 4  of 0.1 mM was prepared by diluting the stock solution 10-fold in water. The assay contained 1.0  53  ml of 0.1 mM nltrocefln-10 mM KP0 , pH 6.9, and 0.1 ml of diluted cell extract. This solution 4  was Immediately mixed in a cuvette and the rate of change of absorption at O.D.482 was recorded. The results were plotted 85 AO.D.482 versus time; only values In the linear range of the plot were used for calculations. A 10" M solution of nltrocefolc acid has an O.D.482nm 4  =1  5 5  (O'Callaghan  et al., 1972). Enzyme activity units: umoles nitrocefoic acid released/min at 23°C.  (E) Assay of protein concentration Protein estimates were made with the Bio-Rad protein assay kit (Bio-Rad Laboratories, Canada) according to the manufacturer's specifications. This method, originally described by Bradford (1976), is a dye-binding assay based on the differential color change of a dye in response to various concentrations of protein. The assay was performed by mixing 160 u.1 of a dilution of the protein sample in water with 640 u.1 of water. To this solution 200 u.1 of the BioRad dye concentrate were added. The solution was then mixed and its absorbance at 595 nm was measured. Standard curves were constructed using bovine serum albumin.  54  RESULTS AND DISCUSSION  I Structure of the gene encoding the exoglucanase of Cellulomonas fimi  (A) Subcloningof thecexoene  (1) Detection of Exg activity coded by subcloned fragments During the early stages of this work, expression of the Exg in recombinant clones W8S determined with an immunological plate screening method (Whittle et al., 1982) or by assaying CMCase activity in cell extracts with the DNS method (Miller, 1959). The immunological screening method is suitable for screening large numbers of transformants, but it is not quantitative. The DNS assay for reducing sugars is a quantitative assay of Exg activity but it has technical disadvantages (Breuil 8nd Saddler, 1985) and the CMC substrate used in this 8ssay is not an optimal substrate for assaying exoglucanases (Gilkes et al., 1984b). In the latter part of these studies it was demonstrated that C. fimi Exg hydro!y2ed para-nitrophenylcellobioside (pNPC) (Gilkes et al., 1984b; Deshpande et al., 1984a) and methylumbelliferylcellobioside (MUCX W.W. Wakarchuk, personal communication; van Tilbeurgh et al., 1982). Wild type E. poll did not display detectable pNPCase or MUCase activity. The Exg cleaved the agluconic bond in pNPC and MUC, releasing para-nitrophenol or methylumbelliferone, respectively (Fig. 3). The release of para-nitrophenol W8S followed spectrophotometries!ly at 410 nm, and provided a simple and rapid method for quantitation of Exg activity in cell extracts. Exg activity in pellets of whole cells W8s also assayed quantitatively by employing procedures described for assaying B-galacosidase (Miller, 1972; see also Materials and Methods). pNPC W8S not suitable for plate screening of recombinant clones due to the insufficient contrast between the plating media and the yellow pNP, which diffused rapidly. I found that MUC was an excellent compound for use in indicator plates to detect Exg activity. Methylumbelliferone fluoresces under long-wave UV irradiation; colonies  55  expressing Exg that were grown on media (LB or M9 agar) containing 10O uM MUC were surrounded by a fluorescent halo when Illuminated with UV light (Fig. 4a). Exg activity was also detected In SDS-polyacrylamlde gels In situ using MUC. Experiments showed that the recombinant Exg retained over 75* of its original activity after exposure for 1 hour to 1 * SDS and 2* 13mercaptoethanol. The stability of Exg to these chemicals suggested that Exg could be detected In denaturing SDS-polyacrylamlde gels. Indeed, positions of Exg activity were Identified in SDSpolyacrylamlde gels after electrophoresis of a recombinant Exg preparation, by soaking the gel for 3 to 5 mlns In 50 mM Trls, pH 8, containing 250 uM MUC, and Illuminating the gel with longwavelength UV light. Two bands of MUCase activity, with apparent molecular weights of 50 and 150 kda, were present In cell extracts of E. coli PM191 (pEC-1.1) (Fig. 4B). Since the cex Insert in pEC-1.1 is large enough to encode a protein of 1^=65,000-70,000 kda, the high molecular weight activity band must be due to aggregates of Exg which are not disrupted by SDS.  (2) Localization and direction of transcription of the cex gene The CK< gene was cloned on a 6.6-kb BamHI fragment of £. flM DNA llgated into the BamHI site of pBR322, giving pEC-1 (Whittle etal., 1982; Gilkes etal., 1984a). E. coll cells containing pEC-1 synthesize and secrete Into the periplasmic space a functional Exg (Gilkes et al., 1984a; Gilkes etal., 1984b). The  of the cex gene product is 51,000 (Gilkes et al., 1984a). A  protein of this size would require a minimum of 1400-1500 bp of coding sequence. The Exg coding region was localized on the 6.6-kb BamHI fragment 8S follows. Deletion mutants of pEC-1 were obtained In vitro bv partial digestion of purified pEC-1 DNA with Sal'. ligation under dilute conditions, and transformation of E. coli C600 (Fig. 5 and 6). Potential Exg-producing clones were identified Immunologically (Whittle et al., 1982). Extracts of positive clones were assayed for CMCase and pNPCase activity and their plasmid DNAs were analyzed by restriction mapping. Strain C600(pEC1.1) produced as much pNPCase activity as strain C600(pEC-1) (Fig. 6). The plasmid pEC-1.1 contained a 2.0-kb BamHI-Sall fragment and two smaller Sail fragments of 0.36  56  Figure 3. Synthetic substrates used in assaying Exg activity.  CH.  CH.OH  L^_  OH  p-nitrophenyl- B -cellobioside (PNPC)  w»  0  0 , 1  [Vv '  o  OH  4-methylumbelliferyl-/?-D-cellobioside  (MUC)  57  Figure 4. Functional detection of Exg on indicator plates (A) and SDS-polyacrylamlde gels (B). (A) £. coJicells containing the plasmld pEC-1.1 were grown overnight on LB-agar plates supplemented with 75 ug ampicillin/ml and 10O pM MUC. The fluorescent haloes surrounding the colonies were visualized with long-wavelength UY light. (B) Crude cell extracts of stationary phase cultures of £. coJi PM191 (pEC-1) were prepared by washing cell pellets in 50 mM TrisOH, pH 7.6, resuspending the cells In the same buffer, disrupting the cells by sonlcatlon and clarifying the extract by centrifugation at 20,000 x g for 30 minutes at 4*C. For electrophoresis, a 7.5* polyacrylamlde separating gel, a 3* stacking gel and the buffer system of Laemmll (1970), all containing 0.1 * SDS, were used. Aliquots of the cell extract were mixed with an equal volume of loading dye mixture (62.5 mM Tris, pH 6.8,2* SDS, 10* glycerol, 5* J3mercaptoethanol and 0.02* bromphenol blue), incubated at 37*C for 30 minutes and electrophoresed at a constant voltage of 120 for 4 hours at 23*C. To prepare the gel for detection of enzyme activity, it was removed from the glass plates, rinsed briefly with 50 volumes of 50 mM Tris, pH 8.0 and then Incubated with gentle shaking In a solution of 50 mM Tris, pH 8.0, containing 250 pM MUC for 5 minutes 8t 23*C. The positions of Exg activity were visualized by the appearance of fluorescent bands under UY light (354 nm wavelength). The M,. of the Exg was determined by marking the position of the fluorescent band of MUCase activity on the gel with pinholes and then staining the gel with Coommassie Brilliant Blue. Lanes 1,2 and 3 contain 4,2 and 1 milliunits of pNPCase activity, respectively. Molecular weights correspond to protein standards rabbit muscle myosin (205,000), E. colifi-oelactosidase(116,000), rabbit muscle phosphorylase b (97,400), bovine serum albumin (66,000), ovalbumin (45,000) and carbonic anhydrase (29,000).  •  58  and 0.24-kb, which were arranged In the same order and orientation as In pEC-1 (Fig. 5 and 6). Removal of the two smaller M l fragments gave pEC-1.15. Since pEC-1, pEC-1.1 and pEC-1.1S gave comparable levels of pNPCase activity In C600, It seemed likely that the 2.0-kb BamHI-Sall Insert of pEC-1.1S contained the complete cex gene coding sequences. Subsequent DNA sequence determinations revealed that the cex gene extended 228-bp beyond the Sail site of the 2.0-kb BamHI-Sall fragment. The BamHI site of pBR3221s located within the tetracycline gene coding region, 320 bp 3' to the tet promoter translational Initiation start site (Backman and Beyer, 1983; Stuber and Bujard, 1981). Insertion of cellulase gene fragments lacking translational termination signals Into the BamHI site of pBR322 would put the Inserted DNA under the control of tet transcription and translation signals. Therefore, It was necessary to localize the signals controlling expression of the cjx gene. The tet promoter can be disrupted by modification of the Hindi 11 site of pBR322 (Boyer et al., 1977). The tet promoter was inactivated by digestion of pEC-1.1 with Eo)RI and Hindi 11, removal of 5" overhanging ends with nuclease S1, and circle reclosure. A recombinant plasmld, designated pEC-1.11, which had lost both the EcoRI and Hindi 11 sites, no longer directed Exg synthesis (Fig. 6); this suggested that pjsx.gene expression In pEC-1.1 was dependent on transcription initiated at the let promoter, and that the direction of transcription was from BamHI to Sail on pEC-1.1. This was confirmed by Inserting the 2.6-kb BamHI-Sall fragment of pEC-1.1 In both orientations downstream from the lacZpo region in the plasmids pUC-12 and pUC-13 (Fig. 5 and 6). In addition to the origin of replication and the B-lactamase gene of pBR322 these plasmids also contain the promoter/operator, the RBS sequence and the amino terminus of the J36al gene from the E. coli lactose operon (Messing, 1983; Vlelra and Messing, 1982). A multiple cloning site is positioned at the fifth and sixth amino terminal codons. pNPCase activity was detected only 1n E. coM JM83 containing pUC 12 with the 2.6-kb C. fimi Insert positioned such that the Sail site was distal to the lacZpo (Fig. 6). This plasmid W8S designated pUC 12-1.1 cex. Furthermore, after transfer of pUC 12-1.1 cex to JM101, an E. coli strain that overproduces the lacl repressor (Messing, 1983), Exg activity increased 40 fold after induction with IPT6, which  59  Figure 5. Subcloning and direction of transcription of the cex.gene. The cex gene is contained on a 6.6 kb BamHI fragment (boxed areas) inserted into the BamHI site of pBR322 to yield pEC-1 (Whittle et al., 1982). Digestion of the BamHI insert by Sail produced 7 fragments of 2.0,1.3, 0.92,0.8,0.72,0.34, and 0.26 kb. These fragments were ordered by the method of Smith and Blrnstiel (1976). The plasmid pEC-1 was first restricted with EpoRI and Hindi il- The Hind' 11 site was then end-labelled with [<x-2p]-(jGTP in the presence of dATP 8nd the Klenow fragment of 3  DNA polymerase I. The 10.9 kb [tx- Pl-H1ndlll-EcoRI fragment was then partially digested with Sail and the products analyzed by agarose gel electrophoresis. The exoglucanase gene was then localized by the in vitro deletion of Sell restriction fragments of pEC-1. Thus, pEC-1 DNA was partially digested with Sail and religated under dilute conditions. £. BiliC600 was transformed with the ligation mix and transformants were screened immunologically (Whittle et al., 1982). The transformant pEC-1.1 contained a 2.0 kbfiamHI-Sall,a 0.37 kb Sail, and a 0.26 kb Sail fragment arranged in the same order and orientation as in pEC-1. pEC-1.1S was obtained by complete digestion of pEC-1.1 with Sail and rellgation under dilute conditions. To construct pUCI 2-1. Icex and pUC 13-1.1 cex^the 2.6 kb BamHI-Sail fragment was removed from pEC-1.1 by complete BajjiHI and partial M l digestion, followed by Isolation of the fragment by electrophoresis in a low melting temperature agarose gel. The purified DNA was then ligated into pUCI 2 and pUCI 3 which had been previously cut with BamHI and Sail - This resulted in the two plasmids, pUCI 2-1. icex and pUC 13-1. Ipex., in which the cex insert is positioned in opposite directions downstream from the lacloo. Only the EJDJJRI , BamHI, Sail, and Hindi 11 sites of the multiple cloning sites of pUCI 2 and pUCI 3 are indicated. pEC-1.1 tap. was obtained by substituting the promoter and the first 289 bp of the tetracycline coding region with the iat promoter fragment from the plasmid pDR540 (Russell and Bennett, 1982). pEC-1.1 was restricted with EjaRI andfcamHIand the large fragment was Isolated. The iac. promoter was gel-purified as an EcoRI-BamHI fragment from pDR540. Ligation of the tac promoter fragment into pEC-1.1 and transformation ofLfflliJMIOI yielded pEC-1. l i s t P t e t . tetracycline promoter; Ptac- lac promoter. Restriction sites: B. BamHI; E.ECQRI ; H3, Hindlll: S, Sail. 52  59a  60  Figure 6. Localization of the pex. gene by in vitro deletion analysis. Restriction sites: BamHI (B); Pstl (P); Sail (S). Only one Pstl site which was used to construct a deletion derivative Is shown in parentheses. Plasmid pEC-1 contains a 6.6-kb C. fimi insert carrying the cex gene. The portions of this fragment that were subcloned Into pBR322 or pUCI 2 and 13 are aligned with respect to their original position 8nd orientation in the parental plasmid pEC-1. The heavy arrows show the direction of transcription of the plasmld promoters for M and M- The Exg activity 1n whole cell extracts containing the various plasmids is given.  pNPCase Promoter Plasmld  pEC-l  Vector  pBR322  Promoter  orientation  kb  4  _ L _  Jet  (P)  pEC-li  pBR322  pEC-l.l  pBR322  pEC-I.IS  pBR322  pUCI2-lcex p U C I 2  -i S  1—r~ S  total s p e c i f i c _|  (nmoles  1,43  3  (+/-)  islOnactiv.)  2.2 Jej.  I.I 5.7  Jac.  pUCI2-Ucex  pUCI2  Joe  pUCI3-|Jce*  pUCI2  IQC  PUCI2-I.IS  PUCI2  lac  pUCI2-Pst (2.$)  pUCI2  lac  pUCI2-P*f  pUCIZ  Joe.  8.0  Ill  activity  pNP/min/nn  0.07 1.5  14.0  5.0  protein)  61  Is an Inducer of the lac promoter. These results showed that the polarity of transcription of the cex gene on the BamHI-Sall Insert of pEC-1.1 and pUC 12-1.1 cex was from the BamHI site to the Sail site. Further deletion analysis of the 2.6-kb BamHI-Sall fragment In plasmld pUC 12-1. Icex mapped the location of the pex gene more precisely (F1g.6). The removal of a 0.64-kb BamHI-Pstl fragment from the cex Insert In pUC 12-1.1 pex, yielded plasmid pUC 12-Pst( 2.6). The deletion of two Sall-MI fragments (0.36 and 0.24-kb) from the Sail end of the pex Insert in pUC 12Pst(2.6), gave plasmld pUCI 2-Pst. Both plasmids pUCI 2-Pst(2.6) and pUCI 2-Pst directed Exg expression. Interestingly, plasmld pUCI 2-Pst, which contained only 1396-bDofC. fimi DNA. yielded higher levels of expression than the plasmids pUC 12-1 cex.and pUCI 2-1.1S. DNA sequence determinations revealed that the proteins encoded by plasmids pUCI 2-Pst(2.6) and pUCI 2-Pst were fusion proteins between Exg and the amlno-terminal portion of B-galactosldase in pUCI 2.  (B) DNA sequence of the pex gene (1) DNA sequence determination The strategy for the determination of the nucleic acid sequence of the 2.6 kb BamHI-Sall Insert In pEC-1.1 is outlined In figure 7. Restriction mapping (Smith and Blrnstlel, 1976) Identified a large number of cleavage sites, most of which were used alone or in combination to obtain fragments for subclonlng and for radioactive end-labelling and chemical sequencing (Maxam and Gilbert, 1980). Chemical sequencing did not yield the entire sequence of both strands of the gene. The missing sequences were determined, and most of the chemically determined sequences were confirmed, by the dideoxy method (Sanger et al., 1977) after cloning specific restriction fragemnts into the M13 vectors mp 10 and mp 11 and the plasmld vector pUC-13 (Messing, 1983). Sequence data from both procedures were necessary to resolve areas of compression and ambiguity, possibly due to the high (71 %) G + C content of the C. fjmi chromosomal DNA.  62  Figure 7. Restriction map and sequencing strategy of the cex gene and Its flanking regions. (A) The location of the exoglucanase coding region. The alanine codon (GCG) is the first amino acid of the mature protein. The deduced translational Initiation (ATG) and termination codons (TGA) are shown. (B) A partial restriction map of the 2.6 kb BamHI-Sall cex gene fragment. The map W8S constructed by the method of Smith and Blrnstiel (1976). pUCI 2-1.1 cex plasmid DNA was digested with BamHI, end-labelled with [cx- P]-dGTP by the Klenow fragment of DNA polymerase I, and then digested with EcoRI. The labelled DNA fragment was then partially digested with the enzymes shown, and the products were analyzed by agarose gel electrophoresis. Only the Avail and EcoRI I sites which were used for sequencing are indicated in parentheses. (C) The sequencing strategy for the 2.6 kbfiamHI-Sail QSSL gene fragment from pEC-1.1. The arrows indicate the origin, direction, and number of nucleotides determined in each sequencing experiment. For Maxam and Gilbert sequencing all fragments were 3'-end-label led. Hence the arrows are in the 3' to 5' direction. Restriction sites: A, Ayall; B, BamHI; Ec, EcoRI; H, Hinfl; He, ttLD£ll;Ps,£sil;Pu £yu.ll;S,Sjll; Sm.Smal:So.Sohl:St.Stul:S3.Sau3al:X.Xhol. 32  J  ZOaa  o o MATURE CEX  (443oo)  B  ,  1  B' Sp  S3  1  1  X  S3  I  MAX AM  a  i  i  i  ^  1  H  FV(A) H  i  1  i r  1  Pv  S3 (Ec)  1  1  'I  'U  1  — n — — I — | 'k  Ps St St Sp S p S m  1  He) H  1  1  1  1  .  S  S3  Sm  'S3 S H  I  S3 HS  N  *•  »• i  H -«  GILBERT  i  -y-  1  -»  H  1  <  r-  H  1 •*  I  >  •«  I—»-|  1  »  <—  1  rI  >  I  -*•  SANGER'S  I  >  I  h  1  1  1  1  >  1_  500  1000  1500  2000  2500 (bp)  63  (2) Nucleic acid sequence The final sequence of 2585 nucleotides containing the entire cex gene is given in Fig. 8. The first nucleotide of the BamHI site has been labelled arbitrarily as nucleotide one. The actual coding sequence of the Exg was located by comparison of the amino acid sequences predicted by the nucleotide sequence with the amino acid sequence of the Exg purified from C. fimi (ML. Langsford, unpublished results). The sequence of the first thirty amino acids at the amino-terminus of the enzyme agreed exactly with the DNA sequence starting at nucleotide 861 in phase 3. The coding region extended for 1329 nucleotides to an in-frame stop codon (T6A) at nucleotide 2190. The deduced amino acid sequence for the Exg product Is shown above the nucleic acid sequence. The amino acid composition predicted by this sequence matched that obtained for the Exg purified from C_. f M (Ml. Langsford, unpublished results). The high G+C content of the C. fimi DNA (71*) predicted large excesses of arglnine and proline residues in the non-coding reading frames 1 and 2. This feature was useful In locating frameshlft errors In the coding region (Bibb et al., 1984). The location of other ORF's in the 5' and 3' flanking DNA sequences of the cex. gene Is discussed below.  (C) Structure of the cex. gene and Its product  Computer analysis of the DNA sequence of the cex.gene coding region and its 5* and 3" flanking regions revealed putative transcriptional and translational gene expression elements, a putative signal peptide, a tandemly repeated 7 amino acid peptide in the carboxyl terminus of the Exg and a highly biased codon usage. These features are summarized schematically in Fig. 9 and discussed in detail below.  (1) Identification of a signal sequence, initiating codon and ribosome binding site The cex gene product is normally secreted as a soluble, extracellular protein (Langsford  12°  9 3 b1=3 CO ^< CT> c o  c3 =3 — c -J 3 2 . cr3<g  •  20  40  60  80  100  120  GATGTCCGCCTGGGCGGCGCGGAGCTTGTCCGTGAGCAGGCGCAACGCGGTCAGTTCCTCGGCGTCGAGGGTTCCGCCGACGAGCCGGCGGATCGAGCGGACGTGCCGGCGCCCGATGTC 140  160  180  200  220  240  GCGTTGGACACGTTGCCCCTCGGGCGTGAGGGTCACCGCGACGCCCCGCGCGTCGCCCACGACGGGACCGCGCGAGACCAGGCCGGCGGCCTCGAGGCGCTCTACCATGCGCGACAGGCT 260  280  300  320  340  360  GGGCTGCGTCAGGAGGCTGCCCTCGCCGAGCTCCTTGATCCGCGCGGTCAGGCCGGGGCAGCGGGACAGCGTGAACAGCACGTCGTACTCGCGGATGGTCAGCGGGTCCCACACGTCGTC 380  400  420  440  460  520  540  560  580  600  TCCCCCTCGGTTGGCTGTCGAAACAGCGACGGTATCGAAACTGCAGGCCAGGGCGGGCCGAAATGATTCAGCACCTCCCGCGGACGGGCCCCCACGTCACAGGGTGCACCCGGCACTGGC 620  640  660  -41  680  700  -30  M  P  R  T  T  P  A  P  G  H  P  720  -20  A  R  G  A  R  T  A  L  R  T  T  _|0 R  R  R  A  A  T  L  V  V  G  A  T  TCGACG^GGAGGAJCATCATGCCTAGGACCACGCCCGCACCCGGCCACCCGGCCCGCGGCGCCCGCACCGCTCTGCGCACGACGCGCCGCCGCGCGGCGACGCTCGTCGTCGGCGCCACGG 740 V  V  L  P  A  0  -I  A  760  +l  A  T  T  L  K  E  780  A  A  D  G  A  G  R  800  D  F  G  F  A  L  D  820 P  N  R  L  S  E  a  s $  zr •  >  3 8  —8 =3 O O  840  A  Q  Y  K  A  I  Q . O  —  §  860 S  E  F  N  L  V  880 V  A  E  N  A  M  K  900 W  D  A  T  E  P  920  S  O  N  S  F  S  F  9*40  G  A  G  D  R  V  G  K  E  L  Y  G  H  1000 T  L  V  W  H  S  Q  1020  A  L  P  D  W  A  K  1040  N  L  N  G  S  A  F  S  A  M  V  N  S  Y  A  A  H  F  E  G  K  V  A  1120 S  W  D  V  V  N  E  1140  V  T  K  V  A  A  F  A  D  G  D  G  P  P  Q  D  S  A  D  1180 F  Q  O  K  L  G  1200  N  G  Y  I  E  T  A  ACTTCGAGGGCAAGGTCGCGTCGTGGGACGTCGTCAACGAGGCGTTCGCCGACGGCGACGGCCCGCCGCAGGACTCGGCGTTCCAGCAGAAGCTCGGCAACGGCTACATCGAGACCGCGT 1220 F  R  A  A  R  A  A  1240 D  P  T  A  K  L  C  1260 I  N  D  Y  N  V  E  1280 G  I  N  A  K  S  N  1300 S  L  Y  D  L  V  1320 K  D  F  K  A  R  TCCGGGCGGCACGTGCGGCGGACCCGACCGCCAAGCTGTGCATCAACGACTACAACGTCGAGGGCATCAACGCGAAGAGCAACTCGCTCTACGACCTCGTCAAGGACTTCAAGGCGCGCG 1340 G  V  P  L  D  C  V  1360 G  F  Q  S  H  L  I  1380 V  G  O  V  P  G  D  1400 F  R  Q  N  L  Q  R  1420 F  A  D  L  G  V  1440 D  V  R  I  T  E  GCGTCCCGCTCGACTGCGTCGGGTTCCAGTCGCACCTCATCGTCGGCCAGGTGCCGGGCGACTTCCGGCAGAACCTGCAGCGGTTCGCGGACCTGGGCGTGGACGTGCGCATCACCGAGC 1460  1480  1500  1520  1540'  &  5  1560  S  =3 "  3  _L  I  5. co  ®a 55. cr =r CDo g : CD _  2 g3  ^ I a - J  £Z O CD r~ t u 3Z3 CO  S  | O  S  0  Q.  &  2^"°  3  T 1080  H  1160  _ CD  &«>  960  D  GCAAGGAGCTGTACGGCCACACGCTCGTCTGGCACTCGCAGCTGCCCGACTGGGCGAAGAACCTCAACGGCTCCGCGTTCGAGAGCGCGATGGTCAACCACGTGACGAAGGTCGCCGACC 1100  -> CD  A  1060  E  CD  § =  § if 8<a C  ACAGCGAGTTCAACCTCGTCGTCGCCGAGAACGCGATGAAGTGGGACGCCACCGAGCCCTCGCAGAACAGCTTCTCCTTCGGCGCGGGCGACCGCGTCGCGAGCTACGCCGCCGACACCG 980  C D  ZJCO  TCGTGCTGCCCGCCCAGGCCGCGACCACGCTCAAGGAGGCCGCCGACGGCGCCGGCCGGGACTTCGGCTTCGCGCTCGACCCCAACCGGCTCTCGGAGGCGCAGTACAAGGCGATCGCCG D  Z  CD ,  CD  CD  o  480  GCGCTGGAACCGTCGCATGAGCGCGACCTTGGCACGAACAGCGACTCCCACGCCTCGGCAGCCTCGCGCGTGGTGGCCTCGCGGGTGCCCGTGGCGCCCGCCGTGCCCTCGACCATCGGC 500  o>  —*  CD c-r CO C D  o GGATCCGCATGCCCGCTCCTCCCTGTCGCCGCCCGCCCCTCGGACGACTTGTCCATCGTGTCCCGTGGGAGGCTCCGACGGATGGGCCCGAGGTCCCGCGGCCGTTCAGGAGGGGTCGGG  CD CO  =r  —' fi.  ~*  ^  m  - J  CD  2.Q  o-  CD  '  . CD W  =J  2.  C T CD  3.  -*>  CO  3  CD  ^. => x  2 . S, C D c 2 c± 8 2,<o  <S » g §.<3.<Q c  S2.  -,  ar c  CD  CD  f°  9- z5  —1  65  Figures (continued) 861 (Langsford, H.L, unpublished results); the corresponding amino acids are underscored. The amino acid numbering starts with an Ala at nucleotide 861 as +1. The putative signal sequence begins at an ATG at nucleotide 738. This is preceded by a possible ribosome binding site (boxed sequence). A possible -35 and -10 promoter region is doubly overlined. In the 3" untranslated region sequences capable of forming stem-and-loop structures are indicated by arrows. cj  o  C J CO  313»  •— *~  CJ CJ  U.  cj cj •- u  <  <  > CJ  o  cj hcj  l-  a  CJ  u u  I— tCJ  u  > o CD ID  ^  CJ  < <  <  cj  CJ  <  s >-  CJ O C J CO  cj  CJ  C5  u <  o > >  ¥  *  >  cj cj K  O  cj < cj  o  t-  u  o  cj <  cj  <  CJ <  u cj u  o CD  -  <  <  cj  -J  cj l -  ¥  <1  o  l/>  o  o UJ *-  cj  CJ (J  cj < CJ  cj  CJ  t-  a U.  >  cj  >  cj  o  _) t -  -  cj  1-  o  CJ  CJ < <  Q  1-  u  <  CJ CJ  O  <  <J <  CJ  cj CJ  -  <  cn  o  CJ  f-  1-  CJ CJ  •-  o.  i-  1-  <  o o  O  cj CJ  g  O  a  a. 1a  a  l/l U.  CJ O C J CD C J CO < — CJ  o  CJ CJ < CJ  a o CJ "V C J CD CJ — < CJ  o  CJ  u . •— h-  o  h-  <  o  CO 01  t-  a  u  < CJ CJ  CJ  h-  2  O  > K  >  2 <  CJ  CJ  CJ  CJ  o  CJ CJ  CJ  < o  CJ CJ  < CJ  ° CJ  14 T c j  CJ CJ  ik C J  CJ CJ  CJ  CJ  CJ CJ o CJ 00 cj O 2 < CN  <  io  CJ ' CJ  ' CJ Ak o  a  u  CJ  CJ CJ < CJ  <  C-l  CJ CJ CJ  <  2  t-  CJ  u  u  <  o K CJ  u < <  CJ CJ CJ CJ  CJ  <  CJ  < < o a cj  < CJ < CJ  CJ  o CD  CJ  >  cj  O t - CD CJ o CJ CN  \~ u < o > <  in CJ  tCJ CJ  CJ CJ CJ CJ  tCJ  CJ CJ  CJ l / l CJ  CJ CJ  CJ  CJ  CJ CJ  CJ  < CJ  <  CJ CJ  <  o  ,  a  CJ  2 < <  CJ  '  'cj < t-  <  H  CJ  cn c j  CJ  CJ CJ  CJ t-  u  u  o O 11 i - in CJ CN CJ CJ CJ H  CJ  CJ CJ  1— CJ CJ  CJ O < CN CJ 1 CJ CN CJ  CJ  o  CJ  o  CJ CD cj i n C J CN Y— CJ  CJ CJ CJ  CJ t~ CJ  CJ  CJ  cj  CJ  CJ  <  CJ  o  CJ CJ CJ  CJ CJ  I- o < CO CJ CN CJ  <  O cj i n t- in < CN CJ  CJ  o  CJ  T CJ *T O CN CJ CJ  O CJ GO ^~ 0 CN CJ CJ  1-  u  CJ  <  CJ  hhCJ CJ CJ  CJ <J <3 CJ u  o t~ o < o  o o  < t- i n  CJ  CJ  CJ  <  CJ  CJ  CJ CJ CJ  CJ  CJ  <  CJ  u  CJ CN  CJ  cj  O CO cn CJ CN CJ CJ  CJ CJ  CN r> CN  J_ CJ  <  CJ CJ  cj o  CJ  u CJ CJ CJ CJ CJ  CJ CJ  < CN CN  CJ  1—  <  CJ  <  tCJ o CJ CD CJ T C J CN CJ  <  •u CJ  o o  CJ  < <  CJ  <  - J  CJ  CJ  O C J *T t- r j CJ CN  CJ  a  CJ  < cj  ' C J  CJ  CJ  (-  CJ CJ  CJ CJ  u  CJ CJ  <  CJ  O  00 TJ CN  •— *— tt— CJ  CJ  CN CN CN  CJ CJ  CJ  <3 <3 CJ  o  u < < cj  CJ  <  CJ CJ  CJ  D  >-  <  CJ CJ  (-  CN  <  CJ I-  1o  CD  CJ  CJ  3  u  CJ  u  *CJ  u  o  CJ  CJ - C J  CJ  CJ  <  CJ CJ  CJ  cn c j  CJ  CJ  u  CJ  CJ CJ CJ u  CJ  o  CJ  CJ  CJ  <  CD  K CJ  CJ  <  <  CJ  CJ D  CJ  O  CN  CJ  CJ  CJ CJ  CJ  CJ  CJ  CJ  o  cj i n  < <  CJ CJ  CJ CJ CJ  CJ  <  o  t-  CJ CJ  O O CJ CJ CJ CN  CJ  a. u  CN  CJ CJ  CJ CJ  in  K  o  CJ  CJ  CJ  t - *T CJ CN CJ CN  CN  ~  o  CJ  CJ  cj o  CJ CJ CJ  CJ < CJ  CJ  <  <  CJ  1—  CJ  u  2  CJ CJ  CJ t-  CJ CJ CJ  u  CJ  ro  t—  cj O  CJ CJ  <z CJ  <  CJ  CJ CJ  1- C J  <  o o CN  cj  in cj <x  o  CJ  C J CD C J CO f - CN  CJ  CJ  CJ  CJ CJ CJ  K  c j CN C J CO CJ -  C J  CJ CJ CJ  <  >- o  O  CJ  t  CJ  CJ  CJ  1  u  CJ CJ u CJ CJ  CD CJ CN C J CN  CJ < CJ  <  CJ CJ  CJ 1-  ^  cj C J CN <  in cj  CJ  <  < O  <I  CJ  CJ CJ  t-  <  tCJ  CJ CJ  CJ CJ  CJ  (-  CN  CJ  CJ  CJ CJ  CJ  <  3  u  CJ < CJ  2  I  1— <u  1/1  CJ CJ <  CJ  <  CJ CJ  U  o  CJ CJ  CJ CJ CJ  < cn  <  CJ  CJ CJ  CJ O CJ CN CJ •— CJ  »—  CJ  CJ  1  CN  CJ  CJ  CJ  CJ  O  CJ  CJ CJ  j L  <  <  u  u u  o u  CJ  CJ CJ  O  CJ CJ CJ  CN  CJ  <  > 1-  <T  CJ  i-  CJ  CJ  CJ CJ  CJ  Q  CJ  1—  CJ  CJ CJ CJ CJ t- CJ  o  <  - J  CJ  <  CJ  cj  <  CN o CN  <  cn c j f-  CJ  o  t- CJ  o  <  CJ  u O CJ CO  h- CN  CN  CJ  CJ CJ  o  u.  CJ  CJ CJ CJ CJ  CJ CJ CJ CJ  CJ CJ  CJ <  CJ  o  CD  1- ^  a cj  > t-  CJ CJ  CJ  a.  CJ  CJ  CJ CJ >- C J < CJ  CJ CJ  *  CJ CJ CJ  3  0. C J  CJ CJ  <  cj  o o  C J CJ CJ 1- C J < CJ  c j rh - *~  u  t-  a  CJ CN  CJ tcj cj  CJ  iu-  CJ  CJ <  in  1- o  a  CJ CJ CJ CJ CJ  a  o.  CJ CJ CJ  > 1-  a  < LO CJ —  < cj < u  K cj  o 1  CJ CJ  cn  o  CJ CJ  o. c j  CJ O O CO  CJ  CJ  t~  CJ CJ  H  o  CJ  CJ CJ  <  t-  CN  u t-  >  CJ CJ < CJ  < CJ  o o  CJ  CJ 1- C J < CJ  a <  3  u  IA  <  UJ  a  <  CJ  CJ <  u u  I >-  CJ  CJ (~ C J <  cn c j  CJ C J C J CD C J CO CJ — < CJ CJ CJ  a  cj o  CJ  CJ  u  -~  < CJ CJ CJ  CJ CJ  CJ  < <  O CP  O t-  CJ  CJ  < JJ C < < CJ  u O  <  u  <  CJ CJ  h-  <  —1  <  cj  t-  cj <  > l~  CJ  CJ  o  t-  CJ  CJ CJ CJ C J  •a c j  CJ  CJ CJ  o  <  CJ  cj «a c j cj cj  o  cj  CJ < K CJ  < CJ  > 1-  u  CJ  >  3  <  cj  <  cj  o  <  <  _1  CJ  <  o  CJ  cj  <  <  I-  CJ  3  cn C J  CJ  I-  o  -  ^  >  CJ  Ni  CJ CJ C J CD  r-  CJ <  ¥  <  2  CJ CJ  a  cj cj  o  h-  •  > •-  3  < CJ  <  cj  i-  C J CN CJ 0) CJ CJ 1- C J < CJ  CJ  z <  a  UJ < cj cj  1- c j  o  -  CJ  CJ CJ CJ CJ  < cj cj cj c j  O  CJ  CJ  > 1-  o  CJ  c j c j co  CJ CJ CJ CJ CJ CJ  CJ CJ  <  o CJ  <  u  <  <  CJ  cj  O C J CO cj i n t - CN CJ CJ tCJ CJ u  66  et al., 1984). Secreted proteins are usually synthesized as precursors with amino-terminal extensions or signal peptides, which are removed by specific enzymes (Blobel and Dobbersteln, 1975; Davis and Tai, 1980). Presumably, the codon at nucleotide 861 corresponded to the first amino acid of the Exg after processing and cleavage of such a signal sequence. Assignment of the start codon assumed that both the RBS and the signal sequence of the cex gene were similar to those described for other Gram-positive becterla (Watson, 1984). The open reading frame extended 360 bp from the 5" side of the codon at nucleotide 861 to nucleotide 501. Between nucleotides 501 and 861 there were two potential Initiator codons In phase with the structural gene: the ATO codons at nucleotides 663 and 738. The AT6 codon at nucleotide 738 had an upstream stretch of nucleotides (AG0A60A) which showed homology to other RBS sequences (Gold et al., 1981; McLaughlin et al., 1981). This sequence was complementary to the CCUCC near the 3' end of the 16S rRNA found In many bacteria (quoted in Murray and Rablnowitz, 1982). The nucleotide sequence 5' to the ATG at position 663 did not show homology with known RBS consensus sequences. Translation Initiating at nucleotide 738 would yield a polypeptide product with a 41 amino acid signal sequence. The putative 41 amino acid signal sequence was composed of a positively charged hydrophlllc amino-terminus, containing 7 arginines, followed by a hydrophobic sequence of 16 amino acids. The carboxyl terminal residue of the signal sequence and the amino terminal residue of the mature protein were alanines. A cleavage site for processing was between these two alanines. Multiple proteolytic cleavage sites have been proposed for the signal sequences of the penicillinase of B. llcheniformis (Nielsen and Lampen, 1982), and the alpha-amylase of B. subtilis (Ohmura et al., 1984). Therefore, there may be a secondary cleavage site. A more detailed description and functional analysis of the signal sequence is contained in the chapter dealing with the secretion of the Exg by E. coli- The predicted molecular weights of the mature (49 kDal) and precursor forms (53 kDal) of the Exg both agree reasonably well with an MY of 51,000 determined in E. coli maxicells (Gilkes etal, 1984a). The M^. for the native C. fimi Exg is 56,000 (Langsford etal., 1984). The discrepancy may be due to glycosylation of the native Exg (Langsford et al., 1984).  CD  kb  promoter "-35"  "-10"  0 i_  exoglucanase  RBS GGAGG  t ATG  1.0  0.5  t  _i  structural  o  tf> c 3 3  gene  CD  i  j alaala  3  a  1.5  (pro-thr-pro-thr- pro- thr-thr)-  3  CO  TGA  B  s  signal peptidase  =3 CD  cleavage site  leader  a -J O  sequence  (41 amino acids)  CD =3  o mature (443  CD*  exoglucanase  o  amino acids)  o. c  CD  8  ir rr CD CD X  J  68 (2) Transcriptional control elements (a) Promoter Scanning of the nucleotide sequence by computer revealed sequences at nucleotides 611616 (TGGCT) and 613-638 (TATCGA) which shared homology with the prokaryotlc "-35" (TT6ACA) and "-10" (TATAAT) consensus promoter sequences (Moran et al., 1982; Rosenberg and Court, 1979). The spacing between these two regions was 16 bp. There were two other sequences starting at positions 395 (TTGATC) and 511 (TTGGCA) which shared homology to the "-35" consensus sequence. However, there were no regions appropriately spaced to the 3' side which resemble"-10" sites. Oreenberg et al. (1986) have Identified a cluster of four transcriptional initiation sites for thecej< gene by Sj nuclease protection experiments located 25 to 29 nucleotides 5' to the ATG codon at position 738-740. The only other putative start codon lies upstream of these transcriptional start sites.  (b) Terminator The sequence after the translation stop signal contained 4 possible hairpin structures, including two overlapping regions of hyphenated dyad symmetry located between nucleotides 2225-2274 and 2259-2301. The inverted repeats are Indicated with arrows in Fig. 8, and the possible secondary structure the largest hairpins could form is depicted in Fig. 10. The first inverted repeat was 11 nucleotides long (2191- 2201) and was positioned precisely at the 3" end of the Exg coding region. It contained two bases of the stop codon. The second possible hairpin structure was 15-bp In length and was positioned 3-bp 3' to the first inverted repeat. Six nucleotides downstream from the second inverted repeat there was a 76-bp region capable of forming two larger hairpin structures which overlap (Fig. 10). The first hairpin (2226-2274) had a 22-bp stem with a 5 base loopout. Eighteen of the 22 stem base-pairs were 0-C. The second putative stem-loop (2259-2301) had a 14-bp stem, 10 of which were 0-C bp. Fourteen bases which formed the 3' leg of the first hairpin could form the 5' leg of the second hairpin. Since transcription stops at nucleotide 2272 of the DNA sequence (Oreenberg et al., 1986), the region  69  must contain a transcriptional termination signal.  (3) Codon usage There was a 98.5* bias towards the use of 6 or C in the third position of those codons used In the Exg gene (Table VII). TheG+C content of the first and second codon positions for the cex gene were 61.4* and 53.3*, respectively. A similar asymmetrical distribution of GC composition (reviewed by Bibb et al., 1984) occurs within the codons of the apjh gene of Streptomvces fradlae (Thompson and Gray, 1983). the leuB oene of Thermus thermoohllus (Kaoawa et al.. 1984), and the endo H gene of Streptomvces Dlicatus (Robblns et al., 1984). In contrast, this GC distribution is not apparent within the codons of E. coli genes (Table VII; Varenne et al., 1984). The codon usage for the preproteln of Exg is shown in Table VIII. The codon usage of the cex gene showed a nonrandom choice of codons. Twenty-six of the possible 61 codons were not used in the cex gene, and 4 codons were used only once. Codon usage In the cjx gene appeared to be quite different from that in E. coli (Table IX) The cex gene employed only 35 codons, whereas E. coli codon usage (Varenne et al., 1984) shows a more general distribution. It Is also evident that the codon usage of E. coli oenes is consistent with intracellular tRNA concentrations In E. coli (Table IX; Ikemura, 1981; Varenne et al., 1984). In contrast, the cex gene employed a number of codons which are used rarely In £. coli genes and which correspond to minor tRNA species. For example, the cex gene had 6 CGG codons and 1 AGO codon for Arg, which are rare codons in E. coli.  (4) A tandemly repeating heptapeptide in the carboxyl terminus of the Exg A novel feature of the deduced amino acid sequence was a peptide of seven amino acids near the carboxyl terminus (Fig. 8 and 9). Nucleotides 1807 to 1869 encoded the amino acid sequence (Pro-Thr-Pro-Thr-Pro-Thr-Thr)3, with the single substitution of a serine for the final threonine in the third repeal The engA gene, which encodes an endoglucanase from C. fjmi. also contains a tandemly repeated Pro-Thr-Pro-Thr-Pro-Thr-Thr sequence, that has been termed the  70  Figure 10. Hypothetical secondary structure which can form in the 3' nontranslated region following the exoglucanase coding region. The numbers and arrows refer to the distance from the first nucleotide of the 6-bp BamHI site of the nucleotide sequence shown in Fig. 6. Dots indicate non-optimal Watson-Crick bp.  C G 2245 -*» C G C U G G U G G C C 22 3 5 G G G C C A C G 5-CCG G  2225  C C G C G « -2 25 5 C G G U A CCG-3 2300 C G "* A U C« A C G G C G C C G C G C G G U G C««•2290 I C G G C . U C A G C C G U U G -2 2 8 0 U  STRUCTURE OF THE GENE ENCODING THE EXOGLUCANASE OF CELLULOMONAS FIMI AND ITS EXPRESSION AND SECRETION IN ESCHERICHIA COLI By GARY PAUL O'NEILL M.Sa.McGill University, 1982 B.Sc.,McG111 University, 1979  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 this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA JUNE 1986 © Gary Paul O'Neill  71  Table VII. Base composition of each position 1n the codons used In the pex gene compared to other genes of high G + C content and E. poll genes.  75 Base composition of coding sequence Codon position  Gene  Base(s)  1  2  3  Reference  Cellulomonas fimi  A  24.0  25.4  0.4  cjx  G  39.5  18.4  41.7  C  21.9  34.9  56.8  T  14.7  21.3  1.0  G+C  61  53  98  6+C  78  43  97  Thompson and 6ray, 1983  6+C  73  49  89  Kagawa etal., 1984  G+C  63  49  92  Robbins etal., 1984  6+C  62  41  53  Varenne etal., 1984  (718 6+C)  8  StreDtomvces fradiae aph. (73?? G+C) Thermus thermoDhllus leuB (69% G+C) StreDtomvces Dlicatus endo H gene (6875 6+C) E. coM (508 G+C^  (a) The value In parentheses Is the overall guanoslne (G) plus cytoslne (C) content of the chromosomal DNA. ( ° ) The overall G plus C content for E. poUchromosomal DNA and for each codon position Is derived from the codon usage of 62 E. colt chromosomal genes (Varenne etal., 1984).  72  Table VIII. Codon utilization for the pre-exoglucanase  aa  No. codon 8  Phe 0  UUU  aa  No. codon  Ser  0  UCU  8  aa  No. codon  Tyr  0  UAU  10  UAC  8  aa  No. codon  Cys  0  UGU  6  UGC  8  22  UUC  8  UCC  0  UUA  0  UCA  Och 0  UAA  Opl  0  UGA  0  UUG  13  UCG  Amb 0  UAG  Trp  12  UGG  0  CUU  1  ecu  His  0  CAU  Arg  1  CGU  16  cue  10  CCC  7  CAC  12  CGC  0  CUA  0  CCA  0  CAA  0  CGA  9  CUG  22  CCG  22  CAG  6  CGG  0  AUU  0  ACU  0  A6U  9  AUC  22  ACC  0  AUA  0  ACA  Met 6  AUG  27  ACG  Val  0  GUU  27  GUC  28  GCC  0  QUA  2  GCA  14  GUG  34  GCG  Leu  lie  Pro  Thr  ALA 2  GCU  Gin  Asn 0  Lys  Ser  23  AAC  9  AGC  0  AAA  Arg 0  AGA  19  AAG  1  AGO  1  GGU  Asp 0  Glu  AAU  GAU  Gly  28  GAC  38  GGC  0  GAA  0  GGA  14  GAG  3  GOG  Number of times the given codon is utilized in the pex gene (total number of codons=484)  8  73  Table IX. Comparison of the codon usage in the cex gene to the average codon usage in 62 £ coll genes and tRNA levels in E. coli. The table lists as a percentage value the occurrence of each triplet in the plus strands of the DNA sequence coding for the cex gene and also the actual number of times each codon was used (* a.a.).The average codon usage in E. coli is a percentage calculation obtained from the cumulative DNA sequence of 62 E. coll genes (Varenne etal., 1984). Twentyone of the 62 E. coli penes analyzed correspond to highly expressed genes and 41 sequences correspond to weakly expressed genes. The total number of codons 1n thecex. genes Is 484 codons. The values for tRNA concentrations are taken from Varenne et al. (1984). Codons that are considered optimal (o) or nonopttmal (x) for translation of E. coli genes are Indicated (Grosjeans andFiers, 1982).  CODON  uuu uuc  AMINO ACID  C. flml exoglucanase # a.a. Z  average codon usage(Z)  E. coll Z tRNA cone.  optimal or nonoptlmal  0 22 0 0  0 4.5 0 0  2 1 0.5 0.6  1.96  UUA UUG  PHE PHE LEU LEU  UCU UCC UCA UCG  SER SER SER SER  0 8 0 13  0 1.7 0 2.7  1.7 1.4 0.3 0.3  3.2 2.78 1.4  CUUCUC CUA CUG  LEU LEU LEU LEU  0 16 0 9  0 3.3 0 1.9  0.5 0.5 0.1 6.2  1.68  X  0.68 5.61  X  ecu  CCC CCA CCG  PRO PRO PRO PRO  1 10 0 22  0.2 2.1 0 4.5  0.3 0.2 0.7 2.5  1.82 0.90 3.08  X  AUU AUC AUA AUG  ILE ILE ILE MET  0 9 0 6  0 1.9 0 1.2  1.7 4.1 0 2.3  5.61  X  0.29 1.68  _  ACU ACC ACA ACG  THR THR THR THR  0 22 0 27  0 4.5 0 5.6  2.0 2.5 0.4 0.5  5.01 4.49 1.73  o o  GUU  VAL VAL VAL VAL  0 27 0 14  0 5.6 0 2.9  3.2 0.9 2.2 1.8  4.01 2.24 5.89  ALA ALA ALA ALA  2 28 2 34  0.4 5.8 0.4 7.0  4.6 1.4 2.8 2.8  5.8 4.07 5.77  cue  CUA GUG  ecu  GCC GCA GCG  1.4  X  0  X  X  o X X X  X  o X  o o o X  X X  o X  o X  0  X  o  0  CODON  AMINO ACID  UAU UAC UAA UAG  TYR TYR OCH AMB  UGU UGC UGA UGG  as  C. f l n l exoglucanase # a.a. Z  E. c o l l average codon Z usage(Z) tRNA cone.  optlaal or nonoptlmal  O 10 0 0  O 2.1 0 0  0.7 1.5  2.80  -  -  CYS 0PL TRP  0 6 0 12  0 1.2 0 2.5  0.3 0.4  1.13  0.7  _  1.68  -  CAU CAC CAA CAG  HIS HIS GLN GLN  0 7 0 22  0 1.4 0 4.5  0.7 1.3 0.9 3.1  2.24  _  CGU CGC CGA CGG  ARG ARG ARG ARG  1 12 0 6  0.2 2.5 0 1.2  3.8 1.9 0.1 0.1  AAU AAC AAA AAG  ASN ASN LYS LYS  0 23 0 19  0 4.8 0 3.9  0.6 3.1 5.2 2.0  AGU AGC AGA AGG  SER SER ARG ARG  0 9 0 1  0 1.9 0 0.2  0.4 1.0 0.1 0  GAU GAC GAA GAG  ASP ASP GLU GLU  0 28 0 14  0 5.8 0 2.9  2.1 3.1 5.0 1.8  GGU GGC GGA CGG  GLY GLY GLY GLY  1 38 0 3  0.2 7.9 0 0.6  4.1 3.1 0.3 0.4  X  o  _  _  _  1.68 2.24  o  5.05  o o  X  X  0.68  X  3.37  X  0 o  5.61  X  1.4  _  0.68  X  _  X  4.49 5.05  o X  6.17  X  0.84 1.40  X  0 X  4  74  Pro-Thr box (Warren et al., 1986; Wong et al., 1986). There is also an Imperfect repeat of 23 amino acids In tanoam close to the carboxyl terminus of the Clostridium thermocellum celA endoglucanase In which 16 of the 23 amino acids of each repeat are Identical (Beguin et al., 1985). These repeated sequences could be Involved in binding two adjacent glucose residues of the cellulose molecule.  (5) Open reading frames in the 5' and 3' flanking sequences of the pex gene In the 5' and 3' nontranslated sequences of the pex gene, open reading frames were found in both the BamHI to Sal' and the Sail to BamHI direction of transcription (Fig. 11). One or more of these open reading frames could encode a polypeptide that begins or ends in one of these open reading frames. The high G+C content of the Exg coding sequence resulted In a scarcity of stop codons (composition: 0 to 3358 GC) and a corresponding abundance of open reading frames.  (6) HydrophilicUy analysis of the Exg and the EngA peptides Figure 12 shows the results of the hydrophlllclty analysis of the precursor forms of the Exg and the related EngA cellulase from C_. fjroi (Warren et al., 1986; Wong et al., 1986). The analysis was performed according to Hopp and Woods (1981). In this method, a numerical hydrophllicity value was assigned to each amino acid. The plots In Fig. 12 were generated by repetitively averaging these values along the peptide chain, where the averaging group length was six amino acids and the 'step' was one amino acid. These average hydrophilicity values versus sequence position were then plotted. Two points were made from these plots. First, the putative signal peptides showed a typical structure with a hydrophilic amino terminus and a hydrophobic carboxyl terminus (Watson, 1984). Second, the amino terminal portion of the mature Exg was hydrophilic compared to its carboxyl terminus. In contrast, the EngA showed an opposite configuration with its amino terminus being hydrophobic. This observation is in agreement with Warren et al. (1986). These authors also concluded that there is considerable amino acid sequence  75  Figure 11. Open reading frame analysis of the DNA sequence for the cex coding region and Its 5' and 3" flanking regions. The locations of all ORF's of at least 20 amino acids (A and A') and only ORF's of at least 20 amino acids starting at ATG's (B and B') are shown for the DNA sequence of the cex coding region and its 5' and 3' flanking regions (see Figure 8). The ORF's in the BamHI to Sal' orientation of the cex. Insert are depicted In figures A and B. ORF's In the Sail to BamHI orientation of the cex insert are depicted in figures A' and B'. The location and orientation of the actual cex coding region is indicated by the heavy arrow and corresponds to the long ORF in reading frame 3 in the BamHI to Sail direction in figure B.  75a  V. of T o t a l 10 1  .  20 1  30 1  .  .  DNA Length  40 50 60 1 . 1 . 1 , reading frame: i  70 1  ,  1  80 ,  1  90 ,  1 00 1  reading -frame: 2  reading -frame: 3  17 open  reading  -frames  found.  B .  1  10 1  1  20  30  I  . 1 •  1 1 I  • r  V. of T o t a l DNA Length 40 50 60 1 , 1 , 1 .  70  reading -frame: i  .  80  i  9p  100  1—  •  » i  reading frame: 2 • reading frame: 3  •  1  ' 5 open r e a d i n g  frames  cex BamHI  1  found.  ^ Sail  76  A' •  10 J1  20  1  '/. of T o t a l  30  1I  I  '  40 1 I  .  I  •  •  . .  DNA Length  1 1  60  . •  1 I  reading frame: i  1.  1  50  70  . ,  80  1 1  1 I  ,  H  .  i  90  1  .  1  j  100 j  j  '  •  reading frame: 2 *l  1  —  ^  h  •  reading frame: 3 •  1  h+  h—• 24 open  10 i  .  20  i  ,  30  i  •  reading  frames  found.  V. of T o t a l DNA Length 40 50 60  i  •  i  .  i  •  70  i  reading frame: i  80  •  i  1  •  90  i  100  •  i  •  reading frame: 2 1  •  r-> I  • reading frame: 3 1  6 open r e a d i n g  A  frames  •  found.  cex  <2> Sail  BamHI  77  conservation between the Exg and the EngA proteins. Both enzymes can be divided into three domains: (1) an irregular domain of low charge density lacking any secondary structure; (11) an ordered domain with higher charge density, secondary structure and a potential active site; and (iii) the Pro-Thr box. These three domains are arranged differently In the Exg and the EngA proteins.  (7) Dot matrix analysis of the sex gene with Itself and with the penA gene Warren et al. (1986) suggested that the cex and the penA genes of C. fimi may have arisen by the shuffling of two conserved sequences and either one or two other sequences. In order to Investigate this possibility, dot matrix comparisons of the DNA sequences of the pex gene and the cenA gene were performed (Fig. 13). The high resolution plots of these two genes detected homology between the Pro-Thr boxes, which appeared as a square of clustered points, at a position corresponding to nucleotides 1015 to 1160 of the cex gene and nucleotides 405 to 540 of the cenA gene (Fig. 13A and B). One other region of DNA sequence homology, Indicated by a series of points on a diagonal between nucleotide 1015 on the pex axis and nucleotide 405 on the penA axis, was evident (Fig. 13AandB). Further analyses (Warren et al., 1986) show that the proteins encoded by the cex and penA gene contain three distinct domains: a short sequence of 20 prolines and threonines (the ProThr box); an Irregular domain lacking secondary structure, which is rich in hydroxyamino 8cids but of low charge density; and an ordered domain having secondary structure, a higher charge density and a potential active site. The Pro-Thr box is conserved almost perfectly in the two enzymes. The irregular domains are 50 % conserved, and the conserved sequences include four Asn-Xaa-Ser/Thr sites. There Is very little conservation 1n the ordered domains, but the potential active sites both have the sequence Glu-Xaay-Asn-Xaeg-Thr; they occur at widely separated sites in the two domains. The order of the domains Is reversed In the two enzymes: irregular /Pro-Thr box/ordered In the endoglucanase; ordered/ Pro-Thr box/irregular in the exoglucanase.  78  Figure 12. Hydrophiliclty analysis of the precursor forms for the C. fjmi Exg (A) and EngA (B). The amino acid sequence of EngA was deduced from the DNA sequence of the cenA gene (Wong et al., 1986). The analysis and hydrophllicity values assigned to each amino acid were as described by Hopp and Woods (1981). The averaging length was six amino 8cids and the step was one amino acid. The locations of the signal peptide and the Pro-Thr box (PT) are Indicated.  78a  Hydrophilicity Plot DMA  v o l : DHfl d a t a f i l e : cex-cdngseqnc  Reading fraae: 1 Peptide •: 1 +1  Starting nt:  * amino a c i d s : 494  Ho I »t: 51234 fluerag i ng I en: 6 Wet charge: -3 i  8.0  I  i  I  i  I  i  I  >  I  I  1  1  I  1  I  1  I  0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 f r a c t i o n o f length  1.0  B Hydrophilicity Plot DNfl  v o l : DNfl d a t a f i l e : cenflcdng  Reading fraee: 1 Peptide •: 1 Starting nt:  +1  * aaino a c i d s : 44Q  Hoi »t: 46679 Ruerag i ng i en: 6 Net charge: +1 I  0.0  i  I  i  I  i  I  i  I  i  I  i  I  i  I  i  i  i  I  i  0.1 0.2 8.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 f r a c t i o n o f length  79  Figure 13. Dot matrix comparisons of the Exg DNA coding sequence with itself and the EngA DNA coding sequence. In figures 13A and 13B the entire nucleotide sequence of the pex gene Is compared with the entire nucleotide sequence of the cenA gene (Wong et al., 1986). In figure 13C, the entire nucleotide sequence of the cex gene Is compared with Itself. In figures 13A and 13C each dot represents the center of a 7-base Identity with a maximum of one mismatch in the homology. In figure 13B each dot represents the center of a 4-base Identity with no mismatches. Regions of homolgy between two sequences appear as diagonal lines at a 45 angle. The diagonal line in figure 13C Is computer-generated. Direct repeats appear as parallel lines across the grid. The numbers on each axes indicate the distance in nucleotides from the first nucleotide of the initiating ATG codon. The A of the ATG start codon Is the first nucleotide. s  79a  DNAttl  (X-axis,  left  DNA#2  (Y-axis,  top  matrix  resolution:  search  element  maximum  #  of  145  290  — > >  length:  DNA  data:cex-cdngseqnc  DNA  data:cenAcdng  x  1000  mismatches:  right): bottom):  (1452  nts)  nts)  1000  7 1  ATG 1  (1347  CEX 435  580  725  TGA 870  1015  1160  1305  1450  80  DNAR1 ( X - a x i s , DNAH2 ( Y - a x i s , matrix  left —> top — >  resolution:  1000  r i g h t ) : DNA d a t a : c e x - c d n g s e q n c ( 1 4 5 2 b o t t o m ) : DNA d a t a : c e n A c d n g ( 1 3 4 7 n t s ) x  search element l e n g t h : 4 maximum II o f m i s m a t c h e s : 0  1000  nts)  81  DNAttl ( X - a x i s , DNA#2 ( Y - a x i s , matrix  le-ft — > top — >  resolution:  1000  r i g h t ) : DNA d a t a : c e x - c d n g s e q n c bottom): DNA d a t a : c e x - c d n g s e q n c x  search element l e n g t h : 7 maximum # o f m i s m a t c h e s : 1  1000  (1452 n t s ) (1452 n t s )  82  II Immunoadsorbent purification of the recombinant Exg  The purification of cellulases, including especially the recombinant C. fimi Exg, for the purpose of their biochemical characterization was interesting for several reasons. First, the C. fimi Exg is glycosylated (Langsford et al., 1984). The glycosylatlon of the C. fimi Exg is a rare example of protein glycosylation by a prokaryote. The role of glycosylatlon of cellulases is not known but it has been suggested to contribute to enzyme stability and activity (Coughlan, 1985b). Second, there is little information on the cellulases regarding their structures, their modes of action, their post-secretional modifications by proteases or their mechanisms of B-1,4 bond cleavage. All of these studies require biologically active pure enzyme. The problem is that the supernantant fluids of cellulolytic organisms contain multiple cellulolytic activities of similar size and enzymatic activity. Furthermore, each of these components can exist in different forms due to differential processing and proteolytic modification (see Introduction). The isolation of the recombinant form of the Exg offered a unique opportunity to compare its properties with those of the native form. The recombinant form of the Exg is not glycosylated because E. coli does not contain those enzymes required for protein glycosylation. In addition, the recombinant Exg is expressed in a cellular background lacking any interefering cellulolytic activities which may complicate purification and identification of the enzyme. Finally, in the initial stages of this work, it was necessary to confirm the molecular size of the recombinant cellulase which had been determined by a maxicell technique (Gilkes et al., 1984a). Thus, I developed a simple and rapid purification protocol for the recombinant Exg which is described below. An immunoadsorbent purification scheme, outlined in Fig. 14, was employed for the isolation of Exg from either osmotic shock fluids or crude cell extracts of E. coli C600 (pEC-1.1). Details concerning the growth of cells, the preparation of crude starting materials and the preparation and use of the immunoadsorbent columns are described in the Materials and Methods. Starting material (whole cell extracts or osmotic shock fluids) was first passed over an S4B-anti-cellulase antibody affinity column prepared by coupling 3g of rabbit anti-cellulase IgG  83  to 90g of CNBr-activated Sepharose 4B. The Exg was retained on the column. After extensive washing of the column with HS-PBS, the retained proteins, which included the Exg, were eluted with 3M NaSCN. Following NaSCN elution, the column was regenerated by washing with 4M guanidine-HCl and HS-PBS. The use of NaSCN for the elution of the Exg from the S4B-anticellulase antibody affinity column was important, because several of the commonly used immunoadsorbent elution buffers partially inactivated the Exg whereas 3M NaSCN had minimal effects on the enzyme activity of the Exg. The elution buffers which had a deleterious effect on the Exg included: (i) O.IM glyclne-HCl, pH 2.6- 0.5M NaCl; (11) 0.1M glycine-HCl, pH 2.6- 50*5 ethylene glycol; (iii) O.IM NaHC03, pH 10.5- 0.5M NaCl; (iv) 0.1M NaHCC^, pH 10.5- 50* ethylene glycol. The chaotropic agents NaSCN and NaCIO^ caused less than a 1051 reduction of Exg activity when used at concentrations of less than 3M. NaSCN was chosen as the elutlng agent because it has a higher chaotropicity than NaClO^. Fractions eluted from the S4B-anti-cellulase antibody affinity column with 3M NaSCN were assayed for pNPCase activity immediately after elution without removal of the NaSCN. Fractions with pNPCase activity were pooled, concentrated by ultrafiltration and dialyzed against PBS. Analysis of this material by SDS-PAGE (Fig. 15, lane C) revealed a major protein band with an M,. of 50,000-51,000 which agreed well with the M,. of 51,000 determined for the recombinant Exg by the maxicell technique. A number of other proteins remained after the S4Banti-cellulase antibody affinity column, which copurified with the Exg, due to their nonspecific adsorption to the immunoadsorbent. A second passage of the eluted Exg fractions over the S4Banti-cellulase antibody affinity column did not eliminate these contaminants (Fig. 15, lane B). The removal of the remaining contaminants was accomplished by negative affinity chromatography on an S4B-anti-E. coli antibody affinity column. This column was prepared by conjugating 0.16 g of rabbit anti-whole E. coli IgO to 4.5g of CNBr-activated S4B. Exg activity was not retained by this column and appeared in the void volume. The purity of the Exg was established by the appearance of a singleband on SDS-PAGE (Fig. 15,laneA). The identity of this band as Exg was confirmed by raising an antiserum to this protein band in a rabbit. The antiserum to this purified protein  84  had Immunological reactivity to purified Exg and also to the cellulases In supernatant fluids of C.fjmi cultures. E. coll proteins in the Exg samples were usually removed by a single passage over the S4B-anti-E. coli antibody affinity column. A detailed study of the purification of the Exg using these immunoadsorbent columns has been carried out by N. Arfman and is presented elsewhere ( Arfman, 1986). The S4B-anticellul8se antibody affinity column in conjunction with a single desalting step has yielded Exg of a purity suitable for amino acid sequence determination and enzyme kinetic studies (Gilkes et al., manuscript In preparation). In conclusion, the Immunoadsorbent chromatography approach provided a simple and fast procedure for the purification of pure, biologically active Exg from E. coll.  85  Figure 14. Scheme for the Immunoadsorbent purification of the recombinant Exg. An osmotic shxkate fluid or a total cell extract was prepared from the cell pellet from a 20 liter log-phase culture of £. collC600( pEC-1.1). This material was then passed sequentially through two immunoadsorbent chromatography columns. The first immunoadsorbent column consisted of rabbit polyclonal anti-cellulase IgG covalently linked to Sepharose 4B. The material bound by this column was eluted with 3M NaSCN. PNPCase activity was recovered only in the NaSCN eluted material. The eluted material was then passed over the anti-whole E. ctfi IgO Immunoadsorbent column to remove contaminating E. coli proteins. The pNPCase activity was contained in the flowthrough material from this column.  86 Figure 15. SDS-polyacrylamlde gel analyses of the Exg purified by Immunoadsorbent chromatography. Lane A, purified Exg after passage over the anti-whole £. coli IgO Immunoadsorbent column; lane 6 and C, the material eluted with NaSCN from the S4B-ant1cellulase immunoadsorbent column after one passage of the crude cell extract (C) or after two sequential passages (B); lane D, periplasmic proteins In shockate fluid of C600(pEC-1.1); lane E, proteins in whole cell extract of C600 (pEC-1.1); MW, molecular weight markers.  A  B  C  D  E  MW  ^•HHHfr  2  9  87  111 Expression of the Exg In E. co_M (A) Introduction My Interest In the Exg was twofold. First, I was Interested In the structure and function of cellulases, about which very little 1s known. Second, cellulases may prove to be of industrial Importance In the conversion of cellulosic waste materials to glucose, which could be fermented to more valuable products. To study the relationship between the structure and function of the Exg or to use the Exg in Industrial applications, It was necessary to isolate active enzyme In large amounts. Since the molecular genetics of Escherichia coll is so advanced and many heterologous proteins had been expressed In this organism, I chose to begin engineering the overproduction of the cex gene product In E. coli. As discussed above, a striking feature of the cex gene was an extreme codon bias. Over 98* of the codons used had 6 or C as the third letter, and of the 484 codons in the cex gene, 26 of the 61 possible codons were not utilized, 4 codons were used once, and 2 codons were used twice. Thus, 98* of the codons In the cex gene were coded for by only 29 of the 61 possible codons. This nonrandom pattern of codon usage Is very different from that found In £. coli and appears to be a characteristic trait of genes from organisms with a high chromosomal 6 + C content (Bibb et al., 1984). The 6 + C content of £.flmichromosomal DNA Is 72*. Several groups have suggested, based on theoretical ( Grantham et al., 1981; Grosjean and F1ers, 1982; Ikemura, 1981) and experimental (Bonekamp etal., 1985; Robinson etal., 1984; Varenne et al., 1984) findings, that codon usage plays an important role in the regulation of gene expression and In heterologous gene expression. The basis of the argument is that for efficient translation, the codons used in the foreign gene should be consistent with the levels of the cognate tRNAs In the host cell. The argument follows that the cognate tRNAs for the codons used in the foreign gene may be so rare in the host cell that translation is slowed. For example, the cex gene h8d 6 CGG codons and 1 AGO codon for Arg , which are rare codons in E. coli genes (Varenne et al., 1984). I n this chapter, I describe the molecular genetic manpulatlons which yielded  88  expression of a recombinant form of the C. Ami Exg In E. coli to a level exceeding 20* of the total cellular protein.  (B) Construction of Exg-expression plasmlds-Transcrlptlonal fusions A set of operon and gene fusions of the cex gene to the various E. coli genes and operons listed in Fig. 16 were constructed for the purpose of overproducing the Exg in E. coli. Attempts to express the C M gene directly under the control of E. coli transcriptional promoters were moderately successful. Originally the cex gene was expressed under the control of the tet promoter (Gilkes et al., 19848; Whittle et al., 1982). Subsequently, an eightfold increase in Exg expression was obtained when the cex gene was placed under the stronger tac promoter ( Fig. 5). Although an eight fold increase of Exg activity in extracts of cells containing pEC-1.1 t§c W8S obtained, I could not detect any new protein bands in total cell extracts by SDS-polyacrylamide gel electrophoretic analysis or by radioactive detection in an E. coli minicell system (Jensen et al., 1984). Since the recombinant Exg h8d roughly the same specific activity as the native C. fimi Exg (N. Gilkes, unpublished observations), the recombinant Exg was present at a level of 0.001 to 0.01 * of the total cellular protein in E. coli. Since abundant mRNA transcripts from the tac promoter fusion were detected( N. Greenberg, unpublished observations), and since the Exg is very resistant to E. cj)li proteases, the lack of high-level Exg expression in these constructs may be attributed to a translational block. Such a block may occur (i) as a result of a lack of optimal translational initiation of the cex gene in E. coli when it is not fused to optimal E. coli translational signals (Gold et al., 1981) or (ii) as a consequence of the biased codon usage in the cex gene ( Grantham et al., 1981; Grosjean and Fiers, 1982; Ikemura, 1981; Robinson et al., 1984; Varenne etal., 1984).  89  Figure 16. Summary of the operon and gene fusions constructed between the cex gene and E. coll genes for the purpose of the overproduction of the Exg in E. coJi. The top box is a schematic of the structure of the cex. gene. Lines 1 to 4 show the E. coli gene expression signals that were fused to the cex. gene. ATG, translational initiation codon; P, promoter; RBS, ribosome binding site. The Exg activities determined by some of these constructions are shown In Fig. 6,17, and 22 and Table X and their structures are described in Fig. 5,19,25,26 and 28.  Promoter  •  R B S  Structural  gene  coding  region  Terminator  Promoters  R.B.S.  Gene fusions  COOH-TGA|  tet. loc. P  L  . P  R  toe •  . P S Y N  Ipp.  synthetic  synthetic  locZ. Ipp. ompA +  amino terminus  troE  90 (C) Construction of Exg-expression plasmids-Translational fusions (1) lacZ fusions I first tested the possibility that the RBS of the cex gene was not recognized efficiently by E. coli ribosomes, since bacterial RBSs differ widely in their strength of translation initiation (Gold et al., 1981). The natural transcriptional and translational regulatory signals of the cex gene were completely replaced with those from the E. coli lac operon. Three different fusions were constructed between the alpha-fragment offiGalIn pUCI 2 and the Exg signal sequence (Fig. 17). The best fusion, pUC 12-1.1 (737), yielded a 15-fold Increase In Exg expression when compared to expression of the Exg under Its own translational Initiation signals 1n plasmid pUCI 2-1. Icex (Fig. 5 and 17). However, the amounts of Exg produced by strains bearing the different fiGal-Exg fusion plasmids varied enormously.fiGal-ExgmRNAs transcribed from the various plasmids differed in their 5" mRNA structure and these differences may have affected mRNA stability, mRNA processing, or rlbosome binding efficiency (Looman et al., 1985). I concluded, however, that an increase 1n Exg activity in E. coli cells could be attained by using an E. coli RBS-ATG translational Initiation sequence in place of the C. fimi RBS.  (2) Expression of the Exg fused to a portable translation initiation site in plasmld pUCI 2 To circumvent the possibility that the native pex RBS does not initiate translation efficiently in E. coli, and to avoid the complications of a hybridfiGal-Exgprotein, the natural pex RBS-ATG was replaced with a synthetic portable translational initiation site (PTIS; de Boer et al., 1983). The PTIS Is a 17-bp DNA fragment containing a 5 base Shine-Dalgarno (S-D; Shine and Dalgarno, 1974) region with a contiguous 3' poly(A) tract followed by an initiator ATG codon eight bases from the S-D region. This sequence (Fig. 18) was determined to be highly efficient for the initiation of translation in E. opji (de Boer et al., 1983). Replacement of the lac RBS in the hybridfiGal-Exgexpression plasmid pUC 12-1.1 (737) with the PTIS to give pUCI 2-1.1 (PTIS) yielded a further 10-fold increase in Exg activity in cell extracts (Fig. 17).  91  The increase in Exg expression obtained by using the PTIS may be attributed to several factors which have been shown to be Important In translation initiation. First, although the RBS sequences are Identical in the PTIS and the pex gene (5'-G0AG6-3'), Its separation from the start codon may be more optimal in the PTIS (8 nucleotides) versus in the cex gene (5 nucleotides) (Fig. 18). In£. poll genes, an RBS-ATG separation of 6 to 12 nucelotldes has been shown to be optimal for translation Initiation (Gold et al., 1981; Kozak et al., 1983). Second, the nucleotide composition between the RBS and start codon of the PTIS consists only of A/T bp. Hall et al. (1982) showed that secondary structures formed between the SD region and regions downstream of the ATG codon can affect translation efficiency. The AT-rlch sequence of the PTIS greatly reduces the opportunity for secondary structure formation with the GC-rlch (72% G + C) cex pane sequences. Third, Stanssens et al (1985,1986) and Looman et al, (1985) have shown that altering the sequence of the 5' ends of mRNAs can dramatically alter the biological half-life of the mRNA and the ability of rlbosomes to initiate translation on the mRNA. Thus, the 5' end of the PTis-cex mRNA may have an increased resistance to degradation, leading to an increased biological half-life, higher steady-state levels of the PTIS-pex mRNA, and Increased translation of the cex gene.  92  Figure 17. DNA sequences of the RBS, translational Initiation site, and amino termini of fusion junctions of BGal-Exg expression plasmids and the level of Exg activity in cell extracts. pUCI21.1 cex and pUC 13-1.1 cex code for unfused pex gene products. The numbering of the codons of the natural cex.gene product in pUCI 2-1. Ipex and pUCI 3-1.ipex begins with the initiating ATG of the signal sequence as -41 and the first codon of the mature Exg as +1. The first pex codon In the BGal-Exg fusions retains its original position number. The deduced amino acid sequence is shown in single letter code over the DNA sequence. The nucleotides and amino acids derived fromfiGalare underlined. Lower case amino acids are of non-lac origin and are derived from the linker region in pUC 12. The restriction sites StyJ, Aval I, and EcoRI I In the amino terminus of the cex. gene were used for fusion of the pex gene to the amino terminus of BGal in pUC12. ( ) Exg activity is expressed as nanomoles of p-nitrophenol released/min/mg of total cell protein. a  Plasmid  Promoter  PUC13-1 lcfix  pUC 12-1.1 ££js.  RBS.  cex  la£  cex  S e q u e n c e o f t h e Exg o r E x g - B G a l f u s i o n  AG6AG6ACATC  A6GAGGACATC  pUC 12-1.1(737)  lac  latZ  AGGAAACAGCT  PUC12-1.K746)  ]a£  latZ  AGGAAACAGCT  Exg a c t i v i t y -3 -2 -1 « l -2 A 0 A A T GCCCAGGCCGCGACC EcoRll  -41 -39 -37 -35 M P R T T P A ATGCCTAGGACCACGCCCGCA Styl Avail -41 -39 -37 -35 M P R T T P A ATGCCTAGGACCACGCCCGCA  -2 -1 •! '2 A Q A A T GCCCAGGCCGCGACC  0.07  -3  (-40) T M I T N S s s p g d P R ATGACCATGATTACGAATTCGAGCTCGCCCGGGGATCCTAGG BamHI Styl (-37) T M I T N S T P ATGACCATGATTACGAATTCGACGCCC EcoRI  1.0  15.0  I 1.0  M)  PUC12-1.K859)  pUC 12-11(PTIS)  l3£  ]2£  l3£l  PTIS  T M I T N S A T AGGAAACAGCT ATGACCATGATTACGAATTCCGCGACC EcoRI (-40) M D P R T T G G A G G A A A A A A T T ATGGATCCTAGGACC BamHI Styl  0.24  146  3  93  Figure 18. Comparison of the nucleotide sequences of the translational Initiation regions of the PTIS and the cex gene. The two sequences are aligned at their ATG start codons at the 3' end. The purlne-rlch rlbosome binding sites are boxed. Homologous bases in the two sequences are marked with a dot.  synthetic portable  translation  initiation site C. fimi  exoglucanase  translation initiation site  f-me1  E. c o l i  3-AAUUUIS G AG GA A A A A A U U A U G - 3 ' • •••• • • •• •  ACAU C 5 - G A C G AG G AG G  AU G- 3'  f-met  94  (D) Plasmids with thermolnduclble control of runaway replication and transcription for the expression of the Exg  (1) Plasmld pCP3cex construction The most successful approach for overproducing the Exg Involved fusing the cex gene to the PTIS and placing these sequences under the transcriptional control of the lambda P|_ promoter contained In the plasmld pCP3. This plasmld Is a derivative of the runaway replication plasmld pKN402 (Remautetal., 1983). The specific procedure I used to construct the P|_ promoter PTIS-Exg gene fusions Is given In Fig. 19 and described In detail In the Materials and Methods. The construction was facilitated by a unique StyJ cleavage site located In codons 2 and 3 of the nucleotide sequence for the precursor form of the Exg. Thus, the entire cex coding sequence , except for the Initiating AT6, and a putative transcriptional terminator, was excised easily. The construction scheme Involved abutting the Exg coding sequence at the second codon of the signal sequence to the Initiating codon of the PTIS. The resulting plasmld, pCP3-cex , coded for the natural Exg except for the Insertion of an Asp codon at the BamHI restriction site, between codons 1 and 2 of the Exg signal sequence (Fig. 19). The plasmld pCP3-cex was transformed Into E. puli C600( pCI857). The plasmld pel857 codes for kanamycin resistance and for the thermo1ab11ecl857 gene product (Remautetal., 1983). At 30°C In the strain C600(pcJ 857) containing pCP3 or pCP3-cex, plasmid replication is regulated normally and transcription from the P|_ promoter is repressed by the cl857 gene product. In contrast, at 41°C, derepression of both plasmid replication and of transcription initiating at P|_ occurs (Fig. 20).  (2) Plasmid pCP3-cex-directed Exg expression Cultures of strain C600 (pcl857) carrying pCP3 and pCP3-pex were grown at 30°C, induced for 90 minutes at 41°C, lysed, and analyzed by SDS-PAOE (Fig. 21). Cell extracts of strains containing pCP3-cex which had been induced expressed a new protein. The Mp of this new  95  Figure 19. Scheme for the construction of the Exg-expresslon plasmids pUCI 2-1.1 (737) and pCP3-cex. See the Materials and Methods for details. The DNA sequence of the RBS and cex coding region Immediately 3' to the P|_ promoter in pCP3-cex Is shown. The functional orientation of the gene coding for B-lactamase (Ap ), Exg, and the las and XP|_ promoters are indicated by arrows. • I .P.T.I.S.; ELD , coding sequence of the peigene signal sequence; EZ3, coding sequence of the mature Exg; • , noncoding sequences of £. fimi DNA. Restriction sites: B, BamHI; E. EcoRI; H3, JHlndMI; S, M l ; St, SJyJ. The deduced amino acid sequence for translated codons Is Indicated In one-letter code above the second nucleotide of each codon. R  96  Figure 20. Diagrammatic representation of the X P|.-Exg expression system.  2)  transcription (strong  promoter,  y-mediated 3) t r a n s l a t i o n  (optimal ond  temperoture-inducible,  ontitermination)  R B S ,R B S - A T G  intervening  configuration  sequence)  97  protein was 58,000 to 60,000 (Fig. 21, lane C). The Mr- of the mature form of Exg Is 56,000 (Gilkes et al., 1984b). The difference in the apparent molecular weights of the recombinant Exg and the native C. fimi Exg was probably due to the lack of removal of the Exg signal peptide by the E. coli signal peptidase. Recent amino acid sequencing results in our laboratory ( Arfman, 1986) show that at low levels of Exg production in £ coli, the signal peptidase correctly processes the signal peptide from the precursor Exg and the Exg is exported to the periplasmic space. However, as discussed below, the pCP5-cex-directed overproduction of Exg lead to its cytoplasmic accumulation. Since signal peptide removal is almost always an essential step in protein translocation (Pugsley and Schwartz, 1985), I expected the cytoplasmic Exg to be the precursor form of the enzyme. Thus, it appeared that the E. coli host did not remove the 42- amino acid Exg signal sequence since the Mr- of the recombinant form of the Exg W8S that predicted for the mature form of the Exg purified from C_. fimi ( Fig. 21, lane B) plus its signal sequence. The new protein made up over 208> of the total cellular protein in induced E. coli cultures as determined by densitometric scanning of the Coomassie blue-stained polyacrylamide gel.  (3) Effect of induction temperature and length of induction on Exg production by oCP3cex. The new protein accumulated rapidly in induced cultures, reaching a maximum level in 45 to 60 minutes based on SDS-PAGE of total cell extracts (Fig. 22B). However, maximal Exg activity was attained 30 minutes after induction and then decreased rapidly (Fig. 22A). This rapid increase and decrease of Exg activity in induced cultures of C600 (pcl857)( pCP3-cex) might be explained by the deposition of the overproduced Exg in an inactive form. Intracellular aggregates could be visualized by conventional light microscopy in induced C600 (pcl857)(pCP3-cex) cells. Presumably, the high intracellular concentration of the new protein lead to its localization in intracellular inclusion bodies in an enzym8tically inactive form  8S  has been previously  reported for other overproduced proteins in E. coli (Marston et al., 1984; Pages et al., 1984;  98  Figure 21. SDS-PAGE analysis of the recombinant pCP3-cjx,-produced Exg and native secreted Exg. The cells were grown at 30"C and induced at 41 *C for 90 minutes as described in the Materials and Methods. Lane A, an enriched Insoluble fraction of recombinant Exg obtained from induced cultures of C600 (pcl857)( pCP3-cej<) 8fter sonication of cells and removal of soluble material by low speed centlfugatlon; lane B, purified mature Exg from culture supernatants of C. fimi; lanes C and E, cell extracts of cells Induced at 4 TC; lanes D and F, cell extracts of cells grown at 30*C; lanes C and D, C600 (pcJ857)(pCP3-ce£); lanes E and F, C600 (pc_l857) (pCP3); laneG, molecular weight standards: rabbit muscle myosin, 205,000; E. colifiGal. 116,000; rabbit muscle phosphorylase B, 97,400; bovine serum albumin, 66,000; ovalbumin, 45,000; carbonic anhydrase, 29,000.  99  Schoner et al., 1985; Winkler et al., 1984). Induction temperatures other than 40 to 42°C resulted in lower yields of the new protein (Fig. 23). (5) Cellular localization of Exg to insoluble aggregates and identification of the Exg To evaluate the identity of the new protein synthesized by the bacterial cells containing the pCP3 -pex recombinant plasmid, the overproduced protein was isolated and tested for enzymatic activity and immunological cross-reactivity with C_. fjmi Exg. The overproduced protein was Isolated in a highly enriched form by low-speed centrifugation of induced cells that had been disrupted by sonication (Fig. 21, lane A and Fig. 24). Aggregates of the new protein sedimented readily and could be solubilized in either 6 M urea or 5 M guanidine-hydrochloride. After solubilization of the aggregates and removal of the solubilizing reagent by dialysis against 50 mli Tris, pH 7.4, Exg activity was detected. The renaturation of the Exg after solubilization in urea or guanidine-hydrochloride and removal of the denaturant by dialysis against 50 mM Tris, pH 7.4, was not complete. The specific activity of this renatured recombinant Exg was approximately 250-fold lower than that of the native Exg. Since the Exg contained 6 cysteines, the renaturation of the recombinant enzyme may have been limited by the rearrangement of disulfide bridges. Thus, I tested the renaturation of the guanidine-hydrochloride-solubilized Exg in a glutathione redox buffer which facilitates formation of disulphide bridges (Ahmed etal., 1975; Winkler etal., 1984). This procedure resulted in a two-fold increase in recovery of enzymatic activity. As indicated above, it seemed likely that the aggregated cytoplasmic enzyme had not been processed. It was possible that the specific activity of this precursor form of the Exg was lower than that of the mature form. The precursor forms of several proteins have been shown to lack biological activity (Pugsley and Schwartz, 1985). For example, the precursor form of B-lactamase is enzymatically inactive and has a substantially different protein conformation as compared with the mature form (Roggenkamp et al., 1985). Furthermore, the enzymatic activity of an endoglucanase from Penicillium ianthinellum. which is  100  Figure 22. Time course of production after Induction of Exg as determined by enzyme activity (A) and SDS-PAGE (B). C600 (pel857) containing pCP3 and pCP3-pex was grown at 30*C and Induced at 41 C as described In the Materials and Methods. Samples (10 ml) were withdrawn from cultures at 0,15,30,45,60,90, and 120 minutes after induction. Cell extracts were prepared and assayed for Exg activity (A) and total cellular proteins were analyzed by SDS-PA6E (B). (-) lanes, C600 (pcJ857)(pCP3); ( + ) lanes, C600 (pcJ857)(pCP3-cei). The time after induction corresponding to each sample Is Indicated at the top of the gel. The position of the Exg Is indicated by an arrow. The positions of molecular weight standards are indicated. #  Time  after  induction (min)  101  Figure 23. SDS-PAGE analysis of proteins synthesized by £. C600 (pcJ857)( pCP3cgx) at 30'C, 34'C, 37°C, 40°Cand 43'C. 50 ml cultures of C600 (pcl857) containing pCP3 (- lanes) or pCP3oB( (+ lanes) were grown at 30*C to an O.D.soo o 0-25 and then divided Into 5 allquots. One aliquot of each strain was grown at the indicated temperature for 240 min. The cells from 10Ou.1 of culture fluid were collected by centrifugation, resuspended in 50ul of SDS-PAGE loading buffer, boiled for 5 min., centrifuged for 5 min. at 23*C and 5-1Oyl of each sample were electrophoresed through a 1058 PAG containing 0.1 % SDS at 100 volts. The gel was then stained with Coomassie brilliant blue. f  30°  MW 205  I—I  -  +  3.7°  34°  -  +  -  +  •  4,0°  -  +  43°  "  +  102  secreted as an Inactive procellulase, Is activated by proteolytic processing (Deshpande et al., 1984b). The Exg could be recovered from E. coli C600 (pcJ857)(pCP3-cex) In a nonaggregated form If the cells were Induced at a temperature of less than 37°C. Presumably at this temperature, the Inactlvation of the c!857 repressor protein was Incomplete, leading to a partial derepression of Pj_, a reduced yield of Exg, and an intracellular Exg concentration that did not lead to Its aggregation. At least some of the antigenic properties of the overproduced protein were similar to those of the authentic C. fimi Exg. The renatured Exg was retained specifically on an Immunoadsorbent chromatography column prepared with a rabbit antl-C. fimi cellulase antiserum and could be eluted with 3 M NaSCN. In contrast, the overproduced Exg was not retained on a rabbit ant1-( whole E. coJi) Immunoglobulin G Immunoadsorbent column (data not shown).  (E) Conclusions I showed that the overproduced E. coli Exg Is enzymatlcally active and exhibits immunological properties similar to those of the natural protein. Presumably, the recombinant Exg Is not glycosylated, since E. coli does not glycosylate proteins. It Is possible that further Increases In expression can be achieved by using different strains of E. coli, or different growth media, or different methods of induction (Mott etal., 1985; Remaut etal., 1983). By obtaining high-level expression of the cex. gene I showed that a prokaryotic gene with a codon usage pattern very dissimilar to that utilized by E. coli can be efficiently overproduced in this organism.  103  Figure 24. SDS-PAGE analysis of the Exg overproduced by pCP3-cjx,and Isolated by low speed centrifugation. Cultures (75 ml) of C600 (pel857) containing pCP3 (- lane) or pCP3-pex (+ lane) were grown at 30'C to an O.D.6oo 0-2 and Induced for Exg production by shifting the culture to 40°C for 90 min.. The cells (10 ml) were harvested by centrifugation, washed in 50 mM Tris, pH 8.0, resuspended in 1.5 ml of 50 mM Tris, pH 8.0, and then disrupted by passage through a French pressure cell at 10,000 psi. The insouble material In the cell extract was then isolated by low speed centrifugation. The cell extracts were centrifuged sequentially at forces of 1000,2000,3000,4000, and 5000 X g (i.e. after each centrifugation, the supernatant fluid was decanted and centrifuged at the next highest centrifugal force). The sedimented material from each centrifugation was resuspended in 300 ul of SDS-PAGE loading buffer and 2 or 8 ul of each sample was electrophoresed through a 10% SDS-PAG. =  104  IV Secretion of the Exg by E. coli  (A) Functional analysis of the Exg signal sequence  The identification of the amino terminus of the mature Exg was based on the correspondence of the sequence of the first 30 amino acids of the major C. fimi Exg with the nucleotide sequence of the cloned cex. gene. Although the translational start codon h8d not been identified experimentally, an ATG codon 41 codons 5' to the first residue of the mature Exg appeared as the most likely site. A putative signal sequence of 41 amino acids was predicted from the nucleotide sequence. I wished to confirm experimentally the limits of this signal sequence. Vectors and transformation protocols were not available for C. fimi. This precluded the genetic analysis of the signal sequence in this organism. However, the cex product is translocated into the E. coli periplasmic space (Gilkes et al., 1984a). This indicated that the signal sequence is functional in E. coli allowing, its analysis in this organism. In this chapter I describe experiments which showed that no more than 37 amino acids of the putative signal sequence were required to direct secretion of a hybrid protein in E. coli.  (1) Construction of hybrid J3Gal-Exg signal sequences I prepared two deletion mutants which lacked the first four or all of the amino acids of the Exg signal sequence. Since these deletions resulted in the elimination of the necessary translational regulatory signals, gene fusions were constructed in which the entire coding sequence of the mature protein and varying lengths of the signal sequence were fused to the lacZpo. the RBS sequence, the translation initiation site, and the sequence encoding the first 6 amino acids of the amino terminus of BGal. It W8S assumed that the ability of the signal peptide to function and to direct the localization of the Exg would be a function of both its length and its amino acid sequence. The gene fusions specified hybrid proteins that were comprised of 6 amino terminal amino acids of BGal linked to 37 or none of the carboxyl-terminal amino acids of the putative signal  105  sequence followed by the complete coding sequence of the mature Exg. Plasmld pUC 12 was the source of the lac operon components (Vlelra and Messing, 1982). The plasmld constructions and the numbering of amino acids in the Exg amino acid sequence are described in the legend to Fig. 25. The plasmid pUC 12-1.1 cex( 748) retained the entire coding sequence of the mature Exg and all but the first four codons of the signal sequence, the latter being replaced by the first six amino acids offiGal.The plasmid pUC 12-1.1 cex( 859) specified a hybrid protein consisting of the first 6 amino 8c1ds offiGalfused to the first codon of the mature, processed Exg sequence. For comparison, the original 6.6-kb BamHI fragment from pEC-1 (Gilkes etal., 1984a) was transferred to pUCI 2 (Vlelra and Messing, 1982). This plasmid, termed pUC12-1cex, had the entire cex gene oriented downstream from the lacZpo of pUC 12 and expressed Exg.  (2) Cellular localization offiGal-Exghybrid proteins Cultures of E. coli PM191 containing pUCI 2- Icex, pUC12-1.1cex( 48), or 7  pUC12-1.1cex(859)were harvested In the mid-exponential phase of growth. The cytoplasmic and periplasmic fractions were prepared by cold osmotic shock (Nossal and Heppel, 1966), and assayed for pNPCase activity. For comparison,fiGal,a cytoplasmic enzyme, andfi-lactamase,a periplasmic enzyme, were also measured. Exg activities of PM 191 cells containing plasmids pUC12- 1cgx> and pUC12-1.1cex(743) appeared largely In the periplasmic space (Table X). In contrast, Exg activity of cells containing pUCI 2-1. 1cex(859) was predominantly in the cytoplasm. This latter fusion is at the +1 codon of the mature, processed Exg sequence. It was obvious that the 37 amino acids of the signal sequence present in the hybrid protein specified by pUC 12-1.1 pex( 748) allowed its secretion in E. coli • The amino acids contributed to the hybrid protein byfiGalhad no detectable effect. The signal sequence of the Exg could not be less than 37 amino acids since there are no in-frame initiation codons between codons -37 and the +1 codon of the mature protein.  106  (3) Conclusions At 41 amino acids In length, the Exg signal peptide Is not unusually long for a prokaryotic signal sequence (Watson, 1984). Gram-posttlve signal sequences are generally longer than those found 1n Gram-negative bacterial proteins (Lampen et al., 1984; Murphy et al., 1984). At least five extracellular proteins from Bacillus and StreDtomvces species are synthesized with signal sequences of 41 to 48 amino acids (Lampen et al., 1984; Robblns et al., 1984; Yang et al., 1983); more complex "prepro" sequences of 75 to 221 amino acids have been reported for proteins of Bacillus and Staphylococcus species (Kovacevlc etal., 1985; Vasantha etal., 1984; Wells etal., 1983). The signal peptides of exported prokaryotic proteins show only small degrees of amino acid homology; however, they do possess similarities. These features Include: (1) an amino terminus consisting of positively charged amino acids, (11) a hydrophobic region directly following the positively charged amino terminus, and (111) an amino acid with a short side chain at the final position of the signal sequence (I nouye etal., 1980; Silhavy etal., 1982; Watson, 1984). The Exg signal peptide shows similar hydrophilic and hydrophobic regions (O'Neill et al., 1986a). The amino terminal end of the Exg signal sequence contains 7 Arg residues, comparable to the 5 Arg present In the amino terminus of the endo-B-N-acetylglucosamlnldase signal sequence from Streptomvcetes Dllcatus (Robblns et al., 1984). The hydrophilic Exg amino terminus is followed by a stretch of 16 amino acids which contains 5 Ala, 4 Val, and 2 Leu, yielding a highly hydrophobic peptide (Kyte and Doollttle, 1982). The processing of several Gram-positive secreted proteins Involves a "prepro" pattern with cleavage of the signal sequence occurring at two or more sites (Kovacevlc et al., 1985; Lampen etal., 1984; Ohmura etal., 1984; Robbins etal., 1984; Yasantha etal., 1984; Wells et al., 1983). It is possible that the Exg signal peptide also undergoes multiple proteolytic cleavages to yield the mature form of the enzyme.  107  It Is Interesting that despite its unusual length, the hybrid J3Gal-Exg signal sequence directed protein export In E. coli. In other experiments (not shown) hybrid signal sequences containing 15 amino acids from the amino terminal of BGal fused to the second amino acid of the Exg signal sequence still directed export of most of the Exg to the periplasm. This is In contrast with previous observations showing that minor alterations, such as point mutations of charged amino acids In the signal sequence, often reduce protein export dramatically (Silhavy et al., 1982).  108  Figure 25. Scheme for the construction of BGal-Exg expression plasmids pUC 12-1.1 cex_( 748) and pUC 12-1. 1pex.( 859). In these two plasmids the coding region for the pex.gene and either 37 codons or no codons of Its signal sequence have been Inserted Into pUC 12 such that (A) the Exg coding sequences were in the correct translational reading frame with the amino-terminus of BGal. The amino acid numbering starts with the first amino acid of the mature Exg as +1. The carboxy- and amino-terminal amino acids of the signal sequence are numbered -1 and -41, respectively. The fusions were made at an Aval I site located at codons -39 and -38 and an EcoRI I site at codons -3, -2, and -1. BGal-Exg fusions were assembled by inserting a DNA fragment coding for the amino-terminus of Exg and a fragment encoding the carboxy-termlnus of the Exg into pUC 12. The Exg fragments were prepared from pUC 12-1.1 pex.. The pex. gene was excised from pUC 12-1.1 cex as a Xhoi-Hindl 11 fragment. This preparation was divided into three aliquots. One aliquot (B) was partially digested with Sau3AI and the products were separated on a low melting temperature agarose gel. A 1633-bp Sau3AI-H1ndlll fragment encoding the carboxy-termlnus of the Exg was recovered. The amino-terminal coding end of this fragment was ligated to either (C) an EcoRIl-Sau3AI pex amino-terminal fragment (codons -1 to +31) or (D) an Ava.ll-Sau.3AI pex 8mino-terminal fragment (codons -37 to +31). The correct translational reading frames were constructed as follows. The second and third aliquots of the purified Xhol -Hjndl 11 cex DNA fragment were digested to completion with either EroRI I (C)or Avail (D). The EcoRI I digested fragments were treated with mung bean nuclease to remove the 5 nucleotide overhang of the EcoRI I site. The Ayal I digested fragments were incubated with dGTP and the Pollk (Klenow) to fill in one nucleotide of the Aval I site. The remaining 2-nucleotlde overhang was removed by treatment with mung bean nuclease. Both the EcoRI I and Aval I -treated fragments were digested to completion with Sau3AI. From these treatments two fragments were Isolated: a 206-bp bluntended Sau3AI fragment spanning the codons -37 to +31 of the pex.gene and 8 95-bp blunt-ended Sau3AI fragment encoding amino 8Cld -1 to + 31 of the cex gene. Fusion of either the "Aval I" or the "EcoRI I" fragments to the filled-in EcoRI site of pUC 12 maintained the correct translational reading frame in the BGal-Exg gene fusion and also recreated the EcoRI site. The various fragments were ligated into pUC 12 as shown. E.coliPM191 dra drm thr leu thi lacY recA56 supE (Meacock and Cohen, 1980) was transformed with the ligation mixture and Individual transformants were screened for gNPCase activity. For both p l 8 s m i d constructions, three pNPCase- positive transformants were selected for restriction analysis. The nucleotide sequence of the fusion regions were confirmed by Sanger's dideoxy sequencing procedure (Sanger et al., 1977) using the reverse sequencing primer (5'-AACAGCTATGACCATG-3') purchased from New England Biolabs, Beverly, MA. The nucleotide sequence and the deduced amino acid sequence of the amino-terminus of the BGal-Exg fusion products are shown. Restriction sites: B (BamHI): Hd (HindiII); Ps (Pst I); S (Sail); Sm (Smal); Ss (Sstl); X (Xbol); ' , (Sau3AI)~f , (EcoRII): f . (Avail). M.C.S., multiple cloning site; N, N-terminusof the mature Exg; single lines, vector DNA; open boxes, 5' and 3* untranslated C. fimi DNA; hatched boxes, signal sequence coding region of the Exg; filled-in boxes, coding region of the mature Exg gene. The direction of transcription of genes is indicated by arrows.  108a  t/Sl/Sm/B  1) X h o l , H i n d m d i g K I 2) isolate 2 3 k b Xhol-Hind m fragment  „.T  ..1 T  Ti  Tt 'TTT TTTTl T,T  ATO  1) E c o R I  3)  I) S a u 3 a l  2) Klenon + dGTP  G ACC ACG G TGC...  p a r t i a l digeet  CC ACG GG T G C nuclease "cC ACG-I98-GCG Sau3al GG T G C - b p - C G C T A G thr... (-37) M l )  3) nwng bean 4)  3) m i . productl of A , B , D 6)  frogment  C C A G GCC GCG G G C G C ... C C G C G ... ' nuclease OS C8C...  G A C C A C G .. GG T G C .  C S c G C - - b p -C G- " ...ala 8  CI)  7  G  TAG  l+M  4) n i l producti ol A , B , C *  «-  5) T 4 D N A l i g a w  T 4 D N A ligoM  Hd/P»/S  ATO  A C C ATO A T T ACQ A AT EcoRI  6 Mr TCC  (-J71 ( - 5 8 Thr P r o ACG CCC.  T  ACC  A T O A T T ACG  A AT EcoRI  6 Mr TCC  TCC  TGC T T A A ... A T T A C G A A T T '• ... T A A T G C T T A A  Hindi!  2 ) isolato l . e k b S o u 3 a - H 3  i) A v o n  A AT  ... A T T A C G  ... T A A 2) Klenow + d T T P , dATP  Ti T i .  A T T ACO  a  (-11 M l Alo Thr OCO ACC .  109  Table X. Localization of Exg, B-lactam.8se, BGal, and the hybrid BGal-Exg in E. coli- The cellular locations of BOal,fl-lactamase,Exg, and BGal-Exg were determined as follows. Cells harboring an appropriate plasmid were grown by inoculating 0.05 ml of a fresh overnight culture into 50 ml of LB, supplemented with 0.58 glycerol and 50 ugAp/ml. At a Klett reading of 60 (blue filter), IPTG was added to a final concentration of 1 mM to induce BGal. The cultures were grown to a final Klett of 180, harvested by centrifugation, washed, and resuspended in 100 mM potassium phosphate buffer, pH 7.0. These cell suspensions were divided into two and one aliquot of each was used for the fractionation of the cytoplasmic and periplasmic compartments by a cold osmotic shock procedure as described previously (Nossal and Heppel, 1966). In this procedure, the cells were collected by centrifugation, exposed to 0.5 M sucrose-1 X 10" M EDTA, 8  4  sedimented and then rapidly dispersed in cold 5 X 10" M MgCl2. The material released in the shock fluid was termed the periplasmic fraction. The cytoplasmic fraction refers to the soluble material obtained from the osmotically shocked cells after breaking the spheroplasts by passage through a French pressure cell at 10,000 psl. The cell extract fractions were prepared by passing the remaining aliquot of the initial cell suspensions through a French pressure cell at 10,000 psl. Cell debris and unbroken cells were removed from the extracts by centrlfugatlon at 50,000 x g for 1 h. p_NPCase activity was determined as previously described (Gilkes et al., 1984b). B-Lactamase activity was assayed spectrophotometrlcally using nitrocefin (O'Callaghan et al., 1972). BGal assays were performed according to Miller (1972). Units of activity: Exg, nmoles p_-n1trophenol released/mln/ml culture;fi-lactamase,nmoles nltrocefolc acid released/min/ml culture x 10^; BGal, nmoles o-nitrophenol released/min/ml culture x 10-5. Numbers in parentheses indicate the percent distribution of activity between the cytoplasmic and periplasmic compartments. To monitor the loss of enzyme activity and protein in the osmotic shock procedure a percent recovery value was calculated. Percent recovery is defined as the sum of the cytoplasmic and periplasmic activities divided by the activity found in the total cell extract x 100. 4  D  109a  TABLE X. Localization of Exg, B-lactamase, BGal, and the hybrid BGal-Exg In E. con.  Activity in nmoles products per min per ml of culture  3  Enzyme  In periplasmic  In cytoplasmic  In cell  Percent  Host [plasmid]  assayed  fraction  fraction  extract  recovery  PM191 [pUC12-l£Sxl  Exg  8.9  89  B-lactamase BGal  PM191 [pUC12-1.l£fix.(748)l Exg B-lactamase BGal  PM191 IpUC12-1.l£§x(859)] Exg B-lactamase BGal  6.8  (77)  1.1  (12.5)  (74)  17.0  (5.8)  3.8  (10)  28.1  (73)  38.3  83  16.0  (80)  3.0  (15)  20.1  95  87.0  (95)  3.7  (4.0)  91.0  99  5.8  (10)  42.5  (76)  56.1  86  0.14  (18)  0.62 (76)  116  (89)  3.2  (2.4)  3.9  (8.7)  35.0  (78)  216  294  0.82 131 45.0  80  94 91 87  15  110  .(B) Use of the OmpA signal peptide for secretion of Exg by E. coli As discussed above, it was deduced on the basis of nucleotide sequence analysis that the precursor of the C.  limi Exg contained an amino-terminal signal peptide of 41 amino acids. This  Gram-positive signal sequence directed the secretion of the Exg to the periplasmic space of E. coli (see Chapter IV(A)). Recent results, obtained following the completion of this project, show that E. coli signal peptidase cleaves the signal peptide from the precursor Exg at the same cleavage site recognized by theC. fimi signal peptidase (N. Arfman, unpublished observations). The localization of the recombinant Exg to the periplasmic space of E. coli is advantageous for the following reasons. First, many foreign proteins expressed in E. coli are degraded rapidly when cytoplasmically located (It8kura etal., 1977 ;Derynck etal., 1984), but the stability of foreign proteins is increased up to 10-fold if they are periplasmically located (Talmadge and Gilbert, 1982). Second, transport to the periplasmic space may be required for formation of disulfide bridges between the 6 cysteines of the Exg. For example, in E. coli disulfide bond formation In the periplasmic protein B-lactamase appears to take place when the protein is translocated from the reducing environment of the cytoplasm to the more oxidizing environment of the periplasm (Pollitt and Zalkin, 1983). Third, the purification of the recombinant Exg from osmotic shock fluids is aided by the fact that only 4* of the total cellular protein is contained in the periplasmic cell compartment (Nossal and Heppel, 1966). However, for the digestion of cellulose by the Exg it is necessary for the enzyme to be extracellular or to be located on the outer cell surface. Since extracellular secretion is normally prevented by the outer membrane of E. coli»localization of the Exg to the outer cell surface employing an appropriate signal peptide, such as the outer membrane protein A (OmpA) signal peptide, is 8 viable alternative. In addition, although the Exg signal peptide functions in E. coli • the Gram-positive structure of the Exg signal peptide, which is very different in length and amino acid content compared to E. coli signal peptides (Watson, 1984), may reduce the efficiency of its processing in E. coli . particularly at high levels of expression. For the reasons discussed above, I sought to replace the natural Exg signal peptide  Ill  with the signal peptide from the OmpA protein of E. coll (Ohraveb et al., 1984). Presumably, a hybrid OmpA-Exg secretory protein may be efficiently secreted, processed and localized to the outer membrane of E. coll. OmpA, the most abundant protein In E. coll. contains a 21 amino acid signal peptide (Movva et al., 1980). Its localization in the outer membrane of E. coli requires a signal peptide and possibly other Information (Freundl et al., 1985). The following experiments describe the construction of plasmld expression vectors encoding a hybrid precursor 1n which codons for the E, coli ompA signal peptide and the exocellular form of the cex. were precisely fused.  (1) Construction of plasmids expressing hybrid OmpA-Exg The plNlllomoA secretion vector, constructed by Inouye and his coworkers (Ohrayeb et al., 1984), was chosen for this work. This plasmld contains: (1) the entire OmpA signal sequence Including Its signal peptidase cleavage site, start codon and ribosome binding site; (11) a multiple cloning site located one codon 3' to the codons specifying the omoA signal peptidase cleavage site; (iil) the lac and lpj. promoters from the lactose and lipoprotein operons of E. coJi positioned in tandem 5' to the ompA translational start site; (Iv) the lac operator region to provide transcriptional regulation of the lgc and Ipj) promoters; and (v) the laci gene which encodes a repressor protein. The objective was to insert a DNA fragment encoding only the mature peptide of the Exg into the pINI I IpjmpA plasmid such that the omoA signal peptide precisely replaced the Exg signal peptide. The practical problem was the lack of restriction endonuclease sites in the codons of the mature amino-terminus of the cex. gene for the excision of a DNA fragment encoding only the mature Exg. Thus, a method was used to generate a blunt end accurately at the first codon in the DNA sequence of the mature Exg. Since the fusions involved protein coding sequences it was necessary to maintain the correct translational reading frame. Extraneous DNA sequences derived from linker DNA sequences or even single nucleotide insertions were not acceptable in the gene fusions. Since there were no suitable restriction endonuclease cleavage  112  sites with which to perform the fusions, I decided to use the approach of Goeddel et al. (1980). In their procedure, a synthetic deoxyollgorlbonucleotlde was used to precisely target the endpolnt of a DNA fragment. However, this procedure requires (1) the purification of a substantial quantity of a double-stranded restriction fragment template, (11) the use of the low 3' to 5" exonucleolytlc activity of the Klenow fragment of DNA polymerase I, and (111) the monitoring of the reaction steps using radloactlvely labeled fragments. Because of these requirements, I sought to Improve the efficiency of the Ooeddel et al. (1980) procedure. The basic biochemical approach I used, outlined In Fig. 26, was to use a synthetic ollgodeoxyrlbonucleotlde annealed to a precise locus on a single-stranded template as a primer for second strand synthesis. The source of template DNA was the ss DNA form of the M13 cloning phage vector developed by Messing (1983) containing a DNA Insert encoding the cex gene. The primed M13 phage DNA was then used as a template for the synthesis of a complementary DNA using the Klenow fragment of DNA polymerase I (Zoller and Smith, 1983). Mung bean nuclease (Laskowskl, 1980) was then used to remove single-stranded regions remaining after elongation and to linearize the double-stranded open circular DNA. Following treatment of the DNA with mung bean nuclease the double-stranded Insert DNA was released from the vector DNA sequences by digestion with a restriction enzyme that recognized a unique site In the DNA 3' to the primer. The generated fragment was then used for In vitro recombinant DNA manipulations. The method required that the cex. DNA first be transferred to an M13 vector (Messing, 1983) such that the antisense strand of the pex insert was incorporated into the plus (packaged) strand of the viral vector. For these studies a 1.96-kb ds BamHI-Sall DNA insert, encoding the entire cex gene, was excised from pUC12-1.1pex(737) and was transferred to Ml 3m pi 1 to y1eldM13mp11pex(737). The direction of transcription of the cex oene was from BamHI to Sail. A primer of 30 nucleotides was annealed to the pex antisense strand on Ml 3m pi lpj9<(737) ata point 130 nucleotides from the BamHI site. The 30-mer primer (5-CX^CCAC6CT(>WGeA60CCecceAC66f^-3^) corresponded to the first 10 amino acids of the mature processed amino terminus of theC. fimi exoglucanase (Flo. 8; M. Langsford, unpublished  113  observations). The Klenow fragment of E. cpJi DNA polymerase I was then used to synthesize a complementary DNA copy of the Insert DNA (Zoller and Smith, 1983). A nicked or gapped ds open circular DNA molecule was produced and was associated with high molecular weight DNA (Fig. 27, lane 4). In the absence of DNA llgase the newly synthesized strand will not be covalently closed (Zoller and Smith, 1983). Two potential problems which may reduce the recovery of the desired product at this step Included (1) the contamination of the Klenow preparation with 5" to 3' exonuclease activity, and (11) strand displacement by the polymerase (Zoller and Smith, 1983; Smith, 1985). The first problem was avoided by using a preparation of the Klenow enzyme free of contaminating 5' to 3" exonuclease activity. The second consideration concerned the potential of the Klenow enzyme to displace the 5' end of the hybridized, elongated oligonucleotide (Smith, 1985), resulting In a heterogeneous population of fragment ends. In my experience this was not a problem. Filamentous phage do contain sequences which block the progression of E. coli DNA polymerase I (Sherman and Gefter, 1976; Smith, 1985). Thus, the Klenow fragment after synthesizing the complementary strand of the Insert could be blocked by sequences on the M13 template and the polymerase would not reach the 5' end of the primer. The synthesis reaction mixture, after phenol extraction and ethanol precipitation of the DNA, was next treated with mung bean nuclease. This enzyme has a high specificity for ss DNA and cleaves ss DNA endonucleolytically to form 5'-phosphate-termini (Kroeker and Kowalskl, 1978; Laskowski, 1980). Mung bean nuclease treatment resulted in the appearance of a lower molecular weight set of fragments (Fig. 27, lane 5). This result Indicated that the gapped circular ds DNA molecules were converted to blunt-ended linear ds DNA molecules by mung bean nuclease. A high molecular weight band of DNA, corresponding to open circular DNA, remained after treatment with mung bean nuclease. This enzyme will not cleave the DNA strand opposite a nick In a duplex and requires ass gap of at least 5 bp In ads DNA (Kroeker and Kowalskl, 1978; Laskowski, 1980). It was possible that the open circular DNAs partly represented a subpopulation of the M13 molecules which had a completely synthesized a second strand and were relatively resistant to mung bean nuclease, whereas the linearized M13 DNAs represented ds circles that were originally gapped and  114  were sensitive to mung bean nuclease. The elongation of the entire pex insert distal to the primer was confirmed by the results shown in Fig. 27, lanes 6 to8. Restriction enzyme sites in the resynthesized DNA were cleavable. Digestion of the M13mp 11 -cex reaction mixture with the Sal' restriction endonuclease, which cleaved at the distal end of the pex gene, and is represented by enzyme 'x' in Fig. 26, yielded a specific digestion product of 110O-bp (Fig. 27, lanes 6 to8). This fragment size was consistent with the length of the desired ds Insert sequence from the priming site to the Sail site in the carboxyl terminus of the pex gene. There is also a background smear of DNA, probably representing partial synthesis products. The ds DNA fragments, after liberation from the vector sequences, were llgated into the EcoRI and the Hindi 11 sites of plasmld pINI 11 ompA. It W8S unnecessary to Isolate the insert sequence from the phage vector DNA, since the M13mp phage do not contain antibiotic resistance genes while cells harboring pINIIIompA could be selected by their plasmld-encoded resistance to ampiclllln.  (2) Expression of OmpA-Exg hybrid proteins in E. poll E. poll PM 191 was transformed with the ligation mixture and individual transformants were screened for pNPC8se activity. Of the 66 transformants tested, 16 had pNPCase activity. Plasmld DNAs from four of these transformants were isolated by the rapid alkaline method (Birnboim and Doily, 1979). Restriction endonuclease analyses confirmed that the plasmid DNAs from these four pNPCase-positive transformants contained Inserts of about 110O-bp which had restriction maps identical to that of the coding sequences of the mature pex gene. The structure of the recombinant plasmid encoding the OmpA-Exg hybrid protein Is shown in Fig. 28. The level of Exg activity in four transfomants, designated plNHIompA-B8. B16, CI 2, and C24, was determined next. The Exg levels produced by plasmids B8 and B16 were 13.1 nmoles pNP released/min/mg protein and 6.2 nmoles pNP released/mln/mg protein for plasmids CI 2 and C24. Maximal Exg activity produced by these plasmids was obtained by addition of 2 mM IPT6 to  115  exponentially growing cultures to induce the tandem lpo_-lac promoters. Little or no activity was detected in the supernatant fluids of these cultures. To examine whether the OmpA-Exg product was secreted, a chloroform osmotic shock procedure was used (Ames etal., 1984). This procedure permeabllizes the outer membrane of E. poll and causes the selective release of periplasmic proteins. Exg activity was detected In the chloroform shockates of all four transformants (pINIIlpmpA-B8, B16, CI 2, and C24), Indicating that hybrid OmpA-Exg proteins were being secreted across the Inner membrane and that the secreted OmpA-Exg was biologically active. Exg activity could also be detected in the membrane fraction of cells disrupted by passage through a French pressure cell. SDS-PAOE analyses of total cell lysates of Induced and noninduced cells carrying pINIIIompA B8, B16, CI 2 and C24, did not reveal any new protein bands. The DNA sequence of the fusion region between the ompA signal peptide and coding sequences of the mature amino-terminus of the cex gene was determined. The nucleotide sequence analysis revealed two classes of gene fusions represented by B8 and B161n one class and C12 and C24 In a second class. In all cases the first 21 8mino acids of the omoA signal peptide and the omoA signal peptidase recognition site were present. However, In plasmids DlNlllompA-B8 and B16 the omoA signal peptide sequences were fused to the codon at position -16 of the cex gene signal peptide. In the plasmids pINIIIgrnpA-Cl 2 and C24 the ompA signal peptide sequences were fused to the codon at position +1 of the coding sequences of the mature Exg. It was interesting that the OmpA-Exg plasmids giving the highest Exg activity were plNlllgmpA-B8 and B16 which encoded a hybrid signal peptide containing the entire ompA signal peptide and 25 carboxyl-terminal amino acids of the cex. gene signal peptide.  (3) Conclusions The coding sequences of the mature pex gene were inserted into the E. coli secretion vector pINI I IgmpA by means of an M13-primer mutagenesis technique. Recombinant plasmids were isolated which had the OmpA signal peptide fused to the amino-terminus of the mature Exg.  116  The hybrid OmpA-Exg protein produced by these plasmids had Exg activity and was secreted across the inner membrane of E. coli. Whether the hybrid OmpA-Exg was localized to the outer membrane and whether the OmpA signal peptide was correctly processed from the OmpA-Exg were not pursued for the two following reasons. First, I had developed, concurrently with the OmpA-Exg secretion plasmids, two other Exg expression vectors, pUC 12-1.1 cex( PTIS)and M13mp 10cex. which yielded very high levels of Exg activity that was secreted to the periplasmic space and passively 'leaked' to the extracellular environment. The native cex signal peptide was utilized In both these vectors. Second, the processing of the signal peptide from the Exg encoded by the plasm Id pUCI 2-1.1cex( PTIS) was shown to be Identical In E. coll and C. fimi ( Arfman. 1986). Hence, using the cex signal peptide, I was able to obtain high levels of Exg production and exocellular accumulation In E. coll.  117  Figure 26. Protocol for the directed access to specific DNA sequences using oligonucleotide primers. The various steps are described In the text. The dashed lines represent M13ssDNA sequences; the solid lines are Insert sequences and the blocked regions are target DNA sequences of the insert sequences. The small open arrow indicates the position of a cleavage site for the restriction endonuclease 'x'. (+), sense strand; (-), anti-sense strand. The sawtooth line represents the synthetic priming oligodeoxyribonucleotide.  3'  )  I) heat denature 12) anneal primer  DNA poll (Klenow) + dNTPs  X  5'  <3>  I mung bean 1 nuclease restriction endonuclease V  cleavage  5' •5'  <  5'  118  Figure 27. Monitoring of the primer extension method by agarose gel electrophoresis. M13 mp 11 -cex DNA at various steps in the procedure was analyzed on a 1.1 % (wt/vol) agarose gel. After electrophoresis, the gel was stained with ethidium bromide, and the DNA fragments were visualized by UV-induced fluorescence. Lane 1, molecular weight markers [ Hindi 11 digest of the E. SJliphage lambda (Sanger et al., 1982) andaHMI digest of plasmld pUCI 3 (Yannlsch-Perron et al., 1985)]; lane 2, M13mpl I-cex ss DNA (0.2 pmoles); lane 3, oligonucleotide-primed M13mp11-a2LSsDNA(0.2pmole);lane4, ol1gonucleot1de-pr1medM13mp11-cex dsDNAafter second strand synthesis (0.2 pmoles); lane 5, mung bean nuclease-treateddsM13mp11-cex DNA (0.2 pmoles); lanes 6,7,8, mung bean nuclease-treated ds M13mp 11 -pjx DNA after Sail digestion with 0.2,0.4, and 0.8 pmoles DNA per l8ne, respectively.  1 2 3 4 5 6 7 8  119  Figure 28. Scheme for the construction of a plasmid vector In which the E. coll ompA signal peptide is fused to the mature amino-terminus of the Exg. The relevant structures of the plasmid pINIIIompA are shown. The direction of transcription for the lipoprotein promoter (lpp ) and the promoter-operator of the lactose operon (lac^ ) are denoted by heavy arrows. The DNA sequence of the cex gene which codes for the mature Exg (solid box) is joined to the DNA sequence of the signal peptide of the omoA gene (stippled box). The amino acids at the signal peptidase cleavage sites of the OmpA, Exg and the hybrid OmpA-Exg are shown. Numbering of amino acids starts at +1 for the first codon of the mature peptides and at -1 for the last carboxyl-terminal codon of the signal peptides. The synthetic deoxyoligoribonucleotide (30-mer; wavy line) is complementary to the coding strand of the Qgx. gene starting with the Thr codon at the second codon of the mature Exg. The JacL gene is represented by an open box. p  0  • ignol p t p t l d a t t ct«ovgg« site  -21 -I ^+1 MCS me! lys lyt 17oa ola. do osnser... I I ompA leader sequence .ompA oligonucleotide  120  LITERATURE CITED Adams, S.P., Kavka, K.S., Wykes, E.J., Holder, S.B., andGalluppi, G.R. (1983) Hindered dlalkylamino nucleoside phosphite reagents In the synthesis of 2 DNA 51 -mers. J. Amer. Chem. Soc. 105: 661-663. Adhya, S. and Gottesman, M. (1978) Control of transcription termination. Ann. Rev. Biochem. 47: 967-996. Ahmed, A. K., Schaffer, S. W., and D. B. Wetlaufer. 1975. Nonenzymlc reactivation of reduced bovine pancreatic rlbonuclease by 8ir oxidation and by glutathione oxidoreduction buffers. J. Biol. Chem. 2§0j 8477-8482. Ames, G.F.-L., Prody, C. and Kustu, S. (1984) Simple, rapid, and quantitative release of periplasmic proteins by chloroform. J. Bact. 160: 1181 -1183. Anba, J., Lazdunski, C. and Pages, J.-M. (1986) L8ck of effect of leader peptidase overproduction on the processing In vivo of exported proteins In Escherichia coli. J. Gen. Microbiol. 132: 689-696. Appleyard, R.K. (1954) Segregation of new lysogenic types during growth of a doubly lysogenic strain derived from Escherichia coll K-12. Genetics 39: 440-452. Arfman, N. (1986) Purification and characterization of two recombinant cellulases from E. coli and the generation of monoclonal antibodies 8gainst a native exoglucanase. M.Sc. Thesis, University of Gronlngen, Grdnlngen, The Netherlands. Armentrout, R.W. and Brown, R.D. (1981) Molecular cloning of genes for cellobiose utilization and their expression in Escherichia coli. Appl. Environ. Microbiol. 41:1355-1362. Atkinson, T., and Smith, M. (1984) Solid ph8se synthesis of oligodeoxyribonucleotides by the phosphite trlester method. In Oligonucleotide synthesis: A practical approach, Gait, M.J., ed., pp. 35-81, IRL Press, Washington, D.C.. Backman, K., and Boyer, H.W. (1983) Tetracycline resistance determined by pBR322 is mediated by one polypeptide. Gene 26:197-203. Ballou, CE. (1982) Yeast cell wall and cell surface. In Molecular Biology of the Yeast Saccharomvces. vol.2: Metabolism and Gene Expression. (Strathern, J., Jones, E.W. and Broach, J.R., eds.) pp. 335-360, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Barras, F., Boyer, M.H., Chambost, J.P., and Chippaux, M. (1984) Construction of a genomic library of Erwlnlachrvsanthemi and molecular cloning of cellulase genes. Mol. Gen. Genet. 197:513-514. Baty, D., Mercereau-Puijalon, 0., Perrin, D., Kourilsky, P. and Lazdunski, C. (1981) Secretion into the bacterial periplasmic space of chicken ovalbumin in Escherichia coli. Gene 16: 79-87.  121  Beguln, P. (1983) Detection of cellulase activity In polyacrylamlde gels using Congo Red stained agar replicas. Anal. Biochem. 131:333-336. Beguin, P., Cornet, P., and Aubert, J,-P. (1985) Sequence of a cellulase gene of the thermophilic bacterium Clostridium thermocellum. J. Bacterlol. 162: 102-105. Beguln, P., Cornet, P. and Millet, J. (1983) Identification of the endoglucanase of the celB.gene of Clostridium thermocellum. Biochimie 65: 495-500. Beguin, P., Gilkes, N.R., Kilburn, D.G., Miller Jr., R.C., O'Neill, G.P. and Warren, RAJ. (1986a) Cloning of cellulase genes. CRC Critical Rev. Biotechnology. (in press). Beguln, P., Rocancourt, M., Chebrou, M.-C. and Aubert, J.-P. (1986b) Mapping of mRNA encoding endoglucanase A from Clostridium thermocellum. Mol. Gen. Genet. 202:251 254. Belasco, J.G., Beatty, J.T., Adams, C.W., von Gabaln, A., and Cohen, S.N. (1985) Differential expression of photosynthesis genes in R. capsulata results from segmental differences in stability within the polycistronic rxcA transcript. Cell 40:171 -181. Bennetzen, J.L., and Hall, B.D. 1982. Codon selection In yeast. J. Biol. Chem. 257:3026-3031. Benson, S.A., Bremer, E. and Sllhavy, T.J. (1984) Intragenic egions required for LamB export. Proc. Natl. Acad. Sci. USA 81:3830-3834. Benson, S.A. and Silhavy, T.J. (1983) Information within the mature LamB protein necessary for localization to the outer membrane of £. cpJi K12. Cell 32: 1325-1335. Bibb, M. S., Findlay, P. R., and Johnson, M.W. 1984. The relationship between base composition and codon usage In bacterial genes and its use for the simple and reliable identification of protein-coding sequences. Gene 30:157-166. Blrnbolm, HI., and Doly, J. (1979) A rapid alkaline extraction procedure for screening of recombinant plasmid DNA. Nucl. Acids Res. 7: 1513-1523. Bisaria, V.S. and Ghose, T.K. (1981) Biodegradation of cellulosic materials: substrates, microorganisms, enzymes and products. Enzyme Mlcrob. Technol. 3:90-104. Blobel.G., and Dobberstein, B. (1975) Transfer of proteins across membranes. I. Presence of proteolytically processed and unprocessed nascent immunoglobulin light chains on membrane-bound ribosomes of murine myeloma. J. Cell. Biol. 67:835-851. Bolivar, F., Rodriguez, R.L., Betlach, M.C., Heyneker, H.L, Boyer, H.W., Crosa, J.H., and Falkow, S. (1977) Construction and characteri28tion of new cloning vehicles, II. A multipurpose cloning system. Gene 2: 95-113. Bollon, A.P. (1984) Recombinant DNA products: Insulin, interferon and growth hormone. CRC Press, Inc., Boc8Raton, Florida. Bonekamp, F., Andersen H.D., Christensen, T., and Jensen, K.F. (1985) Codon-defined ribosomal pausing In Escherichia coll detected by using the pvrE attenuator to probe the coupling between transcription and translation. Nucl. Acids Res. 13:4113-4123.  122 Boyer, H.W., Betlach, M., Bolivar, F., Rodriguez, R.L, Heyneker, H.L., Shine, J., and Goodman, H.M. (1977) The construction of molecular cloning vehicles, in Beers, R.F., and Basset, E.G. (eds.), Recombinant DNA molecules, Vol. 10, Miles Symposium Series. Raven Press, New York, pp. 9-19. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254. Breull, C. and Saddler, J.N. (1985) Comparison of the 3,5-dlnitrosalicyllc acid and NelsonSomogyi methods of assaying for reducing sugars and determining cellulase activity. Enzyme Microb. Technol. 7: 327-332. Broach, J.R., LI, Y.-Y., Wu, L.-C.C. and Jayaram, M. (1983) Vectors for high-level, inducible expression of cloned genes In yeast. In Experimental Manipulation of Gene Expression, (lnouye,M.,ed.),pp. 84-118, Academic Press, New York. Burnett, J.P. (1983) Commercial production of recombinant DNA-derlved products. In Experimental Manipulation of Gene Expression, (Inouye, M., ed.), pp. 259-277 , Academic Press, New York. Chanzy, H. and Henrlssat, B. (1985) Unidirectional degradation of Vajpnlacellulose microcrystals subjected to cellulase action. FEBS Lett. 184:285-288. Chanzy, H. and Henrlssat, B. and Vuong, R. (1984) Colloidal gold labelling of 1,4-B-D-glucan celloblohydrolase adsorbed on cellulose substrates. FEBS Lett. 172: 193-197. Collier, B., Thomas, D.Y., Teather, R.M., Erfle, J.D. and Crosby, W.L. (1984) Cloning and expression in Escherichia coli of cellulase genes from Bacteroides succinooenes. DNA 3: 184-188. Collmer, A., and Wilson, D.B. (1983) Cloning and expression of a Thermomonospora YX endocellulase gene In E. coli. Bio/Technology 1:594-601. Cornet, P., Millet, J., Beguin, P., and Aubert, J.-P. (1983a) Characterization of two eel (cellulose degradation) genes of Clostridium thermocellum coding for endoglucanases. Bio/Technology 1: 589-594. Cornet, P., Tronik, D., Millet, J. and Aubert, J.-P. (1983b) Cloning and expression in Escherichia coli of Clostridium thermocellum penes coding for amino acid synthesis and cellulose hydrolysis. FEMS Microbiol. Letters 16: 137-141. Coughlan, M.P. (1985a) Cellulose hydrolysis: The potentials, the problems and relevant research atGalway. Biochemical Soc. Trans. 13:405-406. Coughlan, M.P. (1985b) The properties of fungal and bacterial cellulases with comment on their production and application. Biotech. Genetic Engineering Rev. 3:39-109. Crosby, B., Collier, B., Thomas, D.Y., Teather, R.M. and Erfle, J.D. (1984) Cloning and expression in Escherichia coli of cellulase genes from Bacteroides succinooenes. In Proceedings of the Fifth Canadian Bioenergy R&D Seminar, Ottawa, Canada, March 26-28, 1984. (Hasnain, S. ed.), pp. 573-576, Elsevier Applied Science Publishers, New York. Cuatrecasas, P., Wilchek, M., and Anflnsen, C. B. (1968) Selective enzyme purification by affinity chromatography. Proc. Natl. Acad. Sci. USA. 61:636-641.  123  Davis, B.D., and Tai, P.C. (1980) The mechanism of protein secretion across membranes. Nature 283: 433-438. de Boer, H., Comstock, L.J. and Vasser, M. (1983a) The tec promoter: A functional hybrid derived from the trp_and lac promoters. Proc. Natl. Acad. Sci. USA 80: 21 -25. deBoer,H.A.,Hui,A., Comstock, L, J., Wong, E., and Vasser, M. (1983b) Portable ShineDalgarno regions- A system for a systematic study of defined alterations of nucleotidesequences within Escherichia coli ribosome binding sites. DNA 2:231 -235. de Boer, H.A. and Shepard, H.M. (1983) Strategies for optimizing foreign gene expression in Escherichia coll. Horizons Biochemistry Biophysics 7:205-248. Derynck, R., Roberts, A.B., Winkler, M.E.,Chen, E.Y. andOoeddel, D.Y. (1984) Human transforming growth factor-*: Precursor structure and expression in Escherichia coli. Cell 38: 287-294. Derynck, R., Hltzeman, R.A., Gray, P.W. and Goeddel, D.Y. (1983) Expression of human interferon-gamma in heterologous systems. In Experimental Manipulation of Gene Expression, (Inouye, M., ed.), pp. 247-258, Academic Press, New York. Detroy, R.W. and St. Julian, G. (1980) Blomass conversion: Fermentation, Chemicals and Fuels. CRCCritical Rev. Microbiol. 10: 203-228. Deshpande, M.V., Eriksson, K.-E. and Pettersson, L.G. (1984a) An assay for selective determination of exo-1,4-fi-glucanases In a mixture of cellulolytlc enzymes. Analyt. Biochem. 138: 481-487. Deshpande, V., Rao, M., Keskar, S. and Mishra, C. (1984b) Occurrence of a procellulase in the culture filtrates of Penlcllllum lanthlnellum. Enzyme Microbial Technol. 6: 371-374. Dreyer, G.B., and Dervan, P.B. (1985) Sequence-specific cleavage of single-stranded DNA: 01igodeoxynucleotide-EDTA-Fe(II) Proc. Natl. Acad. Sci. USA 82:968-972. Dubnau, D. (1983) Molecular cloning in Bacillus subtilis. In Experimental Manipulation of Gene Expression, (Inouye, M., ed.), pp. 33-53, Academic Press, New York. Enari.T.M. (1983) Microbial cellulases. In Microbial Enzymes and Biotechnology. (Fogarty, W.M., ed.) pp. 183-223, Applied Science Publishers, London. Enari, T.-M. and Markkanen, P. (1977) Production of cellulolytic enzymes by fungi. Adv. Biochem. Engineering 5: 3-24. Eriksson, K.E. (1978) Enzyme mechanisms Involved in cellulose hydrolysis by the white rot fungus Sporotrichum oulverulentum. Biotechnol. Bioengin. 20: 317-332. Fahnestock, S.R. and Fisher, K.E. (1986) Expression of the Staphylococcal protein A gene in Bacillus subtilis by gene fusions using the promoter from a Bacillus amvlollauefaclens alpha-amylasegene. J. Bact. 165: 796-804.  124  Fan, L.T., Lee, Y.-H. and Beardmore, D.H. (1980a) Major chemical and physical features of cellulosic materials 8S substrates for enzymatic hydrolysis. Adv. Biochem. Engineering 14: 101-118. Fan, L.T., Lee, Y.-H. and Beardmore, D.H. (1980b) Mechanism of the enzymatic hydrolysis of cellulose: Effects of the major structural features of cellulose on enzymatic hydrolysis. Biotechnology Bioenglneering 22: 177-199. Fllckinger, M.C. (1980) Current biological research in conversion of cellulosic carbohydrates into liquid fuels. How far have we come? Biotechnol. Bioengin. 22 (suppl. 1): 27-48. Freundl, R., Schwarz, H., Klose, M., Movva, N.R. and Henning, U. (1985) The nature of information, required for export and sorting, present within the outer membrane protein OmpA of Escherichia coli K-12. EMBOJ. 4:3593-3598. Ganoza, M.C., Sullivan, P., Cunningham, C., Hader, P., Kofoid, E.C. and Neilson, T. (1982) Effect of b8ses contiguous to AUG on translation initiation. J. Biol. Chem. 257:8228-8232. Gardner, K.H. and Blackwell, J. (1974) The structure of native cellulose. Biopolymers 13: 1975-2001. Garvey, J.S., Cremer, N.E. and Sussdorf, D.H. (1977) Methods in Immunology, third ed., pp. 218-219, W.A. Benjamin, Inc., Reading, Mass. Gentz, R., Langner, A., Chang, A.C.Y., Cohen, S.N. and Bujard, H. (1981) Cloning and analysis of strong promoters is mads possible by the downstream placement of a RNA termination signal. Proc. Natl. Acad. Sci. USA 78: 4936-4940. Gheysen, D., Iserentant, D., Derom, C. and Flers, W. (1982) Systematic alteration of the nucleotide sequence preceding the translational Initiation codon and the effects on bacterial expression of the cloned SV40 small-1 antigen gene. Gene 17:55-63. Ghose, T.K. (1977) Cellulose biosynthesis and hydrolysis of cellulosic substrates. Adv. Biochem. Engineering 6:39-76. Ghose, T.K., Bailey, H.J., Bisaria, V.S., En8ri, T.M., et al. (1983) Measurement of cellulase activities. Final recommendations. Commission on Biotechnology, International Union of Pure and Applied Chemistry. Ghrayeb, J., Kimura, H., Takahara, M., Hsiung, H., M8sui, Y., and Inouye, M. (1984) Secretion cloning vectors In Escherichia coli. EMBOJ. 3: 2437-2442. Gilkes, N.R., Kilburn, D.G., Langsford, Ml., Miller, Jr., R.C., Wakarchuk, W.W., Warren, R.A.J., Whittle, D.J. and Wong, W.K.R. (1984a) Isolation and characterization of Escherichia coli clones expressing cellulase genes from Cellulomonas fimi. J. Gen. Microbiol. 130: 1377-1384. Gilkes, N.R., Langsford, M.L., Kilburn, D.G., Miller, Jr., R.C. and Warren, R.A.J. (1984b) Mode of action and substrate specificities of cellulases from cloned bacterial genes. J. Biol. Chem. 259:10455-10459. Gilkes, N.R., Kilburn, D.G., Miller, Jr., R.C. and Warren, R.A.J. (1984c) A mutant of Escherichia coli that leaks cellulase activity encoded by cloned cellulase genes from Cellulomonas fimi. Bio/Technology 2: 259-263.  125  Goeddel, D.V., Shepard, H.M., Yelverton, E., Leung, D., and Crea, R. (1980) Synthesis of human fibroblast Interferon by £. coli. Nucleic Acids Res. 8: 4057-4074. Gold, L., Prlbnow, D., Schneider, T., Shinedllng, S., Swebelius-Singer, B., and Stormo, 6. (198 J) Translational initiation in prokaryotes. Ann. Rev. Microbiol. 35: 365-403. Gong, C.S. and Tsao, G.T. (1979) Cellulase and biosynthesis regulation. Ann. Reports on Fermentation Processes 3:111 — 139. Grantham, R., Gautler, C., Gouy, M., Jaconzone, M., and Mercier, R. (1981) Codon catalog usage is a genome strategy modulated for gene expressivity. Nucl. Acids Res. 9: r43-r74. Gray, G.L., Baldridge, J.S., McKeown, K.S., Heyneker, H.L. and Chang, C.N. (1985) Periplasmic production of correctly processed human growth hormone In Escherichia coll: natural and bacterial signal sequences are interchangeable. Gene 39:247-254. Greenberg, N.M., Warren, R.A.J., Kilburn, D.G. 8nd Miller Jr., R.C. (1986) Regulation, size and start site determination of the cenA and cex transcripts of Cellulomonas fimi. (manuscript submitted) Grepinet, 0. and Beguin, P. (1986) Sequence of the cellulase gene of Clostridium thermocellum coding for endoglucanase B. Nucl. Acids Res. 14: 1791 -1799. Grosjean, H., and Flers, W. (1982) Preferential codon usage In prokaryotic genes: the optimal codon-anticodon interaction energy and the selective codon usage in efficiently expressed genes. Gene 18: 199-209. Hager, P.W. and Rablnowltz, J.C. (J 985) Inefficient translation of T7 late mRNA by Bacillus subtilis ribosomes-Implications for species-specific translation. J. Biol. Chem. 260: 15163-15167. Hall, M.N., Gabay, J., Debarboullle, M. and Schwartz, M. (1982) A role for mRNA secondary structure in the control of translation initiation. Nature 295: 616-618. Halliwell, G. (1961) The action of cellulolytic enzymes from M. verracaria. Biochemical J. 79: 185-192. Hendy, N.A., Wllke, C.R. and Blanch, H.W. (1984) Enhanced cellulase production In fed-batch culture of Trichoderma reesei C30. Enzyme Microbial Technol. 6: 73-77. Higgins, C.F., Ames, G.F.-L., Barnes, W.M., Clement, J.M. and Hofnung, M. (1982) A novel Interclstronlc regulatory element of prokaryotic operons. Nature 298: 760-762. Hopp, T.P. and Woods, K.R. (1981) Prediction of protein antigenic determinants from amino acid sequences. Proc. Natl Acad. Sci. USA. 78: 3824-3828. Hui, A.H., Hayflick, J., Dinkelspiel, K. and de Boer, H.A. (1984) Mutagenesis of the three bases preceding the start codon of the rj-galactosidase mRNA and Its effect on translation in Escherichia coli. EMBO J. 3: 623-629. Ikemura, T. (1981) Correlation between the abundance of Escherichia coli transfer RNAs and the occurence of the respective codons In its protein genes. J. Mol. Biol. 146: 1-21.  126  Inouye, M. and Halegoua, S. (1980) Secretion and membrane localization of proteins in Escherichia coli .CRCCrit. Rev. Biochem. 7: 339-371. Itakura, K., Hirose, T., Crea, R., Riggs, A.D., Heyneker, HI., Bolivar, F., and Boyer, H.W. (1977) Expression in Escherichia coli of a chemically synthesized gene for the hormone somatostatin. Science 198: 1056-1063. Jensen, K. F., Larsen, J. N. Schack, L., and Sivertsen, A. (1984) Studies on the structure and expression of Escherichia coli pyrC. pvrD. and pvrF using the cloned genes. Eur. J. Biochem. 140:343-352. Johnson, JA, Wong, W.K.R. and Beatty, J.T. (1986) Construction of plasmid expression vectors and their use for expression of cellulase genes In Rhodopseudomonas capsulata. J. Bact. (In press) Johnson, P.H., and Laskowski, Sr., M. (1970) Mung bean nuclease I- 11. Resistance of double stranded deoxyribonucleic acid and susceptibility of regions rich In adenosine and thymidine to enzymatic hydrolysis). J. Biol. Chem. 245: 891 -898. Jollff, G., Beguin, P., Juy, M., Millet, J., Ryter, A., Poljak, R., and Aubert, J.-P. (1986a) Isolation, crystallization and properties of a new cellulase of Clostridium thermocellum overproduced in Escherichia coll. (submitted for publication) Joliff, G., Beguin, P., Millet, J., Aubert, J.-P., Alzari, P., Juy, M. and Poljak, R.J. (1986b) Crystallization and preliminary X-ray diffraction study of an endoglucanase from Clostridium thermocellum. J. Mol. Biol, (in press). Kagawa, Y., Nojima, H., Nakiwa, N., Ishizaka, M., Nakajima,T., Yasuhara, T., Tanaka,T. and Oshlma, T. (1984) High guanoslne plus cytoslne content In the third letter of codons of an extreme thermophile: DNA sequence of the isopropylmalate dehydrogenase of Thermus thermoohilus. J. Biol. Chem. 259: 2956-2960. Kastelein, R.A., Berkhout, B. and van Duin, J. (1983) Opening the closed ribosome-binding site of the lysis cistron of bacteriophage MS2. Nature 305: 741 -743. Kingsman, S.M., Kingsman, A.J., Dobson, M.J., Mellor, J. and Roberts, N.A. (1985) Heterologous gene expression In Saccharomvces cerevislae. Biotech. Genetic Engineering Rev. 3:377416. Kohchi, C. and Toh-e, A. (1985) Nucleotide sequence of Candida pelliculosa Nucl. Acids Res. 13: 6273-6282.  fi-Qlucosidase  gene.  Kolde, Y., Kakamura, A., Uozumi, T., and Beppu, T. (1986) Molecular cloning of a cellulase gene from Bacillus subtilis and its expression in Escherichia coli. Agric. Biol. Chem. 50: 233237. Kolter, R. and Yanofsky, C. (1982) Attenuation in amino acid biosynthetic operons. Annu. Rev. Genet. 16: 113-134. Kotoujansky, A., Diolez, A., Boccara, M., Bertheau, Y., Andro, T., and Coleno, A. (1985) Molecular cloning of Erwlniachrvsanthemi oectinase and cellulase structural genes. EMBOJ. 4: 781-785.  127  Kovacevlc, S., Veal, L.E., Hsung, H.M. and Miller, J.R. (1985) Secretion of Staphylococcal nuclease by Bacillus subtilis. J. Bacteriol. 162: 521 -528. Kozak, M. (1983) Comparison of initiation of protein synthesis in procaryotes, eucaryotes and organelles. Microbiological Rev. 47: 1 -45. Kroeker, W.D. and Kowalskl, D. (1978) Gene-sized pieces produced by digestion of linear duplex DNA with mung-bean nuclease. Biochemistry 17:3236-3243. Kyte, J. and Doolittle, R.F. (1982) A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157: 105-132. Ladlsch, M.R., Lin, K.W., Voloch, M. andTsao, G.T. (1983) Process considerations in the enzymatic hydrolysis of biomass. Enzyme Microb. Technol. 5: 82-102. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. Lamed, R., Setter, E. and Bayer, E.A. (1983) Characterization of a cellulose-binding, cellulasecontaining complex in Clostridium thermocellum. J. Bacteriol. 156:828-836. Lamed, R., Setter, E., Kenig, R. and Bayer, E.A. (1983) The cellulosome- A discrete cell surface organelle of Clostridium thermocellum which exhibits separate antigenic cellulosebinding and various cellulolytic activities. Biotechnol. Bioeng. Symp. 13: 163-181. Lampen, J.O., Wang, W., Mezes, P.S.F. and Yang, Y.-J. (1984)fi-Lactamasesof Bacilli: Nature and processing, in Hoch, J. and Ganesan, A.T. (Eds.), Genetics and Biotechnology of B8c11H, Academic Press, New York , pp. 129-140. Langsford, Ml., Gilkes, N.R., Wakarchuk, W.W., Kilburn, D.G., Miller, Jr., R.C. and Warren, RAJ. (1984) The cellulase system of Cellulomonas fimi. J. Gen. Microbiol. 130: 1367-1376. Laskowski, M. Sr. (1980) Purification and properties of the mung-bean nuclease.Meth. Enzymol. 65:263-276. Llnko, M., Ratto, M., Ylikarl, L. and Bailey, M. (1982) Organisms and enzymes for the hydrolysis of cellulose and xylan. In Proceedings of the International Symposium of Ethanol from Biomass. (Duckworth, H.E. and Thompson, E.A.,eds.) pp. 371-393, The Royal Society of Canad8, Winnipeg. Ljungdahl, L., Pettersson, B., Eriksson, K.E. and Wiegel, J. (1983) A yellow affinity substance Involved In the cellulolytic system of Clostridium thermocellum. Current Microbiology 9: 195-200. Looman, A.C., de Qruyter, M., Voogelaar, A., and van Knippenberg, P.H. (1985) Effects of heterologous rlbosomal binding sites on the transcription and translation of the lacZ gene of Escherichia coli. Gene 37: 145-154. Mackay, R.M., Baird, S., Dove, M.J., Erratt, JA.Gines, M., Moranelli, F., N8sim, A., Willick, G.E., Yaguchi, M. and Seligy, VI. (1985) Glucan8se gene diversity in prokaryotic and eukaryotic organisms. Biosystems 18: 279-292. Mandels, M. (1975) Microbial sources of cellulase. Biotechnol. Bioengin. Symp. 5: 81 -105.  128  M8ndels, M. (1982) Cellulases. p. 35-78. in G. T. Tsao (ed.), Annua) reports on fermentation processes, vol. 5. Academic Press Inc., New York. Mandels, M. (1985) Applications of cellulases. Biochemical Society Transactions 13: 414-416. Maniatis, T., Frltsch, E. F., and Sambrook, J. (1982) Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Marston, F. A. 0., Lowe, P. A., Doel, M. T., Shoemaker, J. M., White, S., and Angal, S. (1984) Purification of calf prochymosln (prorennin) synthesized in Escherichia coll. Bio/Technology 2: 800-804. Masui, Y., Coleman, J. and Inouye, M. (1983) Multipurpose expression cloning vehicles in Escherichia coll. In Experimental manipulation of Gene Expression, (Inouye, M., ed), pp. 15-32, Academic Press, New York. Maxam, A.M., and Gilbert, W. (1980) Sequencing end-labelled DNA with base-specific chemical cleavages, Methods In Enzymology 65:499-560. McCarthy, J.E.G., Sebald, W., Gross, G. and Lammers, R. (1986) Enhancement of translational efficiency by the Escherichia coli atpE translational initiation region: its fusion with two human genes. Gene 41:201-206. McLaughlin, J.R., Murrey, C.L., and Rabinowltz, J.C. (1981) Unique features In the rlbosome binding site sequence of the gram-positive Staphylococcus aureus J3-1actam8se gene. J. Biol. Chem. 256: 11283-11291. Meacock, PA and Cohen, S.N. (1980) Partitioning of bacterial plasmids during cell division: a cis-acting locus that accomplishes stable plasmid inheritance. Cell 20: 529-542. Mendez, B.S. and Gomez, R.F. (1982) Isolation of Clostridium thermocellum auxotrophs. Appl. Environ. Microbiol. 43:495-496. Merlvuorl, H., Slegler, K.M. Sands, J A and Montenecourt, B.S. (1985) Regulation of cellulase biosynthesis and secretion in fungi. Biochemical Sec. Trans. 13: 411 -414. Messing, J. (1983) New M13 vectors for cloning. Methods in Enzymology 101C: 20-78. Michaelis, S. and Beckwith, J. (1982) Mechanisms of incorporation of cell envelope proteins in Escherichia coli. Ann. Rev. Microbiol. 36:435-465. Miller, G.L. (1959) Use of dlnitrosallcyllc acid reagent for determination of reducing sugar. Analyt. Chem. 31: 426-428. Miller, J.H. (1972) Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Millet, J., Petre, D., Beguln, P., Raynaud, 0., and Aubert, J.-P. (1985) Cloning of ten distinct DNA fragments of Clostridium thermocellum coding for cellulases. FEMS Microbiol. Lett. 29: 145-149.  129  Mlyake, Y., Oka, T., Nlshlzawa, T., Misoka, F., Fuwa, T., Yoda, K., Yamasakl, M. and Tamurs, 6. (1985) Secretion of human interferon-alpha induced by using secretion vectors containing a promoter and signal sequence of alkaline phosphatase gene of Escherichia coll . J. Biochem. 97: 1429-1436. Moran, CP. Jr., Lang, N., LeGrice, S.F.J., Lee, 0., Stephens, M., Sonenschein, A.L., Pero, J. and Loslck, R. (1982) Nucleotide sequences that signal the Initiation of transcription and translation in Bacillus subtilis. Mol. Gen. Genet. 186:339-346. Mott, J. E., Grant, R. A., Ho, Y. -S., and Piatt, T. (1985) Maximizing gene expression from plasmld vectors containing the lambda P|_ promoter: Strategies for overproducing transcription termination factor Rho. Proc. Natl. Acad. Sci. USA. 82: 88-92. Movva, N.R., Nakamura, K. and Inouye, M. (1980) Amino acid sequence of the signal peptide of ompA protein, a major outer membrane protein of Escherichia coli. J. Biol. Chem. 255: 27-29. Mullings, R. (1985) Measurement of saccharification by cellulases. Enzyme Microbial Technol. 7: 586-591. Murphy, N., McConnell, D.J. and Cantwell, B.A. (1984) The DNA sequence of the gene and genetic control sites for the excreted B_. subtil lis enzymefl-glucanase.Nucl. Acids Res. 12: 5355-5367. Murphy-Hoi land, K. and Eveleigh, D.E. (1985) Secretion, activity and stability of deglucosylated cellulase of Trichoderma reesei. In Abstracts of the Annual Meeting of the American Society for Microbiology, K-130. Murray, CL., and Rabinowltz, J.C. (1982) Nucleotide sequences of transcription and translation initiation regions in Bacillus phage 029 early genes. J. Biol. Chem. 257: 1053-1062. Nagahari, K., Kanaya, S., Munakata, K., Aoyagi, Y. and Mizushima, S. (1985) Secretion into the culture medium of a foreign gene product from Escherichia coli: use of the ompF gene for secretion of humanfi-endorphin.EMBOJ. 4: 3589-3592. Nelson, N. (1952) A photometric adaptation of the Somogyi method for the determination of glucose. J. Biol. Chem. 153: 376-380. Nichols, B.P. and Yanofsky, C. (1983) Plasmids containing the trp. promoters of Escherichia coli and Serratia marcescens and their use in expressing cloned genes. Methods in Enzymol. 101: 155-164. Nielsen, J.B.K., and Lampen, J.O. (1982) Membrane-bound penicillinase in gram-positive bacteria. J. Biol. Chem. 257: 4490-4495. Nossal, N.G. and Heppel, L.A. (1966) The release of enzymes by osmotic shock from Escherichia coli in exponential phase. J. Biol. Chem. 241: 3055-3062. O'Callaghan, CH., Morris, A., Kirby, S.M. and Shingler, A.H. (1972) Novel method for detection of G-lactamase by using a chromogenic cephalosporin substrate. Antimicrob. Ag. Chemother. 1: 283-288.  130  Ohmura, K., Nakamura, K., Yamazakl, H., Shlroza, T., Yamane, K., Jlgaml, Y., Tanak8, H., Yoda, K., Yamasaki, M. and Tamura, G. (1984) Length and structural effect of signal peptides derived from Bacillus subtilis aloha-amylase on secretion of Escherichia coli 13lactamase in B. subtilis cells. Nucl. Acids Res. 12: 5307-5319. Oka, Y., Sakamoto, S., Miyoshi, K.-l., Fuwa, T., Yoda, K., Yamasaki, M., Tamura, G. and liiyake, T. (1985) Synthesis and secretion of human epidermal growth factor by Escherichia coll. Proc. Natl. Acad. Sci. USA 82: 7212-7216. O'Neill, G., Goh, S.H., Warren, R.A.J., Kilburn, D.G. and Miller Jr., R.C. (1986a) Structure of the gene encoding the exoglucanase of Cellulomonas fimi. Gene 44:325-330. O'Neill, G., Warren, R.A.J., Kilburn, D.G. and Miller Jr., R.C. (1986b) Secretion of a Cellulomonas fimi exoglucanase by Escherichia coli. Gene 44:331-336. O'Neill, G., Warren, RAJ., Kilburn, D.G. and Miller Jr., R.C. (1986c) Overproduction of a cellulase gene with a high guanoslne plus cytoslne content in Escherichia coll. Appl. Environ. Microbiol. 52:000-000. Pages, J.-M., Anba, J., Bernadac, A., Shrgagawa, H., Nakata, A. and Lazdunski, C. (1984) Normal precursors of periplasmic proteins accumulated In the cytoplasm are not exported posttranslationally in Escherichia coli. Eur. J. Biochem. 143: 499-505. Palva, I., Lehtov8ara, P., Kaariainen, L., Sibakov, M.,Contell, K., Schein, C.H., K8shw8gi, K. 8nd Welssmann, C. (1983) Secetlon of Interferon by Bacillus subtilis. Gene 22: 229-235. P8lva, E.P. and Silhavy, T.J. (1984) ]acl fusions to genes that specify exported proteins: A general technique. Mol. Gen. Genet. 194:388-394. Penttila, M.E., Neval8inen, K.M.H., Raynal, A. andKnowles, J.K.C. (1984) Cloning of Aspergillus nloer genes in yeast. Expression of the gene coding Aspergillus 13-olucosldase. Mol. Gen. Genet. 194: 494-499. Petre, D., Beguin, P., Millet, J. and Aubert, J.-P. (1984) Heterologous hybridization of bacterial DNA to the endoglucanases A and B structural genes gelA. and ceJB_ of Clostridium thermocellum Annal. Microbiol. (Inst. Pasteur) 136B: 113-124. Pollitt, S. and Zalkin, H. (1983) Role of primary structure and disulfide bond formation in Blactamase secretion. J. Bacteriol. 153:27-32. Pugsley, A. P., and Schwartz, M. (1985) Export and secretion of proteins by bacteria. FEMS Microbiol. Rev. 32: 3-38. Queen, C. (1983) A vector that uses phage signals for efficient synthesis of proteins in Escherichia coli. J. Mol. Appl. Genetics 2: 1-10. Randall, L.L. and Hardy, S.J.S. (1984) Export of protein in bacteria. Microbiological Rev. 48: 290-298. Ray, P.N. and Pearson, Ml. (1974) Evidence for the post-transcript ional control of the morphogenetic genes of bacteriophage lambda. J.Mol. Biol. 85: 163-175.  131  Ray, P.N. and Pearson, M.L. (1975) Functional inactivation of bacteriophage X morphogenetic gene mRNA. Nature 253:647-650. Raynal, A. and Guerineau, M. (1984) Cloning and expression of the structural gene for Bglucosidase of Kluweromvces fraailis in Escherichia coli and Saccharomvces cerevisiae. Mol. Gen. Genet. 195:108-115. Rees, D.A., Morris, E.R., Thorn, D. and Madden, J.K. (1982) Shapes and Interactions of carbohydrate chains. In Polysaccharides, vol. 1, pp. 195-290, Academic Press, New York. Reese, E.T. (1977) Degradation of polymeric carbohydrates by microbial enzymes. Recent Adv. Phytochemistry 11:311-367. Reese, E.T., Mandels, M. and Weiss, A.H. (1972) Cellulose as a novel energy source. Adv. Biochem. Engineering 2: 181 -200. Reese, E.T., Slu, R.G.H. and Levfnson, H.S. (1950) Biological degradation of soluble cellulose derivatives. J. B8cteriol. 59: 485-497.' Remaut, E., Stanssens, P. and Fiers, W. (1981) Pl8smid vectors for the high-efficiency expression controlled by the P|_ promoter of collphage lambda. Gene 15:81 -93. Remaut, E., Tsao, H., and Fiers, W. (1983) Improved plasmid vectors with a thermoinducible expression and temperature-regulated runaway replication. Gene 22: 103-113. Robbins, P.W.,Trimble, R.B., Wirth, D.F.,Hering,C, Maley, F., Maley,G.F.,Das.,R., Gibson, B.W., Royal, N., and Biemann, K. (1984) Primary structure of the StreDtomvces enzyme endo-B-N-acetylglucosaminidaseH. J. Biol. Chem. 259: 7577-7583. Roberts, D.P., Allen, C, and Lacy, G.H. (1985) Cloning Erwinea carotovarasubsp. carotovara DNA fragments mediating the production of cellulase, protease, and pectolytic enzymes. Phytopathol. 75: 628. Robinson, M., Lllley, R., Little, S., Emtage, J. S., Yarranton.G., Stephens, P., Millican, A., Eaton, M., and Humphreys, G. (1984) Codon usage can affect efficiency of translation in Escherichia coli. Nucl. Acids Res. 12: 6663-6671. Robson, L.M., and Chambllss, G.H. (1986a) Cloning of the Bacillus subtilis DLG B-1,4-glucanase gene and its expression in Escherichia coli and B. subtilis. J. Bact. 165: 612-619. Robson, L.M., and Chambliss, G.H. (1986b) Complete nucleotide sequence of the B-1,4-glucanase gene of Bacillus subtilis. In Abstracts of the Annual Meeting of the American Society for Microbiology, H-5. R., Dargatz, H., and Hollenberg, CP. (1985) Precursor of B-lactamase is enzymatically inactive: accumulation of the preprotein in Saccharomyces cerevisiae. J. Biol. Chem. 260: 1508-1512.  Roggenk8mp,  Roggenkamp, R., Kusterman-Kahn, B. and Hollenberg, CP. (1981) Expression and processing of bacterial B-lactamase in the yeast Saccharomvces cerevisiae. Proc. Natl. Acad. Sci. USA 78: 4466-4470.  132  Rosenberg, FA and Oberkotter, L.Y. (1977) Methods for determining microbial biodegradation In cellulose. In Biodeterloration Investigation Techniques (Walters, A.H., ed.) pp. 243-259. Applied Science Publishers, Barking, Essex, U.K.. Rosenberg, M., and Court, D. (1979) Regulatory sequences involved In the promotion and termination of RNA transcription. Annu. Rev. Genet. 13:319-353. Rosenberg, M., Ho, Y.S., and Shatzman, A. (1983) The use of pKC30 and its derivatives for controlled expression of genes. Methods In Enzymology 101C: 123-137. Rossi J.J., Soberon, X., Marumoto, Y., McMahon, J. and Itakura, K. (1983) Biological expression of an Escherichia coli consensus sequence promoter and some mutant derivatives. Proc. Natl. Acad. Sci. USA 80: 3203-3207. Russell, D. R., and Bennett, G.N. (1982) Construction and analysis of in vivo activity of E. cojl promoter hybrids and promoter mutants that alter the -35 to -10 spacing. Gene 20: 231-243. Ruther, U., Koenen, M., Otto, K., and Muller-Hlll, B. (1981) pUR222, a vector for cloning and rapid chemical sequencing of DNA. Nucl. Acids Res. 9: 4087-4098. Ryu, D.D. and Mandels, M. (1980) Cellulases: biosynthesis and applications. Enzyme Microb. Technol. 2:91-102. Sacco, M., Millet, J. and Aubert, J.-P. (1984) Cloning and expression in Saccharomvces cerevisi8e of a cellulase gene from Clostridium thermocellum. Ann. Microbiol. (Inst. Pasteur) 135A: 485-488. Sanger, F., Coulson, A. R., Hong, G. F., Hill, D. F., Petersen, G. B. (1982) Nucleotide sequence of bacteriophage lambda. J. Mol. Biol. 162: 729-773. Sanger, F., Nicklen, S., and Coulson, A.R. (1977) DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74: 5463-5467. Sashlhara, N., Kudo, T., and Horlkoshl, K. (1984) Molecular cloning and expression of cellulase genes of alkalophilic Bacillus sp. strain N-4 in Escherichia coli. J. Bact. 158:503-506. Schoner, R. G., Ellis, L. F., and Schoner, B.E. (1985) Isolation and purification of protein granules from Escherichia coli cells overproducing bovine growth hormone. Bio/Technology 3: 151-154. Schoner, B.E., Hsiung, H.M., Belagaje, R.M., Mayne, N.G. and Schoner, R.G. (1984) Role of translational efficiency in bovine growth hormone expression in Escherichia coli. Proc. Natl. Acad. Sci. USA 81: 5403-5407. Schwarz, W., Bronnenmeier, K., and Staudenbauer, W.L. (1985) Molecular cloning of Clostridium thermocenum oenes Involved In B-glucan degradation In bacteriophage lambda. Biotechnology Lett. 7: 859-864. Shatzman, A., Ho, Y.-S. and Rosenberg, M. (1983) Use of phage lambda regulatory signals to obtain efficient expression of genes in Escherichia coll. In Experimental manipulation of Gene Expression, (Inouye, M., ed.), pp. 2-14, Academic Press, New York.  133  Sherman, L.A., and Getter, M.L. (1976) Studies on the mechanism of enzymatic DNA elongation by Escherichia coli DNA polymerase II. J. Mol. Biol. 103: 61-76. Shine, J., and Dalgarno, L. (1974) 3'-Terminal sequence of Escherichia coli 16S ribosom8l RNA. Complementarity to nonsense triplets and ribosome binding sites. Proc. Natl. Acad. Sci. USA. 71: 1342-1346. Shiroza, T., Nakazawa, K., Tashiro, N., Yamane, K., Y a n 8 g i , K., Yamasaki, Y., Tamura, 6., Saito, H., Kawade, Y. and Taniguchl, Y. (1985) Synthesis and secretion of biologically active mouse interferon-J3 using a Bacillus subtilis alpha-amvlase secretion vector. Gene 34: 1 -8. Shoemaker,S.P.,Gelfand, D.H., Innis,MA, Kwok,S.Y.,Ladner, M.B. andSchweickart, V. (1984) Recombinant fungal celloblohydrolases. European patent application EP 137,280 European Patent Office. Shoemaker, S., Schweickart, V., Ladner, M., Gelfand, D., Kwok, S., Myambo, K., and Innis, M. (1983) Molecular cloning of exo-celloblohydrolase I derived from Trlchoderma reesei strain L27. Bio/Technology 1:691-696. Silhavy, T.J., Benson, SA 8nd Emr, S.D. (1982) Mechanism of protein localization. Microbiol. Rev. 47:313-344. Singh, A., Lugovoy, J.M., Kohr, W.J. and Perry, L.J. (1984) Synthesis, secretion and processing of alpha-factor-interferon fusion proteins in yeast. Nucl. Acids Res. 12:8927-8938. Skipper, N., Bozzato, R.P., Vetter, D., Davles, R.W. and Hopper, J.E. (1986) Use of the mellbiase promoter and signal peptide to express a bacterial cellulase In yeast. (in press) Skipper, N., Sutherland, M., Davles, R.W., Kilburn, D., Miller Jr., R.C, Warren, A. and Wong, R. (1985) Secretion of a bacterial cellulase by yeast. Science 230:958-960. Smith, H.O. and Blrnstlel, M.L. (1976) A simple method for DNA restriction site mapping. Nucl. Acids Res. 3: 2387-2398. Smith, M. (1985) In vitro mutagenesis. Ann. Rev. Genet. 19: 423-462. Spindler, K.R., Rosser, D.S., and Berk, A.J. (1984) Analysis of adenovirus transforming proteins from early regions 1A and 1B with antisera to inducible fusion antigens produced in Escherichia coli. J. Virol. 49: 132-141. Stanssens, P., Remaut, E. and Fiers, W. (1985) Alterations upstream from the Shine-Dalgarno region and their effect on bacterial expression. Gene 36: 211 -223. Stanssens, P., Remaut, E. and Fiers, W. (1986) Inefficient translation Initiation causes premature transcription termination in the lacZ gene. Cell 44:711 -718. Stern, M.J.,Ames,G.F.-L, Smith,N.H., Robinson, E.C,andHiggins,CF. (1984) Repetitive extragenic palindromic sequences: A major component of the bacterial genome. Cell 37: 1015-1026. Stuber, D. and Bujard, H. (1981) Organization of transcriptional signals in plasmids pBR322 and pACYC 184. Proc. Natl. Acad. Sci. USA 78: 167-171.  134  Talmadge, K. and Gilbert, W. (1982) Cellular location affects protein stability in Escherichia coli • Proc. Natl. Acad. Sci. USA 79: 1830-1833. Talmadge, K., Kaufman, J. and Gilbert, W. (1980) Bacteria mature preproinsulin to proinsulin. Proc. Natl. Acad. Sci. USA 77: 3988-3992. Teather, R.M. and Wood, P.J. (1982) Use of Congo red-polysaccharide Interactions in enumeration and characterization of cellulolytic bacteria from bovine rumen. Appl. Environ. Microbiol. 43:777-780. Teerl, T., Salovourl, I., and Knowles, J. (1983) The molecular cloning of the major cellulase gene from Trichoderma reesei. Bio/Technology 1: 697-699. Thompson, C.J. and Gray, G.S. (1983) Nucleotide sequence of a streptomycete aminoglycoside phosphotransferase gene and its relationship to phosphotransferases encoded by resistance plasmids. Proc. Natl. Acad. Sci. USA 80: 5190-5194. van Gijsegem, F., Toussaint, A. and Schoonejans, E. (1985) In vivo cloning of the pectate lyase and cellulase genes of Erwinea chrvsanthemi. EMBO J. 4: 787-792. van Tllbeurgh, H., Clayssens, M. and de Bruyne, C.K. (1982) The use of 4-methylumbelllferyl and other chomophoric glycosides in the study of cellulolytic enzymes. FEBS Lett. 149: 152-156. Varenne, S., Buc, J., Lloubes, R., and Lazdunskl, C. (1984) Translation Is a non-uniform process: Effect of tRNA availability on the rate of elongation of nascent polypeptide chains. J. Mol. Biol. 180: 549-576. Vasantha, N. and Thompson, L.D. (1986) Secretion of a heterologous protein from Bacillus subtilis with the aid of protease signal sequences. J. Bact. 165: 837-842. Vasantha, N., Thompson, L.D., Rhodes, C, Banner, C, Nagle, J. and Filpula, D. (1984) Genes for alkaline protease and neutral protease from Bacillus amvloliouefaclens contain a large open reading frame between the regions coding for signal sequence and mature protein. J. Bacteriol. 159: 811-819. Vleira, J. and Messing, J. (1982) The pUC plasmids, an M13mp7- derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19: 259-268. Wakarchuk, W.W., Kilburn, D.G., Miller, Jr., R.C. and Warren, RAJ. (1986) Molecular  cloning of an Aarobacterium celloblase In Escherichia coll. Mol. Gen. Genet, (in press).  Wakarchuk, W.W., Kilburn, D.G., Miller, Jr., R.C. and Warren, RAJ. (1984) The preliminary characterization offl-glucosidasesof Cellulomonas fimi. J. Gen. Microbiol. 130: 13851389. Warren, R.A.J., Beck, C.F., Gilkes, N.R., Kilburn, D.G., Langsford, M.L, Miller Jr., R.C, O'Neill, G.P. and Wong, W.K.R. (1986) Sequence conservation and domain shuffling in two cellulases from Cellulomonas fimi. In: Cellucon '86 (in press) Watson, M.E.E. (1984) Compilation of published signal sequences. Nucl. Acids Res. 12:51455164.  135 Wells, J A , Ferrari, E., Henner, D.J., Estell, DA and Chen, E.Y. (1983) Cloning, sequencelng, and secretion of Bacillus amvloliauefaciens cx-amylase in Bacillus subtilis. Nucl. Acids. Res.11:7911-7925. Whitehorn, E.A., Livak, K.J. and Petteway Jr., S.R. (1985) The effects of hybrid ribosomebindlng-site variants on the expression of human Interferon-J3 in Escherichia coli. Gene 36: 375-379. Whittle, D.J., Kilburn, D.G., Warren, RAJ. and Miller, Jr., R.C. (1982) Molecular cloning of a Cellulomonas fimi cellulase oene In Escherichia coll. Gene 17: 139-145. Wledmann, M., Huth, A. and Rapoport, TA (1984) Xenopus oocytes can secrete bacterial betalactamase. Nature (London) 309:637-639. Winkler, M. E., Blaber, M., Bennett,G. L, Holmes, W.,8nd Vehar.GA (1985) Purification and characterization of recombinant urokinase from Escherichia coll. Bio/Technology 3: 990-1000. Wong, W.K.R., Gerhard, B., Gou.Z.M., Kilburn, D.G., Warren, RAJ. and Miller, Jr., R.C. (1986) Characterization and structure of an endoglucanase gene cenA of Cellulomonas fimi-Gene 44: 315-324. Wood, T.M. (1985a) Properties of cellulolytic enzyme systems. Biochemical Society Transactions 13: 407-410. Wood, T.M. (1985b) Aspects of the biochemistry of cellulose degradation, In Cellulose and Its derivatives, chemistry, biochemistry and applications. (Kennedy, J.F., Phillips, G.O., Wedlock, D.J. and Williams, PA, eds.), pp. 173,E111s Harwood Ltd. Publishers, a Division of J. Wiley 8nd Sons, New York. Wood, T.M. and McCrae, S.I. (1979) Synergism between enzymes involved in the solubilization of native cellulose. (Brown, R.D. and Jurasek, L. eds.) Adv. Chemistry 181:181-209. Yaguchi, M., Roy, C, Rollins, C.F., Paice, M.G. and Jurasek, L. (1983) A fungal cellulase shows sequence homology with the active site of hen egg-white lysozyme. Biochem. Biophys. Res. Comm. 116:408-411. Yamada, K. (1977) Recent advances In industrial fermentation in Japan. Biotechnol. Bioengineering 19: 1563-1621. Yang, M., Galizzi, A. and Henner, D. (1983) Nucleotide sequence of the amylase gene from Bacillus subtilis. Nucl. Acids Res. 11: 237-249. Yannlsch-Perron, C, Ylelra, J., and Messing, J. (1985) Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp and pUC vectors. Gene 33: 103-119. Zoller, M.J., and Smith, M. (1983) Oligonucleotide-directed mutagenesis of DNA fragments cloned in M13-derlved vectors, in Methods in Enzymology (Wu, R., Grossman, L., and Moldave, K., eds.), Vol. 100, Part B, pp.468-500, Academic Press, New York.  PUBLICATIONS: O'Neill, 6., Goh, S.H., Warren, RAJ., Kilburn, D.G. and Miller Jr., R.C. (1986) Structure of the Gene Encoding the Exoglucanase of Cellulomonas fimi. Gene 44: 325-330.  O'Neill, G., Warren, R.A.J., Kilburn, D.G. and Miller Jr., R.C. (1986) Secretion of Cellulomonas fimi Exoglucanase bv Escherichia coll. Gene 44: 331-336.  O'Neill, G., Kilburn, D.G., Warren, RAJ. and Miller Jr., R.C. (1986) Overproduction from a Cellulase Gene with a High Guanoslne-Plus-Cytosine Content in Escherichia coli. Applied Environ. Microbiol. 52: 737-743.  Beguin, P., Gilkes, N.R., Kilburn, D.G., Miller, Jr., R.C, O'Neill, G. and Warren, R.A.J.. Cloning of Cellulase Genes. CRC Critical Reviews Biotechnology (in press).  Gilkes, N.R., Langsford, Ml., O'Neill, G., Wakarchuk, W.W., Whittle, D.J., Gerhard, B., Raffle, V., Kilburn, D.G., Miller, Jr., R.C. and Warren, RAJ. (1983) The Cellulase Complex of Cellulomonas fimi and the Cloning of Its Genes in Escherichia coll.. In Proceedings: Biotechnology In the Pulp and Paper Industry, Pira Press, pp. 240-252.  Warren, RAJ., Beck, C.F., Gilkes, N.R., Kilburn, D.G., Langsford, Ml., Miller, Jr., R.C, O'Neill, G.P. and Wong, W.K.R. (1986) Unusual Structural Organization of Two Cellulases from Cellulomonas fimi. Proceedings: Cellucon '86 (in press).  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items