UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

CBD-cellulose : a novel adjuvant system Tekant, Bahar 1991

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

Item Metadata

Download

Media
831-ubc_1992_spring_tekant_bahar.pdf [ 5.53MB ]
Metadata
JSON: 831-1.0086481.json
JSON-LD: 831-1.0086481-ld.json
RDF/XML (Pretty): 831-1.0086481-rdf.xml
RDF/JSON: 831-1.0086481-rdf.json
Turtle: 831-1.0086481-turtle.txt
N-Triples: 831-1.0086481-rdf-ntriples.txt
Original Record: 831-1.0086481-source.json
Full Text
831-1.0086481-fulltext.txt
Citation
831-1.0086481.ris

Full Text

CBD-Cellulose :A Novel Adjuvant SystemByBAHAR TEKANTM.D., Ankara University, 1987A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Microbiology)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAOctober 1991© Bahar Tekant, 1991In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature) Department of ^1--► ICROZ InLor-- ---/The University of British ColumbiaVancouver, CanadaDate^24}`'/101991DE-6 (2/88)iiABSTRACTGlycoprotein D gene of Herpes Simplex Virus Type 1 (gD.1) was clonedin Escherichia coli as a gene fusion to the endoglucanase A cellulosebinding domain (CBD) gene of Cellulomonas fimi. The initial constructproduced a form of CBDcenA-gD.1 fusion protein which severely inhibitedthe growth of E. coli. The inhibitory effect was attributed to the presenceof the transmembrane region of the gD.1 protein. A cassette plasmid wasconstructed to express CBD cenA fused in frame to any other gene. Atruncated gD.1 gene lacking the coding regions for the leader peptide andthe transmembrane/cytoplasmic domains was inserted into this cassette, andthe synthesis of 45.9 kDa fusion protein was shown. The fusion protein wasseverely degraded, and also, associated with the cell membranes.CBDcenA-PhoA, a chimeric protein containing alkaline phosphatase,was used to investigate the potential adjuvant effect of immunization with aCBD-fusion protein bound to cellulose. Immunogens tested includedalkaline phosphatase alone, CBD cenA-PhoA alone, alkaline phosphatase inAvicel (micro crystalline cellulose) solution, CBD cenA -Pho A-Alum(aluminum hydroxide) and CBD cenA-PhoA-Avicel. Mice immunized withCBDcenA-PhoA-Avicel showed the highest anti-alkaline phosphatase andanti-CenA antibody titers. Animals immunized with alkaline phosphatase-Avicel, CBDcenA-PhoA-Alum, CBD cenA-PhoA or alkaline phosphatasealone showed mean secondary anti-alkaline phosphatase antibody titers thatwere 29%, 26%, 17% and 17% of the CBDcenAPhoA-Avicel titers,respectively. Animals immunized with CBD cenA-PhoA-Alum or CBD cenA-PhoA alone showed mean secondary anti-CenA antibody titers that were57% and 42% of the CBDcenA-PhoA-Avicel titers, respectively. The resultsindicate that the use of CBD fusions to present an antigen bound toIIIcellulose provides an effective adjuvant for antibody responses.ivTABLE OF CONTENTSAbstract ^iiTable of contents ^ivList of tables ^viiList of figures^  viiiAbbreviations, nomenclature and symbols ^xiAcknowledgements  xiiiIntroduction ^1Materials and methods ^8I^Bacterial strains, phages and plasmids ^8II^Media ^8III Buffers ^8IV Growth and induction of bacteria^ 10V^DNA maipulations and recombinant DNA techniques^ 10(A) Enzymes, reagents and techniques^ 10(B) Preparation and purification of plasmid andphage DNA^  11(C) Synthetic oligodeoxyribonucleotides ^11(D) Primer annealing reaction and Klenow extention^12(E) Mung bean nuclease reaction^ 13(F) Ligations ^13(G) Preparation of the linker ^14(H) Site directed mutagenesis ^14VI Electrophoresis of DNA^  15VII DNA sequence determination  16VVIII Preparation of cell extracts ^16IX Electrophoresis of proteins ^17X^Western blotting ^17XI Avicel binding assay ^18XII Preparation of immunogens ^19XIII Mice immunization ^20XIV ELISA^  21Results ^22I^Construction of plasmid pUC18-CBD.PT-gD.1^ 22(A) Subcloning of the gD.1 gene into M13mp18 ^22(B) Preparation of the gD.1 gene fragment ^22(C) Preparation of the CBD.PT gene fragment ^25(D) Construction of pUC18-CBD.PT-gD.1^ 28II^Construction of plasmid pUC12-CBD.PT-gD.1^ 29(A) Construction of pUC18-CBD.PT^ 29(B) Subcloning of the CBD.PT gene into M13mp18^33(C) Construction of M13mp18-CBD.PT.IEGR-site directed mutagenesis ^33(D) Construction of M13mp18-CBD.PT-gD.1^ 38(E) Expression of CBD.PT-gD.1 by E. coli JM101(M13mp 18-CBD.PT-gD.1) cells ^41(F) Subcloning of the CBD.PT-gD.1 gene intopUC12^  43III Expression of CBD.PT-gD.1 by E. coli(pUC12-CBD.PT-gD.1) cells ^46(A) Colony blots^  46v i^(B) Western blots ^46^IV Immunization results ^62(A) Experiment 1 ^621. Immunization ^622. ELISA results ^65(B) Experiment 2 ^671. Immunogens ^672. Immunization ^733. ELISA results ^734. Statistical analysis ^80Discussion ^89References ^94viiLIST OF TABLES1 Bacterial strains, phage and plasmids^ 92. Antibody responses in mice immunized with fusionprotein bound to Avicel or controls (Exp.1)^ 663. Primary response of mice immunized with fusionprotein bound to Avicel or controls (Exp.2)^ 74Secondary response of mice immunized with fusionprotein bound to Avicel or controls (Exp.2)^ 745. ANOVA table - primary anti-alkaline phosphatase titers^ 836. ANOVA table - primary anti-CenA titers^ 867. ANOVA table - secondary anti-alkaline phosphatase titers^ 878. ANOVA table - secondary anti-CenA titers^ 88viiiLIST OF FIGURES1. Diagram of CenA ^22. Diagram of CBDcenA-PhoA ^23. Diagram of gD ^44. Scheme for the construction of pUC18-CBD.PT cenicgD.1^ 235. Agarose gel electrophoresis of Ml3mp18-gD.1digested with KpnI and Ball^  246. Monitoring the primer extention method used toprepare the gD.1 fragment, by agarose gelelectrophoresis^  267. Agarose gel electrophoresis showing the preparationof the CBD.PTcenit fragment using the primerextention method^  278. Scheme for the construction of pUC18-CBD.PT„nit^ 309. Agarose gel electrophoresis of pUC18-CBD.PT„ nAdigested with EcoRI^  3110. Agarose gel electrophoresis of Ml3mp18-CBD.PT„nAdigested with SstI  3211. Scheme for the construction of M13mp18-CBD.PT„1A .IEGR^ 3412. Monitoring the primer extention method used for sitedirected mutagenesis, by agarose gel electrophoresis^ 3613. Agarose gel electrophoresis ofMl3mp18-CBD.PT„ nA .IEGR digested with StuI^ 3714. Scheme for the construction of M13mp18-CBD.PT,,,,, A-gD /^ 3915. Agarose gel electrophoresis showing the preparationof the gD.1 fragment using the primer extentionmethod^  40ix16. Agarose gel electrophoresis ofM13mp18-CBD.PTeenA-gD.1 digested with HindIIIor EcoRI^  4217. SDS-PAGE analysis of CBD.PT cenA-gD.1 fusionprotein produced by M13mp18-CBD.PT cenit -gD.1^ 4418. Screening for pUC12-CBD.PT.-gD.1 by colonyblotting^  4719. SDS-PAGE analysis of CBD.PT cenA-gD.1 fusionprotein produced by pUC12-CBD.PT.A-gD.1,in E. coli JM101^  4920. SDS-PAGE analysis of CBD.PT cenA-gD.1 fusionprotein produced by pUC12-CBD.PTcenit-gD.I,in E. coli JM101^  5121. SDS-PAGE analysis of CBD.PT cenA-gD.1 fusionprotein produced by pUC12-CBD.PTcem -gD.1,in E. coli JM101^  5222. SDS-PAGE analysis of CBD.PTcenA-gD.1 fusionprotein produced by pUC12-CBD.PT.4-gD.1,in E. coli JM101^  5423. SDS-PAGE analysis of CBD.PT cenA-gD.1 fusionprotein produced by pUC12-CBD.PTcem-gD.1,in E. coli JM101^  5624. Growth curves of E. coli JM 101 cultures harbouringpUC12-CBD.PTcenA-gD.I^  5725. Growth curves of E. coli JM 101 cultures harbouringpUC12-CBD.PTcenA-gD•1  5826. SDS-PAGE analysis of CBD.PTcenA-gD.1 fusionprotein produced by pUC12-CBD .PT cenAcgD .1 , inxE. coli AB1899, UT 5600, CAG 626, CAG 597 orCAG 629 strains^  6027. SDS-PAGE analysis of CBD.PTcenA-gD.1 fusionprotein produced by pUC12-CBD.PTcenA-gD.1,in E. coli JM 101, BL 21, KS 476 or UT 5600 strains^ 6128. Graphic illustration of 1St set ELISA 0. D. readings-anti-alkaline phosphatase antibodies^  6329. Graphic illustration of 1St set ELISA 0. D. readings -anti-CenA antibodies^  6430. SDS-PAGE analysis of immunogens^ 6831. Graphic illustration of 2nd set ELISA O.D. readings -primary anti-alkaline phosphatase antibodies^ 6932. Graphic illustration of 2nd set ELISA 0. D. readings -primary anti-CenA antibodies^  7033. Graphic illustration of 2nd set ELISA 0. D. readings -secondary anti-alkaline phosphatase antibodies^ 7134. Graphic illustration of 2nd set ELISA 0. D. readings -secondary anti-CenA antibodies^  7235. Graphic illustration of 2nd set ELISA results - primaryanti-alkaline phosphatase titers  7536. Graphic illustration of 2nd set ELISA results - primaryanti-CenA titers^  7637. Graphic illustration of 2nd set ELISA results - secondaryanti-alkaline phosphatase titers^  7738. Graphic illustration of 2nd set ELISA results - secondaryanti-CenA titers^  78x iABBREVIATIONS, NOMENCLATURE AND SYMBOLSaa,^amino acid(s)Ap,^ampicillinAvicel,^microcrystalline cellulose13-gal,^13-galactosidaseCBDCenA,^cellulose binding domain of CenACBDcex ,^cellulose binding domain of CexcenA,^gene coding for the endo-1,4-13-glucanase A ofCellulomonas fimicex,^gene coding for the exo-1,4,-13-glucanaseof Cellulomonas fimidNTP,^deoxyribonucleotide triphosphates (dATP,dCTP, dGTP, dTTP)ds,^double strandedEDTA,^ethylene diamine tetraacetic acidELISA,^enzyme-linked immunoadsorbent assaygD .1 ,^gene coding for the glycoprotein D of HerpesSimplex Type 1IPTG,^isopropyl-13-D-thiogalactopyranosidekbp,^kilobase pair(s)kDa,^kilodalton(s)X, E. coli phage lambdaLB,^luria brothPBS,^phosphate buffered salinePBST,^phosphate buffered saline-tween 20phoA,^gene coding for E. coli alkaline phosphatasePMSF,^phenyl methyl sulfonyl fluorideXIISDS-PAGE, sodium dodecyl sulphate-polyacrylamide gelelectrophoresisss,^single strandedTBS,^tris buffered saline-tween 20TLCK,^N-a-p-Tosyl-L-lysine chloromethyl ketoneX-gal,^5-bromo-4-chloro-3-indoly1-13-D-galactosideXIIIACKNOWLEDGEMENTSI would like to thank Drs. D. G. Kilburn, R. C. Miller Jr. and R. A. J.Warren for allowing me the opportunity to work on this project, and fortheir guidance, encouragement and friendship during my research. I wouldalso like to thank Drs. J. G. Levy and F. Tufaro for taking the time to beon my committee. I would like to thank Dr. Donald Trimbur for hisassistance and support in various aspects and stages of the thesis, Dr. T.Atkinson for oligonucleotide synthesis, Eric Jervis for his assistance withstatistics, and Diane Driver for solidarity. I am grateful to all my fellowcolleagues and friends, especially, Neena Din and Jeffrey Greenwood fortheir friendship, exchanging of ideas, and collaboration; Pat Miller whotransforms the cellulase group into a family, for constant support; and PeteLutwyche, for exchanging a different type of ideas, and inviting me to beon his radio program. Lastly, I would like to thank Argun for hisenthusiasm for my being a scientist. This work was supported by grantsfrom the British Columbia Health and Science Research Foundation. Idedicate this thesis to the memory of my mother, Nuran.1INTRODUCTIONENDOGLUCANASE CenA OF CELLULOMONAS FIMICellulose is a linear polymer composed of glucose subunits linked by13-1,4-glucosidic bonds. In the native state, cellulose molecules form fibers,which are largely composed of compact, crystalline domains separated bymore amorphous regions. Cellulose degradation by bacteria and fungi iscarried out by complex multienzyme systems. The cellulase system of thegram positive bacterium Cellulomonas fimi has been the subject ofinvestigation of this research group over the last ten years.Endoglucanase CenA of Cellulomonas fimi comprises two discrete,functionally independent domains joined by a 23 amino acid linkersequence comprised of proline and threonine residues (the PT box) (Fig.1).The N-terminal domain of 111 amino acids binds the enzyme to cellulose(the cellulose-binding domain, CBD); the C-terminal domain of 284 aminoacids is the catalytic domain (Langsford et al., 1987; Gilkes et al., 1988).The binding mechanisms are yet to be discovered, but it is thought thathydrophobic interactions and hydrogen bonding may be involved. The genefor CenA, cenA, has been cloned in E. coli (Wong et al., 1986). Itsproduct can be purified from cell extracts of E. coli by affinitychromatography on cellulose (Gilkes et al., 1988). E. coli exports asignificant fraction of the CenA it produces to the periplasm (Wong et al.,1986).The properties of the CBD of CenA make it very attractive for usein hybrid proteins which can be purified by affinity chromatography oncellulose. The use of CBD to confer specific adhesive properties to arecombinant protein has been demonstrated using the CBDcenA-PhoANH2 COOH2CBD^PT^catalytic domain1 31^142^165^449Fig. 1 Cellulomonas fimi endoglucanase A (CenA)■ Leader peptide, 0 CBD, ■ PT box, E catalytic domain.NH2 COOH1 31^165 205 222^667Fig. 2 CBDcenA-PhoA fusion protein.■ CenA leader peptide, 0 CBD, ■ PT box, ^ 40 aa of thecatalytic domain, El 17 aa of the TriPhoA,Iffi Alkaline Phosphatase.3fusion protein (Fig.2) (Greenwood et al., 1989). CBD cenA-PhoA fusionprotein is exported to the periplasm of E. coli; and can be purified fromperiplasmic extracts in a single, facile step by binding to cellulose.GLYCOPROTEIN gD OF HERPES SIMPLEX VIRUS TYPE 1The closely related herpes simplex type-1 and type-2 viruses (HSV-1and HSV-2) are the causative agents of various primary and recurrenthuman diseases ranging in severity from minor skin infections throughoften fatal complications (in exposed newborns) and viral encephalitis(Nahmias et al., 1981). In addition, there may be as 9 million cases ofrecurrent genital infections caused primarily by type-2 each year in theUnited States alone.The infectious HSV particles consist of a linear double-strandedDNA genome of 150 kbp in length contained within an icosahedralnucleocapsid that is in turn surrounded by a membrane envelope (Roizmanand Furlong, 1974). Embedded within the virion envelope, are the HSV-specified glycoproteins. (Norrild, 1980). Glycoprotein gD is abundant inthe virion coat and the cytoplasmic membrane of the infected cell, andelicits strong humoral and cellular immune responses in the infected host(Corey and Spear, 1986). Antiserum prepared to HSV-1 gD polypeptide istype-common. Thus, polyvalent gD antiserum (Cohen et al., 1978) andcertain gD monoclonal antibodies (Showalter et al., 1981; Pereira et al.,1980) recognize antigenic determinants (epitopes) and neutralize infectivityof both HSV-1 and HSV-2.The protein coding region of the HSV-1 glycoprotein D (gD.1) genewas mapped, and the nucleotide sequence was determined (Watson et al.,1982). The gD.1 gene has no intervening noncoding sequences, and codesfor 394 residues of gD.1. The first 20 N-terminal amino acids, with the4L^ TMNH2 COOH1^25^340 364 394Fig. 3 Glycoprotein D of Herpes Simplex Virus Type 1.■ Leader peptide, 0 Transmembrane hydrophobic domain,■ Anchor region5exception of arginine at position 7, are hydrophobic or non-polar. Thissequence constitutes part of the signal sequence. The first 25 amino acids ofthe translation product are cleaved off during posttranslational processingsteps to yield the mature glycoprotein (Eisenberg et al., 1984). A furtherstrongly hydrophobic region of 25 amino acids is present at positions 340to 364 which comprises the trans membrane (TM) a-helical region. ThisTM region is followed by a strongly basic region which serves to anchorthe glycoprotein in the membrane (Garoff et al., 1980), (Fig. 3).Antigenically active recombinant gD.1 polypeptides have beenproduced in recombinant bacterial systems (Weis et al., 1983; Amann etal., 1984) as well as in transformed mammalian cell lines (Lasky et al.,1984).The hydrophobic regions at the N-terminus and C-terminus of gD.1have been found to be detrimental to the bacterial hosts (Watson, 1983;Amann et al., 1984a; Steinberg et al., 1986). It was found necessary toremove the sequences coding for these hydrophobic regions for theefficient expression of the gD protein in E. coli (Amann et al., 1984;Steinberg et al., 1986). In addition, some of the recombinant gD.1polypeptides were rapidly degraded in E. coli (Amann et al., 1984a). gD.1could be expressed at high levels as tripartite fusion proteins. cI::gD.1::13galand cro::gD.1::13gal fusion proteins formed inclusion bodies and wereproduced in massive amounts (Brosius, 1984; Amann et al, 1984; Weis etal., 1983).Genetically engineered mammalian cell lines transformed withrecombinant plasmids containing the N-terminal 300 residues but missingthe C-terminal hydrophobic regions were found to synthesize, glycosylateand secrete a gD.1 form which is antigenically similar to native, membranebound form of gD.1 (Lasky et al., 1984).6ADJUVANTSAdjuvants are immunostimulatory substances that potentiate theimmune response against an antigen or vaccine beyond the level achievedby administration of the antigen alone. Traditional live virus vaccinesrequire no adjuvants, as they have the advantage of replicating. Killed virusvaccines are generally sufficiently immunogenic to be effective with noadjuvant or with adjuvants that have limited ability to stimulate immuneresponses. Recently, novel subunit vaccines have been created byrecombinant DNA technology. Such subunit vaccines are polypeptidesrepresenting one or more viral proteins. Because recombinant DNA-derived subunit vaccines involve no infectious agent at any stage ofproduction (except some special approaches like the vaccinia virusexpression system), they are expected to offer significant safety advantagesover virus-derived vaccines. Subunit vaccines may also offer advantages interms of purity, lot-to-lot consistancy, and cost of production. However,subunit vaccines are less immunogenic than whole virus particles andtherefore will require safe adjuvants with significant immunostimulatorycapabilities to obtain their full potential in preventing disease.It is now 65 years since Glenny discovered the immunogenicproperties of diphtheria toxoid precipitated with alum (Glenny et al.,1926), but the mineral gels remain to this day the only adjuvants employedin human vaccines, although the less stringent toxicological requirements ofveterinary vaccines permit a more catholic selection (Bomford, 1985).However, alum does not appear to provide the level of immune stimulationrequired for most recombinant DNA-derived subunit vaccines. Variousstudies with a recombinant gD vaccine (Berman et al., 1985; Sanchez-Pescador et al., 1988) or glycoprotein subunit vaccines prepared fomvirus-infected cells and presented with alum have shown little efficacy7(Thomson et al., 1983; Meignier et al., 1987). More recently, an alumadjuvant glycoprotein subunit vaccine preparation has failed to show anyprotective efficacy in a human clinical trial (Corey et al., 1987).Disadvantages associated with the use of aluminum salts include theinduction of granulomas at the injection site (Turk et al., 1977), lot-to-lotvariation and instability of alum preparations (Edelman, R., 1980), andlack of effective stimulation of cell-mediated immunity, which is especiallyimportant for viral vaccines (Bomford, R., 1980). Despite thesedisadvantages and poor performance with various subunit vaccines, alum isthe only adjuvant currently approved for human use; clearly indicating theneed for a novel, potent and safe adjuvant.This study investigates the potential of cellulose binding domains ofcellulases as tools for augmenting the immune response against antigens. Itwas postulated that an antigen fused to a cellulose binding domain that is inturn bound to cellulose would be more immunogenic than the antigenalone. This adjuvant potential could be investigated by constructing genefusions between the antigens and the cellulose binding domains. Thechimaeric protein products would be purified by affinity chromatographyon cellulose. The immunogenicities of the native antigens, CBD-antigenfusion proteins, and CBD-antigen fusion proteins bound to cellulose, wouldbe tested by animal immunization. The anti-antigen and anti-CBD antibodytiters would then be compared using immunological assays.CBD CenA , the cellulose binding domain of Cellulomonas fimiendoglucanase, was chosen for the construction of the gene fusions. Theantigens involved were gD.1, the glycoprotein D of Herpes SimplexVirus type 1; and PhoA, the alkaline phosphatase of Escherichia coli.Mice were used for immunization, and the immunogenicities werecompared using the enzyme-linked immunoadsorbent assay to measureprimary and secondary antibody titres.8MATERIALS AND METHODSI BACTERIAL STRAINS, PHAGES AND PLASMIDSA list of bacterial strains, phages and plasmids used in these studies isgiven in Table I. Stock cultures of bacteria were maintained at -20°C in LBmedium containing 15% glycerol or at -70°C as DMSO stocks.II MEDIAThe preparation of media and supplements was as described byManiatis et al. (1982) and Messing (1983). Luria broth (LB) contained perlitre; 10 g of Bacto-tryptone, 5 g of Bacto-yeast extract and 10 g of NaCl.B broth contained per litre, 10 g of Bacto-tryptone, 8 g of NaC1 and 1 mlof 1% vitamin B 1 . 2xYT medium contained per litre; 16 g of Bacto-tryptone, 10 g of Bacto-yeast extract and 5 g of NaCl. M9 salts containedper litre; 6 g Na2HPO 4 , 3 g KH2PO4 , 0.5 g NaCl, 1 g NH4C1 with theaddition, after autoclaving, of 2 ml of 1 M MgSO 4 and 0.1 ml of 1 MCaC12 . Minimal medium contained per litre; 100 ml of 10x M9 salts, 10 mlof 20% glucose, 10 ml of 0.01 CaC12 , 1 ml of 1 M MgSO4 , 1 ml of 1%thiamine. Agar plates contained 1.5% agar.III BUFFERSThe composition and preparation of the buffers and solutions9Table 1 Bacrterial strains, phage and plasmidsBacterial Strain Genetic characters^ReferenceE. coil JM101^supE thi A(lac-proAB)  [F' traD36^Messing, 1979proAB lacrl ZAM15]E. coil JM110^thr leu thi lac Y gol K ara tonA tsx^Yanish-Perron et al., 1985dam dcm supE44 A(lac-pro) [F' traD36proAB lacIcl ZAM15]E. coli BL 21E. coil UT 5600E . coil KS 476E. coil CAG 626E. coil CAG 597E. coil CAG 629(relevant genotype) A ompT(relevent genotype) A omp T(relevent genotype) deg P41(relevent genotype) SC122 lon -(relevent genotype) SC122htpR165-Tn10(relevent genotype) SC122lon-htpR165-Tn10Grodberg and Dunn, 1988Earhart et al., 1979Strauch et al., 1989Cooper and Ruettinger, 1975Cooper and Ruettinger, 1975Cooper and Ruettinger, 1975PhageM13mp18PlasmidpUC12pUC18Genetic characterslacGenetic characterslac, ApRlac, ApRReferenceYanisch-Perron et al., 1985ReferenceVieira and Messing, 1982Yanisch-Perron et al., 198510employed in these studies are described by Maniatis et al. (1982).Phosphate-buffered saline (PBS); contained per litre, 8 g NaCl, 0.2 g KC1,0.2 g KH2PO4 , 1.15 g Na2HPO4 .7 H2O and if not sterilized, 0.2 g NaN 3 .PBS-tween (PBST) contained per litre, 0.5 ml of tween-20. Tris-bufferedsaline (TBS) contained per litre; 6.06 g tris-base (final conc. 50mM), 11.7g NaC1 (final conc. 200 mM). The pH was adjusted to 7.5 with dilute HC1.TBS-tween (TBST) contained per litre, 0.5 ml of tween-20. Carbonatecoating buffer, pH 9.6, contained per litre; 1.59 g Na 2CO 3 , 2.93 g NaHCO3and 0.2 g NaN3 . Substrate buffer used for immunoblots contained per 100ml; 1.21 g tris-base (final conc. 100 mM), 0.58 g NaC1 (final conc. 100mM) and 1.02 g MgC12 (final conc. 50 mM). pH was adjusted to 9.5 withdilute HC1.IV GROWTH AND INDUCTION OF BACTERIABacteria were grown in LB or 2xYT medium at 30 or 37°C.Selective antibiotic media contained 100 jig ampicillin / ml. For theinduction of the lac promoter, isopropyl-B-D-thiogalactopyranoside(IPTG) was added to a final concentration of 0.1-1 mM.V DNA MANIPULATIONS AND RECOMBINANT DNATECHNIQUES(A) Enzymes, reagents and techniquesIn general, DNA preparations, restriction enzyme reactions andrecombinant DNA techniques were performed as described by Maniatis et11al. (1982) and Messing (1983). Restriction endonucleases, DNA polymeraseI - Klenow fragment, T4 DNA ligase, mung bean nuclease,deoxyribonucleotides were purchased from Bethesda ResearchLaboratories (BRL), Pharmacia Inc., Boehringer-Mannheim and NewEngland Biolabs (NEB). All of the enzymes were used under the conditionssuggested by the suppliers. Bacterial transformations were carried outusing CaC12 method, according to Maniatis et al. (1982).(B) Preparation and purification of plasmid and phage DNAThe procedures for the growth of phage Ml3mpl 8, purification ofits ds and ss DNA form have been described in detail by Messing (1983).E. coli JM101 or E. coli JM110 was used for the propagation ofM13mp18 phage. For the isolation of plasmid DNA, E. coli JM101containing the appropriate plasmid was grown to a stationary phase in LBwith ampicillin. Cells were collected by centrifugation and plasmid DNAwas isolated by the alkaline lysis method of Birnboim and Doly (1979) asmodified by Maniatis et al. (1982). For large scale isolation of plasmidDNA, 500 ml of LB with ampicillin was inoculated and the cells weregrown to stationary phase. Plasmid DNA was isolated by alkaline lysismethod and purified by ultracentrifugation through CsCl-ethidium bromidegradient.(C) Synthetic oligodeoxyribonucleotidesThe oligo deoxyribonucleotides 5'-AAA TAT GCC TTG GCG GATGCC TCT CTC-3' (24 mer; synthesized for the preparation of the gD.112fragment ), 5'-CGT CGG CGT GGG GGT GGG GGT CGG-3' (24 mer;complementary to the PT box of CenA; synthesized for the preparationof the CBD.PT fragment), 5'-TAG GAA TTC GGT AC-3' (14 mer;synthesized for the linker), 5'-C GAA TTC CTA-3' (10 mer; synthesizedfor the linker), 5'-TCC CCG GGT ACC GAA TTC CTA GGC GCC GGATCC AGG CCT GCC CTC GAT CGG CGT GGG GGT GGG GGT CGG-3' (69mer; synthesized for the site directed mutagenesis), 5'-GTC CAGGAC CGG AAG GTC TTT GCC-3' (24 mer; complementary to the gD.1sequences; synthesized to sequence the junction of CBD cenA-gD.1) weresynthesized chemically on an applied biosystems 380 A DNA synthesizer byT. Atkinson using phosphite triester chemistry, essentially as described(Atkinson and Smith, 1984). All, but the 69 mer oligodeoxyribonucleotidewere purified directly by binding to a Sep-Pak C18 reverse phase cartrigein water, followed by washing with water, and elution with 20%acetonitrile-80% water. The 69 mer oligodeoxyribonucleotide wasseparated from incomplete products by electrophoresis in a 16%polyacrylamide-7 M urea sequencing gel, located by UV-shadowing, andextracted from the gel by the crush and soak method (Atkinson and Smith,1984). It was further purified by using Sep-Pak C18 reverse phasecartrige.(D) Primer annealing reaction and Klenow extentionTwenty pmoles of ss M13mp1 8-cenA or M13mp1 8-gD / wasmixed with a 10 fold molar excess of the appropriate oligonucleotides in200 p1 of Hine II buffer (BRL). The DNA was denatured by heating thesolution in a 1.5 ml microfuge tube at 90°C for 5 minutes and then allowed13to anneal by cooling to 23°C over a period of two hours. The reactionmixture was then supplemented to give final concentrations of 500 gM foreach dNTP and 1 mM dithiothreitol (DTT). Fifteen units of DNApolymerase I (Klenow fragment) (BRL, Gaithersburg, MD) were added tothe reaction. After 20 minutes of incubation at 50°C, a further 15 unites ofDNA polymerase I (Klenow fragment) were added and the incubation wascontinued for another 15 minutes. The reaction mixture was deproteinizedby extraction with phenol/chloroform and chloroform. The DNA wasrecovered by ethanol precipitation and resuspended in 200 pl of mung beannuclease buffer (NEB, Beverly, MA).(E) Mung bean nuclease reactionDNA products from the elongation reactions of M13mp18-cenA orMl3mp18-gD.1 DNA primed with an oligodeoxyribonucleotide weretreated for 30 minutes at 30°C with 50 units of mung bean nuclease in areaction volume of 200 IA of mung bean nuclease buffer (NEB, Beverly,MA). The enzyme was inactivated by the addition of SDS to 0.4% andphenol extraction. The DNA was then precipitated with ethanol andresuspended in the appropriate buffer to be digested with the appropriaterestriction endonuclease.(F) LigationsFor each ligation, the total concentration of all DNA termini (i), andthe effective concentration of both ends of the same molecule in closeproximity (j) were calculated.14i (gM) = (2) (1..tg DNA/111 solution) (1/.66 x)j (JIM) = (5.3 x 10 -4) (48/x) 1 .5Where "x" is the size of the DNA molecule under consideration inkilobases.The ratio of j (vector)/ [ i (vector)+i (insert) was kept between 1and 2, by adjusting the relative amounts of the DNAs; for efficientligations.In general; the total concentration of DNA was 5-10 lig I ml in afinal volume of 20-50 of ligation buffer (BRL); the molar ratio of insertto vector was 3-5:1, and linker to vector was 25:1. For the ligations ofsticky-ended fragments, 10 units of T4 DNA ligase (BRL) were used, andthe reaction mixtures were incubated overnight at 16°C. Up to 400 units ofthe enzyme were used for the ligations of blunt-ended fragments and theincubations were at 4°C.(G) Preparation of the linkerThe 10 mer and 16 mer oligodeoxyribonucleotides were mixed at amolar ratio of 1:1 (3 lag : 5.28 jig) in 30 1.t1 of 5 mM MgC1 2 solution. Themixture was heated in a 1.5 ml microfuge tube at 68°C for 5 minutes. TheDNA was then allowed to anneal by cooling to 23°C over a period of 30minutes. After reaching room temperature, the mixture was removed to4°C.(H) Site directed mutagenesis15Eighteen pmoles of M13mp18-CBD.PT was mixed with 45 pmolesof the 69 mer primer in 200 IA of Hinc II buffer (BRL). The DNA wasdenatured by heating the solution at 90°C for 5 minutes and then allowed toanneal by cooling to 23°C over a period of 2 hours. The reaction mixturewas then supplemented to give final concentrations of 500 p.M for eachdNTP, 5 mM dithiothreitol (DTT), 1 mM ATP, followed by the additionof 15 units of DNA polymerase I (Klenow fragment) (BRL) and T4 DNAligase (BRL). The reaction mixture was incubated at 37°C for 2 hours.Two ngs of the DNA were then used to transform competent E. coliJM101 cells.VI ELECTROPHORESIS OF DNAAgarose gels (0.8 - 1.2%) were used for the analysis, and thepreparation of DNA fragments. Gels contained 1 ethidium bromide perml of TAE (0.04 M Tris-acetate 10.001 M Na 2EDTA) buffer. DNA bandswere visualized by fluorescence using an UV transilluminator. The DNAfragments were recovered from gel slices using Gene CleanTM (BIO 101Inc., La Jolla). Gene CleanTM is a specially formulated silica matrix thatbinds DNA without binding DNA contaminants. Electroelution frompolyacrylamide gel (8%) slice using the dialysis bag technique (Maniatis etal. 1982) was used for the preparation of 0.5 kbp CBD.PT DNA. The 1"dialysis tubing was preboiled in 0.001 M EDTA for 3 minutes. After oneend was clamped, it was filled with 0.1 x TBE (50 mM Tris-OH / 50 mMboric acid / 0.002 M Na 2EDTA) buffer. The excised gel slice was placedagainst one side. The open end was then clamped and the tubing was placed16in electrophoresis apparatus filled with 0.1 x TBE buffer. Two hundredvolts were applied for 45 minutes, followed by the reversal of the currentfor 2 minutes. The buffer containing the DNA was removed from thetubing and the residual ethidium bromide was removed by washing withwater-saturated n-butanol. The DNA was then precipitated with ethanol.VII DNA SEQUENCE DETERMINATIONThe DNA sequences were determined by the Sanger et al. (1977)procedure. Samples from sequencing methods were analysed on 6%denaturing (8 M urea) polyacrylamide gels.VIII PREPARATION OF CELL EXTRACTSThe bacteria were inoculated into 5 ml of medium containingampicillin, and incubated overnight at 30 or 37°C. IPTG (0.1 mM) wasadded at the time of inoculation or 1-2 hours before harvesting. The cellswere harvested by centrifugation and resuspended in 0.5 ml of 50 mMKPO 4 buffer (pH 7.0), 0.02% NaN 3 . The cells were sonicated using aBronson sonifier with a microprobe, followed by centrifugation for 15minutes at 10 K in an Eppendorf microcentrifuge. The supernatants wereremoved. The protease inhibitors PMSF (0.1-1 mM), TLCK (1 mM) andEDTA (3 mM) and pepstatin A (0.1 mM) were added immediately beforeand after sonication. This soluble fraction was used in binding assays. If theactive form of the proteins were not needed, the cells harvested weredirectly resuspended in 250 gl of SDS-PAGE loading buffer (6.25 mMTris-C1, pH 6.8, 2%SDS, 10% glycerol, 5% 2-mercaptoethanol, 0.002%17bromphenol blue), boiled for 10 minutes, and then centrifuged for 5minutes in an Eppendorf microcentrifuge. A total of 2.5 to 5 of thesolubilized material were loaded per slot.IX ELECTROPHORESIS OF PROTEINSProteins were separated by electrophoresis in 10% polyacrylamidegels in the presence of 0.1% SDS. The gels were ran at 200 V in a miniprotein gel apparatus (Bio-Rad) and then stained with Coomassie blue for30 minutes on an orbital shaker, followed by destaining for approximately1 day. The gels were then soaked in 2% glycerol solution for 2 hoursbefore being dried in between two sheets of cellophane presoaked in dH2O.X WESTERN BLOTTINGThe proteins were blotted for 1.5 hours at 500 mA in a Bio-Radblotting apparatus. Detecting CBD.PT : All incubations were for 1 hr at37°C on an orbital shaker. PBS-3% BSA was used for blocking the residualsites and for diluting the antibodies. PBST was used for the washes.Preabsorbed rabbit anti-CenA antibodies and goat anti-rabbit alkalinephosphatase conjugate antibodies were diluted to 1:2000. The blots werewashed three times 10 minutes after the incubations with antibodies. Afterthe last 3 washes, the blots were washed for 10 minutes each, first withPBS and then with 0.1 M Tris-Cl (pH 9.6), followed by the addition of thesubstrate (0.1 ml from 5mg / ml of 5- Bromo-4-Chloro-3-IndolylPhosphate in DMF, 1.0 ml from 1 mg/ml Nitro Blue Tetrazolium in 0.1 M18Tris-Cl (pH 9.6), 0.4 ml from 1M MgC12 in a total volume of 10 ml of 0.1M Tris-Cl (pH 9.6).Detecting gD-1 : The residual sites were blocked for 1 hr at RT withTBST-1% BSA. The blots were then incubated for 1 hr at RT with themouse anti-gD-1 monoclonal antibodies (A 74, 1 mg/ml) diluted to 1 :5000 with TBST. After washing three times 10 minutes with TBST, theblots were incubated for 30 minutes at RT with biotinylated goat anti-mouse IgG (BRL, Gaithersburg, MD; lmg/m1) diluted to 1 : 1000 withTBST. The washes were repeated, and the blots were incubated for another30 minutes at RT with streptavidin-alkaline phosphatase (BRL, 0.88 mg /ml) diluted to 1 : 12000 with TBST. The blots were washed three times 10minutes with TBS, two times 5 minutes with the substrate buffer followedby the addition of the substrate [44^Nitro Blue Tetrazolium (NBT)(Sigma Chem. Co., St louis, MO), 33^5-Bromo-4-Chloro-3-IndolylPhosphate (BCIP) (Sigma Chem. Co)) in 10 ml of substrate buffer]XI AVICEL BINDING ASSAYOne hundred pi of crude cell extract was mixed with 10 mg ofAvicel (FMC International, Ireland) prewashed twice with 1 ml of dH 2Oand twice with 1 ml of 50 mM potassium phosphate buffer (pH 7), in a 1.5microfuge tube . The mixture was incubated at 0-4°C for 1-16 hr withconstant agitation, followed by centrifugation for 10 minutes at 10 5 g in anEppendorf microcentrifuge. The supernatant containing the unboundproteins was transferred. The pellet was washed twice with 0.5 M NaCl andfour times with potassium phosphate buffer, and finally resuspended in 40IA of 2 x SDS-PAGE loading buffer. After boiling for 2 minutes, the19mixture was centrifuged in an Eppendorf microcentrifuge, and thesupernatant containing the bound proteins was transferred.XII PREPARATION OF IMMUNOGENSImmunogens used for the second immunization experiment wereprepared as described below.The concentration of CBDcenA-PhoA in the preparations was 400pg/ml. An equimolar amount of Alkaline Phosphatase was used. Theconcentrations of Avicel and Al(OH) 3 in the preparations were 100 p.g/g1and 1 pz/fil respectively. Dilutions were in PBS buffer.Purification and preparation of CBD cenA-PhoA : CBDcenA-PhoA wasallowed to bind Avicel in PBS buffer. The mixture was incubated at 4°Cfor 16 hrs with constant agitation, followed by centrifugation for 10minutes in an Eppendorf microcentrifuge at 10 K. The supernatantcontaining the unbound proteins was transferred. The pellet was washedtwo times with 0.5 M NaCl, four times with PBS buffer. The CBD cenA-PhoA was then eluted with dH2O. The protein was concentrated by Diafloultrafilters (Amicon, Danvers, MA) and sterilized with Acrodisc (GelmanSciences Inc., Ann Arbor, MI). The final concentration of CBD cenA-PhoAwas 10 pig / pl as measured by Bio-Rad protein assay.Avicel-Ag preparation : Four hundred pig of purified CenA-PhoAwas mixed with 100 mg of Avicel which was prewashed twice with dH 2Oand twice with PBS buffer. The final volume was adjusted to 1000 pi, withPBS. The Ag was allowed to bind Avicel overnight at 4°C on a rotaryshaker. The mixture was then centrifuged in an Eppendorf microcentrifugefor 10 minutes at 10 K. The pellet was washed two times with 0.5 M NaC1,four times with PBS buffer and resuspended at a final volume of 1000 pl in20PBS buffer. In case of Alkaline Phosphatase (from E. coli Type III; SigmaChem. Comp., St louis, MO) as the Ag, the same procedure was applied,except, after the "binding" step the mixture was kept directly for theimmunization.Al(OH)3-Ag preparation : 400 pig of CBDcenA-PhoA (40 Ill) wasmixed with 960 ill of 1.05 mg / ml Al(OH)3 (Connaught Laboratories).Immunogens used at the first set of immunizations (preliminaryimmunizations) have been prepared by Emily Kwan to match the finalconcentrations described above. The major differance at the preparationmethods was that, Avicel was not washed after the CBID cenA -PhoA hasbound.XIII MICE IMMUNIZATIONFive groups of 5 female CBA mice, 5-7 weeks old, were immunizedtwice at a 3-wk interval with 20 lig of CBDcenA-PhoA or an equimolaramount of alkaline phosphatase presented in the various preparations. Acontrol group of 5 mice were not immunized. Immunizations were doneintraperitoneally with injection volume of 50 111. Animals were bled by tailbleeding 2 weeks after the first, and by cardiac puncture 1 week after thefinal immunization for antibody determinations.The blood obtained was kept at 37°C for 1 hr (for clot retraction),then at 4°C overnight, followed by centrifugation for 5 minutes in anEppendorf microcenrifuge. The supernatant (serum) was removed. A finalconcentration of 0.01% NaN 3 was added and the sera were then kept at -20°C.The first set of immunizations : Sixteen female CBA mice, 5-7 weeksold, were immunized twice at a 2-wk interval with 20 14 of CBDcenA-21PhoA or alkaline phosphatase presented in the various preparations, whileanother 4 did not receive any immunization. Immunizations were doneintraperitoneally with an injection volume of 50 p1. Animals were bled 1week after the second immunization.XIV ELISAAntigen, blocking, substrate solutions and mice sera were used at100 ptl / well. Alkaline Phosphatase (Sigma Chem. Comp.) or CBD cenA-PhoA were diluted to a final concentration of 5 p.g/m1 in carbonate coatingbuffer (pH 9.6) (see BUFFERS), and adsorbed to the wells of 96-wellImmulon II plates (Dynatech). The plates were incubated overnight at 4°C.After washing three times with PBST, the blocking solution (PBS-1%BSA) was added and the plates were incubated for 1 hr at 37°C. Theblocking solution was then removed, and the test sera diluted serially withtwo-fold dilutions in the wash buffer were incubated in the Ag-coated wellsfor 1 hr at RT. The plates were washed as before, and Goat Anti-MouseIgG + IgA + IgM - Horseradish Peroxidase Conjugate (BRL) diluted 1 :1000 in wash buffer was added. The incubations were at 37°C for 1 hr.The plates were washed as before and developed with the substrate [0.6 mlof 1% (w/v) 0-Phenylene Diamine in methanol, 60 p1 of 3% (v/v) H202 ina final volume of 30 ml of dH2O]. The colour reaction was stopped after 30minutes by the addition of 501.t1/ well of 4 N H2SO4.22RESULTSI CONSTRUCTION OF PLASMID pUC18-CBD.PT-gD.1(SEE FIGURE 4)(A) Subcloning of the gD.1 gene into M13mp18•^The phage M13mp18 DNA, and the plasmid pUC18-gD.1 carryingthe glycoprotein D gene of herpes simplex virus type-1 (gD.1) weredigested by Hind III and Sst I restriction endonucleases. The 7.2 kbp phageDNA and the 1.8 kbp DNA fragment carrying the gD.1 gene -isolatedfrom a 0.8% agarose gel using Gene CleanTM - were ligated. The ligationmixture was used to transform competent E. coli JM101 cells.Isopropylthiogalactoside, 5-dibromo4-chloro3-indolylgalactoside, soft agarand freshly growing E. coli JM101 cells were added, and the mixture wasplated on B broth plates. The plates were incubated at 37°C.Five of the white plaques were picked, their DNA was extracted bythe alkaline lysis method and digested with Bal I and Kpn I to determinethe orientation of the insert DNA in the constructs. The clones gave 1.5kbp,,and 7.5 kbp sized fragments suggesting that the inserts were in theright orientation. (fig. 5).(B) Preparation of the gD.1 fragmentE. coli JM101 cells were transfected by the phage M13mp18-gD.1M13mp1824 mer primerDNA polymeraseMung bean nucleaseSstIblunt1.7 kbpFig. 4 ConstructionofpUC18-CBD.PT-gD.IEcoRI^bluntpUC18EcoRIS stl23pUC18-cenASstISstI^ SstIpUC18-gD./L CBD PT Catalytic D.M13mp1824 mer primerDNA polymeraseMung bean nucleaseEcoRI SstI IllSstIHindIIITMLigationSstI24FIG 5 Elucidating the orientation of gD.1 DNA in the constructMl3mp18-gD./by Kpn I and Bal I digestion. LANE 1 : Nucleic acid MWstandards, 2c-Hind III fragments (Sanger et al., 1982); 23,130 bp, 9,416 bp,6,557 bp, 4,361 bp, 2,322 bp, 2,027 bp, 0,564 bp, 0,125 bp. LANE 2 :Ml3mp18-gD.1-Kpn I and Bal I fragments, 7.5 and 1.5 kbp (Arrows).25and the culture medium containing the phage particles was used for thepreparation of single stranded (ss) phage DNA.The 24 mer oligodeoxyribonucleotide 5'-AAA TAT GCC TTG GCGGAT GCC TCT-3' - complementary to the gD.1 sequence immediatelydownstream of the leader sequence - was allowed to anneal ss Ml3mp18-gD.1. After extending the primer using the Klenow fragment of DNApolymerase I, the single stranded protrusions were digested with mungbean nuclease. The DNA was then digested with Sst I to give a 1.6 kbpfragment carrying the truncated gD.1 gene. The fragment was thenisolated from a 0.8% agarose gel using Gene C1eanTM (fig. 6).(C) Preparation of CBD.PT fragmentE. coli JM101 cells were transfected by the phage M13mp18-cenA.The culture medium containing the phage particles was used for thepreparation of the ss M13mp18-cenA DNA.The 24 mer oligodeoxyribonucleotide 5'- CGT CGG CGT GGGGGT GGG GGT CGG -3' - complementary to the cenA sequence at thePT box immediately upstream of the catalytic domain - was allowed toanneal ss M13mp18-cenA. After extending the primer using the Klenowfragment of DNA polymerase I, the single stranded protrusions weredigested with mung bean nuclease. The DNA was then digested with EcoR Ito give a 0.5 kbp fragment carrying the CBD.PT gene (Fig. 7). Thisfragment was isolated from a 8% polyacrylamide gel by electroelution.26FIG 6 Monitoring the primer extention. Ml3mp18-gD./ DNA atvarious steps in the procedure was analyzed on a 0.8 % agarose gel. LANE1 : Nucleic acid MW standards, A'-Hind III fragments. LANE 2 :M13mp18-gD.1 ss DNA. LANE 3 : Oligonucleotide-primed M13mp18-gD.1 ds DNA after second strand synthesis. LANE 4 : Mung beannuclease-treated ds M13mp18-gD.1 DNA. Lane 5 : Mung bean nuclease-treated ds M13mp18-gD./ DNA after Sst I digestion (The 1.6 kbp gD.1fragment is marked by an arrow).27FIG 7 Preparation of the CBD.PT gene fragment using the primerextention method. LANE 1 : Nucleic acid MW standards, X -Hind III andEcoRI fragments; 21,226 bp, 5,148 bp, 4,973 bp, 4,268 bp, 3,530 bp,2,027 bp, 1,904 bp, 1,584 bp, 1,375 bp, 0,947 bp, 0,831 bp, 0,564 bp,0,125 bp. LANE 2 : Last stage of preparation of the CBD.PT fragment,mung bean nuclease-treated ds M13mp18-cenA ds DNA after EcoR Idigestion (The 0.5 kbp CBD.PT fragment is marked by an arrow).28(D) Construction of pUC18-CBD.PT-gD.1The plasmid pUC18 was digested with EcoR I and Sst I. TheCBD.PT, gD.1 and 2.65 kbp pUC18 fragments were ligated. The ligationmixture was used to transform competent E. coli JM101 cells. The cellswere plated on LB-Amp-IPTG-X-gal plates and incubated at 37°C.The transformants grew poorly. The colonies were extremely smalleven after prolonged incubation. When inoculated into the liquid mediafor DNA preparation, the transfers failed to grow. These findingssuggested that the the fusion protein was toxic to the host cells. As CenAwas successfully expressed in E. coli previously (Gilkes et al., 1984a,Wong et al., 1986), the toxicity was attributed to the gD.1 componenent ofthe fusion protein. Very hydrophobic as sequences such as the signalpeptide or the transmembrane (TM) region of membrane-boundeucaryotic proteins are often toxic to the bacterial host (Amann etal.,1984a; Brosius, 1984; Remaut et al., 1983; Rose and Shafferman, 1981;Yelverton et al., 1983). gD.1 in the fusion peptide did not have the signalpeptide, but contained the TM region. We concluded that, for an expressionof gD.1 in the bacterial host, the removal of the TM region might becrucial.29II CONSTRUCTION OF PLASMID pUC12-CBD.PT-gD.1(A) Construction of pUC18-CBD.PT(SEE FIGURE 8)The DNA fragment coding for CBD.PT cenA was prepared asdescribed above. The plasmid pUC18 was digested with EcoR I and Kpn I.The 2.65 kbp fragment was recovered from a 0.8% agarose gel using GeneCleanTM. These two fragments and the blunt-ended Kpn I linker wereligated. The linker contained from 5' to 3'; a stop codon, an EcoR I site -to facilitate screening and further subcloning, and the Kpn I sticky end.The ligation mix was used to transform competent E. coli JM101 cells.The cells were plated on LB-Amp-IPTG-X-gal plates and incubated at37°C.The plasmid DNA from three transformants was extracted anddigested with EcoR I to determine the orientation of the insert DNA in theconstructs. The clones gave 0.5 kbp and 2.65 kbp sized fragmentssuggesting that the inserts were in the correct orientation (Fig 9).Sequencing the junctions of these constructs using the dideoxy-sequencing method of Sanger et al. (1977) revealed that, in all threeconstructs, the insertion of the CBD.PT gene was as desired, but the genewas lacking its last codon (ACG). This was attributed to excess digestion bymung bean nuclease.Western blot analysis of the crude cell extracts of these recombinantclones showed that, CBD.PTcenit was expressed at high levels in E. coli(Fig. 20) (Din et al., in press). Its apparent molecular mass (20.0 kDa),relative to standard proteins, was greater than the molecular mass predictedfrom its primary structure (13.6 kDa). The PT box has previously been30Fig. 8Construction ofpUC18-CBD.PTpUC18-cenASstISstI^r—'  L CBD PTM13mp1824 mer primerDNA polymeraseMung bean nucleaseEcoRIpUC18EcoRIKpnI0.5 kbp(ii)EcoRIlac ZLigationTAG GAATTC GGTACATC CTTAAG Cblunt^EcoRI^KpnI(iii)lac Z'^Amp r(i)31FIG 9 Elucidating the orientation of CBD.PT DNA in plasmid pUC18-CBD.PT, by EcoR I digestion. LANE 1 : Nucleic acid MW standards, X-Hind III and EcoR I fragments. LANE 2 : pUC18 digested with EcoR I.LANE 3 : pUC18-CBD.PT-EcoR I fragments, 2.65 kbp and 0.5 kbp(Arrows).32FIG 10 Elucidating the orientation of CBD.PT DNA in the constructM13mp18-CBD.PT by Sst I digestion. LANE 1 : M13mp18-CBD.PT-SstI fragments, 7.2 kbp and 0.5 kbp (Arrows). LANE 2 : M13mp18-CBD.PT undigested. LANE 3 : M13mp18- Sst I fragment, 7.2 kbp.LANE 4 : M13mp18 undigested. LANE 5 : Nucleic acid MW standards, X-Hind III fragments.33shown to cause anomolous electrophoretic migration (Gilkes et al., 1989).Avicel binding assays carried on the crude cell extracts showed that theCBD.PTCenA had the ability of specifically binding to cellulose.(B) Subcloning of the CBD.PT gene into M13mp18CBD.PT was subcloned into M13mp18 to introduce new restrictionsites downstream of the gene.pUC18-CBD.PT and M13mp18 were digested with EcoR I. The 0.5kbp CBD.PT fragment, recovered from a 8% polyacrylamide gel byelectroelution and the linearized M13mp18 were ligated. The ligation mixwas used to transform competent E. coli JM101 cells. After the additionof soft agar, freshly growing E. coli JM101 cells, IPTG and X-gal; themixture was overlayed on LB plates. The plates were incubated at 37°C.DNA from nine transformants was extracted and digested with Sst Ito determine the orientation of the insert. The release of the 0.5 kbp insertby the three of the clones was consistent with the correct orientation (Fig.10).(C) Construction of M13mp18-CBD.PT.IEGR - Site directedmutagenesis. (SEE FIGURE 11)This construct was designed to provide a "cassette" for one stepcreation of fusion proteins with CBD.PT cenA • Fusion proteins containingthe cellulose-binding domains have proven useful for protein purificationand enzyme immobilization (Greenwood et al., 1989; Ong et al.,1989).Their potential use for immune enhancement is to be investigated here.The DNA sequence coding for the Factor Xa recognition site IEGR,34Fig. 11Construction ofMl3mp18-CBD.PT.I.EGRpUC18-CBD.PTir EcoRIEcoRI^EcoRIilL CBD PT0.5 kbpM13 mp 1893 mer primerDNA polymeraseT4 DNA ligase35and a unique Stu I site were introduced to the C-terminus of CBD .PT .Gene fusions could then be created in one step using the Stu I site. Thefusion protein could be digested with Factor Xa to release the purifiedforeign protein. This cassette construct has been used; digestion with FactorXa, and subsequent protein purification was demonstrated (Shen H.,personal comm.).E. coli JM101 cells were transfected by the recombinant phageM13mp18-CBD.PT. The culture medium containing the phage particleswas used for the preparation of single stranded phage DNA.The 69 mer oligodeoxyribonucleotide 5'- TCC CCG GGT ACCGAA TTC CTA GGC GCC GGA TCC AGG CCT GCC CTC GAT CGGCGT GGG GGT GGG GGT CGG -3' synthesized for the introduction ofthe DNA sequence coding for the Factor Xa site (IEGR) and a unique Stu Isite at the 3' end of the CBD .PT gene, immediately upstream of the stopcodon, was allowed to anneal to the ss M13mp18-CBD.PT. The primerextention mixture contained the Klenow fragment of DNA polymerase I,DTT, dNTPs, ATP and T4 DNA ligase (Fig. 12). All of the ss DNA wasextended as observed on an agarose gel. Two ng of the DNA was used totransform competent E. coli JM101 cells, resulting in —300 plaques.The mutagenesis was confirmed by sequencing using the dideoxy-sequencing method of Sanger et al. (1977). The construct was then used totransform competent E. coli JM110 cells for the digestion of the Stu I site(Fig. 13). (The Stu I site was overlapping with the dcm methylase site ofthe E.coli JM101 (Marinus and Morris, 1973; May and Hattman 1975),and was resistent to digestion).36I^2^3FIG 12 Monitoring the site directed mutagenesis. M13mp18-CBD.PTDNA at two steps in the procedure was analyzed on a 0.8 % agarose gel.LANE 1 : Nucleic acid MW standards, X-Hind III fragments. LANE 2 :Ml3mp18-CBD.PT ss DNA. LANE 3 : Oligonucleotide-primedMl3mp18-CBD.PT ds DNA after second strand synthesis andcircularization with T4 DNA ligase.37I 2FIG 13 Evaluation of the site directed mutagenesis product, M13mp18-CBD.PTIIEGR by Stu I digestion. LANE 1 : Nucleic acid MW standards,2-Hind III fragments. LANE 2 : M13mp18-CBD.PTIIEGR undigested.Lane 3 : M13mp18-CBD.PT/IEGR - Stu I fragment, 7.7 kbp (Markedwith an arrow).38(D) Construction of M13mp18-CBD.PT-gD.1(SEE FIGURE 14)E. coli JM110 was transfected with the r phage M13mp18-CBD.PT.IEGR. The RF form of the phage DNA was extracted anddigested with Stu I. The gD.1 fragment was prepared as described above,but digested with Nar I, instead of Sst I, in order to delete the TM region(Fig. 15). The 0.86 kbp gD.1 fragment was isolated from a 0.8% agarosegel using Gene CleanTM. The Nar I end was blunted by a Klenow reaction,and the two fragments were ligated. The ligation mix was used totransform competent E. coli JM101 cells. After the addition of soft agarand freshly growing E. coli JM101 cells, the mixture was plated on LBAmp plates and incubated at 37°C.The transformants were screened both by restriction endonucleaseanalysis and colony blotting. Forty-two plaques were picked randomly;their DNA was extracted by the alkaline lysis method and linearized byHind III digestion. Two of the clones gave the right sized fragment (8.5kbp), while forty others gave the size of the vector (7.7 kbp).Approximately one hundred other transformants were lysed on theplate, with chloroform, and their protein contents were transferred ontonitrocellulose membrane. The transfer was not as successful as it wouldhave been in colony blots, because the soft agar overlay containing theplaques tended to stick to the nitrocellulose membrane and peel off. Theblots were incubated with rabbit anti-gD.1 antiserum and developed withalkaline phosphatase goat anti rabbit antibodies. There was little colourdifference between the transfers, all being extensively pale. Nevertheless,eight of the plaques that gave relatively darker colour reactions werepicked and screened by restriction endonuclease analysis. Two gave theblunt^blunt^bluntlac Z'^Amp r^lac Z' L CBD PT^gD.10.86 kbp(i) Ligation39Fig. 14Construction ofM13mp 18-CBD .PT- gD .1pUC18-gD .1SstIHindIIISstI HindIIIMl3mp18-CBD .PT.IEGRStuITM^LM13mp1824 mer primerDNA polymeraseMung bean nucleaseNanDNA polymerase40FIG 15 Preparation of the gD.1 gene fragment using the primerextention method. LANE 1 : Last stage of preparation of the gD.1fragment; mung bean nuclease-treated Ml3mp18-gD.1 ds DNA ater Nar Idigestion (The 0.86 kbp gD.1 fragment is marked with an arrow). LANE2 : Nucleic acid MW standards, 2k-Hind III fragments.41right sized fragment (8.5 kbp) upon digestion with Hind III, while theother six gave the size of the vector (7.7 kbp). The positive constructswere further examined by digesting with EcoR I and gave 7.2 and 1.4 kbpfragments, corresponding to M13mp 18 and CBD.PT-gD.1, respectively(Fig. 16).The junctions of the positive constructs were sequenced using thedideoxy-sequencing method of Sanger et al. (1977). The universal primerand the specially synthesized primer 5'-GTC CAG GAC CGG AAG GTCTTT GCC-3' were used for sequencing the carboxyl terminus and thejunction of CBD.PT and gD.1, respectively. One of the constructscontained the wrong region of the gD.1 gene. The other three constructscontained the correct gD.1 fragment in the right orientation, but thefragment was lacking its first two codons (AAA TAT). This was attributedto excess digestion by mung bean nuclease.(E) Expression of CBD.PT-gD.1 by E. coli JM101 (M13mp18-CBD.PT-gD.1)E. coli cells harbouring Ml3mp18-CBD.PT-gD.1 were grown inLB medium containing ampicillin (100 µg/ml) and IPTG (0.1 mM) at37°C. The crude cell extracts were prepared either by sonication or boilingthe cells directly in the SDS-PAGE loading buffer.Western blots were performed using anti-CenA polyclonal and anti-gD.1 monoclonal (A74) antibodies. Secondary antibodies used were GoatAnti-Rabbit IgG - Alkaline Phosphatase conjugate and Biotinylated GoatAnti-Mouse IgG, respectively.The fusion protein CBD.PT cenA-gD.1 was expressed in E. coli, butin a severely degraded fashion (Fig. 17). The degradation was mainly in42FIG 16 Restriction digest analysis of M13mp18-CBD.PT-gD.1. LANE 1: Nucleic acid MW standards, X-Hind III fragments. LANE 2 : Nucleic acidMW standards, X -Hind III and EcoR I fragments. LANE 3 : Ml3mpl9digested with Hind III (7.2 kbp) LANE 4 : M13mp18-CBD.PT-gD.1digested with Hind III (8.5 kbp) is marked with an arrow (top) . LANE 5 :Ml3mp18-CBD.PT-gD.1 digested with EcoR I (7.2 and 1.4 kbpfragments) are marked with arrows (middle and bottom).43gD.1 fragment, while the CBD.PTcenA fragment remained mostly intact.The major band revealed with Anti-CenA antibodies was a degradationproduct at the size of CBD.PTcenA (apparent molecular mass, 20.0 kDa)(Fig. 17 B, lanes 3, 4) , The anti-gD.1 antibodies reacted with a whole setof degradation products (Fig. 17 C, lanes 3, 4). Several modifications ofthe growth or sonication conditions; such as adding the IPTG 1 to 2 hoursbefore harvesting, lowering the growth temperature to 30°C, increasing theamount of protease inhibitor PMSF upto 4 mM or including EDTA (3mM)in the sonication buffer (KPO 4), did not increase the amount of intactfusion protein. Phase-contrast microscope analysis revealed no evidence ofinclusion bodies (fusion proteins). Absorption of the soluble fraction withAvicel showed that, the fusion protein and some of the degradationproducts bound specifically to cellulose (Fig 17).(F) Subcloning of the CBD.PT-gD.1 gene into pUC12The CBD.PT-gD.1 gene was subcloned into pUC12 for expression.M13mp18-CBD.PT-gD.1 and pUC12 were digested with Hind III and SStI. The 1.4 kbp CBD.PT-gD.1 gene and the linearized pUC12 wererecovered from a 0.8 % agarose gel and ligated. The ligation mix was usedto transform competent E. coli JM101 cells. The cells were plated onLB-Amp-IPTG-X-gal plates and the plates were incubated at 37°C.DNA from six transformants was extracted and digested with EcoRI and BamH I to elucidate the orientation of the insert in the constructs.The release of a 2 kbp fragment from these constructs was observed on anagarose gel suggesting that the inserts were in the correct orientation (datanot shown).44A.^ B.FIG 17 SDS-PAGE analysis of the fusion protein CBD.PT-gD.1 producedby M13mp18-CBD.PT-gD.1. LANE 1 : Protein MW standards. LANE 2 :Ml3mp18 / E. coli JM101 crude cell extract. LANE 3 : Ml3mp18-CBD.PT-gD.1 I E. coli JM101 crude cell extract. LANE 4 : M13mp18-CBD.PT-gD.1 / E. coli JM101, avicel-bound fraction of the cell extract.LANE 5 : Ml3mp 18-CBD.PT-gD.1 I E. coli JM101, avicel-unboundfraction of the cell extract. (A). Western blot using anti-CenA antibodies.(B). Western blot using anti-gD.1 antibodies. (C). The gel was stained withcoomassie blue (see next page). MW standards correspond to myosin,rabbit muscle (200 kDa); B-galactosidase, E. coli (116 kDa); phosphorilaseB, rabbit muscle (97.4 kDa); albumin, bovine (68 kDa); pyruvate kinase,rabbit muscle (57.2 kDa); glutamate dehydrogenase, bovine liver (53 kDa);ovalbumin (43 kDa); alcohol dehydrogenase, equine liver (41 kDa);glyceraldehyde-3-phosphate dehydrogenase, rabbit muscle (36 kDa);carbonic anhydrase (29 kDa), 13-lactoglobulin (18.4 kDa), lysozyme (14.3kDa). (The fusion protein and the degradation products are marked witharrows).45C.46III EXPRESSION OF CBD.PT-gD.1 BY E. COLI (pUC12-CBD.PT-gD .1)(A) Colony BlotsForty-nine of the colonies containing recombinant plasmids and apUC12 / E. coli JM101 colony as negative control were picked onto LB-Amp-IPTG-X-gal plates and grown overnight at 37°C. The plates werethen inverted over chloroform-saturated paper disks for 15 min at roomtemperature. The nitrocellulose membranes were then placed on top of thelysed colonies, and the protein contents were transferred. The membraneswere then removed and screened with polyclonal anti-CenA andmonoclonal anti-gD.1 antibodies (A74).All of the clones except the negative control, gave positive resultswith both antibodies, showing that the fusion protein was expressed (Fig18).(B) Western BlotsWestern blots were carried out as described in the materials andmethods section.The expression of the fusion protein was similar to that fromM13mp1 8-CBD.PT-gD .1 . Its apparent molecular mass (53.5 kDa),relative to standard proteins, was greater than the molecular mass predictedfrom its primary structure (45.9 kDa). The PT box was previously shownto cause anomolous electrophoretic migration (Gilkes et al., 1989).47FIG 18 Screening for pUC12-CBD.PT-gD.1 construct using colonyblotting method. Fourty-nine transformant E. coli JM101 colonies and asingle colony of pUC12 / E. coli JM101 (marked with "-"), were replicaplated in triplicate. One of the plates was kept as the master copy while thecolonies on other plates were lysed by chloroform. Their protein contentswere transferred onto nitrocellulose membranes and screened for theproduction of CBD.PT-gD.1 using anti-CenA (A), and anti-gD.1 (B)antibodies.48Numerous degradation products were detected by both anti-CenA and anti-gD.1 antibodies when the crude cell extracts were prepared by directextraction of cells in SDS loading buffer. This suggests that the degradationarose from in vivo proteolysis (Fig.s 19, 20). When the cell extracts wereprepared by sonication, low levels of fusion protein were seen. The anti-CenA antibodies revealed a degradation product at the size of CBD.PT(apparent molecular mass 20.0 kDa). Several larger degradation productswere detected with anti-gD.1 antibodies.The protease inhibitors PMSF(0.1-1 mM), TLCK (1 mM), EDTA (3 mM) and pepstatin A (0.1 mM)were included in the sonication buffers (KPO4 , 50 mM) to inhibit E. coliproteases, and the cell extract preparations were carried out at —4°C.These precautions did not lessen the degradation observed, supporting theindication that the proteolysis was occuring in vivo. The amount of fusionprotein and the degradation products were significantly higher if theinduction was late (2 hrs before harvesting) as opposed to early (at the timeof inoculation). The least amount of fusion protein was observed with thecultures that were grown at 37°C and induced early. Growth at 30°Cimproved the level of expression slightly. When the cultures were inducedlate, the amount of fusion protein observed was similar at bothtemperatures, although the extent of degradation was higher at 37°C (Fig.20). The soluble cell fraction obtained by sonication contained muchlesser of the fusion protein than the cell extracts prepared by direct boiling(Fig.s 19, 20). This finding indicated that the fusion protein was associatedwith the insoluble cell fraction. Inclusion bodies could not be seen byphase-contrast microscopy analysis. To locate the fusion protein inside thecell; the cell extacts were subjected to a low speed centrifugation aftersonicating, and then the supernatant was subjected to a high speedcenrifugation (Fig. 21). The supernatant from high speed centrifugationi.e. the soluble fraction (cytoplasmic and periplasmic) contained mainly the49FIG 19 SDS-PAGE analysis of the fusion protein CBD.PT-gD.1 producedby pUC12-CBD.PT-gD.lor M13mp18-CBD.PT-gD.I. The cells wereresuspended in SDS-PAGE loading buffer and boiled (Lanes 1-8); orresuspended in KPO4 buffer and disrupted by sonication (Lanes 9-11), andthe soluble material was subjected to Avicel binding assay. The cultureswere induced at 0 hrs (Lanes 2,3,5,6) or at 15 hrs of growth (Lanes 4, 8,9, 10, 11). LANE 1 : Protein MW standards. LANE 2 : pUC12 IE. coilJM101 crude cell extract. LANE 3 : pUC12-CBD.PT-gD.1 / E.coliJM101 crude cell extract. LANE 4 : Same as Lane 3. LANE 5 :M13mp18-CBD.PT-gD.11 E. coli JM101, crude cell extract. LANE 6 :50pUC18-CBD.PT I E. coli JM101 crude cell extract. LANE 7 : ProteinMW standards. LANE 8 : Same as Lane 3. LANE 9 : Same as Lane 3.LANE 10 : pUC12-CBD.PT-gD.1 / E. coli JM101, avicel-bound fractionof cell extract. LANE 11 : pUC12-CBD.PT-gD.1 / E. coli JM101, avicel-unbound fraction of cell extract. (A). Western blot using anti-CenAantibodies. (B). Western blot using anti-gD.1 antibodies. (The fusionprotein is marked with an arrow).51FIG 20 Western blot analysis of the fusion protein CBD.PT-gD.lproduced by pUC12-CBD.PT-gD .1 , using anti-CenA antibodies.LANE 1: Protein MW standards. LANES 2-4 : Crude cell extract (CCE) of theculture grown at 30° C and induced at 15 hrs of growth. LANE 2, CCEprepared by boiling. LANE 3, CCE prepared using sonication (solublefraction). LANE 4, Culture medium. LANES 5-7 : CCE of the culturegrown at 30° C and induced at 0 hr of growth. LANE 5, CCE prepared byboiling. LANE 6, CCE prepared using sonication (soluble fraction). LANE7, Culture medium. LANES 8-10: CCE of the culture grown at 37° C andinduced at 15 hrs of growth. LANE 8, CCE prepared by boiling. LANE 9,CCE prepared using sonication (soluble fraction). LANE 10, Culturemedium. LANES 11-13 : CCE of the culture grown at 37° C and inducedat 0 hrs of growth. LANE 11, CCE prepared by boiling. LANE 12, CCEprepared using sonication (soluble fraction). LANE 13, Culture medium.(The fusion protein is marked with an arrow).52A.^ B.FIG 21 SDS-PAGE analysis of the fusion protein CBD.PT-gD.l producedby pUC12-CBD.PT-gD.1. After sonication, the cell extract wascentrifuged sequentially at forces of 1,090 X g and 109,000 X g (i. e., afterthe first centrifugation, the supernatant fluid was decanted and cetrifuged atthe higher centrifugal force.) LANE 1 : Protein MW standards. LANE 2 :High speed centrifugation (109,000 X g) supernatant, corresponding to thecytoplasmic and periplasmic fraction. LANE 3 : Low speed centrifugation(1,090 X g) pellet, corresponding to the unbroken cells and inclusionbodies. LANE 4 : High speed centrifugation (109,000 X g) pellet,corresponding to the membrane fragments. (A). The gel was stained withcoomassie blue. (B). Western blot using anti-CenA antibodies. (The fusionprotein is marked with an arrow).5320 kDa degradation product. Both the low speed pellet corresponding tothe inclusion bodies and unbroken cells, and the high speed pelletcorresponding to the membrane fragments, contained the fusion proteinand the degradation products, suggesting that the fusion protein might beassociated with the membranes. The culture supernatant also contained asmall amount of the fusion protein (Fig. 22). However, when the cultureswere harvested at the late log phase, rather than late stationary phase, thefusion protein could not be detected (Fig. 23). Fusion protein in the culturesupernatant may thus result from the release of intracellular proteinsfollowing cell death and lysis. The amount of fusion protein obtained fromthe soluble cell extracts was estimated to be —0.1 1.1g, m1 -1 of culture (bycomparing the fusion protein bands to those of the standards on coomassieblue-stained gels) . When the cells were harvested at late stationary phasethere was —0.05 lag, m1-1 of the fusion protein in the culture supernatant.The timing of induction (Fig. 23) and the temperature (Fig. 20)modifications did not increase the amount of the fusion protein obtainablefrom these fractions, although the amount of the degradation products werealtered. Avicel binding assays carried out on the soluble fraction and theculture supernatant showed that the fusion protein and several of itsdegradation products bound specifically to cellulose, and by doing so, wereconcentrated (Fig. 22). This effect was most significant with the majordegradation product detected by the anti-CenA antibodies (apparentmolecolar mass 20.0 kDa) which probably corresponds to the intactCBD.PTcenA • Growth curves of cultures induced at various times weremeasured to test for possible toxicity of the fusion protein to the host cells(Fig.s 24, 25). Induction with IPTG did not affect the growth of the cellsshowing that the fusion protein was not toxic to the E. coli host. Toovercome the in vivo degradation occuring in E. coli JM101 cells, theconstruct pUC12-CBD.PT-gD.1 was used to transform several protease54FIG 22 SDS-PAGE analysis of the fusion protein CBD.PT-gD.1 producedby pUC12-CBD.PT-gD.1. The cells were disrupted by sonication. LANE1 : Protein MW standards. LANE 2 : Crude cell extract. LANE 3 : Avicel-bound fraction of the cell soluble cell extract. LANE 4 : Avicel-unboundfraction of the cell soluble cell extract. LANE 5 : Same as lane 3. LANE 6: Same as lane 4. LANE 7 : Culture medium. LANE 8 : Avicel-bound55fraction of the culture medium. LANE 9 : Avicel-unbound fraction of theculture medium. LANE 10 : Same as lane 8. LANE 11 : Same as lane 9.(A). The gel was stained with Coomassie blue. (B). Western blot usinganti-CenA antibodies. (C). Western blot using anti-gD.1 antibodies. (Thefusion protein is marked with an arrow).56A.FIG 23 SDS-PAGE analysis of the fusion protein CBD.PT-gD.1 producedby pUC12-CBD.PT-gD.1 induced at various times.The culturesupernatants and the soluble cell fractions were subjected to Avicel bindingassay. The bound fractions were ran on the gel. LANE 1 : Protein MWstandards. (LANE 1* of the coomassie-stained gel corresponds to a secondset of MW standards). LANES 2, 3, 4, 5, Avicel-bound cell extracts.Thecultures were induced at 0, 5, 9 hrs, or none, respectively. LANES 6, 7, 8,9, Avicel-bound culture supernatants.The cultures were induced at 0, 5, 9hrs, or none, respectively. (A). The gel was stained with Coomassie blue.(B). Western blot using anti-CenA antibodies. (The fusion protein and themajor degradation product are marked with arrows).112^3^4^5^61 8 9 1072.00.057HoursFig. 24. The growth curves of E. coli JM101 cultures harbouring theplasmid pUC12-CBD.PT-gD.1 grown at 37° C and induced at differenttimes.1 5 62^3^4Hours730°C IPTG 0 hr30°C IPTG 5 firs30°C IPTG none37°C IPTG 0 hr37°C IPTG 5 firs37°C IPTG none2.01 .00.058Fig. 25. The growth curves of E. coli JM101 cultures horbouring theplasmid pUC12-CBD.PT-gD.1 induced at different times and grown atdifferent temperatures.59deficient mutant strains. E. coli strains KS 476, UT 5600, BL 21, CAG626, CAG 597, CAG 629, deficient of deg P, omp T, omp T, lon, htpR,lon and htpR respectively, were transformed. The amount of fusionprotein obtained from these strains were less than that obtained from theoriginal host E. coli JM101 (Fig.s 26, 27).60FIG 26 Western blot analysis of the fusion protein CBD.PT-gD.1produced by pUC12-CBD.PT-gD.1, using anti-CenA antibodies. LANES1 and 2; E. coli AB 1899 (pUC12-CBD.PT-gD.1), crude cell extract, andculture medium respectively. LANES 3 and 4; E. coli UT 5600 (pUC12-CBD.PT-gD.1), crude cell extract, and culture medium, respectively.LANES 5 and 6; E. coli CAG 626 (pUC12-CBD.PT-gD.1), crude cellextract, and culture medium, respectively. LANES 7 and 8; E. coli CAG597 (pUC12-CBD.PT-gD.1), crude cell extract, and culture medium,respectively. LANES 9 and 10; E. coli CAG 629 (pUC12-CBD.PT-gD.1), crude cell extract, and culture medium, respectively. LANE 11 :Protein MW standards. (The fusion protein is marked by an arrow).61FIG 27 SDS-PAGE analysis of the fusion protein CBD.PT-gD.1 producedby pUC12-CBD.PT-gD.1 .The soluble cell extracts were obtained usingsonication. LANE 1 : Protein MW standards. LANES 2-6 : Crude cellextracts. LANE 2 : E.coli JM 101 (pUC12-CBD.PT-gD.1). LANE 3 :Same as lane 2. LANE 4 : E.coli BL 21 (pUC12-CBD.PT-gD.1). LANE5 : E.coli KS 476 (pUC12-CBD.PT-gD.1). LANE 6 : E.coli UT 5600(pUC12-CBD.PT-gD.1). LANES 7-10: Culture media. LANE 7 : E.coliJM 101 (pUC12-CBD.PT-gD.1). LANE 8 : E.coli BL 21 (pUC12-CBD.PT-gD.1). LANE 9 : E.coli KS 476 (pUC12-CBD.PT-gD.1).LANE 10 : E.coli UT 5600 (pUC12-CBD.PT-gD.1). (A). Western blotusing anti-CenA antibodies. (B). The gel was stained with Coomassie blue.62IV IMMUNIZATION RESULTSThe overall objective of this thesis was to test for an adjuvant effect(if any) when an antigen was bound to cellulose. The use of an antigenfused to a CBD would facilitate this approach. Initially it was planned touse a CBD-gD.1 fusion for these studies. However, because of the difficultyin expressing adequate amounts of this hybrid protein in E. coli, it wasdecided to test the concept using the CBDcenA-PhoA fusion protein (Fig. 3).CBDcenA-PhoA comprises the cellulose binding domain of CenA and thePT box , followed by the 40 N-terminal as of the catalytic domain fused toE. coli alkaline phosphatase (PhoA). The fusion protein binds to Avicel,and can be eluted with water. It can be purified in a single step by affinitychromatography on Avicel (Greenwood et al., 1989).(A) Experiment I1. ImmunizationFour groups of 4 female, CBA mice were immunized twice at a 2-wk interval with 20 Ill of the hybrid protein CBDcenA-PhoA or AlkalinePhosphatase presented in four alternative formulations, i.e. Alk.Pho. (E.coli), CBDcenA-PhoA, CBDcenA-PhoA adsorbed on Avicel, CBD cenA-PhoA adsorbed on Alum. A fifth group of mice served as unimmunizedcontrols. Animals were bled 1 week after the second immunization toA.P.CBD-PhoACBD-PhoA-AvicelCBD-PhoA-AlumControl, no treatm.0.80.70.60.50.40.30.20.10.01.0631.5^2.0^2.5^3.0Log ELISA TitersFig. 28 Anti-Alkaline Phosphatase Antibodies3.5^4.0^4.5-A- CBD-PhoA-w- CBD-PhoA-Avicel- - CBD-PhoA-Alum-o- Control641.71.61.51.41.31.21.11 1.0Fr 0.9.,t1 0.8A 0.70 0.6..ri8' 0.50.40.30.20.10.01 .0^1.5^2.0^2.5^3.0^3.5^4.0^4.5Log ELISA TitersFig. 29 Anti-CenA Antibodies65determine the secondary antibody response.2. ELISA resultsAlkaline phosphatase or CenA specific antibodies in mice sera weremeasured by ELISA using CenA or Alkaline Phosphatase as the coatingantigen. See Materials and Methods for the ELISA procedure. The meanoptic densities of each group is plotted against the log ELISA titers (Fig.s28 and 29), and the mean titers are summerized in Table 2. (The reportedtiter corresponds to the reciprocal of the dilution that produced anabsorbance at 490 nm equal to 50% of the maximal absorbance value ofeach serum).(i) Anti-Alkaline Phosphatase ResponseThe anti-alkaline phosphatase antibody titers measured 1 week afterthe second immunization are summerized in Table 2. Animals that receivedCBDcenA-PhoA-Avicel, developed the highest antibody titers (mean : 382).Animals that received CBDcenA-PhoA-Alum developed the second highestantibody titers (mean : 269), followed by those obtained with alkalinephosphatase (mean : 71); which were 70% and 19% of the titer obtainedwith CBDcenA-PhoA-Avicel, respectively. Those that received CBDcenA-PhoA, or did not receive any treatment, had antibody titers that were 10%and 7% of the level seen with CBD cenA-PhoA-Avicel, respectively (meansof 39 and 27, respectively).(ii) Anti-CenA ResponseThe anti-CenA antibody titers measured 1 week after the secondimmunization are summarized in Table 2. Animals that received CBDcenA-66Table 2Exp. 1 - Antibody responses in mice immunized with fusion proteinbound to Avicel or controls.Group Immunogen n route ELISA Titeranti-APELISA Titeranti-CenA1 AP 4 i. p . 71 ± 29a NAb2 CBDcenA-PhoA 4 i .p . 39 ± 10 60 ± 82 fl3l2cenA-PhoA-Avicel 3 i,p 382 ± 115 637 ± 4704 CBDcenA-PhoA-Alum 4 i.p. 269 ± 46 333 ± 1105 Control, no treatment 4 NA 27 ± 6 28 ± 2a Mean + SDb NA, not applicable67PhoA-Avicel developed the highest antibody titers (mean : 637); followedby those obtained with CBDcenAPhoA-Alum which was 52% of the titerobtained with CBD cenA-PhoA-Avicel (mean ; 333). Animals that receivedor CBDcenA-PhoA, or did not receive any treatment had antibody titers thatwere 9% and 4% of the titer obtained with CBD cenA-PhoA-Avicel (meansof 60 and 28, respectively).(B) Experiment 21. ImmunogensFive groups of immunogens were prepared:(i) Alkaline Phosphatase (E. coli)(ii) Alkaline Phosphatase (E. coli)-Avicel(iii) CBDcenik-PhoA(iv) CBDce,,A-PhoA-Avicel(v) CBDcenA-PhoA-AlumThe alkaline phosphatase and the CBDcenA-PhoA proteins in variousimmunogen preparations were analyzed using SDS-PAGE (Fig. 30) andCBDcenA-PhoA was shown to be attached to avicel or alum, completely.The immunogen alkaline phosphatase-Avicel was prepared exactly asCB D cenA -PhoA-Avicel, except the mixture was not washed after thebinding step as there was no actual binding. This control immunogen wasdesigned to compare the immunogenicity of the antigen bound to Avicelversus the antigen mixed with Avicel, i.e., to elucidate whether binding is68FIG 30 SDS-PAGE analysis of the fusion protein CBD.PTcenA-PhoA oralkaline phosphatase in various immunogen preparations. LANE 1 :Protein MW standards. LANE 2 : CBD.PT cenA-PhoA-Avicel preparation,suspended.(The band runs at 70 kDa corresponds to the intact fusionprotein. The band that runs at 40 kDa corresponds to the PhoA fragment ofthe fusion protein. The presence of the 40 kDa fragment is due to theproteolysis at the junction between the fusion partners). LANE 3 :CBD.PT cenA -PhoA-Avicel preparation, supernatant. LANE 4:CBD.PTcenA-PhoA-Avicel preparation, suspended. LANE 5 : Same as Lane3. LANE 6 : CBD.PTcenA-PhoA preparation. LANE 7 : CBD.PT cenA-PhoA-Alum preparation, supernatant. LANE 8 : Alkaline phosphatase-Avicel preparation, suspended. LANE 9 : Alkaline phosphatasepreparation. The gel was stained with coomassie blue. (The fusion proteinis marked with the arrow at the top, and the alkaline phosphatase is markedwith the arrow at the bottom).1.5 3.0 3.5 4.02.0^2.51.0A.P.A.P.-AvicelCBD-PhoACBD-PhoA-AvicelCBD-PhoA-AlumControl, no treatm.0.50.40.10.0I69Log ELISA TiterFig. 31 Primary Anti-Alkaline Phosphatase Antibodies4.02.0^2.51.5 3.0 3.51.0-A- CBD-PhoA-,- CBD-PhoA-Avicel-4- CBD-PhoA-Alum-o- Control, no treatm.0.50.4I'-' 0.31.4n°in, 0.200.10.070Log ELISA TiterFig. 32^Primary Anti-CenA AntibodiesA.P.- A.P.-Avicel- CBD-PhoA-v- CBD-PhoA-Avicel-0- CBD-PhoA-Alum-0- Control, no treatm.0.5 -0.2 -0.1 -0.0711.0^1.5^2.0^2.5^3.0^3.5^4.0Log ELISA TiterFig. 33^Secondary Anti-Alkaline Phosphatase Antibodies-A- CBD-PhoA-- CBD-PhoA-Avicel-,- CBD-PhoA-Alum-o- Control, no treatm.721.00.90.80.7IR 0.6,...^0.5oA)C.)^0.4.ria00.30.20.10.01.0 1.5^2.0^2.5Log ELISA Titer3.0^3.5^4.0Fig. 34 Secondary Anti-CenA Antibodies73crucial for the response. For detailed explanations for the preparations seematerials and methods.2. ImmunizationFive groups of 5 female, CBA mice were immunized with alkalinephosphatase or the hybrid protein CBDcenA-PhoA presented in fivealternative formulations, while another five were not injected. Animalswere bled 2 weeks after the first, and 1 week after the secondimmunization to determine the primary and secondary antibody responses,respectively.3. ELISA resultsThe primary and secondary alkaline phosphatase or CenA specificantibodies in mice sera were measured by ELISA using CenA or alkalinephosphatase as the coating antigen. (See Materials and Methods for theELISA procedure). The mean optic densities of each group is plottedagainst the log ELISA titers (Fig.s 31,32,33,34). The mean ELISA titersare summerized in Table 3, and the individual titers are shown in figures35, 36, 37 and 38. (The reported titer corresponds to the reciprocal of thedilution that produced an absorbance at 490 nm equal to 50% of themaximal absorbance value of each serum).The reported titer corresponds to the reciprocal of the dilution thatproduced an absorbance at 490 nm equal to 50% of the maximalabsorbance value of each serum.74Table 3Exp. 2 - Primary response of mice immunized with fusion proteinbound to Avicel or controls.Group Immunogen n route ELISA Titeranti-APELISA Titeranti-CenA1 AP 5 i.p. 39+13a NAb2 AP-Avicel 5 i.p. 38±7 NA3 CBDcenA-PhoA 5 i.p. 33 ±7 30±34 CH3Ic) enA-PhoA-Avicel 5 Lp 53+24 32±35 5 i.p. 38±3 25±2CBDcenA-PhoA-Alum6 Control, no treatment 5 NA 29±11 25±4a Mean + SDb NA, not applicableTable 4Exp. 2 - Secondary response of mice immunized with fusion proteinbound to Avicel or controls.Group Immunogen n route ELISA Titeranti-APELISA Titeranti-CenA1 AP 5 i.p. 221+92a NAb2 AP-Avicel 5 i.p. 385±254 NA3 CBDcenA-PhoA 5 i.p. 222±104 50±124 CBDCenA-PhoA-Avicel 5 i.p. 1310±997 119±695 CBDcenA-PhoA-Alum 5 i.p. 350±261 68±56 Control, no treatment 5 NA 29±11 25 ±4a Mean SD.b NA, not applicable.759—8 — oo7 —.',:`4 6 —^o5 —^0 0•.z o o oo,-.1w43—^•0—^0• o•o o0o00o • o00 0802 — 0 0 001 —0^AP^AP^CBD.PhoA CBD.PhoA CBD.PhoA Control-Avicel -Avicel^-Alum no treatmentFig. 35 Primary anti-alkaline phosphatase antibody titers elicited byimmunization of mice with alkaline phosphatase or CBDcenA-PhoApresented with the indicated formulations. Open circles representthe ELISA titers of individual mice. Filled circles represent the meanof each group.76987L'a 6I-2!" 5:=1432000%000 c.0•0 8 go00o010 '0 CBD.PhoA CBD.PhoA-AvicelCBD.PhoA-AlumControlno treatmentFig. 36 Primary anti-CenA antibody titers elicited by immunizatonof mice with CBDcenA-PhoA presented with the indicated formulations.Open circles represent the ELISA titers of individual mice. Filledcircles represent the mean of each group.77320300 0280260240220200180160 0140 0 •12010080 0 060 040200o•oo 080 090000•0 0AP^AP CBD.PhoA CBD.PhoA CBD.PhoA Control-Avicel^-Avicel^-Alum no treatmentFig.37 Secondary anti-alkaline phosphatase antibody titers elicitedby immunization of mice with alkaline phosphatase or CBDcenA-PhoApresented with the indicated formulations. Open circles represent theELISA titers of individual mice. Filled circles represent the mean ofeach group.782624 02220185'.k16 01412 •.'-4-I10ral 8 oo 00 .0 06 00o• 0o4 002 0 0 20 00 '0^CBD-PhoA CBD-PhoA-AvicelCBD-PhoA-AlumControlno treatmentFig. 38 Secondary anti-CenA antibody titers elicited by immunizationof mice with CBDcenA-PhoA presented with the indicated formulations.Open circles represent the ELISA titers of individual mice. Filled circlesrepresent the mean of each group.79(i) Primary Anti-Alkaline Phosphatase ResponseThe anti-Alkaline Phosphatase antibody titers measured 2 weeks afterthe first immunization are shown in Figure 35 and summarized in Table 3.Primary anti-Alkaline Phosphatase response was weak. Animals thatreceived CBDcenA-PhoA-Avicel developed the highest antibody titers(mean : 53). Animals that received Alkaline Phosphatase, AlkalinePhosphatase-Avicel or CBDcenA-PhoA-Alum showed antibody titers thatwere 74%, 72% and 72% of the titer obtained with CBDcenA-PhoA-Avicel(means of 39, 38 and 38 respectively). The lowest antibody titer wasdeveloped by CBDcenA-PhoA which was 60% of the level seen withCBDcenA-PhoA-Avicel (mean : 32).(ii) Primary Anti-CenA ResponseThe anti-CenA antibody titers measured 2 weeks after the firstimmunization are shown in Figure 36 and summarized in Table 3. Theprimary anti-CenA response was very weak. The highest atibody titerswere developed by animals which received the CBD cenA-PhoA-Avicel(mean : 32). Animals that received CBDcenA-PhoA or CBDcenA-PhoA-Alum had antibody titers that were 94% and 78% of the titer obtained withCBDcenA-PhoA-Avicel (means of 30 and 25, respectively).(iii) Secondary Anti-Alkaline Phosphatase ResponseThe anti-alkaline phosphatase antibody titers measured 1 week afterthe second immunization are shown in Figure 37 and summarized in Table4. Animals that received CBD cenA-PhoA-Avicel, developed the highestantibody titers (mean : 1310). Animals which received alkalinephosphatase-Avicel developed the second highest antibody titers (mean :385), followed CBDcenA-PhoA-Alum (mean : 350); 29% and 26% of theCBDcenA-PhoA-Avicel titre, respectively. Those that received CBDcenA-80PhoA or Alkaline Phosphatase, had antibody titers that were 17% of thelevel seen with CBD cenA -PhoA-Avicel (means of 222 and 221,respectively).(iv) Secondary Anti-CenA ResponseThe anti-CenA antibody titers measured 1 week after the secondimmunization are shown in Figure 38 and summerized in Table 4. Anti-CenA antibody titers were much lower than anti-alkaline phosphatase titers.Animals that received CBD cenA-PhoA-Avicel developed the highestantibody titers (mean : 119). Animals that received CBD cenAPhoA-Alum orCBDcenA-PhoA had titers that were 57% and 42% of the titer obtained withCBDcenA-PhoA-Avicel (means of 68 and 50, respectively).4. Statistical analyses of experiment 2The significance of the data was calculated by the ANOVA (analysisof variance) and Tukey H.S.D. multiple comparisons tests using the Statmstatistical software program.The ANOVA test compares the differences within the groups to thedifferences between the groups, and tells whether the latter is significant. Ifthe F-ratio is higher than the value given for those DF values (at the"percentage points of the F distribution : upper 5% points" table in astatistics book), there is significance between the groups (for more detailedexplanations see page 88).The Tukey H.S.D. multiple comparisons test calculates the P valuespairwise, and tells which of the groups are significantly different fromeach other. The P value indicates the probability that the difference is81significant. For a 95% confidence level, the P value has to be 0.05, i.e.,(1-P) x 100 has to be 95%.(i) Primary Anti-Alkaline Phosphatase Response (Table 5)ANOVA revealed no significant differences between groups; i.e.,immunogenicities of various antigen preparations. [The F-ratio (1.646) islower than the value (2.62) given at the "percentage points of the Fdistribution : upper 5% points" table, and the P value (0.186) is higherthan 0.05].(ii) Primary Anti-CenA Response (Table 6)ANOVA revealed significant differences between groups. [The F-ratio (5.559) is higher than the value (3.24) given at the "percentage pointsof the F distribution : upper 5% points" table, and the P value (0.008) islower than 0.05]. Tukey matrix of pairwise comparison probabilitiesshowed that, at a 95% confidence level, antibody titers obtained fromanimals that received the CBDcenA-PhoA-Avicel were significantly higherthan those obtained with CBD cenA-PhoA-Alum or no treatment. [The Pvalues of the pairs 2-3 and 2-4 ( 0.019, each) are lower than 0.05].(iii) Secondary anti-Alkaline Phosphatase Response (Table 7)ANOVA revealed significant differences between groups. [The F-ratio (4.317) is higher than the value (2.62) given at the "percentage pointsof the F distribution : upper 5% points" table, and the P value (0.006) islower than 0.05]. Tukey H.S.D. multiple comparisons showed that, at a95% confidence level, antibody titers obtained from animals that receivedthe CBDcenA-PhoA-Avicel were significantly higher than those obtainedwith Alkaline Phosphatase, CBDcenA-PhoA, CBDcenA-PhoA-Alum or notreatment [The P values of the pairs 4-1, 4-3, 4-5, and 4-6 ( 0.019, 0.019,820.048 and 0.004, respectively) are lower than 0.05].(iv) Secondary Anti-CenA Response (Table 8)ANOVA revealed significant differences between groups. [The F-ratio (5.068) is higher than the value (3.24) given at the "percentage pointsof the F distribution : upper 5% points" table, and the P value (0.012) islower than 0.05]. Tukey H.S.D. multiple comparisons showed that, at a95% confidence level, antibody titers obtained from animals that receivedthe CBDcenA-PhoA-Avicel were significantly higher than those obtainedwith no treatment [The P value of the pair 2-4 (0.008) is lower than 0.05].83Table 5 PRIMARY ANTI-ALKALINE PHOSPHATASE RESPONSEANALYSIS OF VARIANCESource Sum-of-squares^DF^Mean-square^F-ratio^PGroup^1676.967^5^335.393^1.646 0.186Error^4890.000 24^203.750TUKEY HSD MULTIPLE COMPARISONSMATRIX OF PAIRWISE COMPARISON PROBABILITIES1^2 3 4 51 1.0002 1.000^1.0003 0.973^0.992 1.0004 0.650^0.541 0.240 1.0005 1.000^1.000 0.992 0.541 1.0006 0.873^0.935 0.999 0.127 0.9351 Alkaline Phosphatase2 Alkaline Phosphatase-Avicel3 CBDcenA-PhoA4 CBDcenA-PhoA-Avicel5 CBDcenA-PhoA-Alum6 Control, no treatment84Group : Variation between treatments.Error : Variation within treatments.SUM-OF-SQUARES (within-treatment):k^n tSR =^E I (yu - Yd 2t=i i=1k = number of treatments ( k=4 and 6, for anti-CenA and anti-alk.phos.,respectively.nt = number of observations within the tth tratment (5 mice in eachgroup).yti = ith observation in the tth treatment.MEAN-SQUARE (within-treatment):, 2SR =^SR / N-kN = ni+ n2+^+ nk .N-k = DF (Degrees of freedom)(=5+5+5+5+5+5-6 = 24, and 5+5+5+5-4 = 16, for anti-alk.phos. and anti-CenA, respectively).SUM-OF-SQUARES (between treatments):kST=^.1, nt (yt - y) 2t=1y = grand average, sum of all observations devided by the total number ofobservations.85yt = average of the tth treatment.MEAN-SQUARE (between treatments):2ST =^ST / k- 1k-1= DF (Degrees of freedom).k-l= 6-1=5 and 4-1=3, for anti-alk.phos. and anti-CenA, respectively.k = number of treatments.F-RATIO: Mean-square (between treatments) / Mean-square (withintreatments).P: Probability that the difference is significant.(1-P) x 100 = % probability.If P = 0.186, as in Table 5, then the % probability is:(1 - 0.186) x 100 = 81%.86Table 6 PRIMARY ANTI-CEN A RESPONSEANALYSIS OF VARIANCESource Sum-of-squares^DF^Mean-square^F-ratio^PGroup^182.600^3^60.867^5.559 0.008Error^175.200 16^10.950TUKEY HSD MULTIPLE COMPARISONSMATRIX OF PAIRWISE COMPARISON PROBABILITIES1 2 31 1.0002 0.667 1.0003 0.166 0.019 1.0004 0.166 0.019 1.0001 CBDcenA-PhoA2 CBDcenA-PhoA-Avicel3 CBDcenA-PhoA-Alum4 Control, no treatment87Table 7 SECONDARY ANTI-ALKALINE PHOSPHATASE RESPONSEANALYSIS OF VARIANCESource Sum-of-squares^DF^Mean-square^F-ratio^PGroup^5151223.867^5^1030244.773^4.317 0.006Error^5728106.800^24^238671.117TUKEY HSD MULTIPLE COMPARISONSMATRIX OF PAIRWISE COMPARISON PROBABILITIES1^2 3 4 51 1.0002 0.994^1.0003 1.000^0.995 1.0004 0.019^0.062 0.019 1.0005 0.998^1.000 0.998 0.048 1.0006 0.988^0.855 0.988 0.004 0.9011 Alkaline Phosphatase2 Alkaline Phosphatase-Avicel3 CBDcenA-PhoA4 CBDcenA-PhoA-Avicel5 CBDcenA-PhoA-Alum6 Control, no treatment88Table 8 SECONDARY ANTI-CEN A RESPONSEANALYSIS OF VARIANCESource Sum-of-squares^DF^Mean-square^F-ratio^PGroup^23766.550^3^7922.183^5.068 0.012Error^25010.400 16^1563.150TUKEY HSD MULTIPLE COMPARISONSMATRIX OF PAIRWISE COMPARISON PROBABILITIES1 2 31 1.0002 0.060 1.0003 0.891 0.210 1.0004 0.756 0.008 0.3541 CBDcenA-PhoA2 CBDcenA-PhoA-Avicel3 CBDcenA-PhoA-Alum4 Control, no treatment89DISCUSSIONCLONING AND EXPRESSIONThe plasmid pUC18-CBD.PT-gD.1 was constructed, and expressedin E. coli. Upon induction with isopropyl-B-D-thiogalactopyranoside, lysisof the bacterial cultures was observed probably due to the deleterius effectsof the highly hydrophobic sequence at the C-terminus of the gD.1, thetransmembrane region (TM). Very hydrophobic as sequences such as thesignal peptide or the TM region of membrane-bound eucaryotic proteinsare often toxic to the bacterial host (Amann et al.,1984a; Brosius, 1984;Remaut et al., 1983; Rose and Shafferman, 1981; Yelverton et al., 1983).We postulated that, for the expression of gD.1 in the bacterial host, theremoval of not only the signal sequence, but also the TM sequence wascrutial.Another plasmid, pUC12-CBD.PT-gD.1, lacking the TM sequenceof gD.1 was constructed in three steps. Initially, CBD.PT gene wascloned as pUC18-CBD.PT in E. coli. A high level of expression wasobserved upon induction with IPTG. The CBD.PT protein was purified andused for adsorption studies to cellulose (Din et al., 1991; Gilkes et al., inpreparation). Secondly, CBD.PT gene was subcloned into M13mp18, andmanipulated in vitro to introduce the Factor Xa recognition site and aunique restriction endonuclease site to give the recombinant phageM13mp18-CBD.PT-IEGR. This construct provided a "cassette" for onestep creation of fusion proteins with CBD.PT . Fusion proteins containingthe cellulose-binding domains have proven useful for protein purificationand enzyme immobilization (Ong et al., 1989). Their potential use for90immune enhancement is demonstrated here. This cassette construct has beenused; digestion with Factor Xa, and subsequent release of the purifiedforeign protein was demonstrated (Shen, pers. comm.). Thirdly, gD.1gene fragment obtained by in vitro manipulations to remove the signal andthe TM sequences, was ligated into the "cassette" M13mp18-CBD.PT-IEGR to give the recombinant phage M13mp18-CBD.PT-gD.I.Following fusion of these genes, CBD.PT-gD.1 was subcloned into pUC12for expression in E. coli. Expression was observed upon induction withIPTG. The fusion protein was no longer toxic to the host cells indicatingthat the toxicity previously seen was due to the TM region of gD.1 aspostulated. The fusion protein could be purified by binding to cellulose.However, much of the fusion protein appeared to be associated with thehost membranes, and furthermore, severely degraded due to in vivoproteolysis. The amount in the soluble fraction was insufficient for theproduction of material for immunization. The eucaryotic region of therecombinant protein, i.e., gD.1, was the target of the attack of E. coliproteases. Attempts to protect the recombinant protein from proteolysisfailed. Severe degradation of gD.1 in E. coli has been shown by severalgroups (Weis et al., 1983; Amann et al., 1984; Steinberg et al., 1986).The affinity of the CBD.PTcenA-gD.1 for the membrane might berelated to the presence of the CenA leader peptide at the N-terminus of thefusion protein. This leader peptide allows the transport of CenA to theperiplasm of E. coli (Wong et al., 1986). Export to the periplasmfacilitates the recovery of recombinant proteins in an active form. Theseproteins are sometimes even released to the culture media. In this case,export of the gD.1 fusion was incomplete, much of it remaining associatedwith the membrane. A possible solution of this problem would be toexclude the leader sequences from the genetic construct confining therecombinant protein in the cytoplasm. This could be achieved by fusing the91recombinant protein sequences to an N-terminal fragment of a procarioticgene. This approach was followed by Steinberg et al. (1986) who made aseries of X cro::gD.1 gene fusions. None of the plasmids encoding Xcro::gD.1 gene fusions were found to yield significant amounts ofmaterial upon induction with isopropyl-B-D-thiogalactopyranoside. Thissuggests that although the solution described above might preventmembrane association, instability would continue to be a major problem.Nevertheless, flanking on the C-terminus as well as the N-terminus by aprocaryotic protein might protect gD.1 from proteolysis. CBD of C. fimiexoglucanase (CBDcex) which exists at the C-terminus of the enzyme mightbe chosen to make the fusion protein instead of CBD cenA used here. Atripartite fusion protein, such as X cro::gD.1::CBD cex , would be suitable totest the hypothesis. Tripartite fusions with gD.1 containing the entire B-Galhave been obtained as inclusion bodies, and proven stable (Weis et al.,1983; Watson et al., 1983; Amann et al., 1984; Steinberg et al., 1986).Therefore, an alternative would be the construction of X,cro::gD.1::CBD cex ::fl-gal gene fusion. The corresponding fusionprotein might be stable in E. coli as inclusion bodies. After the recovery,the B-Gal could be cleaved off by Factor Xa digestion, and the fusionprotein purified by binding to cellulose.The use of mammalian cell lines as the host for gD.1 expression hasproven advantageous in terms of glycosylation, lack of in vivo proteolysisand secretion (Lasky et al., 1984). Therefore, the best approach would beto switch from the bacterial expression system to a mammalian expressionsystem.92IMMUNOLOGYThe adjuvant effect when an antigen is bound to cellulose via theCBD was shown by immunizing mice with CBD cenA-PhoA adsorbed onAvicel (CBDcenA-PhoA-Avicel). CBDcenA-PhoA-Avicel showed the highestprimary and secondary anti-alkaline phosphatase and anti-CenA antibodytiters (means of 53 and 1310 for the anti-alkaline phosphatase; and, 32 and119 for the anti-CenA antibody titers, respectively) (Tables 3 and 4).Second highest secondary anti-alkaline phosphatase antibody titers weredeveloped by alkaline phosphatase-Avicel control (mean : 385), whichmight indicate a weak non-specific stimulation by Avicel. Alum, the onlyajuvant licenced to be employed in human vaccines, showed a relativelypoor performance (mean : 350). The effect of Alum has shown to bedependent on the type of the antigen, the type of the animal, or the route ofadministration (Bomford, 1986). The antibody titers obtained with alkalinephosphatase and CBD cenA-PhoA were similar (means of 221 and 222respectively). Whether presenting alkaline phosphatase as a fusion proteinhad no effect on its immunogenicity, or the positive effect was neutralizedby the loss of essential epitopes of alkaline phosphatase upon fusion, isunknown.The overall anti-CenA response was poorer than the anti-alkalinephosphatase response (means of secondary anti-CenA antibody titers were119, 68, and 50; for CBDcenA-PhoA-Avicel, CBDcenA-PhoA-Alum andCBDcenA-Ph0A, respectively). This may simply indicate that CBDCenA isless immunogenic than alkaline phosphatase. In the preliminary study (exp.1), the overall anti-CenA response was found to be stronger than the anti-alkaline phosphatase response. This differrence between the results may bedue to the fact that, another batch of CBDcenA-PhoA was used in that study.93The alkaline phosphatase forms dimers, so does the fusion proteinCB DcenA-PhoA. The dimer formation involves interaction between thePhoA regions of the fusion proteins, whereas the CBDCenA regions are notinvolved. The fusion polypeptide CBDCenA-PhoA is sensitive to proteolysisat the junction between the fusion partners. Both of them bind to celluloseif one or both monomers carry a CBD, e. g. CBDCenA-PhoA / CBD cenA-PhoA, or CBDCenA-PhoA / CBDCenA (Greenwood, J. M., personal comm.).Different batches of antigen may contain different ratios of thesefragments depending upon the degree of degradation.We demonstrate here a novel adjuvant system, which is economical,and shows no apparent toxicity to the animals. Its action mechanism is yetto be discovered, but the depot hypothesis proposed for Alum may beeffective here, as well. The system must be tested with different antigens,as the efficiency of adjuvants may alter from antigen to antigen. Thisdepends on the original characteristics of the antigen, as well as the three-dimensional changes that result from fused to another protein, i. e., acellulose binding domain. Fusion proteins obtained with cellulose bindingdomains show differences in their elution properties. Some of the fusionproteins, such as CBDCenA-PhoA used in this study, can be eluted withwater, whereas others require guanidine-HC1 to be eluted. The differencesin the elution properties may also affect the degree of immune stimulationthat can be obtained with various fusion proteins, as it might affect thepersistence of the antigen after administration.94REFERENCESAmann, E., Broker M., Wurm, F. In Chanock, R. M. and Lerner,R. A. (Eds.) Modern approaches to vaccines : Molecular andChemical basis of Virus Virulance and Immunogenicity,Cold Spring Harbor Laboratory, Cold Spring Harbor, NewYork, pp. 183-187 (1984a)Aprile, M. A., and A. C. Wardlaw. Can J. Public Health 57 : 343 (1966)Atkinson, T., and Smith, M. Gait, M. J., ed., pp. 35-81, IRL Press,Washington, D.C. (1984)Berman, P. W., T. Gregory, D. Crase, and L. A. Lasky. Science 278 :290 (1985)Birnboim, H. L., Doly, J. Nucl. Acids Res. 7, 1513- 1523 (1979)Bomford, R., Clin Exp. Immunol. 39 : 435. (1980)Bomford, R. In Animal Cell Biotechnology, New York, AcademicPress (1985)Bomford, R. Progress Towards Better Vaccines. Oxford University Press177-185 (1986)Brosius, J. Gene 27, 161-172 (1984)Cohen, G. H., M. Katze, C. Hydrean-Stern, R. J. Eisenberg. J. Virol.27, 172 (1978)Cooper, S. and Ruettinger, T. Mol. Gen. Genet. 139, 167-176 (1975)Corey, L. and P. G. Spear. N. Engl. J. Med. 314 : 686 (1986)Corey, L., G. Mertz, R. Ashley, and J. Beneditti. In Proceedings of27th Interscience Conference on Antimicrobials andChemotherapeutics, October 4-7 NY. American Society forMicrobiology, Washington, DC, Abstr. 1229, p. 313. (1987)Earhart, C. F., M. Lundrigan, C. L. Pickett, and J. R. Pierce. FEMSMicrobiol. Lett. 6:277-280 (1979)95Edelman, R. Rev. Infect. Dis. 2 : 370 (1980)Eisenberg, R. J., Long, D., Hogue-Angeletti, R., and Cohen, G. H. J.Virol. (1984)Garoff, H., A. M. Frischauf, K. Simons, H. Lehrach, H. Delius. Nature288, 236 (1980)Gilkes, N. R., Kilburn, D. G., Langsford, M. L., Miller Jr., R. C.,Wakarchuk, W. W., Warren, R. A. J., Whittle, D. J. andWong, W. K. R. J. Gen. Microbiol. 130, 1377-1384 (1984a)Gilkes, N. R., Warren, R. A. J., Miller, R. A. J., jr, and Kilburn, D. G.J. Biol. Chem. 263, 10401-10407 (1988)Gilkes, N. R., Kilburn, D. G., Miller, R. C., Jr., and Warren, R. A. J.J. Biol. Chem. 264, 17802-17808 (1989)Gilkes, N. R., Jervis, E., Henrissat, B., Tekant, B., Miller, R. C., Jr,Warren, R. A. J., and Kilburn, D. G. in preparation.Glenny, A. T., Pope, C. G., Waddington, H. and Wallace, U. J. Path.Bact., 29, 31 (1926)Greenwood, J. M., Gilkes, N. R., Kilburn, D. G., Miller, jr, andWarren, R. A. J. FEBS Lett. 244, 127-131 (1989)Grodberg, J. and Dunn, J. J. J. Bacteriol. 170, p. 1245-1253 (1988)Langsford, M. L., Gilkes, N. R., Singh, B., Moser, B., Miller, R. C. jr,Warren, R. A. J., and Kilburn, D. G. FEBS Lett. 225, 163-167 (1987)Lasky, L. A., Dowbenko, D., Simonsen, C., Berman, P. W. ModernApproaches to Vaccines. Cold Spring Habor Laboratory 189-195(1984)Maniatis, T., Fritsch, E. F., and Sambrook, J. Molecular cloning : Alaboratory manual. Cold Spring Harbor Laboratory, ColdSpring Harbor, New York (1982)Marinus, M. G., and Morris, N. R. J. Bacteriol. 114, 1143-1150(1973)May, M. S., and Haman, S. J. Bacteriol. 123, 768-770 (1975)96Meignier, B., T. M. Jourdier, B. Norrild, L. Pereira, and B. Roizman.J. Infect. Dis. 155 : 921 (1987)Messing, J. Recombinant DNA Technical Bulletin, NIH Publication No. 79-99, Vol. 2, No. 2. 43-48 (1979)Messing, J. Methods Enzymol. 101, part C, 20 (1983)Nahmias, A. J., Dannenbarger, J., Wickliffe, C., Muther, J. in TheHuman Herpesviruses : An Interdisciplinary Perspective,(Elsevier, New York, 1981)Norrid, B., Curr. Top. Microbiol. Immunol. 90, 67 (1980)Ong, E., Greenwood, J. M., Gilkes, N. R., Kilburn, D. G., Miller, R.C., Jr., and Warren, R. A. J. Trends in Biotechnology. Vol.7, No. 9, 239-243 (1989)Pereira, L., T. Klassen, J. R. Baringer, ibid. 29, 724 (1980)Remaut, E., Stanssens, P. and Fiers, W. Nucl. Acids Res. 11, 4677-4688 (1983)Rizman, B., and D. Furlong, Compr. Virol. 3, 244 (1974)Rose, J. K., and Shafferman, A. Proc. Natl. Acad. Sci. USA 78, 6670-6674 (1981)Sanchez-Pescador, L., Burke, R. L., Ott, G., Nest, G. V. J. Immunol. 141,1720-1727 (1988)Sanger, F., Niclen, S., and Coulson, A. R. Proc. Natl. Acad. Sci. USA74, 5463-5467 (1977)Sanger, F., Coulson, A. R., Hong, G. F., Hill, D. F., Petersen, G. B. J.Mol. Biol. 162 : 729-773 (1982)Showalter, S. D., M. Zweig, B. Hampar, Infect. Immun. 34, 684(1981)Steinberg, D. A., Watson, R. J., Maiese, W. M. Gene 43, 311-317(1986)Strauch, K. L., Johnson, K., Beckwith, J. J. Bacteriol. 171, 2689-269697(1989)Thomson, T. A., J. Hilfenhaus, H. Moser, and P. S. Morahan. Infect.Immun. 41 : 556 (1983)Turk, J. L., and D. Parker. J. Invest. Dematol. 68 : 336 (1977)Vieira, J. and Messing, J. Gene 19 : 259-268 (1982)Watson, R. J., J. H. Weis, J. S. Salstrom, L. W. Enquist. Science 218,381 (1982)Watson, R. J., Gene 26, 307-312 (1983)Weis, J. H., Enquist, L. W., Salstrom, J. S., and Watson. Nature 302,72-74 (1983)Wong, W. K. R., Gerhard, B., Guo, Z. M., Kilburn, D. G., Warren, R.A. J., and Miller, R. C. jr. Gene 44, 315-324 (1986)Yanish-Perron, C., Vieira, J., and Messing, J. Gene33, 103-119 (1985)Yelverton, E., Norton, S., Obijeski, J. F., and Goeddel, D. V. Science219, 614-620 (1983)

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0086481/manifest

Comment

Related Items