UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Identification and cloning of the genes required for production of the adhesive holdfast organelle of… Yun, Chanyoung 1992

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

Item Metadata


831-ubc_1992_fall_yun_chanyoung.pdf [ 1.97MB ]
JSON: 831-1.0086652.json
JSON-LD: 831-1.0086652-ld.json
RDF/XML (Pretty): 831-1.0086652-rdf.xml
RDF/JSON: 831-1.0086652-rdf.json
Turtle: 831-1.0086652-turtle.txt
N-Triples: 831-1.0086652-rdf-ntriples.txt
Original Record: 831-1.0086652-source.json
Full Text

Full Text

IDENTIFICATION AND CLONING OF THE GENES REQUIRED FOR PRODUCTION OF THE ADHESIVE HOLDFAST ORGANELLE OF THE MARINE CAULOBACTER MCS6 By CHANYOUNGYUN B.Sc, Yonsei University, Seoul, Korea 1988 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Microbiology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1992 © CHANYOUNG YUN, 1992 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, Iagree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publicatioN of this thesis for financial gain shall not be allowed without my written permission. (Signature) Department of Microbiology The University of British Columbia Vancouver, Canada Date Aug. 3 1 , 1992 DE-6 (2/88) ABSTRACT Caulobacters produce an organelle called a holdfast which allows the bacterium to firmly attach to surfaces. Tn5 insertion mutagenesis was used to identify genes affecting holdfast production or function in the marine strain MCS6. 12,000 Tn5-insertion mutants were produced and screened for adhesion defects by a newly developed assay involving the growth of cultures in polystyrene microtitre dish wells and detection of attached cells by staining with crystal violet. Among adhesion-defective mutants, those with multiple polar (pleiotropic) defects were excluded and the remainder were examined for the ability of Wheat Germ agglutinin to label the holdfast region, a feature of the wild-type holdfast. 41 mutants were isolated that produced no detectable or synthesized a reduced amount of holdfast. Southern blot and pulsed field gel electrophoresis (PFGE) analyses of these mutants indicated 11 unique sites of Tn5 insertion, clustered in three regions of the genome. In addition, 71 mutants were found that did not adhere to polystyrene or adhered poorly yet still produced a holdfast organelle as judged by Wheat Germ agglutinin binding. Southern blot and PFGE analyses of 15 of these mutants showed eight Tn5 insertion sites clustered in two regions of the genome. Glass slides treated with silane chemicals (producing surfaces with varying degrees of hydrophobicity and/or hydrophilicity) were used to attempt characterization of this phenotype; unexpectedly no generic pattern of differences in binding was found between the mutants and wild-type Caulobacters. In particular, no reduction in the ii ability of the mutants to bind to hydrophobic surfaces was noted; this might have been expected considering the inability to bind polystyrene. As a means of confirming their unique location and character of each genomic region, representatives of all 5 clusters were chosen and the Tn5-interrupted regions were cloned. The gene segments obtained were in turn used as probes to identify corresponding holdfast-related chromosome regions in a cosmid library of wild-type MCS6 DNA. Positive cosmid clones were transferred into the Tn5 -derived holdfast-defective mutants by conjugation. Among 6 groups tested, 4 of them showed almost full complementation, one showed poor complementation; for one group no complementation was obtained. In sum, several genetic regions responsible for holdfast mediated attachment in marine Caulobacter MCS6 were identified by Tn5 insertion mutagenesis. They were located at several discrete locations on the genome and, via the cloned segments obtained, are now accessible for further genetic analysis. i i i TABLE OF CONTENTS Page Abstract ii List of Figures vi List of Tables vii Abbreviations viii Acknowledgments ix Introduction 1 Materials and Methods 7 Bacterial Strains and Growth Conditions 7 Production of the Tn5 mutant library 7 Analysis of the Tn5 library 1 1 Development of an assay for isolation of holdfast-defective mutants 1 4 Isolation of holdfast-defective mutants 1 4 Analysis of holdfast-defective mutants 1 6 (i) Determining the sites of Tn5 insertion. 1 6 (ii) Determining the phenotypes of holdfast-defective mutants. 1 7 Cloning of holdfast-encoding, Tn5 inserted DNA 2 0 Comparing the holdfast-related gene of MCS6 and Caulobacter crescentus CB2A 2 2 A cosmid library of MCS6 DNA 2 2 iv (i) Production of the library. 2 2 (ii) Complementation testing. 2 3 Results 2 6 Production of Tn5-mutant library 2 6 Development of an assay for holdfast-defective mutants of MCS6 2 7 Detection of holdfast-defective mutants 2 9 Analysis of holdfast-defective mutants 3 1 (i) Determining the sites of Tn5 insertion. 3 1 (ii) Phenotype evaluation of holdfast-defective mutants. 3 6 Cloning of holdfast-related; Tn5 inserted DNA 4 0 Similarity between CB2A and MCS6 DNAs 4 3 Complementation tests 4 3 Discussion 4 9 References 56 v LIST OF FIGURES Figure Page 1. The Tn5 element and probing of Tn5-mutated DNA. 12 2. A Southern blot of Tn5-mutated chromosomal DNA. 2 8 3. A Southern blot comparison of Tn5 containing fragments among holdfast-defective mutant groups. 3 2 4. A Southern blot comparison of Tn5 containing fragments among holdfast-defective mutant groups. 3 3 5. Scoring of cells bound to chemically treated glass slides 3 8 6. A Southern blot comparison of cloned holdfast related DNA fragments with chromosomal DNA digests of holdfast-defective mutant groups that contain linked sites of Tn5 insertion. 4 2 7. A Southern blot of comparison of cloned holdfast related DNA fragments with chromosomal DNA digests of holdfast-defective mutant groups that contain linked sites of Tn5 insertion. 4 4 8. Complementation analysis of cosmid clones using microtitre plate assay. 4 7 vi LIST OF TABLES Table Page 1. Bacterial strains, plasmids and phage used in this study. 8 2. Silanization procedures for covalent modification of slide glasses. 1 9 3. Phenotype analysis of holdfast-defective mutants. 3 0 4. The size of the Tn5 containing chromosomal DNA fragments from the holdfast-defective mutants. 3 5 5. Summary of clones derived from Tn5 interrupted regions. 41 6. Complementation Test. 4 6 vii ABBREVIATIONS DNA EDTA FITC-WGA GlcNac h kb kV LPS MCS min a PFGE RNA sarcosyl SDS SSC SSPYE TE Tn Tris Hi uPD WC deoxyribonucleic acid ethylenediaminetetra-acetic acid fluorescein isothiocyanate-conjugated Wheat Germ agglutinin N-acetylglucosamine hours kilobases kilovolts lipopoly saccharides marine Caulobacter strain minutes ohms pulsed-field gel electrophoresis ribonucleic acid Na+-N-lauryl sarcocine sodium dodecyl sulfate 0.15M Sodium Chloride and 0.015M Sodium Citrate (pH 7.0) sea salt peptone yeast extract lOmM tris-HCl (pH8.0) ImM EDTA transposon tris (hydroxymethyl) aminomethane microliter microfarad wettability coefficient viii ACKNOWLEDGMENTS Thank God for His blessings. I am sincerely thankful to Dr. John Smit for accepting me into his lab, for generously providing direction for three years and for extensive editing of this thesis. I thank Dr. Bert Ely of the University of South Carolina for performing the pulsed-field gel electrophoresis analysis. Special thanks are due to Karoline Lee who prepared the Tn5 mutant library and helped me in bench work, Dr. Harry D. Kurtz, Jr. who provided good advice at numerous stages, and Stephen George Walker and Dr. Wade Bingle who were helpful in providing good advice for my work and thesis. Many thanks to my parents for their love and for supporting me spiritually. Finally, I would like to thank my lovely husband, Yoonsuk Park for his spending all the time with me and for his unwavering support. This thesis is dedicated to all the people who have encouraged me to do this work. ix INTRODUCTION The attachment of bacteria to submerged solid surfaces is of major importance in aquatic environments. For free-living bacteria in nutrient-deficient systems, such as oligotrophic waters and soils, surfaces appear to provide sites where growth and survival can take place (19). Attachment to surfaces in contact with flowing systems such as ships, platforms, dams, pipelines, and heat exchangers ensures that the microbial population is not eliminated by washout. Often this property results in serious fouling problems. For most adhesive bacteria, which cause fouling problems in aquatic environments, adherence occurs by a variety of largely uncharacterized mechanisms (17, 32, 59). Some of these bacteria have been shown to adhere to surfaces, initially in a reversible association, and then eventually in an irreversible adhesion. Reversible adhesion is a time-independent, instantaneous attraction by long-range physical forces which hold bacteria near a surface. The bacteria continue to exhibit Brownian motion and in experimental situations, they can be removed from the surface by the shearing effects of a water jet. In this state, even violent rotational movement of the flagellum can dislodge the bacterium. Long range forces between the counter ions which are loosely attracted to the inert surface and the cell surface appear to mediate the reversible adhesion process. On the other hand; irreversible adhesion is defined as a time-dependent firm adhesion in which the bacteria no longer exhibit 1 Brownian motion and cannot be removed by washing. Polymer bridging by any component external to the cell envelope, such as extracellular polysaccharide polymers, pili, fimbrae, holdfast or flagella can be responsible for the firm anchoring of bacteria to surface seen in irreversible adhesion (17, 32, 33, 59). The adhesion related properties of polysaccharides originating from bacteria in general have been reviewed (59). The most popular general theory is that two types of polymers exist in the adhesion process. Some of the polymers act to keep surfaces together temporarily and are considered to be responsible for initial attachment. Other polysaccharides produced after initial adhesion, provide permanent matrices in which polymers might be cross-linked to form a rigid structure. Chemical analysis of polysaccharides isolated from fouling organisms have mainly identified mannose, glucose, galactose, glucosamine, rhamnose, fucose and glucuronic acid (58, 59). However, it is not clear in these studies whether the polysaccharide analyzed was truly responsible for the bacterium's adhesive capabilities Physicochemical studies of bacterial adhesion suggest that the characteristics of the substratum (surface) (10, 18, 44, 49) and the suspending medium as well as the physiological state of bacteria themselves may be important variables influencing the adhesion processes (17, 19, 32, 59). Surface properties such as surface free energy, hydrophobicity (wettability) and surface charge have been shown to affect bacterial attachment. Most of the bacteria tested have shown a preference for moderately (but not extremely) 2 hydrophobic surfaces. Bacterial adhesion appears also to depend on medium conditions, such as pH, electrolyte concentrations and carbon source. As well, understanding of adhesion mechanism can be confused by the physiological state of bacteria because they are dynamic, living organisms whose physiology can cause their adhesive properties to change over time (17, 32, 58, 59). With respect to the ecological significance of surfaces to aquatic bacteria, two major groups of bacteria are recognized based on their nutrient requirements (42). These are the oligotrophs, capable of growth in low-nutrient environments, and the copiotrophs which require relatively high levels of nutrients for growth. In the initial adhesion process (initial colonization of a surface) in aquatic environments, growth of copiotrophs may lead to a rapid utilization of nutrients accumulating at the interface, thereby creating a nutrient deficient status on the surface that may allow the oligotrophs a competitive advantage (33). Caulobacters are aerobic, oligotrophic, Gram negative, chemoheterotrophic bacteria which are easily isolated from a wide range of locales, including most sources of freshwater, sea water and moist soils (41). Caulobacters are dimorphic bacteria that have an asymmetric pattern of cell division in which two structurally different progeny cells are produced; a motile swarmer cell that carries a single polar flagellum and a nonmotile stalked cell that has a cellular stalk formed by an outgrowth of the cell wall and membranes. The stalk is capable of lengthening in response to nutrient deficient conditions (43). Because of this pattern of growth, 3 Caulobacters have been studied as a model system for investigating both the temporal and spatial regulation of cell differentiation events (37, 38, 50). The ecological role of Caulobacters is likely to use up the few remaining organic matter after such organisms as Pseudomonas have finished their growth (13). Caulobacters, along with several other adhesive bacteria, participate in biofouling phenomena by the production of the holdfast organelle. This structure is positioned at the base of the flagellum in swarmer cells and at the end of the cellular stalk following differentiation of swarmer cells to stalked cells (41, 43). There is no indication that the holdfast serves any other role in addition to adhesion to surfaces (35, 40). The presence of a few polar pili may assist the adhesion process but it seems to be mediated mainly by holdfast. These polar structures and the ability to survive under low nutrient conditions, provides Caulobacters with a full set of functions to live as an ideal biofilm bacterium in dilute nutrient environments (41, 43). Only a small amount of holdfast adhesive is produced by Caulobacters, and the material apparently maintains attachment for the organism for many generations, even in places where significant shear from water flow might be expected (35, 41), suggesting that the holdfast mediates strong, stable adhesion. We argue that these characteristics make the holdfast a good model to define the molecular means used to accomplish strong adhesion. Also an understanding of how Caulobacters attach to surfaces may have 4 broad applicability in evaluation the problems associated with microbial fouling. Little is known about the chemical composition of the holdfast, but all available evidence suggests the holdfast is a complex polysaccharide. Ruthenium red staining properties indicate that it may be an acidic polysaccharide (60), although current chemical analysis has so far failed to confirm that supposition (N. Ravenscroft and J. Smit, unpublished studies). Competition lectin binding experiments between Caulobacter strains that bind Wheat Germ agglutinin (WGA) at their holdfast and oligomers of N-acetyl-glucosamine (GlcNac) demonstrated (35) that stretches of contiguous GlcNac residues occur in the holdfast. This suggests that at least a region within the holdfast is composed of oligomeric GlcNac. However, in a study on the effects of proteolytic and glycolytic enzymes, chitinase and lysozyme resulted in significant degradation of freshwater, but not marine, Caulobacter holdfasts. This suggests that possible regions of oligomeric GlcNac in marine holdfasts are altered with respect to those of freshwater Caulobacters (35). Although oligomers of GlcNac (essentially chitin) are not known for their adhesive properties per se, it is interesting that a GlcNac oligosaccharide secreted by Rhizobium sp. is known to be involved in attachment to and subsequent infection of leguminous plants (29). In previous work, mutants defective in adhesion were generated by UV mutagenesis (40) and Tn5 insertion mutagenesis in C. crescentus CB2A. For the Tn5 insertions, the mutations were clustered in four unlinked groups (36). One mutant group exhibiting 5 a shedding phenotype was examined in considerable detail. The shed holdfast was fully able to attach to surfaces and to the polar region of wild-type cells. A three-gene cluster (hfaAB and hfaC) was found to be involved in attachment of the holdfast to the cell. It was suggested that the protein encoded by the hfaA locus may have a direct role in the attachment of the holdfast to the cell, whereas hfaB may be involved in the positive regulation of hfaC (28). It was concluded that there is a specific mechanism of holdfast attachment to the polar region of the cell, that the attachment appears to be an adhesive event (40), and that a specific genetic region is responsible for this effect (36). We suspect there may be a comparable polar complex and gene set in marine Caulobacter MCS6. The data presented in this study are the first steps in the evaluation of the location and complexity of the genes affecting holdfast production or function in a marine strain, MCS6. 6 MATERIALS AND METHODS Bacterial Strains and Growth Conditions Caulobacters and E. coli strains used in this report are listed in Table 1. Except where noted, Caulobacter crescentus was grown at 30°C in a peptone-yeast extract medium (PYE) (41) with 0.01% CaCl2 and 0.02% MgSO^ and marine Caulobacter strains were grown in 1.5% sea salt peptone-yeast extract medium (1.5% SSPYE) at 25°C. Luria broth medium (31) was used for growing E. coli strains at 37°C. E. coli HB101 cells used for in vitro packaging were grown in LB medium with 0.4% maltose. Antibiotic concentrations used for both E. coli and Caulobacters were: streptomycin, 20 |ig/ml (Sm20); tetracycline 10|ig/ml (TetlO); chloramphenicol 20 u.g/ml (Cm20), for E. coli, 2 |ug/ml (Cm2), for Caulobacter; kanamycin, 50 |ig/ml (Km50); rifampicin, 10 |ig/ml (RflO) and nystatin, 10 | ig/ml (NylO). Bacteriophage T7, used to kill residual donor E. coli cells after conjugation, is a lytic coliphage, propagated in E. coli C600 by standard procedures (31). Production of the Tn5 mutant library Tn5-generated mutations in marine Caulobacter MCS6 were made by the method of Ely and Croft (14). 12,000 Tn5 insertion mutants were produced in order to guarantee 95% confidence that all genes were interrupted by Tn5; this assumes that the marine Caulobacter genome size is approximately the same size as freshwater C. crescentus (4000 kb) (15), that transposition is a 7 Table 1. Bacterial strains, plasmids and phage used in this s tudy . Strain Genotypes Reference E.coli DH5oc F", recAl, hsdRll, supE44, thi-l, endAl, (22) gyrA96, reiki, 08Od/acZAM15, A(/acZYA-argF)U169, X~, pUC plasmid host F"> hsdS20 (rB", mB"), supE44, ara\4, (6) galKl, lacYl, prokl, rpsL20 (Strr), ry/15, leu, mtll, recA13, mcrB, mrr, thi~{l) F", thi-l, thr-l, leuB6, sulll, reck, RP4-2- (51) Tc::Mu, plasmid transfer (tra+) functions cloned into chromosome F~> endkl, thi-l, hsdRll (rk",mk+), reck (11) F", thi-l, thr-l, leuB6, lacYl, tonkll, (23) supE44, X~, general host HB101 SM10 MM294 C600 Caulobacter MCS6 MCS6::Tn5 MCS6-gpl to gpl6 & 7-260 mutants CM260 wild type of marine Caulobacter, Rfr Random Tn5 library mutants, Rfr, Kmr, Sm r Tn5-induced holdfast-defective mutant groups of MCS6, Rfr, Kmr, Sitf a marine Caulobacter holdfast-defective mutant Caulobacter Surface layer-minus variant of wild type crescentus CB2 CB2A P l a s m i d s pBR322 Ap r , Tcr pBR322-Neo r 1.8 kb Hindlll/Bamm fragment of Tn5 cloned into pBR322, Apr, Kmr pSUP2021 pBR325 with Tn5 inserted into Tcr gene and mob site of RP4, Apr, Cmr, Kmr, Mob+ (1) this study this study J. Poindex-ter (52) (4) J. Swindle u n p u b -lished da ta (51) 8 pPR510 Derived from pUC19 and Cmr marker of Tn7725, pBR328 Ap r . Tetr, Cm', derivative of pBR325 and pBR327 pRK2013 Kmr, ColEl replicon, mob site of RK2 pLAFR5 Tc r , broad-host range cosmid pYHFl 13.4 kb Tn5 containing EcoRI fragment cloned from mutant group gpl into the MCS of pPR510, Cmr, Kmr pYHF4 21.7 kb Tn5 containing Clal fragment cloned from mutant gp4 into the pBR328, Kmr> Ap r , Tetr, Cmr pYHF6 19.7 kb Tn5 containing EcoRI fragment cloned from mutant gp6 into the MCS of pPR510, Cmr, Kmr pYHF93 19.7 kb Tn5 containing SstI fragment cloned from mutant 93 into the MCS of pPR510, Kmr, Cmr pCYHFl cosmid clones which has ca. 23 kb MCS6 chromosomal DNA and complement holdfast-defective phenotype of gpl pCYHF6 cosmid clones which has ca. 23 kb MCS6 chromosomal DNA and complement holdfast-defective phenotype of gp6 pCYHF4 cosmid clones which has ca. 23 kb MCS6 chromosomal DNA and complement holdfast-defective phenotype of gp4 pCYHF13 cosmid clones which has ca. 23 kb MCS6 chromosomal DNA and complement holdfast-defective phenotype of gpl3 pCYHF93 cosmid clones which has ca. 23 kb MCS6 chromosomal DNA and complement holdfast-defective phenotype of 93 pCYHF126 cosmid clones which has ca. 23 kb MCS6 chromosomal DNA and complement holdfast-defective phenotype of mutant 126 P h a g e (46) (54) (16) (26) this this this this this this this this this this study study study study study study study study study study T7 Lytic coliphage (7) random event, and that the average gene size is 1 kb. (This estimate was based on the equation N=ln(l-P)/ln(l-f), where N=number of mutants, P=probability, and f=proportion of the genome represented by a single event) (31). The narrow host range plasmid pSUP2021 (51), which cannot be maintained in Caulobacters because it is unable to replicate in non-enteric bacteria, was used as a delivery vehicle for Tn5 (9) in conjugal mating of marine Caulobacter MCS6 with E. coli SM10. The donor strain SM10 carries the transfer genes of the broad host range IncP-type plasmid RP4 integrated in its chromosome. Thus pSUP2021, which contains the Inc P Mob site can be mobilized from the donor to the recipient (56). For conjugations, 108 donor cells or recipient cells (controls) and a mixture of the two (experimental) were combined in several ratios, washed, resuspended in 25 u.1 of 1.5% SSPYE and incubated overnight on sterile nitrocellulose filters (Millipore, Massachusetts). The cells were then resuspended in 0.8 ml of liquid medium and phage T7 was added at multiplicity of infection of 5-10 to lyse residual E. coli donor cells. This mixture was incubated for 1.5 h at 37°C and plated on 1.5% SSPYE (Rf20/ Km50/ Sm20/ NylO); these conditions selected against SM10 and for recovery of Caulobacter cells containing Tn5 insertions. Nystatin was generally added to eliminate an erratic contamination by yeast. After 6 days of growth, Tn5 inserted mutants of MCS6 were pooled and stored at -70°C. (This work was accomplished by Karoline Lee.) 10 Analysis of the Tn5 library Tn5 insertion mutants of MCS6 were selected using both Sm20 and Km50. Kanamycin resistance is a well-known marker for selection of Tn5 insertions while streptomycin resistance is seldom used because the streptomycin resistance determinant, which is separable from the kanamycin resistance determinant on Tn5, is not expressed in enteric bacteria. However in some non-enteric bacteria such as Caulobacters (39, 45), the marker is expressed. Therefore, the genus-specific expression of Smr should be tested whenever Tn5 is used in genetic studies. The frequency of spontaneous mutation of MCS6 to the two drugs (Km50/Sm20) as well as the frequency of Tn5 transposition was assessed by colony hybridization using a 1.8 kb Hindlll/BamHl fragment of Tn5 (Fig. 1), which was an insert in pBR322-Neor, as a probe. About 600 putative transposon-inserted colonies, along with MCS6 as a negative control and E. coli SM10/pSUP2021 as a positive control were transferred to sterile Whatman filter paper which was then washed with 0.5M NaOH for 8 min, 1M Tris-HCl for 10 min, 0.5M Tris-HCl/1.5M NaCl for 4 min, followed by baking for 1.5 h at 80°C. These filters were incubated in prehybridization solution (5x SSC, 200(ig/ml salmon sperm DNA, 0.02% Na+-N-lauryl sarcosine) for 4 h and then 106 cpm of [a-P32]dCTP-labeled probe was injected. Probes were prepared by nick-translation (34) and purified using GENECLEAN™ (Bio 101, Inc. La Jolla, CA) kit. Hybridization was done overnight at 65°C. The filters were then washed twice in lx blot washing solution (3x SSC, 5mM EDTA, 0.1% sarcosyl) for 15 min at 1 1 Tn5 = i = 3: CQ co CQ en re Tf Probe Sa/I-digested DNA Chromosomal DNA XNXNXNXNXNXVS^ mmmmmmm Chromosomal DNA Bflr/ll-digested DNA FIGURE 1. THE TN5 ELEMENT AND PROBING OF TN5-MUTATED DNA: The gray bar indicates the region of Tn5 used as a probe in all Tn5 detection experiments. The gray regions in Sail- and Z?g/II-digested DNA demonstrate the regions that hybridized with the probe. This resulted in predictably lighter and darker bands in the autoradiographs (see text for additional explanation). 12 65°C, and twice with O.lx blot washing solution for 15 min at 65°C. Then the filters were exposed to X-Omat X-ray film (Kodak, New York) at -70°C overnight. Since it has been reported that hot spots for Tn5 insertion have been found in the promoter region of the Tcr gene of pBR322 and in cloned DNA segments (3, 5, 30), the degree of randomness in the production of the Tn5 library was checked. As one criterion for estimating the randomness of Tn5 insertion, the frequency of auxotrophic mutants in the library has been investigated (36). For the C. crescentus CB2A Tn5 insertion mutant library; this approach has indicated reasonable randomness of Tn5 insertion (36). However, since a minimal medium for marine Caulobacter strains has not been defined, the frequency of auxotrophic mutation in the library could not be obtained. Instead, chromosomal DNA from the pooled Tn5 library was isolated, analyzed by Southern blot analysis to determine whether "hot spots" of Tn5 hybridization were noted and if so, whether the DNA fragment sizes in the hot spot region were the same size as the fragments of Tn5-inserted DNA detected in individual holdfast-defective mutants. The isolation of chromosomal DNA and Southern blot analyses were done as described below. In the preparation of the Tn5 library, it was possible that the vector and transposon could have integrated into the chromosomal DNA of MCS6 via non-homologous recombination (14, 51). In order to check this possibility, randomly chosen Tn5 library mutants were patched on the 1.5% SSPYE plates with Cm2 and Km50 as the Cm marker resides on the vector DNA. 13 Development of an assay for isolation of holdfast-defective mutants The cellulose acetate binding assay which was used for detecting holdfast-defective mutants of C. crescentus CB2A (36), was attempted with MCS6 as a method to find holdfast-defective mutants. It was determined that the assay did not work for MCS6, so several additional surfaces were tested. (This was accomplished as a cooperative effort with Karoline Lee). Cellulose acetate, GelBond (polyester film to support agarose gels, FMC products), glass (silanized or not), mylar, and aluminum were tested. Colonies of MCS6, as a positive control, and CM260, as a negative control, were cultured, transferred to these surfaces and then incubated for 30 min in humidified conditions. The materials were then washed with sea water, stained with Coomassie blue (0.1% in 10% isopropanol and glacial acetic acid), acridine orange (0.2%), or crystal violet (0.1%) for 10 min and checked for staining by normal lighting or UV light, as appropriate. Additional surfaces, such as Nylon and Polyethylene with wettability coefficients (the explanation of wettability is given in Table 4) of about 30 to 50 were tested using liquid cultures. 20 u.1 of mid log liquid cultures were applied to the surfaces and tested as above. Ultimately, an assay based on the use of polystyrene microtitre plates with liquid cultures was developed; this is described below. Isolation of holdfast-defective mutants Approximately 12,000 colonies from the frozen stock of the Tn5-insertion MCS6 mutant library were serially diluted and plated. 14 Isolated colonies were inoculated and grown for 2 days in the wells of the polystyrene microtitre plates (Nunc). The liquid cultures were discarded and the plates were washed with a 1.5% sea salts solution using a pressurized water dispenser. The wells were then stained with 0.1% crystal violet for 15 min and washed with distilled water. Cultures containing cells that adhered to the polystyrene wells produced a positive staining reaction (Fig. 8). Potential holdfast-defective mutants did not stick to the well walls. This resulted in a well that did not stain or one that stained faintly. Pleiotropic mutations, such as non-motile mutants or mutants with abnormalities in division were subsequently identified by phase contrast microscopy and were excluded from the group of possible holdfast-defective mutants. In order to directly assess the holdfast of possible holdfast-defective mutants; lectin binding to the holdfast was assayed using fluorescein isothiocyanate (FITC)-conjugated Wheat Germ agglutinin (WGA) (Vector Laboratories Inc. Burlingham, CA), which was known to bind wild-type holdfast, as previously described (35). 0.7 jil of FITC-WGA (5mg/ml) was added to 200 u,l of a mid-logarithmic culture and this mixture was incubated on ice for 30 min. Cells were centrifuged, resuspended with 1 ml of 1.5% SSPYE, centrifuged and resuspended in 20 (il of 50% glycerol and 2% N-propyl gallate in 1.5% seawater. Samples were examined by fluorescence and phase contrast microscopy. 15 Analysis of holdfast-defective mutants (i) Determining the sites of Tn5 insertion. Chromosomal DNA isolations, restriction enzyme digestions, and Southern blot hybridizations were done by standard methods (31). 10 ml of a mid-logarithmic (O.D.6oo=0-4-0.6) culture was centrifuged, washed by resuspension and centrifugation with 0.6 ml of lOmM Tris (pH8.0) and ImM EDTA (TE(10/1)) and resuspended in 0.6ml of TE( 100/10). Lysozyme was added to a final concentration of 300 (ig/ml and the mixture was incubated for 5 min at 37°C. SDS was then added to 1% and the mixture was incubated at 65°C for 10 min. Proteinase K was added to 300 fig/ml and the mixture was incubated at 65°C for 1 h. Nucleic acids were purified by two phenol extractions, two phenol/chloroform/isoamyl alcohol (25:24:1) extractions, one chloroform extraction, followed by ethanol precipitation. Isolated chromosomal DNA was digested with restriction enzymes that cut once (Sail) or twice (Bglll) within the Tn5 element and with enzymes that do not cut within Tn5 (Clal, EcoRI, Kpnl, and Sstl) (2, 9, 25). DNA fragments were separated by electrophoresis on a 0.4% agarose gel with TBE buffer (50mM Tris, 50mM boric acid, 2.5mM EDTA). Nucleic acid preparations were also separated by using pulsed-field gradient gel electrophoresis (PFGE), using published procedures (15, 53) after digestion with Spel or Dral enzymes, which only rarely cut the MCS6 chromosomal DNA. This procedure was done by Bert Ely (University of South Carolina). 16 For Southern blot analyses, DNA was transferred to Hybond-N-nylon membrane (Amersham) by standard procedures (47, 55). Gels were soaked in 0.25M HC1 for 25 min, rinsed with water, soaked in denaturation solution (0.5M NaOH, 1.5M NaCl) for 30 min, soaked in transfer solution (0.25M NaOH, 1.5M NaCl) for 10 min, and the DNA was then transferred by capillary action using transfer solution. The blot was then washed with 2x SSC, air-dried and baked at 80°C for 30 min. Blots were incubated in prehybridization solution (see colony lift procedure) and hybridized overnight with [a -P 3 2 ]dCTP- labe led probes prepared as described above. Blots were then washed twice with 0.3x washing solution (lx SSC, 1.6mM EDTA, 0.006% sarcosyl) for 15 min and then twice with O.lx washing solution for 10 min twice at 55°C. Then blots were dried and exposed to X-ray film. Holdfast mutants that were found by Southern blot analyses to have the same site of Tn5 insertion were collectively referred to as a mutant group and subsequent experiments were performed with a representative mutant from each group. (ii) Determining the phenotypes of holdfast-defective mutants Phenotypes of holdfast-defective mutants were differentiated according to the results of the lectin binding assay, holdfast adhesiveness to microtitre plates, and rosette formation (the capacity for holdfast-proficient cells to stick together at the holdfast region) (41). The latter was determined by phase contrast microscopy. In order to test for the holdfast shedding phenotype, where holdfast is produced but does not remain attached to the cell, a cover 17 slip attachment assay was used as previously described (36). Ethanol washed and autoclaved glass coverslips were added to 50 ml of 1.5% SSPYE and the appropriate mutant inoculum was added. When the O.D.600 was approximately 0.5, the coverslips were rescued and washed with 1.5% SSPYE. Then 0.5 ml of SSPYE containing 0.3 u.1 FITC-WGA was applied to the cover glasses. The cover slips were incubated on ice for 30 min, washed with 1.5% SSPYE, mounted on glass slides and observed by fluorescence microscopy. To investigate the phenotype of holdfast defective mutants which produced faint staining in the microtitre plate assay but still produced an apparently normal holdfast in the lectin binding assay, glass microscope slides were treated with several silanizing reagents which covalently modify the glass surface (Table 2) (8, 48), and subsequently present a variety of chemical substituents on the surface. Squares (0.5cm x 0.5cm) were scored on glass slides with a diamond pencil. The slides were then washed with ethanol and baked at 350°C for 30 min and were treated according to the procedure for a specific siloxane reagent. Treated slides were stored in a dessicator under vacuum until used. Cells of the selected mutant (107 cells in less than 20 u.1 1.5% SSPYE), were applied to squares on the treated slides and incubated for 10, 20, 30, and 60 min in a moisturized chamber. The slides were washed with 1.5% SSPYE and photographs were taken of several fields per incubation condition. Finally, the number of cells per field were scored. 18 Table 2. Silanization procedures for covalent modification of slide glasses s u r f a c e w e t t a b i l i t y coe f f i c i en t s i l a n e (%) c o u p l i n g reaction1* rinse step s o l v e n t i n t e r v a l s o l v e n t s (WC) a Diphenyldichlorosilane (DPS) Trimethyl aminopropyl triethoxy silane (QAP) (Tridecafluoro-1,1,2,2-te t rahydrooctyl)- l -silane (TDF) Trimethyl chloro silane (TMS) 29.1 58.3 6.1 methyl chloride l h 20.1 95% ethanol 15-30 min methyl chloride 1 h methyl chloride 1 h 2X methyl chloride 2X ethanol 2X methyl chloride 2X methyl chloride a The wettability coefficient of a surface was determined by measuring the spread of 25 u.1 drops of a series of solutions of water and methanol (8). v 8 WC= — 4/16^x100 Al/W100 + 1/W80 + 1/W60 + 1/W40 + 1/W30 + 1/W20 + 1/W10 + 1/W0 Where W100=mm drop spread at 100% water, W80=mm drop spread at 80% water in methanol etc. 8=number of solvent concentrations used, 4=minimum possible drop measurement in mm 16=measurable range in mm b All reactions were done at room temperature. Cloning of holdfast-encoding. Tn5 inserted DNA A Tn5 inserted genomic DNA restriction fragment from each representative group of five clusters was cloned into pPR510 (46) or pBR328 (54) and propagated in E. coli DH5a. The Southern blot data was used to select the appropriate sized (10-15 kb) restriction fragments that were shared by all groups in a cluster. In cluster V, a cloned insert could not be obtained to include all linked regions because the groups in this cluster did not have shared fragments in Southern blot analyses. So, mutant 126 was arbitrarily chosen for cloning. Chromosomal DNA isolated from the representative mutant groups was digested with appropriate enzymes and separated by sucrose density gradient centrifugation (31). The gradient (10-40%) was made in polyallomer tubes (Beckman, CA) and ultracentrifuged in a Beckman SW41 swinging bucket rotor at 198,000Xg, 20°C, for 20 h. Fractions were collected, electrophoresed, blotted and probed for the presence of Tn5. DNA from the fractions with the strongest signal hybridization was precipitated with ethanol and resuspended in TE (10 /1) . Vector plasmid DNA was isolated and purified by alkaline lysis and banding by cesium-chloride density gradient centrifugation; it was ligated to the isolated chromosomal fragments by standard methods (31). For cloning of Tn5 inserted DNA from mutant group gp4, a Clal fragment was selected. Since the multiple cloning site of pPR510 does not include a Clal site, the plasmid pBR328 was chosen. 20 After ligation the DNA was diluted ten-fold with distilled water and used for electroporation. Electroporation was done with a Bio-Rad Gene Pulser, operated at 200 ohms, 25 |iFD, and 2.5 kV. Electrocompetent E. coli DH5oc cells for the electroporation were prepared according to published procedures (12, 21). 50 u.1 Portions of electrocompetent cells were thawed and mixed with diluted ligation mixture. No DNA and vector only controls were also done. Immediately after the shock, the cuvette contents were mixed with 1 ml of LB medium, incubated for 30 min at 37°C and plated. Transformants were identified by their resistance to Km50 and Cm20. To confirm that Tn5 inserted fragments had been cloned, colony hybridizations using a Tn5 probe were performed. Isolated plasmids from positive clones were digested with appropriate enzymes to excise the cloned inserts. The cloned DNA and total chromosomal DNA from the mutants was analyzed by Southern blot analysis, using Tn5 as a probe to confirm the cloning. To prove that the sites of Tn5 insertion in certain mutant groups were linked and that chromosomal DNA containing all of these loci was cloned, a Southern blot of each cloned plasmid and chromosomal DNA from each mutant group suspected to contain linked sites of Tn5 insertion was probed with corresponding cloned insert (Tn5 element plus flanking chromosomal DNA). 2 1 Comparing the holdfast-related genes of MCS6 and Caulobacter crescentus CB2A In order to assess the degree of homology between the cloned holdfast related regions of freshwater C. crescentus CB2A and MCS6, the following DNA fragments were used as probes in Southern blot analyses: (i) the Bcll-Bglll fragment including hfaAB, (ii) the Bglll-EcoRl fragment including hfaC transcript region (28) and (iii) the EcoRl fragment of the cloned insert of pHFG5B which could complement holdfast defects in the G5 region of CB2A (Mitchell and Smit; unpublished data). Chromosomal DNAs of MCS6, digested with appropriate enzymes, were analyzed using probes prepared from the DNA fragments. The blots were washed under reduced stringency conditions (twice with O.lx washing solution (0.3x SSC, 0.5mM EDTA, 0.002% sarcosyl) for 10 min, at 45°C) and exposed to X-ray film. Such reduced stringency conditions should permit hybridization with as much as approximately 33% mismatched base pairs, assuming a 67% (G+C) content (61). A cosmid library of MCS6 DNA (i) Production of the library A cosmid library for MCS6 was produced as a means to obtain larger DNA segments containing holdfast-related genes on a single recombinant clone for tests of complementation ability, . The broad host range cosmid cloning vector pLAFR5 (26), which has a double cos cassette, was used by digesting with Seal and 22 BamHI and ligation with size-fractionated (sucrose gradient density centrifugation; approximately 20-25 kb) DNA fragments, from MCS6, generated via partial chromosomal digestion with Sau3A (24). Because of the large average insert size, a cosmid clone bank in pLAFR5 need only contain about 900 members to achieve a confidence of 98% that each gene is represented at least once. The ligations were conducted in 5mM ATP to suppress blunt-end ligation (via the Seal generated termini) so that vector dimers lacking inserts are not packaged in phage X heads. Packaging of ligated cosmids was done with an in vitro packaging extract of A, DNA (Gigapack, Vector Cloning Systems, California). After thawing the extracts, they were mixed with the ligation mixture, incubated at room temperature for 2 h, and then phage dilution buffer (0.1M NaCl, 0.008M MgS0 4-7H 20, 0.05M Tris-HC1 (pH7.5), 0.01% gelatin). Chloroform was added to inhibit bacterial growth. E.coli HB101 cells were grown in LB medium with 0.4% maltose to O.D600=0.5. The culture was centrifuged, resuspended in lOmM MgCl2, and infected immediately with packaged cosmids. Growth of the infected cells on LB plates with Tc selected for cells containing the cosmids. The resulting colonies were pooled and stored at -70°C. (ii) Complementation testing The cloned DNA flanking the Tn5 insertion sites from each of the representative groups was used as probes in colony hybridizations to detect cosmid clones containing holdfast-related genes. The cosmids clones were isolated and plasmid DNA was 23 electroporated into holdfast-defective mutants. For the preparation of electrocompetent cells, MCS6 was grown in 0.6% SSPYE (to minimize carry over of salts (21)). Other preparation steps were the same as for E. coli, except that distilled water washes were replaced with a cold solution of 10 mM MgCh and 5mM CaCh. The Gene Pulser was set at 2.5 kV, 25 u.F and 600 ohms (21). Cosmid preparations (1.25 u.1; 100 ng/u.1) were added to the thawed cells and shocked. Then, 1 ml of 1.5% SSPYE was added and the suspension was incubated at room temperature for 30 min with shaking for 4 h at 25°C. Transformants were selected by resistance to Tc and Km and probed with radio-labeled pLAFR5 fragments to eliminate the possibility of spontaneous drug resistant mutants. Their complementation abilities were examined using the lectin binding assay and by monitoring rosette formation. In addition to electroporation, conjugation by triparental mating was used to transfer cosmid clones into holdfast defective mutants. For the mobilization of cosmids, pRK2013, which contains the RK2 mobilization genes and a ColEl replicon, was used as a helper plasmid (11, 16). E. coli MM294 was the host for pRK2013 and the donor-to-helper-to-recipient ratio for conjugation was 1:1:10. In order to test whether recombination occurred between the inserts of cosmid clones and homologous Tn5 interrupted chromosomal DNA in the complemented cells, curing of cosmid clones was done in the following manner: Complemented cells were inoculated into 1.5% SSPYE (without tetracycline), grown to stationary phase and streaked onto 1.5% SSPYE plates. Single 24 colonies were transferred to 1.5% SSPYE plates with and without tetracycline. Colonies that could not grow with tetracycline were inoculated into liquid medium. These liquid cultures were checked for rosette-forming capability by microscopy. 25 RESULTS Production of Tn5 -mutant library 12,000 Tn5 insertion mutants were produced to guarantee a 95% confidence that all genes were interrupted. This is based on the assumption described in Materials and Methods. Depending on the recipient-to-donor ratio used, the overall conjugation and transposition frequency varied between 10"7 and 10 - 5 per donor. At conjugal mating ratios between 1:1 and 5:1, (as specified in most procedures) selection for Tn5 insertion mutants was 10"7 per donor which is 100 fold lower than that observed in C. crescentus CB2A (36). When the mating ratio in MCS6 was increased to >14:1, however, transposition frequency increased dramatically to 10"5. Spontaneous resistance of MCS6 to Km50 and Sm20 was not de tec table ; approximately 109 CFU/plate were plated on Km50/Sm20, but no spontaneous drug resistant mutants were found. To check for the possibility that integration of the entire pSUP2021 plasmid into chromosomal DNA occurred rather than Tn5 transposition, 288 randomly chosen Tn5 insertion mutants were grown on 1.5% SSPYE plates containing Km50 and Cm2. Only three mutants (1.04%) showed growth, suggesting that single crossover recombination between vector and chromosomal DNA occurred infrequently. Chromosomal DNA from the pool of the Tn5 insertion mutant library and a pool of 56 holdfast defective mutants of MCS6 were probed with Tn5. As shown in Fig. 2, the pool of the Tn5 library 26 produced a generally smeared region that probed positive with Tn5, supporting the supposition that there was no hot spot regions of Tn5 insertions. On the other hand, the probed holdfast-defective mutant chromosomal DNAs which were digested with EcoRI and Sstl (enzymes that do not cut Tn5) demonstrated many bands, indicating that holdfast-related regions were scattered throughout the genome. The absence of strongly dominant bands also supported the notion that Tn5 inserted in a random manner (Fig. 2). Development of an assay for holdfast-defective mutants of MCS6 The cellulose acetate colony binding assay with Coomassie blue staining was investigated as a screen for holdfast-defective mutant colonies of MCS6 in the same manner as was done for C. crescentus CB2A (40). However, colonies of positive controls were not stained. Several surfaces, including polyethylene, Nylon, GelBond, glass (silanized or not), mylar and aluminum and several staining chemicals were tested as alternatives in this assay for detecting holdfast-defective mutants of MCS6. For most surfaces, as with cellulose acetate, no stained spots resulting from colony attachment was seen. In the case of polyethylene and Nylon, which had wettability coefficients in the range of 30 to 50, there was even staining of the negative control, strain CM260. Colony-based assays were replaced by the use of liquid cultures and several surfaces were tested. The use of cellulose acetate stained with 0.2% acridine orange (and visualization by UV light-induced fluorescence) seemed to be the best for detecting 27 FIGURE 2. SOUTHERN BLOT OF TN5-MUTATED CHROMOSOMAL DNA: The blot was probed with the 1.8 kb Hindlll/BamBl fragment of the Tn5 element. All chromosomal DNA was digested with EcoRI/Sstl. (a) The Tn5 insertion library pool (b) a pool of 56 holdfast defective mutants (cluster I to V mutants) 28 holdfast defective mutants among several combinations of surfaces and stains. However, the stain was unstable, somewhat variable from experiment to experiment and not sensitive enough to clearly differentiate mutants. We have no clear explanation as to why liquid culture based approaches resulted in much more detectable binding. However, it is possible that in contrast to C. crescentus CB2A most MCS6 cells in a colony are attached to each other (forming rosettes) and are therefore not available for surface attachment. Finally, the use of multiwell polystyrene dishes with liquid cultures resulted in a successful assay. The wells of the microtitre plates in which holdfast-proficient cells were grown stained strongly with crystal violet. In contrast, the wells were not well stained with holdfast-defective mutants; levels of staining ranged from no staining to weak staining was obtained (Fig. 8). The strength of staining in the microtitre plate assay correlated with the phenotypes detected by the lectin binding assay and rosette formation, as shown in Table 3. Wild-type MCS6 produced strong staining. In contrast, mutants with a negative holdfast phenotype were unstained. Mutants with an altered holdfast phenotype produced medium staining. This correlation suggested that the assay using the polystyrene plates was sensitive enough to detect even intermediate degrees of holdfast production. Detection of holdfast-defective mutants Holdfast-defective mutants were initially isolated by the inability of liquid cultures to bind to polystyrene microtitre plates. 2 9 Table 3. Phenotypic analysis of holdfast-defective mutants C l u s t e r / C o n t r o l MCS6 CM260 CB2A Cluster I Cluster II Cluster III Cluster IV Cluster V G r o u p #of mutants 1 17 3 6 15 10 7 13 14 16 4 93 7 126 177 11 34 179 260 9 1 1 5 1 3 7 2 7 1 3 4 1 3 2 2 1 1 1 Holdfast wi ( - ) • Phenotype Id-type •mutant wild-type (FWC)C no no less no no no no no no less less reduced h< holdfast holdfast holdfast holdfast holdfast holdfast holdfast holdfast holdfast holdfast holdfast 1 or altered aldfast R o s e t t i n g +++ -++++ ----------_ + + + + + + + + M . P . a +++ -++ --+ / -------+ / -+ /-+ + + + + + + + L . B . A . b +++ -++++ --+ /-------+ + / -++ +++ ++ ++ ++ ++ +++ +++ o m a, microtitre plate assay; b, lectin binding assay; c, freshwater Caulobacter 203 mutants (1.69%) from the library of 12,000 Tn5 insertion mutants showed reduced or no binding to the polystyrene. These mutants were further characterized. 91 mutants (45%) of the initially screened mutants were pleiotropic mutants which had multiple defects in the polar region of the cell (20, 27) such as in division problems or loss of motility. These mutants were excluded by microscopic observations. Of the remaining 112 mutants, the lectin binding assay showed that 41 mutants (0.33% of 12,000 Tn5 insertion mutants) had little or no holdfast production. The remaining 71 mutants (0.59% of 12,000 Tn5 insertion mutants) still produced holdfast even though they did not bind to the polystyrene as well as wild-type MCS6 (Table 3). Analysis of holdfast-defective mutants (i) Determining the sites of Tn5 insertion The Tn5 containing restriction fragments were identified by Southern blot analyses after chromosomal DNAs were digested by several enzymes. In the first three clusters of mutants (Table 4, see below for the rationale for grouping and clustering), the chromosomal DNAs were digested with Sail, an enzyme that cuts inside of Tn5 once, and Bglll, which cuts twice, and Southern blot hybridized with the Tn5 specific probe. All Sa/I-digested samples produced two bands, one light and one dark band of variable sizes (Fig. 3) and all BgIll-digested samples produced three hands (Fig. 4), two light bands of variable size and a 3 kb dark band (an internal fragment of Tn5). 3 1 GROUP: i 3 17 6 7 1 0 1 5 1 3 1 4 1 6 15.2 -9 64 -6 6 4 -kb ****•' FIGURE 3. A SOUTHERN BLOT COMPARISON OF TN5 CONTAINING FRAGMENTS AMONG HOLDFAST-DEFECTIVE MUTANT GROUPS: Sa/I DIGESTION OF CHROMOSOMAL DNA. Southern blot probed with the 1.8 kb Hindlll/Bamlil Tn5 fragment. 32 1-5 1-4 k h FIGURE 4. A SOUTHERN BLOT COMPARISON OF TNJ CONTAINING FRAGMENTS AMONG HOLDFAST-DEFECTIVE MUTANT GROUPS: Bglll DIGESTION OF CHROMOSOMAL DNA. Southern blot probed with the 1.8 kb Hindlll/Bamlll Tn5 fragment. 33 The Hindlll-BamHl fragment used as a probe spans most of the Tn5 sequence to the left of the Sail site (Fig. 1), including a small part of the insertion sequence of Tn5. Because of the sequence similarity between the right and left insertion sequences, light and dark bands were produced when hybridized to Sail digests of Tn5-containing DNA (Fig. 3). This provided another means of determining the position and similarity of the Tn5 insertions. Mutants that had identical patterns on the Southern blots analyses were defined as a group; each group represents a unique site of insertion. 11 unique sites of Tn5 insertion were identified. All of the 41 m u t a n t s belonged to one of these groups. When the genomic DNAs of all mutant groups were cleaved by several enzymes that do not cut Tn5 (EcoRl, Sstl, Kpnl, and Clal) and analyzed by Southern blot analysis (Table 4), it could be determined that certain holdfast-related Tn5 insertion groups were clustered: (i) gpl, gp3 and gpl7, (ii) gp6, gpl5, gplO, gp7, gpl3, gpl4 and gpl6, and (iii) gp4. 11 unique sites of Tn5 insertion were clustered in three regions. Groups in cluster I (Fig. 3) produced two bands of similar size on digestion with Sail, supporting the possibility that the groups were closely linked to each other. The groups in the other clusters, however, did not have such a pattern on Sail digestion, suggesting that Tn5 insertion sites of the groups might be scattered within the clusters. The last two mutant categories exhibited reduced binding in the microtitre plate assay but possessed normal-appearing holdfasts 34 Table 4. The size* of the Tn5 containing chromosomal DNA fragments from the holdfast defective mutants. Group Dral Spel EcoRI E-Sb Sstl S-Kc Kpnl K-Cd Clal K-Ee CLUSTER I 1 17 3 215 215 215 140 140 140 3.3 3.3 3.3 3.2 3.2 3.3 18.6 18.6 8.1 18.6 18.6 8.1 >20 >20 >25 4.3 4.3 4.8 4.3 4.3 4.8 3.3 3.3 3.0 CLUSTER II 6 15 10 7 13 16 14 40 40 40 ? 40 40 40 74 74 74 74 115 115 115 9.6 9.6 9.6 9.6 5.0 1.5 1.2 3.8 3.8 3.6 6.6 1.9 1.3 1.2 3.8 3.8 3.6 8.1 2.6 1.3 16.6 3.8 3.8 3.6 3.1 2.6 1.3 9.1 >20 >20 >20 >20 >20 >20 >20 9.6 9.6 9.6 2.3 1.0 9.6 8.1 9.6 9.6 9.6 2.6 1.0 9.6 8.1 9.6 9.6 9.6 2.6 5.0 1.5 1.2 CLUS-TER III 4 310 DF3 5.1 3.8 5.1 5.1 >20 11.0 11.0 5.1 CLUS-TER IV 93 100 370 5.3 3.1. 9.6 9.6 >20 12.6 18.0 6.6 CLUSTER V 7 126 177 11 34 179 260 155 155 155 155 ? 155 155 135 135 105 105 135 135 135 15.6 15.6 3.1 6.3 6.3 3.9 3.9 2.6 7.1 3.1 2.9 3.8 2.4 1.7 2.6 7.1 7.1 7.1 6.3 6.3 1.7 2.6 5.1 7.1 7.1 7.3 6.3 1.7 4.8 10.6 >20 >20 >20 >20 5.6 4.0 2.4 10.1 10.1 2.4 1.6 4.6 4.5 6.3 10.1 10.1 2.4 2.0 7.1 2.3 10.6 3.1 7.1 6.3 2.6 2.3 a The fragment was not resolved, but it was determined that the size was different from those of the other mutants b E-S, EcoRl-Sstl; c S-K, Sstl-Kpnl; d K-C, Kpnl-Clal; e E-K, EcoRl-Kpnl. * The sizes of all the digested fragments do not include the size of TnJ insert. in the lectin binding assay. These mutant categories contained a large number of mutants (71 mutants) (Table 4), 15 of which were randomly chosen and analyzed. There were 8 unique Tn5 insertion sites clustered in 2 regions of the genome. Each mutant was distinct and they were called mutant X instead of groups because no attempt for grouping mutants was made. Except for a few groups, including gpl3, 14, and 16 in cluster II and a number of mutants in cluster IV, all the groups in one cluster had at least one shared restriction fragment. These groups and mutants did not have shared fragments in Southern blot analyses (Table 4). However, PFGE analyses (done by Dr. Ely, Univ. of South Carolina) indicated that they were linked with the other cluster II groups as well as each other on a 40 kb Dral fragment, indicating that they were separate from the other groups, but still on the same Dral fragment. A similar situation occurred with the Cluster V mutants . In sum, Southern blot analyses of all the groups, including the last two categories showed a total of 19 unique Tn5 insertion sites clustered in 5 regions throughout the chromosome. It is possible that more groups and clusters might be identified if all the 71 mutants in cluster IV and V had been analyzed. (ii) Phenotype evaluation of holdfast-defective mutants Three holdfast-defective phenotypes were identified in the analysis (Table 3): (i) no holdfast (ii) less holdfast produced than wild type and (iii) apparent production of a normal holdfast but abnormal 36 adhesion, i.e., reduced or altered holdfast adhesion. The holdfast shedding phenotype (36), in which cells can produce holdfast but fail to firmly attach the organelle to the cells, was not found, despite assays to locate such mutants within the type (iii) phenotype described above. Cells of gp3, gp4 and gpl6 mutants remained attached to the wells of polystyrene plates after washing steps and were stained by crystal violet. But the strength of the staining was decreased when compared with that of MCS6. Also, a few cells with holdfasts were detected in the lectin binding assay. Except for these three groups of mutants, all holdfasts of clusters I, II and III failed to bind to the polystyrene plates; the absence of a holdfast was confirmed by lectin binding assay (Table 3). Clusters IV and V mutants had an interesting phenotype in that they stained polystyrene (microtitre plate wells) less than MCS6 but still produced an equivalent holdfast with identical lectin binding as wild-type. Most of them could form rosettes, although these contained a smaller number of cells than wild type rosettes. To further characterize this phenotype, glass microscope slides were treated with several siloxane chemicals, which produced surfaces with different degrees of hydrophobicity and hydrophilicity (8, 48). The properties of surfaces in hydrophobicity and hydrophilicity can be described in terms of wettability as defined in Table 2. As shown in Fig. 5, mutants of gpl and gp3 showed binding to TMS and QAP (WC of about 20 and 60, respectively). However, the number of cells bound to the surfaces were still very low 37 T3 r—t •I-H P H «5 « 100 I—H 0> •i-H \ C/3 . — I 1—4 01 U TDF(WC6.1) TMS(WC20.1) DPS (WC 29.1) QAP(WC58.3) TDF(WC6.1) TMS(WC20.1) DPS (WC 29.1) QAP(WC58.3) FIGURE 5. SCORING OF CELLS BOUND TO CHEMICALLY TREATED GLASS SLIDES. Each value represents average cell number after counting cells twice or three times and the bar indicates standard error. All the abbreviations are explained in Table 2. 38 because they were holdfast-defective mutants. Since the gpl mutant produced no holdfast, the gp3 mutant (which produced a small amount of holdfast) had higher levels of cell binding than the gpl mutants. It is possible that gpl mutants weakly attached to surfaces via other extracellular polysaccharide polymers (59) and the other mutants bound to the surfaces by some combination of holdfasts and the other extracellular polysaccharide polymers. Most of the cluster IV and V mutants tested, except mutant 7, demonstrated a pattern that was not dramatically different than MCS6 wild type cells, except for a general trend to bind to the TMS surface more than wild-type. The binding of all holdfast-producing strains to a very hydrophobic surface (TDF) was comparable to that of wild-type cells. Mutant 7 was unusual. With respect to the surfaces with wettability coefficients of about 20 and 60, this mutant bound more than wild-type and the other mutants. In contrast, whereas wild type cells preferred to bind to a surface with a wettability of about 30 (in comparison to lower and higher wettability numbers), mutant 7 bound less to this surface. Also, mutant 7 had a different binding pattern than mutants 11 and 126 although all three belonged to cluster V. Furthermore, mutant 93, belonging to cluster IV, had the same binding pattern as mutant numbers 11 and 126. No generic pattern of differences in binding was found between the mutants and wild-type Caulobacters. 39 Cloning of holdfast-related. Tn5 inserted DNA Tn5 inserted DNA fragments were cloned into pPR510 or pBR328 and propagated in E.coli DH5a. The sizes of cloned fragments and plasmids used are shown in Table 5. In determining the fragments to clone, hybridization to chromosomal DNA of all groups (digested with appropriate enzymes) to reveal the presence of Tn5 showed that in most cases a fragment could be selected that was common to all groups in the same cluster. The exceptions to this were gpl3, 14 and 16 in cluster II (Fig. 6), and mutants in cluster V (not shown). For this reason, a second region was selected for Cluster II, the EcoRI fragment of gpl3, and for Cluster V the Sstl fragment of mutant 126 was arbitrarily selected. However, in both cases, clones containing DNA of interest could not be obtained after several attempts. Cloned segments for gpl3 and mutant 126 produced several smaller sized fragments after EcoRI or Sstl digestion, in which only one fragment was supposed to be observed on digestion with EcoRI or Sstl. It was assumed that the products of these gene regions may be toxic to E. coli, resulting in rearrangements . As well, the digested plasmid clones were probed with Tn5. The positive bands that were the same size as the positive chromosomal DNA bands and showing the expected inserts were cloned (Fig. 6). In addition, genomic DNA for each mutant group was probed with cloned inserts of a representative group in each cluster (Fig. 7). Bands of the same size were obtained in these cases, confirming that 40 Table 5. Summary of clones derived from Tn5 interrupted regions. Cluster group enzyme insert size plasmid name insert + plasmid (kb) (kb) Cluster I 1 17 3 EcoRI 9.1 pYHFl 13.4 (pPR510) Cluster II 6 15 10 7 13 14 16 EcoRI EcoRI 15.4 10.8 pYHF6 pYHF13! 19.7 (pPR510) 15.1 (pPR510) Cluster III Cluster IV Cluster V 4 93 126 Clal Sstl SstI 16.8 15.4 13.1 pYHF4 pYHF93 pYHF126* 21.7 (pBR328) 19.7 (pPR510) 17.4 (pPR510) * In these cases there were indications of rearrangements. ' 11 11 1 r '•1 ' ' ; Mil ! pYHFC) pYHM r^ ROl P: 1 17 3 6 15 10 7 13 14 16 4 FIGURE 6. A SOUTHERN BLOT COMPARISON OF CLONED HOLDFAST RELATED DNA FRAGMENTS WITH CHROMOSOMAL DNA DIGESTS OF HOLDFAST-DEFECTIVE MUTANT GROUPS THAT CONTAIN LINKED SITES OF TN5 INSERTIONP: PROBING FOR TN5. pYHFX= the cloned plasmid that contains holdfast related and Tn5 mutated chromosomal DNA cloned from mutant group gpX or mutant X(where X=l to 93). 42 the cloned inserts were from the genomic region that included all the Tn5 insertion sites of mutant groups in each cluster. This solidified the conclusion that those groups were in fact linked. In cluster II of Fig. 7, gpl was used as an example of the expected result obtained with unlinked groups, that is, two bands appeared: one band which hybridized to Tn5 and another one containing unlinked flanking DNA. Similarity between CB2A and MCS6 DNAs By testing the hybridization of CB2A holdfast related genes and MCS6 chromosomal DNA, we had hoped to define to what extent the holdfast of MCS6 is related to that of CB2A. Unfortunately, no hybridization was detected under the least stringent conditions tested (permitting approximately 33% base-pair mismatch). Complementation tests A cosmid library containing 1,200 individual clones of MCS6 genomic DNA was produced. Colony blots for the cosmid library were prepared and used repetitively. About 3 to 7 positive cosmid clones were identified by probing with the cloned flanking DNAs of Tn5 inserted sites of each representative group. In the case of gpl3 and 126, where rearrangements of cloned DNA had apparently occurred, isolated plasmids were digested and used for preparing the probes after confirmation that cloned segments included Tn5 by Southern blot analysis using Tn5 as probe. Conjugal transfer of the cosmid clones to Caulobacter generally 43 '• ! i ;\. i II A'i IV t:*-/VME: JicoRI EcoRI C/al SstI , , | 1 , , , , PI ASMIU- i Ylll ; pYMIfj pYHF4 pYHFO^  ,RO!.P: ! : - "< .S 7 1 4 9^ FIGURE 7. A SOUTHERN BLOT COMPARISON OF CLONED HOLDFAST RELATED DNA FRAGMENTS WITH CHROMOSOMAL DNA DIGESTS OF HOLDFAST-DEFECTIVE MUTANT GROUPS THAT CONTAIN LINKED SITES OF TN5 INSERTION: All probes contain both Tn5 and flanking chromosomal DNA. Cluster I mutants are probed with the cloned inserts of pYHFl; cluster II mutants with the cloned inserts of pYHF6; cluster III, pYHF4; cluster IV, pYHF93. 44 occurred at an efficiency of 10"8 to 10-9 per recipient. The low efficiency probably is the result of triparental mating, since triparental mating conjugation usually occurs with lower efficiency than biparental matings. No spontanous drug resistant mutants were found among transconjugants. Transconjugants were checked by phase-contrast microscopy and examined by the lectin binding assay and microtitre plate assay; most formed rosettes and produced an amount of holdfast comparable to MCS6 (Table 6 & Fig. 8). Cosmids pCYHF6, pCYHF4 and pCYHF126 complemented holdfast defects in holdfast-defective mutants, resulting in the production of similar amounts of holdfast material and forming rosettes comparable to wildtype MCS6, although they were not the same. However, gpl cells with pCYHFl showed unstable complementation, unlike the other complemented cells; the staining in the microtitre plate assay was unstable, producing a variable strength of staining from experiment to experiment. pCYHF13 and pCYHF93 did not completely complement holdfast defects. pCYHF13 resulted in the production of only a small amount of holdfast material. Mutant 93 cells with pCYHF93 did not show any more production of holdfast or rosetting than mutant 93 cells without pCYHF93 and no differences were seen in the microtitre plate assay (Fig. 8). Conjugation proved to be the main tool used for complementation; because of low electroporation efficiency, transformants were obtained for only one group. Difficulties in removal of sea salts during the washing steps in preparing 45 Table 6. Complementation Test C l u s t e r G r o u p R o s e t t i n g M i c r o t i t r e L e c t i n O v e r a l l & & p l a t e b i n d i n g C o m p l e -Contro l Cosmid a s s a y a s s a y m e n t a t i o n MCS6 I I I I I I IV V wild type gp l pCYHFl gp6 pCYHF6 g p l 3 pCYHFB gp4 pCYHF4 93 pCYHF93 126 pCYHF126 -H-+ + + + OT++ + / • + + Yes Yes +/- Yes, but very poor Yes No Yes 46 I 2 ^ 4 5 6 7 8 9 1 1 ) 1 1 1 2 A H (' J) I-f ( y i i FIGURE 8. COMPLEMENTATION ANALYSIS OF COSMID CLONES USING MICROTITRE PLATE ASSAY: The test was done in the same way as the microtitre plate assay used to initially screen for mutants (see Materials and Methods). Well Al contains gpl mutants; well A3, contains gp6; A5, gpl3; A7, gp4; A9, mutant 93; A l l , mutant 126; CI, gpl with pCYHFl; C3, gp6 with pCYHF6; C5, gpl3 with pCYHFB; C7, gp4 with pCYHF4; C9, mutant 93 with pCYHF93; C l l , mutant 126 with pCYHF126; El , MCS6; E3, CM260 47 competent cells and the large size of the cosmid clones are likely the main reasons for the low efficiency of electroporation. (21). In order to check the possibility that complementation of holdfast defects could have occurred by recombination between inserts of cosmids and homologous defective chromosomal DNA, curing of the plasmids was done. Cured cells were readily obtained in all cases and through the curing step, originally complemented cells no longer formed rosettes and became tetracycline sensitive, indicating that loss of plasmid led to loss of complementation. The ease with which tetracycline sensitive derivatives could be selected also indicated that the plasmid had remained as an independently replicating unit. 48 DISCUSSION Tn5 insertion mutagenesis was used to identify the holdfast-related gene regions of marine Caulobacter MCS6. The identified gene regions were cloned to confirm, by complementation analysis, that they were indeed holdfast-related regions and to support future investigation in molecular studies of the holdfast and the adhesion process. Tn5 insertion mutagenesis identified five distinct regions in MCS6 genome that have some function related to holdfast production, assembly or function. The restriction enzyme analyses for Southern blot analyses and PFGE analysis indicate that these five regions do not overlap each other. All of the holdfast defective mutants that were identified had no detectable holdfast, produced less holdfast or had an altered holdfast phenotype. Since holdfast material is present in small amounts, compared to the other polysaccharides of the cell, it is likely to be a significant task to isolate holdfast from the cell polysaccharides. Thus, a mutant exhibiting the shed holdfast phenotype might be useful for isolation of holdfast. However, the holdfast shedding phenotype was not found, unlike the case for the freshwater C. crescentus CB2A (36), despite attempts to specifically identify this class. It is possible that MCS6 has a different mechanism of bridging the holdfast to the cell surface from that of CB2A. Alternatively, holdfasts of marine Caulobacters might be easily and quickly degraded if not attached to the cell surface. 49 Another possibility is that there was no Tn5 insertion in the genomic area where interruption by Tn5 would cause the shedding phenotype. Most of Tn5 insertions resulted in a null or low-producing holdfast phenotype (cluster I, II and III). These regions of the genome are expected to be responsible for synthesis, transport, and/or assembly of the holdfast. It is also possible that some insertions interrupted a positive regulator involved synthesis or assembly of holdfast material. The hfaB region in the CB2A is an example of such a phenotype, albeit, as applied to the holdfast shedding phenotype (28). The mutants in cluster IV and V had an intriguing phenotype. They produced holdfast but showed altered adhesion characteristics. One explanation for this phenotype could be that the holdfast of these mutants consists of different or altered residues or substituents which result in altered binding to various surfaces. Binding assays on siloxane treated glass slides were used to examine this hypothesis. It was expected that a reduced binding affinity to hydrophobic surfaces would occur in these mutants because they were selected on the basis of reduced binding to polystyrene, a relatively hydrophobic surface. Instead, the data indicated that a property of polystyrene other than simple hydrophobicity must be involved in the adhesion deficiency noted in cluster IV and V mutants. It is likely that the holdfast has affinity for specific chemical substituents, and the loss of specific parts of the holdfast has modified the adhesive abilities in specific ways. As the chemical 50 and genetic analysis of cluster IV and V mutants becomes available, a clearer understanding of the differences in binding ability, relative to wild-type cells, should become apparent. A second explanation for the cluster IV and V phenotype involves the mechanism of bridging the holdfast to the cell surface. Some alteration of this "bridge" could be responsible for subtle alterations in the adhesion characteristics of the holdfast proper. Transposon insertions have generated complete breakage of this "bridge" region in freshwater Caulobacter CB2A resulting in a holdfast shedding phenotype (36). However, in the marine Caulobacter MCS6 mutations in the "bridge" would have to be more subtle than a complete breakage of the connection between the cell and the holdfast because none of the cluster IV and V mutants appeared to lose their holdfast in the lectin binding assay. The holdfast-related gene regions of MCS6 were dispersed throughout the genome, similar to the corresponding genes of C. crescentus CB2A. Although Southern blot analyses demonstrated five regions, minimally at least three holdfast related regions of the genome could be present, if linked groups in one cluster are located adjacent to linked groups in another cluster. This, of course, also sets the minimum number of operons. As a formal argument, it is possible that not all the Tn5 inserted mutants were the result of mutation of holdfast-specific genes. One obvious possibility is that interruption of genes responsible for synthesis of specific monosaccharides found in holdfast, but also used in the biosynthesis of other cell 5 1 polysaccharides, might result in a holdfast-deficient phenotype. If this occurred in the present study, the corresponding defect in production of a second polysaccharide (LPS, for example) did not result in a visible defect. Generally, Tn5 insertion in MCS6 genome seemed to be relatively random, probably more so than noted with C. crescentus CB2A (36). An increased number of groups of holdfast-defective mutants, smeared positive bands of Tn5 inserted mutant chromosomal DNA (when it was probed by Tn5) and the existence of pleiotropic mutations (which were not found with CB2A) support the randomness of Tn5 transposition in the genome of MCS6. In preliminary experiments, expression of the streptomycin resistance gene on Tn5 (which is not expressed in E. coli) was noted in MCS6 (39, 45). This was a significant technical aid, allowing for selection for Tn5 acquisition with two drugs. This virtually eliminated the spontaneous drug resistant mutations that occur with selection for transposition with a single drug. In the process of assay development for the detection of holdfast-defective mutants the cellulose acetate assay, which was successfully used for C. crescentus CB2A, was not applicable to holdfast mutant detection in MCS6. One possible explanation for this difference may be that in contrast to C. crescentus CB2A, most of the MCS6 cells in a colony are attached to each other (forming rosettes) and are therefore not available for surface attachment. It is not clear why this difference may be so, but the findings do support the concept that the holdfast of MCS6 is probably different from that of 52 CB2A. This latter point is also supported by previous findings of differential sensitivity to hydrolytic enzymes and in different patterns of lectin binding (35). It is perhaps not surprising that no homology was found between MCS6 chromosomal DNA and CB2A holdfast gene segments. The phylogeny study by 16S rRNA sequence comparisons in marine and freshwater Caulobacters (57) showed that none of the freshwater Caulobacters were closely affiliated with the marine line of descent. MCS6 and CB2A had only 90.1% rRNA similarity which is not a high percentage of similarity, considering that the Caulobacters were determined to be a diverse collection and the most distantly related of the Caulobacters characterized were associated at approximately 88%. For reference in 16S rRNA sequence comparisons, 50% DNA similarity corresponds to approximately 98 to 99% 16S rRNA sequence similarity. We tentatively conclude that at a first approximation CB2A holdfast-related genes are significantly different from those of MCS6. However, our analysis did not involve an extensive or rigorous modification of hybridization stringency conditions and a very weak hybridization signal might well be achievable under some conditions. It is still quite possible that highly conserved regions exist between the holdfast-related genomes of the two species that are too small to be readily detected by hybridization with large probes. Future work in comparing holdfast genes between these species is likely best done after identification and sequence analysis of each set of holdfast genes. 53 Attempts to complement holdfast defects with cloned DNA were positive in four of the six holdfast mutant groups. The complemented cells were almost like wild type. In the case of complementation of gpl with pCYHFl, it is possible that the complementation-produced product could be unstable or imperfect and thus explain the unstableness of complementation. Since unrearranged clones were not obtained in the cloning of holdfast related genes for gpl3 and a rearranged plasmid fragment was used as a probe, it is possible that the cosmid clones detected by this probe did not contain the same genomic region as was surrounding the Tn5 insertion in gpl3 . This might explain the lack of complementation for gpl3. In the case of failure to complement the holdfast defect of mutant 93, it is possible that the cosmid clones obtained did not have a complete transcription unit or translational signals to express the holdfast functions even though the size of the inserts of cosmid clones seemed to be large enough to encompass typical clusters of polysaccharide-encoding gene clusters or transcription units. Although a minor point, we were curious as to whether complementation of holdfast defects by cosmid clones was accomplished by the plasmid themselves or as the result of recombination of the plasmid into the genome region. A plasmid curing step with the complementing strains resulting in loss of complementation has assured us that the complementation seen was in all cases the consequence of independently-replicating cosmids. 54 The cloned segments of Tn5 inserted regions will be useful in the future characterization of holdfast composition, biogenesis and genetic regulation. The cloned Tn5 -inserted fragments can be used as probes to readily retrieve uninterrupted segments of the same region, which require relatively minimal amounts of subcloning prior to sequencing. However, the fragments may be too small to encompass all the whole holdfast related genes in a particular region. In such cases the cosmid clones will be valuable since they are large (average insert size of 23 kb) enough to readily contain all components of functional genetic regions for holdfast. In the future, studies will be continued to identify genetic regulation, mechanisms of holdfast adhesion, and so forth using these clones. Chemical studies on the composition of holdfast will be another area of study for holdfast characterization. However additional efforts should be made to obtain shedding phenotype mutants. This seems to be a key requirement for efficient separation of holdfast from the remainder of the cell in the analysis of the freshwater Caulobacter holdfast and will likely be important for MCS6 as well. 55 REFERENCES 1. Anast, N. and J. Smit. 1988. Isolation and characterization of marine Caulobacters and assessment of their potential for genetic experimentation. Appl. Environ. Microbiol. 54: 809-817. 2. Auswald, E., A. Aludwig and H. Schaller. 19 8 0 . Structural analysis of Tn5. Cold Spring Harbor Symp. Quant. Biol. 45: 107-113. 3. Berg, D. E., M. A. Schmandt and J. B. Lowe. 1 9 8 3 . Specificity of transposon Tn5 insertion. Genetics 105: 8 1 3 -828. 4. Bolivar, F., R. Rodriguez, P. J. Greene, M. C. Betlach, H. L. Heyneker and H. W. Boyer. 1977. Construction and characterization of new cloning vehicles II: A multipurpose cloning system. Gene 2: 95-113. 5. Bossi, L. and M. S. Ciampi. 1981. DNA sequences at the sites of three insertions of the transposable element Tn5 in the histidine operon of Salmonella. Mol. Gen. Genet. 183: 4 0 6 -408. 6. Boyer, H. W. and D. Roulland-Dussoix. 1969. A complementation analysis of the restriction and modification of DNA in E. coli. J. Mol. Biol. 41: 459-472. 7. Brunovskis, I. and W. C. Summers. 1972. The process of infection with coliphage T7. Virology 50: 322-327. 56 8. Burchard, R. P., D. Rittschof and J. Bonaventura. 1 9 9 0 . Adhesion and motility of gliding bacteria on substrata with different surface free energies. Appl. Environ. Microbiol. 5 6 : 2529-2534. 9. De Brujin, F. J. and J. R. Lupski. 1984. The use of transposon Tn5 mutagenesis in the rapid generation of correlated physical and genetic maps of DNA segments cloned into multicopy plasmids - a review. Gene 27: 131-149. 10. Dexter, S. C , J.D. Sullivan, J. W. I l l and S. W. Watson. 1975. Influence of substrate wettability on the attachment of marine bacteria to various surfaces. Appl. Environ. Microbiol. 30: 2 9 8 - 3 0 8 . 11. Ditta, G., S. Stanfield, D. Corbin and D. R. Helinski. 1980. Broad host range DNA cloning system for Gram-negative bacteria: construction of a gene bank of Rhizobium meliloti. Proc. Natl. Acad. Sci. USA 77: 7347-7351. 12. Dower, W. J., J. F. Miller and C. W. Ragsdale. 1988. High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res. 16: 6127-6145. 13. Dworkin, M. 1985. "Caulobacter," p.50-67. In Developmental biology of bacteria. The Benjamin-Cumming Publishing Co., Menlo Park, California. 14. Ely, B. and R. H. Croft. 1982. Transposon mutagenesis in Caulobacter crescentus. J. Bacteriol. 149: 620-625. 57 15. Ely, B. and C. J. Geradot. 1988. Use of pulsed-field-gradient gel electrophoresis to construct a physical map of the Caulobacter crescentus genome. Gene 68: 323-333. 16. Figurski, D. H. and D. R. Helinski. 1979. Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proc. Natl. Acad. Sci. USA 76: 1648-1652 . 17. Fletcher, M. 1987. How do bacteria attach to solid surfaces? Microbiol. Sci. 4: 133-136. 18. Fletcher, M. and G. I. Loeb. 1979. Influence of substratum characteristics on the attachment of a marine Pseudomonad to solid surfaces. Appl. Environ. Microbiol. 37: 67-72. 19. Fletcher, M. and K. C. Marshall. 1981. Advances in microbial ecology, p. 199-236. In K. C. Marshall, Are solid surfaces of ecological significance to aquatic bacteria? Plenum press, New York. 20. Fukuda, A., K. Miyakawa, H. Iida and Y. Okada. 1 9 7 6 . Regulation of polar surface structures in Caulobacter crescentus: pleiotropic mutations affect the coordinate morphogenesis of flagella, pili and phage receptors. Mol.. Gen. Genet. 149: 167-173. 2 1 . Gilchrist, A. and J. Smit. 1991. Transformation of freshwater and marine Caulobacters by electroporation. J. Bacteriol. 173: 921-925. 58 22. Hanahan, D. 1983. DNA cloning, a practical approach, p.109-114. In D. M. Glover, Techniques for transformation of E. coli. IRL Press, Washington D. C. 23 . Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol 166: 557-580. 24. Ish-Horowicz, D. and J. F. Burke. 1981. Rapid and efficient cosmid cloning. Nucleic Acids Research 9: 2 9 8 9 -2998. 25 . Jorgensen, R. A., S. J. Rothstein and W. S. Reznikoff. 1979. A restriction enzyme cleavage map of Tn5 and location of a region encoding neomycin resistance. Mol. Gen. Genet 177: 6 5 - 7 2 . 26. Keen, N. T., S. Tamaki, D. Kobayashi and D. Trollinger. 1988. Improved broad-host-range plasmids for DNA cloning in Gram-negative bacteria. Gene 70: 191-197. 27. Kurn, N., S. Ammer and L. Shapiro. 1974. A pleiotropic mutation affecting expression of polar development events in Caulobacter Crescentus. Proc. Nat. Acad. Sci. USA 71: 3157-3161 . 28. Kurtz, H. D. and J. Smit. 1992. Analysis of a Caulobacter crescentus gene cluster involved in attachment of the holdfast to the cell. J. Bacteriol. 174: 687-694. 29. Lerouge, P., P. Roche, C. Faucher, F. Mailet, G. Truchet, J. C. Prome and J. Denarie. 1990. Symbiotic host-specificity of Rhizobium meliloti is determined by a sulphated 59 and acylated glucosamine oligosaccharide signal. Nature 344: 781-784. 30. Lupski, J. R., P. Gershon, L. S. Ozaki and G. N. Godson. 1984. Specificity of Tn5 insertions into a 36-bp DNA sequence repeated in tandem seven times. Gene 30: 99-106. 3 1 . Maniatis, T., E. F. Fritsch and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold spring harbor laboratory, Cold spring harbor, New York. 32. Marshall, K. C. 1985. Bacterial adhesion, mechanisms and physiological significance, p.133-161. In D. C. Savage and M. Fletcher, Mechanisms of bacterial adhesion at solid-water interfaces. Plenum press, New York. 33 . Marshall, K. C , R. Stout and R. Mitchell. 1 9 7 1 . Mechanism of the initial events in the sorption of the marine bacteria to surfaces. J. Gen. Microbiol. 68: 337-348. 34. Meinkoth, J. and G. M. Wahl. 1987. Nick translation. Methods Enzymol. 152: 91-94. 35 . Merker, R. I. and J. Smit. 1988. Characterization of the adhesive holdfast of marine and freshwater Caulobacters. Appl. Environ. Microbiol. 54: 2078-2085. 36. Mitchell, D. and J. Smit. 1990. Identification of genes affecting the adhesion organelle of Caulobacter crescentus CB2. J. Bacteriol. 172: 5425-5431. 37. Newton, A. and N. Ohta. 1990. Regulation of the cell division cycle and differentiation in bacteria. Ann. Rev. Microbiol. 44: 689-719. 60 38. Newton, A., N. Ohta, E. Huguenel and L.-S. Chen. 1 9 8 5 . Approaches to the study of cell differentiation in Caulobacter crescentus, p. 267-276. In J. A. Hoch and P. Setlow, Molecular biology of microbial differentiation; Proceedings of the ninth international spores conference, Asilomer, Calif. American Society for Microbiology, Washington, D. C . 39. O'Neill, E. A., G. M. Kiely and R. A. Bender. 19 8 4 . Transposon Tn5 encodes streptomycin resistance in nonenteric bacteria. J. Bacteriol. 159: 388-389. 40. Ong, C. J., M. L. Y. Wong and J. Smit. 1990. Attachment of the adhesive holdfast organelles to the cellular stalk of Caulobacter crescentus. J. Bacteriol. 172: 1448-1456. 4 1 .Poindexter, J. S. 1964. Biological properties and classification of the Caulobacter group. Bacteriol. Rev. 2 8 : 231-295 . 42. Poindexter, J. S. 1981. Oligitrophy, p.63-89. In M. Alexander, Advances in microbial ecology. Plenum press, New York. 43 . Poindexter, J. S. 1981. The Caulobacters: ubiquitous unusual bacteria. Microbiol. Rev. 45: 123-179. 44. Pringle, J. H. and M. Fletcher. 1983. Influence of substratum wettability on attachment of freshwater bacteria to solid surfaces. Appl. Environ. Microbiol. 45: 811-817. 45 . Putnoky, P., G. B. Kiss, I. Ott and A. Kondorosi. 19 8 3 . Tn5 carries a streptomycin resistance determinant downstream 6 1 from the kanamycin resistance gene. Mol. Gen. Genet 1 9 1 : 288-294. 46. Quigley, N. B. and P. R. Reeves. 1987. Chloramphenicol resistance cloning vector based on pUC9. Plasmid 17: 54-57. 47 . Reed, K. C. and D. A. Mann. 1985. Rapid transfer of DNA from agarose gel to nylon membranes. Nucleic Acids Res. 1 3 : 7207-7221 . 48. Roberts, D., D. Rittschof, E. Holm and A. R. Schmidt. 1991. Factors influencing initial larval settlement: temporal, spatial and surface molecular components. J. Exp. Mar. Biol. Ecol. 150: 203-211 . 49. Rosenberg, M. 1981. Bacterial adherence to polystyrene: a replica method of screening for bacterial hydrophobicity. Appl. Environ. Microbiol. 42: 375-377. 50. Shapiro, L. 1985. Generation of polarity during Caulobacter cell differentiation. Ann. Rev. Cell Biol. 1: 173-207. 5 1 . Simon, R., U. Priefer and A. Piihler. 1983. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram negative bacteria . Bio/Technology 1: 784-790. 52. Smit, J. and N. Agabian. 1984. Cloning of the major protein of the Caulobacter crescentus periodic surface layer: detection and characterization of the cloned peptide by protein expression assays. J. Bacteriol. 160: 1137-1145. 53 . Smith, C. L., S. R. Klco and C. R. Cantor. 1988. Genome analysis: a practical approach, p.41-72. In K. E. Davies, Pulsed-62 field gel electrophoresis and the technology of large DNA molecules. IRL Press, Washington D. C. 54. Soberon, X., L. Covarrubias and F. Bolivar. 19 8 0 . Construction and characterization of new cloning vehicles: IV. Deletion derivatives of pBR322 and pBR325. Gene 9: 287-305. 55 . Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol 98: 5 0 3 - 5 1 7 . 56. Srivastava, R., V. B. Sinha and B. S. Srivastava. 1 9 8 9 . Chromosomal transfer and in vivo cloning of genes in Vibrio cholerae using RP4::mini-Mu. Gene 75: 253-259. 57. Stahl, D. A., R. Key, B. Flesher and J. Smit. 1992. The phylogeny of marine and freshwater Caulobacters reflects their habitat. J. Bacteriol. 174: 2193-2198. 58. Sutherland, I. W. 1980. Polysaccharides in the adhesion of marine and freshwater bacteria, p.329-338. In R. C. W. Berkley, J. M. Lynch, J. Melling, P. R. Rutter and B. Vincent, Microbial adhesion to surfaces. Ellis Horwood Ltd., Chichester, England. 59. Sutherland, I. W. 1983. Microbial exopolysacharides- their role in microbial adhesion in aqueous systems. CRC Crit. Rev. Microbiol. 10: 173-201. 60. Umbreit, T. H. and J. L. Pate. 1978. Characterization of the holdfast region of wild-type cells and holdfast mutants of Asticcacaulis biprosthecum. Arch. Microbiol. 118: 157-168. 63 61. Wood, N. B., A. V. Rake and L. Shapiro. 1976. Structure of Caulobacter deoxyribonucleic acid. J. Bacteriol. 126: 1305-1315. 64 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items