Molecular Studies on Entry Exclusion in Escherichia coli Minicells

Minicells produced by abnormal cell division in a strain of Escherichia coli (K-12) have been employed here to investigate the phenomenon of "entry exclusion." When purified minicells from strains containing F' or R factors, or both, are mated with radioactive thymidine-labeled Hfr or R(+) donors, the recipient minicells can be conveniently separated from normal-sized donors following mating, and the products of conjugation can be analyzed in the absence of donors and of further growth of the recipients. Transmissible plasmids or episomes are transferred less efficiently to purified minicells derived from strains carrying similar or related elements than to strains without them. Measurement of deoxyribonucleic acid (DNA) degradation and determination of weight-average molecular weights following transfer indicate that degradation of transferred DNA or transfer of smaller pieces cannot account for the comparative reduction in transfer to entry-excluding recipients. Therefore, we conclude that entry exclusion operates to prevent the physical entry of DNA into recipients expressing the exclusion phenotype. The R-produced repressor (product of the drd(+) gene), which represses fertility (i.e., ability to act as donor), reduces exclusion mediated by R or F factor, or both, in matings between strains carrying homologous elements. Furthermore, the data suggest that the presence of the F pilus or F-like R pilus on recipient cells ensures maximum expression of the exclusion phenotype but is not essential for its expression. In contrast to previous suggestions, we found no evidence for a reduction of entry exclusion attributable to the DNA temperature-sensitive chromosomal mutation dnaB(TS).     

Dynamics of plasmid transfer on surfaces

A protocol was developed to study the dynamics of growth and plasmid transfer in surface populations of bacteria. This method allows for quantitative estimates of cell population densities over time, as well as microscopic observations of colony growth and interactions. Using this 'surface slide system' (SSS), the dynamics of the plasmid R1 and its permanently derepressed mutant R1drd19 in surface cultures of Escherichia coli K12 was examined. In surface culture, the stationary-phase cell densities were constant over a wide range of initial cell density (= colony density) and comparable to those obtained in liquid culture. For high initial cell densities, where the cells formed a confluent layer at stationary phase, the kinetics of growth and plasmid transfer was similar to that obtained in liquid culture, and the relative yields of R1drd19 and R1 transconjugants were similar in the two habitats. In surface culture, however, R1drd19 transconjugant yield was profoundly affected, and R1 transfer to a lesser extent, by colony density. In contrast, liquid matings were virtually unaffected by initial cell density. The transfer advantage of the permanently depressed over the repressed plasmid was much less apparent for lower colony densities. I propose a hypothesis for plasmid transfer between colonies that explains these observations as a consequence of the geometry of the surface habitat and the effect of transitory derepression of the synthesis of pili.     

ColB2-K77, a Fertility-Repressed F-Like Sex Factor

The colicinogenic B factor, transferred from Escherichia coli strain K77 (and termed ColB2-K77 or ColB2) to an E. coli K12 F(-) strain, is capable of promoting its own transfer to other K12 F(-) strains at a low rate (from LFC cultures) which can be increased under special conditions (HFC cultures). LFC cultures of K12 (ColB2)(+) F(-) strains show a low level of adsorption of F-specific phage particles which also increases under HFC conditions. The ColB2 factor is thus inferred to be an F-like sex factor which is repressed in its fertility. This repression is concluded to be due to a cytoplasmic repressor since, when ColB2 is present in cells containing an F factor (either autonomous or integrated), F fertility is also repressed as shown by the inability of such (ColB2)(+)F(+) [or (ColB2)(+)Hfr] strains to plaque F-specific phages, and by a reduction in the level of chromosomal transfer from such strains, compared to the corresponding F(+) (or Hfr) control strains. Mutants of the ColB2 factor in which fertility is no longer repressed (fertility derepressed or Fdr mutants) have been isolated. The ColB2Fdr mutant strains do not appear to be able to mobilize chromosomal transfer, although they have acquired F-specific phage sensitivity demonstrable by plaque formation and they transfer their colicin factor at high frequency and are well piliated. The Fdr mutation is presumed to result in the inability to synthesize the cytoplasmic fertility repressor since the ColB2Fdr factor does not repress the fertility of an F factor when present in the same host strain. A fertility-repressed drug resistance factor of the R(f) type is not stable in the presence of a ColB2 factor in the same cell and is eliminated in about 10% of the cells per generation. In contrast, another factor characteristic of the R(i) type is fully compatible with ColB2. Under conditions artificially stabilizing (ColB2Fdr)(+) (Rf)(+) strains, the enhanced fertility of ColB2Fdr is not repressed by the presence of the R factor, nor does the presence of R(f) in the intermediate strain of an HFC (for ColB2) system inhibit the normal increase in ColB2 transmissibility. It is concluded that the repressors of R(f) and ColB2, although both active on F fertility, are different; this may indicate that at least two independently repressible cistrons are involved in the expression of fertility characteristics.     

The role of lipopolysaccharide structure in the recipient cell during plasmid-mediated bacterial conjugation

The role of the lipopolysaccharide (LPS) structure in the recipient cell during bacterial conjugation was studied using a series of well-defined LPS mutations in Salmonella minnesota. The plasmids Flac (IncFI) and R1drd19 (IncFII) transferred at a high frequency to the smooth S218 parent strain and the rough LPS mutants. However, R64drd1 1 (IncI alpha) transferred poorly to the LPS mutants compared with transfer to the smooth LPS parent strain. The decrease in R64drd1 1 transfer frequency correlated with the extent of the defect in LPS structure, suggesting that intact LPS on the recipient is a major requirement for R64drd1 1 mating. Transfer of the P-group plasmid, RK2, was not affected by changes in LPS structure. These results show that plasmids use different cell surface structures during conjugation, and that LPS is particularly important for R64drd1 1 transfer.     

Physical Properties and Mechanism of Transfer of R Factors in Escherichia coli

The physical properties of F-like and I-like R factors have been compared with those of the wild-type F factor in Escherichia coli K-12 unmated cells and after transfer to recipient cells by conjugation. The F-like R factor R538-1drd was found to have a molecular weight of 49 x 10(6), whereas the molecular weight of the I-like R factor R64drd11 was 76 x 10(6). The wild-type F factor, F1, had a molecular weight of 62 x 10(6). When conjugation experiments are performed by using donor strains carrying these derepressed F-like or I-like R factors, the transferred deoxyribonucleic acid can be isolated as a covalently closed circle from the recipient cells. This circular deoxyribonucleic acid was characterized by making use of the observation that the complementary strands of these R factors can be separated in a CsCl-poly (U, G) equilibrium gradient. The results of the strand-separation experiments show that only one of the complementary strands of the R factor is transferred from the donor to the recipient. With both the F-like and I-like R factors, this strand is the heavier strand in CsCl-poly (U, G). These results indicate that even though F-like and I-like R factors differ greatly in many properties (phage specificity, size, compatability, etc.), they are transferred by a similar mechanism.     

Variation in expression of sex factor genes in the Proteus-Providencia group relative to Escherichia coli.

Several instances of anomalous expression of genes introduced from Escherichia coli K-12 into Proteus mirabilis have been described. It is shown here that control of sex pilus synthesis directed by the F-like R factor R1 and its depressed derivatives R1-16 (O-C) and R1-19 (i-minus) is also anomalous in P. mirabilis. Piliation in cells bearing the depressed plasmids is expressed at a lower level than in E. coli K-12, and repression is absent in R1-carrying cells. Preliminary results show a similar effect in Providencia. In Proteus morganii, a similarly reduced level of piliation in R1-16-+ or R1-19-+ cultures is observed, but an intermediate level of repression occurs in R1-+ cultures. Less extensive data suggest that expression of the sex factor genes of an R factor of the N incompatibility group differs far less between E. coli and P. mirabilis hosts. Possible bases for these effects are discussed.     

Genetic and physical characteristics of an enterotoxin plasmid.

We are engaged in the genetic and physical characterization of an enterotoxin (Ent) plasmid, Ent P307, which contains genes for the production of a hear-labile and a heat-stable enterotoxin. We are using an Escherichia coli K-12 strain, 711 (P307), constructed by S. Falkow, which contains no other plasmids besides Ent P307. Our genetic studies have shown that the plasmid is incompatible with the sex factor F, both in the integrated (Hfr) and the autonomous (F-prime) state. Ent P307 can thus be assigned to incompatibility group FI. An R factor, R386, which belongs to the same incompatibility group, was also found to be incompatibile with Ent P307, whereas five other R factors belonging to different incompatibility groups were compatible with Ent P307. In the presence of Ent P307, conjugal transfer and sensitivity to a male-specific phage of a derepressed F-like R factor, R1drd19, were repressed. Ent P307 is, thus, finO+. Presumably, it also causes repression of its own transfer genes since conjugal transfer of Ent P307 could not be demonstrated. Unlike F, it does not restrict the growth of female-specific phage phiII. From physical studies on extracted deoxyribonucleic acid, the molecular weight of Ent P307 was determined to be 54 X 10(6). By electron microscope heteroduplex analysis, the plasmid was found to be homologous with F in four regions, encompassing about half of its length. One long region and two short ones contain genes for conjugal transfer; the other short region carries genes for replication and incompatibility.     

Naturally Occurring R Factor, Derepressed for Pilus Synthesis, Belonging to the Same Compatibility Group as the Sex Factor F of Escherichia coli K-12

A naturally occurring R factor with constitutive pilus synthesis is described which resembles the sex factor F in compatibility and in restricting coliphage T7. Unlike F, it is not cured during growth with acridine orange. Results suggest that the R factor produces repressor of pilus synthesis, to which the operator is insensitive (i(+)o(c)). In this respect it differs from both the F factor (i(-)o(+)) and wild-type F-like R factors (i(+)o(+)).     

finO sequences on conjugally repressed and derepressed F-like plasmids

DNA-DNA hybridization studies have demonstrated the physical relatedness of the fertility inhibition gene, finO, among both FinO+ and FinO- F-like conjugative plasmids, viz. ColV2-K94, R100-1,R1drd19,R1,R6-5, UCR123, R386, p307, R453, R773, and pIP162-1. Furthermore, the data indicate that finO sequences on the FinO- plasmid ColV2-K94 map downstream of the transfer region, within 93.6-95.3 ColV2-K94.     

Pilin-like proteins in the extremely thermophilic bacterium Thermus thermophilus HB27: implication in competence for natural transformation and links to type IV pilus biogenesis

The extreme thermophile Thermus thermophilus HB27 exhibits high frequencies of natural transformation. Although we recently reported identification of the first competence genes in Thermus, the molecular basis of DNA uptake is unknown. A pilus-like structure is assumed to be involved. Twelve genes encoding prepilin-like proteins were identified in three loci in the genome of T. thermophilus. Mutational analyses, described in this paper, revealed that one locus, which contains four genes that encode prepilin-like proteins (pilA1 to pilA4), is essential for natural transformation. Additionally, comZ, a new competence gene with no similarity to known genes, was identified. Analysis of the piliation phenotype revealed wild-type piliation of a pilA1-pilA3Deltakat mutant and a comZ mutant, whereas a pilA4 mutant was found to be completely devoid of pilus structures. These findings, together with the significant similarity of PilA4 to prepilins, led to the conclusion that the T. thermophilus pilus structures are type IV pili. Furthermore, the loss of the transformation and piliation phenotype in the pilA4 mutant suggests that type IV pili are implicated in natural transformation of T. thermophilus HB27.     

Plasmid Specificity of The Origin of Transfer of Sex Factor F

The ability of F-like plasmids to promote transfer from the F origin of transfer was determined. Chromosome transfer was measured from plasmid derivatives of RecA(-) Hfr deletion strains which had lost all the F transfer genes but which in some cases retained, and in others had also lost, the origin sequence. ColV2 and ColVBtrp could initiate transfer from the F origin, but R100-1, R1-19, and R538-1 drd could not. These results can be correlated with the plasmid specificity of the traI components of the different plasmid transfer systems, supporting the hypothesis that the origin of transfer is the site of action of the traI product. Most F-like plasmids, including R1-19 and R538-1 drd, could transfer ColE1, consistent with previous findings that the (plasmid-specific) traI product is not necessary for ColE1 transfer by Flac; ColE1 transfer may be initiated by a ColE1-or host-determined product. R100-1 and R136fin(-) could not transfer ColE1 efficiently, apparently because of differences residing in their pilus-forming genes.     

Electron microscope heteroduplex studies of sequence relations among plasmids of Escherichia coli: II. Structure of drug resistance (R) factors and f factors

The colicinogenic plasmid Col V-K94 and the class of transferable drug-resistance factors that inhibit fertility of the sex factor, F, are believed to be genetically similar to F. Heteroduplexes between these various F-like plasmids were studied by electron microscopy in order to identify and define the DNA segments that contain genes coding for fertility and other common functions. It was found that approximately 44% of the F factor is homologous to several different fi⁺ R factors and to Col V-K94, and that all of the homology between these plasmids and F is restricted to a region comprising only one-half of the F-factor molecule. These data, and previously reported electron microscope mapping of deletions of F factors, suggest that the genes active in the process of transfer are contained in one-half of the F factor, and therefore imply that the other half of F might not be required for expression of its fertility functions.DNA sequence homology among the R factors R6, R100 and R1 was also studied by heteroduplex formation. All of the DNA sequences contained in R100 were observed to be present in R6, and approximately 85% of the DNA sequences of R1 were included in R6. In the $\text{R1}\text{R6}$ heteroduplex, homology is distributed equally in the R-factor region homologous to the F factor, and the region containing antibiotic-resistance determinants (i.e. the segment deleted in the resistance transfer factor unit, RTF). The position of certain drugresistance determinants has been mapped by using various R-factor deletion mutants and correlating the pattern of drug resistance with the sequence homology observed among different R factors. Unlike the other resistance-determinant genes identified, the tetracycline marker has been mapped in the segment of R6 that is largely homologous to the F factor.Inverted repeats, which consist of DNA segments repeated in reverse sequence on the same single strand of DNA, were observed in R6, R100 and Col V-K94. One inverted repeat of R6 is located near the tetraeycline marker. Further duplication of one of the complementary sequences of the inverted repeat, and insertion of this duplication into the plasmid genome apparently inactivates the tetracycline-resistance determinant of R6.     

General organization of the conjugal transfer genes of the IncW plasmid R388 and interactions between R388 and IncN and IncP plasmids.

The complete conjugal transfer gene region of the IncW plasmid R388 has been cloned in multicopy vector plasmids and mapped to a contiguous 14.9-kilobase segment by insertion mutagenesis. The fertility of the cloned region could still be inhibited by a coresident IncP plasmid. The transfer region has been dissected into two regions, one involved in pilus synthesis and assembly (PILW), and the other involved in conjugal DNA metabolism (MOBW). They have been separately cloned. PILW also contains the genes involved in entry exclusion. MOBW contains oriT and the gene products required for efficient mobilization by PILW. MOBW plasmids could also be mobilized efficiently by PILN, the specific pilus of the IncN plasmid pCU1, but not by PILP, the specific pilus of the IncP plasmid RP1.     

In situ monitoring of IncF plasmid transfer on semi-solid agar surfaces reveals a limited invasion of plasmids in recipient colonies

Most natural conjugative IncF plasmids encode a fertility inhibition system that represses transfer gene expression in the majority of plasmid-carrying cells. The successful spread of these plasmids in clinically relevant bacteria has been suggested to be supported by a transitory derepression of transfer gene expression in newly formed transconjugants. In this study, we aimed to monitor the extent of transitory derepression during agar surface matings in situ by comparing plasmid spread of the IncF plasmid R1 and its derepressed mutant R1drd19 at low initial cell densities. A zygotic induction strategy was used to visualize the spatial distribution of fluorescent transconjugants within the heterogeneous environment. Epifluorescence and confocal microscopy revealed different transfer patterns for both plasmids, however, spread beyond the first five recipient cell layers adjacent to the donor cells was not observed. Similar results were observed for other prototypical conjugative plasmids. These results cannot rule out that transitory derepression contributes to the limited R1 plasmid invasion, but other factors like nutrient availability or spatial structure seem to limit plasmid spread.     

Effects of recA Mutations on Pilus Antigenic Variation and Phase Transitions in Neisseria gonorrhoeae

Intragenic recombination between the single complete pilin gene (expression locus) and multiple, distinct, partial pilin gene copies (silent, storage loci) is thought to account for the generation of pilus antigenic diversity and piliation phase (on-off) changes exhibited by Neisseria gonorrhoeae. The mechanisms operating in the genomic rearrangements associated with these forms of pilus variation were investigated through the study of isogenic strains of gonococci bearing either wild-type or altered recA alleles. Examination of the rates of pilus phase variation and the genetic basis for changes in piliation status displayed by these strains show that recA mediated homologous recombination is required for these high frequency events and confirm that the nonpiliated state results from mutations in the expressed pilin gene. In a strain that is deficient in recA mediated homologous recombination, pilus phase variation occurs at a 100-1000-fold reduced rate and results predominantly from one class of spontaneous frameshift mutations within the pilin structural gene.     

The structure of R1drd19: a revised physical map of the plasmid

We have analyzed derivatives of the plasmid R1drd19 carrying the transposon Tn10 by electron microscopy following denaturation and renaturation of the molecules, and by digestion with various restriction enzymes, gel electrophoresis and Southern blotting. We show: 1) that the published restriction map of R1drd19 is inconsistent with our results. We present a modified map which is consistent with our data. 2) that R1drd19 carries a single resident copy of the element IS10 which is normally associated with Tn10 as an inverted repeat, and 3) that R1drd19 carries three copies of the insertion element IS1 in the resistance determinant region.     

Requirements for suppression of a dnaG mutation by an I-type plasmid.

We have confirmed and further characterized the phenomenon of suppression of a dnaG mutation by an I plasmid and determined the requirements for the complete expression of suppressibility. The authentic derepressed mutation of the conjugal fertility system, described earlier, was shown to be one of the requirements. The second requirement is a previously undescribed type of mutation leading to a far higher degree of derepression of this system than the authentic drd mutation. This second mutation leads to extremely high conjugal fertility and pilus production. A third requirement for the complete expression of suppressibility is that the host recA function remain intact. recA function is required not merely for the fulfillment of the other two requirements described above, but also for an additional step or steps leading to its complete expression. The nature of this step(s) is unknown, but it is not insertion of the plasmid into the host genome.     

Partial suppression of the phenotype of Escherichia coli K-12 dnaG mutants by some I-like conjugative plasmids.

Three I-like conjugative plasmids, ColIdrd1, R144drd3, and R64drd11, which are derepressed for functions involved in conjugation, were found to suppress at least partially the phenotype of temperature-sensitive dnaG mutants of Escherichia coli K-12, as judged from the kinetics of deoxyribonucleic acid synthesis at elevated temperature in newly formed and established plasmid-containing strains. In contrast, the corresponding wild-type plasmids and three F-like derepressed conjugative plasmids, F101, R100drd1, and R1drd16, all failed to suppress. Suppression is presumably caused by a different plasmid-determined function from that which promotes survival of ultraviolet-irradiated bacteria, because both the wild-type I-like plasmids and their drd mutants protected irradiated bacteria. One possible interpretation of these results is that the product of a gene carried by certain I-like plasmids can substitute for the bacterial dnaG gene product during ongoing deoxyribonucleic acid replication.     

Twelve pil genes are required for biogenesis of the R64 thin pilus

The IncI1 plasmid R64 produces two kinds of sex pili: a thin pilus and a thick pilus. The thin pilus, which belongs to the type IV family, is required only for liquid matings. Fourteen genes, pilI to -V, were found in the DNA region responsible for the biogenesis of the R64 thin pilus (S.-R. Kim and T. Komano, J. Bacteriol. 179:3594-3603, 1997). In this study, we introduced frameshift mutations into each of the 14 pil genes to test their requirement for R64 thin pilus biogenesis. From the analyses of extracellular secretion of thin pili and transfer frequency in liquid matings, we found that 12 genes, pilK to -V, are required for the formation of the thin pilus. Complementation experiments excluded the possible polar effects of each mutation on the expression of downstream genes. Two genes, traBC, were previously shown to be required for the expression of the pil genes. In addition, the rci gene is responsible for modulating the structure and function of the R64 thin pilus via the DNA rearrangement of the shufflon. Altogether, 15 genes, traBC, pilK through pilV, and rci, are essential for R64 thin pilus formation and function.     

Mutations in R Factors of Escherichia coli Causing an Increased Number of R-Factor Copies per Chromosome

A mutant of the repressed R factor R1a and two mutants of the derepressed R factor R1drd-19 showing a two- to fourfold increase in resistance to all of the antibiotics to which the wild-type R factors mediate resistance were studied. The increased resistance was due to a two- to fourfold increase in the number of R-factor copies per chromosome. The production of drug-metabolizing enzymes was linearly correlated to the gene dosage. There was also a linear correlation between resistance to the drugs and the production of the corresponding enzymes. The mutations were also expressed in Proteus mirabilis PM1. In Proteus, R factors are split into two plasmids, resistance transfer factor and the resistance part. The mutation in one of the mutant R factors seems to be located in the resistance part. A second fi(+) R factor (R100) was introduced into strains already carrying R1drd-19 or the mutant R factor R1drd-19B2. In the first case, R100 and R1drd-19 segregated with equal probability when the bacteria were grown on antibiotic-free medium, whereas, in the second case, R100 was rapidly and preferentially excluded.