Establishment of Escherichia coli cells with an integrated high copy number plasmid

Integration of the pSLT Plasmid into the Salmonella Chromosome Results in a Temperature-Sensitive Growth Defect Due to Aberrant DNA Replication

A mutant of Salmonella enterica serovar Typhimurium was isolated that simultaneously affected two metabolic pathways as follows: NAD metabolism and DNA repair. The mutant was isolated as resistant to a nicotinamide analog and as temperature-sensitive for growth on minimal glucose medium. In this mutant, Salmonella's 94-kb virulence plasmid pSLT had recombined into the chromosome upstream of the NAD salvage pathway gene pncA This insertion blocked most transcription of pncA, which reduced uptake of the nicotinamide analog. The pSLT insertion mutant also exhibited phenotypes associated with induction of the SOS DNA repair system, including an increase in filamentous cells, higher exonuclease III and catalase activities, and derepression of SOS gene expression. Genome sequencing revealed increased read coverage extending out from the site of pSLT insertion. The two pSLT replication origins are likely initiating replication of the chromosome near the normal replication terminus. Too much replication initiation at the wrong site is probably causing the observed growth defects. Accordingly, deletion of both pSLT replication origins restored growth at higher temperatures.IMPORTANCE In studies that insert a second replication origin into the chromosome, both origins are typically active at the same time. In contrast, the integrated pSLT plasmid initiated replication in stationary phase after normal chromosomal replication had finished. The gradient in read coverage extending out from a single site could be a simple but powerful tool for studying replication and detecting chromosomal rearrangements. This technique may be of particular value when a genome has been sequenced for the first time to verify correct assembly.

Establishing a System for Testing Replication Inhibition of the Vibrio cholerae Secondary Chromosome in Escherichia coli

Regulators of DNA replication in bacteria are an attractive target for new antibiotics, as not only is replication essential for cell viability, but its underlying mechanisms also differ from those operating in eukaryotes. The genetic information of most bacteria is encoded on a single chromosome, but about 10% of species carry a split genome spanning multiple chromosomes. The best studied bacterium in this context is the human pathogen Vibrio cholerae, with a primary chromosome (Chr1) of 3 M bps, and a secondary one (Chr2) of about 1 M bps. Replication of Chr2 is under control of a unique mechanism, presenting a potential target in the development of V. cholerae-specific antibiotics. A common challenge in such endeavors is whether the effects of candidate chemicals can be focused on specific mechanisms, such as DNA replication. To test the specificity of antimicrobial substances independent of other features of the V. cholerae cell for the replication mechanism of the V. cholerae secondary chromosome, we establish the replication machinery in the heterologous E. coli system. We characterize an E. coli strain in which chromosomal replication is driven by the replication origin of V. cholerae Chr2. Surprisingly, the E. coli ori2 strain was not inhibited by vibrepin, previously found to inhibit ori2-based replication.

Chromosomal Replication Complexity: A Novel DNA Metrics and Genome Instability Factor

As the ratio of the copy number of the most replicated to the unreplicated regions in the same chromosome, the definition of chromosomal replication complexity (CRC) appears to leave little room for variation, being either two during S-phase or one otherwise. However, bacteria dividing faster than they replicate their chromosome spike CRC to four and even eight. A recent experimental inquiry about the limits of CRC in Escherichia coli revealed two major reasons to avoid elevating it further: (i) increased chromosomal fragmentation and (ii) complications with subsequent double-strand break repair. Remarkably, examples of stable elevated CRC in eukaryotic chromosomes are well known under various terms like "differential replication," "underreplication," "DNA puffs," "onion-skin replication," or "re-replication" and highlight the phenomenon of static replication fork (sRF). To accurately describe the resulting "amplification by overinitiation," I propose a new term: "replification" (subchromosomal overreplication). In both prokaryotes and eukaryotes, replification, via sRF processing, causes double-strand DNA breaks and, with their repair elevating chromosomal rearrangements, represents a novel genome instability factor. I suggest how static replication bubbles could be stabilized and speculate that some tandem duplications represent such persistent static bubbles. Moreover, I propose how static replication bubbles could be transformed into tandem duplications, double minutes, or inverted triplications. Possible experimental tests of these models are discussed.

Homologous Recombination-Experimental Systems, Analysis, and Significance

Homologous recombination is the most complex of all recombination events that shape genomes and produce material for evolution. Homologous recombination events are exchanges between DNA molecules in the lengthy regions of shared identity, catalyzed by a group of dedicated enzymes. There is a variety of experimental systems in Escherichia coli and Salmonella to detect homologous recombination events of several different kinds. Genetic analysis of homologous recombination reveals three separate phases of this process: pre-synapsis (the early phase), synapsis (homologous strand exchange), and post-synapsis (the late phase). In E. coli, there are at least two independent pathway of the early phase and at least two independent pathways of the late phase. All this complexity is incongruent with the originally ascribed role of homologous recombination as accelerator of genome evolution: there is simply not enough duplication and repetition in enterobacterial genomes for homologous recombination to have a detectable evolutionary role and therefore not enough selection to maintain such a complexity. At the same time, the mechanisms of homologous recombination are uniquely suited for repair of complex DNA lesions called chromosomal lesions. In fact, the two major classes of chromosomal lesions are recognized and processed by the two individual pathways at the early phase of homologous recombination. It follows, therefore, that homologous recombination events are occasional reflections of the continual recombinational repair, made possible in cases of natural or artificial genome redundancy.

Two-step method for curing Escherichia coli of ColE1-derived plasmids

To cure Escherichia coli for plasmids derived from the ColE1 replicon advantage is taken of the fact that maintenance of this replicon requires a wild-type allele of polA, encoding DNA polymerase I. Curing is achieved by cotransduction of a mutant polA allele with metE::Tn10, fadAB::Tn10 or other transposon insertions near polA. Reciprocal transduction to Met(+) Pol(+) or to Fad(+) Pol(+) ensures reestablishment of the original genotype except for loss of the plasmid. A set of useful bacterial strains is provided.

Nucleotide sequence, functional characterization and evolution of pFKN, a virulence plasmid in Pseudomonas syringae pathovar maculicola

Pseudomonas syringae pv. maculicola strain M6 (Psm M6) carries the avrRpm1 gene, encoding a type III effector, on a 40 kb plasmid, pFKN. We hypothesized that this plasmid might carry additional genes required for pathogenesis on plants. We report the sequence and features of pFKN. In addition to avrRpm1, pFKN carries an allele of another type III effector, termed avrPphE, and a gene of unknown function (ORF8), expression of which is induced in planta, suggesting a role in the plant-pathogen interaction. The region of pFKN carrying avrRpm1, avrPphE and ORF8 exhibits several features of pathogenicity islands (PAIs). Curing of pFKN (creating Psm M6C) caused a significant reduction in virulence on Arabidopsis leaves. However, complementation studies using Psm M6C demonstrated an obvious virulence function only for avrRpm1. pFKN can integrate and excise from the chromosome of Psm M6 at low frequency via homologous recombination between identical sequence segments located on the chromosome and on pFKN. These segments are part of two nearly identical transposons carrying avrPphE. The avrPphE transposon was also detected in other strains of P. s. pv. maculicola and in P. s. tomato strain DC3000. The avrPphE transposon was found inserted at different loci in different strains. The analysis of sequences surrounding the avrPphE transposon insertion site in the chromosome of Psm M6 indicates that pFKN integrates into a PAI that encodes type III effectors. The integration of pFKN into this chromosomal region may therefore be seen as an evolutionary process determining the formation of a new PAI in the chromosome of Psm M6.

A pot-pourri of plasmid paradoxes: effects of a second copy

Bacterial plasmids are exemplary subjects for study, being conveniently isolated, dissected, reassembled, and introduced into various hosts. Their versatility and power make them eminently worthy of our attention. In what follows I consider some consequences of simply doubling the dosage of particular plasmid genes or of forming a plasmid dimer. These consequences can be perverse, paradoxical, or informative. They bear on questions of cell viability, copy number limitation, clonal homogeneity, check-point control, and the recovery of mutants. They have relevance to biotechnology, evolution and medicine. In reviewing these effects, my motivation is largely to share my enthusiasm for certain kinds of biological narratives, the nature of which is best left for the reader to discern.

Recombination enhancement by replication (RER) in Rhizobium etli

Studies in several organisms show that recombination and replication interact closely. Recombinational repair usually requires associated replication at some stage; moreover, additional replication can induce recombination through either homologous or illegitimate events. In prokaryotes, stimulation of recombination by replication is more dramatic when rolling circle replication is employed. In contrast, theta-type replication induces only a modest increase in recombination frequency. In this article, we show that induction of theta-type replication from a supernumerary origin in the symbiotic plasmid (pSym) of Rhizobium etli leads to a 1000-fold increase in deletion formation on this plasmid. These deletions span 120 kb (the symbiotic region) and have as endpoints the reiterated nitrogenase operons. We have named this phenomenon RER, for recombination enhancement by replication. RER is not affected by the position of the replication origin in the pSym, the direction of advance of the replication fork, or the distance from the origin to the recombining repeats. On the other hand, RER is dependent on an active recA allele, indicating that it is due to homologous recombination. RER displays a strong regionality restricted to the symbiotic region. The similarities and differences of RER with the recombination process observed at the terminus of replication of the Escherichia coli chromosome are discussed.

Towards single-copy gene expression systems making gene cloning physiologically relevant: lambda InCh, a simple Escherichia coli plasmid-chromosome shuttle system

We describe a simple system for reversible, stable integration of plasmid-borne genes into the Escherichia coli chromosome. Most ordinary E. coli strains and a variety of pBR322-derived ampicillin-resistant plasmids can be used. A single genetic element, a lambda phage, is the only specialized vector required. The resultant strains have a single copy of the plasmid fragment inserted stably at the lambda attachment site on the chromosome, with nearly the entire lambda genome deleted.

Site-directed insertion and insertion-deletion mutations in the Escherichia coli chromosome simplified

A procedure to produce an exact chromosomal replica of an insertion or insertion-deletion mutation produced in vitro in a plasmid with a ColE 1 origin of replication is presented. This procedure uses a previously described property of recD mutations (Biek DP, Cohen SN. J. Bacteriol. 1986;167:594-603) and is limited by (1) the compatibility of the new mutation with recD; and (2) the presence of some Escherichia coli DNA flanking the mutation.

Conversion to bidirectional replication after unidirectional initiation from R1 plasmid origin integrated at oriC in Escherichia coli

The cell division phenotypes of Escherichia coli with its chromosome replication driven by oriR (from plasmid R1) were examined by fluorescence microscopy and flow cytometry. Chromosome replication patterns in these strains were followed by marker frequency analyses. In one of the strains, the unidirectional oriR was integrated so that the replication fork moved clockwise from the oriC region, and bacterial growth and division were similar to those of the wild-type parent. The bacteria were able to convert the unidirectional initiation from oriR into bidirectional replication. The site for conversion of uni- to bidirectional replication seemed to be localized and could be mapped genetically within 6 min to the immediate right of the minimal oriC. Replication starting in the counterclockwise direction from the R1 replicon integrated at the same site in the opposite orientation could not be described as either bi- or unidirectional, as no single predominant origin could be discerned from the more or less flat marker frequency pattern. These strains also showed extensive filamentation, irregular nucleoid distribution and the presence of anucleate cells, indicative of segregation and division defects. Comparison among intR1 derivatives differing in the position of the integrated oriR relative to the chromosome origin suggested that the oriC sequence itself was dispensable for the conversion to bidirectionality. However, passage of the replication fork over the 6 min region to the right of oriC seemed important for the bidirectional replication pattern and normal cell division phenotype.

Adaptation to sulfonamide resistance in Neisseria meningitidis may have required compensatory changes to retain enzyme function: kinetic analysis of dihydropteroate synthases from N. meningitidis expressed in a knockout mutant of Escherichia coli

Previously, the effects of three point mutations (at amino acid positions 31, 84, and 194) in the gene coding for a sulfonamide-resistant dihydropteroate synthase of Neisseria meningitidis were analyzed by site-directed mutagenesis. Changes at positions 31 and 194 abolished the phenotypic expression of sulfonamide resistance, while a change at position 84 appeared to be neutral. These studies are here extended to correlate the alterations in phenotype with effects on enzyme kinetics by expressing the cloned meningococcal genes in an Escherichia coli strain that had its dhps gene partially deleted and replaced by a resistance determinant. The most dramatic effects were produced by mutations at position 31. A change from the Leu found in the resistant isolate to a Phe (the residue found in sensitive isolates) led to a 10-fold decrease in the Km and a concomitant drop in the Ki. Changes at position 194 also affected both the Km and Ki but not to the same extent as mutations at position 31. Changing position 84 altered the Km only slightly but significantly. This latter change was interpreted as a compensatory change modulating the function of the enzyme. In another type of resistance gene, 2 amino acid residues, proposed to be an insertion, were deleted, resulting in a sensitive enzyme. However, the resulting Km was raised 10-fold, suggesting that compensatory changes have accumulated in this type of resistance determinant as well. This resistance gene differs by as much as 10% from the sensitive isolates, which makes identification of important mutations difficult.

Recombinational rescue of the stalled DNA replication fork: a model based on analysis of an Escherichia coli strain with a chromosome region difficult to replicate

To examine the physiological effects of DNA replication arrest at the terminus (Ter), we constructed a replication-blocked Escherichia coli strain so that both bidirectional replication forks would be impeded at two flanking Ter sites, one artificial and the other natural. While the blocked strain grew slightly more slowly than a control strain, it had abnormal phenotypes similar to those of E. coli dam mutants, i.e., hyper-Rec phenotype, recA(+)- and recB+ (C+)-dependent growth, and constitutive SOS induction. The observation that these two apparently unrelated mutants cause similar phenotypes led us to design a model. We propose that the following sequential events may occur in both strains. A double-strand (ds) break occurs at the blocked replication fork in the blocked strain and at the ongoing fork in the dam mutant, through which RecBCD enzyme enters and degrades the ds DNA molecule, and the degradation product serves as the signal molecule for SOS induction. When RecBCD enzyme meets an appropriately oriented Chi sequence, its DNase activity is converted to recombinase enzyme, which is able to repair the ds end, recombinationally. this model (i) explains the puzzling phenotype of recA and recB (C) mutants and the SOS-inducing phenotype of polA, lig, and dna mutants under restrictive conditions, (ii) provides an interpretation for the role of the Chi sequence, and (iii) suggests a possible key role for homologous recombination with regard to cell survival following the arrest of DNA replication.

Two enhancer elements for DNA replication of pSC101, par and a palindromic binding sequence of the Rep protein

The minimal replication origin (ori) of the plasmid pSC101 has been previously defined as an approximately 220-bp region by using plasmids defective in the par region, which is a cis-acting determinant of plasmid stability. This ori region contains the DnaA binding sequence, three repeated sequences (iterons), and an inverted repeat (IR) element (IR-1), one of the binding sites of an initiator protein, Rep (or RepA). In the present study, we show that plasmids containing par can replicate at a nearly normal copy number in the absence of IR-1 but still require a region (the downstream region) between the third iteron and IR-1. Because par is dispensable in plasmids retaining IR-1, par and IR-1 can compensate each other for efficient replication. The region from the DnaA box to the downstream region can support DNA replication at a reduced frequency, and it is designated "core-ori." Addition of either IR-1 or par to core-ori increases the copy number of the plasmid up to a nearly normal level. However, the IR-1 element must be located downstream of the third iteron (or upstream of the rep gene) to enhance replication of the plasmid, while the par region, to which DNA gyrase can bind, functions optimally regardless of its location. Furthermore, the enhancer activity of IR-1 is dependent on the helical phase of the DNA double helix, suggesting that the Rep protein bound to IR-1 stimulates the activation of ori via its interaction with another factor or factors capable of binding to individual loci within ori.

Integration of repeated sequences (pBR322) in the Bacillus subtilis 168 chromosome without affecting the genome structure

The Escherichia coli plasmid pBR322 sequence (4363 bp) was integrated at the met, pro, or leuB locus of the Bacillus subtilis chromosome without duplication of the flanking chromosomal regions. The integrated pBR322 was stably maintained as part of the chromosome regardless of its orientation or location. It was found that a DNA segment as large as 17 kb cloned in pBR322 can be readily transferred to the B. subtilis chromosome by transformation. It was demonstrated that a second pBR322 sequence could be effectively introduced at different regions of the chromosome by sequential transformation using chromosomal DNA isolated from a strain that had already acquired a pBR322 sequence at a different locus. Similarly, a third pBR322 sequence could be introduced. By this method, two or three pBR322 sequences can be incorporated at unlinked loci without affecting the overall structure of the B. subtilis genome.

Minimal essential origin of plasmid pSC101 replication: requirement of a region downstream of iterons

The minimal replication origin (ori) of the plasmid pSC101 was defined as an about 220-bp region under the condition that the Rep (or RepA) protein, a plasmid-encoded initiator protein, was supplied in trans. The DnaA box is located at one end of ori, as in other plasmids, like mini-F and P1. The other border is a strong binding site (IR-1) of Rep which is palindromic sequence and lies in an about 50-bp region beyond the repeated sequences (iterons) in ori. This IR-1 is located just upstream of another strong Rep binding site (IR-2), the operator site of the structure gene of Rep (rep), but its function has not been determined. The present study shows that the IR-1 sequence capable of binding to Rep is essential for plasmid replication with a nearly normal copy number. Furthermore, a region between the third iteron and IR-1 is also required in a sequence-specific fashion, since some one-base substitution in this region inactivate the origin function. It is likely that the region also is a recognition site of an unknown protein. Three copy number mutations of rep can suppress any one-base substitution mutation. On the other hand, the sequence of a spacer region between the second and the third iterons, which is similar to that of the downstream region of the third iteron, can be changed without loss of the origin function. The requirement of the region downstream of iterons in pSC101 seems to be unique among iteron-driven plasmid replicons.

Metabolic regulations and biological functions of phospholipids in Escherichia coli

Extensive genetic and biochemical studies in the last two decades have elucidated almost completely the framework of synthesis and turnover of quantitatively major phospholipids in E. coli. The knowledge thus accumulated has allowed to formulate a novel working model that assumes sophisticated regulatory mechanisms in E. coli to achieve the optimal phospholipid composition and content in the membranes. E. coli also appears to possess the ability to adapt phospholipid synthesis to various cellular conditions. Understanding of the functional aspects of E. coli phospholipids is now advancing significantly and it will soon be able to explain many of the hitherto unclear cell's activities on the molecular basis. Phosphatidylglycerol is believed to play the central role both in metabolism and functions of phospholipids in E. coli. The results obtained with E. coli should undoubtedly be helpful in the study of more complicated phospholipid metabolism and functions in higher organisms.

The kilA operon of promiscuous plasmid RK2: the use of a transducing phage (lambda pklaA-1) to determine the effects of the lethal klaA gene on Escherichia coli cells

The kil-kor regulon of promiscuous plasmid RK2 includes the replication initiator gene trfA and several potentially host-lethal kil loci (kilA, kilB, kilC, kilE), whose functions may be involved in plasmid maintenance or broad host range. The kilA locus consists of a single operon of three genes (klaA, klaB, klaC), each of which is lethal when expressed from the klaA promoter in the absence of repressors encoded by korA and korB. In this study, we examined the effects of the unregulated klaA gene on the host cell. Bacteriophage lambda was used to construct a transducing phage (lambda pklaA-1) that allows efficient introduction of the klaA gene into Escherichia coli. Cells lacking korA and korB (to allow uncontrolled expression of klaA) and expressing lambda repressor (to prevent phage lytic growth) are killed by lambda pklaA-1. Cell death is dependent on the klaA structural gene, independent of the SOS system of the host, and is prevented by the presence of korA and korB. lambda pklaA-1 was used to synchronously infect cells lacking korA and korB to determine the effects of klaA on the cells over time. The earliest effects, visible at two hours post-infection, are inhibition of growth of the culture, formation of elongated cells, and striking changes in the appearance of the outer membrane. After four to five hours, the viability of the culture declined sharply and macromolecular synthesis ceased. The distinct class of early events is consistent with the hypothesis that the KlaA polypeptide interacts with a specific target in the host cell.

Introduction of single-copy sequences into the chromosome of Escherichia coli: application to gene and operon fusions

We have developed a general method for the introduction of any cloned sequence into the chromosome of Escherichia coli. This method employs an Hfr strain which carries a fragment of bla (the pBR322 gene imparting ampicillin resistance) between lacI and lacZ. Plasmid-borne inserts which are flanked by sequences from bla and lacZ can be introduced at this locus by homologous recombination. The isolation of recombinants is enhanced by selection for transfer of an integrated copy of the plasmid during conjugation. Once introduced into the chromosome, the inserted sequences can be transferred to other strains by conventional methods such as P1 transduction or conjugation. This method is suitable for the transfer of any cloned sequence to the chromosome and is particularly well suited to the construction of chromosomal gene and operon fusions with lacZ.

ColE1-type vectors with fully repressible replication

We have constructed two cloning vectors, pAM34 and pAM35, derived from pBR322, in which transcription of the replication primer RNA is under control of the lacZpo promoter/operator. These vectors contain a cloning cassette flanked by strong transcriptional terminators. They differ from each other by the presence (pAM34) or absence (pAM35) of gene lacIq. In the presence of repressor LacIq, replication is entirely dependent upon the addition of inducer. This feature allows the temporary maintenance of these plasmids, the construction of strains in which vector derivatives are stably integrated into the chromosome, and the recovery of nucleotide sequences adjacent to cloned fragments. Replication from the integrated plasmid can be adjusted to match the chromosome replication initiation rate required for cell growth in the absence of a functional origin, oriC.

Allelic exchange in Escherichia coli using the Bacillus subtilis sacB gene and a temperature-sensitive pSC101 replicon

To facilitate efficient allelic exchange of genetic information into a wild-type strain background, we improved upon and merged approaches using a temperature-sensitive plasmid and a counter-selectable marker in the chromosome. We first constructed intermediate strains of Escherichia coli K12 in which we replaced wild-type chromosomal sequences, at either the fimB-A or lacZ-A loci, with a newly constituted DNA cassette. The cassette consists of the sacB gene from Bacillus subtilis and the neomycin (kanamycin) resistance gene of Tn5, but, unlike another similar cassette, it lacks IS1 sequences. We found that sucrose sensitivity was highly dependent on incubation temperature and sodium chloride concentration. The DNA to be exchanged into the chromosome was first cloned into derivatives of plasmid pMAK705, a temperature-sensitive pSC101 replicon. The exchanges were carried out in two steps, first selecting for plasmid integration by standard techniques. In the second step, we grew the plasmid integrates under non-selective conditions at 42 degrees C, and then in the presence of sucrose at 30 degrees C, allowing positive selection for both plasmid excision and curing. Despite marked locus-specific strain differences in sucrose sensitivity and in the growth retardation due to the integrated plasmids, the protocol permitted highly efficient exchange of cloned DNA into either the fim or lac chromosomal loci. This procedure should allow the exchange of any DNA segment, in addition to the original or mutant allelic DNA, into any non-essential parts of the E. coli chromosome.

Initiation of chromosomal DNA replication which is stimulated without oversupply of DnaA protein in Escherichia coli

The temperature-sensitive dnaA46 mutation in Escherichia coli can be phenotypically suppressed at 42 degrees C by oversupply of GroELS proteins, and the suppressed cells grow extremely slowly at 30 degrees C. We found that the phenotype of dnaA46 showing this cold sensitivity was dominant over the phenotype of dnaA+, and could not be rescued by introduction of oriC-independent replication systems. These results suggest that the cold sensitivity was not caused by a simple defect in replication. When a growing culture of a dnaA46 strain with a GroELS-overproducing plasmid was shifted from 42 degrees to 30 degrees C in the presence of chloramphenicol, the chromosomal DNA replicated excessively. Initiation of replication occurred at the site of oriC repeatedly four or five times during a 4 h incubation period without concomitant protein synthesis, indicating an excessive capacity for initiation. Such overreplication did not take place at 42 degrees C in the suppressed dnaA46 strain, or at either temperature in GroELS-oversupplied dnaA+ cells. No significant difference was detected between the cellular content of DnaA protein in suppressed cells where the initiation capacity was abnormally high, and that in wild-type cells in which the initiation capacity was normal. Thus, DnaA protein might function in vivo through some phase control mechanism for initiation, apart from a simple regulation by its total amount. A possible mechanism is proposed based on the participation of GroELS proteins in protein folding.

recA gene dependence of replication of the Escherichia coli chromosome initiated by plasmid pUC13 integrated at predetermined sites

Plasmid pUC13 was used to clone DNA fragments of known sites from the chromosome of Escherichia coli. Each chimeric plasmid was introduced individually into the same dnaA46 mutant strain LC381 and suppressive integration (Sin) strains were selected. By means of cotransduction the null mutation recA56 was then introduced into each Sin strain and growth of each recA56 derivative at 42 degrees C was scored. Strains that failed to grow at 42 degrees C depended upon the recA gene for replication. Three factors were shown to limit the viability of LC381 harboring different chimeric plasmids and affect the degree of recA gene dependence of chromosome replication in the Sin strains at 42 degrees C. It is suggested that these three constraints are the consequence of the organization of the E. coli chromosome, particularly the unique ability of terC to retard the progression of replication forks. Two classes of hypotheses concerning the function of the recA gene are considered.

Characterization of malT mutants that constitutively activate the maltose regulon of Escherichia coli

The expression of the maltose regulon of Escherichia coli is controlled by a transcriptional activator, the product of the malT gene, and is induced by the presence of maltose or maltodextrins in the growth medium. We isolated eight mutants with mutations in malT which lead to constitutive expression of the regulon. The nucleotide sequences of the mutated genes revealed that the eight mutations are clustered in two small regions in the first one-third of the malT gene. Two mutated MalT proteins (corresponding to a mutation in each cluster) were purified and examined for in vitro activation of the MalT-dependent malPp promoter. Whereas wild-type MalT activity was absolutely dependent upon the presence of maltotriose, even at high protein concentrations, both mutated proteins were partially active in the absence of this sugar. Indeed, while the activity of the mutated proteins was still increased by maltotriose at low protein concentrations, the proteins were fully active in the absence of maltotriose at high protein concentrations. Both proteins exhibited a fivefold-higher affinity for maltotriose than the wild-type protein did.

The identification of the Escherichia coli ftsY gene product: an unusual protein

The ftsYEX operon in Escherichia coli encodes three proteins, two of which (FtsE and FtsX) are known to be required for cell division. Although FtsE and FtsX have been identified using SDS-PAGE, the FtsY protein has not. We have used in vitro insertion mutagenesis to identify FtsY as a 92 kD polypeptide in maxicell experiments, although predictions from the DNA sequence estimated FtsY to be 54 kD. Results suggest that this disparity could be due to the unusually high percentage of acidic residues within the protein. Complementation tests indicated the presence of a promoter within ftsY required for expression of ftsE and ftsX. The FtsY protein exhibits sequence homology with the SR alpha protein of eukaryotes which is involved in protein secretion. The essential nature of the ftsY gene was also demonstrated.

New method for generating deletions and gene replacements in Escherichia coli

We describe a method for generating gene replacements and deletions in Escherichia coli. The technique is simple and rapid and can be applied to most genes, even those that are essential. What makes this method unique and particularly effective is the use of a temperature-sensitive pSC101 replicon to facilitate the gene replacement. The method proceeds by homologous recombination between a gene on the chromosome and homologous sequences carried on a plasmid temperature sensitive for DNA replication. Thus, after transformation of the plasmid into an appropriate host, it is possible to select for integration of the plasmid into the chromosome at 44 degrees C. Subsequent growth of these cointegrates at 30 degrees C leads to a second recombination event, resulting in their resolution. Depending on where the second recombination event takes place, the chromosome will either have undergone a gene replacement or retain the original copy of the gene. The procedure can also be used to effect the transfer of an allele from a plasmid to the chromosome or to rescue a chromosomal allele onto a plasmid. Since the resolved plasmid can be maintained by selection, this technique can be used to generate deletions of essential genes.

The narK gene product participates in nitrate transport induced in Escherichia coli nitrate-respiring cells

The nucleotide sequence of the Escherichia coli narK gene, which is located in the upstream region of the narCHJI operon, was determined. The narK gene encodes a very hydrophobic protein with 463 amino acid residues (Mr 49,693). A narK deletion mutant, under conditions for the induction of nitrate respiration, was unable to perform nitrate transport. Loss of transport activity was recovered by transforming the mutant with a narK+ plasmid. Thus, we conclude that the narK gene encodes a transmembrane protein participating in nitrate transport. In the narK promoter region, we defined a unique sequence that we designate as a 'nitrate box', functioning as a putative NarL-binding site, in addition to the consensus sequence of the 'anaero-box'.

Identification and characterization of a locus inhibiting extrachromosomal maintenance of the Streptomyces plasmid SLP1

We report here the existence and initial characterization of a genetic locus (imp) that inhibits maintenance of SLP1-derived plasmids as extrachromosomal replicons in a manner distinct from normal incompatibility between autonomous SLP1 replicons. The trans-acting imp function has been localized to a 1.8 kb Eco47III restriction fragment present on integrated SLP1 elements. At least part of this DNA segment is absent from SLP1-derived plasmids. DNA sequence analysis of the imp region indicates that it contains three overlapping open reading frames (ORFs) that may constitute a polycistronic operon. The effects of insertions within the imp region indicate that uninterrupted transcription through all three ORFs is necessary for imp activity.

Genetic analysis of the lytic replicon of bacteriophage P1. II. Organization of replicon elements

The region of bacteriophage P1 DNA containing a lytic (vegetative) replicon has been identified by cloning P1 fragments into a phage lambda vector. We present the sequence of that replicon. Using a novel fusion vector containing two P1 loxP recombination sites, we have developed a transformation assay for replicon function and have used that assay to identify some of the components of the P1 lytic replicon. Among those components is a transcription promoter, P53, whose activity is essential for replicon function. When that promoter is inactivated by the binding of P1 repressor to an operator site, Op53, whose sequence overlaps the promoter, replicon function is blocked. The P53 promoter can be replaced for replicon function by other promoters and, when the lacZ promoter was used, the extent of replication was shown to be proportional to promoter activity. Two open reading frames are located downstream from P53. The promoter-proximal reading frame is 266 amino acid residues long and is not essential for replicon function. In fact, expression of that open reading frame either interferes with plasmid establishment after transformation or is lethal to cells. The promoter-distal reading frame, designated the repL open reading frame, is either 269 or 281 amino acid residues long and is essential for replicon function. Insertion of a Tn5 transposon into the 266 amino acid residue open reading frame inactivates the cloned lytic replicon probably by interfering with the transcription of the repL open reading frame from P53. In P1, this Tn5 insertion mutation completely blocks lytic replication, indicating that the replicon identified here is either the only P1 lytic replicon or, if not, is at least necessary for the function of any other lytic replicon. A four base insertion in the repL open reading frame has largely the same inhibitory effect on phage lytic replication as the Tn5 insertion.

Chromosomal transformation of Escherichia coli recD strains with linearized plasmids

Wild-type Escherichia coli are resistant to genetic transformation by purified linear DNA, probably in part because of exonuclease activity. We demonstrate that E. coli containing a recD mutation could be easily transformed by linearized plasmids containing a selectable marker. The marker was transferred to the chromosome by homologous recombination, whereas plasmid markers not in the region of homology were lost.

The narX and narL genes encoding the nitrate-sensing regulators of Escherichia coli are homologous to a family of prokaryotic two-component regulatory genes

The nucleotide sequence of a 4.4-kilobase SacII-SspI fragment encoding the narXL operon and a part of the narK gene of Escherichia coli has been determined. The narX and narL genes encode proteins of molecular weight 67,275 and 23,927, respectively, and are transcribed from a common promoter, narXp, locating within 429 bases upstream of narX. Transcription from narXp is not significantly induced by nitrate under anaerobiosis, whereas transcription from narK promoter, which overlaps narXp region and is transcribed divergently, is fully induced by nitrate. The N-terminal two-thirds of the NarL protein has extensive homology with those of a diverse set of prokaryotic regulatory proteins, including OmpR, PhoB, SfrA, UhpA, CheY, CheB, NtrC, DctD, FixJ, VirG, SpoOF, and SpoOA. A segment locating in the C-terminal half of the NarL protein seems to have potential most likely to form the helix-turn-helix structure characteristic of a class of DNA-binding protein. The protein is considered to play a role as a transcriptional activator of the nitrate reductase operon, narCHJI, and the narK gene. The C-terminal region of the NarX protein also has homology with other regulatory proteins known as counterparts of two-component regulatory systems, such as EnvZ, PhoR, PhoM, CpxA, NtrB, DctB, FixL, and VirA. Presence of two copies of hydrophobic segments in the N-terminal half of the NarX protein suggests the role as a transmembrane receptor sensing nitrate.

Gene replacement and retrieval with recombinant M13mp bacteriophages

We have developed an allele exchange system for shuttling sequences of DNA to and from their original chromosomal loci. Cloned segments of the histidine operon of Salmonella typhimurium and the lactose operon of Escherichia coli served as target sequences and were used to develop the system. Replacement and retrieval of target sequences used the phage M13mp vectors and proceeded through an M13 lysogen intermediate. The intermediates and products of allele exchange were characterized by genetic and hybridization analyses. Several unique properties of M13 lysogens were exploited to devise positive selections to detect integration and excision. These positive selections were used to manipulate phenotypically silent alleles.

Integrative suppression of dnaA46 by broad host-range plasmid R1162

Replication of plasmid R1162 DNA does not require the product of the dnaA gene. An integrated copy of the plasmid can suppress the temperature-sensitive dnaA46 allele when (1) additional plasmid copies are present in the cytoplasm and (2) an inactive replication origin, generated by deletion, is also present in the chromosome. We propose that the inactive origin sets the rate of initiation of chromosome replication at a level compatible with cell viability, possibly by providing additional binding sites for an R1162-encoded protein that is rate-limiting for plasmid replication.

Synthesis and processing of Escherichia coli TEM-beta-lactamase and Bacillus licheniformis alpha-amylase in E. coli: the role of signal peptidase I

A mutant of Escherichia coli, in which signal peptidase I synthesis can be regulated, was constructed. The mutant was used to study the effects of signal peptidase I limitation on the synthesis and efficiency of processing of two proteins: the periplasmic E. coli TEM-beta-lactamase and Bacillus licheniformis alpha-amylase, which also accumulates in the periplasm of E. coli. Signal peptidase I limitation resulted in reduced rates of processing of pre-beta-lactamase and in strong inhibition of synthesis of alpha-amylase. The data suggest that beta-lactamase is processed post-translationally and that an intimate relationship exists between the synthesis and processing of alpha-amylase.

Disruption of the Escherichia coli cls gene responsible for cardiolipin synthesis

The cls gene of Escherichia coli is responsible for the synthesis of a major membrane phospholipid, cardiolipin, and has been proposed to encode cardiolipin synthase. This gene cloned on a pBR322 derivative was disrupted by either insertion of or replacement with a kanamycin-resistant gene followed by exchange with the homologous chromosomal region. The proper genomic disruptions were confirmed by Southern blot hybridization and a transductional linkage analysis. Both types of disruptants had essentially the same properties; cardiolipin synthase activity was not detectable, but the strains grew well, although their growth rates and final culture densities were lower than those of the corresponding wild-type strains and strains with the classical cls-1 mutation. A disruptant harboring a plasmid that carried the intact cls gene grew normally. The results indicate that the cls gene and probably the cardiolipin synthase are dispensable for E. coli but may confer growth or survival advantages. Low but definite levels of cardiolipin were synthesized by all the disruptants. Cardiolipin content of the cls mutants depended on the dosage of the pss gene, and attempts to transfer a null allele of the cls gene into a pss-1 mutant were unsuccessful. We point out the possibilities of minor cardiolipin formation by phosphatidylserine synthase and of the essential nature of cardiolipin for the survival of E. coli cells.

Construction and characterization of glutaredoxin-negative mutants of Escherichia coli

Deoxyribonucleotides, the precursors of DNA, are formed de novo by ribonucleotide reductase, and in vitro thioredoxin or glutathione plus glutaredoxin have been isolated as hydrogen donors. The in vivo hydrogen donor for ribonucleotide reductase is not known. To study this, the Escherichia coli glutaredoxin gene (255 base pairs) was inactivated by inserting a 2-kilobase kanamycin-resistance fragment into the coding sequence of the cloned gene. The inactivated gene was inserted into the E. coli chromosome and mapped to about 18.5 min. A gene replacement technique was used to obtain a strain, A407, that lacked glutaredoxin by radioimmunoassay and by enzymatic assay with ribonucleotide reductase. Glutaredoxin was found not to be essential for viability of E. coli. Thioredoxin is also not essential for viability, as had been shown earlier, but a double mutant lacking glutaredoxin and thioredoxin could not be obtained by P1 transduction on a defined medium, indicating that either thioredoxin or glutaredoxin is essential. In rich medium, very slowly growing, unstable transductants were obtained that at high frequency gave rise to better growing cells. One such isolate, A410, was shown to still lack glutaredoxin and thioredoxin.

In vivo plasmid construction by integrative recombination within a 156 bp sequence homology

pHG165, a pBR322 copy number derivative of pUC8, has a 38-bp polylinker multiple cloning site located at the 5' end of lac alpha. A further 156 bp, 3' to the multiple cloning site, completes the coding sequence for the production of the beta galactosidase alpha peptide. We describe the use of in vivo plasmid-chromosome cointegrates as a construction method that in this instance has enabled us to cross out the pHG165 multiple cloning site to obtain a wild-type beta-galactosidase sequence for the alpha peptide. Because the DNA sequence available for homologous recombination was only 156 bp in length, the frequency of crosses that removed the multiple cloning site was less than 1 x 10(-9). These crosses were easily obtained, however, after amplification by ampicillin selection.

Direct selection for the exchange of alleles between a plasmid and the Escherichia coli chromosome

Recombination is extensively used in order to move alleles between replicons. The exchange of wild-type chromosomal and mutant plasmid-borne alleles is a two-step process entailing the formation of a cointegrate between the entire plasmid and the chromosome, followed by resolution of such cointegrates to give a mutant chromosome and a plasmid carrying the wild-type chromosomal sequence. Often the cointegrate and the resolved forms cannot be distinguished phenotypically. To enable the direct isolation of the resolved products we have developed a positive selection technique. Cells containing a cointegrated plasmid R1 were constructed by transduction using a P1 lysate prepared from cells harbouring a plasmid comprising a mutant chromosomal allele and the so-called omega fragment which carries an aad (aminoglycoside adenylyltransferase) gene. P1 transduction from the cointegrate strain into an SmD recipient allowed direct selection for the resolved complex, since transduction of the aad gene is lethal to an SmD strain.

Plasmid replication stimulates DNA recombination in Bacillus subtilis

The effects of plasmid replication on the frequency of homologous recombination have been investigated. For that purpose Bacillus subtilis strains that carry in their chromosome directly repeated DNA sequences, and an integrated copy of plasmid pE194 either proximal or distal to the repeats, were constructed. The repeat consists either of 3.9 X 10(3) base pBR322 sequences or 2.1 X 10(3) base B. subtilis chromosomal sequences. As plasmid pE194 is naturally thermosensitive for replication, the activity of the replicon could be regulated. Recombination between the repeated sequences was infrequent (about 10(-4) per generation) when the integrated plasmid did not replicate. It was 20 to 450 times higher when the plasmid was allowed to replicate, provided that the repeats were in the proximity of the plasmid. These results show that plasmid replication stimulates DNA recombination.

Isolation and characterization of a substitution mutation in the ompR gene of Salmonella typhimurium LT2

The expression of the genes ompC and ompF encoding major outer membrane proteins is dependent on the ompR-envZ operon. Here we describe the isolation and characterization of an ompR mutation, a single-base-pair change, that results in an Arg-to-Cys substitution. When present in multiple copies, the mutant allele conferred a dominant OmpC- OmpF+ phenotype. Furthermore, the mutant allele exhibited allele-specific negative complementation with other ompR mutations. This ability, together with its dominant character, suggested that the OmpR protein is capable of multimerization.

Evidence for replicative transposition of Tn5 and Tn9

Two basic types of models, conservative and replicative, have been proposed to account for the mechanism of transposition in bacteria. A method was developed to test these models by positive selection of various transposon-promoted events as galactose-resistant colonies from plasmid-containing cells. The results show that recA plays an important role in the transposition of Tn5 and Tn9 in Escherichia coli. All Tn5-promoted events (cointegrates, deletions and transpositions) are suppressed in recA-, and restored in recA+. In the case of Tn9, however, only transpositions (but not cointegrates or deletions) are diminished in recA-. Therefore, the recA function is required for cointegrate formation by Tn5, and for cointegrate resolution by Tn9. Both Tn5 and Tn9 cointegrates segregate transpositions (which can be seen as sectors on indicator plates) in recA+ hosts. In recA-, the unresolved Tn9 cointegrates undergo a second round of cointegrate formation to excise plasmids bearing galactose-resistant deletions. In growing cultures, the proportion of cointegrates declines steadily while transpositions increase so that, in late stages, cultures are rich in transpositions and contain few cointegrates. This explains the failure of previous workers to identify cointegrates as essential intermediates in transposition. Hence, with the exception of the recA requirement, the mechanism of transposition of these composite transposons is not very different from simple transposons like Tn3. It is concluded that transposition of Tn5 and Tn9 is normally a replicative process.

Host participation in plasmid maintenance: dependence upon dnaA of replicons derived from P1 and F

Nonparticipation of the bacterial dnaA gene in plasmid replication has been assumed to be the general rule. In conditional dnaA mutants of Escherichia coli, only plasmid pSC101 has been shown to have a dnaA requirement. Experiments with dnaA null mutants of E. coli, presented here, show that dnaA plays a critical and direct role in the replication of miniplasmids derived from P1 and F as it does in the initiation of bacterial replication. Evidence is also presented for the existence of a dnaA-independent secondary replicon of P1 that is able to drive bacterial chromosome replication but is inadequate to support the maintenance of P1 as a plasmid in E. coli.

In vivo cloning ad characterization of mutations of the regulatory locus ompR of Escherichia coli K12

The product of the ompR gene of E. coli K12 is a positive regulatory protein, which is needed for the expression of the major outer membrane proteins OmpC and OmpF in E. coli K12. A simple in vivo technique was used to transfer three ompR mutations (ompR101, ompR472, ompR4) onto a multicopy plasmid carrying the wild-type ompR gene. The resulting clones were transformed into wild type and corresponding mutant backgrounds to analyze their effects on ompC and ompF expression. All of the cloned ompR mutant alleles exhibited a dominant OmpC- phenotype in an ompR+ background. In addition negative complementation of ompF expression was observed between chromosomal ompR4 and multicopy ompR101 alleles. The results suggest an interaction between different OmpR molecules and thereby support the idea that OmpR can exist as a multimeric protein.

pHG165: a pBR322 copy number derivative of pUC8 for cloning and expression

During the construction of the Messing pUC plasmid series, the rop(rom) gene of pBR322 which mediates the activity of RNAI was deleted. This has resulted in an elevated copy number for the pUC plasmids which makes the expression of beta-galactosidase activity constitutive in a host containing the Iqtss lac repressor. We describe the construction of a new series of vectors which retain the pUC multiple cloning site (MCS) but in which copy number control has been recovered. In addition, the lac alpha/lac promoter expression region has been inserted into a HpaI cassette. This facilitates the movement of recombinant DNA clones within the MCS. It also increases the complementation activity of the lac alpha peptide by an order of magnitude, allowing selection of recombinants by their Lac- phenotype on MacConkey agar.

RNase H and replication of ColE1 DNA in Escherichia coli

Amber mutations within the rnh (RNase H) gene of Escherichia coli K-12 were isolated by selecting for bacteria capable of replicating in a sup+ background replication-defective cer-6 mutant of the ColE1 replicon. The cer-6 mutation is an alteration of one base pair located 160 nucleotides upstream of the unique replication origin of this plasmid. Subsequently, we determined the DNA alterations present within these mutants. ColE1 DNA replicated in rnh(Am) recA cells, indicating that (i) RNase H, which has been shown to be absolutely required for in vitro initiation of ColE1 DNA replication, is dispensable in vivo, and (ii) ColE1 replication in the absence of RNase H is not dependent on "stable DNA replication," which has been reported to be an alternative mode of chromosomal DNA replication. Another class of bacterial mutations was also isolated. These mutations, named herB, suppressed cer-6 replication in rnh+ bacteria. herB mutations mapped close to the polA gene on the E. coli chromosome and increased the activity of DNA polymerase I. These findings suggest that when the DNA polymerase I has an opportunity to initiate DNA synthesis before RNase H acts, the replication-defective cer-6 mutant or the wild-type ColE1 replicates in E. coli.

Suppression of the Escherichia coli dnaA46 mutation by amplification of the groES and groEL genes

A lambda hybrid phage (lambda Sda1), containing an 8.1 kb EcoRI DNA fragment from the Escherichia coli chromosome, was selected on the basis of its ability to suppress bacterial thermosensitivity caused by the dnaA46 mutation. We have shown that this suppression is due to a recA+-dependent amplification of the 8.1 kb fragment; consistent with this observation, cloning of the 8.1 kb fragment into a high copy number plasmid (pBR325) leads also to suppression of dnaA46. In the suppressed strains growing at high temperature, bidirectional replication starts in or near the oriC region and requires the presence of the DnaA polypeptide. These findings suggest that the overproduction of a gene product(s), encoded by the cloned 8.1 kb fragment, can restore dnaA-dependent initiation of replication at high temperature in the oriC region. Genetic mapping shows that the groES (mopB) and groEL (mopA) genes are located on the 8.1 kb suppressor fragment. Further analysis, including in vitro mutagenesis and subcloning, demonstrates that the amplification of the groES and groEL genes is both necessary and sufficient to suppress the temperature sensitive phenotype of the dnaA46 mutation.

A DNA fragment containing the groE genes can suppress mutations in the Escherichia coli dnaA gene

An 8.2 kb fragment of E. coli chromosomal DNA, when cloned in increased copy number, suppresses the dnaA46 mutation, and an abundant protein of about 68 kd (60 kd when measured by us), encoded by the fragment, is essential for the suppression (Takeda and Hirota 1982). Mapping experiments show that the fragment originates from the 94 min region of the chromosome. It encodes several proteins but only one abundant polypeptide of the correct size, the product of the groEL gene. Suppression by the fragment is allele specific; those mutations which map to the centre of the gene are suppressed. Other initiation mutants including dnaA203, dnaA204, dnaA508, dnaAam, dnaC, dnaP and dnaB252 are not suppressed. Most suppressed strains are cold-sensitive suggesting an interaction between the mutant proteins (or their genes) and the suppressing protein or proteins.

Site-directed mutagenesis of the Escherichia coli chromosome near oriC: identification and characterization of asnC, a regulatory element in E. coli asparagine metabolism

We developed a new method for the specific mutagenization of the E. coli chromosome. This method takes advantage of the fact that a pBR322 plasmid containing chromosomal sequences is mobilizable during an Hfr-mediated conjugational transfer, due to an homologous recombination between the E. coli Hfr chromosome and the pBR322 derivative. Transconjugants are screened with a simple selection procedure for integration of mutant sequences in the chromosome and loss of pBR322 sequences. Using this method we specifically inactivated several genes near the E. coli replication origin oriC. We found that a gene coding for asparagine synthetase A. This regulatory mechanism was investigated in detail by determining in vivo regulation of asnA promoter activity by the 17kD protein under different growth conditions. Results obtained also suggest a general regulatory role of the 17kD protein in E. coli asparagine metabolism. Therefore the 17kD gene is proposed to be renamed asnC.

Genetic mapping and DNA sequence analysis of mutations in the polA gene of Escherichia coli

DNA polymerase I of Escherichia coli provides an excellent model for the study of template-directed enzymatic synthesis of DNA because it is a single subunit enzyme, it can be obtained in large quantities and the three-dimensional structure of the polymerizing domain (the Klenow fragment) has recently been determined (Ollis et al., 1985). One approach to assigning functions to particular portions of the structure is to correlate the altered enzymatic behavior of mutant forms of DNA polymerase I with the change in the primary sequence of the protein. Towards this end we have developed a rapid procedure for mapping any polA mutation to a region no larger than 300 base-pairs within the polA gene. Two series of polA deletion mutants with defined end-points were constructed in vitro and cloned into bacteriophage lambda. These phages can then be used to map precisely E. coli polA mutants. Twelve polA- alleles have been mapped in this way and for nine of them the nature of the mutational change has been determined by DNA sequence analysis. Two of the mutations, polA5 and polA6, which affect the enzyme-DNA interaction, provide evidence for the location of the DNA binding region on the polymerase three-dimensional structure.

Molecular cloning of the cls gene responsible for cardiolipin synthesis in Escherichia coli and phenotypic consequences of its amplification

The cls gene responsible for cardiolipin synthesis in Escherichia coli K-12 was cloned in a 5-kilobase-pair DNA fragment inserted in a mini-F vector, pML31, and then subcloned into a 2.0-kilobase-pair fragment inserted in pBR322. The initial selection of the gene was accomplished in a cls pss-1 double mutant that had lesions in both cardiolipin and phosphatidylserine synthases and required either the cls or the pss gene product for normal growth at 42 degrees C in a broth medium, NBY, supplemented with 200 mM sucrose. The cloned gene was identified as the cls gene by the recovery and amplification of both cardiolipin and cardiolipin synthase in a cls mutant as well as by the integration of a pBR322 derivative into its genetic locus at 27 min on the chromosome of a polA1 mutant. The maxicell analysis indicated that a protein of molecular weight 46,000 is the gene product. The cls gene is thus most likely the structural gene coding for cardiolipin synthase. Hybrid plasmids of high copy numbers containing the cls gene were growth inhibitory to pss-I mutants under the above selective conditions, whereas they inhibited neither the growth of pss-I mutants at 30 degrees C nor that of pss+ strains at any temperature. Amplification of cardiolipin synthase activity was observed, but was not proportional to the probable gene dosage (the enzyme activity was at most 10 times that in wild-type cells), and cardiolipin synthesis in vivo was at the maximum 1.5 times that in wild-type strains, implying the presence in E. coli cells of a mechanism that avoids cardiolipin overproduction, which is possibly disadvantageous to proper membrane functions.