Specific chromosomal sites enhancing homologous recombination in Escherichia coli mutants defective in RNase H

To clone new replication origin(s) activated under RNase H-defective (rnh ^?) conditions in Escherichia coli cells, whole chromosomal DNA digested with EcoRI was to with a Km^r DNA fragment and transformed into an rnh^? derivative host. From the Kmr transformants, we obtained eight kinds of plasmid-like DNA, each of which contained a specific DNA fragment, termed “Hot”, derived from the E. coli genome. Seven of the Hot DNAs (HotA-G) mapped to various sites within a narrow DNA replication termination region (about 280 kb), without any particular selection. Because Hot DNA could not be transformed into a mutant strain in which the corresponding Hot region had been deleted from the chromosome, the Hot DNA, though obtained as covalently closed circular (ccc) DNA, must have arisen by excision from the host chromosome into which it had initially integrated, rather than by autonomous replication of the transformed species. While Hot DNA does not have a weak replication origin it does have a strong recombinational hotspot active in the absence of RNase H. This notion is supported by the finding that Chi activity was present on all Hot DNAs tested and no Hot-positive clone without Chi activity was obtained, with the exception of a DNA clone carrying the dif site.     

RNase H-defective mutants of Escherichia coli: a possible discriminatory role of RNase H in initiation of DNA replication

Summary Mutants of Escherichia coli completely deficient in RNase H activity were isolated by inserting transposon Tn3 into the structural gene for RNase H, rnh, and its promoter. These rnh ^- mutants exhibited the following phenotypes; (1) the mutants grew fairly normally, (2) rnh ^- cells could be transformed with ColE1 derivative plasmids, pBR322 and pML21, though the plasmids were relatively unstable, under non selective conditions, (3) rnh ^- mutations partially suppressed the temperature-sensitive phenotype of plasmid pSC301, a DNA replication initiation mutant derived from pSC101, (4) rnh ^- mutations suppressed the temperature-sensitive growth character of dnaA ^ts mutant, (5) rnh ^- cells showed continued DNA synthesis in the presence of chloramphenicol (stable DNA replication). Based on these findings we propose a model for a role of RNase H in the initiation of chromosomal DNA replication. We suggest that two types of RNA primers for initiation of DNA replication are synthesized in a dnaA/oriC-dependent and-independent manner and that only the dnaA/oriC-dependent primer is involved in the normal DNA replication since the dnaA/oriC independent primer is selectively degraded by RNase H.Abbreviations AP^r ampicillin-resistant - kb kilobase pair(s) - NEM N-ethyl maleimide - Ts temperature-sensitive     

A new type of E. coli recombinational hotspot which requires for activity both DNA replication termination events and the Chi sequence

In E. coli rnh⁻ mutants we identified chromosome-derived, specific DNA fragments termed Hot DNA. When the DNA in the ccc form is integrated into the E. coli genome by homologous recombination to form a directly repeated structure, a striking enhancement of excisional recombination between the repeats occurs. We obtained 8 groups of such Hot DNA, 7 of which were clustered in a narrow region called the replication terminus region (about 280 kb) on the circular E. coli genome. A Ter site can impede the replication fork in a polar fashion. The six Ter sites are approximately symmetrical in the terminus and surrounding region. To block the fork at the Ter site, a protein factor, Ter binding protein encoded in the tau (or tus) gene, is required. In tau⁻ cells, Hot activity of HotA, B, and C DNAs disappears, thereby indicating that the Hot activity is fork arrest-dependent. Other Hot activities were tau-independent. In addition, for at least HotA activity, the presence of Chi, an E. coli recombinational hotspot sequence, is required; the Chi dependent HotA activity was detected in a wild type strain but to a lesser extent than that in the rnh⁻ mutant. To explain the HotA phenomenon at the molecular level, we propose a model in which a ds-break occurs at the replication fork arrested at the Ter site. Our recent data that HOT1, a yeast recombinational hotspot, may also depend on the fork blocking event for activity, suggests that a similar ds-break occurs in both eucaryotes and procaryotes.     

Function of ribonuclease H in initiation of DNA replication in Escherichia coli K-12

Summary Escherichia coli rnh mutants lacking ribonuclease H (RNase H) activity can tolerate deletion of the origin of DNA Replication (oriC) and transposon-insertional inactivation of an initiator gene (dnaA:Tn10). Introduction of the recA200 allele encoding a thermolabile RecA protein intornh ^– dnaA: Tn10 and rnh ^– oriC mutants strains rendered DNA synthesis and colony formation of these mutants temperature sensitive. The temperature sensitivity and the broth sensitivity (Srm^–) of the rnh ^– dnaA: Tn10 recA200 strain was suppressed by the presenceof plasmids (pBR322 derivatives) carrying dnaA ⁺only when the intact oriC site was present on the chromosome. Lack of RNase H activity neither promoted replication of minichromosomes (pOC24 and pasn20) in the absence of required DnaA⁺ protein nor inhibited dnaA ⁺–dependent minichromosome replication. These results led to the conclusion that RNase H is not directly involved in the events leading to initiation of DNA replication at oriC. Rather, it functions as a specificity factor by eliminating certain forms of RNA-DNA hybrids which could otherwise be used to prime DNA replication at sites other than oriC.     

The DNA replication fork blocked at the Ter site may be an entrance for the RecBCD enzyme into duplex DNA.

In Escherichia coli, eight kinds of chromosome-derived DNA fragments (named Hot DNA) were found to exhibit homologous recombinational hotspot activity, with the following properties. (i) The Hot activities of all Hot DNAs were enhanced extensively under RNase H-defective (rnh) conditions. (ii) Seven Hot DNAs were clustered at the DNA replication terminus region on the E. coli chromosome and had Chi activities (H. Nishitani, M. Hidaka, and T. Horiuchi, Mol. Gen. Genet. 240:307-314, 1993). Hot activities of HotA, -B, and -C, the locations of which were close to three DNA replication terminus sites, the TerB, -A, and -C sites, respectively, disappeared when terminus-binding (Tau or Tus) protein was defective, thereby suggesting that their Hot activities are termination event dependent. Other Hot groups showed termination-independent Hot activities. In addition, at least HotA activity proved to be dependent on a Chi sequence, because mutational destruction of the Chi sequence on the HotA DNA fragment resulted in disappearance of the HotA activity. The HotA activity which had disappeared was reactivated by insertion of a new, properly oriented Chi sequence at the position between the HotA DNA and the TerB site. On the basis of these observations and positional and orientational relationships between the Chi and the Ter sequences, we propose a model in which the DNA replication fork blocked at the Ter site provides an entrance for the RecBCD enzyme into duplex DNA.     

Correlation of activity with phenotypes of Escherichia coli partial function mutants of rnh, the gene encoding RNase H

Summary The rnh gene of Escherichia coli encodes RNase H. rnh mutants display at least two phenotypes: (1) they require functional RecBCD enzyme for growth; thus rnh-339::cat recB270 (Ts) and rnh-339::cat recC271 (Ts) strains are temperature sensitive for growth; (2) rnh mutants permit replication that is independent of the chromosomal origin, presumably by failing to remove RNA-DNA hybrids from which extra-original replication can be primed. We report here that manifestation of these two phenotypes occurs at different levels of RNase H function; we have examined partially functional rnh mutants for their in vitro RNase H activity, their ability to rescue viability in recB or recC cells and their ability to permit growth of mutants incapable of using oriC [dnaA (Ts)].     

Function of RNase H in DNA replication revealed by RNase H defective mutants of Escherichia coli

Summary Escherichia coli rnh mutants were isolated using localized mutagenesis and selective measurements of RNase H activity in mutagenized cell extracts with [³H]poly(rC)·poly(dG) as substrate. RNase H activity in extracts of one mutant, ON152 (rnh-91), was undetectable (less than 0.05% of that of wild-type cells). This mutant formed small colonies at 43 °C. At this temperature, accumulation of nascent fragments was more prominent in the rnh-91·polA4113 double mutant than in the polA4113 mutant; however, no accumulation was found in the rnh single mutant at 43° C. Unlike the 1–3 nucleotide primer RNA found on nascent fragments of polA4113 cells, primers from the rnh-91·polA4113 cells ranged from one to about ten bases. These results suggest that the 53 exonuclease activity of DNA polymerase I plays a major role in removal of primer RNA and that RNase H functions in an auxiliary role, excising the 5-portion of longer primers.The rnh mutant supports replication of ColE1-type plasmids. A possible mechanism of replication of such plasmids in rnh mutants and a role of RNase H in the initiation of chromosomal replication are discussed.     

Genetic analysis of constitutive stable DNA replication in rnh mutants of Escherichia coli K12

Summary Escherichia coli rnh mutants deficient in ribonuclease H (RNase H) are capable of DNA replication in the absence of protein synthesis. This constitutive stable DNA replication (SDR) is dependent upon the recA ⁺ gene product. The requirement of SDR for recA ⁺ can be suppressed by rin mutations (for recA⁺-independent), or by lexA(Def) mutations which inactivate the LexA repressor. Thus, there are at least three genetically distinct types of SDR in rnh mutants: recA ⁺-dependent SDR seen in rnh ^- rin⁺ lexA⁺ strains, recA ⁺-independent in rnh ^- rin^- lexA⁺, and recA ⁺-independent in rnh ^- rin⁺ lexA(Def). The expression of SDR in rin ^- and lexA(Def) mutants demonstrated a requirement for RNA synthesis and for the absence of RNase H. The suppression of the recA ⁺ requirement by rin mutations was shown to depend on some new function of the recF ⁺ gene product. In contrast, the suppression by lexA-(Def) mutations was not dependent on recF ⁺. The lexA3 mutation inhibited recA ⁺-dependent SDR via reducing the amount of recA ⁺ activity available, and was suppressed by the recAo254 mutation. The SDR in rnh ^- rin^- cells was also inhibited by the lexA3 mutation, but the inhibition was not reversed by the recAo254 mutation, indicating a requirement for some other lexA ⁺-regulated gene product in the recA ⁺-independent SDR process. A model is presented for the regulation of the expression of these three types of SDR by the products of the lexA ⁺, rin⁺ and recF ⁺ genes.     

Decolorization kinetics of a recombinant Escherichia coli strain harboring azo-dye-decolorizing determinants from Rhodococcus sp.

A 6.3 kb DNA fragment containing genes responsible for azo-dye decolorization was cloned and expressed in Escherichia coli. The resulting recombinant strain E. coli CY1 decolorized 200 mg azo dye (C.I. Reactive Red 22) l^–1 at 28 °C at 8.2 mg g cell^–1 h^–1, while the host (E. coli DH5) had no color-removal activity. Addition of 0.5 mM isopropyl--d-thiogalacto-pyranoside (IPTG) increased the decolorization rate 3.4-fold. The dependence of the decolorization rate on initial dye concentration essentially followed Monod-type kinetics and the maximal rate occurred with the dye at 600 mg l^–1. The decolorization rate of E. coli CY1 was optimal at 40 °C and pH 11. Aeration (increased dissolved O₂ level) strongly inhibited the decolorization, but decolorization occurred effectively under static incubation conditions (no agitation was employed). The CY1 strain also exhibited excellent stability during repeated-batch operations.     

Instability of inhibited replication forks in E. coli

Inhibiting the progress of replication forks in E. coli makes them susceptible to breakage. Broken replication forks are evidently reassembled by the RecBCD recombinational repair pathway. These findings explain a particular pattern of DNA degradation during inhibition of chromosomal replication, the role of recombination in the viability of mutants with displaced replication origin, and hyper-recombination observed in the Terminus of the E. coli chromosome in rnh mutants. Breakage and repair of inhibited replication forks could be the reason for the recombination-dependence of inducible stable DNA replication. A mechanism by which RecABCD-dependent recombination between very short inverted repeats may help E. coli to invert an operon, transcribed in the direction opposite to that of DNA replication, is discussed.     

Properties of two conjugative plasmids mediating tetracycline resistance, raffinose catabolism and hydrogen sulfide production in Escherichia coli.

Summary The tetracycline resistance (Tc) and raffi nose-hydrogen sulfide (Raf-Hys) characters of Escherichia coli D1021 are located on two compatible, conjugative fi^– plasmids named pRSDI and pRSD2, respectively. These were transferred together or separately to the isogenic background of E. coli K12 and isolated as transconjugants. Plasmid molecules isolated as covalently closed circular DNA have been analysed by buoyant density centrifugation, sucrose gradient sedimentation and electron microscopy. pRSDI (Tc) showed a buoyant density of 1.716 g/cm³ (56% GC), the open circular (OC) form had a sedimentation coefficient of 37s. If grown without tetracycline prior to isolation the contour length of most pRSDI molecules was 14.6 (±0.1) m (30.3 × 106 daltons) with a minor species of 13.9 (± 0.1) m owing to a small deletion. Pronounced length variation of pRSDI by growth in the presence of tetracycline owing to gene amplification is subject of a subsequent paper. In contrast, pRSD2 (Raf-Hys) molecules are stable under all conditions tested. The buoyant density was 1.713 g/cm³ (53% GC), the sedimentation coefficient of OGDNA was 58s and the contour length was 40.2 ± 0.6 m (83 × 106 daltons). During exponential growth one copy of pRSD2 per chromosome was found indicating stringent control of plasmid replication. At the onset of stationary growth the copy number rose from one to four per host chromosome indicating relaxation of the replication control. The level of plasmid-coded -galactosidase increases with copy numbers and has been used to test the gene dosage.     

Electroporation and stable maintenance of plasmid DNAs in a biocontrol strain of Pseudomonas syringae

Transformation efficiencies as high as 10⁷ transformants g^–1 DNA have been previously reported for pseudomonads using electroporation protocols established for E. coli with plasmid DNAs prepared from methylation proficient E. coli hosts. We report here a protocol for electroporation of plasmid DNAs into a biocontrol strain of Pseudomonas syringae which could not be electroporated by standard E. coli methods. Transformation efficiencies of 10⁷ or higher were obtained with DNA recovered from initial P. syringae transformation or with DNA prepared from methylation deficient E. coli. Both plasmids used in this study were stably maintained in the absence of selection for at least 50 generations.     

Isolation and nucleotide sequence of an autonomously replicating sequence (ARS) element functional in Candida albicans and Saccharomyces cerevisiae

Summary An 8.6-kb fragment was isolated from an EcoRI digest of Candida albicans ATCC 10261 genomic DNA which conferred the property of autonomous replication in Saccharomyces cervisiae on the otherwise non-replicative plasmid pMK155 (5.6 kb). The DNA responsible for the replicative function was subcloned as a 1.2-kb fragment onto a non-replicative plasmid (pRC3915) containing the C. albicans URA3 and LEU2 genes to form plasmid pRC3920. This plasmid was capable of autonomous replication in both S. cerevisiae and C. albicans and transformed S. cerevisiae AH22 (leu2 ^–) to Leu⁺ at a frequency of 2.15 × 10³ transformants per pg DNA, and transformed C. albicans SGY-243 (ura3) to Ura⁺ at a frequency of 1.91 × 10³ transformants per g DNA. Sequence analysis of the cloned DNA revealed the presence of two identical regions of eleven base pairs (5TTTTATGTTTT3) which agreed with the consensus of autonomously replicating sequence (ARS) cores functional in S. cerevisiae. In addition there were two 10/11 and numerous 9/11 matches to the core consensus. The two 11/11 matches to the consensus, CaARS1 and CaARS2, were located on opposite strands in a non-coding AT-rich region and were separated by 107 bp. Also present on the C. albicans DNA, 538 by from the ARS cores, was a gene for 5S rRNA which showed sequence homology with several other yeast 5S rRNA genes. A sub-fragment (494 bp) containing the 5S rRNA gene (but not the region containing the ARS cores) hybridized to genomic DNAs from a number of yeast species, including S. cerevisiae, C. tropicalis, C. pseudotropicalis, C. parapsilosis, C. kruseii, C. (Torulopsis) glabrata and Neurospora crassa. The 709-bp ARS element (but not the 5S rRNA gene) was necessary for high-frequency transformation and autonomous plasmid replication in both S. cerevisiae and C. albicans.EMBL/GenBank database accession number: X16634 (5S rRNA)     

Human cellular sequences detectable with adenovirus probes

Previous studies suggesting homology between human cellular DNA and the DNAs from adenovirus types 2 and 5 are extended in the present paper. A clone (ChAd^h), isolated from a human genomic DNA library using an adenovirus probe, hybridized to discrete regions of adenovirus 2 DNA, including part of the transforming genes E1a and E1b, as well as to repeated sequences within human DNA. The E1a and E1b genes both hybridize to the same 300 base pair Sau3AI fragment within ChAd^h although there is no obvious homology between E1a and E1b. The Ad 2 E1a gene was also used as a probe to screen other cellular DNAs to determine whether repeated sequences detectable with Ad 2 DNA probes were conserved over long evolutionary periods. Hybridization was detected to the genomes of man, rat, mouse and fruit fly, but not to those of yeast and bacteria. In addition to a smear hybridization, discrete fragments were detected in both rodent and fruit fly DNAs. The experiments reported suggest the existence of two different types of cellular sequences detected by Ad 2 DNA: (1) repeated sequences conserved in a variety of eukaryote genomes and (2) a possible unique sequence detected with an E1a probe different from that responsible for hybridization to repeated sequences. This unique sequence was detected as an EcoRI fragment in mouse DNA and had a molecular size of about 8.8 kb.     

Cloning of a gene involved in cellulose biosynthesis in Acetobacter xylinum: Complementation of cellulose-negative mutants by the UDPG pyrophosphorylase structural gene

Summary Three cellulose-negative (Cel^-) mutants of Acetobacter xylinum strain ATCC 23768 were complemented by a cloned 2.8 kb DNA fragment from the wild type. Biochemical analysis of the mutants showed that they were deficient in the enzyme uridine 5-diphosphoglucose (UDPG) pyrophosphorylase. The analysis also showed that the mutants could synthesize (1-4)-glucan in vitro from UDPG, but not in vivo from glucose. This result was expected, since UDPG is known to be the precursor for cellulose synthesis in A. xylinum. In order to analyze the function of the cloned gene in more detail, its biological activity in Escherichia coli was studied. These experiments showed that the cloned fragment could be used to complement an E. coli mutant deficient in the structural gene for UDPG pyrophosphorylase. It is therefore clear that the cloned fragment must contain this gene from A. xylinum. This is to our knowledge the first example of the cloning of a gene with a known function in cellulose biosynthesis from any organism, and we suggest the gene be designated celA.     

Cloning and expression of Candida albicans ADE2 and proteinase genes on a replicative plasmid in C. albicans and in Saccharomyces cerevisiae.

Summary A plasmid vector (denoted pRC2312) was constructed, which replicates autonomously in Escherichia coli, Saccharomyces cerevisiae and Candida albicans. It contains LEU2, URA3 and an autonomously replicating sequence (ARS) from C. albicans for selection and replication in yeasts, and bla (ampicillin resistance) and ori for selection and replication in E. coli. S. cerevisiae AH22 (Leu^–) was transformed by pRC2312 to Leu^– at a frequency of 1.41 × 10⁵ colonies per g DNA. Transformation of C. albicans SGY-243 (Ura-) to Ura⁺ with pRC2312 resulted in smaller transformant colonies at a frequency of 5.42 × 10³ per g DNA where the plasmid replicated autonomously in transformed cells, and larger transformant colonies at a frequency of 32 per g DNA, in which plasmid integrated into the genome. Plasmid copy number in yeasts was determined by a DNA hybridization method and was estimated to be 15±3 per haploid genome in S. cerevisiae and 2–3 per genome in C. albicans replicative transformants. Multiple tandem integration occurred in integrative transformants and copy number of the integrated sequence was estimated to be 7–12 per diploid genome. The C. albicans ADE2 gene was ligated into plasmid pRC2312 and the construct transformed Ade^– strains of both C. albicans and S. cerevisiae to Ade⁺. The vector pRC2312 was also used to clone a fragment of C. albicans genomic DNA containing an aspartic proteinase gene. C. albicans transformants harboring this plasmid showed a two-fold increase in aspartic proteinase activity. However S. cerevisiae transformants showed no such increase in proteinase activity, suggesting the gene was not expressed in S. cerevisiae.     

Molecular cloning and expression of a proteinase gene from Lactococcus lactis subsp. cremoris H2 and construction of a new lactococcal vector pFX1

The 6.5 kb HindIII DNA fragment of the Lactococcus lactis subsp. cremoris H2 plasmid pDI21 was cloned into Escherichia coli POP13 with NM1149, and also directly into Lactococcus lactis subsp. lactis 4125 using a newly-constructed broad host-range vector pFX1. Proteinase was experessed in both transformed organisms. The proteinase resembles a PI type since it preferentially degraded -casein. The restriction map of the 6.5 kb proteinase gene fragment has minor differences from those of published plamid proteinase genes. High-efficiency electroporation with pFX1 provides a direct approach for gene cloning in lactococci.Abbreviations cfu colony forming units - HEPES N-[2-hydroxyethyl]piperazine-N-[2-ethanesulphonic acid] Dedicated to Prof. Dr. G. Drews on the occasion of his 65th birthday     

Molecular cloning of Bacillus subtilis genes involved in DNA metabolism

Summary Different clones carrying a chromosomal DNA fragment able to transform Bacillus subtilis mutants dnaA13, dnaB19, dnaG5, recG40 and polA42 to a wild-type phenotype were isolated from a library constructed in plasmid pJH101. A recombinant clone carrying a chromosomal fragment able to transform dnaC mutants was obtained from a Charon 4A library. A restriction map of the cloned DNA fragments was constructed. The 11.3 kb cloned DNA fragment of plasmid pMP60-13 containing the wild-type sequence of dnaG5 was shown to transform a recF33 mutant as well.     

Interdependence and nodule specificity of cis-acting regulatory elements in the soybean leghemoglobin lbc3 and N23 gene promoters

Summary A composite plasmid comprising the mini-F and pBR322 replicons was found to inhibit cell growth of a host with conditional mutations in dnaA and rnh under restrictive conditions, where the normal initiation of replication from oriC was inactivated, but the alternative replication initiation from oriK was active. It was further shown that the composite plasrnid inhibited stable DNA replication (SDR) which occurs constitutively in cells mutant for rnh. Neither pBR322 nor mini-F alone produced these inhibitory effects. Deletion analyses revealed that the mini-F segment responsible for the inhibition of both processes was the promoter region of the sopA gene which had been cloned into a site upstream of the bla gene on pBR322 in such an orientation as to cause overexpression of bla. Inserting the promoter of the Escherichia coli lac gene into the same site had the same effect. Introduction of a deletion and a frameshift mutation into bla abolished the inhibition. Thus, the inhibition of growth and SDR appear to be due to overproduction of the bla gene product, -lactamase.     

Initiation of DNA replication in Escherichia coli after overproduction of the DnaA protein

Summary Flow cytometry was used to study initiation of DNA replication in Escherichia coli K12 after induced expression of a plasmid-borne dnaA ⁺ gene. When the dnaA gene was induced from either the plac or the pL promoter initiation was stimulated, as evidenced by an increase in the number of origins and in DNA content per mass unit. During prolonged growth under inducing conditions the origin and DNA content per mass unit were stabilized at levels significantly higher than those found before induction or in similarly treated control cells. The largest increase was observed when using the stronger promoter pL compared to plac. Synchrony of initiation was reasonably well maintained with elevated DnaA protein concentrations, indicating that simultaneous initiation of all origins was still preferred under these conditions. A reduced rate of replication fork movement was found in the presence of rifampin when the DnaA protein was overproduced. We conclude that increased synthesis levels or increased concentrations of the DnaA protein stimulate initiation of DNA replication. The data suggest that the DnaA protein may be the limiting factor for initiation under normal physiological conditions.