Complete Genome Sequence of Halomonas venusta Type Strain DSM 4743, a Moderately Halophilic Marine Bacterium.

Here, we report the genome sequence of Halomonas venusta strain DSM 4743T, a moderately halophilic marine bacterium. This type species genome consists of a 4.3-Mb chromosome, with 3,777 protein-coding genes, 60 tRNA loci, and 6 complete rRNA operons, plus a 6.1-kb plasmid termed p4743-A.

Insights into the history of a bacterial group II intron remnant from the genomes of the nitrogen-fixing symbionts Sinorhizobium meliloti and Sinorhizobium medicae.

Group II introns are self-splicing catalytic RNAs that act as mobile retroelements. In bacteria, they are thought to be tolerated to some extent because they self-splice and home preferentially to sites outside of functional genes, generally within intergenic regions or in other mobile genetic elements, by mechanisms including the divergence of DNA target specificity to prevent target site saturation. RmInt1 is a mobile group II intron that is widespread in natural populations of Sinorhizobium meliloti and was first described in the GR4 strain. Like other bacterial group II introns, RmInt1 tends to evolve toward an inactive form by fragmentation, with loss of the 3' terminus. We identified genomic evidence of a fragmented intron closely related to RmInt1 buried in the genome of the extant S. meliloti/S. medicae species. By studying this intron, we obtained evidence for the occurrence of intron insertion before the divergence of ancient rhizobial species. This fragmented group II intron has thus existed for a long time and has provided sequence variation, on which selection can act, contributing to diverse genetic rearrangements, and to generate pan-genome divergence after strain differentiation. The data presented here suggest that fragmented group II introns within intergenic regions closed to functionally important neighboring genes may have been microevolutionary forces driving adaptive evolution of these rhizobial species.

Identification and characterization of a nodH ortholog from the alfalfa-nodulating Or191-like rhizobia.

Nodulation of Medicago sativa (alfalfa) is known to be restricted to Sinorhizobium meliloti and a few other rhizobia that include the poorly characterized isolates related to Rhizobium sp. strain Or191. Distinctive features of the symbiosis between alfalfa and S. meliloti are the marked specificity from the plant to the bacteria and the strict requirement for the presence of sulfated lipochitooligosaccharides (Nod factors [NFs]) at its reducing end. Here, we present evidence of the presence of a functional nodH-encoded NF sulfotransferase in the Or191-like rhizobia. The nodH gene, present in single copy, maps to a high molecular weight megaplasmid. As in S. meliloti, a nodF homolog was identified immediately upstream of nodH that was transcribed in the opposite direction (local synteny). This novel nodH ortholog was cloned and shown to restore both NF sulfation and the Nif+Fix+ phenotypes when introduced into an S. meliloti nodH mutant. Unexpectedly, however, nodH disruption in the Or191-like bacteria did not abolish their ability to nodulate alfalfa, resulting instead in a severely delayed nodulation. In agreement with evidence from other authors, the nodH sequence analysis strongly supports the idea that the Or191-like rhizobia most likely represent a genetic mosaic resulting from the horizontal transfer of symbiotic genes from a sinorhizobial megaplasmid to a not yet clearly identified ancestor.

Diversity of group II introns in the genome of Sinorhizobium meliloti strain 1021: splicing and mobility of RmInt1.

The number and diversity of known group II introns in eubacteria are continually increasing with the addition of new data from sequencing projects, but the significance of these introns in the evolution of bacterial genomes is unknown. We analyzed the main features of the group II introns present in the genome of the soil microorganism Sinorhizobium meliloti (strain 1021), the nitrogen-fixing symbiont of alfalfa, the DNA sequence of which was recently determined. Strain 1021 harbors three different classes of group II introns: RmInt1, of bacterial class D; SMb2147/SMb21167, which cluster within bacterial class C; and SMa1875, the phylogenetic class of which is uncertain. The group II introns SMb2147/SMb21167 and SMa1875 are widely distributed in S. meliloti, but are present in lower copy numbers than RmInt1. Strain 1021 harbors three copies of RmInt1, which is pSym-specific. Although RmInt1 is spliced in strain 1021, mobility assays suggested that, in contrast to other S. meliloti strains, the genetic background of strain 1021 does not support intron homing events.

Group II introns in the bacterial world.

Group II introns are large catalytic RNA molecules that act as mobile genetic elements. They were initially identified in the organelle genomes of lower eukaryotes and plants, and it has been suggested that they are the progenitors of nuclear spliceosomal introns. Group II self-splicing introns were shown to be present in bacteria in 1993, since when the various bacterial genome sequencing projects have led to a significant increase in the number of group II intron sequences present in databases. However, few of these introns have been characterized, and most were identified on the basis of their intron-encoded protein (IEP), with little data available concerning their ribozyme/RNA structure. Their frequency in prokaryotes is also unknown. We attempt here to provide a first comprehensive review of bacterial group II introns based on recent genome sequencing data and mechanistic studies.

RecA-independent ectopic transposition in vivo of a bacterial group II intron.

RmInt1 is a group II intron of Sinorhizobium meliloti which was initially found within the insertion sequence ISRm2011-2. Although the RmInt1 intron-encoded protein lacks a recognizable endonuclease domain, it is able to mediate insertion of RmInt1 at an intron-specific location in intronless ISRm2011-2 recipient DNA, a phenomenon termed homing. Here we have characterized three additional insertion sites of RmInt1 in the genome of S.meliloti. Two of these sites are within IS elements closely related to ISRm2011-2, which appear to form a characteristic group within the IS630-Tc1 family. The third site is in the oxi1 gene, which encodes a putative oxide reductase. The newly identified integration sites contain conserved intron-binding site (IBS1 and IBS2) and delta' sequences (14 bp). The RNA of the intron-containing oxi1 gene is able to splice and the oxi1 site is a DNA target for RmInt1 transposition in vivo. Ectopic transposition of RmInt1 into the oxi1 gene occurs at 20-fold lower efficiency than into the homing site (ISRm2011-2) and is independent of the major RecA recombination pathway. The possibility that transposition of RmInt1 to the oxi1 site occurs by reverse splicing into DNA is discussed.

Homing of a bacterial group II intron with an intron-encoded protein lacking a recognizable endonuclease domain.

RmInt1 is a functional group II intron found in Sinorhizobium meliloti where it interrupts a group of IS elements of the IS630-Tc1 family. In contrast to many other group II introns, the intron-encoded protein (IEP) of RmInt1 lacks the characteristic conserved part of the Zn domain associated with the IEP endonuclease activity. Nevertheless, in this study, we show that RmInt1 is capable of inserting into a vector containing the DNA spanning the RmInt1 target site from the genome of S. meliloti. Efficient homing was also observed in the absence of homologous recombination (RecA- strains). In addition, it is shown that RmInt1 is able to move to its target in a heterologous host (S. medicae). Homing of RmInt1 occurs very efficiently upon DNA target uptake (conjugation/electroporation) by the host cell resulting in a proportion of invaded target of 11-30%. Afterwards, the remaining intronless target DNA is protected from intron invasion.

Characterization and splicing in vivo of a Sinorhizobium meliloti group II intron associated with particular insertion sequences of the IS630-Tc1/IS3 retroposon superfamily.

By sequence analysis of Sinorhizobium meliloti strain GR4 plasmid pRmeGR4b, we have identified a group II intron named RmInt1 inserted within the insertion sequence ISRm2011-2 of the IS630-Tc1/IS3 retroposon superfamily. Like some other group II introns, RmInt1 possesses, in addition to the structurally conserved ribozyme core, an open reading frame (ORF) with homology to reverse transcriptases. Using a T7 expression system in Escherichia coli, we show that the intron is active in splicing in vivo and that splicing efficiency requires the intron-encoded ORF, which suggests that the putative intron encoded protein has a maturase function. DNA hybridization studies indicate that intron RmInt1 is widespread within S. meliloti native populations and appears to be mostly located within this IS element. Nevertheless, some S. meliloti strains harbour one copy of RmInt1 at a different location. DNA sequence analysis of the 5' exon of one of these heterologous intron insertion sites revealed the presence of a putative IS element closely related to insertion sequence ISRm2011-2. The intron-binding sites (IBS1 and IBS2 motifs) are conserved, although a transition of a G-->A in the IBS1 has occurred. Our results demonstrate an association of intron RmInt1 with particular insertion sequences of the IS630-Tc1/IS3 retroposon superfamily that may have ensured the spread and maintenance of this group II intron in S. meliloti.

Multicopy vectors carrying the Klebsiella pneumoniae nifA gene do not enhance the nodulation competitiveness of Sinorhizobium meliloti on alfalfa.

It has been reported that Sinorhizobium meliloti strains harboring IncQ and IncP multicopy vectors containing constitutively expressed Klebsiella pneumoniae nifA exhibit an increase in nodulation competitiveness on alfalfa (J. Sanjuan and J. Olivares, Mol. Plant-Microbe Interact. 4:365-369, 1991). In our efforts to understand the mechanisms involved, in this work, we have found that the observed enhancement on nodulation competitiveness by IncQ derivatives carrying K. pneumoniae nifA was not dependent on the plasmid-borne nifA activity but on the sensitivity of nonresistant strains to the streptomycin carried over from growth cultures. Furthermore, it was also determined that the nifA of K. pneumoniae on an IncP vector does not have an effect on competitiveness.

Cleavage of p220 by purified poliovirus 2A(pro) in cell-free systems: effects on translation of capped and uncapped mRNAs.

Poliovirus protease 2A(pro) has been obtained in soluble form as a fusion protein with maltose binding protein (MBP). Addition of MBP-2A(pro) to rabbit reticulocyte cell-free systems gives rise to efficient cleavage of the initiation factor of translation p220 (eIF-4G). Translation of capped mRNA encoding the influenza virus NP protein is severely impaired in lysates in which p220 has been proteolytically cleaved. This inhibition is dependent on the concentration of mRNA added to the lysate. Thus, increasing the concentrations of mRNA substantially overcomes the blockade of NP synthesis after p220 cleavage. Notably, translation of uncapped NP mRNA is also compromised in p220-deficient rabbit reticulocyte lysates, suggesting that p220 participates in the translation of both capped and uncapped NP mRNAs. The effects of p220 proteolysis by poliovirus 2A(pro) have also been assayed on luciferase mRNA translation. Three types of mRNAs encoding for luciferase have been examined: capped, uncapped, and mRNA bearing the poliovirus 5' leader region (leader luc mRNA). Synthesis of luciferase directed by any of these mRNAs was inhibited after cleavage of p220 in rabbit reticulocyte lysates. Interestingly, supplementation of the lysate with HeLa cell extracts stimulates leader luc mRNA translation by poliovirus 2A(pro). These results indicate that activation of translation of mRNAs bearing the poliovirus leader region promoted by this poliovirus protease requires a factor present in HeLa cell extracts. These findings agree well with recent experiments implicating p220 not only in protein synthesis directed by capped mRNAs but also in the translation of naturally uncapped mRNAs.

Nod factor-induced expression of leghemoglobin to study the mechanism of NH4NO3 inhibition on root hair deformation.

Nod factors secreted by Rhizobium leguminosarum by, viciae induce root hair deformation, the formation of nodule primordia, and the expression of early nodulin genes in Vicia sativa (vetch). Root hair deformation is induced within 3 h in a small, susceptible zone (+/-2 mm) of the root. NH4NO3, known to be a potent blocker of nodule formation, inhibits root hair deformation, initial cortical cell divisions, and infection thread formation. To test whether NH4NO3 affects the formation of a component of the Nod factor perception-transduction system, we studied Nod factor-induced gene expression. The differential display technique was used to search for marker genes, which are induced within 1 to 3 h after Nod factor application. Surprisingly, one of the isolated cDNA clones was identified as a leghemoglobin gene (VsLb1), which is induced in vetch roots within 1 h after Nod factor application. By using the drug brefeldin A, it was then shown that VsLb1 activation does not require root hair deformation. The pVsLb1 clone was used as a marker to show that in vetch plants grown in the presence of NH4NO3, Nod factor perception and transduction leading to gene expression are unaffected.

Early nodulin gene expression during Nod factor-induced processes in Vicia sativa.

Rhizobium leguminosarum bv. viciae-secreted Nod factors are able to induce root hair deformation, the formation of nodule primordia and the expression of early nodulin genes in Vicia sativa (vetch). To obtain more insight into the mode of action of Nod factors the expression of early nodulin genes was followed during Nod factor-induced root hair deformation and nodule primordium formation. The results of these studies suggested that the expression of VsENOD5 and VsENOD12 is not required for root hair deformation. In the Nod factor-induced primordia both VsENOD12 and VsENOD40 are expressed in a spatially controlled manner similar to that found in Rhizobium-induced nodule primordia. In contrast, VsENOD5 expression has never been observed in Nod factor-induced primordia, showing that the induction of VsENOD5 and VsENOD12 expression are not coupled. VsENOD5 expression is induced in the root epidermis by Nod factors and in Rhizobium-induced nodule primordia only in cells infected by the bacteria, suggesting that the Nod factor does not reach the inner cortical cells.

High level expression in Escherichia coli cells and purification of poliovirus protein 2Apro.

The poliovirus protease 2Apro has been produced to high levels in Escherichia coli using the inducible system that utilizes T7 RNA polymerase. The protease coding sequences that contained an additional AUG to start translation were cloned in pET vectors. Synthesis of 2Apro was induced by IPTG or IPTG plus rifampicin, the levels of the protein made being higher when IPTG alone was used. The expression of the protein is not toxic for E. coli cells and can be readily visualized by Coomassie blue staining of total bacterial protein extracts separated in polyacrylamide gels. Centrifugation of the broken bacterial cells sediments more than 95% of the 2Apro synthesized at a 95% purity level after sarkosyl treatment. Antibodies raised against 2Apro in E. coli recognize a 16K protein in poliovirus-infected cells. In addition, 2Apro shows activity in trans as measured by the cleavage of p220 in HeLa cell extracts and by cleavage of a poliovirus protein substrate that contains the junction between the P1 and P2 polypeptides.

Cloning and inducible synthesis of poliovirus nonstructural proteins.

The poliovirus nonstructural protein-encoding genes have been cloned and expressed in Escherichia coli using the inducible system described by Studier and Moffat [J. Mol. Biol. 189 (1986) 113-130] and Studier [J. Mol. Biol. 219 (1991) 37-44]. The two genes encoding the poliovirus proteases, 2Apro and 3Cpro, were cloned together with their flanking regions in order to test the ability of the polyprotein precursors synthesized to cause proteolytic cleavage and generate mature forms. Both proteases were synthesized and showed activity upon induction in this system. Previously, it had not been possible to produce the three poliovirus nonstructural proteins, 2B, 2C and 3A, and some of their precursors, 2C3AB, 2C3A and 3AB, at high levels in E. coli cells. We report the cloning of their genes using PCR techniques and their efficient expression from pET vectors upon induction with IPTG (isopropyl-beta-D-thiogalactopyranoside). Moreover, some of these proteins, e.g., 3AB, 3A and 2B, are quite toxic for E. coli cells and lysed them upon production. Our results demonstrate the usefulness of this inducible system using the pET vectors to express these toxic poliovirus proteins.

Cell type determines the relative proportions of (-) and (+) strand RNA during poliovirus replication.

To examine the influence of the host cell type on poliovirus RNA synthesis we compared the levels of (-) and (+) strand poliovirus RNA during infection of epithelial (HeLa and HEp-2), leukocytic (U-937, HL-60 and K-562) and nerve (IMR-32) cells. The levels of (-) strand RNA were higher in IMR-32, U-937, K-562 or HL-60 cells than those in HeLa or HEp-2 cells. By contrast, (+) strand RNA content was greater in HeLa or HEp-2 cells. Although significant levels of (+) strand RNA were detected in U-937, K-562 and HL-60 cells, no viral protein synthesis was detected by polyacrylamide gel electrophoretic analysis of metabolically labelled proteins. The molar ratio of poliovirus (-) and (+) RNAs was 2-3 fold higher in IMR-32, U-937 and K-562 cells than in HeLa or HL-60 cells and 5-6 fold higher than in HEp-2 cells. Differentiation of HL-60 cells with a variety of inducers produced differential effects on poliovirus (-) and (+) RNA content and modified the molar ratio of (-)/(+) strand RNAs. These findings indicate that host cell components play a critical role in the regulation of the amount of poliovirus (-) and (+) strand RNAs synthesized during infection.

Restriction of poliovirus RNA translation in a human monocytic cell line.

The infection of the human monocytic cell line U-937 by poliovirus was characterized by a low level of virus production and a slow progression of the cytopathic effect. Infection took place in greater than 99% of the cells as revealed by a limiting dilution assay. No viral protein synthesis was evident in the infected U-937 cells when analyzed by polyacrylamide gel electrophoresis. However, a low level of poliovirus RNA translation was detected by immunofluorescence analysis using a mixture of polyclonal antisera against non-structural proteins. Although there was only a low level of viral protein synthesis, a gradual accumulation of viral mRNA took place in U-937 cells as revealed by RNA blot analysis. Similar results were obtained when the erythroleukemic cell line K-562 was used as a host cell for poliovirus. RNA extracted from infected U-937 cells was efficiently translated in rabbit reticulocyte extracts giving rise to a pattern of viral polypeptides similar to that detected when virion-purified RNA was the template used for the in vitro translation assay, suggesting that the poliovirus RNA present in infected U-937 cells is functional. The existence in U-937 cells of a discriminatory mechanism which differentially interferes with poliovirus RNA translation is discussed.