Molecular cloning of the rfb region of Klebsiella pneumoniae serotype O1:K20: the rfb gene cluster is responsible for synthesis of the D-galactan I O polysaccharide.

Previous chemical analyses identified two structurally distinct O polysaccharides in the lipopolysaccharide of Klebsiella pneumoniae serotype O1:K20 (C. Whitfield, J. C. Richards, M. B. Perry, B. R. Clarke, and L. L. MacLean, J. Bacteriol. 173:1420-1431, 1991). The polysaccharides were designated D-galactan I and D-galactan II; both are homopolymers of galactose. To begin investigation of the synthesis and expression of these O polysaccharides, we have cloned a 7.3-kb region of the chromosome of K. pneumoniae O1:K20, containing the his-linked rfbkpO1 (O-antigen biosynthesis) gene cluster. In Escherichia coli K-12 and Salmonella typhimurium, rfbkpO1 directed the synthesis of D-galactan I but not D-galactan II. The cloned rfbkpO1 genes did not complement a mutation affecting D-galactan II synthesis in K. pneumoniae CWK37, suggesting that another (unlinked) locus is also required for D-galactan II expression. However, plasmids carrying rfbkpO1 did complement a mutation in K. pneumoniae CWK43 which eliminated expression of both D-galactan I and D-galactan II, indicating that at least one function is common to synthesis of both polymers. Synthesis of D-galactan I was dependent on chromosomal galE and rfe genes. Hybridization experiments indicated that the rfbkpO1 sequences from different serotype O1 Klebsiella isolates showed some restriction fragment length polymorphism.     

Expression of two structurally distinct D-galactan O antigens in the lipopolysaccharide of Klebsiella pneumoniae serotype O1.

The lipopolysaccharide (LPS) molecule is an important virulence determinant in Klebsiella pneumoniae. Studies on the serotype O1 LPS were initiated to determine the basis for antigenic heterogeneity previously observed in the O1 side chain polysaccharides and to resolve apparent ambiguities in the reported polysaccharide structure. Detailed chemical analysis, involving methylation and 1H- and 13C-nuclear magnetic resonance studies, demonstrated that the O-side chain polysaccharides of serotype O1 LPS contained a mixture of two structurally distinct D-galactan polymers. The repeating unit structures of these two polymers were identified as [----3)-beta-D-Galf-(1----3)-alpha-D-Galp-(1----] (D-galactan I) and [----3)-alpha-D-Galp-(1----3)-beta-D-Galp-(1----] (D-Galactan II). D-Galactan I polysaccharides were heterogeneous in size and were detected throughout the sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) profile of O1 LPS. In contrast, D-galactan II was confined to the higher-molecular-weight region. The structures of the two D-galactans were not influenced by simultaneous synthesis of a capsular K antigen. Apparently, neither of the D-galactans constitutes a common antigen widespread in Klebsiella spp. as determined by immunochemical analysis. Examination of the LPSs in mutants indicated that expression of D-galactan I can occur independently of D-galactan II. Transconjugants of Escherichia coli K-12 strains carrying the his region of K. pneumoniae were constructed by chromosome mobilization with RP4::mini-Mu. In these transconjugants, the O antigen encoded by the his-linked rfb locus was determined to be D-galactan I, suggesting that genes involved in the expression of D-galactan II are not closely linked to the rfb cluster.     

Relationships between rfb gene clusters required for biosynthesis of identical D-galactose-containing O antigens in Klebsiella pneumoniae serotype O1 and Serratia marcescens serotype O16.

The lipopolysaccharide O antigens of Klebsiella pneumoniae serotype O1 and Serratia marcescens serotype O16 both contain a repeating unit disaccharide of [-->3)-beta-D-Galf-(1-->3)-alpha-D-Galp-(1-->]; the resulting polymer is known as D-galactan I. In K. pneumoniae serotype O1, the genes responsible for the synthesis of D-galactan I are found in the rfb gene cluster (rfbKpO1). We report here the cloning and analysis of the rfb cluster from S. marcescens serotype O16 (rfbSmO16). This is the first rfb gene cluster examined for the genus Serratia. Synthesis of D-galactan I is an rfe-dependent process for both K. pneumoniae serotype O1 and S. marcescens serotype O16. Hybridization experiments with probes derived from each of the six rfbKpO1 genes indicate that the cloned rfbSmO16 cluster contains homologous genes arranged in the same order. However, the degree of homology at the nucleotide sequence level was sufficiently low that hybridization was detected only under low-stringency conditions. rfbABSmO16 genes were subcloned and shown to encode an ABC-2 (ATP-binding cassette) transporter which is functionally identical to the one encoded by the corresponding rfb genes from K. pneumoniae serotype O1. The amino acid sequences of the predicted RfbA and RfbB homologs showed identities of 75.7% (87.9% total similarity) and 78.0% (86.5% total similarity), respectively. The last gene of the rfbKpO1 cluster, rfbFKpO1, encodes a bifunctional galactosyltransferase which initiates the formation of D-galactan I. RfbFKpO1 and RfbFSmO16 are 57.6% identical (with 71.1% total similarity), and both show similarity with RfpB, the galactosyltransferase involved in the synthesis of Shigella dysenteriae type I O-polysaccharide. The G+C contents of the rfbAB genes from each organism are quite similar, and values are lower than those typical for the species. However, the G+C content of rfbFSmO16 (47.6%) was much higher than that of rfbFKpO1 (37.3%), despite the fact that the average for each species (52 to 60%) falls within the same range.     

Role of Rfe and RfbF in the initiation of biosynthesis of D-galactan I, the lipopolysaccharide O antigen from Klebsiella pneumoniae serotype O1.

The 6.6-kb rfb gene cluster from Klebsiella pneumoniae serotype O1 (rfbKpO1) contains six genes whose products are required for the biosynthesis of a lipopolysaccharide O antigen with the following repeating unit structure: -->3-beta-D-Galf-1-->3-alpha-D-Galp-1-->(D-galactan I). rfbFKpO1 is the last gene in the cluster, and its gene product is required for the initiation of D-galactan I synthesis. Escherichia coli K-12 strains expressing the RfbFKpO1 polypeptide contain dual galactopyranosyl and galactofuranosyl transferase activity. This activity modifies the host lipopolysaccharide core by adding the disaccharide beta-D-Galf-1-->3-alpha-D-Galp, representing a single repeating unit of D-galactan I. The formation of the lipopolysaccharide substituted either with the disaccharide or with authentic polymeric D-galactan I is dependent on the activity of the Rfe enzyme. Rfe (UDP-GlcpNAc::undecaprenylphosphate GlcpNAc-1-phosphate transferase) catalyzes the formation of the lipid-linked biosynthetic intermediate to which galactosyl residues are transferred during the initial steps of D-galactan I synthesis. The rfbFKpO1 gene comprises 1,131 nucleotides, and the predicted polypeptide consists of 373 amino acid residues with a predicted M(r) of 42,600. A polypeptide with an M(r) of 42,000 was evident in sodium dodecyl sulfate-polyacrylamide gels when rfbKpO1 was expressed behind the T7 promoter. The carboxy-terminal region of RfbFKpO1 shares similarity with the carboxy terminus of RfpB, a galactopyranosyl transferase which is involved in the synthesis of the type 1 O antigen of Shigella dysenteriae.     

Functional analysis of the galactosyltransferases required for biosynthesis of D-galactan I, a component of the lipopolysaccharide O1 antigen of Klebsiella pneumoniae

D-Galactan I is an O-antigenic polymer with the repeat unit structure [-->3)-beta-D-Galf-(1-->3)-alpha-D-Galp-(1-->], that is found in the lipopolysaccharide of Klebsiella pneumoniae O1 and other gram-negative bacteria. A genetic locus containing six genes is responsible for the synthesis and assembly of D-galactan I via an ATP-binding cassette (ABC) transporter-dependent pathway. The galactosyltransferase activities that are required for the processive polymerization of D-galactan I were identified by using in vitro reactions. The activities were determined with endogenous lipid acceptors in membrane preparations from Escherichia coli K-12 expressing individual enzymes (or combinations of enzymes) or in membranes reconstituted with specific lipid acceptors. The D-galactan I polymer is built on a lipid acceptor, undecaprenyl pyrophosphoryl-GlcpNAc, a product of the WecA enzyme that participates in the biosynthesis of enterobacterial common antigen and O-antigenic polysaccharide (O-PS) biosynthesis pathways. This intermediate is directed into D-galactan I biosynthesis by the bifunctional wbbO gene product, which sequentially adds one Galp and one Galf residue from the corresponding UDP-sugars to form a lipid-linked trisaccharide. The two galactosyltransferase activities of WbbO are separable by limiting the UDP-Galf precursor. Galactosyltransferase activity in membranes reconstituted with exogenous lipid-linked trisaccharide acceptor and the known structure of D-galactan I indicate that WbbM catalyzes the subsequent transfer of a single Galp residue to form a lipid-linked tetrasaccharide. Chain extension of the D-galactan I polymer requires WbbM for Galp transferase, together with Galf transferase activity provided by WbbO. Comparison of the biosynthetic pathways for D-galactan I and the polymannose E. coli O9a antigen reveals some interesting features that may reflect a common theme in ABC transporter-dependent O-PS assembly systems.     

The 5'-proximal hairpin of turnip yellow mosaic virus RNA: its role in translation and encapsidation

The RNA genome of turnip yellow mosaic virus (TYMV) consists of more than 6,000 nucleotides. During a study of the roles of the two hairpins located in its 90-nucleotide 5' untranslated region, it was observed that stabilization of the 5'-proximal hairpin leads to a delay in the development of symptoms on plants. This delay in symptom development for both locally and systemically infected leaves was found to be dependent on a change in the free energy of the hairpin caused by introduced mutations. A protoplast transfection assay revealed that the accumulation of plus-strand full-length RNA and subgenomic RNA, as well as protein expression levels, was affected by hairpin stability. Stabilization of this hairpin inhibited translation. A model is proposed in which a destabilized 5'-proximal hairpin allows maximal translation of the viral proteins. It is suggested that this hairpin may exist in close proximity to the 5' cap as long as its stability is low enough to enable translation. However, at an acidic pH, the hairpin structure becomes more stable and is functionally transformed into the initiation signal for viral packaging. Slightly acidic conditions can be found in chloroplasts, where TYMV assembly is driven by a low pH generated by active photosynthesis.     

Structural analysis of the O-antigen side chain polysaccharides in the lipopolysaccharides of Klebsiella serotypes O2(2a), O2(2a,2b), and O2(2a,2c).

The lipopolysaccharide (LPS) of Klebsiella serotype O2 is antigenically heterogeneous; some strains express multiple antigenic factors. To study this heterogeneity, we determined the structure of the O-antigen polysaccharides in isolates belonging to serotypes O2(2a), O2(2a,2b), and O2(2a,2c), by using composition analysis, methylation analysis, and both 1H and 13C nuclear magnetic resonance spectroscopy. The repeating unit structure of the 2a polysaccharide was identified as the disaccharide [----3)-beta-D-Galf-(1----3)-alpha-D-Galp-(1----] and was identical to D-galactan I, one of two O polysaccharides present in the LPS of Klebsiella pneumoniae serotype O1 (C. Whitfield, J. C. Richards, M. B. Perry, B. R. Clarke, and L. L. MacLean, J. Bacteriol. 173:1420-1431, 1991). LPS from serotype O2(2a,2b) also contained D-galactan I as the only O polysaccharide, suggesting that the 2b antigen is not an O antigen. The LPS of serotype O2(2a,2c) contained a mixture of two structurally distinct O polysaccharides and provides a second example of this phenomenon in Klebsiella spp. One polymer was identical to D-galactan I, and the other polysaccharide, the 2c antigen, was a polymer with a disaccharide repeating unit structure, [----3)-beta-D-GlcpNAc-(1----5)-beta-D-Galf-(1----]. The 2c structure does not resemble previously reported O polysaccharides from Klebsiella spp. Periodate oxidation confirmed that D-galactan I and the 2c polysaccharide are distinct glycans, rather than representing domains within a single polysaccharide chain. Monoclonal antibodies against the 2c antigen indicated that only LPS molecules with the longest O-polysaccharide chains contained the 2c epitope.     

Clonally diverse rfb gene clusters are involved in expression of a family of related D-galactan O antigens in Klebsiella species.

Klebsiella species express a family of structurally related lipopolysaccharide O antigens which share a common backbone known as D-galactan I. Serotype specificity results from modification of D-galactan I by addition of domains of altered structure or by substitution with O-acetyl and/or alpha-D-Galp side groups with various linkages and stoichiometries. In the prototype, Klebsiella serotype O1, the his-linked rfb gene cluster is required for synthesis of D-galactan I, but genes conferring serotype specificity are unlinked. The D-galactan I part of the O polysaccharide is O acetylated in Klebsiella serotype O8. By cloning the rfb region from Klebsiella serotype O8 and analyzing the O polysaccharide synthesized in Escherichia coli K-12 hosts, we show that, like rfbO1, the rfbO8 region directs formation of unmodified D-galactan I. The rfbAB genes encode an ATP-binding cassette transporter required for export of polymeric D-galactan I across the plasma membrane prior to completion of the lipopolysaccharide molecule by ligation of the O polysaccharide to lipid A-core. Complementation experiments show that the rfbAB gene products in serotypes O1 and O8 are functionally equivalent and interchangeable. Hybridization experiments and physical mapping of the rfb regions in related Klebsiella serotypes suggest the existence of shared rfb genes with a common organization. However, despite the functional equivalence of these rfb gene clusters, at least three distinct clonal groups were detected in different Klebsiella species and subspecies, on the basis of Southern hybridization experiments carried out under high-stringency conditions. The clonal groups cannot be predicted by features of the O-antigen structure. To examine the relationships in more detail, the complete nucleotide sequence of the serotype O8 rfb cluster was determined and compared with that of the serotype O1 prototype. The nucleotide sequences for the six rfb genes showed variations in moles percent G+C values and in the values for nucleotide sequence identity, which ranged from 66.9 to 79.7%. The predicted polypeptides ranged from 64.3% identity (78.4% total similarity) to 94.3% identity (98.0% similarity). The results presented here are not consistent with dissemination of the Klebsiella D-galactan I rfb genes through recent lateral transfer events.     

Syntheses of alternating oligo-2'-O-methylribonucleoside methylphosphonates and their interactions with HIV TAR RNA

Oligonucleotide analogues 15-20 nucleotides in length have been prepared, whose sequences are complementary to nucleotides in the upper hairpin of HIV TAR RNA. These alternating oligonucleoside methylphosphonates, mr-AOMPs, contain 2'-O-methylribonucleosides and alternating methylphosphonate and phosphodiester internucleotide linkages. The methylphosphonate and phosphodiester linkages of these oligomers are highly resistant to hydrolysis by exonuclease activity found in mammalian serum and to endonucleases, such as S1 nuclease. The oligomers were prepared using automated phosphoramidite chemistry and terminate with a 5'-phosphate group, which provides an affinity handle for purification by strong anion exchange HPLC. A 15-mer mr-AOMP, 1676, that is complementary to the 5'-side of the TAR RNA hairpin, including the 3-base bulge and 6-base loop region, forms a 1:1 duplex with a complementary RNA 18-mer, mini-TAR RNA. The T(m) of this duplex is 71 degrees C, which is similar to that of the duplex formed by the corresponding all phosphodiester 15-mer. Introduction of two mismatched bases reduces the T(m) by 17 degrees C. The apparent dissociation constant, K(d), for the 1676/mini-TAR RNA duplex as determined by an electrophoretic mobility shift assay at 37 degrees C is 0.3 nM. Oligomer 1676 also binds tightly to the full length TAR RNA target under physiological conditions (K(d) = 20 nM), whereas no binding was observed by the mismatched oligomer. A 19-mer that is complementary to the entire upper hairpin also binds to TAR RNA with a K(d) that is similar to that of 1676, a result that suggests only part of the oligomer binds. When two of the methylphosphonate linkages in the region complementary to the 6-base loop are replaced with phosphodiester linkages, the K(d) is reduced by approximately a factor of 10. This result suggests that interactions between TAR RNA and the oligomer occur initially with nucleotides in the 6-base loop, and that these interactions are sensitive to presence and possibly the chirality of the methylphosphonate linkages in the oligomer. The high affinities of mr-AOMPs for TAR RNA and their resistance to nuclease hydrolysis suggests their potential utility as antisense agents in cell culture.     

Role of the 5'-proximal stem-loop structure of the 5' untranslated region in replication and translation of hepatitis C virus RNA

Sequences of the untranslated regions at the 5' and 3' ends (5'UTR and 3'UTR) of the hepatitis C virus (HCV) RNA genome are highly conserved and contain cis-acting RNA elements for HCV RNA replication. The HCV 5'UTR consists of two distinct RNA elements, a short 5'-proximal stem-loop RNA element (nucleotides 1 to 43) and a longer element of internal ribosome entry site. To determine the sequence and structural requirements of the 5'-proximal stem-loop RNA element in HCV RNA replication and translation, a mutagenesis analysis was preformed by nucleotide deletions and substitutions. Effects of mutations in the 5'-proximal stem-loop RNA element on HCV RNA replication were determined by using a cell-based HCV replicon replication system. Deletion of the first 20 nucleotides from the 5' end resulted in elimination of cell colony formation. Likewise, disruption of the 5'-proximal stem-loop by nucleotide substitutions abolished the ability of HCV RNA to induce cell colony formation. However, restoration of the 5'-proximal stem-loop by compensatory mutations with different nucleotides rescued the ability of the subgenomic HCV RNA to replicate in Huh7 cells. In addition, deletion and nucleotide substitutions of the 5'-proximal stem-loop structure, including the restored stem-loop by compensatory mutations, all resulted in reduction of translation by two- to fivefold, suggesting that the 5'-proximal stem-loop RNA element also modulates HCV RNA translation. These findings demonstrate that the 5'-proximal stem-loop of the HCV RNA is a cis-acting RNA element that regulates HCV RNA replication and translation.     

Recognition elements in rRNA for the tylosin resistance methyltransferase RlmA(II)

The methyltransferase RlmA(II) (formerly TlrB) is found in many Gram-positive bacteria, and methylates the N-1 position of nucleotide G748 within the loop of hairpin 35 in 23S rRNA. Methylation of the rRNA by RlmA(II) confers resistance to tylosin and other mycinosylated 16-membered ring macrolide antibiotics. We have previously solved the solution structure of hairpin 35 in the conformation that is recognized by the RlmA(II) methyltransferase from Streptococcus pneumoniae. It was shown that while essential recognition elements are located in hairpin 35, the interactions between RlmA(II) and hairpin 35 are insufficient on their own to support the methylation reaction. Here we use biochemical techniques in conjunction with heteronuclear/homonuclear nuclear magnetic resonance spectroscopy to define the RNA structures that are required for efficient methylation by RlmA(II). Progressive truncation of the rRNA substrate indicated that multiple contacts occur between RlmA(II) and nucleotides in stem-loops 33, 34 and 35. RlmA(II) appears to recognize its rRNA target through specific surface shape complementarity at the junction formed by these three helices. This means of recognition is highly similar to that of the orthologous Gram-negative methyltransferase, RlmA(I) (formerly RrmA), which also interacts with hairpin 35, but methylates at the adjacent nucleotide G745.     

On loop folding in nucleic acid hairpin-type structures

In a series of studies, combining NMR, optical melting and T-jump experiments, it was found that DNA hairpins display a maximum stability when the loop part of the molecule comprises four or five nucleotide residues. This is in contrast with the current notion based on RNA hairpin studies, from which it had been established that a maximum hairpin stability is obtained for six or seven residues in the loop. Here we present a structural model to rationalize these observations. This model is based on the notion that to a major extent base stacking interactions determine the stability of nucleic acid conformations. The model predicts that loop folding in RNA is characterized by an extension of the base stacking at the 5'-side of the double helix by five or six bases; the remaining gap can then easily be closed by two nucleotides. Conversely, loop folding in DNA is characterized by extending base stacking at the 3'-side of the double helical stem by two or three residues; again bridging of the remaining gap can then be achieved by one or two nucleotides. As an example of loop folding in RNA the anticodon loop of yeast tRNAPhe is discussed. For the DNA hairpin formed by d(ATCCTAT4TAGGAT) it is shown that the loop structure obtained from molecular mechanics calculations obeys the above worded loop folding principles.     

Two RNA species co-purify with RNase P from the fission yeast Schizosaccharomyces pombe.

RNase P activity from Schizosaccharomyces pombe co-purifies with two RNA species. These RNAs are associated with enzyme activity as judged by titrated micrococcal nuclease inactivation experiments. The two RNAs, K1- and K2-RNA, are 285 and 270 nucleotides long, respectively. Both RNAs are transcribed from one gene, present in a single copy in the haploid genome. The primary and a secondary structure of K RNAs have been determined and compared with M1 RNA, their counterpart from Escherichia coli. Very limited sequence homology was observed, and this agrees with the finding that no cross-hybridization with M1 RNA can be detected in a Southern analysis with yeast genomic DNA. However, the secondary structures of K RNA and M1 RNA show the same basic organization and one conserved local motif, the sequence GUG--AGGPu in an exposed hairpin loop.     

Two-dimensional combinatorial screening identifies specific 6'-acylated kanamycin A- and 6'-acylated neamine-RNA hairpin interactions

Herein, we report the RNA hairpin loops from a six-nucleotide hairpin library that bind 6'-acylated kanamycin A (1) and 6'-acylated neamine (2) identified by two-dimensional combinatorial screening (2DCS). Hairpins selected to bind 1 have K(d)'s ranging from 235 to 1035 nM, with an average K(d) of 618 nM. For 2, the selected hairpins bind with K(d)'s ranging from 135 to 2300 nM, with an average K(d) of 1010 nM. The selected RNA hairpin-ligand interactions are also specific for the ligand that they were selected to bind compared with the other arrayed ligand. For example, the mixture of hairpins selected for 1 on average bind 33-fold more tightly to 1 than to 2, while the mixtures of hairpins selected for 2 on average bind 11-fold more tightly to 2 than to 1. Secondary structure prediction of the selected sequences was completed to determine the motifs that each ligand binds, and the hairpin loop preferences for 1 and 2 were computed. For 1, the preferred hairpin loops contain an adenine separated by at least two nucleotides from a cytosine, for example, ANNCNN (two-tailed p-value = 0.0010) and ANNNCN (two-tailed p-value <0.0001). For 2, the preferred hairpin loops contain both 5'GC and 5'CG steps (two-tailed p-value <0.0001). These results expand the information available on the RNA hairpin loops that bind small molecules and could prove useful for targeting RNA.     

The role of essential pyrimidines in the hairpin ribozyme-catalysed reaction.

The hairpin ribozyme is an example of a small catalytic RNA that catalyses the endonucleolytic transesterification of RNA in a highly sequence-specific manner. We have utilised chemical synthesis of RNA to create mutants of the hairpin ribozyme in which a nucleoside analogue replaces one of the essential pyrimidines in the ribozyme. Individual pyrimidine nucleosides were substituted by 4-thiouridine, O4-methyluridine, O2-methyluridine or 2-pyrimidinone-1-beta-d-riboside. To facilitate the synthesis of oligoribonucleotides containing 4-thiouridine, we have devised a new synthetic route to the key intermediate 5'-O-(4, 4'-dimethoxytrityl)-2'-O-tert-butyldimethylsilyl-S-cyanoethyl-4-thiou ridine. The ability of the modified ribozymes to support catalysis was studied and the steady-state kinetic parameters were determined for each mutant. The range of analogues used in this study allows the important functional groups of the essential pyrimidines to be identified. The results demonstrate that each pyrimidine (U41, U42 and C25) plays an important role in hairpin ribozyme catalysis. The findings are discussed in terms of the various models that have been proposed for loop B of the hairpin ribozyme.