Identification of novel immunogenic Mycobacterium tuberculosis peptides that stimulate mononuclear cells from immune donors.

Proteins which are secreted or associated with the cell envelope of Mycobacterium tuberculosis may contain protective T-cell epitopes. Prior to this study, a recombinant clone bank of enzymatically active M. tuberculosis-alkaline phosphatase fusions, were screened for immunogenicity in a murine T-cell model. Five of these were selected for further study, and the IFN-gamma secretion and proliferation of human PBMC from purified protein derivative- (PPD)-positive and PPD-negative donors were measured in response to oligopeptides, Mtb-PhoA fusions and one full-length protein. Epitopes from four of the five selected antigens were immunoreactive in the human model and corresponded to cytochrome d ubiquinol oxidase, cytochrome c oxidase subunit II, MTV005.02 and MTV033.08. Thus, this strategy identified novel human immunogenic peptides as possible candidates for a subunit vaccine.

Mycobacterium tuberculosis efpA encodes an efflux protein of the QacA transporter family.

The Mycobacterium tuberculosis H37Rv efpA gene encodes a putative efflux protein, EfpA, of 55,670 Da. The deduced EfpA protein was similar in secondary structure to Pur8, MmrA, TcmA, LfrA, EmrB, and other members of the QacA transporter family (QacA TF) which mediate antibiotic and chemical resistance in bacteria and yeast. The predicted EfpA sequence possessed all transporter motifs characteristic of the QacA TF, including those associated with proton-antiport function and the motif considered to be specific to exporters. The 1,590-bp efpA open reading frame was G+C rich (65%), whereas the 40-bp region immediately upstream had an A+T bias (35% G+C). Reverse transcriptase-PCR assays indicated that efpA was expressed in vitro and in situ. Putative promoter sequences were partially overlapped by the A+T-rich region and by a region capable of forming alternative secondary structures indicative of transcriptional regulation in analogous systems. PCR single-stranded conformational polymorphism analysis demonstrated that these upstream flanking sequences and the 231-bp, 5' coding region are highly conserved among both drug-sensitive and multiply-drug-resistant isolates of M. tuberculosis. The efpA gene was present in the slow-growing human pathogens M. tuberculosis, Mycobacterium leprae, and Mycobacterium bovis and in the opportunistic human pathogens Mycobacterium avium and Mycobacterium intracellular. However, efpA was not present in 17 other opportunistically pathogenic or nonpathogenic mycobacterial species.

Diagnostic potential of sefA DNA probes to Salmonella enteritidis and certain other O-serogroup D1 Salmonella serovars.

Salmonella enteritidis thin fimbriae, SEF14, were found to be restricted to S. dublin and the predominantly poultry-associated members of the Salmonella O-serogroup D1, S. enteritidis, S. berta, S. gallinarum and S. pullorum, when tested by Western and ELISA analysis from among 90 Salmonella isolates of 42 serovars, as well as from members of several related genera of the Enterobacteriaceae. These five serovars and a single isolate of S. typhi (D1) were also detected by hybridization of genomic DNA from 732 Salmonella isolates of 117 serogroups to gene probes derived from the S. enteritidis sefA (fimbrin gene), sefB (chaperone) or sefC (outer membrane protein) genes encoding proteins involved in SEF14 biosynthesis. None of 250 Enterobacteriaceae or 27 other eubacterial isolates tested hybridized to the sef probes. The sefA, sefB and sefC genes were amplified from these six Salmonella serovars by PCR using primer pairs designed from sefA, sefB or sefC of S. enteritidis. DNA sequencing of sefA genes from these five serovars indicated limited sequence variability among sefA genes and recognition of individual base pairs which could potentially differentiate certain strains of S. enteritidis, S. dublin and S. gallinarum.

Salmonella enteritidis agfBAC operon encoding thin, aggregative fimbriae.

Salmonella enteritidis produces thin, aggregative fimbriae, named SEF17, which are composed of polymerized AgfA fimbrin proteins. DNA sequence analysis of a 2-kb region of S. enteritidis DNA revealed three contiguous genes, agfBAC. The 453-bp agfA gene encodes the AgfA fimbrin, which was predicted to be 74% identical and 86% similar in primary sequence to the Escherichia coli curli structural protein, CsgA. pHAG, a pUC18 derivative containing a 3.0-kb HindIII fragment encoding agfBAC, directed the in vitro expression of the major AgfA fimbrin, with an M(r) of 17,000, and a minor AgfB protein, with an M(r) of 16,000, encoded by the 453-bp agfB gene. AgfA was not expressed from pDAG, a pUC18 derivative containing a 3.1-kb DraI DNA fragment encoding agfA but not agfB. Primer extension analysis identified two adjacent transcription start sites located immediately upstream of agfB in positions analogous to those of the E. coli curlin csgBA operon. No transcription start sites were located immediately upstream of agfA or agfC. Northern (RNA) blot analysis confirmed that transcription of agfA was initiated from the agfB promoter region. Secondary-structure analysis of the putative mRNA transcript for agfBAC predicted the formation of a stem-loop structure (delta Gzero, -22 kcal/mol [-91 kJ/mol]) in the intercistronic region between agfA and agfC, which may be involved in stabilization of the agfBA portion of the agfBAC transcript. agfBAC and flanking regions had a high degree of sequence similarity with those counterparts of the E. coli curlin csgBA region for which sequence data are available. These data are demonstrative of the high degree of similarity between S. enteritidis SEF17 fimbriae and E. coli curli with respect to fimbrin amino acid sequence and genetic organization and, therefore, are indicative of a common and relatively recent ancestry.

The location of four fimbrin-encoding genes, agfA, fimA, sefA and sefD, on the Salmonella enteritidis and/or S. typhimurium XbaI-BlnI genomic restriction maps.

Four fimbrin-encoding genes, fimA (type-1 or SEF21 fimbriae), agfA (thin aggregative or SEF17 fimbriae), sefA (SEF14 fimbriae and sefD (SEF18 fimbriae) from Salmonella enteritidis (Se) 27655-3b were located onto the XbaI-BlnI genomic restriction maps of Salmonella typhimurium (St) LT2 and Se strains SSU7998 and 27655-3b. The XbaI or BlnI genomic fragments carrying these genes were identified by hybridization with labeled oligodeoxyribonucleotides or fimbrin-encoding genes. The fimbrin-encoding genes were not encoded by the virulence plasmids, but were located on chromosomal DNA fragments. The position of each gene on a given XbaI fragment was determined by hybridization of a series of XbaI-digested genomic DNA samples from previously characterized Tn10 mutants of Se and St with its respective probe. The fimA gene mapped near 13 centisomes (Cs) between purE884::Tn10 at 12.6 Cs (11.8 min) and apeE2::Tn10 at 12.8 Cs (12.3 min) beside the first XbaI site at 13.0 Cs in St or between purE884::Tn10 at 12.6 Cs and the XbaI site at 13.6 Cs in Se. The agfA gene mapped near 26 Cs between putA::Tn10 and pyrC691::Tn10 in St, but near 40 Cs between pncX::Tn10 and the XbaI site at 43.3 Cs in Se. This difference in map position was due to the location of agfA near one end of the 815-kb chromosomal fragment inverted between Se and St. The sefA and sefD genes mapped precisely at 97.6 Cs in Se, but were absent from the genome of St LT2. To verify the mapping procedures used herein, tctC was also mapped in both Salmonella serovars. As expected, tctC mapped near 60 Cs in both St and Se, thereby confirming previous studies.

fimA and tctC based DNA diagnostics for Salmonella.

Immunochemical analyses of 85 isolates of 17 Salmonella serovars using polyclonal antiserum to SEF21, the type 1 fimbriae of Salmonella enteritidis, demonstrated antigenic relatedness among both type 1 and type 2 fimbriae of Salmonella. However, anti-SEF21 antiserum was not entirely suitable as a Salmonella diagnostic probe due either to a variability of, or a rare deficiency of, detectable fimbriae. Partial amino acid sequence analyses of the SEF21 structural fimbrin protein revealed 99% homology to Salmonella typhimurium FimA. Therefore, oligonucleotide probes for Salmonella detection were designed following sequencing of S. enteritidis fimA and comparison to the corresponding genes of S. typhimurium, Escherichia coli, Klebsiella pneumoniae and Serratia marcescens. One oligonucleotide probe hybridized to all 612 Salmonella isolates of 89 serovars tested while two other probes detected 97.5% and 99.7% of the isolates. Three consistently weak positive reactions were obtained, therefore, inclusivity was optimized by identification of a Salmonella-specific tctC DNA probe that detected 609 of 612 Salmonella isolates. No hybridization of these Salmonella probes was detected to 250 other Enterobacteriaceae isolates or to 14 other eubacterial species. Therefore, in combination, DNA probes to fimA and tctC proved to be highly reliable diagnostics for Salmonella bacteria. Accordingly, PCR assays targeting fimA and tctC were developed.

DNA-based diagnostic tests for Salmonella species targeting agfA, the structural gene for thin, aggregative fimbriae.

Salmonella enteritidis 27655-3b and a few diarrheagenic Escherichia coli strains produce morphologically and antigenically related, thin, aggregative fimbriae, collectively named GVVPQ fimbriae (S. K. Collinson, L. Emödy, T. J. Trust, and W. W. Kay, J. Bacteriol. 174:4490-4495, 1992). To determine whether GVVPQ fimbriae are common to Salmonella spp. and other enteropathogenic members of the family Enterobacteriaceae, 113 isolates were phenotypically screened for Congo red binding and aggregative colony morphology. Presumptive positive and representative negative strains were examined by Western blotting (immunoblotting) by using antiserum to SEF 17, the native GVVPQ fimbria of S. enteritidis. Only four S. enteritidis strains and six E. coli isolates possessed substantial amounts of GVVPQ fimbriae after 24 h of incubation on T medium. Following 5 days of incubation, 56 of 93 Salmonella isolates (60%) and 1 of 7 additional E. coli clinical isolates possessed detectable levels of GVVPQ fimbriae. Since variable expression of GVVPQ fimbriae was observed among Salmonella isolates and some E. coli strains produced scant amounts, as revealed by immunoelectron microscopy, the ability to produce these fimbriae was evaluated by genotypic screening. The structural gene for the SEF 17 fimbrin, agfA, was amplified by the polymerase chain reaction, cloned, and sequenced to provide a characterized DNA probe. An agfA DNA fragment hybridized strongly to 603 of 604 (99.8%) Salmonella isolates but very weakly to 31 of 266 other members of the family Enterobacteriaceae including 26 of 137 E. coli strains, 3 of 14 Citrobacter spp., and single isolates of Shigella sonnei and Enterobacter cloacae. The agfA DNA probe proved to be a valuable diagnostic tool for Salmonella isolates arrayed on hydrophobic grid membrane filters. Unique agfA sequences were targeted in the development of a polymerase chain reaction assay specific for Salmonella spp.

Characterization of three fimbrial genes, sefABC, of Salmonella enteritidis.

Salmonella enteritidis produces thin, filamentous fimbriae designated SEF14. A 3.9-kb region of a 5.3-kb fragment encoding genes responsible for SEF14 biosynthesis was sequenced and found to contain three genes, sefABC. sefA encoded a novel fimbrin, the structural subunit of SEF14 fimbriae. sefB and sefC encoded proteins homologous to Escherichia coli and Klebsiella pneumoniae fimbrial periplasmic chaperone proteins and fimbrial outer membrane proteins, respectively, and are the first such genes to be characterized from Salmonella spp. in vitro expression directed by the 5.3-kb DNA fragment identified SefA, SefB, and SefC as approximately 14,000-, 28,000-, and 90,000-M(r) proteins, respectively, which correlated with their predicted amino acid sequences. sefB and sefC were not expressed in the absence of sefA. Primer extension analysis of sefABC revealed two major transcription start sites located upstream of sefA. Transcription of sefBC also initiated from the sefA promoter region. Secondary-structure analysis of the mRNA transcript for sefABC predicted the formation of two stable stem-loop structures in the intercistronic region between sefA and sefB indicative of differential regulation of SefA, SefB, and SefC translation. E. coli cells carrying the 5.3-kb DNA fragment of S. enteritidis DNA were unable to assemble distinguishable SEF14 fimbriae; however, immunogold-labelled SEF14 fimbriae were displayed on E. coli clones containing a 44-kb DNA fragment which encompassed the 5.3-kb region. Therefore, sefABC genes make up part of a complex sef operon responsible for the expression and assembly of SEF14 fimbriae.

Thin, aggregative fimbriae mediate binding of Salmonella enteritidis to fibronectin.

The binding of human fibronectin and Congo red by an autoaggregative Salmonella enteritidis strain was found to be dependent on its ability to produce thin, aggregative fimbriae, named SEF 17 (for Salmonella enteritidis fimbriae with an apparent fimbrin molecular mass of 17 kDa). Two other fimbrial types produced by S. enteritidis, SEF 14 and SEF 21, were not responsible for the aggregative phenotype or for fibronectin binding. SEF 17-negative TnphoA mutants which retained the ability to produce SEF 14 and SEF 21 were unable to bind human fibronectin or Congo red and lost the ability to autoaggregate. Only purified SEF 17 but not purified SEF 14 or SEF 21 bound fibronectin in a solid-phase binding assay. Furthermore, only SEF 17 was able to inhibit fibronectin binding to S. enteritidis whole cells in a direct competition enzyme-linked immunosorbent assay. These results indicate that SEF 17 are the fimbriae responsible for binding fibronectin by this enteropathogen.

High-level expression of the Streptomyces clavuligerus isopenicillin N synthase gene in Escherichia coli.

The pcbC gene, which encodes isopenicillin N synthase (IPNS), was subcloned from Streptomyces clavuligerus into Escherichia coli by using the pT7 series of plasmid vectors. The polymerase chain reaction was used to introduce an NdeI site at the translation initiation codon of pcbC, allowing the gene to be inserted behind an E. coli type of ribosome binding site. This construction directed high-level expression of IPNS, but the IPNS was in an inactive form in inclusion bodies. Active IPNS was recovered by solubilizing and renaturing the protein.

Isolation and characterization of a beta-lactamase-inhibitory protein from Streptomyces clavuligerus and cloning and analysis of the corresponding gene.

Culture filtrates of Streptomyces clavuligerus contain a proteinaceous beta-lactamase inhibitor (BLIP) in addition to a variety of beta-lactam compounds. BLIP was first detected by its ability to inhibit Bactopenase, a penicillinase derived from Bacillus cereus, but it has also been shown to inhibit the plasmid pUC- and chromosomally mediated beta-lactamases of Escherichia coli. BLIP showed no inhibitory effect against Enterobacter cloacae beta-lactamase, and it also showed no activity against an alternative source of B. cereus penicillinase. BLIP was purified to homogeneity, and sodium dodecyl sulfate-polyacrylamide gel electrophoresis gave a size estimate for BLIP of 16,900 to 18,000. The interaction between purified BLIP and the E. coli(pUC) beta-lactamase was investigated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and determined to be noncovalent, with an estimated 1:1 molar stoichiometry. The BLIP gene was isolated on a 13.5-kilobase fragment of S. clavuligerus chromosomal DNA which did not overlap a 40-kilobase region of DNA known to contain genes for beta-lactam antibiotic biosynthesis. The gene encoded a mature protein with a deduced amino acid sequence of 165 residues (calculated molecular weight of 17,523) and also encoded a 36-amino-acid signal sequence. No significant sequence similarity to BLIP was found by pairwise comparisons using various protein and nucleotide sequence data banks or by hybridization experiments, and no BLIP activity was detected in the culture supernatants of other Streptomyces spp.

Production of Streptomyces clavuligerus isopenicillin N synthase in Escherichia coli using two-cistron expression systems.

Streptomyces clavuligerus isopenicillin N synthase (IPNS) gene expression was achieved in Escherichia coli by the construction of two-cistron expression systems formed in the high copy number plasmid vector pUC119. These two-cistron constructions were composed of the IPNS gene and its flanking sequences which encoded an upstream open reading frame (ORF), the IPNS ribosome binding site and a putative transcription terminator. No E. coli- like Streptomyces promoter motif was present upstream of the IPNS gene therefore transcriptional regulation of the two-cistron system was provided by the lac promoter of pUC119. Enzymatically active IPNS was detected in E. coli cells harboring the recombinant plasmids thereby providing evidence for the activity of the IPNS ORF and for the feasibility of production of S. clavuligerus IPNS in E. coli. These results indicate that simple two-cistron constructions involving foreign gene flanking sequences may be used to express foreign proteins in E. coli.

Two human genes encoding tRNA(GCCGly).

Two human DNA fragments of 16.7 and 15.5 kb have been selected from a human lambda Charon-4A library by hybridization to an unfractionated tRNA probe. Restriction mapping and Southern and Northern hybridization analyses revealed the presence of a single tRNA-hybridizing region in each of the human DNA fragments. Nucleotide sequence analysis has identified two identical members of the tRNA(GCCGly) gene family. These tRNA(GCCGly) genes encode all of the conserved and semiconserved nucleotides of the tDNA split promoter sequences. Neither gene contains introns or encodes the CCA sequence present on the 3' terminus of mature tRNA. One of these identical tRNA(GCCGly) genes was found to be expressed at a substantially greater efficiency than the other in a HeLa cell lysate in vitro transcription system. No similarity was detected in the nucleotide sequences flanking these genes other than the characteristic, 3' oligo[dT] transcription termination signals and a TCTTT sequence located 7 to 10 bp upstream. These data are consistent with the hypothesis that as yet unidentified tDNA flanking sequences may have an important role in modulating human tRNA gene expression.

Plasmid transformation of Azotobacter vinelandii OP.

Azotobacter vinelandii OP which had been naturally induced to competence by growth in iron- and molybdenum-limited medium was transformed with the broad-host-range cloning vector pKT210. However, the transformation frequency at nearly saturating levels of DNA was 1000-fold lower for pKT210 than for a single chromosomal DNA marker (nif+). Plasmid- and chromosomal-DNA-mediated transformation events were competitive, magnesium-dependent, 42 degrees C-sensitive processes specific to double-stranded DNA, suggesting a common mechanism of DNA binding and uptake. The low frequency of plasmid transformation was not related to restriction of transforming DNA or to the growth period allowed for phenotypic expression. Covalently-closed-circular and open-circular forms of pKT210 transformed cells equally well whereas EcoRI- or HindIII-linearized pKT210 transformed cells with two to three times greater efficiency. Genetic transformation was enhanced 10- to 50-fold when pKT210 contained an insert fragment of A. vinelandii nif DNA, indicating that A. vinelandii possessed a homology-facilitated transformation system. However, all transformants failed to maintain the plasmid-encoded antibiotic resistance determinants, and extrachromosomal plasmid DNA was not recovered from these cells. Flush-ended pKT210 was not active in transformation; however, competent cells were transformed to Nif+ by HincII-digested plasmid DNA containing the cloned A. vinelandii nif-10 marker.

Structure of the Azotobacter vinelandii surface layer.

Electron microscopy of the Azotobacter vinelandii tetragonal surface array, negatively stained with ammonium molybdate in the presence of 1 mM calcium chloride, showed an apparent repeat frequency of 12 to 13 nm. Image processing showed dominant tetrad units alternating with low-contrast cruciform structures formed at the junction of slender linkers extending from corner macromolecules of four adjoining dominant units. The actual unit cell showed p4 symmetry, and a = b = 18.4 nm. Distilled water extraction of the surface array released a multimeric form of the single 60,000 molecular-weight protein (S protein) which constitutes the surface layer. The molecular weight of the multimer was estimated at 255,000 by gel filtration, indicating a tetrameric structure of four identical subunits and suggesting that this multimer was the morphological subunit of the S layer. Tetrameric S protein exhibited low intrinsic stability once released from the outer membrane, dissociating into monomers when incubated in a variety of buffers including those which served as the base for defined media used to cultivate A. vinelandii. The tetramer could not be stabilized in these buffers at any temperature between 4 and 30 degrees C, but the addition of 2 to 5 mM Ca2+ or Mg2+ completely prevented its dissociation into monomers. Circular dichroism measurements indicated that the secondary structure of the tetramer was dominated by aperiodic and beta-sheet conformations, and the addition of Ca2+ did not produce any gross changes in this structure. Only the tetrameric form of S protein was able to reassemble in vitro in the presence of divalent cations onto the surface of cells stripped of their native S layer.

Analysis of a human gene cluster coding for tRNA(GAAPhe) and tRNA(UUULys).

A 13.8-kb fragment of human DNA isolated from a human lambda Charon-4A DNA library was found to contain four human tRNA genes. Nucleotide sequence analysis of approx. 3.7 kb of this segment of human DNA identified two lysine tRNA(UUU) genes identical in coding sequence to a previously reported human lysine tRNA gene [Roy et al., Nucl. Acids Res. 10 (1982) 7313-7322]. The other two tRNA genes were phenylalanine tRNA(GAA) genes, the first to be isolated from a mammalian source. These phenylalanine tRNA(GAA) genes were identical in sequence with the exception of a G/A polymorphism at coordinate 57. None of these tRNA genes contains introns. The tRNA(UUULys) and tRNA(GAAPhe) genes are organized in alternating order and are irregularly spaced, by intergenic regions of approx. 1.0, 2.6 and 5.0 kb, and randomly oriented. There was no evidence to indicate that any of these genes arose by gene duplication, since flanking sequence homology was limited to the putative RNA polymerase III termination signals in the 3'-flanking regions. A mature tRNA-sized product was identified following the transcription of each tRNA gene in a homologous in vitro transcription system. Interestingly, different levels of transcriptional activity of the three identical lysine tRNA genes were observed, suggesting modulation of tDNA expression by extragenic sequences. In addition, a minimum of eight regions of homology to Alu-type repetitive elements were detected in this human DNA fragment, one of which was located 53 bp upstream from a tRNA(GAAPhe) gene.

Regular surface layer of Azotobacter vinelandii.

Washing Azotobacter vinelandii UW1 with Burk buffer or heating cells at 42 degrees C exposed a regular surface layer which was effectively visualized by freeze-etch electron microscopy. This layer was composed of tetragonally arranged subunits separated by a center-to-center spacing of approximately 10 nm. Cells washed with distilled water to remove an acidic major outer membrane protein with a molecular weight of 65,000 did not possess the regular surface layer. This protein, designated the S protein, specifically reattached to the surface of distilled-water-washed cells in the presence of the divalent calcium, magnesium, strontium, or beryllium cations. All of these cations except beryllium supported reassembly of the S protein into a regular tetragonal array. Although the surface localization of the S protein has been demonstrated, radioiodination of exposed envelope proteins in whole cells did not confirm this. The labeling behavior of the S protein could be explained on the basis of varying accessibilities of different tyrosine residues to iodination.

Heat sensitivity of Azotobacter vinelandii genetic transformation.

Heating competent Azotobacter vinelandii at 37 or 42 degrees C resulted in a total loss of competence with no loss of viability. The transformation process was relatively insensitive to heating at either temperature once DNase-resistant DNA binding was nearly complete. Although competent and 42 degrees C-treated cells bound equivalent amounts of [32P]DNA in a DNase-resistant state, no donor DNA marker (nif) or radioactivity was detected in the envelope-free cell lysate of heated cells, suggesting that DNA transport across the cell envelope was a heat-sensitive event. Competence was reacquired in a 42 degrees C-treated culture after 2 h of incubation at 30 degrees C by a process which required RNA and protein syntheses. The release of a surface glycoprotein, required for competence, from cells treated at 42 degrees C occurred in an insufficient amount to account for the total loss of competence. Recovery of competence in 42 degrees C-treated cells and further transformation of competent cells were prevented by the exposure of cells to saturating amounts of transforming DNA. Further DNase-resistant DNA binding, however, still occurred, suggesting that there were two types of receptors for DNase-resistant DNA binding to competent A. vinelandii. DNase-resistant DNA binding was dependent on magnesium ions, and at least one receptor type did not discriminate against heterologous DNA.