- Volume 145, Issue 1, 1999

Summary: The 19 kDa carboxy-terminal domain of Plasmodium yoelii merozoite surface protein-1 (MSP119) was expressed in Salmonella vaccine strains as a carboxy-terminal fusion to fragment C of tetanus toxin (TetC). This study demonstrates that antibodies that recognize disulphide-dependent conformational epitopes in native MSP1 react with the TetC-MSP119 fusion protein expressed in Salmonella. The proper folding of MSP119 polypeptide is dependent on both the Salmonella host strain and the protein to which the MSP119 polypeptide is fused. Serum from mice immunized with Salmonella typhimurium C5aroD expressing TetC-MSP119 recognized native MSP1 as shown by immunofluorescence with P. yoelii-infected erythrocytes. Antibody levels to MSP119 were highest in out-bred mice immunized with S. typhimurium C5aroD carrying pTECH2-MSP119 and antibody was mostly directed against reduction-sensitive conformational epitopes. However, antibody levels were lower than in BALB/c mice immunized with a glutathione S-transferase (GST)-MSP119 fusion protein in Freund's adjuvant, and which were protected against P. yoelii challenge infection. In challenge experiments with P. yoelii the Salmonella-immunized mice were not protected, probably reflecting the magnitude of the antibody response. The results of this study have important implications in the design of live multivalent bacterial vaccines against eukaryotic pathogens.

Summary: Autolytic enzyme profiles of fast- and slow-growing mycobacteria were examined using SDS-PAGE zymography with incorportated mycobacterial peptidoglycan sacculi as substrate. Each species tested (Mycobacterium phlei, Mycobacterium smegmatis, Mycobacterium aurum, Mycobacterium fortuitum and Mycobacterium kansasii) appeared to produce a different set of enzymes on the basis of differing number and molecular masses. A major autolysin from M. phlei was purified to apparent homogeneity by DEAE-cellulose chromatography, preparative gel electrophoresis and Mono Q FPLC. This enzyme had an estimated molecular mass of 38 kDa, an isoelectric point of 5·5 and a pH optimum of pH 7·5. Digestion of purified peptidoglycan by the enzyme resulted in the appearance of reducing sugars, suggesting that the 38 kDa autolysin is a β-glycosidase. Partial internal amino acid sequence of the autolysin was determined and should facilitate identification, cloning and overexpression of the encoding gene.

Summary: Intravenous infection of guinea pigs with the fungus Aspergillus fumigatus resulted in increased levels of chitinase in serum and tissues of the animals. The molecular properties of the enzyme were demonstrated to be different from those of the fungal chitinase, but also from guinea pig lysozyme and β-N-acetylhexosaminidase. Bio-Gel P-100 gel filtration showed that in liver, spleen, heart and lung tissue of control animals there were two molecular mass forms present with apparent molecular masses of 35 kDa and 15 kDa. In brain and serum, only the 35 kDa form was detectable. Kidney showed only the 15 kDa form. Upon infection the 35 kDa form appeared in kidney and increased in the other tissues. When a less pathogenic form of the fungus was used the 35 kDa form remained absent in kidney. In contrast to human serum chitinase, the enzyme from guinea pig serum and tissues did bind to concanavalin A-Sepharose. This was the case for both molecular mass forms. The mode of cleavage of the substrate 4-methylumbelliferyl-tri-N-acetylchitotrioside (MU-[GlcNAc]3, where GlcNAc is N-acetylglucosamine) by the two forms of the enzyme was the same: both [GlcNAc]2 and [GlcNAc]3 were released. The chitinase activity levels in the control tissues showed a large variation in this order: spleen>lung, kidney>liver>heart>brain. The fact that spleen showed the highest chitinase level is in agreement with its major role as a lymphoid organ in cases of systemic infections. The relative increases upon infection were the highest for the tissues that showed low control values.

Summary: The activities of β-oxidation enzymes were measured in extracts of glucose-and triolein-grown cells of Aspergillus niger. Growth on triolein stimulated increased enzyme activity, especially for acyl-CoA dehydrogenase. No acyl-CoA oxidase activity was detected. HPLC analysis after incubation of triolein-grown cell extracts with decanoyl-CoA showed that β-oxidation was limited to one cycle. Octanoyl-CoA accumulated as the decanoyl-CoA was oxidized. β-Oxidation enzymes in isolated mitochondrial fractions were also studied. The results are discussed in the context of methyl ketone production by fungi.

Summary: A 309 bp DNA fragment from Entamoeba histolytica was amplified by PCR using primers derived from the Acanthamoeba castellanii consensus TATA-box binding protein amino acid sequence. The amplified fragment was used to isolate cDNA and genomic DNA clones containing an ORF encoding the complete E. histolytica TATA-box binding protein (Ehtbp, 702 bp, 234 aa, molecular mass 26 kDa). The EhTBP functional domain showed 55% sequence identity to that of Homo sapiens, 54% to A. castellanii and 37% to Plasmodium falciparum TBPs. In Southern blot experiments we detected a single Ehtbp band, which was transcribed as a 1.3 kb mRNA containing a 420 nt 5′ untranslated region. However, the probe hybridized with the 0.8 and 1.5 Mb chromosomes, suggesting that this sequence is diploid. In situ PCR assays showed two signals in 95% of trophozoites, one located in the nucleus and another in EhkO, the novel DNA-containing organelle recently reported. The recombinant E. histolytica TATA-box binding protein was expressed in Escherichia coli. Antibodies against it recognized two proteins of 26 and 29 kDa in E. histolytica nuclear extracts. Confocal microscopy immunofluorescence analysis located the protein in both the nucleus and EhkO.

Summary: Escherichia coli growing under anaerobic conditions produces several molybdoenzymes, such as formate hydrogenlyase (formate to H2 and CO2; hyc and fdhF genes) and nitrate reductase (narGHJI genes). Synthesis of these molybdoenzymes, even in the presence of the cognate transcriptional activators and effectors, requires molybdate in the medium. Besides the need for molybdopterin cofactor synthesis, molybdate is also required for transcription of the genes encoding these molybdoenzymes. In E. coli, ModE was previously identified as a repressor controlling transcription of the operon encoding molybdate transport components (modABCD). In this work, the ModE protein was also found to be a required component in the activation of hyc-lacZ to an optimum level, but only in the presence of molybdate. Mutant ModE proteins which are molybdate-independent for repression of modA-lacZ also restored hyc-lacZ expression to the wild-type level even in the absence of molybdate. Nitrate-dependent enhancement of transcription of narX-iacZ was completely abolished in a modE mutant. Nitrate-response by narG-lacZ and nark-lacZ was reduced by about 50% in a modE mutant. DNase I footprinting experiments revealed that the ModE protein binds the hyc promoter DNA in the presence of molybdate. ModE-molybdate also protected DNA in the intergenic region between narXL and narK from DNase I hydrolysis. DNA sequences (5′ TAYAT 3′ and 5′ GTTA 3′) found in ModE-molybdate-protected modABCD operator DNA were also found in the ModE-molybdate-protected region of hyc promoter DNA (5′ GTTA-7 bp-CATAT 3′) and narX-narK intergenic region (5′ GTTA-7 bp-TACAT 3′). Based on these results, a working model is proposed in which ModE-molybdate serves as a secondary transcriptional activator of both the hyc and narXL operons which are activated primarily by the transcriptional activators, FhIA and NarL, respectively.

Summary: A γ-D-glutamate-meso-diaminopimelate muropeptidase was detected in the vegetative growth phase of Bacillus subtilis 168. It is encoded by the monocistronic lytF operon expressed by the alternative vegetative sigma factor, σD. Sequence analysis of LytF revealed two domains, an organization common to exoproteins of B. subtilis as well as to those from other organisms. The N-terminal domain contains a fivefold-repeated motif attributed to cell wall binding, whilst the C-terminal domain is probably endowed with the catalytic activity. Overrexpression of LytF allowed its purification and biochemical characterization. Inactivation of lytF led to the loss of the cell-wall-bound protein 49′ (CWBP49′) and of the corresponding lytic activity as revealed by renaturation gel assay. Native cell walls prepared from the multiple lytC lytD lytE lytF-deficient mutant did not exhibit any autolysis, whereas walls prepared from a strain endowed with LytF but not with the other three enzymes underwent a slight lysis. Analysis of degradation products of cell wall devoid of teichoic-acid-bound O-esterified D-alanine unambiguously confirmed that LytF cuts the γ-D-glutamate-meso-diaminopimelate bond.

Summary: A 3·5 kb EcoRI-BamHI fragment of Bacillus subtilis chromosomal DNA carrying the ribR gene, involved in regulation of the B. subtilis riboflavin operon, was cloned in the B. subtilis-Escherichia coli shuttle vector pCB20. DNA sequence analysis of this fragment revealed several ORFs, one of which encodes a polypeptide of 230 amino acids with up to 45% sequence identity with FAD synthetases from a number of micro-organisms, such as Corynebacterium ammoniagenes, E. coli and Pseudomonas fluorescens, and also to the ribC gene product of B. subtilis. The ribR gene was amplified by PCR, cloned and expressed in E. coli. Measurement of flavokinase activity in cell extracts demonstrated that ribR encodes a monofunctional flavokinase which converts riboflavin into FMN but not to FAD, and is specific for the reduced form of riboflavin.

Summary: The XPR2 gene from Yarrowia lipolytica encodes an inducible alkaline extracellular protease. Its complex regulation involves pH, carbon, nitrogen and peptones. Two previously identified upstream activating sequence (UAS) regions were analysed in a reporter system, outside the XPR2 context. Fragments from the UAS regions were inserted upstream of a minimal LEU2 promoter directing the expression of a reporter gene. The activity of the hybrid promoters was assessed following integration into the Y. lipolytica genome. This study confirmed the presence of two UASs composed of several interacting elements. Within the distal UAS (UAS1), a TUF/RAP1 binding site exhibited a UAS activity, which was enhanced by the presence of two adjacent repeats, overlapping sites similar to the CAR1 upstream repressing sequence from Saccharomyces cerevisiae. Within the proximal UAS (UAS2), the UAS activity required the interaction of both an ABF1-like binding site and a decameric repeat, containing Aspergillus nidulans PacC site consensus sequences. This decameric repeat was able to mediate repression due to carbon and/or nitrogen sources as well as pH-dependent activation. A study in the context of trans-regulatory mutations in the Y. lipolytica RIM101 gene showed that the PacC-like sites, potential binding sites for YIRim101p, were implicated in the derepression of UAS2-driven expression at neutral-alkaline pH. The in vivo response of the PacC-like decamers to external pH was dependent on the status of the pH-regulated activator YIRim101p, which is homologous to the A. nidulans PacC regulator. The carbon/nitrogen regulation imposed on the decamers was shown to be independent of YIRim101p and to override its effects.

Summary: The groESL bicistronic operon from the enteric pathogen Campylobacter jejuni was cloned and sequenced. It consists of two ORFs encoding proteins with molecular masses of 9·5 and 57·9 kDa, which showed a high degree of homology to other bacterial GroES and GroEL proteins. Northern blot analysis suggested that the groESL operon is transcribed as a bicistronic mRNA, and its steady-state level was markedly increased after temperature upshift. By primer extension assay, one potential transcription start point preceding the groESL genes could be demonstrated, and a putative promoter region compatible with both Escherichia coli and C. jejuni σ70consensus sequences was identified. A conserved inverted repeat, which is believed to be involved in the regulation of the groESL genes, was found between the −10 promoter box and the groES translation start site. The complete coding region of groEL was fused with pET-22b(+) and expressed in E. coli as a His6-tagged recombinant protein (rCjHsp60-His). After purification, the protein was recognized by an anti-HSP60 monoclonal antibody. ELISA and Western immunoblotting experiments showed that IgG and IgA antibody responses against rCjHsp60-His were not significantly increased in sera from 24 patients with sporadic Campylobacter infection when compared to sera from 16 healthy controls.

Summary: The internal spacer region (ISR) between the 16S and 23S rRNS genes of Campylobacter was investigated by PCR fragment length typing and DNA sequencing of clinical and chicken wild-type isolates. PCR fragment length typing showed one fragment of 859 nt in length for the 12 strains of Campylobacter coli investigated. Thirty-six of the Campylobacter jejuni subsp. jejuni strains possessed one fragment, which varied in size between 727 and 802 nt. Three strains showed two fragments between 501 and 923 nt. Strains of C. jejuni subsp. doylei, Campylobacter lari and Campylobacter upsaliensis possessed one or two fragments with lengths different from those of C. coli and C. jejuni subsp. jejuni. DNA sequences were obtained from 54 nt downstream of rrs up to rrl of four strains of C. coli, eight strains of C. jejuni subsp. jejuni, and one strain each of C. jejuni subsp. doylei and C. lari, selected to represent the different biotypes of Campylobacter. ISR lengths determined by PCR fragment length typing and DNA sequencing corresponded for 12 strains. For two strains of C. coli, PCR fragment length typing underestimated ISR lengths by 159 and 193 nt, probably related to incomplete resolution of the distal helical structures, which were not fully denatured during PAGE. For the 14 strains and the published C. jejuni subsp. jejuni sequence, the first 206-211 nt were conserved and included the two tRNA genes in the characteristic tRNAAla to tRNAlle order separated by a short 8-9 nt spacer region. Within the region downstream of tRNAlle, conserved regions were identified which allowed a separation of C. lari from C. coli and C. jejuni but not separation of C. coli from C. jejuni. The 69-282 nt longer variable regions in C. coli strains allowed separation of this species from C. jejuni, confirming results obtained by PCR typing. Certain nucleic acid positions in variable regions were related to the Lior biotypes. Sequence information from ISRs of more strains is needed to ascertain if separation of species and biotypes will be possible for diagnostic purposes.

Summary: A gene cluster containing homologues of the genes cysB, cysJI and cysH was found in the genome of the sulphur-oxidizing purple bacterium Thiocapsa roseopersicina. The nucleotide sequence indicated four open reading frames encoding homologues of 3′-phosphoadenylylsulphate (PAPS) reductase (CysH), sulphite reductase flavoprotein (CysJ) and haem protein (Cysl) subunits, and a transcriptional regulator (CysB). Genes cysJIH are separated by a short cis-active intergenic region from cysB which is transcribed divergently. cysB encodes a polypeptide of 35·9 kDa consisting of 323 amino acid residues with 40% identity to the CysB regulator from enterobacteria. cysH encodes a protein with 239 amino acid residues and a calculated mass of 27·7 kDa; cysJ encodes a protein with 522 amino acid residues and a mass of 57·8 kDa; and cysl encodes a protein with 559 amino acid residues and a mass of 62·3 kDa. The cysJIH gene products have been expressed and used for complementation of cys mutants from Escherichia coli Biochemical analysis. The gene product CysH is a thioredoxin-dependent PAPS reductase (EC 1.8.99.4). It was repressed under photoautotrophic growth using hydrogen sulphide as electron donor and derepressed under conditions of sulphate deficiency. Products of the cysJI genes were identified as the two subunits of NADPH-sulphite reductase (EC 1.8.1.2). cysJ encoded the flavoprotein, with 39% identity to the protein from E. coli, and cysl encoded the haem protein, with 53% identity. A cysl clone was used to complement the corresponding mutant from E. coli and to express enzymically active methylviologen-sulphite reductase.