- Volume 144, Issue 2, 1998

Purple phototrophic bacteria have the ability to capture and use sunlight efficiently as an energy source. In these organisms, photosynthesis is carried out under anaerobic conditions. The introduction of oxygen into a culture growing phototrophically results in a rapid decrease in the synthesis of components of the photosynthetic apparatus and a change to an alternative source of energy, usually derived from the degradation of organic compounds under aerobic conditions (chemoheterotrophy). Switching back and forth between anaerobic (photosynthetic) and aerobic growth requires tight regulation of photosynthetic gene expression at the molecular level. Initial experiments by Cohen-Bazire et al. (1957) showed quite clearly that the regulation of photosynthetic gene expression was in response to two environmental stimuli. The most potent stimulus was oxygen; its presence shut down production of photosynthetic pigments very rapidly. To a lesser extent photosynthetic gene expression responded to light intensity. Low light intensity produced high levels of photosynthetic pigments; high light intensities caused a decrease, but the effect was less dramatic than that observed for oxygen. Since these initial observations were made in Rhodobacter sphaeroides some forty years ago, a great deal has been revealed as to the nature of the genes that encode the various components of the photosynthetic apparatus. Recent progress in the understanding of the regulation of expression of these genes in R. sphaeroides and Rhodobacter capsulatus is the subject of this review.

A 996 bp Aspergillus nidulans cDNA encoding the ASPND1 immunodominant antigen was cloned and expressed in Escherichia coli as a fusion protein with the enzyme glutathione S-transferase (GST) from Schistosoma japonicum. The GST-ASPND1 fusion protein was purified from isolated bacterial inclusion bodies by preparative SDS-PAGE. After cleavage with thrombin, the ASPND1 recombinant antigen (ASPND1r) and the GST protein were separated by SDS-PAGE and immunoblotted with a number of different human sera. The sera from 22 (88%) of 25 patients with an aspergilloma recognized the ASPND1r recombinant antigen on immunoblots. Forty-nine normal human sera and 14 sera from patients with other infections were unreactive. The ASPND1r expressed in E. coli could therefore be used, in combination with previously reported recombinant antigens, as a standardized antigen for serological and clinical diagnosis of Aspergillus-associated diseases.

A sulfite-reductase-type protein was purified from the hyperthermophilic crenarchaeote Pyrobaculum islandicum grown chemoorganoheterotrophically with thiosulfate as terminal electron acceptor. In common with dissimilatory sulfite reductases the protein has an α α β structure and contains high-spin sirohaem, non-haem iron and acid-labile sulfide. The oxidized protein exhibits absorption maxima at 280, 392, 578 and 710 nm with shoulders at 430 and 610 nm. The isoelectric point of pH 8.4 sets the protein apart from all dissimilatory sulfite reductases characterized thus far. The genes for the α- and β-subunits (dsrA and dsrB) are contiguous in the order dsrAdsrB and most probably comprise an operon with the directly following dsrG and dsrC genes. dsrG and dsrC encode products which are homologous to eukaryotic glutathione S-transferases and the proposed α-subunit of Desulfovibrio vulgaris sulfite reductase, respectively. dsrA and dsrB encode 44.2 kDa and 41.2 kDa peptides which show significant similarity to the two homologous subunits DsrA and DsrB of dissimilatory sulfite reductases. Phylogenetic analyses indicate a common protogenotic origin of the P. islandicum protein and the dissimilatory sulfite reductases from sulfate-reducing and sulfide-oxidizing prokaryotes. However, the protein from P. islandicum and the sulfite reductases from sulfate-reducers and from sulfur-oxidizers most probably evolved into three independent lineages prior to divergence of archaea and bacteria.

6-Deoxyerythronolide B synthase (DEBS) is a large multifunctional enzyme that catalyses the biosynthesis of the erythromycin polyketide aglycone. DEBS is organized into six modules, each containing the enzymic domains required for a single condensation of carboxylic acid residues which make up the growing polyketide chain. Module 1 is preceded by loading acyltransferase (AT-L) and acyl carrier protein (ACP-L) domains, hypothesized to initiate polyketide chain growth with a propionate-derived moiety. Using recombinant DNA technology several mutant strains of Saccharopolyspora erythraea were constructed that lack the initial AT-L domain or that lack both the AT-L and ACP-L domains. These strains were still able to produce erythromycin, although at much lower levels than that produced by the wild-type strain. In addition, the AT-L domain expressed as a monofunctional enzyme was able to complement the deletion of this domain from the PKS, resulting in increased levels of erythromycin production. These findings indicate that neither the initial AT-L nor the ACP-L domains are required to initiate erythromycin biosynthesis; however, without these domains the efficiency of erythromycin biosynthesis is decreased significantly. It is proposed that in these mutants the first step in erythromycin biosynthesis is the charging of KS1 with propionate directly from propionyl-CoA.

Albicidins are a family of phytotoxins and antibiotics which play an important role in the pathogenesis of sugarcane leaf scald disease. The albA gene from Klebsiella oxytoca encodes a protein which inactivates albicidin by heat-reversible binding. Albicidin ligand binding to a recombinant AlbA protein, purified by means of a glutathione S-transferase gene fusion system, is an almost instant and saturable reaction. Kinetic and stoichiometric analysis of the binding reaction indicated the presence of a single high affinity binding site with a dissociation constant of 6.4 x 10−8 M. The AlbA-albicidin complex is stable from 4 to 40 °, from pH 5 to 9 and in high salt solutions. Treatment with protein denaturants released all bound albicidin. These properties indicate that AlbA may be a useful affinity matrix for selective purification of albicidin antibiotics. AlbA does not bind to p-nitrophenyl butyrate or α-naphthyl butyrate, the substrates of the albicidin detoxification enzyme AlbD from Pantoea dispersa. The potential exists to pyramid genes for different mechanisms in transgenic plants to protect plastid DNA replication from inhibition by albicidins.

An Escherichia coli knockout ubiCA mutant has been constructed using a gene replacement method and verified using both Southern hybridization and PCR. The mutant, which was unable to synthesize ubiquinone (Q), showed severely diminished growth yields aerobically but not anaerobically with either nitrate or fumarate as terminal electron acceptors. Low oxygen uptake rates were demonstrated in membrane preparations using either NADH or lactate as substrates. However, these rates were greatly stimulated by the addition of ubiquinone-1 (Q-1). The rate of electron transfer to those oxidase components observable by photodissociation of their CO complexes was studied at sub-zero temperatures. In the ubiCA mutant, the reduced form of haemoproteins - predominantly cytochrome b 595-was reoxidized significantly faster in the presence of oxygen than in a Ubi+ strain, indicating the absence of Q as electron donor. Continuous multiple-wavelength recordings of the oxidoreduction state of cytochrome(s) b during steady-state respiration showed greater reduction in membranes from the ubiCA mutant than in wild-type membranes. A scheme for the respiratory electron-transfer chain in E. coli is proposed, in which Q functions downstream of cytochrome(s) b.

Wastewater is often treated using the activated sludge process. Flocculation and subsequent sedimentation of flocs are vital steps in this process that have direct influence on the quality of the effluent water from wastewater treatment plants. Since cells that remain free-living will decrease the quality of the effluent water it is important to understand the mechanisms of bacterial adhesion to flocs. The green fluorescent protein (GFP) was used as a cellular marker to study bacterial adhesion to activated sludge flocs in situ in sludge liquor. Cell surface hydrophobicity (CSH) was shown to be an important factor that determined the relative bacterial adhesion potential. High CSH correlated with high numbers of attached cells. However, the absolute adhesion of two test bacteria to different sludge flocs varied and could not be explained by the floc characteristics. Confocal laser scanning microscopy of GFP-marked cells showed their position in the floc matrix in situ. Hydrophobic cells attached not only on the surface but also within the floc, whereas hydrophilic cells did not. This indicates that cells may penetrate the flocs through channels and pores and increase the effective surface, which in turn makes the clarification of the wastewater effluent more efficient. The addition of polymers is common practice in wastewater treatment and was shown to increase bacterial adhesion to the flocs. A decrease in surface tension caused by addition of DMSO decreased adhesion, indicating the detrimental effect of surfactants on flocculation. An understanding of basic bacterial adhesion and aggregation mechanisms is important for the managment and control of biotechnological wastewater treatment.

Ethanol has been reported to cause mycelial growth in Candida tropicalis Pk233, and mycelial growth has also been shown to be abolished by concomitant addition of myo-inositol. In this study, the process of ethanol-induced mycelial growth in this organism was examined in combination with cytological characterization of actin localization. Cultivation with ethanol gave biphasic growth curves. During the first growth phase (doubling time 2.4 h), there was an accumulation of swollen spherical yeast cells, instead of the oblong ones observed in the control culture, followed by the appearance of spherical daughter cells in chains. Randomly distributed actin patches were observed on these swollen yeast cells and the bud initiation sites of these cells appeared random. These observations suggested that ethanol caused depolarization of cell growth during the first phase. During the second growth phase (doubling time 7.4 h), pseudohyphal cells appeared, projecting from the swollen yeast cells. Activity of chitinase in the control culture rose during the exponential phase. In the ethanol culture the activity stayed at a low level throughout the growth phases. When pseudohyphal cells were transferred to fresh ethanol medium, yeast cells appeared from pseudohyphal filaments and changed their shape to spherical, and filamentation appeared to be inhibited during the first phase. From these observations, an initial effect of ethanol on C. tropicalis cells appeared to be depolarization of cell growth, and the resulting swollen cells grew as polar pseudohyphal cells. In the culture supplemented with both ethanol and inositol, or with both ethanol and sorbitol, the accumulation of swollen cells was not observed and single yeast cells with normal oblong shape were seen throughout the growth phases.

Natural transformation of the soil bacterium Pseudomonas stutzeri JM300 in a non-sterile brown earth microcosm was studied. For this purpose, the microcosm was loaded with purified DNA (plasmid or chromosomal DNA, both containing a high-frequency-transformation marker, his +, of the P. stutzeri genome), the non-adsorbed DNA was washed out with soil extract and then the soil was charged with competent cells (his-1). Both chromosomal and plasmid transformants were found among the P. stutzeri cells recovered from the soil. The number of plasmid transformants increased in a linear fashion with the amount of DNA added [10-600 ng (0.7 g soil)−1]. The observed efficiency of transformation, the time course of transformant formation and the complete inhibition of transformation by DNase I, when added to the soil, were similar to that seen in optimized transformations in nutrient broth. Addition of cells as late as 3 d after loading the soil with plasmid DNA still yielded 3% of the initial transforming activity. This suggests that nucleases indigenous to the soil destroyed the transforming DNA, but at a rate allowing considerable DNA persistence. Transformants were also obtained when intact P. stutzeri cells were introduced into the soil to serve as plasmid DNA donors. Apparently, DNA was released from the cells, adsorbed to the soil material and subsequently taken up by recipient cells. The results indicate that competent cells of P. stutzeri were able to find access to and take up DNA bound on soil particles in the presence of micro-organisms and DNases indigenous to the soil.

The Candida albicans MKC1 gene encodes a mitogen-activated protein (MAP) kinase, which has been cloned by complementation of the lytic phenotype associated with Saccharomyces cerevisiae slt2 (mpk1) mutants. In this work, the physiological role of this MAP kinase in the pathogenic fungus C. albicans was characterized and a role for MKC1 in the biogenesis of the cell wall suggested based on the following criteria. First, C. albicans mkc1Δ/mkc1Δ strains displayed alterations in their cell surfaces under specific conditions as evidenced by scanning electron microscopy. Second, an increase in specific cell wall epitopes (O-glycosylated mannoprotein) was shown by confocal microscopy in mkc1Δ/mkc1Δ mutants. Third, the sensitivity to antifungals which inhibit (1,3)-β-glucan and chitin synthesis was increased in these mutants. In addition, evidence for a role for the MKC1 gene in morphological transitions in C. albicans is presented based on the impairment of pseudohyphal formation of mkc1Δ/mkc1Δ strains on Spider medium and on the effect of its overexpression on Sacch. cerevisiae colony morphology on SLADH medium. Using the two-hybrid system, it was also demonstrated that MKC1 is able to interact specifically with Sacch. cerevisiae Mkk1p and Mkk2p, the MAP-kinase kinases of the PKC1-mediated route of Sacch. cerevisiae, and to activate transcription in Sacch. cerevisiae when bound to a DNA-binding element. These results suggest a role for this MAP kinase in the construction of the cell wall of C. albicans and indicate its potential relevance for the development of novel antifungals.

Recent studies have revealed that fungi possess a mechanism similar to bacterial two-component systems to respond to extracellular changes in osmolarity. In Saccharomyces cerevisiae, SIn1p contains both histidine kinase and receiver (response regulator) domains and acts as an osmosensor protein that regulates the downstream HOG1 MAP kinase cascade. SLN1 of Candida albicans was functionally cloned using an S. cerevisiae strain in which SLN1 expression was conditionally suppressed. Deletion analysis of the cloned gene demonstrated that the receiver domain of C. albicans SIn1p was not necessary to rescue SLN1-deficient S. cerevisiae strains. Unlike S. cerevisiae, a null mutation of C. albicans SLN1 was viable under regular and high osmotic conditions, but it caused a slight growth retardation at high osmolarity. Southern blotting with C. albicans SLN1 revealed the presence of related genes, one of which is highly homologous to the NIK1 gene of Neurospora crassa. Thus, C. albicans harbours both SLN1 and NIK1 type histidine kinases.

The dnaA gene region of Streptococcus pneumoniae was cloned and sequenced. A tRNA gene, seven ORFs and three DnaA box clusters were identified. The order of the genes and intergene regions found was tRNAArg-orf1-DnaA box cluster 3-htrA-spo0J-DnaA box cluster 2-dnaA-DnaA box cluster 1-dnaN-orfX-orfY. Five ORFs are homologous to known bacterial genes. The tRNAArg gene and orf1, also called orfL, have already been described in pneumococci and have been reported to be preceded by the competence regulation locus comCDE. In Escherichia coli, htrA encodes a serine protease. In Bacillus subtilis, spo0J plays a role in sporulation and partition. dnaA encodes an initiator replication protein, very well conserved in several bacteria and dnaN encodes the β subunit of DNA polymerase III in E. coli. The function of orfX is unknown. The N-terminal part of another reading frame, orfY, revealed high homology with a GTP-binding protein. DnaA box clusters were found upstream and downstream from dnaA. The presence of two such clusters suggests that the chromosomal origin of S. pneumoniae is located within this region. The position of dnaA, and therefore the putative origin of replication, were localized on the physical map of S. pneumoniae.

Cleavage of chromosomal DNA from Pseudomonas aeruginosa PAO by SpeI and DpnI has been used together with PFGE and Southern hybridization to establish the map location of the following principal denitrification genes: narGH (encoding the large and small subunits of respiratory nitrate reductase), nirS (cytochrome-cd 1 nitrite reductase), nirE (uroporphyrinogen-III methyltransferase for haem d 1 biosynthesis), norCB (nitric-oxide reductase complex), nosZ (nitrous-oxide reductase) and nosA (an outer-membrane protein and OprC homologue). The study also included several genes related to anaerobic or microaerophilic metabolism: napA (encoding the catalytic subunit of the periplasmic nitrate reductase), ccoN (catalytic subunit of the cytochrome-cbb 3 oxidase), hemN (oxygen-independent coproporphyrinogen-III oxidase), an fnr-like regulatory gene, and azu and fdxA (electron carriers azurin and ferredoxin, respectively). Genes necessary for denitrification are concentrated at 20 to 36 min on the P. aeruginosa chromosome, where they form three separate loci, the nir-nor, nar and nos gene clusters. Genomic DNA of Pseudomonas stutzeri ZoBell was also subjected to SpeI restriction and Southern analysis to assign denitrification genes to individual fragments. A homologue of nosA encoding a putative component of the Cu-processing apparatus for nitrous-oxide reductase was identified. In both P. aeruginosa and P. stutzeri there is evidence for the linkage of anr (fnrA) with hemN and ccoN; and for the presence of a napA gene.

The Bacillus subtilis glpD gene encodes glycerol-3-phosphate (G3P) dehydrogenase. Expression of glpD is mainly controlled by termination/antitermination of transcription at an inverted repeat in the glpD leader. Antitermination is mediated by the antiterminator protein GlpP in the presence of G3P. In this paper, interaction between GlpP and glpD leader mRNA in vivo and in vitro is reported. In vivo, the antiterminating effect of GlpP can be titrated in a strain carrying the glpD leader on a plasmid. GlpP has been purified and gel shift experiments have shown that it binds to glpD leader mRNA in vitro. GlpP is not similar to other known antiterminator proteins, but database searches have revealed an Escherichia coli ORF which has a high degree of similarity to GlpP.

A large cellulolytic enzyme (CelA) with the ability to hydrolyse microcrystalline cellulose was isolated from the extremely thermophilic, cellulolytic bacterium ‘Anaerocellum thermophilum’. Full-length CelA and a truncated enzyme species designated CelA' were purified to homogeneity from culture supernatants. CelA has an apparent molecular mass of 230 kDa. The enzyme exhibited significant activity towards Avicel and was most active towards soluble substrates such as CM-cellulose (CMC) and β-glucan. Maximal activity was observed between pH values of 5 and 6 and temperatures of 95 ° (CM-cellulase) and 85 ° (Avicelase). Cellobiose, glucose and minor amounts of cellotriose were observed as end-products of Avicel degradation. The CelA-encoding gene was isolated from genomic DNA of ‘A. thermophilum’ by PCR and the nucleotide sequence was determined. The celA gene encodes a protein of 1711 amino acids (190 kDa) starting with the sequence found at the N-terminus of CelA purified from ‘A. thermophilum’. Sequence analysis revealed a multidomain structure consisting of two distinct catalytic domains homologous to glycosyl hydrolase families 9 and 48 and three domains homologous to family III cellulose-binding domain linked by Pro-Thr-Ser-rich regions. The enzyme is most closely related to CelA of Caldicellulosiruptor saccharolyticus (sequence identities of 96 and 97% were found for the N- and C-terminal catalytic domains, respectively). Endoglucanase CelZ of Clostridium stercorarium shows 70.4% sequence identity to the N-terminal family 9 domain and exoglucanase CelY from the same organism has 69.2% amino acid identity with the C-terminal family 48 domain. Consistent with this similarity on the primary structure level, the 90 kDa truncated derivative CelA' containing the N-terminal half of CelA exhibited endoglucanase activity and bound to microcrystalline cellulose. Due to the significantly enhanced Avicelase activity of full-length CelA, exoglucanase activity may be ascribed to the C-terminal family 48 catalytic domain.

Apocytochrome c 550 was detected in the periplasm of a new mutant of Paracoccus denitrificans, HN48, that is pleiotropically lacking c type cytochromes, produces reduced levels of siderophores and carries a Tn5 insertion in the ccmF gene for which sequence data, along with that for the contiguous ccmH, are reported. A counterpart to the ccmF gene was found in an archaebacterium but could not be located in the yeast genome, whereas mitochondrial haem lyases in the latter were not present in an archaeobacterial or in eubacterial genomes. A topological analysis for CcmF is presented which indicates at least eleven transmembrane helices, suggesting a role as a transporter; evidence against the substrate being haem is presented but sequence similarity with Escherichia coli γ-aminobutyric acid transporter was identified. Analysis by pulse-chase methodology has shown that, in this and another cytochrome-c-deficient mutant, the apo form of P. denitrificans cytochrome c 550 is much less stable than the holo form, directly demonstrating the presence of a periplasmic degradation system in P. denitrificans that removes non-functional proteins. A variety of phenotypes are observed for P. denitrificans mutated in different ccm genes, thus indicating that the stability of the ccm gene products does not require assembly of a complex of all the Ccm proteins.