Signal peptide hydrophobicity is critical for early stages in protein export by Bacillus subtilis

Signal peptides that direct protein export in Bacillus subtilis are overall more hydrophobic than signal peptides in Escherichia coli. To study the importance of signal peptide hydrophobicity for protein export in both organisms, the alpha-amylase AmyQ was provided with leucine-rich (high hydrophobicity) or alanine-rich (low hydrophobicity) signal peptides. AmyQ export was most efficiently directed by the authentic signal peptide, both in E. coli and B. subtilis. The leucine-rich signal peptide directed AmyQ export less efficiently in both organisms, as judged from pulse-chase labelling experiments. Remarkably, the alanine-rich signal peptide was functional in protein translocation only in E. coli. Cross-linking of in vitro synthesized ribosome nascent chain complexes (RNCs) to cytoplasmic proteins showed that signal peptide hydrophobicity is a critical determinant for signal peptide binding to the Ffh component of the signal recognition particle (SRP) or to trigger factor, not only in E. coli, but also in B. subtilis. The results show that B. subtilis SRP can discriminate between signal peptides with relatively high hydrophobicities. Interestingly, the B. subtilis protein export machinery seems to be poorly adapted to handle alanine-rich signal peptides with a low hydrophobicity. Thus, signal peptide hydrophobicity appears to be more critical for the efficiency of early stages in protein export in B. subtilis than in E. coli.     

Secretion of the Escherichia coli outer membrane proteins OmpA and OmpF in Bacillus subtilis is blocked at an early intracellular step

When the genes coding for the outer membrane (OM) proteins OmpA and OmpF of Escherichia coli are fused to a signal sequence of a bacillar exoenzyme and expressed in Bacillus subtilis they remain cell-bound and the signal sequence is not cleaved. To identify the step of arrest in the export of these proteins we studied their accessibility to protease applied to intact protoplasts; they remained resistant indicating fully intracellular localization. Both proteins appeared associated with the cell membranes in sedimentation and flotation centrifugation experiments. However, OmpA and OmpF proteins synthesized in B. subtilis without a signal sequence were similarly associated with membranes in centrifugation experiments whereas electron microscopy showed the presence of intracytoplasmic inclusion bodies not obviously attached to the cytoplasmic membrane. We conclude that OmpA and OmpF proteins even when provided with a functional signal sequence do not enter the export pathway in B. subtilis, probably owing to lack of a specific export component in B. subtilis.     

Use of alkaline phosphatase fusions to study protein secretion in Bacillus subtilis.

We have constructed a vector designed to facilitate the study of protein secretion in Bacillus subtilis. This vector is based on a translational fusion between the expression elements and signal sequence of Bacillus amyloliquefaciens alkaline protease and the mature coding sequence for Escherichia coli alkaline phosphatase (phoA). We show that export of alkaline phosphatase from B. subtilis depends on a functional signal sequence and that alkaline phosphatase activity depends upon secretion. The vector design facilitates the insertion of heterologous coding sequences between the signal and phoA to generate three-part translational fusions. Such phoA fusions are easily analyzed by monitoring alkaline phosphatase activity on agar plates or in culture supernatants or by immunological detection. Exploitation of this methodology, which has proven to be extremely useful in the study of protein secretion in E. coli, has a variety of applications for studying protein secretion in B. subtilis.     

Secretion activities of Bacillus subtilis alpha-amylase signal peptides of different lengths in Escherichia coli cells

The B. subtilis alpha-amylase promoter and signal peptide are functional in E. coli cells. DNA fragments coding for signal peptides with different lengths (28, 31, 33 and 41 amino acids from the translation initiator Met) were prepared and fused with the E. coli beta-lactamase structural gene. In B. subtilis cells, the sequences of 31, 33 and 41 amino acids were able to secrete beta-lactamase into the surrounding media, but the 28 amino acid sequence was not. In contrast, all of the four sequences were able to export beta-lactamase into the periplasmic space of E. coli cells. Thus, the recognition of the B. subtilis alpha-amylase signal peptide in E. coli cells seems to be different from that in B. subtilis cells.     

Chloramphenicol acetyltransferase, a cytoplasmic protein is incompatible for export from Bacillus subtilis.

Bacillus subtilis cells expressing a hybrid protein (Lvsss-Cat) consisting of the B. amyloliquefaciens levansucrase signal peptide fused to B. pumilus chloramphenicol acetyltransferase (Cat) are unable to export Cat protein into the growth medium. A series of tripartite protein fusions was constructed by inserting various lengths of the Cat sequences between the levansucrase signal peptide and staphylococcal protein A or Escherichia coli alkaline phosphatase. Biochemical characterization of the various Cat protein fusions revealed that multiple regions in the Cat protein were causing the export defect.     

Isolation and molecular genetic characterization of the Bacillus subtilis gene (infB) encoding protein synthesis initiation factor 2.

Western blot (immunoblot) analysis of Bacillus subtilis cell extracts detected two proteins that cross-reacted with monospecific polyclonal antibody raised against Escherichia coli initiation factor 2 alpha (IF2 alpha). Subsequent Southern blot analysis of B. subtilis genomic DNA identified a 1.3-kilobase (kb) HindIII fragment which cross-hybridized with both E. coli and Bacillus stearothermophilus IF2 gene probes. This DNA was cloned from a size-selected B. subtilis plasmid library. The cloned HindIII fragment, which was shown by DNA sequence analysis to encode the N-terminal half of the B. subtilis IF2 protein and 0.2 kb of upstream flanking sequence, was utilized as a homologous probe to clone an overlapping 2.76-kb ClaI chromosomal fragment containing the entire IF2 structural gene. The HindIII fragment was also used as a probe to obtain overlapping clones from a lambda gt11 library which contained additional upstream and downstream flanking sequences. Sequence comparisons between the B. subtilis IF2 gene and the other bacterial homologs from E. coli, B. stearothermophilus, and Streptococcus faecium displayed extensive nucleic acid and protein sequence homologies. The B. subtilis infB gene encodes two proteins, IF2 alpha (78.6 kilodaltons) and IF2 beta (68.2 kilodaltons); both were expressed in B. subtilis and E. coli. These two proteins cross-reacted with antiserum to E. coli IF2 alpha and were able to complement in vivo an E. coli infB gene disruption. Four-factor recombination analysis positioned the infB gene at 145 degrees on the B. subtilis chromosome, between the polC and spcB loci. This location is distinct from those of the other major ribosomal protein and rRNA gene clusters of B. subtilis.     

Synthesis of OmpA protein of Escherichia coli K12 in Bacillus subtilis

We have inserted a C-terminally truncated gene of the major outer membrane protein OmpA of Escherichia coli downstream from the promoter and signal sequence of the secretory alpha-amylase of Bacillus amyloliquefaciens in a secretion vector of Bacillus subtilis. B. subtilis transformed with the hybrid plasmid synthesized a protein that was immunologically identified as OmpA. All the protein was present in the particulate fraction. The size of the protein compared to the peptide synthesized in vitro from the same template indicated that the alpha-amylase derived signal peptide was not removed; this was verified by N-terminal amino acid sequence determination. The lack of cleavage suggests that there was little or no translocation of OmpA protein across the cytoplasmic membrane. This is an unexpected difference compared with periplasmic proteins, which were both secreted and processed when fused to the same signal peptide. A requirement of a specific component for the export of outer membrane proteins is suggested.     

Length and structural effect of signal peptides derived from Bacillus subtilis alpha-amylase on secretion of Escherichia coli beta-lactamase in B. subtilis cells.

The precursor of Bacillus subtilis alpha-amylase contains an NH2-terminal extension of 41 amino acid residues as the signal sequence. The E. coli beta-lactamase structural gene was fused with the DNA for the promoter and signal sequence regions. Activity of beta-lactamase was expressed and more than 95% of the activity was secreted into the culture medium. DNA fragments coding for short signal sequences 28, 31, and 33 amino acids from the initiator Met were prepared and fused with the beta-lactamase structural gene. The sequences of 31 and 33 amino acid residues with Ala COOH-terminal amino acid were able to secrete active beta-lactamase from B. subtilis cells. However beta-lactamase was not secreted into the culture medium by the shorter signal sequence of 28 amino acid residues, which was not cleaved. Molecular weight analysis of the extracellular and cell-bound beta-lactamase suggested that the signal peptide of B. subtilis alpha-amylase was the first 31 amino acids from the initiator Met. The significance of these results was discussed in relation to the predicted secondary structure of the signal sequences.     

Isolation of a secY homologue from Bacillus subtilis: evidence for a common protein export pathway in eubacteria

Genetic and biochemical studies have shown that the product of the Escherichia coli secY gene is an integral membrane protein with a central role in protein secretion. We found the Bacillus subtilis secY homologue within the spc-alpha ribosomal protein operon at the same position occupied by E. coli secY. B. subtilis secY coded for a hypothetical product 41% identical to E. coli SecY, a protein thought to contain 10 membrane-spanning segments and 11 hydrophilic regions, six of which are exposed to the cytoplasm and five to the periplasm. We predicted similar segments in B. subtilis SecY, and the primary sequences of the second and third cytoplasmic regions and the first, second, fourth, fifth, seventh, and tenth membrane segments were particularly conserved, sharing greater than 50% identity with E. coli SecY. We propose that the conserved cytoplasmic regions interact with similar cytoplasmic secretion factors in both organisms and that the conserved membrane-spanning segments actively participate in protein export. Our results suggest that despite the evolutionary differences reflected in cell wall architecture, Gram-negative and Gram-positive bacteria possess a similar protein export apparatus.     

Escherichia coli signal peptides direct inefficient secretion of an outer membrane protein (OmpA) and periplasmic proteins (maltose-binding protein, ribose-binding protein, and alkaline phosphatase) in Bacillus subtilis.

Signal peptides of gram-positive exoproteins generally carry a higher net positive charge at their amino termini (N regions) and have longer hydrophobic cores (h regions) and carboxy termini (C regions) than do signal peptides of Escherichia coli envelope proteins. To determine if these differences are functionally significant, the ability of Bacillus subtilis to secrete four different E. coli envelope proteins was tested. A pulse-chase analysis demonstrated that the periplasmic maltose-binding protein (MBP), ribose-binding protein (RBP), alkaline phosphatase (PhoA), and outer membrane protein OmpA were only inefficiently secreted. Inefficient secretion could be ascribed largely to properties of the homologous signal peptides, since replacing them with the B. amyloliquefaciens alkaline protease signal peptide resulted in significant increases in both the rate and extent of export. The relative efficiency with which the native precursors were secreted (OmpA >> RBP > MBP > PhoA) was most closely correlated with the overall hydrophobicity of their h regions. This correlation was strengthened by the observation that the B. amyloliquefaciens levansucrase signal peptide, whose h region has an overall hydrophobicity similar to that of E. coli signal peptides, was able to direct secretion of only modest levels of MBP and OmpA. These results imply that there are differences between the secretion machineries of B. subtilis and E. coli and demonstrate that the outer membrane protein OmpA can be translocated across the cytoplasmic membrane of B. subtilis.     

Characterisation of preYvaY export reveals differences in the substrate specificities of Bacillus subtilis and Escherichia coli leader peptidases

Translocation, processing and secretion of YvaY, a Bacillus subtilis protein of unknown function, were characterised both in B. subtilis and in Escherichia coli. In its natural host B. subtilis, YvaY was transiently synthesised at the end of the exponential growth phase. It was efficiently secreted into the culture supernatant in spite of a calculated membrane spanning domain in the mature part of the protein. In E. coli, despite the high conservation of Sec-dependent transport components, processing of preYvaY was strongly impaired. To uncover which elements of E. coli and B. subtilis translocation systems are responsible for the observed substrate specificity, components of the B. subtilis Sec-system were co-expressed besides yvaY in E. coli. Expression of B. subtilis secA or secYEG genes did not affect processing, but expression of B. subtilis signal peptidase genes significantly enhanced processing of preYvaY in E. coli. While the major signal peptidases SipS or SipT had a strong stimulatory effect on preYvaY processing, the minor signal peptidases SipU, SipV or SipW had a far less stimulatory effect in E. coli. These results reveal that targeting and translocation of preYvaY is mediated by the E. coli Sec proteins but processing of preYvaY is not performed by E. coli signal peptidase LepB. Thus, differences in substrate specificities of E. coli LepB and the B. subtilis Sip proteins provide the bottleneck for export of YvaY in E. coli. Significant slower processing of preYvaY in absence of SecB indicated that SecB mediates targeting of the B. subtilis precursor.     

Cloning and characterization of Bacillus subtilis homologs of Escherichia coli cell division genes ftsZ and ftsA.

The Bacillus subtilis homolog of the Escherichia coli ftsZ gene was isolated by screening a B. subtilis genomic library with anti-E. coli FtsZ antiserum. DNA sequence analysis of a 4-kilobase region revealed three open reading frames. One of these coded for a protein that was about 50% homologous to the E. coli FtsZ protein. The open reading frame just upstream of ftsZ coded for a protein that was 34% homologous to the E. coli FtsA protein. The open reading frames flanking these two B. subtilis genes showed no relationship to those found in E. coli. Expression of the B. subtilis ftsZ and ftsA genes in E. coli was lethal, since neither of these genes could be cloned on plasmid vectors unless promoter sequences were first removed. Cloning the B. subtilis ftsZ gene under the control of the lac promoter resulted in an IPTGs phenotype that could be suppressed by overproduction of E. coli FtsZ. These genes mapped at 135 degrees on the B. subtilis genetic map near previously identified cell division mutations.     

Sequence comparison of new prokaryotic and mitochondrial members of the polypeptide chain release factor family predicts a five-domain model for release factor structure.

We have recently reported the cloning and sequencing of the gene for the mitochondrial release factor mRF-1. mRF-1 displays high sequence similarity to the bacterial release factors RF-1 and RF-2. A database search for proteins resembling these three factors revealed high similarities to two amino acid sequences deduced from unassigned genomic reading frames in Escherichia coli and Bacillus subtilis. The amino acid sequence derived from the Bacillus reading frame is 47% identical to E.coli and Salmonella typhimurium RF-2, strongly suggesting that it represents B.subtilis RF-2. Our comparison suggests that the expression of the B.subtilis gene is, like that of the E.coli and S. typhimurium RF-2 genes, autoregulated by a stop codon dependent +1 frameshift. A comparison of prokaryotic and mitochondrial release factor sequences, including the putative B.subtilis RF-2, leads us to propose a five-domain model for release factor structure. Possible functions of the various domains are discussed.     

Effect of alteration of charged residues at the N termini of signal peptides on protein export in Bacillus subtilis.

The role of positively charged residues at the N termini of signal peptides in protein export has been studied in Bacillus subtilis. Bacillus signal peptides (alkaline protease [Apr] and neutral protease [Npr] from Bacillus amyloliquefaciens) were altered and fused to mature levansucrase (Lvs). The effects of the various alterations on the export of Lvs in B. subtilis were determined. The replacement of positively charged residues with neutral residues in both Apr and Npr signal peptides resulted in a slight defect in the export of Lvs from B. subtilis. Introduction of a negatively charged residue (aspartic acid) at the N terminus of Npr signal peptide blocked the export of Lvs. However, Apr signal peptide with a net charge of -3 (three aspartic acid residues) was still functional.     

Bacillus subtilis CheN, a homolog of CheA, the central regulator of chemotaxis in Escherichia coli.

The Bacillus subtilis cheN gene was isolated, sequenced, and expressed. It encodes a large negatively charged protein with a molecular weight of approximately 74,000. The predicted protein sequence has 33 to 34% identity with the Escherichia coli and Salmonella typhimurium CheA and Myxococcus xanthus FrzE sequences. These proteins are found to autophosphorylate and are members of the same histidine kinase signal modulating family. CheN has several conserved regions (including the histidine that is phosphorylated in CheA) that coincide with other autophosphorylated signal transducers. A null mutant is defective in attractant-induced methanol formation and shows no behavioral response to chemoeffectors. These results imply that in B. subtilis the mechanism of chemotaxis involves phosphoryl transfer similar to that in E. coli. However, the CheN null mutant mostly tumbles, whereas CheA mutants swim smoothly, and only in B. subtilis does excitation lead to methyl transfer and methanol formation. Thus, the overall mechanism of chemotaxis is different in the two organisms.     

Identification and properties of type I-signal peptidases of Bacillus amyloliquefaciens.

The use of Bacillus amyloliquefaciens for enzyme production and its exceptional high protein export capacity initiated this study where the presence and function of multiple type I signal peptidase isoforms was investigated. In addition to type I signal peptidases SipS(ba) [Meijer, W.J.J., de Jong, A., Bea, G., Wisman, A., Tjalsma, H., Venema, G., Bron, S. & van Dijl, J.M. (1995) Mol. Microbiol. 17, 621-631] and SipT(ba) [Hoang, V. & Hofemeister, J. (1995) Biochim. Biophys. Acta 1269, 64-68] which were previously identified, here we present evidence for two other Sip-like genes in B. amyloliquefaciens. Same map positions as well as sequence motifs verified that these genes encode homologues of Bacillus subtilis SipV and SipW. SipU-encoding DNA was not found in B. amyloliquefaciens. SipW-encoding DNA was also found for other Bacillus strains representing different phylogenetic groups, but not for Bacillus stearothermophilus and Thermoactinomyces vulgaris. The absence of these genes, however, could have been overlooked due to sequence diversity. Sequence alignments of 23 known Sip-like proteins from Bacillus origin indicated further branching of the P-group signal peptidases into clusters represented by B. subtilis SipV, SipS-SipT-SipU and B. anthracis Sip3-Sip5 proteins, respectively. Each B. amyloliquefaciens sip(ba) gene was expressed in an Escherichia coli LepBts mutant and tested for genetic complementation of the temperature sensitive (TS) phenotype as well as pre-OmpA processing. Although SipS(ba) as well as SipT(ba) efficiently restored processing of pre-OmpA in E. coli, only SipS(ba) supported growth at TS conditions, indicating functional diversity. Changed properties of the sip(ba) gene disruption mutants, including cell autolysis, motility, sporulation, and nuclease activities, seemed to correlate with specificities and/or localization of B. amyloliquefaciens SipS, SipT and SipV isoforms.     

Isolation and characterization of Bacillus subtilis genes involved in siderophore biosynthesis: relationship between B. subtilis sfpo and Escherichia coli entD genes.

In response to iron deprivation, Bacillus subtilis secretes a catecholic siderophore, 2,3-dihydroxybenzoyl glycine, which is similar to the precursor of the Escherichia coli siderophore enterobactin. We isolated two sets of B. subtilis DNA sequences that complemented the mutations of several E. coli siderophore-deficient (ent) mutants with defective enterobactin biosynthesis enzymes. One set contained DNA sequences that complemented only an entD mutation. The second set contained DNA sequences that complemented various combinations of entB, entE, entC, and entA mutations. The two sets of DNA sequences did not appear to overlap. AB. subtilis mutant containing an insertion in the region of the entD homolog grew much more poorly in low-iron medium and with markedly different kinetics. These data indicate that (i) at least five of the siderophore biosynthesis genes of B. subtilis can function in E. coli, (ii) the genetic organization of these siderophore genes in B. subtilis is similar to that in E. coli, and (iii) the B. subtilis entD homolog is required for efficient growth in low-iron medium. The nucleotide sequence of the B. subtilis DNA contained in plasmid pENTA22, a clone expressing the B. subtilis entD homolog, revealed the presence of at least two genes. One gene was identified as sfpo, a previously reported gene involved in the production of surfactin in B. subtilis and which is highly homologous to the E. coli entD gene. We present evidence that the E. coli entD and B. subtilis sfpo genes are interchangeable and that their products are members of a new family of proteins which function in the secretion of peptide molecules.     

Expression and characterization of the genes encoding azoreductases from Bacillus subtilis and Geobacillus stearothermophilus

Azoreductases have been characterized as enzymes that can decolorize azo dyes by reducing azo groups. In this study, genes encoding proteins having homology with the azoreductase gene of Bacillus sp. OY1-2 were obtained from Bacillus subtilis ATCC6633, B. subtilis ISW1214, and Geobacillus stearotherophilus IFO13737 by polymerase chain reaction. All three genes encoded proteins with 174 amino acids. The deduced amino acid sequences of azoreductase homologs from B. subtilis ISW1214, B. subtilis ATCC6633, and G. stearotherophilus IFO13737 showed similarity of 53.3, 53.9, and 53.3% respectively to that of Bacillus sp. OY1-2.All three genes were expressed in Escherichia coli, and were characterized as having the decolorizing activity of azo dyes in a beta-NADPH dependent manner. The transformation of several azo dyes into colorless compounds by recombinant enzymes was demonstrated to have distinct substrate specificity from that of azoreductase from Bacillus sp. OY1-2.     

Cloning and analysis of the spc ribosomal protein operon of Bacillus subtilis: comparison with the spc operon of Escherichia coli.

A segment of Bacillus subtilis chromosomal DNA homologous to the Escherichia coli spc ribosomal protein operon was isolated using cloned E. coli rplE (L5) DNA as a hybridization probe. DNA sequence analysis of the B. subtilis cloned DNA indicated a high degree of conservation of spc operon ribosomal protein genes between B. subtilis and E. coli. This fragment contains DNA homologous to the promoter-proximal region of the spc operon, including coding sequences for ribosomal proteins L14, L24, L5, S14, and part of S8; the organization of B. subtilis genes in this region is identical to that found in E. coli. A region homologous to the E. coli L16, L29 and S17 genes, the last genes of the S10 operon, was located upstream from the gene for L14, the first gene in the spc operon. Although the ribosomal protein coding sequences showed 40-60% amino acid identity with E. coli sequences, we failed to find sequences which would form a structure resembling the E. coli target site for the S8 translational repressor, located near the beginning of the L5 coding region in E. coli, in this region or elsewhere in the B. subtilis spc DNA.