Suppression of a signal sequence mutation by an amino acid substitution in the mature portion of the maltose-binding protein.

An unusual spontaneous pseudorevertant of an Escherichia coli strain carrying the signal sequence point mutation malE14-1 was characterized. The suppressor mutation, malE2261, resulted in a single substitution of an aspartyl residue for a tyrosyl residue at position 283 in the sequence of the mature maltose-binding protein. The precursor retained the malE14-1 point mutation in the signal sequence. The pseudorevertant carrying both malE14-1 and malE2261 exported twice the amount of maltose-binding protein as that of the mutant carrying the malE14-1 allele alone but only 18% of the amount exported by a strain producing wild-type maltose-binding protein. A strain carrying the suppressor allele malE2261 in combination with a wild-type signal sequence exported normal quantities of maltose-binding protein to the periplasm. Mature MalE2261 had a Kd for maltose of 27 microM, compared with 3.6 microM for mature wild-type maltose-binding protein. The precursor species than contained both changes resulting from malE14-1 and malE2261 was significantly less stable in the cytoplasm than was the precursor containing only the change encoded by malE14-1.     

Escherichia coli SecB stimulates export without maintaining export competence of ribose-binding protein signal sequence mutants.

Ribose-binding protein (RBP) is exported to the periplasm of Escherichia coli via the general export pathway. An rbsB-lacZ gene fusion was constructed and used to select mutants defective in RBP export. The spontaneous Lac+ mutants isolated in this selection contained either single-amino-acid substitutions or a deletion of the RBP signal sequence. Intact rbsB genes containing eight different point mutations in the signal sequence were reconstructed, and the effects of the mutations on RBP export were examined. Most of the mutations caused severe defects in RBP export. In addition, different suppressor mutations in SecY/PrlA protein were analyzed for their effects on the export of RBP signal sequence mutants in the presence or absence of SecB. Several RBP signal sequence mutants were efficiently suppressed, but others were not suppressed. Export of an RBP signal sequence mutant in prlA mutant strains was partially dependent on SecB, which is in contrast to the SecB independence of wild-type RBP export. However, the kinetics of export of an RBP signal sequence mutant point to a rapid loss of pre-RBP export competence, which occurs in strains containing or lacking SecB. These results suggest that SecB does not stabilize the export-competent conformation of RBP and may affect translocation by stabilizing the binding of pre-RBP at the translocation site.     

Export of unprocessed precursor maltose-binding protein to the periplasm of Escherichia coli cells.

The Escherichia coli maltose-binding protein (MBP) R2 signal peptide is a truncated version of the wild-type structure that still facilitates very efficient export of MBP to the periplasm. Among single amino acid substitutions in the R2 signal peptide resulting in an export-defective precursor MBP (pMBP) were two that replaced residues in the consensus Ala-X-Ala sequence (residues -3 to -1) that immediately precedes the cleavage site. It was suggested that the functional hydrophobic core and signal peptidase recognition sequence of this signal peptide substantially overlap and that these two alterations affect both pMBP translocation and processing. In this study, the export of pMBP by the mutants, designated CC15 and CC17, with these two alterations was investigated further. The pMBP of mutant CC17 has an Arg substituted for Leu at the -2 position. It was found that CC17 cells exported only a very small amount of MBP, but that which was exported appeared to be correctly processed. This result was consistent with other studies that have concluded that virtually any amino acid can occupy the -2 position. For mutant CC15, which exhibits a fully Mal+ phenotype, an Asp is substituted for the Ala at the -3 position. CC15 cells were found to export large quantities of unprocessed, soluble pMBP to the periplasm, although such export was achieved in a relatively slow, posttranslational manner. This result was also consistent with other studies that suggested that charged residues are normally excluded from the -3 position of the cleavage site. Using in vitro oligonucleotide-directed mutagenesis, we constructed a new signal sequence mutant in which Asp was substituted for Arg at the -3 position of an otherwise wild-type MBP signal peptide. This alteration had no apparent effect on pMBP translocation across the cytoplasmic membrane, but processing by signal peptidase was inhibited. This pMBP species with its full-length hydrophobic core remained anchored to the membrane, where it could still participate in maltose uptake. The implications of these results for models of protein export are discussed.     

Protein translocation into Escherichia coli membrane vesicles is inhibited by functional synthetic signal peptides

A synthetic peptide corresponding to the signal sequence of wild type Escherichia coli lambda-receptor protein (LamB) inhibits in vitro translocation of precursors of both alkaline phosphatase and outer membrane protein A into E. coli membrane vesicles (half-maximal inhibition at 1-2 microM). By contrast, the inhibitory effect was nearly absent in a synthetic peptide corresponding to the signal sequence from a mutant strain that harbors a deletion mutation in the LamB signal region and displays an export-defective phenotype for this protein in vivo. Two peptides derived from pseudorevertant strains that arose from the deletion mutant and exported LamB in vivo were found to inhibit in vitro translocation with effectiveness that correlated with their in vivo export ability. Controls indicated that these synthetic signal peptides did not disrupt the E. coli membrane vesicles. These results can be interpreted to indicate that the presequences of exported proteins interact specifically with a receptor either in the E. coli inner membrane or in the cytoplasmic fraction. However, biophysical data for the family of signal peptides studied here reveal that they will spontaneously insert into a lipid membrane at concentrations comparable to those that cause inhibition. Hence, an indirect effect mediated by the lipid bilayer of the membrane must be considered.     

Influence of tat mutations on the ribose-binding protein translocation in Escherichia coli

Proteins are exported across the bacterial cytoplasmic membrane either as unfolded precursors via the Sec machinery or in folded conformation via the Tat system. The ribose-binding protein (RBP) of Escherichia coli is a Sec-pathway substrate. Intriguingly, it exhibits fast folding kinetics and its export is independent of SecB, a general chaperone protein dedicated for protein secretion. In this study, we found that the quantity of RBP was significantly reduced in the periplasm of tat mutants, which was restored by in trans expression of the tatABC genes. Pulse-chase experiments showed that significant amount of wild-type RBP was processed in a secY mutant in the presence of azide (SecA inhibitor), whereas the processing of a slow folding RBP derivative was almost completely blocked under the same conditions. These results would suggest that under the Sec-defective conditions the export of a portion of folded RBP could be rescued by the Tat system.     

Post-translational export of maltose-binding protein in Escherichia coli strains harboring malE signal sequence mutations and either prl+ or prl suppressor alleles

We have studied the export kinetics of the maltose-binding protein (MBP) of Escherichia coli, the malE gene product, when it is synthesized with either a wildtype signal sequence or with a mutationally altered signal sequence that affects the efficiency of secretion to the periplasm. Our results confirm a very rapid export process for the wild-type protein and, in contrast, reveal a relatively slow post-translational mode of export for the altered precursor species. For each different signal sequence mutant, a fraction of the precursor MBP pool that is proportional to the strength of the export defect appears to never exit the cytoplasm. We have also analyzed MBP export in strains harboring prl mutations that suppress malE signal sequence mutations and are thought to somehow alter the specificity of the cell's protein export machinery. The introduction of different prl alleles has no apparent effect on wild-type MBP export but increases both the amount of mutant MBP that is exported and the rate at which this is accomplished. In fact, the presence of two different prl alleles in the same strain can act synergistically in suppressing MBP export defects. The inhibition of total protein synthesis with chloramphenicol can also increase the proportion of pMBP that is post-translationally exported in these strains. A model that describes the initial steps in MBP export is presented.     

Mutations that alter the signal sequence of alkaline phosphatase in Escherichia coli.

A phoA-lacZ gene fusion was used to isolate mutants altered in the alkaline phosphatase signal sequence. This was done by selecting Lac+ mutants from a phoA-lacZ fusion strain that produces a membrane-bound hybrid protein and is unable to grow on lactose. Two such mutant derivatives were characterized. The mutations lie within the phoA portion of the fused gene and cause internalization of the hybrid protein. When the mutations were genetically recombined into an otherwise wild-type phoA gene, they interfered with export of alkaline phosphatase to the periplasm. The mutant alkaline phosphatase protein was found instead in the cytoplasm in precursor form. DNA sequence analysis demonstrated that both mutations lead to amino acid alterations in the signal sequence of alkaline phosphatase.     

A signal sequence is not required for protein export in prlA mutants of Escherichia coli.

The prlA/secY gene, which codes for an integral membrane protein component of the Escherichia coli protein export machinery, is the locus of the strongest suppressors of signal sequence mutations. We demonstrate that two exported proteins of E.coli, maltose-binding protein and alkaline phosphatase, each lacking its entire signal sequence, are exported to the periplasm in several prlA mutants. The export efficiency can be substantial; in a strain carrying the prlA4 allele, 30% of signal-sequenceless alkaline phosphatase is exported to the periplasm. Other components of the E.coli export machinery, including SecA, are required for this export. SecB is required for the export of signal-sequenceless alkaline phosphatase even though the normal export of alkaline phosphatase does not require this chaperonin. Our findings indicate that signal sequences confer speed and efficiency upon the export process, but that they are not always essential for export. Entry into the export pathway may involve components that so overlap in function that the absence of a signal sequence can be compensated for, or there may exist one or more means of entry that do not require signal sequences at all.     

A new genetic selection identifies essential residues in SecG, a component of the Escherichia coli protein export machinery.

The signal sequence of the murine serine protease inhibitor PAI-2 promotes alkaline phosphatase export to the E. coli periplasm. However, high level expression of this chimeric protein interferes with cell growth. Since most suppressors of this toxic phenotype map to secA and secY, growth arrest results from a defective interaction of the chimeric protein with the export machinery. We have characterized suppressors which map in secG, a newly defined gene of the export machinery. All single amino acid substitutions map to three adjacent codons. These secG mutants have a weak Sec phenotype, as determined by their effect on export mediated by wild-type and mutant signal sequences. Whilst a secG disruption allele also confers a weak Sec phenotype, it does not suppress the toxicity of the chimeric protein. This difference results from a selective effect of the secG suppressors on the kinetics of export mediated by the PAI-2 signal sequence. Using a malE signal sequence mutant, which has a Mal-phenotype in secG mutant strains, we have isolated extragenic Mal+ suppressors. Most suppressors map to secY, and several are allele-specific. Finally, SecG overexpression accelerates the kinetics of protein export, suggesting that there are two types of functional translocation complexes: with or without SecG.     

Efficient function of signal peptidase 1 of Escherichia coli is partly determined by residues in the mature N-terminus of exported proteins

Exported proteins require an N-terminal signal peptide to direct them from the cytoplasm to the periplasm. Once the protein has been translocated across the cytoplasmic membrane, the signal peptide is cleaved by a signal peptidase, allowing the remainder of the protein to fold into its mature state in the periplasm. Signal peptidase I (LepB) cleaves non-lipoproteins and recognises the sequence Ala-X-Ala. Amino acids present at the N-terminus of mature, exported proteins have been shown to affect the efficiency at which the protein is exported. Here we investigated a bias against aromatic amino acids at the second position in the mature protein (P2′). Maltose binding protein (MBP) was mutated to introduce aromatic amino acids (tryptophan, tyrosine and phenylalanine) at P2′. All mutants with aromatic amino acids at P2′ were exported less efficiently as indicated by a slight increase in precursor protein in vivo. Binding of LepB to peptides that encompass the MBP cleavage site were analysed using surface plasmon resonance. These studies showed peptides with an aromatic amino acid at P2′ had a slower off rate, due to a significantly higher binding affinity for LepB. These data are consistent with the accumulation of small amounts of preMBP in purified protein samples. Hence, the reason for the lack of aromatic amino acids at P2′ in E. coli is likely due to interference with efficient LepB activity. These data and previous bioinformatics strongly suggest that aromatic amino acids are not preferred at P2′ and this should be incorporated into signal peptide prediction algorithms.     

Two regions of mature periplasmic maltose-binding protein of Escherichia coli involved in secretion.

Six mutations in malE, the structural gene for the periplasmic maltose-binding protein (MBP) from Escherichia coli, prevent growth on maltose as a carbon source, as well as release of the mutant proteins by the cold osmotic-shock procedure. These mutations correspond to insertion of an oligonucleotide linker, concomitant with a deletion. One of the mutations (malE127) affects the N-terminal extension (the signal peptide), whereas the five others lie within the mature protein. As expected, the export of protein MalE127 is blocked at an early stage. This protein is neither processed to maturity nor sensitive to proteinase K in spheroplasts. In contrast, in the five other mutants, the signal peptide is cleaved and the protein is accessible to proteinase K added to spheroplasts. This indicates that the five mutant proteins are, at least in part, exported through the inner membrane. We propose that the corresponding mutations define two regions of the mature protein (between residues 18 and 42 and between residues 280 and 306), which are important for release of the protein from the inner membrane into the periplasm. We discuss the results in terms of possible conformational changes at this late step of export to the periplasm.     

The superoxide dismutase SodA is targeted to the periplasm in a SecA-dependent manner by a novel mechanism

The manganese/iron-type superoxide dismutase (SodA) of Rhizobium leguminosarum bv. viciae 3841 is exported to the periplasm of R. l. bv. viciae and Escherichia coli. However, it does not possess a hydrophobic cleaved N-terminal signal peptide typically present in soluble proteins exported by the Sec-dependent (Sec) pathway or the twin-arginine translocation (TAT) pathway. A tatC mutant of R. l. bv. viciae exported SodA to the periplasm, ruling out export of SodA as a complex with a TAT substrate as a chaperone. The export of SodA was unaffected in a secB mutant of E. coli, but its export from R. l. bv. viciae was inhibited by azide, an inhibitor of SecA ATPase activity. A temperature-sensitive secA mutant of E. coli was strongly reduced for SodA export. The 10 N-terminal amino acid residues of SodA were sufficient to target the reporter protein alkaline phosphatase to the periplasm. Our results demonstrate the export of a protein lacking a classical signal peptide to the periplasm by a SecA-dependent, but SecB-independent targeting mechanism. Export of the R. l. bv. viciae SodA to the periplasm was not limited to the genus Rhizobium, but was also observed in other proteobacteria.     

A functional decaisoleucine-containing signal sequence. Construction by cassette mutagenesis

An alkaline phosphatase signal sequence optimized for formation of a hydrophobic alpha-helix functions very efficiently in the transport process. This mutant contained a core region comprised of 9 consecutive leucine residues (Kendall, D. A., Bock, S. C., and Kaiser, E. T. (1986) Nature 321, 706-708). We have now constructed a second mutant containing a decaisoleucine core region. Isoleucine was chosen because it is an isomer of leucine with comparable hydrophobicity but in synthetic peptides isoleucine favors beta-sheet formation. Surprisingly, this mutant precursor was also processed efficiently, and mature alkaline phosphatase was correctly targeted to the Escherichia coli periplasm. Since the effective length of a beta-strand is extended relative to an alpha-helix, conformational differences should be mirrored by the relative effectiveness of shortened polyisoleucine and polyleucine core regions. However, analysis of two additional mutants containing truncated segments of either polyleucine or polyisoleucine did not reveal any differences and both accumulate as precursors. We conclude that these mutants do not adopt critically different structures. This comparative analysis was facilitated by construction of a new plasmid, CASS3. This plasmid contains unique restriction sites flanking the DNA region coding for the signal sequence hydrophobic core segment. Consequently, the wild type core-encoding region can be readily replaced with synthetic oligonucleotides coding for new structural units and multiple amino acid substitutions can be made without the need for step-wise mutagenesis.     

SecB-independent export of Escherichia coli ribose-binding protein (RBP): some comparisons with export of maltose-binding protein (MBP) and studies with RBP-MBP hybrid proteins.

The efficient export of the Escherichia coli maltose-binding protein (MBP) is known to be SecB dependent, whereas ribose-binding protein (RBP) export is SecB independent. When the MBP and RBP signal peptides were exchanged precisely at the signal peptidase processing sites, the resultant RBP-MBP and MBP-RBP hybrid proteins both were efficiently exported in SecB+ cells. However, only MBP-RBP was efficiently exported in SecB- cells; RBP-MBP exhibited a significant export defect, a finding that was consistent with previous proposals that SecB specifically interacts with the mature moiety of precursor MBP to promote export. The relatively slow, totally posttranslational export mode exhibited by certain mutant RBP and MBP-RBP species in SecB+ cells was not affected by the loss of SecB. In contrast, MBP and RBP-MBP species with similarly altered signal peptides were totally export defective in SecB- cells. Both export-defective MBP and RBP-MBP interfered with SecB-mediated protein export by depleting cells of functional SecB. In contrast, neither export-defective RBP nor MBP-RBP elicited such an interference effect. These and other data indicated that SecB is unable to interact with precursor RBP or that any interaction between these two proteins is considerably weaker than that of SecB with precursor MBP. In addition, no correlation could be established between a SecB requirement for export and PrlA-mediated suppression of signal peptide export defects. Finally, previous studies have established that wild-type MBP export can be accomplished cotranslationally, whereas wild-type RBP export is strictly a posttranslational process. In this study, cotranslational export was not detected for either MBP-RBP or RBP-MBP. This indicates that the export mode exhibited by a given precursor protein (cotranslational versus posttranslational) is determined by properties of both the signal peptide and the mature moiety.     

Diverse effects of mutations in the signal sequence on the secretion of β-lactamase in Salmonella typhimurium

Mutations in the β-lactamase structural gene that alter the signal peptide were used to study secretion into the periplasm of Salmonella typhimurium. Processing and cellular location of mutant gene products were followed by pulse-chase and cell-fractionation experiments and by trypsin accessibility in intact and lysed spheroplasts. The precursor proteins examined never appear as a free species in the periplasm. Two of the signal-sequence mutants accumulate a precursor form that is trypsin-accessible in intact spheroplasts; the precursors synthesized by the remaining mutants resemble wild-type in that they remain trypsin-inaccessible. One of the latter mutants does produce mature protein, but at a very reduced rate. It thus appears that signal-sequence mutations can affect more than one step in the secretion process, and that processing of the signal peptide is not required for the protein to be translocated (at least partially) across the inner membrane.     

Export of active green fluorescent protein to the periplasm by the twin-arginine translocase (Tat) pathway in Escherichia coli

The twin-arginine translocation (Tat) system targets cofactor-containing proteins across the Escherichia coli cytoplasmic membrane via distinct signal peptides bearing a twin-arginine motif. In this study, we have analysed the mechanism and capabilities of the E. coli Tat system using green fluorescent protein (GFP) fused to the twin-arginine signal peptide of TMAO reductase (TorA). Fractionation studies and fluorescence measurements demonstrate that GFP is exported to the periplasm where it is fully active. Export is almost totally blocked in tat deletion mutants, indicating that the observed export in wild-type cells occurs predominantly, if not exclusively, by the Tat pathway. Imaging studies reveal a halo of fluorescence in wild-type cells corresponding to the exported periplasmic form; the GFP is distributed uniformly throughout the cytoplasm in a tat mutant. Because previous work has shown GFP to be incapable of folding in the periplasm, we propose that GFP is exported in a fully folded, active state. These data also show for the first time that heterologous proteins can be exported in an active form by the Tat pathway.     

A novel method for increasing production of mature proteins in the periplasm of Escherichia coli.

A novel strategy to obtain high-level production of mature proteins exported to the periplasm of Escherichia coli is described. It is based on a modified signal sequence generated by insertion of a coding sequence of the polypeptide precursor of interest at the BamHI site of the commercial vector pQE-30 resulting in an addition of a dodeca-peptide (MRGSH6GS) at the N-terminus of the precursor. The modification does not affect correct processing of the modified signal nor proper folding of the target protein, resulting in an untagged native product. The method is simple for avoiding onerous optimization of translation initiation and screening of host stains. The usefulness of this method is illustrated by overexpression of DsbC and DsbA. Induced by 0.01 mM IPTG at 37 degrees C, proteins were overproduced to comprise 20-30% of the total cellular proteins, and more than 95% of the expressed proteins were correctly processed and exported into the periplasm with yields of more than 100 mg per liter culture.     

Molecular cloning and characterization of genes required for ribose transport and utilization in Escherichia coli K-12.

We isolated spontaneous and transposon insertion mutants of Escherichia coli K-12 that were specifically defective in utilization or in high-affinity transport of D-ribose (or in both). Cotransduction studies located all of the mutations near ilv, at the same position as previously identified mutations causing defects in ribokinase ( rbsK ) or ribose transport ( rbsP ). Plasmids that complemented the rbs mutations were isolated from the collection of ColE1 hybrid plasmids constructed by Clarke and Carbon. Analysis of those plasmids as well as of fragments cloned into pBR322 and pACYC184 allowed definition of the rbs region. Products of rbs genes were identified by examination of the proteins produced in minicells containing various rbs plasmids. We identified four rbs genes: rbsB , which codes for the 29-kilodalton ribose-binding protein; rbsK , which codes for the 34-kilodalton ribokinase ; rbsA , which codes for a 50-kilodalton protein required for high-affinity transport; and rbsC , which codes for a 27-kilodalton protein likely to be a transport system component. Our studies showed that these genes are transcribed from a common promoter in the order rbsA rbsC rbsB rbsK . It appears that the high-affinity transport system for ribose consists of the three components, ribose-binding protein, the 50-kilodalton RbsA protein, and the 27-kilodalton RbsC protein, although a fourth, unidentified component could exist. Mutants defective in this transport system, but normal for ribokinase , are able to grow normally on high concentrations of the sugar, indicating that there is at least a second, low-affinity transport system for ribose in E. coli K-12.