A conservative amino acid substitution, arginine for lysine, abolishes export of a hybrid protein in Escherichia coli. Implications for the mechanism of protein secretion

A hybrid protein that comprises the beta-lactamase signal peptide fused precisely to chicken muscle triosephosphate isomerase is not secreted into the periplasm of Escherichia coli. The protein can be secreted, however, if an arginine residue at position 3 of the isomerase is replaced by either a serine or a proline residue. In contrast, replacement of a neighboring lysine residue has no effect on secretion of the protein. Furthermore, if the arginine is removed from position 3 to generate a secreted protein, but is then reintroduced in place of the neighboring lysine, the blockade to secretion is re-established. The singular effect of the arginine residue on secretion does not result from the role this residue plays in the formation or stabilization of the native isomerase structure: mutational alterations remote from the N terminus of the isomerase that prevent the proper folding of the protein do not relieve the block to secretion. The finding that an arginine residue prevents secretion while a lysine residue does not, suggests that basic residues near the mature N terminus of a secreted protein must be deprotonated if orderly export is to occur. This implies that the signal peptide along with the N-terminal portion of the mature protein partitions directly into the lipid bilayer in the course of the secretory process.     

Modulation of the effects of mutations in the basic region of the OmpA signal peptide by the mature portion of the protein

Oligonucleotide-directed site-specific mutagenesis was used to study the structure-function relationship of the positively charged amino terminus of the Escherichia coli outer membrane protein OmpA signal peptide. Mutations were isolated which reduced the overall charge of the amino-terminal region from +2 (wild type) to +1, 0, and -1, as well as one mutation from Thr to Ser at position 4. DNA encoding the wild type and mutant OmpA signal peptides was then fused in-frame to DNA encoding the mature regions of Staphylococcus aureus nuclease A and TEM beta-lactamase. In the case of both the beta-lactamase and nuclease fusions, normal processing was no longer observed when the charge at the amino terminus was reduced to zero or made negative. Differences between the two hybrid proteins were observed in the case of the Thr to Ser mutation. As expected, this mutation had no effect on the beta-lactamase hybrid; however, the processing rate of the nuclease hybrid protein was reduced to nearly one-half. Furthermore, this effect was essentially reversed when a Lys residue at position 3 was deleted. A model is presented which explains the differing effects of a signal peptide mutation on the secretion of different hybrid proteins based on kinetic differences in the translocation of the nuclease and beta-lactamase proteins.     

Defective Escherichia coli signal peptides function in yeast

To investigate structural characteristics important for eukaryotic signal peptide function in vivo, a hybrid gene with interchangeable signal peptides was cloned into yeast. The hybrid gene encoded nine residues from the amino terminus of the major Escherichia coli lipoprotein, attached to the amino terminus of the entire mature E. coli beta-lactamase sequence. To this sequence were attached sequences encoding the nonmutant E. coli lipoprotein signal peptide, or lipoprotein signal peptide mutants lacking an amino-terminal cationic charge, with shortened hydrophobic core, with altered potential helicity, or with an altered signal-peptide cleavage site. These signal-peptide mutants exhibited altered processing and secretion in E. coli. Using the GAL10 promoter, production of all hybrid proteins was induced to constitute 4-5% of the total yeast protein. Hybrid proteins with mutant signal peptides that show altered processing and secretion in E. coli, were processed and translocated to a similar degree as the non-mutant hybrid protein in yeast (approximately 36% of the total hybrid protein). Both non-mutant and mutant signal peptides appeared to be removed at the same unique site between cysteine 21 and serine 22, one residue from the E. coli signal peptidase II processing site. The mature lipo-beta-lactamase was translocated across the cytoplasmic membrane into the yeast periplasm. Thus the protein secretion apparatus in yeast recognizes the lipoprotein signal sequence in vivo but displays a specificity towards altered signal sequences which differs from that of E. coli.     

An alternate pathway for the processing of the prolipoprotein signal peptide in Escherichia coli

Previous studies showed that when the signal sequence plus 9 amino acid residues from the amino terminus of the major lipoprotein of Escherichia coli was fused to beta-lactamase, the resulting hybrid protein was modified, proteolytically processed, and assembled into the outer membrane as was the wild-type lipoprotein (Ghrayeb, J., and Inouye, M. (1983) J. Biol. Chem. 259, 463-467). We have constructed several hybrid proteins with mutations at the cleavage site of the prolipoprotein signal peptide. These mutations are known to block the lipid modification of the lipoprotein at the cysteine residue, resulting in the accumulation of unprocessed, unmodified prolipoprotein in the outer membrane. The mutations blocked the lipid modification of the hybrid protein. However, in contrast to the mutant lipoproteins, the cleavage of the signal peptides for the mutant hybrid proteins did occur, although less efficiently than the unaltered prolipo-beta-lactamase. The mutant prolipo-beta-lactamase proteins were cleaved at a site 5 amino acid residues downstream of the prolipoprotein signal peptide cleavage site. This new cleavage between alanine and lysine residues was resistant to globomycin, a specific inhibitor for signal peptidase II. This indicates that signal peptidase II, the signal peptidase which cleaves the unaltered prolipo-beta-lactamase, is not responsible for the new cleavage. The results demonstrate that the cleavage of the signal peptide is a flexible process that can occur by an alternative pathway when the normal processing pathway is blocked.     

Site-directed mutagenesis of the hydrogenase signal peptide consensus box prevents export of a beta-lactamase fusion protein.

A secretion vector, pVN1, expressing the [NiFe] hydrogenase signal peptide of Desulfovibrio vulgaris Hildenborough fused to beta-lactamase from Escherichia coli was constructed in order to study the unusual characteristics of hydrogenase signal peptides, which share a strictly conserved sequence, the consensus box: R-R-X-F-X-K. Although the hydrogenase signal peptide-beta-lactamase fusion protein was processed much more slowly than the fusion of beta-lactamase with its own signal peptide, the system mimicked several features expected for hydrogenase biosynthesis in E. coli, including increased export under anaerobic conditions. Site-directed mutagenesis of R(-28), the first arginine residue of the consensus box, to a glutamate completely inhibited export and processing of the fusion protein. The same mutation of R(-33), located outside the consensus box, had almost no effect. The data indicate a specific role for the consensus box sequence in the export mechanism for hydrogenase.     

Modification, processing, and subcellular localization in Escherichia coli of the pCloDF13-encoded bacteriocin release protein fused to the mature portion of beta-lactamase.

A fusion between the pCloDF13-derived bacteriocin release protein and beta-lactamase was constructed to investigate the subcellular localization and posttranslational modification of the bacteriocin release protein in Escherichia coli. The signal sequence and 25 of the 28 amino acid residues of the mature bacteriocin release protein were fused to the mature portion of beta-lactamase. The hybrid protein (Mr, 31,588) was expressed in minicells and whole cells and possessed full beta-lactamase activity. Immunoblotting of subcellular fractions revealed that the hybrid protein is present in both the cytoplasmic and outer membranes of E. coli. Radioactive labeling experiments in the presence or absence of globomycin showed that the hybrid protein is modified with a diglyceride and fatty acids and is processed by signal peptidase II, as is the murein lipoprotein. The results indicated that the pCloDF13-encoded bacteriocin release protein is a lipoprotein which is associated with both membranes of E. coli cells.     

The net charge of the first 18 residues of the mature sequence affects protein translocation across the cytoplasmic membrane of gram-negative bacteria

This statistical study shows that in proteins of gram-negative bacteria exported by the Sec-dependent pathway, the first 14 to 18 residues of the mature sequences have the highest deviation between the observed and expected net charge distributions. Moreover, almost all sequences have either neutral or negative net charge in this region. This rule is restricted to gram-negative bacteria, since neither eukaryotic nor gram-positive bacterial exported proteins have this charge bias. Subsequent experiments performed with a series of Escherichia coli alkaline phosphatase mutants confirmed that this charge bias is associated with protein translocation across the cytoplasmic membrane. Two consecutive basic residues inhibit translocation effectively when placed within the first 14 residues of the mature protein but not when placed in positions 19 and 20. The sensitivity to arginine partially reappeared again 30 residues away from the signal sequence. These data provide new insight into the mechanism of protein export in gram-negative bacteria and lead to practical recommendations for successful secretion of hybrid proteins.     

Membrane topology of penicillin-binding protein 3 of Escherichia coli

The beta-lactamase fusion vector, pJBS633, has been used to analyse the organization of penicillin-binding protein 3 (PBP3) in the cytoplasmic membrane of Escherichia coli. The fusion junctions in 84 in-frame fusions of the coding region of mature TEM beta-lactamase to random positions within the PBP3 gene were determined. Fusions of beta-lactamase to 61 different positions in PBP3 were obtained. Fusions to positions within the first 31 residues of PBP3 resulted in enzymatically active fusion proteins which could not protect single cells of E. coli from killing by ampicillin, indicating that the beta-lactamase moieties of these fusion proteins were not translocated to the periplasm. However, all fusions that contained greater than or equal to 36 residues of PBP3 provided single cells of E. coli with substantial levels of resistance to ampicillin, indicating that the beta-lactamase moieties of these fusion proteins were translocated to the periplasm. PBP3 therefore appeared to have a simple membrane topology with residues 36 to the carboxy-terminus exposed on the periplasmic side of the cytoplasmic membrane. This topology was confirmed by showing that PBP3 was protected from proteolytic digestion at the cytoplasmic side of the inner membrane but was completely digested by proteolytic attack from the periplasmic side. PBP3 was only inserted in the cytoplasmic membrane at its amino terminus since replacement of its putative lipoprotein signal peptide with a normal signal peptide resulted in a water-soluble, periplasmic form of the enzyme. The periplasmic form of PBP3 retained its penicillin-binding activity and appeared to be truly water-soluble since it fractionated, in the absence of detergents, with the expected molecular weight on Sephadex G-100 and was not retarded by hydrophobic interaction chromatography on Phenyl-Superose.     

Extracellular secretion of pectate lyase by the Erwinia chrysanthemi out pathway is dependent upon Sec-mediated export across the inner membrane.

The plant pathogenic enterobacterium Erwinia chrysanthemi EC16 secretes several extracellular, plant cell wall-degrading enzymes, including pectate lyase isozyme PelE. Secretion kinetics of 35S-labeled PelE indicated that the precursor of PelE was rapidly processed by the removal of the amino-terminal signal peptide and that the resulting mature PelE remained cell bound for less than 60 s before being secreted to the bacterial medium. PelE-PhoA (alkaline phosphatase) hybrid proteins generated in vivo by TnphoA insertions were mostly localized in the periplasm of E. chrysanthemi, and one hybrid protein was observed to be associated with the outer membrane of E. chrysanthemi in an out gene-dependent manner. A gene fusion resulting in the substitution of the beta-lactamase signal peptide for the first six amino acids of the PelE signal peptide did not prevent processing or secretion of PelE in E. chrysanthemi. When pelE was overexpressed, mature PelE protein accumulated in the periplasm rather than the cytoplasm in cells of E. chrysanthemi and Escherichia coli MC4100 (pCPP2006), which harbors a functional cluster of E. chrysanthemi out genes. Removal of the signal peptide from pre-PelE was SecA dependent in E. coli MM52 even in the presence of the out gene cluster. These data indicate that the extracellular secretion of pectic enzymes by E. chrysanthemi is an extension of the Sec-dependent pathway for general export of proteins across the bacterial inner membrane.     

Effect of OmpA signal peptide mutations on OmpA secretion, synthesis, and assembly.

In previous investigations, we have examined the effect of OmpA signal peptide mutations on the secretion of the two heterologous proteins TEM beta-lactamase and nuclease A. During these studies, we observed that a given signal peptide mutation could affect differentially the processing of precursor OmpA-nuclease or precursor OmpA-lactamase. This observation led us to further investigate the influence of the mature region of a precursor protein on protein export. Preexisting OmpA signal peptide mutations of known secretion phenotype when directing heterologous protein export (nuclease A or beta-lactamase) were fused to the homologous mature OmpA protein. Four signal peptide mutations that have previously been shown to prevent export of nuclease A and beta-lactamase were found to support OmpA protein export, albeit at reduced rates. This remarkable retention of export activity by severely defective precursor OmpA signal peptide mutants may be due to the ability of mature OmpA to interact with the cytoplasmic membrane. In addition, these same signal peptide mutations can affect the level of OmpA synthesis as well as its proper assembly in the outer membrane of Escherichia coli. Two signal peptide mutations dramatically stimulate the rate of precursor OmpA synthesis three- to fivefold above the level observed when a wild-type signal peptide is directing export. The complete removal of the OmpA signal peptide does not result in increased OmpA synthesis. This finding suggests that the signal peptide mutations function positively to stimulate OmpA synthesis, rather than bypass a down-regulatory mechanism effected by a wild-type signal peptide. Overproduction of wild-type precursor OmpA or precursors containing signal peptide mutations which lead to relatively minor kinetic processing defects results in accumulation of an improperly assembled OmpA species (imp-OmpA). In contrast, signal peptide mutations which cause relatively severe processing defects accumulate no or only small quantities of imp-OmpA. All mutations result in equivalent levels of properly assembled OmpA. Thus, a strong correlation between imp-OmpA accumulation and cell toxicity was observed. A mutation in the mature region of OmpA which prevents the proper outer membrane assembly of OmpA was suppressed when export was directed by a severely defective signal peptide. These findings suggest that signal peptide mutations indirectly influence OmpA assembly in the outer membrane by altering both the level and rate of OmpA secretion across the cytoplasmic membrane.     

Altered -3 substrate specificity of Escherichia coli signal peptidase 1 mutants as revealed by screening a combinatorial peptide library

Signal peptidase functions to cleave signal peptides from preproteins at the cell membrane. It has a substrate specificity for small uncharged residues at -1 (P1) and aliphatic residues at the -3 (P3) position. Previously, we have reported that certain alterations of the Ile-144 and Ile-86 residues in Escherichia coli signal peptidase I (SPase) can change the specificity such that signal peptidase is able to cleave pro-OmpA nuclease A in vitro after phenylalanine or asparagine residues at the -1 position (Karla, A., Lively, M. O., Paetzel, M. and Dalbey, R. (2005) J. Biol. Chem. 280, 6731-6741). In this study, screening of a fluorescence resonance energy transfer-based peptide library revealed that the I144A, I144C, and I144C/I86T SPase mutants have a more relaxed substrate specificity at the -3 position, in comparison to the wild-type SPase. The double mutant tolerated arginine, glutamine, and tyrosine residues at the -3 position of the substrate. The altered specificity of the I144C/I86T mutant was confirmed by in vivo processing of pre-beta-lactamase containing non-canonical arginine and glutamine residues at the -3 position. This work establishes Ile-144 and Ile-86 as key P3 substrate specificity determinants for signal peptidase I and demonstrates the power of the fluorescence resonance energy transfer-based peptide library approach in defining the substrate specificity of proteases.     

Intragenic suppressor mutations that restore export of maltose binding protein with a truncated signal peptide

A deletion mutation, malE delta 12-18, removes seven residues from the hydrophobic core of the maltose binding protein (MBP) signal peptide and thus prevents secretion of this protein to the periplasm of E. coli. Intragenic suppressor mutations of malE delta 12-18 have been obtained, some highly efficient in their ability to restore proper MBP export. Twelve independently isolated suppressors represent six unique mutational events. Five result in alterations within the MBP signal peptide; one changes the amino acid at residue 19 of the mature MBP. Analysis of these suppressors indicates that the length of the hydrophobic core is a major determinant of signal peptide function. The experiments further suggest that the hydrophobic core region serves primarily a structural role in mediating protein secretion, and that other sequences outside of this region may be responsible for providing the initial recognition of the MBP nascent chain as a secreted protein.     

Secretion of the Escherichia coli K-12 SheA hemolysin is independent of its cytolytic activity

The Escherichia coli K-12 sheA gene encodes a pore-forming hemolysin that is secreted to the medium by a hitherto unidentified mechanism. To study SheA secretion, we constructed fusions between SheA and the mature form of the periplasmic enzyme beta-lactamase, and performed site-directed mutagenesis on these constructs. The SheA-Bla and Bla-SheA hybrid proteins displayed hemolytic activity and were efficiently exported to the extracellular medium. Our results with mutant hybrid proteins show that secretion of SheA is independent of its cytolytic activity, that secretion is paralleled by a transient leakage of periplasmic contents to the extracellular medium, and that deletion of the 11 C-terminal residues of SheA has no effect on its secretion and cytolytic activity.     

Introduction of basic amino acid residues after the signal peptide inhibits protein translocation across the cytoplasmic membrane of Escherichia coli. Relation to the orientation of membrane proteins

The introduction of positive charges at the amino terminus of the mature domain of secretory proteins resulted in strong inhibition of their translocation across the cytoplasmic membrane of Escherichia coli, both in vitro and in vivo. The model secretory proteins used were OmpF-Lpp chimeric proteins possessing a cleavable or uncleavable signal peptide, beta-lactamase (Bla) and Bla-Lpp chimeric proteins. It is suggested that positively charged residues preceding the hydrophobic domain of the signal peptide have a positive effect, and ones following the hydrophobic domain, a negative effect on the translocation. These findings are discussed in relation to the orientation of membrane proteins, of which positive charges are predominant on the cytoplasmic surface.     

Secretion of beta-lactamase requires the carboxy end of the protein

Synthesis and secretion of beta-lactamase were studied in Salmonella typhimurium infected with P22 phage carrying the structural gene for beta-lactamase (the bla gene) in mutant or wild-type form. The wild-type gene was shown to specify two forms of beta-lactamase which differ in molecular weight by about 2500 daltons. This difference is consistent with removal, predicted on other grounds, of 23 amino-terminal residues (the "signal" sequence). All bla- mutants, including chain-terminating mutants lacking as much as 50% or as little as 10% of the protein, were apparently unaffected in this processing step. Pulse-chase experiments showed that more than 85% of the wild-type (as well as mutant) proteins are synthesized as complete overlength precursors before being processed to their mature forms. Virtually all the mature wild-type protein appears in the periplasmic space whereas a large fraction of the precursor appears in the cytoplasm. In contrast, both the precursor and processed forms of beta-lactamase proteins synthesized by chain-terminating mutants (including one which lacks only 10% of its residues from the carboxy end) are not secreted and apparently remain soluble in the cytoplasm. These results show that the carboxy-terminal amino acid sequence (at least) of beta-lactamase is essential to successful transport across the cytoplasmic membrane, and suggest that the presence (and probably also the act of removal) of the signal sequence does not suffice to ensure secretion.     

Expression in Escherichia coli of the psbO gene encoding the 33 kd protein of the oxygen-evolving complex from spinach.

The cDNA for the 33 kd protein from the oxygen-evolving complex of spinach together with the coding region for the hydrophobic C-terminal part of the transit sequence was cloned into the expression plasmid pDS12/33Ex. The 33 kd protein precursor was expressed in Escherichia coli, secreted into the periplasm and correctly processed to the mature 33 kd protein. Thus the hydrophobic domain of the transit sequence, preceded by a methionine and two lysine residues, can function as a bacterial signal peptide. The periplasmic proteins were released from the cells by osmotic shock and the expressed protein was purified by anion exchange chromatography. The protein was identified by SDS-PAGE and Western blotting. N-terminal sequence analysis showed that the cleavage of the signal peptide occurred at the correct position. The expressed protein could be rebound to CaCl2-washed PSII particles and oxygen evolution was restored in equal amounts by the 33 kd protein from both E. coli and spinach.     

Enhancement of protein translocation across the membrane by specific mutations in the hydrophobic region of the signal peptide.

The hydrophobic region of the signal peptide of the OmpA protein of the Escherichia coli outer membrane was extensively altered in its hydrophobicity and predicted secondary structure by site-specific mutagenesis. The mutated signal peptides were fused to nuclease A from Staphylococcus aureus, and the function of the signal peptide was examined by measuring the rate of processing of the signal peptide. Six of the 12 mutated signal peptides in the nuclease hybrid were processed faster than the wild-type. In particular, the processing of the mutated signal peptide in which the alanine residue at position 9 was substituted with a valine residue was enhanced almost twofold over the processing of the wild-type signal peptide. In addition, the production of nuclease A fused with this mutated signal peptide also increased twofold. However, these effects were not observed when the mutated signal peptide was fused to TEM beta-lactamase. Analysis of the present mutations suggests that both overall hydrophobicity and distinct structural requirements in the hydrophobic region have important roles in signal peptide function.     

Clustering of glycine and NG,NG-dimethylarginine in nucleolar protein C23

Protein C23 (Mr 110 000, pI = 5.5), a major phosphoprotein in the nucleolus of mammalian cells, has been shown to contain 1.3 mol% of NG,NG-dimethylarginine (DMA) [Lischwe, M.A., Roberts, K.D., Yeoman, L.C., & Busch, H. (1982) J. Biol. Chem. 257, 14600-14602]. A tryptic peptide from protein C23 that contains DMA has been isolated and sequenced. Its sequence is Gly-Glu-Gly-Gly-Phe-Gly-Gly-DMA-Gly-Gly-Gly-DMA-Gly-Gly-Phe-Gly-Gly-DMA- Gly-Gly- Gly-DMA-Gly-Gly-DMA-Gly-Gly-Phe-Gly-Gly-DMA-Gly-DMA-Gly-Gly-Phe-Gly-Gly- DMA-Gly-Gly-Phe-DMA-Gly-Gly-DMA-Gly-Gly-Gly-Gly-Asp-Phe-Lys. This peptide contains 34 glycine, 10 DMA, and 6 phenylalanine residues and has clusters of glycine and NG,NG-dimethylarginine interspersed with phenylalanine residues. A similar domain has been found at the amino terminus of a nucleolar protein of Mr 34,000, pI = 8.5. This sequence array may represent a conserved domain characteristic of a certain class of nuclear proteins. All of the methylated arginine residues in protein C23, the 34-kilodalton protein, and myelin basic protein [Carnegie, P.R. (1971) Biochem. J. 123, 57-67] have at least one adjacent glycine. Access of certain arginine methylases to arginine residues may be sterically possible because of the lack of a side chain on the adjacent glycine residue(s).     

Mutational alterations affecting the export competence of a truncated but fully functional maltose-binding protein signal peptide.

The wild-type maltose-binding protein (MBP) signal peptide is 26 amino acids in length. A mutationally altered MBP signal peptide has been previously described that is missing one of the basic residues from the hydrophilic segment and seven residues from the hydrophobic core; however, it still facilitates MBP secretion to the periplasm at a rate and efficiency comparable to those of the wild-type structure. Thus, this truncated signal peptide (designated the R2 signal peptide) must retain all of the essential features required for proper export function. In this study, alterations were obtained in the R2 signal peptide that resulted in an export-defective MBP. For the first time, signal sequence mutations were obtained that resulted in the synthesis of a totally export-defective MBP. As was previously the case for the wild-type signal peptide, the introduction of either charged residues or helix-breaking proline residues adversely affected export function. Despite these similarities, the position of these alterations within the R2 signal peptide, their relative effects on MBP secretion and processing, and an analysis of the ability of various extragenic prl mutations to suppress the secretion defects provide additional insight into the minimal requirements for a functional MBP signal peptide.     

A positive residue in the hydrophobic core of the Escherichia coli lipoprotein signal peptide suppresses the secretion defect caused by an acidic amino terminus.

The signal peptide of secretory proteins requires a basic amino terminus followed by a stretch of hydrophobic residues to effect efficient translocation of precursor proteins. Replacement of the positively charged amino-terminal residues of prolipoprotein by acidic amino acids decreased the rate of precursor translocation (Inouye, S., Soberon, X., Franceschini, T., Nakamura, K., Itakura, K., and Inouye, M. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 3438-3441; Vlasuk, G. P., Inouye, S., Ito, H., Itakura, K., and Inouye, M. (1983) J. Biol. Chem. 258, 7141-7148). We demonstrate here that an arginine residue, but not an aspartate, when localized at position 9 of the hydrophobic region of the lipoprotein signal peptide, is able to suppress intramolecularly the processing defect caused by an acidic amino terminus. Furthermore, when present at position 14 of the signal peptide, this positive residue, but not aspartate, was able to support efficient translocation of unmodified prolipoprotein. This demonstrates that a positive residue can restore the function of a severely defective signal peptide and need not be localized at the amino terminus to do so. Both aspartate and arginine substitution at position 14 of the lipoprotein signal peptide stimulated prolipoprotein synthesis. This effect was position-specific, did not require precursor translocation, and was dominant to the inhibition of synthesis caused by an acidic amino terminus.