A novel complete reconstitution system for membrane integration of the simplest membrane protein

A complete reconstitution system for membrane integration of the simplest protein was developed by means of defined factors. A mutant version of Pf3 coat protein, 3L-Pf3 coat, requires neither signal recognition particle/Sec factors nor a membrane potential for its integration into the cytoplasmic membrane of Escherichia coli. Although 3L-Pf3 coat is spontaneously integrated into liposomes composed of phospholipids, diacylglycerol completely blocks such spontaneous integrations at a physiological level. Under the conditions where spontaneous integration does not occur, 3L-Pf3 coat integration was absolutely dependent on a novel integration-stimulating factor. Combination of the PURE system, an in vitro translation system composed of the purified factors involved in translation in E. coli, with liposomes containing the highly purified integration-stimulating factor revealed multiple cycles of 3L-Pf3 coat integration, achieving the complete reconstitution of membrane integration. Based on the function of the factor, we propose that the factor is named MPIase (Membrane Protein Integrase).     

Conditional lethal mutations separate the M13 procoat and Pf3 coat functions of YidC: different YIDC structural requirements for membrane protein insertion

Conditional lethal YidC mutants have been isolated to decipher the role of YidC in the assembly of Sec-dependent and Sec-independent membrane proteins. We now show that the membrane insertion of the Sec-independent M13 procoat-lep protein is inhibited in a short time in a temperature-sensitive mutant when shifted to the nonpermissive temperature. This provides an additional line of evidence that YidC plays a direct role in the insertion of the Sec-independent M13 procoat protein. However, in the temperature-sensitive mutant, the insertion of the Sec-independent Pf3 phage coat protein and the Sec-dependent leader peptidase were not strongly inhibited at the restricted temperatures. Conversely, using a cold-sensitive YidC strain, we find that the membrane insertion of the Sec-independent Pf3 coat protein is blocked, and the Sec-dependent leader peptidase is inhibited at the nonpermissive temperature, whereas the insertion of the M13 procoat protein is nearly normal. These data show that the YidC function for procoat and its function for Pf3 coat and possibly leader peptidase are genetically separable and suggest that the YidC structural requirements are different for the Sec-independent M13 procoat and Pf3 coat phage proteins that insert by different mechanisms.     

Features of transmembrane segments that promote the lateral release from the translocase into the lipid phase

Topogenic sequences direct the membrane topology of proteins by being recognized and decoded by integral membrane translocases. In this paper, we have compared the minimal sequence characteristics of helical-hairpin, reverse signal-anchor, and stop-transfer sequences in bacterial membrane proteins that use either the YidC or SecYEG translocases for membrane insertion. We find that a stretch composed of 3 leucines and 16 alanines is required for efficient membrane-anchoring of the M13 procoat protein that inserts by a helical hairpin mechanism, and that a stretch composed of only 19 alanines has a detectable membrane-anchoring ability. Similar results were obtained for the reverse signal-anchor sequence of the single-spanning Pf3 coat protein and for stop-transfer segments engineered into leader peptidase. We have also determined the contribution to the apparent free energy of membrane insertion of M13 procoat for all 20 amino acids. The relative order of the contributions is similar to that determined for a stop-transfer sequence in the mammalian endoplasmic reticulum, but the absolute difference between the contributions for the most hydrophobic and most hydrophilic residues is somewhat larger in the E. coli system. These results are significant because they define the features of a membrane protein transmembrane segment that induce lateral release from the YidC and Sec translocases into the lipid bilayer in bacteria.     

Initial steps in protein membrane insertion. Bacteriophage M13 procoat protein binds to the membrane surface by electrostatic interaction.

Bacteriophage M13 procoat protein is synthesized on free polysomes prior to its assembly into the inner membrane of Escherichia coli. As an initial step of the membrane insertion pathway, the precursor protein interacts with the cytoplasmic face of the inner membrane. We have used oligonucleotide-directed mutagenesis to study the regions of the procoat protein involved in membrane binding. We find that there is an absolute requirement for positively charged amino acids at both ends of the protein. Replacing these with negatively charged residues resulted in an accumulation of the precursor in the cytoplasm. We propose that the positively charged amino acids are directly involved in membrane binding, possibly directly to the negatively charged phospholipid head groups. This was tested in vitro with artificial liposomes. Whereas wild-type procoat interacted with these liposomes, we found that procoat mutants with negatively charged amino acids at both ends did not bind. Therefore, we conclude that newly synthesized M13 procoat protein binds electrostatically to the negatively charged inner membrane of E. coli.     

Reconstitution of rapid and asymmetric assembly of M13 procoat protein into liposomes which have bacterial leader peptidase

The leader peptidase of Escherichia coli cleaves a 23-residue leader sequence from M13 procoat to yield mature coat protein in virus-infected cells. We have reconstituted pure leader peptidase into vesicles of E. coli lipids and found that these liposomes are active in the conversion of procoat to coat. Trypsin removes all but 10% of the leader peptidase, yet the vesicles retain nearly full capacity to convert procoat to coat, suggesting that only procoat which inserts across the liposomal membrane is a substrate for leader peptidase. This is confirmed by the finding that over 70% of the coat protein produced by these liposomes spans the membrane. The rate at which leader peptidase inside protease-treated liposomes cleaves externally added procoat is comparable to the rate of procoat cleavage by the same amount of leader peptidase in detergent micelles. Thus, procoat can rapidly integrate across a liposomal membrane and be cleaved to coat protein. These findings confirm the central part of the membrane trigger hypothesis that certain proteins (such as procoat) can cross a bilayer without the aid of a proteinaceous pore or transport system.     

M13 procoat inserts into liposomes in the absence of other membrane proteins

Procoat, the precursor form of the major coat protein of coliphage M13, assembles into the Escherichia coli inner membrane and is cleaved to mature coat protein by leader peptidase. This assembly process has previously been reconstituted using lipids and purified leader peptidase in a cell-free protein synthesis reaction (Watts, C., Silver, P., and Wickner, W. (1981) Cell 25, 347-353; Ohno-Iwashita, Y., and Wickner, W. (1983) J. Biol. Chem. 258, 1895-1900). We now report that procoat can also cross a liposomal membrane composed of only purified phospholipids; leader peptidase is not needed to catalyze insertion. When procoat is synthesized in vitro in the presence of liposomes with encapsulated chymotrypsin, the procoat inserts spontaneously through the membrane and is degraded. The protease was shown by several criteria to be in the lumen of the liposomes. These results demonstrate that the precursor form of an E. coli integral membrane protein can cross a membrane without the aid of leader peptidase or any other membrane proteins.     

A derivative of lipid A is involved in signal recognition particle/SecYEG-dependent and -independent membrane integrations

A cell-free system was developed that allows the correct integration of single and multispanning membrane proteins of Escherichia coli into proteoliposomes. We found that physiological levels of diacylglycerol were required to prevent spontaneous integration into liposomes even of the polytopic mannitol permease. Using diacylglycerol-containing proteoliposomes, we identified a novel integration-stimulating factor. Integration of mannitol permease was dependent on both the SecYEG translocon and this factor and was mediated by signal recognition particle and signal recognition particle receptor. Integration of M13 procoat, which is independent of both signal recognition particle/signal recognition particle receptor and SecYEG, was also promoted by this factor. Furthermore, the factor stimulated the post-translational translocation of presecretory proteins, suggesting that it also mediates integration of a signal sequence. This factor was found to be a lipid A-derived membrane component possessing a peptide moiety.     

YidC is strictly required for membrane insertion of subunits a and c of the F(1)F(0)ATP synthase and SecE of the SecYEG translocase

YidC was previously discovered to play a critical role for the insertion of the Sec-independent M13 procoat and Pf3 coat phage proteins into the Escherichia coli inner membrane. To determine whether there is an absolute requirement of YidC for membrane protein insertion of any endogenous E. coli proteins, we investigated a few representative membrane proteins. We found that membrane subunits of the F(0) sector of the F(1)F(0)ATP synthase and the SecE protein of the SecYEG translocase are highly dependent on YidC for membrane insertion, based on protease mapping and immunoblot analysis. We found that the SecE dependency on YidC for membrane insertion does not contradict the observation that depletion of YidC does not block SecYEG-dependent protein export at 37 degrees C. YidC depletion does not decrease the SecE level low enough to block export at 37 degrees C. In contrast, we found that protein export of OmpA is severely blocked at 25 degrees C when YidC is depleted, which may be due to the decreased SecE level, as a 50% decrease in the SecE levels drastically affects protein export at the cold temperature [Schatz, P. J., Bieker, K. L., Ottemann, K. M., Silhavy, T. J., and Beckwith, J. (1991) EMBO J. 10, 1749-57]. These studies reported here establish that physiological substrates of YidC include subunits of the ATP synthase and the SecYEG translocase, demonstrating that YidC plays a vital role for insertion of endogenous membrane proteins in bacteria.     

Isolation of mutants in M13 coat protein that affect its synthesis, processing, and assembly into phage

The major coat protein (gene 8 protein) of bacteriophage M13 has been studied intensively as a model of membrane assembly, protein packing, and protein-DNA interactions. Because this protein is essential for assembly of the phage, very few mutants have been isolated. We have therefore cloned the gene 8 into a plasmid under control of the araB promoter. In the presence of arabinose, the cloned gene is expressed at a rate comparable to that in an M13-infected cell. Plasmid-derived procoat is inserted across the plasma membrane and processed to coat at a normal rate. The coat can support plaque formation by a defective M13 virus (M13am8) with an amber mutation in its procoat gene. This complementation assay was used to screen the mutagenized, cloned gene 8 for mutants which fail to make fully functional coat. Mutants were obtained which fail to synthesize procoat, which do not convert procoat to mature coat protein, or in which the coat protein is incapable of assembling into infectious virions.     

Processing of preproteins by liposomes bearing leader peptidase

Procoat, the precursor form of M13 coat protein, assembles into sealed liposomes bearing only internally oriented leader peptidase and is processed to yield transmembrane coat protein [Ohno-Iwashita, Y., & Wickner, W. (1983) J. Biol. Chem. 258, 1895-1900]. The precursors of maltose-binding protein and of outer membrane protein A (OmpA) are also processed by these liposomes, showing that these preproteins can at least partially insert across a lipid bilayer. The ability to insert into a bilayer may be a general property of preproteins. The cleavage products, mature OmpA and maltose-binding protein, are not sequestered within the liposomes, suggesting that an additional factor(s) is (are) required for complete translocation. Liposomes were also prepared with leader peptidase in a more physiological, membrane-spanning orientation. These liposomes were also active in the cleavage of externally added procoat, pro-OmpA, and pre maltose-binding protein, though the mature OmpA and maltose-binding protein were still not sequestered within the liposomes. Pretreatment of these liposomes with trypsin cleaved near the amino terminus of the leader peptidase, inactivating the enzyme. The function of this amino-terminal domain, on the opposite side of the membrane from the catalytic domain, is unknown.     

Cysteine residues in the transmembrane regions of M13 procoat protein suggest that oligomeric coat proteins assemble onto phage progeny

The M13 phage assembles in the inner membrane of Escherichia coli. During maturation, about 2,700 copies of the major coat protein move from the membrane onto a single-stranded phage DNA molecule that extrudes out of the cell. The major coat protein is synthesized as a precursor, termed procoat protein, and inserts into the membrane via a Sec-independent pathway. It is processed by a leader peptidase from its leader (signal) peptide before it is assembled onto the phage DNA. The transmembrane regions of the procoat protein play an important role in all these processes. Using cysteine mutants with mutations in the transmembrane regions of the procoat and coat proteins, we investigated which of the residues are involved in multimer formation, interaction with the leader peptidase, and formation of M13 progeny particles. We found that most single cysteine residues do not interfere with the membrane insertion, processing, and assembly of the phage. Treatment of the cells with copper phenanthroline showed that the cysteine residues were readily engaged in dimer and multimer formation. This suggests that the coat proteins assemble into multimers before they proceed onto the nascent phage particles. In addition, we found that when a cysteine is located in the leader peptide at the -6 position, processing of the mutant procoat protein and of other exported proteins is affected. This inhibition of the leader peptidase results in death of the cell and shows that there are distinct amino acid residues in the M13 procoat protein involved at specific steps of the phage assembly process.     

YidC-driven membrane insertion of single fluorescent Pf3 coat proteins

The membrane insertion of single bacteriophage Pf3 coat proteins was observed by confocal fluorescence microscopy. Within seconds after addition of the purified and fluorescently labeled protein to liposomes or proteoliposomes containing the purified and reconstituted membrane insertase YidC of Escherichia coli, the translocation of the labeled residue was detected. The 50-amino-acid-long Pf3 coat protein was labeled with Atto520 and inserted into the proteoliposomes. Translocation of the dye into the proteoliposome was revealed by quenching the fluorescence outside of the vesicles. This allowed us to distinguish single Pf3 coat proteins that only bound to the surface of the liposomes from proteins that had inserted into the bilayer and translocated the dye into the lumen. The Pf3 coat protein required the presence of the YidC membrane insertase, whereas mutants that have a membrane-spanning region with an increased hydrophobicity were autonomously inserted into the liposomes without YidC.     

Recombinant forms of M13 procoat with an OmpA leader sequence or a large carboxy-terminal extension retain their independence of secY function.

The assembly of phage M13 procoat protein into the plasma membrane of Escherichia coli is independent of the secY protein. To test whether this is caused by the unusually small size of procoat, we fused DNA encoding 103 amino acids to the carboxy-terminal end of the procoat gene. The resulting fusion protein, which attains the same membrane-spanning conformation as mature coat protein, still does not require the secY function for membrane assembly. To determine whether the leader sequence governs interaction with the secY protein, we genetically exchanged the leader peptides between procoat and pro-OmpA, a protein which does require secY for its membrane assembly. Each of the resulting hybrid proteins assembles across the plasma membrane, though at a reduced rate. Membrane assembly of the fusion of procoat leader and OmpA required secY function, whereas assembly of the pro-OmpA leader/coat protein fusion was independent of secY. Properties of the entire procoat molecule, rather than its small size or a specific property of its leader peptide determines its mode of membrane assembly.     

The purification of M13 procoat, a membrane protein precursor.

Many membrane proteins and most secreted proteins are initially made as precursors with an N-terminal leader sequence. We now report the isolation of M13 procoat, the precursor of the membrane-bound form of M13 coat protein. There are 40 000 copies of M13 procoat protein/cell during M13 amber 7 virus infection. Purified procoat is quantitatively cleaved by isolated leader peptidase to yield mature-length coat protein. Rabbit antibodies to M13 procoat will precipitate procoat but not coat, suggesting that the antibody molecules are specifically recognizing the leader sequence or the conformation which it induces in the whole procoat molecule.     

Polyoma large T antigen regulates the integration of viral DNA sequences into the genome of transformed cells

The major coat protein of coliphage M13 is an integral protein of the E. coli plasma membrane prior to its assembly into new virus particles. It is generated from its precursor, procoat, by a membrane-bound leader peptidase. We now describe the reconstitution of a highly purified preparation of this enzyme into vesicles of E. coli phospholipids. These vesicles bind procoat made in vitro and procoat isolated from in vitro synthesis. Both the crude and the purified substrates were converted posttranslationally to coat protein. A significant proportion of the coat protein becomes inserted into the vesicle bilayer, with the N terminus facing the vesicle interior and the C terminus exposed to the external medium. These results strongly suggest that highly purified leader peptidase from E. coli and phospholipids are the only components necessary to mediate the binding, processing and insertion of this integral membrane protein.     

Conserved residues of the leader peptide are essential for cleavage by leader peptidase

Gene 8 of bacteriophage M13 codes for procoat, the precursor of its major coat protein. Gene 8 has been cloned into a plasmid and mutagenized. We have isolated mutants of this gene in which procoat is synthesized but is not processed to coat protein. We now describe mutants in the leader region of procoat, at positions -6, -3, and -1 with respect to the leader peptidase cleavage site. These positions are quite conserved among the leader peptides of various pre-proteins. Each of these mutant procoats is synthesized at a normal rate and inserts correctly into the plasma membrane, as judged by its accessibility to protease in intact spheroplasts. Procoat accumulates, largely in its transmembrane form, and is not cleaved to coat. In detergent extracts, the mutant procoats are very poor substrates for added leader peptidase. We conclude that these 3 residues are not conserved for insertion across the membrane but are part of an essential recognition site for the leader peptidase.     

Escherichia coli YidC is a membrane insertase for Sec-independent proteins

YidC is a recently discovered bacterial membrane protein that is related to the mitochondrial Oxa1p and the Alb3 protein of chloroplasts. These proteins are required in the membrane integration process of newly synthesized proteins that do not require the classical Sec machinery. Here we demonstrate that YidC is sufficient for the membrane integration of a Sec-independent protein. Microgram amounts of the purified single-spanning Pf3 coat protein were efficiently inserted into proteoliposomes containing the purified YidC. A mutant Pf3 coat protein with an extended hydrophobic region was inserted independently of YidC into the membrane both in vivo and in vitro, but its insertion was accelerated by YidC. These results show that YidC can function separately from the Sec translocase to integrate membrane proteins into the lipid bilayer.     

Membrane assembly: Posttranslational insertion of M13 procoat protein into E. coli membranes and its proteolytic conversion to coat protein in vitro

The major coat protein (gene 8 product) of bacteriophage M13 is an integral membrane protein during infection of host cells. It is synthesized as a larger precursor (procoat) with a leader sequence of 23 amino acids at its amino terminus. In vivo studies have shown that procoat only inserts into the host-cell plasma membrane after its synthesis is completed. We now demonstrate that procoat can post-translationally insert into inverted cytoplasmic membrane vesicles from E. coli and can be processed proteolytically to yield coat protein. Procoat changes from an assembly-competent substrate to an incompetent (denatured) form within minutes after its synthesis; much of the procoat that accumulates during an hour of in vitro synthesis is therefore denatured. These studies emphasize the importance of stringent criteria for the demonstration of obligate cotranslational assembly.     

Assembly of M13 and M13am8H1R1 procoat protein into microsomes is stimulated by rabbit reticulocyte lysate and ATP

Processing of M13 procoat protein to transmembrane coat protein by dog pancreas microsomes is stimulated by a component of rabbit reticulocyte lysate and ATP. We asked whether this ATP-dependent reaction, involved in membrane assembly of procoat protein in the eukaryotic system, is related to the membrane potential dependent reaction observed for the membrane assembly of procoat protein in E. coli. Specifically, we asked if a mutant procoat protein which had been previously shown to be independent of the membrane potential with respect to its assembly in E. coli (M13am8H1R1 procoat protein) shows a stimulation by reticulocyte lysate and ATP in its assembly into microsomes. Since the mutant procoat protein behaved exactly as the wild type procoat protein in the eukaryotic in vitro system, we propose that the ATP-dependent reaction observed for the eukaryotic system does not substitute for the membrane potential dependent reaction in the prokaryotic system.     

Distinct requirements for translocation of the N-tail and C-tail of the Escherichia coli inner membrane protein CyoA

Inner membrane proteins (IMPs) of Escherichia coli use different pathways for membrane targeting and integration. YidC plays an essential but poorly defined role in the integration and folding of IMPs both in conjunction with the Sec translocon and as a Sec-independent insertase. Depletion of YidC only marginally affects the insertion of Sec-dependent IMPs, whereas it blocks the insertion of a subset of Sec-independent IMPs. Substrates of this latter "YidC-only" pathway include the relatively small IMPs M13 procoat, Pf3 coat protein, and subunit c of the F(1)F(0) ATPase. Recently, it has been shown that the steady state level of the larger and more complex CyoA subunit of the cytochrome o oxidase is also severely affected upon depletion of YidC. In the present study we have analyzed the biogenesis of the integral lipoprotein CyoA. Collectively, our data suggest that the first transmembrane segment of CyoA rather than the signal sequence recruits the signal recognition particle for membrane targeting. Membrane integration and assembly appear to occur in two distinct sequential steps. YidC is sufficient to catalyze insertion of the N-terminal domain consisting of the signal sequence, transmembrane segment 1, and the small periplasmic domain in between. Translocation of the large C-terminal periplasmic domain requires the Sec translocon and SecA, suggesting that for this particular IMP the Sec translocon might operate downstream of YidC.