SEARCH

SEARCH BY CITATION

Keywords:

  • leader peptidase;
  • LepB;
  • signal peptide;
  • Tat;
  • twin-arginine

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

In bacteria, the export of proteins by the twin-arginine translocase (Tat) pathway is directed by cleavable N-terminal signal peptides. We studied the relationship between transport and maturation using a substrate, YedY, that contains an Ala > Leu substitution at the -1 position of the signal peptide. This blocks maturation and leads to the accumulation of a membrane-bound precursor form with the mature domain exposed to the periplasm. Its accumulation does not block transport of other Tat substrates, indicating that exit from the translocation channel has taken place, and the precursor protein is fir mLy integrated into the membrane bilayer. The membrane-integrated nature of the precursor, and complete absence of precursor protein in the periplasm, strongly suggest that the precursor has undergone lateral transfer into the bilayer during translocation. We propose that subsequent proteolytic processing releases the mature protein into the periplasm. A delay in processing results in an inhibition of cell growth, emphasizing a requirement for efficient maturation of Tat substrates.


Abbreviations
GFP

green fluorescent protein

IPTG

isopropyl thio-β-d-galactoside

Tat

twin-arginine translocase

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

In bacteria, the majority of extracytoplasmic proteins are transported across the plasma membrane by either the Sec- or Tat-dependent protein transport pathways (reviewed in Refs [1-3]). These proteins are almost invariably synthesized with a transient N-terminal extension sequence, termed a signal peptide, which is removed by a specific processing peptidase. The maturation of Sec-dependent periplasmic proteins has been extensively studied and the characteristics of signal peptides and their cognate signal peptidases have been established in some detail. Sequence analysis has indicated that typical bacterial signal peptides consist of three domains: a positively charged N-terminal region, a hydrophobic central core domain and a more polar C-terminal region ending with short-chain amino acids at the -3 and -1 positions, relative to the cleavage site [4-6]. The -3 and -1 residues form an important part of the recognition site for the specific peptidase, usually termed leader (or signal) peptidase, that carries out the maturation step.

Signal peptidases have also been studied in some detail. In Escherichia coli, leader peptidase is a membrane-bound protein with the active site on the periplasmic side of the plasma membrane. It is essential for viability and carries out a highly specific removal of signal peptides followoing translocation (reviewed in Ref. [4]). The available evidence indicates that the primary signal peptidase in E. coli, LepB, is responsible for the maturation of both Sec and Tat substrates [7].

The Sec pathway carries out an essential protein export role in bacteria, but most bacteria also contain a second mainstream protein export pathway known as the twin-arginine translocation (Tat) pathway. This has the unique ability to export folded proteins across the tightly sealed, energy-transducing bacterial plasma membrane and it also operates in the chloroplast thylakoid membrane (reviewed in Refs [1-3]). In E. coli, the key Tat components are TatA, TatB and TatC, which appear to form the minimally functional Tat tranlocase. Current models for the Tat translocation pathway propose that the folded substrate protein is targeted to Tat translocation pathway by association between the signal peptide and a TatBC receptor complex (reviewed in Refs [1-3]). This triggers the generation of a translocation channel of TatA protomers to form the full translocation system, although the precise details of the translocation event are poorly understood.

Tat substrates are synthesized with N-terminal signal peptides that contain a key twin-arginine motif [8, 9], but the signal peptides otherwise resemble those of Sec substrates to a remarkable degree – perhaps surprisingly, given that the two types of protein translocase have such different structures and mechanisms. As with Sec signals, Tat signal peptides contain a hydrophobic core domain, with the twin-arginine motif found just prior to this domain, often within a consensus S–R–R–x–F–L–K motif between the N- and H-domains [8]. Tat signal peptides also end with short-chain -3 and -1 residues, typically Ala–Xaa–Ala, and these have been shown to be essential for efficient maturation in thylakoids [10].

Although there are clear similarities between Sec and Tat signal peptides, and they are clearly removed by the same peptidase, it is not clear how Tat signal peptides are presented to the signal peptidease during the overall export process. The vast majority of Tat substrates are soluble periplasmic proteins, and an important question is whether these substrates are transported into the periplasm and then processed to the mature size, or whether they are processed before they leave the membrane environment. A recent study [11] showed that the presence of a ‘Flag’ tag in a Tat signal peptide caused the protein to become embedded in the plasma membrane as unprocessed precursor. This raises the possibility that the protein had become stuck in the membrane en route to the periplasm. In this study, we examined the export and processing of a Tat substrate using a more subtle alteration in the signal peptide, and we used a series of assays to probe the location of the protein. The results suggest that this Tat precursor protein exits the translocase laterally into the plane of the membrane bilayer before undergoing maturation.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Substitution at the -1 position of the signal peptide blocks the maturation of YedY

YedY is a 32 kDa periplasmic molybdenum-containing protein which is predicted to be exported across the membrane by the Tat pathway in E. coli [12, 13]. It is synthesized as a 37 kDa precursor protein with a twin-arginine signal peptide, shown in Fig. 1, that contains features typical of a Tat signal peptide. These include a twin-arginine motif in the N-terminal domain, a hydrophobic core region (underlined) and a more polar C-terminal domain ending with an Ala–Xaa–Ala consensus motif (in this case Ala-His-Ala). YedY was chosen for this study because preliminary studies (not shown) indicated that it was exported relatively rapidly by the Tat pathway. We first set out to confirm that YedY is a bona fide Tat substrate because, although there is experimental evidence that YedY does bear a Tat signal peptide, there is also evidence that the signal peptide can interact with the Sec system [13].

image

Figure 1. Structure of the YedY signal peptide and YedY–A44L variant. Primary structure of the 44-residue signal peptide of E. coli YedY. The twin-arginine motif is shown in larger font, the hydrophobic domain is underlined and the cleavage site is indicated. The terminal residue, Ala44, was substituted by Leu to generate the YedY–A44L variant.

Download figure to PowerPoint

The precursor protein was expressed in wild-type MC4100 cells and a tat null mutant strain, ΔtatABCDE, using the arabinose-inducible pBAD24 plasmid. The cells were then fractionated into cytoplasm, periplasm and membrane fractions (C, P, M), as shown in Fig. 2A. When expressed in wild-type cells, two YedY forms of 37 and 32 kDa are detected by immunoblotting of total cell extracts (Fig. 2A); some nonspecific binding is also observed. Fractionation of the cells shows that the 37 kDa form is in the membrane fraction (M), whereas the smaller form is in the periplasm (P). Very little protein is detected in the cytoplasm (C). In the tat mutant strain, only the larger protein is visible in the membrane fraction (M), and no YedY is detected in the periplasm. These data clearly indicate that the 37 kDa precursor form has been exported to the periplasm and processed to the mature size (32 kDa) in wild-type cells. The complete absence of export in tat mutant cells confirms that pre-YedY is transported exclusively by the Tat pathway.

image

Figure 2. YedY–A44L remains anchored to the membrane after transport by the Tat pathway, or degraded to yield a periplasmic fragment. (A) YedY is a Tat substrate. Pre-YedY was expressed from the pBAD24 plasmid in wild-type MC4100 cells (WT) or in a tat null mutant strain, ΔtatABCDE. Expression was induced for 3 h with arabinose, after which total cell contents were analysed (TC) or the cells were fractionated to yield cytoplasm, membrane and periplasm samples (C, M, P). Samples were immunoblotted using antibodies to the C-terminal 6-His tag on pre-YedY. Pre, precursor; Mat, mature protein. Mobilities of molecular mass markers (in kDa) are shown on the left. (B) YedY–A44L is membrane-bound when expressed from pBAD24, or degraded to yield a periplasmic degradation product when expressed using pEXT22. YedY and YedY–A44L were expressed in wild-type cells using the pBAD24 plasmid (left) or pEXT22 (right). After induction with arabinose/IPTG for 3 h, cells were fractionated to generate cytoplasmic, membrane and periplasmic samples as in (A). DP, degradation product; other symbols as in (A).

Download figure to PowerPoint

In order to study the relationship between translocation and processing, the -1 Ala in the YedY signal peptide (Ala44) was substituted by leucine and the mutated protein (YedY–A44L) was expressed in E. coli. Previous studies on signal peptides have indicated that short-chain residues at the -1 and -3 positions are essential for the efficient maturation of the precursor protein, and substitutions at the -1 position, in particular, can block maturation [5, 10]. YedY–A44L was expressed using both the arabinose-inducible pBAD24 vector and the isopropyl thio-β-d-galactoside (IPTG)-inducible pEXT22 vector. A notable difference is that expression from pBAD24 is tightly regulated (hardly any expression is observed in the absence of arabinose) [14], whereas the pEXT22 plasmid is relatively leaky (induction with IPTG enhances expression approximately twofold over a fairly high basal expression level; data not shown). In each case, the YedY and mutated form were synthesized with a C-terminal 6-His tag to aid the identification of the protein.

Figure 2B shows a typical experiment in which synthesis was induced in wild-type MC4100 cells using the two different vectors. After 3 h of induction, cells were fractionated to generate cytoplasm, membrane and periplasm samples (C, M, P) which were analysed by SDS/PAGE and immunoblotting. The results show that when nonmutated pre-YedY is expressed using either vector, it is exported to the periplasm and processed to the mature size (Mat; also indicated with asterisk), with some pre-YedY found in the membrane fraction (Pre). When synthesized using the pBAD24 plasmid, YedY–A44L is not processed to any significant degree and the protein is found almost exclusively in the membrane fraction as the precursor form. This demonstrates that the substitution of the -1 Ala completely blocks the maturation of YedY.

Interestingly, somewhat different results are obtained when the mutated protein is expressed from the pEXT22 plasmid. Here, the protein is found mostly in the periplasm, but as a degradation product that is slightly smaller than the mature-size protein. There is no evidence of mature-size YedY, which suggests that maturation is again blocked and another, unidentified protease has cleaved the precursor protein to release the degradation product. It is likely that this result reflects the different properties of the plasmids; YedY–A44L is expressed only for a short period from the pBAD24 plasmid, whereas expression from pEXT22 is continuous throughout the bacterial growth period, including several hours before induction commences. We reasoned that this would provide more time for proteases to cleave the unprocessed precursor.

Unprocessed YedY–A44L is membrane-bound with the mature protein exposed to the periplasm

The presence of the YedY–A44L degradation product in the periplasmic fraction (when expressed from pEXT22) suggests that proteolysis of a membrane-bound precursor has resulted in the release of a soluble fragment into the periplasm. This would imply that the mature domain has been efficiently transported to the periplasmic side of the membrane. To test this possibility more directly, we prepared spheroplasts from cells expressing YedY–A44L from pBAD24 and subjected them to proteolysis using proteinase K. The aim was to determine whether the mature domain of the membrane-bound precursor is indeed located on the periplasmic side of the membrane, where it should be accessible to proteinase K. The left-hand panel of Fig. 3A shows data obtained using spheroplasts that were prepared from MC4100 cells expressing YedY–A44L. The precursor form is clearly detected on the immunoblot, together with a faster-migrating band that may result from nonspecific binding. These were analysed without further treatment (Sp), or after incubation with buffer (Buf) or proteinase K (PK). As a control, spheroplasts were treated with proteinase K and subjected to freezing–warming cycles to lyse them and allow access of the protease to cell interior. The data show that the YedY–A44L is fully degraded by proteinase K treatment of spheroplasts, with or without freeze–warming, strongly indicating that the mature protein has been successfully transported to the periplasmic side of the membrane.

image

Figure 3. The mature domain of YedY–A44L is exposed to the periplasm after Tat-dependent export to the periplasm. (A) Left: YedY–A44L was expressed from pBAD24 in wild-type E. coli cells (WT) for 3 h after induction with arabinose. Cells were harvested and YedY–A44L was analysed in total cell contents (TC) or spheroplasts (SP). In the remaining lanes, spheroplasts were treated with buffer (Buf) or proteinase K (PK). Lane PK/T: spheroplasts were subjected to freeze–warming cycles to lyse them before the proteinase K treatment. Right: Nonmutated pre-YedY was expressed in ∆tat null mutant cells and samples were treated in the same manner as the YedY–A44L-expressing cells. ‘Pre’ denotes precursor protein (either YedY–A44L or pre-YedY). (B) YedY–A44L was expressed using the pEXT22 plasmid in wild-type cells or ΔtatABCDE cells (Δtat). After expression, samples of total cells, membrane and periplasm (TC, M, P) were analysed by immunoblotting as in Fig. 2. DP, degradation product, Pre, precursor protein. Molecular mass markers (in kDa) are shown on the left.

Download figure to PowerPoint

As a control, we analysed ∆tatABCDE cells expressing nonmutated pre-YedY in exactly the same manner. In these tat null mutant cells, the pre-YedY would be located only in the cell interior where it should be resistant to proteolysis unless the spheroplasts are freeze–warmed to break them. The data (right) show that pre-YedY is resistant to proteinase K treatment of the spheroplasts but is degraded when the spheroplasts were broken, confirming that the pre-YedY in ∆tatABCDE is on the cytoplasmic side of the membrane. Thus, YedY–A44L is exported to the periplasmic side of the membrane in wild-type cells, where it remains bound to the membrane.

To confirm that YedY–A44L is exported exclusively by the Tat pathway, we expressed it from pEXT22 (as in Fig. 2) in both wild-type cells and ∆tatABCDE cells. Figure 3B shows that in wild-type cells, the protein again appears as a degradation product in the periplasm, whereas in ∆tatABCDE cells this periplasmic product is not observed at all, even with a highly exposed blot such as that shown. Clearly, transport of YedY–A44L is Tat-dependent.

The signal peptide of YedY–A44L is fir mLy embedded in the plasma membrane

YedY–A44L fails to undergo maturation and is found almost entirely in the membrane fraction, which means that the uncleaved signal peptide is attached to the membrane. We first addressed the question of how deeply/fir mLy the signal peptide is actually embedded in the membrane. We set out to determine whether it is fully integrated into the membrane or only peripherally bound, using methods that are used to distinguish between integral and peripheral membrane proteins. The latter are often defined as proteins that can be removed by the treatment with carbonate solutions at high pH [15]. Spheroplasts containing YedY–A44L were treated with sodium carbonate, pH 11.5 or, as a control, Tris/HCl, pH 7.5 (see 'Unprocessed YedY–A44L is membrane-bound with the mature protein exposed to the periplasm'). We also analysed a single-span membrane protein as a control test and chose TatAd for this purpose. TatAd is a single-span component of the Bacillus subtilis Tat system which has been shown to be partly localized in the cytoplasm [16, 17]. It is thus apparently less hydrophobic than many transmembrane proteins, and might be expected to be partially extracted by carbonate treatment. The data (Fig. 4) show that YedY–A44L is in fact very fir mLy integrated into the plasma membrane. The protein is not extracted by treatment of spheroplasts with Na2CO3, and instead remains almost exclusively in the membrane fraction (M) with very little protein in the supernatant (Sup). In the control experiment, TatAd was expressed in E. coli cells and the data show that most of the TatAd remains in the membrane but a sizeable proportion is released by Na2CO3 and found in the supernatant. These data therefore demonstrate that the signal peptide of YedY–A44L is integrated within the membrane, rather than bound loosely to the periphery of the bilayer.

image

Figure 4. The uncleaved signal peptide of YedY–A44L is integrated into the membrane. (Left) YedY–A44L was expressed in wild-type cells for 3 h from pBAD24. Membranes were prepared from spheroplasts and incubated with alkaline carbonate or Tris buffer as detailed in Materials and methods. Samples of total cells (TC), spheroplasts (SP) and the supernatant (sup) or membranes (M) from spheroplast extraction were analysed. (Right) Bacillus subtilis TatAd was synthesized in wild-type E. coli cells and membranes were prepared and analysed by carbonate or Tris washing as for YedY–A44L samples.

Download figure to PowerPoint

Membrane-integration of YedY–A44L occurs at a late stage of the overall translocation process, or after its completion

Because YedY–A44L is fir mLy integrated into the membrane, we considered it possible that the precursor protein had become stuck in the Tat translocase during the translocation process. To date, there have been no indications that proteolytic maturation occurs during the translocation event, but this event, and the relationship between translocation and maturation, have not been studied in any detail. To obtain a clearer picture of this issue we used two approaches. In the first, we simultaneously expressed YedY–A44L from pBAD24 and pre-YedY from pEXT22, and determined whether the former blocked export of the latter. Synthesis of YedY–A44L was induced using arabinose for 3 h, after which synthesis of pre-YedY was induced with IPTG for a further 2.5 h. Although the pEXT22 vector is leaky, as described above, we reasoned that prolonged expression of YedY–A44L from pBAD24 would nevertheless result in a block in pre-YedY export if the mutated form does indeed become stuck in the translocation channel. The results are shown in Fig. 5A. Some precursor-size protein is present in the membrane fraction (M) and this is probably mostly YedY–A44L, perhaps with some unprocessed pre-YedY that has not yet been exported (it is not possible to distinguish between the mutated and nonmutated pre-YedY in this immunoblot). In addition, mature-size YedY is present in the periplasm, which suggests that the export process has not been badly affected by the prolonged synthesis of YedY–A44L.

image

Figure 5. Unprocessed YedY–A44L does not block the Tat translocase. (A) Cells expressed YedY–A44L from pBAD24 and pre-YedY from pEXT22. YedY–A44L synthesis was induced for 3 h using pBAD24 in wild-type cells, after which cells were washed, resuspended in IPTG-containing medium and grown for a further 3 h. Total cell contents, membrane, cytoplasm and periplasm samples were blotted for the presence of YedY + YedY–A44L. Precursor and mature forms of YedY are indicated, together with the degradation product shown in Fig. 2B. (B, C) Wild-type cells expressed YedY–A44L from pEXT22 for 3 h in the presence of IPTG, after which the cells were washed and resuspended in arabinose-containing medium to induce synthesis of TorA–GFP from pBAD24. After a further 2 h, cells were fractionated and cytoplasm, membrane and periplasm fractions were immunoblotted with antibodies to the His tag on YedY (B) or to GFP (C). ‘Pre’ denotes the YedY or TorA–GFP precursor protein, Mat denotes mature GFP protein and mobilities of molecular mass markers (in kDa) are indicated.

Download figure to PowerPoint

An alternative approach is shown in Fig. 5B,C. Here, the YedY–A44L was synthesized for 3 h using the pEXT22 plasmid and we only then induced expression of a chimeric Tat substrate, TorA–green fluorescent protein (GFP), using pBAD24. This would give ample time for the Tat translocon to become blocked by YedY–A44L. The TorA–GFP substrate comprises the signal peptide of TorA linked to GFP, and this protein has been shown to be exported by the Tat pathway [18]. Figure 5B shows the blots for YedY–A44L targeting, and the data resemble those shown above in Fig. 2: the YedY–A44L precursor protein is found in the membrane and no mature protein can be detected. The same fractions were blotted for TorA–GFP (Fig. 5C) and the data show that mature-size GFP is present in the periplasmic fraction. This shows that extended prior expression of YedY–A44L does not block the export of TorA–GFP, and we conclude that the translocon is not saturated with unprocessed YedY–A44L. The available evidence suggests that the protein has completed interaction with the translocon and has moved to a different location, although it remains possible that there is some residual interaction that does not block transport of additional substrate molecules.

When considered together with the above data on the membrane-bound nature of the unprocessed YedY, the results raise the strong possibility that the precursor protein exits the translocon laterally into the membrane bilayer, and then remains in the membrane until processing is carried out. To further probe this possibility we tested whether the precursor protein can be detected, even at low levels, during a time course analysis of expression (induced by arabinose) and export. Figure 6 shows the results of this study, in which membrane and periplasmic samples were analysed over a total induction period of 5 h. With nonmutated pre-YedY, the precursor protein is evident in the membrane fraction after 2 h and the mature protein can be detected in the periplasm by 3 h. YedY–A44L, by contrast, also appears in the membrane after 2 h, and is at maximal levels after 3 h, but is absent at later time points. We assume that this reflects degradation of the mutated protein and clearance from the membrane. With both mutated and nonmutated pre-YedY, a faster-migrating band is evident in the membrane fraction; this may result from proteolytic clipping before export or may stem from nonspecific antibody binding. Importantly, no YedY–A44L is detected in the periplasm at any time point, and these results are consistent with the idea that the protein first enters the membrane before cleavage to the mature size.

image

Figure 6. Mature-size YedY, but not the precursor form, can be detected in the periplasm. Pre-YedY and YedY–A44L were expressed in wild-type cells from pBAD24 and cells were analysed at times (in h) after induction with arabinose. The cells were fractionated and membrane and periplasm samples were analysed. The immunoblots show the presence of precursor forms (Pre) in the membrane fractions and mature-size YedY (Mat) in the periplasm.

Download figure to PowerPoint

Pre-YedY maturation is essential for efficient growth of E. coli cells

The above data show that a Tat substrate can be translocated in the absence of processing, and the overall translocation/maturation pathway is discussed in more detail below. However, additional studies were carried out to assess the overall importance of Tat substrate maturation, after we observed clear effects on cell growth rates in some experiments. We expressed pre-YedY and YedY–A44L from pBAD24 in wild-type cells and measured growth rates before and after induction with arabinose. Analysis of YedY–A44L levels by immunoblotting (Fig. 7A) shows that the uncleavable YedY–A44L protein appears around the 2 h time point, after which it is subjected to degradation by the 4 and 5 h time points (as observed in Fig. 6), with a variety of degradation products apparent. The culture growth rates are shown in Fig. 7B. Cells expressing nonmutated pre-YedY exhibited a typical growth curve as shown in Fig. 7A, with a characteristic lag phase (0–1 h), exponential phase (1–6 h) and stationary phase (6–10 h). However, expression of YedY–A44L yielded different growth curves in which exponential growth suddenly slows dramatically around the 2 h time point. This coincides with the appearance of the YedY–A44L protein as shown in Fig. 7. Growth returns to normal by about the 5 h point, consistent with the disappearance of YedY–A44L. This suggests that the presence of the uncleavable precursor protein is highly detrimental to cell growth.

image

Figure 7. Expression of YedY–A44L inhibits culture growth. (A) Synthesis of YedY–A44L and pre-YedY was induced using pBAD24 in wild-type cells, and samples of cells were analysed at times (in h) after induction. The blots show the presence of precursor and mature forms of YedY and the YedY–A44L protein. (B) Growth rates of the pre-YedY (solid line) and YedY–A44L (dashed line) cultures were monitored by taking D600 readings at the indicated times points. The graph shows an average of three experiments.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

In bacteria, most secretory proteins are synthesized with a signal peptide that initiates export. In E. coli, it appears that a single specific enzyme, leader peptidase (LepB) is responsible for the maturation of most exported precursor proteins. In our study, the -1 Ala residue of the YedY signal peptide was substituted by Leu and the data show that this leads to an essentially complete block in maturation. The result is an accumulation of membrane-bound precursor protein when YedY–A44L is synthesized using the pBAD24 plasmid, or the appearance of a degradation product in the periplasm when pEXT22 is used for expression studies. It is not clear why the mutated precursor is degraded to yield a specific degradation product when synthesized using pEXT22, yet not with pBAD24, but this may reflect slightly differing physiological states when the cells are induced using IPTG in one case or arabinose in the other. The pEXT22-expressed protein is also synthesized over a longer period.

The rapid induction kinetics of the pBAD24 system offer the opportunity to study the maturation of Tat precursors in the context of the overall translocation process, and several points emerge from this study. First, the data indicate that, in the absence of maturation, the mature domain of the precursor protein is transported across the inner membrane, where it is susceptible to natural or exogenously added proteases. Candidate natural proteases include GlpG, DegS, DegP and also others [4]. This is clearly indicated by the appearance of a periplasmic degradation product after synthesis using the pEXT22 plasmid, and by the degradation of the membrane-bound precursor protein when proteinase K is added to spheroplasts after synthesis using the pBAD24 plasmid.

Next, there is evidence that the uncleaved precursor is in a transmembrane state. The YedY–A44L protein is never detected in the periplasm, and is fully resistant to extraction by alkaline carbonate. These results represent strong evidence that the signal peptide is integrated into the membrane bilayer, rather than associating peripherally. Tat signal peptides are known to be membrane-interactive, in the sense that they have a natural affinity for membrane bilayers, but this appears to be reversible because Tat precursor proteins are also found in the bacterial cytoplasm and chloroplast stroma [18-21]. A study on thylakoids showed the presence of a membrane-inserted version of a chimeric Tat precursor protein [22], and this appeared to be a fully inserted early translocation intermediate. However, few studies have addressed the nature of the substrate during the events occuring between initial insertion into the translocase and emergence on the trans side of the membrane. A further study on thylakoids involving mutation of the Tat signal peptide [23] found that unprocessed forms associated with high molecular mass complexes in the membrane, and the authors concluded that Tat substrates can be laterally released from the complexes prior to processing. In bacteria, insertion of a FLAG peptide epitope resulted in an unprocessed form with the mature protein exposed to the periplasm [11]. Our data add further detail on the transport process because: (a) the signal peptide is relatively ‘natural’, containing only a single Ala > Leu mutation at the C-terminus; and (b) we present evidence that the precursor protein cannot be observed in the periplasm, or extracted from the membrane by harsh chemical treatment. Our data also show that the Tat translocase does not become jammed with uncleaved precursor molecules, and they must therefore have exited the translocation channel.

The exclusively membrane-integrated state of the uncleaved YedY precursor protein has important implications for the nature of the translocase itself. The actual structure and nature of the translocation channel is still a matter of debate, but these results show that it is apparently able to allow lateral exit of a near-natural Tat substrate. The results point to an overall export pathway that is summarized in Fig. 8. Export by the Tat pathway is known to occur posttranslationally, and step 1 represents the synthesis of the precursor protein with the mature domain in a folded form. In step 2, the signal peptide binds to the TatBC subunits of the TatABC complex [23], triggering recruitment of the separate TatA complex (reviewed in Refs [1-3]) and initiation of translocation (step 3). Step 4 depicts the completion of translocation of the passenger protein, with the signal peptide still bound at the cis face of the membrane in a loop configuration. There is evidence that both Tat and Sec substrates are transported by a ‘loop’ mechanism in which the N-terminal region of the signal peptide remains bound to the cis side of the membrane while the mature protein is translocated across [24-26].

image

Figure 8. Model for the Tat-dependent export of YedY. Pre-YedY is synthesized in the cytosol (step 1) and binds post-translationally to the TatBC subunits of the TatABC complex (step 2). This triggers recruitment of the separate TatA complex, and translocation of the mature protein through the translocation channel ensues (step 3). In step 4, translocation of the mature domain has been completed and the signal peptide spans the membrane; the black/grey oval represents the N-terminus of the signal peptide, which remains on the cytoplasmic side of the membrane. Step 5 depicts two scenarios; in one (A) the entire precursor protein is transported into the periplasm, after which it inserts into the plasma membrane. The alternative scenario (B) is more consistent with the data from this study; here, the precursor protein exits laterally from the translocation channel into the bilayer, remaining in a transmembrane form. In the normal course of events, signal peptidase cleaves the signal peptide and releases the mature protein into the periplasm (step 6). However, the exposed mature domain can be subjected to proteolysis if maturation is prevented (step 7).

Download figure to PowerPoint

In this report, we have shown that unprocessed pre-YedY is integrated into the membrane, and step 5 shows two possible mechanisms by which this may be achieved. One scenario illustrated in Fig. 8A involves full translocation into the periplasm and subsequent reinsertion into the bilayer from the periplasmic side. Our data are not consistent with this scenario, for reasons outlined above. Instead, we believe that route B applies, whereby the signal peptide never reaches the trans side of the membrane and the precursor protein is released laterally into the lipid bilayer. Step 6 then illustrates the final step for a typical precursor protein, in which the processing peptidase removes the signal peptide to release the mature protein. If this is prevented, as is the case in this study, the outcome may involve eventual proteolytic cleavage of the exposed mature domain as shown in step 7.

It should be emphasized that our data are highly consistent with this ‘lateral exit’ model, but further studies are required for a definitive understanding. For example, we cannot exclude a model whereby precursor molecules do enter the periplasm, but very rapidly intergrate into the bilayer. However, our model is consistent with two studies on the thylakoid Tat system. First, the Rieske Fe/S protein of the cytochrome b6/f complex appears to be a rare natural example of an uncleaved precursor protein, in which the signal peptide anchors the protein to the thylakoid membrane from the lumenal side [27, 28]. Second, a hybrid precursor protein was found to undergo a slow import and maturation pathway that appeared to involve lateral release into the membrane bilayer prior to processing [23]. A number of E. coli inner membrane proteins are also known to be inserted via the Tat pathway [29], and their insertion pathway may well resemble that depicted in this figure for unprocessed pre-YedY.

Finally, our results show that Tat substrate maturation is important for cell viability. The accumulation of unprocessed YedY–A44L has serious adverse affects on growth rate and this clearly indicates that the presence of unprocessed precursors in the membrane causes problems. Similar effects have been noted when Sec substrates accumulate in bacteria [30, 31] and our data thus serve to underline the importance of precursor processing in the overall Tat export pathway.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Bacterial strains, plasmids and growth conditions

All strains and plasmids used are given in Table 1. The constructs described below were expressed using the arabinose-inducible pBAD24 vector and IPTG-inducible pEXT22 vector. Arabinose-resistant derivatives were used as described [14, 32]. Escherichia coli was grown aerobically at 37 °C in Luria–Bertani medium (LB) as described previously [31]. Mediums were supplemented with the following final concentrations: ampicillin (100 mg·mL−1), Kanamycin (50 μg·mL−1) and arabinose (100 mm).

Table 1. Plasmids and strains used in this study
Strain/plasmidRelevant propertiesReference/source
DH5aF_80lacZ_M15_(lacZYA-argF) U169 deoR recA1 endA1 hsdR17 (rk_, mk_) gal-phoA supE44 thi-1 gyrA96 relA1Invitrogen, Carlsbad, CA, USA
MC4100ARAraR, FaraD139DlacU169 rpsL150 relA1 flB5301 deoC1 ptsF25 rbsR [32]
ΔtatABCDEAs MC4100AR; ΔtatABCDE [32, 34]
pCRY1pBAD24 expressing pre-YedY with a 6-His tag under control of pBAD promoterThis study
pCRY5pEXT22 expressing pre-YedY with a 6-His tag under control of lacZ promoterThis study
pXTApEXT22 expressing TorA–GFP with a Strep-tagThis study

Site-directed mutagenesis

Mutagenesis was performed using the Stratagene (La Jolla, CA, USA) QuikChange site-directed mutagenesis kit and confirmed by DNA sequencing. The following primers were used: YedY_A44L_F (5′-GCCTCACGCTGCGCATCTCGATCTGCTTAGC-3′; YedY_A44L_R (5′-GCTAAGCAGATCGAGATGCGCAGCGTGAGGC-3′).

Cell fractionation

Starter cultures (5 mL) were grown overnight and used to inoculate 50 mL of LB (with the appropriate antibiotic selection and arabinose/IPTG to induce the expression of proteins) to a starting D600 of 0.1. Cells were grown to the mid-exponential phase before fractionation as described [35]. Spheroplasts were lysed by sonication, and intact cells and cellular debris were removed by centrifugation (5 min at 10 000 g). Membranes were separated from the cytoplasmic fraction by centrifugation (30 min at 250 000 g). Equal amounts of each fraction were separated by 15% SDS/PAGE and immunoblotted. Proteins expressed in pBAD24 were immunoblotted with His antibodies (Invitrogen, Paisley, UK) and a secondary antibody, horseradish peroxidase anti-mouse IgG conjugate was used. Proteins expressed in pEXT22 were immunoblotted with mouse anti-(strep-tag II) HRP conjugate (IBA, Germany). Both were detected by ECL detection system (Amersham Pharmacia Biotech, Little Chalfont, UK).

Proteolysis of spheroplasts

Spheroplasts were prepared as described and resuspended in proteolysis buffer (10 mm Hepes pH 7.6, 2 mm EDTA). The spheroplasts were then either treated with proteinase K (to a final concentration of 0.25 mg·mL−1) for 30 min on ice, or broken open by 5× free–thaw cycles before treatment with proteinase K.

Carbonate extraction

Spheroplasts were prepared as described and resuspended in 1.0 mL ice-cold 0.1 m pH 11.5 Na2CO3. Samples were incubated on ice for 30 min and pelleted by ultracentrifugation at 10 300 g for 30 min at 4 °C (Beckman TL100.3 rotor). Supernatant was collected and precipitated with 10% trichloroacetic acid, and pellet was dried after the rinse with ice-cold water. The supernatant and pellet were dissolved in 80 μL SDS sample loading buffer. For TatAd extraction tests, the TatAd protein was expressed as in Barnett et al. [16].

Trichloroacetic acid precipitation

Protein samples were concentrated for gel electrophoresis by trichloroacetic acid precipitation. One millilitre of 10% trichloroacetic acid was added and samples were incubated on ice for 30 min before centrifugation at 20 800 g for 15 min (Eppendorf 5417R). The supernatant was removed and 1 mL acetone was added and incubated on ice for 5 min before centrifugation for 10 min. The supernatant was removed and samples were allowed to air dry at room temperature. Samples were resuspended in 80 μL of SDS sample loading buffer.

Time course during arabinose induction

Starter cultures (5 mL) were grown overnight and used to inoculate 50 mL of LB (with the appropriate antibiotic selection and arabinose/IPTG to induce the expression of proteins) to a starting D600 of 0.1. Cells were grown at 37 °C and cell samples were taken every hour for a period of 5 h. At each time point, the same amount of cells measuring by D600 were collected by centrifugation and immediately frozen. Total cell protein content was separated on a 15% SDS/PAGE and immunoblotted as described earlier.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

We are grateful to Cristina Matos for help and advice throughout this work.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  • 1
    Yuan J, Zweers JC, van Dijl JM & Dalbey RE (2010) Protein transport across and into cell membranes in bacteria and archaea. Cell Mol Life Sci 67, 179199.
  • 2
    Robinson C, Matos CF, Beck D, Ren C, Lawrence J, Vasisht N & Mendel S (2011) Transport and proofreading of proteins by the twin-arginine translocation (Tat) system in bacteria. Biochim Biophys Acta 1808, 876884.
  • 3
    Lee P, Tullman-Ercek D & Georgiou G (2006) The bacterial twin-arginine translocation pathway. Annu Rev Microbiol 60, 373395.
  • 4
    Dalbey RE, Wang P & van Dijl J-M (2012) Membrane proteases in the bacterial protein secretion and quality control pathway. Microbiol Mol Biol Rev 76, 311330.
  • 5
    Von Heijne G (1990) The signal peptide. J Membr Biol 115, 195201.
  • 6
    van Roosmalen ML, Geukens N, Jongbloed JD, Tjalsma H, Dubois JY, Bron S, van Dijl J-M & Anne J (2004) Type I signal peptidases of Gram-positive bacteria. Biochim Biophys Acta 1694, 279297.
  • 7
    Lüke I, Handford JI, Palmer T & Sargent F (2009) Proteolytic processing of Escherichia coli twin-arginine signal peptides by LepB. Arch Microbiol 191, 919925.
  • 8
    Stanley NR, Palmer T & Berks BC (2000) The twin arginine consensus motif of Tat signal peptides is involved in Sec-independent protein targeting in Escherichia coli. J Biol Chem 275, 1159111596.
  • 9
    Chaddock AM, Mant A, Karnauchov I, Brink S, Herrmann RG, Klösgen RB & Robinson C (1995) A new type of signal peptide: central role of a twin-arginine motif in transfer signals for the pH-dependent thylakoidal protein translocase. EMBO J 14, 27152722.
  • 10
    Shackleton J & Robinson C (1991) Transport of proteins into chloroplasts. The thylakoidal processing peptidase is a signal-type peptidase with stringent substrate requirements at the -3 and -1 positions. J Biol Chem 266, 1215212156.
  • 11
    Karlsson AJ, Lim H-K, Hansen X, Rocco MA, Bratkowski MA, Ke A & DeLisa MP (2012) Engineering antibody fitness and function using membrane-anchored display of correctly-folded proteins. J Mol Biol 416, 94107.
  • 12
    Tullman-Ercek D, DeLisa MP, Kawarasaki Y, Iranpour P, Ribnicky B, Palmer T & Georgiou G (2007) Export pathway selectivity of Escherichia coli twin arginine translocation signal peptides. J Biol Chem 282, 83068309.
  • 13
    Loschi L, Brokx S, Hills T, Zhang G, Bertero M, Lovering A, Weiner J & Strynadka N (2004) Structural and biochemical identification of a novel bacterial oxidoreductase. J Biol Chem 279, 5039150400.
  • 14
    Guzman L-M, Belin D, Carson MJ & Beckwith J (1995) Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J Bacteriol 177, 41214130.
  • 15
    Molloy MP (2008) Isolation of bacterial cell membranes proteins using carbonate extraction. Methods Mol Biol 424, 397401.
  • 16
    Barnett JP, Eijlander RT, Kuipers OP & Robinson C (2008) A minimal Tat system from a Gram-positive organism: a bifunctional TatA subunit participates in discrete TatAC and TatA complexes. J Biol Chem 283, 25342542.
  • 17
    Jongbloed JD, Martin U, Antelmann H, Hecker M, Tjalsma H, Venema G, Bron S, van Dijl J-M & Müller J (2000) TatC is a specificity determinant for protein secretion via the twin-arginine translocation pathway. J Biol Chem 275, 4135041357.
  • 18
    Thomas JD, Daniel RA, Errington J & Robinson C (2001) Export of active green fluorescent protein to the periplasm by the twin arginine translocase (Tat) pathway in Escherichia coli. Mol Microbiol 39, 4753.
  • 19
    Alami M, Luke I, Deitermann S, Eisner G, Koch HG, Brunner J & Muller M (2003) Differential interactions between a twin-arginine signal peptide and its translocase in Escherichia coli. Mol Cell 12, 937946.
  • 20
    Mould RM & Robinson C (1991) A proton gradient is required for the transport of two lumenal oxygen-evolving proteins across the thylakoid membrane. J Biol Chem 266, 1218912193.
  • 21
    Musser SM & Theg SM (2000) Characterization of the early steps of OE17 precursor transport by the thylakoid Delta pH/Tat machinery. Eur J Biochem 267, 25882598.
  • 22
    Hou B, Frielingsdorf S & Klösgen RB (2006) Unassisted membrane insertion as the initial step in DeltapH/Tat-dependent protein transport. J Mol Biol 355, 957967.
  • 23
    Freilingsdorf S & Klösgen RB (2007) Prerequisites for terminal processing of thylakoidal Tat substrates. J Biol Chem 282, 2445524462.
  • 24
    Shaw AS, Rottier PJM & Rose JK (1988) Evidence for the loop model of signal-sequence insertion into the endoplasmic reticulum. Proc Natl Acad Sci USA 85, 75927596.
  • 25
    Kuhn A, Kiefer D, Koëhne C, Zhu H-Y, Tschantz WR & Dalbey RE (1994) Evidence for a loop-like insertion mechanism of pro-Omp A into the inner membrane of Escherichia coli. Eur J Biochem 226, 891897.
  • 26
    Fincher V, McCaffrey M & Cline K (1998) Evidence for a loop mechanism of protein transport by the thylakoid Delta pH pathway. FEBS Lett 423, 6670.
  • 27
    Karnauchov I, Herrmann R & Klosgen R (1997) Transmembrane topology of the Rieske Fe/S protein of the cytochrome b6/f complex from spinach chloroplasts. FEBS Lett 408, 206210.
  • 28
    Molik S, Karnauchov I, Weidlich C, Herrmann R & Klosgen R (2001) The Rieske Fe/S protein of the cytochrome b6/f complex in chloroplasts: missing link in the evolution of protein transport pathways in chloroplasts? J Biol Chem 276, 4276142766.
  • 29
    Hatzixanthis K, Palmer T & Sargent F (2003) A subset of bacterial inner membrane proteins integrated by the twin-arginine translocase. Mol Microbiol 49, 13771390.
  • 30
    Dalbey RE & Wickner W (1985) Leader peptidase catalyzes the release of exported proteins from the outer surface of the Escherichia coli plasma membrane. J Biol Chem 260, 59255931.
  • 31
    Fikes J & Bassford P (1987) Export of unprocessed precursor maltose-binding protein to the periplasm of Escherichia coli cells. J Bacteriol 169, 23522359.
  • 32
    Bolhuis A, Mathers JE, Thomas JD, Barrett CML & Robinson C (2001) TatB and TatC form a functional and structural unit of the twin-arginine translocase from Escherichia coli. J Biol Chem 276, 2021320219.
  • 33
    Casadaban MJ & Cohen SN (1980) Lactose genes fused to exogenous promoters in one step using a Mu-lac bacteriophage: in vivo probe for transcriptional control sequences. J Mol Biol 138, 179207.
  • 34
    Wexler M, Sargent F, Jack RL, Stanley NR, Bogsch EG, Robinson C, Berks BC & Palmer T (2000) TatD is a cytoplasmic protein with DNase activity. No requirement for TatD family proteins in Sec-independent protein export. J Biol Chem 275, 1671716722.
  • 35
    Randall LL & Hardy SLS (1986) Correlation of competence for export with lack of tertiary structure of the mature species: a study in vivo of maltose-binding protein in E. coli. Cell 46, 921928.