Skip to main content
NCBI home page
The EMBO Journal logoLink to The EMBO Journal
. 2004 Apr 1;23(8):1868–1877. doi: 10.1038/sj.emboj.7600172

Methionine sulfoxide reductases protect Ffh from oxidative damages in Escherichia coli

Benjamin Ezraty 1, Régis Grimaud 1,*, Mohammed El Hassouni 1,†, Daniéle Moinier 1, Frédéric Barras 1,a
PMCID: PMC394232  PMID: 15057280

Abstract

In proteins, methionine residues are primary targets for oxidation. Methionine oxidation is reversed by methionine sulfoxide reductases A and B, a class of highly conserved enzymes. Ffh protein, a component of the ubiquitous signal recognition particle, contains a methionine-rich domain, interacting with a small 4.5S RNA. In vitro analyses reported here show that: (i) oxidized Ffh is unable to bind 4.5S RNA, (ii) oxidized Ffh contains methionine sulfoxide residues, (iii) oxidized Ffh is a substrate for MsrA and MsrB enzymes; and (iv) MsrA/B repairing activities allow oxidized Ffh to recover 4.5S RNA-binding abilities. In vivo analyses reveal that: (i) Ffh synthesized in the msrA msrB mutant contains methionine sulfoxide residues and is unstable, (ii) msrA msrB mutant requires high levels of Ffh synthesis for growth and (iii) msrA msrB mutation leads to defects in Ffh-dependent targeting of MalF. We conclude that MsrA and MsrB are required to repair Ffh oxidized by reactive oxygen species produced by aerobic metabolism, establishing an as-yet undescribed link between protein targeting and oxidation.

Keywords: methionine sulfoxide reductases, oxidative stress, pathogenicity, signal recognition particle

Introduction

The functional alteration of proteins by oxidation emerges as a central cause of various pathologic disorders (Stadtman and Levine, 2000). Reactive oxygen species (ROS) are produced by aerobic metabolism or under specific conditions such as host:parasite interactions and aging process (Stadtman, 1992). The frequent targets of oxidation in proteins are methionine (Met) residues, which are converted to methionine sulfoxide (MetSO) (Vogt, 1995). Methionine sulfoxide reductase A (MsrA), which regenerates Met from MetSO, occurs in most living cells (Brot et al, 1981). The ubiquitous distribution of msrA was expected to correlate with an essential function in cellular physiology. Accordingly, alteration of MsrA synthesis caused global phenotypes such as modified life span in mouse and drosophila (Moskovitz et al, 2001; Ruan et al, 2002), attenuated virulence in bacterial pathogens (Wizemann et al, 1996; Hassouni et al, 1999) or increased sensitivity to a high concentration of exogenously added oxidative agents (Moskovitz et al, 1995, 1998). Surprisingly, however, no major physiological defects were observed in an Escherichia coli msrA mutant grown under normal aerobic growth conditions (Moskovitz et al, 1995; our unpublished data).

Recently, we identified a second ubiquitous methionine sulfoxide reductase encoding gene, referred to as msrB (Grimaud et al, 2001; Kryukov et al, 2002; Olry et al, 2002). Biochemical investigations established that MsrA and MsrB enzymes exhibit different diastereospecificity, the former acting on Met-(S)-SO and the latter on Met-(R)-SO diastereoisomers (Sharov et al, 1999; Kryukov et al, 2002; Olry et al, 2002). This diastereospecificity accounted for the fact that MsrA and MsrB enzymes were required for reduction of all MetSO present in oxidized calmodulin, while the individual action of each allowed reduction of a subset of MetSO only (Grimaud et al, 2001). No drastic phenotypic changes were found to be associated with msrB mutation in E. coli (Grimaud et al, 2001).

In order to investigate the physiological role of the MsrA/B repair pathway in vivo, our strategy was to identify a protein with the following criteria: (i) the function of this protein should be associated with easily scorable phenotype, (ii) this protein should contain a higher than average number of Met residues and (iii) this protein should be conserved throughout evolution (like MsrA and MsrB). The SRP component Ffh fitted all of these criteria, and its conserved role in one of the most fundamental process of cellular life rendered it even more attractive.

SRP catalyzes protein targeting in all cellular organisms (Herskovits et al, 2000; Keenan et al, 2001). The eukaryotic SRP recognizes the NH2-terminal signal sequence of targeted proteins as they emerge from the ribosome. Binding to the SRP arrests polypeptide elongation and mediates docking of the translating ribosome with receptors in the membrane, where translation resumes along with translocation to the lumen of endoplasmic reticulum. In E. coli, the SRP component is essential for viability and efficient protein integration in the inner membrane (Brown and Fournier, 1984; Phillips and Silhavy, 1992; Luirink et al, 1994; Ulbrandt et al, 1997). The E. coli SRP consists of the Ffh protein, the ffs-encoded 4.5S RNA and the FtsY receptor, which are the prokaryotic counterparts of the eukaryotic SRP54, 7SL RNA and SRα components, respectively. Functional conservation between eukaryotic and prokaryotic SRP components is best illustrated by the fact that human SRP54 can interact with bacterium 4.5S RNA, while Ffh substitutes for SRP54 (Poritz et al, 1990). A salient feature of all SRP54 homologs is the occurrence of a Met-rich domain, referred to as M-domain. A decade ago, in the so-called ‘bristles' hypothesis, lateral chains of Met residues were proposed to have a particular role in signal peptide binding (Bernstein et al, 1989). Subsequently, the M domain was shown to contain the 4.5S RNA-binding site and to be required for binding to signal sequences of presecretory proteins (Römisch et al, 1990; Zopf et al, 1990). Structural analysis of the Thermus aquaticus Ffh revealed that the M domain exhibits a deep groove composed of hydrophobic amino acids, including Met residues (Keenan et al, 1998). This groove was proposed to be the signal sequence-binding pocket of SRP. The position of Met residues of the E. coli M domain could be inferred from the T. aquaticus structure. Likewise, Met residues were proposed to line the hydrophobic groove such that they could interact with presecretory proteins (Keenan et al, 1998). Moreover, a detailed structural analysis of the E. coli M domain–RNA complex was carried out (Batey et al, 2000). Noticeably, the Met382 residue that is invariant throughout all prokaryotic and eukaryotic sequences was found to locate within the RNA-binding region of the protein.

In this study, we report that in E. coli MsrA and MsrB enzymes are necessary for protecting SRP from oxidative damages. This finding establishes a functional link between oxidative stress and protein targeting.

Results

Oxidized Ffh is a substrate of MsrA and MsrB

We first tested the hypothesis that oxidation could convert Ffh to a bona fide substrate for MsrA and/or MsrB enzymes. Purified Ffh was submitted to oxidation by H2O2, and subsequently used as a substrate for MsrA and/or MsrB enzymes. Kinetic analyses revealed that MsrA and MsrB exhibited initial rates of NADPH oxidation of 6.4 and 5.35 μM of NADPH oxidized/min, respectively (Figure 1). When MsrA-catalyzed reaction reached completion, MsrB was added and the reaction was found to resume (Figure 1). Similarly, when MsrB-catalyzed reaction reached a plateau, addition of MsrA allowed it to resume (Figure 1). This was similar to what was previously found using oxidized calmodulin as a substrate, and pointed to the existence of two types of MetSO diastereoisomers in oxidized Ffh (Grimaud et al, 2001). When both MsrA and MsrB enzymes were added simultaneously, the initial rate of NADPH oxidation observed was 17.5 μM of NADPH oxidized/min (Figure 1). The total amount of NADPH oxidized with both MsrA and MsrB was the same as that reached when both enzymes were added sequentially (Figure 1). In this experiment, 10 μM Ffh was oxidized, yielding a concentration of MetSO residues in the reaction of 280 μM. As reaching the plateau needed 250 μM NADPH to be oxidized, this indicated that the oxidation protocol was efficient enough to convert most of the 28 Met residues in MetSO, an interpretation that fits with the mass spectrometry data described below (Figure 2). No Msr activity, was observed when nonoxidized SRP was used as a substrate (data not shown).

Figure 1.

Figure 1

Open in a new tab

Oxidized Ffh is a substrate for MsrA and MsrB. Oxidized Ffh was obtained after submitting SRP to H2O2 (50 mM) as described in Materials and methods. The reaction mixtures included NADPH (400 μM), thioredoxin (5 μM), thioredoxin reductase (87 nM) and Ffhox (10 μM). In the first experiment, reaction was carried out first with 1 μM of MsrB (▪) until completion, at which point (indicated by an arrow) 1 μM MsrA (□) was added. In the second experiment, reaction was carried out first with 1 μM of MsrA (♦) until completion, at which point 1 μM MsrB (•) was added. In the third experiment, both MsrA (1 μM) and MsrB (1 μM) were added simultaneously (▴). A schematic representation of the reaction catalyzed by Msr enzymes is shown above the curve.

Figure 2.

Figure 2

Open in a new tab

Oxidation and MsrA/B repair of Ffh. Pure Ffh was oxidized with H2O2 and subsequently repaired by MsrA and MsrB enzymes. Samples were then submitted to tryptic hydrolysis and analyzed by MALDI/TOF-MS. (A) The primary sequence of Ffh is shown. Peptides whose oxidation status was followed during the experiment are indicated in boxes. (B) Predicted mass values of peptides 1–5 described above are indicated. (C) Native Ffh, oxidized Ffh (referred to as Ffhox) and oxidized Ffh repaired with MsrA and MsrB were hydrolyzed and the resulting peptides were separated by MALDI/TOF-MS.

Mass spectrometry analyses of oxidized Ffh and of MsrA/B repaired Ffh

Purified Ffh was oxidized and submitted to repair by MsrA and MsrB. Samples containing native Ffh, oxidized Ffh or MsrA/B-repaired oxidized Ffh were run on SDS–polyacrylamide gel electrophoresis (PAGE). Bands of each sample were cut off the gel, submitted to trypsin digest and the resulting peptides were analyzed by MALDI/TOF-MS (Figure 2). The spectra obtained were scanned for Met-containing peptides. Trypsin hydrolysis was predicted to produce 11 peptides including at least one Met residue. Based on their predicted mass values, five could be identified within the spectra of tryptic digest of native Ffh. These peptides are visualized in the primary sequence shown in Figure 2A. Peptides 3 and 4 exhibited similar theoretical mass values and are expected to give rise to superimposed peaks (Figure 2B). When the spectrum of oxidized Ffh was scanned, mass values of peptides 2, 3, 4 and 5 were increased by 16 Da, corresponding to the gain of one oxygen atom (Figure 2C). No peaks were found at mass values corresponding to the reduced form of these peptides. Two peaks at mass values corresponding, respectively, to reduced and oxidized forms of peptide 1 were visible in the spectrum of oxidized Ffh (Figure 2C). The spectra of MsrA/B-repaired Ffh were found to contain only reduced forms of peptides 1, 3, 4 and 5 (Figure 2C). Both reduced and oxidized forms of peptide 2 were present (Figure 2C). Taken together, these analyses strongly suggested that oxidized Ffh contained MetSO residues that had been converted to Met in the MsrA/B-repaired Ffh.

Mass spectrometry analysis of the oxidation state of Ffh in msrA msrB mutants

We tested the effect of lack of MsrA and MsrB enzymes on the oxidation state of Ffh in vivo. Ffh-(His)6 was overproduced in strains MC4100 and BE53, an isogenic derivative lacking both msrA and msrB genes. Total protein extracts were run on SDS–PAGE and bands corresponding to Ffh were excised, submitted to tryptic digestion and the resulting products were analyzed by mass spectrometry. The spectra of tryptic digest of Ffh synthesized in either background were scanned as described above. Peptides 1, 2, 3 and 4 could be identified. Both reduced and oxidized forms of peptides 2, 3 and 4 were to be found in MC4100 strains while, in contrast, only primarily forms were found in msrA msrB mutants (Figure 3). Both oxidized and reduced forms of peptide 1, which contains the initiator Met residue, were found in MC4100 and mutant strains (Figure 3). Taken together, these results suggested that msrA msrB mutation led to increased oxidation of Ffh. To ascertain that oxidation was due to aerobic metabolism, mass spectrometry analysis was then carried out using MC4100 and mutant strains grown under anaerobiosis. The vast majority of the three peptides analyzed was in a reduced form in the two strains (Figure 3).

Figure 3.

Figure 3

Open in a new tab

Oxidation status of Ffh synthesized in msrA msrB mutant of E. coli. A similar protocol as described in the legend of Figure 2 was followed, except that Ffh was produced in vivo using pBadFfh, either in MC4100 or in isogenic BE53 msrA msrB mutant. Experiments were carried out using either aerobic or anaerobic culture conditions. Cell extracts were run on SDS–PAGE and the band corresponding to Ffh was cut off the gel prior trypsin hydrolysis and MALDI/TOF-MS analysis.

Functional properties of oxidized Ffh and of MsrA/B-repaired Ffh

We investigated whether oxidized Ffh was able to interact with 4.5S RNA. In vitro synthesized 4.5S RNA was labelled with 32P and incubated with increasing concentration of oxidized Ffh (Figure 4). Ffh/4.5S RNA interaction was investigated using a filter-binding assay as described by Suzuma et al (1999). At high concentrations of oxidized Ffh (referred to as Ffhox) (10 μM), only 22.5% of added RNA was retained on the filter. In contrast, 0.5 μM of native Ffh was sufficient to retain over 90% added RNA. This result indicated that oxidized Ffh was altered in its ability to interact with 4.5S RNA.

Figure 4.

Figure 4

Open in a new tab

Interaction of 4.5S RNA with oxidized and repaired Ffh. Oxidized Ffh was obtained after submitting SRP to H2O2 (50 mM) as described in Materials and methods. A subset of it was incubated in the presence of both MsrA and MsrB in the presence of DTT. Radiolabelled 4.5S RNA was obtained by in vitro transcription of the ffs gene in the presence of [α-32P]CTP. Nitrocellulose filter-binding assay was used to examine the 4.5S RNA (0.15 nM) binding activity of Ffh, Ffhox or MsrA/MsrB-repaired Ffhox, as indicated in the graphs. Ffhox stands for oxidized Ffh.

Oxidized Ffh was submitted to repair by MsrA and MsrB and tested for its ability to bind 4.5S RNA (Figure 4). Using 10 μM of MsrA/B-treated oxidized Ffh allowed to retain 73% of added RNA. This showed that repaired Ffh recovered nearly native RNA-binding ability. Comparison with oxidized Ffh indicated that at a high concentration (10 μM), repaired Ffh was three times more efficient than oxidized Ffh in binding RNA. Interestingly, when oxidized Ffh was treated with either MsrA or MsrB alone, approximately 50% RNA was retained at high concentration (10 μM) (data not shown). This study demonstrated that the repair of oxidized Ffh by MsrA/B enzymes permits the recovery of its RNA-binding ability.

Stability of Ffh in msrA msrB mutant

Previous studies showed that Ffh is unstable in vivo when it is dissociated from 4.5S RNA (Jensen and Pedersen, 1994; Park et al, 2002). Therefore, a prediction of both in vivo and in vitro results is that the Ffh should be destabilized in msrA msrB mutants. Thus, MC4100 and BE53 (msrA msrB) strains were grown up to mid-exponential phase, submitted to chloramphenicol treatment, samples taken over time and analyzed by immunoblot for their content in Ffh. Ffh proved to be quite stable in MC4100 strain as ca. 80% of the initial amount was still present after 24 h incubation. In contrast, in BE53 strain, more than 50% of the initial Ffh was degraded in 6 h (Figure 5). These results indicated that a lack of Msr function resulted in destabilization of Ffh.

Figure 5.

Figure 5

Open in a new tab

Importance of msrA msrB for the stability of Ffh in vivo. Cultures of E. coli MC4100 and BE53 (msrA msrB) were grown at 37°C until exponential phase, and chloramphenicol was added to stop synthesis. Aliquots were removed after 1, 3, 6 and 24 h and analyzed by SDS–PAGE, as indicated above the picture. Protein levels were analyzed by Western blotting by using an anti-Ffh antiserum. ECL film was scanned and band intensities were quantified. The percent of remaining Ffh is given above the protein bands in the picture.

Genetic evidence for a physiological connection between MsrA, MsrB and Ffh

In E. coli, as the ffh gene is essential for cell viability, the steady state amount of Ffh must be kept above a threshold concentration. From the decreased stability of Ffh in msrA msrB mutants reported above followed the prediction that cell viability of msrA msrB mutants should require a higher level of Ffh synthesis as compared with the wild type. To test this prediction, we made use of the JM113 strain that contains a chromosomally located arabinose-inducible ffh gene (as unique ffh allele), allowing Ffh synthesis to be controlled. Both the JM113 strain and its isogenic derivative BEjm02 (msrA msrB) grew normally when cultures were fully induced with 0.2% arabinose, that is, synthesizing large amounts of Ffh (Figure 6A). In contrast, when Ffh synthesis was reduced, that is, in the presence of 0.02% arabinose, no growth was observed for the BEjm02 strain, whereas JM113 strain grew normally (Figure 6B). This fitted the prediction that more Ffh needs to be synthesized in msrA msrB backgrounds as compared with JM113. Presumably, the higher demand in Ffh synthesis was related to increased oxidative damage in msrA msrB mutant. To validate this, a similar growth experiment was carried out in the absence of oxygen. Strains were grown under anaerobiosis with trimethyl amine as an electron acceptor. No difference of Ffh synthesis requirement was found between JM113 and BEjm02 strains, as they grew the same at both high and low arabinose concentrations (Figure 6C and D).

Figure 6.

Figure 6

Open in a new tab

Synthetic lethality between the limiting amount of Ffh and msrA msrB mutations. Cultures of E. coli JM113 (λara-ffh) (•) and BEjm02 (λara-ffh msrA msrB) (▪) carrying the pBR322 plasmid were grown in M9 medium with casamino acids supplemented with arabinose 0.2% (A) or arabinose 0.02% (B). OD600 values were recorded. It is important to note that concentrations of arabinose used were found by Siegele and Hue to allow induction of all cells in the population (Siegele and Hu, 1997). Experiments shown in (C) and (D) are identical to (A) and (B), respectively, with strains grown in anaerobiosis using trimethylamine oxide as an electron acceptor.

Importance of MsrA/B for in vivo SRP-dependent protein targeting

To examine the phenotypic consequences of msrA msrB mutations on Ffh-dependent targeting of proteins, the polytopic MalF protein was used as a model. Two MalF–PSBT hybrids previously studied by Tian et al (2000) were used. Hybrid J contains a biotinylation domain that upon membrane insertion locates in the periplasmic compartment, and hybrid K contains a cytoplasmic biotinylation domain. Biotinylation takes place exclusively in the cytoplasm and only biotinylated proteins can interact with streptavidin. Therefore, a hybrid protein will be recognized by streptavidin only if its biotinylation domain localizes in the cytosol. In the MC4100 cells, hybrid J failed to be recognized by streptavidin (Figure 7A). In contrast, inactivation of both msrA and msrB genes resulted in hybrid J biotinylation. Providing msrA or msrB gene in trans allowed the recovery of correct membrane insertion as the J hybrid was no longer biotinylated (Figure 7A). This indicated that the presence of either one of the two Msr enzymes is sufficient to confer an MC4100 integration. This was consistent with the fact that single msrA or msrB mutation did not alter membrane insertion of J hybrid. Moreover, providing ffh gene in trans in msrA msrB backgrounds suppressed mislocation of MalF-PBST J hybrid, as this was later no longer biotinylated (Figure 7B). This confirmed that mislocation of J hybrid in msrA msrB strains was due to limitation in functional Ffh. As expected, the biotinylation pattern of hybrid K was found to be altered in none of the genetic backgrounds as expected. Immunoblot analysis using anti-MalF antibodies showed that similar amounts of J and K hybrid proteins were present in all strains used. Significantly, J hybrid was properly targeted in wt and in msrA msrB strains grown under anaerobiosis, indicating that oxygen was required for targeting defects to be seen (data not shown).

Figure 7.

Figure 7

Open in a new tab

Importance of msrA msrB for proper membrane protein integration. (A) MalF–PSBT fusions J and K were synthesized in MC4100, BE51 (msrA), BE52 (msrB), BE53 (msrA msrB), BE53/pMsrA and BE53/pMsrB strains. Cells were grown at 37°C in M9 medium supplemented with 2 nM biotin and kanamycin (100 μg/ml). Synthesis of MalF–PSBT fusions was induced with 0.5 mM of IPTG, at an OD600 or 0.3. After electrophoresis, the proteins were transferred to a nitrocellulose membrane. To detect biotinylated fusion proteins, streptavidin–horseradish peroxidase conjugate (Amersham Pharmacia) was used. The synthesized fusion proteins were detected using polyclonal anti-MalF. (B) Overproduction of Ffh suppresses translocation defects in msrA msrB mutant. MalF–PSBT fusion J was synthesized in BE53 (msrA msrB) and BE53/pFfh strains and studied as described above.

In order to test whether msrA msrB mutations had a general defect in protein translocation, the efficiency of the Sec pathway was investigated. The fate of three extracytoplasmic proteins was analyzed, namely, the periplasmic proteins MalE and TolB and the outer membrane lipoprotein Pal. Precursor forms of MalE were visible in the secY mutant (Figure 8A). In contrast, only mature forms of MalE were observed in wild type and in msrA msrB mutants, indicating that MalE had crossed the inner membrane equally efficiently in both strains. Similarly, pulse-chase experiments showed that translocation of neither TolB nor Pal was affected in the double mutant (Figure 8B and C).

Figure 8.

Figure 8

Open in a new tab

Sec-dependent translocation is unaffected by msrA msrB mutation. (A) Study of translocation of MalE. Cultures of MC4100, BE53 and CU164 were grown in M9 medium in the presence of maltose at 37°C. Cell extracts were prepared and analyzed by immunoblot using anti-MalE antiserum. A MalE precursor form was visible in secY background (indicated as MalE-P), while only mature forms were visible in MC4100 and BE53 strains. (B) Study of translocation of TolB. The fate of TolB was investigated by pulse-chase experiment. Use of sodium azide-treated cultures allowed to visualize the precursor form (lane 1). Pulse was carried out for 5 min, while samples were taken out at time points 0 s, 30 s and 2 min during the chase period and immunoprecipitated using anti-TolB antibodies. (C) Study of translocation of Pal. The fate of Pal was investigated by pulse-chase experiment using the same protocol as above, using anti-Pal antibodies.

Discussion

Ubiquitous methionine sulfoxide reductases MsrA and MsrB were proposed to be important for cells to resist oxidative stress (Weissbach et al, 2002). Here, we indeed demonstrate that MsrA and B are required to repair oxidatively damaged Ffh, a Met-rich protein conserved throughout procaryotes and eukaryotes. Hence, this study demonstrates that SRP-dependent protein translocation is a target for oxidation, and identifies the MsrA/B-repair pathway as a key actor in protecting this central cellular process.

Our aim was to identify cellular processes, the functioning of which relies on the MsrA/B repair pathway. We chose the SRP component Ffh as a model as it exhibits several criteria of interest (see Introduction), among which is a phylogenic ubiquity similar to that of Msr enzymes. In the first step, a series of chemical and biological studies were carried out in vitro to characterize the effects of oxidation on purified Ffh. Likewise, oxidized Ffh was found to contain MetSO residues and, accordingly, oxidation was shown to convert Ffh to a substrate for MsrA and MsrB enzymes. Subsequently, MetSO-containing Ffh was found to be unable to interact with 4.5S RNA and to form the SRP complex, while MsrA/B activities allowed oxidized Ffh to recover 4.5S RNA-binding abilities. This established the key role of MsrA/B in repairing oxidized Ffh. A second series of experiments were designed to investigate the in vivo situation. Mass spectrometry analysis of tryptic digests showed that Ffh synthesized in aerobically grown E. coli msrA msrB mutants contained MetSO residues as well. This clearly indicated that lack of MsrA/B leads to accumulation of unrepaired, MetSO-containing Ffh, leading to the prediction that no SRP complex would be formed. Two lines of evidences supported this prediction. First, Ffh was found to be unstable in msrA msrB mutants, as previously observed with 4.5S RNA free Ffh. Second, SRP-dependent targeting was found to be hampered in the absence of msrA msrB-encoded enzymes, whereas Sec-dependent targeting was not. Moreover, a case of synthetic lethality was found when combining the reduced level of Ffh and inactivation of msrA msrB genes, giving further credence to a physiological connection between SRP-dependent protein translocation and MsrA/B repair pathway. Thus, both in vivo and in vitro studies converged to identify SRP-mediated protein targeting as a cellular target for MsrA/B repair.

Reduced virulence was found to be due to msrA mutation in several pathogens. Surprisingly, lack of MsrA enzyme led to defects in cell envelope-related process, that is, decreased type I fimbriae-mediated hemagglutination in enteropathogenic E. coli and loss of motility in Erwinia chrysanthemi (Wizemann et al, 1996; Hassouni et al, 1999). MsrA being cytoplasmic, the link with extracytoplasmic functions was unclear. A possibility was that these extracytosolic processes depend upon a cytosolic component, itself dependent upon a functional MsrA enzyme. Our present finding opens a new possibility. Adherence and cell motility depend on inner membrane proteins, the proper membrane integration of which is likely to depend upon SRP pathway. As a working hypothesis, we might envision that host-released ROS alter, among other cellular processes, SRP-dependent protein export. This could in turn reduce the targeting efficiency of virulence-associated factors. Hence, according to this scenario, msrA mutant would exhibit reduced virulence as it could not fully repair SRP. Against this view stands our own finding that msrA mutation alone does not alter SRP targeting and, hence, is unlikely to affect virulence. However, one should consider the fact that host-released ROS are likely to be in much higher amount than those produced by aerobic metabolism. The need for MsrAB repair might be more important in a host context than in a simple lab culture.

In vitro analysis established that both MsrA and MsrB enzymes differ in their diastereospecificity (Sharov et al, 1999; Kryukov et al, 2002; Olry et al, 2002). Yet, under normal laboratory growth conditions, mutating either msrA or msrB was neutral, suggesting that lack of one enzyme species could be compensated for by the presence of the other. In order to account for the neutral effect of mutations in either msrA or msrB, one must first assume that oxidation is essentially a stochastic process, and then, take into account the chemical and functional heterogeneity within a polypeptide population exposed to ROS. For each protein species, ROS exposure should yield a set of polypeptide chains that contain Met residues left nonoxidized, as well as both S- and R-MetSO diastereoisomers, the oxidized status of each Met residue varying from one polypeptide chain to the other. This assumption received experimental support in the present work, as a mixed population of reduced and oxidized forms was observed by mass spectrometry analysis of tryptic digests of in vivo synthesized Ffh. The fate of each polypeptide chain will then depend on the role of each Met residue: oxidation of Met residues located at the surface might make no difference in terms of stability and/or activity, whereas oxidation of structurally or functionally important Met might be harmful. When bacteria face low level of ROS produced endogenously by aerobic metabolism, the size of the oxidized subpopulation might be low and repair by either one of the two, MsrA or MsrB, might be sufficient. This could explain why E. coli strain synthesizing only one Msr exhibits a neutral phenotype under normal growth conditions. Alternatively, biochemical experiments have argued for the existence of an epimerase that could convert Met-R-SO to Met-S-SO (Weissbach et al, 2002). This could account for the neutral phenotype of msrB mutation at least.

If one assumes oxidation to be a stochastic process, the prediction is that the composition of a heterogenous polypeptide population will vary as a function of the ratio between the size of this population and the level of ROS. Experimental validation of this hypothesis was obtained by varying the size of the Ffh population and keeping the level of ROS constant. Thus, under aerobic conditions, we observed synthetic lethality between msrA msrB mutations and reduced Ffh synthesis. This indicated that oxidized Ffh accumulated in the absence of MsrA/B repair function, diminishing the size of the functional Ffh subpopulation. (A complementary evidence was provided by the fact that overproduction of Ffh could suppress MalF targeting defects of msrA msrB strain). Remarkably, when strains were grown in the absence of oxygen, synthetic lethality between reduced Ffh synthesis and msrA msrB mutant disappeared, demonstrating that oxidation was indeed the cause of the shortage in functional Ffh.

Materials and methods

Chemicals

Unless noted otherwise, chemicals were purchased from Sigma. Thioredoxin reductase was a gift from Dr C Williams (University of Michigan, Ann Arbor). Bacteria were grown on Luria-Bertani medium or M9 minimal medium.

Strain construction

The DNA region encompassing msrA gene was cloned into pUC18 plasmid. The resulting plasmid was restricted by HincII enzyme and ligated with a SmaI restricted fragment containing the ΩSpc cassette. The recombinant plasmid was then linearized and electroporated into E. coli KM354 (recJ) strain carrying the pTP223 plasmid (bet gam exo) (Murphy, 1998). SpcR colonies were selected and checked for AmpS phenotype. Polymerase chain reaction (PCR) was then used to check whether recombination occurred at the msrA locus. An msrB deletion-containing strain was constructed by replacing msrB gene with a CamR cassette from plasmid pKD3 in strain BW25113/pKD46, as previously described (Datsenko and Wanner, 2000). Mutations were subsequently transferred into a different strain by transduction with P1 phage (Table I).

Table 1.

Strains and plasmids

  Relevant characteristics Reference or source
Strains
 MG1655 Wild type Lab collection
 JM113 ffh∷KanR λ(arap-ffh, AmpR) MacFarlane and Muller (1995)
 BEjm00 JM113 msrA∷SpcR This study
 BEjm01 JM113 ΔmsrB∷CamR This study
 BEjm02 JM113 msrA∷SpcR ΔmsrB∷CamR This study
 MC4100 araD139 Δ(argF-lac)U169 rpsL150 Lab collection
  relA1 thi fibB5301 deoC1 ptsF25 rbsR  
 BE51 MC4100 msrA∷SpcR This study
 BE52 MC4100 ΔmsrB∷CamR This study
 BE53 MC4100 msrA∷SpcR ΔmsrB∷CamR This study
 CU164 secY39cs, zhd-33∷Tn10 LF Wu
     
Plasmids
 pET21a AmpR Lab collection
 pETffh pET21a carrying ffh gene This study
 pFfh pUC18 carrying ffh gene This study
 pGEMT AmpR Lab collection
 pGEMTffs pGEMT carrying ffs gene This study
Open in a new tab

Plasmid construction

The ffh coding region was amplified by PCR using chromosomal DNA from E. coli strain MG1655 as a template and the oligonucleotides pffhC (5′-GGTACGATCGCTCGAGGCGACCAGGGAAGCCTGGGGG-3′) and pffhN (5′-CTGATAGAATTCCATATGTTTGATAATTTAACCG-3′) as primers. The PCR product was inserted into NdeI- and XhoI-restricted plasmid pET21a (Novagen), yielding plasmid pETffh.

The ffh coding region was amplified from chromosomal DNA from E. coli strain MG1655, using oligonucleotides 5′ coding pffh5, containing an EcoRI site (5′-CTGATAGAATTCTGGGCATGAAGGTTGCCA-3′) and 3′ reverse complement pffh3 containing an SphI site (5′-TAGCTCAGCATGCGCAAACAAGCCAGGCCGA-3′). Note that pffh5 and pffh3 oligonucleotides are located at 520 nt upstream and 10 nt downstream of the ffh coding region. The resulting PCR product was cloned into pUC18 plasmid, previously cut with EcoRI and SphI enzymes, yielding pFfh.

Plasmid pGEMTffs was obtained by cloning ffs gene in the pGEMT vector (Promega). The ffs coding region was amplified by PCR using chromosomal DNA from E. coli strain MG1655 as a template and the oligonucleotides pffs5′ (5′-TGACGTCGCGTTGGTTCTCAACGC-3′) and pffs3′ (5′-TATGCATGGGTGGGGGCCCTGCCAG -3′) as primers. The PCR product was inserted in AatII–NsiI restricted pGEMT plasmid, yielding plasmid pGEMTffs.

Purification of Ffh, MsrA and MsrB proteins

E. coli strain BL21 (DE3)/pETffh cells were grown at 37°C in Luria Bertani containing 100 μg/ml ampicillin. At an OD600 of 0.6, 1 mM isopropyl-1-thio-β-D-galactopyranoside was added, and growth was continued for 2 h. Cells were harvested by centrifugation. The pellet (1.8 g) was resuspended in 7.2 ml of buffer A (25 mM Hepes, pH 7.5, 0.3 M KCl, 10% (v/v) glycerol). Resuspended cells were broken by a single pass through an ice-chilled French pressure cell at 6 tons. The resulting crude extract was centrifuged at 30 000 g for 30 min at 4°C. The supernatant was applied onto a 5 ml Hi-trap column (Amersham Pharmacia Biotech) loaded with nickel and equilibrated with buffer A plus 0.05 M imidazole. Proteins were eluted with a 13 ml gradient imidazole (0.05–0.5 M). Fractions were analyzed by SDS–PAGE. After elution with imidazol, the Ffh-containing fractions were pooled, and the solution was diluted to reach a KCl concentration of 0.1 M. This solution was loaded onto a MonoQ column (0.5 × 5 cm2) (Amersham), equilibrated with buffer B (same as buffer A with 0.1 M KCl). Ffh was eluted with 20 ml gradient ranging from 0.1 to 0.7 M KCl. Spectrophotometric and ethidium bromide-stained agarose gel analyses showed that purified Ffh was 4.5S RNA-free.

MsrA and MsrB proteins were purified by column chromatography as previously described (Grimaud et al, 2001).

In vitro oxidation of Ffh

Ffh (10 mg in buffer B described above) was treated with 50 mM H2O2 for 4 h at room temperature, as described by Levine et al (1996). H2O2 was removed by gel filtration through G25 Sephadex. The released oxidized Ffh was then concentrated by ultrafiltration on ultrafree biomax-5 K (Millipore).

Methionine sulfoxide reductase activity assay

Methionine sulfoxide reductase activity was assayed in buffer A (25 mM Hepes, pH 7.5, 0.3 M KCl, 10% (v/v) glycerol) as described previously (Moskovitz et al, 1996; Grimaud et al, 2001).

Nitrocellulose filter-binding assay

4.5S RNA was obtained by in vitro run-off T7-mediated transcription of plasmid pGEMTffs after cleavage with NsiI restriction endonuclease. 32P-labelled ffs (0.15 nM) was incubated with various concentrations of different forms of Ffh, that is, oxidized Ffh (Ffhox), MsrA-treated Ffhox, MsrB-treated Ffhox and MsrA/MsrB-treated Ffhox, in 10 μl of a buffer containing 10 mM Tris–HCl (pH 7.4), 10 mM MgCl2, 300 mM KCl, 1 unit poly (dIdC) for 1 min at room temperature. Mixtures were then diluted with 0.5 ml of wash buffer (20 mM Tris–HCl (pH 7.4), 10 mM MgCl2, 100 mM NaCl, 10% glycerol). The resulting solutions were immediately passed through nitrocellulose filters (Millipore 0.45 μm HA), which were washed with 1.5 ml of wash buffer as described previously (Suzuma et al, 1999). The levels of radioactivity remaining on the nitrocellulose filters were determined by a liquid scintillation analyser (Packard 2200CA).

Preparation and detection of biotinylated fusion proteins

Detection of biotinylation fusion proteins was carried out as described previously (Tian et al, 2000). Cells were grown overnight at 37°C in M9 medium with appropriate antibiotics. Overnight cultures were spun down, resuspended in the same medium supplemented with 2 nM biotin and kanamycin (100 μg/ml). The MalF–PSBT fusion was induced with 0.5 mM of IPTG, at an OD600 of 0.3. Whole-cell proteins were precipitated after cells reached an OD600 of 0.6 with 6% trichloroacetic acid and resuspended in SDS sample buffer. After electrophoresis, proteins were transferred to a nitrocellulose membrane. To detect biotinylated fusion proteins, streptavidin–horseradish peroxidase conjugate (Amersham Pharmacia) was used. To detect the expressed fusion protein, we used polyclonal anti-MalF-PhoA J (a gift from Jon Beckwith). The ECL system (Amersham Pharmacia) was used for the subsequent detection following the manufacturer's instruction.

Pulse-chase experiments and immunoprecipitation

MC4100 and BE53 carrying the plasmid pFAIII (Isnard et al, 1994) were grown overnight at 37°C in minimal M9 medium added with tetracyclin, glucose (0.2%) and all amino acids, excluding Met and cysteine. Overnight cultures were used to inoculate the same medium (OD600 of 0.02). Cultures were grown with aeration at 37°C until an OD600 of 0.4 was attained. Aliquots (3 ml) were removed from each culture and labelled for 5 min with 50 μCi/ml of Redivue promix [35S] Met. Then, 1.5 mg of cold Met was added and chased for various times (0 s, 30 s, 2 min). Whole-cell proteins were precipitated with 5% trichloroacetic acid. Immunoprecipitation of TolB and Pal was performed as described previously (Kumamoto and Gannon, 1988). Precursor forms were visualized by treating cultures with sodium azide (25 mM) for 5 min prior to the pulse.

Stability of Ffh in msrA msrB mutant

Cultures of E. coli MC4100 and BE53 (msrA msrB) were grown at 37°C in M9 medium with casamino acids. At an OD600 of 0.3, aliquots (1 ml) were removed and chloramphenicol was added (2.5 mg/ml) to the remaining cultures. Aliquots (1 ml) were removed after 1, 3, 6 and 24 h and analyzed by SDS–PAGE. Protein levels were analyzed by Western blotting by using an anti-Ffh antiserum. The ECL film was scanned, and band intensities were quantified using Kodak image station 440CF.

MALDI/TOF-MS analysis of Ffh

Protein bands were excised from the gel, destained and digested with a solution of modified trypsin (10 ng/μl, sequence grade, Promega, Charbonnieres, France) in 25 mM NH4HCO3 (Shevchenko et al, 1996). The mixture was incubated at 37°C for 6 h. ZipTip pipette tips containing C18 reversed phase are used for desalting and concentrating peptides according to the manufacturer's instructions (Millipore corporation, Bedford, MA, USA) and eluted with 70% acetonitrile in water, 0.1% trifluoroacetic acid (v/v). Peptide solutions were mixed with an equal volume of α-cyano-4-hydroxycinnamic acid matrix solution (Sigma, Saint Louis, MO, USA), 10 mg/ml 70% acetonitrile in water, 0.1% TFA and applied to a sample plate for MALDI-MS. Isotope 12C masses were determined in the positive ion reflector mode with MALDI/TOF-MS spectrometer Voyager DE-RP (Applied Biosystems, CA, USA) and internal calibration.

Acknowledgments

Thanks are due to the FB group for fruitful discussions. We thank G Victorero (TAGC, CIML, Marseille) for facilities, J Beckwith, M Müller, LF Wu, R Lloubès for strains and antibodies, and E Bouveret for insightful discussion and careful reading of the manuscript. This work was supported by grants from Université Aix-Marseille II, from CNRS (Programmes Protéome, PICS CNRS-Maroc) and from Conseil Régional PACA. BE was a recipient of fellowships from Ministère de l'Education Nationale and from Fondation de la Recherche Médicale.

References

  1. Batey RT, Rambo RP, Lucast L, Rha B, Doudna JA (2000) Crystal structure of the ribonucleoprotein core of the signal recognition particle. Science 287: 1232–1239 [DOI] [PubMed] [Google Scholar]
  2. Bernstein HD, Poritz MA, Strub K, Hoben PJ, Brenner S, Walter P (1989) Model for signal sequence recognition from amino-acid sequence of 54 K subunit of signal recognition particle. Nature 340: 482–486 [DOI] [PubMed] [Google Scholar]
  3. Brot N, Weissbach L, Werth J, Weissbach H (1981) Enzymatic reduction of protein-bound methionine sulfoxide. Proc Natl Acad Sci USA 78: 2155–2158 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brown S, Fournier MJ (1984) The 4.5 S RNA gene of Escherichia coli is essential for cell growth. J Mol Biol 178: 533–550 [DOI] [PubMed] [Google Scholar]
  5. Datsenko KA, Wanner BL (2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci USA 97: 6640–6645 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Grimaud R, Ezraty B, Mitchell JK, Lafitte D, Briand C, Derrick PJ, Barras F (2001) Repair of oxidized proteins. Identification of a new methionine sulfoxide reductase. J Biol Chem 276: 48915–48920 [DOI] [PubMed] [Google Scholar]
  7. Hassouni M, Chambost JP, Expert D, Van Gijsegem F, Barras F (1999) The minimal gene set member msrA, encoding peptide methionine sulfoxide reductase, is a virulence determinant of the plant pathogen Erwinia chrysanthemi. Proc Natl Acad Sci USA 96: 887–892 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Herskovits AA, Bochkareva ES, Bibi E (2000) New prospects in studying the bacterial signal recognition particle pathway. Mol Microbiol 38: 927–939 [DOI] [PubMed] [Google Scholar]
  9. Isnard M, Rigal A, Lazzaroni JC, Lazdunski C, Lloubes R (1994) Maturation and localization of the TolB protein required for colicin import. J Bacteriol 176: 6392–6396 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Jensen CG, Pedersen S (1994) Concentrations of 4.5S RNA and Ffh protein in Escherichia coli: the stability of Ffh protein is dependent on the concentration of 4.5S RNA. J Bacteriol 176: 7148–7154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Keenan RJ, Freymann DM, Stroud RM, Walter P (2001) The signal recognition particle. Annu Rev Biochem 70: 755–775 [DOI] [PubMed] [Google Scholar]
  12. Keenan RJ, Freymann DM, Walter P, Stroud RM (1998) Crystal structure of the signal sequence binding subunit of the signal recognition particle. Cell 94: 181–191 [DOI] [PubMed] [Google Scholar]
  13. Kryukov GV, Kumar RA, Koc A, Sun Z, Gladyshev VN (2002) Selenoprotein R is a zinc-containing stereo-specific methionine sulfoxide reductase. Proc Natl Acad Sci USA 99: 4245–4250 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kumamoto CA, Gannon PM (1988) Effects of Escherichia coli secB mutations on pre-maltose binding protein conformation and export kinetics. J Biol Chem 263: 11554–11558 [PubMed] [Google Scholar]
  15. Levine RL, Mosoni L, Berlett BS, Stadtman ER (1996) Methionine residues as endogenous antioxidants in proteins. Proc Natl Acad Sci USA 93: 15036–15040 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Luirink J, ten Hagen-Jongman CM, van der Weijden CC, Oudega B, High S, Dobberstein B, Kusters R (1994) An alternative protein targeting pathway in Escherichia coli: studies on the role of FtsY. EMBO J 13: 2289–2296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. MacFarlane J, Muller M (1995) The functional integration of a polytopic membrane protein of Escherichia coli is dependent on the bacterial signal-recognition particle. Eur J Biochem 233: 766–771 [DOI] [PubMed] [Google Scholar]
  18. Moskovitz J, Bar-Noy S, Williams WM, Requena J, Berlett BS, Stadtman ER (2001) Methionine sulfoxide reductase (MsrA) is a regulator of antioxidant defense and lifespan in mammals. Proc Natl Acad Sci USA 98: 12920–12925 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Moskovitz J, Flescher E, Berlett BS, Azare J, Poston JM, Stadman ER (1998) Overexpression of peptide-methionine sulfoxide reductase in Saccharomyces cerevisiae and human T cells provides them with high resistance to oxidative stress. Proc Natl Acad Sci USA 95: 14071–14075 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Moskovitz J, Rahman MA, Strassman J, Yancey SO, Kushner SR, Brot N, Weissbach H (1995) Escherichia coli peptide methionine sulfoxide reductase gene: regulation of expression and role in protecting against oxidative damage. J Bacteriol 177: 502–507 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Moskovitz J, Weissbach H, Brot N (1996) Cloning the expression of a mammalian gene involved in the reduction of methionine sulfoxide residues in proteins. Proc Natl Acad Sci USA 93: 2095–2098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Murphy KC (1998) Use of bacteriophage lambda recombination functions to promote gene replacement in Escherichia coli. J Bacteriol 180: 2063–2071 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Olry A, Boschi-Muller S, Marraud M, Sanglier-Cianferani S, Van Dorsselear A, Branlant G (2002) Characterization of the methionine sulfoxide reductase activities of PILB, a probable virulence factor from Neisseria meningitidis. J Biol Chem 14: 12016–12022 [DOI] [PubMed] [Google Scholar]
  24. Park SK, Jiang F, Dalbey RE, Phillips GJ (2002) Functional analysis of the signal recognition particle in Escherichia coli by characterization of a temperature-sensitive ffh mutant. J Bacteriol 184: 2642–2653 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Phillips GJ, Silhavy TJ (1992) Substitution of fifty four homologue (Ffh) in Escherichia coli with the mammalian 54-kDa protein of signal-recognition particle. Nature 359: 744–746 [DOI] [PubMed] [Google Scholar]
  26. Poritz MA, Bernstein HD, Strub K, Zopf D, Wilhelm H, Walter P (1990) An E. coli ribonucleoprotein containing 4.5S RNA resembles mammalian signal recognition particle. Science 250: 1111–1117 [DOI] [PubMed] [Google Scholar]
  27. Römisch K, Webb J, Lingelbach K, Gausepohl H, Dobberstein B (1990) The 54-kD protein of signal recognition particle contains a methionine-rich RNA binding domain. J Cell Biol 111: 1793–1802 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ruan H, Tang XD, Chen ML, Joiner ML, Sun G, Brot N, Weissbach H, Heinemann SH, Iverson L, Wu CF, Hoshi T, Chen ML, Joiner MA, Heinemann SH (2002) High-quality life extension by the enzyme peptide methionine sulfoxide reductase. Proc Natl Acad Sci USA 99: 2748–2753 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Sharov VS, Ferrington DA, Squier TC, Schöneich C (1999) Diastereoselective reduction of protein-bound methionine sulfoxide by methionine sulfoxide reductase. FEBS Lett 455: 247–250 [DOI] [PubMed] [Google Scholar]
  30. Shevchenko A, Jensen ON, Podtelejnikov AV, Sagliocco F, Wilm M, Vorm O, Mortensen P, Boucherie H, Mann M (1996) Linking genome and proteome by mass spectrometry: large-scale identification of yeast proteins from two dimensional gels. Proc Natl Acad Sci USA 93: 14440–14445 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Siegele DA, Hu JC (1997) Gene expression from plasmids containing the araBAD promoter at subsaturating inducer concentrations represents mixed populations. Proc Natl Acad Sci USA 94: 8168–8172 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Stadtman ER (1992) Protein oxidation and aging. Science 257: 1220–1224 [DOI] [PubMed] [Google Scholar]
  33. Stadtman ER, Levine RL (2000) Protein oxidation. Ann NY Acad Sci 899: 191–208 [DOI] [PubMed] [Google Scholar]
  34. Suzuma S, Hayashi K, Nakamura K, Yamane K (1999) Analysis of Escherichia coli 4.5S RNA binding affinity to Ffh and EF-G. FEMS Microbiol Lett 180: 271–277 [DOI] [PubMed] [Google Scholar]
  35. Tian H, Boyd D, Beckwith J (2000) A mutant hunt for defects in membrane protein assembly yields mutations affecting the bacterial signal recognition particle and Sec machinery. Proc Natl Acad Sci USA 97: 4730–4735 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Ulbrandt ND, Newitt JA, Bernstein HD (1997) The E. coli signal recognition particle is required for the insertion of a subset of inner membrane proteins. Cell 88: 187–196 [DOI] [PubMed] [Google Scholar]
  37. Vogt W (1995) Oxidation of methionyl residues in proteins: tools, targets, and reversal. Free Radic Biol Med 18: 93–105 [DOI] [PubMed] [Google Scholar]
  38. Weissbach H, Etienne F, Hoshi T, Heinemann SH, Lowther WT, Matthews B, St John G, Nathan C, Brot N (2002) Peptide methionine sulfoxide reductase: structure, mechanism of action, and biological function. Arch Biochem Biophys 397: 172–178 [DOI] [PubMed] [Google Scholar]
  39. Wizemann TM, Moskovitz J, Pearce BJ, Cundell D, Arvidson CG, So M, Weissbach H, Brot N, Masure HR (1996) Peptide methionine sulfoxide reductase contributes to the maintenance of adhesins in three major pathogens. Proc Natl Acad Sci USA 93: 7985–7990 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Zopf D, Bernstein HD, Johnson AE, Walter P (1990) The methionine-rich domain of the 54 kd protein subunit of the signal recognition particle contains an RNA binding site and can be crosslinked to a signal sequence. EMBO J 13: 4511–4517 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The EMBO Journal are provided here courtesy of Nature Publishing Group

RESOURCES