Abstract: Signal Recognition Particle (SRP) is a cytosolic particle that is involved in cotranslational transport of presecretory proteins into the endoplasmic reticulum. The Alu domain of SRP, comprising the 9 and 14 kDa subunits and part of the 7S RNA, arrests translation. Here we show that SRP14 inhibits translation and employs a highly charged carboxy terminal oligopeptide in doing so. Furthermore, we observed that incubation of SRP14 with ribosomes leads to displacement of elongation factor eEF2. Thus SRP mediated arrest of translation involves competition of SRP14 with elongation factor eEF2 on the ribosome.
INTRODUCTION
Signal Recognition Particle (SRP) is a cytosolic ribonucleoprotein particle that mediates cotranslational transport of presecretory proteins into the Endoplasmic Reticulum (ER) (Wild et al., 2004). SRP is a cigar shaped particle and comprises six proteinaceous subunits and a 7S RNA (Wild et al., 2004; Siegel and Walter, 1988). Through its 54 kDa subunit (SRP54) SRP recognizes signal peptides of nascent presecretory polypeptides as they emerge from the ribosomal exit tunnel (Kurchalia et al., 1986; Krieg et al., 1986). Through its 9 plus 14 kDa subunits (SRP9, SRP14) and the Alu domain of the 7S RNA SRP arrests elongation until the ribosome/nascent chain-complex has docked to the ER (Siegel and Walter, 1988; Meyer et al., 1982). Although the crystal structure of the SRP9/14 heterodimer had been determined the molecular mechanism of translational arrest remained elusive (Birse et al., 1997; Weichenrieder et al., 2000). A recent structural analysis of SRP-ribosome complexes by cryo-electron microscopy and single particle reconstruction suggested that SRP employs elongation factor mimicry by the Alu domain of SRP, i.e., the SRP9/14 heterodimer plus the Alu domain of the 7S RNA, as a strategy for translational arrest (Halic et al., 2004). However, this analysis did not identify the crucial component within the Alu domain.
ERj1p (originally termed Mtj1p) is an ER membrane protein
with an ER lumenal J-domain and a cytosolic domain that is in contact
with the exit site of translating ribosomes on the ER surface (Dudek et
al., 2002, 2005; Blau et al., 2005). The cytosolic domain of
ERj1p as well as a highly charged oligopeptide that is contained in the
amino terminus of this cytosolic domain (Fig. 1) have
the ability to inhibit translation by an allosteric effect (Dudek et
al., 2002, 2005). Analysis of the primary structures of SRP9 and SRP14
with the protean option of the DNASTAR sequence analysis software identified
an oligopeptide within yeast and mammalian SRP14 that is very similar
to the oligopeptide that was characterized as being responsible for the
ability of ERj1p to inhibit protein synthesis (Fig. 1).
According to the software predictions for the corresponding polypeptides
these peptides additionally have in common a high surface probability
(data not shown). The crystal structure of the mouse SRP9/14 heterodimer,
obtained in the absence of RNA, confirms the surface location of the oligopeptide
(Birse et al., 1997). Strikingly, it had previously been observed
for both yeast and mouse SRP that deletion of the 29 (i.e., amino acid
residues 118 through 146) and 20 (i.e., amino acid residues 91 through
110), respectively, carboxy terminal amino acid residues of SRP14 that
include the highly charged oligopeptide leads to particles that do not
have the ability to arrest translation (Thomas et al., 1997; Mason
et al., 2000). Similarly, deletion of 21 amino terminal amino acid
residues of the cytosolic domain of ERj1p led to loss of the ability to
inhibit translation ((Dudek et al., 2005). Therefore, we proposed
that the employment of specifically targeted positively charged oligopeptides
that interact with specific domains of rRNA is a unifying theme for a
number of ribosomal ligands (Dudek et al., 2005). Here, this hypothesis
was evaluated with respect to SRP14.
| Fig. 1: | Sequences of highly charged peptides that are present in SRP14 (Birse et al., 1997) and ERj1p (Dudek et al., 2002), respectively, as well as in various additional ribosomal effectors. The organisms and the positions of the respective oligopeptides within the protein or precursor protein are indicated. In the case of yeast and murine SRP14 the carboxy terminal amino acid residues that were absent from the deletion mutants (Dudek et al., 2002; Thomas et al., 1997) are presented |
RESULTS
Two recombinant hybrid proteins that are related to mouse SRP14, GST-SRP14 and GST-SRP14ΔC (lacking amino acid residues 91 through 110) were analyzed with respect to their ability to inhibit protein synthesis in reticulocyte lysate. Synthesis of preprolactin was sensitive towards GST-SRP14 (Fig. 2a). The inhibitory effect was specific since its extent correlated reciprocally with the ribosome content of the reticulocyte lysate (Fig. 2b) as we had observed for ERj1p (Dudek et al., 2002). In comparison, GST-SRP14ΔC and GST did hardly or not at all affect protein synthesis (Fig. 2a), thereby confirming the crucial role of the highly charged carboxy terminal oligopeptide. As we also had observed for ERj1p (Dudek et al., 2002), the inhibitory effect of SRP14 was not specific for presecretory proteins, i.e., synthesis of firefly luciferase showed a similar phenotype (data not shown). Thus SRP14 is directly involved in or even responsible for the ability of SRP to arrest translation.
Therefore, we addressed the question of how SRP14 affects translation. When GST-SRP14 was incubated with ribosomes and, subsequently, the ribosomes were reisolated by centrifugation we observed that binding of GST-SRP14 to ribosomes (Fig. 2c) led to the exclusive displacement of a formerly ribosome associated protein (data not shown). In subsequent experiments this protein was characterized as elongation factor eEF2 by amino terminal sequence analysis (data not shown) and by semi-quantitative immunodetection of eEF2 (Fig. 2d). Furthermore, when GST-SRP14ΔC and GST were analyzed in this respect, GST-SRP14ΔC and GST did hardly or not at all bind to ribosomes (Fig. 2c) and there was no displacement of eEF2 (Fig. 2d). Thus the inhibitory effect of SRP14 on translation is due to competition with eEF2 on the ribosome.
DISCUSSION
First of all, our data suggests that SRP14 is responsible for the inhibition of translation by SRP and that SRP inhibits translation by competition of SRP14 with elongation factors for binding to ribosomes. Second, these results support our earlier notion that the employment of specifically targeted positively charged oligopeptides that interact with specific domains of rRNA is a unifying theme for a number of effectors of ribosomes (Dudek et al., 2005). With respect to SRP14, this straightforward interpretation of the SRP14 deletion mutants of SRP had originally been dismissed on the basis of two arguments (Thomas et al., 1997). Firstly, the deletion affected the tertiary structure of the Alu portion of 7S RNA (Thomas et al., 1997). Secondly, according to the crystal structure of the mouse SRP9/14 heterodimer, I) Gly93 and Lys95 of SRP14 form hydrogen bonds with amino acid residues of SRP9, and ii) Leu94 of SRP14 is part of the hydrophobic core between the two proteins (Birse et al., 1997). We favor the idea that the crystal structure reflects a state of the SRP9/14 heterodimer that would be inactive in translational arrest and that in the case of SRP14 the binding of a signal peptide that is present in a nascent presecretory protein as it emerges from the ribosome to SRP54 affects the structure and/or accessibility of the charged oligopeptide and modulates
| Fig. 2: | Effects of SRP14 on ribosomes. GST and the fusion proteins of SRP14
and SRP14ΔC with GST on the amino terminus were produced in E.
coli and purified as described for ERj1p (Dudek et al.,
2002). (a,b) Preprolactin was synthesized in rabbit reticulocyte lysate
(a) or fractionated lysate (b) in the presence of [35S]
methionine as described by Dudek et al. (2002). The translation
reactions contained either buffer, GST or GST fusion proteins of SRP14
or SRP14ΔC (final concentration: 2 μM). After incubation
for various times at 30 °C, the translation reactions were subjected
to SDS-PAGE. The dried gels were analyzed in a phosphorimager. The
amount of full-length radiolabeled preprolactin that was produced
at the end of the buffer control reaction was set to 100 percent.
We note that the concentration of ribosomes and eEF2 in this reticulocyte
based translation reaction is about 150 nM and 300 nM, respectively.
(c,d) GST or GST fusion proteins of SRP14 or SRP14ΔC (final concentration:
1 μM) were diluted into buffer and incubated in the presence
of ribosomes (final concentration: 50 nM), derived from dog pancreas,
for 15 min at 30 °C as described Dudek et al. (2005). Subsequently
the mixture was layered onto a cushion and subjected to centrifugation
for 90 min at 100,000 rpm and 2 °C (Beckman TLA 120.2 rotor). The
ribosomal pellets (p) and postribosomal supernatants (s) were subjected
to SDS/PAGE and subsequent protein staining (c) or subsequent electroblotting
plus immunodetection (d) and densitometric analysis |
The inhibitory activity of SRP14. Accordingly, triggered by binding of a signal peptide to SRP54 the charged oligopeptide within SRP14 could move out of the contact site to SRP9, interact with the ribosome at the eEF2 binding site, and switch off elongation by competing with eEF2. This view is perfectly consistent with the structure of SRP-ribosome complexes (Halic et al., 2004). In the SRP14 mutant SRP particles this conformational change, rather than the absence of the carboxy terminal amino acid residues may have affected the conformation of the Alu portion of 7S RNA. This view is in fact supported by structural data that were obtained for the human SRP9/14 heterodimer in the presence of parts of the 7S RNA (Weichenrieder et al., 2000). Based on these data it was hypothesized that a reversible switch in the folding of the Alu domain of SRP, i.e., the SRP9/14 heterodimer plus its associated Alu portion of the 7S RNA, may permit alternating onset or release of elongation arrest.
ACKNOWLEDGMENTS
This work was supported by the DFG, the Universität des Saarlandes, and the Fonds der Chemischen Industrie.