![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 18, Issue 7, 2636-2645, July 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA 91125
Submitted January 17, 2007;
Revised April 18, 2007;
Accepted April 20, 2007
Monitoring Editor: Reid Gilmore
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Keenan et al., 2001). The functional core of SRP is the SRP54 protein (called Ffh in bacteria) in complex with an SRP RNA, which recognizes the cargo protein and interacts with the SR (called FtsY in bacteria). The protein targeting reaction is regulated by the guanosine-5'-triphosphate (GTP)-binding domains in both SRP54 and SR. SRP recognizes the signal sequence on nascent polypeptides that emerge from a translating ribosome (Walter et al., 1981). The ribosome
nascent chain complex is delivered to the membrane via the interaction of SRP with SR when both proteins are bound with GTP (Gilmore et al., 1982a,b). Upon arrival at the membrane, SRP releases the cargo protein to a protein conducting channel embedded in the membrane (Simon and Blobel, 1991; Gorlich et al., 1992), where the nascent protein is either integrated into the membrane or translocated across the membrane to enter the secretory pathway. GTP hydrolysis is stimulated in the SRPSR complex, which then drives disassembly and recycling of SRP and SR (Connolly et al., 1991).
The SRP and SR GTPases comprise a unique subgroup in the GTPase superfamily (Keenan et al., 2001). Both proteins have a GTPase, "G" domain that shares homology with the classical Ras GTPase fold (Freymann et al., 1997; Montoya et al., 1997). In addition, the SRP-type GTPases contain an N-terminal four-helix bundle, the "N" domain, that packs tightly against the G domain. The G and N domains form a structural and functional unit called the NG domain. Unlike classical signaling GTPases that undergo large conformational changes depending on whether GTP or guanosine- 5'-diphosphate (GDP) is bound, the structures of these GTPases are similar regardless of which nucleotide is bound (Freymann et al., 1999; Montoya et al., 1997; Padmanabhan and Freymann, 2001; Gawronski-Salerno et al., 2006; Reyes and Stroud, unpublished data). Substantial conformational changes occur only when the two GTPases form a complex with one another (Egea et al., 2004; Focia et al., 2004). Most notably, the G and N domains readjust their relative positions such that the N domains of both proteins move closer to the dimer interface and form additional interface contacts to stabilize the complex.
The importance of this N–G domain rearrangement is supported by biochemical analyses. Many mutations at this interface disrupt SRP–SR complex formation and protein targeting (Lu et al., 2001). Interestingly, unlike classical GTPases, free FtsY displays little discrimination between GTP and noncognate nucleotides. In contrast, FtsY acquires substantial nucleotide specificity only when it binds SRP. These results have led to the proposal that during complex formation, FtsY changes from a nondiscriminative, "open" state to a "closed" state in which specific interactions between GTP and active site residues are established (Shan and Walter, 2003). Consistent with these observations, the crystal structure showed that, upon complex formation, the rearrangement at the N–G domain interface brings the nucleotide specificity determinant Asp449 closer to the bound GTP and within hydrogen bonding distance with the amino groups of the guanine ring (Egea et al., 2004). Thus, the N–G domain rearrangement is primarily responsible for the open 
closed conformational change that occurs during SRP–SR complex formation and precisely aligns active site residues with respect to the bound GTP.
The unique structural features of the SRP subgroup of GTPases confer upon them many characteristics that are distinct from canonical GTPases. Most importantly, SRP-type GTPases bind nucleotides much weaker than signaling GTPases and release nucleotides quickly (Moser et al., 1997; Jagath et al., 1998, 2000; Peluso et al., 2001). Therefore, they do not use nucleotide exchange factors to facilitate the conversion from the GDP- to the GTP-bound form. These GTPases also do not use external GTPase-activating proteins; instead, SRP and SR reciprocally activate one another upon complex formation (Powers and Walter, 1995).
A third unique feature of the GTPases engaged in the SRP pathway is the requirement for a universally conserved SRP RNA. Mammalian SRP is a cytosolic ribonucleoprotein complex that consists of six polypeptides and a 7S SRP RNA molecule. Besides SRP54, the other protein components are not conserved, whereas the SRP RNA has been shown to play an indispensable role in protein targeting in all three kingdoms of life. In early biochemical studies on the mammalian SRP, the SRP RNA seemed to be nothing more than a scaffold that holds all the SRP proteins together in a complex (Walter and Blobel, 1982, 1983). The finding that bacteria contain a much simpler SRP, made up solely of a complex of Ffh and the 4.5S SRP RNA, was therefore intriguing. This smaller RNA contains the most phylogenetically conserved region of the SRP RNA, domain IV, which is likely to have been maintained for functional purposes (Poritz et al., 1988; Struck et al., 1988). Subsequently, kinetic analyses of the role of the 4.5S SRP RNA on the GTPase cycles of Ffh and FtsY showed that a major role of this RNA is to accelerate complex formation between the two GTPases. In the absence of the SRP RNA, Ffh–FtsY association is extremely slow, with a rate constant of 5 x 103 M–1 s–1. The SRP RNA accelerates their association kinetics by 400-fold, to a rate that can allow the SRP and SRP receptor to adequately carry out their biological functions, thus accounting for the indispensable role of the SRP RNA in the bacterial, archeal, and eukaryotic SRP pathways.
A novel SRP targeting pathway was discovered in the chloroplast (Schuenemann et al., 1998). cpSRP54 and cpFtsY are the chloroplast homologues of SRP and SR GTPases, respectively (Franklin and Hoffman, 1993; Li et al., 1995; Tu et al., 1999). cpSRP54 recognizes its cargo, the light-harvesting chlorophyll-binding proteins (LHCP), via a protein adaptor cpSRP43 (Tu et al., 2000). Together, cpSRP54 and cpSRP43 deliver the cargo protein from the stroma to the thylakoid membrane via the GTP-dependent interaction between cpSRP54 and cpFtsY (Tu et al., 1999). Surprisingly, the otherwise universally conserved SRP RNA has not been found to date in the chloroplast SRP system. To rationalize the absence of the SRP RNA, we characterized the kinetic and thermodynamic features of the GTPase cycles of cpSRP54 and cpFtsY. We found that, unlike their bacterial and mammalian homologues, the chloroplast SRP and SR GTPases can efficiently interact with one another by themselves. This helps explain why the cpSRP pathway could bypass the requirement for an SRP RNA.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Two additional chromatographic steps (Superdex 75 and monoQ [GE Healthcare]) were added to remove contaminating GTPases. Mutant cpFtsY(D283N) was constructed using the QuikChange procedure (Stratagene, La Jolla, CA), and it was expressed and purified by the same procedure as that for wild-type cpFtsY.
Kinetics
All reactions were carried out at 25°C in assay buffer [50 mM KHEPES, pH 7.5, 150 mM KOAc, 2 mM Mg(OAc)2, 2 mM dithiothreitol, and 0.01% Nikkol]. GTP hydrolysis reactions were followed and analyzed as described previously (Peluso et al., 2001). The general procedures for characterizing the basal and stimulated GTPase reactions between SRP and SR have been described in detail previously (Peluso et al., 2001; Shan and Walter, 2003, 2005), and they are summarized briefly here. The justification for how each microscopic rate constant was derived from these measurements is provided in Supplemental Material.
Basal GTPase or XTPase activities of cpSRP54, cpFtsY, and cpFtsY(D283N) were measured in single turnover reactions as described previously ([GTP] << [E]; Peluso et al., 2001). The dependence of the observed rate constant (kobsd) on protein concentration were fit to Eq. 1,
 | (1) |
The nucleotide affinities of the GTPases were determined using several independent methods. The GTP affinities for cpSRP54 and cpFtsY and the XTP (xanthosine-5'-triphosphate) affinity for cpFtsY(D283N) were obtained from the K1/2 values from fits of the basal GTPase or XTPase reactions to Eq. 1. Because the chemical step is rate limiting for the basal GTPase and XTPase reactions, K1/2 is equal to Kd, the dissociation constant of the nucleotide. The affinities of GDP, 5'-guanylylimido-diphosphate (GppNHp), xanthosine-5'-diphosphate (XDP), and 5'-xanthylylimido-diphosphate (XppNHp) were determined using these nucleotides as inhibitors of the basal GTPase or XTPase reactions (Peluso et al., 2001). With subsaturating protein, the inhibition constant, Ki, is equal to Kd. Finally, the binding of nucleotides to the GTPases was determined directly by using fluorescent N-methyl-anthraniloyl (mant) derivatives of GTP, GDP, and XTP, as described below.
The reciprocally stimulated GTPase reaction between cpSRP54 and cpFtsY was determined in multiple turnover reactions ([GTP] >> [E]) in the presence of a small, fixed amount of cpSRP54 and varying concentrations of cpFtsY, using a GTP concentration that saturates both GTPase sites. The concentration dependence of the observed rate constant (kobsd) is fit to Eq. 2,
 | (2) |
The cpFtsY(D283N)-stimulated GTP hydrolysis from cpSRP54 was also determined in single turnover experiments. The hydrolysis of trace GTP* was monitored in the presence of subsaturating cpSRP54 and varying amounts of cpFtsY(D283N), with 25 µM XTP present to selectively occupy the active site of cpFtsY(D283N). Under these conditions, the third-order reaction GTP* + cpSRP54 + cpFtsY(D283N)·XTP products was followed. The reciprocal reaction, XTP* + cpFtsY(D283N) + cpSRP54·GTP products was determined using an analogous setup, except that the concentration of cpSRP54 was varied, and 25 µM GTP was present to selectively occupy the active site of cpSRP54. The data were fit to Eq. 1 above. Finally, first-order rate constants of the stimulated GTP and XTP hydrolysis reactions from the *GTP·cpSRP54cpFtsY(D283N)·XTP* complex were determined using high concentrations of both proteins (20–80 µM) in the presence of stoichiometric amounts of their respective nucleotides. The reaction time courses were monitored in a quench flow apparatus (Kintek, Pittsford, NY) and fit to a single exponential rate equation to obtain the first-order rate constants.
The effect of XTP on the reaction *GTP·cpSRP54 + cpFtsY(D283N)·GTP* products was determined in the presence of subsaturating concentrations of both proteins and a high concentration of GTP (200 µM) to saturate both active sites. The XTP concentration dependence was fit to Eq. 3,
 | (3) |
is the apparent inhibition constant of XTP determined from this experiment. K is related to the dissociation constant of XTP by Eq. 4,
 | (4) |
and K
are the dissociation constants of XTP and GTP for cpFtsY(D283N), respectively.
Fluorescence
All fluorescence measurements were conducted at 25°C using the single photon counting Fluorolog 3–22 spectrofluorometer (Jobin Yvon, Edison, NJ). Fluorescence emission spectra of mant-derivatives of GTP, GDP, and XTP were acquired using an excitation wavelength of 356 nm. Nucleotide binding affinities were determined by recording the change in fluorescence intensity at 445 nm in the presence of 0.4 to 1 µM mant-nucleotides and increasing concentrations of cpSRP54, cpFtsY, or cpFtsY(D283N). The data were fit to Eq. 5,
 | (5) |
The rate constants for dissociation of mant-GTP and mant-GDP were determined using a pulse-chase experiment as described previously (Jagath et al., 1998). The time course for decay of fluorescence was followed in a stopped flow apparatus (Applied Photophysics, Surrey, United Kingdom) and fit to single exponential functions to obtain the dissociation rate constants.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
Interaction between cpSRP54 and cpFtsY Is Much More Efficient than Classical SRP Systems
In classical SRP systems, complex formation between the SRP and SR GTPases is very slow, and is accelerated 400-fold by the SRP RNA (Peluso et al., 2000, 2001). Once a complex is formed, SRP and SR stimulate each other's GTPase activity, and the rate of this stimulated GTPase reaction within the complex is also accelerated 5- to 10-fold by the SRP RNA (Peluso et al., 2001). Because no SRP RNA has been found in the chloroplast SRP system, we asked whether and how efficiently cpSRP54 and cpFtsY can interact with and activate each other in the absence of an SRP RNA.
To this end, we determined the rate of stimulated GTP hydrolysis reaction in the presence of both cpSRP54 and cpFtsY; GTPase activation in the cpSRP54cpFtsY complex provides a means to monitor complex formation between the two GTPases (Peluso et al., 2000, 2001). To our surprise, cpSRP54 and cpFtsY interact with each other efficiently even in the absence of an SRP RNA (Figure 4, ). The slope of the initial linear portion of the protein concentration dependence, which represents the rate constant of the reaction GTP·cpSRP54 + cpFtsY·GTP products (kcat/Km), is 
400-fold faster than that of the corresponding reaction between the Escherichia coli GTPases in the absence of the SRP RNA (Figure 4, vs. 
). Indeed, this rate constant matches that of the E. coli GTPases in the presence of the 4.5S SRP RNA (Figure 4, 
). The rate constant at saturating protein concentrations, which represents the rate of GTP hydrolysis within the GTP·cpSRP54cpFtsY·GTP complex, is also identical between the chloroplast and the E. coli GTPases in the presence of the SRP RNA ( vs. ), and eightfold faster than that of the E. coli GTPases without the RNA bound ().
|
products (both in the presence and absence of SRP RNA) (Peluso et al., 2001). Therefore, kcat/Km is equal to the association rate constant between the two GTPases. If this were also true for the cpSRP54 and cpFtsY, then the association between cpSRP54 and cpFtsY would be 400-fold faster than their E. coli homologues. Alternatively, kcat/Km is limited by the chemical step instead of complex formation for the chloroplast GTPases. If this were true, then the difference in association rates between the chloroplast and E. coli GTPases would be even greater. Thus, the results in Figure 4 demonstrate that complex formation between the chloroplast SRP and SR GTPases is much more efficient than that of their bacterial and mammalian homologues and therefore do not need the help from an SRP RNA.
cpFtsY Exhibits High Nucleotide Specificity
Association between bacterial SRP and SR GTPases is slow, presumably because significant domain rearrangements are required to form a stable complex, including a change from the open to the closed conformation that is manifested functionally as an increase in the nucleotide specificity of the E. coli FtsY (Shan and Walter, 2003; see Introduction). We hypothesized that the chloroplast SRP GTPases are preorganized in the closed conformation even in the absence of their binding partner, thus reducing the cost for the open closed rearrangement and resulting in a faster rate of protein–protein interaction.
A prediction from this model is that cpFtsY can effectively discriminate between cognate and noncognate nucleotides by itself without the help from cpSRP54. To test this idea, we mutated the conserved specificity determinant Asp283 to an asparagine. This mutation converts many GTPases to XTP-specific proteins by swapping the hydrogen bond between the carboxylate oxygen of Asp and the exocyclic amino group of the guanine ring (Hwang and Miller, 1987; Weijland et al., 1994; Zhong et al., 1995; Bishop et al., 2000). As predicted, wild-type cpFtsY preferentially hydrolyzes GTP. The rate constant of the reaction: GTP* + FtsY GDP + Pi* is 37-fold faster than that of mutant cpFtsY(D283N) (Figure 5A). Similarly, mutant cpFtsY(D283N) hydrolyzes XTP much faster than wild-type cpFtsY (Figure 5B). In contrast, E. coli FtsY exhibits no more than a fourfold difference between wild-type and mutant GTPases in the hydrolysis rates of either nucleotide (Shan and Walter, 2003).
|
). Together, the results in this section show that, unlike its bacterial homologue, the active site of cpFtsY can specifically recognize GTP even in the absence of cpSRP54. This is consistent with the notion that free cpFtsY is already in the closed conformation and pre-organized to interact with cpSRP54.
GTPase Activation between cpSRP54 and cpFtsY Is Reciprocal but Asymmetric
The XTP-specific mutant cpFtsY(D283N) also allowed us to test whether cpSRP54 and cpFtsY reciprocally stimulate the GTPase activity of one another, as is the case for the bacterial system. If this were the case, XTP hydrolysis by cpFtsY(D283N) would be stimulated by cpSRP54, and, conversely, GTP hydrolysis by cpSRP54 would be stimulated by cpFtsY(D283N).
To examine the effect of cpFtsY(D283N) on GTP hydrolysis by cpSRP54, we measured the rate of GTP hydrolysis in the third-order reaction GTP* + cpSRP54 + D283N·XTP GDP + Pi*. As predicted, the rate of GTP hydrolysis is significantly stimulated by the presence of cpFtsY(D283N) (Figure 6A), consistent with the notion that cpFtsY acts as the activating protein for cpSRP54. Analogously, the reciprocal reaction, XTP hydrolysis by cpFtsY(D283N), is significantly stimulated by the presence of cpSRP54 (Figure 6B; the third-order reaction: XTP* + D283N + cpSRP54·GTP XDP + Pi* was followed).
|
10-fold slower than that of XTP hydrolysis from cpFtsY(D283N) (cf. rates in Figure 6, A and B), raising the possibility that nucleotide hydrolyses from the two GTPase sites in the complex are not symmetric. To test this possibility, we formed the GTP·cpSRP54cpFtsY(D283N)·XTP complex by using high concentrations of both proteins and stoichiometric amounts of GTP and XTP, and directly measured the rate constants for hydrolysis of both GTP and XTP from this complex. As shown in Figure 6C, the rate constant for XTP hydrolysis is 3.7 min–1 (squares), over fourfold faster than the rate constant of 0.87 min–1 for GTP hydrolysis (circles). This represents only a lower limit for the difference in hydrolysis rates between the two active sites, because cpFtsY(D283N) bound with GTP is much more active in binding and activating cpSRP54 (see the next section), even though it preferentially binds XTP by itself. Thus, part of the GTP hydrolysis rate observed in Figure 6C (circles) is contributed by an alternative GTP·cpSRP54cpFtsY(D283N)·GTP complex. The actual difference between the hydrolysis rates from the two active sites is larger than that observed in Figure 6C and is closer to the 10-fold difference observed in Figure 6, A and B. This is because the experiments in Figure 6, A and B, measure the rate of third-order reactions, which is determined by the affinities of free cpSRP54 and cpFtsY(D283N) for their respective nucleotides as well as the rates at which GTP and XTP are hydrolyzed from the respective active sites in the complex. Because cpSRP54 and cpFtsY(D283N) exhibit similar affinities for GTP and XTP, respectively (Table 2), the 10-fold difference in the observed reaction rate (Figure 6, A and B) primarily reflects the difference in hydrolysis rate from the two active sites. Thus, like the classical SRP systems, cpSRP54 and cpFtsY act as reciprocal activating proteins for one another, yet unlike their bacterial homologues, nucleotide hydrolyses from the two active sites are asymmetric.
Mutant cpFtsY(D283N) Prefers GTP over XTP upon Complex Formation with cpSRP54
Another intriguing observation from the results in Figure 6C is that the rate constants of the stimulated GTPase and XTPase reactions from the GTP·cpSRP54cpFtsY(D283N)·XTP complex (0.87 and 3.7 min–1, respectively) are over 10-fold slower than that from the wild-type GTP·cpSRP54cpFtsY·GTP complex (Figure 4), even accounting for the fact that two GTP molecules are hydrolyzed in the wild-type complex. Therefore, we suspected that the D283N mutation or the replacement of GTP with XTP renders cpFtsY less active in binding and activating cpSRP54. This is reminiscent of the behavior of an XTP-specific mutant of the E. coli SRP GTPase, SRP(D251N), which is deficient in binding and activating FtsY in its XTP-bound form. Instead, mutant SRP(D251N) can better bind and activate FtsY when bound to the noncognate GTP (Shan and Walter, 2005).
To test whether this is also the case for mutant cpFtsY(D283N), we measured the rate constant for GTP hydrolysis from the GTP·cpSRP54cpFtsY(D283N)·GTP complex when cpFtsY(D283N) is forced to bind its noncognate nucleotide by using a high GTP concentration. When mutant cpFtsY(D283N) is bound with the noncognate GTP, the rate of stimulated GTP hydrolysis is much faster than when it is bound with the cognate XTP (Figure 7A, diamonds vs. squares). The rate constant at saturating protein concentration, which represents the rate constant for GTP hydrolysis from the GTP·cpSRP54 cpFtsY(D283N)·GTP complex, is comparable to that of the wild-type complex (Figure 7A; diamonds vs. circles), suggesting that the GTP·cpSRP54cpFtsY(D283N)·GTP complex achieves the same active conformation as the complex formed by the wild-type proteins. Because a fivefold higher concentration of mutant cpFtsY(D283N) than wild-type cpFtsY is required to reach saturation, complex formation is modestly compromised for GTP-bound cpFtsY(D283N) (Figure 7A; diamonds vs. circles). In contrast, no saturation is observed in the reaction with XTP-bound cpFtsY(D283N) up to 30 µM (squares), indicating that complex formation is significantly compromised when the mutant is bound with its cognate nucleotide. Thus, mutant cpFtsY(D283N) prefers the noncognate GTP over cognate XTP when it forms a complex with cpSRP54.
|
products. If cpFtsY(D283N) is less active in binding and activating the GTPase reaction of cpSRP54 when it is bound with cognate XTP than with noncognate GTP, then addition of XTP, which competes off the GTP bound at the active site of cpFtsY(D283N), should inhibit the stimulated GTPase reaction. As predicted, addition of XTP inhibits this stimulated reaction (Figure 7B). The observed inhibition constant for XTP is 9.0 µM, consistent with the expected value of 8.9 ± 0.9 µM given the affinities of mutant cpFtsY(D283N) for GTP and XTP and the GTP concentration used in this experiment (Eq. 4 in Materials and Methods). This strongly suggests that the binding of XTP to cpFtsY(D283N) is responsible for the observed inhibitory effect. Together, the results in this section show that although cpFtsY(D283N) by itself exhibits a specificity for XTP, this mutant prefers the noncognate GTP for interacting with and stimulating GTP hydrolysis from cpSRP54. Thus, Asp283 and/or the bound GTP play a much more important role than specifying the nucleotide preference of cpFtsY and likely participate in critical interface interactions with cpSRP54 in the cpSRP54cpFtsY complex. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Given these similarities, it is surprising that the otherwise universally conserved SRP RNA, which plays a crucial role in eukaryotic and prokaryotic SRP protein targeting pathways, is missing in the chloroplast SRP pathway. In E. coli, association between the SRP and SR GTPases is extremely slow, with a rate constant of 5 x 103 M–1 s–1 (Peluso et al., 2001). This slow association rate does not seem to be caused by the extended A-domain of E. coli FtsY, because Thermus aquaticus FtsY, which lacks an extended A-domain, also interacts with its binding partner very slowly (Shepotinovskaya and Freymann, 2001). At this rate and the in vivo concentration of these GTPases (nanomolar range), the association between the two GTPases will take hours to complete. The protein targeting reaction, however, occurs on a much faster time scale. A recent report shows that thioredoxin, a small protein with only 105 amino acids, is targeted by the SRP when an appropriate signal sequence is attached (Huber et al., 2005). Because it takes only 3–5 s for complete synthesis of the thioredoxin chain, SRP-dependent protein targeting must occur in <3 s. Thus, the slow interaction kinetics between the SRP and SR GTPases is inappropriate for the targeting reaction. The SRP RNA overcomes this problem by accelerating the association between the two GTPases 400-fold (Peluso et al., 2000, 2001). Another contribution of the SRP RNA is to increase the rate of GTP hydrolysis in the SRPSR complex by 5- to 10-fold (Peluso et al., 2001); GTP hydrolysis is known to drive disassembly and recycling of the SRP and SR (Wilson et al., 1988; Connolly et al., 1991). Here, we showed that cpSRP54 and cpFtsY can interact efficiently with each other even in the absence of an SRP RNA: their association rate is at least as fast as that of their E. coli homologues that contain the SRP RNA, and GTP hydrolysis from the cpSRP54cpFtsY complex also occurs at the same rate as the E. coli GTPase complex in the presence of the RNA. This helps explain how the chloroplast SRP system can bypass the requirement for the SRP RNA.
Why is the protein–protein interaction so efficient between the chloroplast GTPases? Interaction between the bacterial SRP and SR GTPases is slow, presumably due to the requirement for extensive conformational changes during complex formation. One of the important rearrangements is a repositioning of the N–G domain interface, which led to a change of the GTPase site from a floppy, nonspecific open state to a closed state in which active site interactions with the bound nucleotide are established (Shan and Walter, 2003). Thus, one possibility is that cpSRP54 and cpFtsY are preorganized into the closed conformation that is ready to interact with each other. The results herein strongly suggest that this is the case at least for cpFtsY. Free cpFtsY can specifically recognize its cognate nucleotide, in contrast to E. coli FtsY, which acquires nucleotide specificity only when it forms a complex with SRP. Furthermore, cpFtsY exhibits higher affinities for GTP and GDP than its bacterial homologues, with dissociation constants of 2–3 µM instead of 19–30 µM for E. coli FtsY. These observations strongly support the notion that free cpFtsY is preorganized in a closed conformation, and thus can interact with cpSRP54 without paying substantial energetic penalty to rearrange the relative position of the N and G domains. It remains to be seen whether cpSRP54 is similarly preorganized into the closed conformation before interaction with cpFtsY.
It seems that the SRP RNA has been evolved to accelerate the very inefficient interaction between the SRP and SR GTPases in classical SRP pathways. Although models are abundant (Peluso et al., 2000; Buskiewicz et al., 2005; Spanggord et al., 2005), the molecular mechanism by which the SRP RNA acts as a catalyst to accelerate both the formation and disassembly of the SRPSR complex is still poorly understood. It is possible that in the transition state for complex assembly, the SRP RNA may provide a transient tether that facilitates the rearrangement of one or both GTPases into the closed conformation; alternatively, the RNA and the chloroplast GTPases may use completely different mechanisms to attain a faster association kinetics.
Asymmetic Nucleotide Hydrolysis from the cpSRP54cpFtsY Complex
The crystal structure of the T. aquaticus FfhFtsY complex shows that the two GMPPCP molecules are bound at a composite active site formed at the dimer interface (Egea et al., 2004; Focia et al., 2004). Consistent with the composite nature of the active site and the extensive degree of cross-talk between the two GTPase sites, the two nucleotides are hydrolyzed at the same rate from the E. coli SRPFtsY complex. These observations have led to earlier proposals of concerted GTP hydrolysis in the SRPSR complex (Powers and Walter, 1995). In contrast to this notion, we showed here that nucleotide hydrolysis in the cpSRP54cpFtsY complex can be asymmetric, with the nucleotide hydrolyzing 10-fold faster from the cpFtsY than the cpSRP54 active site. This observation argues against a concerted mechanism. Even in the E. coli system, mutant GTPases have been identified in which GTP is hydrolyzed much faster from the SRP than the FtsY active site (Shan et al., 2004). Furthermore, when either of the GTPases is bound with a nonhydrolyzable GTP analogue, it can still activate efficient GTP hydrolysis on its binding partner (Shan, S., unpublished results). Together, these results strongly suggest that hydrolyses of the two GTPs in the SRPSR complex do not proceed through a concerted mechanism or an ordered pathway (i.e., one GTP must be hydrolyzed first before hydrolysis of the second GTP can occur). Rather, each active site can hydrolyze its bound GTP independently.
Even though the nucleotide is hydrolyzed 10-fold slower from cpSRP54 than from cpFtsY(D283N), multiple rounds of XTP hydrolysis from the cpSRP54cpFtsY(D283N) complex is not blocked and occur as efficiently as single turnover reactions (data not shown). Thus, disassembly of the complex must occur on a faster time scale than the second hydrolysis event, implying that SRP and SR can dissociate from one another even when only one of the nucleotides is hydrolyzed. A similar observation was made for the E. coli SRPSR complex (Shan et al., 2004). Together, the data from the E. coli and chloroplast systems suggest that only one GTP hydrolysis event is required to drive disassembly of the SRPSR complex. It remains to be clarified how many GTPs need to be hydrolyzed during each round of protein targeting, and what the precise role of each GTP hydrolysis event is.
The Nucleotide Specificity Determinant of cpFtsY, Asp283, Mediates Molecular Cross-Talk between the Two GTPases
Given the high specificity of cpFtsY(D283N) for XTP, it is surprising to find that this mutant prefers the noncognate GTP over the cognate XTP when it forms a complex with cpSRP54. This strongly suggests that Asp283, in addition to conferring nucleotide specificity to cpFtsY, also contributes to interactions at the dimer interface. The behavior of cpFtsY(D283N) is reminiscent of an XTP-specific mutant of the E. coli SRP GTPase, Ffh(D251N), which also prefers GTP over XTP when it forms a complex with FtsY (Shan and Walter, 2005). The crystal structure confirms that Asp251 makes an important interface contact with Lys390 from FtsY (Egea et al., 2004; Focia et al., 2004). A similar interaction could be formed by Asp283 of cpFtsY with a hydrogen bond donor (–AH) at the interface of the cpSRP54cpFtsY complex (Figure 8A). When cpFtsY(D283N) is bound to XTP, mutation of Asp283 to Asn destroys this interface contact and compromises the interaction between the two GTPases (Figure 8B). In contrast, replacement of XTP with GTP no longer constrains Asn283 in this particular configuration; a rotation around the C
–C
bond can reposition the carbonyl oxygen of Asn283 close to the hydrogen bond donor from cpSRP54, thus restoring this interface contact (Figure 8C). Alternatively, the exocyclic amino group of GTP could directly interact with a hydrogen bond acceptor from cpSRP54 (Figure 8D, -B:), therefore, replacement of GTP with XTP compromises the cpSRP54–cpFtsY(D283N) interaction. In either scenario, our results map the G-IV motif of cpFtsY and its bound nucleotide to the dimer interface between the two GTPases, and demonstrate the presence of extensive cross-talk between the two GTPase sites.
|
), a more thorough kinetic analysis showed that the cpSRP43 protein does not provide additional acceleration of the interaction between these two chloroplast GTPases (Supplemental Figure 2). Therefore, cpSRP43 does not replace all of the functions of the SRP RNA. This novel chloroplast protein may have evolved to mediate other important roles of the SRP RNA in the protein targeting reaction, such as recognition of the cargo protein (Schuenemann et al., 1998; Tu et al., 2000). Analogously, the SRP RNA may have been evolved to interact with ribosomal RNAs during cotranslational protein targeting in the classical SRP pathways (Rinke-Appel et al., 2002; Halic et al., 2004, 2006; Schaffitzel et al., 2006). It is fascinating to speculate on the evolutionary origin of the vast difference in the interaction kinetics between the SRP and SR GTPases from chloroplast versus those from classical SRP pathways, and why cells have evolved the SRP RNA to deal with the inefficient interaction between the SRP and SR GTPases in the classical pathway. An intriguing possibility is that the slow interaction kinetics between the SRP and SR GTPases in classical pathways provides additional opportunities for regulation and for improving fidelity. Compared with the time scale of protein targeting, the interaction kinetics between SRP and SR is still relatively slow even with the RNA present. The ribosome and/or the cargo protein, however, further accelerate the interaction between the two GTPases (Zhang, X., and Shan, S., unpublished data). The SRP RNA, bound in vicinity to the signal-sequence binding site, could mediate this additional stimulation of the SRP–SR interaction in response to cargo binding. In this way, the SRP RNA potentially provides a checkpoint to improve the fidelity of the classical SRP pathway, which needs to sort a diverse array of cellular proteins and to recognize signal sequences with different amino acid composition. In contrast, a much smaller number of proteins needs to be sorted inside the stroma of chloroplast, and the posttranslational cpSRP pathway seems to be dedicated for delivery of the LHCP family of proteins to the thylakoid membrane. More specific binding interactions between cpSRP and its cargo protein can be established to ensure the fidelity of the cpSRP pathway, and this alleviates the need to build in additional fidelity checkpoints by using the SRP RNA as a kinetic regulator.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
The online version of this article contains supplemental material at MBC Online (http://www.molbiolcell.org). 
Address correspondence to: Shu-ou Shan (sshan{at}caltech.edu)
Abbreviations used: GDP, guanosine-5'-diphosphate; GppNHp, 5'-guanylylimido-diphosphate; GTP, guanosine-5'-triphosphate; LHCP, light-harvesting chlorophyll-binding proteins; mant, N-methyl-anthraniloyl; SR, signal recognition particle receptor; SRP, signal recognition particle; XDP, xanthosine-5'-diphosphate; XppNHp, 5'-xanthylylimido-diphosphate; XTP, xanthosine-5'-triphosphate.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Buskiewicz, I., Kubarenko, A., Peske, F., Rodnina, M. V., and Wintermeyer, W. (2005). Domain rearrangement of SRP protein Ffh upon binding 4.5S RNA and the SRP receptor FtsY. RNA 11, 947–957.
Connolly, T., Rapiejko, P. J., and Gilmore, R. (1991). Requirement of GTP hydrolysis for dissociation of the signal recognition particle from its receptor. Science 252, 1171–1173.
Egea, P. F., Shan, S., Napetschnig, J., Savage, D. F., Walter, P., and Stroud, R. M. (2004). Substrate twinning activates the signal recognition particle and its receptor. Nature 427, 215–221.[CrossRef][Medline]
Focia, P. J., Shepotinovskaya, I. V., Seidler, J. A., and Freymann, D. M. (2004). Heterodimeric GTPase core of the SRP targeting complex. Science 303, 373–377.
Franklin, K. E., and Hoffman, N. E. (1993). Characterization of a chloroplast homologue of the 54-kDa subunit of the signal recognition particle. J. Biol. Chem 268, 22175–22180.
Freymann, D. M., Keenan, R. J., Stroud, R. M., and Walter, P. (1997). Structure of the conserved GTPase domain of the signal recognition particle. Nature 385, 361–364.[CrossRef][Medline]
Freymann, D. M., Keenan, R. J., Stroud, R. M., and Walter, P. (1999). Functional changes in the structure of the SRP GTPase on binding GDP and Mg2+GDP. Nat. Struct. Biol 6, 793–801.[CrossRef][Medline]
Gawronski-Salerno, J., Coon, Y.J.S., Focia, P. J., and Freymann, D. M. (2006). X-ray structure of the T. Aquaticus Ftsy: GDP complex suggests functional roles for the C-terminal helix of the SRP GTPases. Proteins 66, 984–995.
Gilmore, R., Blobel, G., and Walter, P. (1982a). Protein translocation across the endoplasmic reticulum: I. Detection in the microsomal membrane of a receptor for the signal recognition particle. J. Cell Biol 95, 463–469.
Gilmore, R., Walter, P., and Blobel, G. (1982b). Protein translocation across the endoplasmic reticulum. II. Isolation and characterization of the signal recognition particle receptor. J. Cell Biol 95, 470–477.
Goforth, R. L., Peterson, E. C., Yuan, J., Moore, M. J., Kight, A. D., Lohse, M. B., Sakon, J., and Henry, R. L. (2004). Regulation of the GTPase cycle in post-translational signal recognition particle-based protein targeting involves cpSRP43. J. Biol. Chem 279, 43077–43084.
Gorlich, D., Prehn, S., Hartmann, E., Kalies, K. U., and Rapoport, T. A. (1992). A mammalian homolog of Sec61p and SecYp is associated with ribosomes and nascent polypeptides during translocation. Cell 71, 489–503.[CrossRef][Medline]
Halic, M., Becker, T., Pool, M. R., Spahn, C.M.T., Grassucci, R. A., Frank, J., and Beckmann, R. (2004). Structure of the signal recognition particle interacting with the elongation-arrested ribosome. Nature 427, 808–814.[CrossRef][Medline]
Halic, M., Blau, M., Becker, T., Mielke, T., Poll, M. R., Wild, K., Sinning, I., and Beckmann, R. (2006). Following the signal sequence from ribosomal tunnel exit to signal recognition particle. Nature 444, 507–511.[CrossRef][Medline]
Huber, D., Boyd, D., Xia, Y., Olma, M. H., Gerstein, M., and Beckwith, J. (2005). Use of thioredoxin as a reporter to identify a subset of Escherichia coli signal sequences that promote signal recognition particle-dependent translocation. J. Bacteriol 187, 2983–2991.
Hwang, Y. W., and Miller, D. L. (1987). A mutation that alters the nucleotide specificity of elongation factor Tu, a GTP regulatory protein. J. Biol. Chem 262, 13081–13085.
Jagath, J. R., Rodnina, M. V., and Wintermeyer, W. (2000). Conformational changes in the bacterial SRP receptor FtsY upon binding of guanine nucleotides and SRP. J. Mol. Biol 295, 745–753.[CrossRef][Medline]
Jagath, J. R., Rodnina, M. V., Lentzen, G., and Wintermeyer, W. (1998). Interaction of guanine nucleotides with the signal recognition particle from Escherichia coli. Biochemistry 37, 15408–15413.[CrossRef][Medline]
Keenan, R. J., Freymann, D. M., Stroud, R. M., and Walter, P. (2001). The signal recognition particle. Annu. Rev. Biochem 70, 755–775.[CrossRef][Medline]
Li, X., Henry, R., Yuan, J., Cline, K., and Hoffman, N. (1995). A chloroplast homologue of the signal recognition particle subunit SRP54 is involved in the posttranslational integration of a protein into thylakoid membranes. Proc. Natl. Acad. Sci. USA 92, 3789–3793.
Lu, Y., Qi, H.-Y., Hyndman, J. B., Ulbrandt, N. D., Teplyakov, A., Tomasevic, N., and Bernstein, H. D. (2001). Evidence for a novel GTPase priming step in the SRP protein targeting pathway. EMBO J 20, 6724–6734.[CrossRef][Medline]
Montoya, G., Svensson, C., Luirink, J., and Sinning, I. (1997). Crystal structure of the NG domain from the signal recognition particle receptor FtsY. Nature 385, 365–368.[CrossRef][Medline]
Moser, C., Mol, O., Goody, R. S., and Sinning, I. (1997). The signal recognition particle receptor of Escherichia coli (FtsY) has a nucleotide exchange factor built into the GTPase domain. Proc. Natl. Acad. Sci. USA 94, 11339–11344.
Padmanabhan, W., and Freymann, D. M. (2001). The conformation of bound GMPPNP suggests a mechanism for gating the active site of the SRP GTPase site. Structure 9, 859–863.[Medline]
Peluso, P., Herschlag, D., Nock, S., Freymann, D. M., Johnson, A. E., and Walter, P. (2000). Role of 4.5S RNA in assembly of the bacterial signal recognition particle with its receptor. Science 288, 1640–1643.
Peluso, P., Shan, S., Nock, S., Herschlag, D., and Walter, P. (2001). Role of SRP RNA in the GTPase cycles of Ffh and FtsY. Biochemistry 40, 15224–15233.[CrossRef][Medline]
Poritz, M. A., Strub, K., and Walter, P. (1988). Human SRP RNA and E. coli. 4.5S RNA contain a highly homologous structural domain. Cell 55, 4–6.[CrossRef][Medline]
Powers, T., and Walter, P. (1995). Reciprocal stimulation of GTP hydrolysis by two directly interacting GTPases. Science 269, 1422–1424.
Rinke-Appel, J., Osswald, M., Knoblauch, K., Mueller, F., Brimacombe, R., Sergiev, P., Avdeeva, O., Bogdanov, A., and Dontsova, O. (2002). Crosslinking of 4.5S RNA to the Escherichia coli ribosome in the presence or absence of the protein Ffh. RNA 8, 612–625.[Abstract]
Schaffitzel, C., Oswald, M., Berger, I., Ishikawa, T., Abrahams, J. P., Koerten, H. K., Koning, R. I., and Ban, N. (2006). Structure of the E. coli signal recognition particle bound to a translating ribosome. Nature 444, 503–506.[CrossRef][Medline]
Schuenemann, D., Gupta, S., Persello-Cartieaux, F., Klimyuk, V. I., Jones, J.D.G., Nussaume, L., and Hoffman, N. E. (1998). A novel signal recognition particle targets light-harvesting proteins to the thylakoid membranes. Proc. Natl. Acad. Sci. USA 95, 10312–10316.
Shan, S., and Walter, P. (2003). Induced Nucleotide Specificity in a GTPase. Proc. Natl. Acad. Sci. USA 100, 4480–4485.
Shan, S., and Walter, P. (2005). Molecular crosstalk between the nucleotide specificity determinant of the SRP GTPase and the SRP receptor. Biochemistry 44, 6214–6222.[CrossRef][Medline]
Shan, S., Stroud, R., and Walter, P. (2004). Mechanism of association and reciprocal activation of two GTPases. Plos Biol 2, e320.[CrossRef][Medline]
Shepotinovskaya, I. V., and Freymann, D. M. (2001). Conformational change of the N-domain on formation of the complex between the GTPase domains of Thermus aquaticus Ffh and FtsY. Biochem. Biophys. Acta 1597, 107–114.
Simon, S. M., and Blobel, G. (1991). A protein-conducting channel in the endoplasmic reticulum. Cell 65, 371–380.
) and in the presence of 5 µM cpSRP54 (
). (C and D) Titration of the fluorescence changes of mant-GTP (C) and mant-GDP (D) in the presence of cpSRP54 (
) or cpFtsY (
