Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stickney, L. M. Articles by Mackie, G. A. Search for Related Content PubMed PubMed Citation Articles by Stickney, L. M. Articles by Mackie, G. A.

 Previous Article  |  Next Article 

Journal of Bacteriology, November 2005, p. 7214-7221, Vol. 187, No. 21
0021-9193/05/$08.00+0     doi:10.1128/JB.187.21.7214-7221.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Function of the Conserved S1 and KH Domains in Polynucleotide Phosphorylase

Leigh M. Stickney, Janet S. Hankins, Xin Miao, and George A. Mackie^*

Department of Biochemistry & Molecular Biology, University of British Columbia, 2350 Health Sciences Mall, Vancouver, British Columbia, Canada V6T 1Z3

Received 7 June 2005/ Accepted 19 August 2005

Top Abstract Introduction Materials and Methods Results Discussion References

 
We have examined the roles of the conserved S1 and KH RNA binding motifs in the widely dispersed prokaryotic exoribonuclease polynucleotide phosphorylase (PNPase). These domains can be released from the enzyme by mild proteolysis or by truncation of the gene. Using purified recombinant enzymes, we have assessed the effects of specific deletions on RNA binding, on activity against a synthetic substrate under multiple-turnover conditions, and on the ability of truncated forms of PNPase to form a minimal RNA degradosome with RNase E and RhlB. Deletion of the S1 domain reduces the apparent activity of the enzyme by almost 70-fold under low-ionic-strength conditions and limits the enzyme to digest a single substrate molecule. Activity and product release are substantially regained at higher ionic strengths. This deletion also reduces the affinity of the enzyme for RNA, without affecting the enzyme's ability to bind to RNase E. Deletion of the KH domain produces similar, but less severe, effects, while deletion of both the S1 and KH domains accentuates the loss of activity, product release, and RNA binding but has no effect on binding to RNase E. We propose that the S1 domain, possibly arrayed with the KH domain, forms an RNA binding surface that facilitates substrate recognition and thus indirectly potentiates product release. The present data as well as prior observations can be rationalized by a two-step model for substrate binding.

Top Abstract Introduction Materials and Methods Results Discussion References

 
The processing and/or degradation of RNAs, including rRNA, tRNA and mRNA, is a critical posttranscriptional regulatory step. In Escherichia coli, several enzymes which participate in RNA processing and degradation are organized into a macromolecular complex, the RNA degradosome (3, 18, 22). Major components of the degradosome include RNase E, an 5'-end-dependent endonuclease, polynucleotide phosphorylase (PNPase), a phosphate-dependent 3' exonuclease, RhlB, a DEAD box RNA helicase, and enolase, an abundant glycolytic enzyme (3, 5, 27). Other proteins associate with the degradosome in apparently substoichiometric quantities, but only RNase E, PNPase, and RhlB are required to reconstitute the activity of the RNA degradosome in vitro (6).

Although their activities are quite different, both RNase E and PNPase share a common structural motif, an S1 (or oligonucleotide/oligosaccharide binding fold) domain, as do RNase G and RNase II (2). The relative locations of the S1 domain in these RNases are shown in Fig. 1a. Its location provides no clue to its function(s) in any of these enzymes. S1 domains are also found in a number of unrelated proteins whose principal common feature is interaction with single-stranded nucleic acids (1, 20, 30). The solution structure of the S1 domain in PNPase has been determined and consists of five antiparallel ß-strands with surface-exposed hydrophobic and basic residues (2). The structure of the S1 domain of RNase E has also been determined recently and displays a similar overall fold (8, 25). Both of the best-characterized mutations in RNase E, rne-1 (G66S) and rne-3071 (L68F), map to ß-3 in the S1 domain (8, 25). These mutations result in thermolability, suggesting that substitutions into this strand destabilize the global fold of RNase E. Consistent with this idea, functional investigations of the S1 domain in RNase E point to its role in facilitating substrate recognition, autoregulation, and subunit assembly (8, 25). The role of this domain in PNPase is not yet clear. Moreover, several point and deletion mutations in the S1 domain constructed during an extensive mutational investigation of PNPase that focused on conserved residues showed minimal effect on PNPase activity and autoregulation (13). Interestingly, the S1 domain is required for the functioning of the type 3 secretion system during the infection of macrophages by Yersinia (24). It is intriguing that the S1 domain in PNPase (residues 619 to 691) is preceded by another RNA binding motif, the KH domain (residues 557 to 591) (Fig. 1b). The presence of tandem RNA binding motifs may provide an extended RNA binding surface on PNPase, as has been proposed for the NusA transcription factor from Thermotoga maritima (32). Such a surface could play a key role in conferring processivity to PNPase or in substrate binding (5, 13, 29). In this regard, a G570D mutation in the KH domain affects the autoregulation of PNPase, consistent with a role in substrate recognition (10) The structure of PNPase from Streptomyces antibioticus shows that the S1 domain, although poorly resolved in its electron density, could not lie in close proximity to the putative active site (28). Thus, the role of the S1 domain in substrate recognition or actual catalysis in either RNase E or PNPase is unresolved.

View larger version (13K): [in this window] [in a new window]

FIG. 1. a. Organization of the S1 RNA binding domains in various RNases. The relative positions of the S1 domain in RNase E, RNase G, RNase II, and PNPase are shown by the cross-hatched box as predicted by Bycroft et al. (2). The coordinates of the domain in each enzyme are given. The N termini of RNase II and PNPase are aligned arbitrarily. b. Domain structure of PNPase and derived truncations. The domain organization of PNPase is based on an alignment of PNPase enzymes (17; see also reference 28). PH refers to the RNase PH-like domain of PNPase, whereas PH' denotes a diverged (inactive) duplication of the PH domain. KH and S1 refer to the KH ("RNP H homology") and S1 (oligonucleotide/oligosaccharide binding fold) domains. The numbers below the schematic give the coordinates of the domain boundaries based on the E. coli enzyme. c. Purity of PNPase and its derivatives. Samples containing 1 µg of purified recombinant protein were separated by SDS-gel electrophoresis and visualized by staining. Lane 1, molecular mass markers; lane 2, wild-type PNPase; lane 3, PNPaseKH; lane 4, PNPase S1; lane 5, PNPase S1+KH.

In this work we have deleted the S1 or KH domain from PNPase, have purified the resultant derivatives, and show that neither domain is required for activity, in agreement with others (13), or for association with RNase E. However, these domains considerably facilitate the activity of PNPase indirectly by promoting substrate binding and product release, possibly in a concerted fashion. We propose a two-step model of substrate recognition to account for our data.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Bacterial strains and plasmids. The vectors pET-11, pET-16b, and pET-24b and the host expression strain BL21(DE3) were obtained from Novagen (Madison, WI). ENS134 [= BL21(DE3) pnp::Tn5] was obtained from Marc Dreyfus. E. coli DH5a and JM109 were the hosts for recombinant plasmid construction. pRE194 (18) was obtained from Sue Lin-Chao via Robert Simons. Plasmids pGC400 (4) and pEP18 (22) served as untagged and tagged sources of PNPase, respectively. Selected portions of the PNPase gene in pGC400 were deleted by the Quick Change method (Stratagene, Inc.). Primer sequences are available upon request. After characterization of candidate mutants, selected plasmids were transformed into BL21(DE3) or BL21(DE3) pnp::Tn5.

Purification of RNase E and PNPase. Flag-Rne was purified from cultures of BL21(DE3) pnp::Tn5/pRE194 grown in 300 ml LB supplemented with kanamycin and carbenicillin at 37°C to early log phase and induced with 0.5 mM IPTG (isopropyl-ß-D-thiogalactopyranoside) for 4 h. Cultures were chilled to 4°C and harvested by centrifugation. Cell pellets were suspended in 1% of the culture volume of buffer A (50 mM Tris-HCl, 10 mM MgCl₂, 60 mM NH₄Cl, 0.5 mM EDTA, 1 mM dithiothreitol [DTT], 5% glycerol, pH 7.6) and lysed by two passages through a French pressure cell (Aminco) at 8,000 lb/in². Lysates were supplemented with DTT to 0.2 mM, DNase I (20 µg/ml), and a cocktail of protease inhibitors; incubated for 15 min on ice; and cleared of debris and unbroken cells by centrifugation at 30,000 x g. The supernatant was diluted with 2 volumes of buffer A and slowly made to 40% of saturation with ammonium sulfate. Precipitated protein was collected by centrifugation, dissolved in 1 ml of buffer A containing 0.2 mM DTT, and dialyzed against two changes of the same buffer to form the AS26 fraction.

Untagged versions of PNPase were purified as described previously (14) using BL21(DE3) pnp::Tn5 as the host and lysis in a French pressure cell. His₆-PNPase was purified from cultures of BL21(DE3)/pEP18 grown similarly, except that the crude cell pellet was suspended in 50 mM Tris-HCl, pH 8.2, containing 500 mM NaCl. Following lysis and clarification of the lysate by centrifugation at 30,000 x g, His₆-PNPase was purified by chromatography on a column packed with Talon resin (BD Biosciences, Inc.) by following the manufacturer's instructions.

Enzyme assays. PNPase activity was assayed by the shortening of SL9A RNA from 85 to 55 residues as described previously (4, 26). The standard assay contained 20 mM Tris-HCl, pH 7.5, 1.5 mM dithiothreitol, 1 mM MgCl₂, 10 mM K-phosphate (pH 7.5), either 20 mM or 250 mM KCl, and 0.8 pmol uniformly labeled SL9A RNA in 40 µl. Assays were initiated by the addition of enzyme (86 fmol wild-type [WT] PNPase or 380 fmol truncated PNPase), incubated at 30°C, and terminated by quenching a portion with 3 volumes of 90% formamide. Samples were separated by electrophoresis in an 8% polyacrylamide gel containing 8 M urea. In all cases, products were quantified using a PhosphorImager (Molecular Dynamics). Minimal degradosomes were reconstituted and assayed on the 375-nucleotide (nt) malEF RNA substrate as described previously (6, 15).

RNA binding assays. Gel mobility shift assays were performed essentially as described by Folichon et al. (9). The RNA ligand, 20 nM SL9A RNA (26), was incubated with increasing amounts of PNPase or its derivatives for 30 min at 37°C in a buffer containing 10 mM Tris-HCl, 1 mM EDTA, 80 mM NaCl, 1% glycerol, pH 8.0. Samples were chilled, loaded onto a nondenaturing gel, and separated in chilled Tris-borate-EDTA buffer at 4°C. Samples were visualized by PhosphorImaging.

Immunoprecipitation and protein blots. Immunoprecipitation with anti-Flag beads and subsequent Western blotting were performed as described previously (6, 21). Samples were denatured by boiling, separated electrophoretically, electroblotted onto a nitrocellulose membrane, and probed with appropriate polyclonal anti-PNP antibodies (1:20,000) and washed.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Truncation of PNPase. Partial proteolysis is a proven method to determine structural boundaries of native, folded proteins (19). Moreover, partial proteolysis of PNPase has been reported to reduce its processivity but not eliminate activity (11, 29). Therefore, we used this approach to probe the domain organization of PNPase. Partially purified N-terminally tagged His₆-PNPase was bound to His · Bind Ni²⁺ resin and then subjected to moderate proteolysis by trypsin or chymotrypsin at ambient temperature (see the legend to Fig. 2). Material remaining on the resin after the protease treatment was eluted and analyzed by Western blotting (Fig. 2). Although tryptic digestion was more complete, both tryptic and chymotryptic digestions yielded a major proteolytic product of 68 kDa, which was retained on the resin and thus encompasses the N terminus (Fig. 2, lanes 2 and 3). The size of this fragment generally agrees with those of previous reports (11, 29). The missing mass corresponds approximately to the combined size of the KH and S1 motifs. These inferences were confirmed by mass spectrometry of an unfractionated partial tryptic digest of PNPase. Major fragments of sizes 60,890 and 16,285 were obtained (data not shown). The former corresponds to residues 1 to 559 (predicted mass, 60,787 Da) and the latter to residues 560 to 711, resulting from a cleavage at R559 in the linker following alpha helix 12 in PNPase (28). The chymotryptic digestion also yields a secondary product of 75 kDa, possibly resulting from partial or complete removal of the S1 domain (Fig. 2, lane 3). The results of partial proteolysis suggest that the N-terminal portion of PNPase (the tandem PH domains plus the linker) (Fig. 1b) is folded as a separate entity independent of the C-terminal RNA binding motifs, in agreement with the crystal structure (28).

View larger version (44K): [in this window] [in a new window]

FIG. 2. Partial proteolysis of PNPase. Partially purified His6-PNPase(200 µg) was adsorbed on 30 µl of His · Bind Ni2+ resin and treated with 10 µg trypsin (lane 2) or chymotrypsin (lane 3) in 25 mM Tris-HCl, 50 mM KCl, 0.1 mM EDTA, pH 7.9, for 1 to 1.5 h at room temperature in a final volume of 50 µl. Digestion was stopped with 10 µl 100 mM phenylmethylsulfonyl fluoride in ethanol. The adsorbed proteins were washed and eluted with 500 µl 20 mM Tris-KCl, 1 M imidazole, 500 mM NaCl, pH 7.9. Portions (100 µl) were precipitated and resolved by electrophoresis on 7% polyacrylamide gels and subjected to Western blotting with anti-His6 antibodies. Arrows in the right margin show untreated PNPase (lane 1; apparent size, 85 kDa) and a major fragment of 68 kDa.

Further dissection of PNPase was achieved by deleting defined segments shown in Fig. 1b. The choice of end points was based in part on those used by Jarrige et al. (13) to facilitate comparison with their data. We found, however, that His₆-PNPase (full length), under otherwise identical conditions and at similar concentrations, was 6- to 10-fold less active than untagged PNPase against a 3'-oligonucleotide-adenylated fragment of the rpsT mRNA (data not shown). Moreover, His₆ PNPase appeared to stall several residues further 3' to a stable stem-loop in the rpsT substrate than WT PNPase (data not shown). For these reasons, the effects of deletion in the KH or S1 domains were assessed using untagged enzymes.

Activities of truncated PNPase derivatives. Wild-type PNPase, PNPaseKH, PNPaseS1, and PNPaseKH+S1 (Fig. 1b) were purified from a derivative of E. coli BL21(DE3) carrying a chromosomal pnp::Tn5 disruption (Fig. 1c) and assayed against a previously characterized synthetic substrate, SL9A RNA (26). Typical time courses of digestion are shown in Fig. 3. Both truncated PNPase derivatives, PNPaseS1 and PNPaseKH, exhibited significant PNPase activity as evidenced by the shortening of the oligo(A) tail in SL9A RNA, in general agreement with other data (13) (Fig. 3b and c). However, unlike those authors, we found that significantly greater amounts of truncated enzyme were required to achieve the same extent of degradation as the wild-type enzyme (Fig. 3; Table 1, second column). PNPaseS1 is about 50-fold less active than the wild type, while PNPaseKH is approximately 19-fold less active. The full C-terminal truncation, PNPaseKH+S1, exhibited approximately 1% of the activity of the wild-type enzyme (Fig. 3d and Table 1). The activity of each enzyme preparation was also assayed at a higher salt concentration (Table 1, third column) with similar results. Under these conditions, PNPaseKH+S1 displayed slightly more relative activity than at a low concentration of salt but was still only 2% as active as the WT enzyme. Other experiments showed that PNPase S1 also exhibits synthetic activity (data not shown).

View larger version (29K): [in this window] [in a new window]

FIG. 3. Time course of PNPase activity against SL9A RNA. Assays of PNPase activity were performed as described in Materials and Methods. Samples were separated by electrophoresis in an 8% polyacrylamide gel containing 8 M urea. Products were visualized and quantified using a PhosphorImager (Molecular Dynamics). The time of digestion in minutes is shown above each lane, while the cartoons in the central margin give the position of the 85-nt SL9A RNA substrate and the 55-nt stalled product. a, WT PNPase; b, PNPaseS1; c, PNPaseKH; d, PNPaseKH+S1.

View this table: [in this window] [in a new window]

TABLE 1. Activities of PNPase and derivatives

We noticed, however, that the truncated forms of PNPase were unable to trim the 3' extension on all of the SL9A RNA substrate molecules to yield the 55-nt limit product, even when sufficient activity and time were allowed. We calculated the number of molecules of SL9A that were shortened per trimer of PNPase after 1 h of incubation when the yield of product was constant. In the case of WT PNPase, all of the substrate molecules were trimmed to 55 nt and each enzyme was calculated to have digested the 3' extensions on six to eight molecules of substrate (Table 1, fourth and fifth columns). The expected yield was 7.5 turnovers per trimer based on the initial concentrations of enzyme and substrate (see Materials and Methods). Likewise, PNPaseKH catalyzed 1.6 turnovers in a low salt concentration (Table 1, fourth column) compared to an expected value of 1.7 (note the higher enzyme concentrations in its assay [see Materials and Methods]). It is, however, somewhat less efficient with a high concentration of salt. In contrast, both PNPaseS1 and PNPaseKH+S1 underwent less than one turnover in a low concentration of salt, although fourfold more enzyme was added to each assay mixture. Indeed, these truncated enzymes effectively stalled on their substrate and were unable to release the partially degraded product. High salt concentrations (Table 1, fifth column) suppress this apparent stalling, as both PNPaseS1 and PNPaseKH+S1 exhibit greater than one turnover and reach about two-thirds of the expected yield.

RNA binding by truncated PNPase. Electrophoretic mobility shift assays were performed with wild-type PNPase and its truncated derivatives to determine the effect of truncation on the enzyme's affinity for a model substrate. Full-length PNPase formed a single complex with SL9A at relatively low RNA/protein ratios (Fig. 4a). It exhibited high affinity for SL9A RNA, with an apparent K_d of 1.4 ± 0.07 nM (n = 3) (Fig. 4a). In contrast, both of its truncated derivatives displayed much weaker affinities for RNA. PNPaseS1 required a significant excess of protein to form a complex with SL9A RNA, which was consistently more diffuse than that formed by the WT enzyme (Fig. 4b). The apparent K_d of this truncation was 55 ± 6 nM (n = 3). Likewise, PNPaseKH also formed a somewhat diffuse set of complexes (Fig. 4c), with an apparent K_d of 40 ± 5 nM. PNPaseKH+S1 appeared to form two complexes. We interpret the first (Fig. 4d) as a 1:1 complex between SL9A and the truncated enzyme, whose apparent affinity for SL9A was 97 ± 7 nM (n = 3). The second, more intense but quite heterogeneous complex forms at higher concentrations of enzyme (Fig. 4d, lanes 4 to 7). We believe that this mixture represents aggregates of protein and RNA-protein complexes.

View larger version (24K): [in this window] [in a new window]

FIG. 4. RNA binding activity of PNPase and derived truncations. Gel mobility shift assays were performed with 20 nM SL9A as the substrate as described in Materials and Methods. The RNA/PNPase ratio (mole/mole) is shown above each lane. a, wild-type PNPase; b, PNPaseS1; c, PNPaseKH; d, PNPaseKH+ S1. The positions of substrate (S) and retarded complexes (C) are shown in the center margin. The asterisk denotes a putative aggregate in panel d.

PNPase-RNase E interactions. In view of their partial dispensability for activity, we also determined whether the KH and S1 domains of PNPase were required for interaction with RNase E. Preliminary experiments indicated that fragments of PNPase tethered to a His · Bind resin were capable of interacting with RNase E (17; data not shown). Confirmation of this observation was sought using coimmunoprecipitation, as interactions between RNase E and PNPase both in vivo and in vitro can be detected readily by this method (6, 21, 31). A fractionated crude extract enriched in Flag-Rne (18) was incubated with purified full-length or truncated PNPase derivatives under conditions that favor reconstitution. Complexes were absorbed to anti-Flag beads, and bound proteins were analyzed by Western blotting (see Materials and Methods). As expected, full-length PNPase bound to Flag-Rne (Fig. 5, lane 2). PNPaseS1, PNPaseKH, and PNPaseKH+S1 also bound to Flag-Rne, with efficiencies equivalent to that of the wild-type enzyme (Fig. 5, lanes 3 to 5). Figure 5, lanes 1 and 6, serve as controls to show the absence of contaminating PNPase in the preparation of Flag-Rne (lane 1) and the specificity of the interaction (lane 6). A similar experiment showed that the purified S1 domain from PNPase (residues 617 to 700) could not interact with Flag-Rne or compete for PNPase binding at a 25-fold molar excess (data not shown). Other, more extensive PNPase truncations (e.g., lacking the N-terminal PH' domain) were also assayed for interaction with RNase E. Although apparent interactions were detected, controls showed that they could not be competed by purified PNPase and were thus nonspecific. Taken together, these experiments show that neither the S1 nor the KH domains is required for the interaction between PNPase and RNase E in the RNA degradosome.

View larger version (5K): [in this window] [in a new window]

FIG. 5. RNase E-PNPase interactions. Coimmunoprecipitation assays were conducted as described in Materials and Methods. All incubation mixtures (100 µl in buffer A) contained 20 µg of an S100 fraction prepared from BL21(DE3) pnp::Tn5 as the carrier, 10 µg of AS26 containing Flag-Rne (lanes 2 to 6), and 2 µg of purified PNPase (or a derivative) as indicated below. Samples were heated for 10 min at 30°C, chilled on ice for 30 min, and then absorbed to anti-Flag-agarose beads (Sigma-Aldrich). Following an extensive washing, bound proteins were eluted and visualized by Western blotting using polyclonal anti-PNPase antibodies. Lane 1, no PNPase; lane 2, WT PNPase; lane 3, PNPaseS1; lane 4, PNPaseKH; lane 5, PNPaseKH+S1; lane 6, WT PNPase but no FLAG-Rne; lane 7, purified PNPase as an internal control.

We also assayed whether truncated derivatives of PNPase could function in a reconstituted minimal degradosome (6). The data in Fig. 6a show the activity of wild-type PNPase against the malEF RNA substrate in the absence of other components of the degradosome. PNPase alone could shorten the input substrate to produce two products, one shortened by 20 residues (the "star" intermediate) and the second by 35 residues to form the 340-nt RSR (repetitive extragenic palindrome-stabilized RNA) intermediate. The latter was not further degraded (Fig. 6a). In contrast, reconstitution of native PNPase with a fragment of RNase E (Rne408) and RhlB to form a minimal degradosome permitted the transient accumulation of the "star" and RSR intermediates in the first few minutes of the assay. Both intermediates were subsequently fully degraded as expected (Fig. 6b) (cf. reference 6). Each of the truncated derivatives of PNPase was substituted for wild-type PNPase in a reconstitution and assayed. All three exhibited similar behaviors (Fig. 6c to e). The malEF RNA was digested slowly, but only the "star" intermediate accumulated to any extent. Importantly, little detectable RSR intermediate could be formed by any of the truncated enzymes. The partially degraded "star" intermediate was not recovered in stoichiometric amounts, a problem that we attribute to trace endonuclease contamination in the preparation of RhlB. The data in Fig. 6f show that PNPaseKH+S1 in the presence of Rne408 can slowly convert the malEF RNA to the "star" intermediate, but not the RSR, with little if any loss to side reactions. We conclude that C-terminal truncations of PNPase compromise both the activity of PNPase and its ability to be coupled to the action of RhlB in a reconstituted minimal RNA degradosome.

View larger version (18K): [in this window] [in a new window]

FIG. 6. Assay of minimal degradosomes reconstituted with truncated PNPase derivatives. Rne408, RhlB, and derivatives of PNPase were purified, reconstituted, and assayed on the 375-nt malEF RNA substrate as described previously (6, 15). Portions of the incubation mixture were removed at the time indicated above each lane, denatured with 3 volumes of 90% formamide, separated electrophoretically on a 6% polyacrylamide gel containing 8 M urea, and quantified using a PhosphorImager. Abbreviations in the central margin indicate the positions of the malEF RNA substrate, the "star" intermediate (355 nt; denoted by an asterisk), and the RSR intermediate (340 nt). a, WT PNPase alone in the absence of Rne408 and RhlB; b, WT PNPase, Rne408 and RhlB; c, PNPaseS1, Rne408 and RhlB; d, PNPaseKH, Rne408 and RhlB; e, PNPaseS1+KH, Rne408 and RhlB; f, PNPaseS1+KH and Rne408.

Top Abstract Introduction Materials and Methods Results Discussion References

 
The S1 and KH domains in PNPase exert subtle effects on activity. Jarrige et al. (13) have thoroughly characterized a number of mutations in PNPase and found that deletion of either the S1 or the KH domain resulted in modest changes in three different measures of enzyme activity. Thus, from the outset, we worked from the hypothesis that the S1 or KH domain might enhance the processivity of PNPase but would not be required for activity against a single-stranded substrate. In agreement with Jarrige et al. (13), our data show that PNPase lacking either the S1 or KH domain does indeed retain phosphorolytic activity against a model substrate. This is consistent with suggestions that the catalytic site of PNPase resides within the core PH domain, also called the second core domain (13, 28). Our data also show that the catalytic core of PNPase retains intrinsic RNA binding activity, in agreement with the finding that a G454D substitution impairs the binding of PNPase to RNA (23). However, the measured rates of product formation by the truncated PNPase derivatives are significantly reduced relative to that of the wild-type enzyme. Elimination of the KH domain reduced specific activity 19-fold, while removal of the S1 domain reduced specific activity almost 50-fold (Table 1). Interestingly, these reductions fully parallelled the loss of affinity for RNA demonstrated by a mobility shift assay. Moreover, to the extent it could be determined, the processivity of the truncated forms of PNPase was unaffected, as discrete, rather than random, end products were produced (Fig. 3). Calculations of product yield showed that each trimer of wild-type PNPase could trim the single-stranded 3' extensions of at least six molecules of the SL9A substrate. In contrast, truncated forms of PNPase lacking the S1 domain (i.e., PNPaseS1 and PNPaseKH+S1) could not process more than one molecule of SL9A during an assay in a low concentration of salt. Significantly, a high salt concentration substantially suppressed this failure. Thus, the truncation of either domain, but particularly the S1 domain, could affect product release and enzyme cycling (see below and the model in Fig. 7). In this regard, it is intriguing that yeast Prp22p, a member of the DEAH family of RNA helicases, contains an S1 domain and is involved in the release of spliced mRNA from the spliceosome (7).

View larger version (21K): [in this window] [in a new window]

FIG. 7. Model for the role of the S1 domain in PNPase. PNPase is depicted as a circle divided into three segments, each representing residues 1 to 541 in one subunit (29). The red boxes in each subunit depict putative substrate binding sites. The S1 and KH domains are represented as solid ellipses. In step 1 of the model, a substrate associates reversibly with the S1 domain of one subunit, establishing an asymmetry within the enzyme-substrate complex. In step 2, the 3' terminus of the substrate migrates to the substrate binding site such that the 3' OH and the neighboring phosphodiester bond lie in the catalytic site. In step 3, phosphorolysis proceeds until the substrate stalls at the base of a stem-loop (26). In step 4, a second molecule of the substrate binds loosely to the S1 domain of the "active" subunit. In step 5, the incoming substrate displaces the stalled product, reinitiating the catalytic cycle.

The S1 and KH domains in PNPase are not required for the assembly of the RNA degradosome. We assessed the ability of truncated PNPase derivatives to interact with RNase E by coimmunoprecipitation in the presence of total protein from a high-speed supernatant as a means of detecting subtle differences in affinity. Nonetheless, all three truncations tested bound equally well to RNase E (Fig. 5). This finding suggests that the binding site for RNase E must lie between residues 1 to 549 of PNPase, i.e., within the region whose structure has been determined (28). Consistent with this interpretation, the G454D mutation mentioned above impairs the stability of the RNA degradosome (23). Attempts were made to identify the site of contact between RNase E and PNPase by creating further truncations in PNPase or by using bifunctional cross-linking, but without success.

Although all three truncation derivatives of PNPase bound to RNase E, their activity in a reconstituted minimal RNA degradosome was significantly reduced. In particular, none of the truncations tested was able to participate in the RhlB-dependent degradation of the highly structured REP element in the malEF RNA substrate. Interpretation of this observation is complicated by the fact that the truncated enzymes exhibit low rates of product release under conditions of the assay (cf. Table 1) and that the substrate is in excess. Nonetheless, it is clear that the presence of RhlB and RNase E is insufficient to promote PNPase-dependent degradation of the REP sequence by the truncated derivatives of PNPase. Subject to the caveat noted above, we propose that the S1 and KH domains of PNPase function in the degradosome to bind and stabilize single-stranded RNA released by RhlB, in effect guiding this RNA to the active site of PNPase (Fig. 7).

Model for substrate binding and retention by PNPase. We propose that there are two RNA binding surfaces on each PNPase monomer: one in the core catalytic domain that confers processivity and one formed by the more peripheral KH and S1 domains. We also assume that only one substrate molecule can gain access to a putative active site in PNPase, although in principle, there are three such sites in the active trimeric enzyme. In our model, we propose that an initial weak, salt-sensitive interaction between PNPase and single-stranded RNAs occurs via one of the tandem KH-S1 domains (step 1 in the model of Fig. 7). A substrate bound to this surface then interacts with the stronger core RNA binding surface in the PH domain, placing the terminal phosphodiester bond in proximity to the active-site residues (step 2 in Fig. 7). This two-step binding process is similar in concept to simple two-state models for the interaction between RNA polymerase and a promoter (16). Phosphorolysis proceeds (step 3 in Fig. 7) until the enzyme encounters a stem-loop barrier. Such a substrate retains its interaction with the core RNA binding surface and thus remains bound to PNPase. Based on our data, we propose that a key role of the S1 domain (acting with the KH domain) is displacement of the stalled substrate. Thus, binding of another molecule of the substrate to one of the unoccupied peripheral KH plus S1 domains, followed by its migration to the corresponding core RNA binding site, permits the second substrate to displace the stalled product of digestion so that the cycle is repeated (steps 4 and 5 in Fig. 7). As depicted in Fig. 7, displacement of a stalled substrate occurs "in cis"—i.e., via the S1 and KH domains of the subunit to which the first substrate is bound. Fresh molecules of substrate could conceivably bind to one of the other S1 and KH domains but would not be able to displace the stalled substrate, and presumably, would not be able to gain access to the active site, which is assumed to be recessed in the central cavity of PNPase (13, 28). In effect, binding of the first substrate induces a functional asymmetry in PNPase that is not erased until a substrate is totally dissociated from its surface. The model also explains why PNPase lacking the S1 domain is unable to engage in more than a single round of substrate shortening. Finally, the model is also consistent with a role for the S1 and KH domains in autoregulation (10, 12, 13).

 
This work was supported by grant MT-5396 from the Canadian Institutes of Health Research to G.A.M.

We thank Rob Edge for his preliminary work and Annie Prud'homme-Généreux for her helpful advice on coimmunoprecipitation. We also thank our colleagues who provided strains.

 
* Corresponding author. Mailing address: Department of Biochemistry & Molecular Biology, University of British Columbia, 2350 Health Sciences Mall, Vancouver, BC V6T 1Z3, Canada. Phone: (604) 822-5943. Fax: (604) 822-5227. E-mail: gamackie{at}interchange.ubc.ca.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Arcus, V. 2002. OB-fold domains: a snapshot of the evolution of sequence, structure and function. Curr. Opin. Struct. Biol. 12:794-801.[CrossRef][Medline]
2
2. Bycroft, M., T. J. Hubbard, M. Proctor, S. M. Freund, and A. G. Murzin. 1998. The solution structure of the S1 RNA binding domain: a member of an ancient nucleic acid-binding fold. Cell 88:235-242.
3
3. Carpousis, A. J., N. F. Vanzo, and L. C. Raynal. 1999. mRNA degradation. A tale of poly(A) and multiprotein machines. Trends Genet. 15:24-28.[CrossRef][Medline]
4
4. Coburn, G. A., and G. A. Mackie. 1998. Reconstitution of the degradation of the mRNA for ribosomal protein S20 with purified enzymes. J. Mol. Biol. 279:1061-1074.[CrossRef][Medline]
5
5. Coburn, G. A., and G. A. Mackie. 1999. Degradation of mRNA in Escherichia coli: an old problem with some new twists. Prog. Nucleic Acids Res. Mol. Biol. 62:55-108.[Medline]
6
6. Coburn, G. A., X. Miao, D. J. Briant and G. A. Mackie. 1999. Reconstitution of a minimal RNA degradosome demonstrates functional coordination between a 3-exonuclease and a DEAD-box RNA helicase. Genes Dev. 13:2594-2603.[Abstract/Free Full Text]
7
7. Company, M., J. Arenas, and J. Abelson. 1991. Requirement of the RNA helicase-like protein PRP22 for release of messenger RNA from spliceosomes. Nature 349:487-493.[CrossRef][Medline]
8
8. Diwa, A. A., X. Jiang, M. Schapira, and J. G. Belasco. 2002. Two distinct regions on the surface of an RNA-binding domain are crucial for RNase E function. Mol. Microbiol. 46:959-969.[CrossRef][Medline]
9
9. Folichon, M., V. Arluison, O. Pellegrini, E. Huntziger, P. Régnier, and E. Hajnsdorf. 2003. The poly(A) binding protein Hfq protects RNA from RNase E and exoribonucleolytic degradation. Nucleic Acids Res. 31:7302-7310.[Abstract/Free Full Text]
10
10. García-Mena, J., A. Das, A. Sánchez-Trujillo, C. Portier, and C. Montañez. 1999. A novel mutation in the KH domain of polynucleotide phosphorylase affects autoregulation and mRNA decay in Escherichia coli. Mol. Microbiol. 33:235-248.[CrossRef][Medline]
11
11. Guissani, A., and C. Portier. 1976. Study on the structure-function relationship of polynucleotide phosphorylase: model of a proteolytic degraded polynucleotide phosphorylase. Nucleic Acids Res. 3:3015-3024.
12
12. Jarrige, A.-C., N. Mathy, and C. Portier. 2001. PNPase autocontrols its expression by degrading a double-stranded structure in the pnp mRNA leader. EMBO J. 20:6845-6855.[CrossRef][Medline]
13
13. Jarrige, A.-C., D. Bréchemier-Baey, N. Mathy, O. Duché, and C. Portier. 2002. Mutational analysis of polynucleotide phosphorylase from Escherichia coli. J. Mol. Biol. 321:397-409.[CrossRef][Medline]
14
14. Jones, G. H., M. F. Symmons, J. S. Hankins, and G. A. Mackie. 2003. Overexpression and purification of untagged polynucleotide phosphorylases. Protein Expr. Purif. 32:202-209.[CrossRef][Medline]
15
15. Mackie, G. A., G. A. Coburn, X. Miao, D. J. Briant, and A. Prud'homme-Généreux. 2001. Preparation of the Escherichia coli Rne protein and reconstitution of the RNA degradosome. Methods Enzymol. 342:346-356.[Medline]
16
16. McClure, W. R. 1985. Mechanism and control of transcription initiation in prokaryotes. Annu. Rev. Biochem. 54:171-204.[CrossRef][Medline]
17
17. Miao, X. 2002. Modular domain organization of RNase E and PNPase. Ph.D. thesis. University of British Columbia, Vancouver, Canada.
18
18. Miczak, A., V. R. Kaberdin, C. L. Wei, and S. Lin-Chao. 1996. Proteins associated with RNase E in a multicomponent ribonucleolytic complex. Proc. Natl. Acad. Sci. USA 93:3865-3869.[Abstract/Free Full Text]
19
19. Muhlberg, A. B., and S. L. Schmidt. 2000. Domain structure and function of dynamin probed by limited proteolysis. Methods 20:475-483.[CrossRef][Medline]
20
20. Murzin, A. G. 1993. OB (oligonucleotide/oligosaccharide binding) fold: common structural and functional solution for non-homologous sequences. EMBO J. 12:681-867.
21
21. Prud'homme-Généreux, A., R. K. Beran, I. Iost, C. S. Ramey, G. A. Mackie, and R. W. Simons. 2004. Physical and functional interactions among RNase E, polynucleotide phosphorylase and the cold-shock protein CsdA: evidence for a ‘cold shock’ degradosome. Mol. Microbiol. 54:1409-1421.[CrossRef][Medline]
22
22. Py, B., H. Causton, E. A. Mudd, and C. F. Higgins. 1994. A protein complex mediating mRNA degradation in Escherichia coli. Mol. Microbiol. 14:717-729.[Medline]
23
23. Regonesi, M. E., F. Briani, A. Ghetta, S. Zangrossi, D. Ghisotti, P. Tortora, and G. Dehò. 2004. A mutation in polynucleotide phosphorylase from Escherichia coli impairing RNA binding and degradosome stability. Nucleic Acids Res. 32:1006-1017.[Abstract/Free Full Text]
24
24. Rosenzweig, J. A., G. Weltman, G. V. Plano, and K. Schesser. 2005. Modulation of Yersinia type three secretion system by the S1 domain of polynucleotide phosphorylase. J. Biol. Chem. 280:156-163.[Abstract/Free Full Text]
25
25. Schubert, M., R. E. Edge, P. Lario, M. A. Cook, N. C. J. Strynadka, G. A. Mackie, and L. P. McIntosh. 2004. Structural characterization of the RNase E S1 domain and identification of its oligonucleotide and dimerization interfaces. J. Mol. Biol. 341:37-54.[CrossRef][Medline]
26
26. Spickler, C., and G. A. Mackie. 2000. Action of RNase II and polynucleotide phosphorylase against RNAs containing stem-loops of defined structure. J. Bacteriol. 182:2422-2427.[Abstract/Free Full Text]
27
27. Steege, D. A. 2000. Emerging features of mRNA decay in bacteria. RNA 6:1079-1090.[Abstract]
28
28. Symmons, M. F., G. H. Jones, and B. F. Luisi. 2000. A duplicated fold is the structural basis for polynucleotide phosphorylase catalytic activity, processivity, and regulation. Struct. Fold Des. 8:1215-1226.[Medline]
29
29. Thang, M. N., L. Dondon, and T. Godefroy-Colburn. 1971. Degradation of Escherichia coli polynucleotide phosphorylase by E. coli endogenous proteases and by trypsin. Biochimie 53:291-302.[Medline]
30
30. Theobald, D. L., R. M. Mitton-Fry, and D. S. Wuttke. 2003. Nucleic acid recognition by OB-fold proteins. Annu. Rev. Biophys. Biomol. Struct. 32:115-133.[CrossRef][Medline]
31
31. Vanzo, N., Y. S. Li, B. Py, E. Blum, C. F. Higgins, L. C. Raynal, H. M. Krisch, and A. J. Carpousis. 1998. Ribonuclease E organizes the protein interactions in the Escherichia coli RNA degradosome. Genes Dev. 12:2770-2781.[Abstract/Free Full Text]
32
32. Worbs, M., G. P. Bourenkov, H. D. Bartunik, R. Huber, and M. C. Wahl. 2001. An extended RNA binding surface through arrayed S1 and KH domains in transcription factor NusA. Mol. Cell 7:1177-1189.[CrossRef][Medline]

---

Journal of Bacteriology, November 2005, p. 7214-7221, Vol. 187, No. 21
0021-9193/05/$08.00+0     doi:10.1128/JB.187.21.7214-7221.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Carzaniga, T., Briani, F., Zangrossi, S., Merlino, G., Marchi, P., Deho, G. (2009). Autogenous Regulation of Escherichia coli Polynucleotide Phosphorylase Expression Revisited. J. Bacteriol. 191: 1738-1748 [Abstract] [Full Text]  
• Shi, Z., Yang, W.-Z., Lin-Chao, S., Chak, K.-F., Yuan, H. S. (2008). Crystal structure of Escherichia coli PNPase: Central channel residues are involved in processive RNA degradation. RNA 14: 2361-2371 [Abstract] [Full Text]  
• Awano, N., Inouye, M., Phadtare, S. (2008). RNase Activity of Polynucleotide Phosphorylase Is Critical at Low Temperature in Escherichia coli and Is Complemented by RNase II. J. Bacteriol. 190: 5924-5933 [Abstract] [Full Text]  
• Buckner, B., Swaggart, K. A., Wong, C. C., Smith, H. A., Aurand, K. M., Scanlon, M. J., Schnable, P. S., Janick-Buckner, D. (2008). Expression and Nucleotide Diversity of the Maize RIK Gene. J Hered 0: esn013v1-esn013 [Abstract] [Full Text]  
• Portnoy, V., Palnizky, G., Yehudai-Resheff, S., Glaser, F., Schuster, G. (2008). Analysis of the human polynucleotide phosphorylase (PNPase) reveals differences in RNA binding and response to phosphate compared to its bacterial and chloroplast counterparts. RNA 14: 297-309 [Abstract] [Full Text]  
• Deikus, G., Bechhofer, D. H. (2007). Initiation of decay of Bacillus subtilis trp leader RNA. J. Biol. Chem. 282: 20238-20244 [Abstract] [Full Text]  
• Amblar, M., Barbas, A., Gomez-Puertas, P., Arraiano, C. M. (2007). The role of the S1 domain in exoribonucleolytic activity: Substrate specificity and multimerization. RNA 13: 317-327 [Abstract] [Full Text]  
• Ramos, C. R. R., Oliveira, C. L. P., Torriani, I. L., Oliveira, C. C. (2006). The Pyrococcus Exosome Complex: STRUCTURAL AND FUNCTIONAL CHARACTERIZATION. J. Biol. Chem. 281: 6751-6759 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stickney, L. M. Articles by Mackie, G. A. Search for Related Content PubMed PubMed Citation Articles by Stickney, L. M. Articles by Mackie, G. A.