| Nucleic Acids Research | Pages |
ATP hydrolysis activity of the DEAD box protein Rok1p is required for in vivo ROK1 function
Introduction
Materials And Methods
Strains and growth conditions
Transformation and DNA manipulation
Site-directed PCR mutagenesis
Plasmid constructions
Purification of wild-type and mutant forms of MBP-Rok1 proteins
SDS-PAGE and western blot analysis
ATPase assay
Results
Preparation of new mutations in the conserved domains of ROK1
In vivo analysis of the rok1 mutations
Purification and ATP hydrolysis activity of Rok1p
Characterization of the ATPase activities of the mutant Rok1 proteins
Discussion
Acknowledgements
References
ATP hydrolysis activity of the DEAD box protein Rok1p is required for in vivo ROK1 function
Received February 18, 1999; Revised and Accepted May 17, 1999
ABSTRACT The yeast ROK1 gene has been initially identified as a high copy plasmid suppressor of the kem1 null mutation and implicated in microtubule-mediated functions. Based on the deduced amino acid sequence of the ROK1 gene, Rok1p has been classified in the DEAD protein family of ATP-dependent RNA helicases. A subsequent report has suggested that Rok1p is required for rRNA processing. We report here the first study on the biochemical activity associated with Rok1p. The MBP-Rok1 hybrid protein was synthesized in Escherichia coli and purified by amylose affinity column and ion exchange chromatography. Rok1p has ATP hydrolysis activity. The significance of the conserved ATPase domains was addressed by generating a series of amino acid substitution mutations in these domains. Both in vivo lethality tests of the mutations and biochemical characterization of the mutant proteins suggest that ATP hydrolysis activity of Rok1p is essential for ROK1 function. The ATPase activity of Rok1p appears to be independent of single-stranded RNA. Furthermore, replacement of the first Arg in the HRIGR domain, the known RNA-binding domain, with Thr, Ile or Lys has no detectable effect on in vivo ROK1 function. The lack of RNA dependency and some of the mutational phenotypes of ROK1differentiate this gene from other members of the family.
INTRODUCTION
Unwinding of secondary RNA structures is an essential step in a variety of cellular processes, including translation, pre-mRNA splicing, RNA transport, ribosome assembly and RNA degradation (1,2). Among proteins known to be involved in these processes, a number of putative ATP-dependent RNA helicases have been identified. The criteria to identify them as ATP-dependent RNA helicases was their strong sequence homology to the prototype RNA helicase, the mammalian translation initiation factor eIF-4A (3-6). The DEAD protein family of RNA helicases is characterized by a central region of eight consensus motifs which include the characteristic DEAD (Asp-Glu-Ala-Asp) box (2). At present, functions have been attributed to four of eight motifs, mainly by the biochemical and mutational analysis of mammalian eIF-4A. The domain AxxGxGKT is a typical ATPase A motif and has been shown to be involved in initial ATP binding (4,7). The predicted ATPase B site (DEAD) is responsible for ATP hydrolysis and couples ATP hydrolysis to helicase activity. The SAT region is essential for RNA unwinding and the basic domain, HRIGRxxR, is required for RNA binding and ATP hydrolysis (4,8). Sequence divergence within the DEAD box gives rise to the DEAH and DExH helicase subgroups, which are more heterogeneous with respect to both sequence and biochemical function (1,3,9). Many viral helicases and the yeast splicing factors Prp2, Prp16, Prp22 and Prp43 are included in these subgroups.
The DEAD protein family includes about 60 proteins that originate from a wide range of organisms from bacteria to humans. Only some of these members (eIF-4A, p68, vasa and An3) have been shown to exhibit both ATP hydrolysis and RNA helicase activities in vitro, but most are regarded as putative ATP-dependent RNA helicases (1,2). For some proteins with no RNA helicase activity demonstrated, an RNA-dependent ATPase activity has been observed, implying that most of the putative RNA helicases could somehow exert their in vivo function in close association with the RNA molecule itself or by modulating the RNA structure. Since most RNA molecules play their critical roles by interacting with proteins or often as components of a ribonucleoprotein complex, putative RNA helicases might participate in regulation of these RNA-protein interactions or alteration of complex formation.
The Rok1 protein of Saccharomyces cerevisiae is a member of the DEAD box RNA helicase family (2,10). The amino acid sequence of Rok1p contains eight highly conserved domains found in the DEAD protein family and shows a striking similarity to a number of proteins included in this family. The ROK1 gene was originally identified as a high copy plasmid suppressor of the kem1 null mutation, which suppresses the kem1 mutation when carried on a high copy number plasmid but not on a low copy number plasmid (11). The KEM1 gene has been reported to affect microtubule and spindle pole body (SPB) functions during conjugation and mitotic cell growth (12,13). The SPB of S.cerevisiaefunctions as a microtubule-organizing center in a manner analogous to the centrosomes in higher eukaryotes (14,15). Microtubules emanating from the SPB participate in mitosis, meiosis and nuclear movement (16). Mutations in KEM1 lead to hypersensitivity to the microtubule-depolymerizing drug benomyl, chromosome loss and defects in SPB duplication or nuclear fusion. The Kem1 protein has subsequently been reported to be a microtubule-associated protein (13). Other groups have also reported that Kem1p (also known as Xrn1p) is a cytoplasmic exoribonuclease with possible roles in mRNA turnover (17-19). A nuclear exoribonuclease identified as Rat1p has been shown to be functionally interchangeable with Kem1p (19,20). Kem1p and Rat1p appeared to be restricted to, and required in, the nucleus and cytoplasm, respectively. Besides the exoribonuclease activity, other diverse functions have been suggested for KEM1, which include a DNA strand exchange activity (Sep1p), a G4 tetraplex-specific nuclease or a superkiller phenotype (SKI1) (21-24). At present, no coherent model has been proposed for this apparently multifunctional Kem1p.
The ROK1 gene, a high copy suppressor of the kem1 mutation, has been suspected of affecting the processes related to the cellular functions of KEM1. Disruption or overexpression of ROK1 causes cell cycle arrest at the unbudded stage (25). kem1 null mutations also cause a significant delay at the unbudded stage with a single SPB or duplicated/but unseparated SPBs (12,13). As in the case of KEM1, other functions related to RNA metabolism have been suggested for ROK1. Synthetic lethal screens using two rRNA processing genes have also identified the ROK1 gene (26). Depletion anlaysis of Rok1p demonstrates that Rok1p is required for the pre-rRNA cleavages which generate 18S rRNA. ROK1 along with KEM1 may play a critical role in pre-rRNA processing or participate in gene expression regulation, either by altering the level of mRNA or by affecting the translation efficiency (19,25). The latter model is supported by the finding that both Rok1p and Kem1p are cytoplasmic proteins (21,27,28). Whether or not their regulatory functions are specific to a subset of genes involved in the cell division cycle or cytoskeletal protein functions remains unclear.
To assess directly the functions of Rok1p as an ATP-dependent RNA helicase, we expressed the ROK1 gene in Escherichia coli as an MBP-Rok1 hybrid protein and purified it for biochemical analysis. Rok1p exhibits ATP hydrolysis activity. Amino acid substitution mutations were introduced in the ATPase A and B motifs and RNA binding motif, HRIGR, by site-directed mutagenesis. Effects of these mutations were analyzed by in vivo functional assays. Mutant Rok1 proteins were purified and their enzymatic activities were examined in comparison with wild-type Rok1p.
MATERIALS AND METHODS
Strains and growth conditions
Escherichia colistrain JM109 [lcub]recA1 supE44 endA1 hsdR17 gyrA96 relA1 thi[Delta](lac-proAB) [F[prime] traD36 proAB+ lacIq lacZ[Delta]M15][rcub] was used for synthesis of MBP-Rok1 hybrid proteins and strain DH5[alpha] [recA1 supE44 endA1 hsdR17 gyrA96 relA1 thi-1 [Delta]lacU169 ([Phi]80lacZ[Delta]M15)] was used to amplify plasmid DNA. Saccharomyces cerevisiae strain JK348 (MAT[alpha]ura3-52 leu2-3,112 trp1-289 can1 cyh2 Gal+ rok1::LEU2 [pJI299- TRP1 CEN1 GAL1::ROK1]) was constructed by using Gal+ strain L2861 from the Fink laboratory. Preparation of the E.coli and yeast media was done by established procedures (29,30). Galactose medium contains 2% filter-sterilized galactose instead of glucose as the sole carbon source.
Transformation and DNA manipulation
Escherichia coli transformation was performed by the method of Mandel and Higa (31). Yeast transformation was carried out according to the lithium acetate method of Ito et al. (32). DNA cloning and sequencing were performed using standard methods (30).
Site-directed PCR mutagenesis
Two DNA fragments overlapping at the targeted region were PCR amplified independently and fused together in a subsequent extension reaction (33). A set of four primers were designed for each mutagenesis. Two complementary mutagenizing primers (primers II and III) contain nucleotide mismatches at each mutation site and two outside primers (primers I and IV) carry appropriate restriction sites for convenience in the subsequent cloning procedure. For the mutagenesis of Gly166 and Lys172 in the ATPase A domain, we used the synthetic oligonucleotides 5[prime]-ttgaggaattcacactgtttatgg-3[prime] and 5[prime]-tctaacaggatccatcattat-3[prime] as outside primers I and IV, respectively, where EcoRI and BamHI sites have been underlined. The mutagenizing primers II and III are 5[prime]-agaccctgtgg-ga(g,a,t)tcccacagggtct-3[prime] and 5[prime]-cgtttggcatgtg(c,t,a)tcccacagggtct-3[prime] for Gly166 and 5[prime]-taaggaatgggctaaaacgtcc(g,a,c)-(g,a,c)accagaccctgtggg-3[prime] and 5[prime]-cccacaggtctggt(g,t,c)(g,t,c)-gacgttaaagcattctta-3[prime] for Lys172. For the mutagenesis of Asp280 in the ATPase B domain, we used oligonucleotide primers I (5[prime]-ttgaggaattcacactgtttatgg-3[prime]), II [5[prime]-caatttatcaagccttca(g,a,c)caaatatcaaatg-3[prime]], III [5[prime]-catttgaataatttg(g,t,c)tgaagctga-taaattg-3[prime]] and IV (5[prime]-tgaagaaatatgagctcaa-3[prime]). For the mutagenesis of Arg 459, we used oligonucleotide primers I (5[prime]-gtataatgatggatcctg-3[prime]), II [5[prime]-ttctaccaatt(g,t,a)tgtgaacataag-3[prime]], III [5[prime]-ctta-tgttcaca(c,t,a)aattggtagaa-3[prime]] and IV (5[prime]-tgaagaatatgagctcaa-3[prime]). The products from two initial PCR reactions were pooled together in equimolar concentrations, denatured and allowed to re-anneal to one another in the overlapping regions. The annealed mutant DNA fragments were extended and subsequently amplified by using the outside primers I and IV. The final PCR products were digested with EcoRI-BamHI (Gly166 and Lys172), EcoRI-SacI (Asp280) or BamHI-SacI (Arg459). The resulting DNA fragments were used to replace a cognate fragment in plasmid pJI221. Mutations as well as PCR-amplified DNA regions were confirmed by DNA sequencing.
Plasmid constructions
Plasmid pJI221, used for PCR mutagenesis, was constructed by isolating a 1.7 kb KpnI-SacI fragment containing ROK1 from pJI216 and inserting it into the KpnI and SacI sites of pRS316 (34). Plasmid pMBP-Rok1 (pJI222) was constructed by using fusion vector pMAL-c2, which contains malE (encoding maltose-binding protein, MBP) under the control of the E.coli tac promoter. A 1.7 kb EcoRI-SalI fragment carrying the ROK1 coding region was PCR amplified from pJI221 to generate an EcoRI site 5[prime] to the start AUG and a SalI site 3[prime] to the UAA stop. The primers were 5[prime]-ggggaattcatggatatttttagagta-3[prime] and 5[prime]-ggggtcgacttatttcgagaaatgttttttttgaaag-3[prime] (EcoRI and SalI restriction sites are underlined). The integrity of the ROK1 sequence was confirmed by DNA sequencing. Mutant plasmids pMBP-G166D, K172L and K172R were constructed by following the same procedure as with wild-type pMBP-Rok1.
Purification of wild-type and mutant forms of MBP-Rok1 proteins
The MBP-Rok1 gene was expressed in E.coli JM109 following IPTG induction. Crude extracts were prepared from 400 ml cultures by sonication with 10 ml of column buffer A (20 mM HEPES, pH 7.2, 200 ml NaCl, 1 mM EDTA, 0.5 mM DTT) containing 0.1 mM PMSF, 1 µg/ml leupeptin, 10 µg/ml aprotinin and 1 µg/ml peptistatin A. All steps were performed at 4°C. The extracts were cleared by centrifugation for 30 min at 10 000 g. The supernatant was diluted with 40 ml of buffer A, applied to an amylose resin column (15 ml volume; New England Biolabs) equilibrated with buffer A and washed with 12 column vol of buffer A. The bound proteins were eluted with buffer A containing 10 mM maltose. The Rok1p-containing fractions were confirmed by SDS-PAGE and western blot analysis with anti-Rok1 antiserum. The fractions containing Rok1p were pooled, diluted with 5 vol of buffer A, loaded onto a 10 ml High S resin column (a strong cation exchanger; BioRad) equilibrated with buffer A containing 200 mM NaCl. Elutions were performed with a 6 vol gradient of 200 mM to 1 M NaCl in buffer A. The Rok1p-containing fractions were dialyzed against 500 vol of 20 mM HEPES, pH 7.2, 0.5 mM DTT, 20% glycerol overnight and stored at -20°C. The protein concentration was determined by a spectrophotometric method using Bradford solution (BioRad).
SDS-PAGE and western blot analysis
Proteins were separated by SDS-PAGE as described (35) and were electrophoretically transferred to nitrocellulose sheets using a semi-dry transfer set (TE70; Hoeffer). Immunoblots were blocked for 2 h with a solution containing 5% instant non-fat milk, 0.2% Tween 20 in TBS (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 0.5% sodium azide). Rok1 antiserum was added (1:1000 dilution) and incubated for 1 h. Rinsed blots were incubated for 1 h with an anti-rabbit IgG conjugated to horseradish peroxidase (1:2000 dilution; Amersham). Proteins were visualized with an ECL kit (Amersham).
ATPase assay
ATP hydrolysis activity was determined by monitoring [[alpha]-32P]ATP hydrolysis by thin layer chromatography (TLC) (36). Reactions were carried out in a total volume of 10 µl which contained 50 mM HEPES, pH 7.0, 2.5 mM MgCl2, 1 µCi [[alpha]-32P]ATP (3000 Ci/mmol) and various amounts of protein sample. After incubation for 15 min at 37°C, reactions were terminated by addition of EDTA to a final concentration of 20 mM. One microliter of the reaction product was analyzed by polyethyleneimine-cellulose TLC in 0.375 M potassium phosphate, pH 3.5. The sheets were dried and exposed on X-ray film.
RESULTS
Preparation of new mutations in the conserved domains of ROK1
Rok1 protein contains eight highly conserved regions found in the DEAD protein family, which include the ATPase A motif (AxxGxGKT), ATPase B motif (DEAD) and RNA-binding motif (HRIGR) (Fig. 1). In our previous work, we generated an amino acid substitution mutation (D280E) of ROK1 which replaced the first Asp with Glu in the DEAD domain (34). The D280E mutation did not complement the lethality caused by the rok1 null mutation, suggesting that the DEAD domain is essential for ROK1 function.
Figure 1. Schematic representations of the conserved motifs in the DEAD box family and Rok1p. Mutations introduced into the conserved regions of Rok1p are shown below the corresponding motifs. The number in each mutation indicates the position of the amino acid residue in Rok1p.
To investigate further the contributions of the conserved amino acid sequences to Rok1 protein function, we have constructed a series of new mutations by site-directed mutagenesis. As summarized in Figure 1, the amino acids in the conserved regions were replaced with several different amino acid residues. Our investigation of the first domain, the ATPase A motif (GxTGxGKT) of Rok1 protein, focused on Gly166 and Lys172. The majority of DEAD proteins have Ala at position Gly166, whereas the DEAH proteins Prp2, Prp16 and Prp22, as well as most GTP-binding proteins, have a Gly at this position (2,4). To determine the significance of Gly166 for Rok1p, it was replaced by Ala (G166A), Val (G166V) or Asp (G166D). Lys172 is invariably present in all nucleotide-binding proteins and is thought to be directly involved in binding to the [beta] and [gamma] phosphates of the bound nucleotide (4). We mutated Lys172 to a conservative amino acid residue Arg (K172R) or to four other amino acid residues (K172G, K172L, K172S or K172W).
The second region of interest is the ATPase B motif (DEAD). Previously, a conservative substitution of the first Asp (D280E) was shown to abolish the ROK1 complementing activity (34). In our present study, we introduced a non-charged hydrophobic residue, Ala or Val, at this position (D280A or D280V).
It has been suggested that the basic domain, HRIGR, is involved in the interaction with RNA (1,8). To determine the significance of this domain in Rok1 function, we replaced the first Arg residue with Thr, Ile or the basic amino acid Lys (R459T, R459I or R459K).
In vivo analysis of the rok1 mutations
To analyze in vivo the effects of amino acid substitution mutations on ROK1 function, we assayed the growth phenotypes of these mutations by using a ROK1 conditional system. Since ROK1 had been shown to be essential for cell viability, a ROK1 conditional system was constructed in such a way that the chromosomal copy of ROK1 was disrupted by the LEU2 gene and a plasmid copy of ROK1 is expressed under the control of a galactose-inducible promoter, GAL1-10. To avoid ROK1 overexpression in the induced condition, we used a centromere-based plasmid pYCplac22 for pGAL-ROK1 construction. The conditional yeast strain grows normally in galactose medium, but is unable to form colonies in glucose medium. The lethal phenotype of this strain in glucose medium was restored when a second plasmid carrying a wild-type ROK1 gene was introduced into the cell.
Each prepared mutant plasmid was transformed into the ROK1 conditional strain and its growth phenotype was assayed in galactose- and glucose-containing medium (Table 1). As for the ATPase A motif (G/A166xTGxGK172T), mutations G166D and K172L showed the most severe growth defect on glucose medium. Replacement of Lys172 by any of four other amino acids, including a basic residue, Arg, led to a slow growth phenotype. These results indicate that the ATPase A motif of Rok1p is essential for in vivo ROK1 function. The Lys residue in the ATPase A motif has not been characterized extensively in the DEAD protein family. A report has shown that substitution of this Lys residue with Asn abolished ATP binding to eIF-4A (37). Various substitution mutations have been introduced at the same Lys residue in the Upf1 protein, a DNA/RNA helicase of another helicase superfamily, which resulted in a lethal effect (38). This Lys residue is thought to be directly involved in nucleotide binding in mammalian Ras and other GTP-binding proteins (39,40). Based on these reasonings and our observations, the Lys172 residue appears to be critical for Rok1p.
Table 1. Effects of rok1 mutations on cell growth in a ROK1 conditional strain
| Mutation | Growth | |
| On galactose | On glucose | |
| Wild-type | + | + |
| G166A | + | + |
| G166D | + | - |
| G166V | + | + |
| K172L | + | - |
| K172G | + | +/- |
| K172R | + | +/- |
| K172S | + | +/- |
| K172W | + | +/- |
| D280A | + | +/- |
| D280V | + | +/- |
| R459T | + | + |
| R459I | + | + |
| R459K | + | + |
The Gly166 residue in the A motif of Rok1p turned out to be interchangeable with Ala (G166A). This result is consistent with the fact that the majority of the DEAD box family have Ala at this position and the reverse substitution Ala->Gly in eIF-4A is neutral (4,41). The presence of Val at this position (G166V), which has been shown to abolish the function of mammalian eIF-4A and yeast TIF1, seems not to affect Rok1p function.
The presence of an Ala or a Val residue at position 280 of the ATPase B motif (D280A or D280V) results in a slow growth phenotype. As for the HRIGR motif, the substitution mutations R459T, R459I and R459K showed almost a wild-type growth pattern. This Arg residue of mammalian eIF-4A has been reported to be critical in ATP hydrolysis and RNA-binding activities and substitution of the same Arg residue of the yeast eIF-4A gene, TIF1, to Thr or Ile are lethal to protein function (4,41). Therefore, our observations led to the speculation that the mode of RNA interaction utilized by Rok1p might be different from those of the eIF-4A proteins.
Purification and ATP hydrolysis activity of Rok1p
Proteins of the DEAD family are thought to exhibit ATP-dependent RNA unwinding activity. At present, this activity has been demonstrated in vitro for only a few members of the family, eIF4-4A, p68, vasa and Xenopus An3 (1,2,9,42,43). To assess the enzymatic activity of Rok1 protein, we purified Rok1p from an E.coli strain expressing the MBP::rok1 recombinant gene under control of the tac promoter. The MBP::rok1 gene contains the full-length ROK1 ORF fused to the C-terminus of the E.coli MBP gene. The recombinant MBP-Rok1 protein of 102 kDa molecuar weight was synthesized in E.coli cells (Fig. 2A, lane 1) and partially purified by affinity chromatography using amylose resin (Fig. 2A, lanes 2-8). This affinity-purified MBP-Rok1p was purified further to >95% homogeneity by cation exchange chromatography (Fig. 2B). Western analysis using authentic anti-Rok1 antibodies verified that the purified major protein band was the Rok1 protein (data not shown).
Figure 2. Purification of MBP-Rok1p. (A) Amylose affinity chromatography of Rok1p. Total protein extract from an induced culture (lane 1) and aliquots (16 µl) of 3 ml fractions eluted from the amylose column (lanes 2-8) were separated by SDS-PAGE and Coomassie blue stained. (B) Cation exchange chromatography of Rok1p. Bound proteins were eluted with a gradient of 200 mM to 1 M NaCl. Aliquots (16 µl) of even numbered fractions (3 ml) 2-18 were separated by SDS-PAGE and Coomassie blue stained. (C) ATP hydrolysis activity. Aliquots (4 µl) of the indicated gradient fractions were assayed for ATPase activity. ATPase assays were performed with 0.3 pmol of [[alpha]-32P]ATP for 15 min.
Aliquots of the Rok1p fractions from cation exchange chromatography were tested for the ability to catalyze the hydrolysis of ATP. ATP hydrolysis was measured as the production of [32P]ADP from [[alpha]-32P]ATP on polyethyleneimine-cellulose TLC plates (36). ATPase activity was detected, showing a single peak of activity that coincided with the abundance of Rok1 protein (Fig. 2C). This result indicates that purified Rok1 protein has ATPase activity. This ATPase activity of Rok1p requires the divalent ion Mg2+ but was not stimulated by the presence of poly(U) single-stranded RNA (Fig. 3B, lanes 4 and 5). We obtained the same results using other single-stranded RNA [poly(A), poly(G) or poly(C); data not shown].
Figure 3. ATPase activity of the wild-type and mutant Rok1p. (A) SPS-PAGE of the purified mutant Rok1p. The mutant forms of MBP-Rok1p were purified by the same procedure as the wild-type. (B) ATPase activity. Lane 1, 0.5 µg MBP; lanes 2 and 3, 0.1 and 0.3 µg wild-type Rok1p, respectively; lane 4, 0.3 µg wild-type Rok1p + 1 µg poly(U); lane 5, 0.3 µg wild-type Rok1p (-MgCl); lanes 6 and 7, 0.1 and 0.3 µg K171L mutant Rok1p, respectively; lanes 8 and 9, 0.1 and 0.3 µg K172R, respectively; lanes 10 and 11, 0.1 and 0.3 µg G166D, respectively.
Characterization of the ATPase activities of the mutant Rok1 proteins
To verify the contributions of the conserved residues of Rok1p to the ATPase activity, we purified the mutant Rok1 proteins K172L, K172R and G166D. These mutations were those with either a lethal or slow growth phenotype in the in vivo analysis. The three mutant Rok1 proteins were purified by following the same procedure as for wild-type Rok1p (Fig. 3A). The K172L mutant protein, which resulted in the most severe growth defect in vivo, showed a drastic reduction in ATP hydrolysis activity (Fig. 3B, lanes 6 and 7), whereas the activity of the K172R mutant protein was similar to the wild-type. The mutation G166D, with a slow growth phenotype, resulted in a 2-fold decrease in ATPase activity. Our enzymatic analyses of the mutant proteins are in good agreement with in vivo phenotypic analyses. These results indicate that the ATPase activity is essential for ROK1 function.
DISCUSSION
We present in this paper the first study on the biochemical activity associated with yeast Rok1 protein, a member of the DEAD box RNA helicase family. The purified MBP-Rok1 hybrid protein has ATP hydrolysis activity. The significance of the conserved ATPase domains was addressed by generating a series of amino acid substitution mutations in these domains and analyzing their phenotypes. Both in vivo lethality tests of the mutations and biochemical characterization of the mutant proteins suggest that the ATPase activity of Rok1p is essential for ROK1 function. The mutations K172L and G166D, which resulted in the most severe growth defect, showed drastic reduction in ATPase hydrolysis activity (96 and 40% reduction, respectively), whereas the mutation K172R, with only a slight defect in growth phenotype, affected the enzymatic activity of Rok1p to a negligible degree. The good agreement between in vivo mutant phenotype and the enzymatic activity for the mutant proteins strongly indicates that the ATP hydrolysis activity of Rok1p is required for its primary role in the cell.
Several mutational analyses have been carried out to investigate the functions of each amino acid residue in the ATPase domains of the DEAD box RNA helicases. Biochemical studies of the mutant proteins have been done mainly on mammalian eIF-4A and genetic in vivo analysis of mutations in the conserved domains done mostly on yeast eIF-4A and recently on the yeast splicing gene PRP28 (4,8,41,44). The data presented in this work are very informative in the regard that both biochemical and genetic in vivo characterization of mutations in the ATPase domains were carried out on Rok1p.
The effects of the substitution mutations in the conserved domains of Rok1p seem to be consistent in general with the data presented previously on other members of the family. However, we can observe slight differences in mutational phenotypes of the ATPase A motif (G/A166xTGxGK172T) or HRIGR domain. The finding that the presence of either Gly or Ala at residue 166 in the ATPase A motif of Rok1p led to a functional protein was somewhat expected. However, Val substitution at this position (G166V) also led to a wild-type phenotype. The same mutations disrupted the functions of both mammalian eIF-4A and yeast TIF1. The Lys172 residue of Rok1p has been suspected of being critical in ATP hydrolysis activity. We observed that only the mutation K172L showed lethality and a reduction in ATPase activity, while other substitutions (K172G, K172R, K172S and K172W) still produced functional proteins. Substitution mutations of the first Arg residue in the HRIGR domain of yeast TIF1 and mouse eIF-4A are lethal to protein function (4,41). But we have shown that replacement of this Arg with Thr, Ile or Lys has no detectable effect on in vivo ROK1 function. The differential phenotypes among proteins of the DEAD box helicase family suggest that the tertiary structures of these proteins in a cell might be diverse and dynamic, especially in association with a nucleotide, ATP, or single-stranded RNA. In general, proteins of the RNA helicase family have distinct N- and C-termini. Rok1p has relatively long ends (165 and 62 amino acids, respectively). These distinct sequences may contibute to the diverse protein structures or provide a domain for interaction with other auxiliary proteins.
The ATPase activity of Rok1p appears to be independent of single-stranded RNA. The addition of a variety of single-stranded RNA at a saturating concentration did not stimulate the ATP hydrolysis activity of Rok1p. It is possible that any RNA co-purified in the Rok1p preparation could provide a high background stimulation of ATPase activity. Treatment of the purified Rok1p with ribonuclease did not give any difference in ATPase activity (data not shown). These observations are particularly interesting because members of the DEAD protein family which have been shown to exhibit ATPase activity are mostly RNA dependent. We have also shown that replacement of the first Arg in the HRIGR domain, the known RNA-binding domain, with Thr, Ile or Lys has no detectable effect on in vivo ROK1 function. This Arg residue of mouse eIF-4A has been reported to be critical in ATP hydrolysis and RNA binding and substitution of the same Arg residue of the yeast eIF-4A gene, TIF1, with Thr or Ile is lethal to protein function (4,41). Therefore, our observations led to the speculation that the mode of RNA interaction utilized by Rok1p might be different from the eIF-4A proteins or that Rok1p might be deficient in RNA-interacting function. An auxiliary interacting factor may be required for this activity.
We attempted to test the RNA helicase activity of Rok1p by using three different double-stranded RNA substrates. Although the HCV NS3 protein, a known RNA helicase used as a positive control in our assays, exhibits RNA unwinding activity, Rok1p did not show any RNA helicase activity (data not shown). Among the DEAD box proteins whose enzymatic activities have been tested in vitro, only a few (eIF-4A, p68, vasa and An3) have been shown to exhibit both ATPase and RNA unwinding activities (1,2,9,42). Some proteins (Prp5p, Prp28p, etc.) have been shown to possess RNA-dependent ATPase activity only (1). It is possible that these putative RNA helicases, including Rok1p, which lack evidence for RNA unwinding activity may require an interacting factor for their enzymatic function. This has been the case for mammalian eIF-4A, whose helicase activity requires an activation cofactor, eIF-4B (42). We are currently investigating a few proteins that appear to interact with Rok1p.
The DEAD protein family of ATP-dependent RNA helicases has been classified by the presence of highly conserved domains and the amino acid similarity to the prototype of this group, eIF-4A. Although about 60 members of this family have been reported (more than 26 proteins are from S.cereivisiae), only a few proteins have been purified and characterized biochemically (44). The genetic and biochemical studies presented here demonstrate a strong link between ATPase activity and in vivo ROK1 function. The lack of RNA dependency and some of the mutational phenotypes of ROK1 differentiate this gene from other members of the family. One possible idea has been raised, that Rok1p may express its functions, especially RNA-dependent functions, in association with other auxiliary factors.
ACKNOWLEDGEMENTS
We are grateful to Professor Joonho Choe and Dr Dong Wook Kim (Korea Advanced Institute of Science and Technology) for useful discussions and technical help. This work was supported by the academic research fund (GE96-91) to J.K. from the Ministry of Education, Republic of Korea.
REFERENCES
*To whom correspondence should be addressed. Tel: +82 42 821 6416; Fax: +82 42 822 7367; Email: jmkim{at}hanbat.chungnam.ac.kr
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. M. S. Elles and O. C. Uhlenbeck Mutation of the arginine finger in the active site of Escherichia coli DbpA abolishes ATPase and helicase activity and confers a dominant slow growth phenotype Nucleic Acids Res., January 17, 2008; 36(1): 41 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Granneman, C. Lin, E. A. Champion, M. R. Nandineni, C. Zorca, and S. J. Baserga The nucleolar protein Esf2 interacts directly with the DExD/H box RNA helicase, Dbp8, to stimulate ATP hydrolysis Nucleic Acids Res., June 13, 2006; 34(10): 3189 - 3199. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Granneman, K. A. Bernstein, F. Bleichert, and S. J. Baserga Comprehensive Mutational Analysis of Yeast DEXD/H Box RNA Helicases Required for Small Ribosomal Subunit Synthesis Mol. Cell. Biol., February 15, 2006; 26(4): 1183 - 1194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Bernstein, S. Granneman, A. V. Lee, S. Manickam, and S. J. Baserga Comprehensive Mutational Analysis of Yeast DEXD/H Box RNA Helicases Involved in Large Ribosomal Subunit Biogenesis Mol. Cell. Biol., February 15, 2006; 26(4): 1195 - 1208. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Rocak, B. Emery, N. K. Tanner, and P. Linder Characterization of the ATPase and unwinding activities of the yeast DEAD-box protein Has1p and the analysis of the roles of the conserved motifs Nucleic Acids Res., February 17, 2005; 33(3): 999 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kikuma, M. Ohtsu, T. Utsugi, S. Koga, K. Okuhara, T. Eki, F. Fujimori, and Y. Murakami Dbp9p, a Member of the DEAD Box Protein Family, Exhibits DNA Helicase Activity J. Biol. Chem., May 14, 2004; 279(20): 20692 - 20698. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-C. Daugeron and P. Linder Characterization and mutational analysis of yeast Dbp8p, a putative RNA helicase involved in ribosome biogenesis Nucleic Acids Res., March 1, 2001; 29(5): 1144 - 1155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Kressler, P. Linder, and J. de la Cruz Protein trans-Acting Factors Involved in Ribosome Biogenesis in Saccharomyces cerevisiae Mol. Cell. Biol., December 1, 1999; 19(12): 7897 - 7912. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||