A minimal region in the NTPase/helicase domain ... - Semantic Scholar

12 downloads 0 Views 206KB Size Report
Donald, R. G. K., Lawrence, D. M. & Jackson, A. O. (1997). The barley stripe mosaic virus ... Howard, A. R., Heppler, M. I., Ju, H. J., Krishnamurthy, K., Payton,.
Journal of General Virology (2006), 87, 3087–3095

DOI 10.1099/vir.0.81971-0

A minimal region in the NTPase/helicase domain of the TGBp1 plant virus movement protein is responsible for ATPase activity and cooperative RNA binding Anna D. Leshchiner, Andrey G. Solovyev, Sergey Yu. Morozov and Natalia O. Kalinina A. N. Belozersky Institute of Physico-Chemical Biology, Moscow State University, Moscow 119992, Russia

Correspondence Natalia O. Kalinina [email protected]

Received 22 February 2006 Accepted 23 May 2006

The TGBp1 protein, encoded in the genomes of a number of plant virus genera as the first gene of the ‘triple gene block’, possesses an NTPase/helicase domain characterized by seven conserved sequence motifs. It has been shown that the TGBp1 NTPase/helicase domain exhibits NTPase, RNA helicase and RNA-binding activities. In this paper, we have analysed a series of deletion and point mutants in the TGBp1 proteins encoded by Potato virus X (PVX, genus Potexvirus) and Poa semilatent virus (PSLV, genus Hordeivirus) to map functional regions responsible for their biochemical activities in vitro. It was found that, in both PVX and PSLV, the Nterminal part of the TGBp1 NTPase/helicase domain comprising conserved motifs I, Ia and II was sufficient for ATP hydrolysis, RNA binding and homologous protein–protein interactions. Point mutations in a single conserved basic amino acid residue upstream of motif I had little effect on the activities of C-terminally truncated mutants of both TGBp1 proteins. However, when introduced into the full-length NTPase/helicase domains, these mutations caused a substantial decrease in the ATPase activity of the protein, suggesting that the conserved basic amino acid residue upstream of motif I was required to maintain a reaction-competent conformation of the TGBp1 ATPase active site.

INTRODUCTION Cell-to-cell movement of plant viruses is enabled by virusencoded non-structural proteins termed ‘movement proteins’ (MPs) (Lucas, 2006). Some plant virus genera belonging to the ‘alpha-like’ supergroup of RNA viruses encode a distinct transport gene module, the triple gene block (TGB). The three partially overlapping TGB genes code for a conserved set of MPs referred to as TGBp1, TGBp2 and TGBp3 (Solovyev et al., 1996; Morozov & Solovyev, 2003). The two smaller proteins (TGBp2 and TGBp3) possess hydrophobic segments and bind to intracellular membrane compartments (Morozov & Solovyev, 2003; Zamyatnin et al., 2004; Haupt et al., 2005; Verchot-Lubicz, 2005; Schepetilnikov et al., 2005). The largest TGB protein, TGBp1, contains an NTPase/helicase sequence domain (Morozov et al., 1989; Koonin & Dolja, 1993). Two types of TGBp1 proteins are distinguished: in the potex-like TGBp1 encoded by filamentous viruses of the Potexviridae, the NTPase/ helicase domain comprises the entire TGBp1 sequence, whereas the hordei-like TGBp1 (rod-shaped viruses of the genera Hordeivirus, Benyvirus, Pomovirus and Pecluvirus) also includes an N-terminal extension domain (Morozov & Solovyev, 2003). 0008-1971 G 2006 SGM

Printed in Great Britain

In accordance with predictions based on sequence analysis, the protein region comprising the NTPase/helicase domain in both potex-like and hordei-like TGBp1s exhibits NTPase activity and RNA helicase activity in vitro (Rouleau et al., 1994; Donald et al., 1997; Kalinina et al., 1996, 2002; Liou et al., 2000). The TGBp1 NTPase/helicase sequence domain is closely related to replicative helicases of ‘alpha-like’ viruses and belongs to helicases of the superfamily 1 (SF1) (Gorbalenya et al., 1989; Gorbalenya & Koonin, 1993; Koonin & Dolja, 1993). This domain has seven typical sequence motifs, of which motif I and motif II correspond to the ‘Walker A’ and ‘Walker B’ sites found in numerous ATPbinding proteins (Gorbalenya & Koonin, 1993; Leipe et al., 2002). Sequence comparisons of TGBp1 with the related DNA helicases of SF1, for which the X-ray structure has been resolved (Caruthers & McKay, 2002), reveal that TGBp1 contains only two of the four structural domains found in SF1 DNA helicases (Kalinina et al., 2002; Morozov & Solovyev, 2003). Experimental data suggest that ATPase activity is required for cell-to-cell movement, particularly for plasmodesmata dilation by the potex-like TGBp1 (Angell et al., 1996; Lough et al., 1998, 2000; Malcuit et al., 1999; Morozov et al., 1999; 3087

A. D. Leshchiner and others

Yang et al., 2000; Howard et al., 2004). The TGBp1 helicase activity is involved in protein/RNA translocation through plasmodesmata to the adjacent cell for both types of viruses with TGB (Morozov & Solovyev, 2003). TGBp1 of Potato virus X (PVX) has also been shown to be a suppressor of RNA silencing (Voinnet et al., 2000). Recently, random mutation analysis of PVX TGBp1 suggested that silencing suppression is essential for the protein to mediate viral cell-to-cell movement (Bayne et al., 2005). TGBp1 was concluded to possess at least two functions required for viral cell-to-cell movement: one is related to the ability of the protein to suppress RNA silencing, while another is responsible for movement per se. The ATPase activity of TGBp1 might be required for suppression of silencing. In this study, we have further investigated TGBp1 functional regions. We have shown that the N-terminal part of the TGBp1 proteins of both virus types (potex-like and hordeilike) including three NTPase/helicase motifs, I, Ia and II, and a small sequence upstream of motif I (a compact region of about 100 amino acid residues) is sufficient for ATPase and cooperative RNA-binding activities. Additionally, the functional activities of a conserved positively charged residue upstream of motif I in the NTPase/helicase domain were analysed.

METHODS Construction of recombinant clones. All recombinant DNA

procedures were carried out by standard methods. PVX TGBp1 (25K protein) with motifs III–VI deleted (25KI–II) was obtained by PCR amplification with primers 59-TTGTGAGCGGATAACAATTTC-39 and 59-CGCTCTAGATTAGTTAGAGTTCCTTGTGGTG-39 using pQE-25K (66His tag-containing recombinant 25K clone in pQE30; Kalinina et al., 1996) as a template. The 39 primer contained an XbaI site (in italics). The PCR product was digested with EcoRI/ XbaI and ligated into vector pQE30 (66 His-tagged expression vector; Qiagen) cut by EcoRI/XbaI. A mutant containing the C-terminal part of TGBp1 of Poa semilatent virus (PSLV) (63K protein) with motifs III–VI deleted (C63KI–II) was obtained similarly: the PSLV C63K coding region was amplified with primers 59-TTGTGAGCGGATAACAATTTC-39 and 59-GCTCTAGATTACAAGAGAATTTCAGCGCTTTC-39 using pQE-C63K (Kalinina et al., 2001) as a template. The PCR product was cloned in the pQE31 vector (Qiagen) using EcoRI/XbaI sites. 25K clones containing point mutations (25KR/A and 25KR/AI–II) were obtained by PCR amplification with the primers 59-ATACCATGGATATTCTCATCAGTAGTTTGAAAAGTTTAGGTTATTCTGCGACTTCC-39 and either 59-AGCGTCTAGACTATGGCCCTGCGCGGACA-39 with pQE-25K or 59-CGCTCTAGATTAGTTAGAGTTCCTTGTGGTG-39 with pQE-25KI–II as templates. The resulting PCR products were cut with NcoI (restriction site shown in bold) and XbaI (in italics) and then ligated into a modified pQE31 vector, where an NcoI site was inserted in the polylinker. To obtain this plasmid, the pQE31 vector was fully digested with BamHI/HindIII, removing the original polylinker. Two partially overlapping primers with a new NcoI site (shown in bold) (59-GATCCCGGGCCCATGGA-39 and 59AGCTTCCATGGGCCCGG-39) were then annealed and ligated into the vector instead of the original polylinker. Two constructs were produced using similar procedures: pQE-25KR/A and pQE-25KR/AI–II. 3088

Point mutations in the C63K protein (mutants C63KK/A and C63KK/AI– II) were introduced by PCR amplification of pQE-C63K with two sets of primers, 59-TTGTGAGCGGATAACAATTTC-39/59-ACGGGCCCTCAGTTGATTTCTTGGTCG-39 (ApaI site in bold) and 59-GAGGGCCCGTGGTTACTCGTTAGGAAC-39/59-GCTCTAGATTACAAGAGAATTTCAGCGCTTTC-39 (XbaI site in italics), respectively. To obtain the construct pQE-C63KK/AI–II, the resulting PCR products were digested with EcoRI/ApaI and ApaI/XbaI, respectively, and then cloned into pQE31 digested with EcoRI/XbaI. For the construct pQEC63KK/A, a slightly different strategy was used. The same PCR products were cut with EcoRI/ApaI and ApaI/SalI (the SalI site is present in the original gene sequence) and then ligated into plasmid pQE-C63K cut with EcoRI/SalI. All plasmids were transformed into Escherichia coli XL-1 cells. All mutants were verified by restriction analyses and sequencing. Expression, purification of His-tagged recombinant TGBp1, SDS-PAGE and Western blot analyses of proteins. E. coli

strain M15 transformed with the recombinant vectors was grown at 37 uC in liquid culture until an OD600 of 0?8–0?9 was reached. Expression of the proteins was induced with 1 mM IPTG followed growth for 2–3 h at 37 uC. The purification of recombinant proteins from cultures was performed using a general procedure described by the manufacturer (Qiagen) for denaturing Ni-NTA chromatography. Proteins were fractionated by SDS-PAGE and stained with Coomassie blue. In several experiments, proteins were transferred onto a nitrocellulose membrane after electrophoresis and probed in a standard Western-blot assay using polyclonal rabbit antibodies raised against the recombinant 25K and 63K proteins. ATP hydrolysis assay. Reaction mixtures contained 10 mM Tris/ HCl, pH 8?0, 10 % glycerol, 1 mM DTT, 5 mM MgCl2, 1 mCi [c-32P]ATP (6000 Ci mmol21) and between 50 and 300 ng dialysed protein in a final volume of 10 ml. Probes were incubated for 1 h at

37 uC and the reaction was stopped by the addition of EDTA to a final concentration of 20 mM. To estimate ATPase activity, the unreacted ATP was precipitated by the addition of 200 ml 7?5 % activated charcoal in 50 mM HCl/5 mM H3PO4; the mixtures were vortexed and allowed to stand for 5 min and the charcoal was then eliminated by centrifuging for 10 min. Half of the supernatant was then analysed by Cherenkov counting for free [32P]phosphate released by hydrolysis of the labelled ATP. The sample incubated without protein was used as a negative control. The total sample activity without incubation and precipitation (minus a negative control sample) was taken as 100 %. ATPase activity of the proteins was expressed as relative activity (percentage hydrolysis of 32P-labelled substrate). Kinetic parameters for ATP hydrolysis were measured in the linear phase of the reaction as described previously (Rakitina et al., 2005). Briefly, the standard [c-32P]ATP-containing reaction mixture was supplemented with 100 or 200 ng protein and different concentrations of unlabelled ATP varying from 5 to 70 mM and 1 mCi (8 nM) [c-32P]ATP. After 30 min incubation, ATP hydrolysis was analysed using the activated charcoal method as described above. Lineweaver– Burke double-reciprocal plots were used to calculate the Km and Vmax of ATP hydrolysis. RNA-binding assay (gel-shift assay). RNA (0?5 mg) from

Tobacco mosaic virus (TMV) was incubated with various concentrations of protein for 15–30 min at room temperature in 15 ml gelshift buffer (10 mM Tris/HCl, pH 8?0, 5 mM MgCl2, 1 mM EDTA, 1 mM DTT and 50 mM NaCl). Samples were analysed in ethidium bromide-containing 1 % agarose gels in 16 Tris/acetate buffer. Application of Hill transformations to the RNA-binding data allows calculations of the Hill coefficient (an indicator of the cooperativity of RNA-binding; Marcos et al., 1999). The dissociation constant Journal of General Virology 87

Activities of N-terminal part of TGBp1 helicase domain (Kd) for the protein–RNA interaction, defined as the concentration of the protein at half saturation (mid-point), was also determined on the basis of these data.

RESULTS The N-terminal region of the TGBp1 NTPase/ helicase domain containing three conserved motifs is sufficient for ATPase activity PVX TGBp1 probably has the simplest NTPase/helicase domain of the SF1 superfamily, because the NTPase/helicase domain of the protein comprises its entire sequence. Our previous studies showed that the C-terminal region of PVX TGBp1 is dispensable for protein ATPase activity (Morozov et al., 1999). To analyse further the enzymic activity of the PVX TGBp1, also termed ‘25K protein’, we constructed a deletion mutant lacking the C-terminal region, including motifs III–VI. The resulting truncated protein, referred to as the 25KI–II mutant, comprised the 96 N-terminal amino acid residues of the 25K protein including the conserved motifs I, Ia and II of the NTPase/helicase domain (Fig. 1a). The mutant and the full-length wild-type 25K proteins were expressed in E. coli as fusions with an N-terminal 66 His tag and purified by Ni-NTA chromatography. Under standard conditions of the ATPase assay, the 25KI–II and wild-type 25K proteins demonstrated similar activity (Fig. 2a). On the other hand, the previously described mutants DCIII-25K and DN-25K (Fig. 1a) (Morozov et al., 1999) were enzymically inactive (Fig. 2a), as expected. The 25KI–II mutant has been constructed by engineering a translational terminator in place of the codon for amino acid residue 97 of the wild-type 25K protein, while in the DCIII-25K mutant, a frame-shift mutation was introduced after residue 84 (Fig. 1a; Morozov et al., 1999). Thus, the frame-shift mutation in the DCIII-25K mutant is located exactly at the NTPase/helicase motif II and the 25K-specific sequence in the ATPase-deficient mutant DCIII-25K is 12 amino acid residues shorter than the enzymically active 25KI–II mutant. (Fig. 1a). Therefore, since further C-terminal truncations of the 25KI–II mutant destroy sequences essential for ATPase activity, we suppose that this mutant contains a minimal set of structural elements required for ATP hydrolysis by the 25K protein. These data point to the fact that the N-terminal region of the TGBp1 is sufficient for proper folding and functioning of the ATPase active centre. To extend these conclusions, a deletion analysis of the NTPase/helicase domain of hordei-like TGBp1 was also carried out. In these experiments, we used the C63K protein, the previously described derivative of PSLV 63 kDa TGBp1 (termed the 63K protein), which comprised the protein NTPase/helicase domain. The C63K protein has been demonstrated to have all the enzymic activities of the fulllength 63K protein (Kalinina et al., 2002). Based on the C63K protein, a C-terminally truncated C63KI–II mutant was constructed (Fig. 1b), which comprised the protein region including the NTPase/helicase motifs I, Ia and II http://vir.sgmjournals.org

Fig. 1. Schematic representation of wild-type and mutant PVX TGBp1 (25K protein) (a) and PSLV TGBp1 (63K protein) (b). Filled boxes mark conserved NTPase/helicase motifs. The number of amino acid residues for each mutant is shown. Substitutions of Arg-15/Ala in the full-length protein (mutant 25KR/A) and in the truncated protein (mutant 25KR/AI–II) (a) and substitutions of Lys-315/Ala in the C63K protein (mutant C63KK/A) and in the truncated protein (mutant C63KK/AI–II) (b) are indicated by arrows. Deletion mutants of the 25K and 63K proteins constructed previously (Morozov et al., 1999; Kalinina et al., 2001) are indicated according to their previous designations (DCIII-25K, DN-25K, DN-63K). Hatched boxes indicate nonviral sequences.

(113 amino acid residues extending from residue 291 to 403 of the full-length 63K protein). Therefore, this mutant was structurally similar to the 25KI–II mutant. The C63KI–II mutant exhibited ATPase activity similar to that of the C63K protein, although the DN-63K mutant (Fig. 1b) (Kalinina et al., 2002) was enzymically inactive (Fig. 2b). Thus, a portion of the TGBp1 proteins of both virus types including three NTPase/helicase motifs, I, Ia and II, and a short sequence upstream of motif I retained ATPase activity. NTPases/helicases of SF1 are known to have high intrinsic ATPase activity and, unlike superfamily 2 helicases, they can be stimulated by the presence of RNA by no more than 2fold. In our case, no essential differences were evident in the ATPase activity of the mutant proteins in the presence or absence of RNA, and ATPase activity of the full-length and truncated mutants was stimulated by 1?4- to 1?7-fold (data not shown). 3089

A. D. Leshchiner and others

Table 1. Kinetic parameters for ATP hydrolysis reactions Kinetic parameters of the PVX and PSLV TGBp1 NTPase/helicase domains and their mutants for ATP hydrolysis were measured in the linear phase of the reaction as described in Methods. Protein PVX TGBp1 (potexvirus) 25K 25KI–II 25KR/A 25KR/AI–II PSLV TGBp1 (hordeivirus) C63K C63KI–II C63KK/A C63KK/AI–II

Km (mM)

Vmax (mmol min”1 mg”1)

11±1?5 13±2?0 26±2?0 21±1?5

3?8±0?2 16±0?5 0?6±0?2 11±0?8

14±1?2 16±1?5 30±1?1 16±1?6

4±0?25 14±0?4 0?4±0?1 15±0?5

full-length 25K and C63K proteins and 16?0±0?5 and 14?0±0?4 nmol min21 mg21 for the 25KI–II and C63KI–II mutants, respectively (Table 1). Since an apparent Vmax is calculated per mg protein, and the molecular masses of the 25KI–II and the C63KI–II mutants are about half the molecular masses of the 25K and C63K proteins, the molar activity, which reflects the velocity of ATP hydrolysis, is approximately 2-fold higher for the truncated mutants compared with the full-length helicase domains. Fig. 2. Comparison of ATPase activities of the PVX and PSLV TGBp1 full-length NTPase/helicase domains and their deletion mutants. ATPase activity of the proteins is presented as relative activity (percentage hydrolysis of 32P-labelled substrate). (a) ATPase activity of the wild-type PVX 25K protein (&) and the truncated 25KI–II (X), DCIII-25K (6) and DN-25K (m) mutants. (b) ATPase activity of the PSLV C63K protein (&) and the truncated C63KI–II (X) and DN-63K (m) mutants.

To characterize the ATPase activities of the 25KI–II and C63KI–II mutants in more detail, we measured the kinetic parameters of ATP hydrolysis. The Km for ATP hydrolysis and an apparent Vmax value were determined as described by Rakitina et al. (2005). For both the 25KI–II and C63KI–II mutants, C-terminal truncation of the NTPase/helicase domain resulted in slightly increased Km values, compared with the 25K and C63K proteins (Table 1). The Km of ATP hydrolysis for the 25K protein was 11±1?5 mM. Deletion of the C-terminal half of the 25K protein affected the Km only slightly (13±2?0 versus 11±1?5 mM), and a similar pattern was observed with the helicase domain of the 63K protein. The Km for the C63K protein was 14±1?2 mM, compared with 16±1?5 mM for the truncated C63KI–II mutant. Thus, in both potex-like and hordei-like TGBp1s, the affinity for ATP at the ATPase active site was similar for the full-length NTPase/helicase domain and for its isolated N-terminal region retaining motifs I, Ia and II. Apparent Vmax values were 3?8±0?2 and 4?0±0?25 nmol min21 mg21 for the 3090

RNA-binding activity of the N-terminal part of NTPase/helicase domains The RNA-binding activity of the 25KI–II and C63KI–II mutants was compared with that of the 25K and C63K proteins in a gel-shift assay used previously for analysis of TGBp1 RNA binding (Kalinina et al., 2001). In particular, equal aliquots of non-specific unlabelled RNA (virion RNA of TMV) were incubated with increasing amounts of protein and then subjected to electrophoresis in ethidium bromidecontaining agarose gels and visualized under UV light. In control experiments, protein–RNA complexes formed by the 25K protein were unable to enter the gel, and all of the RNA was incorporated in such complexes at a protein : RNA molar ratio close to 200 : 1 (Fig. 3a), in agreement with our earlier data (Kalinina et al., 2001). Similar patterns of protein–RNA interactions were found for the 25KI–II mutant, except that fully retarded complexes were formed at a higher protein : RNA molar ratio of approximately 400 : 1 (Fig. 3a). Comparison of RNA binding of the C63KI– II and C63K proteins revealed similar shifts in protein : RNA ratios. All RNA appeared to be bound to the protein at molar ratios of about 140 : 1 for the C63K protein and 350 : 1 for the C63KI–II mutant (Fig. 3b). The N-terminally truncated DN-25K and DN-63K mutants were incapable of binding RNA (Kalinina et al., 2001; Fig. 3). Thus, the N-terminal part of the helicase domain of both the 25K and 63K Journal of General Virology 87

Activities of N-terminal part of TGBp1 helicase domain

(a)

(b) 1: 60

1: 80

1: 100

1: 120

1: 130

1: 150

1: 160

1: 180

RNA

1: 30

1: 50

1: 70

1: 80

25K 1: 50 1: 100 1: 150 1: 200 1: 250 1: 300 1: 320 1: 340 1: 360 1: 400 1: 440 RNA

25KI_II 1: 40

1: 80 1: 140 1: 160 1: 200 1: 240 1: 300 1: 360 RNA

DN-25K

1: 100 1: 120 1: 140 1: 160 1: 170

RNA

C63K 1: 80

1: 120

1: 160

1: 210

1: 260

1: 300

1: 340

1: 380

RNA

C63KI_II 1: 80 1: 120 1: 160 1: 210 1: 260 1: 300 1: 340 RNA

DN-63K

Fig. 3. Gel-shift assay for RNA binding by the full-length and truncated mutants of the PVX TGBp1 (a) and PSLV TGBp1 (b) NTPase/helicase domains. Increasing amounts of proteins were incubated with 0?5 mg TMV RNA and subjected to electrophoresis in 1 % non-denaturing agarose gels. Protein : RNA molar ratios are indicated for each lane. Laser RNA, Sample with no protein added.

proteins seems to be primarily responsible for interactions of this domain with RNA. Analysis of quantified gel-shift data revealed some differences in RNA binding between the full-length NTPase/ helicase domains and their truncated mutants. The dissociation constant (Kd) for protein–RNA interactions, defined as the concentration of the protein at half saturation (mid-point) and calculated on the basis of the gel retardation assay, was 5?4±0?5 and 5?7±0?6 mM for the 25K protein and C63K protein compared with 8?0±0?5 and 9?0±0?5 mM for the 25KI–II and C63KI–II mutants, respectively. Higher Kd values observed for the truncation mutants presumably reflect the fact that more molecules of smaller proteins are required to coat a given RNA fully. Thus, the C-terminal truncation had no significant effect on the efficiency of RNA binding by the NTPase/helicase domains of PVX and PSLV TGBp1. Hill coefficients, indicating the degree of cooperativity of protein–RNA interactions, were calculated for both pairs of proteins (Fig. 4). Slightly increased Hill coefficients indicate that the truncated mutants retained the ability to bind RNA with cooperativity similar to the wild-type helicase domains. No significant differences in protein–RNA interactions could be found under high salt concentrations (200 and 400 mM NaCl; data not shown). Gel-retardation experiments were also carried out with a double-stranded RNA of about 300 bp; there were no significant differences in the efficiency of binding between single-stranded and double-stranded RNA (data not shown). http://vir.sgmjournals.org

The role of a conserved positively charged amino acid residue upstream of motif I in TBGp1 NTPases/helicases Positively charged amino acid residues located upstream of motif I in the TGBp1 of the potexvirus Bamboo mosaic virus (BaMV) have been reported to be crucial for ATPase activity, RNA-binding activity in vitro and virus cell-to-cell movement in vivo (Wung et al., 1999; Liou et al., 2000; Lin et al., 2004). One of these residues, which corresponds to Arg-15 in the PVX 25K protein and Lys-315 in the PSLV 63K

Fig. 4. Cooperativity of RNA–protein interactions for TGBp1s and their mutants (Hill coefficient). The values of the Hill coefficient were calculated from gel-shift RNA-binding data by applications of a Hill transformation. 3091

A. D. Leshchiner and others

protein, is conserved among both potex-like and hordeilike TGBp1s (Morozov & Solovyev, 2003). To study the functional role of these residues, Arg-15 in the 25K protein and Lys-315 in the C63K protein were replaced by Ala, as well as in the respective C-terminally truncated 25KI–II and C63KI–II mutants. The ATPase activity and RNA-binding properties of the resulting mutants 25KR/A, 25KR/AI–II, C63KK/A and C63KK/AI–II (Fig. 1) were analysed. It was found that the ATPase activity of the 25KR/A and C63KK/A point mutants, comprising the full-length NTPase/ helicase domains, decreased considerably compared with that of the 25K and C63K proteins (Fig. 5a, b). The Km values for the 25KR/A and C63KK/A mutants were 26±2?0 and 30±1?1 mM, respectively. The apparent Vmax was

0?6±0?2 and 0?4±0?1 nmol min21 mg21 compared with 3?8±0?2 and 4?0±0?25 nmol min21 mg21 for the fulllength 25K and C63K proteins (Table 1). Thus, the velocity of ATP hydrolysis is about 10-fold lower than that of the wild-type proteins. In contrast, the C-terminally truncated mutants 25KR/AI–II and C63KK/AI–II had ATPase activities similar to those of the 25KI–II and C63KI–II mutants (Fig. 5a, b; Table 1). These data suggest that the introduced substitutions are unlikely to have a direct effect on the intrinsic enzymic activity of TGBp1 ATPase per se. Quantitative evaluation of gel-shift RNA-binding assay data showed no significant difference in Kd between these mutants and the wild-type proteins (data not shown). Interestingly, the cooperativity of RNA binding (determined as Hill coefficients) for the full-length (25KR/A and C63KK/A) and truncated point mutants (25KR/AI–II and C63KK/AI–II) decreased notably compared with the proteins without point mutations (Fig. 4).

Role of the N-terminal part of the TGBp1 NTPase/helicase domain in protein selfinteractions The recombinant 25K protein, as well as the C63K protein, typically migrated in SDS-PAGE as a single band (Kalinina et al., 1996, 2001; Fig. 6a). Interestingly, the 25KI–II and 25KR/AI–II mutants reproducibly gave two bands. The lower band had the size expected for the recombinant protein, whereas the size of the upper band corresponded to the protein dimer (Fig. 6a). In contrast, the deletion mutants DCIII-25K and DN-25K both migrated as single bands (Fig. 6a). These observations indicate that the C-terminally truncated mutants 25KI–II and 25KR/AI–II formed stable dimers in the presence of SDS. Importantly, the intrinsic ability of the isolated TGBp1 N-terminal region for stable self-interaction was confirmed by observation of protein dimers formed by the C63KI–II and C63KK/AI–II mutants (Fig. 6a). Although this is unusual, such enormously stable self-dimers of different proteins have been previously reported by others (Bentley et al., 2002; Gentile et al., 2002; Taliansky et al., 2003).

Fig. 5. Influence of point mutations in a conserved positively charged residue upstream of motif I on the ATPase activity of the PVX and PSLV full-length and truncated NTPase/helicase domains. The conditions of the ATPase assay were the same as described in Fig. 2. (a) Comparison of the ATPase activities of the wild-type 25K protein (X), the full-length point mutant 25KR/A (&) and the truncated point mutant 25KR/AI–II (m). (b) Comparison of the ATPase activity of the wild-type C63K (X), the full-length point mutant C63KK/A (&) and the truncated point mutant C63KK/AI–II (m). 3092

Protein self-interactions were also analysed by Western-blot assays. The proteins were incubated with sample buffer without 2-mercaptoethanol at room temperature and, after electrophoresis and transfer to nitrocellulose membranes, they were probed with antibodies produced against the recombinant 25K and 63K proteins. It should be noted that dimers and oligomers could be found not only in the sample with the C-terminally truncated mutants 25KI–II and C63KI–II, but also in smaller quantities in the sample with wild-type 25K and C63K proteins (Fig. 6b). Therefore, the N-terminal region of the TGBp1 NTPase/ helicase domain could account for homologous protein interactions. Journal of General Virology 87

Activities of N-terminal part of TGBp1 helicase domain

(a) K

63

C

K

25

2

DN

K 25

5K

III-

DC

I

A _I I R/

_ II

KI

25

K

25

K/

_ II I

K

63

C

I

A _I I

K

63

C

47 36 26 20

(b) O 118 85

O D

47 36 M

118 85

O

118

47 36

D

85 47

D 47

26

26

20

20

118 85

O

36

D

26

36

M

20

M

M

26 20

25K

25KI_II

C63K

C63KI_II

Fig. 6. Analysis of the full-length and truncated PVX and PSLV TGBp1 by SDS-PAGE and Western blot assay. (a) Recombinant proteins were separated by 20 % SDS-PAGE and stained with Coomassie blue. The positions of the dimers are indicated by arrowheads and the positions of molecular mass markers are shown on the left (in kDa). (b) Recombinant proteins were subjected to electrophoresis in 20 % SDS-PAGE (truncated mutants) and in 5–15 % gradient SDS-PAGE (fulllength proteins) and then analysed by standard Western-blot assays. Positions of monomers (M), dimers (D) and oligomers (O) are indicated. The positions of molecular mass markers are shown (in kDa).

DISCUSSION It has long been recognized that most ubiquitous cell and viral NTPases, including P-loop NTPases, share two common signatures: the N-terminal Walker A motif with a catalytic Lys residue and the C-terminal Walker B motif with active aspartate and/or glutamate residues (Leipe et al., 2002). The consensus Walker A (G/AXXXXGKT/S, where X is any amino acid) and Walker B (ZZZZD or ZZZZDE, where Z is a hydrophobic amino acid) motifs are included in a common structural b-sheet core for numerous proteins with ATPase activity (Koonin, 1993; Leipe et al., 2002; Vetter & Wittinghofer, 1999). Structurally, P-loop NTPases can be subdivided into several lineages. TGBp1 proteins belong to the RecA/F1-ATPase superclass and, more specifically, fall into helicases of SF1 (Koonin, 1993; Koonin & Dolja, 1993; Leipe et al., 2002). Site-directed mutagenesis and X-ray crystallographic studies revealed that, among the RecA/F1 http://vir.sgmjournals.org

superclass, F1-ATPases possess the minimal set of motifs required for NTP hydrolysis. These motifs represent Walker A and B sites as well as catalytically important signatures (Lobau et al., 1997). It is known that the lysine amino group interacts with MgATP/MgADP phosphates, that the serine or threonine hydroxyl group coordinates a Mg2+ ion (Walker A site), that an aspartic acid in position 1 coordinates a Mg2+ ion and that a glutamic acid in position 2 is a putative catalytic residue (Walker B site) (Caruthers & McKay, 2002). In this paper, we used deletion mutants of the PVX 25K and PSLV 63K proteins (Fig. 1) to demonstrate that the N-terminal region of the TGBp1 NTPase/helicase domain, which contained only three conserved NTPase motifs [I (Walker A), Ia and II (Walker B)], was sufficient for the ATPase function (Fig. 2a, b). Despite evident sequence similarities between the SF1 RNA helicases (including TGBp1) and the SF1 DNA helicases (Koonin, 1993; Koonin & Dolja, 1993), these proteins appear to display different structural requirements for catalysis of ATP hydrolysis. The NTPase activity of DNA helicases was shown to depend not only on motifs I, Ia and II, as was found for TGBp1s, but also on other conserved motifs of the NTPase/helicase domain (Caruthers & McKay, 2002). However, mutational analysis of herpes simplex virus UL5 helicase, also belonging to SF1, showed that changes in motifs III, IV, V and VI did not eliminate ATPase activity of the protein or affect DNA binding. Nevertheless, these mutants completely lacked helicase activity (Graves-Woodward et al., 1997). In 1996, Koonin and Rudd, using computer methods for protein sequence analysis, suggested that the N- and C-terminal domains of SF1 helicases have distinct activities: only the Nterminal domain possesses ATPase activity (Koonin & Rudd, 1996). Later, we confirmed this suggestion experimentally by showing that the ATPase activity of the PVX TGBp1 is not inhibited, but is rather stimulated, by deletion of C-terminal motifs V and VI (Morozov et al., 1999). In the present work, we have defined the enzymically active region for the helicase domains of two types of TGBp1 proteins and have shown that the ‘minimal ATPase subdomain’ includes only three of seven conserved motifs. In addition to ATPase activity, the TGBp1 NTPase/helicase subdomain comprising motifs I, Ia and II retained nucleic acid-binding activity (Fig. 3), which is in agreement with previous data (Kadare et al., 1996; Morozov et al., 1999). BaMV TGBp1 has been shown to have three basic amino acid residues in the region upstream of motif I that are important for RNA binding (Wung et al., 1999; Liou et al., 2000). Our data revealed no significant influence of the conserved basic amino acid residue upstream of motif I on the RNA-binding activity of helicase domains of TGBp1s. It is possible that motif Ia and the following minimally conserved TXGX motif take part in nucleic acid–protein interactions, as has been shown for DNA helicases (Caruthers & McKay, 2002). According to our data, the RNA-binding activity of TGBp1 showed pronounced protein–protein cooperativity (Fig. 4), suggesting involvement of TGBp1 in homologous interactions. Indeed, the 3093

A. D. Leshchiner and others

truncated mutants were able to form significant amounts of stable dimers even under SDS-PAGE conditions (Fig. 6a). Immunoblot analysis revealed that the full-length helicase domain of the PVX and PSLV TGBp1 proteins is capable of forming dimers and oligomers under mild sample treatment conditions (Fig. 6b). In addition, we found that the PVX TGBp1 is capable of self-interactions in a yeast two-hybrid system (unpublished data). Fractionation of the recombinant 25K and C63K proteins and their mutants in sucrosedensity gradients revealed that the proteins could not be detected as monomers and appeared in the form of dimers and oligomers (to be published elsewhere). These observations are in agreement with earlier data on dimer formation in vitro by another potex-like TGBp1 (Wung et al., 1999; Liou et al., 2000) and homologous interactions of hordeilike TGBp1 in the yeast two-hybrid system (Cowan et al., 2002). TGBp1 homologous interactions are consistent with the general notion that many helicases form homodimers or oligomers (Gorbalenya & Koonin, 1993). For example, the activity of TMV replication helicase, which is very closely related to TGBp1 in its amino acid sequence, is dependent on homohexamer formation (Goregaoker & Culver, 2003). Positively charged residues upstream of motif I in the potexlike TGBp1 have been shown previously to be essential for ATPase activity (Wung et al., 1999; Liou et al., 2000). In this paper, we evaluated the functional significance of one of the residues that is conserved in both potex-like and hordei-like TGBp1 proteins. Point mutations of this residue caused a substantial decrease in the ATPase activity of the full-length NTPase/helicase domain. However, these mutations failed to affect the ATPase activity of C-terminally truncated mutants comprising motifs I, Ia and II (Fig. 5). Point mutations reduced the protein cooperativity slightly in protein–RNA interactions (Fig. 4), but did not affect the ability of the mutants to form dimers in SDS-PAGE (Fig. 6a). Thus, we hypothesize that the conserved basic amino acid residue upstream of motif I plays an important role in maintenance of a reaction-competent conformation of the ATPase active centre in the full-length helicase domain. This confirms recent findings that point mutations outside the conserved motifs of the NTPase/helicase domain, presumably affecting protein conformation, have drastic effects on PVX TGBp1 activities in vitro (Bayne et al., 2005). To conclude, we have observed that complete deletion of motifs III–VI in TGBp1 does not inhibit ATPase activity or RNA binding. We also show an important role for the conserved lysine/arginine residue upstream of the Walker A motif in the ATPase activity of potexviral and hordeiviral TGBp1 proteins. Based on these data, the ‘minimal ATPase subdomain’ of TGBp1 proteins has been determined. The engineered PVX and PSLV polypeptide derivatives of TGBp1 represent one of the smallest known proteins to display ATP hydrolysis, RNA binding and the capacity for homologous interactions. 3094

ACKNOWLEDGEMENTS This study was supported by the grant no. 04-04-49356 of the Russian Foundation for Basic Research and in part by a grant of the Leading Scientific Schools of the Russian Ministry of Education and Science.

REFERENCES Angell, S. M., Davies, C. & Baulcombe, D. C. (1996). Cell-to-cell

movement of potato virus X is associated with a change in the sizeexclusion limit of plasmodesmata in trichome cells of Nicotiana clevelandii. Virology 216, 197–201. Bayne, E. H., Rakitina, D. V., Morozov, S. Yu. & Baulcombe, D. C. (2005). Cell-to-cell movement of potato potexvirus X is dependent

on suppression of RNA silencing. Plant J 44, 471–482. Bentley, N. M., LaDu, M. J., Rajan, C., Getz, G. S. & Reardon, C. A. (2002). Apolipoprotein E structural requirements for the formation of SDS-stable complexes with b-amyloid-(1–40): the role of salt

bridges. Biochem J 366, 273–279. Caruthers, J. M. & McKay, D. B. (2002). Helicase structure and

mechanism. Curr Opin Struct Biol 12, 123–133. Cowan, G. H., Lioliopoulou, F., Ziegler, A. & Torrance, L. (2002).

Subcellular localisation, protein interactions, and RNA binding of potato mop-top virus triple gene block proteins. Virology 298, 106– 115. Donald, R. G. K., Lawrence, D. M. & Jackson, A. O. (1997). The barley stripe mosaic virus 58-kilodalton bb protein is a multi-

functional RNA binding protein. J Virol 71, 1538–1546. Gentile, F., Amodeo, P., Febbraio, F., Picaro, F., Motta, A., Formisano, S. & Nucci, R. (2002). SDS-resistant active and

thermostable dimers are obtained from the dissociation of homotetrameric b-glycosidase from hyperthermophilic Sulfolobus solfataricus in SDS. Stabilizing role of the A-C intermonomeric interface. J Biol Chem 277, 44050–44060. Gorbalenya, A. E. & Koonin, E. V. (1993). Helicases: amino acid

sequence comparisons and structure-function relationships. Curr Opin Struct Biol 3, 419–429. Gorbalenya, A. E., Koonin, E. V., Donchenko, A. P. & Blinov, V. M. (1989). Two related superfamilies of putative helicases involved in

replication, recombination, repair and expression of DNA and RNA genomes. Nucleic Acids Res 17, 4713–4730. Goregaoker, S. P. & Culver, J. N. (2003). Oligomerization and

activity of helicase domain of the tobacco mosaic virus 126- and 183-kilodalton replicase proteins. J Virol 77, 3549–3556. Graves-Woodward, K. L., Gottlieb, J., Challberg, M. D. & Weller, S. K. (1997). Biochemical analyses of mutations in the HSV-1 helicase-

primase that alter ATP hydrolysis, DNA unwinding, and coupling between hydrolysis and unwinding. J Biol Chem 272, 4623–4630. Haupt, S., Cowan, G. H., Zeigler, A., Roberts, A. G., Oparka, K. J. & Torrance, L. (2005). Two plant-viral movement proteins traffic in

the endocytic recycling pathway. Plant Cell 17, 164–181. Howard, A. R., Heppler, M. I., Ju, H. J., Krishnamurthy, K., Payton, M. E. & Verchot-Lubicz, J. (2004). Potato virus X TGBp1 induces

plasmodesmata gating and moves between cells in several host species whereas CP moves only in N. benthamiana leaves. Virology 328, 185–197. Kadare, G., David, C. & Haenni, A. L. (1996). ATPase, GTPase and

RNA binding activities associated with the 206-kilodalton protein of turnip yellow mosaic virus. J Virol 70, 8169–8174. Kalinina, N. O., Fedorkin, O. N., Samuilova, O. V., Maiss, E., Korpela, T., Morozov, S. & Atabekov, J. G. (1996). Expression and biochemical

Journal of General Virology 87

Activities of N-terminal part of TGBp1 helicase domain analyses of the recombinant potato virus X 25K movement protein. FEBS Lett 397, 75–78. Kalinina, N. O., Rakitina, D. A., Yelina, N. E. & 9 authors (2001).

RNA-binding properties of the 63 kDa protein encoded by the triple gene block of poa semilatent hordeivirus. J Gen Virol 82, 2569–2578. Kalinina, N. O., Rakitina, D. V., Solovyev, A. G., Schiemann, J. & Morozov, S. Y. (2002). RNA helicase activity of the plant virus

movement proteins encoded by the first gene of the triple gene block. Virology 296, 321–329.

Morozov, S. Yu. & Solovyev, A. G. (2003). Triple gene block:

modular design of a multifunctional machine for plant virus movement. J Gen Virol 84, 1351–1366. Morozov, S. Y., Dolja, V. V. & Atabekov, J. G. (1989). Probable

reassortment of genomic elements among elongated RNA-containing plant viruses. J Mol Evol 29, 52–62. Morozov, S. Y., Solovyev, A. G., Kalinina, N. O., Fedorkin, O. N., Samuilova, O. V., Schiemann, J. & Atabekov, J. G. (1999). Evidence

Koonin, E. V. (1993). A superfamily of ATPases with diverse

for two nonoverlapping functional domains in the potato virus X 25K movement protein. Virology 260, 55–63.

functions containing either classical or deviant ATP-binding motif. J Mol Biol 229, 1165–1174.

Rakitina, D. V., Kantidze, O. L., Leshchiner, A. D., Solovyev, A. G., Novikov, V. K., Morozov, S. Yu. & Kalinina, N. O. (2005). Coat

Koonin, E. V. & Dolja, V. V. (1993). Evolution and taxonomy of

proteins of two filamentous viruses display NTPase activity in vitro. FEBS Lett 579, 4955–4960.

positive-strand RNA viruses: implications of comparative analysis of amino acid sequences. Crit Rev Biochem Mol Biol 28, 375–430. Koonin, E. V. & Rudd, K. E. (1996). Two domains of superfamily I

helicases may exist as separate proteins. Protein Sci 5, 178–180. Leipe, D. D., Wolf, Y. I., Koonin, E. V. & Aravind, L. (2002).

Classification and evolution of P-loop GTPases and related ATPases. J Mol Biol 317, 41–72. Lin, M.-K., Chang, B.-Y., Liao, J.-T., Lin, N.-S. & Hsu, Y.-H. (2004).

Arg-16 and Arg-21 in the N-terminal region of the triple-gene-block protein 1 of Bamboo mosaic virus are essential for virus movement. J Gen Virol 85, 251–259. Liou, D.-Y., Hsu, Y.-H., Wung, C.-H., Wang, W.-H., Lin, N.-S. & Chang, B.-Y. (2000). Functional analyses and identification of two arginine residues

essential to the ATP-utilizing activity of the triple gene block protein 1 of bamboo mosaic potexvirus. Virology 277, 336–344.

Rouleau, M., Smith, R. J., Bancroft, J. B. & Mackie, G. A. (1994).

Purification, properties, and subcellular localization of foxtail mosaic potexvirus 26-kDa protein. Virology 204, 254–265. Schepetilnikov, M. V., Manske, U., Solovyev, A. G., Zamyatnin, A. A., Jr, Schiemann, J. & Morozov, S. Yu. (2005). The hydrophobic

segment of potato virus X TGBp3 a major determinant of the protein intracellular trafficking. J Gen Virol 86, 2379–2391. Solovyev, A. G., Savenkov, E. I., Agranovsky, A. A. & Morozov, S. Y. (1996). Comparisons of the genomic cis-elements and coding

regions in RNA beta components of the hordeiviruses barley stripe mosaic virus, lychnis ringspot virus, and poa semilatent virus. Virology 219, 9–18. Taliansky, M., Roberts, I. M., Kalinina, N., Ryabov, E. V., Raj, S. K., Robinson, D. J. & Oparka, K. J. (2003). An umbraviral protein,

ATPase, roles of three catalytic site residues. J Biol Chem 272, 3648– 3656.

involved in long-distance RNA movement, binds viral RNA and forms unique, protective ribonucleoprotein complexes. J Virol 77, 3031–3040.

Lough, T. J., Shash, K., Xoconostle-Ca´zares, B., Hofstra, K. R., Beck, D. L., Balmori, E., Forster, R. L. S. & Lucas, W. J. (1998).

Verchot-Lubicz, J. (2005). A new cell-to-cell transport model for potexviruses. Mol Plant Microbe Interact 18, 283–290.

Lobau, S., Weber, J., Wilke-Mounts, S. & Senior, A. E. (1997). F1-

Molecular dissection of the mechanism by which potexvirus triple gene block proteins mediate cell-to-cell transport of infectious RNA. Mol Plant Microbe Interact 11, 801–814.

Vetter, I. R. & Wittinghofer, A. (1999). Nucleoside triphosphate-

Lough, T. J., Netzler, N. E., Emerson, S. J., Sutherland, P., Carr, F., Beck, D. L., Lucas, W. J. & Forster, R. L. (2000). Cell-to-cell

Voinnet, O., Lederer, C. & Baulcombe, D. C. (2000). A viral

binding proteins: different scaffolds to achieve phosphoryl transfer. Q Rev Biophys 32, 1–56. movement protein prevents spread of the gene silencing signal in Nicotiana benthamiana. Cell 103, 157–167.

movement of potexviruses: evidence for a ribonucleoprotein complex involving the coat protein and first triple gene block protein. Mol Plant Microbe Interact 13, 962–974.

Wung, C. H., Hsu, Y. H., Liou, D. Y., Huang, W. C., Lin, N. S. & Chang, B. Y. (1999). Identification of the RNA-binding sites of the triple

Lucas, W. J. (2006). Plant viral movement proteins: agents for cellto-cell trafficking of viral genomes. Virology 344, 169–184.

gene block protein 1 of bamboo mosaic potexvirus. J Gen Virol 80, 1119–1126.

Malcuit, I., Marano, M. R., Kavanagh, T. A., De Jong, W., Forsyth, A. & Baulcombe, D. C. (1999). The 25-kDa movement protein of PVX

Yang, Y., Ding, B., Baulcombe, D. C. & Verchot, J. (2000). Cell-to-

elicits Nb-mediated hypersensitive cell death in potato. Mol Plant Microbe Interact 12, 536–543.

cell movement of the 25K protein of potato virus X is regulated by three other viral proteins. Mol Plant Microbe Interact 13, 599–605.

Marcos, J. F., Vilar, M., Perez-Paya, E. & Pallas, V. (1999). In vivo

Zamyatnin, A. A., Jr, Solovyev, A. G., Savenkov, E. I., Germudson, A., Sandgren, M., Valkonen, J. P. T. & Morozov, S. Yu. (2004).

detection, RNA-binding properties and characterization of the RNAbinding domain of the p7 putative movement protein from carnation mottle carmovirus (CarMV). Virology 255, 354–365.

Transient coexpression of individual genes encoded by the triple gene block of potato mop-top virus reveals requirements for TGBp1 trafficking. Mol Plant Microbe Interact 17, 921–930.

http://vir.sgmjournals.org

3095