JVI Accepts, published online ahead of print on 10 September 2014 J. Virol. doi:10.1128/JVI.02148-14 Copyright © 2014, American Society for Microbiology. All Rights Reserved.

1

Binding Interactions between the Encephalomyocarditis Virus Leader and Protein 2A

2 3

Ryan V. Petty1, Holly A. Basta2, Valjean R. Bacot-Davis1, Bradley A. Brown3,

4

and Ann C. Palmenberg1#

5 6

1

7

Madison, Madison, WI, USA

8

2

Current Address: Rocky Mountain College, Billings, MT

9

3

Current Address: GenMark Diagnostics, Carlsbad, CA

Institute for Molecular Virology and Department of Biochemistry, University of Wisconsin –

10 11

# Address correspondence to: Ann Palmenberg, [email protected]

12 13

Key words: EMCV, TMEV, Saffold, Leader, 2A, Ran GTPase

14 15

Word count: 4164 (Body, less References and Figure Legends), 188 (Abstract)

16 17

Running Title: EMCV Leader:2A Interactions

18 19 20 21 22 23

~1~

24

ABSTRACT

25

The Leader (L) and 2A proteins of cardioviruses are the primary anti-host agents

26

produced during infection. For encephalomyocarditis virus (EMCV), the prototype of this genus,

27

these proteins interact independently with key cellular partners to bring about inhibition of active

28

nucleocytoplasmic trafficking and cap-dependent translation, respectively. L and 2A also bind

29

each other and require this cooperation to achieve their effects during infection. Recombinant L

30

and 2A interact with 1:1 stoichiometry at a KD of 1.5 μM. The mapped contact domains include

31

the amino-proximal third of 2A (first 50 amino acids) and the central hinge region of L. This

32

contact partially overlaps the L segment that makes subsequent contact with RanGTPase in the

33

nucleus, and Ran can displace 2A from L. The equivalent proteins from TMEV (BeAn) and

34

Saffold virus interact similarly in any subtype combination, with varying affinities. The data

35

suggest a mechanism whereby L takes advantage of the nuclear localization signal in the COOH-

36

region of 2A to enhance its trafficking to the nucleus. Once there, it exchanges partners in favor

37

of Ran. This required cooperation during infection explains many observed co-dependent

38

phenotypes of L and 2A mutations.

39 40

IMPORTANCE

41

Cardiovirus pathogenesis phenotypes vary dramatically, from asymptomatic, to mild GI

42

distress, to persistent demyelination and even encephalitic death. Leader and 2A are the primary

43

viral determinants of pathogenesis, so understanding how these proteins cooperate to induce such

44

a wide variety of outcomes for the host is of great important and interest to the field of virology,

45

especially those who use TMEV as a murine model for multiple sclerosis.

46 2

47 48

INTRODUCTION The Cardiovirus genus of the Picornaviridae family is divided into several species and

49

subtypes. Among the important members are encephalomyocarditis virus (EMCV), Theiler’s

50

murine encephalomyelitis virus (TMEV) and Saffold virus (1, 2). All have single-stranded,

51

positive-sense RNA genomes encoding single, open reading frames. The polyproteins are

52

cleaved co- and post-translationally by an endogenous 3C protease (3). Unique to this genus, the

53

polyprotein begins with an amino-terminal Leader protein (L) and a centrally located 2A protein

54

that are without homolog or analog in other viruses or cells. Together, they are primarily

55

responsible for almost all cardiovirus anti-host activities (4-8).

56 57

For EMCV, the LE protein is 67 amino acids (aa). Saffold (LS, 71 aa) and TMEV proteins

58

(LT, 76 aa) are slightly longer. The solution structure of Mengo LM (an EMCV strain) has been

59

determined in free form and as bound to RanGTPase, a key cellular participant in L-dependent

60

activities (9). The conformation is primarily random coil, except for an amino-proximal CHCC

61

zinc-finger motif (aa 10-22). The structure of the remainder relies on induced-fit contacts

62

dependent upon specific binding partner(s). The mapped functional units, in addition to the zinc

63

finger, include a central “hinge” region (aa 35-44), essential to Ran interactions, and an “acidic

64

domain” (aa 37-52), that confers an overall pI of 3.8 to the protein (10). The LS and LT homologs

65

are similar, with equivalent low pIs, except they also have short, characteristic “Theilo domain”

66

(13 aa) and “Ser/Thr domain” (12 aa) insertions, configured putatively as linked helices, near

67

their respective C-termini (11).

68

~3~

69

For any LX to function in cells, it must be phosphorylated. The required sites include

70

Tyr41 and Thr47 for LE, Ser57 for LT, and Thr58 for LS. Kinases CK2, SYK and AMPK participate

71

in these modifications, but the precise timing of the reactions and stepwise requirements during

72

infection are not yet clearly understood (11, 12). It is clear, however, that during infection or in

73

recombinant form, the introduction of phosphorylation-competent LE, LS or LT into cells induces

74

a rapid inhibition of active nucleocytoplasmic trafficking (NCT) (8, 12). The mechanism

75

requires, in addition to L phosphorylation, specific L interactions with RanGTPase, a key cellular

76

trafficking regulator. When aided by catalytic amounts of nuclear Ran guanine-nucleotide

77

exchange factor (RCC1), LM binds tightly to Ran (KD of 3 nM), diverting its normal activities

78

into anti-host events (13). The consequence is induced hyperphosphorylation of Phe/Gly-

79

containing nuclear pore proteins (Nup) by cellular kinases in the p38 and Erk1/2 pathways (8,

80

14, 15) and subsequent cessation of active NCT. It has been proposed that LM:Ran complexes

81

achieve this by trapping exportin-bound activated kinases within the nuclear pores (NPC) (9).

82

Since Ran-dependent NCT is essentially shut down, the movement of cellular proteins and RNA

83

through the NPC is reduced to that permitted by diffusion alone. Recombinant LM, LE, LT or LS

84

alone are necessary and sufficient for observing these effects when their genes are transfected

85

into cells (11, 14). During infection, however, cardiovirus L proteins are not the exclusive anti-

86

host activators.

87 88

The functions of protein 2A are not as well characterized. EMCV 2A (143 aa) is

89

translated between the P1 (capsid) and P2/3 (replication) regions of the polyprotein. The protein

90

has a distinctive C-terminal 13-16 aa “scission cassette” (Fig 1) ending with an Asn-Pro-Gly-Pro

91

motif (NPG/P). The unit functions in viral or exogenous contexts, through a co-translational 4

92

ribosome-skipping mechanism, separating otherwise co-joined proteins between the Gly and Pro

93

residues (16). The NPG/P event provides primary scission of cardiovirus polyproteins. The N-

94

terminal release, as with the C-terminal release of L, from an L-P1-2A precursor, is subsequently

95

catalyzed by viral 3Cpro. Antibodies specific to EMCV 2A track the dominant cellular

96

localization to nucleoli during infection, although there is also significant cytoplasmic

97

accumulation (4). The protein has a very basic pI of 9.67, which presumably allows it to remain

98

nucleolar through rRNA binding contacts. Mutagenic mapping has identified a ribosome protein-

99

like nuclear localization signal (NLS) and a C-proximal eIF4E binding site, which partially

100

overlaps the scission cassette sequences, and are common to all known cardioviruses (17).

101

Similar mutations, tested during infection, link the activities of 2A (EMCV) to virus-induced

102

shut-down of cap-dependent translation (4, 18). The protein influences 4EBP1 pathways in

103

certain cell types, and moreover, 2A-deficient viruses can be rescued by chemical inhibitors of

104

mTOR and PI3K, elements required for cap-dependent but not virus-dependent translation (19).

105

During infection, a portion of 2A is found in association with 40S, but not 60S or 80S ribosomal

106

subunits, though no determined mechanism yet links these observations (18).

107 108

Cardiovirus L and 2A interactions with various cellular partners have been the subject of

109

much study and speculation (13, 14, 17, 18). As part of this process, we employed yeast two-

110

hybrid systems to fish out unknown, potential reaction candidates (unpublished). Given their

111

reciprocal pIs, perhaps it should not have been a surprise that both came back as preferred

112

partners of each other. The specificity and required elements for these reactions have now been

113

documented by mutagenesis and biochemistry. Within the virus lifecycle, including the Theilo

114

and Saffold viruses, the mutual L:2A pathways may explain why these proteins’ anti-host

~5~

115

activities should probably not be considered independently. Phenotypes attributed to one protein

116

are, in some steps, co-dependent upon the other.

117 118

MATERIALS AND METHODS

119

Recombinant Constructions. The N-terminal His-tagged GB1 gene for parental plasmid pT-

120

hGB1 originated from a pET30-GBFusion1 vector (a kind gift from John Markley), as excised

121

by PCR using appropriate primers. After digestion with Nco I and Hind III, the amplicon was gel

122

purified, then ligated into pTri-Ex 1.1 (Novagen) using the same restriction sequences. The

123

EMCV 2A gene from pEC9 (20) was amplified in parallel, and then digested with Hind III and

124

Xho I. Plasmid pT-hGB1-2A substituted this fragment into the corresponding sites of parental

125

pT-hGB1. Bacterial expression produces an in-frame His-tagged GB1-2A fusion protein (hGB1-

126

2A). Derivative plasmids, using different primer sets were equivalent, but included only those

127

EMCV 2A sequences encoding amino acids 1-50, 51-100, or 101-143. Expression plasmids for

128

Saffold (SafV-2) and TMEV (BeAn) 2A were of similar configuration and founded on

129

amplicons generated from infectious cDNAs (generous gifts from Dr. Howard Lipton). Leader-

130

GST fusion plasmids for EMCV (LE-GST), Saf-2, (LS-GST) and BeAn (LT-GST) have been

131

described (12), as have GST-LE proteins with substitution mutations, GST-LK35Q, GST-LD37A,

132

and GST-LW40A (21). The sequences of all materials were verified by restriction mapping and

133

Sanger sequencing.

134 135

Protein Purification. For hGB1-2A synthesis, plasmids were transformed into Rosetta

136

BL21(DE3) pLac I cells (Novagen). Single colonies were picked then grown overnight in 2XYT

137

(1% glucose, 34µg/mL chloramphenicol, 50µg/mL ampicillin) at 30ºC. The stocks primed larger 6

138

cultures, which at an OD600 of 0.6, were treated with isopropyl ß-D-1-thiogalactopyranoside (1

139

mM, IPTG). Growth continued (30oC) until harvest at an OD600 of 2.4-3.2. The cells were

140

collected (6,000 x g, 15 min, 4ºC) and frozen at -80ºC. Expressed proteins were extracted after

141

resuspending the pellets in His-2A buffer (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole

142

and 25% v/v glycerol) containing phenyl-methane-sulfonylfluoride (1 mM, PMSF). After

143

incubation with lysozyme (1mg/ml, 30 min, 4oC) the DNA was sheared by sonication. The

144

soluble fraction (20,000 x g, 45 min, 4ºC) was filtered (0.2µM filter, GE Healthcare) then loaded

145

onto a HisTrap HP column (GE Healthcare). Bound proteins were eluted with an imidazole step

146

gradient (20, 60, 120, 250, 500 mM). Relevant fractions were pooled and concentrated, then

147

applied to Sephacryl S-100 columns (GE Healthcare). Separation was by size exclusion (50 mM

148

NaH2PO4, 300 mM NaCl, 25% v/v glycerol, pH 7.4). The proteins were collected, dialyzed

149

(same buffer), concentrated, and then stored at -80ºC. The expression and purification of C-

150

terminal GST-tagged Leader proteins, LE-GST, LT-GST and LS-GST have been described (11),

151

as have protocols for human RanGTPase (N-terminal His-tagged), and (N-terminal GST-tagged)

152

human guanine nucleotide exchange factor, RCC1 (13).

153 154

GST-L:2A interactions within infected lysates. HeLa cells were infected with vEC9 at an MOI of

155

30 in 1x P5. After six hours, the cells were washed three times with PBS followed by incubation

156

with Passive Lysis Buffer (Promega). The resulting whole cell lysates were pre-cleared with

157

glutathione-sepharose beads (GE Life Sciences) overnight at 4oC. They were then incubated with

158

pre-bound GST or GST-L (10 μg) glutathione-sepharose beads for two hours at 37 oC. Beads

159

were washed (5x in PLB), and then boiled, before the proteins were fractionated by SDS-PAGE

~7~

160

and detected by western analyses using murine αGST (Novagen, Product #71097) or α2Α (22)

161

with an anti-murine secondary antibody (Sigma-Aldrich, Product A2554).

162 163

Recombinant Protein Interactions. Protein interaction assays took advantage of the respective

164

GST and hGB1 tags on the LX and 2A recombinant panels. When GST proteins were the baits,

165

they were prebound to glutathione-sepharose 4B beads (10 μl reactions with 50 nM protein, 50

166

mM HEPES, 125 mM NaCl, 0.5% NP40, pH 7.4, 4oC, overnight). The beads were collected (500

167

x g) washed with the same buffer (2x), then incubated (1 hr, 25o C) with increasing amounts of

168

prey protein (e.g. hGB1-2A, 5-100 nmol/sample). For competition experiments between 2A and

169

Ran, the bait protein (GST-LE or mutated variants) was prebound to beads as above, before the

170

incubation (2 hrs) with various prey combinations (at 50 nM). All reactions with Ran also

171

included catalytic amounts of RCC1 (at 1 nM). Values for competition experiments were

172

normalized to the amount of hGB1-2A pulled down by wild-type GST-LE. Reciprocal

173

experiments used hGB1 protein baits bound to Ni+2 charged chelating-sepharose beads in buffer

174

(50 nM protein, in 50 mM HEPES, 400 mM NaCl, 50 mM imidazole, pH 7.4) for the capture of

175

GST-LE preys (5-100 nmol/sample). Binding affinity reactions were similar except for the

176

variable salt concentrations (125-500 mM NaCl). Interspecies Lx-GST (on beads) and hGB1-2A

177

reactions were performed as above. For all reactions, after extensive washing (3x) to reduce

178

background signals, the bead-bound proteins were released by boiling in SDS buffer,

179

fractionated by SDS-PAGE, detected by Coomassie staining, quantitated, and compared to input

180

levels of GST-L or hGB1-2A (ImageQuant software). Alternatively, after transfer to PVDF

181

membranes, the proteins were detected by Western analyses. The antibodies included: murine

182

αGST, goat αRan (Santa Cruz Biotechnologies, Product sc-1156); anti-murine secondary and 8

183

anti-goat secondary (Sigma, Product A5420). The GB1 tag is a derivative of the IgG binding B1

184

domain of the streptococcal protein G (23). Assays to detect this protein (αGB1) need only the

185

murine secondary antibody.

186 187

GST-LE Phosphorylation. Bait (GST-LE) and prey (hGB1-2A) complexes bound to glutathione

188

sepharose 4B beads were established and collected as above, except during the protein capture (1

189

hr, 20oC) the prey concentration varied (2.5, 10 or 40 nM). Once the beads were collected, they

190

were resuspended into (manufacturers’) buffers supplemented with 5.0 μCi [γ-32P] ATP

191

(3,000Ci/mmol, 10mCi/ml), 10 units of CK2 (New England Biolabs), 10.3 units of SYK

192

(SignalChem), or 10 units of both CK2 and SYK. After reaction (37°C, 60 min) the beads were

193

washed (3x) with PBS buffer (plus 500mM NaCl, 0.02% Triton X-100), then boiled in SDS,

194

before protein fractionation by SDS-PAGE. Detection was by silver stain, or phosphorscreen, as

195

visualized with a Typhoon scanner (GE Healthcare).

196 197

Surface Plasmon Resonance. Equilibrium binding studies used a BIAcore 2000 instrument

198

(BIAcore AB, Uppsala, Sweden) loaded with CM5 research grade sensor chips (GE Healthcare).

199

αGST (above) was covalently attached to the chips with amine-coupling chemistry. GST-LE (5

200

µg/ml, 120 nM) diluted in SPR buffer (10 mM bis-tris propane, 100 mM NaCl, 0.005% Tween-

201

20, pH 7.4) was flowed over individual chip cells at a rate of 10 μl/min (75 μg total, 25oC). The

202

buffer was then changed to include hGB1-2A (or iterations) in varying concentrations (10, 20, 50

203

µg/ml, 20 μl/min). The total injection time was 450/600 seconds (120/150 μl total) with a

204

dissociation time of 120 seconds. Chip surfaces were regenerated using 20 mM piperazine (pH

205

9.0) with 2 M KCl. Automatic, parallel reference subtractions were performed with an antibody~9~

206

only lane to account for non-specific and bulk interactions. BIA evaluation software, (version

207

4.1) calculated the normalized binding constants specific to LE and 2A. Association and

208

dissociation rates were determined independently from best-fit curves, using Langmuir

209

calculations at steady-state levels. The slope and y-intercept values, plotted in Excel, recording

210

the concentration of analyte (hGB1-2A) against the kobs, were used to determine the final KD.

211 212

RESULTS

213

LE:2A Interactions. The small size and high charge of cardiovirus LX proteins makes them

214

difficult to detect in experiments involving Western assays unless they are fused to tags like GST

215

(220 aa). Such tags, whether C-linked or N-linked, do not affect the structure or biological

216

activity of LE constructs (14). Likewise, the cardiovirus 2A proteins are relatively insoluble,

217

unless they too are coupled to tags like hGB1 (56 aa) or maltose-binding protein (396 aa, MBP)

218

(unpublished). The combined tags make protein purification easier, and binding studies can take

219

advantage of high specificity commercial reagents. When GST-LE was reacted with infected

220

HeLa cell lysates, the extracted binding partners included 2A (Fig 2) as had been suggested by

221

previous yeast two-hybrid assays (unpublished). The converse experiment with hGB1-2A baits

222

was uninformative because native LE cannot be detected with Westerns. Instead, recombinant

223

GST-LE and hGB1-2A were tested together in reciprocal pull down assays, dependent upon their

224

respective tags, and shown to interact with each other (Fig 3A). In multiple experiments, the

225

“bait” protein captured “prey” in approximate proportion to its solution concentration, reaching

226

saturation at about a 1:2 molar ratio (100 nmol/reaction of prey), regardless of the bead-bound

227

protein. The interactions were not due to either protein’s tag, as these alone were unable to

228

capture cognates. Formation of such complexes withstood the presence of 250-500 mM salt (Fig 10

229

3B), indicating a reasonably specific affinity between the LE and 2A proteins with a strength that

230

could not be due to simple charge:charge interactions (i.e. pI 3.8 verses pI 9.7).

231 232

As a better assessment of this complex, the binding constant was determined by surface

233

plasmon resonance. SPR is essentially a pull down assay using a mass-sensitive chip. In this

234

case, three concentrations of hGB1-2A analyte were reacted over an antibody-fixed GST-LE

235

surface, and the increased mass over 450 or 600 seconds of exposure was recorded in a

236

sensorgram. A plot series is shown in Fig 4A. From these curves, including the decay phase after

237

the analyte is flushed; normalized values for kobs can be calculated for each concentration (Fig

238

4B). These in turn extrapolate to absolute on/off rates [kon=1.4(±0.1)x10-3 M-1s-1, koff

239

=2.1(±0.1)x10-6 s-1] and a KD for the LE:2A reaction, determined here as 1.5±0.1 μM. The

240

shape(s) of the sensogram curves are consistent with 1:1 stoichiometry. Higher order cooperative

241

interactions would have different plots (Fig 4A), and non-linear extrapolated slopes (Fig 4B).

242 243

Homolog Interactions. Among LE, LS and LT sequences (67-71 amino acids, aa) for which there

244

are cDNAs, there is about 29% shared aa identity, and 42% aa similarity (11). The equivalent 2A

245

proteins vary in length from 133-143 aa, and share 14% identity (Fig 1) with 39% similarity. If

246

properly controlled, capture experiments can provide a measure of relative affinity for panels of

247

similar proteins. In this case, a C-terminal tagged LX-GST panel was chosen as baits because

248

Saffold and Theilo Leader proteins become biologically inactive if the tag is attached N-terminal

249

(11). These and cognate hGB1-2A proteins were isolated, quantitated, then reacted in matched

250

samples (Fig 5). In repeated experiments (all data not shown) there was cross-reactivity with

251

every combination, but surprisingly, the 2A from EMCV was always the most reactive with each

~ 11 ~

252

of the Leaders regardless of species. LS-GST and LT-GST bound nearly twice as much of this

253

protein, as they did their homologous counterparts. The EMCV 2A was clearly the preferred

254

binding partner.

255 256

Required 2A Elements. No 2A structure is available for any cardiovirus. For all these sequences

257

though, the C-terminal third of the protein, maintains characteristics of an extended alpha helix

258

(17). The first and second portions are not responsive to structure predictions. These regions are

259

more variable in sequence between species. Most of the basic residues contributing to the pI map

260

in these upstream regions, imparting the clearest pI differential to the first two thirds of the

261

protein (Fig 1). Among 2As, this portion of the EMCV protein is the most basic. The EMCV 2A

262

binding segments making contact with LE were approximated by dividing the gene into

263

fragments encoding residues 1-50, 51-100 and 101-143. The peptides were then expressed with

264

hGB1 tags for solubility. In turn these served as prey in LE-GST capture experiments (Fig 3C).

265

GST-LE was able to pull down a significant portion of fragment 1-50 (73% compared to input),

266

but neither of the other fragments was reactive in this context. Therefore, fragment 1-50 probably

267

contains the dominant 2A determinants for LE interactions, at least as measured in the absence of

268

an intact 2A conformation. Follow up SPR experiments with the hGB1-2A1-50 fragment and

269

GST-LE were inconclusive because the much smaller mass change of the prey did not give

270

reproducible signals, especially at low concentrations.

271 272

LE Partner Competition. In the presence of catalytic amounts (1 nmol/reaction) of RCC1, LE

273

binds RanGTPase at 1:1 stoichiometry with a KD of 3 nM (13). The KD for LE:2A, as determined

274

above by SPR, is much higher (1.5 μM), so in theory, Ran should be able to outcompete 2A if 12

275

the preferred LE bait sites overlap. Ran interacts with the central hinge region of the LE protein

276

(9) within which mutations at K35, D37, and W40 mark the most significant sites (21). 2A (50

277

nmol), Ran (50 nmol) or a mixture of both preys (50 nmol each) were added GST-LE bait,

278

allowed to reach equilibrium (2 hrs), and then assessed for relative 2A binding. All reactions

279

with Ran included catalytic amounts (1 nM) of GST-RCC1. A representative gel series is shown

280

in Fig 6A. The same experiment was repeated (n=4) and averaged to produce the values

281

indicated in Fig 6B. They show that hGB1-2A binding was reduced by 16-26% when the bait

282

GST-LE had any of the key mutations in the hinge region, even in the absence of Ran (white

283

bars). Ran, when present, bound simultaneously to the same GST-LE beads with essentially 1:1

284

stoichiometry (13). The combined preys reduced hGB1-2A binding to the wild-type GST-LE by

285

22%, and clearly, that binding was further weakened with mutant LE sequences, because Ran

286

then displaced even more hGB1-2A (29-60% reduced binding, grey bars). Still, that Ran did not

287

entirely displace hGB1-2A from the bound GST-LE suggests these proteins have partially

288

overlapping, but not mutually exclusive preferences for LE sites. While these mutations were

289

previously shown to reduce L:Ran binding (21) in the absence or RCC1, as shown here, addition

290

of this Ran-activating factor can help overcome some of the mutational inhibition.

291 292

2A Impedes LE Phosphorylation. During infection, LE is sequentially phosphorylated at T47

293

and Y41 by CK2 and SYK enzymes, respectively (12). The addition process, even with

294

recombinant proteins, is obligatory, because mutations which block the CK2 reaction (e.g. T47A)

295

also prevent the SYK reaction at Y41 unless the mutation is a phosphomimetic (e.g. T47E) (12).

296

When added to GST-LE, neither hGB1 nor hGB1-2A prevented the incorporation of 32P, as long

297

as the bait protein had a wild-type CK2 site at T47 (Fig 7A). But when treated with CK2 and then

~ 13 ~

298

SYK, the Y41 site became masked in the presence of hGB1-2A (Fig 7B). The control hGB1 alone

299

did not allow this masking, either on the wild-type GST-LE, or with the phosphomimetic bait,

300

T47E. Therefore, Y41, which lies to the C-terminal side of the LE hinge domain, is among the

301

likely contact sites for 2A binding. This site and T47 are found solvent-exposed when LE binds

302

Ran (9).

303 304 305

DISCUSSION At the earliest stages of a cardiovirus infection, viral proteins are in low abundance. And

306

yet the virus must take swift action to combat innate host antiviral defenses. The LE protein of

307

EMCV achieves this by leveraging a cell kinase-based phosphorylation cascade directed against

308

Phe/Gly-containing nuclear pore proteins (15). The effect is a rapid shutdown of active transport

309

of macromolecules across the NPC (8). Addition of LE to permeabilized cells, or transfection of

310

LE-encoding cDNA into intact cells, can readily demonstrate this effect (8, 15). But in both

311

cases, the viral protein concentrations are effectively much higher than the scant few molecules

312

initially translated from an infecting genome. We previously hypothesized that the viral protein

313

2A, which encodes an active nuclear localization signal, may help shuttle LE to the nuclear rim,

314

thereby placing it directly into contact with RanGTPase, the required LE activation partner (13).

315

This would, however, require a physical interaction between LE and 2A, either directly or

316

indirectly.

317 318

These proteins, encoded at opposite ends of the L-P1-2A precursor, are released by

319

tandem cleavages with 3Cpro almost as soon as the protease is available (24-26). Their respective

320

pIs as the most basic (2A) and acidic (LX) units in the polyprotein should make obvious the 14

321

potential for interaction. Indeed, we demonstrated here that LX and 2A from three different

322

cardioviruses can bind directly in vitro and in any combination, from any virus (Fig 5). The

323

binding is stoichiometric. For EMCV 2A, it can be almost entirely recapitulated with a shorter

324

fragment containing only the first 50 amino acids (Fig 3C). Theilo and Saffold LX cognates

325

reacted with all homologous 2A proteins, but preferred the sequence from EMCV, presumably

326

because that particular 1-50 segment is almost 2 logs more basic than their normal partners (Fig

327

1). When measured by SPR, the EMCV proteins had a KD of 1.5 μM that was partially

328

responsive to salt but the majority of complexes were still able to form at concentrations up to

329

500 mM. Therefore, these proteins must have a degree of specificity in addition to simple

330

charge:charge interactions.

331 332

The preferred partner for LE, Ran, binds with a much lower KD (3 nM). Competitions

333

between 2A and Ran for bead-bound LE, however, suggest that both proteins can be

334

accommodated simultaneously, implying only partially overlapping binding sites. Mutated LE

335

sequences with weaker binding affinities for 2A (e.g. W40A) were more readily displaced by Ran

336

(Fig 6A). Interestingly, LE:2A interactions also clearly masked LE residue Y41, one of two crucial

337

phosphorylation sites for the activity of LE. Phosphorylation is not required for LE interactions

338

with Ran, but without these modifications, the subsequent complex cannot proceed to ternary or

339

quaternary reactions required to trigger the Nup phosphorylation cascade (9). Therefore,

340

logically, 2A cannot remain perpetually bound to LE, during the normal course of events during

341

infection. For complete LE phosphorylation after it is bound to Ran, the 2A must be released.

342

Since LE:Ran interactions are facilitated by the conformational morphing of Ran, as catalyzed by

343

RCC1 tethered to chromatin just inside the nuclear rim, the results are consistent with a 2A-

~ 15 ~

344

dependent trafficking pathway of LE to nuclear RCC1 sites, where it is displaced by Ran.

345

Subsequently, LE, bound to Ran can be dual phosphorylated, and primed for Nup inhibition

346

activities. The freed 2A then presumably proceeds to nucleoli and initiates its independent

347

cellular translation inhibition activities.

348 349

If this scenario is true, it can explain some previously observed experimental anomalies

350

in 2A and LE mutational studies. For example, deletions in LE (27), 2A (17) or chimeric viruses

351

exchanging EMCV and Theilo LX or their 2A (25) typically have incomplete or improperly

352

processed L-P1-2A regions. Presumably, the L-2A interaction, even in this precursor stage, could

353

facilitate proper P1 folding, creating the requisite conformational substrates for sequential

354

reactions with 3Cpro. Without this interaction, disrupted by the deletion or mutation of either

355

protein, 3Cpro would not efficiently process protomers into functional assembly intermediates.

356

The LX:2A binding reactions we tested with EMCV, Saffold and Theilo proteins showed that

357

some chimeric combinations had poorer affinities. For example (Fig 5), LE and Theilo 2A bind to

358

only 10% saturation compared to LE and EMCV 2A. When tested in a virus context it has been

359

reported these same homologous swaps have, as expected, concordant processing and replication

360

defects (25, 28). The results also imply that studies aimed at mutagenesis of LT domains (29, 30),

361

with regard to its assigning nuclear pore activities or effects on cytokine trafficking, could easily

362

cause unintended disruption of 2A-dependent trafficking, or reduced LT:2A affinities, that would

363

manifest as phenotypes with impeded LT localization to the nuclear pore and subsequent shutoff

364

of NCT. All told, our findings show that the anti-host activities of cardiovirus LX and 2A

365

proteins should not be considered independent of one another. Some phenomena previously

366

ascribed solely to 2A or to LX, may result from their affinity for each other. 16

367 368

ACKNOWLEDGEMENTS

369

This work was supported by NIH grant AI017331 to ACP, NIH Training Grants

370

GM07215 to RVP, and CVA347300 to VRB-D. Additional support was from a UW Science and

371

Medicine Graduate Research Scholars Fellowship (VRB-D) and UW Dept. of Biochemistry

372

Wharton Fellowship (RVP). We thank Dr. Howard Lipton for the generous gift of SafV-2 and

373

TMEV (BeAn) cDNAs, Lukasz Ochyl for cloning help, as well as Dr. Darrell McCaslin for his

374

advice and expertise. Instrument data were obtained at the UW Biophysics Instrumentation

375

Facility, which was established with support from the University of Wisconsin and grants BIR-

376

9512577 (NSF) and S10 RR13790 (NIH).

377 378 379

REFERENCES 1.

Pevear, D. C., M. A. Calenoff, E. J. Rozhon, and H. L. Lipton. 1987. Analysis of the

380

complete nucleotide sequence of the picornavirus Theiler's murine encephalomyelitis

381

virus indicates that it is closely related to cardioviruses. J.Virol. 61:1507-1516.

382

2.

Zoll, J., S. E. Hulshof, K. Lanke, F. V. Lunel, W. J. G. Melchers, E. S.-v. d. Ven, M.

383

Roivainen, J. M. D. Galama, and F. J. M. v. Kuppeveld. 2009. Saffold virus, a human

384

Theiler's-like cardiovirus, is ubiquitous and causes infection early in life. PLoS Pathogens

385

5:1-10.

386

3.

Martinez-Salas, E., and M. D. Ryan. 2010. Translation and protein processing, p. 141-

387

161. In E. Ehrenfeld, E. Domingo, and R. P. Roos (ed.), The Picornaviruses. ASM Press,

388

Washington D.C.

~ 17 ~

389

4.

Aminev, A. G., S. P. Amineva, and A. C. Palmenberg. 2003. Encephalomyocarditis

390

viral protein 2A localizes to nucleoli and inhibits cap-dependent mRNA translation. Virus

391

Research 95:45-57.

392

5.

Zoll, J., J. M. D. Galama, F. J. M. van Kuppeveld, and W. J. G. Melchers. 1996.

393

Mengovirus leader is involved in the inhibition of host cell protein synthesis. J.Virol.

394

70:4948-4958.

395

6.

396 397

van Pesch, V., O. van Eyll, and T. Michiels. 2001. The leader protein of Theiler's virus inhibits immediate-early alpha/beta interferon production J.Virol. 75:7811-7817.

7.

Zoll, J., W. J. Melchers, J. M. Galama, and F. J. van Kuppeveld. 2002. The

398

mengovirus leader protein suppresses alpha/beta interferon production by inhibition of

399

the iron/ferritin-mediated activation of NF-kappa B. J.Virol 76:9664-9672.

400

8.

Porter, F. W., and A. C. Palmenberg. 2009. Leader-induced phosphorylation of

401

nucleoporins correlates with nuclear trafficking inhibition of cardioviruses. J.Virol

402

83:1941-1951.

403

9.

Bacot-Davis, V. R., J. J. Ciomperlik, C. C. Cornilescu, H. A. Basta, and A. C.

404

Palmenberg. 2014. Solution structures of Mengovirus leader protein, its phosphorylated

405

derivatives, and in complex with RanGTPase. Proc. Nat. Acad. Sci. USA in press

406

10.

Cornilescu, C. C., F. W. Porter, Q. Zhao, A. C. Palmenberg, and J. L. Markley.

407

2008. NMR structure of the Mengovirus leader protein zinc-finger domain. FEBS Letters

408

582:896-900.

409

11.

Basta, H. A., and A. C. Palmenberg. 2014. AMP-activated protein kinase

410

phosphorylates EMCV, TMEV, and SafV Leader proteins at different sites. Virology

411

262:236-240. 18

412

12.

Basta, H. A., V. R. Bacot-Davis, J. J. Ciomperlik, and A. C. Palmenberg. 2014.

413

Encephalomyocarditis virus Leader is phosphorylated by CK2 and syk as a requirement

414

for subsequent phosphorylation of cellular nucleoporins. J.Virol. 88:2219-2226.

415

13.

Petty, R. V., and A. C. Palmenberg. 2013. Guanine-nucleotide exchange factor RCC1

416

facilitates a tight binding between EMCV Leader and cellular Ran GTPase. J. Virol.

417

87:6517-6520.

418

14.

Porter, F. W., Y. A. Bochkov, A. J. Albee, C. Wiese, and A. C. Palmenberg. 2006. A

419

picornavirus protein interacts with Ran-GTPase and disrupts nucleocytoplasmic

420

transport. Proc.Natl.Acad.Sci.U.S.A 103:12417-12422

421

15.

Porter, F. W., B. Brown, and A. Palmenberg. 2010. Nucleoporin phosphorylation

422

triggered by the encephalomyocarditis virus leader protein is mediated by mitogen-

423

activated protein kinases. J.Virol. 84:12538-12548.

424

16.

Donnelly, M. L., G. Luke, A. Mehrotra, X. Li, L. E. Hughes, D. Gani, and M. D.

425

Ryan. 2001. Analysis of the aphthovirus 2A/2B polyprotein 'cleavage' mechanism

426

indicates not a proteolytic reaction, but a novel translational effect: a putative ribosomal

427

'skip'. J Gen Virol 82:1013-1025.

428

17.

Groppo, R., B. A. Brown, and A. C. Palmenberg. 2011. Mutational analysis of EMCV

429

2A protein identifies a nuclear localization signal and an eIF4E binding site. Virology

430

410:257-267.

431 432

18.

Groppo, R., and A. Palmenberg. 2007. Cardiovirus 2A protein associates with 40S but not 80S ribosome subunits during infection. J. Virol. 81:13067-13074.

~ 19 ~

433

19.

Svitkin, Y. V., H. Hahn, A. C. Gingras, A. C. Palmenberg, and N. Sonenberg. 1998.

434

Rapamycin and wortmannin enhance replication of a defective encephalomyocarditis

435

virus. J.Virol. 72:5811-5819.

436

20.

Martin, L. R., Z. C. Neal, M. S. McBride, and A. C. Palmenberg. 2000. Mengovirus

437

and encephalomyocarditis virus poly(C) tract lengths can affect virus growth in murine

438

cell culture. J. Virol. 74:3074-3081.

439

21.

Bacot-Davis, V. R., and A. C. Palmenberg. 2013. Encephalomyocarditis virus Leader

440

protein hinge domain is responsible for interactions with Ran GTPase. Virology 443:177-

441

185.

442

22.

Aminev, A. G., S. P. Amineva, and A. C. Palmenberg. 2003. Encephalomyocarditis

443

virus (EMCV) proteins 2A and 3BCD localize to nuclei and inhibit cellular mRNA

444

transcription but not rRNA transcription. Virus Research 95:59-73.

445

23.

446 447

Cheng, Y., and D. J. Patel. 2004. An efficient system for small protein expression and refolding. Biochem. Biophys. Res. Commun. 317:401-405.

24.

Hall, D. J., and A. C. Palmenberg. 1996. Mengo virus 3C proteinase: Recombinant

448

expression, intergenus substrate cleavage and localization in vivo. Virus Genes 13:99-

449

110.

450

25.

Zoll, J., F. J. M. van Kuppeveld, J. M. D. Galama, and W. J. G. Melchers. 1998.

451

Genetic analysis of mengovirus protein 2A: its function in polyprotein processing and

452

virus reproduction. J.Gen.Virol. 79:17-25.

453 454

26.

Butterworth, B. E., and R. R. Rueckert. 1972. Kinetics of synthesis and cleavage of encephalomyocarditis virus-specific polypeptides. Virology 50:535-549.

20

455

27.

Dvorak, C. M. T., D. J. Hall, M. Hill, M. Riddle, A. Pranter, J. Dillman, M. Deibel,

456

and A. C. Palmenberg. 2001. Leader protein of encephalomyocarditis virus binds zinc,

457

is phosphorylated during viral infection and affects the efficiency of genome translation.

458

Virology 290:261-271.

459

28.

Paul, S., and T. Michiels. 2006. Cardiovirus leader proteins are functionally

460

interchangeable and have evolved to adapt to virus replication fitness. J. Gen. Virol

461

87:1237-1246.

462

29.

Ricour, C., S. Delhaye, S. V. Hato, T. D. Olenyik, B. Michel, F. J. van Kuppeveld, K.

463

E. Gustin, and T. Michiels. 2009. Inhibition of mRNA export and dimerization of

464

interferon regulatory factor 3 by Theiler's virus leader protein. J Gen Virol 90:177-186.

465

30.

Ricour, C., F. Borghese, F. Sorgeloos, S. V. Hato, F. J. van Kuppeveld, and T.

466

Michiels. 2009. Random mutagenesis defines a domain of Theiler's virus leader protein

467

that is essential for antagonism of nucleocytoplasmic trafficking and cytokine gene

468

expression. J.Virol. 83:11223-11232.

469 470 471 472

FIGURE LEGENDS

473

Figure 1: 2A Protein Sequences EMCV-R (Genbank ABC25550), TMEV BeAn (Swiss-Prot

474

P08544), and Saffold-2 (Genbank AFP86294) were aligned with (Lasergene 9) MegAlign

475

software using the Jotun Hein method. The consensus (Con) required 2 or more identities. Tested

476

protein fragments (LE), their pI values, and functional motifs are indicated.

477

~ 21 ~

478

Figure 2: EMCV-infected HeLa lysates (100 μL) were incubated with either GST or GST-LE

479

(10 μg/reaction). Beads were washed, boiled, and analyzed by Western analysis (αGST and

480

α2Α).

481 482

Figure 3; Pull-down Assays. A) The indicated recombinant “baits” and “preys” were used in

483

reciprocal pull-down assays. The baits were held at 50 nM/reaction, while the prey

484

concentrations varied (5-100 nM/reaction) as described in Methods. Band quantitation is relative

485

to the highest value input (50 nM). B) Similar to A, the association reactions and wash reactions

486

used the indicated salt concentrations. Band detection was by Western analysis (αGST). hGB1(-

487

2A) is recognized by the secondary anti-murine antibody. C) Similar to A, EMCV 2A fragments

488

were reacted with bead-bound LE-GST. Captured prey was detected with Commassie-staining.

489

The right hand marker panel shows the sizes of input 2A fragments, were they to be detected.

490 491

Figure 4: SPR of LE:2A Binding. A) SPR sensorgram curves for hGB1-2A flowed over GST-

492

LE-bound αGST surfaces on CM5 chips at the indicated concentrations. Association phase was

493

either 450 or 600 seconds with a dissociation phase of 120 seconds. Reference subtraction used a

494

lane containing only αGST (not shown). B) BIAevaluation software calculated association and

495

dissociation rates based on sensorgram curves in A, using Langmuir fitting. kobs was plotted

496

against the concentration of hGB1-2A to extract the normalized KD using a best-fit line in Excel.

497 498

Figure 5: Intra- and Interspecies Reactions. LE-GST, LS-GST and LT-GST baits were reacted

499

with hGB1-2A prey from EMCV (E), SafV (S) or TMEV (T). Bead-bound protein was

500

fractionated then detected by Western analysi s with αGST. The secondary (anti-murine) mAb is 22

501

also reactive with GB1 sequences. Band intensity values (hGB1-2A, ImageQuant) were

502

normalized to each respective input lane.

503 504

Figure 6: Ran and 2A Competitions. A) Recombinant GST or GST-LE with substitution

505

mutations K35Q, D37A or W40A were pre-bound to beads as baits, then reacted with equimolar

506

(50 nM) hGB1-2A, hRan (with 1 nM GST-RCC1), or a combination of hGB1-2A plus hRan.

507

Values for the captured proteins were detected by ImageQuant software and normalized to levels

508

of hGB1-2A pulled down by wild-type LE. B) The experiment in A was repeated (n=4), values

509

were averaged over all experiments and plotted to show the variance. Standard T-test

510

significance for plus or minus Ran, are indicated (“n.s.” not significant).

511 512

Figure 7: Phosphorylation of LE:2A Complexes. (A) GST-LE and mutant derivatives (bait) and

513

hGB1-2A or hGB1-2A (prey) bead-bound complexes were reacted with CK2 in the presence of

514

32

515

CK2 and cold ATP as in (12). Fractionated proteins were detected by phosphorimaging.

P-ATP. (B) Similar to A, the SYK plus 32P-ATP reactions were preceded by incubation with

516

~ 23 ~

pI Frag 1-50 E 1- 50 SPNAL---DISRTYPTLHVLIQFNHRGLEVRLFRHGQFWAETRADVILRSKTK 10.6 T 1- 52 NPAALYRIDLFITFTDEFITFDYKVHRRPVLTFRIPGF-GLTPAGRMLVCMGE 8.2 S 1- 52 NPVSIYRVDLFINFSDTVIQFTYKVHGRTVCQYEIPGF-GLSRSGRLLVCMGE 8.8 Con NPxALYRxDLFITFxDxxIxFxYKVHGRxVxxFRIPGF-GLTRAGRxLVCMGE Frag 51-100 10.0 E 51-100 QVSFLSNGNYPSMDSRAPWNPWKNTYQAVLRAEPCRVTMDIYYKRVRPFR 9.6 T 53- 87 QP---AHGPFTSS---------RSLYHVIFTATCSSFSFSIYKGRYRSWK 9.8 S 53- 87 KP---CQLPISTP---------KCFYHIVFTGSRNSFGVSIYKARYRPWK Con QP---xxGPxxSx---------KxxYHxVFTAxxxSFxxSIYKxRYRPWK NLS Frag 101-143 5.5 E 101-146 LPLVQKEWPVREENVFG-LYRIFNAHYAGYFADLLIHDIETNPG /PFMF 6.8 T 88-133 KP-IHDELVDRGYTTFGEFFKAVRGYHADYYRQRLIHDVETNPG /PVQS 6.0 S 88-133 QP-LHDELHDYGFSTFTDFFKAVRDYHASYYKQRLQHDIETNPG /PVQS Con xP-xHDELxDRGxxTFGxFFKAVRxYHAxYYxQRLIHDIETNPG scission cassette eIF4E

2B

at e Ly s

G ST -L E

G ST

Bait: GST-LE-

αGST GST-

2A-

α2A

Prey:

GST-LE

hGB1-2A

hGB1-2A

GST-L E

hGB1-2A

A. Bait:

GST-L

input

Coomassie

GST-LEGSThGB1-2AhGB1-

% 2A : 0 1 18 76 98

Bait: LE-GST

1-

Prey: GST-LE-

GST

hGB1-2Aor frags

1 100 93 85 % 2A:

% 2A: 73

hGB1-2A fragment input

Coomassie

5 25 0 50 0

12

αGST

12

5

mM NaCl

150 51 -1 10 00 114 3

B.

C.

50 51 -1 10 00 114 6

Bait: LE-GST Prey: hGB1-2A

0 0 11 43 102 : % L

0

0

(100) Resonance units

A.

8

50 mg/mL

7 6 5 4

20 mg/mL

3 2

10 mg/mL

1 0 0

1

2

3

5

4

6

7

Time (100 sec)

B. 6 kobs (s-1, 10-3)

ta=450 ta=600

R2=0.99

5 4 3

y=1380x+0.0021 y=konx+koff

2 1

0

5

10

15

hGB1-2A (100 nM)

20

25

Bait: LE-GST LS-GST LT-GST hGB1-2A Prey: E S T E S T E S T E S T LX-GSThGB1-2A% 2A input: 90 60 10 70 40 30 130 60 90

αGST

A. Bait:

B. GST

Significance of 2A:Ran Competition p