1 Human Enterovirus 71 Uncoating Captured at Atomic Resolution 2 ...

JVI Accepts, published online ahead of print on 18 December 2013 J. Virol. doi:10.1128/JVI.03029-13 Copyright © 2013, American Society for Microbiology. All Rights Reserved.

1

Title:

2 3

Human Enterovirus 71 Uncoating Captured at Atomic Resolution Running title: EV71 Uncoating Revealed by X-ray Structures

5 6 7

Ke LYU1, Jie DING1, Jian-Feng HAN2, Yu ZHANG2, Xiao-Yan Wu2, Ya-Ling HE1, Cheng-Feng QIN2, 3, Rong CHEN1#

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

1

Key Laboratory of Molecular Virology and Immunology, Institut Pasteur of Shanghai, Chinese Academy of Sciences, Shanghai 200031, China 2 Department of Virology, State Key Laboratory of Pathogen and Biosecurity, Beijing Institute of Microbiology and Epidemiology, Beijing 100071, China 3 Graduate School, Anhui Medical University, Hefei, China #

corresponding author: Rong Chen, Ph.D., Key Laboratory of Molecular Virology and Immunology, Institut Pasteur of Shanghai, Chinese Academy of Sciences, 320 Yue-Yang Road, Shanghai 200031, CHINA Telephone: (0086) 21-54920651. Fax: (0086) 21-54920651. Email: [email protected]

23 24 25

Abstract: 237 words Text (excluding references and figure legends): 5981 words

1

Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest

4

26 27

Abstract

28 29

Human enterovirus 71 (EV71) is the major causative agent of severe

30

hand-foot-and-mouth

31

characterization of EV71 during its lifecycle can aid in the development of

32

therapeutics against HFMD. Here, we present the atomic structures of the full virion

33

and an uncoating intermediate of a clinical EV71 C4 strain to illustrate the structural

34

changes in the full virion that lead to the formation of the uncoating intermediate

35

prepared for RNA release. Although the VP1 N-terminal regions, observed to

36

penetrate through the junction channel at the quasi 3-fold axis in the uncoating

37

intermediate of coxsackievirus A16, were not observed in the EV71 uncoating

38

intermediate, drastic conformational changes occur in this region, as has been

39

observed in all capsid proteins. Additionally, the RNA genome interacts with the

40

N-terminal extensions of VP1 and residues 32-36 of VP3, both of which are situated

41

at the bottom of the junction. These observations highlight the importance of the

42

junction for genome release. Furthermore, EV71 uncoating is associated with

43

apparent rearrangements and expansion around the 2- and 5-fold axes without obvious

44

changes around the 3-fold axes. Therefore, these structures enabled the identification

45

of hot spots for capsid rearrangements, which led to the hypothesis that the protomer

46

interface near the junction and the 2-fold axis permits the opening of large channels

47

for the exit of polypeptides and viral RNA, which is an uncoating mechanism that is

diseases

(HFMD)

young

children,

and

structural


2

in

48

likely conserved in enteroviruses.

49

Significance

50 Human enterovirus 71 (EV71) is the major causative agent of severe

52

hand-foot-and-mouth diseases (HFMD) in young children. EV71 contains a RNA

53

genome protected from an icosahedral capsid shell. Uncoating is essential in EV71

54

life cycle, characterized by conformational changes in the capsid to facilitate RNA

55

release into host cell. Here we presented the atomic structures of the full virion and an

56

uncoating intermediate of a clinical C4 strain EV71. Structural analysis revealed

57

drastic conformational changes associated with uncoating in all the capsid proteins

58

near the junction at the quasi 3-fold axis, and the protein-RNA interaction at the

59

bottom of the junction in the uncoating intermediate. Significant capsid

60

rearrangements also occur at the icosahedral 2- and 5-fold axis but not at the 3-fold

61

axis. Together we hypothesized that the junction and nearby are hot spots for capsid

62

breaches for the exit of polypeptides and viral RNA during uncoating.

63

3


51

64 65

Introduction Human enterovirus 71 (EV71) is the leading causative agent for severe

67

hand-foot-and-mouth diseases (HFMD) in infants and young children (1), and EV71

68

infection has caused significant morbidity and mortality in the Asia-Pacific regions.

69

During 2008-2012, numerous outbreaks of HFMD occurred in mainland China, with

70

over 2000 fatal cases reported (www.chinacdc.cn). Vaccines and effective drugs

71

remain unavailable.

72

EV71 belongs to Enterovirus species A within the Enterovirus genus of the

73

Picornaviridae (2). The EV71 virion contains a single-stranded positive-sense RNA

74

genome of 7.5 kb. The viral capsid, which has a diameter of approximately 300 Å, is

75

composed of 60 copies each of VP1-VP4 (3, 4) that are organized onto a quasi T=3

76

icosahedral lattice. The capsid proteins VP1, VP2 and VP3 each possess a β-sandwich

77

jelly roll fold and form the outer surface of the capsid shell, whereas VP4 is situated

78

inside the shell. The capsid surface is characterized by a depression encircling each

79

5-fold symmetry axis, which is referred to as the “canyon” that often contains the

80

receptor-binding site in picornaviruses (5, 6). The hydrophobic pocket of VP1, which

81

is located at the bottom of the canyon, contains a lipid moiety termed the “pocket

82

factor”, which likely stabilizes the mature virion. As observed in poliovirus and

83

certain picornaviruses, receptor binding at the junction site triggers the uncoating

84

process, which is characterized by the delivery of the viral genome into the host cell

4


66

compartment for replication and transcription (7). The uncoating is considered to be a

86

multistep process, including the conversion of the mature virion (160S) into an

87

expanded intermediate or A-particle (135S) characterized by the release of the pocket

88

factor and the opening of the 2-fold symmetry axis channels, and the generation of the

89

empty capsid (80S) from the A-particle characterized by the release of the RNA

90

genome (8-11).

91

The structural basis of picornavirus uncoating has been extensively studied

92

using both cryo-EM and X-ray crystallography. The uncoating products can be

93

obtained in vitro from virions by receptor binding and/or exposure to low pH (5.5-6.0)

94

or elevated temperature (3, 12-16). Upon receptor binding, the poliovirus converts to

95

the A-particle, which is associated with irreversible conformational changes,

96

including shifts of the capsid protein β-barrels and externalization of VP4 and the

97

N-terminus of VP1 (8, 17, 18). During coxsackievirus A7 uncoating triggered by heat,

98

both VP4 and RNA are released, and VP1 is rotated, which causes major

99

conformational changes at the interfaces of the capsid proteins VP1, VP2 and VP3

100

(19). The crystal structure of the empty capsid of human rhinovirus type 2 (HRV2),

101

obtained after release of the viral genome, exhibited a key movement around the

102

hydrophobic pocket of VP1 that allowed a coordinated shift of VP2 and VP3. This

103

overall displacement forces a reorganization of the inter-protomer interfaces, resulting

104

in particle expansion and the opening of 2-fold channels in the capsid, which

105

facilitates RNA egress (11).

5


85

The structures of different strains of EV71 have been determined (20-22). By

107

comparing the structures of the mature virion and a naturally occurring empty particle

108

(sedimentation coefficient of 80S), Wang et al. proposed a sensor-adaptor mechanism

109

for EV71 uncoating (21) and found that the largest movement of the polypeptide

110

occurs in residues 230-233 of VP1 at the end of the GH loop and at the beginning of

111

strand H. They referred to this polypeptide region as the adaptor-sensor, which is

112

directly downstream of a region of the GH loop that is external and structurally

113

variable in poliovirus (21). However, the naturally occurring empty particle is

114

composed of VP0 (the precursor of VP2 and VP4), VP1 and VP3, whereas the

115

uncoating empty capsid is composed of VP1, VP2 and VP3. Whether the structure is

116

identical between the two is unknown. Based on cryo-EM studies, Shingler et al.

117

proposed that the EV71 A-particle forms a gateway to allow genome release (23) and

118

found that the diameter of the 2-fold channel in the A-particle is approximately 10 Å,

119

whereas that in the empty capsid is only 6.4 Å. They thus proposed that the 2-fold

120

channel shrinks after RNA release. Recently, a coxsackievirus A16 (CVA16) A-like

121

particle was captured in atomic detail (24), which revealed that a portion of the

122

N-terminal extensions of VP1 is extruded through the capsid, thereby providing novel

123

insights into picornavirus uncoating.

124

In this study, we determined the crystal structures of the full virion and an

125

uncoating intermediate of a clinical C4 strain of EV71 at high resolution. Comparison

126

of these two structures enabled us to map the conformational changes associated with

6


106

127

uncoating and provide a more detailed picture to understand the early steps of EV71

128

uncoating. Furthermore, fitting of the crystal structure of the uncoating intermediate

129

into the cryo-EM reconstruction obtained by Shingler et al. revealed location of the

130

VP2 N-terminus and the interaction between capsid proteins and the RNA genome.


7

Materials and Methods

132

Virion purification

133

EV71 (strain AH08/06, GenBank accession no. HQ611148, with one amino acid

134

change at residue 227 of VP3 (K to Q)) was isolated from a HFMD patient in Anhui,

135

China in 2008. The virus was growing in RD cells in DMEM containing 10% FBS

136

until 90% of cells exhibit cytopathic effect (CPE). About 0.5-1L cell lysate was frozen

137

and thaw three times, then subjected to low speed centrifugation at 8228g for 30min

138

to remove cell debris. The virus supernatant was mixed with 50% polyethylene glycol

139

8000 (PEG8000) and 2M NaCl / PBS (pH7.2) to a final concentration of 5% and

140

200mM, respectively, and let stirring overnight at 4°C. Ultracentrifugation at

141

27000rpm in a SW28 rotor was carried out to spin down virus particles into a 40%

142

sucrose cushion followed by no-brake ultracentrifugation at 26000 rpm in a SW28

143

rotor for 4 hours onto a non-continuous 10-65% sucrose gradient to separate virions

144

from naturally occurring empty particles. Fractions obtained after ultracentrifugation

145

were subject to sodium dodecyl sulfate polyacrylamide gel electrophoresis

146

(SDS-PAGE) analysis. Those fractions containing virions were collected and

147

concentrated by one more round of ultracentrifugation at 28000rpm in a SW28 rotor

148

for 4h. The pelleted virions were resuspended in PBS buffer to a final concentration of

149

about 3 mg/ml. The quality of the virions was examined by negative staining electron

150

microscopy, showing a high homogeneity.

151

8


131

Crystallization and Diffraction Data Collection

153

The purified EV71 virions were subject to screening for crystallization conditions.

154

Crystals were obtained by the vapor diffusion method in hanging drops at 16°C by

155

mixing 2μl of virus (~3 mg/ml) and 2μl of the reservoir solution. The crystallization

156

condition for crystals of full virions was 0.1M NaAc (pH4.5) containing 3.5M sodium

157

formate.

158

Another specific crystallization condition triggered the conversion of the full virion to

159

an uncoating intermediate by unknown reasons and formation of crystals. The

160

crystallization condition for crystals of this uncoating intermediate was 0.1M sodium

161

cacodylate (pH7.0) containing 1.6M sodium acetate.

162

Prior to data collection cryo-protection of the crystals was achieved by resuspending

163

the crystals in the mother liquor with increasing concentrations of glycerol through

164

four steps: 5, 10, 15 and 20% (v/v). The equilibration time at each concentration was

165

at least 30s. These crystals were flash-frozen in liquid nitrogen and used for

166

diffraction data collection on an ADSC Quantum-315 CCD detector at the beamline

167

BL17U1 at Shanghai Synchrotron Radiation Facility. For the full virion, a dataset at

168

3.3Å resolution was collected with monochromatic X-rays (λ=0.97930Å) and a

169

detector to crystal distance of 350mm using an oscillation angle of 0.3° and an

170

exposure time of 2s. For the uncoating intermediate, a dataset at 3.8Å resolution was

171

collected with monochromatic X-rays (λ=0.97914Å) and a detector to crystal distance

172

of 400mm using an oscillation angle of 0.3° and an exposure time of 1s. Indexing,

9


152

173

integration, scaling, post-refinement and reduction of the data were carried out using

174

the HKL2000 software package (25). A total of 50 and 30 diffraction images were

175

used for data processing for the full virion and the uncoating intermediate,

176

respectively.

178

Structure Determination and Refinement

179

Crystals of the uncoating intermediate belong to space group P4232 with 5 copies of

180

the protomer as the asymmetric unit. Thus 5-fold non-crystallographic symmetry was

181

employed during structure determination and refinement. The program GLRF(26) was

182

used to calculate the self-rotation function, and combined with crystal packing

183

analysis to determine the position and orientation of the particle. The structure of

184

poliovirus type 1 empty capsid (PDB ID 1POV) was positioned at the origin with the

185

determined orientation in the unit cell, and used to calculate a 10Å density map as an

186

initial phase model. The phase refinement and extension were carried out using real

187

space averaging taking advantage of the 5-fold non-crystallographic symmetry. The

188

phase extension from 10 Å was carried out gradually in steps of one reciprocal lattice

189

point by iterative cycles of molecular averaging with non-crystallographic 5-fold

190

symmetry, solvent flattening, back transformation using RAVE(27) and CCP4(28).

191

Crystals of EV71 virions belong to space group I23 with 20 copies of the protomer as

192

the asymmetric unit. Thus 20-fold non-crystallographic symmetry was employed

193

during structure determination and refinement. Self-rotation function calculation and

10


177

crystal packing analysis were preformed, which showed that the particle center is not

195

at the origin. The virion structure was then determined by the molecular replacement

196

method using the program package PHENIX (29). To avoid model bias, the crystal

197

structure of a more distantly related picornavirus - coxsackievirus A9 (PDB ID 1D4M)

198

was taken as the search model and a single solution was obtained with log-likelihood

199

gain (LLG) = 11234 and translation function Z score (TFZ) = 53.2. A single solution

200

with LLG = 9098 and TFZ = 74.2 was obtained if the structure of poliovirus type 1

201

empty capsid (PDB ID 1POV) was taken as the search model.

202

After obtaining the initial density map, model building and refinement were carried

203

out iteratively using COOT (30) and PHENIX (29) programs, respectively. The final

204

refinement statistics are summarized in Table 1. The coordinates and structure factors

205

of the uncoating intermediate and the full virion have been deposited in the Protein

206

Data Bank (PDB IDs: 4N43 and 4N53). Figures were drawn and rendered with PyMol

207

(31) and Chimera (32).

208 209

Capsid expansion calculation

210

For the uncoating intermediate, the distances from all the Cα atoms to the particle

211

center were averaged to obtain the capsid radius. For the full virion, these same Cα

212

atoms were used to calculate the capsid radius. To calculate the capsid expansion at

213

the 5-fold axes, residues 142-149 in the DE loop and 183-187 in the FG loop of VP1

214

were taken for calculation. For the 2-fold axes, residues 90-98 and 249-254 of VP2

11


194

and 137-152 of VP3 were considered. For the 3-fold axes, residues 224-232 in the HI

216

loop of VP2 and 203-209 in the HI loop of VP3 were considered. For the junction,

217

residues 193-231 in the GH loop of VP1, residues 130-180 in the EF loop of VP2 and

218

residues 173-192 in the GH loop of VP3 were considered. All these residues were

219

chosen based on visual inspection of the capsid structures.

220 221

Cryo-electron microscopy

222

Aliquots of sample from the hanging drop containing crystals of the uncoating

223

intermediate were blotted onto glow discharged holey carbon grids and plunged into

224

liquid ethane cooled by liquid nitrogen using a FEI Vitrobot mark III freezing robot

225

(FEI). Data were collected in a FEI electron microscope operating at 300 kV with

226

89,000x magnification. Electron micrographs were recorded at a dosage about 10

227

electrons/Å2, with focal settings ranging from 0.9-3μm underfocus. Samples of

228

purified naturally occurring empty particles were also used for cryoEM experiments

229

to show different internal structure features.

230 231

Characterization of the uncoating intermediate

232

The proteolytic sensitivities of the full virion and the uncoating intermediate were

233

assessed by adding 0.5µg of trypsin into in a 5µl solution containing 0.5µg of the

234

corresponding particles. Samples with or without trypsin were incubated at 37°C. The

235

digestion was stopped by dilution of the samples with 1/5 volume of the 5×SDS

12


215

loading buffer followed by boiling of the mixture for 10 min. The samples were

237

subjected to 12% SDS-PAGE, and the VP1 was visualized by western blot using a

238

VP1 antibody (Abnova, Cat. No. PAB7631-D01P).

239

To detect the presence of VP4 in the uncoating intermediate, 0.025U of

240

α-chymotrypsin (Sigma, Cat. No. C4129) was added into a 10µl solution containing

241

1.2µg of the corresponding particles. Samples with or without α-chymotrypsin were

242

incubated at 25°C. The digestion was stopped by dilution of the samples with 1/5

243

volume of the 5×SDS loading buffer followed by boiling of the mixture for 10 min.

244

The samples were subjected to 15% SDS-PAGE, and VP4 was visualized by western

245

blot using a VP4 antibody (Biorbyt, Cat. No. orb10624) with VP2 as a loading control.

13


236

246

Results

247

Crystallization of two forms of EV71 A clinical EV71 C4 strain isolated from Anhui, China was used in this study. The

249

purified virions contained all four structural proteins VP1-4 (Fig. 1A) and retained

250

infectivity, as demonstrated by a cell infection study (data not shown). Crystallization

251

experiments were performed, and crystals were obtained in two different conditions.

252

Unexpectedly, the two determined structures exhibited significant differences, as

253

described in details below.

254 255

Structure of the full virion

256

The crystal structure of the first crystal form, the full virion, was determined at a

257

resolution of 3.3 Å (Fig. 2). The capsid is composed of 60 copies of the protomer.

258

Each protomer is composed of VP1, VP2, VP3 and VP4 (Figs. 3A-E). The quality of

259

the resulting electron density maps enabled the modeling of the polypeptides of VP1

260

(residues 1-297), VP2 (residues 9-254), VP3 (residues 1-242), VP4 (residues 12-69)

261

and the pocket factor (Figs. 3B-F).

262

The structure of the full virion was compared with that reported for the C4 strain

263

(PDB ID 3VBS) (21). Superposition of equivalent Cα atoms in the VP1, VP2, VP3

264

and VP4 proteins resulted in RMSD values of 0.94 Å, 0.48 Å, 0.68 Å and 1.28 Å,

265

respectively, indicating similar capsid protein structures. Amino acid sequence

266

alignment indicated differences in five residues within the capsid protein regions (Fig.

14


248

4). Residue 225 of VP1 in the GH loop near the pocket factor region is Cys (Fig. 3F)

268

rather than Met in 3VBS. All of the clinical EV71 C4 strains contain Cys at residue

269

225 of VP1. The virus strain used to obtain the crystal for structure determination of

270

3VBS was a vaccine strain that had been adapted into Vero cells. During this

271

adaptation process, mutation occurs in the viral genome, leading to a mutation at

272

residue 225 of VP1. Conversely, the virus strain used herein was propagated in RD

273

cells, which more closely resemble the natural host of EV71. Additionally, residue 98

274

of VP1 in the BC loop is Glu rather than Lys in 3VBS, and residue 144 of VP2 in the

275

EF loop is Ser rather than Thr in 3VBS. Both loops are involved in the formation of

276

neutralization epitopes (33-35). VP3 also contains two different residues: residue 93 is

277

Ser rather than Asn in 3VBS and residue 227 is Gln rather than Lys in 3VBS. Residue

278

93 is located on the surface loop (Fig. 5A). None of these mutations cause significant

279

local structural variations.

280 281

Structure of an expanded particle

282

The crystal structure of the second crystal form was determined at a resolution

283

of 3.8 Å (Fig. 2). The average radius in the capsid was found to increase from 130.9Å

284

(the full virion) to 138.5 Å (the expanded particle) with an enlargement of

285

approximately 5.8%. The refined model includes residues 72-296 of VP1, 16-47 and

286

54-250 of VP2 and 1-175 and 189-236 of VP3 (Figs. 3B-D). Residues 48-53 in VP2

287

and 176-188 in VP3 are disordered. The N-terminal region (residues 1-71) of VP1

15


267

becomes disordered in the expanded particle, and the pocket region is empty (Fig. 3F).

289

The disposition of the VP1-3 subunits is largely maintained in the expanded particle

290

in comparison to those in the full virion. The largest structural changes were mapped

291

to the terminal regions and some loops connecting the strands of the β-barrel.

292

Superpositions of the individual VP1, VP2 and VP3 proteins in these two particles

293

resulted in RMSD values of 2.47 Å, 1.69 Å and 2.07 Å for equivalent Cα atoms,

294

respectively (Figs. 3B-D), whereas superpositions of the individual β-barrels of VP1,

295

VP2 and VP3 resulted in RMSD values of 2.06 Å, 1.67 Å and 1.98 Å, respectively.

296

Thus, VP1 and VP3 undergo more substantial conformational changes than VP2. In

297

VP1, in addition to the disordered N-terminal extensions, drastic conformational

298

changes also occur in the C-terminal region (residues 266-297) and the GH loop

299

(residues 191-230), with RMSD values of 2.20 Å and 2.66 Å, respectively. The

300

C-terminal regions wrap around the VP3 of the same protomer, whereas the GH loop

301

lies near the junction at the quasi 3-fold axis (Fig. 5C). In VP2, the overall structure is

302

rather conserved, except that both of the N-terminal extensions and residues 48-53

303

that precede the β-barrel core become disordered. In VP3, a 310-helix in the GH loop

304

(residues 173-192) of VP3 in the full virion located near the junction becomes

305

disordered (Figs. 3C-D). Moreover, significant changes occur in the FG loop

306

(residues 159-164) with an RMSD value of 2.74 Å.

307 308

Identification of the expanded particle as an uncoating intermediate

16


288

The mature virions of some enteroviruses, including poliovirus and human

310

rhinovirus 14, exist in a metastable state, transiently exposing VP4 and the

311

N-terminus of VP1 through hot spots in the capsid near the junction or 2-fold axis in a

312

process called “breathing” (36, 37). Typically, the exposed polypeptides rapidly

313

retract into the particle, and the breathing is a reversible process under physiological

314

conditions. However, under certain conditions such as upon receptor binding and/or

315

acidic pH or elevated temperature, the conformational changes associated with

316

breathing become irreversible (3). The structural features we identified suggest that

317

the first crystal form corresponded to the full virion, whereas the second crystal form

318

corresponded to a capsid-rearranged product. Because the crystals were grown at

319

16°C, breathing, which occurs only at 37°C (36, 37), should not have occurred.

320

Therefore, we speculate that the second crystallization condition triggered the

321

uncoating process of EV71. Generally, enterovirus produces two forms of particles

322

during uncoating: an altered “A-particle” (containing RNA) and an empty capsid

323

(without RNA) (3). To determine whether the second crystal form contained RNA in

324

the interior of the particle, samples from the crystallization drops that contained

325

crystals of the uncoating intermediate were analyzed using cryo-EM. In the cryo-EM

326

images, the internal density corresponded to the RNA genome, whereas the purified

327

naturally occurring empty particle was completely devoid of internal density (Fig. 1B).

328

The presence of the RNA genome in the interior of the particle suggests that this is an

329

uncoating intermediate. The specific crystallization condition triggered the uncoating

17


309

330

process and converted the full virion into the uncoating intermediate. In poliovirus, the A-particle is more susceptible to proteolysis than the intact

332

virion and characterized by the irreversible externalization of the N terminus of VP1

333

(38-40). To assess and compare the proteolytic sensitivities of the full virion and the

334

uncoating intermediate, we performed trypsin digestion experiments. The

335

crystallization hanging drops containing crystals of the uncoating intermediate and the

336

full virion were digested with trypsin, respectively (Fig. 1C). In hanging drops

337

containing crystals of the full virion, VP1 was resistant to trypsin digestion; whereas

338

in hanging drops containing crystals of the uncoating intermediate, a truncated VP1

339

product was observed after trypsin digestion. This suggests the uncoating intermediate

340

obtained in this study has increased protease sensitivity.

341

To determine whether VP4 is still present, hanging drops containing crystals of

342

the uncoating intermediate were digested by α-chymotrypsin and VP4 was visualized

343

by western blot (Fig. 1D). Unexpectedly, most of VP4 was resistant to

344

α-chymotrypsin digestion. Similar results were obtained when trypsin was used for

345

digestion. These studies suggest that most of the VP4 is still associated with the

346

uncoating intermediate and protected from proteolysis.

347 348

Channel expansion at the 2-fold, but not 5-fold, axes in the uncoating

349

intermediate

350

In the virion structure, at the top of each 5-fold axis, a channel with a diameter

18


331

of approximately 13 Å runs from the particle surface towards the particle center. At

352

the top opening of the channel, the BC, EF and HI loops of VP1 form a large

353

protrusion (Fig. 5A). An elaborate network of hydrogen bonds established between

354

Lys182 and Asp185 from a neighboring VP1 maintains the channel in the open

355

conformation and stabilizes this conformation at the 5-fold axes (Fig. 5B). Both of

356

these residues (Lys182 and Asp185) are conserved among enteroviruses in species A

357

and D, whereas they are divergent in species B and C (data not shown). In the

358

structure of the uncoating intermediate, despite significant enlargement of the capsid

359

along the 5-fold axes, no apparent expansion of the 5-fold channel was observed, and

360

the hydrogen bonding network between the 5-fold-related Lys182 and Asp185 is

361

maintained.

362

Conversely, a clear expansion of the 2-fold channels was observed in the

363

uncoating intermediate (Fig. 6). This channel, which is located at the interface of two

364

neighboring pentamers and is surrounded by segments of VP2 and VP3, expands from

365

dimensions of 7.0 Å×27.6 Å in the full virion to 11.7 Å×27.7 Å in the uncoating

366

intermediate, with the channel dimension calculated between the 2-fold-related Cα

367

atoms (Fig. 6). The top of this pore is formed by the C-terminal segment and one

368

α-helix from the CD loop of VP2, an additional α-helix from the EF loop of VP3 and

369

their 2-fold symmetry mates in the particle. The bottom of the pore is formed by two

370

segments from VP2, including a sharp turn at residues 17-18 and a loop region at

371

residues 54-58, both of which precede the β-barrel core structures. This pore opening

19


351

372

at the 2-fold axes was also observed in the uncoating intermediate of CVA16 (24) and

373

the empty capsid of HRV2 that resulted from the release of the genomic RNA (11).

374 375

Structural rearrangements during uncoating Capsid expansion implies rearrangements of the interactions at the interfaces

377

between and within protomers. The junction, which is surrounded by VP1 and VP2

378

from one protomer and VP3 from a neighboring protomer that is related by the 5-fold

379

symmetry, undergoes drastic conformational changes, leading to rearrangements in

380

the interactions between protomers around the 5-fold axes and across the 2-fold axes

381

(Fig. 5C). The junction also undergoes expansion during uncoating, as the distance to

382

the particle center increases from 137.95 Å to 143.06 Å with a change of

383

approximately 3.7%. Additionally, significant structural changes occur at the junction,

384

as observed in the GH loops of VP1 and VP3 (Fig. 5C).

385

The overall organization of the uncoating intermediate is similar to that of the

386

full virion. The distances from the particle center to the surface along the 5-, 3- and

387

2-fold axes are 149.49 Å, 146.44 Å and 137.45 Å, respectively, in the uncoating

388

intermediate, in comparison to 143.11 Å, 146.14 Å and 135.12 Å, respectively, in the

389

full virion. These distances imply an expansion of approximately 4.5%, 0.2% and

390

1.7% in the capsid along the 5-, 3- and 2-fold axes, respectively, indicating a

391

substantially smaller enlargement along the 3-fold axes. The absolute distance shift

392

for every Cα atom between the full virion and the uncoating intermediate was also

20


376

calculated and mapped onto the full virion structure (Fig. 7A). The most dramatic

394

movement lies in VP4 and the terminal regions of VP1, VP2 and VP3 (colored red)

395

(Fig. 7B), with modest changes in VP1 around the 5-fold axes (colored green). The

396

protomer interfaces along the 2-fold axes (predominantly VP2/VP2) and the junction

397

between neighboring 2- and 5-fold axes, in addition to the intra-protomer interfaces

398

(VP1/VP2 and VP1/VP3) (colored cyan), exhibit few changes, and the VP2/VP3

399

portions around the 3-fold axes (colored blue) exhibit the fewest changes (Fig. 7B).

400

Thus, capsid regions around the 2- and 5-fold axes and the junction undergo

401

significant structural changes in comparison to relatively minor variations around the

402

3-fold axes. The uncoating process appears to be a coordinated process with major

403

rearrangements around the 2- and 5-fold axes resulting in local capsid expansion

404

without obvious changes around the 3-fold axes. A similar analysis was conducted

405

using structures of the full virion and the empty capsid after HRV2 uncoating, which

406

exhibited similar structural changes (Fig. 7C).

407 408

Fitting of the crystal structure into the cryo-EM reconstruction of the uncoating

409

intermediate

410

Previously, the cryo-EM structure of the EV71 A-particle of the 1095/Shiga

411

strain was determined at a resolution of 6.3 Å (EMDB: 5465) (23). Five residues that

412

were mapped to capsid proteins were divergent between the 1095 strain (PDB ID

413

4GMP) (22) and our clinical C4 strain (Fig. 4), three of which are linked to EV71

21


393

antigenicity (34). Residue 22 of VP1 is His rather than Arg in 4GMP, and residue 145

415

of VP1 in the surface loop is Glu rather than Gly in 4GMP. Nishimura et al. recently

416

found that the PSGL-1 binding phenotype of EV71 strains is regulated by a single

417

residue (residue 145 of VP1) that maps to the center of the 5-fold mesa (41).

418

Additionally, this residue has been shown to be a determinant for the strain-specific

419

antigenicity of EV71 (42). Residue 126 of VP2 in the β strand E is Val rather than Ile

420

in 4GMP. VP3 also contains two divergent residues: residue 29 is His rather than Tyr

421

in 4GMP, and residue 227 is Gln rather than Lys in 4GMP. Residue 29 is located on

422

the capsid interior surface. None of these residue variations cause significant local

423

structural variations.

424

The crystal structure of our uncoating intermediate was fitted into the A-particle

425

cryo-EM density map (EMDB: 5465, σ=1) (23). The crystal structure agreed well

426

with the cryo-EM structural features, except for a few protruding loops, the most

427

distinct of which is a region in the EF loop of VP2 (residues 136-145) that lies near

428

the junction. Additionally, residues 54-59 of VP2 in the inner side of the 2-fold

429

channel lie outside of the electron density. Interestingly, residues 48-54 of VP2 are

430

disordered in the crystal structure of the uncoating intermediate. We observed a

431

relatively large 2-fold channel, and the capsid interior surface exhibits large patches of

432

negative charge at the 2-fold axis with minimal interspersed positive charge in both

433

the uncoating intermediate and the EV71 A-particle cryo-EM structure.

434

One novel observation is that the visible N-terminus of VP1 (residue 72) and

22


414

residues 32-36 of VP3 in the interior of the capsid, both of which are adjacent to the

436

bottom of junction, interact with the inner RNA density (Fig. 8A). In the full virion,

437

VP4 and the N-terminal extensions of VP1, VP2 and VP3 are packed in layers in the

438

interior of the capsid and interact with the RNA genome. In the uncoating

439

intermediate, it appears that only the N-terminal extensions of VP1 and residues

440

32-36 of VP3 interact with the RNA instead. A structural comparison indicated a lack

441

of obvious shift of this VP3 region during uncoating. Moreover, we observed electron

442

density extending from the visible N-terminus of VP2 (residue 16) into the interior of

443

the capsid, but interaction of VP2 with the RNA genome was not observed (Fig. 8B).

444

The N-terminus of VP1 is externalized at the base of the canyon, as observed

445

from cryo-EM reconstruction analysis of the A-particle. The ordered region of the

446

EV71 uncoating intermediate begins at Ser72 and is located in the center of the quasi

447

3-fold axis in the top view. Based on the correlation between the visible VP1

448

N-termini in the uncoating intermediate and the A-particle electron density map, we

449

speculate that the VP1 N-terminal extensions (residues 1-71) may externalize from

450

the capsid surface in the junction channel.

451 452

Uncoating in EV71 compared with CVA16

453

Recently, the structure of the CVA16 A-like particle was determined at a

454

resolution of 3.0 Å (PDB ID 4JGY) (24). Superposition of equivalent Cα atoms in the

455

VP1, VP2 and VP3 proteins in our uncoating intermediate and 4JGY resulted in

23


435

456

RMSD values of 1.27 Å, 0.89 Å and 1.54 Å, respectively, suggesting similar

457

structures between the EV71 uncoating intermediate and the CVA16 A-like particle

458

(Figs. 9A-D). In both expanded particles, the GH loop of VP3, which is facing the junction

460

channel, becomes disordered (Fig. 9D). Structural differences between the two

461

particles are all mapped to regions near the junction or the 2-fold axis channel. One

462

major variation occurs in the N-terminal extensions of VP1 (residues 62-71) (Figs. 9B

463

and 9E). This region of the polypeptide was found to traverse the capsid in the

464

CVA16 A-like particle, whereas it is disordered in the EV71 uncoating intermediate

465

(Figs. 9B and 9E). Additionally, in the EV71 uncoating intermediate, a loop region

466

(residues 48-53) of VP2 that precedes the β-barrel core structure and runs between the

467

2-fold channel and the junction is disordered, whereas in the CVA16 A-like particle,

468

the GH loops of both VP1 and a portion of the EF loop of VP2 (residues 137-141),

469

which both face the junction channel, are disordered (Fig. 9F).

470

24


459

471

Discussion Previous studies demonstrated that the A-particle and the empty capsid of some

473

enteroviruses can be produced through incubation with the corresponding receptors or

474

after treatment under specific physical conditions. Human scavenger receptor class B

475

member 2 (SCARB2), which is an identified uncoating receptor for EV71, has been

476

shown to convert the EV71 virion into either the 135S particle (43, 44) or an empty

477

capsid that lacks both genomic RNA and VP4 (45) after incubation under acidic

478

conditions. Shingler et al. demonstrated that a mixture of the A-particle and empty

479

capsid can be produced from purified EV71 virions by heating. In this study, we

480

obtained crystals of an uncoating intermediate of EV71 from the purified infectious

481

virions of a clinical EV71 C4 strain under a specific crystallization condition. This

482

intermediate shares some structural features in the capsid with the naturally occurring

483

empty particle composed of VP1, VP3 and VP0, which is the precursor for VP2 and

484

VP4, such as the opening of pores at the 2-fold axes, loss of ordered structures in the

485

VP1 N-terminal extensions and expansion of the capsid. The junction at the quasi

486

3-fold axis (Fig. 6A) in the intermediate structure exhibited drastic conformational

487

changes from that of the full virion. Furthermore, fitting of the crystal structure into

488

the EM density map revealed that the electron density observed between the RNA

489

genome and the capsid proteins (particularly the N-terminal extensions of VP1) is

490

attached to the bottom of the junction. A previous electron tomography analysis of

491

poliovirus during RNA release indicated that the footprint of RNA on the capsid outer

25


472

492

surface was located approximately 20 Å away from a 2-fold axis (18). Collectively,

493

these observations suggest that large pores temporarily open at the junction and the

494

2-fold axes are hot spots for polypeptide and RNA release. In the poliovirus virion, the N-terminus of VP1 is located near the 5-fold axis

496

(46). EM analysis using a Fab directed against the N-terminus of VP1 indicated that

497

the full virion “breathes” with the N-terminus of VP1 (residues 1-53) transiently

498

exposed (47). In the poliovirus A-particle, antibody labeling studies indicated that the

499

location of the N-terminus of VP1 shifts towards the tips of the 3-fold propeller,

500

which lies near the junction, and becomes externalized (48). In the EV71 virion,

501

however, the N-termini of VP1 are located near the 2-fold axis (20, 21). In the EV71

502

A-particle (EMDB: 5465), cryo-EM analysis suggested that the N-terminus of VP1 is

503

externalized at the base of the canyon (Fig. 3) (23). The crystal structure of a CVA16

504

135S-like particle (24) revealed that residues 62-71 of VP1 penetrate through the

505

junction. The present crystal structure of the EV71 uncoating intermediate revealed

506

that the visible N-termini of VP1 lie at the bottom of the junction, implying that the

507

N-terminal extensions of VP1 may be externalized through the junction. Collectively,

508

these observations suggest that the externalization mechanism of the N-termini of

509

VP1 from different enteroviruses may undergo function-driven convergence.

510

A novel observation from this study is that the capsid expands significantly

511

around the 2- and 5-fold axes and the junction between the quasi 3-fold axes, in

512

contrast to the much lower expansion observed around the 3-fold axes. Additionally,

26


495

capsid regions around the 3-fold axes undergo fewer structural perturbations, in

514

contrast to regions around the 2- and 5-fold axes and the junction. Such differences

515

during capsid rearrangement may arise from the significantly fewer interactions

516

present around the 2-fold axes and the quasi 3-fold axes, which makes these regions

517

hot spots for capsid breaches. Because the junction is located between neighboring 2-

518

and 5-fold axis pairs, structural changes may easily propagate into regions around the

519

5-fold axes. It is possible that the appropriate protomer interface near the 2-fold axis

520

channel and the junction may open further to permit RNA release.

521

Another novel observation is that in the EV71 uncoating intermediate only the

522

visible N-terminus (residue 72) of VP1 and residues 32-36 of VP3 in the interior

523

capsid near the junction interact with the inner RNA density. In the full virion, the

524

VP4 and the N-terminal regions of VP1, VP2 and VP3 are packed inside the capsid

525

and directly interact with the RNA genome. The disposition of residues 32-36 in VP3

526

is not greatly altered during uncoating. Such protein-RNA interactions in the

527

uncoating intermediate may facilitate communication from outside of the capsid to the

528

RNA to trigger RNA release or provide an anchored site for RNA release.

529

Enterovirus uncoating occurs in multiple steps. One of the early steps involves

530

the formation of an expanded, altered “A-particle” that is primed for genome release.

531

One of the late steps involves an unknown trigger that results in RNA expulsion,

532

generating an empty capsid. The expanded particle analyzed in this study likely

533

represented an intermediate at the initial stage of uncoating, likely prior to the

27


513

formation of A-particle, as shown by the increased protease sensitivity of VP1 and the

535

retaining of VP4. We addressed the question of how the EV71 capsid conformation

536

changes in the initial stage by comparing the crystal structure of the full virion and the

537

uncoating intermediate and observed specific protein-RNA interactions in the

538

uncoating intermediate, whereas Shingler et al. found that the 2-fold channel regulates

539

genome release in the late stage of uncoating. Collectively, these findings provide a

540

more complete understanding of EV71 uncoating.

541

In summary, together with previous studies, the results presented herein provide

542

a more complete model for enterovirus uncoating. Under certain conditions,

543

conformational changes that are associated with breathing become irreversible to

544

initiate the uncoating process. Conformational changes in the junction region (such as

545

the GH loops of VP1 and VP3), particularly the expulsion of the pocket factor that is

546

buried inside the hydrophobic pocket of VP1, propagate into other regions of the

547

capsid proteins, and the C-terminal region of VP1 shifts away from the VP3 surface to

548

allow more dramatic conformational changes in VP3 to occur. These structural

549

changes permit the opening of the 2-fold channel, leading to rearrangements and

550

expansion at the junction and the 5-fold axes. The extreme N-termini of VP1 and later

551

the VP4 are externalized through the capsid breaches at the protomer interface near

552

the 2-fold axis and the junction, anchoring the amphipathic helices into the membrane

553

and resulting in the formation of the A-particle. The RNA genome, through its

554

interaction with the N-terminal extensions of VP1, is thus poised at the bottom of the

28


534

555

junction, awaiting the trigger for its release to complete the uncoating process (34).

556

One of our future studies is to further characterize the conformational changes at high

557

resolution during the late stages of EV71 uncoating. During revision of our manuscript, Butan et al. presented a high-resolution

559

cryo-EM structure of the 135S particle of type 1 poliovirus(49), which revealed

560

externalization of the N-terminal regions of VP1 near the quasi 3-fold axis. Their

561

study also provided evidence that uncoating is a succession of step-wise change. Thus

562

different stages of uncoating may be sampled in their study from that in our study,

563

which explains different structural features. Nonetheless, their structural analysis on

564

the 135S particle also supported our observation that the junction at the quasi 3-fold

565

axis and the GH loops in VP1 and VP3 play important roles during uncoating.

566 567

29


558

Acknowledgments

569 570 571 572 573 574 575 576 577 578 579

Research in R Chen’s group was supported by 100 Talents´ Program of Chinese Academy of Sciences, Shanghai Pu-Jiang Career Development Award (grant no. 09PJ1411400), 973 Project (grant no. 2010CB912403), and Frontier Research Award from Shanghai Institutes for Biological Sciences-Chinese Academy of Sciences (grant no. 2008KIP105). Research in CF Qin’s lab was supported by Beijing Natural Science Foundation (grants no. 7122129 and no. 7112108) and National Science Foundation of China (grant no. 31270195), and Beijing Nova Program of Science and Technology (grant no. 2010B041). We thank the staff at Shanghai Synchrotron Radiation Facility (Beamline BL17U1) for on-site assistance and the staff at Institute of Biophysics in Chinese Academy of Sciences for Cryo-EM data collection. We also thank Prof. Felix A. Rey from Institut Pasteur for helpful suggestions and discussions.

30


568

References 1. 2.

3. 4. 5.

6.

7. 8.

9. 10.

11.

12. 13.

14.

15.

McMinn, P. C. 2012. Recent advances in the molecular epidemiology and control of human enterovirus 71 infection. Curr Opin Virol 2:199-205. Racaniello, V. R. 2007. Picornaviridae: The Viruses and Their Replication, p. p 795-838. In D. M. H. Knipe, Peter M. (ed.), Fields Virology, vol. I. Tuthill, T. J., E. Groppelli, J. M. Hogle, and D. J. Rowlands. 2010. Picornaviruses. Curr Top Microbiol Immunol 343:43-89. Rossmann, M. G., and J. E. Johnson. 1989. Icosahedral RNA virus structure. Annu Rev Biochem 58:533-573. Rossmann, M. G., E. Arnold, J. W. Erickson, E. A. Frankenberger, J. P. Griffith, H. J. Hecht, J. E. Johnson, G. Kamer, M. Luo, A. G. Mosser, and et al. 1985. Structure of a human common cold virus and functional relationship to other picornaviruses. Nature 317:145-153. Rossmann, M. G. 1989. The canyon hypothesis. Hiding the host cell receptor attachment site on a viral surface from immune surveillance. J Biol Chem 264:14587-14590. Helenius, A. 2007. Virus Entry and Uncoating. In D. M. H. Knipe, Peter M. (ed.), Fields Virology, 5th ed, vol. I. Belnap, D. M., D. J. Filman, B. L. Trus, N. Cheng, F. P. Booy, J. F. Conway, S. Curry, C. N. Hiremath, S. K. Tsang, A. C. Steven, and J. M. Hogle. 2000. Molecular tectonic model of virus structural transitions: the putative cell entry states of poliovirus. J Virol 74:1342-1354. Huang, Y., J. M. Hogle, and M. Chow. 2000. Is the 135S poliovirus particle an intermediate during cell entry? J Virol 74:8757-8761. Tuthill, T. J., D. Bubeck, D. J. Rowlands, and J. M. Hogle. 2006. Characterization of early steps in the poliovirus infection process: receptor-decorated liposomes induce conversion of the virus to membrane-anchored entry-intermediate particles. J Virol 80:172-180. Garriga, D., A. Pickl-Herk, D. Luque, J. Wruss, J. R. Caston, D. Blaas, and N. Verdaguer. 2012. Insights into minor group rhinovirus uncoating: the X-ray structure of the HRV2 empty capsid. PLoS Pathog 8:e1002473. Prchla, E., E. Kuechler, D. Blaas, and R. Fuchs. 1994. Uncoating of human rhinovirus serotype 2 from late endosomes. J Virol 68:3713-3723. Curry, S., C. C. Abrams, E. Fry, J. C. Crowther, G. J. Belsham, D. I. Stuart, and A. M. King. 1995. Viral RNA modulates the acid sensitivity of foot-and-mouth disease virus capsids. J Virol 69:430-438. Arita, M., S. Koike, J. Aoki, H. Horie, and A. Nomoto. 1998. Interaction of poliovirus with its purified receptor and conformational alteration in the virion. J Virol 72:3578-3586. Nurani, G., B. Lindqvist, and J. M. Casasnovas. 2003. Receptor priming of 31


580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621

16.

17.

18.

19.

20. 21.

22.

23.

24.

25.

26.

major group human rhinoviruses for uncoating and entry at mild low-pH environments. J Virol 77:11985-11991. Milstone, A. M., J. Petrella, M. D. Sanchez, M. Mahmud, J. C. Whitbeck, and J. M. Bergelson. 2005. Interaction with coxsackievirus and adenovirus receptor, but not with decay-accelerating factor (DAF), induces A-particle formation in a DAF-binding coxsackievirus B3 isolate. J Virol 79:655-660. Bubeck, D., D. J. Filman, N. Cheng, A. C. Steven, J. M. Hogle, and D. M. Belnap. 2005. The structure of the poliovirus 135S cell entry intermediate at 10-angstrom resolution reveals the location of an externalized polypeptide that binds to membranes. J Virol 79:7745-7755. Bostina, M., H. Levy, D. J. Filman, and J. M. Hogle. 2011. Poliovirus RNA is released from the capsid near a twofold symmetry axis. J Virol 85:776-783. Seitsonen, J. J., S. Shakeel, P. Susi, A. P. Pandurangan, R. S. Sinkovits, H. Hyvonen, P. Laurinmaki, J. Yla-Pelto, M. Topf, T. Hyypia, and S. J. Butcher. 2012. Structural analysis of coxsackievirus A7 reveals conformational changes associated with uncoating. J Virol 86:7207-7215. Plevka, P., R. Perera, J. Cardosa, R. J. Kuhn, and M. G. Rossmann. 2012. Crystal structure of human enterovirus 71. Science 336:1274. Wang, X., W. Peng, J. Ren, Z. Hu, J. Xu, Z. Lou, X. Li, W. Yin, X. Shen, C. Porta, T. S. Walter, G. Evans, D. Axford, R. Owen, D. J. Rowlands, J. Wang, D. I. Stuart, E. E. Fry, and Z. Rao. 2012. A sensor-adaptor mechanism for enterovirus uncoating from structures of EV71. Nat Struct Mol Biol 19:424-429. Cifuente, J. O., H. Lee, J. D. Yoder, K. L. Shingler, M. S. Carnegie, J. L. Yoder, R. E. Ashley, A. M. Makhov, J. F. Conway, and S. Hafenstein. 2013. Structures of the procapsid and mature virion of enterovirus 71 strain 1095. J Virol 87:7637-7645. Shingler, K. L., J. L. Yoder, M. S. Carnegie, R. E. Ashley, A. M. Makhov, J. F. Conway, and S. Hafenstein. 2013. The enterovirus 71 A-particle forms a gateway to allow genome release: a cryoEM study of picornavirus uncoating. PLoS Pathog 9:e1003240. Ren, J., X. Wang, Z. Hu, Q. Gao, Y. Sun, X. Li, C. Porta, T. S. Walter, R. J. Gilbert, Y. Zhao, D. Axford, M. Williams, K. McAuley, D. J. Rowlands, W. Yin, J. Wang, D. I. Stuart, Z. Rao, and E. E. Fry. 2013. Picornavirus uncoating intermediate captured in atomic detail. Nat Commun 4:1929. Z. Otwinowski, W. M. 1997. Processing of X-ray Diffraction Data Collected in Oscillation Mode, p. 307-326, Methods Enzymol, vol. 276. Academic Press, New York. Tong, L., and M. G. Rossmann. 1997. Rotation function calculations with GLRF program. Methods Enzymol 276:594-611. 32


622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663

27.

28. 29.

30. 31. 32.

33.

34.

35.

36.

37.

38.

39.

Kleywegt, G. J. 1999. Experimental assessment of differences between related protein crystal structures. Acta Crystallogr D Biol Crystallogr 55:1878-1884. 1994. The CCP4 suite: programs for protein crystallography. Acta Crystallogr D Biol Crystallogr 50:760-763. Adams, P. D., P. V. Afonine, G. Bunkoczi, V. B. Chen, I. W. Davis, N. Echols, J. J. Headd, L. W. Hung, G. J. Kapral, R. W. Grosse-Kunstleve, A. J. McCoy, N. W. Moriarty, R. Oeffner, R. J. Read, D. C. Richardson, J. S. Richardson, T. C. Terwilliger, and P. H. Zwart. 2010. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr D Biol Crystallogr 66:213-221. Emsley, P., B. Lohkamp, W. G. Scott, and K. Cowtan. 2010. Features and development of Coot. Acta Crystallogr D Biol Crystallogr 66:486-501. Schrodinger, L. 2010. The PyMOL Molecular Graphics System, Version 1.3r1. Pettersen, E. F., T. D. Goddard, C. C. Huang, G. S. Couch, D. M. Greenblatt, E. C. Meng, and T. E. Ferrin. 2004. UCSF Chimera--a visualization system for exploratory research and analysis. J Comput Chem 25:1605-1612. Foo, D. G., R. X. Ang, S. Alonso, V. T. Chow, S. H. Quak, and C. L. Poh. 2008. Identification of immunodominant VP1 linear epitope of enterovirus 71 (EV71) using synthetic peptides for detecting human anti-EV71 IgG antibodies in Western blots. Clin Microbiol Infect 14:286-288. Gao, F., Y. P. Wang, Q. Y. Mao, X. Yao, S. Liu, F. X. Li, F. C. Zhu, J. Y. Yang, Z. L. Liang, F. M. Lu, and J. Z. Wang. 2012. Enterovirus 71 viral capsid protein linear epitopes: identification and characterization. Virol J 9:26. Liu, C. C., A. H. Chou, S. P. Lien, H. Y. Lin, S. J. Liu, J. Y. Chang, M. S. Guo, Y. H. Chow, W. S. Yang, K. H. Chang, C. Sia, and P. Chong. 2011. Identification and characterization of a cross-neutralization epitope of Enterovirus 71. Vaccine 29:4362-4372. Li, Q., A. G. Yafal, Y. M. Lee, J. Hogle, and M. Chow. 1994. Poliovirus neutralization by antibodies to internal epitopes of VP4 and VP1 results from reversible exposure of these sequences at physiological temperature. J Virol 68:3965-3970. Lewis, J. K., B. Bothner, T. J. Smith, and G. Siuzdak. 1998. Antiviral agent blocks breathing of the common cold virus. Proc Natl Acad Sci U S A 95:6774-6778. Fricks, C. E., and J. M. Hogle. 1990. Cell-induced conformational change in poliovirus: externalization of the amino terminus of VP1 is responsible for liposome binding. J Virol 64:1934-1945. De Sena, J., and B. Mandel. 1976. Studies on the in vitro uncoating of poliovirus. I. Characterization of the modifying factor and the modifying reaction. Virology 70:470-483. 33


664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705

40.

41.

42.

43.

44.

45.

46. 47.

48.

49.

50.

De Sena, J., and B. Mandel. 1977. Studies on the in vitro uncoating of poliovirus. II. Characteristics of the membrane-modified particle. Virology 78:554-566. Nishimura, Y., H. Lee, S. Hafenstein, C. Kataoka, T. Wakita, J. M. Bergelson, and H. Shimizu. 2013. Enterovirus 71 Binding to PSGL-1 on Leukocytes: VP1-145 Acts as a Molecular Switch to Control Receptor Interaction. PLoS Pathog 9:e1003511. Lee, H., J. O. Cifuente, R. E. Ashley, J. F. Conway, A. M. Makhov, Y. Tano, H. Shimizu, Y. Nishimura, and S. Hafenstein. 2013. A strain-specific epitope of Enterovirus 71 identified by cryoEM of the complex with Fab from neutralizing antibody. J Virol. Yamayoshi, S., Y. Yamashita, J. Li, N. Hanagata, T. Minowa, T. Takemura, and S. Koike. 2009. Scavenger receptor B2 is a cellular receptor for enterovirus 71. Nat Med 15:798-801. Chen, P., Z. Song, Y. Qi, X. Feng, N. Xu, Y. Sun, X. Wu, X. Yao, Q. Mao, X. Li, W. Dong, X. Wan, N. Huang, X. Shen, Z. Liang, and W. Li. 2012. Molecular determinants of enterovirus 71 viral entry: cleft around GLN-172 on VP1 protein interacts with variable region on scavenge receptor B 2. J Biol Chem 287:6406-6420. Yamayoshi, S., S. Ohka, K. Fujii, and S. Koike. 2013. Functional comparison of SCARB2 and PSGL1 as receptors for enterovirus 71. J Virol 87:3335-3347. Hogle, J. M., M. Chow, and D. J. Filman. 1985. Three-dimensional structure of poliovirus at 2.9 A resolution. Science 229:1358-1365. Lin, J., L. Y. Lee, M. Roivainen, D. J. Filman, J. M. Hogle, and D. M. Belnap. 2012. Structure of the Fab-labeled "breathing" state of native poliovirus. J Virol 86:5959-5962. Lin, J., N. Cheng, M. Chow, D. J. Filman, A. C. Steven, J. M. Hogle, and D. M. Belnap. 2011. An externalized polypeptide partitions between two distinct sites on genome-released poliovirus particles. J Virol 85:9974-9983. Butan, C., D. J. Filman, and J. M. Hogle. 2013. Cryo-electron microscopy reconstruction shows poliovirus 135S particles poised for membrane interaction and RNA release. J Virol. Gouet, P., E. Courcelle, D. I. Stuart, and F. Metoz. 1999. ESPript: analysis of multiple sequence alignments in PostScript. Bioinformatics 15:305-308.

34


706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742

Table 1: Data collection and refinement statistics Fig. 1 Purification of EV71 virions and characterization of the uncoating intermediate. (A) Purification of EV71 virions. The protein composition of the virion that was purified as described in the Materials and Methods was determined using 15% SDS-PAGE analysis. Lane 1: molecular weight marker; Lane 2: purified full virions. The calculated molecular weights of VP1, VP2, VP3 and VP4 are 32.6 kDa, 27.7 kDa, 26.4 kDa and 7.5 kDa, respectively. The purified full virions contain all four structural proteins VP1, VP2, VP3 and VP4. (B) Cryo-EM images of samples from the crystallization drops containing crystals of the uncoating intermediate (left) and purified naturally occurring empty particles (right). The internal density corresponded to the RNA genome in the uncoating intermediate, whereas the empty particle was completely devoid of internal density. The presence of the RNA genome inside the particle suggests that these are uncoating intermediates. (C) The uncoating intermediate is more protease sensitive in comparison with the full virion. The proteolytic sensitivities of the full virion and the uncoating intermediate were assessed by trypsin digestion as described in Materials and Methods. The digested samples were analyzed by western blot using a VP1 antibody. Lane 1: hanging drops containing crystals of the uncoating intermediate. Lane 2: hanging drops containing crystals of the uncoating intermediate digested with trypsin for 1h at 37°C. The truncated VP1 was indicated with the blue arrow head. Lane 3: hanging drops containing crystals of the full virion. Lane 4: hanging drops containing crystals of the full virion digested with trypsin for 1h at 37°C. VP1 in the full virion is resistant to trypsin digestion. (D) The VP4 in the uncoating intermediate is protected from α-chymotrypsin digestion. The uncoating intermediate was treated by α-chymotrypsin digestion as described in Materials and Methods. The digested samples were analyzed by western blot using a VP4 antibody, with VP2 as a loading control. Lane 1: hanging drops containing crystals of the uncoating intermediate were incubated at 25°C. Lane 2: hanging drops containing crystals of the uncoating intermediate digested with α-chymotrypsin for 1h at 25°C. The VP4 in the uncoating intermediate was resistant to protease digestion. Fig. 2 Overall structures of EV71 full virions and the uncoating intermediate. (Left) Radius-colored surface representation of the EV71 full virion viewed along the 2-fold axis. The surface is colored from blue to red according to the distance from the particle center (blue represents the closest). (Middle) Ribbon representations of the full virion (colored red) and the uncoating intermediate (colored blue). Only half of each capsid shell is represented, as a ~80 Å 35


743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784

785 786 787 788 789

slab, to illustrate the expansion of the uncoating intermediate with respect to the full virion. The position of the 2-fold axis of the particle is indicated. (Right) Radius-colored surface representation of the uncoating intermediate viewed along the 2-fold axis. The surface is colored as in the left panel.


36

Fig. 3 Structural changes in the protomer and individual capsid proteins during uncoating. (A) Structural comparison of the protomer in the full virion and the uncoating intermediate. Surface representation of the EV71 full virion viewed along the 2-fold axis with VP1, VP2 and VP3 colored magenta, yellow and cyan, respectively. Cartoon representations of the protomer with VP1, VP2 and VP3 in the uncoating intermediate are colored magenta, yellow and cyan, respectively, whereas their counterparts in the full virion are colored grey. The positions of the icosahedral symmetry elements are indicated. (B) Superposition of VP1. Residues 1-297 are modeled in the full virion and colored red, whereas residues 72-296 are modeled in the uncoating intermediate and colored blue. (C) Superposition of VP2. The proteins are colored as in B. Residues 9-254 and residues 16-47 and 54-250 are modeled in the full virion and the uncoating intermediate, respectively. (D) Superposition of VP3. The proteins are colored as in B. Residues 1-242 and residues 1-175 and 189-236 are modeled in the full virion and the uncoating intermediate, respectively. (E) The structure of VP4, which is colored as in B. Residues 12–69 of VP4 were modeled from well-defined electron density in the full virion. (F) Comparison of the VP1 pockets between the full virion (red, with the pocket factor shown in green) and the uncoating intermediate (blue). Cys225 near the pocket region is shown in yellow. Fig. 4 Structure-based sequence alignments of the capsid proteins VP1, VP2 and VP3 from different EV71 strains. Capsid protein sequences used for the alignment include the clinical EV71 C4 strain used in this study (4N53) and the sequences of two other EV71 strains (denoted as 3VBS and 4GMP) whose capsid proteins have been structurally determined. The secondary structure elements for the EV71 full virion and the uncoating intermediate are shown at the top and bottom of the sequence alignment, respectively. The residue numbers are those in the EV71 full virion. Conserved residues are shown in white with a red background. Helices and strands are labeled according to standard picornaviral nomenclature and are represented by coils and arrows, respectively. The blue triangles indicate the variable residues between 4N53 and 3VBS. The black triangles indicate the variable residues between 4N53 and 4GMP. The VP1 GH loop, VP3 GH loop and residues 48-53 of VP2 are boxed with blue rectangles and correspond to the disordered regions in the structure of the uncoating intermediate. This figure was produced using ESPript (50). Fig. 5 Structural changes at the protomer interface. (A) Top view of the 5-fold axis channel in the virion. The surfaces are colored from blue to red according to their distance from the particle center (blue represents the closest). Four mutations between the clinical C4 strain and 3VBS in the capsid proteins are exposed on the viral surface as indicated. The triangle is drawn around the quasi 3-fold axis (surrounded by VP1 and VP2 from one protomer and VP3 from a neighboring protomer). The variable residues between the clinical C4 strain and 37


790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831

3VBS that are exposed on the capsid surface are colored blue (Glu98 and Cys225 of VP1), cyan (Ser144 of VP2) and black (Ser93 of VP3). (B) Hydrogen bonding network around the 5-fold axis. The amino group in the side chain of Lys182 (colored black) interacts with Asp185 (colored red) of a neighboring VP1 through hydrogen bonds. This interaction network around the 5-fold axis channel is likely conserved among human enterovirus species A and D. (C) Structural changes at the junction during uncoating. The structure of the virion is shown on the left, whereas that of the uncoating intermediate is shown on the right. VP1, VP2 and VP3 are colored magenta, yellow and cyan, respectively. The GH loop of VP1 is colored green, residues 48-53 of VP2 are colored blue and the GH loop of VP3 is colored orange. During uncoating, conformational changes occur in the GH loop of VP1, whereas residues 48-53 in VP2 and the GH loop in VP3 become disordered. Fig. 6 Structural changes at the 2-fold axis channel. VP1, VP2 and VP3 are colored magenta, yellow and cyan, respectively. The 2-fold channel in the full virion is shown on the left, whereas the 2-fold channel in the uncoating intermediate is shown on the right. The 2-fold axis channel expands during uncoating. Fig. 7 Capsid rearrangements during uncoating. (A) The absolute distance shift for every Cα atom (x10) during EV71 uncoating was mapped onto the full virion structure. The surface is colored from red to blue according to the relative distance shift (blue represents the lowest shift). (B) The inner surface of the full virion is colored as in A, showing the internal rearrangements during uncoating. (C) The absolute distance shift for every Cα atom (x10) during HRV2 uncoating was mapped onto the full virion structure. The surface is colored from red to blue according to the relative distance shift (blue represents the lowest shift). Fig. 8 Capsid-RNA interactions in the EV71 uncoating intermediate. (A) EV71 A-particle showing density extending from the capsid shell that interacts with the viral RNA genome. The fitted uncoating intermediate crystal structure is depicted in a ribbon representation with VP1, VP2 and VP3 colored magenta, yellow and cyan, respectively. The A-particle cryo-EM density is depicted as a grey mesh. Residue 72 of VP1 is colored in green and residues 32-36 of VP3 are colored in orange, both of which interact with the inner RNA density. (B) The fitted uncoating intermediate crystal structure is depicted in a ribbon representation with VP1, VP2 and VP3 colored magenta, yellow and cyan, respectively. The A-particle cryo-EM density is depicted as a grey mesh. Residue 16 of VP2 is colored red and does not interact with the inner RNA density.

38


832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873

Fig. 9 Structural comparison between the EV71 uncoating intermediate and the CVA16 A-like particle. (A) Structural comparison of the protomer in the EV71 uncoating intermediate with that in the CVA16 135S-like particle. VP1, VP2, and VP3 in EV71 are colored magenta, yellow and cyan, respectively, and those in CVA16 are colored grey. Thr175 and Tyr189 in VP3 mark the beginning and end of the disordered regions in the determined uncoating intermediate structure. Ala47 and Thr54 in VP2 mark the beginning and end of the disordered region in the determined uncoating intermediate structure. (B) Superposition of VP1. Residues 72-296 of VP1 in EV71 (colored blue) and residues 62-210 and 219-297 in CVA16 (colored orange) are modeled. The ordered region in VP1 begins at Ser72 (colored blue) of the determined uncoating intermediate structure. The ordered region in VP1 of the CVA16 135S-like particle begins at Asn62 (colored orange). (C) Superposition of VP2. The proteins are colored as in B. Residues 16-47 and 54-250 in EV71 and residues 6-136 and 142-249 in CVA16 are modeled. Ala47 and Thr54 indicate the beginning and end of the disordered regions in EV71. Ala136 and Glu142 indicate the beginning and end of the disordered regions in CVA16. (D) Superposition of VP3. The proteins are colored as in B. Residues 1-175 and 189-236 in EV71 and residues 1-179 and 185-236 in CVA16 are modeled. Tyr185 and Ala179 (orange) indicate the beginning and end of the disordered regions in CVA16. (E) Side views showing different positions of the VP1 N-terminus in the EV71 uncoating intermediate (left) and the CVA16 135S-like particle (right). EV71 is colored blue, whereas CVA16 is colored red. The gray surface representation shows the surface of the capsid pentamer from the side view. A stretch of polypeptide was observed to traverse the capsid in CVA16, whereas this region is disordered in EV71. (F) Structures at the junction in CVA16. VP1, VP2 and VP3 are colored magenta, yellow and cyan, respectively. The GH loop in VP1 is colored green, and residues 142-146 in VP2 are colored blue. Portions of the GH loop of VP1 and the EF loop of VP2 (residues 137-141), both facing the junction channel, are disordered.

39


874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904

Table 1: Data collection and refinement statistics Full virion

Uncoating intermediate

I23

P4232

I/σI Completeness (%) Redundancy

591.375 49.84-3.30(3.38-3.30)a 314603/15490 21.6(70.1) 3.7(0.93) 62.2(46.2) 1.5(1.2)

352.416 49.84-3.80(3.87-3.80) 50101/2652 38.4(83.9) 3.3(1.3) 68.0(73.4) 5.0(4.6)

Refinement Resolution (Å) No. reflections

49.84-3.30 314499

49.84-3.80 50069

22.80/26.09 130720 50.47

24.54/27.41 26295 65.65

0.003 0.728

0.002 0.641

Data collection

Rmergeb(%)

Rwork/Rfree(%) No. atoms Average B-factors R.m.s. deviations Bond lengths (Å) Bond angles (°) a

Values in parentheses refer to the highest-resolution shell. 

n

b

R

merge



 I  hkl   I  hkl  i

hkl i1 n

 I  hkl  i

hkl i1

R

 F  hkl   F  hkl   F  hkl  obs

calc

hkl

obs

hkl

The R for the larger “working” set of reflections is referred to as Rwork.

R

free T





h,k,l T

F  h, k, l   k F   F  h, k, l  obs

h,k,l T

T: set of reflections.

calc

obs

(h, k, l)


Space group Cell dimensions a=b=c (Å) Resolution (Å) Unique reflections








