Dec 18, 2013 ... 31 therapeutics against HFMD. Here, we present the atomic .... 128 into the cryo-
EM reconstruction obtained by Shingler et al. revealed location of the. 129 ....
loop of VP2 and 203-209 in the HI loop of VP3 were considered.
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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.
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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.
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
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 i1 n
I hkl i
hkl i1
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)
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
Space group Cell dimensions a=b=c (Å) Resolution (Å) Unique reflections
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest
Downloaded from http://jvi.asm.org/ on December 14, 2017 by guest