1 Characterization of a putative ancestor of coxsackievirus B5 1 2 Maria Gullberg1, Conny Tolf1 ...

53 Pages·2010·5.03 MB·English

JVI Accepts, published online ahead of print on 14 July 2010
J. Virol. doi:10.1128/JVI.00071-10
Copyright © 2010, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.
1 Characterization of a putative ancestor of coxsackievirus B5
2
3 Maria Gullberg1, Conny Tolf1, Nina Jonsson1, Mick N. Mulders2, Carita Savolainen-Kopra3,
4 Tapani Hovi3, Marc Van Ranst4, Philippe Lemey4, Susan Hafenstein5 and A. Michael Lindberg1,*
5
D
o
6 1School of Natural Sciences, Linnaeus University, SE-391 82 Kalmar, Sweden. w
n
lo
7 2Centre for Infectious Disease Control Netherlands, National Institute for Public Health and the a
d
e
d
8 Environment (RIVM), Bilthoven, the Netherlands.
f
r
o
9 3Department of Infectious Disease Surveillance and Control, National Institute for Health and m

h
t
10 Welfare (THL), Helsinki, Finland. tp
:
/
/
11 4Department of Microbiology and Immunology, Rega Institute for Medical Research, Katholieke jvi.
a
s
m
12 Universiteit, Belgium.
.
o
r
13 5Department of Microbiology and Immunology, The Pennsylvania State University College of g
/

o
n
14 Medicine, 500 University Drive, Hershey, Pennsylvania 17033, USA.
J
a
n
15 u
a
r
y
16 *Corresponding author.
1
1
,
17 Mailing address: School of Natural Sciences, Linnaeus University, SE-391 82 Kalmar, Sweden. 2
0
1
18 Phone: [phone]. Fax: [phone]. E-mail: [email protected]. 9

b
y
19 g
u
e
20 Running title: CVB5 phylogenetics and ancestral reconstruction s
t
21
22 Abstract word count: 216
23 Text word count: 5880
1
24 ABSTRACT
25 Like other RNA viruses, coxsackievirus B5 (CVB5) exists as circulating heterogeneous
26 populations of genetic variants. In this study, we present the reconstruction and characterization
27 of a probable ancestral virion of CVB5. Phylogenetic analyses based on capsid protein encoding
28 regions (the VP1 gene of 41 clinical isolates and the entire P1 region of eight clinical isolates) of
D
o
w
29 CVB5 revealed two major co-circulating lineages. Ancestral capsid sequences were inferred from
n
lo
30 sequences of these contemporary CVB5 isolates using maximum likelihood methods. By using a
d
e
d
31 Bayesian phylodynamic analysis, the inferred VP1 ancestral sequence was dated back to 1854
f
r
o
m
32 (1807-1898). In order to study the properties of the putative ancestral capsid, the entire ancestral

h
t
t
33 P1 sequence was synthesized de novo and inserted into the replicative backbone of an infectious p
:
/
/
jv
34 CVB5 cDNA clone. Characterization of the recombinant virus in cell culture showed that fully i.
a
s
m
35 functional infectious virus particles were assembled and that these viruses displayed properties
.
o
r
36 similar to those of modern isolates, in terms of receptor preferences, plaque phenotype, growth g
/

o
n
37 characteristics and cell tropism. This is the first report describing resurrection and
J
a
n
38 characterization of a picornavirus with a putative ancestral capsid. Our approach, including u
a
r
y
39 phylogenetics-based reconstruction of viral predecessors, could serve as a starting point for
1
1
,
40 experimental studies of viral evolution and might also provide an alternative strategy in the 2
0
1
41 development of vaccines. 9

b
y
42 g
u
e
43 Keywords: coxsackievirus; phylogeny; ancestral reconstruction; receptor usage. s
t
2
44 INTRODUCTION
45 The group B coxsackieviruses (CVB, serotype 1-6) were discovered in the 1950s in a
46 search for new poliovirus-like viruses (33, 61). Infections caused by CVBs are often
47 asymptomatic, but may occasionally result in severe diseases of the heart, pancreas and central
48 nervous system (99). CVBs are small icosahedral RNA viruses belonging to the Human
D
o
w
49 Enterovirus B (HEV-B) species within the family Picornaviridae (89). In the positive single-
n
lo
50 stranded RNA genome, the capsid proteins VP1-VP4 are encoded within the P1 region, whereas a
d
e
d
51 the non-structural proteins required for virus replication are encoded within the P2 and P3 region
f
r
o
m
52 (4). The 30 nm capsid has an icosahedral symmetry and consists of 60 copies of each of the four

h
t
t
53 structural proteins. The VP1, VP2 and VP3 proteins are surface exposed, whereas the VP4 p
:
/
/
jv
54 protein lines the interior of the virus particle (82). The coxsackie- and adenovirus receptor i.
a
s
m
55 (CAR), a cell adhesion molecule of the immunoglobulin superfamily, serves as the major cell
.
o
r
56 surface attachment molecule for all six serotypes of CVB (5, 6, 39, 60, 98). Some strains of g
/

o
n
57 CVB1, 3 and 5 also interact with the decay accelerating factor (DAF; CD55), a member of the
J
a
n
58 family of proteins that regulate the complement cascade. However, attachment of CVBs to DAF u
a
r
y
59 alone does not permit infection of cells (6, 7, 59, 85).
1
1
,
60 Picornaviruses exist as genetically highly diverse populations within their hosts, referred to 2
0
1
61 as quasispecies (20, 57). This genetic plasticity enables these viruses to adapt rapidly to new 9

b
y
62 environments, but at the same time, it may compromise structural integrity and enzymatic g
u
e
63 functionality of the virus. The selective constraints imposed on the picornavirus genome are s
t
64 reflected in the different regions used for different types of evolutionary studies. The highly
65 conserved RNA-dependent RNA polymerase (3Dpol) gene is used to establish phylogenetic
66 relationship between more distantly related viruses (e.g. viruses belonging to different genera)
3
67 (38), whereas the variable genomic sequence encoding the VP1 protein is used for classification
68 of serotypes (13, 14, 69, 71, 72).
69 In 1963, Pauling and Zuckerkandl proposed that comparative analysis of contemporary
70 protein sequences can be used to predict the sequences of their ancient predecessors (73).
71 Experimental reconstruction of ancestral character states has been applied in evolutionary studies
D
o
w
72 of several different proteins, e.g. galectins (49), G protein-coupled receptors (52), alcohol
n
lo
73 dehydrogenases (95), rhodopsins (15), ribonucleases (46, 88, 110), elongation factors (32), a
d
e
d
74 steroid receptors (10, 96, 97) and transposons (1, 45, 87). In the field of virology, reconstructed
f
r
o
m
75 ancestral or consensus protein sequences have been used in attempts to develop vaccine

h
t
t
76 candidates for human immunodeficiency virus type 1 (21, 51, 66, 81), but rarely to examine p
:
/
/
jv
77 general phenotypic properties. i.
a
s
m
78 In this study, a CVB5 virus with a probable ancestral virion (CVB5-P1anc) was constructed
.
o
r
79 and characterized. We first analyzed in detail the evolutionary relationship between structural g
/

o
n
80 genes of modern CVB5 isolates and inferred a time scale for their evolutionary history. An
J
a
n
81 ancestral virion sequence was subsequently inferred using a maximum likelihood (ML) method. u
a
r
y
82 This sequence was then synthesized de novo, cloned into a replicative backbone of an infectious
1
1
,
83 CVB5 cDNA clone and transfected into HeLa cells. The hypothetical CVB5-P1anc assembled 2
0
1
84 into functional virus particles that displayed phenotypic properties similar to those of 9

b
y
85 contemporary clinical isolates. This is the first report describing the reconstruction and g
u
e
86 characterization of a fully functional picornavirus with a putative ancestral capsid. s
t
4
87 MATERIALS AND METHODS
88 Cell lines and viruses. African green monkey kidney (GMK), human colon
89 adenocarcinoma (HT29), Chinese hamster ovary (CHO), human lung carcinoma (A549) and
90 human rhabdomyosarcoma (RD) cell lines were purchased from American Tissue Culture
91 Collection. The HeLa Ohio cells were kindly provided by M. Roivainen (Helsinki, Finland).
D
o
w
92 Recombinant CHO cells expressing CAR (CHO-CAR) or DAF (CHO-DAF) were constructed by
n
lo
93 H.-C. Selinka (77, 84). Cells were propagated in Dulbecco´s Modified Eagles Medium (DMEM) a
d
e
d
94 supplemented with 2 mM L-glutamine, 100 U/ml penicillin, 0.1 mg/ml streptomycin and 10%
f
r
o
m
95 newborn calf serum (NCS). Recombinant CHO cells were grown in selective media

h
t
t
96 supplemented with 1 mg/ml G418 (Sigma) for cells expressing CAR and 0.75 mg/ml p
:
/
/
jv
97 Hygromycine B (Invitrogen) for cells expressing DAF. i.
a
s
m
98 The clinical CVB5 isolates used in this study were propagated in GMK cells according to
.
o
r
99 standard procedures (63). The CVB5 strain 1954UK85 (CVB5UK) (109) was provided by J.W. g
/

o
n
100 McCauley, Newbury, UK, and the Dalldorf strain (CVB5D) by R.L. Crowell, Philadelphia, USA
J
a
n
101 (16, 79). As previously described (47), the genome sequence of the CVB5D virus is identical to u
a
r
y
102 the prototype strain Faulkner (CVB5F) (55), except for one amino acid change in the VP1
1
1
,
103 protein. Viral titers of propagated viruses were determined by the tissue culture infectious dose 2
0
1
104 50% (TCID ) method according to standard procedures (41). 9
50
b
y
105 g
u
e
106 Flow cytometry analysis. Flow cytometry analysis was performed as described previously s
t
107 (77). Briefly, CHO, CHO-CAR, CHO-DAF and HeLa cells were stained with an anti-CAR
108 (RmcB) antibody (43) (hybridoma was kindly provided by L. Philipson and R. Pettersson,
109 Karolinska Institute, Sweden: also available from American Tissue Culture Collection: CRL-
110 2379), an anti-DAF (BRIC110) antibody (Cymbus Biotechnologies) or a mouse IgG1 control
5
111 antibody (X0931, Dako). After 1 h incubation at 4°C, cells were washed and stained with a
112 secondary R-phycoerythrin-labeled rabbit anti-mouse antibody (R0439, Dako). Data were
113 acquired using a FACSCalibur™ (Becton Dickinson) and analyzed with CellQuest™ version 3.3
114 software (Becton Dickinson).
115
D
o
w
116 Extraction, amplification and sequencing. Viral RNA was extracted from infected cell
n
lo
117 cultures (QIAamp viral RNA mini kit, Qiagen), reverse transcribed (Superscript III, Invitrogen) a
d
e
d
118 and PCR amplified (PicoMaxx, Stratagene) using virus-specific primers. PCR amplicons were
f
r
o
m
119 visualized in agarose gels and purified (QIAquick gel extraction kit, Qiagen). The nucleotide

h
t
t
120 sequences were determined with an ABI Prism 3130 automated sequencer (Applied Biosystems) p
:
/
/
jv
121 by a primer walking strategy on both strands using the BigDye chemistry (ABI Prism BigDye i.
a
s
m
122 Terminator Cycle Sequencing Ready Reaction kit, version 1.1; Applied Biosystems). Sequences
.
o
r
123 were analyzed using the Sequencher version 4.6 software package (Gene Codes Corporation). g
/

o
n
124
J
a
n
125 Phylogenetic analysis. Viral nucleotide sequences were aligned using ClustalW (94) and u
a
r
y
126 phylogenetic signals were evaluated using likelihood mapping (90). The presence of nucleotide
1
1
,
127 substitution saturation was tested using the approach of Xia et al. (2003) (104). Phylogenetic 2
0
1
128 relationships between the different CVB5 strains, based on the genomic VP1 and P1 regions, 9

b
y
129 were inferred by the ML method as implemented in PhyML (35). Branch support values for g
u
e
130 inferred phylogeny were estimated by non-parametric bootstrapping consisting of 1000 pseudo- s
t
131 replicates (30). The General Time Reversible (GTR) substitution model (53) with a gamma-
132 distributed rate heterogeneity was used for the VP1 analysis, while the same model including the
133 proportion of invariable sites was used for the P1 sequence data. The phylogenetic relationship
134 between viruses was also examined by the neighbor-joining method as implemented in MEGA
6
135 version 4.0 (92). Previously determined CVB5 sequences (CVB5UK, GenBank accession
136 number X67706 and CVB5D (47)) and swine vesicular disease virus (SVDV) sequences
137 (Svdh3jap76, D00435; Svdj1jap73, D16364; Svd27uk72, X54521; Svd1spa93, AF039166 and
138 Svd1net92, AF268065) were included in the phylogenetic analysis of the VP1 gene. The
139 sequences of CVB4 Tuscany (CVB4T, DQ480420) and CVB6 Schmitt (CVB6S, AF114384)
D
o
w
140 were used as an outgroup. Phylogenetic trees were visualized with MEGA 4.0. Root-to-tip
n
lo
141 divergence as a function of sampling time was examined using Path-O-Gen (available at a
d
e
d
142 http://tree.bio.ed.ac.uk/software).
f
r
o
m
143

h
t
t
144 Bayesian evolutionary analysis. We inferred the time scale and tempo of CVB5 evolution p
:
/
/
jv
145 using a Bayesian statistical approach implemented in BEAST (25). This approach employs a full i.
a
s
m
146 probabilistic model of sequence evolution along rooted, time-measured phylogenies with a
.
o
r
147 coalescent prior, using either fixed or relaxed molecular clock model (23, 24). For rapidly g
/

o
n
148 evolving viruses, the molecular clock is calibrated based on divergence accumulation between
J
a
n
149 sequences sampled at different points in time. We used the SRD06 model of nucleotide u
a
r
y
150 substitution (86) with gamma-distributed rate variation among sites, an uncorrelated lognormal
1
1
,
151 relaxed clock model and a Bayesian skyline tree-prior (26). Markov chain Monte-Carlo analyses 2
0
1
152 were run for 10 million generations and diagnosed using Tracer (http://beast.bio.ed.ac.uk/Tracer). 9

b
y
153 The evolutionary history was summarized in the form of a maximum clade credibility tree using g
u
e
154 TreeAnnotator (http://beast.bio.ed.ac.uk/TreeAnnotator) and visualized in FigTree s
t
155 (http://tree.bio.ed.ac.uk/software/figtree). Bayesian credible intervals for continuous parameters
156 are reported as highest posterior density intervals, which are the smallest intervals that contain
157 95% of the posterior distribution.
7
158 Ancestral sequence reconstruction. The ancestral sequences were reconstructed from the
159 CVB5 ingroup taxa of the phylogenetic trees based on the genetic VP1 and P1 regions, but
160 without the reference strains (CVB5D and CVB5UK). These reference strains were not included
161 in the reconstruction because their passage history is unknown. ML ancestral sequences were
162 inferred using a standard codon substitution model (M0, which assumes a homogenous
D
o
w
163 nonsynonymous/synonymous substitution rate among sites and among lineages) as implemented
n
lo
164 in codeml of the PAML package (34, 105). After ML optimization under the codon model, this a
d
e
d
165 approach considers the assignment of a set of characters to all interior nodes at a site as a
f
r
o
m
166 reconstruction and selects the reconstruction that has the highest posterior probability, i.e. so-

h
t
t
167 called joint reconstruction (106). This procedure was efficiently performed using the algorithm p
:
/
/
jv
168 described by Pupko and colleagues (78). A corresponding reconstruction was also performed i.
a
s
m
169 using the best fitting empirical amino acid model. The inference of the ancestor sequence was
.
o
r
170 facilitated by the absence of gaps or evidence of recombination within the genomic P1 region as g
/

o
n
171 confirmed using the Phi-test (12).
J
a
n
172 u
a
r
y
173 Construction of CVB5-P1anc. The complete CVB5D genome was amplified and cloned
1
1
,
174 into the pCR-Script Direct SK(+) vector (Stratagene), by using the AscI and NotI restriction 2
0
1
175 enzyme cleavage sites, as previously described (Fig. 1) (56). In this infectious full-length cDNA 9

b
y
176 clone of CVB5D (pCVB5Dwt), a ClaI site was introduced at nucleotide position 3340 to generate g
u
e
177 a cassette vector (pCVB5D-cas). This modification resulted in one amino acid change in the 2A s
t
178 protein (valine to leucine at amino acid position 17). This substitution was accepted as the
179 leucine-17 is present in the 2A protein of other enteroviruses, including echovirus 30 and
180 echovirus 21. In addition, a synonymous mutation was introduced in the P1 ancestral sequence to
181 remove a SalI site. In order to construct a CVB5 clone with the inferred ancestral capsid sequence
8
182 (pCVB5-P1anc), a pUC57 plasmid containing the ancestral P1 sequence flanked by SalI and ClaI
183 sites was purchased (GenScript). Subsequently, the P1 genomic region of this pUC57 plasmid
184 was amplified and then cloned into pCVB5D-cas. The constructs were propagated in Escherichia
185 coli DH5α and purified (Midiprep kit, Promega). The nucleotide sequences of all constructs were
186 verified by sequencing as described above. D
o
w
187 HeLa cell monolayers were transfected with 2.5 µg of the CVB5 cDNA clones by using n
lo
a
188 Lipofectamine 2000 (Invitrogen) according to the manufacturer´s protocol. Recombinant viruses d
e
d
189 were collected day five post transfection and subsequently sequenced to confirm identity. The f
r
o
m
190 titers of these viruses were determined by TCID assay in HeLa cells. In order to ensure the
50
h
t
t
191 safety of laboratory workers, environment and public, the generation of CVB5-P1anc was p
:
/
/
jv
192 performed under biosafety level 2 containment. i.
a
s
m
193
.
o
r
g
194 Virus infection. Cell monolayers were infected with virus according to standard /

o
n
195 procedures (63). Briefly, subconfluent monolayers of cells grown in 25 cm2 flasks were J
a
n
196 inoculated with viruses at a multiplicity of infection (MOI) of 10 TCID /cell. Following virus u
50 a
r
y
197 adsorption at room temperature for 1 h, the cells were washed three times before addition of 1
1
,

198 DMEM supplemented with 2 mM L-glutamine, 100 U/ml penicillin and 0.1 mg/ml streptomycin. 2
0
1
9
199 Incubation of the virus infection at 37°C in a 7.5% CO atmosphere was continued for five days
2 b
y

200 or until cytopathic effect (CPE) was observed. g
u
e
s
201 Viral replication was quantified by analyses of samples taken immediately after infection t
202 and five days post infection (p.i.) or when a complete CPE was observed. After three cycles of
203 freezing and thawing, the viral titers were determined by TCID assay in HeLa cells as described
50
204 above.
9
205 The molecular evolution of CVB5-P1anc, 151rom70, 4378fin88 and CVB5Dwt was
206 compared after ten serial rounds of infection in GMK, HeLa and RD cells. After the tenth
207 passage, the P1 region of these viruses were sequenced as described above.
208
209 Immunofluorescence. Infected HeLa cells cultured on Lab-TEK II chamber glass slides
D
o
w
210 (Nalge Nunc International) were fixed in 4% formaldehyde for 30 min at 4°C and stained for 1 h
n
lo
211 at room temperature with an enterovirus-specific polyclonal rabbit antiserum (KTL-482) (42). a
d
e
d
212 The primary antibody was visualized with a secondary goat anti-rabbit antibody labeled with
f
r
o
213 Alexa Fluor® 488 (A11034, Molecular probes Inc.). Finally, slides were mounted with m

h
t
t
214 Vectashield (Immunkemi) containing 4',6-diamidino-2-phenylindole (DAPI) and images captured p
:
/
/
jv
215 with an epifluorescence microscope. i.
a
s
m
216
.
o
r
217 Plaque formation assay. A semi-solid gum tragacanth medium was used as previously g
/

o
n
218 described (18) to assess the plaque morphology of viruses. Briefly, confluent monolayers of
J
a
n
219 HeLa cells in six-well plates were incubated with 1 ml virus in 10-fold dilutions for 1 h at 37°C. u
a
r
y
220 Following adsorption, the virus inoculum was aspired and cells were overlaid with DMEM
1
1
,
221 supplemented with 1% NCS, 2 mM L-glutamine, 100 U/ml penicillin, 0.1 mg/ml streptomycin 2
0
1
222 and 0.8% (w/v) gum tragacanth (Sigma). The plaques were visualized by staining cells with a 9

b
y
223 crystal violet-ethanol solution after 48 h incubation at 37°C. g
u
e
224 s
t
225 Viral growth kinetics. To assess the growth kinetics of viruses, HeLa cells in 24-well
226 plates were infected with virus at MOI 10 TCID /cell as described above. Cells and medium
50
227 were harvested and frozen at various time points post infection. Virus titers in collected samples
228 were determined by TCID assay.
50
10