Identification of a novel cosavirus species in faeces of children and its relationship with acute gastroenteritis in China

Accepted Manuscript Identification of a novel cosavirus species in feces of children and its relationship with acute gastroenteritis in China Jie-mei Yu, Yuan-yun Ao, Li-li Li, Zhao-jun Duan PII:

S1198-743X(17)30110-6

DOI:

10.1016/j.cmi.2017.02.018

Reference:

CMI 866

To appear in:

Clinical Microbiology and Infection

Received Date: 1 July 2016 Revised Date:

8 February 2017

Accepted Date: 14 February 2017

Please cite this article as: Yu J-m, Ao Y-y, Li L-l, Duan Z-j, Identification of a novel cosavirus species in feces of children and its relationship with acute gastroenteritis in China, Clinical Microbiology and Infection (2017), doi: 10.1016/j.cmi.2017.02.018. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Identification of a novel cosavirus species in feces of children and its relationship with

2

acute gastroenteritis in China ⊥

4

⊥

Jie-mei Yu , Yuan-yun Ao , Li-li Li, Zhao-jun Duan*

3

Institute for Viral Diseases Control and Prevention, China CDC, Beijing 100052, China

6

*Corresponding author: Zhao-jun Duan E-mail: [email protected]

8

Telephone number: 86-10-63552910

9

Fax number: 86-10-63557757

M AN U

10 11

⊥

These two authors contributed equally to the manuscript.

12

Key words: novel Cosavirus; diarrhea; children; case-control study

14

TE D

13

Running title:

16

Association of a novel cosavirus with diarrhea

19 20 21 22 23 24 25

AC C

18

EP

15

17

SC

7

RI PT

5

ACCEPTED MANUSCRIPT Abstract

27

Objectives: To assess the prevalence of human cosavirus (HCosV) in China and to

28

determine the association of a novel HCosV (Cosa-CHN) with acute gastroenteritis (AGE) .

29

Methods: A case-control study with 461 paired stool samples from diarrhea and healthy

30

children was conducted. Real-time PCR and nested PCR were used to detect the HCosVs.

31

Rapid-amplification of cDNA ends was employed to obtain the ends of the Cosa-CHN.

32

Results: Known HCosVs were detected in two control samples, while Cosa-CHN was

33

detected in eight (1.7%) and six (1.3%) of the case and control samples, respectively. The

34

complete genome of Cosa-CHN comprises 7213 bp. The P1 and P2 regions of the

35

Cosa-CHN were closely related to those of HCosV B, while the P3 region was most similar

36

to that of HCosV D, albeit with low amino acid (aa) identities (66% and 67%, respectively).

37

Phylogenetic analyses of the polyprotein and partial VP3/VP1 regions indicated that

38

Cosa-CHN could be classified as a novel species (tentatively named HCosV G) in

39

cosavirus. There was no significant difference in detection rate (p = 0.59) or mean viral

40

load (p = 0.43) of Cosa-CHN between the cases and controls. Statistical analysis revealed

41

no association between Cosa-CHN and AGE (p = 0.76), and the virus did not exacerbate

42

clinical symptoms.

43

Conclusions: A low prevalence of HCosV was detected, but a novel cosavirus species was

44

found in children with and without gastroenteritis in this study, and the evidence did not

45

support a causative role for the novel virus in pediatric AGE.

EP

TE D

M AN U

SC

RI PT

26

AC C

46 47

Introduction

48

Picornaviridae is a large family of viruses with a naked capsid surrounding a core of

49

single-stranded, positive-sense genomic RNA encoding a single polyprotein. With the

50

advent of high-throughput sequencing, an increasing number of novel picornaviruses have

51

been described. As of July 2013, the family Picornaviridae comprised 26 genera

52

(http://www.ictvonline.org/virusTaxonomy.asp), and included several leading pathogens

53

that affect human and animal health; e.g., polioviruses and hand-foot-and-mouth-disease

54

related enteroviruses.

55

Human cosavirus (HCosV) is a novel virus that was first detected in fecal samples of

ACCEPTED MANUSCRIPT both healthy children and those with nonpolio acute flaccid paralysis in Pakistan in 2008

57

[1]. Subsequently, the virus was found in feces from diverse populations and in sewage

58

water worldwide [2,3,4,5,6,7,8]. The closest relatives of HCosV among the picornaviruses

59

are cardioviruses and senecavirus. The International Committee on the Taxonomy of

60

Viruses (ICTV) states that the genus cosavirus comprises a single type species (species A)

61

(http://www.ictvonline.org/virusTaxonomy.asp). However, five additional species (B to F)

62

of at least 30 genotypes have been reported [1,2,8,9].

RI PT

56

A recent study from Japan suggested HCosV to be a causal pathogen in pediatric

64

patients because of the absence of other common diarrhea-causing viruses [8]. However,

65

according to a previous study, the prevalence of HCosV varies geographically from 0.1% to

66

48%. Moreover, because of frequent viral co-infection and low HCosV loads, the human

67

pathogenicity of cosavirus is controversial [5]. Furthermore, the marked genetic diversity of

68

HCosV complicates disease-association studies, as different viral types are likely associated

69

with different symptoms. Here, we conducted a strict 1:1 paired case-control study to

70

investigate the prevalence of cosavirus in Chinese children and its relationship with acute

71

gastroenteritis (AGE). A novel human cosavirus species (Cosa-CHN) was found in this

72

study to be the most prevalent cosavirus.

73

Materials and Methods

74

Samples

75

Two hundred thirty-nine and two hundred twenty-two pairs of fecal samples from children

76

less than 5 years of age were collected in Lulong, Hebei Province, and Liuyang, Hunan

77

Province, respectively, from 2011 to 2013, as described in our recent study [10]. Briefly,

78

cases were individuals diagnosed by a pediatrician with diarrhea without other diagnosed

79

illness, such as pneumonia. Control subjects were healthy children who did not have

80

diarrhea, fever, vomiting, or a respiratory illness during the previous 1 week. Samples from

81

the controls were collected by village doctor from the children's home, and the maximum

82

time interval between matched case and control sampling was 14 days. The healthy and

83

diarrhea subjects were paired according to age (age groups: 0-5, 6-11, 12-23, 24-35, 36-47

84

and 48-59 months), sex, and area.

85

AC C

EP

TE D

M AN U

SC

63

Only one specimen was obtained per patient. All stool specimens were collected

ACCEPTED MANUSCRIPT within 24 h of hospitalization and stored at –70°C until further analysis. Informed consent

87

was obtained from the parents of all children who provided specimens. The study protocol

88

was approved by a meeting of the Ethics Committee of the National Institute for Viral

89

Disease Control and Prevention, China Center for Disease Control, according to Chinese

90

ethics laws and regulations.

91

Detection of Common Diarrhea-Related Viruses

92

Viral nucleotides were extracted from 10% fecal suspensions in phosphate-buffered saline

93

using the QIAamp Viral RNA MiniKit (Qiagen), and first-strand cDNAs were synthesized

94

using the Superscript II reverse transcriptase (Invitrogen) with random primers. All

95

specimens were tested for rotavirus (RV), human calicivirus (HucV) (norovirus and

96

sapovirus), astrovirus (AstV), and adenovirus (AdV) by real-time PCR using primers,

97

probes, and reaction conditions reported previously [11,12,13,14]. The positivity rates of

98

RV, HuCV, AdV, and AstV in the case and control groups were calculated as described in

99

our previous study [10].

M AN U

SC

RI PT

86

Human Cosavirus Detection

101

HCosV was detected by real-time PCR using primers and conditions described previously

102

[5]. Positive samples were confirmed and genotyped by nested PCR amplification of a

103

904-bp fragment of the viral capsid gene using primers described elsewhere [9]. The

104

products were purified using a QIAquick PCR Purification Kit (Qiagen) and sequenced.

105

One sequence with low homology (aa identity <65%) to previously reported HCosV was

106

detected. The complete genome of this novel virus was amplified. The three sequences

107

were submitted to GenBank under accession numbers KP213320–KP213322.

108

Complete Genomic Amplification

109

To determine the complete genome sequence of Cosa-CHN, we designed primers based on

110

the sequence amplified by nested PCR. Further synthesis was based on the newly amplified

111

sequences. PCR amplifications were performed using a Genome Walking Kit (TaKaRa,

112

Japan) and the extreme 5′ and 3′ ends of the genome were determined using a SMART

113

RACE cDNA Amplification Kit (Clontech) following the manufacturer’s instructions.

114

Sequences were assembled and edited manually to produce the final sequence of the viral

115

genome. Overlapping long fragments (1228–2989 bp) were amplified for final confirmation

AC C

EP

TE D

100

ACCEPTED MANUSCRIPT 116

using ExTaq polymerase.

117

Novel Cosavirus (Cosa-CHN) Detection

119

Samples were screened for the novel cosavirus (Cosa-CHN) by nested PCR using upstream

120

primers targeted to VP3 and downstream primers targeted to VP1. Primers for the first PCR

121

round were CosF1 and CosR1 (supplemental table), and amplification was performed under

122

standard PCR conditions with an initial five cycles of annealing at 56°C followed by 15

123

cycles of annealing at 53°C. The second PCR round used the primers CosF2 and CosR2

124

(supplemental table) under standard PCR conditions with an initial five cycles of annealing

125

at 56°C followed by 35 cycles of annealing at 53°C. Positive and negative controls were

126

included.

M AN U

SC

RI PT

118

Cosa-CHN viral loads in positive samples were quantified by real-time PCR. A

128

forward primer (CosutrF), a reverse primer (CosutrR), and a probe (CosutrProbe) were

129

designed to target the 5′ untranslated region (UTR) of the virus (supplemental table). The in

130

vitro transcribed viral RNA was used as positive control. The reaction conditions were as

131

follows: 50°C for 30 min, 95°C for 15 min, followed by 45 cycles of 95°C for 15 s and

132

60°C for 1 min. After completion of amplification, fluorescence intensity was measured.

133

Sequence and Phylogenetic Analysis

134

PCR products were purified using a QIAquick PCR Purification Kit (Qiagen) and then sent

135

for sequencing. Sequences were determined and analyzed using the DNAStar software

136

package. The SimPlot software (version 3.5.1) was used to align and compare sequences to

137

identify potential recombination of Cosa-CHN with other known cosavirus species.

138

Phylogenetic analysis was performed using nucleotide sequences by the neighbor-joining

139

method and subsequently subjected to bootstrap analysis with 1000 replicates to determine

140

the reliability values at each internal node. Trees were produced using the MEGA software

141

(version 5).

142

Statistical Analysis

143

The statistical significance of differences in mean viral loads between the groups was

144

assessed using Student’s t-test, and that of differences in frequency among areas was

145

evaluated using the Fisher’s exact test. Differences in clinical symptoms between groups

AC C

EP

TE D

127

ACCEPTED MANUSCRIPT 146

were assessed by multiple factor logistic regression. P<0.05 was considered statistically

147

significant. All statistical analyses were performed using SPSS 16.0.

148

Results

150

Detection of HCosV

151

The results of screening real-time PCR showed that of the 461 pairs of samples, 3 (sample

152

IDs: 67c, 87c, and 144) were positive for cosavirus, with viral loads of 1.07 × 104 copies/ml,

153

4.94 × 102 copies/ml, and 2.98 × 104 copies/ml, respectively. Nested PCR and sequencing

154

analysis suggested that samples 67c and 87c had the highest similarity to HCosV A17

155

(Country: Nigeria), with 94 and 95% aa identities, respectively, while sample 144 was most

156

similar to HCosV B1(Country: Pakistan), albeit with only 63% aa identity.

157

Complete Genome of Cosa-CHN

158

The low aa identity to HCosV B1 suggested the presence of a novel cosavirus (named

159

Cosa-CHN), the complete genome of which was sequenced. A total of nine fragments were

160

amplified to determine the complete genome sequence (Figure 1). The full-length genome

161

sequence was deposited in GenBank under the accession number KM516909

162

(NC_025961.1). The identified Cosa-CHN genome comprised 7213 bp (excluding the

163

polyadenylated tract), with a 759-bp 5′ UTR, an open reading frame of 6360 nt (encoding a

164

potential polyprotein precursor of 2120 aa), followed by a 94-bp 3′ UTR and poly(A) tail.

165

The base usage of the Cosa-CHN genome was 28.9% A, 21.5% C, 20.7% G, and 28.9% U,

166

with a relatively low G+C content (42.2%), which is similar to previously reported HCosVs

167

[1]. A methionine start codon at nucleotide position 760 was found in the standard Kozak

168

context (RNNAUGG was AAUAUGG). A hypothetical cleavage map of the Cosa-CHN

169

polyprotein was derived from alignments with other HCosVs (Figure 1).

SC

M AN U

TE D

EP

AC C

170

RI PT

149

The sequence of the P1 and P2 regions of Cosa-CHN exhibited 66% aa identity with

171

those of its closest relative, HCosV B1. However, the P3 region of Cosa-CHN was most

172

similar to that of HCosV D, with 67% aa identity. The 2C and 3CD regions of Cosa-CHN

173

showed 70 and 67% aa identities with those of HCosV B1 and D, respectively. The ICTV

174

states that enteroviruses in the picornaviridae sharing 70% aa identity in P1 and >70% aa

175

identity in the 2C and 3CD regions belong to the same species [15]. Therefore, Cosa-CHN

ACCEPTED MANUSCRIPT 176

is proposed as a novel cosavirus species.

177

Screening and Viral Loads of Cosa-CHN

179

By using the Cosa-CHN-specific detection assay, a total of 461 pairs of samples were

180

screened using nested PCR of the VP3/VP1 region of Cosa-CHN, together with sample ID

181

144, and six (1.3%) controls (sample 67c and 87c that positive for HCosV were not positive

182

for Cosa-CHN) and seven (1.7%) cases were positive for Cosa-CHN. These 13 sequences

183

were submitted to GenBank under accession numbers KP213307–KP213319. All 14

184

sequences of Cosa-CHN (including that of ID 144) with 256 bp spanning the VP3/VP1

185

region showed 99% nt identities. The viruses were detected year-round; their distribution

186

did not differ seasonally. Of the 14 Cosa-CHN-positive subjects, 3 were from Liuyang

187

(South China) and 11 were from Lulong (North China). The age range of

188

Cosa-CHN–positive case subjects was 3–15 months (median = 6.5 months), while that of

189

Cosa-CHN–positive control subjects was 10–36 months (median = 11.5 months). A χ2- test

190

revealed that detection rates differed significantly according to area (p = 0.005), but not

191

between the case and control groups (p = 0.59). The detection rate of Cosa-CHN in

192

Liuyang (southern China) and Lulong (northern China) Provinces differed significantly

193

(4.5% vs. 1.7%).

TE D

M AN U

SC

RI PT

178

The mean viral load of the case group was 1.75 × 103 copies/mL (3.48 × 101–1.07 ×

195

104 copies/mL) vs. 3.36 × 102 copies/mL (1.04 × 102–1.09 × 103 copies/mL) in the control

196

group (Table); a log-normal t-test revealed no difference between the cases and controls (p

197

= 0.43).

198

Recombination and Phylogenetic Analysis of Cosa-CHN

199

Recombination analysis showed no putative inter-species recombination in Cosa-CHN. A

200

phylogenetic analysis using the nucleotide sequences of the VP3–VP1 region (Figure 2a)

201

and complete polyprotein amino acid sequences (Figure 2b) of Cosa-CHN and other known

202

HCosVs and other representative picornaviruses using the neighbor-joining method and

203

1,000 bootstrap replications suggested that Cosa-CHN forms a distinct lineage. The two

204

phylogenetic trees also showed that Cosa-CHN is more closely related to HCosV B than to

205

other previously reported HCosVs.

AC C

EP

194

ACCEPTED MANUSCRIPT 206

Association between Cosa-CHN and AGE Of the eight Cosa-CHN-positive children in the case group, five were co-infected with

208

diarrhea-related viruses (Table). Infection with Cosa-CHN did not exacerbate the clinical

209

symptoms (i.e., proportion of fever, frequency of diarrhea, and duration of diarrhea) of

210

these children (data not shown).

RI PT

207

211

Discussion

213

Cosavirus is a newly established genus in the family Picornaviridae that comprises six

214

genetically distinct species (A–F) [9]. Following its discovery, HCosVs were reported in

215

diverse sample types, including human feces, sewage, and porcine feces, in several

216

countries, such as China, Italy, Thailand, Bolivia, Japan, Tunisia, and Brazil

217

[1,2,3,4,5,6,8,16,17,18,19,20]. However, the prevalence ranges from 0 to 71% depending

218

on the area and population. In this study, we used the real-time PCR method reported in a

219

Brazilian study [5]. Only three samples from healthy children were positive for HCosVs,

220

one of which exhibited the highest similarity to HCosV B, albeit with a low sequence

221

identity. The complete genome of this virus was amplified, and the sequence suggested it to

222

be a novel species of cosavirus, tentatively named Cosa-CHN.

M AN U

TE D

223

SC

212

The genome structure of Cosa-CHN was similar to that of other HCosVs, comprising 5' UTR–structural proteins (VP4, VP2, VP3, and VP1)–non-structural proteins (2A–2C and

225

3A–3D)–3' UTR–poly (A) tail. The length of 5'UTR of Cosa-CHN (759bp) is similar to

226

that of those previously published HCosVs (HCosV B,FJ438907, 749bp; HCosV D,

227

FJ438908, 747bp), it is estimated that the complete sequence of 5'UTR of Cosa-CHN. The

228

cleavage sites in structural proteins, but not non-structural proteins, differ markedly among

229

HCosVs; however, the putative cleavage sites for Cosa-CHN and HCosV-B were identical,

230

with the exception of one difference in VP4 and VP1 (A/D and A/S, respectively). The

231

common motifs in the non-structural proteins of picornaviruses, such as the NTPase and

232

helicase motifs, were also found in Cosa-CHN. However, unlike many picornaviral

233

genomes, which contain a leader (L) protein, the genome of HCosV has no L protein at the

234

5′ end, similar to hepataviruses, enteroviruses, and hepatoviruses [21]. Moreover,

235

Cosa-CHN lacked the VP1 RGD motif, which is present in other HCosVs.

AC C

EP

224

ACCEPTED MANUSCRIPT 236

Sequence analysis revealed that the P1 and P2 regions of Cosa-CHN showed the highest similarity to those of HCosV-B, while the P3 region was most similar to that of

238

HCosV-D; however, all aa identity values were low (<70%). Although picornavirus

239

recombination is frequently observed within, but not between, viral species, with the

240

exception of exchange of the regulatory 5′ UTR region [22,23], the recombinant HCosV

241

D/E strain found in a previous study that exhibited recombination at the P1/P2 junction was

242

designated a single species [9]. Recombination analysis showed no recombination event in

243

Cosa-CHN. Phylogenetic analyses of the polyprotein and VP3/VP1 indicated that this virus

244

forms a separate branch to previous HCosVs, suggesting that Cosa-CHN represents a novel

245

species. The marked genetic diversity of HCosVs suggests a wide range of effects on the

246

host, as is the case for the similarly diverse enteroviruses.

SC

M AN U

247

RI PT

237

Primers for screening HCosV matched HCosV A–F well, but Cosa-CHN less well, as three nucleotides differed in each real time PCR primer (mismatched in position 12/13/15

249

of the positive primer and in position 5/7/8 of the negative primer ). Therefore, it was

250

unsurprising that Cosa-CHN could be amplified using these primers. However, it may be

251

that low Cosa-CHN viral load infections were hard to detect with initial screening primers

252

although this seem unlikely as ID 144 had the lowest viral load. In this study, the detection

253

rate of known HCosVs was 0.2% (2/922), markedly lower than reported in Shanghai, China

254

(3.2% in children with diarrhea and 1.6% in healthy children) [4], but similar to rates

255

reported previously in the UK, Thailand, Japan, and Italy (0.1–0.5%) [6,8,20,22]. To

256

evaluate the prevalence of Cosa-CHN, we designed specific primers; the Cosa-CHN

257

detection rates were 1.3 and 1.7% in the case and control groups, respectively. Furthermore,

258

the detection rate of Cosa-CHN in Liuyang (southern China) and Lulong (northern China)

259

Provinces differed significantly (4.5% vs. 1.7%). Therefore, the prevalence of Cosa-CHN

260

differed according to geographical area. Furthermore, Cosa-CHN-positive samples from

261

both areas were collected year-round, suggestive of no seasonal difference in distribution

262

(data not shown). This is unlike other enteric viruses, such as enterovirus, the prevalence of

263

most of which varies seasonally but not geographically [24].

264 265

AC C

EP

TE D

248

The prevalence of Cosa-CHN, particularly in other countries, should be investigated further. The incidence of HCosV in raw sewage and river water in Japan (71 and 29%,

ACCEPTED MANUSCRIPT respectively) was markedly greater than that reported in this study, which is to be expected

267

as sewage is collection of samples from multiple persons. HCosV was also detected in pig

268

stools in Bolivia, Thailand, and Japan (51.8, 55.4, and 18.9%, respectively). These data

269

suggest that the distribution of cosavirus species may differ among environmental sources,

270

populations and geographical areas. Interestingly, when using the specific primers (both

271

nested PCR primers and real time PCR primers) used in this study to screen the

272

Cosa-CHN-postive samples about three years later, it showed that all the samples were

273

negative for Cosa-CHN. This phenomenon should be caused by the degradation of the virus

274

in the sample, which indicates that this virus is not stable and the future study for this virus

275

should be done using the fresh specimens.

SC

RI PT

266

The pathogenicity of HCosV in humans remains unclear, as the wide genetic diversity

277

and ubiquity of the virus complicates disease-association studies. Nevertheless, HCosV has

278

been detected in subjects with diarrhea; e.g., two previous studies from Japan demonstrated

279

that HCosV might cause diarrhea in pediatric patients [8,25]. In this study, two samples

280

from healthy children were positive for known HCosVs, so we evaluated the relationship

281

between Cosa-CHN and AGE in children. There was no difference in the detection rate and

282

mean viral load of Cosa-CHN between the case and control subjects. However, coinfection

283

with other diarrhea-related viruses was frequent in the Cosa-CHN-positive samples from

284

the case group (5/8). Further statistical analysis indicated that Cosa-CHN coinfection did

285

not exacerbate the clinical symptoms of gastroenteritis. Taken together, these findings

286

suggest that Cosa-CHN does not play a role in gastroenteritis. However, it should be noted

287

that the viral loads were relatively low in this study. A previous study reported that

288

norovirus viral loads lower than a threshold level, it should be regarded as negative; in

289

other word, though the virus was detected, it was not a causal factor in the disease [26].

290

This might also be the case for Cosa-CHN infection, so the possibility that a high load of

291

Cosa-CHN may cause diarrhea cannot be ruled out. Moreover, specific cosavirus genotypes

292

may yet be shown to be associated with specific diseases as is the case for enteroviruses.

293

Further studies involving virus stability, virus isolation, animal experiments, serology and

294

molecular epidemiology should be performed to evaluate the pathogenicity of different

295

cosavirus species and genotypes.

AC C

EP

TE D

M AN U

276

ACCEPTED MANUSCRIPT 296 297

Funding information

298

This work was supported by the National Natural Science Foundation of China (Grant No.

299

81290345).

301

Competing interest

302

All authors declare no conflict of interest.

RI PT

300

303

Author contribution

305

ZJD and JMY designed the experiments; JMY, YYA and LLL performed the experiments;

306

JMY and YYA analyzed the results; JMY wrote the draft manuscript. ZJD revised the

307

manuscript. All the authors reviewed the manuscript.

M AN U

SC

304

308 309

References

310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332

1. Kapoor A, Victoria J, Simmonds P, Slikas E, Chieochansin T, et al. (2008) A highly prevalent and genetically diversified Picornaviridae genus in South Asian children. Proc Natl Acad Sci U S A 105:

TE D

20482-20487.

2. Holtz LR, Finkbeiner SR, Kirkwood CD, Wang D (2008) Identification of a novel picornavirus related to cosaviruses in a child with acute diarrhea. Virol J 5: 159. 3. Blinkova O, Rosario K, Li L, Kapoor A, Slikas B, et al. (2009) Frequent detection of highly diverse variants of cardiovirus, cosavirus, bocavirus, and circovirus in sewage samples collected in the

EP

United States. J Clin Microbiol 47: 3507-3513. 4. Dai XQ, Hua XG, Shan TL, Delwart E, Zhao W. (2010) Human cosavirus infections in children in China. J Clin Virol 48: 228-229.

AC C

5. Stocker A, Souza BF, Ribeiro TC, Netto EM, Araujo LO, et al. (2012) Cosavirus infection in persons with and without gastroenteritis, Brazil. Emerg Infect Dis 18: 656-659.

6. Campanini G, Rovida F, Meloni F, Cascina A, Ciccocioppo R, et al. (2013) Persistent human cosavirus infection in lung transplant recipient, Italy. Emerg Infect Dis 19: 1667-1669.

7. Oude Munnink BB, Canuti M, Deijs M, de Vries M, Jebbink MF, et al. (2014) Unexplained diarrhoea in HIV-1 infected individuals. BMC Infect Dis 14: 22.

8. Okitsu S, Khamrin P, Thongprachum A, Nishimura S, Kalesaran AF, et al. (2014) Detection and molecular characterization of human cosavirus in a pediatric patient with acute gastroenteritis, Japan. Infect Genet Evol 28C: 125-129. 9. Kapusinszky B, Phan TG, Kapoor A, Delwart E (2012) Genetic diversity of the genus Cosavirus in the family Picornaviridae: a new species, recombination, and 26 new genotypes. PLoS One 7: e36685. 10. Yu JM, Ao YY, Liu N, Li LL, Duan ZJ (2015) Salivirus in Children and Its Association with Childhood Acute Gastroenteritis: A Paired Case-Control Study. PLoS One 10: e0130977.

ACCEPTED MANUSCRIPT 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370

11. Chan MC, Sung JJ, Lam RK, Chan PK, Lai RW, et al. (2006) Sapovirus detection by quantitative

371

Figure legends

372

Figure 1. (a) genome organization of Cosa-CHN; (b) whole virus sequencing approach; and

373

(c) protein cleavage map. Nine fragments were used to produce the final genome sequence.

374

The potential cleavage sites of Cosa-CHN are similar to those of HCosV B. The location of

real-time RT-PCR in clinical stool specimens. J Virol Methods 134: 146-153. 12. Kageyama T, Kojima S, Shinohara M, Uchida K, Fukushi S, et al. (2003) Broadly reactive and highly sensitive assay for Norwalk-like viruses based on real-time quantitative reverse transcription-PCR. J Clin Microbiol 41: 1548-1557. 13. Pang XL, Lee B, Boroumand N, Leblanc B, Preiksaitis JK, et al. (2004) Increased detection of rotavirus using a real time reverse transcription-polymerase chain reaction (RT-PCR) assay in stool specimens

RI PT

from children with diarrhea. J Med Virol 72: 496-501.

14. Wolffs PF, Bruggeman CA, van Well GT, van Loo IH. (2011) Replacing traditional diagnostics of fecal viral pathogens by a comprehensive panel of real-time PCRs. J Clin Microbiol 49: 1926-1931. 15. Fauquet. C, Mayo. M, J. M, U. D, LA. B (2005) Virus taxonomy: Eighth report of the International Committee on Taxonomy of Viruses. San Diego, CA: Elsevier Academic Press.

SC

16. Okitsu S, Khamrin P, Hanaoka N, Thongprachum A, Takanashi S, et al. (2016) Cosavirus (family Picornaviridae) in pigs in Thailand and Japan. Arch Virol 161: 159-163.

17. Khamrin P, Chaimongkol N, Malasao R, Suantai B, Saikhruang W, et al. (2012) Detection and molecular characterization of cosavirus in adults with diarrhea, Thailand. Virus Genes 44: 244-246.

M AN U

18. Nix WA, Khetsuriani N, Penaranda S, Maher K, Venczel L, et al. (2013) Diversity of picornaviruses in rural Bolivia. J Gen Virol 94: 2017-2028.

19. Rezig D, Ben Farhat E, Touzi H, Meddeb Z, Ben Salah A, et al. (2015) Prevalence of human cosaviruses in Tunisia, North Africa. J Med Virol 87: 940-943.

20. Kitajima M, Rachmadi AT, Iker BC, Haramoto E, Pepper IL, et al. (2015) Occurrence and genetic diversity of human cosavirus in influent and effluent of wastewater treatment plants in Arizona, United States. Arch Virol 160: 1775-1779.

TE D

21. Yamashita T, Sakae K, Tsuzuki H, Suzuki Y, Ishikawa N, et al. (1998) Complete nucleotide sequence and genetic organization of Aichi virus, a distinct member of the Picornaviridae associated with acute gastroenteritis in humans. J Virol 72: 8408-8412. 22. Santti J, Hyypia T, Kinnunen L, Salminen M (1999) Evidence of recombination among enteroviruses. J Virol 73: 8741-8749.

EP

23. Simmonds P, Welch J (2006) Frequency and dynamics of recombination within different species of human enteroviruses. J Virol 80: 483-493. 24. Xing W, Liao Q, Viboud C, Zhang J, Sun J, et al. (2014) Hand, foot, and mouth disease in China, 2008-12:

AC C

an epidemiological study. Lancet Infect Dis 14: 308-318. 25. Okitsu S, Khamrin P, Thongprachum A, Nishimura S, Kalesaran AF, et al. (2014) Detection and molecular characterization of human cosavirus in a pediatric patient with acute gastroenteritis, Japan. Infect Genet Evol 28: 125-129.

26. Phillips G, Lopman B, Tam CC, Iturriza-Gomara M, Brown D, et al. (2009) Diagnosing norovirus-associated infectious intestinal disease using viral load. BMC Infect Dis 9: 63.

ACCEPTED MANUSCRIPT 375

the initial PCR for HCosV and Cosa-CHN specific RT-nPCR amplicon products were in

376

red and marked as "VP1" and "VP3/VP1" ,respectively.

377

Figure 2. Phylogenetic analysis of Cosa-CHN sequences: (a) polyprotein amino acid

379

sequences; (b) partial VP3/VP1 nucleotide sequences. The tree was constructed using the

380

neighbor-joining method b ny MEGA ver. 5 with 1,000 bootstrap replicates. Sequences

381

from this study are indicated by "•". Cosa-CHN formed a lineage distinct from other known

382

HCosV.

AC C

EP

TE D

M AN U

SC

RI PT

378

ACCEPTED MANUSCRIPT Table 1. Clinical characteristics, viral loads, and related information of eight children in the case group positive for Cosa-CHN. Duration of Patient ID

diarrhea,

Age,

Frequency of

Viral load,

diarrhea, per

copies

Sex

Area

days

Month

Coinfection

Fever

day

Vomiting

(copies/ml)

144

F

Hunan

7

8

HuCV

+

4

-

3.48×101

038

F

Hebei

3

8

-

+

061

F

Hunan

2

6

RV, AdV

+

070

M

Hunan

2

7

AstV

-

102

F

Hunan

1

3

-

N

052

M

Hunan

3

6

-

113

M

Hebei

2

15

RV

073

F

Hebei

15

4

RV

RI PT

No.

-

8.36×101

2

+

1.05×102

6

-

3.46×102

2

-

3.65×102

+

4

-

5.02×102

-

7

-

1.90×103

-

6

+

1.07×104

M AN U

SC

4

Note: Duration of diarrhea indicates the time from the date of admission to the date of

AC C

EP

TE D

discharge from hospital. M, male; F, female; N, no data; +, present; -, absent.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT