Phylogenetic inference of the porcine Rotavirus A origin of the human G1 VP7 gene

Infection, Genetics and Evolution 40 (2016) 205–213

Contents lists available at ScienceDirect

Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid

Research paper

Phylogenetic inference of the porcine Rotavirus A origin of the human G1 VP7 gene Loan Phuong Do a,b, Toyoko Nakagomi a,c, Hiroki Otaki c, Chantal Ama Agbemabiese a, Osamu Nakagomi a,c,⁎, Hiroshi Tsunemitsu d a

Department of Hygiene and Molecular Epidemiology, Graduate School of Biomedical Sciences, Nagasaki University, Nagasaki, Japan Department of Virology, National Institute of Hygiene and Epidemiology, Hanoi, Vietnam Centre for Bioinformatics and Molecular Medicine, Graduate School of Biomedical Sciences, Nagasaki University, Nagasaki, Japan d Dairy Hygiene Research Division, Hokkaido Research Station, National Institute of Animal Health, Sapporo, Hokkaido, Japan b c

a r t i c l e

i n f o

Article history: Received 9 September 2015 Received in revised form 25 February 2016 Accepted 1 March 2016 Available online 4 March 2016 Keywords: Rotavirus Evolution Interspecies transmission VP7 gene G1

a b s t r a c t Rotavirus A (RVA) is an important cause of acute gastroenteritis in children worldwide. The most common VP7 genotype of human RVA is G1, but G1 is rarely detected in porcine strains. To understand the evolutionary relationships between human and porcine G1 VP7 genes, we sequenced the VP7 genes of three Japanese G1 porcine strains; the first two (PRV2, S80B) were isolated in 1980 and the third (Kyusyu-14) was isolated in 2001. Then, we performed phylogenetic and in-silico structural analyses. All three VP7 sequences clustered into lineage VI, and the mean nucleotide sequence identity between any pair of porcine G1 VP7 sequences belonging to lineage VI was 91.9%. In contrast, the mean nucleotide sequence identity between any pair of human G1 VP7 sequences belonging to lineages I–V was 95.5%. While the mean nucleotide sequence identity between any pair of porcine lineage VI strain and human lineage I–V strain was 85.4%, the VP7 genes of PRV2 and a rare porcine-like human G1P[6] strain (AU19) were 98% identical, strengthening the porcine RVA origin of AU19. The phylogenetic tree suggests that human G1 VP7 genes originated from porcine G1 VP7 genes. The time of their most recent common ancestor was estimated to be 1948, and human and porcine RVA strains evolved along independent pathways. In-silico structural analyses identified 7 amino acid residues within the known neutralisation epitopes that show differences in electric charges and shape between different porcine and human G1 strains. When compared with much divergent porcine G1 VP7 lineages, monophyletic, less divergent human G1 VP7 lineages support the hypothesis that all human G1 VP7 genes included in this study originated from a rare event of a porcine RVA transmitting to humans that was followed by successful adaptation to the human host. By contrast, AU19 represents interspecies transmission that terminated in dead-end infection. © 2016 Published by Elsevier B.V.

1. Introduction Rotavirus A (RVA), a species within the genus Rotavirus, family Reoviridae, is a major cause of severe acute gastroenteritis in infants and young children living in countries where rotavirus vaccines are not introduced in the national immunisation programmes (Estes and Greenberg, 2013; Tate et al., 2011). RVA also causes acute enteritis in young piglets and is responsible for a significant proportion of diarrhoea (Amimo et al., 2015; Miyazaki et al., 2013; Marthaler et al., 2014). Each of the host species has its own RVA strains and is known to be susceptible to heterologous strains. Experimentally, human strains infect and cause diarrhoea in gnotobiotic piglets (Hoshino et al., 1995; Gonzalez et al., 2010; Saif et al., 1997), whereas there were numerous reports ⁎ Corresponding author at: Department of Hygiene and Molecular Epidemiology, Graduate School of Biomedical Sciences, Nagasaki University, 1-12-4 Sakamoto, Nagasaki 852-8523, Japan. E-mail address: [email protected] (O. Nakagomi).

http://dx.doi.org/10.1016/j.meegid.2016.03.001 1567-1348/© 2016 Published by Elsevier B.V.

that showed cases in which porcine RVA strains were suspected to infect and cause diarrhoea in children (Komoto et al., 2013; Do et al., 2014; My et al., 2014; Martinez et al., 2014; Zeller et al., 2012; Papp et al., 2013a; Esona et al., 2009; Dong et al., 2013; Matthijnssens et al., 2010b; Wang et al., 2010; Mukherjee et al., 2011; Mladenova et al., 2012; Varghese et al., 2006; Degiuseppe et al., 2013). However, genotypes prevalent among human RVA strains and porcine RVA strains are markedly different; Genotype G1 is the most predominant among human strains but rare among porcine strains. G1 accounted for an average of 40% among human RVA strains before the introduction of rotavirus vaccines (Banyai et al., 2012; Santos and Hoshino, 2005; Gentsch et al., 2005) and 22–33% after the vaccine introduction (Kawai et al., 2012; Doro et al., 2014; Leshem et al., 2014). In sharp contrast, G1 was rarely found in animal rotaviruses and only around 1% of porcine RVA strains are known to possess the G1 genotype (Papp et al., 2013b). During recent surveillance studies in pigs, the G1 genotype was almost never found (Theuns et al., 2014; Pham et al., 2014; Saikruang et al., 2013; Collins et al., 2010). Most of the pig G1

206

L.P. Do et al. / Infection, Genetics and Evolution 40 (2016) 205–213

L.P. Do et al. / Infection, Genetics and Evolution 40 (2016) 205–213

strains were archival strains, isolated in the 1970s–1980s. However, a porcine G1 strain was reported from Slovenia in 2004 (Steyer et al., 2007). We recently reported that a rare G1P[6] human strain isolated from a Japanese infant with diarrhoea in 1997 possessed the genetic background of porcine RVA, and that its VP7 gene was most closely related to the VP7 gene of a British porcine strain detected in 1976 (Do et al., 2014). As a few archived Japanese porcine RVA strains were available, we aimed at exploring the evolutionary relationships between human, porcine and porcine-like human G1 VP7 genes, and sequenced the VP7 genes of two G1 strains isolated in 1980 and another G1 strain isolated in 2001. 2. Materials and methods 2.1. The viruses Three porcine RVA strains analyzed in this study for their VP7 genes were strains RVA/Pig-tc/JPN/S80B/1980/G1P[7], RVA/Pig-tc/JPN/PRV2/ 1980/G1P[7] and RVA/Pig-tc/JPN/Kyusyu-14/2001/G1P[7]. The first two strains were isolated from pigs in Japan in 1980 by using MA104 cells according to the method described later by Sato et al. (1981). The last one, Kyusyu-14, was also isolated from pigs in Japan in 2001 using MA104 cells (Teodoroff et al., 2005). No original stool specimens were available. The isolation in cell culture required a few sequential passages in MA104 cells followed by three-cycles of plaque purifications before the seed stocks were made. The total number of passages from the original isolation was not recorded, but it should be 10 passages at most (as this was the standard practice in rotavirus cell culture). 2.2. Genome sequencing Viral genomic RNAs were extracted from MA104 cells infected with each of these strains using the QIAamp Viral RNA Mini Kit (QIAGEN Sciences, Germantown, MD, USA) according to the manufacturer's instructions. An 8 μl portion of genomic RNA was mixed with random primers and dNTPs in a total volume of 9 μl and denatured at 97 °C for 5 min followed by quenching on ice. To this was added the reversetranscription mixture containing Super Script III Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) in a final reaction volume of 20 μl. The thermal profile included incubation at 25 °C for 5 min, at 42 °C for 60 min and at 70 °C for 15 min for reverse transcription. The VP7 gene was amplified from 2 μl of cDNA with the Beg9 and End9 primers described by Gouvea et al. (1990) by using the GoTaq® Green Master Mix system (Promega Corporation, Madison, WI, USA) at 95 °C for 5 min, and 35 cycles of amplification (at 94 °C for 1 min; at 42 °C for 2 min and at 72 °C for 3 min) followed by the final extension at 72 °C for 8 min. The amplified products were then purified using an ExoSAP-IT purification kit (USB Products, Affymetrix, Cleveland, OH, USA) according to the manufacturer's instructions. Nucleotide sequencing reactions were performed by fluorescent dideoxy chain termination chemistry using the BigDye Terminator Cycle Sequencing Ready Reaction Kit, version 3.1 (Applied Biosystems, Foster City, CA, USA), and nucleotide sequences were determined using an ABI Prism 3730 Genetic Analyzer (Applied Biosystems). 2.3. Phylogenetic analysis We searched the GenBank database with the Basic Local Alignment Search Tool (BLAST) using the prototype Wa strain as a query sequence to collect all available human and animal G1 VP7 genes. Multiple

207

alignments of the nucleotide sequences (135 plus 2 as outgroups) were performed using the MUSCLE programme in the MEGA package (v6.06) (Tamura et al., 2013). While the shortest sequence length we determined was 981 nucleotides, the shortest sequence length for drawing the tree was 906 nucleotides. The pairwise nucleotide identities were calculated using a distance estimation algorithm (p-distance) in the MEGA6 package. The best fit model for nucleotide substitutions was chosen based on the smallest Bayesian Information Criterion score. Phylogenetic analysis of the VP7 gene was conducted by the Maximum Likelihood method with 1000 bootstrap replicates and modelled using Tamura-Nei (TN93) and a discrete Gamma distribution (G) (Nei and Kumar, 2000). Additional phylogenetic relationship among the G1 VP7 genes was inferred using Bayesian method implemented in MrBayes software v.3.2 (Huelsenbeck and Ronquist, 2001). Estimation of the time of the most recent common ancestor (tMRCA) and divergence times of lineages were estimated in the VP7 gene using the Bayesian Markov chain Monte Carlo (MCMC) method in BEAST v1.8.1 (Drummond et al., 2012). The models we used were the TN93 + G nucleotide substitution model, an uncorrelated lognormal relaxed clock model (Drummond et al., 2006) and a coalescent constant size tree prior (Drummond et al., 2005). The uncorrelated lognormal relaxed clock model was chosen over the strict molecular clock model because the relaxed clock model provides a natural means to impose assumptions about rates and times as prior probability densities and accommodate uncertainty in divergence-time estimates (Heath and Moore, 2014). MCMC runs were carried out for 50 million generations and evaluated using Tracer software v1.6 (http://tree.bio.ed.ac.uk/software/tracer/). The resulting maximum clade credibility (MCC) tree was annotated with the TreeAnnotator and viewed with FigTree v1.4.2 (http://tree.bio.ed.ac. uk/software/figtree/).

2.4. In-silico analysis of the VP7 protein structures Amino acid sequences of the VP7 proteins were translated from the nucleotide sequences. We examined the amino acid residues previously identified as involved in antigenic sites and neutralisation domains of the VP7 proteins (Aoki et al., 2009; Dormitzer et al., 2002, 2004). Structural analysis of the VP7 trimer (Protein Data Bank [PDB] number 3FMG) (Aoki et al., 2009) was performed using PyMOL (https://www.pymol.org/). A homology model of the VP7 monomer of porcine RVA strains in lineage VI and a representative human RVA strain, 88H249, which carried the different amino acids from any of porcine RVA were constructed with the crystallographic data of strain RRV (PDB 3GZT [strain RRV]) (Chen et al., 2009). Amino acid alignment was done to find the differences between the target strain and RRV. Then, computational mutation of residues was made using mutate.pl in Multiscale Modeling Tools for Structural Biology (MMTSB) Tool Set (https://mmtsb.org/) (Feig et al., 2004). Because the predicted structures lacked hydrogen atoms, the complete.pl utility in MMTSB Tool Set was used to provide the missing hydrogen atoms for all of the predictions based on the CHARMM27 Force Field. Electrostatic potential of each protein structure was calculated using PDB2PQR server (http://nbcr-222.ucsd.edu/pdb2pqr_2.0.0/) (Dolinsky et al., 2004) and APBS plugin (Baker et al., 2001) in PyMOL based on PARSE Force Field (Sitkoff et al., 1994) and termini proteins were treated as neutral. All figures were visualised by PyMOL.

Fig. 1. The Maximum Likelihood tree (a) and Bayesian phylogenetic tree (b) for the G1 VP7 genes of PRV2, S80B and Kyusyu-14 (indicated with black dots) and AU19 (indicated in black square) with other human rotavirus strains representing each of previously defined lineages I–V. A reference genetic distance (substitutions per site) is indicated with a scale bar at the bottom of each panel. Percent bootstrap support is indicated by the value at each node when the value was 70% or larger, and the posterior probabilities for the Bayesian inference are indicated at the node.

208

L.P. Do et al. / Infection, Genetics and Evolution 40 (2016) 205–213

2.5. Nucleotide sequence accession numbers The VP7 gene sequences of PRV2, S80B and Kyusyu-14 were deposited in the GenBank/EMBL/DDBJ databases under accession numbers LC081149, LC081150, and LC081151, respectively. 3. Results 3.1. Phylogenetic analysis

Table 1 Relationships between and within the G1 VP7 genes carried by human and porcine rotavirus strains.

Within three Japanese porcine rotavirus strains Within lineage VI Within human rotaviruses belonging to lineages I–V Japanese porcine rotavirus strains and lineages I–V Lineage VI and lineages I–V

Average identity (%)

Range of identity (%)

91.7

89.8–96.4

91.9

88.1–97.7

95.5

90.4–100

85.8

82.3–88.5

85.4

82.3–84.5

We determined the VP7 sequences of three Japanese porcine RVA strains, PRV2, S80B, and Kyusyu-14, all detected in Japan but 21 years apart. To place porcine G1 VP7 sequences into the phylogenetic context, a maximum likelihood tree was drawn by including porcine and bovine G1 strains (Bellinzoni et al., 1987, 1990; Steyer et al., 2007; Theuns et al., 2015; Ciarlet and Liprandi, 1994; Blackhall et al., 1992; El-Attar et al., 2001) as well as human G1 strains representing lineages I–V (Arista et al., 2006; Le et al., 2010). In the VP7 phylogenetic trees (Fig. 1) lineages I to V contained all human G1 sequences with a 100% bootstrap support, whereas porcine G1 sequences clustered into two lineages, lineage VI and VII. The three Japanese porcine strains, PRV2, Kyusyu-14 and S80B, clustered into lineage VI together with SW20/21 (detected in England in 1976), RV277 (detected in Belgium in 1977) and a human RVA strain AU19 (detected in Japan in 1997). When pair-wise nucleotide identities were calculated, the nucleotide identity within the three Japanese porcine G1 sequences ranged from 89.8% to 96.4% with a mean of 91.7% whereas the nucleotide identity between any one of PRV2, Kyusyu-14, and S80B and human RVA sequences ranged from 82.3% to 88.5% with a mean of 85.8% (Table 1). The rare human G1P[6] strain AU19 had the highest nucleotide identity of 97.6% with PRV2. While the number of lineages of the human G1 sequences is as many as five, the mean identity within human G1 lineage (the aggregated “lineage” comprising lineages I–V) was 95.5% (range: 90.4%–100%) and that within the porcine G1 lineage (lineage VI) was 91.9% (range: 88.1%–97.7%) whereas the mean distance between human and porcine lineages was 85.4% (range: 82.3%–84.5%) (Table 1). The divergence time of the ancestor of the human lineage (lineages I–V) from the most closely related porcine G1 strain, P21-5, detected in Slovenia was 1915 (Fig. 2). The time of the most recent common ancestor (tMRCA) of the human lineage was estimated to be 1948 and that of lineage VI was estimated to be 1927. The evolutionary rate of the porcine G1 sequences (including AU19) was 0.89 × 10−3 substitutions/ site/year (95% HPD: 2.4 × 10−4–1.6 × 10−3) and that of human sequences was 1.02 × 10−3 substitutions/site/year (95% HPD: 7.6 × 10−4–1.3 × 10−3) (Table 2).

local electric charge (from less negative to negative) and shape (Fig. 4). By contrast, the E145D and V217I substitutions made neither significant change in the local electric charge nor shape. On the other hand, amino acid residues 91 and 238 were not identified by experiments with G1-specific monoclonal antibody escape mutants, but the S91N substitution (Kyusyu-14 vs AU19) and the D238N substitution (SW20/21 vs AU19) changed the local electric charge and shape (Fig. 4). Thus, in silico analysis showed that 4 substitutions, S91N (7-1a), D238N (7-1b), D147N (7-2) and N221D (7-2), likely affected neutralisation specificity of the virus in terms of both local electric charge and shape. When the amino acid sequences of lineage VI porcine strains and 123 human strains belonging to lineages I–V were compared, amino acid substitutions were noted at residues 87, 91, 94, 123, and 291, and residues 217 and 221 in the neutralisation domains 7-1a and 7-2, respectively (Fig. 3c). While porcine rotaviruses had an equal number (3) of T and I at 87, all human strains had T at this residue. At residue 91, all except Kyusyu-14 had N, but 111 (89%) human strains had T and 14 (11%) had N. At residue 94, all porcine strains had N, but 71 (57%) of human strains had N and 54 (43%) had S. At residue 123, all porcine strains had S, but 99 (79%) of human strains had S and 26 (21%) had N. At residue 291, all porcine strains (AU19 excluded) had K, but 76 (61%) of human strains had K and 49 (31%) had R. Regarding neutralisation domain 7-2, 4 porcine strains had I and 2 strains had V at residue 217, 104 (83%) of human strains had M and 21 (17%) had T, whereas all porcine (AU19 excluded) and human strains had N at residue 221. Of these, amino acid residues 94, 217, 221 and 291 were the epitopes identified by experiments with neutralisation escape mutants against G1-specific monoclonal antibodies. The I/V217M/T, D221N and K291R substitutions affected the local electric charge and shape (Fig. 4). Although amino acid residues 91 and 123 were not recognised by G1-specific monoclonal escape mutants, the N91T and S123N mutations changed the local electric charge and shape (Fig. 4).

3.2. In silico structural analysis

4. Discussion

We compared amino acid residues in the neutralisation domains of the VP7 proteins (i) within lineage VI (porcine lineage) and (ii) between lineage VI and the aggregated human lineage (lineages I–V), and performed in-silico analysis to predict how the amino acid substitutions affect the local structure and electric charge. The three neutralisation domains (7-1a, 7-1b and 7-2) which comprise 29 amino acid residues are located on the outer surface of the VP7 trimer are shown in Fig. 3a. When the amino acid sequences of any pair of porcine strains within lineage VI, including porcine-like human AU19, were compared, there were 8 amino acid substitutions in the neutralisation domains; they were T87I, S91N, I125V (7-1a), D238N (7-1b), E145D, D147N, V217I and N221D (7-2) (Fig. 3b). Of these, 4 amino acid residues (145, 147, 217 and 221) were known to be the epitopes selected by neutralisation escape mutants against G1-specific monoclonal antibodies. The D147N substitution (RV277 vs AU19) made an alteration in the local electric charge (from negative to less negative) and shape (Fig. 4). Similarly, the N221D substitution (all lineage VI strains vs AU19) changed the

The determination of the VP7 sequences of three porcine strains, S80B, PRV2, and Kyusyu-14, isolated in Japan 21 years apart doubled the number of G1 sequences belonging to lineage VI where there had been only 3 sequences available. These were British porcine strain SW20/21 and Belgian porcine strain RV277 both possessing P[7] for VP4 genes and porcine-like human strain AU19 possessing P[6] for VP4 (El-Attar et al., 2001; Do et al., 2014; Theuns et al., 2015). A nucleotide sequence diversity of 2.4% revealed in this study between the VP7 gene of AU19 and that of PRV2 is considered small enough given the 17 years of interval in detection and the calculated evolutionary rate of approximately 1 × 10−3 substitutions/site/year. This calculation points to the hypothesis that the VP7 gene of AU19 was derived directly from that of its contemporary porcine RVA strain (Do et al., 2014). The fact that AU19 was the only human strain with its VP7 gene belonging to lineage VI suggests that AU19 represents an example of a porcine RVA strain failing to establish an efficient human-to-human transmission chain after crossing the host species barrier.

L.P. Do et al. / Infection, Genetics and Evolution 40 (2016) 205–213

209

Fig. 2. Maximum clade credibility (MCC) trees for the G1 VP7 sequences of PRV2, S80B and Kyusyu-14 (indicated with black dots) and AU19 (indicated in black square) with other human rotavirus strains representing each of previously defined lineages I–V. MCC trees were constructed using the Bayesian MCMC framework. The years of divergence of each lineage are indicated at each node. Human lineage was collapsed for the sake of simplicity.

Two interesting observations were made from the VP7 gene trees using the maximum likelihood and Bayesian methods (Figs. 1 and 2). First, lineages I–V of G1 human strains clustered with a 100% bootstrap support into a single human clade with tMRCA being estimated to be 1948. Second, the year of divergence of this common ancestor from porcine RVA strains was 1915. Although a close evolutionary relationship between porcine and human Wa-like strains was implicated (Matthijnssens et al., 2008, 2010a; Theuns et al., 2015; Rahman et al., 2007; Teodoroff et al., 2005), the observations made in this study more clearly suggest that all human G1 VP7 sequences were the progeny of one of the genetically diverse porcine G1 strains that jumped into and subsequently adapted successfully to the human host species. The

Table 2 Time of most recent common ancestors for the human and porcine G1 lineages and evolutionary rates of the VP7 genes analysed in this study. Lineage Parameter

Human

Number of sequences 123 Sampling period 1974–2010 Geographical coverage Global 1.01 (0.76–1.30) Evolutionary rate (10−3 substitutions/site/year) (95% HPD) tMRCA for the human lineage 1948 (1928–1962) (95% HPD) tMRCA of the lineage VI (95% HPD) – Year of divergence from porcine RVA 1915 (1868–1947) Note: HPD is the abbreviation for highest posterior density.

Porcine 6 1976–2001 Global 0.89 (0.24–1.57) – 1927 (1894–1956) –

interspecies transmission of a porcine G1 strain to humans was estimated to occur between 1915 and 1948 (Fig. 2). After the divergence, human and porcine G1 sequences evolved in different evolutionary directions as Theuns et al. (2015) described based on their observations on G2 and G9 VP7 genotypes that VP7 genotypes shared between human and pig RVA strains evolved clearly in different evolutionary directions. The transfer of the VP7 gene from an animal RVA on the genetic backbone of human strains through reassortment after interspecies transmission may be a general mechanism by which a new VP7 gene is introduced into human RVA. It was hypothesised that G9 and G12 genotypes in humans were acquired from pigs (Matthijnssens et al., 2010a; Hoshino et al., 2005; Theuns et al., 2015), and the tMRCA of lineage III of G9 and lineage III of G12 were estimated to be as recently as 1989 and 1995, respectively (Matthijnssens et al., 2010a). More recent and nascent example may be G6P[6] strains emerging and spreading in some parts of Africa where Agbemabiese et al. have observed an increased evolutionary rate of the G6 VP7 gene that had been transferred from bovine RVA to human RVA possessing the DS-1-like genetic background (Agbemabiese et al., 2015). A full genome analysis of G8 strains collected in Malawi over a 10 year period revealed that the G8 VP7 genes, which are generally considered to be from bovine origin, were maintained in human population almost exclusively through humanto-human transmission when G8 VP7 gene was transferred to a human DS-1-like genetic backbone (Nakagomi et al., 2013). When amino acid substitutions occurring in the neutralisation domains (7-1a, 7-1b and 7-2) were mapped on the VP7 trimers, the 8 amino acid substitutions within lineage VI porcine rotavirus strains

210

L.P. Do et al. / Infection, Genetics and Evolution 40 (2016) 205–213

Fig. 3. Three-dimensional structure model of the VP7 trimer (PDB entry 3FMG). The antigenic regions are divided in three neutralisation domains, 7-1a, 7-1b and 7-2, which are coloured in red, yellow and orange, respectively. Amino acid residues coloured in green are the residues where amino acid substitutions were observed within PRV strains, whereas amino acid residues coloured in blue are the residues where amino acid substitutions were observed between porcine strains and human strains. Amino acid substitutions surrounded by an oval contour indicate that these residues were identified with G1-specific neutralisation monoclonal escape mutants. (a) Top and side views; (b) amino acid substitutions observed within porcine strains; (c) amino acid substitutions observed between porcine and human strains. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

L.P. Do et al. / Infection, Genetics and Evolution 40 (2016) 205–213

211

Fig. 4. Three-dimensional structure models of the VP7 monomer (PDB entry 3GZT) of human rotavirus AU19, porcine rotavirus strains Kyusyu-14, SW20/21, S80B, and RV277, and human rotavirus 88H249. The colour codes show the electrostatic potential on the surface of the monomer. Negative electric charges are shown in red whereas positive charges are shown in blue. The unit of the electrostatic potential is kT/e, where k is Boltzmann constant, T is the temperature, and e is the charge of an electron. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

(including AU19) were distributed on all three neutralisation domains (Fig. 3b), whereas the 7 substitutions between lineage VI porcine strains and the 123 human strains (excluding AU19) belonging to lineages I–V examined in this study were distributed mostly on two domains (7-1a and 7-2) (Fig. 3c). Furthermore, 4 of the 8 substitutions within lineage VI porcine strains caused changes in electric charges or shape (Fig. 4), and so did 5 of the 7 substitutions between lineage VI porcine strains and any human strains (Fig. 4). Thus, the most of the substitutions are predicted to affect neutralisation specificity of the virus. Theoretically, any one of the diverse G1 VP7 genes of porcine RVA strains possessing different neutralisation specificity can be introduced into the human RVA population, which may result in a new G1 lineage in human RVA. A caution must be exercised, however; possibility remains as to the potential bias in evaluating substitutions in the antigenic regions, because a few changes are known to occur during the course of substantial number of serial passages of wildtype strains in cell cultures (Tsugawa et al., 2014).

Conflict of interest

5. Conclusions

Agbemabiese, C.A., Nakagomi, T., Suzuki, Y., Armah, G., Nakagomi, O., 2015. Evolution of a G6P[6] rotavirus strain isolated from a child with acute gastroenteritis in Ghana, 2012. J. Gen. Virol. 96, 2219–2231. Amimo, J.O., Junga, J.O., Ogara, W.O., Vlasova, A.N., Njahira, M.N., Maina, S., Okoth, E.A., Bishop, R.P., Saif, L.J., Djikeng, A., 2015. Detection and genetic characterization of porcine group A rotaviruses in asymptomatic pigs in smallholder farms in East Africa: predominance of P[8] genotype resembling human strains. Vet. Microbiol. 175, 195–210. Aoki, S.T., Settembre, E.C., Trask, S.D., Greenberg, H.B., Harrison, S.C., Dormitzer, P.R., 2009. Structure of rotavirus outer-layer protein VP7 bound with a neutralizing Fab. Science 324, 1444–1447. Arista, S., Giammanco, G.M., De Grazia, S., Ramirez, S., Lo Biundo, C., Colomba, C., Cascio, A., Martella, V., 2006. Heterogeneity and temporal dynamics of evolution of G1 human rotaviruses in a settled population. J. Virol. 80, 10724–10733. Baker, N.A., Sept, D., Joseph, S., Holst, M.J., McCammon, J.A., 2001. Electrostatics of nanosystems: application to microtubules and the ribosome. Proc. Natl. Acad. Sci. U. S. A. 98, 10037–10041.

We determined the VP7 sequences of three Japanese porcine RVA strains, two of which showed high nucleotide sequence identities (95.3–97.6%) with a rare human G1P[6] strain (AU19) detected in Japan. The G1 VP7 genes and amino acids of porcine RVA strains were diverse despite a small number of porcine G1 strains. Phylogenetic evidence suggests that human G1 strains were the progeny of one of the diverse populations of porcine G1 strains that crossed to another host species, thereby successfully spreading in the human population. Most likely, a reassortment event was needed for this successful spread. By contrast, AU19 represents an example of the interspecies transmission event that terminated in the dead-end infection.

There is no conflict of interest for any author to declare regarding this study. Acknowledgements This work was in part supported by grants from Japan Agency for Medical Research and Development (AMED) (15fk0108020h1302). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.meegid.2016.03.001. References

212

L.P. Do et al. / Infection, Genetics and Evolution 40 (2016) 205–213

Banyai, K., Laszlo, B., Duque, J., Steele, A.D., Nelson, E.A., Gentsch, J.R., Parashar, U.D., 2012. Systematic review of regional and temporal trends in global rotavirus strain diversity in the pre rotavirus vaccine era: insights for understanding the impact of rotavirus vaccination programs. Vaccine 30 (Suppl. 1), A122–A130. Bellinzoni, R.C., Mattion, N.M., Burrone, O., Gonzalez, A., La Torre, J.L., Scodeller, E.A., 1987. Isolation of group A swine rotaviruses displaying atypical electropherotypes. J. Clin. Microbiol. 25, 952–954. Bellinzoni, R.C., Mattion, N.M., Matson, D.O., Blackhall, J., La Torre, J.L., Scodeller, E.A., Urasawa, S., Taniguchi, K., Estes, M.K., 1990. Porcine rotaviruses antigenically related to human rotavirus serotypes 1 and 2. J. Clin. Microbiol. 28, 633–636. Blackhall, J., Bellinzoni, R., Mattion, N., Estes, M.K., La Torre, J.L., Magnusson, G., 1992. A bovine rotavirus serotype 1: serologic characterization of the virus and nucleotide sequence determination of the structural glycoprotein VP7 gene. Virology 189, 833–837. Chen, J.Z., Settembre, E.C., Aoki, S.T., Zhang, X., Bellamy, A.R., Dormitzer, P.R., Harrison, S.C., Grigorieff, N., 2009. Molecular interactions in rotavirus assembly and uncoating seen by high-resolution cryo-EM. Proc. Natl. Acad. Sci. U. S. A. 106, 10644–10648. Ciarlet, M., Liprandi, F., 1994. Serological and genomic characterization of two porcine rotaviruses with serotype G1 specificity. J. Clin. Microbiol. 32, 269–272. Collins, P.J., Martella, V., Sleator, R.D., Fanning, S., O′Shea, H., 2010. Detection and characterisation of group A rotavirus in asymptomatic piglets in southern Ireland. Arch. Virol. 155, 1247–1259. Degiuseppe, J.I., Beltramino, J.C., Millan, A., Stupka, J.A., Parra, G.I., 2013. Complete genome analyses of G4P[6] rotavirus detected in Argentinean children with diarrhoea provides evidence of interspecies transmission from swine. Clin. Microbiol. Infect. 19, E367–E371. Do, L.P., Nakagomi, T., Nakagomi, O., 2014. A rare G1P[6] super-short human rotavirus strain carrying an H2 genotype on the genetic background of a porcine rotavirus. Infect. Genet. Evol. 21, 334–350. Dolinsky, T.J., Nielsen, J.E., McCammon, J.A., Baker, N.A., 2004. PDB2PQR: an automated pipeline for the setup of Poisson–Boltzmann electrostatics calculations. Nucleic Acids Res. 32, W665–W667. Dong, H.J., Qian, Y., Huang, T., Zhu, R.N., Zhao, L.Q., Zhang, Y., Li, R.C., Li, Y.P., 2013. Identification of circulating porcine-human reassortant G4P[6] rotavirus from children with acute diarrhea in China by whole genome analyses. Infect. Genet. Evol. 20, 155–162. Dormitzer, P.R., Sun, Z.Y., Wagner, G., Harrison, S.C., 2002. The rhesus rotavirus VP4 sialic acid binding domain has a galectin fold with a novel carbohydrate binding site. EMBO J. 21, 885–897. Dormitzer, P.R., Nason, E.B., Prasad, B.V., Harrison, S.C., 2004. Structural rearrangements in the membrane penetration protein of a non-enveloped virus. Nature 430, 1053–1058. Doro, R., Laszlo, B., Martella, V., Leshem, E., Gentsch, J., Parashar, U., Banyai, K., 2014. Review of global rotavirus strain prevalence data from six years post vaccine licensure surveillance: is there evidence of strain selection from vaccine pressure? Infect. Genet. Evol. 28, 446–461. Drummond, A.J., Rambaut, A., Shapiro, B., Pybus, O.G., 2005. Bayesian coalescent inference of past population dynamics from molecular sequences. Mol. Biol. Evol. 22, 1185–1192. Drummond, A.J., Ho, S.Y., Phillips, M.J., Rambaut, A., 2006. Relaxed phylogenetics and dating with confidence. PLoS Biol. 4, e88. Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Mol. Biol. Evol. 29, 1969–1973. El-Attar, L., Dhaliwal, W., Howard, C.R., Bridger, J.C., 2001. Rotavirus cross-species pathogenicity: molecular characterization of a bovine rotavirus pathogenic for pigs. Virology 291, 172–182. Esona, M.D., Geyer, A., Banyai, K., Page, N., Aminu, M., Armah, G.E., Hull, J., Steele, D.A., Glass, R.I., Gentsch, J.R., 2009. Novel human rotavirus genotype G5P[7] from child with diarrhea, Cameroon. Emerg. Infect. Dis. 15, 83–86. Estes, M.K., Greenberg, H.B., 2013. In: Knipe, D.M., P. M. H. (Eds.), Rotaviruses. Field Virology 1. Wolters Kluwer Health/Lippincott Williams & Wilkins, Philadelphia, PA, pp. 1347–1401. Feig, M., Karanicolas, J., Brooks 3rd, C.L., 2004. MMTSB Tool Set: enhanced sampling and multiscale modeling methods for applications in structural biology. J. Mol. Graph. Model. 22, 377–395. Gentsch, J.R., Laird, A.R., Bielfelt, B., Griffin, D.D., Banyai, K., Ramachandran, M., Jain, V., Cunliffe, N.A., Nakagomi, O., Kirkwood, C.D., Fischer, T.K., Parashar, U.D., Bresee, J.S., Jiang, B., Glass, R.I., 2005. Serotype diversity and reassortment between human and animal rotavirus strains: implications for rotavirus vaccine programs. J. Infect. Dis. 192 (Suppl. 1), S146–S159. Gonzalez, A.M., Azevedo, M.S., Jung, K., Vlasova, A., Zhang, W., Saif, L.J., 2010. Innate immune responses to human rotavirus in the neonatal gnotobiotic piglet disease model. Immunology 131, 242–256. Gouvea, V., Glass, R.I., Woods, P., Taniguchi, K., Clark, H.F., Forrester, B., Fang, Z.Y., 1990. Polymerase chain reaction amplification and typing of rotavirus nucleic acid from stool specimens. J. Clin. Microbiol. 28, 276–282. Heath, T.A., Moore, B.R., 2014. Bayesian inference of species divergence times. In: Chen, M.-H., Kuo, L., Lewis, P.O. (Eds.), Bayesian Phylogenetics: Methods, Algorithms, and Applications. CRC Press, Taylor and Francis Group, Boca Raton, USA, pp. 277–318. Hoshino, Y., Saif, L.J., Kang, S.Y., Sereno, M.M., Chen, W.K., Kapikian, A.Z., 1995. Identification of group A rotavirus genes associated with virulence of a porcine rotavirus and host range restriction of a human rotavirus in the gnotobiotic piglet model. Virology 209, 274–280.

Hoshino, Y., Honma, S., Jones, R.W., Ross, J., Santos, N., Gentsch, J.R., Kapikian, A.Z., Hesse, R.A., 2005. A porcine G9 rotavirus strain shares neutralization and VP7 phylogenetic sequence lineage 3 characteristics with contemporary human G9 rotavirus strains. Virology 332, 177–188. Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17, 754–755. Kawai, K., O'Brien, M.A., Goveia, M.G., Mast, T.C., El Khoury, A.C., 2012. Burden of rotavirus gastroenteritis and distribution of rotavirus strains in Asia: a systematic review. Vaccine 30, 1244–1254. Komoto, S., Maeno, Y., Tomita, M., Matsuoka, T., Ohfu, M., Yodoshi, T., Akeda, H., Taniguchi, K., 2013. Whole genomic analysis of a porcine-like human G5P[6] rotavirus strain isolated from a child with diarrhoea and encephalopathy in Japan. J. Gen. Virol. 94, 1568–1575. Le, V.P., Chung, Y.C., Kim, K., Chung, S.I., Lim, I., Kim, W., 2010. Genetic variation of prevalent G1P[8] human rotaviruses in South Korea. J. Med. Virol. 82, 886–896. Leshem, E., Lopman, B., Glass, R., Gentsch, J., Banyai, K., Parashar, U., Patel, M., 2014. Distribution of rotavirus strains and strain-specific effectiveness of the rotavirus vaccine after its introduction: a systematic review and meta-analysis. Lancet Infect. Dis. 14, 847–856. Marthaler, D., Homwong, N., Rossow, K., Culhane, M., Goyal, S., Collins, J., Matthijnssens, J., Ciarlet, M., 2014. Rapid detection and high occurrence of porcine rotavirus A, B, and C by RT-qPCR in diagnostic samples. J. Virol. Methods 209, 30–34. Martinez, M., Galeano, M.E., Akopov, A., Palacios, R., Russomando, G., Kirkness, E.F., Parra, G.I., 2014. Whole-genome analyses reveals the animal origin of a rotavirus G4P[6] detected in a child with severe diarrhea. Infect. Genet. Evol. 27, 156–162. Matthijnssens, J., Ciarlet, M., Heiman, E., Arijs, I., Delbeke, T., McDonald, S.M., Palombo, E.A., Iturriza-Gomara, M., Maes, P., Patton, J.T., Rahman, M., Van Ranst, M., 2008. Full genome-based classification of rotaviruses reveals a common origin between human Wa-Like and porcine rotavirus strains and human DS-1-like and bovine rotavirus strains. J. Virol. 82, 3204–3219. Matthijnssens, J., Heylen, E., Zeller, M., Rahman, M., Lemey, P., Van Ranst, M., 2010a. Phylodynamic analyses of rotavirus genotypes G9 and G12 underscore their potential for swift global spread. Mol. Biol. Evol. 27, 2431–2436. Matthijnssens, J., Rahman, M., Ciarlet, M., Zeller, M., Heylen, E., Nakagomi, T., Uchida, R., Hassan, Z., Azim, T., Nakagomi, O., Van Ranst, M., 2010b. Reassortment of human rotavirus gene segments into G11 rotavirus strains. Emerg. Infect. Dis. 16, 625–630. Miyazaki, A., Kuga, K., Suzuki, T., Kohmoto, M., Katsuda, K., Tsunemitsu, H., 2013. Annual changes in predominant genotypes of rotavirus A detected in the feces of pigs in various developmental stages raised on a conventional farm. Vet. Microbiol. 163, 162–166. Mladenova, Z., Papp, H., Lengyel, G., Kisfali, P., Steyer, A., Steyer, A.F., Esona, M.D., IturrizaGomara, M., Banyai, K., 2012. Detection of rare reassortant G5P[6] rotavirus, Bulgaria. Infect. Genet. Evol. 12, 1676–1684. Mukherjee, A., Ghosh, S., Bagchi, P., Dutta, D., Chattopadhyay, S., Kobayashi, N., ChawlaSarkar, M., 2011. Full genomic analyses of human rotavirus G4P[4], G4P[6], G9P[19] and G10P[6] strains from North-eastern India: evidence for interspecies transmission and complex reassortment events. Clin. Microbiol. Infect. 17, 1343–1346. My, P.V.T., Rabaa, M.A., Donato, C., Cowley, D., Phat, V.V., Dung, T.T., Anh, P.H., Vinh, H., Bryant, J.E., Kellam, P., Thwaites, G., Woolhouse, M.E., Kirkwood, C.D., Baker, S., 2014. Novel porcine-like human G26P[19] rotavirus identified in hospitalized paediatric diarrhoea patients in Ho Chi Minh City, Vietnam. J. Gen. Virol. 95, 2727–2733. Nakagomi, T., Doan, Y.H., Dove, W., Ngwira, B., Iturriza-Gomara, M., Nakagomi, O., Cunliffe, N.A., 2013. G8 rotaviruses with conserved genotype constellations detected in Malawi over 10 years (1997–2007) display frequent gene reassortment among strains co-circulating in humans. J. Gen. Virol. 94, 1273–1295. Nei, M., Kumar, S., 2000. Molecular Evolution and Phylogenetics. Oxford University Press, Oxford, New York. Papp, H., Borzak, R., Farkas, S., Kisfali, P., Lengyel, G., Molnar, P., Melegh, B., Matthijnssens, J., Jakab, F., Martella, V., Banyai, K., 2013a. Zoonotic transmission of reassortant porcine G4P[6] rotaviruses in Hungarian pediatric patients identified sporadically ovver a 15 years. Infect. Genet. Evol. 19, 71–80. Papp, H., Laszlo, B., Jakab, F., Ganesh, B., De Grazia, S., Matthijnssens, J., Ciarlet, M., Martella, V., Banyai, K., 2013b. Review of group A rotavirus strains reported in swine and cattle. Vet. Microbiol. 165, 190–199. Pham, H.A., Carrique-Mas, J.J., Nguyen, V.C., Ngo, T.H., Nguyet, L.A., Do, T.D., Vo, B.H., Phan, V.T., Rabaa, M.A., Farrar, J., Baker, S., Bryant, J.E., 2014. The prevalence and genetic diversity of group A rotaviruses on pig farms in the Mekong Delta region of Vietnam. Vet. Microbiol. 170, 258–265. Rahman, M., Matthijnssens, J., Yang, X., Delbeke, T., Arijs, I., Taniguchi, K., Iturriza-Gomara, M., Iftekharuddin, N., Azim, T., Van Ranst, M., 2007. Evolutionary history and global spread of the emerging g12 human rotaviruses. J. Virol. 81, 2382–2390. Saif, L., Yuan, L., Ward, L., To, T., 1997. Comparative studies of the pathogenesis, antibody immune responses, and homologous protection to porcine and human rotaviruses in gnotobiotic piglets. Adv. Exp. Med. Biol. 412, 397–403. Saikruang, W., Khamrin, P., Chaimongkol, N., Suantai, B., Kongkaew, A., Kongkaew, S., Ushijima, H., Maneekarn, N., 2013. Genetic diversity and novel combinations of G4P[19] and G9P[19] porcine rotavirus strains in Thailand. Vet. Microbiol. 161, 255–262. Santos, N., Hoshino, Y., 2005. Global distribution of rotavirus serotypes/genotypes and its implication for the development and implementation of an effective rotavirus vaccine. Rev. Med. Virol. 15, 29–56. Sato, K., Inaba, Y., Shinozaki, T., Fujii, R., Matumoto, M., 1981. Isolation of human rotavirus in cell cultures: brief report. Arch. Virol. 69, 155–160. Sitkoff, D., Sharp, K.A., Honig, B., 1994b. Accurate calculation of hydration free energies using macroscopic solvent models. J. Phys. Chem. 98, 1978–1988.

L.P. Do et al. / Infection, Genetics and Evolution 40 (2016) 205–213 Steyer, A., Poljsak-Prijatelj, M., Barlic-Maganja, D., Jamnikar, U., Mijovski, J.Z., Marin, J., 2007. Molecular characterization of a new porcine rotavirus P genotype found in an asymptomatic pig in Slovenia. Virology 359, 275–282. Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6: molecular evolutionary genetics analysis version 6.0. Mol. Biol. Evol. 30, 2725–2729. Tate, J.E., Burton, A.H., Boschi-Pinto, C., Steele, A.D., Duque, J., Parashar, U.D., 2011. 2008 estimate of worldwide rotavirus-associated mortality in children younger than 5 years before the introduction of universal rotavirus vaccination programmes: a systematic review and meta-analysis. Lancet Infect. Dis. 12, 136–141. Teodoroff, T.A., Tsunemitsu, H., Okamoto, K., Katsuda, K., Kohmoto, M., Kawashima, K., Nakagomi, T., Nakagomi, O., 2005. Predominance of porcine rotavirus G9 in Japanese piglets with diarrhea: close relationship of their VP7 genes with those of recent human G9 strains. J. Clin. Microbiol. 43, 1377–1384. Theuns, S., Desmarets, L.M., Heylen, E., Zeller, M., Dedeurwaerder, A., Roukaerts, I.D., Van Ranst, M., Matthijnssens, J., Nauwynck, H.J., 2014. Porcine group A rotaviruses with heterogeneous VP7 and VP4 genotype combinations can be found together with enteric bacteria on Belgian swine farms. Vet. Microbiol. 172, 23–34.

213

Theuns, S., Heylen, E., Zeller, M., Roukaerts, I.D., Desmarets, L.M., Van Ranst, M., Nauwynck, H.J., Matthijnssens, J., 2015. Complete genome characterization of recent and ancient Belgian pig group A rotaviruses and assessment of their evolutionary relationship with human rotaviruses. J. Virol. 89, 1043–1057. Tsugawa, T., Tatsumi, M., Tsutsumi, H., 2014. Virulence-associated genome mutations of murine rotavirus identified by alternating serial passages in mice and cell cultures. J. Virol. 88, 5543–5558. Varghese, V., Ghosh, S., Das, S., Bhattacharya, S.K., Krishnan, T., Karmakar, P., Kobayashi, N., Naik, T.N., 2006. Characterization of VP1, VP2 and VP3 gene segments of a human rotavirus closely related to porcine strains. Virus Genes 32, 241–247. Wang, Y.H., Kobayashi, N., Nagashima, S., Zhou, X., Ghosh, S., Peng, J.S., Hu, Q., Zhou, D.J., Yang, Z.Q., 2010. Full genomic analysis of a porcine-bovine reassortant G4P[6] rotavirus strain R479 isolated from an infant in China. J. Med. Virol. 82, 1094–1102. Zeller, M., Heylen, E., De Coster, S., Van Ranst, M., Matthijnssens, J., 2012. Full genome characterization of a porcine-like human G9P[6] rotavirus strain isolated from an infant in Belgium. Infect. Genet. Evol. 12, 1492–1500.