Characterization of the Bujaru, Frijoles and Tapara Antigenic Complexes Into the Sandfly Fever Group and Two Unclassified Phleboviruses from Brazil

RESEARCH ARTICLE Nunes-Neto et al., Journal of General Virology 2017;98:585–594 DOI 10.1099/jgv.0.000724

Characterization of the Bujaru, frijoles and Tapara antigenic complexes into the sandfly fever group and two unclassified phleboviruses from Brazil

Joaquim Pinto Nunes-Neto,1† William Marciel de Souza,2,3† Gustavo Olszanski Acrani,4 Marilia Farignoli Romeiro,2 MarcílioJorge Fumagalli,2 Luiz Carlos Vieira,2 Daniele Barbosa de Almeida Medeiros,1 Juliana Abreu Lima,1 Clayton Pereira Silva de Lima,5 Jedson Ferreira Cardoso,5 Luiz Tadeu Moraes Figueiredo,2 Sandro Patroca da Silva,5 Robert Tesh,6 Marcio Roberto Teixeira Nunes1,5,6 and Pedro Fernando da Costa Vasconcelos1,7,*

Abstract The genus Phlebovirus includes the sandfly fever viruses and tick-transmitted uukuviruses. Sandfly fever group viruses have been isolated from various vertebrate species and from phlebotomines and occasionally alternative arthropods, e.g. mosquitoes, or ceratopogonids of the genus Culicoides. Uukuniemi serogroup viruses have been isolated from various vertebrate species and from ticks. Despite the public health importance of some viruses of the genus, the genomic diversity of phleboviruses that could be incriminated as causative of human or veterinary diseases remains underestimated. Here we describe the nearly complete sequences and genomic characterization of two phleboviruses belonging to the Bujaru antigenic complex: the prototype species and the Munguba virus. Furthermore, six previously unclassified phleboviruses isolated in Brazil were also sequenced and characterized: Ambe, Anhanga, Joa, Uriurana, Urucuri and Tapara viruses. The results of the phylogenetic analysis indicated that these viruses group with viruses of three antigenic complexes (Bujaru, Tapara and frijoles clades), with two unclassified phleboviruses. We also performed genomic reassortment analysis and confirmed that there were no events for the viruses described in this study, but we found a new potential reassortment in Medjerda Valley virus, which contains S and L segments of Arbia virus, and probably a unique M segment, both viruses circulate in the same geographic region, indicating these two isolates represent two distinct viruses. This study provides insights into the genetic diversity, classification and evolution of phleboviruses.

INTRODUCTION large segment (LRNA), which encodes the viral RNA- dependent RNA polymerase (RdRp); medium segment At present, more than 350 named bunyaviruses are subdi- (MRNA), which encodes two envelope glycoproteins (Gn vided into order Bunyavirales based on serological, morpho- and Gc) and a non-structural protein (NSm); and small seg- logical and genomic features [1]. Of these, the genus ment (SRNA), with an ambisense coding strategy, which Phlebovirus is composed of approximately 70 named encodes the viral nucleocapsid (N) protein in the negative- viruses, which, based on serological methods, are classified sense orientation and a non-structural protein (NSs) that is into two antigenic groups: the phlebotomus fever (or sand- encoded in the positive sense [2, 3]. fly fever) group and the Uukuniemi group [2]. The genus Phlebovirus includes the sandfly fever viruses and Phleboviruses possesses a genome organization typical of tick-transmitted uukuviruses. Sandfly fever group viruses other members of the order Bunyavirales, with a tripartite have been isolated from various vertebrate species and from negative-sense RNA genome with components named as phlebotomines and occasionally alternative arthropods, e.g.

Received 11 October 2016; Accepted 26 January 2017 Author affiliations: 1Departamento de Arbovirologia e Febres Hemorragicas, Instituto Evandro Chagas, Ananindeua, Para, Brazil; 2Virology Research Center, School of Medicine of Ribeirao Preto of University of Sao~ Paulo, Ribeirao Preto, Sao~ Paulo, Brazil; 3MRC-University of Glasgow Centre for Virus Research, Glasgow, Scotland, UK; 4Universidade Federal da Fronteira Sul, Passo Fundo, Rio Grande do Sul, Brazil; 5Center for Technological Innovation, Instituto Evandro Chagas, Ananindeua, Para, Brazil; 6Department of Pathology, University of Texas Medical Branch, Galveston Texas, USA; 7Department of Pathology, Para State University, Belem, Para, Brazil. *Correspondence: Pedro Fernando da Costa Vasconcelos, [email protected] Keywords: phleboviruses; genetic characterization; sandfly fever group; ungrouped phlebovirus; Brazilian Amazon. Abbreviations: CF, complement fixation; ML, maximum likelihood; RdRp, RNA-dependent RNA polymerase. †These authors contributed equally to this work. Three supplementary figures and three supplementary tables are available with the online Supplementary Material.

000724 ã 2017 The Authors Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66585 On: Tue, 03 Jul 2018 19:08:24 Nunes-Neto et al., Journal of General Virology 2017;98:585–594 mosquitoes, or ceratopogonids of the genus Culicoides [3– NSm with its sequence upstream from the Gn protein, a 5]. On the other hand, viral members of Uukuniemi sero- typical organization of some phleboviruses. The glycopro- complex have been isolated from various vertebrate species teins are predicted to contain transmembrane domains and from ticks and include three phleboviruses of public (TMDs) that are typical of the genus, which is a single TMD health importance: severe fever with thrombocytopenia syn- in Gn, close to the C-terminus of this protein and two drome (SFTS) virus, Heartland and Bhanja viruses [6–9]. TMDs in Gc, one in the N-terminal region and the other one close to the C-terminal region (Figs 2 and S2). Both Despite the public health importance of some viruses of the TMDs in Gn and Gc are responsible for anchoring the gly- genus Phlebovirus, and the recent number of discoveries coproteins on the viral envelope [15]. The asparagine sites and sequencing data of phleboviruses that had been only that are predicted to be N-glycosylated in the M polyprotein previously described through classical virological methods, were also observed, and neither the number of sites nor the genomic diversity of phleboviruses associated with their positions seem to be conserved in these viruses, as can human or veterinary diseases remains underestimated [9– be observed in Fig. 1. These N-glycosylation sites of the pre- 14]. Therefore, it is important to clarify the genetic diversity, dicted protein topologies are located in the polyprotein por- distribution and evolutionary aspects of these viruses. Here, tion, which faces the lumen of the endoplasmic reticulum we describe the nearly complete sequences and characteriza- and of the Golgi apparatus, which could indicate that they tion of the two viruses of the Bujaru antigenic complex, as could be glycosylated during viral replication. However, fur- well as six unassigned phleboviruses isolated in Brazil. ther investigation focusing on biological processes of replication is needed in order to address whether they are RESULTS actually being glycosylated in vivo. Genomic characterization of phlebovirus The SRNA segments of these viruses display the same The genomes of Bujaru, Ambe, Anhanga, Joa, Munguba, encoding strategies, with two ORFs in different directions: Uriurana, Urucuri and Tapara viruses (Table 1) comprise a negative-sense N protein and a positive-sense non- three negative-sense RNA molecules with sizes ranging structural protein (NSs). The sizes of the NP range from from 1628 (SRNA) to 6510 (LRNA) nucleotides with 244 to 250 aa, and those of the NSs between 249 and 289 aa. deduced amino acid lengths similar to those of other phleboviruses (Fig. 1). The L segment of these viruses encodes Phylogenetic and serological relationship of an RdRp of about 2090 aa, with a predicted molecular phleboviruses weight of 238 kDa. Tapara virus contains the largest putative The maximum-likelihood (ML) trees based on nucleotide RdRp, comprising 2120 aa with a predicted molecular or amino acid sequences produced the same clustering pat- weight of 24 165 kDa, and the conserved polymerase activity tern of groups, but with different topologies (Figs 2 and domains consisting of pre-motif A and motifs A through E S3). This fact indicates that the saturation site has not (amino acids 915–922 to 1182–1189), which is conserved adversely affected the phylogenetic inference. The 85 full- in negative sense RNA viral polymerases [Figs 1 and length phlebovirus ORFs were clustered into 18 well- S1 (available in the online Supplementary Material)]. The M supported monophyletic groups; however, many of the polyprotein encoded by the MRNA segment ranges from deeper nodes were unresolved throughout the phylogeny. 1302 to 1429 aa in length, with differences distributed across Eleven of the well-supported clades corresponded to viral the polyprotein sequences. Furthermore, the M polyprotein species that have already been established by the ICTV: contains the characteristic N-terminal signal peptide that is Bujaru, Chagres, candiru, frijoles, Icoaraci, Punta Toro, common to all members of the genus (Fig. 1). The M poly- Rift Valley, Salehabad, sandfly fever Naples, severe fever protein is cleaved into three structural proteins, NSm, Gn with thrombocytopenia syndrome (SFTS) and Uukuniemi. and Gc. All phleboviruses examined in this study possess an Additionally, we found groups that had been previously

Table 1. Names, abbreviations, strain numbers, dates, sources, localities of isolation and accession numbers of the viruses used in this study

Virus name Abbreviation Strain Date of Source of isolate Location Accession numbers isolation

Ambe virus AMBEV BeAr407981 01/14/1982 Phlebotominae sp. Altamira, Para State, Brazil KX611382 to KX611384 Anhanga virus ANHV BeAn46852 10/01/1962 Choloepus brasiliensis Castanhal Forest, Para State, KX611385 to KX611387 Brazil Bujaru virus BUJV BeAn47693 10/26/1962 Proechimys guyannensisoris Utinga Forest, Para State, Brazil KX611388 to KX611390 Joa virus JOAV BeAr371637 03/29/1979 Lutzomyia sp, Altamira, Para State, Brazil KX611391 to KX611393 Munguba virus MUNV BeAr389707 09/20/1980 Lutzomyia umbratilis Almerim, Para State, Brazil KX611394 to KX611396 Tapara virus TAPV BeAn413570 02/04/1983 Phlebotominae sp. Altamira, Para State, Brazil KX611397 to KX611399 Uriurana virus URIV BeAr479776 12/16/1985 Phlebotominae sp. Tucuruí, Para State, Brazil KX611400 to KX611402 Urucuri virus URUV BeAn100049 04/19/1966 Proechimys guyannensis Utinga Forest, Para State, Brazil KX611403 to KX611405

Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66586 On: Tue, 03 Jul 2018 19:08:24 Nunes-Neto et al., Journal of General Virology 2017;98:585–594 (f) (f) (e) (g) (a) (c) (d) (b)

Fig. 1. Schematic view of the genome organization of phlebovirus sequenced reported.

Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66587 On: Tue, 03 Jul 2018 19:08:24 Nunes-Neto et al., Journal of General Virology 2017;98:585–594

(a) (b)

Fig. 2. Maximum-likelihood phylogenetic trees of members of the genus Phlebovirus. Coding sequences of N protein (a), coding sequences of NSs protein (b), M segment (c) and L segment (d). Branches are colour-coded according to group. Horizontal branch lengths are drawn to a scale of nucleotide substitutions per site and according to the approximate likelihood ratio test (aLRT). Phleboviruses sequenced in this study are labelled in red.

Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66588 On: Tue, 03 Jul 2018 19:08:24 Nunes-Neto et al., Journal of General Virology 2017;98:585–594 reported, such as Bhanja, sandfly Sicilian Turkey, and Kar- Pairwise amino acid sequence distance analysis of NP, imabad groups [4, 8, 14, 16]. NSs, glycoprotein and RdRp of eight phleboviruses sequenced in our study along with selected phleboviruses Within the phleboviruses sequenced in this study, the phy- was conducted, and the results of an inter- and intra-clade logenetic tree based on four genes (RdRp, glycoprotein, NP analysis are depicted in Fig. 3 and Table 2. Notably, we and NSs) and CF results reveals that Bujaru and Munguba observed that the phleboviruses transmitted by ticks pre- viruses compose the Bujaru antigenic complex (Fig. 2 and sented a higher divergence rate at the amino acid level Table S2). Interestingly, the results of the phylogenetic anal- than those phleboviruses transmitted by sandflies and ysis indicated that Anhanga virus is a potential member of mosquitoes. All sequenced viruses in this study have simi- Bujaru complex, because this virus shared the same com- lar divergence rates at the amino acid level to those of mon ancestor as this group for both the L and M RNA seg- phleboviruses transmitted by sandflies and mosquitoes. ments (Fig. 2). However, this was not observed for the S However, two exceptions have been observed in the segment and, the complement fixation (CF) test revealed sequence distance analysis of the NSs protein: Rift Valley that Anhanga virus did not present a cross-serological reac- fever and sandfly Sicilian Turkey viruses possess similar tion with other members of the Bujaru antigenic complex divergence rates at the amino acid level to those of phlebo- (Fig. 2 and Table S3). viruses transmitted by ticks (Fig. 3). The evolutionary divergence within phlebovirus groups based on all genes The Uriurana and Tapara viruses form a unique and well- indicated that the groups with lower divergence are Rift established clade closer to the Chagres virus. This result is Valley fever, Aguacate and candiru complexes and the consistent among the four analysed genes with a high statis- major divergences are observed for the tick-borne phlebo- tical support (>97 %), as well as the serological test, which viruses (Table 2). Interestingly, the results observed in ML supports the same conclusion, indicating the basis of a con- trees are similarly found in the evolutionary distance anal- sistent serological relationship between Uriurana and ysis, especially when comparing the phleboviruses based Tapara, which form the clade named as Tapara (Fig. 2 and on the vector (Figs 2 and 3). Table S2). Furthermore, Joa virus forms a unique clade with Salobo virus, indicating that they shared the same common Potential reassortment event analysis ancestor. However, based on antigenic analysis, the Salobo Branching inconsistencies observed in phylogenetic trees virus is a member of the Icoaraci complex, as well the Bel- indicated the possibility of natural reassortment phenom- terra and Icoaraci virus; both these viruses do not yet have ena. Thus, in order to identify the potential reassortment the complete genome sequenced. In addition, the Joa virus events between the sequenced phleboviruses in this report is a member of frijoles complex; the prototype of this com- within the complete coding sequences from other phlebovi- plex also does not yet have its complete genome available ruses, we inspected the nucleotide ML phylogenies for dis- (Fig. 2). cordances in clade clustering between coding sequences of Interestingly, the other two unassigned phleboviruses pre- N, NSs, M and L trees (Fig. 2), together with RDP4 analyses using concatenated coding sequences of the genomes of all sented distinct topology in phylogenetic trees and are not 85 phleboviruses included in this study. From this analysis, clustered in specific clades. The Ambe virus showed a dif- we found a new potential reassortment for Medjerda Valley ferent topology for all segments at both the nucleotide and virus, which has been recently reported [17]; apparently, amino acid levels. The nucleoprotein sequence of Ambe this virus contains S and L segments from Arbia virus, and virus is closely related to the Rift Valley fever group, but an M segment of unknown origin, probably unique not that of the NSs gene, which also presents a different (Table S1). However, we did not find potential reassortment topology in the phylogenetic tree. On the other hand, the events in phleboviruses sequenced in our study. These results of the phylogenic analysis at the nucleotide and results, supported by RDP4 software, indicate that the amino acids levels of M and L segments indicated that sequences of phleboviruses described here are not reassort- Ambe virus shares the same evolutionary origin as the ment candidates. However, additional studies may help to Aguacate/Urucuri and frijoles/Icoaraci complexes, respec- clarify this point. tively (Figs 2 and S3). Likewise, the Urucuri virus possesses a nucleocapsid nucleotide coding sequence phylogenetically DISCUSSION more closely related to that of the Anhanga virus, while its Despite the efforts to understand the diversity in the NSs protein is more closely related to that of Aguacate virus genus Phlebovirus, until now, these viruses have been pri- (Figs 2 and S3). The M segment shared the same evolution- marily classified mainly by serological studies. However, it ary origin as that of Aguacate virus and the L segment is no longer feasible to test newly isolated phleboviruses by forms a unique clade among the sandfly fever Naples and serology against all other known members of the genus sandfly Sicilian Turkey complexes (Fig. 2). Furthermore, because of their abundance and diversity [10, 11, 13]. More- the CF results revealed that Ambe virus and Urucuri virus over, some phleboviruses do not produce readable plaques are serologically distinct from other phleboviruses (Fig. 2 in cell culture nor cause illness in newborn mice [10]. On and Table S3). the other hand, during the last decade, due to the

Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66589 On: Tue, 03 Jul 2018 19:08:24 Nunes-Neto et al., Journal of General Virology 2017;98:585–594 (b) (d) (a) (c)

Fig. 3. Pairwise genetic similarities (amino acid p-distance) among groups of phlebovirus based on the coding sequencing of N protein (a), NSs protein (b), glycoprotein (c) and RdRp (d). The error bars represent the whisker top and whisker bottom.

Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66590 On: Tue, 03 Jul 2018 19:08:24 Nunes-Neto et al., Journal of General Virology 2017;98:585–594

Table 2. Estimates of mean evolutionary divergence over sequence pairs within phlebovirus groups The number of amino acid differences per site from averaging over all sequence pairs within each clade. n/c, Cases in which it was not possible to estimate evolutionary distances.

Group Nucleoprotein Non-Scrutural Glycoprotein Polymerase

Bujaru 0.255 0.533 0.379 0.262 Chagres 0.272 0.433 0.424 0.242 Candiru 0.106 0.359 0.298 0.145 Punta Toro 0.176 0.428 0.279 0.207 Ambe n/c n/c n/c n/c Rift Valley 0.007 0.400 0.013 0.009 Aguacate 0.051 0.100 0.281 0.108 Frijoles 0.321 0.600 0.467 0.316 Karimabad 0.359 0.600 0.278 0.179 Salehabad 0.301 0.343 0.403 0.225 Sandfly Sicilian Turkey 0.152 0.600 0.344 0.195 Urucuri n/c n/c n/c n/c Sandfly fever Naples 0.156 0.529 0.318 0.152 Silverwater 0.310 0.600 0.331 0.201 Bhanja 0.220 0.640 0.307 0.180 SFTS 0.219 0.467 0.226 0.170 Uukuniemi 0.223 0.643 0.294 0.161 availability of new tools, the field of virus taxonomy has The phylogenetic analyses and CF results revealed Uriurana been moving toward a sequence-based taxonomic classifica- and Tapara viruses to be members of an expanded Tapara tion, in particular by the advancement of high-throughput complex within the genus Phlebovirus. Therefore, we sug- sequencing, which offers an option for characterization and gest that the Tapara complex is a novel antigenic complex classification of these viruses [9]. Based on this perspective, within sandfly fever virus, with Uriurana and Tapara we have sequenced the genomes of known phleboviruses to viruses, two previously unclassified phleboviruses. Further- determine their taxonomic relationships; this is one of the more, the phylogenetic relationship between the Tapara several planned publications of our findings. complex and Chagres virus indicates that Uriurana and Tapara viruses may also be involved in human infections Our study suggests the classification of Bujaru, Munguba, [21, 22]. Ambe, Anhanga, Joa, Uriurana, Urucuri and Tapara phleboviruses into three complexes: the Bujaru, Tapara and fri- Salobo virus is a new and previously unclassified member of joles clades. In the Bujaru complex, composed of Bujaru the serogroup, but based on our results, we propose that the and Munguba viruses, the results of phylogenetic analyses Salobo virus should be included in the Icoaraci complex. are in agreement with those of the serological tests, which Furthermore, a previous study has reported the relationship have also been used by the ICTV to classify these viruses between Joa and frijoles viruses in a unique antigenic com- into a unique viral species, the Bujaru phlebovirus [18]. The plex using a complement fixation test, which was confirmed CF assay is an important and valid serological method in by our findings [23]. Considering that a complete coding virology, but these results should be interpreted with sequence for frijoles virus is not available, correlation using caution, especially because cross-reactivity can occur at a both serological and genomic classification was not possible. frequency higher than in haemagglutination and neutraliza- Furthermore, we suppose that sequencing of a larger num- tion assays [9, 19]. ber of phleboviruses, especially viruses classified using serological methods from which a complete genome sequence Despite the Anhanga virus sharing the same common has not been obtained, could further clarify this point. ancestor with Bujaru phlebovirus for the L and M segments, we did not find cross-reactivity in serological tests for these The Ambe and Urucuri viruses are unclassified phlebovi- viruses. Therefore, we believe that Anhanga virus is not a ruses isolated in Brazil. Ambe virus strain BeAr407981 was member of the Bujaru antigenic complex. In addition, these originally isolated from members of the family Phlebotomi- viruses have not been associated with human or veterinary nae collected in a rural area of the municipality of Altamira diseases, but isolates have been reported in sandflies, in Para State, Brazil in 1982. Urucuri virus strain such as Lutzomyia umbratilis, and in mammals, such as BeAr413570 was also isolated from Proechimys guyannensis Proechimys guyannensis oris and Choloepus brasiliensis cap- in 1966 in Utinga Forest, Para State, Brazil. Both viruses tured in Brazil [20]. have not been associated with human or veterinary diseases.

Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66591 On: Tue, 03 Jul 2018 19:08:24 Nunes-Neto et al., Journal of General Virology 2017;98:585–594

Results of phylogenetic analysis and serological tests indi- BeAr371637, Urucuri strain BeAn100049 and Uriurana cated that these viruses do not have a relationship with any strain BeAr479776 were prepared for high-throughput other established antigenic complex. Differences in topolo- sequencing using the pyrosequencing approach [27]. A gies for all RNA segments were observed for these viruses. pyrosequencing library was prepared and used for Therefore, in the light of our results these viruses remain as sequencing on a GS FLX+ pyrosequencer (Roche) at the ungrouped phleboviruses. Center for Technological Innovation in the Evandro Cha- gas Institute, Ministry of Health, Brazil. The cDNA of Here we have shown the genetic characterization and evolu- Bujaru strain BeAn47693 was prepared for high-through- tionary relationships of three antigenic complex (Bujaru, put sequencing using a RAPID module with the TruSeq Tapara and frijoles) and two unclassified phleboviruses Universal adapter (Illumina) and standard multiplex adap- (Ambe and Urucuri viruses) among members of the tors. A paired-end, 150-base-read protocol of the RAPID genus Phlebovirus. Genomic reassortment events were con- module was used for sequencing on a HiSeq 2500 instru- firmed in our analysis, but were not observed for the viruses ment (Illumina) as recommended by the manufacturer’s described in this study [10, 11, 24]. Furthermore, we found protocol. Sequencing was performed at the Life Sciences a novel potential reassortment for Medjerda Valley virus, Core Facility (LaCTAD) of the State University of Campi- which contains S and L segments nearly identical to those of nas (UNICAMP), Brazil. the Arbia virus, and probably a unique M segment, both viruses circulating in the same geographic region [17]. In To assemble Bujaru genomes, reads were first quality-fil- summary, our study provides knowledge of the genetic tered (quality score Q30) using the program FastQC diversity, classification and evolution of phleboviruses. v. 0.11.3 and any adapter sequences were removed using Trim_Galore, Prinseq and Ribopicker [28]. Reads were METHODS assembled by the de novo strategy into contigs using IDBA – Viruses, viral culture and RNA purification UD-1.1.1, SpADES and GARM [29 31]. The genomes of Anhanga, Ambe, Tapara, Munguba, Joa, Urucuri and Uriur- The viral strains used in this study were propagated in Vero ana were assembled using the GS De NovoAssembler pro- cells (CCL81), except the Bujaru virus, which was propa- gram (Newbler v. 3.0).The largest contigs were submitted to gated in C6/36 Aedes albopictus cells as previously described BLAST-based searches to identify possible virus genomes. [8, 25]. The infected cells were incubated for seven to ten Subsequently, the annotations of putative ORF genes were days until visualization of approximately 80 % cytopathic predicted using Geneious 9.1.2 (Biomatters). effect. Then, the viral RNA was extracted using the QIAamp viral RNA extraction kit (Qiagen) as recommended by Prediction of protein domains and functions the manufacturer. Table 1 lists the names, abbreviations, Each of the predicted proteins in Geneious 9.1.2 (Biomat- strain numbers, date, sources and locale of isolation, as well ters) were submitted to TOPCONS webserver [32] for iden- as GenBank accession numbers. tification of transmembrane regions and signal peptide and Serological tests to the NetNglyc 1.0 Server (www.cbs.dtu.dk/services/NetN- Glyc/) for identification of glycosylation sites. The annota- The complement fixation (CF) test was performed by the tions of protein domains in the M segment were performed microtechnique [26]. Sera were inactivated at 60 C for with InterProScanin Geneious 9.1.2 (Biomatters) and Con- 20 min and mixed with test antigen and 2 full units of com- served Domain Database [33]. The protein molecular plement. This first phase of the test was incubated overnight weight for each identified viral protein was predicted at 4 C, after which time the haemolytic system (haemolysin using the Protein Molecular Weight Calculator tool (www. and sheep blood red cells) was added. After 30 min of incu- sciencegateway.org/tools/proteinmw.htm). bation at 37 C and 60 min at 4 C, the plates were then read visually. If the complement is tied up in the first antigen– Phylogenetic analysis antibody reaction, it will not be available for the sheep red Phylogenetic trees were reconstructed using our phlebovi- – blood cell haemolysin reaction and there will be no hae- ruses’ sequences and additionally 77 complete phlebovirus molysis. A negative test would result in haemolysis. CF titres coding sequences (S, M and L) available in the GenBank were recorded as the reciprocals of the highest antibody/ database (www.ncbi.nlm.nih.gov/), yielding a final dataset highest antigen dilutions giving a fixation of complement of 85 viruses. The viral sequences were aligned using the E- value of 3 or 4 on a scale of 0 (complete haemolysis, nega- INS-i algorithm implemented in the MAFFT version 7.158b tive) to 4 (no haemolysis, positive). program [34]. Genome sequencing and assembly Maximum likelihood (ML) phylogenies for all segments The RNA strand synthesis was performed using a cDNA were inferred using PhyML 3.0 software, employing NNI Synthesis System kit and 400 µM Roche Primer ‘random’. (nearest neighbour interchanges) and a general time- The reaction was purified with Agencourt AMPure XP reversible (GTR) model of nucleotide substitution with Reagent (Beckman Coulter). The cDNAs of Anhanga among-site rate heterogeneity parameter (g) and a propor- strain BeAn46852, Ambe strain BeAr407981, Tapara strain tion of invariant sites (I), as determined by the jModelTest BeAr413570, Munguba strain BeAr389707, Joa strain version 2.1.10 program [35, 36].

Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66592 On: Tue, 03 Jul 2018 19:08:24 Nunes-Neto et al., Journal of General Virology 2017;98:585–594

The predicted viral proteins identified in this study were References aligned to complete coding sequences of all phleboviruses 1. Dietzgen RG, Calisher CH, Kurath G, Kuzman IV RLL, Stone DM et al. Virus taxonomy. In: King AMQ, Adams MJ, Carstens using MAFFT software version 7.158b [34]. Amino acid EB and Lefkowitz EJ (editors). Ninth Report of the International sequences were inferred using the LG model [37] substitu- Committe on Taxonomy of Viruses. San Diego, CA, USA: Elsevier; tion models in PhyML 3.0, as determined by the ProtTest 2012. pp. 654–681. version 3.4.2 program [38]. Statistical support for individual 2. Elliott RM, Schmaljohn C. Bunyaviridae. In: Knipe DM and Howley PM (editors). Fields Virology Volume 2. pp. 1244–1282. 6th ed. Phil- nodes was estimated using the approximate likelihood ratio adelphia, PA: Lippincott Williams and Wilkins; 2013. test (aLRT) available in PhyML [39]. 3. Elliott RM, Brennan B. Emerging phleboviruses. Curr Opin Virol Genetic distance and prediction of conserved 2014;5:50–57. motifs 4. Alkan C, Bichaud L, De Lamballerie X, Alten B, Gould EA et al. Sandfly-borne phleboviruses of Eurasia and Africa: epidemiology, The genetic distances among and within clades were calcu- genetic diversity, geographic range, control measures. Antiviral lated based on amino acid alignments for four genes (nucle- Res 2013;100:54–74. ocapsid protein, small non-structural protein, glycoprotein 5. Tesh R. Sandfly fever, Oropouche fever, and other bunyavirus infections. In: Guerrant RL, Walker DH and Weller PF (editors). and RdRp) using p-distance values. Standard error estima- Tropical Infectious Diseases: Principles, Pathogens, and Practice, tions were calculated by the bootstrapping method (1000 2nd ed. London, UK: Elsevier Churchill Livingstone; 2006. pp. 781– replicates) using the MEGA v. 6 software [40]. The results for 783. the genetic distances among clades are presented as box and 6. Zhao L, Zhai S, Wen H, Cui F, Chi Y et al. Severe fever with throm- whisker plots, and genetic distances within clades are shown bocytopenia syndrome virus, Shandong Province, China. Emerg Infect Dis 2012;18:963–965. in a table form. The nucleotide and amino acid identity 7. Savage HM, Godsey MS, Lambert A, Panella NA, Burkhalter KL comparisons for our phlebovirus sequences and also of rep- et al. First detection of heartland virus (Bunyaviridae: Phlebovirus) resentative phleboviruses were performed in Geneious 9.1.2 from field collected arthropods. Am J Trop Med Hyg 2013;89:445– (Biomatters). Potential motifs characteristic of phlebovi- 452. ruses were identified using Geneious 9.1.2 (Biomatters). 8. Matsuno K, Weisend C, Travassos da Rosa AP, Anzick SL, Dahlstrom E et al. Characterization of the Bhanja serogroup Detection of reassortment events viruses (Bunyaviridae): a novel species of the genus Phlebovirus and its relationship with other emerging tick-borne phleboviruses. In order to identify potential reassortment events, the data J Virol 2013;87:3719–3728. were mined for evidence of distinct phylogenetic topologies 9. Palacios G, Savji N, Travassos da Rosa A, Guzman H, Yu X et al. based on the depicted trees at the nucleotide level, as Characterization of the Uukuniemi virus group (Phlebovirus: described above. Also, we concatenated all genes in a single Bunyaviridae): evidence for seven distinct species. J Virol 2013;87: 3187–3195. sequence and performed multiple alignments using the 10. Palacios G, da Rosa AT, Savji N, Sze W, Wick I et al. Aguacate program MAFFT v. 7.158b, as described above. Potential reas- virus, a new antigenic complex of the genus Phlebovirus (family sortment events were then analysed using RDP, GENE- Bunyaviridae). J Gen Virol 2011;92:1445–1453. CONV, Bootscan, MaxChi, Chiamera, SiScan and 3Seq 11. Palacios G, Tesh R, Travassos da Rosa A, Savji N, Sze W et al. methods implemented in RDP4 [41]. Standard program set- Characterization of the Candiru antigenic complex (Bunyaviridae: tings for all methods were used to identify sequences as lin- Phlebovirus), a highly diverse and reassorting group of viruses affecting humans in tropical America. J Virol 2011;85:3811–3820. ear, to obtain phylogenetic evidence, to refine breakpoints 12. Palacios G, Wiley MR, Travassos da Rosa AP, Guzman H, Quiroz and to check alignment consistency. The highest acceptable E et al. Characterization of the Punta Toro species complex (genus P value was set at 0.05, after considering Bonferroni correc- Phlebovirus, family Bunyaviridae). J Gen Virol 2015;96:2079–2085. tion for multiple comparisons. All method-specific program 13. Palacios G, Savji N, Travassos da Rosa A, Desai A, Sanchez-Seco settings remained at their default values. MP et al. Characterization of the Salehabad virus species complex of the genus Phlebovirus (Bunyaviridae). J Gen Virol 2013;94:837– 842. 14. Palacios G, Tesh RB, Savji N, Travassos da Rosa AP, Guzman H Funding information et al. Characterization of the Sandfly fever Naples species com- This work was supported by the Evandro Chagas Institute funds, Con- plex and description of a new Karimabad species complex (genus selho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) Phlebovirus, family Bunyaviridae). J Gen Virol 2014;95:292–300. grants (301641/2010-2), and by the Fundaçao~ de Amparo a Pesquisa do Estado de Sao~ Paulo, Brazil (grant numbers 13/14929-1 and 14/ 15. Halldorsson S, Behrens AJ, Harlos K, Huiskonen JT, Elliott RM 02438-6, and scholarships numbers 12/24150-9, 15/05778-5, 14/ et al. Structure of a phleboviral envelope glycoprotein reveals a 20851-8, 16/01414-1 and 14/15079-4), Universidade Federal do Para consolidated model of membrane fusion. Proc Natl Acad Sci USA and Coordenaçao~ de Aperfeiçoamento de Pessoal de Nível 2016;113:7154–7159. Superior (CAPES). 16. Carhan A, Uyar Y, Ozkaya E, Ertek M, Dobler G et al. Characteriza- tion of a sandfly fever Sicilian virus isolated during a sandfly fever Acknowledgements epidemic in Turkey. J Clin Virol 2010;48:264–269. We thank Amelia P. A. Travassos da Rosa from University of Texas 17. Bichaud L, Dachraoui K, Alwassouf S, Alkan C, Mensi M et al. Iso- Medical Branch, Galveston Texas, USA for performing the serological lation, full genomic characterization and neutralization-based tests and for useful comments, which helped to improve this human seroprevalence of Medjerda Valley virus, a novel sandfly- manuscript. borne phlebovirus belonging to the Salehabad virus complex in northern Tunisia. J Gen Virol 2016;97:602–610. Conflicts of interest 18. Briese T, Alkhovsky S, Beer M, Calisher CH, Remi C et al. Imple- The authors declare that there are no conflicts of interest. mentation of non-Latinized binomial species names in the family

Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66593 On: Tue, 03 Jul 2018 19:08:24 Nunes-Neto et al., Journal of General Virology 2017;98:585–594

Bunyaviridae. International Committee on Taxonomy of Viruses 31. Soto-Jimenez LM, Estrada K, Sanchez-Flores A. GARM: genome 2015. Code assigned: 2015.003aM. assembly, reconciliation and merging pipeline. Curr Top Med Chem 19. Bishop DH, Calisher CH, Casals J, Chumakov MP, Gaidamovich SY 2014;14:418–424. et al. Bunyaviridae. Intervirology 1980;14:125–143. 32. Tsirigos KD, Peters C, Shu N, Kall€ L, Elofsson A. The TOPCONS 20. Tesh RB. The genus Phlebovirus and its vectors. Annu Rev Entomol web server for consensus prediction of membrane protein 1988;33:169–181. topology and signal peptides. Nucleic Acids Res 2015;43:W401– 21. Srihongse S, Johnson CM. Human infections with Chagres virus W407. in Panama. Am J Trop Med Hyg 1974;23:690–693. 33. Marchler-Bauer A, Derbyshire MK, Gonzales NR, Lu S, Chitsaz F et al. 22. Peralta PH, Shelokov A, Brody JA. Chagres virus: a new human CDD: NCBI’s conserved domain database. Nucleic Acids Res isolate from Panama. Am J Trop Med Hyg 1965;14:146–151. 2015;43:D222–D226. 23. Xu F, Liu D, Nunes MR, da Rosa AP, Tesh RB et al. Antigenic and 34. Katoh K, Standley DM. MAFFT multiple sequence alignment soft- genetic relationships among Rift Valley fever virus and other ware version 7: improvements in performance and usability. Mol selected members of the genus Phlebovirus (Bunyaviridae). Am J Biol Evol 2013;30:772–780. Trop Med Hyg 2007;76:1194–1200. 35. Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W et al. 24. Amaro F, Hanke D, Ze-Ze L, Alves MJ, Becker SC et al. Genetic New algorithms and methods to estimate maximum-likelihood characterization of Arrabida virus, a novel phlebovirus isolated in phylogenies: assessing the performance of PhyML 3.0. Syst Biol South Portugal. Virus Res 2016;214:19–25. 2010;59:307–321. 25. Zamree I, Drakes N, Rohani A, Lee HL. Sensitivity of Aedes albo- 36. Darriba D, Taboada GL, Doallo R, Posada D. jModelTest 2: more pictus C6/36 cells line for the detection and infectivity titration of models, new heuristics and parallel computing. Nat Methods 2012; dengue virus. Trop Biomed 2005;22:217–219. 9:772. 26. Beaty BJ, Calisher CH, Shope RE. Arboviruses. In: Schmidt 37. Le SQ, Gascuel O. An improved general amino acid replacement NJ and Emmons RW (editors). Diagnostic Procedures for Viral, matrix. Mol Biol Evol 2008;25:1307–1320. Rickettsial and Chlamydial Infections. Washington, DC: American 38. Darriba D, Taboada GL, Doallo R, Posada D. ProtTest 3: fast selec- Public Health Association; 1989. pp. 797–855. tion of best-fit models of protein evolution. Bioinformatics 2011;27: 27. Shendure J, Ji H. Next-generation DNA sequencing. Nat Biotechnol 1164–1165. 2008;26:1135–1145. 39. Anisimova M, Gascuel O. Approximate likelihood-ratio test for 28. Schmieder R, Lim YW, Edwards R. Identification and removal of branches: a fast, accurate, and powerful alternative. Syst Biol ribosomal RNA sequences from metatranscriptomes. 2006;55:539–552. Bioinformatics 2012;28:433–435. 40. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: 29. Peng Y, Leung HC, Yiu SM, Chin FY. IDBA-UD: a de novo assembler for single-cell and metagenomic sequencing data with highly molecular evolutionary genetics analysis version 6.0. Mol Biol Evol uneven depth. Bioinformatics 2012;28:1420–1428. 2013;30:2725–2729. 30. Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M et al. 41. Martin DP, Murrell B, Golden M, Khoosal A, Muhire B. RDP4: SPAdes: a new genome assembly algorithm and its applications detection and analysis of recombination patterns in virus to single-cell sequencing. J Comput Biol 2012;19:455–477. genomes. Virus Evol 2015;1:5.

Five reasons to publish your next article with a Microbiology Society journal 1. The Microbiology Society is a not-for-profit organization. 2. We offer fast and rigorous peer review – average time to first decision is 4–6 weeks. 3. Our journals have a global readership with subscriptions held in research institutions around the world. 4. 80% of our authors rate our submission process as ‘excellent’ or ‘very good’. 5. Your article will be published on an interactive journal platform with advanced metrics.

Find out more and submit your article at microbiologyresearch.org.

Downloaded from www.microbiologyresearch.org by IP: 200.139.9.66594 On: Tue, 03 Jul 2018 19:08:24