INTRODUCTION
The Bunyaviridae family is divided into five genera; Orthobunyavirus, Phlebovirus, Nairovirus, Hantavirus and Tospovirus [Reference Calisher1]. The largest genus, Orthobunyavirus, is composed of over 150 viruses that infect humans and are transmitted by mosquitoes, midges and ticks from reservoir animals like rodents and livestock. Members of Orthobunyavirus are tri-segmented, negative-sense RNA viruses responsible for mild to severe human and animal diseases. The L (large) segment encodes a large protein that consists of the RNA-dependent RNA polymerase activity for replication and transcription of genomic RNA segments. The M (medium) segment encodes a precursor polypeptide which yields the virion surface glycoproteins Gn and Gc and a non-structural protein NSm, and the S (small) segment encodes the nucleocapsid (NC) and a non-structural protein (NSs) in overlapping reading frames [Reference Soldan and Gonzalez-Scarano2]. Bunyamwera virus is the prototype virus of the Orthobunyavirus genus as well as the Bunyaviridae family of arboviruses. Bunyamwera virus is associated with febrile illness, headache, arthralgia, rash and infrequent central nervous system involvement [Reference Gonzalez, Georges and Monath3]. Viruses of the Orthobunyavirus genus were generally not associated with haemorrhagic symptoms until Ngari virus was implicated in haemorrhagic fever outbreaks in Kenya and Somalia, and retrospectively in Sudan [Reference Bowen4–Reference Gerrard6]. Ngari virus was determined to be a reassortment between two segmented viruses (Bunyamwera and Batai) co-circulating within the same environment. Subsequent analysis of isolated sequences from outbreak samples showed that the L and S segment sequences closely matched those of Bunyamwera virus while the M segment was identical to that of Batai virus [Reference Briese5].
Surveillance activities during Rift Valley fever outbreaks have demonstrated co-circulation of arboviruses including Bunyamwera virus [Reference Crabtree7]. Additionally, ongoing inter-epidemic surveillance activities have indicated continued intense transmission of Bunyamwera virus [Reference Lutomiah8]. Co-circulation of viruses within the same serogroup is likely to provide opportunities for genetic reassortments. Efforts to bring these infections under control and predict their emergence will not be successful unless there is a full understanding of how these viruses are maintained and transmitted within the environment. However, characterization of these emergent arboviral species has been hampered by paucity of genetic sequences making it impossible to accurately estimate their evolutionary trend and public health burden. We have recently isolated Bunyamwera and Ngari viruses from surveillance exercises in Kenya and identified them based on short diagnostic sequences [Reference Lutomiah8, Reference Ochieng9]. Our main objective was to provide complete coding sequence of some of these isolates as well as to investigate their phylogenetic positioning within the Bunyamwera serogroup, based on partial and complete genome sequences of similar virus strains and other selected orthobunyaviruses.
METHODS
Bunyamwera and Ngari virus isolates (Table 1) obtained from previous surveillance exercises in northern Kenya were inoculated into flasks containing confluent Vero cells and observed daily for cytopathic effects (CPE). Viruses were harvested when more than 75% of cells showed CPE. The supernatants were aliquoted into cryotubes and stored at −70 °C, until use.
For RNA extraction, the MagNA Pure LC RNA isolation kit I (Roche Applied Science, USA) was used. cDNA was synthesized using the Transcriptor first-strand cDNA synthesis kit (Roche Applied Science) with random hexamers followed by PCR using Phusion high-fidelity PCR kit (Finnzyme Ltd, Finland) and appropriate primers. Overlapping primers (Supplementary Table S1) for each segment were either designed based on sequences of Bunyamwera, Batai and Ngari viruses available in GenBank or obtained from previous publications [Reference Jost10, Reference Yanase11]. Amplified DNA fragments were visualized by electrophoresis on a 1·5% agarose gel. Amplified DNA fragments were purified and prepared for sequencing using ExoSAP-IT PCR clean-up kit (USB Corp., USA) according to manufacturer's instructions and stored at −20 °C.
Sequencing was performed using sets of overlapping primers for the S, M and L segments as designed previously (Supplementary Table S1) using the Big Dye v. 3·1 kit (Applied Biosystems, USA) and run on the 3500XL Genetic Analyzer (Applied Biosystems). The sequences obtained were cleaned and edited using Bioedit software (USA) for both reads from the forward and reverse primers. Sequences obtained were subjected to Basic Local Alignment Search Tool (BLAST) searches in NCBI GenBank (http://www.ncbi.nlm.nih.gov/blast/Blast) to identify similar sequences. The clean sequences of each segment were aligned against a selection of corresponding segment sequences of Bunyamwera serogroup viruses, including Bunyamwera, Batai and Ngari viruses, using the multiple alignment fast Fourier transform (MAFFT) sequence alignment program [Reference Standley12]. Phylogenetic trees were constructed using maximum likelihood algorithm and analysed with 1000 replicates for bootstrap testing in Molecular Evolutionary Genetics Analysis (MEGA) v. 5·20 software [Reference Tamura13]. La Crosse virus of the California serogroup was used as the out-group for phylogenetic analyses of all three segments of Bunyamwera and Ngari viruses. Nucleotide and amino acid similarities and genetic distances between the Kenyan isolates and selected isolates from diverse regions were computed in MEGA v. 5·20 using the p-distance method [Reference Tamura13].
RESULTS AND DISCUSSION
The complete S, M and L segment coding regions (704, 4302 and 6717 nt, respectively) were sequenced for all five isolates. For this analysis, we excluded the 3′ and 5′ non-coding regions. The gene sequences determined in this study are deposited in GenBank under the following accession numbers (S segment, M segment, L segment):
KM507344, KM507340 and KM507338 (BUNV_MGD_S1_12060_WT);
KM507345, KM507339 and KM507337 (BUNV_GSA_S4_11232_WT);
KM507343, KM514679 and KM507335 (NRIV_TND_S1_19801_WT);
KM507341, KM514677 and KM507336 (NRIV_GSA_TS7_5170_WT);
KM507342, KM514678 and KM507334 (NRIV_ISL_TS2_5242_WT).
Phylogenetic analyses reveal that within all three genomic segments, Bunyamwera virus isolates sequenced in the current study clustered together with an isolate in GenBank from the same locality (BUNV_AMH001130) with strong bootstrap support. The newly sequenced Kenyan Bunyamwera virus isolates possess high nucleotide sequence identities of over 99·0% (S), 99·4% (M) and 99·4% (L) compared to the previous Kenyan isolate (BUNV_AMH001130). Similarly, the newly sequenced Kenyan Bunyamwera virus isolates possess high percentage amino acid sequence identities of over 99·5% (N ORF), 99·0% (NSs ORF), 99·5% (M polyprotein ORF) and 98·8% (L protein ORF) compared to the previous Kenyan isolate. These results indicate that similar Bunyamwera virus strains may be circulating within Kenya and also regionally considering the close association with the 1994 Bunyamwera virus isolate from the Central African Republic (BUNV_ArB29051).
Similarly for Ngari virus, the newly sequenced Kenyan isolates, regardless of the genomic segment, clustered with other Ngari virus isolates (Fig. 1a–c ). The newly sequenced Kenyan Ngari virus isolates were closer to Ngari virus isolates associated with the 1997–1998 haemorrhagic fever outbreak in East Africa (NRIV_9800535 and NRIV_9800521) with strong bootstrap support particularly in analyses of the M and L segments (Fig. 1b , c ). This observation suggests that the current Kenyan Ngari virus isolates may have been introduced into Kenya from the Kenya–Somali border. The newly sequenced Kenyan Ngari virus isolates also possess high nucleotide sequence identities of over 98·5% (S), 97·0% (M) and 97·3% (L) compared to other Ngari virus strains regardless of year and place of isolation. Similarly, the newly sequenced Kenyan Ngari virus strains possess high amino acid sequence identities of over 99·1% (N ORF), 100% (NSs ORF), 98·4% (M polyprotein ORF) and 99·1% (L protein) compared to other Ngari virus isolates. The complete sequence data for the three additional isolates of Ngari virus support previous findings that this virus is a reassortant [Reference Bowen4–Reference Gerrard6].
Of additional interest, we observed temporal/geographical clustering of Ngari isolates in all three phylogenetic trees showing that isolates from the same region or outbreak year generally group together within their own strongly supported independent groupings (Fig. 1a–c ). Our findings suggest a geographical/temporal association with genetic diversity between strains of Ngari virus. Similar observations have been made for Batai virus in which correlation between geographical and genetic diversity has been suggested [Reference Jost10, Reference Huhtamo14]. Batai virus isolates from Europe, Asia and Africa generally cluster independently within strongly supported groups with country-specific viruses clustering closest [Reference Huhtamo14]. However, a more in-depth analysis including identification of signature motifs that might be representative of a given geographical/temporal clade of these viruses is restricted by paucity of genetic sequences in GenBank. Furthermore, there is need for more studies on pathogenesis and distribution of these viruses. While public health efforts have focused on well-characterized viruses such as Rift Valley Fever, West Nile, chikungunya and dengue viruses, the emergence of orthobunyaviruses such as Ngari and Schmallenberg viruses as human and veterinary pathogens, respectively, emphasize the need for in-depth characterization and determination of their true public health impact. Additionally, as more sequences are continually generated, a more extensive survey with regard to sample size and geography is necessary to better understand the distribution of Ngari and Bunyamwera viruses.
SUPPLEMENTARY MATERIAL
For supplementary material accompanying this paper visit http://dx.doi.org/10.1017/S0950268815001338.
ACKNOWLEDGEMENTS
We acknowledge the technical assistance provided by Caroline Tigoi, David Tchouassi, Martin Mbaya and Felix Odhiambo all from the Martin Luscher Emerging Infectious Diseases Laboratory of the International Centre of Insect Physiology and Ecology. We also acknowledge the logistic support of Lillian Igweta, Lisa Omondi and Margaret Ochanda all of Capacity Building, ICIPE. The support through a scholarship to Collins Odhiambo by the Swedish International Development Cooperation Agency (SIDA) through the African Regional Postgraduate Programme in Insect Science (ARPPIS) is greatly acknowledged. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. This paper is published with the permission of the Director of the Kenya Medical Research Institute.
DECLARATION OF INTEREST
None.