|Home | About | Journals | Submit | Contact Us | Français|
A reverse transcription-PCR (RT-PCR) assay was designed, according to previously determined and newly derived genetic data, to target S genomic segments of 47 viruses, including 29 arthropod-borne human pathogens, of the family Bunyaviridae. The analytical sensitivity of the presented assay was evaluated through its application to RNAs extracted from quantitated dilutions of bunyaviruses of interest. Additionally, the assay's analytical specificity was determined through the evaluation of RNAs extracted from selected bunyaviruses and other representative arthropod-borne viruses isolated from a diverse group of host species and temporal and geographic origins. After RT-PCR amplification, DNAs amplified from bunyaviruses of interest were subjected to a novel multiplex sequencing method to confirm bunyavirus positivity and provide preliminary, species-level S segment identification. It is our goal that this assay will be used as a tool for identification and characterization of emergent arthropod-borne bunyavirus isolates of medical import as well as related viruses of the family Bunyaviridae that have not been associated with human illness.
The virus family Bunyaviridae includes more than 300 distinct members, with more than 60 viruses associated with human illness predominantly organized into four genera, Orthobunyavirus, Hantavirus, Nairovirus, and Phlebovirus, according to structural, genetic, and antigenic characteristics (2, 9, 25). Exemplifying the great diversity of species classified within the family, the fifth genus of the family Bunyaviridae, the genus Tospovirus, contains viruses that are known to cause diseases in plants and not animals (2, 9, 29). Viruses of the family Bunyaviridae are highly diverse, with members of each genus additionally classified into individual groups, subtypes, and complexes according to serological character (2, 6, 9, 29). Most bunyaviruses are arthropod-borne, with vectors of transmission including mosquitoes, ticks, culicoid flies, and thrips species (2, 25). However, members of the genus Hantavirus are not associated with an arthropod vector and are thought to be transmitted from infected rodents to humans through aerosolized excreta (2). Bunyaviruses are associated with significant human illnesses worldwide and cause a range of clinical manifestations, including severe pediatric encephalitis (associated with La Crosse virus, of the genus Orthobunyavirus), retinal and hepatic disease (associated with Rift Valley fever virus, of the genus Phlebovirus), and hemorrhagic fever (associated with Crimean-Congo hemorrhagic fever (CCHF) virus, of the genus Nairovirus) (2, 25). Bunyaviruses are enveloped, containing a tripartite genome of mostly negative polarity. The three genomic segments, L, M, and S, can occur in end-hydrogen-bonded circularized forms and minimally encode an RNA-dependent RNA polymerase, envelope glycoproteins (Gn and Gc), and a nucleocapsid (N) protein, respectively (2, 28). The segmented nature of the bunyavirus genome allows the possibility of evolution through antigenic shift. In fact, bunyavirus segment reassortment has been associated with outbreaks of human disease (4, 5, 10). However, the role of segment reassortment in bunyavirus evolution and pathogenicity is largely unknown due to the lack of comprehensive sequence data for members of the family Bunyaviridae (10).
It is believed that arthropod-borne viruses of the family Bunyaviridae will continue to be agents of public health import throughout the 21st century, as has been suggested by their frequent and devastating emergence in recent years (3, 10, 31). In our reference laboratory, we have a specific mission to detect arthropod-borne human pathogens of the family Bunyaviridae as they emerge. To support surveillance, outbreak investigation, and diagnosis, virus isolation is an invaluable tool for the detection of arthropod-borne bunyaviruses. The isolation of arthropod-borne bunyaviruses is performed through the inoculation of either suckling mice or susceptible cells (e.g., Vero cells) with sera or supernatants of homogenates derived from tissues of infected individuals or mosquito pools. Following isolation, identification and characterization of newly derived bunyavirus isolates have been historically performed through the use of predominantly antibody-based methods, including immunofluorescence and plaque reduction neutralization assays (23, 30). While these methods are irreplaceable for the identification of bunyaviruses in a group-specific manner, they are time-consuming and can be limited in their ability to generate unequivocal, species-identifying results due to antibody cross-reactivity and/or limited diversity of available antibodies. In addition, the antibodies required for these assays can be difficult and time-consuming to generate as supplies deplete over time.
The advent of molecular technologies, such as reverse transcription-PCR (RT-PCR), and the application of these technologies in diagnostic and reference laboratories provides a time-efficient alternative to traditional serological methods for the identification of virus isolates. An attractive quality of nucleic acid-based assays is the inherent flexibility of design which allows a researcher to develop an assay with various degrees of specificity, depending on his or her laboratory needs. Additionally, the application of RT-PCR-based molecular consensus assays, designed to detect a group of viruses of interest, followed by nucleotide sequencing for result confirmation and virus speciation has proven a powerful tool for emergent virus identification and discovery when applied to virus isolates in our laboratory (20, 21, 22). However, historically, molecular identification of bunyavirus isolates has been prevented due to limited nucleotide sequence data available for many of these agents and the high level of diversity among bunyaviruses of determined genetic character. Despite these significant challenges, it is proposed that the majority of arthropod-borne human pathogens within the family Bunyaviridae can be identified and characterized by a single RT-PCR-based assay.
To evaluate this assertion, we determined both full- and partial-length S segment primary nucleotide sequence data for 16 bunyaviruses of previously unknown genetic character as targets for assay design and detection prior to the development of the presented assay (15; unpublished data). Here, we describe the development of an RT-PCR-based molecular consensus assay incorporating a diverse set of oligonucleotide primers for the detection of 47 viruses of the Orthobunyavirus, Nairovirus, and Phlebovirus genera of the family Bunyaviridae, including 29 human pathogens of newly derived and previously determined genetic character. Because of the high degree of characterization across genera and the relative conservation within genera, the S segment N open reading frame (ORF) is the genomic target of the presented assay.
The format of the presented RT-PCR-based molecular consensus assay includes the application of two supermixes of oligonucleotides for the detection of RNAs extracted from target viruses of (i) the Orthobunyavirus genus and (ii) the Nairovirus and Phlebovirus genera of the family Bunyaviridae. For S segment species determination, RT-PCR-amplified bunyaviral cDNAs were subjected to a novel multiplex sequencing method followed by NCBI BLAST analyses of newly derived nucleotide sequence data. Based on S segment species determination, this assay is being presented as a tool for the preliminary identification of medically important, and serologically related, viruses of the Orthobunyavirus, Nairovirus, and Phlebovirus genera of the family Bunyaviridae. While there are existing molecular consensus assays for the detection of serogroups of viruses within the family Bunyaviridae (11, 13, 27), to the best of our knowledge, the presented assay detects a larger diversity of bunyaviruses than any previously published effort. It is our hope that this assay will be used as a complement or alternative to standard antibody-based methods for the identification and characterization of bunyaviruses isolated from a variety of geographic locations and sources.
All procedures were performed according to manufacturer instructions unless stated otherwise.
Viruses used in this study were provided by the Arbovirus Diseases Branch of the Division of Vector-Borne Infectious Diseases, National Center for Zoonotic, Vector-Borne, and Enteric Diseases, Centers for Disease Control and Prevention (CDC), and from the World Health Organization arthropod-borne virus reference collection. Viruses selected for evaluation included human pathogens and serologically related viruses of the family Bunyaviridae (Table (Table1).1). Viruses were identified at the species level through serological analyses prior to the beginning of this study. In a few cases, regulations in handling live viruses prevented our ability to work with these agents in our laboratory. For alternatives, we obtained Nairobi sheep disease virus RNA and CCHF virus plasmid-derived RT-PCR controls from George Ludwig (USAMRIID) and Rosemary Sang (Kenya Medical Research Institute), respectively. For quantitation, human pathogens (Table (Table1)1) were subjected to plaque titration of 10-fold dilutions in Vero cells prior to molecular evaluation. Additional viruses of the family Bunyaviridae, as well as representative arthropod-borne viruses of the Alphavirus and Flavivirus genera, of the virus families Togaviridae and Flaviviridae, were evaluated to determine assay specificity (Tables (Tables11 and and22).
A panel of recently derived orthobunyavirus isolates was assembled to evaluate the presented assays' ability to detect target RNAs from a diversity of origins (see Table Table4).4). Isolates from Africa and China were derived from mosquito pools and described and characterized by others prior to their evaluation in our laboratory (8, 24).
Viral RNA was extracted from virus seed by using the QIAamp viral RNA minikit (Qiagen). For specificity testing, extractions were performed on samples ranging in volume from 70 to 140 μl. RNA was eluted in a volume equal to the volume of the starting sample. For sensitivity analysis, extractions were performed on plaque-titrated virus dilutions of 220 μl in volume. RNA was eluted in 110 μl of elution buffer (Qiagen) for a 2× concentration of RNA. Eluted RNA was stored at −70°C until evaluation by RT-PCR amplification.
For amplification and sequencing, oligonucleotide primers (Table (Table3)3) were designed to target previously determined and newly derived S segment sequences of human pathogens of the Orthobunyavirus, Phlebovirus, and Nairovirus genera of the family Bunyaviridae (Tables (Tables11 and and2).2). S segment sequences were aligned by serogroup and genus by using MegAlign software (DNASTAR). According to these alignments, primers were designed manually to target conserved regions shared within the overlapping N and nonstructural ORFs of analyzed orthobunyavirus sequences and the N ORFs of analyzed nairo- and phlebovirus sequences through visual inspection (Fig. (Fig.1;1; Table Table33).
For the detection and amplification of medically important and serologically related viruses of the family Bunyaviridae, the following procedures were followed. RNAs extracted from bunyaviruses of interest and other arthropod-borne viruses (Tables (Tables11 and and2)2) were subjected to RT-PCR amplification with a combination of virus- and serogroup-specific primers designed to detect viruses of the Orthobunyavirus, Nairovirus, and Phlebovirus genera (Table (Table1).1). RT-PCR assays were performed with 20 μl of extracted viral RNA and 50 pmol of each forward and reverse primer listed in mixes for detection of either (i) the Orthobunyavirus genus or (ii) the Nairovirus and Phlebovirus genera (Table (Table3)3) in a 50-μl total reaction volume by using Qiagen OneStep RT-PCR master mix (Qiagen). Reaction mixes were subjected to RT-PCR amplification using the following cycling conditions: 1 cycle of 50°C for 30 min and 95°C for 15 min, followed by 55 cycles of 94°C for 30 s, 55°C for 1 min, and 72°C for 2 min. Reactions were terminated with a final extension step at 72°C for 10 min.
Prior to the selection of presented primers (Table (Table3)3) and assay reaction conditions (see “RT-PCR molecular consensus assay amplification of bunyaviral cDNAs”), numerous alternatives were evaluated (data not shown). The presented oligonucleotide mixes and reaction conditions were selected based on the ability to detect the largest number of bunyaviruses with the highest relative sensitivity.
Products of RT-PCR amplification were evaluated by electrophoresis in a 2.0% agarose gel in 40 mM Tris-acetate-1 mM EDTA buffer. Five microliters of product was analyzed for each amplified sample. Visualization of DNA bands was achieved through ethidium bromide staining and UV transillumination. A sample was determined to be presumptively bunyavirus positive through size-determined migration and visualization of DNA bands, ranging from 210 to 400 bp (Fig. (Fig.2;2; Table Table3),3), on the gel. For DNA purification, the remaining 45 μl of a positive sample was directly purified without the need for gel extraction using a PCR purification kit (Qiagen). As an exception, the presence of multiple bands necessitated gel extraction of the target-sized band from the amplification products of Gabek Forest virus by using a gel extraction kit (Qiagen).
To determine the nucleotide sequence of purified cDNAs amplified from viral RNAs of the Orthobunyavirus, Nairovirus, and Phlebovirus genera, the following steps were taken. DNAs were sequenced with cocktails of either forward or reverse primers targeting (i) the Orthobunyavirus genus or (ii) the Nairovirus and Phlebovirus genera (Table (Table3).3). A total of 3.2 pmol of each listed primer in the appropriate forward or reverse primer mix (Table (Table3),3), along with approximately 50 ng of purified DNA, was added to each reaction by using the ABI BigDye Terminator V3.1 ready reaction cycle sequencing mix (Applied Biosystems). Sequencing reactions were purified using the DyEx 2.0 spin kit (Qiagen). Nucleotide sequences were determined by running purified sequencing reactions on the ABI 3130 genetic analyzer (Applied Biosystems). By these methods, partial S segment sequences were generated for all bunyaviruses amplified by the presented assay (Table (Table11).
Multiple nucleotide sequences generated for each tested bunyavirus (Table (Table1)1) were aligned in both the 5′ and 3′ directions and then edited using SeqMan software (DNASTAR). The resultant consensus sequence was then queried using the NCBI BLAST function of the nucleotide collection database at http://blast.ncbi.nlm.nih.gov (1). For S segment species identification, queries were optimized for somewhat similar sequences (blastn) by using default algorithm parameters. With these methods, sequences that matched the queried sequence with the highest overall maximum value were considered to provide the closest identity to the S segment species of an evaluated DNA. From these data, preliminary species-level identifications of evaluated virus isolates were made.
The GenBank/EMBL/DDBJ accession numbers for the S segment sequences recently determined are EU564827 (Bwamba virus), EU564828 (Pongola virus), EU564829 (Fort Sherman virus), EU564830 (Xingu virus), and EU564831 (Shokwe virus), and newly determined accession numbers are FJ235921 (Wyeomyia virus), FJ235922 (Alenquer virus), FJ235923 (Chandiru virus), FJ235924 (Chagres virus), FJ235925 (Cacao virus), FJ235926 (Itaporanga virus), FJ235927 (Gabek Forest virus), FJ235928 (Turuna virus), FJ235929 (Rio Grande virus), FJ235930 (Nique virus), and GQ166188 (Itaituba virus).
The analytical sensitivity of the bunyavirus molecular consensus assay, as determined through its application to plaque-titrated human pathogens of the family Bunyaviridae, is reported in Table Table1.1. For the Orthobunyavirus genus, the molecular consensus assay detects between 660 PFU/0.1 ml (for Wyeomyia virus) and 0.1 PFU/0.1 ml (for Oropouche virus). For the Phlebovirus and Nairovirus genera, detection limits range from 100 PFU/0.1 ml (for Rift Valley fever virus) to 0.0024 PFU/0.1 ml (for Chagres virus).
The analytical specificity of the bunyavirus molecular consensus assay, as determined through its application to viruses of the families Bunyaviridae, Flaviviridae, and Togaviridae, is presented in Tables Tables11 and and2.2. Forty-seven viruses, including 29 human pathogens, representing six major serogroups of the family Bunyaviridae, were targeted and amplified by the presented assay (Table (Table1).1). Demonstrating fidelity for bunyaviral RNA targets, no evaluated viruses of the Flaviviridae or Togaviridae families were detected by the consensus assay (Table (Table2).2). For all amplified bunyaviral DNAs, multiplex sequencing and NCBI BLAST analyses generated results for S segment nucleotide sequence data that confirmed the serologically derived identity, with the exceptions of Ilesha virus KO/2, Maguari virus 75V3429, and snowshoe hare virus LIEV 13004AKH strains (Table (Table1).1). The S segment species amplified and sequenced from Ilesha virus KO/2, Maguari virus 75V3429, and snowshoe hare virus LIEV 13004AKH strains shared the highest nucleotide sequence identities with various Batai, Cache Valley, and Chatanga virus strains, respectively, according to our analyses (data not shown).
The ability of the bunyavirus consensus assay to detect target viral RNAs extracted from temporally relevant samples was evaluated through its application to a panel of recent isolates of international origins (8, 24; Table Table4).4). The presented molecular consensus assay amplified and identified Bunyamwera, Pongola, and Tahyna virus cDNAs of the Bunyamwera, Bwamba, and California serogroups of the genus Orthobunyavirus from viruses isolated from Kenya and China in 2006 and 2007, respectively (Table (Table4).4). Confirming fidelity for bunyaviral RNAs, other recently derived isolates of the Alphavirus and Flavivirus genera were not amplified by the presented consensus assay (Table (Table44).
To the best of our knowledge, the presented bunyavirus molecular consensus assay has been shown to detect a range of viruses more diverse than can be detected by any other molecular assay designed for the detection of arthropod-borne viruses. In addition, the assay has been shown to have sufficient sensitivity for the detection of virus strains with various titers isolated from a diverse range of host species and geographic and temporal origins (Tables (Tables11 and and4).4). Also, when combined with multiplex sequencing and NCBI BLAST analyses, the presented assay was able to confirm serologically derived identities for all but 3 of 47 amplified viruses (Tables (Tables11 and and4).4). Finally, the assay has been shown to have fidelity for bunyaviral RNAs when applied to arthropod-borne viruses of the families Flaviviridae and Togaviridae (Tables (Tables22 and and44).
In addition to viruses of the Phlebovirus and Orthobunyavirus genera that were detected (Table (Table1),1), some evaluated orthobunyaviruses were not amplified by the presented assay (Table (Table2).2). While we were able to detect Nyando and Guaroa viruses by using alternative primer mixes and reaction conditions relative to those presented here, these alternatives compromised the sensitivity of detection for other medically important agents and, as a result, were abandoned in the final assay design (data not shown). For the group C and Guama serogroup viruses presented in Table Table2,2, no efforts (presented or alternative) generated DNAs for sequencing. These findings are particularly confounding with regard to the group C agents, for which multiple amplification attempts, using oligonucleotide primers designed to target previously published group C, S, and M segments (26), failed to generate DNAs for sequencing. Having tried to eliminate other variables that might contribute to disparities in amplification efficiency, we believe that target nucleotide sequence heterogeneity is responsible for the inability to detect these agents by molecular methods.
The identification of S segment species through multiplex sequencing and NCBI BLAST analyses is a powerful component of the presented molecular consensus assay. However, we regard the determination of virus species through GenBank-derived S segment identity as preliminary for the following reasons. (i) It is often impossible to verify the integrity of sequence data for a given virus species in GenBank. As a result, inaccurate data could potentially cause the misidentification of an S segment species by using the presented methodologies. However, our collective experience using GenBank queries for virus identification has indicated a high degree of integrity for these data in instances where we were able to verify virus identity by alternative methods (Table (Table2)2) (7, 21, 22, 23). (ii) The capacity for genomic segment reassortment among viruses of the family Bunyaviridae allows the possibility that a given isolate possesses genomic segments of heterologous ancestries. Although we did not evaluate them in this study, we consider our findings of Cache Valley, Batai, and Chatanga virus-like S segments, amplified from isolates serologically classified as Maguari, Ilesha, and snowshoe hare virus strains, respectively (Table (Table1),1), evidence of possible segment reassortment. While not yet undertaken, additional experiments will be performed by our group to elucidate the identities of the M and L segments of these virus strains to verify the ancestry of all three genomic segments. Finally, because of the above-described limitations on virus identification by GenBank querying, we recommend that the preliminary identification of a virus species through GenBank-derived S segment identity always be confirmed by additional molecular and/or serological analyses.
The molecular consensus assay is generally more sensitive for the detection of viruses of the Nairovirus and Phlebovirus genera than for the detection of viruses of the Orthobunyavirus genus (Table (Table1).1). We have considered several potential explanations for this difference in sensitivities. (i) From our collective laboratory experience developing and applying assays for the detection of arthropod-borne viruses (12, 14, 16, 17, 18, 19), it has been noted that, in general, increasing the number of different primers in a reaction causes a reduction in overall sensitivity of detection. To accommodate a greater diversity of nucleotide sequence targets, the Orthobunyavirus mix contains more primers than the Nairovirus/Phlebovirus mix does, possibly contributing to the noted differences in sensitivities of detection (Table (Table1).1). (ii) Oligonucleotides designed for the detection of orthobunyaviruses bind target nucleic acid less efficiently than those designed for the detection of nairo- and phleboviruses. In fact, to incorporate a broader diversity of targets, primers designed for the detection of viruses of the Orthobunyavirus genus are generally less specific for the target sequence than are those for the detection of viruses of the Nairovirus and Phlebovirus genera. Therefore, the Orthobunyavirus primers (Table (Table3)3) likely prime RT-PCR amplification less efficiently, resulting in reduced sensitivity. (iii) There could be a difference in the ratio of RNA copies to infectious viral particles in the Vero cell plaque assay for these viruses. If the evaluated phlebo- and nairoviruses have a higher ratio of RNA copies to PFU equivalents in the Vero cell culture system than the evaluated orthobunyaviruses, it would appear that the sensitivity of detection for the former agents is greater. The application of this assay to standards of known RNA copy number is supported by this conjecture. However, because PFU represent the most relevant unit for quantitating infectious virus, we consider experiments designed to determine assay sensitivity through its application to standards of known copy number beyond the necessary scope of this research.
The data presented here (Tables (Tables1,1, ,2,2, and and4)4) demonstrate the utility of the molecular consensus assay for the detection of virus isolates of the family Bunyaviridae. Most promising to us is the ability of the assay to correctly identify temporally relevant human pathogens of the genus Orthobunyavirus, isolated from mosquito pools as part of arbovirus surveillance programs in Kenya and China (Table (Table4)4) (8, 24). Unfortunately, we do not have a comparable panel of recent isolates for the characterization of the Nairovirus/Phlebovirus component of this assay. However, it is believed that the extremely high sensitivities reported for the detection of historically obtained viruses of these genera (Table (Table1)1) are conducive for the detection of newly emergent nairo- and phleboviruses as well. We recommend the incorporation of the presented molecular consensus assay in an algorithm that includes additional molecular and serological methodologies for the comprehensive identification of medically important, and serologically related, virus isolates of the family Bunyaviridae. Additionally, it is our intention that the presented assay might be combined with other previously published (11) or newly derived assays to target multiple segments of the bunyavirus genome. However, research to determine the feasibility of an assay with a designed capability for the detection of reassortant bunyaviruses was not performed in this study. Finally, because this assay has been designed to target conserved regions shared between genetically disparate agents, we believe that additional bunyavirus species besides those evaluated here will be amplified and detected by the presented assay. However, the paucity of genetic information available for many of these agents and extraordinary diversity among viruses of the family Bunyaviridae prevent absolute certainty that the assay will amplify uncharacterized viruses.
We thank Rosemary Sang of the Kenya Medical Research Institute (KEMRI) and George Ludwig (USAMRIID) for the provision of a CCHF virus plasmid and Nairobi sheep disease virus RNA, respectively, that were used in the evaluation of the presented assay. We also thank Jeff Chang for advice on consensus primer design, Mary Crabtree for ongoing collaborative discussion on bunyaviruses, Carol Blair for critical review of the manuscript, George F. Newell, Jr., for editorial advice, and Guo-Dong Liang for Chinese Tahyna virus DNA.
Published ahead of print on 17 June 2009.