Use of a Short Fragment of the C-Terminal E Gene for Detection and Characterization of Two New Lineages of Dengue Virus 1 in India
http://www.100md.com
微生物临床杂志 2006年第4期
Laboratorio de Arbovirus y Enfermedades Víricas Importadas, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Madrid, Spain
Jerome L. and Dawn Greene Infectious Diseases Laboratory, Columbia University, New York, New York
Robert Koch Institute, Berlin, Germany
Centro de Salud Internacional, Hospital Clinic (IDIBAPS), Barcelona, Spain
ABSTRACT
Here we propose the use of a 216-nucleotide fragment located in the carboxyl terminus of the E gene (E-COOH) and a pairwise-based comparison method for genotyping of dengue virus 1 (DENV-1) strains. We have applied this method to the detection and characterization of DENV-1 in serum samples from travelers returning from the tropics. The results obtained with the typing system correlate with the results obtained by comparison of the sequences of the entire E gene of the strains. The approach demonstrates utility in plotting the distribution and circulation of different genotypes of DENV-1 and also suggests the presence of two new clades of Indian strains. The integration of the method with an online database and a typing characterization tool enhances its strength. Additionally, the analysis of the complete E gene of DENV-1 strains suggested the occurrence of a nondescribed recombination event in the China GD23-95 strain. We propose the use of this methodology as a tool for real-time epidemiological surveillance of dengue virus infections and their pathogenesis.
INTRODUCTION
Dengue is the most common and widespread arboviral disease worldwide, with at least 50 million cases recorded every year. Although the majority of infections with dengue viruses (DENV) are asymptomatic, the wide spectrum of disease ranges from mild, self-limited dengue fever to severe and potentially life-threatening dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS). Dengue fever is an acute febrile viral disease, with patients with dengue fever frequently presenting with headache, bone or joint and muscular pain, rash, and leukopenia. DHF is a more virulent form of dengue virus infection and is considered a different clinical entity; it may progress to hypovolemic shock (DSS). Each year nearly 500,000 cases of DHF-DSS require hospitalization, with case-fatality rates as high as 5%, depending on the availability of treatment (39, 40).
The increasing incidence of dengue epidemics in the Americas, Southeast Asia, the Indian subcontinent, and the Western Pacific, along with the growing rate of DHF and DSS cases, is a global public health problem. Urbanization, overpopulation, crowding, poverty, a weakened public health infrastructure, and the globalization of trade and travel have been implicated as causes for this increase (3, 9).
Dengue viruses comprise four different serotypes, and evolutionary and epidemiological studies have facilitated clustering of dengue viruses into genotypes within serotypes (14, 30). It was posited that dengue hemorrhagic fever is a result of the immune enhancement developed after subsequent infection with two different serotypes (10). This is also a plausible mechanism to explain the observed highest risk of DHF and DSS in primary infections in infants with heterotypic immunity due to waning maternal dengue virus antibodies (17). It has also been suggested that virulent strains present an increasing replication rate in human target cells (29); therefore, the virus itself might play an important role in disease severity. Moreover, some genotypes have been associated with DHF epidemics, while others genotypes have been responsible for mild disease in certain areas while causing secondary infections (11, 37, 38). It was posited that these genotypes may present different antigenic epitopes and that these structural differences correlate with pathogenesis (18, 21). Significant differences in disease severity are observed in secondary infections, depending on the origin of the strains (4, 18, 21). Thus, the presence of a specific genotype in an area could be associated with an increased risk for the appearance of DHF and DSS cases.
Accordingly, a surveillance methodology able to identify the presence or introduction of dengue genotypes in areas of endemicity is of great interest for pathogenesis studies and for the establishment of local epidemiological control policies.
Here we describe a rapid, sensitive method for the genotyping of DENV-1, in which a short fragment in the carboxyl terminus of the E gene (E-COOH) is amplified by PCR, sequenced, and analyzed in pairwise comparisons with prototypic DENV sequences representing DENV-1 strains. Application of this method to serum samples obtained from DENV-1-infected travelers who had recently returned to Europe from India suggested the presence of two new DENV-1 lineages. Sequence analysis of the entire E gene confirmed this observation and supported the utility of the method.
MATERIALS AND METHODS
Samples from dengue virus-infected patients. Samples were obtained from travelers with acute DENV-1 infection returning to Europe from regions where dengue is endemic (Table 1). Samples (sera and/or viral culture supernatants) were provided by virology research laboratories of the European Network for Diagnosis of "Imported" Viral Diseases (ENIVD) and Tropical Medicine Clinical Units from the European Network for Tropical Medicine (TropNetEurop). The samples were stored at –80°C until use.
RNA extraction and RT-PCR. Viral RNA was extracted from 140-μl serum samples or viral culture supernatants from DENV-1-infected patients by using the QIAamp viral RNA minikit (QIAGEN) and resuspended in 60 μl of water. Five microliters of RNA extract was subjected to nested reverse transcriptase PCR (RT-PCR; QIAGEN One-Step RT-PCR kit) to amplify a 328-nucleotide (nt) fragment spanning the E-NS1 junction of the DENV genome (5). The RT-PCR mixture contained 1x one-step RT-PCR buffer, 400 mM each deoxynucleoside triphosphate, 400 nM each sense and antisense degenerated primer (primers E/NS1-S and E/NS1-R), and an optimized combination of Omniscript and Sensiscript reverse transcriptases and HotStar Taq DNA polymerase. The RT-PCRs were carried out with a 50-μl reaction volume by using an initial reverse transcription step at 41°C for 45 min, followed by a denaturation and HotStar Taq polymerase activation step (94°C, 15 min) and 40 cycles of denaturation (94°C, 30 s), primer annealing (55°C, 1 min), and primer extension (72°C, 2 min). A final incubation was carried out at 72°C for 5 min. A second amplification reaction (nested PCR) was seeded with 1 μl of the initial amplification product. The 50 μl of the reaction mixture contained 60 mM Tris-HCl (pH 8.5), 2 mM MgCl2, 15 mM (NH4)2SO4, 800 nM each sense and antisense primer (primers E/NS1-SS and EGENE/NS1-RR), and 2.5 U of DNA Taq polymerase (Perkin-Elmer). The samples were subjected to a denaturation step (94°C, 15 min), followed by 40 cycles of denaturation (94°C, 30 s), primer annealing (57°C, 4 min), and primer extension (72°C, 2 min) and a further extension step at 72°C for 5 min.
The samples were also subjected to analysis by a specific DENV-1 nested RT-PCR to amplify the complete E gene. This protocol was performed by using the buffers described for the E-NS1 amplification nested RT-PCR, except for the use of specific primers (primers EGENE1-S, EGENE-R, EGENE1-SS, and EGENE/NS1-RR) that were designed based on published DENV-1 sequences by computer-assisted analysis (MACAW version 32 software, 1995; NCBI). The RT-PCRs were carried out by using an initial reverse transcription step at 48°C for 45 min, followed by a denaturation and HotStar Taq polymerase activation step (94°C, 15 min) and 40 cycles of denaturation (94°C, 30 s), primer annealing (56°C, 2 min), and primer extension (72°C, 2 min). A final incubation was carried out at 75°C for 5 min. In the second amplification reaction, the samples were subjected to a denaturation step (94°C, 5 min), followed by 40 cycles of denaturation (94°C, 30 s), primer annealing (59°C, 2 min), and primer extension (72°C, 2 min) and a further extension step at 75°C for 5 min.
Table 2 shows the sequences and the respective primer positions in the viral genome used for partial (328-nt) or whole (1,700-nt) E gene amplification. The sensitivities were 100 and 10,000 copies, respectively.
Nucleotide sequence analysis. The template for sequencing was obtained by excising the band of interest after size fractionation of the PCR products by agarose gel electrophoresis. Approximately 20 ng of cDNA was sequenced by using 40 pmol oligonucleotide primer and the ABI Prism Dye Terminator cycle sequencing ready reaction kit (Applied Biosystems). The sequence of the E-NS1 fragment was obtained by using a forward primer and a reverse primer flanking the amplification product (Table 2); 11 primers were used to obtain overlapping sequence of the E gene (Table 2). A total of 30 DENV-1 E-NS1 gene (328-nt) and 16 DENV-1 E gene (1,485-nt) sequences were obtained.
Original sequence data were first analyzed by use of CHROMAS software (version 1.3, 1996; C. McCarthy, School of Biomolecular and Biomedical Science, Faculty of Science and Technology, Griffith University, Brisbane, Queensland, Australia); the forward and reverse sequence data for each sample were aligned by using the program SEQMAN (DNASTAR Inc. Software, Madison, Wis.). The consensus sequence was compared and aligned to other samples or DNA database sequences using the program CLUSTAL X, version 1.83 (34).
Sequence analysis of amplified products. A set of 156 DENV-1 sequences (216 nt in length) comprising those from 30 new sequenced strains and 126 collected from GenBank (updated to July 2005) were used for phylogenetic analysis. To test the findings observed with the short fragment, the entire E protein-coding sequence (1,485 nt) was recovered directly from 16 clinical samples by nested RT-PCR and compared in pairwise analyses with DENV-1 complete E 116 sequences collected from GenBank.
Phylogenetic analysis. Phylogenetic analysis was performed by using the best model of nucleotide substitution (according to Modeltest [27] and Tamura and Nei [33], with correction for the proportion of invariable sites of 0.4434 and a gamma distribution of 1.4212). Programs from the MEGA package (version 3) (19) were used to produce phylogenetic trees, reconstructed by the neighbor-joining method. The statistical significance of a particular tree topology was evaluated by bootstrap resampling of the sequences 1,000 times. Pairwise comparisons of the DENV-1 database were done by global alignment by using the algorithm of Needleman and Wunsch (23), implemented by a program from EMBOSS, the European Molecular Biology Open Software Suite (28). An automated program that performs the same analysis is available at and http://aevi.isciii.es.
Recombination event detection. Systematic screening for the presence of recombination patterns was achieved by using the nucleotide alignments and the Recombination Detection Program (RDP; Darren Martin) (22). The algorithms Bootscan (31), MaxChi (32), Chimaera (26), LARD (Likelihood Assisted Recombination Detection) (13), and Phylip Plot (6) were also used for detection of recombination.
Nucleotide sequence accession numbers. The GenBank accession numbers of the nucleotide sequences determined in this study are DQ016630 to DQ016659.
RESULTS
Thirty symptomatic patients with DENV-1 infection were diagnosed and characterized by analysis of the carboxyl terminus of the E gene. The results are detailed in Table 1. The results of the sequence comparison suggested the existence of a new cluster of sequences in patients returning from India. Given the short nature of the fragment that was amplified, the entire sequence of the envelope glycoprotein was obtained for 16 samples.
Phylogenetic analysis using the complete envelope gene. The phylogenetic tree obtained by analysis of a complete data set of all available envelope gene sequences is shown in Fig. 1. The analysis not only allowed the classification within known DENV-1 genotypes, America-Africa (AMAF), Malaysia (MAL), Thailand (THAI), Asia (ASIA), and South Pacific (SP), but also suggested the existence of lineages with distinctive geographical and temporal relationships (16, 20).
The genotype AMAF sequences (Fig. 2) clustered in five well-defined lineages. Lineage AMERICA-1 comprises all known circulating South and Central American strains, as well as a virus (RIOH36589) isolated in Brazil from a traveler returning from Angola (7). Lineage AFRICA includes three East Africa strains. Lineage SOUTH ASIA includes three Myanmar strains circulating in 1976 and 1998. We have designated the remaining two lineages INDIA-1 and INDIA-2. One recombinant strain (Singapore-S275), with parents from different lineages, remains separated from the rest of the lineages.
The ASIA genotype sequences (Fig. 3) are clustered in three lineages: ASIA-1, which contains Japanese and old Hawaiian strains; ASIA-2, which includes strains recovered from 1980 to 2001 in China, Thailand, Laos, Myanmar, Taiwan, and Djibouti; and ASIA-3, which contains the most recent isolates (1993 to 2004) from Myanmar, Thailand, Cambodia, China, Malaysia, and Indonesia.
The SP genotype (Fig. 4) represents four different lineages: SP-1, which comprises isolates from Indonesia (1998), Polynesia, and Micronesia; SP-2, which comprises isolates from Indonesia (1998 to 2002) and Australia (1983); SP-3, which comprises strains from the Philippines (1995 to 2004) and a recent Indonesian isolate (SC01); and SP-4, which contains the oldest strains from the Philippines (1974 to 1984). One strain from China (GD23-95-China-1995), three isolates from America, and one isolate from Thailand clustered independently. Again, as described above, they are recombinants and their parents belong to different lineages.
The genotype MAL includes only the sylvatic Malaysian isolate. The genotype THAI clusters old Thailand isolates recovered from 1954 to 1964. Both MAL and THAI genotypes may be extinct, as no recent isolates have been found.
Recombination analysis. Recombination has been reported in the envelope gene in all DENV serotypes (35, 41). Because recombination events may occur at various positions in the envelope gene, phylogenetic trees based on partial envelope sequence data may be misleading. To address this concern we searched for evidence of recombination using the full envelope sequence. We detected previously reported recombination between lineages AMERICA-1 and ASIA-3 strains in S275-Singapore-1990, and between lineages AMERICA-1 and SP-3 in strains Caribbean 495-Mexico1985, Caribbean 925-Mexico1985, CV1636-Jamaica1977, and AHF82-80-Thailand1982 (35). Additionally, we found evidence of a new recombination event in the GD23-95-China-1995 sequence. This virus appears to be a recombinant of the ASIA-3 and SP-1 lineages (Fig. 5). All the recombinant strains appeared as outliers in our phylogenetic tree (Fig. 2 and 4).
Phylogenetic analysis of the carboxyl terminus of the envelope gene. Although phylogenetic assignment of dengue viruses is classically based on the sequence of the entire E gene, this approach is not practical for a real-time epidemiological surveillance of dengue virus infections and their pathogenesis, where high throughput is essential. More sequence data are available for the carboxyl terminus than the amino terminus of the E gene; furthermore, the carboxyl terminus contains regions of sequence conservation at positions 1850 to 2200 and 2500 to 2700 that are ideal for consensus PCR. We tested the ability of the carboxyl-terminal segment between positions 2196 and 2411 (BR/90 strain; GenBank accession no. AF226685) to predict phylogenetic assignments based on the complete E gene sequence. The topologies obtained were essentially the same. Genotypes ASIA, MAL, and THAI showed the same clustering. The changes in topology observed in genotypes AMAF and SP are explained by recombination events. Since in all the recombination events described the 5' breakpoint is located approximately 350 nt into the E gene and the 3' end breakpoint occurs in the NS1 gene, the analysis of the 3' portion of the E gene should allow detection of the inserted recombinant sequence. Consistent with this observation, strain GD23-95-China-1995, a recombinant of the ASIA-3 and SP-1 lineages, groups with the SP-1 lineage. Similarly, the cluster that includes recombinant strains from the AMERICA-1 and SP-3 lineages appears as a member of the SP-3 lineage; and the S275-Singapore strain, a recombinant between AMERICA-1 and ASIA-3 lineages, appears as a member of the AMERICA-1 lineage.
Pairwise analysis by use of local alignments. An all-against-all pairwise sequence comparison was done with the DENV-1 sequences to assess the potential for the establishment of a simple program for the classification of DENV-1 genotypes and lineages similar to programs that we have created for mumps virus (25) and adenovirus (1). Sequence analysis of the 216-nt region at the carboxyl terminus amplified by PCR with consensus primers SEQAS/E/NS1-SS and SEQBR/EGENE/NS1-RR (Table 2) was sufficient to allow determination of the genotype and the lineage (Fig. 6). The validity of this method was confirmed by analysis of variance by comparing the scores of the sequence comparisons within genotypes or lineages to the scores of comparisons between genotypes. A software tool has been developed to classify DENV-1 sequences by introducing the target sequence (either the complete sequence or the carboxyl E gene sequence) and is available at and http://aevi.isciii.es.
DISCUSSION
Viral nucleotide sequence data and phylogenetic methods have been widely used to understand genetic relationships between dengue viruses as well as their epidemiology (29). Others have tried to establish phylogenetic relationships based on a partial sequence, notably, the intersection between E and NS1 (111 nt from the 3' end of E and 129 nt from the 5' end of NS1) (30) and the amino terminus of E (2, 30). At the time that these regions were proposed for use for phylogenetic analyses, only a few sequence data were available for optimization and validation and the phenomenon of recombination in dengue viruses had not been thoroughly evaluated. Although dengue virus classification had been consistent when the full envelope sequence is used, amplification and sequencing analysis of nearly 1,500 nt are not feasible for real-time epidemiological surveillance. Here we describe a 216-nt fragment located in the carboxyl terminus of the E gene of DENV-1 that can be readily amplified by consensus PCR to delineate DENV-1 genotypes. The classification data obtained with the shorter sequence are similar to those obtained with the complete E gene, with one caveat, i.e., detection of recombination events.
Three (2, 15, 24) to five (7, 29) different genotypes of DENV-1 have been proposed. Our analysis of complete E gene sequences confirmed the presence of five genotypes and the clustering of sequences below the genotype level that correlate with the geographical origin and/or time of isolation and appear to represent distinct lineages with strong bootstrap support (Fig. 1).
A correlation between genotypes and pathogenesis had been suggested. It would be of interest to observe if the potential to cause DHF is a property shared at the genotype level or at the lineage level. The AMAF genotype has been described as a genotype with a high epidemiological impact due to its spread and its potential to cause DHF (29). Remarkably, two new clusters of Indian strains belonging to the genotype AMAF (lineages INDIA-1 and INDIA-2) were observed in our analysis. One lineage (INDIA-1) is related to American strains, while INDIA-2 is closer to the African and South Asiatic strains. Two different lineages are currently circulating in the subcontinent. From our data it could be inferred that this genotype is currently circulating in extensive areas both in the Americas (the AMERICA-1 lineage) and on the Indian subcontinent (the INDIA-1 and INDIA-2 lineages).
Given the geographical or temporal association among the sequences belonging to the same lineage, a classification based on lineages facilitates surveillance and tracking of dengue virus isolates. For example, in genotype ASIA, three lineages (the ASIA-1, ASIA-2, and ASIA-3 lineages) have been detected. The ASIA-2 lineage had been detected in Myanmar, Cambodia, and Thailand. We detected two members of ASIA-2 in patients returning from Kenya; this may suggest the introduction of this lineage into East Africa.
During the course of reviewing published sequence data, we discovered evidence of recombination between the ASIA-3 and the SP-1 lineages in GD23-95-China 1995. Since this virus appears to remain apart from the rest of the clusters (Fig. 5) and was isolated in 1995, it is possible that this lineage did not become established. Although recombination is accepted in Japanese encephalitis and St. Louis encephalitis viruses (8, 36), recombination in DENV has been controversial. The original reports of recombination in DENV by Holmes et al. (12) and Worobey et al. (41) have been challenged (7, 29). The only unequivocal way to resolve the scientific question with respect to GD23-95-China 1995 would be to obtain the isolate to independently confirm the results of sequence and phylogenetic analyses.
The use of a consensus PCR based on a short fragment of the E protein coupled to a similarity analysis by using a pairwise alignment method demonstrate that the DENV-1 genotype and lineage can be accurately defined. This method offers some additional advantages in microbiology laboratories, in contrast to the application of classical methods of molecular epidemiology (e.g., sequencing of the complete envelope from cell culture isolates). Among those advantages, we count the use of clinical samples as the source of viral RNA; the amplification of a short sequence to identify the serotype, the genotype, and the lineage, thus facilitating the acquisition of information in a very short time; the application of pairwise comparison analysis and the proposed online typing tool to identify DENV-1 genotypes, which minimizes the need for extensive phylogenetic analysis; and the use of an automatically updated database for comparison and a global analysis (e.g., all available DENV sequences are used each time), which minimizes the possibility of the most common mistake in phylogenetic analysis (e.g., use of a biased set of sequences for comparison). Only in the case that a new strain could not be clearly associated with a described genotype and/or lineage should the complete E gene be analyzed to identify the appearance of a new genotype. The detection and characterization of two new lineages of DENV-1 in India and the description of the spread of genotype ASIA, lineage ASIA-2, in East Africa clearly show the usefulness of this approach. This approach has been demonstrated to be equally useful for the genotyping of the four dengue virus serotypes (data not shown).
The use of this methodology for the identification of dengue virus infection could greatly contribute to the acquisition of dengue epidemiological data worldwide, detection of the appearance and spread of new genotypes and lineages, and detection of the spread and circulation of those already described and, more deeply, to a basic understanding of dengue virus pathogenesis.
ACKNOWLEDGMENTS
We are grateful to the members of ENIVD and TropNetEurop for providing the DENV isolates. This study would not have been possible without their help. We thank F. Molero for technical assistance with the processing of samples.
M.N. and A.T. are ENIVD participants . Other ENIVD participants include P. Cassinotti and D. Schultze (Institute für Klinische Mikrobiologie und Immunologie, St. Gallen, Switzeralnd), M. Van Esbroek (Institute of Tropical Medicine, Antwerp, Belgium), A. Fomsgaard and L. Vinner (Statens Serum Institute, Copenhagen, Denmark), and M. Grandadam (Unite de Virologie Tropicale Parc du Pharo, Marseille Armees, France). J.G. is TropNetEurop participant. Other TropNetEurop participants include S. Puente (Hospital Carlos III, Madrid, Spain), O. Whichmann (Institute for Tropical Medicine, Berlin, Germany), M. Schunk (Medizinische Klinik Klinikum der Universitt, Munich, Germany), and R. Lopez-Velez (Hospital Ramon y Cajal, Madrid, Spain). M.N. is a representative of the ENIVD Dengue Study Workgroup, and J.G. is a representative of the TropNetEurop Dengue Study Workgroup. A.T., J.G., and C.D., as well as S. Puente, are researchers of the RICET Network (Red de Investigacion Cooperativa en Enfermedades Tropicales C03/04).
C.D. is under contract by an agreement between the Public Health Division of the Spanish Ministry of Health (DGSP-MSC) and the Instituto de Salud Carlos III (ISCIII) for the development of the Hemorrhagic Viral Fevers Surveillance and Control Program. This work has received financial support from the Instituto de Salud Carlos III (ISCIII) through research project grants (grants MPY 1194/02 and C03/04). G.P. and W.I.L. are supported by the Ellison Medical Foundation and the National Institutes of Health (grants AI 51292 and U54 AI57158 to W.I.L.).
Both authors contributed equally.
REFERENCES
Casas, I., A. Avellon, M. Mosquera, O. Jabado, J. E. Echevarria, R. H. Campos, W. I. Lipkin, P. Perez-Brea, and G. Palacios. 2005. Molecular identification of adenoviruses in clinical samples by analyzing a partial hexon genomic region. J. Clin. Microbiol. 43:6176-6182.
Chungue, E., O. Cassar, M. T. Drouet, M. G. Guzman, M. Laille, L. Rosen, and V. Deubel. 1995. Molecular epidemiology of dengue-1 and dengue-4 viruses. J. Gen. Virol. 76(Pt 7):1877-1884.
Clarke, T. 2002. Dengue virus: break-bone fever. Nature 416:672-674.
Cologna, R., P. M. Armstrong, and R. Rico-Hesse. 2005. Selection for virulent dengue viruses occurs in humans and mosquitoes. J. Virol. 79:853-859.
Domingo, C., G. Palacios, M. Niedrig, M. Cabrerizo, O. Jabado, N. Reyes, W. I. Lipkin, and A. Tenorio. 2004. A new tool for the diagnosis and molecular surveillance of dengue infections in clinical samples. Dengue Bull. 28:87-95.
Felsenstein, J. 1989. PHYLIP: phylogeny inference package (version 3.2). Cladistics 5:164-166.
Goncalvez, A. P., A. A. Escalante, F. H. Pujol, J. E. Ludert, D. Tovar, R. A. Salas, and F. Liprandi. 2002. Diversity and evolution of the envelope gene of dengue virus type 1. Virology 303:110-119.
Gould, E. A., X. de Lamballerie, P. M. Zanotto, and E. C. Holmes. 2003. Origins, evolution, and vector/host coadaptations within the genus Flavivirus. Adv. Virus Res. 59:277-314.
Gubler, D. J. 2002. Epidemic dengue/dengue hemorrhagic fever as a public health, social and economic problem in the 21st century. Trends Microbiol. 10:100-103.
Halstead, S. B. 1981. The Alexander D. Langmuir Lecture. The pathogenesis of dengue. Molecular epidemiology in infectious disease. Am. J. Epidemiol. 114:632-648.
Halstead, S. B., T. G. Streit, and J. G. Lafontant. 2001. Haiti: absence of dengue hemorrhagic fever despite hyperendemic dengue virus transmission. Am. J. Trop. Med. Hyg. 65:180-183.
Holmes, E., M. Worobey, and A. Rambaut. 1999. Phylogenetic evidence for recombination in dengue virus. Mol. Biol. Evol. 16:405-409.
Holmes, E. C. 1998. Molecular epidemiology of dengue virus—the time for big science. Trop. Med. Int. Health 3:855-856.
Holmes, E. C., and S. S. Burch. 2000. The causes and consequences of genetic variation in dengue virus. Trends Microbiol. 8:74-77.
Hwang, K. P., P. Y. Chu, Y. C. Tung, H. L. Wang, Y. Y. Yueh, Y. C. Wu, C. Chin, and K. H. Lin. 2003. Molecular epidemiological study of dengue virus type 1 in Taiwan. J. Med. Virol. 70:404-409.
Kanesa-Thasan, N., G. J. Chang, B. L. Smoak, A. Magill, M. J. Burrous, and C. H. Hoke, Jr. 1998. Molecular and epidemiologic analysis of dengue virus isolates from Somalia. Emerg. Infect. Dis. 4:299-303.
Kilks, S. C., S. Nimmanitya, A. Nisalak, and D. S. Burke. 1988. Evidence that maternal antibodies are important in the development of dengue haemorrhagic fever in infants. Am. J. Trop. Med. Hyg. 38:411-449.
Kochel, T. J., D. M. Watts, S. B. Halstead, C. G. Hayes, A. Espinoza, V. Felices, R. Caceda, C. T. Bautista, Y. Montoya, S. Douglas, and K. Russell. 2002. Effect of dengue-1 antibodies on American dengue-2 viral infection and dengue hemorrhagic fever. Lancet 360:310-312.
Kumar, S., K. Tamura, I. B. Jakobsen, and M. Nei. 2001. MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17:1244-1245.
Kurane, I., and T. Takasaki. 2001. Dengue fever and dengue haemorrhagic fever: challenges of controlling an enemy still at large. Rev. Med. Virol. 11:301-311.
Leitmeyer, K. C., D. W. Vaughn, D. M. Watts, R. Salas, I. Villalobos de Chacon, C. Ramos, and R. Rico-Hesse. 1999. Dengue virus structural differences that correlate with pathogenesis. J. Virol. 73:4738-4747.
Martin, D., and E. Rybicki. 2000. RDP: detection of recombination amongst aligned sequences. Bioinformatics 16:562-563.
Needleman, S. B., and C. D. Wunsch. 1970. A general method applicable to the search for similarities in the amino acid sequence of two proteins. J. Mol. Biol. 48:443-453.
Nuegoonpipat, A. A., A. Berlioz-Arthaud, V. Chow, T. Endy, K. Lowry, Q. Mai le, T. U. Ninh, A. Pyke, M. Reid, J. M. Reynes, S. T. Su Yun, H. M. Thu, S. S. Wong, E. C. Holmes, and J. Aaskov. 2004. Sustained transmission of dengue virus type 1 in the Pacific due to repeated introductions of different Asian strains. Virology 329:505-512.
Palacios, G., O. Jabado, and D. Cisterna. 2005. Molecular identification of mumps virus genotypes from clinical samples: standardized method of analysis. J. Clin. Microbiol. 43:1869-1878.
Posada, D., and K. A. Crandall. 2001. Evaluation of methods for detecting recombination from DNA sequences: computer simulations. Proc. Natl. Acad. Sci. USA 98:13757-13762.
Posada, D., and K. A. Crandall. 1998. MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817-818.
Rice, P., I. Longden, and A. Bleasby. 2000. EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet. 16:276-277.
Rico-Hesse, R. 2003. Microevolution and virulence of dengue viruses. Adv. Virus Res. 59:315-341.
Rico-Hesse, R. 1990. Molecular evolution and distribution of dengue viruses type 1 and 2 in nature. Virology 174:479-493.
Salminen, M. O., J. K. Carr, D. S. Burke, and F. E. McCutchan. 1995. Identification of breakpoints in intergenotypic recombinants of HIV type 1 by bootscanning. AIDS Res. Hum. Retrovir. 11:1423-1425.
Smith, J. 1992. Analyzing the mosaic structure of genes. J. Mol. Evol. 34:126-129.
Tamura, K., and M. Nei. 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol. Biol. Evol. 10:512-526.
Thompson, J., T. Gibson, F. Plewniak, F. Jeanmougin, and D. Higgins. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876-4882.
Tolou, H. J., P. Couissinier-Paris, J. P. Durand, V. Mercier, J. J. de Pina, P. de Micco, F. Billoir, R. N. Charrel, and X. de Lamballerie. 2001. Evidence for recombination in natural populations of dengue virus type 1 based on the analysis of complete genome sequences. J. Gen. Virol. 82:1283-1290.
Twiddy, S. S., and E. C. Holmes. 2003. The extent of homologous recombination in members of the genus Flavivirus. J. Gen. Virol. 84:429-440.
Watts, D. M., K. R. Porter, and P. Putvatana. 1999. Failure of secondary infection with American genotype dengue 2 to cause dengue hemorrhagic fever. Lancet 354:1431-1434.
White, N. J. 1999. Variation in virulence of dengue virus. Lancet 354:1401-1402.
World Health Organization. 2002, posting date. Dengue and dengue haemorragic fever. Fact sheet 117. World Health Organization, Geneva, Switzerland.
World Health Organization. 1997. Dengue haemorrhagic fever. Diagnosis, treatment, prevention and control, 2nd ed. World Health Organization, Geneva, Switzerland.
Worobey, M., A. Rambaut, and E. C. Holmes. 1999. Widespread intra-serotype recombination in natural populations of dengue virus. Proc. Natl. Acad. Sci. USA 96:7352-7357.(C. Domingo, G. Palacios, )
Jerome L. and Dawn Greene Infectious Diseases Laboratory, Columbia University, New York, New York
Robert Koch Institute, Berlin, Germany
Centro de Salud Internacional, Hospital Clinic (IDIBAPS), Barcelona, Spain
ABSTRACT
Here we propose the use of a 216-nucleotide fragment located in the carboxyl terminus of the E gene (E-COOH) and a pairwise-based comparison method for genotyping of dengue virus 1 (DENV-1) strains. We have applied this method to the detection and characterization of DENV-1 in serum samples from travelers returning from the tropics. The results obtained with the typing system correlate with the results obtained by comparison of the sequences of the entire E gene of the strains. The approach demonstrates utility in plotting the distribution and circulation of different genotypes of DENV-1 and also suggests the presence of two new clades of Indian strains. The integration of the method with an online database and a typing characterization tool enhances its strength. Additionally, the analysis of the complete E gene of DENV-1 strains suggested the occurrence of a nondescribed recombination event in the China GD23-95 strain. We propose the use of this methodology as a tool for real-time epidemiological surveillance of dengue virus infections and their pathogenesis.
INTRODUCTION
Dengue is the most common and widespread arboviral disease worldwide, with at least 50 million cases recorded every year. Although the majority of infections with dengue viruses (DENV) are asymptomatic, the wide spectrum of disease ranges from mild, self-limited dengue fever to severe and potentially life-threatening dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS). Dengue fever is an acute febrile viral disease, with patients with dengue fever frequently presenting with headache, bone or joint and muscular pain, rash, and leukopenia. DHF is a more virulent form of dengue virus infection and is considered a different clinical entity; it may progress to hypovolemic shock (DSS). Each year nearly 500,000 cases of DHF-DSS require hospitalization, with case-fatality rates as high as 5%, depending on the availability of treatment (39, 40).
The increasing incidence of dengue epidemics in the Americas, Southeast Asia, the Indian subcontinent, and the Western Pacific, along with the growing rate of DHF and DSS cases, is a global public health problem. Urbanization, overpopulation, crowding, poverty, a weakened public health infrastructure, and the globalization of trade and travel have been implicated as causes for this increase (3, 9).
Dengue viruses comprise four different serotypes, and evolutionary and epidemiological studies have facilitated clustering of dengue viruses into genotypes within serotypes (14, 30). It was posited that dengue hemorrhagic fever is a result of the immune enhancement developed after subsequent infection with two different serotypes (10). This is also a plausible mechanism to explain the observed highest risk of DHF and DSS in primary infections in infants with heterotypic immunity due to waning maternal dengue virus antibodies (17). It has also been suggested that virulent strains present an increasing replication rate in human target cells (29); therefore, the virus itself might play an important role in disease severity. Moreover, some genotypes have been associated with DHF epidemics, while others genotypes have been responsible for mild disease in certain areas while causing secondary infections (11, 37, 38). It was posited that these genotypes may present different antigenic epitopes and that these structural differences correlate with pathogenesis (18, 21). Significant differences in disease severity are observed in secondary infections, depending on the origin of the strains (4, 18, 21). Thus, the presence of a specific genotype in an area could be associated with an increased risk for the appearance of DHF and DSS cases.
Accordingly, a surveillance methodology able to identify the presence or introduction of dengue genotypes in areas of endemicity is of great interest for pathogenesis studies and for the establishment of local epidemiological control policies.
Here we describe a rapid, sensitive method for the genotyping of DENV-1, in which a short fragment in the carboxyl terminus of the E gene (E-COOH) is amplified by PCR, sequenced, and analyzed in pairwise comparisons with prototypic DENV sequences representing DENV-1 strains. Application of this method to serum samples obtained from DENV-1-infected travelers who had recently returned to Europe from India suggested the presence of two new DENV-1 lineages. Sequence analysis of the entire E gene confirmed this observation and supported the utility of the method.
MATERIALS AND METHODS
Samples from dengue virus-infected patients. Samples were obtained from travelers with acute DENV-1 infection returning to Europe from regions where dengue is endemic (Table 1). Samples (sera and/or viral culture supernatants) were provided by virology research laboratories of the European Network for Diagnosis of "Imported" Viral Diseases (ENIVD) and Tropical Medicine Clinical Units from the European Network for Tropical Medicine (TropNetEurop). The samples were stored at –80°C until use.
RNA extraction and RT-PCR. Viral RNA was extracted from 140-μl serum samples or viral culture supernatants from DENV-1-infected patients by using the QIAamp viral RNA minikit (QIAGEN) and resuspended in 60 μl of water. Five microliters of RNA extract was subjected to nested reverse transcriptase PCR (RT-PCR; QIAGEN One-Step RT-PCR kit) to amplify a 328-nucleotide (nt) fragment spanning the E-NS1 junction of the DENV genome (5). The RT-PCR mixture contained 1x one-step RT-PCR buffer, 400 mM each deoxynucleoside triphosphate, 400 nM each sense and antisense degenerated primer (primers E/NS1-S and E/NS1-R), and an optimized combination of Omniscript and Sensiscript reverse transcriptases and HotStar Taq DNA polymerase. The RT-PCRs were carried out with a 50-μl reaction volume by using an initial reverse transcription step at 41°C for 45 min, followed by a denaturation and HotStar Taq polymerase activation step (94°C, 15 min) and 40 cycles of denaturation (94°C, 30 s), primer annealing (55°C, 1 min), and primer extension (72°C, 2 min). A final incubation was carried out at 72°C for 5 min. A second amplification reaction (nested PCR) was seeded with 1 μl of the initial amplification product. The 50 μl of the reaction mixture contained 60 mM Tris-HCl (pH 8.5), 2 mM MgCl2, 15 mM (NH4)2SO4, 800 nM each sense and antisense primer (primers E/NS1-SS and EGENE/NS1-RR), and 2.5 U of DNA Taq polymerase (Perkin-Elmer). The samples were subjected to a denaturation step (94°C, 15 min), followed by 40 cycles of denaturation (94°C, 30 s), primer annealing (57°C, 4 min), and primer extension (72°C, 2 min) and a further extension step at 72°C for 5 min.
The samples were also subjected to analysis by a specific DENV-1 nested RT-PCR to amplify the complete E gene. This protocol was performed by using the buffers described for the E-NS1 amplification nested RT-PCR, except for the use of specific primers (primers EGENE1-S, EGENE-R, EGENE1-SS, and EGENE/NS1-RR) that were designed based on published DENV-1 sequences by computer-assisted analysis (MACAW version 32 software, 1995; NCBI). The RT-PCRs were carried out by using an initial reverse transcription step at 48°C for 45 min, followed by a denaturation and HotStar Taq polymerase activation step (94°C, 15 min) and 40 cycles of denaturation (94°C, 30 s), primer annealing (56°C, 2 min), and primer extension (72°C, 2 min). A final incubation was carried out at 75°C for 5 min. In the second amplification reaction, the samples were subjected to a denaturation step (94°C, 5 min), followed by 40 cycles of denaturation (94°C, 30 s), primer annealing (59°C, 2 min), and primer extension (72°C, 2 min) and a further extension step at 75°C for 5 min.
Table 2 shows the sequences and the respective primer positions in the viral genome used for partial (328-nt) or whole (1,700-nt) E gene amplification. The sensitivities were 100 and 10,000 copies, respectively.
Nucleotide sequence analysis. The template for sequencing was obtained by excising the band of interest after size fractionation of the PCR products by agarose gel electrophoresis. Approximately 20 ng of cDNA was sequenced by using 40 pmol oligonucleotide primer and the ABI Prism Dye Terminator cycle sequencing ready reaction kit (Applied Biosystems). The sequence of the E-NS1 fragment was obtained by using a forward primer and a reverse primer flanking the amplification product (Table 2); 11 primers were used to obtain overlapping sequence of the E gene (Table 2). A total of 30 DENV-1 E-NS1 gene (328-nt) and 16 DENV-1 E gene (1,485-nt) sequences were obtained.
Original sequence data were first analyzed by use of CHROMAS software (version 1.3, 1996; C. McCarthy, School of Biomolecular and Biomedical Science, Faculty of Science and Technology, Griffith University, Brisbane, Queensland, Australia); the forward and reverse sequence data for each sample were aligned by using the program SEQMAN (DNASTAR Inc. Software, Madison, Wis.). The consensus sequence was compared and aligned to other samples or DNA database sequences using the program CLUSTAL X, version 1.83 (34).
Sequence analysis of amplified products. A set of 156 DENV-1 sequences (216 nt in length) comprising those from 30 new sequenced strains and 126 collected from GenBank (updated to July 2005) were used for phylogenetic analysis. To test the findings observed with the short fragment, the entire E protein-coding sequence (1,485 nt) was recovered directly from 16 clinical samples by nested RT-PCR and compared in pairwise analyses with DENV-1 complete E 116 sequences collected from GenBank.
Phylogenetic analysis. Phylogenetic analysis was performed by using the best model of nucleotide substitution (according to Modeltest [27] and Tamura and Nei [33], with correction for the proportion of invariable sites of 0.4434 and a gamma distribution of 1.4212). Programs from the MEGA package (version 3) (19) were used to produce phylogenetic trees, reconstructed by the neighbor-joining method. The statistical significance of a particular tree topology was evaluated by bootstrap resampling of the sequences 1,000 times. Pairwise comparisons of the DENV-1 database were done by global alignment by using the algorithm of Needleman and Wunsch (23), implemented by a program from EMBOSS, the European Molecular Biology Open Software Suite (28). An automated program that performs the same analysis is available at and http://aevi.isciii.es.
Recombination event detection. Systematic screening for the presence of recombination patterns was achieved by using the nucleotide alignments and the Recombination Detection Program (RDP; Darren Martin) (22). The algorithms Bootscan (31), MaxChi (32), Chimaera (26), LARD (Likelihood Assisted Recombination Detection) (13), and Phylip Plot (6) were also used for detection of recombination.
Nucleotide sequence accession numbers. The GenBank accession numbers of the nucleotide sequences determined in this study are DQ016630 to DQ016659.
RESULTS
Thirty symptomatic patients with DENV-1 infection were diagnosed and characterized by analysis of the carboxyl terminus of the E gene. The results are detailed in Table 1. The results of the sequence comparison suggested the existence of a new cluster of sequences in patients returning from India. Given the short nature of the fragment that was amplified, the entire sequence of the envelope glycoprotein was obtained for 16 samples.
Phylogenetic analysis using the complete envelope gene. The phylogenetic tree obtained by analysis of a complete data set of all available envelope gene sequences is shown in Fig. 1. The analysis not only allowed the classification within known DENV-1 genotypes, America-Africa (AMAF), Malaysia (MAL), Thailand (THAI), Asia (ASIA), and South Pacific (SP), but also suggested the existence of lineages with distinctive geographical and temporal relationships (16, 20).
The genotype AMAF sequences (Fig. 2) clustered in five well-defined lineages. Lineage AMERICA-1 comprises all known circulating South and Central American strains, as well as a virus (RIOH36589) isolated in Brazil from a traveler returning from Angola (7). Lineage AFRICA includes three East Africa strains. Lineage SOUTH ASIA includes three Myanmar strains circulating in 1976 and 1998. We have designated the remaining two lineages INDIA-1 and INDIA-2. One recombinant strain (Singapore-S275), with parents from different lineages, remains separated from the rest of the lineages.
The ASIA genotype sequences (Fig. 3) are clustered in three lineages: ASIA-1, which contains Japanese and old Hawaiian strains; ASIA-2, which includes strains recovered from 1980 to 2001 in China, Thailand, Laos, Myanmar, Taiwan, and Djibouti; and ASIA-3, which contains the most recent isolates (1993 to 2004) from Myanmar, Thailand, Cambodia, China, Malaysia, and Indonesia.
The SP genotype (Fig. 4) represents four different lineages: SP-1, which comprises isolates from Indonesia (1998), Polynesia, and Micronesia; SP-2, which comprises isolates from Indonesia (1998 to 2002) and Australia (1983); SP-3, which comprises strains from the Philippines (1995 to 2004) and a recent Indonesian isolate (SC01); and SP-4, which contains the oldest strains from the Philippines (1974 to 1984). One strain from China (GD23-95-China-1995), three isolates from America, and one isolate from Thailand clustered independently. Again, as described above, they are recombinants and their parents belong to different lineages.
The genotype MAL includes only the sylvatic Malaysian isolate. The genotype THAI clusters old Thailand isolates recovered from 1954 to 1964. Both MAL and THAI genotypes may be extinct, as no recent isolates have been found.
Recombination analysis. Recombination has been reported in the envelope gene in all DENV serotypes (35, 41). Because recombination events may occur at various positions in the envelope gene, phylogenetic trees based on partial envelope sequence data may be misleading. To address this concern we searched for evidence of recombination using the full envelope sequence. We detected previously reported recombination between lineages AMERICA-1 and ASIA-3 strains in S275-Singapore-1990, and between lineages AMERICA-1 and SP-3 in strains Caribbean 495-Mexico1985, Caribbean 925-Mexico1985, CV1636-Jamaica1977, and AHF82-80-Thailand1982 (35). Additionally, we found evidence of a new recombination event in the GD23-95-China-1995 sequence. This virus appears to be a recombinant of the ASIA-3 and SP-1 lineages (Fig. 5). All the recombinant strains appeared as outliers in our phylogenetic tree (Fig. 2 and 4).
Phylogenetic analysis of the carboxyl terminus of the envelope gene. Although phylogenetic assignment of dengue viruses is classically based on the sequence of the entire E gene, this approach is not practical for a real-time epidemiological surveillance of dengue virus infections and their pathogenesis, where high throughput is essential. More sequence data are available for the carboxyl terminus than the amino terminus of the E gene; furthermore, the carboxyl terminus contains regions of sequence conservation at positions 1850 to 2200 and 2500 to 2700 that are ideal for consensus PCR. We tested the ability of the carboxyl-terminal segment between positions 2196 and 2411 (BR/90 strain; GenBank accession no. AF226685) to predict phylogenetic assignments based on the complete E gene sequence. The topologies obtained were essentially the same. Genotypes ASIA, MAL, and THAI showed the same clustering. The changes in topology observed in genotypes AMAF and SP are explained by recombination events. Since in all the recombination events described the 5' breakpoint is located approximately 350 nt into the E gene and the 3' end breakpoint occurs in the NS1 gene, the analysis of the 3' portion of the E gene should allow detection of the inserted recombinant sequence. Consistent with this observation, strain GD23-95-China-1995, a recombinant of the ASIA-3 and SP-1 lineages, groups with the SP-1 lineage. Similarly, the cluster that includes recombinant strains from the AMERICA-1 and SP-3 lineages appears as a member of the SP-3 lineage; and the S275-Singapore strain, a recombinant between AMERICA-1 and ASIA-3 lineages, appears as a member of the AMERICA-1 lineage.
Pairwise analysis by use of local alignments. An all-against-all pairwise sequence comparison was done with the DENV-1 sequences to assess the potential for the establishment of a simple program for the classification of DENV-1 genotypes and lineages similar to programs that we have created for mumps virus (25) and adenovirus (1). Sequence analysis of the 216-nt region at the carboxyl terminus amplified by PCR with consensus primers SEQAS/E/NS1-SS and SEQBR/EGENE/NS1-RR (Table 2) was sufficient to allow determination of the genotype and the lineage (Fig. 6). The validity of this method was confirmed by analysis of variance by comparing the scores of the sequence comparisons within genotypes or lineages to the scores of comparisons between genotypes. A software tool has been developed to classify DENV-1 sequences by introducing the target sequence (either the complete sequence or the carboxyl E gene sequence) and is available at and http://aevi.isciii.es.
DISCUSSION
Viral nucleotide sequence data and phylogenetic methods have been widely used to understand genetic relationships between dengue viruses as well as their epidemiology (29). Others have tried to establish phylogenetic relationships based on a partial sequence, notably, the intersection between E and NS1 (111 nt from the 3' end of E and 129 nt from the 5' end of NS1) (30) and the amino terminus of E (2, 30). At the time that these regions were proposed for use for phylogenetic analyses, only a few sequence data were available for optimization and validation and the phenomenon of recombination in dengue viruses had not been thoroughly evaluated. Although dengue virus classification had been consistent when the full envelope sequence is used, amplification and sequencing analysis of nearly 1,500 nt are not feasible for real-time epidemiological surveillance. Here we describe a 216-nt fragment located in the carboxyl terminus of the E gene of DENV-1 that can be readily amplified by consensus PCR to delineate DENV-1 genotypes. The classification data obtained with the shorter sequence are similar to those obtained with the complete E gene, with one caveat, i.e., detection of recombination events.
Three (2, 15, 24) to five (7, 29) different genotypes of DENV-1 have been proposed. Our analysis of complete E gene sequences confirmed the presence of five genotypes and the clustering of sequences below the genotype level that correlate with the geographical origin and/or time of isolation and appear to represent distinct lineages with strong bootstrap support (Fig. 1).
A correlation between genotypes and pathogenesis had been suggested. It would be of interest to observe if the potential to cause DHF is a property shared at the genotype level or at the lineage level. The AMAF genotype has been described as a genotype with a high epidemiological impact due to its spread and its potential to cause DHF (29). Remarkably, two new clusters of Indian strains belonging to the genotype AMAF (lineages INDIA-1 and INDIA-2) were observed in our analysis. One lineage (INDIA-1) is related to American strains, while INDIA-2 is closer to the African and South Asiatic strains. Two different lineages are currently circulating in the subcontinent. From our data it could be inferred that this genotype is currently circulating in extensive areas both in the Americas (the AMERICA-1 lineage) and on the Indian subcontinent (the INDIA-1 and INDIA-2 lineages).
Given the geographical or temporal association among the sequences belonging to the same lineage, a classification based on lineages facilitates surveillance and tracking of dengue virus isolates. For example, in genotype ASIA, three lineages (the ASIA-1, ASIA-2, and ASIA-3 lineages) have been detected. The ASIA-2 lineage had been detected in Myanmar, Cambodia, and Thailand. We detected two members of ASIA-2 in patients returning from Kenya; this may suggest the introduction of this lineage into East Africa.
During the course of reviewing published sequence data, we discovered evidence of recombination between the ASIA-3 and the SP-1 lineages in GD23-95-China 1995. Since this virus appears to remain apart from the rest of the clusters (Fig. 5) and was isolated in 1995, it is possible that this lineage did not become established. Although recombination is accepted in Japanese encephalitis and St. Louis encephalitis viruses (8, 36), recombination in DENV has been controversial. The original reports of recombination in DENV by Holmes et al. (12) and Worobey et al. (41) have been challenged (7, 29). The only unequivocal way to resolve the scientific question with respect to GD23-95-China 1995 would be to obtain the isolate to independently confirm the results of sequence and phylogenetic analyses.
The use of a consensus PCR based on a short fragment of the E protein coupled to a similarity analysis by using a pairwise alignment method demonstrate that the DENV-1 genotype and lineage can be accurately defined. This method offers some additional advantages in microbiology laboratories, in contrast to the application of classical methods of molecular epidemiology (e.g., sequencing of the complete envelope from cell culture isolates). Among those advantages, we count the use of clinical samples as the source of viral RNA; the amplification of a short sequence to identify the serotype, the genotype, and the lineage, thus facilitating the acquisition of information in a very short time; the application of pairwise comparison analysis and the proposed online typing tool to identify DENV-1 genotypes, which minimizes the need for extensive phylogenetic analysis; and the use of an automatically updated database for comparison and a global analysis (e.g., all available DENV sequences are used each time), which minimizes the possibility of the most common mistake in phylogenetic analysis (e.g., use of a biased set of sequences for comparison). Only in the case that a new strain could not be clearly associated with a described genotype and/or lineage should the complete E gene be analyzed to identify the appearance of a new genotype. The detection and characterization of two new lineages of DENV-1 in India and the description of the spread of genotype ASIA, lineage ASIA-2, in East Africa clearly show the usefulness of this approach. This approach has been demonstrated to be equally useful for the genotyping of the four dengue virus serotypes (data not shown).
The use of this methodology for the identification of dengue virus infection could greatly contribute to the acquisition of dengue epidemiological data worldwide, detection of the appearance and spread of new genotypes and lineages, and detection of the spread and circulation of those already described and, more deeply, to a basic understanding of dengue virus pathogenesis.
ACKNOWLEDGMENTS
We are grateful to the members of ENIVD and TropNetEurop for providing the DENV isolates. This study would not have been possible without their help. We thank F. Molero for technical assistance with the processing of samples.
M.N. and A.T. are ENIVD participants . Other ENIVD participants include P. Cassinotti and D. Schultze (Institute für Klinische Mikrobiologie und Immunologie, St. Gallen, Switzeralnd), M. Van Esbroek (Institute of Tropical Medicine, Antwerp, Belgium), A. Fomsgaard and L. Vinner (Statens Serum Institute, Copenhagen, Denmark), and M. Grandadam (Unite de Virologie Tropicale Parc du Pharo, Marseille Armees, France). J.G. is TropNetEurop participant. Other TropNetEurop participants include S. Puente (Hospital Carlos III, Madrid, Spain), O. Whichmann (Institute for Tropical Medicine, Berlin, Germany), M. Schunk (Medizinische Klinik Klinikum der Universitt, Munich, Germany), and R. Lopez-Velez (Hospital Ramon y Cajal, Madrid, Spain). M.N. is a representative of the ENIVD Dengue Study Workgroup, and J.G. is a representative of the TropNetEurop Dengue Study Workgroup. A.T., J.G., and C.D., as well as S. Puente, are researchers of the RICET Network (Red de Investigacion Cooperativa en Enfermedades Tropicales C03/04).
C.D. is under contract by an agreement between the Public Health Division of the Spanish Ministry of Health (DGSP-MSC) and the Instituto de Salud Carlos III (ISCIII) for the development of the Hemorrhagic Viral Fevers Surveillance and Control Program. This work has received financial support from the Instituto de Salud Carlos III (ISCIII) through research project grants (grants MPY 1194/02 and C03/04). G.P. and W.I.L. are supported by the Ellison Medical Foundation and the National Institutes of Health (grants AI 51292 and U54 AI57158 to W.I.L.).
Both authors contributed equally.
REFERENCES
Casas, I., A. Avellon, M. Mosquera, O. Jabado, J. E. Echevarria, R. H. Campos, W. I. Lipkin, P. Perez-Brea, and G. Palacios. 2005. Molecular identification of adenoviruses in clinical samples by analyzing a partial hexon genomic region. J. Clin. Microbiol. 43:6176-6182.
Chungue, E., O. Cassar, M. T. Drouet, M. G. Guzman, M. Laille, L. Rosen, and V. Deubel. 1995. Molecular epidemiology of dengue-1 and dengue-4 viruses. J. Gen. Virol. 76(Pt 7):1877-1884.
Clarke, T. 2002. Dengue virus: break-bone fever. Nature 416:672-674.
Cologna, R., P. M. Armstrong, and R. Rico-Hesse. 2005. Selection for virulent dengue viruses occurs in humans and mosquitoes. J. Virol. 79:853-859.
Domingo, C., G. Palacios, M. Niedrig, M. Cabrerizo, O. Jabado, N. Reyes, W. I. Lipkin, and A. Tenorio. 2004. A new tool for the diagnosis and molecular surveillance of dengue infections in clinical samples. Dengue Bull. 28:87-95.
Felsenstein, J. 1989. PHYLIP: phylogeny inference package (version 3.2). Cladistics 5:164-166.
Goncalvez, A. P., A. A. Escalante, F. H. Pujol, J. E. Ludert, D. Tovar, R. A. Salas, and F. Liprandi. 2002. Diversity and evolution of the envelope gene of dengue virus type 1. Virology 303:110-119.
Gould, E. A., X. de Lamballerie, P. M. Zanotto, and E. C. Holmes. 2003. Origins, evolution, and vector/host coadaptations within the genus Flavivirus. Adv. Virus Res. 59:277-314.
Gubler, D. J. 2002. Epidemic dengue/dengue hemorrhagic fever as a public health, social and economic problem in the 21st century. Trends Microbiol. 10:100-103.
Halstead, S. B. 1981. The Alexander D. Langmuir Lecture. The pathogenesis of dengue. Molecular epidemiology in infectious disease. Am. J. Epidemiol. 114:632-648.
Halstead, S. B., T. G. Streit, and J. G. Lafontant. 2001. Haiti: absence of dengue hemorrhagic fever despite hyperendemic dengue virus transmission. Am. J. Trop. Med. Hyg. 65:180-183.
Holmes, E., M. Worobey, and A. Rambaut. 1999. Phylogenetic evidence for recombination in dengue virus. Mol. Biol. Evol. 16:405-409.
Holmes, E. C. 1998. Molecular epidemiology of dengue virus—the time for big science. Trop. Med. Int. Health 3:855-856.
Holmes, E. C., and S. S. Burch. 2000. The causes and consequences of genetic variation in dengue virus. Trends Microbiol. 8:74-77.
Hwang, K. P., P. Y. Chu, Y. C. Tung, H. L. Wang, Y. Y. Yueh, Y. C. Wu, C. Chin, and K. H. Lin. 2003. Molecular epidemiological study of dengue virus type 1 in Taiwan. J. Med. Virol. 70:404-409.
Kanesa-Thasan, N., G. J. Chang, B. L. Smoak, A. Magill, M. J. Burrous, and C. H. Hoke, Jr. 1998. Molecular and epidemiologic analysis of dengue virus isolates from Somalia. Emerg. Infect. Dis. 4:299-303.
Kilks, S. C., S. Nimmanitya, A. Nisalak, and D. S. Burke. 1988. Evidence that maternal antibodies are important in the development of dengue haemorrhagic fever in infants. Am. J. Trop. Med. Hyg. 38:411-449.
Kochel, T. J., D. M. Watts, S. B. Halstead, C. G. Hayes, A. Espinoza, V. Felices, R. Caceda, C. T. Bautista, Y. Montoya, S. Douglas, and K. Russell. 2002. Effect of dengue-1 antibodies on American dengue-2 viral infection and dengue hemorrhagic fever. Lancet 360:310-312.
Kumar, S., K. Tamura, I. B. Jakobsen, and M. Nei. 2001. MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17:1244-1245.
Kurane, I., and T. Takasaki. 2001. Dengue fever and dengue haemorrhagic fever: challenges of controlling an enemy still at large. Rev. Med. Virol. 11:301-311.
Leitmeyer, K. C., D. W. Vaughn, D. M. Watts, R. Salas, I. Villalobos de Chacon, C. Ramos, and R. Rico-Hesse. 1999. Dengue virus structural differences that correlate with pathogenesis. J. Virol. 73:4738-4747.
Martin, D., and E. Rybicki. 2000. RDP: detection of recombination amongst aligned sequences. Bioinformatics 16:562-563.
Needleman, S. B., and C. D. Wunsch. 1970. A general method applicable to the search for similarities in the amino acid sequence of two proteins. J. Mol. Biol. 48:443-453.
Nuegoonpipat, A. A., A. Berlioz-Arthaud, V. Chow, T. Endy, K. Lowry, Q. Mai le, T. U. Ninh, A. Pyke, M. Reid, J. M. Reynes, S. T. Su Yun, H. M. Thu, S. S. Wong, E. C. Holmes, and J. Aaskov. 2004. Sustained transmission of dengue virus type 1 in the Pacific due to repeated introductions of different Asian strains. Virology 329:505-512.
Palacios, G., O. Jabado, and D. Cisterna. 2005. Molecular identification of mumps virus genotypes from clinical samples: standardized method of analysis. J. Clin. Microbiol. 43:1869-1878.
Posada, D., and K. A. Crandall. 2001. Evaluation of methods for detecting recombination from DNA sequences: computer simulations. Proc. Natl. Acad. Sci. USA 98:13757-13762.
Posada, D., and K. A. Crandall. 1998. MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817-818.
Rice, P., I. Longden, and A. Bleasby. 2000. EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet. 16:276-277.
Rico-Hesse, R. 2003. Microevolution and virulence of dengue viruses. Adv. Virus Res. 59:315-341.
Rico-Hesse, R. 1990. Molecular evolution and distribution of dengue viruses type 1 and 2 in nature. Virology 174:479-493.
Salminen, M. O., J. K. Carr, D. S. Burke, and F. E. McCutchan. 1995. Identification of breakpoints in intergenotypic recombinants of HIV type 1 by bootscanning. AIDS Res. Hum. Retrovir. 11:1423-1425.
Smith, J. 1992. Analyzing the mosaic structure of genes. J. Mol. Evol. 34:126-129.
Tamura, K., and M. Nei. 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol. Biol. Evol. 10:512-526.
Thompson, J., T. Gibson, F. Plewniak, F. Jeanmougin, and D. Higgins. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876-4882.
Tolou, H. J., P. Couissinier-Paris, J. P. Durand, V. Mercier, J. J. de Pina, P. de Micco, F. Billoir, R. N. Charrel, and X. de Lamballerie. 2001. Evidence for recombination in natural populations of dengue virus type 1 based on the analysis of complete genome sequences. J. Gen. Virol. 82:1283-1290.
Twiddy, S. S., and E. C. Holmes. 2003. The extent of homologous recombination in members of the genus Flavivirus. J. Gen. Virol. 84:429-440.
Watts, D. M., K. R. Porter, and P. Putvatana. 1999. Failure of secondary infection with American genotype dengue 2 to cause dengue hemorrhagic fever. Lancet 354:1431-1434.
White, N. J. 1999. Variation in virulence of dengue virus. Lancet 354:1401-1402.
World Health Organization. 2002, posting date. Dengue and dengue haemorragic fever. Fact sheet 117. World Health Organization, Geneva, Switzerland.
World Health Organization. 1997. Dengue haemorrhagic fever. Diagnosis, treatment, prevention and control, 2nd ed. World Health Organization, Geneva, Switzerland.
Worobey, M., A. Rambaut, and E. C. Holmes. 1999. Widespread intra-serotype recombination in natural populations of dengue virus. Proc. Natl. Acad. Sci. USA 96:7352-7357.(C. Domingo, G. Palacios, )