| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In French Polynesia, an alarming increase in dengue transmission has been observed since the dengue-2 (DEN-2) outbreak in 1971, in which there were more than 35,000 cases, including the first three deaths recorded.5 Subsequently, dengue outbreaks involving all four dengue serotypes (DEN-1, DEN-2, DEN-3, and DEN-4) have occurred: DEN-1 in 1975–1976, 1988–1989, and 2001; DEN-4 in 1979; DEN-3 in 1989–1990, and DEN-2 in 1996–1997.6,7 No co-circulation of the four serotypes has been observed in French Polynesia.
To determine the origin and evolution of dengue viruses and to examine virus diversity, sequence analysis of the pre-membrane (prM) locus,8,9 the envelope (E) gene, the E/non-structural 1 (NS1) junction region, and other genes such as nonstructural 3 (NS3) gene has been undertaken by several groups. With regard to DEN-1 viruses, studies suggest there up to five DEN-1 virus genotypes or groups that cluster geographically.10–12
The viral E protein of dengue virus contains the sites required for virus binding and penetration into susceptible cells. Within the E protein, neutralizing epitopes are localized with two potential N-linked glycosylation sites, Asn-67 and Asn-153. Mutations of these N-linked glycosylation sites may affect virus-mediated membrane fusion and neurovirulence.13,14 Therefore, most sequence analysis of the dengue viruses have been performed on this gene.
For surveillance and epidemiologic studies, we compared the entire DEN-1 virus E gene sequences derived from French Polynesian strains isolated during the 1988–1989 and 2001 epidemics with those described previously and available in the GenBank library (Table 1
). The E gene sequence of the Singapore S275/90 strain (SIN90) included in this study was that revised on July 5, 2002 and is also available on the GenBank library. No deletion or extra cysteine residues at amino acid positions 389–390 were observed in this latter strain.
|
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Extraction of RNA and reverse transcriptase-polymerase chain reaction (RT-PCR). The virus genomic RNA was extracted from either acute-phase sera from a DF patient directly (FP89) or from the culture media virus from infected C6/36 cells collected after one passage (FP01) using the QIAamp viral RNA mini kit (Qiagen Inc., Hilden, Germany) according to manufacturer’s instructions. Briefly, a 140 µL aliquot of serum or infected cellular supernatant was mixed with the buffer AVL-carrier RNA, incubated for 10 minutes, and loaded onto the spin column. This was followed by two washes with buffer AW. The final nucleic acid extracts were obtained in a total volume of 60 µL of elution buffer AVE.
Oligonucleotide primers (both sense and antisense) used in the amplification and sequencing protocols were designed with the OMIGA 1.1 software (Oxford Molecular Ltd., Oxford, United Kingdom). For the entire E gene, three primer pairs were used as follows: D1/732 (5'-GAAACAAGAAC-CGAAACG-3') and D1/1509 (5'-CCATCTCAT-TAAAGTCCAGC-3'), D1/1399 (5'-CAATTGCAACCAT-AACACC-3') and D1/2106 (5'-TGCTTCCTTTCTT-GAACC-3'), and D1/1966 (5'-GATTGATAACAGC-CAATCC-3') and D1/2685 (5'-CAACAACATCTCCTA-CAACC-3'). The nucleic acid sequences of these three E selected regions of DEN-1 virus were amplified using the OneStep RT-PCR Kit (Qiagen Inc.). Briefly, 10 µL of extracted RNA were added to 40 'L of an RT-PCR mixture as recommended by the manufacturer. Reverse transcription was performed at 50°C for 30 minutes, followed directly by 30 cycles of amplification at 94°C for 30 seconds, 54°C for one minute, and 72°C for one minute, and a final extension at 72°C for 10 minutes. The RT-PCR was performed in single-tube (0.2 'L) in a thermocycler (Model Mastercycler'Gradient 5331; Eppendorf, Hamburg, Germany). Electrophoresis of the amplified DNA products was then performed on a 1.5% (w/v) agarose gel and staining with ethidium bromide (0.5 µg/mL). The expected band was sliced from the gel, and purified using the QIAquick gel extraction kit (Qiagen Inc).
Sequencing of DNA. Sequencing reactions were performed using the BigDyeTM Terminator Cycle Sequencing Ready Reaction Kit (Perkin-Elmer Corp.-Applied Biosystems Inc., Warrington, United Kingdom). Unincorporated dideoxy terminators were removed using the DyeEx Spin columns (Qiagen Inc). The products were analyzed using an automated ABI Prism 310 genetic analyzer (Applied Biosystems, Foster City, CA). Both strands of the cDNA PCR products were sequenced.
Phylogenetic analysis.
The overlapping nucleic acid sequences obtained from individual sequencing reactions were combined for analysis and edited using the AutoAssembler 2.1 software (Perkin-Elmer Corp.-Applied Biosystems Inc.). Multiple alignment of sequences from French Polynesian isolates and the published E DEN-1 virus sequences accessed from the GENBANK library (Table 1
) was performed using the CLUSTAL W 1.8116 algorithm with default parameters. A phylogenetic tree was constructed using the neighbor joining method17 using maximum likelihood distance parameters (Kimura two-parameter formula) using the Molecular Evolutionary Genetics Analysis software package.18 Bootstrap analysis with 1,000 replicates was used to determine the robustness of the tree and the evolutionary relationship of the viruses.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sequence comparisons of the entire E gene of the 31 DEN-1 virus isolates showed a maximum divergence of 10.2% between BR971 and FP01, which translates into a 1.6% divergence at the protein level. The minimum nucleotide divergence of 0.2% was observed between two Brazilian 1997 strains, BR972 and BR974, with 0.2% amino acid divergence (Table 2). It is interesting to note that the FP01 strain is the most divergent DEN-1 virus isolate at the gene level (9.2–10.2%), while the Australian strain AUS83 is the most divergent at the protein level (1.8–3.6%).
|
According to the classification of Rico-Hesse10 in which maximum genetic distance within a given genotypic group or subgroup was <6%, five genotypes were observed from the complete sequence of the E gene of 31 DEN-1 virus strains involved worldwide. Phylogenetic analysis of the E gene sequences using the neighbor-joining method (Figure 1) showed these five clusters and confirmed the results of previous studies.12 These include cluster I, the first dengue-1 virus strains from Japan, Hawaii, Asia, and Djibouti; cluster II, Thai strains; cluster III, the Malaysian sylvatic strain; cluster IV, strains from southeast Asia, the south Pacific, and Australia; and cluster V, strains from the Americas, Africa, and southeast Asia.
|
). Within a genotype, nucleotide divergence among viruses was 4.0% for genotype I, ranging from 0.3% to 5.4% and 0.2% to 5.3% for genotypes IV and V, respectively. Within genotype V, FGA89 and ARU85 are the most closely related strains to the FP89 isolate, with 0.8% divergence at the nucleotide level and 0.6% and 0.4% divergence at amino acid level, respectively. Moreover, within genotype IV, DEN-1 virus strain FP01 showed the highest (5.4%) nucleotide divergence rate when compared with the THA80 strain (AHF82–80 strain) isolated in Thailand in 1980 and the lowest (3.3%) nucleotide divergence rate when compared with the A88 strain, an Indonesian strain isolated in 1988 from a DHF patient.19 It is interesting to note the lowest divergence (3.6%) at nucleotide level of this strain with the revised S275/90 strain.
The deduced amino acid sequences of the entire E gene were aligned and compared. The E proteins have retained an amino acid sequence similarity greater than 96% over the studied 46-year period. These data suggest that domains responsible for predicted flavivirus protein architecture and/or biologic function are strictly conserved. More than 80% of the nucleotide mutations were silent and among the other mutations, most are transitions.
Within the E glycoprotein, 59 of 495 amino acid changes were observed when the worldwide DEN-1 virus strains were compared. The amino acid sequence changes are summarized in Table 3.
|
). The five amino acid substitutions at positions 37 (domain I), 339, 352, 379 (domain III), and 439 (transmembrane region) are non-conservative. Domain III (amino acids 302–404) is thought to play an important role in cell attachment and tropism. The impact of these changes on biologic properties and disease outcome is not known. The six other amino acid substitutions at positions 88, 114, 287, 297, 369, and 473 are conservative.
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Studies of dengue virus molecular evolution and sequence data were performed to determine phylogenetic relationships of the viruses within each serotype. A threshold of 6% divergence is currently used to separate different genotypes within a dengue serotype.10,21 Several phylogenetic studies analyzing a large number of DEN-1 virus genomic sequences have been published.10,11 They analyzed either a 240 nucleotide sequence in the E/NS1 junction region of the genome or a 180 nucleotide sequence in the E gene and suggested that there were potentially five or three DEN-1 virus genotypes, respectively. A recent study, based on nucleotide sequences of the entire E gene,12 showed that DEN-1 viruses can be classified into five genotypes or monophyletic groups.
The present study confirmed the existence of these five genotypes with variations in the classification of the Singapore S275/90 strain. This strain belongs to genotype IV instead of genotype V, as in the last analysis.12 No deletion at nucleotide positions 1,045–1,047 was observed in the July 5, 2002 revised sequence (1,485 nucleotides instead of 1,482 nucleotides). Moreover, the THA80 (AHF82–80) strain isolated in 1980 in Thailand was assigned to genotype IV in this study, which is different from the analysis of Goncalvez and others,12 in which Thai80 (genotype I) corresponded to another isolate, the PUO 359 strain, also isolated in 1980 in Thailand. American and African strains were distinct from other isolates. The Malaysian sylvatic monkey isolate represents one distinct genotype, although bootstrap values supporting these groupings were low (Figure 1).
Bootstrap analysis showed that while several of the major nodes of the tree were not well supported (values 
60%), many others were robust (values 
70%).22 In most phylogenetic analyses, bootstrap values provide highly conservative estimates of the probability of correctly inferring the corresponding clusters or clades.
The phylogram generated by the sequencing of the entire E gene showed that the French Polynesian isolate FP89 belonged to subtype V. The high similarity obtained by nucleotide and amino acid sequence between ARU85 and FP89 (99.2% and 99.6%, respectively) confirmed the American origin of DEN-1 viruses isolated in 1989 in French Polynesia, which were previously demonstrated by comparing a 180-basepair fragment in the E gene.11 For the latest DEN-1 virus involved in the epidemic in 2001 in French Polynesia, it was assigned to genotype IV. Pairwise comparisons of strains isolated in Indonesia in 1988 (A88) and in French Polynesia in 2001 (FP01) showed a nucleotide divergence of 3.3% for the entire E gene. This observation suggests that the recent and severe epidemic in French Polynesia was probably due to the introduction of genotype IV DEN-1 viruses that were previously involved in an outbreak of DHF in Jakarta, Indonesia in 1988,23 rather than the reemergence of a previously strain circulating in the area at a lower level between epidemics. It must be noted that the recombinant strain24 S275/90 (SIN90) is also closely related to the FP01 isolate with a 3.6% nucleotide divergence.
Several studies have purported to show that dengue virus genotypes differ in their ability to cause severe dengue.25,26 The two French Polynesian strains isolated in 1989 and 2001 belong to two different genotypes, V and IV, respectively. Epidemiologic differences in disease severity were observed between the DEN-1 epidemics in French Polynesia in 1989 and 2001: in 1989, only DF was reported whereas DHF/DSS occurred at a rate of 2.7 per 1,000 people in 2001. In addition, eight fatal cases were reported among children in the 2001 epidemic. Among the severe hospitalized cases, laboratory tests necessary to determine infection status (primary versus secondary) were performed for 156 cases. A total of 9.6% were primary infections and 73% of these were in individuals less than one year of age (Laille M and others, unpublished data). High-level transmission of this DEN-1 virus was also observed in Hawaii, the Samoan Islands,27 Fiji, and New Caledonia from 2001 to 2003.(Fijian and New Caledonian Health Authorities, unpublished data). The observed increased severity of DF since 2001 in the Pacific region might be associated with the introduction of this new genotype (IV) from Indonesia, as well as the Southeast Asian genotype DEN-2, which has been reported to potentially cause severe disease.28
No relatedness between nucleotide sequence and disease severity could be demonstrated when the French Polynesian E sequences determined from DF cases were compared. Sequence analysis of full-length of French Polynesian virus type 1 strain genomes (from DF and DHF/DSS cases) should be investigated to identify nucleotide and/or amino acid differences responsible for the virulent genotype.
The E protein sequence of geographically diverse DEN-1 virus strains is highly conserved, with an amino acid similarity of at least 96% (Table 2). Conservation of the 12 cysteine residues involved in disulfide bridge formation was observed. The glycine-rich internal element, the flavivirus fusion domain, (amino acids 98–111), is conserved among DEN-1 virus E proteins, as well as the neutralizing epitopes localized in the E protein with two potential N-linked glycosylation sites (Asn-67 and Asn-153).
Nucleotide changes were generally conservative, leading to few amino acid changes (Table 3). The impact of these amino acid changes on biologic properties and pathogenic properties is not known. Nevertheless, in the FP89 and FP01 strains and among the 11 amino acid substitutions, three significant changes were observed within domain III (amino acids 302–404) of the E protein: T 339 S, I 352 V, and I 379 V (Table 4). This domain seems to interact with cellular receptors for virus attachment and entry. Comparison of these three changes with the consensus deduced amino acid sequence generated from all included strains showed that the changes T 339 S and I 352 V may be responsible in part for the observed differences in terms of disease severity and incidence of DHF/DSS. Another observation was that DEN-1 virus FP01 had an N
D non-conservative change at position 37 (domain I) and a VI significant amino acid substitution at position 439 (transmembrane region) in the E protein.
Analyzing phylogenetic relationships of dengue virus strains associated with epidemics of DF compared with DHF/DSS, including our French Polynesian strains, may contribute to a better understanding of the pathogenesis of a dengue virus infection, particularly if the amino acid changes showed by sequence analysis may be related to an increase in virulence. Characterization of genetic determinants for virulence would provide further support for the hypothesis that epidemics of DHF/DSS may be caused by the circulation of viral strains with increased virulence28,29 and expand the range of pathogenic properties for increased transmittability (high viremia).
Received October 28, 2003. Accepted for publication May 28, 2004.
We are grateful to Dr. Taiana Darius for help and advice with phylogenetic analysis. We also thank Dr. Allison Imrie for reviewing the manuscript.
Authors’ address: Manola Laille and Claudine Roche, Laboratoire de Recherche en Virologie Médicale, Institut Louis Malardeé, PO Box 30, 98713 Papeete Tahiti, Polynésie Française, Telephone: 689-41-64-68, Fax: 689-43-15-90, E-mails: mlaille{at}ilm.pf and croche{at}ilm.pf.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
