Evolutionary relationships of endemic/epidemic and sylvatic dengue viruses

Phylogenetic relationships and differential selection pressures among genotypes of dengue-2 virus

To elucidate the processes controlling the emergence and spread of dengue-2 virus (DEN-2) we examined the evolution of viral isolates sampled from both local (Viet Nam) and global populations. Our phylogenetic analysis, incorporating envelope (E) glycoprotein sequences from 147 isolates of DEN-2, provided a more complete picture of viral diversity, with a newly defined "Cosmopolitan" genotype having a near global distribution and two other genotypes restricted to Asia. By analyzing rates of synonymous and nonsynonymous substitution we determined that genotypes have experienced different selection pressures, with some evidence of positive selection in the Cosmopolitan genotype and one of the two Asian genotypes, but that the transition from sylvatic to human transmission was not accompanied by adaptive evolution of the E gene. Although there was no association between selection pressures acting on the E gene and proposed virulence differences among genotypes, some putatively selected amino acid sites have previously been implicated in changing viral pathogenicity, most notably E-390, and may also affect transmittability. These findings have implications for the future spread of DEN-2.

Phylogenetic analysis of the envelope protein (domain III) of dengue 4 viruses

OBJECTIVE

To evaluate the genetic variability of domain III of envelope (E) protein and to estimate phylogenetic relationships of dengue 4 (Den-4) viruses isolated in Mexico and from other endemic areas of the world.

MATERIAL AND METHODS

A phylogenetic study of domain III of envelope (E) protein of Den-4 viruses was conducted in 1998 using virus strains from Mexico and other parts of the world, isolated in different years. Specific primers were used to amplify by RT-PCR the domain III and to obtain nucleotide sequence. Based on nucleotide and deduced aminoacid sequence, genetic variability was estimated and a phylogenetic tree was generated. To make an easy genetic analysis of domain III region, a Restriction Fragment Length Polymorphism (RFLP) assay was performed, using six restriction enzymes.

RESULTS

Study results demonstrate that nucleotide and aminoacid sequence analysis of domain III are similar to those reported from the complete E protein gene. Based on the RFLP analysis of domain III using the restriction enzymes NIa III, Dde I and Cfo I, Den-4 viruses included in this study were clustered into genotypes 1 and 2 previously reported.

CONCLUSIONS

Study results suggest that domain III may be used as a genetic marker for phylogenetic and molecular epidemiology studies of dengue viruses. The English version of this paper is available too at: http://www.insp.mx/salud/index.html.

Molecular epidemiology of dengue type 2 virus in Venezuela: evidence for in situ virus evolution and recombination

Epidemic outbreaks of dengue fever (DF) were first recorded in Venezuela in 1978 and were followed by the emergence of dengue haemorrhagic fever (DHF) outbreaks in 1989. To gain a better understanding of the nature of these epidemics, the complete envelope (E) gene sequence of 34 Venezuelan dengue type 2 (DEN-2) viruses, isolated between 1997 and 2000 was determined. Of these isolates, 16 were from patients with DF and 17 were from patients diagnosed with DHF. There were no diagnostic sequence differences between them, suggesting that the E gene alone does not determine disease severity. These sequence data were also used in phylogenetic comparisons with a global sample of DEN-2 viruses, including strains collected previously from Venezuela. This analysis revealed that the ancestors of the Venezuelan viruses were Asian in origin, implying that a DEN-2 virus strain from this region was introduced into Venezuela and the wider Caribbean region during the late 1970s or the early 1980s. The phylogenetic trees further indicate that evolution of DEN-2 virus in Venezuela has occurred in situ, with differentiation into a number of distinct but co-circulating lineages, rather than the repeated introduction of new strains from other localities. By incorporating additional sequence data from the virus capsid, premembrane and membrane genes, evidence is provided that a single Venezuelan strain sequenced previously, designated Mara4, is a recombinant virus, incorporating genome sequence from Venezuelan and Asian parental viruses.

Evolutionary change of predicted cytotoxic T cell epitopes of dengue virus

Peptides bound by human class I major histocompatibility complex molecules and presented to cytotoxic T cells (CTL) were predicted by a computer algorithm, and changes in these predicted CTL epitopes (pCTLE) were reconstructed by maximum parsimony in a phylogeny of complete genotypes of dengue virus, West Nile virus (WNV), and Japanese encephalitis virus (JEV). In a clade including dengue virus serotypes 1 and 3 (D1V and D3V), pCTLE were lost over evolutionary time at a greater rate than new ones were gained. A similar but less pronounced trend was seen in the other dengue serotypes but not in WNV and JEV. The loss of pCTLE predicted to be restricted by HLA-B(*)2705 was particularly pronounced in D1V and D3V. Since dengue is endemic in humans, while WNV and JEV are not, these results are consistent with CTL-driven selection on dengue viruses, particularly, D1V and D3V.

Nucleotide sequencing and serological evidence that the recently recognized deer tick virus is a genotype of Powassan virus

Deer tick virus (DTV) is a recently recognized North American virus isolated from Ixodes dammini ticks. Nucleotide sequencing of fragments of structural and non-structural protein genes suggested that this virus was most closely related to the tick-borne flavivirus Powassan (POW), which causes potentially fatal encephalitis in humans. To determine whether DTV represents a new and distinct member of the Flavivirus genus of the family Flaviviridae, we sequenced the structural protein genes and 5' and 3' non-coding regions of this virus. In addition, we compared the reactivity of DTV and POW in hemagglutination inhibition tests with a panel of polyclonal and monoclonal antisera, and performed cross-neutralization experiments using anti-DTV antisera. Nucleotide sequencing revealed a high degree of homology between DTV and POW at both nucleotide (>80% homology) and amino acid (>90% homology) levels, and the two viruses were indistinguishable in serological assays and mouse neuroinvasiveness. On the basis of these results, we suggest that DTV should be classified as a genotype of POW virus.

Interferon inhibits dengue virus infection by preventing translation of viral RNA through a PKR-independent mechanism

Previously, we demonstrated that pretreatment of cells with interferon (IFN) alpha + gamma or beta + gamma inhibited dengue virus (DV) replication. In this study, experiments were performed to better define the mechanism by which IFN blocks the accumulation of dengue virus (DV) RNA. Pretreatment of human hepatoma cells with IFN beta + gamma did not significantly alter virus attachment, viral entry, or nucleocapsid penetration into the cytoplasm. The inhibitory effect of IFN was retained even when naked DV RNA was transfected directly into cells, confirming that steps associated with viral entry were not the primary target of IFN action. Biosynthetic labeling experiments revealed that IFN abolished the translation of infectious viral RNA that occurred prior to RNA replication. Subcellular fractionation experiments demonstrated that IFN did not significantly alter the ability of viral RNA to attach to ribosomes. The antiviral effect of IFN appeared independent of the IFN-induced, double-stranded RNA-activated protein kinase (PKR) and RNase L, as genetically deficient PKR- RNase L- cells that were infected by DV retained sensitivity to inhibition by IFN. We conclude that IFN prevents DV infection by inhibiting translation of the infectious viral RNA through a novel, PKR-independent mechanism.

Phylogenetic relationships of flaviviruses correlate with their epidemiology, disease association and biogeography

Phylogenetic analysis of the Flavivirus genus, using either partial sequences of the non-structural 5 gene or the structural envelope gene, revealed an extensive series of clades defined by their epidemiology and disease associations. These phylogenies identified mosquito-borne, tick-borne and no-known-vector (NKV) virus clades, which could be further subdivided into clades defined by their principal vertebrate host. The mosquito-borne flaviviruses revealed two distinct epidemiological groups: (i) the neurotropic viruses, often associated with encephalitic disease in humans or livestock, correlated with the Culex species vector and bird reservoirs and (ii) the non-neurotropic viruses, associated with haemorrhagic disease in humans, correlated with the Aedes species vector and primate hosts. Thus, the tree topology describing the virus-host association may reflect differences in the feeding behaviour between Aedes and Culex mosquitoes. The tick-borne viruses also formed two distinct groups: one group associated with seabirds and the other, the tick-borne encephalitis complex viruses, associated primarily with rodents. The NKV flaviviruses formed three distinct groups: one group, which was closely related to the mosquito-borne viruses, associated with bats; a second group, which was more genetically distant, also associated with bats; and a third group associated with rodents. Each epidemiological group within the phylogenies revealed distinct geographical clusters in either the Old World or the New World, which for mosquito-borne viruses may reflect an Old World origin. The correlation between epidemiology, disease correlation and biogeography begins to define the complex evolutionary relationships between the virus, vector, vertebrate host and ecological niche.

Genetic variation in the 3' non-coding region of dengue viruses

The 3' non-coding region (3'NCR) of strains of dengue 1 (DEN 1), DEN 2, DEN 3, and DEN 4 viruses, isolated in different geographical regions, was sequenced and compared to published sequences of the four dengue viruses. A total of 50 DEN 2 strains was compared: 7 West African strains, 3 Indonesian mosquito strains, 1 Indonesian macaque isolate, and 39 human isolates from Southeast Asia, the South Pacific, and the Caribbean and Americas. Nucleotide sequence alignment revealed few deletions and no repeat sequences in the 3' NCR of DEN 2 viruses and showed that much of the 3' NCR was well conserved. The strains could be divided into two groups, sylvatic and human/mosquito/macaque, based on nucleotide sequence homology. A hypervariable region was identified immediately following the NS5 stop codon, which involved a 2-10 nucleotide deletion in human, mosquito, and macaque isolates compared with the sylvatic strains. The DEN 2 3'NCR was also compared with 3'NCR sequences from strains of DEN 1, DEN 3, and DEN 4 viruses. DEN 1 was found to have four copies of an eight nucleotide imperfect repeat following the NS5 stop codon, while DEN 4 virus had a deletion of 75 nucleotides in the 3'NCR. We propose that the variation in nucleotide sequence in the 3'NCR may have evolved as a function of DEN virus transmission and replication in different mosquito and non-human primate/human host cycles. The results from this study are consistent with the hypothesis that DEN viruses arose from sylvatic progenitors and evolved into human epidemic strains. However, the data do not support the hypothesis that variation in the 3'NCR correlates with DEN virus pathogenesis.

Identification of arboviruses and certain rodent-borne viruses: reevaluation of the paradigm

Diagnostic and epidemiologic virology laboratories have in large part traded conventional techniques of virus detection and identification for more rapid, novel, and sensitive molecular methods. By doing so, useful phenotypic characteristics are not being determined. We feel that the impact of this shift in emphasis has impaired studies of the biology of viruses. This position paper is a plea to the scientific and administrative communities to reconsider the importance of such information. We also suggest a revised paradigm for virus isolation and characterization and provide a rationale for accumulating biologic (phenotypic) information.