| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Short Communication |
1 Center for Infectious Disease Dynamics, Department of Biology, The Pennsylvania State University, Mueller Laboratory, University Park, PA 16802, USA
2 Department of Virology, US Army Medical Component-Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand
3 Military Infectious Diseases Research Program, US Army Medical Research and Materiel Command, Fort Detrick, MD 21702, USA
4 Queen Sirikit National Institute of Child Health, Bangkok, Thailand
Correspondence
Edward C. Holmes
ech15{at}psu.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT MAIN TEXTREFERENCES |
|---|
Supplementary figures and tables are available in JGV Online.
|
MAIN TEXT |
|---|
|
TOPABSTRACTMAIN TEXTREFERENCES |
|---|
). While the majority of dengue virus (DENV) infections may be asymptomatic, a significant proportion of patients experience a self-limited febrile disease (DF), sometimes of sufficient seriousness to require hospitalization, and a minority develop serious dengue disease, often classified as dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS).
The mechanisms that underpin DENV pathogenesis remain uncertain. The most cited hypothesis is that severe disease results from antibody-dependent enhancement following sequential infections with multiple DENV serotypes (DENV-1 to DENV-4) (Halstead et al., 1980). However, it is possible that some cases of severe dengue disease that present as secondary infections are in fact caused by an overactive T-cell response (Mongkolsapaya et al., 2003). Alternatively, it is possible that particular strains of DENV have acquired mutations that increase their virulence. Specifically, an ‘Asian’ genotype of DENV-2 is thought to be more often associated with severe dengue disease in the Americas than the previously circulating ‘American’ genotype (Rico-Hesse et al., 1997). Although differences in the ability of DENV strains to infect different cell lines have been described in vitro (Cologna & Rico-Hesse, 2003), whether these explain large-scale epidemiological patterns is unclear.
The genome of DENV comprises a single, positive-sense RNA molecule of approximately 11 kb translated as a single polyprotein. The coding region is flanked by a 5'-untranslated region (UTR), of approximately 94 nt, and a 3'-UTR of 388–462 nt (Chambers et al., 1990). The 3'-UTR contains several conserved elements, designated CS1 (complementary or cyclization sequences), CS2 and RCS2 (repeated CS2), as well as a 3'-long stable hairpin (LSH) structure, which is conserved among all members of the family Flaviviridae (Hahn et al., 1987). The 3'-UTR is thought to play a pivotal role in viral biology (Alvarez et al., 2005a, b; Blackwell & Brinton, 1995; Chiu et al., 2005; Durbin et al., 2001; Hahn et al., 1987; Men et al., 1996; Proutski et al., 1997a; Yu & Markoff, 2005; Zeng et al., 1998). In particular, it has been proposed that the 3'-UTR forms conserved secondary structures that interact with viral and host nucleic or protein factors to form a complex involved in the regulation of transcription and replication of minus-strand RNA, and which may also enhance the efficiency of translation.
It has been suggested that there is an association between specific RNA secondary structures in the 3'-UTR of DENV and disease outcome (Leitmeyer et al., 1999; Mangada & Igarashi, 1997; Proutski et al., 1997b). For example, it has been proposed that Asian and American genotypes of DENV-2 exhibit diagnostic differences in the 3'-UTR (and 5'-UTR) such that it is a primary determinant of DHF/DSS (Leitmeyer et al., 1999). However, viral isolates associated with DF and DHF can have the same 3'-UTR sequence and hence predicted secondary structures (Mangada & Igarashi, 1998; Rodriguez-Roche et al., 2005; Shurtleff et al., 2001).
The role played by the 3'-UTR in determining the severity of dengue disease therefore remains unclear. A limitation of some previous studies is that they were carried out with either a small sample of viral isolates or only considered part of the 3'-UTR. To alleviate such problems, we explored the association between the structure of the DENV 3'-UTR and disease outcome using genome data from all four DENV serotypes, collected from a well-defined population in Thailand over a period of 30 years (Nisalak et al., 2003). These sequences were combined with a larger set of 3'-UTR sequences from all four DENV serotypes sampled on a global basis.
The consensus sequences of 32 complete genomes of DENV-1 to DENV-4 were sampled from dengue patients admitted to the Queen Sirikit National Institute of Child Health (QSNICH), Bangkok, Thailand from 1973 to 2003. Background epidemiological and methodical data are described elsewhere (Klungthong et al., 2004; Zhang et al., 2005, 2006). To complement these data, we compiled 61 complete DENV genome sequences from GenBank for which information on disease status was also available: DENV-1=7 (5 DF and 2 DHF/DSS), DENV-2=36 (14 DF and 22 DHF/DSS), DENV-3=18 (14 DF and 4 DHF/DSS), DENV-4=0. This provided sequences from a wide range of geographical locations and was termed our ‘global’ dataset. GenBank accession numbers and other relevant information are given in Supplementary Table S1 (available in JGV Online).
For each of the 32 Thai DENV isolates, we extracted the complete envelope (E) gene (1479–1485 bp) and 3'-UTR (388–462 nt) sequences. Sequence alignments within and among serotypes were constructed using the CLUSTAL X 1.8 program (Thompson et al., 1997) and checked manually. To determine the DENV genotype of each sequence, phylogenetic trees were inferred from their E gene sequences using the maximum-likelihood (ML) method available in the PAUP* package (Swofford, 2003). In all cases the GTR+I+
4 substitution model was employed with all parameters estimated from the data.
The 3'-UTR secondary structures were estimated using the MFOLD package (Zuker, 2003; Mathews et al., 1999). As MFOLD gives multiple meta-stable structures of a certain minimum energy level, we calculated two structures for each isolate, denoted ‘str1’ and ‘str2’. In all cases str2 is a suboptimal structure. We assume that the smaller the variance between these two structures the more reliable the prediction. The default parameters were used in all other cases. To ensure further that our structures were reliable, we also employed the RNA-fold program available in the Vienna RNA package (Hofacker et al., 1994), which has proven useful in predicting flavivirus 3'-UTR structures (Rauscher et al., 1997).
We undertook a comparative analysis of predicted 3'-UTR RNA secondary structures from a large sample of Thai and global isolates of DENV. In most cases, only small differences were observed between the two meta-stable structures predicted by MFOLD, and the structures predicted by MFOLD and the Vienna Package were generally consistent.
The extent of intra-serotype diversity differed considerably between serotypes, with the greatest diversity seen in DENV-4 (33 variable sites), followed by DENV-1 (26 variable sites), DENV-2 (24 variable sites) and DENV-3 (13 variable sites). Most variable sites are located in the hypervariable region (positions nt 0–146 in our four-serotype alignments) and may therefore affect the subsequent folding considerably. Indeed, the hypervariable region exhibits extensive size variation between serotypes, with a length of 136 nt in DENV-1, 121 nt in DENV-2, 116 nt in DENV-3 and only 48 nt in DENV-4.
We directed most attention to the 3'-LSH. As this structure is critical for virus viability, it is expected to be present in all viruses studied. In total, a full-length 3'-LSH was absent in 17 of 93 sequences from the global dataset, in either str1 or str2, or the structures predicted by the Vienna Package (Supplementary Table S1 available in JGV Online). In these cases, only part of the 3'-LSH structure was identified, comprising 
30–40 nt, instead of the expected 100 nt. Of the isolates with shortened 3'-LSH structures, DENV-1 isolates ThD1-0442/80 and ThD1-0673/80 are notable in that they are invariant in the 3'-LSH region itself, but exhibit 11 unique mutations upstream of the 3'-LSH, which may have disturbed this structure (Supplementary Fig. S1 available in JGV Online). Similarly, DENV-4 isolate ThD4-0734/00 lacks a complete 3'-LSH, again probably due to upstream sequence interference. The possible influence of upstream mutations sits in contrast to most models of 3'-UTR secondary structure, although it is unclear whether these aberrant 3'-LSH structures could be stabilized by cellular/viral proteins and/or long-range RNA–RNA interactions in vivo.
We next examined the disease association of each of the 3'-UTR structures. For the Thai data, DENV-1 is the most illuminating serotype (Fig. 1a). Our previous phylogenetic analysis (Zhang et al., 2005) revealed that 8 of 10 available sequences can be divided into four pairs. Of these, pairs ThD1-0442/80–ThD1-0673/80 and ThD1-0097/94–ThD1-0488/94 possess identical 3'-UTR sequences within each pair. However, one member of each pair is associated with DF and the other with DHF. Pairs ThD1-0008/81–ThD1-0336/91 and ThD1-0049/01–ThD1-S0102/01 exhibit a single nucleotide difference within each pair, but possess identical secondary structures. Again, one sequence of each pair is associated with DF and the other with DHF. The two unpaired isolates possess secondary structures that are unique within DENV-1 (data not shown), and are associated with different clinical outcomes: ThD1-S0081/82 with DF and ThD1-CN0323/91 with DHF.
|
). A very similar structure is found in isolates ThD2-0017/98 and ThD2-0026/88. Sequence ThD2-0284/90 has less similarity in secondary structure with the other viruses associated with DF and is the only representative of the Asian–American genotype. Four of five DHF isolates possess the same secondary structure, including structure MA (Fig. 2a). However, these four isolates have the same secondary structure as DF isolates ThD2-0263/95 and ThD2-0055/99, again showing that there is no clear association between secondary structure and disease syndrome. Finally, the remaining sequence from the DHF group (ThD2-0498/84) has very different structure (data not shown).
|
), while all DENV-3 isolates associated with DHF have different structures. However, isolate ThD3-0055/93 associated with DHF, has the same sequence, and hence RNA secondary structure, as the three DF strains. The two remaining DHF isolates (ThD3-0010/87 and ThD3-1687/98) have different structures (Supplementary Fig. S2 available in JGV Online), although they possess the same 3'-LSH as the DF strains.
Finally, for DENV-4, isolates ThD4-0017/97 (DF) and ThD4-0476/97 (DHF) have the same predicted structures (Fig. 1c). Further, ThD4-0087/77 associated with DHF and ThD4-0734/00 associated with DF share the same secondary structure (Supplementary Fig. S3 available in JGV Online). The structures of two viruses – ThD4-0348/91 associated with DF and ThD4-0485/01 associated with DHF – are very different from the others, containing only two modules shared with other members of this serotype (regions shown in red in Fig. 1c). Finally, for all serotypes, the general lack of association between the 3'-UTR and disease was not affected if only isolates taken from primary or secondary infections were analysed.
We performed a similar analysis on our global dataset of 93 strains, although no extra data were available for DENV-4. In DENV-1 and DENV-3, we again found no clear association between secondary structure and disease severity. We determined 14 groups of sequences (with at least two sequences in each group and totalling 41 isolates), in which each group is composed of isolates with the same secondary structure but with differing disease associations (Supplementary Table S2 available in JGV Online).
Because the largest number (46) of isolates is available for DENV-2, we considered their case in more detail (Supplementary Table S3 available in JGV Online). Overall, we recognized six majority (M) structures in the 3'-UTR of DENV-2, denoted MA, MB, MC, MD, ME and MF, and which differ from each other in small modules of no more than 30 nt (Fig. 2; see Supplementary Fig. S4 for structures MD, ME, MF available in JGV Online). These M structures comprise 72 % of all predictions and all other conformations are minority ones, shared by up to three isolates only.
These data provide two notable observations. First, different genotypes are often associated with different 3'-UTR secondary structures. Most notably, structures MA, MC and MF are only present in the Asian I and Asian II genotypes, while structure MB is found in Asian–American viruses and structure MD is only observed in cosmopolitan genotype viruses. Similarly, as noted previously, viruses of the American genotype have a 3'-UTR structure very different from those seen in Asian viruses (Leitmeyer et al., 1999), do not contain the M structures MA to MF, and in most cases (12/14) do not possess a complete 3'-LSH (Fig. 2b).
Second, there is seemingly a significant association between the M structures and DHF/DSS: of the 33 isolates with M structures, 24 are associated with DHF/DSS, whereas only four DHF/DSS isolates are predicted to have non-M structures (
2=6.89; P=0.009). However, many of the DHF/DSS strains were sampled from the same population during the same epidemic and hence do not represent independent variables. We therefore performed a phylogenetic analysis of the DENV-2 data using their associated E gene sequence as a marker of evolutionary history. This revealed that many of the DHF/DSS isolates form a monophyletic group that was contemporaneous in time and space (Supplementary Fig. S5 available in JGV Online), suggesting that these isolates are descendants of a single virus associated with DHF. Because of such pseudo-replication, we corrected our statistical analysis by retaining only one isolate from this clade. This resulted in a marginally non-significant association (2=3.4; P=0.065). The possible association between disease syndrome and the DENV-2 3'-UTR therefore needs to be investigated with a larger number of viral isolates sampled from independent outbreaks.
The lack of a consistent association between the secondary structure of the 3'-UTR and disease syndrome has a number of important implications. Most fundamentally, if there is a virological basis to the clinical outcome of DENV infection, then it is likely to reflect the action of multiple gene loci that may interact in a complex manner. As such, specific 3'-UTR structures alone are unlikely to be a strong predictor of clinical outcome. It is also evident that, because of the high level of intra-host genetic variation observed in DENV (Aaskov et al., 2006; Lin et al., 2004; Wang et al., 2002), the viral population within individual hosts will in reality comprise a spectrum of secondary structures, perhaps with differing interactions and disease associations. Indeed, the free energies between the meta-stable structures of MFOLD are generally very small and it is possible that they will be able to transform among each other, forming a dynamic equilibrium between alternative structural conformations. Finally, even though the association between 3'-UTR sequence/structure and viral virulence is weak, it is clear that some mutations in this region may have a major impact on viral fitness. In particular, it is striking that those viral genomes that lack a full-length 3'-LSH – most members of the American genotype of DENV-2 and representatives of genotype III of DENV-3 – have both suffered extinction, or near extinction, in their respective host populations (Rico-Hesse et al., 1997; Zhang et al., 2005).
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTMAIN TEXTREFERENCES |
|---|
Alvarez, D. E., De Lella Ezcurra, A. L., Fucito, S. & Gamarnik, A. V. (2005a). Role of RNA structures present at the 3'UTR of dengue virus on translation, RNA synthesis, and viral replication. Virology 339, 200–212.[CrossRef][Medline]
Alvarez, D. E., Lodeiro, M. F., Ludueña, S. J., Pietrasanta, L. I. & Gamarnik, A. V. (2005b). Long-range RNA-RNA interactions circularize the dengue virus genome. J Virol 79, 6631–6643.
Blackwell, J. L. & Brinton, M. A. (1995). BHK cell proteins that bind to the 39 stem-loop structure of the West Nile virus genome RNA. J Virol 69, 5650–5658.[Abstract]
Chambers, T. J., Hahn, C. S., Galler, R. & Rice, C. M. (1990). Flavivirus genome organization, expression, and replication. Annu Rev Microbiol 44, 649–688.[CrossRef][Medline]
Chiu, W. W., Kinney, R. M. & Dreher, T. W. (2005). Control of translation by the 5'- and 3'-terminal regions of the dengue virus genome. J Virol 79, 8303–8315.
Cologna, R. & Rico-Hesse, R. (2003). American genotype structures decrease dengue virus output from human monocytes and dendritic cells. J Virol 77, 3929–3938.
Durbin, A. P., Karron, R. A., Sun, W. & 10 other authors (2001). Attenuation and immunogenicity in humans of a live dengue virus type-4 vaccine candidate with a 30 nucleotide deletion in its 3'-untranslated region. Am J Trop Med Hyg 65, 405–413.[Abstract]
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.[CrossRef][Medline]
Hahn, C. S., Hahn, Y. S., Rice, C. M., Lee, E., Dalgarno, L., Strauss, E. G. & Strauss, J. H. (1987). Conserved elements in the 3' untranslated region of flavivirus RNAs and potential cyclization sequences. J Mol Biol 198, 33–41.[CrossRef][Medline]
Halstead, S. B., Porterfield, J. S. & O'Rourke, E. J. (1980). Enhancement of dengue virus infection in monocytes by flavivirus antisera. Am J Trop Med Hyg 29, 638–642.
Hofacker, I. L., Fontana, W., Stadler, P. F., Bonhoeffer, S., Tacker, M. & Schuster, P. (1994). Fast folding and comparison of RNA secondary structures. Monatsh Chem 125, 167–188.[CrossRef]
Klungthong, C., Zhang, C., Mammen, M. P., Jr, Ubol, S. & Holmes, E. C. (2004). The molecular epidemiology of dengue virus serotype 4 in Bangkok, Thailand. Virology 329, 168–179.[CrossRef][Medline]
Leitmeyer, K. C., Vaughn, D. W., Watts, D. M., Salas, R., Villalobos de Chacon, I., Ramos, C. & Rico-Hesse, R. (1999). Dengue virus structural differences that correlate with pathogenesis. J Virol 73, 4738–4747.
Lin, S. R., Hsieh, S. C., Yueh, Y. Y., Lin, T. H., Chao, D. Y., Chen, W. J., King, C. C. & Wang, W. K. (2004). Study of sequence variation of dengue type 3 virus in naturally infected mosquitoes and human hosts: implications for transmission and evolution. J Virol 78, 12717–12721.
Mangada, M. N. M. & Igarashi, A. (1997). Sequences of terminal non-coding regions from four dengue-2 viruses isolated from patients exhibiting different disease severities. Virus Genes 14, 5–12.[Medline]
Mangada, M. N. M. & Igarashi, A. (1998). Molecular and in vitro analysis of eight dengue type 2 viruses isolated from patients exhibiting different disease severities. Virology 244, 458–466.[CrossRef][Medline]
Mathews, D. H., Sabina, J., Zuker, M. & Turner, D. H. (1999). Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. J Mol Biol 288, 911–940.[CrossRef][Medline]
Men, R., Bray, M., Clark, D., Chanockm, R. M. & Lai, C. J. (1996). Dengue type 4 virus mutants containing deletions in the 3' noncoding region of the RNA genome: analysis of growth restriction in cell culture and altered viremia pattern and immunogenicity in rhesus monkeys. J Virol 70, 3930–3937.[Abstract]
Mongkolsapaya, J., Dejnirattisai, W., Xu, X.-N. & 11 other authors (2003). Original antigenic sin and apoptosis in the pathogenesis of dengue hemorrhagic fever. Nat Med 9, 921–927.[CrossRef][Medline]
Nisalak, A., Endy, T. P., Nimmannitya, S. & 7 other authors (2003). Serotype-specific dengue virus circulation and dengue disease in Bangkok, Thailand from 1973 to 1999. Am J Trop Med Hyg 68, 191–202.
Proutski, V., Gould, E. A. & Holmes, E. C. (1997a). Secondary structure of the 3'-untranslated region of flaviviruses: similarities and differences. Nucleic Acids Res 25, 1194–1202.
Proutski, V., Gaunt, M. W., Gould, E. A. & Holmes, E. C. (1997b). Secondary structure of the 3'-untranslated region of yellow fever virus: implications for virulence, attenuation and vaccine development. J Gen Virol 78, 1543–1549.[Abstract]
Rauscher, S., Flamm, C., Mandl, C. W., Heinz, F. X. & Stadler, P. F. (1997). Secondary structure of the 3'-noncoding region of flavivirus genomes: comparative analysis of base pairing probabilities. RNA 3, 779–791.[Abstract]
Rico-Hesse, R., Harrison, L. M., Salas, R. A. & 7 other authors (1997). Origins of dengue type 2 viruses associated with increased pathogenicity in the Americas. Virology 230, 244–251.[CrossRef][Medline]
Rodriguez-Roche, R., Alvarez, M., Gritsun, T., Halstead, S., Kouri, G., Gould, E. A. & Guzman, M. G. (2005). Virus evolution during a severe dengue epidemic in Cuba, 1997. Virology 334, 154–159.[CrossRef][Medline]
Shurtleff, A. C., Beasley, D. W. C., Chen, J. J. Y. & 9 other authors (2001). Genetic variation in the 3' non-coding region of dengue viruses. Virology 281, 75–87.[CrossRef][Medline]
Swofford, D. L. (2003). PAUP*. Phylogenetic analysis using parsimony (*and other methods). Version 4. Sunderland, MA: Sinauer Associates.
Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F. & Higgins, D. G. (1997). The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25, 4876–4882.
Wang, W.-K., Lin, S. R., Lee, C.-M., King, C. C. & Chang, S.-C. (2002). Dengue type 3 virus in plasma is a population of closely related genomes: quasispecies. J Virol 76, 4662–4665.
Yu, L. & Markoff, L. (2005). The topology of bulges in the long stem of the flavivirus 3' stem-loop is a major determinant of RNA replication competence. J Virol 79, 2309–2324.
Zeng, L., Falgout, B. & Markoff, L. (1998). Identification of specific nucleotide sequences within the conserved 3'-SL in the dengue type 2 virus genome required for replication. J Virol 72, 7510–7522.
Zhang, C., Mammen, M. P., Jr, Chinnawirotpisan, P., Klungthong, C., Rodpradit, P., Monkongdee, P., Nimmannitya, S., Kalayanarooj, S. & Holmes, E. C. (2005). Clade replacements in dengue virus serotypes 1 and 3 are associated with changing serotype prevalence. J Virol 79, 15123–15130.
Zhang, C., Mammen, M. P., Jr, Chinnawirotpisan, P. & 7 other authors (2006). Structure and age of genetic diversity of dengue type-2 virus (DENV-2) in Thailand. J Gen Virol 87, 873–883.
Zuker, M. (2003). MFOLD web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res 31, 3406–3415.
Received 1 March 2006;
accepted 24 April 2006.
This article has been cited by other articles:
|
|
 |
|
  S. Tajima, Y. Nukui, T. Takasaki, and I. Kurane Characterization of the variable region in the 3' non-translated region of dengue type 1 virus J. Gen. Virol., August 1, 2007; 88(8): 2214 - 2222. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
