Virus Research 135 (2008) 320���325 Contents lists available at ScienceDirect Virus Research j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / v i r u s r e s Short communication Genetic analysis of Dengue 3 virus subtype III 5 and 3 non-coding regions Ricardo L.A. Silva a,���, Aravinda M. de Silva b,1, Eva Harris c,2, Gene H. MacDonald a a Department of Infectious Diseases, St. Jude Children���s Research Hospital, 332 N. Lauderdale, Memphis, TN 38105, USA b Department of Microbiology and Immunology, CB #7290, University of North Carolina, Chapel Hill, NC 27599, USA c Division of Infectious Diseases, School of Public Health, 1 Barker Hall #424, University of California, Berkeley, CA 94720-7354, USA a r t i c l e i n f o Article history: Received 21 January 2008 Received in revised form 29 February 2008 Accepted 7 March 2008 Available online 23 April 2008 Keywords: Dengue 3 virus Non-coding region Nucleotide sequence Phylogenetic analysis a b s t r a c t The emergence of dengue hemorrhagic fever in Sri Lanka in 1989 correlated with the appearance of a genetic variant of Dengue 3 virus (DENV-3) subtype III (group B), closely related to the endemic group A variant. We studied the 5 and 3 non-coding regions (NCRs) of 15 DENV-3 subtype III isolates from Sri Lanka, Nicaragua and Martinique and found variability in the 3 NCRs. This included an 11-nucleotide insertion common to all isolates examined and two nucleotide differences that segregated viruses geo- graphically into an American and a Sri Lankan group. Comparisons also identified three nucleotide differences shared by all group A Sri Lankan DENV-3 III isolates linked to mild disease epidemics but not group B Sri Lankan and group B-associated American isolates linked to severe disease epidemics. Clustering of the Latin American/Caribbean isolates with Sri Lankan group B DENV-3 in phylogenetic analyses supports the proposed common East African origin for all these strains and confirms the use of the 3 NCR for molecular epidemiologic studies of DENV-3. The differences in the 3 NCRs reported here, as well as potential alterations in the structural and non-structural coding genes, may contribute to the increased pathogenicity of group B DENV-3. �� 2008 Elsevier B.V. All rights reserved. Dengue viruses (DENV) are the most medically important arthropod-borne viruses, and the diseases they cause, dengue fever (DF) and dengue hemorrhagic fever/dengue shock syndrome (DHF/DSS), are a major public health problem in tropical and sub- tropical regions (Monath and Heinz, 1996). Prior to 1989, there were few cases of DHF in the Indian subcontinent, despite the co-circulation of all four DENV serotypes however, after 1989, there was a sharp increase in the incidence of DHF/DSS in Sri Lanka (Messer et al., 2002). Phylogenetic analysis of Dengue 3 virus (DENV-3) subtype III isolates, using a 966-nucleotide (nt) fragment spanning part of capsid, preM/M and part of the envelope (E) genes, from before and after 1989 allowed the identification of two DENV- 3 III variants, named group A and group B, whose presence in Sri Lanka correlated with mild and severe disease epidemics, respec- tively (Messer et al., 2003). Numerous studies have suggested that the sequence and sec- ondary structures of the 5 and 3 non-coding regions (NCRs) of flaviviruses play a role in viral replication and that differences ��� Corresponding author. Present address: Instituto Nacional de Cancer, Rua Andre Cavalcanti, 37, Rio de Janeiro, RJ, 20.231-080, Brazil. Tel.: +55 21 3233 1338 fax: +55 21 3233 1453. E-mail addresses: rlsilva@inca.gov.br (R.L.A. Silva), desilva@med.unc.edu (A.M. de Silva), eharris@berkeley.edu (E. Harris). 1 Fax: +1 919 962 8103. 2 Fax: +1 510 642 6350. in these regions between strains may influence viral virulence (Leitmeyer et al., 1999 Cologna and Rico-Hesse, 2003 reviewed in Clyde et al., 2006). Due to a limited amount of nucleotide sequence information, DENV-3 NCRs have not been well characterized. In this study, we describe the analysis of 5 and 3 NCRs from eight novel and seven published strains of DENV-3 subtype III, isolated from humans and representing two different geographical regions (Table 1). These include eight isolates from the Americas and the Caribbean and seven isolates from Sri Lanka (Harris et al., 2000 Messer et al., 2003 Peyrefitte et al., 2003). The Sri Lankan isolates sequenced in this study represent two different DENV-3 III genetic subgroups from the period before (group A) and after (group B) the emergence of DHF in 1989. Most of these viruses were obtained from the Centers for Disease Control and Prevention (Fort Collins) from Dr. Duane Gubler (Table 1). Two additional isolates from Nicaragua were collected from patients during a DHF outbreak in Nicaragua (Harris et al., 2000). These were compared to seven previously published DENV-3 III isolates, as well as DENV serotypes 1, 2 and 4 and two related flaviviruses, West Nile virus (WNV) and Japanese encephalitis virus (JEV) (Table 1). All viruses sequenced in this study were passaged a maximum of three times in C6/36 mosquito cells. The sequences of the 5 and 3 NCRs of DENV-3 strains from different geographic regions were compared. Viral RNA was extracted from culture media using the QiaAmp RNeasy Mini Kit (Qiagen, Valencia, CA), decapped using pyrophos- 0168-1702/$ ��� see front matter �� 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.virusres.2008.03.007

R.L.A. Silva et al. / Virus Research 135 (2008) 320���325 321 Table 1 Source of virus used for genetic analysis Name Serotype/subtype/group Year Location Original strain ID and source of virus Passage history Sequence source GenBank accession # 83SriLan2 3/IIIA 1983 Sri Lanka D1306, D. Gubler, CDC, Fort Collins C6 36 p3 This study AY585845 85SriLan 3/III/A 1985 Sri Lanka 073, D. Gubler, CDC, Fort Collins C6 36 p3 This study AY585847 89SriLan2 3/III/A 1989 Sri Lanka D2863, D. Gubler, CDC, Fort Collins C6 36 p3 This study AY585846 89SriLan1 3/III/B 1989 Sri Lanka D2783, D. Gubler, CDC, Fort Collins C6 36 p3 This study AY585851 93 SriLan1 3/III/B 1993 Sri Lanka D5231, D. Gubler, CDC, Fort Collins C6 36 p3 This study AY585848 97SriLan1 3/III/B 1997 Sri Lanka L57, A. deSilva, UNC-Chapel Hill C6 36 p3 This study AY585852 98Nicara1 3/III/B L a 1998 Nicaragua 6845, E. Harris, UC-Berkeley C6 36 p3 This study AY585849 98Nicara3 3/III/B L 1998 Nicaragua 7071, E. Harris, UC-Berkeley C6 36 p3 This study AY585850 00Mart1 3/III/B L 2000 Martinique D3Mart1234, Peyrefitte et al., 2003 Peyrefitte et al., 2003 AY099337 00SriLan 3/III/B L 2000 Sri Lanka D3SriLa1266, Peyrefitte et al., 2003 Peyrefitte et al., 2003 AY099336 00Mart3 3/III/B L 2000 Martinique D3Mart1567, Peyrefitte et al., 2003 Peyrefitte et al., 2003 AY099343 00Mart2 3/III/B L 2000 Martinique D3Mart1706, Peyrefitte et al., 2003 Peyrefitte et al., 2003 AY099344 01Mart1 3/III/B L 2001 Martinique D3Mart2012, Peyrefitte et al., 2003 Peyrefitte et al., 2003 AY099345 01Mart3 3/III/B L 2001 Martinique D3Mart2023, Peyrefitte et al., 2003 Peyrefitte et al., 2003 AY099346 01Mart2 3/III/B L 2001 Martinique D3Mart2336, Peyrefitte et al., 2003 Peyrefitte et al., 2003 AY099347 D3-H87 3/I 1956 Philippines GenBank M93130 D1 1 1990 Singapore GenBank M87512 D2 2 1983 Jamaica GenBank M20558 D4 4 1981 Dominica GenBank AF326573 JEV b ��� ��� ��� GenBank M55506 WNV c ��� ��� ��� GenBank AF196835 a B L group B-like defined by genetic analysis in this study and Messer et al. (2003) . b JEV Japanese encephalitis virus. c WNV West Nile virus. phatase (Epicentre, Madison, WI) and ligated with RNA ligase (Epicentre) (Mandl et al., 1991). First strand synthesis was performed using Superscript III reverse transcriptase (Invitro- gen, Carlsbad, CA) and a reverse primer at position nt 1088 (5 -CTTCCTTAGGGTCGCCAGTTG-3 ). PCR amplification across the junction was carried out using a 3 -end sense primer at nt 10,185 (5 -TAGGCAATGAAGAGTTTCTGGACTACATGC-3 ) and a 5 -end anti- sense primer at nt 223 (5 -GCTATGAACGCCATAACCAATTTCATT-3 ) to generate a PCR product of 742 bp, spanning the ligated 5 ���3 termini. PCR products were purified using the QIAquick PCR Kit (Quiagen) and cloned using the TOPO TA Cloning Kit (Invitrogen). To resolve the problem that arises upon sequence determination from cloned cDNA, whereby rare nucleotides that are present in the pool of viral RNA or mutations that may be introduced during cDNA synthesis or PCR amplification may be detected, 20 clones for each strain were sequenced across the 5 ���3 NCR in the forward and reverse directions, and a consensus sequence resulting from the alignment of the sequences obtained for each strain was used for subsequent analysis. There was some degree of nucleotide vari- ability among the 20 clones sequenced for each isolate, which were scattered throughout the 3 NCR, including the highly conserved 3 terminus. However, no consistent changes were observed among clones derived from different isolates. The biological importance of these variations is not clear, since some of these mutations lie in the very 3 -end of the NCR, which is essential for DENV viability in culture, favoring the hypothesis that these changes are either arti- facts of the methodology or non-viable viral particles produced in cell culture. The 5 NCR (93 nt) of all DENV-3 III isolates was highly con- served, with only a single nucleotide difference observed between D3Mart2336 and the other isolates (nt 36 C ��� T). The 5 cyclization sequence required for efficient replication, at positions 132���139, was also conserved (Hahn et al., 1987 Khromykh et al., 2001). Comparison of the DENV-3 III consensus sequence with D3-H87, a subtype I isolate considered the DENV-3 prototype, revealed a single nucleotide difference (nt 39 A ��� G). This high degree of iden- tity in the 5 NCR most likely represents a strong selection pressure for the function of this region. Although significant sequence dif- ferences have been reported in the 5 NCR between flaviviruses of different serogroups, sequence conservation between members of the same serogroup is significantly higher (Brinton and Dispoto, 1988). A similar degree of sequence conservation in the 5 NCR has also been reported for the DENV-2 serotype (Mangada and Igarashi, 1997). Sequence analysis of the entire 3 NCR (443 nt) of the 15 DENV-3 III strains, as well as D3-H87, revealed a 116-nt hypervariable region just past the NS5 stop codon, followed by a highly conserved region (Fig. 1). Hypervariable regions just 3 of the stop codon, followed by highly conserved sequence, are a common feature of flaviviruses (Gritsun et al., 1997 Poidinger et al., 1996). All DENV-3 III isolates contained an 11-nucleotide insertion (nt 10,275) following the stop codon, as compared to D3-H87. This insertion, previously reported by Peyrefitte et al. (2003), appears to be a common feature of all DENV-3 subtype III isolates from the Latin America/Caribbean and Sri Lankan regions, supporting the Sri Lankan origin of the DENV-3 subtype III circulating in the Americas. The three Sri Lankan DENV-3 III group A isolates shared complete sequence identity throughout the 3 NCR. Three nucleotide differences were consistently observed between the Sri Lankan group A and group B isolates (including Latin American/Caribbean strains), which fell outside the highly conserved A1���A4 regions at nt 10,274 (G ��� A), nt 10,307 (T ��� C) and nt 10,436 (A ��� G) (Fig. 1). The group A variants that were endemic to Sri Lanka prior to 1989 were associated with very low incidences of DHF. The novel appearance of the closely related group B iso- late correlated with a severe outbreak of DHF (Messer et al., 2003).