Journal of Virological Methods 113 (2003) 19–28 Evidence for in vitro falsely-primed cDNAs that prevent specific detection of virus negative strand RNAs in dengue-infected cells: improvement by tagged RT-PCR Christophe N. Peyrefitte a,b , Boris Pastorino a,b , Maël Bessaud a,b , Hugues J. Tolou a,b , Patricia Couissinier-Paris a,b,∗ a Unité de Virologie Tropicale, Institut de Médecine Tropicale du Service de Santé des Armées, BP 46, Parc du Pharo, 13998 Marseille Armées, France b Université de la Méditerranée, EA 3292, IFR 48, Marseille Cedex, France Received 26 February 2003; received in revised form 7 July 2003; accepted 8 July 2003 Abstract The identification of cell types replicating dengue viruses is an important step towards the understanding of the pathophysiology of dengue severe forms. Since the detection of negative strand viral RNAs is the more reliable marker of active replication for single-strand positive sense RNA viruses, we reassessed the specificity of RT-PCR assays already developed to detect dengue negative strand RNAs. Studying mammalian Vero cells infected by a dengue-2 strain, it was shown that falsely-primed cDNAs are generated in vitro during the reverse transcription step and are amplified subsequently by PCR. Since this may compromise the specificity of existing RT-PCR systems, we developed a tagged RT-PCR assay and addressed the role of some critical factors in such a system. Optimization of the negative strand-specific tagged RT-PCR allowed to resolve the problems due to the PCR amplification of falsely-primed cDNAs. Using this assay it was possible to detect specifically negative strand RNAs as soon as 3h after Vero cells have been exposed to the dengue-2 strain and we showed that this system is highly specific. Thus, the present dengue negative strand-specific tagged RT-PCR assay may help to reassess viral replication in the context of dengue pathophysiology. © 2003 Elsevier B.V. All rights reserved. Keywords: Flavivirus; Dengue; Replication; Negative strand RNAs; Tagged RT-PCR 1. Introduction Dengue viruses are members of the Flaviviridae family, that exist as four distinct serotypes all responsible for differ- ent clinical manifestations in many tropical and sub-tropical regions of the world (Guzman and Kouri, 2002). The geo- graphical spread (McBride and Bielefeldt-Ohmann, 2000) and the dramatic increase of severe forms of dengue infec- tion (WHO, 2000), i.e. dengue haemorrhagic fever (DHF) and dengue shock syndrome (DSS), have lead many re- searchers to investigate different aspects of the pathophys- iology of dengue diseases. However, a number of questions related to the pathophysiological events remain to be elu- ∗ Corresponding author. Tel.: +33-491-150154; fax: +33-491-150172. E-mail address: imtssa.vro@wanadoo.fr (P. Couissinier-Paris). cidated. In particular, the definitive identification of cell types replicating dengue viruses either in vivo or in vitro, is still a matter of discussions (Bhamarapravati, 1997; Diamond et al., 2000). Moreover, the possibility that differ- ent clinical manifestations of dengue infection may reflect at least partly the ability of dengue virus strains to replicate with different efficiencies in host cells, due to structural differences in regions critical for replication, still remains to be established (Mangada and Igarashi, 1998; Leitmeyer et al., 1999; Pandey and Igarashi, 2000; Ishak et al., 2001). Since dengue virus genome is single-strand positive sense RNA, the viral replication cycle requires the syn- thesis of negative strand RNAs, that are present as part of double-strand complexes formed by both positive and neg- ative RNA strands, known as replicative intermediate (RI) or replicative forms (RF) in infected cells (Cleaves et al., 0166-0934/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0166-0934(03)00218-0