ASSESSMENT OF HEPATITIS-C VIRUS SEQUENCE COMPLEXITY BY ELECTROPHORETIC MOBILITIES OF BOTH SINGLE-STRANDED AND DOUBLE-STRANDED DNAS

To assess genetic variation in hepatitis C virus (HCV) sequences accur ately, we optimized a method for identifying distinct viral clones wit hout determining the nucleotide sequence of each clone. Twelve serum s amples were obtained from seven individuals soon after they acquired H CV during a prospective study, and a 452-bp fragment from the E2 regio n was amplified by reverse transcriptase PCR and cloned. Thirty-three cloned cDNAs representing each specimen were assessed by a method that combined heteroduplex analysis (HDA) and a single-stranded conformati onal polymorphism (SSCP) method to determine the number of clonotypes (electrophoretically indistinguishable cloned cDNAs) as a measure of g enetic complexity (this combined method is referred to herein as the H DA+SSCP method). We calculated Shannon entropy, incorporating the numb er and distribution of clonotypes into a single quantifier of complexi ty. These measures were evaluated for their correlation with nucleotid e sequence diversity. Blinded analysis revealed that the sensitivity ( ability to detect variants) and specificity (avoidance of false detect ion) of the HDA+SSCP method were very high. The genetic distance (mean +/- standard deviation) between indistinguishable cloned cDNAs (intra clonotype diversity) was 0.6% +/- 0.9%, and 98.7% of cDNAs differed by < 2%, while the mean distance between cloned cDNAs with different pat terns was 4.0% +/- 3.2%. The sensitivity of the HDA+SSCP method compar ed favorably with either HDA or the SSCP method alone, which resulted in intraclonotype diversities of 1.6% +/- 1.8% and 3.5% +/- 3.4%, resp ectively. The: number of clonotypes correlated strongly with genetic d iversity (R-2, 0.93), but this correlation fell off sharply when fewer clones were assessed. This HDA+SSCP method accurately reflected nucle otide sequence diversity among a large number of viral cDNA clones, wh ich should enhance analyses to determine the effects of viral diversit y on HCV-associated disease. If sequence diversity becomes recognized as an important parameter for staging or monitoring of HCV infection, this method should be practical enough for use in laboratories that pe rform nucleic acid testing.