A set of reference sequences for the hepatitis C genotypes 4d, 4f, and 4k covering the full open reading frame

Abstract

Infection with genotype 4 of the Hepatitis C virus is common in Africa and the Mediterranean area, but has also been found at increasing frequencies in injection drug users in Europe and North America. Full length viral sequences to characterize viral diversity and structure have recently become available mostly for subtype 4a, and studies in Egypt and Saudi Arabia, where high proportions of subtype 4a infected patients exist, have begun to establish optimized treatment regimens. However knowledge about other subtype variants of genotype 4 present in less developed African states is lacking. In this study the full coding region from so far poorly characterized variants of HCV genotype 4 was amplified and sequenced using a long range PCR technique. Sequences were analyzed with respect to phylogenetic relationship, possible recombination and prominent sequence characteristics compared to other known HCV strains. We present for the first time two full-length sequences from the HCV genotype 4k, in addition to five strains from HCV genotypes 4d and 4f. Reference sequences for accurate HCV genotyping are required for optimized treatment, and a better knowledge of the global viral sequence diversity is needed to guide vaccines or new drugs effective in the world wide epidemic.

Figure 1

Strategy for PCR amplification of the entire HCV open reading frame. A genotyping fragment in NS5B and two additional fragments 1 and 3 provide small sequence islands that enable generation of strain specific primers for the amplification of large hemi-genome fragments.

Figure 2

Neighbour joining tree based on partial NS5B sequences (nucleotide positions 7939–8269 relative to the HCV genotype 1a reference sequence H77) of HCV strains presented in this manuscript (shaded in grey), and sequences provisionally assigned as HCV subtype 4 references [Simmonds et al., 2005]. The numbers of bootstrap replicates supporting relevant nodes (total 1000 replicates) are indicated. Clusters of novel and reference nucleotide sequences suggest classification of the new strains as HCV subtypes 4d, 4f and 4k.

Figure 3

Neighbour joining tree based on published nucleotide sequences covering the full HCV open reading frame, with novel full-length sequences shaded in grey. The numbers of bootstrap replicates supporting relevant nodes (total 1000 replicates) are indicated. Full-length sequences provisionally assigned to subtypes 4d, 4f and 4k form distinct clusters within genotype 4 separate from other published subtypes, thus confirming their definition as separate subtypes according to consensus criteria [Simmonds et al., 2005]. HCV genotypes, subtypes and NCBI reference numbers are given for each sequence.

Figure 4

Simplot graph displaying pairwise nucleotide similarities between the test sequence 4k.EU392171 and full length sequences of all sequences included in Figure 3. Pairwise similarities are plotted based on calculations from a 200 nucleotide window sliding in 20 nucleotide steps across the genome. As illustrated, only the sequence of the second genotype 4k patient EU392173 shows consistently higher nucleotide similarities with 4k.EU392171 compared to all other sequences, with no evidence for recombination between this genotype 4k sequence and any other HCV genotype.

Figure 5

Amino acid alignment of clonal sequences from patient EU392169 documenting a mixed infection with two different viral quasispecies with and without a two amino acid insertion in position 575 and 576.

Figure 6

Amino acid alignment of reference sequences and novel strains in the NS5A region, including the Interferon Sensitivity Determining Region (ISDR, shaded in grey). Novel strains and noticeable parts of the sequences are marked in grey. Dashes indicate gaps or deletions. Amino acid positions are indicated relative to the HCV genotype 1a reference sequence H77.