Considerable information about the structural features of the HCV-encoded RdRp is now available from the crystal structure of the enzyme (Ago et al., 1999; Bressanelli et al., 1999; Lesburg et al., 1999) (Ch. 16). However, the lack of a highly productive tissue culture system that consistently supports virus replication (Ch. 11) has made it difficult to isolate replication complexes in order to study NS5B function in its appropriate context. For this reason, a number of laboratories have isolated recombinant HCV NS5B for functional studies as well as for drug screening (Behrens et al., 1996; Al et al., 1997, 1998; Lohmann et al., 1997; Yamashita et al., 1998; Hagehorn et al., 2000). Some of the initial attempts to isolate NS5B from recombinant insect cells or E. coli showed that a large percentage was insoluble. To solve this problem, NS5B was isolated under denaturing conditions and then refolded into active enzyme (Yuan et al., 1997), where it was shown to replicate HCV RNA as well as globin mRNA, poly(A), poly(C) and other templates in the absence of additional virus or host factors (Behrens et al., 1996; Al et al., 1997, 1998; Lohmann et al., 1997). Secondary structural analysis of HCV NS5B showed that it had a hydrophobic carboxy-terminus, which may explain its hydrophobicity and membrane association in HCV replication complexes. Removal of the hydrophobic 21 carboxy-terminal amino acid residues from NS5B facilitated purification and increased the yield of functional enzyme (Yamashita et al., 1998).