Based on variation in nucleotide sequence within restricted regions in the putative C (core) gene of hepatitis C virus (HCV), four groups of HCV have been postulated in a panel of 44 HCV isolates. They were provisionally designated types I, II, III and IV. A method for typing HCV was developed, depending on the amplification of a C gene sequence by polymerase chain reaction using a universal primer (sense) and a mixture of four type-specific primers (antisense). HCV types were determined by the size of the products specific to each of them. Type II was found in HCV samples from 131 (82%) of 159 blood donors, more often than in those from 48 (60%) of 80 patients with non-A, non-B (NANB) liver disease in Japan (P< 0-01). In 11 haemophiliacs who had received imported coagulation factor concentrates, type I was found in five, as against type II in four. Double infection with two different HCV types was found in two patients with chronic NANB liver disease (types I and II; II and III) and two haemophiliacs (types I and II; I and III). HCV types were identical in mother and baby in each of two examples of perinatal transmission, and were also identical in donor and recipient in a case of accidental needle exposure.

The complete nucleotide sequence of a hepatitis C virus derived from plasma of a human carrier in Japan was determined. The cDNA of the isolate (HC-J6) contained 9481 nucleotides and an additional T stretch of 30 to 108 nucleotides at the 3' end, and had one large open reading frame coding for a polyprotein of 3033 amino acids. It differed by 31.8 to 32.1% in the nucleotide sequence and by 27-4 to 27-7 % in the amino acid sequence from an American isolate and two Japanese isolates previously reported. Among these four isolates, the 5' non-coding region of 329 to 341 nucleotides was well conserved (>93% identity), whereas the 3' non-coding region of 39 to 45 nucleotides (T stretches not included) was more variable (>30% identity). An excellent degree of conservation of the 5' non-coding region would reflect its pivotal role in replication, and primers deduced from this region could be applied for the sensitive and specific detection of viral RNA by polymerase chain reaction. Due to a high degree of similarity in the amino acid sequence of the putative core protein (>90%), antigen probes deduced from it would be suitable for the serological diagnosis of HCV infection. Low sequence similarity in the putative envelope protein (> 53 % identity), however, would have to be taken into account in considering the immunoprophylaxis of HCV infection.