The second envelope protein (E2) of the hepatitis G virus (HGV) was expressed in Chinese hamster ovary (CHO) cells and showed a molecular weight of approximately 60 to 70 kd, with 15 to 25 kd of the size contributed by N-linked glycosylation. An enzyme-linked immunosorbent assay (ELISA) using HGV-E2 was developed to test for antibodies to this protein (anti-E2) in human sera. High sensitivity was achieved by developing monoclonal antibodies (mAbs) to HGV-E2, which were used as capture antibodies in the ELISA. Our studies revealed that 16% of healthy Spanish blood donors were exposed to HGV, indicating that additional routes of viral transmission besides parenteral exposure might exist. An even higher prevalence of exposure to HGV (52%-73%) was found in several groups at risk of parenteral exposure to infectious agents, i.e., intravenous drug users, transfusion history, hemophiliacs, and hepatitis C virus (HCV)-positive patients. Most anti-E2-positive patients were HGV-RNA-negative and vice versa, indicating an inverse correlation of these two viral markers. A panel of 16 posttransfusion patients followed for up to 16 years revealed that patients who develop an anti-E2 response become HGV-RNA-negative, while patients who do not develop anti-E2 are persistently infected. Immunity to HGV seems to be long-lasting, because circulating antibody to E2 could still be detected 14 years after seroconversion. Sequence comparisons showed that E2 is highly conserved among isolates collected worldwide, indicating that immune escape variants are not common in HGV infections. This reflects on a molecular level why HGV infections usually are cleared spontaneously by the host. However, possible mechanisms of HGV persistence, as found in some patients, remain to be elucidated.