A human type C retrovirus, designated HTLV, previously was isolated from or identified in some patients with leukemias and lymphomas of mature T lymphocytes. HTLV is genetically and serologically distinct from any known animal retroviruses. The absence of HTLV proviral sequences in DNA from normal humans showed that HTLV is not a ubiquitous endogenous (germ-line transmitted) virus ofhumans. Antibodies to HTLV core proteins have been identified in some people with T-cell neoplasias and are particularly prevalent in Japanese with adult T-cell leukemia, suggesting that HTLV is acquired horizontally. However, it was possible that HTLV is transmitted through the germ line of some (possibly rare) families and is then expressed in the HTLVpositive malignancies. An opportunity to study this question was provided by the development of several T-cell lines and a B-cell line from one HTLV-positive patient with a cutaneous T-cell lymphoma. Here we report that HTLV proteins or nucleic acids (or both) are found in three independently derived T-cell lines, all shown by HLA typing to have originated from the patient. In contrast, the B-cell line, the identity of which was also ascertained by HLA typing, contained no detectable HTLV protein, RNA, or proviral DNA. Because the sensitivity of the latter assay is more than sufficient to detect one proviral equivalent per haploid genome, the results indicate that HTLV was not transmitted to this patient through the germ line but rather was acquired by infection.The discovery (1,2) and purification (3) of T-cell growth factor (TCGF) has made it possible to grow relatively mature human T cells from normal (1, 2) and neoplastic (4) sources routinely in liquid suspension culture for long periods. We have identified a human type C retrovirus in cultured T cells from several patients with leukemia-lymphoma. of relatively mature T cells. In some instances, the virus (called HTLV) has been constantly produced by these T cells (5, 6) which have become independent of TCGF and established as permanent cell lines, apparently because of eventual constitutive release of TCGF (7). In two instances, the virus has been characterized in detail (5, 6). HTLV is not an endogenous virus ubiquitous in human germ lines (8) and is readily distinguishable from the known animal retroviruses, including those from subhuman primates (8-10). Antibodies to two HTLV proteins, p24 and p19, have also been found in sera from a few other patients with T-cell malignancies in the United States (11, 12), in a high percentage of patients with adult T-cell leukemias and other leukemias in Japan (13,14), and, more recently, in T-cell malignancies from many other parts of the world (unpublished data).Recently, workers in Japan have also shown that a retrovirus is associated with these diseases in southern Japan (15-17), and we have ascertained that this virus is HTLV (or a closely related strain) (unpublished data). Here we show that HTLV is present in some of the T cells in three independently derived T-cell lines fr...