Peripheral blood lymphocytes from patients with the acquired immunodeficiency syndrome (AIDS) or with signs or symptoms that frequently precede AIDS (pre-AIDS) were grown in vitro with added T-cell growth factor and assayed for the expression and release of human T-lymphotropic retroviruses (HTLV). Retroviruses belonging to the HTLV family and collectively designated HTLV-III were isolated from a total of 48 subjects including 18 of 21 patients wih pre-AIDS, three of four clinically normal mothers of juveniles with AIDS, 26 of 72 adult and juvenile patients with AIDS, and from one of 22 normal male homosexual subjects. No HTLV-III was detected in or isolated from 115 normal heterosexual subjects. The number of HTLV-III isolates reported here underestimates the true prevalence of the virus since many specimens were received in unsatisfactory condition. Other data show that serum samples from a high proportion of AIDS patients contain antibodies to HTLV-III. That these new isolates are members of the HTLV family but differ from the previous isolates known as HTLV-I and HTLV-II is indicated by their morphological, biological, and immunological characteristics. These results and those reported elsewhere in this issue suggest that HTLV-III may be the primary cause of AIDS.

A novel human B-lymphotropic virus (HBLV) was isolated from the peripheral blood leukocytes of six individuals: two HTLV-III seropositive patients from the United States (one with AIDS-related lymphoma and one with dermatopathic lymphadenopathy), three HTLV-III seronegative patients from the United States (one with angioimmunoblastic lymphadenopathy, one with cutaneous T-cell lymphoma, and one with immunoblastic lymphoma), and one HTLV-III seronegative patient with acute lymphocytic leukemia from Jamaica. All six isolates were closely related by antigenic analysis, and sera from all six virus-positive patients reacted immunologically with each virus isolate. In contrast, only four sera from 220 randomly selected healthy donors and none from 12 AIDS patients without associated lymphoma were seropositive. The virus selectively infected freshly isolated human B cells and converted them into large, refractile mono- or binucleated cells with nuclear and cytoplasmic inclusion bodies. HBLV is morphologically similar to viruses of the herpesvirus family but is readily distinguishable from the known human and nonhuman primate herpesviruses by host range, in vitro biological effects, and antigenic features.

Kaposi's sarcoma (KS) is frequently associated with human immunodeficiency virus-1 (HIV-1) infection. Supernatants from HIV-1-infected T cells carrying the CD4 antigen promote the growth of cells derived from KS lesions of AIDS patients (AIDS-KS cells), and the HIV-1 tat gene, introduced into the germ line of mice, induces skin lesions closely resembling KS. Here we report that the tat gene product (Tat) is released from both HIV-1-acutely infected H9 cells and tat-transfected COS-1 cells. These Tat-containing supernatants specifically promote growth of AIDS-KS cells which are inhibited by anti-Tat antibodies; recombinant Tat has the same growth-promoting properties. Therefore a viral regulatory gene product can be released as a biologically active protein and directly act as a growth stimulator. These and previous data indicate that extracellular Tat could be involved in the development or progression, or both, of KS in HIV-1-infected individuals.

Human T-cell leukemia virus type III (HTLV-III) was recently identified as the probable etiologic agent of the acquired immune deficiency syndrome (AIDS). Here it is shown that, in human T-cell lines infected with HTLV-III, gene expression directed by the long terminal repeat sequence of this virus is stimulated by more than two orders of magnitude compared to matched uninfected cells. The rate of transcription of the HTLV-III long terminal repeat is more than 1000 times that of the SV40 early promoter in one infected cell line. Thus, HTLV-III, like HTLV-I, HTLV-II, and the bovine leukemia virus, is characterized by trans -activation of transcription in infected cells. The efficiency of trans -activation in the case of HTLV-III may account, at least in part, for the virulent nature of HTLV-III infection.