Infection of the HTLV-II-bearing T-cell line C3 with HTLV-III/LAV is highly permissive and lytic

Acquired immunodeficiency syndrome-like illness associated with systemic Mycoplasma fermentans infection in a human immunodeficiency virus-negative homosexual man

A 35-year-old homosexual man developed a composite nodal Kaposi's sarcoma and peripheral T-cell lymphoma that were associated with a peripheral blood CD4-positive lymphocyte count of only 43/mm3. The patient subsequently developed Pneumocystis carinii pneumonitis and eventually died due to disseminated Cryptococcus neoformans. Numerous premortem tests for the presence of human immunodeficiency virus (HIV) types 1 and 2 were negative by the enzyme-linked immunosorbent assay, Western blot, viral isolation, and polymerase chain reaction techniques. Postmortem evaluations for HIV-1, HIV-2, human T-cell lymphotropic virus (HTLV)-I, and HTLV-II also were negative by polymerase chain reaction, immunofluorescence assays, and viral isolation. A systemic infection by Mycoplasma fermentans, however, was documented by immunohistochemistry and polymerase chain reaction in premortem and postmortem tissues. This recently recognized human pathogen has produced systemic infections in patients with the acquired immunodeficiency syndrome (AIDS) and in previously healthy non-AIDS patients who characteristically have a fulminant flu-like illness. Additionally, M fermentans has enhanced the cytopathic effect of HIV in in vitro studies and has produced fatal wasting illnesses with terminal lymphopenia in inoculated adult silvered leaf monkeys. This report is the first description of an association between M fermentans infection and an AIDS-like illness in an HIV-negative individual. The etiology of the severe immunosuppression in this patient and the associated role of M fermentans remain to be determined by further investigations.

CD4-independent binding of HIV-1 to the B lymphocyte receptor CR2 (CD21) in the presence of complement and antibody

Complement and antibody contribute to infection-enhancement and possible expanded cellular tropism of HIV-1 in vitro through a process requiring complement receptors. Until now, however, the ability of HIV-1 to bind complement receptors has not been documented or characterized. We investigated whether antibody and complement permitted HIV-1 to bind to the B lymphocyte receptor, CR2 (CD21), in an effort to learn more about infection-enhancement, and also because CR2 can mediate B cell proliferation and antigen localization in lymphoid organs in other systems. HIV-1 incubated with antibody and fresh human serum as a source of complement bound approximately 10-fold greater to cells expressing CR2 than to HIV-1-permissive cells lacking this receptor. A similar effect was observed using cells which expressed CR2 but no CD4. This binding was minimal in heat-inactivated and C3-deficient sera, and was significantly reduced by the anti-CR2 MoAb, OKB7, but not by the anti-CD4 MoAb, OKT4a. Thus, complement and antibody acted in concert to facilitate the binding of HIV-1 to CR2 independently of CD4. CD4-independent binding of HIV-1 to CR2 was not sufficient to produce infection in Raji-3 cells. Titres of antibodies mediating CR2 binding correlated with antibody titres as measured by immunofluorescence (P < 0.01) and infection-enhancement (P < 0.05) but were discordant with titres of neutralizing antibodies, a result consistent with the utilization of CR2 for enhanced infection of cells. The ability of complement and antibody to facilitate the binding of HIV-1 to CR2 in the absence of CD4 provides new insights into mechanisms of HIV-1-induced immunopathogenesis and infection-enhancement.

Selective antiviral activity of synthetic soluble L-tyrosine and L-dopa melanins against human immunodeficiency virus in vitro

Melanins are pigments found in hair, skin, irides of the eye, and brain. Their functions in mammals include protection from exposure to sunlight, camouflage from predators, sexual recognition within species, and possible electron transfer reactants. Most natural melanins exist in an insoluble form, which is one reason there is little information on the biological properties of soluble melanins. Here, synthetic soluble melanins were obtained by chemical oxidation of L-tyrosine or spontaneous oxidation of L-beta-3,4-dihydroxyphenylalanine (L-dopa). Replication of human immunodeficiency virus types 1 and 2 (HIV-1 and HIV-2) was inhibited by soluble melanin in two human lymphoblastoid cell lines (MT-2 and H9) and in phytohemagglutinin-stimulated human T cells. Effective concentrations of 0.15-10 micrograms/ml had no cell toxicity. Melanin blocked infection by cell-free virus and interfered with HIV-induced syncytium formation and cytopathic effects when fusion-susceptible, uninfected cells, were mixed with chronically infected cells. Melanin also impeded the HIV-1 envelope surface glycoprotein, and T cell specific monoclonal antibody leu-3a (CD4), but not leu-5b (CD2), from binding to the surface of MT-2 cells. No effect on HIV-1 reverse transcriptase activity in viral lysates was observed. These results identify a unique biological property of melanin, and suggest that soluble melanins may represent a new class of pharmacologically active substances which should be further investigated for potential therapeutic utility in the treatment of Acquired Immune Deficiency Syndrome (AIDS).

Co-infection with HIV-I and HTLV-I/II In Intravenous Drug Users in Suburban New York City, With Comparison to Other Geographic Areas

Co-infection with human T-lymphotropic virus types I or II (HTLV-I or -II) may be a cofactor in the progression of human immunodeficiency virus (HIV) infection. We assessed the frequency of simultaneous infection with these retroviruses among intravenous drug users from Westchester County, N.Y., a suburb of New York City. Comparison was made with similar studies in the United States and Europe.

Complement-mediated antibody-dependent enhancement of HIV-1 infection requires CD4 and complement receptors

In this study it is demonstrated that complement-mediated antibody-dependent enhancement (C'-ADE) of HIV-1 infection in vitro is blocked by murine monoclonal antibodies to CD4 and complement receptor type 2 (CR2) while HIV-1 infection in the absence of C'-ADE is blocked by anti-CD4 but not anti-CR2 monoclonal antibodies. The anti-CR2 murine monoclonal antibody, OKB7, blocked C'-ADE of HIV-1 infection at concentrations greater than 1 microgram/ml. The anti-CD4 monoclonal antibody, OKT4a, but not OKT4f blocked C'-ADE at concentrations greater than 0.06 microgram/ml. HIV-1 infections were quantitated by cytopathic effect, indirect immunofluorescence, and reverse transcriptase release. It appears from these in vitro studies that C'-ADE of HIV-1 infection requires both CD4 and complement receptors while HIV-1 infection in the absence of antibody and complement requires only CD4.

Spectrum of biological properties of human immunodeficiency virus (HIV-1) isolates

In vitro assessment of biological properties of 14 independent isolates of human immunodeficiency virus type 1 (HIV-1) was performed in order to gain insight into the spectrum of behavioral diversity of HIV-1s and to attempt to identify phenotypic traits that may be eventually correlated with in vivo pathogenesis. All of these biologically cloned isolates were found to spread very slowly in most cell cultures, requiring 8-10 weeks for virus to spread from a few infected cells to around 10(5) cells. If viral synergistic activity was also present, as in HTLV-1-infected cells, HIV-1 spread was greatly accelerated. The isolates varied in their cellular tropisms, having as much as 100,000-fold difference in their tropisms for various human CD4-positive cell lines. Several HIV isolates were dual-tropic for both T and promonocytic cells, but some of these isolates did not readily infect U937 promonocytes while readily infecting THP-1 promonocytes. Both the slow spread and extreme tropisms of HIV-1 isolates have practical implications for titering HIVs and for initiating any studies examining the interaction between a given isolate and any given cell. Some isolates did not score readily by reverse transcriptase assays while others did and this did not reflect the amount of infectious virus produced. These findings raise questions about the reliability of HIV quantitation by RT assay. The HIV isolates further varied in their ability to kill and/or fuse cells, whereas some induced cytopathology more efficiently in a given cell line than others, even though the latter appeared to replicate as well. Finally, most isolates killed cells without syncytia formation, demonstrating that cell-to-cell fusion is a minor mechanism of cytopathology. The properties observed for each HIV isolate appeared to be stable phenotypes for that virus and the diversity of biological behavior raises the possibility that independent HIV isolates may differ in their virulence properties in vivo as well.

Antibody-dependent enhancement of human immunodeficiency virus type 1 (HIV-1) infection in vitro by serum from HIV-1-infected and passively immunized chimpanzees

Based on recent reports of antibody-dependent enhancement of human immunodeficiency virus type 1 (HIV-1) infection in vitro by serum from HIV-1-infected humans, sera from HIV-1 antibody-positive chimpanzees (Pan troglodytes) was evaluated for enhancing activity in an in vitro infection assay that uses MT-2 cells (a human lymphoblastoid cell line). Although fresh chimpanzee serum was found to have pronounced infection-enhancing properties in the absence of antibody to HIV-1, this effect was abolished by heat inactivation (57 degrees C, 1 hr) or treatment with cobra venom anticomplementary protein. Heat-inactivated, HIV-1 antibody-positive chimpanzee serum could enhance HIV-1 infection of MT-2 cells in vitro when combined with fresh, normal human serum. By serial serum samples from three HIV-1-infected chimpanzees, HIV-1 antibody-positive chimpanzees are shown to develop enhancing antibodies early in infection (2 mo postchallenge), whereas neutralizing antibodies develop later. Over the course of HIV-1 infection, this enhancing activity decreases while neutralizing activity increases, suggesting a possible role for enhancing and neutralizing activities in HIV-1 pathogenesis. The enhancing activity of an IgG fraction used to passively immunize chimpanzees against HIV-1 infection is shown to be present at dilutions as high as 1:65,000, offering an interesting possible reason for the failure of passive immunization to protect chimpanzees from HIV infection. These results suggest that serum from HIV-1-immunized chimpanzees might be tested to determine whether current HIV-1 candidate vaccines induce production of antibodies that mediate antibody-dependent enhancement of HIV-1 infection in this in vitro assay.

Antibody-independent, complement-mediated enhancement of HIV-1 infection by mannosidase I and II inhibitors

Human immunodeficiency virus type 1 (HIV-1) infectivity and cytopathic effect require proper maturation of the viral envelope glycoprotein carbohydrate moieties. We have found that fresh human serum enhances the infectivity of HIV-1 in MT-2 cell infection assays when virus is synthesized in the presence of the mannosidase I inhibitor, 1-deoxymannojirimycin, or the mannosidase II inhibitor, swainsonine, but has no enhancing effect on virus synthesized in the presence of the glucosidase I inhibitors, castanospermine and 1-deoxynojirimycin, or the glucosidase II inhibitor, bromoconduritol. Enhanced infections were characterized by cytopathic effect, antigen synthesis and reverse transcriptase release, all which occurred sooner than in control-infected cultures. This enhancement of infection was also observed in C1q-deficient serum but was not observed in serum that was heat-inactivated or depleted of complement components C3 or factor B, thus suggesting a requirement for the alternate pathway of complement.

Augmentation of human immunodeficiency virus type 1 gene expression by tumor necrosis factor alpha

It is now well established that cytokines are involved in the regulation of gene expression from HIV-1 LTR. The present study provides evidence that TNF-alpha stimulates HIV-1 gene expression and that the enhancer sequence within the HIV-1 LTR is involved in the stimulation. These results support the idea that immunologic stimulation and infection may trigger the development of clinical AIDS in individuals latently infected with HIV-1.

Inhibition of HIV-1 proviral DNA synthesis and RNA accumulation by mismatched dsRNA

The antiviral activity of mismatched dsRNA of the form poly(I):poly(C12-U)n (Ampligen) against the human immunodeficiency virus type 1 (HIV-1) was investigated by RNA-RNA and RNA-DNA hybridizations. Mismatched dsRNA delayed the appearance of newly transcribed HIV-1 RNA as detected by liquid dot-blot hybridization in cultures of H9 T-lymphoblastoid cells following virus challenge. The appearance of proviral DNA as detected by Southern hybridization following virus challenge in H9 cells was also delayed. Mismatched dsRNA had no effect in syncytium inhibition assays performed by fusing MT-2 cells with H9/HTLV-IIIB cells. These results suggest that the in vitro anti-HIV-1 activity of mismatched dsRNA occurs, at least in part, at an early stage in the viral replication cycle following initial gp120-CD4 binding.

A hybridoma producing human monoclonal antibody specific for glycoprotein 120 kDa of human immunodeficiency virus (HIV-1

A stable hybridoma producing anti-HIV human monoclonal antibody (HMCA) was generated by fusing CD3-depleted human splenic lymphocytes from an HIV sero-positive donor with the mouse myeloma cell line P3x63AgU1. The resultant hybridoma has been secreting IgG1, lambda chain for over nine months at a rate of 2.5 micrograms/10(6)cells/day. The HMCA shows specific reactivity in ELISA using HIV-infected cell lysates. Immunofluorescence tests have indicated that this HMCA binds specifically to the surface of H9 and C3 HIV/HTLVIIIb infected cells, HIV/N1T infected CEM cells and to MoT cells infected with an HIV clinical isolate. Western blotting revealed recognition of glycoproteins 120 and 160 kDa of HIV by the HMCA. Although this HMCA demonstrated no neutralizing activity, the production of an anti-HIV HMCA specific for glycoprotein 120 kDa indicates the possibility that a neutralizing HMCA can be developed as further fusions with lymph nodes and spleens from HIV positive donors are performed.

Role of protein N-glycosylation in pathogenesis of human immunodeficiency virus type 1

Human immunodeficiency virus type 1 (HIV-1), the retrovirus responsible for acquired immunodeficiency syndrome (AIDS), contains two heavily glycosylated envelope proteins, gp120 and gp41, which mediate attachment of virions to glycosylated cell surface receptor molecules (CD4 antigens) and appear to be responsible for syncytium formation and associated cytopathic effects of this virus. A comprehensive study of the effects of N-linked glycoprotein processing inhibitors on HIV-1 replication, infectivity, cytopathicity, target-cell infectibility, syncytium formation, and gp120 electrophoretic mobility was conducted to assess the importance of protein glycosylation in the pathogenesis of HIV-1 in vitro. The electrophoretic mobility of gp120 was decreased when gp120 was synthesized in the presence of castanospermine or 1-deoxynojirimycin (inhibitors of glucosidase I), increased when gp120 was synthesized in the presence of 1-deoxymannojirimycin (mannosidase I) or swainsonine (mannosidase II), and unaffected when gp120 was synthesized in the presence of bromoconduritol (glucosidase II). Inhibition by tunicamycin (lipid-linked oligosaccharide precursor synthesis), castanospermine, 1-deoxynojirimycin, and 1-deoxymannojirimycin attenuated HIV-1 infectivity and blocked HIV-1-induced syncytium formation and cytopathicity, whereas bromoconduritol and swainsonine failed to have such effects. None of the inhibitors interfered with virus replication in acutely infected cells or affected the ability of target cells to form syncytia with untreated HIV-1-infected cells. These results demonstrate that protein N-glycosylation is critical to the pathogenesis of HIV-1 at the levels of viral infectivity and cytopathicity but not at the level of virus replication or of host-cell infectibility.

Concomitant infection of HTLV-I and HIV-1: prevalence of IgG and IgM antibodies in Washington, D.C. area

Serum samples collected from four groups of individuals in the Washington, D.C. area were examined for the presence of IgG and IgM classes of antibody reacting against HTLV-I and HIV-1. These four groups were: (1) healthy adults with negative premarital VDRL test for syphilis (n = 113), (2) miscellaneous common disease patients (n = 155), (3) drug abusers (n = 130), and (4) homosexual men (n = 187). The former two groups are considered to be low-risk groups, and the latter two, high-risk groups. The prevalence of IgG antibody on ELISA/Western blot tests for these groups were respectively: (1) 5.3%/1.8%, (2) 5.2%/1.9%, (3) 13.9%/4.6%, and (4) 4.3%/1.6% for HTLV-I, and (1) 2.7%/0.9%, (2) 4.5%/0%, (3) 12.3%/5.4%, and (4) 8.0%/5.9% for HIV-1. Instances of possible concomitant infection as shown by the presence of antibodies against both HTLV-I and HIV-1 were found only in the latter two high-risk groups, i.e. two (1.5%) in group (3), and three (1.6%) in group (4) as confirmed by both Western blot and immunofluorescence tests. Out of 97 sera collected from drug abusers in 1985-86 which had IgG antibody by Western blot test against HIV-1, 23 (23.7%) were HTLV-I antibody positive by ELISA test (Group 5), and 8 of these were confirmed by Western blot test. Among these 8 persons, IgM antibody against HTLV-I was found in 2, while that against HIV-1 was positive in 7 persons.(ABSTRACT TRUNCATED AT 250 WORDS)

Natural killer cell infection and inactivation in vitro by the human immunodeficiency virus

Cytolytic activity of human mononuclear peripheral blood leukocytes from healthy donors, cultured in interleukin-2 conditioned medium, was abrogated by in vitro infection with the lymphadenopathy associated virus (LAV) isolate of the human immunodeficiency virus (HIV). Although viral antigens are not expressed in cultured cells until 14 days postinfection, cytolytic activity was lost as early as 3 days after infection. Loss of cytolytic function was not a result of the release of suppressive factors from either infected cells or uninfected CEM cells since supernatants from neither infected cultures nor CEM cell cultures had any inhibitory effects on the function of uninfected cells. Cultured lymphocytes expressing Leu 11b were also shown to express HIV antigens via immunofluorescence after 14 days in culture. These results suggest that natural killer (NK) cells, as defined by expression of Leu 11b, were infected by HIV in vitro and the loss of lytic function was likely a direct consequence of that infection.

Evaluation of antiviral drugs and neutralizing antibodies to human immunodeficiency virus by a rapid and sensitive microtiter infection assay

A 96-well microtiter infection assay for the human immunodeficiency virus (HIV) is described. The assay utilizes human T-cell lymphotropic virus type I-immortalized MT-2 cells as targets for infection and requires only 4 to 5 days for completion. Cytolysis was quantitated by vital dye uptake of poly-L-lysine-adhered cells as an endpoint for infection. The assay's efficacy was proven by the sensitive and accurate assessment of several known anti-HIV agents including two inhibitors of reverse transcription (3'-azido-3'-deoxythymidine and 2',3'-dideoxycytidine), three biological response modifiers (recombinant interferons alpha and beta and mismatched double-stranded RNA), a direct inactivator of HIV virions (amphotericin B), and neutralizing antibodies from two HIV-positive human subjects. Evaluation of data was facilitated by computer-assisted analysis. This assay provides a means for rapid, sensitive, and inexpensive large-scale in vitro testing of potential anti-HIV therapeutic regimens and quantitation of HIV-neutralizing antibody titers.

Mismatched dsRNA (ampligen) induces protection against genomic variants of the human immunodeficiency virus type 1 (HIV-1) in a multiplicity of target cells

Mismatched double-stranded RNA of the form r(I)n.r(C12-U)n (Ampligen) has been shown to be active against human immunodeficiency virus type 1 (HIV-1) using CEM and C3 cells as targets for infection by the highly similar HIV-1 isolates HTLV-IIIB and LAV (Montefiori, D.C. and Mitchell, W.M., 1987, Proc. Natl. Acad. Sci. U.S.A., 84, 2985-2989). The scope of Ampligen's anti-HIV-1 activity was examined in this study using the genetically divergent HIV-1 isolate HTLV-IIIRF, two additional target T-cell lines, H9 and MT-2, and a monocyte/macrophage cell line, U937. As judged by indirect immunofluorescence, reverse transcriptase activity and vital dye uptake, Ampligen was active against HTLV-IIIRF in H9, MT-2, C3 and U937 cells in addition to being active against HTLV-IIIB in U937 cells. A minimum of 1 h preincubation of cells (MT-2) with Ampligen was required for maximum activity. These results suggest that Ampligen's potential clinical efficacy may not be limited by either the highly variable nature or host cell range of HIV-1.

Persistent coinfection of T lymphocytes with HTLV-II and HIV and the role of syncytium formation in HIV-induced cytopathic effect

We previously demonstrated a high permissiveness of HTLV-II-transformed T lymphocytes (C3) to human immunodeficiency virus (HIV) infection in vitro, and that this infection results in the lysis of cells (D.C. Montefiori and W.M. Mitchell (1986) Virology 155, 726-731). We now show that a small percentage of HIV-infected C3 cells resist cell lysis, grow continuously in culture, and express antigens of both viruses. High levels of reverse transcriptase activity found in the culture fluid of these coinfected cells were associated with the presence of fully infectious HIV and an absence of detectable infectious HTLV-II. Virus production in C3 cells coinfected with the HIV isolate HTLV-III was approximately threefold greater than in C3 cells coinfected with the HIV isolate LAV, a result which suggests that HIV genomic diversity may give rise to differences in replicative capacities. Lysis resistance was found to be a cellular-determined function in that HIV produced in cultures of C3/HTLV-III cells retained the capacity to elicit a lytic response upon repeated infection. Small syncytia (less than five nuclei) were rarely observed in cultures of C3 and nonlytic C3/HIV cells, whereas large syncytia (more than five nuclei) were in abundance during the lytic phase of coinfection, a result which supports a role for syncytium formation in the mechanism of HIV-induced cytopathic effects. The results of these studies further demonstrate that there exists a lack of viral interference by HTLV-II infection, and that HTLV-II-transformed lymphocytes could act as a chronic reservoir for HIV in vivo.

Antiviral activity of mismatched double-stranded RNA against human immunodeficiency virus in vitro

The biological response modifier r(I)n.r(C12-U)n, referred to here as mismatched double-stranded (ds) RNA, was examined for antihuman immunodeficiency virus (HIV) activity in vitro because of its known antiviral activity and ability to induce interferon (IFN) in other biological systems [Carter, W. A., Strayer, D. R., Hubbell, H. R. & Brodsky, I. (1985) J. Biol. Response Modif. 4, 495-502]. We found that cultures of the highly HIV-permissive T-cell line C3 were afforded significant protection from HIV infection when incubated in growth media supplemented with mismatched dsRNA at 10-50 micrograms/ml prior to virus challenge. Similar results were obtained at 50 micrograms of mismatched dsRNA per ml in cultures of the T-lymphoblastoid cell line CEM. Infections were monitored by indirect immunofluorescence of cells for viral p24 antigen expression, reverse transcriptase activity in culture fluids for virus production, and vital dye uptake for cytopathic effect. Antiviral activity was increased by the continued presence of mismatched dsRNA in cultures following virus challenge. A one-time exposure to mismatched dsRNA (50 micrograms/ml) provided greater antiviral activity than either a one-time exposure to recombinant IFN-alpha [250 international units (IU)/ml], IFN-beta (250 IU/ml), or IFN-gamma (50 IU/ml) in cultures of CEM cells, or a one-time exposure to a combination of all three IFNs (150 IU each per ml) in cultures of C3 cells. Mismatched dsRNA at 50 micrograms/ml had no effect on cell division, RNA and protein synthesis, or virus replication in all T-cell lines examined. A clear distinction between the activities of mismatched dsRNA and IFN was the ability of IFN to suppress the in vitro replication of HIV that occurred at IFN concentrations (150 IU each of alpha, beta, and gamma per ml) that provided less antiviral activity than mismatched dsRNA (50 micrograms/ml). The results of these in vitro studies suggest a potential therapeutic value for mismatched dsRNA in the treatment of acquired immunodeficiency syndrome (AIDS).

Mismatched double-stranded RNA (ampligen) reduces concentration of zidovudine (azidothymidine) required for in-vitro inhibition of human immunodeficiency virus

'Ampligen', a non-toxic, mismatched polymer of double-stranded RNA with antiviral and immunomodulatory activities reduced the concentration of zidovudine (azidothymidine, AZT; 'Retrovir', Wellcome) required for inhibitory activity against human immunodeficiency virus (HIV) in vitro. At the higher doses of AZT tested, the virustatic activity observed seemed to have a synergistic virustatic relation with ampligen. Thus, combined therapy with ampligen and AZT can be expected to be more beneficial than AZT alone to patients with the acquired immunodeficiency syndrome (AIDS) or AIDS-related complex since AZT regimens that seem to be clinically effective are associated with considerable toxicity.

Evidence that mannosyl residues are involved in human immunodeficiency virus type 1 (HIV-1) pathogenesis

The Human Immunodeficiency Virus (HIV), the causative agent of AIDS, is thought to bind to T4+ (CD4+) target cells through the heavily glycosylated gp120 envelope glycoprotein. Plant lectins bind glycoproteins through noncovalent interaction with specific hexose residues; therefore, lectins were evaluated for their ability to inactivate HIV in vitro. The mannose-specific lectins concanavalin-A and succinyl concanavalin-A completely inactivated HIV while lentil lectin, wheat germ agglutinin, and phytohemagglutinin-P substantially inactivated HIV. BS-II, Vicia villosa (hairy vetch), and Ptilota plumosa (red marine algae) failed to alter the infectibility of HIV. Neither simple stearic hindrance, viral aggregation, nor lectin-cell interactions served to explain this phenomenon. Glycoprotein glycosylation was evaluated by differential lectin binding as well as molecular weight changes in gp120 when virus was produced in the presence of swainsonine, a glycosylation inhibitor. Lentil lectin bound gp120 better than concanavalin-A, suggesting the majority of glycosylation sites are fucosylated. The apparent molecular weight of gp120 was reduced by swainsonine, although HIV infectivity and concanavalin-A inactivation were retained. Thus, at least some N-glycosylation sites are complex-type glycoproteins but regions external to the (GlcNAc)2(Man)3 "core" pentasaccharide region are not required for HIV infectivity. It appears that the site or sites involved are nonfucosylated, high mannose and/or biantennary, nonsialylated, N-glycosylated regions of gp120 or gp41. Alternatively, they may be in close approximation to such carbohydrate regions.