Complementation of murine cells for human immunodeficiency virus envelope/CD4-mediated fusion in human/murine heterokaryons

Monoclonal antibodies define a cellular antigen involved in HTLV-I infection

The exact mechanism by which the human T cell leukemia viruses (HTLV) infects target cells remains unclear; although some molecules have been identified to be important in viral infection and entry. To investigate these phenomena, we generated a panel of monoclonal antibodies (MAb) against a B cell line (BJAB-WH) which is highly permissive for infection with HTLV. These MAb have been used to further characterize the membrane molecules important for HTLV infection. Three of these MAb designated 4.2.3, 3.3.10, and 11.2.3 were capable of inhibiting syncytium formation induced in human B and T cell lines (i.e., BJAB-WH and SupT-1, respectively) by co-culture with HTLV-I infected MT-2 cells. All of these MAbs immunoprecipitated a 80-85 kDa antigen from the lysates of metabolically labeled BJAB-WH but not from BJAB-CC/84, a noninfectible target cell. The binding of these MAb with different HTLV target cells was analyzed and compared with binding of polyclonal monospecific antisera to the same cell lines. A 80-85 kDa membrane glycoprotein was isolated with an immunoaffinity chromatographic column constructed with MAbs 4.2.3 and 3.3.10. This cellular antigen was capable of inhibiting HTLV I/MT-2 induced fusion. This is the first direct demonstration that a 80-85 kDa cellular glycoprotein is directly involved in HTLV I/II infection and syncytium formation.

HIV-1-resistance phenotype conferred by combination of two separate inherited mutations of CCR5 gene

BACKGROUND

Despite multiple exposures to HIV-1, some individuals remain uninfected, and their peripheral-blood mononuclear cells (PBMC) are resistant to in-vitro infection by primary HIV-1 isolates. Such resistance has been associated with a homozygous 32-base-pair deletion (delta 32) in the C-C chemokine receptor gene CCR5. We examined other mutations of the CCR5 gene that could be associated with resistance to HIV-1 infection.

METHODS

We assessed the susceptibility of PBMC to in-vitro infection by HIV-1 isolates that use the CCR5 as the major coreceptor for viral entry in 18 men who had frequent unprotected sexual intercourse with a seropositive partner. We also did genotypic analysis of CCR5 alleles. One of the 18 exposed but uninfected men (who we refer to as ExU2) showed total resistance to in-vitro infection by CCR5-dependent viruses, and was found to carry a CCR5 delta 32 allele and a single point mutation (T-->A) at position 303 on the other allele. To find out whether the CCR5 mutation was restricted to ExU2's family or existed in the general population, we did genetic analyses of the CCR5 genotype in ExU2's father and sister and also in 209 healthy blood donors who were not exposed to HIV-1.

FINDINGS

The m303 mutation found in ExU2 introduced a premature stop codon and prevented the expression of a functional coreceptor. The family studies revealed that the m303 mutant allele was inherited as a single mendelian trait. Genotype analysis showed that three of the 209 healthy blood donors were heterozygous for the mutant allele.

INTERPRETATION

We characterise a new CCR5 gene mutation, present in the general population, that prevents expression of functional coreceptors from the abnormal allele and confers resistance to HIV-1 infection when associated to the delta 32 CCR5 mutant gene.

Antiviral activity of derivatized dextrans on HIV-1 infection of primary macrophages and blood lymphocytes

The present study demonstrates at the molecular level that dextran derivatives carboxymethyl dextran benzylamine (CMDB) and carboxymethyl dextran benzylamine sulfonate (CMDBS), characterized by a statistical distribution of anionic carboxylic groups, hydrophobic benzylamide units, and/or sulfonate moieties, interact with HIV-1 LAI gp120 and V3 consensus clades B domain. Only limited interaction was observed with carboxy-methyl dextran (CMD) or dextran (D) under the same conditions. CMDBS and CMDB (1 microM) strongly inhibited HIV-1 infection of primary macrophages and primary CD4+ lymphocytes by macrophage-tropic and T lymphocyte-tropic strains, respectively, while D or CMD had more limited effects on M-tropic infection of primary macrophages and exert no inhibitory effect on M- or T-tropic infection of primary lymphocytes. CMDBS and CMDB (1 microM) had limited but significant effect on oligomerized soluble recombinant gp120 binding to primary macrophages while they clearly inhibit (> 50%) such binding to primary lymphocytes. In conclusion, the inhibitory effect of CMDB and the CMDBS, is observed for HIV M- and T-tropic strain infections of primary lymphocytes and macrophages which indicates that these compounds interfere with steps of HIV replicative cycle which neither depend on the virus nor on the cell.

The ectodomain of the human T-cell leukemia virus type 1 TM glycoprotein is involved in postfusion events

To examine the contribution of the transmembrane envelope glycoprotein (TM) to the infectivity of the human T-cell leukemia virus type 1 (HTLV-1), single amino acid substitutions were introduced throughout its ectodomain. The mutated envelopes were tested for intracellular maturation and for functions, including ability to elicit syncytium formation and ability to mediate cell-to-cell transmission of the virus. Three major phenotypes, defining three functionally distinct regions, were identified. (i) Mutations causing defects in intracellular maturation of the envelope precursor are mostly distributed in the central portion of the TM ectodomain, containing the immunosuppressive peptide. This region, which includes vicinal cysteines thought to form an intramolecular disulfide bridge, is probably essential for correct folding of the protein. (ii) Mutations resulting in reduced syncytium-forming ability despite correct intracellular maturation are clustered in the amino-terminal part of the TM ectodomain, within the leucine zipper-like motif. Similar motifs with a propensity to form coiled-coil structures have been implicated in the fusion process driven by other viral envelope proteins, and HTLV-1 may thus conform to this general rule for viral fusion. (iii) Mutants with increased syncytium-forming ability define a region immediately amino-terminal to the membrane-spanning domain. Surprisingly, these mutants exhibited severe defects in infectivity, despite competence for fusion. Existence of this phenotype indicates that capacity for cell-to-cell fusion is not sufficient to ensure viral entry, even in cell-to-cell transmission. The ectodomain of the TM glycoprotein thus may be involved in postfusion events required for full infectivity of HTLV-1, which perhaps represents a unique feature of this poorly infectious retrovirus.

HIV-1 coreceptor activity of CCR5 and its inhibition by chemokines: independence from G protein signaling and importance of coreceptor downmodulation

HIV-1 infection requires the presence of specific chemokine receptors on CD4+ target cells to enable the fusion reactions involved in virus entry. CCR5 is a major fusion coreceptor for macrophage-tropic HIV-1 isolates. HIV-1 entry and fusion are mediated by the viral envelope glycoprotein (Env) and are inhibited by CCR5 ligands, but the mechanisms are unknown. Here, we test the role of G protein signaling and CCR5 surface downmodulation by two separate approaches: direct inactivation of CCR5 signaling by mutagenesis and inactivation of G(i)-type G proteins with pertussis toxin. A CCR5 mutant lacking the last 45 amino acids of the cytoplasmic C-terminus (CCR5306) was created that was expressed on transfected cells at levels comparable to cells expressing CCR5 and displayed normal chemokine binding affinity. CCR5 ligands induced calcium flux and receptor downmodulation in cells expressing CCR5, but not in cells expressing CCR5306. Nevertheless, CCR5 or CCR5306, when coexpressed with CD4, supported comparable HIV-1 Env-mediated cell fusion. Consistent with this, treatment of CCR5-expressing cells with pertussis toxin completely blocked ligand-induced transient calcium flux, but did not affect Env-mediated cell fusion or HIV-1 infection. Also, pertussis toxin did not block chemokine inhibition of Env-mediated cell fusion or HIV-1 infection. However, chemokines inhibited Env-mediated cell fusion less efficiently for CCR5306 than for CCR5. We conclude that the C-terminal domain of CCR5 is critical for G protein signaling and receptor downmodulation from the surface, but that neither function is required for CCR5 fusion coreceptor activity. The contrasting phenotypes of CCR5 and CCR5306 suggest that coreceptor downmodulation and direct blockage of Env interaction sites both contribute to chemokine inhibition of HIV-1 infection.

Enhancement of human immunodeficiency virus type 1 envelope-mediated fusion by a CD4-gp120 complex-specific monoclonal antibody

The entry of human immunodeficiency virus type 1 (HIV-1) into cells is initiated by binding of the viral glycoprotein gp120-gp41 to its cellular receptor CD4. The gp120-CD4 complex formed at the cell surface undergoes conformational changes that may allow its association with an additional membrane component(s) and the eventual formation of the fusion complex. These conformational rearrangements are accompanied by immunological changes manifested by altered reactivity with monoclonal antibodies specific for the individual components and presentation of new epitopes unique to the postbinding complex. In order to analyze the structure and function of the gp120-CD4 complex, monoclonal antibodies were generated from splenocytes of BALB/c mice immunized with soluble CD4-gp120 (IIIB) molecules (J. M. Gershoni, G. Denisova, D. Raviv, N. I. Smorodinsky, and D. Buyaner, FASEB J. 7:1185-1187 1993). One of those monoclonal antibodies, CG10, was found to be strictly complex specific. Here we demonstrate that this monoclonal antibody can significantly enhance the fusion of CD4+ cells with effector cells expressing multiple HIV-1 envelopes. Both T-cell-line-tropic and macrophage-tropic envelope-mediated cell fusion were enhanced, albeit at different optimal doses. Furthermore, infection of HeLa CD4+ (MAGI) cells by HIV-1 LAI, ELI1, and ELI2 strains was increased two- to fourfold in the presence of CG10 monoclonal antibodies, suggesting an effect on viral entry. These findings indicate the existence of a novel, conserved CD4-gp120 intermediate structure that plays an important role in HIV-1 cell fusion.

HIV coreceptor downregulation as antiviral principle: SDF-1alpha-dependent internalization of the chemokine receptor CXCR4 contributes to inhibition of HIV replication

Ligation of CCR5 by the CC chemokines RANTES, MIP-1alpha or MIP-1beta, and of CXCR4 by the CXC chemokine SDF-1alpha, profoundly inhibits the replication of HIV strains that use these coreceptors for entry into CD4(+) T lymphocytes. The mechanism of entry inhibition is not known. We found a rapid and extensive downregulation of CXCR4 by SDF-1alpha and of CCR5 by RANTES or the antagonist RANTES(9-68). Confocal laser scanning microscopy showed that CCR5 and CXCR4, after binding to their ligands, are internalized into vesicles that qualify as early endosomes as indicated by colocalization with transferrin receptors. Internalization was not affected by treatment with Bordetella pertussis toxin, showing that it is independent of signaling via Gi-proteins. Removal of SDF-1alpha led to rapid, but incomplete surface reexpression of CXCR4, a process that was not inhibited by cycloheximide, suggesting that the coreceptor is recycling from the internalization pool. Deletion of the COOH-terminal, cytoplasmic domain of CXCR4 did not affect HIV entry, but prevented SDF-1alpha-induced receptor downregulation and decreased the potency of SDF-1alpha as inhibitor of HIV replication. Our results indicate that the ability of the coreceptor to internalize is not required for HIV entry, but contributes to the HIV suppressive effect of CXC and CC chemokines.

Multiple extracellular domains of CCR-5 contribute to human immunodeficiency virus type 1 entry and fusion

Human immunodeficiency virus type 1 (HIV-1) entry is governed by the interaction of the viral envelope glycoprotein (Env) with its receptor. The HIV-1 receptor is composed of two molecules, the CD4 binding receptor and a coreceptor. The seven-membrane-spanning chemokine receptor CCR-5 is one of the coreceptors used by primary isolates of HIV-1. We demonstrate that the mouse homolog of CCR-5 (mCCR-5) does not function as an HIV-1 coreceptor. A set of chimeras of human CCR-5 and mCCR-5 was studied for Env-induced cell fusion and HIV-1 infection. Using the HIV-1ADA envelope glycoprotein in a syncytium formation assay, we show that replacement of any fragment containing extracellular domains of mCCR-5 by its human counterparts is sufficient to allow Env-induced fusion. Conversely, replacement of any fragment containing human extracellular domains by its murine counterpart did not lead to coreceptor function loss. These results show that several domains of CCR-5 participate in coreceptor function. In addition, using a panel of primary nonsyncytium-inducing and syncytium-inducing isolates that use CCR-5 or both CXCR-4 and CCR-5 as coreceptors, we show that the latter dual-tropic isolates are less tolerant to changes in CCR-5 than strains with a more restricted coreceptor use. Thus, different strains are likely to have different ways of interacting with the CCR-5 coreceptor.

Role of the first and third extracellular domains of CXCR-4 in human immunodeficiency virus coreceptor activity

The CXCR-4 chemokine receptor and CD4 behave as coreceptors for cell line-adapted human immunodeficiency virus types 1 and 2 (HIV-1 and HIV-2) and for dual-tropic HIV strains, which also use the CCR-5 coreceptor. The cell line-adapted HIV-1 strains LAI and NDK and the dual-tropic HIV-2 strain ROD were able to infect CD4+ cells expressing human CXCR-4, while only LAI was able to infect cells expressing the rat homolog of CXCR-4. This strain selectivity was addressed by using human-rat CXCR-4 chimeras. All chimeras tested mediated LAI infection, but only those containing the third extracellular domain (e3) of human CXCR-4 mediated NDK and ROD infection. The e3 domain might be required for the functional interaction of NDK and ROD, but not LAI, with CXCR-4. Alternatively, LAI might also interact with e3 but in a different way. Monoclonal antibody 12G5, raised against human CXCR-4, did not stain cells expressing rat CXCR-4. Chimeric human-rat CXCR-4 allowed us to map the 12G5 epitope in the e3 domain. The ability of 12G5 to neutralize infection by certain HIV-1 and HIV-2 strains is also consistent with the role of e3 in the coreceptor activity of CXCR-4. The deletion of most of the amino-terminal extracellular domain (e1) abolished the coreceptor activity of human CXCR-4 for ROD and NDK but not for LAI. These results indicate that HIV strains have different requirements for their interaction with CXCR-4. They also suggest differences in the interaction of dual-tropic HIV with CCR-5 and CXCR-4.

Role of the amino-terminal extracellular domain of CXCR-4 in human immunodeficiency virus type 1 entry

We have studied the role of the N-terminal extracellular domain of the human immunodeficiency virus type 1 (HIV-1) coreceptor, CXCR-4, in the entry and fusion of syncytium-inducing strains of HIV-1. Progressive deletions were introduced in the N-terminal extracellular domain of CXCR-4 and the effect on infection by different isolates was tested. Infection of cells expressing the different CXCR-4 deletion mutants by HIV-1 LAI and 89.6 was reduced only about twofold. In contrast, the HIV-1 GUN-1 and RF isolates were substantially more impaired in their ability to mediate cell-free infection and cell-cell fusion. Since LAI and RF are T-cell line-tropic viruses while 89.6 and GUN-1 are dual tropic, no clear correlation between tropism and requirements for CXCR-4 N-terminal sequences emerged. We also introduced point mutations at the two N-linked glycosylation sites. The isolates tested (LAI, RF, GUN-1, and 89.6) were not affected by the removal of predicted N-linked glycosylation sites in CXCR-4. We conclude that distinct virus strains interact differently with the CXCR-4 coreceptor and that the N-terminal extracellular domain is not the sole functional domain important for HIV-1 entry.

Potent inhibition of HIV-1 infectivity in macrophages and lymphocytes by a novel CCR5 antagonist

The chemokine receptors CXCR4 and CCR5 have recently been shown to act as coreceptors, in concert with CD4, for human immunodeficiency virus-type 1 (HIV-1) infection. RANTES and other chemokines that interact with CCR5 and block infection of peripheral blood mononuclear cell cultures inhibit infection of primary macrophages inefficiently at best. If used to treat HIV-1-infected individuals, these chemokines could fail to influence HIV replication in nonlymphocyte compartments while promoting unwanted inflammatory side effects. A derivative of RANTES that was created by chemical modification of the amino terminus, aminooxypentane (AOP)-RANTES, did not induce chemotaxis and was a subnanomolar antagonist of CCR5 function in monocytes. It potently inhibited infection of diverse cell types (including macrophages and lymphocytes) by nonsyncytium-inducing, macrophage-tropic HIV-1 strains. Thus, activation of cells by chemokines is not a prerequisite for the inhibition of viral uptake and replication. Chemokine receptor antagonists like AOP-RANTES that achieve full receptor occupancy at nanomolar concentrations are strong candidates for the therapy of HIV-1-infected individuals.

Human immunodeficiency virus strains differ in their ability to infect CD4+ cells expressing the rat homolog of CXCR-4 (fusin)

A clade B strain of human immunodeficiency virus type 1 (HIV-1(LAI)) could infect CD4+ cells expressing human CXCR-4 (fusin) or its rat homolog with similar efficacy. By contrast, cells expressing rat CXCR-4 were not permissive to HIV-1(NDK) (clade D), HIV-2(ROD), or HIV-1(LAI) with chimeric envelope protein gp120 bearing the V3 domain from HIV-1(NDK). The reciprocal chimeric gp120 (HIV-1(NDK) with V3 from HIV-1(LAI)) could mediate infection of cells expressing either human or rat CXCR-4. Genetically divergent HIV strains have different requirements for interaction with the CXCR-4 coreceptor, and the gp120 V3 domain seems to be involved in this interaction.

HIV type 1 V3 peptide constructs act differently on HIV type 1 infection of peripheral blood lymphocytes and macrophages

We have previously shown that a multibranched peptide construct derived from the tip of the B clade V3 loop consensus sequence (MPBC1: [GPGRAF]8-[K]4-[K]2-K-beta A-OH), but not V3 monomer peptides, inhibit human immunodeficiency virus type 1 (HIV-1) infection and syncytium formation of CD4+ T cells from immortalized lines. Here, we show that MBPC1 attaches to normal peripheral blood lymphocytes (PBLS) and monocytes but not to erythrocytes. While treatment with 5 microM MBPC1 had no significant antiviral effect on HIV-1Ba-L infection of monocyte-derived macrophages as assessed by p24 production in culture supernatants, this dose inhibited both HIV-1Ba-L and HIV-1LAI infection of PBLs. Virus production was inhibited up to 90% when MBPC1 was added to PBLs immediately after the virus, and was inhibited about 50% when it was added after 3 days; no effect was noted when it was added 7 days postinfection. MBPC1 did not affect PBL growth or IL-2 receptor and CD4 surface expression level. These observations suggest a selective antiviral effect of MBPC1 on CD4+ T lymphocytes and they provide additional circumstantial evidence that HIV-1 enters lymphocytes and monocytes by different mechanisms.

Macrophage tropism of human immunodeficiency virus type 1 and utilization of the CC-CKR5 coreceptor

The recent identification of the CC-CKR5 beta chemokine receptor as a major cofactor for entry of macrophage-tropic isolates of human immunodeficiency virus type 1 (HIV-1) raises the question of whether macrophage tropism is determined by utilization of this chemokine receptor. We observe that in addition to macrophage-tropic isolates of clades A, B, and E, macrophage-tropic isolates of clade F also utilize the CC-CKR5 molecule for entry. However, using single-round replication-competent reporter viruses carrying the envelope genes of T-cell line-tropic or macrophage-tropic phenotypic recombinant and mutant HIV-1 strains in infection of stable cell lines that coexpress the CD4 and chemokine receptors, we were unable to establish a strict correlation between macrophage tropism and utilization of the CC-CKR5 chemokine receptor. This latter finding suggests that a cofactor other than CC-CKR5 serves to determine entry into primary macrophages.

Chemokine receptors as fusion cofactors for human immunodeficiency virus type 1 (HIV-1)

CD4 is the primary cellular receptor for human immunodeficiency virus type 1 (HIV-1), but is not sufficient for entry of HIV-1 into cells. After a decade-long search, the cellular coreceptors that HIV-1 requires in conjunction with CD4 have been identified as members of the chemokine receptor family of seven-transmembrane G-protein coupled receptors. The discovery of distinct chemokine receptors that support entry of T-cell tropic (CXCR-4) and macrophage tropic HIV-1 strains (CCR-5) explains the differences in cell tropism between viral strains, the inability of HIV-1 to infect most nonprimate cells, and the resistance of a small percentage of the population to HIV-1 infection. Further understanding of the role of chemokine receptors in viral entry may also help explain the evolution of more pathogenic forms of the virus, viral transmission, and HIV-induced pathogenesis. These recent discoveries will aid the development of strategies for combating HIV-1 transmission and spread, the understanding of HIV-1 fusion mechanisms, and the possible development of small animal models for HIV-1 drug and vaccine testing.

Transcytosis of infectious human immunodeficiency virus across a tight human epithelial cell line barrier

Contact between various epithelial cell lines and HIV chronically infected mononuclear cell lines results in a massive and rapid budding of HIV virions toward the epithelium followed by their internalization into epithelial endosome-like structures. Here it is shown that as early as 30 minutes after apical contact, primary virus isolates generated from primary peripheral blood leukocytes from HIV-infected patients can cross an epithelial cell line barrier using transcytosis, the characteristic epithelial transcellular vesicular pathway. As the next step in the spread of infection, transcytosed HIV particles can productively infect mononuclear cells located at the basolateral side of the epithelial barrier. These observations suggest an alternative, rapid and efficient mechanism for transmission of HIV across an intact epithelial barrier.

SPC3, a V3 loop-derived synthetic peptide inhibitor of HIV-1 infection, binds to cell surface glycosphingolipids

Synthetic multibranched peptides derived from the V3 domain of human immunodeficiency virus type 1 (HIV-1) gp120 inhibit HIV-1 entry into CD4+ and CD4- cells by two distinct mechanisms: competitive inhibition of HIV-1 binding to CD4-/GalCer+ colon cells and postbinding inhibition of HIV-1 fusion with CD4+ lymphocytes. In the present study, we have characterized the cellular binding sites for the V3 peptide SPC3, which possesses eight V3 consensus motifs GPGRAF radially branched on a neutral polyLys core matrix. These binding sites are glycosphingolipids that share a common structural determinant, i.e., a terminal galactose residue with a free hydroxyl group in position 4: GalCer/sulfatide on CD4-/GalCer+ colon cells; LacCer and its sialosyl derivatives GM3 and GD3 on CD4+ human lymphocytes. These data suggest that the V3 peptide binds to the GalCer/sulfatide receptor for HIV-1 gp120 on HT-29 cells and thus acts as a competitive inhibitor of virus binding to these CD4- cells, in full agreement with previously published virological data. In contrast, SPC3 does not bind to the CD4 receptor, in agreement with the data showing that the peptide inhibits HIV-1 infection of CD4+ cells by acting at a postattachment step. The binding of SPC3 to LacCer, GM3, and GD3, expressed by CD4+ lymphocytes, suggests a role for these glycosphingolipids in the fusion process between the viral envelope and the plasma membrane of CD4+ cells. Since the multivalent peptide can theoretically bind to several of these glycosphingolipids, we hypothesize that the resulting cross-linking of membrane components may affect the fluidity of the plasma membrane and/or membrane curvature, altering the virus-cell fusion mechanism.

Regions in beta-chemokine receptors CCR5 and CCR2b that determine HIV-1 cofactor specificity

Macrophage-tropic (M-tropic) HIV-1 strains use the beta-chemokine receptor CCR5, but not CCR2b, as a cofactor for membrane fusion and infection, while the dual-tropic strain 89.6 uses both. CCR5/2b chimeras and mutants were used to map regions of CCR5 important for cofactor function and specificity. M-tropic strains required either the amino-terminal domain or the first extracellular loop of CCR5. A CCR2b chimera containing the first 20 N-terminal residues of CCR5 supported M-tropic envelope protein fusion. Amino-terminal truncations of CCR5/CCR2b chimeras indicated that residues 2-5 are important for M-tropic viruses, while 89.6 is dependent on residues 6-9. The identification of multiple functionally important regions in CCR5, coupled with differences in how CCR5 is used by M- and dual-tropic viruses, suggests that interactions between HIV-1 and entry cofactors are conformationally complex.

Cofactors provide the entry keys. HIV-1

The identification of the cofactors required for HIV-1 entry into cells promises to provide new insights into viral transmission and pathogenesis, and opens new avenues for AIDS therapy and prophylaxis.

Human immunodeficiency virus type 1 membrane fusion mediated by a laboratory-adapted strain and a primary isolate analyzed by resonance energy transfer

Previous studies of human immunodeficiency virus type 1 (HIV-1) envelope glycoprotein-mediated membrane fusion have focused on laboratory-adapted T-lymphotropic strains of the virus. The goal of this study was to characterize membrane fusion mediated by a primary HIV-1 isolate in comparison with a laboratory-adapted strain. To this end, a new fusion assay was developed on the basis of the principle of resonance energy transfer, using HeLa cells stably transfected with gp120/gp41 from the T-lymphotropic isolate HIV-1LA1 or the macrophage-tropic primary isolate HIV-1JR-FL. These cells fused with CD4+ target cell lines with a tropism mirroring that of infection by the two viruses. Of particular note, HeLa cells expressing HIV-1JR-FL gp120/gp41 fused only with PM1 cells, a clonal derivative of HUT 78, and not with other T-cell or macrophage cell lines. These results demonstrate that the envelope glycoproteins of these strains play a major role in mediating viral tropism. Despite significant differences exhibited by HIV-1JR-FL and HIV-1LAI in terms of tropism and sensitivity to neutralization by CD4-based proteins, the present study found that membrane fusion mediated by the envelope glycoproteins of these viruses had remarkably similar properties. In particular, the degree and kinetics of membrane fusion were similar, fusion occurred at neutral pH and was dependent on the presence of divalent cations. Inhibition of HIV-1JR-FL envelope glycoprotein-mediated membrane fusion by soluble CD4 and CD4-IgG2 occurred at concentrations similar to those required to neutralize this virus. Interestingly, higher concentrations of these agents were required to inhibit HIV-1LAI envelope glycoprotein-mediated membrane fusion, in contrast to the greater sensitivity of HIV-1LAI virions to neutralization by soluble CD4 and CD4-IgG2. This finding suggests that the mechanisms of fusion inhibition and neutralization of HIV-1 are distinct.

A seven-transmembrane domain receptor involved in fusion and entry of T-cell-tropic human immunodeficiency virus type 1 strains

Entry of human immunodeficiency virus type 1 (HIV-1) into cells requires binding to CD4 and fusion with a cellular membrane. Fusion does not occur in most nonhuman cells even when they express human CD4, indicating that one or more human accessory factors are required for virus infection. Recently, a seven-transmembrane domain protein has been shown to serve as an accessory factor for T-cell-tropic (T-tropic) HIV-1 isolates (Y. Feng, C. C. Broder, P. E. Kennedy, and E. A. Berger, Science 272:872-877, 1996). Here we show that expression of this glycoprotein, termed fusin, in murine, feline, simian, and quail cell lines, in conjunction with human CD4, rendered these cells fully permissive for HIV-1 envelope glycoprotein (Env)-mediated membrane fusion. Expression of CD4 or fusin alone did not permit fusion. In addition, introduction of fusin and CD4 into a human cell line, U87MG, that is resistant to HIV-1 induced syncytium formation and to infection by HIV-1 when expressing CD4 alone made this cell line permissive for Env-mediated cell-cell fusion. Fusion was observed only with T-tropic Env proteins. Macrophage-tropic (M-tropic) Env proteins from the SF162, ADA, and Ba-L HIV-1 strains did not fuse with cells expressing fusin and CD4, suggesting that M-tropic viruses utilize an accessory molecule other than fusin. Finally, coexpression of fusin and CD4 made both a murine and feline cell line susceptible to virus infection by T-tropic, but not M-tropic, HIV-1 strains.

Cell type-specific fusion cofactors determine human immunodeficiency virus type 1 tropism for T-cell lines versus primary macrophages

Work in this laboratory previously demonstrated that the tropism of different human immunodeficiency type 1 isolates for infection of human CD4+ continuous cell lines (e.g., T-cell lines and HeLa-CD4 transformants) versus primary macrophages is associated with parallel intrinsic fusogenic specificities of the corresponding envelope glycoproteins (Envs). For T-cell line-tropic isolates, it is well established that the target cell must also contain a human-specific fusion cofactor(s) whose identity is unknown. In this study, we tested the hypothesis that the Env fusion specificities underlying T-cell line versus macrophage tropism are determined by distinct cell type-specific fusion cofactors. We applied a recombinant vaccinia virus-based reporter gene assay for Env-CD4-mediated cell fusion; the LAV and Ba-L Envs served as prototypes for T-cell line-tropic and macrophage-tropic isolates, respectively. We examined CD4+ promyeloctic and monocytic cell lines that are infectible by T-cell line-tropic isolates and become susceptible to macrophage-tropic strains only after treatment with differentiating agents. We observed parallel changes in fusion specificity: untreated cells supported fusion by the LAV but not the Ba-L Env, whereas cells treated with differentiating agents acquired fusion competence for Ba-L. These results suggest that in untreated cells, the block to infection by macrophage-tropic isolates is at the level of membrane fusion; furthermore, the differential regulation of fusion permissiveness for the two classes of Envs is consistent with the existence of distinct fusion cofactors. To test this notion directly, we conducted experiments with transient cell hybrids formed between CD4-expressing nonhuman cells (murine NIH 3T3) and different human cell types. Hybrids formed with HeLa cells supported fusion by the LAV Env but not by the Ba-L Env, whereas hybrids formed with primary macrophages showed the opposite specificity; hybrids formed between HeLa cells and macrophages supported fusion by both Envs. These results suggest the existence of cell type-specific fusion cofactors selective for each type of Env, rather than fusion inhibitors for discordant Env-cell combinations. Finally, analyses based on recombinant protein expression and antibody blocking did not support the proposals by others that the CD44 or CD26 antigens are involved directly in the entry of macrophage-tropic isolates.

Human immunodeficiency virus type 1 entry into murine cell lines and lymphocytes from transgenic mice expressing a glycoprotein 120-binding mutant mouse CD4

Human CD4, the receptor for the gp120 envelope glycoprotein of HIV-1, is the route for viral entry into CD4+ cells; other cellular factors may cooperate with CD4 to facilitate HIV-1 entry into human cells. Human CD4 expressed on murine cells does not readily mediate HIV-1 entry, which may reflect a functional incompatibility of human CD4 with murine cellular components. We postulated that a HIV-1 gp120-binding mutant murine CD4 (L3T4) possessing a minimal number of human amino acid residues could facilitate HIV-1 entry into rodent cells, unlike human CD4. This hypothesis led us to develop a series of murine L3T4 mutants that bear human CD4 gp120-binding region amino acid residues while retaining most L3T4 epitopes. HeLa cell transfectants expressing gp120-binding mutant L3T4 proteins could be infected with HIV-1. Three mouse cell lines expressing these L3T4 mutant proteins could also be infected with HIV-1 as determined by PCR techniques that detect viral DNA and spliced RNAs. Lectin-stimulated polymorphonuclear leukocytes from transgenic mice (SBL mouse) expressing a gp120-binding L3T4 mutant protein were infected with HIV-1 at the same frequency as lectin-stimulated human peripheral blood lymphocytes as determined by in situ PCR analyses. Supernatant p24gag and reverse transcriptase levels in HIV-infected mouse cell cultures, however, were routinely at background levels, unlike HIV-infected human cell cultures. Thus, gp120-binding mutant L3T4 proteins mediate viral entry in all mouse cells that were tested, but high-level viral replication is absent in these cells.

Transgenic cell lines for detection of animal viruses

Rapid diagnostic assays based on direct detection of viral antigen or nucleic acid are being used with increasing frequency in clinical virology laboratories. Virus culture, however, remains the only way to detect infectious virus and to analyze clinically relevant viral phenotypes, such as drug resistance. Growth of viruses in cell culture is labor intensive and time-consuming and requires the use of many different cell lines. Transgenic technology, together with increasing knowledge of the molecular pathways of virus replication, offers the possibility of using genetically modified cell lines to improve virus growth in cell culture and to facilitate detection of virus-infected cells. Genetically modifying cells so that they express a reporter gene only after infection with a specific virus can allow the detection of infectious virus by rapid and simple enzyme assays such as beta-galactosidase assays without the need for antibodies. Although transgenic cells have recently been successfully used for herpes simplex virus detection, much more work needs to be done to adapt this technology to other human viral pathogens such as cytomegalovirus and respiratory viruses. This review offers some strategies for applying this technology to a wide spectrum of animal viruses.

CC CKR5: a RANTES, MIP-1alpha, MIP-1beta receptor as a fusion cofactor for macrophage-tropic HIV-1

Human immunodeficiency virus-type 1 (HIV-1) entry requires fusion cofactors on the CD4+ target cell. Fusin, a heterotrimeric GTP-binding protein (G protein)-coupled receptor, serves as a cofactor for T cell line-tropic isolates. The chemokines RANTES, MIP-1alpha, and MIP-1beta, which suppress infection by macrophage-tropic isolates, selectively inhibited cell fusion mediated by the corresponding envelope glycoproteins (Envs). Recombinant CC CKR5, a G protein-coupled receptor for these chemokines, rendered CD4-expressing nonhuman cells fusion-competent preferentially with macrophage-tropic Envs. CC CKR5 messenger RNA was detected selectively in cell types susceptible to macrophage-tropic isolates. CC CKR5 is thus a fusion cofactor for macrophage-tropic HIV-1 strains.

The beta-chemokine receptors CCR3 and CCR5 facilitate infection by primary HIV-1 isolates

We examined the ability of chemokine receptors and related G protein-coupled receptors to facilitate infection by primary, clinical HIV-1 isolates. CCR5, when expressed along with CD4, the HIV-1 receptor, allowed cell lines resistant to most primary HIV-1 isolates to be infected. CCR3 facilitated infection by a more restricted subset of primary viruses, and binding of the CCR3 ligand, eotaxin, inhibited infection by these isolates. Utilization of CCR3 and CCR5 on the target cell depended upon the sequence of the third variable (V3) region of the HIV-1 gp120 exterior envelope glycoprotein. The ability of various members of the chemokine receptor family to support the early stages of HIV-1 infection helps to explain viral tropism and beta-chemokine inhibition of primary HIV-1 isolates.

A dual-tropic primary HIV-1 isolate that uses fusin and the beta-chemokine receptors CKR-5, CKR-3, and CKR-2b as fusion cofactors

Here, we show that the beta-chemokine receptor CKR-5 serves as a cofactor for M-tropic HIV viruses. Expression of CKR-5 with CD4 enables nonpermissive cells to form syncytia with cells expressing M-tropic, but not T-tropic, HIV-1 env proteins. Expression of CKR-5 and CD4 enables entry of a M-tropic, but not a T-tropic, virus strain. A dual-tropic primary HIV-1 isolate (89.6) utilizes both Fusin and CKR-5 as entry cofactors. Cells expressing the 89.6 env protein form syncytia with QT6 cells expressing CD4 and either Fusin or CKR-5. The beta-chemokine receptors CKR-3 and CKR-2b support HIV-1 89.6 env-mediated syncytia formation but do not support fusion by any of the T-tropic or M-tropic strains tested. Our results suggest that the T-tropic viruses characteristic of disease progression may evolve from purely M-tropic viruses prevalent early in virus infection through changes in the env protein that enable the virus to use multiple entry cofactors.

HIV-1 entry into CD4+ cells is mediated by the chemokine receptor CC-CKR-5

The beta-chemokines MIP-1alpha, MIP-1beta and RANTES inhibit infection of CD4+ T cells by primary, non-syncytium-inducing (NSI) HIV-1 strains at the virus entry stage, and also block env-mediated cell-cell membrane fusion. CD4+ T cells from some HIV-1-exposed uninfected individuals cannot fuse with NSI HIV-1 strains and secrete high levels of beta-chemokines. Expression of the beta-chemokine receptor CC-CKR-5 in CD4+, non-permissive human and non-human cells renders them susceptible to infection by NSI strains, and allows env-mediated membrane fusion. CC-CKR-5 is a second receptor for NSI primary viruses.

Expression of HIV env gene in a human T cell line for a rapid and quantifiable cell fusion assay

Human immunodeficiency virus type 1 (HIV-1) envelope glycoproteins present at the surface of infected cells are known to mediate fusion with CD4-positive target cells. In this study we have developed a novel Env-expressing cell line for investigating the fusion process in a biologically significant system. Cell surface expression of the HIV-1 env gene, isolated from the highly fusogenic strain SF33, was obtained in the CD4-negative T cell line A2.01. To render the system versatile and efficient, HIV-1 regulatory proteins Tat and Rev were supplied in trans. The presence of Env at the cell surface was shown by cytofluorometry and immunofluorescence and precursor processing of gp160 to gp120/gp41 was demonstrated by Western blot. The fusion capacity of A2.01-Env cells was assessed by coculture with CD4-positive T lymphocytes or the fusion indicator cell line, HeLa-CD4-LTR-beta-Gal. By coincubation with CD4-positive T cells such as SupT1, A2.01-Env cells were observed to mediate rapidly numerous well-defined syncytia in a reproducible fashion. By expressing Tat, they also had the capacity to trans-activate the LTR-linked reporter beta-Gal gene following fusion with HeLa-CD4-LTR-beta-Gal cells. The fusion-inhibiting anti-CD4 monoclonal antibodies Q425 and Q428 were used to block specifically Env-mediated fusion with CD4-positive cells and to demonstrate application of this system to the search for potential fusion-blocking agents. Our system thus offers a biologically significant model for studying fusion events with the advantages of being rapid, reproducible and versatile.

Exceptional fusogenicity of Chinese hamster ovary cells with murine retroviruses suggests roles for cellular factor(s) and receptor clusters in the membrane fusion process

Chinese hamster ovary (CHO) cells are naturally resistant to infection by amphotropic and ecotropic murine retroviruses, but they become susceptible after expressing corresponding receptors rRAM-1 and mCAT-1, respectively, and they then form abundant syncytia when exposed to these viruses. The fusogenic activities of CHO cell clones increase much more strongly with levels of receptor expression than do their susceptibilities to infection, suggesting that the assembly of receptor clusters may limit syncytium formation. However, other cell lines are not fusogenic, even if they express larger amounts of receptors. Our results suggest that a factor that is relatively abundant or active in CHO cells may functionally interact with rRAM-1 and mCAT-1 in a pathway that enables receptor-bearing membranes to fuse with membranes that contain viral envelope glycoproteins. In the case of CHO/rRAM-1 cells, syncytia form at foci of amphotropic 4070A virus infection by fusion-from-within of infected with uninfected cells. This fusogenic propensity is a sole property of the uninfected CHO/rRAM-1 cells, which fuse in cocultures with any cells infected with 4070A virus. With CHO/mCAT-1 cells, fusogenicity is even greater and involves fusion-from-without by ecotropic virion particles. In contrast to infection, which behaves as expected for a process limited by ecotropic virus attachment to single receptors, fusion-from-without increases dramatically for cells that express the highest levels of mCAT-1. We propose that infection and syncytium formation are limited at distinct steps of a common pathway that requires virus binding to a single receptor, assembly of multivalent virus-receptor complexes, structural changes in viral envelope glycoproteins, and membrane fusion. The limiting step in syncytium formation is a cellular process that depends on receptor clustering and is relatively active in CHO cells.

HIV-1 entry cofactor: functional cDNA cloning of a seven-transmembrane, G protein-coupled receptor

A cofactor for HIV-1 (human immunodeficiency virus-type 1) fusion and entry was identified with the use of a novel functional complementary DNA (cDNA) cloning strategy. This protein, designated "fusin," is a putative G protein-coupled receptor with seven transmembrane segments. Recombinant fusin enabled CD4-expressing nonhuman cell types to support HIV-1 Env-mediated cell fusion and HIV-1 infection. Antibodies to fusin blocked cell fusion and infection with normal CD4-positive human target cells. Fusin messenger RNA levels correlated with HIV-1 permissiveness in diverse human cell types. Fusin acted preferentially for T cell line-tropic isolates, in comparison to its activity with macrophagetropic HIV-1 isolates.

Anti-idiotypic antibody to the V3 domain of gp120 binds to vimentin: a possible role of intermediate filaments in the early steps of HIV-1 infection cycle

Although the CD4 molecule is the major cellular receptor for human immunodeficiency virus (HIV), several lines of evidence suggest participation of additional molecules that are engaged after the binding of HIV to the CD4 receptor and that may facilitate viral entry into the target cell. Some of the post-CD4 binding, perfusion events involve the third hypervariable region (V3 loop) of the viral envelope protein gp120. To identify cellular proteins that interact with the V3 loop, we chose as a probe an antiidiotypic monoclonal antibody (MAb), anti-id2, which was prepared against the neutralizing MAb 110.4 that binds the V3 domain in the envelope glycoprotein gp120 of the LAI isolate of HIV-1. Anti-id2 reacted specifically with a 55- to 60-kDa protein in human T cell and monocytoid cell lines, and in a mouse melanoma cell line. This protein was identified immunologically and by protein sequence analysis as vimentin, an intermediate filament protein of lymphoid and other cells of mesodermal origin. Antiserum raised against vimentin inhibited nuclear translocation of HIV-1 DNA following infection of monocytes and CD4+ T cells with live virus, and reduced the amount of HIV-1 gag-specific RNA in the nuclei of monocytes following inoculation with HIV-1 pseudovirions. These data suggest that vimentin may participate in the early steps of HIV-1 replication, perhaps during the uptake of HIV-1 preintegration complexes into the nuclear compartment.

A truncated herpes simplex virus thymidine kinase phosphorylates thymidine and nucleoside analogs and does not cause sterility in transgenic mice

Dividing eukaryotic cells expressing the herpes simplex virus type 1 thymidine kinase (TK) gene are sensitive to the cytotoxic effect of nucleoside analogs such as acyclovir or ganciclovir (GCV). Transgenic mice with cell-targeted expression of this conditional toxin have been used to create animals with temporally controlled cell-specific ablation. In these animal models, which allow the study of the physiological importance of a cell type, males are sterile. In this study, we showed that this phenomenon is due to testis-specific high-level expression of short TK transcripts initiated mainly upstream of the second internal ATG of the TK gene. This expression is DNA methylation independent. To obtain a suicide gene that does not cause male infertility, we generated and analyzed the properties of a truncated TK (delta TK) lacking the sequences upstream of the second ATG. We showed that when expressed at sufficient levels, the functional properties of delta TK are similar to those of TK in terms of thymidine or GCV phosphorylation. This translated into a similar GCV-dependent toxicity for delta TK- or TK-expressing cells, both in vitro and in transgenic mice. However, delta TK behaved differently from TK in two ways. First, it did not cause sterility in delta TK transgenic males. Second, low-level delta TK RNA expression did not confer sensitivity to GCV. The uses of delta TK in cell-specific ablation in transgenic mice and in gene therapy are discussed.

Phorbol ester-induced down modulation of tailless CD4 receptors requires prior binding of gp120 and suggests a role for accessory molecules

The entry of human immunodeficiency virus type 1 into cells proceeds via a fusion mechanism that is initiated by binding of the viral glycoprotein gp120-gp41 to its cellular receptor CD4. Species- and tissue-specific restrictions to viral entry suggested the participation of additional membrane components in the postbinding fusion events. In a previous study (H. Golding, J. Manischewitz, L. Vujcic, R. Blumenthal, and D. Dimitrov, J. Virol. 68:1962-1968, 1994), it was found that phorbol myristate acetate (PMA) inhibits human immunodeficiency virus type 1 envelope-mediated cell fusion by inducing down modulation of an accessory component(s) in the CD4-expressing cells. The fusion inhibition was seen in a variety of cells, including T-cell transfectants expressing engineered CD4 receptors (CD4.401 and CD4.CD8) which are not susceptible to down modulation by PMA treatment. In the current study, it was found that preincubation of A2.01.CD4.401 cells with soluble monomeric gp120 for 1 h at 37 degrees C primed them for PMA-induced down modulation (up to 70%) of the tailless CD4 receptors. The gp120-priming effect was temperature dependent, and the down modulation may have occurred via clathrin-coated pits. Importantly, nonhuman cell lines expressing tailless CD4 molecules did not down modulate their CD4 receptors under the same conditions. The gp120-dependent PMA-induced down modulation of tailless CD4 receptors could be efficiently blocked by the human monoclonal antibodies 48D and 17B, which bind with increased avidity to gp120 that was previously bound to CD4 (M. Thali, J. P. Moore, C. Furman, M. Charles, D. D. Ho, J. Robinson, and J. Sodroski, J. Virol. 67:3978-3988, 1993). These findings suggest that gp120 binding to cellular CD4 receptors induces conformational changes leading to association of the gp120-CD4 complexes with accessory transmembrane molecules that are susceptible to PMA-induced down modulation and can target the virions to clathrin-coated pits.

Fusogenic selectivity of the envelope glycoprotein is a major determinant of human immunodeficiency virus type 1 tropism for CD4+ T-cell lines vs. primary macrophages

We investigated the relationship between the fusion selectivity of the envelope glycoprotein (env) and the tropism of different human immunodeficiency virus type 1 (HIV-1) isolates for CD4+ human T-cell lines vs. primary macrophages. Recombinant vaccinia viruses were prepared encoding the envs from several well-characterized HIV-1 isolates with distinct cytotropisms. Cells expressing the recombinant envs were mixed with various CD4+ partner cell types; cell fusion was monitored by a quantitative reporter gene assay and by syncytia formation. With CD4+ continuous cell lines as partners (T-cell lines, HeLa cells expressing recombinant CD4), efficient fusion occurred with the envs from T-cell line-tropic isolates (IIIB, LAV, SF2, and RF) but not with the envs from macrophage-tropic isolates (JR-FL, SF162, ADA, and Ba-L). The opposite selectivity pattern was observed with primary macrophages as cell partners; stronger fusion occurred with the envs from the macrophage-tropic than from the T-cell line-tropic isolates. All the envs showed fusion activity with peripheral blood mononuclear cells as partners, consistent with the ability of this cell population to support replication of all the corresponding HIV-1 isolates. These fusion selectivities were maintained irrespective of the cell type used to express env, thereby excluding a role for differential host cell modification. We conclude that the intrinsic fusion selectivity of env plays a major role in the tropism of a HIV-1 isolate for infection of CD4+ T-cell lines vs. primary macrophages, presumably by determining the selectivity of virus entry and cell fusion.

CD26 expression correlates with entry, replication and cytopathicity of monocytotropic HIV-1 strains in a T-cell line

Experiments to identify cell determinants involved in HIV-1 tropism revealed a specific decrease in the expression of the T-cell activation antigen CD26 after monocytotropic (M-tropic) but not T-cell line-tropic (T-tropic) virus infection of the PM1 T-cell line. The level of CD26 expression in single-cell clones of PM1 correlated with the entry rate and cytopathicity of M-tropic HIV-1 variants, resulting in preferential survival of cells with low CD26 levels after infection. Experiments with recombinant viruses showed that the third hypervariable region of the envelope gp120 plays an important role in this selection process. This study identifies CD26 as a key marker for M-tropic human immunodeficiency virus type 1 (HIV-1) infection and suggests a mechanism for the early loss of CD26-expressing cells in HIV-1-infected individuals.

Influence of transmembrane domains on the fusogenic abilities of human and murine leukemia retrovirus envelopes

The envelopes of two highly divergent oncoviruses, human T-cell leukemia virus type 1 (HTLV-1) and Friend murine leukemia virus (F-MuLV), have distinct patterns of cellular receptor recognition, fusion, and syncytium formation. To analyze the influence of the transmembrane envelope subunit (TM) on fusogenic properties, we substituted either the entire TM or distinct domains from F-MuLV for the corresponding domains in the HTLV-1 envelope. Parental, chimeric, and truncated envelopes cloned into a eukaryotic expression vector were monitored for fusogenic potential in human, rat, and murine indicator cell lines by using a quantitative assay. This highly sensitive assay allowed us to assess the fusogenic properties and syncytium-forming abilities of the HTLV-1 envelope in murine NIH 3T3 cells. All chimeric envelopes containing extracellular sequences of the F-MuLV TM were blocked in their maturation process. Although deletions of the HTLV-1 cytoplasmic domain, alone and in combination with the membrane-spanning domain, did not prevent envelope cell surface expression, they impaired and suppressed fusogenic properties, respectively. In contrast, envelopes carrying substitutions of membrane-spanning and cytoplasmic domains were highly fusogenic. Our results indicate that these two domains in F-MuLV and HTLV-1 constitute structural entities with similar fusogenic properties. However, in the absence of a cytoplasmic domain, the F-MuLV membrane-spanning domain appeared to confer weaker fusogenic properties than the HTLV-1 membrane-spanning domain.

Virus receptors: implications for pathogenesis and the design of antiviral agents

A virus initiates infection by attaching to its specific receptor on the surface of a susceptible host cell. This prepares the way for the virus to enter the cell. Consequently, the expression of the receptor on specific cells and tissues of the host is a major determinant of the route of entry of the virus into the host and of the patterns of virus spread and pathogenesis in the host. This review emphasizes the virus-receptor interactions of human immunodeficiency virus, the rhinoviruses, the herpesviruses, and the coronaviruses. These interactions are often found to be complex and dynamic, involving multiple sites or factors on both the virus and the host cell. Also, the receptor may play an important role in virus entry per se in addition to its role in virus binding. In the cases of human immunodeficiency virus and the rhinoviruses, ingenious approaches to therapeutic strategies based on inhibiting virus attachment and entry are under development and in clinical trials.

Proteinase-resistant factors in human erythrocyte membranes mediate CD4-dependent fusion with cells expressing human immunodeficiency virus type 1 envelope glycoproteins

Murine CD4+ cells are resistant to human immunodeficiency virus type 1 (HIV-1) entry and to fusion with cells expressing HIV-1 envelope glycoproteins (Env). The role of human-specific factors in Env/CD4-mediated fusion is shown by the ability of transient cell hybrids formed between CD4+ murine cells and human HeLa cells to fuse with Env+ cells. Fusion events were observed when other human cells, including erythrocytes, were substituted for HeLa cells in the hybrids. Experiments with erythrocyte ghosts showed that the factors allowing Env/CD4-mediated fusion are located in the plasma membrane. These factors were fully active after extensive digestion of erythrocytes with proteinase K or pronase. Nonprotein components of human plasma membranes, possibly glycolipids, could therefore be required for Env/CD4-mediated fusion and virus entry.

Heterokaryons formed between a rat myeloma and a mouse fibroblast are permissive for entry of HIV type 1

Expression of CD4 is not sufficient for cellular susceptibility to HIV-1 entry. The majority of nonprimate cells appear to lack a factor that plays an accessory role to CD4 during virus-cell fusion. We have previously reported CD4-dependent entry in rat myeloma Y3 cells, and here we show that Y3 and CD4-expressing mouse L929 cells fuse spontaneously to produce heterokaryons susceptible to entry by HIV-1. The results suggest that a putative accessory factor necessary for entry of HIV-1 into cells is expressed by the nonprimate Y3 cell line.

Factors involved in entry of the human immunodeficiency virus type 1 into permissive cells: lack of evidence of a role for CD26

It has been proposed recently that the cell surface peptidase CD26 acts in concert with CD4, the human immunodeficiency virus (HIV) primary receptor molecule, to mediate HIV entry into permissive cells. We have failed to detect significant levels of CD26 cell surface expression and enzymatic activity in a number of commonly propagated human CD4+ cell lines, although CD26 mRNA was present at very low levels, as detected by reverse transcription PCR. No relationship existed between the expression of CD26 and the ability of these cells to be infected with HIV or to fuse to form syncytia. We have tested two inhibitors of CD26 enzymatic activity and several anti-CD26 monoclonal antibodies and found that they inhibit neither HIV infection nor HIV-induced syncytium formation. NIH 3T3 cells stably transfected with the cDNAs for human CD4 and CD26 expressed these molecules at the cell surface and had CD26 enzymatic activity. Inoculation of the double transfectants with HIV did not result in virus entry above the background level, as verified by PCR amplification of viral DNA. We were unable to recover infectious virus from the HIV-inoculated NIH 3T3 double transfectants either by transfer of supernatants or by cocultivation with human CD4+ indicator cells. Moreover, the transfectants did not fuse with HIV-infected cells to form syncytia, nor were syncytia observed in HIV-inoculated cultures. These results are inconsistent with the CD26 molecule being a cofactor for entry of HIV in CD4+ cells.

Characterization of human immunodeficiency virus type 1 gp120 binding to liposomes containing galactosylceramide

Human immunodeficiency virus type 1 (HIV-1) infects some cell types which lack CD4, demonstrating that one or more alternative viral receptors exist. One such receptor is galactosylceramide (GalCer), a glycosphingolipid distributed widely in the nervous system and in colonic epithelial cells. Using a liposome flotation assay, we found that the HIV-1 surface glycoprotein, gp120, quantitatively bound to liposomes containing GalCer but not to liposomes containing phospholipids and cholesterol alone. Binding was saturable and was inhibited by preincubating liposomes with anti-GalCer antibodies. We observed less efficient binding of gp120 to liposomes containing lactosylceramide, glucosylceramide, and galactosylsulfate, whereas no binding to liposomes containing mixed gangliosides, psychosine, or sphingomyelin was detected. Binding to GalCer was rapid, largely independent of temperature and pH, and stable to conditions which remove most peripheral membrane proteins. By contrast, gp120 bound to lactosylceramide could be removed by 2 M potassium chloride or 3 M potassium thiocyanate, demonstrating a less stable interaction. Removal of N-linked oligosaccharides on gp120 did not affect binding efficiency. However, as previously observed for CD4 binding, heat denaturation of gp120 prevented binding to GalCer. Finally, binding was critically dependent on the concentration of GalCer in the target membrane, suggesting that binding to glycolipid-rich domains occurs and that GalCer conformation may be important for gp120 recognition.