The morphology of human immunodeficiency virus (HIV) by negative staining

Shear-reversible clusters of HIV-1 in solution: stabilized by antibodies, dispersed by mucin

IMPORTANCE

The phenomenon of reversible clustering is expected to further nuance HIV immune stealth because virus surfaces can escape interaction with antibodies (Abs) by hiding temporarily within clusters. It is well known that mucin reduces HIV virulence, and the current perspective is that mucin aggregates HIV-1 to reduce infections. Our findings, however, suggest that mucin is dispersing HIV clusters. The study proposes a new paradigm for how HIV-1 may broadly evade Ab recognition with reversible clustering and why mucin effectively neutralizes HIV-1.

HIV-1 is more dependent on the K182 capsid residue than HIV-2 for interactions with CPSF6

The HIV-1 capsid (CA) utilizes CPSF6 for nuclear entry and integration site targeting. Previous studies demonstrated that the HIV-1 CA C-terminal domain (CTD) contains a highly conserved K182 residue involved in interaction with CPSF6. In contrast, certain HIV-2 strains possess a substitution at this residue (K182R). To assess whether CA-CPSF6 interaction via the CA CTD is conserved among primate lentiviruses, we examined resistance of several HIV-1- and HIV-2-lineage viruses to a truncated form of CPSF6, CPSF6-358. The results demonstrated that viruses belonging to the HIV-2-lineage maintain interaction with CPSF6 regardless of the presence of the K182R substitution, in contrast to the case with HIV-1-lineage viruses. Our structure-guided mutagenesis indicated that the differential requirement for CA-CPSF6 interaction is regulated in part by residues near the 182nd amino acid of CA. These results demonstrate a previously unrecognized distinction between HIV-1 and HIV-2, which may reflect differences in their evolutionary histories.

Taxonomy

This chapter addresses the classification and taxonomy of viruses with special attention to viruses that show pneumotropic properties. Information provided in this chapter supplements that provided in other chapters in Parts II–V of this volume that discuss individual viral pathogens.

Second-site compensatory mutations of HIV-1 capsid mutations

The human immunodeficiency virus (HIV) capsid (CA) protein assembles into a hexameric lattice that forms the mature virus core. Contacts between the CA N-terminal domain (NTD) of one monomer and the C-terminal domain (CTD) of the adjacent monomer are important for the assembly of this core. In this study, we have examined the effects of mutations in the NTD region associated with this interaction. We have found that such mutations yielded modest reductions of virus release but major effects on viral infectivity. Cell culture and in vitro assays indicate that the infectivity defects relate to abnormalities in the viral cores. We have selected second-site compensatory mutations that partially restored HIV infectivity. These mutations map to the CA CTD and to spacer peptide 1 (SP1), the portion of the precursor Gag protein immediately C terminal to the CTD. The compensatory mutations do not locate to the molecularly modeled intermolecular NTD-CTD interface. Rather, the compensatory mutations appear to act indirectly, possibly by realignment of the C-terminal helix of the CA CTD, which participates in the NTD-CTD interface and has been shown to serve an important role in the assembly of infectious virus.

HIV-1 capsid protein forms spherical (immature-like) and tubular (mature-like) particles in vitro: structure switching by pH-induced conformational changes

The viral genome and replicative enzymes of the human immunodeficiency virus are encased in a shell consisting of assembled mature capsid protein (CA). The core shell is a stable, effective protective barrier, but is also poised for dissolution on cue to allow transmission of the viral genome into its new host. In this study, static light scattering (SLS) and dynamic light scattering (DLS) were used to examine the entire range of the CA protein response to an environmental cue (pH). The CA protein assembled tubular structures as previously reported but also was capable of assembling spheres, depending on the pH of the protein solution. The switch from formation of one to the other occurred within a very narrow physiological pH range (i.e., pH 7.0 to pH 6.8). Below this range, only dimers were detected. Above this range, the previously described tubular structures were detected. The ability of the CA protein to form a spherical structure that is detectable by DLS but not by electron microscopy indicates that some assemblages are inherently sensitive to perturbation. The dimers in equilibrium with these assemblages exhibited distinct conformations: Dimers in equilibrium with the spherical form exhibited a compact conformation. Dimers in equilibrium with the rod-like form had an extended conformation. Thus, the CA protein possesses the inherent ability to form metastable structures, the morphology of which is regulated by an environmentally-sensitive molecular switch. Such metastable structures may exist as transient intermediates during the assembly and/or disassembly of the virus core.

Biochemical and structural analysis of isolated mature cores of human immunodeficiency virus type 1

Mature human immunodeficiency virus type 1 (HIV-1) particles contain a cone-shaped core structure consisting of the internal ribonucleoprotein complex encased in a proteinaceous shell derived from the viral capsid protein. Because of their very low stability after membrane removal, HIV-1 cores have not been purified in quantities sufficient for structural and biochemical analysis. Based on our in vitro assembly experiments, we have developed a novel method for isolation of intact mature HIV-1 cores. Concentrated virus suspensions were briefly treated with nonionic detergent and immediately centrifuged in a microcentrifuge for short periods of time. The resuspended pellet was subsequently analyzed by negative-stain and thin-section electron microscopy and by immunoelectron microscopy. Abundant cone-shaped cores as well as tubular and aberrant structures were observed. Stereo images showed that core structures preserved their three-dimensional architecture and exhibited a regular substructure. Detailed analysis of 155 cores revealed an average length of ca. 103 nm, an average diameter at the base of ca. 52 nm, and an average angle of 21.3 degrees. There was significant variability in all parameters, indicating that HIV cores are not homogeneous. Immunoblot analysis of core preparations allowed semiquantitative estimation of the relative amounts of viral and cellular proteins inside the HIV-1 core, yielding a model for the topology of various proteins inside the virion.

Development of HIV/AIDS vaccine using chimeric gag-env virus-like particles

We attempted to develop a candidate HIV/AIDS vaccine, by using unprocessed HIV-2 gag pr45 precursor protein. We found that a 45 kDa unprocessed HIV-2 gag precursor protein (pr45), with a deletion of a portion of the viral protease, assembles as virus-like particles (VLP). We mapped the functional domain of HIV-2 gag VLP formation in order to find the minimum length of gag protein to form VLP. A series of deletion mutants was constructed by sequentially removing the C-terminal region of HIV-2 gag precursor protein and expressed truncated genes in Spodoptera frugiperda (SF) cells by infecting recombinant baculoviruses. We found that deletion of up to 143 amino acids at the C-terminus of HIV-2 gag, leaving 376 amino acids at the N-terminus of the protein, did not affect VLP formation. There is a proline-rich region at the amino acid positions 373 to 377 of HIV-2 gag, and replacement of these proline residues by site-directed mutagenesis completely abolished VLP assembly. Our data demonstrate that the C-terminal p12 region of HIV-2 gag precursor protein, and zinc finger domains, are dispensable for gag VLP assembly, but the presence of at least one of the three prolines at amino acid positions 373, 375 or 377 of HIV-2NIH-Z is required for VLP formation. Animals immunized with these gag particles produced high titer antibodies and Western blot analyses showed that anti-gag pr45 rabbit sera react with p17, p24 and p55 gag proteins of HIV-1. We then constructed chimeric gag genes, which carry the hypervariable V3 region of HIV-1 gp120, because the V3 loop is known to interact with chemokine receptor as a coreceptor, and known to induce the major neutralizing antibodies and stimulate the cytoxic T lymphocyte responses in humans and mice. We expressed chimeric fusion protein of HIV-2 gag with 3 tandem copies of consensus V3 domain that were derived from 245 different isolates of HIV-1. In addition, we also constructed and expressed chimeric fusion protein that contains HIV-2 gag with V3 domains of HIV-1IIIB, HIV-1MN, HIV-1SF2 and HIV-1RF. The chimeric gag-env particles had a spherical morphology, and the size was slightly larger than that of a gag particle. Immunoprecipitation and Western blot analyses show that these chimeric proteins were recognized by HIV-1 positive human sera and antisera raised against V3 peptides, as well as by rabbit anti-gp120 serum. We obtained virus neutralizing antibodies in rabbits by immunizing these gag-env VLPs. In addition, we found that gag-env chimeric VLPs induce a strong CTL activity against V3 peptide-treated target cells. Our results indicate that V3 peptides from all major clades of HIV-1 carried by HIV-2 gag can be used as a potential HIV/AIDS vaccine.

Determination of the size of HIV using adenovirus type 2 as an internal length marker

The size of a virion is a key criterion to its proper classification and may have implications in many practical aspects. Size determinations by thin section electron microscopy often result in length aberrations of more than 10% because of a number of preparative and instrumental inaccuracies, e.g. specimen shrinkage or swelling and unreliable calibration. Using adenovirus type 2 as an isometric size marker for internal calibration, we have determined the diameters of mature and immature HIV-1 to be 110 to 128 and 132 to 146 nm, respectively. The marker had been used either as a purified particle suspension added to the HIV producing culture, or adenovirus had been propagated together with HIV by infecting HIV producing cells. Using well characterized isometric markers, e.g. an icosahedral virus, in thin section electron microscopy appears to be a suitable technique for viral size determinations.

Inner architecture of human and simian immunodeficiency viruses

The cores of human and simian immunodeficiency viruses (HIV and SIV) were observed by negative staining after isolation of the core with Nonidet P40 and glutaraldehyde. Four kinds of cores were found: asymmetric and symmetric sectoral shapes, a bar shape, and a triangular shape. These results were confirmed by the examination of ultrathin sections of whole virions. In some virions, the connection between the core and the envelope was observed after freeze fracturing. Its structure was considered to be characteristic of an intermediate stage of viral maturation. The HIV-1 core was reacted with anti-HIV-1 p24 mouse monoclonal antibody.

Small virus-like particles bud from the cell membranes of normal as well as HIV-infected human lymphoid cells

Electron microscopy (EM) of cell sections showed cell associated virus-like particles (VLP), 50-60 nm in diameter, budding from the membrane of human lymphoid cells in culture. The particles had an envelope continuous with the cell membrane and a dense core that almost filled the particle. Particles 70-80 nm in diameter with prominent external spikes were found in the culture medium by negative staining (medium-associated VLP). Cell-associated VLP were also present in cord lymphocytes, both on initial separation and after culture with or without foetal calf serum, and therefore were considered to be endogenous to the cells and were not bovine diarrhoea virus. VLP were observed in most of the lymphoid cell lines examined. VLP were also found less frequently in established human tumour and nontumour cell lines. Both cell-associated and medium-associated VLP were also present in HIV infected cell cultures, and they could be distinguished from HIV by their characteristic morphology and smaller size. It was not determined whether the 2 types of particle represented the same entity, or whether they were defective virus particles or a cellular secretory product.

Chimeric gag-V3 virus-like particles of human immunodeficiency virus induce virus-neutralizing antibodies

A 41-kDa unprocessed human immunodeficiency virus 2 (HIV-2) gag precursor protein that has a deletion of a portion of the viral protease assembles as virus-like particles by budding through the cytoplasmic membrane of recombinant baculovirus-infected insect cells. We have constructed six different combinations of chimeric genes by coupling the truncated HIV-2 gag gene to the neutralizing domain (V3) or the neutralizing and the CD4 binding domains (V3+CD4BD) of gp120 env gene sequences from HIV-1 or HIV-2. The env gene sequences were inserted either into the middle of the gag gene or at the 3' terminus of the gag gene. Virus-like particles were formed by chimeric gene products only when the env gene sequences were linked to the 3' terminus of the gag gene. Insertion of env gene sequence in the middle of the gag gene resulted in high-level chimeric gene expression but without the formation of virus-like particles. Three different chimeric genes [gag gene with HIV-1 V3 (1V3), gag gene with HIV-2 V3 (2V3), and gag gene with HIV-2 V3+CD4BD (2V3+CD4BD)] formed virus-like particles that were secreted into the cell culture medium. In contrast, the HIV-1 V3+CD4BD/HIV-2 gag construct did not form virus-like particles. The chimeric gag-env particles had spherical morphology and the size was slightly larger than that of the gag particles, but the chimeric particles were similar to the mature HIV particles. Western blot analysis showed that the gag-env chimeric proteins were recognized by antibodies in HIV-positive human serum and rabbit anti-gp120 serum. Rabbit anti-gag 1V3 and anti-gag 2V3 sera reacted with authentic gp120 of HIV-1 and HIV-2, respectively, and neutralized homologous HIV infectivity. Our results show that precursor gag protein has potential as a carrier for the presentation of foreign epitopes in good immunological context. The gag protein is highly immunogenic and has the ability to carry large foreign inserts; as such, it offers an attractive approach for HIV vaccine development.

Unusual features of protein interaction in human immunodeficiency virus (HIV) virions

The treatment of HIV-1 virions with ionic and nonionic detergents (NP 40, octylglucoside, Na deoxycholate) resulted in an effect unusual for enveloped viruses: instead of solubilization of glycoproteins, the core protein p24 was solubilized while envelope glycoproteins with other structural proteins were found in subviral particles. These data are consistent with a model of HIV structural organization in which glycoproteins are included in the matrix formed by the protein p 17 and suggest that p24 is neither involved in the matrix nor closely bound to any viral proteins.

Production of human immunodeficiency virus (HIV)-like particles from cells infected with recombinant vaccinia viruses carrying the gag gene of HIV

We constructed a recombinant vaccinia virus carrying the entire gag and pol genes of human immunodeficiency virus type 1 (HIV-1). The main gene product detected in the lysates of infected CV-1 and SW480 cells was the gag precursor protein. However, in the culture fluid of infected SW480 cells, but not of infected CV-1 cells, reverse transcriptase (RT) activity was detected. The highest RT activity was found at a density of 1.15 g/ml and this fraction contained many round particles with diameters of 100-150 nm. In contrast to the infected cell lysates, the particles contained the processed gag and pol proteins, suggesting that particle formation may be a prerequisite for efficient processing of the gag precursor by the HIV protease encoded in the pol gene. Particles were also recovered from the culture fluid of SW480 cells infected with another recombinant vaccinia virus carrying only the gag gene. These particles contained the unprocessed gag precursor, indicating that the gag precursor alone was sufficient for particle production.

Morphogenesis and morphology of HIV. Structure-function relations

Fine structure and antigenic make-up analysis of HIV were combined in a 2D model, from which functional aspects can be deduced. On the envelope 72 probably trimeric surface knobs (gp120) are connected to the virion via the transmembrane protein gp41. Gp120 is shed during ageing of the virion, but host cell antigens stay firmly anchored to the envelope. Underneath the envelope, p17 forms the matrix protein layer, while the capsid of the double cone shaped core is built up of p24. The relation between biochemical findings and morphogenesis and maturation of HIV as well as aspects of pathogenesis and vaccination are discussed.

Recovery of antigenically reactive HIV-2 cores

Negative staining studies of human immunodeficiency virus (HIV) have been hampered by the fragile nature of the particles. Although detergent treatment is capable of releasing cores from HIV-2 particles, these are unstable and do not retain morphological integrity. Addition of glutaraldehyde will stabilise these structures but, if used at too high a concentration, will destroy their antigenicity. This study shows that if both detergent and glutaraldehyde are used in correct proportions, antigenically reactive cores can be recovered from HIV-2 cell cultures. More specifically we show that a mixture of 0.1% Nonidet P40 and 0.1% glutaraldehyde produces preparations of HIV-2 cores that are suitable for immune electron microscopy. These cores reacted positively, that is, formed immune complexes, with both human HIV-2 antisera and a mouse monoclonal antibody that, although directed against p24 (HIV-1), reacts also with p25 (HIV-2).