Effect of amino acid changes in the V1/V2 region of the human immunodeficiency virus type 1 gp120 glycoprotein on subunit association, syncytium formation, and recognition by a neutralizing antibody.

The contributions of the first and second variable regions of the human immunodeficiency virus type 1 gp120 glycoprotein to envelope glycoprotein structure, function, and recognition by a neutralizing antibody were studied. Several mutants with substitutions in the V2 loop demonstrated complete dissociation of the gp120 and gp41 glycoproteins, suggesting that inappropriate changes in V2 conformation can affect subunit assembly. Some glycoproteins with changes in V1 or V2 were efficiently expressed on the cell surface and were able to bind CD4 but were deficient in syncytium formation and/or virus entry. Recognition of gp120 by the neutralizing monoclonal antibody G3-4 was affected by particular substitutions affecting residues 176 to 184 in the V2 loop. These results suggest that the V1/V2 variable regions of the human immunodeficiency virus type 1 gp120 glycoprotein play a role in postreceptor binding events in the membrane fusion process and can act as a target for neutralizing antibodies.     

Characterization of conserved human immunodeficiency virus type 1 gp120 neutralization epitopes exposed upon gp120-CD4 binding.

Interaction with the CD4 receptor enhances the exposure on the human immunodeficiency type 1 gp120 exterior envelope glycoprotein of conserved, conformation-dependent epitopes recognized by the 17b and 48d neutralizing monoclonal antibodies. The 17b and 48d antibodies compete with anti-CD4 binding antibodies such as 15e or 21h, which recognize discontinuous gp120 sequences near the CD4 binding region. To characterize the 17b and 48d epitopes, a panel of human immunodeficiency virus type 1 gp120 mutants was tested for recognition by these antibodies in the absence or presence of soluble CD4. Single amino acid changes in five discontinuous, conserved, and generally hydrophobic regions of the gp120 glycoprotein resulted in decreased recognition and neutralization by the 17b and 48d antibodies. Some of these regions overlap those previously shown to be important for binding of the 15e and 21h antibodies or for CD4 binding. These results suggest that discontinuous, conserved epitopes proximal to the binding sites for both CD4 and anti-CD4 binding antibodies become better exposed upon CD4 binding and can serve as targets for neutralizing antibodies.     

Separation of the complex DNA binding domain of EBNA-1 into DNA recognition and dimerization subdomains of novel structure.

EBNA-1 is essential for replication of the latent episomal form of the Epstein-Barr virus genome and is involved in regulation of viral latency promoters. EBNA-1 activity is mediated through direct DNA binding. The DNA binding and dimerization functions of EBNA-1 have previously been located to a carboxy-terminal domain, amino acids (aa) 459 to 607. To identify and define the subdomains for these two functions, we created an extensive series of deletions and point mutations in an EBNA-1 (aa 408 to 641) background. The ability of the EBNA-1 mutants to heterodimerize with a wild-type EBNA-1 (aa 459 to 641) Immunoprecipitation assays with a monoclonal antibody, EBNA.OT1x, that recognizes EBNA-1 (aa 408 to 641) but not EBNA-1 (aa 459 to 641). These experiments revealed that mutations affecting dimerization occurred over two separate regions, aa 501 to 532 and aa 554 to 598. DNA binding was tested in mobility shift assays against a panel of oligonucleotide-binding sites. Dimerization was a prerequisite for DNA binding. The DNA recognition domain was localized to a separate region, aa 459 to 487, upstream of the dimerization domain. EBNA-1 variants carrying substitutions at aa 467 and 468 and at aa 477 gave a pattern of binding to mutant oligonucleotide probes that implicates these particular amino acids in DNA recognition. EBNA-1 appears to utilize novel mechanisms for both DNA recognition and dimerization since neither domain conforms to previously described structural motifs.     

Antibodies to discontinuous or conformationally sensitive epitopes on the gp120 glycoprotein of human immunodeficiency virus type 1 are highly prevalent in sera of infected humans.

We have used an indirect-capture enzyme-linked immunosorbent assay to quantitate the reactivity of sera from human immunodeficiency virus type 1 (HIV-1)-infected humans with native recombinant gp120 (HIV-1 IIIB or SF-2) or with the gp120 molecule (IIIB or SF-2) denatured by being boiled in the presence of dithiothreitol with or without sodium dodecyl sulfate. Denaturation of IIIB gp120 reduced the titers of sera from randomly selected donors by at least 100-fold, suggesting that the majority of cross-reactive anti-gp120 antibodies present are directed against discontinuous or otherwise conformationally sensitive epitopes. When SF-2 gp120 was used, four of eight serum samples reacted significantly with the denatured protein, albeit with ca. 3- to 50-fold reductions in titer. Only those sera reacting with denatured SF-2 gp120 bound significantly to solid-phase-adsorbed SF-2 V3 loop peptide, and none bound to IIIB V3 loop peptide. Almost all antibody binding to reduced SF-2 gp120 was blocked by preincubation with the SF-2 V3 loop peptide, as was about 50% of the binding to native SF-2 gp120. When sera from a laboratory worker or a chimpanzee infected with IIIB were tested, the pattern of reactivity was reversed, i.e., there was significant binding to reduced IIIB gp120, but not to reduced SF-2 gp120. Binding of these sera to reduced IIIB gp120 was 1 to 10% that to native IIIB gp120 and was substantially decreased by preincubation with IIIB (but not SF-2) V3 loop peptide. To analyze which discontinuous or conformational epitopes were predominant in HIV-1-positive sera, we prebound monoclonal antibodies (MAbs) to IIIB gp120 and then added alkaline phosphatase-labelled HIV-1-positive sera. MAbs (such as 15e) that recognize discontinuous epitopes and compete directly with CD4 reduced HIV-1-positive sera binding by about 50%, whereas neutralizing MAbs to the C4, V2, and V3 domains of gp120 were either not inhibitory or only weakly so. Thus, antibodies to the discontinuous CD4-binding site on gp120 are prevalent in HIV-1-positive sera, antibodies to linear epitopes are less common, most of the antibodies to linear epitopes are directed against the V3 region, and most cross-reactive antibodies are directed against discontinuous epitopes, including regions involved in CD4 binding.     

Secretion of streptavidin from Bacillus subtilis.

Streptavidin is an extracellular tetrameric protein produced by Streptomyces avidinii. A series of hybrid gene fusions consisting of Bacillus signal peptide coding regions fused to the mature streptavidin sequence were constructed. B. subtilis strains harboring these plasmids accumulate a tetrameric streptavidin in the growth medium. The properties of the streptavidin produced by B. subtilis are similar to those of the streptavidin produced by S. avidinii. B. subtilis strains carrying the various fusions can be grown to a high cell density in a biotin-free medium. Thus, B. subtilis represents an alternate host system for the production of streptavidin.     

Fluorescence dequenching kinetics of single cell-cell fusion complexes.

In an earlier paper which models the cell-cell (or virus-cell) fusion complex as two partial spherical vesicles joined at a narrow neck (Rubin, R. J., and Yi-der Chen. 1990. Biophys. J. 58:1157-1167), the redistribution by diffusion of lipid-like molecules through the neck between the two fused cell surfaces was studied. In this paper, we extend the study to the calculation of the kinetics of fluorescence increase in a single fusion complex when the lipid-like molecules are fluorescent and self-quenching. The formalism developed in this paper is useful in deducing fusion activation mechanisms from cuvette fluorescence measurements in cell-cell fusion systems. Two different procedures are presented: 1) an exact one which is based on the exact local density functions obtained from diffusion equations in our earlier study; and 2) an approximate one which is based on treating the kinetics of transfer of probes between the two fused cells as a two-state chemical reaction. For typical cell-cell fusion complexes, the fluorescence dequencing curves calculated from the exact and approximate procedures are very similar. Due to its simplicity, the approximate method should be very useful in future applications. The formalism is applied to a typical cell-cell fusion complex to study the sensitivity of dequenching curves to changes in various fusion parameters, such as the radii of the cells, the radius of the pore at the fusion junction, and the number of probes initially loaded to the complex.     

Neutron structure factors of in-vivo deuterated amorphous protein C-phycocyanin

Neutron powder diffraction measurements of fully deuterated protein C-phycocyanin have been made at three temperatures, 295, 200, and 77 K, using dry and partially hydrated samples. The average coherent structure factors and the corresponding radial distribution functions d(r) are determined. The changes in d(r) functions observed in hydrated samples depend strongly on the level of hydration and most of these changes are due to water-protein interactions. At 0.365 gram D₂O per gram of protein, the water crystallized into hexagonal ice at 200 K and below, but at 0.175 gram D₂O per gram of protein, no crystallization of water was observed. At the higher hydration a peak appears in the radial distribution function which indicates that the average distance of the water molecule in the first hydration shell from the amino acid residues is 3.5 Å.     

Measurements of conformational changes during adhesion of lipid and protein (polylysine and S-layer) surfaces

The adhesion forces between various surfaces were measured using the "surface forces apparatus" technique. This technique allows for the thickness of surface layers and the adhesion force between them to be directly measured in controlled vapor or liquid environments. Three types of biological surfaces were prepared by depositing various lipid-protein monolayers (with thicknesses ranging from 1 to 4 nm) on the inert, molecularly smooth mica surface: (i) hydrophobic lipid monolayers; (ii) amphiphilic polyelectrolyte surfaces of adsorbed polylysine; and (iii) deposited bacterial S-layer proteins. The adhesion, swelling, and wetting properties of these surfaces was measured as a function of relative humidity and time. Initial adhesion is due mainly to the van der Waals forces arising from nonpolar (hydrophobic) contacts. Following adhesive contact, significant molecular rearrangements can occur which alter their hydrophobic-hydrophilic balance and increase their adhesion with time. Increased adhesion is generally enhanced by (i) increased relative humidity (or degree of hydration); (ii) increased contact time; and (iii) increased rates of separation. The results are likely to be applicable to the adhesion of many other biosurfaces, and show that the hydrophobicity of a lipid or protein surface is not an intrinsic property of that surface but depends on its environment (e.g., on whether it is in aqueous solution or exposed to the atmosphere), and on the relative humidity of the atmosphere. It also depends on whether the surface is in adhesive contact with another surface and-when considering dynamic (nonequilibrium) conditions-on the time and previous history of its interaction with that surface. (c) 1993 John Wiley & Sons, Inc.     

A molecular,isozyme and morphological map of the barley (Hordeum vulgare) genome

A map of the barley genome consisting of 295 loci was constructed. These loci include 152 cDNA restriction fragment length polymorphism (RFLP), 114 genomic DNA RFLP, 14 random amplified polymorphic DNA (RAPD), five isozyme, two morphological, one disease resistance and seven specific amplicon polymorphism (SAP) markers. The RFLP-identified loci include 63 that were detected using cloned known function genes as probes. The map covers 1,250 centiMorgans (cM) with a 4.2 cM average distance between markers. The genetic lengths of the chromosomes range from 124 to 223 cM and are in approximate agreement with their physical lengths. The centromeres were localized to within a few markers on all of the barley chromosomes except chromosome 5. Telomeric regions were mapped for the short (plus) arms of chromosomes 1, 2 and 3 and the long (minus) arm of chromosomes 7.This research was also supported by other members of the NABGMP: K. Kasha, Department of Crop Science, University of Guelph, Guelph, Ontario, Canada NIG 2W1; W. Kim, Agriculture Canada Research Station, 195 Dafoe Road, Winnipeg, Manitoba, Canada R3T 2M9; A. Laroche, Agriculture Canada Research Station, P.O. Box 3000 Main, Lethbridge, Alberta, Canada,TU 4B1; S. Molnar, Plant Research Centre Agriculture Canada, Central Experimental farm, Ottawa, Ontario, Canada K1A 0C6; G. Scoles, Department of Crop Science, University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N OWOThis research is part of the North American Barley Genome Mapping Project, R. A. Nilan and K. Kasha, Coordinator and Associate Coordinator, respectively Permanent address: Department of Plant Genetics, NI Vavilov Institute of General Genetics, Russian Academy of Sciences, Moscow