The Ebola filoviruses are aggressive pathogens that cause severe and often lethal hemorrhagic fever syndromes in humans and nonhuman primates. To date, no effective therapies have been identified. To analyze the entry and fusion properties of Ebola virus, we adapted a human immunodeficiency virus type 1 (HIV-1) virion-based fusion assay by substituting Ebola virus glycoprotein (GP) for the HIV-1 envelope. Fusion was detected by cleavage of the fluorogenic substrate CCF2 by ␤-lactamase-Vpr incorporated into virions and released as a result of virion fusion. Entry and fusion induced by the Ebola virus GP occurred with much slower kinetics than with vesicular stomatitis virus G protein (VSV-G) and were blocked by depletion of membrane cholesterol and by inhibition of vesicular acidification with bafilomycin A1. These properties confirmed earlier studies and validated the assay for exploring other properties of Ebola virus GP-mediated entry and fusion. Entry and fusion of Ebola virus GP pseudotypes, but not VSV-G or HIV-1 Env pseudotypes, were impaired in the presence of the microtubule-disrupting agent nocodazole but were enhanced in the presence of the microtubule-stabilizing agent paclitaxel (Taxol). Agents that impaired microfilament function, including cytochalasin B, cytochalasin D, latrunculin A, and jasplakinolide, also inhibited Ebola virus GP-mediated entry and fusion. Together, these findings suggest that both microtubules and microfilaments may play a role in the effective trafficking of vesicles containing Ebola virions from the cell surface to the appropriate acidified vesicular compartment where fusion occurs. In terms of Ebola virus GP-mediated entry and fusion to various target cells, primary macrophages proved highly sensitive, while monocytes from the same donors displayed greatly reduced levels of entry and fusion. We further observed that tumor necrosis factor alpha, which is released by Ebola virus-infected monocytes/macrophages, enhanced Ebola virus GP-mediated entry and fusion to human umbilical vein endothelial cells. Thus, Ebola virus infection of one target cell may induce biological changes that facilitate infection of secondary target cells that play a key role in filovirus pathogenesis. Finally, these studies indicate that pseudotyping in the HIV-1 virion-based fusion assay may be a valuable approach to the study of entry and fusion properties mediated through the envelopes of other viral pathogens.

Natural alpha interferon (IFN-␣)-producing cells (IPCs

In view of the cationic amphipathic structure of tachyplesin I and antiparallel beta-sheet as a general DNA binding motif, DNA binding of the antimicrobial peptide has been examined. Several footprinting-like techniques using DNase I protection, dimethyl sulfate protection, and bleomycin- (BLM-) induced DNA cleavage were applied in this study. Some distinct footprints with DNase I are detected, and also the sequence-specific cleavage mode of the BLM-Fe(II) complex clearly is altered in the presence of tachyplesin I. In addition, methylation of the N-7 residue of guanine situated in the DNA major groove is not entirely inhibited (or activated) by tachyplesin I. The results suggest that tachyplesin I interacts with the minor groove of DNA duplex. Disappearance of the footprints by dithiothreitol-treated tachyplesin I and Ala-tachyplesin strongly suggests a significant contribution of secondary structure containing an antiparallel beta-sheet to the DNA binding of tachyplesin I. This is the first report on DNA interaction with a small peptide which contains a unique antiparallel beta-sheet structure. The mechanism for antimicrobial action of tachyplesin I has also been inferred.

Transcription factor Sp1 has three tandem repeats of a Cys2His2-type zinc finger motif and specifically binds to GC box DNA. We investigated the interaction of Sp1 with GC box DNA by several footprinting techniques. Methylation of four guanine bases in the sequence 5'-GGGCG-3' is strongly protected by Sp1 binding, whereas a guanine base flanked at the 5' end by the above sequence is extremely hypermethylated. Methylation interference experiments explicitly show that four guanine bases from the guanine-rich strand, and one from the cytosine-rich strand, in the sequence 5'-GGGCG-3' are crucially required for GC box recognition by Sp1. In footprinting using the 1,10-phenanthroline-copper complex, binding of Sp1 clearly alters the cleavage patterns by the metal complex. Footprints of the protein did not cover the full length of each GC box sequence, and the protein strongly masked scission in the sequence 5'-GCGG(A/G)(G/A)-3'. In cleavage of GC box DNA by the bleomycin-iron complex, Sp1 binding induces new cutting at a 5'-GA-3' site within the box. The results indicate that (i) the three zinc fingers do not contribute equivalently to the binding of Sp1 to the GC box, namely, important base contacts arise from the second and third fingers, and (ii) the protein binding induces local but significant structural distortion of the 3' region of the guanine-rich strand in the GC box. These features are clearly distinct from those of Zif268 and Krox20, which are three-zinc-finger proteins closely related to Sp1.

Certain types of chemokine receptors have been identified as coreceptors for HIV-1 infection. The process of viral entry is initiated by the interaction between an envelope protein gp120 of HIV-1, CD4, and one of the relevant coreceptors. To understand the precise mechanism of the Env-mediated fusion and entry of HIV-1, we examined whether the V3 region of gp120 of T-cell line tropic (T-tropic) virus directly interacts with the coreceptor, CXCR-4, by using five synthetic V3 peptides: two cyclized V3 peptides (V3-BH10 and V3-ELI) which correspond to the V3 regions of the T-tropic HIV-1 IIIB and HIV-1 ELI strains, respectively, a linear V3 peptide (CTR36) corresponding to that of HIV-1 IIIB strain; and cyclized V3 peptides corresponding to that of the macrophage-tropic (M-tropic) HIV-1 ADA strain (V3-ADA) or the dualtropic HIV-1 89.6 strain (V3-89.6). FACScan analysis with a CXCR-4+ human B-cell line, JY, showed that V3-BH10, V3-ELI, and V3-89.6 but not CTR36 or V3-ADA blocked the binding of IVR7, an anti-CXCR-4 monoclonal antibody (MAb), to CXCR-4 with different magnitudes in a dose-dependent manner, while none of the V3 peptides influenced binding of an anti-CD19 MAb at all. Next, the effects of the V3 peptides on SDF-1β-induced transient increases in intracellular Ca2+ were investigated. Three V3 peptides (V3-BH10, V3-ELI, and V3-89.6) prevented Ca2+mobilization. Furthermore, the three peptides inhibited infection by T-tropic HIV-1 in a dose-dependent manner as revealed by an MTT assay and a reverse transcriptase assay, while the other peptides had no effects. These results present direct evidence that the V3 loop of gp120 of T-tropic HIV-1 can interact with its coreceptor CXCR-4 independently of the V1/V2 regions of gp120 or cellular CD4.