Research ArticleVIROLOGY

Triterpenoids manipulate a broad range of virus-host fusion via wrapping the HR2 domain prevalent in viral envelopes

See allHide authors and affiliations

Science Advances  21 Nov 2018:
Vol. 4, no. 11, eaau8408
DOI: 10.1126/sciadv.aau8408
  • Fig. 1 Schematic representation of virus-host membrane fusion via a common trimer-of-hairpins by which EBOV, HIV, IAV, and other viruses with envelopes enter cells.

    (A) Pathway for converting distinct viral envelopes into the common trimer-of-hairpins conformation, which brings the viral and host membranes into close proximity for fusion. The interaction of HR2 with HR1 via hydrophobic residues, mainly at positions a and d, plays a critical role in the formation of the trimer-of-hairpins. (B) Lead compound and probe discovered among triterpenoid products that can inhibit EBOV-host fusion. 293T and A549 cells were used as the host cells.

  • Fig. 2 Identification of the EBOV envelope GP as the target of the triterpenoid leads.

    (A) Scheme representing the mechanism by which the triterpenoid probe Y18 covalently captures the target protein via sequential photocrosslinking capture and click-mediated biotin harvest. (B) Western blot analyses of the cross-linked protein with Y18 using anti-GP, anti-HA, or anti-VSVG antibodies. The packaged cell lysates, pretreated with 50 μM Y0, Y11, or Y12 for 30 min, were incubated with 5 μM Y18 and then exposed to UV light (365 nm) before being treated with azide-containing biotin. (C) Mass spectrum of the captured protein with the cross-linking site identified by peptide mapping. m/z, mass/charge ratio. (D) SPR characterization of the affinity between the triterpenoid compounds, Y11 and Y18, and the HR2 peptide, which was immobilized on an SA sensor chip. Y12 and E-64d served as negative controls. (E) SPR characterization of the effects of the triterpenoid compounds on HR1-HR2 interaction. HR2 was allowed to flow across the HR1 chip surface in the absence or presence of the lead compound; EboIZN39IQ and E-64d served as positive and negative controls, respectively.

  • Fig. 3 Characterization of the spatial interactions within the triterpenoid-HR2 complex.

    (A) Overlapped HR2 acylamino 1H NMR spectra in the absence and presence of Y11 to differentiate the residues interacting with the triterpenoid lead. (B) NMR TOCSY spectra of the HR2 peptide (KIDQIIHDF) in the absence or presence of Y11. Upon the addition of Y11 (ratio of HR2/Y11 as 1:1), the signals (blue arrows) of I627 β-H and F630 α-H from HR2 shifted. (C) Assignment of each of the intermolecular NOE signal from the triterpenoid-HR2 complex. (D) Global HR2-Y11 complex inferred from the NOE signals plus docking simulation (Protein Data Bank: 5JQ3). The intermolecular NOEs are indicated by orange dashed lines.

  • Fig. 4 Identification of HIV GP41 HR2 as the target domain of the triterpenoid lead compounds.

    (A) Structures of the anti-HIV triterpenoid lead Y19 and its photocrosslinking probe Y20. (B) Characterization of the inhibitory activity of the triterpenoids Y19 and Y20 on the infectivity of HIV-1 pseudovirions. The concentrations of Y19 and Y20 used were 1 μM. CEM 4 was used as the host cell. (C) Identification of HR2 as the photocrosslinked domain by MS. The HIV GP41 protein was photoaffinity labeled with the Y20 probe and then analyzed by peptide mapping. A molecular weight increase of 772.5598, corresponding to photoactivated Y20, was observed for the peptide 623-WNNMTWMEWDREINNYTSLIHSLIEESQNQQEK-655. The cross-linking site at M626 was confirmed by the MALDI-MS analysis of tryptic digests. (D) Structural representations of Y19 and Y20 binding to the HIV GP41 HR2 domain (Protein Data Bank: 4TVP), as inferred from docking simulations. The -COOH group of the lead compounds lies in the polar pocket generated by D624, N625, and T627, and the 3-OH group of the lead compounds lies near the polar residues S649 and Q650. The diazirine group of Y20 is in close proximity to M626, the residue that is cross-linked by Y20. Multiple hydrophobic interactions are observed between the lead compounds and their binding sites, namely, W628, W631, I635, Y638, I642, L645, and L646.

  • Fig. 5 Identification of HR2 in influenza HA2 as the target domain for triterpenoid leads.

    (A) Structures of the anti-influenza triterpenoid lead Y3 and its photocrosslinking probe Y21. (B) Characterization of the inhibitory activity of the triterpenoids Y3 and Y21 on the infectivity of influenza HA–HIV pseudovirions (IAVpp). A549 was used as the host cell. (C) Identification of HR2 as the photocrosslinked domain by MS. The influenza HA protein was photoaffinity labeled with the Y21 probe and then analyzed by peptide mapping. A molecular weight increase of 825.5027, corresponding to photoactivated Y21, was observed in the peptide 471-NNAKEIGNGCFEFYHK-486 [influenza A/California/04/2009 (H1N1)]. The cross-linking site was within a residue between N471 and I476 based on the MALDI-MS analysis of tryptic digests. (D) Structural representations of Y3 and Y21 binding to the HR2 domain of influenza HA (Protein Data Bank: 3LZG) according to both the docking simulation and the alanine screening data. The amino acid residues that are involved in interactions with the lead compounds are labeled. The Y21 diazirine group forms a hydrogen bond with N672, which is located near the cross-linking site.

  • Fig. 6 The structural similarity of HR2 provides the basis by which triterpenoids block virus–cell membrane fusion.

    (A) Common helical wheel representation of the six-helix bundle, the post-fusion form of viral fusion proteins, and helical wheel representations of the HR2 domains of EBOV GP2, HIV GP41, and IAV HA, with the amino acid residues identified. The residues at positions a and d are hydrophobic residues, whereas the residues at positions b, c, e, f, and g are polar residues. The residues at positions a and d of the HR1 domain form the trimer interface, and the HR2 domain folds into the groove of the HR1 trimer via its hydrophobic face formed by the a and d residues. The residues involved in triterpenoid binding (depicted in orange) are primarily distributed at positions a, d, e, and g. (B) Shared mechanism for the triterpenoid-mediated inhibition of membrane fusion between cells and the EBOV, HIV, or influenza virus. The triterpenoids bind to the HR2 domain of a class I viral fusion protein, disrupting its interaction with the HR1 trimer. Thus, membrane fusion between the virus and the host cell is inhibited.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/11/eaau8408/DC1

    Supplementary Materials and Methods

    Fig. S1. Discovery of pentacyclic triterpene glycoconjugates as EBOV entry inhibitors.

    Fig. S2. Time-of-addition experiments to clarify the stage at which lead compounds blocked EBOV entry.

    Fig. S3. Characterization of the affinity of triterpenoid compounds to N-terminal HR1, C-terminal HR2, and their effect on HR1-HR2 interaction.

    Fig. S4. Characterization of the spatial interactions within the triterpenoid-HR2 complex according to NMR spectra.

    Fig. S5. Docking simulation and biological mapping supporting a structural model of the triterpenoid lead compound–HR2 complex.

    Fig. S6. Characterization of the affinity of triterpenoid compounds Y19 and Y20 to HIV HR2 and HR1 and their effect on HR1-HR2 interactions.

    Fig. S7. Identification of HR2 in influenza HA2 as the domain targeted by the triterpenoid leads.

    Fig. S8. Production and characterization of the HR2 peptide (KIDQIIHDF)–specific polyclonal antibody.

    Fig. S9. The structure-activity relationship of triterpenoids against viruses according to our study.

    Table S1. Broad antiviral spectra of the tested compounds against various EBOV subtypes and MARV.

    Movie S1. The EBOV-host membrane fusion via a trimer-of-hairpins.

    Movie S2. Pentacyclic triterpene lead compounds inhibit virus-host membrane fusion by targeting the HR2 of virus envelope protein.

  • Supplementary Materials

    The PDF file includes:

    • Supplementary Materials and Methods
    • Fig. S1. Discovery of pentacyclic triterpene glycoconjugates as EBOV entry inhibitors.
    • Fig. S2. Time-of-addition experiments to clarify the stage at which lead compounds blocked EBOV entry.
    • Fig. S3. Characterization of the affinity of triterpenoid compounds to N-terminal HR1, C-terminal HR2, and their effect on HR1-HR2 interaction.
    • Fig. S4. Characterization of the spatial interactions within the triterpenoid-HR2 complex according to NMR spectra.
    • Fig. S5. Docking simulation and biological mapping supporting a structural model of the triterpenoid lead compound–HR2 complex.
    • Fig. S6. Characterization of the affinity of triterpenoid compounds Y19 and Y20 to HIV HR2 and HR1 and their effect on HR1-HR2 interactions.
    • Fig. S7. Identification of HR2 in influenza HA2 as the domain targeted by the triterpenoid leads.
    • Fig. S8. Production and characterization of the HR2 peptide (KIDQIIHDF)–specific polyclonal antibody.
    • Fig. S9. The structure-activity relationship of triterpenoids against viruses according to our study.
    • Table S1. Broad antiviral spectra of the tested compounds against various EBOV subtypes and MARV.

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.avi format). The EBOV-host membrane fusion via a trimer-of-hairpins.
    • Movie S2 (.avi format). Pentacyclic triterpene lead compounds inhibit virus-host membrane fusion by targeting the HR2 of virus envelope protein.

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article