Sulfoglycodendrimer Therapeutics for HIV-1 and SARS-CoV-2

Abstract

Hexavalent sulfoglycodendrimers (SGDs) are synthesized as mimics of host cell heparan sulfate proteoglycans (HSPGs) to inhibit the early stages in viral binding/entry of HIV-1 and SARS-CoV-2. Using an HIV neutralization assay, the most promising of the seven candidates are found to have sub-micromolar anti-HIV activities. Molecular dynamics simulations are separately implemented to investigate how/where the SGDs interacted with both pathogens. The simulations revealed that the SGDs: 1) develop multivalent binding with polybasic regions within and outside of the V3 loop on glycoprotein 120 (gp120) for HIV-1, and consecutively bind with multiple gp120 subunits, and 2) interact with basic amino acids in both the angiotensin-converting enzyme 2 (ACE2) and HSPG binding regions of the Receptor Binding Domain (RBD) from SARS-CoV-2. These results illustrate the considerable potential of SGDs as inhibitors in viral binding/entry of both HIV-1 and SARS-CoV-2 pathogens, leading the way for further development of this class of molecules as broad-spectrum antiviral agents.

Keywords: HIV‐1; SARS‐CoV‐2 receptor binding domain; glycodendrimers; molecular dynamics.

Figure 1

A) Synthesis of seven SGDs containing common mono, di, and trisaccharides. a) TsCl, pyridine, CH₂Cl₂, overnight, room temperature. b) K₂CO₃, acetonitrile, reflux, 48 h. c) TFA, CH₂Cl₂, room temperature, 2 h. d) TBTU, DIPEA, DMSO, pH = 9, overnight. e) SO₃‐pyridine, DMF, 0 °C, 1 h. B) Summary of structural data for Compounds 12d,e–18d,e.

Figure 2

Summary of all biological data for SGDs, Compounds 12e–18e. General notations: NA, Assay not conducted. Colors: Green‐Significant activity, Gray‐No significant activity, Yellow‐Positive assay controls. ^aELISA assays report the average IC₅₀ value obtained for a minimum of two assays for each compound and evaluated concentrations of 0–400µg mL⁻¹ in duplicate wells. ^bMST K_d values were obtained from triplicate measurements of serial dilutions (0.01–50 000 nmol) of the SGDs evaluated. ^cEach sample concentration was tested in duplicate. For SGDs 12e–18e, the presence of a greater than (>) sign indicates that no reduction in the luciferase signal was observed at any concentration tested (0.02–50 µg mL⁻¹) compared to no test sample. For DS and CHO1‐31, sample concentrations ranged from 0.01–25 µg mL⁻¹. ^dFor the cytotoxicity assessment, 12e–18e and DS were tested for reduction of relative luminescence units (RLU) after 48 h as compared to a no compound control. The concentrations presented are as follows: 0–50 µg mL⁻¹ for 14e, 16e and 17e and 0–25 µg mL⁻¹ for 12e, 13e, 15e, 18e, and DS, respectively. No measurable toxicity was observed for 12e–18e as evidenced by the > sign in front of the highest concentration tested, while for DS, a 45% drop in the living cell population was observed at 50 nm concentration, as measured by the presence of ATP.

Figure 3

MST summary. Left panel: Diagram illustrating the general MST experiment process. At the top, a quartz capillary loaded with sample is heated with a laser focused on a narrow region of the capillary tube. Upper left quadrant: The initial random distribution of fluorescently‐tagged protein in the bound and unbound states with the glycodendrimer at equilibrium. Upper right quadrant: Once the laser is turned on, the temperature gradient will cause the bound and unbound proteins to migrate out of the heated area at different rates. Lower left quadrant: This depicts the heated region of the capillary after the differential migration of the bound/unbound proteins. Lower right quadrant: The molecules return to a random distribution in the capillary after the laser is turned off. Right panel: MST results for selected SGDs plus positive control DS. The labeled‐gp120 concentration is 200 nm for all trials, and the y‐axis represents the MST responses at the 5 s mark after heating.

Figure 4

Different SGDs attached to one gp120 trimer: A) Cellobiose‐SGD (12e); B) Lactose‐SGD (15e); C) Maltotriose‐SGD (17e); D) Melibiose‐SGD (18e). Only two SGDs with the most favorable binding configurations are shown in each case. Proteins are shown as a white surface, SGDs are represented in atomistic details, and the gp120 residues interacting with SGDs (within 3 Å) are shown in licorice. The coloring scheme: V3 loop–purple; SGD atoms: C–cyan, O–red, S–yellow, N–blue, H–omitted; basic residue–blue, acidic residue–red, polar residue–green, nonpolar residue–white. Water and ions are omitted for better visualization. E) Percentage of different interactions contributing to the SGD‐gp120 binding (left bar: quantified over the whole protein; right bar: quantified only with V3 loop. Polar–white: interaction with polar amino acids; hydrophobic–red: interaction with hydrophobic amino acids; charge–blue: interaction with positively charged amino acids.

Figure 5

Lactose SGD (15e) binding with RBD of SARS‐CoV‐2. A) Two views (90° rotation) of 15e binding with the ACE2 binding region. B) Two views (90° rotation) of 15e binding with the purported HSPG binding region. Coloring scheme: SGD: C‐cyan, O‐red, S‐yellow, N‐blue, H‐omitted, SARS‐CoV‐2 interacting amino acids: Arg‐pink, Lys‐grey. Water and ions are omitted for better visualization. C) Percentage of different interactions contributing to the binding to the RBD of SARS‐CoV‐2 averaged over all the binding modes. Polar–white: interaction with polar amino acids; hydrophobic–red: interaction with hydrophobic amino acids; charge–blue: interaction with positively charged amino acids. C‐cellobiose (12e); L‐lactose (15e). D) Free energy of binding for GDs (12d, 15d) and SGDs (12e, 15e) with RBD of SARS‐CoV‐2 (two trials for each system). Top: refers to the initial placement of the GD/SGD near the ACE2 binding region. Mid: Refers to the initial placement of the GD/SGD near the HSPG binding region.