Structural Analysis of the H3 Protein in Monkeypox Virus

(A) Phylogenetic tree depicts the evolutionary relationships within the Poxviridae family, highlighting MPXV, (blue circle), VARV, (red circle), and VACV, (green circle). (B) Schematic shows MPXV adhesion to a cell, with H3 and fusion complexes depicted. (C) The amino acid sequence and structure of MPXV H3 displays β-strands (yellow arrows) and α-helices (pink cylinders). Mg(II) binding sites (green) and potential HS binding motifs (blue underlines) are shown. The novel three-helical domain (240-282), is highlighted in yellow. (D) AlphaFold2 prediction of MPXV H3 Structure shownin two orientations. Regions matching VACV H3’s known structure are in blue; novel elements predicted are in yellow, with all potential GAG-binding motifs highlighted.

Molecular Dynamics and Docking Analyses of H3-HS Interactions

(A-B) Cartoons show docking results of heparan sulfate (HS) with H3 motifs 1, 2, 3, and the helical domain, respectively. Panels (B) show RMSD values from 1µs MD simulations, color-coded to match the configurations in (A). (C) Schematics illustrates the reaction coordinate in umbrella sampling, highlighting HS dissociation from H3 with a green force curve. (D) Presents binding free energies for HS-H3 interactions in motifs 1, 2, 3, and the helical region, color-coded respectively. (E) A free energy landscape map from a 1000 ns REMD simulation shows HS binding configurations to the helical and Mg(Ⅱ) regions. (F) Panel provides salt bridge formation statistics between HS and H3’s basic amino acids, with a bar chart of average formations. (G) A surface plot shows the frequency of salt bridge formations within H3, with areas of frequent formations in blue. (H) Detailed views of HS-H3 interactions, with the left image showing salt bridges and the right image displaying electrostatic interactions with Mg(Ⅱ). This panel illustrates the impact on HS binding stability to H3 following the removal of Mg(Ⅱ) during the simulation. (J) The effects of mutating all basic amino acids in the helical domain on the binding stability of HS are shown. (K) The “palm-binding” model is depicted where HS is secured by the helical “fingers” and interacts with the Mg(Ⅱ) “palm.”

Charge Characteristics and Structural Analysis of H3 Protein

(A) The heatmap shows the amino acid charge distribution in the H3 protein across 66 Poxviridae viruses, following multiple sequence alignment. Blue indicates areas with more positive charges, and red indicates more negative charges. The accompanying curve shows the average charge of all amino acids in Poxviridae H3. (B) Logo plot of the amino acid sequence of the helical region, highlighting the conservation of basic amino acids at specific positions. (C) The surface charge analysis of H3 from the MPXV, VACV, VARV, and cowpox viruses (CPXV) with the helical domain showing a significantly positive charge. (D) Schematic of the single-molecule force spectroscopy unfolding experiment on H3, illustrating the unfolding process of the helical region (yellow) followed by the main body (blue). (E) Representative curves of H3 unfolding, color-coded to show the helical region (yellow), main body (blue), and full-length (purple) unfolding. (F) Histograms depict the force spectroscopy signals for helical domain, main body, and full-length unfolding, with ΔLc statistics provided. The inset shows a Gaussian fit of unfolding forces.

Analysis of the Helical Domain’s Interaction with HS at Cellular Level

(A) Schematic of the cell force spectroscopy experiment setup shows three scenarios: wild-type H3 on an AFM tip interacting with HS on CHO-K1 cells, mutation of all basic amino acids in the H3 helical region to serine, and cells treated with HS hydrolase to remove surface HS before testing. (B) Force-extension curves depict interactions for the wild type, mutant, and control groups, marked with blue and red asterisks for dissociation events. An inset shows the optical microscope positioning the AFM probe. (C) Histograms of dissociation signals comparing the wild type, mutant, and control groups, with an inset detailing the surface distribution of dissociation forces. (D) Statistical graph showing binding probabilities for different groups, highlighting significant differences determined by t-tests (p<1e-5). (E) Flow cytometry results illustrate interactions of wild-type (WT) and Uncharged H3 fused with eGFP with CHO-K1 cells, alongside surface HS removed control (green) and cell-only control (Blank, grey). (F) Statistical analysis of flow cytometry data, showing significant differences between groups as determined by t-tests. *****, P<1e-5

Development and Testing of Protein Inhibitors Targeting the H3 Helical Domain

(A) Diagrams depict protein inhibitors designed to target the H3 helical region, created using RFdiffusion. Sequences capable of folding into the target scaffold structures were generated using ProteinMPNN, and were validated through AlphaFold2, followed by 500 ns MD simulations for structural stability and interaction scoring. (B) FCM analysis demonstrates the inhibitory effect of AI-Poxblock723 at various concentrations. (C) BLI confirms direct interaction between AI-Poxblock723 and the H3 helical domain. (D) Graphs display the inhibitory effect of the indicated AI-Poxblocks on MPXV infection of Vero E6 cells, with quantitative analysis of virus-infected foci provided on the right.

Structural Prediction of Monkeypox Virus H3 Protein by AlphaFold2.

In the protein cartoon diagram, the colors represent the pLDDT scores output by AlphaFold2, with blue indicating a higher prediction confidence. The helical domain is displayed in blue, indicating a region where the pLDDT score is greater than 80, signifying a high confidence in structural prediction.

Structural Formula of HS.

The composition of HS used in the main text consists of -[IdoA2S-GlcNS6S-IdoA-GlcNS(3,6S)]5-, containing a total of 20 repeating monosaccharide units.

Docking Results of H3 with HS.

The docking area of HS was confined near different motifs by setting the docking box. The figure displays the docking results for four motifs, with the top four docking outcomes in each motif area selected based on the best scores (kCal/mol) from AutoDock Vina for subsequent MD simulations.

Umbrella Sampling Calculation of the Binding Free Energy of HS-H3.

The conformation of HS binding simultaneously to the helical domain and Mg2+, obtained from replica exchange simulations, was further subjected to umbrella sampling to calculate the binding free energy. The stretching process is illustrated in (A), with the distribution of the reaction coordinate shown in (B). (C) presents the PMF (Potential of Mean Force) profile obtained after calculations using the Weighted Histogram Analysis Method (WHAM).

Evolution of Salt Bridge Formation in HS During MD Simulations.

This represents the variation in the number of salt bridges formed between HS and all basic amino acid side chain of H3 as the simulation progresses. The lighter areas indicate the formation of a greater number of salt bridges.

AlphaFold2 Structural Prediction of 66 Poxviridae Virus H3 Proteins.

The yellow portion represents the helical domain, while the blue portion signifies the main body of H3.

Analysis Results of Full Sequence Analysis for 66 Poxviridae Virus H3 Proteins.

The logo plot illustrates the probability of amino acid occurrence within the H3 sequences, where larger font sizes indicate a higher occurrence probability, denoting greater conservation. Acidic amino acids, basic amino acids, polar amino acids, and non-polar amino acids are represented in red, blue, green, and black, respectively.

Protein Immobilization in AFM-SMFS Experiments.

Peptide GL-ELP20-Cys is modified on a glass substrate (or AFM tip) that has been functionalized with maleimide through a Michael addition reaction. The N-terminus GL can undergo an enzymatic ligation reaction with the C-terminus NGL of the target protein in the presence of ligase OaAEP1, thereby immobilizing the target protein on the substrate or AFM tip.

Artificial Design of H3 Inhibitors.

The gray structure represents the protein backbone structure designed by RFdiffusion, while the purple structure is the sequence generated by ProteinMPNN based on that backbone, followed by complex prediction results from AlphaFold2. The two structures align well, with an RMSD less than 1.5 nm.

AlphaFold2-predicted structures of the H3 and inhibitor complexes.

From left to right are AI-Poxblock302, AI-Poxblock602, AI-Poxblock614, AI-Poxblock723, and AI-Poxblock761. All inhibitors are depicted in purple, and the helical domain of H3 are marked in yellow.

MD simulations of H3-Inhibitor complex.

The H3 and Inhibitor complex underwent a 500 ns MD simulation, resulting in the variation of RMSD (nm) over time (ns).

Flow cytometry analysis demonstrates the efficacy of the inhibitors.

The graph shows relative fluorescence intensity, indicating that in the presence of 10 μM AI-Poxblock723, the binding of H3(1-282)-eGFP to CHO-K1 cells is reduced compared to the control group (gray).

AI-Poxblock723 and H3(1-239) BLI experiment.

BLI curves of three concentrations of H3(1-239) (20000 nM, 10000 nM, and 2500 nM) with AI-Poxblock723 loaded on the sensor. The curves could not be fitted to obtain the kD value.

CD spectroscopy experiment of AI-Poxblock723.

CD spectroscopy verifies the designed inhibitor with a α-helical structure.

Sequence comparison of the H3 helical domain among representative orthopoxvirus strains MPXV, VACV, VARV, and CPXV.

Basic amino acid residues are highlighted in blue, and differences relative to the MPXV sequence are shaded in yellow.

The inhibitory effect of indicated inhibitors on VACV infection of Vero E6 cells.

Virus-infected foci and counts were shown on the left.