Computationally Designed Epitope-Mediated Imprinted Polymers versus Conventional Epitope Imprints for the Detection of Human Adenovirus in Water and Human Serum Samples

Detection of pathogenic viruses for point-of-care applications has attracted great attention since the COVID-19 pandemic. Current virus diagnostic tools are laborious and expensive, while requiring medically trained staff. Although user-friendly and cost-effective biosensors are utilized for virus detection, many of them rely on recognition elements that suffer major drawbacks. Herein, computationally designed epitope-imprinted polymers (eIPs) are conjugated with a portable piezoelectric sensing platform to establish a sensitive and robust biosensor for the human pathogenic adenovirus (HAdV). The template epitope is selected from the knob part of the HAdV capsid, ensuring surface accessibility. Computational simulations are performed to evaluate the conformational stability of the selected epitope. Further, molecular dynamics simulations are executed to investigate the interactions between the epitope and the different functional monomers for the smart design of eIPs. The HAdV epitope is imprinted via the solid-phase synthesis method to produce eIPs using in silico-selected ingredients. The synthetic receptors show a remarkable detection sensitivity (LOD: 102 pfu mL–1) and affinity (dissociation constant (Kd): 6.48 × 10–12 M) for HAdV. Moreover, the computational eIPs lead to around twofold improved binding behavior than the eIPs synthesized with a well-established conventional recipe. The proposed computational strategy holds enormous potential for the intelligent design of ultrasensitive imprinted polymer binders.


Computational evaluations
The adenovirus epitope was taken from the crystal structure of HAdV fibre knob protein (PDB: 6HCN), located in the capsid with a residue sequence of AKLTLVLTKCGSQILATVSVLA (419-440) adopting a β -hairpin structure.The terminal ends are methylated.The protein was solvated in a water box and neutralized with two Cl -ions.All simulations were performed on GPUs with a 2 fs time step under periodic boundary conditions with the particle-mesh-Ewald method for electrostatic interactions, a cutoff of 12 Å for the van der Waals interaction and hydrogens constrained with the SHAKE algorithm.The setup was minimized for 40 ps, heated from 0 to 100 K (NVT) in 60 ps and then from 100 to 300 K (NPT) in 80 ps and finally pre-equilibrated at 300 K (NPT) for 120 ps.From the last pre-equilibrated structure, 1000 ns MD simulations were performed at 300 K (NVT).For functional monomer selection, the epitope structure resulting from equilibration was randomly surrounded by 10 copies of each of the functional monomer (acrylic acid, methacrylic acid, 4(5)-vinyl imidazole, acrylamide, methacrylamide) while solvated and neutralized.The H-bonds and salt bridges formed between the epitope and the monomer were monitored for performance evaluation.Two functional monomers demonstrating the highest number of bonds were further simulated with the epitope in combination while the epitope total monomer ratio remained as 1:20.

Adenovirus epitope synthesis and characterization
To best of our knowledge, adenovirus epitope was synthesized in a research lab for the first time in this work using solid-phase peptide synthesis technique by following two major steps: (i) loading and (ii) coupling of Fmoc/tBuprotected amino acids 1 .
(ii)Coupling of Fmoc/ t Bu-protected amino acids: To 1 gram of the resin (≈0.3 mmol g -1 ), 4 equivalents of amino acid, 4 equivalents of TBTU (O-(benzotriazole-1-yl)-N,N,N′,N′-tetramethyluronium tetrafluoroborate) and 10 equivalents of DIPEA in DMF was added and shaken for 30 min.The adenovirus peptide with different amino acid sequences were synthesized using this protocol with an alternating sequence of Fmoc-deprotections and amino acid couplings.To orthogonally synthesize linear adenovirus peptide with aspartic acid modifications, the Boc-Ala-OH at N-terminal was used instead of Fmoc-Ala-OH and the Fmoc-Lys(Alloc)-OH was used instead of Fmoc-Lys(Ac)-OH.After removal of the Alloc group in the presence of Pd(PPh3)4 and phenylsilane, the Fmoc-Asp(tBu)-OH was coupled on the side chain.
After removal of the last Fmoc group, the resin was transferred to a 5 mL syringe with frit and cap.After addition of the cleavage cocktail (trifluoroacetic acid (TFA), H2O, triethylsilane (TES), 95:2.5:2.5), the syringe was shaken for 1 h.The peptide was precipitated in ice cold diethyl ether and centrifuged.The supernatant was removed, and the precipitate was washed with diethyl ether twice.The peptide was resolved in MeCN/H2O (1:4) and lyophilized.Crude peptide variants were purified by reversed-phase high performance liquid chromatography (HPLC).Collected fractions containing the products were pooled and lyophilized to afford the adenovirus peptides as powder solid.

Preparation of adenovirus samples
Human adenovirus serotype 5 was grown on HEK-293 (ATCC CRL-1573) cells and harvested 3 days after infection.Cells were lysed through 3 repeated freeze-thaw cycles, virus particles were purified by 2 rounds of cesium chloride (Merck, Darmstadt, Germany) density gradient centrifugation as previously described 2 .Cesium chloride was removed by gel filtration on Sephadex G-25 (Pharmacia, Uppsala, Sweden) equilibrated with injection buffer (3 mM KCl, 1 mM MgCl2, phosphate buffered saline (PBS) and 10% glycerol).Subsequently, concentrated virus particle suspensions were passed through a 0.45 µm filter and stored in aliquots at -80 °C for further use.Virus titers were determined by titration of 10fold serial dilutions on HEK-293 cells.Additionally, particles counts were assessed by spectrophotometry as described previously 3 .Concentrated particle suspensions contained 2-5 × 10 11 particles per ml.Prior to further use, virus stocks were diluted to a final particle concentration of 1x10 9 plaque forming unit (PFU) mL -1 and fully inactivated by the addition of 0.5 % glutaraldehyde (Applichem, Darmstadt, Germany) in molecular grade water (Thermo Scientific, Waltham, MA, USA) and incubation for >48 hours at 8 °C.

Preparation of control viruses
Equine arteritis virus (EAV) and Herpes simplex virus type 1 (HSV-1) were propagated on RK13 [ATCC CCL-37] and Vero [ATCC CCL-81] cells, respectively, using Dulbecco's Modified Eagle's Medium (DMEM) (PAN-Biotech, Germany) supplemented with 1% Penicillin/Streptomycin. Ultra-filtration of both viruses was carried out using Vivaspin ® 2 Centrifugal Concentrator (MWCO 50 KDa; Sartorius, Stonehouse, UK) according to manufacturer's instructions. 4Briefly, cellular debris was removed by centrifugation for 5 min, 10000 g at 4 °C.Two steps of 2 ml virus stock centrifugation at 4°C (4000 g) for 20 minutes followed by 2 times washing steps with PBS were conducted using Vivaspin concentrator columns.A final volume of approximately 200 μL virus concentrate was collected and titrated on the respective cell line for each virus.Final concentration of 1x10 9 PFU mL -1 for HSV-1 and 1.3 × 10 8 PFU mL -1 for EAV were achieved, prior to further use virus was inactivated by the addition of 25 % Glutaraldehyde to a final concentration of 0.5 % and incubation for >48 hours at 8 °C.
Severe acute respiratory coronavirus 2 (SARS-CoV-2) particles were obtained from passage 3 of an early 2020 SARS-CoV-2 B.1 outbreak isolate (BetaCoV/ Germany/ BavPat1/2020) propagated on Vero E6 cells.All work handling infectious SARS-CoV-2 was performed under appropriate BSL-3 safety conditions (Freie Universität Berlin, Institut für Virologie).VeroE6 cells (ATCC CRL-1586) were cultured in minimal essential medium (MEM; PAN Biotech, Aidenbach, Germany) supplemented with 10% fetal bovine serum (PAN Biotech, Aidenbach, Germany), 100 IU mL -1 penicillin G and 100 µg mL -1 streptomycin (Carl Roth, Karlsruhe, Germany).Virus particles were obtained 48 hours post infection of confluent Vero E6 cells.For virus harvest, cell culture flasks were frozen at -80°C and thawed to obtain crude lysate which was centrifuged for 10 min at 4°C to remove cellular debris and titrated to obtain virus titers on Vero E6 cells as previously described 2 .Virus containing supernatant was subsequently inactivated by 24 hours of incubation in 4% neutral buffered formaldehyde at room temperature and exported from the BSL-3 lab for further purification by ultracentrifugation using a sucrose gradient as previously described 5 .

Surface plasmon resonance-based affinity evaluation
Gold surface of sensor chip (Cytiva, Germany) was modified with 2mM 11-Mercaptoundecanoic acid solution and conjugated with HAdV-specific computational eIPs (2µL min -1 , 30 minutes) following the 4 minutes activation of surface with 0.2 M EDC and 0.05 M NHS mixture.The unoccupied sites on the sensor were blocked with 100 µg mL -1 bovine serum albumin (BSA) and 0.1 mM ethanolamine treatment.HAdV samples were prepared in 10 mM phosphate buffer saline with 0.05% Tween for a concentration range of 10 3 -10 7 pfu mL -1 and injected to sensor for 9 minutes with 120 s of dissociation step.The resulting sensogram was evaluated with Biacore X100 evaluation software for steady-state affinity calculation.

Electrochemical characterization
Cyclic voltammetry (CV) and square-wave voltammetry (SWV) techniques were employed for characterization of eIP conjugated sensor and consecutive HAdV binding event.The three-electrode setup including Pt counter electrode, Ag/AgCl reference electrode, and gold working electrode was utilized with PalmSens4 potentiostat (Belltec, Lüdenscheid, Germany).A potential range of −0.2 to 0.8 V with a scan rate of 0.05 V s -1 was applied for CV measurements.SWV measurements were taken with a range of applied potentials of −0.3 to 0.8 V at an amplitude of 0.05 V and a frequency of 5 or 10 Hz.Gold electrodes were cleaned prior to each experiment following a previously reported method 6 .All measurements were taken in the presence of redox probe solution (10 mM K3Fe(CN6) in 0.1 M KCl) at room temperature.

FigureFigure
Figure S2.H-bonds formed between acrylic acid and the epitope during molecular dynamics simulation.

Figure
Figure S5.H-bonds formed between acrylamide and the epitope during molecular dynamics simulation.

Figure
Figure S6.H-bonds formed between methacrylamide and the epitope during molecular dynamics simulation.

Figure S7 .
Figure S7.Number of contacts for A) acrylic acid and B) methacrylic acid during molecular dynamics simulations.Salt bridges between acrylic acid and epitope's C) lysine and D) serine residues.

Figure
Figure S8.A) Chemical structure of aspartic acid modified (shown in blue) adenovirus epitope.B) Gradient system for the peptide with their respective retention time and ESI-HR-MS data.UV-detection at λ = 210 nm; buffer A: 0.1% HCOOH in H2O; buffer B: 0.1% HCOOH in MeCN.Size distribution of eIPs C) by intensity, and D) by number (n=3).E) Average zeta potential of eIPs.
Figure S9.FTIR spectrum of in-silico designed eIP.