Research Article www.small-journal.com Graphene Oxide Nanosheets Interact and Interfere with SARS-CoV-2 Surface Proteins and Cell Receptors to Inhibit Infectivity Mehmet Altay Unal, Fatma Bayrakdar, Hasan Nazir, Omur Besbinar, Cansu Gurcan, Neus Lozano, Luis M. Arellano, Süleyman Yalcin, Oguzhan Panatli, Dogantan Celik, Damla Alkaya, Aydan Agan, Laura Fusco, Serap Suzuk Yildiz, Lucia Gemma Delogu, Kamil Can Akcali, Kostas Kostarelos,* and Açelya Yilmazer* 1. Introduction Nanotechnology can offer a number of options against coronavirus disease 2019 (COVID-19) acting both extracellularly and intracellularly to the host SARS-CoV-2 has spread all over the world, over 192 countries with more than cells. Here, the aim is to explore graphene oxide (GO), the most studied 2D 146 million confirmed cases and 3.1 mil- nanomaterial in biomedical applications, as a nanoscale platform for interac- lion deaths.[1] Even though there are tion with SARS-CoV-2. Molecular docking analyses of GO sheets on interac- numerous preclinical and clinical studies tion with three different structures: SARS-CoV-2 viral spike (open state – 6VYB ongoing, unfortunately, there is cur- or closed state – 6VXX), ACE2 (1R42), and the ACE2-bound spike complex rently no specific preventive or curative treatment for coronavirus disease 2019 (6M0J) are performed. GO shows high affinity for the surface of all three struc- (COVID-19).[2,3] As discussed in various tures (6M0J, 6VYB and 6VXX). When binding affinities and involved bonding reports, nanotechnology and nanoma- types are compared, GO interacts more strongly with the spike or ACE2, com- terials could provide solutions for the pared to 6M0J. Infection experiments using infectious viral particles from four fight against COVID-19 in different ways, different clades as classified by Global Initiative on Sharing all Influenza Data including: i) detection of viral particles; ii) (GISAID), are performed for validation purposes. Thin, biological-grade GO protection of human exposure from viral particles via personal protection equip- nanoscale (few hundred nanometers in lateral dimension) sheets are able to ment; iii) inactivation of viral particles by significantly reduce copies for three different viral clades. This data has dem- capture onto surfaces or different envi- onstrated that GO sheets have the capacity to interact with SARS-CoV-2 sur- ronments; and iv) inactivation of viral face components and disrupt infectivity even in the presence of any mutations particles through molecular surface inter- on the viral spike. GO nanosheets are proposed to be further explored as a actions with novel therapeutic drug mole- cules (e.g., small molecules or antibodies) nanoscale platform for development of antiviral strategies against COVID-19. and vaccines.[4] Dr. M. A. Unal, O. Besbinar, C. Gurcan, D. Celik, D. Alkaya, O. Besbinar, C. Gurcan, O. Panatli, D. Celik, D. Alkaya, A. Agan, Prof. K. C. Akcali, Dr. A. Yilmazer Dr. A. Yilmazer Stem Cell Institute Department of Biomedical Engineering Ankara University Ankara University Balgat, Ankara 06520, Turkey Golbasi, Ankara 06830, Turkey Dr. F. Bayrakdar, S. Yalcin, Dr. S. Suzuk Yildiz E-mail: ayilmazer@ankara.edu.tr Ministry of Health General Directorate of Public Health Dr. N. Lozano, Dr. L. M. Arellano, Prof. K. Kostarelos Microbiology References Laboratory Catalan Institute of Nanoscience and Nanotechnology (ICN2) Sihhiye, Ankara 06430, Turkey UAB Campus Bellaterra Dr. H. Nazir Barcelona 08193, Spain Department of Chemistry E-mail: kostas.kostarelos@icn2.cat Ankara University Dr. L. Fusco, Dr. L. G. Delogu Tandogan, Ankara 06100, Turkey Department of Biomedical Sciences The ORCID identification number(s) for the author(s) of this article University of Padua can be found under https://doi.org/10.1002/smll.202101483. Padua 35122, Italy © 2021 The Authors. Small published by Wiley-VCH GmbH. This is an Prof. K. C. Akcali open access article under the terms of the Creative Commons Attribution Department of Biophysics License, which permits use, distribution and reproduction in any medium, Faculty of Medicine provided the original work is properly cited. Ankara University Sihhiye, Ankara 06230, Turkey DOI: 10.1002/smll.202101483 Small 2021, 17, 2101483 2101483 (1 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com Since the early phases of the pandemic, computational (e.g., MoS2 and WSe2), transition metal carbides (MXenes, models have been used to screen for drug molecules previ- e.g., Ti3C2), hexagonal boron nitride (hBN), and graphitic ously approved for other pathologies and disease conditions carbon nitride (g-C3N4) have been developed.[28–31] Due to their to be potentially repurposed against COVID-19.[5,6] Via these unique physicochemical properties including optical and elec- approaches, drugs such as remdesivir or toremifene have been tronic properties, along with their diverse chemical composi- reported to show high binding affinities against SARS-CoV-2 tion and physicochemical properties including their good proteins, including the spike protein or main protease.[7–10] Fol- biocompatibility, these nanomaterials have attracted attention lowing these in silico and further pre-clinical evaluations, these in various disciplines, including biomedical sciences.[32,33] drugs have been moved toward clinical trials, and remdesivir Currently, applications such as anticancer therapeutics, mul- has recently been approved by the Food and Drug Administra- timodal bioimaging, cancer theranostics, biosensing, tissue tion for the treatment of COVID-19 requiring hospitalization.[11] engineering, and antimicrobial coatings have been proposed During these computational simulations, different approaches and studied. Out of these 2D materials, graphene oxide (GO) is can be used: Docking analysis is one of those, used to illumi- the most extensively explored in biomedicine. Various groups nate the interaction processes of molecules that are spatially have shown how GO sheets interact with biological systems interacting to each other. The most critical information to be and fully characterize their toxicological, biodegradability, obtained from this process is to identify the ligand(s) that bind and tissue distribution profiles by detailed in vitro and in vivo to protein epitopes most efficiently and determine the energy experiments.[34–38] values of such binding interactions. In light of this information, The current study aims to offer a combined in silico and in one can anticipate to reveal new target proteins and the ways in vitro analysis to interrogate whether GO sheets can be consid- which potential drug molecules could interact with them.[12,13] ered as a nanomaterial platform able to interact effectively with Molecular dynamics simulations (MDS) have mainly been used specific SARS-CoV-2 surface proteins and receptors leading for conformation analysis of proteins. Conformational change to an inhibitory action. Molecular docking experiments have processes take place in a short period of time (nanosecond to been performed to understand how GO interacts with the viral microsecond) and are impossible to follow experimentally. spike, the ACE2 cell receptor, and the spike-ACE2 complex and MDS enables us to visualize such experimentally challenging identify the binding parameters that govern these interactions. events. From such analysis, the aim has been to elucidate the Finally, a proof-of-concept study was performed in vitro (using role that amino acids play in ligand interactions and which Vero E6 cell cultures) to experimentally evaluate the effect of amino acids provide structural stability to the protein.[14,15] GO on the inhibition of wild-type SARS-CoV-2 infectivity. Four Several viral mechanisms can be shown as targets including different infectious viral clades, as classified by the Global Ini- the blockade of structural proteins responsible for entry into tiative on Sharing All Influenza Data (GISAID) platform, have human cells or inhibition of important viral enzymes respon- been used in order to better understand the inhibition of viral sible for genome replication or viral assembly.[2,3] Receptor infection in the presence of mutations in viral proteins. binding domains (RBD) of the viral spike play a critical role in SARS-CoV-2 and ACE2 fusion, so this is a very relevant target complex for viral inhibition.[16] As described above, various ther- 2. Results apeutic strategies have been suggested to inhibit this interac- tion. Among some of the clinically tested drugs, chloroquine/ It has been shown that 6M0J is the ACE2-bound form of SARS- hydroxychloroquine, arbidol, ribavirin, pensiclovir, favipiravir, CoV-2 spike protein. According to the UniProt (The Universal nafamostat, nitazoxanide, and camostat mesylate have been Protein Resource) database (https://www.uniprot.org/uniprot/ shown to decrease virus–host cell binding or cellular internali- P0DTC2), the 437–508 receptor-binding motif area of the 319– zation or release.[17–22] Delineating the mechanisms of virus 541 amino acid receptor binding domain (RBD) of spike pro- and cell interactions and viral life cycle, is not only important teins binds to the human-ACE2 protein (Figure 1A). The amino for the advancement of antiviral therapeutics, but also for the acids 30–41, 82–84, and 353–357 of ACE2 proteins are involved development of protective technologies from these viruses. For in this interaction (https://www.uniprot.org/uniprot/Q9BYF1). this reason, antiviral agents which show higher binding affinity Examining the affinity of GO for the 6M0J could allow us to and neutralization activity of viral particles have been included examine its effect on both the spike and ACE2 binding. The in the fabrication of personal protective equipment, including docking calculation results of GO and 6M0J are shown in face masks. Copper, zinc, and polyethylenimine (PEI) are Figure 2. Considering the docking results of GO with the 6M0J, among such agents with antiviral activity that have been previ- GO showed affinity in 4 regions of the spike-ACE2 pair between ously reported to increase inactivation efficiency in masks by values −9.1 kcal mol−1 to −8.4 kcal mol−1. However, three of these improved viral entrapment without any toxicity to the user.[23–27] binding conditions (1st, 2nd, and 4th) are particularly important Following the discovery of graphene, different classes of since they cover the entire SARS-CoV-2 infection process (spike, 2D materials including transition-metal dichalcogenides ACE2 and interface). When the highest binding affinity of GO- Spike-ACE2 docking (−9.1 kcal mol−1) was examined in detail, GO showed strong affinity against amino acids of the spike- Prof. K. Kostarelos RBD via 2 hydrophobic, 12 hydrogen, and 1 charge interaction Nanomedicine Lab with ASP427, a total of 15 bonds, and as well as 6 van der Waals National Graphene Institute and Faculty of Biology Medicine & Health The University of Manchester forces which all explained the strong interaction with the ACE2 AV Hill Building, Manchester M13 9PT, United Kingdom (Figure 3). On closer examination, (Table S2, Supporting Infor- E-mail: kostas.kostarelos@manchester.ac.uk mation) GO could act as a multiple H-donor and H-acceptor Small 2021, 17, 2101483 2101483 (2 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com Figure 1. A) Schematic representation of viral spike and ACE2. SARS-CoV-2 binds to host cell receptor ACE2 through spike protein. Receptor binding domain (RBD) plays an important role during this interaction. Created with BioRender.com. B) Characterisation of GO material. I) Height AFM image (dimension: 5×5; scale bar: 1 µm) with insert of height cross-section profile along the indicated region in the height AFM image; II) corresponding lateral dimension distribution analysis (number of analyzed flakes: 356); III) SEM micrograph (scale bar: 1 µm); and IV) corresponding lateral dimen- sion distribution (n = 116 analyzed flakes). with strong H-bonds ranging from 1.74 Å to 2.93 Å. Moreover, GO:6M0J docking results are also shown in Figure S2, Table S3, electrostatic-pi–anion interaction of 427 ASP spike residue with and Figure S3, Table S4, respectively). Overall, it could be con- GO is important for both increasing the binding strength and cluded that GO strongly bound on the surface of 6M0J. affecting the load density in the close chemical environment. In The 3-stranded glycoproteins 6VYB and 6VXX of SARS- addition, GO formed van der Waals bonds with the following CoV-2 are those binding to ACE2 and correspond to open and amino acid residues of ACE2: PRO321, MET383, ALA384, closed states of the SARS-CoV-2 spike, respectively. The interac- PHE55, ARG559, and ALA387. The 2nd and 4th positions of the tion of GO with these proteins is shown in Figure 4A,B and the Small 2021, 17, 2101483 2101483 (3 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com During a SARS-CoV-2 infection event, blocking of the interac- tion between the virus and host cells could be achieved by directly acting on the viral particles, the host cell receptors or the viral particles bound to the host receptor. To interrogate such pos- sible interactions, we performed docking simulations between GO and ACE. Figure 4C shows the docking results of GO with 1R42. When Figures 2 and 4C were evaluated comparatively, it was seen that the binding positions of GO and spike on the ACE2 do not overlap. However, when binding affinity values are taken into account for these docking analyses, the difference between them was 0.8 kcal mol−1, indicating that GO binds more strongly to ACE2. This difference between binding strengths comes from the excess and diversity of the bonds formed between the GO and the ACE2 (Figures S3b and S6b, Supporting Information). When the bonding interactions were compared (Tables S4 and S8, Supporting Information), 7 hydrogen bonds and 2 hydrophobic interactions are obtained between GO and ACE2 binding (6M0J), while 12 hydrogen bonds, 2 hydrophobic, and 1 electrostatic inter- action are calculated for ACE2. Therefore, it can be concluded that GO bound more strongly to the ACE2. Based on the in silico investigations above, GO had been shown to have a strong affinity toward both the viral spike and the ACE2. In order to interrogate the possible implications of such interactions experimentally, we performed in vitro tests using infectious, wild-type SARS-CoV-2 virions. The biological- grade GO material used in the studies has been developed for biomedical applications by our laboratories and are cur- rently extensively explored in the context of various labs and projects. The characterization data of the specific batch of GO materials used in these studies are described in Figure 1B; and Figure S1, Supporting Information, in agreement with our pre- vious reports.[39–43] The GO sheets used were thin (1–3 carbon layers) and of a few hundred nanometers in lateral dimension (100–400nm), with proven biocompatibility and purity. Prior to testing antiviral activity, the possible cytotoxic impact of GO exposure to Vero E6 cells has been evaluated. As can be seen from Figure S7, Supporting Information, material did not show any significant toxicity up to 100 µg mL−1 in vitro in Vero Figure 2. The top nine docking results of GO and 6M0J. The binding E6 cells. Four different viral genotypes that have been previ- regions of GO are numbered considering the binding affinity values. GO ously deposited in GISAID platform, all corresponding to a dif- colored with green represents the highest binding affinity. Binding affini- ties of these identified regions are given on the right. ferent clade, were used for viral infection experiments. During in vitro experimentation, two different protocols were followed: affinity values are given in Table S4, Supporting Information. As a pre-infection and a post-infection protocol. In the pre-infec- seen in Figure 4A,B, GO has the potential to interact with each tion protocol, Vero E6 cells were first treated with the GO at region of open and closed sate glycoproteins. The affinity values different concentrations, and after 2 h viral particles at MOI of of these interactions ranged between −10.5, −9.4 and −9.0, 0.01 were added into the cell culture media. During the post- −8.3 kcal mol−1, respectively (Table S5, Supporting Information). infection protocol, cells were first incubated with the infectious These high affinity values indicate that the bond strength should virions and GO at different concentrations were added later also be high. In addition, when the binding positions of GO in onto the infected cell culture. Clade GR showed the highest and glycoproteins were examined, this indicated also an effective significant inhibition of infection in response to GO treatment interaction, especially in relation to the RBD regions of the gly- (Figure 5A), reaching nearly a four-log reduction in viral copy coproteins. GO acted as a strong hydrogen donor and acceptor number following the pre-infection protocol. On the other hand, (Tables S6 and S7; Figure S4–S6, Supporting Information) S and GR showed significant inhibition only at the highest GO on interaction with both glycoproteins in the range of 2.214– dose of 100 µg mL−1 (Figure 5B and 5C). The clade named as 3.326 Å. In addition, GO has been shown to produce electro- “other” was the least responsive to GO-mediated inhibition, static interactions with both proteins (6VYB: A: LSY417; 6VXX: although still a significant reduction of viral copy number was B: LYS462, C: ARG273, B: GLU465, C: ASP88) (Figures S4d achieved at 100 µg mL−1 GO concentration (Figure 5D). Plaque- and S5d, Supporting Information), while the chemical surface forming assays were also performed in order to further validate character of GO can change the charge load density. the observed antiviral activity of GO against SARS-CoV-2 from Small 2021, 17, 2101483 2101483 (4 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com Figure 3. Detailed investigation of GO and 6M0J docking analysis. A) GO-6M0J docking result, binding affinity −9.1 kcal mol−1; B) 2D map of 6M0J amino acids bonding interactions with GO. Spike residues: TYR369, LYS378, CYS379, TYR380, GLY381, VAL382, SER383, ARG408, ALA411, PRO412, GLY413, GLN414, THR415, ASP427, ASP428, PHE429. ACE2 residues: PRO321, MET383, ALA384, ALA383, PHE555, ARG559; C) H-bonding: pink shows donors and green shows acceptors; D) Charge interaction with ASP427; E) Hydrophobicity (alkyl and pi-alkyl) interactions, LYS378. the GR clade. As can be seen in Figure 5E, treating the Vero E6 Finally, we went back to the in silico platform and have deline- cells with GO prior to viral infection with the GR clade resulted ated the effect of mutations found in the four different viral gen- in an inhibition profile from which the median inhibitory con- otypes used in the in vitro study. Molecular docking was applied centration (IC50) was found to be around 30 µg mL−1. Whereas to better explain the experimental results obtained in vitro. The for the post-infection protocol, a higher IC50 value (45 µg mL−1) mutation profile of all genotypes can be found in Table S9, Sup- was obtained (Figure 5F). Arbidol which has been shown to be porting Information. GO is expected to mainly interact with the an effective drug in disrupting viral uptake in Vero E6 cells[18] viral surface proteins and/or host cell surface receptors. For this has been also used as a “benchmark” therapeutic agent in our reason, while considering the mutations on the viral genome, study. When the Vero E6 cells were pretreated with Arbidol we mainly focused on the mutations of the spike protein. Both at different dilutions, inhibition of viral infection could be the GR and GH clades have a mutation at the spike structure, observed both via qRT-PCR and plaque assays (Figure S8, Sup- which is D614G. The GR clade showed more sensitivity against porting Infomration). This data supports that in both assays, the GO compared to the GH clade, pointing out the importance which have been also extensively used in literature,[20,21,44] the of mutations found in the downstream viral proteins. However, GO antiviral activity could also be demonstrated. When the pre- all of these mutations are on the non-structural genes in the infection and post-infection protocols were compared, it could GH clade and their impact on overall viral infectivity has not yet be seen that the inhibition level of viral infection was higher been studied enough. The S clade, on the other hand, has a dif- during the pre-infection protocol, as evident by both qRT-PCR ferent spike mutation known as G1251V which is not within the and plaque assays. This agreed with the in silico analysis, since receptor binding domain of the viral spike, and therefore is not GO had been shown to have higher affinity toward the spike or expected to play an important role in the spike recognition. The ACE2 (a situation mimicked by pre-infection) compared to the clade represented as “other” didn’t show any mutation on the spike-bound ACE2 (a situation mimicked by post-infection). viral spike, representing a native form of the spike protein. We Small 2021, 17, 2101483 2101483 (5 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com Figure 4. Docking of GO against open/closed state of Spike and IR42. A) GO with open state of spike (6VYB), 2nd affinity (highlighted in green) value corresponds to −10.4 kcal mol−1; B) GO with closed state of spike (6VXX), 4th and 7th affinity values correspond to −8.7 and −8.4 kcal mol−1, respec- tively. C) The top nine docking results of GO and 1R42 are given on the right. The binding regions of GO are numbered considering the binding affinity values. GO colored with green represents the highest binding affinity. have performed molecular docking of GO on interaction with 3. Discussion the mutated spike D614G, which is one of the mostly reported mutations causing increased infectivity of SARS-CoV-2 due Nanotechnology offers a number of possibilities for antiviral to its effect on the receptor binding domain of the spike pro- activity, both outside and inside the host cells. Several nano- tein. To examine the effect of the D614G mutation on the spike technology-based platforms have already been successful in protein, first, a structural prediction was made by replacing preclinical studies to counter a variety of human viral patho- the 614th amino acid Asp in the B chain of 6VXX and 6VYB gens such as HIV, human papilloma virus, herpes simplex, with Gly. After structural prediction, the structural compar- and respiratory viruses.[45–47] Moreover, nanoscale systems ison – RMSD values between wild type and mutant structures can potentially increase the effectiveness of drugs and other were calculated to be 7.869 Å and 7.919 Å for 6VXXwt-6VXXmu bioactive molecules by reducing the required effective dose, and 6VYBwt-6VYBmu, respectively. The structural changes therefore dramatically improve the therapeutically effective between the wild-type and the mutant viruses occurred in the drug toxicity thresholds. In the literature, there are numerous RBD regions of the spike protein (Figure 6A). This structural reports of antiviral drug delivery platforms suitable for different change altered the donor–acceptor properties of the amino acid viral diseases and targets. Such platforms include liposomes, chain (Figure 6B) and consequently led to an increased binding emulsions, de÷rimers, micelles, solid-lipid hydrogel based strength and interaction between the GO sheets and the spike nanocarriers, polymeric nanoparticle, carbon-based and 2D proteins (Figure 6C). materials.[48–50] Small 2021, 17, 2101483 2101483 (6 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com Figure 5. In vitro evaluation of GO mediated viral inhibition. Vero E6 cells were treated with GO (5, 10, 50, and 100 µg mL−1) and SARS-CoV-2 viral particles (MOI 0.1) according to a pre-infection or post-infection protocol. Four different viral genotypes were used which belong to A) GR, B) S, C) GH, and D) other clades according to GISAID. After 5 days, cell culture supernatants were used to quantify viral copy numbers by qRT-PCR. Vero E6 cells were treated with GO (5, 10, 50, and 100 µg mL−1) and SARS-CoV-2 viral particles (MOI 0.1) from clade GR according to E) pre-infection or F) post-infection protocol. % of inhibition was plotted following plaque assay. * p < 0.05, ***p < 0.001 compared to virus control. The field of nanotechnology has recently caught the atten- example, Baweja et al. reported that GO and reduced GO (rGO) tion of computational analysis in order to help scientists were able to inhibit the α-helix to β-sheet transition of amyloid offer better understanding of nanomaterial interactions with beta (Aβ) peptide, which has been implicated in the pathogen- biological systems.[4] Based on the data generated by such esis of Alzheimer’s disease.[55] In another study, Putri el al. in silico analyses, nanotechnology-based therapeutic approaches studied via computational analysis the thermally responsive could be improved.[51–53] Among 2D materials, GO is the most behavior of a polymer-GO complex in the design of a sensor studied one using such computational models to depict interac- with an “on/off” switch upon binding to a cancer cell marker tions with various biological moieties, mainly proteins.[54–61] For at its lower critical solution temperature.[61] In silico approaches Small 2021, 17, 2101483 2101483 (7 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com Figure 6. The effect of the D614G mutation on the interaction between GO and spike. A) Comparison of wild type and mutation structures of 6VXXwt- 6VXXmu and 6VYBwt-6VYBmu (colors – tawny-brown: wt; cyan: mutant) shows that the structural change occurred at the RBD. B) The D614G mutation effect on B wire of 6VXX and 6VYB are obtained. C) Molecular docking of GO against the mutated form of 6VXX and 6VYB was performed and binding affinities (ΔG kcal mol−1) of 6VYBmu and 6VXXmu were calculated. Small 2021, 17, 2101483 2101483 (8 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com are becoming valuable tools to better understand the thera- cellular nanosponges which display cellular receptors of the peutic potential of nanomaterial-drug conjugates, including 2D virus on the surface, to demonstrate that these nanosponges materials. were able to neutralize viral particles resulting in inhibition of Considering the urgent need to offer as many options as pos- infection.[74] In contrast, studies evaluating the antiviral activity sible in managing the SARS-CoV-2 pandemic, such computa- of 2D materials are very limited. In one of two studies to do tional methods are especially important to guide the rational so, Raval et al. reported a simple, initial computational anal- design of new nanoscale systems on interaction with the critical ysis showing interaction between graphene and the receptor- SARS-CoV-2 viral components. In our study, GO was shown binding domain of spike complexed with its receptor ACE2. The to have affinity towards the spike protein, ACE2 receptors and molecular simulation data using pristine multi-layer graphene spike-ACE2 complex. However, when the binding affinities reported interactions with SARS-CoV-2 proteins, but no experi- and types of bonds were compared, GO was found to interfere mental work was offered to validate the computational obser- more strongly to the viral spike (6VYB or 6VYB) and the ACE2 vations.[75] However, inhibition of viral infection can happen (1R42) before binding to the virus ligand, compared to 6M0J. not only at the spike – ACE2 complex, but also at the spike or Based on this in silico observation, when cells were exposed to ACE2. As can be seen in our study, comparing docking analysis GO prior to viral treatment in the context of a pre-infection pro- at different protein domains is crucial to better evaluate the tocol closer to a preventive clinical senario, pronounced viral inhibitory effect of nanomaterials. In the context of graphene inhibition was observed compared to a post-infection protocol or GO incorporated in personal protective equipment (PPE), De in vitro. In combination, in silico and in vitro analyses in this Maio et al. recently reported that GO could reduce SARS-CoV-2 study emphasized the importance of the correlation between infectivity in vitro.[76] However, the GO concentration range computational and experimental methodologies in evaluating selected is excessive and not realistic for the purposes of anti- the antiviral activity of nanoscale platforms suspended in physi- viral therapeutics. Furthermore, given that there are different ologically relevant aqueous solutions. viral clades spreading among patient populations, it was impor- 2D nanomaterials, due to their extremely large surface area, tant to evaluate the effectiveness of graphene or GO functional- can be superior carriers for antiviral drug delivery purposes ized PPEs against multiple infectious viral genotypes. compared to other materials with different structural confor- In our present study, it is shown that GO can lead to reduced mations and dimensions. 2D nanomaterials indeed have been SARS-CoV-2 infectivity in 3 out of 4 infectious viral clades explored in various preclinical studies for the delivery of cyto- tested. Differences in viral inhibition among viral clades could toxic agents, such as MTX, DOX or 5-FU (among others) to be attributed to the mutations found in the viral genotypes. cancer cells for chemotherapeutic applications.[62–64] Although When GO was docked against one of the mutated regions, much attention has been placed on cancer therapy, there are enhanced binding affinity was observed. For example, D614G studies suggesting that these materials are also promising can- mutation in the viral spike, which has also been reported to didates for anti-microbial therapies. Experimental studies have cause higher infectivity among human populations, was found shown that the interaction between graphene-related 2D mate- to cause a structural change at 6VXX.[77–80] Based on this knowl- rials and bacteria, viruses and fungi could lead to strong anti- edge and the experimental data obtained in vitro, when GO was bacterial and antiviral activities.[65] For example graphene oxide computationally docked against one of these mutated spike (GO) derivatives have been shown to compete with the cell sur- regions, enhanced binding affinity was observed. We postulate face receptor heparan sulfate in binding herpes simplex virus that further interference of all viral genotype mutations that type-1 (HSV-1).[66] Another study reported the broad-spectrum may be encountered in human populations can be achieved antiviral activity of GO against pseudorabies virus (PRV, a DNA via appropriate engineering the GO material properties via sur- virus) and porcine epidemic diarrhea virus (PEDV, an RNA face functionalization. In addition to the D614G mutation, it virus). According to these reports, GO significantly suppressed has been recently reported in GISAID platform that the other the infection of PRV and PEDV at non-cytotoxic concentra- most common receptor binding mutations S477N (part of large tions.[67] Deokar et al. also showed the design and synthesis of Melbourne outbreak from clade GR and some Central Euro- sulfonated magnetic nanoparticles functionalized with reduced pean clade GH clusters), N439K (the long lasting UK outbreak graphene oxide (SMRGO) to capture and photothermally with clade G and the European spillover), N501Y (the new UK destroy HSV-1.[68] In a more recent study by Donskyi et al., a variant VUI-202012/01 in clade GR and a recent clade GH out- series of graphene derivatives with defined polyglycerol sulfate break in South Africa) and Y453F (the mink adaptation) as well and fatty amine functionalities have been synthesized and their as combinations of these mutations with deletions alter the interactions with HSV-1 were investigated. When 2D sheets surface of spike protein. These changes will certainly affect its were functionalized with C6- and C9-alkyl chains, they showed affinity for host receptors, as well as the antiviral nanomate- efficient inhibition of HSV-1 without any significant toxicity in rials being developed. Therefore, our findings actually show the VeroE6 cells, suggesting that antiviral agents against HSV-1 can importance of considering different viral genotypes/mutations be obtained by controlled and stepwise functionalization of gra- that will arise also in the future, in order to better understand phene sheets.[69] the effect of nanomaterials tested against SARS-CoV-2. Overall, Recently, several groups started reporting the antiviral activi- our observations suggest that GO can be considered a prom- ties of different nanomaterials against the novel SARS-CoV-2 ising candidate to be used as an antiviral platform nanomaterial virus. Nanoparticles containing silver, aluminum nitride or in the design of either PPE able to capture and retain viral par- copper have been suggested to inhibit SARS-CoV-2 infec- ticles[76] for disease prevention, or as an antiviral drug delivery tivity.[70–73] In another approach, Zhang et al. constructed system for therapeutic purposes. Small 2021, 17, 2101483 2101483 (9 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com 4. Experimental Section were collected on several spots after irradiation with a power of 1 mW for 10 s and using a 50x objective to focus on the sample. Samples were Molecular Docking: The crystal structures of 6M0Ja, 1R42b, 6VXXc, prepared by drop casting of 20 µL onto glass coverslip and then, dried and 6VYBc were obtained from the Research Collaboratory for Structural overnight. The data were collected and analyzed with Origin software. Bioinformatics Protein Data Bank (RSCB PDB).[81–83] These protein The intensity ratio ID/IG was obtaining by taking the peak intensities domains have been identified as: 6M0J: SARS-CoV-2 spike receptor-binding without baseline correction. domain bound with ACE2, 1R42: native human Angiotensin Converting Zeta Potential Measurements: Zeta potential (ζ) was measured by a Enzyme (ACE2), 6VXX: SARS-CoV-2 spike glycoprotein (closed state) Zeta-sizer Nano ZS (Malvern instruments) equipped with disposable and 6VYB: spike ectodomain structure (open state). Prior preparation capillary cells at the ICN2 Molecular Spectroscopy and Optical of protein domains (screening, removing waters, small molecules, and Microscopy Facility. Water dispersant settings for refractive index and more), docking analysis, and image processing after analysis were all viscosity, and automatic analysis were used for all GO measurements performed by Chimera version 1.14 and Discovery Studio 2020, which (20 µg mL−1). Each sample was measured three times at room have been well-established in literature.[84,85] The free-standing GO surface temperature. (with functional groups on both sides of the layer) is optimized using UV–Vis Spectroscopy: Absorbance was evaluated by using a Nanodrop modified AMBER force field and MD simulations. The split in the oxygen 2000c spectrophotometer (Thermo Scientific) at room temperature groups of GO is about: Epoxy oxygen 15% Hydroxy oxygen 10% Carboxy using a Hellma QS Quartz micro cuvette. GO samples were prepared in (edge groups) 5% (% calculated with respect to the total carbon atoms water in a concentration range of 2.5–20 µg mL−1. on the graphene layer). GO and protein models prepared in PDB format X-Ray Photoemission Spectroscopy: XPS data was obtained using a were converted to PDBQT format with the addition of Gasteiger partial Phoibos 150 (SPECS, GmbH) electron spectrometer equipped with a charges via the AutoDock program version 4.2,[86] and grid box mapping hemispherical analyzer, operating under ultrahigh-vacuum conditions, parameters were also determined (Table S1, Supporting Information). In and with an Al Kα (hν = 1486.74 eV) X-ray source, at the ICN2 order to observe the affinity of GO against the whole protein structure, Photoemission Spectroscopy Facility. Charge effects on the samples were grid box parameters were adjusted to cover the whole protein. Finally, removed by taking the C1s line from adventitious carbon at 284.6 eV. molecular docking analysis of GO and protein prepared in PDBQT format Samples were prepared by deposition of 20 µg of GO material onto 5×5 were performed using AutoDock-Vina,[87] which is a sensitive and high- silicon wafers (Ted Pella) and dried overnight. In order to estimate the performance docking software. As a result of the calculations, nine results photoelectron peak intensities, CasaXPS software (http://www.casaxps. with the highest binding affinity were evaluated. For mutant analysis, the com) was used. program Robetta was used for B chain structural prediction of 6VYB and Cell Culture: African green monkey kidney Vero E6 cell line was 6VXX proteins (https://robetta.bakerlab.org). purchased from ATCC and maintained in DMEM media containing Material (Graphene Oxide) Synthesis and Characterisation: Synthesis of 10% FBS and 1% antibiotics. Four local SARS-CoV-2 isolates (hCoV-19/ Biological-Grade GO: Biological-grade GO was synthesized from graphite Turkey/HSGM-302/2020 (Clade GR), hCoV-19/Turkey/HSGM-439/2020 powder (Sigma–Aldrich, UK) by the modified Hummers’ method as (Clade other), hCoV-19/Turkey/HSGM-1049/2020 (Clade GH), and previously described.[88] Depyrogenation of all the glassware involved hCoV-19/Turkey/HSGM-1027/2020 (Clade S) with GISAID accession in the synthesis and the use of non-pyrogenic material was used to numbers of EPI_ISL_437313|2020-03-27, EPI_ISL_437307|2020-03-25, obtain endotoxin-free GO material suitable for biological applications.[38] EPI_ISL_437309|2020-03-26 and EPI_ISL_437317|2020-03-27, respectively) Briefly, 18.4 mL of sulfuric acid was added to the mixture of the reaction was used in this study. Viruses were propagated in Vero E6 cells by using containing 0.8 g of graphite and 0.4 g of sodium nitrate. Then, 2.4 g of DMEM media containing 2% FBS and 1% antibiotics. All virus-related potassium permanganate was added very slowly. After 30 min, 37.5 mL experiments were performed at Biosafety Level 3 laboratories. of water was transferred slowly and the reaction was kept at 98 °C for Viral Infection: Vero E6 cells were seeded onto 96-well plates at 30 min. Afterward, 112.5 mL of water was added and also 12.5 mL of 30% confluency. Cells were either incubated with GO (5, 10, 50, 100 µg mL−1) hydrogen peroxide to stop the reaction. The consequent GO purification or SARS-CoV-2 (MOI 0.1) in a pre-infection or post-infection protocol. was performed by centrifugation steps until the pH of the supernatant In pre-infection protocol, cells were first treated with 100 µL media was neutral. GO material was obtained after 5 min sonication and then containing GO and after 2 h, viral particles were added onto the purified by centrifugation.[39] culture media. In a post-infection protocol, cells were first infected with Atomic Force Microscopy: An Asylum MFP-3D atomic force microscope SARS-CoV-2 (100 µL) and following 2 h of incubation, GO was added (Oxford instruments) operating in standard air-tapping mode and into the wells. After the final treatment, plates were incubated at 5% CO2 equipped with silicon probes (Ted Pella) with a resonance frequency at 37 °C. Plates were monitored for cytopathic activity for 5 days. of 300 kHZ and a nominal force of 40 N m−1 was used to characterize qRT-PCR Analysis: At the end of day 5, cell culture supernatants were the surface. Samples were prepared by drop casting of 20 µL of GO collected. Total RNA was isolated via MPLC total nucleic acid isolation suspension (100 µg mL−1) onto a freshly cleaved mica surface (Ted Pella) kit (Roche) using automated MagNA Pure LC Instrument (Roche). One- previously covered with 20 µL of poly-L-lysine 0.01% solution (Sigma- step qRT-PCR was performed using the transcriptor one-step RT-PCR Kit Aldrich), subsequently washed with water, and then dried overnight at (Roche) by using 5 µL of samples per each 20 µL reaction volume. In room temperature. Images were processed using Gwyddion software order to calculate viral copies per µL, a standard curve was constructed (http://gwyddion.net, version 2.56). Distribution analysis of 5 µm x 5 µm by using five different standards of known copy numbers according to height AFM images to obtain the lateral dimensions was performed the viral N gene. using ImageJ software (https://imagej.nih.gov). Plaque Assay: Vero E6 cells were seeded onto 48-well plates at Scanning Electron Microscopy: SEM images were recorded on a confluency. Cells were either incubated with GO (5, 10, 50, 100 µg mL−1) Magellan 400L field emission scanning electron microscope (Oxford or SARS-CoV-2 (MOI 0.1) in a pre-infection or post-infection protocol. instruments) at the ICN2 Electron Microscopy Unit, which was equipped After 1h, carboxymethyl cellulose (CMC) overlay medium (300 µL) was with an Everhart-Thornley as secondary electrons detector and using added into each well. At the end of 4 days, plates were fixed with 10% an acceleration voltage of 20 kV and beam current of 0.1 nA. In each formaldehyde and stained with crystal violet. Plaques were counted for sample, 20 µL of GO material (100 µg mL−1) were deposited on an each condition and plaque forming units were calculated. Based on Ultrathin C on Lacey C grid; any excess of material was removed and pfu values, % of inhibition was calculated for each treatment group dried overnight at room temperature. The lateral dimension distribution compared to the virus control group. was obtained by measuring the flakes using ImageJ software. Statistical Analysis: All values were expressed as mean ±ST.D. Raman Spectroscopy: Raman spectra were acquired with a confocal Comparison between groups was performed by one-way ANOVA, Raman microscope (Witec) at room temperature coupled to 632 nm followed by a Tukey’s post hoc multiple comparisons by the statistics laser excitation and using a grating of 600 g nm−1. Single Raman spectra program GraphPad. At least three independent replicated were analyzed. Small 2021, 17, 2101483 2101483 (10 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com Supporting Information [6] C. Harrison, Nat. Biotechnol. 2020, 38, 379. [7] Y. Han, P. Král, ACS Nano 2020, 14, 5143. Supporting Information is available from the Wiley Online Library or [8] S. L. Senanayake, Future Drug. Discovery 2020, 2, 2. from the author. [9] S. Sadegh, J. Matschinske, D. B. Blumenthal, G. Galindez, T. Kacprowski, M. List, R. Nasirigerdeh, M. Oubounyt, A. Pichlmair, T. D. Rose, M. Salgado-Albarrán, J. Späth, A. Stukalov, N. K. Wenke, K. Yuan, J. K. Pauling, J. Baumbach, Nat. Commun. 2020, 11, Acknowledgements 3518. M.A.U. and F.B. contributed equally to this work. L.G.D. acknowledges [10] A. Trezza, D. Iovinelli, A. Santucci, F. Prischi, O. Spiga, Sci. Rep. the Starting Grant 2020 from the Department of Biomedical Science, 2020, 10, 13866. University of Padua. K.K. acknowledges financial support from the [11] Y. Zhou, F. Wang, J. Tang, R. Nussinov, F. Cheng, Lancet Global UKRI EPSRC Programme Grant 2D-Health (project EP/P00119X/1) and Health 2020, 2, e667. the European Union Horizon 2020 Graphene Flagship project under [12] X.-Y. Meng, H.-X. Zhang, M. Mezei, M. Cui, Curr. Comput. - Aided grant agreement GrapheneCore2 785219. A.Y. is thankful to the Turkish Drug Des. 2011, 7, 146. Academy of Sciences (TUBA) for financial support under the young [13] L. Zheng, A. A. Alhossary, C.-K. Kwoh, Y. Mu, in Encyclopedia of investigator programme. M.A.U., H.N., K.C.A., and A.Y. would like to Bioinformatics and Computational Biology (Ed. S. Ranganathan, M. acknowledge the funding by the Scientific and Technology Council of Gribskov, K. Nakai, C. Schönbach), Academic Press, Oxford 2019, Turkey (TUBITAK) under the grant 18AG020. The authors would like p. 550. to acknowledge the following Groups at ICN2: Advanced Electronic [14] K. Rasool, M. Helal, A. Ali, C. E. Ren, Y. Gogotsi, K. A. Mahmoud, Materials and Devices (Prof. Jose A. Garrido) regarding access to ACS Nano 2016, 10, 3674. their AFM and Raman instrumentation, and the Supramolecular NanoChemistry and Materials (Prof. Daniel Maspoch) for access to their [15] A. Kukol, Molecular Modeling of Proteins, Humana Press, Totowa, NJ UV–vis equipment. 2008. [16] J. M. Sanders, M. L. Monogue, T. Z. Jodlowski, J. B. Cutrell, JAMA, J. Am. Med. Assoc. 2020, 18, 1824. [17] C. Wu, Y. Liu, Y. Yang, P. Zhang, W. Zhong, Y. Wang, Q. Wang, Y. Xu, Conflict of Interest M. Li, X. Li, M. Zheng, L. Chen, H. Li, Acta Pharm. Sin. B 2020, 10, 766. The authors declare no conflict of interest. [18] M. Wang, R. Cao, L. Zhang, X. Yang, J. Liu, M. Xu, Z. Shi, Z. Hu, W. Zhong, G. Xiao, Cell Res. 2020, 30, 269. [19] A. Zumla, J. F. Chan, E. I. Azhar, D. S. Hui, K. Y. Yuen, Nat. Rev. Drug Discovery 2016, 15, 327. Data Availability Statement [20] M. Hoffmann, H. Kleine-Weber, S. Schroeder, N. Krüger, T. Herrler, S. Erichsen, T. S. Schiergens, G. Herrler, N. H. Wu, A. Nitsche, The data that support the findings of this study are available from the corresponding author upon request. M. A. Müller, C. Drosten, S. Pöhlmann, Cell 2020, 181, 271. [21] J. Liu, R. Cao, M. Xu, X. Wang, H. Zhang, H. Hu, Y. Li, Z. Hu, W. Zhong, M. Wang, Cell Discov. 2020, 6, 16. [22] M. A. Unal, C. V. Bitirim, G. Y. Summak, S. Bereketoglu, I. C. Zeytin, Keywords O. Bul, C. Gurcan, D. Aydos, E. Goksoy, E. Kocakaya, Z. Eran, M. Murat, N. Demir, J. Somers, E. Demir, H. Nazir, S. A. Ozkan, antiviral therapeutics, COVID-19, in silico, in vitro, viral mutations A. Ozkul, A. Azap, A. Yilmazer, K. C. Akcali, Can. J. Physiol. Phar- macol. 2021, 99, 1. Received: March 13, 2021 [23] G. Borkow, S. S. Zhou, T. Page, J. Gabbay, PLoS One 2010, 5, Published online: May 14, 2021 e11295. [24] J. S. Oxford, R. Lambkin, M. Guralnik, R. A. Rosenbloom, M. P. Petteruti, K. Digian, C. Lefante, Am. J. Ther. 2007, 14, 455. [25] M. S. Nestor, N. Swenson, A. Macri, M. Manway, P. Paparone, [1] COVID-19 Dashboard by the Center for Systems Science and J. Clin. Aesthetic Dermatol. 2016, 3, 25. Engineering (CSSE) at Johns Hopkins University (JHU), https:// [26] B. De Gusseme, L. Sintubin, L. Baert, E. Thibo, T. Hennebel, coronavirus.jhu.edu/map.html (accessed: April 2021). G. Vermeulen, M. Uyttendaele, W. Verstraete, N. Boon, Appl. [2] B. Shanmugaraj, K. Siriwattananon, K. Wangkanont, Environ. Microbiol. 2010, 76, 1082. W. Phoolcharoen, Asian Pac. J. Allergy Immunol. 2020, 1, 10. [27] G. Tiliket, D. L. Sage, V. Moules, M. Rosa-Calatrava, B. Lina, [3] Y. F. Tu, C. S. Chien, A. A. Yarmishyn, Y. Y. Lin, Y. H. Luo, Y. T. Lin, J. M. Valleton, Q. T. Nguyen, L. Lebrun, Chem. Eng. J. 2011, W. Y. Lai, D. M. Yang, S. J. Chou, Y. P. Yang, M. L. Wang, S. H. Chiou, 173, 341. Int. J. Mol. Sci. 2020, 21, 7. [28] S. Z. Butler, S. M. Hollen, L. Cao, Y. Cui, J. A. Gupta, H. R. Gutiérrez, [4] C. Weiss, M. Carriere, L. Fusco, I. Capua, J. A. Regla-Nava, T. F. Heinz, S. S. Hong, J. Huang, A. F. Ismach, E. Johnston- M. Pasquali, J. A. Scott, F. Vitale, M. A. Unal, C. Mattevi, Halperin, M. Kuno, V. V. Plashnitsa, R. D. Robinson, R. S. Ruoff, D. Bedognetti, A. Merkoçi, E. Tasciotti, A. Yilmazer, Y. Gogotsi, S. Salahuddin, J. Shan, L. Shi, M. G. Spencer, M. Terrones, F. Stellacci, L. G. Delogu, ACS Nano 2020, 14, 6383. W. Windl, J. E. Goldberger, ACS Nano 2013, 7, 2898. [5] L. Riva, S. Yuan, X. Yin, L. Martin-Sancho, N. Matsunaga, L. Pache, [29] K. Kalantar-zadeh, J. Z. Ou, T. Daeneke, M. S. Strano, M. Pumera, S. Burgstaller-Muehlbacher, P. D. De Jesus, P. Teriete, M. V. Hull, S. L. Gras, Adv. Funct. Mater. 2015, 25, 5086. M. W. Chang, J. F.-W. Chan, J. Cao, V. K.-M. Poon, K. M. Herbert, [30] J. Zhu, P. Xiao, H. Li, S. A. C. Carabineiro, ACS Appl. Mater. Inter- K. Cheng, T.-T. H. Nguyen, A. Rubanov, Y. Pu, C. Nguyen, A. Choi, faces 2014, 6, 16449. R. Rathnasinghe, M. Schotsaert, L. Miorin, M. Dejosez, T. P. Zwaka, [31] H. Taheri, M. A. Unal, M. Sevim, C. Gurcan, O. Ekim, A. Ceylan, K.-Y. Sit, L. Martinez-Sobrido, W.-C. Liu, K. M. White, et al., Nature Z. Syrgiannis, K. C. Christoforidis, S. Bosi, O. Ozgenç, M. J. Gómez, 2020, 586, 113. M. Turktas Erken, Ç. Soydal, Z. Eroğlu, C. V. Bitirim, U. Cagin, F. Arı, Small 2021, 17, 2101483 2101483 (11 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com A. Ozen, O. Kuçuk, L. G. Delogu, M. Prato, Ö. Metin, A. Yilmazer, [59] D. P. Tran, V. T. Lam, T. L. Tran, T. N. S. Nguyen, H. T. Thi Tran, Small 2020, 16, 1904619. Phys. Chem. Chem. Phys. 2018, 20, 19240. [32] R. Kurapati, K. Kostarelos, M. Prato, A. Bianco, Adv. Mater. 2016, 28, [60] H. Vovusha, S. Sanyal, B. Sanyal, J. Phys. Chem. Lett. 2013, 4, 3710. 6052. [61] A. D. Putri, B. T. Murti, S. Kanchi, M. I. Sabela, K. Bisetty, A. Tiwari [33] Y. Chen, C. Tan, H. Zhang, L. Wang, Chem. Soc. Rev. 2015, 44, 2681. Inamuddin, A. M. Asiri, Sci. Rep. 2019, 1, 7873. [34] A. F. Rodrigues, L. Newman, N. Lozano, S. P. Mukherjee, B. Fadeel, [62] R. Kurapati, K. Kostarelos, M. Prato, A. Bianco, Adv. Mater. 2017, 29, C. Bussy, K. Kostarelos, 2D Mater. 2018, 5, 035020. 6052. [35] S. Vranic, K. Kostarelos, Biomedical Nanotechnology, Methods in [63] L. Fusco, A. Gazzi, G. Peng, Y. Shin, S. Vranic, D. Bedognetti, Molecular Biology, Vol. 1570, Springer, Berlin 2017, pp.287. F. Vitale, A. Yilmazer, X. Feng, B. Fadeel, C. Casiraghi, L. G. Delogu, [36] A. F. Rodrigues, L. Newman, D. Jasim, S. P. Mukherjee, J. Wang, Theranostics 2020, 10, 5435. I. A. Vacchi, C. Ménard-Moyon, A. Bianco, B. Fadeel, K. Kostarelos, [64] A. Gazzi, L. Fusco, A. Khan, D. Bedognetti, B. Zavan, F. Vitale, C. Bussy, Adv. Sci. 2020, 7, 1903200. A. Yilmazer, L. G. Delogu, Front. Bioeng. Biotechnol. 2019, 7, [37] S. P. Mukherjee, G. Gupta, K. Klöditz, J. Wang, A. F. Rodrigues, 295. K. Kostarelos, B. Fadeel, Small 2020, 16, 1907686. [65] Z. Wang, W. Zhu, Y. Qiu, X. Yi, A. von dem Bussche, A. Kane, [38] S. P. Mukherjee, N. Lozano, M. Kucki, A. E. Del Rio-Castillo, H. Gao, K. Koski, R. Hurt, Chem. Soc. Rev. 2016, 45, 1750. L. Newman, E. Vázquez, K. Kostarelos, P. Wick, B. Fadeel, PLoS One [66] M. Sametband, I. Kalt, A. Gedanken, R. Sarid, ACS Appl. Mater. 2016, 11, e0166816. Interfaces. 2014, 6, 1228. [39] A. F. Rodrigues, L. Newman, N. Lozano, S. P. Mukherjee, B. Fadeel, [67] S. Ye, K. Shao, Z. Li, N. Guo, Y. Zuo, Q. Li, Z. Lu, L. Chen, Q. He, C. Bussy, K. Kostarelos, 2D Mater. 2018, 5, 035020. H. Han, ACS Appl. Mater. Interfaces. 2015, 7, 21571. [40] S. P. Mukherjee, B. Lazzaretto, K. Hultenby, L. Newman, [68] A. R. Deokar, A. P. Nagvenkar, I. Kalt, L. Shani, Y. Yeshurun, A. F. Rodrigues, N. Lozano, K. Kostarelos, P. Malmberg, B. Fadeel, A. Gedanken, R. Sarid, Bioconjugate Chem. 2017, 28, 1115. Chem 2018, 4, 334. [69] I. S. Donskyi, W. Azab, J. L. Cuellar-Camacho, G. Guday, A. Lippitz, [41] S. Vranic, A. F. Rodrigues, M. Buggio, L. Newman, M. R. H. White, W. E. S. Unger, K. Osterrieder, M. Adeli, R. Haag, Nanoscale 2019, D. G. Spiller, C. Bussy, K. Kostarelos, ACS Nano 2018, 12, 11, 15804. 1373. [70] G. C. Tremiliosi, L. G. P. Simoes, D. T. Minozzi, R. I. Santos, [42] I. de Lázaro, S. Vranic, D. Marson, A. F. Rodrigues, M. Buggio, D. C. B. Vilela, E. L. Durigon, R. R. G. Machado, D. S. Medina, A. Esteban-Arranz, M. Mazza, P. Posocco, K. Kostarelos, Nanoscale L. K. Ribeiro, I. L. V. Rosa, M. Assis, J. Andrés, E. Longo, 2019, 11, 13863. L. H. Freitas-Junior, bioRxiv. 2020, 2020.06.26.152520. [43] I. de Lázaro, P. Sharp, C. Gurcan, A. Ceylan, M. Stylianou, T. Kisby, [71] G. Pezzotti, E. Ohgitani, M. Shin-Ya, T. Adachi, E. Marin, Y. Chen, S. Vranic, K. Barr, H. Taheri, A. Ozen, C. Bussy, A. Yilmazer, F. Boschetto, W. Zhu, O. Mazda, bioRxiv. 2020, 2020.06.19.159970. K. Kostarelos, Adv. Ther. 2021, 4, 2000109. [72] J. Hasan, A. Pyke, N. Nair, T. Yarlagadda, G. Will, K. Spann, [44] W. Dai, B. Zhang, X.-M. Jiang, H. Su, J. Li, Y. Zhao, X. Xie, Z. Jin, P. K. D. V. Yarlagadda, ACS Biomater. Sci. Eng. 2020, 6, J. Peng, F. Liu, C. Li, Y. Li, F. Bai, H. Wang, X. Cheng, X. Cen, S. Hu, 4858. X. Yang, J. Wang, X. Liu, G. Xiao, H. Jiang, Z. Rao, L.-K. Zhang, [73] S. S. Jeremiah, K. Miyakawa, T. Morita, Y. Yamaoka, A. Ryo, Y. Xu, H. Yang, H. Liu, Science 2020, 368, 1331. Biochem. Biophys. Res. Commun. 2020, 533, 195. [45] T. Yadavalli, D. Shukla, Nanomedicine 2017, 1, 219. [74] Q. Zhang, A. Honko, J. Zhou, H. Gong, S. N. Downs, J. H. Vasquez, [46] L. Singh, H. G. Kruger, G. E. M. Maguire, T. Govender, R. H. Fang, W. Gao, A. Griffiths, L. Zhang, Nano Lett. 2020, 20, R. Parboosing, Ther. Adv. Infect. Dis. 2017, 4, 105. 5570. [47] J. A. Jackman, J. Lee, N.-J. Cho, Small 2016, 12, 1133. [75] B. Raval, A. K. Srivastav, S. K. Gupta, U. Kumar, S. K. Mahapatra, [48] F. D. Cojocaru, D. Botezat, I. Gardikiotis, C. M. Uritu, G. Dodi, P. N. Gajjar, I. Banerjee, J. Biomol. Struct. Dyn. 2020, 1. L. Trandafir, C. Rezus, E. Rezus, B. I. Tamba, C. T. Mihai, Pharma- [76] F. De Maio, V. Palmieri, G. Babini, A. Augello, I. Palucci, G. Perini, ceutics 2020, 12, 171. A. Salustri, M. De Spirito, M. Sanguinetti, G. Delogu, L. G. Rizzi, [49] M. Milovanovic, A. Arsenijevic, J. Milovanovic, T. Kanjevac, G. Cesareo, P. Soon-Shiong, M. Sali, M. Papi, medRxiv. 2020, N. Arsenijevic, Antimicrobial Nanoarchitectonics: From Synthesis 2020.09.16.20194316. to Applications (Ed. A. M. Grumezescu), Elsevier, Amsterdam, the [77] Q. Li, J. Wu, J. Nie, L. Zhang, H. Hao, S. Liu, C. Zhao, Netherlands 2017, p. 383. Q. Zhang, H. Liu, L. Nie, H. Qin, M. Wang, Q. Lu, X. Li, Q. Sun, [50] X. Dong, W. Liang, M. J. Meziani, Y.-P. Sun, L. Yang, Theranostics J. Liu, L. Zhang, X. Li, W. Huang, Y. Wang, Cell 2020, 182, 2020, 10, 671. 1284. [51] T. Lunnoo, J. Assawakhajornsak, T. Puangmali, J. Phys. Chem. C. [78] J. Ogawa, W. Zhu, N. Tonnu, O. Singer, T. Hunter, A. L. Ryan, 2019, 123, 3801. G. M. Pao, bioRxiv. 2020, 2020.07.21.214932. [52] S. Kunjiappan, T. Panneerselvam, S. Govindaraj, P. Parasuraman, [79] I. O. Omotuyi, O. Nash, O. B. Ajiboye, C. G. Iwegbulam, S. Baskararaj, M. Sankaranarayanan, S. Arunachalam, E. B. Oyinloye, O. A. Oyedeji, Z. A. Kashim, K. Okaiyeto, J. Comput. E. Babkiewicz, A. Jeyakumar, M. Lakshmanan, Adv. Anticancer Chem. 2020, 41, 2158. Agents Med. Chem. 2019, 16, 1966. [80] S. Isabel, L. Graña-Miraglia, J. M. Gutierrez, C. Bundalovic-Torma, [53] P. A. Wijeratne, V. Vavourakis, Interface Focus 2019, 3, 20180063. H. E. Groves, M. R. Isabel, A. Eshaghi, S. N. Patel, J. B. Gubbay, [54] E. Peng, N. Todorova, I. Yarovsky, ACS Omega 2018, 3, 11497. T. Poutanen, D. S. Guttman, S. M. Poutanen, Sci. Rep. 2020, 1, [55] L. Baweja, K. Balamurugan, V. Subramanian, A. Dhawan, J. Mol. 14031. Graphics Modell. 61, 2015, 175. [81] J. Lan, J. Ge, J. Yu, S. Shan, H. Zhou, S. Fan, Q. Zhang, X. Shi, [56] D. Chen, C. Copeland, D. Majumdar, S. Roszak, J. Leszczynski, Q. Wang, L. Zhang, X. Wang, Nature 2020, 581, 215. Struct. Chem. 2017, 28, 379. [82] P. Towler, B. Staker, S. G. Prasad, S. Menon, J. Tang, T. Parsons, [57] S. S. K. Mallineni, J. Shannahan, A. J. Raghavendra, A. M. Rao, D. Ryan, M. Fisher, D. Williams, N. A. Dales, M. A. Patane, J. M. Brown, R. Podila, ACS Appl. Mater. Interfaces. 2016, 8, 16604. M. W. Pantoliano, J. Biol. Chem. 2004, 279, 17996. [58] T. Sumaryada, M. S. Gunawan, S. Perdana, S. Arjo, A. Maddu, [83] A. C. Walls, Y. J. Park, M. A. Tortorici, A. Wall, A. T. McGuire, Biosensors 2019, 9, 18. D. Veesler, Cell 2020, 181, 281. Small 2021, 17, 2101483 2101483 (12 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH www.advancedsciencenews.com www.small-journal.com [84] E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch, [86] G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner, R. K. Belew, D. M. Greenblatt, E. C. Meng, T. E. Ferrin, J. Comput. Chem. 2004, D. S. Goodsell, A. J. Olson, J. Comput. Chem. 2009, 30, 2785. 25, 1605. [87] O. Trott, A. J. Olson, J. Comput. Chem. 2010, 2, 455. [85] Biovia Discovery Studio, https://www.discngine.com/discovery- [88] H. Ali-Boucetta, D. Bitounis, R. Raveendran-Nair, A. Servant, studio (accessed: May 2020). J. Van den Bossche, K. Kostarelos, Adv. Healthcare Mater. 2013, 2, 433. Small 2021, 17, 2101483 2101483 (13 of 13) © 2021 The Authors. Small published by Wiley-VCH GmbH