Silver Nanoparticle-Decorated Reduced Graphene Oxide Nanomaterials Exert Membrane Stress and Induce Immune Response to Inhibit the Early Phase of HIV-1 Infection

www.advmatinterfaces.de 2201996 (1 of 15) © 2022 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH Silver Nanoparticle-Decorated Reduced Graphene Oxide Nanomaterials Exert Membrane Stress and Induce Immune Response to Inhibit the Early Phase of HIV-1 Infection Soumajit Mukherjee, Zuzana Bytesnikova, Sophie Martin, Pavel Svec, Andrea Ridoskova, Jana Pekarkova, Cendrine Seguin, Jean-Luc Weickert, Nadia Messaddeq, Yves Mély, Lukas Richtera, Halina Anton,* and Vojtech Adam* DOI: 10.1002/admi.202201996 1. Introduction Graphene oxide (GO) has been at the forefront of carbon nanomaterial research over the past two decades. Due to their unique properties, such as a large surface area, high tensile strength, the presence of modifiable surface groups, and good biocompatibility, graphene derivatives have been used to expand several areas of research including electronics, materials science, nonlinear optics, and biotechnologies.[1–8] GO is formally derived from graphene, which is a single layer of carbon atoms bonded in a 2D hexagonal lattice.[9,10] The chemical oxidation and exfoliation of graphite produce GO-exhibiting surface groups, such as carboxyl, hydroxyl, epoxy, and carbonyl groups, that offer an excellent opportunity to covalently bind biomolecules, drugs, or fluorophores. The exact composition and number of these groups are variable and depend on the synthesis pathway.[2,11] The chemical Graphene-based 2D nanomaterials exhibit unique physicochemical, electric, and optical properties that facilitate applications in a wide range of fields including material science, electronics, and biotechnology. Recent studies have shown that graphene oxide (GO) and reduced graphene oxide (rGO) exhibit antimicrobial effects on bacteria and viruses. While the bactericidal activity of graphene-based nanomaterials is related to mechanical and oxidative damage to bacterial membranes, their antiviral activity has been less explored. Currently available experimental data are limited and suggest mechanical disruption of viral particles prior to infection. In this study, the antiviral properties of reduced GO-based nanocomposites decorated with Ag nanoparticles (rGO-Ag) are evidenced against human immunodeficiency virus-1 pseudovirus used as an enveloped virus model. By combining biochemical and original single virus imaging approaches, it is shown that rGO-Ag induces peroxidation of pseudoviral lipid membrane and that consequent alteration of membrane properties leads to a reduction in cell entry. In addition, rGO-Ag is found to be efficiently internalized in the host cell leading to the elevated expression of proinflammatory cytokines. Altogether, the presented results shed new light on the mechanisms of rGO-Ag antiviral properties and confirm the high potential of graphene derivatives as an antimicrobial material for biomedical applications. © 2022 The Authors. Advanced Materials Interfaces published by WileyVCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. ReseaRch aRticle The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/admi.202201996. S. Mukherjee, Z. Bytesnikova, P. Svec, A. Ridoskova, L. Richtera, V. Adam Department of Chemistry and Biochemistry Mendel University in Brno Zemedelska 1, Brno 61300, Czech Republic E-mail: [email protected] S. Martin, Y. Mély, H. Anton Laboratory of Bioimaging and Pathologies CNRS UMR 7021 Faculty of Pharmacy University of Strasbourg 74, route du Rhin, Illkirch-Graffenstaden 67400, France E-mail: [email protected] J. Pekarkova Department of Microelectronics Faculty of Electrical Engineering and Communication Brno University of Technology Technicka 3058/10, Brno 61600, Czech Republic J. Pekarkova, V. Adam Central European Institute of Technology Brno University of Technology Purkynova 656/123, Brno 61200, Czech Republic C. Seguin Laboratory of Design and Application of Bioactive Molecules CNRS UMR 7199 Faculty of Pharmacy University of Strasbourg 74, route du Rhin, Illkirch-Graffenstaden 67400, France J.-L. Weickert, N. Messaddeq Imaging Center Institute of Genetics and Molecular and Cellular Biology (IGBMC) 1 Rue Laurent Fries, Illkirch-Graffenstaden 67400, France Adv. Mater. Interfaces 2023, 10, 2201996 www.advancedsciencenews.com © 2022 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 2201996 (2 of 15) www.advmatinterfaces.de reduction of GO reduces the surface functional groups to produce reduced graphene oxide (rGO).[12,13] The atomic-thin layer 2D structure of GO-derived materials is particularly beneficial for biomedical use as the flexibility of the nanomaterial allows it to fold in tight spaces in biological systems. Recent studies have shown that GO and rGO exhibit inhibiting effects on bacteria as well as viruses.[14–19] The biocidal activity of GO was first described in 2010.[20,21] Since, the effects of graphene-based materials were investigated for a variety of microorganisms, including Pseudomonas aeruginosa, Escherichia coli, herpes simplex virus (HSV), pseudorabies virus (PRV), porcine epidemic diarrhea virus (PEDV), porcine reproductive and respiratory syndrome virus (PRRSV), and severe acute respiratory syndrome coronavirus (SARS CoV-2).[16,17,19,22–25] Despite active research in the field, the underlying mechanisms of the antimicrobial activity of GO-based materials are still debated, possibly because of their dependence on the synthesis paths determining the physicochemical properties of the final materials. The size, charge density, and presence of functional groups determine the interactions with biological molecules and hence their potential biological functions.[19,24] Several studies have shown that the sharp edges of 2D GO nanosheets damage microbial external membranes (viral envelopes), leading to the release of intracellular content and infection inhibition. This nano-knife concept has been discussed in studies on E. coli, Staphylococcus aureus, HSV, PRRSV, and PEDV.[14,16,20,26] Less destructive effects, such as GO attachment to the external membranes and lipid extraction, have also been suggested.[14,19,26–28] Moreover, since GO sheets are flexible, they can envelope bacterial or cellular surfaces, and thus, impede pathogen cell entry.[24,29] The antimicrobial activities of GO-derived materials can be enhanced by coupling them with silver nanoparticles (Ag NPs). Indeed, Ag NPs have attracted attention due to their antibacterial, antifungal, and antiviral properties.[18,30,31] Interestingly, Ag NPs and multilayer GO have been shown to stimulate cell proliferation and migration, which can be profitable for biomedical applications such as tissue engineering or wound healing.[32,33] However, due to their high chemical reactivity, Ag NPs aggregate or agglomerate in biological media, which reduces their overall efficacy.[34,35] To overcome this limitation, GO is used as a stabilizer that covalently binds Ag NPs through its surface groups. Ag-decorated GO nanocomposites (GO-Ag) show antiviral activities against enveloped (feline coronavirus) and non-enveloped (infectious bursal disease virus) viruses.[36] In a recent report, Du et al. showed that GO-Ag efficiently inhibits PRRSV and PEDV infection by acting on viral particles prior to entry. Transmission electron microscopy (TEM) images of the viral particles incubated with GO-Ag revealed structural damage caused by the sharp edges of the GO sheets. Interestingly, GO-Ag treatment was also reported to stimulate the innate immune response in host cells (MARC-145).[17] Although the antiviral activities of GO and its derivatives are well established, reports on their inhibitory mechanisms remain rare. In this context, we aimed to depict the antiviral activity of rGO-Ag-based nanomaterials. Human immunodeficiency virus-1 (HIV-1) pseudotyped lentivirus (PSV) was used as enveloped virus model. Compared to the GO-Ag reported in previous antiviral studies, rGO-Ag is more hydrophobic and contains fewer oxygen groups. Therefore, we assume that the interaction with the virus will impact the viral lipid membrane and efficiently inhibit the infectivity of the viral particles. To verify this hypothesis, a series of biochemical and microscopy-based experimental approaches were used to evaluate the effect of rGO-Ag on HIV-1 infection and to unveil the mechanism of this inhibition. 2. Results 2.1. Preparation and Characterization of rGO-Ag GO sheets were synthesized according to the modified Hummer’s method and used as a starting nanomaterial for the preparation of nanocomposites (NCs) with Ag NPs. Figure S1, Supporting Information, shows the scanning electron microscopy (SEM) and atomic force microscopy (AFM) images of the obtained GO as a nanomaterial with a large sheet-like structure and smooth surfaces. Figure1a shows the SEM images of rGO reduced by sodium borohydride in mild (200mg) and stronger (400 mg) conditions in the presence of silver nitrate for the preparation of two distinct NCs, rGO-Ag_L and rGO-Ag_H. In all images, rGO is observed with Ag NPs distributed on its surface. The TEM images of both NCs confirmed the nanoscale size of the Ag NPs (Figure 1b). Energy-dispersive X-ray spectroscopy (EDS) was used to confirm the elemental compositions of all synthesized nanomaterials (Figure S2, Supporting Information). Elemental mapping of the nanoparticles by TEM and scanning transmission electron microscopy (STEM) is demonstrated in the insets, where green represents silver and red represents carbon (Figures1c,d). AFM was used to study the topographies of the synthesized rGO-Ag_L and rGO-Ag_H. Figure2a shows the 2D images of the NCs demonstrating their single-layer character, whereas the 3D images (Figure 2b) demonstrate the thickness of the rGO sheets with distributed Ag NPs. Raman spectroscopy was used to confirm the reduction after the addition of different amounts of sodium borohydride. The D (≈1335cm−1) and G (1610cm−1) peaks showed different intensities for rGO-Ag_L and rGO-Ag_H (Figure2c). The G band is a result of the in-plane vibrations of sp2-bonded carbon atoms, while the D band is due to the out-of-plane vibrations attributed to the presence of structural defects. The intensity ratio of the D/G bands (ID/IG) is thus a measure of the defects in the graphene structure. The ID/IG ratio of rGO-Ag_L was lower than that of rGO-Ag_H, indicating a stronger reduction of rGO in the sample, in which more reducing agent was used (Table1).[37] The Raman spectroscopy data are in accordance with the X-ray photoelectron spectroscopy (XPS) analysis that was performed to determine the rGO-Ag surface composition and the Ag oxidation state. Figure S3, Supporting Information, shows the characteristic wide-scale XPS spectra for the rGO-Ag samples. The high-resolution spectra of the C 1s, O 1s, and Ag 3d regions, presented in Figure3a,b show the covalent bond structures in the nanomaterial. The peak values are indicative of the type of covalent bonds present in the synthesized rGO-Ag samples. The high-resolution spectra of rGO-Ag_L and rGO-Ag_H show three peaks in the C 1s region, indicating the presence of sp2 carbon at 284.71 and 284.78 eV, COH at 286.83 and 286. 90eV, and COC at 288.21 and 288.28eV, respectively Adv. Mater. Interfaces 2023, 10, 2201996 21967350, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202201996 by Technical University In Brno, Wiley Online Library on [04/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com © 2022 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 2201996 (3 of 15) www.advmatinterfaces.de (Figure 3ai,bi).[38] The O 1s regions of rGO-Ag_L and rGOAg_H presented two main components: CO at 531.01 and 531.07 eV and COH at 532.61 and 532.67 eV, respectively (Figure 3aii,bii).[39] The high-resolution XPS spectra of the Ag 3d peaks of rGO-Ag_L and rGO-Ag_H showed two peaks attributed to Ag (3d5/2) and Ag (3d3/2) centered at 368.38, 368.42, 374.38, and 374.38 eV, respectively (Figure 3aiii,biii). The binding energies of these photoelectron peaks are characteristic of metallic Ag, indicating the presence of Ag(0) on the surface of the rGO sheets and confirming the Ag oxidation state.[40] The main chemical species and their atomic concentrations are summarized in Table1. To evaluate the stability and overall surface charge of the synthesized rGO-Ag nanomaterials, their zeta potential was measured. All rGO-Ag samples were found to possess a strongly negative electric charge, favoring a highly stable NC dispersion (Table1). 2.2. rGO-Ag NCs Show Effective Virucidal Activity against HIV-1 PSV The antiviral activities of the graphene-based NCs were evaluated by monitoring the infection of HeLa and THP-1 cells with an HIV-1-based PSV. This model is widely used to study the early phase of viral infection, from cell entry until the integration of the viral genome into host cell DNA. The PSV used Figure 1. Physical characterization of the synthesized rGO-Ag. a) SEM images of rGO-Ag_L (i) and rGO-Ag_H (ii) showing rGO sheets decorated with Ag NPs, scale bars 2µm, b) TEM images of rGO-Ag_L (i) and rGO-Ag_H (ii) showing the presence of nanoscale Ag NPs (scale bars 10nm), c,d) TEM (i), STEM (ii), and EDS (iii) mapping of rGO-Ag_L and rGO-Ag_H, respectively, scale bars 50nm. The STEM images and EDS mapping show the presence of Ag NPs as bright or green spots, respectively, scale bars 50nm. Figure 2. Characterization of single-layer rGO-Ag surfaces. a) 2D AFM images of rGO-Ag_L (i) and rGO-Ag_H (ii), b) 3D AFM images of rGO-Ag_L (i) and rGO-Ag_H (ii) showing the single-layer morphology and thickness, and c) Raman spectra of rGO-Ag_L and rGO-Ag_H showing the reduction state of rGO. Adv. Mater. Interfaces 2023, 10, 2201996 21967350, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202201996 by Technical University In Brno, Wiley Online Library on [04/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com © 2022 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 2201996 (4 of 15) www.advmatinterfaces.de in this study was non-replicative since the viral genome was replaced by luciferase (PSV/luc)- or eGFP (PSV/gfp) genes, whose expression serves as a reporter of infection. To increase cell tropism, the HIV-1 envelope protein was replaced by a vesicular stomatitis virus glycoprotein (VSV-G) that ensured cell entry via endocytosis, independent of the presence of CD4 receptors on the cell surface. To explore the impact of the graphene-based NCs on viral infectivity, the viral particles (PSV/luc) were incubated with a high concentration (100 µg mL−1) of rGO-Ag for different time periods ranging from 5 to 90 min at room temperature. Following the incubation, the samples were centrifuged to remove the larger particles and aggregates, and the PSV-containing supernatants were deposited on the cells. 24 h postinfection (h.p.i.), the cells were lysed, and luciferase expression was measured. As shown in Figure4a, the luciferase assay revealed that virus infectivity was significantly reduced in contact with rGO-Ag. This effect increased with time and reached a plateau at 90 min of PSV/rGO-Ag contact. Therefore, 90 min of incubation was performed in further experiments. Next, we evaluated the dependence of infection inhibition on the rGO-Ag concentration. For this, HIV-1 PSV particles (PSV/ luc) were incubated with increasing concentrations of rGO-Ag ranging from 1to 2000µgmL−1 for 90 min, followed by centrifugation and supernatant deposition on HeLa cells. Infection was measured 24h.p.i. by luciferase activity. As shown in Figure4b,c, the PSV infectivity was reduced by 50% (IC50) when exposed to 48µgmL−1 of rGO-Ag (based on the dry weight of the material), which corresponds to 22µgmL−1 of Ag content. PSV incubation with 100µgmL−1 rGO-Ag decreased the viral infectivity to less than 10%. The reduction state of rGO-Ag did not influence the observed virucidal effect, as rGO-Ag_L and rGO-Ag_H showed similar inhibitory activities (Figure S4, Supporting Information). rGO-Ag_L was selected for further study and is henceforth referred to as rGO-Ag. The virucidal activity of rGO-Ag was then compared to that of Ag NPs and GO sheets used separately (Figure4b,c). As shown in Figure4b, the Ag NPs showed weak antiviral activity, reducing infection by only 30% even at very high concentrations. In contrast, PSV incubation with GO resulted in a significant reduction in infectivity with an IC50 value of 210µgmL−1 (Figure4c). The comparison of the inhibition caused by GO and Ag NPs separately Table 1. Physicochemical characterization of the synthesized rGO-Ag samples. Parameter Method rGO-Ag_L rGO-Ag_H Surface area [µm2]SEM 0.005–4.700, avg. 0.18 0.03–11.70, avg. 3.70 Thickness [µm] AFM 0.04–1.11 0.04–0.37 Degree of defects ID/IGRaman 1.11 1.20 Surface charge [mV] Zeta potential −42.9 −81.4 C sp2 (%) XPS 45.20 47.89 COH, COC and CO in C 1s region [%] XPS 43.89, 9.92, 0.00 41.48, 9.68, 0.00 CO, COH, H2O in O 1s region [%] XPS 16.75, 83.25, 0.00 14.84, 85.16, 0.00 Ag (3d3/2) and Ag (3d5/2) in Ag 3d region [%] XPS 39.98, 60.02 39.98, 60.02 Figure 3. High-resolution XPS spectra of different rGO-Ag NC samples fitted with CasaXPS software (version 2.3.22) showing surface characteristics and bond structures. C 1s, O 1s, and Ag3d region peaks of rGO-Ag_L (a) and rGO-Ag_H (b). The peak values indicate the type of covalent bonds present in the rGO-Ag samples. Adv. Mater. Interfaces 2023, 10, 2201996 21967350, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202201996 by Technical University In Brno, Wiley Online Library on [04/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com © 2022 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 2201996 (5 of 15) www.advmatinterfaces.de with the inhibition caused by rGO-Ag indicated a clear synergistic effect. To investigate the dependence of the antiviral effect of rGO-Ag on the cell line, the same experiments were repeated with THP-1 cells. As shown in Figure4d, a very similar infection inhibition was observed in THP-1 cells, with an IC50=43µgmL−1. Next, the effect of rGO-Ag on viral infectivity was tested using PSV/gfp. Supernatant samples containing this PSV were prepared by the abovementioned method in the presence of increasing amounts of rGO-Ag and used to infect HeLa cells and THP-1 cells. At 24h.p.i., the cells were either harvested for Figure 4. Virucidal effect of rGO-Ag NCs on HIV-1 PSV infection in HeLa and THP-1 cells. a) Time dependence of the virucidal effect of 100µgmL−1 rGO-Ag NCs. b) Inhibition of viral infection in HeLa cells by exposure to rGO-Ag NCs and Ag NPs. The concentration refers to the silver concentration in both types of nanomaterials. c) Inhibition of viral infection in HeLa cells by exposure to GO and rGO-Ag NCs. The concentration refers to the total weight of the nanomaterial. d) Inhibition of viral infection in THP-1 cells by exposure to rGO-Ag NCs. e) Detection of eGFP in the cell lysate after incubation with PSV/gfp pre-incubated with increasing concentrations of rGO-Ag. f) Fluorescence images of HeLa cells infected with HIV-1 PSV/gfp and g) fraction of HeLa cells expressing eGFP 24h.p.i,.mean value ± STD, n= 2000 cells were analyzed for each condition. h) Fluorescence images of THP-1 cells infected with HIV-1 PSV/gfp and i) fraction of THP-1 cells expressing eGFP, mean value ± STD, n= 2000 cells were analyzed for each condition. Adv. Mater. Interfaces 2023, 10, 2201996 21967350, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202201996 by Technical University In Brno, Wiley Online Library on [04/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com © 2022 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 2201996 (6 of 15) www.advmatinterfaces.de whole-cell extract preparation or processed for fixation prior to imaging or flow cytometry. The fluorescence-activated cell sorting (FACS) analysis of the THP-1 cells showed a significant decrease in the median GFP fluorescence intensity in cells treated with rGO-Ag compared to that in untreated cells. As the cell population with a significant fluorescent signal was relatively low (≈10%) (Figure S5, Supporting Information), the samples were observed by fluorescence microscopy, which ensures increased sensitivity compared to FACS. As demonstrated in Figure 4f–h, both the number of GPF-expressing cells and the eGFP content in the cell lysate were significantly reduced in samples treated with rGO-Ag. Whole-cell lysate immunoblotting revealed that the GFP expression was less than 50% with exposure to 100µg mL−1 of rGO-Ag compared to the control, and no expression was observed with 200µgmL−1 of NCs (Figure4e). A similar result was observed when GFP expression was studied by fluorescence microscopy. The imaging of multiple slides revealed that at 24h.p.i., ≈70% of HeLa cells were infected and expressed GFP (Figure4f,g). Exposing PSV particles to increasing concentrations of rGO-Ag reduced the fraction of eGFP-expressing cells to less than 20% (for 200µgmL−1 rGO-Ag), indicating a significant loss of viral infectivity. In THP-1 cells infected with PSV/gfp for 30h, ≈30% of the cells expressed GFP in the control samples, whereas less than 10% of cells were infected (Figure4h,i) when the PSV particles were exposed to 100µgmL−1 of rGO-Ag. Altogether, our data showed that PSV incubation with rGO-Ag resulted in the potent inhibition of viral infectivity. 2.3. Effects of rGO-Ag NCs on Cellular Viability Since the viral infectivity of the HIV-1 PSV particles was measured by the expression of reporter genes, it was necessary to verify that the infected cells were perfectly viable and that the observed decrease in Luc and GFP levels was not caused by cytotoxic effects. During sample preparation, the PSV was incubated with the rGO-Ag (rGO-Ag_p). Then, the samples were centrifuged, and the PSV-containing supernatants (rGO-Ag_s) were deposited on the cells. The initial sample and the supernatant shown in Figure5a clearly indicate that the majority Figure 5. Characterization of the rGO-Ag in the supernatant and cellular toxicity analysis. a) Visual comparison of rGO-Ag whole particles (rGO-Ag_p) and supernatant (rGO-Ag_s) in PBS at 100µgmL−1, b) SEM image showing the sheet-like morphology of rGO and the Ag NP distribution on the surface, c) TEM, STEM, and EDS mapping showing the nanoscale Ag NP distribution on the rGO sheet, d) AFM analysis of rGO-Ag in the supernatant depicting the single-layer topography of the nanomaterial. Cell viability of HeLa (e) and THP-1 (f) cells after incubation with the supernatants containing PSV/luc and increasing concentrations of rGO-Ag NCs. Adv. Mater. Interfaces 2023, 10, 2201996 21967350, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202201996 by Technical University In Brno, Wiley Online Library on [04/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com © 2022 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 2201996 (7 of 15) www.advmatinterfaces.de of rGO-Ag were pelleted and removed during centrifugation. Therefore, we first analyzed the PSV-containing supernatants used for infections. AAS analysis confirmed that the supernatant contained ≈5% of the initially present rGO-Ag (data not shown). The size of rGO-Ag_s was investigated by SEM and TEM coupled with STEM-EDS (Figure 5b,c) and the surface topography was visualized by AFM (Figure5d). These analyses revealed that the NCs in the supernatants were significantly smaller than the originally synthesized NCs shown in Figure1. The NC size distribution according to the measured surface area is given in Figure S6, Supporting Information. In addition, TEM/STEM-EDS analysis confirmed the presence of Ag NPs on the nanosheet surfaces. These observations confirmed that the PSV-containing supernatants used for infection contained nano to micrometer-sized rGO-Ag sheets. To determine the potential cytotoxicity of these rGO-Ag sheets, the viability of HeLa and THP-1 monocytic cells was measured by MTT assay after 24 h of incubation with PSV particles and NC-containing supernatants. The cells were cultured and infected under the same conditions as the previous infectivity experiments. Briefly, the PSV particles (PSV/luc) were incubated in the presence of various concentrations of rGO-Ag, GO, and Ag NPs for 90 min, followed by centrifugation, and the supernatants were deposited on the cells for 24h. Afterward, MTT assays were performed to assess cell viability. As shown in Figure5e, the relative survival rate of HeLa cells when exposed to all types of NCs was greater than 80%, suggesting good biocompatibility. The THP-1 cells, which are more sensitive to culture conditions, showed higher cell death when exposed to supernatants containing high NC concentrations (initial concentration of 200µgmL−1), whereas the relative cell viability was greater than 80% at 100µgmL−1 of rGO-Ag NCs (Figure5f). Since it was experimentally difficult to measure the Ag NPs and GO concentrations in the supernatant, to avoid any bias, all curves in Figure5e,f are plotted with initial sample concentrations. Moreover, the cellular morphology did not change in our experiments. These observations clearly indicate that in both cell lines, infection inhibition was observed under noncytotoxic conditions. Therefore, the decrease in luciferase and GFP expression did not result from cell death but reflected a real suppression of viral infectivity. 2.4. Mechanism of the rGO-Ag Antiviral Activity To understand the mechanism underlying the antiviral activity of the rGO-Ag NCs, we investigated a few possible hypotheses. GO sheets naturally bind biomolecules through different surface groups that establish covalent or electrostatic interactions.[41] Although the chemical reduction of GO removes most oxygen-containing groups from the surface, a significant amount of oxygen-rich functional groups is left to bind biomolecules, as seen from the XPS analysis (Figure3). Moreover, Ag NPs can bind to the SH groups of the VSV-G envelope glycoproteins. All these interactions may facilitate binding between the NCs and viral particles during incubation. We therefore hypothesized that a significant fraction of the PSV particles might be bound and pelleted together with the rGO-Ag during the centrifugation step. Consequently, the observed decrease in Luc expression simply reflects this loss of PSV particles. In addition, the interactions of GO and GO-Ag with microorganisms or viruses were previously shown to rupture their membranes, leading to the release of their internal content. Similar rGO-Ag-induced rupture of the pseudoviral particles could possibly cause the decrease in infection observed in our experiments. To check the possible mechanical disruption or the removal of the viral particles from the samples by binding to rGO-Ag, we examined the total supernatant PSV content. The viral particles were incubated for 90 min with 50, 100, and 200µgmL−1 rGO-Ag or left untreated. The samples were centrifuged, and the total p24 protein amount in the supernatant was measured by ELISA. The p24 protein is a component of the HIV-1 capsid structure that protects the viral core. As indicated in Figure6a, exposure to rGO-Ag and subsequent centrifugation caused only a limited loss of p24 protein. When incubated with 200µgmL−1 of rGO-Ag (conditions that lead to the complete loss of infectivity), the supernatant still contained ≈77% of the initially present p24. However, the total p24 content only provided information about the presence of capsid protein in the sample but did not distinguish between mature capsid in intact viral particles from capsid protein that might be released upon disruption of the viral core or from the capsid domain of non-processed Gag polyprotein. To check for the presence of intact PSV particles in the supernatant, ultracentrifugation was performed to separate the complete PSV particles from the potentially released viral proteins. The pellet was lysed and immunoblotted with an antibody against the HIV-1 integrase protein. Figure6b shows that there was no significant reduction in the amount of integrase in the rGO-Ag-exposed samples compared to control. These results clearly showed the presence of mature PSV particles in the supernatant, highlighting that exposure to rGO-Ag did not alter the viral core or pull down viral particles during centrifugation. To further evaluate the effect of rGO-Ag on viral particles, the morphology of the PSVs was observed by SEM. PSV/luc were treated with 100µgmL−1 rGO-Ag for 90 min. The samples were then centrifuged, and the supernatants were deposited on collagen-coated coverslips. As shown in Figure6c, the spherical shape of the viral particles was unaltered upon exposure to the NCs. Changes in the PSV particle size were also negligible. rGO-Ag-treated viral particles were also imaged at the surface of HeLa cells after 1h of incubation at 4 °C. At this low temperature, the viruses can attach to the surface receptors of the host cells, but their cellular entry via endocytosis is inhibited. The SEM images in Figure6c show that upon exposure to rGOAg, the viral particles were still able to attach to the cells, and no significant difference in their number or morphology was observed. However, since sample preparation for SEM requires numerous washing steps, it is impossible to estimate the fraction of PSV particles lost during sample processing and hence to conclude about any potential decrease in the number of PSVs attached on the cell surface. In any case, the observed viral particles in the rGO-Ag treated samples were not significantly affected compared to non-treated ones and were still bound to the cell surface. Altogether, our data do not evidence any significant binding between rGO-Ag NCs and viral particles, refuting the hypothesis that a loss of infectious viral particles from the Adv. Mater. Interfaces 2023, 10, 2201996 21967350, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202201996 by Technical University In Brno, Wiley Online Library on [04/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com © 2022 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 2201996 (8 of 15) www.advmatinterfaces.de Figure 6. Mechanism of the antiviral effect of rGO-Ag NCs against HIV-1 PSV. a) Quantification of p24 viral antigen concentration in the supernatants by ELISA. b)Immunoblot analysis of integrase protein in the supernatants of PSV treated with increasing concentrations of rGO-Ag. c) SEM images of PSVs incubated in the presence of 100µgmL−1 rGO-Ag or left untreated. d) Confocal images of HeLa cells infected with IN-sfGFP-containing HIV-1 PSV (PSV/IN-gfp) for 2h. Cell nuclei were stained with Hoechst. White arrows indicate intracellular fluorescent PSV particles. e) Quantification of PSV/IN-gfp particles detected in the middle planes of the cells. The fluorescent spots larger than twice the resolution limit (500nm) that correspond to aggregates were excluded from the analyzed data; n= 85 for each condition, the box represents 25–75% of the data, the whisker represents the SD, and the line shows the mean value. f–i) HEK 293T cells or PSV/luc were treated with various concentrations of rGO-Ag or left untreated and Adv. Mater. Interfaces 2023, 10, 2201996 21967350, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202201996 by Technical University In Brno, Wiley Online Library on [04/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com © 2022 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 2201996 (9 of 15) www.advmatinterfaces.de supernatant or the mechanical disruption of the virus by NCs might inhibit viral infection. Next, we used fluorescent HIV-1 PSV particles (PSV/INsfGFP)[42] to examine their cellular entry by fluorescence microscopy. PSV particles were treated with 100 µg mL−1 of rGO-Ag or left untreated. The samples were centrifuged, and HeLa cells were infected with the collected supernatant for 2h to allow the viral particles to enter the cells. The samples were fixed and imaged by confocal microscopy. PSVs expressing INsuperfolder green fluorescent protein (sfGFP) were detected as fluorescent spots in the cytoplasm of the infected cells. The number of detected viral particles was counted in the middle plane of the cells to avoid counting the PSVs at the cell surface. As shown in Figure6d,e, HeLa cells infected with nontreated PSVs contained on average 3.8 cytoplasmic spots, whereas only 1.25 viral particles were detected when the PSVs were preincubated with rGO-Ag (Figure6e). These data suggest that rGO-Ag effectively reduces viral entry into host cells. It has been previously reported that graphene-based nanomaterials show oxidative activity and may trigger reactive oxygen species (ROS) generation and lipid peroxidation in mammalian cells.[27,43] Therefore, the NC-induced alteration of the PSV lipid envelope can potentially affect the binding of PSV to the host cell surface and its cell entry. To test this hypothesis, the level of lipid peroxidation resulting from incubation with rGO-Ag NCs was measured by Liperfluo labeling. Liperfuo dye emits fluorescence specifically in the presence of lipid peroxides. These experiments were first performed on HEK 293T cells since our model PSV particles are enveloped in a lipid membrane that is recovered from the plasma membrane of HEK 293T cells during viral budding. Cells were incubated with 100 µg mL−1 rGO-Ag for 3h, followed by washing and labeling with Liperfluo dye. As shown in the confocal images and corresponding plot in Figure6f,g, the presence of rGO-Ag led to an increase in Liperfluo fluorescence in the cells, indicating that rGO-Ag induced the lipid peroxidation of their membranes. Similarly, HIV-1 PSV particles were incubated with rGO-Ag for 90 min, followed by ultracentrifugation and staining with Liperfluo dye. The PSV particles were then loaded into an Ibidi channel slide and imaged with confocal and epifluorescence microscopes. The mean Liperfluo intensity of individual PSV particles showed a significant increase for the rGO-Ag treated sample (Figure 6h). Interestingly, we also observed very different motility of treated and nontreated viral particles. In the nontreated samples, numerous viral particles were attached to the glass coverslip surface, while in the rGO-Ag treated samples, the majority of the viral particles diffused in the buffer, and only a few of them were attached. Figure6j (and Videos S1 and S2, Supporting Information) represents the trajectories of individual viral particles in the control and rGO-Ag treated samples. Altogether, these results indicate that rGO-Ag induces lipid peroxidation and modifies the physicochemical properties of the viral external membrane. In conclusion, our data suggest that rGO-Ag does not disrupt or remove PSV particles from the samples during the centrifugation step. However, they cause lipid peroxidation, which could alter the properties of the viral and cellular membranes, preventing viral entry. 2.5. Cellular Uptake of rGO-Ag and Stimulation of the Cell Proinflammatory Response It was previously reported that cell membrane interaction and GO-Ag internalization can stimulate cell signaling, which may lead to the inhibition of the viral infection.[17,44] To explore this possible contribution to our observations, we examined the cellular uptake of rGO-Ag by HeLa cells. Since rGO-Ag nanomaterials emit fluorescence,[45] it was possible to visualize them inside the cells by confocal microscopy. The supernatant of rGO-Ag was prepared as mentioned above and incubated with cells for 3 or 24h. Figure7a reveals the presence of fluorescent spots in the rGO-Ag treated cells, while no fluorescence was detected in the control sample. As shown in the higher-magnification image, rGO-Ag was clearly present in the cell cytoplasm. These observations indicate that rGO-Ag is efficiently taken up by the cells and can be detected in the cell cytoplasm after 3h of incubation. Next, we examined whether the preincubation of rGO-Ag with cells had any effect on viral infection. HeLa and THP-1 cells were exposed to supernatants containing different concentrations of rGO-Ag for 3 and 24 h followed by extensive washing to remove all unbound nanoparticles. The cells were then infected with PSV/luc, and their luciferase activity was measured after 24 h. As demonstrated in Figure 7b, a short (3 h) preincubation of HeLa cells with rGO-Ag did not affect viral infection. In contrast, when cells were incubated with rGO-Ag for 24h prior to infection, their luciferase activity was reduced to 68% and 39% for the HeLa and THP-1 cells, respectively (Figure7c,d). Since the PSV was not in contact with the rGO-Ag materials before infection, we suggest that these results show an alternate inhibitory effect of rGO-Ag that occurs most likely during the intracellular steps of the viral cycle. As mentioned earlier, one of the possible mechanisms responsible for rGO-Ag-induced antiviral activity could be related to innate immunity activation. We, therefore, studied the effect of rGO-Ag on the cellular inflammatory response at the transcriptional level. THP-1 and HeLa cells were incubated with supernatants containing rGO-Ag (samples initially containing 50µgmL−1) for 24h. This incubation was followed by total RNA extraction and two-step quantitative reverse transcription–polymerase chain reaction (qRT–PCR) assays. We found that rGO-Ag strongly stimulated the production of proinflammatory cytokine mRNAs, such as TNF-α, IL-8, IL-1β, and IL-6, in the THP-1 and HeLa cells. As demonstrated in stained with Liperfluo dye (final concentration 1µmolL−1) in serum-free media. f) Confocal images of HEK 293T cells stained with Liperfluo dye. g) Representative graph of the mean fluorescence intensities in the cell area; n = 70 for each condition, the box represents 25–75% of the data, and the whisker represents the SD. h) Confocal images of the PSV/luc stained with Liperfluo dye. i) Mean fluorescence intensities of fluorescent spots corresponding to individual viruses; n= 400 for each condition, the box represents 25–75% of the data, the whisker represents the standard deviation (SD), and ***p< 0.001. j) Trajectories of rGO-Ag-treated or untreated PSV/luc. Adv. Mater. Interfaces 2023, 10, 2201996 21967350, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202201996 by Technical University In Brno, Wiley Online Library on [04/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License