Biomimetic Copper Nanoparticles Coated with ACE2‐Overexpressing Membranes for Selective SARS‐CoV‐2 Neutralization and Disinfection

ABSTRACT

The SARS‐CoV‐2 spike protein facilitates viral entry into host cells by binding to the human angiotensin‐converting enzyme 2 (ACE2) receptor. To exploit this mechanism for therapeutic intervention, a liposome fusion‐induced membrane exchange (LIME) strategy to generate biomimetic membrane‐integrated liposomes (MILs) from ACE2‐overexpressing mammalian cells was developed. Using engineered HeLa cells as a model, MILs have been successfully harvested that retained native surface proteins, including ACE2, as confirmed by immunogold TEM and Western blot analysis. These ACE2‐presenting MILs were then coated onto copper nanoparticles (Cu NPs), creating biomimetic Cu@MIL nanostructures with dual functions, including selective viral capture via ACE2‐mediated binding and neutralization, as well as potent antiviral activity from Cu NP disinfection. This synergistic platform effectively camouflages the nanomaterial with host‐mimetic membranes, conferring targeted viral neutralization and disinfection capabilities. Our findings highlight the potential of Cu@MIL nanoparticles as a decoy‐plus antiviral therapeutic for SARS‐CoV‐2, offering a promising strategy to combat COVID‐19 and future pandemics of receptor‐specific pathogens.

By fusing ACE2‐overexpressing membrane‐integrated liposome (MIL) with copper nanoparticles, the biomimetic Cu@MIL nanostructures were created that directly hijack the SARS‐CoV‐2 entry pathway. These particles combine potent virus‐targeting precision with intrinsic antiviral activity, achieving rapid neutralization and disinfection. This breakthrough strategy opens a powerful avenue for effective COVID‐19 therapy.

1. Introduction

The coronavirus disease 2019 (COVID‐19) was caused by the infectious causative agent named severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2). As a result, over seven million deaths worldwide occurred during the spread period [1]. SARS‐CoV‐2 is a type of RNA beta coronavirus and contains a single‐stranded RNA genome. The coronavirus gets its name from the crown‐like structure, which contains spike (S) proteins. The trimeric S protein is a class 1 fusion protein cleaved into soluble S1 and S2 subunits that are non‐covalently associated. The S1 subunit contains a receptor‐binding domain (RBD), which helps virus conjugation with infected lung epithelial cells by binding to the angiotensin‐converting enzyme 2 (ACE2) receptor of the cell surface [2, 3, 4]. In contrast, the S2 subunit acquires a fusion mechanism to integrate the viral membrane with the host cell membrane. As a result, this crown structure, with the binding feature of the ACE2 receptor, is typically the focus of research aimed at developing novel vaccine and medication candidates [5].

The physiological function of the ACE2 receptors is to degrade angiotensin II. Existing excess angiotensin II by downregulating ACE2 receptors after SARS‐CoV‐2 infection leads to systemic hypercytokinemia, serious pulmonary inflammation, and the deadly complications of COVID‐19 disease [6, 7]. Recent research has shown that some approaches can block the viral interaction with ACE2 entry receptors on host cells, including neutralizing S‐specific antibodies [8, 9], and rhACE2 [10, 11], thereby preventing COVID‐19 infection. Although many high‐affinity monoclonal antibodies have already been identified in recovered patients to treat COVID‐19 disease, some have not shown favorable efficacy for hospitalized patients [12, 13], and others have lost efficacy against new variants, such as the delta variant [14]. Monoclonal antibodies targeting specific epitopes of SARS‐CoV‐2 antigens have limited efficacy in controlling current and upcoming mutant strains [15, 16].

Recently, copper‐based nanoparticles have been widely used in nanomedicine, food preservation, coatings, and the textile industries. Copper nanoparticles (Cu NPs) in various forms, including CuO, Cu₂O, CuS, CuI, and Cu⁰, exhibit antimicrobial and antiviral properties [17]. In contrast to antibiotics, copper influences multiple cellular or viral targets, leading to membrane and capsid instability, reactive oxygen species (ROS) generation, protein degradation, enzyme inhibition, iron‐sulfur (Fe–S) center displacement, and genotoxic effects [18, 19, 20, 21, 22]. In many cases, viruses are mainly prone to copper‐mediated toxicity because viruses lack a metal response or repair mechanisms, such as metal‐induced persistence mechanisms that exist in bacteria (Table S1) [23]. Considering the ongoing and future pandemics induced by diverse variants, the use of copper nanoparticles has emerged as a potential strategy to overcome viral transmission [24].

Biomimetic nanoparticles are generated via membrane‐nanoparticle fusion, yielding a hybrid membrane‐core interface that preserves the native composition, lateral fluidity, and functional proteins of the source cell membrane. This hybrid architecture enables nanoparticles to inherit key biological functions, such as specific targeting and immune‐evasion behavior [25]. This approach offers key advantages, including excellent biocompatibility, prolonged circulation, targeted delivery, and improved tissue penetration [26, 27, 28, 29]. This membrane camouflage enables precise delivery of therapeutic agents to diseased sites, with broad applications in infectious diseases, cancer, and tissue repair [30, 31, 32, 33]. Wang and co‐workers developed erythrocyte‐cancer hybrid membrane‐coated copper sulfide nanoparticles with enhanced tumor targeting and inhibition [34]. In another study, biomimetic polymer nanoparticles delivered membrane antigens to antigen‐presenting cells, thereby activating the immune response against the tumor [35]. More types of biomimetic nanomaterials for various applications were shown in Table S2. Leveraging these traits, biomimetic nanoparticles hold strong potential for selective antiviral therapy.

In this study, the novel liposome fusion‐induced membrane exchange (LIME) strategy was successfully applied to the mammalian cell system (engineered HeLa cells overexpressing ACE2), enabling the large‐scale production of bioactive membrane‐integrated liposomes (MILs) that retain their original surface characteristics (ACE2 expression) (Scheme 1). This method was previously demonstrated to successfully extract outer membrane components from Shewanella oneidensis MR‐1, preserving their native proteins and exhibiting remarkable electrochemical activity [36]. These ACE2‐presenting MILs were then used to camouflage Cu NPs, thus yielding Cu@MIL nanostructures. By integrating the virus‐targeting decoy capability of host‐derived membranes with the intrinsic antiviral properties of copper nanoparticles, this biomimetic nanoplatform can effectively neutralize and disinfect SARS‐CoV‐2, offering a promising strategy for COVID‐19 treatment and broader antiviral applications, such as antiviral coatings, biosensing, and medical masks.

SCHEME 1.

Schematic diagram of biomimetic copper nanoparticles coated with ACE2‐expressed membrane‐integrated liposome secreted from the engineered HeLa cells, and its antiviral application in selective SARS‐CoV‐2 neutralization and disinfection.

2. Results and Discussion

2.1. ACE2‐Overexpression on Engineered Mammalian Cells

In the present study, normal HeLa and engineered ACE2‐overexpressing HeLa cancer cells were selected as the cell models for further harvesting their extracellular vesicles (EVs), which would be used for biomimetic nanodrug fabrication and evaluation of their application for SARS‐CoV‐2 neutralization and disinfection. The engineered mammalian cell with specific expression of ACE2 receptor was successfully achieved on the HeLa cancer cell system by Wang et al. in 2004 [37]. The primary reason for using HeLa cells with very low ACE2 expression is for subsequent experimental comparisons. Immunofluorescence staining data confirmed the low ACE2 status in the parental HeLa cell, where a profound fluorescent signal was observed in the engineered HeLa cell line, reflecting its overexpression of the ACE2 receptor (Figure 1a).

FIGURE 1.

The scalable production of the mammalian cell‐derived vesicles. (a) Immunofluorescence image of parental HeLa and ACE2‐HeLa cell line. (Red: ACE2; Blue: Hoechst 33342; (scale bar = 50 µm). (b) Schematic illustration of the approach to MIL preparation using the LIME technique. (c) The picture of pellet collected after ultracentrifugation method using biogenic process and LIME technique (white dotted circle) [groups: (i) biogenic HeLa EVs, (ii) HeLa MILs, (iii) biogenic ACE2‐HeLa EVs, (iv) ACE2‐HeLa MILs]. (d) Transmission electron microscopy (TEM) images of HeLa MILs (left) and ACE2‐HeLa MILs (right). (e) Hydrodynamic size distribution of EVs and MILs (n = 3). (f) Zeta potential of EVs and MILs (n = 3). (g) Protein concentration of EVs and MILs. (h) The protein‐index yield of ACE2‐HeLa MILs as a function of liposome‐dependent volume (n = 3). (i) Cell viability of HeLa cells treated with positive control (w/o FBS), negative control (w/ FBS), 200, 500, 1000, and 1500 µL of liposome. (^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; n.s. = no significance).

2.2. Scalable Production of Cell‐Derived Vesicles

Considering the biosafety concerns associated with utilizing cancer cells, the purified extracellular vesicles with complete membrane characteristics were prioritized for antiviral applications over the direct use of cells or cell lysates. However, the yield of biogenic EVs secreted from host mammalian cells is too low, resulting in an unfavorable development of the related applications [38]. Here, a highly fluid liposome composed of 80% DOPC and 20% DOPE was applied to promote vesicle production from cells, inspired by our previous study, which indicated a feasible strategy for scalable vesicle production, termed the liposome fusion‐induced membrane exchange (LIME) process (Figure 1b) [36]. The uniform spherical 100 nm liposomes were successfully fabricated through the regular extrusion method (Figure S1). Excitingly, the significant pellet products from the ultracentrifuged supernatant of HeLa and engineered HeLa cell cultures with highly fluid liposome co‐incubation for 24 h were obtained, implying the initially successful LIME treatment for enlarging the vesicle secretion (Figure 1c). A clear comparison, with no visible pellet on the bottom of the control tube, showcases the limited biogenic vesicle production from the cell culture without liposome co‐incubation. Considering the essential differences of vesicular products from biogenic and LIME‐mediated pathways, the LIME‐based artificial vesicle is collectively called a membrane‐integrated liposome (MIL).

The transmission electron microscope (TEM) images reveal the HeLa and engineered HeLa cell‐derived MILs, respectively abbreviated as HeLa MILs and ACE2‐HeLa MILs, with a clear vesicular structure (Figure 1d). Whereas the TEM images of HeLa EVs and ACE2‐HeLa EVs exhibit a typical spherical morphology (Figure S2). Additionally, the hydrodynamic particle diameters of ACE2‐negative and ACE2‐positive MILs are 129.1 ± 7.1 and 127.4 ± 7.8 nm, respectively, as determined by a dynamic light scattering (DLS) analyzer (Figure 1e). Moreover, a significant shift in zeta potential was observed from the crude liposomes to both types of MILs, indicating the incorporation of biological membrane proteins onto the MIL surface after the LIME process (Figure 1f). The amount of proteins in the biogenic vesicle, called EVs, and MILs from both cell lines was quantified using a protein assay and the standard BSA‐based calibration curve (Figure S3). The protein concentrations in HeLa MILs and ACE2‐HeLa MILs are 1.1 and 1.3 mg/mL, respectively (Figure 1g). Both MIL samples contained higher protein levels than biogenic EVs, confirming that the alternative LIME approach can yield more cell‐derived vesicles than the biogenic process. Consistently, nanoparticle tracking analysis (NTA) revealed a more than 10‐fold increase in particle concentration in the MIL samples compared to the corresponding biogenic EV samples derived from equal cell numbers and 24 h of incubation (Figure S4), aligning with the higher yields of MILs indicated by protein quantification. The protein concentration was clearly dependent on the liposome volume (Figure 1h). Compared with the control group, which showed almost negligible protein levels, a gradual increase in protein concentration was observed as the liposome volume increased from 200 µL to 1,500 µL. Notably, no further increase in protein concentration occurred after adding 2,000 µL of liposome, implying that the membrane protein production capacity had reached saturation. Furthermore, cell viability was evaluated using the MTT assay to ensure that MIL production via the FBS‐free LIME technique, using various liposome concentrations, did not harm HeLa and engineered HeLa cells (Figure 1i). All groups showcased very high viability and limited effects on cell growth (less than 20%), indicating that the process is low‐cytotoxic and that the resulting MILs are not apoptotic bodies or damaged cell fragments [36].

2.3. Biomimetic Activities of ACE2‐Expressing Membrane‐Integrated Liposomes

To confirm the surface presentation of ACE2 on the engineered HeLa cell‐derived MILs, immunogold TEM was employed using anti‐ACE2 primary antibody and 10 nm secondary antibody‐labeled gold nanoparticle (Figure 2a). The results demonstrate the successful binding of gold nanoparticles to the ACE2‐HeLa MILs, evidencing the presence of ACE2 receptors on the MIL surface, which is essential for their function as a biomimetic decoy nanocarrier targeting SARS‐CoV‐2. Although ACE2‐HeLa‐derived biogenic EVs exhibit excellent ACE2 surface expression, their limited yield still hinders further biotechnological applications. Sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) analysis was then applied to reveal distinct protein profiles among the tested samples (Figure 2b). ACE2‐HeLa MILs (lane 5) exhibited a clear band corresponding to the ACE2 protein (105–110 kDa) [16]. Notably, the ACE2 protein band was absent in both the HeLa cell lysate (lane 2) and HeLa MILs (lane 4). Using western blotting, the presence of ACE2 protein was confirmed in ACE2‐HeLa cell lysates, biogenic EVs, and MILs, indicating successful incorporation of ACE2 proteins into MILs generated from ACE2‐expressing cells. (Figure 2c).

FIGURE 2.

The evaluation of the biomimetic activities of MILs. (a) Immunogold TEM showing ACE2 expression on ACE2‐HeLa‐derived biogenic EVs as well as on ACE2‐HeLa MILs. Red arrows indicate gold‐labelled antibodies bound to ACE2 on the surface of the vesicles. (b) SDS‐PAGE was used to investigate the protein profile of the cell lysates and MIL samples derived from HeLa and ACE2‐HeLa cells. (c) Western Blotting was used to confirm the ACE2 expression in the cell lysates, biogenic EVs, and MILs samples.

2.4. Antiviral Copper Nanoparticles

Based on the superior antiviral activity of copper, stable copper nanoparticles with hydrophobic TOPO and ODA surface coatings (Cu@TOPO/ODA NPs) were synthesized using the thermal decomposition method for subsequent application in the present study [39]. The observation of Cu@TOPO/ODA NPs in the TEM image indicated the uniform spherical morphology with narrow size distributions of 96.9 ± 11.7 nm (Figure 3a). The hydrophobic Cu@TOPO/ODA NPs in the oil phase were then transferred into the water phase through a ligand exchange reaction, in which the hydrophobic surfactants were replaced by hydrophilic polyvinylpyrrolidone (PVP), thus formatting the Cu@PVP NPs hydrated in the aqueous phase (Figure 3b). The TEM image of Cu@PVP NPs revealed no obvious changes in morphology and particle size after the surfactant replacement process, highlighting the excellent stability of this copper colloid. Ultraviolet–visible (UV–vis) spectroscopy was then applied to measure the characteristic absorption peak of Cu@TOPO/ODA and Cu@PVP NPs at 585 nm, indicating the high optical stability of the resulting Cu colloids with different surfactants (Figure 3c). Moreover, DLS analysis further determined the hydrodynamic diameters of Cu@TOPO/ODA and Cu@PVP to be 110.0 ± 3.6 and 122.3 ± 5.4 nm, respectively, reflecting the enlarged hydration range resulting from the PVP polymer deposition on the surface of the NPs (Figure 3d). Zeta potential measurements revealed that Cu@TOPO/ODA NPs had a strong positive surface charge (+35 mV), while Cu@PVP NPs exhibited a negative potential (−26 mV) (Figure 3e). Fourier‐transform infrared (FTIR) spectroscopy was employed to analyze the chemical composition of Cu@TOPO/ODA and Cu@PVP NPs (Figure 3f). The FTIR peak at 1156 cm⁻¹ is attributed to the P═O bond from TOPO, whereas the peaks at 1290 and 2853–2930 cm⁻¹ contributed to the C─N and C─H stretching of ODA, respectively. The peaks at 1290 and 1720 cm⁻¹ corresponded to the C─N and C═O bonds of PVP, indicating the successful phase transfer of Cu NPs. In their cases, a distinct peak near 538 cm⁻¹ was observed for Cu─O bonds, thereby confirming the successful synthesis of Cu‐based NPs. X‐ray diffraction (XRD) analysis was performed to determine the crystal composition of Cu@TOPO/ODA and Cu@PVP NPs. The diffraction patterns of both matched well with that of pristine Cu (JCPDS 04–0836), indicating high crystalline purity and no crystal changes after surface modification (Figure 3g) [40]. In addition, the reservation of a single Cu@PVP NPs in the high‐resolution transmission electron microscope (HR‐TEM) image showed the featured (200) lattice fingerprint of copper and homogenous copper element distribution, while its electron diffraction result indicated a single crystal characteristic with clear (111) and (200) diffraction spots (Figure 3h–j). Moreover, X‐ray photoelectron spectroscopy (XPS) further confirmed the copper configuration of Cu@TOPO/ODA and Cu@PVP NPs (Figures S5 and S6) [39]. The Cu⁰/Cu²⁺ ratios of Cu@TOPO/ODA and Cu@PVP NPs are 78.5/21.5% and 54.8/45.2%, respectively, indicating that the inevitable copper oxidation occurred during the ligand exchange process.

FIGURE 3.

Characterizations of the copper colloids. TEM images of (a) Cu@TOPO/ODA and (b) Cu@PVP NPs. The inserted photos are their colloidal images in a two‐phase solution (Top phase: toluene; bottom phase: water). (c) Ultraviolet–visible (UV–vis) spectra, (d) Hydrodynamic diameter measurement, (e) Zeta potential analysis, and (f) FTIR spectra of Cu@TOPO/ODA and Cu@PVP NPs. (g) X‐ray diffraction pattern of copper standard (JCPDS no. 04–0836), Cu@TOPO/ODA, and Cu@PVP NPs. (h) High‐resolution TEM imaging of a single Cu@PVP NP with Cu element analysis, (i) Area‐selected high‐resolution observation, and (j) Electron diffraction pattern along with (001) zone axis. The yellow and white frames indicated the observed regions for element analysis and area‐selected high‐resolution TEM imaging, respectively.

2.5. Membrane Encapsulation of Antiviral Copper Nanoparticles

A biomimetic Cu colloid, designed as a decoy‐plus antiviral therapeutic for SARS‐CoV‐2, was fabricated by coating a hydrophilic copper core with ACE2‐presenting MILs. While the conventional extrusion method can achieve membrane coating by forcing nanoparticles through a membrane filter, where the fluidity of the cell membrane and mechanical shear promote phospholipid encapsulation. However, this approach is labor‐intensive and time‐consuming [41]. As an alternative, a cyclic sonication‐freezing method was employed, yielding a satisfactory and cost‐effective result [36]. In this process, ultrasonic waves disrupt the vesicular membrane of MIL into fragments, which spontaneously self‐assemble around the Cu nanoparticles under low‐temperature conditions, resulting in the large‐scale fabrication of MIL‐encapsulated Cu colloids (Cu@MIL NPs) via a rapid and facile strategy.

TEM imaging confirmed that ACE2‐HeLa Cu@MIL NPs exhibited a uniform spherical morphology with a size distribution of 110.7 ± 8.5 nm (statistical analysis of over 100 Cu@MIL particles for average diameter calculation) (Figure 4a). The apparent size variation observed in the TEM image can be attributed to differences in particle orientation, membrane deformation during sample drying, and local contrast variations arising from the thin MIL coating. These effects are commonly observed in membrane‐coated nanoparticles and do not reflect the intrinsic non‐homogeneity of the prepared sample [42]. DLS showed that the average hydrodynamic diameter increased from 122.3 ± 5.4 to 140.6 ± 14.6 nm, and the size distribution remained uniform after membrane coating (Figure 4b; Figure S7). Furthermore, from the comparison of particle size obtained by TEM and DLS, the hydrodynamic size is slightly larger than the TEM‐derived particle size, because of the presence of a hydration layer and surface coating in the dispersed state in solution. Zeta potential measurements revealed that a negative shift in surface charge was observed after ACE2‐HeLa MIL encapsulation of Cu colloids (Figure 4b). Notably, the zeta potential of ACE2‐HeLa Cu@MIL NPs reached −16.3 mV, closely matching that of MIL alone, suggesting full membrane coverage with consistent surface characteristics between MIL and Cu@MIL NPs, which was identified as the optimal condition. No significant difference in optical features at 585 nm between Cu@PVP and Cu@MIL NPs, indicating no adverse effects to the copper core during the sonication‐mediated membrane coating process (Figure S8).

FIGURE 4.

Encapsulation of antiviral copper core with ACE2 membrane. (a) TEM image of Cu@MIL NPs. The inserted figure shows an enlarged NP, marked with a white frame. (b) Hydrodynamic diameter and zeta potential of Cu NPs before and after MIL encapsulation. (c) SDS‐PAGE analysis of membrane protein profile from cell lysate, Cu@MIL, and Cu@PVP NPs. (d) FTIR spectra of ACE2‐HeLa MIL and Cu@MIL NPs.

Additionally, XPS analysis further verified the successful MIL coating (Figure S9). The P 2p spectrum of Cu@PVP showed noisy features, whereas Cu@MIL displayed a distinct peak at 133.2 eV, indicating the presence of phospholipids. SDS‐PAGE analysis confirmed the successful integration of major bioactive components (proteins), of ACE2‐HeLa, including vimentin and prekeratin [43], into Cu@MIL NPs (Figure 4c). Consistent with this, FTIR spectra of MIL and Cu@MIL NPs showed characteristic CH₂ stretching vibrations at 2921 and 2853 cm⁻¹, a C═O stretching band at 1650 cm⁻¹ attributed to the presence of proteins, and a broad band between 1050 cm⁻¹ corresponding to phospholipid vibrations (Figure 4d) [36]. The stability of MIL‐encapsulated Cu NPs dispersed in PBS was further evaluated by monitoring hydrodynamic diameter, nanoparticle dispersion images, and TEM over 7 days, demonstrating good colloidal stability under physiological conditions (Figure S10).

2.6. Selective SARS‐CoV‐2 Neutralization and Disinfection

Before evaluating the therapeutic potency of the biomimetic antiviral copper nanoparticles, the biosafety profile was first assessed. The MRC‐5 and 3T3 cells were treated with Cu@PVP and Cu@MIL NPs for 24 h, respectively. These treatments did not exhibit any obvious cytotoxic effects up to a 20‐ppm concentration of Cu on the normal cell lines, indicating their potential safety and good biocompatibility (Figure 5a). To evaluate the therapeutic potency of the biomimetic antiviral copper nanoparticles, a lentivirus‐based SARS‐CoV‐2 spike protein expression pseudo‐virus system expressing spike protein (Omicron variants, BA.2) and GFP was used. The TEM image of spike protein‐expressing SARS‐CoV‐2 pseudo‐virus is shown in Figure S11. The results showed that the pseudo‐virus could selectively infect ACE2‐overexpressing ACE2‐HeLa cells in a dose‐dependent manner, whereas the parental HeLa cells with low ACE2 expression were difficult to infect (Figure 5b,c). To more precisely calculate the IC₅₀ value, a higher viral load was used (10 µL virus per reaction) to boost the infection rate. The data showed that ACE2‐HeLa Cu@MIL nanoparticles effectively block pseudovirus infection in a dose‐dependent manner (the IC₅₀ value was 2.09 ppm) (Table S3), when about 90% infection inhibition was shown at concentrations of 3.75 ppm (Figure 5d,e). On the other hand, the ACE2‐Hela MIL showed partial neutralization capability (about 40% infection inhibition) at higher concentration (5 µg/mL which equivalent to membrane protein amount of 10 ppm Cu@MIL NPs) but no significant inhibition was found at lower concentration (1.67 µg/mL which equivalent to membrane protein amount of 3.34 ppm Cu@MIL NPs) (Figure S12). Lee and co‐workers demonstrated the biodistribution, accumulation, and clearance of Cu NPs and Cu ions in their study. The Cu NPs showed systemic uptake, and their levels in blood and tissues rose gradually [44]. The cytotoxicity assay in the current study showed that Cu@PVP and Cu@MIL were safe up to 20 ppm Cu for normal cells, whereas in the pseudovirus assay, a much lower concentration (3.75 ppm Cu) of Cu@MIL NPs was used to neutralize the virus, thereby confirming that our Cu@MIL is highly biocompatible at lower concentrations and demonstrating its clinical potential for disinfecting SARS‐CoV‐2.

FIGURE 5.

The therapeutic potential of the biomimetic antiviral copper nanoparticles. (a) viability of (left) MRC5 and (right) 3T3 cells after exposure to Cu@PVP and Cu@MIL NPs at different concentrations of dosages for 24 h. (b) The results of the infection were determined by analyzing the ratio of GFP‐expressing cells after infection. (Scale bar: 100 µm) (c) The infection rate of parental HeLa and ACE2 overexpressing ACE2‐HeLa cells was quantified. (d) The therapeutic potential of the biomimetic antiviral copper nanoparticles was evaluated (10 µL of pseudo‐virus was used for infection per reaction). (Scale bar: 100 µm) (e) The therapeutic potential of the biomimetic antiviral copper nanoparticles was quantified. (^* p < 0.05; ^** p < 0.01; ^**** p < 0.0001; n.s. = no significance).

Compared with recent anti‐SARS‐CoV‐2 nanomaterials, the central distinction of the Cu@MIL system lies in its biomimetic and dual‐mode antiviral design (Table S4). Unlike single‐ligand NP systems, ACE2‐presenting MILs provide a multivalent receptor array that enhances spike protein binding through cooperative interactions, more closely mimicking the native host cell membrane [16]. In parallel, the Cu NP core contributes an additional receptor‐independent antiviral mechanism via copper‐induced viral inactivation. Mechanistically, SARS‐CoV‐2 is particularly susceptible to copper‐mediated toxicity due to the absence of effective metal‐response and repair pathways. Copper nanoparticles can inactivate viruses through both dissolution‐independent mechanisms, in which stable NPs directly interact with viral components, and ion‐mediated pathways, where released Cu⁺/Cu²⁺ induces capsid damage, protein denaturation, and viral genome degradation [45, 46]. In addition, Cu⁺ catalyzes the generation of reactive oxygen species, including superoxide and hydroxyl radicals, to further contribute to viral inactivation [47]. These complementary mechanisms together provide clear proof of principle for the proposed antiviral strategy. In the future, the novel LIME technique shows considerable potential to extract various MILs from other engineered or target cells and coat them onto specific nanoparticles of interest to target other disease types.

Beyond the current proof‐of‐concept antiviral model, this strategy also offers several broader and concrete antiviral applications. First, MIL‐coated nanomaterials could be incorporated into antiviral surface coatings, where virus‐binding receptors displayed on the MIL membrane enable active virus capture rather than passive inactivation alone [48]. Second, MILs derived from receptor‐engineered cells may serve as functional recognition elements for rapid viral prescreening and biosensing platforms, in which receptor‐virus binding events can be translated into optical or electrochemical‐based readouts [49]. Third, the scalability of the LIME technique enables integration of MIL‐functionalized nanomaterials into antiviral personal protective equipment, such as next‐generation face masks or filter layers, where the MIL membrane provides selective viral adsorption while the inorganic core contributes additional antiviral activity [50, 51]. From a translational perspective, cost is a major differentiating factor. Based on our current laboratory‐scale estimation, the production cost of Cu@MIL is approximately 8.5 USD/mg, which will be significantly decreased after scale‐up production and will be substantially lower than reported costs for monoclonal antibodies (9‐300 USD/mg) and will be similar to that of the protein‐functionalized nanosystems (∼10–40 USD/mg) (Table S5). This cost advantage, together with scalable MIL production enabled by the LIME technique, supports the feasibility of Cu@MIL for high‐volume applications, such as antiviral coatings, filtration materials, and disposable protective devices.

3. Conclusion

In the present study, the scalable production of vesicles (MIL) from an ACE2‐expressing HeLa cell line using the LIME process marked the first successful application of LIME technology in a mammalian cell system. On the other side, spherical shape hydrophobic Cu NPs were prepared, and ligand exchange as well as surface modification were successfully performed to obtain water‐soluble Cu@PVP NPs with antiviral properties. Next, biomimetic nanoparticles were prepared by camouflaging ACE2‐HeLa MILs on Cu@PVP NPs, thus termed as Cu@MIL NPs. These ACE2‐presenting Cu@MIL NPs exhibit excellent virus‐targeting capabilities of MIL and tremendous antiviral properties of copper, enabling them to neutralize and disinfect the SARS‐CoV‐2 virus, thereby presenting an effective approach for COVID‐19 therapy.

4. Experimental Section

4.1. Materials

Copper (I) bromide (CuBr, 98%) was purchased from Alfa Aesar. Octadecylamine (ODA, C₁₈H₃₅N, 90%) was obtained from Acros Organics. Trioctylphosphine oxide (TOPO, C₂₄H₅₁OP, 98%) was bought from Thermo Scientific. Oleylamine (C₁₈H₃₇N, >80%) was obtained from Macklin. Cetyltrimethylammonium bromide (CTAB, C₁₉H₄₂BrN, ≥98%), 3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide (MTT), fetal bovine serum (FBS), penicillin‐streptomycin (P‐S), and polyvinylpyrrolidone (PVP, (C₆H₉NO)n, M.W. = 55 000) were purchased from Sigma–Aldrich. 1,2‐Dioleoyl‐sn‐glycero‐3‐phosphocholine (DOPC, C₄₄H₈₄NO₈P, >99%) and 1,2‐dioleoyl‐sn‐glycero‐3‐phosphoethanolamine (DOPE, C₄₁H₇₈NO₈P, >99%) were obtained from Avanti. Bovine serum albumin (BSA) was purchased from Bio Basic. Dulbecco's Modified Eagle Medium (DMEM) was obtained from GeneDireX. Deionized water from a Millipore purification system was used in all experiments.

4.2. Characterizations

Transmission electron microscopy (TEM, Hitachi, H‐7500) and High‐resolution transmission electron microscopy (HR‐TEM, JEOL, JEM‐2100F) were used to observe the morphology of the NPs. An X‐ray diffractometer (XRD, Bruker, D8 ADVANCE) was used to determine the crystallographic structure of Cu‐based NPs. The copper concentration of the materials was measured using an atomic absorption spectrometer (AA, Thermo Scientific, iCE 3000 series). Ultraviolet–visible spectrometry (UV‐vis, Analytik Jena, Specord 200 Plus) was used to characterize the optical properties of the samples. X‐ray photoelectron spectroscopy (XPS, Kratos, Axis ultra DLD) was used to determine their elemental composition, chemical states, and oxidation level. The dynamic light scattering analyzer (DLS, Otsuka Electronics, ELSZ‐2000) was used to measure the zeta potential and the hydrodynamic diameters of all NPs. A Fourier‐transform infrared spectroscopy (FTIR, Bruker, Alpha 1) was used to determine the vibration spectra of each sample. To assess cytotoxicity and protein concentration, a microplate reader (Thermo Scientific, Multiskan SkyHigh) was used. A nanoparticle tracking analyzer (NTA, Malvern Instruments Ltd, NanoSight LM10) was used to detect the particle number of each colloidal solution.

4.3. Synthesis of Copper Nanoparticles

A fixed amount of 0.05 g of CuBr, 1.0 g of TOPO, and 0.08 g of ODA were mixed with 20 mL of oleylamine in a round‐bottom flask. The reaction mixture was degassed by continuous vacuum pumping for at least 30 min. In the next step, the apparatus was filled with argon and heated in a heating jacket to 300°C at a rate of 20°C/min, maintained for 10 min, until spherical TOPO/ODA‐modified Cu NPs (Cu@TOPO/ODA NPs) formed. Afterward, the reaction solution was cooled to room temperature and centrifuged at 8 000 rpm for 5 min. The supernatant was discarded, and the Cu@TOPO/ODA NPs were washed three times with toluene, each time centrifuging at 8 000 rpm for 10 min. The resulting Cu@TOPO/ODA NPs were stored in oleylamine to prevent oxidation during prolonged storage [39].

4.4. Ligand Exchange of Copper Nanoparticles

The aqueous phase was prepared by dissolving 0.6 g of CTAB and 0.06 g of PVP in 20 mL of deionized water. For the oil phase, Cu@TOPO/ODA NPs were dispersed in 100 µL of toluene at a copper concentration of 10,000 ppm. The oil phase was then added dropwise to the aqueous phase under sonication for 20 min. During this process, the hydrophobic TOPO and ODA on the surface of NPs were gradually replaced by hydrophilic PVP, thus resulting in the production of PVP‐modified Cu NPs (Cu@PVP NPs) and transferring the NPs into the aqueous phase [39]. Afterward, the products were centrifuged at 8,000 rpm for 10 min. The supernatant was discarded, and the Cu@PVP NPs were washed three times with ethanol. Finally, the resulting Cu@PVP NPs were stored in water for use in subsequent experiments.

4.5. Preparation of Liposomes

The stock solution of DOPC and DOPE was prepared in chloroform at a concentration of 10 mg/mL and stored at −20°C. Initially, the DOPC and DOPE were mixed in a volume ratio of 0.504 and 0.118 mL. The mixed solution was dried using nitrogen purging, resulting in a multilayered lipid film at the bottom of the glass vial. Then, the lipid layers were homogeneously dispersed in 800 µL of PBS and thoroughly mixed, followed by a standard extrusion treatment using an extruder (Avanti) equipped with polycarbonate filters of 100 nm pores to produce homogeneous liposomes. The resulting 100 nm liposomes were stored in PBS at 4°C for further experimental use.

4.6. Cell Culture

The human cervical cancer (HeLa) cell line and the engineered ACE2‐overexpressed HeLa cell line were purchased from the Japanese Collection of Research Bioresources. The cells were cultured in DMEM supplemented with 10% FBS and 1% penicillin‐streptomycin at 37°C in a humidified atmosphere with a 5% CO₂ environment.

4.7. Preparation of Membrane‐Integrated Liposomes

The liposome fusion‐induced membrane exchange approach was applied to get the MIL for the target cells. Normal HeLa and ACE2‐overexpressed HeLa cells, at a cell density of 1 × 10⁶, were seeded and cultured in a 10‐cm cell culture dish for 24 h of incubation. After that, the medium was removed and the cell dish was washed with PBS in triplicate. Then, 8 mL of FBS‐free DMEM medium containing 200, 500, 1,000, 1,500, and 2,000 µL of liposomes (2.2 mg/100 µL) was added to the cell dish, followed by gentle shaking for an additional 24 h of incubation. During this process, the MILs were gradually secreted from the cell. Afterward, the cell medium containing MILs was carefully collected and filtered through a 0.45 µm syringe filter (Millipore) to remove the suspended cell bodies and side products. The filtered supernatant containing MILs was then subjected to ultracentrifugation at 50,000 rpm at 4°C for 4 h. The collected pellet was dispersed in PBS and stored at −20°C for further use in subsequent experiments.

4.8. Preparation of Membrane‐Coated Cu Nanoparticles

The Cu@PVP NPs (200 ppm at copper concentration) dispersed in 500 µL PBS were mixed with 400 µL of the solution of MILs (1 mg/mL) derived from HeLa and ACE2‐overexpressing HeLa cells. Then, the mixture solution was sonicated for 10 min to disassemble the vesicular MILs, followed by freezing in an ice bath for 5 min to self‐assemble the MILs component onto the surface of the Cu@PVP NPs, thus producing the MIL‐coated Cu NPs. This sonication‐freezing process was repeated in triplicate to ensure the full encapsulation of Cu@PVP NPs by MILs. After that, the Cu@MIL NPs were collected by centrifugation at 13,000 rpm for 5 min and washed twice with PBS solution. Furthermore, the Cu@MIL NPs were again centrifuged twice at 5,000 rpm for 10 min. The resulting products were dispersed in PBS and stored at 4°C for further use in subsequent experiments.

4.9. Protein Quantification

The protein concentration of MILs was determined using a Bio‐Rad protein assay kit (Bio‐Rad Laboratories). The as‐prepared BSA solutions with known concentrations were applied for calibration curve preparation. The Bio‐Rad working solution was prepared by diluting the stock solution 5 times. Afterward, 200 µL of Bio‐Rad solution was added to 5 µL of BSA standard solution or the MILs solution with an unknown concentration, and then the mixture was incubated for 10 min. Finally, the optical density at 570 nm of each sample was measured by a microplate reader. The unknown concentration of MILs was further determined by using a calibration curve, where protein concentration was plotted as a function of optical density.

4.10. Protein Electrophoresis

A sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) assay was performed to separate and identify the specific proteins from the samples. Initially, the electrophoresis tank was filled with Tris‐MOPS‐SDS running buffer. A commercially available SDS gel (4%–12% PAGE) was put inside the tank. In the next step, 4 µL of 6X Laemmli SDS sample buffer was mixed well with 20 µL of the sample solutions of MILs, Cu@PVP NPs, and Cu@MIL NPs, and all the mixtures were incubated in a dry bath at 95°C for 5 min. After that, 10 µL of the sample mixture was loaded into the SDS‐PAGE gel, and the electrophoresis process was conducted at 120 V for 40 min. Afterward, the gel was washed several times with deionized water to remove the SDS buffer, and then stained with Brilliant Blue R for 20 min to visualize whole proteins on the gel. Finally, the image of the gel with clear protein bands was taken after the gel was completely washed with deionized water.

4.11. Immunogold TEM Imaging

For TEM analysis, before loading with the samples, the nickel carbon‐coated grids (Electron Microscopy Sciences) were subjected to glow discharge treatment. The samples were loaded onto the grids and blocked with Basic Blocking Solution (Electron Microscopy Sciences). The samples were then stained with anti‐ACE2 antibody (R&D Systems). After washing with the Basic Blocking Solution, the samples were stained with a 10 nm nanogold‐conjugated secondary antibody (Abcam), and again washed with Basic Blocking Solution, then the samples were fixed with 2.5% glutaraldehyde (Sigma–Aldrich). After washing with deionized water, the samples were stained with 2% uranyl acetate (Merck & Co., Inc.). The samples were then air‐dried and examined with a JEM‐1400 Transmission Electron Microscope (JEOL).

4.12. Western Blotting

For cell, biogenic EV, and MIL lysis, the samples were incubated on ice in whole‐cell extract lysis buffer for 30 min. The lysates were then centrifuged at 12,000 rpm for 10 min, and the protein concentration was measured using the Bradford assay (Bio‐Rad Laboratories, Inc.). For Western blot analysis, lysates were boiled for 5 min with sample buffer before being separated on SDS‐polyacrylamide gels. Proteins were transferred to polyvinylidene difluoride membranes (Merck & Co., Inc.) and blocked with 5% nonfat milk in TBST buffer. Using the electrochemiluminescence kit (Amersham Pharmacia Biotech, Inc.), the specific binding antibodies were detected. The specific antibodies against Clatherin (BD Biosciences), ACE2 (Abcam Limited), TSG101 (Santa Cruz Biotechnology), and Hsp90 (Stressgen Biotechnologies) were used.

4.13. Immunofluorescence Staining

Indirect immunofluorescence staining was performed following fixation of the cells with paraformaldehyde (Merck). Anti‐ACE2 antibody was purchased from R&D Systems. The secondary anti‐goat Alexa‐594‐conjugated antibodies (Thermo Fisher Scientific Inc.) were used as counterstaining when Hoechst 33342 DNA staining was employed. The cells were washed and retained in PBS before analysis using an Image Xpress Micro Confocal High‐Content Imaging System (Molecular Device).

4.14. Pseudo‐Virus Infection Assay

To mimic SARS‐CoV‐2 infection, a lentivirus‐based system expressing the SARS‐CoV‐2 spike protein was used. The pseudo‐virus expressing spike protein and GFP (Omicron variants, BA.2) was purchased from LumiSTAR Biotechnology, Inc. To evaluate the infection rate of cells with different ACE2 expression statuses and the disinfection potential of nanoparticles, the results of infection were determined by analyzing the GFP‐expressing cell ratio after infection using an Image Xpress Micro Confocal High‐Content Imaging System (Molecular Device). In brief, different volumes of virus, with or without preincubation with nanoparticles for 4 h, were added to the cells after a 24 h seeding period. After 24 h incubation, the virus‐containing medium was replaced with fresh culture medium, and the cells were cultured for an additional 24 h before the ratio of GFP‐expressing cells was analyzed.

4.15. Statistical Analyses

Statistical analyses were performed using Origin 2025. Comparisons between groups were carried out using Student's t‐test or one‐way analysis of variance (ANOVA). A p‐value of less than 0.05 was considered statistically significant.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: adhm70952‐sup‐0001‐SuppMat.docx.

Supporting File 2: adhm70952‐sup‐0002‐DataFile.pdf.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.