The following study evaluates the toxicological profile of a selected metal-organic framework utilizing electric cell-substrate impedance sensing (ECIS), a real-time, high-throughput screening technique.
Metal-organic frameworks (MOFs) are hybrids formed through the coordination of metal ions and organic linkers in organic solvents. The implementation of MOFs in biomedical and industrial applications has led to concerns regarding their safety. Herein, the profile of a selected MOF, a zeolitic imidazole framework, was evaluated upon exposure to human lung epithelial cells. The platform for evaluation was a real-time technique (i.e., electric cell-substrate impedance sensing [ECIS]). This study identifies and discusses some of the deleterious effects of the selected MOF on the exposed cells. Furthermore, this study demonstrates the benefits of using the real-time method versus other biochemical assays for comprehensive cell evaluations. The study concludes that observed changes in cell behavior could hint at possible toxicity induced upon exposure to MOFs of different physicochemical characteristics and the dosage of those frameworks being used. By understanding changes in cell behavior, one foresees the ability to improve safe-by-design strategies of MOFs to be used for biomedical applications by specifically tailoring their physicochemical characteristics.
Metal-organic frameworks (MOFs) are hybrids formed through the combination of metal ions and organic linkers1,2 in organic solvents. Due to the variety of such combinations, MOFs possess structural diversity3, tunable porosity, high thermal stability, and high surface areas4,5. Such characteristics make them attractive candidates in a variety of applications, from gas storage6,7 to catalysis8,9, and from contrast agents10,11 to drug delivery units12,13. However, the implementation of MOFs into such applications has raised concerns relative to their safety to both the users and the environment. Preliminary studies have shown, for instance, that cellular function and growth change upon the exposure of cells to metal ions or linkers used for MOF synthesis1,14,15. For instance, Tamames-Tabar et al. demonstrated that ZIF-8 MOF, a Zn-based MOF, was leading to more cellular changes in a human cervical cancer cell line (HeLa) and a mouse macrophage cell line (J774) relative to Zr-based and Fe-based MOFs. Such effects were presumably due to the metal component of ZIF-8 (i.e., Zn), which could potentially induce cell apoptosis upon framework disintegration and Zn ion release1. Similarly, Gandara-Loe et al. demonstrated that HKUST-1, a Cu-based MOF, caused the highest reduction in mouse retinoblastoma cell viability when used at concentrations of 10 µg/mL or greater. This was presumably due to the Cu metal ion incorporated during the synthesis of this framework, which, once released, could induce oxidative stress in the exposed cells15.
Moreover, analysis showed that the exposure to MOFs with different physicochemical characteristics could lead to varying responses of exposed cells. For instance, Wagner et al. demonstrated that ZIF-8 and MIL-160 (an Al-based framework), used in the exposure of an immortalized human bronchial epithelial cell, led to cellular responses dependent on frameworks' physicochemical properties, namely hydrophobicity, size, and structural characteristics16. Complementarily, Chen et al. demonstrated that a concentration of 160 µg/mL MIL-100(Fe) exposed to human normal liver cells (HL-7702) caused the largest loss in cellular viability, presumably due to the metal component of this specific framework (i.e., Fe17).
While these studies categorize MOFs' deleterious effects on cellular systems based on their physicochemical characteristics and exposure concentrations, thus raising potential concerns with framework implementation, especially in biomedical fields, most of these evaluations are based on single time point colorimetric assays. For instance, it was shown that when (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) tetrazolium (MTT) and water-soluble tetrazolium salt (WST-1) assays were used, these biochemical reagents could lead to false positives upon their interactions with the particles that the cells were also exposed to18. The tetrazolium salt and neutral red reagents were shown to possess a high adsorption or binding affinity onto the surfaces of the particles, resulting to agent signal interference19. Moreover, for other types of assays, such as flow cytometry, which was previously shown to be used for assessing changes in cells exposed to MOFs20,21, it was shown that major issues have to be circumvented if a viable analysis of particles' deleterious effects is to be considered. In particular, detection ranges of the particles' sizes, especially in mixed populations like the ones offered by MOFs or references of the particles used for calibration before cellular changes, have to be addressed22. It was also shown that the dye used during cell labeling for such cytometry assays could also interfere with the nanoparticles that the cells were exposed to23.
The goal of this study was to use a real-time, high-throughput evaluation assay to assess changes in cell behavior upon exposure to a select MOF. Real-time evaluations can help provide insights into time-dependent effects, as related to the windows of exposures16. Further, they provide information on changes in cell-substrate interactions, cell morphology, and cell-cell interactions, as well as how such changes depend on the physicochemical properties of the materials of interest and exposure times24,25 respectively.
To demonstrate the validity and applicability of the proposed approach, human bronchial epithelial (BEAS-2B) cells, ZIF-8 (a hydrophobic framework of zeolitic imidazolate16), and electric cell-substrate impedance sensing (ECIS) were used. BEAS-2B cells represent a model for lung exposure26 and have been previously used to evaluate changes upon the exposure of cells to nanoclays and their thermally degraded byproducts26,27,28, as well as assess the toxicity of nanomaterials, such as single-walled carbon nanotubes (SWCNTs)18. Furthermore, such cells have been used for more than 30 years as a model for pulmonary epithelial function29. ZIF-8 was chosen due to its wide implementation in catalysis30 and as contrast agents31 for bioimaging and drug delivery32, and thus for the extended potential for lung exposure during such applications. Lastly, ECIS, the noninvasive, real-time technique, was previously used to evaluate changes in cell adherence, proliferation, motility, and morphology16,26 as a result of a variety of interactions between analytes (both materials and drugs) and exposed cells in real-time16,18,28. ECIS uses an alternating current (AC) to measure the impedance of cells immobilized on gold electrodes, with the impedance changes giving insights into changes in resistance and capacitance at the cell-gold substrate interface, barrier function as induced by cell-cell interactions, and over-cell layer coverage of such gold electrodes33,34. Using ECIS allows quantitative measurements at a nanoscale resolution in a noninvasive, real-time manner26,34.
This study assesses and compares the simplicity and ease of evaluation of MOF-induced changes in cellular behavior in real-time with single-point assay evaluations. Such a study could be further extrapolated for evaluating cell profiles in response to exposure to other particles of interest, thus allowing for safe-by-design particle testing and subsequent helping with implementation. Moreover, this study could complement genetic and cellular assays that are single-point evaluations. This could lead to a more informed analysis of the deleterious effects of particles on the cellular population and could be used for screening such particles' toxicity in a high-throughput manner16,35,36.
1. ZIF-8 synthesis
2. ZIF-8 collection
3. ZIF-8 surface morphology (scanning electron microscopy [SEM])
4. ZIF-8 elemental composition
5. Cell culture
6. Cell counting
7. ZIF-8 dose preparation
8. Half-maximal inhibitory concentration (IC 50)
9. Electrical cell-substrate impedance sensing (ECIS)
10. Data analysis
11. Statistical analysis
Using a common in vitro model cell line39 (BEAS-2B), this study aimed to demonstrate the feasibility and applicability of ECIS to assess changes in cell behavior upon exposure to a lab-synthesized MOF. These changes assessment was complemented by analysis through conventional colorimetric assays.
The physicochemical characteristics of the framework were first evaluated to ensure the reproducibility of the methods employed, the validity of the obtained results, and the pertinent discussions of such results. The analysis of the surface morphology of ZIF-8, for instance, was performed through SEM; images showed that the frameworks displayed a rhombic dodecahedron morphology40 and had an average size of 62.87 nm ± 9.61 nm (Figure 1A). The elemental composition of the MOFs was determined viaEDX spectroscopy. The results showed that the main composition of ZIF-8 consisted of C, N, and Zn (i.e., the makeup of the imidazolate linker and the metal ion, Zn16) (Figure 1B). Previous X-ray diffraction showed that ZIF-8 crystalline phases have identified specific peaks at 10.33°, 12.8°, 14.7°, 16.5°, and 18°, with such peaks being attributed to planes (002), (112), (022), (013), and (222), respectively, 16,41.
For the proposed cell behavior screening, the IC50 (concentration where ZIF-8 inhibits cell growth by 50%) was determined29. For this, cells were exposed to ZIF-8 for 48 h at doses ranging from 0-950 µg/mL (Figure 2A). The dose-response graph of exposed cells is displayed in Figure 2B, with the subsequent recorded IC50 of 35.7 µg/mL. In addition, the analysis revealed a dose-dependent trend in the viability of cells upon exposure to the ZIF-8.
Upon determining the IC50, ECIS analysis was performed. Briefly, ECIS was used as a real-time screening strategy of BEAS-2B exposed to ZIF-8 MOFs at IC50, as well as below and above this concentration, namely 15.7 µg/mL (C1), 35.7 µg/mL (C2), 55.7 µg/mL (C3), and 75.7 µg/mL (C4) respectively, with the cells being exposed for a total period of 72 h. The inclusion of ECIS was envisioned to allow insight into determining changes to cell coverage, morphology, and viability, all in real-time. The ECIS results were recorded as changes in resistance and changes in cell-substrate interactions, namely the alpha parameter42.
The resistance trends are displayed in Figure 3A as changes in the profiles of the cells treated with ZIF-8 compared to the control (black line) (i.e., unexposed cells). Specifically, the analysis showed that the wells containing cells on the gold electrodes exposed to C2-C4 doses displayed an initial slight increase in resistance immediately after cell exposure to the frameworks (all relative to the control [i.e., unexposed cells]). These initial changes were subsequently followed by sharp decreases in recorded resistances, with such decreases being prevalent between 6-8 h from the exposure (Figure 3A). A complete loss in resistance was observed after 14-18 h from the exposure time. Cells exposed to doses below the IC50 value displayed resistances, like the wells with control cells, until about 16-18 h from the exposure when resistance losses appear. Also, during the cellular exposure, a continuous signal perturbation of the signal of the wells used in the experiments was observed. These perturbances persisted during the analysis of the alpha parameter (Figure 3B), with this analysis further showing that exposed cells change their cell-substrate interactions with profiles similar to the resistance ones. Moreover, such changes were dependent of the time from the exposure and the dose used in such exposure.
Figure 1: SEM image and average elemental composition ZIF-8 particles. (A) Representative SEM image of ZIF-8 particles. (B) Average elemental composition of ZIF-8 particles (± standard deviation [SD] bars). Please click here to view a larger version of this figure.
Figure 2: Cell viability. (A) Cell treatment schematic (created with Biorender.com). (B) Viability of BEAS-2B cells exposed to ZIF-8 particles at doses ranging from 0-950 µg/mL; this analysis was used to determine the IC50 (± SD bars; ***p = 0.0001 and **** p < 0.0001 relative to the control). Please click here to view a larger version of this figure.
Figure 3: Representative cellular resistance and changes in cellular attachment. (A) Representative cellular resistance of BEAS-2B cells exposed to ZIF-8 particles below, at, and above the IC50 concentration. (B) Changes in the cellular attachment (i.e., changes in the alpha parameter) of BEAS-2B cells when exposed to ZIF-8 particles below and above the IC50 concentration. Please click here to view a larger version of this figure.
Previous analysis showed that ECIS could be used to assess the behavior of cells exposed to analytes (i.e., carbon nanotubes35, drugs43, or nanoclays16). Furthermore, Stueckle et al. used ECIS to evaluate the toxicity of BEAS-2B cells exposed to nanoclays and their byproducts and found that the cellular behavior and attachment were dependent on the physicochemical characteristics of such materials42. Herein, we proposed to determine the possible changes of BEAS-2B cells in response to exposure to MOFs, to thus contribute to the body of knowledge aimed to differentiate any deleterious effects of ZIF-8 on cellular systems in vitro, all in a high-throughput and real-time manner.
The analysis first allowed for the evaluation of framework properties to thus help correlate the changes in cellular behavior to the physicochemical properties of the tested MOFs44. Specifically, upon cellular exposure to MOFs with regular geometries of similar elemental composition (see results above), the analysis showed changes in cellular behavior, presumably resulting from the uptake of the frameworks and their profile cellular degradation. Specifically in terms of the uptake, previous analysis of toxicity on hydrophobic materials, such as this framework, showed that when such materials are exposed to cells, their uptake is more disruptive to a cellular membrane than the uptake of their hydrophilic counterparts presumably due to their ability to remove lipids from the membrane's structural bilayer16,45 during their cellular translocation thus leading to membrane dysfunctions or plasma membrane perturbations46,47. In particular, lipid removal was shown to lead to changes in membrane integrity, cell signaling48, and an increase in cellular uptake49, with a downside of cell function effects. For instance, Farcal et al. demonstrated that a hydrophobic TiO2 nanomaterial had an increased cytotoxic effect when compared to its hydrophilic counterpart in exposed alveolar macrophages50. In addition, Wagner et al., demonstrated that hydrophobic ZIF-8 caused more disruption to BEAS-2B cells when compared to hydrophilic MIL-16016.
The recorded changes in cell behavior are presumably due to disturbances in cell-substrate interactions26 and/or cell viability and proliferation16 onto the electrodes as a result of cell exposure to the selected framework. For instance, Wagner et al. evaluated similar parameters in cells exposed to nanoclays. The authors found that a complete loss in cell-substrate interaction was observed over the course of the 72 h experiments. In addition, the authors demonstrated a time-dependent effect on cell viability and proliferation when the BEAS-2Bs were exposed to their selected nanoclays26. While the particles being tested herein are MOFs, the possible extended association with the effects seen for nanoclays is feasible, as some of the physicochemical characteristics of these particles (namely size, hydrophobicity, and composition) are similar to those of the MOFs . Furthermore, Stueckle et al. also demonstrated time- and treatment-dependent cellular behavior when nanoclays were exposed to immortalized BEAS-2B cells. Others have demonstrated that ZIF-8, at a dose of 100 µg/mL, cause BEAS-2B cells to have a complete loss in the cell monolayer and cell viability, presumably due to changes in cell-substrate interactions16. It is, however, noted that in such studies, the ZIF-8 particles were obtained under different synthesis conditions (i.e., 10 min versus 24 h used in this study), with such synthesis conditions subsequently resulting in different shapes/morphologies of the frameworks to thus support previous conclusions. This shows that the reaction time6, temperature2, and solvent44 used during particle synthesis influence their shape, size, and composition profiles, and subsequently lead to differential effects and behaviors in exposed cells.
The analysis also showed that cells exposed to ZIF-8 doses below the IC50 value displayed resistances similar to those of the controls, presumably since such doses were not significant enough to induce ECIS-observable cell transformation in the conditions and timeframe these experiments were undertaken in. However, at doses above IC50, a complete loss in cellular resistance was observed and only within a few hours after the exposure, most likely because of changes in cell viability and induced possible changes in cell-substrate interactions16 (Figure 3A). These claims are supported by previous studies that showed that non-viable cells change their shape and detach from the substrates. For instance, Verma et al. used a real-time impedance sensing technique to assess the cytotoxicity of nanoclays in a human alveolar epithelial cell line (A549). It was shown that the platelet-type nanoclays caused an increase in the number of detached and deformed cells when compared to the tubular type nanoclays51. In addition, Xiao et al. demonstrated that the cytoskeleton of fibroblastic V79 cells was compromised when exposed to cadmium chloride, due to an influx of water, extracellular ions, and cytotoxic chemicals from the culture medium. The changes in cell cytoskeleton led to the loss of cell-substrate interactions. Furthermore, when benzalkonium chloride was exposed to V79 cells, there was a significant decrease in cellular resistance, due to damage to the cellular membrane that the exposure caused52.
Also, changes in resistance can be due to the cell reactions to the physicochemical characteristics of ZIF-8. In particular, ZIF-8 is a hydrophobic framework; previous analysis has shown that hydrophobicity can cause an increase in toxicity16. Metal ions, such as Zn, have also been previously shown to induce a higher degree of toxicity when compared to other metals like Zr and Fe1 namely, possibly contributing to higher cellular death.
It was further shown that all cells recorded an initial decrease in resistance right after exposure, followed by a subsequent increase. Eldawud et al. demonstrated similar effects when SWCNTs were exposed to BEAS-2B cells. The authors interpreted the increase in cellular resistance as being due to the BEAS-2B cells overcoming the disturbances caused by materials' exposure and subsequently regaining their integrity18 all while eliminating electrode drift28, to thus ensure reproducibility of the data analysis (see ECIS handbook).
Even though this study showed that ECIS could be a valuable tool to use in cell behavior screening, especially due to the high-throughput feature, it is recommended that the technique does not replace single time point assays when a full picture of a material's deleterious assessment is desired. In particular, the single point assay could further allow for the evaluation of cell transformation at their genetic level45,53, as shown by Eldawud et al., for instance for BEAS-2B cells exposed to nanodiamonds and through fluorescence-activated cell sorting (FACS) 45. Yu et al., also used FACS to evaluate human colon cancer cells (HCT0116) exposed to hyaluronic acid-modified mesoporous silica nanoparticles (HA-MSNs)54.
The results presented suggest that the route of change is dependent on the MOFs' physicochemical characteristics, as well as the time and dosing of such frameworks used in exposures. The results and methods described herein further emphasize the importance of real-time measurements and their advantages in understanding the changes in cellular profiles. These can potentially be supplemented with additional biochemical assays to evaluate the mechanism of toxicity, thus leading to safe-by-design strategies of MOFs that have no deleterious effects on cell behavior, recorded as cell attachment, cell-cell interactions, or cell-substrate interactions.
The authors have nothing to disclose.
This work was funded in part by the National Institute of General Medical Sciences (NIGMS) T32 program (T32 GM133369) and the National Science Foundation (NSF 1454230). Additionally, WVU Shared Research Facilities and Applied Biophysics assistance and support are acknowledged.
4-[3-(4-idophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene disulfonate (WST-1 assay) | Roche | 5015944001 | |
0.25% Trypsin-EDTA (1x) | Gibco | 25255-056 | |
100 mm plates | Corning | 430167 | |
1300 Series A2 biofume hood | Thermo Scientific | 323TS | |
2510 Branson bath sonicator | Process Equipment & Supply, Inc. | 251OR-DTH | |
2-methylimidazole, 97% | Alfa Aesar | 693-98-1 | |
5 mL sterile microtube | Argos Technologies | T2076S-CA | |
50 mL tubes | Falcon | 352098 | |
96W10idf well plates | Applied Biophysics | 96W10idf PET | |
96-well plates | Fisherbrand | FB012931 | |
Biorender | Biorender | N/A | |
Countess cell counting chamber slides | Invitrogen | C10283 | |
Countess II FL automated cell counter | Life Technologies | C0916-186A-0303 | |
Denton Desk V sputter and carbon coater | Denton Vacuum | N/A | |
Dimethly sulfoxide | Corning | 25-950-CQC | |
DPBS/Modified | Cytiva | SH30028.02 | |
Dulbecco's modified Eagle medium | Corning | 10-014-CV | |
ECIS-ZΘ | Applied Biophysics | ABP 1129 | |
Excel | Microsoft | Version 2301 | |
Falcon tubes (15 mL) | Corning | 352196 | |
Fetal bovine serum | Gibco | 16140-071 | |
FLUOstar OPTIMA plate reader | BMG LABTECH | 413-2132 | |
GraphPad Prism Software (9.0.0) | GraphPad Software, LLC | Version 9.0.0 | |
HERAcell 150i CO2 Incubator | Thermo Scientific | 50116047 | |
Hitachi S-4700 Field emission scanning electron microscope equipped with energy dispersive X-ray | Hitachi High-Technologies Corporation | S4700 and EDAX TEAM analysis software | |
ImageJ software | National Institutes of Health | N/A | |
Immortalized human bronchial epithelial cells | American Type Culture Collection | CRL-9609 | |
Isotemp freezer | Fisher Scientific | ||
Methanol, 99% | Fisher Chemical | 67-56-1 | |
Parafilm sealing film | The Lab Depot | HS234526A | |
Penicillin/Steptomycin | Gibco | 15140-122 | |
Sorvall Legend X1R Centrifuge | Thermo Scientific | 75004220 | |
Sorvall T 6000B | DU PONT | T6000B | |
Trypan blue, 0.4% solution in PBS | MP Biomedicals, LLC | 1691049 | |
Vacuum Chamber | Belart | 999320237 | |
Zinc Nitrate Hexahydrate, 98% extra pure | Acros Organic | 101-96-18-9 |