ZnO nanoparticle preparation route influences surface reactivity, dissolution and cytotoxicity

Abstract

ZnO nanoparticles (nZnO) are commonly used in nanotechnology applications despite their demonstrated cytotoxicity against multiple cell types. This underscores the significant need to determine the physicochemical properties that influence nZnO cytotoxicity. In this study, we analyzed six similarly sized nZnO formulations, along with SiO₂-coated nZnO, bulk ZnO and ZnSO₄ as controls. Four of the nZnO samples were synthesized using various wet chemical methods, while three employed high-temperature flame spray pyrolysis (FSP) techniques. X-ray diffraction and optical analysis demonstrated the lattice parameters and electron band gap of the seven nZnO formulations were similar. However, electrophoretic mobility measures, hydrodynamic size, photocatalytic rate constants, dissolution potential, reactive oxygen species (ROS) production and, more importantly, the cytotoxicity of the variously synthesized nZnO towards Jurkat leukemic and primary CD4⁺ T cells displayed major differences. Surface structure analysis using FTIR, X-ray photoelectron spectroscopies (XPS) and dynamic light scattering (DLS) revealed significant differences in the surface-bound chemical groups and the agglomeration tendencies of the samples. The wet chemical nZnO, with higher cationic surface charge, faster photocatalytic rates, increased extracellular dissolution and ROS generation demonstrated greater cytotoxicity towards both cell types than those made with FSP techniques. Furthermore, principal component analysis (PCA) suggests that the synthesis procedure employed influences which physicochemical properties contribute more to the cytotoxic response. These results suggest that the synthesis approach results in unique surface chemistries and can be a determinant of cellular cytotoxicity and oxidative stress responses.

Figure 1

np-induced toxicity values for (a) jurkat leukemic cell and (b) normal primary cd4⁺ t cell viability at 24 hours after treatment with the wet chemical method (green bars; wet), flame spray pyrolysis (blue bars; fsp) nzno formulations and the sio₂-fspr and bulk controls (black bars; control). the white labels on the histogram bars depict the ic₅₀ values obtained for the indicated sample. the histogram bars were ordered from lowest to highest ic₅₀ for both cell types to depict the synthesis method trends observed for the np-induced toxicity. cultures were treated concurrently with varying concentrations of nzno dispersed in nanopure water/rpmi for 24 hours and cell viability was evaluated (means ± standard error, minimum of n = 3) using alamar blue staining (jurkat cells) or flow cytometry with pi staining (cd4⁺ t cells). statistical analysis was performed using repeated measures analysis of variance and model-based means post hoc test (p < 0.05) with differing letters denoting statistical significance. linear contrast models were used to determine statistical significance between the wet chemical, flame spray pyrolysis and control samples.

Figure 2

nzno samples generated ros in a formulation-dependent manner. mitochondrial superoxide generation by all nzno at 24-hour post treatment with 32.4 μg/ml zno using flow cytometry and mitosox™ red staining. statistical analysis was performed using repeated measures analysis of variance and model-based means post hoc test (p < 0.05) with differing letters denoting statistical significance.

Figure 3

surface property characterization for the powdered samples ((a) and (b)). and catalytic activity plots ((c) and (d)) depicting the uv/vis monitored fluorescence of the model sulfo-rhodamine b dye and the average dye degradation kinetic values obtained for the evaluated treatment conditions. ftir spectra for (a) wet chemical synthesis methods, and (b) heat treatment methods illustrate the peaks observed and the corresponding wavenumber values. the graph in (c) is representative of the time-dependent plots obtained and demonstrate the photocatalytic decomposition of sulfo-rhodamine b dye in nanopure water after treatment with eg nps. the histogram in (d) depicts the average catalytic rate constants (k, min^-1) for the nzno and control samples. statistical analysis in (d) was performed using repeated measures analysis of variance and model-based means post hoc test (p < 0.05) with differing letters denoting statistical significance. histogram bars represent n=3 replicates with error bars indicating s.e.

Figure 4

nzno formulations display similar dissolution kinetic trends in cellular media at 4 and 24 hours. extracellular (ec) zn²⁺ concentrations (μg/ml) measured via icp-ms (bars and left y-axis) and intracellular zn²⁺ concentrations measured via flow cytometry and expressed as mean fluorescence intensity (mfi) of the zinc specific dye fluozin-3 am (line graphs and right y-axis) evaluated at 4 hours (top graph) and 24 hours (bottom graph). control samples, designated as the left most bar or line graph symbol were rpmi-based cellular media (ec assay) and nt cells (intracellular (ic) assay). statistical analysis was performed for the extracellular zinc measurements (histogram bars) using repeated measures analysis of variance and model-based means post hoc test (p < 0.05) with differing letters denoting statistical significance. histogram bars or line graph circles represent the average of n = 4 replicates with error bars depicting s.e.

Figure 5

cell-associated (ca) zn²⁺ concentrations (μg/l) measured at 4- and 24-hours via icp-ms. a control sample designated as the left most grey bar, were non-treated cells grown in cellular media. statistical analysis was performed using repeated measures analysis of variance and model-based means post hoc test (p < 0.05) with differing letters denoting statistical significance. histogram bars represent the average of n = 4 replicates with s.e. error bars.

Figure 6

representative samples highlighting the ftir region from 1800 to 350 cm^-1 and the zno peak deconvolution. the figure includes (a) ftir spectra from 1800 to 350 cm^-1 for the eg nzno sample at 4 hours (blue) and 24 hours (red) and peak deconvolution of the broad ftir band from 750-350 cm^-1 for the eg nzno sample at (b) 4-hour and (c) 24-hour time points. samples were introduced to cellular media at a concentration of 32 μg/ml and incubated for the indicated time points. after incubation, the dispersions were centrifuged and the precipitate retained and dried overnight at 60 °c.

Figure 7

the integrated area ratio of zno to po₄^3- from the ftir spectra and the atomic concentration ratio of zn/p from the xps survey spectra of the insoluble zinc amorphous precipitates isolated from nzno dispersions in cellular media post incubated for 4 and 24 hours. the left-hand side of the figure represents the integrated area ratio of zno to po₄^3- from the ftir spectra (line graphs with circles and first right y-axis) and the atomic concentration ratio of zn/p from the xps survey spectra (line graphs with triangles and second right y-axis) evaluated at (a) 4 hours and (b) 24 hours. the faded histogram bars represented the ca zinc results presented in fig. 5 and are included for reference. the right-hand side of the figure represents the xps spectra from 1027 ev to 1018.5 ev illustrating the deconvolution of the zn2p_3/2 peak for the (c) eg nzno as prepared sample and the (d) eg nzno in cellular media at the 4 hour time point.

Figure 8

the time-dependent deposition nzno onto the bottom of a well in a 96-well culture plate as determined by isdd dosimetry modelling for the individual nzno formulations. the total amount of nzno introduced to the cellular media was 6.48 μg (32.4 μg/ml introduced into 0.2 ml of rpmi-based cellular media). the curves in (a) represent the calculated nzno deposition over a 30-hour period. the histogram bars (left y-axis) in (b) demonstrate the modelled concentration of nzno deposited at the 24-hour time point with the half white/half black circles (first right y-axis) correlating to hydrodynamic size and the red triangles (second right y-axis) representing the specific surface area (ssa) as measured by bet. error bars in (b) represent s.e. with n = 3 replicates.

Figure 9

pca model-generated values for both pc1 and pc2 and a graphical representation of the pc scores and jurkat cells ic₅₀ values for each evaluated sample. the table in (a) represents the loading values (× 100) for each measured variable, eigenvalues and the percent variation explained for each pc. the indicated abbreviations refer to hydrodynamic size (hydrosize), intracellular zinc [ic zn²⁺], cell-associated-zinc [ca zn²⁺], and extracellular zinc [ec zn²⁺] concentrations. the green cubes (wet chemical methods), blue spheres (fsp method) and black tetrahedrons (controls samples) depicted in the 3d graph in (b) represent pc1 scores versus ic₅₀ values versus pc2 scores. the colored ellipses denote sample grouping based on similar pc1 and/or pc2 scores.

Scheme 1

illustration depicting the nzno dissolution processes and interactions with cells after np treatment.