A Novel Nanosafety Approach Using Cell Painting, Metabolomics, and Lipidomics Captures the Cellular and Molecular Phenotypes Induced by the Unintentionally Formed Metal-Based (Nano)Particles

Abstract

Additive manufacturing (AM) or industrial 3D printing uses cutting-edge technologies and materials to produce a variety of complex products. However, the effects of the unintentionally emitted AM (nano)particles (AMPs) on human cells following inhalation, require further investigations. The physicochemical characterization of the AMPs, extracted from the filter of a Laser Powder Bed Fusion (L-PBF) 3D printer of iron-based materials, disclosed their complexity, in terms of size, shape, and chemistry. Cell Painting, a high-content screening (HCS) assay, was used to detect the subtle morphological changes elicited by the AMPs at the single cell resolution. The profiling of the cell morphological phenotypes, disclosed prominent concentration-dependent effects on the cytoskeleton, mitochondria, and the membranous structures of the cell. Furthermore, lipidomics confirmed that the AMPs induced the extensive membrane remodeling in the lung epithelial and macrophage co-culture cell model. To further elucidate the biological mechanisms of action, the targeted metabolomics unveiled several inflammation-related metabolites regulating the cell response to the AMP exposure. Overall, the AMP exposure led to the internalization, oxidative stress, cytoskeleton disruption, mitochondrial activation, membrane remodeling, and metabolic reprogramming of the lung epithelial cells and macrophages. We propose the approach of integrating Cell Painting with metabolomics and lipidomics, as an advanced nanosafety methodology, increasing the ability to capture the cellular and molecular phenotypes and the relevant biological mechanisms to the (nano)particle exposure.

Keywords: additive manufacturing; high-content screening (HCS); inflammation; multivariate analysis; nanoparticle emissions; new approach methodologies (NAMs); targeted metabolomics.

Figure 2

Physicochemical characterization of the (nano)particles unintentionally emitted at the metal AM occupational settings (AMPs). (A–F) Scanning electron microscopy (SEM) images demonstrating the size and shape characteristics of the collected AMPs. The blue arrows indicate the spherical micron-sized particles (possibly the feedstock material, based on size/shape/chemical composition), the yellow arrow highlights the micron-sized particles with an altered shape, and the green arrows indicate the large and irregularly shaped particle clusters. The red rectangular spaces in (A,C,E) are magnified in (B,D,F). The red arrows in (C,E) show the nanosatellites attached to the surface of the micron-sized particles. The white dashed areas in (C,E) demonstrate the alterations in the micron-sized particle surface topography. The green arrows in (E,F) show the tendency of the micron-sized particles to adhere to a large number of nanoparticle aggregates/agglomerates. The transmission electron microscopy (TEM) image and the graph in (F), upper right corner, report the nanoparticle size distribution and shape. (G) SEM combined with the energy dispersive spectroscopy (EDS) analysis of the bulk chemical composition (relative metal composition of Fe, Cr, Mn, Mo, Al, Si and V) of the AMPs (see also Figure S1). (H) X-ray photoelectron spectroscopy (XPS) analysis of the relative oxidized metal composition (Fe, Cr, Mn) of the outermost surface of the AMPs (both the micron-sized particles (surface) and the nanoparticles (surface/bulk)).

Figure 3

Impact of the (nano)particles unintentionally emitted at the metal AM occupational settings (AMPs) on the cell viability/metabolic activity, ROS production, and internalization. Plate-based assays with the U-2 OS cells exposed to a range of AMP concentrations (0–100 µg/mL) discloses that: (A) AMPs did not impair the cell viability (blue color—Hoechst 33342 labelling viable nuclei; green color—Image-iT DEAD Green viability stain labelling unviable nuclei), however, AMPs reduced the metabolic activity of cells, as shown in the Alamar blue assay (graph on the right). RFI—relative fluorescence units; (B) AMPs induced the ROS production and exerted oxidative stress in the exposed cells (blue color—Hoechst 33,342 nuclear labelling; red color—CellROX Deep Red ROS labelling). The graph on the right is a quantitative summary of the relative fluorescence obtained from the cells after the ROS labelling. The fluorescent signal was quantified using ImageJ software and the data is reported as the mean value of N = 150 cells, per condition. CTCF—corrected total cell fluorescence. * p < 0.05; *** p < 0.001. (C) Unexposed U-2 OS cells observed under the scanning electron microscope (SEM) at 10,000× magnification. (D–F) U-2 OS cells exposed to 50 µg/mL AMPs for 24 h and imaged under SEM. Red arrows indicate the AMP aggregates/agglomerates associated with the cell’s outer surface (D) or partially covered by the cell membrane (E). Energy dispersive spectroscopy (EDS) spectra in the upper right corners (E,F) indicate the composition of the electron-dense AMP areas. Pt—platinum.

Figure 4

Cell Painting labelling patterns in the U-2 OS cells. Representative images of the control and cells exposed to 0.156, 0.313, 0.625, 1.25, 2.5, 5, 25, 50, and 100 µg/mL of the (nano)particles unintentionally emitted at the metal AM occupational settings (AMPs), live-labeled for the mitochondria (Mito), fixed, permeabilized, and labeled with the remaining fluorescent probes for the nuclei (DNA), actin/Golgi/plasma membrane (AGP), endoplasmic reticulum (ER), and RNA/nucleoli (RNA). Distinct morphological effects of the AMPs observed qualitatively, are evident in each channel, with the exception of the DNA-related features, where the morphological changes can be observed only to a lower extent. All images were acquired at a 20× magnification.

Figure 5

Quantitative summary of the morphological effects for the (nano)particles unintentionally emitted at the metal AM occupational settings (AMPs). Treatment-level feature data were normalized and scaled per plate (N = 3), the batch effect was corrected, and then averaged. Three heatmaps were established, corresponding to the morphological features organized by compartment: (A) cytoplasm (962 features), (B) nuclei (976 features), and (C) cell (998 features); by feature group: correlation, intensity, radial distribution, and texture; and by fluorescent channel: nuclei (DNA), actin cytoskeleton/Golgi/plasma membrane (AGP), endoplasmic reticulum (ER), RNA/nucleoli (RNA), and mitochondria (Mito). The colors represent the fold change in each measured feature, with respect to the unexposed control cells. The rows correspond to the individual concentrations of the AMPs (0.156–100 µg/mL). Exposure concentrations are in descending order from top to bottom. Columns represent the individual morphological features. Data were derived from 435,678 single cell profiles distributed across six technical replicate wells of three microplates/biological replicates (N = 18 wells in total). (D) Boxplots for the feature Texture_Entropy_AGP_5_02_256 are shown as an example of a feature that is clearly dependent on the AMP concentration (x-axis). (E) Uniform manifold approximation and projection (UMAP) was employed at the image level on the median of the morphological profiles of the U-2 OS cells exposed to the AMPs. Color legend indicates the AMP concentrations (µg/mL). (F) Results for the sparse partial least squares discriminant analysis (sPLS-DA) predictions averaged over three cross-validation runs, as a scatterplot. Predicted concentration class (y-axis) is shown as dependent on the true concentration class (x-axis). Circle radius visualizes the frequencies for the respective result, with numbers as text for values larger than 20. Sensitivity for the predictions of large AMP concentrations was 82%, the specificity was 94%. Concentrations of AMPs were considered “small” if less than, or equal to 1.25 µg/mL. Clear concentration dependency, with the predictive potential for the new samples, could be found for the AMP concentrations larger than 2.5 µg/mL.

Figure 6

Effects of the (nano)particles unintentionally emitted at the metal AM occupational settings (AMPs) on the lipidome and polar metabolome. The fold changes (blue and red) of the top 25 out of 73 identified lipids, via the untargeted lipidomics (A) and the top 25 out of 63 identified polar metabolites, via the targeted metabolomics (C) are represented as feature-clustered heatmaps. Each column within the heatmaps represents one of three biological replicates with two technical repetitions. SiO₂ was used as a positive control. Volcano plots show the up-regulation (red) or down-regulation (blue) of the lipids (B) and polar metabolites (D) after 24 h of exposure to 25 μg/mL AMPs. The log2 (FC—fold change) of the relative abundance of the lipids/polar metabolites in the AMP-exposed cells and in the control cells. The y-axis represents the −log10 (p-value) between the exposed and control samples. Results reported in the volcano plots (B,D) are a summary of three biological replicates with two technical repetitions.

Figure 1

Schematic overview of the experimental design. To investigate the toxic effects of the (nano)particles unintentionally emitted at the metal AM occupational settings (AMPs), on human cells, the study was designed in three main phases. (A) Extensive AMPs’ physicochemical characterization, performed using a wide spectrum of analytical methods to support the data interpretation and comparability. In addition, the AMP dispersions were tested for the presence of the endotoxin. (B) Plate-based assays were employed to understand the AMP effects on the cell viability, the ROS production, and internalization. The high-content screening (HCS) by a Cell Painting assay, followed by the univariate, unsupervised multivariate, and supervised multivariate analyses, were performed to understand the impact of AMPs on the cells’ morphological profiles (U-2 OS cells). Mito—mitochondria; ER—endoplasmic reticulum; AGP—actin, Golgi, plasma membrane. (C) Lipidomic and metabolomic analyses were performed in a co-culture model (A549/THP-1 cells), mimicking the lung tissue as a potential key target organ for AMPs, and to close the knowledge gap on the AMP MoAs. Figure created with Biorender.com.