Search

Introducing particles as additives, specifically engineered nanoparticles, in the food industry has improved food properties. Since 2021, alongside the presence of these added particles, there has been a mandatory requirement to disclose if those additives are nanomaterials in the ingredient list of food products. However, detecting and characterizing nanomaterials is time-consuming due to their small sizes, low concentrations, and diverse food matrices. We present a streamlined analytical process to detect the presence of inorganic particles in food, applicable for food regulation and control. Using X-ray Fluorescence Spectrometry for screening enables quick categorization of inorganic particles labeling accuracy, distinguishing products with and without them. For the former, we develop matrix-independent digestion and introduce time-effective statistics to evaluate the median particle size using a reduced number of particles counted, ensuring accurate “nano” labeling. Through the implementation of this work, our objective is to simplify and facilitate verifying the proper labeling of food products.

Understanding the interaction between cells and nanoparticles (NPs) is important to understand the hazard associated with nanoparticles. This requires quantifying and interpreting dose-response relationships, and cells cultured in vitro and exposed to particle dispersions mainly rely on mathematical models that estimate the received nanoparticle dose. Yet, none of these models consider that aqueous cell-culture media wet the inner surface of hydrophilic open wells, which results in a curved fluid-air interface called meniscus. Here we address in detail the impact of the meniscus on nanoparticle dosimetry. We present experiments, build an advanced mathematical model, and find that the presence of meniscus may bring about systematic errors that must be considered to advance reproducibility and harmonization. Finally, we also propose simple practical solutions to this problem, such as covering the air-liquid interface with a permeable lid or soft rocking of the cell culture well plate.

Air-liquid interface (ALI) lung cell models cultured on permeable transwell inserts are increasingly used for respiratory hazard assessment requiring controlled aerosolization and deposition of any material on ALI cells. The approach presented herein aimed to assess the transwell insert-delivered dose of aerosolized materials using the VITROCELL® Cloud12 system, a commercially available aerosol-cell exposure system. An inter-laboratory comparison study was conducted with seven European partners having different levels of experience with the VITROCELL® Cloud12. A standard operating procedure (SOP) was developed and applied by all partners for aerosolized delivery of materials, i.e., a water-soluble molecular substance (fluorescence-spiked salt) and two poorly soluble particles, crystalline silica quartz (DQ12) and titanium dioxide nanoparticles (TiO2 NM-105). The material dose delivered to transwell inserts was quantified with spectrofluorometry (fluorescein) and with the quartz crystal microbalance (QCM) integrated in the VITROCELL® Cloud12 system. The shape and agglomeration state of the deposited particles were confirmed with transmission electron microscopy (TEM). Inter-laboratory comparison of the device-specific performance was conducted in two steps, first for molecular substances (fluorescein-spiked salt), and then for particles. Device- and/or handling-specific differences in aerosol deposition of VITROCELL® Cloud12 systems were characterized in terms of the so-called deposition factor (DF), which allows for prediction of the transwell insert-deposited particle dose from the particle concentration in the aerosolized suspension. Albeit DF varied between the different labs from 0.39 to 0.87 (mean (coefficient of variation (CV)): 0.64 (28%)), the QCM of each VITROCELL® Cloud 12 system accurately measured the respective transwell insert-deposited dose. Aerosolized delivery of DQ12 and TiO2 NM-105 particles showed good linearity (R2>0.95) between particle concentration of the aerosolized suspension and QCM-determined insert-delivered particle dose. The VITROCELL® Cloud 12 performance for DQ12 particles was identical to that for fluorescein-spiked salt, i.e., the ratio of measured and salt-predicted dose was 1.0 (29%). On the other hand, a ca. 2-fold reduced dose was observed for TiO2 NM-105 (0.54 (41%)), which was likely due to partial retention of TiO2 NM-105 agglomerates in the vibrating mesh nebulizer of the VITROCELL® Cloud12. This inter-laboratory comparison demonstrates that the QCM integrated in the VITROCELL® Cloud 12 is a reliable tool for dosimetry, which accounts for potential variations of the transwell insert-delivered dose due to device-, handling- and/or material-specific effects. With the detailed protocol presented herein, all seven partner laboratories were able to demonstrate dose-controlled aerosolization of material suspensions using the VITROCELL® Cloud12 exposure system at dose levels relevant for observing in vitro hazard responses. This is an important step towards regulatory approved implementation of ALI lung cell cultures for in vitro hazard assessment of aerosolized materials.

Machine learning model to predict the outcome (Z-average and Polydispersity index) of ultrasonication experiments with nanoparticle dispersions based on the particle properties and process parameters

1. “Characterization_Nanoparticles.zip” Representative transmission electron microscopy images of the particles (.tif format), and image analyses of size distribution (.csv format). Dynamic light scattering measurement files of the primary nanoparticles and the agglomerated dispersions (.csv format). 2. “AdditionalData_DLS.zip” Auto-correlation functions, Cumulants analyses, distribution fit, intensity-weighted particle size distribution, volume weighted particle size distribution and the quality report for each measurement (.pdf format, by Malvern Panalyticals DLS instrument). 3. “RawData_DLS.zip” Compiled raw data of all DLS measurements that were used in the ML modeling or testing process (.txt format). 4. “Sonication_Energy_Calibration.zip” Calorimetric data used to calibrate the effective delivered sonication energy (.csv format). 5. “ModelTest_Data.zip” Data of all measurement the model was tested with (.csv format).

Machine learning model to predict the outcome (Z-average and Polydispersity index) of sonication experiments with nanoparticle dispersions based on the particle properties and process parameters

Additional file 1: Figure S1A. Comparison of the absorbance and emission spectra of the different LysoTracker probes. Figure S1B. Comparison of the emission spectra of the different (nano)particles. Figure S1C. Comparison of the absorbance spectra of the different (nano)particles. Figure S2. Confocal microscopy images of different LysoTracker probes. Figure S3. Transmission electron micrographs (TEM) of the different silica (nano)particles. Figure S4A. Size distributions of 59 nm SiO2-BDP FL NP and 59 nm SiO2-RhoB NP measured by dynamic light scattering. Figure S4B. Size distributions of 119 nm SiO2-Cy5 particles and 920 nm SiO2-Cy5 particles measured by dynamic light scattering. Figure S5A and S5B. Representative images of individual cells from different experiments for identification and contour with the corresponding histogram of the complete image acquired for each individual channel (nanoparticles and lysosomes). Figure S6. Representative bright field image of live cells (A) and fluorescent microscopy image of fixed cells (B) showing a population of individual cells used for colocalization analysis. Figure S7. Representative images of LysoTracker Red probe and Lamp-2 stainings in a single cell. Figure S8. Comparison of Pearson’s and Manders’ coefficients between live and fixed cells. Figure S9. Comparison of raw integrated intensity between live and fixed cells. Video S1. Continuous live cell imaging of J774A.1 cells. Script S1. Raw integrated densities. Script S2. Colocalization analysis.

Cellular surface recognition and behavior are driven by a host of physical and chemical features which have been exploited to influence particle–cell interactions. Mechanical and topographical cues define the physical milieu which plays an important role in defining a range of cellular activities such as material recognition, adhesion, and migration through cytoskeletal organization and signaling. In order to elucidate the effect of local mechanical and topographical features generated by the adsorption of particles to an underlying surface on primary human monocyte‐derived macrophages (MDM), a series of poly(N‐isopropylacrylamide) (pNIPAM) particles with differing rigidity are self‐assembled to form a defined particle‐decorated surface. Assembly of particle‐decorated surfaces is facilitated by modification of the underlying glass to possess a positive charge through functionalization using 3‐aminopropyltriethoxysilane (APTES) or coating with poly(L‐lysine) (PLL). MDMs are noted to preferentially remove particles with higher degrees of crosslinking (stiffer) than those with lower degrees of crosslinking (softer). Alterations to the surface density of particles enabled a greater area of the particle‐decorated surface to be cleared. Uniquely, the impact of particle adsorption is evinced to have a direct impact on topographical recognition of the surface, suggesting a novel approach for controllably affecting cell‐surface recognition and response.