Table 2.
Overview of the key findings regarding the state of science in in vitro nanotoxicity testing of food-grade nanomaterials, categorized by nanomaterial type
| First author | Year | Test system | Dose range (administered) | Nanomaterial grade | PCM characterization | Standardized dispersion and characterization | Dose range rationale and dosimetry | Dissolution biokinetics | Main conclusions from study | Ref |
|---|---|---|---|---|---|---|---|---|---|---|
| Titanium dioxide | ||||||||||
| Kirsten Gerloff | 2009 | Caco-2 cells | 20 and 80 μg/cm2 for 4 h and 24 h | Not reported | Manufacturer provided | No standard dispersion protocol specified | Not reported | Not reported | Food-related nanoparticles potentially hazardous. All nanoparticles (TiO2, SiO2, CB, and ZnO) except MgO exhibit cytotoxicity. ZnO and SiO2 induce DNA damage while SiO2 and CB cause glutathione depletion | [80] |
| Brian A. Koeneman | 2010 | Caco-2 cells | Acute dose of 10, 100 and 1000 μg/ml. Chronic dose of 100 and 1000 μg/ml |
Not reported | Primary particle size and SSA (provided by manufacturer), SEM, DLS, zeta potential | Not reported | Not reported | Not reported | TiO2 nanoparticles can potentially translocate through epithelial lining (at low levels) by transcytosis and induce sub-lethal effects – microvilli reorganization and intracellular calcium increase in Caco-2 cells | [56] |
| Kirsten Gerloff | 2012 | Caco-2 cells | 20 and 80 μg/cm2 for 4 h and 24 h | Not reported | XRD, SSA by BET method, XRF analysis, TEM, and DLS in DI water and cell culture media | No standard dispersion protocol specified | Not reported | Not reported | Anatase/rutile TiO2 nanoparticles show higher toxicity per unit surface area than pure anatase | [68] |
| Matthieu Fisichella | 2012 | Caco-2 cells | 10 to 100 μg/ml for 4 h, 24 h and 72 h | TiO2 STNPs widely used in sunscreens | DLS in DI water and culture medium, zeta potential, TEM | Not reported | Dose range based on accidental high exposures, but dosimetry not taken into consideration | Not reported | Surface-treated TiO2, which have a strong tendency to agglomerate in complex media, show no toxic effects on Caco-2 cells after exposures up to 72 h | [74] |
| Yun Zhao | 2013 | Human primary epidermal keratinocytes | 50 fg/ml to 500 μg/ml for 24 h | Not reported | TEM, DLS | No standard dispersion protocol specified | Not reported | Not reported | TiO2 nanoparticles induce autophagy in addition to cell viability loss in human primary epidermal keratinocytes | [58] |
| Christie McCracken | 2013 | C2BBe1 cells | 10 μg/cm2 for short-term (24 h) and long-term (29 exposure cycles) exposure | Not reported | DLS, zeta potential, TEM, DRIFTS, XRD | No standard dispersion protocol specified | Not reported | AAS to measure Zn2+ from ZnO nanoparticles in stomach phase | C2BBe1 cells internalize TiO2, SiO2 and ZnO nanoparticles but show mild toxicity only upon exposure to ZnO nanoparticles. TiO2 nanoparticles exposed to simulated digestion environment induce mild toxic effects | [59] |
| Isabella De Angelis | 2013 | Caco-2 cells | 1, 2.5, 5, 10 and 20 μg/cm2 for 6 h and 24 h | Not reported | DLS, zeta potential, SEM, TEM, ICP-MS | No standard dispersion protocol specified | Not reported | ICP-MS to measure amount of Zn or Ti in cells | ZnO nanoparticles, in contrast to TiO2 nanoparticles, induce significant toxicity in Caco-2 cells by increasing intracellular ROS levels, pro-inflammatory cytokine (IL-8) and releasing Zn2+ ions | [60] |
| Kirsten Gerloff | 2013 | Caco-2 cells | 0.3125, 1.25, 5, 20 and 80 μg/cm2 for 4 h and 24 h | Not reported | TEM, ICP-OES, DLS, SLD | No standard dispersion protocol specified | Not reported | Not reported | Undifferentiated Caco-2 cells more sensitive to the toxic effects exerted by SiO2 and ZnO nanoparticles than differentiated Caco-2 cells | [127] |
| Xin-Xin Chen | 2013 | Caco-2 cells, GES-1 cells | 10, 25, 50, 100 and 200 μg/ml for 24 h | Nanoparticles extracted from commercially available chewing gums | XRD, TEM-EDS, SEM, NTA | Not reported | Not reported | Not reported | More than 93% of TiO2 in chewing gums is in nano form and ~ 95% of nano-TiO2 particles end up being swallowed. Nano-TiO2 relatively safe for GES-1 and Caco-2 cells | [16] |
| Zhangjian Chen | 2014 | V79 cells | 5, 10, 20, 50 and 100 μg/ml for 6 h, 24 h and 48 h | Not reported | Previously characterized [96] | No standard dispersion protocol specified | Not reported | Not reported | TiO2 nanoparticles induce significant increase in DNA strand breaks, % Tail DNA and HPRT gene locus mutation frequency | [62] |
| James J. Faust | 2014 | C2BBe1 cells | 0.35, 3.5 and 35 μg/ml for 24 h | Food grade TiO2 and TiO2 extracted from chewing gums | XPS, XRD, TEM, DLS, zeta potential | No standard dispersion protocol specified | Not reported | Not reported | Food grade TiO2 nanoparticles disrupt brush border epithelium independent of sedimentation | [26] |
| Emilie Brun | 2014 | Caco-2 cells, co-culture of Caco-2 and HT29-MTX cells (mucus-secreting epithelium), co-culture of Caco-2 and Raji B cells (follicle-associated epithelium) | 50 and 2000 μg/ml for 48 h | Not reported (self-synthesized) | SSA by BET, XRD, TEM, agglomeration state (DLS), zeta potential, XAS | Nanoparticle suspensions pulse sonicated at 28% amplitude – corresponding power measured using a calorimetric procedure [104] | Dose range based on worst case scenario, but dosimetry not taken into consideration | Not reported | TiO2 nanoparticles pass through follicle-associated epithelium model only and their intracellular accumulation depends highly on the cell model – higher in Goblet and M cells than in enterocytes. Intracellular TiO2 does not dissolve and shows higher biopersistance |
[46] |
| Constantinos Gitrowski | 2014 | Caco-2 cells | 1 mg/L for 0 h, 2 h, 4 h, 6 h, 8 h and 24 h | Not reported | TEM and NTA in water | No standard dispersion protocol specified | Not reported | Not reported | Caco-2 cells show characteristic active uptake of Ti from TiO2 nanoparticle exposures, which is dependent on the crystal form of the nanomaterial | [109] |
| Birgit J. Teubl | 2015 | Buccal mucosa (ex vivo), Human buccal epithelial cells (TR146) | 50, 100, 150 and 200 μg/ml for 4 h and 24 h | One pigment-grade TiO2. Not reported for the other two TiO2 |
TEM, DLS, FTIR, laser diffraction analysis, surface hydrophobicity by RB adsorption method | Nanoparticle suspensions ultra-sonicated from 1 to 24 h to evaluate the optimal method to ensure lowest mean particle sizes | Not reported | Not reported | TiO2 nanoparticles tend to aggregate in saliva but available nano-TiO2 gets internalized in the oral cavity within 10 min. Although no effect on viability and membrane integrity, internalized TiO2 triggers ROS production in the cells of buccal epithelium after short-time incubation | [148] |
| Magdiel I. Setyawati | 2015 | SW480, DLD-1 and NCM460 cells | 62.5, 250 and 1000 μM for 24 h | Not reported | FETEM, hydrodynamic size (DLS), zeta potential | No standard dispersion protocol specified | Not reported | Not reported | Among ZnO, TiO2 and SiO2, ZnO nanoparticles were the most cytotoxic to all three intestinal cell types. Different cellular responses among the three cell types owes to their different genetic landscape | [64] |
| Zheng-Mei Song | 2015 | Caco-2 cells | 50 and 200 μg/ml for 24 h | Food additive TiO2 and regular TiO2 | XRF, XRD, TEM, hydrodynamic size (DLS), zeta potential, FTIR spectroscopy | Not reported | Not reported | Not reported | Native and digestion fluid pretreated TiO2 nanoparticles get internalized by Caco-2 cells but not toxic to Caco-2 cells/monolayers. The possibility of TiO2 nanoparticles translocating through Caco-2 monolayers is low | [99] |
| Saeko Tada-Oikawa | 2016 | THP-1 and Caco-2 cells | 1, 10, 25 and 50 μg/ml for 24 h and 72 h | Not reported | Hydrodynamic size (DLS), TEM, zeta potential | Nanoparticle suspensions were sonicated based on a standardized protocol [101] | Not reported | Not reported | Anatase TiO2 nanoparticles induce inflammatory response by upregulating IL-1β and IL-8 production in THP-1 and Caco-2 cells, respectively | [69] |
| Maria G. Ammendolia | 2017 | HT29 cells | 1, 2.5, 5 and 20 μg/cm2 for 6 h, 24 h and 48 h | Not reported | TEM, SEM, hydrodynamic diameter (DLS), PdI. SSA and purity (provided by manufacturer) | No standard dispersion protocol specified | Not reported | Not reported | TiO2 nanoparticles do not induce cytotoxicity or changes in mitochondrial membrane potential but cause dose-dependent oxidative stress that decreases at 24 h. TiO2 nanoparticles, in combination with IGF-1, induce higher cell proliferation as compared to TiO2 nanoparticles alone | [125] |
| William Dudefoi | 2017 | MET-1 bacterial community | 100 and 250 ppm for 48 h | Two food-grade TiO2 and one P25 TiO2 | TEM, XRD, isoelectric point, SSA by BET, XPS | Not applicable | Dose range based on the amount of TiO2 found in the intestine after ingestion of 1–2 pieces of gum or candy | Not applicable | TiO2 nanoparticles do not significantly alter the human gut microbiota by showing little impact on a defined anaerobic gut microbial community MET-1, as assessed through bacterial respiration, fatty acid profiles and phylogenetic composition | [45] |
| Silicon dioxide | ||||||||||
| Kirsten Gerloff | 2009 | Caco-2 cells | 20 and 80 μg/cm2 for 4 h and 24 h | Not reported | Manufacturer provided | No standard dispersion protocol specified | Not reported | Not reported | Food-related nanoparticles potentially hazardous. All nanoparticles (TiO2, SiO2, CB, and ZnO) except MgO exhibit cytotoxicity. ZnO and SiO2 induce DNA damage while SiO2 and CB cause glutathione depletion | [80] |
| Helge Gehrke | 2012 | HT29 cells | 0.03, 0.31, 1.56, 3.13, 15.6, 31.3, 93.8 and 156.3 μg/cm2 for 24 h, 48 h and 72 h | Not reported | TEM, DLS, zeta potential | No standard dispersion protocol specified | Not reported | Not reported | SiO2 nanoparticle stimulate HT29 cell proliferation whereas cytotoxicity depends on its concentration and size, and FCS (Fetal calf serum) content of the cell culture medium | [57] |
| Christie McCracken | 2013 | C2BBe1 cells | 10 μg/cm2 for short-term (24 h) and long-term (29 exposure cycles) exposure | Not reported | DLS, zeta potential, TEM, DRIFTS, XRD | No standard dispersion protocol specified | Not reported | AAS to measure Zn2+ from ZnO nanoparticles in stomach phase | C2BBe1 cells internalize TiO2, SiO2 and ZnO nanoparticles but show mild toxicity only upon exposure to ZnO nanoparticles. TiO2 nanoparticles exposed to simulated digestion environment induce mild toxic effects | [59] |
| Kirsten Gerloff | 2013 | Caco-2 cells | 0.3125, 1.25, 5, 20 and 80 μg/cm2 for 4 h and 24 h | Not reported | TEM, ICP-OES, DLS, SLD | No standard dispersion protocol specified | Not reported | Not reported | Undifferentiated Caco-2 cells more sensitive to the toxic effects exerted by SiO2 and ZnO nanoparticles than differentiated Caco-2 cells | [127] |
| Yi-Xin Yang | 2014 | GES-1 cells, Caco-2 cells | 10, 25, 50, 100, 200, 400 and 600 μg/ml for 24 h, 48 h and 72 h | Food additive SiO2 nanoparticles | XRD, TEM, SSA by BET, hydrodynamic size (DLS), zeta potential, XRF, FTIR | No standard dispersion protocol specified | Not reported | Not reported | At higher concentrations, food additive SiO2 nanoparticles enter cells and inhibit cell growth by cell cycle arrest | [128] |
| Magdiel I. Setyawati | 2015 | SW480, DLD-1 and NCM460 cells | 62.5, 250 and 1000 μM for 24 h | Not reported | FETEM, hydrodynamic size (DLS), zeta potential | No standard dispersion protocol specified | Not reported | Not reported | Among ZnO, TiO2 and SiO2, ZnO nanoparticles were the most cytotoxic to all three intestinal cell types. Different cellular responses among the three cell types owes to their different genetic landscape | [64] |
| Zinc oxide | ||||||||||
| Kirsten Gerloff | 2009 | Caco-2 cells | 20 and 80 μg/cm2 for 4 h and 24 h | Not reported | Manufacturer provided | No standard dispersion protocol specified | Not reported | Not reported | Food-related nanoparticles potentially hazardous. All nanoparticles (TiO2, SiO2, CB, and ZnO) except MgO exhibit cytotoxicity. ZnO and SiO2 induce DNA damage while SiO2 and CB cause glutathione depletion | [80] |
| Christie McCracken | 2013 | C2BBe1 cells | 10 μg/cm2 for short-term (24 h) and long-term (29 exposure cycles) exposure | Not reported | DLS, zeta potential, TEM, DRIFTS, XRD | No standard dispersion protocol specified | Not reported | AAS to measure Zn2+ from ZnO nanoparticles in stomach phase | C2BBe1 cells internalize TiO2, SiO2 and ZnO nanoparticles but show mild toxicity only upon exposure to ZnO nanoparticles. TiO2 nanoparticles exposed to simulated digestion environment induce mild toxic effects | [59] |
| Isabella De Angelis | 2013 | Caco-2 cells | 1, 2.5, 5, 10 and 20 μg/cm2 for 6 h and 24 h | Not reported | DLS, zeta potential, SEM, TEM, ICP-MS | No standard dispersion protocol specified | Not reported | ICP-MS to measure amount of Zn or Ti in cells | ZnO nanoparticles, in contrast to TiO2 nanoparticles, induce significant toxicity in Caco-2 cells by increasing intracellular ROS levels, pro-inflammatory cytokine (IL-8) and releasing Zn2+ ions | [60] |
| Kirsten Gerloff | 2013 | Caco-2 cells | 0.3125, 1.25, 5, 20 and 80 μg/cm2 for 4 h and 24 h | Not reported | TEM, ICP-OES, DLS, SLD | No standard dispersion protocol specified | Not reported | Not reported | Undifferentiated Caco-2 cells more sensitive to the toxic effects exerted by SiO2 and ZnO nanoparticles than differentiated Caco-2 cells | [127] |
| Yanli Wang | 2014 | GES-1 cells, Neural stem cells | 15 μg/ml for 24 h | Not reported | XRD, TEM, XRF, hydrodynamic size (DLS) in water and cell culture medium, zeta potential | Not reported | Not reported | Not reported | Higher rate of dissolution of ZnO nanoparticles in the presence of Vitamin C aggravate the toxic effects of ZnO nanoparticles | [63] |
| Magdiel I. Setyawati | 2015 | SW480, DLD-1 and NCM460 cells | 62.5, 250 and 1000 μM for 24 h | Not reported | FETEM, hydrodynamic size (DLS), zeta potential | No standard dispersion protocol specified | Not reported | Not reported | Among ZnO, TiO2 and SiO2, ZnO nanoparticles were the most cytotoxic to all three intestinal cell types. Different cellular responses among the three cell types owes to their different genetic landscape | [64] |
| Iron oxide | ||||||||||
| Wen Zhang | 2010 | Caco-2 cells | 100, 200 and 300 μg/ml from 5 to 45 min | Not reported (self-synthesized) | DLS and TEM | Not reported | Dose range not justified but adsorption kinetics taken into consideration | Not reported | Adsorption of hematite nanoparticles on Caco-2 cells is size dependent. Longer exposures induce tight junction disruption, and microvilli reorganization and detachment | [98] |
| Madhavi Kalive | 2012 | Caco-2 cells | 1, 10 and 100 ppm from 5 to 28 days | Not reported (self-synthesized) | DLS, PdI and zeta potential in DI water and culture medium, ICP-MS | No standard dispersion protocol specified | Not reported | Not reported | Hematite nanoparticles potentially induce structural changes in the Caco-2 epithelium and the effects at cellular and genetic level are size-dependent | [66] |
(alphabetical): AAS Atomic absorption spectroscopy, BET Brunauer-Emmett-Teller, CB Carbon black, DLS Dynamic light scattering, DRIFTS Diffuse reflectance infrared Fourier transform spectroscopy, FTIR Fourier transform infrared spectroscopy, ICP-MS Inductively-coupled plasma mass spectrometry, ICP-OES Inductively-coupled plasma optical emission spectrometry, IGF-1 Insulin-like growth factor 1, MET-1 Microbial ecosystem therapeutic-1, NTA Nanoparticle tracking analysis, PdI Polydispersity index, SEM Scanning electron microscopy, SLD Static light diffraction, SSA Specific surface area, STNPs Surface treated nanoparticles, TEM Transmission electron microscopy, TEM-EDS Transmission electron microscopy-energy dispersive spectroscopy, XAS X-ray absorption spectroscopy, XPS X-ray photoelectron spectroscopy, XRD X-ray diffraction, XRF X-ray fluorescence