Effects of Zinc Oxide Nanoparticles on Model Systems of the Intestinal Barrier

Abstract

Zinc oxide nanoparticles (ZnO NP) are often used in the food sector, among others, because of their advantageous properties. As part of the human food chain, they are inevitably taken up orally. The debate on the toxicity of orally ingested ZnO NP continues due to incomplete data. Therefore, the aim of our study was to examine the effects of two differently sized ZnO NP (<50 nm and <100 nm primary particle size; 123-614 µmol/L) on two model systems of the intestinal barrier. Differentiated Caco-2 enterocytes were grown on Transwell inserts in monoculture and also in coculture with the mucus-producing goblet cell line HT29-MTX. Although no comprehensive mucus layer was detectable in the coculture, cellular zinc uptake was clearly lower after a 24-h treatment with ZnO NP than in monocultured cells. ZnO NP showed no influence on the permeability, metabolic activity, cytoskeleton and cell nuclei. The transepithelial electrical resistance was significantly increased in the coculture model after treatment with ≥307 µmol/L ZnO NP. Only small zinc amounts (0.07-0.65 µg/mL) reached the basolateral area. Our results reveal that the cells of an intact intestinal barrier interact with ZnO NP but do not suffer serious damage.

Keywords: Caco-2; HT29-MTX; Transwell; barrier integrity; coculture; nanoparticles; toxicity; zinc oxide.

Figure 1

Representative microscopic images of Caco-2 cells (a), HT29-MTX cells (b) and Caco-2/HT29-MTX coculture cells (c) stained with alcian blue.

Figure 2

Relative metabolic activity of Caco-2 monoculture (a) and Caco-2/HT29-MTX coculture cells (b) after 24 h treatment with ZnO NP and ZnCl₂. SC: solvent control (5% Millipore water); PC: positive control (0.1% Triton X-100 + 10 mM EGTA apical, 10 mM EGTA basolateral). Data are normalized to the untreated control (=100%; dashed line) and expressed as mean + standard deviation; n = 4 for (a), n = 3 for (b). Significant differences compared to the untreated control (* p  ≤  0.05) were obtained by one-way analysis of variance/Ryan–Einot–Gabriel–Welsh post hoc test.

Figure 3

Relative TEER of Caco-2 monoculture (a) and Caco-2/HT29-MTX coculture cells (b) after 24 h treatment with ZnO NP and ZnCl₂. UC: untreated control; SC: solvent control (5% Millipore water); PC: positive control (0.1% Triton X-100 + 10 mM EGTA apical, 10 mM EGTA basolateral). Dashed line (=100%) implies TEER before incubation. Data are expressed as mean + standard deviation; n = 16. Significant differences compared to UC (* p  ≤  0.05) were obtained by one-way analysis of variance/Ryan–Einot–Gabriel–Welsh post hoc test.

Figure 4

Basolateral FITC-dextran concentrations of Caco-2 monoculture (a) and Caco-2/HT29-MTX coculture cells (b) after 24 h of treatment with ZnO NP and ZnCl₂ and subsequent apical addition of FITC-dextran. UC: untreated control; SC: solvent control (5% Millipore water); PC: positive control (0.1% Triton X-100 + 10 mM EGTA apical, 10 mM EGTA basolateral). Data are expressed as mean + standard deviation; n = 3. Significant differences compared to UC (* p  ≤  0.05) were obtained by one-way analysis of variance/Ryan–Einot–Gabriel–Welsh post hoc test.

Figure 5

Representative images of fluorescence staining with phalloidin (green) and DAPI (blue) after 24 h incubation with cell culture medium (untreated control), 5% Millipore water (solvent control), 0.1% Triton X-100 + 10 mM EGTA apical, 10 mM EGTA basolateral (positive control), 614 µmol/L ZnO NP (<50 nm or <100 nm) and ZnCl₂; Caco-2 monocultured cells are shown above and Caco-2/HT29-MTX cocultured cells are shown below.

Figure 6

Zinc amount of Caco-2 monoculture (a) and Caco-2/HT29-MTX coculture cells (b) after 24 h of treatment with ZnO NP and ZnCl₂. UC: untreated control. Data are expressed as mean + standard deviation; n = 3. Significant differences compared to UC (* p  ≤  0.05) were obtained by one-way analysis of variance/Ryan–Einot–Gabriel–Welsh post hoc test.

Figure 7

Zinc content of apical and basolateral supernatants of Caco-2 monoculture (a–c) and Caco-2/HT29-MTX coculture cells (d–f) after 24 h of treatment with ZnO NP and ZnCl₂ compared to the stock dispersions. Data are expressed as mean + standard deviation; n = 4. Significant differences between zinc amounts before (stock) and after incubation (apical + basolateral; * p  ≤  0.05; ** p  ≤  0.01) were obtained by unpaired t-tests.