Final Report Summary - NANOTRAFFIC (Do small things lead to big problems? Mechanism of uptake and toxicity of metal nanoparticles in intestinal cells)
The overarching objective of the NanoTraffic project was to investigate the uptake and excretion capacity and intracellular mechanisms of toxicity of metal-based nanoparticles (NPs) using as model system a fish intestinal-hepatic cell in vitro system from rainbow trout. The human colon adenocarcinoma cell line (CaCo2) was additionally used for comparisons and especially to exploit the possibility to use human specific antibodies. A major aim was to investigate if metal NPs influence essential metal homeostasis and copper transporter function. These two aspects were evaluated measuring the intracellular concentration of essential metals (Cu, Zn and Fe) along with the concentrations of exposure metals (e.g. Ag). Intracellular concentrations were linked to transcriptional (copper transporters and metal detoxification genes) and post-translational (ATP7A protein trafficking) changes elicited in cells on exposure to metal-NPs.
The project is divided into three steps: 1) establish polarized in vitro preparations and assess cell viability; 2) quantify uptake and intracellular fate of metal-NPs; and 3) evaluate cellular responses to metal-NPs.
Step 1: The intestinal-hepatic cell culture system was developed. This system enables the evaluation of transport of NPs (or dissolved ions from the NPs) across the intestinal epithelia and the evaluation of effects that these have on the intestinal cells and liver cells growing underneath.
Confocal microscopy analyses showed that RTgutGC grown on transwells for 3-4 weeks develop tight junctions and the cytoskeleton (actin) takes on a morphology typical of polarised epithelial cells. Moreover, RTgutGC cells develop a trans epithelial electrical resistance (TEER) comparable to the fish intestine in vivo. Other features of the fish intestine have also been shown in RTgutGC cells, including expression of Na/K- and V- ATPase at the mRNA and protein level and Villin at the mRNA level.
Step 2 & 3: The initial particle of choice was a citrate-coated silver NP (cit-AgNP; nominal size: 19 nm). Silver-based NPs are particularly well suited to test the effect of metal-NPs on copper homeostasis because ionic silver is known to be transported intracellularly by Cu transporter proteins. In addition, two titanium dioxide NPs were selected as a non-dissolving NP. These were the commercially available Aeroxide P25 (P25-TiO2NPs; size: 21 nm) and a fluorescently labelled TiO2NPs (Al-TiO2NPs; size: 25 nm). The agglomeration behaviour of all NPs was analysed by dynamic light scattering (DLS) in different media: distilled water, a growth medium (L15std; Leibovitz L15 supplemented or not with 5% Foetal Bovine Serum, FBS) and in exposure medium, L15ex, which is L15 without amino acids and vitamins. The agglomeration for all NPs was higher in L15ex (~700 nm for cit-AgNP and ~2000 nm for P25- and Al- TiO2NPs) than in the protein rich L15std (~200 nm for cit-AgNP and ~400 nm for P25- and Al- TiO2NPs), suggesting that proteins might have a role in steric stabilization of NPs agglomerates.
Cell viability and effective concentrations causing 50 % effects (EC50) have been determined in cells using a multiple endpoint assay previously developed in the host lab. Three parameters of cell viability are measured simultaneously on the same set of cells: metabolic activity, membrane and lysosomes integrity. The cytotoxicity of the cit-AgNPs or AgNO3 increased greatly in L15ex in which amino acids and proteins are removed. Remarkably, the endpoint most affected (lower EC50 values) for cit-AgNPs was lysosome integrity (Neutral Red). This result led to the hypothesis that lysosomes are a key target of cit-AgNPs in these cells. This hypothesis was confirmed by two additional experiments: (1) using a lysosome specific molecular probe, LysoTracker, an identical concentration response was obtained as for Neutral Red and (2) by Scanning electron microscopy (SEM) where cit-AgNPs could be visualised inside lysosomes. Elemental analyses by Energy-dispersive X-ray spectroscopy (EDX) confirmed that electron dens particles contained silver (not shown). Both of the TiO2-NPs were not toxic at all doses tested (up to 1.25 mM). However, internalization of the NPs could be shown directly by SEM for P25-TiO2NPs and indirectly by measuring fluorescence in cells exposed to Al-TiO2NPs (not shown). Differently from cit-AgNPs, P25-TiO2NPs accumulated more in endosomes than in lysosomes.
The intestinal-hepatic cell culture system has been exploited to evaluate transport of silver NP or silver ions across RTgutGC cells. Using the viability assays described above it was determined that applying apically 1 μM Ag as AgNO3 or cit-AgNP for 24 hours was non-toxic to the RTgutGC cells while applying 3.4 μM AgNO3 or 10 μM cit-AgNP resulted in a 15% reduction in viability. After exposure, the accumulation of Ag was measured by ICP-MS in RTgutGC and RTL-W1 cells, as well as in the apical and basolateral media. While exposures of 1 μM AgNO3 or AgNP resulted in identical accumulation of Ag in RTgutGC, Ag was transported across more efficiently and accumulated more in RTL-W1 cells on exposure to AgNO3. A reduction in essential elements (Cu, Zn and Fe) was measured in RTgutGC cells exposed to 3.4 μM of AgNO3. In addition, a parallel set of cells was exposed to the same conditions but mRNA levels were measured in cells by qPCR after exposures. Even though identical amounts of Ag accumulated in RTgutGC cells exposed to 1 μM of AgNO3 or cit-AgNP, cells exposed to AgNO3 induced almost twice as much MTb than cells exposed to cit-AgNPs. In addition, cells exposed to AgNO3 induced Zn and Cu transporters, ZnT1 and ATP7B respectively, while cells exposed to cit-AgNP did not. AgNO3 exposure results in activation of MTF1 regulated genes MT and ZnT1 while cit-AgNP appear not to affect this process probably due to different uptake mechanisms, endocytosis vs transporter mediated and slower release of ionic Ag. Others have also shown that Cu and Fe efflux pumps might be regulated by MTF1. This effect could therefore explain the reduction in essential metals in cells exposed to 3.4 μM AgNO3.
Copper ATPase (ATP7A) function was evaluated via monitoring of its trafficking behaviour in human gut cells (Caco2). Exposure to CuSO4 or AgNO3 resulted in redistribution of ATP7A from the Golgi Network to the cells’ periphery. However cit-AgNP did not induce such effect, supporting the hypothesis that at this conditions cit-AgNP do not trigger alterations in Cu homeostasis.
Overall, this study provides new mechanistic understanding on the accumulation and toxicity of Ag and Ti based NPs in RTgutGC cells. The fish intestinal-hepatic cell system has been characterized and proved to be an informative tool to study metal-NP uptake and toxicity. The characterization of the RTgutGC cells will also be of great value for fish toxicologists and physiologists and has potential to became the model of fish intestine.
The project is divided into three steps: 1) establish polarized in vitro preparations and assess cell viability; 2) quantify uptake and intracellular fate of metal-NPs; and 3) evaluate cellular responses to metal-NPs.
Step 1: The intestinal-hepatic cell culture system was developed. This system enables the evaluation of transport of NPs (or dissolved ions from the NPs) across the intestinal epithelia and the evaluation of effects that these have on the intestinal cells and liver cells growing underneath.
Confocal microscopy analyses showed that RTgutGC grown on transwells for 3-4 weeks develop tight junctions and the cytoskeleton (actin) takes on a morphology typical of polarised epithelial cells. Moreover, RTgutGC cells develop a trans epithelial electrical resistance (TEER) comparable to the fish intestine in vivo. Other features of the fish intestine have also been shown in RTgutGC cells, including expression of Na/K- and V- ATPase at the mRNA and protein level and Villin at the mRNA level.
Step 2 & 3: The initial particle of choice was a citrate-coated silver NP (cit-AgNP; nominal size: 19 nm). Silver-based NPs are particularly well suited to test the effect of metal-NPs on copper homeostasis because ionic silver is known to be transported intracellularly by Cu transporter proteins. In addition, two titanium dioxide NPs were selected as a non-dissolving NP. These were the commercially available Aeroxide P25 (P25-TiO2NPs; size: 21 nm) and a fluorescently labelled TiO2NPs (Al-TiO2NPs; size: 25 nm). The agglomeration behaviour of all NPs was analysed by dynamic light scattering (DLS) in different media: distilled water, a growth medium (L15std; Leibovitz L15 supplemented or not with 5% Foetal Bovine Serum, FBS) and in exposure medium, L15ex, which is L15 without amino acids and vitamins. The agglomeration for all NPs was higher in L15ex (~700 nm for cit-AgNP and ~2000 nm for P25- and Al- TiO2NPs) than in the protein rich L15std (~200 nm for cit-AgNP and ~400 nm for P25- and Al- TiO2NPs), suggesting that proteins might have a role in steric stabilization of NPs agglomerates.
Cell viability and effective concentrations causing 50 % effects (EC50) have been determined in cells using a multiple endpoint assay previously developed in the host lab. Three parameters of cell viability are measured simultaneously on the same set of cells: metabolic activity, membrane and lysosomes integrity. The cytotoxicity of the cit-AgNPs or AgNO3 increased greatly in L15ex in which amino acids and proteins are removed. Remarkably, the endpoint most affected (lower EC50 values) for cit-AgNPs was lysosome integrity (Neutral Red). This result led to the hypothesis that lysosomes are a key target of cit-AgNPs in these cells. This hypothesis was confirmed by two additional experiments: (1) using a lysosome specific molecular probe, LysoTracker, an identical concentration response was obtained as for Neutral Red and (2) by Scanning electron microscopy (SEM) where cit-AgNPs could be visualised inside lysosomes. Elemental analyses by Energy-dispersive X-ray spectroscopy (EDX) confirmed that electron dens particles contained silver (not shown). Both of the TiO2-NPs were not toxic at all doses tested (up to 1.25 mM). However, internalization of the NPs could be shown directly by SEM for P25-TiO2NPs and indirectly by measuring fluorescence in cells exposed to Al-TiO2NPs (not shown). Differently from cit-AgNPs, P25-TiO2NPs accumulated more in endosomes than in lysosomes.
The intestinal-hepatic cell culture system has been exploited to evaluate transport of silver NP or silver ions across RTgutGC cells. Using the viability assays described above it was determined that applying apically 1 μM Ag as AgNO3 or cit-AgNP for 24 hours was non-toxic to the RTgutGC cells while applying 3.4 μM AgNO3 or 10 μM cit-AgNP resulted in a 15% reduction in viability. After exposure, the accumulation of Ag was measured by ICP-MS in RTgutGC and RTL-W1 cells, as well as in the apical and basolateral media. While exposures of 1 μM AgNO3 or AgNP resulted in identical accumulation of Ag in RTgutGC, Ag was transported across more efficiently and accumulated more in RTL-W1 cells on exposure to AgNO3. A reduction in essential elements (Cu, Zn and Fe) was measured in RTgutGC cells exposed to 3.4 μM of AgNO3. In addition, a parallel set of cells was exposed to the same conditions but mRNA levels were measured in cells by qPCR after exposures. Even though identical amounts of Ag accumulated in RTgutGC cells exposed to 1 μM of AgNO3 or cit-AgNP, cells exposed to AgNO3 induced almost twice as much MTb than cells exposed to cit-AgNPs. In addition, cells exposed to AgNO3 induced Zn and Cu transporters, ZnT1 and ATP7B respectively, while cells exposed to cit-AgNP did not. AgNO3 exposure results in activation of MTF1 regulated genes MT and ZnT1 while cit-AgNP appear not to affect this process probably due to different uptake mechanisms, endocytosis vs transporter mediated and slower release of ionic Ag. Others have also shown that Cu and Fe efflux pumps might be regulated by MTF1. This effect could therefore explain the reduction in essential metals in cells exposed to 3.4 μM AgNO3.
Copper ATPase (ATP7A) function was evaluated via monitoring of its trafficking behaviour in human gut cells (Caco2). Exposure to CuSO4 or AgNO3 resulted in redistribution of ATP7A from the Golgi Network to the cells’ periphery. However cit-AgNP did not induce such effect, supporting the hypothesis that at this conditions cit-AgNP do not trigger alterations in Cu homeostasis.
Overall, this study provides new mechanistic understanding on the accumulation and toxicity of Ag and Ti based NPs in RTgutGC cells. The fish intestinal-hepatic cell system has been characterized and proved to be an informative tool to study metal-NP uptake and toxicity. The characterization of the RTgutGC cells will also be of great value for fish toxicologists and physiologists and has potential to became the model of fish intestine.