Low uptake of silica nanoparticles in Caco-2 intestinal epithelial barriers

Cellular barriers, such as the skin, the lung epithelium or the intestinal epithelium, constitute one of the first obstacles facing nanomedicines or other nanoparticles entering organisms. It is thus important to assess the capacity of nanoparticles to enter and transport across such barriers. In this work, Caco-2 intestinal epithelial cells were used as a well-established model for the intestinal barrier, and the uptake, trafficking and translocation of model silica nanoparticles of different sizes were investigated using a combination of imaging, flow cytometry and transport studies. Compared to typical observations in standard cell lines commonly used for in vitro studies, silica nanoparticle uptake into well-developed Caco-2 cellular barriers was found to be very low. Instead, nanoparticle association to the apical outer membrane was substantial and these particles could easily be misinterpreted as internalised in the absence of imaging. Passage of nanoparticles through the barrier was very limited, suggesting that the low amount of internalised nanoparticles was due to reduced uptake into cells, rather than a considerable transport through them.


Protein corona
To isolate and characterise hard corona proteins, solutions of 100 μg/ml 50 and 150 nm SiO 2 -NPs in 0%, 10%, 55% and 80% FBS in HBSS were prepared. After  Free dye electrophoresis To screen nanoparticles for the eventual presence of a fraction of free labile dye, the silica nanoparticles used for this study (50 and 150 nm; non-modified surface) together with commercial nanoparticles of similar sizes (50 and 100 nm Kisker silica nanoparticles; non-modified surface) were subjected to SDS-PAGE electrophoresis, following methods previously described [1][2][3]. All the different SiO 2 -NP batches were tested at 25 μg/ml in de-ionised water at room temperature, and the samples loaded on the gel by mixing the particle solution with 1/3 volume of SDS-DTT loading buffer S3 (10:1). Samples were run at 130 V for 1 h and a gel fluorescence image was obtained using a Typhoon scanner.

Transepithelial electrical resistance
Transepithelial electrical resistance (TEER) measurements were performed in order to determine the integrity of the Caco-2 barriers. As described in the Experimental section of the main text, Caco-2 barriers were equilibrated in HBSS buffer for 30-60 min, and then exposed to the nanoparticles dispersed in HBSS supplemented with 0%, 10%, 55% and 80% FBS at 37 °C for 6 h. Then the cells were carefully washed with pre-warmed HBSS three times, and the TEER immediately measured using an Endome chamber (World-Precision Instruments Inc., New Haven, CT, USA). Barriers treated with 0%, 10%, 55% and 80% FBS HBSS solutions only (without nanoparticles) were also characterised after 6 h exposure in order to discriminate the effects of the serum from that of the nanoparticles. Untreated controls in cDMEM were also measured. The resistance of blank transwells was subtracted from all values.
The results are presented as the change of the TEER value with respect to that of untreated cells in cDMEM (TEER of treated / TEER of untreated). The average TEER of three replicates is shown and error bars represent the standard error (SEM).

Transport study
Transport studies were performed in transwell systems on Caco-2 barriers cultured for 21 days. After replacing the medium with pre-warmed fresh HBSS buffer (0.5 ml and 1.5 ml in the apical and basal chamber, respectively), the Caco-2 monolayer was incubated at 37 °C for 60 min in an orbital shaker (Titramax 1000) at 100 RPM. The  Table 1. These examples follow the naive assumption that incubation with more proteins results in more enriched protein coronas formed on the surface of the nanoparticles.
Interestingly, more complicated behaviour can, however, be seen for other proteins.
Thus, in some cases, adsorption decreases at the highest serum concentration (for S9 example, band 1-2 for 150 nm SiO 2 -NPs), while other proteins exhibit the complete reverse behaviour and are less abundant as serum concentration increases (for example, band 10-11 for 50 nm SiO 2 -NPs). This is consistent with previous reports for similar nanoparticles in plasma [5], where it was also found that the corona composition varied with protein content. Such observations demonstrate the complexity of the corona formation, while emphasizing the importance of fully specifying the exposure conditions in nanoparticle studies, since exposure to different coronas may result in different outcomes on cells. results are from three independent experiments, exhibiting quantitative differences but, generally, the same trend of a larger nanoparticle association to barriers cultured for 4 days compared to those cultured for 21 days, as well as a trend of larger association in the absence of serum. The asterisk (*) in panel c indicates that the actual value is higher than the one measured and presented due to the limited range of the instrument.
Note that the two nanoparticles are loaded with different amounts of fluorescent dye so no (direct) absolute comparison may be made between the two sizes. S14 Figure S9: Nanoparticle association with Caco-2 barriers after removal of nanoparticle source. Caco-2 barriers (cultured for 21 days) were exposed for 6 h to three independent experiments, exhibiting quantitative differences but, generally, the same trend of a lower nanoparticle association after culturing in the absence of nanoparticles. S15 Figure S10: Surface association of 50 nm SiO 2 -NPs exposed for 6 h to Caco-2 barriers cultured for 21 days. A large amount of nanoparticles was found outside the cells, while very few were found intracellularly. Figure S11: Confocal fluorescence image of a cross-section of Caco-2 barriers exposed to 50 nm SiO 2 -NPs. Caco-2 barriers were exposed to 100 μg/ml 50 nm SiO 2 -NPs for 6 h in medium with 80 % FBS. The nanoparticles are shown in green.
Lysosomes and nuclei were stained by LAMP1 antibody (red) and DAPI (blue), respectively. The arrow indicates nanoparticle fluorescence close to the basal membrane, which could be suggestive of a nanoparticle (or several within one fluorescent spot) potentially having or being about to translocate the barrier. S16 Figure S12: Nanoparticle transport across Caco-2 barriers cultured for 21 days.