Impact of polymer-modified gold nanoparticles on brain endothelial cells: exclusion of endoplasmic reticulum stress as a potential risk factor

Abstract

A library of polymer-coated gold nanoparticles (AuNPs) differing in size and surface modifications was examined for uptake and induction of cellular stress responses in the endoplasmic reticulum (ER stress) in human brain endothelial cells (hCMEC/D3). ER stress is known to affect the physiology of endothelial cells (ECs) and may lead to inflammation or apoptosis. Thus, even if applied at non-cytotoxic concentrations ER stress caused by nanoparticles should be prevented to reduce the risk of vascular diseases and negative effects on the integrity of barriers (e.g. blood-brain barrier). We exposed hCMEC/D3 to twelve different AuNPs (three sizes: 18, 35, and 65 nm, each with four surface-modifications) for various times and evaluated their effects on cytotoxicity, proinflammatory mediators, barrier functions and factors involved in ER stress. We demonstrated a time-dependent uptake of all AuNPs and no cytotoxicity for up to 72 h of exposure. Exposure to certain AuNPs resulted in a time-dependent increase in the proinflammatory markers IL-8, MCP-1, sVCAM, sICAM. However, none of the AuNPs induced an increase in expression of the chaperones and stress sensor proteins BiP and GRP94, respectively, or the transcription factors ATF4 and ATF6. Furthermore, no upregulation of the UPR stress sensor receptor PERK, no active splicing product of the transcription factor XBP1 and no upregulation of the transcription factor CHOP were detectable. In conclusion, the results of the present study indicate that effects of different-sized gold nanoparticles modified with various polymers were not related to the induction of ER stress in brain microvascular endothelial cells or led to apoptosis.

Keywords: BiP; blood-brain barrier; cell stress; tight junction proteins; unfolded protein response.

Figure 1. Uptake of various gold nanoparticles by endothelial cells after 24 hours of exposure.

hCMEC/D3 were exposed to 100 µg / mL AuNPs for 24 hours. (A - D) Cells were fixed and membranes were stained with anti-CD31 antibody (red). Nuclei were stained with Hoechst dye (blue). Agglomerates of AuNPs were detected as black dots (optical/fluorescence microscopy, Delta Vision, 60x; scale bar: 10 µm). (A) 35 nm – glucosamine; (B) 35 nm – hydroxypropylamine; (C) 35 nm – taurine; (D) 35 nm – PEG. (E) Cells exposed to various gold nanoparticles were analyzed for internalized gold nanoparticles by ICP-AES. Amounts of AuNPs were calculated per cell. Results are shown as means ± SEM of four independent experiments each performed in triplicate (n = 4). *: P < 0.05, **: P > 0.01; ***: P < 0.001 (TWOway ANOVA with Tukey t-test). (F – I) Cells were exposed to AuNPs and analyzed by transmission electron microscopy (scale bar: 1.9 µm). Higher magnifications are shown in F’ – I’ (scale bar: 200 nm). (F) @glucosamine; (G) @hydroxypropylamine; (H) @taurine; (I) @PEG.

Figure 2. Cytokine secretion of hCMEC/D3 after treatment with different gold nanoparticles.

hCMEC/D3 were treated with 150 µg / mL AuNPs of different sizes and surface modifications. The concentrations of proinflammatory mediators ((A) IL-8, (B) MCP-1, (C) sICAM, (D) sVCAM) were determined using ELISA after 24h, 48h and 72h. TNFα treated cells were used as positive control. Untreated cells were used as control (Ctrl). The data represents the means ± SEM of two independent experiments each performed in triplicate (n = 2). *: P < 0.05, **: P > 0.01; ***: P < 0.001 (TWOway ANOVA with Dunnett t-test).

Figure 3. Expression of tight junction proteins in hCMEC/D3 after treatment with different gold nanoparticles

hCMEC/D3 were treated with 150 µg / mL gold nanoparticles of different sizes and surface modifications for 48 hours. The relative quantification of TJ proteins was determined using real-time PCR. Untreated cells have been used as control. The data represents the means ± SEM. of two independent experiments each performed in triplicate (n = 2). *: P < 0.05 (ONEway ANOVA with Dunnett t-test).

Figure 4. Impact of AuNP treatment on the splicing of transcription factor XBP1 mRNA as an indication of UPR activation.

hCMEC/D3 were treated with various AuNPs for 4 hours. Cells were analyzed for XBP1 mRNA splicing. (A) PCR was performed with XBP1 primers that amplify both variants of XBP1, unspliced (u) and spliced (s) variants. Cells treated with 10 µg / mL (i) or 150 µg / mL (ii) AuNPs were analyzed. Samples were plotted in the following order: glucosamine, hydroxypropylamine, taurine, PEG coated AuNPs. (A’) Tunicamycin was used as positive control for XBP1 splicing. (B) For the same samples (150 µg / mL) the amount of spliced XBP1 was quantified using real-time PCR. Untreated cells have been used as control. The data represents the means ± SEM of two independent experiments each performed in triplicate (n = 2). (TWOway ANOVA with Dunnett t-test).

Figure 5. Expression of UPR-related genes upon exposure to AuNPs.

hCMEC/D3 were treated with various 150 µg / mL AuNPs for 4 and 24 hours. Real-time PCR was used to quantify the expression of UPR related genes: (A) BiP, (B) Grp94, (C) ATF4, (D) PERK, (E) IRE1α, (F) ATF6. The data represents the means ± SEM of two independent experiments each performed in triplicate (n = 2). *: P < 0.05, **: P > 0.01 (TWOway ANOVA with Dunnett t-test).

Figure 6. Expression of CHOP in brain endothelial cells after AuNP treatment.

hCMEC/D3 were treated with various 150 µg / mL AuNPs for various time points. The expression of CHOP was quantified by real-time PCR. Data are shown as relative quantification (RQ). Untreated control was set to 1. The data represents the means ± SEM of two independent experiments each performed in triplicate (n = 2). (TWOway ANOVA with Dunnett t-test).