Interaction of Iron Oxide Nanoparticles with Macrophages Is Influenced Distinctly by “Self” and “Non-Self” Biological Identities

Upon contact with biological fluids like serum, a protein corona (PC) complex forms on iron oxide nanoparticles (IONPs) in physiological environments and the proteins it contains influence how IONPs act in biological systems. Although the biological identity of PC–IONP complexes has often been studied in vitro and in vivo, there have been inconsistent results due to the differences in the animal of origin, the type of biological fluid, and the physicochemical properties of the IONPs. Here, we identified differences in the PC composition when it was derived from the sera of three species (bovine, murine, or human) and deposited on IONPs with similar core diameters but with different coatings [dimercaptosuccinic acid (DMSA), dextran (DEX), or 3-aminopropyl triethoxysilane (APS)], and we assessed how these differences influenced their effects on macrophages. We performed a comparative proteomic analysis to identify common proteins from the three sera that adsorb to each IONP coating and the 10 most strongly represented proteins in PCs. We demonstrated that the PC composition is dependent on the origin of the serum rather than the nature of the coating. The PC composition critically affects the interaction of IONPs with macrophages in self- or non-self identity models, influencing the activation and polarization of macrophages. However, such effects were more consistent for DMSA-IONPs. As such, a self biological identity of IONPs promotes the activation and M2 polarization of murine macrophages, while a non-self biological identity favors M1 polarization, producing larger quantities of ROS. In a human context, we observed the opposite effect, whereby a self biological identity of DMSA-IONPs promotes a mixed M1/M2 polarization with an increase in ROS production. Conversely, a non-self biological identity of IONPs provides nanoparticles with a stealthy character as no clear effects on human macrophages were evident. Thus, the biological identity of IONPs profoundly affects their interaction with macrophages, ultimately defining their biological impact on the immune system.


Study of Corona Formation in different types of biological sera.
To study the process of PC formation, we incubated APS-, DEX-or DMSA-IONPs (125 μg Fe/ml) in DMEM supplemented with a 10% FBS, MS or HS at 37 °C for different times (0, 1, 3, 5, 10, 24, 48 or 72 h). The hydrodynamic size of the IONPs incubated with the different sera was measured by DLS using a NanoSizer ZS (Malvern) 1 .
IONP treatment and toxicity assay. Cells were exposed to the IONPs (125 μgFe/ml unless otherwise stated) and their viability was determined with the colorimetric PrestoBlue assay

Analysis of IONP uptake by macrophage cells
Perls' Prussian blue staining. For iron staining, the cells were washed with PBS after incubation with the APS-, DEX-or DMSA-IONPs (125 μgFe/ml) in DMEM or RPMI with 10% of FBS, MS or HS. After IONP treatment, the cells were fixed in 4% paraformaldehyde (PFA) for 15 min, permeabilized with 0.05% Triton X-100 for 5 min, stained with an equal volume of HCl 4% and potassium ferrocyanide trihydrate 4% for 30 min, and counterstained with neutral red 0.5% for 2 min. The cells were then washed with distilled water, air-dried and mounted in medium (7.7% gelatin and 54% glycerol). Images were acquired on an Olympus IX70 inverted bright field microscope with a 63x oil objective.
Confocal microscopy imaging of internalized IONPs. The macrophage cells were grown on poly-lysine coated coverslips for 24 hours and then exposed to APS-, DEX-or DMSA-IONPs (125 μgFe/ml) for 24 h in the different sera. Subsequently, the cells were treated for 2 h at 37 °C with LysoTracker green (1:400 in culture medium: Invitrogen) and then for 30 min at RT with WGA (1:200 in PBS: Invitrogen). The cells were then washed, fixed with 4% PFA (15 min) and then counterstained for 10 min at RT with DAPI (1:500 in PBS: Sigma). Finally, the samples were mounted in Fluoromount G (ThermoFisher) and images were acquired on a confocal multispectral Leica TCS SP5 system, with a 63X/1.4 NA oil objective and a 3X zoom.

List of primers for real-time quantitative PCR (RT-qPCR)
The primers used (all from Sigma) are indicated in Table S1.

SUPPLEMENTARY RESULTS
Physicochemical characterization of IONPs. The main physicochemical characteristics of the IONPs used in this study are summarized in Figure S1, characterizing their hydrodynamic size in water.

Evaluation of IONPs stability with different coatings according to species of the serum.
The stability of all three IONPs was evaluated by obtaining DLS measurements after the incubation of the APS-, DEX-or DMSA-IONPs in medium alone or in medium (DMEM or RPMI) supplemented with different biological sera for 24 h. In Figure S3, there are no signs of agglomeration during the 24 h period as the distribution is homogeneous. We did not detect IONPs sedimentation caused by large agglomerates of IONPs; we did not detect multiple peaks in the hydrodynamic size distributions that would indicate hetero-aggregate formation, as shown below. The results obtained are shown in Figure S3. In terms of the classes of proteins identified, three in particular were selected: apolipoproteins, complement proteins and immunoglobulins. Again, focusing on these confirmed that the greatest differences in the PC are marked by the biological serum and not the IONP coating ( Figure S4B). S7 IONP toxicity. The IONP treatment and type of sera did not affect RAW 264.7 and THP1 cell viability in PrestoBlue assays ( Figure S5) and hence, the optimal concentration selected was 125 μg/ml for both cell types. Quantification of the iron concentration in THP1 cells by ICP-OES. The data (mean ± SD) are representative of three independent experiments. One-way analysis of variance (ANOVA) and a Student's t-test were used to assess the ICP-OES data, and the asterisks indicate significant differences: *p < 0.05 and **p < 0.01. Legend: FBS (fetal bovine serum), MS (mouse serum) and HS (human serum).
To verify the uptake of the IONPs with different coatings observed by ICP-OES in macrophage cells, Prussian Blue iron staining with and neutral red counterstaining was performed. After a 24 h incubation with the IONPs in the different sera, we observed the internalization differed depending on the IONP coating and the type of biological serum (see Figure S7). In summary, for RAW264.7 cells we observed differences in the influence of PC derived from FBS and MS sera for both markers used (Figure S8A, B), however, in THP1 cells no differences were shown in the expression of CD80 and CD86 derived from the influence of PC formation in different biological sera (Figure S8C, D).
Influence of the type of biological serum on macrophage migration. The differences in RAW macrophage cell migration were analyzed depending on the type of biological serum with which the culture medium was supplemented. To analyze the migration of RAW cells in DMEM supplemented with 10% FBS or MS, a wound closure assay was used and migration was analyzed over 24h. The directional migration index (DMI) was obtained using the formula indicated in the Materials and Methods 2 . These results showed significant differences in the migration of RAW 264.7 cells when cultured in medium supplemented with 10% FBS or MS, with enhanced migration when cells were cultured in medium supplemented with FBS (see Figure S9). Figure S9. The directional migration index (DMI) of RAW264.7 cells cultured in medium supplemented with different sera. The DMI was measured with Image J software from three independent experiments. One-way analysis of variance (ANOVA) and a Student's t-test were used and the asterisks indicate significant differences: *p < 0.05, **p < 0.01, and ***p < 0.001.

S11
Analysis of mitochondrial morphology according to the type of biological serum used after exposure to DMSA-IONPs. To analyze ROS production, the mitochondrial morphology was assessed and striking changes were observed, mainly in the RAW 264.7 murine macrophages when exposed to DMSA-IONPs in medium supplemented with FBS or MS. Figure S10. Study of mitochondrial morphology in RAW 264.7 cells treated with DMSA-IONPs with different PCs depending on the type of biological serum. Representative TEM images of mitochondria (in red) in RAW 264.7 cells exposed to DMSA-IONPs in medium supplemented with 10% FBS or MS. Scale bar: 2-5 µm.