Dual Fractions Proteomic Analysis of Silica Nanoparticle Interactions with Protein Extracts

Abstract

Dual-fraction proteomics reveals a novel class of proteins impacted by nanoparticle exposure.

Background: Nanoparticles (NPs) interact with cellular proteomes, altering biological processes. Understanding these interactions requires comprehensive analyses beyond solely characterizing the NP corona.

Methods: We utilized a dual-fraction mass spectrometry (MS) approach to analyze both NP-bound and unbound proteins in Saccharomyces cerevisiae sp. protein extracts exposed to silica nanoparticles (SiNPs). We identified unique protein signatures for each fraction and quantified protein abundance changes using spectral counts.

Results: Strong correlations were observed between protein profiles in each fraction and non-exposed controls, while minimal correlation existed between the fractions themselves. Linear models demonstrated equal contributions from both fractions in predicting control sample abundance. Combining both fractions revealed a larger proteomic response to SiNP exposure compared to single-fraction analysis. We identified 302/56 proteins bound/unbound to SiNPs and an additional 196 "impacted" proteins demonstrably affected by SiNPs.

Conclusion: This dual-fraction MS approach provides a more comprehensive understanding of nanoparticle interactions with cellular proteomes. It reveals a novel class of "impacted" proteins, potentially undergoing conformational changes or aggregation due to NP exposure. Further research is needed to elucidate their biological functions and the mechanisms underlying their impact.

Keywords: corona; mass spectrometry; protein extracts; protein–nanoparticle interactions; proteomics; silica nanoparticles.

Figure 1

Adsorption isotherm of YPE on silica NP (1 g.L⁻¹) in phosphate buffer (100 mM, pH7) conducted by depletion. Ten samples at YPE concentration from 2.5 × 10⁻² to 2 g.L⁻¹ were incubated with silica NP using a ThermoMixer^® at 20 °C for 3 h (cycles of 15 s at 800 rpm followed by 285 s at rest). Samples were centrifuged at 20 °C, 20,000 rpm for 10 min, and the supernatant concentration (unbound proteins) was determined using the absorbance at 205 nm with an absorption coefficient of 31 L.g⁻¹.cm⁻¹. Horizontal and vertical error bars represent the standard error of the mean.

Figure 2

Calibration curve for the YPE concentration with SDS 1%. Concentration levels are determined using the absorbance at 205 nm with an absorption coefficient of 31 L.g⁻¹.cm⁻¹. The blue curve depicts the linear regression model fitted to the data points. Horizontal and vertical error bars represent the standard error of the mean.

Figure 3

Spectral counts correlation plot between the pellet, supernatant and control fractions. Pearson correlation coefficients between replicates are calculated and depicted as squares wherein the size and color (scale given on the right of the plot) depend on its value. Average correlation coefficient values between fraction pairs are indicated on the lower triangular matrix.

Figure 4

Spectral Counts (SCs) linear regression model. This plot depicts the linear regression as a blue line between the cumulated SC of each protein in the pellet and supernatant, with the SC of these proteins in the control (see Equation (2)). Besides the regression line, each dot depicts the related SC in the control (y-axis), cumulated pellet, and supernatant (x-axis). The color of the dot is red when the protein is more abundant in the pellet than in the supernatant, and otherwise, it is blue.

Figure 5

Venn diagram of the number of detected proteins in the pellet, supernatant and control fractions. The relative percentage is indicated in parentheses.