Supervised Learning and Mass Spectrometry Predicts the in Vivo Fate of Nanomaterials

Abstract

The surface of nanoparticles changes immediately after intravenous injection because blood proteins adsorb on the surface. How this interface changes during circulation and its impact on nanoparticle distribution within the body is not understood. Here, we developed a workflow to show that the evolution of proteins on nanoparticle surfaces predicts the biological fate of nanoparticles in vivo. This workflow involves extracting nanoparticles at multiple time points from circulation, isolating the proteins off the surface and performing proteomic mass spectrometry. The mass spectrometry protein library served as inputs, while blood clearance and organ accumulation were used as outputs to train a supervised deep neural network that predicts nanoparticle biological fate. In a double-blinded study, we tested the network by predicting nanoparticle spleen and liver accumulation with upward of 94% accuracy. Our neural network discovered that the mechanism of liver and spleen uptake is due to patterns of a multitude of nanoparticle surface adsorbed proteins. There are too many combinations to change these proteins manually using chemical or biological inhibitors to alter clearance. Therefore, we developed a technique that uses the host to act as a bioreactor to prepare nanoparticles with predictable clearance patterns that reduce liver and spleen uptake by 50% and 70%, respectively. These techniques provide opportunities to both predict nanoparticle behavior and also to engineer surface chemistries that are specifically designed by the body.

In situanalysis of liposome hard and soft protein corona structure and composition in a single label-free workflow

“…These attempts are complicated by possible nanomaterial-induced conformational changes in the bound proteins, 7 which may escape the traditional methods of preclinical toxicology and can only be observed in in vivo. 8 It was recently demonstrated that the corona controls NP interactions with immune cells and pre-coating with artificial coronas prepared in vitro 15 or in vivo 16 can be employed to extend their blood circulation times and control biodistribution. Identical NPs may also form coronas with different sizes and compositions in different individuals, due to the effect of individual variation and disease on plasma composition or protein structure.…”

Biological Behavior Regulation of Gold Nanoparticles via the Protein Corona

“…Based on the results thus far, we see some contradictory observations, suggesting that it remains great necessity to further unravel how the PC formation impacts the GNP biological behaviors. Furthermore, the recent successes in the combined photodynamic therapy and photothermal therapy of cancer by PC‐formed GNRs in small animals, [ 73 ] a prolonged stay of GNPs in cells (24 h) induced by the specific adsorption of the concanavalin A (ConA) using a glucosamine surface modifier, [ 74 ] and the protein evolution study on the NP surface and its application in predicting the biological fate of GNPs in vivo [ 75 ] demonstrate that the utilizing and modulating the nature of the PC on the GNP surface would be an important strategy to regulate the performance of GNP‐based biosystems in the future.…”

“…Influencing liver accumulation and hepatotoxicity [43a] HSA or apo E Increased organ targeting [66a] Tunable plasma lipoproteins (TPLs) Improved transportation across the bloodbrain barrier [67a] In vivo PC in rats Different half-life, spleen and liver accumulation dependent on patterns of a multitude of NP surface-adsorbed proteins [75] Tumor targeting BSA/HSA Nonsmall cell lung cancer (NSCLC) cell lineand colorectal cancer cell (HCT116)-specific target [4c] 10%, 50%, and 100% FBS Diverse targeting ability according to ligand property (RGD vs transferrin) and particle size against human brain glioma U87cells [72] Mouse serum Enhanced drug delivery and cancer treatment in Call 27 oral squamous cell carcinoma (OSCC) cell-based small animal model in vivo [73] and its application in predicting the biological fate of GNPs in vivo [75] demonstrate that the utilizing and modulating the nature of the PC on the GNP surface would be an important strategy to regulate the performance of GNP-based biosystems in the future.…”

Role of Artificial Intelligence and Machine Learning in Nanosafety

“…[62] Image profiling using open source CellProfiler and a CNN-based algorithm ilastik, [64] were employed by Ilet et al to study nanoparticle distributions. [65] A particularly impressive use of ML in nanosafety was published recently by Lazerovits et al [66] They conducted experiments aimed at understanding the adsorption of blood proteins on nanoparticles immediately after intravenous injection, how this interface changes during circulation, and how it alters to distribution of nanoparticles in vivo. They showed that the evolution of proteins on nanoparticle surfaces predicts the biological fate of nanoparticles in vivo.…”