Proteome encoded determinants of protein sorting into extracellular vesicles

Abstract

Extracellular vesicles (EVs) are membranous structures released by cells into the extracellular space and are thought to be involved in cell-to-cell communication. While EVs and their cargo are promising biomarker candidates, protein sorting mechanisms of proteins to EVs remain unclear. In this study, we ask if it is possible to determine EV association based on the protein sequence. Additionally, we ask what the most important determinants are for EV association. We answer these questions with explainable AI models, using human proteome data from EV databases to train and validate the model. It is essential to correct the datasets for contaminants introduced by coarse EV isolation workflows and for experimental bias caused by mass spectrometry. In this study, we show that it is indeed possible to predict EV association from the protein sequence: a simple sequence-based model for predicting EV proteins achieved an area under the curve of 0.77±0.01, which increased further to 0.84±0.00 when incorporating curated post-translational modification (PTM) annotations. Feature analysis shows that EV associated proteins are stable, polar, and structured with low isoelectric point compared to non-EV proteins. PTM annotations emerged as the most important features for correct classification; specifically palmitoylation is one of the most prevalent EV sorting mechanisms for unique proteins. Palmitoylation and nitrosylation sites are especially prevalent in EV proteins that are determined by very strict isolation protocols, indicating they could potentially serve as quality control criteria for future studies. This computational study offers an effective sequence-based predictor of EV associated proteins with extensive characterisation of the human EV proteome that can explain for individual proteins which factors contribute to their EV association.

Protein Disulfide Isomerase-Enriched Extracellular Vesicles from Bladder Cancer Cells Support Tumor Survival and Malignant Transformation in the Bladder

“…The mechanisms that underlie the packaging of oxidized PDI into EVs by BC cells warrant further clari cation. While the loading of protein cargos within EVs may be partially stochastic, ubiquitination, palmitoylation, and other post-translational modi cations can facilitate the sorting of certain proteins into EVs (46,47). It remains to be determined whether these modi cations are relevant for PDI loading into EVs and whether the oxidized form of PDI is preferentially packaged into EVs under varying levels of ER stress.…”

Deciphering Protein Secretion from the Brain to Cerebrospinal Fluid for Biomarker Discovery

“…Entries with no associated canonical protein sequence or a sequence containing nonstandard amino acids were discarded. Human proteins that have never been detected by any mass spectrometry study according to the ProteomicsDB were excluded as for these proteins it would not be possible to conclude if they are not identified by mass spectrometry studies because of their absence or undetectability. , Subsequently, 2079 brain elevated proteins were retained. We annotated this filtered brain elevated HPA proteome as CSF secreted (positive class) and as non-CSF secreted (negative class) if the proteins were present or absent in the CSF proteome (Figure ).…”

“…The PROSITE protein domain database and the ScanProsite tool were used to identify group enriched sequence pattern motifs and annotate the motif-containing proteins. , The use of sequence-based features largely obtained from prediction tools was preferred over curated database annotations to limit the annotation bias in our results . However, curated protein annotations of ectodomain shedding , and extracellular vesicle (EV) association were taken from previous studies on these properties and investigated to support the interpretation of our model’s findings. Note, however, that these latter annotations were not included in the feature data set used for training the machine learning model.…”

Viral Components Trafficking with(in) Extracellular Vesicles

“…Additionally, it was observed that higher-order oligomerization of proteins increases sorting of the resulting protein aggregates into MVB [69], although the mechanism is unclear. Informatic modelling also identified protein sequences and post-translational modifications that may promote EV loading [70][71][72]. However, there is no experimental evidence for such a scenario.…”