Nanoscale imaging of tumor cell exosomes by expansion single molecule localization microscopy (ExSMLM).

Tumor cell exosomes play a very important role in the process of tumor cell proliferation and metastasis. However, due to the nanoscale size and high heterogeneity of exosomes, in-depth understanding of their appearance and biological characteristics is still lacking. Expansion microscopy (ExM) is a method that embeds biological samples in a swellable gel to physically magnify the samples to improve the imaging resolution. Before the emergence of ExM, scientists had invented several super-resolution imaging techniques that could break the diffraction limit. Among them, single molecule localization microscopy (SMLM) usually has the best spatial resolution (20-50 nm). However, considering the small size of exosomes (30-150 nm), the resolution of SMLM is still not high enough for detailed imaging of exosomes. Hence, we propose a tumor cell exosomes imaging method that combines ExM and SMLM (i.e. Expansion SMLM, denoted as ExSMLM), which can realize the expansion and super-resolution imaging of tumor cell exosomes. In this technique, immunofluorescence was first performed to fluorescently label the protein markers on the exosomes, then the exosomes were polymerized into a swellable polyelectrolyte gel. The electrolytic nature of the gel made the fluorescently labeled exosomes undergo isotropic linear physical expansion. The expansion factor obtained in the experiment was about 4.6. Finally, SMLM imaging of the expanded exosomes was performed. Owing to the improved resolution of ExSMLM, nanoscale substructures of closely packed proteins were observed on single exosomes, which has never been achieved before. With such a high resolution, ExSMLM would have a great potential in detailed investigation of exosomes and exosome-related biological processes.

Predicting Alzheimer's Disease Conversion From Mild Cognitive Impairment Using an Extreme Learning Machine-Based Grading Method With Multimodal Data.

Identifying patients with mild cognitive impairment (MCI) who are at high risk of progressing to Alzheimer's disease (AD) is crucial for early treatment of AD. However, it is difficult to predict the cognitive states of patients. This study developed an extreme learning machine (ELM)-based grading method to efficiently fuse multimodal data and predict MCI-to-AD conversion. First, features were extracted from magnetic resonance (MR) images, and useful features were selected using a feature selection method. Second, multiple modalities of MCI subjects, including MRI, positron emission tomography, cerebrospinal fluid biomarkers, and gene data, were individually graded using the ELM method. Finally, these grading scores calculated from different modalities were fed into a classifier to discriminate subjects with progressive MCI from those with stable MCI. The proposed approach has been validated on the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort, and an accuracy of 84.7% was achieved for an AD prediction within 3 years. Experiments on predicting AD conversion from MCI within different periods showed similar results with the 3-year prediction. The experimental results demonstrate that the proposed approach benefits from the efficient fusion of four modalities, resulting in an accurate prediction of MCI-to-AD conversion.