EV trapping - Raman characterization of single tumor-derived extracellular vesicles
Due to the COVID-19 crisis measures the PhD defence of Agustin Enciso Martinez will take place online without the presence of an audience.
The PhD defence can be followed by a live stream.
Agustin Enciso Martinez is a PhD student in the research group Medical Cell Biophysics (MCBP). His supervisors are prof.dr. L.W.M.M. Terstappen and dr. C. Otto from the Faculty of Science and Technology.
Cancer is one of the leading causes of mortality and morbidity around the world, second only to cardiovascular disease. Early detection, effective monitoring of the disease and optimal treatments may bring cancer from a deadly to a chronic controllable disease. Liquid biopsies may effectively detect and monitor cancer through the analysis of body fluids such as blood. Both under pathological and normal physiological conditions, cells release extracellular vesicles (EVs) for communication and waste removal purposes. EVs are particles enclosed by a lipid membrane that enable cellular communication by transferring messages encoded in the biomolecules that make up the EVs, such as lipids, proteins, nucleic acids and sugars. Because cancer cells release EVs that can enter the blood stream, a fraction of the particles in blood will be of tumor origin. Such tumor-derived EVs (tdEVs) may contain specific information about their parental cancer cells and may reflect the state of the patient. Thus, identifying and characterizing tdEVs can provide relevant information for cancer diagnosis, selection of optimal treatment and monitoring treatment response.
The detection and characterization of tdEVs in blood has been, however, challenging due to (1) their small size, which may be as small as 30 nm in diameter, (2) their size and density overlap with other more abundant EVs and non-EV particles in blood, such as lipoprotein particles and (3) the limited knowledge on their chemical composition. Therefore, the utilization of tdEVs as cancer biomarkers will largely benefit from techniques that are able to detect single biological nanoparticles and disclose their physical and chemical features, which can be used to distinguish tdEVs from other particles and gain a better understanding about them.
The aim of this work is to detect and characterize biological nanoparticles in blood, specifically tdEVs, at the single particle level. Hence, this thesis explores various methods that enable, in a novel way, the detection and chemical characterization of individual particles and the discrimination of tdEVs from other EVs and non-EV particles, such as lipoprotein particles, in a label-free manner.
One method explored is the correlation of scanning electron microscopy (SEM) and Raman spectroscopy data that enables the acquisition of high-resolution SEM images and the spatial correlation with chemical information as obtained from Raman micro-spectroscopic imaging. We developed a sample preparation protocol and workflow to study biological samples with correlative SEM-Raman. As a starting point, SEM-Raman characterization was performed on cancer and non-cancer cells (chapter 3). High resolution SEM images of non-labelled and non-metal coated cells as well as Raman images were acquired. Cell morphology is then correlated with chemical information and, by performing multivariate analysis on the Raman spectra, cells of different origins are identified. Next, we downscaled the SEM-Raman approach to EVs (chapter 4) and developed a multi-modal platform for the specific capture of tdEVs on antibody-functionalized stainless-steel substrates. We performed a multi-modal analysis that correlates size, morphology and chemical information of single tdEVs by combining SEM, Raman and atomic force microscopy (AFM) imaging. The correlation of SEM and Raman images enabled to distinguish tdEVs from contaminant particles.
Another method is the development of optical trapping and synchronized Rayleigh and Raman scattering (OT-sRRs) for the detection and characterization of single biological nanoparticles, such as tdEVs, directly in suspension and in a label-free manner. Using a single laser beam, single EVs in suspension are trapped in the laser focal spot. The scattered Rayleigh signal is used to detect when an individual particle is trapped and the synchronously acquired Raman signal enables to assign a Raman spectrum to each trapped particle, thereby, disclosing global chemical composition of each individual particle. The automated continuous trapping and release of single particles enables to probe multiple single particles in the sample fluid. The OT-sRRs method was first validated with monodisperse beads and then applied to EVs (chapter 5). Next, the method was applied to EVs of different origins, including tdEVs, as well as to lipoprotein particles, which are a major component of blood plasma (chapter 6). We showed the Raman fingerprint of each particle type and associated it with relevant biomolecules. By performing principal component analysis (PCA) on a dataset consisting of Raman spectra of all the particles measured, we were able to distinguish tdEVs from red blood cell-derived EVs and lipoprotein particles in a label-free manner and directly in suspension. Additionally, based on the Rayleigh scattering, we estimated the size range of particles measured by our OT-sRRs method.
After having analyzed tdEVs from prostate cancer cell lines and plasma of healthy donors, we analyzed plasma of metastatic castration-resistant prostate cancer (mCRPC) patients and compared it to healthy controls at the individual particle level by means of OT-sRRs (chapter 7). PCA and hierarchical cluster analysis showed that the particles of some patients had nucleic acids that contributed significantly to their Raman fingerprint. Although this pilot study aimed at detecting and characterizing single tdEVs in mCRPC patients, significant differences were observed in terms of the general particle profile in the plasma fraction of some patients compared to healthy controls.
An interesting finding derived from studying cells and EVs with correlative SEM-Raman was the interaction of organosilicon compounds with biological membranes (chapter 8). We show some results that indicate the incorporation of organosilicon compounds in lipid membranes, which may affect sample preparation, experimental outcome and perhaps human health.
In conclusion, this thesis describes the implementation of various novel methods to study biological nanoparticles in blood, from cancer cells to tdEVs and from model nanoparticles to nanoparticles in the plasma of cancer patients. These developments open an avenue not only to exploit the potential of tdEVs as cancer biomarkers, but also to study other particles in body fluids and, with that, the general nanoparticle profile, which may be affected under pathological conditions such as cancer.