Skip to main content
To KTH's start page

Multiplexed absolute quantification of blood plasma proteins using targeted proteomics

Time: Fri 2022-06-10 10.00

Location: J3:12 Nanna Svartz, BioClinicum, Solnavägen 30, Solna

Video link: https://kth-se.zoom.us/j/63718728720

Language: English

Subject area: Biotechnology

Doctoral student: David Kotol , Systembiologi, Science for Life Laboratory, SciLifeLab

Opponent: Professor Roman Zubarev,

Supervisor: Professor Mathias Uhlén, Systembiologi, Science for Life Laboratory, SciLifeLab, Albanova VinnExcellence Center for Protein Technology, ProNova; Dr. Edfors Fredrik, Systembiologi

Export to calendar

QC 2022-05-16

Abstract

Proteins are the molecular building blocks of all living organisms. They are the functioning actors in metabolism and communication, and they give specific architecture and purpose to all tissues. Their abundance is dynamic and can be interpreted and associated to different physiological states or diseases. Blood plasma, the liquid component of blood, is a proximal fluid to all organs in the human body. As a consequence, it contains an immense amount of information about wide variety of biological processes. This makes it a great sample to study but also demanding due to the complexity and dynamic range of protein concentrations. A promising technology capable of extracting the information is targeted proteomics. Especially in combination with heavy labelled standards it combines liquid chromatography and tandem mass spectrometry (LC-MS/MS) into state-of-the-art quantification on an absolute scale. This toolbox is capable of detecting the smallest perturbations thanks to its precision and multiplexing capability. In this thesis, I describe our efforts to develop and use a novel platform for targeted proteomics suitable for clinical applications with high demands on accuracy and reliability.

In Paper I, a deep dive into the resource of standards coming from The Human Protein Atlas was performed. It was possible to identify their great analytical performance and generate LC-MS/MS coordinates for over 25,000 proteotypic peptides from over 10,000 proteins. A proof-of-concept experiment was designed based on the screening results where 23 proteins in blood plasma were absolutely quantified. The data allowed us to identify distinct clusters of individuals based on their LPA, APOE, SERPINA5 and TFRC levels. 

Paper II directly builds on the findings from Paper I and applies them on a well-defined cohort of 94 individuals whose blood plasma samples were collected over one-year period in four different time-points. The longitudinal protein profiling allowed to assess the variability of plasma proteome within individuals with the developed assay. The applied assay included 52 target proteins with mainly actively secreted liver proteins and 21 FDA approved targets. The results were benchmarked with clinical data for APOA1 and APOB and showed high correlation. This suggests that targeted proteomics and multiplex absolute quantification has an appealing potential as an alternative to antibody-based assays.

In Paper III, the labelled standards underwent a thorough investigation in regard to their stability and reproducibility of quantification. Also, a novel formulation of standards was devised. Here, the standards were stored in chaotropic agent and subsequently dried down. This allows for addition only protocols, upscaling, increased reproducibility, and storage at ambient temperature for at least one month with retained analytical performance. This novel workflow was used for multiplex quantification of 100 clinically relevant protein targets in blood plasma using almost 300 peptides with median coefficient of variation below 10%.

Paper IV expands on Paper III and applies the technology to profile 1,800 individuals in a pan-cancer study. The blood plasma was collected from individuals suffering from one of 15 different cancer types. The applied assay targeted 253 proteins and was able to quantify 1,013 peptides with low variation across the four months of analysis time. Several potential biomarkers for multiple myeloma were identified with the most useful targets consisting of the complement complex C1 and the JCHAIN and CD5L proteins.

urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-312258