Systematic assessment of antibody selectivity in plasma based on a resource of enrichment profiles

Claudia Fredolini,Janne Lehtiö,Laura Sanchez-Rivera,Sanna Byström,Rui M Branca,Jochen M Schwenk,Peter Nilsson,Davide Tamburro,Marina Ioannou,Fredrik Pontén

doi:10.1038/s41598-019-43552-5

Abstract

There is a strong need for procedures that enable context and application dependent validation of antibodies. Here, we applied a magnetic bead assisted workflow and immunoprecipitation mass spectrometry (IP-MS/MS) to assess antibody selectivity for the detection of proteins in human plasma. A resource was built on 414 IP experiments using 157 antibodies (targeting 120 unique proteins) in assays with heat-treated or untreated EDTA plasma. For each protein we determined their antibody related degrees of enrichment using z-scores and their frequencies of identification across all IP assays. Out of 1,313 unique endogenous proteins, 426 proteins (33%) were detected in >20% of IPs, and these background components were mainly comprised of proteins from the complement system. For 45% (70/157) of the tested antibodies, the expected target proteins were enriched (z-score ≥ 3). Among these 70 antibodies, 59 (84%) co-enriched other proteins beside the intended target and mainly due to sequence homology or protein abundance. We also detected protein interactions in plasma, and for IGFBP2 confirmed these using several antibodies and sandwich immunoassays. The protein enrichment data with plasma provide a very useful and yet lacking resource for the assessment of antibody selectivity. Our insights will contribute to a more informed use of affinity reagents for plasma proteomics assays.

Highlights

Antibodies are important tools used in a wide range of assays within life science, but there is a growing awareness about the importance to carefully validate the data generated[1]
The study included mostly polyclonal binders from the Human Protein Atlas (HPA) and monoclonal antibodies from mouse and other species
This study describes a resource that was built from proteins enriched from plasma and it was applied for the determination of antibody selectivity in plasma

Summary

Introduction

Antibodies are important tools used in a wide range of assays within life science, but there is a growing awareness about the importance to carefully validate the data generated[1]. We found z-scores a convenient statistical approach for the scope of our study, as they determine whether a specific sample represented or deviated from the populations of samples population or if it deviates This approach allowed us to identify the endogenous proteins that were either most uniquely or commonly enriched by each antibody in the analyzed plasma sample. Statistics using z-scores have been widely used in clinical population studies[19] and for the analysis of omics data types[20,21,22] but not in the contest of the analysis of immunoprecipitation data These scores built the foundation of our resource that we used for the systematic evaluation of antibodies selectivity in plasma by IP-MS/MS, where data analysis is often complicated by the high number of proteins commonly identified in each experiment

Objectives

Methods

Results

Conclusion