Label-Free and Standard-Free Absolute Quantitative Proteomics Using the “Total Protein” and “Proteomic Ruler” Approaches

Abstract

Understanding biological systems and their variation upon stimuli requires knowledge on their composition, primarily including information on organization and dynamics of proteomes. The total protein approach (TPA) is a label- and standard-free method for absolute protein quantitation of proteins using large-scale proteomic data. The method relies on the assumption that the total MS signal from all identified proteins in the dataset reflects—in a biochemical sense—the total protein and the MS signal from a single protein corresponds its abundance in the studied sample. The method offers an easy way to quantify thousands of protein per sample. A related method, the “Proteomic Ruler,” enables conversion of the protein abundance data calculated by TPA to compute numbers of protein copies per cell. TPA and the Proteomic Ruler are powerful tools for studying dynamics of cell architecture.

Introduction

In the bottom-up proteomics, there are two general approaches allowing identification of changes or differences between biological systems. The first one provides relative information on alteration between samples. This is most frequently assessed by differential labeling of samples with stable isotopes, which allows comparison of abundances of peptides and in consequence proteins between samples. Typical examples of such approaches are metabolic labeling (SILAC and Super SILAC) and chemical labeling involving iTRAQ and TMT technologies. In addition, there are several label-free relative quantitation approaches. Relative changes between independent analyses are difficult to compare and the relative data cannot be directly correlated to a physiological status of a biological system. In contrast this is possible using “absolute” quantitation methods. In these approaches stable isotope-labeled standards, usually peptides at known concentrations, are spiked into samples before MS analysis. Measurement of the proportion between the abundances of peptide originating from the sample and the standard allows calculation of the peptide concentration. The concentration of single or averaged concentrations of different peptides is used for assessing protein titers. This approach is also known as the “targeted,” because the standards have to be designed before proteomic analyses and therefore such analyses can cover only a limited number of proteins. To circumvent this technological constrain several label-free computational methods for absolute quantification of proteins have been proposed.

Label-free methods comprise approaches using intensities of MS¹ spectra, such as iBAQ (Schwanhausser et al., 2011, Schwanhausser et al., 2013) and Top3 (Silva, Gorenstein, Li, Vissers, & Geromanos, 2006) and those based on counting of MS² spectra, the emPAI (Ishihama et al., 2005) and APEX (Braisted et al., 2008) methods (Fig. 1). Each of these label-free methods requires a biochemical input for calculation of the protein abundances. This is either determination of the total amount of the analyzed sample or a use of protein standards with defined concentrations. A direct comparison of these methods showed that the TOP3 methods are the most reliable one (Ahrne, Molzahn, Glatter, & Schmidt, 2013).

In contrast to these methods, another computational procedure, the “total protein approach” (TPA), allows absolute protein quantitation without any biochemical input. In the TPA method calculation of protein abundance is based on spectral intensities acquired in the large-scale proteomic analyses (Wisniewski et al., 2012, Wisniewski and Rakus, 2014). The method does not require any specific knowledge on the sample and is standard free (Fig. 1). Therefore it can be applied to any large dataset including archival data. In large-scale analyses a related method, the “Proteomic Ruler,” allows conversion of TPA abundance values into protein copy numbers per cell (Wisniewski, Hein, Cox, & Mann, 2014) (Fig. 2).

This chapter describes principles and applications of TPA and the Proteomic Ruler for absolute quantitative proteomics (Fig. 2). Combining both categories allows studying proteomes beyond the almost trivial “differential display” type analyses. In addition, consistency of the absolute proteomic data with other biochemical parameters such as total protein, nucleic acid content, or enzymatic activity provides novel research tools for studying biological systems.