Cell-Free Protein Expression System for Generating Stable Isotope-labeled Proteins

Rapidly express protein standards for quantitative mass spectrometry

by Ryan Bomgarden, Ph.D.1; Derek Baerenwald, B.S.2; Eric Hommema, M.S.1; Scott Peterman, Ph.D.3; John Rogers, Ph.D.1 - 10/11/12

Stable isotope-labeled peptides are routinely used as internal standards for mass spectrometry (MS) quantification of enzymatically digested protein samples. However, stable isotope-labeled proteins are considered better standards because they control for MS sample-preparation loss and digestion efficiency variation.¹ Traditional in vivo expression systems, such as ¹⁵N-labeled E. coli or SILAC (i.e., stable isotope labeling using amino acids in cell culture), have been used to express recombinant heavy proteins; however, these systems are limited in expressing toxic or insoluble proteins, require two to three days for protein expression, and can have low yield of functional proteins. Also, because in vivo systems use stable isotope-labeled cell lines, all proteins in the cell become isotopically labeled, leading to high waste and cost.

In vitro translation (IVT) is an alternative protein expression method that uses a cellular-extract system to transcribe DNA into mRNA, which is subsequently translated into protein. Most IVT systems use prokaryotic (e.g., bacteria) or non-human eukaryotic (e.g., rabbit reticulocyte) cell extracts;² however, these systems lack the components needed for proper folding and modification of human proteins, have low expression yields, and inefficiently incorporate stable isotope-labeled amino acids. In this study, we used the Thermo Scientific 1-Step Heavy Protein IVT Kit (Part No.88331) to express stable isotope-labeled protein standards for quantitative MS.

The 1-Step Heavy Protein IVT Kit uses a novel human cell-free lysate^3,4 to rapidly express stable isotope-labeled proteins (Figure 1). We used recombinant green fluorescent protein (GFP) to optimize protein expression and heavy amino acid concentrations. Titration and time-course experiments using HeLa lysates supplemented with heavy amino acids showed that amino acid concentrations greater than or equal to 1mM and incubation times longer than four hours were required for optimal protein expression and stable-isotope incorporation of > 90% (Figure 2A and 2B). No difference in GFP protein expression was observed in lystates supplemented with heavy or light amino acids (data not shown).

Figure 2. Heavy green-fluorescent protein expression. Panel A: GFP was expressed for 20 hours with increasing amounts of heavy arginine (Arg10) and lysine (Lys8), and analyzed as described in Figure 1. Panel B: GFP expression with time, showing corresponding heavy-isotope incorporation.

To validate the human heavy IVT system for producing human recombinant proteins, six recombinant proteins were expressed and purified using GST or 6xHIS affinity purification. Although all proteins were expressed as indicated by a western blot (Figure 3A), only four of the six proteins were recovered with high yield after purification and sample preparation. As determined by MS analysis of heavy and light peptides, all expressed proteins had stable-isotope incorporation equal to or greater than 90% (Figure 3B and Figure 4).

Figure 4. Human proteins expressed by in vitro translation have high isotope incorporation. Panel A: MS spectra of stable isotope-labeled GAPDH peptide AGAHLQGGAk. Panel B: Mass spectrometry spectra of stable isotope-labeled cyclin D1 peptide AYPDANLLNDr.

Ideal protein standards are those that are identical to their endogenous counterparts. Expressing recombinant proteins in human cell-free extract systems can aid proper protein folding and post-translational modifications.² During the purification of one human protein, Bcl-2-associated death promoter (BAD), we observed co-purification of light 14-3-3σ with the heavy protein (Figure 5A). This protein-protein interaction is known to be mediated by 14-3-3 binding of serine/threonine phosphorylation motifs (Figure 5B).⁵ MS analysis of the IVT-expressed protein identified three out of four Akt consensus phosphorylation sites (Figure 5C). Overall, these results indicate that recombinant BAD expressed using this human cell-free expression system has functional protein modifications and interactions.

The 1-Step Heavy Protein IVT Kit was used to rapidly express six different full-length, stable-isotope labeled human proteins with >90% isotope incorporation in less than 8 hours. Heavy proteins expressed using this system had functional protein modifications and interactions, which makes them ideal MS standards for endogenous protein quantitation.

Protein expression and purification: For each gene, full-length cDNAs were expressed as C-terminal fusion proteins using the 1-Step Heavy Protein IVT Kit (Part No. 88331). Depending on the C-terminal affinity tag, expressed proteins were purified using the Thermo Scientific Pierce GST Spin Purification Kit (Part No. 16106) or the Thermo Scientific HisPur Cobalt Purification Kit (Part No. 90091). Purified protein samples were separated by SDS-PAGE and stained using Thermo Scientific Pierce GelCode Blue Stain Reagent (Part No. 24590). Gel slices containing each protein were destained, reduced, alkylated and digested to peptides using trypsin for 4-16 hours. The digested peptides were desalted using Thermo Scientific C18 Stage Tips (Cat # SP201), lyophilized and reconstituted with 0.1% trifluoroacetic acid.

Expression of isotopically labeled proteins was performed in IVT reactions using a custom amino acid mix supplemented with the stable isotope-labeled amino acids ¹³C₆¹⁵N₂ L-lysine and ¹³C₆¹⁵N₄ L-arginine.  All reactions were incubated at 30°C for 8-16 hours, unless otherwise noted. Recombinant HIS-GFP protein fluorescence was measured using a GFP standard curve with a Tecan Safire* Fluorometer.

LC-MS/MS Analysis: A NanoLC-2D high-pressure liquid chromatograph (HPLC) with a Thermo Scientific PepMap C18 Column (75 µm ID x 20 cm) was used to separate peptides using a 5-40% gradient (A: water, 0.1% formic acid; B: acetonitrile, 0.1% formic acid) at a flow rate of 300nL/min for 40 minutes. A Thermo Scientific LTQ Orbitrap XL ETD Mass Spectrometer was used to detect peptides using a top-six experiment consisting of single-stage MS followed by acquisition of six MS/MS spectra with collision-induced dissociation (CID) to aid in protein identification.

Data Analysis: MS spectra were searched for matches with a custom human SWISSProt database using Thermo Scientific Proteome Discoverer Software version 1.3, and the SEQUEST* search engine. Static modifications included carbamidomethyl with methionine oxidation. Lysine-8 and arginine-10 were used as dynamic modifications. SILAC ratios were based on the area under the curve (AUC) for each heavy and light peptide, and to determine stable-isotope incorporation.

1. Hanke, S., et al. (2007). Absolute SILAC for accurate quantitation of proteins in complex mixtures down to the attomole level. J Proteom Res 7:1118-30.
2. Ciccimaro, E., et al. (2009). Absolute quantification of phosphorylation on the kinase activation loop of cellular focal adhesion kinase by stable isotope dilution liquid chromatography/mass spectrometry. Anal Chem 81(9):3304-13.
3. Mikami, S., et al. (2008). A human cell-derived in vitro coupled transcription/translation system optimized for production of recombinant proteins. Protein Expr Purif 62(2):190-8.
4. Stergachis, A., et al. (2011). Rapid empirical discovery of optimal peptides for targeted proteomics. Nat Meth 8(12):1041-3.
5. Tzivion, G., et al. (2001). 14-3-3 proteins; bringing new definitions to scaffolding. Oncogene20(44):6331-8.

The data in this article were previously presented at the 2012 American Society for Mass Spectrometry annual meeting in a poster entitled "Development of a human cell-free expression system to generate stable isotope-labeled protein standards for quantitative mass spectrometry".