Quantitative Analysis of the AKT/mTOR Pathway Using Multiplex Immunoprecipitation and Targeted Mass Spectrometry

Examine signaling pathways with targeted proteomics

(See a list of the products featured in this article.)

The AKT/mTOR pathway plays a central role in tumor progression and cancer-drug resistance, and therefore the quantitative measurement of this pathway’s protein expression and posttranslational modifications (PTMs) is vital to cancer research [1]. A major limitation when measuring AKT/mTOR pathway protein levels is the lack of rigorously validated methods and reagents, as well as a reliance on semiquantitative results from western blot analyses. Because many biologically relevant proteins are present in vanishingly small quantities, immunoprecipitation (IP) is commonly used as a tool for enriching protein [2,3]. Mass spectrometry (MS) is increasingly becoming the detection method of choice for determining protein abundance and identifying PTMs.

The IP-MS workflow, which combines IP steps with subsequent analysis by MS, can be used to enrich signaling proteins, benchmark antibody performance, and reveal protein– protein interactions [4]; see “Comprehensive strategy for antibody validation” in BioProbes 75 Journal of Cell Biology Applications. Multiplex IP coupled with targeted MS (mIP-tMS) further enhances this workflow by simultaneously quantifying multiple proteins and their phosphorylation states in a specific signaling pathway. Here we demonstrate the mIP-tMS methodology by analyzing a specific set of protein targets in the AKT/mTOR pathway.

Singleplex IP-MS assay development for AKT/mTOR pathway proteins

To begin our analysis of the AKT/mTOR pathway, we selected 11 pathway proteins—10 of which have at least one phosphorylation site—for antibody validation by IP-MS (Table 1). These targets were chosen to align with commercially available reagents for western blot analysis (WB), enzyme-linked immunosorbent assays (ELISA), and Luminex® bead-based multiplex immunoassays (Luminex assays). Figure 1 shows the general workflow for mIP-tMS assay development.

The first step was to validate antibodies by IP-MS and identify quantotypic peptides for each AKT/mTOR pathway protein. Antibody candidates were screened by IP-MS to determine their effectiveness—both their ability to immunoprecipitate AKT pathway proteins and their usefulness when combined with MS. The IP-enriched samples were then analyzed by LC-MS to qualitatively identify targets of interest, interacting proteins, PTMs, and quantotypic peptides. We performed the LC-MS analysis using a Thermo Scientific Dionex UltiMate 3000 RSLCnano System and Thermo Scientific Q Exactive HF Hybrid Quadrupole-Orbitrap Mass Spectrometer, and the data were analyzed with Thermo Scientific Proteome Discoverer 1.4 software to assess percent sequence coverage, unique peptides, area, and PTMs for each pathway protein.

We identified the unique tryptic peptides with the best analytical characteristics for each protein target, and selected these quantotypic peptide sequences for the synthesis of stable isotope–labeled AQUA peptides (using HeavyPeptide AQUA Ultimate custom services) that serve as internal standards. Heavy AQUA peptides for each protein target were further evaluated for linearity, reproducibility, accuracy, dynamic range, quantitation limits, and recovery.

Finally, absolute quantitation of each protein target was achieved by utilizing a standard curve with serial dilutions of each heavy AQUA peptide across three orders of magnitude, in conjunction with targeted MS (tMS) using the parallel reaction monitoring (PRM) method implemented on the mass spectrometer. The tMS data were analyzed with Skyline software (MacCoss Lab Software, University of Washington) to determine the limit of quantitation (LOQ) from the calibration curve and the target analyte concentration from unknown samples.

Table 1. Total and phosphorylated AKT/mTOR pathway protein targets selected for analysis.

Enrichment of AKT/mTOR pathway protein targets

Table 2 shows the enrichment of low-abundance AKT/mTOR pathway protein targets by the singleplex IP-MS method. IGF stimulation has been shown to activate AKT/mTOR pathway signaling through phosphorylation cascades [5]. AKT/mTOR pathway protein targets were immunoprecipitated from unstimulated and IGF-stimulated A549 lysates for MS analysis using the IP-MS methodology described above. When compared with neat lysates (combined lysates from the unstimulated and IGF-stimulated cells that were not immunoprecipitated), the IP-enriched samples and LC-MS analysis allowed us to identify a significantly larger number of unique peptides (Table 2).

The effect of IGF stimulation on AKT/mTOR pathway proteins was assessed by comparing the abundance of signaling proteins, interacting partners, and PTMs for each protein target. Protein isoforms and interacting partners were identified for total AKT, IGF1R, and mTOR targets. Relevant phosphorylation sites were detected for phosphorylated AKT1, AKT2, mTOR, IGF1R, and PRAS40 targets. Relative abundance of each pathway protein was determined using the total area of all unique peptides identified for each target. A larger number of unique peptides for phosphorylated AKT and IGF1R were observed for IGF-stimulated A549 lysates compared with unstimulated samples. These results demonstrate that beyond simply identifying protein targets from whole lysates, the singleplex IP-MS method can also be applied to interrogate AKT/mTOR pathway signaling events in the context of IGF stimulation.

Table 2. Enrichment of low-abundance AKT/mTOR pathway proteins by IP-MS.

Validation of mIP-tMS assays

The singleplex IP-MS analysis described above highlighted antibodies that successfully enrich their intended targets, interacting proteins, and PTMs; however, it required that we perform separate IPs for each of the 10 phosphorylated (and 11 total) protein targets from the AKT/mTOR pathway. Alternatively, mIP-tMS analysis should allow simultaneous enrichment and quantitation of all targets from a single IP. Focusing on the AKT/mTOR proteins that we identified earlier, we validated the use of 10-plex phospho and 11-plex total mIP-tMS assays with unstimulated and IGF-stimulated MCF7 lysates. mIP was carried out using biotinylated antibodies and the Thermo Scientific Pierce MS-Compatible Magnetic IP Kit (streptavidin). For the tMS analysis, standard curves were generated using 18 spiked-in AQUA heavy peptides for 12 targets (including both AKT1 and AKT2) and then employed for absolute quantitation of AKT/mTOR proteins. These mIP-tMS assays identified 11 proteins in the multiplex phospho assay and 12 proteins for the multiplex total assay (Table 3). We found that in addition to the IRS1 protein, PI3K subunits were identified in IGF-stimulated MCF7 cell lysates. Upregulation of phosphorylated AKT1, AKT2, and IGF1R was observed upon IGF stimulation.

Table 3. Validation of phospho and total mIP-tMS assays.

Benchmarking mIP-tMS assays

Several singleplex or multiplex immunoassay-based techniques exist for quantifying proteins of interest from biological samples. We sought to compare the mIP-tMS assay with established immunoassays in order to benchmark its analytical performance. We analyzed 11 phosphorylated and 12 total AKT/mTOR pathway proteins in unstimulated and IGF-stimulated HCT116, A549, and MCF7 cells using mIP-tMS, WB, ELISA, and Luminex assays (Figure 2). mIP-tMS assays allowed absolute quantitation for all 12 total and 11 phosphorylated targets in low- to subnanogram concentrations across all cell lines. The comparative summary of the four techniques showed target-dependent correlation, and the WB method disagreed most often with the other three methods (Figure 2). Variability across techniques for some targets can be the result of antibody specificity. Phosphorylated targets showed lower correlation compared to total protein abundances from the AKT/mTOR pathway.

Figure 2. Technology benchmarking of the AKT/mTOR pathway. Comparison of mIP-tMS assays with current immunoassay techniques to quantitate AKT-mTOR pathway protein targets from unstimulated and IGF-stimulated A549, HCT116, and MCF7 lysates were performed. Western blot, ELISA, and Luminex assays were performed according to manufacturer’s instructions. mIP-tMS assays for 11 total and 10 phosphorylated targets were performed in PRM mode using the Thermo Scientific Q Exactive HF Hybrid Quadrupole-Orbitrap Mass Spectrometer. Good correlation was observed for (A) total AKT and (B) phospho IGF1R for mIP-tMS assays, ELISA, and Luminex assays but not western blot. RQ = relative quantitation; AQ = absolute quantitation.

Conclusions

Using the mIP-tMS method developed for the AKT/mTOR pathway, we verified antibody selectivity and assessed interactions and off-targets for AKT/mTOR pathway proteins. Validated antibodies and optimized mIP-tMS workflows combine the benefits of target enrichment, selectivity, and flexibility to better interrogate complex biological interactions. The multiplexing capabilities of mIP-tMS assays provide an effective strategy for increasing sample throughput without sacrificing detection limits.

References

1. Logue JS, Morrison DK (2012) Genes Dev 26:641–650.
2. Gingras AC, Gstaiger M, Raught B et al. (2007) Nat Rev Mol Cell Biol 8:645–654.
3. Carr SA, Abbatiello SE, Ackermann BL et al. (2014) Mol Cell Proteomics
13:907–917.
4. Ackermann BL (2012) Clin Chem 58:1620–1622.
5. Kodama Y, Baxter RC, Martin JL (2002) Am J Respir Cell Mol Biol 27:336–344.