Effects of MEK inhibitors GSK1120212 and PD0325901 in vivo using 10-plex quantitative proteomics and phosphoproteomics

Abstract

Multiplexed isobaric tag based quantitative proteomics and phosphoproteomics strategies can comprehensively analyze drug treatments effects on biological systems. Given the role of mitogen-activated protein/extracellular signal-regulated kinase (MEK) signaling in cancer and mitogen-activated protein kinase (MAPK)-dependent diseases, we sought to determine if this pathway could be inhibited safely by examining the downstream molecular consequences. We used a series of tandem mass tag 10-plex experiments to analyze the effect of two MEK inhibitors (GSK1120212 and PD0325901) on three tissues (kidney, liver, and pancreas) from nine mice. We quantified ∼ 6000 proteins in each tissue, but significant protein-level alterations were minimal with inhibitor treatment. Of particular interest was kidney tissue, as edema is an adverse effect of these inhibitors. From kidney tissue, we enriched phosphopeptides using titanium dioxide (TiO2 ) and quantified 10 562 phosphorylation events. Further analysis by phosphotyrosine peptide immunoprecipitation quantified an additional 592 phosphorylation events. Phosphorylation motif analysis revealed that the inhibitors decreased phosphorylation levels of proline-x-serine-proline (PxSP) and serine-proline (SP) sites, consistent with extracellular-signal-regulated kinase (ERK) inhibition. The MEK inhibitors had the greatest decrease on the phosphorylation of two proteins, Barttin and Slc12a3, which have roles in ion transport and fluid balance. Further studies will provide insight into the effect of these MEK inhibitors with respect to edema and other adverse events in mouse models and human patients.

Keywords: Barttin; Cell biology; GSK1120212; Multiplexing; PD0325901; Phosphoproteomics.

Figure 1. Experimental overview using TMT10-plex quantitation

A) The 3x3+1 strategy consists of triplicate analyses of vehicle-, GSK1120212-, and PD0325901-treated mice plus a cross-over channel comprising a mix of all samples allowing for the comparison across three separate TMT experiments and thus between tissue types and inhibitor treatments. B) Experimental details. Mouse tissues were harvested and proteolyzed. For protein expression profiling, peptides were labeled with TMT reagents and then combined as shown in Panel A. Labeled peptides were separated by basic pH reverse-phase (BpRP) chromatography with fraction collection. Fractions (12) were analyzed by an LC-MS³ method using a 3-hr gradient. For phosphorylation analysis, two different experiments were collected. First, phosphopeptides were first enriched using TiO₂ and then labeled with TMT reagents. Second, phosphotyrosine-containing peptides were enriched via anti-phosphotyrosine antibodies. The unbound phosphopeptide fraction was analyzed via BpRP chromatography (12 fractions) for general phosphorylation. The affinity isolated phosphotyrosine sample was analyzed by a single 3-hr LC-MS³ analysis.

Figure 2. Summary of protein expression data

A) Summary of total peptides, unique peptides and proteins quantified in each tissue at <1% FDR. B) Distributions of log₂-transformed fold change ratios for three tissues comparing the drug treatment to control. V, vehicle; G, GSK1120212: PD, PD0325901.

Figure 3. Summary of TiO₂ phosphorylation enrichment data

A) Phosphopeptides identified by TiO₂ (titanium dioxide enrichment) and pY IP (phosphotyrosine immunoprecipitation). B) Heat map of clustered phosphorylation changes in kidney tissue. K-means clustering of significantly changing sites from kidney tissue of mice treated with vehicle, GSK1120212, and PD0325901. Gene ontology (GO) enrichment terms for each of the clusters are shown alongside the clusters, as are bar charts showing effects of inhibitor treatments for representative sites. Motifs that are enriched for the inhibitor treatments are illustrated on the left. V, vehicle; G, GSK1120212: PD, PD0325901; SN, signal-to-noise; *, p-value<0.05 versus vehicle as determined via ANOVA and post-hoc Tukey’s test.

Figure 4. Protein expression and phosphorylation analysis for BSND protein

A) Relative abundance of site S298 showing decreased phosphorylation upon either drug treatment. B) Site S162 phosphorylation shows no alteration in phosphorylation levels. C) Normalized TMT signal to noise ratio (S/N) for BSND protein expression does not change. *, p-value<0.05 versus vehicle as determined via ANOVA and post-hoc Tukey’s test.

Figure 5. Enrichment of kidney phosphotyrosine (pY)-containing peptides via immunoprecipitation

Left: Heat map of significant phosphotyrosine-containing peptide changes following enrichment. Right: Histograms of TMT signal-to-noise (SN) intensity for example phosphotyrosine sites. V, vehicle; G, GSK1120212; PD, PD0325901; *, p-value<0.05 versus vehicle control as determined via ANOVA and post-hoc Tukey’s test.