Middle-Down Mass Spectrometry Reveals Activity-Modifying Phosphorylation Barcode in a Class C G Protein-Coupled Receptor

Abstract

G protein-coupled receptors (GPCRs) are the largest family of membrane receptors in humans. They mediate nearly all aspects of human physiology and thus are of high therapeutic interest. GPCR signaling is regulated in space and time by receptor phosphorylation. It is believed that different phosphorylation states are possible for a single receptor, and each encodes for unique signaling outcomes. Methods to determine the phosphorylation status of GPCRs are critical for understanding receptor physiology and signaling properties of GPCR ligands and therapeutics. However, common proteomic techniques have provided limited quantitative information regarding total receptor phosphorylation stoichiometry, relative abundances of isomeric modification states, and temporal dynamics of these parameters. Here, we report a novel middle-down proteomic strategy and parallel reaction monitoring (PRM) to quantify the phosphorylation states of the C-terminal tail of metabotropic glutamate receptor 2 (mGluR2). By this approach, we found that mGluR2 is subject to both basal and agonist-induced phosphorylation at up to four simultaneous sites with varying probability. Using a PRM tandem mass spectrometry methodology, we localized the positions and quantified the relative abundance of phosphorylations following treatment with an agonist. Our analysis showed that phosphorylation within specific regions of the C-terminal tail of mGluR2 is sensitive to receptor activation, and subsequent site-directed mutagenesis of these sites identified key regions which tune receptor sensitivity. This study demonstrates that middle-down purification followed by label-free quantification is a powerful, quantitative, and accessible tool for characterizing phosphorylation states of GPCRs and other challenging proteins.

Figure 1

Development of a novel middle-down proteomic assay. (A) Overview of middle-down MS assay. (B) Overview of isolation, fragmentation, and PRM of phosphorylated precursors for relative quantification of isomers, an example is given for a diphosphorylated tail. (C) Sequence of human mGluR2 (accession #Q14416) cleavage product and potential phosphorylation sites (red) used for PRM studies. Residue numbering in (C) is based on the endogenous sequence reported in accession #Q14416-01.

Figure 2

mGluR2 is subject to co-occurring phosphorylations. (A) Intact mass spectra of the mGluR2 C-terminus (4+ charge state). (B) Relative summed intensity across the LC-timescale (%) of intact phosphorylated C-terminal tails before l-glutamic acid treatment. Unmodified mGluR2 is denoted in gray, monophosphorylated mGluR2 in light blue, diphosphorylated mGluR2 in dark blue, triphosphorylated mGluR2 in green, and tetraphosphorylated mGluR2 in gold.

Figure 3

mGluR2 is subject to basal isomeric phosphorylation states. (A) Sequence of mGluR2 cleavage product and potential phosphorylation sites (red) used for PRM studies, b-type and y-type fragment ions utilized for PRM studies are denoted as black flags. Fragment ion numbering is based on the number of residues in the cleavage product (P2 = PA830 in accession #Q14416-1). Phosphorylation content was localized between regions A, B, C, and D for the (B) mono-, (C) di-, (D) and triphosphorylated populations. The intact precursor corresponding to each set of plots is shown pictorially on the left. Color fill denotes the number of phosphorylations localized within a given region using fragment ions from the PRM assay. No phosphorylation is denoted with gray, monophosphorylation is denoted with light blue, diphosphorylation is denoted with dark blue, and triphosphorylation is denoted with green. For the higher stoichiometries, regions A and B were combined based on fragment ion coverage. *For the triphosphorylated population, y₁₉ to y₁₆ were used to calculate occupancy in region D + T857. Error bars represent the standard error of the mean (N = 3 biological replicates and 3 technical replicates).

Figure 4

mGluR2 activation enhances intracellular phosphorylation abundance. (A) Normalized intensity of intact phosphorylated C-terminal tails over 1 mM l-glutamic acid treatment (0–60 min). The intensity for a given phosphorylated stoichiometry is normalized to the intensity of the unmodified mGluR2 tail in the corresponding LC–MS run. Values for each phosphoform are denoted by color. Values for the monophosphorylated population are shown in light blue, the diphosphorylated population is shown in dark blue, the triphosphorylated population is shown in green, and the tetraphosphorylated population is shown in gold. (B) Each phosphorylated stoichiometry is provided as an individual inset. Error bars represent the standard error of the mean (N = 3 biological replicates and 3 technical replicates). Brackets indicate a statistical comparison between two data points using a Wilcoxon–Mann–Whitney test. Significant differences are denoted as follows: * = p < 0.05, ** = p ≤ 0.01, *** = p ≤ 0.001, **** = p ≤ 0.0001.

Figure 5

mGluR2 activation dynamically shifts positional phosphorylation abundance over time. (A) The difference in log 2 (fragment ion intensity) for unmodified and monophosphorylated fragment ions derived from the monophosphorylated precursor population shows statistically significant changes in phosphorylation position over 1 mM l-glutamic acid treatment (0 to 60 min). The population of averaged fragment ions is listed above each box plot. These ratios of fragment ions can be converted into relative percentages showing phosphorylation content within a defined region. Relative phosphorylation content was calculated for the (B) mono- and (C) diphosphorylated precursor populations. Color fill denotes the number of phosphorylations localized within a given region using fragment ions from the PRM assay. No phosphorylation is denoted with gray, monophosphorylation is denoted with light blue, and diphosphorylation is denoted with dark blue. Error bars represent the standard error of the mean (N = 3 biological replicates and 3 technical replicates). Brackets indicate a statistical comparison between two data points using a Wilcoxon–Mann–Whitney test. Significant differences are denoted as follows: * = p < 0.05, ** = p ≤ 0.01, *** = p ≤ 0.001, **** = p ≤ 0.0001.

Figure 6

Intracellular phosphorylation states fine-tune mGluR2 signaling. (A) Dose–response curves for wild-type mGluR2 (blue); T832E, S833E, and S837E mGluR2 (orange); T832A, S833A, and S837A mGluR2 (purple); S867E, T868E, T869E, S870E, and S871E mGluR2 (pink); and S867A, T868A, T869A, S870A, and S871A mGluR2 (green). Quantification of (B) half-maximal effective concentration (EC₅₀) and (C) maximal efficacy (E_MAX) for l-glutamic acid. (D) Coulombic electrostatic potential map for CryoEM structure 7MTS; inset shows the intracellular pocket of mGluR2 complexed with G_i. Error bars represent the standard error of the mean (N = 3 biological replicates and 2 technical replicates). Asterisks indicate a statistical comparison using a one-way ANOVA with multiple comparisons to WT. Significant differences are denoted as follows: *** = p ≤ 0.001, **** = p ≤ 0.0001.