Amino acid metabolites that regulate G protein signaling during osmotic stress

Abstract

All cells respond to osmotic stress by implementing molecular signaling events to protect the organism. Failure to properly adapt can lead to pathologies such as hypertension and ischemia-reperfusion injury. Mitogen-activated protein kinases (MAPKs) are activated in response to osmotic stress, as well as by signals acting through G protein-coupled receptors (GPCRs). For proper adaptation, the action of these kinases must be coordinated. To identify second messengers of stress adaptation, we conducted a mass spectrometry-based global metabolomics profiling analysis, quantifying nearly 300 metabolites in the yeast S. cerevisiae. We show that three branched-chain amino acid (BCAA) metabolites increase in response to osmotic stress and require the MAPK Hog1. Ectopic addition of these BCAA derivatives promotes phosphorylation of the G protein α subunit and dampens G protein-dependent transcription, similar to that seen in response to osmotic stress. Conversely, genetic ablation of Hog1 activity or the BCAA-regulatory enzymes leads to diminished phosphorylation of Gα and increased transcription. Taken together, our results define a new class of candidate second messengers that mediate cross talk between osmotic stress and GPCR signaling pathways.

Fig 1. Phosphorylation of Gpa1 in response to osmotic stress occurs in a Hog1-dependent, pH-independent manner.

Western blot analysis reveals that Gpa1 and Snf1 are phosphorylated (p-Gpa1 and p-Snf1) in response to (A) glucose (Glc) limitation (“High” = 2% glucose, “Low” = 0.05% glucose), (B) heat shock (42°C), or (C) osmotic stress (0.5 M KCl). Note that Hog1 is phosphorylated (p-Hog1) in response to heat shock or osmotic stress, but not glucose limitation. Intracellular pH (insets) decreases in response to glucose limitation or heat shock (shaded area), but not osmotic stress. Hog1 catalytic activity (hog1^K52R) is required for phosphorylation of Gpa1 but not Snf1. Diploid, control cells lacking Gpa1. reg1Δ, control cells lacking Gpa1 phosphatase. Hog1, Snf1, and Load correspond to gels probed with Hog1, poly-His, and G6PDH antibodies, respectively. Data were quantified based on band intensity, and are presented as mean ± standard deviation, N = 3.

Fig 2. Global metabolomics analysis identifies candidate second messengers of osmotic stress.

(A) Metabolites from wild-type and hog1Δ cells, untreated or treated for 20 minutes with 0.5 M KCl, were extracted and then analyzed by GC-MS and LC-MS/MS. (B) 296 unique metabolites were identified. Venn diagram of metabolites that increase >2-fold in response to osmotic stress (n = 28), in cells that express Hog1 (n = 13) or both (n = 3). (C) Heat map of metabolites that increase in salt-treated wild-type compared to unstressed wild-type cells (left column, top), and increase in salt-treated wildtype, but not salt-treated hog1Δ cells (right column, bottom). Colored arrows indicate 2-hydroxy carboxylic acid derivatives of the BCAAs valine (HIV, green), leucine (HIC, red), and isoleucine (HMVA, blue). (D, Top) Relative abundance of the three BCAA derivatives and (D, Bottom) their parent amino acids. Data presented as mean ± standard deviation, N = 5.

Fig 3. BCAA derivatives are necessary for a full response to osmotic stress.

(A) BCAAs are converted to 2-keto acids by the BCAA transaminases, Bat1 and Bat2. The 2-keto acids are subsequently reduced to the 2-hydroxy acids, and ultimately exported by the fusel acid transporter, Pdr12. (B, C) Genetic ablation of BAT1 or BAT2, or (B, E) loss of MAPK phosphorylation consensus sites (bat1^5A bat2^3A) leads to reduced Gpa1 phosphorylation. (B, D) Genetic ablation of PDR12 does not affect Gpa1. Data presented as mean ± standard deviation, * indicates p ≤ 0.05, N = 3.

Fig 4. BCAA derivatives promote Gpa1 phosphorylation.

(A) Ectopic addition of 2-hydroxyisocaproate (HIC) promotes phosphorylation of Gpa1 but not Hog1 or Snf1. (B) Ectopic addition of the BCAA derivatives promotes Gpa1 phosphorylation while intracellular pH is unaffected (inset). (C, D) Ectopic addition of HIC promotes Gpa1 phosphorylation in wild-type and Hog1-deficient cells. Osmotic stress promotes Gpa1 phosphorylation only in wild-type cells. Data presented as mean ± standard deviation, N = 3.

Fig 5. BCAA derivatives do not bind directly to Gα proteins.

¹H-¹⁵N 2D HSQC NMR spectra of Gαi-GDP alone (black) or in the presence of 25-fold excess (A) HIV, (B) HIC, or (C) HMVA (color) reveal no discernable peak shifts. (D) Spectral overlay of Gαi-GDP at pH 6.0 (magenta) and pH 7.0 (black) is presented as a positive control. Inset, magnified view of a subset of resonances showing pH-dependent spectral changes.

Fig 6. The AMPK kinase Elm1 phosphorylates Gpa1 upon BCAA derivative addition.

Gpa1 phosphorylation after ectopic addition of 30 mM HIC is abrogated in cells lacking the AMPK kinase ELM1, or all three AMPK kinases (ΔΔΔ). Data presented as mean ± standard deviation, N = 3.

Fig 7. BCAA derivatives diminish MAPK-dependent gene transcription.

(A) Addition of BCAA derivatives or KCl dampens α-factor pheromone-induced gene transcription (P_FUS1-GFP). The dampening capacity of each BCAA derivative is correlated with Gpa1 phosphorylation (see Fig 4). (B) Genetic ablation of the AMPK kinases increases basal gene transcription, consistent with reduced Gpa1 phosphorylation [21]. Correspondingly, the kinase mutants abrogate any ability of the metabolites to suppress basal signaling and limit their ability to suppress pheromone signaling (41% reduction in wildtype vs. 26% reduction in the kinase mutant strain). Data are presented as mean ± standard deviation, N = 4.