Quantitative Non-canonical Amino Acid Tagging (QuaNCAT) Proteomics Identifies Distinct Patterns of Protein Synthesis Rapidly Induced by Hypertrophic Agents in Cardiomyocytes, Revealing New Aspects of Metabolic Remodeling

Abstract

Cardiomyocytes undergo growth and remodeling in response to specific pathological or physiological conditions. In the former, myocardial growth is a risk factor for cardiac failure and faster protein synthesis is a major factor driving cardiomyocyte growth. Our goal was to quantify the rapid effects of different pro-hypertrophic stimuli on the synthesis of specific proteins in ARVC and to determine whether such effects are caused by alterations on mRNA abundance or the translation of specific mRNAs. Cardiomyocytes have very low rates of protein synthesis, posing a challenging problem in terms of studying changes in the synthesis of specific proteins, which also applies to other nondividing primary cells. To study the rates of accumulation of specific proteins in these cells, we developed an optimized version of the Quantitative Noncanonical Amino acid Tagging LC/MS proteomic method to label and selectively enrich newly synthesized proteins in these primary cells while eliminating the suppressive effects of pre-existing and highly abundant nonisotope-tagged polypeptides. Our data revealed that a classical pathologic (phenylephrine; PE) and the recently identified insulin stimulus that also contributes to the development of pathological cardiac hypertrophy (insulin), both increased the synthesis of proteins involved in, e.g. glycolysis, the Krebs cycle and beta-oxidation, and sarcomeric components. However, insulin increased synthesis of many metabolic enzymes to a greater extent than PE. Using a novel validation method, we confirmed that synthesis of selected candidates is indeed up-regulated by PE and insulin. Synthesis of all proteins studied was up-regulated by signaling through mammalian target of rapamycin complex 1 without changes in their mRNA levels, showing the key importance of translational control in the rapid effects of hypertrophic stimuli. Expression of PKM2 was up-regulated in rat hearts following TAC. This isoform possesses specific regulatory properties, so this finding indicates it may be involved in metabolic remodeling and also serve as a novel candidate biomarker. Levels of translation factor eEF1 also increased during TAC, likely contributing to faster cell mass accumulation. Interestingly those two candidates were not up-regulated in pregnancy or exercise induced CH, indicating PKM2 and eEF1 were pathological CH specific markers. We anticipate that the methodologies described here will be valuable for other researchers studying protein synthesis in primary cells.

Fig. 1.

Scheme of method combining pSILAC with AHA labeling. See Experimental Procedures for further information.

Fig. 2.

Optimization of pSILAC conditions. A, ARVC and HEK293 cells were maintained as described in methods; 30 min prior to adding [³⁵S]methionine. DMEM was replaced by M199 medium and some cells were treated with cycloheximide (35.54 μmol/L) where indicated. 18 h later, cells were lysed and analyzed for [³⁵S]methionine incorporation as described in the Methods. B, Twelve different tissues from a single rat were pulverised in liquid nitrogen, then lysed with RIPA buffer. Protein concentrations were measured by the BCA method and 20 μg of protein from each were subject to Western blots. C, HeLa cells and ARVC were labeled with AHA for 18 h. Cells were then lysed, and subjected to the 'click' reaction, the alkyne carrier being alkyne-agarose resin. Corresponding amounts of input lysate and supernatant after click reaction were clicked to alkyne-biotin and visualized by Streptavidin, Alexa Fluor® 680. D, ARVC were pre-incubated in M199 medium with different concentrations of methionine (25% = 25.13 μmol/L, 50% = 50.26 μmol/L) for 30 min, AHA or normal methionine were then added into the medium immediately after treatment, After different time periods as indicated, cells were lysed, the same amount of total proteins from each group were subjected to click reaction. The alkyne carrier here is biotin-alkyne. After Click reaction, 40 μl of reaction product were analyzed by SDS-PAGE followed by Streptavidin, Alexa Fluor® 680 visualization. The bar graph shows quantification for newly synthesized protein minus negative control (lane 3). E, ARVC were incubated in M199 medium with low concentration of methionine (25.13 μmol/L) for 30 min. Some were then stimulated by insulin or PE as indicated and [³⁵S]methionine was added to the medium immediately. 48 h later, cells were lysed and 20 μg of total protein were subjected to SDS-PAGE and Coomassie blue staining, whereas newly synthesized proteins in the dried gel were visualized by autoradiography. The bar graph shows quantification for newly synthesized protein:total protein from three independent experiments. Significance was determined by Student's t test.

Fig. 3.

Hierarchical clustering analysis in heat map format of analyzed proteins. A, Insulin and PE increase the synthesis of a range of proteins via the mTORC1 pathway, as their synthesis is inhibited by rapamycin. The molecular function gene ontology of these proteins were manually curated with ExPASy bioinformatics resource portal (http://www.expasy.org) and mapped to glycolysis, energy and metabolism, fatty acids beta-oxidation, amino acid metabolism, ATP synthesis, translation factors, cytoskeleton/sarcomere proteins, heat shock proteins or other/unclassified proteins. B, The ten proteins whose synthesis is increased most strongly by insulin in comparison to PE. The data are plotted as (change in synthesis caused by insulin- change caused by PE)/change caused by PE as a percentage.

Fig. 4.

A, Direct Protein Interaction Network (PIN) in silico analysis of all differentially expressed proteins after insulin and PE activation (B) PIN in silico analysis of all differentially expressed proteins after insulin + rapamycin and PE + rapamycin activation. PIN analysis was performed with MetaCore and curated with DAVID gene ontologies of subcellular localization. The analyzed proteins are color-coded as red and blue, indicating their up- or down-regulation, respectively. The red, blue and gray lines denote a positive, negative or unspecified regulation, respectively.

Fig. 5.

Validation of pSILAC results in ARVC. A, ARVC were starved in M199 medium containing low methionine (25.13 μmol/L) for 30 min. Cells were then stimulated with insulin or PE as indicated. [³⁵S]methionine was added into the medium immediately after treatment and, 48 h later, cells were lysed and HSP60 was IP'd from 250 μg total lysate. IPs were subjected to SDS-PAGE and Coomassie brilliant blue staining, whereas newly synthesized HSP60 in the dried gel was visualized by autoradiography. B, ARVC were treated and lysed as in panel A. Endogenous ACO2 was then IP'd from 250 μg of cell lysate. IP's were divided into two equal portions; one was analyzed by Western blot to verify the IP efficiency, whereas the other was used for verification of pSILAC result by Click reaction as described in Methods. C, ARVC were isolated and maintained as described in Methods. Cells were transferred to low-methionine medium (25.13 μmol/L) 30 min prior to treatment. ARVC were treated with PE or insulin where indicated. Normal methionine or AHA was added into the medium immediately after treatment; 48 h later, cells were lysed and the same amounts of total protein were subjected to Western blots. D, ARVC were isolated, maintained and treated as in panel C, 48 h later, cells were lysed, total RNA was extracted and subjected to RT-qPCR for the Aco2, Hspd1, Mb, Desmin, and Acadl mRNAs. E, Summary depiction of changes in protein synthesis rates and mRNA levels of selected candidate genes. F, 5′-UTRs of the Desmin, Aldoa, and Jup mRNAs.

Fig. 6.

Expression of PKM2 and eEF1G is up-regulated in TAC animal model. A, B, TAC operation was as described in EXPERIMENTAL PROCEDURES. At two- and six-week time points, six experimental rats and six age-paired controls were sacrificed; left ventricles were pulverised in liquid nitrogen, then lysed with RIPA buffer, and protein concentration was measured by the BCA method. 20 μg of protein from each heart was subjected to Western blots. C, Quantification of PKM2 and eEF1G from panels A, B. D, E, Control and TAC rats were as described in panel A (six in each group). At two- and six-week time points, six experimental rats and six age-paired controls were sacrificed. Total RNA was extracted from the left ventricles by TRIzol and analyzed by RT-qPCR.

Fig. 7.

Chronic activation of mTORC1 in MEFs increases the expression of JUP, ADLOA, ACO2, and MDH2 proteins. Wild-type and TSC2^−/− MEFs were cultured in complete DMEM. A, After lysis, 20 μg of total protein were subjected to Western blot using the indicated antibodies. B, After lysis, total RNA was extracted and subjected to RT-qPCR analysis for Mdh2, Jup, Aco2, Aldoa, Hspd1, and Rps6. C, Quantification and summary table of panels A and B.

Fig. 8.

Metabolic remodelling model reflected in the distinct patterns of protein synthesis induced by the PE and Insulin in cardiomyocytes, (see Discussion for details).