Crotonylation of key metabolic enzymes regulates carbon catabolite repression in Streptomyces roseosporus

Abstract

Due to the plethora natural products made by Streptomyces, the regulation of its metabolism are of great interest, whereas there is a lack of detailed understanding of the role of posttranslational modifications (PTM) beyond traditional transcriptional regulation. Herein with Streptomyces roseosporus as a model, we showed that crotonylation is widespread on key enzymes for various metabolic pathways, and sufficient crotonylation in primary metabolism and timely elimination in secondary metabolism are required for proper Streptomyces metabolism. Particularly, the glucose kinase Glk, a keyplayer of carbon catabolite repression (CCR) regulating bacterial metabolism, is identified reversibly crotonylated by the decrotonylase CobB and the crotonyl-transferase Kct1 to negatively control its activity. Furthermore, crotonylation positively regulates CCR for Streptomyces metabolism through modulation of the ratio of glucose uptake/Glk activity and utilization of carbon sources. Thus, our results revealed a regulatory mechanism that crotonylation globally regulates Streptomyces metabolism at least through positive modulation of CCR.

Fig. 1. Crotonylation in S. roseosporus.

a Immunoblot of S. roseosporus lysate. Twenty μg of total protein were loaded, and the crotonylated proteins were detected with anti-Kcr monoclonal antibody (1:2000) while Coomassie blue staining was used for the loading control. b Immunoblot for crotonylation with anti-Kcr monoclonal antibody (1:2000) in wild-type strain cultured in TSB with additional concentration of sodium crotonate (pH 7.4) or sodium acetate (pH 7.4) as indicated for 24 h. Twenty μg of total protein were loaded, and Coomassie blue staining was used for the loading control. c LC-MS analysis of cellular crotonyl-CoA and acetyl-CoA levels from cells cultured as in b. The data represent mean peak area ±SD of three independent experiments. P value was calculated with Student’s t test (ns means P > 0.05).

Fig. 2. Crotonyl-proteomic analysis of S. roseosporus proteins.

a Procedure diagram for crotonyl-proteomics in S. roseosporus. b Statistics of crotonylated proteins from S. roseosporus. c, d Gene ontology functional classification of the identified crotonylation proteins based on biological processes (c) and molecular functions (d). e Carbon metabolism pathways in Streptomyces. Crotonylated enzymes are highlighted in red.

Fig. 3. Crotonylation regulates S. roseosporus metabolism.

a KEGG pathway-enrichment analysis for lysine-crotonylated proteins. b Pyruvate production assays with addition of NaCr or NaAc. S. roseosporus wild type was cultured in TSB supplemented with NaCr (pH 7.4) or NaAc (pH 7.4) with the indicated concentrations for 24 h, and the intracellular pyruvate concentration was determined. P value was calculated with Student’s t test (n = 4). c Morphological phenotype of S. roseosporus wild type on R5 with or without 10 mM NaCr (pH 7.4) for 2 or 6 days. d Daptomycin production in wild type supplemented with 10 mM NaCr. The daptomycin yield was measured after culturing 120 h. Experiments were performed in triplicate. P value was calculated with Student’s t test (n = 3).

Fig. 4. Identification of crotonylation enzymes.

Immuno-blot assays of crotonylation with anti-Kcr monoclonal antibody (1:2000) from S. roseosporus strain wild type (WT), two decrotonylase mutants ΔcobB (a) and Δhdac (b), and a crotonyl-transferase mutant Δkct1 (c) with anti-Kcr monoclonal antibody in right, together with Coomassie blue staining in left for the loading control.

Fig. 5. Glk is crotonylated at two conserved lysine (K) residues.

a MS/MS analysis of crotonylation on Glk K89. Crotonylation was calculated based on the molecular weight difference values between y3 and y2 (458.26 − 262.14 = 196.12), and b6 and b4 (788.39 − 477.25 = 311.14). b MS/MS analysis of crotonylation on Glk K91, which was calculated based on the molecular weight difference values between y5 and y4 (727.41 − 531.29 = 196.12), and b7 and b6 (916.49 − 720.37 = 196.12). c Glk protein alignment from S. roseosporus (Sro), S. griseus (Sgr), and S. coelicolor (Sco). Two crotonylated residues (Kcr sites) along with conserved motifs are shown. Residues involved in the glucose-binding mechanism were labeled with asterisks.

Fig. 6. Glk is reversibly crotonylated by CobB and Kct1.

a Bacterial two-hybrid analysis of the interaction between Glk and CobB or Kct1. Glk was constructed in pT18, while CobB or Kct1 in pT25. E. coli strain BTH101 containing the hybrids were cultured on LB for 48 h and photographed. b, c In vitro assays for Glk crotonylation by Kct1 (b) and decrotonylation by CobB (c). Reactions were analyzed by western blot with the indicated antibodies (1:2000). The relative intensity of each sample is also shown. In b, the reaction time is 1 h, while in c, the reaction time for decrotonylation is: lane 1, 1.5 h; lane 2, 1 h; lane 3, 1.5 h.

Fig. 7. Crotonylation regulates Glk activity.

a In vitro assays of crotonylation on the Glk catalytic activity. Glk_cr was prepared from the in vitro reaction with Glk and Kct1 in Fig. 3b. Glk mutant isoforms were purified from E. coli. Statistical significance was determined by two-tailed unpaired Student’s t test (n = 4). b, c Western blot analysis of crotonylation level of Glk in vivo. S. roseosporus wild type and ΔcobB and Δkct1 mutants expressing ermEp*-3flag-glk were cultured in the YEME medium, and mycelia were taken at the time points indicated. 3FLAG-Glk was immune-precipitated from the lysate, and western blots were performed with anti-FLAG M2 or anti-Kcr rabbit antibody (1:2000). d Glucose kinase activity assays of the lysate from wild type and ΔcobB and Δkct1 mutants. Bacterial strains were cultured in the YEME medium, and the kinase activity was measured every 12 h. Experiments were performed in triplicate.

Fig. 8. Crotonylation regulates carbon utilization in S. roseosporus.

a Residual glucose in the medium of S. roseosporus wild type and ΔcobB and Δkct1 mutants. Bacterial strains were cultured in the YEME medium, and glucose concentration was measured every 12 h. Experiments were performed in triplicate. b Quantitative assays of fold change of galK gene expression in Δglk mutant complemented with glk. Cells were cultured in MM media with different carbon sources. Gal galactose, glc glucose. Experiments were performed in quadruple. c Quantitative assays of fold change of galK gene expression in wild type and ΔcobB and Δkct1 mutants. Cells were cultured in YEME medium with 4% glucose for 18 or 24 h. Experiments were performed in triplicate. d Quantitative assays of fold change of galK gene expression in Δglk complemented with glk and its mutants (K8991Q, K8991R). Cells were cultured in MM media with different carbon sources for 24 h. Gal galactose, glc glucose. Experiments were performed in triplicate. e Dry weight of wild type and ΔcobB and Δkct1 mutants in the YEME culture without glucose but with galactose as the sole carbon source.