Fig 2: Fbxo7 knockdown in T-ALL cells reduces the proportion of inactive monomer/dimer PFKP and promotes glycolysis. (A) Lysates from cells expressing a control or Fbxo7 shRNA, and with Fbxo7 overexpression, were analyzed by immunoblot and quantified (CCRF-CEM n = 4, HEK293T n = 3). Data are presented as mean ± SD. (B) Image of representative immunoblot of PFKP half-life in CCRF-CEM cells expressing a control or Fbxo7 shRNA and treated with CHX for up to 24 h (n = 2). (C) Cells were treated with DMSO or 10 µM MG132 for 4 h and lysed. Proteins were resolved by SDS-PAGE to immunoblot for PFKP, and levels were quantified (CCRF-CEM n = 2, HEK293T n = 3). Data are presented as mean ± SD. (D) Lysates from CCRF-CEM cells with control or Fbxo7 shRNA were passed over a 200-kD molecular weight cutoff filter to separate PFKP monomers (86 kD) and dimers (172 kD) from tetramers (344 kD). AMP and citrate were added to lysates as controls to respectively promote or dissociate PFKP tetramers. Total lysates and <200 kD fractions were resolved by SDS-PAGE and analyzed by immunoblot (representative n = 4). (E) Quantification of PFKP in <200 kD fraction as in D (n = 4). Data are presented as mean ± SD. (F) Glucose metabolism in CCRF-CEM (top) and Jurkat E6 (bottom) cells expressing a control or Fbxo7 shRNA was analyzed by the Agilent Seahorse Glycolytic Rate Assay (n = 4). Basal glycolysis refers to the physiological rate, whereas compensatory glycolysis reflects the glycolytic capacity when mitochondrial respiration is inhibited. (G) Chronos gene dependency scores for FBXO7 in cancer cell lines of various lineages, plotted using data publicly available on the DepMap portal (https://DepMap.org). A Chronos score of 0 indicates that a gene is nonessential, while -1 is comparable to the median of all pan-essential genes. Blood & lymphocyte n = 111, peripheral nervous system [PNS] n = 32, soft tissue n = 44, uterus n = 34. ***, P < 0.005; ****, P < 0.001; *****, P < 0.0005. Source data are available for this figure: SourceData F3.

Fig 3: Fbxo7 promotes the Cdk6-independent ubiquitination and Cdk6-dependent phosphorylation of PFKP. (A) Fbxo7 immunoprecipitation from CCRF-CEM cells treated with 1 µM palbociclib or 0.1 µM of either of two different Cdk6 PROTACs for 24 h (Rana et al., 2019; Su et al., 2019). PFKP co-immunoprecipitation (IP) was analyzed by immunoblot (n = 2). (B) HEK293T cells with a control or Fbxo7 shRNA were transfected and treated with 1 µM palbociclib for 24 h as indicated. Cell lysates were resolved by SDS-PAGE and analysed by immunoblot (n = 2). (C) Cdk6 immunoprecipitation from CCRF-CEM cells expressing control or Fbxo7 shRNA. PFKP co-immunoprecipitation was analyzed by immunoblot (n = 1). (D) Cdk6 kinase activity in CCRF-CEM cells expressing control or Fbxo7 shRNA and treated with 1 µM palbociclib for 24 h as indicated. Kinase activity is measured in a dose-dependent manner by fluorescence of the Cdk6 biosensor (n = 1). (E) Diagram of model whereby Fbxo7 promotes both the ubiquitination (Ub) and phosphorylation (P) of PFKP. Source data are available for this figure: SourceData F2.

Fig 4: Cdk6 immunoprecipitation (IP) from MOLT-4 cells expressing control or Fbxo7 shRNA. PFKP co-immunoprecipitation was analyzed by immunoblot (n = 1). Source data are available for this figure: SourceData FS2.

Fig 5: Effect of SAE1/UBA2-mediated SUMOylation inhibition on pyruvate kinase M2 activity in RA FLSs.RA FLSs were transfected with siRNA-1 and siRNA-3 for SAE1 (siSAE1-1, siSAE1-3) or UBA2 (siUBA2-1, siUBA2-3) or control siRNA (siC) for 48 hours or pretreated with GA (150 µM) for 24 hours and then stimulated with or without TNF-a (10 ng/mL) for 24 hours. Protein expression was measured by Western blot. (A) Effect of SAE1/UBA2 inhibition on PK activity. Enzyme activity was detected by the indicated kits in the Methods section. Data are expressed as the mean ± SD from 4 independent experiments. (B) Effect of SAE1 or UBA2 inhibition on protein expression of HK2, PFKP, and PKM2. (C and D) Effect of SAE1 or UBA2 knockdown (C) or GA treatment (D) on TNF-a–induced PKM2 phosphorylation. (E) Effect of SAE1 or UBA2 inhibition on TNF-a–induced nuclear translocation of PKM2. For cellular IF staining, PKM2 (green) and nuclei (blue) were evaluated using fluorescence microscopy, and representative images from 3 independent experiments are shown. Original magnification, ×400. (F) Effect of SUMO-1 knockdown on TNF-a–induced PKM2 phosphorylation. (G and H) Effect of SAE1 (G) or UBA2 (H) knockdown on TNF-a–induced phosphorylation of PKM2 in SUMO-1–overexpressing RA FLSs. (I and J) Effect of SAE1 or UBA2 knockdown on PKM2 and SUMO-1 interaction. The immunoprecipitates were probed with anti-PKM2 (upper panels) or anti–SUMO-1 (lower panels) antibody. Immunoprecipitations were also performed with IgG as a negative control. Input controls are shown as indicated (left panel). (K) Effect of the K336R mutation on PKM2 SUMO-1 modification. MH7A, a human RA FLS cell line, was transfected with or without a PKM2 plasmid carrying a mutation of K336 to arginine (K336R). Input controls are shown as indicated (left panel). (L) Effect of the K336R mutation on TNF-a–induced PKM2 phosphorylation. Data (C, D, F, G, H, and L) are expressed as the mean ± SD of densitometry quantification of Western blot results from at least 3 independent experiments. *P < 0.05 versus siC or DMSO or VECTOR+siC or WT; #P < 0.05 versus TNF-a + siC or TNF-a + DMSO or TNF-a + OE SUMO-1 + siC or TNF-a + WT; &P < 0.05 versus OE SUMO-1+ siC, by Student’s t test (for comparisons between DMSO and GA in panel B) or 1-way ANOVA (for other groups in panel B and other panels).