Fig 2: NOTCH1 is target of miR-200s and miR-34s. a Putative binding sites of miR-34b-5p, miR-200b and miR-200c in NOTCH1 3′UTRs (TargetScan). Asterisks indicate nucleotides substituted in miR-34s and miR-200s predicted target sites to perform the luciferase assays. b NOTCH1 3′UTRs regulate luciferase activity dependent on miR-200b, miR-200c, miR-34b* and miR-34c-5p in SH-SY5Y cells (WT wild type, MUT mutant, P P value). Renilla luciferase activity was normalized on the firefly luciferase activity of the pSICHECK2 vector. c Western blot analysis of NOTCH1 (BD, 552466 and Santa Cruz, sc-6014R), β-actin (ACTB) and vinculin (VCL) after microRNAs transfection in SH-SY5Y cells; the full-length (FL, ~300 kDa) and NOTCH1 transmembrane fragment (NTM, ~120 kDa) are indicated. Cells were collected at 48 h from miRNA transfection. Normalization with densitometric analysis is shown. d Real-time quantitative reverse transcription PCR analysis for the NOTCH1 transcriptional target genes HES1, HES5 and NOTCH1 (*P < 0.05; **P < 0.005; ***P < 0.0005, ND not detectable)

Fig 3: Restoration of MTP function after re-expression. Plasmids pcDNA3 or pcDNA3-hMTP-FLAG were transfected into three different MKO cell lines. After 48 h, (A) media were used to measure apoB100 secretion using ELISA. B: Cells were lysed in buffer K. Cell lysates (30 μg) were resolved on 8% SDS-PAGE and blotted on to nitrocellulose membranes. MTP protein was detected using polyclonal anti-hMTP antibodies (top). Beta-actin was used as loading control (middle). The apoB100 in media were detected by resolving 30 μl of the media on 6% SDS-PAGE gel, transferring to membrane, and probing with anti-ApoB monoclonal antibodies (1D1) (bottom). C: MTP’s TAG transfer activity was measured using 50 μg of crude cell lysate. D: The FLAG-tagged MTP protein was purified using anti-FLAG M2 affinity gel from transfected cells. The purified MTP protein was resolved on 8% SDS-PAGE gel and transferred to nitrocellulose membrane. The MTP protein (top) and its partner PDI (bottom) were detected using polyclonal anti-hMTP antibodies (Abcam) and anti-PDI monoclonal antibodies (Invitrogen), respectively. apoB100, apolipoprotein B100; PDI, protein disulfide isomerase.

Fig 4: The 3-Hydroxybutyrate suppresses ROS and apoptosis in endothelial cells. (A–D,F):, human umbilical vein endothelial cells (HUVECs) with addition of phosphate-buffered saline (Con) or norepinephrine (NE) (A; n = 6, 14; scale bar = 50 μm); NE or NE + SC79(AKT activator) (B; n = 4, 4; scale bar = 50 μm); NE or NE + Arg (C; n = 4, 4; scale bar = 50 μm); NE or NE + NO donor (D; n = 4, 4; scale bar = 50 μm), or NE or NE + 3-Hydroxybutyrate (3-HB) (F; n = 4,4; scale bar = 50 μm). (E) Dihydroethidium (DHE) staining of HUVECs administrated with NE or NE + 3-Hydroxybutyrate (3-HB) (scale bar = 10 μm), and the quantified data analyzed as relative DHE ratio (DHE area was divided with nuclear number/view, and subjected to further quantification studies) (n = 4, 4, 4). (G) Western blot detecting signals from pAKT, AKT or beta Actin (Actin) in HUVECs administrated with insulin together with NE or NE + 3-HB. Right panel is the quantified data showing relative pAKT signal (n = 6, 6, 6) (signal was analyzed as pAKT/AKT/Actin). (H) Scheme summarizing our findings. Empagliflozin (EMPA) activates the AKT/eNOS/NO pathway and suppresses EC apoptosis through an increase of 3-Hydroxybutyrate and reduction of reactive oxygen species (ROS). This change leads to enhancement of capillarization in the left ventricle (LV) during pressure overload and ameliorates cardiac dysfunction. Data were analyzed by a 2-tailed Student’s t test (A–D,F), or by 2-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison test (E,G). *P < 0.05, **P < 0.01. Values are shown as the mean ± SEM. NS not significant. Small circle indicates outlier, triangle indicates abnormal value. Analyses were performed with and without these values; when the results of both analyses were nonsignificant, the difference was described as NS.

Fig 5: VPS34 activity is elevated upon PIKfyve inhibition in Inpp4b-deficient cells.A, Inpp4b+/+ or Inpp4b−/− MEFs treated with vehicle or apilimod 10 nM for 48 h, followed by immunoblotting against Inpp4b, VPS34, or Beta Actin. B, Inpp4b+/+ or Inpp4b−/− MEFs treated with vehicle or apilimod 10 nM for 48 h, followed by VPS34 immunoprecipitation and kinase assay to monitor VPS34 activity. C, Inpp4b+/+ MEF transiently expressing LAMP1-mCherry and treated with vehicle or 500 nM VPS34-IN1 for 48 h followed by PtdIns(3)P immunostain. D, quantification from (C) of total cell PtdIns(3)P fluorescence signal or (E) PtdIns(3)P fluorescence signal overlayed on LAMP1-mCherry positive regions within a cell. The scale bar represents 20 μm, zoomed inset: 5 μm. F, Inpp4b+/+ or Inpp4b−/− MEF transiently expressing LAMP1-mCherry and treated with vehicle or apilimod 10 nM or VPS34-IN1 500 nM at various combinations for 48 h. G, quantification of LAMP1-positive vacuoles greater than 1.5 μm in diameter per cell and (H) mean vacuole diameter (μm). The scale bar represents 20 μm. Data represent ± SD from three independent experiments with 25 to 30 cells assessed per treatment condition per experiment. Statistical significance was measured by ANOVA and multiple Student’s t test and represented as ∗ (p < 0.05). INPP4B, inositol polyphosphate 4-phosphatase type II; MEF, mouse embryonic fibroblast; PtdIns, phosphoinositide; PtdIns(3)P, phosphatidylinositol-3-monophosphate; PIKfyve, Phosphoinositide Kinase, FYVE-Type Zinc Finger Containing.