Fig 2: The plp1-CreERT transgene preferentially drives expression of DTR in tSCs at the NMJ. Young adult NMJs expressing plp1-CreERT and DTR transgenes were stained with BTX to label AChRs (A) and for S100 to label tSCs (asterisks) and myelinating SCs (B). DTR stain is robust in the soma of each tSC (asterisks), but weak in their processes (C). In myelinating SCs, DTR stain seems to be concentrated in paranodal microvilli (thin arrows, C) and in small spots in each nucleus (thick arrows, C). (D) Composite image of A–C. Fiji and Microsoft Powerpoint were used to generate this figure. Scale bar 20 µm.

Fig 3: BHLHE40-knockdown reduced hypoxia-induced expression of a panel of cytokines and growth factors. a Heatmaps of cytokines and growth factors whose hypoxia-induced expression (1% O2 at 6 h or 48 h, fold-change = 1.5 in two independent experiments) was diminished by BHLHE40-knockdown (KD) in LM cells. The gene expression levels were determined using the Illumina Human HT-12 expression BeadChips. Normalized (quantile normalization) hybridization signals were log2 transformed and standardized by genes across experiment conditions to generate the heatmap. b Heatmaps of a subset of genes list in a whose expression was affected by BHLHE40-KD in LM cells exposed to hypoxia combined with low (1 mM) glucose (1%O2/LG, 4 h). The gene expression levels were determined using the Affymetrix Human Gene 1.0 ST array. c Heatmaps of hypoxia-induced genes whose expression was not significantly affected by BHLHE40-KD in LM cells as determined by the Illumina Human HT-12 expression BeadChips. d Expression of luciferase reporters driven by hypoxia-responsive elements of ITGA6 or LDHA was not affected by BHLHE40 knockout (KO) by CRISPR/Cas9 editing in MDA-MB-231 cells, in the absence or presence of exogenous HIF1A. Luciferase activities were normalized to co-transfected CMV-ß-galactosidase and presented as mean ± SD (n = 6). e Expression of genes in control LM empty vector (EV) and LM BHLHE40-KD cells exposed to 1%O2/LG (4 h). mRNA expression levels were determined by qPCR, normalized to RPL13A, and presented as mean ± SD (n = 6). *p < 0.05 (n = 6, 1%O2/LG vs. untreated control), **p < 0.05 (n = 6, KD vs. EV), one-way ANOVA followed by Tukey’s post-hoc tests. f mRNA and protein expression levels of HBEGF and CTGF in primary xenograft tumors, determined by qPCR and immunoblotting, respectively. *p < 0.05 (n = 6, KD vs. EV), Student’s t test. Representative immunoblotting images of three tumors of KD or EV cells are presented

Fig 4: Exosomic secretion of HBEGF was reduced by BHLHE40 depletion in MDA-MB231 or TR cells. a BHLHE40-knockout (KO) by CRISPR/Cas9 editing in MDA-MB-231 cells reduced the total number of exosomes and the amount of HBEGF protein in exosomes in comparison with empty vector (EV) control cells. b BHLHE40-knockdown (KD) by shRNA in tamoxifen-resistant (TR) cells reduced the total number of exosomes and the amount of HBEGF protein in exosomes in comparison with TR EV cells. Exosomes were purified from conditioned medium of 5 × 106 cells cultured in medium supplemented with exosome-free serum, either under normal culture condition or exposed to 1% O2/low glucose (LG) for 6 h. Exosomes were purified using the ExoQuick-TC solution and quantified under a fluorescent microscope after being labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE) using the Exo-Glow labeling kit, which is designed to exclude background particles. Exosome number in the bar graph is presented as mean number of exosomes per field ± SD (total of nine fields from three independent experiments were examined). The presence of HBEGF and exosome markers in purified exosomes (3 µg protein/lane) or whole cell extracts (WCE; 30 µg protein/lane) were detected by immunoblotting. **p < 0.05 (n = 9, 1%O2/LG vs. control), *p < 0.05 (n = 9, KO or KD vs. EV), one-way ANOVA followed by Tukey’s post-hoc tests

Fig 5: BHLHE40 activates gene expression by sequestering histone deacetylase (HDAC)1 and HDAC2 from genome DNA binding in MDA-MB-231 cells exposed to hypoxia and low glucose (1%O2/LG, 4 h). a BHLHE40-knockout (KO) diminished dissociation of HDAC1 and HDAC2 from the promoter region of HBEGF in MDA-MB-231 cells exposed to 1%O2/LG (4 h), as determined by chromatin immunoprecipitation (ChIP) followed by qPCR of the HBEGF promoter region (-529 to -372 from the transcription start site). HBEGF promoter binding activity of HDAC1 or HDAC2 was calculated as: (DNA amount in anti-HADC IP complex – DNA amount in control IgG IP complex)/DNA amount in 1% input. *p < 0.05 (n = 6, 1%O2/LG vs. control), **p < 0.05 (n = 6, KO vs. EV). b 1%O2/LG treatment increased interactions between BHLHE40 and HDAC1/2 in the soluble cellular fraction of MDA-MB-231 empty vector (EV) cells. Protein-protein interaction was detected by reciprocal co-immunoprecipitation (IP)/immunoblotting (IB) analysis. c HDAC inhibition induced expression of BHLHE40 target genes. Cells were exposed to hypoxia (1% O2) or HDAC inhibitors (BRD6688 10 µM or TSA 2 µM) for 24 h. mRNA expression levels were determined by qPCR, normalized to RPL13A, and presented as mean ± SD (n = 6). *p < 0.05 (n = 6, treated vs. untreated control cells), one-way ANOVA followed by Tukey’s post-hoc tests