Fig 1.
(a) Homer annotation analysis of BMAL1 ChIP-seq peak location shows frequency of distribution across the genome in undifferentiated (UNDIFF) and lactogen differentiated (DIFF) HC11 cultures. 3UTR, 3’ untranslated region; TTS, terminal transcription site; pseudo, pseudogene; 5UTR, 5’ untranslated region. (b) Venn diagram illustrating the overlap and number of unique Ensembl gene IDs of protein coding genes with transcriptional start site nearest to BMAL1 ChIP-seq peaks in the two undifferentiated (UNDIFF1 and UNDIFF2) and two differentiated (DIFF1 and DIFF2) samples.
Table 1.
Representative categories enriched with protein coding genes nearest to BMAL1 ChIP-seq overlapping peaks.
Fig 2.
Schematic showing cellular location of BMAL1 targets downstream of CREB1 and estradiol generated using Ingenuity Pathway Analysis tools, with the addition of insulin and Fzd1 genes.
Gene names and symbols are given here and can be found with more detail in S16 Table. ATP-binding cassette subfamily A (Abca1, Abca8a, Abcg2); Activin A receptor (Acvr2a), ADAM metallopeptidase with thrombospondin type 1 (Adamts5, 18) and ADAM like decysin (Adamdec1), angiotensin II receptor (Agtr1b), A kinase anchoring protein (Akap9,13), angiopoietin (Angpt1), inorganic phosphate transporter (Ankh), amyloid beta precursor protein (APP), bone morphogenetic protein receptor type 1B (Bmpr1b), bisphosphate nucleotidase (Bpnt2) retinoic acid inducible (Brinp2), chromosome 2 open reading frame 88 (C2orf88), serum/glucocorticoid regulated kinase (Sgk3), calcium dependent secretion activator 2 (Cadps2), cellular communication network factor 3 (Ccn3), cadherins (Cdh8, 10,11, 13), cyclin dependent kinase (Cdk6, 14), CF transmembrane conductance regulator (Cftr), cholinergic receptor muscarinic 2 (Chrm2), cannabinoid receptor 1 (Cnr1), collagens (Col1a2, 3a1, 5a2, 11a1), complement C3d receptor 2 (Cr2), cAMP responsive element binding protein 1 (Creb1), cysteine rich transmembrane BMP regulator 1 (Crim1) crystallin zeta (Cryz), cystatin (Cst5), catenin alpha 2 (Ctnna2), chemokine receptor (Cxcr4), cytochrome P450 family (Cyp26b1, 51a1), DAB adaptor protein 2 (Dab2), dickkopf WNT signaling pathway inhibitor 2 (Dkk2), discs large MAGUK scaffold protein 2 (Dlg2), dynein light chain roadblock-type 2 (Dynlrb2), EGF like repeats and discoidin domains 3 (Edil), ephrin B2 (Efnb2), ectonucleotide pyrophosphatase (Ennpp2), erb-b2 receptor tyrosine kinase 4 (Erbb4), ELKS/RAB6-interacting/CAST family member 2 (Erc2), estrogen related receptor gamma (Esrrg), Fas associated via death domain (Fadd), fibroblast growth factor (Fgf9), four and a half LIM domains 2 (Fhl2), FERM domain containing 4B (Frmd4b), fucosyltransferase 8 (Fut8), polypeptide N-acetylgalactosaminyltransferase 7 (Galnt7), galanin receptor (Galr1), growth associated protein 43 (Gap43), glial cell derived neurotrophic factor (Gdnf), growth hormone receptor (Ghr), glutaminase (Gls), glutamate-ammonia ligase (Glul), G protein subunit gamma 12 (Gng12), glycerol-3-phosphate acyltransferase (Gpam), G protein-coupled receptor 63 (Gpr63), GPRIN family member 3 (Gprin3), growth factor receptor bound protein 10 (Grb10), glutamate ionotropic receptors (Gria1, Gria3, Grin2a), hyaluronan synthase 2 (Has2), histone deacetylase 5 (Hdac5), hepatocyte growth factor (Hgf), 15-hydroxyprostaglandin dehydrogenase (Hpgd), interleukin 15 (Il15), interleukin receptor (Il7r), inositol polyphosphate-4-phosphatase type II B (Inpp4b), insulin (Ins1), integrin subunit beta 1 (Itgb1), jagged canonical Notch ligand (Jag1), janus kinase 1 (Jak1), potassium voltage-gated channel (Kncab1, d2, d3, q5), kinase insert domain receptor (Kdr), Kruppel like factor 4 (Klf4), leptin (Lep), luteinizing hormone/choriogonadotropin receptor (Lhcgr), LIM domain 7 (Lmo7), lipoprotein lipase (Lpl), latent transforming growth factor beta binding protein 1 (Ltbp1), mab-21 like 1 (Mab21l1), mitotic arrest deficient 2 like 1 (Mad2l1), MAF bZIP transcription factor (Maf), membrane associated guanylate kinase (Magi1,2), mal, T cell differentiation protein 2 (Mal2), mitogen-activated protein kinase (Map2k6,k14), MyoD family inhibitor domain containing (Mdfic), MAM domain containing glycosylphosphatidylinositol anchor 1 (Mdga1), mesoderm specific transcript (Mest), myosin (Myo1b), myoferlin (Myof), NDUFA4 mitochondrial complex associated (Ndufa4), Nectin cell adhesion molecule (Nectin3), neural growth regulator (Negri), neuronal differentiation 6 (Neurod6), sodium/potassium transporting ATPase interacting (Nkain2), natriuretic peptide receptor 3 (Npr3), neuronal pentraxin 2 (Nptx2) neuropeptide Y (Npy) and receptor (Npr1r), nuclear receptors (Nr3c2-mineralcorticoid receptor; Nr5a2- liver receptor 1, an essential transcriptional regulator of lipid metabolism), neuropilin 2 (Nrp2), nudix hydrolase 7 (Nudt7), neurexophilin 1 (Nxph1), olfactomedin 4 (Ofm4), opioid receptor kappa 1 (Oprk1), PBX homeobox 1 (Pbx1), phosphodiesterase 1C (Pde1c), phospholipase A2 (Pla2g4a), plexin (Plxna2, a4), peripheral myelin protein 22 (Pmp22), PPARG coactivator 1 alpha (Ppargc1a), protein phosphatase 1 regulatory subunit 3B (Ppp1r3b), protein phosphatase 1 catalytic subunit alpha (Ppp3ca), prickle planar cell polarity protein 2 (Prickle2), prolactin (Prl8a2), prolactin receptor (Prlr), polypyrimidine tract binding protein 2 (Prbp2), prostaglandin E receptor (Ptger2,3,4), protein tyrosine kinase 2 (Ptk2), pentraxin (Ptx3), RALY RNA binding protein like (Ralyl), Rap2b, RB binding protein 8 (Rbbp8), retinol dehydrogenase 10 (Rdh10), regulator of G protein signaling (Rgs7, 18), roundabout guidance receptor 1 (Robo1), RP1 axonemal microtubule associated (Rp1), R-spondin 2 (Rspo2), receptor transporter protein 4 (Rtp4), RUNX1 partner transcriptional co-repressor 1 (Runx1t1), ryanodine receptor 2 (Ryr2), sphingosine-1-phosphate receptor 1 (S1pr1), Samhd1, SATB homeobox 1 (Satb1), selenoprotein P (Selnop), semaphorin 3A (Sema3a), shootin 1 (Shtn1), solute carrier (SLC1a3, 4a4, 6a11, 8a1), slit guidance ligand 2 (Slit2), sphingomyelin phosphodiesterase 1 (Smpd1), sushi, nidogen and EGF like domains 1 (Sned1), sortilin related VPS10 domain containing receptor 3 (Sorcs3), SPHK1 interactor, AKAP domain containing (Sphkap), SPARC-osteonectin (Spock3), sprouty related EVH1 domain, Steap1, containing 2 (Spred), serine/threonine kinase 3 (Stk3), tachykinin precursor 1 (Tac1), threonyl-tRNA synthetase 1 (Tars1), Tcb1d9, TBL1X receptor 1 (Tbl1xr1), tissue factor pathway inhibitor 2 (Tfpi2), thyroglobulin (Tg), thyroid hormone beta (Thrb), toll like receptor 4 (Tlr4), transmembrane (Tm4sf1, Tmeff2, Tmem74), TNF (Tnfsf11) and TNF receptor (Tnfrsf11b), tensin 3 (Tns3), translocation associated membrane protein 1 (Tram1), thyroid releasing hormone receptor (Trhr), tripartite motif containing 24 (Trim24), transcriptional repressor GATA binding 1 (Trps1), UDP glycosyltransferase 8 (Ugt8), vascular endothelial growth factor C (Vegfc), Wnt2, Wnt2b, Xkr4, zinc finger (Zbtb18, Zfhx3, Zfpm2, Zmat3).
Table 2.
Representative categories enriched with potential BMAL1 target genes identified using ChIP-seq that overlapped with genes that showed circadian rhythms of expression in mammary glands and BMAL1 targets identified in hepatic tissue.
Fig 3.
(a)Western blot analysis of BMAL1 protein abundance in HC11 and monoclonal colonies (1A, 1B, 1C, 2A, 2B, 2C, 2D) established from CRISPR-CAS transfected with guide RNA (gRNA) targeting BMAL1 gene post monoclonal selection. BMAL1 is absent in colonies 1C and 2D and decreased in the remaining colonies compared to WT HC11. Monoclonal 1C culture was used in all subsequent experiments, and referred to as BMAL1-KO. Data are representative of two western blots. (b) Temporal analysis of Per2 expression in WT HC11 (solid line) and BMAL1-KO (dashed line) cultures. For this experiment cells were grown to confluence in growth media. Media was changed to lactogen media for 2 hr to synchronize clocks, as we previously have shown this synchronizes clocks in HC11 cells [5]. At completion of 2 hr lactogen treatment (time 0 hr), cells were rinsed with PBS and cultured in growth media for remainder of the experiment. Cells were collected for isolation of total RNA every 4 hr over a 48 hr period beginning at 0 hr. Per2 was measured with RT-qPCR, and levels were expressed relative to mean ΔCT of HC11 across all time points. Cosinor analysis found mesor (-0.12 and -0.68), amplitude (1.58 and 0.70), acrophase (0.78 and -0.56), R2 (0.85 and 0.47) and p-value (8.05 E-5 and 0.04) of fit to a 24 hr rhythm, respectively, for HC11 and BMAL1-KO lines. Data represent n = 3 wells/line and 2 experimental replicates. (c) BMAL1 protein abundance in UNDIFF and DIFF cultures measured using ELISA. Data are expressed as mean μg of BMAL1/mg protein ± standard deviation of three samples per treatment; * indicates difference at p<0.05. Data represent n = 3 protein isolates/line/state of differentiation.
Fig 4.
Effect of BMAL1 CRISPR-CAS knockout (BMAL1-KO) on growth and metabolic activity in culture.
(a) Eight day growth curve analysis was performed by plating 100,000 cells/well on day 0 in 6 well dish of wild type HC11(black line) and BMAL1-KO (gray line), two wells/treatment were collected and counted every two days; values are mean ± standard deviation across five experiments. Two way ANOVA found that line and day significantly affected (p<0.05) number of cells. Data represent five experimental replicates, with n = 2 wells/line per experiment. (b) MTT assay was performed by plating HC11 (black bars) and BMAL1-KO (gray bars) cells at 10,000 cells/well in a 96-well plate; on days 2, 4, 6 and 8 of culture, MTT assay was performed. ANOVA found line and day significantly affected NADH levels; values are mean ± standard deviation; * indicates difference between lines at p<0.05 across 3 replicate experiments. Data represent three replicate experiments collected from n = 3 wells/line/day. (c) The intensity of MTT staining per cell across three images on each day in HC11 (black) and BMAL1-KO (gray) cultures was quantified. Values are mean intensity per cell ± standard deviation. A significant difference at p<0.05 is indicated by *. (d) Images of cells were captured following staining with the MTT assay on day 2 and 6 of culture. (e) Cells were collected from 100 mm dishes (plating density was 100,000 cells/ml) for fluorescence activated cell sorting (FACS) to determine percent of cells in G1/G0 and S/G2/M phases following labeling with propidium iodide across 8 days of culture. Values are mean across 5 experiments, with ANOVA analysis finding that day had an effect (p<0.05) on proportion of cells in phases, but there was no difference between HC11 (black) and BMAL1-KO (gray). (f) FACS analysis for dead or dying cells (cells or events with <2N) in HC11 (black) and BMAL1-KO (gray). Values are percent of total events ± standard error; * indicates difference at p<0.05. (g) RT-qPCR analysis of Ccnd1 and Sod3 expression in undifferentiated (UNDIFF) cultures of HC11 (black) and BMAL1-KO (gray) cells. Values are mean across triplicate samples and two experimental replicates, normalized to express fold-change relative to mean of HC11 ± standard deviation using delta-delta cycle threshold method; Student t-test analysis * indicates difference between lines at p<0.05. (h) Reactive oxygen species (ROS) assay of HC11 and BMAL1 cells on day 3 and 4 of culture. Two-way ANOVA found day and line affected (p<0.05) ROS levels; * indicates difference between lines at p<0.05 across 3 replicate experiments n = 3 wells/line/day; values are mean of arbitrary units (AU) ± standard deviation.
Fig 5.
RT-qPCR analysis of (a) Tph1 and (b) Slc6a4 (Sert) in undifferentiated (UNDIFF) and lactogen differentiated (DIFF) HC11 (black) and BMAL1-KO (gray) cultures. Values are mean across triplicate samples and two experimental replicates. To calculate relative difference, data were normalized to mean ΔCT of HC11 and fold change difference was determined using the delta-delta CT method; ANOVA and post-hoc Tukey test analysis indicated with differing letter reflecting difference at p<0.05. (c) ChIP-qPCR analysis of Slc6a4 (Sert) promoter region using primers that targeted two sites that contained E-box sequences beginning at -42 and -1282 nucleotide bases upstream of the transcriptional start site in undifferentiated HC11 cultures; values are mean across four samples, normalized to express fold-change relative to mean CT of mock. A 2-fold difference was considered a positive ChIP.
Fig 6.
RT-qPCR analysis of (a) Ppara and (b) Csn2 in undifferentiated (UNDIFF) and lactogen differentiated (DIFF) HC11 and BMAL1-KO cultures. Values are mean across triplicate samples and two replicate experiments, normalized to express fold-change relative to mean of HC11 ± standard deviation using delta-delta cycle threshold method. Note the y-axis in Ppara is fold change, whereas y-axis for Csn2 is log base 2 of fold change due to large induction. ANOVA and post-hoc Tukey test analysis findings is indicated by differing letter reflecting difference at p<0.05. (c) ELISA quantification of CSN2 protein in UNDIFF and DIFF HC11 (black) and BMAL1-KO (gray) cultures. Values are mean concentration ± standard deviation across triplicate samples. ANOVA and post-hoc Tukey test analysis findings is indicated by differing letter reflecting difference at p<0.05. (d) RT-qPCR analysis of Prlr in UNDIFF and lactogen DIFF HC11 (black) and BMAL1-KO (gray) cultures. Values are mean across triplicate samples, normalized to express fold-change relative to mean of HC11 ± standard deviation using delta-delta cycle threshold method. Note the y-axis is fold change. ANOVA and post-hoc Tukey test analysis findings is indicated by differing letter reflecting difference at p<0.05. (e) Images of two and a half dimensional drip gel cultures of HC11 and BMAL-KO cells taken with phase-contrast microscopy after 7 days of incubation in lactogen media. Cells were plated at 13,000 cells/well.