Supplementary Methods Plasmids 1 Bam Sal Bam Sal...Aug 22, 2011  · 2 34. Tothova Z, et al. (2007) FoxOs are critical mediators of hematopoietic stem cell resistance to physiologic

1

Supplementary Methods

Plasmids. pBabe DN-FOXO1-HA-neo was constructed by PCR using a pBabe FOXO1 template (1) and the following primers: GCGCGGATCCATGGCCGAGGCGCCTCAG(forward), GCGCGTCGACTTAAGCGTAGTCTGGGACGTCGTATGGGTATGCAGCTCTTCTCCTAGGAG (reverse). The PCR product was gel purified, digested with BamHIand SalI, and then ligated into pBabe neo, which had been digested with BamHI and SalI and then dephosphorylated. pBabe RFP-DN-FOXO1-HA neo was constructed bydigesting pBabe DN-FOXO1-HA neo with BamHI and SalI and ligating into a similarly digested pBabe N-terminal RFP neo vector. pBabe HA-Runx1 neo was constructed byPCR using a mouse Runx1 template (Open Biosystems) and the following primers: GCGCAGATCTATGTACCCATACGACGTCCCAGACTACGCTGCTTCAGACAGCATTTTTGAGTC (forward), GCGCGAATTCTCAGAAGCATTCACAGTTTCC (reverse). The PCR product was gel purified, digested with BglII and EcoRI, and then ligated into pBabeneo, which had been digested with BamHI and EcoRI and then dephosphorylated. All constructs were verified by sequencing.

Lentiviral RNA interference. Lentiviruses were prepared in 293T cells (ATCC) by triple transfection of the pLKO.1 construct together with pCMV delta R8.2 + pCMV VSV-G(Addgene) or psPAX2 + pMD.2G (Addgene) according to standard procedures. Viral particles were collected at 24 and 48 hr, pooled, and passed through a 0.45 µm filterbefore use. For lentiviral infection of MCF10A-5E cells, 50,000 cells were seeded in a 6-well dish overnight and then infected overnight with 0.5 ml virus + 0.5 ml MCF10Agrowth medium (2) containing polybrene at a final concentration of 8 µg/ml. Infected cells were refed with 2 ml growth medium for ~8 hr and then reinfected overnight. 24 hrafter the second infection, cells were split into 10-cm plates and selected in growth medium containing 2 µg/ml puromycin for 3–5 days.

Retroviral overexpression. For retroviral infection of MCF10A-5E cells, 50,000 cells were seeded in a 6-well dish overnight and then infected overnight with 0.5 ml virus +0.5 ml MCF10A growth medium (2) containing polybrene at a final concentration of 8 µg/ml. Infected cells were refed with 2 ml growth medium for ~8 hr and then reinfectedovernight. Doubly infected cells were refed with 2 ml growth medium for ~8 hr and then reinfected overnight. 24 hr after the third infection, cells were split into 10-cm platesand selected in growth medium containing 300 µg/ml G418 for 4–7 days.

Coverslip immunofluorescence. For coverslip immunofluorescence, MCF10A-5E cells were plated at 25,000 cells/cm2 overnight and then treated with 20 µM LY294002(Calbiochem) or 0.1% DMSO for 5 hr. Coverslips were fixed with 3.7% paraformaldehyde for 15 min, washed 3 × 5 min in PBS, and then blocked for 1 hr at room temperaturewith 1× Western Blocking Reagent (Roche) in PBS + 0.3% Triton X-100. After blocking, coverslips were incubated overnight at room temperature with 1× Western BlockingReagent (Roche) in PBS + 0.3% Triton X-100 containing anti-FOXO1 (1:100, Cell Signaling) or anti-FOXO3 (1:200, Upstate). Coverslips were washed 3 × 5 min in PBS andincubated for 1 hr at room temperature with 1× Western Blocking Reagent (Roche) in PBS + 0.3% Triton X-100 containing Alexa 488-conjugated goat anti-mouse (1:200,Invitrogen) and Alexa 555-conjugated goat anti-rabbit (1:200, Invitrogen). Coverslips were washed 3 × 5 min in PBS and counterstained with 0.5 µg/ml DAPI (Sigma) for 5min at room temperature. After two 5 min washes in PBS, coverslips were mounted with 0.5% n-propyl gallate in PBS + 90% glycerol (3).

Supplementary References

1. Nakamura N, et al. (2000) Forkhead transcription factors are critical effectors of cell death and cell cycle arrest downstream of PTEN. Mol Cell Biol 20(23):8969-8982.2. Debnath J, Muthuswamy SK, & Brugge JS (2003) Morphogenesis and oncogenesis of MCF-10A mammary epithelial acini grown in three-dimensional basement

membrane cultures. Methods 30(3):256-268.3. Giloh H & Sedat JW (1982) Fluorescence microscopy: reduced photobleaching of rhodamine and fluorescein protein conjugates by n-propyl gallate. Science

217(4566):1252-1255.4. Li P, et al. (2003) AKT-independent protection of prostate cancer cells from apoptosis mediated through complex formation between the androgen receptor and FKHR.

Mol Cell Biol 23(1):104-118.5. Christian M, et al. (2002) Cyclic AMP-induced forkhead transcription factor, FKHR, cooperates with CCAAT/enhancer-binding protein beta in differentiating human

endometrial stromal cells. J Biol Chem 277(23):20825-20832.6. Schuur ER, et al. (2001) Ligand-dependent interaction of estrogen receptor-alpha with members of the forkhead transcription factor family. J Biol Chem 276(36):33554-

33560.7. Seoane J, Le HV, Shen L, Anderson SA, & Massague J (2004) Integration of Smad and forkhead pathways in the control of neuroepithelial and glioblastoma cell

proliferation. Cell 117(2):211-223.8. Hirota K, et al. (2003) Hepatocyte nuclear factor-4 is a novel downstream target of insulin via FKHR as a signal-regulated transcriptional inhibitor. J Biol Chem

278(15):13056-13060.9. Kim JJ, et al. (2003) Regulation of insulin-like growth factor binding protein-1 promoter activity by FKHR and HOXA10 in primate endometrial cells. Biol Reprod 68(1):24-

30.10. Foucher I, et al. (2002) Hoxa5 overexpression correlates with IGFBP1 upregulation and postnatal dwarfism: evidence for an interaction between Hoxa5 and Forkhead

box transcription factors. Development 129(17):4065-4074.11. Nemoto S, Fergusson MM, & Finkel T (2004) Nutrient availability regulates SIRT1 through a forkhead-dependent pathway. Science 306(5704):2105-2108.12. Zhao HH, et al. (2001) Forkhead homologue in rhabdomyosarcoma functions as a bifunctional nuclear receptor-interacting protein with both coactivator and corepressor

functions. J Biol Chem 276(30):27907-27912.13. Zawel L, et al. (1998) Human Smad3 and Smad4 are sequence-specific transcription activators. Mol Cell 1(4):611-617.14. Li J, Chang HW, Lai E, Parker EJ, & Vogt PK (1995) The oncogene qin codes for a transcriptional repressor. Cancer Res 55(23):5540-5544.15. el-Deiry WS, Kern SE, Pietenpol JA, Kinzler KW, & Vogelstein B (1992) Definition of a consensus binding site for p53. Nat Genet 1(1):45-49.16. Klein-Hitpass L, Ryffel GU, Heitlinger E, & Cato AC (1988) A 13 bp palindrome is a functional estrogen responsive element and interacts specifically with estrogen

receptor. Nucleic Acids Res 16(2):647-663.17. Seiler-Tuyns A, et al. (1986) Identification of estrogen-responsive DNA sequences by transient expression experiments in a human breast cancer cell line. Nucleic Acids

Res 14(22):8755-8770.18. Roche PJ, Hoare SA, & Parker MG (1992) A consensus DNA-binding site for the androgen receptor. Mol Endocrinol 6(12):2229-2235.19. Osada S, Yamamoto H, Nishihara T, & Imagawa M (1996) DNA binding specificity of the CCAAT/enhancer-binding protein transcription factor family. J Biol Chem

271(7):3891-3896.20. Jiang G & Sladek FM (1997) The DNA binding domain of hepatocyte nuclear factor 4 mediates cooperative, specific binding to DNA and heterodimerization with the

retinoid X receptor alpha. J Biol Chem 272(2):1218-1225.21. Pellerin I, Schnabel C, Catron KM, & Abate C (1994) Hox proteins have different affinities for a consensus DNA site that correlate with the positions of their genes on the

hox cluster. Mol Cell Biol 14(7):4532-4545.22. Mangelsdorf DJ & Evans RM (1995) The RXR heterodimers and orphan receptors. Cell 83(6):841-850.23. Pierrou S, Hellqvist M, Samuelsson L, Enerback S, & Carlsson P (1994) Cloning and characterization of seven human forkhead proteins: binding site specificity and

DNA bending. EMBO J 13(20):5002-5012.24. Meyers S, Downing JR, & Hiebert SW (1993) Identification of AML-1 and the (8;21) translocation protein (AML-1/ETO) as sequence-specific DNA-binding proteins: the

runt homology domain is required for DNA binding and protein-protein interactions. Mol Cell Biol 13(10):6336-6345.25. Sandri M, et al. (2004) Foxo transcription factors induce the atrophy-related ubiquitin ligase atrogin-1 and cause skeletal muscle atrophy. Cell 117(3):399-412.26. Stitt TN, et al. (2004) The IGF-1/PI3K/Akt pathway prevents expression of muscle atrophy-induced ubiquitin ligases by inhibiting FOXO transcription factors. Mol Cell

14(3):395-403.27. Essaghir A, Dif N, Marbehant CY, Coffer PJ, & Demoulin JB (2009) The transcription of FOXO genes is stimulated by FOXO3 and repressed by growth factors. J Biol

Chem 284(16):10334-10342.28. Al-Mubarak B, Soriano FX, & Hardingham GE (2009) Synaptic NMDAR activity suppresses FOXO1 expression via a cis-acting FOXO binding site: FOXO1 is a FOXO

target gene. Channels (Austin) 3(4):233-238.29. de Keizer PL, et al. (2010) Activation of forkhead box O transcription factors by oncogenic BRAF promotes p21cip1-dependent senescence. Cancer Res 70(21):8526-

8536.30. van den Heuvel AP, Schulze A, & Burgering BM (2005) Direct control of caveolin-1 expression by FOXO transcription factors. Biochem J 385(Pt 3):795-802.31. Ramaswamy S, Nakamura N, Sansal I, Bergeron L, & Sellers WR (2002) A novel mechanism of gene regulation and tumor suppression by the transcription factor

FKHR. Cancer Cell 2(1):81-91.32. Bakker WJ, et al. (2004) FoxO3a regulates erythroid differentiation and induces BTG1, an activator of protein arginine methyl transferase 1. J Cell Biol 164(2):175-184.33. Kops GJ, et al. (2002) Forkhead transcription factor FOXO3a protects quiescent cells from oxidative stress. Nature 419(6904):316-321.

2

34. Tothova Z, et al. (2007) FoxOs are critical mediators of hematopoietic stem cell resistance to physiologic oxidative stress. Cell 128(2):325-339.35. Ovcharenko I & Nobrega MA (2005) Identifying synonymous regulatory elements in vertebrate genomes. Nucleic Acids Res 33(Web Server issue):W403-407.36. Janes KA, Wang CC, Holmberg KJ, Cabral K, & Brugge JS (2010) Identifying single-cell molecular programs by stochastic profiling. Nat Methods 7(4):311-317.37. Whittaker SR, Walton MI, Garrett MD, & Workman P (2004) The Cyclin-dependent kinase inhibitor CYC202 (R-roscovitine) inhibits retinoblastoma protein

phosphorylation, causes loss of Cyclin D1, and activates the mitogen-activated protein kinase pathway. Cancer Res 64(1):262-272.38. Bridge AJ, Pebernard S, Ducraux A, Nicoulaz AL, & Iggo R (2003) Induction of an interferon response by RNAi vectors in mammalian cells. Nat Genet 34(3):263-264.39. Li HH, et al. (2007) Atrogin-1 inhibits Akt-dependent cardiac hypertrophy in mice via ubiquitin-dependent coactivation of Forkhead proteins. J Clin Invest 117(11):3211-

3223.40. Schafer ZT, et al. (2009) Antioxidant and oncogene rescue of metabolic defects caused by loss of matrix attachment. Nature 461(7260):109-113.

Table S1. Binding sites for reported FOXO-interacting transcription factors.

AR C/EBPβ ER FOXG1 HNF-4 HOX p53 RAR Smad3/4 FOXO RUNX1Group 1 FBXO32 0 0 0 0 0 1 3 0 0 9 14

FOXO1 0 0 0 0 0 2 0 0 0 10 6FOXO3 0 0 0 0 0 1 1 0 0 9 9CCNI 0 1 0 0 0 1 3 0 0 11 11BCL2L13 0 0 0 0 0 1 1 0 0 11 15CDKN1A 0 0 0 0 0 1 8 0 1 4 8CAV1 0 0 0 0 0 1 3 0 0 20 5

Group 2 SOX4 0 0 0 0 0 1 4 0 0 7 5BTG1 0 0 0 1 0 2 0 0 0 20 8CDKN1C 0 0 0 0 0 0 6 0 0 4 9SESN1 0 0 0 0 0 1 1 0 0 11 5

FOXOs are reported to interact with the above transcription factors according to the following references provided by the Human Protein Reference Database: AR (4), C/EBPβ(5), ER (6), FOXG1 (7), HNF-4 (8), HOX (9, 10), p53 (11), RAR (12), Smad3/4 (7).

Table S2. Consensus binding sites used

Transcriptionfactor

Consensus Reference

Smad3/4 GTCTAGAC (13)FOXG1 TGTAAACAAA (14)p53 RRRCWWGYYY (15)ER GGTCANNNTGACC (16, 17)AR GGWACANNNTGTTCT (18)C/EBPβ RTTGCGYAAY (19)HNF-4 AGGTCANAGGTCA (20)HOX STAATTG (21)RAR AGGTCANAGGTCA (22)FOXO RTAAAYA (23)RUNX1 TGYGGT (24)

The following standard IUPAC abbreviations are used: R =A/G; Y = C/T; W = A/T; S = G/C

Table S3. Literature support for validated and reported FOXO target genes.

Evidence for induction by FOXO References

FBXO32 Induced by FOXOs during muscle atrophy (25, 26)FOXO1/3 Induced by constitutively active FOXO3. Promoter bound by

endogenous FOXO3(27, 28)

CDKN1A Induced by FOXOs upon TGFβ stimulation or oxidative stress (7, 29)CAV1 Induced by constitutively active FOXO3 (30)SOX4 Induced by constitutively active FOXO1 (31)BTG1 Induced by constitutively active FOXOs and during erythroid

differentiation(31, 32)

SOD2 Induced by FOXOs during oxidative stress (33, 34)SEMA3C Induced by constitutively active FOXO1 (31)

Target genes were considered validated if their control by FOXOs was published by at least two independentgroups. Target genes were considered reported if their control by FOXOs appeared in one publication.

Table S4. Gene ontology (GO) enrichment among genes with FOXO–RUNX1 binding modules

GO biological process Total number ofgenes

Expected numberof genes

Observed number of genes among thosewith FOXO–RUNX1 module

Foldenrichment

p value

development 1827 63 97 1.54 7.10e-06cell surface receptor linked signaltransduction

1188 41 69 1.68 1.14e-05

system development 465 16 35 2.18 1.31e-05nervous system development 460 16 34 2.14 2.53e-05organ development 567 20 39 1.99 3.31e-05cell communication 2983 103 139 1.35 4.92e-05neurophysiological process 643 22 41 1.85 1.13e-04signal transduction 2700 93 126 1.35 1.23e-04anatomical structure development 1306 45 70 1.55 1.28e-04neuron differentiation 85 3 11 3.75 1.58e-04

Genes with FOXO and RUNX1 consensus binding sites within 35 bp of one another were identified with SynoR (35) and tested for enrichment of specific GO biologicalprocesses. GO biological processes are ordered based upon the relative significance of their enrichment.

Fig. S1. MCF10A-5E cells express FOXO1 and FOXO3 but not FOXO4. RNAwas purified from MCF10A-5E cells or HT-29 colon adenocarcinoma cells andanalyzed for FOXO1, FOXO3, and FOXO4 expression by PCR. Blank (B) andno reverse transcription (RT) samples were included as negative controls, andGAPDH was included as a representative high-abundance gene. RNA from HT-29 cells was used to verify the FOXO4 primers.

Fig. S2. Quantitatively accurate qPCR measurement of BCL2L13 by small-sample cDNA amplification. Samples containing 100 microdissected cells wereserially diluted, amplified by small-sample poly(A) PCR (36), and quantified byqPCR. Data are shown as the median ± range of three replicate small-sampleamplifications. Amplification efficiencies (E) based on a log-linear fit of the3–100-cell dilutions (red line) are listed along with primer efficiencies (Ep)calculated by serially diluting the template before qPCR. Equivalent validation ofthe other genes examined in this study was reported in (36).

3

Fig. S3. Systematic pairwise correlation analysisof single-cell gene expression by RNA FISH. (A-S) The loading-normalized fluorescenceintensities of each gene pair were correlated forsingle cells of three independent acini to quantifythe extent of coregulation as described in Fig. 2A–C. Single-cell correlations (Rsingle cell) havebeen ordered from highest to lowest and aresummarized in Fig. 2D. (T) A log-normaldistribution is an appropriate statistical model forsingle-cell gene expression as measured byRNA FISH. Log-normal probability plot of thenormalized and median-centered single-cellfluorescence intensities of UBC (a housekeepinggene not used as a loading control, orange)compared to a log-normal distribution (dashed)as described in Fig. 2 E and F. Single-cell UBCfluctuations are indistinguishable from a log-normal distribution.

Fig. S4. Kinetics of the FOXO gene panel doesnot require 3D culture explicitly. (A) qPCRmeasurements for the FOXO gene panel for theindicated days of 3D morphogenesis. Data aretaken from Fig. 1B for comparison. (B) qPCRmeasurements for the FOXO gene panel for theindicated days of 2D culture in morphogenesismedium. MCF10A-5E cells were seeded at12,500 cells/cm2 in morphogenesis assaymedium and refed every fourth day (2).Validated or reported FOXO target genes areshown in red or orange respectively. Loadingcontrols for qPCR normalization are shown ingray, and genes not previously implicated asFOXO targets are shown in black. Data areshown as the mean ± s.e.m. of (A) triplicate or(B) quadruplicate biological samples.

Fig. S5. RUNX1 phosphorylation on S276

requires CDK activity but not ERK activity.MCF10A-5E acini were treated for 24 hr with theCDK inhibitor roscovitine or the MEK1/2 inhibitorU0126 and fixed at d10. Frozen sections werestained for (A) Rb phosphorylation on T821 (p-Rb), a site that is responsive to CDK inhibition(37); (B) phospho-ERK1/2 (p-ERK1/2); or (C)Runx1 phosphorylation on S276 (p-Runx1). Scalebar is 20 µm.

4

Fig. S6. Validation of an shRNA hairpin for RUNX1. (A) Knockdown of RUNX1 measured by qPCR. MCF10A-5E cells were infected with pLKO.1 vector control or shRUNX1and analyzed for RUNX1 levels. (B) Expression of the shRUNX1 hairpin does not induce an interferon response as indicated by the interferon-response gene OAS1 (38). (C)shRUNX1 causes a uniform decrease in RUNX1 at the single-cell level. Monolayers of MCF10A-5E cells infected with the indicated lentiviruses were stained for RUNX1 andimaged by widefield immunofluorescence. The exposure time and scaling of images is matched to allow comparison of fluorescence intensities. Scale bar is 25 µm.

Fig. S7. Single-cell coregulation between FOXO groups is created upon RUNX1knockdown. (A and B) Representative genes from Group 1 (BTG1) and Group 2(CAV1) were analyzed by RNA FISH as described in Fig. 2 A–C and in the Experi-mental Procedures. Single-cell correlation (Rsingle cell) is shown with 90% confidenceintervals in parenthesis. Scale bar is 25 µm.

Fig. S8. Validation of a dominant-negative FOXO1. (A) FOXOs are localized to thenucleus after prolonged LY294002 treatment. MCF10A-5E monolayers were treatedwith 20 µM LY294002 for 5 hr, stained with the indicated antibodies, and imaged bywidefield immunofluorescence. Scale bar is 25 µm. (B) MCF10A-5E cells wereinfected with an HA-tagged DN-FOXO or vector control and analyzed for HA byimmunoblotting. β-tubulin was used as a loading control. (C) Most LY-inducedFOXO target genes are inhibited by DN-FOXO1. MCF10A-5E monolayers weretreated with 20 µM LY294002 for 5 hr, and population-level mRNA measurementswere performed by qPCR. Data are shown as the mean ± s.e.m. of triplicatebiological samples. FBXO32 induction is not repressed in DN-FOXO1 cells, possiblybecause of positive feedback between FBXO32 and FOXOs (39).

Fig. S9. Comparison of DCFDA staining patterns in 3D acini of the MCF10A-5Eclone and parental MCF10A cells. On the indicated day of 3D culture, acini werelabeled with DCFDA (green) as described in the Methods and counterstained withDRAQ-5 (blue) to label nuclei. Note that parental MCF10A cells do not showDCFDA staining in matrix-attached cells at d7, consistent with (40). However, bothcell types show non-uniform DCFDA staining in matrix-attached cells at d10 whenstochastic profiling was performed (36). Scale bar is 20 µm.

Fig. S10. RUNX1 loss is not strongly associated with FOXO1 upregulation in breast-cancer specimens that are not triple negative. (A) RUNX1 and FOXO1 mRNAexpression is weakly anticorrelated in non-triple-negative breast cancers. Correlatedprobesets from the GEO #GSE6861 dataset were standardized as z-scores andaveraged to give the estimated relative levels of RUNX1 and FOXO1. The 10th and90th percentiles of RUNX1 expression (dashed boxes) were further analyzed in B.(B) Tumors with the lowest RUNX1 mRNA expression do not have significantlyupregulated FOXO1 mRNA expression. Box-and-whisker plots from the samples inthe dashed boxes in A are shown with notches indicating 90% nonparametricconfidence intervals. Significance for the increase in FOXO1 expression wasdetermined by the Wilcoxon rank-sum test.