1 RUNX1, a factor mutated in ,

2 controls the fate of ER-positive mammary luminal cells

3

1,2 1,2,3,* 1,2,* 1,2,4 4 Maaike P.A. van Bragt , Xin Hu , Ying Xie , and Zhe Li

5

6

7

8

1 9 Division of Genetics, Brigham and Women's Hospital, Boston, Massachusetts 02115,

10 USA

2 11 Department of Medicine, Harvard Medical School, Boston, Massachusetts 02115, USA

3 12 School of Life Sciences, Jilin University, Changchun 130012, China

13

14

4 15 Corresponding author (E-mail: [email protected])

16 *Equal contributions

17

18 Competing interests statement: The authors declare no competing interests

19

20

21

1 22 Abstract

23 RUNX1 encodes a RUNX family (TF) and was recently identified as a

24 novel mutated in human luminal breast cancers. We found that Runx1 is expressed

25 in all subpopulations of murine mammary epithelial cells (MECs) except the secretory

26 alveolar luminal cells. Conditional knockout of Runx1 in MECs by MMTV-Cre led to a

27 decrease in luminal MECs, largely due to a profound reduction in the estrogen

28 (ER)-positive mature luminal subpopulation, a phenotype that could be rescued by loss of

29 either Trp53 or Rb1. Mechanistically RUNX1 represses Elf5, a master regulatory TF gene

30 for alveolar cells, and regulates mature luminal TF/co-factor (e.g., Foxa1 and

31 Cited1) involved in the ER program. Collectively, our data identified a key regulator of the

+ + 32 ER luminal lineage whose disruption may contribute to development of ER luminal

33 breast cancer when under the background of either TP53 or RB1 loss.

34

2 35 Introduction

36 RUNX1, RUNX2 and RUNX3, and their common non-DNA-binding partner CBFβ,

37 form a small family of heterodimeric transcription factors (TFs) referred to as Core-

38 Binding Factors (CBFs) (Speck and Gilliland, 2002). They are best known as master

39 regulators of cell fate determination in blood, bone and neuron, respectively (Chuang et

40 al., 2013). RUNX1 is a master regulator of hematopoietic stem cells and multiple mature

41 blood lineages. Translocations and involving both RUNX1 and CBFB are

42 frequently found in human (Speck and Gilliland, 2002). Recently, key roles of

43 this family of TFs in epithelial cells and solid tumors also started to emerge (Chuang et

44 al., 2013; Scheitz and Tumbar, 2013; Taniuchi et al., 2012). In particular, in breast

45 cancer, recent whole-genome and whole-exome sequencing studies have consistently

46 identified point mutations and deletions of RUNX1 in human luminal breast cancers

47 (Banerji et al., 2012; Ellis et al., 2012; Network, 2012). In addition, mutations in CBFB

48 were also identified in luminal breast cancers from these studies. Its gene product CBFβ

49 is critical for enhancing DNA binding by RUNX TFs through allosteric regulation (Bravo et

50 al., 2001; Tahirov et al., 2001). Thus, we hypothesized that RUNX1, together with CBFβ,

51 might play a key role in mammary epithelial cell (MEC) lineage determination as a master

52 regulatory TF, and that loss of this normal function might contribute to breast cancer

53 development.

54 There are two major epithelial cell lineages in the mammary gland (MG), luminal

55 lineage (including ductal and alveolar luminal cells) and basal lineage (the mature cell

56 type in the basal lineage is myoepithelial cell) (Figure 1A). These two types of MECs are

57 produced by multipotent mammary stem cells (MaSCs, which are basal cells) during

58 embryonic development or upon MEC transplantation to cleared mammary fat pads

59 (Shackleton et al., 2006; Spike et al., 2012; Stingl et al., 2006), but in adult MGs they

3 60 appear to be maintained largely by their distinct lineage-specific stem cells, based on

61 lineage tracing studies (Tao et al., 2014; Van Keymeulen et al., 2011). The gene

62 regulatory network that must be in place to orchestrate lineage specification and

63 differentiation of stem cells into mature MEC types remains largely elusive, although a

64 number of key TFs have been identified in recent years, e.g., GATA3 has been shown as

65 a master regulator for both ductal and alveolar luminal cells (Asselin-Labat et al., 2007;

66 Kouros-Mehr et al., 2006); ELF5 was identified as a master regulator of alveolar cells

67 (Choi et al., 2009; Oakes et al., 2008); SLUG (SNAIL2) was shown as a master regulator

68 of MaSCs, and it could reprogram differentiated MECs to transplantable MaSCs, together

69 with another TF, SOX9 (Guo et al., 2012). In this work, we asked whether RUNX1 is an

70 integral part of this transcription network and how its mutations contribute to breast

71 tumorigenesis. By using genetic, cellular and molecular approaches, we found that

72 RUNX1 is a key regulator of (ER)-positive mature ductal luminal cells,

+ 73 and that loss of RUNX1 may contribute to development of ER luminal breast cancer

74 when under the background of either TP53 or RB1 loss.

75 Results

76 Runx1 is expressed in all MEC subsets except in alveolar luminal cells

77 We first measured expression levels of all three Runx genes and their common co-factor

- + hi 78 gene Cbfb in freshly sorted basal epithelial cells (Lin CD24 CD29 ) and luminal epithelial

- + lo 79 cells (Lin CD24 CD29 ) (Figure 1A) from adult virgin female mice by quantitative RT-PCR

80 (qRT-PCR). Results showed that Runx1 is the predominantly expressed Runx gene in

81 both luminal and basal cells (Figure 1B). Immunohistochemical (IHC) staining further

82 confirmed expression of RUNX1 protein in these two major MEC types in adult virgin

83 MGs (Figure 1C). However, RUNX1 expression is largely absent in alveolar luminal cells

4 84 (ALs) that start to emerge during pregnancy (Figure 1D-E). In the lactating gland, the only

85 MEC type that still expresses RUNX1 is the myoepithelial cell (Figure 1F-G). Upon

86 involution, RUNX1 expression is restored to a pattern resembling that of the virgin gland

87 (Figure 1H). Additionally, we performed microarray expression profiling of sorted subsets

- + hi - 88 of MECs, including basal cells (Lin CD24 CD29 ), luminal progenitors (LPs, Lin

+ lo + - + lo - 89 CD24 CD29 CD61 ), mature luminal cells (MLs, Lin CD24 CD29 CD61 , mainly

90 represent ductal luminal cells in virgin MGs), and alveolar luminal cells [ALs, i.e., MECs

91 genetically marked by Wap-Cre at mid-gestation; Wap-Cre is a transgenic mouse line

92 with Cre expression under the control of the Whey acidic protein (Wap) , a milk

93 protein promoter (Wagner et al., 1997)]. Estimation of Runx1 levels based on this

94 microarray data set confirmed its expression in all MEC subsets except in ALs (Figure 1I).

95 We examined Runx1 expression levels in different subsets of MECs in several additional

96 published microarray datasets (Asselin-Labat et al., 2010; Lim et al., 2010; Meier-Abt et

97 al., 2013) and further confirmed this expression pattern (Figure 1-figure supplement 1A-

98 C); in particular, in the pregnant MGs, Runx1 was also found expressed in basal MECs,

99 but not in luminal MECs (mainly ALs) (Figure 1-figure supplement 1C). Lastly, by qRT-

100 PCR, we verified that Runx1 was indeed expressed in sorted LPs and MLs, but not in

101 Wap-Cre-marked ALs (Figure 1J).

102

103 Loss of Runx1 in MECs affects multiple MEC subsets

104 The RUNX1 mutations identified from the recent sequencing studies of human breast

105 cancers include point mutations, frame-shift mutations and deletions (Banerji et al., 2012;

106 Ellis et al., 2012; Network, 2012). We analyzed the breast cancer-associated missense

107 mutations of RUNX1 to determine whether they lead to loss-of-function of RUNX1 (Figure

108 2A). Based on a previous alanine-scanning site-directed mutagenesis study (Li et al.,

5 109 2003), we found that these missense mutations either affect amino acid residues in

110 RUNX1 that directly contact DNA, or disrupt the overall fold of its DNA-binding RUNT

111 domain or abolish its binding to CBFβ, both of which would also perturb its DNA-binding

112 (Figure 2B). Thus, similar to RUNX1 deletions, the point mutations also lead to loss-of-

113 function of RUNX1, due to disrupted DNA-binding ability. Therefore, we asked whether

114 and how loss of Runx1 could affect development of normal MECs.

-/- 115 Runx1 mice died during mid-gestation mainly due to hemorrhages in the central

116 nervous system and are thus not suitable to determine the effect of Runx1-loss on MG

117 development (Okuda et al., 1996; Wang et al., 1996). We therefore used a conditional

L/L 118 knockout allele of Runx1 (Runx1 ) (Li et al., 2006b). To facilitate characterization of

119 Runx1-null MECs, we bred in a conditional Cre-reporter, Rosa26-Stop-YFP (R26Y).

120 Cross of the floxed Runx1 mice with the R26Y reporter mice and MMTV-Cre transgenic

121 mice allowed us to simultaneously disrupt Runx1 in MECs and mark the targeted cells by

122 Yellow Fluorescent Protein (YFP) (Figure 3A). Lineage analysis revealed that in virgin

123 MGs, MMTV-Cre mainly targeted MECs in the luminal lineage, but it could also lead to

124 Cre-mediated recombination in a portion of basal MECs (Figure 3B). By fluorescence-

+ L/L 125 activated cell sorting (FACS), we isolated YFP MECs from MMTV-Cre;Runx1 ;R26Y

L/+ and +/+ 126 females and MMTV-Cre;Runx1 ;R26Y control females and by qRT-PCR, we

+ L/L 127 confirmed loss of Runx1 expression in YFP MECs from MMTV-Cre;Runx1 ;R26Y

L/L 128 females (Figure 3C). Whole-mount analysis of MGs from MMTV-Cre;Runx1 ;R26Y

129 virgin females or dams on lactation day-0 did not reveal any obvious gross morphological

L/L 130 abnormalities, although a portion (3 out of 7) of MMTV-Cre;Runx1 ;R26Y females

131 exhibited a slight delay in expansion of their ductal trees during pubertal growth (Figure 3-

L/L 132 figure supplement 1A). Surprisingly however, none of the MMTV-Cre;Runx1 ;R26Y

133 dams were able to successfully nurse their pups (Figure 3-figure supplement 1B). Most of

6 134 their pups died within 24 hours postpartum and no milk spots were observed in them

+/+(or L/+) 135 compared to pups from MMTV-Cre;Runx1 ;R26Y dams. A closer examination of

L/L 136 MGs of lactating MMTV-Cre;Runx1 ;R26Y females revealed milk stasis and an

137 increasing number of cytoplasmic lipid droplets (Figure 3-figure supplement 1C). Similar

138 phenotypes have also been observed in Runx1 conditional knockout mice with Krt14-Cre

139 (i.e., Cre-expressing transgenic mice under the control of the Keratin 14 promoter) (Krt14-

L/L 140 Cre;Runx1 , data not shown), and in a number of genetically engineered mice with

141 defects in myoepithelial cell contraction and milk ejection (Haaksma et al., 2011; Li et al.,

142 2006a; Plante et al., 2010; Weymouth et al., 2012). Since Runx1 is only expressed in

143 myoepithelial cells at this stage (Figure 1D-G, I, and Figure 1-figure supplement 1C), we

144 reasoned that the nursing defects observed are most likely due to a disrupted function of

145 RUNX1 in myoepithelial cells. Additional studies are required to determine this.

L/L 146 In MMTV-Cre;Runx1 ;R26Y virgin females, we found that the percentages of the

147 YFP-marked MEC population (representing Runx1-null MECs) were significantly reduced

+/+ 148 when compared to those of the MMTV-Cre;Runx1 ;R26Y control females (Figure 3D-E).

149 Furthermore, the ratios of the YFP-marked luminal to basal subsets were also

L/L 150 significantly reduced in MMTV-Cre;Runx1 ;R26Y females (Figure 3D, F); this could be

151 due to an expansion of the YFP-marked basal population, or a reduction in the YFP-

+ 152 marked luminal population, or both. However, since the overall population of YFP MECs

L/L 153 in MMTV-Cre;Runx1 ;R26Y females was reduced (Figure 3D-E), the reduction in the

+ 154 YFP luminal/basal ratio was most likely due to a reduction in the YFP-marked Runx1-null

155 luminal population.

156

+ 157 Loss of Runx1 leads to a profound reduction in ER mature luminal cells

158 Recent studies suggest that most breast cancers, including both basal-like and luminal

7 159 subtypes, may originate from luminal cells, rather than from basal MaSCs (Keller et al.,

160 2012; Lim et al., 2009; Molyneux et al., 2010; Proia et al., 2011). Furthermore, RUNX1

161 and CBFB mutations have only been found in the luminal subtype of human breast

162 cancers (Banerji et al., 2012; Ellis et al., 2012; Network, 2012) and our data so far

163 showed that loss of Runx1 appeared to lead to a reduction in the luminal population

164 (Figure 3D-F), we therefore examined the role of RUNX1 in luminal MECs (from which

165 luminal breast cancers may originate).

166 To determine the overall defects of Runx1-null luminal MECs, we first profiled the

+ L/L 167 transcriptomes of YFP Runx1-null luminal cells (sorted from MMTV-Cre;Runx1 ;R26Y

+ 168 females) and control YFP Runx1-wild type (WT) luminal cells (sorted from MMTV-

+/+ 169 Cre;Runx1 ;R26Y) by microarray. By gene set enrichment analysis [GSEA

170 (Subramanian et al., 2005)], we observed significant enrichment of a previously

171 generated LP signature and downregulation of a ML signature in Runx1-null luminal cells

172 (Figure 4A). These LP and ML signatures were generated previously based on subset-

173 specific genes conserved in the corresponding human and mouse MEC subpopulations

174 (Lim et al., 2010). Furthermore, we also observed significant enrichment of multiple gene

175 sets related to signaling in Runx1-null luminal cells in relation to Runx1-WT luminal

176 cells (Figure 4-figure supplement 1A), possibly indicating a stress response in these

177 mutant luminal cells in vivo. Lastly, we examined expression levels of a number of TF/co-

178 factor genes known to be part of the transcription network that regulates specification and

179 maintenance of luminal MECs. In our microarray data, we found that Elf5, a TF gene

180 critically required in the alveolar cell lineage and a LP marker (Choi et al., 2009; Lim et

181 al., 2010; Oakes et al., 2008), was upregulated in Runx1-null luminal cells, whereas

182 several ductal luminal TF/co-factor genes (e.g. Gata3, Foxa1, Esr1, Cited1) were

183 downregulated (Figure 4-figure supplement 1B). When validated by qRT-PCR, we found

8 184 that although expression of Gata3 did not seem to be significantly affected in luminal cells

185 upon Runx1-loss, expression levels of Foxa1, Esr1 and Cited1 were downregulated in

186 Runx1-null luminal cells, whereas Elf5 expression was upregulated (Figure 4B).

187 Our microarray data for the entire luminal population suggested that Runx1-loss in

188 luminal MECs might lead to either a global block in luminal differentiation or loss of a

189 mature luminal MEC subpopulation. To determine this, we performed FACS analysis of

- + lo 190 the Lin CD24 CD29 luminal lineage. Intriguingly, we found that in both pubertal and adult

L/L + 191 virgin MMTV-Cre;Runx1 ;R26Y females, the YFP Runx1-null ML subpopulation defined

- + lo - - - 192 based on CD14 and c-Kit staining (Lin CD24 CD29 CD14 c-Kit , referred to as CD14 c-

- 193 Kit MLs hereafter) (Asselin-Labat et al., 2011) was significantly reduced, whereas the

+ - + lo + + + 194 YFP Runx1-null LP subpopulation (Lin CD24 CD29 CD14 c-Kit , referred to as CD14 c-

+ 195 Kit LPs hereafter) was increased (Figure 4C-D, Figure 4-figure supplement 1C-D). The

196 reduction in the ML subpopulation was further confirmed in adult virgin MMTV-

L/L - lo - 197 Cre;Runx1 ;R26Y females based on CD61 staining (MLs: Lin CD29 CD61 ) (Asselin-

198 Labat et al., 2007) (Figure 4-figure supplement 1E-F).

+ L/L 199 The residual YFP MECs in the ML gate in MMTV-Cre;Runx1 ;R26Y virgin

200 females could either represent a truly Runx1-null ML subpopulation (but reduced in

201 percentage), or represent YFP-marked MLs that have escaped Cre-mediated disruption

L + - 202 of the Runx1 allele (thus not truly Runx1-null). To determine this, we sorted YFP CD14

- L/L +/+ 203 c-Kit MLs from MMTV-Cre;Runx1 ;R26Y virgin females and MMTV-Cre;Runx1 ;R26Y

+ + + 204 control females; as an internal control, we also sorted YFP CD14 c-Kit LPs from the

+ 205 same animals. By qRT-PCR, we found that while Runx1 expression in YFP LPs from

L/L 206 MMTV-Cre;Runx1 ;R26Y females was significantly reduced, its expression in YFP-

+ 207 marked MLs was only slightly reduced (Figure 4E). This data suggested that many YFP

L/L 208 MLs in MMTV-Cre;Runx1 ;R26Y females might have escaped Cre-mediated excision in

9 L +/+ +/- 209 at least one copy of their Runx1 alleles (thus they were either Runx1 or Runx1 ). The

210 data thus also suggests that RUNX1 is essential for the emergence or maintenance of

211 the ML lineage.

212 A recent study demonstrated that the luminal cell compartment of the mouse MG

+ + 213 could be further resolved into non-clonogenic ER MLs, as well as clonogenic ER LPs

- + 214 and ER LPs based on CD49b and Sca1 staining; the ER LPs may represent progenitors

+ - 215 for ER MLs whereas the ER LPs probably represent alveolar progenitors (Shehata et

216 al., 2012). We examined these luminal subpopulations in the YFP-gated luminal

L/L 217 population in MMTV-Cre;Runx1 ;R26Y virgin females. We found that compared to their

+/+ - 218 corresponding subpopulations in MMTV-Cre;Runx1 ;R26Y control females, the CD49b

+ + L/L 219 Sca1 ER ML subpopulation was significantly reduced in MMTV-Cre;Runx1 ;R26Y

+ - - 220 females, whereas the CD49b Sca1 ER LP subpopulation was significantly increased;

+ + + 221 the CD49b Sca1 ER LP subpopulation was not significantly altered (Figure 4F-G). Of

+ 222 note, since the overall YFP luminal population was significantly reduced in MMTV-

L/L - 223 Cre;Runx1 ;R26Y females (Figure 3D-F), the increase in the ER LP subpopulation

+ 224 might be mainly due to a reduction in the ER ML subpopulation (thus proportionally

- 225 increased the percentage of the ER LP subset), rather than a significant expansion of

- + 226 ER LPs per se; similarly, although the percentage of the ER LP subpopulation was not

+ + 227 significantly changed, the absolute number of YFP ER LPs could still be reduced (due

+ 228 to an overall reduction in YFP luminal MECs). In support of this, we measured Runx1

229 expression in these three luminal MEC subpopulations. We found that whereas the YFP-

- + + 230 marked ER LP subset had a profound reduction in Runx1 expression, the YFP ER LP

+ + 231 subset exhibited a partial reduction in Runx1 transcripts, and the YFP ER ML

232 subpopulation had almost no loss of Runx1 expression (Figure 4-figure supplement 1G),

+ + 233 suggesting RUNX1 is required for both ER LPs and ER MLs. Collectively, our data

10 + 234 suggests that RUNX1 is required for the development or maintenance of the ER luminal

+ 235 lineage, and it is particularly essential for the ER MLs.

236

+ 237 Reduction in the ER luminal subpopulation upon Runx1 disruption can be rescued

238 by loss of either Trp53 or Rb1

239 From recent whole-genome/exome sequencing studies, RUNX1 and CBFB mutations

240 were only identified in the luminal subtype of human breast cancers (Banerji et al., 2012;

+ 241 Ellis et al., 2012; Network, 2012), which are typically ER . Paradoxically our data so far in

+ 242 the murine model suggests that loss-of-function of Runx1 leads to a reduction in ER

L/L 243 luminal MECs in vivo. Furthermore, we have followed MMTV-Cre;Runx1 ;R26Y females

244 for at least 18 months and have not observed any mammary tumor development in them.

245 This can be explained by a possibility in which RUNX1-mutant breast cancer originates

+ 246 from ER luminal MECs and Runx1 disruption alone actually leads to loss of the cell-of-

247 origin of such cancer. We hypothesized that additional genetic events might be needed to

+ 248 cooperate with RUNX1 loss to promote development of luminal breast cancer from ER

249 luminal MECs.

250 Interestingly, one recent sequencing study unveiled that pathway signatures of

251 RB1 , TP53 mutation, and RUNX1 mutation are co-associated with human

252 luminal B breast tumors (Ellis et al., 2012). Furthermore, by carefully examining luminal

253 breast cancer cases with RUNX1 mutations, we noticed >50% of them are accompanied

254 by mutations or deletions in either TP53 or RB1 genes (Network, 2012). Lastly, our

255 microarray data for luminal MECs suggested that loss of Runx1 might lead to activation of

256 the p53 pathway in luminal cells in general (Figure 4-figure supplement 1A). Based on

257 these observations, we hypothesized that loss of Runx1 in luminal MECs perturbs the

+ 258 fate of ER MLs, possibly leading to a stress response and subsequently upregulation of

11 259 the p53 pathway, which then triggers cell cycle arrest (or apoptosis); this would cause the

+ - 260 Runx1-null (YFP ) MLs to be outcompeted by Runx1-WT (YFP ) MLs. If this is the case,

261 then either disruption of the p53 pathway or activation of cell cycle by Rb1-loss might

+ 262 rescue the phenotype of ER ML cell loss upon Runx1 disruption. To test this, we bred

L/L L/L 263 MMTV-Cre;Runx1 ;R26Y mice to Trp53 or Rb1 conditional knockout mice (Trp53 or

L/L 264 Rb1 ). In the resulting compound mice, we were only able to follow MMTV-

L/L L/L L/L L/L 265 Cre;Runx1 ;Trp53 ;R26Y or MMTV-Cre;Runx1 ;Rb1 ;R26Y females for ~4-5 months

266 or ~9-10 months, respectively, due to lethality possibly caused by hematopoietic

267 malignancies (as MMTV-Cre has leaky expression in bone marrow hematopoietic cells).

268 Nevertheless, we were able to analyze MEC subpopulations in their MGs. Upon MMTV-

269 Cre-induced Trp53 or Rb1 loss alone, the percentages of YFP-marked MECs increased

+ 270 dramatically so that the majority of MECs in their MGs became YFP (Figure 5A,

271 increased from ~20-30% to ~70-90%), suggesting a growth advantage for Trp53-null or

- 272 Rb1-null MECs (in relation to their Trp53-WT or Rb1-WT YFP neighbors). However, the

273 percentages of the YFP-marked luminal population were both reduced (Figure 5A, middle

274 and bottom left plots compared to upper left plot, green circles). Interestingly, disruption

275 of Runx1 either together with Trp53 or with Rb1 significantly increased the percentages of

+ 276 the YFP luminal population [Figure 5A-C, increased from ~4% (Trp53-loss alone) to

277 ~11% (Runx1/Trp53-loss) (Figure 5B), and from ~12% (Rb1-loss alone) to ~23%

278 (Runx1/Rb1-loss) (Figure 5C), respectively]. Of particular note, the percentage of the

279 YFP-marked ML subpopulation, which was dramatically reduced upon Runx1-loss alone

280 (Figure 5A, upper right plot, red circle), was reverted back to almost the normal level

281 upon simultaneous loss of Runx1 together with either Trp53 or Rb1 [Figure 5A, middle

282 right plot for Trp53 (5B for quantification), bottom right plot for Rb1 (5C for quantification),

+ 283 red circles]. To verify the presence of ER MECs in the MGs of these compound female

+ 284 mice, we performed IHC staining for ERα and could indeed detect abundant ERα luminal

12 L/L L/L L/L L/L 285 MECs in both MMTV-Cre;Runx1 ;Trp53 ;R26Y and MMTV-Cre;Runx1 ;Rb1 ;R26Y

286 compound females (Figure 5-figure supplement 1A-B, since the majority of MECs in their

+ + 287 MGs were YFP , most of these ERα MECs should represent MECs with simultaneous

+ 288 loss of Runx1 and Trp53 or Rb1). As the residual YFP MECs in the ML gate from

L/L 289 MMTV-Cre;Runx1 ;R26Y females (Runx1-loss alone) appear to have escaped Cre-

L 290 mediated excision in at least one Runx1 allele (Figure 4E), we wanted to determine

+ 291 whether YFP MLs in these compound mice had undergone (or escaped) Cre-mediated

L 292 excision of their Runx1 alleles. By qRT-PCR analysis, we observed more than 50%

293 reduction in the Runx1 expression level in the YFP-marked ML subpopulation sorted from

L/L L/L 294 MMTV-Cre;Runx1 ;Rb1 ;R26Y females (Figure 5D), suggesting a significant portion of

+ L 295 these YFP MLs should have undergone biallelic excision of their Runx1 alleles.

296

297 RUNX1 controls transcription of select target genes in vitro

+ 298 Since Runx1-loss leads to a reduction in ER MLs and the residual MECs present in the

- - - + 299 CD14 c-Kit or CD49b Sca1 ML gate appear to have escaped Cre-mediated disruption of

300 Runx1 (Figure 4, Figure 4-figure supplement 1), it is technically challenging to study how

+ 301 RUNX1 controls the fate of ER luminal cells at the molecular level in this mouse model

302 directly. Therefore we first performed molecular studies in human breast cancer cell lines

+ 303 MCF7 and T47D. Although both cell lines are ER luminal breast cancer cell lines, a key

304 difference between them at the molecular level is that MCF7 cells express WT p53,

305 whereas T47D cells carry a TP53 missense mutation (nonfunctional p53) (Schafer et al.,

306 2000). Interestingly, despite multiple attempts, we were only able to obtain RUNX1

307 knockdown (kd) stable lines from TP53-mutant T47D cells, but not from TP53-WT MCF7

308 cells. This observation suggests that a similar genetic interaction between RUNX1-loss

+ 309 and TP53-loss may also operate in human ER luminal breast cells. We therefore used

+ 310 T47D cells as our cell line model to study how RUNX1 controls the fate of ER luminal

13 311 breast epithelial cells. By Western blot, we found that upon RUNX1 kd, the protein level

312 of ELF5 was increased, whereas the protein levels of both ERα and FOXA1 were

313 reduced (Figure 6A).

314 ELF5 is a master regulator of alveolar cells, a cell type in which Runx1 is not

315 expressed (Figure1 D-J, Figure 1-figure supplement 1C). Interestingly, it was shown

316 previously that RUNX1 is a direct target of ELF5 and is repressed by it, based on

317 chromatin immunoprecipitation (ChIP) analysis (Kalyuga et al., 2012). In our microarray

318 data for sorted MEC subsets, as well as those publicly available microarray data sets we

319 analyzed, we could always observe a largely opposite expression pattern of Elf5 and

320 Runx1 (Figure 1I, Figure 1-figure supplement 1). In both basal cells and MLs in which

321 Runx1 is highly expressed, Elf5 is not; Elf5 expression is greatly elevated in ALs whereas

322 Runx1 expression is repressed. This negative correlation in their expression levels could

323 be further confirmed in the HC11 cell line model. While both Runx1 and Elf5 were

324 expressed in uninduced HC11 cells, upon induction of alveolar differentiation, the ELF5

325 protein level was increased, whereas the RUNX1 protein level was reduced (Figure 6-

326 figure supplement 1A). To determine whether ELF5 is also a direct target of RUNX1, we

327 performed ChIP analysis on T47D cells and identified significant binding of RUNX1 to

328 multiple evolutionary conserved RUNX binding sites in the ELF5 (Figure 6B). The

329 RUNX1 binding was particularly profound in an region ~17kb upstream of the

330 ELF5 transcription start site (ECR-1, Figure 6B).

331 Since RUNX1 kd in T47D cells led to downregulation of ER and FOXA1 (Figure

332 6A), we asked whether ESR1 (encoding ER) and FOXA1 are direct targets of RUNX1.

333 We identified a RUNX binding motif in the ESR1 control region ~1.4kb upstream of its

334 transcription start site, as well as several RUNX binding motifs in the FOXA1 control

335 region ~1.4-1.9kb upstream of its transcription start site. By ChIP assay, we confirmed

14 336 significant binding of RUNX1 to the -1.6kb and -1.9kb motifs in the FOXA1 locus (Figure

337 6C); however, RUNX1 binding to the -1.4kb region in the ESR1 locus was not statistically

338 significant (Figure 6-figure supplement 1B). From our ChIP analysis, we also found that

339 RUNX1 could bind to a highly conserved RUNX binding site in the promoter region of

340 CITED1 (Figure 6D, Figure 6-figure supplement 1C). CITED1 encodes a selective co-

341 activator for estrogen-dependent transcription, which potentially regulates the sensitivity

342 of luminal cells to estrogen (Yahata et al., 2001). Although in T47D cells, CITED1 protein

343 level appeared unchanged upon RUNX1 kd (Figure 6A), we found that in murine models,

-/- 344 a subset of genes downregulated in Cited1 MGs (Howlin et al., 2006) were also

345 downregulated in Runx1-null luminal MECs (Figure 6-figure supplement 2). As analyzed

346 above, these downregulated genes (based on microarray) may reflect a reduction in the

347 ML subpopulation in both models, suggesting loss of Cited1 or Runx1 in MGs may lead to

+ 348 a similar phenotype (e.g., reduced ER MLs). Collectively, this data indicates that Cited1

349 may still represent a target gene of RUNX1 in vivo and thus we also included it in the

350 subsequent studies.

351

352 RUNX1 represses Elf5 and regulates mature luminal TF/co-factor genes involved in

353 the ER program in vivo

354 To determine whether RUNX1 regulates expression of these transcription regulators in

+ 355 primary cells in vivo, we took advantage of the rescue of Runx1-null ER luminal MECs

356 by Trp53 or Rb1-loss (Figure 5) and measured expression of these genes in FACS-

+ L/L L/L 357 sorted YFP luminal MEC subsets. We used MMTV-Cre;Runx1 ;Rb1 ;R26Y double

L/L 358 mutant and MMTV-Cre;Rb1 ;R26Y single mutant females for this analysis, as MMTV-

L/L L/L 359 Cre;Runx1 ;Trp53 ;R26Y double mutant females often exhibit early lethality. When

360 comparing double mutants (with Rb1/Runx1-loss) to single mutants (with Rb1-loss

+ + 361 alone), we found that both Elf5 and Esr1 appeared upregulated in ER LPs and ER MLs

15 362 (based on CD49 and Sca1 staining) from double mutants (Figure 7A-B, left plots), and

+ 363 Foxa1 and Cited1 were downregulated in the rescued double mutant ER MLs (Figure

364 7B, left plot). Since we cannot rule out a possibility in which Rb1-loss in MECs also

365 affects expression of these genes, we compared their expression in double mutants to

366 matched WT females as well. From this comparison, we found that both Elf5 and Esr1

+ 367 were also upregulated and Foxa1 and Cited1 were slightly downregulated in ER LPs and

+ 368 ER MLs from double mutants (Figure 7A-B, right plots). Furthermore, as we showed

L/L + 369 above, in MMTV-Cre;Runx1 ;R26Y females, the ER LP subset appeared less affected

+ 370 by Runx1-loss compared to the ER ML subset (Figure 4F-G), and our expression

+ 371 analysis further showed that Runx1 expression was partially reduced in ER LPs (Figure

372 4-figure supplement 1G). We therefore asked whether there is any correlation of reduced

373 Runx1 expression to changes in expression of other genes in this MEC subset.

+ 374 Interestingly, we found that in Runx1 mutant ER LPs, both Elf5 and Esr1 were

375 upregulated, and Foxa1 and Cited1 also appeared slightly upregulated (Figure 7-figure

376 supplement 1A).

377 Thus, from both cell line and in vivo expression analyses, the gene that exhibits

378 the most consistent change upon Runx1-loss is Elf5, which appears to be a target gene

+ 379 of RUNX1 repressed by it in ER luminal MECs. Intriguingly, we found that upregulation

+ 380 of Elf5 upon Runx1 disruption is not restricted to ER luminal cells and/or the Rb1-loss

+ + + - 381 genetic background. In LPs defined based on either CD14 c-Kit or CD49b Sca1 where

382 Elf5 is abundantly expressed, loss of Runx1 further increased their Elf5 expression

383 (Figure 7-figure supplement 1B). Strikingly, in basal MECs where Elf5 is normally not

384 expressed (Figure 1I, Figure 1-figure supplement 1), loss of Runx1 led to its profound

385 upregulation (Figure 7-figure supplement 1B). These data suggest that in normal MGs,

386 RUNX1 represses expression of Elf5 in almost all MEC subsets in which Runx1 is

387 expressed.

16 + 388 Our in vivo observation that Esr1 is upregulated in ER luminal MECs upon

389 Runx1-loss is both unexpected [as the in vitro data in T47D cells shows its

390 downregulation upon RUNX1 kd (Figure 6A)] and intriguing. Three possibilities may

391 explain this observation. First, similar to Elf5, RUNX1 may repress Esr1 transcription in

392 vivo and therefore loss of Runx1 may lead to de-repression of Esr1 expression. Although

393 we cannot rule out this possibility entirely, we argue that this mechanism is least likely, as

394 in our ChIP analysis, RUNX1 only exhibited a weak binding to a putative RUNX binding

395 motif we identified in the Esr1 locus (Figure 6-figure supplement 1B). The downregulation

396 of ERα in T47D cells upon RUNX1 kd is likely to be indirect [e.g., due to RUNX1-loss-

397 induced upregulation of ELF5, as overexpression of ELF5 in T47D cells can also

398 suppress ER expression (Kalyuga et al., 2012)]. Secondly, Esr1 upregulation appears

+ - - 399 restricted to the Runx1-null ER luminal MECs; in ER LPs and ER basal MECs, we did

400 not observe de-repression of Esr1 expression upon Runx1 disruption (Figure 7-figure

401 supplement 1C). This is apparently different from regulation of Elf5 by RUNX1, in which

402 loss of Runx1 leads to expression of Elf5 even in basal cells in which Elf5 is normally not

403 expressed (Figure 7-figure supplement 1B). This is also different from a recent finding of

404 repression of Esr1 by ID4, as loss of Id4 leads to widespread upregulation of Esr1

405 expression in both luminal and basal MECs (Best et al., 2014). A second possibility is that

406 under the Rb1 (or Trp53)-null background, Rb1 (or Trp53) loss not only rescues the

+ 407 Runx1-null ER luminal MECs, but also leads to their hyperproliferation. Our FACS

408 analysis for YFP-marked MECs in these animals supports this notion. In double mutant

+ 409 females, we not only observed a slight increase in the percentages of total YFP MECs

+ 410 (Figure 5A), but also increase in both the YFP luminal subset and in particular, the YFP-

+ + 411 marked ER LP and ER ML subpopulations (Figure 5A-C, Figure 7-figure supplement

412 2A). Further supporting this notion, we found that in TP53-mutant T47D cells, kd of

413 RUNX1 leads to a significant increase in their proliferation (Figure 7-figure supplement

17 + 414 2B). Lastly, we quantified ERα luminal MECs in MGs with either Runx1/Rb1-loss or

+ 415 Runx1/Trp53-loss and found that both double mutants contained significantly more ERα

416 MECs than single Rb1-loss or Trp53-loss mutants (Figure 7-figure supplement 2C-D).

417 Since our expression analysis was based on FACS-sorted MEC subsets (e.g., LP, ML)

+ - 418 and each subset may represent a mixture of both ER and ER MECs (with different

+ 419 proportions), a change in this proportion, due to overrepresentation of the rescued ER

420 luminal cells in the FACS-sorted MEC subpopulations from Rb1/Runx1-null double

421 mutants, may contribute to the ostensibly higher Esr1 transcript signals in double mutant

+ + + 422 ER LPs and ER MLs (Figure 7A-B). Furthermore, if these rescued ER luminal cells

423 represent normal MLs, we expect to observe a similar increase in the signals of the ER-

424 related luminal genes Foxa1 and Cited1. However, this was not the case (Figure 7A-B),

425 suggesting that Foxa1 and Cited1 may be target genes of RUNX1 and loss of Runx1 may

426 reduce their expression, but not that of Esr1. A third possibility for the observed higher

+ 427 Esr1 expression in both Runx1 mutant ER LPs (Figure 7-figure supplement 1A) and

+ 428 Runx1/Rb1-null ER luminal MECs (Figure 7A-B) is a compensatory upregulation of Esr1

429 expression. Loss of Runx1 may perturb the ER regulatory network, leading to Esr1

430 upregulation by disrupting a negative feedback loop (Lee et al., 2014). Alternatively, loss

+ 431 of Runx1 may lead to production of a population of abnormally fated ER luminal MECs

432 with de-repressed (upregulated) Elf5 and downregulated Foxa1 and Cited1, all of which

433 can lead to a reduction in estrogen sensitivity and/or output from ER signaling (Hurtado et

434 al., 2011; Kalyuga et al., 2012; Yahata et al., 2001), and subsequently a compensatory

435 upregulation of Esr1 expression.

436 The higher Esr1 signal makes it challenging to accurately quantify any potential

437 changes in expression levels of ER-related mature luminal genes upon Runx1 disruption,

438 by simply comparing their expression in double to single mutants or to WT controls (as

439 the matched MEC subsets based on FACS sorting may have different cell compositions,

18 + 440 if Runx1-null ER luminal MECs become over-populated). Interestingly, we noticed that in

L/L L/L 441 MLs from MMTV-Cre;Runx1 ;Rb1 ;R26Y females, Runx1 reduction became more

442 profound in older females. We therefore similarly monitored changes in expression of

443 other TF/co-factor genes over time in animals with the same genotype. This strategy may

444 allow us to control for changes introduced by differences in cell

445 populations or genetic backgrounds. By using this strategy, we found that in Rb1-null

446 single mutants, Esr1, Foxa1 and Cited1 were upregulated, and Elf5 was downregulated

- - 447 in their CD14 c-Kit MLs when they aged; however, in MLs from double mutants, following

448 Runx1 reduction, although Esr1 was upregulated to a similar level (to that of single

449 mutants), Foxa1 and Cited1 were not, and Elf5 was even further upregulated (Figure 7C).

+ + 450 We also observed a similar trend of changes for Foxa1, Cited1 and Elf5 in CD14 c-Kit

451 LPs from the same animals (Figure 7D). Of note, Esr1 expression in these LPs appeared

452 further upregulated in older females, possibly due to hyperproliferation of the rescued

+ - 453 ER luminal cells within this largely ER LP subpopulation [The LP subset defined based

+ + + 454 on CD14 c-Kit contains a small number of ER cells (Shehata et al., 2012)]. Overall, the

455 data from this time course study further supports that Elf5, Foxa1 and Cited1 are target

+ 456 genes of RUNX1 in ER luminal cells. In addition, this data also suggests that there may

+ high low low 457 exist a population of Runx1-null Elf5 Esr1 Foxa1 Cited1 abnormally differentiated

458 luminal MECs that are rescued for proliferation under the Rb1-loss background.

459 Collectively, our in vitro and in vivo expression analysis coupled with ChIP

460 analysis suggest that Elf5 is a key target gene of RUNX1 repressed by it in MECs.

461 RUNX1 also positively regulates expression of mature luminal TF/co-factor genes

462 involved in the ER program such as Foxa1 and Cited1, but it may not regulate

463 transcription of Esr1 directly.

464

465

19 466 Discussion

+ 467 In this study, we identified RUNX1 as a key regulator of ER luminal MECs. RUNX1 may

468 control the in vivo fate of this MEC subpopulation by optimizing activation of the ML gene

469 expression program (e.g., regulating key ML TF/co-factor genes, Foxa1 and Cited1) and

470 repression of the program for the alternative cell fate choice (i.e., repressing the key TF

471 gene for alveolar cells, Elf5) (Figure 8A). All three RUNX1 target genes we identified from

472 this study are in agreement with previous genetic studies in mice. Forced expression of

473 Elf5 in MECs led to slower ductal elongation due to inhibited cell proliferation (Oakes et

-/- -/- 474 al., 2008). Similarly, both Foxa1-loss (Foxa1 ) and Cited1-loss (Cited1 ) in MECs impair

475 ductal elongation and invasion by affecting ductal luminal MEC expansion/apoptosis

+ L/L 476 (Bernardo et al., 2010; Howlin et al., 2006). In ER LPs from MMTV-Cre;Runx1 ;R26Y

477 females with partial Runx1 reduction, we observed an abnormal expression pattern of

478 these genes (i.e., upregulation of both Elf5 and Esr1, and slight upregulation of Foxa1

479 and Cited1, Figure 7-figure supplement 1A). Although the currently defined LPs (largely

480 based on FACS and in vitro clonogenic assays, Figure 1A) have been proposed as

481 progenitor cells for both ductal and alveolar luminal cells (Visvader, 2009), they are

482 nevertheless a mixed population of heterogeneous progenitor cells that exhibit plasticity

483 in vitro (Shehata et al., 2012); thus it remains largely unsolved whether separate ductal or

+ 484 alveolar LPs give rise to their corresponding mature luminal cell types (i.e., ER MLs or

- 485 ER ALs, respectively), or whether bi-potent LPs give rise to either MLs or ALs, under the

+ 486 physiological setting. In the ER luminal lineage, although our data does not directly

- + + 487 address whether there is a hierarchy among ER LPs, ER LPs and ER MLs and whether

+ + 488 ER LPs are direct precursors for ER MLs, the abnormal expression pattern of key

+ 489 transcription regulators in ER LPs upon Runx1 reduction may be explained by a

+ 490 possibility in which a portion of them are committed for differentiation to ER MLs by

20 491 upregulating Esr1; but due to Runx1-loss, Elf5 is not repressed and Foxa1 and Cited1 are

492 not sufficiently upregulated in them, potentially leading to an abnormal population of

+ high low low + 493 Elf5 Esr1 Foxa1 Cited1 ML-like cells retained in the ER LP FACS gate. A small

+ 494 number of such abnormal Runx1-null ER ML-like cells may also present in the ML FACS

+ 495 gate. These abnormal ML-like cells are probably also the ER luminal subpopulation

496 rescued for proliferation upon loss of Rb1 (or Trp53). As another support for this notion,

- + 497 we measured expression levels of these TF/co-factor genes in the ER LP, ER LP and

+ 498 ER ML subsets sorted from WT animals and found that Elf5 expression trends down,

- + 499 whereas expression of Esr1, Foxa1 and Cited1 similarly trends up from ER LPs to ER

+ + + 500 LPs and then to ER MLs, and that Runx1 is also elevated from ER LPs to ER MLs

501 (Figure 7-figure supplement 2E). This expression pattern suggests that differentiation of

+ 502 ER luminal MECs requires coordinated expression of all these factors and Runx1-loss

503 may disrupt their coordinated expression.

504 Among RUNX1 target genes, the repressed Elf5 is of particular interest, as it

505 encodes a master regulatory TF for the alternative cell fate of the milk-secreting alveolar

- 506 lineage (mainly ER cells) in which Runx1 is not expressed (Figure 1D-G,I-J). We showed

507 that ELF5 is a direct target of RUNX1 and is repressed by it (Figures 6A-B and 7). Thus,

508 combined with the previous observation in which RUNX1 was reciprocally shown as a

509 direct target repressed by ELF5 (Kalyuga et al., 2012), these data suggest that RUNX1

510 and ELF5 are two master regulators for mutually exclusive cell fate choices (i.e., ductal

511 versus alveolar fates) by antagonizing each other’s transcription program [e.g., RUNX1

512 promotes the ER program (this study), whereas ELF5 suppresses it (Kalyuga et al.,

513 2012)] (Figure 8A), in a way similar to the GATA1-PU.1 paradigm for regulating the

514 choice between erythroid and myeloid fates (Huang et al., 2007).

+ 515 Intriguingly, RUNX1 not only represses Elf5 expression in ER luminal cells, but

516 also in all other MEC subsets where Runx1 is expressed (Figure 7-figure supplement

21 517 1B). The de-repression of Elf5 in basal MECs is of particular interest. Recently it was

- 518 found that RUNX1 protein expression correlates with poor prognosis in ER breast

519 cancer, and more specifically in triple-negative breast cancer (TNBC) (Ferrari et al.,

- 520 2014). Furthermore, RUNX1 was also found associated with super-enhancers in an ER

521 breast cancer cell line (Hnisz et al., 2013). As super-enhancers often associate with key

522 oncogenes in cancer cells (Loven et al., 2013), these recent findings suggest that RUNX1

- 523 may also play an oncogenic role in ER breast cancers. The link between RUNX1 and

- 524 ELF5 in basal MECs may explain a potential oncogenic role of RUNX1 in ER breast

525 cancer/TNBC, as it was shown previously that SNAI2 (encodes SLUG) is a target of

526 ELF5 repressed by it (Chakrabarti et al., 2012). Thus, it is possible that RUNX1

- 527 expression in ER breast cancer cells may repress ELF5 expression, leading to de-

528 repression (thus upregulation) of SNAI2 expression, which then promotes epithelial-

529 mesenchymal transition (EMT) and aggressiveness of breast cancer cells. Interestingly, it

530 was shown recently that Snai2-null mice exhibit a nursing defect, due to failed milk

531 ejection caused by defects in basal/myoepithelial cell differentiation (Phillips et al., 2014).

532 In Runx1-null mice, upregulation of Elf5 in basal MECs may lead to repression of Snai2,

533 which may provide an explanation for the similar nursing defect we have observed in our

534 Runx1 conditional knockout mice (Figure 3-figure supplement 1B-C).

535 RUNX1 likely represses transcription of ELF5 in ductal luminal cells by recruiting

536 repression complexes to chromatin, such as the Polycomb Repressive Complex 1

537 (PRC1). Recently it was shown that RUNX1 binds a core component of PRC1, BMI1, in

538 hematopoietic cells (Yu et al., 2012). Intriguingly, in MECs, complete loss of Bmi1 not

539 only led to reduced activity of MaSCs, but also caused premature lobuloalveolar

540 differentiation in non-pregnant females (i.e., a phenotype resembling ELF5-induced

541 alveolar differentiation from LPs), and conversely, overexpression of Bmi1 inhibited

542 lobuloalveolar differentiation induced by pregnancy hormones (i.e., a phenotype

22 543 mimicking Elf5-loss) (Pietersen et al., 2008). Thus, it is tempting to speculate that RUNX1

544 may repress transcription of ELF5 via epigenetic regulation by recruiting the BMI1-

545 containing PRC1 complex. Interestingly, ectopic expression of RUNX2, another RUNX

546 family TF, under the control of the MMTV promoter in MMTV-Runx2 transgenic mice, led

547 to compromised lobuloalveolar development during lactating, resulting in failure of milk

548 production (McDonald et al., 2014). We suspect that ectopic Runx2 expression in MECs

549 may also recruit BMI1/PRC1, leading to constitutive repression of Elf5, which impairs

550 alveolar differentiation.

551 In luminal breast cancer, our study provides strong evidence to support that

+ 552 RUNX1 plays a key role in this breast cancer subtype as a tumor suppressor in ER

553 ductal luminal cells, which may be their cells of origin. All three RUNX TFs have been

554 shown to play context-dependent roles in breast cancer development as either tumor

555 suppressors or oncogenes (Chimge and Frenkel, 2013). Among them, RUNX3 is also a

556 tumor suppressor as it is often inactivated in human breast cancers and loss of one copy

557 of Runx3 led to spontaneous mammary tumor development in a portion of aged female

558 mice (Huang et al., 2012). The tumor suppressor role of RUNX3 in breast cancer is

559 explained by its ability to inhibit ERα-dependent transactivation by reducing the stability of

560 ERα (Huang et al., 2012). In contrast, RUNX2 mainly exhibits oncogenic roles in breast

561 cancer by promoting invasiveness and metastasis via its target, SNAI2 (Chimge et al.,

562 2011); however it may also play a tumor suppressor role in breast cancer by antagonizing

563 ERα (thus, similar to RUNX3) (Chimge et al., 2012). Here we showed that RUNX1, the

+ 564 most abundantly expressed RUNX TF in MECs, controls the fate of ER luminal cells in

565 part by upregulating FOXA1 and CITED1, and by repressing ELF5. Furthermore, RUNX1

566 has also been shown as a novel tethering factor for recruiting ERα to its genomic sites for

567 ER-mediating transcriptional activation (Stender et al., 2010). Estrogen signaling has dual

568 roles in MECs and breast cancer cells; on one hand it has an oncogenic role by

23 + 569 promoting proliferation of ER luminal breast cancer cells, on the other hand it also has a

570 tumor suppressor role by promoting MEC differentiation and inhibiting metastasis of

571 breast cancer cells (Chimge and Frenkel, 2013). The tumor suppressor role of RUNX2

572 and RUNX3 mainly relates to the antagonism between RUNX2/3 and the cancer-

573 promoting program of ER signaling, whereas the tumor suppressor role of RUNX1 largely

574 correlates to its ability to positively regulate the tumor-suppression program of ER

575 signaling. This further explains the restriction of RUNX1 mutations [and only RUNX1

576 mutations, as RUNX2 and RUNX3 mutations are not found in breast cancer (Banerji et

+ 577 al., 2012; Ellis et al., 2012; Network, 2012)] to ER luminal breast cancers. This notion is

578 also consistent with a previous observation in which RUNX1 was found among a 17-gene

579 signature associated with metastasis as a gene downregulated in metastasis-prone solid

580 tumors, including breast cancer (Ramaswamy et al., 2003).

581 Our data allow us to propose the following model for the clinical observation in

582 which RUNX1 loss-of-function mutations and deletions have only been identified in

+ 583 human ER luminal breast cancers, often accompanied by mutations or copy number

584 losses in TP53 or RB1 genes (Ellis et al., 2012; Network, 2012). In patients, it is likely that

+ 585 loss of RUNX1 function similarly impairs the normal fate of ER ductal luminal cells, due

586 to a crippled ER program. This molecular change may cause cellular stress, leading to

587 activation of the p53 pathway and subsequently cell cycle arrest and/or apoptosis; as a

+ 588 result, abnormally differentiated RUNX1 mutant ER luminal cells are outcompeted by

589 their WT neighbors in vivo. However, loss of TP53 or RB1 would relieve the cell cycle

590 arrest or positively activate cell cycle, respectively, and/or rescue apoptosis in them,

+ 591 leading to rescue of the RUNX1 mutant ER luminal cells. Upon acquisition of additional

+ 592 mutations, the RUNX1/TP53-mutant or RUNX1/RB1-mutant ER premalignant luminal

+ 593 cells eventually progress to ER luminal breast cancer (Figure 8B). Of note, germline

594 mutations of RUNX1 that result in haploinsufficiency of RUNX1 can lead to an autosomal

24 595 dominant disorder referred to as familial platelet disorder with a propensity to acute

596 myeloid (FPD/AML) (Song et al., 1999). Interestingly, in one study that

597 characterized three FPD/AML pedigrees, it was found that one female patient with

598 FPD/AML also developed a breast cancer two years after AML was diagnosed, and no

599 other tumors were observed in all three pedigrees (Preudhomme et al., 2009). Although

600 the sample size for this study was too small, it certainly raises an intriguing question as to

601 whether germline mutations of RUNX1 predispose FPD/AML patients to luminal breast

602 cancer, but only under a background of either TP53 or RB1 loss.

+ 603 In summary, we identified RUNX1 as a key regulator of the ER luminal lineage.

+ 604 Loss of RUNX1 may contribute to development of ER luminal breast cancer under a

605 background of either TP53 or RB1 loss, and upon cooperation with other additional

606 oncogenic events.

607

608 Materials and methods

609 Mice

L/L 610 Mice carrying the floxed Runx1 allele (Runx1 ) (Li et al., 2006b) were bred with mice

611 carrying a conditional Cre-reporter, R26Y. Subsequently these mice were bred with mice

612 that drive expression of Cre recombinase under the control of the mouse mammary tumor

613 virus (MMTV) promoter (MMTV-Cre), and with mice carrying the floxed Trp53 allele

L/L L/L 614 (Trp53 ) or floxed Rb1 allele (Rb1 ). For studying Runx1 disruption in basal MECs, Cre

615 transgenic mice under the control of the Keratin 14 promoter (Krt14-Cre) were also used.

616 Mice were obtained from JAX (R26Y: 006148; MMTV-Cre: 003553) or the MMHCC

617 repository (Krt14-Cre: 01XF1; Wap-Cre: 01XA8) or were a generous gift from Dr. Stuart

L/L L/L 618 Orkin [Trp53 and Rb1 (Walkley et al., 2008)]. All animal experiments and procedures

619 were approved by our Institutional Animal Care and Use Committee (IACUC).

25 620

621 Whole-mount, histology and immunohistochemistry

622 Whole-mounts of MGs of pubertal, adult virgin or lactation day-0 mice were fixed and

623 processed as previously described (Jones et al., 1996). For histology and

624 immunohistochemical staining, MGs were fixed in 10% formalin and embedded in

625 paraffin. For RUNX1 or ERα detection, antigen retrieval (Citrate buffer pH6.0, 20 minutes

626 boil in microwave oven) was performed prior to incubation with an anti-RUNX1 antibody

627 (2593-1, Epitomics, Burlingame, CA) or an anti-ERα antibody (SC-542, Santa Cruz

628 Biotechnology, Dallas, TX). Signal was detected using the impress reagent kit and DAB

629 substrate (MP-7401 and SK-4100, Vector Laboratories, Burlingame, CA).

630

631 Mammary gland cell preparation, flow cytometric analysis and cell sorting

632 Thoracic and inguinal mammary glands were dissected from pubertal or adult virgin

633 female mice and cell suspensions were prepared as previously described (Shackleton et

634 al., 2006). Flow cytometric analysis was performed with the DXP11 analyzer (Cytek,

635 Fremont, CA) or the Accuri C6 analyzer (BD Biosciences, San Jose, CA). FACS sorting

636 was performed with a FACSAria sorter (BD Biosciences). Data was analyzed with FlowJo

637 (Tree Star, Ashland, OR) or CFlow (BD Biosciences). Antibodies used for FACS were

638 purchased from eBiosciences (San Diego, Ca) and included, CD24-eFluor450, CD24-

639 eFluor605, CD29-APC, CD61-PE, c-Kit-PE-CY7, CD14-PE, CD49b-PE, Sca1-APC and

640 biotinylated CD31, CD45 and TER119 [i.e., lineage (Lin) markers], as well as

641 Streptavidin-PerCP-CY5.5. We also used a Sca1-APC-CY7 antibody purchased from BD

642 biosciences (San Jose, CA).

643

644 Microarray analysis and quantitative RT-PCR

645 Total RNA from sorted subsets of MECs was prepared by the RNeasy kit (Qiagen,

26 646 Valencia, CA) and amplified with the Ovation RNA Amplification System V2 (Nugen, San

647 Carlos, CA). YFP-marked luminal cells were sorted from adult virgin MMTV-

L/L +/+ 648 Cre;Runx1 ;R26Y or MMTV-Cre Runx1 ;R26Y littermates. Normal MEC subsets,

649 including MaSCs, LPs, and MLs, were sorted from WT C57/B6 adult virgin females;

+ 650 alveolar luminal cells (ALs) were sorted as YFP cells from Wap-Cre;R26Y females at

651 mid-gestation. Mouse Genome 430 2.0 Array (Affymetrix, Santa Clara, CA) was used to

652 generate the expression profiles. All arrays were normalized by dCHIP and analyzed by

653 GSEA as described (Subramanian et al., 2005), using MSigDB database v3.1

654 (http://www.broadinstitute.org/gsea/msigdb/index.jsp). For qRT-PCR, cDNA was

655 generated with Omniscript (Qiagen) according to the manufacture’s protocol and real-

656 time PCR was performed using FastStart SYBR Green Master (Roche). ddCt method

657 was used for normalization to the control group and to the endogenous control (Hprt).

658 Primers are listed in supplementary file 1.

659

660 ChIP analysis

661 Cells were cross-linked with 1% formaldehyde at room temperature for 10 minutes,

662 quenched with 0.125M glycine for 5 minutes and washed with PBS, harvested by

663 scraping and lysed in cell lysis buffer (0.1%SDS; 0.5%NP40; 1mM EDTA; 10mM Tris-

664 HCl, pH 7.4; 0.5% NaDOC). 200-1000bp DNA fragments were obtained after sonication.

665 After 10 minutes centrifugation at speed at 4°C, supernatant was used for IP

666 overnight at 4°C. 30μl Dynabeads Protein G beads (Invitrogen) and 1μg antibody were

667 used for each IP. One tenth of lysate was saved as input. The following antibodies were

668 used: rabbit anti-RUNX1 (ab92336, Abcam), rabbit IgG (sc-2027, Santa Cruz

669 Biotechnology). The beads were washed twice with the following buffers, 3 minutes each:

670 low-salt buffer (0.1% SDS; 1% Triton X-100; 1mM EDTA; 10mM Tris-HCl, pH 7.4; 300mM

671 NaCl; 0.1% NaDOC), high-salt buffer (0.1% SDS; 1% Triton X-100; 1 mM EDTA; 10mM

27 672 Tris-HCl, pH 7.4; 500mM NaCl; 0.1% NaDOC), LiCl buffer (10mM Tris–HCl pH 8, 0.25M

673 LiCl, 1mM EDTA pH 8, 1% NP-40, 1% NaDOC) and TE. Precipitated materials were

674 eluted with 300 μl elution buffer (1% SDS, 0.1M NaHCO3, 50mM Tris-HCl pH 8 and

675 10mM EDTA). Chromatin was reverse-crosslinked by adding 12 μl of 5M NaCl and

676 incubated overnight at 65°C. DNA was obtained after RNaseA treatment, protease K

677 treatment, phenol/chloroform extraction, and ethanol precipitation. DNA was analyzed by

678 qPCR, normalized to the input DNA. Primers are listed in supplementary file 1.

679

680 RUNX1 knockdown, Western blot and proliferation assay

681 shRNAs for RUNX1 were purchased from Open Biosystems (shRNA sequences are

682 listed in supplementary file 1, data from a pool of TRCN0000013659-D1 and

683 TRCN0000013662-D4 was shown). After lentiviral infection and puromycin selection,

684 stable shRNA-expressing cell lines were generated. For Western blotting, whole-cell

685 extracts were prepared by boiling cells for 10 min at 95°C in SDS sample buffer [50mM

686 Tris (pH 6.8), 100mM DTT, 2% SDS, 0.1% bromophenol blue, 10%glycerol]. Cell lysates

687 were then resolved by SDS-PAGE. β-actin (Fisher Lab) was used as a loading control.

688 Primary antibodies (RUNX1: Abcam ab92336, ELF5: Abcam ab77007, CITED1: Abcam

689 ab92550, ERα: Santa Cruz Biotechnology sc-8002, FOXA1: Santa Cruz Biotechnology

690 sc-6553) were detected using HRP-conjugated anti-rabbit antibodies and visualized using

691 enhanced chemiluminescence detection (ECL reagents from Fisher Lab). Proliferation of

692 T47D cells was determined by absorbance of alamarBlue, following manufacturer’s

5 693 protocol (Invitrogen Lot155363SA). 110 Cells were seeded in 96-well plate, and were

694 measured after 3 or 5 days in culture. 1/10 volume of alamarBlue reagent was directly

695 added to cells in culture medium, incubated for 4 hours at 37°C. Absorbance of

696 alamarBlue was monitored at 570 nm, using 600 nm as a reference wavelength

697 (normalized to the 600 nm value).

28 698

699 Statistical analysis

700 The results were reported as Mean ± S.E.M. unless otherwise indicated, and Student’s t-

701 Tests were used to calculate statistical significance.

702

703 Accession numbers

704 The microarray expression profiling data sets generated in this manuscript have been

705 deposited to the GEO database under the following accession numbers: GSE47375 (for

706 Runx1) and GSE47376 (for normal MEC subsets) or as SuperSeries GSE47377.

707

708 Acknowledgements

709 We are grateful to Y. Qiu and G. Losyev for expert technical assistance with FACS

710 sorting, Y. Shao and E. Fox for microarray, R. Bronson for histology, A. Cantor and H.

L/L L/L 711 Huang for providing Runx1 mice, S. Orkin and A. Chen for providing Trp53 and

L/L 712 Rb1 mice, and L. Ellisen for providing HC11 cells. This research was supported by a

713 Pathway-to-Independence K99/R00 grant from NCI (CA126980), a Seed Grant from

714 Harvard Stem Cell Institute, Milton Fund Award from Harvard University, Hearst

715 Foundation Young Investigator Award from Brigham and Women’s Hospital (BWH), and

716 Start-up Fund from BWH to Z.L.; Z.L. is also supported by NIH R01 HL107663.

717

718 References

719 Asselin-Labat, M.L., Sutherland, K.D., Barker, H., Thomas, R., Shackleton, M., Forrest,

720 N.C., Hartley, L., Robb, L., Grosveld, F.G., van der Wees, J., et al. (2007). Gata-3 is an

721 essential regulator of mammary-gland morphogenesis and luminal-cell differentiation. Nat

722 Cell Biol 9, 201-209.

29 723 Asselin-Labat, M.L., Sutherland, K.D., Vaillant, F., Gyorki, D.E., Wu, D., Holroyd, S.,

724 Breslin, K., Ward, T., Shi, W., Bath, M.L., et al. (2011). Gata-3 negatively regulates the

725 tumor-initiating capacity of mammary luminal progenitor cells and targets the putative

726 tumor suppressor caspase-14. Mol Cell Biol 31, 4609-4622.

727 Asselin-Labat, M.L., Vaillant, F., Sheridan, J.M., Pal, B., Wu, D., Simpson, E.R., Yasuda,

728 H., Smyth, G.K., Martin, T.J., Lindeman, G.J., et al. (2010). Control of mammary stem cell

729 function by steroid hormone signalling. Nature 465, 798-802.

730 Banerji, S., Cibulskis, K., Rangel-Escareno, C., Brown, K.K., Carter, S.L., Frederick, A.M.,

731 Lawrence, M.S., Sivachenko, A.Y., Sougnez, C., Zou, L., et al. (2012). Sequence

732 analysis of mutations and translocations across breast cancer subtypes. Nature 486, 405-

733 409.

734 Bernardo, G.M., Lozada, K.L., Miedler, J.D., Harburg, G., Hewitt, S.C., Mosley, J.D.,

735 Godwin, A.K., Korach, K.S., Visvader, J.E., Kaestner, K.H., et al. (2010). FOXA1 is an

736 essential determinant of ERalpha expression and mammary ductal morphogenesis.

737 Development 137, 2045-2054.

738 Best, S.A., Hutt, K.J., Fu, N.Y., Vaillant, F., Liew, S.H., Hartley, L., Scott, C.L., Lindeman,

739 G.J., and Visvader, J.E. (2014). Dual roles for Id4 in the regulation of estrogen signaling

740 in the mammary gland and ovary. Development 141, 3159-3164.

741 Bravo, J., Li, Z., Speck, N.A., and Warren, A.J. (2001). The leukemia-associated AML1

742 (Runx1)--CBF beta complex functions as a DNA-induced molecular clamp. Nat Struct Biol

743 8, 371-378.

744 Chakrabarti, R., Hwang, J., Andres Blanco, M., Wei, Y., Lukacisin, M., Romano, R.A.,

745 Smalley, K., Liu, S., Yang, Q., Ibrahim, T., et al. (2012). Elf5 inhibits the epithelial-

746 mesenchymal transition in mammary gland development and breast cancer metastasis

747 by transcriptionally repressing Snail2. Nat Cell Biol 14, 1212-1222.

30 748 Chimge, N.O., Baniwal, S.K., Little, G.H., Chen, Y.B., Kahn, M., Tripathy, D., Borok, Z.,

749 and Frenkel, B. (2011). Regulation of breast cancer metastasis by Runx2 and estrogen

750 signaling: the role of SNAI2. Breast Cancer Res 13, R127.

751 Chimge, N.O., Baniwal, S.K., Luo, J., Coetzee, S., Khalid, O., Berman, B.P., Tripathy, D.,

752 Ellis, M.J., and Frenkel, B. (2012). Opposing effects of Runx2 and estradiol on breast

753 cancer cell proliferation: in vitro identification of reciprocally regulated gene signature

754 related to clinical letrozole responsiveness. Clin Cancer Res 18, 901-911.

755 Chimge, N.O., and Frenkel, B. (2013). The RUNX family in breast cancer: relationships

756 with estrogen signaling. Oncogene 32, 2121-2130.

757 Choi, Y.S., Chakrabarti, R., Escamilla-Hernandez, R., and Sinha, S. (2009). Elf5

758 conditional knockout mice reveal its role as a master regulator in mammary alveolar

759 development: failure of Stat5 activation and functional differentiation in the absence of

760 Elf5. Dev Biol 329, 227-241.

761 Chuang, L.S., Ito, K., and Ito, Y. (2013). RUNX family: Regulation and diversification of

762 roles through interacting . Int J Cancer 132, 1260-1271.

763 Ellis, M.J., Ding, L., Shen, D., Luo, J., Suman, V.J., Wallis, J.W., Van Tine, B.A., Hoog, J.,

764 Goiffon, R.J., Goldstein, T.C., et al. (2012). Whole-genome analysis informs breast

765 cancer response to aromatase inhibition. Nature 486, 353-360.

766 Ferrari, N., Mohammed, Z.M., Nixon, C., Mason, S.M., Mallon, E., McMillan, D.C., Morris,

767 J.S., Cameron, E.R., Edwards, J., and Blyth, K. (2014). Expression of RUNX1 correlates

768 with poor patient prognosis in triple negative breast cancer. PLoS One 9, e100759.

769 Guo, W., Keckesova, Z., Donaher, J.L., Shibue, T., Tischler, V., Reinhardt, F., Itzkovitz,

770 S., Noske, A., Zurrer-Hardi, U., Bell, G., et al. (2012). Slug and Sox9 cooperatively

771 determine the mammary stem cell state. Cell 148, 1015-1028.

31 772 Haaksma, C.J., Schwartz, R.J., and Tomasek, J.J. (2011). Myoepithelial cell contraction

773 and milk ejection are impaired in mammary glands of mice lacking smooth muscle alpha-

774 actin. Biol Reprod 85, 13-21.

775 Hnisz, D., Abraham, B.J., Lee, T.I., Lau, A., Saint-Andre, V., Sigova, A.A., Hoke, H.A.,

776 and Young, R.A. (2013). Super-enhancers in the control of cell identity and disease. Cell

777 155, 934-947.

778 Howlin, J., McBryan, J., Napoletano, S., Lambe, T., McArdle, E., Shioda, T., and Martin,

779 F. (2006). CITED1 homozygous null mice display aberrant pubertal mammary ductal

780 morphogenesis. Oncogene 25, 1532-1542.

781 Huang, B., Qu, Z., Ong, C.W., Tsang, Y.H., Xiao, G., Shapiro, D., Salto-Tellez, M., Ito, K.,

782 Ito, Y., and Chen, L.F. (2012). RUNX3 acts as a tumor suppressor in breast cancer by

783 targeting . Oncogene 31, 527-534.

784 Huang, S., Guo, Y.P., May, G., and Enver, T. (2007). Bifurcation dynamics in lineage-

785 commitment in bipotent progenitor cells. Dev Biol 305, 695-713.

786 Hurtado, A., Holmes, K.A., Ross-Innes, C.S., Schmidt, D., and Carroll, J.S. (2011).

787 FOXA1 is a key determinant of estrogen receptor function and endocrine response. Nat

788 Genet 43, 27-33.

789 Jones, F.E., Jerry, D.J., Guarino, B.C., Andrews, G.C., and Stern, D.F. (1996). Heregulin

790 induces in vivo proliferation and differentiation of mammary into secretory

791 lobuloalveoli. Cell Growth Differ 7, 1031-1038.

792 Kalyuga, M., Gallego-Ortega, D., Lee, H.J., Roden, D.L., Cowley, M.J., Caldon, C.E.,

793 Stone, A., Allerdice, S.L., Valdes-Mora, F., Launchbury, R., et al. (2012). ELF5

794 suppresses estrogen sensitivity and underpins the acquisition of antiestrogen resistance

795 in luminal breast cancer. PLoS Biol 10, e1001461.

32 796 Keller, P.J., Arendt, L.M., Skibinski, A., Logvinenko, T., Klebba, I., Dong, S., Smith, A.E.,

797 Prat, A., Perou, C.M., Gilmore, H., et al. (2012). Defining the cellular precursors to human

798 breast cancer. Proc Natl Acad Sci U S A 109, 2772-2777.

799 Kouros-Mehr, H., Slorach, E.M., Sternlicht, M.D., and Werb, Z. (2006). GATA-3 maintains

800 the differentiation of the luminal cell fate in the mammary gland. Cell 127, 1041-1055.

801 Lee, J., Tiwari, A., Shum, V., Mills, G.B., Mancini, M.A., Igoshin, O.A., and Balazsi, G.

802 (2014). Unraveling the regulatory connections between two controllers of breast cancer

803 cell fate. Nucleic Acids Res 42, 6839-6849.

804 Li, S., Chang, S., Qi, X., Richardson, J.A., and Olson, E.N. (2006a). Requirement of a

805 myocardin-related transcription factor for development of mammary myoepithelial cells.

806 Mol Cell Biol 26, 5797-5808.

807 Li, W., Ferguson, B.J., Khaled, W.T., Tevendale, M., Stingl, J., Poli, V., Rich, T.,

808 Salomoni, P., and Watson, C.J. (2009). PML depletion disrupts normal mammary gland

809 development and skews the composition of the mammary luminal cell progenitor pool.

810 Proc Natl Acad Sci U S A 106, 4725-4730.

811 Li, Z., Chen, M.J., Stacy, T., and Speck, N.A. (2006b). Runx1 function in hematopoiesis is

812 required in cells that express Tek. Blood 107, 106-110.

813 Li, Z., Yan, J., Matheny, C.J., Corpora, T., Bravo, J., Warren, A.J., Bushweller, J.H., and

814 Speck, N.A. (2003). Energetic contribution of residues in the Runx1 Runt domain to DNA

815 binding. J Biol Chem 278, 33088-33096.

816 Lim, E., Vaillant, F., Wu, D., Forrest, N.C., Pal, B., Hart, A.H., Asselin-Labat, M.L., Gyorki,

817 D.E., Ward, T., Partanen, A., et al. (2009). Aberrant luminal progenitors as the candidate

818 target population for basal tumor development in BRCA1 mutation carriers. Nat Med 15,

819 907-913.

820 Lim, E., Wu, D., Pal, B., Bouras, T., Asselin-Labat, M.L., Vaillant, F., Yagita, H.,

821 Lindeman, G.J., Smyth, G.K., and Visvader, J.E. (2010). Transcriptome analyses of

33 822 mouse and human mammary cell subpopulations reveal multiple conserved genes and

823 pathways. Breast Cancer Res 12, R21.

824 Loven, J., Hoke, H.A., Lin, C.Y., Lau, A., Orlando, D.A., Vakoc, C.R., Bradner, J.E., Lee,

825 T.I., and Young, R.A. (2013). Selective inhibition of tumor oncogenes by disruption of

826 super-enhancers. Cell 153, 320-334.

827 McDonald, L., Ferrari, N., Terry, A., Bell, M., Mohammed, Z.M., Orange, C., Jenkins, A.,

828 Muller, W.J., Gusterson, B.A., Neil, J.C., et al. (2014). RUNX2 correlates with subtype-

829 specific breast cancer in a human tissue microarray, and ectopic expression of Runx2

830 perturbs differentiation in the mouse mammary gland. Dis Model Mech 7, 525-534.

831 Meier-Abt, F., Milani, E., Roloff, T., Brinkhaus, H., Duss, S., Meyer, D.S., Klebba, I.,

832 Balwierz, P.J., van Nimwegen, E., and Bentires-Alj, M. (2013). Parity induces

833 differentiation and reduces Wnt/Notch signaling ratio and proliferation potential of basal

834 stem/progenitor cells isolated from mouse mammary epithelium. Breast Cancer Res 15,

835 R36.

836 Molyneux, G., Geyer, F.C., Magnay, F.A., McCarthy, A., Kendrick, H., Natrajan, R.,

837 Mackay, A., Grigoriadis, A., Tutt, A., Ashworth, A., et al. (2010). BRCA1 basal-like breast

838 cancers originate from luminal epithelial progenitors and not from basal stem cells. Cell

839 Stem Cell 7, 403-417.

840 Network, T.C.G.A. (2012). Comprehensive molecular portraits of human breast tumours.

841 Nature 490, 61-70.

842 Oakes, S.R., Naylor, M.J., Asselin-Labat, M.L., Blazek, K.D., Gardiner-Garden, M., Hilton,

843 H.N., Kazlauskas, M., Pritchard, M.A., Chodosh, L.A., Pfeffer, P.L., et al. (2008). The Ets

844 transcription factor Elf5 specifies mammary alveolar cell fate. Genes Dev 22, 581-586.

845 Okuda, T., van Deursen, J., Hiebert, S.W., Grosveld, G., and Downing, J.R. (1996).

846 AML1, the target of multiple chromosomal translocations in human leukemia, is essential

847 for normal fetal liver hematopoiesis. Cell 84, 321-330.

34 848 Phillips, S., Prat, A., Sedic, M., Proia, T., Wronski, A., Mazumdar, S., Skibinski, A.,

849 Shirley, S.H., Perou, C.M., Gill, G., et al. (2014). Cell-State Transitions Regulated by

850 SLUG Are Critical for Tissue Regeneration and Tumor Initiation. Stem Cell Reports 2,

851 633-647.

852 Pietersen, A.M., Evers, B., Prasad, A.A., Tanger, E., Cornelissen-Steijger, P., Jonkers, J.,

853 and van Lohuizen, M. (2008). Bmi1 regulates stem cells and proliferation and

854 differentiation of committed cells in mammary epithelium. Curr Biol 18, 1094-1099.

855 Plante, I., Wallis, A., Shao, Q., and Laird, D.W. (2010). Milk secretion and ejection are

856 impaired in the mammary gland of mice harboring a Cx43 mutant while expression and

857 localization of tight and adherens junction proteins remain unchanged. Biol Reprod 82,

858 837-847.

859 Preudhomme, C., Renneville, A., Bourdon, V., Philippe, N., Roche-Lestienne, C., Boissel,

860 N., Dhedin, N., Andre, J.M., Cornillet-Lefebvre, P., Baruchel, A., et al. (2009). High

861 frequency of RUNX1 biallelic alteration in secondary to familial

862 platelet disorder. Blood 113, 5583-5587.

863 Proia, T.A., Keller, P.J., Gupta, P.B., Klebba, I., Jones, A.D., Sedic, M., Gilmore, H.,

864 Tung, N., Naber, S.P., Schnitt, S., et al. (2011). Genetic predisposition directs breast

865 cancer phenotype by dictating progenitor cell fate. Cell Stem Cell 8, 149-163.

866 Ramaswamy, S., Ross, K.N., Lander, E.S., and Golub, T.R. (2003). A molecular

867 signature of metastasis in primary solid tumors. Nat Genet 33, 49-54.

868 Schafer, J.M., Lee, E.S., O'Regan, R.M., Yao, K., and Jordan, V.C. (2000). Rapid

869 development of tamoxifen-stimulated mutant p53 breast tumors (T47D) in athymic mice.

870 Clin Cancer Res 6, 4373-4380.

871 Scheitz, C.J., and Tumbar, T. (2013). New insights into the role of Runx1 in epithelial

872 stem cell biology and pathology. J Cell Biochem 114, 985-993.

35 873 Shackleton, M., Vaillant, F., Simpson, K.J., Stingl, J., Smyth, G.K., Asselin-Labat, M.L.,

874 Wu, L., Lindeman, G.J., and Visvader, J.E. (2006). Generation of a functional mammary

875 gland from a single stem cell. Nature 439, 84-88.

876 Shehata, M., Teschendorff, A., Sharp, G., Novcic, N., Russell, A., Avril, S., Prater, M.,

877 Eirew, P., Caldas, C., Watson, C.J., et al. (2012). Phenotypic and functional

878 characterization of the luminal cell hierarchy of the mammary gland. Breast Cancer Res

879 14, R134.

880 Song, W.J., Sullivan, M.G., Legare, R.D., Hutchings, S., Tan, X., Kufrin, D., Ratajczak, J.,

881 Resende, I.C., Haworth, C., Hock, R., et al. (1999). Haploinsufficiency of CBFA2 causes

882 familial thrombocytopenia with propensity to develop acute myelogenous leukaemia. Nat

883 Genet 23, 166-175.

884 Speck, N.A., and Gilliland, D.G. (2002). Core-binding factors in and

885 leukaemia. Nat Rev Cancer 2, 502-513.

886 Spike, B.T., Engle, D.D., Lin, J.C., Cheung, S.K., La, J., and Wahl, G.M. (2012). A

887 mammary stem cell population identified and characterized in late embryogenesis reveals

888 similarities to human breast cancer. Cell Stem Cell 10, 183-197.

889 Stender, J.D., Kim, K., Charn, T.H., Komm, B., Chang, K.C., Kraus, W.L., Benner, C.,

890 Glass, C.K., and Katzenellenbogen, B.S. (2010). Genome-wide analysis of estrogen

891 receptor alpha DNA binding and tethering mechanisms identifies Runx1 as a novel

892 tethering factor in receptor-mediated transcriptional activation. Mol Cell Biol 30, 3943-

893 3955.

894 Stingl, J., Eirew, P., Ricketson, I., Shackleton, M., Vaillant, F., Choi, D., Li, H.I., and

895 Eaves, C.J. (2006). Purification and unique properties of mammary epithelial stem cells.

896 Nature 439, 993-997.

897 Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L., Gillette, M.A.,

898 Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., et al. (2005). Gene set

36 899 enrichment analysis: a knowledge-based approach for interpreting genome-wide

900 expression profiles. Proc Natl Acad Sci U S A 102, 15545-15550.

901 Tahirov, T.H., Inoue-Bungo, T., Morii, H., Fujikawa, A., Sasaki, M., Kimura, K., Shiina, M.,

902 Sato, K., Kumasaka, T., Yamamoto, M., et al. (2001). Structural analyses of DNA

903 recognition by the AML1/Runx-1 Runt domain and its allosteric control by CBFbeta. Cell

904 104, 755-767.

905 Taniuchi, I., Osato, M., and Ito, Y. (2012). Runx1: no longer just for leukemia. Embo J 31,

906 4098-4099.

907 Tao, L., van Bragt, M.P.A., Laudadio, E., and Li, Z. (2014). Lineage Tracing of Mammary

908 Epithelial Cells Using Cell-Type-Specific Cre-Expressing Adenoviruses. Stem Cell

909 Reports 2, 770-779.

910 Van Keymeulen, A., Rocha, A.S., Ousset, M., Beck, B., Bouvencourt, G., Rock, J.,

911 Sharma, N., Dekoninck, S., and Blanpain, C. (2011). Distinct stem cells contribute to

912 mammary gland development and maintenance. Nature 479, 189-193.

913 Visvader, J.E. (2009). Keeping abreast of the mammary epithelial hierarchy and breast

914 tumorigenesis. Genes Dev 23, 2563-2577.

915 Wagner, K.U., Wall, R.J., St-Onge, L., Gruss, P., Wynshaw-Boris, A., Garrett, L., Li, M.,

916 Furth, P.A., and Hennighausen, L. (1997). Cre-mediated gene in the mammary

917 gland. Nucleic Acids Res 25, 4323-4330.

918 Walkley, C.R., Qudsi, R., Sankaran, V.G., Perry, J.A., Gostissa, M., Roth, S.I., Rodda,

919 S.J., Snay, E., Dunning, P., Fahey, F.H., et al. (2008). Conditional mouse osteosarcoma,

920 dependent on p53 loss and potentiated by loss of Rb, mimics the human disease. Genes

921 Dev 22, 1662-1676.

922 Wang, Q., Stacy, T., Binder, M., Marin-Padilla, M., Sharpe, A.H., and Speck, N.A. (1996).

923 Disruption of the Cbfa2 gene causes necrosis and hemorrhaging in the central nervous

924 system and blocks definitive hematopoiesis. Proc Natl Acad Sci U S A 93, 3444-3449.

37 925 Weymouth, N., Shi, Z., and Rockey, D.C. (2012). Smooth muscle alpha actin is

926 specifically required for the maintenance of lactation. Dev Biol 363, 1-14.

927 Yahata, T., Shao, W., Endoh, H., Hur, J., Coser, K.R., Sun, H., Ueda, Y., Kato, S.,

928 Isselbacher, K.J., Brown, M., et al. (2001). Selective coactivation of estrogen-dependent

929 transcription by CITED1 CBP/p300-binding protein. Genes Dev 15, 2598-2612.

930 Yu, M., Mazor, T., Huang, H., Huang, H.T., Kathrein, K.L., Woo, A.J., Chouinard, C.R.,

931 Labadorf, A., Akie, T.E., Moran, T.B., et al. (2012). Direct recruitment of polycomb

932 repressive complex 1 to chromatin by core binding transcription factors. Mol Cell 45, 330-

933 343.

934

935

38 936 Figure legends

937 Figure 1. Expression pattern of Runx1 in murine MGs. (A) Schematic diagram of a

938 simplified version of the MEC hierarchy. MECs can be separated into the luminal and

939 basal lineages. Major MEC subpopulations, their names and name abbreviations, as well

940 as their marker expression patterns are shown. Note “luminal progenitor (LP)” has been

941 used to refer to progenitor cells for the luminal lineage defined based on either CD61

942 (Asselin-Labat et al., 2007), or CD14 and c-Kit (Asselin-Labat et al., 2011), or CD49b (Li

943 et al., 2009), and is therefore a mixture of overlapping progenitor cell populations, and

944 may include common or separate progenitors for ductal and alveolar luminal cells. (B)

945 qRT-PCR analysis of Runx1, Runx2, Runx3 and Cbfb transcripts isolated from luminal

946 and basal cells of adult virgin female mice. (C-H) IHC staining for RUNX1 on sections of

947 MGs at different developmental stages: (C) adult virgin, (D-E) midgestation (the region

948 highlighted in D is shown in E), (F-G) lactation (the region highlighted in F is shown in G),

949 and (H) after involution. Arrows and arrowheads indicate RUNX1-expressing luminal and

950 basal cells, respectively; * indicates lumen. Scale bars = 20μm. (I) Relative expression

951 values of indicated genes determined by microarray analysis of the indicated MEC

952 subpopulations isolated from the MGs of adult virgin female mice. ALs were isolated as

+ 953 YFP cells from Wap-Cre;R26Y females (i.e., MECs genetically marked by the Wap-Cre

954 transgene) during midgestation. Affymetrix probes used to estimate expression of each

955 indicated gene are 1419555_at, 1422864_at, 1448886_at, 1435663_at, 1449031_at and

956 1418496_at for Elf5, Runx1, Gata3, Esr1, Cited1, and Foxa1, respectively. (J) Runx1

957 expression levels were confirmed in sorted LPs, MLs and ALs (as in I) by qRT-PCR.

958 The following figure supplement is available for figure 1:

959 Figure supplement 1. Expression analysis of Runx1 and other select luminal

960 transcription factor (TF) genes based on microarray.

39 961

962 Figure 2. Analysis of RUNX1 mutations. RUNX1 somatic missense mutations identified

963 in human breast cancers disrupt its DNA binding either directly (disrupting direct DNA

964 contact) or indirectly (disrupting the overall protein fold of its DNA-binding RUNT domain

965 or disrupting CBFβ binding). (A) RUNX1 full-length protein sequence; RUNT domain is

966 highlighted in blue. The three amino acid residues affected by point mutations in luminal

967 breast cancers [based on (Ellis et al., 2012)] are shown in red. Several additional

968 missense mutations [based on (Network, 2012; Taniuchi et al., 2012)] are also

969 highlighted with red font. (B) How these missense mutations affect RUNX1 DNA binding

970 is predicted based on a previous structural and biochemical analysis of the RUNT domain

971 (Li et al., 2003).

972

973 Figure 3. Runx1 loss leads to a reduction in the luminal MEC population. (A)

974 Schematic representation of the Runx1 conditional knockout allele in which its 4 is

975 flanked by loxP sites, as well as the R26Y conditional Cre-reporter. STOP: transcriptional

976 stopper cassette. Subsequent breeding with MMTV-Cre resulted in mice in which

977 selected subsets of MECs express YFP and lack expression of functional RUNX1. (B)

- + + 978 FACS gating strategy for detecting lin YFP (lin: lineage markers) MECs as well as YFP

- + lo - + hi 979 lin CD24 CD29 luminal (Lu) and lin CD24 CD29 basal (Ba) MECs in MMTV-Cre;R26Y

980 females. Str: stromal cells. (C) qRT-PCR analysis confirming loss of Runx1 expression in

+ L/L 981 YFP MECs sorted from MMTV-Cre;Runx1 ;R26Y females (L/L). (D) FACS analysis

- + - + 982 showing the reduced lin YFP MEC population (left plots), as well as the reduced lin YFP

L/L 983 luminal population (right plots), in MMTV-Cre;Runx1 ;R26Y female compared to those in

+/+ - + 984 MMTV-Cre;Runx1 ;R26Y control female. (E-F) The percentages of lin YFP MEC

- + 985 population (E), as well the ratios of luminal/basal subpopulations among the lin YFP gate

40 L/L 986 (F), are significantly reduced in MMTV-Cre;Runx1 ;R26Y females (n=9) (L/L) compared

+/+ 987 to those in MMTV-Cre;Runx1 ;R26Y control females (n=10) (+/+). P values: *: p0.05;

988 error bars represent Mean ± S.E.M.

989 The following figure supplement is available for figure 3:

990 Figure supplement 1. Conditional knockout study of Runx1 in murine MGs.

991

+ 992 Figure 4. Runx1 disruption leads to a profound reduction in ER MLs. (A) GSEA

993 enrichment plots showing correlation of the expression profiles of Runx1-null or WT

994 luminal MECs with previously published conserved human and mouse signatures of LPs

995 (left) or MLs (right) (Lim et al., 2010). (B) qRT-PCR validation of TF/co-factor genes

996 known to play roles in luminal lineage specification and maintenance. RNA was isolated

+ 997 from YFP Runx1-null and WT primary luminal MECs. (C) FACS plots of expression of

+ 998 CD14 and c-Kit, two LP markers (Asselin-Labat et al., 2011), in the gated YFP luminal

- + lo L/L 999 MECs (Lin CD24 CD29 ) of adult MMTV-Cre;Runx1 ;R26Y virgin females and MMTV-

+/+ - - 1000 Cre;Runx1 ;R26Y control females. Note the CD14 c-Kit mature luminal (ML)

1001 subpopulation was largely lacking in the lower right plot. (D) Quantification of the

1002 percentages of the ML and LP subpopulations as indicated in C showing significant

L/L 1003 reduction in the ML subpopulation in MMTV-Cre;Runx1 ;R26Y females (n=13)

+/+ 1004 compared to that in MMTV-Cre;Runx1 ;R26Y control females (n=10). (E) qRT-PCR

1005 analysis showing significantly reduced Runx1 expression in the LP subpopulation, but not

L/L 1006 in the ML subpopulation in MMTV-Cre;Runx1 ;R26Y females. (F) FACS plots of

+ 1007 expression of CD49b, a LP marker, and Sca1, an ER ML marker (Shehata et al., 2012)

+ - + + 1008 in the gated YFP luminal MEC population. Note the CD49b Sca1 ER ML subpopulation

+ - - 1009 was dramatically reduced, whereas the CD49b Sca1 ER LP subpopulation was

L/L 1010 increased in MMTV-Cre;Runx1 ;R26Y females. (G) Quantification of the percentages of

+ + - 1011 the ER ML, ER LP and ER LP subpopulations as indicated in F showing significant

41 + L/L 1012 reduction in the ER ML subpopulation in MMTV-Cre;Runx1 ;R26Y females (n=4)

+/+ 1013 compared to those in MMTV-Cre;Runx1 ;R26Y control females (n=4). P values: *:

1014 p0.05; #: p0.005; ^: p0.0005; NS=not significant; error bars represent Mean ± S.E.M.

1015 The following source data and figure supplement are available for figure 4:

1016 Source data 1. Gene sets from the MSigDB database C2-CGP (chemical and genetic

1017 perturbations, v3.1) enriched in Runx1-null luminal cells.

1018 Source data 2. Gene sets from the MSigDB database C2-CGP (chemical and genetic

1019 perturbations, v3.1) enriched in Runx1-WT luminal cells.

1020 Source data 3. Gene sets from the MSigDB database C2-CP:KEGG (KEGG gene sets,

1021 v3.1) enriched in Runx1-null luminal cells.

L/L 1022 Figure supplement 1. Analysis of the luminal phenotype in MMTV-Cre;Runx1 ;R26Y

1023 females.

1024

+ 1025 Figure 5. Reduction in ER MLs upon Runx1 disruption can be rescued by Trp53 or

- + - + 1026 Rb1 loss. (A) FACS analysis showing total lin YFP MEC population, lin YFP luminal

- + 1027 population, and lin YFP ML and LP subpopulations (from left to right for each genotype,

1028 an example of the gating strategy is indicated in the bottom left plots), in female mice with

- + 1029 the indicated genotypes. Those highlighted in green show increased lin YFP luminal

1030 populations upon loss of both Runx1 and Trp53 or Rb1 (middle and bottom right plots,

1031 respectively) compared to those of Trp53 or Rb1 loss alone (middle and bottom left plots,

- + 1032 respectively); those highlighted in red show increased lin YFP ML subpopulations upon

1033 loss of both Runx1 and Trp53 or Rb1 (middle and bottom right plots, respectively)

1034 compared to that of Runx1 loss alone (upper right plot). Lu: luminal; Ba: basal; LP:

1035 luminal progenitor; ML: mature luminal cell. (B-C) Quantifications for the percentages of

1036 each indicated subpopulation in A under either the Trp53-loss (B) or Rb1-loss

1037 background (C); in (B) Trp53-loss alone (n=5), Trp53/Runx1-loss (n=3); in (C) Rb1-loss

42 1038 alone (n=5), Rb1/Runx1-loss (n=5). (D) qRT-PCR analysis showing significantly reduced

1039 Runx1 expression in both the YFP-marked ML and LP subpopulations in MMTV-

L/L L/L 1040 Cre;Runx1 ;Rb1 ;R26Y females. P values: *: p0.05; ^: p0.0005; NS=not significant;

1041 error bars represent Mean ± S.E.M.

1042 The following figure supplement is available for figure 5:

+ 1043 Figure supplement 1. Abundant ER MECs are present in both Runx1/Trp53-null and

1044 Runx1/Rb1-null MGs.

1045

+ 1046 Figure 6. RUNX1 controls transcription of select target genes in human ER breast

1047 cancer cells. (A) Western blot showing upregulation of ELF5 and downregulation of ERα

1048 and FOXA1 upon RUNX1 knockdown in T47 luminal breast cancer cells. (B) ChIP

1049 analysis showing significant binding of RUNX1 to multiple ECRs (evolutionary conserved

1050 regions) with RUNX-binding sites in the ELF5 locus in T47D cells. (C) ChIP analysis

1051 showing significant binding of RUNX1 to the -1.6kb and -1.9kb regions of FOXA1 in T47D

1052 cells. RUNX1 binding to the -1.4kb region is marginally significant (p=0.08). (D) ChIP

1053 analysis showing significant binding of RUNX1 to the promoter region (-0.64kb) of

1054 CITED1 in T47D cells. See also Figure 6-figure supplement 1C for RUNX1 binding to the

1055 same region in MCF7 cells. In (B-D), RUNX1-binding motifs (highlighted in red) and their

1056 flanking sequences are shown; note RUNX1-binding motifs in ECR-1 and ECR-3 of ELF5

1057 (B) and the -0.64kb site of CITED1 are in the reverse strand. P values: *: p0.05; #:

1058 p0.005; NS=not significant; error bars represent Mean ± S.E.M.

1059 The following source data and figure supplements are available for figure 6:

1060 Source data 1. Gene sets from Cited1 knockout females.

1061 Figure supplement 1. Relation of RUNX1 to other luminal TFs.

43 1062 Figure supplement 2. GSEA results showing enrichment of genes downregulated in

1063 Cited1 knockout MGs in Runx1-null luminal MECs.

1064

+ 1065 Figure 7. Target genes of RUNX1 in ER luminal MECs revealed by in vivo

1066 expression analysis. (A-B) qRT-PCR analysis showing changes in expression of the

+ + + 1067 indicated genes in sorted YFP ER LPs (A) and ER MLs (B) (based on CD49b and

L/L L/L 1068 Sca1 expression) from 4-5 months old MMTV-Cre;Rb1 ;Runx1 ;R26Y double mutant

L/L 1069 females, compared to those from either MMTV-Cre;Rb1 ;R26Y single mutant females

1070 (left plots) or MMTV-Cre;R26Y WT females (right plots). (C-D) qRT-PCR analysis

- - + + 1071 comparing expression of the indicated genes in CD14 c-Kit MLs (C) and CD14 c-Kit LPs

L/L L/L 1072 (D) from 7-month to 2-month old MMTV-Cre;Rb1 ;Runx1 ;R26Y (Runx1 L/L) and

L/L 1073 MMTV-Cre;Rb1 ;R26Y (Runx1 +/+) females. Expression was normalized to those of the

1074 corresponding 2-month old double or single mutant females, respectively. Error bars:

1075 Mean ± S.E.M.

1076 The following figure supplements are available for figure 7:

1077 Figure supplement 1. Loss of Runx1 in vivo leads to changes in expression of ER

1078 program-related genes.

+ 1079 Figure supplement 2. RUNX1 reduction leads to hyperproliferation of abnormal ER

1080 luminal cells in a context-dependent manner.

1081

+ 1082 Figure 8. Model for the role of RUNX1 in ER mammary luminal cells and luminal

1083 breast cancer. (A) Relative expression levels of key TFs in different subsets of MECs

1084 are indicated [“+++”, “++”, “+”, “+/-”, “-” indicate highest to low to no expression, based on

1085 Figure 1I) and our single cell profiling data for sorted MECs (M.P.A.v.B. and Z.L.,

1086 unpublished data)]. RUNX1 or ELF5 controls the ductal or alveolar luminal cell fate,

+ 1087 respectively, by antagonizing each other. RUNX1 further controls the fate of ER ductal

44 1088 luminal MECs by regulating the ER program via modulating FOXA1 and CITED1

1089 expression. (B) Genetic interaction between loss of RUNX1 and loss of either TP53 or

+ 1090 RB1 plays a key role in the development of RUNX1-mutant ER luminal breast cancer.

1091

1092 Supplementary File 1. Primers and shRNAs used in this study (All sequences from 5’ to 3’)

45 1 Supplemental figure legends

2

3 Figure 1 – Figure supplement 1. Expression analysis of Runx1 and other select

4 luminal transcription factor (TF) genes based on microarray

5 Relative expression values of indicated genes determined by microarray analysis of the

6 indicated mammary epithelial cell (MEC) subpopulations isolated from the mammary

7 glands (MGs) of adult female mice from multiple published data sets. (A) Based on GEO

8 database accession # GSE40875 (only showing the nulliparous subsets). Basal_CD49hi

- lo - 9 and Basal_CD49flo are two subpopulations in the lin CD24 Sca1 basal lineage based on

10 higher or lower CD49f expression, respectively. (B) Based on GEO database accession #

11 GSE19446. (C) Based on GEO database accession # GSE20402.

12

13 Figure 3 – Figure supplement 1. Conditional knockout study of Runx1 in murine

14 MGs

15 (A) Whole-mount carmine staining of inguinal MGs of pubertal (top), adult virgin mice

16 (middle) and lactation day-0 dams (bottom). Scale bars indicate 3mm. Black arrow

17 indicates the main duct coming from the nipple; * indicates the lymph node. Blue arrow

18 indicates the distance between the front of the expanding ductal tree and the lymph node

L/L 19 (used as a reference point). (B) Weaning record showing MMTV-Cre;Runx1 females

20 failed to nurse their pups. (C) Hematoxylin & eosin (H&E) staining of sections of MGs

+/+ L/L 21 from MMTV-Cre;Runx1 (left) and MMTV-Cre;Runx1 (right) dams on lactation day-0.

22 Note the increasing number of cytoplasmic lipid droplets and milk in the lumen of the

L/L 23 MMTV-Cre;Runx1 dam. Arrowheads point to lipid droplets.

24

1 25 Figure 4 – Figure supplement 1. Analysis of the luminal phenotype in MMTV-

L/L 26 Cre;Runx1 ;R26Y females

27 (A) GSEA analysis of Runx1-null and WT luminal MECs showing enrichment of several

28 gene sets related to the p53 signaling pathway in Runx1-null luminal MECs compared to

29 WT luminal MECs. (B) Relative expression values of the indicated genes determined by

30 microarray analysis of Runx1-null and WT luminal MECs. Affymetrix probes used to

31 estimate expression of each indicated gene are 1448886_at, 1419555_at, 1418496_at,

32 1449031_at, and 1460591_at for Gata3, Elf5, Foxa1, Cited1, and Esr1, respectively. (C)

- - + 33 FACS plots of CD14 and c-Kit expression in the lin luminal (upper plots) and lin YFP

+/+ 34 luminal (bottom plots) MECs of 5-week old MMTV-Cre;Runx1 ;R26Y and MMTV-

L/L - - - 35 Cre;Runx1 ;R26Y females showing reduced CD14 c-Kit ML subpopulation within the lin

+ L/L 36 YFP luminal population in MMTV-Cre;Runx1 ;R26Y female (bottom right plot). (D)

37 Quantifications (for C) of the percentages of ML or LP subpopulations within the indicated

- 38 gates showing significant reduced ML and increased LP subpopulations within the lin

+ L/L 39 YFP luminal gate in 5-week old MMTV-Cre;Runx1 ;R26Y females (n=4) (L/L) compared

+/+ 40 to those in 5-week old MMTV-Cre;Runx1 ;R26Y control females (n=4) (+/+). (E) FACS

- lo - 41 plots showing the reduced YFP-marked lin CD29 CD61 ML subpopulation in adult

L/L +/+ 42 MMTV-Cre;Runx1 ;R26Y virgin females (n=2) compared to MMTV-Cre;Runx1 ;R26Y

43 control females (n=3). (F) Quantifications for the basal, LP and ML subpopulations based

44 on CD61 and CD29 staining in E. (G) qRT-PCR analysis showing significantly reduced

- + 45 Runx1 expression in the ER LP subpopulation, partial Runx1 reduction in the ER LP

+ 46 subpopulation, and no reduction in the ER ML subpopulation from MMTV-

L/L 47 Cre;Runx1 ;R26Y females, based on CD49b and Sca1 staining. P values: ^: p≤0.0005;

48 NS=not significant; error bars represent Mean ± S.E.M.

49

2 + 50 Figure 5 – Figure supplement 1. Abundant ER MECs are present in both

51 Runx1/Trp53-null and Runx1/Rb1-null MGs

+ 52 (A-B) IHC staining for ERα showing abundant ER MECs (brown cells) in MMTV-

L/L 53 Cre;Runx1 females upon simultaneous loss of either Trp53 (A) or Rb1 (B).

54

55 Figure 6 – Figure supplement 1. Relation of RUNX1 to other luminal TFs

56 (A) Western blot showing downregulation of RUNX1 and upregulation of ELF5 protein

57 levels upon induced alveolar differentiation in HC11 cells. (B) ChIP analysis showing

58 relatively weak binding of RUNX1 to the -1.4kb region (RUNX1-binding site in the reverse

59 strand) of ESR1 in T47D cells (p=0.1). (C) ChIP analysis showing significant binding of

60 RUNX1 to the CITED1 promoter in MCF7 cells. Genomic location of a highly conserved

61 RUNX1-binding site (in the reverse strand) is shown. P values: *: p≤0.05; NS=not

62 significant; error bars represent Mean ± S.E.M.

63

64 Figure 6 – Figure supplement 2. GSEA results showing enrichment of genes

65 downregulated in Cited1 knockout MGs in Runx1-null luminal MECs.

66 Two gene sets representing genes downregulated in Cited1-null MGs [based on (Howlin

67 et al., 2006; McBryan et al., 2007)] are also downregulated (i.e., negatively enriched) in

68 Runx1-null luminal MECs in relation to WT luminal MECs. FDR q-val ≤ 0.25 is considered

69 as significant (Subramanian et al., 2005).

70

71 Figure 7 – Figure supplement 1. Loss of Runx1 in vivo leads to changes in

72 expression of ER program-related genes.

+ 73 (A) Changes in expression of the indicated genes in the ER LP subpopulation from

L/L 74 MMTV-Cre;Runx1 ;R26Y females (Runx1 L/L) compared to MMTV-Cre;R26Y WT

3 - 75 females (Runx1 +/+). (B) Elf5 expression is de-repressed upon Runx1 loss in all ER

+ - - + + 76 MEC subsets examined, including CD49b Sca1 ER LPs, CD14 c-Kit LPs enriched for

- - 77 ER cells (both sorted from MMTV-Cre;R26Y females) and ER basal MECs (sorted from

78 Krt14-Cre;R26Y females). (C) Compared to A, Esr1 expression is not increased in

+ - - + + - - 79 CD49b Sca1 ER LPs, CD14 c-Kit LPs (enriched for ER cells) and ER basal MECs with

+ + 80 Runx1-loss. Note Esr1 expression in CD14 c-Kit LPs with Runx1-loss is slightly

+ + + 81 elevated, possibly due to the fact that a small portion of CD14 c-Kit LPs are ER cells

82 (Shehata et al., 2012), which have higher Esr1 expression, as shown in A.

83

84 Figure 7 – Figure supplement 2. RUNX1 reduction leads to hyperproliferation of

+ 85 abnormal ER luminal cells in a context-dependent manner.

- + + 86 (A) Quantification of the percentages of the ER LP, ER LP and ER ML subpopulations

L/L L/L 87 sorted from MMTV-Cre;Rb1 ;Runx1 ;R26Y females (n=4) compared to those from

L/L + 88 MMTV-Cre;Rb1 ;R26Y control females (n=4) showing slight upregulation of the ER LP

+ L/L L/L 89 and ER ML subpopulations from MMTV-Cre;Rb1 ;Runx1 ;R26Y females. (B)

90 Knockdown (kd) of RUNX1 in T47D luminal breast cancer cells, which carry a TP53

91 mutation, increased their proliferation. (C-D) Quantification of IHC staining for ERα

+ 92 showing significant increase in the numbers of ER luminal MECs in MMTV-

L/L L/L L/L L/L 93 Cre;Rb1 ;Runx1 ;R26Y (C) and MMTV-Cre;Trp53 ;Runx1 ;R26Y (D) double mutant

94 females compared to their corresponding single mutant control females. (E) qRT-PCR

- + + 95 analysis showing expression of indicated genes in the ER LP, ER LP and ER ML

96 subpopulations from MMTV-Cre;R26Y WT females. Expression was normalized to that of

+ 97 the ER LP subpopulation (=1). P values: *: p≤0.05; #: p≤0.005; ^: p≤0.0005; error bars

98 represent Mean ± S.E.M. (error bars represent Mean ± S.D. in B).

99

100

4 101 References for Supplemental figure legends:

102

103 Howlin, J., McBryan, J., Napoletano, S., Lambe, T., McArdle, E., Shioda, T., and Martin,

104 F. (2006). CITED1 homozygous null mice display aberrant pubertal mammary ductal

105 morphogenesis. Oncogene 25, 1532-1542.

106 McBryan, J., Howlin, J., Kenny, P.A., Shioda, T., and Martin, F. (2007). ERalpha-CITED1

107 co-regulated genes expressed during pubertal mammary gland development: implications

108 for breast cancer prognosis. Oncogene 26, 6406-6419.

109 Shehata, M., Teschendorff, A., Sharp, G., Novcic, N., Russell, A., Avril, S., Prater, M.,

110 Eirew, P., Caldas, C., Watson, C.J., et al. (2012). Phenotypic and functional

111 characterization of the luminal cell hierarchy of the mammary gland. Breast Cancer Res

112 14, R134.

113 Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L., Gillette, M.A.,

114 Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., et al. (2005). Gene set

115 enrichment analysis: a knowledge-based approach for interpreting genome-wide

116 expression profiles. Proc Natl Acad Sci U S A 102, 15545-15550.

117

5 A B Luminal lineage (Lin-CD24+CD29lo) CD14-;c-Kit- Luminal CD61-;CD49b- progenitor (LP; Sca1+; ER+ mixture) Mature luminal cell (Ductal) (ML) CD14+;c-Kit+ CD61-;Wap+; CD61+;CD49b+ ER- ER- or + Alveolar luminal cell (AL)

Basal progenitor Myoepithelial cell

C D E

F G H

I J (LP) (ML) (AL) A MASDSIFESFPSYPQCFMRECILGMNPSRDVHDASTSRRFTPPSTALSPGKMSEALPLGAPDAGAALAGK!

LRSGDRSMVEVLADHPGELVRTDSPNFLCSVLPTHWRCNKTLPIAFKVVALGDVPDGTLVTVMAGNDENY! R166Q G168E R169K! SAELRNATAAMKNQVARFNDLRFVGRSGRGKSFTLTITVFTNPPQVATYHRAIKITVDGPREPRRHRQKL!

DDQTKPGSLSFSERLSELEQLRRTAMRVSPHHPAPTPNPRASLNHSTAFNPQPQSQMQDTRQIQPSPPWS!

YDQSYQYLGSIASPSVHPATPISPGRASGMTTLSAELSSRLSTAPDLTAFSDPRQFPALPSISDPRMHYP!

GAFTYSPTPVTSGIGIGMSAMGSATRYHTYLPPPYPGSSQAQGGPFQASSPSYHLYYGASAGSYQFSMVG!

GERSPPRILPPCTNASTGSALLNPSLPNQSDVVEAEGSHSNSPTNMAPSARLEEAVWRPY!

B

RUNX1 Additional Additional Affected Mutation Mutation Mutation missense missense missense amino disrupts perturbs disrupts mutations mutations mutations acid DNA contact! Runt CBFβ based on based on based on residue domain contact! Ellis et al! Taniuchi TCGA, based on fold! ! et al! Nature 2012! Li et al! R166Q! R139! Yes! G168E! G141! Yes! R169K! R142! Yes! S73C! S73! Yes! R174Q! R174! Yes! G108D! G108! Yes! N112S! N112! Yes! L134P! L134 (not Close to a studied)! critical DNA contact residue R135! D171G! D171! Yes! A B C YFP-

YFP+

D MMTV-Cre;Runx1+/+;R26Y lin- Lu 24.4% 65.5% E F Ba 25 * 3 *

YFP 28.3% 20 2.5 CD24

FSC CD29 15 2

MMTV-Cre;Runx1L/L;R26Y cells YFP+ 1.5 10 lin- Lu 14.5% 1 49.7% 5 cells Lin-YFP+ Ba 0.5 % of Lin- YFP 43% 0 in luminal/basal ofRatio 0

CD24 Runx1 Runx1 +/+ L/L Runx1 Runx1 FSC CD29 +/+ L/L A B Runx1 +/+ Runx1 L/L

C +/+ D MMTV-Cre;Runx1 ;R26Y 100 Runx1 +/+ lin-YFP+ lin-YFP+ luminal ^ 80 Runx1 L/L 60 ^ 40 CD14 CD24 20 indicated gate gate indicated % of cells in the in % of cells 0 CD29 c-Kit ML LP

MMTV-Cre;Runx1L/L;R26Y E Runx1 +/+ lin-YFP+ lin-YFP+ luminal Runx1 L/L CD14 CD24

CD29 c-Kit F G 80 Runx1 +/+ # MMTV-Cre; MMTV-Cre; +/+ L/L Runx1 L/L Runx1 ;R26Y Runx1 ;R26Y 60 lin-YFP+ luminal lin-YFP+ luminal # ER+ 11.5% ER+ 40 ML ER+ 14.4% ML ER+ LP NS LP 34.2% 20

7.09% gate indicated ER- ER- in the % of cells Sca1 LP Sca1 LP 0 39.0% 71.6% ER+ ML ER+ LP ER- LP CD49b CD49b A MMTV-Cre;Runx1+/+;R26Y MMTV-Cre;Runx1L/L;R26Y

41.9% 85.6% 27.4% 67.1% 14.5% 49.7% Lu LP Lu LP Ba Ba 22.8% 43% ML ML 44.7% 6.2%

MMTV-Cre;Runx1+/+;Trp53L/L;R26Y MMTV-Cre;Runx1L/L;Trp53L/L;R26Y

4.6% 21.3% 10.9% 20% 76.6% Lu LP 86.3% Lu LP Ba Ba 42.5% 26% ML ML 37.2% 50.8%

MMTV-Cre;Runx1+/+;Rb1L/L;R26Y MMTV-Cre;Runx1L/L;Rb1L/L;R26Y lin- 10.1% 32.5% 27.9% 27.4% 77.6% Lu LP 83.9% Lu LP Ba Ba 21.3% 9.7% ML ML 38.9% 37.1% CD14 CD24 YFP FSC CD29 c-Kit B Under MMTV-Cre;Trp53L/L;R26Y C Under MMTV-Cre;Rb1L/L;R26Y D Under MMTV-Cre;Rb1L/L;R26Y Runx1 +/+ 70 Runx1 +/+ 60 Runx1 +/+ NS Runx1 L/L * Runx1 L/L NS Runx1 L/L 60 50 1.6 NS NS 1.4 * ^ 50 40 40 1.2 NS 30 * 1 30 Runx1 0.8 20 20 * 0.6 10 10 expression 0.4 indicated gate indicated gate Relative % ofcells in the 0 % ofcells in the 0 0.2 Luminal Basal ML LP Luminal Basal ML LP 0 lin-YFP+ lin-YFP+ Luminal lin-YFP+ lin-YFP+Luminal ML LP A B Control RUNX1 kd ELF5 RUNX1 -20kb -10kb -1kb +1 3’UTR ELF5 -17.4kb ECR-1 -11.3kb ECR-2 ECR-3 GCAAGAAACACATTCCT AGAAATGTGTCAGACAG CTGGAAAACACATGAAT ERα 50 # IgG 40 RUNX1 FOXA1 30 # 20 * CITED1 10

β-ACTIN Fold enrichment 0 ECR-1 ECR-2 ECR-3 non-specific region C FOXA1 D CITED1 -3kb -2kb -1kb +1 -1kb +1 +1kb +2kb -1.6kb -1.9kb -1.4kb -0.64kb AATTGTGTGTCAGAAAC TGGATGACCACAGAACA AGAAATGTGTTTTAACT CTGTGTGTGTTTCTGAGTGTCTGTGACCATGTGTTTGTTGT 7 # NS IgG 6 6 IgG RUNX1 5 # * 5 RUNX1 4 4 3 3 2 2

Fold enrichment 1 1

Fold enrichment 0 0 -0.64kb non-specific region -1.4kb -1.6kb -1.9kb non-specific region A ER+ LPs (CD49b+Sca1+ ) MMTV-Cre;Rb1L/L 16 14 MMTV-Cre (WT) 14 MMTV-Cre;Rb1L/L; 12 12 Runx1L/L MMTV-Cre;Rb1L/L; 10 Runx1L/L 10 8 8 6 6 4 Relative epression Relative 4 Relative Expression Relative 2 2 0 0 Runx1 Elf5 Esr1 Foxa1 Cited1 Runx1 Elf5 Esr1 Foxa1 Cited1

B ER+ MLs (CD49b-Sca1+ ) MMTV-Cre;Rb1L/L MMTV-Cre (WT) 6 4 MMTV-Cre;Rb1L/L; 3.5 MMTV-Cre;Rb1L/L; 5 Runx1L/L Runx1L/L 3 4 2.5 3 2 2 1.5 1 Relative expression Relative Relative expression Relative 1 0.5 0 0 Runx1 Elf5 Esr1 Foxa1 Cited1 Runx1 Elf5 Esr1 Foxa1 Cited1

C Runx1 Esr1 Foxa1 Cited1 Elf5 Runx1 +/+ ML Runx1 +/+ ML 3 Runx1 +/+ ML 3 Runx1 +/+ ML 4 4.5 3 Runx1 +/+ ML Runx1 L/L ML 4 Runx1 L/L ML Runx1 L/L ML Runx1 L/L ML Runx1 L/L ML 3 3.5 2 2 3 2 2.5 2 2 1 1 1.5 1 1 1 0.5

0 0 expression Relative 0 expression Relative 0 0 Relative expression Relative Relative expression Relative 2 months 7 months expression Relative 2 months 7 months 2 months 7 months 2 months 7 months 2 months 7 months D Runx1 Esr1 Foxa1 Cited1 Elf5 4 Runx1 +/+ LP 6 Runx1 +/+ LP 3 Runx1 +/+ LP 4 4 Runx1 +/+ LP Runx1 +/+ LP Runx1 L/L LP Runx1 L/L LP Runx1 L/L LP Runx1 L/L LP 5 Runx1 L/L LP 3 3 3 2 4 2 2 2 3 1 2 1 1 1 1 Relative expression Relative Relative expression Relative 0 expression Relative 0 0 0 0 Relative expression Relative 2 months 7 months 2 months 7 months 2 months 7 months expression Relative 2 months 7 months 2 months 7 months A Elf5 - Repress Repress Elf5 ++ Runx1 ++ Runx1 + Foxa1 + Fox1 +/- Cited1 + Cited1 +/- Esr1 + Activate Esr1 +/- Mature ductal Repress luminal cell Elf5 - (ML, ER+) Elf5 +++ Runx1 +++ Luminal Runx1 - progenitor Alveolar Esr1 - (LP; mixture, luminal Foxa1 - Repress ER+ or -) Mammary cell (AL, Cited1- stem cell ER-) Repress (MaSC, ER-) Elf5 - Basal progenitor Runx1 +++ (putative), ER- Myoepithelial cell, ER- B ER+ luminal Cell cycle Reduction Loss of cell fate Activation of arrest; in ER+ RUNX1 impaired; p53 pathway cellular stress apoptosis luminal MECs Relief of cell Loss of cycle arrest; TP53 rescue from Rescue and apoptosis hyperproliferation Cell cycle of ER+ luminal Loss of activation; MECs RB1 rescue from Additional apoptosis mutations

Luminal breast cancer