Involvement of a Transcription Factor, Nfe2, in Breast Cancer Metastasis to Bone

cancers

Article Involvement of a Transcription factor, Nfe2, in Breast Cancer Metastasis to Bone

Di Zhang 1, Sadahiro Iwabuchi 2 , Tomohisa Baba 1, Shin-ichi Hashimoto 2, Naofumi Mukaida 1 and So-ichiro Sasaki 1,*

1 Division of Molecular Bioregulation, Cancer Research Institute, Kanazawa University, Kakuma-Machi, Kanazawa, Ishikawa 920-1192, Japan; [email protected] (D.Z.); sergenti@staff.kanazawa-u.ac.jp (T.B.); mukaida@staff.kanazawa-u.ac.jp (N.M.) 2 Department of Molecular Pathophysiology, Institute of Advanced Medicine, Wakayama Medical University, 811-1 Kimiidera, Wakayama 641-8509, Japan; [email protected] (S.I.); [email protected] (S.-i.H.) * Correspondence: s_sasaki@staff.kanazawa-u.ac.jp; Tel.: +81-76-674-6736

Received: 24 September 2020; Accepted: 13 October 2020; Published: 16 October 2020

Simple Summary: Breast cancer patients are frequently complicated by bone metastasis, which deteriorates their life expectancy. Bone metastasis is treated with the drugs targeting osteoclast activation, which is mostly observed at the metastasis sites. However, these drugs cannot directly inhibit cancer cell growth and therefore, a novel therapeutic strategy is required to impede cancer cell proliferation at bone metastasis sites. Here, we proved that a transcription factor, NFE2, was expressed selectively in breast cancer cells at bone metastasis sites and contributed crucially to their enhanced proliferation therein, by activating Wnt pathway. Thus, NFE2 can be a novel molecular target to treat breast cancer bone metastasis.

Abstract: Patients with triple negative breast cancer (TNBC) is frequently complicated by bone metastasis, which deteriorates the life expectancy of this patient cohort. In order to develop a novel type of therapy for bone metastasis, we established 4T1.3 clone with a high capacity to metastasize to bone after orthotopic injection, from a murine TNBC cell line, 4T1.0. To elucidate the molecular mechanism underlying a high growth ability of 4T1.3 in a bone cavity, we searched for a novel candidate molecule with a focus on a transcription factor whose expression was selectively enhanced in a bone cavity. Comprehensive gene expression analysis detected enhanced Nfe2 mRNA expression in 4T1.3 grown in a bone cavity, compared with in vitro culture conditions. Moreover, Nfe2 gene transduction into 4T1.0 cells enhanced their capability to form intraosseous tumors. Moreover, Nfe2 shRNA treatment reduced tumor formation arising from intraosseous injection of 4T1.3 clone as well as another mouse TNBC-derived TS/A.3 clone with an augmented intraosseous tumor formation ability. Furthermore, NFE2 expression was associated with in vitro growth advantages of these TNBC cell lines under hypoxic condition, which mimics the bone microenvironment, as well as Wnt pathway activation. These observations suggest that NFE2 can potentially contribute to breast cancer cell survival in the bone microenvironment.

Keywords: bone metastasis; breast cancer; transcription factor; Wnt pathway

1. Introduction Breast cancer is the most prevalent cancer in females with 280,000 new cases annually in the United States alone [1]. Although prognosis has been progressively improved in advanced countries due to early diagnosis and advanced treatment [2], more than 40,000 breast cancer patients succumb annually

Cancers 2020, 12, 3003; doi:10.3390/cancers12103003 www.mdpi.com/journal/cancers Cancers 2020, 12, 3003 2 of 18 in the United States [1], mostly due to metastasis. A meta-analysis study revealed that a median of 12% of patients with stage I–III breast cancer developed bone metastases during a median follow-up of 60 months and that more than 50% patients with metastatic breast cancer had bone metastasis [3]. At bone metastasis sites, cancer cells are presumed to interact with the bone microenvironment to cause bone destruction, where osteoclast activation has a crucial role [4]. Thus, bone metastasis is frequently treated with bisphosphonates and/or denosumab, a monoclonal antibody against the receptor activator of NF-κB (RANK) ligand, to inhibit osteoclast activation, thereby controlling bone metastasis [5]. These drugs have disease-modifying effects but are mostly palliative due to their indirect effects on intraosseous cancer cells [5]. It is, therefore, necessary to develop a novel type of agents, which can inhibit the growth of cancer cells in a bone cavity, based on the understanding the molecular and cellular mechanisms underlying bone metastasis. With the interaction with a myriad of host resident cells, cancer cells metastasize to distant organs through multiple steps: growth at the primary site, intravasation from the primary tumor site, dissemination through systemic circulation, extravasation into the metastatic organ, and growth at the metastatic organ [6]. Moreover, in order to form a metastatic focus, tumor cells should adapt to the microenvironment of a metastatic organ, which exhibits different cell components including immunocompetent cells from the original organ [7]. Intra-cardiac injection of breast cancer cell lines [8] or injection of human breast cancer cell lines into the mammary fat pads of immune-deficient mice [9] was employed as a breast cancer bone metastasis model but could not reproduce the whole steps of bone metastasis. Thus, using a murine triple-negative breast cancer (TNBC) cell line, 4T1.0, we previously established 4T1.3 clone which could metastasize spontaneously to a bone cavity after its orthotropic injection into mammary fat pads of immunocompetent mice [10]. A subsequent analysis revealed that 4T1.3 clone exhibited constitutively enhanced expression of a chemokine, Ccl4, and the enhanced capacity of 4T1.3 clone to metastasize to bone was due to neither accelerated growth at its primary site nor augmented migration to bone, but was ascribed to its higher ability to grow in the bone microenvironment [10]. RNAseq was conducted to elucidate the molecular mechanisms underlying the enhanced capability of 4T1.3 clone to grow in a bone cavity and, as a consequence, revealed the changes in the expression of a wide variety of genes in intraosseously growing 4T1.3 clone, even when compared with intraosseously growing 4T1.0. Inspired by the potential involvement of several transcription factors such as Runx-2 [11] and Bach1 in breast cancer bone metastasis [12], we searched for an additional transcription factor(s), which is associated with augmented proliferation of 4T1.3 clone in a bone cavity. We finally provided evidence to indicate that Nfe2, a transcription factor involved in normal and malignant hematopoiesis [13–17], could provide breast cancer cells with an advantage to grow efficiently in a bone cavity, thereby accelerating breast cancer bone metastasis.

2. Results

2.1. Transcription Factor Expression in 4T1.3 Grown in a Bone Cavity We previously observed that 4T1.3 cells exhibited a higher growth capacity compared with 4T1.0 cells [10]. Consistently, intraosseous injection of 4T1.3 cells formed a larger tumor mass with enhanced cell proliferation, as evidenced by increased Ki-67-positive cell numbers, compared with that of 4T1.0 cells (Figure1a,b). Moreover, intraosseous injection of 4T1.0 and 4T1.3 cells gave rise to similar CD31-positive areas and VEGF expression in intraosseous tumor sites (Figure1c,d). Intraosseous injection of either 4T1.0 or 4T1.3 increased intraosseous CD51-positive osteoclast numbers to similar extents, compared with untreated mice (Figure1e). Consistently, the injection of either clone resulted in similar mRNA expression levels of Rank, Rank ligand (Rankl), and osteoprotegerin (Opg) (Figure1f), which are presumed to be involved in osteoclastogenesis. These observations favor the notion that a higher intraosseous growth capacity of 4T1.3 cells can be ascribed to their intrinsic enhanced growth capacity but neither to augmented neovascularization nor to exaggerated osteoclast activation. Hence, Cancers 2020, 12, 3003 3 of 18 Cancers 2020, 12, x FOR PEER REVIEW 3 of 18 intraosseouswe focused growth on a transcription ability on 4T1.3 factor(s) cells, which and searched can confer for intraosseous a candidate growthtranscription ability factor(s) on 4T1.3 that cells, fulfilledand searched all three for conditions a candidate (Figure transcription 2a) and narrowed factor(s) that down fulﬁlled to 13 transcription all three conditions factor genes (Figure (Figure2a) and 2bnarrowed). A subsequent down to 13 qRT transcription-PCR analysis factor detected genes (Figure significantly2b). A subsequent augmented qRT-PCR expression analysis of detected three transcriptionsigniﬁcantly factors, augmented Lmo2, expression Myb and Nfe2 of three, in 4T1.3 transcription grown in factors, a boneLmo2 cavity, Myb thanand in thatNfe2 ,under in 4T1.3 in vit grownro culturein a bone conditions, cavity than among in that under thesein vitro 13 culture transcription conditions, factors among (Figure these 13 2c). transcription Consistently, factors immunohistochemical(Figure2c). Consistently, (IHC) immunohistochemical analysis detected(IHC) these analysis three transcriptiondetected these factor three transcription proteins in factor the intraosseousproteins in thetumor intraosseous foci arising tumor from fociintraosseous arising from injection intraosseous of 4T1.3 injection clone (Figure of 4T1.3 2d). clone (Figure2d). Ki-67

(a) (b) 2 4T1.0 4T1.3

0.0005t

2.5 3 ) 3

r 0.0004

(

7 3 2.5 3

H 2.0 6

n - 0.0003

2 2 K

s 2.0

f s 1.5

e e o 0.0002

r r 2 2

p p

s 1.5 e

e 1.0 e e

e e

0.0001 1 1

p p

u v 1.0 v

e e

0.5 t t t 1 1

0.0000 e

a a

l l

v v

e e

0.5 4T1.0 4T1.3t (c) t (d) (e) t 0(f)

0.0 0 a

R R

a a

l l l CD51

e VEGF e

CD31 t

t Rant k Rankl Opg Rank Rankl Opg Rank Rankl Opg

r r 0 0.0 0 R (%) R R (%)

(%) p

m m 4T1.0-Control2.5 4T1.0-Nfe2Rank Rankl Op4T1.3-sh-Scrg 3 44T1.3-sh-Nfe2 4T1.0-in3 BM 4T1.3-inRank BMRankl Opg

μ Rank Rankl Opg

8 μ

H H 2.0

a t 4T1.0-in BM 4T1.3-in BM

t a 4T1.0-Control 4T1.0-Nfe2 4T1.3-sh-Scr 4T1.3-sh-Nfe2

e e 2.0

r r 3

o 6 a

a o 1.5

)

) ) 2 2

s s 1.5

(

e e

5 3 4 2

1.0 p

p p

D D 1.0

e e

C C 1 1

v v v 1

2 o o 0.5

0.5 t

t t

a a

l l

e e

a a 0 0.0 0 0 R 0 R R

0.0 R

R 4T1.0 4T1.3Rank Rankl Op4T1.0g 4T1.3 Rank Rdankl .0 Op.g3 Rank Rankl Opg te 1 1 a T T 4T1.0-Control 4T1.0-Nfe2 4T1.3-sh-Scre 4 4T1.3-sh-Nfe24 4T1.0-in BM 4T1.3-in BM tr n u FigureFigure 1. 1.(a)( aH&E) H&E staining staining of of4T1.0 4T1.0 or or4T1.3 4T1.3 tumor tumor foci foci in ina bone a bone cavity. cavity. Tibiae Tibiae were were obtained obtained from from micemice 7 days 7 days after after intraosseous intraosseous injection injection of 4T1.0 of 4T1.0or 4T1.3 or cells 4T1.3 for cells H&E for staining. H&E staining. Representative Representative images fromimages 5 independent from 5 independent mice are mice shown are shown here. Scar here. bars, Scar bars,200 μm. 200 µ (bm.) Enhanced (b) Enhanced cell cell proliferation proliferation at at intraosseousintraosseous tumor tumor sites. sites. Tibiae Tibiae were were obtained obtained from from mice mice 7 7 days days after after intraosseous intraosseous injection ofof 4T1.04T1.0 or or4T1.3 4T1.3 cellscells andand werewere subjectedsubjected to to anti-Ki67-antibody anti-Ki67-antibody staining. staining. All All values values represent represent mean mean+ +SD SD (n(n= = 3). 3).* *p p< <0.05. 0.05. ( c(,cd,d)) Quantitative Quantitative analysis analysis of of CD31- CD31 and- and VEGF-positive VEGF-positive areas areas in in 4T1.0 4T1.0 or or 4T1.3 4T1.3 tumor tumor foci fociin in a bonea bone cavity. cavity. Tibiae Tibiae were were obtained obtained from from mice mice 7 days 7 days after after intraosseous intraosseous injection injection of 4T1.0 of 4T1.0 or 4T1.3 or 4T1.3cells cells and wereand were subjected subjected to anti-CD31 to anti-CD31 (c) or anti-VEGF (c) or anti- (VEGFd) antibody (d) antibody staining. staining. All values All represent values representmean + meanSD (n += SD3 to(n 5).= 3 (toe) 5). Quantitative (e) Quantitative analysis analysis of CD51-positive of CD51-positive areas areas in tumor-bearing in tumor-bearing bones. bones.Tibiae Tibiae were were obtained obtained from from mice mice untreated untreated or 7 or days 7 days after after intraosseous intraosseous injection injection of 4T1.0 of 4T1.0 or 4T1.3 or 4T1.3cells cell ands and were were subjected subjected to to anti-CD51 anti-CD51 antibody antibody staining. staining. AllAll values represent mean mean ++ SDSD (n ( n= 4).= 4). * p*

Cancers 2020, 12, 3003 4 of 18 Cancers 2020, 12, x FOR PEER REVIEW 4 of 18

(a) ① Expression pattern (b) ① 1811 4T1.0 or 4T1.3 in vitro in BM genes FACS Aria 4T1.0 RNA-Seq (Parental) Tumor cells 4T1.3 - + ② 656 ③ 1206 (CD45 CD326 ) (bone-tropic) ◎ genes genes

② ③ Breast cancer patients

Metastatic Metastatic Ccl4 tissues in tissues in +1 Bone Lung App Ar Ets1 Gata1 Gata2 -2000 Lmo2 Max Myb Nf1 Nfe2 Transcription factor binding site Pbx1 Rora Sf1

r 4T1.3 in a bone cavity

p 4T1.3 - in BM

H 1000

t 800

n 600

s 400

r 200

x 10

t 3

e 0 R r 1 1 2 2 x 1 2 1 a 1 p s a a o a b f e x r f p A t t t y f b o S A E a a m M M N G G L N P R

FigureFigure 2. (a )2. Schematic (a) Schematic representation representation ofof thethe procedures toto identify identify candidate candidate Transcription Transcription factors. factors. (b) A( totalb) Aof total 13 transcription of 13 transcription factors factors fulfilled fulfilled 3 conditions 3 conditions described described in Figure in Figure2a. ( c 2a.) mRNA (c) mRNA expression level ofexpression 13 extracted level transcriptionof 13 extracted factorstranscription in 4T1.3 factors cells, in which4T1.3 cells, grew which in a bonegrew marrowin a bone cavitymarrow or were cavity or were cultured in vitro. Total RNAs were extracted from 4T1.3 cells, either isolated from the cultured in vitro. Total RNAs were extracted from 4T1.3 cells, either isolated from the bone marrow bone marrow cavity at day 7 after intraosseous injection, or cultured in vitro. The expression levels of cavity at day 7 after intraosseous injection, or cultured in vitro. The expression levels of 13 transcription 13 transcription factors were determined by qRT-PCR. The mean and SD values were calculated from factors3 animals were determined or 3 independentl by qRT-PCR.y in vitro cultured The mean cells. and (d) Immunohistochemical SD values were calculated analysis fromof 4T1.3 3 animalscells or 3 independentlyin a bone cavity.in vitroTibiaecultured were obtained cells. from (d) mice Immunohistochemical 7 days after intraosseous analysis injection of of 4T1.3 4T1.3 cellscells and in a bone cavity.were Tibiae subjected were to obtained IHC staining from by mice using 7 anti days-pan after cytokeratin, intraosseous anti-LMO2, injection anti-MYB, of 4T1.3 or anti cells-NFE2 and were subjectedantibodies. to IHC The staining tumor by foci using were anti-pan indicated cytokeratin, as the positive anti-LMO2, areas of anti-MYB, pan cytokeratin, or anti-NFE2 which were antibodies. The tumordemarcated foci were by dot indicated lines. Representative as the positive images areas from of pan 5 independent cytokeratin, mice which are were shown demarcated here. Insets by dot lines.indicate Representative enlarged imagesarea of tumor from foci. 5 independent Scar bars, 100 mice μm arein main shown images, here. 50 Insets μm in indicateinsets. enlarged area of tumor foci. Scar bars, 100 µm in main images, 50 µm in insets. 2.2. Enhanced Intraosseous Growth by Nfe2-Expressing Breast Cancer Cells 2.2. EnhancedWe next Intraosseous examined Growth the effects by Nfe2-Expressing of the gene transduction Breast Cancer of Lmo2, Cells Myb, or Nfe2 on intraosseous Wegrowth next of examined parental 4T1.0 the cells, effects which of thegrow gene less e transductionfficiently in a bone of Lmo2 cavity, Mybcompared, or Nfe2 withon 4T1.3 intraosseous [10]. The transfection with individual transcription factors resulted in the establishment of clones, which growth of parental 4T1.0 cells, which grow less efficiently in a bone cavity compared with 4T1.3 [10]. exhibited enhanced expression of the corresponding transcription factors as evidenced by qRT-PCR The transfection(Figure 3a). When with these individual clones were transcription injected directly factors into resulteda bone cavity, in theonly establishment the Nfe2-expressing ofclones, whichclone exhibited formed enhancedsignificantly expression larger tumor of foci the compared corresponding with a control transcription vector-transfected factors asone evidenced (Figure by qRT-PCR3b) together (Figure with3a). enhanced When these NFE2 clones protein wereexpression injected at tumor directly sites (Figureinto a 3c).bone However, cavity, Nfe2 only the Nfe2-expressingtransfection clone failed formed to change significantly in vitro largermRNA tumor express fociion compared of Rank, Ranklwith, a orcontrol Opg vector-transfected(Figure 3d). one (FigureConsistently,3b) together NFE2 proteins with enhanced were detected NFE2 at proteintumor sites expression arising from at tumorintraosseous sites (Figureinjection3 c).of 4T1.3 However, Nfe2 transfectioncells but not th failedose of to4T1.0 change cells (Figurein vitro 3e)mRNA. On the expressioncontrary, no differences of Rank, Rankl were ,observed or Opg on(Figure the 3d). growth rates between Nfe2-expressing and control 4T1.0 clones when they were injected into mouse Consistently, NFE2 proteins were detected at tumor sites arising from intraosseous injection of 4T1.3 mammary fat pads (Figure 3f). Moreover, Nfe2 mRNA was expressed to a larger extent in 4T1.3 cells butobtained not those from of intraosseous 4T1.0 cells tumor (Figure sites3e). but On not the mammary contrary, no fat di padfferences sites (Figure were observed 3g). These on the growthobservations rates between wouldNfe2 indicate-expressing that NFE2 and could control confer 4T1.0 an advantage clones whento grow they only were in a bone injected cavity into but mouse mammarynot at fatthe padsprimary (Figure sites,3 mammaryf). Moreover, fat pads.Nfe2 In mRNA order to was evaluate expressed the clinical to a relevance larger extent of Nfe2 in 4T1.3 obtainedexpression, from intraosseous we evaluated tumor the data sites on overall but not survival mammary of breast fat padcancer sites patients. (Figure Patients3g). These with a observations higher would indicate that NFE2 could confer an advantage to grow only in a bone cavity but not at the primary sites, mammary fat pads. In order to evaluate the clinical relevance of Nfe2 expression, we evaluated the data on overall survival of breast cancer patients. Patients with a higher expression of Nfe2 exhibited an unfavorable prognosis (Figure3h), suggesting the clinical significance of Nfe2 expression among breast cancer patients. Cancers 2020, 12, x FOR PEER REVIEW 5 of 18

Cancers 2020,expression12, 3003 of Nfe2 exhibited an unfavorable prognosis (Figure 3h), suggesting the clinical significance 5 of 18 of Nfe2 expression among breast cancer patients.

r p 18

t H 4T1.0-Control (%) (%) (%) (%) (%) (%) (%) (%) (%) r (b)

y (a) y

t ** t

i 15 18 4T1.0-Lmo240 40 40 40 i 40 40 40 40 40

4T1.0-Control v

o ** o

c ** c

t 4T1.0-Myb c

12 s 15 n.s.4T1.0-Lmo2n.s. n.s. n.s. n.s. n.s.

s n ** ** **

M e 30 30 30 30 30 o 4T1.0-Nfe2 30 30 30 30

** t

B 4T1.0-Myb B

B 9 r

a a 3

a 2.5 a 3

e 4T1.0-Nfe2

** H

e 6

r r 20 20

r 20 20 r

r 20

20 r 20 e 20 20

)

p t ** 2.0

n 6 n 3

+ + 3

+ 2.5 +

(

e 10 10 10 10 10 10 10 o 10 10 2

o o o 2 o

o o

t t

t 0 t

3 s

R s 1.5

a 2.0

e e

Lmo2 Myb Nfe2 C

p p 2 p 2

0 s

R 0 0 0 0 0 0 0 0 0 x

s s 1.5 1.0

e e

l r l 1 Lmo2 Myb r Nfe2 2 l 2 2 l b1l bl b l 2 l 2l 2

e e o o o o o o o o o o o o p e e e p r r p r r yr yr y r f r fr f

v t t t v t t t t t x t

m m m i

i 1.0 i n n n n Mn Mn M n N n N

e n N

t e 0.5 e - - t L L L t - - - - o - o -o 1 - o 0 o 0o 0 o 1 o o 0 0 0 a

a . a . . e . . 0 e 0 e C C 0C C C C C C. . . C l l - - .- - 1- l 1- 1 - 1 - -1 1

v 1 1 v 1 0 0 0 0 0 e

e 0 e 0 0 T T T i 0

i i . T . . T . . . . T. .T T 0.5 T 4 4 t t 1 4 1 1 t 4 1 1 41 1 41 14 4 0 0.0 4 0 R

R R

a a T T T a T T T T T T

l l (d) 4 4 4 l (e)4 4 4 4 4 4

e (c) e e Rank Rankl Opg Rank Rankl 0 Opg Rank Rankl Opg 0.0 0 R

R R Rank Rankl 4T1.0-ControlOpg 4T1.0-Nfe2Rank Rankl 4T1.3-sh-ScrOpg 4T1.3-sh-Nfe2Rank Rankl 4T1.0-inOpg BM 4T1.3-in BM

4T1.0-inp BM 4T1.3-in BM

p 4T1.0-Controlp 2.5 4T1.0-Nfe2 4T1.3-sh-Scr3 4T1.3-sh-Nfe2 3

2.0

o o 4T1.0

i 4T1.0-Nfe2 i 2 2

s s 1.5

e e

p p

x x 1.0

e e

1 1

v v

0.5 t

t t

a a

l l

e e e 0 0.0 0 R 4T1.0-Control R 4T1.3R Rank Rankl Opg Rank Rankl Opg Rank Rankl Opg 4T1.0-Control 4T1.0-Nfe2 4T1.3-sh-Scr 4T1.3-sh-Nfe2 4T1.0-in BM 4T1.3-in BM (f) (g) (h)

) 900 3 4T1.0-Control m 750

( 4T1.0-Nfe2 e 600

u l 450

r 300

u 150

T 0 0 2 4 6 8 10 12 14 16 18 20 Days Figure 3. (Figurea) Expression 3. (a) Expression levels oflevels transcription of transcription factor factor mRNA mRNA expression expression in inthe the used used transfectants. transfectants. 4T1.0 cells were transfected4T1.0 cells were with transfected each with transcription each transcription factor-expressing factor-expressing or or controlcontrol vector vector and andtotal RNAs total RNAs were were extracted from the resultant transfectants. The mean and SD values were calculated from 3 extracted fromindepen thedent resultant experiments transfectants.. ** p < 0.01. The (b) Intraosseous mean and tumor SD values formation were of calculatedLmo2-, Myb-, from or Nfe2 3 independent- experiments.overexpressing ** p < 0.01. 4T1.0 (b cell) Intraosseous lines. Lmo2-, Myb tumor- or Nfe2 formation-expressing ofor Lmo2control-, vectorMyb--,transfected or Nfe2 -overexpressing4T1.0 4T1.0 cell lines.cells (5.0Lmo2 × 10-,3 cells)Myb were- or Nfe2injected-expressing into tibiae of or mice. control Seven vector-transfected days after the inoculation, 4T1.0 tibiae cells were (5.0 103 cells) removed and subjected to anti-pan cytokeratin antibody staining to detect tumor foci formation. All × were injected into tibiae of mice. Seven days after the inoculation, tibiae were removed and subjected to values represent mean + SD (n = 5 to 9). ** p < 0.01, n.s., not significant. (c) Immunohistochemical anti-pan cytokeratinanalysis of Nfe2 antibody-expressing staining or control to detect vector-transfected tumor foci 4T1.0 formation. cells in a bone All values cavity. Tibiae represent were mean + SD (n = 5 to 9).obtained ** p < from0.01, mice n.s., 7 not days signiﬁcant. after intraosseous (c) Immunohistochemical injection of Nfe2-expressing analysis or control of Nfe2 vector-expressing- or control vector-transfectedtransfected 4T1.0 cells 4T1.0 and were cells subjected in a bone to anti cavity.-pan cytokeratin Tibiae wereor anti- obtainedNFE2 antibodies from staining. mice 7 days after Pan-cytokeratin-positive tumor areas are demarcated by dot lines. Representative images from 5 intraosseous injection of Nfe2-expressing or control vector-transfected 4T1.0 cells and were subjected independent mice are shown here. Insets indicate enlarged area of tumor foci. Scar bars, 100 μm in to anti-panmain cytokeratin images, 50 μm or in anti-NFE2 insets. (d) Determin antibodiesation of staining. mRNA expression Pan-cytokeratin-positive level of m Rank, Rankl, or tumor Opg areas are demarcatedin byNfe2 dot-expressing lines. Representative or control vector- imagestransfected from 4T1.0 5 independentcells. Total RNAs mice were are extracted shown either here. from Insets indicate enlarged areaNfe2-expressing of tumor or foci. control Scar vector bars,-transfected 100 µm 4T1.0 in cells main cultured images, in vitro 50. µAllm values in insets. represent (d )me Determinationan + SD (n = 3). (e) Immunohistochemical analysis of 4T1.0 or 4T1.3 cells in a bone cavity. Tibiae were of mRNAobtained expression from levelmice 7 ofdays m afterRank intraosseous, Rankl, or injectionOpg in ofNfe2 4T1.0-expressing or 4T1.3 cells and or controlwere subjected vector-transfected to 4T1.0 cells.anti Total-pan RNAs cytokeratin were or extracted anti-NFE2 eitherantibodies from stainiNfe2ng. -expressing Pan-cytokeratin or-positive control tumor vector-transfected areas are 4T1.0 cells cultureddemarcatedin vitro by. dot All lines. values Representative represent images mean from+ SD 5 independent (n = 3). ( micee) Immunohistochemical are shown here. Insets analysis of 4T1.0 orindicate 4T1.3 enlarged cells in area a boneof tumor cavity. foci. Scar Tibiae bars, 100 were μm obtainedin main images, from 50mice μm in 7insets. days (f) after Tumor intraosseous growth rates of Nfe2-expressing or control vector-transfected 4T1.0 cells upon their injection into injection ofmammary 4T1.0 or fat 4T1.3pads. Nfe2 cells-expressing and were or control subjected vector- totransfected anti-pan 4T1.0 cytokeratin cells (1.5 × 10 or5) were anti-NFE2 injected antibodies staining. Pan-cytokeratin-positiveinto mammary fat pads and tumor tumor growth areas was aredetermined demarcated at the indicated by dot time lines. points. Representative All values images µ from 5 independent mice are shown here. Insets indicate enlarged area of tumor foci. Scar bars, 100 m in main images, 50 µm in insets. (f) Tumor growth rates of Nfe2-expressing or control vector-transfected 4T1.0 cells upon their injection into mammary fat pads. Nfe2-expressing or control vector-transfected 4T1.0 cells (1.5 105) were injected into mammary fat pads and tumor growth was determined at the × indicated time points. All values represent mean + SD (n = 5). (g) Expression level of Nfe2 mRNA in 4T1.3 cells, which were cultured in vitro, grew at mammary fat pads or in a bone marrow cavity. Total RNAs were extracted from 4T1.3 cells, which were cultured in vitro (In vitro), were isolated from primary tumor at day 9 after orthotopic injection (In MFP) or from a bone marrow cavity at day 7 after intraosseous injection (In BM). All values represent mean + SD (n = 3). * p < 0.05 (h)Eﬀects of NFE2 expression level on the prognosis of breast cancer patients. Overall survival of patients with breast cancer (GSE16446) was analyzed based on NFE2 expression level by analyzing the Kaplan–Meier Plotter database. Cancers 2020, 12, x FOR PEER REVIEW 6 of 18

represent mean + SD (n = 5). (g) Expression level of Nfe2 mRNA in 4T1.3 cells, which were cultured in vitro, grew at mammary fat pads or in a bone marrow cavity. Total RNAs were extracted from 4T1.3 cells, which were cultured in vitro (In vitro), were isolated from primary tumor at day 9 after orthotopic injection (In MFP) or from a bone marrow cavity at day 7 after intraosseous injection (In BM). All values represent mean + SD (n = 3). * p < 0.05 (h) Effects of NFE2 expression level on the prognosis of breast cancer patients. Overall survival of patients with breast cancer (GSE16446) was Cancersanalyzed2020, 12 based, 3003 on NFE2 expression level by analyzing the Kaplan–Meier Plotter database. 6 of 18

2.3. Reduced Intraosseous Growth of Breast Cancer Cells by Nfe2 Expression Ablation 2.3. Reduced Intraosseous Growth of Breast Cancer Cells by Nfe2 Expression Ablation We next examined the effects of Nfe2 gene expression ablation on intraosseous growth. We next examined the effects of Nfe2 gene expression ablation on intraosseous growth. Additionally, Additionally, from another mouse TNBC cell line, TS/A, we established TS/A.3 clone, which from another mouse TNBC cell line, TS/A, we established TS/A.3 clone, which exhibited a higher exhibited a higher capacity to grow in a bone cavity but not at mammary fat pads, compared with its capacity to grow in a bone cavity but not at mammary fat pads, compared with its parental clone parental clone (Figure 4a,b). Nfe2 shRNA reduced its gene expression in 4T1.3 and TSA.3 clones (Figure4a,b). Nfe2 shRNA reduced its gene expression in 4T1.3 and TSA.3 clones (Figure4c) and (Figure 4c) and protein expression in 4T1.3 (Figure 4d) with few effects on in vitro mRNA expression protein expression in 4T1.3 (Figure4d) with few e ffects on in vitro mRNA expression levels of Rank, levels of Rank, Rankl, and Opg (Figure 4e). Concomitantly, Nfe2 shRNA transduction significantly Rankl, and Opg (Figure4e). Concomitantly, Nfe2 shRNA transduction significantly reduced the tumor reduced the tumor growth of 4T1.3 and TS/A.3 clones upon their intraosseous injection (Figure 4f) growth of 4T1.3 and TS/A.3 clones upon their intraosseous injection (Figure4f) with few e ffects on with few effects on their tumor formation upon their injection into mammary fat pads (Figure 4g). their tumor formation upon their injection into mammary fat pads (Figure4g). These observations These observations would indicate that NFE2 provided breast cancer cells with a growth advantage would indicate that NFE2 provided breast cancer cells with a growth advantage in a bone cavity but in a bone cavity but not at primary sites. not at primary sites.

t (a) (b) (c) t

(%)

n 1.2 4T1.3 1.2 TS/A.3 i 0.6 Nfe2 Nfe2

)

400

s ** 3

o l TS/A

m 1.0 1.0

m TS/A.3

(

300 i i 0.8 0.8

o 0.4 s

t 0.6 t 0.6

r 200 r

r 3

o r 2.5 3

H r 0.4

0.4 H

0.2

3 o

3 o 2.5 o

100 i

t 0.2 2.0 0.2

e 2.0 i

n 2

n n 2

s p 0.0 0.0 R

s 0.0 0 o 1.5

e e e 2 r 2

r r 2 r r 2

0 2 4 6 8 10 12 14 s

s P TS/A TS/A.3 1.5 c e c e

p p

S f S f e

e e

- - x

r r N 1.0 N

A - e A - e

Days p A A p 1 1 N N e

x x 1.0 R

e R N N v

v v

e e

i h R i h R 1 1

0.5 t

t t s s e

e e - h - h

a a

3 s 3 s v

v v

l (d) . - l . - l

i (e) i 1 3

3 e

0.5 t

e e Tt . /A . t

1 a a a 0 0.0 A 0 R 4 R S R

l / l T T

4 S e e RTank Rankl e Opg Rank Rankl 0 Opg Rank Rankl Opg

0.0 t 0

R R

r r shRNA-Scr

p 4T1.0-in BM 4T1.3-in BM

p p Rank Rankl 4T1.0-ControlOpg 4T1.0-Nfe2Rank Rankl 4T1.3-sh-ScrOpg 4T1.3-sh-Nfe2Rank Rankl Opg shRNA-Scr 2.5 3 3

4T1.0-Control shRNA4T1.0-Nfe2-Nfe2 4T1.3-sh-Scr 4T1.3-sh-Nfe2 4T1.0-in BM 4T1.3-in BM

2.0

i i 2 2

s s 1.5

e e

p p

x x 1.0

e e

1 1

v v

0.5 t

t t

a a

l shRNA-Nfe2 l l

e e e 0 0.0 0 R

R R Rank Rankl Opg Rank Rankl Opg Rank Rankl Opg 4T1.0-Control 4T1.0-Nfe2 4T1.3-sh-Scr 4T1.3-sh-Nfe2 4T1.0-in BM 4T1.3-in BM

M (f) M (g)

(%) B

(%) 4T1.3 TS/A.3

)

) n 900 1500

i 0.6

3 4T1.3-shRNA-Scr 3 TS/A.3-shRNA-Scr i 0.15

l **

l l * 750 4T1.3-shRNA-Nfe2 TS/A.3-shRNA-Nfe2

m m 1200

(

r r 600

m 0.4 m o 0.10 900

l 450 l

f 600

o o 300

o 0.05 0.2

o o 300

u i 150

p p 0.0 0 0

0.00 o

r r 0 3 6 9 12 15 18 21 0 3 6 9 12 15 18 21

P r P r 2 2 c e c e -S f -S f Days A -N A -N N A N A Days Days R N R N h R h R -s h -s h .3 -s .3 -s 1 .3 A .3 T 1 / 4 T S /A 4 T S Figure 4. (a) IntraosseousT injection of TS/A and TS/A.3 clones. Tibia was obtained from mice 9 days Figure 4. (a) Intraosseous injection of TS/A and TS/A.3 clones. Tibia was obtained from mice 9 days after intraosseous injection of TS/A or TS/A.3 cells and was subjected to ﬂow cytometric analysis. after intraosseous injection of TS/A or TS/A.3high cells and was subjected to flow cytometric analysis. The The proportions of CD45− CD24− CD44 tumor cells among total cells were determined as described proportions of CD45- CD24- CD44high tumor cells among total cells were determined as described in in the Materials and Methods section. All values represent mean + SD (n = 7). ** p < 0.01. (b) Tumor the Materials and Methods section. All values represent mean + SD (n = 7). ** p < 0.01. (b) Tumor growth rates of TS/A and TS/A.3 cells upon their injection into mammary fat pads. TS/A or TS/A.3 cells (1.5 105) were injected into mammary fat pads and tumor growth was determined at the × indicated time points. All values represent mean + SD (n = 4). (c) Nfe2 expression level in Nfe2 shRNA-transfected 4T1.3 or TS/A.3 clones. Either 4T1.3 or TS/A.3 clones were transfected with Nfe2 or scrambled shRNA-expressing vectors. Total RNAs were extracted from the resultant transfectants and were subjected to qRT-PCR to quantify Nfe2 mRNA. (d) Immunohistochemical analysis of Nfe2 shRNA- Cancers 2020, 12, 3003 7 of 18

transfected 4T1.3 cells in a bone cavity. Tibiae were obtained from mice at day 9 after intraosseous injection of Nfe2 shRNA- or scrambled shRNA-transfected 4T1.3 cells and were subjected anti-pan cytokeratin, or anti-NFE2 antibodies staining. Pan-cytokeratin-positive tumor areas are demarcated by dot lines. Representative images from 5 independent mice are shown here. Insets indicate enlarged area of tumor foci. Scar bars, 100 µm in main images, 50 µm in insets. (e) Determination of mRNA expression level of Rank, Rankl, or Opg in Nfe2 shRNA-transfected 4T1.3 cells. Total RNAs were extracted either from Nfe2 shRNA- or scrambled shRNA-transfected 4T1.3 cells cultured in vitro. The mRNA expression levels were determined by qRT-PCR. All values represent mean + SD (n = 3). (f) Intraosseous tumor formation of Nfe2 shRNA-transfected 4T1.3 or TS/A.3 clones. Tibiae were obtained from mice 9 days after intraosseous injection of Nfe2 shRNA- or scrambled shRNA-transfected 4T1.3 or TS/A.3 cells and were subjected to ﬂow cytometric analysis. All values represent mean + SD (n = 7 to 8). * p < 0.05. (g) Tumor growth rates of Nfe2 shRNA-transfected 4T1.3 or TS/A.3 clones upon their injection into mammary fat pads. Nfe2 shRNA- or scrambled shRNA-transfected 4T1.3 or TS/A.3 clones (1.5 105) × were injected into mammary fat pads and tumor growth was determined at the indicated time points. All values represent mean + SD (n = 5). 2.4. Contribution of NFE2 to Breast Cancer Cell Survival under Hypoxic and Anchorage-Independent Conditions In order to delineate the molecular mechanism underlying NFE2-mediated enhanced intraosseous tumor formation, we examined the eﬀects of NFE2 expression on in vitro proliferation. Reduced Nfe2 expression did not impair the in vitro proliferation and survival of 4T1.3 or TS/A.3 clone under normoxic and anchorage-dependent conditions (Figure5a,c). On the contrary, Nfe2 shRNA treatment reduced the growth advantage of 4T1.3 or TS/A.3 clone under hypoxic and anchorage-independent conditions (Figure5b,d), the conditions which resemble the bone microenvironment. Consistently, Nfe2 gene transduction provided 4T1.0 clone with a growth advantage under hypoxic and anchorage-independent conditions but not under normoxic and anchorage-dependent ones (Figure5e,f). These observations would indicate that NFE2 could enhance the growth of breast cancer cells in a bone cavity where hypoxia predominates [18]. However, hypoxia-inducible factor (Hif)1α expression was not augmented in 4T1.3 cells grown in a bone cavity, compared with those under normoxic in vitro culture conditions (Figure5g). Moreover, Nfe2 gene transduction into 4T1.0 cells did not increase the expression level of hypoxia-related genes including Hif1α, Hif2α, Myc, and Nrf2 compared with a control clone even under hypoxic conditions (Figure5h,i). Thus, NFE2-mediated growth enhancement may not be ascribed to the enhanced expression of these hypoxia-related transcription factors.

2.5. Wnt Pathway Activation in Nfe2-Mediated Enhanced Intraosseous Tumor Formation No evidence to indicate the direct effects of NFE2 on breast cancer cell proliferation [14,17], prompted us to conduct gene set enrichment analysis (GSEA) to delineate biological pathways and processes that are associated with NFE2-mediated enhanced intraosseous tumor formation. The analysis revealed the enrichment in Wnt pathway in Nfe2-expressing 4T1.0, compared with its control clone (Figure6a). Then, we validated this result by using qRT-PCR. Nfe2-expressing 4T1.0 clone displayed enhanced gene expression of Wnt-related molecules including Axin2 and Lef1 compared with the control 4T1.0 clone (Figure6b). Similarly, the expression of Wnt-related molecules, such as Axin2, Lbh, and Lef1, was augmented in intraosseously growing 4T1.3 and TS/A.3 cells, compared with intraosseously growing 4T1.0 and TS/A cells, respectively (Figure6c,d). We finally examined the involvement of the Wnt pathway in NFE2-mediated growth promotion by using a specific Wnt/β-catenin inhibitor, LF3. The inhibitor did abrogate enhancement in cell proliferation of Nfe2-expressing 4T1.0 cells under hypoxic and anchorage-independent conditions (Figure6e,f). Altogether, NFE2 expression can activate the Wnt pathway, which can confer a growth advantage on tumor cells when they grow in a bone cavity with a low oxygen content. Cancers 2020, 12, 3003 8 of 18

(a) (b) : shRNA Scr (×104) (%) (×104) (%) 1000 :shRNANfe2 100 40 100 **

75 30 75 Rate 4T1.3 600 l

va 50 20 50 ** i * 10 25

25 numbers Cell Survival Rate Cell numbers 200 Surv 0 0 0 0 244872 0244872 0 244872 0 244872 (c) (d) (×104) (%) (×104) (%) 1000 : shRNA Scr 100 40 100 * :shRNANfe2 ** 75 30 75 TS/A.3 600 * 50 20 50 numbers

l * l 10 25 Survival Rate Survival Cell numbers Ce 25 200 Survival Rate 0 0 0 0 244872 0 244872 0244872 0244872 (e) (f) 4 (×10 ) : Control 4 (%) (%) (×10 ) 100 1000 :Nfe2 100 40 s -expressing r ** 30 * 75

75 Rate 4T1.0 600 l va 50 20 * i 50 numbers v l r l ** 25

10 Su Cell numbe

Ce 25 200 Rate Survival 0 0 0 0 244872 0 244872 0244872 0 244872 Hours Hours Hours Hours HIF-1α (g) 1.2 (h) 2.0 (i) 2.5 2.0 1.5 0.8 1.5 1.0 1.0 0.4 0.5 0.5

0.0 Relative expression to Hprt to expression Relative 0.0 0.0 Relative expression to Hprt Relative expression Hprt to expression Relative Hif1α Hif2α Myc Nrf2 Hif1α Hif2α Myc Nrf2 ro HIF1a HIF2a Myc Nrf2 HIF1a HIF2a Myc Nrf2 vit In BM In 4T1.0-Control 4T1.0-Nfe2 4T1.0-Control 4T1.0-Nfe2

Figure 5. (a,b) In vitro proliferation rates of Nfe2 shRNA- or scrambled shRNA-transfected 4T1.3 cells under normoxic (a) or hypoxic and anchorage-independent conditions (b). (c,d) In vitro proliferation rates of Nfe2 shRNA- or scrambled shRNA-transfected TS/A.3 cells under normoxic (c) or hypoxic and anchorage-independent conditions (d). (e,f) In vitro proliferation rates of Nfe2-expressing or control vector-transfected 4T1.0 cells under normoxic (e) or hypoxic and anchorage-independent

conditions (f). Each of the cell lines were cultured under normoxic (20% O2) or hypoxic (1% O2) and (a)anchorage-independent conditions,(b) respectively. Cell numbers(c) and survival rates(d) under each condition

were determined by using trypan blue4 exclusion assay.30 All values represent mean20 + SD (n = 3). 0.3 Gene set: Wnt_UP.V1_UP * * * ** * p0.2< 0.05, ** p < 0.01.Size: 114, ( gNES:) Expression 1.48 level of* Hif1α mRNA expression in 4T1.3 cells, which were cultured NOM p-value: 0.0185 3 15 in0.1 vitro , or grewFDR in q-value: a bone 0.12251 cavity. Total RNAs were* extracted20 from 4T1.3 cells, which were cultured 0.0 2 10 in-0.1 vitro (In vitro) or grew in a bone cavity (In BM) at day10 7 after intraosseous injection. All values

Enrichment Score 5 * represent-0.2 mean + SD (n = 3). (h,i) Determination1 of mRNA expression level of hypoxia-related genes in 0 0 to Hprt Relative expression 0 Hprt to expression Relative Nfe2-expressing or control vector-transfected to Hprt expression Relative 4T1.0 cells, which was cultured under normoxic or hypoxic Axin2 Lbh Lef1 Axin2 Lbh Lef1 Axin2 Lbh Lef1 T 4T1.0-Control and4T1.0- anchorage-independentNfe2 conditions. Total RNAs were extracted4T1.0 from4T1.3 Nfe2 -expressingTS/A or TS/A.3 control 4T1.0-Control vector-transfected 4T1.0 cells cultured under4T1.0-Nfe2 normoxic (h) or hypoxic and anchorage-independent (e)conditions (i), respectively. All values represent mean(f)+ SD (n = 3).

(a) (b) : shRNA Scr (×104) (%) (×104) (%) 1000 :shRNANfe2 100 40 100 **

75 30 75 Rate 4T1.3 600 l

va 50 20 50 ** i * 10 25

75 Rate 4T1.0 600 l va 50 20 * i 50 numbers v l r l ** 25

10 Su Cell numbe

Ce 25 200 Rate Survival 0 0 0 0 244872 0 244872 0244872 0 244872 Hours Hours Hours Hours HIF-1α (g) 1.2 (h) 2.0 (i) 2.5 2.0 1.5 0.8 1.5 1.0 1.0 0.4 0.5 0.5

Cancers 2020, 12, 3003 9 of 18

(a) (b) (c) (d)

4 30 20 0.3 Gene set: Wnt_UP.V1_UP * * * ** 0.2 Size: 114, NES: 1.48 * NOM p-value: 0.0185 3 15 0.1 FDR q-value: 0.12251 * 20 0.0 2 10 -0.1 10

Enrichment Score 5 * -0.2 1 0 0 to Hprt Relative expression 0 Hprt to expression Relative Relative expression to to Hprt expression Relative Axin2 Lbh Lef1 Axin2 Lbh Lef1 Axin2 Lbh Lef1 T 4T1.0-Control 4T1.0-Nfe2 4T1.0 4T1.3 TS/A TS/A.3 4T1.0-Control 4T1.0-Nfe2 (e) (f)

1 Figure 6. (a) Gene set enrichment analysis of Nfe2-expressing or control vector-transfected 4T1.0 cells in vitro. Total RNAs were extracted from Nfe2-expressing or control vector-transfected 4T1.0 cells under in vitro culture conditions and were subjected to bulk RNA sequencing. (b) Determination of mRNA expression of Wnt-related molecules in Nfe2-expressing or control vector-transfected 4T1.0 cells in vitro. Total RNAs were extracted from Nfe2-expressing or control vector-transfected 4T1.0 cells under in vitro culture conditions. All values represent mean + SD (n = 3). * p < 0.05. (c) Determination of mRNA expression of Wnt-related molecules in 4T1.0 or 4T1.3 cells, which grew in a bone cavity. Total RNAs were extracted from 4T1.3 cells or its parental 4T1.0 cells, which were isolated from a bone cavity at day 9 after intraosseous injection as described in the Materials and Methods section. Comparative CT values were calculated by normalizing to each value of 4T1.0 cells cultured under in vitro conditions. All values represent mean + SD (n = 3). * p < 0.05, ** p < 0.01. (d) Determination of mRNA expression of Wnt-related molecules in TS/A or TS/A.3 cells, which grew in a bone cavity. Total RNAs were extracted from TS/A.3 cells or its parental TS/A cells, which were isolated from a bone cavity at day 9 after intraosseous injection. Comparative CT value were calculated by normalizing to each value of TS/A cells cultured under in vitro conditions. All values represent mean + SD (n = 2 to 3). * p < 0.05. (e,f) In vitro proliferation rates of Nfe2-expressing or control vector-transfected 4T1.0 cells in the presence of LF3 under normoxic (e) or hypoxic and anchorage-independent conditions (f). Each of

the cell lines were cultured with LF3 or vehicle control, DMSO, under normoxic (20% O2) or hypoxic (1% O2) and anchorage-independent conditions, respectively. Cell numbers and survival rates were determined by using trypan blue exclusion assay. All values represent mean + SD (n = 3). ** p < 0.01.

3. Discussion Like metastasis to other organs, bone metastasis is based on a multi-step process from tumor growth at primary sites to tumor formation in bone [19]. Accumulating evidence indicates the involvement of several transcription factors, Runx-2 [11] and Bach1 [12], in breast cancer metastasis to bone. RUNX-2 can promote breast cancer bone metastasis by increasing integrin α5-mediated colonization [20] or inducing RANKL-mediated osteoclast activation [21]. Bach1 can promote bone metastasis by enhancing the expression of metalloproteinase 1 and a chemokine receptor, CXCR4 [12]. On the contrary, we previously demonstrated that a higher capacity of 4T1.3 clone to metastasize to bone can be ascribed to its higher ability to grow in a bone cavity but neither to enhancement in its colonization, induction of osteoclast activation, nor Cxcr4 expression [10]. Hence, we presumed that an additional transcription factor(s) may contribute to 4T1.3 clone metastasis to bone by binding cis-elements of Ccl4 gene, which had an indispensable role in bone metastasis and exhibited enhanced Cancers 2020, 12, 3003 10 of 18 expression [10]. Through combined dry bioinformatics and wet analyses, Nfe2 has finally been identified as a transcription factor, which can provide breast cancer cells with an intraosseous growth advantage. NFE2 was originally discovered as a DNA binding protein present in extracts of erythroid cells [22] and was subsequently characterized as a hematopoietic-specific basic-leucine zipper protein which dimerizes with the small Maf subunits [23], thereby binding activator protein (AP)-1-like recognition site [24]. NFE2 was initially presumed to be an indispensable transcription factor for correct regulation of porphobilinogen deaminase (PBGD) gene [25], coding for the third enzyme of the heme biosynthesis. Moreover, NFE2 can regulate α- and β-globin gene transcription [26,27]. Neonates exhibit severe anemia probably compounded by concomitant bleeding due to severe thrombocytopenia, but surviving adults exhibit only mild anemia with compensatory reticulocytosis and splenomegaly [28]. Thus, NFE2 deficiency may be compensated by other transcription factors involved in erythroid genesis and globin gene expression. However, NFE2-deficient mice develop severe thrombocytopenia arising from the intrinsic defects in the megakaryocyte lineage [29], indicating its indispensable role in megakaryocyte biogenesis. In contrast, mice expressing an Nfe2 transgene in hematopoietic cells exhibit many features of myeloproliferative disorders (MPN) including thrombocytosis and leukocytosis, and often develop acute myeloid leukemia [14] or myelosarcoma [17]. Consistently, MPN patients display Nfe2 overexpression arising from either recurrent truncating mutation [15] or epigenetic changes mediated by the histone demethylase JMJD1C [16]. These observations would indicate the crucial roles of aberrant Nfe2 expression in some types of leukemogenesis. NFE2 expression is presumed to be restricted to hematopoietic cells such as erythroid cells, megakaryocytes and granulocytes [30,31]. On the contrary, we provided the first definitive evidence to indicate that NFE2 was robustly expressed in mouse breast cancer cells grown in a bone cavity. Moreover, we proved that NFE2 could confer a growth advantage on breast cancer cells under hypoxic and anchorage-independent conditions, which mimic the microenvironment in a bone cavity. The studies on Nfe2 transgenic mice proved that Nfe2 gene overexpression could increase colony formation of several types of hematopoietic progenitor cells, but without additional elucidation on its molecular mechanisms [14,17]. Hence, we conducted GSEA on Nfe2-overexpression 4T1.0 clone and demonstrated, together with subsequent qRT-PCR analysis, the enrichment of Wnt pathway molecules in 4T1.3 clones grown in a bone cavity as well as Nfe2-expressing breast cancer cells. The Wnt pathway is involved in various aspects of carcinogenesis, such as cell proliferation, resistance to cell death, and angiogenesis [32]. A Wnt inhibitor abrogated NFE2-mediated enhancement in cell proliferation and survival observed on Nfe2-overexpressing cells which were cultured under hypoxic conditions. Given that bone cavity is generally hypoxic, it is likely that the activated Wnt pathway is associated with enhanced breast cancer cell survival in the bone cavity. NF-E2-related factor (Nrf2) is a master regulatory transcription factor of cytoprotective genes [33], and its aberrant expression is frequently observed in cancer cells and can provide an advantage for their growth [34]. Nrf2 is also a basic-leucine zipper DNA-binding protein and shows a structural similarity with Nfe2 even at its DNA-binding domain [35]. Like NFE2, Nrf2 heterodimerizes with the small Maf subunit [36], and binds to anti-oxidant responsive element (ARE) or electrophile-responsive element (EpRE) to induce transcription of target genes [37]. Moreover, several lines of evidence indicate that aberrant Nrf2 expression could activate Wnt pathways in several types of carcinogenesis [38,39]. The consensus ARE/EpRE sequence comprises 50-TGACNNNGC-30 [37], which is highly similar to the consensus-binding sequence for Nfe2 [40]. Hence, it is tempting to speculate that NFE2 might activate similar sets of target genes as Nrf2 did and eventually activate the Wnt pathway. Analyses on the public database, GeneCards® (https://www.genecards.org), together with our RNAseq on intraosseously grown 4T1.3 and 4T1.0 cells, demonstrated that NFE2 can regulate the expression of several bone metastasis-related transcription factors including Runx2 [11], Bach1 [12], and Esr1 [41]. Thus, NFE2 may be a hub of the transcription factors involved in bone metastasis and consequently its blockade may be effective to treat bone metastasis. NFE2 and Nrf2 bind with a similar region of the small Maf subunit to form a heterodimer and to eventually exert their functions [13,33]. Cancers 2020, 12, 3003 11 of 18

As ML385, a thiazole-indoline compound, can specifically inhibit the interaction between Nrf2 and Maf [42], it may be able to inhibit the interaction between NFE2 and Maf, thereby blocking NFE2 activities. Alternatively, inhibition of Nfe2 expression can be employed to treat bone metastasis of breast cancer with aberrant Nfe2 expression. As reported on Runx-2 [43], the treatment with siRNA may be feasible to decrease Nfe2 expression. Moreover, Nfe2 transcription could be enhanced by another transcription factor, Runx-1 [44]. Thus, Nfe2 expression can be reduced by the modulation of the expression and/or functions of this transcription factor. Furthermore, Nfe2 expression in hematopoietic cells can be enhanced also by several secreted bioactive molecules such as activin A [45] and platelet-derived growth factor (PDGF) [46]. As our preliminary observations demonstrated that Pdgf mRNA expression was increased in fibroblasts present in 4T1.3-growing bone cavity, fibroblast-derived PDGF may enhance Nfe2 expression in bone metastasis foci. Hence, PDGF blockade may be also effective to treat bone metastasis by decreasing Nfe2 expression. Nevertheless, more detailed analysis on the functions and expression mechanisms of NFE2 will be warranted to establish a successful strategy to treat bone metastasis by targeting NFE2.

4. Materials and Methods

4.1. Cell Lines A mouse TNBC cell line, BALB/c-derived 4T1 (CRL-2539), was purchased in 2007 from American Type Culture Collection. Two clones, parental 4T1.0 clone and 4T1.3 clone with a high capacity to metastasize to bone, were established from 4T1 as previously described [10]. TS/A, another mouse TNBC cell line established from BALB/c mammary adenocarcinoma [47], was obtained in 2010 from Dr. Nanni. To establish a TS/A clone with a high growth capacity in the bone marrow cavity, parental TS/A cells (2.5 103 cells in 20 µL HBSS) were injected into a bone marrow cavity of BALB/c mouse tibiae. × Eleven days after the tumor inoculation, total bone marrow cells were collected and cultured for 3 weeks in medium containing 50–100 µg/mL G418. The resultant cells, TS/A.1 clone, were injected again into a bone marrow cavity of a new mouse tibiae, and this cycle was repeated three times to obtain TS/A.3 clone. All these cell lines were cultured at 37 ◦C under 5% CO2 in a RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS) and used within 10 passages of receiving from the source or the establishment of the clone.

4.2. Mice Eight-week old speciﬁc pathogen-free BALB/c mice were obtained from Charles River Laboratories (Yokohama, Japan) and were kept under the speciﬁc pathogen-free conditions. All the animal experiments in this study complied with the Guideline for the Care and Use of Laboratory Animals of Kanazawa University with the approval by the Committee on Animal Experimentation of Kanazawa University (AP-163706).

4.3. Antibodies Mouse anti-mouse pan cytokeratin antibody (BioLegend, San Diego, CA, USA), rabbit anti-mouse LMO2 and rabbit anti-mouse NFE2 antibodies (Novus Biologicals, Centennial CO, USA), rabbit anti-mouse c-MYB antibody (Bioss Antibodies, Woburn, MA, USA), rat anti-mouse Ki-67 antibody (Dako Japan, Kyoto, Japan), mouse anti-human CD51 antibody (BD Biosciences, San Diego, CA, USA), rabbit anti-mouse CD31 antibody (Abcam, Toronto, ON, Canada), and goat anti-mouse VEGF antibody (Santa Cruz, Sana Cruz, CA, USA) were used as the primary antibodies for immunohistochemical analysis. The following antibodies were used as the primary antibodies for ﬂow cytometry analysis; Allophycorin (APC)-labeled rat anti-mouse CD326 and peridinin–chlorophyll–protein complex–cyanine 5.5 (PerCP/Cy5.5)-labeled rat anti-mouse CD24 antibodies (BioLegend), FITC- or PerCP/Cy5.5-labeled rat anti-mouse CD45 antibodies (eBioscience, San Diego, CA, USA), APC-labeled rat-anti mouse CD44 antibody (BD Biosciences). Cancers 2020, 12, 3003 12 of 18

4.4. RNAseq Analysis Total RNAs were extracted from the cell lines and its quality was conﬁrmed using an Agilent 2200 TapeStation (Agilent Technologies, Palo Alto, CA, USA). RNA Integrity Numbers (RIN) of all samples were 10.0, and were subjected to RNAseq analysis as previously described. GSEA was executed through the GenePattern server [48–50]. As the metric for ranking genes, log2_ratio_of_means was exploited, and the gene set size was set to 20 to 200.

4.5. Procedures to Identify Candidate Transcription Factors Candidate transcription factors were selected as those that fulfilled all three of the following conditions (Figure2a). The first condition is that the candidate transcription factor gene expression should be confirmed to be more abundant in 4T1.3 cells grown in a bone cavity than 4T1.0 cells grown in a bone cavity with the use of RNAseq analysis on tumor foci arising from intraosseous injection of 4T1.0 or 4T1.3 cells [10]. The second condition is that the binding site of the candidate transcription factors should be verified to be localized in the region from 2 kbp upstream to 200 bp downstream of the transcription start site of mouse Ccl4 gene, which is abundantly expressed in 4T1.3 cells [10], with the use of TFBIND (http://tfbind.hgc.jp) at a cut-off value higher than 0.85. The third condition is that the candidate transcription factor gene should be proven to be expressed more abundantly in bone metastasis sites than lung metastasis sites in GSE14020, which is a dataset of genes expressed in various metastatic sites of breast cancer patients.

4.6. Establishment of Cell Lines Constitutively-Expressing Transcription Factors Full length construct of mouse Lmo2 (MMM1013-202769386), Myb (MMM1013-202763262) and Nfe2 cDNAs (MMM1013-202804796) were purchased from Dharmacon (GE Healthcare Dharmacon, Milwaukee, WI, USA). Each protein coding portion is cloned into the pIRES vector (Takara Bio USA, Mountain View, CA, USA, PT3266-5) using suitable restriction sites. The resultant expression vectors were transduced into 4T1.0 clone using JetPRIME DNA and siRNA transfection reagent (Polyplus transfection, Illkirch, France). Then, the cells were cultured in medium containing G418 (400 µg/mL) for 7 days, to obtain constitutively expressing clones.

4.7. Establishment of Cell Lines Expressing shRNA MISSION®shRNA Lentiviral Transduction Particles (Sigma Aldrich, St. Louis, MO, USA, SHCLNV-NM_008685) were used to knockdown the Nfe2 gene expression in the 4T1.3 and TS/A.3 clones. shRNAs were generated against the mouse Nfe2 mRNA (TRC Number: TRCN0000081983 and TRCN0000424174) and were packaged into the lentiviral particles. shRNAs targeting scrambled RNAs (Nontarget; SHC002) were prepared with the use of pLK01-puro (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer’s instructions. Stably gene knocked-down clones were selected in the presence of puromycin (3.5 µg/mL for 4T1.3 cell line and 6 µg/mL for TS/A.3 cell line).

4.8. Intraosseous Injection Tumor cell suspensions (5.0 103 4T1.0-derived cells or 10 103 TS/A-derived ones in 20 µL HBSS) × × were injected into a bone marrow cavity of tibiae as previously described [51]. Seven or nine days after the injection, tibiae were removed for a subsequent analysis. In order to delineate transcription factor expression, the resected tibiae were subjected to the immunostaining with anti-LMO2, -MYB or -NFE2 antibodies by using the ABC Elite Kit (Vector Laboratories, Burlingame, CA, USA) and peroxidase substrate 3, 30-diaminobenzidine kit (Vector Laboratories) in an essentially similar way as described previously [10]. To evaluate intraosseous tumor mass arising from 4T1.0-derived clones, the resected mouse tibiae were immunostained with anti-pan cytokeratin antibodies and the ratios of pan-cytokeratin-positive to whole bone marrow areas were determined with the use of a microscopy system (BZ-X700) and Keyence Analysis Software (Keyence Japan, Osaka, Japan), as previously Cancers 2020, 12, 3003 13 of 18 described [10]. To evaluate osteoclast numbers in a bone cavity, resected mice tibiae were subjected to the immunostaining with anti-CD51 antibody, and the ratios of CD51-positive areas to the whole bone marrow areas were determined. To evaluate tumor cell proliferation and angiogenesis in a bone cavity, resected mouse tibiae were subjected to immunostaining with anti-Ki-67, -CD31 or -VEGF antibodies, and the ratios of each of the positive cells or positive areas to the whole tumor foci areas were determined. For the enumeration of tumor cell numbers in bone marrow, single cell suspensions were prepared from the resected tibial bones, by sequentially flushing with the RPMI medium and removing erythrocytes with ammonium chloride lysis buffer. The resulting single cell suspensions were incubated with either the combination of FITC-labeled anti-CD45, PerCP/Cy5.5-labeled anti-CD24, and APC-labeled anti-CD44 antibodies, or the combination of PerCP/Cy5.5-labeled anti-CD45 and APC-labeled anti-CD326 antibodies, for 30 min at 4 ◦C. Dead cells were removed from acquired data with a fixable viability Dye (eBioscience). The stained cells were acquired on a FACS Canto System II (BD Biosciences) and analyzed using FlowJo software (BD Biosciences). 4T1.0-derived and + high TS/A-derived tumor cells were defined as CD45−CD326 and CD45−CD24−CD44 cells, respectively. In another series of experiments, 4T1.0-derived tumor cells were isolated and collected by using FACS Aria cell sorter (BD Biosciences) and subjected to RNA extraction immediately.

4.9. In Vitro Cell Proliferation Assay The cell suspensions (2.0 105 cells in 10 mL) were seeded in 100 mm dish and cultured at 37 C × ◦ for the indicated time intervals. Live cell numbers were determined by using trypan blue exclusion assay while survival rates were calculated as the ratios of live cell numbers to the sum of live and dead cells. In some experiments, cell suspensions (2.0 105 cells in 500 µL) were added to each well of an ® × EZ-BindShut 24-well plate (IWAKI Glass, Chiba, Japan) and incubated at 37 ◦C under hypoxic (1% O2) and anchorage-independent conditions for the indicated time points to determine live cell numbers and survival rates. In another series of experiments, Nfe2-expressing or control vector-transfected 4T1.0 cells were seeded and cultured in normoxic (20% O2) and anchorage-dependent, or hypoxic and anchorage-independent conditions in the presence of Wnt inhibitor, LF3 (30 µM, SelleckChem, Houston, TX, USA) or DMSO to determine live cell numbers and survival rates. In these experiments, culture media were not changed until the end of the experiments.

4.10. Tumor Growth at the Primary Site The cell suspensions (1.5 105 cells in 100 µL) were injected into the secondary mammary fat pads × and tumor growth was evaluated by measurement with calipers every 2 to 3 days. Tumor volumes were calculated as a b2/2, where a and b mean the long and the short diameters of the tumor, × respectively. To collect tumor cells from primary tumor tissues, single cell suspensions were prepared from the resected primary tumors 9 days after orthotopic injection. Primary tumors were cut into small pieces and treated with 1 mg/mL collagenase D (Roche, Basle, Switzerland) and 40 µg/mL bovine pancreas-derived deoxyribonuclease I (Sigma-Aldrich) for 20 min and 37 ◦C with constant shaking. The resulting mixtures were filtered through a 100 µm cell strainer (Corning, Corning, NY, USA) to obtain single cell suspensions. Erythrocytes were removed with ammonium chloride lysis buffer. The resulting single cell suspensions were incubated with the combination of FITC-labeled anti-CD45, PerCP/Cy5.5-labeled anti-CD24, PE-labeled anti-CD44, and APC-labeled anti-CD326 antibodies, for 30 min at 4 ◦C. Dead cells were removed from acquired data with a fixable viability Dye + high (eBioscience). Tumor cells were defined as CD45−CD326 CD24−CD44 cells and were collected by using a FACS Aria cell sorter (BD Biosciences). Isolated tumor cells were subjected to total RNAs extraction immediately.

4.11. qRT-PCR Analysis Total RNAs were extracted from cell lines or tumors with the use of RNeasy Mini Kit (Qiagen, Chatsworth, CA, USA) and subjected to qRT-PCR by using the primers listed in AppendixA Table A1, Cancers 2020, 12, 3003 14 of 18 as described previously [52]. Expression levels of the target genes were analyzed through the comparative threshold cycle method (∆∆CT). The hypoxanthine guanine phosphoribosyltransferase (Hprt) were used as an internal control.

4.12. Clinical Database Analysis Overall survival (GSE16446) was evaluated by analyzing the data deposited in the Kaplan–Meier Plotter database (https://kmplot.com/analysis/) and was shown using Kaplan–Meier curve. The cohort was divided by using the cutoﬀ value 6 in analysis to set the patients into 2 groups, high and low NFE2-expressing groups.

4.13. Library Construction and Sequencing Total RNAs were obtained from the in vitro cultured parental or Nfe2-expressing 4T1 cells by a RNeasy Plus Mini Kit (Qiagen), following the instructions. The quality of RNAs was evaluated using an Agilent 4200 TapeStation (Agilent Technologies), and the concentration was measured using a NanoDrop (Thermo Fisher Scientific, Waltham, MA, USA). The libraries for sequencing were created using TruSeq Stranded mRNA (Illumina, San Diego, CA, USA), following the procedure, and their quality and quantities were measured by using an Agilent 4200 TapeStation, a Qbit Fluorometer (Thermo Fisher Scientific) and a KAPA Library Quantification kit (Roche). The average size of the libraries was 292–340 bp, and high-throughput sequencing for 6 samples was performed by NextSeq 500/550 High Output Kit v2.5 (75 cycles, 40/40 cycles, pair-end). All datasets have been deposited at the National Center for Biotechnology Information, Gene Expression Omnibus under accession number 154,541 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE154541). The result of bulk RNA sequencing was calculated by using a CLC Genomics Workbench (Filgen, Inc., Aichi, Japan). Then we investigated c6 (c6.all.v7.1.symbols.gmt) gene sets in the current study. The significant results of GSEA were identified using the criteria: FDR p-value < 0.05 and FDR < 0.25.

4.14. Statistical Analysis The means + SD were calculated for all parameters determined. Statistical signiﬁcance was evaluated using one-way ANOVA, followed by the Tukey–Kramer post-hoc test or the Mann–Whitney U test. p values of less than 0.05 were considered statistically signiﬁcant.

5. Conclusions In this study, we revealed that a transcription factor, Nfe2, plays an indispensable role in breast cancer metastasis to bone, by conferring growth advantages to breast cancer cells in the bone microenvironment. Moreover, Nfe2 expression was associated with enhanced expression of Wnt-related molecules, suggesting that NFE2 can regulate the Wnt pathway, thereby contributing to breast cancer growth in a bone cavity.

Author Contributions: D.Z. and S.-i.S. conducted the experiments using cell lines and mice. S.I., S.-i.H., and S.-i.S. conducted gene expression analysis. T.B., N.M., and S.-i.S. planned the research project and wrote the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Acknowledgments: This work is partially supported by Grant-in-Aid for Scientific Research (C) (20K07567) from the Japan Society for the Promotion of Science (So-ichiro Sasaki and Naofumi Mukaida). Conflicts of Interest: The authors declared no conflict of interest. Cancers 2020, 12, 3003 15 of 18

Appendix A

Table A1. Nucleotide sequence of the primers used for qRT-PCR.

Gene Species Size Forward Primer (50-30) Reverse Primer (30-50) Hprt mouse 90 TCCTCCTCAGACCGCTTTT CCTGGTTCATCATCGCTAATC Myb mouse 79 GCTGGAGTTGCTCCTGATGT GCTGCAAGTGTGGTTCTGTG Nfe2 mouse 126 GTACTCAGAGGCAGCGCTCA CAACCTTTCAGGCACCTGTGG Lmo2 mouse 70 GCAGCTGGATGATTCGCTCT CCTCCACGCTCAAGGACAGT App mouse 167 CCCAAGGCCTCATCATGTGT GCTCGTAGATCACACGGAGG Ets1 mouse 175 CGGTCAGCGGGAATTTGAGA TGATGGTGAGAGTCGGCTTG Sf1 mouse 135 CCATGCAGCACAACCCAAAT AGGCGTACTTCCCAGGTACT Pbx1 mouse 132 TGTCACAGCCACCAATGTGT TTGAGTGACTGCACGCTCAT Max mouse 142 GGCTTGTTGTTGTCGGTGAC AAACCTCGGTTGCTCTTCGT Nf1 mouse 199 CTTCTTCCTGCGACTGCGTT TCTGCGACAGACGTCAACAT Gata1 mouse 72 TTGGGATCACCCTGAACTCG GGTTGAACCTGGGCTTGTTG Gata2 mouse 176 ACCCCTATCCCGTGAATCCG CCACCTCCATGGTCCACTAC Ar mouse 105 GCTCACCAAGCTCCTGGATT TCAGGAAAGTCCACGCTCAC Rora mouse 170 GCCGAGGTATCTCAGTCACG AGGACAGGAGTAGGTGGCAT Axin2 mouse 184 AGCCTGTTTATCCCATCACG TGAGGCTTCACGTGCATTAG Lbh mouse 125 GATCGGCTGAGATGACCGAG GCAGCAACGGTCAAAGTCTG Lef1 mouse 165 CGGGAAGAGCAGGCCAAATA AGCTGTCATTCTGGGACCTG

References

1. Siegel, R.L.; Miller, K.D.; Jemal, A. Cancer statistics, 2020. CA 2020, 70, 7–30. [CrossRef][PubMed] 2. Dafni, U.; Tsourti, Z.; Alatsathianos, I. Breast cancer statistics in the european union: Incidence and survival across european countries. Breast Care 2019, 14, 344–353. [CrossRef] 3. Body, J.J.; Quinn, G.; Talbot, S.; Booth, E.; Demonty, G.; Taylor, A.; Amelio, J. Systematic review and meta-analysis on the proportion of patients with breast cancer who develop bone metastases. Crit. Rev. Oncol. Hematol. 2017, 115, 67–80. [CrossRef][PubMed] 4. Maurizi, A.; Rucci, N. The Osteoclast in Bone Metastasis: Player and Target. Cancers 2018, 10.[CrossRef] [PubMed] 5. Coleman, R.; Gnant, M.; Morgan, G.; Clezardin, P. Effects of bone-targeted agents on cancer progression and mortality. J. Natl. Cancer Inst. 2012, 104, 1059–1067. [CrossRef][PubMed] 6. Fidler, I.J.; Ellis, L.M. The implications of angiogenesis for the biology and therapy of cancer metastasis. Cell 1994, 79, 185–188. [CrossRef] 7. Langley, R.R.; Fidler, I.J. The seed and soil hypothesis revisited–the role of tumor-stroma interactions in metastasis to different organs. Int. J. Cancer. J. Int. Du Cancer 2011, 128, 2527–2535. [CrossRef] 8. Kang, Y.; Siegel, P.M.; Shu, W.; Drobnjak, M.; Kakonen, S.M.; Cordón-Cardo, C.; Guise, T.A.; Massagué, J. A multigenic program mediating breast cancer metastasis to bone. Cancer Cell 2003, 3, 537–549. [CrossRef] 9. Xia, T.S.; Wang, J.; Yin, H.; Ding, Q.; Zhang, Y.F.; Yang, H.W.; Liu, X.A.; Dong, M.; Du, Q.; Ling, L.J.; et al. Human tissue-specific microenvironment: An essential requirement for mouse models of breast cancer. Oncol. Rep. 2010, 24, 203–211. [CrossRef] 10. Sasaki, S.; Baba, T.; Nishimura, T.; Hayakawa, Y.; Hashimoto, S.; Gotoh, N.; Mukaida, N. Essential roles of the interaction between cancer cell-derived chemokine, CCL4, and intra-bone CCR5-expressing fibroblasts in breast cancer bone metastasis. Cancer Lett. 2016, 378, 23–32. [CrossRef] 11. Vishal, M.; Swetha, R.; Thejaswini, G.; Arumugam, B.; Selvamurugan, N. Role of Runx2 in breast cancer-mediated bone metastasis. Int. J. Biol. Macromol. 2017, 99, 608–614. [CrossRef][PubMed] 12. Davudian, S.; Mansoori, B.; Shajari, N.; Mohammadi, A.; Baradaran, B. BACH1, the master regulator gene: A novel candidate target for cancer therapy. Gene 2016, 588, 30–37. [CrossRef][PubMed] 13. Gasiorek, J.J.; Blank, V. Regulation and function of the NFE2 transcription factor in hematopoietic and non-hematopoietic cells. Cell. Mol. Life Sci. 2015, 72, 2323–2335. [CrossRef][PubMed] Cancers 2020, 12, 3003 16 of 18

14. Kaufmann, K.B.; Grunder, A.; Hadlich, T.; Wehrle, J.; Gothwal, M.; Bogeska, R.; Seeger, T.S.; Kayser, S.; Pham, K.B.; Jutzi, J.S.; et al. A novel murine model of myeloproliferative disorders generated by overexpression of the transcription factor NF-E2. J. Exp. Med. 2012, 209, 35–50. [CrossRef] 15. Jutzi, J.S.; Bogeska, R.; Nikoloski, G.; Schmid, C.A.; Seeger, T.S.; Stegelmann, F.; Schwemmers, S.; Grunder, A.; Peeken, J.C.; Gothwal, M.; et al. MPN patients harbor recurrent truncating mutations in transcription factor NF-E2. J. Exp. Med. 2013, 210, 1003–1019. [CrossRef] 16. Peeken, J.C.; Jutzi, J.S.; Wehrle, J.; Koellerer, C.; Staehle, H.F.; Becker, H.; Schoenwandt, E.; Seeger, T.S.; Schanne, D.H.; Gothwal, M.; et al. Epigenetic regulation of NFE2 overexpression in myeloproliferative neoplasms. Blood 2018, 131, 2065–2073. [CrossRef] 17. Jutzi, J.S.; Basu, T.; Pellmann, M.; Kaiser, S.; Steinemann, D.; Sanders, M.A.; Hinai, A.S.A.; Zeilemaker, A.; Bojtine Kovacs, S.; Koellerer, C.; et al. Altered NFE2 activity predisposes to leukemic transformation and myelosarcoma with AML-specific aberrations. Blood 2019, 133, 1766–1777. [CrossRef] 18. Spencer, J.A.; Ferraro, F.; Roussakis, E.; Klein, A.; Wu, J.; Runnels, J.M.; Zaher, W.; Mortensen, L.J.; Alt, C.; Turcotte, R.; et al. Direct measurement of local oxygen concentration in the bone marrow of live animals. Nature 2014, 508, 269–273. [CrossRef] 19. Weilbaecher, K.N.; Guise, T.A.; McCauley, L.K. Cancer to bone: A fatal attraction. Nat. Rev. Cancer 2011, 11, 411–425. [CrossRef] 20. Li, X.Q.; Lu, J.T.; Tan, C.C.; Wang, Q.S.; Feng, Y.M. RUNX2 promotes breast cancer bone metastasis by increasing integrin α5-mediated colonization. Cancer Lett. 2016, 380, 78–86. [CrossRef] 21. Kim, B.; Kim, H.; Jung, S.; Moon, A.; Noh, D.Y.; Lee, Z.H.; Kim, H.J.; Kim, H.H. A CTGF-RUNX2-RANKL Axis in Breast and Prostate Cancer Cells Promotes Tumor Progression in Bone. J. Bone Miner. Res. 2020, 35, 155–166. [CrossRef][PubMed] 22. Mignotte, V.; Wall, L.; Deboer, E.; Grosveld, F.; Romeo, P.H. Two tissue-specific factors bind the erythroid promoter of the human porphobilinogen deaminase gene. Nucleic Acids Res. 1989, 17, 37–54. [CrossRef] [PubMed] 23. Andrews, N.C.; Kotkow, K.J.; Ney, P.A.; Erdjument-Bromage, H.; Tempst, P.; Orkin, S.H. The ubiquitous subunit of erythroid transcription factor NF-E2 is a small basic-leucine zipper protein related to the v-maf oncogene. Proc. Natl. Acad. Sci. USA 1993, 90, 11488–11492. [CrossRef][PubMed] 24. Andrews, N.C.; Erdjument-Bromage, H.; Davidson, M.B.; Tempst, P.; Orkin, S.H. Erythroid transcription factor NF-E2 is a haematopoietic-specific basic-leucine zipper protein. Nature 1993, 362, 722–728. [CrossRef] 25. Mignotte, V.; Eleouet, J.F.; Raich, N.; Romeo, P.H. Cis- and trans-acting elements involved in the regulation of the erythroid promoter of the human porphobilinogen deaminase gene. Proc. Natl. Acad. Sci. USA 1989, 86, 6548–6552. [CrossRef] 26. Loyd, M.R.; Okamoto, Y.; Randall, M.S.; Ney, P.A.Role of AP1/NFE2 binding sites in endogenous alpha-globin gene transcription. Blood 2003, 102, 4223–4228. [CrossRef] 27. Onishi, Y.; Kiyama, R. Interaction of NF-E2 in the human β-globin locus control region before chromatin remodeling. J. Biol. Chem. 2003, 278, 8163–8171. [CrossRef] 28. Shivdasani, R.A.; Orkin, S.H. Erythropoiesis and globin gene expression in mice lacking the transcription factor NF-E2. Proc. Natl. Acad. Sci. USA 1995, 92, 8690–8694. [CrossRef] 29. Lecine, P.; Villeval, J.L.; Vyas, P.; Swencki, B.; Xu, Y.; Shivdasani, R.A. Mice lacking transcription factor NF-E2 provide in vivo validation of the proplatelet model of thrombocytopoiesis and show a platelet production defect that is intrinsic to megakaryocytes. Blood 1998, 92, 1608–1616. [CrossRef] 30. Ney, P.A.; Andrews, N.C.; Jane, S.M.; Safer, B.; Purucker, M.E.; Weremowicz, S.; Morton, C.C.; Goff, S.C.; Orkin, S.H.; Nienhuis, A.W. Purification of the human NF-E2 complex: cDNA cloning of the hematopoietic cell-specific subunit and evidence for an associated partner. Mol. Cell Biol. 1993, 13, 5604–5612. [CrossRef] 31. Toki, T.; Itoh, J.; Arai, K.; Kitazawa, J.; Yokoyama, M.; Igarashi, K.; Yamamoto, M.; Ito, E. Abundant expression of erythroid transcription factor P45 NF-E2 mRNA in human peripheral granurocytes. Biochem. Biophys. Res. Commun. 1996, 219, 760–765. [CrossRef] 32. Zhong, Z.; Yu, J.; Virshup, D.M.; Madan, B. Wnts and the hallmarks of cancer. Cancer Metastasis Rev. 2020. [CrossRef] 33. Taguchi, K.; Yamamoto, M. The KEAP1-NRF2 System in Cancer. Front. Oncol. 2017, 7, 85. [CrossRef] Cancers 2020, 12, 3003 17 of 18

34. Ohta, T.; Iijima, K.; Miyamoto, M.; Nakahara, I.; Tanaka, H.; Ohtsuji, M.; Suzuki, T.; Kobayashi, A.; Yokota, J.; Sakiyama, T.; et al. Loss of Keap1 function activates Nrf2 and provides advantages for lung cancer cell growth. Cancer Res. 2008, 68, 1303–1309. [CrossRef] 35. Moi, P.; Chan, K.; Asunis, I.; Cao, A.; Kan, Y.W. Isolation of NF-E2-related factor 2 (Nrf2), a NF-E2-like basic leucine zipper transcriptional activator that binds to the tandem NF-E2/AP1 repeat of the β-globin locus control region. Proc. Natl. Acad. Sci. USA 1994, 91, 9926–9930. [CrossRef] 36. Itoh, K.; Chiba, T.; Takahashi, S.; Ishii, T.; Igarashi, K.; Katoh, Y.; Oyake, T.; Hayashi, N.; Satoh, K.; Hatayama, I.; et al. An Nrf2/small Maf heterodimer mediates the induction of phase II detoxifying enzyme genes through antioxidant response elements. Biochem. Biophys. Res. Commun. 1997, 236, 313–322. [CrossRef] 37. Hirotsu, Y.; Katsuoka, F.; Funayama, R.; Nagashima, T.; Nishida, Y.; Nakayama, K.; Douglas Engel, J.; Yamamoto, M. Nrf2-MafG heterodimers contribute globally to antioxidant and metabolic networks. Nucleic Acids Res. 2012, 40, 10228–10239. [CrossRef] 38. Gong, X.; Yi, J.; Carmon, K.S.; Crumbley, C.A.; Xiong, W.; Thomas, A.; Fan, X.; Guo, S.; An, Z.; Chang, J.T.; et al. Aberrant RSPO3-LGR4 signaling in Keap1-deficient lung adenocarcinomas promotes tumor aggressiveness. Oncogene 2015, 34, 4692–4701. [CrossRef] 39. Manigandan, K.; Manimaran, D.; Jayaraj, R.L.; Elangovan, N.; Dhivya, V.; Kaphle, A. Taxifolin curbs NF-κB-mediated Wnt/β-catenin signaling via up-regulating Nrf2 pathway in experimental colon carcinogenesis. Biochimie 2015, 119, 103–112. [CrossRef] 40. Mignotte, V.; Navarro, S.; Eleouet, J.F.; Zon, L.I.; Romeo, P.H. The extinction of erythroid genes after tetradecanoylphorbol acetate treatment of erythroleukemic cells correlates with down-regulation of the tissue-specific factors NF-E1 and NF-E2. J. Biol. Chem. 1990, 265, 22090–22092. 41. Cosphiadi, I.; Atmakusumah, T.D.; Siregar, N.C.; Muthalib, A.; Harahap, A.; Mansyur, M. Bone Metastasis in Advanced Breast Cancer: Analysis of Gene Expression Microarray. Clin. Breast Cancer 2018, 18, e1117–e1122. [CrossRef] 42. Singh, A.; Venkannagari, S.; Oh, K.H.; Zhang, Y.Q.; Rohde, J.M.; Liu, L.; Nimmagadda, S.; Sudini, K.; Brimacombe, K.R.; Gajghate, S.; et al. Small molecule inhibitor of NRF2 selectively intervenes therapeutic resistance in KEAP1-deficient NSCLC tumors. ACS Chem. Biol. 2016, 11, 3214–3225. [CrossRef] 43. Pratap, J.; Wixted, J.J.; Gaur, T.; Zaidi, S.K.; Dobson, J.; Gokul, K.D.; Hussain, S.; Van Wijnen, A.J.; Stein, J.L.; Stein, G.S.; et al. Runx2 transcriptional activation of Indian Hedgehog and a downstream bone metastatic pathway in breast cancer cells. Cancer Res. 2008, 68, 7795–7802. [CrossRef] 44. Glembotsky, A.C.; Bluteau, D.; Espasandin, Y.R.; Goette, N.P.; Marta, R.F.; Marin Oyarzun, C.P.; Korin, L.; Lev, P.R.; Laguens, R.P.; Molinas, F.C.; et al. Mechanisms underlying platelet function defect in a pedigree with familial platelet disorder with a predisposition to acute myelogenous leukemia: Potential role for candidate RUNX1 targets. J. Thromb. Haemost. 2014, 12, 761–772. [CrossRef] 45. Lee, W.H.; Chung, M.H.; Tsai, Y.H.; Chang, J.L.; Huang, H.M. Interferon-γ suppresses activin A/NF-E2 induction of erythroid gene expression through the NF-κB/c-Jun pathway. Am. J. Physiol. Cell Physiol. 2014, 306, C407–C414. [CrossRef] 46. Chui, C.M.; Li, K.; Yang, M.; Chuen, C.K.; Fok, T.F.; Li, C.K.; Yuen, P.M. Platelet-derived growth factor up-regulates the expression of transcription factors NF-E2, GATA-1 and c-Fos in megakaryocytic cell lines. Cytokine 2003, 21, 51–64. [CrossRef] 47. Lollini, P.L.; de Giovanni, C.; Eusebi, V.; Nicoletti, G.; Prodi, G.; Nanni, P. High-metastatic clones selected in vitro from a recent spontaneous BALB/c mammary adenocarcinoma cell line. Clin. Exp. Metastasis 1984, 2, 251–259. [CrossRef] 48. Mootha, V.K.; Lindgren, C.M.; Eriksson, K.F.; Subramanian, A.; Sihag, S.; Lehar, J.; Puigserver, P.; Carlsson, E.; Ridderstråle, M.; Laurila, E.; et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat. Genet. 2003, 34, 267–273. [CrossRef] 49. Reich, M.; Liefeld, T.; Gould, J.; Lerner, J.; Tamayo, P.; Mesirov, J.P. GenePattern 2.0. Nat. Genet. 2006, 38, 500–501. [CrossRef] 50. Subramanian, A.; Tamayo, P.; Mootha, V.K.; Mukherjee, S.; Ebert, B.L.; Gillette, M.A.; Paulovich, A.; Pomeroy, S.L.; Golub, T.R.; Lander, E.S.; et al. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. USA 2005, 102, 15545–15550. [CrossRef] Cancers 2020, 12, 3003 18 of 18

51. Baba, T.; Naka, K.; Morishita, S.; Komatsu, N.; Hirao, A.; Mukaida, N. MIP-1alpha/CCL3-mediated maintenance of leukemia-initiating cells in the initiation process of chronic myeloid leukemia. J. Exp. Med. 2013, 210, 2661–2673. [CrossRef] 52. Sasaki, S.; Baba, T.; Shinagawa, K.; Matsushima, K.; Mukaida, N. Crucial involvement of the CCL3-CCR5 axis-mediated ﬁbroblast accumulation in colitis-associated carcinogenesis in mice. Int. J. Cancer 2014, 135, 1297–1306. [CrossRef]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).