-
[CANCER RESEARCH 61, 261–269, January 1, 2001]
A Role for CCAAT/Enhancer Binding Protein b-Liver-enriched
Inhibitory Proteinin Mammary Epithelial Cell Proliferation 1
Cynthia A. Zahnow,2 Robert D. Cardiff, Rodolfo Laucirica, Daniel
Medina, and Jeffrey M. RosenDepartments of Cell Biology [C. A. Z.,
D. M., J. M. R.] and Pathology [R. L.] and The Methodist Hospital
[R. L.], Baylor College of Medicine, Houston, Texas 77030, and
Centerfor Comparative Medicine, University of California at Davis,
Davis, California 95616 [R. D. C.]
ABSTRACT
The transcription factor, CCAAT/enhancer binding protein
b(C/EBPb), regulates the expression of genes involved in
proliferation andterminal differentiation. Dimerization of the
dominant-negative C/EBPb-liver-enriched inhibitory protein (LIP)
isoform with the C/EBP b-liver-enriched activating protein (LAP)
isoform inhibits the transcriptionalactivation of genes involved in
differentiation. Consequently, an increasein LIP levels may inhibit
terminal differentiation and lead to proliferation.C/EBPb-LIP and
LAP are crucial for mammary gland development(G. W. Robinsonet
al.,Genes Dev.,12: 1907–1916, 1998; T. N. Seagroveset al., Genes
Dev.,12: 1917–1928, 1998) and are also overexpressed inbreast
cancer (B. Raughtet al., Cancer Res.,56: 4382–4386. 1996; C.
A.Zahnow et al.,J. Natl. Cancer Inst., 89: 1887–1891, 1997);
however, littleis known about how these isoforms differentially
regulate cell cycle pro-gression. To address this question,
C/EBPb-LIP was overexpressed inboth the mammary glands of
transgenic mice and in cultured TM3mammary epithelial cells. Here
we report that the involuted mammaryglands from transgenic mice
overexpressing C/EBPb-LIP contain bothfocal and diffuse alveolar
hyperplasia and, less frequently, contain mam-mary intraepithelial
neoplasias (high grade) and invasive and noninvasivecarcinomas.
Likewise, cultured TM3 cells, stably expressing C/EBPb-LIP,showed
an increase in proliferation and foci formation attributable to
areentry into S-phase during cellular confluence. These results
demonstratethat C/EBPb-LIP can induce epithelial proliferation and
the formation ofmammary hyperplasias and suggest that a
C/EBPb-LIP-initiated growthcascade may be susceptible to additional
oncogenic hits, which could resultin the initiation and progression
of neoplasia.
INTRODUCTION
Although a majority of breast cancer research has focused
onstudies of advanced tumors and metastases, the molecular
mecha-nisms responsible for the regulation of normal mammary gland
de-velopment and the initiation of premalignant disease are still
not wellunderstood. Breast cancer originates primarily in the
normal mam-mary epithelium of the terminal ducts and has been
hypothesized toinvolve the clonal expansion of an initiated cell
into an epithelialhyperplasia prior to local invasion of the
mammary stroma. Themolecular changes that occur during this
progression include theamplification and/or overexpression of
transcription factors, growthfactors, and growth factor receptors
or the silencing of tumor sup-pressor genes, which can then act to
disrupt the delicate balancebetween cellular proliferation,
terminal differentiation, and pro-grammed cell death. C/EBP3b is
one such transcription factor, which
has been implicated in cell cycle regulation and plays an
importantrole in mammary gland proliferation and
differentiation.
C/EBPb is a member of the C/EBP family of transcription
factorsthat bind to specific DNA sequences as homo- and
heterodimers andaffect the transcription of target genes involved
in proliferation anddifferentiation. Six C/EBP genes have thus far
been identified(C/EBPa, C/EBPb, C/EBPg, C/EBPd, C/EBPe, andC/EBPz),
and allof the genes are intronless except forC/EBPe andC/EBPz.
Transcrip-tion of C/EBPb results in a single mRNA that can be
translated intofour isoforms: LAP (full-length LAP-Mr 38,000 and
LAP-Mr 35,000);LIP (LIP-Mr 20,000); and a smallerMr 16,000 isoform.
The predom-inant isoforms expressed in the mouse mammary gland are
theMr35,000 andMr 20,000 family members. Several different
mechanismshave been described to account for the differential
expression of theC/EBPb isoforms: (a) a leaky ribosome scanning
mechanism (1); (b)the interaction of a CUG repeat binding protein
(CUGBP1) with the59 region of C/EBPb mRNA (2); (c) a mechanism
involving theevolutionary conserved upstream open reading frame of
the 59 regionof C/EBPb mRNA and the eukaryotic translation
initiation factorseIF-2a and eIF-4E (3); and (d) specific
proteolytic cleavage in hema-topoietic progenitor cells present in
mouse liver (4). All C/EBPbfamily members share a strong homology
in the COOH-terminal,leucine-rich dimerization domain (bZIP) and
the DNA-binding basicregion. The truncated C/EBPb-LIP isoform,
translated from the thirdAUG, lacks most of thetrans-activation
domain and can, therefore,dimerize and bind to DNA but is unable to
activate gene transcription.Because of an increased DNA affinity of
the C/EBPb-LIP isoform,this inhibition of transcriptional activity
can occur even at substoi-chiometric ratios of LIP:C/EBP, thereby
suggesting a dominant-negative function for C/EBPb-LIP (1). Thus,
the LAP:LIP ratio,rather than their absolute amounts, may be an
important indicator oftranscriptional activity by C/EBPb.
Dimerization of bZIP proteins canoccur in the absence of DNA but is
a prerequisite for DNA binding(5). Additionally, dimers of bZIP
proteins are usually unstable whennot bound to DNA and will rapidly
dissociate back to monomers (6).
C/EBPb is vital for development of the mouse mammary gland
(7,8). As demonstrated in the C/EBPb knockout mouse, mammaryglands
contain enlarged, undeveloped ducts that have a low prolifer-ative
rate and decreased tertiary branching. C/EBPb-LAP expressionis
detectable throughout murine mammary gland development and isin
contrast to C/EBPb-LIP expression levels, which are highest
duringpregnancy (proliferative state) and reduced in the virgin
(mice,4months of age) gland and lactating gland (8, 9). TheC/EBPa
andC/EBPd genes are also expressed in the murine mammary
gland.Although C/EBPa mRNA is expressed throughout mammary
devel-opment, C/EBPa is not essential for mouse mammary gland
devel-opment (8). Additionally, knockout mice have been generated
forC/EBPd (10), but a mammary gland phenotype has not been
reported.Nevertheless, the C/EBPd transcript is overexpressed
during involu-tion of the mouse mammary gland (11, 12), and cell
culture studies
Received 6/23/00; accepted 10/26/00.The costs of publication of
this article were defrayed in part by the payment of page
charges. This article must therefore be hereby
markedadvertisementin accordance with18 U.S.C. Section 1734 solely
to indicate this fact.
1 This research was supported by Grant CA 16303 from the
National Cancer Institute(to J. M. R.), Grant JB-0014 from the
State of California Breast Cancer Research Program(to R. D. C.),
and Contract DAMD17-96-1-6086, a postdoctoral fellowship from
theDepartment of Defense (to C. A. Z.).
2 To whom requests for reprints should be addressed, at Johns
Hopkins ComprehensiveCancer Center, Room 542, 1650 Orleans Street,
Baltimore, MD 21231.
3 The abbreviations used are: C/EBP, CCAAT/enhancer binding
protein; WAP, wheyacidic protein; LAP, liver-enriched activating
protein; LIP, liver-enriched inhibitoryprotein; MTS,
3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxy-phenyl)-2-(4-sulfonyl)-2H-tetrazolium;
BrdUrd, bromodeoxyuridine; FACS, fluorescence-activated cell
sorter;
MIN, mammary intra-epithelial neoplasia; HAN, hyperplastic
alveolar nodule; CMV,cytomegalovirus; Rb, retinoblastoma; HOG,
hyperplastic outgrowth.
261
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
have determined that its predominant role in mammary gland
devel-opment is in growth arrest of mammary epithelial cells (13,
14).
Consequently, the differential expression pattern of the
C/EBPbisoforms suggests a dual and opposing role in mammary gland
devel-opment and the importance of the LAP:LIP ratio as a cell
cycleswitch, resulting either in cellular differentiation or
proliferation.Although C/EPBb-LIP is also overexpressed in breast
cancer and isassociated with biological predictors of poor
survival, such as loss ofestrogen and progesterone receptor
expression, increased cellular pro-liferation, aneuploidy, and poor
histological and nuclear grades (15),its role in tumorigenesis is
unknown. Taken together, these observa-tions have led to the
hypothesis that overexpression of the C/EBPb-LIP isoform in the
mammary gland can result in epithelial cellproliferation that may
render the mammary gland more susceptible toadditional oncogenic
hits, resulting in the initiation and progression ofneoplasia.
Persistent, aberrant expression of the C/EBPb-LIP isoformin these
neoplasms may contribute to an increased growth rate andresult in a
more proliferative or aggressive tumor. To test this hypoth-esis,
complementary approaches have been used to study the
overex-pression of C/EBPb-LIP in both transgenic mice and
mammaryepithelial cell lines. Our studies have demonstrated that
C/EBPb-LIPoverexpression is associated with increased epithelial
proliferation,resulting in mammary hyperplasias and the stochastic
development ofinfrequent carcinomas.
MATERIALS AND METHODS
Transfection and Maintenance of TM3 Cells. TM3 cells were grown
andmaintained using HEPES buffered DMEM/F-12 growth medium
containing2% fetal bovine serum, 10mg/ml insulin, L-glutamine, 5
ng/ml epidermalgrowth factor, and 5mg/ml gentamicin sulfate (16).
At 20–40% confluence,cells were stably transfected with pCIneo-LIP
or pCIneo (as control) usingSuperfect (Qiagen). Stably transfected
cells were cloned using cloning cylin-ders (PGC Scientifics) and
maintained with 0.2 mg of G418 per ml growthmedia.
MTS Cell Proliferation Assay. Five independent LIP clones and
fiveindependent neomycin control clones were plated in
quadruplicate into a96-well tissue culture format at a density of
23 105 cells/well. The number ofviable or proliferative cells was
determined for days 1, 3, 5, 7, 9, and 12 ofculture using the
CellTiter 96 Aqueous Non-Radioactive Cell ProliferationAssay from
Promega, according to the manufacturer’s instructions.
Afterapplication of the MTS reagent, the cells were incubated for
2–3 h at 37°C, andabsorbance at 490 nm was measured using a Dynex
Technologies ELISA platereader. Data were plotted as fold change in
growth rate.
BrdUrd Staining for FACS Analysis. Two independent LIP clones
andtwo independent neomycin control clones were plated into 100-mm
tissueculture dishes at a density of 13 106 cells/plate. At days 3,
7, 10, and 15 ofculture, the cells were pulse labeled for 15 min
with 10mM BrdUrd (AmershamLife Science), washed with Hanks’ medium,
and removed from the plate usingthe enzyme Dispase II (Boehringer
Mannheim). The pellet was resuspended in200 ml of Hanks’ medium,
and the cells were fixed by the addition of 70%ethanol while
vortexing to avoid cell clumping. Cells were stored at 4°C in70%
ethanol until collection of the last time point. Approximately 43
106
cells were removed from the initial pellet and incubated for 10
min with 3 mlof pepsin (0.04% in 0.10N HCl) on a rocker at room
temperature. Aftercentrifugation (1200 rpm for 5 min), the pepsin
supernatant was aspirated, 3 mlof 2 N HCl were added to a vortexed
pellet, and the mixture was incubated for20 min at 37°C. After
incubation, 6 ml of 0.1M sodium borate were addedwhile vortexing,
and the cells were pelleted. After aspiration of the superna-tant,
6 ml of PBST-B (PBS with 0.5% Tween 20, 0.5% BSA) were added
whilevortexing, the cells were pelleted, the supernatant was
aspirated, and 1 ml ofPBS containing 1 unit of DNase-free RNase was
added and incubated for 30min at 37°C. The nuclei were again
pelleted, the supernatant was removed, and20 ml of anti-BrdUrd FITC
and 100ml of PBST were added. The nuclei wereincubated for 1 h in
the dark at room temperature, and 3 ml of PBST-B wereadded while
vortexing, the nuclei were pelleted, the supernatant was
aspirated,
and propidium iodide (Sigma) was added for a final concentration
of 5 or 10mg/ml in PBST-B. Nuclei were stored at 4°C overnight and
examined 1 daylater by FACS analysis.
Active Caspase-3 Determination.Active-caspase-3 levels were
deter-mined both by a fluorogenic assay and FACS analysis. TM3
cells stablyexpressing either PCI-neo-LIP or PCI-neo as control
were plated at an equaldensity of 1.53 106/100-mm plate and
cultured for 3, 7, 10, and 15 days. Ateach time point, cells were
harvested and processed by two methods: scrapingand freezing of the
cell pellet for the fluorogenic assay; or digestion withDispase
(Boehringer Mannheim), followed by fixation with 4%
paraformal-dehyde for FACS analysis in the Flow Cytometry Core Lab
(Baylor College ofMedicine, Houston, TX). Active caspase-3 was
determined using either Ac-DEVD-AMC Caspase-3 (CPP32) fluorogenic
substrate or phycoerythrin-con-jugated polyclonal rabbit
anti-active caspase-3 according to the manufacturer’sinstructions
(PharMingen).
Plasmid Construction: WAP-LIP-WAP. The first step in the
generationof this construct was theEcoRI linearization and Klenow
fill-in of a pBlue-script SKII(1) plasmid containing 843 bp of rat
WAP 39sequence, with aportion of the third exon, the third intron,
all of the fourth exon, and 70 bp of39 flanking DNA. The second
step included the removal of;865 bp of anNcoI/XhoI cDNA fragment
(LIP) from the COOH-terminal region of ratC/EBPb (MSV/C/EBPb,
kindly provided by Dr. S. McKnight, University ofTexas
Southwestern, Dallas, TX). This cDNA insert contains only the
thirdtranslation initiation Met codon and encodes a full-length
protein for LIP andnot the LAP isoforms. After fill-in with Klenow,
the LIP cDNA fragment wasligated to a position immediately 59to the
39 WAP sequence in pSCPTSKII(1). In the third step, the LIP-WAP
39construct was excised using bothKpnI andSpeI, filled-in with
Klenow, andXbaI linkers were attached. ThisLIP-WAP39 fragment was
then ligated to anXbaI-linearized WAP 59frag-ment, which consists
of 982 bp of a rat WAP 59 promoter fragment (2949 to11) and WAP
59untranslated region (from11 to 133). The integrity of
theWAP-LIP-WAP construct was confirmed by sequencing the WAP-LIP
bound-aries. The WAP-LIP-WAP construct was removed from pSCPT
SKII(1) bydigestion withBstXI andKpnI producing a vector (2.9 kb)
and insert fragment(2.75 kb) that were similar in size. Further
digestion usingPvuI, which cutsonly the vector, allowed complete
size fractionation and separation usingagarose gel electrophoresis.
The DNA was further purified and concentrated ona silica matrix
(Glassmilk; Geneclean). Transgenic mice (FVB inbred) weregenerated
by the transgenic core facility at Baylor College of Medicine.
Plasmid Construction: PCI-neo-LIP. To construct PCI-neo-LIP,
an865-bp cDNA, which codes only LIP, was excised from the
WAP-LIP-WAPconstruct usingXbaI andEcoRV and directionally cloned
into the PCI-neoplasmid (Promega) at theXbaI andSmaI restriction
sites using T4 DNA ligase.
Analysis of Tail DNA. Hot-start PCR reactions (25ml) were
performedusing a bottom and top mix initially separated by a wax
barrier. The bottommix, containing 1mg of genomic tail DNA, 1 mM
MgCl, 0.2 mM deoxynucle-otide triphosphates, 10% DMSO, and
103Promega thermocycle buffer in afinal volume of 14.5ml, was
heated to 90°C for 10 min to denature the DNAand melt the wax
pellet and then cooled to 4°C. The top mix, containing 12.5pmol of
each primer, 103Promega thermocycle buffer, and 2.5 units of
Taqpolymerase (Promega) in a final volume of 10.5ml, was added to
the top of thehardened wax barrier and allowed to mix with the
bottom reagents by heatingto 94°C. The reaction profile consisted
of 30 cycles of 1 min at 94°C, 2 minat 60°C, and 3 min at 72°C.
After the final cycle, the samples were incubatedat 72°C for an
additional 5 min. Reactions were performed in a DNA thermo-cycler
(Perkin-Elmer). The PCR products were resolved on a 1.5% agarose
gel.The sequences of the synthetic oligonucleotides used in the PCR
reactionswere as follows (59to 39): rWAP11 (F),
ATCAGTCATCACTTGCCTGC-CGCCG; and LIP 1574 (R),
GTGTGTTGCGTCAGTCCCGTGTCCA.
Protein Extraction and Western Blot Analysis. Tissue and/or
cells weredisrupted in RIPA buffer [50 mM Tris-Cl (pH 7.4), 1%
NP40, 0.25% desoxy-cholate, 150 mM NaCl, 1 mM EGTA, and 0.2% SDS]
containing the followingkinase, phosphatase, and protease
inhibitors: 1 mM NaVO3, 1 mM NaF, 1 mMNa2MoO4, 10 nM okadaic acid,
and 1mg/ml each of benzamidine, aprotinin,soybean trypsin
inhibitor, and antipain. Aliquots of these lysates containing100 mg
of protein were electrophoresed on denaturing SDS
12%-polyacryl-amide minigels and then transferred to polyvinylidene
difluoride membranes(Millipore, Bedford, MA) overnight at 75 mA.
Blots were blocked 90 min inTBST [20 mM Tris (pH 7.5), 150 mM NaCl,
and 0.5% Tween 20] containing
262
A ROLE FOR C/EBPb-LIP IN MAMMARY PROLIFERATION
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
3% nonfat dry milk (Carnation, Glendale, CA) and then incubated
for 90 minin this solution containing antibodies (0.5 ng/ml; Santa
Cruz) prepared againstC/EBPb. Blots were washed with TBST (without
milk) three times for 5–10min each, with agitation. Blots were then
incubated for 60 min in blockingsolution containing 200 ng/ml
biotinylated donkey antirabbit immunoglobulin(Amersham, Little
Chalfont, England) and washed. Lastly, blots were incu-bated for 30
min in blocking solution containing 40 ng/ml
streptavidin-horseradish peroxidase (Oncogene Science, Uniondale,
NY) and washed asbefore. Enhanced chemiluminescence (Hyperfilm;
Amersham) and chemifluo-rescence reagents (Storm Fluoroimager;
Molecular Dynamics) were used forvisualization per the
manufacturer’s instructions.
Tissue. Approximately 43 glands from lactating mice (day 10 of
lactation)were examined. Lactating mothers were separated from pups
2 h prior toexcision of the mammary glands to reduce the
variability associated withsuckling and milk stasis. Inguinal
glands were fixed in 10% neutral bufferedformalin for ;6 h,
embedded in paraffin, sectioned at 5mm, deparaffinizedthrough a
graded series of xylenes and alcohols, rehydrated in water,
andstained with H&E. Involuted, inguinal, and thoracic mammary
glands wereremoved and examined from 22 transgenic LIP mice and 14
control mice(nontransgenic siblings and wild-type FVBs), 6–32
months of age. All micewere multiparous, and most had undergone
involution.3 months beforebiopsy; however, a few mice were
permitted to involute for a shorter intervalof 14 days prior to
biopsy. Tissues were processed in a manner identical to thatfor
lactating tissue. Whole-mount analysis was performed as described
bySeagroveset al. (8).
RESULTS
Overexpression of C/EBPb-LIP in Transgenic Mice. Sequencesin the
rat WAP promoter and 39 untranslated region were used tocreate a
transgenic construct, WAP-LIP-WAP (Fig. 1A), which pref-erentially
targets high levels of C/EBPb-LIP expression to the mam-mary gland
in mice, starting at about days 7–10 of pregnancy andextending
throughout lactation (17). Seven WAP-LIP-WAP foundermice were
generated. One female founder did not produce live off-spring, and
the other three female founders were mosaic and did notexpress
theLIP transgene in their mammary glands. The six remain-ing
founders were bred further, and transgene expression was detectedin
lactating glands of the F1 generation from three of the six
founderlines (6067, 6074, and 6070) by Western blot analysis (Fig.
1B) andby reverse transcription-PCR (data not shown).
Unfortunately, be-cause of limitations of the currently available
C/EBPb antibodies,transgenic LIP expression could not be detected
via immunocyto-chemistry for two reasons: (a) the antibody
recognizes the COOHterminus and cannot distinguish between the
C/EBPb-LIP and LAPisoforms; and (b) the antibody cannot
discriminate between endoge-
nous mouse C/EBPb-LIP and transgenic rat C/EBPb-LIP because
theproteins are.98% similar in amino acid identity. The level
oftransgene expression was relatively constant in subsequent
genera-tions, as evidenced by the similar levels of transgene
expressionobserved in the F1 as well as in the F5 generation (data
not shown).Although the construct is not epitope-tagged, the
transgenic C/EBPb-LIP protein is developmentally distinguishable
from endogenousC/EBPb-LIP during lactation, because the native LIP
isoform is notexpressed during lactation and is primarily expressed
during preg-nancy (Fig. 1B, FVB lane). Consequently, it was
hypothesized thatany phenotypic effects resulting from the
overexpression of C/EBPb-LIP would, therefore, be most readily
detected during lactation andsubsequent involution.
The mammary glands from;22 C/EBPb-LIP transgenic mice and21
control mice (nontransgenic siblings and wild-type FVBs)
corre-sponding to days 1–18 of lactation were examined at both the
grossand microscopic levels. C/EBPb-LIP transgenic mice did not
experi-ence any difficulties in nursing their pups, and no
histological abnor-malities were observed in the mammary glands of
lactating mice.Next, involuted glands from 22 transgenic and 14
control mice (non-transgenic siblings and wild-type FVBs), 6–32
months of age, wereexamined for abnormalities. Mammary gland
neoplasia was observedin 9% (2 of 22) of transgenic mice and
included two invasive carci-nomas (Fig. 2A) and three MINs (high
grade; Fig. 2B) from one27-month-old mouse. MINs comprise a variety
of intraluminal epi-thelial proliferations with atypical cytology,
includingin situ carci-nomas (18). Additionally, the gland from a
20-month-old mousecontained a highly proliferative, poorly
differentiated carcinoma (Fig.2D), and the gland contralateral to
the tumor contained diffuse alve-olar hyperplasia (Fig. 2C). A more
thorough, blind examination (bytwo independent researchers, R. D.
C. and D. M.) of a subset of the 22involuted glands revealed that
30–40% (3 of 10 or 4 of 10) oftransgenic mice contained two
distinct forms of mammary hyperpla-sias known as focal hyperplastic
alveoli or HAN (Fig. 3,A andB) anddiffuse alveolar hyperplasia
(Fig. 3,C andD). The epithelial cells inthese hyperplasias are
characterized by their large shape and activenuclei with open
chromatin, large nucleoli, and a high rate of mitosis,as evidenced
by an abundance of mitotic figures. In contrast to normalepithelial
cells that have undergone a delayed involution, hyperplasticcells
contain very little lipid and are not actively secreting.
Thedescribed neoplasias and hyperplasias were observed in both the
6067and 6074 lines, and no tumors or hyperplasias were observed
inage-matched, nontransgenic siblings or wild-type FVB mice.
Fig. 1. Structure of the WAP transgenic construct and detection
ofexpression by Western blot analysis.A, the transgene was
constructedusing 949 bp of rat WAP 59noncoding sequence and the
first 33 bp ofthe open reading frame, followed by 865 bp of a
C/EBPb rat cDNAfragment that codes only for the C/EBPb-LIP isoform.
An additional843 bp of rat WAP sequence containing part of exon
III, intron C, andexon IV plus 70 bp of 39flanking DNA was
positioned immediately 39to the cDNA (see “Materials and Methods”
for further details).B,representative Western blot of mammary gland
extracts prepared from(10-day) lactating, F1 generation, female
mice from the followingtransgenic founder lines: 6067, 6074, 6070,
6060, and 6065. Thetransgenic (LIP) construct was detected in three
lines (6067, 6074, and6070), and the endogenous (LAP) isoform was
detectable in all mice.A nontransgenic (FVB) mouse was included in
the Western blot todemonstrate that endogenous C/EBPb-LIP protein
levels are not de-tectable during lactation. Cross-reactive
material (CRM), which isobserved in mammary gland extracts from
C/EBPb knockout mice (8),serves as an internal loading
standard.
263
A ROLE FOR C/EBPb-LIP IN MAMMARY PROLIFERATION
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
Overexpression of C/EBPb-LIP in Cultured Mammary Epithe-lial
Cells. To investigate the molecular mechanisms responsible forthe
proliferation and hyperplasias associated with C/EBPb-LIP
over-expression, cell cycle studies were initiated in cultured
mammaryepithelial cells. Two considerations helped to determine
which mam-mary epithelial cell line was used for testing the
effects of overex-pression of C/EBPb-LIP on cell growth and
tumorigenicity: (a) thatthe endogenous levels of C/EBPb-LIP were
low; and (b) that the cellline should exhibit normal,
nontumorigenic growth patterns. The TMcell lines were established
from hyperplastic alveolar outgrowths,which resulted from thein
vivo transplantation of FSK cell lines (16).These lines are
maintained both in culture (in vitro) and as mammarytransplants,
which grow (in vivo) either as hyperplastic outgrowths ortumors.
TM3 (HOG) is a slow-growing, hyperplastic alveolar out-growth that
is ovarian hormone dependent and infrequently progressesinto tumors
when maintained beyond transplant generation 16 (19). Incontrast,
(low passage,,10) TM3 in vitro cultures do not producesuccessful
outgrowths after transplantation into cleared mammary fat
pads, have low endogenous C/EBPb levels as compared with a
moreneoplastic TM line (TM6; Ref. 20), exhibit a marked dependence
onepidermal growth factor for growth (21), and contain a mutant
p53(Ser
233 ins; Refs. 22 and 23), which has been associated with a
higher
rate (4–7%) of apoptosis (19). Consequently, both the TM3
out-growths and the TM3 cell line fit the necessary criteria for
containingrelatively low levels of endogenous C/EBPb and exhibiting
normal orweakly tumorigenic growth patterns.
The TM3 cell line was stably transfected with either
CMV-drivenPCIneo-LIP or PCIneo (the vector without LIP cDNA insert,
ascontrol), and stable cellular clones were generated. In the
parental,nontransfected cells, endogenous C/EBPb protein levels
were ob-served to change with growth. The expression of the native
LIPisoform was consistently observed to be higher when the cells
wereexponentially growing than when the cells were contact
inhibited orconfluent (Fig. 4,left panel). Consequently, CMV-driven
expressionof C/EBPb-LIP from the PCIneo-LIP construct was easily
detectedduring confluence (Fig. 4,middle panel). Passage number had
little
Fig. 2. The mammary glands from transgenic C/EBPb-LIP mice
develop invasive carcinoma and high-grade MIN.A, representative
photomicrograph (320) of one of two invasivecarcinomas from mouse
(7869, case 1). These H&E-stained lesions contained extensive
fibrosis, infiltrated by small cords of atypical epithelial cells
(seearrow) with large pleomorphicnuclei and scattered mitoses.B,
the same gland from mouse (7869, case 1) also contained three
high-grade MINs. The lesion represented inB (320) consists of
several expanded alveoli(large arrow) filled with hyperchromatic,
atypical nuclei with prominent nucleoli and abnormal mitotic
figures. The oval profiles typically form a cribriform-like pattern
(small arrows),as is often observed in ductal carcinomain situ,and
a dense lymphocytic infiltrate (w) is present in the upper portions
of the micrograph. (CandD, 320). Mouse (case 35) developeda mammary
carcinoma with some foci of squamous metaplasia.C, the gland
contralateral to the tumor and depicts a profile of incomplete
involution or diffuse hyperplasia. The contoursof the residual
alveoli are described as rounded (arrows), as opposed to the
angulated acini present in a normally regressed gland (see Fig. 3,C
andD). Duct ectasia was also commonlyobserved in involuted glands.
The large carcinoma (D) is composed of nests (short arrow) and
cords (long arrow) of very hyperchromatic cells in a dense
connective tissue stroma.The sizes of the cords vary. The tumor
cells have large, pleomorphic nuclei with prominent and multiple
nucleoli but with delicate chromatin. The cytoplasm is amphophilic,
and themitotic rate is very high.
264
A ROLE FOR C/EBPb-LIP IN MAMMARY PROLIFERATION
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
effect on loss or gain of endogenous C/EBPb expression, as shown
bycomparison of the first panel containing the earlier passage
parentalTM3 line and the last panel (Fig. 4), which represents a
late-passageTM3 clonal line expressing only neomycin. Endogenous
C/EBPb-LAP levels were more variable during confluence, but
endogenousC/EBPb-LIP levels were usually low and were never
observed toexceed the expression levels for C/EBPb-LAP during
confluence inTM3 cells.
TM3 clones stably expressing CMV-driven LIP and
vector-only,neomycin controls were randomly chosen and tested for
proliferativepotential using an MTS cell proliferation kit. The
clones displayeddifferent growth rates, but the five clonal lines
expressing LIP were,on average, twice as proliferative as the five
control (Neo) clonal lines(Fig. 5). The fold change in growth was
determined by dividing thenumber of proliferating, viable cells (as
measured by the amount ofabsorbance at 490 nm) at days 12, 9, 7, 5,
and 3 by the value for
proliferation at day 1 for each clone. To determine whether
thisincrease in cell number or growth was attributable to an
increase in thenumber of cells entering S phase, cells were pulse
labeled withBrdUrd at 3, 7, 10, and 15 days of culture and analyzed
by flowcytometric (FACS) analysis (Fig. 6B). The data indicate that
expres-sion of LIP in TM3 cells facilitates entry into S-phase and
DNAsynthesis. Both the LIP-expressing cells and the control cells
exhib-ited similar levels of BrdUrd incorporation during
exponential growth(day 3) and early confluence (day 7); however,
the LIP-expressingcells did not remain contact inhibited, and by
day 15 of culture, atleast 10% of the cells had re-entered the cell
cycle, were proliferating,and formed foci as compared with the
neomycin control cells, whichremained a monolayer (Fig. 6,A andB).
Interestingly, CMV-drivenLIP expression does not coincide,
temporally, with the renewedgrowth and reentry of the LIP clones
into the cell cycle (Fig. 6C).Nuclear LIP expression, as well as
the LIP:LAP ratio, was higher in
Fig. 3. Involution in the mammary glands of transgenic CEBPb-LIP
mice is incomplete and characterized by diffuse and focal
hyperplasia.A, whole-mount analysis of the involutedmammary gland
from mouse 8/98#2, case 15, showing both focal and diffuse
hyperplasia. The focal hyperplasia or HAN is evident as a
grape-like cluster on theright sideof the photo(arrow). H&E
analysis of the HAN (arrow) is presented inB (320). The diffuse
hyperplasia resembles a delayed or incomplete regression. The
residual alveoli are characterized byan abnormally round appearance
as opposed to the more normal, collapsed, and angulated acini
observed after a normal regression.C (320) shows a normal pattern
of mammary glandregression (mouse 2160, case 16) with collapsed and
angulated acini, often containing lipid droplets (arrow;inset,340).
D (320), a diffuse alveolar hyperplasia (mouse 6074 invT,case 19).
The rounded alveoli do not contain lipid (arrow;inset,340) and are
either filled with cells or contain multilayers of epithelial cells
possessing large, active nuclei with anopen chromatin and large
nucleolus (not visible in this magnification).
265
A ROLE FOR C/EBPb-LIP IN MAMMARY PROLIFERATION
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
the LIP-expressing clones during the first week of culture than
duringthe second week (Fig. 6C). It is highly unlikely that the
proliferationobserved in these cells can be attributed to clonal
variation, becausefoci formation was observed in at least four
clonal lines stablyexpressing C/EBPb-LIP and was never observed in
the control cells.In addition, the increase in cellular
proliferation was not accompaniedby a decrease in apoptosis. The
TM3 clones were assayed for changesin caspase-3 activity using two
independent methods (see “Methodsand Materials”), and the levels of
active caspase-3 were found not tobe significantly decreased in the
proliferating LIP-expressing cells ascompared with the control
(Neo) cells (data not shown).
Transplantation of LIP-overexpressing TM 3 Cells into the
FatPads of BALB/c Mice. To determine whether the TM3 cells
main-tained their proliferative growth potentialin vivo, stably
expressingLIP and vector control (Neo) cells (13 106 cells) were
transplantedinto the right and left inguinal mammary fat pads,
respectively, ofvirgin, syngeneic BALB/c mice. An inherent
difficulty in cell line/transplantation experiments is that many
nontransfected, high-passagemammary epithelial cell lines will
spontaneously form tumors aftertransplantation into a cleared
mammary fat pad.4 This may be attrib-utable to the fact that during
immortalization, these cells have lostexpression of p16, p53, or
other cell cycle regulators. Thus, thisexperiment was not designed
to examine the oncogenic capacity ofC/EBPb-LIP but rather to test
the reproducibility of LIP-overexpress-ing cells to proliferate in
the mammary fat padin vivo as well as onplastic. Because clonal
selection of the TM3 cells resulted in higherpassage lines, it was
expected that the CMV-driven LIP expressionmight generate larger
more proliferative tumors with a decreasedlatency. Accordingly,
palpable tumors were detectable in the LIP butnot the Neo
transplants 6 weeks after transplantation. Although notpalpable,
the Neo cells also formed some histologically identifiablesmall
tumors. Examination of the transplanted fat pads, via
H&Estaining of paraffin-embedded sections, revealed that the
LIP-express-ing transplants either grew out into large,
undifferentiated tumors thatcompletely filled the fat pad (four of
seven) or did not grow out at all(three of seven). In contrast, the
vector control (Neo) transplants grewas small palpable tumors
(three of seven) or undifferentiated cellmasses that did not fill
the fat pad but with additional time couldgenerate palpable tumors
(four of seven). Analysis of the transplantsdemonstrated that the
LIP tumors were approximately four timeslarger, as determined by
tumor volume (mm3) and wet weight (g, datanot shown) than the
vector control (Neo) outgrowths and tumors (Fig.7C). The larger
size is suggestive of a more proliferative tumor. Thiswas confirmed
by the detection of 10-fold more mitotic figures in the
LIP than in the control tumors or outgrowths (Fig. 7,A andB).
Thus,evidence from bothin vitro tissue culture andin vivo
transplantationstudies demonstrate that overexpression of
C/EBPb-LIP in mammaryepithelial cells results in increased
proliferation.
DISCUSSION
These results have demonstrated that overexpression of
C/EBPb-LIP in the mammary glands of transgenic mice as well as in
mammaryepithelial cells cultured on plastic results in increased
epithelial cellproliferation. These mammary hyperplasias may,
therefore, be inher-ently more susceptible to additional oncogenic
“hits” resulting in thestochastic formation of infrequent tumors,
as was observed in 9% ofthe C/EBPb-LIP transgenic mice. Expression
of C/EBPb-LIP hasbeen observed in many rodent mammary tumors and
some humanbreast cancers and may increase the number of
proliferative cells,potentially resulting in more highly
proliferative and aggressive tu-mors. Consequently, overexpression
of C/EBPb-LIP may be an im-portant indicator for breast epithelium
at risk for hyperplasia andcancer.
The incidence of hyperplastic and neoplastic lesions in our
WAP-LIP-WAP mice are in agreement with several other published
reportsof genetically engineered mice bearing WAP-driven
transgenes. Forexample, in WAP-stromelysin transgenic mice, 6–24
months of age,20% of mice contained atypical proliferative lesions
and 7.4% devel-oped mammary carcinomas (24). Transgenes driven by
the WAPpromoter are preferentially expressed in alveolar epithelial
cells and,to a lesser extent, in ductal epithelial cells in the
mammary gland (17,25, 26). The rat WAP promoter used in our
transgenic study isminimally active during each estrous cycle in
virgin as well as inmultiparous females but is maximally expressed
starting at day 10 ofpregnancy and extending throughout lactation
(17, 26, 27). Conse-quently, the induction of hyperplasias by
WAP-LIP-WAP probablyoccurs during pregnancy and/or lactation but
may be detectable onlyafter involution, following the regression of
the surrounding normalalveolar epithelium. Attempts to detect an
early, LIP-induced, prolif-erative response during pregnancy were
unsuccessful because it wasdifficult to distinguish small increases
in proliferation from the highlyproliferative background during
pregnancy (data not shown). WAP-driven C/EBPb expression after
involution was not detectable byWestern blots and could not be
localized by immunohistochemistry asdiscussed previously. Thus, the
hyperplasias observed in the involuted
4 D. Medina, personal communication.
Fig. 4. C/EBPb-LIP and LAP levels are elevated in exponentially
growing TM3 cellsbut are decreased as the cells become contact
inhibited during confluence. Western blotanalysis of whole-cell
protein extracts from TM3 cells, which were either
exponentiallygrowing (exp) or confluent and contact inhibited
(conf), is shown. The parental TM3 cellsrepresent the early-passage
cells from which the LIP and Neo clones were derived. LIPand Neo
clones refer to clonal TM3 lines that stably express either
C/EBPb-LIP orneomycin (Neo) as the control.
Fig. 5. Overexpression of LIP causes TM3 cell proliferation. TM3
clones stablyexpressing LIP (n5 5; L1, L3, LL5, L6, and L9) and
vector-only controls (n5 5; N2,N3, N5, N6, and N10) were randomly
chosen, plated at equal density, and tested forproliferative
potential using an MTS cell proliferation kit (Promega). The
various clonesdisplayed different growth rates, but the clones
overexpressing LIP were, on average,twice as proliferative as the
cells without exogenous LIP. The fold change in growth
wasdetermined by dividing the number of proliferating viable cells
(as measured by theamount of absorbance at 490 nm) at days 12, 9,
7, 5, and 3 by the value for proliferationat day 1 for each clone.
Bars, SE.
266
A ROLE FOR C/EBPb-LIP IN MAMMARY PROLIFERATION
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
tissue are either no longer dependent on LIP expression or are
main-tained by the expression of LIP in a small subset of cells,
possibly asa result of limited transgene expression that may occur
during eachestrous cycle. Likewise, in the cell culture studies,
CMV-driven LIPexpression was higher during exponential growth (day
3) and earlyconfluence (day 7) than during days 10 and 15 of
confluence, whenLIP-induced proliferation is evident. Although
these TM3 cells havebeen isolated as subclones that stably express
C/EBPb-LIP, it ispossible that the subset of cells that forms foci
and proliferate duringlate confluence has higher levels of
C/EBPb-LIP expression than theadjacent cells, which are less
proliferative.
Targeted dominant-negative constructs are especially difficult
tooverexpress in transgenic mice, because the transgene has the
poten-
tial to negatively regulate its own promoter. In fact, only one
othertransgenic study thus far has successfully overexpressed a
dominant-negative, C/EBP-related protein (28). Consequently, WAP
regulatorysequences were chosen to target transgene expression to
the mammarygland in this study because these sequences did not
contain anyknown, functional C/EBP consensus sites. However,
subsequent anal-ysis of milk protein gene expression from the
mammary glands ofC/EBPb knockout mice has demonstrated that loss of
C/EBPb candramatically reduce the levels of both WAP mRNA and
protein (7, 8).Similarly, when our WAP-LIP-WAP mice were crossed
withC/EBPb-knockout mice, expression of theLIP transgene was
reducedor was nondetectable (data not shown). Taken together, these
datademonstrate that C/EBPb is indeed important in the regulation
ofWAP, and that autoregulatory effects of LIP may account for
themoderate levels of transgene expression and subtle phenotype
ob-served in these mice. Additionally, variegated or sectored
localizationof gene expression in transgenic mice can also account
for variationsin the level of transgene expression (29). Several
different transgenicmouse studies, including our studies with
WAP-driven transgenes,have demonstrated that cells expressing the
transgene often appear asscattered clusters, leading to a
variegated pattern of gene expression
Fig. 6. Overexpression of C/EBPb-LIP in TM3 mammary epithelial
cells results inincreased growth and foci formation.A, micrographs
(34) of confluent TM3 monolayersgrown on plastic at days 10 and 15
of culture. Note the presence of foci in theLIP-expressing clonal
line (upper panels) and the absence of foci in the control lines
(Neo,neomycin-expressing only;lower panels). The LIP-expressing
cells are also smaller andmore crowded in appearance than the
control cells.B, LIP and Neo clonal lines (n5 2,each) were plated
at equal density, and at days 3, 7, 10, and 15 of culture, the
cells werepulse labeled for 15 min with 10mM BrdUrd, harvested, and
analyzed by FACS analysisfor percentage of BrdUrd incorporation.
Although foci formation was also observed in atleast four other
LIP-expressing clones, the BrdUrd analysis was conducted with only
twoof these clones, and consequently SEs could not be determined.C,
Western blot analysisof cytoplasmic (Lanes C) and nuclear (Lanes N)
extracts from LIP expressing TM3 clone(LIP1) and control clone (Neo
10) at 7 and 15 days of culture. TM6 cells serve as apositive
control (1) for C/EBPb-LIP and LAP, and cross-reactive material
(CRM) isindicated on the blot.
Fig. 7. Transplantation of TM3 cells stably overexpressing
C/EBPb-LIP result in moreproliferative tumors in vivo.
Approximately 1 3 106 TM3 cells stably expressingC/EBPb-LIP were
transplanted into the right, cleared inguinal fat pad of BALB/c
mice,and an equal number of non-LIP-expressing control cells
(Neo/control) were transplantedinto the contralateral gland.
Mammary glands were harvested 6 weeks after transplanta-tion, fixed
overnight in 10% neutral buffered formalin, and processed via
standardmethods for paraffin sectioning.A, H&E micrograph (340)
of a C/EBPb-LIP-expressingtransplant that grew out as a poorly
differentiated tumor. Numerous mitotic figures(boxed) are visible
within this one high-powered field. The tumors derived from
C/EBPb-LIP-expressing cells were approximately four times larger,
as determined by tumorvolume or wet weight (C), and contained 10
times more mitotic figures/10 high-poweredfields than the
outgrowths and smaller tumors derived from the control cells (B).
The totalnumber of mitotic figures/10 high-powered fields (HPF) was
determined by a pathologist(R. L.) as a blind comparative
study.Bars,SE.
267
A ROLE FOR C/EBPb-LIP IN MAMMARY PROLIFERATION
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
(30, 31). This may also account, in part, for the
focalversusdiffusepattern of hyperplasia observed in the
WAP-LIP-WAP mice. Further-more, the timing of transgene expression
may also be an importantfactor because WAP-driven transgenes are
not expressed until pu-berty. If it were possible to selectively
target C/EBPb-LIP expressionto the mammary gland either during
early ductal development or invirgin transgenic mice, one might
expect to observe a very different ormore severe phenotype.
It is generally accepted that breast cancer originates in the
terminalduct lobular unit (32). Although the mouse mammary gland
does notcontain a terminal duct lobular unit, an equivalent
structure would bethe tertiary branches that give rise to the
alveoli. HANs can form inthis region, as was observed in the
WAP-LIP-WAP transgenic mice.HAN is a low-grade, focal alveolar
hyperplasia that persists in theinvoluted mammary gland and has
been experimentally proven viatransplantation experiments to be a
precancerous, clonal lesion withhigh malignant potential (33–35).
Squamous metaplasia, inflamma-tion, or lymphocytic infiltration,
also frequently present in the invo-luted glands of WAP-LIP-WAP
mice, has been proposed to be anormal repair response of the
mammary gland to the hormonal chal-lenges and damage caused by
multiple pregnancies (18).
Numerous reports in tissues other than the mammary gland
supportthe observation that C/EBPb-LIP plays a proliferative role
in cellcycle control. In adipocytes, C/EBPb and C/EBPd have been
shownto induce C/EBPa expression, which arrests the ongoing
proliferationand facilitates terminal cell differentiation (36,
37). Moreover, over-expression of C/EBPb-LIP results in continued
proliferation and isable to inhibit the adipocyte conversion into
the differentiated pheno-type (38). Similarly, a recent study has
demonstrated that the intro-duction of C/EBPb-LIP via retroviral
gene transfer into 3T3-L1 cellsresults in proliferation, foci
formation, and a loss of contact inhibition(3). Although C/EBPa is
primarily responsible for regulating terminaldifferentiation in
hepatocytes (39), cellular proliferation in Hep G2hepatoma cells is
not blocked by C/EBPa expression but is abrogatedby C/EBPb-LAP
(40). In adult hepatocytes, differentiation and pro-liferation are
mutually exclusive (40), and during rat postnatal devel-opment, the
levels of LAP in liver nuclei are elevated much more thanthose of
LIP (1). This is suggestive that the LAP:LIP ratio is importantfor
differential regulation of gene expression and differentiation in
theadult liver. In contrast, during hepatocyte proliferation after
partialhepatectomy, C/EBPa levels decline, but both C/EBPd and
C/EBPblevels increase. In fact, C/EBPa:C/EBPb heterodimers are
replacedwith C/EBPb homodimers during the early G1 period after
partialhepatectomy (41, 42).
C/EBP family members have been historically described as
DNA-binding proteins; however, the C/EBPs are also capable of
protein-protein interactions with cell cycle proteins such as Rb
and p21. TheC/EBPb-LIP and LAP isoforms can directly interact with
the SV40Tantigen domain of hypophosphorylated Rb (43). This
transient butdirect interaction with Rb increases DNA binding and
transactivationpotential of the C/EBPb isoforms, and depending on
the ratio ofLIP:LAP, may inhibit the transactivation potential of
LAP to tran-scribe genes involved in cellular differentiation (43).
Additionally, anin vivo analysis in the liver of C/EBPa knockout
mice showed thatC/EBPa and p21 interact via protein-protein
interactions to stabilizep21 levels (44). At the transcriptional
level, studies in rat hepatomacells have demonstrated that C/EBPa
can bind to the canonical C/EBPDNA binding site in the p21
cyclin-dependent kinase inhibitor gene,resulting in the elevation
of p21 expression, the inhibition of cyclin-dependent
kinase-dependent Rb phosphorylation, and the induction ofcell cycle
arrest at G1 (45–47). Similarly, in human colorectal cancercell
lines, C/EBPb has been shown to increase p21 transcription, butit
was not determined whether the C/EBPb isoforms have opposing
effects on p21 regulation (48). However, in primary cultures
ofkeratinocytes, the deletion of theC/EBPb gene did not alter
expres-sion of p21 (49). Consequently, the regulation of p21 by
C/EBPb maybe a tissue-specific process. Further investigation of
p21 regulation byLIP in mammary epithelial cells is clearly
warranted. These observa-tions are important, because the canonical
C/EBP DNA binding site inthep21gene promoter should be capable of
binding all of the C/EBPs,including C/EBPb. The C/EBPs have
identical binding specificities,and the hierarchy of DNA binding
affinities for the C/EBP consensussequence is C/EBPb . C/EBPa .
C/EBPd (50). If C/EBPb-LIP wereto dimerize with C/EBPa or form
homodimers with itself, the tran-scriptional regulation of p21
might be inhibited, resulting in phospho-rylation of Rb and
progression through the G1-S transition. Thisprovides a potential
mechanism by which C/EBPb-LIP might induceentry into S-phase.
Alternatively, alterations in p21 or other cyclin-dependent
kinaseinhibitors may result in changes in apoptosis. Although no
decreasesin apoptosis were observed in the clonally selected TM3
cells, mod-ulation of the LIP:LAP ratio may result in increased
apoptosis inmammary epithelial cells5 or a rescue from apoptosis by
matrixdetachment in intestinal epithelial cells.6 Effects on
apoptosis may betissue specific and dependent on the amount of LIP
present in thecells. Failure to obtain TM3 clones that highly
express LIP, may bethe result of induction of apoptoin in induction
of apoptosis in theseclones during selection. Thus, LIP expression
may potentially regulatecell proliferation and/or apoptosis,
depending on the cell type andcell-substratum interactions.
In conclusion, these studies indicate that the overexpression
ofC/EBPb-LIP in mammary epithelial cells promotes proliferation
andthe development of hyperplasias. The data also support the
hypothesisthat LIP overexpression may stimulate a growth cascade,
which maybe susceptible to additional oncogenic hits and result in
the stochasticformation of tumors. The elucidation of the molecular
mechanisms bywhich C/EBPb-LIP regulates cell cycle progression may,
therefore,be critical for defining protein targets associated with
premalignancyand neoplastic progression.
ACKNOWLEDGMENTS
We thank Liz Hopkins for assistance with histology, Jeff Scott
for help withFACS analysis, Jason Gay for assistance with surgical
techniques, ShirleySmall for mouse husbandry, and Frances Kittrell
for helpful discussions abouttechniques related to mammary
epithelial cell lines and mice.
REFERENCES
1. Descombes, P., and Schibler, U. A liver-enriched
transcriptional activator protein,LAP, and a transcriptional
inhibitory protein, LIP, are translated from the samemRNA. Cell,
67: 569–579, 1991.
2. Timchenko, N. A., Welm, A. L., Lu, X., and Timchenko, L. T.
CUG repeat bindingprotein (CUGBP1) interacts with the 59region of
C/EBPb mRNA and regulatestranslation of C/EBPb isoforms. Nucleic
Acids Res,27: 4517–4525, 1999.
3. Calkhoven, C. F., Muller, C., and Leutz, A. Translational
control of C/EBPa andC/EBPb isoform expression. Genes Dev.,14:
1920–1932, 2000.
4. Welm, A. L., Timchenko, N. A., and Darlington, G. J. C/EBPa
regulates generationof C/EBPb isoforms through activation of
specific proteolytic cleavage. Mol. Cell.Biol., 19: 1695–1704,
1999.
5. Turner, R., and Tjian, R. Leucine repeats and an adjacent DNA
binding domainmediate the formation of functional cFos-cJun
heterodimers. Science (WashingtonDC), 243: 1689–1694, 1989.
6. Patel, L. R., Curran, T., and Kerppola, T. K. Energy transfer
analysis of Fos-Jundimerization and DNA binding. Proc. Natl. Acad.
Sci. USA,91: 7360–7364, 1994.
7. Robinson, G. W., Johnson, P. F., Hennighausen, L., and
Sterneck, E. The C/EBPbtranscription factor regulates epithelial
cell proliferation and differentiation in themammary gland. Genes
Dev.,12: 1907–1916, 1998.
5 M. Bissell, personal communication.6 J. Brugge, personal
communication.
268
A ROLE FOR C/EBPb-LIP IN MAMMARY PROLIFERATION
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
8. Seagroves, T. N., Krnacik, S., Raught, B., Gay, J.,
Burgess-Beusse, B., Darlington,G. J., and Rosen, J. M. C/EBPb, but
not C/EBPa, is essential for ductal morphogen-esis, lobuloalveolar
proliferation, and functional differentiation in the mouse mam-mary
gland. Genes Dev.,12: 1917–1928, 1998.
9. Raught, B., Liao, W. S. L., and Rosen, J. M. Developmentally-
and hormonally-regulated C/EBP isoforms influenceb-casein gene
expression. Mol. Endocrinol.,9:1223–1232, 1995.
10. Tanaka, T., Yoshida, N., Kishimoto, T., and Akira, S.
Defective adipocyte differen-tiation in mice lacking the C/EBPb
and/or C/EBPd gene. EMBO J.,16: 7432–7443,1997.
11. Gigliotti, A. P., and DeWille, J. W. Lactation status
influences expression of CCAAT/enhancer binding protein isoform
mRNA in the mouse mammary gland. J. Cell.Physiol.,174: 232–239,
1998.
12. Sabatakos, G., Davies, G. E., Grosse, M., Cryer, A., and
Ramji, D. P. Expression ofthe genes encoding CCAAT-enhancer binding
protein isoforms in the mouse mam-mary gland during lactation and
involution. Biochem. J.,334: 205–210, 1998.
13. O’Rourke, J., Yuan, R., and DeWille, J.
CCAAT/enhancer-binding protein-d (C/EBP-d) is induced in growth-
arrested mouse mammary epithelial cells. J. Biol.Chem.,272:
6291–6296, 1997.
14. O’Rourke, J. P., Newbound, G. C., Hutt, J. A., and DeWille,
J. CCAAT/enhancer-binding proteind regulates mammary epithelial
cell G0 growth arrest and apoptosis.J. Biol. Chem.,274:
16582–16589, 1999.
15. Zahnow, C. A., Younes, P., Laucirica, R., and Rosen, J. M.
Overexpression ofC/EBPb-LIP, a naturally occurring, dominant-
negative transcription factor, in humanbreast cancer. J. Natl.
Cancer Inst.,89: 1887–1891, 1997.
16. Kittrell, F. S., Oborn, C. J., and Medina, D. Development of
mammary preneoplasiasin vivo from mouse mammary epithelial cell
linesin vitro. Cancer Res.,52: 1924–1932, 1992.
17. Bayna, E. M., and Rosen, J. M. Tissue-specific, high level
expression of the rat wheyacidic protein gene in transgenic mice.
Nucleic Acids Res.,18: 2977–2985, 1990.
18. Cardiff, R. D., Anver, M. R., Gusterson, B. A.,
Hennighausen, L., Jensen, R. A.,Merino, M. J., Rehm, S., Russo, J.,
Tavassoli, F. A., Wakefield, L. M., Ward, J. M.,and Green, J. E.
The mammary pathology of genetically engineered mice: theconsensus
report and recommendations from the Annapolis meeting.
Oncogene,19:968–988, 2000.
19. Bonnette, S. G., Kittrell, F. S., Stephens, L. C., Meyn, R.
E., and Medina, D.Interactions of apoptosis, proliferation and host
age in the regression of the mousemammary preneoplasia, TM3,
carrying an unusual mutation in p53. Carcinogenesis(Lond.), 20:
1715–1720, 1999.
20. Raught, B., Gingras, A-C., James, A., Medina, D., Sonenberg,
N., and Rosen, J. M.Expression of a translationally regulated,
dominant-negative CCAAT/enhancer-bind-ing proteinb isoform and
up-regulation of the eukaryotic translation initiation factor2a are
correlated with neoplastic transformation of mammary epithelial
cells. CancerRes.,56: 4382–4386, 1996.
21. Medina, D., Kittrell, F. S., Oborn, C. J., and Schwartz, M.
Growth factor dependencyand gene expression in preneoplastic mouse
mammary epithelial cells. Cancer Res.,53: 668–674, 1993.
22. Jerry, D. J., Ozbun, M. A., Kittrell, F. S., Lane, D. P.,
Medina, D., and Butel, J. S.Mutations in p53 are frequent in the
preneoplastic stage of mouse mammary tumordevelopment. Cancer
Res.,53: 3374–3381, 1993.
23. Ozbun, M. A., Jerry, D. J., Kittrell, F. S., Medina, D., and
Butel, J. S. p53 mutationsselectedin vivo when mouse mammary
epithelial cells form hyperplastic outgrowthsare not necessary for
establishment of mammary cell linesin vitro. Cancer
Res.,53:1646–1652, 1993.
24. Sternlicht, M. D., Lochter, A., Sympson, C. J., Huey, B.,
Rougier, J. P., Gray, J. W.,Pinkel, D., Bissell, M. J., and Werb,
Z. The stromal proteinase MMP3/stromelysin-1promotes mammary
carcinogenesis. Cell,98: 137–146, 1999.
25. Li, B., Greenberg, N., Stephens, L. C., Meyn, R., Medina,
D., and Rosen, J. M.Preferential overexpression of a
172Arg[arrow]Leu mutant p53 in the mammarygland of transgenic mice
results in altered lobuloalveolar development. Cell GrowthDiffer,
5: 711–721, 1994.
26. Kordon, E. C., McKnight, R. A., Jhappan, C., Hennighausen,
L., Merlino, G., andSmith, G. H. Ectopic TGFb1 expression in the
secretory mammary epitheliuminduces early senescence of the
epithelial stem cell population. Dev. Biol.,168:47–61, 1995.
27. Robinson, G. W., McKnight, R. A., Smith, G. H., and
Hennighausen, L. Mammaryepithelial cells undergo secretory
differentiation in cycling virgins but require preg-nancy for the
establishment of terminal differentiation. Development
(Camb.),121:2079–2090, 1995.
28. Moitra, J., Mason, M. M., Olive, M., Krylov, D., Gavrilova,
O., Marcus-Samuels, B.,Feigenbaum, L., Lee, E., Aoyama, T.,
Eckhaus, M., Reitman, M. L., and Vinson, C.Life without white fat:
a transgenic mouse. Genes Dev.,12: 3168–3181, 1998.
29. Dobie, K., Mehtali, M., McClenaghan, M., and Lathe, R.
Variegated gene expressionin mice. Trends Genet,13: 127–130,
1997.
30. Dobie, K. W., Lee, M., Fantes, J. A., Graham, E., Clark, A.
J., Springbett, A., Lathe,R., and McClenaghan, M. Variegated
transgene expression in mouse mammary glandis determined by the
transgene integration locus. Proc. Natl. Acad. Sci.
USA,93:6659–6664, 1996.
31. Festenstein, R., Tolaini, M., Corbella, P., Mamalaki, C.,
Parrington, J., Fox, M.,Miliou, A., Jones, M., and Kioussis, D.
Locus control region function and hetero-chromatin-induced position
effect variegation. Science (Washington DC),271:1123–1125,
1996.
32. Cardiff, R. D., and Wellings, S. R. The comparative
pathology of human and mousemammary glands. J. Mammary Gland Biol.
Neoplasia,4: 105–122, 1999.
33. DeOme, K. B., Faulkin, L. J., Jr., Bern, H. A., and Blair,
P. E. Development ofmammary tumors from hyperplastic alveolar
nodules transplanted into gland-freemammary fat pads of female C3H
mice. Cancer Res.,19: 515–520, 1959.
34. Beuving, L. J. Mammary tumor formation within outgrowths of
transplanted HANSfrom carcinogen treated rats. J. Natl. Cancer
Inst.,40: 1287–1289, 1968.
35. Medina, D. Preneoplasia in mammary tumorigenesis.In: R. D.
A. M. Lippman (ed.),Mammary Tumor Cell Cycle, Differentiation and
Metastasis, pp. 37–69. Norwell,MA: Kluwer Academic Publishers,
1996.
36. Cao, Z., Umek, R. M., and McKnight, S. L. Regulated
expression of three C/EBPisoforms during adipose conversion of
3T3–L1 cells. Genes Dev.,5: 1538–1552,1991.
37. Umek, R. M., Friedman, A. D., and McKnight, S. L.
CCAAT-enhancer bindingprotein: a component of a differentiation
switch. Science (Washington DC),251:288–292, 1991.
38. Yeh, W-C., Cao, Z., Classon, M., and McKnight, S. L. Cascade
regulation of terminaladipocyte differentiation by three members of
the C/EBP family of leucine zipperproteins. Genes Dev.,9: 168–181,
1995.
39. Lekstrom-Himes, J., and Xanthopoulos, K. G. Biological role
of the CCAAT/enhancer-binding protein family of transcription
factors. J. Biol. Chem.,273: 28545–28548, 1998.
40. Buck, M., Turler, H., and Chojkier, M. LAP (NF-IL6), a
tissue-specific transcrip-tional activator, is an inhibitor of
hepatoma cell proliferation. EMBO J.,13: 851–860,1994.
41. Diehl, A. M., and Yang, S. Q. Regenerative changes in C/EBPa
and C/EBPbexpression modulate binding to the C/EBP site in the
c-fos promoter. Hepatology,19:447–456, 1994.
42. Rana, B., Xie, Y., Mischoulon, D., Bucher, N. L., and
Farmer, S. R. The DNA bindingactivity of C/EBP transcription factor
is regulated in the G1 phase of the hepatocytecell cycle. J. Biol.
Chem.,270: 18123–18132, 1995.
43. Chen, P. L., Riley, D. J., Chen-Kiang, S., and Lee, W-H.
Retinoblastoma proteindirectly interacts with and activates the
transcription factor NF-IL6. Proc. Natl. Acad.Sci. USA,93: 465–469,
1996.
44. Timchenko, N. A., Harris, T. E., Wilde, M., Bilyeu, T. A.,
Burgess-Beusse, B. L.,Finegold, M. J., and Darlington, G. J.
CCAAT/enhancer binding protein alpharegulates p21 protein and
hepatocyte proliferation in newborn mice. Mol. Cell. Biol.,17:
7353–7361, 1997.
45. Ramos, R. A., Nishio, Y., Maiyar, A. C., Simon, K. E.,
Ridder, C. C., Ge, Y., andFirestone, G. L.
Glucocorticoid-stimulated CCAAT/enhancer-binding proteina
ex-pression is required for steroid-induced G1 cell cycle arrest of
minimal-deviation rathepatoma cells. Mol. Cell. Biol.,16:
5288–5301, 1996.
46. Cha, H. H., Cram, E. J., Wang, E. C., Huang, A. J., Kasler,
H. G., and Firestone, G. L.Glucocorticoids stimulatep21 gene
expression by targeting multiple transcriptionalelements within a
steroid responsive region of the p21waf1/cip1 promoter in
rathepatoma cells. J. Biol. Chem.,273: 1998–2007, 1998.
47. Cram, E. J., Ramos, R. A., Wang, E. C., Cha, H. H., Nishio,
Y., and Firestone, G. L.Role of the CCAAT/enhancer binding
protein-a transcription factor in the glucocor-ticoid stimulation
of p21waf1/cip1gene promoter activity in growth-arrested
rathepatoma cells. J. Biol. Chem.,273: 2008–2014, 1998.
48. Chinery, R., Brockman, J. A., Peeler, M. O., Shyr, Y.,
Beauchamp, R. D., and Coffey,R. J. Antioxidants enhance the
cytotoxicity of chemotherapeutic agents in colorectalcancer: a
p53-independent induction of p21WAF1/CIP1 via C/EBPb [see
comments].Nat. Med.,3: 1233–1241, 1997.
49. Zhu, S., Oh, H. S., Shim, M., Sterneck, E., Johnson, P. F.,
and Smart, R. C. C/EBPbmodulates the early events of keratinocyte
differentiation involving growth arrest andkeratin 1 and keratin 10
expression. Mol. Cell. Biol.,19: 7181–7190, 1999.
50. Osada, S., Yamamoto, H., Nishihara, T., and Imagawa, M. DNA
binding specificityof the CCAAT/enhancer-binding protein
transcription factor family. J. Biol. Chem.,271: 3891–3896,
1996.
269
A ROLE FOR C/EBPb-LIP IN MAMMARY PROLIFERATION
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
2001;61:261-269. Cancer Res Cynthia A. Zahnow, Robert D.
Cardiff, Rodolfo Laucirica, et al. Inhibitory Protein in Mammary
Epithelial Cell Proliferation
-Liver-enrichedβA Role for CCAAT/Enhancer Binding Protein
Updated version
http://cancerres.aacrjournals.org/content/61/1/261
Access the most recent version of this article at:
Cited articles
http://cancerres.aacrjournals.org/content/61/1/261.full#ref-list-1
This article cites 49 articles, 33 of which you can access for
free at:
Citing articles
http://cancerres.aacrjournals.org/content/61/1/261.full#related-urls
This article has been cited by 23 HighWire-hosted articles.
Access the articles at:
E-mail alerts related to this article or journal.Sign up to
receive free email-alerts
Subscriptions
Reprints and
[email protected] at
To order reprints of this article or to subscribe to the
journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take
you to the Copyright Clearance Center's (CCC)
.http://cancerres.aacrjournals.org/content/61/1/261To request
permission to re-use all or part of this article, use this link
on July 10, 2021. © 2001 American Association for Cancer
Research.cancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/content/61/1/261http://cancerres.aacrjournals.org/content/61/1/261.full#ref-list-1http://cancerres.aacrjournals.org/content/61/1/261.full#related-urlshttp://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/61/1/261http://cancerres.aacrjournals.org/