Wnt factors: Genes, functions and cooperative signaling in mammary cell transformation

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. U M I films
the text directly from the original or copy submitted. Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality illustrations
and photographs, print bleedthrough, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send UM I a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and continuing
from left to right in equal sections with small overlaps.
ProQuest Information and Learning
300 North Zeeb Road, Ann Arbor, M l 48106-1346 USA
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
WNT FACTORS: GENES, FUNCTIONS AND COOPERATIVE SIGNALING IN
MAMMARY CELL TRANSFORMATION
by
JunQing Qian
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPFTY
(MOLECULAR MICROBIOLOGY AND IMMUNOLOGY)
May 2002
Copyright 2002 JunQing Qian
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
U M I Number: 3073836
Copyright 2002 by
Qian, JunQing
All rights reserved.
___ ®
UMI
UMI Microform 3073836
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
The Graduate School
U niversity Park
LOS ANGELES, CALIFORNIA 90089-1695
This dissertation, w ritten b y
JUIMQING QIAM_____________
Under the direction o f h£*X. D issertation
Committee, and approved b y a ll its members,
has been presented to and accepted b y The
Graduate School, in partial fulfillm ent o f
requirements fo r the degree o f
DOCTOR OF PHILOSOPHY
M siy L O i 2002
DISSERT A TION COMMITTEE
7 T r ..... *. Chai rperson
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
This dissertation is dedicated with love and gratitude to my parents, Jin-Hui
Qian and Xiao-Ming Guo, whose encouragement and support made this dissertation
possible. To my husband V. Jiayi Ma, lovely daughter Julianna JQ Ma and the little
one, for bringing me the meaning o f love and life. To my parents-in-law, Guo
Zhong Ma and Hao En Jia, and my aunt’s family, Chen-Ying Wang, Xiao-Yue Guo
and Zheng Wang, for their understanding and support.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
I wish to express my gratitude to a lot of people for their help and support
during my dissertation work. First of all, I would like to thank my mentor, Dr.
Gregory M. Shackleford for providing the environment to do active research and for
his advice and guidance through out my graduate studies. The valuable experience
and the affection for science I acquired under his direction will greatly benefit my
future career.
I also want to express my thanks to the other members o f my Dissertation
Committee, Dr. Michael Lai, Dr. Joseph Landolph, and Dr. Cheng-Ming Chuong for
their encouragement and scientific input at all stages of this study.
I would like to acknowledge the contributions of Zhaorong Jiang in Chapter I
for the initial cloning o f the exon 2-4 of Wnt9b cDNA, and Paige Heaphy for part of
the chromosomal mapping of Wnt9b.
I would also like to acknowledge the contributions of our collaborator, Dr.
Yves A. DeClerck’s laboratory, in Chapter III for the zymogram of MMPs and
Northern blotting o f MMP-3, and the sharing o f experimental materials and
expertises.
I am grateful to those who generously provided experimental materials: Dr.
Michael Reth for exon-trap vector pML53In, Dr. Donald Kohn for GPE ecotropic
retrovirus packaging cells, Dr. Jan Kitajewski for pBS/W ntl-HA plasmid, Dr. D.M.
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Omitz for pMirb vector, Dr. Tai-Lan Tuan for BB3103, Dr. Han Clevers for TOP-
Luc reporter construct, and Dr. Y-H Song for pHyg plasmid.
I would like to thank my friends, Dr. Kai-Jin Wu for her comments on my
dissertation and support to my research, Dr. YuHong Song for her over-a-decade of
friendship and encouragement on my graduate studies from the day one, and Dr.
Xiaobing Jiang for his friendship and help on my dissertation.
Very special thanks to my parents-in-law, Hao En Jia and Guo Zhong Ma, for
their everyday support during my preparation of this dissertation. To the family of
my aunt, Chen-Ying Wang, Xiao-Yue Guo and Zheng Wang, for their constant
encouragement.
Last of all, my sincerest thanks to my family, Jiayi and Julianna for their love
and understanding that motivated me to finish this dissertation. I am extremely
grateful to my parents, Jin-Hui Qian and Xiao-Ming Guo, for their true love, support
and sacrifice. They are the source of my strength and spirit for the past and for the
future.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
DEDICATION
ACKNOWLEDGEMENTS iii
LIST OF TABLES/FIGURES vi
ABSTRACT viii
ABBREVIATIONS X
I MOUSE WNT9B TRANSFORM ING ACTIVITY, 1
TISSUE-SPECIFIC EXPRESSION AND EVOLUTION
Abstract
Introduction
Materials and Methods
Results and Discussion
References
II W N T10A AND W N T10B IN C E L L U L A R 41
TRANSFORMATION
Introduction
Materials and Methods
Results
Discussion
References
III COOPERATIVE FACTORS IN W NT1-INDUCED 72
CELLULAR TRANSFORMATION AND SIGNALING
Introduction
Materials and Methods
Results
Discussion
References
IV Bibliography 113
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES AND FIGURES
Figure 1 Amino acid sequence alignment o f Wnt proteins 15
Table 1 Percent Amino Acid Identities and Similarities o f Mouse 18
Wnt9b and Wnt9a with Other Wnt Proteins
Figure 2 Dendrograms of Wnt proteins 2 1
Figure 3 Chromosomal mapping of Wnt9b 25
Figure 4 Adult and fetal expression of mouse Wnt9b 29
Figure 5 Wnt9b causes small foci in C57MG cells 32
Figure 6 Wnt9b expression causes mild morphological changes in 34
C57MG cells
Figure 7 Focus formation assay by overexpressing HA-tagged Wnt 53
proteins in C57MG cells
Figure 8 Morphological comparisons of C57MG cells expressing 54
HA-tagged Wnt genes
Figure 9 Focus formation assay of overexpressing Wnt proteins by 57
the LNCX retroviral infections
Figure 10 Morphological comparison of C57MG cells infected with 58
LNCX or LNC-Wnt viruses
Figure 11 Northern Blot analysis of Wnt gene expression in infected 59
cells
Figure 12 The addition of basic FGF to the WntlO focus formation 62
assay
Figure 13 WntlOa and WntlOb proteins do not trigger Wnt/p-catenin 65
signaling
Figure 14 Wnt signaling pathways and a hypothesized role o f MMPs 75
in Wnt signaling
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 15
Figure 16
Figure 17
Figure 18
Figure 19
Figure 20
Figure 21
Figure 22
Figure 23
Acidic FGF and basic FGF enhance the Wnt 1-induced 86
focus-formation
The presence of BB3103 inhibits the W ntl-induced focus 88
formation of C57MG cells
The existence of MMP activities in C57MG culture 91
MMP-7 and MMP-3 are expressed in G418-selected pools 93
of both Wntl producing cells and empty vector control
cells
CT clones respond to W ntl signal induced either by 98
transient infection or co-cultivation with W ntl producing
cells
The MMP inhibitor GM6001 partially inhibits Wnt 100
signaling
A G 3340 partially inhibits W nt/P-catenin signal 102
transduction
MMP-3 collaborates with W ntl in Wnt signaling and 104
cellular transformation
MMP-7 does not positively affect Wnt/p-catenin signaling 106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
Wnt genes encode a large family o f secreted signaling molecules with critical
functions in animal development and potential oncogenic activities when
misregulated. Described in this dissertation are the structural and/or functional
characterization o f recently discovered Wnt genes and the identification of a novel
cooperation in Wnt signaling. The full-length cDNA of mouse Wnt9b (also called
WntlS by others) was isolated, and the full-length sequence of its close relative,
Wnt9a (also called Wntl4), was deduced from unannotated genomic DNA sequence
databases. The Wnt9b gene was mapped by interspecific backcross mapping to 63
cM on chromosome 11. Analyses o f the W nt9b and W nt9a sequences and
chromosomal locations strongly suggest that these genes resulted from a vertebrate
gene duplication event and that both genes are orthologs of Wnt9 genes from shark
and hagfish. Adult expression of Wnt9b is primarily restricted to the kidney, and
fetal expression was detected throughout all developmental stages tested. Wnt9b
overexpression in C57MG mammary epithelial cells caused small transformed foci
and moderate morphological transformation as compared with W ntl. The
transforming abilities of WntlOa and WntlOb genes, recently identified in our
laboratory, were tested in similar assays and found not to have transforming
activities. A minimal addition of fibroblast growth factor (FGF) proteins, which are
strong oncogenic collaborators o f Wnts in vivo, marginally promoted focus
formation by WntlOa and WntlOb. Contrary to W ntl, WntlOa and WntlOb did not
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cause morphological changes in PC 12 neuronal cells and did not trigger Wnt/P-
catenin signal transduction in C57MG cells. A minimal addition of FGFs showed
collaboration with Wntl by enhancing focus formation in C57MG cells. Lastly, a
novel cooperation between Wntl and matrix metalloproteinases (MMPs) has been
identified in this study. We found a significant cooperation between Wntl and
MMP-3 in C57MG cell morphological transformation and Wnt/P-catenin signaling
assays. The inhibition o f MMP activity by synthetic MMP inhibitor compounds
disrupted this cooperation, as well as signaling by W ntl alone, confirming an
apparent involvement o f MMPs in Wnt signaling and suggesting that they may have
antioncogenic activity in Wnt-induced tumorigenesis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABBREVIATIONS
AP-1 Activator Protein-1
APC Adenomatous Polyposis Coli
CamKil Ca2 + -calmodulin-dependent protein Kinase II
DMEM Dulbecco's Modified Eagle's Medium
FCS Fetal Calf Serum
FGF Fibroblast Growth Factor
GSK-3 Glycogen Synthase Kinase-3
IRES Internal Ribosome Entry Site
LTR Long Terminal Repeat
MoMLV Moloney Murine Leukemia Virus
MMP Matrix Metalloproteinase
MMTV Mouse Mammary Tumor Virus
PKC Protein Kinase C
RFLP Restriction Fragment Length Polymorphism
SDS Sodium dodecyl sulfate
TCF/LEF T Cell Factor/Lymphoid Enhancer Factor
TPA/PMA l2-0-tetradecanoylphorbol-13-acetate/4-a-Phorbol
12-myristate 13-acetate
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER I
MOUSE WNT9B TRANSFORMING ACTIVITY, TISSUE-SPECIFIC
EXPRESSION AND EVOLUTION
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
The Wnt family of secreted factors has oncogenic potential and important
roles as developmental regulators. We report an analysis o f mouse Wnt9b (also
called W ntl5), including its cDNA, Jackson backcross mapping, epithelial cell
transforming activity, tissue and fetal expression patterns and evolution. We also
deduced the full-length amino acid sequence o f its close relative, Wnt9a (also called
Wnt 14) from unannotated genomic DNA sequences in GenBank. Full-length
comparisons among Wnt amino acid sequences provide evidence that Wnt9b and
Wnt9a are close paralogs and that both genes are orthologs of Wnt9 genes from shark
and hagfish. Mapping Wnt9b to the Jackson BSS backcross panel places it distal to
BRCAl and Adam 11 and proximal to Kcnj2 and Sox9 on chromosome 11. Sequence
comparisons and linkages of two pairs o f Wnt genes ( Wnt9a and Wnt3a at 32 cM on
chromosome 11, and Wnt9b and Wnt3 at 63 cM on chromosome 11) suggest that
they arose from the relatively recent duplication of a single ancestral Wnt gene pair,
confirming the paralogous relationship o f Wnt9a and Wnt9b. Wnt9b expression is
primarily restricted to the kidney in the mouse with lower levels detected in the
preputial gland, liver and mammary gland. Testing o f staged whole mouse fetuses
from 9.5 to 17.5 days o f gestation showed expression at all stages with a peak at day
10.5. No significant elevation of Wnt9b expression was detected in 29 mouse
mammary tumor virus-induced tumors. Overexpression o f Wnt9b in C57MG
mammary epithelial cells caused small transformed foci in cell monolayers and a
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
moderate morphological transformation in pooled colonies when compared with
Wntl.
Keywords: Wnt proto-oncogene, kidney gene expression, fetal gene expression,
mammary epithelial cell transformation, mouse chromosome 11, evolution
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION
The Wnt genes encode a large family of secreted cysteine-rich glycoproteins
that have important roles in short-range cell-cell communication in a variety of
processes during development and in adults [1,2]. They are found in a wide variety
of metazoans from invertebrates to man and are involved in a range o f activities that
includes cell proliferation, differentiation, cell movement, cell fate determination and
tumorigenesis [1,2]. Through the analysis of experimental and natural Wnt mutants,
Wnts have been shown to play critical roles in development o f the central nervous
system, reproductive tract, kidneys, placenta and limbs in vertebrates [1,2].
The overexpression o f several Wnt genes has been associated with a variety
of human cancers, although direct evidence o f involvement, such as amplification or
rearrangement, has not been found [3]. Despite this, the oncogenic potential of Wnt
genes has been shown and studied in a number o f ways. Several Wnt genes have
been found to be proto-oncogene targets for insertional mutagenesis by mouse
mammary tumor virus (MMTV) in mammary tumors from infected animals
including the founding member o f this family, Wntl, which was originally
discovered in such studies [4-6]. Some, but not all, Wnts can also morphologically
transform certain epithelial cell lines in culture, notably the C57MG mouse
mammary cell line, from an epithelial to a fibroblastic morphology [7]. In addition,
Wnts can also alter the growth properties o f some fibroblast cell lines and can
dramatically change the morphology, growth factor responsiveness and gene
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
expression patterns of PC12 pheochromocytoma cells [8-11]. Transgenic mice
expressing Wnts in the mammary gland have more directly demonstrated the
oncogenic potential of Wnt genes [12,13]. Such mice usually display extensive
mammary gland hyperplasia and stochastically develop adenocarcinomas.
Transgenic mice infected with MMTV have shown that fibroblast growth factors are
strong oncogenic collaborators with Wnts in mammary tumorigenesis [6,14,15].
Wnt proteins are normally associated with the extracellular matrix and signal
in an autocrine and paracrine manner through specific cell surface receptors of the
Frizzled family of seven-transmembrane proteins. In the classical Wnt signaling
pathway, stimulation of a Frizzled receptor by a Wnt causes, via disheveled, the
suppression of glycogen synthase kinase 3P and the stabilization o f cytoplasmic (3 -
catenin, which then complexes with TCF/LEF transcription factors and other
proteins to stimulate or repress transcription of target genes [1,2]. The stabilization
of P-catenin by Wnts is strongly correlated with their ability to transform C57MG
cells in culture [16]. P-catenin is normally in a complex with several other proteins
in the cytoplasm, including the APC and Axin proteins. Mutations in either APC,
Axin or P-catenin can cause stabilization o f P-catenin and activation of the pathway.
Many such mutations have been detected in human cancers and have further
implicated the Wnt signaling pathway in human tumorigenesis [3,17]. Evidence is
accumulating for a second Wnt signaling pathway that does not utilize p-catenin, but
rather functions through calcium-calmodulin kinase II and protein kinase C [18].
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
This pathway appears to be used by those Wnts that do not transform C57MG cells
in culture, as exemplified by Wnt5a [18].
Given the critical roles that Wnt proteins and their signaling pathways play in
multiple processes in development and their proven oncogenic potential, it is
important to identify the full extent o f the Wnt gene family and to characterize their
functions. Here we have isolated a fUll-length Wnt9b cDNA from the mouse,
characterized its transforming potential in C57MG cells and present the gene's
expression in adult mouse tissues and in staged fetuses. A partial mouse cDNA and
a full-length human cDNA of this gene have been reported recently and called W ntl5
and WNT14B, respectively [19,20]. Comparisons o f our full-length Wnt9b cDNA
sequence with other Wnts together with genomic analyses o f its locus provide strong
evidence that this gene is a mouse ortholog o f Wnt9 genes from fish described in
1992 [21]. Here, we will term the genes previously called Wntl5/14B as Wnt9b in
accordance with standard Wnt nomenclature.
MATERIALS AND METHODS
Isolation o f mouse Wnt9b cDNA.
A PCR-based amplification o f BALB/c mouse genomic DNA using
degenerate primers (correspond to the conserved Wnt amino acids CKCHG and
FHWCC) was employed to isolate members o f the Wnt gene family essentially as
described previously [22]. In addition, DNA was first digested by one of four
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
restriction enzymes (EcoR I, Acc I, Sac I or Sac II) to preclude amplification of Wnts
that are commonly isolated by this procedure. A 330-bp fragment of Wnt9b was
obtained from EcoR I-digested template and given the laboratory name CF2. To
isolate a cDNA by exon trapping and determine the exon structure o f this gene, we
cloned a 10 kb Bgl II genomic fragment. A partial mouse genomic library was
constructed by ligating 9 kb to 12 kb Bgl II fragments to Bam H I-digested and
phosphatased Lambda DASH vector arms (Stratagene, La Jolla, CA). Ligated DNAs
were packaged (Gigapack Gold, Stratagene, La Jolla, CA) and titered, and 3xl05
plaques were plated on P2-392 bacteria and screened with a radiolabled CF2 probe.
A positive plaque clone was purified, and the DNA insert was subcloned into
pBluescript (Stratagene, La Jolla, CA). For exon trapping, a 6.5 kb Xho I fragment
containing the 5' side o f the CF2 sequence was subcloned into the exon-trap vector
pML53In [36] (gift from Michael Reth) in both orientations. The two constructs
were separately transfected into Cos7 cells using Lipofectamine (Life Technologies,
Inc., Rockville, MD), and total RNA was extracted 48 hours later. A cDNA
fragment containing the second, third and part of the fourth exon of Wnt9b was
amplified by RT-PCR using vector-specific primers and cloned. The rest of fourth
exon sequence, which encodes the C-terminus o f Wnt9b, was obtained by
sequencing the appropriate part of the 10 kb genomic DNA. The first exon o f Wnt9b
was isolated via a nested PCR cloning strategy using the Marathon Ready mouse
kidney cDNA library (Clontech, Palo Alto, CA) as the PCR template, similar to a
procedure used previously [37]. The first round PCR was performed using the GC
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Advantage PCR kit (Clontech, Palo Alto, CA) according to the manual with the
outside primer pair: API (supplied by Clontech, Palo Alto, CA) and gene specific
primer (GSP) 1 (5’-TCCCCGCGGGGACAGCTTCAGTAGGTCAC-3'), which
anneals to exon 2 o f Wnt9b. To obtain gene specific products, a second round PCR
was perform ed using the nested prim ers AP2 and GSP2 (5'-
TCCCCGCGGGGAAGGGTGTCAGGACC-3'). A -2 0 0 bp fragment was
specifically amplified, cloned into pBluescript and sequenced to confirm that a
potential initiation codon and signal peptide were present. The full-length Wnt9b
cDNA was amplified directly from the kidney cDNA library using primers from the
translation initiation codon to the stop codon, and products from three PCRs were
cloned into pBluescript and sequenced. The Wnt9b cDNA GenBank accession
number is AF469004.
Identification o f sequences for human WNT9B and mouse and human fVnt9a.
The nucleotide and amino acid sequences o f the mouse Wnt9b cDNA and
chicken Wnt9aU4 [19] were used in blastn and tblastn searches of the various
GenBank databases (nr, htgs, gss) containing human and mouse genomic DNAs.
Candidate clones containing human WNT9B and human and mouse Wnt9a were
identified, and coding exon boundaries were determined by comparison with the
coding domains o f murine Wnt9b and chicken Wnt9aH4 and by the localization of
consensus 5'and 3’ splice sequences. Accession numbers of DNAs containing human
WNT9B are ACO15855 and NT_010763; the latter sequence was also found to
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
contain WNTS. Human WNT9A was found in accession number AL360269 and both
human WNT9A and WNT3A were found in NU_004908. Mouse Wnt9a and Wnt3a
were found in AC025579.
Sequence analyses.
Amino acid sequences were aligned using CLUSTAL W (v. 1.81) with
default parameters on a European Molecular Biology Laboratory web server
(http://www.ebi.ac.uk). Alignment outputs from this source were used to create a
dendrogram using TreeTop [38] at a node of the European Molecular Biology
Network (http://www.genebee.msu.su) [39]. Signal peptides were predicted using
SignalP V2.0 (http://www.cbs.dtu.uk) [40]. Core sequences of Drosophila and
mouse Wnt proteins in Fig.2C were prepared essentially as described by Graba et al
[25]. Core sequences consisted of the region extending from the first highly
conserved cysteine found in most Wnts from mice and Drosphila (except W ntl,
Wnt8a, Wnt8b, wg/wingless, Dwnt6, Dwnt8 and DwntlO) to the final conserved
cysteine at the C-terminus. Obvious insertions, which lack general Wnt homology,
were also deleted [25]. The exact sequences used are available upon request.
Mouse Wnt9b chromosomal mapping.
DNAs from C57BL/6 and Mus spretus mice (obtained from The Jackson
Laboratory) were analyzed for restriction fragment-length polymorphisms (RFLPs)
by Southern blotting using a Wnt9b cDNA probe. This probe hybridized to Eco RV
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
fragments o f approximately 7.5 kb for C57BL/6 and 12.5 kb for M. spretus. DNAs
of 94 mice from The Jackson Laboratory interspecific backcross panel [(C57BL/6JEi
x SPRET/Ei)Fl x SPRET/Ei], called Jackson BSS [27], were tested for the presence
or absence of the 7.4-kb C57BL/6-specific fragment. Some backcross DNA typings
were confirmed using a PCR-based assay in which a 3.2-kb Wnt9b PCR product was
differentially digested with Eag 1(1.1 and 2.1-kb fragments for C57BL/6; uncut for
M. spretus) and Bgl II (uncut for C57BL/6; 1.2 and 2.0-kb fragments for M. spretus).
The PCR primers used were 5'-TCCCCCGGGCTGACCGGTCGTGAGTCC-3' and
5'-TCCCCCGGGAGGCCTAGCGCTTGCAGGT-3', which anneal to the second and
fourth exons, respectively, of Wnt9b. The resulting data were submitted to The
Jackson Laboratory for linkage analysis. The raw mapping data may be viewed at
th e fo llo w in g World Wide Web address:
http://www.jax.org/resources/documents/cmdata.
RNA isolation, Northern blotting and RT-PCR.
Total RNA from adult and fetal mouse tissues or cultured cells was isolated
by the LiCl-Urea method as described [41]. Poly(A)+ RNA was isolated from
approximately 300 pg of total tissue RNA using an oligo-d(T) cellulose column
(Stratagene, La Jolla, CA). One-half o f each poly(A)+ RNA sample was separated
on an agarose gel and blotted to a nylon membrane (Pharmacia, Piscataway, NJ) as
previously described [11]. The membranes were hybridized to a 32P-labeled probe
generated from a 1.0 kb fragment containing exon 2-4 o f the Wnt9b cDNA. The
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
membranes were exposed to Kodak Bio-MR film at -80°C with Kodak Bio-MS
intensifying screen for 5 days. Membranes were then stripped of probe and
rehybridized with a glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probe to
assess relative loading.
Total RNAs from confluent pooled G418-selected cells were isolated using
Trizol (Life Techologies, Inc., Rockville, MD) according to the manufacturer's
instructions. Five pg was used for Northern blot analysis using a Wnt9b cDNA
probe as described above, and the Northern blots were exposed to film for 24-hours.
After stripping the previous probe, the membranes were sequentially rehybridized
and exposed to films using Wntl cDNA and GAPDH probes.
For reverse transcriptase-polymerase chain reactions (RT-PCRs), 1 pg of
total fetal RNA was treated with DNase I (Life Technologies, Inc., Rockville, MD)
for 10 minutes prior to RT reactions to degrade any residual genomic DNA. Treated
RNA samples were equally divided into two tubes, and RT reactions were performed
in the presence or absence of RT (Superscript II; Life Technologies, Inc., Rockville,
MD) according to manufacturer's instructions using random nanomers as primers. A
32-cycle PCR was then performed using one-twentieth of the RT reaction with
mouse Wnt9b primers (Forward: 5'-GCAAGTGCCACGGTGTGTCA-3'; Reverse:
5'-TCCCCCGGGAGGCCTAGCGCTTGCAGGT-3’; product size: 464 bp). Mouse
(3-actin primers (Forward: 5'-GTATGGAATCCTGTGGCATCC-3'; Reverse: 5'-
TACGCAGCTCAGTAACAGTCC-3'; product size: 350 bp) were included with the
Wnt9b primers in the same PCR tubes to normalize the cDNA amounts. The final
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
concentration o f Wnt9b and actin primers was 0.4 |xM and 0.05 |iM, respectively.
Neither PCR product accumulated to saturation under these conditions. One third of
the PCR products were then separated in a 2% agarose gel, stained with ethidium
bromide and photographed. The intensity o f each DNA band was analyzed using
SigmaGel software (Jandel— Scientific Instrument Services, Ringoes, NJ) and
plotted using Microsoft Excel.
Preparation o f Retroviral Stocks:
Mouse Wnt9b and Wntl [42] cDNAs were cloned into the Not I and Sal I
sites of pLNCX [43], a retroviral expression vector in which the gene o f interest is
driven by a cytomegalovirus promoter and the neomycin resistance gene is driven by
a Moloney murine leukemia virus long terminal repeat (MoMLV-LTR) for the
selection o f G4l8-resistant colonies. The cloned inserts were sequenced to confirm
authenticity. These two constructs and the empty pLNCX plasmid were transfected
into GPE ecotropic retrovirus packaging cells [44] (gift from Dr. Donald Kohn,
University o f Southern California, Los Anegeles, CA) using Lipofectamine (Life
Technologies, Inc., Rockville, MD). Transfected cells were grown in Dulbecco's
modified Eagle's medium (DMEM, Irvine Scientific, Santa Ana, CA) containing
10% fetal calf serum (FCS) (Omega Scientific, Tarzana, CA) and antibiotics and
were selected in this medium supplemented with 700 pg/ml (net active conc.) G418
(Geneticin, Life Technologies, Inc., Rockville, MD) for 10-14 days. The resulting
G 4l8-resistant colonies were pooled for continuous production o f viruses. The
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
supernatants containing infectious virions were collected from 70-100% confluent
cells 24-48 hours after changing to fresh medium without G418 and either used
directly or stored at -20°C or -80°C for future use.
Transformation Assays:
C57MG mouse mammary epithelial cells were grown in normal
medium— DMEM supplemented with 10% FCS and 10 pg/ml insulin (Sigma, St.
Louis, MO) with antibiotics. Infected C57MG cells were tested for transformation in
parallel using a focus-formation assay and a morphological analysis of G418-
selected cells. Approximately 2-4 x 10s C57MG cells were seeded in 10 cm plates
the day before infection. For the focus assay, C57MG cells were infected with
viruses and grown to near-confluence at which time the medium was changed to
DMEM containing 0.1% FCS and 0.1 pg/ml insulin. Cells were allowed to remain
at confluence in this medium for 5-10 days. Any resulting foci were scored and
typical foci were photographed. For the morphological analysis assay, infected cells
underwent G418 selection beginning at 48 hours after infection. At least 200 G418-
resistant colonies were pooled from each infection and grown in medium with G418
until a 10-cm dish of cells was obtained. Cells were then seeded in normal medium
at 70-90% confluence, and the medium was switched to DMEM containing 0.1%
FCS and 0.1 pg/ml insulin within 12 hours. The morphology o f the cells was scored
and documented by phase-contrast photomicroscopy 5 days later.
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
RESULTS AND DISCUSSION
Mouse Wnt9b cDNA Isolation
As an extension of a previous study to isolate potentially novel mouse Wnt
genes [22], we used degenerate primers corresponding to the conserved amino acids
CKCHG and FHWCC to amplify a 330-bp Wnt-related fragment from mouse DNA.
We used a combination of exon-trapping and RT-PCR to obtain the missing portions
o f this cDNA, and a full-length copy was finally obtained by PCR from a mouse
kidney cDNA library. We termed the gene Wnt9b due to the strong similarity of its
deduced amino acid sequence to those of the Wnt9 genes from shark and hagfish
[21]. A partial mouse cDNA from one exon of this gene that matches our cDNA has
been reported and termed Wntl5 [19]. Further rationale for the Wnt9b name, as well
as the b suffix, is explained later in this report. Wnt9b has the four-exon structure of
most mammalian Wnt genes and encodes a 359-amino acid protein that displays the
characteristic features o f Wnt proteins including an N-terminal signal peptide, one
potential N-linked glycosylation site and all 23 of the highly conserved cysteine
residues (Fig. 1). This full-length Wnt9b cDNA enabled us to better characterize the
relationship o f this gene with other Wnt family members and to examine its
biological activity.
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 1. Amino acid sequence alignment of Wnt proteins. Predicted amino acid
sequences of human and mouse Wnt9b and Wnt9a were aligned with mouse Wntl
and a partial shark Wnt9 using Clustal W (v. 1.81). Dots represent areas that are
beyond the boundaries of the protein or sequences that are not yet identified (for
shark Wnt9), and dashes represent gaps in the alignment. Aligned residues that are
identical or similar in at least 66% of the sequences are shaded with a black or gray
background, respectively; additionally, any residues that are similar to a block of
identical residues are shaded in gray. Similar amino acids were grouped as follows:
I, L, M, V; F, W, Y; H, K, R; A, G; E, D, N, Q; S, T; P; C. Asterisks denote the 23
cysteines conserved among most Wnt proteins. A tilde (~) denotes the potential N-
linked glycosylation site of Wnt9a and Wnt9b. The region o f potential signal
peptidase cleavage sites for Wnt9a and Wnt9b as predicted by SignalP V2.0 is
overlined. Arrows indicate the cysteine spacings that are specific to the Wnt9 group.
Accession numbers: Wnt9b, AF469004; shark Wnt9, P28107; W ntl, P04426.
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
m ou se
human
human
m ou se
s h a r k
m ou se
W nt9a
WNT9A
WNT9B
H n t9b
Wnt9
W n tl
LLARWjj
PLARWy
. .MRPPPAy
MRPAPAu
O LTS<3l^PS
O LTH illStPS
L C L % fe — AAA
LCLg^gP^lAAA
. M ^W A L L PS W V S T TL L jJi'3r3J»A A Li|(lH S SG R W W G IV N IA SS T N L L T 48
m ou se W nt9a
human WNT9A
human WNT9B
m ou se W nt9b
s h a r k Wnt9
m ou se W n tl
f
SM SAjE
s m s a S
hl'glK
h l ' g l S I
DSKSLQLVLEPSLQL 1 3 5 jy c^ lR L I j|3 N ^ [L H S V S G G L Q S A V R ^ ^ (W ^ Q 98
m ou se W nt9a
human WNT9A
human WMT9B
m ou se W nt9b
s h a r k Wnt9
m ou se W n tl
— *
m ouse
human
human
m ou se
s h a r k
m ouse
W nt9a
WNT9A
WNT9B
W nt9b
W nt9
W n tl gsSQ jpR R R G PG G P D SIiggB gS'jQ rD F G R L jB R E aV O gSE K G R jjQ jfL K 198
m ou se W nt9a
human WMT9A
human WNT9B
m ou se W nt9b
s h a r k Wnt9
m ou se W n tl NL 3 J |I E A gR TT VPS EMRQE 3 Q g |g ( | D 248 [VTLRAV
m ou se W nt9a
human WNT9A
human WMT9B
m ou se W nt9b
s h a r k Wnt9
m ou se W n tl GNRGSNRASRA
RGRASGS
RGRASGA
RQGSLT
KPGGPA
KHSYTL
RLEPEDPAH3> —
m ou se W nt9a M .A G — RP
human WNT9A 1: : I
S lag— r p
human WNT9B 9 t P 3 — KX
m ouse W nt9b
1 : : ]
S t p ] — EX
s h a r k Wnt9
1 : : (
S i p 3— k x
m ou se W n tl Q rx g sR L G
m ou se Wnt 9 a P
human WNT9A P
human WNT9B S
m ou se W nt9b s
s h a r k Wnt9 p
m ou se W n tl R
AGAIS
AGAXS
SSSPALDG
LD 293
X.D 293
285
287
70
293
- H I
P H Q ^ p g s ^ U ig rH T R V L H E ^ .. 370
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Wnt9b Sequence Comparisons with Other Wnt Proteins
In addition to the reported partial Wnt9b cDNA (called WntlS) and the
human ortholog called WNT14B, the closest known Wnt to mouse Wnt9b is the
human WNT14 gene [19,23]. We deduced the mouse W ntl4 amino acid sequence
from unannotated sequences in GenBank to allow full-length comparisons o f Wnt9b
to other mouse Wnt sequences (Fig. I). As detailed below, we propose that W ntl4
has sufficient sequence and structural similarity to Wnt9b to be termed its close
paralog, which in Wnt nomenclature would be called Wnt9a [21,24]. Therefore, we
will hereafter refer to this gene as Wnt9a.
The mouse Wnt9b and Wnt9a amino acid sequences were compared with all
other mouse Wnt proteins as well as with the closely related Wnt9 sequences from
shark and hagfish, and the percent identities and similarities were calculated (Table
1). Mouse Wnt9b was most highly related to Wnt9a (61% identity), which is the
same percent identity as between the mouse paralogs Wnt8A and Wnt8B and similar
to that of mouse paralogs WntlOa and WntlOb (62% identity) (not shown). The
degree of sequence identity between the full-length mouse Wnt9b or Wnt9a protein
to all other mouse Wnt proteins is much lower, ranging from 32% and 42% (Table
1). These identity levels suggest that Wnt9a and Wnt9b are appropriately named as
a/b paralogs of each other, consistent with the nomenclature of other Wnts [21,24].
Comparisons o f Wnt9b to all known Wnt sequences from other species
revealed that it is most highly related to the partial sequences o f shark and hagfish
Wnt9 (66% and 58% identity, respectively); Wnt9a was also highly related to these
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
sequences (59% and 56% identity, respectively) (Table 1 and Fig. 1). Hagfish and
shark lineages are thought to have diverged approximately 510 and 440 million years
ago, respectively [21], which presumably explains the slightly lower percent
identities to hagfish Wnt9. Since these identity levels are similar to, or higher than,
those between a number of known Wnt orthologs (e.g., mouse and zebraflsh Wnt8a
[61%], mouse and zebrafish Wntl 1 [61%], mouse and frog Wnt8a [64%] and mouse
and zebrafish Wntl Ob [65%], among others), they suggest that Wnt9a and Wnt9b
from mammals are indeed orthologs of shark and hagfish Wnt9 genes. Furthermore,
the lineages of zebrafish and frog diverged more recently (420 and 365 million years
ago, respectively) than those o f shark and hagfish [21], so the identity levels
exhibited between mouse and shark (and hagfish) provide even stronger evidence of
an orthologous relationship. Moreover, tblastn searches o f all GenBank databases
(including the complete Fugu rubripes genome) using shark and hagfish Wnt9
sequences as queries returned only sequences with obvious Wnt9b or Wnt9a
similarity as the closest matches, suggesting that no genes other than Wnt9b and
Wnt9a (except, perhaps, lineage-specific duplicates of these genes) are likely to
emerge as orthologs of shark and hagfish Wnt9.
TABLE 1
Percent Amino Acid Identities and Similarities of
Mouse Wnt9b and Wnt9a with Other Wnt Proteins
Wnt9b Wnt9a
Wntl 38/44 36/43
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE 1
(continued)
Wnt9b Wnt9a
Wnt2 40/46 39/48
Wnt2b/13 40/47 38/47
Wnt3 34/40 34/41
Wnt3a 32/39 34/42
Wnt4 37/45 37/44
Wnt5a 35/43 34/44
Wnt5b 37/44 35/43
Wnt6 40/46 35/44
Wnt7a 36/44 35/43
Wnt7b 35/43 36/44
Wnt8a 35/45 34/42
Wnt8b 36/42 36/43
Wnt9a 61/69
Wnt9b 61/69
Wnt 10a 39/47 36/44
Wnt 10b 38/45 38/48
W ntll 37/44 35/42
W ntl6 42/48 36/45
Wnt9-shark partial 66/73 59/70
Wnt9-hagfish partial 58/63 56/67
Note. All indicated proteins are full-length mouse sequences, except shark and
hagfish Wnt9. Percent identities and similarities (first and second numbers,
respectively) were determined using the Gap program (Genetics Computer
Group, Inc.). The two highest sets o f values in each column are boldfaced.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To gain additional perspective on the relationship o f Wnt9b and Wnt9a to
other Wnts, we aligned all mouse Wnts (full-length sequences) using Clustal W
program and generated a dendrogram from this alignment (Fig. 2A). This
dendrogram illustrates the relationship o f Wnt9a to Wnt9b, which is similar in
closeness to that of three other pairs of close Wnt paralogs: Wnt 10a and Wnt I Ob,
W nt8a and Wnt8b, and Wnt2 and Wnt2b. This supports our use o f the a/b
nomenclature to describe the mouse Wnt9a and Wnt9b genes. We also generated a
dendrogram o f partial sequences from all known mouse and shark Wnts (and hagfish
Wnt9) corresponding to the known regions of shark Wnts (approximately the C-
terminal third of the full-length proteins) [21] (Fig. 2B). This dendrogram shows
that shark and hagfish Wnt9 is most closely related to Wnt9b, as also shown in Table
I, and that this relationship is at least as close as that between the shark and mouse
orthologs of Wnt 10a. These analyses support the notion that Wnt9a and Wnt9b are
paralogs with a close evolutionary history and that they are orthologs of shark and
hagfish Wnt9. Additionally, we used dendrograms to examine the relationship o f
Wnt9b and Wnt9a to Drosophila Wnt proteins, of which there are a total of seven.
This analysis revealed that Drosophila Wnt4 has significant sequence similarity to
Wnt9a and Wnt9b and suggests that the Wnt9!b genes have a direct evolutionary link
with Drosophila Wnt4 (Fig. 2C). This sequence similarity between Drosophila Wnt4
and the Wnt9 proteins from shark and hagfish suggests that the respective genes are
likely orthologs, as has been proposed previously [25]
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 2. Dendrograms of Wnt proteins. Multiple sequence alignments were
performed using CLUSTAL W (v. 1.81), and a phylogenetic tree was produced with
the TreeTop algorithm. (A) Dendrogram o f full-length mouse Wnt amino acid
sequences showing the close relationship of Wnt9a and Wnt9b. (B) Dendrogram of
partial Wnt amino acid sequences from all mouse and shark Wnts (and hagfish
Wnt9) suggesting an orthologous relationship between shark and hagfish Wnt9 and
mouse Wnt9b and Wnt9a. These aligned partial sequences correlate with those
present in the shark Wnt9 sequence in Fig. I. (C) Dendrogram of core sequences
(prepared as described in Materials and Methods) of all Drosophila and mouse Wnt
proteins suggesting that that Drosophila Wnt4 may be the Drosophila ortholog of
mouse Wnt9a and Wnt9b. All other Drosophila Wnts sorted with their presumed
orthologs-the Drosophila nomenclature does not always reflect orthology with
vertebrate Wnts.
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
■ Wnt8a
• Wnt8b
-W ntl
■ Wnt6
■ Wnt2
■ Wnt2b
■ WntSa
• WntSb
■ Wnt7a
• Wnt7b
• Wntl 6
- Wnt3
• Wnt3a
• Wnt4
• Wntl 1
- WntlOa
■Wntl Ob
■ Wnt9a
• Wnt9b
Wntl
Wntl (sh)
Wnt3
Wnt3b (sh)
Wnt3a
Wnt3a (sh)
Wnt4
Wnt6
WntlOa
WntlOa (sh)
Wntl Ob
Wnt2
Wnt2b
Wnt5a
WntSa (sh)
WntSb
Wnt5b (sh)
Wnt7a
Wnt7a (sh)
Wnt 7b
Wnt7b (sh)
Wntl 6
Wntl 1
Wnt8a
Wnt8b
Wnt9b
Wnt9 (sh)
Wnt9a
Wnt9 (hf)
Wnt2
Wnt2b
WntSa
WntSb
WntS (Dm)
Wg (Dm)
Wntl
Wnt3
Wnt3a
Wnt4
Wnt7a
Wnt7b
Wnt2 (Dm)
Wnt16
Wntl 1
Wnt8a
Wnt8b
Wnt9a
Wnt9b
Wnt4 (Dm)
Wnt6
Wnt6 (Dm)
WntlOa
Wntl Ob
WntlO (Dm)
Wnt8 (Dm)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
An examination of the amino acid sequences o f Wnt9b and Wnt9a reveals
that they have two short amino acid deletions between the eleventh and thirteenth
conserved cysteines that create an unusual cysteine spacing among Wnt proteins
(denoted by arrows in Fig. I). As noted earlier for W ntl5(9b), this area contains
deletions of two and four residues relative to almost all other Wnt proteins described
thus far [19]. We examined the sequences in this region o f 135 individual Wnt
proteins from various species from nematodes to humans, and this spacing was found
only in Wnt9b and Wnt9a from mammals and in Wnt9 from shark and hagfish. The
spacing is otherwise invariant among all chordate Wnts. Only Drosophila Wnt4 had
similar deletions of eight and four residues at these two positions [25], and mom2
from C. elegans had deletions of seven and six amino acids [26]. This remarkably
unique structural similarity among the Wnt9 proteins of mammals and fish, together
with the sequence similarity discussed above, suggests that they share a close
evolutionary history and further supports the categorization o f these genes as
orthologs.
Wnt9b Chromosomal Location
We mapped the mouse Wnt9b gene to chromosome 11 by Southern blotting
and PCR analysis o f Wnt9b polymorphisms in 94 interspecific backcross animals
using the Jackson Laboratory backcross mapping resource [27]. Wnt9b mapped just
distal (I. I cM +/- l.l cM) to a cluster of mapped genes that includes A d a m ll, Bcral,
Gast, Gcn512, Gm, H apl, Hcrt, Mpp3, Pea3, Psme3, Tubg and Mpmv8 (Fig. 3 and
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
data not shown). The W nt9b location corresponds to position 63.0 cM o f
ch ro m o so m e 11 on th e M ouse G enom e D a ta b a se m ap
(http://www.informatics.jax.org/searches/linkmap_form.shtml) [28]. Another Wnt
gene, Wnt3, has also been found to map closely to Wnt9b [19], and we confirmed
this in our analysis of unannotated mouse genomic sequences in GenBank, in which
we found Wnt9b and W nt3 within the same contig (accession no. AL596108)
approximately 24.4 kb from each other in a 5'-to-5' orientation. We queried
GenBank with the sequences of Wnt9a and Wnt3a, the close paralogs of Wnt9b and
Wnt3, and also found them both within a single contig (accession no. AC025579)
approximately 16 kb from each other, again in a 5'-to-5' orientation. This finding
suggests that the Wnt9b-Wnt3 and Wnt9a-Wnt3a gene pairs arose by duplication of a
single ancestral Wnt gene pair.
Evidence suggests that the genomes o f vertebrates have experienced two
large-scale gene duplications, possibly genome duplications, during and after their
divergence from simple chordates approximately 500-550 million years ago [21,29].
These duplications have resulted in paralogous genes with relatively similar amino
acid sequences. The nomenclature of Wnt genes takes into account the close
relationship of Wnts that experienced these duplication events, and such genes are
given identical numbers but with a/b suffixes [21,24]. Given the close sequence
relationship of the Wnt9b-Wnt3 and Wnt9a-Wnt3a pairs to each other, and the fact
that they apparently arose by the relatively recent duplication o f an ancestral pair, it
now seems clear that Wnt9a and Wnt9b are close Wnt paralogs warranting the a/b
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 3. Chromosomal mapping of Wnt9b. Localization of mouse Wnt9b to
chromosome 11 by interspecific backcross mapping. (A) The Jackson BSS
interspecific backcross showing part of Chromosome 11 with loci linked to Wnt9b.
The map is depicted with the centromere toward the top. The percent recombination
(cM) +/- the standard error (SE) between adjacent loci is given to the left of the
chromosome bar. The order of loci mapping to the same position is arbitrary. (B)
Segregation of loci in 94 mice in the backcross. Loci are listed in order with the
most proximal at the top. The black boxes represent the C57BL6/JEi allele and the
white boxes the SPRET/Ei allele. The number o f animals with each haplotype is
given at the bottom of each column o f boxes. Missing typings were inferred from
surrounding data where assignment was unambiguous.
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A
Jackson BSS Chr 11
CM+/-SE 7
r
r
’ Csf3, Crk7, Hoxb7, Hoxb13
.1 + /-1 .1
■ Adam11, Brcal, Gast, Hap1
.1 + /-1.1
- Wnt9b, D11Mit126
.1 +/- 1.1
■ D11MH10, D11MH127
.1 +/- 1.1
■ D11Mit181, D11Mit333
.1 +/- 1.1
- D11Mit11, Kcnj2, Ods
.1 +/- 1.1
- D11MH101, Sox9
b
C s/3
Brcal
Wnt9b
D u m t i o ■ □ □ □ □ □ □ ■
D 11 M H181
Kcnj2
Sox9
47 41 1 1 1 1 1 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
nomenclature. As the last large-scale gene duplication event is proposed to have
occurred at or before the divergence o f jawed vertebrates [29], it is tempting to
speculate that the shark genome also contains a paralog of the known Wnt9 gene.
From the dendrogram above (Fig. 2B), this second gene might be predicted to be the
direct ortholog o f Wnt9a.
Wnt9b Expression in Adult Mouse Tissues, During Fetal Development and in
Mouse Mammary Tumors
The expression patterns of Wnts in adult mice differ markedly, and their
functions in adult organs apart from roles in development are not well established.
We examined the expression of mouse Wnt9b by Northern blotting of poly(A)+
RNA from 18 adult organs, as well as newborn kidney and cultured C57MG mouse
mammary epithelial cells. We found that, in the adult, transcripts of Wnt9b are most
abundant in kidney (Fig. 4A). The newborn kidney sample displayed a significantly
higher level of Wnt9b RNA than did the adult. Other adult expression sites include
the male preputial gland, the female mammary gland, liver and low but detectable
levels in brain. W nt9b transcripts were not detected in C57MG cells, which
morphologically respond to many Wnt signals.
The Wnt genes are well known for their importance as short distance
signaling molecules during development. We examined fetal samples from 9.5 to
17.5 days of gestation using reverse transcriptase-PCR (RT-PCR) analysis and found
Wnt9b transcripts throughout the entire period tested (Fig. 4B). By examining the
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
RT-PCR products of Wnt9b and fi-actin at times prior to the plateau phase of PCR,
we were able to estimate that expression of Wnt9b peaks at approximately 10.5 days
of gestation (Fig. 4C). The expression pattern suggests that Wnt9b may function in
several adult tissues, particularly in kidney, and during at least the latter half of fetal
development. Although we did not examine specific fetal tissues in this study, given
the high level o f Wnt9b mRNA in newborn kidney, it seems likely that this tissue is a
major site of expression in the developing fetus. It is possible, therefore, that Wnt9b
functions in concert with the Wnt4 in kidney development [30].
Several Wnt genes, such as W ntl, Wnt3 and WntlOb, have been found to be
transcriptionally activated by mouse mammary tumor virus (MMTV) insertion
mutations [4-6]. To determine if Wnt9b might be a target for MMTV insertions in
tumors, we examined Wnt9b expression in a group of MMTV-induced mammary
tumors from BALB/c mice by RT-PCR. Among the 29 tumor samples tested, 17
were positive for Wnt9b transcripts. However, after normalizing to fi-actin, the
levels o f Wnt9b transcripts in tumor samples were very low compared to the level in
adult kidney or to levels normally observed for MMTV activated genes (data not
shown). Thus, Wnt9b is not likely to be a major target for MMTV insertional
mutagenesis in BALB/c mice. O f course, this result does not exclude the possibility
that Wnt9b could be involved in mammary tumorigenesis in other contexts, such as
in certain transgenic mice or in MMTV-induced tumors from other genetic
backgrounds.
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4. Adult and fetal expression of mouse Wnt9b. (A) Northern Blot analysis
o f adult tissue expression. Poly(A)+ RNAs from 21 pooled mouse tissue samples
and one cultured cell sample were subjected to northern blotting and hybridization
with a Wnt9b cDNA probe and, after probe removal, with a GAPDH probe. (B) RT-
PCR analysis o f fetal expression. Staged mouse fetal RNA samples from 9.5 to 17.5
days postcoitum (dpc) were tested for Wnt9b expression by RT-PCR (lanes marked
RT+). Lanes marked RT- represent RT reactions lacking reverse transcriptase,
which served as negative controls. (5-actin cDNA was co-amplified with Wnt9b in
all tubes as a control for RNA concentration. (C) The density o f each band in (B)
was analyzed using Sigma Gel software (Jandel, Scientific Instrument Services,
Ringoes, NJ), and the plotted result shows the peak o f Wnt9b fetal expression at 10.5
dpc.
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A
Wnt9b
G A PD H
* 1.4 kb
d p c 9-5 10.5 11.5 12.5 13.5 14.5 15.5 16.5 17.5 § ^
RT - + - + - + - + + s
N Wnt9b
(464bp)
U p-actin
(350bp)
c
Fetal Expression of Wnt9b
; ^ & H I § £ & ESI m i ^ ^
9.5 10.5 11.5 12.5 13.5 14.6 15.5 16.5 17.5 kidney
days postcoitum
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Wnt9b Transforming Activity
We tested the transformation potential o f Wnt9b by determining whether
Wnt9b induces morphological changes on C57MG mouse mammary epithelial cells
using a focus formation assay and at confluence. The cDNA encoding full-length
Wnt9b protein was cloned into the LNCX retroviral vector and the resulting viral
stocks, as well as IFwf/-containing and empty vector control stocks, were used to
infect C57MG cells. The cells infected with the LNCX vector alone, as well as
uninfected cells, formed no obvious foci in confluent monolayers (Figs. 5A and 5B),
and the cells infected with the Wntl viruses, as the positive control, formed large foci
(Figs. 5C and 5D), as expected. The cells infected with the Wnt9b cDNA viruses
formed small foci in multiple independent infections (Figs. 5E-5H), suggesting that
Wnt9b can transform C57MG cells but is not as potent as Wntl in this assay.
In parallel to the focus formation assay, infected cells were also selected in
G418 for general morphological analysis. More than 200 G4l8-resistant colonies
infected with either the Wnt9b, W ntl or empty viruses were pooled and grown to
confluence. Cells infected by W ntl viruses showed transformation from a flat
cuboidal morphology to a spindle-shaped elongated morphology and lost contact
inhibition (Fig. 6A, panels c and d). Parental cells and cells infected by LNCX
vector alone remained flat and cuboidal (Fig. 6A, panels and b). Cells infected by
Wnt9b showed that a majority o f cells changed to a disorganized spindle-shaped
elongated morphology, but they failed to lose contact inhibition (Fig. 6A, panels e
and f). These data show that C57MG cells can respond morphologically to Wnt9b
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 5. Wnt9b causes small foci in C57MG cells. CS7MG cells were infected
with the retroviral vectors noted below and maintained afterward in medium with
0.1% FCS for 5-7 days; pictures were taken o f typical foci. Uninfected parental
C57MG cells (A) and cells infected with the empty vector LNCX (B) did not
develop foci. Cells infected with W ntl retroviruses were transformed and formed
large foci (C,D). Cells infected with Wnt9b retroviruses were mildly transformed
and formed small foci (E,F,G,H)
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
stimulation. However, Wnt9b does not strongly transform C57MG cells in terms o f
focus formation and morphology at confluence.
To confirm that the exogenous Wnts are expressed in the selected C57MG
cells, total cellular RNA was extracted from the pooled cells and subjected to
Northern blot analysis. The W ntl or Wnt9b transcripts were readily detected (Fig.
6B), whereas no such transcripts were detected in cells infected with the empty
vector. Thus, although the two Wnt genes were expressed at comparable levels, they
differed in their ability to transform C57MG cells.
Including this study, 12 o f the 19 known mouse Wnts have now been tested
for transforming activity in C57MG cells [7,16,31,32]. These studies have shown
that Wnts fall into three general groups-highly transforming, moderately/weakly
transforming and non-transforming. Our results show that Wnt9b is a moderately
transforming Wnt. The reasons for the differences in the potency o f Wnts in
transforming C57MG cells have not been fully elucidated. However, a simple
explanation for the lack of transforming activity by some Wnts is that appropriate
receptors or co-receptors for such Wnts are not expressed at sufficient levels in these
cells. Alternatively, the appropriate receptors may be present, but these Wnts and
their receptors may use the so-called Wnt/Ca2 + pathway [18], which is independent
o f the beta-catenin pathway used by transforming Wnts. This pathway has been
shown to be used in nonmammalian systems by Wnt5a, a nontransforming Wnt. A
third possibility is that C57MG cells may express sufficient quantities o f an
antagonistic factor that inhibits the Wnt activity being assayed. Several examples of
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 6. Wnt9b expression causes mild morphological changes in C57MG
cells. (A) Pooled G4l8-resistant colonies o f Wnts or vector-infected cells were
grown to near confluence, maintained at confluence in medium with 0.1% FCS for 5
days and photographed. Uninfected parental C57MG cells (a) and cells infected with
the empty vector (b) remained flat and cuboidal in shape and stay confluent as a
monolayer. Wntl vector-infected cells (c, d) exhibited an elongated spindle shape
and lost contact inhibition. Wnt9b vector-infected cells (e, f) showed a milder
transformed phenotype than W ntl-infected cells. (B) Northern blot analysis of
Wnt9b and W ntl in the G4l8-selected cell pools from panel A. Total RNA (5 pig
each) was electrophoresed, blotted and sequentially hybridized with probes for
Wnt9b (upper panel), Wntl (middle panel) and GAPDH (lower panel). Blots were
stripped of probe between hybridizations. CON, cells infected with empty vector
only.
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A
sr T jf ^
5 ^ ? ~.?ivtr>>. N
M = rt? - ■ ■ .• > • • ivAi •
B
CON Wntl Wnt9b
M Wnt9b
M Wntl
4 0 1 1 * - * * * & - < GAPDH
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
such factors exist, including the Frzb [33] and Dickkopf [34] proteins as well as
some nontransforming Wnts themselves [35]. The secretory attribute o f these
proteins might allow the specificity to inhibit only certain transforming Wnts or their
receptors, as would be required to explain the lack of, or weak, transforming activity
by a subset of Wnts.
In conclusion, we report that Wnt9b has transforming potential in C57MG
cells and show its expression during fetal development and in a restricted set o f
tissues in the adult, especially the kidney. We additionally analyze the relationship
o f this gene with other Wnt family members and conclude that Wnt9b and Wnt9a,
respectively, are close paralogs o f each other and are orthologs o f the Wnt9 genes
previously isolated from shark and hagfish. Among the evidence for these the
paralogous and orthologous relationships are (i) a relatively close percentage identity
with each other and with their orthologs in fish (within percentages commonly used
for other Wnt paralogs and orthologs), (ii) a unique cysteine pattern that groups these
genes together in a restricted structural set, (iii) the presence o f these genes in
duplicated chromosomal segments, making them close paralogs, and (iv) the lack o f
any other candidates related to shark and hagfish Wnt9 in the nearly completely
sequenced genomes of several other vertebrates.
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
1. Wodarz, A., and Nusse, R. (1998). Mechanisms o f Wnt signaling in
development. Annu. Rev. Cell Dev. Biol. 14: 59-88.
2. Moon, R. T., Brown, J. D., and Torres, M. (1997). WNTs modulate cell fate
and behavior during vertebrate development. Trends Genet. 13: 157-162.
3. Polakis, P. (2000). Wnt signaling and cancer. Genes Dev. 14: 1837-1851.
4. Nusse, R., and Varmus, H. E. (1982). Many tumors induced by the mouse
mammary tumor virus contain a provirus integrated in the same region o f the host
genome. Cell 31: 99-109.
5. Roelink, H., Wagenaar, E., Lopes da Silva, S., and Nusse, R. (1990). Wnt-3,
a gene activated by proviral insertion in mouse mammary tumors, is homologous to
int-l/Wnt-1 and is normally expressed in mouse embryos and adult brain. Proc. Natl.
Acad. Sci. USA 87: 4519-4523.
6. Lee, F. S., Lane, T. F., Kuo, A., Shackleford, G. M., and Leder, P. (1995).
Insertional mutagenesis identifies a member of the Wnt gene family as a candidate
oncogene in the mammary epithelium o f int-2/Fgf-3 transgenic mice. Proc. Natl.
Acad. Sci. U S A 92: 2268-2272.
7. Brown, A. M., Wildin, R. S., Prendergast, T. J., and Varmus, H. E. (1986). A
retrovirus vector expressing the putative mammary oncogene int-1 causes partial
transformation of a mammary epithelial cell line. Cell 46: 1001-1009.
8. Bafico, A., Gazit, A., Wu-Morgan, S. S., Yaniv, A., and Aaronson, S. A.
(1998). Characterization o f Wnt-1 and Wnt-2 induced growth alterations and
signaling pathways in NIH3T3 fibroblasts. Oncogene 16: 2819-2825.
9. Bradbury, J. M., Niemeyer, C. C., Dale, T. C., and Edwards, P. A. (1994).
Alterations o f the growth characteristics o f the fibroblast cell line C3H 10T1/2 by
members o f the Wnt gene family. Oncogene 9:2597-2603.
10. Shackleford, G. M., Willert, K., Wang, J., and Varmus, H. E. (1993). The
Wnt-1 proto-oncogene induces changes in morphology, gene expression, and growth
factor responsiveness in PC12 cells. Neuron 11: 865-875.
11. Erdreich-Epstein, A., and Shackleford, G. M. (1998). Differential expression
o f Wnt genes in normal and flat variants o f PC 12 cells, a cell line responsive to
ectopic Wntl expression. Growth Factors 15: 149-158.
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12. Tsukamoto, A. S., Grosschedl, R., Guzman, R. C., Parslow, T., and Varmus,
H. E. (1988). Expression of the int-1 gene in transgenic mice is associated with
mammary gland hyperplasia and adenocarcinomas in male and female mice. Cell 55:
619-625.
13. Lane, T. F., and Leder, P. (1997). Wnt-10b directs hypermorphic
development and transformation in mammary glands of male and female mice.
Oncogene 15: 2133-2144.
14. Shackleford, G. M., MacArthur, C. A., K.wan, H. C., and Varmus, H. E.
(1993). Mouse mammary tumor virus infection accelerates mammary carcinogenesis
in Wnt-l transgenic mice by insertional activation of int-2/Fgf-3 and hst/Fgf-4. Proc.
Natl. Acad. Sci. U S A 90: 740-744.
15. MacArthur, C. A., Shankar, D. B., and Shackleford, G. M. (1995). Fgf-8,
activated by proviral insertion, cooperates with the W nt-l transgene in murine
mammary tumorigenesis. J. Virol. 69: 2501-2507.
16. Shimizu, H., et al. (1997). Transformation by Wnt family proteins correlates
with regulation of beta-catenin. Cell Growth Differ. 8: 1349-1358.
17. Taipale, J., and Beachy, P. A. (2001). The Hedgehog and Wnt signalling
pathways in cancer. Nature 411: 349-354.
18. Kuhl, M., Sheldahl, L. C., Park, M., Miller, J. R., and Moon, R. T. (2000).
The Wnt/Ca2 + pathway: a new vertebrate Wnt signaling pathway takes shape. Trends
Genet. 16: 279-283.
19. Bergstein, I., et al. (1997). Isolation of two novel WNT genes, WNT 14 and
WNT15, one of which (WNT15) is closely linked to WNT3 on human chromosome
17q2l. Genomics 46:450-458.
20. Kirikoshi, H., Sekihara, H., and Katoh, M. (2001). Molecular cloning and
characterization of WNT14B, a novel member o f the WNT gene family. Int. J.
Oncol. 19:947-952.
21. Sidow, A. (1992). Diversification o f the Wnt gene family on the ancestral
lineage of vertebrates. Proc. Natl. Acad. Sci. U S A 89: 5098-5102.
22. Wang, J., and Shackleford, G. M. (1996). Murine WntlOa and WntlOb:
cloning and expression in developing limbs, face and skin of embryos and in adults.
Oncogene 13: 1537-1544.
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23. Saitoh, T., Hirai, M., and Katoh, M. (2001). M olecular cloning and
characterization of WNT3A and WNT14 clustered in human chromosome lq42
region. Biochem. Biophys. Res. Commun. 284: 1168-1175.
24. Gavin, B. J., McMahon, J. A., and McMahon, A. P. (1990). Expression of
multiple novel Wnt-l/int-1-related genes during fetal and adult mouse development.
Genes Dev. 4: 2319-2332.
25. Graba, Y., et al. (1995). DWnt-4, a novel D rosophila Wnt gene acts
downstream o f homeotic complex genes in the visceral mesoderm. Development
121:209-218.
26. Rocheleau, C. E., et al. (1997). Wnt signaling and an APC-related gene
specify endoderm in early C. elegans embryos. Cell 90: 707-716.
27. Rowe, L. B., et al. (1994). Maps from two interspecific backcross DNA
panels available as a community genetic mapping resource. Mamm. Genome. 5: 253-
274.
28. Blake, J. A., Eppig, J. T., Richardson, J. E., Bult, C. J., and Kadin, J. A.
(2001). The Mouse Genome Database (MGD): integration nexus for the laboratory
mouse. Nucleic Acids Res. 29: 91-94.
29. Sidow, A. (1996). Gen(om)e duplications in the evolution o f early
vertebrates. Curr. Opin. Genet. Dev. 6: 715-722.
30. Stark, K.., Vainio, S., Vassileva, G., and McMahon, A. P. (1994). Epithelial
transformation o f metanephric mesenchyme in the developing kidney regulated by
Wnt-4. Nature 372:679-683.
31. Wong, G. T., Gavin, B. J., and McMahon, A. P. (1994). Differential
transformation of mammary epithelial cells by Wnt genes. Mol. Cell Biol. 14: 6278-
6286.
32. Christiansen, J. H., Monkley, S. J., and Wainwright, B. J. (1996). Murine
WNT11 is a secreted glycoprotein that morphologically transforms mammary
epithelial cells. Oncogene 12: 2705-2711.
33. Moon, R. T., Brown, J. D., Yang-Snyder, J. A., and Miller, J. R. (1997).
Structurally related receptors and antagonists compete for secreted Wnt ligands. Cell
88: 725-728.
34. Zorn, A. M. (2001). Wnt signalling: antagonistic Dickkopfs. Curr. Biol. 11:
R592-595.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35. Torres, M. A., et al. (1996). Activities o f the Wnt-1 class o f secreted
signaling factors are antagonized by the Wnt-5A class and by a dominant negative
cadherin in early Xenopus development. J. Cell Biol. 133: 1123-1137.
36. Auch, D., and Reth, M. (1990). Exon trap cloning: using PCR to rapidly
detect and clone exons from genomic DNA fragments. Nucleic Acids Res. 18: 6743-
6744.
37. Shackleford, G. M L, Ganguly, A., and MacArthur, C. A. (2001). Cloning,
expression and nuclear localization of human NPM3, a m em ber o f the
nucleophosmin/nucleoplasmin family of nuclear chaperones. BMC Genomics 2: 8.
38. Chumakov, K. M., and Iushmanov, S. V. (1988). [A principle o f maximum
topological similarity in molecular systematics]. Mol. Gen. Mikrobiol. Virusol.: 3-9.
39. Brodsky, L. I., et al. (1995). GeneBee-NET: Internet-based server for
analyzing biopolymers structure. Biochemistry 60: 923-928.
40. Nielsen, H., Engelbrecht, J., Brunak, S., and von Heijne, G. (1997).
Identification o f prokaryotic and eukaryotic signal peptides and prediction of their
cleavage sites. Protein Eng. 10: 1-6.
41. Shackleford, G. M., and Varmus, H. E. (1987). Expression o f the proto
oncogene int-l is restricted to postmeiotic male germ cells and the neural tube of
mid-gestational embryos. Cell 50: 89-95.
42. Fung, Y. K., Shackleford, G. M., Brown, A. M., Sanders, G. S., and Varmus,
H. E. (1985). Nucleotide sequence and expression in vitro o f cDNA derived from
mRNA of int-l, a provirally activated mouse mammary oncogene. Mol. Cell. Biol. 5:
3337-3344.
43. Miller, A. D., and Rosman, G. J. (1989). Improved retroviral vectors for gene
transfer and expression. Biotechniques 7: 980-982, 984-986, 989-990.
44. Chong, H., Starkey, W., and Vile, R. G. (1998). A replication-competent
retrovirus arising from a split-function packaging cell line was generated by
recombination events between the vector, one of the packaging constructs, and
endogenous retroviral sequences. J. Virol. 72: 2663-2670.
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER II
WNT10A AND WNTIOB IN CELLULAR TRANSFORMATION
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION
As introduced in Chapter I, Wnt genes encode a large family o f secretory but
extracellular matrix-associated short distance signaling molecules. The biological
functions o f Wnt proteins in animals include cell fate determination, cell growth and
differentiation, and are indispensable for animal development [1,2]. If misregulated,
Wnts and their signaling components can be responsible for, or associated with, the
formation of many kinds of tumors in animals and humans[3]. To better understand
this gene family, many laboratories have been actively engaged in searching for the
new members of this family. The efforts o f our laboratory resulted in the discovery
of three new members: Wnt9b (described in Chapter I), Wnt 10a and Wnt 10b [4].
Wnt 10a and Wnt 10b are two distinct but closely related Wnt genes (62%
amino acid identity) [4]. WntlOb has two alternatively spliced forms— WntlOb,
which encodes the full-length protein, and WntlOb-S, which contains a 25% central
deletion [4]. Both Wnt 10a and WntlOb genes are expressed in many tissues of adults
and during embryonic stages. The major adult expression sites for Wnt 10a include
brain, adrenal gland, thymus, spleen, mammary gland and, at lower levels, all other
organs. WntlOb transcripts are more concentrated in lung and uterus. Their
functions in these adult organs are not yet clear. Both Wnt 10a and WntlOb are
expressed in embryos, like most of the Wnts, suggesting that they also function in
animal development [4].
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
WntlOa and WntlOb have been related to a variety o f developmental
processes. In zebrafish, WntlOa has been suggested to function in regionalization of
the brain during embryogenesis [5]. WntlOa has also been shown to be stimulated
by NF-kappa B signaling at the pre-B to immature B cell transition [6]. Both
WntlOa and WntlOb may be involved in hair follicle morphogenesis [7,8]. WntlOb
has been shown to function during hematopoiesis[9,10] in the expansion of multi
lineage progenitor cells. WntlOb has also been found to be important in animal tooth
development [11,12]. WntlOb together with its putative receptor Fzd5 may be
responsible for yolk sac and placental angiogenesis, where loss of function causes
defects in yolk sac function [13]. Adipogenesis and skin development and function
are other processes that involve WntlOb functions [14,15].
Both WntlOa and WntlOb are associated with tumorigenesis and multiple
cancers. In humans, WNT10A up-regulation has been reported in esophageal,
gastric and colorectal cancer [16]. WNT10A has also been reported to be
upregulated by tumor necrosis factor alpha (TN Fa) or Helicobacter pylori in the
MKN45 gastric cancer cell line [16]. High level expression of WNT10A was
detected in cancer cell lines such as SW480 (colorectal cancer), HL-60
(promyelocytic leukemia) and Raji (Burkitt’s lymphoma) [17].
Additional evidence has been gathered that WntlOb is involved in
tumorigenesis. Like Wntl, WntlOb is a target for MMTV insertional mutagenesis in
developing mammary tumors [18]. In a transgenic mouse model that targets ectopic
expression of WntlOb to the mammary glands, WntlOb causes mammary gland
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hyperplasia, which progresses to the formation of mammary adenocarcinomas, thus
confirming its roles in mouse mammary tumorigensis [19]. Additionally, WntlOb
cooperates with Fgf3 in forming mammary tumors in transgenic mice [20]. Human
W NT 1 OB is expressed in some breast carcinomas but not normal or benign
proliferating breast tissues; however no WNT10B gene amplification or
rearrangement was found in those carcinomas [21]. Human WNT 1 OB is also
expressed in some endometrial carcinoma cell lines but not in normal endometrial
epithelial and stromal cells [22]. Expression or elevated expression of WNT10B has
also been reported in several other human cancer cells lines [23]. In the MKN45
human gastric cancer cell line, upon the stimulation o f TN Fa (frequently activated in
gastric mucosa by bacterial infection), WNT 1 OB is one o f the major factors that is
upregulated and believed to lead to gastric carcinogenesis [24].
Besides animal models and the study of Wnt expression in tumors and cancer
cell lines, an important analysis of Wnts is whether Wnt proteins are able to cause
transformation of cultured cells. Several cell lines are known to morphologically
respond to Wnt stimulation. RAC1 lc, a mammary cell line, can be transformed by
W ntl or Wingless, and these transformed cells are tumorigenic in nude mice [25].
PC 12, a rat pheochromocytoma cell line, is changed from a round and barely
adherent cell type to a flat adherent cell type upon Wntl stimulation [26]. C57MG
was the first cell line found to be morphologically transformed by a Wnt [27], and is
the most widely used cell line for Wnt transformation assays. C57MG, an
immortalized mammary epithelial cell line derived from normal mouse mammary
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
gland [28], changes from a epithelial cell morphology to a fibroblastic cell
morphology with loss o f contact inhibition [27]. However, these Wnt-transformed
cells are not tumorigenic in nude mice. Wntl was the first Wnt reported to be able to
transform C57MG in both autocrine and paracrine manners, which cofirmed that
Wnt functioned as a secretory protein exerting short-distance effects [29]. Most of
the known Wnt family members have been tested for their abilities of inducing
C57MG transformation [30-32]. Although Wnts share a conserved sequence and
structural sim ilarities, their transforming abilities on C57MG cells differ
dramatically. Based on their abilities to morphologically transform C57MG cells,
Wnts can be roughly grouped into three groups— highly transforming, moderately
transforming and non-transforming [30]— although the grouping of Wnts is slightly
different in different studies [30,32].
It appears now that Wnts may initiate two different signaling pathways—the
Wnt/P-catenin pathway and the Wnt/Ca2 + pathway— depending on the Wnt and,
possibly, the cells involved [33]. The Wnt signaling will be discussed in more detail
in Chapter III of this dissertation. W ntl, a highly transforming Wnt, initiates the
Wnt/P-catenin pathway and induces TCF (T cell factor) dependent transcription.
Wnt5a, a non-transforming Wnt, uses the Wnt/Ca2 + pathway and can function as an
antagonist to Wntl signaling [33]. Thus, the transforming ability of Wnts not only
gives a hint of their roles in tumorigenesis, but also are very useful in studying the
signaling pathway that is specific for the Wnt under study. For the new Wnt genes
identified in this laboratory, the study o f their transforming potential and signaling
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pathway will give a better understanding o f their roles in tumorigenesis.
Furthermore, the factors that affect the transforming capacity and their effects on
other cell lines such as PC 12 have been studied as well.
MATERIALS AND METHODS:
Cell lines and reagents
C57MG mouse mammary epithelial cells, derived from a normal mammary
gland [28], were grown in Dulbecco's modified Eagle's medium (DMEM; Irvine
Scientific, Santa Ana, CA) containing 10% fetal calf serum (FCS) (Omega
Scientific, Tarzana, CA) and 10 pg/ml insulin (Sigma, St. Louis, MO) with penicillin
and streptomycin. All cells were grown in humidified incubators with 7% CO2 at
37°C unless otherwise stated.
PC 12 cells, a rat neuronal cell line established from a transplantable adrenal
pheochromocytoma [34], were grown in DMEM containing 10% horse serum, 5%
FCS and antibiotics.
Acidic fibroblast growth factor (aFGF) and basic fibroblast growth factor
(bFGF) (Calbiochem, San Diego, CA) stock solutions (10 pg/ml) were prepared in
PBS, and aliquots were kept at -80°C before use.
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The HA (hemagglutin) epitope tagged Wnt expression constructs using pMirh
mammalian expression vectors
HA, an eleven amino acid epitope tag, was tagged to Wnt proteins at the C-
terminals. The coding region of WntlOa was directly amplified from pBS-WntlOa
cDNA using P/i/-PCR (Stratagene, La Jolla, CA) with primers annealing to the start
codon (5 ’-CCATGGGCAGCGCCCACCCT-3’> and the stop codon (5 ’-
GGCCATGGACTTGCAGACGCTGACCCA-3’). The primers add Nco I sites at
both the start and stop codons, which allows the WntlOa coding region to be cloned
in-frame into the pBS/Wntl-HA plasmid (gift from Dr. Jan Kitajewski, Columbia
University, NY). The resulting plasmid contains the Wntl leader sequence followed
by WntlOa coding region with an in-frame HA tag and stop codon. The entire region
was then cloned into pMirb (gift of D.M. Omitz, Washington University, St. Louis,
MO), that produces a bicistronic mRNA with an internal ribosome entry site (IRES)
for translation o f the downstream neomycin phosphotransferase gene. In the
pMirb/WntlOa-HA plasmid, transcription is under the control o f MoMLV-LTR
(Moloney murine leukemia virus LTR), and WntlOa-HA is translated from the
initiation site.
WntlOb and an internally deleted alternative splice variant WntlOb-S have
Nco I sites at 818T (WntlOb) and 534T ( WntlOb-S). These “T” nucleotides were
mutated via PCR to C without changing the coded amino acid. The 5’ region was
amplified using primers W10F1 (5’-GGCTTCGCCATGGTGGAGGAGCCC-3’) at
the start codon and WI0B1 (5’-TTGGCAGCTGCCTGACGTTCCGTGG-3’) at the
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T-C mutation region. The 3’ portion was amplified using primers W10F2 (5’-
AAGCGGAAGTGCAAATGCCACGGA-3’), which is complementary to primer
W 10B I at the m u tatio n reg io n , and p rim er W 10B2 ( 5 ’-
GGCCATGGATTTACACACATTGACCCA-3’) at the stop codon. The PCR
products were then annealed and PCR-extended to serve as the template. The PCR
with primers W10FI and W10B2 generated the complete coding region o f WntlOb
or WntlOb-S containing no internal Nco I sites. The fragments were then cloned into
pBS/W ntl-H A and subsequently cloned into pMirb sim ilarly to generate
pMirb/WntlOb-HA and pMirb/WntlOb-S-HA.
Transformation assays using pMirb transfection
Actively growing C57MG cells were seeded into six well plates 12 hours
before transfection. Cells were transfected with the Wnt-containing pMirb vector or
the empty pMirb vector using Lipofectamine (Life Technologies, Inc., Rockville,
MD) according to the manufacturer’s manual. The original transfected cells were
then plated at about 5-10% confluence into 10cm tissue culture dishes (Coming,
Acton, MA). At around 70% confluence, the medium was changed to medium
containing 0.1% FCS and 0.1 pg/ml insulin and the cells were incubated for focus
formation. Duplicate plates underwent selection in normal medium supplemented
with 700 pg/ml (net active conc.) G418 (Geneticin, Life Technologies, Inc.,
Rockville, MD) for 10-14 days. The G4l8-resistant colonies control for transfection
efficiency and were pooled as stably transfected cells. These pooled cells were later
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
grown to confluence, and the morphology o f the cells in normal medium was scored
and documented by phase-contrast photomicroscopy.
Preparation o f LNCX retroviral stocks
pLNCX is a retroviral expression vector in which the gene o f interest
(denoted as the symbol X) is driven by a cytomegalovirus promoter and the
neomycin resistance gene is driven by a MoMLV-LTR for the selection o f G418-
resistant colonies. Mouse WntlOa, WntlOb, WntlOb-S and W ntl cDNAs were
previously cloned into the pLNCX by former lab member, Jianwei Wang, to
generate pLNC-WntlOa (pLNCX containing WntlOa at position “X”), pLNC-
WntlOb, pLNC-WNtlOb-S and pLNC-W ntl. These constructs and the empty
pLNCX plasmid were transfected into \j/2 ecotropic retrovirus packaging cells using
Lipofectamine (Life Technologies, Inc., Rockville, MD). Transfected cells were
grown in DMEM containing 10% FCS (Omega Scientific, Tarzana, CA), antibiotics
and 600 pg/ml (net active conc.) G418 (Geneticin, Life Technologies, Inc.,
Rockville, MD) for 10-14 days. The G4l8-resistant colonies were pooled for
continuous production of viruses. The supernatants containing infectious virions
were collected from 70-100% confluent cells 24-48 hours after changing to fresh
medium without G418 and either used directly or aliquoted and stored at -20°C or -
80°C for future use.
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Transformation assays using LNCX retroviral infection
Infected C57MG cells were tested for transformation in parallel using a
focus-formation assay and a morphological analysis o f G418-selected cells as
described in Chapter I. Briefly, for the focus assay, C57MG cells were infected with
viruses and grown to near-confluence, then changed to medium containing 0.1%
FCS and 0.1 p.g/ml insulin. Cells remained at confluence in this medium for 5-10
days. Any resulting foci were scored, and typical foci were photographed. For the
morphological analysis assay, infected cells underwent G418 selection, and G418-
resistant colonies were pooled and grown in medium with G418. Cells were then
seeded into normal medium at 70-90% confluence, and the medium was changed to
medium containing 0.1% FCS and 0.1 |ig/ml insulin within 12 hours (The doubling
time of C57MG cells is about 18 hours). The morphology of the cells was scored
and documented by phase-contrast photomicroscopy 5 days later.
For the growth factor treatment o f cells in the focus formation assay, the
medium (with 0.1% FCS) contained either 1 ng/ml bFGF, 1 ng/ml aFGF, 10 |ig/ml
insulin or PBS as solvent control.
Northern blot analysis
Total RNAs from confluent pooled G418-selected cells were isolated using
Trizol (Life Technologies, Inc., Rockville, MD) according to the manufacturer's
instructions. Ten micrograms of total RNA were separated on an agarose gel and
blotted to a nylon membrane (Pharmacia, Piscataway, NJ) as previously described
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
[35]. The membranes were hybridized to a 32P-labeled probe generated from a
WntlOa cDNA. The membranes were exposed to Kodak X-omat film at -80°C for 48
hours. Membranes were then stripped of probe and sequentially rehybridized with
WntlOb or Wntl cDNA probes and exposed to films. Ethidium bromide-stained 18s
rRNA was used as loading control.
PC12 cell infection using LNCX retroviruses
Actively growing PC 12 cells were dislodged by pipetting and subcultured at
about 5-10% confluence into 10 cm culture dishes 18 hours before infection. Virus
stocks were diluted in fresh medium to a final volume of 4 ml and polybrene was
added to 8 pig/ml. Viruses were left on cells over 12 hours and then changed to fresh
medium. After recovery in fresh medium for 48 hours, PC 12 underwent G418
selection at a concentration of 1000 pg/ml for over 14 days until mock infected cells
were all dead. The cell morphology of the colonies was compared in the original
plates.
RESULTS
HA-tagged WntlO in the pMirb Expression System does not Morphologically
Transform C57MG Cells in Culture
To study the transforming ability o f the W ntlOa and W ntlO b genes, a
transfection model was used. C57MG cells were transfected with either pMirb/Wnt-
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
HA or pMirb empty vector, and the resulting cells were assayed for focus formation
and for morphological changes. In the focus formation assay (Fig. 7), cells
transfected with pBS DNA (Fig. 7A) and the pMirb empty vector (Fig. 7B) (negative
controls) all maintained the typical C57MG epithelial cell morphology with contact
inhibition. Cells remained as a monolayer and did not form foci. pMirb/Wntl-HA
transfected cells (Fig. 7C) formed strong foci in the cell monolayer, and cells at the
centers of the foci were elongated and morphologically transformed. pMirb/WntlOa-
HA (Fig. 7D), pMirb/WntlOb-HA (Fig. 7E) and pMirb-WntlOb-S-HA (Fig. 7F)
transfected cells did not form foci, although very mild disturbances of cell
morphology were occasionally observed.
Transfected cells were also selected for G418 resistance, and the resulted
colonies were pooled for morphological comparison (Fig. 8). The advantage of this
pMirb expressing system is that only one bicistronic transcript is generated, and
neomycin resistance could be used as a marker for W nt expression. Thus, the
selected cells should have the expression o f the W nt-H A genes introduced.
Unselected parental C57MG cells (Fig. 8A) and selected empty vector transfected
cells (Fig. 8B) were flat and maintained the typical cuboidal epithelial cell
morphology. pMirb/Wntl-HA transfected cells (Fig. 8C) showed transformation
from the flat cuboidal morphology to a spindle-shaped elongated morphology and
lost contact inhibition by forming multiple layers o f elongated cells to higher cell
density. pMirb/WntlOa-HA (Fig. 8D), pMirb/WntlOb-HA (Fig. 8E) and
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 7. Focus formation assay by overexpressing HA-tagged Wnt proteins in
C57MG cells. C57MG cells were transfected with pMirb mammalian cell
expression constructs for overexpression o f HA epitope-tagged Wnts. Cells were
grown without selection for G418 resistance, and allowed to remain at confluence in
low- serum medium for 5-10 days. (A) pBS transfected cells and (B) pMirb empty
vector transfected cells were negative controls. Cells exhibited the typical flat
cuboidal C57MG cell morphology, and no foci were observed. (C) Cells transfected
with pM irb/W ntl-H A , where a typical focus in the dish is shown. (D)
pMirb/WntlOa-HA transfected cells, (E) pMirb/Wnt 1 Ob-HA transfected cells and (F)
pMirb/WntlOb-S-HA transfected cells were similar to pBS transfected cells or empty
vector transfected cells; no foci were observed except minor morphological
disturbances.
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 8. Morphological comparisons of C57MG cells expressing HA-tagged
Wnt genes. After transfection with pMirb or pMirb/Wnt-HA, the C57MG cells were
selected in G418 and were subsequently grown as pools o f 200-400 infected
colonies. (A) Unselected parental cells and (B) pMirb empty vector transfected cells
in normal medium were flat with the characteristic cuboidal shape for CS7MG cells.
(C) pMirb/Wntl-HA transfected cells, as positive control, show typical transformed
cellular morphology. (D) Cells transfected with pMirb/WntlOa-HA, (E) cells
transfected with pMirb/WntlOb-HA and (F) cells transfected with pMirb/WntlOb-S-
HA were all basically similar to the negative controls. All cells are shown at full
confluence in 10% FCS-10 pg/ml insulin-DMEM.
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pM irb/W ntlOb-S-HA (Fig. 8F) transfected cells remained as monolayers at
confluence similar to that o f the parental cells and pMirb transfected cells.
In the focus formation assay, WntlOa-HA and the two isoforms of WntlOb-
HA failed to induce any foci whereas Wntl-HA was able to induce foci. Similarly,
in the morphological comparisons o f drug-selected pooled colonies, WntlOa-HA and
WntlOb-HA did not exhibit the transformed morphology as Wntl-HA did. Thus, as
compared to W ntl-H A, WntlOa-HA and the two variants o f WntlOb-HA do not
morphologically transform C57MG mammary epithelial cells.
Overexpression of WntlOa, WntlOb and WntlOb-S Using the LNCX Retroviral
Expression System do not Transform C57MG Cells
C57MG cells are very resistant to transfection, which may be due to their
very thick extracellular matrix. In addition, I occasionally observed a slight
morphological disturbance in cells transfected with the pMirb vector as compared
with parental cells. To confirm the transformation results I observed using the pMirb
system, I repeated the focus formation and morphological comparison experiments
using the LNCX retroviral expression system. The various retroviral constructs
containing the three Wnt cDNAs without an epitope-tag or the empty vector were
first transfected into the \|/2 packaging cell line. Viruses were collected to infect
C57MG cells. As expected, the infection efficiency o f C57MG cells was much
higher than their transfection efficiency.
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In the focus assay (Fig. 9), mock-infected cells (Fig. 9A) and cells infected
by the empty vector LNCX (Fig. 9B) remained flat in a monolayer and no focus was
ever observed (0.1% FCS in medium). Cells infected by LNC-Wntl viruses formed
strong foci as demonstrated (Fig. 9C). However, cells infected by LNC-WntlOa
viruses (Fig. 9D), LNC-WntlOb viruses (Fig. 9E) and LNC-WntlOb-S viruses had
morphologies essentially the same as the negative controls where cells remained flat
and did not form foci. Thus in this focus assay, WntlOa and WntlOb did not cause
C57MG transformation as did W ntl. This result is consistent with the result using
the pMirb system.
Duplicate plates of the same infected cells were treated with G418, and the
drug-resistant colonies were pooled. Pooled cells were allowed to grow to near
confluence, the medium was changed to low-serum medium (0.1% FCS), and the
cell morphologies were compared. The unselected parental cells (Fig. 10A) and cells
infected by the empty vector LNCX viruses (Fig. I OB) (negative controls) were flat
and remained as a monolayer. The LNC-Wntl infected cells (Fig. IOC) (positive
control) were transformed and piled up in multiple cell layers as we observed in the
pMirb system. The morphology of the LNC-WntlOa (Fig.lOD), LNC-WntlOb (Fig.
10E) or LNC-WntlOb-S (Fig. 10F) infected cells was similar to that o f the negative
controls. Cell morphologies in medium containing 10% FCS and 10 ng/ml insulin
were the same (not shown). Thus, overexpression o f WntlOa and the two isoforms
o f WntlOb using the LNCX retroviral system did not cause morphological
transformation of C57MG cells.
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 9. Focus form ation assay of overexpressing W nt proteins by the LNCX
retroviral infections. C57MG cells were infected with LNCX or LNC-Wnt viruses,
grown without selection to near confluence and allowed to remain at confluence in
low-serum medium for 5-10 days. (A) Cells in mock infection without viruse and
(B) cells infected by the empty vector viruses served as negative controls. Cells
remained as a flat monolayer where growth was inhibited by contact inhibition. (C)
Cells infected by LNC-W ntl viruses (positive control) formed large foci and a
typical focus is shown. (D) Cells infected by LNC-WntlOa viruses, (E) LNC-
Wnt I Ob or (F) LNC-Wnt I Ob-S were similar to the negative controls; although some
cells were elongated, they were never be able to form a focus.
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 10. Morphological comparison of C57MG cells infected with LNCX or
LNC-Wnt viruses. After infection with LNCX or different LNC-Wnt viruses, the
C57MG cells were selected in G418 and subsequently grown as pools o f 200-400
infected colonies. Cell morphology was compared at confluence in 0.1% FCS
medium. (A) Unselected parental cells or (B) the empty vector LNCX infected cells
were flat and formed single cell layer. (C) Cells infected by LNC-Wntl viruses
shows typical transformed cellular morphology. (D) The cells infected by LNC-
WntlOa, or (E) LNC-WntlOb or (F) LNC-WntlOb-S were flat and grew as a single
layer just as the negative controls.
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To verify the production of Wnt transcripts, total RNAs from G418-selected
infected cells were extracted and analyzed by Northern Blot analysis. Exogenous
expression o f the Wnt genes was readily detected using 10 fig o f total RNA (Fig. 11),
and no such expression was detected in the LNCX empty vector infected cells.
Probe
W ntl cDNA
WntlOa cDNA
WntlOb cDNA
18s rRNA,
EB staining
Figure li: Northern Blot analysis of Wnt gene expression in infected cells. Total
RNA samples were extracted from pooled G418 selected cells, and 10 |ig were gel-
separated and blotted to a nylon membrane. The blot was sequentially hybridized
with a WntlOa probe, WntlOb probe, and W ntl probe (as indicated on the right).
Proper transcripts were detected in the corresponding samples. No such transcripts
were detected in the empty vector LNCX negative control lane and in the sample
from LNC-Wnt9b, which serves as negative control for probe cross-reaction.
Ethidium bromide stained 18s rRNA served as loading control.
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Based on the results from both the pMirb plasmid system and the LNCX
retroviral system, WntlOa, WntlOb and WntlOb-S proteins do not induce focus
formation or significant morphological changes in C57MG cells as compared to
W ntl.
Low Concentration of FGFs may Cause Limited Enhancement of Wnt-induced
Focus Formation
Since WntlOa and WntlOb failed to induce the formation o f foci, I tested
some factors that may promote their transforming abilities. FGFs are well known
Wnt collaborators in tumorigenesis and animal development [20,36-39]. In a
transgenic mouse model, WntlOb has been previously shown to cooperate with Fgf2
in forming mammary tumors [20]. To test FGF’s effect on WntlOa and WntlOb
transformation, we chose acidic FGF and basic FGF to be added into the media. At
higher concentration, both FGFs were able to cause morphological alteration in
C57MG cells, which suggests the existence of their complete signaling components
in C57MG cells. Furthermore, aFGF is able to interact with all four FGF receptors
and initiates signal transduction, thus obviating the requirement for any particular
receptor profile in C57MG cells. Since a high concentration o f FGF itself is
sufficient to induce C57MG transformation, we chose to use only a very low
concentration (1 ng/ml), which does not cause obvious morphological changes in
these cells. A few small foci were observed in WntlOa and WntlOb infected cells
(Fig. 12, middle panels), but the incidence o f foci is below one out o f 200 colony
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
forming units (CFU, as calculated using the titers o f viral stocks). No change was
observed in LNC-WntlOb-S virus infected plates (Fig. 12, bottom panels). On the
contrary, the addition o f a normal concentration o f insulin (10 |xg/ml, typically
included in C57MG growth medium), generally increased the focus forming
background in all plates, and no significant differences in background foci numbers
were observed between empty vector control and WntlOa or WntlOb (not shown). It
appears that a low concentration of an FGF may cause some limited enhancement of
WntlOa and WntlOb focus formation in C57MG cells. Given the oncogenic
cooperation o f Wnts and FGFs in vivo [20,36,37,39], refinement of this cell culture
assay may allow a more detailed molecular analysis o f this cooperation.
Overexpression of WntlOa and WntlOb in PC12 Cells does not Cause a Flat Cell
Morphology as does Wntl
W ntl has been shown to cause PC 12 cells to change from a normal round
morphology to a flat adherent morphology [26]. The expression o f several genes
was altered by Wntl in this morphological transition [26,35]. Since WntlOa is
expressed in neuronal organs like brain, the overexpression o f WntlOa may be able
to affect the PC 12 cells of neuronal origin. I infected round PC 12 cells with LNC-
WntlOa, LNC-WntlOb, LNC-WntlOb-S viruses, LNC-Wntl positive control viruses
or LNCX empty vector control viruses. Infected cells were directly selected in the
original plates, and the colonies that appeared were compared for their morphology
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 12. The addition of basic FGF to the WntlO focus formation assay. Basic
FGF was added into the assay at final concentration of 1 ng/ml when the cells were
switched to low-serum medium. General cell morphology was not changed (top
panel) at this concentration. Small foci were observed in LNC-WntlOa plates (upper
middle panels) and LNC-WntlOb plates (lower middle panels), but not in the LNC-
WntlOb-S infected plates (bottom panels). The addition o f acidic FGF at the same
concentration yielded similar results (not shown).
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LNCX
WntlOa
WntlOb
Wnt10b-S
bFGF
PBS (1 ng/ml)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
after all the mock-infected cells died. The overexpression o f either WntlOa or
WntlOb apparently does not change PC 12 morphology like Wntl does (not shown).
Overexpression of WntlOa and WntlOb Genes does not Trigger the P-
catenin/TCF Signaling Pathway
A C57MG TOP-luciferase reporter cell line (CT cells, described in Chapter III) was
used to determine whether WntlOa or WntlOb proteins activate P-catenin/TCF
signaling by measuring luciferase activity. CT cells were infected with LNC-
WntlOa, LNC-WntlOb, LNC-WntlOb-S, LNC-Wntl positive control or LNCX
negative control viruses. After recovery in fresh medium for six hours, cells were
placed in serum-free DMEM medium for 24 hours before cells were lysed. Cell
lysates were assayed for luciferase activity and normalized by protein concentrations.
The reporter activities (as demonstrated by RLU, relative luciferase unit) of WntlOa
or WntlOb infected cells were similar to that of LNCX cells, whereas W ntl cells
generated a seven fold increase of reporter activity (Fig. 13). Similar experiments
using different CT clones generated similar W ntl activation, however, at different
levels (not shown), which is probably due to the clonal variation among these clones.
The insertion sites of the reporter construct may account for some o f this variation.
In conclusion, the overexpression of WntlOa and the two isoforms of WntlOb
does not morphologically transform C57MG, alter PC 12 morphology, or initiate P-
catenin/TCF signaling as W ntl does.
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
W n tlO a a n d W n tlO b p ro te in s d o n o t tr ig g e r
W n t/b e ta -c a te n in sig n a lin g
2 0 0 0 0
16000
16000
14000
3 12000
- J
O S toooo
eooo
6000
4000
2 0 0 0
c
L N C X W n t l W n t l OA W n t l OB W n t l O B -S
F igure 13. W ntlO a and W ntlO b proteins do not trig g er W nt/P-catenin
signaling. CT7 cells were transiently infected by LNC-WntlOa, LNC-WntlOb,
LNC-WntlObs, LNC-Wntl positive control or LNCX negative control viruses. The
cell lysates were tested for luciferase activity, normalized by protein concentrations
and plotted using Microsoft Excel. The error bars were generated by the Excel-
STDEV function.
DISCUSSION
Although WntlOa and WntlOb have been repeatedly demonstrated to be
involved in or associated with tumorigenesis [16-19,21-24,40], our results show that
WntlOa and WntlOb do not transform C57MG cells in culture and apparently do not
initiate P-catenin/TCF signal transduction. In addition, they appear not to induce a
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
flat morphology in PC 12 cells, contrary to W ntl. These results suggest that these
genes belong to the non-transforming group of Wnt genes. As noticed previously in
a similar research, transformation o f cultured cells does not necessarily correlate
with tumorigenesis, as demonstrated by Wnt7b, which is transforming but does not
promote tumor formation [41]. In the case of WntlOa and WntlOb, tumorigenicity
may not correlate with transformation in vitro due to several possibilities as
discussed in Chapter I. Cell transforming capacity o f Wnt may be affected by
several factors such as (i) the availability of the frizzled receptors and/or coreceptors
specific to the Wnt in question, (ii) the utilization of the Wnt/Ca2 + pathway and (iii)
the presence o f antagonistic factors that could inhibit such Wnt activity. The
profiling of frizzled expression and Wnt antagonists in C57MG cells, and the in-
depth characterization of Wnt signaling pathways will greatly help in understanding
Wnt transformation in vitro and tumorigenesis.
In addition, the lack of proper cooperative factors in cell culture as compared
to in vivo could also be a reason that WntlOa or WntlOb do not transform C57MG
cells. I tried adding a cooperating factor of Wnt, like FGF, into the focus formation
assay. The result demonstrated a limited enhancement of focus formation in WntlOa
and WntlOb infected cells but not in WntlOb-S cells. Due to the limited amount of
FGF that can be added to C57MG cells without causing morphological changes,
future in vitro evidence for Wnt and FGF cooperation may be gained by other means,
like the Wnt inducible luciferase reporter cell lines CT cells described in Chapter III.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
However, FGFs may cooperate with Wnts by means other than stimulating the Wnt
signaling pathways.
PC 12 cells, upon W ntl stimulation, change cell morphology, change
responsiveness to nerve growth factor [26], activate P-catenin pathway [42] and alter
gene expression [26,35,43,44]. My result shows that PC 12 cells cannot be altered by
retroviral introduction o f WntlOa or WntlOb, which maybe due to the reasons noted
above, such as receptors and/or coreceptors, signaling pathways and inhibitory
factors. In addition, most of the Wnt genes have not been tested in PC 12 cells, and
thus it may not be uncommon that PC 12 cells do not respond to certain Wnts.
I have generated a reporter cell line by transfecting C57MG cells with a Wnt-
responsive reporter construct (detail described in Chapter III) to allow the status of
Wnt/P-catenin signaling pathway to be assessed under various experimental
conditions. Although Wntl virus infection triggers reporter gene expression as
expected, WntlOa or WntlOb does not. The inability o f these genes to stimulate this
pathway may also be due to the lack of receptors or coreceptors or the presence o f
inhibitory factors. It is also possible that WntlOa and WntlOb, and non-transforming
Wnts like Wnt5a, may use a second (presumably non-transforming) Wnt pathway,
called Wnt/Ca2 + pathway, instead of the Wnt/(3-catenir. pathway. However, a
conclusive explanation cannot yet be made for the lack o f transformation by WntlOa
and WntlOb, since the possibilities like the lack o f receptors or the presence of
inhibitors have not been excluded, and the Wnt/Caz+ pathway has not been assessed
yet.
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
1. Cadigan, K. M., and Nusse, R. (1997). Wnt signaling: a common theme in
animal development. Genes Dev 11: 3286-3305.
2. Wodarz, A., and Nusse, R. (1998). Mechanisms o f Wnt signaling in
development. Annu Rev Cell Dev Biol 14: 59-88.
3. Polakis, P. (2000). Wnt signaling and cancer. Genes Dev 14: 1837-1851.
4. Wang, J., and Shackleford, G. M. (1996). Murine WntlOa and WntlOb:
cloning and expression in developing limbs, face and skin of embryos and in
adults. Oncogene 13: 1537-1544.
5. Kelly, G. M., Lai, C. J., and Moon, R. T. (1993). Expression of wntlOa in the
central nervous system of developing zebrafish. Dev Biol 158: 113-121.
6. Li, J., et al. (2001). Novel NEMO/IkappaB kinase and NF-kappa B target
genes at the pre-B to immature B cell transition. J Biol Chem 276: 18579-
18590.
7. Reddy, S., et al. (2001). Characterization o f Wnt gene expression in
developing and postnatal hair follicles and identification of Wnt5a as a target
of Sonic hedgehog in hair follicle morphogenesis. Mech Dev 107: 69-82.
8. St-Jacques, B., et al. (1998). Sonic hedgehog signaling is essential for hair
development. Curr Biol 8: 1058-1068.
9. Austin, T. W., Solar, G. P., Ziegler, F. C., Liem, L., and Matthews, W.
(1997). A role for the Wnt gene family in hematopoiesis: expansion of
multilineage progenitor cells. Blood 89: 3624-3635.
10. Van Den Berg, D. J., Sharma, A. K., Bruno, E., and Hoffman, R. (1998).
Role of members o f the Wnt gene family in human hematopoiesis. Biood 92:
3189-3202.
11. Sarkar, L., and Sharpe, P. T. (1999). Expression o f Wnt signalling pathway
genes during tooth development. Mech Dev 85: 197-200.
12. Dassule, H. R., Lewis, P., Bei, M., Maas, R., and McMahon, A. P. (2000).
Sonic hedgehog regulates growth and morphogenesis o f the tooth.
Development 127:4775-4785.
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13. Ishikawa, T., et al. (2001). Mouse Wnt receptor gene Fzd5 is essential for
yolk sac and placental angiogenesis. Development 128: 25-33.
14. Ross, S. E., et al. (2000). Inhibition o f adipogenesis by Wnt signaling.
Science 289:950-953.
15. Saitoh, A., Hansen, L. A., Vogel, J. C., and Udey, M. C. (1998).
Characterization o f Wnt gene expression in murine skin: possible
involvement o f epiderm is-derived Wnt-4 in cutaneous epithelial-
mesenchymal interactions. Exp Cell Res 243: 150-160.
16. Kirikoshi, H., Sekihara, H., and Katoh, M. (2001). Up-regulation of
WNT10A by tumor necrosis factor alpha and Helicobacter pylori in gastric
cancer. Int J Oncol 19: 533-536.
17. Kirikoshi, H., Sekihara, H., and Katoh, M. (2001). WNT10A and WNT6,
clustered in human chromosome 2q35 region with head-to- tail manner, are
strongly coexpressed in SW480 cells. Biochem Biophys Res Commun 283:
798-805.
18. Callahan, R. (1996). MMTV-induced mutations in mouse mammary tumors:
their potential relevance to human breast cancer. Breast Cancer Res Treat 39:
33-44.
19. Lane, T. F., and Leder, P. (1997). Wnt-lOb directs hypermorphic
development and transformation in mammary glands o f male and female
mice. Oncogene 15: 2133-2144.
20. Lee, F. S., Lane, T. F., Kuo, A., Shackleford, G. M., and Leder, P. (1995).
Insertional mutagenesis identifies a member of the Wnt gene family as a
candidate oncogene in the mammary epithelium o f int-2/Fgf-3 transgenic
mice. Proc Natl Acad Sci U S A 92: 2268-2272.
21. Bui, T. D., et al. (1997). A novel human Wnt gene, WNT 1 0B, maps to I2ql3
and is expressed in human breast carcinomas. Oncogene 14: 1249-1253.
22. Bui, T. D., Zhang, L., Rees, M. C., Bicknell, R., and Harris, A. L. (1997).
Expression and hormone regulation of Wnt2, 3, 4, 5a, 7a, 7b and 10b in
normal human endometrium and endometrial carcinoma. Br J Cancer 75:
1131-1136.
23. Hardiman, G., Kastelein, R. A., and Bazan, J. F. (1997). Isolation,
characterization and chromosomal localization o f human WNT10B.
Cytogenet Cell Genet 77: 278-282.
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24. Saitoh, T., Kirikoshi, H., Mine, T., and Katoh, M. (2001). Proto-oncogene
WNT I OB is up-regulated by tumor necrosis factor alpha in human gastric
cancer cell line MKN45. Int J Oncol 19: 1187-1192.
25. Rijsewijk, F., van Deemter, L., Wagenaar, E., Sonnenberg, A., and Nusse, R.
(1987). Transfection o f the int-1 mammary oncogene in cuboidal RAC
mammary cell line results in m orphological transform ation and
tumorigenicity. Embo J 6: 127-131.
26. Shackleford, G. M., Willert, K., Wang, J., and Varmus, H. E. (1993). The
Wnt-1 proto-oncogene induces changes in morphology, gene expression, and
growth factor responsiveness in PC 12 cells. Neuron 11: 865-875.
27. Brown, A. M., Wildin, R. S., Prendergast, T. J., and Varmus, H. E. (1986). A
retrovirus vector expressing the putative mammary oncogene int-1 causes
partial transformation o f a mammary epithelial cell line. Cell 46: 1001-1009.
28. Howard, D. K., Schlom, J., and Fisher, P. B. (1983). Chemical carcinogen-
mouse mammary tumor virus interactions in cell transformation. In Vitro 19:
58-66.
29. Jue, S. F., Bradley, R. S., Rudnicki, J. A., Varmus, H. E., and Brown, A. M.
(1992). The mouse Wnt-1 gene can act via a paracrine mechanism in
transformation of mammary epithelial cells. Mol Cell Biol 12: 321-328.
30. Wong, G. T., Gavin, B. J., and McMahon, A. P. (1994). Differential
transformation of mammary epithelial cells by Wnt genes. Mol Cell Biol 14:
6278-6286.
31. Christiansen, J. H., Monkley, S. J., and Wainwright, B. J. (1996). Murine
WNT11 is a secreted glycoprotein that morphologically transforms mammary
epithelial cells. Oncogene 12:2705-2711.
32. Shimizu, H., et al. (1997). Transformation by Wnt family proteins correlates
with regulation of beta-catenin. Cell Growth Differ 8: 1349-1358.
33. Miller, J. R., Hocking, A. M., Brown, J. D., and Moon, R. T. (1999).
Mechanism and function of signal transduction by the Wnt/beta-catenin and
Wnt/Ca2+ pathways. Oncogene 18: 7860-7872.
34. Greene, L. A., and Tischler, A. S. (1976). Establishment o f a noradrenergic
clonal line o f rat adrenal pheochromocytoma cells which respond to nerve
growth factor. Proc Natl Acad Sci U S A 73: 2424-2428.
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER III
COOPERATIVE FACTORS IN WNT1-INDUCED CELLULAR
TRANSFORMATION AND SIGNALING
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION
Wntl induced focus formation in C57MG mouse mammary epithelial cells is
a traditional model to study Wnt oncogenic activity in vitro [1-3]. C57MG cells
normally grow as a contact-inhibited monolayer and typically display a flat,
cubiodal-like epithelial morphology. However, upon Wnt stimulation, these cells
changed to an elongated morphology and overcome contact inhibition. Cells
infected with retrovirus expressing Wntl will pile up in multiple layers and form
visible foci in the monolayer o f uninfected cells. These foci expand and cells in the
centers will eventually detach from the plate. This experimental model is an
important tool to study oncogenic gene activity in cell culture. Different Wnts may
have different effects on C57MG cells as demonstrated in the previous two chapters,
and these differing effects may be due to the use o f alternate signaling pathways, as
discussed below. In addition, other factors may also play important roles in
determining the formation of foci. This chapter focused on one of the factors that
affect this assay and the mechanism o f its action.
The elucidation o f the Wnt signaling pathway is the key to understanding the
functional mechanism o f Wnts in both development and tumorigenesis. The latest
model of Wnt signaling (Figure 14 A) suggests the existence of at least two distinct
Wnt pathways [4,5]: the Wnt/P-catenin pathway and the Wnt/Ca2+ pathway. In the
Wnt/P-catenin pathway, as currently understood, Wnt proteins function as secretory
factors that bind to seven-transmembrane receptors o f the Frizzled family on the cell
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
surface. Through activation o f dishevelled, the glycogen synthase kinase-3 (GSK.-3)
activity is inhibited. Thus, p-catenin phosphorylation by GSK-3 is inhibited. The
hypophosphorylated P-catenin, unlike the phosphorylated version, is resistant to
proteosome-mediated protein degradation and accumulates in the cytoplasm. The
stabilized free P-catenin in the cytoplasm can form a complex with the transcription
factor T cell factor (TCF; also called lymphoid enhancer factor or LEF) and result in
the transcription o f target genes controlled by TCF binding sequences in their
promoter regions [5-7]. This Wnt/p-catenin pathway has been implicated in many
cancers including those of breast and colon [8,9].
An alternative pathway is called the Wnt/ Ca2 + pathway [10-12], utilized by
Wnt5a, is the pathway that involves Ca2+. The activation o f this pathway results in
the intracellular release o f Ca2 + through phospholipase C (PLC) and inositol-1,4,5-
triphosphate (IP3). The Ca2 + activates the Ca2 + -sensitive enzymes, Ca2 + -calmodulin-
dependent protein kinase II (CamKII) and protein kinase C (PKC). The cellular
responses are quite different from those o f Wnt/p-catenin pathway and some of them
are antagonistic to P-catenin pathway. This pathway is not as well characterized as
the Wnt/p-catenin pathway, and mainly studied in Dr. Moon’s laboratory (University
o f Washington, Seattle, WA) using zebrafish and xenopus as major model systems.
It is generally believed that cancer formation is a multistep process consisting
o f the accumulation of genetic mutations over a period o f time. Statistical and
molecular analyses have gathered evidence demonstrating the cooperation o f
oncogenes in this process [13]. As demonstrated in animal models, Wnt genes often
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
^ Inhibitory factors
Cooperative factors
[Ca2 + ]
( P-catenin )
Cellular responses
IM M P No Wnt Wnt
Wnt
Plasm a Membrane
Frizzled ♦
D ish ev eled
Frizzled
o o
r?0
Proteolysis
T C F
N uclear Membrane
Target
G e n e s I
5 F 5 r T «ra«‘
TCF-^ — L G e n e s
? M M P
Figure 14. Wnt signaling pathways and a hypothesized role of MMPs in Wnt
signaling. (A) The canonical Wnt/p-catenin pathway (left) and the Wnt/Ca2 +
pathway (right) (figure adapted from reference [4]). The horizontal arrows and
question marks highlight the factors that may affect signaling or pathway choice.
These factors include the different Wnts, Frizzled receptors, cellular contexts and/or
existing factors that inhibit or cooperate with either o f the pathways. (B) A
hypothesized role of MMPs in the Wnt/p-catenin pathway. MMP cleavage of E-
cadherin may release P-catenin, which could contribute to the Wnt-induced
accumulation of free cytoplasmic P-catenin and thus potentiate Wnt signaling. The
possible induction of MMPs by Wnt signaling could enhance the cooperation further.
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cooperate with FGF genes in tumorigenesis [14-19]. The molecular mechanism of
this collaboration is not clear. The FGFs are a family o f more than twenty
structurally related secretory growth factors [20]. Like Wnts, members of FGF
family are also the targets o f MMTV insertional activation in mammary
tumorigenesis [21,22]. MMTV infection of W ntl transgenic mice would preferably
activates FGF genes in mammary tumors [15,16]. Similarly, Wnt genes are activated
by MMTV insertions in FGF transgenic mice [18].
The FGF signal is transduced into the cells via its receptor. FGF receptors
(FGFR1-4) are transmembrane receptor tyrosine kinases that have different affinity
towards different FGF ligands and dimerize after ligand binding [23,24], The
dimerization is required for receptor kinase activity, and the resulting signal
transduction events, which utilize MAP kinases, lead to the transcriptional activation
of target genes. In this section, the collaboration between Wnts and FGFs was tested
in C57MG cellular transformation system.
P-catenin is the key component in Wnt/P-catenin signaling pathway. Without
Wnt stimulation, the majority of P-catenin proteins in the cells are bound to the E-
cadherin at the cell membrane and the free cytoplasmic P-catenin proteins are rapidly
degraded upon GSK-3 phosphorylation. Wnt signaling causes the cytoplasmic
accumulation o f free P-catenin through the inhibition of GSK-3 activity. Only the
free P-catenin enters the nucleus and couples with TCF to activate the transcription
of Wnt target genes. Thus, any attribution to this free P-catenin pool would
theoretically cooperate with Wnt signaling, and the release of E-cadherin-bound P-
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
catenin would be one of them. One family o f such molecules implicated in Wnt
signaling through this way is the family o f matrix metalloproteinases (MMPs).
MMPs are a class o f extracellular matrix-degrading enzymes and were first identified
as metastasis related factors [25]. Their roles in metastasis have been extensively
studied, whereas little is known about their potential roles in early tumorigenesis.
There is indirect evidence suggesting that MMPs could potentially
collaborate with Wnts in oncogenesis. First, two MMPs have been shown to cleave
E-cadherin [26-28], which may release P-catenin to the cytoplasm and augment the
stablization of P-catenin by Wnt signaling as mentioned above. Second, MMP-7
(Matrilysin) has been shown to be upregulated by P-catenin in colorectal cancer and
intestinal tumors [29,30]. MMP-7 is normally an epithelial cell-specific MMP
detected in mammary tumors [29,30]. In addition, since c-jun has recently been
shown to be upregulated by P-catenin in colorectal cells [31], other MMPs may also
be regulated by Wnt through their c-jun (A P-l, activator protein-1) responsive
promoters. MMP-3, also known as Stromelysin 1, is normally secreted by the
stromal cells in breast tissue and is able to cleave E-cadherin in vitro [26,28].
Overexpression o f a constitutively active form of MMP-3 in mice has been shown to
be able to accelerate mammary tumorigenesis[32].
No direct evidence shows that MMPs can potentiate Wnt signaling. One
important objective of my study is to use an in vitro cell culture system to study new
signaling components or cooperative factors like MMPs in Wnt signal transduction.
Although MMPs could participate in Wnt signaling in multiple ways, one working
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hypothesis is that MMPs may function in Wnt signaling by cleaving the extracellular
domains of cadherins to release intracellular (3-catenin for the formation o f (3-
catenin/TCF transcription complexes (Figure 14 B). If we find that MMPs are
upregulated by the Wnt/p-catenin pathway in mammary epithelial cells as they are in
colon cancer cells [30], this could further potentiate Wnt signaling in our model.
MATERIALS AND METHODS
Cell lines, reagents and constructs
C57MG cells were grown in DMEM containing 10% FCS and 10 pg/ml
insulin (normal medium) unless stated otherwise.
PC 12 cells expressing Wntl (PCl2/W ntl) and the empty vector control PC 12
cells (PC12/LNCX) were previously generated and characterized as described [33].
Their culture conditions were described in Chapter II.
Rati cells, an embryonic 3T3 like fibroblast cell line from the Fischer rat
[34], were grown in DMEM containing 10% FCS and antibiotics. Rati cells
expressing Wntl (Rati/W ntI) and empty vector control cells (Ratl/LNCX) were
previously generated in Dr. Shackleford’s laboratory.
293 cells, a cell line of primary human embryonal kidney transformed by
sheared human adenovirus type 5 DNA [35], were grown in DMEM with 10% FCS
and antibiotics. 293 cells expressing W ntl (293/W ntl) and empty vector control
cells (293/LNCX) were generated by infecting 293 cells with viruses from PA317
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
amphotropic packaging ceils. Infected cells were selected with 700 pg/ml G418, and
the drug-resistant colonies were pooled. Northern blot analysis was performed as
described in previous chapters to verify the transgene expression.
MMP inhibitors: GM6001 (also known as Ilomastat or N-[(2R)-2-
(hydroxamidocarbonylmethyl)-4-methylpentanoyl]-L-tryptophan methylamide) was
purchased from Chemicon (Temecula, CA) and was added to the medium at a final
concentration o f 20 pM. AG3340 was obtained from our collaborator, Dr. Yves
DeClerck (Childrens Hospital Los Angeles, CA), and was added to the medium at a
final concentration o f 100 nM. BB3103 [NI -(2,2-Dimethyl- 1-methylcarbamoyl-
propyl)-N4-hydroxy-2-isobutyl-3-(thiophene-2-sulfonyl-methyl)-succinamide] was
obtained from Dr. Tai-Lan Tuan (Childrens Hospital Los Angeles, CA) and added to
the medium at a final concentration o f 200 nM.
Purified recombinant MMP3 proprotein was purchased from Transduction
Laboratories (Lexington, Kentucky). MMP3 proprotein was activated prior to use
with APMA [4-(Acetoxymercurio) aniline p-Aminophenylmercuric acetate; Sigma,
St. Louis, MO) according to the manufacturer’s instructions. The final concentration
o f MMP3 added to the medium was 100 ng/ml. Purified recombinant activated
MMP7 protein was purchased from Chemicon (Temecula, CA) and added to the
medium at a final concentration o f 100 ng/ml.
Constructs: TOP-Luc was obtained from Dr. Hans Clevers (Netherlands)
[36]. The luciferase expression is under the control of five TCF binding sites and a
minimal promoter and is consequently upregulated by Wnt/(3-catenin-TCF signaling.
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The pHyg cotransfection plasmid was obtained from Dr. Y-H Song (UCLA, CA) and
the hygromycin B resistant gene was driven by a HSV-thymidine kinase (TK.)
promoter.
Infection o f C57MG cells with retroviral vectors and treatment with MMP
inhibitor—BB3103
Twelve hours before infection, C57MG cells were seeded at 10% confluence
in normal medium. The cells were infected by the retroviral vector LNCX carrying
Wntl or the empty vector LNCX for 12 hours and then incubated in fresh normal
medium for 24 hours after infection. When the cells grew to near confluence, the
medium was switched to DMEM medium containing 0.1% FCS and 0.1 pg/ml
insulin. The MMP inhibitor BB3103 (200 nM) or DMSO (as solvent control) was
also added at this time. Photos were taken six days after the addition of these
reagents.
Gelatin and casein zymograms
In 10-cm dishes, C57MG cells were infected with W ntl viruses or LNCX
empty viruses as described above and incubated 24 hours in fresh normal medium.
Cells were switched to 4 ml o f serum-free plain DMEM medium per dish, and
conditioned medium was collected 48 hours later. Otherwise, infected cells were
selected in G418 (700 |ig/ml), grown to near confluence and conditioned media were
collected 48 hours after switching to serum-free medium. Conditioned media were
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
either directly used or concentrated about 20-fold using Centricon 10 ultra-filters
(Amicon, Beverly, MA).
For the gelatin zymogram, conditioned media samples were mixed with
Laemmli sample buffer without reducing reagents and incubated at 37°C for 20
minutes. Samples were then separated on 8% sodium dodecyl sulfate (SDS)-
polyacrylamide gel containing 1 mg/ml gelatin (Sigma, St. Louis, MO). Upon
finishing electrophoresis, gels were washed twice in 2.5% Triton X-100, 30 minutes
each time, and then incubated at 37°C overnight in 56 mM Tris-HCl, pH 7.5,
containing 10 mM CaCli. Gels were stained in Coomassie Blue R-250 and
destained. Finally, gels were dried in a cellulose frame (Novex, Carlsbad, CA) after
equilibrating for 20 minutes in 30% methanol containing 5% glycerol.
For the casein zymogram, samples were separated in precast 4-16%
prestained Gradient Blue Casein Zymogram gel (Novex, Carlsbad, CA) (good for
protein sizes o f 10-200 kDa). After electrophoresis, gels were washed and incubated
as for the gelatin gels and directly visualized or dried.
Western blotting
C57MG cells that have been infected and G4l8-selected were grown to near
confluence in normal medium, except that one dish o f LNCX cells was treated with
PMA (4-a-Phorbol 12-myristate 13-acetate, Sigma, St. Louis, MO) at a final
concentration o f 100 ng/ml 24 hours prior to harvest. Cells were washed three times
with prechilled PBS and lysed in 25 mM Tris-HCl, pH7.5, containing 0.2% SDS and
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10 mM EDTA. Protein concentration was measured using the Bio-Rad DC protein
assay kit and used to normalize loading. Samples containing 10 ng o f protein were
separated on 10% SDS-polyacrylamide slab gels and transferred to nitrocellulose
membranes (Amersham, Piscataway, NJ). Membranes were then blocked for I hour
with TBST (50 mM Tris-HCl, pH7.5, 100 mM NaCl, 0.2% Tween 20) containing
5% non-fat milk and then incubated overnight at 4°C with polyclonal antibody
against MMP-7 (1:2,000 dilution, Chemicon, Temecula, CA) in the same solution.
Blots were washed with TBST five times for two minutes each wash, and blocked
with TBST containing 5% non-fat milk. Blots were incubated for 45 minutes with
horseradish peroxidase-conjugated, sheep anti-rabbit antibody (1:10,000 dilution,
Transduction Laboratories, Lexington, Kentucky) in the block solution. Bands were
visualized with an enhanced chemiluminescence detection kit (Amersham,
Piscataway, NJ) according to the manufacturer’s instructions.
Establishment o f CT (CS7MG-TOP) reporter cell lines
C57MG cells were seeded in six-well plates in normal medium 12 hours
before transfection. The cells were co-transfected with TOP-Luc and pHyg at 10:1
molecular ratio, respectively, using Lipofectamine. After incubated in fresh
medium, the cells were seeded into 10-cm dishes in normal medium with 200 ng/ml
hygromycin B (Life Technologies, Inc., Rockville, MD). Approximately 20 days
later, selected colonies were picked to 24-well plates (Coming, Acton, MA) and
grown in medium with hygromycin B. These clones were named CTl through
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CT34. Cells were expanded and frozen as stocks and subsequently underwent
screening for TOP-Luc integrated reporter cell lines as described below.
To screen the CT clones for those that contain inducible luciferase
expression, CT cells were seeded into six-well plate at 70% confluence, and
switched to serum-free plain DMEM medium the day after seeding. Cells were then
either given no treatment or treated with 10 mM LiCl or 20% serum and 10 mM LiCl
24 hours after medium switch. Cells were lysed 24 hours later and analyzed using
the Luciferase Assay System (Promega, Madison, WI). The relative luciferase units
(RLU) were then obtained by normalizing the luciferase reading with the protein
concentration as measured using the Bio-Rad NC protein assay kit. The luciferase
activity of CT cells should be induced by LiCl treatment and even greater induction
should be detected when serum is also included [37]. Lithium can inhibit GSK-3
activity, which mimics the Wnt signaling and thus activates downstream components
o f the Wnt signaling pathway [38].
Transient infection o f CT reporter cell lines
LNC-Wntl and LNCX empty vector viruses were collected as described in
Chapter II. CT cells were seeded into six-well plates at 50% confluence 12 hours
prior to infection. Cells were then infected with viruses in duplicates for 12 hours
and then allowed to recover in fresh medium for another 12 hours. Cells were
switched into serum-free plain DMEM medium 24 hours before cells were lysed and
analyzed for RLU as described above.
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Cocultivation using CT reporter cell lines
Wntl producing cells and corresponding vector control cells were generated
previously via LNC-Wntl or LNCX infection, G418 selection and pooling o f more
than 200 colonies. Wntl producing cells or control cells generated from four kinds
o f parental cells— PC 12, 293, Rati and C57MG— were used for coculture
experiments.
CT cells and Wntl producing cells were grown in appropriate medium to
near confluence. Cells were trypsinized and dispersed into single cells and counted
using a Counter Cell Counter. 3xl04 cells each o f CT cells and either Wntl
producing cells or control cells were evenly seeded into each well of six-well plates.
Cells were switched into serum-free plain DMEM medium within twelve hours and
treated with reagents six hours later. Cells were lysed 24 hours later and analyzed
for luciferase activity and normalized by protein concentration.
RESULTS
FGFs Promote Wntl-induced Focus Formation
FGFs and Wnts are strong collaborators in mouse mammary gland
tumorigenesis, as demonstrated in transgenic mice and by MMTV infection [15-19].
No such collaboration has been demonstrated in cell culture. In Wntl-induced focus
formation experiments, I added a very low concentration (1 ng/ml) of FGF proteins
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in the medium and observed an enhancement of focus formation (Figure 15). Since
FGF itself is able to transform C57MG cells at high concentration and even has a
stronger effect than W ntl, only a very low concentration could be used. At this
concentration used, no obvious morphological changes in control cells were
observed. Acidic FGF (aFGF) and basic FGF (bFGF) were chosen for their ability to
initiate FGF signaling in C57MG cells and for their availability. Using the dark
phase o f a phase contrast microscope at low magnification, the dark area is the flat
(nontransformed) cell monolayer and the bright areas show the round or elongated
spindle-shaped cells that have piled up into multiple cell layers (Fig. 15 A). In both
aFGF and bFGF-treated plates, focus formation was enhanced as demonstrated by
larger and brighter foci at both dilutions of Wntl viruses. At a higher magnification
using normal phase contrast, larger and more numerous foci were again observed
(Fig. 15B). Due to the limitation o f the concentration of FGF proteins that could be
used, this model cannot assess the collaboration between FGF and Wnt at higher
FGF concentrations, which might better mimic the in vivo conditions. However, the
addition o f insulin (normally included in C57MG culture medium) or serum
generally increased the focus forming ability non-specifically (not shown).
Nevertheless, a specific synergistic morphological response was observed.
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 15. Acidic FGF and basic FGF enhance the Wntl-induced focus-
formation. Acidic and basic FGF was added into the medium at a final
concentration o f 1 ng/ml at the time of switching the cells into low-serum medium.
No obvious cell morphological changes were observed in empty vector LNCX
infected cells (left panels) at this concentration. The addition of aFGF (middle panels
o f both A and B) and bFGF (bottom panels o f both A and B) increased the size and
number of Wntl-induced foci as compared to the PBS (the solvent control) treated
cells (upper panels o f A and B).
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A.
PBS
aFGF
1 ng/ml
bFGF
1 ng/ml
B.
PBS
aFGF
1 ng/ml
bFGF
1 ng/ml
LNCX
Wntl
5x10* cfu/ml
Wntl
105 cfu/ml
LNCX
Wntl
5X1CH C fU /m |
Wntl
10s cfu/ml
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V E C -T Q B = L M C X -
DMSO BB3103
W . N . T - 1 .
DMSO BB3103
Figure 16: The presence of BB3103 inhibits the W ntl-induced focus formation
of C57M G cells. The C57MG cells were seeded in normal medium at 10%
confluence, and infected by retroviruses to express W ntl. When the cells grew to
near confluence, the medium was switched to a low-serum medium with BB3103
(low concentration at 200 nM, whereas the suggested effective range is 100 nM-10
pM) or DMSO as solvent control. Photos were taken six days after the addition of
these reagents. The low magnification photos were taken with a 4X objective lens in
dark phase where rounded cells are bright and flat cells are dark. The high
magnification photos were taken with a 10X objective lens in normal phase where
the cell shape is readily seen. The presence of MMP inhibitor BB3103 (BB3103
panels) at low concentration in the culture medium is able to dramatically inhibit the
Wntl-induced formation and expansion of foci (DMSO panels).
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MMP Inhibitor BB3103 Inhibits the Wntl-induced Focus Formation of C57MG
Cells.
Wntl induced focus formation in C57MG ceils is the major experimental
system I used to study Wnt oncogenic activity in vitro. As proposed in the
introduction section (Fig. 14 B), we suspect MMPs might be able to cooperate with
Wnts in mammary tumorigenesis. We tested the hypothesis using the same C57MG
focus formation assay. In previous experiments, we observed that the factors in the
medium appeared to play important roles in the formation o f these foci. Since there
may be MMPs existing in the cell culture medium, we used MMP inhibitor to treat
the cells asking if the W ntl-induced focus formation could be affected. As
suspected, the addition of the MMP-specific inhibitor BB3103 significantly inhibited
the Wntl-induced focus formation (Figure 16). The BB3103 treated cells were less
transformed, stayed in the monolayer and generally form smaller foci, whereas
solvent control DMSO-treated cells formed typical strong Wntl foci that eventually
detached from the plates at their centers. Thus, the MMP function was required for
complete Wntl focus formation and expansion.
The Existence of MMPs in the C57MG Culture Media
There are more than 20 members of the MMP family, some of which can be
detected by gelatin or casein zymograms [25]. The gelatin zymogram is able to
detect MMP-2 and MMP-9 activity, whereas the normal casein zymogram is easily
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
able to detect MMP-3, and the casein aymogram using the gradient blue casein gel is
sensitive enough to barely detect MMP-7 activity. C57MG cells were transiently
infected with Wntl viruses or LNCX viruses, and one LNCX plate was treated with
TP A [12-O-tetradecanoylphorbol-13-acetate, here a similar compound with the same
effects— 4-a-Phorbol 12-myristate 13-acetate (PMA)— was used] before serum-free
conditioned media were collected. The conditioned media were analyzed with a
gelatin zymogram (Fig. 17A-i). MMP-2 activity was present in the media without a
significant change upon Wntl stimulation. Using 4-16% gradient blue casein gel,
the MMP-7 activity in the conditioned serum-free media collected from LNCX,
Wntl or TPA treated LNCX cells was barely visible (Fig. l7B-ii).
Pooled G4l8-selected C57MG cells that were infected or transfected with
Wntl-expression or control vectors were given to Dr. DeClerck’s laboratory and
analyzed by gelatin zymogram (Fig. 17A-ii). The serum-free conditioned media
from W ntl-expressing and transformed C57MG cells or empty vector cells were
collected, and the conditioned media were analyzed by gelatin zymogram. MMP-2
and MMP-9 activity were detected (Fig. 17A-ii). Same samples were also analyzed
in Dr. DeClerck’s Laboratory by casein zymogram showing the existence o f MMP-3
(Fig. l7B-i). MMP-2, -3, -9 and -7 are present in the medium; however, no
significant activity change was detected after W ntl stimulation.
MMP-7 is a small molecule (proform 28 kDa and active form 19 kDa) [39]. A
normal 8% SDS-PAGE gel containing 1% casein is likely to miss the MMP-7 band,
and the activity detected by the gradient casein zymogram was not conclusive
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 17. The existence of MMP activities in C57MG culture. (A -i)
Conditioned media collected from transiently infected cells with or without TPA
(PMA) treatment were collected and analyzed with a gelatin zymogram. In
unconcentrated samples, 8 |il LNCX (lane I) medium and 12 |xl Wntl medium (lane
2) were paired according to protein concentration where the 12 |xl pair (lanes 2 and
3) were normalized by cell numbers. In concentrated (~20x) samples, 12 |il samples
were loaded to each lane according to cell numbers. The conditioned medium from
NIH3T3 cells served as positive control for MMP-2. (A-ii) Conditioned media
collected from G418 selected pooled or cloned cells were also analyzed with gelatin
zymogram in Dr. DeClerck’s laboratory. MMP-2 and MMP-9 activities were
detected in all samples without significant changes. (B-i) The same samples were
also analyzed with a casein zymogram in Dr. DeClerck’s laboratory. MMP-3 activity
was detected in most o f the samples except parental cells. (B-ii) Infected and
selected pools o f LNCX and W ntl cells were grown and treated with TPA.
Conditioned media were analyzed in a precast 4-16% gradient prestained blue casein
gel (molecular weight range from 10-200 kDa). MMP-7 activity was barely visible.
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A. Gelatin Zymogram
i)
MMP-2
unconcentrated concentrated
<
£ L ^
8 ? 53 S 3 s =
Z G O G O > N z ^ Z '> > (N
— 5 ^ , J C J C . » S .
C M C M
M 5 -
ii)
MMP-9
MMP-2
Wntl Clones
o _ —
rt Z r ~
N C O ^
O O C O
UL LL LLI
B. Casein Zymogram
i)
MMP-3
Wntl Clones
N C O ^
o o C O
U . u_ L U
H )
MMP-7
LN C X Wntl LN C X -TPA
19KDa
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 18. MMP-7 and MMP-3 are expressed in G418-selected pools of both
Wntl producing cells and empty vector control cells. (A) G418-selected pools of
C57MG cells infected with LNCX or W ntl viruses were grown to near confluence
and switched to serum-free medium; one LNCX plate was treated with TPA (PMA
100 ng/ml). Cells were lysed 24 hours later and samples o f 10 fig protein were
separated on a 10% SDS-PAGE gel and transferred to a membrane. The blot was
hybridized to MMP-7 polyclonal antibody. (B) Selected pools or clones and
parental cells were grown and total RNA was extracted. Total RNA of 20 p.g was
separated and transferred to a nylon membrane. The blot was hybridized to a
radioactive labeled MMP-3 or Wntl cDNA probe. This Northern blot analysis was
performed in Dr. DeClerck’s laboratory.
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A. Western Blot
LNCX
MMP-7
W ntl
LNCX-
TPA
28 KDa
19 KDa
B. Northern Blot
MMP-3
Wntl
W nti Clones
N r t ^
o o co
LL LL U J
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
regarding quantitation. Therefore to detect the presence o f MMP-7 quantitatively,
Western blotting analysis was carried out (Figure 18A). Results confirmed the
presence of MMP-7 in the cultures, but no significant differential expression was
observed. Northern blot analysis for MMP-3 transcripts was also performed in Dr.
DeClerck’s laboratory (Figure 18B) and confirmed the expression of MMP-3.
These results show that MMPs are indeed produced by C57MG cells and
therefore are available to participate in Wnt signaling as we proposed. We did not
observe significant induction of any of the five MMPs tested here by W ntl, but this
is not a prerequisite for their involvement in Wnt signaling. Moreover, many other
untested MMPs are potential players in Wnt signaling.
The Generation of CT (C57MG TOP) Wnt Reporter Cell Lines
C57MG is the most widely used cell line in the study of Wnt functions, since
it morphologically responds to Wnt signals and preserves all the Wnt/P-catenin
signaling components. However, C57MG is also a cell line that is difficult to work
with because of its resistance to a lot o f experimental procedures including
transfection. In addition, its viscous extracellular matrix can interfere with the
collection o f conditioned medium from these cells, especially from older cultures.
Although other cell lines have been used in the study o f Wnts, they commonly lack a
morphological response to Wnt and they may even lead to alternative signal
transduction due to the different cellular context. Thus C57MG is still the best
cellular model for studying Wnt signaling in mammary tumorigenesis.
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To bypass the transfection difficulties, I established several cloned C57MG-
TOP (CT) reporter cell lines with the reporter construct stably integrated into the
cellular chromosome. The TOP-Luc reporter construct used here has five TCF
binding sites in the promoter region followed by luciferase reporter gene. The
luciferase expression should correspond to the Wnt/P-catenin signal.
Wntl was introduced to the cells either by transient infection with W ntl
retroviruses or by coculture with Wntl producing cells. Transiently infected CT
cells were starved in serum-free medium over 24 hours and then lysed for luciferase
activity (Fig. 19A). The reporter cells were also cocultured with Wnt-1 producing
cells, starved in serum-free medium for over 24 hours and analyzed for luciferase
activity (Fig. 19B). Significant inductions were detected using such cloned cell lines
in both testing systems.
MMP Inhibitor GM6001 is able to Partially Inhibit Wnt-1 Signaling
Based on our hypothesis, MMP inhibitor should show inhibition of Wnt
signaling if Wnt signaling indeed requires or collaborates with MMP. Using the
established CT reporter cells, we induced the W ntl signal with or without the MMP
inhibitor GM6001 and tested for the reporter activity. Preliminary data show that the
inhibitor was able to partially inhibit Wnt signaling through the (3-catenin/TCF
pathway up to 30% (Figure 20). The inhibition o f W ntl signaling by GM6001 is
readily detected in both the transient infection (Fig. 20A) and the cell coculture (Fig.
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20B) testing systems. These results suggest the Wnt/p-catenin-TCF signaling
pathway is enhanced in the presence of MMP activity.
MMP Inhibitor AG3340 Inhibits Wntl Signaling
GM6001 is an inhibitor o f some, but not all, MMPs. GM6001 is most potent for
collagenases [MMP-8 (Ki 0.1 nM) and MMP-1 (Ki 0.4 nM)] and gelatinases [MMP-
9 (Ki 0.2 nM), MMP-2 (Ki 0.5 nM)]. However, it is less effective against
stromelysin (MMP-3 Ki 27 nM). In addition, cell culture normally requires a
concentration up to 25 jiM, which may even be higher for C57MG due to the viscous
extracellular matrix. AG3340, a synthetic MMP inhibitor that is more specific to
MMP-2 and MMP-3 and which requires relatively low concentration, was also used
to test the inhibitory effect on Wnt/p-catenin signal transduction. Similar to the
results with GM6001, preliminary data show that AG3340 is also able to partially
inhibit Wnt signaling in CT7 cells cocultured with W ntl-producing C57MG cells
(Figure 21). The combination of AG3340 and GM6001 exhibits slightly greater
inhibition o f Wnt/p-catenin signaling, suggesting that their effects are additive and
that multiple MMPs may contribute to Wnt signaling.
MMP3 Potentiates Wntl Signaling
Evidence suggests that MMP-3 is a candidate MMP for involvement in Wnt
signaling. MMP-3 has been shown to be capable o f E-cadherin cleavage, which
could release E-cadherin-bound P-catenin to the cytoplasm [26,28]. Wntl signaling
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 19. CT clones respond to Wntl signal induced either by transient
infection or co-cultivation with Wntl-producing cells. (A) CT14 and CT20 cells
were transiently infected by W ntl or LNCX viruses and starved 24 hours before
harvest. Cell lysates were analyzed for luciferase activity and normalized by protein
concentrations. A 1.5-7 fold induction was observed in ten CT clones. (B) CT cells
were cocultured with different Wntl-producing cells and starved 24 hours before
harvest. Cell lysates were analyzed and the results were plotted in Excel. The error
bars or data ranges were generated using the STDEV function in Excel.
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A. Transient Infection
CT14 Cion* Tran* iont Infoction
180000
160000
140000
120000
100000
80000
60000
40000
20000
0
mm
8 K - 5 C
r n m m
LNCX Wht1
CT20 Clono T ransiontlnfoction
9000
8000
7000
6000
D 5000
C C 4000
3000
2000
1000
0
LNCX Wntl
B. Cell Co-cultivation
CT7-PC12 and CT14-293 Co-cultivation
90000
80000
70000
60000
3 50000
* 40000
30000
20000
10000
0
« 0 ? T * > 'v *
H * f £ 3 g
v-r.jLsi;
□ LNCX
■ Wntl
CT19-Rat1 and CT20-CS7MG Co-
cultivation
K U P S i
w m & k
□ LNCX
Q W W 1
CT7-PC12 CT14-293
CT19-Rat1 CT20-C57MG
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 20. The MMP inhibitor GM6001 partially inhibits Wnt signaling. (A)
CT cells were infected with Wntl or LNCX viruses and allowed to recover in fresh
medium. Cells were then starved in serum-free medium and treated with GM6001 at
a final concentration o f 20 |iM 24 hours before harvest. Cell lysates were tested for
luciferase activity and normalized RLU was plotted using Microsoft-Excel. (B) CT
cells were cocultured with different Wntl-producing cells at 1:1 ratio. Twelve hours
after seeding, cells were starved for six hours before the treatment with the inhibitor.
Cells were lysed 24 hours later and analyzed. Results were plotted accordingly.
GM6001 exhibited partial inhibition of Wntl/P-catenin signaling.
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A. Transient Infection
CT7 Clone Transient
Infection
180000
160000
140000
120000
100000
80000
60000
40000
20000
0
iK £!© g|g
ife ■ * - r > ”• * , ! « ' ? , < « » « * { • ' ! . X
LNCX Wntl Wnt1-
GM6001
CT20 Clone Transient
Infection
9000
8000
7000
6000
s 5000
« 4000
3000
2000
1000
0
^gj|^v«v*iaag
t m m m m
S E W S
LNCX Wntl Wnl1-
GM6001
B. Cell Co-cultivation
CT7-PC12 and CT14-293 Co-cultivation
SONS
fit,\-t‘ P.
m m m r n m im m
0 LNCX
■ LNCX-GM6Q01
■ W ntl
■ Wnt1-GM6001-
CT7-PC12 CT14-293
CT19-Rat1 and CT20-C57MG Co-cultivation
W m M W -
a LNCX
LNCX-GM6001
Wntl
W ntl -GM6001
• - . ,-Jv ;■ £ $ # * “L '* , v & • S T - s i- £ - it t f Z iC , & » * £ :
CT 19-Ratl CT20-C57MG
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
AG3340 Inhibits Wnt-reporter Activity
120000
100000
80000
ID
- I 60000
0£
40000
20000
0
LNCX LNCX- Wntl W ntl- Wnt1-
AG3340 A63340 AG+GM
aiijPPM p
sjflj^gj& sgaw w fcaaB ^^
Sif#tStS®l«
|P«4§i
JiS^3iis
1 1
m&smi
1 ' - % J
Figure 21. AG3340 partially inhibits W nt/f)-catenin signal transduction. CT7
cells were cocultured with Wntl-producing cells or LNCX control cells. Cells were
starved, treated (100 nM final concentration of AG3340) and analyzed as in Figure
20. LNCX cells have basal level o f luciferase activity (left two bars). Wntl induces
approximately a 4-fold increase in this clone (middle bar). AG3340 inhibits the
induction, and the combination o f AG3340 and GM6001 further inhibits the
induction (right two bars).
could then stabilize these free (3-catenin proteins to induce greater effects in the cells.
To test for potential cooperation between Wntl and MMP-3, commercially available
purified MMP-3 was added to CT reporter cells when cocultivated with Wntl
producing cells. Preliminary results show a cooperation between W ntl and MMP-3
(Figure 22). The addition o f MMP-3 to LNCX coculture did not produce an obvious
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
induction o f luciferase expression, whereas the addition to the W ntl coculture
initiated a greater induction as compared to Wntl without MMP-3 (Fig. 22A). The
addition of AG3340 was able to abolish the additional induction caused by MMP-3.
The presence o f MMP-3 to the cells also promotes more transformed phenotype in
Wntl coculture (Fig. 22B), and this effect could be reversed by the addition of
AG3340. Thus, the addition of MMP inhibitor AG3340 was able to inhibit the
cooperation between Wntl and MMP-3.
MMP-7 (Matrilysin) is an epithelial cell-specific MMP detected in mammary
tumors and can be induced by Wnt signaling [29,30]. In addition, MMP-7, like
MMP-3, has been shown to cleave E-cadherin in vitro [27,28] and could be detected
in C57MG culture. Recombinant pre-activated MMP-7 was added to the medium at
a concentration of 100 ng/ml (the same concentration as MMP-3 used) for 30 hours,
and the effect on Wnt/P-catenin signaling was tested. The reporter activity, however,
was not significantly altered by MMP-7 (Figure 23). At this stage, without testing
higher concentrations of MMP-7, I cannot rule out the possibility that the MMP-7
concentration was not sufficient to affect Wnt signaling in C57MG cells.
Overexpression of the activated form of MMP-7 directly in the assay system may be
a preferred approach to meet the concentration requirement.
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 22. MMP-3 collaborates with W ntl in Wnt signaling and cellular
transformation. (A) Reporter CT cells were seeded together with Wntl-producing
or control C57MG cells and incubated for 12 hours in 10% serum medium. Cells
were then treated with AG3340 in serum-free medium for 24 hours. Activated
MMP-3 protein was then added to the medium, incubated for an additional 30 hours,
and cells were lysed for luciferase activity. After normalized by protein
concentrations, the RLU was plotted using Excel. (B) Pictures were taken of the cell
morphologies before the cells in (A) were lysed. The morphological effect o f MMP-
3 on these cells was reversed by AG3340.
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A
MMP3 potentiates Wnt signaling and can be inhibited
by M M P inhibitors
3 5 0 0 0
3 0 0 0 0
2 5 0 0 0
^ 20000
5 ?
1 5 0 0 0
10000
5 0 0 0
0
•dw ff
I P
te a s m
fevgsS BB V ^ S T ;®
fe s H R tea® fe a a S S S
r;~.'; ■ ? > > . / - » ;
N O N E A G 3 3 4 0 M M P 3 N O N E A G 3 3 4 0 M M P 3 M M P 3 * AG
LNCX Wntl
NONE MMP3 AG3340
Vector
LNCX
MMP3+AG3340
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MMP-7 at 100 ng/ml does not Significantly Affect W n tl Signaling
M M P 7 i n W n t / b e t a - c a t e n i n s i g n a l i n g
120000
100000
00000
60000
40000
20000 —
T i 1 --l" " ' I 1 1 --“ “ --1 -- I --1 1 I
NONE MMP7 MMP7- NONE AG3340 AG+GM MMP 7 MMP7- MMP7-
AG+GM AG AG+GM
LNCX Wntl
Figure 23. MMP-7 does not positively affect W nt/^-catenin signaling. CT7 cells
were cocultured with Wntl-producing cell or control cells and treated with pre
activated MMP-7 proteins (100 ng/ml final concentration, the same as MMP-3) for
30 hours in serum free medium similar to the experiment in Figure 22. Results were
plotted with Excel.
DISCUSSION
FG F genes and Wnt genes are well-known collaborators in tumorigenesis as
demonstrated in several experimental animal models [15-19]. For the first time in
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cell culture, we observed cooperative effects between FGFs and Wnts in a focus
formation assay, in spite o f the limited effects which is possibly due to the low
concentration o f FGF proteins that could be used. The limitations o f the focus assay
can probaly be resolved by using the CT luciferase reporter cell lines generated here.
In addition, the more quantitative nature o f the luciferase assay has obvious
advantages over the less quantitative focus formation assay. Pilot experiments
suggest that FGFs may enhance Wnt signaling in these cells, but results are not yet
conclusive.
The establishment of CT reporter cell lines is going to greatly facilitate future
studies on Wnt signaling. With the cocultivation system, Wnt or other signals are
readily introduced to the reporter cells with controllable dosage and time. However,
with an internal control like an integrated Renilla luciferase construct, the CT cell
lines could be improved. The better Wnt signaling reporter cell lines use the RJLU
generated by Renilla luciferase activity in the same test tube which is not only
convenient but also more accurate than a separate protein assay.
MMPs were identified as novel Wnt collaborators in this study and
demonstrated significant effects on Wnt signaling. However, the evidence was
largely indirect, since it was primarily obtained by the use o f synthetic MMP
inhibitors. The inhibition spectrum of these experimental reagents has not yet been
well characterized. Although purified MMP-3 protein added to cell cultures
stimulated Wnt signaling, the effective concentration, half-life and time course
response o f these proteins have not yet been established, partly due to the prohibitive
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cost o f the proteins. Alternatively, ectopic expression of preactivated forms of
MMP-3 and MMP-7 directly in the cell culture assay system may provide more
information on the collaboration. Furthermore, there are more than 20 members of
the MMP family members that have been identified. Only a small subset of the
MMPs have been tested so far, and thus it is possible that one o f the other MMPs
actually collaborates more specifically with W ntl. In addition, their effects on
WntlOa, Wnt I Ob and Wnt9b would be interesting to look at, since at least some of
these Wnts did not show significant activity in the luciferase reporter assay by
themselves.
Finally, whether MMPs and Wnts collaborate in vivo would be an interesting
topic to pursue in the future. Since MMP activity is also involved in tumor
progression and the in vivo environment of mammary epithelial cells normally
contains MMP activity, the cooperation between Wnts and MMPs may drive cells
undergo further oncogenic changes. Wntl by itself can only partially transform
C57MG cells, i.e., the cell growth properties and morphology can be changed by
W ntl, but the cells cannot grow in anchorage-independent conditions or form tumors
in nude mice. It would be interesting to test these, and other, parameters of
oncogenic transformation in C57MG cell lines that overexpress both MMP and Wnt
genes. If results suggest an oncogenic cooperation, future studies using bitransgenic
mice will provide us further proof of Wnt-MMP cooperation. The results from this
study will broaden our understanding of the Wnt/P-catenin pathway as well as the
mechanisms o f mammary tumorigenesis involving both Wnts and MMPs.
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
1. Wong, G. T., Gavin, B. J., and McMahon, A. P. (1994). Differential
transformation of mammary epithelial cells by Wnt genes. Mol Cell Biol 14:
6278-6286.
2. Brown, A. M., Wildin, R. S., Prendergast, T. J., and Varmus, H. E. (1986). A
retrovirus vector expressing the putative mammary oncogene int-1 causes
partial transformation of a mammary epithelial cell line. Cell 46: 1001-1009.
3. Shimizu, H., et al. (1997). Transformation by Wnt family proteins correlates
with regulation o f beta-catenin. Cell Growth Differ 8: 1349-1358.
4. Kuhl, M., Sheldahl, L. C., Park, M., Miller, J. R., and Moon, R. T. (2000).
The Wnt/Ca2+ pathway: a new vertebrate Wnt signaling pathway takes
shape. Trends Genet 16: 279-283.
5. Miller, J. R., Hocking, A. M., Brown, J. D., and Moon, R. T. (1999).
Mechanism and function of signal transduction by the Wnt/beta-catenin and
Wnt/Ca2+ pathways. Oncogene 18: 7860-7872.
6. Dierick, H., and Bejsovec, A. (1999). Cellular mechanisms o f wingless/Wnt
signal transduction. Curr Top Dev Biol 43: 153-190.
7. Arias, A. M., Brown, A. M., and Brennan, K. (1999). Wnt signalling:
pathway or network? Curr Opin Genet Dev 9: 447-454.
8. Polakis, P. (2000). Wnt signaling and cancer. Genes Dev 14: 1837-1851.
9. Barker, N., and Clevers, H. (2000). Catenins, Wnt signaling and cancer.
Bioessays 22: 961-965.
10. Liu, X., et al. (1999). Activation o f a frizzled-2/beta-adrenergic receptor
chim era promotes Wnt signaling and differentiation o f mouse F9
teratocarcinoma cells via Galphao and Galphat. Proc Natl Acad Sci U S A 96:
14383-14388.
11. Slusarski, D. C., Corces, V. G., and Moon, R. T. (1997). Interaction of Wnt
and a Frizzled homologue triggers G-protein-linked phosphatidylinositol
signalling. Nature 390: 410-413.
12. Slusarski, D. C., Yang-Snyder, J., Busa, W. B., and Moon, R. T. (1997).
Modulation o f embryonic intracellular Ca2+ signaling by Wnt-5A. Dev Biol
182: 114-120.
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13. Hunter, T. (1997). Oncoprotein networks. Cell 88: 333-346.
14. Kwan, H., et al. (1992). Transgenes expressing the W nt-l and int-2 proto
oncogenes cooperate during mammary carcinogenesis in doubly transgenic
mice. Mol Cell Biol 12: 147-154.
15. Shackleford, G. M., MacArthur, C. A., Kwan, H. C., and Varmus, H. E.
(1993). Mouse mammary tumor virus infection accelerates mammary
carcinogenesis in W nt-l transgenic mice by insertional activation of int-
2/Fgf-3 and hst/Fgf-4. Proc Natl AcadSci U S A 90: 740-744.
16. MacArthur, C. A., Shankar, D. B., and Shackleford, G. M. (1995). Fgf-8,
activated by proviral insertion, cooperates with the Wnt-1 transgene in
murine mammary tumorigenesis. J Virol 69: 2501-2507.
17. Li, Y., Hively, W. P., and Varmus, H. E. (2000). Use o f MMTV-Wnt-l
transgenic mice for studying the genetic basis o f breast cancer. Oncogene 19:
1002-1009.
18. Lee, F. S., Lane, T. F., Kuo, A., Shackleford, G. M., and Leder, P. (1995).
Insertional mutagenesis identifies a member o f the Wnt gene family as a
candidate oncogene in the mammary epithelium of int-2/Fgf-3 transgenic
mice. Proc Natl Acad Sci U S A 92: 2268-2272.
19. van Leeuwen, F., and Nusse, R. (1995). Oncogene activation and oncogene
cooperation in MMTV-induced mouse mammary cancer. Semin Cancer Biol
6: 127-133.
20. Omitz, D. M., and Itoh, N. (2001). Fibroblast growth factors. Genome Biol 2.
21. Peters, G., Kozak, C., and Dickson, C. (1984). Mouse mammary tumor virus
integration regions int-l and int-2 map on different mouse chromosomes. Mol
Cell Biol A: 375-378.
22. Callahan, R., and Smith, G. H. (2000). M M TV-induced mammary
tumorigenesis: gene discovery, progression to malignancy and cellular
pathways. Oncogene 19: 992-1001.
23. Goldfarb, M. (2001). Signaling by fibroblast growth factors: the inside story.
Sci STKE 2001: E37.
24. Omitz, D. M., et al. (1996). Receptor specificity of the fibroblast growth
factor family. J Biol Chem 271: 15292-15297.
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25. Duffy, M. J., Maguire, T. M., Hill, A., McDermott, E., and O'Higgins, N.
(2000). Metalloproteinases: role in breast carcinogenesis, invasion and
metastasis. Breast Cancer Res 2: 252-257.
26. Lochter, A., et al. (1997). Matrix metalloproteinase stromelysin-l triggers a
cascade o f m olecular alterations that leads to stable epithelial-to-
mesenchymal conversion and a premalignant phenotype in mammary
epithelial cells. J Cell Biol 139: 1861-1872.
27. Davies, G., Jiang, W. G., and Mason, M. D. (2001). Matrilysin mediates
extracellular cleavage o f E-cadherin from prostate cancer cells: a key
mechanism in hepatocyte growth factor/scatter factor-induced cell-cell
dissociation and in vitro invasion. Clin Cancer Res 7: 3289-3297.
28. Noe, V., et al. (2001). Release o f an invasion promoter E-cadherin fragment
by matrilysin and stromelysin-1. J Cell Sci 114: 111-118.
29. Crawford, H. C., et al. (1999). The metalloproteinase matrilysin is a target of
beta-catenin transactivation in intestinal tumors. Oncogene 18: 2883-2891.
30. Brabletz, T., Jung, A., Dag, S., Hlubek, F., and Kirchner, T. (1999). beta-
catenin regulates the expression o f the matrix metalloproteinase-7 in human
colorectal cancer. Am J Pathol 155: 1033-1038.
31. Mann, B., et al. (1999). Target genes of beta-catenin-T cell-factor/lymphoid-
enhancer-factor signaling in human colorectal carcinomas. Proc Natl Acad
Sci U S A 96: 1603-1608.
32. Stemlicht, M. D., Bissell, M. J., and Werb, Z. (2000). The matrix
metalloproteinase stromelysin-l acts as a natural mammary tumor promoter.
Oncogene 19: 1102-1113.
33. Shackleford, G. M., Willert, K., Wang, J., and Varmus, H. E. (1993). The
Wnt-1 proto-oncogene induces changes in morphology, gene expression, and
growth factor responsiveness in PC 12 cells. Neuron 11: 865-875.
34. Topp, W. C. (1981). Normal rat cell lines deficient in nuclear thymidine
kinase. Virology 113: 408-411.
35. Graham, F. L., Smiley, J., Russell, W. C., and N aim , R. (1977).
Characteristics o f a human cell line transformed by DNA from human
adenovirus type S. J Gen Virol 36: 59-74.
I ll
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36. Korinek, V., et al. (1997). Constitutive transcriptional activation by a beta-
catenin-Tcf complex in APC-/- colon carcinoma. Science 275: 1784-1787.
37. Chen, R. H., Ding, W. V., and McCormick, F. (2000). Wnt signaling to beta-
catenin involves two interactive components. Glycogen synthase kinase-
3beta inhibition and activation of protein kinase C .J Biol Chem 275: 17894-
17899.
38. Hedgepeth, C. M., et al. (1997). Activation o f the Wnt signaling pathway: a
molecular mechanism for lithium action. Dev Biol 185: 82-91.
39. Woessner, J. F., Jr. (1992). The small matrix metalloproteinase of the rat
uterus. Matrix Suppl 1: 58-63.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER IV
BIBLIOGRAPHY
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Arias, A. M., A. M. Brown, et al. (1999). “Wnt signalling: pathway or network?”
Curr Opin Genet Dev 9(4): 447-54.
Auch, D. and M. Reth (1990). “Exon trap cloning: using PCR to rapidly detect and
clone exons from genomic DNA fragments.” Nucleic Acids Res. 18(22): 6743-4.
Austin, T. W., G. P. Solar, et al. (1997). “A role for the Wnt gene family in
hematopoiesis: expansion of multilineage progenitor cells.” Blood 89(10): 3624-35.
Bafico, A., A. Gazit, et al. (1998). “Characterization o f Wnt-1 and Wnt-2 induced
growth alterations and signaling pathways in NIH3T3 fibroblasts.” Oncogene
16(21): 2819-25.
Barker, N. and H. Clevers (2000). “Catenins, Wnt signaling and cancer.” Bioessays
22(11): 961-5.
Bergstein, I., L. M. Eisenberg, et al. (1997). “Isolation of two novel WNT genes,
WNT14 and WNT15, one of which (WNT15) is closely linked to WNT3 on human
chromosome 17q2l.” Genomics 46(3): 450-8.
Blake, J. A., J. T. Eppig, et al. (2001). “The Mouse Genome Database (MGD):
integration nexus for the laboratory mouse.” Nucleic Acids Res. 29(1): 91-4.
Bocchinfuso, W. P., W. P. Hively, et al. (1999). “A mouse mammary tumor virus-
Wnt-1 transgene induces mammary gland hyperplasia and tumorigenesis in mice
lacking estrogen receptor-alpha.” Cancer Res. 59(8): 1869-76.
Brabletz, T., A. Jung, et al. (1999). “beta-catenin regulates the expression o f the
matrix metalloproteinase-7 in human colorectal cancer.” Am J Pathol 155(4): 1033-
8.
Bradbury, J. M., C. C. Niemeyer, et al. (1994). “Alterations of the growth
characteristics o f the fibroblast cell line C3H 10TI/2 by members o f the Wnt gene
family.” Oncogene 9(9): 2597-603.
Bradley, R. S., P. Cowin, et al. (1993). “Expression o f Wnt-1 in PC 12 cells results in
modulation o f plakoglobin and E-cadherin and increased cellular adhesion.” J Cell
Biol 123(6 Pt 2): 1857-65.
Brodsky, L. I., A. L. Drachev, et al. (1993). “A novel method o f multiple alignment
o f biopolymer sequences.” Biosystems 30(1-3): 65-79.
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Brodsky, L. I., V. V. Ivanov, et al. (1995). “GeneBee-NET: Internet-based server for
analyzing biopolymers structure.” Biochemistry 60: 923-928.
Brown, A. M., R. S. Wildin, et al. (1986). “A retrovirus vector expressing the
putative mammary oncogene int-1 causes partial transformation of a mammary
epithelial cell line.” Cell 46(7): 1001-9.
Bui, T. D., T. O'Brien, et al. (1998). “High expression of Wnt7b in human superficial
bladder cancer vs invasive bladder cancer.” Br. J. Cancer 77(2): 319-24.
Bui, T. D., J. Rankin, et al. (1997). “A novel human Wnt gene, WNT10B, maps to
12ql3 and is expressed in human breast carcinomas.” Oncogene 14(10): 1249-53.
Bui, T. D., L. Zhang, et al. (1997). “Expression and hormone regulation of Wnt2, 3,
4, 5a, 7a, 7b and 10b in normal human endometrium and endometrial carcinoma.” Br
J Cancer 75(8): 1131-6.
Cadigan, K. M. and R. Nusse (1997). “Wnt signaling: a common theme in animal
development.” Genes Dev. 11(24): 3286-305.
Callahan, R. (1996). “MMTV-induced mutations in mouse mammary tumors: their
potential relevance to human breast cancer.” Breast Cancer Res Treat 39(1): 33-44.
Callahan, R. and G. H. Smith (2000). “MMTV-induced mammary tumorigenesis:
gene discovery, progression to malignancy and cellular pathways.” Oncogene 19(8):
992-1001.
Calvo, R., J. West, et al. (2000). “Altered HOX and WNT7A expression in human
lung cancer.” Proc. Natl. Acad. Sci. U S A 97(23): 12776-81.
Chen, R. H., W. V. Ding, et al. (2000). “Wnt signaling to beta-catenin involves two
interactive components. Glycogen synthase kinase-3beta inhibition and activation of
protein kinase C.” J Biol Chem 275(23): 17894-9.
Chong, H., W. Starkey, et al. (1998). “A replication-competent retrovirus arising
from a split-function packaging cell line was generated by recombination events
between the vector, one of the packaging constructs, and endogenous retroviral
sequences.” J Virol 72(4): 2663-70.
Christiansen, J. H., S. J. Monkley, et al. (1996). “Murine WNT 11 is a secreted
glycoprotein that morphologically transforms mammary epithelial cells.” Oncogene
12(12): 2705-11.
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chumakov, K. M. and S. V. lushmanov (1988). “[A principle of maximum
topological similarity in molecular systematics].” Mol. Gen. Mikrobiol. Virusol.(3):
3-9.
Crawford, H. C., B. M. Fingleton, et al. (1999). “The metalloproteinase matrilysin is
a target of beta-catenin transactivation in intestinal tumors.” Oncogene 18(18): 2883-
91.
Cygan, J. A., R. L. Johnson, et al. (1997). “Novel regulatory interactions revealed by
studies o f murine limb pattern in Wnt-7a and En-l mutants.” Development 124(24):
5021-32.
Dahmen, R. P., A. Koch, et al. (2001). “Deletions of AXIN1, a component of the
WNT/wingless pathway, in sporadic medulloblastomas.” Cancer Res. 61(19): 7039-
43.
Dassule, H. R., P. Lewis, et al. (2000). “Sonic hedgehog regulates growth and
morphogenesis of the tooth.” Development 127(22): 4775-85.
Dassule, H. R. and A. P. McMahon (1998). “Analysis of epithelial-mesenchymal
interactions in the initial morphogenesis of the mammalian tooth.” Dev Biol 202(2):
215-27.
Davies, G., W. G. Jiang, et al. (2001). “Matrilysin mediates extracellular cleavage of
E-cadherin from prostate cancer cells: a key mechanism in hepatocyte growth
factor/scatter factor-induced cell-cell dissociation and in vitro invasion.” Clin Cancer
Res 7(10): 3289-97.
Dierick, H. and A. Bejsovec (1999). “Cellular mechanisms of wingless/Wnt signal
transduction.” Curr Top Dev Biol 43: 153-90.
Duffy, M. J., T. M. Maguire, et al. (2000). “Metalloproteinases: role in breast
carcinogenesis, invasion and metastasis.” Breast Cancer Res 2(4): 252-7.
Erdreich-Epstein, A. and G. M. Shackleford (1998). “Differential expression of Wnt
genes in normal and flat variants of PC 12 cells, a cell line responsive to ectopic
Wntl expression.” Growth Factors 15(2): 149-58.
Feamhead, N. S., M. P. Britton, et al. (2001). “The ABC o f APC.” Hum. Mol.
Genet. 10(7): 721-33.
Fritzsch, B., D. H. Nichols, et al. (1995). “Development of midbrain and anterior
hindbrain ocular motoneurons in normal and Wnt-1 knockout mice.” J. Neurobiol.
27(4): 457-69.
116
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fung, Y. K., G. M. Shackleford, et al. (1985). “Nucleotide sequence and expression
in vitro o f cDNA derived from mRNA o f int-1, a provirally activated mouse
mammary oncogene.” Mol. Cell. Biol. 5(12): 3337-44.
Gavin, B. J., J. A. McMahon, et al. (1990). “Expression of multiple novel Wnt-l/int-
1-related genes during fetal and adult mouse development.” Genes Dev. 4(12B):
2319-32.
Goldfarb, M. (2001). “Signaling by fibroblast growth factors: the inside story.” Sci
STKE 2001(106): E37.
Graba, Y., K. Gieseler, et al. (1995). “DWnt-4, a novel Drosophila Wnt gene acts
downstream of homeotic complex genes in the visceral mesoderm.” Development
121(1): 209-18.
Graham, F. L., J. Smiley, et al. (1977). “Characteristics o f a human cell line
transformed by DNA from human adenovirus type 5.” J Gen Virol 36(1): 59-74.
Greene, L. A. and A. S. Tischler (1976). “Establishment of a noradrenergic clonal
line o f rat adrenal pheochromocytoma cells which respond to nerve growth factor.”
Proc Natl Acad Sci U S A 73(7): 2424-8.
Hardiman, G., R. A. Kastelein, et al. (1997). “Isolation, characterization and
chromosomal localization of human WNT10B.” Cytogenet Cell Genet 77(3-4): 278-
82.
Hedgepeth, C. M., L. J. Conrad, et al. (1997). “Activation of the Wnt signaling
pathway: a molecular mechanism for lithium action.” Dev Biol 185(1): 82-91.
Howard, D. K., J. Schlom, et al. (1983). “Chemical carcinogen-mouse mammary
tumor virus interactions in cell transformation.” In Vitro 19(1): 58-66.
Huguet, E. L., J. A. McMahon, et al. (1994). “Differential expression of human Wnt
genes 2, 3, 4, and 7B in human breast cell lines and normal and disease states of
human breast tissue.” Cancer Res. 54(10): 2615-21.
Hunter, T. (1997). “Oncoprotein networks.” Cell 88(3): 333-46.
lozzo, R. V., I. Eichstetter, et al. (1995). “Aberrant expression o f the growth factor
Wnt-5A in human malignancy.” Cancer Res. 55(16): 3495-9.
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ishikawa, T., Y. Tamai, et al. (2001). “Mouse Wnt receptor gene Fzd5 is essential
for yolk sac and placental angiogenesis.” Development 128(1): 25-33.
Janson, K., E. D. Cohen, et al. (2001). “Expression o f DWnt6, DWntlO, and DFz4
during Drosophila development.” Mech Dev 103(1-2): 117-20.
Jiang, Z. and G. M. Shackleford (1999). “Mouse mammary tumor virus carrying a
bacterial supF gene has wild-type pathogenicity and enables rapid isolation of
proviral integration sites.” J. Virol. 73(12): 9810-5.
Jue, S. F., R. S. Bradley, et al. (1992). “The mouse Wnt-1 gene can act via a
paracrine mechanism in transformation of mammary epithelial cells.” Mol Cell Biol
12(1): 321-8.
Katoh, M. (2002). “Regulation o f WNT3 and WNT3A mRNAs in human cancer cell
lines NT2, MCF- 7, and MKN45.” Int J Oncol 20(2): 373-7.
Kelly, G. M., C. J. Lai, et al. (1993). “Expression o f wntlOa in the central nervous
system o f developing zebrafish.” Dev Biol 158(1): 113-21.
Kengaku, M., J. Capdevila, et al. (1998). “Distinct WNT pathways regulating AER
formation and dorsoventral polarity in the chick limb bud.” Science 280(5367):
1274-7.
Kirikoshi, H., S. Inoue, et al. (2001). “Expression of WNT10A in human cancer.” Int
J Oncol 19(5): 997-1001.
Kirikoshi, H. and M. Katoh (2002). “Molecular cloning and characterization o f
mouse W ntl4b, clustered with mouse Wnt3 in mouse chromosome 11.” Int J Mol
Med 9(2): 135-9.
Kirikoshi, H., H. Sekihara, et al. (2001). “WNT10A and WNT6, clustered in human
chromosome 2q35 region with head-to-tail manner, are strongly coexpressed in
SW480 cells.” Biochem. Biophys. Res. Commun. 283(4): 798-805.
Kirikoshi, H., H. Sekihara, et al. (2001). “Molecular cloning and characterization of
WNT14B, a novel member of the WNT gene family.” Int. J. Oncol. 19(5): 947-52.
Kirikoshi, H., H. Sekihara, et al. (2001). “Up-regulation o f WNT10A by tumor
necrosis factor alpha and Helicobacter pylori in gastric cancer.” Int J Oncol 19(3):
533-6.
118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Kirikoshi, H., H. Sekihara, et al. (2001). “Molecular cloning and characterization of
human WNT11 .” Int J Mol Med 8(6): 651-6.
Kispert, A., S. Vainio, et al. (1998). “Wnt-4 is a mesenchymal signal for epithelial
transformation of metanephric mesenchyme in the developing kidney.” Development
125(21): 4225-34.
Korinek, V., N. Barker, et al. (1997). “Constitutive transcriptional activation by a
beta-catenin-Tcf complex in APC-/- colon carcinoma.” Science 275(5307): 1784-7.
Kuhl, M., L. C. Sheldahl, et al. (2000). “The Wnt/Ca2+ pathway: a new vertebrate
Wnt signaling pathway takes shape.” Trends Genet. 16(7): 279-83.
Kwan, H., V. Pecenka, et al. (1992). “Transgenes expressing the Wnt-1 and int-2
proto-oncogenes cooperate during mammary carcinogenesis in doubly transgenic
mice.” Mol Cell Biol 12(1): 147-54.
Lako, M., S. Lindsay, et al. (2001). “Characterisation of Wnt gene expression during
the differentiation of murine embryonic stem cells in vitro: role o f Wnt3 in
enhancing haematopoietic differentiation.” Mech Dev 103(1-2): 49-59.
Lane, T. F. and P. Leder (1997). “Wnt-10b directs hypermorphic development and
transformation in mammary glands of male and female mice.” Oncogene 15(18):
2133-44.
Lee, F. S., T. F. Lane, et al. (1995). “Insertional mutagenesis identifies a member of
the Wnt gene family as a candidate oncogene in the mammary epithelium of int-
2/Fgf-3 transgenic mice.” Proc. Natl. Acad. Sci. U S A 92(6): 2268-72.
Lejeune, S., E. L. Huguet, et al. (1995). “Wnt5a cloning, expression, and up-
regulation in human primary breast cancers.” Clin. Cancer Res. 1(2): 215-22.
Li, J., G. W. Peet, et al. (2001). “Novel NEMO/IkappaB kinase and NF-kappa B
target genes at the pre-B to immature B cell transition.” J Biol Chem 276(21):
18579-90.
Li, Y., W. P. Hively, et al. (2000). “Use o f MMTV-Wnt-1 transgenic mice for
studying the genetic basis of breast cancer.” Oncogene 19(8): 1002-9.
Liu, X., T. Liu, et al. (1999). “Activation o f a frizzled-2/beta-adrenergic receptor
chimera promotes Wnt signaling and differentiation of mouse F9 teratocarcinoma
cells via Galphao and Galphat.” Proc Natl Acad Sci U S A 96(25): 14383-8.
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Lochter, A., S. Galosy, et al. (1997). “M atrix metalloproteinase stromelysin-1
triggers a cascade o f molecular alterations that leads to stable epithelial-to-
mesenchymal conversion and a premalignant phenotype in mammary epithelial
cells.” J Cell Biol 139(7): 1861-72.
MacArthur, C. A., D. B. Shankar, et al. (1995). “Fgf-8, activated by proviral
insertion, cooperates with the Wnt-1 transgene in murine mammary tumorigenesis.”
J. Virol. 69(4): 2501-7.
Mann, B., M. Gelos, et al. (1999). “Target genes o f beta-catenin-T cell-
factor/lymphoid-enhancer-factor signaling in human colorectal carcinomas.” Proc
Natl Acad Sci U S A 96(4): 1603-8.
McWhirter, J. R., S. T. Neuteboom, et al. (1999). “Oncogenic homeodomain
transcription factor E2A-Pbxl activates a novel WNT gene in pre-B acute
lymphoblastoid leukemia.” Proc. Natl. Acad. Sci. U S A 96(20): 11464-9.
Miller, A. D. and G. J. Rosman (1989). “Improved retroviral vectors for gene
transfer and expression.” Biotechniques 7(9): 980-2, 984-6, 989-90.
Miller, J. R., A. M. Hocking, et al. (1999). “Mechanism and function o f signal
transduction by the Wnt/beta-catenin and Wnt/Ca2+ pathways.” Oncogene 18(55):
7860-72.
Moon, R. T., J. D. Brown, et al. (1997). “WNTs modulate cell fate and behavior
during vertebrate development.” Trends Genet. 13(4): 157-62.
Moon, R. T., J. D. Brown, et al. (1997). “Structurally related receptors and
antagonists compete for secreted Wnt ligands.” Cell 88(6): 725-8.
Naylor, S., M. J. Smalley, et al. (2000). “Retroviral expression o f Wnt-1 and Wnt-7b
produces different effects in mouse mammary epithelium.” J Cell Sci 113(Pt 12):
2129-38.
Nielsen, H., J. Engelbrecht, et al. (1997). “A neural network method for
identification o f prokaryotic and eukaryotic signal peptides and prediction o f their
cleavage sites.” Int. J. Neural. Syst. 8(5-6): 581-99.
Nielsen, H., J. Engelbrecht, et al. (1997). “Identification o f prokaryotic and
eukaryotic signal peptides and prediction o f their cleavage sites.” Protein Eng. 10(1):
1- 6 .
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Noe, V., B. Fingleton, et al. (2001). “Release of an invasion promoter E-cadherin
fragment by matrilysin and stromelysin-1.” J Cell Sci 114(Pt 1): 111-118.
Nusse, R. (2001). “An ancient cluster of Wnt paralogues.” Trends Genet 17(8): 443.
Nusse, R. and H. E. Varmus (1982). “Many tumors induced by the mouse mammary
tumor virus contain a provirus integrated in the same region of the host genome.”
Cell 31(1): 99-109.
Omitz, D. M. andN. Itoh (2001). “Fibroblast growth factors.” Genome Biol 2(3).
Omitz, D. M., J. Xu, et al. (1996). “Receptor specificity of the fibroblast growth
factor family.” J Biol Chem 271(25): 15292-7.
Parr, B. A. and A. P. McMahon (1998). “Sexually dimorphic development of the
mammalian reproductive tract requires Wnt-7a.” Nature 395(6703): 707-10.
Patapoutian, A. and L. F. Reichardt (2000). “Roles of Wnt proteins in neural
development and maintenance.” Curr. Opin. Neurobiol. 10(3): 392-9.
Peters, G., C. Kozak, et al. (1984). “Mouse mammary tumor virus integration
regions int-1 and int-2 map on different mouse chromosomes.” Mol Cell Biol 4(2):
375-8.
Polakis, P. (2000). “Wnt signaling and cancer.” Genes Dev. 14(15): 1837-51.
Porfiri, E., B. Rubinfeld, et al. (1997). “Induction of a beta-catenin-LEF-1 complex
by wnt-1 and transforming mutants of beta-catenin.” Oncogene 15(23): 2833-9.
Ramachandran, B., K. Houben, etal. (1993). “Differential expression of transporters
for norepinephrine and glutamate in wild type, variant, and WNT 1-expressing PC 12
cells.” J Biol Chem 268(32): 23891-7.
Reddy, S., T. Andl, et al. (2001). “Characterization of Wnt gene expression in
developing and postnatal hair follicles and identification of Wnt5a as a target of
Sonic hedgehog in hair follicle morphogenesis.” Mech Dev 107(1-2): 69-82.
Reya, T., M. O'Riordan, et al. (2000). “Wnt signaling regulates B lymphocyte
proliferation through a LEF-1 dependent mechanism.” Immunity 13(1): 15-24.
Rijsewijk, F., L. van Deemter, et al. (1987). “Transfection o f the int-1 mammary
oncogene in cuboidal RAC mammary cell line results in morphological
transformation and tumorigenicity.” Embo J 6(1): 127-31.
121
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rocheleau, C. E., W. D. Downs, et al. (1997). “Wnt signaling and an APC-related
gene specify endoderm in early C. elegans embryos.” Cell 90(4): 707-16.
Roelink, H., E. Wagenaar, et al. (1990). “Wnt-3, a gene activated by proviral
insertion in mouse mammary tumors, is homologous to int-1/Wnt-1 and is normally
expressed in mouse embryos and adult brain.” Proc Natl Acad Sci U S A 87(12):
4519-23.
Ross, S. E., N. Hemati, et al. (2000). “Inhibition o f adipogenesis by Wnt signaling.”
Science 289(5481): 950-3.
Rowe, L. B., J. H. Nadeau, et al. (1994). “Maps from two interspecific backcross
DNA panels available as a community genetic mapping resource.” Mamm. Genome.
5(5): 253-74.
Saitoh, A., L. A. Hansen, et al. (1998). “Characterization of Wnt gene expression in
murine skin: possible involvement o f epidermis-derived Wnt-4 in cutaneous
epithelial- mesenchymal interactions.” Exp Cell Res 243(1): 150-60.
Saitoh, T., M. Hirai, et al. (2001). “Molecular cloning and characterization of
WNT3A and WNT14 clustered in human chromosome lq42 region.” Biochem.
Biophys. Res. Commun. 284(5): 1168-75.
Saitoh, T., H. Kirikoshi, et al. (2001). “Proto-oncogene WNT10B is up-regulated by
tumor necrosis factor alpha in human gastric cancer cell line MKN45.” Int J Oncol
19(6): 1187-92.
Sarkar, L. and P. T. Sharpe (1999). “Expression o f Wnt signalling pathway genes
during tooth development.” Mech Dev 85(1-2): 197-200.
Satoh, S., Y. Daigo, et al. (2000). “AXIN1 mutations in hepatocellular carcinomas,
and growth suppression in cancer cells by virus-mediated transfer o f AXIN1.” Nat.
Genet. 24(3): 245-50.
Shackleford, G. M., A. Ganguly, et al. (2001). “Cloning, expression and nuclear
localization o f human NPM3, a member o f the nucleophosmin/nucleoplasmin family
of nuclear chaperones.” BMC Genomics 2(1): 8.
Shackleford, G. M., C. A. Mac Arthur, et al. (1993). “Mouse mammary tumor virus
infection accelerates mammary carcinogenesis in Wnt-1 transgenic mice by
insertional activation of int-2/Fgf-3 and hst/Fgf-4.” Proc. Natl. Acad. Sci. U S A
90(2): 740-4.
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Shackleford, G. M. and H. E. Varmus (1987). “Expression of the proto-oncogene int-
1 is restricted to postmeiotic male germ cells and the neural tube o f mid-gestational
embryos.” Cell 50(1): 89-95.
Shackleford, G. M., K. Willert, et al. (1993). “The Wnt-1 proto-oncogene induces
changes in morphology, gene expression, and growth factor responsiveness in PC 12
cells.” Neuron 11(5): 865-75.
Shimizu, H., M. A. Julius, et al. (1997). “Transformation by Wnt family proteins
correlates with regulation of beta-catenin.” Cell Growth Differ. 8(12): 1349-58.
Sidow, A. (1992). “Diversification of the Wnt gene family on the ancestral lineage o f
vertebrates.” Proc. Natl. Acad. Sci. U S A 89(11): 5098-102.
Sidow, A. (1996). “Gen(om)e duplications in the evolution o f early vertebrates.”
Curr. Opin. Genet. Dev. 6(6): 715-22.
Slusarski, D. C., V. G. Corces, et al. (1997). “Interaction of Wnt and a Frizzled
homologue triggers G-protein-linked phosphatidylinositol signalling.” Nature
390(6658): 410-3.
Slusarski, D. C., J. Yang-Snyder, et al. (1997). “M odulation o f embryonic
intracellular Ca2+ signaling by Wnt-5A.” Dev Biol 182(1): 114-20.
Smalley, M. J. and T. C. Dale (2001). “Wnt signaling and mammary tumorigenesis.”
J. Mammary Gland Biol. Neoplasia 6(1): 37-52.
Smith, K., T. D. Bui, et al. (1999). “Up-regulation o f macrophage wnt gene
expression in adenoma-carcinoma progression of human colorectal cancer.” Br. J.
Cancer 81(3): 496-502.
St-Jacques, B., H. R. Dassule, et al. (1998). “Sonic hedgehog signaling is essential
for hair development.” Curr Biol 8(19): 1058-68.
Stark, K., S. Vainio, et al. (1994). “Epithelial transformation o f metanephric
mesenchyme in the developing kidney regulated by Wnt-4.” Nature 372(6507): 679-
83.
Stemlicht, M. D., M. J. Bissell, et al. (2000). “The matrix metalloproteinase
stromelysin-1 acts as a natural mammary tumor promoter.” Oncogene 19(8): 1102-
13.
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Taipale, J. and P. A. Beachy (2001). “The Hedgehog and Wnt signalling pathways in
cancer.” Nature 411(6835): 349-54.
Topp, W. C. (1981). “Normal rat cell lines deficient in nuclear thymidine kinase.”
Virology 113(1): 408-11.
Torres, M. A., J. A. Yang-Snyder, et al. (1996). “Activities of the W nt-l class of
secreted signaling factors are antagonized by the Wnt-5A class and by a dominant
negative cadherin in early Xenopus development.” J Cell Biol 133(5): 1123-37.
Tsukamoto, A. S., R. Grosschedl, et al. (1988). “Expression of the int-l gene in
transgenic mice is associated with mammary gland hyperplasia and adenocarcinomas
in male and female mice.” Cell 55(4): 619-25.
Tucker, A. S., D. J. Headon, et al. (2000). “Edar/Eda interactions regulate enamel
knot formation in tooth morphogenesis.” Development 127(21): 4691-700.
Van Den Berg, D. J., A. K. Sharma, et al. (1998). “Role of members of the Wnt gene
family in human hematopoiesis.” Blood 92(9): 3189-202.
van Leeuwen, F. and R. Nusse (1995). “Oncogene activation and oncogene
cooperation in MMTV-induced mouse mammary cancer.” Semin Cancer Biol 6(3):
127-33.
Vider, B. Z., A. Zimber, et al. (1996). “Evidence for the involvement of the Wnt 2
gene in human colorectal cancer.” Oncogene 12(1): 153-8.
Wang, J. and G. M. Shackleford (1996). “Murine WntlOa and WntlOb: cloning and
expression in developing limbs, face and skin of embryos and in adults.” Oncogene
13(7): 1537-44.
Webster, M. T., M. Rozycka, et al. (2000). “Sequence variants of the axin gene in
breast, colon, and other cancers: an analysis of mutations that interfere with GSK3
binding.” Genes Chromosomes Cancer 28(4): 443-53.
Wodarz, A. and R. Nusse (1998). “Mechanisms of Wnt signaling in development.”
Annu Rev Cell Dev Biol 14: 59-88.
Woessner, J. F., Jr. (1992). “The small matrix metalloproteinase of the rat uterus.”
Matrix Suppl 1: 58-63.
Wong, G. T., B. J. Gavin, et al. (1994). “Differential transformation o f mammary
epithelial cells by Wnt genes.” Mol. Cell Biol. 14(9): 6278-86.
124
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Wunder, J. S., K. Eppert, et al. (1999). “Co-amplification and overexpression of
CDK4, SAS and MDM2 occurs frequently in human parosteal osteosarcomas.”
Oncogene 18(3): 783-8.
Zhuang, L., Z. Zhang, et al. (1999). “Aberrant expression of growth factor Wnt-5A
in six urinary malignant cell lines.” Chin. Med. J. (Engl) 112(3): 251-5.
Zorn, A. M. (2001). “Wnt signalling: antagonistic Dickkopfs.” Curr. Biol. 11(15):
R592-5.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Asset Metadata

Creator Qian, JunQing (author)

Core Title Wnt factors: Genes, functions and cooperative signaling in mammary cell transformation

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Molecular Microbiology and Immunology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biology, cell,biology, molecular,health sciences, oncology,OAI-PMH Harvest

Language English

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-229367

Unique identifier UC11339111

Identifier 3073836.pdf (filename),usctheses-c16-229367 (legacy record id)

Legacy Identifier 3073836.pdf

Dmrecord 229367

Document Type Dissertation

Rights Qian, JunQing

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA