‘ Hi.
{-1.
w .. “Luz J ..

.2
n:

L I;
1.9. an;
112 2

g... .
a? _

 

 

 

- -
:sifiaz ‘huu.

.. .. v... .. l
00 . . ‘ 1.311....

iwfimwmwnh
. f-

z
.34. In 2.9! A
(39...? -2.

:13}; J. .
nuwfltnuuuu ”Ilka

.. ’9 v. I.)

 

THEL

\

J 003

This is to certify that the

dissertation entitled

CHARACTERIZATION OF THE PUTATIVE TUMOR SUPPRESSOR
GENE 5T7, A NOVEL MEMBER OF THE LON-DENSITY
LIPOPROTEIN RECEPTOR SUPERFAMILY

presented by

Michele Ann Battle

has been accepted towards fulfillment
of the requirements for

Ph.D. degreeinCell and Molecular Biology

 

Major professor

»/ . 7 ,,
Wacfl/ A: fifiél/
Date 401cc ”tr/4;» //roZUfloZ,

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

 

LIBRARY "
Michigan State
University

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

[LEE—11299!

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDue.p65-p.15

CHARACTERIZATION OF THE PUTATIVE TUMOR SUPPRESSOR
GENE 877, A NOVEL MEMBER OF THE LOW-DENSITY
LIPOPROTEIN RECEPTOR SUPERFAMILY
BY

Michele Ann Battle

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Cell and Molecular Biology Program

2002

ABSTRACT
CHARACTERIZATION OF THE PUTATIVE TUMOR SUPPRESSOR
GENE ST7, A NOVEL MEMBER OF THE LOW-DENSITY
LIPOPROTEIN RECEPTOR SUPERFAMILY
By
Michele Ann Battle
It is widely accepted that cancer develops through a multi-step process
involving a series of genetic and/or epigenetic changes that convert a normal cell
into a cancer cell. Tumors originate from a single cell that has acquired all
characteristics necessary to be able to produce a malignant tumor, and the cells
of the tumor are the progeny of this first malignant cell. In an effort to identify the
number and nature of these changes, McCormick and Maher (Risk Anal. 14: 257,
1994) established a model system in which normal human fibroblasts in culture
can be transformed into malignant fibroblasts by acquiring a series of genetic
changes, each conferring a growth advantage that allows sequential clonal
expansion. Using this model system and differential mRNA display to compare
an infinite life span, non-tumorigenic human fibroblast cell strain to one of its
carcinogen-transformed, malignant derivatives led to the identification of a novel
gene, designated ST7, that was not expressed in the malignant derivative and
was absent or expressed at a low level in a series of malignant cell lines derived
from patients’ tumors. Subsequent discovery of a closely related protein
suggested that ST7 is a novel member of the low-density lipoprotein receptor
(LDLR) superfamily. Although proteins of this superfamily are best known as

endocytic receptors, recent studies demonstrate that the cytoplasmic domains of

several members interact with adaptor and scaffold proteins implicated in signal
transduction. To evaluate ST7's relationship to the proteins of this superfamily, I
used proteomic tools to analyze the functional motifs present in ST7. The data
indicate that ST7 is, indeed, a member of an LDLR subfamily and reveal that
ST7's cytoplasmic domain contains motifs implicated in endocytosis and signal
transduction. Using the yeast two-hybrid system, I determined that ST7's
cytoplasmic domain, like that of some other LDLR superfamily proteins, interacts
with several proteins related to signal transduction and/or endocytosis. To
determine if ST7 can suppress tumor formation, a malignant cell line, SHAC,
derived from a patient's fibrosarcoma was transfected with a plasmid encoding
ST7 or a truncated form of ST7 protein, i.e., its extracellular and transmembrane
domains, but only the first ten amino acids of its 343-residue cytoplasmic domain.
The tumorigenicity of cell strains expressing these proteins was determined by
injection into athymic mice. SHAC cells overexpressing full-length ST7 formed
tumors at a significantly lower frequency than that of control cells. Moreover, the
tumors that did develop exhibited a much greater latency. In contrast,
overexpression of truncated ST7 failed to inhibit the tumor-fonning ability of
SHAC cells. These data support the hypothesis that $17 is a tumor suppressor.
Furthermore, the fact that expression of a truncated form of ST7 failed to inhibit
the tumor-fanning ability of SHAC cells strongly suggests that ST7’s cytoplasmic

domain plays an important role in the protein’s function as a tumor suppressor.

DEDICATION

With love, I dedicate my dissertation to Mom,

Dad, Amy, Kelly, Lori, Scott, Andre, and Marshall.

ACKNOWLEDGMENTS

I thank Dr. Veronica Maher and Dr. Justin McCormick for providing me
with the privilege of carrying out my dissertation research in the Carcinogenesis
Laboratory. I must credit Veronica as the reason I chose to attend Michigan
State University. During our first meeting, she immediately impressed me with
her enthusiasm and energy. As her student, I’ve learned that her enthusiasm
and energy are neverending. Her love of science and her dedication to her work
inspire me. I also offer sincere thanks to Justin for his guidance in my research
and in my career. I will remember his many words of encouragement and the
cookies he provides to weary graduate students! I am also very grateful for
guidance provided by Drs. Jerry Dodgson, Laura McCabe, John Wang, Katheryn
Meek and Leslie Kuhn. From the time I began my research in the
Carcinogenesis Laboratory in May 1997 to the present, the members of the
Carcinogenesis Laboratory provided me not only with helpful scientific advice
and instruction but also with their friendship. I am especially thankful for the
friendship of Kim-Hien T. Dao, Susanne Kleff, Dan Appledorn, Kristin McNally,
Piro Lito, Sandra O’Reilly, Hongyan Liang, Ziqiang Li and Shikha Gupta. Several
undergraduate students helped me with my dissertation research and I thank
each of them, with special thanks to Emily Reamer. I thank my parents and my
sisters Amy, Kelly, and Lori for their constant encouragement. Finally, I thank
my husband Scott. When my experiments were successful, he celebrated with

me. When my experiments failed, he convinced me that I would solve the

problem. In the final months of preparing my dissertation, although he was
tremendously busy as a new assistant professor in Milwaukee, he found time to

return to East Lansing to help me complete this task.

vi

TABLE OF CONTENTS

LIST OF TABLES .................................................................................................. x
LIST OF FIGURES .............................................................................................. xi
ABBREVIATIONS ............................................................................................... xii
INTRODUCTION .................................................................................................. 1
REFERENCES ..................................................................................................... 5
CHAPTER I: LITERATURE REVIEW ................................................................. 6
I. A review of the molecular mechanisms of tumorigenesis ................................. 6
A. Six types of phenotypic changes typically involved in cancer ...................... 6
1. The ability to proliferate without mitogenic signals .................................... 6
a. The proto-oncogene MYC ................................................................ 7
b. The proto-oncogene RAS ................................................................. 9
2. The ability to evade growth inhibitory signaling ....................................... 12
a. The tumor suppressor gene RB ..................................................... 12
b. The tumor suppressor gene p53 .................................................... 15
3. The ability to evade apoptosis ................................................................ 16
4. The ability to proliferate limitlessly (the acquisition of an
infinite lifespan) ...................................................................................... 19
5. The ability to initiate and to sustain angiogenesis ................................... 22
6. The ability to invade surrounding tissues and to metastasize ................. 28
a. Integrins, cellular adhesion and signaling proteins that play
a critical role in invasion and metastasis ........................................ 30
b. Matrix metalloproteinases, a family of proteases that plays
a critical role in invasion and metastasis ........................................ 33
B. The Multi-Step Model of Tumorigenesis .................................................... 34
1. Epidemiological evidence ....................................................................... 34
2. Colorectal tumorigenesis ........................................................................ 35
3. MSU-1 lineage as a model for tumorigenesis ......................................... 41
a. Generation and characterization of the MSU-1 lineage .................. 41
b. Using the MSU-1 lineage to study the multi-step process of
tumorigenesis in human fibroblasts ............................................... 46
II. The LDLR superfamily proteins and their relationship to ST7 ......................... 49
A. The LDLR Superfamily ............................................................................... 49
1. LDLR ...................................................................................................... 50
2. VLDLR and ApoER2 ............................................................................... 57
3. LRP1 and LRP1B ................................................................................... 61
4. LRP5 and LRP6 ...................................................................................... 72
5. LRP3 and LRP9, two ST7-related proteins ............................................. 78
B. The relationship between ST7 and the LDLR superfamily ......................... 82
REFERENCES ................................................................................................... 84

vii

Chapter II: ST7 is a novel member of the low-density lipoprotein receptor (LDLR)
superfamily with a cytoplasmic tail that interacts with proteins related to signal

transduction pathways ...................................................................................... 1 15
ABSTRACT ....................................................................................................... 1 17
INTRODUCTION .............................................................................................. 118
EXPERIMENTAL PROCEDURES .................................................................... 121
Materials ........................................................................................................ 121
Cell Culture .................................................................................................... 121
Isolation of ST7 cDNA ................................................................................... 121
DNA sequencing ........................................................................................... 122
Yeast two-hybrid assay ................................................................................. 122
Oligonucleotides ............................................................................................ 124
Construction of mammalian expression plasmids ......................................... 125
Co—immunoprecipitation of proteins from transiently transfected 293-T cells 127
RESULTS ......................................................................................................... 129
Isolation of ST7 cDNA from a human heart muscle cDNA library .................. 129
ST7 is a novel member of the LDLR family of proteins ................................. 129
Analysis of the ST7 protein ........................................................................... 131
Identification of proteins that interact with the cytoplasmic domain of the ST7
protein ........................................................................................................ 140
Confirmation of the protein-protein interactions within human cells ............... 143
Identification of the RACK1, MIBP, and SARA binding domains within ST7’s
cytoplasmic domain .................................................................................... 147
DISCUSSION .................................................................................................... 158
REFERENCES ................................................................................................. 165
Appendix: Expression of ST7 Protein Inhibits the Tumor-Fonning Ability of a
Human Fibrosarcoma-derived Cell Line ............................................................ 171
ABSTRACT ....................................................................................................... 1 73
INTRODUCTION .............................................................................................. 175
MATERIALS AND METHODS .......................................................................... 178
Materials ........................................................................................................ 178
Cells and cell culture conditions .................................................................... 178
Stable transfection SHAC cells ...................................................................... 179
Preparation of cell Iysates ............................................................................. 179
Preparation of plasma-membrane enriched fractions .................................... 179
Western blotting analysis .............................................................................. 180
Tumorigenicity studies ................................................................................... 180
RESULTS ......................................................................................................... 182
Construction of expression plasmids encoding V5/His-tagged full-length or a
V5/His-tagged truncated form of ST7 ......................................................... 182
Expression of full-length ST7 in a human fibrosarcoma-derived cell line,
but not the truncated form of ST7, inhibits tumor forming activity ............... 182

viii

The ST7-V5lHis and ST7tr1-V5/His proteins are overexpressed in the SHAC

cell strains generated ................................................................................. 187
DISCUSSION .................................................................................................... 194
REFERENCES ................................................................................................. 197

LIST OF TABLES
Chapter II
Table I. Structure of the ST7 gene ................................................................. 130

Table II. Putative PKC and PKA phosphorylation sites in ST7’s
cytoplasmic domain .......................................................................... 139

Table III. Identification of proteins that interact with the cytoplasmic
domain of ST7 ................................................................................... 142

Table IV. Interaction of prey proteins with truncated ST7 bait proteins ............ 153

Appendix

Table 1. Expression of ST7 inhibits the tumor-forming ability of SHAC cells 186

LIST OF FIGURES

Chapter I

Figure 1. The multi-step and clonal model of carcinogenesis: colon cancer. ..... 39

Figure 2. The MSU-1 lineage. ............................................................................ 43
Figure 3. Structures of representative proteins of the LDLR superfamily. .......... 53
Chapter II

Figure 1. Structures of representative proteins of the LDLR superfamily. ........ 133
Figure 2. The cytoplasmic domain of ST7 (residues 498-840) ......................... 135

Figure 3. ST7 interacts with RACK1 p, MIBPp, and SARAp in human cells ...... 145
Figure 4. ST7 interacts with full-length RACK1 and MIBP in human cells ....... 149

Figure 5. Illustration of the truncated ST7 cytoplasmic domain bait proteins used
to determine the location of the MIBP, SARA, and RACK1 binding sites in

ST7’s cytoplasmic domain ......................................................................... 151
Figure 6. The first ten amino acids of the ST7 cytoplasmic domain may suffice

for MIBP binding ........................................................................................ 157
Appendix

Figure 1. Presence or absence of V5/His-tagged proteins in SHAC
transfectants. ............................................................................................. 1 85

Figure 2. Expression of endogenous ST7 protein in normal human fibroblasts,
SHAC cells, and SHAC-derived transfectants ........................................... 190

Figure 3. The V5/His tagged ST7 proteins encoded by the transfected
plasmids are localized in the membrane of the SHAC cell recipients ........ 193

xi

APAF1 ,
APC,
apoB,
apoE,
BCL-2,
bHLHZ,
BPDE,
cdk,
cDNA,
CUB,

DCC,
DISC.
DPC4,
EGF,
EGFR,
ERK,
FAK,
FAP,
F H,
FISH,
FZ,

ABBREVIATIONS

epoptotic erotease activating factor-1
edenomatous eolyposis eoli
epelipoprotein B

emlipoprotein g

B-eell lymphoma 2

_b_asic helix-foop-nelix _z_ipper
eenzo(a)eyrene-g_iol-epoxide
eyclin-gependent kinases
eomplementary _D_I;I_A_

Complement factor C1s/C1 r, Urchin embryonic growth factor, Bone
morphogenetic protein

_deleted in eolorectal earcinoma
geath-jnducing eignaling eomplex
geleted in pancreatic earcinoma 4
epidermal growth factor

epidermal growth factor [eceptor
extracellular signal-[egulated kinase
focal edhesion kinase

familial edenomatous polyposis
familial hypercholesterolemia
fluorescent in eitu hybridization

frigzled

xii

GAP,
GEF,
GPI,
G-protein,
GSK-3B,
GTPase,
HAT,
HDL,
HNPCC.
IDL,
LDL,
LDLR,
LDLRA,
LRP,
MAP,
MAPK,
Max,
MDM2,
MEGF7,
MEK,
MIBP,
MLCK,
MMP,

QTPase ectivating protein

GDP exchange factor
glycosylphosphatidyI-fnositol
guanine-nucleotide binding m
glycogen eynthase kinase-gp
guanosine friphosphate hydroleee
pistone ecetylfransferese

pigh-Qensity lipoprotein

pereditary ponpolyposis polorectal eancer
intermediate-gensity lipoprotein
low-gensity lipoprotein

low-gensity lipoprotein feceptor
low-density lipoprotein feceptor class A
LDLR-[elated protein
mitogen-ectivated protein
mitogen-ectivated protein kinase
Myc-essociated factor x

murine gouble minute clone 2

multiple epidermal growth factor-like domains protein 1
_MAPIERK kinase

muscle fntegrin pinding protein

myosin light ehain kinase

matrix metalloproteinase

xiii

MMR,
NSCLC.
OPPG,
PI3K,
PAI-1,
PDGF,
PDGFR,
PEX,
PKA,
PKC,
PTB,
RACK1,
RB,
SARA,
SD,
SH2,
8H3,
SOS,
TGF,
TNF,
tPA,
TRRAP,
TSP-1,

DNA mismatch fepair

hon-emall eell lung earcinoma
psteoporosis-pseudoglioma
phosphatidylfnositol g-hinase
plasminogen ectivator inhibitor type-1
platelet-gerived growth factor
platelet-gerived growth factor [eceptor
hemopexin

protein hinase A

protein hinase Q

phosphofyrosine hinding

[eceptor for ectivated protein C hinase
[etinohlastoma

_S_MAD enchor for [eceptor ectivation
eynthetic gropout

Src homology 2

§rc homology 3

eon pf eevenless

fransforming growth factor
fumor-hecrosis factor

fissue plasminogen ectivator
hansforming/tfanscription domain essociated protein

fhromboepondin-1

xiv

uPA, prokinase plasminogen ectivator

uPAR, prokinase-type plasminogen ectivator [eceptor
VEGF, yascular endothelial growth factor
VLDL, yery [ow-gensity lipoprotein

VLDLR, yery [ow—gensity lipoprotein feceptor

INTRODUCTION

It is widely accepted that human tumors originate from a single cell that
has acquired all of the properties necessary to become malignant, and replication
of that cell gives rise to all of the cells of the tumor. Theories of carcinogenesis
must explain how the “first” tumor cell acquired the appropriate genetic and/or
epigenetic changes required to render it malignant, which for tumors in adults are
estimated to be five or more. Although tumors arise in virtually all tissues of the
human body and each type of tumor displays some unique characteristics, the
basic steps responsible for the conversion of a normal cell into a malignant cell
are believed to be common. In a recent review, Hanahan and Weinberg (2000)
propose six essential phenotypic characteristics that must be acquired if a normal
cell is to become a malignant cell: i) the ability to proliferate without mitogenic
signals; ii) the ability to evade growth-inhibitory signals; iii) the ability to evade
apoptosis; iv) the ability to proliferate limitlessly; v) the ability to initiate and to
sustain angiogenesis; and vi) the ability to invade tissue and to metastasize.
Each successive change thwarts an “anticancer defense mechanism” inherent in
every cell. The first section of Chapter I reviews the current model of the
molecular mechanisms of the multi-step process of tumorigenesis, focusing on
the six characteristics that are required for a cell to become tumorigenic. The
second section reviews the structure and function of several key members of the
low-density lipoprotein receptor (LDLR) superfamily and the relationship of the
protein encoded by the novel gene ST7, discovered in the Carcinogenesis

Laboratory in 1997, to this superfamily.

Chapter II consists of a manuscript describing my work characterizing the
ST7 protein and identifying proteins that interact with its cytoplasmic domain. In
1997, when the novel gene, designated ST7, was identified by Jing Qing using
differential mRNA display, she recognized that the ST7 protein had the
characteristics of a transmembrane protein, but at that time no other proteins with
significant similarity to ST7 had been reported. Approximately one year later,
Yamamoto and his associates, using degenerate Oligonucleotides corresponding
to the highly conserved region of the ligand-binding domains found in the
proteins of the LDLR superfamily, discovered a novel LDLR-related gene, which
they designated LRP3 (Ishii et al., 1998). Our search of databases for proteins
structurally related to ST7, after the sequence of LRP3 had been submitted,
revealed a very strong similarity between LRP3 and ST7, enabling us to
recognize that ST7, although it differed significantly from the LDLR prototype,
could be a novel member of the LDLR superfamily. Although proteins of the
LDLR superfamily are best known as endocytic receptors (reviewed by Krieger
and Herz, 1994 and Hussain et al., 1999), recent studies show that the
cytoplasmic domains of several LDLR superfamily proteins interact with a variety
of adaptor and scaffold proteins implicated in signal transduction (reviewed by Li
et al., 2001; Herz and Bock, 2002; van der Geer, 2002). To evaluate ST7's
relationship to the proteins of the LDLR superfamily, I used proteomic tools
including similarity searches, sequence alignments, pattern and profile searches,
and post-translational modification prediction programs to analyze the functional

motifs present in ST7. Such analyses showed that ST7 is, indeed, a novel

member of the LDLR superfamily and that like several other LDLR superfamily
proteins, ST7’s cytoplasmic domain contains several motifs implicated in
endocytosis and/or signaling. To identify proteins that interact with ST7’s
cytoplasmic domain, I used the yeast two-hybrid system. These studies showed
that ST7’s cytoplasmic domain interacts with several proteins related to signal
transduction and/or endocytosis, suggesting that ST7 functions in these
pathways.

The appendix describes research I carried out to test the hypothesis that
ST7 functions as a tumor suppressor. A tumorigenic cell line derived from a
patient’s fibrosarcoma, designated SHAC, was transfected with a plasmid
encoding full-length ST7 or an ST7 protein with a severely truncated cytoplasmic
domain, i.e., containing ST7’s extracellular and transmembrane domains, but
only the first 10 of the 343 amino acids that comprise its cytoplasmic domain.
The tumorigenicity of independent transfectants expressing these proteins, as
well as that of independent vector-control transfectants and non-ST7-expressing
control transfectants, was assessed by subcutaneous injection of these various
cell strains into athymic mice. Tumors formed at a significantly lower frequency
in athymic mice injected with SHAC cells expressing full-length ST7 than in
athymic mice injected with any of the control cell strains. Any tumors that did
develop exhibited a much greater latency. These data support the hypothesis
that ST7 is a tumor suppressor. The fact that expression of a truncated form of
817 failed to inhibit the tumor-forming ability of SHAC cells strongly suggests that

the cytoplasmic domain of ST7 plays an important role in the protein’s function as

a tumor suppressor. These studies are still in progress. When the final
experiments have been completed, a manuscript describing the results will be

submitted to Cancer Research.

REFERENCES

Hanahan, D., and Weinberg, R. A. (2000). The hallmarks of cancer. Cell 100, 57-
70.

Herz, J., and Bock, H. H. (2002). Lipoprotein receptors in the nervous system.
Annu Rev Biochem 71, 405-434.

Hussain, M. M., Strickland, D. K., and Bakillah, A. (1999). The mammalian low-
density lipoprotein receptor family. Annu Rev Nutr 19, 141-172.

Ishii, H., Kim, D. H., Fujita, T., Endo, Y., Saeki, S., and Yamamoto, T. T. (1998).
cDNA cloning of a new low-density lipoprotein receptor-related protein and
mapping of its gene (LRP3) to chromosome bands 19q12-q13. 2. Genomics 51,
132-135.

Krieger, M., and Herz, J. (1994). Structures and functions of multiligand
lipoprotein receptors: macrophage scavenger receptors and LDL receptor-related
protein (LRP). Annu Rev Biochem 63, 601-637.

Li, Y., Cam, J., and Bu, G. (2001). Low-density lipoprotein receptor family:
endocytosis and signal transduction. Mol Neurobiol 23, 53-67.

Qing, J., Maher, V. M., Tran, H., Argraves, W. S., Dunstan, R. W., and
McCormick, J. J. (1997). Suppression of anchorage-independent growth and
matrigel invasion and delayed tumor formation by elevated expression of fibulin-
1D in human fibrosarcoma-derived cell lines. Oncogene 15, 2159-2168.

Qing, J., Wei, 0., Maher, V. M., and McCormick, J. J. (1999). Cloning and
characterization of a novel gene encoding a putative transmembrane protein with

altered expression in some human transformed and tumor-derived cell lines.
Oncogene 18, 335-342.

van der Geer, P. (2002). Phosphorylation of LRP1: regulation of transport and
signal transduction. Trends Cardiovasc Med 12, 160-165.

CHAPTER I

LITERATURE REVIEW

I. A review of the molecular mechanisms of tumorigenesis

A. Six types of phenotypic changes typically involved in cancer

Although tumors arise in virtually all tissues of the human body and each
type of tumor displays some unique characteristics, the basic steps responsible
for the conversion of a normal cell into a malignant cell are believed to be
common. In a recent review, Hanahan and Weinberg (2000) discuss the
molecular mechanisms that give rise to cancer. They propose six essential
phenotypic characteristics that must be acquired if a normal cell is to become a
malignant cell: i) the ability to proliferate without mitogenic signals; ii) the ability
to evade growth-inhibitory signals; iii) the ability to evade apoptosis; iv) the ability
to proliferate limitlessly; v) the ability to initiate and to sustain angiogenesis; and
vi) the ability to invade tissue and to metastasize. Each successive change
thwarts an “anticancer defense mechanism” inherent in every cell Hanahan and

Weinberg, 2000). A discussion of these six hallmarks of cancer follows.

1. The ability to proliferate without mitogenic signals

Normal human cells proliferate in response to mitogenic signals. Such
signals include soluble growth factors, extracellular matrix components, and cell-
cell interactions (Hanahan and Weinberg, 2000). Unregulated cellular
proliferation, often accomplished by mutating oncogenes, is one hallmark of

tumor cells. Several types of oncogenic alterations can result in deregulation of

growth pathways, allowing these cells to proliferate in the absence of mitogenic
signals. Tumor cells can synthesize growth factors that act in an autocrine
fashion to stimulate their growth. Mutations in growth factor receptors can cause
constitutive activation of these receptors. Mitogenic pathways can also become
constitutively active because of mutations in downstream molecules responsible
for transmitting mitogenic signals from the growth factor receptor to the cell’s

nucleus.

a. The proto-oncogene MYC

The study of transforming retroviruses has provided insight into human
cancer. Many cellular proto-oncogenes were identified because they are
homologues of viral oncogenes. The c-MYC proto-oncogene was identified as
the cellular homologue of the viral oncogene v-MYC present in the genome of the
avian myelocytomatosis retrovirus (Vennstrom et al., 1982). The c-MYC proto-
oncogene plays a causal role in many types of human tumors (reviewed by
Facchini and Penn, 1998). For example, translocations between chromosome 8
and chromosomes 2, 14, or 22, which have been observed in Burkitt’s
lymphoma, position the c-MYC proto-oncogene in close proximity to
immunoglobulin enhancer elements, resulting in up-regulation of c-MYC gene
expression (Solomon et al., 1991). Additionally, deregulation of c-MYC
expression has been observed in breast carcinomas, colorectal carcinomas,
cervical carcinomas, small cell lung carcinomas, osteosarcomas, glioblastomas,
and melanomas (reviewed by Pelengaris et al., 2002).

The c-MYC gene encodes a transcription factor in the basic helix-loop-

helix zipper (bHLHZ) family (Kato et al., 1990). Through its bHLHZ domain, Myc
heterodimerizes with Max, another bHLHZ protein (Blackwood and Eisenman,
1991; Kato et al., 1992; Amati et al., 1993a, 1993b). Myc-Max heterodimers bind
to specific DNA promoter sequences to activate the transcription of genes that
stimulate cellular proliferation and growth and to repress the transcription of
genes that inhibit terminal differentiation. The interaction of Myc—Max
heterodimers with other proteins in the cell determines whether transcriptional
activation or repression occurs. For example, Myc-Max heterodimers activate
transcription of the CYCLIN DZ gene by recruiting the transforming/transcription
domain associated protein (TRRAP) to the CYCLIN DZ promoter (Bouchard et
al., 2001). TRRAP is a component of a chromatin remodeling complex
containing histone acetyltransferase (HAT) activity (McMahon et al. 2000).
Histone acetylation changes the chromatin conformation, enabling transcription
factors to gain access to promoter sequences (reviewed by Peterson, 2002).
This facilitates the activation of CYCLIN DZ expression by Myc-Max
heterodimers. Conversely, to inhibit the expression of p15’NK‘B and p21WAF 1,
Myc-Max heterodimers sequester the transcription factor Miz1 (Staller et al.,
2001; Herold et al. 2002). The interaction of Myc-Max with Miz1 prevents Miz1
from binding to p300, a co-activator required for Miz1 transactivation of the
p15’NK‘B and pZ1WAF’ promoters.

The Mad family of bHLHZ transcription repressor proteins antagonizes
Myc function (Ayer et al., 1993; reviewed by Grandori et al., 2000). By binding to

Max, Mad proteins prevent the formation of Myc-Max heterodimers. Mad-Max

heterodimers bind to the same DNA promoters as Myc-Max heterodimers (Lutz
et al., 2002). Mad-Max heterodimers, however, recruit histone deacetylases to
promoters, and subsequent histone deacetylation causes chromatin remodeling
that represses transcription (Ayer et al., 1995; Schreiber-Agus et al., 1995).
Expression of the c-MYC gene is important for cell cycle progression,
specifically for G1 to S progression (Steiner et al., 1995; Berns et al., 1997).
Myc-Max heterodimers induce expression of several key cell cycle regulatory
proteins including cyclin DZ, cyclin dependent kinase 4 (cdk4), and CUL1
(reviewed by Pelengaris et al., 2002). Cyclin DZ-cdk4 complexes bind to and
sequester the cell cycle inhibitor p27KIP 1. The p27“P1 protein normally binds to
and inhibits cyclin E—cde complexes. Sequestration of p27“?1 frees cyclin E-
cdk2 complexes to phosphorylate the cell cycle inhibitor pr. Such
phosphorylation of pr causes it to release bound EZF family transcription
factors, which then act to stimulate cell cycle progression. (The role of pr in the
cell cycle is discussed in detail below.) The CUL1 protein inhibits p27K'P1 by

targeting it for proteosomal degradation (O’Hagan et al., 2000).

b. The proto-oncogene RAS

Two of the three RAS proto-oncogenes, H-RAS and K-RAS, were
identified as the cellular homologues of the viral oncogenes carried by the
Harvey and Kirsten murine sarcoma viruses, respectively (Shih et al., 1979;
Chang et al., 1982; Parada et al., 1982; Pulciani et al., 1982; Santos et al., 1982;
Der et al., 1982). The N-RAS proto-oncogene was isolated from a

neuroblastoma cell line (Shimizu et al., 1983). Mammalian cells ubiquitously

express at least one of these three Ras isoforms (reviewed by Macaluso et al.,
2002). The likely importance of Ras proteins in tumorigenesis is demonstrated
by the presence of RAS genes with activating mutations in at least 30% of
human tumors (reviewed by Bos et al., 1989). Point mutations in codon 12, 13,
59, or 61 are most commonly observed in human tumors (Reddy et al., 1982;
Taparowsky et al., 1982; Tong et al., 1989; Krengel et al., 1990; Macaluso et al.,
2002). Such mutations disrupt the GTPase activity of the protein, producing a
constitutively active Ras protein, i.e., GTP-bound Ras.

Each RAS proto-oncogene encodes a 21 kD protein that functions as a
monomeric guanine-nucleotide binding protein (G-protein) (reviewed by
Barbacid, 1987). As a monomeric G-protein, Ras binds to and hydrolyzes
guanosine triphosphate (GTP) via an intrinsic guanosine triphosphate hydrolase
(GTPase) activity (Shih et al., 1980; Gibbs et al., 1984; Temeles et al. 1985; de
Vos et al., 1988). The hydrolysis of GTP yields an inactive GDP-bound protein.
Two classes of proteins facilitate the cycling of Ras between its active, GTP-
bound state and its inactive, GDP-bound state. GTPase activating proteins, or
GAPs, stimulate GTPase activity, leading to the conversion of active, GTP-bound
Ras to inactive, GDP-bound Ras; GDP exchange factors, or GEFs, assist in the
dissociation of GDP from the inactive protein so that GTP can bind (reviewed by
Takai et al., 2001). The ability of Ras to cycle between active and inactive states
enables it to act as a molecular switch for the pathways it regulates. Post-
translational modifications of the Ras protein target it to the plasma membrane

(Hancock et al., 1990; reviewed by Newman and Magee, 1993 and Macaluso et

10

al., 2002). These modifications include the famesylation and subsequent
methylation of cysteine 186 and the palmitoylation of cysteine residues between
amino acids 165-186.

Binding of growth factors to their cognate cellular receptors, which contain
intracellular tyrosine kinase domains, leads to receptor dimerization and
autophosphorylation at specific tyrosine residues (reviewed by Heldin et al.,
1995). The receptor’s phosphotyrosine residues serve as docks for cytoplasmic
scaffold proteins. As their name suggests, the scaffold proteins provide
additional binding sites for proteins required to transmit mitogenic signals
(Pawson and Scott, 1997). One such scaffold protein is Grb2, which interacts
with phosphotyrosines in the cytoplasmic tails of growth factor receptors via its
Src homology 2 (SH2) domain (Lowenstein et al., 1992; Matuoka et al., 1993).
The GrbZ protein also binds to proline residues in the Ras—specific GDP
exchange factor (GEF) SOS through its Src homology 3 (SH3) domain (Chardin
et al., 1993). This interaction places SOS in close proximity to its substrate,
GDP-bound Ras. The interaction of SOS with GDP-bound Ras stimulates the
dissociation of GDP from Ras, thereby enabling the Ras protein to bind to GTP
and become active. GTPase activating proteins (GAPs) then associate with
GTP-bound Ras to stimulate GTP hydrolysis. While in its active, GTP-bound
state, Ras can bind to and stimulate its effectors. There are three main Ras
effector pathways: 1) the RAF1/MEK/ERK pathway; 2) the Pl3K/Akt pathway;
and 3) the Ral-GDS/Ral GTPase/RIP1 pathway (reviewed by Takai et al., 2001;

Macaluso et al., 2002). Activation of these pathways by Ras leads to the

11

expression of cellular genes that stimulate cell cycle progression, that promote
cellular survival over apoptosis, and that regulate cytoskeletal organization. For
example, activation of the RAF1/MEK/ERK pathway by Ras leads to transcription
of the CYCLIN D1 gene (reviewed by Robinson and Cobb, 1997). D—type cyclins
interact with cyclin-dependent kinases (cdks) and inactivate the cell cycle
inhibitor pr, thereby stimulating cell cycle progression. (The role of pr in the

cell cycle is discussed in detail below.)

2. The ability to evade growth inhibitory signaling

Normal human cells possess numerous genes encoding proteins that
function to protect the cells from inappropriate cellular proliferation. For example,
when contact is made with other cells, signals to stop proliferation are
transmitted. Damage to the genome initiates signaling cascades designed to halt
the cell cycle, allowing time for DNA repair to occur before S-phase of the cell
cycle. If cellular DNA repair mechanisms fail to adequately repair the damage,
activation of programmed cell death (apoptosis) can occur. In order to gain the
ability to proliferate uncontrollably and form tumors, cells must not only activate

oncogenes but must also subvert these protective growth inhibitory pathways.

a. The tumor suppressor gene RB

Retinoblastoma is a cancer of the retina of the eye. There are two forms
of retinoblastoma, familial and sporadic. The familial form of retinoblastoma is
characterized by retinal tumors that occur in both eyes. Each eye may have one

or more tumors. These tumors almost always develop before the age of five.

12

The sporadic form usually results in a single retinal tumor that develops later in
life. Knudson (1971) proposed a “two-hit” model of tumorigenesis to explain the
occurrence of both familial and sporadic retinoblastoma. According to this
model, retinoblastoma arises because of loss of a protein required for negative
regulation of retinal cell growth. To inactivate the function of this protein,
mutations must occur in both alleles of the gene. Loss of the ability to produce
functional protein results in the unregulated proliferation of the retinal cells with
this defect. Knudson referred to these protective genes as anti-oncogenes;
scientists now refer to them as tumor suppressor genes. The familial form of
retinoblastoma involves the inheritance of a gennline mutation in one allele and
the subsequent acquisition of a somatic mutation in the other allele, whereas the
sporadic form involves the acquisition of somatic mutations In both alleles. The
probability that a person born with two wild-type alleles will acquire somatic
mutations in both alleles to cause the development of retinoblastoma is very low
because both mutations must occur in a single cell for a tumor to arise.
However, the probability that a person born with a gerrnline mutation in one allele
will acquire a somatic mutation in the other allele in a single cell is significantly
higher because every cell in the retina of this individual contains the first
mutation. This explains why familial retinoblastoma occurs early in life and why
the affected individuals typically have bilateral tumors. Moreover, individuals who
inherit a gerrnline mutation are predisposed to bone and soft tissue sarcomas
later in life (reviewed by Nevins, 2001).

Studies of retinoblastoma led to the identification the first tumor

13

suppressor gene, designated the retinoblastoma (RB) gene (Friend et al., 1986;
Fung et al., 1987; Lee et al., 1987). Although first associated with
retinoblastoma, mutations in the RB gene have been observed in many human
tumors including osteosarcomas, small cell lung carcinomas, and breast
carcinomas (reviewed by Nevins, 2001). Moreover, the Rb pathway is altered in
greater than 80% of human cancers (reviewed by Ortega et al., 2002). As
predicted by Knudson, the RB gene encodes a protein that plays a critical role in
negatively regulating cellular proliferation. Specifically, the Rb protein serves as
a cell cycle gatekeeper by regulating the cell’s transition from G1 phase of the
cell cycle into S phase. Phosphorylation of Rb regulates its function.
Unphosphorylated Rb binds to members of the EZF family of transcription
factors, preventing these factors from activating transcription of EZF-responsive
genes. Such genes encode proteins required for cell cycle progression
(reviewed by Trimarchi and Lees, 2002). Therefore, EZF proteins must be active
in order for the cell cycle to progress. Mitogenic signals lead to the activation of
D-family cyclins and their associated kinases, cyclin-dependent kinases 4 and 6
(cdk4 and cdk6) (reviewed by Sears and Nevins, 2002). Activated cyclin D-
cdk4/6 complexes phosphorylate Rb. Phosphorylation of Rb disrupts its
interaction with the E2F proteins, freeing the EZF proteins to activate
transcription. Rb inactivation is enhanced by expression of two EZF responsive
genes, the CYCLIN E and CDKZ genes. Cyclin E and cde form a complex that
further phosphorylates Rb. Tumor cells that have lost functional Rb protein

exhibit unrepressed EZF activity, and this provides one mechanism to evade

14

growth inhibitory signaling.

b. The tumor suppressor gene p53

Another key protein involved in protecting cells from malignancy is the
tumor suppressor p53. This protein was first identified in 1979 as a cellular
protein with a high binding affinity for the SV40 large T antigen (Lane and
Crawford, 1979; Linzer and Levine, 1979). Simultaneously, DeLeo et al. (1979)
showed that the p53 protein is highly expressed in chemically-transformed
sarcoma-derived cells. Although p53 was first classified as an oncogene,
subsequent work by several groups demonstrated that the original p53 cDNA
isolated encoded mutant p53 protein and that wild-type p53 strongly suppresses
growth and inhibits transformation (Hinds et al., 1989; Finlay et al., 1989; Eliyahu
et al., 1989; Baker et al., 1989).

Consistent with its potent ability to inhibit cell growth and to induce
apoptosis, normal cells maintain a very low level of p53 protein (reviewed by
Levine, 1997). The ubiquitin ligase MDMZ binds to p53 protein causing the
ubiquitination of p53 and subsequent degradation of ubiquitinated p53 by the
proteosome (Haupt et al., 1997; Kubbutat et al., 1997; Midgley and Lane, 1997).
Forces that interrupt the MDMZ-p53 interaction can lead to stabilization of the
p53 protein. For example, DNA damage induces several protein kinases,
including Chk1, Cth, ATM, and ATR, that phosphorylate p53 to destabilize its
interaction with MDM2, thereby preventing p53 degradation (reviewed by
Vousden, 2002). Interaction of the ARF protein with MDMZ also results in the

stabilization of p53 (Stott et al., 1998). Once stabilized, p53 translocates to the

15

nucleus where it can activate the transcription of genes that cause cell cycle
arrest (e.g., pZ1WAF1’0’V l), apoptosis (e.g., BAX), and/or DNA repair (e.g.,
GADD45). In a developing tumor, p53 also regulates the transcription of
inhibitors of angiogenesis (e.g., TSP1). Loss of functional p53 protein, therefore,
can lead to the propagation of abnormal cells.

Mutations that cause the inactivation of the p53 protein have been
observed in approximately 50% of human tumors (reviewed by Hollstein et al.,
1994). In a recent review, Vousden and Lu (2002) note that most of the efforts to
identify mutations in p53 examined only the sequence of exons five through
eight. Recent analyses of the entire p53 coding sequence show that mutations
occur in other regions, suggesting that the true frequency of p53 mutations in
human tumors may be significantly higher. Furthermore, even if human tumors
contain wild-type p53 protein, its function can be lost because of amplifications of
the MDMZ gene or the related MDMX gene (Oliner et al., 1992; Riemenschneider
et al., 1999). Amplification of these genes leads to a higher steady-state level
of the proteins (Ramos et al., 2001). Increased MDMZ protein causes increased
p53 degradation. The MDMX protein inhibits p53’s function as a transcription
factor, thereby preventing p53-mediated responses to cellular stresses (Shvarts

et al., 1996; Finch et al., 2002; Sabbatini and McCormick, 2002).

3. The ability to evade apoptosis

Apoptosis, a well—organized and highly regulated process through which
cells undergo programmed cell death, is characterized by the cell’s autodigestion

of its contents (reviewed by Igney and Krammer, 2002). Proteases digest

16

nuclear envelope proteins and cytoskeletal proteins, and nucleases digest the
genomic material. Both lead to the formation of apoptotic bodies that are rapidly
phagocytosed. Embryonic development relies upon apoptosis to shape the
developing organism (Alberts et al., 1994). Apoptosis also provides a key
cellular defense mechanism against tumorigenesis because it eliminates
abnormal cells from the body’s tissues. If abnormal cells become resistant to
apoptosis, these cells continue to proliferate, which potentially leads to the
acquisition of additional genetic changes that convert these cells into tumor cells.

There are two well-characterized apoptotic pathways, the extrinsic
pathway and the intrinsic pathway (reviewed by Ashkenazi, 2002; Igney and
Krammer, 2002). Death receptors, members of the tumor-necrosis factor (TNF)
receptor superfamily, mediate apoptosis through the extrinsic pathway. These
receptors contain a protein motif known as a death domain that interacts with
intracellular death-domain containing adaptor proteins (Hofmann and Tschopp,
1995; Feinstein et al., 1995). A death-inducing signaling complex, or DISC,
assembles at the receptor’s death domain (Kischkel et al., 1995). The critical
players in the DISC are the cysteine aspartyI-specific proteases known as
caspases (reviewed by Shi, 2002). Activation of the initiator caspases 8 and 10
in the DISC leads to activation of executioner caspases (caspases 3, 6, and 7)
that carry out apoptosis by cleaving death substrates.

III-defined death-inducing signals activate the intrinsic apoptotic pathway,
which is mediated by the pro-apoptotic members of the BCLZ family of proteins

(reviewed by Cory and Adams, 2002). These pro-apoptotic proteins can be

17

further divided into two subfamilies, the BAX family, which includes BAX, BAK
and BOK, and the BH3 family, which includes proteins such as BID, BIM, BIK,
and BAD. The pro-apoptotic BAX and BAK proteins function at the mitochondrial
membrane, where they induce the release of cytochrome c, a key step in the
intrinsic apoptotic pathway (Jurgensmeier et al, 1998; O’Connor et al., 1998;
Korsmeyer et al., 2000). Subsequent to cytochrome c release, a complex
containing cytochrome c, APAF1 (apoptotic protease activating factor-1), ATP,
and procaspase 9 assembles in the cytosol (Li et al., 1997). This complex is
known as the apoptosome. Cleavage of procaspase 9 to yield the active
caspase occurs in the apoptosome, leading to the activation of executioner
caspases, such as caspase 3, that then cleave death substrates (Li et al., 1997).
The BCLZ protein family also contains anti-apoptotic proteins, including
BCL-Z and BCL-XL, which antagonize the pro-apoptotic factors (reviewed by
Cory and Adams, 2002). Cells receive specific survival signals, which include
growth factors, cytokines, hormones, and adhesion molecules, and these signals
activate cellular signal transduction pathways that cause the inhibition of pro-
apoptotic factors. For example, activation of the PI3K/Akt signaling pathway by
specific types of survival signals inhibits apoptosis (reviewed by Vivanco and
Sawyers, 2002). Specifically, Datta et al. (1997) showed that Akt phosphorylates
BAD, a pro-apoptotic protein of the BCLZ family, and this event suppresses
apoptosis. As a pro-apoptotic protein, BAD dimerizes with the anti-apoptotic
factor BCL—XL to inhibit this factor (Yang et al., 1995; Ottilie et al., 1997; Kelekar

et al., 1997). When phosphorylated by Akt, BAD no longer binds to BCL—XL,

18

thereby freeing BCL-XL to carry out its role as an anti-apoptotic protein.
Finucane et al. (1999) demonstrated that BCL—XL promotes survival through its
ability to prevent BAX-induced mitochondrial release of cytochrome c.

Several mechanisms can enable tumor cells to evade apoptosis. Up-
regulation of the expression of anti-apoptotic proteins favors cell survival. For
example, BCL-2 is overexpressed in B-cell follicular lymphoma (Cleary et al.,
1986; Tsujimoto and Croce, 1986). Conversely, down-regulation of the
expression of pro-apoptotic proteins or the acquisition of inactivating mutations in
genes encoding pro-apoptotic proteins also favors cell survival. For example,
inactivating mutations in the pro-apoptotic BAX gene have been observed in
human tumors (Rampino et al., 1997; Meijerink et al., 1998; Molenaar et al.,
1998). The inactivation of tumor suppressor genes in combination with the
activation of cellular proto-oncogenes in tumor cells can lead to resistance to
apoptosis through the repression of pro-apoptotic signaling pathways and the
stimulation of anti-apoptotic signaling pathways (reviewed by Igney and
Krammer, 2002). For example, as previously noted, p53 is a tumor suppressor
gene that can initiate apoptosis. lnactivating mutations in p53, therefore, inhibit
the cell’s ability to activate apoptosis. Moreover, up—regulation of the PI3K
pathway through constitutively active Ras or constitutively active growth factor

receptors stimulates survival pathways, allowing tumor cells to evade apoptosis.

4. The ability to proliferate limitlessly (the acquisition of an

infinite lifespan)

Human cells in culture have a finite lifespan. In 1961, Hayflick and

19

Moorhead were the first to demonstrate that human embryonic fibroblasts in
culture divide only 40—60 times. Their studies of cultured human fibroblasts also
demonstrated that the cells have an innate ability to count their cell divisions. For
example, if human cells are frozen at cell division 20, the cells will begin cell
division 21 when returned to culture. The cells remember their cell division
history (reviewed in Hayfiick, 2000).

At the end of a cell’s replicative lifespan, growth is arrested and the cell
enters a state known as replicative senescence. Senescent human fibroblasts
are metabolically active but unable to divide (Hayflick and Moorhead, 1961).
They have a distinct large, flat morphology and are usually multinucleated.
Although the precise mechanism by which cells count cell divisions remains
elusive, clues about the signals that drive a cell into senescence emerged with
discovery of telomeres.

Naked chromosomal ends are susceptible to degradation, recombination,
or end-to-end chromosome fusions, and these events can cause the loss of
genetic material, genomic instability, or cell death (reviewed by Kim et al., 2002;
Hackett and Greider, 2002). Moreover, loss of chromosomal DNA occurs with
each cycle of DNA replication because the 3’ ends of linear chromosomal DNA
cannot be replicated completely, and this is referred to as the “end replication
problem” (Watson, 1972; Olovnikov, 1973). Protection of vulnerable
chromosomal DNA ends is accomplished by telomeres, which consist of simple
hexameric DNA repeats and a complex of telomeric DNA binding proteins

(Blackburn and Challoner, 1984; Shampay et al., 1984; Blackburn et al., 1989;

20

reviewed by Blackburn, 1991). In human cells, telomeres range from 10—15 kb in
length (reviewed by Hackett and Greider, 2002). Erosion of telomeres occurs
with each cycle of DNA replication. Loss of the repetitive telomeric DNA
prevents the loss of adjacent DNA that contains genes. Telomeres can be
maintained by the activity of telomerase, a reverse transcriptase that synthesizes
new telomeric DNA (Greider and Blackburn, 1985, 1989; Shippen Lentz and
Blackburn, 1989, 1990; Strahl and Blackburn, 1996; Lingner et al., 1997; Cech et
al., 1997). Most adult somatic human cells, however, lack telomerase activity
(Kim et al., 1994).

Senescent cells have short telomeres, leading to the attractive theory that
telomeric length provides the cell division counting mechanism noted by Hayflick
and Moorhead (reviewed in Hayflick, 2000). Several lines of evidence support
this theory. First, Harley et al. (1990) showed that telomeres in normal human
diploid fibroblasts shorten as the cells are cultured. Second, Allsopp et al. (1992)
showed that replicative capacity in culture correlates with initial telomere length.
Cells in culture with long telomeres can undergo more cell divisions than those in
culture with short telomeres. Apparently, the telomeric structure itself conveys
signals that count the number of cell divisions. A minimal length of telomeric
DNA is required so that the complement of telomeric binding proteins can
assemble and form a functional telomere. When telomeres reach this critical
minimum length in normal cells, telomeric structure is disrupted, and the p53 and
pr pathways are activated to initiate senescence (reviewed by Kim et al.,

2002). The exact mechanism for the activation of p53 and pr as a result of

21

telomeric loss remains to be uncovered.

Normal human fibroblasts have a finite lifespan, and such cells have never
been reported to become malignantly transformed. It is argued that the reason
for this is that they do not undergo a sufficient number of cell replications to
acquire all of the necessary phenotypic changes required to form tumors. In
contrast, tumorigenic human fibroblasts have an infinite lifespan, suggesting that
the acquisition of an infinite lifespan is a necessary prerequisite for tumor
development. Loss of the p53 and pr tumor suppressor pathways provides
one mechanism for human fibroblasts to acquire an infinite lifespan (reviewed by
Kim et al., 2002). When telomeres reach a minimal critical length in p53 and pr
deficient cells, genomic instability ensues. Such instability favors a genetic
change that leads to stabilization of telomeres, thereby promoting survive. For
example, induction of telomerase expression provides one way to stabilize
telomeres. Approximately 85-90% of human tumor cells display telomerase
activity (Kim et al., 1994; Ducray et la. 1999; Klingelhutz, 1999; Hahn and
Meyerson, 2001; Kang and Park, 2001). The remaining human tumor cells
lacking telomerase activity are also capable of maintaining telomeres through an
alternative recombination-based mechanism (ALT) (Bryan et al., 1997; Dunham

et al., 2002).

5. The ability to initiate and to sustain angiogenesis

Tumor tissue, like all tissue, requires an adequate blood supply. As tumor
tissue expands, so must its blood supply. When the cells of a tumor lack

necessary oxygen and nutrients and waste products accumulate, the cells die,

22

blocking further expansion of the tumor. Therefore, angiogenesis, which refers to
the sprouting of new capillaries from existing blood vessels, must occur (Gilbert,
1997). Angiogenesis is a complex process that involves several steps (reviewed
by Hanahan and Folkman, 1996; Webb and Vande Woude, 2000). Local
degradation of the basement membrane surrounding the endothelial cells of pre-
existing vessels occurs. The basement membrane, composed of extracellular
matrix proteins including laminins, collagens, and proteoglycans, provides an
acellular support for epithelial, mesenchymal, and endothelial cells (reviewed by
Hood and Cheresh, 2002). The endothelial cells of the vessels then change
shape and invade the surrounding stroma, forming a migrating column.
Proliferating endothelial cells make up the leading edge of the migrating column.
A region of differentiated endothelial cells follows the leading edge. The
differentiated cells, which no longer proliferate, change shape and tightly adhere
to each other to form the lumen of a new capillary tube. Fusion of the sprouting
tubes results in the formation of loops that circulate blood to the surrounding
fissue.

In the early 1970s, Folkman and colleagues provided the first evidence
that tumor cells must recruit blood vessels in order to survive. Specifically,
implantation of tumor fragments or tumor cells from culture at avascular sites in
rabbit corneal tissue results in the growth of new capillaries into the tumor
(Gimbrone et al., 1972). Inhibition of angiogenesis dramatically reduces the
growth of the Implanted tumor tissue or cells (Gimbrone et al., 1972). Additional

studies showed that a lack of adequate vasculature causes tumor cells to

23

undergo necrosis and/or apoptosis (Brem et al., 1976; Holmgren et al., 1995;
Parangi et al., 1996).

Studies of three transgenic mouse models of cancer (islet cell carcinoma,
dermal fibrosarcoma, and epidermal squamous cell carcinoma) as well as
studies of two human cancers (breast carcinoma and cervical carcinoma) provide
additional evidence that one of the phenotypic requirements of a developing
tumor is the activation of angiogenesis (Hanahan, 1985; Lacey et al., 1986;
Kandel et al. 1991; Sippola-Thiele et al., 1989; Arbeit et al., 1994; Hurlin et al.,
1995; Coussens et al., 1996; Weidner et al., 1991, 1992; Brown et al., 1995;
Guidi et al., 1994, 1995; Smith-McCune and Weidner, 1994). In these types of
cancer, histologically distinct intermediates mark the course of the conversion of
a normal cell into a tumor cell. By measuring the blood vessel density and the
level of certain immunological markers of angiogenesis (e.g. von Willebrand
factor) in these intermediates, researchers determined that the early lesions are
usually avascular, whereas the malignant tumors that arise are highly vascular.
For example, the development of dermal fibrosarcomas in a transgenic mouse
model is marked by at least three stages: mild fibromatosis, aggressive
fibromatosis, and fibrosarcomas (SippoIa-Thiele et al., 1989). Benign
hyperproliferation of dermal fibroblasts residing below the epidermis
characterizes both types of fibromatosis. Greater cell density, an increased
mitotic index, and a general thickening of the dermal layer distinguish aggressive
fibromatosis from mild fibromatosis (Kandel et al., 1991). Fibrosarcomas, the

malignant lesions, protrude from the skin and have an even greater cell density

24

and an anaplastic character compared with benign fibromatoses. Using
immunohistochemical staining of von Willebrand factor, a protein expressed
specifically in endothelial cells, Kandel et al. (1991) measured the degree of
vascularization in normal dermis, mild fibromatoses, aggressive fibromatoses,
and fibrosarcomas. The normal dermal tissue and mild fibromatosis tissue show
a low level of vascularization. In contrast, aggressive fibromatosis tissue and
fibrosarcomas exhibit a high degree of vascularization, indicating that the
activation of angiogenesis occurs during the transition from mild fibromatosis to
aggressive fibromatosis. Such a change contributes to the evolution of a
malignant tumor.

These studies led to the theory of an “angiogenic switch” (reviewed by
Hanahan and Folkman, 1996). During tumor development, a change occurs in
the cells that “switches on” angiogenesis, thereby converting an avascular tumor
to an angiogenlc tumor. Additional evidence, such as the studies of Bouck and
colleagues presented below, indicates that the angiogenic switch is not quite as
simple as “on” or “off". Rather, the formation of new blood vessels becomes
increasingly prevalent during tumorigenesis (reviewed by Webb and Vande
Woude, 2000). A balance between pro-angiogenic and anti-angiogenic factors
actually controls the angiogenic switch. In normal tissue and avascular tumor
tissue, anti-angiogenic factors outweigh pro-angiogenic factors, and the genesis
of new capillaries does not occur. In angiogenic tumors, pro-angiogenic factors
dominate, and new capillaries sprout. The studies of Bouck and colleagues,

which defined the mechanism for the acquisition of an angiogenic phenotype by

25

Li-Fraumeni fibroblasts during their conversion to tumorigenicity, illustrate the
balance model of the angiogenic switch (Dameron et al., 1994a and 1994b,
Bouck et al., 1996; Volpert et al., 1997). Fibroblast cell lines derived from Li-
Fraumeni patients contain one wild-type p53 allele and one mutant p53 allele.
Passaging of these cells results in loss of the wild-type p53 allele in a subset of
the population. The cells lacking wild-type p53 protein acquire an infinite life-
span. lntroduction of the H-RAS oncogene into these cells followed by
appropriate selection can yield tumorigenic cell lines (Bischoff et al., 1991).
However, less than half of the Ras—transformed Li-Fraumeni fibroblast cell lines
are tumorigenic. This indicates that additional changes must occur in the Ras-
transfonned Li-Fraumeni cell lines to produce tumorigenic cell lines (Bischoff et
al., 1991). The angiogenic switch necessary in the Li-Fraumeni model of
tumorigenesis occurs in two steps through two key mediators of angiogenesis,
thrombospondin-1 (TSP-1) and vascular endothelial growth factor (VEGF). First,
loss of wild-type p53 protein in these cells results in the loss of TSP-1 expression
and secretion and in an increase in VEGF expression and secretion (Dameron et
al., 1994a, 1994b; Volpert et al., 1997). Second, expression of oncogenic H-Ras
protein in the p53 null Li-Fraumeni cells further increases VEGF expression and
secretion. In this model system, the secretion of the anti-angiogenic factor TSP-
1 decreases as the secretion of the pro-angiogenic factor VEGF increases
(Volpert et al., 1997). Therefore, neovascularization occurs because the factors
that stimulate angiogenesis override those that would inhibit it.

TSP-1 belongs to a multi-protein family that also includes

26

thrombospondin-Z, -3, -4, and -5 (reviewed by de Fraipont et al., 2001). TSP-1
and TSP-Z constitute a TSP subfamily. Secreted as disulfide-bonded
homotrimers, TSP-1 and TSP-2 associate with molecules in the extracellular
matrix as well as with cellular receptors. Both TSP-1 and TSP-2 have been
shown to be potent inhibitors of angiogenesis in numerous cell types (reviewed
by Adams, 2001). Because TSP-1 has been studied in more detail than TSP-2,
the discussion that follows will emphasize the current model for TSP-1’s inhibition
of angiogenesis. TSP-1 regulates several properties of endothelial cells
including proliferation, motility, cell-cell interactions, and morphogenesis
(reviewed by Cavallaro and Christofori, 2001). Jimenez et al. (1996)
demonstrated that TSP-1 inhibits angiogenesis by causing endothelial cells to
undergo apoptosis. Briefly, TSP-1 binds to the extracellular domain of the C036
receptor. The interaction of TSP-1 and CD36 recruits the Src-family non-
receptor tyrosine kinase Fyn to the cytoplasmic tail of CD36. The downstream
consequence of the TSP-1-CD36-Fyn interaction is the activation of the caspase
3-like protein. The signaling intermediates connecting Fyn to the caspase 3-like
protein have not been identified. Activation of the caspase 3-Iike protein leads to
phosphorylation of p38 and translocation of phospho-p38 from the cytosol to the
nucleus. Presumably, nuclear localized phospho-p38 stimulates the transcription
of pro-apoptotic genes. Production of pro-apoptotic proteins then induces
apoptosis. Subsequent to this study, Volpert et al. (2002) demonstrated that the
pro-apoptotic Fas/Fas ligand pathway is required for TSP-I-mediated apoptosis

in endothelial cells. Activation of p38 by the TSP-1 pathway leads to increased

27

expression of the Fas ligand (FasL), which binds to its death-inducing receptor
Fas to activate the extrinsic apoptotic pathway. Interestingly, Volpert et al.
(2002) also demonstrated that treatment of endothelial cells with pro-angiogenic
factors, such as VEGF, increases the amount of Fas receptor on the cell surface,
thereby sensitizing the angiogenic endothelial cells to apoptosis. In contrast, pro-
angiogenic factors like VEGF also promote cell survival and induce the
expression anti-apoptotic proteins. Again, the idea of a balance comes into play.
When endothelial cells are exposed to a limiting amount of pro-angiogenic
factors, apoptosis occurs. When pro-angiogenic factors increase to a critical

amount, survival overrides apoptosis and new blood vessels form.

6. The ability to invade surrounding tissues and to metastasize

The metastasis of primary tumors accounts for 90% of cancer deaths
(reviewed by Hanahan and Weinberg, 2000). Metastasis refers to the spread of
tumor cells from their site of origin to new sites in the body (Alberts et al., 1994).
For metastasis to occur, tumor cells must enter the circulatory system (reviewed
by Hood and Cheresh, 2002 and Chambers et al., 2002). To gain access to the
circulatory system, tumor cells infiltrate the basement membrane that surrounds
the tumor itself and the endothelial cells of blood vessels. After the invading
tumor cells successfully penetrate these basement membranes, the tumor cells
enter the circulatory system and travel to the tissue where the secondary tumor
will arise. At this secondary site, the tumor cells must again penetrate the
basement membrane of the vasculature and of the target tissue. Once the tumor

cells have successfully invaded the target tissue, the cells must initiate and

28

sustain growth to give rise to a secondary tumor. It is important to note that there
is an intimate link between metastasis and angiogenesis. As discussed above,
angiogenesis involves the local degradation of the basement membrane
surrounding the endothelial cells of pre-existing vessels and the subsequent
invasion of the surrounding tumor tissue by the endothelial cells. Because the
vasculature in this angiogenic environment is compromised, tumor cells can
more easily enter the circulatory system and metastasize.

Certain cancers seem to have preferred sites for metastasis (Weiss et al.,
1986; Weiss, 1992; Radinsky et al., 1995; Radinsky and Ellis, 1996). For
example, breast carcinomas often metastasize to bone, liver, brain, and lungs.
Prostate cancer primarily metastasizes to bone. Colorectal carcinomas
predominately metastasize to the liver. Two main ideas emerged to explain
organ-specific metastases. The “seed and soil” model is attributed to Steven
Paget (1889). This theory proposes that although cancer cells (“seeds”) spread
virtually everywhere throughout the body, only specific organs (“soil”) provide
adequate conditions for those specific types of tumor cells to develop into
secondary tumors. The second model, proposed by James Ewing (1928),
attributes that path of blood flow in the body as the reason for certain tumor types
to metastasize to specific organs. In their review of metastasis, Chambers et al.
(2002) explain that experimental evidence supports both theories. The pattern of
blood flow in the body adequately explains only 66% of metastases; the
remaining metastases may be explained by interactions between the tumor cells

and the new environment.

29

The processes of invasion and metastasis are very complex and require
the cooperation and coordination of many cellular signaling pathways. The tumor
cells switch from a proliferative to a migratory program in order to move from the
primary tumor site into the bloodstream. Once in the bloodstream, the cells
activate survival pathways. At the secondary tumor site, the cells must then
switch from a migratory to a proliferative program to establish the metastatic
lesion. Cellular adhesion molecules, cytoskeletal proteins, and proteases play a

critical role in this process.

a. Integrins, cellular adhesion and signaling proteins that play
a critical role in invasion and metastasis

Heterodimeric cell surface glycoproteins known as integrins regulate a
cell’s interaction with the surrounding extracellular matrix (reviewed by Juliano,
2002). There are 18 known integrin a-subunits and 8 known integrin B-subunits
expressed in human cells that form non-covalently heterodimers, each with
unique ligand binding properties. Each (1 and 8 subunit is composed of a large
extracellular domain, which contains ligand binding motifs, a transmembrane
domain, and a relatively short cytoplasmic domain, which can interact with
intracellular molecules (reviewed by Liu et al., 2000).

To promote a migratory phenotype, alterations in integrin expression
patterns occur in tumor cells. Expression of integrins that promote migration in a
specific cell type increases while expression of those that inhibit migration
decreases. For example, invasive melanomas express a high level of the av83

integrin at the invasive front of the tumor and in angiogenic blood vessels,

3O

whereas the premalignant lesions and quiescent blood vessels express a low
level of this integrin (Felding-Habemlann et al., 1992; Brooks et al., 1994).
Breast carcinoma cells decrease expression of specific integrin subunits,
including 01, (13, as, 81, and B4 (reviewed by Mizejewski, 1999). Alternatively,
reorganization of the cellular distribution of integrins to promote migration can
occur. For example, in colon carcinoma cells, the laminin-binding integrin (1684
moves from stable cellular junctions known as hemidesmosomes to migratory
regions of the cells, where it functions in the formation and stabilization filopodia
and lamellae (Rabinovitz and Mercurio, 1997).

The interaction of integrins and extracellular matrix proteins causes
integrin clustering and the formation of focal contacts or focal adhesions
(reviewed by Juliano, 2002). Various intracellular proteins bind to the integrin
cytoplasmic tall at focal adhesions to transduce signals (reviewed by Hynes,
2002). Most interestingly, integrins not only convey signals from the extracellular
environment to the intracellular environment, known as outside-in signaling, but
they also convey signals from the intracellular environment to the extracellular
environment, known as inside-out signaling.

Outside-in signaling modulates the cell’s response to extracellular stimuli.
Intracellular proteins such as the actin binding proteins a-actinin, talin, and
paxillin and the scaffold proteins RACK1 and Grb2 bind to integrin cytoplasmic
domains in response to an extracellular matrix protein binding to integrin
extracellular domains (Otey et al., 1990; Pavalko et al, 1991; Horwitz et al., 1986;

Knezevic et al., 1996; Pfaff et al., 1998; Schaller et al., 1995; Tanaka et al., 1996;

31

Chen et al., 2000; Liu et al., 1999; Liu and Ginsberg, 2002; Liliental and Chang,
1998; Law et al., 1996). Kinases including focal adhesion kinase (FAK) and the
Src family of non-receptor tyrosine kinases also interact with integrin cytoplasmic
domains in response to an extracellular matrix protein binding to integrin
extracellular domains (Schaller et al., 1992, 1999; Hildebrand et al. 1993; Chen
et al., 2000). These interactions link integrins to the actin cytoskeleton and to
kinases, two key mediators of cellular motility. For example, FAK associates with
the cytoplasmic domain of integrins at focal adhesions and becomes activated by
autophosphorylation of tyrosine 397 (reviewed by Parsons et al., 2002).
Phosphorylated FAK can then activate several signaling cascades depending
upon which intracellular proteins bind to it. One branch involves the recruitment
of the scaffolding protein GrbZ to the focal adhesion. This leads to Ras activation
and subsequent activation of the RAF1IMEK/ERK pathway. In addition to
activating gene transcription, Erks directly regulate enzymes required for cellular
motility. Erk phosphorylates myosin-Iight-chain kinase (MLCK), a kinase that
regulates contractile force in cells (Klemke et al., 1997). Activation of myosin
through MLCK generates force that pulls the cell fonlvard toward new adhesive
contacts and breaks old contacts.

Inside-out signaling modulates the cell’s adhesiveness, the strength of the
interaction between integrins and the extracellular matrix (Huttenlocher et al.,
1996; Palecek et al., 1997). Proteins that bind to integrin cytoplasmic domains
can affect integrin quaternary structure, thereby influencing the integrin’s affinity

and avidity for its ligands. An intermediate level of adhesiveness promotes

32

migration. A high level of adhesiveness immobilizes cells, whereas a low level of
adhesiveness fails to provide an adequate amount of traction, rendering

migration inefficient.

b. Matrix metalloproteinases, a family of proteases that plays
a critical role in invasion and metastasis

To gain access to the circulatory system, tumor cells infiltrate the
basement membranes surrounding the tumor itself and the endothelial vessel
cells. The matrix metalloproteinase (MMP) family of proteases plays a critical
role in the regulated degradation of basement membrane (reviewed by Nagese
and Woessner, 1999, Egeblad and Werb, 2002; Hood and Cheresh, 2002).
Specifically, two members of the family, MMPZ and MMP9, are closely linked to
invasion and metastasis. MMPZ and MMP9 degrade type IV collagen, the
predominant component of the basement membrane. Increased MMP protein
has been detected on stromal cells in proximity to invasive melanoma cells, on
metastatic breast carcinoma cells, and on invading angiogenic endothelial cells
(Monteagudo et al., 1990; Pyke et al., 1992; Brooks et al., 1996; Deryugina et al.,
1997). The fact that MMP protein is found not only on tumor cells but also on
surrounding stromal and endothelial cells emphasizes the importance of the
interplay of multiple cell types in the tumor microenvironment. Signals from
surrounding cells can promote or inhibit tumor growth.

In addition to being a barrier to invasion and metastasis, the extracellular
matrix also serves as a track along which the cells migrate (Hood and Cheresh,

2002). Therefore, the proteolytic processing of the basement membrane must be

33

tightly regulated to prevent over-degradation of the extracellular matrix track.
MMPs are secreted by cells as inactive enzymes (pro-MMPs), and activation of
pro-MMPs occurs on the cell surface (reviewed by Nagese, 1997). For example,
the cell surface receptor membrane-type 1 MMP (MT1-MMP) contains a
protease domain, and it cleaves the pro-peptide from pro-MMPZ to yield mature,
active MMP2 (Murphy et al., 1999). Active MMPZ can then associate with the
integrin avl33, thereby localizing active MMPZ at the invasive front of the tumor
(Brooks et al. 1996; Deryugina et al., 2000; Hofrnann et al., 2000). One
consequence of MMPZ activity is the generation of MMPZ inhibitory fragments
known as hemopexin (PEX) fragments, which bind to MMPZ preventing MMPZ-
dvl33 interactions (Brooks et al., 1998). Therefore, a negative feedback loop
provides one mechanism to regulate MMPZ-mediated proteolysis of the
extracellular matrix.

B. The Multi-Step Model of Tumorigenesis

1. Epidemiological evidence

It is widely accepted that cancer develops via a multi-step process, which
involves a series of genetic changes that cause a normal cell to evolve into a
cancer cell. Using a mathematical model to estimate the number of changes
required to cause human carcinomas, Peto (1977) concluded that a very large
number of rate-limiting cellular changes is not necessarily required to convert an
epithelial cell into a carcinoma cell. In fact, he estimated that approximately 6
changes are required. Subsequently, Renan (1993) extended the work of Peto.

He analyzed 28 types of human malignancies and found that approximately 7-12

34

genetic changes are required for human tumors to develop. The presence of
intermediates between normal cells and malignant cells, as is seen in colon
cancer, further supports the multi-step model of tumorigenesis (Kinzler and
Vogelstein, 1996). Exposure to carcinogens does not lead to the immediate
occurrence of cancer. For example, individuals who smoke cigarettes develop
lung cancer only after decades of exposure to the carcinogens in cigarette smoke
(Peto, 1977). Moreover, the incidence of cancer in a population increases with
age (Dix, 1989). Both observations support the contention that cancer develops
via a multi-step process, which involves a series of genetic changes that occur

over time to convert a normal cell into a cancer cell.

2. Colorectal tumorigenesis

Because the process of tumorigenesis in colorectal epithelium is marked
by distinct morphological changes, colorectal cancer provides an excellent model
for studying the multi-step nature of tumorigenesis. Normal colon epithelia,
dysplastic adenomas (benign tumors), and colon carcinomas can be
distinguished and isolated to determine a genetic profile for each type of tissue.
Through molecular analyses of tissues representing each stage of colorectal
tumorigenesis, researchers demonstrated that at least seven genetic events
must occur to give rise to colorectal carcinomas (reviewed by Kinzler and
Vogelstein, 1996).

Study of familial adenomatous polyposis (FAP), an autosomal dominant
genetic disorder that predisposes patients to colorectal cancer, led to the

discovery of one key tumor suppressor gene involved in preventing colorectal

35

cancer (Kinzler and Vogelstein, 1996). FAP patients develop hundreds to
thousands of colorectal tumors known as adenomatous polyps (Gardner, 1951,
1972). A single polyp, which is benign, poses no immediate danger to the
patient’s life. The sheer number of polyps that develop, however, guarantees
that a cell in at least one benign polyp will accrue additional genetic changes to
progress to malignancy. In view of this situation, colorectal cancer develops in
virtually 100% of FAP patients unless the colon is removed, and such diagnoses
occur approximately 25 years earlier than in individuals who develop sporadic
colorectal cancer (reviewed by Fearon and Jones, 1992; Kinzler and Vogelstein,
1996; Marsh and Zori, 2002).

FAP patients inherit a gerrnline inactivating mutation in one allele of the
adenomatous polyposis coli (APC) gene (Groden et al., 1991; Nishisho et al.,
1991). The acquisition of an inactivating somatic mutation in the other APC allele
results in the loss of functional APC protein, which causes dysplasia in colorectal
epithelium (Kinzler and Vogelstein, 1996). The APC protein exerts its tumor
suppressive power by negatively regulating a component of the Wnt signaling
pathway, thereby inhibiting this proliferation promoting pathway (reviewed by
Moon et al., 2002). (The Wnt pathway is reviewed in Section II.) The APC
protein forms a complex with axin and glycogen synthase kinase-3B (GSK-BB)
that binds to B-catenin. The formation of a quaternary complex of B-
catenin/APCIaxin/GSK-3B enables GSK-SB to phosphorylate B-catenin. Such
phosphorylation targets B—catenin for proteosomal degradation (reviewed by

Bienz, 2002). lnactivating mutations in the APC gene, which result in loss of

36

functional APC protein, have been observed in greater than 80% of all colorectal
cancers, supporting the argument that APC is an important gatekeeper in
colorectal epithelium (Kinzler and Vogelstein, 1996). Moreover, loss of functional
APC protein appears to be an early event in colorectal tumorigenesis because
such loss has been detected in dysplastic aberrant crypt foci (ACF), the earliest
observable colorectal neoplastic lesions (Jen at al., 1994).

Activating mutations of the proto-oncogene K-RAS have been detected
in adenomas, which occur later in the development of colorectal cancer (Kinzler
and Vogelstein, 1996). It is believed that ACFs give rise to adenomas. As
discussed above, the Ras family of small G-proteins activates cellular
proliferation. Constitutive activation of Ras small G-proteins through activating
mutations can enable cells to proliferate in the absence of mitogenic signals.
Such a change likely plays a causal role in the progression of an ACF to an
adenoma.

Loss of heterozygosity at chromosomes 18q and 17p occurs even later in
the progression of colorectal tumorigenesis (Kinzler and Vogelstein, 1996).
These chromosomal aberrations have been detected in large adenomas and in
carcinomas. Two putative tumor suppressor genes have been identified at the
18q locus: the deleted in colorectal carcinoma (DCC) gene and the deleted in
pancreatic carcinoma (DPC4/SMAD4) gene (Fearon et al., 1990 and Hahn et al.,
1996). The DOC gene encodes a protein containing immunoglobulin superfamily
repeats (reviewed by Graziano et al., 2002; Calvert and Frucht, 2002). Such

domains may indicate a role for DCC in cell to cell or cell to matrix interactions.

37

Figure 1. The multi-step and clonal model of carcinogenesis: colon cancer.
The multi-step and clonal model of carcinogenesis predicts that tumors originate
from a single cell that has acquired all of the necessary genetic changes to be
able to produce a malignant tumor. All cells of the tumor are the progeny of this
first malignant cell. This diagram illustrates how this model applies to human
colon cancer. Activating mutations in colon cancer-related oncogenes (e.g.,
RAS) and inactivating mutations in colon cancer-related tumor suppressor genes
(e.g., APC, DCC,/DPC4, and p53) provide a growth advantage to the cell,
thereby enabling clonal expansion. Each arrow indicates a single cell that
acquired a genetic change (e.g., loss of APC); each triangle represents a large
population of colon epithelial cells that arose through clonal expansion of the cell
that has acquired one or more colon cancer-related mutation(s). The
histologically distinct types of colon tissues that emerge during this process are

indicated below the triangles. (Adapted from Kinzler and Vogelstein, 1996)

38

_. 952“.

«805050 qu:oo< «80:23 «60:33 83:25.5 £23525
8a.. 3835.8... 23m 053395 .5502

 

 

 

 

 

 

 

 

nmn VOA—o mg 6 0m<
“—0 500 “—0
mac.— Co 30..

wood

39

Mehlen et al. (1998) and Chen et al. (1999) report that DCC may play a role in
inducing apoptosis. The DPC4 gene encodes the SMAD4 protein, a key
mediator of the TGF-beta signaling cascade. This signaling pathway acts as a
negative regulator of cellular proliferation in colon epithelia. SMAD4
heterodimerizes with receptor-activated SMADs, and these heterodimers
translocate to the nucleus to regulate transcription of TGF-beta responsive genes
(reviewed by Weinstein et al., 2000). The specific genes regulated depend upon
which receptor-activated SMAD interacts with SMAD4. Loss of functional
SMAD4 protein prevents transcription of TGF-responsive genes, which encode
proteins to repress cellular proliferation. The gene implicated at the chromosome
17p locus is the tumor suppressor gene p53. lnactivating mutations in the p53
gene, which results in loss of functional p53 protein, have been observed in more
than 50% of colorectal cancers (reviewed by Markowitz et al., 2002). As
discussed above, because of its potent ability to inhibit cell growth and to induce
apoptosis, p53 is a key protein involved in protecting cells from malignancy.
Studies of hereditary nonpolyposis colorectal cancer (HNPCC), another
disease that predisposes individuals to colon cancer, shed additional light onto
the process of colorectal tumorigenesis. Although HNPCC patients develop
adenomas at a rate similar to non-affected individuals, these adenomas almost
assuredly progress to malignancy (reviewed by Lynch et al., 1993, 2000). The
accelerated rate of tumorigenesis evident in HNPCC patients occurs because of
loss of function of proteins involved in DNA mismatch repair (MMR). Individuals

affected by HNPCC inherit an inactivating mutation in one allele of a MMR gene.

40

Gerrnline mutations have been observed in five different MMR genes: hMSHZ,
hMSH6, hMLH1, hPMS1, and hPMSZ (Leach et al., 1993; Fishel et al., 1993;
Bronner et al., 1994; Papadopoulos et al., 1994; Nicolaides et al., 1994; Miyaki et
al., 1997; Akiyama et al. 1997). The acquisition of a somatic mutation in the
other allele yields a cell that cannot perform MMR. As in FAP patients, colorectal
cancer diagnoses occur much earlier in HNPCC patients compared with
individuals who develop sporadic colorectal cancer (Kinzler and Vogelstein,
1996). Because DNA repair is compromised, mutations can quickly accumulate
in the colorectal epithelial cells of the adenomas. Mutations that occur in the
genes involved in colorectal tumorigenesis lead to the conversion of adenomas
to carcinomas. HNPCC illustrates how deficiency in DNA repair can accelerate

the tumorigenesis process.

3. MSU-1 lineage as a model for tumorigenesis
a. Generation and characterization of the MSU-1 lineage
To study the multi-step process of tumorigenesis in human fibroblasts, the
Carcinogenesis Laboratory created the MSU-1 cell lineage as a model system
(Morgan et al., 1991). The MSU-1 lineage contains non-tumorigenic, diploid,
finite lifespan human fibroblasts through tumorigenic, nearly diploid, infinite life
span human fibroblasts (Figure 2). To establish the MSU-1 lineage, normal
human fibroblasts derived from the foreskin of a neonate were put into culture.
This cell line, designated LG1, was transfected with plasmid containing both the
viral oncogene v-MYC and the drug resistance marker neomycin (Morgan et al.,

1991). Drug resistant, clonal populations were propagated, and virtually all of

41

Figure 2. The MSU-1 human fibroblast lineage. To establish the MSU-1
lineage, normal human fibroblasts derived from the foreskin of a neonate,
designated LG1, were transfected with plasmid containing the viral oncogene v-
MYC. Clonal populations expressing v-Myc were propagated, and virtually all of
these populations eventually entered a senescent state and died. A few cells,
probably a clonal population, escaped senescence and continued to proliferate.
This surviving infinite life span cell strain was designated MSU-1.0. Subsequent
studies by our laboratory showed that telomerase is active in the MSU-1.0 cell
strain, whereas telomerase is inactive in its parental cell line LG1. During
propagation of the MSU-1.0 cell strain, a faster growing variant cell strain arose
and overgrew the culture. This infinite life span variant cell strain was designated
MSU-1.1. Karyotype analysis of the MSU-1.1 cell strain showed that it contains
45 chromosomes including two marker chromosomes. In addition to an infinite
lifespan, the MSU-1.1 cell strain is also partially growth factor independent.
When MSU-1.0 or MSU-1.1 cells are injected in athymic mice, neither forms
tumors. However, carcinogen treatment or oncogene transfection of MSU-1.1
cells followed by the isolation of focus-derived MSU-1.1 cells can produce cells
that are tumorigenic in athymic mice. In this figure, each arrow indicates a single
cell that acquired genetic change or changes (e.g., expression of v-Myc,
activation of telomerase, or the presence of two marker chromosomes); each
triangle represents a large population of human fibroblasts that arose through

clonal expansion of the cell that acquired one or more genetic changes.

42

 

 

 

 

N 239".

 

 

 

 

 

 

8:: .832: 35.2: 35.2: 3:5: 3:5...
25> a E . ..
your». "gnaw“ swam: c.7222 assess; 5..
-msoom
o yvv
0 s7
e 003% so? $99. 4%
0.90 «.0 coo «sew cove 0
owe « coo
case «J.

43

these populations eventually entered a senescent state and died. A few cells,
probably a clonal population, escaped senescence and continued to proliferate.
The surviving infinite life span cell strain was designated MSU-1.0. Karyotype
analysis of MSU-1.0 demonstrated that the cell strain is diploid and has no visible
chromosomal anomalies.

Transformed human cells often behave differently in culture than their
non-transformed counterparts. Properties such as an infinite lifespan, the loss of
contact inhibited growth, the loss of growth factor dependence, the loss of
anchorage-dependent growth, and the ability to form tumors in an appropriate
host distinguish transformed cells from non-transfon'ned cells (reviewed by
Pelengaris et al., 2002). Although the MSU-1.0 cell strain has an infinite lifespan,
it does not display the other properties of transformation. It is contact inhibited,
growth factor dependent, and anchorage dependent. It does not form tumors
when subcutaneously injected into athymic mice.

The fact that only one v-MYC expressing cell clone escaped senescence
to yield an infinite life span cell strain indicates that expression of the v-MYC
oncogene alone was insufficient to cause immortalization. Rather, at least one
additional genetic change was required to produce the immortalized human
fibroblast cell strain MSU-1.0. Subsequent studies by our laboratory show that
telomerase is active in the MSU-1.0 cell strain, whereas telomerase is inactive in
its parental cell line LG1. Additional studies have demonstrated that expression
of the transcription factor Sp1 is approximately 2-3 fold higher in MSU-1.1 cells

compared with LG1 cells (2. Lou, unpublished studies). Since both Myc and Sp1

44

function as transcription factors for the telomerase gene, the up—regulation of Sp1
together with Myc expression may be responsible for the immortalization of a
single clone of v-MYC expressing LG1 cells (Kyo et al., 2000).

During propagation of the MSU-1.0 cell strain, a faster growing variant cell
strain arose and overgrew the culture. This infinite life span variant cell strain
was designated MSU-1.1. Karyotype analysis of the MSU—l .1 cell strain showed
that it contains 45 chromosomes including two marker chromosomes. In addition
to an infinite lifespan, the MSU-1.1 cell strain is also partially growth factor
independent. MSU-1.1 does not display the other markers of transformation; it is
anchorage dependent, contact inhibited, and does not form tumors when
subcutaneously injected into athymic mice. However, carcinogen treatment
(gamma irradiation, UV irradiation, benzo(a)pyrene-dioI-epoxide, methyl-
nitrosourea, or ethyl-nitrosourea) or oncogene transfection (H-RAS or N-RAS) of
MSU-1.1 followed by the appropriate selection method can produce cells that
form tumors in athymic mice (Fry et al., 1986; Hurlin et al., 1989; Yang et al.,
1992; Lin et al., 1995; O’Reilly et al., 1998; Boley et al., 2000). Attempts to
generate tumorigenic MSU-1.0 derivatives using carcinogens or oncogenes have
consistently failed, indicating that the genetic changes that converted MSU-1.0
into MSU-1.1 are critical in the multi-step process of tumorigenesis. Thus, the
MSU-1 lineage provides a set of isogenic human fibroblast cell lines/strains at
varying steps of the tumorigenesis pathway. These cell lines/strains can be

utilized to dissect the molecular mechanisms of carcinogenesis.

45

b. Using the MSU-1 lineage to study the multi-step process of
tumorigenesis in human fibroblasts

The work of Qing et al. (1997, 1999) of this laboratory demonstrates the
usefulness of the MSU-1 lineage in studying the multi-step process of
tumorigenesis. Because the MSU-1 lineage contains isogenic cell lines/strains,
the gene expression pattern of non-tumorigenic MSU-1.1 cells can be compared
with the gene expression pattern of a tumorigenic MSU-1.1 derivative. Such
comparisons provide information about the genetic changes required to convert a
non-tumorigenic MSU-1.1 cell into a tumorigenic cell line. Genes with up-
regulated expression in the tumorigenic cell line represent potential oncogenes;
genes with down-regulated expression in the tumorigenic cell line represent
potential tumor suppressor genes. To carry out such a comparison, Qing et al.
(1997, 1999) used differential mRNA display. The gene expression profiles of
two cell strains/lines of the MSU-1 lineage, MSU-1.1 and L210 6AISB1, were
compared. MSU-1.1 is an infinite lifespan, nearly diploid human fibroblast cell
strain. As discussed above, oncogene transfection or carcinogen treatment of
MSU-1.1 followed by appropriate selection can produce cells that form tumors in
athymic mice. The cell line L210 6AISB1 is one such derivative. To generate the
L210 6AISB1 cell line, MSU-1.1 cells were treated with the carcinogen
benzo(a)pyrene-dioI-epoxide (BPDE), and BPDE-transformed cells were
selected using a focus assay. The tumorigenic potential of the focus-derived cell
strains was assayed by subcutaneous injection of the focus-derived cells into

athymic mice. Tumors that formed in the mice were removed and cultured to

46

yield tumor—derived MSU-1.1 cell lines. The tumorigenic MSU-1.1 derivative cell
line L210 6AISB1 was derived from a single tumor formed in an athymic mouse
by a single focus-derived cell strain. Through the comparison of MSU-1.1 cells
with L210 6AISB1 cells, Qing et al. (1997, 1999) identified two cDNAs strongly
down-regulated in the tumorigenic cell line L210 6AISB1. One cDNA isolated
encodes the fibulin-1D extracellular matrix protein. The other cDNA isolated
encodes a novel protein designated ST7.

Fluorescent in situ hybridization (FISH) analysis demonstrated that the
ST7 gene is located on the q arm of human chromosome 8 (8q22.2-23.1) (Qing
et al., 1999). Examination of the tissue distribution of ST7 expression by
Northern blotting analysis showed that it is widely expressed in human organs
(Qing et al., 1999). ST7 mRNA is most abundant in heart and skeletal muscle,
less abundant in brain, lung, pancreas and placenta, and nondetectable in liver
and kidney.

To determine if loss of ST7 expression is a general characteristic of
tumorigenic cells, Qing et al. (1999) performed Northern and Western blotting
analyses of a set of 31 tumor-derived cell lines consisting of 15 cell lines derived
from patients’ tumors (sarcomas and carcinomas) and 16 cell lines derived from
tumors formed in athymic mice by malignantly transformed MSU-1.1 cells; 92%
of the former and 38% of the latter express very low or non-detectable ST7.
Based on this data, Qing et al. (1999) concluded that ST7 may play an important
role in the transformation of human fibroblasts.

When the novel gene ST7 was first identified, it was recognized that it had

47

the characteristics of a transmembrane protein, but at that time no other proteins
with significant similarity to ST7 had been reported. Approximately one year
later, Yamamoto and his associates, using degenerate Oligonucleotides
corresponding to the highly conserved region of the ligand-binding domains
found in the proteins of the LDLR superfamily, discovered a novel LDLR-related
gene, which they designated as the LRP3 gene (Ishii et al., 1998). Our search of
the databases for proteins structurally related to ST7, after the sequence of LRP3
had been submitted, revealed a very strong similarity between LRP3 and ST7,
enabling us to recognize that ST7, although it differed significantly from the LDLR
prototype, was also a novel member of the LDLR superfamily. In 2000,
Sugiyama et al. identified the murine LDLR-related protein 9 (LRP9), another
protein homologous to both ST7 and LRP3. The identification of LRP9 supported
the classification of ST7 as a novel member of the LDLR superfamily and
suggested that ST7, LRP3 and LRP9 constitute a subfamily of the LDLR
superfamily. In light of our finding that ST7 is an LDLR-related protein, Section II
of this literature review contains a discussion of the structure and function of
several key LDLR superfamily members. Section II also discusses the

relationship of ST7 to LDLR-related proteins.

48

II. The LDLR superfamily proteins and their relationship to ST7

A. The LDLR Superfamily

The low-density lipoprotein receptor superfamily is composed of 14
proteins. The superfamily can be divided into core members and more distantly
related subfamilies. The seven core members of the superfamily are: 1) the low-
density lipoprotein receptor (LDLR) (Yamamoto et al., 1984); 2) the very low-
density lipoprotein receptor (VLDLR) (T akahashi et al., 1992); 3) the
apolipoprotein E receptor 2 (ApoER2) (Kim et al., 1996; Novak et al., 1996); 4)
the low—density lipoprotein receptor-related protein 1 (LRP1), which is also
referred to as the 02-macroglobulin receptor (Herz et al., 1988; Strickland et al.,
1990); 5) the low density lipoprotein receptor-related protein 2 (LRPZ), which is
also referred to as megalin (Saito et al., 1994); 6) the low—density lipoprotein
receptor-related protein 1B (LRP1B) (Liu et al., 2000); and 7) the multiple
epidermal growth factor-like domains protein 7 (MEGF7), which is also referred
to as human LRP4 (Nakayama et al., 1998). The low-density lipoprotein
receptor-related protein 5 (LRP5) (Dong et al., 1998; Hey et al., 1998) and the
low-density lipoprotein receptor-related protein 6 (LRP6) (Brown et al., 1998)
comprise one LDLR subfamily. Another subfamily contains ST7, LRP3, and
marine LRP9. Finally, LR11/SORLA, which is also referred to as human LRP9,
(Jacobsen et al., 1996; Morwald et al., 1997) and corin, which is also referred to
as mun'ne LRP4, (Tomita et al., 1998; Yan et al., 1999) are two more distantly
related members of the LDLR superfamily.

Classically, members of the LDLR superfamily such as LDLR, VLDLR,

49

ApoER2, LRP1, and LRPZ have been defined as endocytic receptors for a
plethora of diverse ligands, including lipoproteins and proteases (reviewed by
Strickland et al., 1995; Hussain et al., 1999; Howell and Herz, 2001). Recent
studies, however, reveal a new role for these proteins, namely as signal
transducers (reviewed by Li et al., 2001a; Herz and Bock, 2002; van der Geer,

2002)

1. LDLR

LDLR, the prototype receptor of the superfamily, has been studied since
the early 1970s, when Michael S. Brown and Joseph L. Goldstein undertook the
task of determining the molecular mechanism underlying the autosomal dominant
genetic disorder familial hypercholesterolemia (FH). Patients suffering from FH
have an increased concentration of plasma cholesterol and are susceptible to
heart attacks very early in life (reviewed by Brown and Goldstein, 1986).
Cholesterol exists in the bloodstream in a complex with transport plasma
lipoproteins. There are four major classes of plasma lipoproteins: low-density
lipoprotein (LDL), very low-density lipoprotein (VLDL), intermediate-density
lipoprotein (IDL), and high-density lipoprotein (HDL) (Brown and Goldstein,
1986). Of these four lipoproteins, LDL and HDL transport the majority of
cholesterol (Brown and Goldstein, 1986). In 1974, Brown and Goldstein provided
the first evidence for the existence of plasma lipoprotein receptors on the cell
surface. Specifically, they demonstrated that 125I-LDL binds to normal fibroblasts
but does not bind to FH fibroblasts, indicating that the FH fibroblasts lack a

functional LDL receptor (LDLR) (Brown and Goldstein, 1974a and 1974b).

50

Subsequent studies on the fate of the 125l-LDL bound to the cell surface revealed
that LDLR mediates uptake of LDL by cells via a process termed receptor-
mediated endocytosis (Brown and Goldstein, 1986). Using fibroblasts derived
from an FH patient that could bind to 125l-LDL but could not internalize ‘zsl-LDL,
they determined that LDLR internalization requires the clustering of receptors in
coated pits (Anderson et al., 1976; 1977a; 1977b). Brown and Goldstein
received the Nobel Prize in Physiology or Medicine in 1985 for their pioneering
work in unraveling the role LDL and LDLR in cholesterol homeostasis.

In 1984, Yamamoto et al. identified the human LDLR cDNA. Examination
of the tissue distribution of LDLR expression in various adult human tissues by
Northern blotting analysis shows that it is most abundant in liver and adrenal
tissue (reviewed by Hussain et al., 1999). LDLR is expressed at a lower level in
heart, placenta, lung, kidney, ovary, testis, intestine, brain, and skeletal muscle.
Expression is not detectable in adipose tissue. The human LDLR gene is located
on the p arm of chromosome 19 (19p13.1-13.3) (Lindgren et al., 1985). The
protein contains six important structural features: 1) a signal sequence; 2)
cysteine-rich, complement-like domains known as low-density lipoprotein
receptor class A (LDLRA) domains; 3) an epidermal growth factor (EGF)
precursor homology domain; 4) an O-linked glycosylation domain; 5) a single
transmembrane domain; and 6) a cytoplasmic domain containing a signal to
direct endocytosis (Figure 3) (reviewed by Brown and Goldstein, 1986; Krieger
and Herz, 1994; Hussain et al., 1999; Herz and Bock, 2002).

Proteins destined for the transport through the endoplasmic reticulum to

51

Figure 3. Structures of representative proteins of the LDLR superfamily
(adapted from Strickland et al., 2002). The types of domains in these proteins
and the organization of these domains divide the superfamily into at least four
subfamilies: 1) LDLR, VLDLR, and ApoER2; 2) LRP1, LRP1B, and LRPZ; 3)
LRP5 and LRP6; 4) ST7, LRP3, and murine LRP9. The cytoplasmic domains of
these proteins contain motifs implicated in endocytosis and/or signal
transduction, which are not shown. (Images in this dissertation are presented in

color.)

52

n 2:2“.

 

Equu £53336 5
52.63 532.3253 335.6 o
82.2 a:_:_8:ou.n._.>>> o

 

 

85 Seas. .882 3.33 .
Iooolmul I I . «I2 25 52:8 <53 l
m 58.8 mac -
roooizz F5
m 33%.. 8:253 .255 I
58'... :2 85

 

 

Iooimlgxz one...

12 «am.—

 

fz 2%...

:2 egg;
zoooll I2 «582
zooollm 12 53>

I000! IZ Kn—On—

53

the Golgi apparatus, lysosomes, cell membrane, or extracellular space contain
N-terminal signal sequences (reviewed by Claros et al., 1997). These
sequences consist of 15-25 residues characterized by a short, positively charged
N-terminal region, a hydrophobic core, and a more polar C-terrninal region. A
cleavage site often follows the signal sequence, and an enzyme known as a
signal peptidase cleaves the signal sequence peptide to yield a mature protein.
In the case of LDLR, the first 21 amino acids of the protein provide the signal
sequence, which is cleaved from the protein on its path from the endoplasmic

reticulum to the plasma membrane (Brown and Goldstein, 1986).

LDLR contains seven LDLRA ligand binding domains at its N-terminus,
arranged as a cluster of 4 repeats and a cluster of 3 repeats separated by a short
linker (Yamamoto et al., 1984). Each LDLRA domain is composed of
approximately 40 residues with two distinct features. First, six conserved
cysteine residues form three intramolecular disulfide bonds; second, a conserved
triad of serine, aspartic acid, and glutamic acid resides at the C-terminus of each
domain (Yamamato et al., 1984; Daly et al., 1995). In LDLR and other members
of the LDLR superfamily, the LDLRA domains function in ligand binding
(Yamamoto et al., 1984; Stldhof et al., 1985a, 1985b). For example, these
domains in LDLR bind to apolipoprotein E (apoE) and apolipoprotein B-100
(apoB-100) containing lipoprotein particles (Esser et al., 1988; Russell et al.,
1989). van Driel et al. (1987) showed that the binding of these ligands to LDLR
is Ca2*-dependent. Daly et al. (1995a) determined the first three-dimensional

structure of an LDLRA domain. Their work indicates that LDLRA domains fold to

54

form a B-hairpin structure followed by a series of 8 turns and that the conserved
serine, aspartic acid and glutamic acid residues at the C-terminus of the domain,
as well as additional acidic residues, are clustered on one face of the protein.
Subsequently, the three-dimensional structures of several more LDLRA domains
were determined, both from LDLR and from the LDLR-related protein 1 (LRP1)
(Daly et al., 1995b; Fass et al., 1997; Huang et al., 1999; North et al., 2000;
Simonovic et al., 2001). All additional structures agreed with that of Daly et al.
(1995a). It is thought all LDLRA domains fold in the same general manner, but
that each individual LDLRA domain has a unique surface with which ligands
interact. These interactions may rely on the acidic clusters, on accessible
hydrophobic residues, or on the combination of acidic and hydrophobic residues
(Huang et al., 1999). These unique surfaces most likely provide domain
specificity. For example, LDLR utilizes different combinations of its seven ligand
binding domains to interact with apoE or apoE-100 lipoprotein particles. LDLRA
domains 3-7 are required for apoB-100 binding, whereas LDLRA domain 5 is
required for apoE binding (Esser et al., 1988; Russell et al., 1989).

EGF precursor homology domains consist of three EGF-type cysteine rich
repeats and YWTD repeats that constitute a YWTD domain (reviewed by Krieger
and Herz, 1994). The EGF precursor homology domain of LDLR functions in the
acid-dependent dissociation of ligands from internalized LDLR and in recycling of
the receptor (Davis et al., 1987). The EGF-type cysteine repeats span
approximately 40 residues and contain six conserved cysteine residues that form

three intramolecular disulfide bonds. Two of the three EGF-type cysteine repeats

55

(A and B) are separated from the third repeat (C) via the YWTD domain (Krieger
and Herz, 1994). Repeats A and B bind calcium (Malby et al., 2001). The three-
dimensional structure of the A and B repeats from LDLR has been determined.
Saha et al. (2001) showed that these repeats have an elongated, rod-like
arrangement. In an analysis of the YWI'D repeats from a number of proteins
including LDLR, Springer (1998) redefined the boundaries of the YWTD repeat
regions. Previously, researchers reported YWTD repeats occur in clusters of
five. Springer defined a YWTD domain, which consists of six contiguous YWTD
repeats. He proposed that these repeats are not autonomous and that they must
fold together to form a single domain with a six-bladed B—propellor structure.
Jean at al. (2001) determined the three-dimensional structure of the YWTD
domain in LDLR and confirmed that this domain indeed folds as Springer (1998)
suggested.

The extracellular domain of LDLR functions to bind to ligands; the
intracellular domain directs endocytosis (reviewed by Brown and Goldstein,
1986; Krieger and Herz, 1994; Hussain et al., 1999). In 1990, Chen et al.
reported that the sequence NPW found in the cytoplasmic domain of LDLR is
required for clathrin-mediated endocytosis of the low density lipoprotein receptor
(LDLR). This sequence was more generally defined as NPXY since the residue
found at position three could be any amino acid. Further work indicated that a
larger six-residue sequence, FDNPW, is actually required by LDLR for
internalization (Collawn et al., 1991). In a 1993 review, Trowbridge et al. specify

a more general tyrosine-based internalization motif, namely aromatic-X—X—X-X-

56

aromatic, with arginine and proline perhaps preferred at the third and fourth
positions, respectively (aromatic-X-N-P-X-aromatic). (For the sake of simplicity,
LDLR-type tyrosine-based endocytic signal will be referred to as an NPXY motif.)
Davis et al. (1986) identified a point mutation in LDLR isolated from fibroblasts of
a patient suffering from familial hypercholesterolemia. This mutation alters the
LDLR endocytic signal, converting the critical tyrosine residue of the signal to
cysteine (FDNPW to FDNPVC). They demonstrated that the replacement of this
tyrosine by cysteine significantly decreased the internalization of 125I-LDL.

In summary, LDLR binds to lipoprotein cholesterol transporters in the
bloodstream through its interaction with either apoE or apoE-100 and then
internalizes the bound lipoproteins (reviewed by Nykjaer and Willnow, 2002).
Uptake of these lipoprotein particles from the bloodstream provides cells with the
cholesterol necessary to maintain membrane integrity and to synthesis steroid
hormones (Brown and Goldstein, 1986). Defects in LDLR function cause
aberrations in cholesterol metabolism and cholesterol homeostasis. In addition,
Trommsdorff et al. (1998) demonstrated that the cytosolic adaptor proteins Dab1
and Fe65 bind to the NPXY motif in the cytoplasmic domain of LDLR. No
functional significance has been uncovered for this interaction. LDLR has not

been implicated as a receptor that directly transduces signals.

2. VLDLR and ApoER2

In a search for LDLR related proteins, Takahashi et al. (1992) identified
the rabbit VLDLR protein. Briefly, they used low stringency conditions to screen

an LDLR—subtracted rabbit heart cDNA library with a probe derived from the

57

LDLR sequence. Expression of one of the isolated cDNAs in an LDLR—deficient
Chinese hamster ovary (CHO) cell line enabled the cDNA-expressing cells to
bind to and internalize apoE-containing lipoproteins, VLDL, B—VLDL, and
intermediate density lipoprotein (lDL) but not LDL. Examination of the
distribution of VLDLR expression in various adult rabbit tissues by Northern
blotting analysis shows that it is most abundant in heart, muscle, and adipose
tissue and that it is also expressed at a lower level in spleen, lung, brain, kidney,
adrenal tissue, testis, and small intestine. In contrast to LDLR, VLDLR is barely
detectable in liver. In 1994, both Sakai et al. and Oka et al. identified the human
VLDLR cDNA. Sakai et al. (1994) isolated two VLDLR isoforms. These
researchers found that the pattern of VLDLR expression in human tissues
matches the pattern of VLDLR expression in rabbit tissue. The human VLDLR
gene is located on the p arm of chromosome 9 (9p24) (Oka et al., 1994). The
human and rabbit VLDLR proteins are virtually identical. Moreover, the structure
of the VLDLR protein is remarkably similar to that of the LDLR protein (Figure 3).
VLDLR contains eight LDLRA domains at its N-terminus, arranged as a cluster of
5 repeats and a cluster 3 repeats separated by a short linker (reviewed by
Hussain et al., 1999). As described above, LDLR contains only seven LDLRA
domains at its N-terminus. The presence or absence of the O-linked
glycosylation motif distinguishes the two VLDLR isoforms (Sakai et al., 1994).
Both isoforms bind VLDLR ligands, so the loss of this motif does not appear to
affect VLDLR function (Hussain et al., 1999).

In 1996, Kim et al. identified the human ApoER2 cDNA by screening a

58

human placenta cDNA library using PCR with degenerate primers corresponding
to the highly conserved internalization signal found in LDLR and VLDLR.
Expression of the ApoER2 cDNA in an LDLR-deficient CHO cells enables the
cells to bind to and internalize apoE-enriched B—VLDL efficiently. These studies
also show that ApoER2 binds to LDL, but with low affinity. Examination of the
distribution of ApoER2 expression in various adult human tissues by Northern
blotting analysis shows that it is mainly expressed in brain and in placenta (Kim
et al., 1996). A low level of expression is detectable in ovary and testis (reviewed
by Hussain et al., 1999). Expression is undetectable in heart, lung, liver, skeletal
muscle, kidney, and pancreas. The human ApoER2 gene is located on the p
arm of chromosome 1 (1p34) (Kim et al., 1997). The ApoER2 cDNA encodes a
protein very similar to LDLR and VLDLR. The ApoER2 protein contains seven
LDLRA domains at its N-terminus, arranged as clusters of 5 repeats and Z
repeats separated by a linker (Figure 3).

The similarity between structures of VLDLR, ApoER2, and LDLR and the
fact that VLDLR and ApoER2 bind to and internalize lipoproteins suggests that
VLDLR and ApoER2 may function similarly to LDLR, namely in lipoprotein
metabolism (Takahashi et al., 1992; Oka et al., 1994; Kim et al., 1996). More
recently, however, the VLDR and ApoER2 proteins have been implicated in an
important signal transduction pathway that regulates brain development
(reviewed by Herz, 2002; Rice and Curran, 2001). The first clue that LDLR
superfamily members function in the regulation of brain development was

provided by Trommsdorff et al. (1998), who demonstrated that the cytosolic

59

adaptor protein mDab1 binds to the NPXY sequence found in the cytoplasmic
domains of both LDLR and LRP. Previously, Howell et al. (1997) showed that
Dab1 plays a critical role in brain development. The dab1 knock-out mouse
exhibits abnormal cellular layering in the cerebral cortex, hippocampus, and
cerebellum. This phenotype very closely matches that of the reeler mouse,
implicating Dab1 as a player in the Reelin pathway. The reeler mouse, which
arose spontaneously in a population of “snowy-bellied” mice, lacks the large
extracellular protein Reelin (D’Arcangelo et al., 1995; reviewed by Rice and
Curran, 2001). These knock-out mice showed that Reelin and Dab1 function in
the same pathway in the developing brain, but the receptor connecting the
extracellular Reelin protein to the intracellular Dab1 adaptor protein was not
known. The fact that Dab1 binds to the NPXY sequence in LDLR superfamily
members made these proteins interesting candidates for the Reelin receptor.
Finally, characterization of the vIdIr/apoerZ double knock-out mouse showed that
its phenotype is virtually indistinguishable from that of the reeler mouse,
suggesting that VLDLR and ApoER2 serve as receptors for Reelin (Trommsdorff
et al., 1999). Further studies show that both Reelin and Dab1 bind to VLDR and
ApoER2 (Trommsdorff et al., 1999; D’Arcangelo et al., 1999; Hiesberger et al.,
1999). Moreover, Reelin must interact with VLDLR and ApoER2 for tyrosine
phosphorylation of Dab1 to occur (D’Arcangelo et al., 1999; Hiesberger et al.,
1999). Phosphorylation of Dab1 is required to transmit Reelin signals (Howell et
al., 2000).

In their recent review, Rice and Curran (2001) synthesize the available

60

data regarding the Reelin signaling pathway and outline the following model for
the role of VLDLR and ApoER2 in this pathway. Reelin binds to the extracellular
domains of VLDLR and ApoER2. Such binding has two major consequences,
the internalization of Reelin and the activation of a tyrosine kinase signaling
cascade resulting in the phosphorylation of Dab1 protein bound to the
cytoplasmic domain of the receptors. ApoE inhibits this signaling cascade,
presumably by binding to VLDLR and ApoER2 thereby preventing Reelin binding
to the receptors. The tyrosine kinase required for Dab1 phosphorylation has not
yet been identified. Neither VLDLR nor ApoER2 possess tyrosine kinase activity.
Endocytosis of the receptor complex might play a key role in this cascade by
shuttling Dab1 from the membrane into the cytosol where a tyrosine kinase can
act on it. This theory is intriguing because it links the classical function of LDLR
superfamily proteins, namely endocytosis, with a novel function, namely signal
transduction. Alternatively, VLDLR and ApoER2 may interact with another cell
surface receptor that itself has tyrosine kinase activity or that recruits a tyrosine

kinase.

3. LRP1 and LRP1B

The goal of Herz et al. (1988) was to identify additional receptors for
apoE-containing lipoproteins. Because Yamamoto et al. (1984) showed that the
conserved acidic residues in the LDLRA ligand-binding domains mediate the
interaction of LDLR and apoE, Herz et al. (1988) designed Oligonucleotides
complementary to these highly conserved acidic regions in LDLR and screened a

mouse leukocyte library to isolate cDNAs containing such conserved regions.

61

One partial cDNA isolated from the mouse library encodes eight LDLRA
domains. They used this cDNA as an initial probe for cDNA walking in a human
liver cDNA library and isolated eight cDNAs covering approximately 15 kB.
Together, these cDNAs encode a 4544 residue protein designated LRP1.
Subsequent sequencing of the az-macroglobulin receptor protein revealed that it
is identical to LRP1 (Strickland et al., 1990). Examination of the expression
pattern of LRP1 in various adult human tissues demonstrated that LRP1 is most
abundant in liver, brain, and lung, moderately abundant in adrenal tissue, and
least abundant in intestine, kidney, placenta, ovary, and testis (reviewed by
Hussain et al., 1999). The human LRP1 gene is located on the q arm of
chromosome 12 (12q13.1-14.3) (Myklebost et al., 1989; Hilliker et al., 1992).

Like other members of the LDLR superfamily, LRP1 contains five key
structural elements: 1) a signal sequence; 2) cysteine-rich, complement-like
domains known as low-density lipoprotein receptor class A (LDLRA) domains; 3)
epidermal growth factor (EGF) precursor homology domains; 4) a single
transmembrane domain; and 5) a cytoplasmic domain containing a signal to
direct endocytosis (Figure 3) (reviewed by Krieger and Herz, 1994). Unlike
LDLR, VLDLR, and ApoER2, LRP1 does not contain an O-linked glycosylation
domain. LRP1 is approximately five times larger than the LDLR, VLDLR, and
ApoER2 proteins. It is synthesized as a single polypeptide chain of 600 kD.
Cleavage of this polypeptide into two subunits, a 515 kD subunit and a 85 kD
subunit, occurs in the Golgi apparatus, presumably by the protease furin. At the

cell surface, the two LRP1 subunits associate by strong noncovalent interactions.

62

The large 515 kD subunit contains the majority of the protein’s extracellular
domain; the smaller 85 kD subunit mainly contains the protein’s transmembrane
and cytosolic domains.

The extracellular region of LRP1 contains 31 LDLRA ligand-binding
domains, 22 EGF-type cysteine-rich repeats, and 8 YWTD domains (Figure 3)
(Herz et al., 1988). The LDLRA domains are arranged in four clusters containing
2, 8, 10, and 11 LDLRA repeats. Twelve of the 22 EGF-type cysteine-rich
repeats, together with four of the eight YWTD domains, constitute four EGF-
precursor homology domains. A single EGF-precursor homology domain follows
each cluster of LDLRA domains. Four EGF-type cysteine-rich repeats and the
four remaining YWTD domains form incomplete EGF-precursor homology
domains, each comprised of one YWTD domain followed by one EGF-type
cysteine rich repeat. A single incomplete EGF-precursor homology domain
immediately precedes the second LDLRA cluster, and three incomplete EGF-
precurscr homology domains immediately precede the third LDLRA cluster. The
remaining six EGF-type cysteine-rich repeats link the extracellular domain to the
protein’s transmembrane domain. The protein’s cytoplasmic domain contains
two copies of the NPXY endocytic signal (Herz et al., 1988). The LRP1
cytoplasmic tail also harbors two additional types. of internalization signals, a
single YXXO internalization motif and two dileucine motifs (Li et al., 2000). The
function of these internalization motifs is discussed below.

Although first identified as an apoE-binding receptor, subsequent studies

of LRP1 revealed that its extracellular domain recognizes greater than 30

63

different ligands (reviewed by Herz and Strickland, 2001; Strickland et al. 2002).
These ligands include lipoproteins, proteinases, proteinase-inhibitor complexes,
extracellular matrix proteins, bacterial toxins, and viruses. Most LRP1 ligands
interact with the second and fourth LDLRA domain clusters (Moestrup et al.
1993; Nykjaer et al., 1994; Horn et al., 1997; Neels et al., 1999). The three-
dimensional structures of two individual LDLRA domains located in the second
LDLRA cluster of in LRP provide insights into the mechanism by which one
protein interacts with so many different ligands (Huang et al., 1999; Dolmer et al.,
2000). In general, the primary structure of each individual LDLRA domain is
highly variable, with the exception of the six conserved cysteine residues and the
conserved acidic residues at the C-terminus of each domain. The conserved
residues enable LDLRA domains to fold similarly. The highly variable nature of
the other residues in each LDLRA domain provides each domain with a unique
surface with which ligands interact. For example, each domain has a unique
charge density that dictates ligand binding (Huang et al., 1999; Dolmer et al,
2000). Recently, Mikhailenko et al. (2001) demonstrated that LDLRA clusters
may cooperate in ligand binding. Specifically, binding of 02-macroglobulin
requires both the first and second LDLRA domain clusters.

It has been well—established that LRP1 acts as an endocytic receptor for
its many ligands (reviewed by Krieger and Herz, 1994; Hussain et al., 1999).
Ligand binding leads to internalization of the LRP1-ligand complex through
clathrin-coated pits (Czekay et al., 2001). Upon internalization, a decrease in

endosomal pH occurs, and the bound ligand dissociates. The receptor is then

64

recycled to the cell surface and the ligand is degraded in the lysosome. In
contrast to LDLR in which the NPXY signal directs LDLR-mediated endocytosis,
the single YXXO motif in the LRP cytoplasmic tail acts as the dominant signal for
LRP1-mediated endocytosis (Li et al., 2000). One of the two dileucine motifs
also functions to a lesser degree in directing endocytosis. Subsequently, Li et al.
(2001b) also demonstrated that protein kinase A (PKA) mediates phosphorylation
of a serine residue in the LRP1 cytoplasmic domain. Mutation of this serine
residue to either alanine or threonine abolishes LRP1 phosphorylation by PKA.
Moreover, these LRP1 mutants exhibit a reduced rate of endocytosis and less
efficient delivery of ligands for degradation, suggesting that PKA-mediated
phosphorylation of LRP1 regulates endocytosis.

Recent evidence suggests that LRP1 not only functions in endocytosis but
also in signal transduction (Trommsdorff et al., 1998; Gotthardt et al., 2000;
Barnes et al., 2001; Boucher et al., 2002; Loukinova et al., 2002; Lutz et al.,
2002). Trommsdorff et al. (1998) and Gotthardt et al. (2000) demonstrated that
the cytoplasmic tail of LRP1 binds to multiple adaptor and scaffold proteins.
These investigators utilized the yeast two-hybrid assay system, GST-fusion
protein pull-down assays, and co-immunoprecipitation to discover and confirm
these interactions. Trommsdorff et al. (1998) showed that the cytosolic adaptor
proteins Dab1 and Fe65 bind to the NPXY motif in the cytoplasmic domain of
LRP. Gotthardt et al. (2000) detected eleven proteins that interact with the
cytoplasmic tail of LRP1 (e.g., JIP-1, JIP-Z, lCAP-1, PSD-95). These LRP-1

interacting proteins are known to be involved in cellular processes such as "the

65

regulation of synaptic transmission, activation and modulation of MAP kinase
pathways, local organization of the cytoskeleton, cell adhesion, and endocytosis"
(Gotthardt et al., 2000).

Barnes et al. (2001) demonstrated that LRP1 is tyrosine—phosphorylated in
v-Src transformed fibroblasts and that tyrosine-phosphorylated LRP1 interacts
with the cytosolic adaptor protein Shc. They propose that LRP1 provides a link
between the Src and Ras pathways in v-Src transformed fibroblasts. Specifically,
myristylation of the non-receptor tyrosine kinase Src targets it to insert in the
plasma membrane. This event localizes Src in proximity to the LRP1
cytoplasmic domain, and Src phosphorylates LRP1. Tyrosine-phosphorylated
LRP1 then recruits Shc to the membrane by binding to Shc’s phosphotyrosine
binding (PTB) domain. At the membrane, Src can phosphorylate Shc.
Phosphorylated Shc binds to the scaffolding protein Grb2, which recruits SOS
leading to the activation of the Ras pathway.

Finally, Boucher et al. (2002) and Loukinova et al. (2002) showed that
LRP1 serves as a co-receptor for the growth factor PDGF beta. Specifically,
these studies showed that LRP1 directly interacts with PDGF beta and that
treatment of fibroblasts with PDGF beta induces phosphorylation of tyrosine
residues in the cytoplasmic domain of LRP1. Moreover, only LRP1 molecules
localized in caveolae are phosphorylated upon PDGF treatment. Caveolae are
invaginations of the plasma membrane that are rich in cholesterol, GPI-anchored
proteins, and several receptor and nonreceptor tyrosine kinases (reviewed by

Pelkman and Helenius, 2002). These membrane structures are involved in

66

mediating endocytosis through a clathrin-independent mechanism. Moreover,
the fact that so many signaling molecules reside in caveolae suggests that these
structures may also function in regulating signal transduction. In addition to
LRP1, both PDGF receptors (PDGFR) and Src family kinases can be localized to
caveolae (Boucher et al., 2002). These studies of LRP1 phosphorylation further
demonstrated a requirement for the kinase activity of the PDGFR because
PDGFR-specific kinase inhibitors block LRP1 phosphorylation. However,
wortmannin, an inhibitor of PI3K, also completely blocks LRP1 phosphorylation,
suggesting that the PDGFR kinase does not directly phosphorylate LRP1.
Instead, Boucher et al. (2002) proposed that a PDGFR/PI3K activated tyrosine
kinase phosphorylates LRP1. The Src family kinases fit this profile. The work of
Barnes et al. (2001) described immediately above supports the conclusion that
Src family kinases phosphorylate LRP1. Moreover, Loukinova et al. (2002)
showed that a Src family-specific kinase inhibitor blocks LRP1 phosphorylation.
Taken together, these studies suggest the following model for LRP1’s interaction
with the PDGF beta pathway. Homodimers of the PDGF beta growth factor bind
to the PDGFR causing receptor dimerization and autophosphorylation of
tyrosines in the receptor’s cytoplasmic tail. Activation of PDGFR leads to
activation of Src family kinases co-Iocalized with PDGFR at caveolae. Because
LRP1 also binds to PDGF beta homodimers and both PDGFR and LRP1 reside
in caveolae, PDGFR and LRP1 co-localize at the cell surface. This brings LRP1
in close proximity to active Src family kinases, which phosphorylate LRP1.

Subsequently, phosphorylated LRP1 recruits Shc to the membrane. This leads

67

to activation of the She and Shc—mediated signaling pathways. Therefore, the
localization of these signaling molecules to caveolae directly contributes to this
pathway, suggesting an intimate link between signaling and endocytosis.

Using representational difference analysis (RDA), a technique for
comparing complex but highly related genomes to discern losses or gains in
chromosomal DNA, Lisitsyn et al. (1995) identified seven loci in the human
genome that appear to be homozygously deleted in a variety of human tumor cell
lines. One such homozygously deleted locus observed in both a renal cancer
cell line and a bladder cancer cell line mapped to chromosome Zq21.2. Liu et al.
(2000) proposed that chromosome 2q21.2 contains a candidate tumor
suppressor gene. Using exon trapping analysis and 5’ and 3’ RACE, they
identified another novel member of the LDLR superfamily, LRP1B (initially
designated LRP1-DH). Liu et al. (2000a, 2000b) then examined a panel of non-
small cell lung carcinoma (NSCLC) cell lines for alterations in the LRP1B gene.
They reported inactivation of the LRP1B gene in approximately 45% of the cell
lines studied. Inactivation of the LRP1B gene can occurs through homozygous
deletions within the LRP1B gene. Point mutations in the LRP1B alleles were
also observed. Finally, some cell lines expressed abnormal truncated LRP1B
transcripts. In contrast, none of the small cell lung cancer cell lines studied had
any observable defects in the LRP1B sequence, and only one such line
contained abnormal LRP1B transcripts. These findings strongly suggest that
inactivation of LRP1B plays an important in the development of non-small cell

lung cancer, but not the development of small cell lung cancer. Most recently,

68

Langbein et al. (2002) also documented alterations of the LRP1B gene in high
grade urothelial cancer.

LRP1 and LRP1B share 59% amino acid identity (Liu et al., 2000b).
Moreover, the LRP1 and LRP1B proteins have virtually identical structures
(Figure 3). LRP1B contains one additional LDLRA domain in its fourth cluster,
and it has a small sequence insertion in its cytoplasmic domain between the two
NPXY domains. The expression pattern of LRP1B differs from that of LRP1. As
described above, LRP1 expression is most abundant liver, lung, and brain
(reviewed by Hussain et al., 1999); LRP1B expression is most abundant in
thyroid gland, salivary gland, and brain (Liu et al., 2001). LRP1B transcripts are
also detectable by RT-PCR in heart, kidney, lung, liver, and bladder, although
expression in these tissues is low.

Because of the strong structural similarity between LRP1 and LRP1B, Liu
et al. (2001) proposed that LRP1B binds to LRP1 ligands. Therefore, they
assessed the ability of three LRP1 ligands, urokinase plasminogen activator
(uPA), tissue plasminogen activator (tPA), and plasminogen activator inhibitor
type-1 (PAI-1), to bind to LRP1B. Both uPA and tPA are serine proteases, and
PAl-1 is a serine protease inhibitor that targets uPA and tPA (reviewed by
Andreasen et al., 2000). As discussed above, LRP1 contains two major ligand
binding sites, the second and fourth LDLRA clusters. Because previous work
demonstrated that an LRP1 minireceptor containing the fourth LDLRA cluster,
the transmembrane domain, and the cytoplasmic tail binds to and internalizes

ligands, Liu et al. (2001) constructed a similar LRP1 B minireceptor, containing its

69

fourth LDRLA cluster, transmembrane domain, and cytoplasmic tail. LRP1-null
CHO cells expressing the LRP1B minireceptor bind to and internalize uPA, tPA,
and PAl-1, however, the kinetics of ligand endocytosis differs between LRP1 and
LRP1B. Specifically, LRP1B internalizes ligands much more slowly than LRP1.
The finding that LRPlB serves as a receptor for uPA and PAI-1 raises the
possibility that LRP1B regulates the uPA system. It is well established that the
uPA system plays a key role in the invasion and metastasis of tumor cells as well
as in angiogenesis (reviewed by Dano et al., 1999; Andreasen et al., 2000). To
understand the role LRP1B plays in regulating this system, it is first necessary to
review how LRP1 regulates the uPA system. Much work has shown that LRP1
mediates the function of the uPA system (Nykjaer et al., 1992; Kounnas et al.,
1993; Conese et al., 1995; Nykjaer et al., 1997; Weaver et al., 1997; Czekay et
al., 2001; also reviewed in Herz and Strickland, 2001). Three recent reviews, two
by Andreasen et al. (1997, 2000) and one by Herz and Strickland (2001), explain
the uPA proteolytic system and its interaction with LRP1. Briefly, uPA is
synthesized as a single chain zymogen (pro-uPA) and proteolysis of pro-uPA
yields an active two-chain enzyme (uPA). Interaction of uPA with its receptor
uPAR, a 55-kDa glycosylphosphatidyl-inositol (GPI) -anchored cell surface
protein, enhances the conversion of pro-uPA to uPA by concentrating pro-uPA
on the cell surface. Receptor-bound uPA then cleaves plasminogen to produce
the active protease plasmin, which degrades many components of the
extracellular matrix and may be involved in activating matrix metalloproteinases

(MMPs). The serine proteinase inhibitor PAI-l binds to receptor-bound uPA. A

70

conformational change occurs in PAI-1 upon binding to uPA, revealing a high
affinity binding site for LRP1. Interaction of the PAl-1:uPA:uPAR complex with
LRP1 triggers rapid endocytosis of the complex. uPAR and LRP1 are recycled,
and the uPA and PAI-1 molecules are degraded. In summary, LRP1 functions to
regenerate unoccupied uPAR on the cell surface. Unoccupied uPAR then binds
to pro-uPA to begin the proteolytic cascade again. Therefore, the presence of
LRP1 on the cell surface promotes uPA-mediated cellular responses, such as
cellular motility.

In contrast, the presence of LRP1B on the cell surface inhibits the uPA
proteolytic cascade (Liu et al., 2002). CHO cells expressing an LRP1B
minireceptor exhibit a 90% reduction in endocytosis of PAl-1zuPAzuPAR
complexes compared with CHO cells expressing a comparable LRP1
minireceptor. Moreover, the LRP1 B-expressing cells display a decreased rate of
cellular migration compared with the LRP1-expressing cells. Therefore,
accumulation of LRP1B:PAI-1:uPA:uPAR complexes on the cell surface inhibits
the regeneration of unoccupied uPAR, the activation of plasminogen by uPA, and
cellular migration. Liu et al. (2002) propose that LRP1B functions a tumor
suppressor by negatively regulating proteolytic cascades necessary for invasion
and metastasis. Therefore, loss of LRP1B protein would contribute to the
regeneration of free uPAR because PAI-1:uPA:uPAR complexes would be free to
bind to another receptor with faster endocytic kinetics. Ultimately, this causes a
higher level of uPA activity on the surface of cells lacking LRP1B. As discussed

above, uPA activity leads to degradation of the extracellular matrix and activation

71

of MMPs. Such activity contributes to tumor invasion and metastasis.

4. LRP5 and LRP6

In their search for candidate genes in an insulin-dependent diabetes
mellitus locus (IDDM4) at chromosome 11q13, Hey et al. (1998) identified a
cDNA encoding the LDLR-related protein 5 (LRP5). Specifically, they
constructed a contig containing the IDDM4 locus and used high-throughput
sequence analysis to identify genes within this region. Simultaneously, Dong et
al. (1998) identified the LRP5 cDNA in their screen for novel genes expressed in
human osteoblasts. Examination of the tissue distribution of LRP5 expression in
various adult human tissues by Northern blotting analysis shows that it most
abundant in liver, pancreas, prostate, placenta, and small intestine (Hey et al.,
1998). It is less abundant in heart, lung, skeletal muscle, kidney, spleen, thymus,
testis, and colon. A very low level of LRP5 mRNA is detectable in brain and
leukocytes.

Brown et al. (1998) used a region of the LRP5 nucleotide sequence to
screen a mouse liver cDNA library for LRP5-related sequences and isolated a
cDNA encoding a part of the novel LDLR-related protein LRP6. To isolate the
human LRP6 cDNA, they used the mouse LRP6 cDNA as probe to screen a
human kidney cDNA library. Examination of the tissue distribution of LRP6
expression in various adult human tissues by Northern blotting analysis shows
that it is most abundant in heart, brain, placenta, lung, kidney, pancreas, spleen,
testis, and ovary. It is expressed at a lower level in liver, skeletal muscle,

prostate, and colon. LRP6 mRNA ls undetectable in thymus, small intestine, and

72

leukocytes. The human LRP6 gene is located on the p arm of chromosome 12
(12p11.2-13.3Z4) (Brown et al., 1998).

The LRP5 and LRP6 proteins share 82% amino acid similarity, as
determined by the protein-protein BLAST program (www.ncbi.nlm.nih.gov). Most
importantly, these proteins consist of the same types of functional motifs
arranged in an identical configuration (Figure 3). The N-terminus of each protein
contains a putative signal sequence to direct the protein to the plasma
membrane, residues 125 in LRP5 and 1-19 in LRP6. These signal peptides are
predicted to be cleaved from the mature protein. Each mature protein consists of
four key structural features linking these LRPs to the LDLR superfamily: 1)
components of EGF-precursor homology domains; 2) LDLRA ligand-binding
domains; 3) a single transmembrane domain; 4) a cytoplasmic tall with a signal
to direct endocytosis (Hey et al., 1998; Brown et al., 1998). The arrangement of
these conserved protein domains in LRP5 and LRP6, however, differs from that
of the core members of the LDLR superfamily (Figure 3). The N-terminus of both
LRP5 and LRP6 contain four YWTD domains and four EGF-type cysteine rich
repeats. A single EGF-type cysteine rich repeat follows each YWTD domain.
Therefore, the extracellular domain of LRP5 and LRP6 contains four incomplete
EGF-precursor homology domains. Three LDLRA ligand-binding domains follow
the incomplete EGF-precursor homology domains.

The cytoplasmic tails of LRP5 and LRP6 are highly conserved between
these two proteins, but are much less similar to the other LDLR family members.

The serine and proline rich cytoplasmic tails of LRP5 and LRP6 lack the

73

conserved NPXY internalization sequence found in other LDLR family members.
These proteins, however, contain two other motifs implicated in directing
receptor-mediated endocytosis (Hey et al., 1998; Brown et al, 1998). Both LRP5
and LRP6 contain a tyrosine-based YXXO internalization motif, which was first
identified as the signal required for clathrin-mediated endocytosis of the
transferrin receptor and the cation independent mannose 6-phosphate receptor
(Jing et al., 1990; Collawn et al., 1990;Canfield et al., 1991; Lobel et al., 1989;
Jadot et al., 1992). In addition to this tyrosine-based motif, LRP5 and LRP6
contain a dileucine motif, a category of sorting signals distinct from the tyrosine-
based motifs. It has been reported that dileucine motifs play two roles in protein
sorting (Trowbridge et al., 1993; Rapoport et al., 1998). First, dileucine motifs
are recognized at the plasma membrane and direct clathrin-mediated
endocytosis. Second, dileucine motifs are recognized in the trans-Golgi network
(TGN) and direct protein sorting from the TGN to Iysosomes or secretory
vesicles.

The unique arrangement of the extracellular domains in LRP5 and LRP6
as well as their unique cytoplasmic tails indicates that these proteins constitute a
subfamily of the LDLR superfamily. No studies have been done to demonstrate
that LRP5 and LRP6 function in receptor-mediated endocytosis. However,
several groups recently demonstrated that LRP5 and LRP6 act as essential
players in the Wnt signaling pathway (Wehrli et al., 2000; Tamai et al., 2000;
Pinson et al., 2000; Mac et al., 2001). In their studies of the Wingless (Wg)

pathway, the Drosophila counterpart of the Wnt pathway, Wehrli et al. (2000)

74

isolated the arrow gene and determined that its gene product is essential for all
Wg signaling pathways. The arrow protein is homologous to the human and
murine LRP5 and LRP6 proteins. Tamai et al. (2000) showed that ectopic
expression of the human LRP6 protein in Xenopus embryos induces expression
of Wnt-responsive genes, leading to dorsal axis duplication and neural crest
formation. They also demonstrated that LRP6 binds to Wnt-1 and that the LRP6—
Wnt complex associates with Frizzled (Fz). There are at least ten Fz proteins
that constitute a family of receptors for Wnt glycoproteins (reviewed by Wodarz
and Nusse, 1998). The findings of Tamai et al. (2000) suggest that LRP5 and
LRP6, together with Fz, act as co-receptors for Wnt family glycoproteins. Pinson
et al. (2000) provided additional evidence supporting the evolutionarily conserved
role of LRP5 and LRP6 as Wnt co-receptors. Specifically, they isolated mice with
a disrupted LRP6 gene and found that these mice, which lack functional LRP6
protein, die at birth because of severe developmental defects, including
malformation of the skeleton, limbs, eyes, central nervous system, and urogenital
tract. The phenotype of LRP6 knock-out mice mimics the phenotype of Wnt null
mice (Pinson et al., 2000). LRP6 null mice, however, do not demonstrate all of
the phenotypes of Wnt null mice, and the defects observed in the LRP6 null mice
are generally less severe than those observed in Wnt null mice. The presence of
LRP5 may partially compensate for the lack of LRP6 in these mice. LRP5 null
mice exhibit a different phenotype compared with LRP6 null mice. These mice
are viable and fertile but display defects in eye vascularization and in skeletal

development (Kato et al., 2002). Interestingly, LRP5 heterozygote mice also

75

displayed a mild defect in skeletal development. The activity of the Wnt-
responsive transcription factor Lef1 in osteoblasts derived from LRP5 null mice
was lower than in osteoblasts derived from normal mice (Kato et al., 2002). This
suggests that LRP5 acts to mediate Wnt signaling in osteoblasts.

Mutations in the human LRP5 gene result in defects in bone density and
eye development (Gong et al., 2001). Individuals homozygous for inactivating
mutations in the LRP5 gene suffer from the autosomal recessive genetic disorder
osteoporosis-pseudogIioma (OPPG) syndrome. Children born with OPPG
syndrome have a very low bone density and are prone to bone fractures and
skeletal and ocular deformities. Similar to the LRP5 heterozygous mice, OPPG
carriers (individuals with one mutant LRP5 allele and one wild-type LRP5 allele)
exhibit a lower bone density than non-carriers. Gong et al. (2001) also provide
evidence of a functional interaction between LRP5 and Wnt family glycoproteins.
Recently, two other groups reported that LRP5 mutations cause the autosomal
dominant high-bone-mass (HBM) trait (Little et al., 2002; Boyden et al., 2002).

Although the studies described above show that LRP5 and LRP6 act as
co-receptors for Wnt glycoproteins, they do not explain how these co-receptors
transduce Wnt signals. Specifically, studies of the Wnt signaling pathway show
that the binding of Wnt to its co-receptors leads to the stabilization of B-catenin
protein, the translocation of stabilized B-catenin protein from the cytosol to the
nucleus, the activation of the TCF/LEF family of transcription factors by B—catenin
protein, and the transcription of TCF/LEF responsive genes (reviewed by Wodarz

and Nusse, 1998; Polakis, 2000; Zorn, 2001). An important breakthrough in

76

understanding how these co-receptors transduce Wnt signals comes from the
work of Mao et al. (2001), who demonstrated that LRP5 binds to the scaffolding
protein axin, thereby promoting B—catenin stabilization. Axin contains binding
sites for three important intracellular Wnt pathway proteins: B-catenin, glycogen
synthase kinase-3B (GSK-3B), and APC (Nakamura et al., 1998; Kishida et al.,
1998). In unstimulated cells, a complex of axin, B-catenin, GSK—3B, and APC
forms (reviewed by Bienz, 2002; Moon et al., 2002). GSK-3B phosphorylates l3-
catenin, which targets B-catenin for ubiquitination and subsequent proteosomal
degradation. Mao et al. (2001) proposed that the binding of Wnt to LRP5 causes
axin to translocate from the cytosol, where it promotes the degradation of [3-
catenin, to the cell membrane, where it no longer promotes the degradation of [3-
catenin. Therefore, B—catenin accumulates in the cytosol. This leads to the
translocation of B—catenin from the cytosol to the nucleus and subsequent
activation of Wnt-responsive genes. They also showed that localization of axin at
the membrane promotes the destabilization and degradation of axin.
Up-regulated activation of gene transcription by B-catenin occurs in human
cancer cells through disruption of the Wnt signaling pathway. Several genes in
the Wnt pathway have been shown to be proto-oncogenes (B-CA TEN/N) or
tumor suppressor genes (APC, AXIN). Monoallelic activating mutations in the B-
CATENIN gene have been observed in colorectal tumors, gastric tumors,
hepatocellular carcinomas, thyroid tumors, endometrial ovarian tumors, prostate
tumors, and melanomas (reviewed by Polakis, 2000). These mutations usually

affect the APC-interaction domain of B-catenin such that mutant B-catenin can

77

not interact with APC (Morin et al., 1997; Rubinfeld et al. 1997). Therefore, it is
not targeted for proteosomal degradation. As described above, biallelic
inactivating mutations in the APC gene have been observed in greater than 80%
of all colorectal cancers, supporting the argument that APC is an important
gatekeeper in colorectal epithelium (Kinzler and Vogelstein, 1996). Biallelic
inactivating mutations in the axin gene have been observed in human
hepatocellular carcinomas (Satoh et al., 2000). It will not be surprising if
mutations in LRP5 and LRP6 are observed in human tumors. Although
mutations in F2 have not been documented in human tumors, overexpression of
F2 has been observed in human esophageal tumor cells but not in matched
normal cells (Tanaka et al., 1998). Such overexpression leads to increased
nuclear localization of B—catenin and a concomitant increase in transcription of

Wnt-responslve genes.

5. LRP3 and LRP9, two ST7-related proteins

When the novel gene ST7 was first identified, it was recognized that it had
the characteristics of a transmembrane protein, but at that time no other proteins
with significant similarity to ST7 had been reported. Approximately one year
later, Yamamoto and his associates discovered a novel protein with a high
degree of similarity to ST7 (Ishii et al., 1998). To isolate novel members of the
LDLR superfamily, they designed Oligonucleotides corresponding to a highly
conserved region of the ligand-binding domains found in the LDLR superfamily
proteins and used these Oligonucleotides to amplify novel LDLR-like cDNAs from

a rat liver cDNA library. One cDNA isolated contained DNA sequences encoding

78

LDLR ligand-binding domains, and it was designated LDLR-related protein 3
(LRP3). Using the rat LRP3 cDNA as a probe, they then isolated the human
LRP3 cDNA from a library constructed from the human liver cell line HepG2.
Using FISH analysis, Ishii et al. (1998) determined that the LRP3 gene is located
on the q arm of human chromosome 19 (19q12-13.2). Examination of the tissue
distribution of LRP3 expression by Northern blotting analysis shows that it is
highly expressed skeletal muscle, ovaries, heart, brain, liver, pancreas, prostate,
and small intestine and weakly expressed in testis, colon, and leukocytes. A
trace amount of LRP3 transcript is detectable in placenta, lung, kidney, spleen,
and thymus. Finally, Ishii et al. (1998) reported that LRP3 protein fails to bind to
B—VLDL and the receptor associated protein (RAP), two proteins that bind to
other members of the LDLR superfamily. Therefore, they concluded that LRP3
may function in a pathway other than lipoprotein metabolism. No further studies
describing LRP3 have been reported.

In 2000, Sugiyama et al. identified the murine LDLR-related protein 9
(LRP9), a protein homologous to both ST7 and LRP3. To isolate novel members
of the LDLR superfamily, they used a novel signal sequence trap method to
screen a mouse lymphocyte cDNA library. This procedure takes advantage of
the fact that cell membrane receptors contain signal peptides in their protein
sequences that direct the receptors to their final destination in the plasma
membrane. Briefly, cDNA fragments from the mouse lymphocyte cDNA library
were fused to a constitutively active, signal-sequence deficient mpl cytokine

receptor. These hybrid cDNAs were expressed in a murine interleukin-3-

79

dependent pro-B cell line. The presence of an in-frame signal sequence in the
lymphocyte cDNA fragment causes targeting of the constitutively active mpl
cytokine receptor to the plasma membrane. Cells with the constitutively active
mpl cytokine receptor on the cell surface can grow in the absence of interleukin-
3; cells lacking the constitutively active mpl cytokine receptor on the cell surface
undergo apoptosis. Therefore, surviving cells contained lymphocyte cDNAs that
encode signal sequence peptides. Once recovered from the surviving pro-B
cells, the cDNA fragments were used to isolate the full-length cDNAs of interest,
including LRP9. Examination of the tissue distribution of LRP9 expression in
various adult murine tissues by Northern blotting analysis shows that it is highly
expressed in heart, lung, liver, and kidney, weakly expressed in brain and spleen,
and not expressed in skeletal muscle and testis (Sugiyama et al., 2000).
Examination of the tissue distribution of LRP9 expression in various adult human
tissues by Northern blotting analysis using the mouse LRP9 probe shows a
slightly different expression pattern (Sugiyama et al., 2000). The highest
expression of LRP9 transcripts in the human tissues examined occurs in the
kidney. Skeletal muscle and heart tissue have slightly lower levels of LRP9
mRNA. Trace amounts of LRP9 mRNA are detectable in leukocyte, lung,
placenta, small intestine, liver, spleen, thymus, and colon. LRP9 mRNA is
undetectable in brain. Sugiyama et al. (2000) also demonstrated that apoE-
enriched B—VLDL stimulates cholesteryl ester formation in LDLR-deficient CHO
cells expressing HA—tagged LRP9. Based on this observation, they concluded

that LRP9 mediates the cellular uptake of apoE-enriched B-VLDL. They did not,

80

however, provide evidence of a direct interaction between apoE-enriched B-VLDL
and LRP9. Finally, they used radiation hybrid mapping to localize the human
LRP9 gene to the q arm of chromosome 14 (14q11.2). The human homologue of
the murine LRP9 cDNA has not yet been isolated, and no further studies
describing LRP9 have been reported. (The protein referred to as human LRP9 in
the literature is not the homologue of the murine protein. It is the LR11/SORLA
protein.)

The protein structures of LRP3 and LRP9 are highly conserved. LRP3
and LRP9 consist of the same types of functional motifs arranged in an identical
configuration (Figure 3). Similar to other LDLR superfamily members, the N-
terrninus of each protein contains a putative signal sequence to direct the protein
to the plasma membrane, residues 1-30 in LRP3 and residues 1-17 in LRP9
(Ishii et al., 1998; Sugiyama et al., 2000). These signal peptides are predicted to
be cleaved from the protein. The extracellular domain of each mature protein
consists of two key structural features. First, like other LDLR superfamily
members, LRP3 and LRP9 contain LDLRA ligand-binding domains. There are
five LDLRA domains present in LRP3 and four present in LRP9. In LRP3, these
domains are arranged in two clusters, one containing two domains and the other
containing three domains. In LRP9, the first cluster contains only one domain
and the second cluster contains three domains. Second, unlike other LDLR
superfamily proteins, LRP3 and LRP9 contain two Complement factor C1s/C1r,
Urchin embryonic growth factor, Bone morphogenetic protein (CUB) domains.

One CUB domain precedes each LDRLA cluster. Each CUB domain is

81

composed of approximately 110 residues with four conserved cysteine residues
that form two intramolecular disulfide bonds. These domains are found in a wide
variety of unrelated proteins and are believed to participate in protein-protein
interactions (reviewed by Bork and Beckmann, 1993; Christensen and Bim,
2002). The three-dimensional structure of the CUB domain present in bovine
acidic seminal fluid protein (aSFP), a member of the sperrnadhesin protein
family, has been determined to be a barrel-like structure with two five-stranded 6-
sheets joined by surface exposed B-tums (Romao et al., 1997). CUB domain
containing proteins have been shown to form dimers. For example, the three
dimensional structure of a heterodimer of two porcine seminal plasma
sperrnadhesins, PSP-l and PSP-ll, shows that these proteins interact through the
CUB domain present in each (Varela et al., 1997). Similar to other LDLR family
members, a single transmembrane domain follows the extracellular domain of
LRP3 and LRP9, and the cytoplasmic tails of LRP3 and LRP9 contain several
putative signals to direct endocytosis. Specifically, these proteins contain at least
one NPXY-like sequence, a YXXO sequence, and at least one dileucine motif.
The cytoplasmic tails of these proteins are also notably rich in serine, threonine,
and proline residues.

B. The relationship between ST7 and the LDLR superfamily

The identification of LRP3 and LRP9 enabled our group to classify ST7 as
a novel member of the LDLR superfamily. (Chapter II of this dissertation contains
a detailed analysis of the ST7 protein structure.) The ST7 protein shares 66%

amino acid similarity with LRP3 and 46% amino acid similarity with LRP9, as

82

determined by the protein-protein BLAST program (www.ncbi.nlm.nih.gov). The
N-termini of S17, LRP3 and LRP9 contain the same types of functional domains,
namely LDLRA domains and CUB domains, arranged in an identical
configuration (Figure 3). The juxtamembrane regions of the ST7, LRP3, and
LRP9 cytoplasmic domains are also highly conserved. The other members of
the LDLR superfamily are more distantly related to the ST7 subfamily proteins
(reviewed by Howell and Herz, 2001). Specifically, the extracellular domains of
the other LDLR superfamily proteins contain EGF-precursor homology domains,
which are not present in the ST7 subfamily proteins, but lack CUB domains,
which are present in the ST7 subfamily proteins. There are, however, three
features that link the ST7 subfamily to all other members of the LDLR
superfamily. First, all proteins contain signal sequences directing transport to the
cell membrane. Second, all proteins contain LDLRA ligand-binding clusters.
Finally, all proteins contain sequences implicated in receptor-mediated

endocytosis.

83

REFERENCES

Adams, J. C. (2001). Thrombospondins: multifunctional regulators of cell
interactions. Annu Rev Cell Dev Biol 17, 25-51.

Alberts, B., Bray, D., Lewis, J., Raff,M. Roberts, K., and Watson, JD. (1994)
Molecular biology of the cell.3rd ed. New York: Garland Publishing, Inc.

Akiyama, Y., Tsubouchi, N., and Yuasa, Y. (1997). Frequent somatic mutations
of hMSH3 with reference to microsatellite instability in hereditary nonpolyposis
colorectal cancers. Biochem Biophys Res Commun 236, 248-252.

Allsopp, R. C., Vaziri, H., Patterson, C., Goldstein, S., Younglai, E. V., Futcher,
A. B., Greider, C. W., and Harley, C. B. (1992). Telomere length predicts
replicative capacity of human fibroblasts. Proc Natl Acad Sci U S A 89, 10114-
10118.

Amati, B., Brooks, M. W., Levy, N., Littlewood, T. D., Evan, G. I., and Land, H.

(1993a). Oncogenic activity of the c-Myc protein requires dimerization with Max.
Cell 72, 233-245.

Amati, B., Littlewood, T. D., Evan, G. l., and Land, H. (1993b). The c-Myc protein
induces cell cycle progression and apoptosis through dimerization with Max.
Embo J 12, 5083-5087.

Anderson, R. G., Goldstein, J. L., and Brown, M. S. (1976). Localization of low
density lipoprotein receptors on plasma membrane of normal human fibroblasts

and their absence in cells from a familial hypercholesterolemia homozygote. Proc
Natl Acad Sci U S A 73, 2434-2438.

Anderson, R. G., Brown, M. S., and Goldstein, J. L. (1977a). Role of the coated
endocytic vesicle in the uptake of receptor-bound low density lipoprotein in
human fibroblasts. Cell 10, 351-364.

Anderson, R. G., Goldstein, J. L., and Brown, M. S. (1977b). A mutation that
impairs the ability of lipoprotein receptors to localise in coated pits on the cell
surface of human fibroblasts. Nature 270, 695-699.

Andreasen, P. A., Kjoller, L., Christensen, L., and Duffy, M. J. (1997). The
urokinase-type plasminogen activator system in cancer metastasis: a review. Int
J Cancer 72, 1-22.

Andreasen, P. A., Egelund, R., and Petersen, H. H. (2000). The plasminogen

activation system in tumor growth, invasion, and metastasis. Cell Mol Life Sci 57,
25—40.

84

Arbeit, J. M., Munger, K., Howley, P. M., and Hanahan, D. (1994). Progressive
squamous epithelial neoplasia in K14-human papillomavirus type 16 transgenic
mice. J Virol 68, 4358-4368.

Ashkenazi, A. (2002). Targeting death and decoy receptors of the tumour-
necrosis factor superfamily. Nat Rev Cancer 2, 420-430.

Ayer, D. E., Kretzner, L., and Eisenman, R. N. (1993). Mad: a heterodimeric
partner for Max that antagonizes Myc transcriptional activity. Cell 72, 211-222.

Ayer, D. E., Lawrence, Q. A., and Eisenman, R. N. (1995). Mad-Max
transcriptional repression is mediated by ternary complex formation with
mammalian homologs of yeast repressor Sin3. Cell 80, 767-776.

Baker, S. J., Fearon, E. R., Nigro, J. M., Hamilton, S. R., Preisinger, A. C.,
Jessup, J. M., vanTuinen, P., Ledbetter, D. H., Barker, D. F., Nakamura, Y., and
et al. (1989). Chromosome 17 deletions and p53 gene mutations in colorectal
carcinomas. Science 244, 217-221.

Barbacid, M. (1987). ras genes. Annu Rev Biochem 56, 779-827.

Barnes, H., Larsen, 8., Tyers, M., and van Der Geer, P. (2001). Tyrosine-
phosphorylated low density lipoprotein receptor-related protein 1 (Lrp1)
associates with the adaptor protein SHC in SRC-transforrned cells. J Biol Chem
276, 19119-19125.

Berns, K., Hijmans, E. M., and Bernards, R. (1997). Repression of c-Myc
responsive genes in cycling cells causes G1 arrest through reduction of cyclin
E/CDKZ kinase activity. Oncogene 15, 1347-1356.

Bienz, M. (2002). The subcellular destinations of APC proteins. Nat Rev Mol Cell
Biol 3, 328-338.

Bischoff, F. 2., Strong, L. C., Yim, S. O., Pratt, D. R., Siciliano, M. J., Giovanella,
B. C., and Tainsky, M. A. (1991). Tumorigenic transformation of spontaneously

immortalized fibroblasts from patients with a familial cancer syndrome. Oncogene
6, 183-186.

Blackburn, E. H., and Challoner, P. B. (1984). Identification of a telomeric DNA
sequence in Trypanosoma brucei. Cell 36, 447-457.

Blackburn, E. H. (1991). Structure and function of telomeres. Nature 350, 569—
573.

Blackburn, E. H., Greider, C. W., Henderson, E., Lee, M. S., Shampay, J., and

Shippen-Lentz, D. (1989). Recognition and elongation of telomeres by
telomerase. Genome 31, 553-560.

85

Blackwood, E. M., and Eisenman, R. N. (1991). Max: a helix-loop-helix zipper
protein that forms a sequence-specific DNA-binding complex with Myc. Science
251,1211-1217.

Boley, S. E., McManus, T. P., Maher, V. M., and McCormick, J. J. (2000).
Malignant transformation of human fibroblast cell strain MSU-1.1 by N-methyl-N-

nitrosourea: evidence of elimination of p53 by homologous recombination.
Cancer Res 60, 4105-4111.

Bork, P., and Beckmann, G. (1993). The CUB domain. A widespread module in
developmentally regulated proteins. J Mol Biol 231, 539-545.

Bos, J. L. (1989). ras oncogenes in human cancer: a review. Cancer Res 49,
4682-4689.

Bouchard, C., Dittrich, O., Kiermaier, A., Dohmann, K., Menkel, A., Eilers, M.,
and Luscher, B. (2001). Regulation of cyclin D2 gene expression by the
Myc/Max/Mad network: Myc-dependent TRRAP recruitment and histone
acetylation at the cyclin DZ promoter. Genes Dev 15, 2042-2047.

Boucher, P., Liu, P., Gotthardt, M., Hiesberger, T., Anderson, R. G., and Herz, J.
(2002). Platelet-derived growth factor mediates tyrosine phosphorylation of the
cytoplasmic domain of the low Density lipoprotein receptor-related protein in
caveolae. J Biol Chem 277, 15507-15513.

Bouck, N., Stellmach, V., and Hsu, S. C. (1996). How tumors become
angiogenic. Adv Cancer Res 69, 135-174.

Boyden, L. M., Mao, J., Belsky, J., Mitzner, L., Farhi, A., Mitnick, M. A., Wu, 0.,
lnsogna, K., and Lifton, R. P. (2002). High bone density due to a mutation in
LDL-receptor-related protein 5. N Engl J Med 346, 1513-1521.

Brem, S., Brem, H., Folkman, J., Finkelstein, D., and Patz, A. (1976). Prolonged
tumor dormancy by prevention of neovascularization in the vitreous. Cancer Res
36, 2807-281 2.

Bronner, C. E., Baker, S. M., Morrison, P. T., Warren, G., Smith, L. G., Lescoe,
M. K., Kane, M., Earabino, C., Lipford, J., Lindblom, A., and et al. (1994).
Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with
hereditary non-polyposis colon cancer. Nature 368, 258-261.

Brooks, P. C., Montgomery, A. M., Rosenfeld, M., Reisfeld, R. A., Hu, T., Klier,
G., and Cheresh, D. A. (1994). lntegrin alpha v beta 3 antagonists promote tumor

regression by inducing apoptosis of angiogenic blood vessels. Cell 79, 1157-
1 164.

Brooks, P. C., Stromblad, S., Sanders, L. 0., von Schalscha, T. L., Aimes, R. T.,
Stetler-Stevenson, W. G., Quigley, J. P., and Cheresh, D. A. (1996). Localization

86

of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction
with integrin alpha v beta 3. Cell 85, 683-693.

Brooks, P. C., Silletti, 8., von Schalscha, T. L., Friedlander, M., and Cheresh, D.
A. (1998). Disruption of angiogenesis by PEX, a noncatalytic metalloproteinase
fragment with integrin binding activity. Cell 92, 391-400.

Brown, L. F., Berse, B., Jackman, R. W., Tognazzi, K., Guidi, A. J., Dvorak, H. F.,
Senger, D. R., Connolly, J. L., and Schnitt, S. J. (1995). Expression of vascular
permeability factor (vascular endothelial growth factor) and its receptors in breast
cancer. Hum Pathol 26, 86-91.

Brown, M. 8., Dana, S. E., and Goldstein, J. L. (1974). Regulation of 3-hydroxy-
3-methylglutaryl coenzyme A reductase activity in cultured human fibroblasts.
Comparison of cells from a normal subject and from a patient with homozygous
familial hypercholesterolemia. J Biol Chem 249, 789-796.

Brown, M. S., and Goldstein, J. L. (1974). Familial hypercholesterolemia:
defective binding of lipoproteins to cultured fibroblasts associated with impaired
regulation of 3-hydroxy-3-methylglutaryl coenzyme A reductase activity. Proc
Natl Acad Sci U S A 71 , 788-792.

Brown, M. S., and Goldstein, J. L. (1986). A receptor-mediated pathway for
cholesterol homeostasis. Science 232, 34-47.

Brown, S. D., Twells, R. 0., Hey, P. J., Cox, R. D., Levy, E. R., Soderman, A. R.,
Metzker, M. L., Caskey, C. T., Todd, J. A., and Hess, J. F. (1998). Isolation and
characterization of LRP6, a novel member of the low density lipoprotein receptor
gene family. Biochem Biophys Res Commun 248, 879-888.

Bryan, T. M., Marusic, L., Bacchetti, S., Namba, M., and Reddel, R. R. (1997).
The telomere lengthening mechanism in telomerase-negative immortal human
cells does not involve the telomerase RNA subunit. Hum Mol Genet 6, 921-926.

Calvert, P. M., and Frucht, H. (2002). The genetics of colorectal cancer. Ann
Intern Med 137, 603-612.

Cavallaro, U., and Christofori, G. (2001). Cell adhesion in tumor invasion and
metastasis: loss of the glue is not enough. Biochim Biophys Acta 1552, 39-45.

Cech, T. R., Nakamura, T. M., and Lingner, J. (1997). Telomerase is a true
reverse transcriptase. A review. Biochemistry (Mosc) 62, 1202-1205.

Chambers, A. F., Groom, A. C., and MacDonald, l. C. (2002). Dissemination and
growth of cancer cells in metastatic sites. Nat Rev Cancer 2, 563-572.

Chang, E. H., Gonda, M. A., Ellis, R. W., Scolnick, E. M., and Lowy, D. R. (1982).
Human genome contains four genes homologous to transforming genes of

87

Harvey and Kirsten murine sarcoma viruses. Proc Natl Acad Sci U S A 79, 4848-
4852.

Chardin, P., Camonis, J. H., Gale, N. W., van Aelst, L., Schlessinger, J., Wigler,
M. H., and Bar-Sagi, D. (1993). Human Sos1: a guanine nucleotide exchange
factor for Ras that binds to GRB2. Science 260, 1338-1343.

Chen, L. M., Bailey, D., and Fernandez-Valle, C. (2000). Association of beta 1
integrin with focal adhesion kinase and paxillin in differentiating Schwann cells. J
Neurosci 20, 3776-3784.

Chen, W. J., Goldstein, J. L., and Brown, M. S. (1990). NPXY, a sequence often
found in cytoplasmic tails, is required for coated pit-mediated internalization of
the low density lipoprotein receptor. J Biol Chem 265, 3116-3123.

Chen, Y. Q., Hsieh, J. T., Yao, F., Fang, B., Pong, R. C., Cipriano, S. C., and
Krepulat, F. (1999). Induction of apoptosis and GZ/M cell cycle arrest by DCC.
Oncogene 18, 2747-2754.

Christensen, E. |., and Birn, H. (2002). Megalin and cubilin: multifunctional
endocytic receptors. Nat Rev Mol Cell Biol 3, 256-266.

Claros, M. G., Brunak, S., and von Heijne, G. (1997). Prediction of N-temlinal
protein sorting signals. Curr Opin Struct Biol 7, 394-398.

Cleary, M. L., Smith, S. D., and Sklar, J. (1986). Cloning and structural analysis
of cDNAs for bcl-Z and a hybrid bcl-2limmunoglobulin transcript resulting from
the t(14;18) translocation. Cell 47, 19-28.

Collawn, J. F., Stangel, M., Kuhn, L. A., Esekogwu, V., Jing, S. Q., Trowbridge, l.
S., and Tainer, J. A. (1990). Transferrin receptor internalization sequence YXRF
implicates a tight turn as the structural recognition motif for endocytosis. Cell 63,
1061 -1 072.

Collawn, J. F., Kuhn, L. A., Liu, L. F., Tainer, J. A., and Trowbridge, l. S. (1991).
Transplanted LDL and mannose-6-phosphate receptor internalization signals
promote high-efficiency endocytosis of the transferrin receptor. Embo J 10, 3247-
3253.

Conese, M., Nykjaer, A., Petersen, C. M., Cremona, O., Pardi, R., Andreasen, P.
A., Gliemann, J., Christensen, E. l., and Blasi, F. (1995). alpha-2 Macroglobulin
receptor/Ldl receptor-related protein(Lrp)-dependent internalization of the
urokinase receptor. J Cell Biol 131, 1609-1622.

Cory, S., and Adams, J. M. (2002). The Bcl2 family: regulators of the cellular life-
or-death switch. Nat Rev Cancer 2, 647-656.

88

Coussens, L. M., Hanahan, D., and Arbeit, J. M. (1996). Genetic predisposition
and parameters of malignant progression in K14-HPV16 transgenic mice. Am J
Pathol 149, 1899-1917.

Czekay, R. P., Kuemmel, T. A., Orlando, R. A., and Farquhar, M. G. (2001).
Direct binding of occupied urokinase receptor (uPAR) to LDL receptor-related
protein is required for endocytosis of uPAR and regulation of cell surface
urokinase activity. Mol Biol Cell 12, 1467-1479.

Daly, N. L., Djordjevic, J. T., Kroon, P. A., and Smith, R. (1995a). Three-
dimensional structure of the second cysteine-rich repeat from the human low-
density lipoprotein receptor. Biochemistry 34, 14474-14481.

Daly, N. L., Scanlon, M. J., Djordjevic, J. T., Kroon, P. A., and Smith, R. (1995b).
Three-dimensional structure of a cysteine-rich repeat from the low—density
lipoprotein receptor. Proc Natl Acad Sci U S A 92, 6334-6338.

Dameron, K. M., Volpert, O. V., Tainsky, M. A., and Bouck, N. (1994a). Control of
angiogenesis in fibroblasts by p53 regulation of thrombospondin-1. Science 265,
1582-1584.

Dameron, K. M., Volpert, O. V., Tainsky, M. A., and Bouck, N. (1994b). The p53
tumor suppressor gene inhibits angiogenesis by stimulating the production of
thrombospondin. Cold Spring Harb Symp Quant Biol 59, 483-489.

Dano, K., Romer, J., Nielsen, B. S., Bjorn, S., Pyke, C., Rygaard, J., and Lund, L.
R. (1999). Cancer invasion and tissue remodeling—cooperation of protease
systems and cell types. Apmis 107, 120-127.

D'Arcangelo, G., Miao, G. G., Chen, S. C., Scares, H. 0., Morgan, J. l., and
Curran, T. (1995). A protein related to extracellular matrix proteins deleted in the
mouse mutant reeler. Nature 374, 719-723.

D'Arcangelo, G., Homayouni, R., Keshvara, L., Rice, D. S., Sheldon, M., and
Curran, T. (1999). Reelin is a ligand for lipoprotein receptors. Neuron 24, 471-
479.

Datta, S. R., Dudek, H., Tao, X., Masters, 8., Fu, H., Gotoh, Y., and Greenberg,
M. E. (1997). Akt phosphorylation of BAD couples survival signals to the cell-
intrinsic death machinery. Cell 91, 231-241.

Davis, C. G., Lehrman, M. A., Russell, D. W., Anderson, R. G., Brown, M. S., and
Goldstein, J. L. (1986). The J.D. mutation in familial hypercholesterolemia: amino

acid substitution in cytoplasmic domain impedes internalization of LDL receptors.
Cell 45, 15-24.

89

Davis, C. G., van Driel, l. R., Russell, D. W., Brown, M. S., and Goldstein, J. L.
(1987). The low density lipoprotein receptor. Identification of amino acids in
cytoplasmic domain required for rapid endocytosis. J Biol Chem 262, 4075-4082.

de Fraipont, F., Nicholson, A. C., Feige, J. J., and Van Meir, E. G. (2001).
Thrombospondins and tumor angiogenesis. Trends Mol Med 7, 401-407.

de Vos, A. M., Tong, L., Milburn, M. V., Matias, P. M., Jancarik, J., Noguchi, S.,
Nishimura, S., Miura, K., Ohtsuka, E., and Kim, S. H. (1988). Three-dimensional
structure of an oncogene protein: catalytic domain of human c-H-ras p21.
Science 239, 888-893.

DeLeo, A. 3., Jay, 6., Appella, E., Dubois, G. 0., Law, L. W., and Old, L. J.
(1979). Detection of a transformation-related antigen in chemically induced
sarcomas and other transformed cells of the mouse. Proc Natl Acad Sci U S A
76, 2420-2424.

Der, C. J., Krontiris, T. G., and Cooper, G. M. (1982). Transforming genes of
human bladder and lung carcinoma cell lines are homologous to the ras genes of
Harvey and Kirsten sarcoma viruses. Proc Natl Acad Sci U S A 79, 3637-3640.

Deryugina, E. l., Bourdon, M. A., Luo, G. X., Reisfeld, R. A., and Strongin, A.
(1997). Matrix metalloproteinase-2 activation modulates glioma cell migration. J
Cell Sci 110 (Pt 19), 2473-2482.

Deryugina, E. l., Bourdon, M. A., Jungwirth, K., Smith, J. W., and Strongin, A. Y.
(2000). Functional activation of integrin alpha V beta 3 in tumor cells expressing
membrane-type 1 matrix metalloproteinase. Int J Cancer 86, 15-23.

Dix, D. (1989) The role of aging in cancer incidence: an epidemiological study. J.
Gerontolo. 44, 10-18.

Dolmer, K., Huang, W., and Gettins, P. G. (2000). NMR solution structure of
complement-like repeat CR3 from the low density lipoprotein receptor-related
protein. Evidence for specific binding to the receptor binding domain of human
alpha(2)-macroglobulin. J Biol Chem 275, 3264-3269.

Dong, Y., Lathrop, W., Weaver, 0., Qiu, Q., Cini, J., Bertolini, D., and Chen, D.
(1998). Molecular cloning and characterization of LR3, a novel LDL receptor
family protein with mitogenic activity. Biochem Biophys Res Commun 251, 784-
790.

Ducray, C., Pommier, J. P., Martins, L., Boussin, F. D., and Sabatier, L. (1999).
Telomere dynamics, end-to—end fusions and telomerase activation during the
human fibroblast immortalization process. Oncogene 18, 4211-4223.

Dunham, I. (2002). Lessons from the sequence of human chromosome 22. Ernst
Schering Res Found Workshop, 31-50.

90

Egeblad, M. and Werb, Z. (2002). New functions for the matrix
metalloproteinases in cancer progression. Nat Rev Cancer 2, 161-174.

Eliyahu, D., Michalovitz, D., Eliyahu, S., Pinhasi-Kimhi, O., and Oren, M. (1989).
Wild-type p53 can inhibit oncogene-mediated focus formation. Proc Natl Acad
Sci U S A 86, 8763-8767.

Esser, V., Limbird, L. E., Brown, M. S., Goldstein, J. L., and Russell, D. W.
(1988). Mutational analysis of the ligand binding domain of the low density
lipoprotein receptor. J Biol Chem 263, 13282-13290.

Ewing, J. (1928) In: Neoplastic Diseases. A Treatise on Tumors, W. B. Saunders
Co., Philadelphia & London, 77-89.

Facchini, L. M. and Penn, L. 2. (1998). The molecular role of Myc in growth and
transformation: recent discoveries lead to new insights. Faseb J 12, 633-651.

Fass, D., Blacklow, S., Kim, P. S., and Berger, J. M. (1997). Molecular basis of
familial hypercholesterolaemia from structure of LDL receptor module. Nature
388, 691 -693.

Fearon, E. R., and Vogelstein, B. (1990). A genetic model for colorectal
tumorigenesis. Cell 61, 759-767.

Fearon, E. R., and Jones, P. A. (1992). Progressing toward a molecular
description of colorectal cancer development. Faseb J 6, 2783-2790.

Feinstein, E., Kimchi, A., Wallach, D., Boldin, M., and Varfolomeev, E. (1995).
The death domain: a module shared by proteins with diverse cellular functions.
Trends Biochem Sci 20, 342-344.

Felding-Habennann, B., Mueller, B. M., Romerdahl, C. A., and Cheresh, D. A.
(1992). Involvement of integrin alpha V gene expression in human melanoma
tumorigenicity. J Clin Invest 89, 2018-2022.

Finch, R. A., Donoviel, D. B., Potter, 0., Shi, M., Fan, A., Freed, D. D., Wang, C.
Y., Zambrowicz, B. P., Ramirez-Solis, R., Sands, A. T., and Zhang, N. (2002).
mdmx is a negative regulator of p53 activity in vivo. Cancer Res 62, 3221-3225.

Finlay, C. A., Hinds, P. W., and Levine, A. J. (1989). The p53 proto-oncogene
can act as a suppressor of transformation. Cell 57, 1083-1093.

Finucane, D. M., Bossy-Wetzel, E., Waterhouse, N. J., Cotter, T. G., and Green,
D. R. (1999). Bax-induced caspase activation and apoptosis via cytochrome c
release from mitochondria is inhibitable by Bcl-xL. J Biol Chem 274, 2225-2233.

Fishel, R., Lescoe, M. K., Rao, M. R., Copeland, N. G., Jenkins, N. A., Garber, J.,
Kane, M., and Kolodner, R. (1993). The human mutator gene homolog MSHZ

91

and its association with hereditary nonpolyposis colon cancer. Cell 75, 1027-
1038.

Friend, 8. H., Bernards, R., Rogelj, S., Weinberg, R. A., Rapaport, J. M., Albert,
D. M., and Dryja, T. P. (1986). A human DNA segment with properties of the
gene that predisposes to retinoblastoma and osteosarcoma. Nature 323, 643-
646.

Fry, D. G., Milam, L. D., Maher, V. M., and McCormick, J. J. (1986).
Transformation of diploid human fibroblasts by DNA transfection with the v-sis
oncogene. J Cell Physiol 128, 313-321.

Fung, Y. K., Murphree, A. L., T'Ang, A., Qian, J., Hinrichs, S. H., and Benedict,
W. F. (1987). Structural evidence for the authenticity of the human
retinoblastoma gene. Science 236, 1657-1661.

Gardner, E. J. (1951) A genetic and clinical study of intestinal polyposis, a
predisposing factor for carcinoma of the colon and rectum. Am. J. Hum. Genet.
3: 167-176.

Gardner, E. J. (1972) Discovery of the Gardner syndrome. Birth Defects Orig.
Art. Ser. VIII: 48-51.

Gibbs, J. B., Sigal, l. S., Poe, M., and Scolnick, E. M. (1984). Intrinsic GTPase
activity distinguishes normal and oncogenic ras p21 molecules. Proc Natl Acad
Sci U S A 81, 5704—5708.

Gilbert, SF. (1997) Developmental biology. 5th ed. Sunderland, MA: Sinauer
Associates, Inc.

Gimbrone, M. A., Jr., Leapman, S. B., Cotran, R. S., and Folkman, J. (1972).
Tumor dormancy in vivo by prevention of neovascularization. J Exp Med 136,
261-276.

Gong, Y., Vikkula, M., Boon, L., Liu, J., Beighton, P., Ramesar, R., Peltonen, L.,
Somer, H., Hirose, T., Dallapiccola, B., et al. (1996). Osteoporosis-pseudoglioma
syndrome, a disorder affecting skeletal strength and vision, is assigned to
chromosome region 11q12-13. Am J Hum Genet 59, 146-151.

Gong, Y., Slee, R. B., Fukai, N., Rawadi, G., Roman-Roman, S., Reginato, A. M.,
Wang, H., Cundy, T., Glorieux, F. H., Lev, D., et al. (2001). LDL receptor-related
protein 5 (LRP5) affects bone accrual and eye development. Cell 107, 513-523.

Gotthardt, M., Trommsdorff, M., Nevitt, M. F., Shelton, J., Richardson, J. A.,
Stockinger, W., Nimpf, J., and Herz, J. (2000). Interactions of the low density
lipoprotein receptor gene family with cytosolic adaptor and scaffold proteins
suggest diverse biological functions in cellular communication and signal
transduction. J Biol Chem 275, 25616-25624.

92

Grandori, C., Cowley, S. M., James, L. P., and Eisenman, R. N. (2000). The
Myc/Max/Mad network and the transcriptional control of cell behavior. Annu Rev
Cell Dev Biol 16, 653-699.

Graziano, F., Cascinu, S., Staccioli, M. P., Catalano, V., Rossi, M. C., Baldelli, A.
M., Giordani, P., Muretto, P., and Catalano, G. (2001). Potential role and
chronology of abnormal expression of the Deleted in Colon Cancer (DCC) and
the p53 proteins in the development of gastric cancer. BMC Cancer 1, 9.

Greider, C. W., and Blackburn, E. H. (1985). Identification of a specific telomere
terminal transferase activity in Tetrahymena extracts. Cell 43, 405-413.

Greider, C. W., and Blackburn, E. H. (1989). A telomeric sequence in the RNA of
Tetrahymena telomerase required for telomere repeat synthesis. Nature 337,
331-337.

Groden, J., Thliveris, A., Samowitz, W., Carlson, M., Gelbert, L., Albertsen, H.,
Joslyn, G., Stevens, J., Spirio, L., Robertson, M., and et al. (1991). Identification
and characterization of the familial adenomatous polyposis coli gene. Cell 66,
589-600.

Guidi, A. J., Fischer, L., Harris, J. R., and Schnitt, S. J. (1994). Microvessel
density and distribution in ductal carcinoma in situ of the breast. J Natl Cancer
Inst 86, 614-619.

Guidi, A. J., Abu-Jawdeh, G., Berse, B., Jackman, R. W., Tognazzi, K., Dvorak,
H. F., and Brown, L. F. (1995). Vascular permeability factor (vascular endothelial

growth factor) expression and angiogenesis in cervical neoplasia. J Natl Cancer
Inst 87, 1237-1245.

Hackett, J. A., and Greider, C. W. (2002). Balancing instability: dual roles for
telomerase and telomere dysfunction in tumorigenesis. Oncogene 21, 619-626.

Hahn, S. A., Schutte, M., l-loque, A. T., Moskaluk, C. A., da Costa, L. T.,
Rozenblum, E., Weinstein, C. L., Fischer, A., Yeo, C. J., Hruban, R. H., and
Kern, S. E. (1996). DPC4, a candidate tumor suppressor gene at human
chromosome 18q21.1. Science 271, 350-353.

Hahn, W. C., and Meyerson, M. (2001). Telomerase activation, cellular
immortalization and cancer. Ann Med 33, 123-129.

Hanahan, D. (1985). Heritable formation of pancreatic beta-cell tumours in
transgenic mice expressing recombinant insulin/simian virus 40 oncogenes.
Nature 315, 115-122.

Hanahan, D., and Folkman, J. (1996). Patterns and emerging mechanisms of the
angiogenic switch during tumorigenesis. Cell 86, 353-364.

93

Hanahan, D., and Weinberg, R. A. (2000). The hallmarks of cancer. Cell 100, 57-
70.

Hancock, J. F., Paterson, H., and Marshall, C. J. (1990). A polybasic domain or
palmitoylation is required in addition to the CAAX motif to localize p21ras to the
plasma membrane. Cell 63, 133-139.

Harley, C. B., Futcher, A. B., and Greider, C. W. (1990). Telomeres shorten
during ageing of human fibroblasts. Nature 345, 458-460.

Haupt, Y., Maya, R., Kazaz, A., and Oren, M. (1997). Mdm2 promotes the rapid
degradation of p53. Nature 387, 296-299.

Hayflick, L., and Moorhead, P. (1961). The serial cultivation of human diploid cell
strains. Exp. Cell Res. 25, 585.

Hayflick, L. (2000). The illusion of cell immortality. Br J Cancer 83, 841-846.

Heldin, C. H. (1995). Dimerization of cell surface receptors in signal transduction.
Cell 80, 213-223.

Herold, S., Wanzel, M., Beuger, V., Frohme, C., Beul, D., Hillukkala, T., Syvaoja,
J., Saluz, H. P., Haenel, F., and Eilers, M. (2002). Negative regulation of the
mammalian UV response by myc through association with Miz-1. Mol Cell 10,
509-521.

Herz, J., Hamann, U., Rogne, S., Myklebost, O., Gausepohl, H., and Stanley, K.
K. (1988). Surface location and high affinity for calcium of a 500-kd liver
membrane protein closely related to the LDL-receptor suggest a physiological
role as lipoprotein receptor. Embo J 7, 4119-4127.

Herz, J., and Strickland, D. K. (2001). LRP: a multifunctional scavenger and
signaling receptor. J Clin Invest 108, 779-784.

Herz, J., and Bock, H. H. (2002). Lipoprotein receptors in the nervous system.
Annu Rev Biochem 71 , 405-434.

Hey, P. J., Twells, R. 0., Phillips, M. S., Yusuke, N., Brown, S. D., Kawaguchi,
Y., Cox, R., Guochun, X., Dugan, V., Hammond, H., etal. (1998). Cloning of a
novel member of the low-density lipoprotein receptor family. Gene 216, 103—111.

Hiesberger, T., Trommsdorff, M., Howell, B. W., Goffinet, A., Mumby, M. C.,
Cooper, J. A., and Herz, J. (1999). Direct binding of Reelin to VLDL receptor and
ApoE receptor 2 induces tyrosine phosphorylation of disabled-1 and modulates
tau phosphorylation. Neuron 24, 481-489.

94

Hildebrand, J. D., Schaller, M. D., and Parsons, J. T. (1993). Identification of
sequences required for the efficient localization of the focal adhesion kinase,
pp125FAK, to cellular focal adhesions. J Cell Biol 123, 993-1005.

Hilliker, C., Van Leuven, F., and Van den Berghe, H. (1992). Assignment of the
gene coding for the alpha Z-macroglobulin receptor to mouse chromosome 15
and to human chromosome 12q13-q14 by isotopic and nonisotopic in situ
hybridization. Genomics 13, 472-474.

Hinds, P, Finlay, C. and Levine, A.J. (1989) Mutation is required to activate the
p53 gene for cooperation with the ras oncogene and transformation. J. Virol. 63,
739-746.

Hofmann, K., and Tschopp, J. (1995). The death domain motif found in Fas (Apo-
1) and TNF receptor is present in proteins involved in apoptosis and axonal
guidance. FEBS Lett 371, 321-323.

Hofmann, U. B., Westphal, J. R., Van Kraats, A. A., Ruiter, D. J., and Van
Muijen, G. N. (2000). Expression of integrin alpha(v)beta(3) correlates with
activation of membrane-type matrix metalloproteinase-1 (MT1-MMP) and matrix
metalloproteinase-2 (MMP-2) in human melanoma cells in vitro and in vivo. Int J
Cancer 87, 12-19.

Hollstein, M., Rice, K., Greenblatt, M. S., Soussi, T., Fuchs, R., Sorlie, T., Hovig,
E., Smith-Sorensen, B., Montesano, R., and Harris, C. C. (1994). Database of

p53 gene somatic mutations in human tumors and cell lines. Nucleic Acids Res
22, 3551-3555.

Holmgren, L., O'Reilly, M. S., and Folkman, J. (1995). Dorrnancy of
micrometastases: balanced proliferation and apoptosis in the presence of
angiogenesis suppression. Nat Med 1, 149-153.

Hood, J. D. and Cheresh, D. A. (2002). Role of integrins in cell invasion and
migration. Nat Rev Cancer 2, 91-100.

Horn, l. R., van den Berg, B. M., van der Meijden, P. Z., Pannekoek, H., and van
Zonneveld, A. J. (1997). Molecular analysis of ligand binding to the second
cluster of complement-type repeats of the low density lipoprotein receptor-related
protein. Evidence for an allosteric component in receptor-associated protein-
mediated inhibition of ligand binding. J Biol Chem 272, 13608-13613.

Horwitz, A., Duggan, K., Buck, C., Beckerle, M. C., and Burridge, K. (1986).
Interaction of plasma membrane fibronectin receptor with talin—a transmembrane
linkage. Nature 320, 531-533.

Howell, B. W., Gertler, F. B., and Cooper, J. A. (1997). Mouse disabled (mDab1):
a Src binding protein implicated in neuronal development. Embo J 16, 121-132.

95

Howell, B. W., and Herz, J. (2001). The LDL receptor gene family: signaling
functions during development. Curr Opin Neurobiol 11, 74-81.

Huang, W., Dolmer, K., and Gettins, P. G. (1999). NMR solution structure of
complement-like repeat CR8 from the low density lipoprotein receptor-related
protein. J Biol Chem 274, 14130-14136.

Hurlin, P. J., Maher, V. M., and McCormick, J. J. (1989). Malignant
transformation of human fibroblasts caused by expression of a transfected T24
HRAS oncogene. Proc Natl Acad Sci U S A 86, 187-191.

Hurlin, P. J., Foley, K. P., Ayer, D. E., Eisenman, R. N., Hanahan, D., and Arbeit,
J. M. (1995). Regulation of Myc and Mad during epidermal differentiation and
HPV-associated tumorigenesis. Oncogene 11, 2487-2501.

Hussain, M. M., Strickland, D. K., and Bakillah, A. (1999). The mammalian low-
density lipoprotein receptor family. Annu Rev Nutr 19, 141-172.

Huttenlocher, A., Ginsberg, M. H., and Horwitz, A. F. (1996). Modulation of cell
migration by integrin-mediated cytoskeletal linkages and ligand-binding affinity. J
Cell Biol 134, 1551-1562.

Hynes, R. (2002). lntegrins: bidirectional, allosteric signaling machines. Cell 110,
673-687.

Igney, F. H., and Krammer, P. H. (2002). Immune escape of tumors: apoptosis
resistance and tumor counterattack. J Leukoc Biol 71 , 907-920.

Ishii, H., Kim, D. H., Fujita, T., Endo, Y., Saeki, 8., and Yamamoto, T. T. (1998).
cDNA cloning of a new low-density lipoprotein receptor-related protein and
mapping of its gene (LRP3) to chromosome bands 19q12-q13. 2. Genomics 51 ,
132-135.

Jacobsen, L., Madsen, P., Moestrup, S. K., Lund, A. H., Tommerup, N., Nykjaer,
A., Sottrup-Jensen, L., Gliemann, J., and Petersen, C. M. (1996). Molecular
characterization of a novel human hybrid-type receptor that binds the alphaZ-
macroglobulin receptor-associated protein. J Biol Chem 271, 31379-31383.

Jadot, M., Canfield, W. M., Gregory, W., and Kornfeld, S. (1992).
Characterization of the signal for rapid internalization of the bovine mannose 6-
phcsphate/insulin-like growth factor-ll receptor. J Biol Chem 267, 11069-11077.

Jen, J., Powell, S. M., Papadopoulos, N., Smith, K. J., Hamilton, S. R.,
Vogelstein, B., and Kinzler, K. W. (1994). Molecular determinants of dysplasia in
colorectal lesions. Cancer Res 54, 5523-5526.

96

Jeon, H., Meng, W., Takagi, J., Eck, M. J., Springer, T. A., and Blacklow, S. C.
(2001 ). Implications for familial hypercholesterolemia from the structure of the
LDL receptor YWTD-EGF domain pair. Nat Struct Biol 8, 499-504.

Jimenez, M. D., Guerrero, M., Marenco, J. L., Sanchez-Burson, J., and Garcia-
Bragado, F. (1996). Antiphospholipid antibodies and movement disorders. Rev
Rhum Engl Ed 63, 227-228.

Jing, S. 0., Spencer, T., Miller, K., Hopkins, 0., and Trowbridge, l. S. (1990).
Role of the human transferrin receptor cytoplasmic domain in endocytosis:
localization of a specific signal sequence for internalization. J Cell Biol 110, 283-
294.

Juliano, R. L. (2002). Signal transduction by cell adhesion receptors and the
cytoskeleton: functions of integrins, cadherins, selectins, and immunoglobulin-
superfamily members. Annu Rev Pharmacol Toxicol 42, 283-323.

Jurgensmeier, J. M., Xie, Z., Deveraux, Q., Ellerby, L., Bredesen, D., and Reed,
J. C. (1998). Bax directly induces release of cytochrome c from isolated
mitochondria. Proc Natl Acad Sci U S A 95, 4997-5002.

Kandel, J., Bossy-Wetzel, E., Radvanyi, F., Klagsbrun, M., Folkman, J., and
Hanahan, D. (1991). Neovascularization is associated with a switch to the export
of bFGF in the multistep development of fibrosarcoma. Cell 66, 1095-1104.

Kang, M. K., and Park, N. H. (2001). Conversion of normal to malignant
phenotype: telomere shortening, telomerase activation, and genomic instability
during immortalization of human oral keratinocytes. Crit Rev Oral Biol Med 12,
38—54.

Kato, G. J., Barrett, J., Villa-Garcia, M., and Dang, C. V. (1990). An amino-
terrninal c-myc domain required for neoplastic transformation activates
transcription. Mol Cell Biol 10, 5914-5920.

Kato, G. J., Lee, W. M., Chen, L. L., and Dang, C. V. (1992). Max: functional
domains and interaction with c-Myc. Genes Dev 6, 81-92.

Kato, M., Patel, M. S., Levasseur, R., Lobov, l., Chang, B. H., Glass, D. A., 2nd,
Hartmann, C., Li, L., Hwang, T. H., Brayton, C. F., etal. (2002). bea1-
independent decrease in osteoblast proliferation, osteopenia, and persistent
embryonic eye vascularization in mice deficient in Lrp5, a Wnt coreceptor. J Cell
Biol 157, 303-314.

Kelekar, A., Chang, B. S., Harlan, J. E., Fesik, S. W., and Thompson, C. B.

(1997). Bad is a BH3 domain-containing protein that forms an inactivating dimer
with Bcl-XL. Mol Cell Biol 17, 7040-7046.

97

Kim, D. H., Iijima, H., Goto, K., Sakai, J., Ishii, H., Kim, H. J., Suzuki, H., Kondo,
H., Saeki, S., and Yamamoto, T. (1996). Human apolipoprotein E receptor 2. A
novel lipoprotein receptor of the low density lipoprotein receptor family
predominantly expressed in brain. J Biol Chem 271, 8373-8380.

Kim, D. H., Magoori, K., lnoue, T. R., Mao, C. C., Kim, H. J., Suzuki, H., Fujita,
T., Endo, Y., Saeki, S., and Yamamoto, T. T. (1997). Exonflntron organization,
chromosome localization, alternative splicing, and transcription units of the
human apolipoprotein E receptor 2 gene. J Biol Chem 272, 8498-8504.

Kim, H., Farris, J., Christman, S. A., Kong, B. W., Foster, L. K., O’Grady, S. M.,
and Foster, D. N. (2002). Events in the immortalizing process of primary human

mammary epithelial cells by the catalytic subunit of human telomerase. Biochem
J 365, 765-772.

Kim, N. W., Piatyszek, M. A., Prowse, K. R., Harley, C. B., West, M. 0., Ho, P. L.,
Coviello, G. M., Wright, W. E., Weinrich, S. L., and Shay, J. W. (1994). Specific

association of human telomerase activity with immortal cells and cancer. Science
266, 2011-2015.

Kinzler, K. W., and Vogelstein, B. (1996). Lessons from hereditary colorectal
cancer. Cell 87, 159-170.

Kischkel, F. C., Hellbardt, S., Behrrnann, l., Gerrner, M., Pawlita, M., Krammer,
P. H., and Peter, M. E. (1995). Cytotoxicity-dependent APO-1 (Fas/0095)-
associated proteins form a death-inducing signaling complex (DISC) with the
receptor. Embo J 14, 5579-5588.

Kishida, S., Yamamoto, H., Ikeda, S., Kishida, M., Sakamoto, I., Koyama, S., and
Kikuchi, A. (1998). Axin, a negative regulator of the wnt signaling pathway,
directly interacts with adenomatous polyposis coli and regulates the stabilization
of beta-catenin. J Biol Chem 273, 10823-10826.

Klemke, R. L., Cal, 8., Giannini, A. L., Gallagher, P. J., de Lanerolle, P., and
Cheresh, D. A. (1997). Regulation of cell motility by mitogen-activated protein
kinase. J Cell Biol 137, 481-492.

Klingelhutz, A. J. (1999). The roles of telomeres and telomerase in cellular
immortalization and the development of cancer. Anticancer Res 19, 4823-4830.

Knezevic, l., Leisner, T. M., and Lam, S. C. (1996). Direct binding of the platelet
integrin alphallbbeta3 (GPIIb-Illa) to talin. Evidence that interaction is mediated
through the cytoplasmic domains of both alphallb and beta3. J Biol Chem 271,
16416-16421.

Knudson, A. 6., Jr. (1971). Mutation and cancer: statistical study of
retinoblastoma. Proc Natl Acad Sci U S A 68, 820-823.

98

Korsmeyer, S. J., Wei, M. C., Saito, M., Weller, 8., Oh, K. J., and Schlesinger, P.
H. (2000). Pro-apoptotic cascade activates BID, which oligomerizes BAK or BAX
into pores that result in the release of cytochrome 0. Cell Death Differ 7, 1166-

1 173.

Kounnas, M. Z., Henkin, J., Argraves, W. S., and Strickland, D. K. (1993). Low
density lipoprotein receptor-related protein/alpha Z-macroglobulin receptor
mediates cellular uptake of pro-urokinase. J Biol Chem 268, 21862-21867.

Krengel, U., Schlichting, L., Scherer, A., Schumann, R., Frech, M., John, J.,
Kabsch, W., Pai, E. F., and Wittinghofer, A. (1990). Three-dimensional structures
of H-ras p21 mutants: molecular basis for their inability to function as signal
switch molecules. Cell 62, 539-548.

Krieger, M., and Herz, J. (1994). Structures and functions of multiligand
lipoprotein receptors: macrophage scavenger receptors and LDL receptor-related
protein (LRP). Annu Rev Biochem 63, 601-637.

Kubbutat, M. H., Jones, S. N., and Vousden, K. H. (1997). Regulation of p53
stability by MdmZ. Nature 387, 299-303.

Kyo, S., Takakura, M., Taira, T., Kanaya, T., Itoh, H., Yutsudo, M., Ariga, H., and
Inoue, M. (2000). Sp1 cooperates with c-Myc to activate transcription of the
human telomerase reverse transcriptase gene (hTERT). Nucleic Acids Res 28,
669-677.

Lacey, M., Alpert, S., and Hanahan, D. (1986). Bovine papillomavirus genome
elicits skin tumours in transgenic mice. Nature 322, 609-612.

Lane, D. P., and Crawford, L. V. (1979). T antigen is bound to a host protein in
SV40-transformed cells. Nature 278, 261-263.

Langbein, S., Szakacs, O., Wilhelm, M., Sukosd, F., Weber, 8., Jauch, A., Lopez
Beltran, A., Alken, P., Kalble, T., and Kovacs, G. (2002). Alteration of the LRP1B
gene region is associated with high grade of urothelial cancer. Lab Invest 82,
639-643.

Law, D. A., Nannizzi-Alaimo, L., and Phillips, D. R. (1996). Outside-in integrin
signal transduction. Alpha llb beta 3-(GP llb Illa) tyrosine phosphorylation
induced by platelet aggregation. J Biol Chem 271, 10811-10815.

Leach, F. S., Nicolaides, N. C., Papadopoulos, M, Liu, B., Jen, J., Parsons, R.,
Peltomaki, P., Sistonen, P., Aaltonen, L. A., Nystrom-Lahti, M., and et al. (1993).
Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell
75, 1215-1225.

99

Lee, W. H., Bookstein, R., Hong, F., Young, L. J., Shew, J. Y., and Lee, E. Y.
(1987). Human retinoblastoma susceptibility gene: cloning, identification, and
sequence. Science 235, 1394-1399.

Levine, A. J. (1997). p53, the cellular gatekeeper for growth and division. Cell 88,
323-331.

Li, R, Nijhawan, D., Budihardjo, l., Srinivasula, S. M., Ahmad, M., Alnemri, E. S.,
and Wang, X. (1997). Cytochrome c and dATP-dependent formation of Apaf-
1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91 , 479-489.

Li, Y., Marzolo, M. P., van Kerkhof, P., Strous, G. J., and Bu, G. (2000). The
YXXL motif, but not the two NPXY motifs, serves as the dominant endocytosis
signal for low density lipoprotein receptor-related protein. J Biol Chem 275,
17187-17194.

Li, Y., Cam, J., and Bu, G. (20013). Low-density lipoprotein receptor family:
endocytosis and signal transduction. Mol Neurobiol 23, 53-67.

Li Y., van Kerkhof P., Marzolo M.P., Strous G.J., and Bu G. (2001b).
Identification of a major cyclic AMP-dependent protein kinase A phosphorylation
site within the cytoplasmic tail of the low-density lipoprotein receptor-related
protein: implication for receptor-mediated endocytosis. Mol Cell Biol. 21, 1185-
95.

Liliental, J., and Chang, D. D. (1998). Rack1, a receptor for activated protein
kinase C, interacts with integrin beta subunit. J Biol Chem 273, 2379-2383.

Lin, C., Maher, V. M., and McCormick, J. J. (1995). Malignant transformation of
human fibroblast strain MSU-1.1 by v-fes requires an additional genetic change.
Int J Cancer 63, 140-147.

Lindgren, V., Luskey, K. L., Russell, D. W., and Francke, U. (1985). Human
genes involved in cholesterol metabolism: chromosomal mapping of the loci for
the low density lipoprotein receptor and 3-hydroxy-3-methylglutaryI-coenzyme A
reductase with cDNA probes. Proc Natl Acad Sci U S A 82, 8567-8571.

Lingner, J., Hughes, T. R., Shevchenko, A., Mann, M., Lundblad, V., and Cech,
T. R. (1997). Reverse transcriptase motifs in the catalytic subunit of telomerase.
Science 276, 561-567.

Linzer, D. l., and Levine, A. J. (1979). Characterization of a 54K dalton cellular
SV40 tumor antigen present in SV40-transformed cells and uninfected embryonal
carcinoma cells. Cell 17, 43-52.

Lisitsyn, N. A. (1995). Representational difference analysis: finding the
differences between genomes. Trends Genet 11, 303-307.

100

Little, R. D., Carulli, J. P., Del Mastro, R. G., Dupuis, J., Osborne, M., Folz, C.,
Manning, S. P., Swain, P. M., Zhao, S. C., Eustace, B., et al. (2002). A mutation
in the LDL receptor-related protein 5 gene results in the autosomal dominant
high-bone-mass trait. Am J Hum Genet 70, 11-19.

Liu, C. X., Musco, S., Lisitsina, N. M., Forgacs, E., Minna, J. D., and Lisitsyn, N.
A. (2000a). LRP-DIT, a putative endocytic receptor gene, is frequently inactivated
in non-small cell lung cancer cell lines. Cancer Res 60, 1961-1967.

Liu, C. X., Musco, S., Lisitsina, N. M., Yaklichkin, S. Y., and Lisitsyn, N. A.
(2000b). Genomic organization of a new candidate tumor suppressor gene,
LRP1B. Genomics 69, 271-274.

Liu, C. X., Li, Y., Obermoeller-McCormick, L. M., Schwartz, A. L., and Bu, G.
(2001). The putative tumor suppressor LRP1 B, a novel member of the low
density lipoprotein (LDL) receptor family, exhibits both overlapping and distinct
properties with the LDL receptor-related protein. J Biol Chem 276, 28889-28896.

Liu, S., Thomas, S. M., Woodside, D. 6., Rose, D. M., Kiosses, W. B., Pfaff, M.,
and Ginsberg, M. H. (1999). Binding of paxillin to alpha4 integrins modifies
integrin-dependent biological responses. Nature 402, 676-681.

Liu, S., and Ginsberg, M. H. (2000). Paxillin binding to a conserved sequence
motif in the alpha 4 integrin cytoplasmic domain. J Biol Chem 275, 22736-22742.

Lobel, P., Fujimoto, K., Ye, R. D., Griffiths, G., and Kornfeld, S. (1989). Mutations
in the cytoplasmic domain of the 275 kd mannose 6-phosphate receptor
differentially alter lysosomal enzyme sorting and endocytosis. Cell 57, 787-796.

Loukinova, E., Ranganathan, S., Kuznetsov, S., Gorlatova, N., Migliorini, M. M.,
Loukinov, D., Ulery, P. G., Mikhailenko, l., Lawrence, D. A., and Strickland, D. K.
(2002). Platelet-derived growth factor (PDGF)-induced tyrosine phosphorylation
of the low density lipoprotein receptor-related protein (LRP). Evidence for
integrated co-receptor function betwenn LRP and the PDGF. J Biol Chem 277,
15499-15506.

Lowenstein, E. J., Daly, R. J., Batzer, A. G., Li, W., Margolis, B., Lammers, R.,
Ullrich, A., Skolnik, E. Y., Bar-Sagi, D., and Schlessinger, J. (1992). The SH2 and
SH3 domain-containing protein GRBZ links receptor tyrosine kinases to ras
signaling. Cell 70, 431-442.

Lutz, C., Nimpf, J., Jenny, M., Boecklinger, K., Enzinger, C., Utermann, G., Baier-
Bitterlich, G., and Baier, G. (2002). Evidence of functional modulation of the
MEKK/JNK/cJun signaling cascade by the low density lipoprotein receptor-
related protein (LRP). J Biol Chem 277, 43143-43151.

Lutz, W., Leon, J., and Eilers, M. (2002). Contributions of Myc to tumorigenesis.
Biochim Biophys Acta 1602, 61-71.

101

Lynch, H. T., Smyrk, T. 0., Watson, P., Lanspa, S. J., Lynch, J. F., Lynch, P. M.,
Cavalleri, R. J., and Boland, C. R. (1993). Genetics, natural history, tumor
spectrum, and pathology of hereditary nonpolyposis colorectal cancer: an
updated review. Gastroenterology 104, 1535-1549.

Lynch, H. T., and Lynch, J. F. (2000). Hereditary nonpolyposis colorectal cancer.
Semin Surg Oncol 18, 305-313.

Macaluso, M., Russo, G., Cinti, C., Bazan, V., Gebbia, N., and Russo, A. (2002).
Ras family genes: an interesting link between cell cycle and cancer. J Cell
Physiol 192, 125-130.

Malby, 6., Pickering, R., Saha, S., Smallridge, R., Linse, S., and Downing, A. K.
(2001). The first epidermal growth factor-like domain of the low-density

lipoprotein receptor contains a noncanonical calcium binding site. Biochemistry
40, 2555-2563.

Mac, J., Wang, J., Liu, B., Pan, W., Farr, G. H., 3rd, Flynn, C., Yuan, H., Takada,
S., Kimelman, 0., Li, L., and Wu, D. (2001). Low-density lipoprotein receptor-
related protein-5 binds to Axin and regulates the canonical Wnt signaling
pathway. Mol Cell 7, 801-809.

Markowitz, S. D., Dawson, D. M., Willis, J., and Willson, J. K. (2002). Focus on
colon cancer. Cancer Cell 1, 233-236.

Marsh, 0., and Zori, R. (2002). Genetic insights into familial cancers- update and
recent discoveries. Cancer Lett 181, 125-164.

Matuoka, K., Shibasaki, F., Shibata, M., and Takenawa, T. (1993). Ash/Grb-Z, a
SH2ISH3-containing protein, couples to signaling for mitogenesis and
cytoskeletal reorganization by EGF and PDGF. Embo J 12, 3467-3473.

McMahon, S. B., Wood, M. A., and Cole, M. D. (2000). The essential cofactor
TRRAP recruits the histone acetyltransferase hGCN5 to c-Myc. Mol Cell Biol 20,
556-562.

Mehlen, P., Rabizadeh, S., Snipas, S. J., Assa-Munt, N., Salvesen, G. S., and
Bredesen, D. E. (1998). The DCC gene product induces apoptosis by a
mechanism requiring receptor proteolysis. Nature 395, 801-804.

Meijerink, J. P., Mensink, E. J., Wang, K., Sedlak, T. W., Sloetjes, A. W., de
Witte, T., Waksman, G., and Korsmeyer, S. J. (1998). Hematopoietic
malignancies demonstrate Ioss-of-function mutations of BAX. Blood 91, 2991-
2997.

Midgley, C. A., and Lane, D. P. (1997). p53 protein stability in tumour cells is not

determined by mutation but is dependent on MdmZ binding. Oncogene 15, 1179-
1 189.

102

Mikhailenko, I., Battey, F. D., Migliorini, M., Ruiz, J. F., Argraves, K., Moayeri, M.,
and Strickland, D. K. (2001). Recognition of alpha Z-macroglobulin by the low
density lipoprotein receptor-related protein requires the cooperation of two ligand
binding cluster regions. J Biol Chem 276, 39484-39491.

Miyaki, M., Konishi, M., Tanaka, K., Kikuchi-Yanoshita, R., Muraoka, M., Yasuno,
M., lgari, T., Koike, M., Chiba, M., and Mori, T. (1997). Gerrnline mutation of
MSH6 as the cause of hereditary nonpolyposis colorectal cancer. Nat Genet 17,
271-272.

Mizejewski, G. J. (1999). Role of integrins in cancer: survey of expression
patterns. Proc Soc Exp Biol Med 222, 124-138.

Moestrup, S. K., Holtet, T. L., Etzerodt, M., Thogersen, H. C., Nykjaer, A.,
Andreasen, P. A., Rasmussen, H. H., Sottrup—Jensen, L., and Gliemann, J.
(1993). Alpha 2-macroglobulin-proteinase complexes, plasminogen activator
inhibitor type-1-plasminogen activator complexes, and receptor-associated
protein bind to a region of the alpha 2-macroglobulin receptor containing a cluster
of eight complement-type repeats. J Biol Chem 268, 13691-13696.

Molenaar, J. J., Gerard, B., Chambon-Pautas, C., Cave, H., Duval, M., Vilmer,
E., and Grandchamp, B. (1998). Microsatellite instability and frameshift mutations
in BAX and transforming growth factor-beta RII genes are very uncommon in
acute lymphoblastic leukemia in vivo but not in cell lines. Blood 92, 230-233.

Monteagudo, C., Merino, M. J., San-Juan, J., Liotta, L. A., and Stetler-Stevenson,
W. G. (1990). lmmunohistochemical distribution of type IV collagenase in normal,
benign, and malignant breast tissue. Am J Pathol 136, 585-592.

Moon, R. T., and Miller, J. R. (1997). The APC tumor suppressor protein in
development and cancer. Trends Genet 13, 256-258.

Morgan, T. L., Yang, D. J., Fry, D. G., Hurlin, P. J., Kohler, S. K., Maher, V. M.,
and McCormick, J. J. (1991). Characteristics of an infinite life span diploid human
fibroblast cell strain and a near-diploid strain arising from a clone of cells
expressing a transfected v-myc oncogene. Exp Cell Res 197, 125-136.

Morin, P. J., Sparks, A. B., Korinek, V., Barker, N., Clevers, H., Vogelstein, B.,
and Kinzler, K. W. (1997). Activation of beta-catenin-ch signaling in colon cancer
by mutations in beta-catenin or APC. Science 275, 1787-1790.

Morwald, S., Yamazaki, H., Bujo, H., Kusunoki, J., Kanaki, T., Seimiya, K.,
Morisaki, N., Nimpf, J., Schneider, W. J., and Saito, Y. (1997). A novel mosaic
protein containing LDL receptor elements is highly conserved in humans and
chickens. Arterioscler Thromb Vasc Biol 17, 996-1002.

103

Murphy, G., Stanton, H., Cowell, S., Butler, 6., Knauper, V., Atkinson, S., and
Gavrilovic, J. (1999). Mechanisms for pro matrix metalloproteinase activation.
Apmis 107, 38-44.

Myklebost, O., Arheden, K., Rogne, S., Geurts van Kessel, A., Mandahl, N.,
Herz, J., Stanley, K., Heim, S., and Mitelman, F. (1989). The gene for the human
putative apoE receptor is on chromosome 12 in the segment q13-14. Genomics
5, 65-69.

Nagase, H. (1997). Activation mechanisms of matrix metalloproteinases. Biol
Chem 378, 151-160.

Nagase, H., and Woessner, J. F., Jr. (1999). Matrix metalloproteinases. J Biol
Chem 274, 21491-21494.

Nakamura, T., Hamada, F., Ishidate, T., Anai, K., Kawahara, K., Toyoshima, K.,
and Akiyama, T. (1998). Axin, an inhibitor of the Wnt signalling pathway, interacts
with beta-catenin, GSK—3beta and APC and reduces the beta-catenin level.
Genes Cells 3, 395-403.

Nakayama, M., Nakajima, D., Nagase, T., Nomura, N., Seki, N., and Ohara, O.
(1998). Identification of high-molecular-weight proteins with multiple EGF-like
motifs by motif-trap screening. Genomics 51, 27-34.

Neels, J. G., van Den Berg, B. M., Lookene, A., Olivecrona, G., Pannekoek, H.,
and van Zonneveld, A. J. (1999). The second and fourth cluster of class A
cysteine-rich repeats of the low density lipoprotein receptor-related protein share
ligand-binding properties. J Biol Chem 274, 31305-31311.

Nevins, J. R. (2001). The Rb/EZF pathway and cancer. Hum Mol Genet 10, 699-
703.

Newman, C. M., and Magee, A. I. (1993). Posttranslational processing of the ras
superfamily of small GTP-binding proteins. Biochim Biophys Acta 1155, 79-96.

Nicolaides, N. C., Papadopoulos, M, Liu, B., Wei, Y. F., Carter, K. C., Ruben, S.
M., Rosen, C. A., Haseltine, W. A., Fleischmann, R. D., Fraser, C. M., and et al.
(1994). Mutations of two PMS homologues in hereditary nonpolyposis colon
cancer. Nature 371, 75-80.

Nishisho, l., Nakamura, Y., Miyoshi, Y., Miki, Y., Ando, H., Horii, A., Koyama, K.,
Utsunomiya, J., Babe, 8., and Hedge, P. (1991). Mutations of chromosome 5q21
genes in FAP and colorectal cancer patients. Science 253, 665-669.

North, C. L., and Blacklow, S. C. (2000). Solution structure of the sixth LDL-A
module of the LDL receptor. Biochemistry 39, 2564-2571.

104

Nykjaer, A., Petersen, C. M., Moller, B., Jensen, P. H., Moestrup, S. K., Holtet, T.
L., Etzerodt, M., Thogersen, H. C., Munch, M., Andreasen, P. A., and et al.
(1992). Purified alpha 2-macroglobulin receptor/LDL receptor-related protein
binds urokinaseplasminogen activator inhibitor type-1 complex. Evidence that
the alpha Z-macroglobulin receptor mediates cellular degradation of urokinase
receptor-bound complexes. J Biol Chem 267, 14543-14546.

Nykjaer, A., Kjoller, L., Cohen, R. L., Lawrence, D. A., Gliemann, J., and
Andreasen, P. A. (1994). Both pro-uPA and uPA: PAI-1 complex bind to the
alpha Z-macroglobulin receptor/LDL receptor-related protein. Evidence for
multiple independent contacts between the ligands and receptor. Ann N Y Acad
Sci 737, 483-485.

Nykjaer, A., Conese, M., Christensen, E. l., Olson, D., Cremona, O., Gliemann,
J., and Blasi, F. (1997). Recycling of the urokinase receptor upon internalization
of the uPAzserpin complexes. Embo J 16, 2610-2620.

Nykjaer, A., and Willnow, T. E. (2002). The low-density lipoprotein receptor gene
family: a cellular Swiss army knife? Trends Cell Biol 12, 273-280.

O'Connor, L., Strasser, A., O'Reilly, L. A., Hausmann, 6., Adams, J. M., Cory, S.,
and Huang, D. C. (1998). Bim: a novel member of the Bcl-2 family that promotes
apoptosis. Embo J 17, 384-395.

O'Hagan, R. C., Ohh, M., David, G., de Alboran, I. M., Alt, F. W., Kaelin, W. G.,
Jr., and DePinho, R. A. (2000). Myc-enhanced expression of Cul1 promotes
ubiquitin-dependent proteolysis and cell cycle progression. Genes Dev 14, 2185-
2191.

Oka, K., Tzung, K. W., Sullivan, M., Lindsay, E., Baldini, A., and Chan, L. (1994).
Human very-low-density lipoprotein receptor complementary DNA and deduced
amino acid sequence and localization of its gene (VLDLR) to chromosome band
9p24 by fluorescence in situ hybridization. Genomics 20, 298-300.

Oliner, J. D., Kinzler, K. W., Meltzer, P. 8., George, D. L., and Vogelstein, B.
(1992). Amplification of a gene encoding a p53-associated protein in human
sarcomas. Nature 358, 80-83.

Olovnikov, A. M. (1973). A theory of marginotomy. The incomplete copying of
template margin in enzymic synthesis of polynucleotides and biological
significance of the phenomenon. J Theor Biol 41, 181-190.

O'Reilly, 3., Walicka, M., Kohler, S. K., Dunstan, R., Maher, V. M., and
McCormick, J. J. (1998). Dose-dependent transformation of cells of human

fibroblast cell strain MSU-1.1 by cobalt-60 gamma radiation and characterization
of the transformed cells. Radiat Res 150, 577-584.

105

Ortega, S., Malumbres, M., and Barbacid, M. (2002). Cyclin D-dependent
kinases, INK4 inhibitors and cancer. Biochim Biophys Acta 1602, 73-87.

Otey, C. A., Pavalko, F. M., and Burridge, K. (1990). An interaction between
alpha-actinin and the beta 1 integrin subunit in vitro. J Cell Biol 111, 721-729.

Ottilie, S., Diaz, J. L., Home, W., Chang, J., Wang, Y., Wilson, G., Chang, 8.,
Weeks, 8., Fritz, L. C., and Oltersdorf, T. (1997). Dimerization properties of
human BAD. Identification of a BH-3 domain and analysis of its binding to mutant
BCL-Z and BCL-XL proteins. J Biol Chem 272, 30866-30872.

Paget, S. (1989). The distribution of secondary growths in cancer of the breast.
1889. Cancer Metastasis Rev 8, 98-101.

Palecek, S. P., Loftus, J. C., Ginsberg, M. H., Lauffenburger, D. A., and Honlvitz,
A. F. (1997). lntegrin-ligand binding properties govern cell migration speed
through cell-substratum adhesiveness. Nature 385, 537-540.

Papadopoulos, N., Nicolaides, N. C., Wei, Y. F., Ruben, S. M., Carter, K. C.,
Rosen, C. A., Haseltine, W. A., Fleischmann, R. D., Fraser, C. M., Adams, M. D.,
and et al. (1994). Mutation of a mutL homolog in hereditary colon cancer.
Science 263, 1625-1629.

Parada, L. F., Tabin, C. J., Shih, C., and Weinberg, R. A. (1982). Human EJ
bladder carcinoma oncogene is homologue of Harvey sarcoma virus ras gene.
Nature 297, 474-478.

Parangi, S., O'Reilly, M., Christofori, G., Holmgren, L., Grosfeld, J., Folkman, J.,
and Hanahan, D. (1996). Antiangiogenic therapy of transgenic mice impairs de
novo tumor growth. Proc Natl Acad Sci U S A 93, 2002-2007.

Parsons, M., Keppler, M. D., Kline, A., Messent, A., Humphries, M. J., Gilchrist,
R., Hart, l. R., Quittau-Prevostel, C., Hughes, W. E., Parker, P. J., and N9, T.
(2002). Site-directed perturbation of protein kinase 0- integrin interaction blocks
carcinoma cell chemotaxis. Mol Cell Biol 22, 5897-5911.

Pavalko, F. M., Otey, C. A., Simon, K. O., and Burridge, K. (1991). Alpha-actinin:
a direct link between actin and integrins. Biochem Soc Trans 19, 1065-1069.

Pawson, T., and Scott, J. D. (1997). Signaling through scaffold, anchoring, and
adaptor proteins. Science 278, 2075-2080.

Pelengaris, S., Khan, M., and Evan, G. (2002). c-myc: more than just a matter of
life and death. Nat Rev Cancer 2, 764-776.

Pelkmans, L., and Helenius, A. (2002). Endocytosis via caveolae. Traffic 3, 311-
320.

106

Peterson, C. L. (2002). Chromatin remodeling: nucleosomes bulging at the
seams. Curr Biol 12, R245-247.

Peto, R. (1977) Epidemiology, multistage models, and short term mutagenicity
tests. In: Origins of Human Cancer, Cold Spring Harbor Laboratory, Cold Spring
Harbor, NY, 1403-1428.

Pfaff, M., Liu, S., Erle, D. J., and Ginsberg, M. H. (1998). Integrin beta
cytoplasmic domains differentially bind to cytoskeletal proteins. J Biol Chem 273,
6104-6109.

Pinson, K. l., Brennan, J., Monkley, S., Avery, B. J., and Skarnes, W. C. (2000).
An LDL-receptor-related protein mediates Wnt signalling in mice. Nature 407,
535-538.

Polakis, P. (2000). Wnt signaling and cancer. Genes Dev 14, 1837-1851.

Pulciani, S., Santos, E., Lauver, A. V., Long, L. K., and Barbacid, M. (1982).
Transforming genes in human tumors. J Cell Biochem 20, 51-61.

Pyke, C., Ralfkiaer, E., Huhtala, P., Hurskainen, T., Dano, K., and Tryggvason,
K. (1992). Localization of messenger RNA for Mr 72,000 and 92,000 type IV
collagenases in human skin cancers by in situ hybridization. Cancer Res 52,
1336-1341 .

Qing, J., Maher, V. M., Tran, H., Argraves, W. S., Dunstan, R. W., and
McCormick, J. J. (1997). Suppression of anchorage-independent growth and
matrigel invasion and delayed tumor formation by elevated expression of fibulin-
1D in human fibrosarcoma-derived cell lines. Oncogene 15, 2159-2168.

Qing, J., Wei, D., Maher, V. M., and McCormick, J. J. (1999). Cloning and
characterization of a novel gene encoding a putative transmembrane protein with
altered expression in some human transformed and tumor-derived cell lines.
Oncogene 18, 335-342.

Rabinovitz, I., and Mercurio, A. M. (1997). The integrin alpha6beta4 functions in
carcinoma cell migration on Iaminin-1 by mediating the formation and
stabilization of actin-containing motility structures. J Cell Biol 139, 1873-1884.

Radinsky, R. (1995). Molecular mechanisms for organ-specific colon carcinoma
metastasis. Eur J Cancer 31A, 1091-1095.

Radinsky, R., and Ellis, L. M. (1996). Molecular determinants in the biology of
liver metastasis. Surg Oncol Clin N Am 5, 215-229.

Ramos, Y. F ., Stad, R., Attema, J., Peltenburg, L. T., van der Eb, A. J., and
Jochemsen, A. G. (2001). Aberrant expression of HDMX proteins in tumor cells
correlates with wild-type p53. Cancer Res 61, 1839-1842.

107

Rampino, M, Yamamoto, H., Ionov, Y., Li, Y., Sawai, H., Reed, J. C., and
Perucho, M. (1997). Somatic frameshift mutations in the BAX gene in colon
cancers of the microsatellite mutator phenotype. Science 275, 967-969.

Rapoport, l., Chen, Y. C., Cupers, P., Shoelson, S. E., and Kirchhausen, T.
(1998). Dileucine—based sorting signals bind to the beta chain of AP-1 at a site
distinct and regulated differently from the tyrosine-based motif-binding site. Embo
J 17, 2148-2155.

Reddy, E. P., Reynolds, R. K., Santos, E., and Barbacid, M. (1982). A point
mutation is responsible for the acquisition of transforming properties by the T24
human bladder carcinoma oncogene. Nature 300, 149-152.

Renan, M. J. (1993). How many mutations are required for tumorigenesis?
Implications from human cancer data. Mol Carcinog 7, 139-146.

Rice, D. S., and Curran, T. (2001). Role of the reelin signaling pathway in central
nervous system development. Annu Rev Neurosci 24, 1005-1039.

Riemenschneider, M. J., Buschges, R., Wolter, M., Reifenberger, J., Bostrom, J.,
Kraus, J. A., Schlegel, U., and Reifenberger, G. (1999). Amplification and
overexpression of the MDM4 (MDMX) gene from 1q32 in a subset of malignant
gliomas without TP53 mutation or MDMZ amplification. Cancer Res 59, 6091-
6096.

Robinson, M. J., and Cobb, M. H. (1997). Mitogen-activated protein kinase
pathways. Curr Opin Cell Biol 9, 180-186.

Romao, M. J., Kolln, l., Dias, J. M., Carvalho, A. L., Romero, A., Varela, P. F.,
Sanz, L., Topfer-Petersen, E., and Calvete, J. J. (1997). Crystal structure of
acidic seminal fluid protein (aSFP) at 1.9 A resolution: a bovine polypeptide of
the sperrnadhesin family. J Mol Biol 274, 650-660.

Rubinfeld, B., Albert, I., Porfiri, E.,‘ Munemitsu, S., and Polakis, P. (1997). Loss of
beta-catenin regulation by the APC tumor suppressor protein correlates with loss
of structure due to common somatic mutations of the gene. Cancer Res 57,
4624-4630.

Russell, D. W., Brown, M. S., and Goldstein, J. L. (1989). Different combinations
of cysteine-rich repeats mediate binding of low density lipoprotein receptor to two
different proteins. J Biol Chem 264, 21682-21688.

Sabbatini, P., and McCormick, F. (2002). MDMX inhibits the p300/CBP-mediated
acetylation of p53. DNA Cell Biol 21, 519-525.

Saha, S., Boyd, J., Werner, J. M., Knott, V., Handford, P. A., Campbell, I. D., and
Downing, A. K. (2001). Solution structure of the LDL receptor EGF-AB pair: a

108

paradigm for the assembly of tandem calcium binding EGF domains. Structure
(Camb) 9, 451-456.

Saito, A., Pietromonaco, S., Loo, A. K., and Farquhar, M. G. (1994). Complete
cloning and sequencing of rat gp330/"megalin," a distinctive member of the low
density lipoprotein receptor gene family. Proc Natl Acad Sci U S A 91 , 9725-
9729.

Sakai, J., Hoshino, A., Takahashi, S., Miura, Y., Ishii, H., Suzuki, H.,
Kawarabayasi, Y., and Yamamoto, T. (1994). Structure, chromosome location,

and expression of the human very low density lipoprotein receptor gene. J Biol
Chem 269, 2173-2182.

Santos, E., Tronick, S. R., Aaronson, S. A., Pulciani, S., and Barbacid, M. (1982).
T24 human bladder carcinoma oncogene is an activated form of the normal
human homologue of BALB- and Harvey-MSV transforming genes. Nature 298,
343-347.

Satoh, S., Daigo, Y., Furukawa, Y., Kato, T., Miwa, N., Nishiwaki, T., Kawasoe,
T., lshiguro, H., Fujita, M., Tokino, T., et al. (2000). AXIN1 mutations in
hepatocellular carcinomas, and growth suppression in cancer cells by virus-
mediated transfer of AXIN1. Nat Genet 24, 245-250.

Schaller, M. D., Borgman, C. A., Cobb, B. S., Vines, R. R., Reynolds, A. B., and
Parsons, J. T. (1992). pp125FAK a structurally distinctive protein-tyrosine kinase
associated with focal adhesions. Proc Natl Acad Sci U S A 89, 5192-5196.

Schaller, M. D., Otey, C. A., Hildebrand, J. D., and Parsons, J. T. (1995). Focal
adhesion kinase and paxillin bind to peptides mimicking beta integrin cytoplasmic
domains. J Cell Biol 130, 1181-1187.

Schaller, M. D., Hildebrand, J. D., and Parsons, J. T. (1999). Complex formation
with focal adhesion kinase: A mechanism to regulate activity and subcellular
localization of Src kinases. Mol Biol Cell 10, 3489-3505.

Schreiber-Agus, N., Chin, L., Chen, K., Torres, R., Rao, G., Guida, P., Skoultchi,
A. l., and DePinho, R. A. (1995). An amino-terminal domain of Mxi1 mediates
anti-Myc oncogenic activity and interacts with a homolog of the yeast
transcriptional repressor SIN3. Cell 80, 777-786.

Sears, R. C., and Nevins, J. R. (2002). Signaling networks that link cell
proliferation and cell fate. J Biol Chem 277, 11617-11620.

Shampay, J., Szostak, J. W., and Blackburn, E. H. (1984). DNA sequences of
telomeres maintained in yeast. Nature 310, 154-157.

Shi, Y. (2002). Mechanisms of caspase activation and inhibition during apoptosis.
Mol Cell 9, 459-470.

109

Shih, T. Y., Weeks, M. 0., Young, H. A., and Scolnick, E. M. (1979). p21 of
Kirsten murine sarcoma virus is thermolabile in a viral mutant temperature
sensitive for the maintenance of transformation. J Virol 31 , 546-546.

Shih, T. Y., Papageorge, A. G., Stokes, P. E., Weeks, M. O., and Scolnick, E. M.
(1980). Guanine nucleotide-binding and autophosphorylating activities
associated with the p21src protein of Harvey murine sarcoma virus. Nature 287,
686-691.

Shimizu, K., Goldfarb, M., Suard, Y., Perucho, M., Li, Y., Kamata, T., Feramisco,
J., Stavnezer, E., Fogh, J., and Wigler, M. H. (1983). Three human transforming
genes are related to the viral ras oncogenes. Proc Natl Acad Sci U S A 80, 2112-
21 16.

Shippen-Lentz, D., and Blackburn, E. H. (1989). Telomere terminal transferase
activity from Euplotes crassus adds large numbers of I l I IGGGG repeats onto
telomeric primers. Mol Cell Biol 9, 2761-2764.

Shippen-Lentz, D., and Blackburn, E. H. (1990). Functional evidence for an RNA
template in telomerase. Science 247, 546-552.

Shvarts, A., Steegenga, W. T., Riteco, N., van Laar, T., Dekker, P., Bazuine, M.,
van Ham, R. C., van der Houven van Oordt, W., Hateboer, G., van der Eb, A. J.,
and Jochemsen, A. G. (1996). MDMX: a novel p53-binding protein with some
functional properties of MDMZ. Embo J 15, 5349-5357.

Simonovic, M., Dolmer, K., Huang, W., Strickland, D. K., Volz, K., and Gettins, P.
G. (2001). Calcium coordination and pH dependence of the calcium affinity of
ligand-binding repeat CR7 from the LRP. Comparison with related domains from
the LRP and the LDL receptor. Biochemistry 40, 15127-15134.

Sippola-Thiele, M., Hanahan, D., and Howley, P. M. (1989). Cell-heritable stages
of tumor progression in transgenic mice harboring the bovine papillomavirus type
1 genome. Mol Cell Biol 9, 925-934.

Smith-McCune, K. K., and Weidner, N. (1994). Demonstration and
characterization of the angiogenic properties of cervical dysplasia. Cancer Res
54, 800-804.

Solomon,E., Borrow, J., and Goddard, AD. (1991) Chromosome abberations
and cancer. Science, 254, 1153-1160.

Springer, T. A. (1998). An extracellular beta-propeller module predicted in
lipoprotein and scavenger receptors, tyrosine kinases, epidermal growth factor
precursor, and extracellular matrix components. J Mol Biol 283, 837-862.

Staller, P., Peukert, K., Kiennaier, A., Seoane, J., Lukas, J., Karsunky, H., Moroy,
T., Bartek, J., Massague, J., Hanel, F., and Eilers, M. (2001). Repression of

110

p15lNK4b expression by Myc through association with Miz-1. Nat Cell Biol 3,
392-399.

Steiner, P., Philipp, A., Lukas, J., Godden-Kent, D., Pagano, M., Mittnacht, S.,
Bartek, J., and Eilers, M. (1995). Identification of a Myc-dependent step during
the formation of active G1 cyclin-cdk complexes. Embo J 14, 4814-4826.

Stott, F. J., Bates, 8., James, M. C., McConnell, B. B., Starborg, M., Brookes, S.,
Palmero, l., Ryan, K., Hara, E., Vousden, K. H., and Peters, G. (1998). The
alternative product from the human CDKNZA locus, p14(ARF), participates in a
regulatory feedback loop with p53 and MDMZ. Embo J 17, 5001-5014.

Strahl, C., and Blackburn, E. H. (1996). Effects of reverse transcriptase inhibitors

on telomere length and telomerase activity in two immortalized human cell lines.
Mol Cell Biol 16, 53-65.

Strickland, D. K., Ashcom, J. D., Williams, S., Burgess, W. H., Migliorini, M., and
Argraves, W. S. (1990). Sequence identity between the alpha Z-macroglobulin
receptor and low density lipoprotein receptor-related protein suggests that this
molecule is a multifunctional receptor. J Biol Chem 265, 17401-17404.

Strickland, D. K., Kounnas, M. Z., and Argraves, W. S. (1995). LDL receptor-
related protein: a multiligand receptor for lipoprotein and proteinase catabolism.
Faseb J 9, 890-898.

Strickland, D. K., Gonias, S. L., and Argraves, W. S. (2002). Diverse roles for the
LDL receptor family. Trends Endocrinol Metab 13, 66-74.

Sudhof, T. C., Goldstein, J. L., Brown, M. S., and Russell, D. W. (1985a). The
LDL receptor gene: a mosaic of exons shared with different proteins. Science
228, 815-822.

Sudhof, T. 0., Russell, D. W., Goldstein, J. L., Brown, M. S., Sanchez-Pescador,
R., and Bell, G. I. (1985b). Cassette of eight exons shared by genes for LDL
receptor and EGF precursor. Science 228, 893-895.

Sugiyama, T., Kumagai, H., Morikawa, Y., Wada, Y., Sugiyama, A., Yasuda, K.,
Yokoi, N., Tamura, S., Kojima, T., Nosaka, T., et al. (2000). A novel low-density
lipoprotein receptor-related protein mediating cellular uptake of apolipoprotein E-
enriched beta-VLDL in vitro. Biochemistry 39, 15817-15825.

Takahashi, S., Kawarabayasi, Y., Nakai, T., Sakai, J., and Yamamoto, T. (1992).
Rabbit very low density lipoprotein receptor: a low density lipoprotein receptor-
Iike protein with distinct ligand specificity. Proc Natl Acad Sci U S A 89, 9252-
9256.

Takai, Y., Sasaki, T., and Matozaki, T. (2001). Small GTP-binding proteins.
Physiol Rev 81 , 153-208.

111

Tamai, K., Semenov, M., Kato, Y., Spokony, R., Liu, C., Katsuyama, Y., Hess, F.,
Saint-Jeannet, J. P., and He, X. (2000). LDL-receptor-related proteins in Wnt
signal transduction. Nature 407, 530-535.

Tanaka, S., Akiyoshi, T., Mori, M., Wands, J. R., and Sugimachi, K. (1998). A
novel frizzled gene identified in human esophageal carcinoma mediates
APC/beta-catenin signals. Proc Natl Acad Sci U S A 95, 10164-10169.

Tanaka, T., Yamaguchi, R., Sabe, H., Sekiguchi, K., and Healy, J. M. (1996).
Paxillin association in vitro with integrin cytoplasmic domain peptides. FEBS Lett
399, 53-58.

Taparowsky, E., Suard, Y., Fasano, O., Shimizu, K., Goldfarb, M., and Wigler, M.
(1982). Activation of the T24 bladder carcinoma transforming gene is linked to a
single amino acid change. Nature 300, 762-765.

Temeles, G. L., Gibbs, J. B., D'Alonzo, J. S., Sigal, I. S., and Scolnick, E. M.
(1985). Yeast and mammalian ras proteins have conserved biochemical
properties. Nature 313, 700-703.

Tomita, Y., Kim, D. H., Magoori, K., Fujino, T., and Yamamoto, T. T. (1998). A
novel low-density lipoprotein receptor-related protein with type II membrane
protein-like structure is abundant in heart. J Biochem (Tokyo) 124, 784-789.

Tong, L. A., de Vos, A. M., Milburn, M. V., Jancarik, J., Noguchi, S., Nishimura,
S., Miura, K., Ohtsuka, E., and Kim, S. H. (1989). Structural differences between
a ras oncogene protein and the normal protein. Nature 337, 90-93.

Trimarchi, J. M., and Lees, J. A. (2002). Sibling rivalry in the EZF family. Nat Rev
Mol Cell Biol 3, 11-20.

Trommsdorff, M., Borg, J. P., Margolis, B., and Herz, J. (1998). Interaction of
cytosolic adaptor proteins with neuronal apolipoprotein E receptors and the
amyloid precursor protein. J Biol Chem 273, 33556-33560.

Trommsdorff, M., Gotthardt, M., Hiesberger, T., Shelton, J., Stockinger, W.,
Nimpf, J., Hammer, R. E., Richardson, J. A., and Herz, J. (1999).
Reeler/Disabled-like disruption of neuronal migration in knockout mice lacking
the VLDL receptor and ApoE receptor 2. Cell 97, 689-701.

Trowbridge, l. S., Collawn, J. F., and Hopkins, C. R. (1993). Signal-dependent
membrane protein trafficking in the endocytic pathway. Annu Rev Cell Biol 9,
129-161.

Tsujimoto, Y., and Croce, C. M. (1986). Analysis of the structure, transcripts, and

protein products of bcI-Z, the gene involved in human follicular lymphoma. Proc
Natl Acad Sci U S A 83, 5214-5218.

112

van der Geer, P. (2002). Phosphorylation of LRP1: regulation of transport and
signal transduction. Trends Cardiovasc Med 12, 160-165.

Varela, P. F., Romero, A., Sanz, L., Romao, M. J., Topfer-Petersen, E., and
Calvete, J. J. (1997). The 2.4 A resolution crystal structure of boar seminal
plasma PSP-llPSP-II: a zona pellucida-binding glycoprotein heterodimer of the
spemladhesin family built by a CUB domain architecture. J Mol Biol 274, 635-
649.

Vennstrom, B., Sheiness, D., Zabielski, J., and Bishop, J. M. (1982). Isolation
and characterization of c-myc, a cellular homolog of the oncogene (v-myc) of
avian myelocytomatosis virus strain 29. J Virol 42, 773-779.

Vivanco, l., and Sawyers, C. L. (2002). The phosphatidylinositol 3-Kinase AKT
pathway in human cancer. Nat Rev Cancer 2, 489-501.

Volpert, O. V., Dameron, K. M., and Bouck, N. (1997). Sequential development of
an angiogenic phenotype by human fibroblasts progressing to tumorigenicity.
Oncogene 14, 1495-1502.

Volpert, O. V., Zaichuk, T., Zhou, W., Reiher, F., Ferguson, T. A., Stuart, P. M.,
Amin, M., and Bouck, N. P. (2002). lnducer-stimulated Fas targets activated
endothelium for destruction by anti-angiogenic thrombospondin-1 and pigment
epithelium-derived factor. Nat Med 8, 349-357.

Vousden, K. H. (2002). Activation of the p53 tumor suppressor protein. Biochim
Biophys Acta 1602, 47-59.

Vousden, K. H., and Lu, X. (2002). Live or let die: the cell’s response to p53. Nat
Rev Cancer 2, 594-604.

Watson, J. D. (1972). Origin of concatemeric T7 DNA. Nat New Biol 239, 197-
201.

Weaver, A. M., Hussaini, I. M., Mazar, A., Henkin, J., and Gonias, S. L. (1997).
Embryonic fibroblasts that are genetically deficient in low density lipoprotein
receptor-related protein demonstrate increased activity of the urokinase receptor
system and accelerated migration on vitronectin. J Biol Chem 272, 14372-14379.

Webb, C. P., and Vande Woude, G. F. (2000). Genes that regulate metastasis
and angiogenesis. J Neurooncol 50, 71-87.

Wehrli, M., Dougan, S. T., Caldwell, K., O'Keefe, L., Schwartz, 8., Vaizel-
Ohayon, D., Schejter, E., Tomlinson, A., and DiNardo, S. (2000). arrow encodes
an LDL-receptor-related protein essential for Wingless signalling. Nature 407,
527-530.

113

Weidner, N., Semple, J. P., Welch, W. R., and Folkman, J. (1991). Tumor

angiogenesis and metastasis-correlation in invasive breast carcinoma. N Engl J
Med 324, 1-8.

Weidner, N., Folkman, J., Pozza, F., Bevilacqua, P., Allred, E. N., Moore, D. H.,
Meli, S., and Gasparini, G. (1992). Tumor angiogenesis: a new significant and

independent prognostic indicator in early-stage breast carcinoma. J Natl Cancer
Inst 84, 1875-1887.

Weinberg, R. A. (1991). Tumor suppressor genes. Science 254, 1138-1146.

Weinstein, M., Yang, X., and Deng, C. (2000). Functions of mammalian Smad
genes as revealed by targeted gene disruption in mice. Cytokine Growth Factor
Rev 1 1, 49-58.

Weiss, L., Grundmann, E., Torhorst, J., Hartveit, F., Moberg, l., Eder, M.,
Fenoglio-Preiser, C. M., Napier, J., Home, C. H., Lopez, M. J., and et al. (1986).
Haematogenous metastatic patterns in colonic carcinoma: an analysis of 1541
necropsies. J Pathol 150, 195-203.

Weiss, L. (1992). An analysis of the incidence of myocardial metastasis from
solid cancers. Br Heart J 68, 501-504.

Wodarz, A., and Nusse, R. (1998). Mechanisms of Wnt signaling in development.
Annu Rev Cell Dev Biol 14, 59-88.

Yamamoto, T., Davis, C. G., Brown, M. S., Schneider, W. J., Casey, M. L.,
Goldstein, J. L., and Russell, D. W. (1984). The human LDL receptor: a cysteine-
rich protein with multiple Alu sequences in its mRNA. Cell 39, 27-38.

Yan, W., Sheng, N., Seto, M, Morser, J., and Wu, Q. (1999). Corin, a mosaic
transmembrane serine protease encoded by a novel cDNA from human heart. J
Biol Chem 274, 14926-14935.

Yang, D., Louden, C., Reinhold, D. S., Kohler, S. K., Maher, V. M., and
McCormick, J. J. (1992). Malignant transformation of human fibroblast cell strain
MSU-1.1 by (+-)-7 beta,8 alpha-dihydroxy—9 alpha,10 alpha-epoxy-7,8,9,10-
tetrahydrobenzo [a]pyrene. Proc Natl Acad Sci U S A 89, 2237-2241.

Yang, E., Zha, J., Jockel, J., Boise, L. H., Thompson, C. B., and Korsmeyer, S. J.
(1995). Bad, a heterodimeric partner for BcI-XL and Bcl-Z, displaces Bax and
promotes cell death. Cell 80, 285-291.

Zorn, A. M. (2001). Wnt signalling: antagonistic Dickkopfs. Curr Biol 11, R592-
595.

114

Chapter II

$17 is a novel member of the low-density lipoprotein receptor (LDLR)

superfamily with a cytoplasmic tail that interacts with proteins related to

signal transduction pathways.

Michele A. Battle, Veronica M. Maher, and J. Justin McCormick1

Carcinogenesis Laboratory, Cell and Molecular Biology Program, Department of
Biochemistry and Molecular Biology. and Department of Microbiology and
Molecular Genetics, Michigan State University, East Lansing, Michigan 48824-

1 302

' This work was supported by United States Department of Health and Human
Services grants AG11026, CA82885, and CA09726Z.

1To whom requests for reprints should be addressed, Carcinogenesis
Laboratory, Food Safety and Toxicology Building, Michigan State University, East
Lansing, MI 48824-1302. Phone (517) 353-7785; Fax (517) 353-9004; E-mail:

mccormi1@msu.edu.

115

2 Manuscript in preparation, M.A. Battle, S. O’Reilly, V.M. Maher, and J. J.
McCormick. Expression of ST7 protein inhibits the tumor-forming ability of a
human fibrosarcoma-derived cell line.

3 The abbreviations used are: LDLR, low-density lipoprotein receptor; LRP,
lipoprotein receptor-related protein; LDLRA, low-density lipoprotein receptor type
A; CUB, Complement factor C1s/C1r, Urchin embryonic growth factor, Bone
morphogenetic protein; RACK1, receptor for activated protein kinase C; SARA,
SMAD anchor for receptor activation; MIBP, muscle integrin binding protein;

SNAPIN, SNARE-associated protein.

116

ABSTRACT

Previous studies identified the novel putative tumor suppressor ST7.
Subsequent discovery of two proteins closely-related to ST7 strongly suggests
that ST7 is a novel member of the low-density lipoprotein receptor (LDLR)
superfamily. It has been proposed that ST7 and its two related proteins
constitute an LDLR subfamily. Although proteins of the LDLR superfamily are
best known as endocytic receptors, recent studies of several such proteins
demonstrate that their cytoplasmic domains interact with a variety of adaptor and
scaffold proteins implicated in signal transduction pathways. To evaluate ST7’s
relationship to the proteins of the LDLR superfamily, we used proteomic tools to
analyze the functional motifs present in ST7. The data confirm that ST7 is,
indeed, a member of an LDLR subfamily, which also contains LDLR-related
protein 3 (LRP3) and murine LRP9. We also found that ST7’s cytoplasmic
domain contains several motifs implicated in endocytosis and signaling. Use of
the yeast-two hybrid system to identify proteins that associate with ST7’s
cytoplasmic domain revealed that ST7’s cytoplasmic domain, like those of other
proteins in the LDLR superfamily, interacts with proteins related to signal
transduction and/or endocytosis, strongly suggesting that ST7 functions in these

two pathways.

117

INTRODUCTION

It is generally accepted that each human cancer is clonal in origin, i.e.,
begins with a single cell that has acquired all of the properties necessary to form
a tumor. Every cell of the tumor is the progeny of this first cell. Theories of
carcinogenesis must explain how the “first” tumor cell acquired the appropriate
genetic and/or epigenetic changes required to render it malignant, which for
tumors in adults are estimated to be five or more. To try to define the number
and nature of such changes, McCormick and Maher (1996) developed a model
system in which normal human fibroblasts in culture can be transformed into
malignant fibroblasts by acquiring a series of genetic changes, each conferring a
growth advantage that allows sequential clonal expansion.

In 1997, using differential mRNA display to compare an infinite life span,
non-tumorigenic human fibroblast cell strain, designated MSU-1.1, to a
carcinogen-transformed, malignant MSU-1.1 derivative, McCormick and his
colleagues identified a novel gene, which they designated ST7 (Qing et al.,
1999). Compared to the level of expression in the parental cell strain, MSU1.1,
ST7 was found to be strongly down-regulated in the tumorigenic cell line.
Additional malignant cell lines also were found to have a low level of ST7,
suggesting that ST7 is a novel tumor suppressor gene. Recent studies from this
laboratory demonstrated that overexpression of ST7 protein inhibited the tumor-
forming ability of a human fibrosarcoma-derived cell line, whereas
overexpression of a truncated ST7 protein that lacked most of the C-terrninus

failed to do soz. These latter data strongly suggest that the C-terrninus of ST7

118

plays a critical role in the protein’s function as a tumor suppressor.

When the novel gene ST7 was first identified, it was recognized that it had
the characteristics of a transmembrane protein, but at that time no other proteins
with significant similarity to ST7 had been reported. Approximately one year
later, Yamamoto and his associates, using degenerate Oligonucleotides
corresponding to the highly conserved region of the ligand-binding domains
found in the proteins of the LDLR3 superfamily, discovered a novel LDLR-related
gene, which they designated as the LRP3 gene (Ishii et al., 1998). Our search of
the databases for proteins structurally related to ST7, after the sequence of LRP3
had been submitted, revealed a very strong similarity between LRP3 and ST7,
enabling us to recognize that ST7, although it differed significantly from the LDLR
prototype, was also a novel member of the LDLR superfamily. Although proteins
of the LDLR superfamily are best known as endocytic receptors (see Krieger and
Herz, 1994; Hussain et al., 1999; Strickland et al., 2002 for review), recent
studies show that the cytoplasmic domains of LRP1, LRPZ, LRP5, LRP6, LDLR,
VLDLR, and ApoER2 interact with a variety of adaptor and scaffold proteins
implicated in signal transduction (see Li et al., 2001; Herz and Bock, 2002; van
der Geer, 2002 for review).

To evaluate ST7’s relationship to the proteins of the LDLR superfamily, we
used proteomic tools including similarity searches, sequence alignments, pattern
and profile searches, and post-translational modification prediction programs to
analyze the functional motifs present in ST7. Such analyses showed that ST7 is,

indeed, a novel member of the LDLR superfamily and that similar to other LDLR

119

proteins, ST7’s cytoplasmic domain contains several motifs implicated in
endocytosis and/or signaling. To identify intracellular proteins that interact with
ST7’s cytoplasmic domain, we used the yeast two-hybrid system. We found that
ST7’s cytoplasmic domain interacts with several proteins related to signal
transduction and/or endocytosis, suggesting that ST7 functions in these

pathways.

120

EXPERIMENTAL PROCEDURES

Materials—The sources of specific materials used are: restriction enzymes
and T4 DNA ligase, New England Biolabs or lnvitrogen; shrimp alkaline
phosphatase, Promega; Pfu polymerase, Stratagene; Oligonucleotides, the
Michigan State University Macromolecular Structure, Sequencing, and Synthesis
Facility; yeast culture medium, Clontech; and antibiotics, Sigma or Roche
Molecular Biochemicals.

Cell Culture—The HEK 293-T cells were cultured in high-glucose
Dulbecco’s modified Eagle’s medium containing 10% (vlv) fetal calf serum
(HyClone), 100 units/ml penicillin, and 100 ug/ml streptomycin in a 37°C
humidified incubator with 5% CO; in air.

Isolation of ST7 cDNA—To isolate a full-length ST7 cDNA, the
GeneTrapperTM cDNA Positive Selection System (lnvitrogen) and SuperScriptTM
human heart cDNA library (lnvitrogen) were used according to the
manufacturer’s directions. Briefly, single-stranded phagemid DNA was
generated by digesting double-stranded phagemid DNA (5 pg) with
bacteriophage f1 gene ll protein (a site-specific F1 endonuclease) and E.ccli
exonuclease III. A biotinylated ST7-specific probe (5’-
TTGCTCTTGCT'ITI'CCTCGCTGGGG-3’) was hybridized with single-stranded
DNA for 1h at 37’C. Following hybridization, streptavidin paramagnetic beads
were used to capture ST7 cDNA-probe complexes. After several washes,
single-stranded DNA bound to the beads was eluted. To produce double-

stranded DNA, captured DNA was repaired using the provided repair enzyme

121

 

and the same unbiotinylated ST7-specific oligonucleotide used in the capture
procedure. Repaired DNA was electroporated into ultracompetent ElectroMax
DH10B E. coli (lnvitrogen) and plated on LB agar plates containing 100 pg/ml
ampicillin. DNA isolated from several colonies was screened for the presence of
a full-length ST7 cDNA insert (~3.0 Kb) by digesting the DNA with several
restriction endonucleases and sequencing the 5’ and 3’ ends of the DNA. The
phagemid containing ST7 was designated pCMV—SPORT-ST7.

DNA sequencing—The sequence of DNA samples was determined in one
of three ways: manually determined using the dideoxy chain termination method
with a TAQuenoe® cycle sequencing kit (US Biochemicals) or Fidelity DNA
sequencing kit (Oncor), using primers radiolabeled by T4 polynucleotide kinase
and v-32P dATP; automatically sequenced using a Visible Genetics Gene Clipper
sequencer with either a ThermoSequenase Cy 5.5 terminator cycle sequencing
kit or a ThermoSequenase primer (Cy 5.0 or Cy 5.5) cycle sequencing kit
(Amersham); or sequenced by the Michigan State University Genomics
Technology Support Facility.

Yeast two-hybrid assay— (The MATCHMAKER GAL4 Two-Hybrid System
3 (Clontech) was used. The ST7 cytoplasmic bait plasmids were constructed
using the pGBKT7 bait vector. To generate the bait plasmids, the appropriate
region of ST7’s cytoplasmic domain was amplified from pCMV-SPORT-Sl7
using Pfu polymerase and the appropriate Oligonucleotides. 817mm contains
residues 498-840; ST7cym, residues 498-667; ST7cyt03, residues 668-840;

ST7cyto4, residues 498-652; ST7¢yto5, residues 498-560; and Schtos, residues

122

561-667. The purified PCR products (QlAquick spin PCR purification kit, Qiagen)
were digested with restriction endonucleases EcoRI and Xhol and subcloned into
EcoRI and 83!] sites of the bait vector pGBKT7. The S. cerevisiae strain AH109,
which contains the HIS3, ADEZ, and MEL1 reporter genes, was transformed
with the bait plasmid using the lithium acetate method described in the
MATCHMAKER instruction manual (Clontech). Transformants containing the
bait plasmid were selected by plating the cells on Minimal Synthetic Dropout (SD)
agar lacking tryptophan (SDI-trp). A MATCHMAKER skeletal muscle cDNA
library (Clontech) was amplified following the standard semi-solid amplification
procedure described in the MATCHMAKER instruction manual, and plasmid DNA
was prepared using the QIAGEN plasmid giga kit. To detect ST7-interacting
proteins, AH109-ST7cyto1 cells were transformed with the amplified library.
Transformants were plated onto medium-stringency SDI-leul-trpl-his agar. After
incubation at 30°C for several days, colonies were streaked onto SDI-leul-trp and
SDI-leul-trpI-his agar plates. These yeast colonies were subjected to a higher
stringency test, i.e., they were replica plated onto SDI-leuI-trpl-hisl-ade agar
plates. The colonies judged to grow well at high stringency conditions were re-
streaked for further analysis. Plasmid DNA was isolated from yeast colonies
using the Zymoprep yeast plasmid miniprep kit (Zymo Research). Prey plasmids
were rescued by transforming E. coli DH5ol with the plasmid DNA isolated from
yeast and by selecting these transforrnants on LB agar containing ampicillin at
100 uglml. The prey plasmids isolated from E. coli (QIAprep spin plasmid DNA

miniprep kit, Qiagen) were sequenced using automated DNA sequencing

123

technology (Visible Genetics). To map various binding domains in ST7, bait
plasmids encoding truncated forms of the ST7 cytoplasmic domain were
engineered, and AH109 strains expressing the truncated bait proteins were
transformed with the pACTZ prey plasmids isolated in the original yeast-two
hybrid screen. As a negative control, these strains were also transformed with a
prey plasmid encoding the SV40 large T-antigen. Transformants were plated on
quadruple drop-out agar (SDI-leul-trpl-hisl-ade). Growth on this medium
indicated that the truncated STAY,o protein interacted with the prey protein of
interest.

Oligonucleotides—Primer sequences (5’-3’) used to construct the
plasmids for the yeast two hybrid screen and for co-immunoprecipitation studies
follow. Restriction sites used for cloning are in boldface type, stop codons are in
italics, and template sequences are underlined: 5’ primer for ST7cyto1, ST7cym,
ST7cyto4, and ST7cyt05 , CGGAATTCAflT‘ITATTCTCTGAGA; 3’ primer for
ST7cyto1 and ST7cyt03, CGCTCGAGCTAACAAAGTAAQAAAGCCTC; 5’ primer
for ST7cyto3, CGGAA TTCGCGACAGTAGGAGCATGT; 3’ primer for ST7cym and
ST7cyt06, CGCTC GAG TTA‘I‘I'CCACTGCCGTI'GTGGG; 3’ primer for ST7cyto4,
CGCTCGAGTTAQTGAGTATGATTI’CT; 3’ primer for ST7cytos, CGCTCGAG-
TTAQQLQAGATTI'TCQAA; 5’ primer for ST7cyt05, CGGAATI'CCCTAGC-
GGTACGATCT. Primer sequences (5’-3’) for subcloning of HA-tagged partial
prey proteins into pcDNA6-V5/His follow: 5’ primer for HA-RACKip, HA-MIBPp,
HA-SARAp, HA-o-actinin-Zp, HA-SNAPINp, and HA-myotilin, AAGCGGCCGC-

QQCACCATGGCTTACCCA; 3’ primer for HA-RACK1p, HA-MIBPp, HA-SARAp,

124

HA—o—actinin-Zp, HA-SNAPINp, and HA-myotilin, CGACTAGTGAGATGGTGC—
AQGATGCACAGTTG. Primer sequences for cloning full-length myotilin cDNA
into the prey plasmid pACTZ: 5’ primer, CGGAATTCTAAGCATGTT-
TAAEAQAACGI; 3’ primer, CGCTCGAG TTAAAGTTgTTCfigTFTCATAG.
Primer sequences for isolating full-length RACK1 cDNA: 5’ primer, AAGATATC-
CATQACTGACflGATQ/igg; 3’ primer, AAGTCGACCTAGCGTGTGCC-
AATGGTCAC; Primer sequences for isolating full-length MIBP cDNA: 5’ primer,
AAGATATCCATGAAGCTCATCGTGGGC. Primer sequences for constructing
pcDNA6-ST7-V5/His, pcDNA6-S‘l7m-V5/His, and pcDNA6-S'I7trz-V5/His: 5’
primer, which hybridizes to pCMV-SPORT vector sequence upstream of an
EcoRI site and the ST7 start codon, TAGGTGACACTATAGAAG_G-
TACGCCTGCAG; 3’ for ST7-V5lHis, CGGAATTCTCGT/LCCAAACA—
AAGTAACAAAQ; 3’ primer for ST7tr1-V5/His, CGCTCGAGIQTTLCAAAC;
ATTCTCAG; 3’ primer for ST7m-V5/His, CGCTCGAGC'I'I'ACAAGT—
ACATCCCAA.

Construction of mammalian expression plasmids—CD NAs encoding the
prey proteins of interest were PCR amplified from the appropriate pACTZ prey
plasmids using Pfu polymerase. The 5’ primer used hybridizes with the pACTZ
HA epitope sequence located upstream of the cDNA insertion site; the 3’ primer
used hybridizes with sequence on pACTZ downstream of the cDNA insertion site.
Therefore, each PCR products contains the entire partial prey cDNA with an N-
terrninal HA-tag. The GAL4 activation domain was not amplified in this

procedure. Most PCR products (QIAquick spin PCR purification kit, Qiagen)

125

were digested with Spa! and subcloned the EcoRV and Xbal sites of the
mammalian expression vector pcDNA6-V5lHis (lnvitrogen). One PCR product
was left undigested, and this blunt-ended insert was subcloned into the EcoRV
site of pcDNA6-V5/His. The primers were designed to contain a stop codon at
the 3’ end of each PCR product so that the proteins would not be V5lHis tagged.
Any full-length cDNA inserts isolated from the yeast library were used to
generate an expression plasmid containing the entire open reading from with an
N-terminal HA tag. First, the entire open reading frame was amplified from the
pACTZ plasmid containing the ORF using Pfu polymerase. The purified PCR
product was digested with EcoRI and Xhol and subcloned into the corresponding
sites in pACTZ. The HA-tagged, full-length prey protein was amplified from this
plasmid and subcloned into pcDNA6-V5/His in the same manner as the others.
Additional full-length cDNAs were amplified from the MATCHMAKER skeletal
muscle cDNA library using Pfu polymerase and the appropriate primers. The
purified PCR products were digested with EcoRV and Sell and subcloned into
the EcoRV and Xhol sites of pcDNA6-HA. To construct pcDNA6-HA, the HA
epitope was excised from pcDNA6-HA-MIBPp using EcoRI and subcloned into
the EcoRI site of pcDNA6-V5/His. Again, because the primers were designed to
contain the endogenous stop codons, these proteins were not tagged by V5lHis.
To construct mammalian expression plasmids containing full-length V5lHis
tagged ST7 or truncated V5lHis tagged ST7, the appropriate region of ST7 was
amplified from pCMV-SPORT-ST7 using Pfu polymerase and the appropriate

oligonucleotides. The purified PCR products (QIAquick spin PCR purification kit,

126

Qiagen) were digested with restriction endonucleases EcoRl (ST7) or with EcoRI
and Xhol (ST7tr1 and STA-2) and subcloned into EcoRI or EcoRI and Xhol sites
pcDNA6-V5/His A (lnvitrogen). Elimination of the endogenous stop codons
caused the proteins encoded by these cDNAs to have C-terminal V5 and His
epitope tags.

Co-immunoprecipitation of proteins from transiently transfected 293-T
cells—293-T cells (1.5 x 106) were plated in 60-mm diameter tissue culture
dishes one day prior to transfection. Cells were transfected with a total of 10 pg
of plasmid DNA using LipofectAMlNE (lnvitrogen) following the manufacturer’s
instructions. Forty-eight hours after transfection, the cells were washed once in
cold PBS and collected in cold lysis buffer (50 mM Tris-HCI, pH 7.2, 150 mM
NaCl, 0.5% NP-40, 50 mM NaF, 1 mM Na3VO4, 1 mM benzamidine, 1 mM
phenylmethylsulfonyl fluoride, 25 pglml aprotinin, and 25 pg lml leupeptin). The
cells were incubated in the lysis buffer for 30 min on ice with periodic gentle
vortexing. The protein extracts were clarified by centrifugation (25,000 x g) for 30
min at 4°C. The supernatants were transferred to fresh tubes and centrifuged
(25,000 x g) for an additional hour 4°C. Total protein concentration was
determined using the Coomassie protein assay reagent (Pierce) following the
manufacturer’s instructions. For immunoprecipitation, protein extracts (250-500
pg) were pre-cleared using normal rabbit serum (0.25 pg) (Santa Cruz
Biotechnology) and protein A-agarose beads (20 pl) (Santa Cruz Biotechnology)
for 30 min at 4°C with end-over-end mixing. The pre-cleared supernatants were

incubated with 2 pg rabbit polyclonal anti-HA antibody (Y-11, Santa Cruz

127

 

Biotechnology) for 2 h at 4°C with end-over-end mixing. A control reaction was
carried out for each protein pair in which no HA antibody was included in the
immunoprecipitation reaction. Protein A-agarose beads (20 pl) were added, and
the reactions were incubated for an additional hour at 4°C with end-over-end
mixing. The pellets were washed three times with a buffer containing 25 mM
Tris-HCI, pH 7.4, 30 mM MgCIz, 40 mM NaCl, 1.0% NP-40, and once with a
buffer containing 25 mM Tris-HCI, pH 7.4, 30 mM MgCl2, 40 mM NaCl. Following
the washes, the agarose pellets were resuspended in 40 pl of 1X Laemelli SDS-
PAGE sample buffer. Denatured immunoprecipitated proteins were separated by
SDS-PAGE and transferred to Immobilon-P membrane (Millipore). Such
membranes were blocked for 2 h with Tris-buffered saline containing 0.1%
Tween-20 (T BST) and 5% non-fat milk. Blocked membranes were incubated
with a monoclonal V5 antibody (lnvitrogen) and a monoclonal HA antibody (F-7,
Santa Cruz Biotechnology) diluted in the blocking solution for at least 2 h at room
temperature and with the appropriate HRP-conjugated secondary antibody
(Santa Cruz Biotechnology or Sigma) diluted in blocking solution for 1 h at room
temperature. SuperSignal West Pico chemiluminescent substrate (Pierce) was

used according to the manufacturer’s instructions to detect proteins.

128

RESULTS

Isolation of ST7 cDNA from a human heart muscle cDNA library—
Because the ST7 cDNA fragments originally isolated had been obtained from
multiple cDNA libraries using PCR with relatively low fidelity DNA polymerases,
there was a possibility that they contained PCR generated mutations. Therefore,
we isolated ST7 cDNA from a human heart cDNA library using the GeneTrapper
cDNA Positive Selection System. Comparison of this DNA sequence to those of
Qing et al. (1999) revealed that this ST7 cDNA is the shorter of the two known
ST7 isoforms. It lacks 57 nucleotides at its 5’ end (nucleotides 79 -136 of
GenBank accession number NM_013437). To determine if this isoform was the
product of alternative splicing of the transcript, we used the human genome
BLAST program to search the human genomic DNA database for the ST7
sequence. The ST7 gene was localized to an approximately 100 Kb region
human chromosome 8q (Locus Link ID 29967), in agreement with the fluorescent
in situ hybridization data of Qing et al. (1999) showing ST7’s location to be
chromosome 8q22.2-23.1. This sequence contains seven putative exons (Table
l). The 57 nucleotides missing in the short isoform correspond to the second
exon, strongly suggesting that the short isoform is a product of alternative
splicing of the transcript.

ST7 is a novel member of the LDLR family of proteins—The similarity
between ST7 and LRP3 strongly suggested that ST7 is a novel member of the
LDLR superfamily. Comparison of residues 27-604 of ST7 with residues 42-627

of LRP3 using BLAST revealed that the two proteins are 50% identical and 66%

129

 

Table I

 

 

Structure of the ST7 gene
Exon Intron
size size
Exon Intron — exon boundaries (bp) Exon — intron boundaries (kb)
1 79 GGGGTGTACG gtaagtgtcc 55,9
2 ttaattttag GAAATGGTGC 57 GTGTCAACTG gtaagtcatt 22.8
3 tatcttccag CTTGTGGAGA 136 TACTATAAGG gtaattctac 9.4
4 tttttttcag TTTTCAGGAT 203 TATTTTTCAG gtgtgttttt 12.4
5 atcttcacag GGAAATCTGA 1105 TTGAAAGAAG gtcagtatca 1,8
6 ctctcaaaag ATCATTTGAA 133 ACCTAATCAG gtatattgca 3.5
7 atttttgtag GCTTCTGTTT 866 ACTTTGTTAG

 

130

 

 

similar. After this research had been carried out, Sugiyama et al. (2000) reported
the discovery a murine LDLR-related protein, LRP9, which closely resembles
ST7 and LRP3. They suggested that LRP9, LRP3, and ST7 constitute a novel
subfamily of the LDLR superfamily. Comparison of ST7 with residues 28-698 of
murine LRP9 showed that the proteins are 35% identical and 46% similar. The
most striking characteristics common to all three proteins are: 1) their
extracellular domains contain the same types of functional domains arranged in
an identical configuration (Figure 1); 2) the juxtamembrane regions of their
cytoplasmic domains are highly conserved (Figure 2A), strongly suggesting that
this region is functionally important; and 3) their cytoplasmic tails contain putative
signals for endocytosis and sequences that may participate in signal
transduction. These highly related features are strong evidence for considering

that ST7, LRP3, and LRP9 comprise an LDLR subfamily.

Analysis of the ST7 protein—Qing et al. (1999) predicted that ST7
encodes a type I single-pass transmembrane protein. To gain additional
information regarding ST7’s protein structure, its sequence was further analyzed
using proteomic analysis tools available through the ExPASy Molecular Biology
Server (www.expasy.ch) The protein’s putative extracellular domain, residues 1
to 473, contains a signal sequence peptide motif (residues 127) as determined
by the SignalP prediction program (Nielsen et al., 1997). Such motifs are
characteristic of proteins targeted to the plasma membrane (Claros et al., 1997).
This region also contains five LDLRA domains arranged in two clusters and two

CUB domains, as determined using ScanProsite (Gattiker et al., 2002) (Figure 1).

131

 

Figure 1. Structures of representative proteins of the LDLR superfamily
(adapted from Strickland et al., 2002). The types of domains in these proteins
and the organization of these domains divide the superfamily into at least four
subfamilies: 1) LDLR, VLDLR, and ApoER2; Z) LRP1, LRP1B, and LRPZ; 3)
LRP5 and LRP6; 4) ST7, LRP3, and murine LRP9. The cytoplasmic domains of
these proteins contain motifs implicated in endocytosis and/or signal transduction,

which are not shown. (Images in this dissertation are presented in color.)

132

 

:ooolul'l'fz

Iooolilzz

 

:08 ll :2
1000!“ 12
:ooolml :2

P 2:9“.

an»... act—:2
nami—

Cb

cam.—

mam.—

 

c_eEou 353.330

53.2. 5.833053 cox—=70
Eon! 9.585995;
«cone. 9:..u0m

52:2. (1.5..

52:00 mac

e258 8538 .25.»

 

 

 

:2
f2

IZ

«mm.—

m Em.—

Em...

133

Figure 2. The cytoplasmic domain of ST7 (residues 498-840). A. Alignment
of the cytoplasmic domains of ST7, LRP3, and murine LRP9. Regions of identity
in these proteins are red, and regions of similarity are blue. The consensus
sequence is shown beneath the alignment in black. Uppercase letters designate
amino acids conserved in all three proteins. Lowercase letters designate amino
acids conserved in two of the three proteins. Note the high degree of sequence
conservation in the juxtamembrane region (498-560 of ST7) and in the extreme
carboxy terminal region of these proteins. B. Putative endocytic and signaling
motifs identified in ST7’s cytoplasmic domain. The YXXO internalization motif is
indicated by a double-underline. The NPXY-like internalization motif is shown by
a dotted underline. A single underline denotes the three dileucine motifs.
Potential phosphorylation sites are boxed. The four PDZ domain binding motifs
are shown in boldface italic font. A line is drawn above the WW domain binding

motif. (Images in this dissertation are presented in color.)

134

A

ST7

LRP3

LRP9
consensus

ST7

LRP3

LRP9
consensus

ST7

LRP3

LRP9
consensus

ST7

LRP3

LRP9
consensus

ST7

LRP3

LRP9
consensus

ST7

LRP3

LRP9
consensus

498
521
446

558
581
525

615
633
576

675
659
598

735
681
627

795
725
668

KLYSLRMFERRSFETQLSRVEAELLRREAPPSYGQLIAQGLIPPVEDFPVCSPNQASVLE
KLYSLRTQEYRAFETQMTRLEAEFVRREAPPSYGQLIAQGLIPPVEDFPVYSASQASVLQ
KLYAIRTQEYSIRAP—LSRMEAEIVQQQAPPSYGQLIAQGAIPPVEDFPTENPNDNSVLG
KLYisthyr FetqlsRvEAElvrreAPPSYGQLIAQGlIPPVEDFPV spnanVL

NLR--LAVRSQLG-FTSVRLPMAGRSSNIWNRIFNFARSRHSGSLALVSADGDEVVPSQS
NLR--TAMRRQMRRHASRRGPSRRRLGRLWNRLFHRPRAPR-GQIPLLTAAR----PSQ-
NLPSLLQILRQDMTPGGTSGGRRRQRGRSVRRLVRRLR--RWGLLPRTNT ------ PAR-
NLR laverl as rgp rrr griwanfhr R r G lplvsa Psq

TSREPERNHTHRSLFSVESDDTDTENERRDMAGASGGVAAPLPQKVPPTTAVEATVGACA

--------------- TVLGDGFLQP--—-----APG--AAPD----PPAPLMD--TG---
------------------ APETRSQ—---—-------—VTPS---—VPSEALDDSTG---
sv gddt t a g aaP th avd ttG

SSSTQSTRGGHADNGRDVTSVEPPSVSPARHQLTSALSRMTQGLRWVRFTLGRSSSLSQN
----- STR---AAGDR---——-PPS-AP-------—---------------GRAPEVGPS
-—--HACEGG-AVGGQDGEQ-APP--LP ----------------------- IKTPIPTPS

strgg A ggrd PPs P grsp lsps

QSPLRQLDNGVSGREDDDDVEMLIPISDGSSDFDVNDCSRPLLDLASDQGQGLRQPYNAT

GPPLP ------ SGLRDPE---CR-PV ----- DKDRKVCREPLADGPAP-ADAPREPCSAQ
TLPALA ----- TVSETPG- ----- PL ------ PSVPVESS-LLSGVVQVLRGRLLP-SLW
P1 39 edpd Pi d dv vcs ledg g g r P sa
NPG ------ VRPSNRDGPCERCGIVHTAQIPDTCLEVTLKNETSDDEALLLC
DPHP---QVSTASSTLGPHSPEPLGVCRNPPPPCSPML---EASDDEALLVC

SPGPTWTQTGTHTTVLSPEDEDDVLLLP-LAEPEVWVV---EA-EDEPLLA-
ng q t as lgP e e ivv qipdpcl v1 EasdDEaLLlc

Figure 2

135

 

 

498 KLYSLRMFER RIFETQLEjRV EAEthREAP PMIAQG LIPPvEpgpy
548 QSPNQA'VLE NLRLAVRSQL GFTSVRLPMA GRs'vIWNRI FNFW

598 SLALVSADGD svv|-ol- EPERNHEHRS LF‘IVESDDTD EENERRDMAG

648 ASGGVAAPLP QKVPP'IEAVE ATVGACASS. To. RGGHAD NGRDvEsjvap

698 -| IPARHQL TSALSRMTQG LRWVRFELGR s. INQSP LRQLDNGVIG
748 REDDDDVEML IPI. .FII DVNDCSRPLL DLASDQGQGL RQPYNATNPG

798 VRPIVRDGPC ERCGIVHTAQ IPDTCLEVEL KNEIIpDEAL LLC

Figure 2 (cont’d)

136

The first LDLRA cluster contains two domains (146-182 and 195-236) and the
second cluster contains three domains (355-392, 393-430, and 431 -467). Each
LDLRA domains consists of ~40 residues, with six conserved cysteine residues
that form three intramolecular disulfide bonds and with conserved serine, aspartic
acid, and glutamic acid residues at the domain’s C-tenninus (Yamamoto et al.,
1984). All of the members of the LDLR superfamily contain LDLRA domain
repeats (reviewed by Hussain et al., 1999; Strickland et al., 2002). In many of
these proteins, such domains have been demonstrated to function as ligand
binding sites (Yamamoto et al., 1984). A single CUB domain (residues 28-140
and 240-353) precedes each LDLRA domain cluster (Figure 1). CUB domains
are composed of ~110 amino acids with four conserved cysteine residues that
form two intramolecular disulfide bonds and are believed to participate in protein-
protein interactions (reviewed by Bork and Beckmann, 1993; Christensen and
Birn, 2002). Exon 2 of ST7, which is lacking in the isoform used in our studies,
encodes protein sequence between the signal sequence peptide and the first
CUB domain. Therefore, loss of this exon's sequence does not affect any of the
functional domains present in ST7’s extracellular domain. The other ST7 isoform
simply contains 19 additional amino acids immediately upstream of the first CUB
domain.

The putative extracellular domain of ST7 is followed by a single
hydrophobic transmembrane helix flanked by positively charged residues
(residues 474-497) (Qing et al., 1999). The C-terminal cytoplasmic tail of the

protein (residues 498-840) contains several sequence motifs related to

137

 

endocytosis (Figure 2B). These endocytic signaling motifs include an YXXg
motif (YGQL, 530-533), a NPXY-like motif (EDFPVC, 543-548), and three
potential dileucine repeats (521-522, 776-777, and 837-839). In these consensus
sequences, X represents virtually any amino acid and Q represents any amino
acid with a bulky hydrophobic group (see Trowbridge et al., 1993 for a review of
endocytic signals). Several putative signaling motifs were also found. Four PDZ
domain binding motifs, (SIT )XV (SRV, 515-517; TSV, 570-572; TAV, 664-666;
and TSV, 693-995), and a WW-domain binding motif, PPXY (PPSY, 527-530),
are present (Figure ZB). Moreover, the cytoplasmic domain of ST7 is notably
rich in proline, serine, and threonine residues. Proline-rich regions are often
involved in protein-protein interactions (Kay et al., 2000). Because the
phosphorylation of residues in the tails of receptors can regulate their interactions
with intracellular proteins, we analyzed ST7’s cytoplasmic tail for putative serine,
threonine, and tyrosine phosphorylation sites using NetPhos 2.0 prediction
software (Blom et al., 1999), available at the ExPASy webserver. Twenty-five
serines and seven threonines were predicted to be potential sites of
phosphorylation (based on a score of ~0.5 or greater) (Figure 2B). Six of the 32
putative phosphorylation sites, three serines and three threonines, are located
within the consensus sequence for phosphorylation by PKC (Table II). Eight
sites, five serines and three threonines, are located within the consensus
sequence for phosphorylation by PKA (Table II). Serine and threonine residues
constituting three of the four PDZ domain-binding motifs, as well as the serine

residue present in the WW domain-binding motif, were also predicted to be

138

Table II

Putative PKC and PKA phosphorylation sites in ST7’s cytoplasmic domain

 

 

 

Position Sequence Kinase
509 FERR§FETQ PKA
581 AGRS§NIWN PKA
596 RSRH§GSLA PKA
615 PSQSISREP PKC
624 ERNHIHRSL PKA, PKC
680 SSTQ§TRGG PKC
693 GRDVISVEP PKA
724 WVRFILGRS PKA
730 GRSS§LSQN PKA
746 DNGV_S_GRED PKC
801 GVRP§NRDG PKA, PKC
826 CLEVILKNE PKC

 

139

potential phosphorylation sites. The three tyrosines present in ST7’s cytoplasmic
domain were not predicted to be phosphorylated. However, one of these
tyrosine residues is shared by the WW domain-binding motif and the YXXfl
internalization motif (PPS-Y-GQL), and several groups have reported
phosphorylation of tyrosines in the context of either motif (Klingmuller et al.,
1996; Sotgia et al., 2001; Macias et al., 2002).

Identification of proteins that interact with the cytoplasmic domain of the
ST7 protein—Because tumorigenicity studies conducted in this laboratory
suggest a functional role for ST7’s cytoplasmic tail in inhibiting tumor growth2 and
because our analysis of ST7’s cytoplasmic domain revealed several potential
endocytic and signal transduction motifs, we used the yeast-hybrid system to
identify proteins that interact with the ST7’s cytoplasmic C-terrninus (described
under “Experimental Procedures”). Of 57 cDNAs isolated, 30 different cDNAs
were identified. Six cDNAs, encoding proteins of the greatest interest, were
chosen for further studies. These are: 1) RACK1, a WD domain-containing
protein implicated as a key cellular scaffold protein that interacts with multiple
cellular receptors (Ron et al., 1994, 1999; Chang et al., 2001); 2) MIBP, an
integrin binding protein believed to participate in signal transduction pathways
regulating myogenesis (Li et al., 1999); 3) SARA, a FWE domain-containing
cellular scaffold protein that shuttles SMADZ and SMAD3 to the cytoplasmic
domains of TGFB receptors (T sukazaki et al., 1998); 4) o-actinin-Z, a
cytoskeletal protein thought to link plasma membrane receptors to the

cytoskeleton (Galliano et al., 2000); 5) myotilin, a novel cytoskeletal protein that

140

interacts with a-actinin-Z. Defects in myotilin cause limb-girdle muscular
dystrophy type 1A (Salmikangas et al., 1999; Hauser et al., 2000, 2002); and 6)
SNAPIN, a novel binding partner for the SNAP-25 component of the SNARE
vesicular transport complex (llardi et al., 1999). To verify that these cDNAs
encode the proteins responsible for the original interactions observed in our
yeast two-hybrid screen, AH109 yeast cells containing the ST7cyto1 bait plasmid
were transformed with each of the six prey plasmids, and the transformants were
plated on selection medium. Each of the proteins tested interacted with the
cytoplasmic domain of ST7, as detected by reporter gene activity.

Polymerase chain reaction (PCR) was used to subclone each of the six
prey cDNAs into the mammalian expression vector pcDNA6-V5lHis. An N-
terminal HA-epitope tag was included to facilitate detection of the proteins, and a
stop codon was included at the 3' end of each cDNA to exclude the V5 and His
epitope tags from the resulting proteins. Table III indicates the cDNAs isolated
from yeast. Those encoding only a portion of the prey protein are designated by
a subscript “p”. Because three of the four myotilin cDNAs isolated from yeast
contained the entire open reading frame and some 5’ upstream untranslated
sequence, the entire myotilin open reading frame was subcloned into the
expression vector. Sequencing of each insert to verify the fidelity of the PCR
reactions revealed that we had isolated the longer isoform of the MIBP cDNA
(GenBank accession number AK001663), which contains a 132 nucleotide in-
frame insertion between nucleotides 328 and 329 of the short MIBP coding

sequence (GenBank accession number NM_014446). The sequences of the two

141

Table III

Identification of proteins that interact with the cytoplasmic domain of ST7

To identify ST7-interacting proteins, the complete ST7 cytoplasmic domain was used
as bait in a yeast two-hybrid screen of a human skeletal muscle cDNA library. The
majority of positive cDNAs identified encode a partial prey protein, designated by a
subscript “p”. Column four lists the region of each prey protein encoded by isolated
cDNAs. Three of the four myotilin prey plasmids isolated contained a cDNA that includes

the entire open reading frame and some of the 5’ untranslated sequence (5’UTR)

 

 

Clones Independent Prey protein

Name isolated clones isolated size Function/pathway

MIBPp 1 1 92-COOH Signal transduction
65-COOH

RACK1p 2 2 Signal transduction
106-COOH

SARAp 1 1 731 -COOH Signal transduction
5'UTR-COOH

Myotilin 4 2 Actin cytoskeleton
250-COOH

or-Actinin-2p 1 1 307-COOH Actin cytoskeleton

Synaptic
SNAPI N, 2 1 1 7-COOH transmission

 

142

MIBP proteins are identical except for an additional 44 residues in the longer
form. Sequencing also demonstrated that a variant SARA cDNA had been
isolated. To date, three SARA transcript variants have been observed and are
reported in GenBank. Our partial cDNA, which lacks the 81 nucleotides of exon
14, represents a fourth SARA transcript variant.

Confirmation of the protein-protein interactions within human cells—
Protein-protein interactions identified using the yeast two-hybrid assay are only
considered valid if verified by other biochemical methods. Therefore, we used
co-immunoprecipitation to determine whether the six prey proteins identified
interact with ST7’s cytoplasmic domain in human cells. 293-T cells were
transiently co-transfected with plasmids encoding an HA-tagged partial prey
protein and ST7-V5lHis protein, and the HA-tagged prey protein was
immunoprecipitated from Iysates of these cells. The immunoprecipitation
products were analyzed by Western blotting with a monoclonal V5 antibody to
detect ST7-V5lHis and a monoclonal HA antibody to detect the HA-tagged prey
proteins. The ST7-V5lHis protein was co-immunoprecipitated with HA-RACK1p
(Figure 3A), HA-MlBPp (Figure 3B), and HA-SARAp (Figure 30), indicating an
interaction between these protein pairs. ST7-V5lHis failed to co-
immunoprecipitate with l-lA-o-actininzp, HA-SNAPle, and HA-myotilin. An
example is shown in Figure SD. Failure to detect an interaction between these
three prey proteins and ST7-V5lHis by co-immunoprecipitation does not exclude
these as ST7-interacting proteins because the co-immunoprecipitation approach

is less sensitive than the yeast two-hybrid assay.

143

Figure 3. ST7 interacts with RACK1 p, MIBPp, and SARAp in human cells. A.
Western blotting analysis of HA immunoprecipitation reactions from 293-T cells
transiently transfected with ST7-V5lHis, HA-RACKp, or both ST7-V5lHis and HA-
RACKp. The upper panel was probed with anti-V5 antibody; the lower panel was
probed with anti-HA antibody. Lanes designated with a minus (-) sign contain
negative control IP reactions, i.e., immunoprecipitation reactions without HA
antibody; lanes designated with a plus (+) sign contain immunoprecipitation
reactions with anti-HA; lanes designated as input contain 25 pg of the lysate
used in the immunoprecipitation reactions; lanes designated as blank were not
loaded. B. Same as A except cells were transfected with ST7-V5lHis, HA-MIBPp,
or both ST7-V5lHis and HA-MIBPp; C. Same as A except cells were transfected
with ST7-V5lHis, HA-SARAp, or both ST7-V5lHis and HA-SARAp; D. Same as A

except cells were transfected with both ST7-V5lHis and HA-d-actinin-Zp.

144

ills

ras

ain

ion

ate

not

 

 

 

 

 

«SW-V5lHis

 

 

 

«HA-RACK1,,

 

 

 

 

 

 

 

 

 

 

 

145

A
sn-verrs
Plasmids and
Tmnsfwt°d= sn-verrs HA-RAcm, HA-RACK1P
x .8 .8 .3
IP: HA HA 5 HA 5, ‘3; HA 5 g
- + E - + E s - + E 5
we: V5 «- .
WB: HA - - - -
B
sn-verrs
Plasmids and
Transfected: ST7-V5lHis HA-MIeP, HA-MreP,
x a X a 8
IP:HA HA 5 3 HA 5 3 HA 5 a
- + E E + E s - + E S.
we: vs "' .
we: HA - a. - 4.
Figure 3

4— ST7-V5lHis

«- HA-MIBPP

 

ST7-V5lHis
Plasmids

and
Transfected: ST7-V5lHis PIA-SARA, HA-SARAP

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

x. on x u .8 a
rP:HALA_§a-H__§3M§3
- + E 5 - + E 5 - + E 5
W8: V5 . u Jesw-vsrms
. .u‘
we: HA O _- . .<-HA-SARAP
ST7-V5lHis
Plasmids and
Transfected: HA-a-actinln-ZP
IP: HA HA_ i
- + 5
W8: v5 . «sn-veris
we: HA ~+HA-fl-actiflifl-2p
Figure 3 (cont’d)

146

The interaction of these proteins with ST7 may simply have been missed as the
result of being weaker or transient. Whether full-length RACK1 and MIBP
proteins interact with ST7 was determined as follows. cDNAs encoding
full-length RACK1 and MIBP were isolated from the MATCHMAKER skeletal
muscle cDNA library using PCR with appropriate oligonucleotides and subcloned
into pcDNA6-HA. The endogenous stop codon was retained in each full-length
cDNA so that the resulting proteins would not be tagged by the V5 and His
epitopes. We sequenced the HA-RACK1 and HA-MIBP cDNA inserts and
verified that the PCR reactions were faithful. 293-T cells were transiently co-
transfected with plasmids encoding HA-RACK1 or HA-MIBP and ST7-V5lHis,
and the HA-RACK1 or HA-MIBP protein was immunoprecipitated from lysates of
these cells. The immunoprecipitation products were analyzed by Western
blotting with a monoclonal V5 antibody and a monoclonal HA antibody. ST7-
V5/His was co-immunoprecipitated with both HA-RACK1 (Figure 4A) and HA-
MIBP (Figure 48), indicating that the full-length RACK1 and MIBP proteins
interact with ST7. The ability of full-length SARA to interact with ST7 has not yet
been determined because using PCR, we were unable to isolate a cDNA
containing the full-length variant SARA.

Identification of the RACK 1 , MIBP, and SARA binding domains within
ST7’s cytoplasmic domain—To identify the specific region within ST7's
cytoplasmic domain required for its interaction with RACK1, MIBP, and SARA,
we used the yeast two-hybrid assay with truncated forms of the ST7 cytoplasmic

domain as bait (Figure 5). AH109 strains expressing the truncated ST7Wto bait

147

Figure 4. ST7 interacts with full-length RACK1 and MIBP in human cells. A.
Western blotting analysis of HA immunoprecipitation reactions from 293-T cells
transiently transfected with ST7-V5lHis, HA-RACK, or both ST7-V5lHis and HA-
RACK. The upper panel was probed with anti-V5 antibody; the lower panel was
probed with anti-HA antibody. Lanes designated with a minus (-) sign contain
negative control lP reactions, i.e., immunoprecipitation reactions without HA
antibody; lanes designated with a plus (+) sign contain immunoprecipitation
reactions with anti-HA; lanes designated as input contain 25 pg of the lysate
used in the immunoprecipitation reactions; lanes designated as blank were not
loaded. B. Same as A except cells were transfected with ST7-V5lHis, HA-MIBP,

or both ST7-V5lHis and HA-MIBP.

148

ST7-V5IH is

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

and
A ST7-V5IHis HA-RACK1 m
Plasmids
Transfected:
X on! X a .2
IP: HA HA 5 3, HA 5 3 HA 5 g
- + a 5 - + E 5. - + E g-
WB: V5 - EST‘I—Vflflls
we: HA " "" -9 " «HA-RACK1
B
ST7-V5IHls
Plasmids and
Transfectod: ST7-V5IHis HA-MIBP HA-MIBP
x x .. x to
IP: HA HA 5 HA 5 3 HA 5 a
- + B - + E 5 - + E 5
WB: V5 «SW-VSIHts
“’3 ”A - " - db ~—HA—MIeP
Figure4

149

Figure 5. Illustration of the truncated ST7 cytoplasmic domain bait proteins
used to determine the location of the MIBP, SARA, and RACK1 binding
sites in ST7’s cytoplasmic domain. The appropriate bait-containing AH109
strains were transformed with an MIBPp, SARAp, or RACK1p prey plasmid. As a
negative control, the bait-containing AH109 strains were also transformed with a
prey plasmid encoding the SV40 large T-antigen, which does not bind to $17.
Transformants were plated on the most stringent medium, SDI-trpl-leul-hisl-ade.
Growth on this medium indicated that the bait and prey proteins tested were able

to interact within yeast cells.

150

lde.

able

Truncated ST7 Cytoplasmic Domains

317::th
3mm,
$176M,
3mm,
ST7cyeos
$176,,“

498_ 840
498_ 667

668— 840
498—625

498_560

561 _667

Figure 5

151

proteins were transformed with the pACT2 prey plasmids isolated in the original
yeast-two hybrid screen. As a negative control, we transformed these AH109
strains with a prey plasmid encoding the SV40 large T-antigen, which does not
interact with ST7. Transformants were plated on selective agar (SD/-
leul-trpl-hisl-ade). Failure of the transformants to grow indicated that the
truncated ST7Wt0 protein encoded by the bait plasmid failed to bind to the prey
protein. Table IV summarizes the results of these studies. As previously
determined, the ST7cyto1 bait protein (residues 498-840) was able to interact with
RACK1, MIBP, and SARA, but not with the negative control SV40 large T-
antigen. The ST7cyto1 bait was divided into an N-terrninal ST7¢mz fragment
(residues 498-667) and a C-tenninal ST7cytoa fragment (residues 668-840).
RACK1, MIBP, and SARA were each able to interact with Schtoz. As expected,
the negative control SV40 large T-antigen failed to interact with ST7cyr02. We
found that the ST7cyt03 fragment alone activated reporter gene expression in
AH109, thereby rendering it uninforrnative in these studies. Because our
proteomic analysis of ST7 indicated that the juxtamembrane region of the
cytoplasmic tail may constitute a functionally important domain, we further
divided the N-terminal ST7cyt02 fragment into three fragments, ST7cyto4 (residues
498-625), ST7cyms (residues 498-560), and ST7cyt05 (residues 561-625). Again,
the negative control SV40 large T-antigen protein did not interact with ST7cyro4,
Sch05, and Schos. RACK1, MIBP, and SARA each interacted with ST7¢yto4
and ST7cytos. Neither MIBP nor SARA interacted with the ST7cyt05, whereas

RACK1 was able to interact with it. ST7cytoe lacks the highly conserved residues

152

Table IV

Interaction of prey proteins with truncated ST7 bait proteins

Full-length and truncated ST7 cytoplasmic domains (shown in Figure 5) were used as bait to
identify the MIBP, SARA, and RACK1 binding sites in ST7. SV40 large T-antigen (SV40 T-Ag),
which does not interact with ST7’s cytoplasmic domain, served as the negative control. Plus (+)
signs indicate that an ST7 bait protein interacted with the prey protein tested; minus (-) signs
indicate that these proteins did not interact. NA, not applicable because the ST7¢ym bait protein

self-activated the yeast reporter genes.

 

 

 

 

 

 

 

 

BINEY MlBPp SARAP RACK1, SV40 T-Ag
ST7cyto1 + + + -
ST7cym + + + -
sncytoa NA NA NA NA
STEM + + + -
SUM + + + -
SWW - - + .

 

 

 

 

 

 

153

of the juxtamembrane region of ST7’s cytoplasmic domain. These results
demonstrate that the membrane-proximal 63 amino acids of $17 contain the
critical residues involved in its interaction with MIBP and SARA. Because
RACK1 interacted with all of the bait proteins spanning residues 498-667, the
critical residues for its interaction with ST7 must be contained in these truncated

proteins.

Having mapped the MIBP binding domain to the membrane proximal
region of ST7, co-immunoprecipitation studies using transiently co-transfected
293-T cells were carried out to determine whether MIPB interacts with additional
ST7 proteins truncated within its juxtamembrane region. Specifically, we
constructed expression plasmids encoding V5lHis-tagged ST7tr1, a truncated
protein containing ST7’s extracellular and transmembrane domains, but only the
first ten residues of its 343 amino acid cytoplasmic domain (residues 1-507), and
V5lHis-tagged ST7"; a truncated protein containing the ST7’s extracellular and
transmembrane domains, but only the first lysine of its cytoplasmic domain
(residues 1-498). Sequencing verified that the truncated ST7 cDNAs had been
cloned in-frame with the epitope tags. 293-T cells were transiently co-transfected
with plasmids encoding HA-tagged full-length MIBP and ST7-V5lHis, ST7tr1-
V5/His, or ST7tr2-V5/His. HA-MIBP was immunoprecipitated from lysates of the
transiently co-transfected cells. The immunoprecipitation products were
analyzed by Western blotting with a monoclonal V5 antibody to detect V5-tagged
ST7 proteins and a monoclonal HA antibody to HA-MIBP. Both ST7-V5lHis and

the ST7tr1-V5lHis were strongly co-immunoprecipitated with HA-MIBP (Figure 6),

154

indicating an interaction between these protein pairs. Somewhat surprising was
the finding, as shown in Figure 6, that the ST7tr2-V5/His protein was also co-
immunoprecipitated with HA-MIBP, although to a much lesser extent compared

with ST7-V5lHis and ST7tr1-V5lHis.

155

_ Figure 6. The first ten amino acids of the ST7 cytoplasmic domain may
suffice for MIBP binding. Western blotting analysis of HA immunoprecipitation
reactions from 293-T cells transiently transfected with ST7-V5lHis and HA-MlBP,
ST7tr1-V5lHis and HA-MIBP, or ST7trz-V5/His and HA-MIBP. The upper panel
was probed with anti-V5 antibody to detect both the full-length and truncated V5-
tagged ST7 proteins; the lower panel was probed with anti-HA antibody to detect
HA-MIBP. . Lanes designated with a minus (-) sign contain negative control lP
reactions, i.e., immunoprecipitation reactions without HA antibody; lanes
designated with a plus (+) sign contain immunoprecipitation reactions with anti-
HA; lanes designated as input contain 25 pg of the lysate used in the

immunoprecipitation reactions; lanes designated as blank were not loaded.

156

llmay

 

 

   

   

 

 

Jitation
-MlBP,
'panel
:ed V5
detect
itrol lP
. ST7-V5lHis ST7tr1-V5lHis SWfl-V5MI8
lanes Plasmids and and and
Transfected: HA-MIBP HA-MIBP HA-MIBP
. x -“ s “ '5
lhanli IP: HA HA 5 HA 5 ,1 HA 5 a
- + E - + T s - + E =
in the
. «sn-verrs
WB: V5
" sum-verrs
.m '~— or
sun-vsrms
WB: HA +HA-MIBP

 

 

 

 

Figure 6

157

 

DISCUSSION

Based on our analysis of ST7’s protein sequence, we conclude that ST7 is
a member of the low-density lipoprotein receptor (LDLR) superfamily. ST7
contains four critical structural motifs linking it to this superfamily: 1) an N-
terminal signal sequence; 2) multiple LDLRA domains arranged in clusters in the
extracellular region of the protein; 3) a single transmembrane domain; and 4) a
cytoplasmic tail containing motifs believed to function in endocytosis and/or
signaling. Moreover, the striking similarity between the structures of ST7, LRP3,
and murine LRP9, shown in Figures 1 and 2B, indicates that these three proteins
constitute an LDLR subfamily. Our findings agree with and strengthen those of
Sugiyama et al. (2000), who first suggested that these proteins comprise a
subfamily of the LDLR superfamily.

The extracellular region of every LDLR superfamily protein contains
various numbers of LDLRA domain repeats arranged in clusters. These domains
function in ligand binding (Yamamoto et al., 1984; reviewed by Hussain et al.,
1999; Strickland et al., 2002). CUB domains are found in a diverse array of
functionally unrelated extracellular proteins (reviewed by Bork and Beckmann,
1993). For example, the peripheral membrane protein cubilin, anchored in the
external side of the plasma membrane through its N-terminus, contains 27
extracellular CUB domains that function in ligand binding (reviewed by
Christensen and Birn, 2002). The LDLRA and CUB domains present in ST7 may
very likely enable ST7 to bind to extracellular ligands. At this point, no ST7

ligands have been identified, but ongoing work in our laboratory is directed

158

toward this goal. Members of the LDLR superfamily also contain endocytic
signals in their cytoplasmic domains. The presence of putative endocytic motifs
in ST7’s cytoplasmic tail (Figure 28) suggest that, like other members of the
LDLR superfamily, ST7 functions in endocytosis. In addition to putative
internalization signals, we found (see Figure 28 and Table II) that ST7’s
cytoplasmic domain contains several putative signaling motifs as well as sites of
potential phosphorylation. The presence of such motifs suggests that ST7 also
plays a role in signal transduction. This is in agreement with recent studies that
show that several other LDLR family proteins function in signal transduction
(Trommsdorff et al., 1998, 1999; Gotthardt et al., 2000; Wehrli et al., 2000; Tamai
et al., 2000; Pinson et al., 2000; Mac et al., 2001; Barnes et al., 2001; Boucher et
al., 2002; Loukinova et al., 2002; Lutz et al., 2002).

The data presented in Figures 3 and 4 further support the hypothesis that
the cytoplasmic domain of ST7, like other members of the LDLR superfamily,
functions in signal transduction. Both RACK1 and SARA have been shown to be
involved in signal transduction pathways. RACK1 was first identified as a
shuttling protein for activated protein kinase C (PKC) (Ron et al., 1994; 1999).
Because ST7’s cytoplasmic tail contains six potential PKC phosphorylation sites
(Table II), it may be that RACK1 shuttles activated PKC to ST7 to phosphorylate
ST7. RACK1 consists of seven WD domains that fold to form a B, propeller
structure (Wall et al., 1995). Such a structure may allow RACK1 to bind several
proteins simultaneously, thereby enabling it to function as a scaffold protein (Wall

et al., 1995). Consistent with this structure, RACK1 has been shown to interact

159

with a wide variety of proteins in addition to PKC. These include the non-
receptor tyrosine kinases Src and Fyn (Chang et al., 1998, 2001, 2002; Yaka et
al., 2002), the CAMP-specific phosphodiesterase PDE4D5 (Yarwood et al.,
1999), the Ras GTPase activating protein p120GAP (Koehler and Moran, 2001),
the angiotensin receptor associated protein AGTRAP (Wang et al., 2002), and
the signal transducer and activator of transcription 1 protein (Usacheva et al.,
2001). RACK1 also associates with the cytoplasmic domains of cell surface
receptors such as 81 integrins (Liliental and Chang, 1998), the insulin-like growth
factor I receptor (lGF-IR) (Hermanto et al., 2002; Kiely et al., 2002), the protein
tyrosine phosphatase PTPp (Mourton et al., 2001), and the NRZB subunit of the
NMDA receptor (Yaka et al., 2002). Because RACK1 interacts with a plethora of
cellular proteins, it is believed to play a key role in regulating a variety of cellular
signaling pathways (Chang et al., 2001).

The FWE domain-containing protein SARA (also known as MADH-
interacting protein, MADHIP) was first identified by Tsukazaki et al. (1998). They
found that SARA serves as a scaffold protein to link unphosphorylated SMAD2/3
to active TGFB receptor heterodimers. By localizing SMADZ/3 to the membrane,
SARA facilitates phosphorylation of these SMADs by the type I receptor’s kinase
domain (Tsukazaki et al., 1998). Phosphorylated SMAD2/3 dissociate from
SARA and associate with SMAD4 to translocate to the nucleus and regulate
gene expression. More recently, it was determined that SARA localizes to early
endosomes, suggesting that endocytosis plays a role in this SARA-mediated

signaling pathway (ltoh et al., 2002; Penheiter et al., 2002; Hayes et al., 2002).

160

 

Although little is known about MIBP, it too appears to be involved in signal
transduction. Li et al. (1999) first identified MIBP using a yeast two-hybrid assay
to screen for proteins able to bind to the cytoplasmic tails of B1-integrins. They
found that myocytes express a high level of MIBP prior to differentiation and a
low level of MIBP after differentiation and that overexpression of MIBP in
myocytes inhibits cell fusion and differentiation (Li et al., 1999). Therefore, MIBP
likely plays an important role in signal transduction pathways regulating myocyte
fusion and differentiation.

Our data showing that ST7’s cytoplasmic domain binds to proteins related
to signal transduction is consistent with the emerging role of LDLR superfamily
proteins as signal transducers (reviewed in Strickland et al., 1995; Herz, 2001,
2002; van der Geer, 2002). For example, Trommsdorff et al. (1998, 1999) and
D’Arcangelo et al. (1999) demonstrated that VLDLR and ApoER2 are co-
receptors for the extracellular protein Reelin. Binding of Reelin to these
receptors has two major consequences, the internalization of Reelin and the
activation of a tyrosine kinase signaling cascade. This results in the
phosphorylation of Dab1 protein bound to the cytoplasmic domains of these
receptors. Alterations in this pathway cause dysfunction in brain development
(Trommsdorff et al., 1999). Additionally, several groups demonstrated that
LRP5 and LRP6 participate in the Wnt signaling pathway (Wehrli et al., 2000;
Tamai et al., 2000; Pinson et al., 2000; Gong et al., 2001; Mac et al., 2001).
Both proteins serve as co-receptors for extracellular Wnt glycoproteins. Mao et

al. (2001) demonstrated that cytoplasmic domain of LRP5 binds to the

161

scaffolding protein axin to promote B-catenin stabilization and Wnt responsive
gene expression. Mice lacking LRP6 die at birth as a result of severe
developmental defects, including malformation of the skeleton, limbs, eyes, and
urogenital tract (Pinson et al., 2000). Mutations in the human LRP5 gene result
in defects in bone density and eye development (Gong et al., 2001; Little et al.,
2002; Boyden et al., 2002). Boucher et al. (2002) and Loukinova et al. (2002)
showed that LRP1 serves as a co-receptor for PDGF BB and that the adaptor
protein Shc binds to a phosphotyrosine residue in LRP1’s cytoplasmic domain
upon PDGF BB stimulation. Finally, yeast two-hybrid studies demonstrated that
the cytoplasmic tails of LRP1, LRP2, LDLR, VLDLR, and ApoER2 interact with
proteins known to be involved in MAP kinase signal transduction, synaptic
transmission regulation, cytoskeletal organization, cellular adhesion, and
endocytosis, adding further support to the idea that members of the LDLR
superfamily participate not only in endocytosis but also in signaling (Gotthardt et
al., 2000). The ST7-interacting proteins described in this paper can be classified
into many of the same categories as those that interact with other LDLR
superfamily members.

Our yeast two-hybrid studies also demonstrated that both SARA and
MIBP bind to the juxtamembrane region of ST7’s cytoplasmic domain (Figure 5
and Table IV). Because the bait construct containing ST7 residues 668-840 was
able to self-activate the reporter genes, we cannot exclude the possibility that this
region contains additional MIBP and SARA binding sites. However, the presence

of MIBP and SARA binding sites in the juxtamembrane region of ST7 supports

162

our hypothesis that this region, which is highly conserved among the ST7
subfamily proteins, is important for ST7 function. Moreover, we determined that
the first ten amino acids of ST7’s cytoplasmic domain may suffice for MIBP
binding (Figure 6). Based on signal intensity, we conclude that the MIBP binds
equally well to ST7 and ST7tr1. Somewhat surprising was the fact that a small
amount of ST7t,2 associated with MlBP. Because the cytoplasmic tail of this
truncated protein contains only one residue, we did not expect it to interact with
MIBP. It is always possible that the remaining lysine residue plays a critical role
in the ST7-MIBP interaction, and by itself is able to interact with ST7, but does so
much less efficiently. Another possible explanation is that residues of the
transmembrane helix domain become exposed because of a conformational
change when ST7 is active and participate in the ST7-MIBP interaction. The
growth medium used in these studies, which included 10% fetal bovine serum,
may contain ligands that bind to ST7’s extracellular domain to activate the
receptor and induce such conformational changes. Studies of integrin
cytoplasmic domains suggest that ligand binding does induce conformational
changes that affect the position of residues within and near the membrane
(Armulik et al., 1999; Lu et al., 2001). Because the cells used in our co-
immunoprecipitation experiments contain endogenous ST7 protein, it is possible
that the latter protein was responsible for making it appear as if MIBP interacts
with ST7"; If dimerization of endogenous ST7 with ST7"; were to occur, the
observed interaction could have resulted from MIBP associating with the full-

length cytoplasmic tail of endogenous ST7, instead of the truncated cytoplasmic

163

 

tail. Therefore, immunoprecipitation of MIBP would cause both endogenous and
truncated ST7 to be co-precipitated. We prefer this possibility as the most likely
explanation of our results with ST7"; lf dimerization explained that interaction,
then it is possible that dimerization also explains the interaction between ST7tr1
and MIBP. However, if it is responsible for both results, then we would have
expected to see an equivalent amount of 817m and ST7tr2 pulled-down by MIBP.
This was not the case (Figure 6).

ST7 contains at least one RACK1 binding site in the region spanning
residues 498-667 (Table IV and Figure 5). If there is only one binding site for
RACK1 in this region, it must fall between residues 498-625 because these
residues enabled RACK1 to associate with ST7. Moreover, the observation that
RACK1 interacts with both Sme and ST7m06, two bait proteins that do not
include any common residues, suggests that each of these proteins contain part
of the RACK1 binding site. We cannot exclude the possibility that there is more
than one RACK1 binding site in this region of ST7’s cytoplasmic tail, and each
bait protein contains one or more complete sites. Neither can our results rule out
the possibility that residues 668-840 of ST7 contain one or more RACK1 binding

sites.

164

REFERENCES

Arrnulik, A., Nilsson, |., von Heijne, G., and Johansson, S. (1999). Determination
of the border between the transmembrane and cytoplasmic domains of human
integrin subunits. J Biol Chem 274, 37030-37034.

Barnes, H., Larsen, 8., Tyers, M., and van Der Geer, P. (2001). Tyrosine-
phosphorylated low density lipoprotein receptor-related protein 1 (Lrp1)
associates with the adaptor protein SHC in SRC-transforrned cells. J Biol Chem
276, 19119-19125.

Blom, N., Gammeltoft, S., and Brunak, S. (1999). Sequence and structure-based
prediction of eukaryotic protein phosphorylation sites. J Mol Biol 294, 1351-1362.

Bork, P., and Beckmann, G. (1993). The CUB domain. A widespread module in
developmentally regulated proteins. J Mol Biol 231, 539-545.

Boucher, F., Liu, P., Gotthardt, M., Hiesberger, T., Anderson, R. G., and Herz, J.
(2002). Platelet-derived growth factor mediates tyrosine phosphorylation of the
cytoplasmic domain of the low Density lipoprotein receptor-related protein in
caveolae. J Biol Chem 277, 15507-15513.

Boyden, L. M., Mao, J., Belsky, J., Mitzner, L., Farhi, A., Mitnick, M. A., Wu, D.,
lnsogna, K., and Lifton, R. P. (2002). High bone density due to a mutation in
LDL-receptor-related protein 5. N Engl J Med 346, 1513-1521.

Chang, B. Y., Chiang, M., and Cartwright, C. A. (2001). The interaction of Src
and RACK1 is enhanced by activation of protein kinase C and tyrosine
phosphorylation of RACK1. J Biol Chem 276, 20346-20356.

Chang, B. Y., Conroy, K. B., Machleder, E. M., and Cartwright, C. A. (1998).
RACK1, a receptor for activated C kinase and a homolog of the beta subunit of G
proteins, inhibits activity of src tyrosine kinases and growth of NIH 3T3 cells. Mol
Cell Biol 18, 3245-3256.

Chang, B. Y., Harte, R. A., and Cartwright, C. A. (2002). RACK1: a novel
substrate for the Src protein-tyrosine kinase. Oncogene 21, 7619-7629.

Christensen, E. |., and Birn, H. (2002). Megalin and cubilin: multifunctional
endocytic receptors. Nat Rev Mol Cell Biol 3, 256-266.

Claros, M. G., Brunak, S., and von Heijne, G. (1997). Prediction of N-terminal
protein sorting signals. Curr Opin Struct Biol 7, 394-398.

D'Arcangelo, G., Homayouni, R., Keshvara, L., Rice, D. S., Sheldon, M., and
Curran, T. (1999). Reelin is a ligand for lipoprotein receptors. Neuron 24, 471-
479.

165

 

Galliano, M. F., Huet, C., Frygelius, J., Polgren, A., Wewer, U. M., and Engvall,
E. (2000). Binding of ADAM12, a marker of skeletal muscle regeneration, to the
muscle-specific actin-binding protein, alpha -actinin-2, is required for myoblast
fusion. J Biol Chem 275, 13933-13939.

Gattiker A., Gasteiger E. and Bairoch A. (2002) ScanProsite: a reference
implementation of a PROSITE scanning tool. Applied Bioinfonnatics 12107-108.

Gong, Y., Slee, R. B., Fukai, N., Rawadi, G., Roman-Roman, S., Reginato, A. M.,
Wang, H., Cundy, T., Glorieux, F. H., Lev, D., etal. (2001). LDL receptor-related
protein 5 (LRP5) affects bone accrual and eye development. Cell 107, 513-523.

Gotthardt, M., Trommsdorff, M., Nevitt, M. F., Shelton, J., Richardson, J. A.,
Stockinger, W., Nimpf, J., and Herz, J. (2000). Interactions of the low density
lipoprotein receptor gene family with cytosolic adaptor and scaffold proteins

suggest diverse biological functions in cellular communication and signal
transduction. J Biol Chem 275, 25616-25624.

Hauser, M. A., Conde, C. B., Kowaljow, V., Zeppa, G., Taratuto, A. L., Torian, U.
M., Vance, J., Pericak-Vance, M. A., Speer, M. C., and Rosa, A. L. (2002).
myotilin Mutation Found in Second Pedigree with LGMD1A. Am J Hum Genet 71.

Hauser, M. A., Horrigan, S. K., Salmikangas, P., Torian, U. M., Viles, K. D.,
Dancel, R., Tim, R. W., Taivainen, A., Bartoloni, L., Gilchrist, J. M., et al. (2000).
Myotilin is mutated in limb girdle muscular dystrophy 1A. Hum Mol Genet 9,
2141-2147.

Hayes, 8., Chawla, A., and Corvera, S. (2002). TGFbeta receptor internalization
into EEA1-enriched early endosomes: role in signaling to Smad2. J Cell Biol 158,
1239-1249.

Herrnanto, U., Zong, C. S., Li, W., and Wang, L. H. (2002). RACK1, an insulin-
like growth factor I (lGF-l) receptor-interacting protein, modulates lGF-l-
dependent integrin signaling and promotes cell spreading and contact with
extracellular matrix. Mol Cell Biol 22, 2345-2365.

Herz, J., and Bock, H. H. (2002). Lipoprotein receptors in the nervous system.
Annu Rev Biochem 71 , 405-434.

Herz, J., and Strickland, D. K. (2001). LRP: a multifunctional scavenger and
signaling receptor. J Clin Invest 108, 779-784.

Hussain, M. M., Strickland, D. K., and Bakillah, A. (1999). The mammalian low-
density lipoprotein receptor family. Annu Rev Nutr 19, 141-172.

llardi, J. M., Mochida, S., and Sheng, 2. H. (1999). Snapin: a SNARE-associated
protein implicated in synaptic transmission. Nat Neurosci 2, 119-124.

166

 

Ishii, H., Kim, D. H., Fujita, T., Endo, Y., Saeki, S., and Yamamoto, T. T. (1998).
cDNA cloning of a new low-density lipoprotein receptor-related protein and
mapping of its gene (LRP3) to chromosome bands 19q12-q13. 2. Genomics 51 ,
132-135.

ltoh, F., Divecha, N., Brocks, L., Oomen, L., Janssen, H., Calafat, J., ltoh, S., and
Dijke Pt, P. (2002). The FYVE domain in Smad anchor for receptor activation
(SARA) is sufficient for localization of SARA in early endosomes and regulates
TGF-beta/Smad signalling. Genes Cells 7, 321-331.

Kay, B. K., Williamson, M. P., and Sudol, M. (2000). The importance of being
proline: the interaction of proline-rich motifs in signaling proteins with their
cognate domains. Faseb J 14, 231-241.

Kiely, P. A., Sant, A., and O'Connor, R. (2002). RACK1 is an insulin-like growth
factor 1 (lGF-1) receptor-interacting protein that can regulate lGF-1-mediated Akt
activation and protection from cell death. J Biol Chem 277, 22581-22589.

Klingmuller, U., Bergelson, S., Hsiao, J. G., and Lodish, H. F. (1996). Multiple
tyrosine residues in the cytosolic domain of the erythropoietin receptor promote
activation of STAT5. Proc Natl Acad Sci U S A 93, 8324-8328.

Koehler, J. A., and Moran, M. F. (2001). RACK1, a protein kinase C scaffolding
protein, interacts with the PH domain of p120GAP. Biochem Biophys Res
Commun 283, 888-895.

Krieger, M., and Herz, J. (1994). Structures and functions of multiligand
lipoprotein receptors: macrophage scavenger receptors and LDL receptor-related
protein (LRP). Annu Rev Biochem 63, 601-637.

Li, J., Mayne, R., and Wu, C. (1999). A novel muscle-specific beta 1 integrin
binding protein (MIBP) that modulates myogenic differentiation. J Cell Biol 147,
1391-1398.

Li, Y., Cam, J., and Bu, G. (2001). Low-density lipoprotein receptor family:
endocytosis and signal transduction. Mol Neurobiol 23, 53-67.

Liliental, J., and Chang, D. D. (1998). Rack1, a receptor for activated protein
kinase C, interacts with integrin beta subunit. J Biol Chem 273, 2379-2383.

Little, R. D., Carulli, J. P., Del Mastro, R. G., Dupuis, J., Osborne, M., Folz, C.,
Manning, S. P., Swain, P. M., Zhao, S. C., Eustace, B., etal. (2002). A mutation
in the LDL receptor-related protein 5 gene results in the autosomal dominant
high-bone-mass trait. Am J Hum Genet 70, 11-19.

Loukinova, E., Ranganathan, S., Kuznetsov, S., Gorlatova, N., Migliorini, M. M.,

Loukinov, D., Ulery, P. G., Mikhailenko, l., Lawrence, D. A., and Strickland, D. K.
(2002). Platelet-derived growth factor (PDGF)-induced tyrosine phosphorylation

167

 

of the low density lipoprotein receptor-related protein (LRP). Evidence for
integrated co-receptor function betwenn LRP and the PDGF. J Biol Chem 277,
15499-15506.

Lu, C., Takagi, J., and Springer, T. A. (2001). Association of the membrane
proximal regions of the alpha and beta subunit cytoplasmic domains constrains
an integrin in the inactive state. J Biol Chem 276, 14642-14648.

Lutz, C., Nimpf, J., Jenny, M., Boecklinger, K., Enzinger, C., Utermann, G., Baier-
Bitterlich, G., and Baier, G. (2002). Evidence of Functional Modulation of the
MEKKIJNK/cJun Signaling Cascade by the Low Density Lipoprotein Receptor-
related Protein (LRP). J Biol Chem 277, 43143-43151.

Macias, M. J., Wiesner, S., and Sudol, M. (2002). W and SH3 domains, two
different scaffolds to recognize proline-rich ligands. FEBS Lett 513, 30-37.

Mao, J., Wang, J., Liu, 8., Pan, W., Farr, G. H., 3rd, Flynn, C., Yuan, H., Takada,
S., Kimelman, D., Li, L., and Wu, D. (2001). Low—density lipoprotein receptor-
related protein-5 binds to Axin and regulates the canonical Wnt signaling
pathway. Mol Cell 7, 801-809.

McCormick, J. J., and Maher, V. M. (1996). Analysis of the multistep nature of
human carcinogenesis utilizing human fibroblasts. Radiat. Oncol. Invest. 3, 387-
391.

Mourton, T., Hellberg, C. B., Burden-Gulley, S. M., Hinman, J., Rhee, A., and
Brady-Kalnay, S. M. (2001). The PTPmu protein-tyrosine phosphatase binds and
recruits the scaffolding protein RACK1 to cell-cell contacts. J Biol Chem 276,
14896-14901 .

Nielsen, H. Engelbrecht, J., Brunak S., and von Heijne, G. (1997) Identification of
prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites
Protein Engineering 10, 1-6.

Penheiter, S. G., Mitchell, H., Garamszegi, N., Edens, M., Dore, J. J., Jr., and
Leof, E. B. (2002). lnternalization-dependent and -independent requirements for

transforming growth factor beta receptor signaling via the Smad pathway. Mol
Cell Biol 22, 4750-4759.

Pinson, K. l., Brennan, J., Monkley, S., Avery, B. J., and Skarnes, W. C. (2000).
An LDL-receptor-related protein mediates Wnt signalling in mice. Nature 407,
535-538.

Qing, J., Wei, D., Maher, V. M., and McCormick, J. J. (1999). Cloning and
characterization of a novel gene encoding a putative transmembrane protein with

altered expression in some human transformed and tumor-derived cell lines.
Oncogene 18, 335-342.

168

 

Ron, D., Chen, C. H., Caldwell, J., Jamieson, L., Orr, E., and Mochly-Rosen, D.
(1994). Cloning of an intracellular receptor for protein kinase C: a homolog of the
beta subunit of G proteins. Proc Natl Acad Sci U S A 91 , 839-843.

Ron, D., Jiang, Z., Yao, L., Vagts, A., Diamond, I, and Gordon, A. (1999).
Coordinated movement of RACK1 with activated betallPKC. J Biol Chem 274,
27039-27046.

Salmikangas, P., Mykkanen, O. M., Gronholm, M., Heiska, L., Kere, J., and
Carpen, O. (1999). Myotilin, a novel sarcomeric protein with two lg-like domains,
is encoded by a candidate gene for limb-girdle muscular dystrophy. Hum Mol
Genet 8, 1329-1336.

Sotgia, F., Lee, H., Bedford, M. T., Petrucci, T., Sudol, M., and Lisanti, M. P.
(2001). Tyrosine phosphorylation of beta-dystroglycan at its WW domain binding
motif, PPxY, recruits SH2 domain containing proteins. Biochemistry 40, 14585-
14592.

Strickland, D. K., Gonias, S. L., and Argraves, W. S. (2002). Diverse roles for the
LDL receptor family. Trends Endocrinol Metab 13, 66-74.

Strickland, D. K., Kounnas, M. Z., and Argraves, W. S. (1995). LDL receptor-
related protein: a multiligand receptor for lipoprotein and proteinase catabolism.
Faseb J 9, 890-898.

Sugiyama, T., Kumagai, H., Morikawa, Y., Wada, Y., Sugiyama, A., Yasuda, K.,
Yokoi, N., Tamura, S., Kojima, T., Nosaka, T., et al. (2000). A novel low-density
lipoprotein receptor-related protein mediating cellular uptake of apolipoprotein E-
enriched beta-VLDL in vitro. Biochemistry 39, 15817-15825.

Tamai, K., Semenov, M., Kato, Y., Spokony, R., Liu, C., Katsuyama, Y., Hess, F.,
Saint-Jeannet, J. P., and He, X. (2000). LDL-receptor-related proteins in Wnt
signal transduction. Nature 407, 530-535.

Trommsdorff, M., Borg, J. P., Margolis, B., and Herz, J. (1998). Interaction of
cytosolic adaptor proteins with neuronal apolipoprotein E receptors and the
amyloid precursor protein. J Biol Chem 273, 33556-33560.

Trommsdorff, M., Gotthardt, M., Hiesberger, T., Shelton, J., Stockinger, W.,
Nimpf, J., Hammer, R. E., Richardson, J. A., and Herz, J. (1999).
Reeler/Disabled-like disruption of neuronal migration in knockout mice lacking
the VLDL receptor and ApoE receptor 2. Cell 97, 689-701.

Trowbridge, l. S., Collawn, J. F., and Hopkins, C. R. (1993). Signal-dependent
membrane protein trafficking in the endocytic pathway. Annu Rev Cell Biol 9,
129-161.

169

Tsukazaki, T., Chiang, T. A., Davison, A. F., Attisano, L., and Wrana, J. L.
(1998). SARA, a FWE domain protein that recruits Smad2 to the TGFbeta .
receptor. Cell 95, 779-791.

Usacheva, A., Smith, R., Minshall, R., Baida, G., Seng, S., Croze, E., and
Colamonici, O. (2001). The WD motif-containing protein receptor for activated
protein kinase C (RACK1) is required for recruitment and activation of signal
transducer and activator of transcription 1 through the type I interferon receptor. J
Biol Chem 276, 22948-22953.

van der Geer, P. (2002). Phosphorylation of LRP1: regulation of transport and
signal transduction. Trends Cardiovasc Med 12, 160-165.

Wall, M. A., Coleman, D. E., Lee, E., lniguez-Lluhi, J. A., Posner, B. A., Gilman,
A. G., and Sprang, S. R. (1995). The structure of the G protein heterotrimer Gi
alpha 1 beta 1 gamma 2. Cell 83, 1047-1058.

Wang, W., Huang, Y., Zhou, Z., Tang, R., Zhao, W., Zeng, L., Xu, M., Cheng, C.,
Gu, S., Ying, K., et al. (2002). Identification and characterization of AGTRAP, a
human homolog of murine Angiotensin II Receptor-Associated Protein (Agtrap).
Int J Biochem Cell Biol 34, 93-102.

Wehrli, M., Dougan, S. T., Caldwell, K., O'Keefe, L., Schwartz, 8., Vaizel-
Ohayon, D., Schejter, E., Tomlinson, A., and DiNardo, S. (2000). arrow encodes
an LDL-receptor-related protein essential for Wingless signalling. Nature 407,
527-530.

Yaka, R., Thornton, C., Vagts, A. J., Phamluong, K., Bonci, A., and Ron, D.
(2002). NMDA receptor function is regulated by the inhibitory scaffolding protein,
RACK1. Proc Natl Acad Sci U S A 99, 5710-5715.

Yakabi, K., Arimura, T., Koyanagi, M., Uehigashi, Y., R0, 8., Minagawa, Y., and
Nakamura, T. (2002). Effects of interleukin-8 and Helicobacter pylori on
histamine release from isolated canine gastric mucosal mast cells. J
Gastroenterol 37, 10-16.

Yamamoto, T., Davis, C. G., Brown, M. S., Schneider, W. J., Casey, M. L.,
Goldstein, J. L., and Russell, D. W. (1984). The human LDL receptor: a cysteine-
rich protein with multiple Alu sequences in its mRNA. Cell 39, 27-38.

Yarwood, S. J., Steele, M. R., Scotland, G., Houslay, M. D., and Bolger, G. B.
(1999). The RACK1 signaling scaffold protein selectively interacts with the
CAMP-specific phosphodiesterase PDE4D5 isoform. J Biol Chem 274, 14909-
14917.

170

Appendix

Expression of ST7 Protein Inhibits the Tumor-Forming Ability of a

Human Fibrosarcoma-derived Cell Line1

Michele A. Battle, Sandra O’Reilly, Thomas Mullaney, Veronica M. Maher, and J.

Justin McCormick2

Carcinogenesis Laboratory, Cell and Molecular Biology Program, Department of
Microbiology and Molecular Genetics and Department of Biochemistry and
Molecular Biology, Michigan State University, East Lansing, Michigan 48824-
1302

1 This work was supported by United States Department of Health and Human
Services grants AG11026, CA82885, and CA097262.

2 To whom requests for reprints should be addressed at Carcinogenesis
Laboratory, Food Safety and Toxicology Building, Michigan State University, East
Lansing, MI 48824-1302. Phone (517) 353-7785; Fax (517) 353-9004; E-mail:

mccormi1@msu.edu.

171

3 Manuscript in preparation, Battle, M.A., Maher, V.M., and McCormick, J.J. ST7
is a novel member of the LDLR superfamily with a cytoplasmic tail that interacts
with intracellular proteins related to signal transduction pathways.

4 The abbreviations used are: LDLR, low-density lipoprotein receptor; LRP,
lipoprotein receptor-related protein; LDLRA, low-density lipoprotein receptor type
A; CUB, Complement factor C1s/C1r, Urchin embryonic growth factor, Bone

morphogenetic protein; RACK1, receptor for activated protein kinase C.

172

 

ABSTRACT

Previous studies using differential mRNA display to compare the infinite life
span, non-tumorigenic human fibroblast cell strain MSU-1.1 to a carcinogen
transformed, malignant MSU-1.1 derivative identified a novel gene, ST7.
Compared to its parental strain, expression of ST7 was strongly down-regulated
in the tumorigenic cells. A low level of ST7 expression was also observed in
additional malignant human cell lines. Subsequent studies revealed that ST7 is a
novel member of the low-density lipoprotein receptor (LDLR) superfamily, several
proteins of which have been shown to participate in both endocytosis and signal
transduction. To determine if ST7 acts as a tumor suppressor, the human
fibrosarcoma-derived malignant cell line SHAC was transfected with plasmids
encoding V5lHis-tagged full-length ST7 or V5lHis-tagged truncated ST7, which
contains its extracellular and transmembrane domains, but only the first 10 of the
343 amino acids of its cytoplasmic domain. The tumorigenicity of two
independent full-length ST7-V5lHis-expressing transfectants, two independent
transfectants expressing the V5lHis-tagged truncated form of ST7, two
independent vector-control transfectants, and two independent non-ST7-
expressing control transfectants was assessed by subcutaneous injection of
these cell strains into athymic mice. The results showed that overexpression of
full-length ST7 inhibits the tumor-forming ability of SHAC cells. Tumors
developed at a significantly lower frequency than that of the control cell
lines/strains. In contrast, overexpression of the truncated form of ST7 failed to

inhibit the tumor-forming ability of SHAC cells, supporting the hypothesis that

173

 

$17 is a tumor suppressor and that its cytoplasmic domain plays a role in the

protein's function as a suppressor.

174

 

INTRODUCTION

It is generally thought that tumors originate from a single cell that has acquired
all of the properties necessary to become malignant, and replication of that cell
gives rise to all of the cells of the tumor. Theories of carcinogenesis must
provide an understanding of how the original tumor cell gained the necessary
genetic and/or epigenetic changes to become malignant. At least five such
changes are estimated to be required. To determine the number and nature of
these changes, McCormick and his colleagues (reviewed by McCormick and
Maher, 1996) established a model system in which normal human fibroblasts in
culture can be transformed into malignant fibroblasts through the acquisition of a
series of genetic changes, each providing a growth advantage that enables
sequential clonal expansion. Using this model system, McCormick and his
colleagues (Qing et al., 1997) carried out differential mRNA display to compare
an infinite life span, non-tumorigenic human fibroblast cell strain, designated
MSU-1.1, to a carcinogen-transformed, malignant derivative of MSU-1.1 cells.
This comparison identified a novel putative tumor suppressor gene, ST7, which
was predicted to encode a transmembrane protein (Qing et al., 1999). ST7
expression in the tumorigenic cell line was significantly lower than that of the
non-tumorigenic parental cell strain. Moreover, several other malignant human
cell lines have a low level of ST7, suggesting that ST7 deficiency is a common
characteristic of tumorigenic cells. Recent studies in this laboratory3 revealed
that ST7 is a member of a subfamily of the LDLR4 superfamily of plasma

membrane receptors. Its extracellular domain contains an N-terrninal signal

175

 

sequence to target it to the membrane, five LDLRA domains, and two CUB
domains. lts cytoplasmic tail, which is notably rich in serine, threonine, and
proline, contains numerous putative phosphorylation sites and several motifs
implicated in endocytosis and signal transduction.

Another member of the LDLR superfamily, LRP1B, has been identified as a
candidate tumor suppressor gene. Liu et al. (2000a, 2000b) showed that the
LRP1B gene is inactivated in ~45% of non-small cell lung cancer cell lines
studied. They detected homozygous deletions within the LRP1B gene, and point
mutations in the LRP1B alleles. Finally, some of the cell lines studied expressed
abnormal truncated LRP1B transcripts. In contrast, none of the small cell lung
cancer cell lines studied had any observable defects in the LRP1B sequence,
and only one such line contained abnormal LRP1B transcripts. These findings
strongly suggest that inactivation of LRP1B plays an important in the
development of non-small cell lung cancer, but not in the development of small
cell lung cancer. Most recently, Langbein et al. (2002) also documented
alterations of the LRP1B gene in high grade urothelial cancer. In light of these
data and of the earlier study of Qing et al. (1999) showing that ST7 expression is
low in tumorigenic human fibroblasts, we tested the hypothesis that ST7
functions to suppress tumorigenicity. A tumorigenic cell line derived from a
patient’s fibrosarcoma, designated SHAC, was transfected with a plasmid
encoding full-length ST7 or an ST7 protein with a severely truncated cytoplasmic
domain, i.e., it contains ST7’s extracellular and transmembrane domains, but

only the first 10 of the 343 amino acids that comprise its cytoplasmic domain.

176

 

The tumorigenicity of independent transfectants expressing these proteins as
well as of independent vector-control transfectants and non-S‘I7-expressing
control transfectants was assessed by subcutaneous injection of these various
cell strains into athymic mice. Tumors formed at a significantly lower frequency
in athymic mice injected with SHAC cells expressing full-length ST7 than in
athymic mice injected with control cell strains. Any tumors that developed did so
with a greater latency. These data support the hypothesis that ST7 suppresses
the tumor-forming ability of SHAC cells. Furthermore, expression of a truncated
form of ST7 failed to inhibit the tumor-forming ability of SHAC cells, suggesting
that the cytoplasmic domain of ST7 plays an important role in the protein's

function as a tumor suppressor.

177

 

MATERIALS AND METHODS

Materials. The sources of specific materials used are: supplemented calf
serum and fetal calf serum, Hyclone; restriction enzymes and T4 DNA ligase,
New England Biolabs or lnvitrogen; shrimp alkaline phosphatase, Promega; Pfu
polymerase, Stratagene; oligonucleotides, the Michigan State University
Macromolecular Structure, Sequencing, and Synthesis Facility; protease and
phosphatase inhibitors, Sigma; SeeBlue® Plus 2 prestained protein molecular
weight standards, lnvitrogen; polyclonal ST7 antibody and monoclonal Na‘lK”
ATPase antibody, Affinity BioReagents; monoclonal V5 antibody, lnvitrogen;
monoclonal actin antibody, Sigma; and horseradish peroxidase-conjugated
secondary antibodies, Santa Cruz Biotechnology or Sigma.

Cells and cell culture conditions. Diploid human fibroblast cell lines SL89
and LG1 were derived from the foreskin of normal neonates. They were cultured
in Eagle’s minimal essential medium containing 0.2 mM L-aspartic acid, 0.2 mM
L-serine, 1.0 mM sodium pyruvate, 10% supplemented calf serum, penicillin (100
units/ml), and streptomycin (100 pg/ml) (growth medium) in a 37°C humidified
incubator with 5% C02 in air. SHAC cells, a cell line derived from a patient’s
fibrosarcoma, and all SHAC-derived cell strains were similarly cultured in growth
medium, supplemented with 1 pg/ml hydrocortisone. The human embryonic
kidney-derived cell line HEK 293-T was similarly cultured in high-glucose
Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf serum

and containing penicillin and streptomycin.

178

 

Stable transfection SHAC cells. Lipofectamine (lnvitrogen) was used
according to the manufacturer’s instructions. Blasticidin (10 pglml) (lnvitrogen)
was used to select stably transfected independent clonal populations.

Preparation of cell lysates. Cells were washed in cold phosphate-buffered
saline, collected in lysis buffer (50 mM Tris-HCI, pH 7.2, 150 mM NaCl, 50 mM
NaF, 0.5% NP-40, 1 mM Na3VO4, 1 mM benzamidine, 1 mM
phenylmethylsulfonyl fluoride, 25 pg/ml aprotinin, and 25 pg/ml leupeptin). Cells
in lysis buffer were incubated on ice for 30 min. Lysates were clarified by
centrifugation (25,000 x g) for 30 min at 4°C. The concentration of protein in
each lysate was determined using the Coomassie protein assay reagent (Pierce)
following the manufacturer’s instructions, and aliquots of each lysate containing
50 pg of protein were denatured in 5X Laemelli sample buffer (0.125 M Tris-HCI,
pH 6.8, 25% glycerol, 5% sodium dodecyl sulfate, 0.0125% bromophenol blue,
and 25% B—mercaptoethanol as the reducing agent).

Preparation of plasma-membrane enriched fractions. Cells were washed
in phosphate-buffered saline, collected in hypotonic lysis buffer (50 mM Tris-HCI,
pH 7.4, 35 mM NaF, 5 mM MgCl2, 1mM EGTA, 1 mM N33VO4, 1 mM sodium
pyrophosphate, 1 mM phenylmethylsulfonyl fluoride, 25 pg/ml aprotinin), and
incubated on ice for 10 min. Cells were homogenized by 30 strokes in a dounce
homogenizer. Homogenates were centrifuged (500 x g) for 5 min at 4°C.
Supernatants were centrifuged (16,900 x g) for 1 h at 4°C. Pellets, which are
enriched with plasma-membrane proteins, were resuspended in RIPA buffer (50

mM Tris-HCI, pH 7.5, 150 mM NaCl, 1% Triton X-100, 0.1% sodium dodecyl

179

 

sulfate, 0.5% sodium deoxycholate, 1mM EDTA, 1 mM Na3VO4, 2 mM
phenylmethylsulfonyl fluoride, and 25 pg/ml aprotinin). To obtain cytosolic,
soluble fractions, the supernatants were clarified by centrifugation (100,000 x g)
for 2 h at 4°C. The concentration of protein in each fraction was determined
using a bicinchoninic acid (BCA) protein assay reagent kit (Pierce) following the
manufacturer’s instructions. Aliquots of each plasma membrane-enriched
fraction and cytosolic, soluble fraction containing 20 pg of protein were denatured
in 5X Laemelli sample buffer.

Western blotting analysis. Denatured proteins were separated by sodium
dodecyl sulfate polyacrylamide gel electrophoresis and transferred to Immobilon-
P membrane (Millipore). The membranes were blocked with Tris-buffered saline
containing 0.1% Tween-20 and 5% non-fat milk (blocking solution) for 2 h at
room temperature and incubated with the appropriate primary antibody diluted in
blocking solution for at least 2 h at room temperature. They were then incubated
with an appropriate secondary antibody diluted in blocking solution for 1 h at
room temperature. To detect proteins on the membrane, SuperSignal West Pico
chemiluminescent substrate (Pierce) was used according to the manufacturer’s
instructions.

Tumorigenicity studies. The ability of cells to form tumors was tested in
athymic mice (BALB/c). For each cell line/strain tested, two sites per athymic
mouse (shoulder and flank or both flanks) were subcutaneously injected with 1.0
x 10‘5 cells. Mice were examined weekly for tumor formation, and the tumor

dimensions were measured using calipers. Tumor volume (in cm3) was

180

 

estimated from the equation used to calculate the volume of a sphere. Tumor
latency was defined in these experiments as the time (wk) required for a tumor to
reach a volume of 0.5 cm3. For statistical analysis, the cell strains were divided
into three groups: group 1) ST7-V5lHis expressing transfectants (ST7-1 and
ST7-2); group 2) ST7tr1-V5/His transfectants cell strains (ST7tr1-1 and ST7tr1-2);
and group 3) non-ST7-expressing control and vector-control transfectants (NC-1,
NC-2, VC-1, and VC-2). The average latency for each group of cell strains was
computed. A t-test was used to compare the average latency observed for the
full-length ST7-expressing group with that observed for the group expressing the
truncated form of ST7 and with that observed for non-ST7-expressing control and
vector-control group, i.e., group 1 compared with group2 and group 1 compared

with group 3.

181

 

RESULTS

Construction of expression plasmids encoding V5lHis-tagged full-length
or a V5lHis-tagged truncated form of ST7. To construct an expression
plasmid encoding an ST7-V5lHis fusion protein, a donor plasmid containing ST7
cDNA was used as a template for PCR to generate an ST7 PCR product
containing a mutated stop codon (T AG—+TTG). This PCR product was digested
with EcoRI and subcloned into the EcoRI digested, dephosphorylated expression
vector pcDNA6-V5/HisA, in-frame with the vector’s C-terrninal V5 and His epitope
tags. The resulting plasmid, which contains a gene encoding a selective marker
(blasticidin), was designated pcDNA6-ST7-V5/His. To construct an expression
plasmid encoding a V5lHis-tagged truncated form of ST7 protein, the same
donor plasmid was used as the template for PCR amplification of the region of
ST7 encoding amino acids 1 through 507. The resulting plasmid, designated
pcDNA-ST7m-V5/His, encodes a protein containing the extracellular and
transmembrane domains of ST7, but only the first 10 of the 343 residues of its
cytoplasmic domain. Sequencing verified the fidelity of the PCR reactions.

Expression of full-length ST7 in a human fibrosarcoma-derived cell line,
but not the truncated form of ST7, inhibits tumor forming activity. To test
the hypothesis that ST7 functions as a tumor suppressor, we transfected SHAC
cells, a tumorigenic cell line derived from a patient’s fibrosarcoma, with the ST7-
V5lHis expression plasmid, with the ST7tr1-V5lHis expression plasmid, or with an
empty plasmid as a control. Transfectants were selected for resistance to

blasticidin. Independent drug resistant clonal populations were isolated,

182

 

expanded, and screened for expression of the appropriate V5lHis tagged protein
using Western blotting analysis with a V5-specific monoclonal antibody. Figure 1
shows the level of expression of V5lHis tagged full-length or truncated ST7
protein in such transfectants. The first two lanes contain protein isolated from
cells transfected with the pcDNA6-ST7-V5IHis plasmid, but that did not express
the encoded V5lHis-tagged ST7 protein (non-ST7-expressing control
transfectants). We observed two V5-reactive bands in the Iysates from the ST7tr1-
1 and ST7tr1-2 cell strains. We interpret the lower band visible in lanes 5 and 6 to
represent a product of proteolytic degradation.

These six SHAC-derived cell strains (shown in Figure 1), as well as two
vector-transfected control cell strains, VC-1 and VC-2, i.e., SHAC-derived cell
strains transfected with the empty pcDNA6-V5/HisA plasmid, and the parental
non-transfected SHAC cell line were tested for their ability to form tumors in
athymic mice. The cells were subcutaneously injected at two sites per mouse,
and the latency of tumor formation was defined as time required for a tumor to
reach 0.5 cm3 in volume. The results are shown in Table 1. Because the mice
were injected at two sites, the first tumor to arise sometimes grew faster than the
other, most likely because of having a better blood supply. In those cases where
one tumor reached 0.5 cm3 in volume before the other, the mouse had to be
sacrificed. Even though the second tumor had not reached 0.5 cm3 in volume,
examination by the pathologist was able to determine that the smaller tumor was
malignant. Such tumors were excluded from the tumors analyzed for latency

(Table 1, columns 2-4), but were included in the total number of tumors per sites

183

 

Figure 1. Presence or absence of V5lHis-tagged proteins in SHAC transfectants.
Protein (50 pg) was separated in a 10% SDS-polyacrylamide gel and transferred
to PVDF membrane for Western blotting analysis with the anti-V5 monoclonal
antibody. Lanes 1 and 2, the non-ST7-expressing control cell strains NC-1 and
NC-2; Lanes 3 and 4, the ST7-V5lHis-expressing cell strains ST7-1 and ST7-2,
which showed a reduction in tumor-forming ability; Lanes 5 and 6, the ST7tr1-
V5lHis expressing cell strains ST7tr1-1 and ST7tr1-2. The lower of the two bands
present in lanes 5 and 6 is most likely a product of proteolytic degradation. Actin

expression was measured as a loading control.

184

transfectants
Id transferred
5 monoclori
ins N01 and
1 and 817-2
6, the ST7,-
he two bands

rdation. Actin

 

. - «ST7-V5IH is

«snm -ver is

 

 

 

«Actin

 

 

 

Figure 1

185

 

Table 1 Expression of ST7 inhibits the tumor-forming ability of SHAC cells

 

 

 

Cells Tumors (2 0.5 cm“) Latency range Average Total tumors per
tested“ per sites injected (wk)b latency injected sitesc
SHAC 7/8 46 4.8 8/8

VC-1 9/14 7-9 7.6 13/14

VC-2 4/6 4.5-8 5.8 6/6

NC-1 9/14 3-5.5 3.9 14/14

NC-2 12/14 4-7.5 5.8 14/14

ST7m-1 16/16 3.5-5.5 4.7 16/16
ST7m-2 15/16 4.5-7.5 5.7 16/16
ST7-1 6/14 10.5-16.5 13.6“ 9/14
ST7-2 10/22 1023 15.2‘I 10/22

 

' SHAC, parental non-transfected cells; VC-1&-2, vector-control transfectants; NC-1&-2, non-ST7-
expressing control transfectants; ST7tr1-1&2, transfectants expressing a V5lHis-tagged truncated
form of ST7; ST7-1&2, transfectants expressing V5lHis-tagged full-length ST7.

” Time required for a tumor to reach 0.5 cm3 in volume.

° See Results for explanation of this column.

d The difference in average latency between the ST7-expressing transfectants and the non-ST7-
expressing control or vector-control transfectants is highly significant, p < 0.0001; the same is
true of the difference between the ST7-expressing transfectants and the transfectants expressing

the truncated form of ST7.

186

injected (Table 1, column 5). As expected, the parental SHAC cell line as well as
the four non-ST7-expressing or vector control cell strains formed tumors at
virtually all sites injected with a very short latency. The same was true for the cell
strains expressing a truncated ST7 protein. In contrast, the ST7-1 and ST7-2 cell
strains, which express full-length ST7-V5lHis protein, formed tumors at only
about half of the sites injected, and when tumors developed, the time to reach
0.5 cm3 (latency) was significantly greater (p < 0.0001) than the time required for
the non-ST7-expressing cells and those expressing a truncated ST7 protein to
form such tumors.

The ST7-V5lHis and ST7tr1-V5lHis proteins are overexpressed in the
SHAC cell strains generated. As shown in Figure 1, the level of expression of
the truncated V5lHis-tagged ST7 protein is higher than that of full-length V5lHis-
tagged ST7 protein. Even if the lower band present in the lysates of the
transfectants expressing the truncated form of ST7 represents a degraded form
of this protein, the level of expression of the intact form of V5/His-tagged
truncated ST7 is higher than that of V5lHis-tagged full-length ST7. To compare
the level of exogenous full-length ST7-V5lHis protein with that of endogenous
ST7 protein, we first used the polyclonal antibody generated in 1997 by Qing et
al. (1999). It proved no longer active. Therefore, we had antibody prepared in
rabbits using another peptide from the cytoplasmic domain of ST7. This
antibody, directed against residues 617-632 of ST7, is significantly more specific
to ST7 than the anti-ST7 polyclonal antibody in 1997. To test the specificity of

this antibody, protein from HEK 293-T cells transiently transfected with an ST7-

187

 

V5lHis expression plasmid or a plasmid encoding a non-tagged ST7 protein, i.e.,
the donor plasmid referred to above or a derivative designated pcDNA6-ST7,
was used for Western blotting analysis. A membrane containing three identical
sets of samples was divided into three sections and probed with the new anti-
ST7 polyclonal antibody, its corresponding pre-immune serum, or an anti-V5
monoclonal antibody. As expected, the pre-immune serum did not recognize any
proteins; the polyclonal ST7 antibody recognized both the tagged and non-
tagged proteins; and the anti-V5 monoclonal antibody recognized the V5lHis
tagged protein at the same position as that recognized by the new anti-ST7
antibody (data not shown). This new ST7 antibody was then used to compare
the level of expression of full-length V5lHis-tagged ST7 in the two isolated SHAC
transfectants ST7-1 and ST7-2 with the level of expression of endogenous ST7
found in foreskin-derived human fibroblast cell lines, vector-control SHAC
transfectants, or non ST7-expressing control SHAC transfectants. Parental
SHAC cells were also included. The results, shown in Figure 2, indicate that the
new ST7 antibody recognized the V5lHis-tagged full-length ST7 in the SHAC
transfectants ST7-1 and ST7-2 (Figure 2, lanes 8 and 9). This protein migrates
at a slower rate than endogenous ST7 found in normal foreskin-derived cell lines
(lanes 1 and 2). No such larger protein band was observed in the non-ST7-
expressing control transfectants (Figure 2, lanes 6 and 7). However, in SHAC
cells and its derivative transfectants, the new ST7 antibody also recognized a
protein that migrated at the same position as that found in normal diploid human

fibroblasts.

188

 

Figure 2. Expression of endogenous ST7 protein in normal human fibroblasts,
SHAC cells, and SHAC-derived transfectants. Protein (50 pg) was separated in
a 10% SDS-polyacrylamide gel, transferred to a PVDF membrane, and probed
with a polyclonal anti-ST7 antibody. Lanes 1-2, foreskin-derived fibroblasts;
Lane 4, SHAC cells; Lanes 5 and 6, empty-vector control transfected cell strains;
Lanes 7 and 8, non-ST7-expressing control SHAC transfectants; Lanes 9 and 10,
ST7-V5lHis expressing SHAC transfectants. Actin expression was used as a

loading control.

189

In fibroblasts,
separated in
, and probed
d fibroblasts.
d cell strains
168 9 and II.

ismwea

  
 

 

 

 

+ ST7-V5lHis
‘ ‘- ST7

fl:- Actin

 

 

Figure 2

190

Plasma membrane localization of ST7-V5lHis and ST7tr1-V5lHis proteins.
To determine whether exogenous ST7-V5lHis and ST7tr1-V5lHis proteins localize
to cellular membranes, as predicted by our analysis of the structure of ST7
protein‘, we isolated plasma membrane-enriched and soluble fractions from the
ST7-1, ST7tr1-2, and VC-2 SHAC transfectants and analyzed them by Western
blotting using the anti-V5 antibody. Both ST7-V5lHis and ST7tr1-V5lHis were
detected in the appropriate plasma membrane-enriched fractions and were not in
the matching soluble fractions, indicating that V5lHis tagged full-length ST7 and
V5lHis tagged truncated S'l7 localize to the plasma membrane (Figure 3). As a
control, we also analyzed these fractions by blotting with an antibody against
Na+/K+ ATPase, an integral membrane protein. As expected, this latter protein
was detected in the plasma membrane-enriched fractions and was not in the

soluble fractions.

191

Figure 3. The V5lHis tagged ST7 proteins encoded by the transfected plasmids
are localized in the membrane of the SHAC cell recipients. Plasma membrane-
enriched (M) and soluble, cytosolic (S) fractions were isolated from a vector
control transfected SHAC cell strain (lanes 1 and 2); from a transfectant
overexpressing the truncated form of ST7 (lanes 3 and 4); and from a
transfectant overexpressing full-length ST7 (lanes 5 and 6). A total of 20 pg of
each fraction was separated in a 10% SDS-polyacrylamide gel, transferred to
PVDF membrane, and probed with anti-V5 antibody (upper panel) and an
antibody specific for a membrane protein, i.e., anti-Na‘VK+ ATPase antibody

(lower panel). The latter protein has multiple isoforms.

192

led plasmids
Imembrane
am a vector
transfectant
and from
.I of 20 p901
'anslerred to
nel) and an

ase antibody

 

 

 

. «ST7-V5lHis

«$17,, -V5IHis

 

 

 

 

]>Na”IK* ATPase

 

Figure 3

193

DISCUSSION

The results of the present study, as shown in Table 1, support the hypothesis
that ST7 is a novel tumor suppressor. Overexpression of V5lHis-tagged full-
length protein inhibited the tumor-forming ability of SHAC cells, a cell line derived
from a patient’s fibrosarcoma. Moreover, overexpression of a severely truncated
form of ST7, which contains its extracellular and transmembrane domains, but
lacks the majority of its cytoplasmic domain, failed to inhibit the tumor-forming
ability of SHAC cells. Therefore, we conclude that ST7 functions as a tumor
suppressor in SHAC cells. Moreover, the data indicate that tumor suppression
occurs as a result of ST7 protein function rather than merely protein
overexpression, since the overexpressed truncated protein did not inhibit
tumorigenicity. These results also strongly suggest that the cytoplasmic domain
of ST7 plays a critical role in the protein’s function as a tumor suppressor.

When we began our studies to test the hypothesis that ST7 functions as a
tumor suppressor, we chose to use the human fibrosarcoma-derived cell line
SHAC because research carried out in this laboratory in 1997 showed that SHAC
cells fail to express endogenous ST7 protein (Qing et al., 1999). However, when
we used a different polyclonal antibody to examine ST7 expression in both
normal human diploid fibroblast cell lines and in SHAC and SHAC-derived cell
strains, we detected endogenous ST7 protein in all of those cell lines/strains
(Figure 2). Because there is no matched normal human cell line to use as the
standard against which the ST7 expression level in SHAC cells can be

compared, it is not possible to determine whether or not ST7 is down-regulated in

194

 

SHAC cells. However, the tumor data in Table 1 demonstrates that if the ST7
protein present in SHAC cells is functional, then its level is not sufficient to
prevent these cells from rapidly forming tumors. Increasing the amount of full-
length ST7 protein in these cells clearly inhibited their tumor-forming ability and
greatly increased tumor latency- It is possible that the ST7 protein present in
SHAC cells is non-functional. For example, point mutations in one or both of the
alleles of the ST7 gene could result in the synthesis of an inactive or mislocalized
protein. Such mutations have been shown to inactivate LDLR, the prototype
member of the LDLR superfamily (reviewed by Brown and Goldstein, 1986), and
to cause familial hypercholesterolemia, an autosomal dominant disorder resulting
in an increased concentration of plasma cholesterol and early onset heart attacks
(Davis et al., 1986). At present, we cannot assess ST7 function in human
fibroblasts. Moreover, ST7 belongs to an LDLR subfamily3 that also contains two
closely related proteins, LRP3 and LRP9. The functions of LRP3 and LRP9 have
not yet been determined, but the striking similarity of these proteins with ST7
suggests that this subfamily of proteins has overlapping functions in human cells.
Therefore, it is possible that SHAC cells are deficient in one or both of the other
members of this subfamily and that overexpression of ST7 restores the function
lost by LRP3 and/or LRP9 deficiency.

Both ST7-V5lHis and ST7m-V5/His proteins were found in plasma membrane-
enriched cellular fractions and not in soluble, cytosolic cellular fractions, strongly
suggesting that these proteins are localized in the plasma membrane (Figure 3).

This agrees with Qing et al. (1999), who predicted that ST7 is a type I

195

transmembrane protein. Our related studies of ST7’s cytoplasmic domain3
provide at least one possible explanation for ST7’s ability to suppress tumors.
Using the yeast two-hybrid system, we isolated proteins that interact with ST7’s
cytoplasmic domain. One such ST7-interacting protein identified was RACK1, a
scaffold protein that interacts with a plethora of cellular proteins and is believed
to play a key role in regulating a variety of cellular signaling pathways (reviewed
by Schechtman and Mocth-Rosen, 2001). Chang et al. (1998) showed that
RACK1 binds to the nonreceptor tyrosine kinase Src and negatively regulates its
kinase activity, thereby inhibiting Src-mediated proliferation of NIH3T3
fibroblasts. We also observed that ST7 overexpression inhibits the proliferation
of SHAC cells (data not shown). It is possible that ST7, through its interaction
with RACK1, plays a role in RACK1-mediated inhibition of Src-induced cellular
proliferation. Binding of RACK1 to ST7’s cytoplasmic tail would place RACK1 at
the membrane in proximity to active Src, enabling RACK1 to bind to and inhibit
Src. Inhibition of such a proliferative pathway by ST7 would provide cells with
protection against abnormal cellular replication. Therefore, loss of ST7 function
would disrupt this protective growth inhibitory pathway and provide tumor cells
with a growth advantage. A better understanding of the mechanism through
which ST7 functions to inhibit tumorigenesis may provide insights into the

diagnosis and treatment of cancer.

196

 

REFERENCES

Bork, P., and Beckmann, G. (1993). The CUB domain. A widespread module in
developmentally regulated proteins. J Mol Biol 231 , 539-545.

Boucher, F., Liu, P., Gotthardt, M., Hiesberger, T., Anderson, R. G., and Herz, J.
(2002). Platelet-derived growth factor mediates tyrosine phosphorylation of the
cytoplasmic domain of the low Density lipoprotein receptor-related protein in
caveolae. J Biol Chem 277, 15507-15513.

Brown, M. S., and Goldstein, J. L. (1986). A receptor-mediated pathway for
cholesterol homeostasis. Science 232, 34-47.

Chang, B. Y., Chiang, M., and Cartwright, C. A. (2001). The interaction of Src
and RACK1 is enhanced by activation of protein kinase C and tyrosine
phosphorylation of RACK1. J Biol Chem 276, 20346-20356.

Chang, B. Y., Conroy, K. B., Machleder, E. M., and Cartwright, C. A. (1998).
RACK1, a receptor for activated C kinase and a homolog of the beta subunit of G
proteins, inhibits activity of src tyrosine kinases and growth of NIH 3T3 cells. Mol
Cell Biol 18, 3245-3256.

Chang, B. Y., Harte, R. A., and Cartwright, C. A. (2002). RACK1: a novel
substrate for the Src protein-tyrosine kinase. Oncogene 21 , 7619-7629.

Christensen, E. l., and Birn, H. (2002). Megalin and cubilin: multifunctional
endocytic receptors. Nat Rev Mol Cell Biol 3, 256-266.

Davis, C. G., Lehrman, M. A., Russell, D. W., Anderson, R. G., Brown, M. S., and
Goldstein, J. L. (1986). The J.D. mutation in familial hypercholesterolemia: amino

acid substitution in cytoplasmic domain impedes internalization of LDL receptors.
Cell 45, 15-24.

Hussain, M. M., Strickland, D. K., and Bakillah, A. (1999). The mammalian low-
density lipoprotein receptor family. Annu Rev Nutr 19, 141-172.

Liu, C. X., Li, Y., Obermoeller-McCormick, L. M., Schwartz, A. L., and Bu, G.
(2001). The putative tumor suppressor LRP1 B, a novel member of the low
density lipoprotein (LDL) receptor family, exhibits both overlapping and distinct
properties with the LDL receptor-related protein. J Biol Chem 276, 28889-28896.

Liu, C. X., Musco, 8., Lisitsina, N. M., Forgacs, E., Minna, J. D., and Lisitsyn, N.
A. (2000a). LRP-DIT, a putative endocytic receptor gene, is frequently inactivated
in non-small cell lung cancer cell lines. Cancer Res 60, 1961-1967.

Liu, C. X., Musco, S., Lisitsina, N. M., Yaklichkin, S. Y., and Lisitsyn, N. A.
(2000b). Genomic organization of a new candidate tumor suppressor gene,
LRP1 B. Genomics 69, 271-274.

197

 

Loukinova, E., Ranganathan, S., Kuznetsov, S., Gorlatova, N., Migliorini, M. M.,
Loukinova, D., Ulery, P. G., Mikhailenko, l., Lawrence, D. A., and Strickland, D.
K. (2002). Platelet—derived growth factor (PDGF)-induced tyrosine
phosphorylation of the low density lipoprotein receptor-related protein (LRP).
Evidence for integrated co-receptor function betwenn LRP and the PDGF. J Biol
Chem 277, 15499-15506.

McCormick, J. J., and Maher, V. M. (1996). Analysis of the multistep nature of
human carcinogenesis utilizing human fibroblasts. Radiat. Oncol. Invest. 3, 387-
391.

Qing, J., Maher, V. M., Tran, H., Argraves, W. S., Dunstan, R. W., and
McCormick, J. J. (1997). Suppression of anchorage-independent growth and
matrigel invasion and delayed tumor formation by elevated expression of fibulin-
1D in human fibrosarcoma-derived cell lines. Oncogene 15, 2159-2168.

Qing, J., Wei, D., Maher, V. M., and McCormick, J. J. (1999). Cloning and
characterization of a novel gene encoding a putative transmembrane protein with
altered expression in some human transformed and tumor-derived cell lines.
Oncogene 18, 335-342.

Schechtman, D., and Mochly-Rosen, D. (2001). Adaptor proteins in protein
kinase C-mediated signal transduction. Oncogene 20, 6339-6347.

Strickland, D. K., Gonias, S. L., and Argraves, W. S. (2002). Diverse roles for the
LDL receptor family. Trends Endocrinol Metab 13, 66-74.

198

 

R

‘MICHIGAN STATf UNIVERSIIVTY LIB
llilWilli!llllllliilllilllillllilllllllll!
3 1293 02372 7

illlilill

419