MATERIALS AND METHODS FOR IDENTIFYING AGENTS THAT
MODULATE NORRIN, NORRIN MTMETICS, AND AGENTS IDENTIFIED
THEREBY
REFERENCE TO A SEQUENCE LISTING
The sequence listing submitted herewith containing SEQ ID NOS: 1-28 are
incorporated herein by reference for all purposes.
FIELD OF THE INVENTION
The present invention relates to materials and methods of regulating the Norrin
gene or the Norrin protein, Norrin mimetics, or agents that interact with Norrin and
thereby modulate its activity in the LRP5/Norrin/Frizzled4 complex.
BACKGROUND
Mutations in Norrin or Norrie disease protein (NDP or Ndph) leads to Norrie
disease (ND), an X-linked recessive neurological syndrome (Berger et al., 1992 Nat.
Genet. 1: 199-203; and Chen et al., 1992 Nat. Genet. 1: 204-208; for example, NCBl
Accession Nos. AAH29901, BC029901, and CAA46639 for human sequences, and
CAA58725, CAA63134, and X83794 for Mus museulus sequences). The gene
encoding NDP is located at Xp11.4 on the human genome. The disease characteristics
include retinal dysplasia, blindness, and mental retardation. NDP knockout mice have
an eye phenotype, which resembles human Norrie disease (Rhem et al., 2002 J.
Neuroscience 22:4286-4292) and show failure in retinal angiogenesis. The gene is
also tied to Coats disease (retinal telangiectasis), X-linked exudative vitreoretinopathy
(EVRX), and advanced retinopathy of prematurity (ROP).
NDP mutations can also cause X-linked form of Familial Exudative
Vitreoretinopathy (FEVR) that harbors some of the ND symptoms. FEVR is also
caused by mutations in Fri2zled4 (Fz4) gene that encodes one of the ten serpentine
seven-transmembrane receptors of Wnt-signaling (Robitaille et al., 2002 Nat. Genet.
32: 326-330).
The Wnt family consists of 19 members, and they show high affinity
interaction with 10 Frizzled (Fz) proteins. Different Wnts have different affinities for
various Frizzled proteins and may engage different pathways (Wu et al., 2002 J. Biol.
Chem. 277: 41762 - 41769). The Wnt-canonical pathway involves beta-catenin
stabilization through interaction with Frizzled and its co-receptor lipoprotein related
receptor protein 5 or 6 (LRP5/LRP6). In patients and mice, the loss of function
mutations in LRP5 also show vascular eye defects in addition to osteoporosis and give

rise to osteoporosis-pseudoglioma syndrome (OPPG) (Gong et al., 2001 Cell 207:
513-523; and Kato et al., 2002 J. Cell Biol. 157: 303-314).
Norrin can induce the Wnt-beta-catenin pathway (Xu et al., 2004 Cell 116:
883-895). Norrin functions much like a Wnt, because, both require an Fz receptor and
an LRP5/LRP6 co-receptor for signaling, and both bind predominantly to cysteine rich
domain (CRD) of Fz with nanomolar affinity (Hsieh et al., 1999 Proc. Nat V. Acad.
Sci. 96: 3546-3551; Wu et al., 2002 J. Biol. Chem. 277: 41762-41769; and Xu et al.,
2004 Cell 116: 883-895). Norrin is a cysteine-rich small protein that gets secreted; it
forms disulfide—linked oligomers and remains associated with the cell surface and
extra cellular matrix (Perez-Vilar et al., 1997 J. Biol Chem. 272: 33410-33415).
From sequence comparisons and modeling studies, it has been suggested that Norrin
has a tertiary structure similarity to TGF-beta (Meitinger et al., 1993 Nat. Genet. 5:
376-380), von Willebrand's factor, and mucin (Meindl et al., 1992 Nat. Genet. 2: 139-
143). The Norrin gene is expressed predominantly in brain and retina (Berger et al.,
1992 Nat. Genet. 1: 199-203; and Chen et al., 1992 Nat. Genet. 1: 204-208).
Drug candidates for treating diseases related to bone remodeling are constantly
being sought. The human adult skeleton is in a dynamic state, being continually
broken down and reformed by the coordinated actions of osteoclasts and osteoblasts
on trabecular (also called cancellous) bone surfaces and in haversian systems.
Disruption of bone remodeling can lead to diseases and conditions such as
osteoporosis (postmenopausal osteoporosis, glucocorticoid-induced osteoporosis,
transplantation induced osteoporosis, and juvenile), rickets, osteomalacia, tumor
induced osteomalacia, hypophosphatasia, Paget's disease, and others. Thus, more
fully elucidating the pathways controlling bone remodeling and identifying targets in
those cascades are useful for developing agents that modulate the targets are needed.
SUMMARY
The materials and methods described herein provide a greater elucidation of
Norrin's involvement in the Wnt pathway via its interaction with LRP5 and 6 and
Frizzled4 for bone remodeling and lipid metabolism modulation, and generally
provide assays using Norrin and compounds which interact with Norrin (e.g. Norrin
agonists) to screen for compounds that are useful in modulating bone disorders and
lipid modulation. Also contemplated are methods and materials for identifying Norrin
mimetics, as well as other mimetics of the LRP5/Frizzled4/Norrin complex.
Accordingly, an aspect is directed to a method of identifying an agent that
modulates bone or a lipid comprising: (a) having a Frizzled4 protein or biologically
active Frizzled4 polypeptide with a LRP5 protein or biologically active LRP5 protein
in the presence of the agent; and (b) determining whether said agent is a agent that
interacts with Frizzled4 and/or LRP5 or a biologically active polypeptide of Frizzled4

or LRP5, and modulates at least one parameter of bone and/or a lipid in the presence
of the agent. One aspect has the Frizzled4 protein or biologically active Frizzled4
polypeptide fragment is linked to the LRP5 protein or biologically active polypeptide
fragment of LRP5; this can be in the form of a fusion protein. The agent being
screened by this method can be a Norrin mimetic, as well as agonists and antagonists
ofNorrin.
Another aspect is directed to a method of identifying an agent that modulates
bone metabolism or lipid metabolism comprising: (a) having a Norrin protein or a
biologically active Norrin polypeptide fragment and a Frizzled4 protein or biologically
active polypeptide fragment fused to LRP5 and/or LRP6 proteins or biologically
active polypeptide fragments thereof in the presence of the agent; and (b) measuring in
vitro or in vivo at least one parameter of bone modulation and/or lipid modulation to
identify the agent that modulates bone metabolism or lipid metabolism.
The parameters of bone modulation for any of the discussed methods can be
any one or more of bone density, bone strength, trabecular number, bone size, or
tissue connectivity, or any combination thereof. The parameters of lipid modulation
discussed in any of these screening methods can include a change in the level of HDL,
VLDL, cholesterol, triglyceride, apoE, or LDL. Another aspect of screening agents in
these methods is to see whether they alter expression patterns of genes associated with
lipid metabolism or bone metabolism. For example, do they alter expression of one or
more of COX-2, Jun, Fos, cyclin Dl, Wnt10B, SFRPl, connexin 43, eNOS, Wnt10B,
cyclin Dl, Frizzled2, and WISP2 is modulated.
The methods can further include Dkk protein or a biologically active Dkk
polypeptide fragment and/or a Kremen protein a biologically active Kremen
polypeptide fragment and/or a Wnt protein or a biologically active Wnt polypeptide
fragment.
In yet another aspect, a method is contemplated for identifying an agent that
modulates a Norrin-Frizzled4 activity comprising: (a) having the agent, a Norrin
protein or biologically active polypeptide fragment of Norrin, and a Frizzled4 protein
or biologically active polypeptide fragment of Frizzled 4 fused to LRP5 and/or LRP6
or a biologically active polypeptide fragment of LRP5 and/or LRP6, or fused to a
ligand binding domain (LBD) containing polypeptide fragment of LRP5 or LRP6 and:
(i) a Kremen protein or biologically active polypeptide fragment of
Kremen; and/or
(ii) a Dkk protein or biologically active polypeptide fragment of Dkk;
and
(b) determining whether the agent modulates a Norrin-Frizzled4 activity. The agents
can be a Norrin mimetic, Norrin agonist, Dkk antagonist, or a Kremen antagonist. It
can also be for assessing Frizzled4 agonists and LRP5 agonists.

For certain of these methods, the proteins or biologically active polypeptide
fragments of the proteins can be affixed on a substrate, such as PVDF or
nitrocellulose.
A farther aspect contemplates a method of identifying an agent that regulates
bone modulation or lipid modulation comprising: (a) administering the agent to a cell
expressing Frizzled4 and LRP5, wherein Frizzled4 is a Frizzled4 protein or
biologically active polypeptide fragment of Frizzled4, and LRP5 is a LRP5 protein or
a biologically active polypeptide fragment of LRP5; (b) determining whether said
administration of the test agent modulates a LRP5-Frizzled4 interaction; and (c)
determining whether the agent modulates a bone parameter and/or a lipid parameter.
One aspect of this method contemplates that the cell does not express Norrin, which is
useful for identifying Norrin mimetics. Another aspect has the cell expressing a non-
endogenous Frizzled4, LRP5, LRP6, and/or Norrin and using the cell to, for example,
identify Norrin agonists.
Another aspect contemplates a method of identifying an agent that regulates
bone modulation or lipid modulation comprising: (a) administering a test agent to a
cell expressing LRP5, Norrin and Frizzled4, wherein LRP5 is a LRP5 protein or a
biologically active polypeptide fragment of LRP5, Norrin is a Norrin protein or a
biologically active Norrin polypeptide, and Frizzled4 is a Frizzled4 protein or a
biologically active polypeptide fragment of Frizzied4; (b) determining whether said
administration of the test agent modulates Norrin-Frizzled4 interaction; and (c)
determining whether the test agent modulates a parameter of bone modulation or lipid
modulation. This contemplates the cell optionally expressing a non-endogenous
Norrin, LRP5, and/or Frizzled4. Alternatively, the cell may not express an
endogenous Norrin, LRP5, LRP6, and/or Frizzled4.
Any of the agents tested can be Norrin mimetics, Dkk antagonists, or Kremen
antagonists, as well as Frizzled4 agonists and mimetics, Norrin agonists, and LRP5
agonists.
In any of the methods or cells contemplated the cells can be vertebrate cells.
Vertebrate cells can include but are not limited to bone cells, kidney cells,
mesenchymal cells, adipocytes, preadipocytes, or Xenopus cells.
In any of the methods, kits, cells/cell lines discussed, the Dkk can be Dkkl,
Dkk2, Dkk3, or Dkk4, or a biologically active polypeptide of Dkkl, Dkk2, Dkk3, or
Dkk4. Likewise, for Kremen, when Kremen is cited in any aspect it can be Kremen 1
or Kremen2, or a biologically active polypeptide of Kremenl or Kremen2. Also the
Wnt can be any of Wnt1 to Wnt19 {e.g., Wnt1, Wnt3, Wnt3a, or Wnt10b) or a
biologically active fragment of any of these.
Cell lines for use with any of the methods and kits can include but are not
limited to KHOS/NP cells, KHOS-240S cells, KHOS-321H cells, DSDh cells, VA-

ES-BJ cells, 7F2 cells, U2OS cells, HOSTE85 cells, ROS cells, MC3T3-E6 cells,
UMR-106 cells, Saos2 cells, MG63 cells, HOB cells, mesenchymal stem cells (e.g.,
human adult mesenchymal stem cells), C3H10T1/2 cells, HEK293A cells, or
HEK293T cells.
Animals are also contemplated for use in screening the reagents for
modulating bone metabolism and lipid metabolism. Animal models include
transgenic animals. For example, the animal can be an LRP5 or HBM expressing
transgenic animal. Alternatively the animal may be knockout animals wherein one or
more of LRP5, LRP5, Norrin, a Dkk, a Kremen, a Wnt, and Frizzled4 are knocked
out. Alternatively, the animals also contemplate combined knockouts and introduced
genes. The animals can be any vertebrates such as Xenopus or mice.
Yet another embodiment contemplates a kit for identifying an agent which
modulates Norrin-Frizzled4 activity comprising: (a) a series of cells incapable of
expressing Norrin that are co-transfected with nucleic acids encoding Frizzled4 or a
biologically active polypeptide fragment of Frizzled4 and LRP5 or a biologically
active polypeptide fragment of LRP5; (b) optionally a Dkk nucleic acid for co-
expression in a series of cells co-expressing Frizzled4 or the biologically active
polypeptide fragment of Frizzled4 and LRP5 or the biologically active polypeptide
fragment of LRP5; (c) optionally a Kremen nucleic acid for co-expression in a series
of cells co-expressing Frizzled4 or a biologically active polypeptide fragment of
Frizzled4 and LRP5 or a biologically active polypeptide fragment of LRP5, and/or for
co-expression in a series of cells co-expressing Frizzled4 or a biologically active
polypeptide fragment of Frizzled4, LRP5 or a biologically active polypeptide
fragment of LRP5, and Dkk or a biologically active fragment of Dkk; (d) optionally a
LRP6 nucleic acid for co-expression in a series of cells co-expressing Frizzled4 or a
biologically active polypeptide fragment of Frizzled4 and LRP5 or a biologically
active polypeptide fragment of LRP5; and (e) optionally a Wnt nucleic acid for co-
expression in a series of cells co-expressing Frizzled4 or a biologically active
polypeptide fragment of Frizzled4 and LRP5 or a biologically active polypeptide
fragment of LRP5.
Yet another aspect contemplated Is a cell or a cell line lacking a native Norrin
and which expresses a non-native LRP5 and a non-native Frizzled4, wherein the non-
native LRP5 is a non-native LRP5 protein and the non-native Frizzled4 is a non-
native Frizzled4 protein, wherein the LRP5 is the complete protein or a biologically
active polypeptide fragment of LRP5 and Frizzled4 is the complete protein or a
biologically active polypeptide fragment of Frizzled4, and Norrin is the complete
protein or a biologically active polypeptide fragment of Norrin. Expression of these
proteins can be transient or stable expression. Another aspect contemplates non-
native (non-endogenous) expression of LRP5, LRP6, Frizzled4, a Dkk, and/or a

Kremen, wherein any of these can be whole protein or biologically active polypeptide
fragments thereof.
Another aspect contemplates the agents identified by any one of the methods
above alone or in a pharmaceutical composition with suitable pharmaceutically
acceptable excipients and/or carriers. The agent can be used to treat a lipid disorder
and or a bone disorder or used to formulate a medicament for use in treating one of
these disorders.
It is to be understood that both the foregoing general description and the
following detailed description are exemplary and explanatory and are intended to
provide further explanation of the invention as claimed.
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings, which are included to provide a further
understanding of the materials and methods disclosed and are incorporated in and
constitute a part of this specification, illustrate embodiments.
FIG. 1. Norrin activates TCF-signaling in U2OS osteoblast-like cells but not
in HEK-293A cells. Transient co-transfection assays were conducted in human
embryonic kidney (HEK) HEK-293A cells and U2OS cells with TCF-Iuci and renilla
reporters. The bar graph represents the ratio of luminescence signals of TCF-luci to
renilla signals that normalize the responses between various transfections using
different cDNA constructs in pcDNA vector.
FIG. 2. Frizzled4 (Fz4) co-transfection induces Norrin mediated TCF-signal
in HEK-293A cells and enhances that in U2OS cells. The bar graph shows the results
of TCF-Iuci responses obtained using transient transfections of HEK-293A and U2OS
cells with various combinations of cDNAs.
FIG. 3. Norrin induced LRP5-Fz4-TCF-signal can be inhibited synergistically
by Dkkl and Kremen2 in U2OS cells.
FIG. 4. Norrin mediated TCF signal with LRP5-G171V mutant (HBM
phenotype) is less inhibited than that with LRP5 in presence of Dkkl and Kremen2.
DETAILED DESCRIPTION
Since it was intriguing that three genes, NDP, Frizzled4 (Fz4), and lipoprotein
related receptor protein 5 (LRP5), were found to be involved in vasularization of the
retina, investigations on these line revealed that they do interact at the molecular level;
and Norrin is a ligand of LRP5-Fz4 complex. It is interesting to note that the Norrin-
mediated activation of LRP5/6 involves Fz4 and not the other five members of the Fz
family (i.e., mFZ3, hFZ5, mFz6, mFz7, and mFz8). However, Norrin has specificity
to Fz4 and does not show any significant sequence homology to Wnts.
LRP5 mutations in humans and in mice have revealed the pivotal role that

LRP5 and Wnt-signal play in bone metabolism (Gong et al., 2001 Cell 207: 513-523;
Kato et al., 2002 J. Cell Biol. 157: 303-314; Boyden et al., 2002 N. Engl. J. Med. 346:
1513-1521; and Little et al., 2002 Am. J. Hum. Genet. 70: 1-19). The G171V (high
bone mass or "HBM" type) mutation and other such mutations in the first propeller
domain of LRP5 results in a decreased affinity of the HBM variants to the Dikkopfl
(Dkkl) protein as compared to that with the wild-type LRP5 (Boyden et al., 2002 N.
Engl. J. Med. 346; and Ai et al., 2005 Mol. Cell. Biol. 25: 4946-4955). The HBM
mutation leads to decreased inhibition by Dkkl, and the activation of HBM mutant
mediated Wnt-beta-catenin signals in vitro. This phenomenon is speculated to be the
underlying molecular mechanism important to high bone mass (HBM) phenotype in
humans and in transgenic mice with HBM-type mutations (Babij et al., 2003 J. Bone
Mineral Res. 18: 960-974).
Dkkl is one of the secreted antagonists of LRP5/6-Wnt signal (Glinka etal.,
1998 Nature 391: 357-362; Kawano et al., 2003 J. Cell. Sci. 116: 2627-2634; and
Bafico et al., 2001 Nat: Cell Biol. 3: 683-686). In addition, Dkkl in the presence of
Kremen 1/2, another type of single pass transmembrane receptor, enhances the
inhibition of LRP5/6-TCF signal mediated through Wnt (Mao et. al., 2002 Nature
417: 664-667) by internalization of the LRP5-Dkkl-Krernen ternary complex.
Kreirtens form the ternary complex at the cell surface with LRP5/6 and Dkkl to
facilitate their internalization or endocytosis. Kremens facilitate rapid endocytosis of
LRP5 and LRP6 from the cell membrane and thereby block LRP5/6-Wht signaling.
The materials and methods disclosed herein arose from speculation that the expression
pattern of these interactors in a given cell type can regulate Wnt-signaling or bring
additional specificity to LRP5/6 function in cells, such as osteoblasts. Tt is to be noted
that unless specifically set forth, in all instances wherein Dkkl is referenced, any of
the other Dkks may be substituted alone or in combination.
Analysis of the four splice variants of Kremen2 (Krm2) revealed a variant
lacking 44 amino acids at the carboxy terminus, which can enhance Dkkl mediated
inhibition of LRP5/6 (B. Mao et al., "Kremen proteins are Dickkopf Receptors that
Regulate Wnt/beta-Catenin Signaling," 2002 Nature 417(6889): 664-667). Maximal
effects of Dkkl enhancement is seen with a full-length Krni2 clone. Krrn2 activity is
mediated by its interaction with the second cysteine-rich domain of Dkkl. Krm2 can
also convert the LRP6-Wnt signal activator Dkk2, into an inhibitor in HEK-293A
cells. It is to be noted that unless specifically set forth, in all instances wherein
Kremen2 is referenced, Kremen 1 may be substituted alone or in combination with
Kremen2.
The materials and methods disclosed herein are directed to the functional
interactions between Norrin, Frizzled4, LRP5, or HBM variants of LRP5, e.g.,
G171V. As described herein, Norrin enhances modestly the TCF-signal of the

G171V-LRP5 mutant over the signal observed with LRP5 in U2OS bone cells.
Norrin also leads to a decreased inhibition of the pathway by Dkkl and/or Kremen2.
Disclosed herein are materials and methods for the use of Norrin as a screening agent
for finding reagents that are Norrin mimetics and Norrin agonists. Such Norrin
modulating agents and Norrin mimetics may be useful for bone modulation. Norrin
modulators and Norrin mimetics include but are not limited to small chemical
molecules, polypeptides, peptides, siRNAs, and immunoglobulins.
1. Abbreviations and Definitions
1.1 Abbreviations
The following abbreviations have been used in the specification. Although
these acronyms and abbreviations may have different meanings in other arts, they are
as indicated below, or as separately distinguished in the specification.
ACP5 acid phosphatase 5
Akt-3 protein kinase B (PKB) or RAC-PK
AlPASE alkaline phosphatase
ALPHA amplified luminescent proximity homogenous assay
APE adaptor-related protein 1
AP1B1 adaptor protein complex AP-1, beta 1 subunit
AX1N axin
b.i.d. bis in die (twice daily)
BGN bone specific biglycan
BMC bone mineral content
BMD bone mineral density
BMPl bone morphogenetic protein 1
BMP4 bone morphogenetic protein 4
BMU bone remodeling unit
BSA bovine serum albumin
BTG2 B-cell translocation gene 2, antiproliferative
CBFB core binding factor beta
CCND1 cyclinDl
CCND3 cyclin D3
CCNI cyclin I
cDNA complimentary DNA
CELSR2 cadherin EGF LAG seven-pass G-type receptor 2
CFP cyan fluorescent protein
CHUK/IKK alpha conserved helix-loop-helix ubiquitous kinase, IkB kinase
alpha
CK1 alpha casein kinase 1, alpha 1

CKB creatine kinase, brain
CNKl connector enhancer of KSR-like
Co11Al collagen, type 1, alpha 1
Col3A 1 collagen, type 3, alpha 1
Col6A3 collagen, type VI, alpha 3
Connx43 Connexin 43
COX-2 cyclooxygenase^
CRABP2 cellular retinoic acid binding protein IT
CRD cysteine rich domain
CSFIR colony stimulating factor 1 receptor
CSPG2 chondroitin sulphate proteoglycan
CTGF connective tissue growth factor
CTSK cathepsin K
CX3CR1 chemokine (C-X3-C) receptor 1
Cyclin Dl see also CCNDl
DELTEX deltex homolog 2 (Drosophild), see EphB2
Dkk Dikkopf
Dkkl Dikkopfl
Dkk2 Dikkopf2
Dkk3 Dikkopf3
Dkk4 Dikkopf4
DMSO dimethyl sulphoxide
dsRNA double stranded RNA
DVLl disheveled, dsh homolog (Drosophild)
DXA dual X-ray absorptiometry
EDTA ethylenediaminetetra acetic acid
EGTA ethylene gIycol-O-O'-bis(2-amino-ethyl)-N,N;N,N'-tetraacetic
acid
eNOS excitable nitric oxide synthase
EPHB2 connector enhancer of KSR-like (Drosophila kinase suppressor
of ras)
EPHB6 Eph receptor B6
ERBB3 GRO1 oncogene
ERK. also known as mitogen activated protein kinase p44/42
(MAPK)
EVRX X-linked exudative vitreoretinopathy
FAP fibroblast activation protein, alpha
FBLNl fibulin 1
FBS fetal bovine serum

FEVR familial exudative vitreoretinopathy
FGF-2 fibroblast growth factor 2 (basic)
FGF-7 Fibroblast growth factor 7 (keratinocyte growth factor)
FOS FBJ murine osteosarcoma viral oncogene homolog
FOSL1 Fos-like antigen 1
FRET fluorescent resonance energy transfer
Frizzled2 Frizzled (Drosophila) homolog 2, also called FZD2
Fz Frizzled
Fz4 Frizzled4
FZD2 Frizzled (Drosophila) homolog 2
FZD4 Frizzled homolog 4
G171V glycine to valine mutation at position 171 of human LRP5
GADD45A growth arrest and DNA-damage inducible, alpha
GADD45B growth arrest and DNA-damage inducible 45, beta
GADD45G growth arrest and DNA-damage inducible 45, gamma
GAS6 growth arrest-specific 6
GFP green fluorescent protein
GJA1 gap junction membrane channel protein alpha 1 (also known as
Connexin 43)
GJB3 gap junction membrane channel protein beta 3
GSK-3 glycogen synthase kinase-3
GSK-3a glycogen synthase kinase-3, alpha isoform
GSK-3P glycogen synthase kinase-3, beta isoform
HBM high bone mass phenotype
HDL high density lipoprotein
HEK human embryonic kidney
HERPUDl homocysteine-inducible, endoplasmic reticulum stress-
inducible, ubiquitin-Hke domain member 1
HRT hormone replacement therapy
i.m. intramuscular
i.v. intravenous
IDB2 inhibitor of DNA binding 2
EDB3 insulin-like growth factor 2 (somatomedin A)
IGF2R insulin-like growth factor 2 receptor
IGFBP6 insulin-like growth factor binding protein 6
iGSK GSK inhibitor
iGSK-3 GSK-3 inhibitor
IL-1 interleukin-1
ELIRI interleukin-1 receptor, type I

ILIRLI interleukin-1 receptor-like 1
DL4RA interleukin 4 receptor, alpha
IL-6 interleukin-6
ITGA5 integrin alpha 5 (fibronectin receptor alpha)
ITGB5 integrin, beta
ITGBLl integrin, beta-like 1
JNK c-jun amino kinase pathway
JUN v-jun avian sarcoma virus 17 oncogene homolog
JUNDl Jun proto-oncogene related gene dl
Kremen Kringle coding gene making the eye and the nose
Krml Kremen 1
Krm2 Kremen2
LBD ligand binding domain of LRP5, LRP6, HBM
LDL low density lipoprotein
LDLR low density lipoprotein receptor
LOX lysyl oxidase
LRP5 low density lipoprotein receptor related protein 5
LRP6 low density lipoprotein receptor-related protein 6
LSPl lymphocyte-specific protein 1
LUM lumican
mAb monoclonal antibody
MAPK mitogen activated protein kinase (p42,44) (ERK)
MAPKAPK2 mitogen-activated protein kinase-activated protein kinase 1,
also called MK2
MCC mutated in colorectal cancers
MDSC mesenchyme derived stem cells
MET met proto-oncogene (hepatocyte growth factor receptor)
MMP-14 matrix metalloproteinase 14
MMP-9 matrix metalloproteinase 9
MSXl homeo box, msh-like 1
MYBLl v-myb myeloblastosis viral oncogene homolog (avian)-like 1
MYC v-myc avian myelocytomatosis viral oncogene homolog
MYCS Myc-Iike oncogene, s-myc protein
NCAMl neural cell adhesion molecule 1
ND Norrie disease
NDP Norrie disease protein (alsoNdph)
NFATCl nuclear factor of activated T-cells, cytoplasmic 1
NFKBl nuclear factor of kappa light chain gene enhancer in B-cells 1,
p105

Non-TG non-transgenic
NOS3 nitric oxide synthase 3, also known as eNOS
NR4A1 nuclear receptor subfamily 4, group A, member 1
OGN osteoglycin
OPG osteoprotegerin
OPPG osteoporosispseudoglioma syndrome
OSMR oncostatin M receptor
PCOLCE procollagen c-proteinase enhancer protein
PDGFA Cluster Incl. M29464:PIatelet derived growth factor alpha
PDGFRA platelet-derived growth factor receptor alpha polypeptide
PKA protein kinase A
PKC protein kinase C
PLAT tissue-type plasminogen activator, t-PA
PNA peptide nucleic acid
PRDC protein related to DAC and Cerberus
PTGIS prostaglandin synthase
PTGS post transcriptional gene silencing
PTGSl prostaglandin-endoperoxide synthase 1, also called COX-1
PTGS2 prostaglandin-endoperoxide synthase 2 (prostaglandin G/H
synthase or cyclooxygenase 2) or COX-2
PTH parathyroid hormone
RAMP3 receptor (calcitonin) activity modifying protein 3
RANK receptor activator of NF-kB
RANKL receptor activator of NF-kB ligand
RJLUs relative luciferase units
RNAi KNA interference
ROP retinopathy of prematurity
RUNXl runt related transcription factor 1
RUNX2/CBFA1 runt related transcription factor 2
S100AIO calcium binding protein similar to calpactin
SDCl syndecan 1
SDF1 stromal derived factor 1
SERM selective estrogen receptor modulator
SERPINE1 serine (or cysteine) proteinase inhibitor, clade E (nexin,
plasminogen activator inhibitor type 1), member 1
SFRP1 secreted frizzled-related protein 1
SFRP4 secreted frizzled-related protein 4
shRNA short hairpin RNA
siRNA short interfering RNA

SPARC sparc/osteonectin
SPARCLl SPARC-like 1 (mast9, hevin)
SPPl secreted phosphoprotein 1
SPR surface plasmon resonance
STATl signal trandsducer and activator of transcription 1
STAT3 RIKEN cDNA 1110034C02 gene
TANK TRAF family member-associated Nf-kappa B activator
TCF T cell factor
TG transgenic
TGFBl transforming growth factor, beta 1
TGFBR2 transforming growth factor, beta receptor II
TGF-β tumor growth factor beta
THBD thrombomodulin
THBSI thrombospondin 1
TIEG TGFB inducible early gene
TIMPl tissue inhibitor of metal loproteinase
TIMP2 tissue inhibitor of metalloproteinase 2
TIMP3 tissue inhibitor of metalloproteinase 3
TNF tumor necrosis factor
TNFRSF10B tumor necrosis factor receptor superfamily, member 10b
TNFRSF11B tumor necrosis factor receptor superfamily, member 11 b
(osteoprotegerin)
TNFSFl 1 tumor necrosis factor (ligand) superfamily, member 11 (see
RANKL)
TOB1 transducer of ErbB-2.1
TRAF3 TNF receptor-associated factor 3
TUNEL terminal deoxynucleotidyl transferase dUTP nick end labeling
UNK_D83402 prostaglandin 12 (prostacyclin) synthase
VCAMl vascular cell adhesion molecule 1
VEH vehicle
VLDL very low density lipoprotein
WIF Wnt inhibitory factor
WISPl WNTl inducible pathway protein 1
WISP2 WNTl inducible signaling pathway protein 2
Wnt 3A wingless-type MMTV integration site family member
Wnt wingless-type MMTV integration site family {e.g., Wntl to
Wnt 19)
Wntl OB wingless-type MMTV integration site family member 1 OB
Wnt6 wingless-type MMTV integration site family member 6

YFP yellow fluorescent protein
1.2 Definitions
In accordance with this detailed description, the following abbreviations and
definitions apply. It must be noted that as used herein, the singular forms "a", "an",
and "the" include plural referents unless the context clearly dictates otherwise. Thus,
for example, reference to "a mimetic" includes a plurality of such mimetics, and
reference to "the dosage" includes reference to one or more dosages and equivalents
thereof known to those skilled in the art, and so forth.
Unless defined otherwise, all technical and scientific terms used herein have
the same meaning as commonly understood by one of ordinary skill in the art. The
following terms are provided below. The genes referred to herein are meant to
include the accessions numbers referenced as well as other sequences not specified.
By "animal" is meant any vertebrate. "Animal" includes "mammals."
Preferred mammals include livestock animals (e.g., ungulates, such as cattle, buffalo,
horses, sheep, pigs and goats), as well as rodents (e.g., mice, hamsters, rats and guinea
pigs), canines, felines, primates (e.g., chimpanzees, orangutans, humans), lupines,
camelids, and cervidae. Other vertebrates include avians (e.g., chickens, ducks, geese,
fowl), amphibians .(e.g., Xenopus) and ichthyes (fish).
By "Norrin" is meant to include all vertebrate forms of Norrin, and all its
polypeptide and nucleic acid forms. "Norrin" is also referred to as ND, Norrie disease
protein, Norrin precursor, NDP, Ndp, and Norrie disease protein homology (Ndph).
Norrin variants are also contemplated.
By "Norrin activity" would be a Norrin activity as it involves the Wnt
signaling pathway and its interaction with Frizzled4 and LRP5/6. This would include
Norrin's interaction with Frizzled4 (Fz4) and its enhancement of LRP5 activity. The
only Frizzled protein Norrin interacts with is Frizzled4. However, the Wnts discussed
herein can be used in the assay systems to compare the specificity of any molecule that
would modulate Norrin-Frizzled4 and LRP5 and LRP6 interaction in the Wnt
signaling system. Thus, by "Norrin modulating agent" is an agent that modulates a
Norrin activity, wherein the Norrin activity is part of Wnt signaling. A preferred
Norrin activity is regulation of bone remodeling and/or lipid modulation. With regard
to lipid modulation, see U.S. Application No. 09/578,900. The contents of U.S.
Application No. 09/578,900 are incorporated herein by reference for all purposes in its
entirety. Norrin modulating agents include agonists and antagonists of bone activity
and/or lipid levels. A Norrin agonist, for example, would enhance bone growth in a
subject when administered.
By "Dkk" is meant to include all vertebrate forms of Dkk1, Dkk2, Dkk3, and
Dkk4 and all nucleic acid and polypeptide forms. By "Dkkl" is also referred to as
Dickkopf-1, Dickkopf related protein-1 precursor, Dkk-1, DKK-1, hDkk-1 (for the

human form of Dkk1), AK, and UNQ492/PRO1008. By "Dkk2" is also meant to
include Dickkopf-2, Dickkopf related protein-2 precursor, Dkk-2, DKK-2, hDkk-2
(for the human form of Dkk2), and UNQ682/PRO1316. By "Dkk3" is also meant to
include Dickkopf-3, Dickkopf related protein-3 precursor, Dkk-3, hDkk-3 (human
form of Dkk3), REIC, and UNQ258/PRO295. By "Dkk4" is meant to include
Dickkopf-4, Dickkopf related protein-4 precursor, Dkk-4, DKK-4, and hDkk-4 (for
the human form of Dkk4). Dkk variants are also contemplated. Dkk modulating
agents would include Dkk antagonists and agonists.
By "Kremen" is meant to include all vertebrate forms of Kremenl and
Kremen2, and all the nucleic acid and polypeptide forms. "Kremenl" is also referred
to as Dickkopf receptor, FLJ31863, KREMEN, Kringle-containing protein marking
the eye and the nose, Kringle containing transmembrane protein 1, and KRMl.
"Kremen2" is also referred to as Dickkopf receptor 2, Kremen protein 2 precursor,
Kringle-containing protein marking the eye and the nose, KRM2, MGC10791,
MGC16709, and the Kremen2 form associated with the Mammalian Gene Collection
(MGC) Program Team, 2002 "Generation and initial analysis of more than 15, 000
full length human and mouse cDNA sequences," Proc. Nat'l Acad. Sci. USA 99(26):
16899-16903. Kremen variants are also contemplated. Kremen modulating agents
would include Kremen antagonists and agonists.
By "LRP5" or "low density lipoprotein receptor-related protein 5" is meant to
include all vertebrate nucleic acid and polypeptide forms of LRP5. Other names for
LRP5 and related homologs include BMNDl, Zmaxl, HGNC:8152, low-density
lipoprotein receptor-related protein 5 precursor, LR3, LRP7, OPPG, OPS, and
VBCH2. "LRP5" is also known as "arr" in Drosophila and has the following
additional synonyms: BEST:CK00539, CK00539, i(2)kO8131, LDLR-like, LRP,
LRP5/6, and LRP6. For example, one "HBM" or "high bone mass" variant of LRP5
has a single amino acid change in the polypeptide form from a glycine to a valine at
position 171 in the human polypeptide sequence. There is a similar mutation at
position 170 of the mouse sequence. Additional homologs in other vertebrates can be
determined in other species given the three-dimensional propeller domains. Use of
HBM contemplates inclusion of the G171V variant and its homolog from other
vertebrate species. An HBM variant is one that produces a high bone mass phenotype,
which results from a mutation in LRP5 other than the G171V change. For the Zmaxl
and HBM forms, see U.S. Patent No. 6,770,461, which is incorporated herein in its
entirety for all purposes. Thus, an example of a wild-type variant or homolog of
LRP5 is Zmaxl.-When reference is made to LRP5, all forms ofLRP5 including a
wild type variant or homolog are also contemplated. Variants of LRP5 and HBM are
also contemplated. LRP5 modulating agents would include agonists and antagonists
of LRP5. Also contemplated are LRP5 mimetics.

By "LRP6" or "low density lipoprotein receptor-related protein 6" is meant to
include all vertebrate nucleic acid and polypeptide forms of LRP6. LRP6 is also
referred to as low-density lipoprotein receptor-related protein 6 precursor. Variants of
LRP6 are also contemplated. When reference is made to LRP6, all forms of LRP6
including a wild type variant or homolog are also contemplated. When discussing the
Frizzled4/LRP5 complex, the complex is also meant to include Frizzled4/LRP6 and
Frizzled4/LRP5/LRP6. LRP6 modulating agents include LRP6 agonists and
antagonists. LRP6 mimetics are also contemplated herein for use in modulating the
Wnt pathway in a manner to enhance bone growth.
By "Wnt" is meant to include any Wnt (wingless-type MMTV integration site
family member) protein and nucleic acid including those of Wntl -Wntl 9. Exemplary
Wnt forms include Wntl (also known as wingless-type MMTV integration site family
member 1, INTl, and Wnt-1 proto-oncogene protein precursor), Wnt3 (also known as
wingless-type MMTV integration site family member 3, INT4, and Wnt-3 proto-
oncogene protein precursor), Wnt3a (also known as wingless-type MMTV integration
site family member 3A and Wnt-3a protein precursor), and Wnt 10b (also known as
wingless-type MMTV integration site family member 10B, Wnt-10b protein
precursor, WNT-12, Wnt-12, and WNT-I2). Variants of any of the Wnt forms are
also contemplated. Wnt modulating agents include Wnt agonists and antagonists.
By "variant" is meant to include a form of a nucleic acid encoding a protein,
wherein the protein has biological activity in the Wnt cascade, and is involved in
modulation of bone metabolism and lipid metabolism. This can include augmented
variants of LRP5, such as the G171V variant that produces a high bone mass in the
human expressing this protein.
By "biologically active fragment", "polypeptide fragment", and "biologically
active polypeptide" are meant a biologically active fragment of LRP5, LRP6, HBM,
Kremenl, Kremen2, any Dkk, any Wnt, and Norrin, wherein such activity modulates
the Wnt pathway, and preferably the Wnt pathway with regard to bone development,
bone modulation, and/or metabolism of a lipid. These are domains of the complete
proteins that are involved with Wnt signaling, and thereby Wnt pathway induced
modulation of lipids and/or bone development. For example, biologically active
polypeptides of LRP5 and LRP6 can be the extracellular portion of those proteins
{e.g., amino acids 1-1376 of human LRP5 (GenBank Accession No. NP_002326),
Zmaxl, or HBM). Additionally, for human LRP5, which is 1615 amino acids in
length, other domains with biological activity may can include the transmembrane
domain (amino acids 1385 to 1407), the cytoplasmic domain (amino acids 1408 to
1615), and the extracellular domain (amino acids 1-1384 or 20-1384 if the first 19
amino acids of the signal peptide are removed). For human LRP6, which is 1613
amino acids long, would have analogous domains: extracellular domain (amino acids

1-1370 or 20-1370 if the first 19 amino acids of the signal peptide are removed),
transmembrane domain (amino acids 1371-1393), and the cytoplasmic domain (amino
acids 1394-1613). The extracellular cysteine rich domain (CRD) of Frizzled4 has
been shown to interact with Norrin, i.e., amino acids 36-165 (Accession No.
IPR000024; GenBank Accession No. NP_036325; Xu et al., 2004 Cell 116: 883-895);
thus a biologically active polypeptide of Frizzled4 could contain the CRD. A
biologically active polypeptide of Norrin could include the CRD domain of Norrin,
e.g., amino acids 15-150. In another example, it has been reported that for a Dkk
protein to bind to Kremenl or Kremen2, the entire extracellular domain is required,
e.g., amino acids 1-362 for human Kremen2 (GenBank Accession No. BAC00872).
Thus, biologically active portions of Kremenl and Kremen2 would contain at least the
extracellular domain, as well as longer sequences thereof. For Dkkl, for example, a
biologically active polypeptide would contain at least the C-termina! cysteine rich
domain (amino acids 183-266 for human Dkkl; GenBank Accession No.
AAQ89364). It is known that the C-termina! cysteine rich domain is involved in the
binding of LRP5 and LRP6 to Kremen2. Thus, for any biologically active polypeptide
of a Dkk protein, the polypeptide could contain at least the cysteine rich domain of
each Dkk. However, other examples include polypeptides containing the cysteine rich
domain of a Dkk protein as well as, for example in Dkkl, sequences both to the N-
terminal and/or C-terminal ends of the cysteine rich domain of Dkkl. Similar
sequences would be contemplated for the other Dkks. Such biologically active
fragments can also include complete proteins minus one or more amino acids at either
the carboxy terminus, or amino terminus, or within the polypeptide that forms the
protein, but which have the same activity as the full-length protein and wherein such
biologically active polypeptide fragments do not act as blocking inhibitor when
compared with activity induced by the full-length polypeptide.
By a "lipid parameter" is meant to include, but is not limited to an in vitro or
in vivo measured parameter to analyze a change of lipid concentration based on
exposure to a reagent. The lipid parameter can include measurement of apoE, HDL,
LDL, VLDL, triglyceride, cholesterol, the number of adipocytes, a change in
adipocyte gene expression, or a combination of these parameters. A lipid parameter is
also meant to include ratios of, for example HDL: VLDL. If studying in vivo changes,
lipid profiles can be done such as fasting lipid profiles (total cholesterol, triglycerides,
LDL and HDL) to assess modulation, of lipid levels due to administration of a test
reagent.
By "lipid disorders", "lipid diseases," and "lipid conditions" which may be
mediated by Norrin are meant to include but are not limited to familial lipoprotein
lipase deficiency, familial apoprotein CII deficiency, familial type 3
hyperlipoproteinemia, familial hypercholesterolemia, familial hypertriglyceridemia,

multiple lipoprotein-type hyperlipidemia, elevated lipid levels due to dialysis and/or
diabetes, and elevated lipid levels of unknown etiologies.
"Bone development" generally refers to any process involved in the change of
bone over time, including, for example, normal development, changes that occur
during a disease state, and changes that occur during aging or changes in hormonal
pattern. This may refer to structural changes and dynamic rate changes such as
growth rates, resorption rates, bone repair rates, and etc. "Bone development
disorder" particularly refers to any disorders in bone development including, for
example, changes that occur during disease states and changes that occur during
aging. Bone development may be progressive or cyclical in nature. Aspects of bone
that may change during development include, for example, mineralization, formation
of specific anatomical features, and relative or absolute numbers of various cell types.
Other bone disorders contemplated that may not be tied to development include but
are not limited to age related loss of bone, bone fractures (e.g., hip fracture, Colle's
fracture, vertebral crush fractures), chondrodystrophies, drug-induced disorders (e.g.,
osteoporosis due to administration of glucocorticoids or heparin, and osteomalacia
due to administration of aluminum hydroxide, anticonvulsants, or glutethimide), high
bone turnover, hypercalcemia, hyperostosis, osteogenesis imperfecta, osteomalacia,
osteomyelitis, osteoporosis, Paget's disease, osteoarthritis, and rickets.
"Bone modulation" or "modulation of bone formation" refers to the ability to
affect any of the physiological processes involved in bone remodeling, as will be
appreciated by one skilled in the art, including, for example, bone resorption and
appositional bone growth, by amongst other things, osteoclastic and osteoblastic
activity, and may comprise some or all of bone formation and development as used
herein.
Bone is a dynamic tissue that is continually adapting and renewing itself
through the renewal of old or unnecessary bone by osteoclasts and the rebuilding of
new bone by osteoblasts. The nature of the coupling between these processes is
responsible for both the modeling of bone during growth as well as the maintenance
of adult skeletal integrity through remodeling and repair to meet the everyday needs of
mechanical usage. There are a number of diseases that result from an uncoupling of
the balance between bone resorption and formation. With aging there is a gradual
"physiologic" imbalance in bone turnover, which is particularly exacerbated in women
due to menopausal loss of estrogen support that leads to a progressive loss of bone.
As bone mineral density falls below population norms, there is a consequential
increase in bone fragility and susceptibility to spontaneous fractures. For every 10
percent of bone that is lost, the risk of fracture doubles. Individuals with bone mineral
density (BMD) in the spine or proximal femur 2.5 or more standard deviations below
normal peak bone mass are classified as osteoporotic. However, osteopenic

individuals with BMD between 1 and 2.5 standard deviations below the norm are also
at risk.
Bone is measured by several different forms of X-ray absorptiometry. All of
the instruments measure the inorganic or bone mineral content of the bone. Standard
DXA measurements give a value that is an areal density, not a true density
measurement by the classical definition of density (mass/unit volume). Nevertheless,
this isthe type of measurement used clinically to diagnose osteoporosis. However,
while BMD is a major contributing factor to bone strength, as much as 40% of bone
strength stems from other factors including but not limited to: (1) bone size (i.e.,
larger diameters increase organ-level stiffness, even in the face of lower density); (2)
the connectivity of trabecular structures; (3) the level of remodeling (remodeling loci
are local concentrators of strain); and (4) the intrinsic strength of the bony material
itself, which in turn is a function of loading history (i.e., through accumulated fatigue
damage) and the extent of collagen cross-linking and level of mineralization. There is
good evidence that all of these strength/fragility factors play some role in osteoporotic
fractures, as do a host of extraskeletal influences as well (such as but not limited to
fall patterns, soft tissue padding, and central nervous system reflex responsiveness).
Additional analytical instruments can be used to address these features of
bone. For example, the pQCT allows measurement of separate trabecular and cortical
compartments for size and density. The μCT (micro CT) provides quantitative
information on architectural features such as trabecular connectivity. The μCT also
gives a true bone density measurement With these tools, the important non-BMD
parameters can be measured for diagnosing the extent of disease and the efficacy of
treatments. Current treatments for osteoporosis are based on the ability of drugs to
prevent or retard bone resorption. Although newer anti-resorptive agents are proving
to be useful in the therapy of osteoporosis, they are viewed as short-term solutions to
the more definitive challenge to develop treatments that will increase bone mass
and/or the bone qualify parameters mentioned above. Thus, bone modulation may be
assessed by measuring parameters such as bone mineral density (BMD) and bone
mineral content (BMC) by pDXA X-ray methods, bone size, thickness or volume as
measured by X-ray, bone formation rates as measured, for example, by calcien
labeling, total, trabecular, and mid-shaft density (as measured by pQCT and/or μCT
methods), connectivity and other histological parameters (as measured by μCT
methods), mechanical bending and compressive strengths (as preferably measured in
femur and vertebrae respectively). Thus, measurable parameters include but are not
limited to bone density, bone strength, trabecular number, bone size, and bone tissue
connectivity. Due to the nature of these measurements, each may be more or less
appropriate for a given situation as the skilled practitioner will appreciate.
Furthermore, parameters and methodologies such as a clinical history of freedom front

fracture, bone shape, bone morphology, connectivity, normal histology, fracture repair
rates, and other bone quality parameters are known and used in the art. Most
preferably, bone quality may be assessed by the compressive strength of vertebra
when such a measurement is appropriate. Bone modulation may also be assessed by
rates of change in the various parameters. Most preferably, bone modulation is
assessed at more than one age. Compounds can be assessed over any one or more of
the parameters listed herein for determining modulation of bone density.
"Normal bone density" refers to a bone density within two standard deviations
of a Z score of 0 in the context of the HBM linkage study. In a general context, the
range of normal bone density parameters is determined by routine statistical methods.
A normal parameter is within about 1 or 2 standard deviations of the age and sex
normalized parameter, preferably about 2 standard deviations. A statistical measure
of meaningfulness is the P value which can represent the likelihood that the associated
measurement is significantly different from the mean. Significant P values are P <
0.05, 0.01, 0.005, and 0.001, preferably at least P <0.01.
The terms "force", "load", "stress" and "strain" are used interchangeably herein
and are related to the principles of force, which in mechanics is any action that tends
to maintain or alter the position of a body or to distort it and this term is used
interchangeably with load in this document. Force as a measure per unit area is
defined as "stress," and is also referred to herein as "mechanical stress" and can be
classified as compressive, tensile or shear depending on how the forces (load) are
applied. Specifically, compressive stresses are developed if loads are applied so that
the material becomes shorter, whereas tensile stresses are developed when the
material is stretched. Shear stresses are developed when one region of a material
slides relative to an adjacent region. The result of stress is defined as deformation and
the percentage of the relative deformation or change in length is termed "strain". If for
example a material is stretched to 101% of its original length it has a strain of 0.01 or
1%. Since strain has no units it is either reported as relative deformation where a
strain of 0.01 is equal to 1% deformation or in terms of microstrain where 10,000
microstrain is equal to 0.01 strain or 1% deformation (Turner et al., 1993 Bone, 14:
595-608).
By "test agent," and "test reagent" is meant to include small compounds,
compositions, peptides, mimetics, polypeptides, siRNAs, and immunoglobulins.
Compositions include combinations of two or more active compounds, wherein one or
more of the active compounds are Wnt pathway (cascade) modulators.
By "immunoglobulins" is meant to include antibodies and antibody fragments.
As used herein, the term "antibody" is meant to refer to complete, intact antibodies,
diabodies, and antibody fragments such as Fab fragments, Fab', and F(ab)2 fragments.
Complete antibodies include monoclonal antibodies (mAb), such as murine

monoclonal antibodies, chimeric antibodies, humanized antibodies, primatized
antibodies, and human antibodies. The production of antibodies and genetically
engineered or enzymatically produces portions of antibodies and the organization of
the genetic sequences that encode such molecules are well known and are described,
for example, in Harlow et al., ANTIBODIES: A LABORATORY MANUAL, Cold Spring
Harbor Laboratory, Cold Spring Harbor, N.Y. (1988); Harlow et al., USING
ANTIBODIES: A LABORATORY MANUAL, (Cold Spring Harbor Press, New York, 1998);
and Breitling et al., RECOMBINANT ANTIBODIES (Wiley-Spektrum, 1999), which are
incorporated herein by reference for all purposes. Immunoglobulins also include
fragments such as scFv.
By "immunologically active" is meant any immunoglobulin protein or
fragment thereof which recognizes and binds to an antigen. Preferably, the
immunologically active protein or fragment thereof modulates the antigen to which it
binds. For example, if it binds to Norrin or to a ligand of Norrin, the immunologically
active protein or fragment thereof would modulate Norrin activity or the activity of the
Norrin ligand.
"Single-chain Fvs" ("scFvs") are recombinant antibody fragments consisting
of only the variable light chain (VL) and variable heavy chain (VH) covalently
connected to one another by a polypeptide linker. Either VL or VH may be the NH2 -
terminal domain. The polypeptide linker may be of variable length and composition
so long as the two variable domains are bridged without serious steric interference.
Typically, the linkers are comprised primarily of stretches of glycine and serine
residues with some glutamic acid or lysine residues interspersed for solubility.
"Diabodies" are dimeric scFvs. The components of diabodies typically have
shorter peptide linkers than most scFvs, and they show a preference for associating as
dinners.
An "Fv" fragment is an antibody fragment that consists of one VH and one VL
domain held together by non-covalent interactions. The term "dsFv" is used herein to
refer to an Fv with an engineered intermolecular disulfide bond to stabilize the VH-VL
pair.
A "F(ab')2" fragment is an antibody fragment essentially equivalent to that
obtained from immunoglobulins (typically IgG) by digestion with the enzyme pepsin
at pH 4.0-4.5. The fragment may also be recombinantly produced.
A "Fab" fragment is an antibody fragment essentially equivalent to that
obtained by reduction of the disulfide bridge or bridges joining the two heavy chain
pieces in the F(ab')2 fragment. The Fab' fragment may also be recombinantly
produced.
The term "protein-capture agent" means a molecule or a multi-molecular
complex, which can bind a protein to itself. Protein-capture agents preferably bind

their binding partners in a substantially specific manner. Protein-capture agents with a
dissociation constant (KD) of less than about 10-6 are preferred (e.g., 10-7, 10-8, 10-9,
10-10). Antibodies or antibody fragments are highly suitable as protein-capture agents.
Antigens may also serve as protein-capture agents, since they are capable of binding
antibodies. A receptor that binds a protein ligand is another example of a possible
protein-capture agent. Protein-capture agents are understood not to be limited to
agents, which only interact with their binding partners through non-covalent
interactions. Protein-capture agents may also optionally become covalently attached
to the proteins, which they bind. For instance, the protein-capture agent may be
photo-crosslinked to its binding partner following binding.
The term "binding partner" means a protein that is bound by a particular
protein-capture agent, preferably in a substantially specific manner. In some cases,
the binding partner may be the protein normally bound in vivo by a protein that is a
protein-capture agent. In other embodiments, however, the binding partner may be the
protein or peptide on which the protein-capture agent was selected (through in vitro or
in vivo selection) or raised (as in the case of antibodies). A binding partner may be
shared by more than one protein-capture agent. For instance, a binding partner that is
bound by a variety of polyclonal antibodies may bear a number of different epitopes.
One protein-capture agent may also bind to a multitude of binding partners (for
instance, if the binding partners share the same epitope).
"Conditions suitable for protein binding" means those conditions (in terms of
salt concentration, pH, detergent, protein concentration, temperature, etc.") which
allow for binding to occur between a protein and its binding partner in solution.
Preferably, the conditions are not so lenient that a significant amount of non-specific
protein binding occurs.
An "array" is an arrangement of entities in a pattern on a substrate. Although
the pattern is often a two-dimensional pattern, the pattern may also be a three-
dimensional pattern for a greater application of the material to the array substrate.
The term "substrate" refers to the bulk, underlying, and core material of the
arrays of the invention. The substrate is the material to which nucleic acids,
antibodies, immunoglobulins and other compounds are affixed.
By "transgenic animal" is meant an animal harboring in its germ line a gene or
nucleic acid that has been introduced by cDNA technology. This can be, for
examples, introduction of human genes into rodents or a mouse gene in a mouse. The
term can include knock-out animals and knock-in animals and combinations, for
example wherein an animal has had its wild-type gene knocked out and then replaced.
The replaced gene can be the native wild-type gene, a cognate gene from another
animal such as a human gene, or a variant such as LRP5. The introduced gene can
also be under control of an inducible promoter. The HBM variant cDNA can be a

native variant or non-native variant. For example, the human HBM variant of G171V
can be introduced into a mouse. Alternatively, the mouse counterpart to G171V can
also be introduced into a mouse yielding a transgenic HBM mouse expressing a native
HBM variant. A transgenic animal is not meant to include transgenic humans, but can
include non-human primates and other animals. The transgenic animal can have
knocked out or introduced any one or more Dkk, Norrin, LRP5, LRP6, Kremen, Wnt,
or Frizzled4. A transgenic animal is contemplated to be a non-human animal, but can
include non-human primates.
By "LRP5 transgenic animal" is means to include an animal expressing both
the native and a cDNA form of LRP5 or only a cDNA form of LRP5 if the animal has
the native form of LRP5 removed or incapable of function (knocked out). The cDNA
form of LRP5 may be under an inducible element. The animal can be one wherein the
native gene is knocked out and a native or non-native LRP5 has been introduced, or
knocked-in. These knock-in animals again can have the genes preferably under
inducible control.
By an "HBM transgenic animal" is meant an animal wherein the native LRP5
is present or knocked out and a cDNA encoding the HBM variant is present.
By "effective amount" or "dose effective amount" or "therapeutically effective
amount" is meant an amount of an agent which modulates a biological activity of
Norrin sufficient to modulate a bone parameter and/or a lipid parameter.
The term "recognizes and binds," when used to define interactions of antisense
nucleotides, siRNAs (small inhibitory RNA), or shRNAs (short hairpin RNAs) with a
target sequence, means that a particular antisense, siRNA, or shRNA sequence is.
substantially complementary to the target sequence, and thus will specifically bind to a
portion of an mRNA encoding polypeptide. As such, typically the sequences will be
highly complementary to the mRNA target sequence, and will have no more than 1, 2,
3, 4, 5, 6, 7, 8, 9, or 10 base mismatches throughout the sequence. In many instances,
it may be desirable for the sequences to be exact matches, i.e. be completely
complementary to the sequence to which the oligonucleotide specifically binds, and
therefore have zero mismatches along the complementary stretch. Highly
complementary sequences will typically bind quite specifically to the target sequence
region of the mRNA and will therefore be highly efficient in reducing, and/or even
inhibiting the translation of the target mRNA sequence into polypeptide product.
Substantially complementary oligonucleotide sequences will be greater than
about 80 percent complementary (or "% identity") to the corresponding mRNA target
sequence to which the oligonucleotide specifically binds, and will, more preferably be
greater than about 85 percent complementary to the corresponding mRNA target
sequence to which the oligonucleotide specifically binds. In certain aspects, as
described above, it will be desirable to have even more substantially complementary

oligonucleotide sequences for use in the practice of the invention, and in such
instances, the oligonucleotide sequences will be greater than about 90 percent
complementary to the corresponding mRNA target sequence to which the
oligonucleotide specifically binds, and may in certain embodiments be greater than
about 95 percent complementary to the corresponding mRNA target sequence to
which the oligonucleotide specifically binds, and even up to and including 96%, 97%,
98%, 99%, and even 100% exact match complementary to the target mRNA to which
the designed oligonucleotide specifically binds.
Percent similarity or percent complementary of any of the disclosed sequences
may be determined, for example, by comparing sequence information using the GAP
computer program, version 6.0, available from the University of Wisconsin Genetics
Computer Group (UWGCG). The GAP program utilizes the alignment method of
Needleman and Wunsch, 1970 J. Mol. Biol. 48(3): 443-53. Briefly, the GAP program
defines similarity as the number of aligned symbols (i.e., nucleotides or amino acids)
which are similar, divided by the total number of symbols in the shorter of the two
sequences. The preferred default parameters for the GAP program include: (1) a
unary comparison matrix (containing a value of 1 for identities and 0 for non-
identities) for nucleotides, and the weighted comparison matrix of Gribskov and
Burgess (1986 Nucleic Acids Res. 14(1): 327-34), (2) a penalty of 3.0 for each gap and
an additional 0.10 penalty for each symbol in each gap; and (3) no penalty for end
gaps.
By "mimetic" is meant a molecule that performs the same function or behaves
similarly to the mimicked agent or has an activity that is enhanced to that of the agent
being mimicked. For example, a Norrin mimetic would interact with LRP5 and/or
LRP6 and Frizzled4 as the Norrin polypeptide does and modulate bone mass and/or
lipid levels like Norrin or at an enhanced level to that observed for Norrin. For
example, the mimetic could induce a high bone mass like phenotype, such as observed
for the HBM phenotype (which results for example from the G171V mutation in the
human LRP5 polypeptide, or the cognate location in another vertebrate LRP5). The
mimetic molecule can be a polypeptide, peptide, immunoglobulin, or a small chemical
compound.
By "reporter element" is meant a polynucleotide that encodes a polypeptide
capable of being detected in a screening assay. Examples of polypeptides encoded by
reporter elements include, but are not limited to, lacZ, GFP, YFP (or other fluorescent
reporter), luciferase, and chloramphenicol acetyltransferase.
By "cell" or "host cell" is meant to include vertebrate cells, as well as yeast
cells or certain prokaryotic cells for use in screening assays. For example, a suitable
cell may be a yeast cell in a yeast two hybrid assay.
By "bone cell" is meant to include cells from tissue culture ("cultured cell") or

cells obtained from bone tissue. Such cells include but are not limited to osteoblasts,
preosteoblasts, osteoprogenitor cells, osteoclasts, osteocytes, mesenchymal stem cells,
any of the cells discussed herein, or any combination thereof. By bone tissue would
mean to include a combination of these cells, as may be obtained from a bone biopsy.
By "Dkk antagonist" is meant to include but not limited to monoclonal or
polyclonal antibodies or immunogenically active fragments thereof, peptide aptamers,
a GSK binding protein, an antisense molecule to a GSK nucleic acid, an RNA
interference molecule, a morpholino oligonucleotide, a peptide nucleic acid (PNA), a
ribozyme, and a peptide that can inhibit Dkk activity in the Wnt pathway.
Likewise, by "Kremen antagonist" is meant to include but not limited to
monoclonal or polyclonal antibodies or immunogenic active fragments thereof,
peptide aptamers, an RNA interference molecule, a morpholino oligonucleotide, a
peptide nucleic acid (PNA), a ribozyme, and a peptide that can inhibit Kremen activity
in the Wnt pathway.
By "Wnt 3 A agonist" is meant to include reagents which can up regulate Wnt
3A synthesis and/or activity. By "Wnt 3A mimetic" is meant a molecule that mimics
Wnt3A activity. By "Wnt 3A variant" would include any functional variant which
when administered with load can-enhance activation with a Wnt/p-catenin response.
The term "fusion protein" refers to a protein composed of two or more
polypeptides that, although typically not joined together in their native state, are
joined by their respective amino and carboxyl termini through a peptide linkage to
form a single continuous polypeptide. It is understood that the two or more
polypeptide components can either be directly joined or indirectly joined through a
peptide linker/spacer.
2. Assays for Screening Test Agents Which Modulate Norrin
The materials and methods disclosed herein are directed in part to methods of
screening agents that modulate NDP genes or the Norrin proteins encoded by those
genes or identifying Norrin mimetics and Norrin agonists. The assays are also
directed to materials and methods of screening agents that modulate reagents that
interact with Norrin proteins, identifying Norrin agonists, and identifying Norrin
mimetics. These could be reagents that by binding with Frizzled4, modulate Norrin
activity. These can also be reagents which by modulating Dkkl activity (either the
gene or the protein) modulate Norrin activity. Another example would be Kremen2 ,
. wherein modulation of Kremen2 (either the gene or the protein) would modulate
Norrin activity. In the instances of Dkkl and Kremen2, preferably the reagent
modulating the activity of these compounds would be an antagonist of Dkkl or
Kremen2. Preferably, the assays would result in reagents that also modulate LRP5
activity via Frizzled4 and Norrin interaction. Preferable LRP5 modulation would be

in the form of enhanced activity, such as that produced by an agonist, or a mimetic
{e.g., a Norrin mimetic or Frizzled4 mimetic).
Assay systems can include a step wherein test agents are screened for their
ability to bind to Frizzled4, Norrin, Dkkl, Kremen2, LRP5, or act as a Norrin
mimetic. This can be any system both involving substrates or free in solution,
wherein binding of the test agents to any of the aforementioned substrates is allowed
to occur and then assayed to determined binding. Test agents can be admixed with
Frizzled4, Norrin, Dkkl, Kremen2, or LRP5 under physiological conditions (e.g., pH
about 7.0 to about 7.4; 24°C to about 40°C) for a sufficient period of time to permit
binding, e.g., about 1 minute to 6 hours.
Test agents can be screen for binding as discussed above or can be candidates
from any chemical library. The test agents can then be assayed in a cell-based assay
system. Such a cell-based assay system can be one wherein the cells are transiently or
stably transfected with a nucleic acid encoding at least one of the following: Frizzled4,
Norrin, Dkkl, Kremen2, or LRP5, or any combination thereof. Thus, cells would
individually express at least Frizzled4 and Norrin, as well as the remaining three
genes. There would also be a series of cells stably or transiently co-transfected with
the following combinations of nucleic acids:
(a) Norrin and LRP5 and/or LRP6
(b) Norrin, a Dkk {e.g., Dkk 1 to Dkk4) and LRP5 and/or LRP6
(c) Norrin, a Kremen {e.g., Kremen 1 or 2) and LRP5 and/or LRP6
(d) Norrin, a Kremen, a Dkk, and LRP5 and/or LRP6
(e) Frizzled4 and Norrin;
(f) Frizzled4, Norrin, and LRP5
(g) Frizzled4, Norrin, and a Dkk (e.g., Dkkl to Dkk4);
(h) Frizzled4, Norrin, and a Kremen (e.g., Kremen 1 and/or 2);
(i) Frizzled4, Norrin, a Dkk, and Kremen2;
(j) Frizzled4, Norrin, LRP5, and Dkkl;
(k) Frizzled4, Norrin, LRP5, and Kremen2; and/or
(1) Frizzled4, Norrin, LRP5, Dkkl, and Kremen2, and/or
(m) or any combination.
Also contemplated for any of the above combinations are LRP6, HBM, other Dkks
(e.g., Dkk2, Dkk3, and/or Dkk4), Wnts, and/or Kremen 1.
It would be understood by one of ordinary skill that such an assay system may
also require vector controls, wherein the vector is that in which the nucleic acid
encoding any of the above proteins is operably linked for expression in the cells. The
vector control can consist of the transient or stable transfection of cells with only the
vector and/or with no vector. Transient transfection and stable transfection of cells
can be performed using techniques known in the art. See, e.g., Sambrook et al.,

MOLECULAR CLONING: A LABORATORY MANUAL, volumes 1-3 (3rd ed., Cold Spring
Harbor Press, NY 2001) or any of the prior editions by Sambrook. Co-transfections
can be prepared either transiently or stably as known in the art. Sambrook et al., 200)
Nucleic acids encoding Frizzled4, Norrin, LRP5, Dkkl, and Kremen are listed
in part below along with their associated protein sequences. The embodiments of this
application are not limited to the sequences disclosed herein.



Additional HBM and LRP5 sequences (e.g., Zmaxl) are disclosed in U.S.
Application Nos. 09/544,398 (now U.S. Patent No. 6,770,461) and 10/240,851, which
are herein incorporated by reference in their entirety for all purposes.
The cells which can be transfected can be any mammalian cell line. Preferable
cell lines are human cell lines, especially when using nucleic acids which encode
human proteins for any of the above. Thus, transfections can occur for mouse nucleic
acids in mouse lines or for human nucleic acids in human lines. Cell lines can be
bone cell lines, kidney cell lines stem cell lines from humans or other vertebrates.
Exemplary kidney cell lines include but are not limited to HEK-293 cells (ATCC®
No. CRL-1573) and HepG2 cells. Exemplary bone cell lines include but are not
limited to KHOS/NP (R-970-5) (ATCC® No. CRL-1544), KHOS-240S (ATCC® No.
CRL-1545), KHOS-321H (ATCC® No. CRL-1546), DSDh (ATCC® No. CRL-
2131), VA-ES-BJ (ATCC® No. CRL-2138), 7F2 (ATCC® No. CRL-12557), U-2 OS
(also known as U2OS; ATCC® No. HTB-96), HOSTE85, ROS, MC3T3-E6, UMR-
106, Saos2, MG63, and HOBs. Exemplary stem cell lines include but are not limited
to human adult mesenchymal stem cells (CambrexBioscience) and the mouse stem
cell line, C3H10T1/2 (ATCC).
The nucleic acids encoding any of the proteins would include the open reading
frames (ORFs), as well as any transcriptional information necessary for transcription
and translation. The nucleic acids encoding the proteins would in turn be operably
1 inked to a vector suitable for stable and/or transient transfection in a cell. Suitable
vectors include but are not limited to TK-renilla, pcDNA3.1 (Invitrogen), and pUSE
(Upstate Biotech). Other operable vectors capable of expression in vertebrate cells
may be used.

Any reporter system that provides information on the regulation of genes and
their associated proteins can be utilized, including but not limited to TK-renilla, P-
galactosidase (β-gal), alkaline phosphatase, green fluorescent protein (GFP), or other
fluorescent protein marker. A preferred system, as described herein, is the
combination of TCF-Iuci and TK-renilla as described in the examples. Other reporter
and vector combinations operative in vertebrate cells may also be utilized.
In one aspect, the relative amounts of Norrin or a Norrin interacting protein of
a cell population that has been exposed to the agent to be tested is compared to an un-
exposed control cell population. Antibodies can be used to monitor the differential
expression of the protein in the different cell populations. Cell lines or populations
are exposed to the agent to be tested under appropriate conditions and time. Cellular
lysates may be prepared from the exposed cell line or population and a control,
unexposed cell line or population. The cellular lysates are then analyzed with the
probe, as would be known in the art. See, e.g., Ed Harlow and David Lane,
ANTIBODIES: A LABORATORY MANUAL (Cold Spring Harbor, NY, 1988) and Ed
Harlow and David Lane, USING ANTIBODIES: A LABORATORY MANUAL (Cold Spring
Harbor, NY 1998).
For example, N- and C- terminal fragments of Norrin can be expressed in
bacteria and used to search for proteins which bind to these fragments. Fusion
proteins, such as His-tag or GST fusion to the N- or C-terrninal regions of Norrin (or
to biologically active domains of Norrin) or a whole Norrin protein can be prepared.
These fusion proteins can be coupled to, for example, Talon or Glutathione-Sepharose
beads and then probed with cell lysates to identify molecules which bind to Norrin.
Prior to lysis, the cells may be treated with purified Wnt proteins, RNA, or drugs
which may modulate Wnt signaling or proteins that interact with downstream
elements of the Wnt pathway. Lysate proteins binding to the fusion proteins can be
resolved by SDS-PAGE, isolated and identified by, for example protein sequencing or
mass spectroscopy, as is known in the art. See, e.g., PROTEIN PURIFICATION
APPLICATIONS: A PRACTICAL APPROACH (Simon Roe, ed., 2nd ed. Oxford Univ. Press,
2001) and "Guide to Protein Purification" in Meth. Enzymology vol. 182 (Academic
Press, 1997).
The activity of Norrin, a Norrin mimetic, a Norrin interacting protein (e.g.,
Norrin agonist or Norrin antagonist), or a complex of Norrin with LRP5/LRP6/HBM
and/or a Kremen protein or Dkk protein may be affected by compounds which
modulate the interaction between Norrin and a Norrin interacting protein, and/or
Norrin and LRP5/LRP6/HBM, Norrin and/or Frizzled4, a Dkk protein, or a Kremen
protein. Provided herein are methods and research tools for the discovery and
characterization of these compounds. The interaction between Norrin or a Norrin
mimetic and a Norrin/Frizzled4 interacting protein and/or Norrin and LRP5/6/HBM,

and Norrin/Dkk, and Norrin/Kremen may be monitored in vivo and in vitro. Similar
assays can also be used for assessing Norrin agonists and antagonists. Compounds
which modulate the stability of a Norrin/Fz4 complex are potential therapeutic
compounds.
Example in vitro methods include: binding LRP5/6/HBM, Norrin/Fz4, or a
Norrin/Fz4 interacting protein to a sensor chip designed for an instrument such are
made by Biacore (Uppsala, Sweden) for the performance of a plasmon resonance
spectroscopy observation. For example, using this method, a chip with one of
"Norrin/Fz4, a Norrin/Fz4 interacting protein, or LRP5/LRP6 can be first exposed to
the other under conditions which permit them to form a complex. A test compound is
then introduced, and the output signal of the instrument provides an indication of any
effect exerted by the test compound. By this method, compounds may be rapidly
screened. This method can be used for any Norrin/Fz combination with LRP5, LRP6,
HBM, any Dkk, any Kremen, any Wnt, and any combination thereof.
Another, in vitro, method is exemplified by the SAR-by-NMR methods
(Shuker et al., 1996 Science 274: 1531-4). For example, a Norrin/Fz4 binding domain
and/or LRP5/LRP6/HBM LBD can be expressed and purified as l5N-labeled protein
by expression in labeled media. The labeled protein(s) are allowed to form the
complex in solution in a nuclear magnetic resonance (NMR) sample tube. The
heteronuclear correlation spectrum in the presence and absence of a test compound
provides data at the level of individual residues with regard to interactions with the
test compound and changes at the protein-protein interface of the complex. This
method can be used with any Norrin/Frizzled4 combination with LRP5, LRP6, HBM,
any Dkk, any Kremen, any Wnt, and any combination thereof.
One of skill in the art knows of many other protocols, e.g., affinity capillary
electrophoresis (Okun et al., 2001 J. Biol. Chem. 276: 1057-1062), fluorescence
spectroscopy, electron paramagnetic resonance, etc., which can also be used to
monitor the modulation of a complex and/or measure binding affinities for complex
formation in the presence and absence of a test agent for any of the above listed
combination of proteins or biologically active fragments thereof.
Protocols for monitoring the modulation of a Norrin/Frizzled4 interaction, a
Norrin/LRP5/Frizzled4 interaction, or a Norrin mimetic's interaction with any one or
more of LRP5, LRP6, HBM, Kremen 1, Kremen2, any Dkk, and any Wnt can be
performed using a yeast hybrid protocol. The yeast two- or more hybrid method may
be used to monitor the modulation of a complex by monitoring the expression of
genes activated by the formation of a complex of fusion proteins of Norrin/Frizzled4
and/or any of the above-listed other proteins. If using LRP5, LRP6, or HBM, then the
complete protein can be used or the ligand binding domains (LBDs) or portions of the
beta propeller containing the YWTD repeats. Nucleic acids according to the invention

which encode the interacting Norrin and Frizzled4 or Norrin and LRP5/LRP6/HBM
LBD domains are incorporated into bait and prey plasmids. The yeast two hybrid
(Y2H) method or yeast hybrid method for three or more proteins is performed in the
presence of one or more test compounds. The modulation of the complex is observed
by a change in expression of the complex activated gene. It will be appreciated by one
skilled in the art that test compounds can be added to the assay directly or, in the case
of proteins, can be co-expressed in the yeast with the bait and prey compounds.
Similarly, fusion proteins of Norrin and Norrin interacting proteins can also be used in
an Y2H screen to identify other proteins which modulate the Norrin/Frizzled4
complex (such as Dkk, Kremen, other negative regulators, and positive regulators).
Yeast hybrid technologies are known in the art. See for example, LI ZHU AND
GREGORY J. HANNON, YEAST HYBRID TECHNOLOGIES (2000).
Assay protocols such as these may be used in methods to screen for
compounds, drugs, treatments which modulate the Norrin/Frizzled4 complex, whether
such modulation occurs by competitive binding, acting as a Norrin mimetic, or by
altering the structure of the Norrin/Frizzled4 complex, or by stabilizing or
destabilizing the protein-protein interface. It may be anticipated that peptide aptamers
may competitively bind, although induction of an altered binding site structure by
steric effects is also possible. As used herein, a biological or pathological process
modulated by Norrin/Frlzz!ed4 and the Norrin/Fz4/LRP5 complex may include
binding of Norrin to Frizzled4, or to a protein that interacts with the
Norrin/Frizzled4/LRP5 complex, or prevents Dkk and/or Kremen down regulation of
the Norrin/Frizzled4 complex. This can include compounds that interact with the
Norrin or modulate synthesis of the proteins involved with the complex as well as
Norrin mimetics.
Further bone-related markers may be observed such as alkaline phosphatase
activity, osteocalcin production, or mineralization in addition to other bone related
factors that can be assessed in conjunction with the biochemical analysis of
modulation of the Norrm/Frizzled4/LRP5 complex, as discussed herein.
Pathological processes refer to a category of biological processes that produce
a deleterious effect. For example, expression or up-regulation of expression of LRP5
or LRP6 and/or Dkk and/or a Dkk interacting protein may be associated with certain
diseases or pathological conditions. As used herein, an agent is said to modulate a
pathological process when the agent alters the process from its base level in the
subject to a statistically significant level. For example, the agent may reduce the
degree or severity of the process mediated by that protein in the subject to which the
agent was administered. For instance, a disease or pathological condition may be
prevented, or disease progression modulated by the administration of agents which

reduce or modulate in some way the expression or at least one activity of a protein of
the invention.
As Frizzled4/Norrin and LRP5/LRP6 (as well as Kremen, Dkk, and Wnt) are
involved directly and/or indirectly in bone mass modulation, one embodiment of this
invention is to use "Norrin/Frizzled4 complex and Norrin/Frizzled4 complex ligands as
a method of diagnosing a bone condition or disease. Certain markers are associated
with specific Wnt signaling conditions (e.g., TCF/LEF activation). Diagnostic tests
for bone conditions may include the steps of testing a sample or an extract thereof for
the presence of Dkk or Dkk interacting protein nucleic acids {i.e., DNA or RNA),
oligomers or fragments thereof or protein products of TCF/LEF regulated expression.
For example, standard in situ hybridization or other imaging techniques can be
utilized to observe products of Wnt signaling.
Also discussed herein are methods and materials for modulating bone
development or bone loss conditions. Inhibition of bone loss may be achieved by
inhibiting or modulating changes in the Norrin/Frizzled4 complex and thereby the
Wnt signaling pathway. For example, absence of Norrin activity or increased Dkkl
activity may be associated with low bone mass. Increased activity Norrin and
Frizzled4 may be associated with high bone mass. Therefore, modulation of
Norrin/Frizzled4 activity will in turn modulate bone mass. Modulation of a Dkk's
interaction with the Norrin/Frizzled4 complex via agonists and antagonists is one
embodiment of a method to regulate bone development.
The agents of the present invention can be provided alone, or in combination
with other agents that modulate a particular pathological process. As used herein, two
agents are said to be administered in combination when the two agents are
administered simultaneously or are administered independently in a fashion such that
the agents will act at the same time.
The agents of the present invention can be administered to a non-human test
animal for example via parenteral, subcutaneous (s.c), intravenous (i.v.),
intramuscular (i.m.), intraperitoneal (i.p.), transdermal or buccal routes. Alternatively,
or concurrently, administration may be by the oral route. The dosage administered
will be dependent upon the age, health, and weight of the recipient, kind of concurrent
treatment, if any, frequency of treatment, and the nature of the effect desired.
The present invention further provides compositions containing one or more
agents that modulate expression or at least one activity of Norrin or the
Norrin/Frizzled4 complex or which act as a Norrin mimetic. While individual needs
vary, determination of optimal ranges of effective amounts of each component is
within the skill of the art. Typical dosages of the active agent, which include a Norrin
mimetic or an agent that mediates Norrin, a Norrin interacting protein, or a ligand of
the Norrin/Frizzled4 complex (or Norrin/Frizzled4/LRP5 complex, which is also

contemplated throughout wherein Norrin/Frizzled4 complexes are discussed), may
comprise from about 0.0001 to about 50 mg/kg body weight. The preferred dosages
may comprise from about 0.001 to about 50 mg/kg body weight. The most preferred
dosages may comprise from about 0.1 to about 1 mg/kg body weight. In an average
human of 70 kg, the range would be from about 7 μg to about 3.5 g, with a preferred
range of about 0.5 mg to about 5 mg (and for example any 0.1 mg value within this
range).
In addition to the pharmacologically active agent, the compositions of the
present invention may contain suitable pharmaceutically acceptable carriers
comprising excipients, carriers, and auxiliaries which facilitate processing of the
active compounds into preparations which can be used pharmaceutical ly for delivery
to the site of action. Suitable formulations for parenteral administration include
aqueous solutions of the active compounds in water-soluble form, for example, water-
soluble salts. In addition, suspensions of the active compounds as appropriate oily
injection suspensions may be administered. Suitable lipophilic solvents or vehicles
include fatty oils (e.g., vegetable oils such as sesame oil), or synthetic fatty acid esters
(e.g., ethyl oleate or triglycerides). Aqueous injection suspensions may contain
substances which increase the viscosity of the suspension and include but are not
limited to sodium carboxymethyl cellulose, sorbitol and/or dextran. Optionally, the
suspension may also contain stabilizers. Liposomes and other non-viral vectors can
also be used to encapsulate the agent for delivery into the cell.
The pharmaceutical formulation for systemic administration according to the
invention may be formulated for enteral, parenteral, or topical (top) administration.
Indeed, all three types of formulations may be used simultaneously to achieve
systemic administration of the active ingredient.
Suitable formulations for oral administration include hard or soft gelatin
capsules, pills, tablets, including coated tablets, elixirs, suspensions, syrups or
inhalations and controlled release forms thereof.
Potentially, any compound which binds and thereby modulates Norrin, a
Norrin mimetic, a Norrin ligand, or the Norrin/Frizzled4 complex may be a
therapeutic compound. In one embodiment of the invention, a peptide or nucleic acid
aptamer according to the invention is used in a therapeutic composition. Such
compositions may comprise an aptamer, or aNorrin/Frizzled4 fragment, unmodified
or modified. In another embodiment, the therapeutic compound comprises aNorrin-
orNorrin/Frizz!ed4-compIex-interacting protein, or biologically active fragment
thereof.
Nucleic acid aptamers have been used in compositions for example by
chemical bonding to a carrier molecule such as polyethylene glycol (PEG), which may
facilitate uptake or stabilize the aptamer. A di-alkylglycerol moiety attached to an

RNA can be used to embed the aptamer in liposomes, thus stabilizing the compound.
Incorporating chemical substitutions (i.e., changing the 2'-OH group of ribose to a 2'-
NH in RNA confers ribonuclease resistance) and capping, etc. can prevent breakdown.
Several such techniques are discussed for RNA aptamers in Brody and Gold, 2000
Rev. Mol. Biol. 74:3-13.
Peptide aptamers may be used in therapeutic applications by the introduction
of an expression vector directing aptamer expression into the affected tissue such as
for example by retroviral delivery, by encapsulating the DNA in a delivery complex or
simple by naked DNA injection. Or, the aptamer itself or a synthetic analog may be
used directly as a drug. Encapsulation in polymers and lipids may assist in delivery.
The use of peptide aptamers as therapeutic and diagnostic agents is reviewed by
Hoppe-Syler and Butz, 2000 J. Mol. Med. 78: 426-430.
In another aspect, the structure of a constrained peptide aptamer of the
invention may be determined such as by NMR or X-ray crystallography (Cavanagh et
al., PROTEIN NMR SPECTROSCOPY: PRINCIPLES AND PRACTICE, Academic Press, 1996;
Drenth, PRINCIPLES OF PROTEIN X-RAY CRYSTALLOGRAPHY, Springer Verlag, 1999).
Preferably, the structure can be determined in complex with the target protein. A
small molecule analog is then designed according to the positions of functional
elements of the 3D structure of the aptamer. (GUIDEBOOK ON MOLECULAR MODELING
IN DRUG DESIGN, Cohen, Ed., Academic Press, 1996; MOLECULAR MODELING AND
DRUG DESIGN (TOPICS IN MOLECULAR AND STRUCTURAL BIOLOGY), Vinter and
Gardner Eds., CRC Press, 1994). Methods are provided herein for the identifying and
designing effective and specific drugs that modulate the activity of Norrin, act a
Norrin mimetics, Norrin interacting proteins, Norrin/Frizzled4 interacting proteins,
and the Norrin/Frizzled4 complex. Small molecule mimetics of the peptide aptamers
are also encompassed within the scope.
2.1 Cell-Based Norrin Functional Reporter Assay
The TCF-reporter assays described in the examples below can be developed
into screening assays either to identify Norrin mimetics (FIGS. 1 and 2) or to identify
antagonists of Norrin-signal inhibitors, such as antagonists of Dkkl/Kremen2 (FIG.
3). In both types of assays when preformed in bone or non-bone cell types, the active
molecules would enhance the TCF-luciferase signals or other suitable signal method.
2.2 Norrin/LRP5/LRP6 and PKK/LRF5/LRP6/Kremen Assays
Another method that can be used to screen for reagents that modulate Norrin's
interaction with Frizzled4/LRP5/LRP6 is via an enzyme linked immunosorbent assay
(ELISA) assay. Two possible permutations of this assay are exemplified, but others
can also be utilized. For example, LRP5 can be immobilized to a solid surface, such
as a tissue culture plate well. One skilled in the art would recognize that other
supports, such as but not limited to a nylon or nitrocellulose membrane, a silicon chip,

a glass slide, beads, etc. can be substituted and utilized. One manner of doing this can
be to have the form of LRP5/LRP6/Fz4 as a fusion protein, wherein the extracellular
domain of LRP5/LRP6/Fz4 is fused to the Fc portion of a human IgG or other IgG.
The LRP5/LRP6/Fz4-Fc fusion protein can be produced in Chinese hamster ovary
(CHO) cell (or another suitable cell line), wherein the fusion proteins are extracted
from the cell lines or the media. The isolated LRP5/6/Fz4-Fc fusion protein can be
immobilized on the solid surface via anti-human Fc antibody or by Protein-A or
Protein G-coated plates, for example. The substrate can then be washed to remove
any non-bound protein. Conditioned media containing secreted Norrin protein or
secreted Norrin-epitope tagged protein (or purified Norrin, purified Norrin-epitope
tagged protein, Norrin mimetic, or fragment containing a biologically active portion of
Norrin involved in bone modulation) can be incubated in the wells or containers.
Alternatively, a test reagent can be incubated with the fixed fusion protein in order to
screen for Norrin mimetics. The binding of Norrin or Norrin mimetic to
LRP5/LRP6/Fz4 can be assessed using antibodies to either Norrin or to an epitope tag.
For example, a Norrin-V5 epitope tagged protein or fragment thereof can be detected
with anti-V5 antibody. This assay system can then be used for example to identify
Norrin mimetics, Norrin agonists, and immunoglobulins that bind in a manner like
Norrin to the LRP5/LRP6/Fz4 fusion protein. Assays can be, for example, in the form
of a competitive assay, tagged test reagents, and the like. These assay systems can
also be utilized with HBM. The assay can also be modified to have washings that
include a Dkk, a Kremen, and/or a Wnt protein or biologically active fragment
thereof, when screening for test agents that modulate the interaction and formation of
complexes between these proteins and polypeptides.
Alternatively, the Norrin protein or a biologically active fragment thereof (all
references to Norrin protein assumes that a biologically active fragment can also be
used) or a Norrin mimetic, which is involved in bone modulation could be directly
fused to a detection marker, such as alkaline phosphatase. Here the detection of the
Norrin-LRP5/LRP6/Fz4 interaction can be directly investigated without subsequent
antibody-based experiments. The bound Norrin or Norrin mimetic is detected in an
alkaline phosphatase assay or other detection assay. If the Norrin-alkaline
phosphatase fusion protein is bound to the immobilized LRP5/LRP6/Fz4, alkaline
phosphatase activity would be detected in a colorimetric, radioactive, or fluorescent
readout. As a result, one can assay the ability of small molecule compounds to alter
the binding of Norrin to LRP5/LRP6/Fz4 using this system or whether the test reagent
is a Norrin mimetic or Norrin agonist. For example, compounds, when added with
Norrin (or epitope-tagged Norrin) to each well of the plate, can be scored for their
ability to modulate the interaction between Norrin and LRP5/LRP6/Fz4 based on the
signal intensity of bound Norrin present in the well after a suitable incubation time

and washing. The assay can be calibrated by doing competition experiments with
unlabeled Nonrin or with a second type of epitope-tagged Norrin. Any molecule that
is able to modulate (e.g., enhance) the Norrin-LRP5/LRP6/Fz4 interaction may be a
suitable therapeutic candidate, more preferably an osteogenic therapeutic candidate or
a candidate capable of modulating a lipid (e.g., ApoE, LDL, HDL, VLDL,
triglyceride, cholesterol). Such molecules include small chemical compounds,
peptides, and immunoglobulins; (antibodies, antibody fragments) all can be examined
using such an assay system.
2.3 Norrin-LRP5/6/Fz4 Homogenous Assay
Another method to investigate modulation of protein-protein interactions is via
Fluorescence Resonance Energy Transfer (FRET). FRET is a quantum mechanical
process, where a fluorescent molecule, the donor, transfers energy to an acceptor
chromophore molecule which is in close proximity. Similarly an Amplified
Luminescent Proximity Homogenous Assay (ALPHA screen) also can be used to
evaluate Norrin-LRP5/6 or Fz4 interaction domains and function ofNorrin mimetics
in the Fz4/LRP5 complex. Such systems have been successfully used in the literature
to characterize the intermolecular interactions between LRP5 and Axin (see, e.g.,
Maio et al., Molec. Cell Biol. 7: 801-9). There are many different fluorescent tags
available for such studies and there are several ways to fluorescently tag the proteins
of interest. For example, CFP (i.e., cyan fluorescent protein) and YFP (Le., yellow
fluorescent protein) can be used as donor and acceptor, respectively. Fusion proteins,
with a donor and an acceptor, can be engineered, expressed, and purified or
conjugated to specific donor and acceptor beads.
For instance, in FRET type assays, purified Norrin proteins, or biologically
active polypeptides thereof, or agents being screened as Norrin mimetics can be fused
to CFP (or another fluorescent protein), and purified LRP5/6/Fz4 protein or
biologically active polypeptides thereof (e.g., LBD, or beta propeller containing
domain), fused to YFP can be generated and purified using standard approaches. If
Norrin-CFP and LRP5/Fz4-YFP are in close proximity, the transfer of energy from
CFP to YFP will result in a reduction of CFP emission and an increase in YFP
emission. Energy is supplied with an excitation wavelength of 450 nm, and the
energy transfer is recorded at emission wavelengths of 480 nm and 570 nm. The ratio
of YFP emission to CFP emission provides a gauge for changes in the interaction
between Norrin (or Norrin mimetic) and LRP5/Fz4. This system is amenable for
screening small molecule compounds that may alter the Norrin-LRP5/Fz4 protein-
protein interaction and activity in the Wnt cascade. Compounds that enhance or
disrupt the interaction would be identified by an increase or decrease respectively in
the ratio of YFP emission to CFP emission. Such compounds that modulate the
LRP5/Fz4 interaction in the same fashion as Norrin would then be considered

candidate Norrin mimetic molecules. Agents would also be screened for those which
enhance Norrin-like activity. These assay systems can be further modified with
different fluorescent proteins to include Kremen, Dkk, and/or Wnt in various
combinations. Further characterization of the compounds can be done using the TCF-
luciferase or Xenopus embryo assays to elucidate the effects of the compounds on
functional Norrin signaling.
2.4 Yeast Hybrid Assays
The two-hybrid, three-hybrid or other yeast hybrid system is extremely useful
for studying protein:protein interactions. See, e.g., Chien et al., 1991 Proc. Nat'}
Acad. Sci. USA 88: 9578-82; Fields et al., 1994 Trends Genetics 10: 286-92; Harper et
al., 1993 Cell 75: 805-16; Vojtek et al., 1993 Cell 74: 205-14; Luban et al., 1993 Cell
73: 1067-78; Li et al., 1993 FASEB J. 7: 957-63; Zang et al., 1993 Nature 364: 308-
13; Golemis etal., 1992 Mol Cell. Biol. 12: 3006-14; Sato et al., 1994 Proc. Nat'I.
Acad. Sci. USA 91: 9238-42; Coghlan et al., 1995 Science 267: 108-111; Kalpana et
al., 1994 Science 266:2002-6; Helps et al., 1994 FEBSLett. 340: 93-8; Yeung et al.,
1994 Genes & Devel. 8:2087-9; Durfee et al, 1993 Genes & Devel. 7: 555-569;
Paetkau etal., 1994 Genes & Devel. 8: 2035-45; Spaargaren etal., 1994 Proc. Nat'I.
AcadSci. USA 91: 12609-13; Ye etal., 1994 Proc. Nat V Acad. Sci. USA 91: 12629-
33; and U.S. Patent Nos. 5,989,808; 6,251,602; and 6,284,519.
Variations of the system are available for screening yeast phagemid (see, e.g.,
Harper, CELLULAR INTERACTIONS AND DEVELOPMENT: A PRACTICAL APPROACH, 153-
179 (1993);and Elledge et al., 1991 Proc.Nat'l Acad. Sci. USA 88: 1731-5), or
plasmid (Bartel, 1993 Cell 14: 920-4); Finley et al., 1994 Proc. Nat'l Acad. Sci. USA
91: 12980-4) cDNA libraries to clone interacting proteins, as well as for studying
known protein pairs.
The success of the two-hybrid system relies upon the fact that the DNA
binding and polymerase activation domains of many transcription factors, such as
GAL4, can be separated and then rejoined to restore functionality (Morin et al., 1993
Nuc. Acids Res. 21: 2157-63). While these examples describe two-hybrid screens in
the yeast system, it Is understood that a two-hybrid screen may be conducted in other
systems such as mammalian cell lines. The invention is therefore not limited to the
use of a yeast two-hybrid system, but encompasses such alternative systems.
Yeast strains with integrated copies of various reporter gene cassettes, such as
for example GAL→LacZ, GAL→HIS3 or GAL→URA3 (Bartel, IN CELLULAR
INTERACTIONS AND DEVELOPMENT: A PRACTICAL APPROACH, 153-179 (1993); Harper
et al., 1993 Cell 75: 805-16; Fields et al., 1994 Trends Genetics 10: 286-92) are co-
transformed with two plasmids, each expressing a different fusion protein. One
plasmid encodes a fusion between protein "X" and the DNA binding domain of, for
example, the GAL4 yeast transcription activator (Brent et al., 1985 Cell 43: 729-36;

Ma et al., 1987 Cell 48: 847-53; Keegan et al., 1986 Science 231: 699-704), while the
other plasmid encodes a fusion between protein "Y" and the RNA polymerase
activation domain of GAL4 (Keegan et al., 1986). The plasmids are transformed into
a strain of the yeast that contains a reporter gene, such as lacZ, whose regulatory
region contains GAL4 binding sites. If proteins X and Y interact, they reconstitute a
functional GAL4 transcription activator protein by bringing the two GAL4
components into sufficient proximity to activate transcription. It is well understood
that the role of bait and prey proteins may be alternatively switched and thus the
embodiments of this invention contemplate and encompass both alternative
arrangements.
Either hybrid protein alone must be unable to activate transcription of the
reporter gene. The DNA-binding domain hybrid must be unable to activate
transcription, because it does not provide an activation function; and the activation
domain hybrid must be unable to activate transcription, because it cannot localize to
the GAL4 binding sites. Interaction of the two test proteins reconstitutes the function
of GAL4 and results in expression of the reporter gene. The reporter gene cassettes
consist of minimal promoters that contain the GAL4 DNA recognition site (Johnson et
al., 1984 Mol Cell. Biol 4: 1440-8; Lorch et al., 1984 J. Mol. Biol. 186: 821-824)
cloned 5' to their TATA box. Transcription activation is scored by measuring either
the expression of β-galactosidase (or other reporter) or the growth of the transformants
on minimal medium lacking the specific nutrient that permits auxotrophic selection
for the transcription product, e.g., URA3 (uracil selection) or HIS3 (histidine
selection). See, e.g., Bartel, 1993; Durfee et al., 1993 Genes & Devel. 7: 555-569;
Fields et al., 1994 Trends Genet. 10: 286-292; and U.S. Patent No. 5,283,173.
Generally, these methods include two proteins to be tested for interaction
which are expressed as hybrids in the nucleus of a yeast cell. One of the proteins is
fused to the DNA-binding domain (DBD) of a transcription factor, and the other is
fused to a transcription activation domain (AD). If the proteins interact, they
reconstitute a functional transcription factor that activates one or more reporter genes
that contain binding sites for the DBD. Exemplary two-hybrid assays are Norrin,
Norrin/Frizzled4, or Frizzled4/LRP5 fusions.
3. In vivo Methods of Assaying Agents
In addition to the in vitro methods identified herein, the methods and materials
can further include use of animals to study the effect of test agents screened and
identified by in vitro analysis. For example, transgenic animals wherein one (or more)
of Norrin, Kremen (Kremen 1 and/or 2), Dkk (Dkkl, Dkk2, Dkk3, and/or Dkk4),
LRP5, LRP6, HBM, Wnt (Wntl to Wntl9), and Frizzled4 genes are introduced as
cDNAs. Examples of LRP5 and HBM transgenic animals can be found in

International PCT Application No. PCT/US02/14876 and U.S. Application No.
10/680,287. The subject matter of these applications is incorporated herein by
reference in their entirety for all purposes.
Thus, in one aspect, after the steps of screening the test agent against any of
the transfected cell lines discussed above and/or after agents have been tested to see if
they bind to any of Dkk, Norrin, Frizzled4, LRP5, LRP6, HBM, Wnt, and/or Kremen,
these test agents can also be assessed in vivo. Adding the step of testing reagents in
vivo adds a validation step to the tests obtained by any of the means discussed in
Section 2 supra. Reagents can be administered to the animals via any means of
administration suitable for the compound, e.g., oral, intravenous, intramuscular,
intraperitoneal, cutaneous, and the like. Administration may depend on the
formulation of the test compound. For example, small inhibitory RNAs (siRNAs) and
immunoglobulins may get administered intravenously rather than orally. Small
chemical compounds may be administered orally or intravenously. Amounts of the
test compound would be administered based on. a weight basis for the animal.
Animals could also be utilized to test bioavailability and degradation products
of the compounds.
Animals would be administered the test compounds over a period of days,
weeks, or months. Administration can be daily, weekly, bimonthly, monthly and the
like. Animals can have the agent administered alone or in conjunction with exercise,
which causes strain on the bones of the animal. Discussion of how strain can be
placed on the animals' bones is described in International PCT Application No.
PCT/US2004/17951. The contents of this application are incorporated herein by
reference in its entirety for all purposes.
For example, the pDXA can be measured in wild-type and transgenic animals
that are administered various dosages of agents. For example, wild-type and
transgenic mice are anesthetized, weighed and whole-body X-ray scans of the skeleton
generated using the LUNAR small animal PIXImus device. Scans can be performed
when the mice are weaned (i.e., at 3 weeks of age) and repeated at 2 week intervals.
Wild-type animals can be scanned at 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23,27, and 29
weeks. Scanning of transgenic animals can be performed for periods up to 17 weeks.
Scans can be analyzed for BMD (bone mineral density), BMC (bone mineral content),
TTM (total tissue mass), and percent (%) fat for various body regions.
Additionally or alternatively, faxitron radiographs of the above animals can be
obtained. For example, following pDXA scanning of anesthetized animals, an
additional X-ray can be taken using a Faxitron device allowing measurement of bone
size.
Additionally or alternatively, calcein labeling can be performed on the above
animals. For example, animals can be dosed with calcein at 15 mg/kg animal weight

on two consecutive occasions. The first dose can be given 9 days before the animal is
euthanized, and the second dose given two days prior to animal euthanasia.
Measurement of bone formation can then be determined.
Certain types of ex vivo analysis of the above animals can also optionally be
performed. For example, RNA isolation can be done from tissue, pQCT and microCT
(μCT), histology, bending strength analysis, compressive strength analysis of vertebra,
and serum analysis can be performed. For example, RNA can be isolated from tibia
and other tissues using TRIzol7 to determine mRNA expression. pQCT analysis of
any of the above animals can be performed by obtaining a femur and cleaning it of
soft tissue. The femur can then be stored in 70% ethanol for determination of total
and trabecular density of the distal metaphysis and cortical density of the mid-shaft
then determined.
Analysis of the animal's femur can also be used to determine trabecular
indices of the distal metaphysis.
Optionally, histological analysis can be performed on any of the above
animals. For example, the femur of a mouse can be used to determine bone area and
static and dynamic parameters of the distal metaphysis. Alternatively or additionally,
immunohistochemistry can be performed (e.g., in situ hybridization of osteogenic
markers and TUNEL staining of cells undergoing apoptosis).
Any of the above animals can also have their bones examined for bending
strength or compressive strength of vertebra. For bending strength, the animal's
femur (or other suitable bone) can be cleaned of soft tissue and stored at about -20°C
prior to analysis of 3-point bending strength of the mid-shaft. Compressive strength
can be measured by removing the spine of, for example, a mouse from T10 to L6 or
L7. Soft tissue is left on the spine which is then frozen at about -20°C until analysis.
Compressive strength is frequently measured at the L5 vertebra.
For purposes of lipid analysis, serum from animals can be assessed. For
example, animals can be euthanized and serum prepared from the blood to measure
total cholesterol, triglycerides, osteocalcin, and other biochemical surrogate markers.
Gene transcript and expression modulation can also be assessed on animals.



Any one or more of the above referenced genes, in any combination, can be assessed
for changes in expression due to administration of a test agent, control agent, stress on
the bone or bone cell, and 50 forth. The gene profile produced can then be assessed in
conjunction with the affect the agent has on the Norrin-Frizzled4 or Frizzled4-LRP5
complex and its activity in the Wnt pathway. For example, the gene/protein profile
obtained by the agent that modulated the Norrin-Frizzled4-LRP5 complex or a Norrin
mimetic produces a profile like that observed either with the administration of a Dkk
antagonist or Kremen antagonist or by stress on the bone or bone cell.
Bone load and comparing bone load to modulation by an agent can also be
performed in vitro. For example, gravitational load can be used to induce stress on
any of the bone cells discussed herein, and the profile of gene expression of any one or
more of the genes or any combination of the following genes can be assessed as part
of a bone stress profile. The bone stress profile can be assessed to see whether, for
example, a Norrin mimetic induces an enhance bone stress profile or decreases the
inhibition of Dkk and/or Kremen on the genes of a bone stress profile.

Determination of whether the administered test agents can induce a bone
modulating effect can be assessed by X-ray for change of bone density or by animal
sacrifice and examination of cortical bone as described in U.S. Application No.
10/680,287 and International PCT Application No. PCT/US2004/17951, or any of the
methods described herein or known in the art. The contents of these applications are
incorporated herein by reference in their entirety for all purposes.
4. Methods of Studying Bone Loading In vitro
One aspect of the invention is the study of the effect of bone load in vitro and
means by which the benefits of bone load (i.e., increased bone mineralization) can be
enhanced. Studying bone load enhancement can be done both in vivo (as discussed
above) and in vitro. Bone load enhancement can be first performed in vitro followed
then with in vivo experiments, such as those discussed above.
Consequently, one aspect of the invention involves placing cells under
conditions, which simulate load stimuli. There are several methods available for
placing strain on cell cultures to mimic the bone load response observed in vivo.
These methods include but are not limited to fluid shear, hydrostatic compression,
uniaxial stretch, biaxial stretch, gravitational loading and load induced using a
Flexercell®, or equivalent system.
4.1 Bone Load Stimuli
Preferred genes which are modulated by a bone load stimuli, such as those
provided by any of the above methods, include but are not limited to SFRPl,
connexin, W1SP2 43, CCNDl, WntlOb, Jun, Fos, PTGS2 (COX-2), and eNOS.

Additional genes that can be monitored for increases in their activity (e.g., increased
mRNA transcripts and protein) as reflected in many of the Tables herein. At least six
genes that have been shown to be consistently up-regulated in response to bone load
(i.e., Jun, Fos, eNOS, SFRP1, COX-2 and Connexin 43) are also enhanced by the
addition of an agent which activates the Wnt pathway. Other genes, such as Wnt2, are
not enhanced by the addition of reagents that activated the Wnt pathway (e.g., GSK-3
inhibitors and Wnt 3A and its agonists, mimetics, and variants) and only respond to
bone load. Thus, one aspect would include using such in vitro systems to study
enhancement of the stress profile genes in response to, for examples, a Norrin
mimetic, a Norrin agonist, a Frizzled4 mimetic, or a Frizzled4 mimetic.
4.1.1 Fluid Shear Stimulus
One method of inducing bone load is by fluid shear. Fluid shear can utilize a
cone plate viscometer that generates continuous laminar shear by a stirring
mechanism. Alternatively, a flow loop apparatus can produce such shear in a parallel
flow culture chamber. The latter method and apparatus is exemplified by the
Streamer system produced by Flexcell International Corporation. The flow loop
apparatus also is known to produce a reproducible and consistent stimulus. The only
drawbacks are that the end points are typically short-lived and whether these changes
impact the function of differentiated osteoblasts (Basso et al., 2002 Bone 30(2): 347-
51).
4.1.2 Hydrostatic Compression Stimulus
A second method of inducing bone load is use of hydrostatic compression.
Hydrostatic compression can utilize compressed air to generate a continuous or
intermittent force that is believed to localize the force specifically to regions where the
cells interact with the extracellular matrix protein/adhesion proteins.
4.1.3 Uniaxial Stretch Stimulus
A third means of inducing bone load in vitro is use of a uniaxial stretch
stimulus. The uniaxial stretch method utilizes stretch force in one direction. The
method involves growing cells in a tissue culture on a treated strip of polystyrene film
or other film, which is fixed to a flexible layer of silicone. The layer of silicone is
further attached to two metal bars. The metal bars can be manipulated relative to each
other using an electromagnet or some other moving means. This method does not
create any fluid shear. The lack of fluid shear makes this method less preferred,
because interstitial fluid flow may play a larger role in bone remodeling than
mechanical stretch. Accordingly, this method may not fully mimic what occurs in
vivo despite the reproducible and consistent stimulus produced (Basso et al., 2002
Bone 30(2): 347-51).

4.1.4 Biaxial Stretch Stimulus
Biaxial stretch is essentially the Flexercell® system discussed herein. This
method uses a collagen coated silastic membrane upon which the cells are grown.
The plates are then placed in a special tray, which is attached to a vacuum pump. The
vacuum pump stretches and relaxes the membrane, by stretching or otherwise
distorting the cell membrane. Additionally, any media or fluid movement will further
add fluid shear.
4.1.5 Gravitational Load Stimulus
Gravitational loading is another method by which bone load can be induced in
vitro. Essentially, force is placed on the cells causing the cells to flatten. For
additional details, see for example, Hatton et al., 2003 J. Bone & Min. Res. 18(1): 58-
66; and Fitzgerald et al., 1996 Exp. Cell. Res. 228: 168-71. Specifically, the cells are
grown on plates or cover slips and then are exposed to increasing G forces.
4.1.6 Flexercell® Stimulus
One preferred method for assessing reagent-based enhancement of the Wnt
pathway and bone mineralization is using the Flexercell® system, a biaxial stretch
stimulus. Briefly, bone cells (e.g., MC3T3 cells) are exposed to about 3,400 με.
Loads of about 50 με to about 5,000 με (and any value in between) can be used as well
for mechanical load stimuli. Any stimulus in this range mimics physiological bone
load stimuli. Stimuli above 5,000 με result in pathophysiological loads, and therefore
are not preferred. The cells also can be exposed to a Wnt pathway modulator (e.g., a
GSK inhibitor) prior to exposure to biaxial stretch.
The genes up-regulated by the administration of the load alone or with a GSK-
3 inhibitor include, but are not limited to COX-2, eNOS, connexin 43, Fos, Jun,
WISP2, Wnt10b, Cyclin D1, and SFRPl. The expression profile obtained in vitro
from the Flexercell® studies mimics the in vivo loading gene expression profile {i.e.,
RNA analysis performed on cells from HBM TG mice tibia wherein the mice were
subjected to bone load using a four-point system). Thus, this mechanical load assay,
or the use of other mechanical load means with the variety of cell lines disclosed
herein, can be used to identify small molecules, peptides, immunoglobulins, and the
like that modulate, and preferably activate, the canonical Wnt pathway and which
mimic the HBM phenotype. A Norrin mimetic can produce the same response as
Norrin or an enhanced response, like the enhanced response of the HBM variant of
LRP5 of increased bone mass. Thus, using this system would be helpful for screening
reagents that enhance the up-regulated genes of the stress profile in an HBM-Iike
manner as well as acting in a manner equivalent to wild-type Norrin.
The in vitro methods of inducing mechanical stress stimuli on cells can also be
used to study cell proliferation and apoptosis, which is relevant to bone remodeling
and the need for osteoblast and osteoclast proliferation and osteoclast resorption. For

example, HBM and unaffected osteoblastic cells can be seeded into bioflex 6-well
plates and cultured for 2-3 days in growth media containing 10% FBS until the cells
are about 60% confluent. Twenty-four hours prior to mechanical loading, the media is
replaced with 1 mL of basal media containing about 2 to about 4% FBS. The cells are
then subjected to about 50 to about 5,000 με of load for about 1 to about 5 hours. The
cells can be further studied for reagents that are Norrin mimetics or which are agonists
of the LRP5/Norrin/Frizzled4 complex (e.g., Norrin agonists, Frizzled4 agonists, or
LRP5 agonists) as well as antagonists thereof.
Following load, the cells are cultured for an additional period of time.
Subsequently, cell number and proliferation can be assessed using a number of
commercial assays or assays known in the art, including but not limited to [3H]-
thymidine incorporation, 5-bromo-2'-deoxyuridine (BrdU) incorporation, 3-(4,5
dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tra2olium
salts (MTS) assay, TUNEL assay (i.e., terminal deoxynucleotidyltransferase dUTP
nick end labeling) or Annexin V assay.
The following genes can be analyzed with regard to the profile. In another
embodiment, Wnt antagonists can be screened or used to treat individuals wherein
bone demineralization (e.g., osteopetrosis) is needed. Wnt antagonists include but are
not limited to Dkk1 antagonists, and Kremen antagonists. Norrin agonists, and Norrin
mimetics along with Frizzled4 agonists and mimetics, Wnt agonists and mimetics, and
LRP5 and LRP6 agonists and mimetics can also be assessed under this system.

The materials and methods relating to the protein and nucleic acid arrays for
bone load are discussed in greater detail in International PCT Application No.
PCT/US2004/17951, which is herein incorporated in its entirety for all purposes.
4.2 Functional Evaluation of Norrin in Xenopus
Xenopus embryos are an informative and well-established in vivo assay system
to evaluate the modulation of Wnt signaling see, e.g., McMahon et al., 1989 Cell 58:
1075-84; Smith and Harland 1991 Cell 67: 753-65, reviewed in Wodarz and Nusse,

1998 Annu. Rev. Cell. Dev. Biol. 14: 59-88).
Modification of the Wnt signaling pathway by impacting the Norrin-Frizzled4-
LRP5 complex can be visualized by examining the embryos for a dorsalization
phenotype (duplicated body axis) after RNA injection into the ventral blastomere at
the 4- or 8-cell stage. On the molecular level, phenotypes can be analyzed by looking
for expression of various marker genes in stage 10.5 day embryos. Such markers
would include general endoderm, mesoderm, and ectoderm markers as well as a
variety of tissue-specific transcripts.
Analysis of the embryos can be done using RT-PCR/TaqMan® and can be
done on whole embryo tissue or in a more restricted fashion (microdissection).
Because this system is very flexible and rapid, by injecting combinations of
transcripts, such as Norrin, LRP5/LRP6 and Fz4, the mechanism of Norrin signaling
pathway can be dissected. Previous studies have demonstrated that LRP6 alone or in
combination with LRP5 + Wnt5a were able to induce axis duplication (dorsalization)
in this system (Tamai et al., 2000 Nature 407: 530-35). Once the Norrin signaling is
established, it can be used to evaluate by Dkk and Kremen antagonists and Norrin
agonists and Norrin mimetics.
4.2.1 Constructs for Xenopus Expression (Vector pCS2+)
Norrin, LRP5/6, Fz4, Dkk (e.g., Dkk1), Wnt, and Kremen1/2 cDNAs can be
subcloned into a vector, such as pCS2+, in the sense orientation with respect to the
vector SP6 promoter. The pCS2+ vector contains an SV40 virus polyadenylation
signal and T3 promoter sequence (for generation of antisense mRNA) downstream of
the insert. Other vectors can also be utilized for expression of the proteins, in any
combination.
4.2.2 mRNA Synthesis and Microinjection Protocol
mRNA for microinjection into Xenopus embryos can be generated by in vitro
transcription using the cDNA constructs in the pCS2+ vector for example as described
above as template. RNA is synthesized using the Ambion mMessage mMachine high
yield capped RNA transcription kit (Ambion Cat. #1340) following the manufacturer's
specifications for the Sp6 polymerase reactions. RNA products can be brought up to a
final volume of 50 μL in sterile, glass-distilled water and purified over Quick Spin
Columns for Radiolabeled RNA Purification using a G50-Sephadex column (Roche,
Cat. #1274015) following the manufacturer's specifications. The resulting eluate was
finally extracted with phenol:chloroform:isoamyl alcohol and isopropanol precipitated
using standard protocols (Sambrook et al., 1989). Final RNA volumes are usually
approximately 50 μL. RNA concentration can be determined by absorbance values at
260 nm and 280 nm. RNA integrity can be visualized by ethidium bromide staining
of denaturing (formaldehyde) agarose gel electrophoresis (Sambrook et al., 1989).
Various amounts of RNA (about 2 pg to about 1 ng) are injected into the ventral

blastomere of the 4- or 8-cell Xenopus embryo. These protocols are described in
Moon et al., 1989 Technique-J. Meth. Cell & Mol. Biol. 1: 76-89, and Peng, 1991
Meth. Cell. Biol. 36: 657-62.
Molecules identified as modulating Norrin function or which act as Norrin
mimetics in any of the assays described herein can be further validated using animal
models or other in vitro screening assays.
4.3 Evaluation of Norrin for Osteogenic Effect in Mesenchymal Stem
Cells
Human mesenchymal stem cells (hMSCs) (Cambrex Bio Science,
Walkersville, MD) and mouse stem cells (e.g., C3H10T1/2, ATCC) can be induced to
differentiate into mineralized bone nodules (Jaiswal et al., 1997 J. Cell Biol. 64: 295-
312) or adipose tissues (Pettinger et al., 1999 Science 284: 143-147) in vitro by
osteogenic or adipogenic medium respectively. Wnt signal activation enhances
osteogenesis and inhibits adipogenesis in hMSCs. Norrin-Fz4-LRP5 mediated
signaling is expected to provide a similar type of differentiation patterns in hMSCs.
Thus, identifying Norrin mimetics and Norrin agonists using hMSCs (or other MSC
cell from another vertebrate) is a screening assay contemplated.
In addition to human mesenchymal stem cells, stem cells from other vertebrate
animals can also be used. Mesenchymal stem cells are progenitor cells to various
bone cells (e.g., osteoblasts) as well as to adipocytes {see, e.g., Bennett et al., 2005
Proc Nat'l Acad-Sci. USA 102(9): 3324-3329). Alternatively, more differentiated
cells can be substituted such as preosteoblasts, osteocytes, and mature osteoblasts. It
should be noted that instances wherein the cell line is indicated to be a human derived
cell line, analogous cells from another vertebrate animal can be substituted.
43.1 Evaluation of Osteogenic Activity by Norrin Over Expression
Briefly, Norrin can be added to hMSCs (passage 3-6) as a purified protein or
as part of conditioned medium, or expressed by infecting hMSCs using viral vectors.
Alternatively, Norrin, Norrin agonists, and Norrin mimetics can be added to the
hMSCs along with the osteogenic medium (growth medium supplemented with 10
nM dexamethasone, 50 μg/mL L-ascorbic acid and 5 mM beta-glycerophosphate).
After about 1 to about 3 weeks of incubation along with appropriate control medium
at about 37°C, and by weekly replenishment of fresh medium with or without Norrin,
the osteogenic activity can be measured by standard techniques. For example, the
osteogenic activity can be measured by staining the cells for alkaline phosphatase
(AlkPhos) protein expression, determining the enzymatic activity of AlkPhos,
induction of AlkPhos or osteocalcin mRNAs and detection of mineralization by
Alizerin Red or von-Kossa stains along with appropriate controls.

4.3.2 Evaluation of Modulation of Adipogenesis byNorrin Over
Expression
Norrin, Norrin agonists, or Norrin mimetics can be added to hMSCs by
expressing using viral or other types of vectors along with the adipogenic
differentiation medium (i.e., growth medium containing 10 nM dexamethasone, 50
μg/mL L-ascorbic acid phosphate, 500 μM isobutylmethylxanthine and 60 μM
indomethacin) for 1-3 weeks. The effect of Norrin, Norrin agonists, or Norrin
mimetics on adipocyte modulation can be determined by the alteration of the
expression of adipogenic marker genes (e.g., Adipsin) or by staining the cells with oil
red O reagent.
Changes in expression due to the administration of test agents can be
performed using DNA array technology. Such technology is already used to test for
obesity and diabetes. Therefore, one aspect would be to screen test agents using DNA
array technology developed for obesity. See, e.g., Nadler et al., 2000 Proc. Natl.
Acad. Sci. 97(21): 11371-11376, and the genes indicated for lipid metabolism,
secreted proteins, as well as the other genes with decreased or increased expression
associated with obesity. The cells used in conjunction with the DNA array technology
can be mesenchymal cells, adipocytes, or preadipocytes, or other cells discussed
herein.
In vivo changes in lipid levels and adipogenesis can be measured by a variety
of different tests. Biood and serum can be collected and analyzed for blood chemistry.
Thus, Norrin mimetics or Norrin agonists can be screened for the effects in vivo.
Likewise, Dkk inhibitors and/or Kremen inhibitors can be screened for their impact on
Norrin (in the presence or absence of a Norrin agonist) or Norrin mimetic activity with
the LRP5/LRP6/Frizzled4 complex.
4.3.3 Evaluation of Osteogenesis and Adipogenesis Modulation by
Norrin Gene Knock Down
By using an osteogenic or an adipogenic medium, hMSCs will differentiate
into osteogenic or adipogenic lineages. Small hairpin RNAs of Norrin, Frizzled4 or
LRP5 can be used as a control to demonstrate the enhancing effects of Norrin and
Frizzled4 on the pathway and to show the impact inhibition of these proteins has on
adipogenesis and osteogenesis. For example, in presence of osteogenic medium and
the infection of viral vector containing Norrin shRNA, Norrin gene transcription can
be blocked and the differentiation of hMSCs into osteoblasts or the production of
mineralized bone nodules that can be detected by various methods indicated above.
Gene knockdown in tissue culture and in vivo can be attained by sequence-
specific DNA or RNA analogs that can block the activity of selected single-stranded
genetic sequences. Examples of such approaches include antisense oligonucleotide
technology and the introduction of a homologous double-stranded RNA (dsRNA) or

short interfering RNA (siRNA), which is also called post-transcriptional gene
silencing (PTGS) or RNA interference (RNAi). This can be achieved by introducing
into the cell siRNAs specific to the given target gene mRNAs via shRNAs (short
hairpin RNAs) using various viral gene-delivery vectors or by transfecting plasmid
vectors. Methods of performing RNA interference and gene silencing are known as
discussed for example in Meister and Tuschl, 2004 "Mechanisms of gene silencing by
double-stranded RNA," Nature 431: 343-349; Dorsett and Tuschl, 2005 "siRNAs:
applications in functional genomics and potential as therapeutics," Nat. Rev. RNA
Interference Collection 40-51 and the references cited therein. Once the siRNA is
introduced, the extent of target gene knockdown is measured by standard techniques
including qRT-PCR, Northern Blots for RNAs or by Western blots for protein
expression.
5. Kits for Testing Agents Which Modulate Norrin
Another aspect contemplates kits for testing agents which modulate Norrin
activity, and preferably for agents, which through modulation of Norrin activity,
modulate the Wnt pathway. These kits can be used to screen for Norrin mimetics and
Norrin agonists, as well as other mimetics and agonists of the LRP5/Norrin/Frizzled4
complex.
Contemplated kits would include cells and nucleic acids encoding at least
Norrin and Frizzled4. Preferably, there would be nucleic acids encoding LRP5, Dkk
(any of the Dkks), HBM, and/or Kremen (Kremen 1 and 2), as well as Norrin and
Frizzled4 and/or any combination thereof. LRP6 and Wnts can also be included.
Nucleic acids encoding the above polypeptides would be operably linked to a vector.
Vector only would also be preferably included for control purposes. The kits could be
styled for either transient transfection use or for stably transfecting cells.
Alternatively the kits can contain purified proteins, of any of the above
proteins for use in vitro assay systems, such as those described here. They can include
substrates such as nitrocellulose, ELSA plates, or other suitable substrates.
The kits would preferably include an assay appropriate reporter system,
whether alkaline phosphatase, on or more fluorescent proteins and the like.
In one aspect, the kit could come with frozen cell lines for use in screening. In
another aspect, the kit could come with instructions listing appropriate, previously
tested cells that are suitable for the in vitro assays described herein.
In another aspect, the kit could come with the various reporters, enzymes, and
reagents necessary for detecting the reporter used to detect the modulation. For
example, if using the TCF-luci and TK-renilla assay system, the kit could come with
TCF-luci and TK-renilla luciferases and detection reagents. Kits could also come
with transgenic animals, wherein a cDNA for Dkk, Norrin, LRP5, LRP6, HBM,

Kremen, Wnt, and/or Frizzled4 has been introduced into an animal(s). Kits can
include a series of such animals for elucidation of activity for a particular test reagent.
6. Cell Lines
Another aspect the preparation of cell lines which do not express Norrin and/or
Frizzled4. These cells lines can then have transiently or stably expressed non-native
forms of LRP5, LRP6, Frizzled, Norrin, Dkk, Kremen, Wnt, and Norrin in any of the
combinations discussed herein. Therefore, the cell lines can be used to screen
reagents that are Norrin mimetics or Norrin agonists. For example, a cell line which
has non-native (non-endogenous) forms of LRP5 and Frizzled4 expressed and lacks
Norrin, there would be no means by which to activate the Wnt pathway through the
LRP5-Frizzled4-Norrin mechanism. However, with the introduction of a Norrin
mimetic, the pathway would be activated. Such cell lines would be useful controls for
identifying "Norrin mimetics. The cells could then have additional non-endogenous
transcripts of Dkk and/or Kremen introduced with screening examined for test agents
that modulate Dkk and/or Kremen interaction with the Frizzled4-LRP5/6-Norrin
complex. Using this process, Dkk antagonists and/or Kremen antagonists can be
identified. Introduction of a non-endogenous Norrin in one of these Norrin-free lines
can be used to assess Norrin agonist on the Norrin-LRP5-Frizzled4 interaction.
Stable and transient expression of the nucleic acids encoding any of the
proteins or biologically active polypeptide fragments thereof can be accomplished by
means known in the art. See, e.g., R. IAN FRESHNEY, CULTURE OF ANIMAL CELLS: A
MANUAL OF BASIC TECHNIQUE (2000) and JOSEPH SAMBROOK AND DAVID W.
RUSSELL, MOLECULAR CLONING: A LABORATORY MANUAL (3ri ed. 2001).
Cells which do not have endogenous Norrin include, but are not limited to,
kidney cells. Therefore, kidney cells provide a useful tool for screening Norrin
mimetics. Alternatively, cell lines can be prepared that are knock downs where one or
more of the genes encoding LRP5, LRP6, Norrin, a Wnt, a Dkk, a Kremen, and/or
Frizzled4 are knocked out such that an endogenous polypeptide can no longer be
synthesized. This procedure can be carried out on any cell, such as but not limited to
adipocytes, preadipocytes, mesenchymal cells, various bone cells, and kidney cells.
It will be apparent to those skilled in the art that various modifications and
variations can be made in the materials and methods described herein without
departing from the spirit or scope of the invention. Thus, it is intended that the
present invention cover the modifications and variations of this invention provided
they come within the scope of the appended claims and their equivalents.

EXAMPLES
EXAMPLE 1: Norrin/Wnt-TCF Signal Assay
Norrin clone isolation. cDNA was cloned by standard PCR method from the
IMAGE clones. Specifically, the Norrin open reading frame (ORF) sequence from the
NCBI (NM_000266) was used to search for available IMAGE clones. Clone
#5179578 was identified as a predicted full length cDNA. The IMAGE clone was
purchased from Open Biosystems (Huntsville, AL). The ORF was amplified by
standard PCR techniques using the following primers:
5'-CATATGAATTCACCATGAGAAAACATGTACTAGCTGCATC-3' (SEQ ID
NO:1) (which brings an EcoR1 site for cloning as well as a consensus Kozak
immediately 5' of the initiating ATG) and
5'-GATATGCGGCCGCTCTAGATCAGGAATTGCATTCCTCGCAGTG-3'SEQ
ID NO:2) (which brings both an XbdI and NoiI site following the stop codon). The
resulting PCR. product was digested with EcoRI and NoiI and cloned into the EcoRI
and NolI sites of PCDNA3.1 (Invilrogen). Positive isolates were identified by
restriction digest and confirmed byDNA sequence analysis to match the published
sequence (Accession No. NM_000266).
Kremen clone isolation. The PCR amplified full length fragment was
subcloned into pcDNA3.1 vector at EcoRVBamHI restriction enzyme sites. The
isolated sequence was then verified to match the published human Kremen2 sequence
(Accession Nos. NM_172229/ AB086405.1, and NP_757384.1). cDNA was isolated
from human osteoblast-like U2OS cell line total RNA. Total RNA from about
2.5x106 U2OS cells was purified using the RNeasy kit (Qiagen, Valencia, CA)
following the protocol of the manufacturer. Kremen2 cDNA isolate was amplified
following standard PCR methods. The PCR primers used were:
5' primer: 5'-GGACGAATTCACCATGGGGACACAAGCCCTGCAG-5' (SEQ ID
NO:3) 3' primer: 5'-CTCGCTCATCTCCGCTCTCTGAGGATCCCAGG-3' (SEQ
ID NO:4). PCR amplified full length fragment was subcloned into pcDNA3.1 vector
at EcoRVBamHl restriction enzyme sites and the entire sequence was verified to
match the published human Kremen2 sequence (Accession Nos. NM_172229/
AB086405.1, NP_757384).
Dkk clone isolation. A human cDNA with GenBank accession number
AF127563 was available in the public database. Using this sequence, PCR primers
were designed to amplify the open reading frame with a consensus Kozak sequence
immediately upstream of the initiating ATG. Oligos 117162 (5'-
CAATAGTCGACGAATTCACCATGGCTCTGGGCGCAGCGG-31; SEQ ID NO:5)
and 117163(5'-
GTATTGCGGCCGCTCTAGATTAGTGTCTCTGACAAGTGTGAA-3';SEQID
NO:6) were used to screen a human uterus cDNA library by PCR. The resulting PCR

product was purified, subcloned into pCRII-TOPO (Invitrogen Corp.), sequence
verified, and digested with EcoRVXhoI. This insert was subcloned into the pCS2+
vector at the EcoRl-XhoI sites.
A full length cDNA encoding human Dkk2 was isolated to investigate the
specificity of the Zmax/LRP5/HBM interaction with the Dkk family of molecules.
DkkI was identified in yeast as a potential binding partner of Zmax/LRP5/HBM.
DkkI has also been shown in the literature to be an antagonist of the Wnt signaling
pathway, while Dkk2 is not (Krupnik et al, 1999). The Dkk2 full length cDNA
serves as a tool to discriminate the specificity and biological significance of
Zmax/LRP5/HBM interactions with the Dkk family (e.g., Dkk1, Dlck2, Dkk3, Dkk4,
Soggy, their homologs and variant, etc.). A human cDNA sequence for Dkk2
(GenBank Accession No. NM_014421) was available in the public database. Using
this sequence, PCR primers were designed to amplify the open reading frame with a
consensus Kozak sequence immediately upstream of the initiating ATG. Oligos
51409 (5'- CTAACGGATCCACCATGGCCGCGTTGATGCGG-3'; SEQ 1DNO.7)
and (5'-GATTCGAATTCTCAAATTTTCTGACACACATGG-3'; SEQ 1D NO:8)
were used to screen human embryo and brain cDNA libraries by PCR. The resulting
PCR product was purified, subcloned into pCRJI-TOPO, sequence verified, and
digested with BamHVEcoRI. This insert was subcloned into the pCS2+vector at the
BamHl-EcoRI sites. For additional discussion regarding Dkkl and Dkk2 clones, see
International PCT application PCT/US02/15982. Similar constructions for Dkk3 and
Dkk4 can be prepared using the sequences as referred to herein.
LRP6 clones. Full length LRP6 was isolated from the pED6dpc4 vector by
Xhol-Xbal digestion. The full length cDNA was reassembled into the Xhol-Xbal sites
of pCS2+. Insert orientation was confirmed by DNA sequencing.
LRP5 (Zmaxl) andHBM. Insert cDNA was isolated from the full length
cDNA retrovirus constructs (with optimized Kozak sequences) by BglU-EcoRl
digestion and subcloned into the BaniHl-EcoRI sites of the pCS2+ vector. For more
details on the LRP5 and HBM constructs, see U.S. Patent No. 6,770,461 and
International PCT Application PCT/USO1/16946 entitled "Regulating lipid levels via
the Zmaxl or HBM gene."
Wnt clones. The Wnt genes utilized in the experiments shown herein were
obtained as follows. Ten different full length Wnt cDNAs were purchased from
Upstate Biotechnology (Lake Placid, NY) in the vector pUSEamp(+). The genes are:
Wntl (Cat. No. 21-121), Wnt2 (Cat. No. 21-122), Wnt3 (Cat. No. 21-123), Wnt3a
(Cat. No. 21-124), Wnt4 (Cat. No. 21-125), Wnt5a (Cat. No. 21-133), Wnt5b (Cat.
No. 12-126), Wnt6 (Cat. No. 21-127), Wnt7a (Cat. No. 21-128) and, Wnt7b (Cat. No.
21-129). Inserts were released by Xbal digestion and subcloned into the Xba\ site of
the pCS105 vector for Xenopus expression. Orientation was confirmed by sequence

analysis.
Full length Wntl 1 cDNA was isolated from 1x106 human osteoblast cells
(HOBs, passage #13) by RT-PCR using GCRich Kit (Clontech, Mountain View, CA)
and the following specific primers: (Forward): 5'-
GGGAATTCGCGACGATGAGGGCGCGGCCGCA-3' (SEQ IDNO:9) (includes
EcoRI site) and (Reverse): 5'-
GGGCGGCCGCAGGGCCTCACTTGCAGACATAGC-3' (SEQ ID NO:10)
(includes Noil site). The RT-PCR generated DNA fragment was digested with EcoRI
and Noil and inserted into the EcoRl and Noil created site of the pcDNA3.l vector.
The full length sequence was verified to match the published Wntl 1 sequence
(Accession No. Y12692).
Full length cDNAs for other Wnt genes can be obtained by standard PCR
techniques from various human cDNA library sources using public sequence to design
primers to amplify the open reading frame of the gene and facilitate subcloning into
the pCS 105 vector or pcDNA3.1 type mammalian vector or other suitable mammalian
vector. Suitable primers for other Wnt genes are presented in the Table 1 below. "F"
stands for "forward" primer and "R" for "reverse" primer.



Reporter Assay. The TCF assay involves a 16x-TCF reporter (containing 16
copies of Wnt-beta-catenin signal responsive TCF element with basal TK-promoter)
and Luciferase gene. The construct contains 16 copies of the TCF binding sites
placed upstream of a minimal TK (thymidine kinase) promoter and the luciferase gene
in pGL3 vector (Promega, Madison, WI). The sequence of the four TCF binding sites
(see paired sequences below) was generated by oligonucleotide synthesis approach
and contains the following sequence with Nhel and Xhol restriction enzyme sites at
the 5' and 3' en'ds respectively. The underlined doinains indicate the TCF binding
sites. Both strands are provided. When the two strands anneal, Nhel (5') and Xhol
(3') compatible restriction sites are introduced for further cloning and they contain the
TCF binding sites (SEQ ID NOS: 27 and 28 respectively):
5' -CTAGCGAGAACAAAGGAGATTCAAAGGflGATCAAAGGAGATCftAAGGACTAGTTC-3'
3' -TCGAGAACTAGTCCTTTGATCTCCTTTGATCTCCTTTGAATCTCCTTTGTTCTCG-5'
TK-renilla (Promega Corp., WI) as internal assay normalization control; and
pcDNA vector-based constructs of Norrin, Wntl, Wnt3a, Dkkl, and Kremen2 cDNAs
as discussed above, and a Frizzled4 construct (Origene Tech. Inc.; Rockville, MD) are
also part of the assay. Other vectors capable of expressing different forms of
vertebrate Norrin, Wntl, Wnt3a, Dkkl, and Kremen2 can be substituted.
The clones individually containing these genes are co-transfected into Human
Embryonic Kidney (HEK)-293A cells (ATCC, Manassas, VA) or a human
osteosarcoma derived bone/osteoblast-like cell line, U2OS (ATCC). The cells were
cultured in Dulbecco's Minimum Essential Media (DMEM) (Invitrogen) or RPMI
media (Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (FBS),
1% glutamax (Invitrogen), and 1% penicillin/streptomycin (Invitrogen).
HEK-293A cells (50,000 cells per well) or U2OS cells (25,000 cells per well)
were plated in 96-well plates. After 24 hours of incubation after plating
(approximately 80-90% confluent), the media was replaced with 100 μL of fresh
serum-free OPTIM media (Gibco/BRL). Both cell types were transfected with the
16x-TCF(TK)-Firefly Luciferase (0.3 pg/well) and TK-Renilla-Iuciferase (0.06
μg/well) using Lipofectamine 2000 transfection reagent (Promega; Madison, WI)
according to manufacturer's instructions. Experiments were performed in duplicate.

The test cDNA constructs were transfected at different concentrations, as
needed. About 0.0005 μg/well to 0.05 μg/well cDNA of each of the constructs were
used. The transfection was performed using Lipofectamine™ 2000 (Invitrogen)
according to manufacturer's instructions. The DNA mix and reagent was incubated
for 30 min. 50 μg/well of the DNA-reagent mix was added to the 100 μL of OPTIM
media. The cells then were incubated for four hours at 37°C. The transfection
medium was replaced with fresh 150 μL of DMEM or RPMI media on the HEK 293 A
and U2OS cells, respectively. After 20-24 hours of incubation at 37°C in a CO2
incubator, the media was removed. The transfected cell monolayers were lysed by
adding 150 μL of IX lysis buffer of Dual Luci Reagent (Promega Corp.).
After 10 minutes, 20 μL of the lysate was transferred into a new 96-well
white-plate (Packard/Costar). Cell lysates were mixed with 100 μg/well of LARII
buffer (Dual Luci Reagent) and the Relative Luciferase Units (RJLUs) were measured
using a Packard Topcount NXT™ luminescence counter (Meriden, CT). This was
followed by the addition of 100 μg/well of "stop & glo" reagent (Dual Luci Reagent),
and the internal assay control renilla luciferase was measured using the Packard
Topcount NXT™ luminescence counter.
The ratio oFTCF-firefly-luci to renilla was calculated and is presented as bar
graphs in the FIGS. 1-4. The experiments were done in quadruplicate with standard
deviations calculated for the degree of error shown.
FIG. 1 indicates that when Norrin cDNA was transfected into human U2OS
cells, it gave approximately a 2- to 3-fold induction (pink slant lined bar) of TCF
signal in the absence of Wnt cDNA transfection. However, the HEK-293A
transfected cells did not produce any significant TCF-signal activation. FIG. 1 also
shows that the co-transfection of Norrin and LRP5, one of its co-receptors into HEK-
293A cells, did not activate the TCF-signal (purple checkered bar).
Interestingly, when the vectors containing the genes for Norrin (Nr) or LRP5
(L5) were co-transfected into U2OS cells, TCF-signal was enhanced. The materials
and methods utilized are as described above. See FIG. 1, right hand side. When both
. Norrin and LRP5 were co-transfected in U2OS cells, TCF signal was synergistically
enhanced (right most bar, FIG. 1) - approximately 6-fold over vector-alone (control).
Since Norrin requires the Frizzled4 (Fz4) receptor in addition to the LRP5/6 co-
receptor, the data implies that U2OS cells contain Fz4, while HEK-293A cells lack
the Fz4. Without Fz4, Norrin cannot induce TCF-signal and hence the lack of
response in the HEK-293A cells (left hand side of FIG. 1).
EXAMPLE 2: Fz4 is Required For Norrin Signaling
]n order to evaluate the Fz4-Norrin interaction, the Fz4 and LRP5 cDNAs
were transfected into both cell types which were also transfected with Norrin.

Transfections were performed as discussed above in Example 1. Detection of TCF
was performed and constructs used are as discussed in Example 1. Data was obtained
in quadruplicate with the statistical analysis seen being a calculation of the standard
deviation.
The HEK-293A cells show that there is no TCF response for vector only (V),
LRP5 only (L5), Fz4 only (F4), or LRP5 and Norrin (L5+Nr). The addition of Norrin
and Fz4 (F4+Nr) yield approximately a 6-fold increase in TCF over vector or Fz4
alone. The data in FIG. 2 demonstrates that in HEK-293A cells, the Junctional Fz4
was the limiting factor for Norrin-TCF-signal activation. Tt also indicates that in
HEK-293A cells, the Norrin-Fz4 interaction probably utilizes the endogenous LRP5/6
receptors of the cells. In addition, the co-transfection of LRP5 along with Fz4 and
Norrin (L5+F4+Nr) in HEK-293 cells results in further enhancement of TCF-signal,
up to 16-fold over LRP5-only (L5) and Norrin-only (Nr) transfected HEK293A cells.
See left side of FIG. 2.
In contrast, the same tests were conducted with U2OS cells. The U2OS cells
yielded significantly greater TCF activity by co-transfection of Fz4 and Norrin
(F4+Nr) over vector alone (V), LRP5 alone (L5)5 Frizzled4 alone (FA), and U2OS
cells co-transfected with LRP5 and Norrin (L5+Nr). See FIG. 2. The response in
U2OS cells was further enhanced when Fz4, LRP5, and Norrin (F4+L5+Nr) were all
co-transfected into the cells (about 2-fold to about 6-fold). Such data tn U2OS cells
probably indicates the presence of endogenous Wnt-signal components including Fz4
and LRP5/6 receptors.
EXAMPLE 3: Norrin induced LRP5-Fz4-TCF Signal Can Be Inhibited
SvnergisticaHv by Dkkl and Kremen2 in U2OS Cells
U2OS cells were transfected or co-transfected with blank vector (V), LRP5
(L5), Frizzled4 (Fz4), Norrin (Nr), Kremen2 (Krm2), or Dkkl as indicated in FIG. 3
according to the assay procedures set forth in Example 1. Results were obtained in
quadruplicate and the standard of deviation calculated therefrom.
FIG. 3 displays the ratio of TCF-luci to renilla signal modulation in U2OS-
TCF assays when transfected with the various cDNA constructs. On the right side of
FIG. 3, the bars indicates that transfection of LRP5 (L5), Fz4 (Fz4), or Norrin (Nr),
gave up to about 2- to 5-fold induction over vector control. However, co-transfection
of both Fz4 and Norrin (Fz4 +Nr) resulted in about a 15-fold induction of TCF signal.
The effect of Fz4 and Norrin was further enhanced by the expression of the
LRP5 (L5) cDNA (i.e., tallest and darkest bar). The maximal activity of Fz4+Nr+L5
was inhibited partially by co-transfecting the cells with Dkkl. When both Dkkl and
Krm2 were added to cells co-transfected with LRP5, Fz4, and Norrin (L5+Fz4+Nr),
the TCF-signal almost completely inhibited (right side, bar on far right).

The results displayed in FIG. 3 indicate that Norrin-LRP5-Fz4 mediated TCF-
signal can be inhibited by Dkkl. Dkkl is considered to be an efficient inhibitor of
Wnt-LRP5-Fz induced Wnt-canonical pathway. Interestingly, Kremen2 (Krm2)
addition synergizes with Dkkl and in turn resulted in blocking the Norrin action of
enhancing the Wnt pathway.
EXAMPLE 4: Norrin mediated TCF-sienals with HBM and LRP5 are
differentially inhibited by Dkkl and Kremen2
A comparison of Norrin-TCF-signal modulation by LRP5 or its gain of
function mutant, HBM, was studied in cDNA transfected U2OS cells. The results are
displayed in FIG. 4. Transfection of U2OS cells with HBM cDNA gave a slightly
greater TCF-signal than transfection with LRP5 cDNA. See left side of FIG. 4, which
shows vector only (V), LRP5-only (L5), HBM only (H), Frizzled4-only (Fz4), and
Norrin (Nr). Constructions and conditions are as described for Example 1. The
experiment was performed in quadruplicate with the standard of error calculated there
from.
Co-transfection of the U2OS cells with Norrin and Fz4, as well as either LRP5
(Nr+Fz4+L5) or HBM (Nr+Fz4+H), resulted in maximal TCF-signal with both the
LRP5 and HBM cDNAs, i.e., about 25-fold over basal activity. Dkkl co-transfection
with the Norrin, Frizzled4, LRP5, or Norrin, Frizzled4, and HBM combinations,
resulted in about a 38-40% inhibition of TCF signal. The Dkkl inhibition was further
enhanced with the addition of Kremen2 (Krm2). See right hand side of FIG. 2, two-
right most bars.
The comparative Norrin-TCF-signal analysis implies that LRP5 mutation
G171V mediated Norrin-Fz4-TCF signal confers a partial resistance to the inhibitory
action of Kremen2 and Dkkl. This interesting observation is quite similar to the
results previously observed with Wnt3a and Wntl mediated TCF-signal with LRP5
and HBM in presence of Dkkl and Kremen2. It has been reported that LRP5-Wnt-
TCF signaling modulates osteogenesis, while the HBM mutation leads to high bone
mass phenotype in humans and in transgenic mice. Based on these results, it is likely
that Norrin, as a more specific ligand of LRP5-Fz4 complex than Wnis, plays a
significant role in bone metabolism.
In each of the examples above, Dkkl can be substituted with Dkk2, Dkk3,
and/or Dkk4. Additionally, in examples using Kremen2, it would be understood that
Kremenl could be substituted and used in the same assay. Additionally, the proteins
or biologically active polypeptides can be introduced by co-transfections or by the
addition of the purified protein and/or conditioned media containing the protein and/or
biologically active polypeptides.
With the in vitro data discussed supra, Norrin has been shown to enhance the

LRP5-Frizzled4 mediated Wnt-canonical pathway by activating TCF-reporter in
U2OS bone cells and not in HEK-293A cells without the transfection of Frizzled4.
LRP5-mediated Wnt signaling is important in bone formation/maintenance as
evidenced by high bone mass ("HBM") phenotype in LRP5-G171V mutation in
humans and in transgenic animals. The data presented also shows that Norrin
mediated TCF-signal in presence of LRP5-G171V (HBM) mutant is less sensitive to
Dkkl mediated inhibition as compared to that with LRP5. Since it is hypothesized
that decreased inhibition due to G171 V mutation in LRP5 is one of the causes of
HBM phenotype, we would expect that Norrin, its expression, its induction, and/or
Norrin mimetics could enhance bone formation or maintenance in vivo. Thus, a
Norrin knockout in vivo could show osteopenia. The above assays are representative
assays for use in screening, inter alia, Norrin mimetics, Norrin agonists, and Frizzled4
agonists.
All references cited herein are incorporated by reference herein in their entirety
for all purposes.

WE CLAIM:-
1. A method of screening an agent that modulates bone metabolism or lipid
metabolism comprising: (a) having a Norrin protein or a biologically active Norrin
polypeptide fragment and a Frizzled4 protein or a biologically active polypeptide
fragment of Frizzled4 fused to LRP5 and/or LRP6 proteins or biologically active
polypeptide fragments of LRP5 and/or LRP6 in the presence of the agent; and (b)
measuring at least one parameter of bone modulation and/or lipid modulation to
screen for the agent that modulates bone metabolism or lipid metabolism.
2. The method of claim 1, wherein the agent is a Norrin mimetic.
3. The method of claim 1 or 2, wherein the parameter of bone metabolism modulation
is of bone density, bone strength, trabecular number, bone size, or tissue
connectivity, or any combination thereof.
4. The method of any of claims 1 to 3, wherein the parameter of lipid metabolism
modulation is a change in the level of HDL, VLDL, cholesterol, triglyceride, appE, or
LDL.
5. The method of any of claims 1 to 4, wherein the bone metabolism parameter
measured is altered expression of one or more of COX-2, Jun, Fos, cyclin D1,
Wnt/10B, SFRP1, connexin 43, eNOS, Wnt10B, cyclin D1, Frizzled2, and WISP2 is
modulated.
6. The method of any of claims 1 to 5, wherein the bone metabolism parameter
measured is altered expression of one or more of COX-2, Jun, Fos, cyclin D1,
Wnt10B, SFRP1, connexin 43, and eNOS is modulated.
7. The method of any of claims 1 to 4, further comprising a Dkk protein or a
biologically active Dkk polypeptide fragment.
8. The method of any of claims 1 to 4, further comprising a Kremen protein or a
biologically active Kremen polypeptide fragment.
9. The method of claim 8, further comprising a Dkk protein or a biologically active
Dkk polypeptide fragment.

10. The method of any of claims 1 to 4, further comprising a Wnt protein or a
biologically active Wnt polypeptide fragment.
11. A method of screening an agent that modulates a Norrin-Frizzled4 activity
comprising: (a) having the agent, a Norrin protein or biologically active polypeptide
fragment of Norrin, and a Frizzled4 protein or biologically active polypeptide fragment

of Frizzled4 fused to LRP5 and/or LRP6 or biologically active polypeptide fragment of
LRP5 and/or LRP6, or fused to a ligand binding domain (LBD) containing polypeptide
fragment of LRP5, and: (i) a Kremen protein; and/or (ii) a Dkk protein; and (b)
determining whether the agent modulates a Norrin-Frizzled4 activity.
12. The method of claim 11, wherein the agent is a Norrin mimetic, Dkk antagonist,
or a Kremen antagonist.
13. A method of identifying an agent that regulates bone modulation or lipid
modulation comprising: (a) administering the agent to a cell expressing Frizzled4 and
LRP5, wherein Frizzled4 is a Frizzled4 protein or a biologically active Frizzled4
polypeptide, and LRP5 is a LRP5 protein or a biologically active polypeptide of LRP5;
(b) determining whether said administration of the agent modulates a LRP5-Frizzled4
interaction; and (c) determining whether the agent modulates a bone parameter or a
lipid parameter.
14. The method of claim 13, wherein the agent is a Norrin mimetic, a Dkk antagonist,
or a Kremen antagonist.
15. The method of claim 13, wherein the cell does not express Norrin.
16. A method of screening an agent that regulates bone modulation or lipid
modulation comprising: (a) administering the agent to a cell expressing LRP5, Norrin
and Frizzled4, wherein LRP5 is a LRP5 protein or a biologically active LRP5
polypeptide, Norrin is a Norrin protein or a biologically active Norrin polypeptide, and
Frizzled4 is a Frizzled4 protein or a biologically active polypeptide of Frizzled4; (b)
determining whether said administration of the agent modulates Norrin-Frizzled4
interaction; and (c) determining whether the agent modulates a parameter of bone
modulation or lipid modulation.
17. The method of claim 16, wherein the cell expresses a non-endogenous Norrin,
LRP5, and/or Frizzled4.
18. The method of claim 16, wherein the cell does not express an endogenous
Norrin.
19. The method of claim 16, wherein LRP5 is co-expressed as a fusion polypeptide
with Frizzled4.
20. The method of claim 16, wherein the cells are bone cells, kidney cells,
mesenchymal cells, adipocytes, preadipocytes, or Xenopus cells.

21. The method of claim 16, wherein another series of cells co-expresses Norrin-
Frizzled4, and Dkk, and determining whether said agent modulates Dkk inhibition of
Norrin-Frizzled4.
22. The method of claim 16, wherein another series of cells co-expresses Norrin,
Frizzled4, and Kremen, and determining whether said agent modulates Kremen
inhibition of Norrin-Frizzled4.
23. The method of claim 21, wherein the Dkk is Dkk1, Dkk2, Dkk3, or Dkk4, or a
biologically active polypeptide of Dkk1, Dkk2, Dkk3, or Dkk4.
24. The method of claim 8 or 22, wherein Kremen is Kremeni or Kremen2, or a
biologically active polypeptide of Kremeni or Kremen2.
25. The method of claim 16, wherein the cells are bone cells, adipocytes,
preadipocytes, stem cells, or kidney cells.
26. The method of claim 16, further comprising the step of administering the agent to
an animal, and determining whether said agent induces a change in bone mass in
said animal.
27. The method of claim 10, wherein Wnt is Wnt 1 to Wnt 19.
28. A cell or a cell line lacking a native Norrin and which expresses a non-native
LRP5 and a non-native Frizzled4, wherein the non-native LRP5 is a non-native:
LRP5 protein or a biologically active fragment thereof and the non-native Frizzled4 is
a non-native Frizzled4 protein or a biologically active fragment thereof.
29. The cell line of claim 28, wherein the non-native LRP5 and/or the non-native
Frizzled4 are stably expressed.
30. The cell line of claim 28, which further expresses a non-native Dkk and/or a non-
native Kremen or biologically active polypeptide fragments of a non-native Dkk or
non-native Kremen.
31. The cell line of claim 28, wherein the non-native Dkk is Dkk1, Dkk2, Dkk3, or
Dkk4 and the non-native Kremen is Kremeni or Kremen2.

The specification discloses materials and methods for screening and identifying reagents, which modulate Norrin
activity as it relates to Wnt pathway signaling. Preferably, agents identified thereby modulate bone remodeling and/or lipid levels,
and can be Norrin mimetics and Norrin agonists, as well as other agonists and mimetics of the LRP5/Norrin/Frizzled4 complex.