WO 2006/133164 PCT/US2006/021878
ANTI-TRKB MONOCLONAL ANTIBODIES AND USES THEREOF
PRIORITY INFORMATION
The present application claims the benefit of U.S. Serial No. 60/687,705, filed
June 6, 2005, the entire contents of which are hereby incorporated by reference.
BACKGROUND OF THE INVENTION
Trk tyrosine kinase receptors are multi-domain single-transmembrane
receptors that play an important role in a wide spectrum of neuronal responses
including survival, differentiation, growth and regeneration. They are high affinity
receptors for neurotrophins, a family of protein growth factors, which includes nerve
growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-
3) and neurotrophin-4/5 (NT-4/5). The neurotrophins share highly conserved
structural features, yet their unique amino acid sequences allow each member to elicit
high affinity interactions with the extracellular domain of specific Trk receptors,
namely TrkA, B or C. Thus, NGF is a preferred ligand for TrkA; BDNF and NT-4/5
are preferred iigands for TrkB; and NT-3 has been shown to bind TrkC, although it
also appears to bind TrkA and TrkB with lower affinities.
Among the Trk receptors, the role of TrkB has been well characterized in the
central nervous system (CNS). TrkB are widely distributed in the brain, including in
the neocortex, hippocampus, striatum, olfactory formation and brainstem. Using
BDNF as a cognate ligand, the indispensable roles of TrkB in neuronal survival,
differentiation and neuroregeneration have been shown in a number of
neurodegenerative models, including stroke, spinal cord injury, axotomy and ALS.
Through various signaling analyses and receptor knockout systems, the
responses of BDNF have been shown to depend on the binding and activation of
TrkB. As noted, Trk receptors are multi-domain single-transmembrane proteins.
They consist of an extracellular ligand binding domain, a transmembrane region, and
an intracellular tyrosine kinase domain. The extracellular domain is composed of a
leucine-rich motif flanked by two cysteine clusters and two immunoglobulin(Ig)-like
domains. Trk receptors have been shown to interact with their ligands mainly through

WO 2006/133164 PCT/US2006/021878
the second Ig-like domain, although contribution of other regions such as a leucine-
rich motif and the first Ig domain in ligand-docking has been proposed. The crystal
structures of the ligand binding domains of Trk receptors as well as ligand-receptor
complexes have been resolved. The ligand-receptor interface appears to consist of
two patches; one for a conserved binding motif shared among all neurotrophins and
the other specific for each neurotrophin.
Upon the binding of neurotrophins to Trk receptors, receptor dimerization and
subsequent conformational changes occur, which are believed to lead to activation of
the intraceilular tyrosine kinase domain. There are several conserved tyrosines in the
intracellular domain of Trk receptors. Phosphorylation of the autoregulatory loop of
the kinase domain activates the kinase activity, and phosphorylation of other residues
promotes signaling by creating docking sites for adaptor proteins that couple these
receptors to intracellular signaling cascades, including Ras/extracellular signal
regulated kinase (ERK) protein kinase pathway, the PI3K/Akt kinase pathway and
phospholipase C-gamma. Although somewhat overlapping, these individual
pathways are involved in discrete biological activities: Ras/MAPK regulates neuronal
differentiation and proliferation, PI3K/Akt pathways control actin dynamics and
survival and PLC gamma is involved in calcium mobilization.
To date, there exist no successful examples of reagents that act as potent and
selective in vivo agonists of TrkB. While BDNF, as a recombinant protein, has been
shown to increase neuronal survival and neuroregeneration in a number of CNS
degenerative models in vitro and in vivo, the outcomes of BDNF protein therapy in
clinics have been negative, most likely because BDNF has a short in vivo half-life.
There is therefore a need in the art for pharmaceutical reagents that act as potent and
selective in vivo agonists of TrkB.
SUMMARY OF THE INVENTION
The present invention provides monoclonal antibodies for human TrkB. In
certain embodiments the inventive antibodies bind and activate human TrkB. In
certain embodiments the inventive antibodies are selective for human TrkB in that
they bind preferentially to TrkB over human TrkA or human TrkC. In some
2

WO 2006/133164 PCT/US2006/021878
embodiments the inventive monoclonal antibodies cross-react with murine Trkfi.
The invention also provides pharmaceutical compositions that comprise one or
more of these antibodies. The invention further provides methods for preparing the
inventive monoclonal antibodies. Humanized or veneered versions of the inventive
antibodies are also encompassed as are the hybridomas that produce the inventive
antibodies.
The invention further provides methods of using these monoclonal antibodies
as agonists of TrkB, In certain embodiments the monoclonal antibodies are used as
agonists of human TrkB. According' to such embodiments, the antibodies may be
used to treat conditions which require TrkB activation, including neurological
conditions.
The invention yet further provides methods for detecting TrkB in a sample and
methods for purifying TrkB from a sample using the inventive antibodies.
DESCRIPTION OF THE DRAWING
Figure 1 depicts TrkB-specific antibody activities in immune sera from five
immunized mice (M1-M5). Immune sera binding results from an ELISA assay with a
recombinant protein that includes the extracellular domain of human TrkB fused to
the Fc region of human IgGl (rhTrkB-EDC-Fc) are shown in (A). Immune sera
binding results from an ELISA assay with a recombinant protein that includes the
extracellular domain of murine TrkB (rmTrkB-EDC) are shown in (B). The extent to
which immune sera could block the interaction of rhTrkB-EDC-Fc and BDNF in a
competition ELISA assay are shown in (C). Immune sera binding to surface rhTrkB
on HEK-293 cells are shown in (D).
Figure 2 depicts control results that were obtained with neurotrophins in a
luciferase activity assay that uses HEK-293 cells which express surface rhTrkB.
These control results show that the assay can selectively represent the activation of
TrkB.
Figure 3 depicts results that were obtained with immune sera in the same
luciferase activity assay as Fig. 2. These results show that incubation of hTrkB cells
with immune bleeds significantly increased the luciferase signals in a dose-dependent
3

WO 2006/133164 PCT/US2006/021878
manner.
Figure 4 depicts results that were obtained with certain TrkB antibodies in the
same luciferase activity assay as Fig. 2. All of the antibodies caused dose-dependent
increases in the signal with EC50 in the range of 10-10 (M). The maximum signal
window was - 7 fold over basal for most of the antibodies, which was comparable to
the response induced by BDNF at 200 ng/ml (6.2 fold over basal).
Figure 5 depicts the results that were obtained with certain TrkB antibodies in
a neurite outgrowth assay. The assay used human neuroblastoma SY5Y cells, which
are known to express TrkB upon neuronal differentiation. Addition of BDNF and of
most antibodies promoted neurite outgrowth as demonstrated by increases in the
neurite length (A) and the number of branch points (B). Representative images of a
few treatment groups are shown in (C). The results from full-dose response analyses
obtained for a subset of these antibodies are shown in (D).
Figure 6 depicts the results that were obtained with certain TrkB antibodies in
a neuroprotection assay. The assay measures the survival of differentiated SY5Y
cells following serum withdrawal injury. Results showed that BDNF and several
antibodies protected differentiated SY5Y cells, demonstrating dose-dependent
increases in cell viability.
Figure 7 depicts some of the results that were obtained when three anti-TrkB
antibodies, 18C3,29D7 and 17D11, which were shown to bind to recombinant murine
TrkB (rmTrkB), were tested in rat cerebellar granule neuron (CGN) cultures for
activity against the endogenous rat TrkB receptor. Results are shown for a neurite
outgrowth assay and a neuroprotection assay (29D7 only).
Figure 8 depicts the results that were obtained when certain anti-TrkB
antibodies were evaluated in Western analysis for induction of TrkB
autophosphorylation. All antibodies led to robust TrkB phosphorylation, and these
effects were antagonized by treatment with the kinase inhibitor K252a, indicating that
these TrkB-binding antibodies caused activation of TrkB.
Figure 9 depicts the results from a FACS TrkA binding assay (A) and a TrkA
luciferase activity assay (B) that were obtained with certain anti-TrkB antibodies.
These results show that inventive antibodies do not bind or activate human TrkA
4

WO 2006/133164 PCT/US2006/021878
cells.
Figure 10 depicts the results that were obtained when certain anti-TrkB
antibodies were evaluated in a TrkC luciferase activity assay. These results show that
inventive antibodies do not activate human TrkC cells.
Figure 11 depicts the rodent model of neonatal hypoxia-ischemia (HI) that is
based on the Levine procedure.
Figure 12 depicts the amino acid sequence of human TrkB (SEQ ID NO:1).
Figure 13 depicts the amino acid sequence of murine TrkB (SEQ ID NO:2).
Figure 14 depicts the amino acid sequence of rat TrkB (SEQ ID NO.3).
Figure 15 depicts the amino acid sequence of chicken TrkB (SEQ ID NO;4).
Figure 16 depicts the amino acid sequence of human TrkA (SEQ ID NO:5).
Figure 17 depicts the amino acid sequence of human TrkC (SEQ ID NO:6).
Figure 18 shows the phage binding ELISA data from a representative set of
TrkB positive scFv clones (BMV library panning selection).
Figure 19 shows the binding specificity of TrkB selected scFv antibodies
tested using the FACS assay. The data shows that TrkB selected scFv antibodies
react with membrane associated TrkB in a specific manner (filled histograms) with no
cross-reactivity to control cells (open histograms).
Figure 20 show ELISA binding results for 29D7 IgG antibody and its Fab
fragment with human (A) and mouse (B) TrkB. Both total IgG and Fab of 29D7
showed dose-dependent binding activities to human and mouse TrkB. However, the
ED50 value of the Fab was about 100 fold lower than that of intact 27D7. Neither an
isotype control IgGl nor its Fab fragment bound to TrkB.
Figure 21 shows luciferase activity in HEK-293 cells measured 16 hours post-
treatment with 29D7 IgG or 29D7 Fab. Both total IgG and Fab of 29D7 induced
dose-dependent luciferase activities, indicating activation of TrkB. However, the
EC50 value of 29D7 Fab was about 27 fold higher than that of 27D7 IgG (0.083 nM
and 2.28 nM for 27D7 IgG and 29D7 Fab, respectively). Neither an isotype control
IgGl nor its Fab fragment had effects on TrkB activation.
Figure 22 shows results from the epitope mapping analysis of anti-TrkB
5

WO 2006/133164 PCT/US2006/021878
monoclonal antibodies 17D11,29D7,7F5,11E1 and 19E12. The 17D11 monoclonal
antibody recognizes loop-3 of the IgG-2 segment of human TrkB (KNEYGKD, SEQ
IDNO:7, amino acids 364 to 370 of SEQ ID NO:1); and loop-1 of the IgG-2 segment
of human TrkB (KGNPKP, SEQ ID NO:8, amino acids 308 to 313 of SEQ ID NO:1).
The 29D7, 7F5,11E1 and 19E12 monoclonal antibodies all recognize loop-3 of the
IgG-1 segment of human TrkB (ENLVGED, SEQ ID NO:10, amino acids 269 to 275
of SEQ ID NO:1); and may also recognize loop-1 of the IgG-1 segment of human
TrkB (AGDPVP, SEQ ID NO:11, amino acids 221 to 226 of SEQ ID NO:1).
Figure 23 compares the stains that were obtained from the striatum and
hippocampus for different treatments after hypoxic-ischemic (HI) injury.
Figure 24 compares the total brain tissue loss (A) and sub-regional tissue loss
(B) for different treatments after HI injury. The data are shown as mean and standard
error of the mean (S.E.M.). BDNF and anti-TrkB monoclonal antibody 29D7 showed
significant dose-dependent protection of the brain from HI injury. The protection was
not localized to any specific sub-region of the brain but greatest protection was
generally observed in the cortex.
Figure 25 shows the results of a DEVD-AMC cleavage assay that was used to
determine caspase-3 activities for different treatments after HI injury. The results for
different regions of the brain are shown as mean ± S.E.M. BDNF and anti-TrkB
monoclonal antibody 29D7 blocked caspase-3 activation caused by HI injury.
Figure 26 shows immunoblots of protein samples taken from mice given
different treatments after HI injury. The samples were separated by SDS-PAGE and
subjected to immunoblotting with antibodies against certain caspase-3 substrates
(PARP and a-spectrin). Antibodies against P-actin were used as controls to verify
equal loading of proteins. The results show that cleavage of these caspase-3
substrates was inhibited by BDNF and anti-TrkB monoclonal antibody 29D7 (C =
contra and I = ipsi sides of carotid ligation).
Figure 27 shows immunoblots of protein samples taken from normal mice
after an intracerebroventricular injection of anti-TrkB antibody 29D7 (or vehicle as
control). Samples were taken 1,2,6,12 and 24 hours after injection. Samples were
separated by SDS-PAGE and subjected to immunoblotting with antibodies specific to
6

WO 2006/133164 PCT/US2006/021878
proteins that are phosphorylated as a result of TrkB activation (phospho-ERK1/2 and
phospho-AKT). Antibodies against P-actin were used as controls to verify equal
loading of proteins. The results show that anti-TrkB monoclonal antibody 29D7
induced time-dependent phosphorylation of ERK1/2 and AKT (data from
hippocampal and cortical samples are shown and are representative of results obtained
from four other samples).
Figure 28 is a densitometric quantification of the ERK phosphorylation shown
in Figure 27.
Figure 29 is a densitometric quantification of the AKT phosphorylation shown
in Figure 27.
Figure 30 shows vehicle and 29D7 treated brain tissues that have been fixed
and immunoflourescently labeled with an antibody for the neuron-specific marker
NeuN (left) and an anti-phospho-ERK1/2 antibody (center). The right hand figures
show these two merged. ERK1/2 phosphorylation is significantly stronger in the
29D7 treated samples.
Figure 31 shows the time-course of ERK1/2 activation in the cortical and
hippocampal tissues following intracerebroventricular injection of anti-TrkB
monoclonal antibody 29D7.
Figure 32 shows the dose-dependent inhibition of neurite outgrowth of
primary cerebellar granule neurons caused by (A) MAG and (B) myelin.
Figure 33 shows the reversal of (A) MAG- and (B) myelin-mediated neurite
inhibition that is caused by anti-TrkB monoclonal antibody 29D7.
DESCRIPTION OF CERTAIN EMBODIMENTS
The present invention stemmed in part from the realization that monoclonal
antibodies that specifically bind to the extracellular domains of the TrkB receptor
might dimerize TrkB and be sufficient to induce the activation of the receptor and
biological responses similar to those mediated by BDNF.
Monoclonal antibodies in general represent a unique class of proteins that
have diverse utilities in research, medical diagnosis, and the clinical treatment of
disease. Advantageously, monoclonal antibodies are thought to have greater
7

WO 2006/133164 PCT/US2006/021878
pharmacokinetic stability than recombinant proteins such as BDNF thereby allowing
for sustainable pharmacological effects that have a significant benefit for clinical
applications. For example, the current clinically approved uses of monoclonal
antibody medications include prevention of organ transplant rejection, treatment of
cancers (e.g., breast, colon, non-Hodgkin's lymphoma, leukemia), rheumatoid
arthritis, prophylaxis against respiratory syncytial virus disease, Crohn's disease,
percutaneous coronary intervention, and asthma. Other medical applications for
treating a variety of human diseases with monoclonal antibodies are also currently in
clinical trials. The biological effects of these monoclonal antibodies are generally
conveyed by blocking or neutralizing molecular events exerted by endogenous
ligands, thus primarily acting as specific high-affinity antagonists. In the present
invention, monoclonal antibodies are used as agonists that mimic the biological
effects of receptor-ligand interactions.
1. Monoclonal antibodies and hybridomas
In one aspect, the present invention provides monoclonal antibodies that bind
human TrkB (SEQ ID NO:1). In certain embodiments, these antibodies bind human
TrkB (SEQ ID NO:1) with an ED50 in the range of about 10 pM to about 500 nM, for
example in the range of about 10 pM to about 1 nM, about 10 pM to about 100 pM, or
about 10 pm to about 50 pM. As used herein the term "about" is defined to
encompass variations of ±15%.
In one embodiment the inventive antibodies are also agonists of human TrkB.
In certain embodiments, these antibodies activate human TrkB (SEQ ID NO:1) with
an EC50 in the range of about 10 pM to about 500 nM, for example in the range of
about 10 pM to about 1 nM, about 10 pM to about 100 pM, or about 10 pM to about
50 pM.
In one embodiment, inventive antibodies are selective for human TrkB (SEQ
ID NO: 1) in that they bond preferentially to TrkB (SEQ ID NO: 1) over human TrkA
(SEQ ID NO:5) or human TrkC (SEQ ID NO:6). In some embodiments, inventive
antibodies do not bind to human TrkA or human TrkC. As defined herein, an
inventive antibody "does not bind" human TrkA (SEQ ID NO:5) or human TrkC
8

WO 2006/133164 PCT/US2006/021878
(SEQ ID NO:6) if it exhibits no detectable binding with these receptors at
concentrations above about 1 nM, for example above about 10 nM, above about 100
nM or above about 1 µM.
In one embodiment, inventive antibodies are selective for human TrkB in that
they do not activate human TrkA (SEQ ID NO:5) or human TrkC (SEQ ID NQ:6).
As defined herein, an inventive antibody "does not activate" human TrkA (SEQ ID
NO:5) or human TrkC (SEQ ID NO.6) if it causes no detectable activation with these
receptors at concentrations above about 1 nM, for example above about 10 nM, above
about 100 nM or above about 1 µM.
In one embodiment, inventive antibodies block the binding between BDNF
and human TrkB (SEQ ID NO:1) with, an IC50 in the range of about 100 pM to about
500 nM, for example in the range of about 100 pM to about 1 nM, or about 100 pM to
about 500 pM. In other embodiments these antibodies do not block the binding
between BDNF and human TrkB (SEQ ID NO: 1). As defined herein, an inventive
antibody "does not block" the binding between BDNF and human TrkB (SEQ ID
NO:1) if it exhibits no detectable blocking activity at concentrations above about 1
nM, for example above about 10 nM, above about 100 nM or above about 1 µM.
In certain embodiments the antibodies of the invention belong to an IgG
isotype, e.g., the IgGl, IgG2a or IgG2b isotype.
In another aspect, the present invention provides monoclonal antibodies with
any of the properties from above that further bind and/or activate mouse TrkB (SEQ
ID NO:2). In certain embodiments, these antibodies bind and/or activate mouse TrkB
(SEQ ID NO:2) with an ED50 in the range of about 10 pM to about 500 nM, for
example in the range of about 10 pM to about 1 nM, including in the range of about
10 pM to about 500 pM and the range of about 10 pM to about 100 pM.
In another aspect, the present invention provides monoclonal antibodies with
any of the properties from above that further bind one or more specific epitopes of
human TrkB (SEQ ID NO: 1) and optionally one or more specific epitopes of mouse
TrkB (SEQ ID NO:2). In certain embodiments, these antibodies bind one or both
epitopes of human TrkB with the sequences KNEYGKD (SEQ ID NO:7, amino acids
364 to 370 of SEQ ID NO: 1) and KGNPKP (SEQ ID NO:8, amino acids 308 to 313
9

WO 2006/133164 PCT/US2006/021878
of SEQ ID NO:1). In one embodiment, these antibodies also bind one or both
epitopes of mouse TrkB with the sequences KNEYGKD (SEQ ID NO:7, amino acids
364 to 370 of SEQ ID NO.2) and RGNPKP (SEQ ID NO;9, amino acids 308 to 313
of SEQ ID NO:2). In other embodiments, the present invention provides antibodies
that bind an epitope of human TrkB with the sequence ENLVGED (SEQ ID NO.10,
amino acids 269 to 275 of SEQ IDNO:1) and optionally an epitope of human TrkB
with the sequence AGDPVP (SEQ ID NO:11, amino acids 221 to 226 of SEQ ID
NO:1). In one embodiment, these antibodies also bind an epitope of mouse TrkB with
the sequence ENLVGED (SEQ ID NO: 10, amino acids 269 to 275 of SEQ ID NO:2),
In yet another aspect, the present invention provides hybridomas that produce
any of these monoclonal antibodies. For example, hybridomas that were deposited
with the ATCC on August 18, 2005 and have been given ATCC patent deposit
designations PTA-6948 (17D11) and PTA-6949 (29D7) are provided.
In still another aspect, the present invention provides monoclonal antibodies
that are produced by the hybridomas that were deposited with the ATCC on August
18,2005 and have been given ATCC patent deposit designations PTA-6948 (17D11)
and PTA-6949 (29D7), respectively, The present invention also provides antibodies
that block the binding of these antibodies and therefore share the same binding
epitope on human TrkB (SEQ ID NO: 1).
2. Preparation of monoclonal antibodies and hybridomas
It is to be understood that the monoclonal antibodies of the invention can be
prepared by any known method. For example, they can be prepared using synthetic,
recombinant or hybridoma technology (e.g., as described in Antibodies: A Laboratory
Manual, Ed. by E. Harlow and D. Lane, Cold Spring Harbor Laboratory Press, 1988
or Monoclonal Antibodies: Principles and Practice by J.W. Goding, Academic Press,
1996). In particular it will be appreciated that the inventive antibodies can be
prepared by initially immunizing an animal with human TrkB or a derivative thereof
(e.g., a recombinant protein that includes the extracellular domain of human TrkB)
and then preparing monoclonals from suitably prepared hybridomas.
In one aspect, the monoclonal antibodies of the present invention are prepared
10

WO 2006/133164 PCT/US2006/021878
by standard hybridoma technology using at least two protein immunogens that are
derived from TrkB proteins of different species. For example, the immunogens may
include a first immunogen that was derived from human TrkB and a second
immunogen that was derived from a non-human TrkB. Preferably the immunogens
each include the extracellular domain of TrkB. In certain embodiments the at least
two immunogens are combined as a mixture for mimuruzation purposes.
In one embodiment, the first of these immunogens is a recombinant protein
that includes the extracellular domain (ECD) of human TrkB. The ECD of human
TrkB is comprised of amino acid residues C32-H430 from the full length protein
(which is set forth as SEQ 1DNO:1 in Fig. 12, GenBank Accession No. NP_00617i).
Note that all protein and nucleic acid sequences that are present as of the filing date in
the GenBank, SwissProt, EMBL databases or any other publicly available database
are incorporated herein by reference (including without limitation the sequences of
the neurotrophins, e.g., NGF, BDNF, NT-3, NT-4/5, etc.). The second immunogen is
a recombinant protein that includes the extracellular domain (ECD) of murine TrkB.
The ECD of murine TrkB is comprised of amino acid residues C32-H429 from the
full length protein (which is set forth as SEQ ID NO:2 in Fig. 13, GenBank Accession
No. PI5209). Those skilled in the art will appreciate that suitable immunogens that
include these domains can be prepared using standard recombinant technology (e.g.,
see Protocols in Molecular Biology Ed. by Ausubel et al., John Wiley & Sons, New
York, NY, 1989 and Molecular Cloning: A Laboratory Manual Ed. by Sambrook et
al., Cold Spring Harbor Press, Plainview, NY, 1989, the contents of which are
incorporated herein by reference).
In certain embodiments, the first and second immunogens are administered in
a ratio (by weight) that is greater than about 1. For example, the ratio may be about 2,
3,4,5,6,7, 8,9,10 or more. In one embodiment the ratio is greater than about 5. In
one embodiment the ratio is about 10. In one embodiment the first and second
immunogens are administered simultaneously as a mixture.
In one embodiment, one or both immunogens include versions of their
respective extracellular domains that differ slightly from the naturally-occurring
domains. For example, amino acids at the N- or C-terminus of the extracellular
11

WO 2006/133164 PCT/US2006/021878
domain may be missing. Alternatively a few amino acids within the naturally-
occurring sequence may be mutated. Preferably these mutations are conservative
substitutions. It is to be understood that these non-naturally occurring versions should
include an amino acid sequence that is at least 95%, preferably at least 96%, more
preferably at least 97%, yet more preferably at least 98% and even more preferably at
least 99% identical to the extracellular domains that are found in SEQ ID NO: 1 or
SEQ ID NO:2.
In certain embodiments, the first and/or second immunogens do not include
any of the amino acids that are found outside of the extracellular domain of TrkB
(e.g., they do not include the amino acids that are found in transmembrane and/or
intracelluiar domains of TrkB), In certain embodiments, the immunogens may
include one or more terminal amino acids that are absent from the naturally occurring
TrkB proteins. In particular, terminal amino acids may be added to increase
expression of the recombinant protein, as a consequence of the vector used for
expression, etc. In addition, amino acid segments that are absent from the protein
allergen may be added to the amino and/or carboxyl terminus of the recombinant
protein, e.g., tags for purification, labels for detection, tags that increase the solubility
of the recombinant allergen, tags that increase the stability of the'immunogens, fusion
with an unrelated protein or adjuvant carrier protein, etc. A proteolytic cleavage site
may be introduced at the junction of the added amino acid segment and the
recombinant protein terminus to enable removal of the added segment after the
recombinant protein has been purified, absorbed, etc. Common terminal
modifications used in recombinant technology are described in Current Protocols in
Molecular Biology Ed. by Ausubel et al., John Wiley & Sons, New York, NY, 1989
and Molecular Cloning; A Laboratory Manual Ed. by Sambrook et al., Cold Spring
Harbor Press, Plainview, NY, 1989.
In one embodiment, the first and second immunogens have the same
composition as the two immunogens that are described in the Examples and that were
obtained from R&D systems, Inc. (Cat. No. 397-TR/CF and 1494-TB/CF,
respectively).
In certain embodiments, the first immunogen is as described above but the
12

WO 2006/133164 PCT/US2006/021878
second immunogen includes the extracellular domain (ECD) from a non-human TrkB
species other than murine (e.g., rat, chicken, rabbit, etc.). The ECD of rat TrkB is
comprised of amino acid residues C32-H429 from the full length protein (which is set
forth as SEQ ID NO:3 in Fig. 14, GenBank Accession No. NP_036863). The ECD of
chicken TrkB is comprised of amino acid residues C32-T428 from the full length
protein (which is set forth as SEQ ID NO:4 in Fig. 15, GenBank Accession No.
CAA54468).
Once suitable immunogens have been prepared, the immunogens are injected
into any of a wide variety of animals (e.g., mice, rats, rabbits, etc). In one
embodiment the immunogens are injected into mice as described in the Examples.
For example, the immunogens are injected subcutaneously and intraperitoneally with
complete Freund's adjuvant. In certain embodiments, each animal is injected
subcutaneously at multiple different sites. In this step, the recombinant proteins may
serve as immunogens without further modification. Alternatively, a superior immune
response may be elicited if the recombinant proteins are joined to an adjuvant carrier
protein, such as bovine serum albumin or keyhole limpet hemocyanin (KLH). The
immunogens are injected into the animal host, preferably according to a
predetermined schedule incorporating one or more booster immunizations and the
animals are bled periodically. For example, in certain embodiments, one or more
booster immunizations are administered intravenously. Binding between immune
sera and the first (and optionally the second) immunogen is then optionally assessed
to confirm that a suitable titer of antibodies has been raised.
Monoclonal antibodies that are specific for one or both of the immunogens
may be prepared by any standard method. For example, the technique of Kohler and
Milstein, Eur. J. Immunol. 6:511,1976 and improvements thereto may be used.
Briefly, these methods generally involve the preparation of immortal cell lines
capable of producing antibodies having the desired specificity. Such cell lines may be
produced, for example, from spleen cells obtained from one or more animals
immunized as described above. The spleen cells are then immortalized by, for
example, fusion with a myeloma cell fusion partner, preferably one that is syngeneic
with the immunized animal. In certain embodiments animals with suitable sera are
13

WO 2006/133164 PCT/US2006/021878
boosted one or more times with the first and/or second immunogen before the spleen
cells are removed. A variety of fusion techniques may be employed. For example,
the spleen cells and myeloma cells may be combined with a nonionic detergent for a
few minutes and then plated at low density on a selective medium that supports the
growth of hybrid cells, but not myeloma cells. A certain selection technique uses
HAT (hypoxanthine, aminopterin, thymidine) selection. After a sufficient time,
usually about 1 to 2 weeks, colonies of hybrids are observed. Single colonies are
selected and their culture supematants tested for binding activity against the first (and
optionally the second) immunogen as described below.
Monoclonal antibodies may be isolated from the supematants of growing
hybridoma colonies. In addition, various techniques may be employed to enhance the
yield, such as injection of the hybridoma cell line into the peritoneal cavity of a
suitable vertebrate host, such as a murine. Monoclonal antibodies may then be
harvested from the ascites fluid or the blood. Contaminants may be removed from the
antibodies by conventional techniques, such as chromatography, gel filtration,
precipitation and extraction. The antibody isotypes can be determined using standard
methods. As discussed in the Examples and shown in Table 1, we have prepared
specific murine antibodies belonging to isotypes IgGl, IgG2a and IgG2b using these
methods.
3. Characterization of antibody binding
In certain embodiments, inventive antibodies are characterized for their
binding activities to human TrkB (e.g., using ELISA and/or FACS as described in the
Examples), In certain embodiments binding to human TrkB proteins that are
expressed on a cell surface may also be assessed (e.g., using HEK293 cells as
described in the Examples). Preferably, inventive antibodies are also tested for their
cross-species binding activity (e.g., with the second immunogen). This allows
monoclonal antibodies that bind TrkB from both species to be identified. These
antibodies are of interest since they can be tested in animal models with the
knowledge that they can also be applied in human clinical trials.
In certain embodiments it may prove advantageous to further characterize the
14

WO 2006/133164 PCT/US2006/021878
binding properties of any given monoclonal antibody. In particular, one may use a
competition assay (e.g., an ELISA) to determine whether the antibodies block the
interaction of TrkB and BDNF. One may also assess whether the antibodies bind
non-human TrkB and/or human TrkA or TrkC.
Mapping of the relative antibody binding epitopes on TrkB (human or other)
may also be conducted, e.g., by examining the activity of each individual antibody in
blocking the binding of other antibodies to TrkB. For example, the observation that
two antibodies block each other's binding suggests these antibodies may bind to the
same epitope or overlapping epitopes on TrkB,
4. Characterization of antibody function
In certain embodiments, inventive antibodies are characterized for their
functional ability to activate human TrkB. Any agonist assay may be used. The
Examples describe an exemplary luciferase assay that has been shown to selectively
represent the activation of TrkB. TrkB autophosphorylation (e.g., as measured by
Western blot) can also be used as a measure of TrkB activation.
Alternatively or additionally, the human TrkB agonist activity of the inventive
antibodies can be assessed in an assay that involves endogenous TrkB (e.g., in human
neuroblastoma SY5Y cells). As described in the Examples, such assays test for the
ability of each antibody to promote neurite growth or to increase survival of
differentiated cells following injury (e.g., serum withdrawal injury). In certain
embodiments the dose-dependent effects of the inventive antibodies on neurite growth
and/or cell viability are measured.
In certain embodiments the purified monoclonal antibodies are also
characterized for their functional ability to activate non-human TrkB (e.g., murine,
rat, chicken, rabbit, etc.). The Examples describe an assay in which the antibodies
were tested in rat cerebellar granule neuron (CGN) cultures for activity against the
endogenous rat TrkB receptor. As with the human cells a neurite outgrowth assay and
a neuroprotection assay can be performed. Other useful assays are known in the art
and will be recognized by those skilled in the art.
In yet other embodiments and as described in the Examples, the purified
15

WO 2006/133164 PCT/US2006/021878
monoclonal antibodies are further characterized for their functional ability to activate
human TrkA and/or TrkC,
5. Humanized and veneered antibodies
When using an inventive antibody for therapeutic purposes it may prove
advantageous to use a humanized or veneered version of the antibody of interest to
reduce any potential immunogenic reaction. In general, humanized or veneered
antibodies minimize unwanted immunological responses that limit the duration and
effectiveness of therapeutic applications of non-human antibodies in human
recipients.
A number of methods for preparing humanized antibodies comprising an
antigen binding portion derived from a non-human antibody have been described in
the art. In particular, antibodies with rodent variable regions and their associated
complementarity-determining regions (CDRs) fused to human constant domains have
been described (e.g., see Winter et al., Nature 349:293,1991; Lobuglio et al., Proc.
Nat. Acad. Sci. USA 86:4220,1989; Shaw et al., J. Immunol. 138:4534, 1987; and
Brown et al., Cancer Res. 47:3577,1987). Rodent CDRs grafted into a human
supporting framework region (FR) prior to fusion with an appropriate human antibody
constant domain (e.g., see Riechmann et al., Nature 332:323, 1988; Verhoeyen et al.,
Science 239:1534,1988; and Jones et al. Nature 321:522,1986) and rodent CDRs
supported by recombinantly veneered rodent FRs have also been described (e.g., see
EPO Patent Pub. No. 519,596).
Completely human antibodies are particularly desirable for therapeutic
treatment of human patients. Such antibodies can be produced using transgenic mice
that are incapable of expressing endogenous immunoglobulin heavy and light chains
genes, but which can express human heavy and light chain genes (e.g., see Lonberg t
and Huszar Int. Rev. Immunol. 13:65-93,1995 and U.S. Patent Nos. 5,545,806;
5,569,825; 5,625,126; 5,633,425; and 5,661,016).
Veneered versions of the inventive antibodies may also be used in the methods
of the present invention. The process of veneering involves selectively replacing FR
residues from, e.g., a murine heavy or light chain variable region, with human FR
16

WO 2006/133164 PCT/US2006/021878
residues in order to provide an antibody that comprises an antigen binding portion
which retains substantially all of the native FR protein folding structure. Veneering
techniques are based on the understanding that the antigen binding characteristics of
an antigen binding portion are determined primarily by the structure and relative
disposition of the heavy and light chain CDR sets within the antigen-association
surface (e.g., see Davies et al., Ann. Rev. Biochem. 59:439,1990). Thus, antigen
association specificity can be preserved in a humanized antibody only wherein the
CDR structures, their interaction with each other and their interaction with the rest of
the variable region domains are carefully maintained. By using veneering techniques,
exterior (e.g., solvent-accessible) FR residues which are readily encountered by the
immune system are selectively replaced with human residues to provide a hybrid
molecule that comprises either a weakly immunogenic, or substantially non-
immunogenic veneered surface.
6. Single-chain antibodies
Single chain antibodies can also be prepared based on the inventive antibodies.
For example, a single-chain antibody (scFv) can be engineered as described in, for
example, Colcher et al., Ann. N Y Acad Sci. 880:263-80, 1999; and Reiter, Clin.
Cancer Res. 2:245-52,1996. Specific methods are described in the Examples. The
single-chain antibody can be dimerized or multimerized to generate multivalent
antibodies having specificities for different epitopes of human TrkB.
7. Pharmaceutical compositions
Monoclonal antibodies of the invention may be administered neat in order to
activate TrkB in accordance with the present invention. More commonly, however,
they are administered in the context of a pharmaceutical composition, that contains a
therapeutically effective amount of one or more antibodies together with one or more
other ingredients known to those skilled in the art for formulating pharmaceutical
compositions.
As used herein, the terms "pharmaceutically effective amount" or
"therapeutically effective amount" mean the total amount of each active ingredient of
17

WO 2006/133164 PCT/US2006/021878
the pharmaceutical composition or method that is sufficient to show a meaningful
patient benefit, i.e., treatment, prevention or amelioration of a condition which
requires TrkB activation. When applied to an individual active ingredient that is
administered alone, the term refers to that ingredient alone. When applied to a
combination of active ingredients, the term refers to combined amounts of the active
ingredients that result in the therapeutic effect, whether administered in combination,
serially or simultaneously.
In certain embodiments of the invention, inventive antibodies are administered
with a weekly dose in the range of about 0.1 to about 1000 mg/kg body weight, or
about 1 to about 500 mg/kg body weight, in certain embodiments about 10 to about
300 mg/kg body weight. Doses may be administered as a single regimen or as a
continuous regimen divided by two or more doses over the course of a day or week.
Delivery may be as a bolus or in certain embodiments as a gradual infusion (e.g., by
injection over 30 mins). In certain embodiments one or more higher doses (e.g., 2,3
or 4 fold higher) may be administered initially followed by one or more lower
maintenance doses. The higher dose(s) may be administered at the onset of treatment
only or at the beginning of each treatment cycle. These dosage levels and other
dosage levels herein are for intravenous or intraperitoneal administration. The skilled
person will readily be able to determine the dosage levels required for a different
route of administration. It will be appreciated that, in general, the precise dose used
will be as determined by the prescribing physician and will depend not only on the
weight of the subject and the route of administration, but also on the age of the subject
and the severity of the symptoms.
Additional ingredients useful in preparing pharmaceutical compositions in
accordance with the present invention include, for example, carriers (e.g, in liquid or
solid form), flavoring agents, lubricants, solubilizers, suspending agents, fillers,
glidants, compression aids, binders, tablet-disintegrating agents, encapsulating
materials, emulsifiers, buffers, preservatives, sweeteners, thickening agents, coloring
agents, viscosity regulators, stabilizers or osmo-regulators, or combinations thereof.
Liquid pharmaceutical compositions preferably contain one or more
monoclonal antibodies of the invention and one or more liquid carriers to form
18

WO 2006/133164 PCT/US2006/021878
solutions, suspensions, emulsions, syrups, elixirs, or pressurized compositions.
Pharmaceutically acceptable liquid carriers include, for example water, organic
solvents, pharmaceutically acceptable oils or fat, or combinations thereof. The liquid
carrier can contain other suitable pharmaceutical additives such as solubilizers,
emulsifiers, buffers, preservatives, sweeteners, flavoring agents, suspending agents,
thickening agents, colors, viscosity regulators, stabilizers or osmo-regulators, or
combinations thereof. If the liquid formulation is intended for pediatric use, it is
generally desirable to avoid inclusion of alcohol.
Examples of liquid carriers suitable for oral or parenteral administration
include water (preferably containing additives such as cellulose derivatives such as
sodium carboxymethyl cellulose), alcohols or their derivatives (including monohydric
alcohols or polyhydric alcohols such as glycols) or oils (e.g., fractionated coconut oil
and arachis oil). For parenteral administration the carrier can also be an oily ester
such as ethyl oleate and isopropyl myristate. The liquid carrier for pressurized
compositions can be halogenated hydrocarbons or other pharmaceutically acceptable
propellant.
Solid pharmaceutical compositions preferably contain one or more solid
carriers, and optionally one or more other additives such as flavoring agents,
lubricants, solubilizers, suspending agents, fillers, glidants, compression aids, binders
or tablet-disintegrating agents or an encapsulating material. Suitable solid carriers
include, for example, calcium phosphate, magnesium stearate, talc, sugars, lactose,
dextrin, starch, gelatin, cellulose, methyl cellulose, sodium carboxymethyl cellulose,
polyvinylpyrrolidine, low melting waxes or ion exchange resins, or combinations
thereof, In powder pharmaceutical compositions, the carrier is preferably a finely
divided solid which is in admixture with the finely divided active ingredient. In
tablets, the active ingredient(s) are generally mixed with a carrier having the
necessary compression properties in suitable proportions, and optionally, other
additives, and compacted into the desired shape and size.
In some embodiments of the invention, pharmaceutical compositions are
provided in unit dosage form, such as tablets or capsules. In such form, the
composition is sub-divided in unit dose containing appropriate quantities of the active
19

WO 2006/133164 PCT/US2006/021878
ingredient(s). The unit dosage forms can be packaged compositions, for example
packeted powders, vials, ampoules, pre-filled syringes or sachets containing liquids.
The unit dosage form can be, for example, a capsule or tablet itself, or it can be an
appropriate number of any such compositions in package form.
Thus, the present invention also provides a pharmaceutical composition in unit
dosage form for activating TrkB, where the composition contains a therapeutically
effective unit dosage of at least one monoclonal antibody of the invention. As one
skilled in the art will recognize, the certain therapeutically effective unit dosage will
depend on the method of administration.
The present invention also provides a therapeutic package for dispensing the
monoclonal antibodies of the invention to an individual being treated for a condition
which requires TrkB activation. In some embodiments, the therapeutic package
contains one or more unit dosages of at least one inventive monoclonal antibody, a
container containing the one or more unit dosages, and labeling directing the use of
the package for treatment. In certain embodiments, the unit dose is in tablet or
capsule form. In some cases, each unit dosage is a therapeutically effective amount.
8. Other pharmaceutical agents
According to the present invention, monoclonal antibodies of the invention
may be administered alone to modulate TrkB activity. Alternatively the antibodies
may be administered in combination with (whether simultaneously or sequentially)
one or more other pharmaceutical agents useful in the treatment, prevention or
amelioration of one or more other conditions (including symptoms, disorders, or
diseases) which require TrkB activity.
For example, other pharmaceutical agents that can modulate TrkB activity
may be used in combination with the monoclonal antibodies of the invention,
including other activators of TrkB. U.S. Patent Nos. 5,770,577; 6,077,829; 6,723,701
and 6,800,607 (each incorporated herein by reference in their entirety) describe
BDNF derivatives and compositions that may be useful in accordance with the
practice of the present invention.
20

WO 2006/133164 PCT/US2006/021878
Additionally or alternatively, the monoclonal antibodies may be used in
conjunction with other pharmaceutical agents that are useful in the treatment,
prevention or amelioration of neurological disorders and diseases. In certain
embodiments, the monoclonal antibodies are combined with agents that are useful in
the treatment, prevention or amelioration of disorders and diseases-caused by injuries
to the nervous system (e.g., by wound, surgery, ischemia, infection, metabolic
diseases, malnutrition, malignant tumor, toxic drugs, etc.). It is to be understood that
any suitable agent known in the art may be used, including those listed in the
Physicians, Desk Reference, 55th Edition, 2001, published by Medical Economics
Company, Inc. at Monvale, NJ, the relevant portions of which are incorporated herein
by reference.
9. Therapeutic uses
In one aspect, inventive antibodies are useful for treating conditions (including
symptoms, disorders, or diseases) which require activation of TrkB. Such methods
involve administering a therapeutically effective amount of one or more inventive
antibodies to the individual. In certain embodiments, the invention provides methods
for treating neurological conditions. For example, and without limitation, inventive
antibodies may be used to treat individuals with a nervous system that has been
injured by wound, surgery, ischemia, infection, metabolic diseases, malnutrition,
malignant tumor, toxic drug, etc. Specific examples include stroke, spinal cord
injury, traumatic brain injury, retinal degeneration and axotomy. The inventive
antibodies may also be used to treat disorders such as attention-deficit hyperactivity
disorder (ADHD), depression and age-associated mental impairment (i.e., by
providing cognitive enhancement). The inventive antibodies may also be used to treat
congenital or neurodegenerative conditions including Alzheimer's disease,
Parkinson's disease, Huntington's chorea, amyotrophic lateral sclerosis (ALS) and
conditions related to these. The benefits of TrkB activation in treating non-
neurological diseases such as cancer and diabetes has also been described and the
inventive antibodies may therefore find utility in such contexts (e.g., U.S. Patent Nos.
21

WO 2006/133164 PCT/US2006/021878
5,877,016 and 6,800,607 describe the benefits of TrkB activation for treating cancer
and diabetes, respectively).
The methods of this invention are useful for treating the conditions described
herein in adults and children. They may also be utilized for veterinary applications,
particularly including canine and feline applications. If desired, the methods herein
may also be used with farm animals, such as ovine, bovine, porcine and equine
breeds.
Inventive methods involve delivery of inventive monoclonal antibodies via
any appropriate route of administration including, for example, parenteral,
intravenous, topical, nasal, oral (including buccal or sublingual), rectal or other
modes. In general, the antibodies may be formulated for immediate, delayed,
modified, sustained, pulsed, or controlled-release delivery.
In certain embodiments, the antibodies are formulated for delivery by
injection. In such embodiments, administration may be, for example, intracaveraous,
intravenous, intra-arterial, intraperitoneal, intrathecal, intraventricular, intraurethral,
intrasternal, intracranial, intramuscular or subcutaneous, or via by infusion or
needleless injection techniques. For such parenteral administration, the antibodies of
the invention may be prepared and maintained in conventional lyophylized
formulations and reconstituted prior to administration with a pharmaceutically
acceptable saline solution, such as a 0.9% saline solution. The pH of the injectable
formulation can be adjusted, as is known in the art, with a pharmaceutically
acceptable acid, such as methanesulfonic acid. Other acceptable vehicles and solvents
that may be employed include Ringer's solution and U.S.P. In addition, sterile, fixed
oils are conventionally employed as a solvent or suspending medium. For this
purpose any bland fixed oil can be employed including synthetic mono- or
diglycerides. In addition, fatty acids such as oleic acid are used in the preparation of
injectables. The injectable formulations can be sterilized, for example, by filtration
through a bacterial-retaining filter, or by incorporating sterilizing agents in the form
of sterile solid compositions which can be dissolved or dispersed in sterile water or
other sterile injectable medium prior to use.
22

WO 2006/133164 PCT/US2006/021878
In order to prolong the effect of the inventive antibody, it may be desirable to
slow its absorption from an intramuscular or subcutaneous injection. Delayed
absorption of such an administered antibody may be accomplished by dissolving or
suspending the agent in an oil vehicle. Injectable depot forms are made by forming
microencapsule matrices of the antibody in biodegradable polymers such as
polylactide-polyglycolide. Depending upon the ratio of antibody to polymer and the
nature of the particular polymer employed, the rate of antibody release can be
controlled. Examples of other biodegradable polymers include poly(orthoesters) and
poly(anhydrides). Depot injectable formulations are also prepared by entrapping the
antibodies in liposomes or micro emulsions which are compatible with body tissues.
For application topically to the skin, the antibodies can be formulated as a
suitable ointment containing the active ingredient suspended or dissolved in, for
example, a mixture with one or more of the following: mineral oil, liquid petrolatum,
white petrolatum, propylene glycol, polyoxyethylene polyoxypropylene compound,
emulsifying wax and water. Alternatively, they can be formulated as a suitable lotion
or cream, suspended or dissolved in, for example, a mixture of one or more of the
following: mineral oil, sorbitan monostearate, a polyethylene giycol, liquid paraffin,
polysorbate 60, cetyl esters wax, cetearyl alcohol, 2-octyldodecanol, benzyl alcohol
and water.
The inventive antibodies can also be administered intranasally or by inhalation
and are conveniently delivered in the form of a dry powder inhaler or an aerosol spray
presentation from a pressurized container, pump, spray, atomiser or nebuliser, with or
without the use of a suitable propellant, e.g. dichlorodifluoromethane,
trichlorofluoromethane, dichlorotetrafiuoroethane, a hydrofluoroalkane, carbon
dioxide or other suitable gas. In the case of a pressurized aerosol, the dosage unit may
be determined by providing a valve to deliver a metered amount. The pressurized
container, pump, spray, atomiser or nebuliser may contain a solution or suspension of
the antibody, e.g., using a mixture of ethanol and the propellant as the solvent, which
may additionally contain a lubricant, e.g., sorbitan trioleate. Capsules and cartridges
(made, for example, from gelatin) for use in an inhaler or insufflator may be
23

WO 2006/133164 PGT/US2006/021878
formulated to contain a powder mix of the antibodies of the invention and a suitable
powder base such as lactose or starch.
For inventive methods utilizing oral delivery, such delivery may be
accomplished using solid or liquid formulations, for example in the form of tablets,
capsules, multi-particulates, gels, films, ovules, elixirs, solutions or suspensions. In
certain embodiments, the monoclonal antibodies are administered as oral tablets or
capsules. Such preparations may be mixed chewable or liquid formulations or food
materials or liquids if desirable, for example to facilitate administration to children, to
individuals whose ability to swallow tablets is compromised, or to animals.
Compositions for rectal administration are preferably suppositories which can
be prepared by mixing the inventive antibodies with suitable non-irritating excipients
or carriers such as cocoa butter, polyethylene glycol or a suppository wax which are
solid at ambient temperature but liquid at body temperature and therefore melt in the
rectal vault and release the antibodies. Retention enemas and rectal catheters can also
be used as is known in the art. Viscosity-enhacing carriers such as hydroxypropyl
cellulose are also certain carriers of the invention for rectal administration since they
facilitate retention of the pharmaceutical composition within the rectum. Generally,
the volume of carrier that is added to the pharmaceutical composition is selected in
order to maximize retention of the composition. In particular, the volume should not
be so large as to jeopardize retention of the administered composition in the rectal
vault.
10. Diagnostic uses
In another aspect, inventive antibodies may be used for detecting TrkB in a
sample (e.g., in order to diagnose a disorder characterized by over or under expression
of TrkB). According to such methods an inventive antibody is combined with a
sample under conditions to allow specific binding. The specific binding is then
detected thereby indicating the presence of TrkB in the sample.
The sample can be derived from a body fluid (e.g., from cerebrospinal fluid,
blood, serum, urine, etc.) or an extract of cells or tissue (e.g., a biopsy sample).
Detection of binding can be facilitated by coupling (i.e., physically linking) the
24

WO 2006/133164 PCT/US2006/021878
antibody to a detectable substance (i.e., antibody labeling). Examples of detectable
substances include various enzymes, prosthetic groups, fluorescent materials,
luminescent materials, bioluminescent materials, and radioactive materials. Examples
of suitable enzymes include horseradish peroxidase, alkaline phosphatase, β-
galactosidase, or acetylcholinesterase; examples of suitable prosthetic group
complexes include streptavidin/biotin and avidin/biotin; examples of suitable
fluorescent materials include umbelliferone, fluorescein, fluorescein isothiocyanate,
rhodamine, dichlorotriazmylamine fluorescein, dansyl chloride or phycoerythrin; an
example of a luminescent material includes luminol; examples of bioluminescent
materials include luciferase, luciferin, and aequorin; and examples of suitable
radioactive material include 125I,131I, 35S or 3H.
A variety of protocols for measuring antibody binding, including ELISA and
FACS, are known in the art and provide a basis for diagnosing altered or abnormal
levels of TrkB expression. ELISA and FACS are further described in the Examples,
Normal or standard values for TrkB expression are established by combining samples
taken from normal individuals with an inventive antibody under conditions suitable
for complex formation. The amount of standard complex formation may be
quantified by various methods depending on the nature of the detectable substance.
Preferably the amount of standard complex formation is quantified by photometric
means. Levels of TrkB expression in samples from diseased individuals are then
compared with the standard values. Deviation between standard and diseased values
establishes the parameters for diagnosing disease.
In certain embodiments, the inventive methods may be used to diagnose
neurological conditions that are characterized by over or under expression of TrkB.
For example, and without limitation, inventive methods may be used to identify
nervous systems that have been injured by wound, surgery, ischemia, infection,
metabolic diseases, malnutrition, malignant tumor, toxic drug, etc. Specific examples
include stroke, spinal cord injury, traumatic brain injury, retinal degeneration and
axotomy. The inventive methods may also be used to diagnose disorders such as
attention-deficit hyperactivity disorder (ADHD), depression and age-associated
mental impairment (i.e., by providing cognitive enhancement). The inventive
25

WO 2006/133164 PCT/US2006/021878
methods may also be used to diagnose congenital or neurodegenerative conditions
including Alzheimer's disease, Parkinson's disease, Huntington's chorea,
amyotrophic lateral sclerosis (ALS) and conditions related to these.
11. Purification uses
The invention also provides a method of using an antibody to purify TrkB
from a sample comprising combining an inventive antibody with a sample under
conditions to allow specific binding, thereby producing an antibody-TrkB receptor
complex, separating the antibody-TrkB receptor complex from the remainder of the
sample and then separating the antibody from the TrkB receptor, thereby obtaining a
purified TrkB receptor. It will be appreciated that the inventive antibodies can be
used to isolate TrkB by any standard technique, such as affinity chromatography or
immunoprecipitation.
EXAMPLES
The present invention is further illustrated and supported by the following
examples. However, these examples should in no way be considered to further limit
the scope of the invention. To the contrary, one having ordinary skill in the art would
readily understand that there are other embodiments, modifications, and equivalents
of the present invention without departing from the spirit of the present invention
and/or the scope of the appended claims.
EXAMPLE 1
This example describes the preparation and in vitro characterization and
testing of a plurality of TrkB antibodies.
Materials and Methods
Immunogens
Murine anti-TrkB antibodies were prepared using a mixture of two protein
immunogens: a first recombinant protein that includes the extracellular domain (ECD)
of human TrkB (rhTrkB-ECD) (R&D systems, Inc., Cat. No. 397-TR/CF) and a
26

WO 2006/133164 PCT/US2006/021878
second recombinant protein that includes the extracellular domain of murine TrkB
(rmTrkB-ECD) (R&D system, Inc., Cat. No. 1494-TB/CF).
The extracellular domain of human TrkB is comprised of amino acid residues
C32-H430 of the full length protein (which is set forth as SEQ ID NO:1 in Fig. 12,
GenBank Accession No. NP_006171). rhTrkB-ECD was expressed in murine
myeloma cell line NSO. The calculated molecular mass of monomeric rhTrkB-ECD
is 44 kDa; however, when glycosylated it migrates as a broad band of 80-100 kDa in
SDS-PAGE under reducing conditions.
The extracellular domain of murine TrkB is comprised of amino acid residues
C32-H429 of the full length protein (which is set forth as SEQ ID NO:2 in Fig. 13,
GenBank Accession No, P15209). In the mhTrkB-ECD used for this Example, this
sequence is flanked by an N-terminal human CD33 signal peptide which is cleaved
during expression and a C-terminai His Tag. rhTrkB-ECD was also expressed in
murine myeloma cell line NSO. The calculated molecular mass of the monomeric
mhTrkB-ECD is 45.9 kDa; however, when glycosylated it migrates as a broad band of
75-100 kDa in SDS-PAGE under reducing conditions.
Immunization schedules
Five 8-week old female BALB/c mice were immunized with 10 µg of rhTrkB-
ECD that was pre-mixed with complete Freund's adjuvant (CFA). The mixture was
separated into portions that were injected subcutaneously and intraperitoneally 4 times
biweekly (i.e., at weeks 0,2, 4, and 6). The mice were also immunized at week 7 by
subcutaneous and intraperitoneal injection with 1 ug of rmTrkB-ECD that was pre-
mixed with complete Freund's adjuvant (CFA). Mice bleeds were collected at weeks
5 and 7 and antibody responses in the sera were evaluated.
Generation of murine anti-TrkB monoclonal antibodies (mAbs)
Three of the five mice were additionally boosted intravenously with 10 µg
rhTrkB and 1 µg rmTrkB 3 days prior to the cell fusion (which occurred at week 12).
Splenocytes from these three mice were fused with murine myeloma cells
P3X63Ag8.653 (ATCC, Cat. No. CRL-1580) at 4:1 ratio using 50% polyethylene
27

WO 2006/133164 PCT/US2006/021878
glycol (MW 1500) (Roche Diagnostics Corp., Cat. No. 783641). After fusion, cells
were seeded and cultured in 96-well plates at 1x105 cells/well in selection medium
(RPMI1640 containing 20% FBS and 5% Origen) (IGEN International, Inc., Cat. No.
210001), 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, lxHEPES
and lxHAT (hypoxanthine-aminopterin-thymidine) (Sigma, Cat. No. H0262).
Hybridoma supernatants were screened for binding with rhTrkB by ELISA and
staining on rhTrkB-expressing HEK293 stable cells by FACS analysis (see below).
The hybridoma supernatants selected to be positive were subsequently tested for
agonist activities on rhTrkB using a luciferase assay (see below). Selected
hybridomas were subcloned four times by serial dilutions and once by FACS sorting
(see below). Conditional medium was harvested from the stable hybridoma culture.
Prosep-A (Montage Antibody Purification Spin columns, Millipore, Cat. No. P36486)
was used to purify IgG from the hybridoma conditional medium. The Ig class of each
mAb was determined with a murine mAb isotyping kit (IsoStrip, Boehringer
Mannheim Corp., Cat. No. 1493027).
ELISAs
To measure the presence of TrkB-specific antibodies, 96-well plates
(Maxisorp, Nunc) were coated with 1 µg/ml rhTrkB-EDC-Fc (R&D system, Cat. No.
688-TK) or rmTrkB-Fc (R&D system) and incubated overnight at 4 °C. After
washing the plates and blocking the wells with PBS (10 mM sodium phosphate, 150
mMNaCl, pH 7.2) containing 1% BSA and 0.05% Tween-20,100 ul of diluted
immune serum or hybridoma supernatants were added and incubated for 1 hr at room
temperature. The plates were washed, and the bound anti-TrkB antibodies were
detected using peroxidase conjugated goat anti-murine IgG (H+L) (PIERCE, Cat. No.
31434) followed by incubation with the substrate TMB (BioFX Laboratories, Cat. No.
TMBW_1000-01). The absorbance values were determined at 450 ran in a
spectrophotometer.
To determine the mAb concentration in the hybridoma supernatant, 96-well
plates were coated with 1 µg/ml goat anti-murine IgG (Fcγ) (PIERCE, Cat. No.
31123) in PBS and incubated overnight at 4 °C. After washing and blocking the wells
28

WO 2006/133164 PCT/US2006/021878
with PBS containing 1% BSA and 0.05% Tween-20,100 µl of diluted hybridoma
supernatants were added for 1 hr at room temperature. Purified isotype-matching
murine IgG was used as a standard for the quantification of anti-TrkB IgG
concentrations. The plates were washed, and HRP labeled goat anti-murine IgG-Fc
were added and incubated for 1 hr at room temperature. After washing the wells, the
substrate TMB was added. Absorbance was determined at 450 nm.
Characterization of relative antibody binding epitopes using competition binding
To determine how the binding of anti-TrkB IgG to TrkB protein affects the
BDNF interaction with TrkB, a 96-well plate was coated with 0.3 µg/ml of BDNF
(R&D system, Cat. No. 248-BD/CF) in PBS and incubated overnight at 4 °C. After
washing and blocking the wells with PBS containing 1% BSA and 0.05% Tween-20,
100 µl of pre-incubated hybridoma supernatant (or diluted immune serum) mixture
with rhTrkB-EDC-Fc was added to the plate and incubated for 1 hr at room
temperature. After washing the plate, peroxidase conjugated goat anti-human IgG,
(Fey) (PIERCE, Cat. No. 31416) was added and incubated for 1 hr at room
temperature. The wells were washed and the substrate TMB was added. Absorbance
was determined at 450 nm.
To map the relative antibody binding epitopes on rhTrkB, a 96-well plate was
coated with 1 |ig/ml of each individual TrkB-specific mAb in PBS and incubated
overnight at 4 °C. After washing and blocking the wells with PBS containing 1%
BSA and 0.05% Tween-20, 100 ul of mixture of each individual pre-incubated TrkB
mAb (20 µg/ml) and rhTrkB-EDC-Fc (0.1 µg/ml) was added to the wells and
incubated for 1 hr at room temperature. The plate was washed and incubated for 1 hr
with peroxidase conjugated goat ant-human IgG, (Fey). After washing the plate, the
substrate TMB was added. Absorbance was determined at 450 nm.
Luciferase assay
A stable line of HEK-293 cells expressing rhTrkB was generated by
transfecting HEK-293 cells with pcDNA-hTrkB (full length, see GenBank Accession
No. NM_06180). Transfected cells were selected in the presence of hygromycin for
29

WO 2006/133164 PCT/US2006/021878
2 wks with limited dilutions. Following initial evaluation, a single clone of cells was
chosen for the study. For the luciferase assay, cells were plated at 1.5x104 cells/well
in 100 µl growth medium in 96-well plates. The next day, cells were treated with 10
µl of 10x final concentration of BDNF or test antibodies. Luciferase activities were
measured 16 hr after the treatments using the Promega Steady-Glo assay kit according
to the manufacturer's protocol. In brief, media was replaced with 100 µl of PBS and
100 µl of Steady-Glo reagent was added. After sealing the plates with TopSeal, the
plates were shaken at Titer Plate Shaker at speed ~ 5 for 5 minutes and then
luminescence was measured using a TopCount NXT v2.13 instrument (Packard).
FACS analysis
HEK-293 cells expressing rhTrkB were detached from the plates with PBS
containing 5 mM EDTA and transferred in to 5 ml Falcon tubes (Becton Dickinson,
Cat. No. 352063) with 2x105 cells per tube. Cells were washed once with PBS by
centrifuging at 800 rpm at 4 °C for 3 min, and incubated for 30 min at 4 °C with 100
µl of hybridoma culture supernatant, purified antibodies or immune serum diluted in
PBS with 1% FBS. The cells were washed 3 times with 1 ml PBS containing 1% FBS
and incubated for 30 min at 4 °C in the dark with PE labeled goat anti-murine IgG,
F(ab')2 fragment (DAKO Corporation, Cat. No. R0480) in PBS containing 1% FBS.
Cells were washed three times again and re-suspended in 250 µl PBS containing 1%
FBS. Popidium iodide was used for detection of dead cells, which were excluded
from analysis. The fluorescence of 5000 cells/tube was counted by a FACScan flow
cytofiuorometer (Becton Dickinson).
TrkB autophosphorylation assay
HEK-293 cells expressing rhTrkB were plated in 24 well plates at 2x105
cells/well in a DMEM growth media. The next day, cells were incubated in serum-
free DMEM for 90 min, then stimulated with BDNF or testing antibodies at different
concentrations for 30 min at 37 °C. After washing with PBS once, cells were lysed in
Laemmli Sample Buffer (Bio-Rad) preheated at 95 °C. Lysates were run through
QIAshredder column (Qiagen) and 20 µl of samples were resolved on 4-12% Bis-Tris
30

WO 2006/133164 PCT/US2006/021878
gel (Invitrogen). Following transfer to nitrocellulose membranes, phosphorylated
TrkB bands were detected using PY490 phospho-Trk-specific antibody (1:100, Cell
Signaling, Cat. No. 9141) followed by incubation with HRP-conjugated anti-rabbit
secondary antibody (Molecular Probes). The signals were developed using ECL plus
kit (Amersham).
Neurite outgrowth assay
Human neuroblastoma SH-SY5Y cells were grown in DMEM:F12(1:1)
supplemented with 2 mM L-glutamine, 15 % FBS and pen/strep. For the neurite
outgrowth assay, cells were plated in 96 well tissue culture plates at the density of
4xl03 cells/well and incubated with 10 µM all-trans-retinoic acid (RA) to induce
neuronal differentiation. On the third day, the media was replaced with fresh growth
media with or without BDNF or testing antibodies, as indicated in the results. After
three additional days in culture, cells were fixed by IC-Fix for 30 min at room
temperature and further processed for immunostaining of beta-Tubulin III. First, cells
were permeabilized by brief incubation in 0.2% Triton in phosphate-buffered solution
(TPBS). Then samples were incubated in 1.5% normal goat serum (NGS) in TPBS
for 30 min to block non-specific binding, followed by incubation with anti-beta-
Tubulin III mAb (Tujl, 1:1000, Covance) in 1.5% NGS/TPBS. Tujl signals were
detected using Alexa 488 murine anti-goat antibody (1:500, Molecular Probes) and
neurite outgrowth was analyzed using Cellomics arrayscan.
For the measurement of neurite promoting effects in the primary neurons, rat
or murine cerebellar granular neuron (CGN) cultures were prepared. Briefly, the
cerebellum was dissected from animals postnatal day 7 and cut into small pieces. The
tissue was treated with papain (Worthington Biochemical Corp.) for 30 min at 37 °C
and dispersed by gentle trituration. Following centrifugation at 300 g for 5 min,
dissociated cells were reconstituted in Neurobasal medium containing B27
supplement, 0.5 mM l-glutamine, and 25 mM potassium chloride and plated in 96
well Biocoat plates precoated with poly-d-lysin (BD bioscience) at a density of
1.2x104 cells/well. Cells were treated with BDNF or testing antibodies for 24 hrs,
fixed with IC-Fix for 30 min at room temperature and processed for Tuj 1
31

WO 2006/133164 PCT/US2006/021878
immunostaining as described above.
Cell survival assay
SH-SY5Y cells were plated in 96 well plates at 1x104 cells/well and incubated
with RA (10 µM) to induce neuronal differentiation. After 3 days, the culture media
was switched to growth media without serum (serum-free media) and cells were
treated with BDNF, testing antibodies or vehicle. Following two additional days in
culture, cell viability was measured by MTT assay using the CellTiter 96 Non-
Radioactive Cell Proliferation Assay Kit (Promega) according to the manufacturer's
protocol.
Neuroprotection assay
Rat or murine CGN cultures were prepared from 7 day-old pups as described
above and plated in 96 well Biocoat plates precoated with poly-d-lysin (BD
bioscience) at a density of 7.3x104 cells/well. 24 hrs after plating, cells were
subjected to potassium serum deprivation (KSD) injury, which is known to result in
significant death of cerebellar granular neurons. Sister cultures were co-treated with
BDNF or testing antibodies. Following 24 hr incubation, cell viability was measured
using a CellTiter 96 Non-Radioactive Cell Proliferation Assay Kit (Promega).
Results
Evaluation of antibody responses from TrkB immunized mice
To evaluate the specific immune responses to TrkB, the five immunized mice
(M1-M5) were bled one week following the third and fourth immunizations. High
anti-TrkB antibody titers in the serum were determined by ELISA and FACS analysis
in both the third and the fourth bleeds. Fig. 1 represents results from the bleeds after
the fourth immunization. All five mice generated high titers recognizing both
rhTrkB-ECD and rmTrkB-ECD in ELISA (Figs. 1A and IB).
The location of antibody binding epitopes on hTrkB relative to the BDNF
binding site was also evaluated. Results from competition ELISA showed that the
immune bleeds can block rhTrkB-EDC-Fc and BDNF interactions in an order of M2
32

WO 2006/133164 PCT/US2006/021878
= M5 > M3 > M4 = M1 (Fig. 1C). Interestingly, the efficacies of antibodies in
blocking the rhTrkB-BDNF interaction correlated with the antibody binding titers
(compare Figs. 1A and 1C). All immune bleeds were also shown to bind to cell
surface expressed rhTrkB on HEK-293 cells as demonstrated by FACS analysis (Fig.
ID).
Following the observation of high titer specific binding of immune sera to
hTrkB, these immune bleeds were evaluated in a luciferase reporter assay to test their
agonistic activities. The luciferase activity in HEK-293 cells expressing rhTrkB has
been shown to selectively represent the activation of TrkB (Fig. 2). Incubation of
these cells with immune bleeds significantly increased the luciferase signals in a dose-
dependent manner (Fig. 3). Again, the efficacy of agonistic activities was in an order
of M2 = M5 > M3 > M4 = M1. The three mice with the highest antibody titers and
robust agonist activities (M2, M3 and M5) were chosen for subsequent hybridoma
generation.
Production of monoclonal antibodies
As described previously, the three selected mice (M2, M3 and M5) were
boosted intravenously with 10 µg of rhTrkB-ECD and 1 µg of rmTrkB-ECD three
days prior to the fusion. From the first round of hybridoma screening, ninety-four
clones were picked up which bound strongly to rhTrkB-EDC-Fc in an ELISA assay.
Twenty of these clones cross-reacted with rmTrkB-ECD in an ELISA assay. FACS
analysis confirmed that fifty-four clones bound to rhTrkB expressed on the surface of
HEK-293 cells.
The TrkB agonist activity of each clone was tested using a luciferase assay
and seventeen hybridoma clones with highest activities were selected for further
characterization. Following stabilization of these clones, three rounds of subcloning
with serial dilution, and one round of subcloning with FACS sorting, the monoclonal
antibodies from each culture conditional medium were collected and purified by using
ProSep-A (Montage Antibody Purification Kits). The IgG isotypes of each antibody
were determined by Murine Isotyping test kit (Table 1). The antibody concentrations
were determined by Murine IgG quantification ELISA and all seventeen clones
33

WO 2006/133164 PCT/US2006/021878
produced good levels of Ig in the culture supernatants.
TABLE 1

Characterization of monoclonal antibodies
The purified TrkB-specific monoclonal antibodies were characterized for their
binding activities to hTrkB by ELISA. Most of the antibodies bound to rhTrkB with
high binding affinities (ED50=10-11(M)), except for a clone 29D7, for which the ED50
dropped from 10-11(M) to 10-10(M) after purification (data not shown). Two clones
12F4 and 18C8 lost their binding activities after purification and were therefore
deselected from the priority list (and thus were not included in Table 1). Using FACS
analyses, all of the fifteen remaining TrkB binding mAbs were also shown to
specifically bind to hTrkB expressed on the surface of HEK293 cells. Antibodies
were also tested for their cross-species binding activities to rmTrkB by ELISA. While
most of the antibodies were found to bind rmTrkB weakly, clones 17D11,18C3 and
29D7 were found to bind rmTrkB with ED50= 10-10-11(M) (Table 2).
A competition ELISA assay was used to determine whether the antibodies
block the interaction of rhTrkB and BDNF. Clone 29D7 showed no blocking activity,
34

WO 2006/133164 PCT/US2006/021878
while clones 17D11,18C3 and 19E12 partially blocked rhTrkB-BDNF interaction.
All of the other clones blocked rhTrkB-BDNF interaction with IC50 = 3-5 x 10-10 (M)
(Table 2).
TABLE 2

Mapping of the relative antibody binding epitopes on rhTrkB was conducted
by examining the activity of each individual antibody in blocking the binding of other
antibodies to rhTrkB. For example, the observation that two antibodies blocked each
other's binding with rhTrkB suggests these antibodies may bind to the same epitope
or overlapping epitopes on rhTrkB (Table 3). In Table 3, the pre-bound antibodies are
presented in each row while the competing (i.e., coating) antibodies are presented in
each column. The results showed that clones 11E1,19E12 and 29D7 may recognize
unique epitopes. Clones 17D11 and 18C3 appear to compete for the same binding
site. All the remaining clones competed with each other and may share the same
binding epitope.
35

WO 2006/133164 PCT/US2006/021878


WO 2006/133164 PCT/US2006/021878
Luciferase activities
The purified TrkB-specific antibodies were examined using a luciferase assay
to demonstrate agonist activities. All of the antibodies caused dose-dependent
increases in the signal with EC50 in the range of 10-10 (M) (Fig. 4). The maximum
signal window was ~ 7 fold over basal for most of the antibodies, which was
comparable to the response induced by BDNF at 200 ng/ml (6.2 fold over basal).
Functional activities of mAbs
To test if the TrkB-binding antibodies have functional agonist activities
mediated by endogenous TrkB receptor activation, neurite promoting effects were
evaluated following treatments with these antibodies. Based on the species specific
binding characteristic of the antibodies, we first utilized human neuroblastoma SY5Y
cells, which are known to express TrkB upon neuronal differentiation. Consistent
with previous reports, we observed that the addition of BDNF promoted neurite
outgrowth as demonstrated by increases in the neurite length and the number of
branch points (Figs. 5A and 5B). Importantly, most antibodies also significantly
increased neurite outgrowth with efficacies comparable or even superior to BDNF
(Figs. 5A and 5B). Examples of representative images of a few treatment groups are
shown in Fig. 5C.
Seven TrkB antibodies with highest activities (6E2,7F5,11E1,16E11,
17D11,19E12 and 29D7) were selected for full dose-response analysis. They
induced dose-dependent increases in the neurite growth with EC50 of around 10-10 (M)
(Fig. 5D, data was not obtained for 6E2). These antibodies were also tested for their
ability to increase survival of differentiated SY5Y cells following serum withdrawal
injury. Results showed that BDNF and several antibodies protected differentiated
SY5Y cells, demonstrating dose-dependent increases in cell viability (Fig. 6).
Two TrkB antibodies, 17D11,18C3 and 29D7, which were shown to bind to
rmTrkB, were also tested in rat cerebellar granule neuron (CGN) cultures for activity
against the endogenous rat TrkB receptor. In both the neurite outgrowth assay and the
neuroprotection assay, only 29D7 showed activities comparable to those of BDNF,
while 17D11 and 18C3 were inactive (Fig. 7 and data not shown). It may suggest that
37

WO 2006/133164 PCT/US2006/021878
17D11 and 18C3 recognize an epitope on murine TrkB that contains different amino
acids in rat TrkB.
TrkB phosphorylation analysis
Seven TrkB antibodies (6E2, 7F5,11E1,16E11, 17D11, 19E12 and 29D7)
selected based on all activities described above were evaluated in Western analysis for
induction of TrkB autophosphorylation. All antibodies led to robust TrkB
phosphorylation, and these effects were antagonized by treatment with the kinase
inhibitor K252a, indicating that these TrkB-binding antibodies caused activation of
TrkB (Fig. 8).
Two TrkB antibodies, 17D11,18C3 and 29D7, which were shown to bind to
rmTrkB, were also tested in rat cerebellar granule neuron (CON) cultures for activity
against the endogenous rat TrkB receptor. In both the neurite outgrowth assay and the
neuroprotection assay, only 29D7 showed activities comparable to those of BDNF,
while 17D11 and 18C3 were inactive (Fig. 7 and data not shown). It may suggest that
17D11 and 18C3 recognize an epitope on murine TrkB that contains different amino
acids in rat TrkB.
TrkB phosphorylation analysis
Seven TrkB antibodies (6E2,7F5,11E1,16E11,17D11,19E12 and 29D7)
selected based on all activities described above were evaluated in Western analysis for
induction of TrkB autophosphorylation. All antibodies led to robust TrkB
phosphorylation, and these effects were antagonized by treatment with the kinase
inhibitor K252a, indicating that these TrkB-binding antibodies caused activation of
TrkB (Fig. 8).
ATCC deposits
The hybridomas that produced the 17D11 and 29D7 TrkB antibodies were
deposited with the ATCC on August 18,2005 and have been given ATCC patent
deposit designations PTA-6948 and PTA-6949, respectively.
38

WO 2006/133164 PCT/US2006/021878
EXAMPLE 2
This example describes further in vitro characterization and testing of a
plurality of TrkB antibodies. In particular, this example describes experiments that
were performed to assess the TrkB vs. TrkA and TrkC specificity of some of the
antibodies of Example 1.
Materials and Methods
FACS analysis
HEK-293 cells expressing human TrkA were detached from the plates with
PBS containing 5 mM EDTA and transferred in to 5 ml Falcon tubes (Becton
Dickinson, Cat. No. 352063) with 2x103 cells per tube. Cells were washed once with
PBS by centrifuging at 800 rpm at 4 °C for 3 min, and incubated for 30 min at 4 °C
with 100 µl of hybridoma culture supernatant or immune serum diluted in PBS with
1% FBS. The cells were washed 3 times with 1 ml PBS containing 1% FBS and
incubated for 30 min at 4 °C in the dark with PE labeled goat anti-murine IgG, F(ab')2
fragment (DAKO Corporation, Cat. No. R0480) in PBS containing 1% FBS. Cells
were washed three times again and re-suspended in 250 µl PBS containing 1% FBS.
Popidium iodide was used for detection of dead cells, which were excluded from
analysis. The fluorescence of 5000 cells/tube was counted by a FACScan flow
cytofluorometer (Becton Dickinson).
Luciferase assay
Stable lines of HEK-293 cells expressing human TrkA (or human TrkC) were
prepared. Transfected cells were selected in the presence of hygromycin for 2 wks
with limited dilutions. Following initial evaluation, a single clone of cells was chosen
for the study. For the luciferase assay, cells were plated at 1.5x104 cells/well in 100
µl growth medium in 96-well plates. The next day, cells were treated with 10 µl of
10x final concentration of NGF (or NT-3) or test antibodies. Luciferase activities
were measured 16 hr after the treatments using the Promega Steady-Glo assay kit
according to the manufacturer's protocol. In brief, media was replaced with 100 µl of
PBS and 100 µl of Steady-Glo reagent was added. After sealing the plates with
39

WO 2006/133164 PCT/US2006/021878
TopSeal, the plates were shaken at Titer Plate Shaker at speed ~ 5 for 5 minutes and
then luminescence was measured using a TopCount NXT v2.13 instrument (Packard).
Results
FACS analysis of monoclonal antibodies
The neat conditioned media for each of the TrkB-specific monoclonal
antibodies of Table 1 were characterized for their binding activities to human TrkA by
FACS. All of the tested antibodies failed to bind to human TrkA. Specifically, the
antibodies were tested at concentrations ranging from about 5 to about 60 µg/ml (see
specific concentrations for each antibody in Table 1, these correspond to
concentrations in the range of about 30 to about 500 nM) and each antibody failed to
show any detectable binding. The FACS data is shown in Fig. 9A.
Luciferase activities
A subset of the purified TrkB-specific antibodies of Table 2 were also
examined using a TrkA or TrkC luciferase assay to assess agonist activities. None of
the tested antibodies activated TrkA or TrkC. Specifically, the antibodies were tested
at concentrations of up to about 3 µg/ml (about 20 nM) and in each case the
antibodies failed to cause detectable increase above basal (see Fig. 9B for TrkA and
Fig. 10 for TrkC). In contrast, NGF induced a ~ 6 fold increase over basal for TrkA at
300 ng/ml (Fig. 9A) and NT-3 induced a ~ 6 fold signal increase over basal for TrkC
at 300 ng/ml (Fig. 10).
EXAMPLE 3
This example describes the in vivo testing of a plurality of TrkB antibodies in
a rodent model of neonatal hypoxia-ischemia (HI).
Materials and Methods
Animals and surgical procedures
The rodent model of neonatal hypoxia-ischemia (HI) was based on the Levine
procedure that is set forth in Fig. 11 (e.g., see Levine, Am. J. Pathol 36:1-17,1960;
40

WO 2006/133164 PCT/US2006/021878
Rice et al., Ann. Neurol, 9:131-141,1981 and Gidday et al., Neurosci. Lett. 168:221-
224,1994 each of which is incorporated herein by reference). Briefly, pups at P7
were anesthetized with 2.5% halothane and the left common carotid artery was
permanently ligated. After the incisions were sutured, pups were returned to the
home cage for recovery and feeding. Two hours later, pups were placed in individual
containers, through which humidified 8% oxygen was flowed. After 2.5 hrs of a
hypoxic ischemic period, the pups were returned to their home cages. For treatment,
animals received a 5 µl intracerebroventricular injection of 0.1 or 0.3 nmole of either
BDNF, anti-TrkB monoclonal antibody 29D7, control IgGl or vehicle just prior to the
hypoxic insult.
Assessment of brain tissue loss
At P14, brain sections were prepared from some of the mice to determine
damage caused by the HI injury. The coronal sections of the striatum, cortex and
hippocampus were stained with cresyl violet and the percent area loss in the lesioned
hemisphere was compared with the intact area. Fig. 23 compares the stains that were
obtained from the striatum and hippocampus for different treatments. Figs. 24A and
24B compare the total brain tissue loss and sub-regional tissue loss, respectively, for
different treatments. The data are shown as mean and standard error of the mean
(S.E.M.). BDNF and anti-TrkB monoclonal antibody 29D7 showed significant dose-
dependent protection of the brain from HI damage. The protection was not localized
to any specific sub-region of the brain but greatest protection was generally observed
in the cortex.
Biochemical analysis
24 hours after the HI injury, brain sections were prepared from a subset of the
mice for biochemical analysis. Hippocampal and cortical brain tissues were
dissected, lysed and subjected to several biochemical assays.
A DEVD-AMC cleavage assay was used to determine caspase-3 activities
(e.g., see Nagase et al., Immunol Lett. 84:23,2002 incorporated herein by reference).
The results for different regions of the brain are shown in Fig. 25 for mice that were
41

WO 2006/133164 PCT/US2006/021878
given 0.1 nmole of different test reagents (mean ± S.E.M.). BDNF and anti-TrkB
monoclonal antibody 29D7 blocked caspase-3 activation caused by HI injury.
Protein samples (30 µg/lane) from mice treated with 0.3 nmole of different test
reagents were separated by SDS-PAGE and subjected to immunoblotting with
antibodies against certain caspase-3 substrates (PARP and α-spectrin). Antibodies
against P-actin were used as controls to verify equal loading of proteins. The results
shown in Fig. 26 demonstrate inhibition of cleavage of these caspase-3 substrates by
BDNF and anti-TrkB monoclonal antibody 29D7 (C = contra and I = ipsi sides of
carotid ligation).
Follow up experiments
The following are prophetic follow up experiments that could be performed on
the mice of this study. The performance of the remaining mice in spatial learning and
memory tests could be assessed (e.g., without limitation the tests described in Almli et
al., Exp. Neurology 166:99-114,2000 incorporated herein by reference). For
example, the mice could be tested between P20-P30, Optionally, the performance
could be assessed over several days during that period or even at later time points.
Treatment protocols that produce a positive outcome (as measured by a
reduction in brain tissue loss, an improvement in spatial learning and memory or
otherwise) could be repeated over an even greater range of dosages. Alternatively or
additionally, treatment protocols could be repeated with different modes of
administration (e.g., intraperitoneal administration, intravenous administration, etc.);
with different start points (e.g., before ligation, immediately after hypoxia, with a
variable delay after hypoxia, etc.); and/or with different duration or frequency (e.g.,
daily treatment for 2 wks post injury, etc.).
42

WO 2006/133164 PCT/US2006/021878
EXAMPLE 4
This example describes the preparation and testing of human single-chain Fv
(scFv) antibodies against human TrkB. Human antibody phage display libraries (CS,
BMV and DP-47; Cambridge Antibody Technology) containing single-chain Fv
fragments were selected for TrkB binding using the following methods.
Materials and Methods
Selection of antibodies by panning
10 µg of hTrkB-EDC-Fc was coated on Nunc Maxisorp plate in 100 µl and
left overnight at 4 °C. A phage library was pre-blocked (50 µl phage aliquot added to
50 µl of 6% skimmed milk in 2xPBS) and negatively selected against PSGL-Fc
(Wyeth, Lot. No. 00H25M004) for 1 hr at room temperature to deplete phage reactive
against Fc. Deselected phage were transferred to the target protein-coated plate
(hTrkB-EDC-Fc) and incubated at room temperature for 2hrs. Wells were washed 10
times with PBS/0.1% Tween 20 and 5 times with PBS. Bound phage were eluted
with 50 µl/well of freshly made 100 mM TEA (140 µl of TEA in 10 ml of ultra pure
water). Eluted phage were neutralized using 25 ,ul of sterile 1 M Tris-HCl pH 7.5 .
Phage were infected into 10 ml of a mid-log (O.D at 600 nm = 0.5) of E.coli TGI
cells. Transformed cells were spread onto a 2xTYAG agar Bioassay plate and
incubated overnight at 30 °C.
Soluble phase selection (Biotin selection)
Phage antibodies were selected using biotinylated hTrkB-EDC-Fc according
to the protocol described above with the following exceptions. Pre-blocked phage
were deselected first against 100 nM biotinylated PSGL-Fc followed by positive
selection on 100 nM biotinylated hTrkB-EDC-Fc. Human TrkB specific phage were
captured using pre-blocked magnetic streptavidin beads (Dynabeads M-280
streptavidin, Cat. No. 112.06). Beads were washed 10 times with PBS/0.1% Tween
20 and 3 times with PBS. Bound phage were eluted with 200 µl of freshly made 100
mM TEA (140 µl of TEA in 10 ml of ultra pure water) and neutralized using 100 µl
of sterile 1 M Tris-HCl pH 7.5 to the eluted phage to neutralize the TEA. Phage were
43

WO 2006/133164 PCT/US2006/021878
infected into E.coli and propagated as described in the previous step.
Phage rescue
E. coli infected with phage were scrapped off the Bioassay agar plates and
mixed with 10 ml 2xTYAG per Bioassay plate (2xTY with 100µg/ml Amp and 2%
glucose). 20 ml 2xTYAG was inoculated with 100 µl of cell suspension and grown at
37 °C (300 rpm) to OD at 600 ran = 0.3-0.5, E. coli were superinfected with 3.3 ul of
MK13KO7 helper phage and incubated at 37 °C (150 rpm) for 1 hr. Superinfected
cells were re-suspended in 20 ml 2xTYAK medium (2xTY/100 µg/ml Amp/50 µg/ml
Kanamycin) and grown overnight at 25 °C at 280 rpm. E. coli were spun at 3500 rpm
for 15 min and the supernatant containing the phage were used for the next round of
selection.
Phage antibody preparation for ELISA
Single colonies of infected E. coli were picked into microtitre wells (Costat
Cellwells) containing 150 mi of 2xTYAG media (2% glucose) per well. Clones were
allowed to grow at 37 °C (100-120 rpm) for 5-6 hrs (OD at 600 nm = 0.5). M13K07
helper phage stock (1013 pfu/ml) was diluted 1:1000 with 2xTYAG medium and 20 µl
was added to each well. The wells were incubated at 37 °C (100 rpm) for 1 hr, Plates
were centrifuged at 3200 rpm for 10 min, the supernatant was removed and the cells
were resuspended in 150 µl of 2xTYAK medium. Cultures were grown overnight at
25 °C (120 rpm). The next day, plates were centrifuged at 3200 rpm for 15 min and
the supernatant was transferred to a fresh plate for the phage ELISA assay.
Phage ELISA
ELISA plates were coated with 50 ul per well of 1 (µg/ml of hTrkB-EDC-Fc
(BSA as a control) in PBS at 4 °C overnight. Wells were rinsed 3 times with PBS and
blocked with 300 µl per well of PBS/3 % skimmed milk at room temperature for 1 hr.
Phage were blocked with equal volume of PBS/6% skimmed milk and incubated at
room temperature for 1 hr. 50 µl of blocked phage were added to ELISA wells and
44

WO 2006/133164 PCT/US2006/021878
incubated at room temperature for 1 hr. Plates were washed 3 times with PBS/0.1%
Tween followed by 3 times with PBS. 50 µl of HRP-Mouse anti-M13 antibody in
PBS/3 % skimmed milk (1:5000, Amersham Pharmacia biotech, Cat. No. 27-9421-
01) was added to each well and incubated at room temperature for 1 hr. Plates were
washed as before and 50 µl TMB substrate was added to each well and developed for
2-5 min. The reaction was stopped by adding 50 µl of 0.5 M sulphuric acid and the
absorbance was read at 450 nm.
Single-chain Fv preparation for FACS
Single colonies were picked and grown in deepwell microtiter plates
containing 0.9 ml 2xTYAG media at 37 °C for 5-6 hrs. scFv expression was induced
by adding IPTG to a final concentration of 0.02 mM in 2xTY medium and growth at
30 °C overnight. E. coli were harvested after overnight growth and osmotically
shocked with 150 µl TES buffer diluted 1.'5 in water and incubated on ice for 30 min.
Cells were centrifuged and the supernatant transferred to a fresh plate for the FACS
assay. 293 cells stably transfected with TrkB were stained with scFv prepared as
described. Cells were incubated on ice for 30 min and then washed with PBS buffer.
Single-chain Fv was detected using a solution containing 9E10 anti-myc antibody
diluted 1:1000 followed by addition of PE conjugated anti mouse IgG-Fc antibody
diluted 1:500. Stained cells were analysed on a Bectin-Dickinson Flow Cytometer.
Results
PhageELISA
Table 4 shows the results from phage ELISA assays from the different phage
library selections (panning and soluble selections). Fig. 18 shows the phage binding
ELISA data from a representative set of TrkB positive clones (BMV library panning
selection). The majority of TrkB selected clones showed strong and specific
reactivity against TrkB-EDC-Fc and not to the control protein PSGL-Fc.
45

WO 2006/133164 PCT/US2006/021878
TABLE 4

Antigen (selection) Library phage ELISA
(% positives) Sequence
(% diversity)
TrkB-EDC-Fc
(panning) CS
BMV
DP-47 78
58
45 53
23
39
Biotinylated
TrkB-EDC-Fc (soluble) CS
BMV
DP-47 3
6
10 100
50
50
After two rounds of panning selection, 78% of randomly picked clones from
the CS library were positive for TrkB-EDC-Fc binding but not for PSGL-Fc binding.
Likewise, 3% of clones were positive after two rounds of the soluble selections. The
numbers for the BMV and DP-47 libraries were 58% and 45% positive clones from
panning selection and 6% and 10% from soluble selection. Using DNA sequencing,
we confirmed that 53% of the CS panning clones were unique whereas 100% of the
CS soluble selection clones were unique. The numbers for the BMV and DP-47
libraries were 23% and 39% unique from panning selections and 50% and 50%
unique from soluble selections.
FACS analysis
Fig. 19 shows the binding specificity of TrkB selected scFv antibodies tested
using the FACS assay. The data shows that TrkB selected scFv antibodies react with
membrane associated TrkB in a specific manner (filled blue histograms) with no
cross-reactivity to control cells (open green histograms).
46

WO 2006/133164 PCT/US2006/021878
EXAMPLE 5
This example compares the binding activities of the 29D7 TrkB antibody and
a 29D7 Fab fragment against human and mouse TrkB. Mouse (rhTrkB-EDC-Fc) and
human (rmTrkB-EDC) TrkB were coated on ELISA plates as described above in
Example 1,
Figs. 20A and 20B show the ELISA binding results for human and mouse
TrkB, respectively. Both the 29D7 IgG antibody and its Fab fragment have dose-
dependent binding activities to human and mouse TrkB, However, the ED50 value of
the Fab was about 100 fold lower than that of intact 27D7. Neither an isotype control
IgGl nor its Fab fragment bound to TrkB.
EXAMPLE 6
This example compares the agonistic activities of the 29D7 TrkB antibody and
the Fab fragment of Example 5 against human TrkB. Agonistic activities were
examined using a luciferase assay and HEK-293 cells which express surface rhTrkB
as described in Example 1.
HEK-293 cells were treated with 29D7 IgG or 29D7 Fab at indicated
concentrations and accumulating luciferase activities were measured 16 hours post-
treatment. As shown in Fig. 21, both total IgG and Fab of 29D7 induced dose-
dependent luciferase activities, indicating activation of TrkB. However, the EC50
value of 29D7 Fab was about 27 fold higher than that of 27D7 IgG (0.083 nM and
2.28 nM for 27D7 IgG and 29D7 Fab, respectively). Neither an isotype control IgGl
nor its Fab fragment had effects on TrkB activation.
EXAMPLE 7 .
This example describes the epitope mapping analysis of certain inventive
TrkB monoclonal antibodies. The mapping was performed against linear, single
looped and double looped peptides that were deduced from sequences within the
human TrkB extracellular domain.
As shown in Fig. 22, the 17D11 monoclonal antibody recognizes loop-3 of the
IgG-2 segment of human TrkB (KNEYGKD, SEQ ID NO:7, amino acids 364 to 370
47

WO 2006/133164 PCT/US2006/021878
of SEQ ID NO:1); and loop-1 of the IgG-2 segment of human TrkB (KGNPKP, SEQ
ID NO:8, amino acids 308 to 313 of SEQ ID NO.l). The mouse sequence for the
fanner epitope (KNEYGKD, SEQ ID NO:7, amino acids 364 to 370 of SEQ ID
NO:2) is identical to the human sequence. The mouse sequence for the latter epitope
(RGNPKP, SEQ ID NO:9, amino acids 308 to 313 of SEQ ID NO:2) differs by one
amino acid.
As shown in Fig. 22, the 29D7,7F5,11E1 and 19E12 monoclonal antibodies
all recognize loop-3 of the IgG-1 segment of human TrkB (ENLVGED, SEQ ID
NO;10, amino acids 269 to 275 of SEQ ID NO:1); and may also recognize loop-1 of
the IgG-1 segment of human TrkB (AGDPVP, SEQ ID NO:11, amino acids 221 to
226 of SEQ ID NO: 1). The mouse sequence for the former epitope (ENLVGED,
SEQ ID NO: 10, amino acids 269 to 275 of SEQ ID NO:2) is identical to the human
sequence. The mouse sequence for the latter epitope (GGDPLP, SEQ ID NO: 12,
amino acids 221 to 226 of SEQ ID NO:2) differs by two amino acids.
EXAMPLE 8
This example describes in vivo TrkB activation experiments that were
performed with the anti-TrkB antibody 29D7. Briefly, pups at P7 received a 5 µl
intracerebroventricular injection of 0.3 nmole of either anti-TrkB monoclonal
antibody 29D7 or vehicle. Brain tissues were then dissected 1,2,6,12 and 24 hours
after injection. Tissues were lysed and equal amounts of protein samples (30 fig/lane)
were separated by SDS-PAGE and subjected to immunoblotting with antibodies
specific to proteins that are phosphorylated as a result of TrkB activation (phospho-
ERK1/2 and phospho-AKT). Antibodies against P-actin were used as controls to
verify equal loading of proteins. The results of Fig. 27 show that anti-TrkB
monoclonal antibody 29D7 induced time-dependent phosphorylation of ERK1/2 and
AKT (data from hippocampal and cortical samples are shown and are representative
of results obtained from four other samples).
Fig. 28 is a densitometric quantification of the ERK phosphorylation shown in
Fig. 27. Fig. 29 is a densitometric quantification of the AKT phosphorylation shown
in Fig. 27. Fig. 30 shows vehicle and 29D7 treated brain tissues that have been fixed
48

WO 2006/133164 PCT/US2006/021878
and immunoflourescently labeled with an antibody for the neuron-specific marker
NeuN (left) and an anti-phospho-ERK1/2 antibody (center). The right hand figures
show these two merged. ERK1/2 phosphorylation is significantly stronger in the
29D7 treated samples. Fig. 31 shows the time-course of ERK1/2 activation in the
cortical and hippocampal tissues following intracerebroventricular injection of anti-
TrkB monoclonal antibody 29D7.
EXAMPLE 9
This example describes a MAG/myelin-induced neurite inhibition assay that
was performed with the anti-TrkB antibody 29D7. Briefly, aliquots of 50 µl
recombinant rat MAG(l-5) or purified rat myelin were added to 96 well flat bottom
tissue culture plates and air-dried overnight at room temperature. The total amounts
of MAG(l-5) or myelin were 0.25-0.5 µg per well, unless indicated differently. Next
day, the MAG- or myelin-coated plates were treated with 50 µl of poly-D-lysine (17
µg/ml) for 1.5 hours, followed by incubation with a media containing 10 % FBS for 1
hour at 37 °C, After aspiration of the media, rat CGN cells were plated at a density of
8,000 cells/well in B27-supplemented Neurobasai growth media containing 200 mM
L-Glutamine, 2 M KC1, and 100 U/ml penicillin and streptomycin. When indicated,
treatment reagents were added at the time of cell plating. Neurons were grown in a 37
°C incubator equilibrated with 5 % CO2- Approximately 20 hours post plating, cells
were fixed with 4 % paraformaldehyde and proceeded for Tuj 1 staining. Cells were
permeabilized with 0.2 % Triton X/PBS (TPBS) for 5 minutes at room temperature
followed by incubation with 1.5 % normal goat serum in TPBS (S-TPBS) for 30
minutes to block non-specific binding. An aliquot of anti-Neuronal Class III b-
Tubulin monoclonal antibody (Tujl, 1:1000; Covance # MMS-435P) was added to
the cells. After 1 hour incubation at room temperature, unbound antibodies were
washed with PBS three times and an aliquot of Alexa Fluor 488 mouse anti-goat IgG
antibody (1:500, Molecular Probe # A-l 1001) was added to visualize the signals.
Hoechst 33342 (Molecular Probe # H-3570,2 µg/ml) was included to label the
nucleus. After washing, plates were sealed and analyzed for neurite growth using a
Cellomics array scan. Typically around 300 cells from 9 fields were analyzed per
49

WO 2006/133164 PCT/US2006/021878
well and each treatment was conducted in quadruplets.
MAG and myelin inhibit neurite outgrowth of primary cerebellar granule
neurons in cultures. Increasing concentrations of recombinant rat MAG(I-5) or
purified rat myelin were coated on 96 well tissue culture plates and the neurite
extension of primary neurons were measured at 20 hours as described above. As
shown in Figure 32, (A) MAG and (B) myelin resulted in dose-dependent inhibition
of neurite outgrowth.
CGN neurons were then plated on either MAG or myelin-coated plates with
control or with a range of concentrations of the TrkB antibody 29D7. Neurite
extension was measured 20 hours post-treatment. As shown in Figure 33, 29D7 led to
reversal of (A) MAG- and (B) myelin-mediated neurite inhibition.
OTHER EMBODIMENTS
Other embodiments of the invention will be apparent to those skilled in the art
from a consideration of the specification or practice of the invention disclosed herein.
It is intended that the specification and examples be considered as exemplary only,
with the true scope of the invention being indicated by the following claims.
50

WO 2006/133164 PCT/US2006/021878
CLAIMS
We claim:
1. A monoclonal antibody that binds human TrkB.
2. The monoclonal antibody of claim 1, wherein the binding with human TrkB
has an ED50 in the range of about 10 pM to about 500 nM.
3. The monoclonal antibody of claim 1, wherein the binding with human TrkB
has an ED50 in the range of about 10 pM to about 1 nM.
4. The monoclonal antibody of claim 1, wherein the binding with human TrkB
has an ED50 in the range of about 10 pM to about 100 pM.
5. The monoclonal antibody of claim 1, wherein the binding with human TrkB
has an ED50 in the range of about 10 pM to about 50 pM.
6. The monoclonal antibody of claim 1, wherein the monoclonal antibody
activates human TrkB with an EC50 in the range of about 10 pM to about 500 nM.
7. The monoclonal antibody of claim 1, wherein the monoclonal antibody
activates human TrkB with an EC50 in the range of about 10 pM to about 1 nM.
8. The monoclonal antibody of claim 1, wherein the monoclonal antibody
activates human TrkB with an EC50 in the range of about 10 pM to about 100 pM.
9. The monoclonal antibody of claim 1, wherein the monoclonal antibody
activates human TrkB with an EC50 in the range of about 10 pM to about 50 pM.
10. The monoclonal antibody of claim 1, wherein the monoclonal antibody does
51

WO 2006/133164 PCT/US2006/021878
not bind and/or does not activate human TrkA.
11. The monoclonal antibody of claim 10, wherein the monoclonal antibody
exhibits no detectable binding with human TrkA and/or does not cause any detectable
activation of human TrkA at antibody concentrations above about 1 nM.
12. The monoclonal antibody of claim 10, wherein the monoclonal antibody
exhibits no detectable binding with human TrkA and/or does not cause any detectable
activation of human TrkA at antibody concentrations above about 10 nM.
13. The monoclonal antibody of claim 10, wherein the monoclonal antibody
exhibits no detectable binding with human TrkA and/or does not cause any detectable
activation of human TrkA at antibody concentrations above about 100 nM.
14. The monoclonal antibody of claim 10, wherein the monoclonal antibody
exhibits no detectable binding with human TrkA and/or does not cause any detectable
activation of human TrkA at antibody concentrations above about I µM.
15. The monoclonal antibody of claim 1, wherein the monoclonal antibody does
not bind and/or does not activate human TrkC.
16. The monoclonal antibody of claim 15, wherein the monoclonal antibody
exhibits no detectable binding with human TrkC and/or does not cause any detectable
activation of human TrkC at antibody concentrations above about 1 nM.
17. The monoclonal antibody of claim 15, wherein the monoclonal antibody
exhibits no detectable binding with human TrkC and/or does not cause any detectable
activation of human TrkC at antibody concentrations above about 10 nM.
18. The monoclonal antibody of claim 15, wherein the monoclonal antibody
exhibits no detectable binding with human TrkC and/or does not cause any detectable
52

WO 2006/133164 PCT/US2006/021878
activation of human TrkC at antibody concentrations above about 100 nM.
19. The monoclonal antibody of claim 15, wherein the monoclonal antibody
exhibits no detectable binding with human TrkC and/or does not cause any detectable
activation of human TrkC at antibody concentrations above about 1 µM.
20. The monoclonal antibody of claim 10, wherein the monoclonal antibody does
not bind and/or does not activate human TrkC.
21. The monoclonal antibody of claim 1, wherein the monoclonal antibody blocks
the binding between BDNF and human TrkB.
22. The monoclonal antibody of claim 21, wherein the monoclonal antibody
blocks the binding between BDNF and human TrkB with an IC50 in the range of about
100 pM to about 500 nM.
23. The monoclonal antibody of claim 21, wherein the monoclonal antibody
blocks the binding between BDNF and human TrkB with an IC50 in the range of about
100 pM to about 1 nM.
24. The monoclonal antibody of claim 21, wherein the monoclonal antibody
blocks the binding between BDNF and human TrkB with an IC50 in the range of about
100 pM to about 500 pM.
25. The monoclonal antibody of claim 1, wherein the monoclonal antibody does
not block the binding between BDNF and human TrkB.
26. The monoclonal antibody of claim 25, wherein the monoclonal antibody
exhibits no detectable blocking activity at antibody concentrations above about 1 nM.
27. The monoclonal antibody of claim 25, wherein the monoclonal antibody
53

WO 2006/133164 PCT/US2006/021878
exhibits no detectable blocking activity at antibody concentrations above about 10
nM.
28. The monoclonal antibody of claim 25, wherein the monoclonal antibody
exhibits no detectable blocking activity at antibody concentrations above about 100
nM.
29. The monoclonal antibody of claim 25, wherein the monoclonal antibody
exhibits no detectable blocking activity at antibody concentrations above about 1 uM.
30. The monoclonal antibody of claim 1, wherein the monoclonal antibody has an
IgGl isotype.
31. The monoclonal antibody of claim 1, wherein the monoclonal antibody has an
IgG2a isotype.
32. The monoclonal antibody of claim 1, wherein the monoclonal antibody has an
IgG2b isotype.
33. The monoclonal antibody of any one of claims 1-32, wherein the monoclonal
antibody binds and/or activates murine TrkB.
34. The monoclonal antibody of claim 33, wherein the binding with murine TrkB
has an ED50 in the range of about 10 pM to about 500 nM.
3 5. The monoclonal antibody of claim 33, wherein the binding with murine TrkB
has an ED50 in the range of about 10 pM to about 1 nM.
36. The monoclonal antibody of claim 33, wherein the binding with murine TrkB
has an ED50 in the range of about 10 pM to about 500 pM.
54

WO 2006/133164 PCT/US2006/021878
37. The monoclonal antibody of claim 33, wherein the binding with murine TrkB
has an ED50 in the range of about 10 pM to about 100 pM.
3 8. The monoclonal antibody of claim 1, wherein the monoclonal antibody binds
one or both of the epitopes of human TrkB with sequences KNEYGKD (SEQ ID
NO:7) and KGNPKP (SEQ ID NO:8).
39. The monoclonal antibody of claim 38, wherein the monoclonal antibody
further binds one or both of the epitopes of mouse TrkB with sequences KNEYGKD
(SEQ ID NO.7) and RGNPKP (SEQ ID NO;9).
40. The monoclonal antibody of claim 1, wherein the monoclonal antibody binds
an epitope of human TrkB with the sequence ENLVGED (SEQ ID NO:10) and
optionally an epitope of human TrkB with the sequence AGDPVP (SEQ ID NO:11).
41. The monoclonal antibody of claim 40, wherein the monoclonal antibody
further binds an epitope of mouse TrkB with sequence ENLVGED (SEQ ID NO: 10).
42. A monoclonal antibody that is produced from the hybridoma deposited with
the ATCC on August 18,2005 and having ATCC patent deposit designation PTA-
6948 or from a progenitor cell thereof.
43. A monoclonal antibody that is produced from the hybridoma deposited with
the ATCC on August 18, 2005 and having ATCC patent deposit designation PTA-
6949 or from a progenitor cell thereof.
44. The monoclonal antibody of claim 1, wherein the monoclonal antibody binds
to the same epitope as the monoclonal antibody of claim 42.
45. The monoclonal antibody of claim 1, wherein the monoclonal antibody binds
to the same epitope as the monoclonal antibody of claim 43.
55

WO 2006/133164 PCT/US2006/021878
46. A hybridoma that produces a monoclonal antibody according to any one of
claims 1-41.
47. A hybridoma deposited with the ATCC on August 18,2005 and having ATCC
patent deposit designation PTA-6948 or a progenitor cell thereof.
48. A hybridoma deposited with the ATCC on August 18,2005 and having ATCC
patent deposit designation PTA-6949 or a progenitor cell thereof.
49. A method comprising:
immunizing an animal with at least two protein immunogens that were derived
from TrkB proteins of different species; and
fusing spleen cells from the immunized animal with myeloma cells to produce
a hybridoma.
50. The method of claim 49, wherein the at least two protein immunogens include
a first immunogen that was derived from human TrkB and a second immunogen that
was derived from a non-human TrkB.
51. The method of claim 50, wherein the first immunogen includes the
extracellular domain of human TrkB and the second immunogen includes the
extracellular domain of a non-human TrkB.
52. The method of claim 51, wherein the second immunogen includes the
extracellular domain of a non-human TrkB selected from the group consisting of
murine TrkB, rat TrkB and chicken TrkB.
53. The method of claim 52, wherein the second immunogen includes the
extracellular domain of murine TrkB.
56

WO 2006/133164 PCT/US2006/021878
54. The method of claim 50, wherein the first immunogen and the second
immunogen are, administered in the immunizing step in amounts (by weight) that have
a ratio that is greater than about 1.
55. The method of claim 54, wherein the ratio is greater than about 5.
56. The method of claim 55, wherein the ratio is about 10.
57. The method of claim 51, wherein the first and second imrmmogens do not
include any of the amino acids that are found outside of the extracellular domain of
TrkB.
58. A hybridoma that was made according to the method of any one of claims 49-
57.

59. The method of any one of claims 49-57 further comprising:
generating a monoclonal antibody from the hybridoma.
60. A monoclonal antibody that was made according to the method of claim 59.
61. A method of activating human TrkB in an individual, comprising:
administering to an individual, a therapeutically effective amount of a
monoclonal antibody of any one of claims 1-41.
62. The method of claim 61, wherein said individual is suffering from a
neurological condition which requires activation of TrkB.
63. The method of claim 61, wherein the nervous system of said individual has
been injured by a wound, surgery, ischemia, infection, a metabolic disease,
malnutrition, a malignant tumor or a toxic drug.
57

WO 2006/133164 PCT/US2006/021878
64. The method of claim 61, wherein said individual has suffered a stroke, a spinal
cord injury, or an axotomy.
65. The method of claim 61, wherein said individual is suffering from a congenital
or neurodegenerative condition.
66. The method of claim 65, wherein said condition is selected from the group
consisting of Alzheimer's disease, Parkinson's disease, Huntington's chorea, and
amyotrophic lateral sclerosis (ALS).
67. The method of claim 61, wherein said monoclonal antibody is administered to
said individual parenterally.
68. The method of claim 61, wherein said monoclonal antibody is administered to
said individual intravenously or interperitoneally.
69. A method comprising:
combining a monoclonal antibody of any one of claims 1-41 with a sample
that includes an amount of human TrkB under conditions that allow specific binding
between the monoclonal antibody and human TrkB; and
detecting the specific binding, thereby indicating the presence of human TrkB
in the sample.
70. A method comprising:
combining a monoclonal antibody of any one of claims 1-41 with a sample
that includes an amount of human TrkB under conditions that allow specific binding
between the monoclonal antibody and human TrkB, thereby producing an antibody-
TrkB complex;
separating the antibody-human TrkB complex from the remainder of the
sample; and
separating the antibody from human TrkB, thereby obtaining purified human
TrkB.
58

The present invention provides monoclonal antibodies for human TrkB. In certain embodiments the inventive antibodies bind and activate human TrkB. In certain embodiments the inventive antibodies are selective for human TrkB in that they do not bind (or activate) human TrkA or human TrkC. In some embodiments the inventive monoclonal antibodies cross-react with murine TrkB. Humanized or veneered versions of the inventive antibodies are also encompassed. Pharmaceutical compositions that
comprise inventive antibodies are provided as are methods for preparing the inventive antibodies and methods of using these for
treatment, detection or purification purposes.