U.S. patent application number 10/310674 was filed with the patent office on 2003-09-04 for peptide or protein containing a c '-d loop of the cd28 receptor family.
Invention is credited to Hanke, Thomas, Hunig, Thomas, Luhder, Fred.
Application Number | 20030166860 10/310674 |
Document ID | / |
Family ID | 26010744 |
Filed Date | 2003-09-04 |
United States Patent
Application |
20030166860 |
Kind Code |
A1 |
Hunig, Thomas ; et
al. |
September 4, 2003 |
Peptide or protein containing a C '-D loop of the CD28 receptor
family
Abstract
The invention relates to a protein or peptide comprising the
C'-D loop of a member of the CD28 family, uses thereof and mAbs
obtainable therefrom.
Inventors: |
Hunig, Thomas; (Wuerzburg,
DE) ; Luhder, Fred; (Wuerzburg, DE) ; Hanke,
Thomas; (Wuerzburg, DE) |
Correspondence
Address: |
Mark D. Wieczorek, Ph.D., Esq.
Post Office Box 70072
San Diego
CA
92167
US
|
Family ID: |
26010744 |
Appl. No.: |
10/310674 |
Filed: |
December 4, 2002 |
Current U.S.
Class: |
530/350 |
Current CPC
Class: |
A61P 37/04 20180101;
C07K 14/70521 20130101 |
Class at
Publication: |
530/350 |
International
Class: |
C07K 014/74 |
Foreign Application Data
Date |
Code |
Application Number |
Dec 4, 2001 |
DE |
101 60 516 |
Jan 10, 2002 |
DE |
102 00 714 |
Claims
1. A protein or peptide comprising the C'-D loop of a member of the
CD28 family or containing a peptide analogous thereto or containing
a mimicry compound thereto, not however a member of the CD28
family.
2. A peptide according to claim 1, wherein the ends thereof are
bound to one binding position each of a substrate, wherein the
binding positions of the substrate are spatially arranged with
regard to each other according to the binding positions for the
C'-D loop in CD28, wherein the C'-D loop or the peptide being
analogous thereto is fixed in a three-dimensional configuration
according to the C'-D loop in CD28, the bound C'-D loop or the
peptide being analogous thereto being freely accessible for
antibodies, and wherein the substrate is not a member of the CD28
family without a C'-D loop or a natural peptide being analogous
thereto of the respective CD28 family member.
3. A peptide or protein according to claim 1 or 2, comprising an
amino acid sequence seq.-ID 41, seq.-ID 1 or 5-10, not however
human CD28, seq.-ID 42, seq.-ID 2 or 12-17, not however human
CTLA-4, seq.-ID 43, seq.-ID 3 or 19-24, not however human ICOS, or
seq.-ID 44, seq.-ID 4 or 26-31, not however human PD-1.
4. A peptide or protein or mimicry compound according to one of
claims 1 to 3, obtainable by that one or more prospective proteins
or peptides are subjected to a binding test with one of the
monoclonal antibodies (mAbs) 9D7 or 5.11A, and binding peptides or
proteins are selected.
5. A nucleic acid coding for a peptide according to one of claims 1
to 4 or for a protein containing this peptide, not however coding
for a member of the CD28 family.
6. A vector comprising a nucleic acid according to claim 5 operably
linked to a suitable promotor for expression in a target cell line
being transfected with the vector.
7. Use of a peptide or protein or mimicry compound thereto
according to one of claims 1 to 4 for producing a pharmaceutical
composition for the modulation of the T cell proliferation.
8. Use of a peptide according to one of claims 1 to 4 or of a
protein containing this peptide or of a mimicry compound thereto,
optionally lacking the binding site for costimulatory mAbs, in a
method for producing mAbs which superagonistically modulate the
proliferation of T cells of several to all sub-groups, wherein a
non-human mammal is immunized with the protein or peptide or the
mimicry compound thereto, wherein from the non-human mammal cells
are taken, and hybridoma cells are produced from the cells, and
wherein the thus obtained hybridoma cells are selected such that in
their culture supernatant mAbs are contained which bind to the C'-D
loop of the peptide or protein or to the mimicry compound
thereto.
9. Use of a peptide according to one of claims 1 to 4 or of a
protein containing this peptide or of a mimicry compound thereto,
optionally lacking the binding site for costimulatory mAbs, in a
screening method for the identification of substances
superagonistically modulating the proliferation of T cells of
several to all sub-groups, wherein a prospective substance or a
mixture of prospective substances is subjected to a binding assay
with the peptide or protein or mimicry compound, and wherein
substances binding to the peptide or protein or mimicry compound
are selected, in particular mAbs and/or mimicry compounds.
10. Hybridoma cells producing mAbs binding to a peptide or protein
according to one of claims 1 to 4, in particular as filed under the
DSM numbers DSM ACC2531 (9D7 or 9D7G3H11) or DSM ACC2530 (5.11A or
5.11A1C2H3).
11. mAbs obtainable from hybridoma cells according to claim 10 or
mAbs which are coded at least partially by one or more of the
sequences seq.-ID 33, 35, 37 and/or 39, or mAbs which contain or
consist of one or more sequences seq.-ID 34, 36, 38 and/or 40 or of
FIG. 10 or sequences being homologous thereto.
12. Use of a mAb according to claim 11 or of a mimicry compound
thereto or of a substance obtainable from a screening method
according to claim 9 for producing a pharmaceutical composition for
the treatment of diseases with pathologically reduced CD4 T cell
counts, in particular AIDS.
13. Use of a mAb according to claim 11 or of a mimicry compound
thereto or of a substance obtainable from a screening method
according to claim 9 for producing a pharmaceutical composition for
the treatment following stem cell transplantations after chemo or
radio therapy of leukemic diseases.
14. Use of a mAb according to claim 11 or of a mimicry compound
thereto or of a substance obtainable from a screening method
according to claim 9 for producing a pharmaceutical composition for
multiplying and/or qualitatively influencing immune reactions after
vaccinations.
15. Use of a mAb according to claim 11 or of a mimicry compound
thereto or of a substance obtainable from a screening method
according to claim 9 for producing a pharmaceutical composition for
the treatment of autoimmune-inflammatory diseases.
Description
FIELD OF THE INVENTION
[0001] The invention relates to a protein or peptide containing a
partial sequence of a member of the CD28 receptor family, a nucleic
acid coding for such a peptide, a plasmid containing such a nucleic
acid, hybridoma cells forming monoclonal antibodies (mAbs) binding
to such a peptide, mAbs obtainable from such hybridoma cells, and
methods of use of the peptide and the mAbs.
DEFINITIONS
[0002] Monoclonal antibodies are antibodies being produced by
hybrid cell lines (so-called hybridomas) typically resulting from
the fusion of a B cell of animal or human origin producing
antibodies with a suitable myeloma tumor cell.
[0003] The amino acid sequence of human CD28 is known under the
accession no. NM.sub.--006139.
[0004] The amino acid sequence of human CTLA-4 is known under the
accession no. L15006.
[0005] The amino acid sequence of human ICOS is known under the
accession no. AJ277832.
[0006] The amino acid sequence of human PD-1 is known under the
accession no. U64863.
[0007] The C'-D loop of CD28 comprises the amino acids 52 to 66 of
the above CD28 sequence (numbering according to FIG. 7, see also
Ostrov, D. A. et al.; Science (2000), 290:816-819). The term C'-D
loop will in the following also comprise any partial sequences
thereof.
[0008] A loop or a binding site arranged therein is freely
accessible, if there is for a defined binding partner no steric
hindrance for the binding site in the loop by the sequences or
molecules outside of the loop.
[0009] Activation of T lymphocytes is the increase of metabolic
activity, increase of the cell volume, synthesis of immunologically
important molecules and initiation of the cell division
(proliferation) of T lymphocytes as a response to an external
stimulation. Inhibition is the opposite process. For example are
such processes caused by occupation of the CD28 molecule on T cells
by special CD28-specific monoclonal antibodies. The activation of
the T lymphocytes with the described side effects is part of the
physiologic immune reaction, in pathologic situations however there
may be lost control thereof (lympho-proliferative diseases), or may
be insufficient (immunodeficiency).
[0010] Modulation of the proliferation of T cells is either the
increase of the activity (for a pathologically insufficient
activation) or reduction or inhibition of the activity (for
pathologically lympho-proliferative diseases).
[0011] Several sub-groups of the T cells means at least sub-groups
of CD4 and CD8 T cells expressing CD 28.
[0012] An analogous peptide is a peptide the amino acid sequence of
which differs from that one of the peptide to which it is
analogous, which binds however a defined binding partner with at
least the same affinity. Deviations in the sequence may be
deletions, substitutions, insertions and elongations. An analogous
peptide will usually comprise a tertiary (partial) structure and/or
exposition being very similar to the peptide, in a (cell surface)
protein, and otherwise only needs to comprise or form a binding
site for the defined binding partner in the section analogous to
the immediate binding section of the peptide.
[0013] A mimicry compound of a mAb is a natural or synthetic
chemical structure behaving in a binding assay as a defined mAb
mimicrying the mimicry compound.
[0014] A mimicry compound of a C'-D loop is a natural or synthetic
chemical structure to which specifically bind mAbs being
superagonistic and specific for a member of the CD28 family.
[0015] The term mAbs comprises, in addition to structures of the
usual Fab/Fc constructions, also structures consisting of or
comprising the Fab fragment only. It is also possible to
exclusively use the variable region, the fragment of the heavy
chains being connected in a suitable manner, for instance also by
means of synthetic bridge molecules, with the fragment of the light
chain, in such a way that the binding regions of the chains form
the antibody binding site. The term antibody also comprises
(complete) chimeric and humanized antibodies.
[0016] Superagonistic modulation of the proliferation of T cells
means that no costimulation, i.e. no further binding event in
addition to a binding of a mAb or of a mimicry compound to a member
of the CD28 family is required for the stimulation or inhibition of
the proliferation.
[0017] A screening method comprises the use of a target, for
instance a partial sequence from CD28, one or more known or unknown
substances being contacted with the target and a binding event
being detected or not detected. In the case of the detection of a
binding event, the substance is selected. In the case of the use of
a mixture of substances, typically a deconvolution follows a
selection of the mixture for the purpose of the determination of
binding components in the selected mixture.
[0018] As a CD28 family is designated a group of T cell surface
receptors having an immuno-regulatory activity. This may be either
stimulating, as in the case of the CD28, or inhibiting, as in the
case of the CTLA-4. To the CD28 family belong CD28, CTLA-4, PD-1
and ICOS.
[0019] A substrate can be soluble, insoluble and/or immobilized. A
substrate can be formed of any natural or synthetic molecules, for
instance of amino acid chains, among others. In this respect a
protein or a peptide in the terminology of this text needs not
necessarily be a protein or a peptide according to the conventional
definition. Usually, however, a protein or peptide according to the
terminology used herein is also a protein or peptide in the usual
terminology.
BACKGROUND OF THE INVENTION AND PRIOR ART
[0020] For the understanding of the invention, first the following
technological background is important. The activation of resting T
cells for the proliferation and functional differentiation first
requires the occupation of two surface structures, so-called
receptors: i.e. of the antigen receptor having a different
specificity from cell to cell and being necessary for the detection
of antigens, for instance viral fission products; and of the CD28
molecule expressed in an identical manner on all resting T cells
with the exception of one sub-group of the human CD8 T cells, said
CD28 molecule binding in to ligands on the surface of other cells.
This is called the costimulation of the antigen-specific immune
reaction by CD28. In cell culture, these processes can be simulated
by occupying the antigen receptor and the CD28 molecule by suitable
mAbs. In the classical system of the costimulation, neither the
occupation of the antigen receptor nor of the CD28 molecule alone
will lead to the T cell proliferation, the occupation of both
receptors is however effective. This observation has been made for
T cells of man, mouse and rat.
[0021] There are known, however, also mAbs that can alone initiate
the T cell proliferation. Such a superagonistic (that is
independent from the occupation of the antigen receptor) activation
of resting T cells has been observed in the following systems: in
the document Brinkmann et al., J. Immunology, 1996, 156: 4100-4106,
it has been shown that a very small fraction (5%) of human T
lymphocytes carrying the surface marker CD45 RO being typical for
resting T lymphocytes, is activated by the CD28-specific mAb 9.3
normally requiring costimulation with addition of the growth factor
interleukin-2 (IL-2) without occupation of the antigen receptor. In
the document of Siefken et al., Cellular Immunology, 1997, 176:
59-65, it has been shown that a CD28-specific mAb produced in a
conventional manner, i.e. by immunization of mice with human cells,
can activate in cell culture a sub-group of human T cells without
occupation of the antigen receptor for the proliferation, if CD28
is occupied by this mAb and the cell-bound mAbs are in addition
cross-linked with each other by further antibodies. It is common to
the in this respect known antibodies that only a small fraction of
the T cells can be activated.
[0022] In the document of Tacke et al., Eur. J. Immulog., 1997, 17:
239-247, two kinds of CD28-specific monoclonal antibodies with
different functional properties have been described: costimulatory
mAbs costimulating the activation of resting T cells with a
simultaneous occupation of the antigen receptor only; and
superagonistic mAbs being able to activate T lymphocytes of all
classes in vitro and in animal experients for the proliferation,
without occupation of the antigen receptor. Both in this respect
known mAbs originate from an immunization with cells, on the
surface of which rat CD28 is expressed, and which are obtainable by
different selections directed to their respective properties.
Finally, from WO 98/54225 is known another superagonistic mAb,
namely CMY-2.
[0023] The in this respect known superagonistic mAbs do not meet in
their stimulatory effect the requirements with regard to the
strength of the activating effect or the width of the activated
sub-populations of T lymphocytes or do not have the required human
specificity.
TECHNICAL OBJECT OF THE INVENTION
[0024] The invention is based therefore on the technical object to
provide means, by use of which superagonistic compounds can be
found which bind to one or several members of the CD28 family and
have an improved stimulatory or inhibiting effect, as well as to
specify such compounds.
[0025] The Findings the Invention is Based On.
[0026] The invention is based on the examination of the binding
regions of superagonistic mAbs at CD28 as well as the interaction
found in these experiments of the C'-D loop of CD28 with
superagonistic mAbs. Further, the invention is based on the finding
that a corresponding binding region for superagonistic mAbs can be
found in the other members of the CD28 family, namely there, too,
the C'-D loops. From this basic findings, various aspects for
technical teachings of the invention can be deducted.
[0027] Basics of the Invention and Preferred Embodiments.
[0028] For achieving the technical object, the invention teaches a
protein or peptide comprising the C'-D loop of a member of the CD28
family, or comprising a peptide being analogous thereto or
comprising a mimicry compound thereto, not however a member of the
CD28 family. In other words, the essential element of a protein or
peptide according to the invention is the C'-D structure (or of an
analogous/mimicry substance thereto), and that irrespective of
whether and which sequences follow on both sides of the loop. It is
only essential that the loop structure is sufficiently exposed, in
order to offer access for superagonistic mAbs or mimicry compounds
and to prevent in the case of the specific binding possibility a
binding not for steric reasons. According to the invention, it is
surprising that all found superagonistic CD28-specific mAbs bind to
the C'-D loop, whereas the not superagonistic CD28-specific mAbs do
not bind thereto. By a single group of target structures with
regard to their spatial distribution on members of the CD28 family,
thus prospective effective substances against a number of different
diseases are made accessible.
[0029] In the embodiment of a peptide, in particular of an
oligopeptide (4 to 9 amino acids) or a polypeptide (10-100 amino
acids) or of a mimicry compound thereto, it is preferred that the
ends thereof are each bound to a binding position of a substrate,
the binding positions of the substrate being spatially arranged
with regard to each other according to the binding positions for
the C'-D loop, the C'-D loop or the peptide being analogous thereto
or the mimicry compound thereto being fixed in a three-dimensional
configuration according to the C'-D loop, the bound C'-D loop or
the peptide being analogous thereto or the mimicry compound thereto
being freely accessible for antibodies or mimicry compounds
thereto, and the substrate not being a member of the CD28 family
without a C'-D loop. Thus a three-dimensional structure permitting
a binding with superagonistic substances is provided.
[0030] In detail, a peptide or protein according to the invention
may comprise an amino acid sequence seq.-ID 41 (human CD28 loop),
not however be human CD28, seq.-ID 42 (human CTLA-4 loop), not
however be human CTLA-4, seq.-ID 43 (human ICOS loop), not however
be human ICOS, or seq.-ID 44 (human PD-1 loop), not however be
human PD-1. One or two amino acids may be added according to FIG. 7
to the 3' end and/or the 5' end. But partial sequences thereof may
also be comprised in the peptides according to the invention, for
instance according to the sequences seq.-ID 1 to 4, respectively.
The seq.-ID 5 to 10 indicate variants of the human CD28 loop. The
seq.-ID 12 to 17 indicate variants of the human CTLA-4 loop. The
seq.-ID 19 to 24 indicate variants of the human ICOS loop. The
seq.-ID 26 to 31 indicate variants of the human PD-1 loop. One or
more amino acids of the sequence 11 may be added according to FIG.
7a to one of the sequences 1, 7 or 9. One or more amino acids of
the sequence 18 may be added according to FIG. 7a to one of the
sequences 2, 14 or 16. One or more amino acids of the sequence 25
may be added according to FIG. 7a to one of the sequences 3, 21 or
23. One or more amino acids of the sequence 32 may be added
according to FIG. 7a to one of the sequences 4, 28 or 30. The above
sequences are sections according to the invention, to which
superagonistic mAbs will specifically bind. It can in particular be
seen, when comparing the sequences, that the primary structure of
the loop is specific for the respective family members. By
selection of the C'-D loop of a specific member and thus by
application of substances having specificity for this selected
loop, thus alternatively an activation or an inhibition of the
proliferation can be obtained.
[0031] A (CD28-specific) protein or peptide according to the
invention or a mimicry compound thereto can be identified by that
one or more prospective proteins, peptides or mimicry compounds are
subjected to a binding test with e.g. one of the mAbs 9D7 or 5.11A,
and binding peptides are selected. The mentioned mAbs are new
superagonistic CD28-specific mAbs, which are described in detail in
the experimental section hereof. By means of corresponding mAbs
with specificity for a C'-D loop of another CD28 family member,
proteins, peptides or mimicry compounds according to the invention
and being specific for the other members can be identified. Such
corresponding mAbs may be obtained in an analogous manner.
[0032] The invention further relates to a nucleic acid coding for a
peptide according to the invention or for a protein comprising such
a peptide, not however coding for a member of the CD28 family, and
to a vector, e.g. plasmid, comprising such a nucleic acid, operably
linked to a suitable promotor.
[0033] The peptide, protein according to the invention or a mimicry
compound thereto according to the invention can be used in a method
for producing mAbs which superagonistically modulate the
proliferation of T cells of several to all sub-groups, a non-human
mammal being immunized with the protein or peptide or the mimicry
compound thereto, from the non-human mammal cells being taken,
hybridoma cells being produced from the cells, and such obtained
hybridoma cells being selected, the culture supernatant of which
contains mAbs, which bind to the C'-D loop of the protein or
peptide or the mimicry compound thereto, such hybridoma cells and
mAbs obtainable with such hybridoma cells. Human mAbs according to
the invention can alternatively however also be produced by that B
lymphocytes are selected which bind to the loop, and that their
expressed immunoglobulin genes are cloned. Furthermore, human mAbs
can be isolated from phage libraries. The average man skilled in
the art is without any problems in a position, using his knowledge,
to execute such alternative methods, so that no detailed
description is needed here.
[0034] By using the invention, new superagonistic CD28 family
specific mAbs and/or mimicry compounds can however also be
identified. Therefore the invention also relates to the use of a
peptide, of a protein according to the invention or of a mimicry
compound thereto according to the invention in a screening method
for the identification of substances superagonistically modulating
the proliferation of T cells of several to all sub-groups, a
prospective substance or a mixture of prospective substances being
subjected to a binding assay with the peptide or protein or mimikry
compound thereto, and substances binding to the peptide or protein
or mimikry compound thereto being selected. In principle, any
conventional binding assay can be used. Of special importance may
be here the search for mimicry compounds, since these are typically
so-called small molecules having pharmacological advantages over
macromolecules. In such a screening method for mimicry compounds, a
substance library can be screened with high throughput. Both
aforementioned uses may be carried out with a native CD 28 receptor
family member as well.
[0035] A peptide, protein or mimicry compound thereto according to
the invention as well as the mAbs or mimicry compounds thereto
according to the invention have therapeutic relevance, since
thereby lymphoproliferative diseases may be treated by inhibition
of the proliferation, as well as immunodeficiency diseases by
activation of the proliferation. The induction of effector
functions, e.g. secretion of effector substances, is also possible.
This is achieved by selection or design of the mAb or of the
mimicry compound according to a specificity and high affinity for a
specific member of the CD28 family. If a higher
specificity/affinity is desirable, for instance, for controlling
surprising side effects, the process may be such that a second
ligand in addition to the mAb or the mimicry compound with
specificity for the special family member is searched, and the
second ligand is linked, after an analysis of the relative spatial
positions of the bound two ligands with respect to each other, by a
bridging molecule with the mAb or the mimicry compound. The
determination of the position of two ligands with respect to each
other after binding to a target can for instance be made by X-ray
structure analysis or multi-dimensional NMR correlation
spectroscopy, for instance .sup.15N/.sup.1H NMR. A second ligand
can be determined by conventional screening methods, the special
CD28 family member being used as a target. It is understood that
the second ligand does not bind at the C'-D loop, but spaced
thereto. On the other hand it is possible that a peptide, protein
or mimicry compound according to the invention thereto hot having
an otherwise physiological effect, competitively binds natural
ligands of the members of the CD28 family, and thus creates a
reverse effect by prevention of a pathologically caused natural
activation or inhibition.
[0036] Therefore the invention relates on one hand also to the use
of a peptide, protein or mimicry compound thereto according to the
invention for producing a pharmaceutical composition for the
modulation of the physiological T cell proliferation, wherein a
such compound is optionally mixed with suitable carrier and
auxiliary compounds and galenically prepared for the desired mode
of administration, e.g. injection i.v or i.p.
[0037] On the other hand, therefore, the invention relates to the
use of a mAb according to the invention or a mimicry compound
thereto according to the invention for producing a pharmaceutical
composition for the treatment of diseases with pathologically
reduced CD4 T cell numbers, in particular AIDS, for producing a
pharmaceutical composition for the treatment following stem cell
transplantations after chemo or radio therapy of leukemic diseases,
for producing a pharmaceutical composition for multiplying and/or
qualitatively influencing immune reactions after vaccinations, and
for producing a pharmaceutical composition for the treatment of
autoimmune-inflammatory diseases. Mixing and galenic preparation is
performed in an according manner.
[0038] The invention finally relates to therapeutic methods,
wherein to a person suffering from a proliferative or
immunodeficient disease, a pharmaceutical composition according to
the invention is administered.
[0039] In the following, the invention is described in more detail
with respect to examples representing embodiments only. There
are:
[0040] FIG. 1: stimulation of T lymphocytes from the rat with
various CD28-specific mAbs (a: costimulation, b: superagonistic,
i.e. direct stimulation),
[0041] FIG. 2: a sequence comparison between mouse, rat and human
CD28 in the section of the C'-D loop (in box),
[0042] FIG. 3: experimental results for the localization of the
binding site of superagonistic mAbs at the CD28 molecule of the
rat,
[0043] FIG. 4: binding of various human CD28-specific mAbs to CD28
(a) and costimulatory (b) and superagonistic, i.e. directly
stimulatory (c) activity of the mAbs of FIG. 4a,
[0044] FIG. 5: binding tests that show that human CD28-specific
superagonistic mAbs specifically bind to the C'-D loop,
[0045] FIG. 6: a three-dimensional representation of CD28 with
marking of the C'-D loop,
[0046] FIG. 7: sequence alignment under emphasis of the C'-D loop
or analogous structures, for various members of the CD28
family,
[0047] FIG. 8: experiments for the activation of cells by means of
human CD28-specific mAbs and mutated mouse CD28 molecules,
[0048] FIG. 9: representation of the sequences seq.-ID 33-40 (a-h),
and
[0049] FIG. 10: humanized variable domain of the mAb 5.11A (light
chain: VLC5.11=a; heavy chain: VHC5.11=b)
[0050] FIG. 1 shows the stimulation of freshly isolated T
lymphocytes from the rat in the form of a 3H thymidine
incorporation. The approach corresponds to the one described in
document WO98/54225, to which reference to a full extent is made
here and in the following, and the scope of disclosure of which is
hereby incorporated in the present text. In FIG. 1a is shown the
costimulation, i.e. in all wells, T cell receptors (TCR) specific
mAbs were bound to the plastic surface. For lack of costimulation,
the negative control (uppermost row) does not show any
incorporation. Costimulation is then given by the addition of
CD28-specific mAbs in a soluble form. All the different
CD28-specific mAbs used are shown. This series of various
CD28-specific mAbs originate from an approach of the immunization
and production of hybridoma cell lines described in WO98/54225
already. These are culture supernatants containing sufficient
CD28-specific mAbs for a saturating binding to the 2.times.10
cultivated cells. From FIG. 1a can be taken that all of these mAbs
are able to activate in a costimulating manner, i.e. to induce in
presence of the anti-TCR mAbs the thymidine incorporation. In FIG.
1b is shown the stimulation in absence of TCR specific mAbs. This
experiment has also been performed as described in document
WO98/54225. It can be seen that only two mAbs are capable to
stimulate the T lymphocytes in absence of a TCR signal. These mAbs
have thus superagonistic activity.
[0051] Furthermore it has been examined whether costimulatory and
superagonistic CD28-specific mAbs bind to different sections of the
CD28 molecule. The mAbs were produced by immunization of mice with
CD28 from the rat; as expected, they all do not react with mouse
CD28 (not shown). Since the mAbs can thus only recognize such
sections of the rat CD28 molecule which differ from that of the
mouse, first a sequence comparison between CD28 from the mouse and
the rat was performed (see FIG. 2, upper part). The differences
between the two species are highlighted. For the designation of the
amino acids, the one-letter code was used. As prototypes for a
conventional rat CD28-specific mAb was used JJ319, for a
superagonistic mAb JJ316 (see WO98/54225).
[0052] In FIG. 3, the mapping of the binding region is shown.
Expression plasmids were constructed, wherein one part of the
extracellular domain of CD28 originates from the mouse, another one
from the rat. This is shown by bars or lines, respectively; on the
right-hand side thereof the binding of the mAbs JJ316 and JJ319 to
mouse fibroblasts (L929 cells) is represented, which have been
transfected with these expression plasmids. In the first two lines
of FIG. 3 (m/r and r/m 1-37), the binding of both antibodies to the
"right-hand" half of the sequence is mapped: Both will bind if this
originates from the rat. In the reverse construct (rm CD28 1-37,
left-hand rat, right-hand mouse), there is no binding. In the third
line (m/r CD28 1-66) it is shown that JJ316 does not bind anymore,
whereas the still present part of the rat sequence ("right-hand")
is still sufficient for a detection by JJ319. Thus, the two mAbs
detect different epitopes on the CD28 molecule, and the binding of
the superagonist JJ316 is located in that region which originated
in the construct of the first line, not however in the construct of
the third line from the rat. A clear candidate therefor is the
section in the box of FIG. 2.
[0053] In lines 4 and 5 of FIG. 3, therefore first two and then
three amino acids were so modified in this region of the mouse CD28
molecule that they will now represent the rat sequence. By this
"transplantation" of three amino acids only, the binding capability
for mAb JJ316, not however that of JJ319 could be transferred. In
Table 1 the binding data for the whole group of CD28-specific mAbs
are summarized. There exists a unique correlation: a significant
binding to the C'-D loop of the rat which has been created by
transfer of the amino acid positions 62, 64 and 65 to the CD28
molecule of the mouse, could be found for the two superagonistic
mAbs JJ316 and 5S38 17 only, not however for the conventional (only
costimulatory) mAbs. A costimulatory mAb (5S35) detects the epitope
in the box to a very weak degree and binds very strongly to the
"conventional" epitope.
[0054] The next figures deal with superagonistic human-specific
mAb. These have also been produced in mice, thus do not react with
the CD28 molecule of the mouse. The mice were immunized with human
CD28-transfected A20/J mouse B lymphoma cells (see WO98/54225) and
additionally boostered prior to the fusion with commercially
available human CD28-Fc fusion protein (bought from R and D
Systems) . In a series of fusion experiments, 24 out of several
thousand cell lines were identified which produce human
CD28-specific mAbs (binding to mouse L929 cells expressing human
CD28, not however to untransfected L929 cells), in an analogous
manner to the screen in document WO98/54225. From these, two showed
the desired superagonistic activity (9D7 and 5.11A), whereas all
new mAbs have a conventional costimulatory activity. In the
following, in particular these two superagonistic mAbs will be
described. As an example for a conventional human CD28-specific
mAb, the also newly generated mAb 7.3B6 is used.
[0055] FIG. 4a shows that the used preparations of the three new
mAbs bind comparatively well and also with comparable titer to
human T lymphocytes. An experiment is shown, wherein freshly
isolated mononuclear cells from the human blood (so-called PBMC)
were first treated with different dilution steps of the used mAbs
on ice; then they were washed, and the bound mAb was made visible
by a secondary antibody marked by a fluorescence dye, said
secondary antibody specifically detecting the bound mouse mAbs. By
using another mAb detecting human CD4 T cells, and to which a
second fluorescence dye was bound, the binding of the titrated mAbs
could be determined by electronic gating selectively for the CD4 T
lymphocytes. By "MFI" is given the average fluorescence intensity
being a measure for the amount of the bound CD28-specific mAb. The
concentrations represent 3-fold dilutions of a standardized
original preparation. It is fully normal that in this test the
highest concentration gives a weaker signal than the following
titration steps; this has to do with the avidity (bivalent binding)
of mAbs and does not play any role in the context discussed
here.
[0056] In FIG. 4b and c the capability of these three new
CD28-specific mAbs to stimulate--in presence and absence of a TCR
signal--freshly isolated human T cells to growth is compared.
Again, a 3H thymidine incorporation is shown, as described above
for the rat. For FIG. 4b, the wells were coated with a mAb which
reacts with the human TCR/CD3 complex. Thus, the costimulation was
measured. It can be seen that the proliferation without
costimulation with one the mAbs fails to appear (negative control),
all three antibodies are however capable to stimulate the cell
division. For FIG. 4c, the absence of a TCR/CD3-specific mAb was
selected. Only the antibodies 9D7 and 5.11A could stimulate in a
superagonistic way.
[0057] After the epitope for superagonistic mAb for the rat has
been defined, and two new superagonistic mAbs having specificity
for human CD28 have been isolated, it has been checked whether
these mAbs bind to the corresponding position of the human CD28
molecule. As can be seen from FIG. 2, the CD28 molecules of mouse
and man differ in numerous positions. Based on the mapping of the
superagonistic epitope for the rat, it has therefore been directly
checked whether the binding site for the superagonistic epitope on
human CD28 to the CD28 molecule of the mouse can be achieved by
"transplantation" of the five amino acids of this homologous
region. The results are shown in FIG. 5. Taking into account the
background of the homogeneously represented mouse sequence for the
extracellular domain of the CD28 molecule (center) the exchanged
(mouse to man) amino acid positions are shown as lines (bottom).
The numbers at the side indicate in addition the individual
positions and mutations (F60V means for instance that at position
60 the phenylalanine of the mouse has been replaced by a valine of
the human sequence). Adjacent thereto, the binding of the three
examined mAbs is represented. As the figure shows, all three mAbs
detect human CD28, only the two mAbs 9D7 and 5.11A however react
with the mouse molecule, into which the five amino acids of the
human CD28 have been transplanted at the crucial position. Taking
into account the great variety of differences, this specific
generation of the reactivity is surprising and confirms to a full
extent the findings, derived from the experiments with rat CD28,
that superagonistic mAbs must bind to a specific, namely exactly
this site of the molecule.
[0058] In FIG. 6 is shown a three-dimensional model of the CD28
molecule. The newly identified binding region is highlighted. It
corresponds to the sequence in the box in FIG. 2. Concerning its
structure, the extracellular domain of CD28 belongs to the
so-called immunoglobulin super family being characterized by two
superimposed 8 sheets as a basic structure. The labeling of these
bands follows a pattern as given in the literature. It is important
for the representation shown here that the region identified as an
epitope for superagonistic CD28-specific mAbs in rat and mouse is
described by "C'-D loop". It has thus been shown that mAbs having
specificity for the C'-D loop of the CD28 molecule show
superagonistic activity, thus can be used for the activation of T
lymphocytes in the meaning of the document WO98/54225. The
superagonistic activity of C'-D loop-specific mAbs in rat and man
shows that therein not the sequence of the epitope, but its
position or shape is important.
[0059] CD28 belongs to a family of cell surface receptors with
immuno-regulatory activity. This is either stimulating (CD28, ICOS)
or inhibiting (CTLA-4, PD-1). In FIG. 7, the sequences of the known
members of the CD28 family are shown in the sense of an
"alignment". The C'-D loop for CD28 is highlighted. Analogous loops
of the other molecules (in box) are a correspondingly favorable
target structure for the development of superagonistic ligands. It
should be noted, with regard to FIG. 7, that "-" is a gap in the
alignment, i.e. the amino acids following thereto are immediately
connected to each other.
[0060] In the experiments of FIG. 8, it has been examined whether
mAbs according to the invention having specificity for the C'-D
loop of the rat or man do not only bind to the mouse CD28 with
"transplanted" C'-D loop of the rat or man (see FIGS. 3 and 5), but
whether there is really an activation. For this purpose, T tumor
cells of the mouse, BW, were transfected either with the construct
of FIG. 3, line 5 (rat C'-D loop transfer) or with the construct of
FIG. 5, line 3 (human C'-D loop). The activation of these cells is
not measured by cell division (they proliferate anyway), but by the
production of the cytokine IL-2. FIG. 8 shows first that without
stimulation there is no IL-2 production (negative control). The
stimulation with a T cell receptor-specific mAb induces IL-2
production (positive control). FIG. 8a shows the results when using
the superagonistic mAb JJ316 of the rat, whereas FIG. 8b shows the
results for the human C'-D loop-specific mAb 5.11A. In either case
the respective cell lines are stimulated for the IL-2 production.
As expected, there is however no stimulation by "conventional"
CD28-specific mAbs, since these do not only not bind to the C'-D
loop, but cannot detect the construct at all, because they are
specific for rat or human-specific sequences not contained in the
construct.
[0061] In FIG. 9a is shown the nucleic acid sequence of the
variable region of the light chain of a mAb 9D7 according to the
invention (seq.-ID 33). FIG. 9b shows the peptide coded thereby
(seq.-ID 34). FIG. 9c shows the nucleic acid sequence of the
variable region of the heavy chain of this mAb (seq.-ID 35). FIG.
9d is the peptide coded thereby (seq.-ID 36).
[0062] In FIG. 9e is represented the nucleic acid sequence of the
variable region of the heavy chain of a mAb 5.11A according to the
invention (seq.-ID 37). FIG. 9f shows the peptide coded thereby
(seq.-ID 38). FIG. 9g shows the nucleic acid sequence of the
variable region of the light chain of this mAb (seq.-ID 39). FIG.
9h shows the peptide coded thereby (seq.-ID 40). FIG. 10 shows the
humanized variable domain of the mAb 5.11A. FIG. 10a is the light
chain, and FIG. 10b the heavy chain.
1TABLE I Binding of anti-rat CD28 mAbs to mouse and rat CD28 and
various CD28 mutants mCD28, S62P m/rCD28 mAb Mouse CD28 Rat CD28
A64V, EG5G Mva11269I Control -- -- -- -- JJ316 -- + + + + + + --
JJ319 -- + + + -- + + + 5S28 -- + + -- + + + 5S38.17 -- + + + + + +
-- 5S247 -- + + + -- + + + 5G40/3 -- + + + -- + + + 5G87 -- + + --
+ + 5G111 -- + + -- + + 5G35 -- + + + + + + +
[0063]
Sequence CWU 1
1
44 1 6 PRT Artificial Novel Sequence 1 Val Tyr Ser Lys Thr Gly 1 5
2 5 PRT Artificial Novel Sequence 2 Phe Leu Asp Asp Ser 1 5 3 6 PRT
Artificial Novel Sequence 3 Val Ser Ile Lys Ser Leu 1 5 4 6 PRT
Artificial Novel Sequence 4 Gln Pro Gly Gln Asp Cys 1 5 5 4 PRT
Artificial Novel Sequence 5 Tyr Ser Lys Thr 1 6 5 PRT Artificial
Novel Sequence 6 Tyr Ser Lys Thr Gly 1 5 7 5 PRT Artificial Novel
Sequence 7 Val Tyr Ser Lys Thr 1 5 8 6 PRT Artificial Novel
Sequence 8 Tyr Ser Lys Thr Gly Phe 1 5 9 7 PRT Artificial Novel
Sequence 9 Val Tyr Ser Lys Thr Gly Phe 1 5 10 5 PRT Artificial
Novel Sequence 10 Ser Lys Thr Gly Phe 1 5 11 8 PRT Artificial Novel
Sequence 11 Gly Asn Tyr Ser Gln Gln Leu Gln 1 5 12 4 PRT Artificial
Novel Sequence 12 Leu Asp Asp Ser 1 13 5 PRT Artificial Novel
Sequence 13 Leu Asp Asp Ser Ile 1 5 14 5 PRT Artificial Novel
Sequence 14 Phe Leu Asp Asp Ser 1 5 15 6 PRT Artificial Novel
Sequence 15 Leu Asp Asp Ser Ile Cys 1 5 16 7 PRT Artificial Novel
Sequence 16 Phe Leu Asp Asp Ser Ile Cys 1 5 17 5 PRT Artificial
Novel Sequence 17 Asp Asp Ser Ile Cys 1 5 18 8 PRT Artificial Novel
Sequence 18 Tyr Met Met Gly Asn Glu Leu Thr 1 5 19 4 PRT Artificial
Novel Sequence 19 Ser Ile Lys Ser 1 20 5 PRT Artificial Novel
Sequence 20 Ser Ile Lys Ser Leu 1 5 21 5 PRT Artificial Novel
Sequence 21 Val Ser Ile Lys Ser 1 5 22 6 PRT Artificial Novel
Sequence 22 Ser Ile Lys Ser Leu Lys 1 5 23 7 PRT Artificial Novel
Sequence 23 Val Ser Ile Lys Ser Leu Lys 1 5 24 5 PRT Artificial
Novel Sequence 24 Ile Lys Ser Leu Lys 1 5 25 8 PRT Artificial Novel
Sequence 25 Lys Thr Lys Gly Ser Gly Asn Thr 1 5 26 4 PRT Artificial
Novel Sequence 26 Pro Gly Gln Asp 1 27 5 PRT Artificial Novel
Sequence 27 Pro Gly Gln Asp Cys 1 5 28 5 PRT Artificial Novel
Sequence 28 Gln Pro Gly Gln Asp 1 5 29 6 PRT Artificial Novel
Sequence 29 Pro Gly Gln Asp Cys Arg 1 5 30 7 PRT Artificial Novel
Sequence 30 Gln Pro Gly Gln Asp Cys Arg 1 5 31 5 PRT Artificial
Novel Sequence 31 Gly Gln Asp Cys Arg 1 5 32 9 PRT Artificial Novel
Sequence 32 Leu Ala Ala Phe Pro Glu Asp Arg Ser 1 5 33 321 DNA
Artificial Novel Sequence 33 gatatccaga cgacacagac tacatcctcc
cgttctgcct ctctgggaga cagagtcacc 60 atcagttgca gggcaggtca
ggacattagt aattatttaa actggtatca gcagaaacca 120 gatggaactg
ttaagctcct gatctactac acatcaagat tacactcagg agtcccatca 180
aggttcagtg gcagtgggtc tggaacagat tattctctca ccattagcaa cctggagcaa
240 gaagatattg ccacttactt ttgccaacag ggtcatacgc ttccgtggac
gttcggtgga 300 ggcaccaagc tggaaatcaa a 321 34 107 PRT Artificial
Novel Sequence 34 Asp Ile Gln Thr Thr Gln Thr Thr Ser Ser Leu Ser
Ala Ser Leu Gly 1 5 10 15 Asp Arg Val Thr Ile Ser Cys Arg Ala Gly
Gln Asp Ile Ser Asn Tyr 20 25 30 Leu Asn Trp Tyr Gln Gln Lys Pro
Asp Gly Thr Val Lys Leu Leu Ile 35 40 45 Tyr Tyr Thr Ser Arg Leu
His Ser Gly Val Pro Ser Arg Phe Ser Gly 50 55 60 Ser Gly Ser Gly
Thr Asp Tyr Ser Leu Thr Ile Ser Asn Leu Glu Gln 65 70 75 80 Glu Asp
Ile Ala Thr Tyr Phe Cys Gln Gln Gly His Thr Leu Pro Trp 85 90 95
Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys 100 105 35 363 DNA
Artificial Novel Sequence 35 gatgtgcagc ttcaggagtc gggacctggc
ctggtgaaac cttctcagtc tctgtccctc 60 acctgcactg tcactggcta
ctcaatcacc agtgattatg cctggaactg gatccggcag 120 tttccaggaa
acaaactgga gtggatgggc tacataagat acagtggtag tactagctac 180
aatccatctc tcaaaagtcg aatctctatc actcgagaca catccaagaa ccagttcttc
240 ctgcagttga attctgtgac tactgaggac acagccacat attactgtgc
aagagattgg 300 ccgcgaccga gctactggta cttcgatgtc tggggcgcag
ggaccacggt caccgtctcc 360 tca 363 36 121 PRT Artificial Novel
Sequence 36 Asp Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Lys Pro
Ser Gln 1 5 10 15 Ser Leu Ser Leu Thr Cys Thr Val Thr Gly Tyr Ser
Ile Thr Ser Asp 20 25 30 Tyr Ala Trp Asn Trp Ile Arg Gln Phe Pro
Gly Asn Lys Leu Glu Trp 35 40 45 Met Gly Tyr Ile Arg Tyr Ser Gly
Ser Thr Ser Tyr Asn Pro Ser Leu 50 55 60 Lys Ser Arg Ile Ser Ile
Thr Arg Asp Thr Ser Lys Asn Gln Phe Phe 65 70 75 80 Leu Gln Leu Asn
Ser Val Thr Thr Glu Asp Thr Ala Thr Tyr Tyr Cys 85 90 95 Ala Arg
Asp Trp Pro Arg Pro Ser Tyr Trp Tyr Phe Asp Val Trp Gly 100 105 110
Ala Gly Thr Thr Val Thr Val Ser Ser 115 120 37 360 DNA Artificial
Novel Sequence 37 caggtccaac tgcagcagtc cggacctgag ctggtgaagc
cggggacttc agtgaggatt 60 tcctgcgagg cttctggcta caccttcaca
agctactata tacactgggt gaaacagagg 120 cctggacagg gacttgagtg
gattggatgt atttatcctg gaaatgtcaa tactaactat 180 aatgagaagt
tcaaggacaa ggccacactg attgtagaca catcctccaa cactgcctac 240
atgcagctca gcagaatgac ctctgaggac tctgcggtct atttctgtac aagatcacac
300 tacggcctcg actggaactt cgatgtctgg ggcgcaggga ccacggtcac
cgtctcctca 360 38 120 PRT Artificial Novel Sequence 38 Gln Val Gln
Leu Gln Gln Ser Gly Pro Glu Leu Val Lys Pro Gly Thr 1 5 10 15 Ser
Val Arg Ile Ser Cys Glu Ala Ser Gly Tyr Thr Phe Thr Ser Tyr 20 25
30 Tyr Ile His Trp Val Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile
35 40 45 Gly Cys Ile Tyr Pro Gly Asn Val Asn Thr Asn Tyr Asn Glu
Lys Phe 50 55 60 Lys Asp Lys Ala Thr Leu Ile Val Asp Thr Ser Ser
Asn Thr Ala Tyr 65 70 75 80 Met Gln Leu Ser Arg Met Thr Ser Glu Asp
Ser Ala Val Tyr Phe Cys 85 90 95 Thr Arg Ser His Tyr Gly Leu Asp
Trp Asn Phe Asp Val Trp Gly Ala 100 105 110 Gly Thr Thr Val Thr Val
Ser Ser 115 120 39 321 DNA Artificial Novel Sequence 39 gacatccaga
tgaaccagtc tccatccagt ctgtctgcat cccttggaga cacaattacc 60
atcacttgcc atgccagtca aaacatttat gtttggttaa actggtacca gcagaaacca
120 ggaaatattc ctaaactctt gatctataag gcttccaacc tgcacacagg
cgtcccatca 180 aggtttagtg gcagtggatc tggaacaggc ttcacattaa
ccatcagcag cctgcagcct 240 gaagacattg ccacttacta ctgtcaacag
ggtcaaactt atccgtacac gttcggaggg 300 gggaccaagc tggaaataaa a 321 40
107 PRT Artificial Novel Sequence 40 Asp Ile Gln Met Asn Gln Ser
Pro Ser Ser Leu Ser Ala Ser Leu Gly 1 5 10 15 Asp Thr Ile Thr Ile
Thr Cys His Ala Ser Gln Asn Ile Tyr Val Trp 20 25 30 Leu Asn Trp
Tyr Gln Gln Lys Pro Gly Asn Ile Pro Lys Leu Leu Ile 35 40 45 Tyr
Lys Ala Ser Asn Leu His Thr Gly Val Pro Ser Arg Phe Ser Gly 50 55
60 Ser Gly Ser Gly Thr Gly Phe Thr Leu Thr Ile Ser Ser Leu Gln Pro
65 70 75 80 Glu Asp Ile Ala Thr Tyr Tyr Cys Gln Gln Gly Gln Thr Tyr
Pro Tyr 85 90 95 Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys 100
105 41 15 PRT Artificial Novel Sequence 41 Gly Asn Tyr Ser Gln Gln
Leu Gln Val Tyr Ser Lys Thr Gly Phe 1 5 10 15 42 15 PRT Artificial
Novel Sequence 42 Tyr Met Met Gly Asn Glu Leu Thr Phe Leu Asp Asp
Ser Ile Cys 1 5 10 15 43 15 PRT Artificial Novel Sequence 43 Lys
Thr Lys Gly Ser Gly Asn Thr Val Ser Ile Lys Ser Leu Lys 1 5 10 15
44 16 PRT Artificial Novel Sequence 44 Leu Ala Ala Phe Pro Glu Asp
Arg Ser Gln Pro Gly Gln Asp Cys Arg 1 5 10 15
* * * * *