U.S. patent application number 11/114233 was filed with the patent office on 2007-09-13 for altered antibodies and their preparation.
This patent application is currently assigned to BURROUGHS WELLCOME CO.. Invention is credited to Michael Ronald Clark, Stephen Paul Cobbold, Scott David Gorman, Herman Waldmann.
Application Number | 20070212753 11/114233 |
Document ID | / |
Family ID | 10682322 |
Filed Date | 2007-09-13 |
United States Patent
Application |
20070212753 |
Kind Code |
A1 |
Gorman; Scott David ; et
al. |
September 13, 2007 |
Altered antibodies and their preparation
Abstract
An altered antibody chain is produced in which the CDR's of the
variable domain of the chain are derived from a first mammalian
species. The framework-encoding regions of DNA encoding the
variable domain of the first species are mutated so that the
mutated framework-encoding regions encode a framework derived from
a second different mammalian species. The or each constant domain
of the antibody chain, if present, are also derived from the second
mammalian species. An antibody which is capable of binding to human
CD4 antigen is also provided together with a pharmaceutical
composition comprising the antibody.
Inventors: |
Gorman; Scott David;
(Cambridge, GB) ; Clark; Michael Ronald;
(Cambridge, GB) ; Cobbold; Stephen Paul;
(Cambridge, GB) ; Waldmann; Herman; (Cambridge,
GB) |
Correspondence
Address: |
FINNEGAN, HENDERSON, FARABOW, GARRETT & DUNNER;LLP
901 NEW YORK AVENUE, NW
WASHINGTON
DC
20001-4413
US
|
Assignee: |
BURROUGHS WELLCOME CO.
Research Triangle Park
NC
|
Family ID: |
10682322 |
Appl. No.: |
11/114233 |
Filed: |
April 26, 2005 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
08459655 |
Jun 2, 1995 |
|
|
|
11114233 |
Apr 26, 2005 |
|
|
|
08030175 |
May 17, 1993 |
6767996 |
|
|
08459655 |
Jun 2, 1995 |
|
|
|
Current U.S.
Class: |
435/69.1 ;
424/130.1; 435/320.1; 435/326; 530/387.1; 536/23.53 |
Current CPC
Class: |
C07K 2319/00 20130101;
C07K 2317/24 20130101; C07K 16/2812 20130101; A61K 38/00 20130101;
A61P 37/00 20180101 |
Class at
Publication: |
435/069.1 ;
435/320.1; 435/326; 424/130.1; 530/387.1; 536/023.53 |
International
Class: |
C12P 21/06 20060101
C12P021/06; C07H 21/04 20060101 C07H021/04; A61K 39/395 20060101
A61K039/395; C07K 16/18 20060101 C07K016/18; C12N 5/06 20060101
C12N005/06 |
Foreign Application Data
Date |
Code |
Application Number |
Sep 17, 1990 |
GB |
90 20282.1 |
Sep 16, 1991 |
WO |
PCT/GB91/01578 |
Claims
1. A process for the preparation of an antibody chain in which
complementarity determining regions (CDRs) of the variable domain
of the antibody chain are derived from a first mammalian species
and the framework of the variable domain and, if present, the or
each constant domain of the antibody chain are derived from a
second different mammalian species, which process comprises: (i)
mutating the framework-encoding regions of DNA encoding a variable
domain of an antibody chain of the said first species such that the
mutated framework-encoding regions encode the said framework
derived from the said second species; and (ii) expressing the said
antibody chain utilising the mutated DNA from step (i); the
mutation in step (i) being such that an antibody incorporating the
antibody chain expressed in step (ii) retains the binding
capability of the antibody from which the CDRs are derived.
2. A process according to claim 1, wherein the framework-encoding
regions of DNA encoding the variable domain of an antibody heavy
chain are mutated in step (i).
3. A process according to claim 1 or 2, wherein the
framework-encoding regions of DNA encoding the variable domain of
an antibody light chain are mutated in step (i).
4. A process according to any one of the preceding claims, wherein
the said first species is rat or mouse.
5. A process according to any one of the preceding claims, wherein
the said second species is human.
6. A process according to any one of the preceding claims,
comprising: (a) determining the nucleotide and predicted amino acid
sequence of a variable domain of a selected antibody chain of the
said first species; (b) determining the antibody framework to which
the framework of the said domain is to be altered; (c) mutating the
framework-encoding regions of DNA encoding the said variable domain
such that the mutated framework-encoding regions encode the
framework determined upon in step (b). (d) linking the mutated DNA
obtained in step (c) to DNA encoding a constant domain of the said
second species and cloning the DNA into an expression vector; and
(e) introducing the expression vector into a compatible host cell
and culturing the host cell under such conditions that antibody
chain is expressed.
7. A process according to claim 6, in which about the most
homologous framework of an antibody chain of a different species is
selected in step (b) as the framework to which the framework of the
said variable domain is to be altered.
8. A process according to any one of the preceding claims, wherein
the antibody of the said first species is a CD4 antibody.
9. A process according to any one of the preceding claims, wherein
the said antibody chain is co-expressed with a complementary
antibody chain and antibody comprising the said two chains is
recovered.
10. An antibody which is capable of binding to human CD4 antigen,
in which the CDRs of the light chain of the antibody have the amino
acid sequences: TABLE-US-00004 CDR1: LASEDIYSDLA CDR2: NTDTLQN
CDR3: QQYNNYPWT
in which the CDRs of the heavy chain of the antibody have the amino
acid sequences: TABLE-US-00005 CDR1: NYGMA CDR2: TISHDGSDTYFRDSVKG
CDR3: QGTIAGIRH, and
in which the framework of the variable domain and, if present, the
or each constant domain of each chain are derived from a mammalian
non-rat species.
11. An antibody according to claim 10, in which the mammalian
non-rat species is human.
12. An antibody according to claim 11, in which the variable domain
framework of the heavy chain is homologous to the heavy chain
variable domain framework of the protein KOL.
13. An antibody according to claim 12, in which the heavy chain
variable region has the amino acid sequence shown in the upper line
in FIG. 10 or 12.
14. An antibody according to claim 11, in which the variable domain
framework of the heavy chain is homologous to the heavy chain
variable domain framework of the protein NEW.
15. An antibody according to claim 14, in which the heavy chain
variable region has the amino acid sequence shown in the upper line
of FIG. 6 or 7.
16. An antibody according to any one of claims 11 to 15, in which
the variable domain framework of the light chain is homologous to
the variable domain framework of the protein REI.
17. An antibody according to claim 16, in which the light chain has
the amino acid sequence shown in the upper line of FIG. 3.
18. An antibody which is capable of binding to human CD4-antigen in
which the CDRs of the light and heavy chain are derived from a
non-human mammalian species and the framework of the variable
domain and, if present, the or each constant domain of each chain
are human, the said antibody having glycosylation characteristic of
CHO cells.
19. A pharmaceutical composition comprising a pharmaceutically
acceptable carrier or diluent and, as active ingredient, an
antibody as claimed in any one of claims 10 to 18.
Description
[0001] The present invention relates to altered antibodies and
their preparation. The invention is typically applicable to the
production of humanised antibodies.
[0002] Antibodies typically comprise two heavy chains linked
together by disulphide bonds and two light chains. Each light chain
is linked to a respective heavy chain by disulphide bonds. Each
heavy chain has at one end a variable domain followed by a number
of constant domains. Each light chain has a variable domain at one
end and a constant domain at its other end. The light chain
variable domain is aligned with the variable domain of the heavy
chain. The light chain constant domain is aligned with the first
constant domain of the heavy chain. The constant domains in the
light and heavy chains are not involved directly in binding the
antibody to antigen.
[0003] The variable domains of each pair of light and heavy chains
form the antigen binding site. The domains on the light and heavy
chains have the same general structure and each domain comprises a
framework of four regions, whose sequences are relatively
conserved, connected by three complementarity determining regions
(CDRs). The four framework regions largely adopt a beta-sheet
conformation and the CDRs form loops connecting, and in some cases
forming part of, the beta-sheet structure. The CDRs are held in
close proximity by the framework regions and, with the CDRs from
the other domain, contribute to the formation of the antigen
binding site.
[0004] The preparation of an altered antibody in which the CDRs are
derived from a different species than the framework of the
antibody's variable domains is disclosed in EP-A-0239400. The CDRs
may be derived from a rat or mouse monoclonal antibody. The
framework of the variable domains, and the constant domains, of the
altered antibody may be derived from a human antibody. Such a
humanised antibody elicits a negligible immune response when
administered to a human compared to the immune response mounted by
a human against a rat or mouse antibody. Humanised CAMPATH-1
antibody is disclosed in EP-A-0328404.
[0005] We have now devised a new way of preparing an altered
antibody. In contrast to previous proposals, this involves altering
the framework of a variable domain rather than the CDRs. This
approach has the advantages that it does not require a pre-existing
cDNA encoding, for example, a human framework to which to reshape
and that it is technically easier than prior methodologies.
[0006] Accordingly, the present invention provides a process for
the preparation of an antibody chain in which the CDRs of the
variable domain of the antibody chain are derived from a first
mammalian species and the framework of the variable domain and, if
present, the or each constant domain of the antibody chain are
derived from a second different mammalian species, which process
comprises:
[0007] (i) mutating the framework-encoding regions of DNA encoding
a variable domain of an antibody chain of the said first species
such that the mutated framework-encoding regions encode the said
framework derived from the said second species; and
[0008] (ii) expressing the said antibody chain utilising the
mutated DNA from step (i).
[0009] A variable domain of either or both chains of an antibody
can therefore be altered by:
[0010] (a) determining the nucleotide and predicted amino acid
sequence of a variable domain of a selected antibody chain of the
said first species;
[0011] (b) determining the antibody framework to which the
framework of the said variable domain is to be altered;
[0012] (c) mutating the framework-encoding regions of DNA encoding
the said variable domain such that the mutated framework-encoding
regions encode the framework determined upon in step (b);
[0013] (d) linking the mutated DNA obtained in step (c) to DNA
encoding a constant domain of the said second species and cloning
the DNA into an expression vector; and
[0014] (e) introducing the expression vector into a compatible host
cell and culturing the host cell under such conditions that
antibody chain is expressed.
[0015] The antibody chain may be co-expressed with a complementary
antibody chain. At least the framework of the variable domain and
the or each constant domain of the complementary chain generally
are derived from the said second species also. A light chain and a
heavy chain may be co-expressed. Either or both chains may have
been prepared by the process of the invention. Preferably the CDRs
of both chains are derived from the same selected antibody. An
antibody comprising both expressed chains can be recovered.
[0016] The antibody preferably has the structure of a natural
antibody or a fragment thereof. The antibody may therefore comprise
a complete antibody, a (Fab').sub.2 fragment, a Fab fragment, a
light chain dimer or a heavy chain. The antibody may be an IgG such
as an IgG1, IgG2, IgG3 or IgG4 IgM, IgA, IgE or IgD. Alternatively,
the antibody may be a chimaeric antibody of the type described in
WO 86/01533.
[0017] A chimaeric antibody according to WO 86/01533 comprises an
antigen binding region and a non-immunoglobulin region. The antigen
binding region is an antibody light chain variable domain or heavy
chain variable domain. Typically, the chimaeric antibody comprises
both light and heavy chain variable domains. The non-immunoglobulin
region is fused at its C-terminus to the antigen binding region.
The non-immunoglobulin region is typically a non-immunoglobulin
protein and may be an enzyme region, a region derived from a
protein having known binding specificity, from a protein toxin or
indeed from any protein expressed by a gene. The two regions of the
chimaeric antibody may be connected via a cleavable linker
sequence.
[0018] The invention is preferably employed to humanise an
antibody, typically a monoclonal antibody and, for example, a rat
or mouse antibody. The framework and constant domains of the
resulting antibody are therefore human framework and constant
domains whilst the CDRs of the light and/or heavy chain of the
antibody are rat or mouse CDRs. Preferably all CDRs are rat or
mouse CDRs. The antibody may be a human IgG such as IgG1, IgG2,
IgG3, IgG4; IgM; IgA; IgE or IgD carrying rat or mouse CDRs.
[0019] The process of the invention is carried out in such a way
that the resulting antibody retains the antigen binding capability
of the antibody from which it is derived. An antibody is reshaped
according to the invention by mutating the framework-encoding
regions of DNA coding for the variable domains of the antibody.
This antibody and the reshaped antibody should both be capable of
binding to the same antigen.
[0020] The starting antibody is typically an antibody of a selected
specificity. In order to ensure that this specificity is retained,
the variable domain framework of the antibody is preferably
reshaped to about the closest variable domain framework of an
antibody of another species. By "about the closest" is meant about
the most homologous in terms of amino acid sequences. Preferably
there is a homology of at least 50% between the two variable
domains.
[0021] There are four general steps to reshape a monoclonal
antibody. These are:
[0022] (1) determining the nucleotide and predicted amino acid
sequence of the starting antibody light and heavy chain variable
domains;
[0023] (2) designing the reshaped antibody, i.e. deciding which
antibody framework region to use during the reshaping process;
[0024] (3) the actual reshaping methodologies/techniques; and
[0025] (4) the transfection and expression of the reshaped
antibody.
[0026] These four steps are explained below in the context of
humanising an antibody. However, they may equally well be applied
when reshaping to an antibody of a non-human species.
Step 1: Determining the Nucleotide and Predicted Amino Acid
Sequence of the Antibody Light and Heavy Chain Variable Domains
[0027] To reshape an antibody only the amino acid sequence of
antibody's heavy and light chain variable domains needs to be
known. The sequence of the constant domains is irrelevant because
these do not contribute to the reshaping strategy. The simplest
method of determining an antibody's variable domain amino acid
sequence is from cloned cDNA encoding the heavy and light chain
variable domain.
[0028] There are two general methods for cloning a given antibody's
heavy and light chain variable domain cDNAs: (1) via a conventional
cDNA library, or (2) via the polymerase chain reaction (PCR). Both
of these methods are widely known. Given the nucleotide sequence of
the cDNAs, it is a simple matter to translate this information into
the predicted amino acid sequence of the antibody variable
domains.
Step 2: Designing the Reshaped Antibody
[0029] There are several factors to consider in deciding which
human antibody sequence to use during the reshaping. The reshaping
of light and heavy chains are considered independently of one
another, but the reasoning is basically similar for each.
[0030] This selection process is based on the following rationale:
A given antibody's antigen specificity and affinity is primarily
determined by the amino acid sequence of the variable region CDRs.
Variable domain framework residues have little or no direct
contribution. The primary function of the framework regions is to
hold the CDRs in their proper spacial orientation to recognize
antigen. Thus the substitution of rodent CDRs into a human variable
domain framework is most likely to result in retention of their
correct spacial orientation if the human variable domain is highly
homologous to the rodent variable domain-from which they
originated. A human variable domain should preferably be chosen
therefore that is highly homologous to the rodent variable
domain(s).
[0031] A suitable human antibody variable domain sequence can be
selected as follows: [0032] 1. Using a computer program, search all
available protein (and DNA) databases for those human antibody
variable domain sequences that are most homologous to the rodent
antibody variable domains. This can be easily accomplished with a
program called FASTA but other suitable programs are available. The
output of a suitable program is a list of sequences most homologous
to the rodent antibody, the percent homology to each sequence, and
an alignment of each sequence to the rodent sequence. This is done
independently for both the heavy and light chain variable domain
sequences. The above analyses are more easily accomplished if
customized sub-databases are first created that only include human
immunoglobulin sequences. This has two benefits. First, the actual
computational time is greatly reduced because analyses are
restricted to only those sequences of interest rather than all the
sequences in the databases. The second benefit is that, by
restricting analyses to only human immunoglobulin sequences, the
output will not be cluttered by the presence of rodent
immunoglobulin sequences. There are far more rodent immunoglobulin
sequences in databases than there are human. [0033] 2. List the
human antibody variable domain sequences that have the most overall
homology to the rodent antibody variable domain (from above). Do
not make a distinction between homology within the framework
regions and CDRs. Consider the overall homology. [0034] 3.
Eliminate from consideration those human sequences that have CDRs
that are a different length than those of the rodent CDRs. This
rule does not apply to CDR 3, because the length of this CDR is
normally quite variable. Also, there are sometimes no or very few
human sequences that have the same CDR lengths as that of the
rodent antibody. If this is the case, this rule can be loosened,
and human sequences with one or more differences in CDR length can
be allowed. [0035] 4. From the remaining human variable domains,
the one is selected that is most homologous to that of the rodent.
[0036] 5. The actual reshaped antibody (the end result) should
contain CDRs derived from the rodent antibody and a variable domain
framework from the human antibody chosen above. Step 3: The Actual
Reshaping Methodologies/Techniques
[0037] A cDNA encoding the desired reshaped antibody is preferably
made beginning with the rodent cDNA from which the rodent antibody
variable domain sequence(s) was originally determined. The rodent
variable domain amino acid sequence is compared to that of the
chosen human antibody variable domain sequence. The residues in the
rodent variable domain framework are marked that need to be changed
to the corresponding residue in the human to make the rodent
framework identical to that of the human framework. There may also
be residues that need adding to or deleting from the rodent
framework sequence to make it identical to that of the human.
[0038] Oligonucleotides are synthesised that can be used to
mutagenize the rodent variable domain framework to contain the
desired residues. Those oligonucleotides can be of any convenient
size. One is normally only limited in length by the capabilities of
the particular synthesizer one has available. The method of
oligonucleotide-directed in vitro mutagenesis is well known.
[0039] The advantages of this method of reshaping as opposed to
splicing CDRs into a human framework are that (1) this method does
not require a pre-existing cDNA encoding the human framework to
which to reshape and (2) splicing CDRs is technically more
difficult because there is usually a large region of poor homology
between the mutagenic oligonucleotide and the human antibody
variable domain. This is not so much a problem with the method of
splicing human framework residues onto a rodent variable domain
because there is no need for a pre-existing cDNA encoding the human
variable domain. The method starts instead with the rodent cDNA
sequence. Also, splicing framework regions is technically easier
because there is a high degree of homology between the mutagenic
oligonucleotide and the rodent variable domain framework. This is
true because a human antibody variable domain framework has been
selected that is most homologous to that of the rodent.
[0040] The advantage of the present method of reshaping as opposed
to synthesizing the entire reshaped version from scratch is that it
is technically easier. Synthesizing a reshaped variable domain from
scratch requires several more oligonucleotides, several days more
work, and technical difficulties are more likely to arise.
Step 4: The Transfection and Expression of the Reshaped
Antibody
[0041] Following the mutagenesis reactions to reshape the antibody,
the cDNAs are linked to the appropriate DNA encoding light or heavy
chain constant region, cloned into an expression vector, and
transfected into mammalian cells. These steps can be carried out in
routine fashion. A reshaped antibody may therefore be prepared by a
process comprising:
[0042] a) preparing a first replicable expression vector including
a suitable promoter operably linked to a DNA sequence which encodes
at least a variable domain of an Ig heavy or light chain, the
variable domain comprising framework regions from a first antibody
and CDRs comprising at least parts of the CDRs from a second
antibody of different specificity;
[0043] b) if necessary, preparing a second replicable expression
vector including a suitable promoter operably linked to a DNA
sequence which encodes at least the variable domain of a
complementary Ig light or heavy chain respectively;
[0044] c) transforming a cell line with the first or both prepared
vectors; and
[0045] d) culturing said transformed cell line to produce said
altered antibody.
[0046] Preferably the DNA sequence in step a) encodes both the
variable domain and the or each constant domain of the antibody
chain, the or each constant domain being derived from the first
antibody. The antibody can be recovered and purified. The cell line
which is transformed to produce the altered antibody may be a
Chinese Hamster Ovary (CHO) cell line or an immortalised mammalian
cell line, which is advantageously of lymphoid origin, such as a
myeloma, hybridoma, trioma or quadroma cell line. The cell line may
also comprise a normal lymphoid cell, such as a B-cell, which has
been immortalised by transformation with a virus, such as the
Epstein-Barr virus. Most preferably, the immortalised cell line is
a myeloma cell line or a derivative thereof.
[0047] Although the cell line used to produce the altered antibody
is preferably a mammalian cell line, any other suitable cell line,
such as a bacterial cell line or a yeast cell line, may
alternatively be used. In particular, it is envisaged that E.
coli-derived bacterial strains could be used.
[0048] It is known that some immortalised lymphoid cell lines, such
as myeloma cell lines, in their normal state secrete isolated Ig
light or heavy chains. If such a cell line is transformed with the
vector prepared in step (a) it will not be necessary to carry out
step (b) of the process, provided that the normally secreted chain
is complementary to the variable domain of the Ig chain encoded by
the vector prepared in step (a).
[0049] However, where the immortalised cell line does not secrete
or does not secrete a complementary chain, it will be necessary to
carry out step (b). This step may be carried out by further
manipulating the vector produced in step (a) so that this vector
encodes not only the variable domain of an altered antibody light
or heavy chain, but also the complementary variable domain.
[0050] Alternatively, step (b) is carried out by preparing a second
vector which is used to transform the immortalised cell line. This
alternative leads to easier construct preparation, but may be less
preferred than the first alternative in that it may not lead to as
efficient production of antibody.
[0051] In the case where the immortalised cell line secretes a
complementary light or heavy chain, the transformed cell line may
be produced for example by transforming a suitable bacterial cell
with the vector and then fusing the bacterial cell with the
immortalised cell line by spheroplast fusion. Alternatively, the
DNA may be directly introduced into the immortalised cell line by
electroporation or other suitable method.
[0052] An antibody is consequently produced in which CDRs of a
variable domain of an antibody chain are homologous with the
corresponding CDRs of an antibody of a first mammalian species and
in which the framework of the variable domain and the constant
domains of the antibody are homologous with the corresponding
framework and constant domains of an antibody of a second,
different, mammalian species. Typically, all three CDRs of the
variable domain of a light or heavy chain are derived from the
first species.
[0053] The present process has been applied to obtain an antibody
against human CD4 antigen. Accordingly, the invention also provides
an antibody which is capable of binding to human CD4 antigen, in
which the CDRs of the light chain of the antibody-have the amino
acid sequences: TABLE-US-00001 CDR1: LASEDIYSDLA CDR2: NTDTLQN
CDR3: QQYNNYPWT,
[0054] in which the CDRs of the heavy chain of the antibody have
the amino acid sequences: TABLE-US-00002 CDR1: NYGMA CDR2:
TISHDGSDTYFRDSVKG CDR3: QGTIAGIRH, and
in which the framework of the variable domain and, if present, the
or each constant domain of each chain are derived from a mammalian
non-rat species.
[0055] The antibody preferably has the structure of a natural
antibody or a fragment thereof. The antibody may therefore comprise
a complete antibody, a (Fab').sub.2 fragment, a Fab fragment, a
light chain dimer or a heavy chain.
[0056] The antibody may be an IgG such as IgG1, IgG2, IgG3 or IgG4
IgM, IgA, IgE or IgD. Alternatively, the antibody may be a
chimaeric antibody of the type described in WO 86/01533.
[0057] A chimaeric antibody according to WO 86/01533 comprises an
antigen binding region and a non-immunoglobulin region. The antigen
binding region is an antibody light chain variable domain or heavy
chain variable domain. Typically the chimaeric antibody comprises
both light and heavy chain variable domains. The non-immunoglobulin
region is fused at its C-terminus to the antigen binding region.
The non-immunoglobulin region is typically a non-immunoglobulin
protein and may be an enzyme region, a region derived from a
protein having known binding specificity, from a protein toxin or
indeed from any protein expressed by a gene. The two regions of the
chimaeric antibody may be connected via a cleavable linker
sequence.
[0058] The invention is preferably employed to humanise a CD4
antibody such as a rat or mouse CD4 antibody. The framework and the
constant domains of the resulting antibody are therefore human
framework and constant domains whilst the CDRs of the light and/or
heavy chain of the antibody are rat or mouse CDRs. Preferably all
CDRs are rat or mouse CDRs. The antibody may be a human IgG such as
IgG1, IgG2, IgG3, IgG4; IgM; IgA; IgE or IgD carrying rat or mouse
CDRs.
[0059] Preferably the framework of the antibody heavy chain is
homologous to the corresponding framework of the human antibody KOL
(Schmidt et al, Hoppe-Seyler's Z. Physiol. Chem., 364 713-747,
1983). The sixth residue of framework 4 in this case is suitably
Thr or Pro, preferably Thr. This residue is the 121st residue in
the KOL antibody heavy chain variable region (Schmidt et al, 1983),
and is identified as residue 108 by Kabat (Kabat et al, "Sequences
of proteins of immunological interest", US Dept of Health and Human
Services, US Government Printing Office, 1987). Alternatively, the
framework of the antibody heavy chain is homologous to the
corresponding framework of the human antibody NEW (Saul et al, J.
Biol. Chem. 253: 585-597, 1978). The final residue of framework 1
in this case is suitably Ser or Thr, preferably Ser. This residue
is at position 30 (Kabat et al, 1987). Preferably the framework of
the antibody light chain is homologous to the variable domain
framework of the protein REI (Epp et al, Eur. J. Biochem., 45,
513-524, 1974).
[0060] The framework regions of one or both chains of a CD4
antibody can be reshaped by the present process. Alternatively, one
or both chains of a CD4 antibody may be reshaped by the procedure
described in EP-A-0239400. The procedure of EP-A-0239400 involves
replacing CDRs rather than the replacement of frameworks. The CDRs
are grafted onto a framework derived from a mammalian non-rat
species, typically a human. This may be achieved by
oligonucleotide-directed in vitro mutagenesis of the CDR-encoding
regions of an antibody chain, light or heavy, from a mammalian
non-rat species. The oligonucleotides in such an instance are
selected so that the resulting CDR-grafted antibody has the light
chain CDRs 1 to 3 and the heavy chain CDRs 1 to 3 shown above.
[0061] The reshaped CD4 antibody can be used to induce tolerance to
an antigen. It can be used to alleviate autoimmune diseases such as
rheumatoid arthritis. It can be used to prevent graft rejection.
Tolerance to a graft such as an organ graft or a bone marrow
transplantation can be achieved. Also, the reshaped CD4 antibody
might be used to alleviate allergies. Tolerance to allergens could
be achieved.
[0062] The CD4 antibody may be depleting or non-depleting. A
depleting antibody is an antibody which depletes more than 50%, for
example from 90 to 99%, of target cells in vivo. A non-depleting
antibody depletes fewer than 50%, for example, from 10 to 25% and
preferably less than 10% of target cells in vivo. A CD4 antibody
may be administered alone or may be co-administered with a
non-depleting or depleting CD8 antibody. The CD4 antibody,
depleting or non-depleting, and CD8 monoclonal antibody, depleting
or non-depleting, may be administered sequentially in any order or
may be administered simultaneously. An additional antibody, drug or
protein may be administered before, during or after administration
of the antibodies.
[0063] A CD4 antibody and, indeed, a CD8 antibody as appropriate
are given parenterally, for example intravenously. The antibody may
be administered by injection or by infusion. For this purpose the
antibody is formulated in a pharmaceutical composition further
comprising a pharmaceutically acceptable carrier or diluent. Any
appropriate carrier or diluent may be employed, for example
phosphate-buffered saline solution.
[0064] The amount of non-depleting or depleting CD4 and, if
desired, CD8 antibody administered to a patient depends upon a
variety of factors including the age and weight of a patient, the
condition which is being treated and the antigen(s) to which it is
desired to induce tolerance. In a model mouse system from 1 .mu.g
to 2 mg, preferably from 400.mu.g to 1 mg, of a mAb is administered
at any one time. In humans from 3 to 500 mg, for example from 5 to
200 mg, of antibody may be administered at any one time. Many such
doses may be given over a period of several weeks, typically 3
weeks.
[0065] A foreign antigen(s) to which it is desired to induce
tolerance can be administered to a host before, during, or after a
course of CD4 antibody (depleting or non-depleting) and/or CD8
antibody (depleting or non-depleting). Typically, however, the
antigen(s) is administered one week after commencement of antibody
administration, and is terminated three weeks before the last
antibody administration.
[0066] Tolerance can therefore be induced to an antigen in a host
by administering non-depleting or depleting CD4 and CD8 mAbs and,
under cover of the mAbs, the antigen. A patient may be operated on
surgically under cover of the non-depleting or depleting CD4 and
CD8 mAbs to be given a tissue transplant such as an organ graft or
a bone marrow transplant. Also, tolerance may be induced to an
antigen already possessed by a subject. Long term specific
tolerance can be induced to a self antigen or antigens in order to
treat autoimmune disease such as multiple sclerosis or rheumatoid
arthritis. The condition of a patient suffering from autoimmune
disease can therefore be alleviated.
[0067] The following Example illustrates the invention. In the
accompanying drawings:
[0068] FIG. 1: shows the nucleotide and predicted amino acid
sequence of rat CD4 antibody light chain variable region. The
number of the first and last amino acid or nucleotide in each line
is indicated in the left and right margins, respectively. Base
pairs 1-269 (HindIII-PvuII) and 577-620 ([BglII/BclI]-BamHI) are
part of the vector M13V.sub.KPCR3, while base pairs 270-576 are
from the PCR product of the CD4 antibody light chain variable
region (V.sub.L). CDRs (boxes) were identified by comparison to
known immunological sequences (Kabat et al, "Sequences of proteins
of immunological interest, US Dept of Health and Human Services, US
Government Printing Office, 1987).
[0069] FIG. 2: shows the nucleotide and predicted amino acid
sequence of the reshaped CAMPATH-1 antibody light chain cDNA. The
number of the first and last amino acid or nucleotide in each line
is indicated in the left and right margins, respectively. CDRs are
identified by boxes.
[0070] FIG. 3: shows the nucleotide and predicted amino acid
sequence of the reshaped CD4 antibody light chain cDNA
CD4V.sub.LREI. The number of the first and last amino acid or
nucleotide in each line is indicated in the left and right margins,
respectively. CDRs are identified by boxes.
[0071] FIG. 4: shows the nucleotide and predicted amino acid
sequence of rat CD4 antibody heavy chain variable region. The
number of the first and last amino acid or nucleotide in each line
is indicated in the left and right margins, respectively. CDRs are
identified by boxes. Base pairs 1-272 (HindIII-PstI) and 603-817
(BstEII-BamHI) are part of the vector M13V.sub.HPCR1, while base
pairs 273-602 are from the PCR product of the CD4 antibody heavy
chain variable region (V.sub.H).
[0072] FIG. 5: shows the nucleotide and predicted amino acid
sequence of the reshaped CAMPATH-1 antibody heavy chain cDNA. The
number of the first and last amino acid or nucleotide in each line
is indicated in the left and right margins, respectively. CDRs are
identified by boxes.
[0073] FIG. 6: shows the nucleotide and predicted amino acid
sequence of the reshaped CD4 antibody heavy chain cDNA
CD4V.sub.HNEW-Thr.sup.30. The number of the first and last amino
acid or nucleotide in each line is indicated in the left and right
margins, respectively. CDRs are identified by boxes.
[0074] FIG. 7: shows the nucleotide and predicted amino acid
sequence of the reshaped CD4 antibody heavy chain cDNA
CD4V.sub.HNEW-Ser.sup.30. The number of the first and last amino
acid or nucleotide in each line is indicated in the left and right
margins, respectively. CDRs are identified by boxes.
[0075] FIG. 8: shows the heavy chain variable (V) region amino acid
sequence of the human myeloma protein KOL. CDRs are identified by
boxes. This sequence is taken from the Swiss-Prot protein sequence
database.
[0076] FIG. 9: shows the nucleotide and predicted amino acid
sequence of the reshaped CD4 antibody heavy chain V region
CD4V.sub.HKOL-Pro.sup.113. The number of the first and last amino
acid or nucleotide in each line is indicated in the left and right
margins, respectively. CDRs are identified by boxes.
[0077] FIG. 10: shows the nucleotide and predicted amino acid
sequence of the reshaped CD4 antibody heavy chain V region
CD4V.sub.HKOL-Pro.sup.113 without immunoglobulin promoter. The
number of the first and last amino acid or nucleotide in each line
is indicated in the left and right margins, respectively. CDRs are
identified by boxes.
[0078] FIG. 11: shows the nucleotide and predicted amino acid
sequence of the reshaped CD4 antibody heavy chain V region
CD4V.sub.HKOL-Thr.sup.113. The number of the first and last amino
acid or nucleotide in each line is indicated in the left and right
margins, respectively. CDRs are identified by boxes.
[0079] FIG. 12: shows the nucleotide and predicted amino acid
sequence of the reshaped CD4 antibody heavy chain V region
CD4V.sub.HKOL-Thr.sup.113 without immunoglobulin promoter. The
number of the first and last amino acid or nucleotide in each line
is indicated in the left and right margins, respectively. CDRs are
identified by boxes.
[0080] FIG. 13: shows the results of an ELISA that compares the
avidity of YNB46.1.8 and CD4V.sub.HKOL-Thr.sup.113 antibodies. The
X-axis indicates the concentration (.mu.g/ml) of YNB46.1.8
(triangles) or CD4V.sub.HKOL-Thr.sup.113 (circles) antibody. The
Y-axis indicates the optical density at 492 nanometers.
EXAMPLE
1. Materials and Methods
[0081] Isolation of monoclonal antibody. The rat-derived anti-human
CD4 antibody, clone YNB46.1.8 (IgG.sub.2b, kappa light chain
serotype), was the result of fusion between a rat splenocyte and
the Lou strain rat myeloma cell line Y3-Ag 1.2.3 (Galfre et al,
Nature, 277: 131-133, 1979) and was selected by its binding to a
rat T cell line NB2-6TG stably transfected with an expression
vector containing a complementary DNA (cDNA) encoding the human CD4
antigen (Madden et al, Cell, 42: 93-104, 1985). Antibody was
purified by high pressure liquid chromatography (HPLC).
[0082] Isolation of Antibody Variable Regions. cDNAs encoding the
V.sub.L and V.sub.H regions of the CD4 antibody were isolated by a
polymerase chain reaction (PCR)-based method (Orlandi et al, PNAS
USA, 86: 3833-3837, 1989) with some modifications. Total RNA was
isolated from hybridoma cells by the guanidine thiocyanate method
(Chirgwin et al, Biochemistry, 18: 5294, 1979), and poly(A).sup.+
RNA was isolated by passage of total RNA through and elution from
an oligo(dT)-cellulose column (Aviv and Leder PNAS USA 69: 1408,
1972). Poly(A).sup.+ RNA was heated at 70.degree. C. for 5 minutes
and cooled on ice just prior to use. A 25 .mu.l first strand
synthesis reaction consisted of 5 .mu.g poly(A).sup.+ RNA, 250
.mu.M each dNTP, 50 mM Tris.HCl (pH 8.2 at 42.degree. C.), 10 mM
MgCl.sub.2, 100 mM KCl, 10 mM dithiothreitol, 23 units reverse
transcriptase (Anglian Biotec, Colchester, U.K.), 3.5 pmoles of the
V.sub.L region-specific oligonucleotide primer V.sub.K1FOR
[5'-d(GTT AGA TCT CCA GCT TGG TCC C)] or the V.sub.H
region-specific primer V.sub.H1FOR-B [5,-d(TGA GGA GAC GGT GAC CGT
GGT CCC TTG GCC)], and incubated for 5 minutes at 20.degree. C. and
then 90 minutes at 42.degree. C.
[0083] Subsequent 50 .mu.l PCR amplifications consisted of 5 .mu.l
of the first strand synthesis reaction (unpurified), 500 .mu.M each
DNTP, 67 mM Tris-HCl (pH 8.8 at 25.degree. C.), 17 mM
(NH.sub.4).sub.2SO.sub.4, 10 mM MgCl.sub.2, 20 .mu.g/ml gelatin, 5
units TAQ DNA polymerase (Koch-Light, Haverhill, U.K.), and 25
pmoles (each) of primers V.sub.K1FOR and V.sub.K1BACK [5'-d(GAC ATT
CAG CTG ACC CAG TCT CCA)] for the V.sub.L region or V.sub.H1FOR-B
and the mixed primer V.sub.H1BACK [5'-d(AG GT(CG) (CA)A(GA) CTG CAG
(GC)AG TC(TA) GG)] for the V.sub.H region. Reactions were overlayed
with mineral oil and subjected to 30 cycles of 1.5 minutes at
95.degree. C. (denaturation), 1.5 minutes at 37.degree. C.
(V.sub.L) or 50.degree. C. (V.sub.H; annealing), and 3 minutes at
72.degree. C. (extension) with a Techne PHC-1 programmable cyclic
reactor. The final cycle contained a 10 minute extension time.
[0084] The samples were frozen at -20.degree. C. and the mineral
oil (a viscous liquid at -20.degree. C.) was removed by aspiration.
The aqueous phases were thawed, and PCR products were purified by
electrophoresis in 2% agarose gels, and then double digested with
either PvuII and BglII (V.sub.L) or PstI and BstEII (V.sub.H)
restriction enzymes, and cloned into the PvuII and BclI restriction
sites of the vector M13V.sub.KPCR3 (for V.sub.L region; Orlandi et
al, 1989) or the PstI and BstEII restriction sites of the vector
M13V.sub.HPCR1 (for V.sub.H region). As described in the results,
V.sub.L region clones were first screened by hybridisation to a
.sup.32P-labeled oligonucleotide probe [5'-d(GTT TCA TAA TAT TGG
AGA CA)] specific for the CDR2 of the Y3-Ag 1.2.3 V.sub.L region.
V.sub.L region clones not hybridising to this probe and V.sub.H
region clones were sequenced by the dideoxy chain termination
method (Sanger et al, PNAS USA 74: 5463, 1977).
[0085] Reshaped Light Chain Variable Region and Expression Vector
Construct.
[0086] The reshaped light chain was constructed by
oligonucleotide-directed in vitro mutagenesis in an M13 vector by
priming with three oligonucleotides simultaneously on a 748 base
single-stranded cDNA template encoding the entire V.sub.L and kappa
constant (C.sub.K) regions of the reshaped CAMPATH-1 antibody
(Reichmann at al, Nature 332: 323-327, 1988). The three
oligonucleotides [5'-d(AGA GTG ACC ATC ACC TGT CTA GCA AGT GAG GAC
ATT TAC AGT GAT TTA GCA TGG TAC CAG CAG AAG CCA), 5'-d (CTG CTG ATC
TAC AAT ACA GAT ACC TTG CAA AAT GGT GTG CCA AGC AGA TTC), 5'-d(ATC
GCC ACC TAC TAC TGC CAA CAG TAT AAC AAT TAT CCG TGG ACG TTC GGC CAA
GGG ACC)] were designed to replace each of the three CDRs in the
REI-based human antibody V.sub.L region framework that is part of
the reshaped CAMPATH-1 antibody V.sub.L region (Reichmann et al,
1988). A clone containing each of the three mutant oligonucleotides
was identified by nucleotide sequencing and was subcloned into the
HindIII site of the expression vector pH.beta.APr-1 (Gunning et al,
PNAS, 84: 4831-4835, 1987) which also contained a dihydrofolate
reductase gene (Ringold et al, J. Mol. Appl. Genet. 1: 165-175,
1981) driven by a truncated SV40 promoter.
[0087] Reshaped Heavy Chain Variable Regions Based on the Variable
Region Framework of the Human Antibody NEW, and Expression Vector
Constructs.
[0088] Two versions of the NEW-based reshaped heavy chain were
created, CD4V.sub.HNEW-Thr.sup.30 and CD4V.sub.HNEW-Ser.sup.30. The
CD4V.sub.HNEW-Thr.sup.30 version (FIG. 6) encodes a threonine
residue at position 30 while the CD4V.sub.HNEW-Ser.sup.30 version
(FIG. 7) encodes a Ser residue at position 30. As a matter of
convenience, CD4V.sub.HNEW-Thr.sup.30 was created first by
oligonucleotide-directed in vitro mutagenesis in the vector M13mp18
by priming with three oligonucleotides simultaneously on a 1467
base single-stranded cDNA template (FIG. 5) encoding the entire
heavy chain of the reshaped CAMPATH-1 antibody (Reichmann et al,
1988). The three oligonucleotides [5'-d(TCT GGC TTC ACC TTC ACC AAC
TAT GGC ATG GCC TGG GTG AGA CAG CCA CCT), 5'-d(GGT CTT GAG TGG ATT
GGA ACC ATT AGT CAT GAT GGT AGT GAC ACT TAC TTT CGA GAC TCT GTG AAG
GGG AGA GTG), 5'-d(GTC TAT TAT TGT GCA AGA CAA GGC ACT ATA GCT GGT
ATA CGT CAC TGG GGT CAA GGC AGC CTC)] were designed to replace each
of the three complementarity determining regions (CDRs) in the
NEW-based V.sub.H region that is part of the reshaped CAMPATH-1
antibody (Reichmann et al, 1988). A clone (FIG. 6) containing each
of the three mutant oligonucleotides was identified by nucleotide
sequencing. CD4V.sub.HNEW-Ser.sup.30 was created second by
oligonucleotide-directed in vitro mutagenesis in the vector M13mp18
by priming with a single oligonucleotide on the 1458 base
single-stranded cDNA template (FIG. 6) encoding
CD4V.sub.HNEW-Thr.sup.30. The oligonucleotide [5'-d(GCT TCA CCT TCA
GCA ACT ATG GCA T)] was designed to mutate the residue at position
30 from threonine [ACC] to serine [AGC]. A clone (FIG. 7)
containing this mutant oligonucleotide was identified by nucleotide
sequencing. Double-stranded forms of the clones
CD4V.sub.HNEW-Thr.sup.30 and CD4V.sub.HNEW-Ser.sup.30 were
subcloned as HindIII fragments into the HindIII site of the
expression vector pNH316. The vector pNH316 is a modified version
of the vector pH.beta.APr-1 (Gunning et al, PNAS, 84: 4831-4835,
1987) which was engineered to contain a neomycin resistance gene
driven by a metallothionine promoter.
[0089] Reshaped Heavy Chain Variable Regions Based on the Variable
Region Framework of the Human Antibody KOL, and Expression Vector
Constructs
[0090] Two versions of the KOL-based reshaped heavy chain were
created, CD4V.sub.HKOL-Thr.sup.113 and CD4V.sub.HKOL-Pro.sup.113.
The CD4V.sub.HKOL-Thr.sup.113 version encodes a threonine residue
at position 113 (FIG. 11) while the CD4V.sub.HKOL-Pro.sup.113
version encodes a proline residue at position 113 (FIG. 9). As a
matter of convenience, CD4V.sub.HKOL-Thr.sup.113 was created first
by oligonucleotide-directed in vitro mutagenesis of single-stranded
DNA template containing the 817 base HindIII-BamHI fragment
encoding the V.sub.H region of the rat CD4 antibody (FIG. 4) cloned
into M13mp18 by priming simultaneously with five oligonucleotides
[5'-d(CAC TCC CAG GTC CAA CTG GTG GAG TCT GGT GGA GGC GTG GTG CAG
CCT GG), 5'-d(AAG GTC CCT GAG ACT CTC CTG TTC CTC CTC TGG ATT CAT
CTT CAG TAA CTA TGG CAT G), 5'-d(GTC CGC CAG GCT CCA GGC AAG GGG
CTG GAG TGG), 5'-d(ACT ATC TCC AGA GAT AAT AGC AAA AAC ACC CTA TTC
CTG CAA ATG G), 5'-d(ACA GTC TGA GGC CCG AGG ACA CGG GCG TGT ATT
TCT GTG CAA GAC AAG GGA C)] which were designed to replace the rat
framework regions with the human framework regions of KOL. A clone
containing each of the five mutant oligonucleotides was identified
by nucleotide sequencing. CD4V.sub.HKOL-Pro.sup.113 was created
second by oligonucleotide-directed in vitro mutagenesis of
single-stranded DNA template containing the 817 base HindIII-BamHI
fragment encoding CD4V.sub.HKOL-Thr.sup.113 cloned into M13mp18 by
priming with the oligonucleotide [(5'-d(TGG GGC CAA GGG ACC CCC GTC
ACC GTC TCC TCA)]. A clone containing this mutant oligonucleotide
was identified by nucleotide sequencing.
[0091] The immunoglobulin promoters were removed from the
double-stranded DNA forms of clones encoding
CD4V.sub.HKOL-Thr.sup.113 (FIG. 11) and CD4V.sub.HKOL-Pro.sup.113
(FIG. 9) by replacing (for both versions) the first 125 bp
(HindIII-NcoI) with a HindIII-NcoI oligonucleotide linker fragment
[5'-d(AGC TTT ACA GTT ACT GAG CAC ACA GGA CCT CAC) and its
overlapping complement 5'-d(CAT GGT GAG GTC CTG TGT GCT CAG TAA CTG
TAA)]. The resultant clones, CD4V.sub.HKOL-Thr.sup.113 (FIG. 12)
and CD4V.sub.HKOL-Pro.sup.113 (FIG. 10), now 731 bp HindIII-BamHI
fragments, were separately subcloned into the HindIII and BamHI
cloning sites of the expression vector pH.beta.APr-1-gpt (Gunning
et al, PNAS USA 76, 1373, 1987) into which had been cloned the
human IgG1 constant region gene (Bruggemann et al, J. Exp. Med.
166, 1351-1361, 1987) at the BamHI site. Thus, when transfected and
expressed as antibody heavy chains (see below), these reshaped
V.sub.H regions are linked to human IgG1 constant regions.
[0092] Fluorescence Activated Cell Sorter (FACS) Analysis
[0093] The relative affinities of the reshaped antibodies to bind
the CD4 antigen were estimated by FACS analysis. The CD4-expressing
cells used in this analysis were a cloned rat T cell line NB2-6TG
stabily transfected with an expression vector containing a
complementary DNA (cDNA) encoding the human CD4 antigen (Maddon et
al, Cell, 42, 93-104, 1985). Cells were stained with the
appropriate reshaped antibody followed by fluorescein-conjugated
sheep anti-human antibodies (Binding Site Ltd., Birmingham, UK).
Control staining (see Table 1) consisted of no antibody present
during the first stage of cell staining. Mean cellular fluorescence
was determined with an Ortho FACS.
[0094] Antibody Avidity Analysis
[0095] The relative avidities of the rat YNB46.1.8 antibody and the
reshaped CD4V.sub.HKOL-Thr.sup.113 antibody were estimated by an
enzyme-linked immunosorbent assay (ELISA). Microtiter plates were
coated with soluble recombinant CD4 antigen (Byrn et al, Nature,
344: 667-670, 1990) at 50 ul/well, 10 ug/ml, and then blocked with
100 ul/well phosphate buffered saline (PBS) containing 1.0% bovine
serum albumin (BSA). Antibodies were diluted in PBS containing 0.1%
BSA, and added to wells (50 ul/well) for 45 minutes at room
temperature. Biotinylated CD4V.sub.HKOL-Thr.sup.113 antibody (10
ul/well; 20 ug/ml final concentration) was then added to each well
for an additional 45 minutes. Wells were washed with PBS containing
0.1% BSA, and then 50 ul streptavidin-biotinylated horseradish
peroxidase complex (Amersham; Aylesbury, UK) diluted 1:1,000 was
added to each well for 30 minutes. Wells were washed with PBS
containing 0.1% BSA, and 100 ul substrate (25 mM citric acid, 50 mM
disodium hydrogen phosphate, 0.1% (w/v) o-phenylene diamine, 0.04%
(v/v) 30% hydrogen peroxide) was added to each well. Reactions were
stopped by the addition of 50 ul/well 1.0 M sulfuric acid. Optical
densities at 492 nanometers (OD.sub.492) were determined with an
ELISA plate reader.
[0096] Transfections.
[0097] Dihydrofolate reductase deficient chinese hamster ovary
(CHO.sup.DHFR-) cells (10.sup.6/T-75 flask) were cotransfected as
described (Wigler et al, PNAS USA 76, 1373, 1979) with 9 .mu.g of
heavy chain construct and 1 .mu.g of the light chain construct.
Transfectants were selected in medium containing 5% dialysed foetal
bovine serum for 2 to 3 weeks, and antibody-secreting clones were
identified by ELISAs of conditioned media. Antibody was
concentrated and purified by protein-A Sepharose (Trade Mark)
column chromatography.
2. Results
[0098] Cloning of Light and Heavy Chain Variable Region cDNAs.
[0099] cDNAs encoding the V.sub.L and V.sub.H regions from CD4
antibody-secreting hybridoma cells were isolated by PCR using
primers which amplify the segment of mRNA encoding the N-terminal
region through to the J region (Orlandi et al, 1989). V.sub.L and
V.sub.H region PCR products were subcloned into the M13-based
vectors M13V.sub.KPCR3 and M13V.sub.HPCR1, respectively. Initial
nucleotide sequence analysis of random V.sub.L region clones
revealed that most of the cDNAs encoded the V.sub.L region of the
light chain expressed by the Y3-Ag 1.2.3 rat myeloma cell line
(Crowe et al, Nucleic Acid Research, 17: 7992, 1989) that was used
as the fusion partner to generate the anti-CD4 hybridoma. It is
likely that the expression of the Y3-Ag 1.2.3 light chain mRNA is
greater than that of the CD4 antibody light chain, or the Y3-Ag
1.2.3 light chain mRNA is preferentially amplified during the
PCR.
[0100] To maximize the chance of finding CD4 V.sub.L region cDNAs,
we first screened all M13 clones by hybridisation to a
.sup.32P-labeled oligonucleotide probe that is complementary to the
CDR 2 of Y3-Ag 1.2.3 (Crowe et al, Nucleic Acid Research, 17: 7992,
1989). Subsequent sequence analysis was restricted to M13 clones
which did not contain sequence complementary to this probe. In this
manner, two cDNA clones from independent PCR amplifications were
identified that encoded identical V.sub.L regions. Nucleotide
sequence analysis of random V.sub.H region PCR products revealed a
single species of V.sub.H region cDNA. Two V.sub.H cDNA clones from
independent PCR amplifications were found to contain identical
sequences except that the codon of residue 14 encoded proline [CCT]
in one clone while the second clone encoded leucine [CTT] at the
same position.
[0101] According to Kabat et al 1987, 524 of 595 sequenced V.sub.H
regions contain a proline residue at this position, while only 6
contain leucine. We have therefore chosen the proline-encoding
clone for illustration (see below). As residue 14 lies well within
the first V.sub.H framework region and not in a CDR, it is unlikely
to contribute directly to antigen binding, and the ambiguity at
this position did not affect the subsequent reshaping strategy.
Thus, we have not investigated this sequence ambiguity further.
[0102] The cDNA sequences and their predicted amino acid sequences
are shown in FIGS. 1 and 4. As no additional V.sub.L or V.sub.H
region-encoding clones were found, it was assumed that these
sequences were derived from the CD4 antibody genes.
[0103] Construction of Reshaped Antibodies.
[0104] Our goal was to investigate the importance of selecting the
appropriate human V region framework during reshaping. Two
reshaping strategies were employed.
[0105] First Reshaping Strategy.
[0106] In the first strategy, we created a reshaped antibody that
incorporated the CDRs from the rat-derived CD4 antibody and the
same human V region framework sequences that we had previously
successfully used for the reshaped CAMPATH-1 antibody, namely an
REI-based framework for the V.sub.L region and an NEW-based
framework for the V.sub.H region (Reichmann et al, 1988). This was
accomplished by oligonucleotide-directed in vitro mutagenesis of
the six CDRs of the reshaped CAMPATH-1 antibody light and heavy
chain cDNAs shown in FIGS. 2 and 5, respectively. The resultant
reshaped CD4 antibody light chain (FIG. 3) is called CD4V.sub.LREI.
Two versions of the NEW-based reshaped CD4 antibody heavy chain
were created: CD4V.sub.HNEW-Thr.sup.30 (FIG. 6) encoding a
threonine residue at position 30 (in framework 1) and
CD4V.sub.HNEW-Ser.sup.30 (FIG. 7) encoding a serine residue at
position 30. These two different versions were created because the
successfully reshaped CAMPATH-1 antibody heavy chain bound antigen
well whether position 30 encoded a threonine or serine residue
(Reichmann et al, 1988), and we chose to test both possibilities in
this case as well.
[0107] Second Reshaping Strategy
[0108] In the second reshaping strategy, we have reshaped the CD4
antibody V.sub.H region to contain the V.sub.H region framework
sequences of the human antibody KOL. Of all known human antibody
V.sub.H regions, the overall amino acid sequence of the V.sub.H
region of KOL is most homologous to the rat CD4 antibody V.sub.H
region. The V.sub.H regions of the human antibodies KOL and NEW are
66% and 42% homologous to the rat CD4 antibody V.sub.H region,
respectively.
[0109] Two versions of the KOL-based reshaped CD4 antibody heavy
chain V region were created that differ by a single amino acid
residue within the fourth framework region:
CD4V.sub.HKOL-Pro.sup.113 (FIG. 10) encodes a proline residue at
position 113 and CD4V.sub.HKOL-Thr.sup.113 (FIG. 12) encodes a
threonine residue at position 113. CD4V.sub.HKOL-Pro.sup.113 is
"true to form" in that its framework sequences are identical to
those of the KOL antibody heavy chain V region (FIG. 8).
[0110] Of all known human antibody V.sub.L regions, the overall
amino acid sequence of the V.sub.L region of the human light chain
NEW is most homologous (67%) to the rat CD4 antibody V.sub.L
region. Thus, the identical reshaped light chain, CD4V.sub.LREI
(described above), that was expressed with the NEW-based reshaped
CD4 antibody heavy chains CD4V.sub.HNEW-Thr.sup.30 and
CD4V.sub.HNEW-Ser.sup.30, is also expressed with the KOL-based
reshaped CD4 antibody heavy chains CD4V.sub.HKOL-Pro.sup.113 and
CD4V.sub.HKOL-Thr.sup.113. This is advantageous because expression
of the same reshaped light chain with different reshaped heavy
chains allows for a direct functional comparison of each reshaped
heavy chain.
[0111] To summarise, four different reshaped antibodies were
created. The reshaped light chain of each antibody is called
CD4V.sub.LREI. The reshaped heavy chains of the antibodies are
called CD4V.sub.HNEW-Thr.sup.30, CD4V.sub.HNEW-Ser.sup.30,
CD4V.sub.HKOL-Pro.sup.113 and CD4V.sub.HKOL-Thr.sup.113,
respectively. Each of the reshaped heavy chains contain the same
human IgG1 constant region. As each reshaped antibody contains the
same reshaped light chain, the name of a reshaped antibody's heavy
chain shall be used below to refer to the whole antibody (heavy and
light chain combination).
[0112] Relative Affinities of the Reshaped Antibodies
[0113] The relative affinities of the reshaped antibodies were
approximated by measuring their ability to bind to CD4
antigen-expressing cells at various antibody concentrations. FACS
analysis determined the mean cellular fluorescence of the stained
cells (Table 1).
[0114] It is clear from this analysis that the reshaped CD4
antibodies bind to CD4 antigen to varying degrees over a broad
concentration range. Consider Experiment 1 of Table 1 first.
Comparing CD4V.sub.HKOL-Thr.sup.113 antibody to
CD4V.sub.HNEW-Thr.sup.30 antibody, it is clear that both antibodies
bind CD4.sup.+ cells when compared to the control, reshaped
CAMPATH-1 antibody. However, CD4V.sub.HKOL-Thr.sup.113 antibody
binds CD4.sup.+ cells with far greater affinity than
CD4V.sub.HNEW-Thr.sup.30 antibody. The lowest concentration of
CD4V.sub.HKOL-Thr.sup.113 antibody tested (2.5 ug/ml) gave a mean
cellular fluorescence nearly equivalent to that of the highest
concentration of CD4V.sub.HNEW-Thr.sup.30 antibody tested (168
ug/ml). Experiment 2 demonstrates that CD4V.sub.HNEW-Ser.sup.30
antibody may bind CD4.sup.+ cells somewhat better than
CD4V.sub.HNEW-Thr.sup.30. Only 2.5 ug/ml CD4V.sub.HNEW-Ser.sup.30
antibody is required to give a mean cellular fluorescence nearly
equivalent to 10 ug/ml CD4V.sub.HNEW-Thr.sup.30 antibody.
Experiment 3 demonstrates that CD4V.sub.HKOL-Thr.sup.113 antibody
may bind CD4.sup.+ cells somewhat better than
CD4V.sub.HKOL-Pro.sup.113 antibody.
[0115] From these assays, it is clear that the KOL-based reshaped
antibodies are far superior to the NEW-based reshaped antibodies
with regards to affinity towards CD4.sup.+ cells. Also, there is a
lesser difference, if any, between CD4V.sub.HNEW-Thr.sup.30
antibody and CD4V.sub.HNEW-Ser.sup.30 antibody, and likewise
between CD4V.sub.HKOL-Thr.sup.113 antibody and
CD4V.sub.HKOL-Pro.sup.113 antibody. A ranking of these reshaped
antibodies can thus be derived based on their relative affinities
for CD4.sup.+ cells:
CD4V.sub.HKOL-Thr.sup.113>CD4V.sub.HKOL-Pro.sup.113>>CD4V.sub.HN-
EW-Ser.sup.30>CD4V.sub.HNEW-Thr.sup.30
[0116] It should be restated that each of the reshaped CD4
antibodies used in the above experiments have the identical heavy
chain constant regions, and are associated with identical reshaped
light chains. Thus observed differences of binding to CD4.sup.+
cells must be due to differences in their heavy chain V
regions.
[0117] Relative Avidities of the Rat YNB46.1.8 Antibody and the
Reshaped CD4V.sub.HKOL-Thr.sup.113 Antibody
[0118] The relative avidities of the rat YNB46.1.8 antibody and the
reshaped CD4V.sub.HKOL-Thr.sup.113 antibody were estimated by
ELISA. In this assay, the ability of each antibody to inhibit the
binding of biotinylated CD4V.sub.HKOL-Thr.sup.113 antibody to
soluble recombinant CD4 antigen was determined. Results of an
experiment are shown in FIG. 13. The inhibition of binding of
biotinylated CD4V.sub.HKOL-Thr.sup.113 antibody was linear for both
the unlabeled CD4V.sub.HKOL-Thr.sup.113 and YNB46.1.8 antibodies
near the optical density of 0.3. The concentrations of
CD4V.sub.HKOL-Thr.sup.113 and YNB46.1.8 antibodies that give an
optical density of 0.3 are 28.7 and 1.56 ug/ml, respectively. Thus
the avidity of the YNB46.1.8 antibody can be estimated to be
28.7/1.56 or about 18 times better than that of
CD4V.sub.HKOL-Thr.sup.113 antibody. It should be noted that this
experiment only provides a rough approximation of relative
avidities, not affinities. The rat YNB46.1.8 antibody contains a
different constant region than that of the
CD4V.sub.HKOL-Thr.sup.113 antibody, and this could affect how well
the antibodies bind CD4 antigen, irrespective of their actual
affinities for CD4 antigen. The actual affinity of the reshaped
antibodies for CD4 antigen may be greater, lesser, or the same as
the YNB46.1.8 antibody. The other reshaped antibodies
CD4V.sub.HKOL-Pro.sup.113, CD4V.sub.HNEW-Ser.sup.30, and
CD4V.sub.HNEW-Thr.sup.30 have not yet been tested in this assay.
TABLE-US-00003 TABLE 1 Mean cellular fluorescence of CD4.sup.+
cells stained with reshaped antibodies Concentration Mean cellular
Reshaped Antibody (.mu.g/ml) Fluorescence Experiment 1.
CD4V.sub.HKOL-Thr.sup.113 113 578.0 CD4V.sub.HKOL-Thr.sup.113 40
549.0 CD4V.sub.HKOL-Thr.sup.113 10 301.9 CD4V.sub.HKOL-Thr.sup.113
2.5 100.5 CD4V.sub.HNEW-Thr.sup.30 168 97.0
CD4V.sub.HNEW-Thr.sup.30 40 40.4 CD4V.sub.HNEW-Thr.sup.30 10 18.7
CD4V.sub.HNEW-Thr.sup.30 2.5 10.9 CAMPATH-1 100 11.6 CAMPATH-1 40
9.4 CAMPATH-1 10 9.0 CAMPATH-1 2.5 8.6 CONTROL -- 9.0 Experiment 2.
CD4V.sub.HNEW-Thr.sup.30 168 151.3 CD4V.sub.HNEW-Thr.sup.30 40 81.5
CD4V.sub.HNEW-Thr.sup.30 10 51.0 CD4V.sub.HNEW-Thr.sup.30 2.5 39.3
CD4V.sub.HNEW-Ser.sup.30 160 260.2 CD4V.sub.HNEW-Ser.sup.30 40
123.5 CD4V.sub.HNEW-Ser.sup.30 10 68.6 CD4V.sub.HNEW-Ser.sup.30 2.5
49.2 CONTROL -- 35.8 Experiment 3. CD4V.sub.HKOL-Pro.sup.113 100
594.9 CD4V.sub.HKOL-Pro.sup.113 40 372.0 CD4V.sub.HKOL-Pro.sup.113
10 137.7 CD4V.sub.HKOL-Pro.sup.113 2.5 48.9
CD4V.sub.HKOL-Thr.sup.113 100 696.7 CD4V.sub.HKOL-Thr.sup.113 40
631.5 CD4V.sub.HKOL-Thr.sup.113 10 304.1 CD4V.sub.HKOL-Thr.sup.113
2.5 104.0 CONTROL -- 12.3
[0119]
Sequence CWU 1
1
* * * * *