U.S. patent application number 14/683835 was filed with the patent office on 2015-12-17 for hla-e binding.
This patent application is currently assigned to ISIS INNOVATION LIMITED. The applicant listed for this patent is ISIS INNOVATION LIMITED. Invention is credited to David S.J. Allan, Veronique M. Braud, Andrew J. McMichael, Christopher A. O'Callaghan, Graham S. Ogg.
Application Number | 20150361180 14/683835 |
Document ID | / |
Family ID | 10823165 |
Filed Date | 2015-12-17 |
United States Patent
Application |
20150361180 |
Kind Code |
A1 |
Braud; Veronique M. ; et
al. |
December 17, 2015 |
HLA-E BINDING
Abstract
The invention relates to a method of testing a compound for
biological activity, which method comprises providing cells
expressing one of the CD94/NKG2 family of receptors, contacting the
cells with recombinant HLA-E under binding conditions in the
presence of the test compound, and determining whether the presence
of the compound affects the binding of HLA-E to the cells. The
HLA-E property of binding to CD94/NKG2 receptors on NK cells and a
subset of CD8+ T cells is useful for targeting CD94/NKG2+ cells for
a variety of purposes such as identification, isolation, killing or
inactivation.
Inventors: |
Braud; Veronique M.;
(Shrivenham, GB) ; Allan; David S.J.; (Oxford,
GB) ; Ogg; Graham S.; (Oxford, GB) ;
O'Callaghan; Christopher A.; (Pasadena, CA) ;
McMichael; Andrew J.; (Beckley, Oxford, GB) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
ISIS INNOVATION LIMITED |
OXFORD |
|
GB |
|
|
Assignee: |
ISIS INNOVATION LIMITED
OXFORD
GB
|
Family ID: |
10823165 |
Appl. No.: |
14/683835 |
Filed: |
April 10, 2015 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12109990 |
Apr 25, 2008 |
|
|
|
14683835 |
|
|
|
|
09555555 |
Sep 25, 2000 |
7410767 |
|
|
PCT/GB98/03686 |
Jun 1, 2000 |
|
|
|
12109990 |
|
|
|
|
Current U.S.
Class: |
800/13 ;
435/375 |
Current CPC
Class: |
A61P 15/08 20180101;
C07K 2317/76 20130101; A61P 31/18 20180101; A61P 31/12 20180101;
C07K 16/2896 20130101; C07K 16/2851 20130101; G01N 2333/70539
20130101; A61P 37/02 20180101; G01N 33/56977 20130101; A61P 37/06
20180101; A61P 15/00 20180101; G01N 2333/7056 20130101; A61P 31/00
20180101; A61P 37/04 20180101; A61P 31/22 20180101; A61P 35/02
20180101; A61P 35/00 20180101; A61P 37/00 20180101 |
International
Class: |
C07K 16/28 20060101
C07K016/28 |
Foreign Application Data
Date |
Code |
Application Number |
Dec 4, 1997 |
GB |
9725764.6 |
Claims
1. An in vitro method of protecting a cell or inhibiting HLA-E
mediated inhibition of NK cell activity comprising: (a) selecting
an antibody that interferes with HLA-E-binding to an inhibitory
CD94/NKG2 receptor from the group consisting of an anti-CD94
antibody, an anti-NKG2A antibody and an anti-HLA-E antibody; and
(b) contacting an NK or T cell expressing an inhibitory CD94/NKG2
receptor with the antibody in vitro.
2. The method of claim 1, wherein the inhibitory CD94/NKG2-receptor
is NKG2A.
3. The method of claim 1, wherein the antibody is an anti-CD94
antibody.
4. The method of claim 1, wherein the antibody is an anti-NKG2A
antibody.
5. The method of claim 1, wherein the antibody is an anti-HLA-E
antibody.
6. The method of claim 1, wherein the compound is a multimer of
HLA-E comprising two or more HLA-E molecules.
7. The method of claim 6, wherein the multimer has an enhanced
binding capability compared to non-multimeric HLA-E.
8. The method of claim 6, wherein the multimer comprises
recombinant soluble HLA-E molecules.
9. The method of claim 8, wherein the recombinant soluble HLA-E
molecules are attached via a linker molecule.
10. An in vitro method of preventing HLA-E medicated inhibition of
NK cell activity, said method comprising: (a) selecting an
anti-NKG2A antibody; and (b) contacting an NK or T cell expressing
an inhibitory CD94/NKG2 receptor with the anti-NKG2A antibody in
vitro.
11. A transgenic animal comprising a recombinant DNA encoding HLA-E
having an HLA-B8 leader sequence.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of U.S. patent
application Ser. No. 12/109,990, which was filed Apr. 25, 2008 and
is a continuation of U.S. patent application Ser. No. 09/555,555
(issued on Aug. 12, 2008 as U.S. Pat. No. 7,410,767), which was
filed Sep. 25, 2000 and was a section 371 National Stage of
PCT/GB98/03686, which was filed Dec. 4, 1998 and claimed priority
to GB9725764.6, which was filed Dec. 4, 1997, all which are
incorporated herein by reference as if fully set forth.
FIELD OF INVENTION
[0002] This invention relates to methods of identifying, targeting
and of isolating a group of CD94+ cells, in particular a group of
CD94+ cells including natural killer (NK) cells and a subset of T
cells. The invention also relates to methods of targeting
functional moieties such as toxins to the CD94+ cells. The
invention further relates to multimeric complexes of HLA-E for use
in the methods.
BACKGROUND
[0003] Human leukocyte antigen-E (HLA-E) is a nonclassical MHC
class Ib molecule of very limited polymorphism. Human nonclassical
MHC class Ib molecules (which include HLA-E, HLA-F and HLA-G) are
homologous to classical MHC class Ia molecules (HLA-A, HLA-B and
HLA-C) but are characterised by limited polymorphism and low cell
surface expression (reviewed in Shawar et al 1994 Annu. Rev.
Immunol. 12: 839). The mouse MHC class Ib molecule Qa-1 shares some
characteristics with HLA-E in that it displays a broad tissue
distribution and some structural similarities in the
peptide-binding groove (reviewed in Soloski et al 1995 Immunol.
Rev. 147: 67).
[0004] Whereas the function of the classical MHC class I molecules
in presenting peptides derived from cytosolic proteins to CD8+ T
cells is well-established, the function of nonclassical MHC
molecules remains unknown, in particular for HLA-E. HLA-E is
transcribed in most tissues. Recently, we have shown that HLA-E is
able to bind in its peptide-binding site peptides derived from MHC
class I leader sequences at positions 3 to 11 (Braud et al 1997
Eur. J. Immunol. 27: 1164-1169). The optimum binding peptide is a
nonamer. Using alanine and glycine substitutions, it was
established that there are primary anchor residues at positions 2
and 9 of the peptide and secondary anchor residues at position 7
and possibly position 3. The literature suggests that HLA-E is
localised in the endoplasmic reticulum (ER) and might have a role
in the loading of peptides onto classical MHC class I molecules in
a similar way to HLA-DM for MHC class II molecules. Neither mouse
cells transfected with HLA-E and human .beta.2microglobulin
(.beta.2m) nor the 721.221 cell line which only expresses HLA-E and
HLA-F, show surface expression of HLA-E (Ulbrecht et al, J.
Immunol. 1992 149: 2945-2953 and J. Exp. Med. 1992 176:
1083-1090).
[0005] Assembly of MHC class I molecules occurs in the endoplasmic
reticulum (ER) and requires peptide translocation through the
transporter associated with antigen processing (TAP) (reviewed in
Cerundolo et all 1996 In HLA and MHC genes, molecules and function,
Edited by Browning M, McMichael A. Oxford: Bios. Scientific
Publisher Ltd; 193-223). In human cells, newly synthesized MHC
class I heavy chains associate with calnexin which is later
displaced by the association of .beta.2m. Following dissociation of
calnexin, class I-.beta.2m heterodimers are stably associated with
another ER resident protein, calreticulin. Another molecule,
tapasin, which is associated with TAP and with MHC class I
calreticulin complexes, acts as a bridge between them. MHC class I
association with TAP facilitates peptide binding and the class I
molecules are released and exported to the cell surface upon stable
loading of peptide.
[0006] Natural killer (NK) cells are cytotoxic cells which have the
morphology of large granular lymphocytes and are normally defined
by their activity. They use recognition systems which are not yet
clearly understood. Recognition of tumour cell lines and
virally-infected cells is however driven by the absence of MHC
class I at the target cell surface (some MHC class I molecules
interact with specific NK receptors). NK cell cytotoxicity is
mediated by the interaction between Fas and Fas ligand or by the
release of the contents of the intracellular granules including the
pore forming protein perforin and the serine protease granzyme B.
NK cells are generally but not exclusively CD3- and CD56+. They may
also be CD16+ and some are also CD8+. Certain CD8+ T cells have an
NK cell-like function, in that they are able to kill MHC class I
negative cells.
[0007] NK cells express receptors that interact with MHC class I
and serve to inhibit or activate NK cell-mediated cytotoxicity. The
killer cell immunoglobulin-like receptors (KIR), which are members
of the immunoglobulin superfamily, make up one such group of
receptors.
[0008] Another NK cell inhibitory receptor which has a similar
effect is the CD94/NKG2A receptor from the C-type lectin
superfamily, which is expressed at the cell surface as a
heterodimer of CD94 covalently associated with NKG2A. CD94 also
associates with other members of the NKG2 family, which consists of
four closely related molecules NKG2A, B, C and E and two more
distantly related molecules NKG2D and F. CD94/NKG2A and B are both
inhibitory NK cell receptors which interact with MHC class I to
inhibit NK cell lysis, while CD94/NKG2C is a stimulatory NK cell
receptor which interacts with MHC class I to perform an NK cell
triggering function.
[0009] The CD94/NKG2C activator receptor contains a third subunit
DAP12, which is expressed as a disulphide-bonded homodimer and
interacts with NKG2C via charged residues in the transmembrane
domains (Lanier, 1998, Nature 391:703; Lanier et al, 1998, Immunity
8:693). DAP12 is necessary for efficient transport of the
CD94/NKG2C complex to the cell surface. Ligation of CD94 on
CD94/NKG2C/DAP12 transfectants causes tyrosine phosphorylation of
DAP12, suggesting that it induces cellular activation via
DAP12.
[0010] Four genes encode the NKG2 glycoproteins: NKG2A, NKG2C,
NKG2E and NKG2D/F (Houchins et al, 1991, J. Exp. Med. 173:1017;
Plougastel et al, 1997, Eur. J. Immunol. 27: 2835). NKG2A and B are
alternative splicing products (differing by an 18 amino acid
segment immediately outside the transmembrane region). NKG2C is
highly homologous to NKG2A and B with 94% homology in the external
C terminal domain and 56% homology through the internal and
transmembrane regions. NKG2D is distantly but significantly related
(21% homology) as too is NKG2F.
[0011] It has recently been discovered that HLA-E is stably
expressed at a low level on the surface of cells. Its expression at
the surface correlates with co-expression of human MHC class I
molecules which possess a peptide in their leader sequence capable
of binding to HLA-E. Loading of these signal sequence-derived
peptides is TAP and Tapasin-dependent and HLA-E assembly appears to
be similar to classical MHC class I assembly (Braud et al 1998
Current Biology 8:1-10).
[0012] It has now also been discovered that HLA-E binds to NK cells
expressing receptors CD94/NKG2A, B and C. The majority of NK cells
express CD94/NKG2 receptors and the majority of NK cells are
capable of binding HLA-E. HLA-E also binds to a small subset of T
cells expressing CD94/NKG2 heterodimers.
[0013] Furthermore, surface expression of HLA-E provides protection
against killing by CD94/NKG2A+ NK cells.
[0014] In addition, it has been discovered that multimeric HLA-E
molecules bind strongly to NK cells and the T cell subset
expressing CD94/NKG2.
SUMMARY
[0015] In its broadest sense, the invention provides the use of the
interaction between HLA-E and NK cells and/or a subset of T cells,
to identify and/or target and/or isolate those cells; and HLA-E in
a suitable form for such use.
[0016] The invention provides in one aspect a method of causing an
interaction of CD94/NKG2+ cells, which method comprises contacting
the cells with HLA-E under binding conditions.
[0017] The invention thus encompasses in one embodiment a method of
identifying the presence of CD94/NKG2+ NK cells and T cells in a
sample, which method comprises contacting the sample with HLA-E
under suitable binding conditions and detecting binding of HLA-E to
the cells.
[0018] In another embodiment, the method according to the invention
provides a method of selecting for CD94/NKG2+ cells, in particular
NK cells and a subset of T cells, from a sample, which method
comprises contacting the sample with HLA-E under binding conditions
and separating cells bound to the HLA-E from the mixture.
[0019] In still another embodiment there is provided a method of
killing or inactivating NK cells and a subset of T cells, which
method comprises contacting the cells with HLA-E under binding
conditions and carrying out targeted killing on the bound cells.
Any targeted killing method may be used, for example NK cells may
be identified by detecting bound HLA-E, and then destroyed by use
of a laser, or the HLA-E may carry a toxic moiety which kills or
inactivates the cells to which the HLA-E binds.
[0020] In yet another embodiment, the invention provides a method
of modifying NK cell activity against a potential target cell, by
expressing HLA-E at the surface of the target cell. In this
embodiment, the binding of the CD94/NKG2 receptors to the cell
surface HLA-E thus causes an interaction between the CD94/NKG2+
cells and the HLA-E bearing cells. In the case of CD94/NKG2A or B
receptors, this will be an inhibitory interaction the effect of
which is to protect the HLA-E bearing cell from killing by the NK
cells. In the case of CD94/NKG2C receptors there may be an NK cell
stimulatory effect. In the case of NK cells expressing both
inhibitory and activator CD94/NKG2 receptors, the overall effect of
an interaction with HLA-E at the target cell surface is an
inhibitory one, since the inhibitory receptors override the
stimulatory receptors.
[0021] Thus, the invention involves the use of the newly discovered
HLA-E-CD94/NKG2 receptor binding partnership for a variety of
possible purposes. The interaction in the method according to the
invention may be simply the binding of the CD94/NKG2 receptors to
the HLA-E. Alternatively or additionally the interaction of the
CD94/NKG2+ cells with the HLA-E may give rise to an effect on the
activity of the CD94/NKG2+ cells, such as an inhibitory effect.
[0022] In further aspects, the invention provides CD94/NKG2+ cells
isolated by the method according to the invention; and a population
of cells depleted of CD94/NKG2+ cells by the method according to
the invention.
[0023] In another aspect the invention provides a non-human
mammalian cell which expresses HLA-E at the cell surface by virtue
of a nucleic acid encoding HLA-E integrated into the genome of the
cell. The nucleic acid encoding HLA-E is a heterologous nucleic
acid in the sense that it is not found in those cells in nature.
The invention also provides recombinant animals comprising such
cells, which animals include transgenic animals which contain
HLA-E-encoding nucleic acid material in their somatic and germ
cells, as well as animals which are recipients of a transplant from
such transgenic animals.
[0024] In another aspect the invention provides a method of testing
a compound for biological activity, which method comprises: [0025]
(i) providing cells expressing CD94/NKG2 receptors at the cell
surface; [0026] (ii) contacting the cells with HLA-E in the
presence of the test compound; and [0027] (iii) determining whether
the presence of the compound affects the binding of HLA-E to the
cells; and compounds identified by the method as being compounds
which affect the binding of HLA-E to CD94/NKG2 receptors.
[0028] Preferably the cells expressing CD94/NKG2 receptors in the
method according to this aspect of the invention do not naturally
express the receptors, and most preferably they are non-human
cells. The cells are preferably stable transfectants, that is to
say they contain nucleic acid material expressing CD94/NKG2 stably
integrated into their genome.
[0029] Compounds such as antibodies, in particular monoclonal
antibodies, may be screened by the method for a particular desired
property. Compounds may be identified by the method which interfere
specifically with the interaction between HLA-E and CD94/NKG2A but
not CD94/NKG2C, or vice versa. Antibodies with such specificity
will be useful, for example to enable CD94/NKG2A NK cells to be
distinguished from CD94/NKG2C cells. Therapeutic uses for
antibodies which inhibit the binding of HLA-E to CD94/NKG2
receptors are also envisaged, for example in bone marrow
transplantation. It can be difficult to find a matched human donor
and allogeneic cells might not possess MHC class I ligands that can
engage the inhibitory receptors of all NK cells. An antibody which
specifically blocks the binding of NK cells to the activator
receptor CD94/NKG2C will therefore be useful. Another example of a
possible therapeutic use of the antibodies is in the treatment of
certain autoimmune diseases.
[0030] In another aspect the invention provides a multimer of HLA-E
comprising two or more HLA-E molecules, said multimer having
enhanced binding capability compared to non-multimeric HLA-E,
optionally labeled with a signal moiety. The multimers of HLA-E
contain at least two subunits which have the binding properties of
HLA-E receptors (that is they bind CD94/NKG2 receptors), linked
together to produce a bi-functional or multifunctional species.
Each subunit comprises all or a substantial part of the
extracellular region of HLA-E, generally at least the .alpha.1,
.alpha.2 and .alpha.3 HLA-E domains, together with P2
microglobulin, and a suitable peptide in the peptide binding
groove. Preferred HLA-E multimers are tetramers, but other
multimers for example dimers, trimers and multimers containing 5,
6, 7 etc. HLA-E molecules are not excluded.
[0031] In another aspect the invention provides a recombinant HLA-E
coupled to a toxic agent. The purpose of the toxic agent is to kill
or inactivate cells to which the recombinant HLA-E binds. It is
sufficiently toxic for that purpose. The toxic agent preferably has
a localised effect, that is it preferably does not affect
surrounding cells to which the HLA-E is not bound. Preferably the
recombinant HLA-E is in the form of a multimer.
[0032] In another aspect the invention provides a method of
preventing HLA-E-mediated inhibition of NK cell activity. The
method comprises selecting a compound that interferes with
HLA-E-binding to an inhibitory CD94/NKG2 receptor and contacting an
NK or T cell expressing an inhibitory CD94/NKG2 receptor with the
compound.
BRIEF DESCRIPTION OF THE DRAWINGS
[0033] FIGS. 1A-1D illustrate a series of graphs showing HLA-E
tetramer binds NK cells and a subset of T cells. Flow cytometry
analysis on gated peripheral blood lymphocytes from normal EBV
seropositive donor VB using (FIG. 1A) HLA-E tetramer refolded
around the leader sequence peptide residues 3-11 from HLA-B*0801 or
(FIG. 1C) HLA-A2 tetramer refolded around the Epstein Barr Virus
(EBV) lytic cycle BMLF1 259-267 peptide epitope. The phenotypes of
(FIG. 1B) HLA-E tetramer or (FIG. 1D) HLA-A2 tetramer binding
lymphocytes were further investigated in triple color stains as
indicated. Percentages in each quadrant are represented by the
cross in the upper right.
[0034] FIGS. 2A-2B illustrate a series of graphs showing HLA-E
tetramer staining is inhibited by anti-CD94 antibodies. In FIG. 2A,
peripheral blood lymphocytes from normal donor SRJ stained with the
anti-CD94 antibody HP3D9 (1/50 dilution of ascites) followed by
FITC-anti-mouse IgG (Fab')2 (Sigma); HLA-E tetramer PE alone; or
HLA-E tetramer-PE in the presence of HP3D9 (1/50) which inhibited
HLA-E tetramer staining. In FIG. 2B, the NK cell line NKL
expressing the NK receptor CD94/NKG2A but none of the KIR molecules
stained with the anti-CD94 antibody DX22 (1 mg) followed by
PE-anti-mouse IgG; HLA-E tetramer-PE; or HLA-E tetramer-PE in the
presence of 1 mg of DX22 antibody which inhibited HLA-E tetramer
staining. Percentages in each quadrant are listed in the upper
right.
[0035] FIGS. 3A-3B illustrate a series of graphs showing HLA-E
binds to NK cell CD94/NKG2A, CD94/NKG2B and CD94/NKG2C receptors
but not to CD94 or NKG2 alone. FIG. 3A illustrates P815 cells
stably transfected with pBJ-neo vector containing human CD94 cDNA
or NKG2B cDNA. Cells were stained with PE-control mouse IgG1
(cMIgG1) or IgG2b (cMIgG2b), anti-CD94 antibody DX22-PE, anti-NKG2A
and B antibody DX20-PE, or HLA-E tetramer-PE. FIG. 3B illustrates
293T cells stably transfected with CD94 were transiently
transfected with NKG2A, NKG2B, and NKG2C. Flow cytometry staining
was performed using rabbit preimmune serum (cRIgG) 1/500 final
dilution or rabbit anti-CD94/NKG2 heterodimer serum
(anti-CD94/NKG2) 1/500 final dilution, both followed by
FITC-antirabbit IgG, or with HLA-E tetramer-PE.
[0036] FIGS. 4A-4B illustrate a series of graphs showing HLA-E
mediates inhibition of NK cells through interaction with CD94/N
KG2A. In FIG. 4A, lysis of 721.221 cells expressing HLA-B*5801,
HLA-G or a chimeric molecule (GLS-B*5801) containing the HLA-G
leader sequence and the extracellular, transmembrane, and
cytoplasmic domains of HLA-B*5801 by a representative NK-cell clone
expressing the CD94/NKG2A receptor. Assays were performed at an
effector to target ratio of 0.5:1, in the presence of control
immunoglobulin (clg), anti-CD94 (DX22), or anti-HLA class I (DX17)
at 5 .mu.g ml.sup.-1. In FIG. 4B, lysis of 721.221 cells expressing
mouse CD80 or a chimeric molecule (B7LS-mCD80) containing the
HLA-B*0702 leader sequence and the extracellular, transmembrane,
and cytoplasmic domains of mouse CD80 by two representative NK-cell
clones expressing the CD94/NKG2A receptor. Assays were performed at
an effector-to-target ratio of 1:1 in the presence of control
immunoglobulin (clg) or anti-CD94 (DX22) at 10 .mu.g ml.sup.-1.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0037] The invention is useful for diagnostic purposes and in
general for monitoring diseases. Detection and/or quantization of
NK cells or a subpopulation of NK cells, or a sub-population of T
cells (which may be further identified by co-staining with anti-CD8
or anti-CD4 antibody or antibodies or ligands to other T cell
markers) will be useful in a variety of conditions, including the
following:
[0038] (i) Cancer. Lymphomas and Leukaemias (Particularly Large
Granular Cell Leukaemias).
[0039] NK cells are believed to have an anti-tumour cell activity.
A marker for progress of therapy, or simple prognosis, can be
provided by monitoring NK cell numbers and optionally their state
of activation. This will provide an extremely simple test. The
methods described herein can be used to determine or estimate NK
cell numbers in a sample taken from a patient. The numbers of NK
cells and T cells with an NK-like activity (CD94/NKG2+) can be
estimated by the use of antibodies against markers of NK cells
(e.g, CD56 or CD16) or antibodies against T cell markers (e.g.,
CD3). The state of activation of the NK cells can be investigated
by co-staining with antibodies to activation markers.
Alternatively, the activation state of NK cells can be assessed in
a functional assay in vitro. The NK cells are isolated from the
sample and their cytolytic activity and/or ability to produce
cytokines (e.g., interferon-y and/or TNF-a) is assessed, either
directly or after a short period in culture.
[0040] (ii) Infections
[0041] NK cell numbers may change during viral or other infections
and knowing their numbers could be of great value for example in
HIV infected patients. It will be of particular interest to monitor
NK cell numbers in cytomegalovirus (CMV) infections. CMV has
sequences in its proteins that are capable of affecting HLA-E
expression. More specifically, these CMV sequences induce cell
surface expression of HLA-E in the virus host cell.
[0042] (iii) Pregnancy
[0043] There is interest in the role of NK cells in the placenta,
in the prevention of rejection of the foetus. The invention
provides a means to monitor NK cells in the placenta.
[0044] (iv) Transplantation
[0045] NK cells may be involved in transplant rejection and
graft-versus-host disease (GVHD) after bone marrow transplantation.
Monitoring of NK cells may be of value in patient management.
[0046] (v) Immunodeficiency
[0047] The diagnostic use of HLA-E can be extended to the detection
of new immunodeficiency syndromes, either inherited or acquired,
which exhibit lower or higher than normal NK cell levels. Some
treatments may be toxic or stimulatory to NK cells.
[0048] (vi) Autoimmune Diseases
[0049] It will be useful to monitor NK cells in autoimmune diseases
such as systemic lupus erythematosus, diabetes, thyroid diseases,
vitiligo, rheumatoid arthritis etc.
[0050] (vii) Following Treatment
[0051] The invention also enables the monitoring of secondary
effects of any treatment which could lead to up or down regulation
of CD94/NKG2+ NK cells.
[0052] Although NK cells are specifically referred to in the
examples of diseases and conditions above, monitoring of the T cell
subset expressing receptors recognised by HLA-E is also
included.
[0053] The invention also has a range of therapeutic applications.
Examples include the following:
[0054] (i) Enhancing NK Levels
[0055] The invention provides a method of selecting HLA-E binding
NK cells or T cells from a mixed cell population. The selected
cells can be expanded in vitro and returned to the patient. Such
treatment may be effective in some serious infections or cancers
where a growth deficiency of these cells is associated with poor
prognosis.
[0056] (ii) Removing NK Cells
[0057] The invention provides methods and means for removal of
HLA-E binding NK cells and T cells, for example from bone marrow to
be used as donor bone marrow in transplantation. T cell depletion
of bone marrow using other techniques is already known and is
effective. HLA-E coupled to a toxin could also be used to destroy
HLA-E binding cells in vivo.
[0058] (iii) Expressing HLA-E at the surface of cells
[0059] The invention provides a method and means for inhibiting or
stimulating NK cell activity, by providing HLA-E at the surface of
cells. NK cells play a role in transplant rejection. Therefore, by
ensuring that HLA-E is expressed at the cell surface in a
xenotransplant organ or tissue, the likelihood of rejection will be
reduced and rejection may be avoided altogether.
[0060] Multimers of HLA-E in accordance with the invention can take
a variety of possible forms. The HLA-E molecules may be associated
with one another via a linker molecule. Alternatively or
additionally the HLA-E molecules may be attached to larger entities
such as membrane structures, for example liposomes. Suitable linker
molecules include multivalent attachment molecules such as avid in,
streptavidin and EXTRAVIDIN (modified egg white avidin) each of
which has four binding sites for biotin. Thus, biotinylated HLA-E
molecules can be formed into multimer complexes of HLA-E having up
to four HLA-E binding sites. The number of HLA-E molecules in the
resulting complex will depend upon the ratio of HLA-E molecules to
linker molecules used to make the complexes, and also on the
presence or absence of any other biotinylated molecules such as
biotinylated signal moieties or toxic moieties. Preferred complexes
are trimeric or tetrameric HLA-E complexes.
[0061] Other cross-linking techniques suitable for preparing HLA-E
dimers or multimers include, for example, antibody cross-linking to
produce dimers, and techniques using cross-linking agents such as
flurogenic compounds or other chemical cross-linking agents.
[0062] Described herein in detail is a method of making HLA-E
tetramers by biotinylating recombinant HLA-E heavy chains
(.alpha.1, .alpha.2 and .alpha.3 domains) and constructing
complexes by refolding the HLA-E heavy chains in vitro with
.beta.2m molecules and with a synthetic leader peptide from HLA-B.
In the method described, biotinylation sites are engineered into
the C terminus of the HLA-E heavy chain and the recombinant HLA-E
molecules are biotinylated by means of a suitable biotinylating
enzyme. The multivalent linker molecule, in this case EXTRAVIDIN
(modified egg white avidin), is then added. Variations of this
tetramerisation method are envisaged. For example, the .beta. 2m
could be chemically biotinylated at its seven lysine residues, four
of which are in an appropriate position to allow complexing.
[0063] Whatever form of HLA-E is used, a suitable peptide for
binding in the HLA-E peptide groove will be required. Tables 1 and
2 show examples of possible peptides found at residues 3 to 11 of
MHC class I leader sequences. The tables indicate which of these
are found to bind to HLA-E in vitro. Other peptides which bind to
HLA-E may be used, for example peptides from viral proteins which
may be the same as or different to MHC class I leader sequence
peptides. Although nonamer peptides are usually the optimum size
for binding to HLA-E, it is possible that peptides which are
slightly shorter or slightly longer, e.g., by one or two residues
at one or both ends, will also work. A suitable assay for
identifying synthetic peptides which bind to HLA-E is described in
Braud et al 1997 Eur. J. Immunol. 27:1164-1169.
[0064] Labeling of the HLA-E may be carried out by any suitable
method. Described herein is tetrameric HLA-E labeled with
phycoerythrin via the EXTRAVIDIN (modified egg white avidin) used
to form the tetrameric complex. A variety of other signal moieties
may be employed at a variety of possible sites on the HLA-E monomer
or multimer, and the HLA-E may be labeled before, during or after
multimerisation. Useful signal moieties commonly employed for
labeling proteins include radioactive (e.g., 32-P or 33-P or 35-S),
fluorescent (e.g. FITC) and enzymatic (e.g. horseradish peroxidase)
labels. These and other detectable labels may be employed in the
invention.
[0065] HLA-E is also provided according to the invention linked to
a support. The HLA-E thus immobilised is particularly useful for
capture purposes, that is for capturing and separating CD94/NKG2+
cells. A mixture containing or suspected of containing CD94/NKG2+
cells is brought into contact with the immobilised HLA-E under
binding conditions and unbound material is removed. CD94/NKG2+
cells may then be recovered. Examples of suitable solid supports
include but are not limited to support bodies or particles which
may be for example in the form of beads, wells, tubes and plates.
The HLA-E may be immobilised by a variety of means well-known in
the art and may be attached to the support by means of a suitable
linker molecule.
[0066] HLA-E is also provided according to the invention attached
to an effector agent such as a toxic moiety, the purpose of which
may be the targeted depletion of CD94/NKG2+ cells in a mixed
population of cells in vitro or in vivo. The toxic moiety is
preferably a toxin which selectively kills the cells to which the
HLA-E is bound. Suitable toxins can be derived from natural sources
such as plants (for example ricin, abrin) or bacteria (for example
diphtheria toxin, Pseudomonas exotoxin), or they may be synthetic
or synthetic versions of naturally occurring toxins. In one
particular example, the toxin is coupled to HLA-E by means of a
linker molecule that introduces a disulphide bond between the HLA-E
and the toxin (Vitetta et al, Biol. Therapy of Cancer, 1991,
482-495; Myers et al, J. Immunol. Methods, 1991, 131: 221-237).
Thioether or other related compounds may be used to form
non-reducible linkages. In another example, the toxin is expressed
as a fusion protein with the HLA-E molecule, by splicing the gene
encoding the toxin to the gene encoding the ligand to create a
recombinant fusion protein (Pastan et al, Annu. Rev. Biochem.,
1992, 61: 331-354). Such fusion proteins may also include a
biotinylation site to enable the creation of tetrameric or
multimeric complexes of the toxin and HLA-E.
[0067] In general, the HLA-E described herein for use in detecting
and separating and targeting toxins to CD94/NKG2+ cells will be a
soluble HLA-E molecule. The term "soluble" is used herein in the
manner in which it is conventionally used in the art in relation to
cell surface receptors. A soluble form of a cell surface receptor
is usually derived from the native form by deletion of the
transmembrane domain. The protein may be truncated by removing both
the cytoplasmic and the transmembrane domains, or there may be
deletion of just the transmembrane domain with part or all of the
cytoplasmic domain being retained. The important thing is that the
desired extracellular function of the receptor is retained, which
in this case is the CD94/NKG2-binding capability of the HLA-E
binding domain. Wild type HLA-E may be modified to achieve the
desired form by proteolytic cleavage, or by expressing a
genetically engineered truncated or partially deleted form.
[0068] HLA-E for use in the invention which is present at surface
of a cell on the other hand is preferably, although not
necessarily, a membrane-bound form of HLA-E containing a
transmembrane domain. It may or not also have a cytoplasmic domain.
In order to get HLA-E expressed at the surface of a cell which does
not normally express HLA-E at its surface, it is necessary to
provide a peptide which binds in the HLA-E peptide binding groove.
The peptide may be derived either from an HLA leader sequence which
is permissive for HLA-E expression, or it may be derived from
another source which expresses peptides that bind to HLA-E and
induce its expression. Such other sources include viruses which
escape NK cell-mediated cytotoxicity by encoding a peptide which
binds to HLA-E and induces its expression. For example, the human
cytomegalovirus (HCMV) encodes a protein known as UL40 (Accession
No. p16780) which possesses a peptide capable of binding to HLA-E
(see fourth leader sequence peptide listed in Table 1).
[0069] Thus, HLA-E expression at the surface of a non-human
mammalian cell may be achieved for example by co-transfecting the
cell with nucleic acid encoding HLA-E and another HLA class I which
has a leader sequence peptide capable of binding to HLA-E.
Alternatively the cell may be co-transfected with nucleic acids
encoding HLA-E and UL40 of HCMV. The inventors have shown that when
transfected with UL40, cells expressing HLA-E intracellularly are
induced to express HLA-E at the cell surface. Instead of
co-transfecting the cells with two different sequences, the cells
can be transfected with a single vector containing both sequences,
or more conveniently, with a chimeric nucleic acid which encodes a
fusion protein of HLA-E with the peptide. This chimeric nucleic
acid may be for example a recombinant nucleic acid encoding HLA-E,
in which the leader sequence of HLA-B8 replaces the leader sequence
of HLA-E. In another example, the nucleic acid encodes HLA-E linked
to the peptide via a linker of sufficient length to allow the
peptide to locate in the peptide-binding groove while also
remaining covalently attached to the HLA-E molecule. In this
instance, the peptide is not a part of the leader sequence which is
cleaved off, and is instead linked to the mature HLA-E molecule
which is expressed at the cell surface. Advantageously, the peptide
is linked to the HLA-E heavy chain via the HLA-E N-terminus (the
.alpha.-1 domain). Such MHC-linked peptides are described in the
published literature (see for example Mottez et al., J. Exp. Med.,
1995,181:493-502). It may also be advantageous to express the HLA-E
heavy chain and 2m as a fusion protein. Such fusions are described
in the literature (see for example Toshitani et al., PNAS, 1996,
93:236). Thus, the HLA-E-peptide molecule may be expressed from a
single coding sequence.
[0070] For the provision of HLA-E at the cell surface in organs or
tissues to be transplanted into another species, suitable
transgenic animals can be produced. Techniques for generating
transgenic animals such as transgenic pigs or rats or mice are well
known in the art. For the purposes of the invention, nucleic acid
material which expresses HLA-E and a suitable HLA-E binding peptide
in the recipient organism is introduced into cells of the organism
at the appropriate early stage of development. Individual animals
which express HLA-E at the surface of their cells are then
selected. Organs or tissues from those animals provide xenogeneic
material for transplantation.
[0071] The absence of HLA-E in xenogeneic cells makes them
susceptible to NK cell-mediated lysis because their MHC class I
molecules fail to be recognised by human killer cell inhibitory
receptors. This is evident from inducing expression of HLA-Cw0301
on porcine endothelial cells, which protects the porcine cells
against xenogeneic cytotoxicity mediated by NK cells expressing the
inhibitory NK receptors binding to HLA-Cw0301 (Seebach et al, J.
Immunol., 1997, 159:3655). There is evidence that NK cells play a
role in the cellular immune response against xenografts (reviewed
in Kaufman et al, Ann. Rev. Immunol., 1995, 13:339 and Bach et al,
Immunol. Today, 1996, 17:379). The endothelium is the first site of
contact between a vascularized xenograft and the recipient immune
system. It has been shown that human NK cells adhere to the
vascular endothelium and infiltrate into the xenogeneic organs
(Kirk et al, Transplantation, 1993, 56:785; Inverardi, Immunol.
Rev.,1994 141:71) and that NK cells directly activate porcine
endothelial cells (Goodman et al, Transplantation, 1996, 61:763).
It has been observed that xenogeneic human anti-porcine
cytotoxicity in vitro includes an important MHC unrestricted
contribution from polyclonal NK populations (Inverardi, Immunol.
Rev., 1994, 141:71; Kirk et al, Transplantation, 1993, 55:294;
Seebach et al, Xenotransplantation, 1996, 3:188). It has also been
shown that transgenic mice expressing a killer cell inhibitory
receptor from the immunoglobulin superfamily, CD158, which
recognises HLA-Cw3, are prevented from rejecting H-2 mismatch bone
marrow grafts which express the cognate MHC class I HLA-Cw3 allele
(Cambiaggi et al, PNAS, 1997, 94(15):8088-92. Thus, an HLA-E
transgenic animal can be used to provide organs which will not be
prone to attack by human NK cells expressing CD94/NKG2A
receptors.
[0072] Xenogeneic transplantation will also involve the use of
other mechanisms and/or reagents for the purpose of improving
transplant survival. In particular, immunosuppressive agents may be
employed. Immunosuppressive drugs are commonly used in
transplantation therapy.
[0073] Currently, the primary animal species proposed as sources of
xenografts are pigs and baboons (and possibly cows). Examples of
particular treatments currently under consideration are: [0074]
treatment of Parkinson's disease by implantation of porcine foetal
neuronal tissue; [0075] treatment of diabetes mellitus by
implantation or infusion of encapsulated porcine pancreatic islet
cells; [0076] treatment of hepatic failure by perfusion through or
implantation of whole pigs livers; (see Deacon, Nat. Med., 1997,
3:350; Tibell, Transplant. Proc., 1994, 26:762; Cramer, Transplant.
Proc., 1995, 27:80).
[0077] It will be evident that in addition to any special features
such as the absence of transmembrane and/or cytoplasmic domains, or
the presence of biotinylation sites required for multimerisation,
the recombinant HLA-E used in the invention may have other features
which make it different to native HLA-E. For example, the
recombinant HLA-E may have deletions or insertions or altered
residues compared to native HLA-E, which result in improved
properties such as enhanced binding capability or improved
stability, for use in accordance with the invention. HLA-E having
improved stability at elevated temperatures, such as temperatures
over 4.degree. C. and/or over room temperature and/or at or around
37.degree. C., will be of particular interest.
[0078] HLA-E in recombinant form as described herein is also
provided in formulations suitable for in vivo use. Such
formulations comprise a pharmaceutically acceptable diluent or
carrier.
[0079] HLA-E itself exhibits very little polymorphism. The
sequences for two different alleles of HLA-E can be found in the
following data base locations: E*0101 at M20022 (arg in the residue
at position 107); E*01031 at M32507 (glycine residue at position
107).
[0080] Some further, known techniques the principles of which may
be additionally applied in the separation, identification or
targeted killing methods of the invention are discussed below.
[0081] Some known separation methods will be adaptable for use in
the separation or isolation of CD94/NKG2+ cells using HLA-E. For
example, T cell populations can be isolated by use of
antibody-coated plates. The antibodies are specific for particular
cell-surface markers. Cell separation can be a negative selection
process or a positive selection process (Wysocki et al, 1978 PNAS
75:2840-2848). HLA-E coated plates may be used to separate
CD94/NKG2+ cells.
[0082] Immunomagnetic purification of a T cell subpopulation can
also be realised using suitable antibodies coated on magnetic
beads, in a negative or a positive selection process (Funderud et
al, 1987 in Lymphocytes: A Practical Approach Oxford University
Press, New York 5561). HLA-E-coated beads may be similarly employed
for selection of CD94/NKG2+ NK cells and T cells.
[0083] FACS (fluorescence activated cell sorting) techniques may
also be employed. Cell sorting of fluorescence-labeled cells uses
flow cytometry to monitor the expression of specific intracellular
and cell surface molecules and sort cell populations (Fleisher et
al, 1988 Cytometry 9:309-315).
[0084] Techniques which may be used in accordance with the
invention for selective depletion or targeted killing of CD94/NKG2+
cells in a mixed cell population include
antibody/complement-mediated cytotoxicity. Using a
complement-fixing antibody, the cells expressing the marker
recognised by the antibody can be lysed in presence of complement
(Bianco et al, 1970 J. Exp. Med. 132:702-720). For example, an
anti-HLA-E antibody may be employed to selectively destroy cells to
which HLA-E is bound.
TABLE-US-00001 TABLE 1 Examples of peptides generated from MHC
class I leader sequences at residues 3-11 Leader sequence peptide
(3-11) Binding to from MHC class I HLA-E in vitro -VMAPRTLVL +
-VMAPRTLLL + -VMAPRTVLL + -VMAPRTLIL + -VMAPRTLFL + -VMGPRTLVL +/-
-VTAPRTVLL - -VTAPRTLLL - -VMPPRTLLL + -VMEPRTLIL - -VMAPRALLL -
[SEQ ID NOS: 1 - 11]
EXAMPLES
Example 1
[0085] Construction of HLA-E Tetrameric Complexes
[0086] HLA-E-tetrameric complexes were constructed by refolding
recombinant HLA-E and .beta.2m molecules in vitro with a synthetic
peptide (VMAPRTVLL) [SEQ ID NO 3] derived from residues 3-11 of the
signal sequence of HLA-B*0801. A biotinylation site was engineered
in the C terminus of the HLA-E heavy chain, allowing HLA-E/.beta.
2m/peptide complexes to be enzymatically biotinylated using E.coli
BirA enzyme and conjugated with phycoerythrin (PE)-labeled
EXTRAVIDIN (modified egg white avidin) to create tetrameric
complexes. HLA-A and -B tetramic complexes have proved to be very
efficient at specifically binding to T cell receptors on
antigen-specific CD8+ T cells from peripheral blood in vitro
(Altman et al, 1996 Science 274: 94-96).
[0087] Methods
[0088] HLA-E was cloned by RT-PCR with primers C007 and C006 from
RNA extracted from monocytes of an HLA-E*0101 homozygous
individual. The N terminal nucleotide sequence was synonymously
altered by PCR mutagenesis using the primers C017 and C006 to
optimise protein expression from the pGMT7 vector in E. coli. The
coding sequence for the extracellular portion of HLA-E (residues
1276) was amplified using the primers C017 and CO23 and recloned
into a pGMT7 derivative to produce the expression plasmid C00092
which contains the BirA recognition and biotinylation site in frame
at the 3' end of the HLA-E heavy chain. Primers were:
TABLE-US-00002 [SEQ ID NO: 12] COO6
gtgggctaagcttacggcttccatctcagggtgacgggctc [SEQ ID NO: 13] COO7
ctacgggcatatggtagatggaaccctccttttactctcc [SEQ ID NO: 14] CO17
ccgtacctcgagcatatgggttctcattattaaaatattttcata
cttctgtatctagacccggccg [SEQ ID NO: 15] CO23
tggtgtctagaggatcctggcttccatctcagggtgacgggctcg
[0089] HLA-E tetrameric complexes were generated essentially as
described (Altman et al 1996). Briefly, HLA-E and 132m proteins
were over expressed in E. coli strains BL21 (DE3) pLysS and XA90
respectively, purified from inclusion bodies, solubilised into a
urea solution, then refolded by dilution in vitro with a synthetic
peptide (VMAPRTVLL) [SEQ ID NO 3] from HLA-B*0801 leader sequence
(Research Genetics). HLA-E heavy chain/.beta.2m/peptide complexes
were biotinylated with BirA enzyme, purified by FPLC and MONO-Q
anion exchange chromatography, then complexed in a 4:1 molar ratio
with EXTRAVIDIN-PE (Sigma) (modified egg white avidin-PE).
Example 2
[0090] Binding of HLA-E Tetramers
[0091] Peripheral blood mononuclear cells (PBMC) from 9 normal
donors were stained with HLA-E tetramer prepared as described in
Example 1 and compared to staining observed with an HLA-A2 tetramer
refolded with Epstein Barr Virus (EBV) lytic cycle BMLF1 259-267
peptide epitope (Steven et al, 1997 J.Exp. Med. 185: 1605-17). A
high frequency of lymphoid cells were stained with the HLA-E
tetramer (range 2 to 11%) (FIG. 1A), whereas the HLA-A2 tetramer
generally stained 0 to 0.8% of the lymphocytes in EBV-seropositive
donors (FIG. 1C). By setting an electronic gate on the lymphocytes
binding HLA-E tetramer, we observed that a large proportion were NK
cells (typically 40 to 80% CD3-, CD56+) but a significant subset
were T cells (typically 15 to 50% CD3+), some of which were also
expressing CD56 (FIG. 1B). About 2% of the lymphocytes binding
HLA-E tetramer were CD4+ T cells, and about 5% were CD19+ B cells,
but these could represent non-specific binding because of similar
staining with the HLA-A2 tetramer (data not shown). The HLA-A2
tetramer did not bind to CD56+ cells but, in EBV-seropositive
donors, bound to EBV specific CD3+, CD8+ T cells (FIG. 1D),
confirming previous studies on the specificity of MHC-tetrameric
complexes for T cells bearing a specific T cell receptor (Altman et
al 1996).
[0092] HLA-E tetramer staining was abolished when the PBMC and the
tetramer were incubated in the presence of the antibody HP3D9
(Aramburu et al, 1990 Immunol. 144: 3238-47) against CD94, an NK
cell receptor belonging to the C-type lectin superfamily (Chang et
al, 1995 Eur J. Immunol. 25: 2433-37) (FIG. 2A). As the antibody
HP3D9 was diluted, the HLA-E tetramer staining was restored (data
not shown). The interaction between HLA-E and CD94 was also
confirmed by staining a number of well-characterised CD94+ NK
clones with HLA-E tetramer and demonstrating that another anti-CD94
mAb (DX22) (Phillips et al, 1996 Immunity 5:163-172) completely
inhibited HLA-E tetramer binding (FIG. 2B, and data not shown). No
staining with HLA-A2 tetramer was found on CD94+ NK clones (data
not shown).
[0093] To characterise further the NK receptor interacting with
HLA-E, we stained P815 and 293T cells transfected with these
receptors. No HLA-E tetramer staining was observed on P815 stably
transfected with CD94 alone or NKG2B alone (FIG. 3A), nor on 293T
transiently transfected with CD94 or NKG2A alone (data not shown).
In contrast, HLA-E tetramer bound to 293T cells cotransfected with
CD94 and NKG2A, CD94 and NKG2B, or CD94 and NKG2C (FIG. 3B).
Expression of the heterodimers on these transfectants was monitored
using a polyclonal rabbit serum that reacts with CD94/NKG2A, NKG2B
and NKG2C heterodimers (Lazetic et al, 1996 J. Immunol. 157:
4741-45). This result was confirmed using mouse pre-B Ba/F3 cells
stably transfected with CD94/NKG2C or NK clones expressing the
inhibitory receptor CD94/NKG2A (FIG. 2B and data not shown).
Carbohydrates on HLA-E are not necessary for binding, as the
recombinant HLA-E used to make the tetramer was produced in E.
coli. This is quite surprising given that both CD94 and NKG2
proteins are members of the C-type lectin superfamily. Carbohydrate
residues may form additional points of interaction increasing the
affinity of binding.
[0094] We have also shown that HLA-E does not interact with other
killer cell inhibitory cell receptors (KIR) as no staining with the
HLA-E tetramer was observed on Ba/F3 cells transfected with KIR2DL1
(NKAT1 or p58), KIR2DL3 (NKAT2 or p58), KIR3DL1 (NKAT3 or p70),
KIR3DL2 (NKAT4 or p70/140), KIR2DS2 (NKAT5 or p50), KIR2DL2 (NKAT6
or p58) or KIR2DS4 (NKAT8 or p50) (Lanier et al, 1997 Immunol. Rev.
155: 145154). Furthermore, staining of PBMC with HLA-E tetramer was
not blocked by antibodies against any of these MR receptors: EB6
(anti-KIR2DL1), GL183 (anti-KIR2DL3, -KIR2DS2, -KIR2DL2), DX9
(anti-KIR3DL1), or 5.133 (anti-KIR3DL1, -KIR3DL2) (data not shown).
Thus, the CD94/NKG2 receptors appear to be unique and specific
receptors for HLA-E recognition.
[0095] We have previously reported that HLA-E, like the mouse Qa-1
molecule (Aldrich et al, 1994 Cell 79: 649-658; DeCloux et al, 1997
J. Immunol. 158: 2183-2191; Cotterill et al, 1997 Eur. J. Immunol.
27: 21232132) can bind signal sequence-derived peptides from MHC
class I molecules in vitro (Braud et al 1997) and recently showed
that HLA-E cell surface expression is regulated by the binding of
such peptides (Braud et al 1998). Most HLA-A and HLA-C alleles
possess a leader peptide 3-11 that binds to HLA-E whereas only a
third of HLA-B alleles do. The remaining B alleles have a Threonine
at position 2 in the peptide instead of a Methionine. This
substitution at a primary anchor residue disrupted peptide binding
to HLA-E as measured in an in vitro peptide binding assay
previously described (Braud et al 1997) (Table 2). Transfection of
MHC class I alleles which have a leader peptide capable of binding
to HLA-E into HLA-A, -B, -C, -G negative 721.221 cells resulted in
expression of the endogenous HLA-E on the cell surface of 721.221
cells. When the leader sequence peptide was not capable of binding,
no such upregulation of HLA-E at the cell surface was observed.
[0096] It has been shown previously that NK cells expressing an
inhibitory CD94/NKG2A receptor do not kill 721.221 cells
transfected with certain HLA-A, -B, -C, or -G alleles, but are able
to lyse these transfectants in the presence of neutralising
anti-CD94 or anti class I antibodies (Phillips et al, 1996 Immunity
5: 163-172; Sivori et al, 1996 Eur. J. Immunol. 26:2487-2492;
Sivori et al, 1996 Transplant 28: 3199-3203). A striking
correlation between the presence of an HLA class I leader sequence
peptide capable of binding to HLA-E causing its surface expression
and the specificity of the CD94/NKG2A inhibitory receptor is shown
in Table 2. All the MHC class I alleles which, upon transfection,
protect 721.221 cells from killing by CD94/NKG2A+ NK clones have a
peptide capable of binding to HLA-E. Similarly, all HLA alleles
incapable of protecting against these clones lack an HLA-E binding
leader peptide. Together with the direct evidence for physical
interaction between HLA-E and CD94/NKG2A, these results indicate
that inhibition by the CD94/NKG2A receptor is mediated by
recognition of HLA-E rather than a broad range of HLA-A, -B, and -C
molecules. In further support of this, the HLA-A2 tetramer refolded
around a Tax peptide epitope of human T-cell lymphotropic virus
HTLV1 (Garboczi et al 1996 Nature 384:134-141) did not bind to
CD94/NKG2A transfectants or NK cells expressing CD94/NKG2
receptors, despite the fact that HLA-A2 has been shown to have a
protective effect against CD94/NKG2A+ NK clones and HLA-A2 target
cell protection can be reversed in the presence of anti-CD94 or
anti-class I antibodies. Furthermore we confirmed, by
immunoprecipitation that the anti-class I antibody DX 17, which
inhibits interactions between class I molecules and CD94/NKG2A,
also recognizes HLA-E.
[0097] It has recently been demonstrated that recognition of
721.221 target cells by CD94/NKG2A+ NK clones can be inhibited by
transfection of HLA-G, another nonclassical class I molecule mainly
expressed on trophoblast cells (Soderstrom et al, 1997 J. Immunol.
159: 1072-1075; Perez Villar et al, 1997 J. Immunol. 158:
5736-5743; Pende et al, 1997 Eur. J. Immunol. 27: 1875-1880).
However, HLA-G also possesses a leader sequence peptide capable of
binding to HLA-E and 721.221-G transfectants express a significant
level of HLA-E. Similarly, Reyburn et al 1997 (Nature 386: 514-517)
recently reported that human cytomegalovirus encodes a viral
protein (UL18), with similarity to MHC class I, that can protect
721.221 cells from NK cell lysis, possibly involving CD94
receptors. Whether these observations can be explained by binding
of HLA-G or UL18 leader peptides to the endogenous HLA-E molecules
in 721.221 is under investigation.
[0098] HLA-E also binds to CD94/NKG2C which has been shown to
activate cytolytic activity in NK cell transfectants (Houchins et
al, 1997 J. Immunol. 158: 3603-3609) indicating that HLA-E is
involved in regulating NK cell-mediated cytotoxicity via both
CD94/NKG2A and CD94/NKG2B inhibitory NK cell receptors and
CD94/NKG2C stimulatory NK cell receptors. Our present results
demonstrate a novel role for a non-classical class I molecule HLA-E
and identify its predominant receptor. It remains to be determined
whether the strong preference of HLA-E for binding signal
sequence-derived peptides is simply to permit expression of HLA-E
or whether it is implicit in recognition by CD94/NKG2
receptors.
TABLE-US-00003 TABLE 2 Presence of .221 cells Inhibition leader
sequence Concentration of tranfected of killing peptide HLA leader
peptide required with by CD94/ capable of sequence to obtain HLA
class I NKG2A + binding to peptide 50% of binding alleles NK
clones* HLA-E.sctn. (residues 3-11) to HLA-E.dagger. .221 - -
.dagger-dbl. .221-A*0201 + + VMAPRTLVL 0.06 .mu.M .221-A*0211 + +
VMAPRTLVL 0.06 .mu.M .221-A*2501 + + VMAPRTLVL 0.06 .mu.M
.221-A*2403 + + VMAPRTLVL 0.06 .mu.M .221-A*3601 + + VMAPRTLLL 0.3
.mu.M .221-B*0702 + + VMAPRTVLL 0.06 .mu.M .221-Cw*0102 + +
VMAPRTLIL 0.3 .mu.M .221-Cw*0401 + + VMAPRTLIL 0.3 .mu.M
.221-Cw*0304 + + VMAPRTLIL 0.3 .mu.M .221-Cw*0801 + + VMAPRTLIL 0.3
.mu.M .221-G + + VMAPRTLFL 0.3 .mu.M .221-B*1501 - - VTAPRTVLL
>100 .mu.M .221-B*5101 - - VTAPRTVLL >100 .mu.M .221-B*5801 -
- VTAPRTVLL >100 .mu.M .221-B*4601 - - VTAPRTVLL >100 .mu.M
.221-B*5401 - - VTAPRTLLL >100 .mu.M .221-B*5501 - - VTAPRTLLL
>100 .mu.M [SEQ ID NOS: 1, 1, 1, 1, 2, 3, 4, 4, 4, 4, 5, 7, 7,
7, 7, 8, 8, respectively] *Results published by Phillips et al 1996
.dagger.A peptide binding assay was developed in vitro. Results are
expressed as a ratio of optical densities referred to as percentage
of binding to HLA-E (Braud et al 1997). .dagger-dbl.The HLA-A, -B,
-C, and -G negative.221 cells express HLA-E and HLA-F which have a
shorter leader sequence and lack the appropriate peptide capable of
binding to HLA-E. .sctn.The presence of a leader sequence peptide
capable of binding to HLA-E upregulates HLA-E surface expression as
measured on .221 and .221 cells transfected with HLA-A or-B alleles
using the antibody DT9 recognizing HLA-E and HLA-C alleles
[0099] Figure Legends for Example 2
[0100] FIG. 1 HLA-E tetramer binds NK cells and a subset of T
cells
[0101] Flow cytometry analysis on gated peripheral blood
lymphocytes from normal EBV seropositive donor VB using (A) HLA-E
tetramer refolded around the leader sequence peptide residues 3-11
from HLA-B*0801 or (C) HLA-A2 tetramer refolded around the Epstein
Barr Virus (EBV) lytic cycle BMLF1 259-267 peptide epitope (Steven
et al 1997). The phenotypes of (B) HLA-E tetramer or (D) HLA-A2
tetramer binding lymphocytes were further investigated in triple
colour stains as indicated. Percentages in each quadrant are
represented by the cross in the upper right. Within the total CD3-,
CD56+ NK cell population, 10.3% of cells bound HLA-E tetramer, and
within the total CD3+ T cell population, 2.2% of cells bound HLA-E
tetramer. In contrast, less than 0.2% of CD3-, CD56+ cells bound
HLA-A2 tetramer, whereas 1% of CD3+ T cells bound HLA-A2
tetramer.
[0102] FIG. 2 HLA-E tetramer staining is inhibited by anti-CD94
antibodies
[0103] (A) Peripheral blood lymphocytes from normal donor SRJ were
stained with the anti-CD94 antibody HP3D9 (Aramburu et al 1990)
(1/50 dilution of ascites) followed by FITC-anti-mouse IgG (Fab')2
(Sigma); HLA-E tetramer PE alone; or HLA-E tetramer-PE in the
presence of HP3D9 (1/50) which inhibited HLA-E tetramer
staining.
[0104] (B) The NK cell line NKL (Robertson et al, 1996 Exp.
Haematol. 24: 406-415) expressing the NK receptor CD94/NKG2A but
none of the KIR molecules was stained with the anti-CD94 antibody
DX22 (Phillips et al 1996) (1 mg) followed by PE-anti-mouse IgG;
HLA-E tetramer-PE; or HLA-E tetramer-PE in the presence of 1 mg of
DX22 antibody which inhibited HLA-E tetramer staining. Percentages
in each quadrant are listed in the upper right. The HLA-A2 tetramer
refolded around the HTLV1 Tax peptide (Garbocz et al 1996) did not
bind to NKL (data not shown).
[0105] FIG. 3 HLA-E binds to NK cell CD94/NKG2A, CD94/NKG2B and
CD94/NKG2C receptors but not to CD94 or NKG2 alone.
[0106] (A) P815 cells were stably transfected with pBJ-neo vector
containing human CD94 cDNA (Chang et al 1995) or NKG2B cDNA
(Houchins et al 1991 J. Exp. Med. 173: 1017-20). Cells were stained
with PE-control mouse IgG1 (cMIgG1) or IgG2b (cMIgG2b), anti-CD94
antibody DX22-PE, anti-NKG2A and B antibody DX20-PE, or HLA-E
tetramer-PE. Neither P815 transfectant stained with HLA-E tetramer
or HLA-A2 HTLV1 Tax peptide tetramer.
[0107] (B) 293T cells stably transfected with CD94 were transiently
transfected with NKG2A, NKG2B, and NKG2C (Lazetic et al 1996). Flow
cytometry staining was performed using rabbit preimmune serum
(cRIgG) 1/500 final dilution or rabbit anti-CD94/NKG2 heterodimer
serum (anti-CD94/NKG2) 1/500 final dilution, both followed by
FITC-antirabbit IgG, or with HLA-E tetramer-PE. No staining with
HLA-A2-HTLV1 tax peptide tetramer was observed. Staining of
293T-CD94 cells cotransfected with a control plasmid were not
stained by HLA-E tetramer or the rabbit anti-CD94/NKG2 serum (data
not shown).
Example 3
[0108] Transfection of cells with HLA-E-Binding Leader Sequences to
Enable HLA-E Expression and Protection against NK Cell Clones
[0109] Methods
[0110] Human NK-cell clones were established and cultured as
described (Litwin et al, 1993 J. Exp. Med. 178: 1321-1336).
Cytotoxicity assays were performed as described (Phillips et al,
1996 Immunity 5: 163172). A chimeric cDNA containing the leader
segment of HLA-G and the extracellular, transmembrane, and
cytoplasmic domains of HLA-B*5801 was generated by PCR using the
following oligonucleotide primers: sense primer 1,
TABLE-US-00004 [SEQ ID NO: 16]
5'-GCGTCTAGAATGGTGGTCATGGCACCCCGA-3'; antisense primer 1, [SEQ ID
NO: 17] 5'-CATGGAGTGGGAGCCGGCCCAGGTCTCGGT-3'; sense primer 2, [SEQ
ID NO: 18] 5'-GGCTCCCACTCCATGAGGTAT-3'; and [SEQ ID NO: 19]
antisense primer 2, 5'-AAGCTTTCAAGCTGTGAGAGACA-3'.
[0111] PCR was performed using a wild-type HLA-G cDNA as a template
with primer set 1 and using wild-type HLA-B*5801 cDNA as a template
with primer set 2. Products from these PCR reactions were mixed and
used as templates for a subsequent reaction with sense primer 1 and
antisense primer 2. The product was digested with Xbal and HindlIl
and ligated into the pBJneo vector.
[0112] A chimeric cDNA containing the leader segment of HLA-B*0702
and the extracellular, transmembrane and cytoplasmic domains of
mouse CD80 (or B7-1) was generated by PCR using the following
oligonucleotide primers:
TABLE-US-00005 sense primer 3, [SEQ ID NO: 20]
5'-ACCGAGACCTGGGCCGTTGATGAACAACTG-3'; antisense primer 3, [SEQ ID
NO: 21] 5'-GCAAGCTTCTAAAGGAAGACGGTCTGTTC-3'; sense primer 4, [SEQ
ID NO: 22] 5'-GGGCGTCGACCCGGACTCAGAATCTCCTCAGACGCCGAG-3'; and
antisense primer 4, [SEQ ID NO: 23]
5'-CAGTTGTTCATCAACGGCCCAGGTCTCGGT-3'.
[0113] PCR was performed using a wild-type mouse CD80 cDNA as a
template with primer set 3 and using wild-type HLA-B*0702 cDNA as a
template with primer set 4. Products from these PCR reactions were
mixed and used as templates for a subsequent reaction with sense
primer 4 and antisense primer 3. The product was digested with Sall
and HindlIl and ligated into the pBJneo vector. PCR products were
verified by sequencing. 721.221 B-lymphoblastoid cells were
transfected with the wild-type and chimeric cDNAs and selected as
described (Litwin et al 1993).
[0114] Results
[0115] To determine whether the presence of an HLA-E binding leader
peptide that induces surface expression of HLA-E is enough to
provide protection against CD94/NKG2A+ NK-cell clones, a chimeric
complementary DNA (GLS-B*5801) was generated. It contained the
leader segment of HLA-G (from which a peptide can bind to HLA-E;
Table 2) and the extracellular, transmembrane and cytoplasmic
domains of HLA B*5801 (an HLA molecule that is not implicated in
recognition by CD94/NKG2A receptors--Phillips et al, 1996 Immunity
5:163-172). Stable 721.221 cell line transfectants were selected
and analysed for susceptibility to lysis by NK-cell clones
expressing CD94/NKG2A receptors. As shown in FIG. 4a, an NK-cell
clone expressing a CD94/NKG2A receptor efficiently killed
untransfected 721.221 cells as well as 721.221 cells transfected
with wild-type HLA-B*5801. However, protection against
NK-cell-mediated lysis was conferred by expression of the chimeric
GLS-B*5801 molecule but reversed in the presence of antibodies
against either CD94 or HLA class I molecules. A chimeric cDNA
(B7LS-mCD80) containing the leader segment of HLA-B*0702 (with a
peptide that can bind to HLA-E; Table 2) and the extracellular,
transmembrane and cytoplasmic domains of mouse CD80 was transfected
into 721.221 cells and tested for lysis by CD94/NKG2A+ NK-cell
clones. CD80 is an adhesion cell surface molecule expressed on
activated B and T cells and macrophages. In this experiment CD80 is
used as an irrelevant control molecule to show that only the leader
sequence of MHC class I molecules is necessary to upregulate HLA-E
and induce a protective effect. There was less lysis of 721.221
cells expressing the B7LS-mCD80 molecule but not of cells
expressing wild-type CD80, and protection was reversed by anti-CD94
but not control antibodies (FIG. 4b). These results indicate that
an HLA-E binding leader peptide alone is enough to protect 721.221
cells from lysis by NK-cell clones expressing inhibitory
CD94/NKG2A-type receptors.
[0116] These results provide further confirmation of HLA-E as a
ligand for CD94/NKG2A.
[0117] Figure Legends for Example 3
[0118] FIG. 4 HLA-E mediates inhibition of NK cells through
interaction with CD94/N KG2A.
[0119] (A) Lysis of 721.221 cells expressing HLA-B*5801, HLA-G or a
chimeric molecule (GLS-B*5801) containing the HLA-G leader sequence
and the extracellular, transmembrane, and cytoplasmic domains of
HLA-B*5801 by a representative NK-cell clone expressing the
CD94/NKG2A receptor. Assays were performed at an effector to target
ratio of 0.5:1, in the presence of control immunoglobulin (clg),
anti-CD94 (DX22), or anti-HLA class I (DX17) at 5 .mu.g
ml.sup.-1.
[0120] (B) Lysis of 721.221 cells expressing mouse CD80 or a
chimeric molecule (B7LS-mCD80) containing the HLA-B*0702 leader
sequence and the extracellular, transmembrane, and cytoplasmic
domains of mouse CD80 by two representative NK-cell clones
expressing the CD94/NKG2A receptor. Assays were performed at an
effector-to-target ratio of 1:1 in the presence of control
immunoglobulin (clg) or anti-CD94 (DX22) at 10 .mu.ml.sup.-1.
Example 4
[0121] Isolation of CD94/NKG2+ Cells by Fluorescence Activated Cell
Sorting
[0122] Peripheral blood mononuclear cells (PBMC) were obtained from
venous blood which had been taken from donors into tubes containing
Heparin. Briefly, blood samples were diluted 1:1 with serum free
RPMI-1640 and 10 ml of diluted blood was laid onto a 5 ml
Ficoll-Hypaque gradient. After a centrifugation at 1200 rpm for 30
minutes, the PBMC at the interface were carefully removed and
washed twice in RPM I. The first centrifugation was performed for
10 minutes at 2000 rpm and the second for 10 minutes at 1200 rpm to
remove most platelets. PBMC were then diluted in RPM and kept in
sterile medium while processed.
[0123] Binding of HLA-E tetramer was monitored by flow cytometry
and cells were sorted. PBMC (5 x 106) were incubated for 15 minutes
at 37.degree. C. followed by 15 minutes at 4.degree. C. with 12
.mu.l of HLA-E tetramer labeled with phycoerythrin (PE). CD3
monoclonal antibody labeled with FITC was then added for another 15
minutes at 4.degree. C. Cells were then washed twice and sorted on
a FACScan, which measures fluorescent light emission and separates
distinct cell populations by electrostatic-deflection (electronic
cell sorting). Single cells or subsets of cells stained by HLA-E
tetramer were collected in sterile 96 well plates and put in
culture. NK cells (CD3-, HLA-E tet+) and T cells (CD3+, HLA-E tet+)
were grown in Yssel's medium (Yssel et al, 1984, J. Immunol.
Methods, 72:2199) in the presence of irradiated feeder cells (PBMC
and JY BCL), 0.1 .mu.g/ml of PHA, and 100 U/ml or 10 U/ml of IL-2
respectively. Autologous cells processed in this way are suitable
for reinjection into patients. FACS techniques can also be used to
count cells for quantization purposes.
[0124] Suitable methods for culturing NK cells and clones are
described in Litwin et al, J. Exp. Med, 1993,178:1321-1336. Methods
for maintaining T cells are described in Dunbar et al, Current
Biol., 1998, 8(7) 413 and Nixon et al, Nature, 1988. 336:
484-487.
Example 5
[0125] Isolation of CD94/NKG2+ Cells using HLA-E-coated beads
[0126] Cells expressing CD94/NKG2 receptors were isolated with
HLA-E-streptavidin coated DYNABEADS (superparamagnetic particles).
DYNABEADS M-280 Streptavidin are magnetic beads coated with
streptavidin. Soluble HLA-E was engineered with a biotinylation
site for BirA enzyme at the C terminus of HLA-E heavy chain and
refolded with 132 microglobulin and a synthetic peptide derived
from residues 3-11 of the signal sequence of some HLA molecules
(described in Braud et al, 1998, Nature, 391:795). These HLA-E
monomers were biotinylated using BirA enzyme and conjugated to
DYNABEADS M-280 using a standard protocol. Biotinylated HLA-E were
incubated with PBS-washed DYNABEADS M280 for 30 minutes at
4.degree. C. with bidirectional mixing (2 .mu.g HLA-E/10.sup.7
DYNABEADS). The beads were collected by placing the tube in a DYNAL
Magnetic Particle Concentrator (MPC) and the supernatant was
removed. The beads were washed 5 times in the same way. HLA-E
coated M-280 DYNABEADS were then mixed with isolated PBMC (obtained
as described in Example 3) (10.sup.7 beads/ml) and incubated for 20
minutes at 4.degree. C. with gentle rotation. The tube was placed
in a MPC (magnetic bead concentrator) and left to rest for 2
minutes. The supernatant was removed and the cells attached to the
beads washed 5 times. Cells were then grown as described in Example
3.
[0127] NB: To deplete PBMC of cells expressing receptors for HLA-E,
the supernatant is kept and the beads discarded.
Example 6
[0128] Targeted Killing of NK Cells
[0129] Recombinant biotinylated HLA-E is prepared as described in
Example 1. A mixture containing biotinylated HLA-E and a
biotinylated toxic agent such as the enzyme perforin in a molecular
ratio of 3:1 is combined with EXTRAVIDIN (modified egg white
avidin) to produce multimeric HLA-E linked to the toxic agent. A
PBMC sample from a human donor is prepared according to standard
techniques and contacted with the HLA-E reagent, resulting in
killing of the CD94/NKG2+ cells present in the sample. CD94/NKG2
negative cells are recovered.
Example 7
[0130] Xenotransplantation
[0131] A recombinant DNA expressing HLA-E in which the leader
sequence of HLA-E was replaced by the leader sequence of HLA-B8 was
generated. This chimeric cDNA contains the leader sequence of
HLA-B, and the extracellular (.alpha.1, .alpha.2 and .alpha.3), the
transmembrane and the cytoplasmic domains of HLA-E. It was
generated using the following oligonucleotide primers:
TABLE-US-00006 sense primer A: [SEQ ID NO: 24]
5'-CTCGGCGGCCCTGGCCCTGACCGAGACCTGGGCGGGCTCCCACTCC TTG-3' antisense
primer B: [SEQ ID NO: 25]
5'-TTCTGTCTAGATTACAAGCTGTGAGACTCAGACCCCTG-3' sense primer C: [SEQ
ID NO: 26] 5'-CTGACCGAATTCGCCGCCACCATGCTGGTCATGGCGCCCCGAACCG
TCCTCCTGCTGCTCTCGGCGGCCCTGGCC-3'
[0132] The PCR was performed using the cDNA of HLA-E as a template
with primers A and B. The product from that PCR was then used as a
template for a subsequent reaction with primers B and C. The last
product was digested with EcoRl and Xbal and ligated into the
expression vector pcDNA3.
[0133] Transgenic animals are then produced as follows. Females are
superovulated, mated to fertile males and sacrificed the following
day. Zygotes with two pronuclei are recovered and one of the
pronuclei is microinjected with the DNA expressing the HLA-E-HLA-B
leader sequence construct. Surviving embryos are reimplanted into
pseudopregnant foster females and DNA samples from new borns are
evaluated for the presence of the foreign gene (HLA-E construct).
These techniques are described in detail in the literature (e.g.,
Guide to techniques in mouse development, P. Wassarman and M
DePamphilis, Methods in Enzymology (Academic Press), Section X:
Transgenic animals: pronuclear injection (p747-802) and Section Xl:
Transgenic animals: embryonic stem cells and gene targeting
(p803-932)).
Example 8
[0134] Stable transfection of CD94 and NKG2 Genes into Mammalian
Cells
[0135] Mouse cells (P815, L cells) were sequentially transfected by
electroporation or calcium phosphate DNA precipitation
respectively, with a mammalian expression vector pcDNA3 (neomycin
resistance gene) containing CD94 cDNA and either NKG2A or NKG2C
with DAP12.
[0136] NKG2A and NKG2C cDNA was cloned into the expression vector
pcDNA3.1/hygro vector (containing the hygromycin resistance gene)
and DAP12 was cloned into the expression vector pcDNA3.1/zeo
(zeomycin resistance). P815 cells were electroporated with 500
.mu.F, 0.25 volt, and selection (G418 and hygromycin and zeomycin)
was added 2 days later. Cells expressing a high level of receptors
were sorted by flow cytometry. L cells were transfected by calcium
phosphate DNA precipitation and selected in the presence of G418,
hygromycin and zeomycin. Transfectants were cloned by limiting
dilution and cell surface expression of the CD94/NKG2 receptors was
monitored using specific antibodies.
[0137] The stable transfectants are useful for the identification
of antibodies or other agents that interfere with HLA-E binding to
CD94/NKG2. Agents which specifically interfere with HLA-E binding
to either inhibitory CD94/NKG2 receptors (e.g., CD94/NKG2A) or
stimulatory CD94/NKG2 receptors (e.g., CD94/NKG2C) can be
identified by performing binding assays using two different
transfectants. A list of known antibodies which interfere with
HLA-E binding to CD94/NKG2 is given below:
[0138] Antibodies which Block the Interaction between HLA-E and
CD94/NKG2 Receptors: [0139] 1--Anti-HLA-E:--3D12 (Lee et al, 1998,
J. Immunol. 160: 4951 [0140] 2--Anti-CD94:--HP3D9 (Perez-Villar et
al, 1995, J. Immunol. 154:5779) commercialised by Pharmingen [0141]
HP-3B1 (Aramburu et al, 1990, J. Immunol. 144:3238) commercialised
by Immunotech [0142] X1A85 (Sivori et al, 1996, Eur. J. Immunol.
26:2487) [0143] DX22 (Phillips et al, 1996, Immunity, 5: 163) DNAX
[0144] 3--Anti-NKG2A:--Z199 (Carreto et al, 1997, Eur. J. Immunol.
27:563) commercialised by Immunotech [0145] Z270 (Sivori et al,
1996, Eur. J. Immunol. 26: 2487)
[0146] The techniques described in this example may also be used to
transfect mammalian cells such as murine L cells with nucleic acids
encoding HLA-E as described herein.
Sequence CWU 1
1
2618PRTArtificialPeptides generated from MHC leader sequences 1Val
Met Ala Pro Arg Thr Leu Leu 1 5 29PRTArtificialPeptides generated
from MHC leader sequences 2Val Met Ala Pro Arg Thr Leu Leu Leu 1 5
39PRTArtificialPeptides generated from MHC leader sequences 3Val
Met Ala Pro Arg Thr Val Leu Leu 1 5 49PRTArtificialPeptides
generated from MHC leader sequences 4Val Met Ala Pro Arg Thr Leu
Ile Leu 1 5 59PRTArtificialPeptides generated from MHC leader
sequences 5Val Met Ala Pro Arg Thr Leu Phe Leu 1 5
69PRTArtificialPeptides generated from MHC leader sequences 6Val
Met Gly Pro Arg Thr Leu Val Leu 1 5 79PRTArtificialPeptides
generated from MHC leader sequences 7Val Thr Ala Pro Arg Thr Val
Leu Leu 1 5 89PRTArtificialPeptides generated from MHC leader
sequences 8Val Thr Ala Pro Arg Thr Leu Leu Leu 1 5
99PRTArtificialPeptides generated from MHC leader sequences 9Val
Met Pro Pro Arg Thr Leu Leu Leu 1 5 109PRTArtificialPeptides
generated from MHC leader sequences 10Val Met Glu Pro Arg Thr Leu
Ile Leu 1 5 119PRTArtificialPeptides generated from MHC leader
sequences 11Val Met Ala Pro Arg Ala Leu Leu Leu 1 5
1241DNAArtificialPrimer 12gtgggctaag cttacggctt ccatctcagg
gtgacgggct c 411340DNAArtificialPrimer 13ctacgggcat atggtagatg
gaaccctcct tttactctcc 401468DNAArtificialPrimer 14ccgtacctcg
agcatatggg ttctcattct ttaaaatatt ttcatacttc tgtatctaga 60cccggccg
681545DNAArtificialPrimer 15tggtgtctag aggatcctgg cttccatctc
agggtgacgg gctcg 451630DNAArtificialPrimer 16gcgtctagaa tggtggtcat
ggcaccccga 301730DNAArtificialPrimer 17catggagtgg gagccggccc
aggtctcggt 301821DNAArtificialPrimer 18ggctcccact ccatgaggta t
211923DNAArtificialPrimer 19aagctttcaa gctgtgagag aca
232030DNAArtificialPrimer 20accgagacct gggccgttga tgaacaactg
302129DNAArtificialPrimer 21gcaagcttct aaaggaagac ggtctgttc
292239DNAArtificialPrimer 22gggcgtcgac ccggactcag aatctcctca
gacgccgag 392330DNAArtificialPrimer 23cagttgttca tcaacggccc
aggtctcggt 302449DNAArtificialPrimer 24ctcggcggcc ctggccctga
ccgagacctg ggcgggctcc cactccttg 492538DNAArtificialPrimer
25ttctgtctag attacaagct gtgagactca gacccctg
382675DNAArtificialPrimer 26ctgaccgaat tcgccgccac catgctggtc
atggcgcccc gaaccgtcct cctgctgctc 60tcggcggccc tggcc 75
* * * * *