Copyright © 2000 by Lippincott Williams & Wilkins, Inc.
Vol. 70, 662–667, No. 4, August 27, 2000
Printed in U.S.A.
AN IMMUNOSUPPRESSIVE AND ANTI-INFLAMMATORY HLA
CLASS I-DERIVED PEPTIDE BINDS VASCULAR CELL ADHESION
XUEFENG LING,2TOMOAKI TAMAKI,2YUN XIAO,2SALAR KAMANGAR,2CAROL CLAYBERGER,2,3
DAVID B. LEWIS,2AND ALAN M. KRENSKY2,4
Division of Immunology and Transplantation Biology, Department of Pediatrics, and
the Department of Cardiothoracic Surgery, Stanford University, California 94305
Background. A synthetic peptide corresponding to
residues 75–84 of HLA-B2702 modulates immune re-
sponses in rodents and humans both in vitro and in
Methods. We used a yeast two-hybrid screening, an
in vitro biochemical method, and an in vivo animal
Results. Two cellular receptors for this novel immu-
nomodulatory peptide were identified using a yeast
two-hybrid screen: immunoglobulin binding protein
(BiP), a member of the heat shock protein 70 family,
and vascular cell adhesion molecule (VCAM)-1. Identi-
fication of BiP as a ligand for this peptide confirms
earlier biochemical findings, while the interaction
with VCAM-1 suggests an alternative mechanism of
action. Binding to the B2702 peptide but not to closely
related variants was confirmed by ligand Western blot
analysis and correlated with immunomodulatory ac-
tivity of each peptide. In mice, an ovalbumin-induced
allergic pulmonary response was blocked by in vivo
administration of either the B2702 peptide or anti-
Conclusions. We propose that the immunomodula-
tory effect of the B2702 peptide is caused, in part, by
binding to VCAM-1, which then prevents the normal
interaction of VCAM-1 with VLA-4.
HLA molecules are important in antigen presentation and
as targets for transplant rejection (1). However, the recent
discoveries of killer inhibitory receptors (KIR) (2) and immu-
nomodulatory effects of HLA-derived peptides (3) indicate
that HLA molecules also act to suppress the immune re-
sponse. We previously described a series of peptides corre-
sponding to linear sequences of HLA molecules with pro-
found immunomodulatory effects in vitro and in vivo in both
animal models and humans (4–11). One of these peptides,
corresponding to the ?1 alpha helix of HLA-B2702, blocks cell
mediated cytotoxicity in an allele unrestricted manner. The
immunomodulatory effects of the B2702 peptide are entirely
attributable to the carboxyl terminal 10 amino acids (resi-
dues 75–84), and an inverted dimer peptide (B2702.84–75/
75–84) is the most potent variant peptide tested (12). A
monomeric B2702 peptide blocks cell-mediated immunity in
transplantation patients treated with otherwise standard
therapy in a placebo controlled phase II trial (13). To eluci-
date the mechanism of action of this peptide, molecules bind-
ing to the B2702 peptide were purified by standard immuno-
affinity techniques. The major proteins isolated by this
method were the constitutive and inducible forms of heat
shock protein (HSP) 70 (13), but it remains unclear exactly
how the interaction of the B2702 peptide with HSP 70 affects
cell mediated lysis. In this report, the yeast-two hybrid sys-
tem, developed by Field and co-workers (14), was used to
identify additional cellular ligands for the peptide. Immuno-
globulin binding protein (BiP), a member of the HSP70 fam-
ily, and VCAM-1 were identified in this manner, and the
importance of this vascular cell adhesion molecule (VCAM-1)
interaction was evaluated in a mouse allergy model.
MATERIALS AND METHODS
HLA-derived peptides. All peptides were synthesized and purified
as described (10). The amino acid sequence of the B2702.84–75/75–
84, B2702.84–75T/75–84T, and B0701.84–75/75–84 peptides are
shown in Table 1. For biochemical analyses, a biotin group was
attached to the amino terminus using N-hydroxyl-succinimidyl-ester
(NHS) activated biotin (NHS-LC-Biotin II: Pierce Chemical Co.,
Rockford, IL). Biotinylation did not affect the immunomodulatory
activity of these peptides (12).
Plasmid construction and yeast two-hybrid screen. The MATCH-
MAKER Two-Hybrid System 2 kit (Clontech Laboratories, Inc., Palo
Alto, CA) was used for yeast-two hybrid analysis. An oligonucleotide
encoding the B2702.84–75/75–84 sequence was cloned into the car-
boxyl terminal end of the GAL4 DNA binding domain of the plasmid
vector pAS1 (a gift from S. J. Elledge, Baylor College of Medicine),
generating plasmid pXL-23 (referred to as the DNA-binding domain
[DNA-BD]/B2702 vector construct and as “bait”). mRNA was ex-
tracted from human peripheral blood lymphoctyes PBLs (Stanford
University Blood Bank) 6 days after phytohemagglutinin (PHA)
activation. cDNA was synthesized using random and oligo(dT) prim-
ing and inserted into the plasmid pACT2 (Clontech Laboratories,
Inc.) (complexity ? 106, referred to as the activation domain [AD]/
library vector construct). The library and pXL-23 plasmid were co-
transformed into yeast strain Y190 (from Clontech), and screens
were performed as suggested by the manufacturer. Potential inter-
acting clones isolated in the library screen were retransformed with
pXL-23 or with other “bait” plasmid constructs (DNA-BD/CDK2,
DNA-BD/lamin, DNA-BD/p53, DNA-BD/syntaxin). Clones which al-
lowed positive yeast two-hybrid interactions with pXL-23 but not
other “bait” plasmids were designated positive. Library plasmids
were isolated from positive clones (15), amplified, and sequenced
(Sequenase 1.0 DNA Sequencing Kit, U.S. Biochemical, Cleveland,
1This work was supported by grants from the National Institutes
of Health (PO1 AI41520 to A.M.K. and C.C. and R01 AI44699 to
D.B.L.), the Walter V. and Idun Berry Award (X.L.), and Howard
Hughes summer fellowships (T.T. and S.K.).
2Division of Immunology and Transplantation Biology.
3Department of Cardiothoracic Surgery.
4Address correspondence to: Alan M. Krensky, M.D., Department
of Pediatrics, Stanford University, Stanford, CA 94305.
OH). The DNA sequences generated were analyzed for homologic
features in the GenBank database (http://www.ncbi.nlm.nih.gov).
Preparation of cell lysates. PBL were isolated by Ficoll-Hypaque
density centrifugation and cultured for 3 days in RPMI-1640 supple-
mented with 10% fetal calf serum (Hyclone Laboratories, Logan,
UT), 2 mM L-glutamine, 100 U/ml penicillin, 100 ?g/ml streptomy-
saline (PBS) and resuspended them in 300 ?l of ice-cold Tris/saline/
?l of ice-cold lysis buffer (2% Triton X-100, 1 mM PMSF in TSA) to the
cell solution and stirred for 1 hr at 4°C. The cell solution was centri-
fuged for 10 min at 4000 ?g, and the supernatant was transferred to a
fresh tube. We added 100 ?l of 5% sodium deoxycholate (Na-DOC) TSA,
and the mixture was incubated for 10 min at 4°C.
Precipitation of peptide-binding proteins. The cell lysate was pre-
cleared with ImmunoPure? Immobilized Streptavidin (Pierce), bio-
tinylated B2702.84–75/75–84 or biotinylated B2702.84–75T/75–84T
(40 ?M final concentration) was added to 200 ?l aliquots, and the
slurry was incubated at 4°C for 2 hr. We then added 50 ?l of
ImmunoPure? Immobilized Streptavidin and incubated the mixture
for 1 hr. The mixture was centrifuged and the streptavidin agarose
precipitate was washed three times with 1 ml of the lysis buffer.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-
PAGE), membrane transfer, and immunoblotting. We added 30 ?l of
SDS-PAGE sample buffer (62.5 mM Tris-HCl, pH 6.8, 20% glycerol,
2% SDS, 5% ?-mercaptoethanol) to the precipitated streptavidin
agarose and boiled the mixture, and eluted proteins were separated
on SDS-PAGE (10% acrylamide gel). The proteins were transferred
to a polyvinyldifluoride (PVDF) membrane (BioRad, Hercules, CA).
The membrane was incubated with mouse anti-HSP70 monoclonal
antibody (mAb) (Sigma), diluted 1:1000, washed, and incubated with
horseradish peroxidase-conjugated rabbit anti-mouse antibody (Ab)
(Amersham) diluted 1:2000. Immunodetection using enhanced
chemiluminescent method (ECL, Amersham) was performed accord-
ing to the manufacturer’s instruction.
Ligand Western blot. We separated 0.1 ?g of recombinant human
vascular cell adhesion molecule-1 (rVCAM-1, R&D System, Minne-
apolis, MN), recombinant HSP70 (amino acids 386–646, a gift from
D. Hanson, Stanford University), and OVA (ICN Biomedicals, Inc.)
using SDS-PAGE and transferred them to a PVDF membrane. The
membrane was washed twice for 10 min in Tris-buffered saline (TBS)
with 0.1% Tween 20 (TBS-T) and once for 30 min in TBS with 3%
NP-40, blocked for 2 hr with 1% bovine serum albumin (BSA) (Sig-
ma) in TBS at room temperature, and washed again with TBS-T for
10 min. The hybridization solution was 1% BSA in TBS-T supple-
mented with 10 ?M of biotinylated peptide. The membrane was
hybridized overnight in a sealed bag at 4°C, washed with TBS-T at
room temperature once for 15 min, and twice for 5 min. The mem-
brane was then incubated with ExtrAvidin?-Peroxidase (Sigma) di-
luted to 1:2000 in 1% BSA TBS-T for 1 h and washed again once for
15 min and twice for 5 min. The bands were detected by ECL
(Amersham). The membrane was then stripped with 100 mM ?-mer-
captoethanol, 2% SDS, 62.5 mM Tris-HCl, pH 6.7 at 50°C for 30 min
and probed with either goat polyclonal anti-human VCAM-1 Ab
(R&D Systems) and HRP-conjugated rabbit anti-goat immunoglob-
ulin (Ig) antibody (Sigma) or mouse anti-HSP70 Ab (Sigma) and HRP-
conjugated rabbit anti-mouse Ig Ab, or rabbit anti-OVA Ab and goat
anti-rabbit Ig Ab (Amersham). ECL was used for immunodetection.
Mouse allergic pulmonary disease. Mouse allergic pulmonary dis-
ease was induced as previously reported (16). Female BALB/c mice
were purchased from Jackson Laboratories (Bar Harbor, ME) and
were housed under specific pathogen-free conditions. Mice received
single i.p. injections of 100 ?g alum-precipitated crystalline OVA
(Pierce) in 200 ?l PBS on days 1 and 14 and intranasal (i.n.) doses
consisted of 50 ?g of OVA in 50 ?l of PBS on days 14, 25, 26, and 27
(16). Two hr before each i.n. dose of OVA, 50 ?l of PBS containing 5
?g of peptide, or 0.2% dimethyl sulfoxide (DMSO) (a concentration
equivalent to that in the peptide solutions), was given i.n.. Anti-
VLA-4 Ab-treated mice were given 25 ?g of VLA-4 (CD49d) PS/2
mAb (Southern Biotechnology Associates, Inc., Birmingham, AL) in
50 ?l of PBS i.n. 2 hr before each i.n. dose of OVA. Twenty-four hr
after the last i.n. administration, mice were killed and BAL was
performed on the left lung. BAL fluid cell counts and staining for
eosinophils were performed as previously described (16).
The yeast two-hybrid screen identifies a HSP70 family
member, BiP, as a B2702.84–75/75–84-binding protein. Us-
ing the B2702.84–75/75–84 peptide fusion construct (“bait”)
and a cDNA library from activated PBL in a yeast two-hybrid
screen, 17 positive clones were identified. Clones were ex-
cluded if they did not demonstrate yeast two-hybrid binding
specificity in the ?-galactosidase specificity assay. An HSP70
family member, BiP, was identified as a B2702.84–75/75–84-
binding protein. The region of the BiP molecule identified
was residues 466–654, and its interaction with the B2702
peptide is specific. The isolated BiP clone did not bind to
irrelevant “bait” constructs, including CDK2, lamin, p53, and
syntaxin. Sequence analysis (GCG software, bestfit program)
revealed residues 466–654 of BiP and residues 386–640
HSP70 (B2702 peptide binding domain in HSP70, Clayberger
and Krensky, unpublished) were highly homologous (Fig. 1).
The interaction of the B2702 peptide with HSP proteins was
also demonstrated biochemically. Biotinylated B2702.84–75/
TABLE 1. Sequences of class I HLA-derived peptides
Y R L A I R L N E R R E N L R I A L R Y
Y R L A T R L N E R R E N L R T A L R Y
Y G R L N R L S E R R E S L R N L R G Y
TABLE 2. Clones identified by the yeast two-hybrid screen
Homology to sip110b
aPositive in ? galactosidase colony-lift assay and His 3 expression
as judged by growth on plates containing 3-aminotriazole.
bVery short open reading frame.
LING ET AL.
August 27, 2000
75–84 peptide and streptavidin agarose were used to precip-
itate B2702 peptide-binding proteins from PBL lysates (3
days after activation with PHA). Western blot analysis of the
precipitated material with anti-HSP70 antibody revealed
bands at 70 kDa and 74 kDa (Fig. 2). Whereas the functional
peptide (B2702.84–75/75–84) bound HSP70 proteins, a non-
functional peptide in which threonine was substituted for
isoleucine (B2702.84–75T/75–84T) bound only weakly to
HSP70 proteins. Thus, the yeast two-hybrid system confirms
previous findings that this immunosuppressive peptide in-
teracts with members of the HSP70 family (12).
B2702.84–75/75–84 peptide binds specifically to VCAM-1
in vitro. VCAM-1 was another ligand for the B2702 peptide
identified in a yeast two-hybrid screen. The portion of
VCAM-1 isolated corresponds to amino acids 75–482, span-
ning the middle of domain 1 through domain 5 of the
VCAM-1 molecule (Fig. 3). To verify the interaction of
VCAM-1 and the B2702 peptide, ligand Western blot analy-
sis was performed. Samples of recombinant VCAM-1, the
peptide-binding domain of HSP70 (amino acids 386–646),
and OVA were separated by SDS-PAGE and transferred to
PVDF membrane. Membranes were then incubated with bi-
otinylated peptides (B2702.84–75/75–84, B2702.84–75T/75–
84T, B0701.84–75/75–84), and, after extensive washing, pro-
tein-peptide interactions were visualized with streptavidin
peroxidase. The B2702.84–75/75–84 bound to both VCAM-1
and HSP70, whereas the nonfunctional B2702.84–75T/75–
84T peptide did not bind to VCAM-1 and bound only weakly
to HSP70 (Fig. 4). Another HLA-derived peptide that does
not affect cytotoxicity, B0701.84–75/75–84, failed to bind
either VCAM-1 or HSP70. None of the peptides bound to
OVA. Thus, the ligand Western blot both confirms the inter-
action of the B2702 peptide with VCAM-1 and demonstrates
that the interaction is direct, not requiring intermediate
factors. The correlation between in vitro VCAM-1-B2702 pep-
tide binding and in vivo immunosuppressive effects suggests
that the peptide-VCAM-1 interaction may be involved in the
in vivo functional effects of the peptide.
Both the B2702.84–75/75–84 peptide and anti-VLA-4 an-
tibody block the immune response in a murine allergic
asthma model. The specific interaction of the B2702.84–75/
75–84 peptide and VCAM-1 indicated that the peptide may
exert its functional effects in part through blockade of cell-
cell adhesion. Therefore, the effect of the B2702.84–75/75–84
peptide on an allergic pulmonary inflammatory response was
evaluated in mice. The number of leukocytes and eosinophils
infiltrating the airways was significantly reduced in mice
that had been given B2702.84–75/75–84 peptide i.n. 2 hours
before challenge with i.n. OVA (Fig. 5; P?0.001 vs. the
DMSO control by ANOVA). Administration of the B0701.84–
75/75–84 peptide had no effect (P?0.05). In contrast, i.n.
administration of anti-VLA-4 mAb showed a similar, signif-
icant decrease in the allergic pulmonary response (P?0.001
vs. DMSO control by ANOVA).
A synthetic peptide corresponding to the ?1 alpha helix of
HLA-B2702 has immunomodulatory effects both in vitro and
in vivo (7, 12, 17–20). Using the yeast two-hybrid system, we
show here that this peptide binds to BiP and VCAM-1. Pre-
viously our group used biochemical techniques to demon-
strate that HSP70 family members bound to the B2702 pep-
tide in a sequence specific manner (12). Identification of
VCAM-1 as an additional ligand is sequence specific and corre-
lates with in vitro and in vivo functional activity. Both the
B2702 peptide and anti-VLA-4 antibody inhibit pulmonary al-
lergic disease in a mouse model, suggesting that disruption of
the normal VCAM-1-VLA-4 interaction underlies some of the
immunomodulatory effects of the B2702 peptide in vivo.
VCAM-1 plays an essential role in leukocyte adhesion and
infiltration into inflammatory sites (21–24). VCAM-1 expres-
sion is induced by proinflammatory cytokines, including in-
terleukin-1, TNF, and interleukin-4 (21, 22, 25, 26). Up-
FIGURE 1. Sequence homology of regions of BiP and HSP70.
Residues 386–540 of HSP70 contain the B2702 binding do-
main. The region of BiP protein identified in the yeast-two
hybrid system is residues 466–654.
FIGURE 2. Precipitation of B2702-binding proteins using the
biotinylated B2702 peptides and streptavidin agarose. The
membrane was probed with anti-HSP70 antibody.
Vol. 70, No. 4
regulation of VCAM-1 on vascular endothelium leads to the
recruitment of VLA-4-expressing leukocytes to the site of
antigenic stimulation. The importance of this interaction in
cell adhesion, signal transduction, and cellular activation has
been documented in several model systems and targeted for
therapeutic intervention (27). Administration of mAb recog-
nizing VCAM-1 or VLA-4 inhibits disease in a variety of
models, including porcine lung disease and murine cardiac
transplantation (28–32). Administration of soluble VCAM-1
reduces inflammation in diabetes and other models (33).
Human VCAM-1 is a type I (extracellular N-terminus)
transmembrane glycoprotein consisting of seven Ig domains
(34). The region of VCAM-1 found to bind B2702.84–75/
75–84 in the yeast two-hybrid screen (residues 75–482) in-
cludes portions of domain 1 through domain 4 (Fig. 3). Do-
mains 1 and 4, especially the conserved IDSP motifs, are
crucial for VLA-4 binding (35, 36). These IDSP motifs are
functionally related to the RGD motif, a common signal for
integrin binding to extracellular matrix proteins like fi-
bronectin. Residues 75–482 of VCAM-1 include the IDSP
sequence of domain 4 but not the IDSP motif from domain 1.
A synthetic peptide corresponding to the IDSP motif did not
affect binding of the B2702 peptide to recombinant VCAM-1
(data not shown), indicating either that the B2702 peptide
does not bind to this motif or that additional residues con-
tained in the 75–482 region contribute to the interaction.
The sequence specificity of the B2702 peptide binding to
VCAM-1 in vitro and inhibition of lymphocyte function both
in vitro and in vivo suggest that this finding is functionally
relevant. The established role of VCAM-1 in allograft rejec-
tion and the ability of mAbs to inhibit graft rejection suggest
that the HLA-derived peptide may function at least in part in
a similar manner to disrupt the VCAM-1 -VLA-4 interaction.
The established role of VCAM-1 in a broad range of inflam-
matory processes led us to test the effect of the B2702 peptide
in an OVA-induced asthma model in mice. This model is
characterized by prominent infiltration of eosinophils into
the airway, similar to what is observed in severe allergen-
FIGURE 4. Ligand Western blot of re-
combinant VCAM-1, HSP70 peptide-
binding domain, and OVA probed
with: (a) biotinylated B2702.84–75/
75–84, (b) biotinylated B2702.84–
B0701.84–75/75–84, or (d) no pep-
FIGURE 3. VCAM-1 domain structure. The underlined region indicates the region identified in the yeast-two hybrid system.
Each domain consists of one unit of Ig-like structure. The IDSP sequences (****) in domains 1 and 4 are identified as the VLA-4
(?4?1 integrin) binding sequence.
LING ET AL.
August 27, 2000
induced asthma in humans (16). MAbs specific for either
VCAM-1 or VLA-4 inhibit eosinophil and T lymphocyte infil-
tration in mouse trachea, whereas anti-LFA-1 and ICAM-1
antibodies have weaker effects (37). Administration of the
B2702 peptide significantly decreased total leukocyte and
eosinophil infiltration compared with the control and in a
manner similar to anti-VLA-4. This result supports the
model that the B2702 peptide functions at least in part
through the VCAM-1 -VLA-4 interaction.
The mechanism of action of the B2702 peptide is likely not
limited to blockade of leukocyte adhesion and infiltration.
The VCAM-1 -VLA-4 interaction has also been implicated as
a co-stimulatory pathway (38, 39). The results to date sug-
gest that effects on T cell activation and anergy induction are
also relevant to the mechanism of action of the B2702 pep-
tide. The selective binding to HSP70 proteins may also be
important in this regard, although the precise role of this
interaction remains unclear. In any case, HLA derived pep-
tides are a potentially important new class of drugs capable
of inducing and maintaining immunologic tolerance. The role
of HLA molecules and peptides in down-regulation of the
immune response, including inflammatory reactions, is im-
portant both biologically and therapeutically.
Acknowledgments. We thank R. Jung for peptide synthesis, D.
Chen and H. Zhang for excellent technical assistance.
1. Germain R. MHC-dependent antigen processing and peptide
presentation: providing ligands for T lymphocyte activation.
Cell 1994; 76: 287.
2. Lanier LL. Natural killer cell receptors and MHC class I inter-
actions. Curr Opin Immunol 1997; 9: 126.
3. Krensky AM, Clayberger C. HLA derived peptides as a novel
strategy for the prevention of allograft rejection. Exp Opin
Invest Drugs 1996; 5: 809.
4. Parham P, Clayberger C, Zorn SL, Ludwig DS, Schoolnik GK,
Krensky AM. Inhibition of alloreactive cytotoxic T lymphocytes
by peptides from the alpha 2 domain of HLA-A2. Nature 1987;
5. Clayberger C, Parham P, Rothbard J, Ludwig DS, Schoolnik GK,
Krensky AM. HLA-A2 peptides can regulate cytolysis by hu-
man allogeneic T lymphocytes. Nature 1987: 330: 763.
6. Krensky AM, Lyu SC, Pouletty P, Benjamin R, Parham P, Clay-
berger C. Peptides corresponding to the CD8 binding region of
HLA class I block the differentiation of cytotoxic T lymphocyte
precursors. Transplant Proc 1993; 25: 483.
7. Clayberger C, Lyu SC, Pouletty P, Krensky AM. Peptides corre-
sponding to T-cell receptor-HLA contact regions inhibit class
I-restricted immune responses. Transplantation Proc 1993; 25:
8. Clayberger C, Lyu SC, DeKruyff R, Parham P, Krensky AM.
Peptides corresponding to the CD8 and CD4 binding domains
of HLA molecules block T lymphocyte immune responses in
vitro. J Immunol 1994; 153: 946.
9. Nisco S, Vriens P, Hoyt G, et al. Induction of allograft tolerance
in rats by an HLA class-I-derived peptide and cyclosporine A.
J Immunol 1994; 152: 3786.
10. Clayberger C, Lyu SC, Pouletty P, Krensky AM. Peptides corre-
sponding to T-cell receptor-HLA contact regions inhibit class
I-restricted immune responses. Transplant Proc 1993; 25: 477.
11. Boytim ML, Lyu SC, Jung R, Krensky AM, Clayberger C. Inhibi-
tion of cell cycle progression by a synthetic peptide corresponding
to residues 65–79 of an HLA class II sequence: functional simi-
larities but mechanistic differences with the immunosuppressive
drug rapamycin. J Immunol 1998; 160: 2215.
12. Noessner E, Goldberg JE, Naftzger C, Lyu SC, Clayberger C,
Krensky AM. HLA-derived peptides which inhibit T cell func-
tion bind to members of the heat-shock protein 70 family. J
Exp Med 1996; 183: 339.
13. Giral M, Cuturi MC, Nguyen JM, et al. Decreased cytotoxic
activity of natural killer cells in kidney allograft recipients
treated with human HLA-derived peptide. Transplantation
1997; 63: 1004.
14. Fields S, Song O. A novel genetic system to detect protein-
protein interactions. Nature 1989; 340: 245.
15. Robzyk K, Kassir Y. A simple and highly efficient procedure for
rescuing autonomous plasmids from yeast. Nucleic Acids Res
1992; 20: 3790.
16. Zhang Y, Lamm WJE, Albert RK, Chi EY, Henderson WR Jr,
Lewis DB. Influence of the route of allergen administration
and genetic background on the murine allergic pulmonary
response. Am J Resp Crit Care Med 1997; 155: 661.
17. Clayberger C, Rosen M, Parham P, Krensky AM. Recognition of
an HLA public determinant (Bw4) by human allogeneic cyto-
toxic T lymphocytes. J Immunol 1990; 144: 4172.
18. Krensky AM, Clayberger C. The induction of tolerance to alloan-
tigens using HLA based synthetic peptides. Curr Opin Immu-
nol 1994; 6: 791.
19. Buelow R, Veyron P, Clayberger C, Pouletty P, Touraine JL.
Prolongation of skin allograft survival in mice following ad-
ministration of ALLOTRAP. Transplantation 1995; 59: 455.
20. Cuturi M-C, Josien R, Douilard P, et al. Prolongation of alloge-
neic heart graft survival in rats by administration of a peptide
(a.a. 75–84) from the alpha 1 helix of the first domain of
HLA-B701. Transplantation 1995; 59: 661.
21. Osborn L. Leukocyte adhesion to endothelium in inflammation.
Cell 1990; 62: 3.
22. Springer TA. Adhesion receptors of the immune system. Nature
1990; 346: 425.
23. Pelletier RP, Ohye RG, Vanbuskirk A, et al. Importance of en-
dothelial VCAM-1 for inflammatory leukocytic infiltration in
vivo. J Immunol 1992; 149: 2473.
24. Nakajima H, Nishimura T, Yoshida S, Iwamoto I. Role of vascu-
lar cell adhesion molecule 1/very late activation antigen 4 and
intercellular adhesion molecule 1/lymphocyte function-associ-
FIGURE 5. The effect of anti-VLA-4 mAb, the B2702.84–75/
75–84 peptide, and the irrelevant peptide on leukocyte infil-
tration into the airways in the mouse OVA-induced asthma
model. The number of leukocytes in BAL fluid was signifi-
cantly reduced in the anti-VLA-4 mAb and B2702.84–75/75–84-
treated groups compared with the DMSO control (P<0.001 by
ANOVA), where the cell counts of the irrelevant peptide-
treated group and the DMSO control group did not differ
significantly (P>0.05). The percentage of leukocytes that
were eosinophils ranged between 61% and 78% for all groups,
and ANOVA showed no significant difference between leuko-
cytes that were eosinophils and those that were not.
Vol. 70, No. 4
ated antigen 1 interactions in antigen-induced eosinophil and
T cell recruitment into the tissue. J Exp Med 1994; 179: 1145.
25. Osborn L, Hession C, Tizard R, et al. Direct expression cloning of
vascular cell adhesion molecule 1, a cytokine-induced endothe-
lial protein that binds to lymphocytes. Cell 1989; 59: 1203.
26. Gearing AJ, Newman W. Circulating adhesion molecules in dis-
ease. Immunol Today 1993; 14: 506.
27. Foster CA. VCAM-1/alpha 4-integrin adhesion pathway: thera-
peutic target for allergic inflammatory disorders. J Allergy
Clin Immunol 1996; 98: S270.
28. Chisholm PL, Williams CA, Lobb RR. Monoclonal antibodies to
the integrin alpha 4 subunit inhibit the murine contact hyper-
sensitivity response. Eur J Immunol 1993; 23: 682.
29. Weg VB, Williams TJ, Lobb RR, Nourshargh S. A monoclonal
antibody recognizing very late activation antigen-4 inhibits
eosinophil accumulation in vivo. J Exp Med 1993; 177: 561.
30. Pretolani M, Ruffie C, Lapa e Silva JR, Joseph D, Lobb RR, Var-
gaftig BB. Antibody to very late activation antigen 4 prevents
antigen-induced bronchial hyperreactivity and cellular infiltra-
tion in the guinea pig airways. J Exp Med 1994; 180: 795.
31. Tsukamoto K, Yokono K, Amano K, et al. Administration of
monoclonal antibodies against vascular cell adhesion molecule/
very late antigen-4 abrogates predisposing autoimmune diabe-
tes in NOD mice. Cell Immunol 1995; 165: 193.
32. Fryer AD, Costello RW, Yost BL, et al. Antibody to VLA-4, but not
to L-selectin, protects neuronal M2 muscarinic receptors in anti-
gen-challenged guinea pig airways. J Clin Invest 1997; 99: 2036.
33. Jakubowski A, Ehrenfels BN, Pepinsky RB, Burkly LC. Vascular
cell adhesion molecule-Ig fusion protein selectively targets ac-
tivated alpha 4-integrin receptors in vivo: inhibition of auto-
immune diabetes in an adoptive transfer model in nonobese
diabetic mice. J Immunol 1995; 155: 938.
34. Chothia C. The molecular structure of cell adhesion molecules.
Annu Rev Bioche 1997; 66: 823.
35. Vonderheide RH, Tedder TF, Springer TA, Staunton DE. Resi-
dues within a conserved amino acid motif of domains 1 and 4 of
VCAM-1 are required for binding to VLA-4. J Cell Biol 1994;
36. Clements JM, Newham P, Shepherd M, et al. Identification of a
key integrin-binding sequence in VCAM-1 homologous to the
LDV active site in fibronectin. J Cell Sci 1994; 107: 2127.
37. van Seventer GA, Newman W, Shimizu Y, et al. Analysis of T cell
stimulation by superantigen plus major histocompatibility
complex class II molecules or by CD3 monoclonal antibody:
costimulation by purified adhesion ligands VCAM-1, ICAM-1,
but not ELAM-1. J Exp Med 1991; 174: 901.
38. Damle NK, Klussman K, Linsley PS, Aruffo A. Differential co-
stimulatory effects of adhesion molecules B7, ICAM-1, LFA-3,
and VCAM-1 on resting and antigen-primed CD4? T lympho-
cytes. J Immunol 1992; 148: 1985.
39. Schlegel PG, Vaysburd M, Chen Y, Butcher EC, Chao NJ. Inhi-
bition of T cell costimulation by VCAM-1 prevents murine
graft-versus-host disease across minor histocompatibility bar-
riers. J Immunol 1995; 155: 3856.
Received 21 October 1999.
Accepted 17 March 2000.
Copyright © 2000 by Lippincott Williams & Wilkins, Inc.
Vol. 70, 667–673, No. 4, August 27, 2000
Printed in U.S.A.
ROLE AND REGULATION OF PIG CD59 AND MEMBRANE
COFACTOR PROTEIN/CD46 EXPRESSED ON PIG AORTIC
CARMEN W. VAN DEN BERG,2,3COLIN RIX,4S. MELANIE HANNA,5
JOSE M. PEREZ DE LA LASTRA,5AND B. PAUL MORGAN5
Departments of Pharmacology, Therapeutics and Toxicology, Cardiology, and Medical Biochemistry, U.W.C.M.,
Heath Park, Cardiff CF144XN, United Kingdom
Background. Hyperacute rejection in xenotrans-
plantation is caused by activation of complement (C)
on endothelium. We have previously shown that puri-
fied C-regulators of the pig (CD59 and membrane co-
factor protein [MCP]) are efficient regulators of hu-
man C (HuC). The aim of this study was to clarify the
role of endogenously expressed C-regulatory mole-
cules on pig endothelium in the protection against
Methods. Porcine aortic endothelial cells (PAEC)
were harvested and cultured for various passages.
PAEC were examined for the expression of endoge-
nous pig CD59 and MCP by flow cytometry. PAEC
were assessed for their susceptibility to lysis by HuC.
The effect of phorbol 12-myristate 13-acetate and var-
ious cytokines on the expression of MCP and CD59 and
C-susceptibility was assessed.
Results. Primary PAEC showed an initial high level
of expression of pig CD59, however, upon culturing,
CD59 levels decreased dramatically to about 20% after
1This work was supported by a Royal Society grant to C.W.B. and
a Wellcome Trust grant to B.P.M. J.M.P.L. is a recipient of a Marie
2Department of Pharmacology, Therapeutics and Toxicology.
3Address correspondence to: Carmen W. van den Berg, Depart-
ment of Pharmacology, Therapeutics and Toxicology, Wales Heart
Research Institute, University of Wales College of Medicine, Heath
Park, Cardiff CF144XN, UK. E-mail: firstname.lastname@example.org.
4Department of Cardiology.
5Department of Medical Biochemistry.
VAN DEN BERG ET AL.
August 27, 2000