close

Вход

Забыли?

вход по аккаунту

?

Negatively charged residues interacting with the p4 pocket confer binding specificity to DRB1.178038120401

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 38, No. 12, December 1995, pp 1744-1753
0 1995, American College of Rheumatology
1744
NEGATIVELY CHARGED RESIDUES INTERACTING WITH THE
p4 POCKET CONFER BINDING SPECIFICITY TO DRB 1*0401
SUSAN L. WOULFE, CHRISTINE P. BONO, MICHELLE L. ZACHEIS, DAWN A. KIRSCHMANN,
TROY A. BAUDINO, CRAIG SWEARINGEN, ROBERT W. KARR, and BENJAMIN D. SCHWARTZ
Objective. To identify critical residues involved in
the binding of a selective peptide to DRB1*0401.
Methods. The binding of peptides to native or
site-directed mutant DR molecules was evaluated using
enzyme-linked immunosorbent assay and flow cytometry.
Results. Amino acid substitutions at DR and
peptide residues, which were predicted to contribute to
interactions within the DR p4 pocket, had the greatest
effects on the specificity of binding.
Conclusion. Differences in the peptide-binding
repertoires of DR molecules may contribute to associations with autoimmune diseases.
Immune responses are determined, in part, by
the assortment of HLA class I1 genes inherited by an
individual (1). Recognition of peptide-class I1 molecular complexes by CD4-positive T cells initiates the
cascade of events that result in an immune response
and, occasionally, an autoimmune response (2). Of all
the HLA class I1 molecules, those encoded by the
DRBl genes are the most polymorphic (3). The polymorphic residues of the DRpl chains are located in 3
distinct hypervariable regions in the primary sequence
(4). As molecular modeling studies have predicted
(5,6), and crystallographic analyses of the HLA-DR1
molecule have shown (7,8), these polymorphic residues are predominantly located on one-half of the
molecule, composed of both the floor (beta-pleated
sheet structure) and one side (alpha-helical structure)
of the peptide-binding groove. While some of the
Supported in part by NIH grant AI-32764.
Susan L. Woulfe, PhD, Christine P. Bono, BS, Michelle L.
Zacheis, BS, Dawn A. Kirschmann, PhD, Troy A. Baudino, BS,
Craig Swearingen, BS, Robert W. Karr, MD, Benjamin D.
Schwartz, MD, PhD: G . D. Searle & Co., St. Louis, Missouri.
Address reprint requests to Susan L. Woulfe, PhD, G. D.
Searle/Monsanto Co., 700 Chesterfield Village Parkway North, Box
AA4C, St. Louis, MO 63198.
Submitted for publication April 24, 1994; accepted in revised form June 26, 1995.
polymorphic residues located on the side of the groove
directly interact with the T cell receptor (9,10), the
polymorphic floor residues and several of the side
residues appear to either interact with the peptide or
influence peptide conformation (1 1-14).
Polymorphic differences between class I1 molecules contribute to selective immune responsiveness
by influencing the repertoire of peptides which bind to
the class I1 molecules (14). As such, class I1 polymorphisms almost certainly play a role in the development
of autoimmune diseases (15). For example, individuals
who inherit genes that encode DR4 or DR1 molecules
are more likely to develop rheumatoid arthritis (RA)
than individuals who do not express these molecules
(16). Because the expression of certain DR-peptide
complexes could be etiologic for RA, much effort has
been expended in the identification of specific binding
motifs. Several approaches have previously been utilized to analyze these peptides that bind to DR4 and DR1
molecules, including sequencing naturally bound peptides (17,18), screening phage display libraries (19,20),
and evaluating large numbers of synthetic peptides
(21-23). These studies have identified several loose
binding motifs of residues that are acceptable at key
positions in the peptide (17-23). The motifs of peptides
that bind to DR4 and DR1 molecules are distinct from
those reported to bind to other DR alleles (17-26).
The association of RA with certain DR4 subtypes and DR1 makes the characterization of the
peptides that bind selectively to these alleles of considerable interest. The DR4 subtypes that are associated with RA are Dw4 (DRB1*0401), Dw14 (a public
specificity encoded by DRB 1*0404 and DRB 1*0408),
and Dw15 (DRB1*0405), but not DwlO (DRBl"0402)
or Dw13 (DRB1*0403) (27). Although these DR4 subtypes differ by only 1-4 residues (3,they may bind
distinct complements of peptides. In this study, we
characterized the binding of a peptide that binds
selectively to DRB 1*0401-encoded molecules, and
1745
SELECTIVE PEPTIDE BINDING TO DRBl"0401 MOLECULES
Table 1. Polymorphic residues of selected DRBI* allele-encoded proteins
Amino acid position
B1 allele (DR type or DR4 subtype)
0401 (Dw4)
0101 (DRI)
1101 (DR11)
1501 (DR2)
0301 (DR3)
09
10
11
12
13
26
28
30
31
33
31
E
W
W
E
Q
Y
Y
V
L
S
P
S
K H
- F
T
S
- R
T
S
F
L
-
D
E
-
Y
C
-
F
I
-
H
N
N
Y
S
_
-
_
-
0401 (Dw4)
0402 (Dw 10)
0403 (Dw13)
0404 (Dw 14)
0408 (Dw 14)
0405 (Dw15)
identified critical peptide and DR residues that contribute to this interaction.
MATERIALS AND METHODS
Peptides. Native hemagglutinin 307-3 19 peptide
(HA307-3 19, PKYVKQNTLKLAT) (28), mutated HA307319 peptide with a glutamic acid substituted for tyrosine at
position 309 (HA309E, PKEVKQNTLKLAT) (21), native
Mycobacterium leprae 18-kd protein 37-49 peptide (ML3749, EEFVVEFDLPGIK) (29), native human immunodeficiency virus glycoprotein 41 peptide (gp41, QARILAVERYLKDQ) (30), and biotinylated derivatives (B-HA, B-ML,
and B-gp41) were synthesized in the Department of Medicinal Chemistry (G. D. Searle & Co.) using an automated
peptide synthesizer (Applied Biosystems, Foster City, CA)
with standard Boc-amino acid coupling protocols (31). Peptide ASMLIO (FAADAAAAAA) was synthesized in the
Department of Immunology and Glycobiology (G. D. Searle
& Co.) using standard Fmoc chemistry. Some of the peptide
resins were biotinylated at the N-terminus by coupling with
succinimidyl-6-(biotinarnido)hexanoate (Pierce, Rockford,
IL). All of the peptides were purified by reverse-phase high
performance liquid chromatography (HPLC) (Waters DeltaPak C-18). HPLC-purified ML37-49 peptides, containing
single alanine substitutions and A6MLlO (FAAEFDAAAA),
were purchased from Macromolecular Resources (Fort Collins, CO). The compositions of all peptides were verified by
mass spectroscopy.
Cell lines. The DR homozygous B lymphoblastoid
cell lines, GM03164 (DRB1*0401), GM03104 (DRB1*0101),
GM03161 (DRB1*1501), and GM03098A (DRB1*0301), were
obtained from the National Institutes of General Medical
Sciences Human Genetic Mutant Cell Repository (Camden,
NJ). The DR homozygous B lymphoblastoid Swei cell line
(DRB1*1101) was originally obtained from Dr. John Hansen
(Fred Hutchinson Cancer Research Center, Seattle, WA).
The B lymphoblastoid cell lines were maintained in RPMI
1640 media supplemented with 15% fetal calf serum (Hyclone, Logan, UT) and 2 mM glutamine (Mediatech, Wash-
_
-
61
10
71
13
L
I
F
Q
D
D
K
R
E
-
A A T G
- - - G
G
- v
G R N V
D
-
L
1
Q
D
-
-
-
K
E
R
R
R
R
51
-
- - - - N
-
-
s
-
-
-
-
-
-
14
A
E
-
-
77
86
G
V
V
V
-
ington, DC). The production of mouse L cell transfectants,
which express DRB1*0401, DRB1*0402, DRB1*0403, and
DRB1*0404, and site-directed mutants of DRB 1*0401, was
as previously described (12). Transfectants were maintained
in Dulbecco's modified Eagle's medium supplemented with
10% fetal calf serum, 2 mM glutamine, 1% penicillin/
streptomycin solution, and 0.25 mglml G418 (Gibco BRL,
Gaithersburg, MD).
Cellular peptide-binding assay. The binding of biotinylated peptides to cells was performed as previously described (12). Briefly, for studies using B lymphoblastoid cell
lines, 3 x lo5 cells were incubated with 10 /.M B-HA alone
or with a combination of 10 pA4 B-HA and varying concentrations (1-300 CLM) of competitor peptides. For studies
using L cell transfectants, the cells were incubated with 10
pA4 B-ML, B-HA, or B-gp41 in the presence or absence of
500 pA4 competitor peptides. All incubations were performed overnight at 37°C in V-bottom 96-well tissue culture
plates in Excell-300 serum-free media (JRH Biosciences,
Lenexa, KS). The cells were washed 3 times with phosphate
buffered saline containing 1% bovine serum albumin and
0.02% sodium azide, and incubated with StreptavidinQuantum Red conjugate (Sigma, St. Louis, MO) for 30
minutes. After 3 additional washes, the cells were analyzed
by flow cytometry (FACScan; Becton Dickinson, Mountain
View, CA). The data were collected and analyzed using the
Lysis I1 data analysis program (Becton Dickinson) to obtain
the mean channel fluorescence (MCF) on a linear scale for
each sample. For studies using the B lymphoblastoid cell
lines, a minimum signal-to-noise ratio of 5 was calculated
by dividing the MCF for cells incubated with B-HA and
Streptavidin-Quantum Red, compared with StreptavidinQuantum Red alone. The ability of native peptides to compete with biotinylated peptide binding to the B lymphoblastoid cell lines was calculated as the percentage of specific
inhibition using the formula
100 x
MCF (B-HA with competitor peptide) MCF (fluorescein isothiocyanate [FITC] avidin alone)
MCF (B-HA with no competitor peptide) MCF (FITC avidin alone)
WOULFE ET AL
1746
For studies using transfected cells, the binding of biotinylated peptides was normalized to class I1 expression, as
determined by FITC-L243 binding. Increases or decreases of
~ Z f o l din biotinylated peptide binding to the site-directed
mutant molecules over the wild-type DRB 1*0401 molecules
were considered meaningful.
Competition capture assay. The binding of biotinylated peptide to purified DRB1*0401 molecules was performed using an LB3.1 capture enzyme-linked immunosorbent assay (ELISA), as previously described (32). Results
are expressed as the concentration of competitor peptide
that produced a 50% inhibition
in binding of B-HA;
values are the average from 2 experiments. Increases or
decreases of 22-fold in biotinylated peptide binding to
DRB 1*0401 molecules were considered meaningful.
DRB1*0401 model. A homology model of DRB1*0401
was constructed using coordinates of the DRB1*0101 molecule (kindly provided by Dr. Jerry Brown, Harvard University, Cambridge, MA), as previously described (12).
RESULTS
Peptide 3749, derived from M leprae 18-kd
protein (ML3749), binds specifically to DRB1*0401expressing cells. DRBl*OlOl, DRB1*1501, DRB1*0301,
DRBf*0401, and DRBI*llOI share general sequence
homology, but differ in sequence in r l of the 3
hypervariable regions (Table 1). The ability of ML3749, HA307-319, and HA309E to compete for binding
with B-HA was examined using B lymphoblastoid cell
lines that express these molecules (Figure 1). The cell
lines were incubated with both B-HA and increasing
concentrations of competitor peptides. ML3749 competed for binding with biotinylated HA307-3 19 in a
dose-dependent manner only on the DRB1*0401 cell
line. The ML37-49 peptide inhibited binding of 10 pI4
B-HA by 50% at a concentration of 25 pI4. As
expected (21,33), native HA307-3 19 peptide competed
for binding with B-HA on each of the cell lines, while
the HA309E peptide did not compete dose dependently.
ML3749 binds specifically to cells that express
DR4 subtype DRB1*0401. DR4 subtypes share common first and second hypervariable domains, but differ
in the third hypervariable region (Table 1). To evaluate
the binding of ML37-49 to DR4 subtypes, L cell
transfectants expressing DRB 1*0401, DRB l"0402,
DRB 1*0403, and DRB I *0404 molecules were incubated with B-ML (Figure 2, first 3 bars), B-HA, or
B-gp41 (see ref. 12; data not shown). Detectable B-ML
binding to the DRB1*0401 cells (relative potency of 1)
was observed, but minimal, if any, binding to the other
DR4 subtypes was detected using this method. These
results demonstrate that ML37-19 binding was sensitive to amino acid substitutions in the third hypervariable region.
Negatively charged residues, located in polymorphic regions of DR4 subtype and DR mutant molecules,
correlate with lack of ML37-49 peptide binding. The
specificity of B-ML binding was further evaluated
using a panel of transfectants expressing DRB1*0401
site-directed mutations. As shown in Figure 2, B-ML
binding was nominally affected by the amino acid
substitutions present in the site-directed mutant molecules. Substitution of histidine at position 13 of the /3
chain with phenylalanine (p13 H-F) inhibited B-ML
binding by >2-fold, while substitution at this position
with tyrosine (p13 H-Y) increased binding by >2-fold.
B-ML binding was enhanced 10-fold by substitution of
pchain residue 28 from aspartic acid to histidine (p28
D-H). Replacement of aspartic acid at position 57 with
serine (p57 D-S) increased binding by 2.5-fold. Substitution of lysine at position 71 with glutamic acid (p71
K-E) decreased peptide binding >2-fold, while substitution at this position with arginine (p71 K-R) increased binding 5-fold. Substitution of residue 74 with
glutamic acid (p74 A-E) had little meaningful effect on
binding, while substitution with glutamine (p74 A-Q)
increased binding 5-fold. Replacement of glycine at
residue 86 with valine (p86 G-V) increased peptide
binding 4-fold. Amino acid substitutions at pchain
residues 11, 30, 37, and 70, or a-chain residues 24 and
31, had minimal effects on B-ML binding. These
results demonstrate that the residues at pchain positions 13, 28, 71, 74, and 86 were important in binding
the B-ML peptide.
Phenylalanine at position 3 of ML3749 peptide
is a key anchoring residue. The role of the amino acid
residues at each position within the ML3749 peptide
was evaluated using 13 mutant peptides containing a
single alanine substitution at each position. The ability
of the alanine-scanning mutants to compete for binding
with biotinylated ML37-49 peptide was determined by
ELISA (Table 2). Substitution of phenylalanine at
position 3 by alanine completed the abrogated binding
to DRB 1*0401. This result indicated that phenylalanine was a primary anchor residue. Alanine substitutions at positions 4, 7, and 9 decreased binding between 2- and 4-fold. Substitutions at positions 6, 12,
and 13 increased binding from 2-fold to 4-fold. Alanine
substitutions at residues 1, 2, 5, 8, and 11 had little or
no effect on peptide binding. Therefore, ML37-49
residues 4, 6, 7, 9, 12, and 13 may form secondary
contacts with DRB1*0401.
SELECTIVE PEPTIDE BINDING TO DRB l"0401 MOLECULES
1747
100
90:
A. DRB1'0101
90
80:
80
c
5
4
60
4
2
504
I+
40:
5
70
9
504
I-I
40:
2
a
P
30
30
20
20
*ol 0
i
10
100
competitor peptide (phi)
c.
80
1000
1(0 0
DRB1'0301
0:
1
10
100
competitor peptide (pM)
1
10
100
competitor peptide (phi)
1000
1
. . ....
*
*
1
.
9
..
'...I
*
. , .*.
10
100
cornpatitor peptide (phi)
1000
90:
80;
70:
c
4
4
60-
c
2
501
Figure 1. Analysis of ML37-49 binding to DR-expressing cells
using the cellular peptide-binding assay. Varying concentrations of
competitor peptides, HA307-319 (B), ML3749 (A), or HA309E
(O), were used to inhibit the binding of B-HA to cells expressing
DRB1*0101 (A), DRBI*ISOI (B), DRB1*0301 (C), DRB1*0401 (D),
or DRB1*1101 (E).Results of a representative experiment are
shown.
+I
40<
a
1
10
100
competitor peptide (pM)
1( 00
1748
WOULFE ET AL
Glutamic acid at position 6 and aspartic acid at
position 8 contribute to the selective binding of
ML37-49 to DRB1*0401. The effects of the alanine
substitutions on the specificity of binding were examined using L cell transfectants that express various DR
molecules. The molecules expressed by the transfectants used in this experiment share homology in 2 1 of
their hypervariable regions with DRB1*0401 (Table 1).
Relative Potency
0
0
0
I-.
w
c.
c.
0
DRB1* 0402
DRBl * 0403
Table 2.
Binding of alanine-substituted peptides to DRBl*O401
Peptide
Sequence
Go(nM)*
~
ML3749
1E
2E
3F
4v
5v
6E
IF
8D
9L
1OP
11G
121
13K
EEFVVEFDLPGIK
AEFVVEFDLPGIK
EAFVVEFDLPGIK
EEAVVEFDLPGIK
EEFAVEFDLPGIK
EEFVAEFDLPGIK
EEFVVAFDLPGIK
EEFVVEADLPGIK
EEFVVEFALPGI K
EEFVVEFDAPGIK
EEFVVEFDLAGIK
EEFVVEFDLPAIK
EEFVVEFDLPGAK
EEFVVEFDLPGIA
62.5
52.5
46.5
>100,000
140
115
15
I65
37.5
I90
260
39
26.5
16
* Values are the concentration of competitor peptide that produced
a 50% inhibition in binding (I(&).
DRBl * 0404
alpha 24 F-H
alpha 3 1 I-E
beta 11 V-S
beta 11 V-G
beta 13 H-F
beta 13 H-Y
beta 28 D-H
beta 30 Y-L
beta 31 Y-S
beta 3 1 Y-F
beta 57 D-S
beta 70 Q-0
beta 70 Q-R
beta 7 1 K-E
beta I 1 K-R
beta 14 A-E
beta 74 A-Q
beta 86 G-V
Figure 2. Binding of B-ML to DR4 subtypes, DRB1*0401,
DRB1*0402, DRB1*0403, DRB1*0404, and DRB1*0401 mutant molecules, determined by competition capture enzyme-linked immunosorbent assay. Results are expressed as the relative potency of
binding compared with the binding to the DRB1*0401 cell line.
B-ML binding to the DRB1*0401 transfectant resulted in an average
mean channel fluorescence (MCF) of 53 compared with the avidin
control MCF of 19; L243 binding resulted in an average MCF of 618
compared with the IgGCL2 control MCF of 19 (average of 7
experiments). Dashed lines indicate an increase or decrease in
binding of ~ 2 - f o l dData
.
shown are averages of 2 values pooled from
several experiments.
DRB1*0101 and DRBl"0401 are highly homologous in
the third hypervariable region, DRBl*O402 and
DRB1*0401 share identical first and second hypervariable regions, and DRB1*1101 and DRB1*0401 are
highly homologous in the second hypervariable region.
The 13 alanine-scanning mutants were tested for their
ability to compete with biotinylated HA307-3 19 for
binding to these transfectants (Figure 3). As expected,
all alanine mutants, except the position 3 mutant,
competed with biotinylated HA307-3 19 for binding to
DRB1*0401. In addition, the position 6 mutant peptide, but not the native ML37-49 peptide, inhibited the
binding of the biotinylated peptide to the DRBl*O402
cell line. The position 8 mutant, but not the native
ML3749 peptide, inhibited binding to the DRB1*0101
cell line. These results suggest that both glutamic acid
at position 6 and aspartic acid at position 8 of the
ML3749 peptide conferred specificity of peptide binding to DRB I *0401.
Negative charge at position 6 confers binding
selectivity. The previous experiments identified peptide residues 3, 6, and 8 as important in binding or
selectivity for the DRBl*0401 molecule. To determine
if residues 6 and 8 were sufficient to impart selectivity,
a decapeptide, A6ML 10 (FAAEFDAAAA), substituted with alanines at positions for which the side
chain interactions were predicted to be unimportant in
conferring binding selectivity, was evaluated for its
ability to bind to several DR types (Figures 4A-D).
The truncated alanine-rich peptide maintained the
ability to selectively compete for binding with B-HA
on DRBl*0401 cells, but not on DRB1*0101,
DRB1*0301, or DRB1*1101. In addition, A8ML10
(FAADAAAAAA) also competed for binding on
SELECTIVE PEPTIDE BINDING TO DRB 1*0401 MOLECULES
SEQUENCE
DRB1.
0401
I
I
DRBV
0402
I
I
I
DRB1.
0101
I
I
I
0
DRB1*
1101
I
I
I
ML37-49
EEFVVEFDLPGIK
1E-A
AEFVVEFDLPGIK
2E-A
EAFVVEFDLPGIK
3F-A
EEAVVEFDLPGIK
0
9
0
0
4V- A
EEFAVEFDLPGIK
60.5
12.5
0
0
5V-A
EEFVAEFDLPGIK
53
6
13
0
6E-A
EEFVVAFDLPGIK
70.5
49.5
7.5
0
7F-A
EEFVVEADLPGIK
50
5.5
0
Q
8D-A
EEFVVEFALPGIK
74.5
14.5
66
4.5
I
54.5
66
68.5
4.5
14.5
14
0
0
0
0
0
I
I
1
I
9L-A
10P-A
-
11G A
13K - A
121 A
EEFVVEFDLAGIK
52.5
I
3.5
I
0
[
0
EEFVVEFDLPAIK
59
12
30
0
EEFVVEFDLPGAK
65.5
10.5
21
0
EEFVVEFDLPGIA
HA307-319 PKYVKQNTLKLAT
HASOSE
1
PKEVKQNTLKLAT
I
62
I
11.5
I
7
I
I
0
71.5
56
83
41.5
0
19
0
24
I
Figure 3. Binding of alanine-substituted peptides to transfectants
expressing DRB 1 *0401, DRBl*0402, DRBl*OlOl, and DRBl*l101,
as determined by cellular peptide-binding assay. The values shown
are the average of 2 experiments and are expressed as the percentage inhibition of B-HA binding. See Results for significance of
shading.
DRB1*0401 molecules, but not on DRB1*1101 molecules (Figures 4E and F). These results confirm the
importance of glutamic acid/aspartic acid at position
6 of the ML37-49 peptide (position 4 of the decapeptides) in conferring specificity and selectivity to
DRB 1 *0401 .
DISCUSSION
The ability of peptides to bind specifically to a
given DR molecule is governed by the polymorphic
residues of the DR molecule. Most of these polymorphic DR residues are clustered together and define
pockets that interact with amino acid side chains of the
bound peptide (8). In this study, we analyzed the
binding of ML37-49 as a tool to probe the importance
1749
of individual peptide and DR residues in allele-specific
binding interactions.
Our initial characterization of the binding of
peptide ML37-49 to DR homozygous B cell lines
suggested that this peptide bound selectively to
DRB 1 *0401. Although the flow cytometry assay permits the analysis of peptide binding to multiple DR
proteins without the need for purification of each type
of molecule, it is a relatively insensitive method of
assay compared with the ELISA (33). Therefore, we
cannot rule out the possibility that ML37-49 or any of
the substituted peptides might bind to detergentpurified DR molecules at low pH (5.5) in the ELISA,
without purifying and analyzing each molecule. Nevertheless, no binding of ML37-49 to cells that do not
express DRB 1 *0401 was detected, suggesting that the
peptide either does not bind or binds with low affinity
under the assay condition tested.
As predicted from published binding motifs
(17-24), an aromatic residue near the N-terminus of
the peptide was shown to be essential for peptide
binding. Substitution of phenylalanine at residue 3
with alanine abrogated peptide binding. These results
suggest that phenylalanine at residue 3 of ML37-49
corresponds to position pl of the DR binding motif (8).
Phenylalanine at residue 3 of ML37-49 is predicted to
interact with pocket 1 (Figure 9, composed of nonpolymorphic a-chain residues 24, 3 1 , 32, 43, nonpolymorphic pchain residues 89 and 90, and polymorphic
@chain residues 85 and 86 (8). Interestingly, substitution of glycine at position 86 with valine did not
prevent ML37-49 peptide binding, although this substitution abrogated B-HA binding (12). Thus, valine at
position 86 appears to prevent the binding of HA307319, which contains a tyrosine anchor, but is of less
importance in the binding of ML37-49, which contains
a phenylalanine anchor.
Substitution of residue 6 with alanine increased
the binding of the peptide to DRB1*0401 and changed
the range of its binding specificity. The alanine replacement at position 6 permitted peptide binding to
DRB 1 *0402. DRB 1 *0401 and DRB 1 *0402 molecules
have identical first and second hypervariable regions,
but have nonconservative differences in the third
hypervariable region. The ability of DRBl*O401, but
not DRB1*0402, to bind ML37-49 indicates that residues in the third hypervariable region were responsible for the differential binding. DRB 1 *0402 molecules
contain 2 negatively charged amino acids located in
the “polymorphic” positions, p70 and p71.
1750
WOULFE ET AL
100
100
A. DRB1.0101
8
'
4
u
80
8
70
'
70
4
u
60
50
60
50
40
40
*
B. DRB1'0301
90
80
301
*
30
20
20
10
]
o
l0
1
10
100
1
1000
10
5
100
11 30
100
1000
C. DRB1*0401
90
80
8
'1
4
U
*
70
60
50
40
30
20
10
0
1
10
80.
70;
E.
DRB1.0401
c
702
:
$ 601
3
4
r
,
:
3;' 5 0 1
:
40'
d 30;
* 20'
10:
1
10
100
1000
competitor peptide (pM)
00
Figure 4. Analysis of A6MLIO and A8MLlO binding to DR-expressing cells using the cellular peptide-binding assay. Varying concentrations
of competitor peptides, HA307-319 (closed squares), A6MLIO (FAAEFDAAAA) (closed circles), HA309E (closed triangles), or A8MLIO
(FAADAAAAAA) (closed diamonds), were used t o inhibit the binding of B-HA t o cells expressing DRB1*0101 (A), DRB1*0301 (B),
DRB1*0401 (C and E), or DRBI*IIOI (D and F). Percentage of specific inhibition was calculated as described for Figure 2. Results of a
representative experiment are shown.
SELECTIVE PEPTIDE BINDING TO DRB 1*0401 MOLECULES
1751
Figure 5. Structural model of the DRB1*0401 molecule. The a chain is shown as the striped ribbon. The p chain is depicted as
a solid ribbon. The a carbons of polymorphic residues are illustrated as the large numbered balls. The a carbons of selected
nonpolymorphic (r- and Pchain residues are shown as small numbered balls. Five pockets that accommodate side chains of bound
peptide (see ref. 8) are numbered 1 , 4, 6, 7, and 9, and are enclosed by the dotted lines.
Residue 6 of the ML37-49 peptide is predicted
to extend into the p4 pocket, as described by Stern et
a1 (8). As shown in Figure 5 , the glutamic acid at
peptide residue 6 can potentially interact with pchain
polymorphic residues 13, 26, 28, 70, 71, 74, and 78,
which comprise the p4 pocket. One possible explanation for the failure of ML37-49 to bind to DRB 1*0402
is that the negatively charged glutamic acid at residue
6 of ML37-39 is incompatible with the negative
charges located at positions p70 and p71 within the p4
pocket of DRB1*0402. In addition, 2 alanine-rich
peptides, A6ML10 and A8MLI0, both containing a
phenylalanine anchor at position 1 and a negatively
charged amino acid at position 4, retained sufficient
information to allow selective binding to DRB 1*0401,
which contains 2 positively charged residues at positions p70 and p71. Our findings extend the observations of Jardetzky et a1 (24) that the majority of peptide
side chains are not required for high-affinity peptide
binding, by identifying residues involved in the specificity of binding. Interestingly, other approaches aimed at
identifying DRB 1*0401-specific peptide binding motifs
failed to predict the binding of negatively charged
residues at the p4 position of the peptide (17-23).
1752
Substitution of residue 8 with alanine also
broadened the binding specificity of the substituted
peptide relative to the parent peptide. Replacing aspartic acid at position 8 with alanine permitted binding
to both DRB1*0401 and DRB1*0101, but did not permit
binding to DRBl”O402 or DRB 1* 1101. DRB 1*0401 and
DRB1*0101 are highly homologous in the third hypervariable region, but differ in the first and second. The
ability of ML37-49 to bind to DRB1*0401, but not to
DRB1*0101, suggests that residues in the first or
second hypervariable regions were responsible for the
differential binding. Therefore, residues in the first or
second hypervariable regions almost certainly interact
with residue 8 of ML37-49.
Residue 8 of ML37-49 is predicted to extend
into the p6 pocket (Figure 5). Aspartic acid at residue
8 is modeled to interact with nonpolymorhic a-chain
residues 11, 62, 65, and 66, and polymorphic Pchain
residues 9 and 11. It is likely that tryptophan at
position pS, leucine at position pl 1, and/or cysteine at
position j330 of the DRB1*0101 molecule create a
pocket that is capable of accepting an alanine but not
an aspartic acid residue.
The DR residues involved in binding to
ML37-49 were evaluated using cell lines that express
DRB 1*0401 mutants. The majority of effects involved
residues that contribute to the p4 pocket. Nonconservative substitutions at positions p13, p28, p71, p74,
and p86 altered the binding of ML37-49. In general,
substitutions that introduced a positive charge into the
p4 binding pocket (such as the aspartic acid to histidine
substitution at p28) increased binding. In contrast, the
introduction of a negative charge (lysine to glutamic acid
at p71), or removing a positive charge (histidine to
phenylalanine at p13) in the pocket decreased ML37-49
binding to the DR molecules (Figure 2).
Results obtained using transfectants that express DRB 1*0401 site-directed mutant molecules were
not always predictive of the results obtained using
transfectants that express wild-type DR4 subtype
molecules. For example, ML37-49 bound poorly to
DRB 1*0403 transfectants. The protein encoded by the
DRB 1*0403 differs sequentially from DRB 1*0401 by
an R at p71, an E at p74, and a V at pS6. Surprisingly,
the transfectant that expressed wild-type or DRBl*0401
site-directed mutant molecules containing an E at p74
bound ML37-49 equally well. This result was unexpected, since the introduction of a negative charge into
the p4 pocket failed to prevent peptide binding. It is
possible that either DRB1*0401 p74 does not contact
the side chain of residue 6 of ML37-49, or that the
WOULFE ET AL
interactions between DR residues and a given peptide
ligand side chain may affect interactions elsewhere in
the molecules.
ML37-49 also bound poorly to the DRB1*0404
molecule, which differs from DRB 1*0401 only by an R
at p71 and a V at pS6. However, cells expressing
site-directed mutant molecules containing single
amino acid substitutions at these corresponding positions both bound ML37-49 peptide better than the
wild-type molecule. These results further suggest that
the effects of single substitutions are not necessarily
additive, and that the interactions between DR residues and a given peptide ligand side chain may affect
interactions elsewhere in the molecules.
The importance of a given position within a
peptide ligand, with its interaction with DRB 1*0401
molecule side chains, may vary from peptide to peptide (12), as indicated in the present study. However,
our experiments using the selectively binding
ML3749 peptide demonstrate that the peptide residues that were predicted to contribute to interactions
within the DRB1*0401 p4 pocket had the greatest
effects on binding specificity. Thus, the p4 pocket is
capable of influencing the peptide-binding repertoire of
DR molecules. DR1 and DR4 subtypes Dw4, Dw14,
and Dw15 are associated with the autoimmune disease
RA, while DR4 subtypes DwlO and Dw13 are not. It is
not known whether the association of certain DR4
subtypes with RA is due to the binding of a putative
arthritogenic peptide by these subtypes, or is the result
of another mechanism. However, our data are consistent with the hypothesis that differences in the peptidebinding repertoire of these DR molecules account for
their differential association with RA.
Our results are also consistent with the “shared
epitope hypothesis” in suggesting a role for residues
67-74 in mediating disease, but do not rule out contributions from other polymorphic residues, such as
position 86, in shaping the peptide repertoire. For
example, an arthritogenic peptide that contains a negatively charged residue at the p4 position, in the
context of residues at other positions in the peptide,
might bind to the RA-associated molecules, but may
be prevented from binding to the non-RA-associated
molecules. It may be possible to utilize the knowledge
of an aspartic acid-glutamic acid interaction at p4 to
design potential RA-associated allele-specific therapeutic agents, which act by blocking the DR peptidebinding groove.
SELECTIVE PEPTIDE BINDING TO DRB l”0401 MOLECULES
ACKNOWLEDGMENTS
The authors thank Dr. X.-T. Fu for use of the transfectants that express molecules with mutated achains, and Dr.
J. Welply, Dr. s. Howard, M. Zupec, and J. Bullock for
synthesis and purification of the peptides. We also thank Dr.
w.
and J . Gorka for helpful advice and discussions.
REFERENCES
1. Schwartz BD: The human major histocompatibility human
leukocyte antigen complex. In, Immunology. Seventh edition.
Edited by DP Stites, A1 Terr. East Norwalk, CT, Appleton &
Lange, 1991
2. Nepom GT: MHC class 11 molecules and autoimmunity. Annu
Rev lmmunol 9:493-525, 1991
3. Bodmer JG, March SGE, Albert ED, Bodmer WF, Dupont B,
Erlich HA, Mach B, Mayr WR, Parham P, Sasazuki T,
Schreuder GMT, Strominger JL, Svejgaard A, Teraski PI:
Nomenclature for factors of the HLA system. Hum Immunol
31:18&194, 1990
4. Marsh SGE, Bodmer JG: HLA class I1 nucleotide sequences.
Hum Immunol 35:l-17, 1992
5. Brown JH, Jardetzky T, Saper MA, Samraoui B, Bjorkman PJ,
Wiley DC: A hypothetical model of the foreign antigen binding
site of class I1 histocompatibility molecules. Nature 332:845850, 1988
6. Gorga JC: Structural analysis of class I1 major histocompatibility complex proteins. Crit Rev Immunol 11:305-335, 1992
7. Brown J H , Jardetzky TS, Gorga JC, Stern LJ, Urban RG,
Strominger JL, Wiley DC: Three-dimensional structure of the
human class 11 histocompatibility antigen HLA-DRI . Nature
364:33-39, 1993
8. Stern LJ, Brown JH, Jardetzky TS, Gorga JC, Urban RG,
Strominger J L , Wiley DC: Crystal structure of the human class
11 MHC protein HLA-DRl complexed with an influenza virus
peptide. Nature 368:215-221, 1994
9. Barber LD, Bal V, Lamb JR, O’Hehir RE, Yendle J , Hancock
RJT, Lechler RI: Contribution of T-cell receptor-contacting and
peptide-binding residues of the class I1 molecule HLA-DR4
Dw 10 to serologic and antigen-specific T-cell recognition. Hum
Immunol 32:110-118, 1991
10. Krieger JI, Karr RW, Grey HM, Yu W-Y, O’Sullivan D,
Batovsky L, Zheng Z-L, Colon SM, Gaeta FCA, Sidney J ,
Albertson M, Del Guercio M-F, Chestnut RW, Sette A: Single
amino acid changes in DR and antigen define residues critical for
peptide-MHC binding and T cell recognition. J Immunol 146:
2331-2340, 1991
11. Zelikzewski D, Golvano JJ, Gaudebout P, Dorval I, Freidel C,
Gebuhrer L, Betuel H, Borras-Cuesta F , Sterkers G: Implication of HLA-DR residues at positions 67, 71, and 86 in interaction between HLA-DRll and peptide HA306-320. J Immunol
151:6237-6247, 1993
12. Fu X-T, Bono CP, Woulfe SL, Swearingen C, Summers NL,
Sinigaglia F , Sette A, Schwartz BD, Karr RW: Pocket 4 of the
HLA-DR (a,pl*0401) molecule is a major determinant of T cell
recognition of peptide. J Exp Med 181:915-926, 1995
13 Newton-Nash DK, Eckels DD: Differential effect of polymorphism at HLA-DRl @chain positions 85 and 86 on binding and
recognition of DRl-restricted antigenic peptides. J Immunol
150: 1813-1821, 1993
14. Rothbard JB, Gefter ML: Interactions between immunogenic
peptides and MHC proteins. Annu Rev Immunol 9:527-565,
1991
15. Schiffenbauer J, Schwartz BD: The HLA complex and its
relationship to rheumatic diseases. Immunol Rheum Dis 13:463485, 1987
1753
16. Hansen JA, Nelson JL: Autoimmune diseases and HLA. Crit
Rev Immunol 10:307-328, 1990
17. Chicz RM, Urban RG, Lane WS, Gorga JC, Stern LJ, Vignali
DAA, Strominger JL: Predominant naturally processed peptides bound to HLA-DRI are derived from MHC-related molecules and heterogeneous in size. Nature 358:764768, 1992
18. Kropshofer H , Max H , Miiller CA, Hesse F , Stevanovic s, Jung
G, Kalbacher H: Self-peptide released from class I1 HLA-DRI
exhibits a hydrophobic two-residue contact motif. J Exp Med
17511799-1803, 1992
19. Hammer J , Takacs B, Sinigaglia F: Identification of a motif for
HLA-DRl binding peptides using M13 display libraries. J Exp
Med 176:1007-1013, 1992
20. Hammer J , Valsasnini P, Tolba K , Bolin D, Higelin J, Takacs B,
Sinigalia F: Promiscuous and allele-specific anchors in HLADR-binding peptides. Cell 74: 197-203, 1993
21. Sette A, Sidney J , Oseroff C, del Guercio M-F, Southwood S,
Arrhenius T, Powell MF, Colon SM, Gaeta FCA, Grey HM:
HLA DR4 Dw4-binding motifs illustrate the biochemical basis
of degeneracy and specificity in peptide-DR interactions. J
lmmunol 151:3163-3 170, 1993
22. Marshall KW, Liu AF, Canales J, Perahia B, Jorgensen B,
Gantozos RD, Aguilar B, Devaux B, Rothbard JB: Role of the
polymorphic residues in HLA-DR molecule in allele-specific
binding of peptide ligands. J Immunol 152:49574946, 1994
23. O’Sullivan D, Arrhenius T, Sidney J, Del Guercio M, Albertson
M, Wall M, Oseroff C, Southwood S, Colon SM, Gaeta FCA,
Sette A: On the interaction of promiscuous antigenic peptides
with different DR alleles: identification of common structural
motifs. J Immunol 147:2663-2669, 1991
24. Jardetzky TS, Gorka JC, Busch R, Rothbard J, Strominger JL,
Wiley DC: Peptide binding to HLA-DRl: a peptide with most
residues substituted to alanine retains MHC binding. EMBO J
9:1797-1803, 1990
25. Sidney J, Oseroff C, Southwood S , Wall M, Ishiokd G, Koning
F. Sette A: DRB1*0301 molecules recognize a structural motif
distinct from the one recognized by most DRPl alleles. J
Immunol 149:2634-2640, 1992
26. Chicz RM, Urban RJ, Gorga JC, Vignali DAA, Lane WS,
Strominger JL: Specificity and promiscuity among naturally
processed peptides bound to HLA-DR alleles. J Exp Med
178:2747, 1993
27. Gregersen PK: HLA associations with rheumatoid arthritis: a
piece of the puzzle. J Rheumatol Suppl 19:7-11, 1992
28. Nobusawa E , Aoyama T, Kato H, Suzuki Y, Tateno Y,
Nakajima K: Comparison of complete amino acid sequences
and receptor-binding properties among 13 serotypes of hemagglutinins of influenza A viruses. Virology 182:475485, 1991
29. Oftung F , Shinnick TM, Mustafa AS, Lundin KEA, Godal R,
Nerland AH: Heterogenity among human T cell clones recognizing an HLA-DR4, DwCrestricted epitope from the 18-kDa
antigen of Mycobacterium leprae defined by synthetic peptides.
J Immunol 144:147&1483, 1990
30. Hammond SA, Obah E, Stanhope P, Monell CR, Strand M,
Robbins FM, Bias WB, Karr RW, Koenig S, Siliciano RF:
Characterization of a conserved T cell epitope in HIV-1 gp41
recognized by vaccine-induced human cytolytic T cells. J lmmunol 146:1470-1477, 1991
31. Gross E , Meienhofer J: The Peptides. New York, Academic
Press, 1979
32. Kirschmann DA, De Ciechi PA, Bono CP, Zacheis ML,
Schwartz BD, Woulfe SL: Utilization of soluble fusion proteins
for induction of T cell proliferation. Cell lmmunol 160:193-198,
1995
33. Busch R, Rothbard JB: Detection of peptide-MHC class I1
complexes on the surface of intact cells. J lmmunol Methods
13411-22, 1990
Документ
Категория
Без категории
Просмотров
1
Размер файла
1 037 Кб
Теги
178038120401, drb1, pocket, residue, negatively, interactiv, confers, specificity, binding, charge
1/--страниц
Пожаловаться на содержимое документа