close

Вход

Забыли?

вход по аккаунту

?

Epitope analysis of the major reactive region of the 100-kd protein of PM-Scl autoantigen.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 39, No. 9, September 1996, pp 1588-1595
8 1996, American College of Rheumatology
1588
EPITOPE ANALYSIS OF THE MAJOR REACTIVE REGION OF
THE 100-kd PROTEIN OF PM-Scl AUTOANTIGEN
QUN GE, YAJUAN WU, JUDITH A. JAMES, and IRA N. TARGOFF
Objective. To localize the epitope(s) bound by
anti-PM-Scl antibodies in the N-terminal half of the
100-kd protein, the major antigen of the PM-Scl
complex.
Methods. Investigations were performed by immunoblotting 20 anti-PM-Scl positive sera against bacterially expressed, polymerase chain reaction-derived
deletion mutants of the S1 fragment (amino acids
11-437), enzyme-linked immunosorbent assay (ELISA)
screening against synthesized serial octapeptides, and
ELISA screening, with anti-PM-Scl positive sera,
against a synthesized 21-amino acid peptide covering
the active region.
Results. Anti-PM-Scl positive sera retained full
immunoblot activity with fragment 207-436 and most
activity with fragment 11-241, but had markedly decreased activity against fragments 236-436 and 11-212,
indicating a major epitope in the aa 207-241 region.
Fusion proteins with smaller fragments localized this
activity between aa 226 and aa 246. Of 42 anti-S1positive, anti-PM-Scl positive sera tested by ELISA
against a synthetic peptide of this region, 36 were
definitely positive, 4 borderline, and 2 negative. Similar
activity was seen with a peptide from which proline 228
was deleted. Three additional epitope areas were found
in S1, but each reacted with only a few sera. Anti-PMPresented in part at the 57th National Scientific Meeting of the
American College of Rheumatology, San Antonio, TX, November 1993.
Supported in part by NIH grants AR-32214, AR-42474, and
AI-21568, by Department of Veterans Affairs Medical Research
Funds, and by a grant from the Scleroderma Federation and United
Scleroderma Foundation.
Qun Ge, MD, Yajuan Wu, MD: Oklahoma Medical Research
Foundation, Oklahoma City; Judith A. James, MD, PhD: University of
Oklahoma Health Sciences Center and Oklahoma Medical Research
Foundation, Oklahoma City; Ira N. Targoff, MD: University of Oklahoma Health Sciences Center, Veterans Affairs Medical Center, and
Oklahoma Medical Research Foundation, Oklahoma City.
Address reprint requests to Ira N.Targoff, MD, Arthritis/
Immunology Section, Oklahoma Medical Research Foundation, 825
NE 13th Street, Oklahoma City, OK 73104.
Submitted for publication December 12, 1995; accepted in
revised form March 29, 1996.
Scl positive sera did not react with any octapeptide
spanning the major epitope area (aa 207-246).
Conclusion. The main immunoblot epitope of the
PM-Scl 100-kd protein is within a central area of 21 aa
(aa 226-246), but is longer than the usual linear epitope.
This peptide may be useful in patient testing. Three
minor epitopes in S1 may also be recognized by some sera.
Anti-PM-Scl autoantibody is strongly associated
with an overlap syndrome of polymyositis or dermatomyositis with scleroderma (1). Clinical signs of both
conditions are seen in 50-70% of patients with anti-PMScl (2-4), and anti-PM-Scl is found in 24% of patients
with myositis/scleroderma overlap (3). Most others with
anti-PM-Scl have isolated scleroderma or myositis. Patients with anti-PM-Scl have a high frequency of arthritis, interstitial lung disease, and Raynaud’s phenomenon, and a very high frequency of HLA-DR3 (3-5).
Other myositis- or scleroderma-specific autoantibodies
are generally absent in these patients, indicating that
they are an immunologically distinct subgroup (2,6,7).
Thus, anti-PM-Scl is important for clinical diagnosis and
patient classification.
PM-Scl is a nucleolar antigen whose function has
not been established (8,9). Immunoprecipitation (IPP)
with anti-PM-Scl positive serum shows at least 11 proteins that appear to constitute a complex, with no
associated nucleic acid (3,9,10). More than 90% of sera
identified as anti-PM-Scl positive by IPP consistently
react with the 100-kd protein, and >50% react with the
70-75-kd protein, with no cross-reaction between these
proteins (7,11,12). Previously, a complementary DNA
(cDNA) derived from human thymocytes encoding the
PM-Scl 100-kd protein was isolated and sequenced, with
a predicted protein of 860 amino acids and molecular
mass of 98.87 kd (7). A second cDNA encoding the
PM-Scl 100-kd protein derived from HeLa cells has also
been described, with a predicted protein of 885 aa, that
was identical to the thymocyte form except for a 75-bp
insert in the HeLa 100-kd protein after aa 695 (12).
1589
MAJOR EPITOPE OF PM-Scl ANTIGEN
Recombinant forms of both proteins reacted with antiPM-Scl positive sera.
The origin of this and other scleroderma- or
myositis-associated autoantibodies, and their role in the
pathogenesis of the disease, are unknown. However,
many, including anti-PM-Scl, have high specificity for
particular clinical syndromes, suggesting an important
relationship to fundamental disease processes (1,13).
Identification of key reactive regions of the molecule
may assist in elucidating initial events in antibody production, particularly in pursuit of certain hypotheses
such as molecular mimicry. Thus, localization of important epitopes of autoantigens has been of significant
interest.
Two regions of PM-Scl 100-kd protein that are
reactive with autoantibodies have previously been identified by studies of recombinant protein fragments produced in Escherichia coli (12,14), Of 3 Sma I restriction
fragments, S1 (including the N-terminal437 aa) reacted
by immunoblot (IB) with 41 of 42 sera that had reacted
with the whole recombinant protein, and the intensity of
reaction of most individual sera was similar to that of
whole protein (14). The central S2 fragment (aa 439749) also reacted with 39 sera by IB, but reaction with S1
was often estimated to be stronger (S1 reaction greater
than S2 reaction in 28 sera). Some anti-PM-Scl positive
sera reacted with the C-terminal S3 portion, but none
reacted most strongly with this region. Using proteins
expressed from H i m 11, Rsa I, and Pvu I1 restriction
fragments, that study localized the major reactivity of
the N-terminal S1 fragment to between aa 156 and aa
312 (14). This area included a region of high hydrophilicity and predicted surface probability that was a good
candidate for the epitope site.
Since the N-terminal S1 region carries the major
reactivity of the 100-kd protein for most anti-PM-Scl
positive sera, this region was analyzed further in the
present study, to better characterize the reactive
epitope(s).
MATERIALS AND METHODS
Sera. Sera were determined to be anti-PM-Scl positive
by double immunodiffusion against concentrated calf thymus
extract (2) and by IPP from HeLa cell extract (14,lS). The 20
anti-PM-Scl positive sera tested against deletion fragments
were selected based on availability and known reactivity with
the N-terminal S1 (aa 11-437) fragment by immunoblot and
the H2 (aa 11-312) fragment by enzyme-linked immunosorbent assay (ELISA), as determined in previous studies (14J6).
An additional 24 anti-PM-Scl positive sera were used for
ELISA testing, based on availability.All but 1was known to be
Fragments
s1 (AA 11.437)
A1 (AA 11-281)
A2 (AA 1 1 -241)
A3 (AA 11-212)
A4 (AA 11-164)
A5 (AA 11-144)
A6 (AA 151-436)
~7 (AA 207-436)
A8 (AA 236436)
A9 (AA 283-436)
/---
1
Epitopes
A10 (AA 207-257)
All (AA
A12 (AA
A13 (AA
A1 4 (AA
A15 (AA
2
-
3
4
207-246)
207-241)
226257)
236257)
226246)
Figure 1. Diagram of deletion mutants used in this study. Each
protein was produced by expression of polymerase chain reaction
(PCR)-derived complementary DNA (cDNA) fragments in Escherichiu coli. S1 fragment was used as PCR template. S1 is the largest
Srnu I restriction fragment of the human PM-Scl 100-kd cDNA
(full-length HeLa protein = 885 amino acids). The “Fragments”
column indicates the fragment name, followed, in parentheses, by the
amino acids (AA) included in the fragment. To the right is the
schematic representation of the region included, and the relationship
of each fragment to the full S l fragment. Proteins S1 and Al-A9
included the amino acids indicated plus a 6-histidine tag; fragments
A10-Al5 included the amino acids indicated as fusion proteins with
mouse dihydrofolate reductase. The “Epitopes” row schematically
shows the proposed localization of the 4 epitopes identified in this
study (see text for rationale for placement).
reactive with H2 by IB. Control sera from normal subjects and
patients with other known autoantibodies were obtained from
the serum bank of the Oklahoma Medical Research
Foundation.
Preparation of deletion mutants. In the previous study
(14), a cDNA encoding the whole 100-kd protein (labeled
PCR-A) was prepared by polymerase chain reaction (PCR)
with 5’ and 3‘ gene-specific primers, using reverse-transcribed
HeLa messenger RNA as template. Fragment S1 was the
largest fragment obtained by digestion of PCR-A with Sma I.
In this region, the thymocyte and HeLa sequences were
identical.
For the present study, five 5’ and four 3‘ S1-fragment
deletion mutants, defined as in Figure 1, were prepared by
PCR amplification (performed as previously described [7]),
using gene-specific primers and the S1 fragment as template.
Each of the 9 deletion fragments was isolated independently by
agarose gel electrophoresis, subcloned into pQE plasmid, and
transformed into E coZi strain MlS/pREP4 using the Qiaexpressionist system (Qiagen, Chatsworth, CA) as previously
described (14). The constructs used produced independent
proteins except for a 6-histidine tag. The insert-containing
1590
plasmid was identified by screening of mini-plasmid preparations. Colonies harboring plasmid with an insert of expected
size were selected for small-scale expression, induced with
isopropylthio-P-galactoside (IPTG). In addition, 6 internal
fragments, defined in Figure 1, were also generated by PCR
with gene-specific primers, and subcloned into the pQE vector
using constructs that generated fusion proteins with mouse
dihydrofolate reductase, which contributed 19.5 kd to these
proteins.
Immunoblotting. Proteins produced by E coli from
expression of PCR fragments were tested for reaction with
anti-PM-Scl positive sera from 20 different patients, and with
sera from 4 normal subjects and 6 myositis patients with other
autoantibodies. E coli lysates were prepared in sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer (50 mM Tris HCI, pH 6.8, with 5% P-mercaptoethanol, 2% SDS, 0.01% bromphenol blue, and 10% glycerol),
subjected to SDS-PAGE (12% for deletions, 15% for fusion
proteins), and transferred to nitrocellulose as previously described (7). Blocking was done with 5% nonfat milk in Tris
buffered saline. After incubation with 1:200 dilutions of test
anti-PM-Scl positive sera and control sera, the blots were
developed with goat anti-human IgG/alkaline phosphatase
conjugate (Sigma, St. Louis, MO) and BCIP/NBT substrate
(Sigma).
Synthesis of peptides and peptide ELISA. Serial octapeptides including all possible octapeptides within the
area from aa 207 to aa 246 were synthesized on solid-phase
plastic pins as previously described (17,18). Each octapeptide
was tested by ELISA at 1:lOO dilution against 5 standard
anti-PM-Scl positive sera and 2 control sera, and developed
with a goat anti-human IgG (y-chain specific)-alkaline
phosphatase conjugate and phosphatase substrate system, as
described (17,18). Control wells with reactive La (19) and
Sm (17) peptides were included and tested with standard
sera.
After reactivity was localized within a 21-aa region
by expression of fragments in E coli, this region was synthesized as a multiple antigenic peptide (MAP; Applied Biosystems, Foster City, CA) and labeled MAP4c. It was tested
in a standard ELISA, with plates coated with peptide at 5
pglml, 100 $/well (0.5 &well), blocked with 0.2% bovine
serum albumin in phosphate buffered saline, incubated with
patient or control serum at 1:200 dilution, and developed as for
the pin ELISA. Each serum was tested in wells with and
without antigen, and the background without antigen was
subtracted. Values were considered elevated if the activity in
the serum was more than 2 standard deviations above the
mean in the normal control sera tested simultaneously and
additional controls tested subsequently. A 20-aa peptide
(MAP4b) was similarly synthesized, but without the proline at
the third peptide position (aa 228 in the whole protein), and
was similarly tested.
The hydrophilicity and surface probability estimates
were performed previously (14), according to the KyteDoolittle and Emini analyses, with the Peptidestructure program of the Sequence Analysis Software Package (Genetics
Computer Group, Madison, WI). Correlations were calculated
using Microsoft Excel, and probability by t-test.
GE ET AL
Anti-P M- Sc I
+
Ctrl
A1
5
1
90
15
20
25
30
A2
1
5
10
20
15
30
25
A3
1
5
lo
15
20
25
30
1
5
lo
15
P
25
30
15
20
A4
A5
1
5
lo
25
30
Figure 2. Immunoblot with recombinant 3' deletion mutants. Lysates
of Escherichia coli, each expressing one of the 3' deletion mutants
(AI-A5, as in Figure l), were electrophoresed and immunoblotted. In
lanes 1-20, each lane of each blot was developed with anti-PM-Scl
positive serum from 1 of the 20 patients. Lanes 21-24, Normal control
(Ctrl) sera. Lanes 25-30, Anti-PM-Scl negative myositis patient sera.
RESULTS
Each of the PCR-derived cDNA fragments was
successfully expressed in E coli in substantial amounts,
as demonstrated by the appearance of a new protein of
appropriate size in IPTG-induced E coli lysates. Fragment A1 (aa 11-281) reacted with all 20 anti-PM-Scl
positive sera, including 15 that showed strong (3-4+)
reactivity, and with no control sera (Figure 2 and Table
1). Fragment A2 (aa 11-241) reacted more weakly with
9 of the 20 sera, but 10 sera still had a 3-4+ reaction,
and 17were positive. A much greater drop in activity was
seen with fragment A3 (aa 11-212); although 9 sera still
had activity, only 1 reacted a t 3+, and none a t 4+
(Figure 2). Reactivity with A3 was weaker than reactivity
with A1 in 19 of the 20 sera, and weaker than reactivity
with A2 in 15 sera. This pointed to the aa 212-241 region
1591
MAJOR EPITOPE OF PM-Scl ANTIGEN
Table 1. Reactivity of anti-PM-Scl positive sera with bacterially expressed proteins*
Protein
A4
A5
A6
A7
3+
2+
3+
1+
4+
-
3+
-
4+
4+
4+
4+
3+
3+
4+
4+
4+
3+
4+
3+
Serum
A1
A2
1
4+
3+
4+
4+
3+
2+
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
4+
3+
3+
3+
4+
2+
4+
2+
4+
3+
3+
2+
2+
1+
4+
A3
-
2+
4+
1+
-
*
-
-
2
1+
4+
-
3+
-
2
4+
-
-
-
-
4+
4+
2+
1+
3+
-
3+
2+
-
17
10
7
3
-
2+
4+
4+
4+
-
3+
-
4+
3+
-
~
-
4+
4+
4+
4+
4+
4+
3+
4+
4+
4+
4+
4+
4+
4+
2+
4+
-
3+
-
4+
20
18
0
0
20
20
20
19
20
20
20
15
-
4+
4+
4+
4+
3+
4+
4+
4+
4+
4-t
4+
1
1
Total
3-4+
4+
4+
4+
4+
4+
4+
-
4+
4+
4+
3+
4+
4+
-
-
-
3+
3i
4+
2+
-
4+
4+
4+
41-
4+
3i
4+
3+
4+
3+
-
2+
All
3+
4+
4+
-
2+
A10
4+
-
4+
A9
4+
-
20
A8
4 i
~
~
4+
A12
-
A13
A14
A15
4+
4+
4+
0
0
20
19
20
19
4+
4+
4+
4+
4+
4+
4+
4+
4+
4+
4+
4+
2+
4+
3+
4+
4+
4+
4+
4+
4+
4+
4+
4+
4+
4+
4+
4+
4+
2+
4+
3+
4+
4+
4+
4+
4+
* Immunoblot staining was scored as 0-4+. Total = total showing any positive reaction (C = faint stain of uncertain significance, considered
negative); 3-4+ = total showing strong (3+ or 4+) reactivity.
as containing the major epitope (epitope 3 in Figure 1).
It was not yet clear whether the decrease in activity
between A1 and A2 reflected a second epitope between
241 and 281, or whether the deletion affected the major
epitope. However, the remaining activity against A3
protein in 9 sera strongly suggested the existence of 1 or
more additional epitopes in the A3 region.
Five of the 9 sera that reacted with A3 reacted
even more strongly with fragment A4 (aa 11-164), and 2
others were positive. However, 4 weak A3 reactors
showed negative reactivity with A4. This further suggested a minor second epitope (epitope 2), located
within the A4 region. The marked decrease in activity
from fragment A4 to fragment A5 (aa 11-144), which
reacted significantly with only 1 serum (at 3+), pointed
to the aa 145-164 area for the site of epitope 2. The
reaction of serum 4 with A5 indicated an uncommon
additional epitope in this area. Since a previous study
had shown no reactivity of serum 4 or 42 other anti-PMScl sera with fragments of either aa 11-89 or aa 90-155
(14), the most likely place for this epitope (epitope 1 in
Figure 1) was at the juncture of these fragments (across
aa 89-90).
Fragments A6 (aa 151-436) and A7 (aa 207-436)
were fully active (Figure 3 and Table l),but, consistent
with the reactivity of the 3’ deletion mutants, fragment
A8 (236-436) lost most activity, reacting with only 2 of
20 anti-PM-Scl positive sera. Since no serum reacted
with fragment A9 (aa 283-436), the 2 sera were apparently reacting with an epitope(s) between aa 236 and aa
283 (epitope 4 in Figure 1).
These findings suggested that the major epitope
was between aa 207 and aa 241, which included the
region of high surface probability (aa 219-227) predicted as a possible site for that epitope. However,
synthesized peptides covering this region (aa 207-227
and aa 216-236) did not react with any anti-PM-Scl
positive sera by ELISA, and ELISA reaction of antiPM-Scl with H2 fragment was not inhibited by the
peptides tested at concentrations of up to 200 pg.
Therefore, the region was studied further using the
internal fragments defined in Figure 1, expressed as
fusion proteins.
Protein A l l (aa 207-246) reacted with all 20
anti-PM-Scl positive sera (Figure 4), but A12, without
aa 242-246, had no activity (Table 1).This implied that
aa 242-246 were crucial, despite the activity seen with
fragment A2, a difference that may relate to the presence of the 11-212 region in A2. Reactivity with fusion
proteins A13 and A14, along with A8, showed that the
region between aa 226 and aa 236 is required for the
major epitope (Table 1). Region 226-246 therefore
GE ET AL
1592
appeared to encompass the major epitope; supporting
this, fusion protein A15 (aa 226-246) reacted strongly
with all 20 sera (Figure 4). Since the 2 sera that reacted
with A8 also reacted with A14 but not A12, epitope 4
could be further localized to the 241-257 region.
In order to localize the epitope further, 5 antiPM-Scl positive sera and 2 normal control sera were
tested against all serial octapeptides synthesized on
solid-phase pins between aa 207 and aa 246, by ELISA
(results not shown). There was no consistent, significant
area of shared reactivity by anti-PM-Scl positive sera
compared with normal subjects, indicating that the major epitope required more than the minimal 6-8 amino
acids of a strict sequential epitope.
In contrast, a synthesized peptide from aa 226 to
aa 246 (MAP~c),tested by standard ELISA, reacted
with all 20 selected anti-PM-Scl positive sera, confirming the need for the larger peptide (Figure 5). This also
excluded the possibility that absence of a posttranslational modification resulted in the lack of reaction with
the serial octapeptides. Thirty-seven of the 44 anti-PM-
Ctrl
Anti-PM-ScI
4
As
5
1
10
15
20
25
30
20
25
30
A7
1
5
10
15
1
5
10
15
A8
20
25
30
20
25
30
As
1
5
10
15
Figure 3. Immunoblot with recombinant 5’ deletion mutants. Lysates
of Eschen’chia coli, each expressing one of the 5‘ deletion mutants
(A6-A9, as in Figure l), were tested as in Figure 2, against the same
set of anti-PM-Scl positive sera (lanes 1-20) and controls (Ctrl) (lanes
21-30).
An t i - P M - S c l
4-
I
5
I 0
20
25
30
I
5
LO
I0
I5
30
I-=
A1 o
1s
A1 1
15
A1 s
1
5
10
15
10
IS
30
Figure 4. Immunoblot with recombinant fusion proteins carrying internal fragments. Lysates of Escherichia coli, each expressing one of
the internal fragments defined in Figure 1, were tested as in Figure 2,
against the same set of anti-PM-Scl positive sera (lanes 1-20) and
controls (Ctrl) (lanes 21-30). Only selected internal fragments are
shown; see Table 1 for results with others.
Scl positive sera tested (84%), including all used for
fragment testing, were positive, with values >0.291
optical density (OD) units (4 SD above the mean in
normal control sera) and ranging from 0.396 to 1.451
OD units at the screening dilution used. Values in the 1
serum that had been negative by IB against S1 (at 0.259
OD units) (14) and in 3 other anti-PM-Scl positive sera
were less than 4 SD, but more than 2 SD, above the
mean in controls (i.e., values > 0.160), and were considered “borderline,” and values in 3 (7%) were negative.
Overall, the correlation between ELISA activity against
MAP4c and that against protein H2 measured in a
previous study (16), for 43 anti-PM-Scl positive sera
tested in both systems, was r = 0.787 (P < 0.005).
Peptide MAP4b (226-246 minus proline 228)
was equally active, with 41 of the 43 tested sera (95%)
having values higher than 0.165 (4 SD above the control
mean for this peptide). The value in the IB-negative
serum was 0.268 OD units, and the other values ranged
from 0.386 to 1.414 units (Figure 5). The correlation
between MAP4c and MAP4b activity was r = 0.879 (P <
0.005). All 4 anti-PM-Scl positive sera that had borderline reactivity and the 1that was negative against MAP4c
were positive (>0.165 OD units) against MAP4b. However, 2 of 44 anti-PM-Scl positive sera (5%) were clearly
nonreactive with this epitope in this form (below the
control mean + 2 SD value of 0.091 units), despite
MAJOR EPITOPE OF PM-Scl ANTIGEN
A: ELISA Against
PM-Scl PePtide MAP4c
A
. . ._
..
. . ._
. ._
..
. . .-
........ ..
A
A
NORMAL
N=38
OTHERS
N=41
02
0
PM-Scl
N=44
B. ELISA Against
PM-Scl Peptide MAP4b
. . . . . . . . . . . . .
1.4
v)
1.2
p.
0
3-p
. . . . . . . . . . . . . . . . . . . . . . . . . .
0.8
.8
0.6
c
0.21
.'
. . . . . . . . . . . . . . . . . . . . . . . .
....
..........................
..' . - 4 . .. . . . . . . . . . . . . . . . . . . . . . . . .
fi
.I
1
.........................
1 *
PM-Scl
N=43
NORMAL
N=38
OTHERS
N=4 1
Figure 5. Results of the synthetic peptide enzyme-linked immunosorbent assay (ELISA). Activity of anti-PM-Scl positive sera and control
sera from normal subjects or patients with other autoantibodies
(OTHERS), at a screening dilution of 1:200, was measured against A,
the 21-amino acid MAP4c (aa 226-246) and B, the 20-amino acid
MAP4b (aa 226-246 with proline 228 deleted) (the shortest active
peptide identified in this study).
significant IB and ELISA activity against the whole
recombinant protein and the SUH2 fragments in previous studies (14,16).
Seventy-nine control sera were tested against
each peptide, including 38 normal sera and 41 with other
autoantibodies (Figure 5 ) (reactive with Ro, La, U1
RNP, Sm, Jo, PL-7, PL-12, Mi, KJ, and others). Against
M A P ~ c 3, controls (2 normal, 1 anti-Ro positive) were
borderline reactive and 1 (normal) was in the positive
range (0.326). Against MAP4b, 5 controls (1 anti-Jo-1
positive, 1 anti-PL-7, positive, 1systemic lupus erythematosus patient, and 2 normal) were borderline reactive
1593
and 1 (anti-RNP/Sm positive) had elevated reactivity.
Other anti-RNP/Sm positive sera were negative against
both peptides. Each control serum that had borderline
or elevated reactivity against 1 peptide was negative
against the other.
In the 14 sera tested, preincubation with 0.1 pg of
MAP4c inhibited anti-H2 ELISA activity by a mean of
77% (range 43-99%) (results not shown). There was
little additional inhibition with increased peptide (mean
inhibition up to 86% with 100 pg). Those that were
negative against fragment A3 tended to inhibit more
than those that were positive (mean 73% for 5 A3reactive sera, versus 93% for 3 A3-nonreactive sera, at
0.1 pg). Less inhibition was seen in the ELISA of the
whole 100-kd protein; the mean inhibition was 39%
(range 0-80%) with 0.1 pg, and 57% (range 17-91%)
with 100 pg.
DISCUSSION
This study demonstrated that the major immunoblot epitope of the N-terminal half of the PM-Scl 100-kd
protein is contained within the aa 226-246 region, with
contributions from both the aa 226-236 and aa 242-246
portions. Ninety-five percent of anti-PM-Scl positive
sera tested reacted with this epitope by ELISA, and the
strength of this reaction correlated with that of reaction
with the whole N-terminal portion. Other epitopes were
present on the N-terminal portion of the molecule, but
each reacted with only a small percentage of anti-PMScl positive patient sera.
The synthesized 20- and 21-aa peptides were
active, but none of the synthesized serial peptides of 8
amino acids within this area was reactive. This places the
minimum-sized fragment that can carry the epitope
between 9 aa and 20 aa. If removal of aa 228 from
peptide MAP4b actually implies that aa 226-227 are
also unnecessary, it would narrow the maximum required region to 18 amino acids. Conceivably, additional
amino acids could be deleted. Nevertheless, a need for a
degree of conformation or discontinuity is suggested.
This conformation would have to be easily achieved,
since the epitope readily forms spontaneously in the
synthesized peptide and after denaturation during
blotting.
There was an apparent contradiction between the
high activity of the A2 protein, low activity of A3, and
absent activity of A12. Comparison of A2 and A3
demonstrates that epitopes 1 and 2 cannot explain the
activity of A2. The inactivity of fusion protein A12
demonstrates that no new epitope is present between
212 (end of A3) and 226 (start of major epitope region).
1594
One possible explanation is that the N-terminal 207 aa
promote formation of the proper conformation in the
major epitope region when present, but are not required, at least for some sera, if the 242-246 region is
present. This would imply that aa 242-246 do not
directly participate in the antibody binding site itself, at
least for some sera. The partial reduction, but not
elimination, of the activity of A 2 supports this interpretation. Since some sera do lose most activity against A 2
(for example, sera 3 and 8 in Figure 2), and there is
variation in the amount of loss, there may be slight
variation in the precise binding site, or the conformational requirement, between sera.
The area of high hydrophilicity and surface probability (aa 219-227, peaking at 220) in the N-terminal
portion was suggested as the possible site of the major
epitope, since it would more likely be exposed on the
native molecule (12,14). Studies using the deletion fragments were prepared with this possibility in mind, and
the results seemed to support this hypothesis, pointing to
the aa 207-241 region. However, most of the region
subsequently determined to include the active site was
outside the high surface probability area, and actually
included an area of negative hydrophilicity in its center
(aa 234-240). Also, fragment A l l (207-241) demonstrates that unlike the N-terminal region in A2, the
hydrophilic area alone is not able to restore activity to
the 226-241 region. Considering the proximity of the
active region to the hydrophilic area, it is possible that
the true epitope region is also on the surface, or
otherwise exposed, in the native conformation.
The 226-246 epitope is a predominant epitope in
the sense that it is shared by most anti-PM-Scl positive
patients, and it is the strongest epitope for most antiPM-Scl positive sera. Previous studies had shown that
the S1 region is predominant in the molecule as a whole,
for most sera (14). This study demonstrated that it
predominates in the S1 region, both in frequency and in
estimated IB strength. This was supported by the results
of inhibition studies with the peptide, which showed that
reaction with this epitope accounted for a large portion
of the reaction with the H2/S1 region, even for sera that
showed reaction with other epitopes.
The presence of a predominant epitope may have
implications regarding the origin of the antibodies. It is
consistent with the molecular mimicry hypothesis, in
which a response to a foreign protein, such as that of an
infectious agent, leads to shared reactivity with a crossreactive epitope. However, no viral or bacterial protein
with sequence similarity to the aa 226-246 area was
identified in the Genbank or EMBL database. The best
GE ET AL
match was that of aa 235-240 of the major epitope with
aa 2470-2475 of human dystrophin (with similarity of
231-233 with 2467-2469), of interest since it is a skeletal
muscle protein. One alternative to mimicry at this site
would be if the region were particularly immunogenic,
resulting in a shared epitope regardless of the mechanism for immunization. The proximity to the hydrophilic
region may be a factor contributing to this.
The significance of the finding that 2 sera did not
react with this epitope is unclear. If the epitope were a
site of mimicry, it would suggest that the exceptional
sera either lost activity that was originally present,
required more conformation than provided by the peptide, or developed the anti-PM-Scl response through
another site or different mechanism. Future studies will
be needed to determine whether patients with anti-PMScl that does not react with the peptide are clinically
different from other patients with the antibody.
Other PM-Scl epitopes must also be considered
in such hypotheses. Studies have demonstrated significant reactivity of most anti-PM-Scl positive sera with an
epitope in the S2 region (aa 439-749), and more than
half have significant reactivity with the 70-75-kd PM-Scl
protein. In addition, we identified epitope 2 in this study,
at least part of which is in the aa 144-164 range, that
reacted with 7 of 20 anti-PM-Scl positive sera. If aa
226-246 is the initial site of immunization, by whatever
mechanisms, reaction with these other epitopes may
represent spreading of the response to other areas in
individual patients. It is possible that there are additional epitopes that are purely conformational, and not
represented on the bacterially synthesized proteins. The
existence of such epitopes was suggested in previous
studies (14), and they may be more important than
226 -246.
The ELISA using the synthesized MAP4c or
MAP4b peptides showed promise as a possible clinical
test for detection and/or measurement of anti-PM-Scl.
Synthesized peptides have been useful for the study of
other autoantibodies, such as anti-ribosomal P (20).
They are convenient to produce and avoid the risk of
false-positive results due to reaction with proteins from
E coli or other producing cells. Most anti-PM-Scl positive sera tested were strongly positive in this ELISA.
Even the serum that was negative by IB was positive in
this test, an indication of its sensitivity for sera that
recognize this epitope. Unfortunately, 2 anti-PM-Scl
positive sera failed to bind to this peptide. It is possible
that combining this peptide with another region, possibly
epitope 2 identified in the present study, or the epitope
MAJOR EPITOPE OF PM-Scl ANTIGEN
of the S2 fragment (14), may allow identification of such
exceptional anti-PM-Scl positive sera.
In a previous study, ELISA with whole PM-Scl
100-kd protein was used for detection of anti-PM-Scl,
and similarly, some sera failed to react (16). However,
the high concentrations of the epitope possible with the
MAP ELISA may increase sensitivity for sera with
limited reactivity. Further study will be needed to assess
its clinical utility.
In conclusion, this analysis demonstrated an interesting predominant epitope, more complex than a
simple linear epitope, and located adjacent to a strongly
hydrophilic area. Study of this epitope may provide clues
to the origin of anti-PM-Scl antibodies. A synthesized
form of this epitope may be useful and convenient as an
antigen for autoantibody testing.
ACKNOWLEDGMENTS
The authors wish to thank Drs. Paul Plotz, Frederick
Miller, Chester Oddis, and Morris Reichlin for referral of
patient sera, and Jody Gross for assistance with solid-phase
peptide synthesis. Longer peptides were synthesized a t the
Molecular Biology Resource Facility of the William K. Warren
Research Institute, University of Oklahoma Health Sciences
Center.
REFERENCES
1. Targoff IN: Immune manifestations of inflammatory muscle disease. Rheum Dis Clin North Am 20:857-880, 1994
2. ReicNin M, Maddison PJ, Targoff IN, Bunch T, Arnett FC, Sharp
G, Treadwell E, Tan EM: Antibodies to a nuclear/nucleolar
antigen in patients with polymyositis-overlap syndrome. J Clin
Immunol4:40-44, 1984
3. Oddis CV, Okano Y, Rudert WA, Trucco M, Duquesnoy RJ,
Medsger TA Jr: Serum autoantibody to the nucleolar antigen
PM-Scl: clinical and immunogenetic associations. Arthritis Rheum
35:1211-1217, 1992
4. Marguerie C, Bunn CC, Copier J, Bernstein RM, Gilroy JM, Black
CM, So AK, Walport MJ: The clinical and immunogenetic features of patients with autoantibodies to the nucleolar antigen
PM-Scl. Medicine (Baltimore) 71:327-336, 1992
5. Genth E, Mierau R, Genetzky P. von Miihlen CA, Kaufmann S ,
1595
von Wilmowsky H, Meurer M, Krieg T, Pollmann H-J, Hart1 PW:
Immunogenetic associations of scleroderma-related antinuclear
antibodies. Arthritis Rheum 33:657-665, 1990
6. Reichlin M, Arnett FC Jr: Multiplicity of antibodies in myositis
sera. Arthritis Rheum 27:1150-1156, 1984
7. Ge Q, Frank MB, O'Brien CA, Targoff I N Cloning of a complementary DNA coding for the 100-kD antigenic protein of the
PM-Scl autoantigen. J Clin Invest 90:559-570, 1992
8. Targoff IN, Reichlin M: Nucleolar localization of the PM-Scl
antigen. Arthritis Rheum 28:226-230, 1985
9. Reimer G, Scheer U, Peters J-M, Tan EM: Immunolocalization
and partial characterization of a nucleolar autoantigen (PM-Scl)
associated with polymyositis/scleroderma overlap syndromes. J Immunol 137:3802-3808, 1986
10. Gelpi C, Alguero A, Angeles Martinez M, Vidal S, Juarez C,
Rodriguez-Sanchez JL: Identification of protein components reactive with anti-PM/Scl autoantibodies. Clin Exp Immunol81:5964,1990
11. Alderuccio F, Chan EKL, Tan EM: Molecular characterization of
an autoantigen of PM-Scl in the polymyositis/sclerodermaoverlap
syndrome: a unique and complete human cDNA encoding an
apparent 75-kD acidic protein of the nucleolar complex. J Exp
Med 173:941-952, 1991
12. Bluthner M, Bautz FA: Cloning and characterization of the cDNA
coding for a polymyositis-scleroderma overlap syndrome-related
nucleolar 100-kD protein. J Exp Med 176:973-980, 1992
13. Rothfield N F Autoantibodies in scleroderma. Rheum Dis Clin
North Am 18:483-498, 1992
14. Ge Q, Wu Y, Trieu EP, Targoff IN: Analysis of the specificity of
anti-PM-Scl autoantibodies. Arthritis Rheum 37:1445-1452, 1994
15. Targoff IN: Autoantibodies to aminoacyl-transfer RNA synthetases for isoleucine and glycine: two additional synthetases are
antigenic in myositis. J Immunol 144:1737-1743, 1990
16. Ge Q, Wu Y, Targoff kN: Detection and quantification of human
anti-PM-Scl autoantibodies using recombinant PM-Scl protein
(abstract). Arthritis Rheum 36 (suppl 9):S150, 1993
17. James JA, Harley JB: Linear epitope mapping of an Sm BB'
polypeptide. J Immunol 148:2074-2079, 1992
18. Scofield RH, Dickey WD, Jackson KW, James JA, Harley JB: A
common autoepitdpe near the carboxyl terminus of the 60-kD Ro
ribonucleoprotein: sequence similarity with a viral protein. J Clin
Immunol 11:378-388, 1991
19. James JA, Gross T, Scofield RH, Harley JB: Immunoglobulin
epitope spreading and' autoimmune disease after peptide immunization: Sm B/B'-derived PPPGMRPP and PPPGIRGP induce
spliceosome autoimmunity. J Exp Med 181:453-461, 1995
20. Elkon KB, Skelly S, Parnassa AP,Moller W, Danho W, Weissbach
H, Brot N: Identification and chemical synthesis of a ribosomal
protein antigenic determinant in systemic lupus erythematosus.
Proc Natl Acad Sci U S A 835'419-7423, 1986
Документ
Категория
Без категории
Просмотров
0
Размер файла
787 Кб
Теги
major, epitopes, autoantigen, protein, 100, analysis, regions, reactive, scl
1/--страниц
Пожаловаться на содержимое документа