close

Вход

Забыли?

вход по аккаунту

?

Epitopes of immunoreactive myelin basic protein in human cerebrospinal fluid.

код для вставкиСкачать
Epitopes of Immunoreactive Myelin Basic
Protein in Human Cerebrospinal Fluid
John N. Whitaker, MD, Manjula Gupta, PhD, and Olive F. Smith, BS
~
T o define in more detail the features of the immunoreactive myelin basic protein (MBP) present in cerebrospinal fluid
(CSF) of humans following acute injury to central nervous system myelin, the epitopes of MBP recognized by three
different antisera, each capable of detecting immunoreactive MBP in CSF, were examined. All three antisera reacted
well with human MBP and human MBP peptide 45-89. Only in radioimmunoassays in which the MBP peptide 45-89
served as the radioligand could clearly elevated values of immunoreactive MBP be measured in CSF specimens from 5
patients with multiple sclerosis during or immediately after an exacerbation. The two antisera that reacted well with
MBP peptide 80-89 resulted in higher levels of immunoreactive MBP measured in CSF. An epitope present in human
MBP peptide 80-89 but sharing a conformation with both MBP and MBP peptide 45-89 is present in CSF following
acute central nervous system myelin damage in multiple sclerosis.
Whitaker JN, Gupta M, Smith OF: Epitopes of immunoreactive myelin basic protein in human
cerebrospinal fluid. Ann Neurol 20:329-336, 1986
Myelin basic protein (MBP) is a protein of 18,500
molecular weight and approximately 170 residues that
constitutes 30% of central nervous system myelin proteins 16, 191. Antigenic material that is cross-reactive
with MBP appears in cerebrospinal fluid (CSF) of persons for a period of approximately 7 to 14 days following acute injury to central nervous system myelin 18,
27,331. The presence of immunoreactive MBP in CSF
may serve as an index of demyelination in multiple
sclerosis (MS) (reviewed in 1351). In MS the immunoreactive MBP in CSF exists in a range of sizes
smaller than intact MBP 11, 181, with the epitopes
from the region encompassing residues 45-89 occurring as the dominant antigenic form 127, 301.
Many antisera to MBP fail to detect immunoreactive
MBP in CSF. While some of this limitation may be
related to the low affinity of certain antisera to MBP
and resultant insensitivity of immunoassays in which
these antisera are used, the antigenic determinants or
epitopes of MBP recognized also appear to be crucial
for an antiserum to detect MBP-like material in CSF.
In previous studies of antisera, it was shown that antisera reactive with epitopes in the carboxyl half of MBP
peptide 45-89' were needed for the immunoreactive
MBP in CSF to be detected 1301. These studies used a
small number of MBP peptides, two of the most important, i.e., bovine MBP peptide 68-88 and guinea
pig MBP peptide 79-88, being derived from nonhuman sources. In the present investigation, properties of
the antisera detecting MBP-Like material in CSF have
been further examined using a large number of human
MBP peptides of the correct sequence.
From the Neurology and Research Services, the Birmingham and
Memphis Veterans Medical Center; the Departments of Neurology
and Cell Biology, the University of Alabama at Birmingham, Birmingham, AL; and the Department of Immunopathology of The
Cleveland Clinic Foundation, Cleveland, OH.
*The numbering system for amino acid residues conforms with the
sequences published for human [6],bovine 14, 131, and guinea pig
1231MBP. This differs from the convention previously used in publications from this laboratory in which the numbering of residues was
always related to the amino acid sequence of bovine MBP. Hence,
human MBP peptides 45-89 and 80-89 are comparable to bovine
MBP peptides 43-88
79-88, respectively.
Received Oct 16, 1985, and in revised form Dec 11, 1985. Accepted for publication Dec 21, 1985.
Materials and Methods
Preparation of M B P and M B P Peptides
Human MBP was prepared from delipidated post-mortem
human brain by means of acid extraction and carboxymethylcellulose chromatography [31}. Components 1 and 3 1111
were obtained, desalted on a column of Sephadex G-25
coarse, equilibrated with 0.1% acetic acid, and lyophilited.
Component 1 of MBP was further purified by reverse-phase
high-performance liquid chromatography (RP-HPLC) using
a solvent system of trifluoroacetic acid and acetonitrile [17}.
Human MBP peptides 1-44, 45-89, and 90-170 were prepared from human MBP by means of digestion with bovine
brain cathepsin D [301. The fractions containing human
MBP peptides 1-44, 45-89, and 90-170 were individually
Address reprint requests to Dr Whitaker, Department of Neurology, University of Alabama at Birmingham, University Station, Birmingham, AL 35294.
329
desalted over a column of Sephadex G-50 equilibrated with
0.1 M NH4HC03 1341. The fraction containing MBP peptide 45-89 was further purified by isocratic elution on RPHPLC to generate MBP peptides 45-88 and 45-87 1151.
Guinea pig and bovine MBP peptides 43-88 were similarly
prepared but were not further processed by RP-HPLC.
Guinea pig MBP peptide 77-88 was prepared by trypsin
digestion 132’).
Small MBP peptides were synthesized by Peninsula Laboratories and shown by the producer to have the expected
amino acid residues; they gave a single peak by RP-HPLC.
Some of these peptides have been previously described [28].
On the basis of the corrected sequence of human MBP peptide 45-89 1141, peptides 67-81, 76-85, and 80-87 and related peptides containing all or part of the sequence of 80-87
were prepared using the corrected residues of histidineglycine at position 77-78 and glutamic acid-asparagine at
residues 83-84. The human MBP peptides synthesized by
Peninsula Laboratories were as follows: 5 3-67,64-? 5,67-8 1,
76-85, 76-87, 77-87, 78-89, 77-87, 80-87, 80-85, 81-87,
82-87,82-87,82-70,82-91,82-92,82-73,84-89, and 84-96.
In addition to these MBP peptides, the human MBP peptide
80-87, previously used 1281 and shown to be incorrect because it contained glutamine-aspartic acid at residues 83-84
[141, was also used for certain studies. The peptide of bovine
MBP containing residues 77-88 was used after synthesis by
Peninsula Laboratories.
Human MBP peptides 44-67 and 68-89 were prepared by
solid-phase synthesis subjected to hydrofluoric acid to remove blocked residues 1241 and separated from reactants by
gel filtration on Sephadex G-25 equilibrated with 0.1 M
acetic acid at 25°C. The eluted peptides were lyophilized and
further purified by RP-HPLC using a trifluoroacetic acidacetonitrile system [171. Fractions were collected and analyzed by amino acid analysis and total sequencing from the
amino terminal [34].The results (data not shown) confirmed
the accuracy and purity of the human MBP peptides as 43-69
and 68-89.
Polyactyhmide Gel Electrophoresis
Fig I , Polyacrylamide disc gel electrophoresis at pH 2.5 in the
presence of urea of the (1) human myelin basic protein (MBP)
preparation used for preparation of imrnunogenfor rabbit 79
and sheep 45, ( 2 ) human MBP preparation usedfor immunization of GI,( 3 )human MBP usedas radioligand and inhibitor
standard in doudle-antibody radioimmunoasJay, and ( 4 )human
MBP peptide 45-89. Twenty-Jivemicrograms per gel. Cathode
at bottom. (Amido Schwan.)
Polyacrylamide disc gel electrophoresis was performed at p H
2.5 in the presence of urea {12}.
Preparation of Antisera
The three antisera used were prepared with different immunogens and by different immunization protocols. They
were selected from a large group of antisera available. The
protocols used will be described, but similar protocols have
been followed for other animals and have not led to the
production of antisera similarly useful in the detection of
MBP-like material in CSF.
Rabbit 77 (R79) and sheep 45 (S45) antisera were immunized with a conjugate of human MBP linked by carbodiimide to albumin [28). The MBP used for the immunogen was isolated from the postmortem brain tissue of a
person with MS [27].This MBP used contained components
Z to 4 {I I} and a variety of peptide breakdown products that
were not separated from the remainder of the apparently
intact MBP (Fig 1) but were included in the conjugation
reaction. R77 received injections of human MBP linked to
rabbit serum albumin and S45 received injections of human
330 Annals of Neurology
Vol 20
No 3
MBP linked to ovalbumin. The immunization of these animals and their reactivities have been previously described
[27, 301. The third antiserum, referred to as G I , was obtained from a New Zealand white rabbit immunized subcutaneously in multiple sites on the back with 250 pg of
human MBP (Fig l), prepared by the batch method 112) in
incomplete Freund’s adjuvant. Three weeks later this rabbit
was boosted with 200 k g of human MBP in complete
Freund’s adjuvant and subsequent boostings were carried out
at monthly intervals with 250 pg of human MBP in incomplete Freund’s adjuvant. The antiserum used was from a
bleeding taken after four boosting immunizations with incomplete Freund’s adjuvant.
For the collection of antisera from each of the animals,
blood was removed, allowed to clot at room temperature
overnight, separated from the clot and blood by centrifugation at 3,000 rpm for 10 minutes, and frozen at -20°C or
below prior to use. The sera were diluted and used without ,
heating or other treatment.
September 1786
Preparation of Radioligands
Human MBP component 1 I l l ] and human MBP peptide
45-89 C151 were radiolabeled with ''I by lactoperoxidasecatalyzed iodination. The procedures for iodination and separation of unbound 1251 have been previously described 1271.
Immunoassay
A double-antibody radioimmunoassay (DA-RIA) technique
was used 127, 301. The total volume of the reaction mixture
was 950 p1. The titer of antisera was determined in a 3-day
radioimmunoprecipitation in the DA-RIA and the inhibition
by CSF, MBP, or MBP peptides was detected by means of a
nonequilibrium 4-day DA-RIA [301. All peptides were examined in triplicate and all peptides giving detectable inhibition were analyzed in at least two separate assays. Each MBP
peptide was stored in buffer [27), thawed, diluted to a range
of concentrations, and used only once. For determination of
the inhibition for relative comparison, data were assessed by
the logit-log method 1221 and stated in molar terms with the
molecular weights based on 100 per amino acid residue in
the peptides.
CSF Specimens
CSF was removed by lumbar puncture and stored in plastic
vials (Pierce, Rockford, IL) at - 40°C or colder until time of
assay. Permission for using CSF in this study was obtained
according to Institutional Review Board clearance. The clinical diagnosis and clinical conditions were ascertained by one
of the authors (J. N. W.). The dilutions of CSF were made in
buffers containing 0.2 M Tris acetate, p H 7.2, containing
0.2% methylated bovine serum albumin.
Results
Polyacvlamide Gel Electrophoresis
B y polyacrylamide gel electrophoresis, the human
MBP preparations used as immunogens were shown to
contain a mixture of materials in addition to intact
MBP (Fig 1). Lower-molecular-weight fragments with
little remaining MBP were noted in the MBP from the
patient with MS (Fig 1, gel l ) , and both higher- and
lower-molecular-weight bands were noted in the
batch-prepared MBP (Fig 1, gel 2). The MBP used as
radioligand contained minor bands migrating beyond
MBP and a band, presumably the dimer of MBP, migrating more slowly (Fig l, gel 3). The human MBP
peptide 45-89 revealed one band (Fig 1, gel 4 ) with or
without the RP-HPLC step [ l 5 ] .
Titers of Antisera
Each of the antisera from R79, G1, and S45 was
reacted in a 3-day DA-RIA with radioiodinated human
MBP or human MBP peptide 45-89. The dilutions at
which precipitation of the antigens occurred differed
among the three antisera, but for each antiserum was
always much higher when the radioligand was human
MBP than when it was human MBP peptide 45-89
(Fig 2).
Immunoassays with Radioligand of MBP
Dilutions of the antisera giving 30 to 50% precipitation of human MBP were then used in a DA-RIA
05
RECIPROCAL DILUTION
Fig 2. Precipitation of 12rI-labeledmyelin basic protein (MBP}
(solid circles) and 1251-labeledMBP peptide 45-89 (open circles) by varying dilutions of antiserum from rctbbit 79.
Table 1 . Ferntomoles of MBP and MBP Peptides Neededjir
Inhibition of Reaction with Radiolabeled Human MBP
Antiserum"
~~
Rabbit
79
GI
Inhibitors
(31%)
(33%)
Sheep
45
(32%)
MBP
MBP peptides
1-44
45-79
80-89
90-170
143
115
23
>106
>lo5
>lo6
>lo6
>106
> 106
>106
>105
>lo6
>lo6
6,453
10,900
"Values indicate amount, in ferntomoles, of inhibitor required to
produce 50% inhibition between the antiserum (rabbit 79,
1 : 12,000; G1, 1 : 10,000; sheep 45, 1:40,000) and radioligand of
human MBP.
bThe amount of radioligand precipitated by antibody in the dilution
used in the assay.
MBP = myelin basic protein.
under nonequilibrium conditions to determine the
ability of selected MBP peptides to displace binding of
radioiodinated human MBP (Table 1). The results,
plotted as to the mole amounts of MBP or MBP peptide producing displacement, indicated that the three
antisera reacted well with human MBP but had little
reactivity with human MBP peptides 1-44, 45-89,
90-170, or 80-89 (Table 1). In the DA-RIAs with
two (G1 and S45) of the three antisera, very low
amounts of immunoreactive MBP could be detected
with radiolabeled MBP as the radiofigand, but none
could be detected with R79 (Table 2).
Immunoassays with Radioligand of Human
MBP Pepti& 45-85,
When the same 3 antisera were used in a DA-RIA
with the radioligand being radioiodinated human MBP
Whitaker et al: Epitopes of MBP in Human CSF
331
Table 3 , lnhibition of Reaction with Radiolabeled Human
MBP Peptide 45-89 by MBP and MBP Peptides
Table 2. Concentration of lmmunoreactiue MBP in CSF
Measured as Cross-Reactive with Human MBP or Human
M B P Peptide 45-89
Antiseruma
CSF Specimenb
3
Antiseruma
Radioligand
1
2
Rabbit 79
MBP
0
MBPpeptide
34.0
0
0
12.0 1.5
G1
Sheep45
45-89
MBP
2.4
M B P p e p t i d e 26.0
45-89
MBP
0.3
MBP peptide 6.2
45-89
4
Rabbit
Inhibitors
5
0
0
ND 52.0
1.2 2.2
2.0 2.5
ND
34.0
0.6 0.1 0.5
1.4 1.4 1.6
0.6
10.0
2.2
5.5
Hu MBP
MBP peptidesb
HU45-89
HU45-88
GP 43-88
BO 43-88
"Antisera diluted to give 30 to 50% precipitation of radioligand. For
radiolabeled MBP: rabbit 79, 1: 12,000; G1, 1: 10,000; sheep 45,
1:40,000. For radiolabeled MBP peptide 45-89; rabbit 79, 1:800;
G1, 1:2,000; sheep 45, 1:8,000.
bValues indicate concentration in nanograms per milliliter of CSF of
immunoreactive MBP measured against a standard of the antigen
homologous to the radioligand.
MBP = myelin basic protein; CSF
not done.
=
cerebrospind fluid; ND
=
79
Sheep
45
G1
1.0
1.o
> 104
4.5
4.0
0.7
0.9
1 .o
1.0
344
3.7
3.7
2.8
> lo4
> 104
HU80-89,'
1.1
0.8
Hu 80-89:
GP 79-88
BO 79-88
> 104
> lo4
>lo4
> 104
4.9
11.8
1.2
2.2
> 104
> 104
HU68-89
HU76-89
HU 77-89
HU78-89
HU79-89
HU81-89
0.6
5.1
12.0
6.7
22.5
0.8
1.7
4.0
5.0
8.0
402
1,082
> 10"
> 10*
> 104
211
> 104
> lo4
peptide 45-89, displacement of radioligand was affected nearly equally by human MBP peptide 45-89
and intact MBP (Table 3). R79 and G1 showed strong
reactivity with human MBP peptide 80-89, whereas
S45 failed to react with this peptide (Table 3). Immunoreactive MBP could be detected in CSF in DARIAs with all 3 antisera when human MBP peptide
45-89 was the radioligand (Table 2). In the DA-RIA
with a radioligand of human peptide 45-89, G1 and
R79 measured much higher values of CSF MBP-like
material than did S45 (Table 2). The differences were
less at lower values and greater at higher levels. R79
was less sensitive for detecting low levels than was
G1 but reflected higher levels in the same CSF (Table
2).
"Numbers indicate the molar ratio of amount of MBP or MBP
peptide relative to human MBP peptide 45-89 required to produce
50% inhibition between the antiserum (rabbit 79, 1:800; G1,
1:2,000; sheep 45, 1:8,000) and radioligand of human MBP peptide 45-89. Amount of unlabeled human MBP peptide 45-89
needed for 50% inhibition varied between 70 to 90 fmol for rabbit
79, 90 to 100 fmol for G1, and 45 to 65 fmol for sheep 45.
peptides tested up to lo5 fmol and that failed to inhibit the
reaction between antisera and radiolabeled human MBP peptide
45-89 were human MBP peptides 44-69, 53-67, 64-75, 69-81,
76-85, 80-85, 82-87, 82-89, 82-90, 82-91, 82-92, 82-93, 83-89,
84-89, and 84-96.
'80-89, is the peptide with the correct sequence of histidine-glycine
at positions 77-78.
d80-89, is the peptide with the incorrect sequence of glycinehistidine at positions 77-78.
Epitopes in H u m a n M B P Peptide 45-89 Recognized
Sy the Antisera
that the bovine and guinea pig MBP peptides had not
been further purified by RP-HPLC [ l 5 } . None of the
3 antisera reacted with MBP peptides 44-69, 53-69,
64-75, 69-81, or 76-85.
The reactions for R79 and G1 were also similar in
showing that both antisera reacted with MBP peptides
76-89, 77-89, 78-89, and 79-89, although less well
than with MBP peptide 80-89. Comparison of molar
displacement values for MBP peptides 76-89, 77-89,
78-89, and 79-89 revealed that for both R79 and G1
the reaction was best with peptide 80-89, which
showed some reactivity (Table 3), and no reaction occurred with all of the other MBP peptides tested
(Table 3). In particular, MBP peptides from the region
of MBP 80-89 but which contained less than the full
decapeptide such as MBP peptides, 80-85, 82-87,
82-89, and 84-89, failed to produce displacement
Because of the differences in the 3 antisera and because of the greater usefulness of a radioligand of human MBP peptide 45-89 in detecting MBP-like material in CSF by DA-RIA, more detailed studies were
performed with a variety of MBP peptides in DA-RIA
with human MBP peptide 45-89 as the radioligand. As
shown in Table 3, the reaccivities of R79 and G1 were
similar in reacting well with human MBP, human MBP
peptide 45-89, and human MBP peptide 68-89, poorly
with human MBP peptide 45-88, and showing good
reactivity with human MBP peptide 80-89 as well as
MBP peptides 43-88 and 79-88 of bovine and guinea
pig origin. The difference in the displacement by
bovine and guinea pig MBP peptide 43-88 compared
to human MBP peptide 45-89 is influenced by the fact
332 Annals of Neurology
MBP = myelin basic protein; H u = human; Bo = bovine, G P =
guinea pig.
Vol 20 No 3 S e p t e m b e r 1986
- 2 +
+1-
0
:
t
0
0-
0,
-1-
-20
2
4
6
.
8
0
2
Ln AMOUNT (femtomoles)
Ln VOLUME (microliters)
Fig 3. Displacement of radiolabeled human myelin basic protein
(MBPI peptide 45-89 from rabbit 79 in double-antibody
radioimmunoassay by vuying amounts of human cerebrospinal
/hid (CSF)from Patients 1 (open circles, dashed line) and 3
(solid circles, dashed line) (Table 2) and human MBP (solid
circles, solid line), human MBP peptide 45-89 (open circles,
solid line), and human MBP peptide 80-89 (triangles). Each
CSF point is an average of duplicate results and each antigen
displacement is the average of triplicate results.
of radioligand. MBP peptides containing a portion of
MBP peptide 82-89 or various MBP peptides and additional residues at the carboxyl terminal of MBP
peptide 80-89, such as MBP peptides 82-90, 82-91,
82-92, 82-93, and 84-96, also failed to produce displacement (Table 3).
In contrast to these results with R79 and G1, which
were similar to one another, the reaction with S45 was
different in that it reacted better with human MBP
peptide 45-88 and marginally with MBP peptide 68-87
(Table 3). S45 failed to react not only with human
MBP peptide 80-89 or other synthetic MBP peptides
but also with bovine and guinea pig MBP peptides
43-88.
Reactivity of CSF MBP-Like Material
in Dilution Assays
Because of the observed differences in the detection of
immunoreactive MBP in CSF, the relative parallel or
nonparallel nature of the displacement measured in
the assays with the 3 different antisera and human
radiolabeled MBP peptide 45-89 was examined. The
principle underlying this analysis is that the similar or
identical nature of 2 or more antigens will be reflected
by the degree of similarity of their reaction with the
same antiserum. Thus, the closer their parallelism of
displacement of radioligand from antibody, the more
nearly identical the antigens are presumed to be.
When one of these antigens, e.g., the MBP peptide, is
known with certainty, then the identity of the antigen
being characterized, e.g., the immunoreactive MBP in
CSF, can be inferred. All 3 antisera had approximately
the same detection limit of 50 to 100 pglassay tube of
.
.
.
6
4
8
Ln AMOUNT (ferntomoles)
Ln VOLUME (microliters)
Fig 4. Displacement of radiolabeled human myelin basic protein
(MBP)peptide 45-89 from GI in double-antibody radioimmunoassay by varying amounts of human cerebrospinaljuid
(CSF)from Patients 1 (open circles, dashed line) and 3 (solid
circles, dashed line) (Table 2) and human MBP (solid circles,
solid line), human MBP peptide 45-89 (open circles, solid
line), and human MBP peptide 80-89 (triangles). Each CSF
point is an average of duplicate results and each antigen displacement is the average of triplicate results.
+1-
8
0-
m
60
-J
-1-
\
-20
,
,
v
2
4
.
6
.
.
a
Ln AMOUNT (ferntomoles)
Ln VOLUME (microliters)
Fig 5 . Displacement of radiolabeled human myelin basic protein
(MBP)peptide 45-89 from sheep 45 in double-antibody
radioimmunoassay by vaying amounts of human cerebrospinal
jluid (CSF)from Patients I (open circles, dashed line) and 3
(solid circles, dashed line) (Table21 and human MBP (solid
circles, solid line) and human MBP peptide 45-89 (open circles, solid line). Each CSF paint is an average of duplicate
results and each antigen dispkzcemnt is the average of triplicate
results.
antigenic material cross-reactive with MBP peptide
45-89, but the displacement of radioligand from R79
and G 1 was greater at high concentrations than the
displacement from S45. This is presumably the explanation for the higher levels detected with R79 and G1
as compared with S45 (Table 2). The immunoreactive
material in the two CSF specimens diluted in near parallel with each other in assays with any of the three
antisera (Figs 3-5). They showed the greatest paralWhitaker et al: Epitopes of MBP in Human CSF 333
lelism with MBP or MBP peptide 45-89 but little or
none with MBP peptide 80-89. The parallelism was
least with S45, which did not react with MBP peptide
80-89. These results were interpreted to indicate that
the dominant epitope was in the region of MBP containing residues 80-87 but in a specified conformation
existing in MBP and MBP peptide 45-89.
Discussion
Delineation of the specific features of MBP-like material appearing in CSF after central nervous system myelin injury may be of importance for several reasons.
First, since the MBP measured in CSF has not been
validated by a nonimmunological test such as a bioassay, determination of the size and epitopes of the immunoreactive MBP would contribute to its further
characterization. Second, knowledge of the features of
the MBP-like material present should assist in standardizing its assay or improving the sensitivity for its
detection. Third, because the MBP-like material in
CSF presumably is the substrate that is further degraded by proteinases to generate smaller peptides of
MBP in blood o r urine, information about it would
indicate what small peptides might possibly be produced. Fourth, related to the ability of selected MBP
peptides to induce experimental allergic encephalomyelitis and provoke immune responses 1161, the nature
of the MBP present may suggest the type of MBP
peptide to use in testing patients with MS or other
disorders for cellular or humoral immunity to MBP.
Fifth, characteristics of the MBP present may imply
the sites of MBP cleaved and inferentially the type of
proteinase responsible for the degradation.
Attempts to elucidate the properties of MBP in CSF
have consisted of determining size [l, 7 , 18, 261,
epitopes [ 3 , 9 , 2 7 , 301, or the parallelism of dilution of
CSF MBP-like material with MBP 15, 21). Although a
large form of MBP has been reported to be present in
the CSF of patients with MS [7, 261, two investigations
of the gel filtration of MBP-like material in CSF reactive with MBP peptide 45-89 11) or MBP Cl8) have
demonstrated a spectrum of sizes of MBP-like material
in CSF, with larger forms predominating in CSF from
persons with cerebral infarction and smaller forms existing in CSF from persons with acute-phase MS. Studies of the epitopes of MBP present have shown that
the dominant epitope present was MBP peptide 45-89
[27],especially an epitope toward the carboxyl terminal of the peptide 130). Although antigenic material
cross-reactive with bovine MBP peptide 89-169 has
been reported to be present in CSF C37, the report has
not been confirmed C27, 301. The dilutional analysis of
CSF MBP-like material has demonstrated a moderate
nonparallelism with MBP [ 5 , 21).
In the present study, the features of the immunoreactive MBP in CSF were further defined
334 Annals of Neurology
Vol 20
through a detailed analysis of the epitopes of MBP
recognized by three antisera that measure immunoreactive MBP in CSF but at different concentrations.
All three antisera detected little or no MBP-like material in assays with intact MBP as radioligand. The
same three antisera, at lower dilutions, used in assays
with human MBP peptide 45-89 as radioligand measured MBP-like material in CSF. None of the three
antisera reacted with any of the MBP peptides from
the middle and amino terminal half of MBP peptide
45-89. Thus, no reaction occurred with MBP peptides
44-69, 53-69, 64-75, 69-81, or 76-85.
The two antisera that exhibited the highest amounts
of MBP-like material in CSF showed similar features
in recognizing an epitope for which the decapeptide
80-89 was required. The reduced recognition of
bovine MBP peptide 79-88 and guinea pig MBP peptide 79-88 and the marked diminution in reactivity of
human MBP peptide 81-89 imply a major influence
for threonine at residue 80, which is replaced by
proline in bovine MBP and serine in guinea pig
MBP c6, 23). Neither R79 nor G1 reacted with MBP
peptide 80-87 containing the erroneous sequence of
glutamine-aspartic acid at residues 83-84 110, 141.
The reactivities of R79 and G1 were not entirely identical in that R79 showed greater preference for an
epitope more restricted to MBP peptide 80-89 and
more affected by the addition of one to four residues
to the amino side of residue 80. G1 also was less able
to recognize the epitope in MBP peptide 81-89 and
was less dependent on residue 89 than was R79. This
suggests that G1 recognized an epitope of different
conformation or, if recognizing a sequential epitope,
positioned less toward the carboxyl end of MBP peptide 80-89. The slightly higher values of MBP-like
material in CSF detected by R79 compared with G1
suggests that the epitope recognized by R79 is the
dominant epitope present.
The third antiserum tested, S45, also detected
MBP-like material in CSF but at considerably lower
levels than did either R79 or G1. Although all three
antisera shared a nearly equal reaction with human
MBP and MBP peptide 45-89 when human MBP peptide 45-89 was the radioligand, S45 recognized an
epitope that was not matched by any of the small peptides of MBP tested and depended much less on the
residue at position 89. This suggests that S45 reacted
with an epitope shifted further away from the carboxyl
end of human MBP peptide 45-89 and existing in a
conformation present only in larger MBP peptides.
The lack of reaction of S45 with bovine MBP 43-88
and guinea pig MBP peptide 43-88 indicates that the
epitope recognized by S45 is affected to a greater extent by minor sequence changes in the MBP sequences of bovine and guinea pig MBP in residues 76-80 as
compared with human MBP 16, 14, 23). The sequence
No 3 September 1986
in guinea pig MBP is serine-glutamine-(-)(-)-arginine-serine,
the sequence in bovine MBP
is alanine-glutamine-histidine-glycine-arginine-proline, and the sequence in human MBP is serine(-)-hstidine-glycine-arginine-threonine
{ 12, 141.
The ability of S45 to detect some MBP-like material in CSF, even though at lower levels, is presumably
related to the high affinity that S45 has for human
MBP and human MBP peptide 45-89. A difference in
affinity may also explain why other antisera and immunoassays using MBP as radioligand successfully detect MBP-like material in CSF { 5 , 8, 201.
The present analysis of epitopes and parallelism of
displacement strongly suggests that a decapeptide from
the carboxyl portion of human MBP peptide 45-89
and in a conformation shared by human MBP, human
MBP peptide 45-89, and human peptide 80-89 represents a dominant epitope of MBP-like material in CSF
after central nervous system myelin injury. Since most,
if not all, of the surface of a protein may be immunogenic {2}, the conformation of MBP {25] or MBP
peptides may play a central role in both the induction
of antibody successfully used in the immunoassays and
in the detection of MBP-like material in CSF. Another
conclusion having clinical relevance that can be drawn
from this study is that the absolute values of MBP or
MBP peptide detected by immunochemical means will
vary depending on the antiserum as well as the assay
used. Standardization of assay procedures and of reagents will be required to give meaning to the numerical values of immunoreactive MBP in body fluids.
This research was supported by a grant (NS 21357) from the National Institutes of Health and by the Research Program of the
Veterans Administration.
We thank Dr Jerome M. Seyer for human MBP synthetic peptides
44-69 and 68-89 and Mrs Sara Baker and Ms Carol Smitherman for
secretarial assistance in preparation of the manuscript.
References
1. Bashir RM, Whitaker JN: Molecular features of immunoreactive myelin basic protein in cerebrospinal fluid of persons with
multiple sclerosis. Ann Neurol 7:50-57, 1980
2. Benjamin DC, Berzofsky JA, East IJ, et al: The antigenic structure of proteins: a reappraisal. Annu Rev Immunol 2:67-101,
1982
3. Biber A, Englert D, Dommasch D, Hempel K. Myelin b,asic
protein in cerebrospinal fluid of patients with multiple sclerosis
and other neurological diseases. J Neurol 225:231-236, 1981
4. Brostoff SW, Reuter W, Hichens M, Eylar EH: Specific cleavage of the A1 protein with cathepsin D. J Biol Chem 249:559567, 1974
5. Burgisser PH, Mathieu JM, DeTribolet N, Gautier E: Dosage
de la proteine basique de la myeline dans le liquide cephalorachidien au cours &affections neurologiques. Schweiz Med
Wochenschr 112:643-647, 1982
6. Carnegie PR, Moore WJ: Myelin basic protein. In Bradshaw
RA, Schneider DM (eds): Proteins of the Nervous System, 2nd
ed. New York, Raven, 1980, pp 119-143
7. Carson JH, Barbarese E, Braun PE, McPherson TA: Components in multiple sclerosis cerebrospinal fluid that are detected
by radioimmunoassay for myelin basic protein. Proc Natl Acad
Sci USA 75:1976-1978, 1978
8. Cohen SR, Herndon RM, McKhann GM: Myelin basic protein
in cerebrospinal fluid as an indicator of active demyelination.
Radioimmunoassay of myelin basic protein in spinal fluid. An
index of active demyelination. N Engl J Med 295:1454-1457,
1976
9. Cohen SR, Jubelt B, Brooks BR, et al: A diagnostic index of
active demyelination: myelin basic protein in cerebrospinal
fluid. Ann Neurol8:25-31, 1980
10. Deibler GE, Krutzsch HC, Martenson RE: A reinvestigation of
the amino acid sequences of bovine, rabbit, monkey, and human
myelin basic proteins. J Biol Chem 260:472-474, 1985
11. Deibler GE, Martenson RE: Chromatographic fractionation of
myelin basic protein. Partial characterization and methylarginine
contents of the multiple forms. J Biol Chem 248:2392-2396,
1973
12. Deibler GE, Martenson RE, Kies MW: Large scale preparation
of myelin basic protein from central nervous tissue of several
mammalian species. Prep Biochem 23139-165, 1972
13. Eylar EH, Brostoff S, Hashim G, er al: Basic A1 protein of the
myelin membrane. The complete amino acid sequence. J Chem
Biol246:5770-5784, 1971
14. Gibson BW, Gilliom RD, Whitaker JN, Biemann K. Amino
acid sequence of human myelin basic protein peptide 45-89 as
determined by mass spectrometry. J Biol Chem 259:50285031, 1984
5. Gilliom RD, Whitaker JN, SeyerJM: Separation of myelin basic
protein peptide 43-88 and its fragments by analytic and preparative high-performance liquid chromatography. J Chromatogr
254:211-218, 1983
6. Hashim GA: Myelin basic prorein: structure, function and antigenic determinants. Immunol Rev 39:60-107, 1978
7. Heinemann MA, Whitaker JN, Seyer JM: The separation and
isolation of human myelin basic protein peptides by reversephase high performance liquid Chromatography in volatile buffers. J Chromatogr 329295-299, 1985
18. Karlsson B, Aliing C: Molecular size of myelin basic protein immunoreactivityinspinalfluid.JNeuroimmunolG: 141-150,1984
19. Norton WT: Formarion, structure and biochemistry of myelin.
In Seigel GJ, Albers RW, Agranoff BW, Katzman RT (eds):
Basic Neurochemistry, 3rd ed. Boston, Little, Brown, 1981, pp
63-92
20. Ohta M, Matsubara F, Konishi T, Nishitani H: Radioimmunoassay of myelin basic protein in cerebrospinal fluid and its clinical
application to patients with neurological diseases. Life Sci
27:1069-1074, 1980
21. Palfreyman JW, Thomas SGT, Ratcliffe JG: Radioimmunoassay
of human myelin basic protein in tissue extract, cerebrospinal
fluid and serum and its clinical application to patients with head
injury. Clin Chem Acta 82:259-270, 1978
22. Rodbard D, Jiwald JE: Computer analysis of radioligand assay
and radioimmunoassay data. In Diczfalusy E (ed): Steroid Assay
by Protein Binding, Karolinska Symposium on Methods in Reproductive Endocrinology. Stockholm, Karolinska Institutet, pp
79-103, 1970
23. Shapira R, McKneally SS, Chou F, Kibler RF: Encephalitogenic
fragment of myelin basic protein. Amino acid sequence of
bovine rabbit, guinea pig, monkey, and human fragment. J Biol
Chem 246:4630-4640, 1971
24. Stewart JM, Young JD: Solid Phase Peptide Synthesis. San
Francisco, Freeman, 1969
25. Stoner G L Predicted folding of P-structure in myelin basic
protein. J Neurochem 43:433-447, 1984
26. Sunderland SM, McPherson TA. Big basic protein: detection in
Whitaker et d.Epitopes of MBP in Human CSF 335
27.
28.
29.
30.
31.
normal and MS brains as well as in CSF from MS patients. Clin
Res 28:700A, 1980
Whitaker JN: Myelin encephalitogenic protein fragments in cerebrospinal fluid of persons with multiple sclerosis. Neurology
(NY) 27:911-920, 1977
Whitaker JN: The antigenic reactivity of small fragments
derived from human myelin basic protein peptide 43-88. J Immunol 129:2729-2733, 1982
Whitaker JN: Indicators of disease activity in multiple sclerosis.
Studies of myelin basic protein-like materials. Ann NY Acad
Sci 436:140-150, 1984
Whitaker JN, Bashir RM, Chou C-HJ, a b l e r RF: Antigenic
features of myelin basic protein-like material in cerebrospinal
fluid. J Immunol 124:1148-1153, 1980
Whitaker JN, Chou C-HJ, Chou FC-H, Kibler RF: Antigenic
336 Annals of Neurology
Vol 20
32.
33.
34.
35.
No 3 September 1986
determinants of bovine myelin encephalitogenic protein recognized by rabbit antibody to myelin encephalitogenic protein. J
Biol Chem 250:9106-9111, 1975
Whitaker JN, Chou C-HJ, Chou FC-H, a b l e r R F Molecular
internalization of a region of myelin basic protein. J Exp Med
146:317-333, 1977
Whitaker JN, Lisak RP, Bashir RM, et al: Immunoreactive myelin basic protein in the cerebrospinal fluid in neurological disorders. Ann Neurol 7:58-64, 1980
Whitaker JN, Seyer JM: The sequential limited degradation of
bovine myelin basic protein by bovine brain cathepsin D. J Biol
Chem 254:6956-6963, 1979
Whitaker JN, Synder DS: Myelin components in the cerebrospinal fluid in diseases affecting central nervous system myelin.
Clin Allergy Immunol 2:469-482, 1982
Документ
Категория
Без категории
Просмотров
3
Размер файла
920 Кб
Теги
epitopes, basic, protein, myelin, immunoreactivity, human, fluid, cerebrospinal
1/--страниц
Пожаловаться на содержимое документа