close

Вход

Забыли?

вход по аккаунту

?

The specificity of antibodies to the Fab В╨Ж2 fragment of human IgG.

код для вставкиСкачать
562
THE SPECIFICITY OF ANTIBODIES TO THE
F(ab’)2 FRAGMENT OF HUMAN
RALPH HEIMER, LAWRENCE D. WOLFE, and JOHN L. ABRUZZO
The specificity of IgG anti-F(ab’)2 antibodies was
examined in unfractionated sera of patients with rheumatoid arthritis and also with affinity-purified antibody
preparations. Examination of the sera by an enzymelinked immunosorbent assay, using pooled human
F(ab’h fragments absorbed to microtiter plates, revealed that IgG anti-F(ab’)* antibodies cross-react with
human and rabbit IgG and rabbit F(ab’)z. IgG antiF(ab’)z antibodies were purified by affinity chromatography and, when tested by a fluid-phase inhibition
enzyme-linked immunosorbent assay, were found to be
of 2 types. One fraction, similar to pepsin agglutinator,
reacted with human F(ab’)z fragment alone. The other
fraction was cross-reactive with human IgG and yet
failed to react with idiotopes on Fab or epitopes on Fc
fragments. The IgG anti-F(ab’)2 antibodies we purified
had no reactivity toward a human immune complex
prepared from tetanus toxoid and antitoxoid.
Antibodies to F(ab‘):! fragments of pooled human IgG are elevated in patients with rheumatoid
arthritis (RA) (1-7), patients with subacute bacterial
endocarditis (2,7), the elderly (8), and patients with
infertility linked to anti-sperm antibody production (9).
Some of these antibodies appear to be cytotoxic to B
From the Department of Biochemistry and the Department
of Medicine, Jefferson Medical College of Thomas Jefferson University. Philadelphia, Pennsylvania.
Supported by NIH grant AM-26409 and by a grant from the
Eastern Pennsylvania Chapter of the Arthritis Foundation.
Ralph Heimer, PhD; Lawrence D. Wolfe, PhD; John L.
Abruzzo, MD.
Address reprint requests to Ralph Heimer, PhD, Department of Biochemistry, Jefferson Medical College of Thomas Jefferson University, 1020 Locust Street, Philadelphia, PA 19107.
Submitted for publication May 21, 1984; accepted in revised
form November 7, 1984.
Arthritis and Rheumatism, Vol. 28, No. 5 (May 1985)
lymphocytes (10). Cross-reactivity with F(ab’), fragments of IgG of other species has been reported
recently (11-14). Previous studies have stressed circumstantial evidence that some of these antibodies
recognize idiotopes present in F(ab’>:! fragments prepared from pooled human IgG (3,6). Other antibodies
of this family react only with F(ab’), but not with
intact IgG or Fab fragments, indicating recognition of
epitopes accessible only after some structural modification of IgG (1,2,15). Previous studies of specificity
of anti-F(ab’), antibodies, however, are difficult to
interpret, because of use of assay procedures which
are susceptible to interference by rheumatoid factors
(RF) (1-3,7,16). The assay difficulties are compounded
by the problems of separating RF from purified antiF(ab’), antibodies (3). We directed our research to
preparing purified anti-F(ab’), antibodies free of RF,
foregoing, however, attempts to fully recover a subset
“hidden” within complexes (6) and possibly associated with antiidiotypic activity. This report describes a
purification protocol and details specificity studies of
purified antibodies with a panel of inhibitors using
isotype-specific enzyme-linked immunosorbent assays
(ELISA).
MATERIALS AND METHODS
Sera. Sera in this study were obtained from 7 patients
with classic RA who were seropositive for RF and from 2
seronegative, apparently healthy volunteers. The sera were
kept at -20°C until use.
Antisera. Affinity-purified goat anti-human IgM and
anti-human IgG antibodies (Tago, Inc., Burlingame, CA)
were absorbed with pooled F(ab’)z fragments of human IgG
to make them specific for the Fc fragments of the above
isotypes. Rabbit anti-human IgG (Fd fragment-specific) was
ANTIBODY SPECIFICITY TO F(ab')*
obtained from Calbiochem-Behring, La Jolla, CA. Rabbit
anti-goat IgG and goat anti-human IgG (Fc fragment-specific) were obtained from Cappel Laboratories, West Chester,
PA, The specificity of these antibodies was confirmed by
immunoelectrophoresis. Affinity-purified antibodies were
conjugated to alkakine phosphatase (Sigma, St. Louis, MO)
with glutaraldehyde according to the recommendations of
the manufacturer, to obtain an approximate enzyme activity
of 500 units/mg.
Assays. ELISA for anti-F(ab')z antibodies and RF
were carried out in rigid flat-bottom polystyrene microtiter
plates (Immulon; Dynatech Laboratories, Inc., Alexandria,
VA). F(ab')2 for anti-F(ab')2 antibody assays and Fc for RF
assays, each at 500 ng/well, were adsorbed for 12 hours.
After washing with Buffer A (O.05M Tris [hydroxymethyl]
aminomethane, 0.15M NaCI, 1% sodium azide, pH 1.4)
containing 0.05% Tween 20 (Buffer A-Tween), sera or
purified antibodies (diluted in 0.1% egg albumin) were allowed to bind for 2 hours.
For an assay of RF, 0.01 units of goat anti-human
IgM or rabbit anti-human Fd fragment linked to alkaline
phosphatase was added for 2 hours, washed with Buffer ATween 20, and further incubated with 0.1 mg/well p-nitrophenyl phosphate (Sigma).
For IgM and IgG anti-F(ab')z antibody assays, we
adapted the radioimmunoassay described previously ( 5 ) to
conditions for ELISA. In this method, an excess unlabeled
goat anti-Fc fragment antibody was interposed between
binding of anti-F(ab')z antibodies and the addition of the
enzyme-conjugated antibody. This step minimizes interference by RF. After a 2-hour incubation with goat anti-Fc
antibody, the presence of the anti-F(ab')2 antibodies was
determined with alkaline phosphatase-linked rabbit antigoat IgG and/or goat anti-human IgM. In inhibition experiments, highly purified IgG anti-F(ab')2 antibody was measured directly with enzyme-linked goat anti-human Fc. All
ELISA were quantitated by measuring absorbance at 405 nm
with a Titertek Multiscan Monitor (Flow Laboratories,
McLean, VA).
Isotype-specific radioimmunoassays (RIA) for IgG
anti-F(ab')z were carried out as described previously ( 5 ) .
Antigens and inhibitors. Human IgG was purified
from pooled Cohn fraction I1 (United States Biochemical
Corp., Cleveland, OH) or individual serum by passage
through a DEAE-AffiBlue ion exchange column (Bio-Rad
Laboratories, Richmond, CA), pH 8.0, in 0.02M K2HP04
buffer. All preparations were absorbed with pooled IgG-Sepharose to remove RF and other antibodies reacting with
intact IgG. Rabbit IgG was purified from serum by adsorption to protein A-Sepharose.
F(ab')2fragments were prepared from purified IgG of
pooled Cohn fraction I1 which had been digested with pepsin
and gel-filtered through Sephadex G-200 (Pharmacia Fine
Chemicals, Piscataway, NJ) as described previously ( 5 ) . The
F(ab')z preparations were examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under
nonreducing conditions and found to give a single broad
band. Preparations were also tested on microtiter plates with
anti-Fc antibodies conjugated with alkaline phosphatase,
without finding detectable amounts of undigested IgG and
Fc' fragments. The F(ab')? preparations were absorbed with
563
IgG-Sepharose and F(abr)2-Sepharoseprior to use as antigen
or inhibitor to remove RF and anti-F(ab')* antibodies.
Fab fragments of human IgG were obtained from
Rockland, Inc. (Gilbertsville, PA). The material was tested
by SDS-PAGE under reducing and nonreducing conditions
and found to be free of intact IgG or Fc fragments. The
material reacted with rabbit anti-human IgG (Fd fragmentspecific), but not with goat anti-human IgG (Fc fragmentspecific).
Fc fragments, prepared by plasmin cleavage of
pooled human IgG, were obtained from Alpha Therapeutics
Corp. (L,os Angeles, CA). On SDS-PAGE, in the absence of
reducing agents, the antigen appeared to be a single material
with an estimated molecular weight of 5 x lo4. The material
did not react with anti-human F(ab')2 antibodies by ELISA.
Light chains were prepared from pooled human IgG
by a modification of the method of Fleischman et al (17).
Reduction was carried out for 1 hour at room temperature
with 10 m M dithiothreitol in 0.5M Tris-HCI, pH 8.0. Twenty
millimoles of iodoacetamide was added and the solution was
incubated for 30 minutes at 4"C, dialyzed against 1M acetic
acid, and then fractionated on Sephadex G-200. Light chains
were dialyzed against Buffer A and shown by SDS-PAGE to
be free of intact IgG and free gamma chains.
Immune complexes were prepared from tetanus toxoid obtained from the Massachusetts Department of Public
Health, Boston. The material was radiolabeled with '"Na by
the method of Fraker and Speck (18) and moved on
SDS-PACE in the presence of 2-mercaptoethanol as a major
component at approximately 180,000 daltons. Antibody to
tetanus toxoid was purified from human tetanus immune
globulin (Wyeth Laboratories, Inc., Marietta, PA) on a
column of tetanus toxoid conjugated to CNBr-activated
Sepharose. Soluble immune complexes were formed by
mixing 50 pg of tetanus toxoid with 350 pg of affinitypurified antitoxoid. This ratio (Ab6: Agl) appears to correspond to the equivalence point when studied by precipitin
reactions in the presence of 3% polyethylene glycol (PEG).
The complexes remained soluble in the absence of PEG.
Purification of anti-F(ab')Z antibodies. The purification procedure is outlined in Figure 1. Briefly, 2 ml of serum
from patients with RA or 6 ml of serum from healthy
volunteers was diluted fivefold in Buffer A-Tween and
incubated for 2 hours at room temperature with 10 mg of Fc
fragments of IgG covalently linked to 1 gm CNBr-activated
Sepharose. After washing with Buffer A-Tween, bound
anti-Fc antibodies were eluted with glycine-HC1 buffer
(O.IM, pH 2.0), and neutralized immediately with 3M Tris.
The fraction that had not bound to Fc-Sepharose was
incubated for 2 hours at room temperature with 10 mg
F(ab')2 covalently linked to CNBr-activated Sepharose.
After washing with Buffer A-Tween, bound material
was eluted sequentially with acetate buffer (O.IM, pH 5.0).
followed by glycine-HCI buffer. One milligram of egg albumin was added to each milliliter of eluate. Fractions eluted
from imniunoadsorbents were concentrated in a Savant
Speed Vac Concentrator and 200-pl aliquots of these fractions were layered on a 10-40% sucrose gradient in Buffer A.
The samples were ultracentrifuged in a Beckman SW 50L
rotor at 30,000 revolutions per minute for 18 hours at 4°C.
Fractions were collected by puncturing the bottom of the
HEIMER ET AL
564
Serun (diluted 1/51
1
-
+
Fc
Sepharose
absorbed to Fc
not absorbed
F(ab‘)?
-
Sepharose
not absorbed
absorbed to Fcab’)?
(Fraction 11)
-
I
pH 2.:
-
Sepharose
elution
Fraction I
Sepharose
A
pH 5 . 0 e l u t i o n
Fraction I l l
pH 2 . 0 e l u t i o n
.,
Fraction IV”
Figure 1. Purification of anti-F(ab‘)2antibodies. Fractions I and 1V
were further purified by sucrose density gradient ultracentrifugation
prior to testing specificity with various inhibitors.
tube and were screened by ELISA for IgM and IgG antibodies to F(ab’)z or Fc fragments. The IgG concentration of
some of the fractions was determined by ELISA, using
microtiter wells containing adsorbed protein A , and adsorbed IgG was detected with alkaline phosphatase-conjugated F(ab’)z fragments of goat anti-human IgG.
RESULTS
Specificity of IgG anti-F(ab’)2 antibodies in serum. To test for the specificity of these antibodies with
IgG isotype-specific RIA, RA patient sera, diluted
1 :5 , were absorbed with various antigens (1 mg each)
covalently linked to Sepharose CL-4B. Following
incubation at 4°C for 18 hours, the absorbed sera were
tested for residual anti-F(ab’)* antibody activity by
comparing antibody levels with those of controls prepared from similarly diluted sera which had been
exposed to egg albumin-Sepharose.
The loss of reactivity due to exposure to
F(ab’)*-Sepharose was set at 100% inhibition. The
results (Table 1) indicate that more anti-F(ab‘)2antibody was consistently absorbed by human F(ab‘)*-Sepharose than by human IgG, rabbit IgG, or rabbit
F(ab’)*-Sepharose. Since the 5 sera tested contained
considerable amounts of RF, it remained uncertain
whether reactivity with human and rabbit IgG was due
to RF itself, or whether a considerable portion of
anti-F(ab’)* antibodies was directed against epitopes
shared by human and rabbit IgG and their F(ab’)*
fragments.
Purification of IgG anti-F(ab’)* antibodies.
When F(ab’)2-Sepharose was used for the purification
of anti-F(ab’)* antibodies from diluted sera which
contained large amounts of RF, preparations obtained
were heavily contaminated with RF. A more successful purification protocol was eventually devised (Figure 1) in which sera, diluted 1 : 5 , were exposed initially
to Fc-Sepharose to remove substantial amounts of RF.
The sera were then exposed to F(ab‘)2-Sepharose from which anti-F(ab’)* antibodies and further
quantities of RF were removed by a 2-step elution.
The progress of the purification was followed by
isotype-specific ELISA, in wells coated with Fc fragments for estimating RF and in wells coated with
F(ab‘)* fragments for estimating anti-F(ab‘)2antibodies. The results of the fractionation of 2 ml of serum
from patient SL, who had RA and high levels of RF,
are detailed in Table 2. A substantial portion of the
initial IgG anti-F(ab’)* antibody activity (titer 1 :320)
was recovered from F(ab’)2-Sepharoseafter elution at
pH 2.0 (fraction IV, titer 1 :80). This fraction was free
of IgG-RF. Considerable IgG anti-F(ab’)2 antibody
activity was recovered also in the initial absorption
step with Fc-Sepharose by elution at pH 2.0 (fraction
I, titer 1 :40).
Using our fractionation protocol (Figure 1 and
Table 2), IgG anti-F(ab’)* antibodies were prepared
from the sera of 2 RA patients (including SL) and from
Table 2. The fractionation of serum from rheumatoid arthritis
patient SL, using Fc-Sepharose followed by F(ab‘)2-Sepharose*
Titer
Table 1. Percent inhibition of serum IgG anti-F(ab’)2 antibody
activity by absorption with human IgG-Sepharose, rabbit IgGSepharose, and rabbit F(ab’)2-Sepharose*
Patient
Human IgGSepharose
Rabbit IgGSepharose
Rabbit F(ab’)*Sepharose
LH
LB
JC
FB
DH
46
19
32
17
19
26
39
18
3
13
14
59
18
1
8
* The activity of serum IgG a ~ ~ t i - F ( a bantibody
‘)~
following absorption with the above listed antigens was compared with that of human
F(ab’)2-Sepharose-exposedserum (100% inhibition).
Initial serum
Fraction I
Fraction I1
Fraction 111
Fraction IV
40,960
640
10,240
1,280
40
640
40
320
40
<5
640
20
40
20
20
320
40
<5
<5
80
* Fractionation was done by the same method as in Figure
1.
Enzyme-linked immunosorbent assay performed on fractions
adjusted to initial serum volume. The titers were determined by the
amount of dilution of sample required for an absorbance of 0.500 at
405 nm. RF = rheumatoid factor.
ANTIBODY SPECIFICITY TO F(ab’)2
Table 3.
Serum
source
Purification of IgG anti-F(ab’)2antibodies*
Titer
Initial
Fraction I
Fraction IV
SL (RA)
320
40
80
DW (RA)
640
80
320
KH (normal)
40
<S
<5
SP (normal)
320
I60
20
* The titers were determined by the amount of dilution of sample
required for an absorbance of 0.500 at 405 nm in enzyme-linked
immunosorbent assay for IgG anti-F(ab’)? antibody; fractions refer
to material eluted at pH 2.0 from Fc-Sepharose (fraction I ) and
F(ab’)?-Sepharose (fraction IV) (see Figure I). Volumes of all
fractions were adjusted to correspond to that of initial serum. RA =
rheumatoid arthritis.
2 normal sera (Table 3 ) . It can be seen that IgG
anti-F(ab‘)2 antibody activity was recovered to a
substantial extent in fraction IV in the 2 RA patients’
sera. In contrast, anti-F(ab’)? antibody recovery was
best in fraction 1 in the 2 normal sera.
Specificity of IgG anti-F(ab’)* antibodies recovered in fraction IV. IgG anti-F(ab’)2antibodies purified
as indicated in Figure 1, and representing material
recovered as fraction IV, were purified further by
sucrose density gradient ultracentrifugation. Fractions
devoid of IgM anti-F(ab’)? or IgM anti-Fc antibody
were tested with a panel of potential inhibitors. The
concentration of each antibody preparation chosen
(1.6 pmoles) was on the linear portion of a dose-response curve obtained by ELISA. The concentration
of all inhibitors was fixed at 560, 280, 140, and 70-fold
mo1a.r excess relative to antibody. The assays were
carried out at room temperature for 2 hours in the
presence of 1 mg/ml of egg albumin in Buffer ATween.
The amount of IgG anti-F(ab’)?bound to microtiter wells coated with F(ab’)2 fragments was measured with enzyme-conjugated goat anti-human Fc
fragment. Absorbance at 405 nm, obtained when inhibitors were used in the absence of anti-F(ab’)2 antibodies, ‘were subtracted from test results. Figure 2 gives
the results of 1 set of experiments and is representative
of 3 separate experiments with the same 2 RA sera (SL
and DW). Only human F(ab’)?fragment was an effective inhibitor. Pooled human IgG and rabbit F(ab‘)z
fragrnents were relatively poor inhibitors and light
chains, Fab and Fc fragments, and rabbit IgG did not
inhibit at all. An immune complex prepared from
tetanus toxoid-antitoxoid had no greater inhibitory
activity than the affinity-purified anti-tetanus toxoid
IgG itself.
The same 2 preparations (patients SL and DW)
565
were also tested in the presence of autologous IgG
which had been pre-absorbed with pooled F(ab’)?-Sepharose and pooled Fc-Sepharose. No inhibition was
found. When the 2 antibody preparations were added
separately to a normal serum which had been diluted
1 : 10 and absorbed to remove endogenous anti-F(ab‘)?
antibodies, levels of antibody activity also remained
unchanged.
The 2 purified IgG anti-F(ab’)2antibody preparations (fraction 1V) obtained from the sera of normal
individuals (RH and SP) and tested in concentrations
equal to those in Figure 2 reacted similarly to the 2
preparations obtained from patients with RA, with
respect to lack of inhibition by light chains, Fab and Fc
r
A
E
C
v)
0
*
t
W
0
z
a
.8
1
m
U
0
v)
m
a
I
2
I
4
I
I
I
6
8
10
INHIBITOR (MOLES x
1
lo-’’)
Figure 2. Inhibition of IgG anti-F(ab’)?antibodies from rheumatoid
arthritis patients DW (A) and SL (B) by pooled human F(ab’)z
rabbit F(ab’)? fragments (0-0)-and hufragments (0-O),
man IgG (W---W).
IgG anti-F(ab‘)Zantibody in fraction IV (further
purified by sucrose density gradient ultracentrifugation) was examined by enzyme-linked immunosorbent assay in the presence of 113900 pmoles inhibitor (70-560-fold excess with respect to antibody
present).
HEIMER ET AL
566
fragments, and rabbit IgG. Relative to the inhibition
obtained with F(ab’), fragments (loo%), 1 preparation
(from SP) was 40% inhibited by a 560-fold molar
excess of rabbit F(ab’), fragment and preparations
from SP and RH were inhibited 50-60% by a 560-fold
molar excess of pooled human IgG.
Specificity of IgG anti-F(ab’)2antibodies recovered in fraction I. Fraction I was also subjected to
sucrose density gradient ultracentrifugation to remove
IgM antibodies. Each of these preparations, however,
retained small amounts of IgG-RF (titers 5 1 :40). IgG
anti-F(ab’), antibody reactivity was not inhibited in
the presence of 560-fold and 140-fold molar excess of
pooled human Fab and Fc fragments or rabbit IgG and
rabbit F(ab’)*fragments. Unlike the fraction IV preparation, similar amounts of pooled human IgG and antitetanus toxoid antibody inhibited the IgG anti-F(ab‘)l
antibody activity between 29 and 80%, relative to
F(ab’),, whose inhibitory action was set at 100%
(Table 4). The immune complex prepared from tetanus
toxoid-antitoxoid was no more inhibitory than the
antibody itself.
DISCUSSION
Serum IgG anti-F(ab’), antibody activity appears to be directed against epitopes present on the
surface, as well as in the interior of the F(ab’), portion
of IgG. This heterogeneity is seen when sera are
analyzed for this antibody after absorption with immobilized human or rabbit IgG and F(ab’)* fragments
thereof. There appears to be considerable variation
among human sera in response to these absorptions, as
some serum antibodies are extensively cross-reactive
with rabbit IgG and rabbit F(ab’)2fragments (Table 1).
Cross-reactivity with F(ab’), fragments of IgG of other
species has been reported recently (11-14).
The effectiveness of absorbing sera that contain
soluble IgG with immobilized IgC may, however, be
Table 4. Percent inhibition of IgG anti-F(ab’), antibody activity in
fraction I*
Serum
source
Pooled
human IgG
Anti-tetanus
toxoid antibody
SL (RA)
DW (RA)
SP (normal)
29
78
42
34
39
80
* The numbers represent the percent inhibitory activity tested at
400-fold molar excess relative to input antibody in fraction 1, and
also relative to inhibition by human F(ab’)2fragment (100%). RA =
rheumatoid arthritis.
compromised. Additionally, endogenous RF may also
affect the #absorptions. To better define the specificities of IgG anti-F(ab’), antibodies, attempts were
made to purify these, preferably free of RF, and then
to test specificity in the presence of known quantities
of soluble inhibitors.
Initially we encountered difficulties, as did other investigators (3), in obtaining purified anti-F(ab’),
antibodies without concomitant RF impurities. Eventually we developed a purification protocol in which
we made use of the fact that RF are usually low affinity
antibodies, which can be removed from solid-phase
IgG at pH 5.0. This selective elution facilitated the
purification of anti-F;(ab’)l antibodies free of IgG-RF
activity (Table 2). The activity of the highly purified
IgG anti-F(ab’), antibody preparations was virtually
unaffected by the presence of a 560-fold molar excess
of pooled human IgG (Figure 2), human Fab and Fc
fragments, light chains, as well as a soluble complex of
tetanus toxoid and human anti-tetanus toxoid at
equivalence. The anti-F(ab’)* antibody activity was
also unaffected when added to 1 : 5 diluted serum,
depleted of RF and anti-F(ab’), antibody by appropriate absorptions. Moreover, rabbit IgG and their F(ab’):
fragments were also ineffective inhibitors.
It would appear that nearly all IgG anti-F(ab’),
antibodies from the 2 patients with RA present in
fraction IV, and accounting for substantial antibody
activity when compared with whole serum (Table 3).
are directed against epitopes uncovered by pepsin
digestion at pH 4.0. These epitopes are inaccessible to
antibody in the intact IgG molecule. To that extent,
fraction IV from these 2 sera behaved as if it contained
mainly the previously described pepsin agglutinator
(PA), but the activity varied from the original description (2) and a later study (7) of these antibodies by
remaining unreactive with immune complexes. In previous work, the specificity of PA was tested in whole
serum using only insoluble immune complexes. Thus.
other anti-F(ab’)* antibodies and RF may have influenced the analysis. On the basis of our findings, a
revised definition of PA would have these antibodies
react with F(ab’), fragments exclusively.
The IgG anti-F(ab’)2antibodies offraction IV of
the 2 normal individuals appeared to be heterogeneous, since they exhibited PA activity and considerable cross-reactivity with IgG. In view of the lack of
reactivity with Fab fragments, it is unlikely that the
cross-reacting antibodies are antiidiotypic antibodies.
Because of lack of reactivity with Fc fragments, the
location of the reactive epitope for the subset of
567
ANTIBODY SPECIFICITY TO F(ab‘)2
anti- F(ab‘), antibody reacting with intact IgG may be in
the hinge region.
Fraction I, obtained in our purification procedure by absorption to Fc-Sepharose, also contained
consiiderable quantities of IgG anti-F(ab’)* antibodies
(TabKe 3). The latter may co-purify with IgM-RF which
can bind immune complexes. including those which
may contain anti-F(ab’)* antibodies (6). When fraction
I was further purified by sucrose density gradient
ultracentrifugation, all of the IgM-RF activity was
removed, leaving IgG-RF and IgG anti-F(ab’)* antibodies which were demonstrable by separate ELISA,
using, either Fc or F(ab’)* fragments absorbed to
microtiter wells. In ELISA, anti-F(ab’)* antibody activity was unaffected by the addition of 560-fold excess
Fab ,and Fc fragments, as well as the tetanus toxoid
immune complex preparation. Despite such lack of
reactivity, varying degrees of inhibition were observed
when the preparations were tested in the presence of
pooled IgG or an affinity-purified antitoxoid IgG preparation (Table 4). Fraction I, therefore, was quite
similar to fraction IV, except that there was less PA
present and, more importantly, there were antibodies
cross-reacting with intact IgG. Because of lack of
reactivity with Fab fragments, fraction I, like fraction
IV, contains no measurable antiidiotypic antibodies.
In a recent report (6), it was suggested that
antibody activity to reduced and alkylated pooled
human F(ab’)* fragments was detectable both in the
free form and also in complexes as “hidden antibody”
and that free antibody predominated over hidden
antiblody in RA sera but not in normal sera. It was
suggested that the hidden antibodies might be the
result of interaction between idiotypes and antiidiotypes. Our failure to encounter a significant contribution to anti-F(ab’)* activity by antiidiotypic antibodies
among the affinity-purified fractions may be attributed
to our purification procedure, since we made no
attempts to dissociate by acid hidden anti-F(ab’)*
antibodies.
The emphasis in this report has been on IgG
anti-F(ab’)* antibodies. This is not to say that JgM
anti-E;(ab’)*antibodies are less important, but rather
that IgM anti-F(ab’)*antibodies were not readily purified by our procedures. A report on the specificity of
these antibodies in whole serum has been published
elsewhere (13).
In summary, on the basis of our results, albeit
with a limited sample, we suggest that IgG anti-F(ab’)2
antibodies make up a heterogeneous population. A
substantial portion of the IgG anti-F(ab’)* anti-
body activity is due to PA. Moreover, PA is redefined
here as antibodies reacting with epitopes which become accessible on pepsin digestion but remain inaccessible when on intact IgG or IgG immune complexes. A subset of IgG anti-F(ab’)* antibodies, especially
those found in fraction I, appears to be cross-reactive
with human IgG, but because of lack of reactivity with
Fab fragments cannot be considered to be antiidiotypic. Sera also contain other IgG anti-F(ab’)*antibodies
with cross-reactivity for rabbit IgG and rabbit F(ab’)*
fragments, but these could not be purified effectively
by affinity chromatography using Fc-Sepharose followed by F(ab’)2-Sepharose.In contrast to the work of
others (3,6,16), we found no evidence suggestive of
reactivity with idiotypes. This may be explained in
part by our use of pooled and absorbed F(ab’)z fragments as immunoabsorbents and test antigens.
REFERENCES
1. Osterland CK, Harboe M, Kunkel HG: Anti-gammaglobulin factors in human sera revealed by enzymatic
splitting of anti-Rh antibodies. Vox Sang 8: 113-152,
1963
2. Harboe M, Rau B, Aho K: Properties of various antigammaglobulin factors in human sera. J Exp Med
121503-5 19, 1965
3. Nasu H , Chia DS, Knutson DW, Barnett EV: Naturally
occurring human antibodies to the F(ab‘)z portion of
IgG. Clin Exp Immunol 42:378-386, 1980
4. Natvig JB: Heterogeneity of anti-gammaglobulin factors
detected by pepsin digested human gamma-G-globulin.
Acta Pathol Microbiol Scand 66:369-381, 1966
5. Heimer R, Wolfe LD, Abruzzo JL: IgM and IgG antiF(ab’)2antibodies in rheumatoid arthritis and systemic
lupus erythematosus. Arthritis Rheum 25: 1298-1306,
1982
6. Birdsall HH, Lidsky MD, Rossen RD: Anti-Fab‘ antibodies in rheumatoid arthritis: measurement of the relative quantities incorporated in soluble immune complexes in sera and supernatants from cultured peripheral
blood lymphocytes. Arthritis Rheum 26: 148t-1492, 1983
7. Davey MP, Korngold L: Association of anti-F(ab’)z
antibodies (pepsin agglutinators) with immune complexes as determined by enzyme linked immunosorbent
assays. Int Arch Allergy Appl Immunol 67:278-283,
1982
8. Lawrence TG Jr, Williams RC Jr: Studies of human
anti-gamma globulin factors reacting with pepsin-digested gamma globulins. J Exp Med 125:233-248, 1967
9. Mathur S, Khansari N , Genco PV, Williamson HO,
HEIMER ET AL
10.
11.
12.
13.
14.
Perry LC, Cracker AJ, Rust PF, Fudenberg HH: Association of antibodies to Fab with sperm antibodies and
HLA (abstract). Fed Proc 42: 1209, 1983
Nasu H, Chia D, Iwaki Y , Barnett EV, Terasaki PL:
Cytotoxicity of anti-Fab antibodies against B lymphocytes in rheumatoid arthritis. Arthritis Rheum 24: 12781282, 1981
OIds LC, Miller JJ: Anti-F(ab’)2antibodies that interfere
with interpretation of the anti-C3 assay for immune
complexes in children with rheumatic diseases. Arthritis
Rheum 27:330-336, 1984
Wolfe LD, Abruzzo JL, Heimer R: Specificity of antiF(ab’)z antibodies (abstract). Fed Proc 42:931, 1983
Wolfe LD, Abruzzo JL, Heimer R: Specificities of IgM
anti-F(ab’)* antibodies. Immunol Commun 13: 15-27,
1984
Ling NR, Drysdale P: Antibodies in human sera to
F(ab’)zfragments of monoclonal and polyclonal IgG. Int
Arch Allergy Appl Immunol 66:459-463, 1981
15. Waller M , Blaylock K: Further studies on the anti-globulin factors in human serum to the pepsin digested
fragment of the Ri anti-Rh antibody. J Immunol 97:438443, 1966
16. Nasu H, Chia DS, Taniguchi 0, Barnett EV: Characterization of anti-F(ab’)>antibodies in SLE patients: evidence for crossreacting auto-anti-idiotynic antibodies.
Clin Immunol Immunopathol 25:80-90, 1982
17. Fleischman JB, Pain R, Porter RR: Reduction of gamma
globulins. Arch Biochem Biophys (suppl) I: 174-187,
1961
18. Fraker PH, Speck JC Jr: Protein and cell membrane
iodinations with sparingly soluble chloroamide, 1,3,4.6
tetrachloro-3a, 6a-diphenylglycoluril. Biochem Biophys
Res Comm 80:849-857, 1978
Документ
Категория
Без категории
Просмотров
6
Размер файла
644 Кб
Теги
antibodies, igg, specificity, fragmenty, fab, human
1/--страниц
Пожаловаться на содержимое документа