вход по аккаунту


Igm and igg anti-fab В Ъ2 antibodies in rheumatoid arthritis and systemic lupus erythematosus.

код для вставкиСкачать
Radioimmunoassays for anti-F(ab’)z antibodies,
which feature the use of goat anti-human Fc antibody
for correcting potentiation of IgM anti-F(ab’)z antibody
titers by endogenous IgM anti-Fc antibodies (rheumatoid factors), are described. Individuals with classic
rheumatoid arthritis had significantly more IgM antiF(ab’)2 antibody (P < 0.001) and IgG anti-F(ab’)2
antibody (P = 0.05) than did individuals with systemic
lupus erythematosus or normal volunteers. There is
some similarity in patterns of isotype distribution of
anti-F(ab’)z antibodies and rheumatoid factors.
Elevated levels of antibodies to epitopes located on the Fab or F(ab’);! portions of IgG are often
found in the sera of patients with rheumatoid arthritis
(RA) and other diseases (1-6). The biologic significance of these findings is just beginning to emerge.
For example, it was shown only recently that
anti-F(ab’);! antibodies react with allotypic and/or idiotypic epitopes on the surface of subpopulations of B
lymphocytes of normal donors (7). There is also recent
evidence that some anti-F(ab’)2 activity is directed
against a repertoire of idiotypes that is commonly
included in pooled human IgG (5). Earlier it was
suggested that anti-F(ab’)2antibodies, like the classic
anti-Fc rheumatoid factors (RF), arise from stimulaFrom the Department of Biochemistry and Medicine,
Thomas Jefferson University, Philadelphia, PA.
Supported in part by grant AM-26409 from the National
Institutes of Health.
Ralph Heimer, PhD: Professor of Biochemistry; Lawrence
D. Wolfe, PhD; John L . Abruzzo, MD: Professor of Medicine,
Director of Division of Rheumatology.
Address reprint requests to Ralph Heimer, PhD, Thomas
Jefferson University, 1020 Locust Street, Philadelphia, PA 19107.
Submitted for publication February 12, 1982; accepted in
revised form June 25. 1982.
Arthritis and Rheumatism, Vol. 25, No. 11 (November 1982)
tion of autologous antigen-antibody complexes (2).
The relationship of anti-F(ab’)2 to anti-Fc antibodies,
however, has remained unclear, perhaps because the
technology of detection of R F was well ahead of that
of anti-F(ab’)2 antibodies. Quantitation of R F of various isotypes dates back to 1967 (8), and reliable
radioimmunoassays (RIA) for IgM and IgG R F have
since been described (9,lO). An effective RIA for antiF(ab’)*antibodies, however, was onlyreported recently
The present report is the first on separate
detection of IgM and IgG anti-F(ab’)2 antibodies. Our
RIA was designed to overcome interference of endogenous RF, a problem inherent in direct tests for class
specific antibodies (1 1-16). The data obtained by our
RIA suggest some similarity in isotype patterns of antiF(ab’)zantibody and R F when sera of patients with RA
are compared with sera of patients with systemic lupus
erythematosus (SLE) or with sera of normal donors.
Sera. Sera of patients and healthy volunteers were
kept at -20°C and -70°C until use. Patients with RA and
SLE fulfilled American Rheumatism Association criteria
Buffers. Buffer A consisted of 0.05M Tris (hydroxymethyl) aminomethane, 0.15M sodium chloride, and 0.1%
sodium azide at pH 7.4. Buffer A-Tween was made by
adding 0.05%, by volume, of Tween 20 to Buffer A. Phosphate buffered saline contained 1 part of O.15M phosphate
buffer, pH 7.2, and 9 parts physiologic saline.
Latex fixation. The tests were performed in tubes
with the use of latex particles (Dow Chemical, Midland, MI)
coated with human IgG (Cohn Fraction 11, United States
Biochemical Corp., Cleveland, OH) (19).
Preparation of F(ab’)*fragments. One hundred mi&
grams of Cohn Fraction I1 were digested with 2 mg pepsin
(Worthington Biochemical Co., Freehold, NJ) for 24 hours
at 37°C in pH 4.0 acetate buffer. The digest, dialyzed in
Buffer A, was fractionated on a 2.4 x 100-cm column of
Sephadex G-200 (Pharmacia Fine Chemicals, Piscataway ,
NJ), previously calibrated with IgG and human serum albumin. Fractions eluting in the range of the F(ab’Iz fragment
were pooled and concentrated by negative pressure dialysis.
The solution was absorbed for 4 hours at 4°C with goat antihuman Fc-Sepharose to remove undigested IgG, and with
human IgG-Sepharose to remove anti-IgG antibody activity.
When the resulting preparation, which contained 34 mg
protein, was tested at 1 mg/ml by immunoelectrophoresis in
agarose gel against horse anti-human antiserum (Behring
Diagnostics, Somerville, NJ) and against goat anti-human
Fab antiserum (Hyland Laboratories, Los Angeles, CA), it
was found to contain F(ab’h only. Furthermore, the preparation showed no Fc content when placed into microtiter
wells and reacted in an RIA with ”’I-goat anti-Fc antibody
(Cappel Laboratories, Cochranville, PA) or with a monoclonal IgM (mIgM) R F followed by ‘251-goatanti-human IgM
antibody (Tago, Inc., Burlingame, CA).
Preparation of Fc fragments. One hundred milligrams
of Cohn Fraction 11, dissolved in 0.05M phosphate buffer,
pH 7.1, 0.005M cysteine, and 0.002M ethylenediamine tetraacetate (EDTA) was incubated at 37°C for 18 hours with 1
mg papain (Worthington Biochemical Corp.) The solution
was dialyzed against phosphate buffered saline for 24 hours
and applied to a 2 . 5 ~ 9 0 - c mcolumn of Sephacryl S-300
(Pharmacia Fine Chemicals). Fractions corresponding to the
position of the Fab and Fc mixture but free of whole IgG
were pooled and exposed to 2 ml protein A-Sepharose
(Pharmacia Fine Chemicals). After being washed, 3M sodium thiocyanate in Buffer A was used to elute the absorbed
material. Following immediate dialysis, the solution was
further absorbed with anti-FabSepharose. Immunoelectrophoresis of a 1-mg/ml sample against horse anti-human
serum revealed only 1 band, which was identified as Fc by
goat anti-Fc antiserum (Cappel Laboratories). The yield of
highly purified Fc fragment was 16 mg, i.e., approximately
50% of the expected yield.
Preparation of mIgM rheumatoid factor. The serum
of a patient with Waldenstrom macroglobulinemia was kindly supplied by Dr. Steven Hauptman, Jefferson Medical
College, Philadelphia, PA. Serum was absorbed on IgGSepharose and released by 0.1M glycine buffer, pH 2.9. The
solution was passed through a column of Sephadex G-200
and the major fraction, which eluted in the void volume, was
found to contain IgM, when placed into microtiter wells
reacting with I2’I-anti-IgM but not with lZ5I-anti-IgG antibody. It reacted with anti-kappa but not with anti-lambda
antisera. Twelve micrograms of protein gave a titer of 1/800
in the latex fixation test.
Absorption with F(ab‘),-, Fc- and keyhole limpet
haemocyanin (KLH)- conjugates. Conjugates were prepared
at levels of 10 mg protein/l gm dry weight cyanogen bromide-activated Sepharose (Pharmacia Fine Chemicals), according to instructions supplied by the manufacturer. Over
90% binding was observed with all preparations. Sera diluted
1 :30 in Buffer A-Tween were shaken vigorously in capped
vials for 2% hours at room temperature in ratios of 50 ml of
undiluted serum to 100 Sepharose-bound protein. The suspensions were freed of immune adsorbent by filtration.
Radiolabeling. Goat anti-human Fc antibody (Cappel
Laboratories) was purified by affinity chromatography with
Fc-Sepharose. Rabbit anti-goat IgG (Cappel Laboratories)
was similarly purified with goat IgG-Sepharose. Material
eluted at 4°C with 2M sodium thiocyanate in Buffer A was
immediately dialyzed and further purified by exposure to
F(ab‘)z-Sepharose. Affinity purified goat anti-IgM (Tago,
Inc., Burlingame, CA) was additionally purified by exposure
to F(ab’)2-Sepharose. Iodination was carried out with Na’251
(Amersham Corp., Arlington Heights, IL) by the method of
Fraker and Speck (20) with IodoGen (Pierce Chemical Co.,
Rockford, IL) at 4°C for 30 minutes. Unreacted I2’I was
removed by dialysis. Specific activities varying from 0.2 to
0.3 mCi/mg protein were obtained. Eighty to 90% of the
protein had specific antibody activity, as determined by
adsorption to antigen bound to Sepharose.
Radioimmune assay for anti-F(ab’)2 antibodies. Two
hundred microliters of a F(ab’), solution, which contained 25
pg in 1 ml of Buffer A, was added to the wells of flexible Vbottom microtiter plates (Dynatech Laboratories, Inc., Alexandria, VA) and kept overnight at room temperature in a
moist atmosphere. The wells were then washed 3 times with
Buffer A-Tween, and 100 pl of serially diluted serum samples was added. The solutions were then incubated for 3
hours at room temperature. After washing the wells with
Buffer A-Tween 3 times, we added 100 pl of goat anti-human
Fc antiserum (Cappel Laboratories, Inc.) that had been
absorbed with F(ab’)*-Sepharose and diluted 1/100 in Buffer
A-Tween. After a 3-hour incubation at room temperature,
the wells were again washed 3 times with Buffer A-Tween.
IgG anti-F(ab’), antibody was detected after overnight incubation with 5 x lo4 counts per minute (cpm) of affinity
purified ‘251-rabbitanti-goat IgG antibody. IgM anti-F(ab’)z
antibody was detected after overnight incubation with 5 x
lo4 cpm of affinity purified ”’I-goat anti-human IgM antibody. The plates were washed 3 times with Buffer A-Tween,
dried, and cut with hot Nichrome wire.
Individual wells were counted in a Searle gamma
counter, Model 1285. The data for each serum sample were
plotted as cpm of antibody bound versus the log of the
reciprocal of the dilution. The titer of IgG anti-F(ab’)2
antibody was defined as that dilution which caused binding
of 10% of the initially applied radiolabeled antibody. The
titer of IgM anti-F(ab‘)*antibody was defined as the dilution
which caused binding of 4% of the initially applied radiolabeled antibodies. Titers were read directly from semilog
plots. The intraassay coefficient of variation, established by
performing 10 assays each on 3 sera, was less than 10%.
Statistical analysis. An analysis of variance was performed on the log titers of each of the patient groups used in
this study, followed by a test of the differences between pairs
of means by the method of Scheffe (21). Correlations were
determined by the method of Pearson.
Amount of serum. To determine the effect of
serum dilution o n detection of IgG and IgM anti-
presumably specific for the F(ab’), target was eluted
with 2M sodium thiocyanate in Buffer A, dialyzed, and
reconstituted to original volume with Buffer A-Tween.
These preparations had over 80% of the initial antiF(ab’), antibody activity, but they lacked activity in
plastic wells coated with 50 pg KLH or wells that
received no protein coat. In a further control experiment, sera absorbed with KLH-Sepharose lost insignificant amounts of anti-F(ab’)2 antibody activity.
Effect of mIgM RF on RIA. Direct measurement
of IgM anti-F(ab’), antibody activity is subject to
interference by IgM RF when the latter is also present
in a test serum. Potentiation of the IgM anti-F(ab’)2
antibody titer occurs because IgM RF can react with
IgG anti-F(ab’)2-F(ab’)2 complexes that adhere to the
wells of microtiter plates. To measure the magnitude
of the potentiation, we added increasing amounts of
mIgM RF to a normal serum, diluted 1/20 in Buffer
A-Tween. After the wells were washed 3 times, IgM
anti-F(ab’)2antibody titers were measured by adding 5
X lo4 cpm of 1251-anti-human IgM antibody to each
well. The linear increase of apparent IgM anti-F(ab’)2
antibody after varying amounts of mIgM RF were
added can be seen in Figure 2.
Also shown in Figure 2 is the effect of mIgM RF
on IgG anti-F(ab’)z antibody titers. In this case, when
‘2S1-anti-human IgG was added to the wells in place of
‘251-anti-human IgM antibody, a linear decline of
F(ab’), antibody, normal and RA sera were serially
diluted in Buffer A-Tween and tested by the RIA. A
relatively linear correspondence between serum dilution and IgM anti-F(ab’)2 antibody extended over a
dilution range of 1/5 to 1/160. Data for an RA serum
showing one of the highest antibody titers and for a
normal serum with the lowest antibody titer are shown
in Figure 1A. Data for IgG anti-F(ab’)z antibody for a
representative RA and normal serum are shown in
Figure 1B. A linear response was seen over the entire
range of serum dilutions tested. Titers were calculated
from similar dilution profiles on the basis of the serum
dilution required to give an arbitrary amount of binding. It is apparent in Figure 1A that a comparison of
normal and RA IgM anti-F(ab’), antibody titers requires a binding level of 2,000 cpm, or 4% of input
cpm. Most normal sera would not show a titer at a
higher binding level. Conversely, IgG anti-F(ab’)2 titers are compared most accurately when they are
calculated as the serum dilution required to bind 5,000
cpm, or 10% of input cprn (Figure 1B).
Specificity of the RIA. We exposed serum samples, 3 of high titer and 1 of low titer, to F(ab’)2Sepharose by shaking them in capped glass vessels at
room temperature for 3 hours. Approximately 1 ml of
serum was used for each milligram of bound F(ab’),.
After absorption, there was >95% reduction in both
IgM and IgG anti-F(ab’)2 antibody titer. Antibody
0 8
- -.
+ t)
Figure 1. Binding of IgM (A) and IgG (B) antibodies to F(ab’), at increasing dilutions of rheumatoid
and normal (A-A)
serum, determined by radioimmunoassay with 5 x lo4 counts per minute
(cpm) of ‘251-anti-human IgM or ’*‘I rabbit anti-goat Fc antibody.
1 3
0 3 0 . 6
Figure 2. The effects of addition of mIgM rheumatoid factor (RF) to
a 1/20 dilution of normal serum on the binding of IgM anti-F(ab’)*
(0-0) and IgG anti-F(ab’)* (0-0) antibody to F(ab’),, determined by radioimmunoassay in the absence of goat anti-human Fc
antibody with 5 x lo4 counts per minute (cpm) of ‘2SI-anti-human
IgM or ‘251-anti-human Fc antibody.
apparent IgG anti-F(ab’)2 antibody was seen. This
result was consistent with the previous experiment,
namely, that IgM R F might be expected to block the
access of 12’I-anti-human IgG to the IgG anti-F(ab’)2
antibody adhering to plastic-bound F(ab’),.
Removal of interference by mIgM RF. The following tests were done to establish that the potentiation of IgM anti-F(ab’)2 antibody titers by mIgM R F is
abolished by adding a competing high-affinity antiserum to human Fc. First, increasing amounts of 12’1mIgM R F were allowed to bind to microtiter wells
coated with human IgG. Then a 1/100 dilution of goat
anti-human Fc antiserum (5 pg specific antibody/well)
was added to the wells. After we washed the wells
again, residual I2’I-mIgM R F was counted. As can be
seen in Figure 3, when 1.8 pg of labeled R F was
applied to the wells (equivalent to an R F concentration
of 360 pg/ml of serum), 45 ng of the labeled R F bound.
However, further incubation with goat anti-human Fc
antiserum resulted in the removal of 40 ng (nearly
90%) of the bound ‘*’I-mIgM RF. Thus, goat antihuman Fc antiserum was effective in competing with
mIgM R F for human IgG.
In a second test, we determined the effect of
addition of unlabeled mIgM R F on the apparent binding levels of IgM anti-F(ab’)2 antibodies. Figure 4
shows these antibody levels with and without the
addition of goat anti-human Fc antiserum for 4 RA and
2 normal sera in the presence or absence of added
mIgM RF. In 5 of the 6 sera tested, the addition of
mIgM R F (360 pg/ml of serum) led to a variable
potentiation of the apparent binding of IgM antibody
to F(ab’)2. In each of the RA sera tested, the addition
of a M O O dilution of goat anti-human Fc antiserum not
only removed all of the potentiation attributed to
mIgM R F but also reduced the binding of apparent
IgM anti-F(ab’)2 antibody to levels lower than those
obtained when these sera were tested in the absence of
goat anti-human Fc antiserum. In the 2 normal sera
tested, goat anti-Fc did not depress binding levels
beyond those obtained in the absence of goat antihuman Fc antiserum. In the normal sera, however,
goat anti-human Fc antiserum did remove most of the
potentiation caused by the addition of mIgM RF.
These results indicate that goat anti-human Fc antiserum can remove most of the potentiation by mIgM R F
at concentrations equivalent to 360 @g/ml of serum.
IgM and IgG anti-F(ab’)2antibody titers of a test
population. Apparent IgM anti-F(ab’)z titers for 20
normal sera, 22 SLE sera, and 23 RA sera were
determined in the presence of goat anti-human Fc
251 -mlgM R F ,M I C R O G R A M S I W E 11
Figure 3. Removal of ‘251-mIgM rheumatoid factor (RF) by goat
anti-human Fc antibody. Increasing amounts of ‘251-mIgM R F were
added to microtiter wells coated with human IgG. After allowing the
labeled RF to bind, the wells were washed. Goat anti-human Fc
antiserum, diluted 1/100, was added to appropriate wells ( 5 CL&
specific antibody/well). After incubation with the goat antiserum,
the wells were washed and the residual ‘’’I-mlgM RF was counted.
antibody by RIA on the basis of the serum dilution
required to bind 4% of input cpm I2’I-anti-IgM antibody. The geometric mean titer was 1/4.2 for the
normal group, 1/6.4 for the SLE group, and 1/35 for the
RA group, with coefficients of variation of 2.0, 3.5 and
2.93, respectively. The titers of RA sera were significantly higher than those of either of the other groups
(P< 0.001), and the normal and SLE groups did not
differ significantly from each other (Figure 5A).
Apparent IgG anti-F(ab’)2 antibody titers for 20
normal sera, 23 SLE sera, and 24 RA sera were
determined by RIA in the presence of goat anti-human
Fc antibody, on the basis of the serum dilution required to bind 10% of input cpm of 1251
anti-goat Fc
antibody. The geometric mean titer was 1/7.3 for the
normal group, 1/8.1 for the SLE group, and 1/17 for the
RA group, with coefficients of variation of 1.8, 2.8,
and 1.8, respectively. There was a significant difference between normal and RA groups (P= 0.05, Figure
5B). As can be seen in Figure 6, there was a weak but
significant correlation between IgM and IgG antiF(ab’), antibody titers for individual sera (r = 0.48, P <
As can be seen in Figure 7, there was also a
weak but significant correlation between latex agglutination titers and IgM anti-F(ab’), antibody titers (r =
0.43, P = 0.05). However, no significant correlation
was observed between latex agglutination titers and
IgG anti-F(ab’)2 antibody titers (r = 0.19, P > 0.10).
initially applied at a concentration equivalent to 360
pg/ml of serum (Figure 3), which is approximately
twice the mean amount of IgM R F found in an RA
population (10). The amount of goat anti-human Fc
added, therefore, appears to be effective for sera
having high RF titers. Upon removal of the IgM RF by
goat anti-human Fc antibody, IgM anti-F(ab‘)2 antibody is evaluated in our RIA with affinity-purified
‘*’I-anti-human IgM (Figure 8D). As a consequence of
the addition of goat anti-human Fc antibody (Figure
8D), IgG anti-F(ab’)2antibody must now be measured
indirectly with ‘251-anti-goat IgG antibody.
R A 2115,120
L T = 1 / 320
We have designed an RIA for the detection of
IgM and IgG anti-F(ab’), antibodies, which react with
pooled human F(ab‘)2fragments that adhere to microtiter wells (Figure 8A). The presence of endogenous
IgM RF in a test serum causes a variable amount of
potentiation of the apparent IgM anti-F(ab‘)*antibody
titer. The proposed mechanism for this potentiation is
presented schematically in Figure 8B, which shows
the attachment of IgM RF to the IgG anti-F(ab’),
antibody that has reacted with the F(ab’), fragments.
To remove this potentiation, we interposed a second
step, in which we added excess goat anti-human Fc
antibody after the initial incubation with human test
serum. Goat anti-human Fc antibody displaces IgM
RF (Figure 8C>, but has no effect on the IgM antiF(ab’), antibody that has bound to the target antigen.
In support of this mechanism, we found that
adding 5 pg of goat anti-human Fc antibody to each
well displaced 90% of the 1251-mIgM RF that had
bound to human IgG. This labeled mIgM RF was
mlgM RF
4tNORM 1
mlgM R F
Figure 4. Reduction of apparent IgM anti-F(ab’), antibody levels
by goat anti-human Fc antibody. Four RA and 2 normal sera, diluted
1/20, with or without the addition of mIgM RF (360 pg/ml of
undiluted serum) were added to microtiter wells coated with F(ab’),.
After allowing binding of antibodies, the wells were washed and
incubated with or without goat anti-human Fc antiserum, diluted
1/100. After washing, IgM anti-F(ab’), antibodies were measured by
adding 5 x lo4 cpm of ‘251-anti-human IgM. RA = rheumatoid
arthritis; RF = rheumatoid factor. NORM = normal serum. Latex
agglutination titers (L.T.)are shown for each of the sera tested. IgG
anti-F(ab’), antibody titers of RAI-RA4 and NI and N2 sera are I /
100, 1/22, 1/17, 1/20, 1/10, and 1/21.
0 .
0 .
0 .
0 .
0 .
0 .
0 .
a - a a
Figure 5. Serum IgM (A) and IgG (B) anti-F(ab’)Zantibody titers determined by radioimmunoassay for sera from normals
(N)and patients with rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE). The titer for each serum was
defined as the serum dilution required to bind 4% [IgM anti-F(ab’)* antibody] or 10% [IgG anti-F(ab’)z antibody] of the 5 X
lo4 input counts per minute. The horizontal bars show the geometric mean for each group. The mean IgM anti-F(ab’)Z
antibody titer for RA falls 1.94 normal SD above the normal mean ( P < O.OOl), and the mean IgG anti-F(ab‘), antibody titer
for RA falls 0.82 normal SD above the normal mean ( P = 0.05).
50 -
o m
50 100
IgM ANlI-F(ab))
1 / SERUM D t L U l I b
Figure 6. IgM and IgG anti-F(ab’)Zantibody titers show a weak but
significant correlation (r = 0.48, P < 0.05) for a panel of 26
rheumatoid arthritis sera, for which both determinations were done
at the same time.
We have also tested the effect of adding unlabeled mIgM R F on the apparent IgM anti-F(ab’)2
antibody titers of sera with endogenous RF and sera
lacking them (Figure 4). The variable potentiation of
IgM anti-F(ab’)* antibody could not be related directly
to titers of endogenous RF and IgG anti-F(ab’)2antibody and probably depended additionally on total
available IgG and immune complex content of the test
sera. Goat anti-human Fc antiserum, however, removed all the potentiation caused by addition of mIgM
RF, regardless of whether the sera contained endogenous RF. The goat anti-human Fc antiserum, moreover, reduced the apparent IgM anti-F(ab’)2antibody
titers of all the RA sera below the antibody levels seen
when no mIgM RF was added. The amount of reduction of the IgM anti-F(ab% antibody levels again was
not directly related to either latex agglutination or IgG
anti-F(ab’)2 antibody titers, suggesting the involvement of a combination of factors.
We have shown that the addition of mIgM RF
to serum also causes the reduction of apparent IgG
400 200 -
- *
fied polyclonal IgG RF or mIgG RF preparations are
The major finding with our RIAs was that
individuals with classic RA had significantly more IgM
anti-F(ab’)2antibody ( P < 0.001) than individuals with
active SLE or normal volunters. Elevated IgG antiF(ab’);! antibodies were also observed in these patients,
although the difference between normal and RA sera
was far less pronounced (P = 0.05). Because it is
difficult to prepare highly purified IgM and IgG antiF(ab’)2antibodies that are uncontaminated by RF, we
currently are unable to estimate absolute amounts of
these antibodies in test sera. IgM and IgG anti-F(ab’)2
antibody titers for individual sera (Figure 6) were
significantly correlated (r = 0.48, P < 0.05), suggesting
a distribution of these antibody classes similar to those
of classic R F (10).
The possible importance of anti-F(ab’)* antibodies was discussed previously with data obtained by
an RIA (5). The test was based on polyethylene glycol
facilitated precipitation by anti-F(ab’)2 antibodies of
1251-labeled aggregates of F(ab’), prepared from
pooled human IgG. Because of the design of the test,
direct measurements of the antibody class reacting
with F(ab’)2could not be made. In that system, 72% of
the RA sera tested precipitated over 33% of the F(ab’)2
Figure 7. IgM anti-F(ab’)z antibody titers and latex agglutination
titers for a group of 22 rheumatoid arthritis sera show a weak but
significant correlation (r = 0.43, P = 0.05)
anti-F(ab’)2 antibody levels (Figure 2 ) . To date, we
have not definitively studied the correction of this
effect with the addition of goat anti-human Fc antibody. If the mechanism for the interference by IgM RF
in these measurements is as shown in Figure 8B, then
goat anti-human Fc antibody would also correct for
the underestimation of IgG anti-F(ab’)2 antibody by
removing IgM RF from the Fc region of bound IgG
antibodies. This hypothesis is not readily tested, however, since it seems inappropriate to compare IgG antiF(ab’)2 titers obtained by indirect measurements by
1251-anti-goatIgG with those obtained by direct measurement of ‘251-anti-human IgG antibody.
We do not now know whether goat anti-Fc
antibody displaces human IgG RF from some of the
complexes formed by the reaction of anti-F(ab’)2antibodies with F(ab’),. It is unlikely, though, that IgG RF
will interfere with IgM anti-F(ab’)2antibody measurements to the same extent as IgM RF. Direct studies of
the problem are contemplated as soon as highly puri-
Figure 8. The radioimmunoassay for the detection of IgM and IgG
anti-F(ab’), antibodies. Concave line = microtiter well. A, IgM and
IgG anti-F(ab’), antibodies present in test sera bind to microtiter
wells coated with pooled human F(ab’)2 fragment. B, IgM RF
present in a test serum binds to IgG anti-F(ab’), antibodies, potentiating the apparent IgM anti-F(ab’), antibody titer. C, The addition of
excess goat anti-human Fc antibody displaces the IgM RF from the
Fc region of IgG. D, Corrected binding levels of anti-F(ab’),
antibodies are now determined with lZ5I-anti-human IgM antibody
or 1251-anti-goatIgG antibody.
aggregates. In contrast, only 1 1% of the sera from SLE
patients and 3% of the normal sera were positive by
this criterion.
Our results are in agreement with the study
cited above, and it is likely that we are measuring
similar antibodies. Our RIA, however, shows that
differences in titer between RA and normal sera are
larger for IgM anti-F(ab’)2 antibody than for IgG antiF(ab’), antibody. Since Nasu et a1 ( 5 ) used an assay in
which IgM anti-F(ab’)2antibody may be more efficient
than IgG anti-F(ab’), antibody in crosslinking the
F(ab’), aggregates, it is possible that the differences
they observed between normal and RA sera also can
be attributed to IgM ar~ti-F(ab’)~
Finding IgM and IgG anti-F(ab’)2 antibodies
significantly higher in RA sera when compared with
SLE or normal sera is also of interest because it would
be desirable to compare the antibody class distributions of anti-F(ab’)2 antibody with those of RF. While
IgM RF long had been considered to be present in
greater quantity than IgG RF sera, recently reported
data suggest that the levels of both IgM and IgG R F
are significantly higher in RA than in normal sera
(8,10,22,23) and, in fact, that there is more IgG RF
than IgM RF in RA sera (10). It would appear that the
distribution patterns of anti-F(ab’)2antibodies and RF
are similar. It is also of interest that there was a weak,
although significant correlation (r = 0.43, P = 0.05)
between IgM anti-F(ab’)2 and latex fixation titers in
our RA test population. Therefore, it is possible that
anti-F(ab’)2antibodies and RF are produced by similar
antigenic stimuli, as had been previously suggested
In tests measuring antibody to antigen other
than F(ab’),, previous workers have abolished interference by R F by absorbing test sera with solidified
IgG or Fc (11-15). Absorption with Fc-Sepharose to
remove RF, however, does not work with anti-F(ab’)2
antibodies. A large portion of these antibodies probably also absorbs to the Fab sites of IgG RF that are
being absorbed on Fc-Sepharose. Since laborious absorption is no longer required, the RIA method described in this paper appears useful and practical for
assessing isotype specificities of antibodies to most
antigens. The RIA is easily performed and is designed
to guard against false-positive results when sera with
moderate or large amounts of RF are assayed.
patients with SLE, David A. Miller for technical assistance,
and Susan Hanson for preparation of the manuscript.
We wish to thank Dr. Hyman Menduke for advice on
statistics, Dr. Ralph J. DeHoratius for serum samples of
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,
2. Harboe M, Rau B, Aho K: Properties of various antigammaglobulin factors in human sera. J Exp Med
121503-5 19, 1965
3. Natvig JB: Heterogeneicity of anti-gammaglobulin factors detected by pepsin-digested human gamma-G-globulin. Acta Pathol Micobiol Scand 66:369-381, 1966
4. Waller M, Blaylock K: Further studies on the antiglobulin factors in human serum to the pepsin digested
fragment of the Ri anti-Rh antibody. J Immunol97:438443, 1966
5. Nasu H, Chia DS, Knutson DW, Barnett EV: Naturally
occurring human antibodies to the F(ab’)2 portion of
IgG. Clin Exp Immunol42:378-386, 1980
6. Milgrom F, Dubiski S, Wozniczko G: Human sera with
“anti-antibody’’ activity. Vox Sang 1: 172-183, 1956
7. Nasu H, Chia D, Iwaki Y Barnett EV, Terasaki PI:
Cytotoxicity of anti-Fab antibodies against B lymphocytes in rheumatoid arthritis. Arthritis Rheum 24: 12781282, 1981
8. Torrigiani G, Roitt IM: Antiglobulin factors in sera from
patients with rheumatoid arthritis and normal subjects.
Ann Rheum Dis 26:334-340, 1967
9. Carson DA, Lawrence S, Catalan0 MA, Vaughan JH,
Abraham G: Radioimmunoassay of IgG and IgM rheumatoid factors reacting with human IgG. J Immunol
119:295-300, 1977
10. Wernick R, LoSpalluto JJ, Fink CW, Ziff M: Serum IgG
and IgM rheumatoid factors by solid phase radioimmune
assay. Arthritis Rheum 24:1501-1511, 1981
11. Knez V, Stewart JA, Ziegler DW: Cytomegalovirus
specific IgM and IgG response in humans studied by
radioimmune assay. J Immunol 117:2006-2013, 1976
12. Meurman OH, Ziola BR: IgM class rheumatoid factor
interference in the solid phase radioimmunoassay of
rubella-specific IgM antibodies. J Clin Pathol 3 1 :483487, 1978
13. Kangro HO, Pattison JR, Heath RB: The detection of
rubella-specific IgM antibodies by radioimmunoassay.
Br J Exp Pathol 59577-583, 1978
14. Repetti CF, Gill TJ 111: Radioimrnunoassay of IgM and
IgG antitetanus toxoid antibody. J Immunol Methods
37:153-163, 1980
15. Vejtorp M: The interference of IgM rheumatoid factor in
enzyme-linked immunosorbent assays of rubella IgM
and IgG antibodies. J Virol Methods 1:1, 1980
16. Leinikki PO, Shekarchi I, Dorsett P, Sever J: Determination of virus-specific IgM antibodies by using ELISA:
elimination of false-positive results with protein ASepharose absorption and subsequent IgM antibody
assay. J Lab Clin Med 92:849-857, 1978
17. Ropes MW, Bennett GA, Cobb S, Jacox R, Jessar RA:
1958 revision of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 9: 175-176, 1958
18. Cohen AS, Reynolds WE, Franklin EC, Kulka JP,
Ropes MW, Schulman LE, Wallace SL: Preliminary
criteria for the classification of systemic lupus erythematosus. Bull Rheum Dis 21:643-648, 1971
19. Singer JM, Plotz CM: Latex fixation test: application to
the serologic diagnosis of rheumatoid arthritis. Am J
Med 21:888-892, 1956
20. Fraker PH, Speck JC Jr: Protein and cell membrane
iodinations with sparingly soluble chloroamide, 1 , 3 , 4 , 6
tetrachloro3a, 6a-diphenylglycoluril. Biochem Biophys
Res Commun 80:849-857, 1978
21. Colquhoun D: Lectures on Biostatistics. Oxford, Clarendon Press, 1971, pp 210-212
22. Pope RM, McDuffy SJ: IgG rheumatoid factor: relationship to seropositive rheumatoid arthritis and absence in
seronegative disorders. Arthritis Rheum 22:988-997,
23. Palosuo T, Milgrom F: IgG rheumatoid factor: detection
by enzyme immunoassay in rheumatoid arthritis and
normal subjects. J Immunol Methods 47:171-181, 1981
Без категории
Размер файла
750 Кб
lupus, antibodies, igm, systemic, arthritis, igg, erythematosus, anti, fab, rheumatoid
Пожаловаться на содержимое документа