close

Вход

Забыли?

вход по аккаунту

?

Studies on heterophile antibodies in rheumatoid arthritis.

код для вставкиСкачать
634
STUDIES ON HETEROPHILE
ANTIBODIES IN RHEUMATOID
ARTHRITIS
TOMOE NISHIMAKI, KYOICHI KANO, and FELIX MILGROM
Sera and synovial fluids of patients with rheumatoid arthritis were studied for the presence of heterophile
antibodies to sheep and bovine erythrocytes by means of
hemolysis in agar gel. It was demonstrated that 18 of 146
sera had hemolytic antibody titers of 160 or more; all 18
(12%) against sheep and 8 (6%) against bovine erythrocytes. Of 31 synovial fluids examined, 5 showed hemolysin
titers of 40 or more; all 5 (16%) against sheep and 3 (10%)
against bovine erythrocytes. These heterophile antibodies
were shown to belong to IgM and/or IgG class. Absorption and inhibition studies revealed that antibodies of 10
positive sera and 2 synovial fluids were of Forssman specificity and antibodies of 6 sera and 3 synovial fluids were of
Hanganutziu-Deicher specificity. Two remaining sera
were shown to contain a mixture of Forssman antibodies
and immune anti-B antibodies.
In 1949, Heller and his associates (1) noted that
titers of agglutinins against sheep red blood cells
(SRBC) increase in sera of patients with rheumatoid
arthritis (RA) at the active stage of the disease. This
From the Department of Microbiology, State University of
New York at Buffalo, School of Medicine, Buffalo, New York 142 14.
Supported by Research Grants No. IM-24B from the American Cancer Society, No. RG-892B2 from the National Multiple Sclerosis Society, and the Richard W. and Mae Stone Goode Estate.
Address reprint requests to Dr. Felix Milgrom, Department
of Microbiology, State University of New York at Buffalo, Buffalo,
NY 14214.
Submitted for publication December 19, 1977; accepted
in revised form March 21. 1978.
Arthritis and Rheumatism, Vol. 21, No. 6 (July-August 1978)
observation, however, has been largely ignored, even
though several investigators including ourselves made
similar observations (unpublished data).
Our previous studies (2) have shown that 26 of 63
sera from patients with lymphomas or leukemias contained SRBC lysins resembling Forssman (F) antibodies. Furthermore, we have recently demonstrated
that Hanganutziu-Deicher (H-D) antibodies which were
originally described as antibodies engendered by injections of foreign species sera (3-5) also appear in sera of
patients with various diseases who had never received
such injections (6).
The purpose of this study was to investigate heterophile antibodies in sera (RAS) as well as in synovial
fluids (RASF) of RA patients.
MATERIALS AND METHODS
RAS and RASF. RAS and RASF were kindly supplied
by Dr. K. Wicher of the Erie County Laboratory, Buffalo,
New York, and Dr. F. A. Green of the Veterans Administration Hospital, Buffalo, New York. Immune anti-human blood
group B serum originating from a group A volunteer D M who
received injections of B substance of horse origin was kindly
supplied by Dr. J. F. Mohn of this department. For controls,
sera of 20 healthy staff members of this department and 100
random sera of blood donors were used. All sera were heatinactivated for 30 minutes at 56OC prior to use.
Solubilized Heterophile Antigens. Solubilized heterophile H-D and Paul-Bunnell (P-B) (7) antigens were kindly
prepared by Dr. J. M. Merrick of this department following
procedures described previously (8). F antigen was also prepared by Dr. J. M. Merrick. To this end, neutral glycolipids
HETEROPHILE ANTIBODIES IN RA
isolated from guinea pig kidneys were purified by a technique
based on the procedure described in (9).
Reduction and Alkylation of RAS and RASF. To 0.5 ml
of properly diluted RAS or RASF, an equal volume of 0.1
M 2-mercaptoethanol (2-ME) was added. The mixture was
incubated for 1 hour at 37°C. Then the preparation was
dialyzed at 4°C against 0.02 M iodoacetamide solution and
against PBS, each for 24 hours.
Hemolysis in Agar Gel. This test was performed following the previously described procedure (10-12). SRBC,
bovine red blood cells (BRBC), and human blood group A or
B erythrocytes were suspended in medium 199 at a concentration of 1.25%. One volume of the cell suspension was added to
six volumes of 0.7% agar solution at 47°C and the mixture was
spread immediately o n prewarmed microscope slides. After
the agar layer solidified, 5 pl droplets of RAS or RASF were
placed on its surface. The slides were incubated in a moist
chamber for 60 minutes a t 37”C, washed and incubated again
for 60 minutes with a 1 : 10 diluted rabbit serum as a source of
complement. T h e slides were then washed, air dried, and fixed
with ethanol. Antibody titer was expressed as the reciprocal of
the highest dilution of the tested specimens that gave a definite
hemolysis.
Absorption Experiments. Washed sediments of homogenized guinea pig kidneys, SRBC, BRBC, or human blood
group A or B erythrocytes were used for absorption. The tissue
sediment or packed erythrocytes were mixed with an equal
volume of appropriately diluted RAS or RASF. The mixtures
were incubated for 30 minutes a t 20”C, centrifuged at 4000g
for 5 minutes and the supernatant was recovered. The procedure was repeated three times using fresh absorbing materials.
Inhibition of Hemolysis. To 5Opl of a RAS containing
eight units of hemolysin against SRBC and/or BRBC was
added an equal volume of the solubilized heterophile antigen
at various concentrations. The mixture was incubated at 20°C
for 60 minutes and tested for hemolysis in agar gel.
635
examined, 5 (16%) had titers of 40 or more for SRBC
and 3 (10%) for BRBC. The 5 “SRBC positive” RASF
included all 3 “BRBC positive” RASF.
To determine specificities of SRBC lysins in RAS
and RASF, absorption experiments were carried out. At
first, 10 RAS and 2 RASF were tested that gave positive
lytic tests with SRBC but not with BRBC (Table I ) .
Absorption of all these specimens with guinea pig kidney reduced their titers for SRBC to less than 40. In
contrast, absorption with BRBC had no significant effect on the titers. It appeared, therefore, that SRBC lysins
under study were F antibodies. Seven RAS and the two
RASF were treated with 2-ME. Such treatment abolished hemolytic activities of 3 RAS and 1 RASF.
Table 2 presents results obtained in absoprtion
experiments on 6 RAS and 3 RASF which were positive
against both SRBC and BRBC. As seen in the table,
absorption of these specimens with SRBC, BRBC, and
guinea pig kidney abolished their activities against both
SRBC and BRBC. These results would indicate that
lytic antibodies of this group were of H-D specificity.
Treatment of 4 RAS and 3 RASF with 2-ME reduced
lytic titers for SRBC to less than 40 in 2 sera and all 3
RASF. Titers of BRBC lysins of these 4 RAS, however,
were not much affected by 2-ME treatment, whereas
lytic titers for BRBC were reduced to less than 40 in all 3
RASF treated.
Table 1. Heterophile Antibodies in Rheumatoid Arthritis Sera ( R A S )and
Synouial Fluids (RASF). Group I
RESULTS
To establish “baseline” titers of normal human
sera in hemolysis in agar gel, sera of 20 healthy staff
members of this department and 100 random blood
donors were examined. Prior to testing, sera were absorbed with a mixture of human blood group A and B
erythrocytes. Of these sera, 20 had titers of 80 in reactions with SRBC; the titers of the remaining sera were 40
or less. Only four sera showed a lytic titer of 40 in
reactions with BRBC; the remaining sera had titers of 20
or less. On the basis of these results, as well as results of
our previous studies (6,12), sera with hemolytic titers of
160 or more were considered to be positive.
Subsequently, RAS and RASF were titrated
against SRBC and BRBC by means of hemolysis in agar
gel. It was found that of 146 RAS, 18 (12%) had lytic
titers of 160 o r more for SRBC and 8 (6%) had such
titers for BRBC. The 18 “SRBC positive” sera included
all 8 “BRBC positive” sera. Of 31 RASF specimens
Lytic Titers Against
RAS no. 161
130
22
19
144
52
141
54
31
147
RASF no. 29
8
A
A’
2560
2560
1280
1280
640
320
320
320
160
160
640
320
<40
1280
<40
I60
80
80
<40
NDt
ND
ND
80
<40
SRBC*
B
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
C
D
I280
2560
640
640
320
160
160
80
40
40
640
160
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
BRBC*
A
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
* A =- Unabsorbed specimens; A’ = unabsorbed specimens treated
with 2-mercaptoethanol; B = specimens absorbed with SRBC;
C = specimens absorbed with BRBC; D = specimens absorbed with
guinea pig kidney.
t Not done.
‘
NISHIMAKI ET AL
636
Table 2. Hererophile Antibodies in Rheumatoid Arthritis Sera ( R A S )and Synovial Fluids (RASF). Group 11
Lytic Titers Against
SRBC*
A
RAS no. 103
143
153
I04
I32
16
RASFno. 6
18
16
320
320
160
160
160
160
320
160
40
A'
80
80
<40
NDt
<40
ND
<40
<40
<40
BRBC*
B
C
D
A
A'
B
C
D
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
1280
1280
1280
640
320
320
ND
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
640
320
160
2560
640
80
160
ND
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
*A
= Unabsorbed specimens; A' = unabsorbed specimens treated with 2-mercaptoethanol; B = specimens absorbed with SRBC; C = specimens absorbed with BRBC; D = specimens absorbed with guinea
pig kidney.
t Not done.
Possible potentiation of hemolytic activities of
the heterophile antibodies by rheumatoid factor (RF)
was considered, but it was excluded in the present study.
As shown in Table 2, the treatment of some RAS such as
RAS 103 and 132 with 2-ME had very little effect on
titers of BRBC lysins, whereas their R F activity was
abolished almost completely. Furthermore, an experiment was performed in which RAS with potent R F but
no heterophile antibodies was added to nonrheumatoid
sera containing IgG lysins for SRBC and/or BRBC.
These were normal human sera, pathological nonrheumatoid sera, and sera of human volunteers immunized with SRBC. In none of these experiments did the
hemolytic titers of the sera tested increase and in several
instances decrease of the titers was noted.
Absorption studies by use of human erythrocytes
were also performed. It was found that F antibodies of
the first group and H-D antibodies of the second group
were not removed by either blood group A or B erythrocytes (not shown in the tables).
To ascertain the F nature of SRBC lysins of the
first group and the H-D nature of lytic antibodies of the
second group, inhibition tests with solubilized heterophile antigens were performed (Table 3). For these
experiments two sera, RAS 130 of the first group and
RAS 103 of the second group, were selected. The lytic
activity of RAS 130 against SRBC was completely inhibited by F antigen, but it was not affected by H-D or
P-B antigen. In contrast, lytic activity of the RAS 103
against BRBC was inhibited completely by H-D antigen
but not by P-B or F antigens.
The two remaining positive sera, RAS 123 and
RAS 155, which were positive with both SRBC and
BRBC showed a reaction pattern different from the one
encountered in the previously discussed two groups of
RAS and RASF. Absorption of these sera with SRBC
or guinea pig kidney abolished their activity against
SRBC and BRBC, whereas absorption with BRBC removed only BRBC but not SRBC lysins from these sera.
It appeared, therefore, that these two sera contained
both F and H-D antibodies. It was, however, shown
Table 3. Inhibirion of Hemolytic Acriuities of Rheumatoid Arthritis
Sera ( R A S ) by Solubilized Heterophile Antigens
Reactions of RAS 130
with SRBC
after Incubation with
Dilution
of
Inhibitor* H - D t
1 to:
I
4
16
64
256
1024
4096
Saline
++
++
++
++
++
++
++
Reactions of RAS 103
with BRBC
after Incubation with
P-B$
Fg
H-D
P-B
F
++
++
++
++
++
++
++
++
-
-
++
++
++
++
++
++
++
++
++
++
++
++
++
++
++
-
* Original concentration I mg/ml.
t Hanganutziu-Deicher antigen.
$ Paul-Bunnell antigen.
9 Forssman antigen.
-
-
-
+
637
HETEROPHILE ANTIBODIES IN RA
subsequently that absorption of these two RAS with
human B but not A erythrocytes removed BRBC lysins
almost completely, indicating that BRBC lysins of these
two sera might be primarily directed against human B
antigen. RAS 123 came from a group A patient and
RAS 155 from a group 0 patient.
In further experiments, RAS 123 and 155, along
with an immune anti-B serum, were studied for their
activity against SRBC, BRBC, and human B erythrocytes. Results of these studies are presented in Table 4 in
which for brevity sake, only RAS 123 was included
(RAS 155 gave almost identical results). As seen in the
table, SRBC lysins of RAS 123 could be absorbed with
SRBC or with guinea pig kidney but not with BRBC,
human A or human B erythrocytes; weak SRBC lysins
in the anti-B serum showed a similar absorption pattern.
This indicated that SRBC lysins of these sera were most
likely F antibodies. BRBC lysins of RAS 123 and the
anti-B serum could be absorbed with SRBC, BRBC,
guinea pig kidney, or human B erythrocytes but not with
A erythrocytes. Group B lysins of these sera could be
absorbed with BRBC, guinea pig kidney, or human B
erythrocytes. Absorption of these sera with SRBC reduced titers of group B lysins but failed to remove them
completely. These results indicated that BRBC lysins of
the two RAS tested were antibodies to human group B
erythrocytes.
DISCUSSION
The results of the present studies confirmed the
original observation of Heller et al. ( I ) , by showing that
a significant proportion of RAS and RASF contained
SRBC lysins of high titers. It was also shown that some
of these RAS and RASF contained BRBC lysins.
Table 4. Comparison of Hemolytic Activities of a Rheumatoid Arthritis
Serum #I23 ( R A S ) and an Imrnune A n t i 4 Serum D M
I
Lytic Titers Against
SR BC
BRBC
Human BRBC
Serum
RAS
DM
RAS
DM
RAS
DM
Unabsorbed
Absorbed with:
Sheep RBC
Bovine RBC
Guineapig RBC
Human"A"RBC
Human"B"RBC
>640
128
160
128
160
32
<40
320
<40
320
>640
<4
64
<4
<40
<40
<40
160
<40
<4
<4
<4
64
<4
40
<40
<40
160
<40
4
<4
<4
32
<4
128
64
The specificities of these heterophile antibodies
were determined in 18 positive RAS and 5 positive
RASF by absorption and inhibition studies. SRBC lysins in 10 RAS and 2 RASF were shown to be F antibodies. These antibodies did not act upon BRBC, they
could be absorbed by guinea pig kidney but not by
BRBC, and they were inhibited completely by solubilized F antigen originating from guinea pig tissues (1315). Heterophile antibodies reacting with both SRBC
and BRBC found in 6 RAS and 3 RASF were proved to
be H-D antibodies. Evidence was presented that their
activity could be abolished by absorption with SRBC,
BRBC, and guinea pig kidney and inhibited by solubilized H-D antigen but not by F or P-B antigen (3,4,16).
The titers of these heterophile antibodies were
reduced in most instances after treatment with 2-ME
and in some instances the antibodies were abolished
completely. This would indicate that the heterophile
antibodies under investigation belonged to IgM and/or
IgG class.
In addition to above mentioned RAS and RASF,
we encountered two exceptional RAS with SRBC and
BRBC lysins. Absorption studies on these sera demonstrated that they contained a mixture of F antibodies
and antibodies to human blood group B erythrocytes.
Whereas sheep hemolysins of these sera could be absorbed by guinea pig kidney but not BRBC or human
erythrocytes, BRBC lysins could be absorbed by SRBC,
guinea pig kidney, and human erythrocytes of group B
but not A. The fact that BRBC lysins of these RAS and
immune anti-B sera could be absorbed by SRBC while
their group B lysins could not be abolished completely
by SRBC (Table 4) may be explained by assuming that
anti-B antibodies of these sera consisted of two different
types. The one type, presumably immune antibodies,
was directed against antigens shared not only by BRBC
and guinea pig tissues but also by SRBC whereas corresponding antigens for the other type, presumably natural antibodies, were not shared by SRBC (l7,18).
The antigenic stimuli responsible for formation
of heterophile, F and H-D antibodies by RA patients,
still remain to be determined. Since some bacteria such
as certain strains of shigellae and pneumococci are
known to possess F or F-like antigen (19), it is possible
that the F antibodies demonstrated in RA patients
might have been engendered by bacterial infections in
the course of clinical or subclinical infection (20).
To our knowledge, H-D antigen has not been
shown on microorganisms. As mentioned previously
( 6 ) , it is tempting t o speculate that this antigen may
appear as a neoantigen in the pathologic tissues as a
638
NISHIMAKI ET AL
result of derepression of s o m e enzymes t h a t may be
inactive under physiologic conditions.
F u t h e r studies will h a v e to establish the possible
role of these heterophile antibodies in pathogenesis of
RA. F o r m a t i o n of circulating immune complexes by
these antibodies is currently under investigation.
ACKNOWLEDGMENTS
The authors express their appreciation to Margaret
Grybel, Susan Heubusch, and Mary H u for their excellent
technical assistance.
REFERENCES
1. Heller G. Jacobson AS, Kolodny MH: A modification of
2.
3.
4.
5.
6.
7.
the hemagglutination test for rheumatoid arthritis. Proc
SOCExp Biol Med 72316-323, 1949
Milgrom F, Kano K, Fjelde A, Bloom ML: Heterophile
antigens and antibodies in transplantation and tumors.
Transpl Proc 7901-207, 1975
Hanganutziu M: HCmagglutinines hCttrogtnCtiques aprts
injection de serum de cheval. C R Seanc SOCbiol91: 14571459, 1929
Deicher H: o b e r die Erzeugung heterospezifischer Haemagglutinine durch lnjektion artfremden Serums. Z Hyg
1061561-579, 1926
Pirofsky B, Ramirez-Mateos JC, August A: “Foreign
serum” heterophile antibodies in patients receiving antithymocyte antisera. Blood 42385-392, I973
Kasukawa R, Kano K, Bloom ML, Milgrom F: Heterophile antibodies in pathologic human sera resembling
antibodies stimulated by foreign species sera. Clin Exp
Immunol 25:122-132, 1976
Paul JR, Bunnell WW: Presence of heterophile antibodies
in infectious mononucleosis. Amer J Med Sci 183:90-104,
1932
8. Merrick J M , Schifferle R, Zadarlik K, Kano K, Milgrom
F: Isolation and partial characterization of the heterophile
antigen of infectious mononucleosis from bovine erythrocytes. J Supramolec Struc 6:275-290, 1977
9. Saito T , Hakomori S: Quantitative isolation of total glycosphingolipids from animal cells. J Lipid Res 12:257259, 1971
10. Milgrom F, Loza U: Complement-mediated hemolysis in
agar gel. J I’mmunol 98:415-423, 1966
11. Juji T, Kano K, Milgrom F: Antibody induced lysis of
nucleated cells in agar gel. Intern Arch Allergy Appl Immunol 41:739-753, 1971
12. Malave I, Kano K, Milgrom F: Reactions of murine lymphoma cells with infectious mononucleosis sera. J Immuno1 110:439-443, 1973
13. Schiff F: Heterogenic hemagglutinins in man following
therapeutic injections of immune sera produced in rabbits.
J Immunol 33:305-313, 1937
14. Flory LL: Relationship between anti-A and anti-Forssman. Immunology 14:787-798, 1968
15. Alving C R , Joseph KC, Wistar T: Influence of membrane
composition on the interaction of a human monoclonal
“anti-Forssman” immunoglobulin with liposomes. Biochemistry 13:4818-4824, 1974
16. Stuart CA, Griffin AM, Wheeler KM, Batty S: A thermostable antigen in beef-cells. Proc SOC Exp Biol (NY)
3 4 2 1 2-2 15, 1936
17. Friedenreich V, With S: Uber B-Antigen und B-Antikorper bei Menscheo und Tieren. Z ImmunForsch
78:152-172, 1933
18. Joysey VC: The relation between animal and human blood
groups. Brit Med Bull 15:158-164, 1959
19. Springer GF: Blood-group and Forssman antigenic determinants shared between microbes and mammalian
cells, Progress in Allergy. Vol 15. Basel, Karger, 1971, pp
9-77
20. Christian CL: Rheumatoid arthritis: etiological considerations. Arthritis Rheum 7:455-466, 1964
Документ
Категория
Без категории
Просмотров
1
Размер файла
415 Кб
Теги
antibodies, heterophile, arthritis, studies, rheumatoid
1/--страниц
Пожаловаться на содержимое документа