896

PLATELET AGGREGATION AND AGGREGATION INHIBITION BY DIFFERENT ANTIGLOBULINS AND ANTIGLOBULIN COMPLEXES FROM SERA OF PATIENTS WITH RHEUMATOID ARTHRITIS

P. C. FINK, U. PIENING, M. FRICKE, and H. DEICHER Antiglobulin complexes were isolated from rheumatoid arthritis (RA) sera in order to investigate their ability to aggregate human platelets and to influence platelet aggregation induced by heat-aggregated IgG. Eleven of 38 seropositive RA sera showed a significant platelet aggregation (PA) titer as compared to 40 normal control sera. No correlation was observed between sheep T cell agglutination titers and P A titers of individual sera. Ten of 22 RA antiglobulin preparations containing IgC, IgM, and IgA antiglobulins and antiglobulin complexes also showed positive P A tests. Such preparations were able to inhibit P A brought about by heat-aggregated human IgG. When IgG and IgM antiglobulins were tested separately, only IgM antiglobulins showed this inhibitory effect, whereas IgC antiglobulins were inactive. Neither IgC nor IgM antiglobulins induced P A alone. Reassociated IgC antiglobulin complexes consisting of carefully prepared IgC and isolated IgC antiglobulins were able to induce PA. Platelet aggregation by RA sera was thus shown to be due to IgC From the Division of Clinical Immunology and Blood Transfusion, DeDartment of Medicine. Medical School. Hannover. Germany. Supported by the Deutsche Forschungsgemeinschaft, SFB 54, Projekt G3. Address reprint requests to Prof. Dr. med. H. Deicher, Abteilung fur klinische Immunologie und Transfusionsmedizin, Medizinische Klinik, Medizinische Hochschule, Postfach 6 10180, D-3000 Hannover 61, Germany. Submitted for publication December 1, 1978; accepted in revised form January 9, 1979. Arthritis and Rheumatism, Vol. 22, No. 8 (August 199)

antiglobulin complexes present in the sera. It can be concluded from these experiments that the composition of antiglobulin complexes in individual sera comprised of antiglobulins of different classes and IgC “antigen” is responsible for PA results in vitro. In pathophysiologic terms, the reaction of IgC antiglobulin complexes with thrombocytes in vivo may be an important step in the chain of events leading to generalized vascular damage and deposition of immune complexes in vessel walls which is thought to be responsible for the development of generalized vasculitis in RA. Activation of both human and rabbit platelets by immune complexes has been documented by a number of groups ( 1-5). Preformed antigen-antibody complexes, heat-aggregated human IgG, and IgG insolubilized with bisdiazotized benzidine, as well as IgG adsorbed to latex particles, have all been shown to induce aggregation of human platelets. This aggregation is associated with the release of vasoactive substances, nucleotides, and calcium ions (3). Certain differences regarding the reactivity of Platelets from different Species have been observed during these investigations (2). In particular, the immune complex induced damage of rabbit Platelets appears to depend On the activation of the alternate pathway of complement activation (2). For human platelets, however, Pfueser and Liischer (4) have shown that by immune complexes, aggregated IgG, and collagen occurs without

PLATELET AGGREGATION IN RA

participation of additional factors, such as complement components. Henson and Spiegelberg (5) have demonstrated that aggregated IgG, aggregated Fc fragments, and aggregated IgG myeloma proteins activate platelets, whereas monoclonal proteins of other Ig classes (IgM, IgA, IgD, IgE) as well as (Fab'), fragments did not cause platelet damage as measured by sertonin release. Circulating immune complexes of different size and composition have been detected in sera of patients with rheumatoid arthritis (RA) (6-8) as well as in sera of other chronic rheumatic disorders and chronic viral infections (9). They have also been shown deposited in vessel walls and in the perivascular synovial tissue as well as in rheumatoid synovia (10-13). In RA, the majority of these immune complexes are composed of IgG and different types of antiglobulins belonging to several Ig classes (14-17). We have therefore investigated the ability of antiglobulin complexes isolated from RA sera to aggregate human platelets as well as to influence platelet aggregation (PA) induced by heat-aggregated IgG. The results show that IgG and IgM antiglobulins have different properties regarding both effects.

MATERIALS AND METHODS Sera. Thirty-eight sera obtained from seropositive (sheep cell agglutination [SCAT] titer > 1 :64)patients with definite or classic RA according to the standards of the American Rheumatism Association (18) (mean age 45 f 15 years, sex distribution fema1e:male = 2: I). Both patient and SCAT negative control sera from healthy blood group 0 persons were kept at -70°C. They were heated at +56"C for 30 minutes and subsequently centrifuged at 1,200g at +2OoC for 20 minutes before use in PA tests. Preparation of antiglobulins. A crude globulin preparation was obtained from approximately 70 ml of decomplemented RA serum by two successive precipitations with 40% ammonium sulfate (final concentration). The precipitate was dissolved in 0.01M phosphate buffered saline (PBS), pH 7.2 (one-third of the original serum volume), and dialyzed overnight against 5 liters of 0.1M NaHCO, buffer, pH 8.2. An immunoadsorbent was then prepared with 200 mg of human IgG (obtained from Prof. Dr. H. Hoppe, Zentralinstitut fur das Blutspendewesen der Freien und Hansestadt Hamburg), heated at +63"C for 30 minutes (19) in 0.1M NaHCO, buffer, pH 8.2, and 15 grn cyanogenbromide activated Sepharose 4B (Deutsche Pharmacia, Frankfurt) as the carrier. Dialyzed crude globulin preparations were incubated with such an immunoadsorbent and rotated end-over-end for 12 hours at +20°C. The mixture was then packed on a column (1.5 x 31 cm), and unbound proteins were washed away with 0.1M NaHCO, buffer, pH 8.2, until the absorbance of the washing fluid at 254 nm was below 0.02. The bound protein representing the antiglobulin preparation was eluted from the immuno-

897

adsorbent with 0.1M sodium acetate buffer (SAB), pH 4.4, immediately dialyzed against PBS, pH 7.2, and concentrated to 0.2- 1.5 mg/ml by negative pressure dialysis ( 15). Fractionation of antiglobulin preparations. One-half milliliter samples of individual antiglobulin preparations were applied to a calibrated Sephadex G-200 column (0.9 x 150 mm) equilibrated with SAB, pH 4.4. Protein eluted with SAB from the gel was collected by using an LKB 2070 UltroRac I1 fraction collector and a Uvicord I1 UV-absorption meter. The fractionated material was immediately dialyzed overnight at +4"C against PBS, pH 7.2, and assayed for Ig contents and serologic activity as described below. Platelet aggregation (PA) technique. PA tests were carried out as described by Penttinen et a1 (20) with slight modifications. The separation of fresh human platelets was performed by differential centrifugation. Five hundred milliliters of ACD-stabilized blood from healthy donors of blood group 0 were centrifuged at 800 rprn for 20 minutes at +20"C in a plastic double bag. Platelet-rich plasma (PRP) was transferred into the second bag, acidified to pH 6.6 with ACD-A (8 m1/100 ml), and again centrifuged at 2,200-2,400 rpm for 20 minutes at +2OoC. Supernatant plasma was removed; the platelets were carefully resuspended in approximately 30 ml of autologous plasma and placed into plastic tubes. The tubes were again centrifuged at 1,200 rpm for 20 minutes at +2OoC, the supernatant plasma was sucked off, and the platelet sediment was resuspended in PBS, pH 7.2, free of Ca++and Mg++ (21). The platelet count was then adjusted to 2 X 10S/mm3by appropriate dilution. Fifty microliters of this platelet suspension were then incubated with 50 pl of antiglobulin fractions, sera, or serial dilutions thereof in siliconized microtiter wells with u-shaped bottoms at +4"C for 18-20 hours. Heat-aggregated human IgG (5 mg/ml dissolved in PBS, pH 7.2, heated for 30 minutes at +63"C) served as a positive control and buffer alone as negative control. PA was read with the naked eye by using dark ground illumination, as well as phase contrast microscopy. Immunochemical analysis. Immunoglobulin concentrations (IgG, IgA, and IgM) were measured by radial immunodiffusion technique using LC-Partigen plates (Behringwerke, Marburg/Lahn, Germany). Serologic reactions. Serologic reactivity of the preparations was assayed by using a commercial latex slide test (LFT) (Latex-Rf-Reagens, Behringwerke, Marburg/Lahn, Germany) and a conventional sensitized sheep cell agglutination technique (22). Sedimentation analysis. To obtain an estimate of the molecular size of materials contained in the antiglobulin preparations, sedimentation velocity runs were performed with a Beckman Spinco Model E ultracentrifuge equipped with a photoelectronic absorption scanner and Schlieren optical systems. Sedimentation profiles of the preparations were determined by using An-FTi and An-HTi rotors, a speed of 60,000 rpm at +20°C, and ultraviolet absorption optics at 280 nm (kindly provided by Dr. Riesner, Lehrstuhl fur physiologische Chemie 11, Hannover). Statistical analysis. PA titers were expressed as the negative logarithm of the highest dilution giving positive aggregation (e.g. 1 :32 = -log 1.5 1). Statistical analysis was performed by using Student's t test.

FINK ET AL

898

RESULTS

Table 2. Analytical ultracentrifugation profile of an isolated antiglobulin preparation (patient LB)

Immunochemical and physical analysis of antiglobulin preparations and isolated antiglobulins. Antiglobulin preparations giving positive serologic reactions for “rheumatoid factor” activity (i.e. “seropositive” preparations) were isolated from 22 RA sera by using the method described above. Immunochemical analysis revealed the presence of IgG and IgM in all, and IgA in the majority of these materials (Table 1). When the preparations were analyzed in sedimentation velocity runs, the expected heterogeneity of such preparations was detected. At pH 7.2, a typical preparation contained approximately 50% material with a sedimentation coefficient of 17.5S, about 10% with a sedimentation coefficient of 6.6S, about 25% intermediate complexes (6,23) with a sedimentation component about 1 lS, and finally approximately 14% macromolecular material with a sedimentation component above 20s (6). A representative example is shown in Table 2. At pH 4.4, the complexes present were readily dissociated, revealing two main components as constituents of the complexes with sedimentation components of 6.6s and 17.5S, representing IgG and IgM. Accordingly, separation of the components of the antiglobulin preparations at pH 4.4 on Sephadex G-200 columns gave two fractions. The first

Sedimentation coefficients (S”20, w)

Table 1. Immunoglobulin content of eluates derived from affinity chromatographedRA sera

Patients AL SI PA FO

LM KR

vs I0

KN

ss

AS RZ

xv LB HB

SM KR

IM FM KG AJ SB

kG

IgM mg/100 ml

0.52 4.9 1.8 3.6 2.0 5.1 0.23 2.6 3.4 4.8 5.0 5.3 4.6 8.2 3.7 6.0 2.8 6.3 4.1 1.2 0.7 6.9

10.7 6.8 6.8 3.8 8.9 9.7 4.9 5.4 8.7 9.5 10.4 8.2 4.9 5.2 4.3 7.9 9.2 8.9 6.2 7.6 4. I 3.4

mg/100 ml’

IgA mg/100 mlt 1.26 1.4 0.6

I .a

I .2

I .o 1.1 0.8 0.4

NT 0.2

0.8 I .2 -

NT -

0.9 1.5 1.3 -

* Determined by single radial immunodiffusion technique.

t No detection of IgA; NT = not tested.

7% of total

% of total

pH 7.2

protein content*

6.6 S (10.6 S) 17.5 S (>18.0 S)

25 51 13

I1

pH 4.4 1.5 S 6.6 S 17.5 S (>17.5 S)

protein content 6 60 32 2

* Total protein content of the antiglobulin preparation: 0.84 mg/ ml.

fraction contained IgM and gave positive reactions in LFT at +22”C, thus showing the characteristics of classic “rheumatoid factor.” The second peak consisted of IgG, was negative in latex agglutination at room temperature, and represented IgG antiglobulins and possibly some IgG “antigen” dissociated from the complexes present in the original material. An example of the fractionated IgM and IgG antiglobulins is shown in Figure 1. PA and aggregation inhibition by RA sera, antiglobulin preparations, and isolated antiglobulins. Eleven of 38 investigated RA sera induced significant PA in comparison to 40 normal control sera, i.e. titers at or above 1.50. The mean titer of the controls was 0.98 & 0.25, whereas the group of RA sera gave a mean titer of 1.25 f 0.66, the highest titers reaching 3.3 1. No correlation was found between the PA titer and the “rheumatoid factor” titer as measured by the SCAT. Similar results were obtained with the antiglobulin preparations: 10 of 22 sera induced PA with a mean titer of 0.9, the remaining 12 being negative in this test. The aggregating capacity of the heterogeneous antiglobulin preparations were thus similar to that of RA sera (Figure 2). Wager et a1 (24) have reported that cryoglobulin IgM antiglobulins inhibited IgG complex-induced PA. Twenty-one individual antiglobulin preparations were therefore tested for their ability to interfere with the PA caused by heat-aggregated IgG (5 mg/ml at 63OC for 30 minutes) at a standard antiglobulin concentration of 0.1 mg/ml. Seventeen of the preparations clearly showed inhibition of PA (Figure 3). No attempts were made to compare PA and PA inhibition capacities of individual antiglobulin preparations. The participation of IgM and IgG antiglobulins and antiglobulin complexes in PA and P A inhibition. When isolated IgM and IgG antiglobulins were tested for their ability to inhibit PA, only IgM antiglobulins were able to block PA by heat-aggregated IgG, whereas

PLATELET AGGREGATION IN RA

100

n 0

1

2

3

4

5

6

7

8

9 10 11 12 mlElution volume

Figure 1. Chromatographic separation of IgM and IgG antiglobulins on Sephadex G-200. Elution with 0.1Msodium acetate buffer, pH 4.4. Reactions of single fractions with specific IgM and IgG antisera as well as in a latex slide test are indicated in the upper part of the diagram.

IgG antiglobulins were totally inactive in this respect (Figure 4). In contrast to a proportion of the unfractionated antiglobulin preparation, neither isolated IgM nor IgG antiglobulins gave measurable PA. The concentration of fractionated IgM and IgG antiglobulins was in the range of 0.04 to 0.21 mg/ml (mean concentration of IgM antiglobulins 0.14 mg/ml, and of IgG antiglobulins 0.12 mg/ml). Because of the shortage of purified material and in view of the similar range of concentrations of the individual preparations, the separated antiglobulins were used for PA assays without further manipulation. Thus the experiments reported so far showed that fractionated antiglobulins were unable to aggregate platelets, whereas a proportion of RA sera gave positive PA, as did the unfractionated antiglobulin preparations containing antiglobulins and antiglobulin complexes. In further experiments, the ability of immune complexes reassociated from isolated IgG antiglobulins and carefully prepared IgG was studied. IgM antiglobulins were not tested because of their inhibitory capacity for PA by heat-aggregated IgG (see above). IgG at a concentration of 0.3 mg/ml was incubated with 10 different IgG antiglobulins separated as described at a mean protein concentration of 0.08 mg/ml (ranging from 0.03 to 0.21 mg/ml) for 30 minutes at room temperature. As shown in Figure 5, 7 of 10 such preparations gave positive PA with a mean PA titer of 0.42 f 0.47, the highest titer being 1.51.

899

have been shown to be inactive in this respect (5). Since RA sera contain antiglobulin complexes of different composition (6,14,15), PA in a proportion of these sera was an expected finding. Recently Torisho (27) also described PA in RA sera. In the light of some recent findings, this effect may not be just a laboratory phenomenon. As described earlier for immune complexes in synovia (28), investigations by Zubler et a1 (29) support the notion that circulating immune complexes may also be of pathogenic significance in RA. These authors found a positive correlation between the presence of circulating immune complexes as measured by the Clq binding test and elevated serum levels of C3d, pointing toward complement consumption in vivo by immune complexes present in circulating blood, indicating that an active disseminated immune complex disease associated with a generalized vasculitis (19,30), a well known

c .c

.o C

Activation and aggregation of platelets by immune complexes containing IgG antibodies and by aggregated IgG have been described by several authors (25,26), whereas aggregates consisting of IgM or IgA

3.01

m

n=22

0

6

2.71

0

m

1.51

1.20

0.90

'~

0.m

f

s m

w

DISCUSSION

antiglobulins from RA sera (0.1 mg/ml)

E

Q

~

00

00

.,-

00

0-

^ ^

-

-

0000 0000

FINK ET AL

900

complication of severe RA, is present. However, although IgG, IgM, and C3 have been found localized in vessel walls of RA (31), suggesting immune complex deposition as the cause for vascular involvement in RA, the true importance of this observation in terms of pathophysiologic events has not been established, nor has the mechanism resulting in immune complex influx into vessel walls been elucidated. The observation of PA by a fraction of seropositive RA sera containing antiglobulins and antiglobulin complexes is in agreement with earlier work (32). Ten of 22 antiglobulin preparations containing serologically active antiglobulins and immune complexes were shown to aggregate platelets. On the other hand, partial inhibition of PA induced with heat-aggre-

+

6.02

?!

n=6

5.42

.-0

m

6p -9

n=3 p

Platelet aggregation and aggregation inhibition by different antiglobulins and antiglobulin complexes from sera of patients with rheumatoid arthritis.

896 PLATELET AGGREGATION AND AGGREGATION INHIBITION BY DIFFERENT ANTIGLOBULINS AND ANTIGLOBULIN COMPLEXES FROM SERA OF PATIENTS WITH RHEUMATOID ARTHR...
645KB Sizes 0 Downloads 0 Views