Total hemolytic complement CH50 and second component of complement C В╨Ж2hu activity in serum and synovial fluid.код для вставкиСкачать
Total Hemolytic Complement (CH,,) and Second Component of Complement (C’2hu)Activity in Serum and Synovial Fluid By GEORGE FOSTIROPOULOS, K. FRANK AUSTENAND KURTJ. BLOCH The presence of the second component of human complement in synovial fluid was demonstrated by a stoichiometric titration using EAC’lagp4ap cells; fourth component and a moiety of third component were identified by immunoelectrophoresis using specific antiserum. The concentration of total hemolytic complement, C’2, and C‘4 was strikingly diminished in synovial fluid from patients with rheumatoid arthritis compared to their concentration in synovial fluid from patients with traumatic, degenerative or gouty effusions. These findings suggest that reduction in total complement activity was due to activation of C’1 with subsequent utilization of its substrates, C’2 and C’4. Le presentia del secunde componente de complemento human in liquido synovial esseva demonstrate per un titration stoichiometric con le us0 de cellulas EAC’lagp,4gp. Le quarte componente e un parte del tertie componente esseva identificate per immunoelectrophorese con le us0 de antisero specific. Le concentration de complemento hemolytic total, de C’2, e de C’4 esseva reducite frappantemente in le liquido synovial de patientes con arthritis rheumatoide in comparation con lor concentration in liquido synovial ab pitientes con effusiones traumatic, degeneratori, o guttose. Iste constatationes suggere que reductiones in le activitate de complemento total esseva le effect0 de un activation de C’l con le utilisation subsequente de su substratos C’2 e C’4. c OMPLEMENT ( C ’ } is required for immune cytoto,xic injury such as immune hemolysis,lg2 the serum bactericidal reaction,.? or ‘the lysis of certain tumor celk4s5 The specificity of the cytotoxic reaction is determined by the antibody involved. Complement refers to a group of nonspecific serum factors (complement components) which interact in a certain sequence so as to destroy cells previously sensitized by antibody. Electron microscopic data indicate that the final step in the cytotoxic reaction is the production of holes in the cell membrane.6 Complement may also participate in other immunologic phenomena such as phagocytosis,7,8 the Arthus reaction,9J0 and experimental nephritis.ll Available data suggest that phagacytosis and the Arthus reaction may not require the entire complement sequence involved in cell lysis.’3-’4 Evidence for the participation of complement in pathogenesis of human disease has usually been obtained by studying changes in serum complement From the Department of Medicine, Harzjard Madical School and the Medical Services, Massachusatts General Hospital, Boston, Massachusetts. This is publication No. 386 of the Robert W. Looett Memorial Group for the Study of Diseases Causing Deformities, Harvard Medical School, at the Massachusetts General Hospital. Supported b y Grants A-3564 and AM-04502 from the National Institute of Arthritis and Metabolic Diseases, National lnstitutes of Health and by Grant Al-04536 from the National Institutes of Allergy and Infectious Diseases. 219 ARTHRITIS AND RHEUMATISM, VOL. 8, No. 2 (APRIL),1965 220 FOSTIROPOULOS, AUSTEN, BLOCH activity. The cytotoxic activity of whole serum may be conveniently assayed using sensitized sheep erythrocytes. The sequence in which the components of guinea pig complement act in immune hemolysis has been r e v i e ~ e d , ~ J ~ , ~ ~ and will be briefly considered only to introdrrce the necessary terminology and to stress the theoretical shortcoming of the whole complement titer. Following the union of antibody ( A ) with certain antigenic sites ( S ) on the red cell ( E ) , these sensitized receptor sites react with the complement system in a definite sequence. The interaction of the first complement component with the sensitized red cell (EA) converts the first component from a precursor (C’lp) to an active form (C’la).Ii The interaction with the fourth complement component (C’4) follows after the firstls forming the EAC’la,4 cell. The EAC’la,4 cell next reacts with the second complement component ( C 2 ) . As a result of the enzymatic action of cell bound C‘la on C’21e320an inactive fragment (C’Zi) appears in the fluid phase and an active fragment (C’Za) is bound to the cell (EAC‘la,4,2a). Once the cell achieves the EAC’la,4,2a state it can either interact with C’3 factors to yield an irreversibly damaged cell or it can decay back to the EAC’la,4 state.21J2 The C’2a lost during the decay reaction cannot be recovered as a hemolytically active component, however the EAC’la,4,2a state can be regenerated by utilizing additional C’2. In this manner an antigen-antibody complex can deplete the serum of C’2 without the participation of C‘3 factors. Unless the C’2 titer becomes limiting the reduction in concentration of this component will not be adequately reflected in the whole complement titer ( C‘H50).23,24 The level of whole complement activity has been shown to be reduced in systemic lupus erythematosus2b and one of the C’3 factors, PlC globulin has been identified at the site of the renal lesion in that In rheumatoid arthritis the level of whole serum complement is normal or slightly whereas the titer in rheumatoid joint effusions is strikingly diminished compared to the concentration in synovial fluid of gouty or osteoarthritic joint^.^^,^^ In order to further explore the mechanism of this apparent intraarticular complement depletion in rheumatoid arthritis, measurement of whole hemolytic complement in blood and synovial fluid was coupled with assays for certain components, C’2, C’4 (PlE globulin) and C’3a ( P l C globulin). The demonstration of intra-articular depletion of C’2 and C’4 without a reduction in the concentration of these components in serum, supports the contention that in certain articular disease there is local activation of the complement system. MATERIAL AND METHODS Serum and synovial fluid was obtained from 70 patients. Fifty-six patients were included in this study after careful examination and follow-up permitted a definite diagnosis of their disease. Clinical details of each patient group are presented in the appropriate .section under Results. Synovial fluid was obtained by needle aspiration of joint effusions; in most instances fluid was obtained from the knee joint. Nearly all traumatic and osteoarthritic fluids were obtained by needle puncture of the joint capsule exposed during surgery. Hemorrhagic fluids were not tested. Synovial fluids were kept at room temperature for 30-60 minutes followed by centrifugation at 3000 r.p.m. for 20 minutes, at 0”. Aliquots of the super- 221 CH50 AND C‘2”‘’ ACTIVITY IN SERUM natant fluid were either tested immediately or stored at -70”. Viscous fluids were centrifuged at 12,000 to 20,000g in order to sediment most of the mucin. In general, venous blood was obtained from the patients coincident with the joint aspiration and processed in the same manner as the synovial fluid. Blood for control values was obtained from 20 healthy volunteers. M e w r e m e n t of the tolaole complement titer (C’H5&. Erythrocytes in Alsever’s buffer were supplied either by Probio Inc., Nyack, N. Y. or Baltimore Biological Laboratoria (BBL), and commercial hemolysin was obtained from BBL. Complement titration was performed according to the macro method described by Mayer,ao results are expressed in terms of 50 per cent hemolytic units (C’H,,). Calcium was omitted from the NaC1-Veronal-gelatin (0.1 per cent) buffer used in this procedure.25 Titrations of hemolytic complement performed with serum initially diluted 1:12(E or 1:160 yielded at least four tubes in which 20 per cent to 80 per cent of sensitized cells were lysed. In order to obtain similar results with synovial fluid, it was often necessary to test several dihitions. If a synovial fluid initially diluted 1 : l O failed to yield significant lysis, then the sample is reported as containing fewer than 2 C’H,, units per ml. The serum C’H,, titer of 20 healthy volunteers ranged from 32 to 44 units per ml. iMemrement of the Second Component of Human Complement (C’2kll). The second component of human complement in serum or synovial fluid was assayed by the two step procedure of Austen and Beer.23 Human serum or synovial fluid and sensitized sheep erythrocytes in the EAC’lagp,4gp state werp incubated at 30” for tmax time. The term tmax refers to the time at which the greatest number of sensitized receptor sites are in the SAC‘lag~’,4gp,2hl1 state. Lysis of these cells was accomplished by adding guinea pig serum treated with ethylenediaininetetraacetate as a source of C‘3 factors. The conversion of SAC’lagg,4-~sites to SAC’lagD,4 gp,2’ll1 sites is stoichometrically related to the relative conZentration of C’2hll in the reaction mixture. The reciprocal of the sernm or synovial fluid dilution which produces one SAC’lagP.4 gP,211L1 site per cells is arbitrarily defined as the number of C ‘ 2 h ~units per ml. of serum or synovial fluid. The titration is linear even with C‘2’1u titcrs as low as 1 per cent of normal or as high as 200 per cent of normal.23 Identical results were obtained in this assay with duplicate serum samples and the C‘2hu titers of serum aliquots assayed on five consecutive days v‘aried by less than 10 per cent. On the other hand, there was appreciable variation in the absolute C’211u titer of a standard reference serum meawied with different lots of EAC’lagp,4gp cells.23 For this reason a standard reference serum was included with each C’211“ assay. The absolute value for each unknown serum or synovial fluid was corrected for any deviation of the reference Serum from its assigned value of 650 C‘PUunits per ml. The coirected C’2”u titer of 20 healthy volunteers ranged from 350 to 670 (mean = 510) units per ml. Serum samples used in the c‘2111‘ assay were generally diluted in two-fold steps from 1:200 to 1:3200; synovial fluid was diluted 1:50 through 1:1600. Because of the limited quantity available, lesser dilutions of synovial fluid were not used. Samples failing to give significant lysis at 1:50 were reported as containing fewer than 5 C’2h~unit3 per ml. synovial fluid. Detection of p l E (C‘4) and p l C Globulin (C’3a) by Immunaelectrophoresis. Specific rabbit antiserum to B1E and p l C globulin was kindly supplied by Dr. Hans Muller-Eberhard. Both the agar gel and the Verona1 electrophoresis buffer contained 0.01M disodium ethylenediaminetetraacetate to limit activation of the first component of complement during iminunoelectrophoretic analysis. The latter was performed as described by Scheidegger.31 Additional Procedures. The latex fixation test for rhenmatoid factor was carried out on serum and synovial fluid by the method of Singer and Pl0t2.32 Estimation of the leukocyte concentration in synovial fluid and the mucin clot test were performed as described by Ropes and Bauer.3s Total protein in synovial fluid was determined by biuret reaction. RESULTS Trneiniatic Arthritis. Paired synovial fluid and serum specimens were obtained from five male patients with traumatic arthritis (table 1).In cases E. G. 222 FOSTIROPOULOS, AUSTEN, BLOCH Table 1.-Total Hemolytic Complement and Second Component of Complement Activity in Serum and Synovial Fluid of Patients with Traumatic Arthritis and Degenerative Joint Disease Synovial Fluid Serum ~ No. Pt. 1 2 3 4 5 M. L.* J. W.* E.G. J. C.* S.K. Duration of Effusion 5 wks. 8 wku. 6 days 4 wk s . 1 Vk. Protein Mnein 9% Type 5.8 3.8 4.75 5.5 4.2 WBC per mm.? L.F.T. Xecip. C’Ha titer u/ml. Traumntic Arthritis 900 0 1 0 1 640 0 I 250 1 340 0 1 1350 0 ~ L.F.T. Recip. titer C‘2 u/ml. C’Hso u/ml. C’2 u/ml. 330 244 314 184 540 344 545 640 N.D. 0 0 0 0 0 810 384 980 660 606 574 475 0 0 0 0 0 0 0 27 29 30 33 36 N.D. 41 43 37 40 45 21 22 23 24 36 37 38 358 232 360 386 214 249 300 42 38 40 40 44 3R 53 Dcgeneratiue .Joint Disease 1 2 3 4 5 6 7 A.B.* C. D.* P.R.* A.D.* O.M. D.F. F.G. lyear 1 year 6 mo. 5mo. 5 mo. 3 mo. 1 mo. 3.8 3.2 3.0 4.5 3.8 3.7 4.6 1 1 1 1 1 1 2 2000 560 580 2100 460 520 350 0 0 0 0 0 0 0 *Synovial fluid obtained during joint surgery; serum obtained one or two days later. = not done. N.D. and S. K. joint fluid was obtained by aspiration within one week of the OCcurrence of trauma to the knee joint. In the other 3 cases, fluid was obtained during surgical excision of ruptured menisci. There was no evidence of current systemic illness or of an antecedent joint disease in these patients. Joint fluids contained relatively few leukocytes, were highly viscous and had a good mucin content. The latex fixation test was negative on serum and synovial fluid. Sera of these patients contained total hemolytic C’ activity ranging from 3745 units per ml. and C’2 activity ranging from 344-640 units per ml. Total hemolytic complement activity of synovial fluid ranged from 27-36 units per ml. and second component activity ranged from 184 to 330 units per ml. Degenerative Joint Disease. The six women and one man in this group had characteristic osteoarthritis involving either hip or knee joints (table 1 ) . These patients had no evidence of an active systemic disease and their erythrocyte sedimentation rate was normal. Synovial fluid characteristics were similar to those of the preceding group. The latex fixation test was negative in serum and synovial fluid. With one exception, patient F. G., the number of C’H50units per ml. of serum fell within the normal range. Two patients, A. B. and P. R., had an increased number of C’2 units per ml. of serum. In both cases, serum was obtained within one or two days of joint surgery and the elevated C‘2 levels probably reflect the acute effect of surgery on serum ~ o m p l e m e n t .In ~~ the patients with osteoarthritis, synovial fluid CH50values ranged from 21 to 38 units per ml. and C’2 from 214 to 386 units per ml. Gouty Arthritis. Serum and synovial fluid was obtained from six male patients within one week of the onset of an acute attack of gout (table 2 ) . These synovial fluids had a high concentration of leukocytes and a poor mucin content. All patients had an elevated concentration of serum uric acid and polariscopic examination of synovial fluid revealed the presence of sodium ~ AND 5 ~ ~0 ’ ACTIVITY 2 ~ IN~ SERUM ~ 223 Table 2.-Total Hemolytic Complement and Second Component of Complement Activity in Serum and Synovial Fluid of Patients with Acute Gout and Reiter’s Syndrome Synovial Fluid Duration of No. Pt. Effusion Protein Mucin g% Type WBC per mm.3 __ L.F.T. Recip. C~HW, C’2 titer u/ml. u/ml. C‘Hs u/ml. Serum 0 2 u/ml. L.F.T. Recip. titer 300 720 650 0 0 0 Acute Gout H.F. 1 wk. 1 wk. 10 days I 2 3 4 5 6 A.D. H.P. F. B. F. M. Zdays 6.4 10 days 1 wk. 6.46 5.45 4 4 3 3 3 3 1 2 T.M. D. C. 15 days 10 days 5.2 6.0 3 2 s. s. 6.75 4.6 3.7 30300 4000 3850 36000 45700 20050 0 0 0 0 0 21 25 26 29 137 277 290 35 45 N.D. 350 440 55 51 48 45 66 90 39 45 349 325 50 52 N.D. 0 928 1300 0 1100 470 0 0 Reiter’s Syndrome 8000 8050 0 0 urate crystals in phagocytes. In almost every patient in this group, the number of C’H50 and C’2 units per ml. of serum exceeded the normal range. The synovial fluid CHoo,2 1 4 5 units per ml., and C‘2 activity, 137440 units per ml., resembled that of the preceding two groups. Reiter’s Syndrome. In two young adults manifesting the classical triad of Reiter’s syndrome, accompanied by recent joint effusions, serum total hemolytic complement activity again exceeded the normal range (table 2). The C’H50 and C’2 activity of synovial fluid in these two patients was less than that of serum and resembled the values obtained with synovial fluid in the preceding groups. Rheumatoid Arthritis. The seventeen women and seven men included in this group had classical or definite rheumatoid arthritis as defined by A.R.A. criteria. All had active disease and joint effusions studied had been present from two weeks to several months. These patients were all receiving moderate to large doses of salicylates; in addition, cases E. T., I. M., and S. S. were receiving small doses of corticosteroids. One patient was receiving phenylbutazone (C.A.) and another (A.F.) chrysotherapy in addition to salicylates. None of the patients had received intra-articular injections of corticosteroids for several months prior to the joint aspiration. The titer of whole serum complement (table 3 ) was within the normal range in 20 patients, slightly diminished in two, and significantly reduced to 16 C’HC0units in one instance, patient F. T. The C’2 titer of serum was within the normal range in 16 patients, elevated in two, and significantly reduced in four, the titers in patients F. T., P. W., A. F., and E. C. being 7.2, 156, 226, and 250 units ml. respectively. The elevation in serum C’2 titer in patient E. T. is very likely related to the presence of a urinary tract infection,24 while the increased titer in the serum of patient J. M. is probably due to recent surgical trauma.34 The striking reduction in serum whole complement and C’2 in patient F. T. may be attributable to a m y l o i d o ~ i s ,or~ ~to the presence of considerable serum cryoprot e i n ~ The . ~ ~ other three patients with moderately reduced serum C’2 titers 224 FOSTIROPOULOS, AUSTEN, BLOCH Table 3.-Total Hemolytic Complement and Second Component of Complement Activity in Serum and Synovial Fluid of Patients with Rheumatoid Arthritis --.____ No. Pt. 1 P. W. J. A. E. C. 2 3 4 G fi 7 8 !I 10 11 12 13 14 15 16 17 18 19 20 21 22 23 G. v. E. T. F. T. C . G. G. A. I. M. S. D. M.N. w. P. M. S. M. G . 0 . c. M.R. A. F. P. F. s. w. L. A. J.M.’ s. s. C. A . Duration of Effusion Protein sr/o Muein Type 6mo. 3 mo. 2 mo. 5 mo. 5 mo. 2 mo. 2 wks. 3 mo. 3 mo. 3 mo. 2mo. 2 mo. 6 mo. 3 mo. 2 mo. 12mo. 3 mo. 4 mo. 3 mo. 3 mo. 3mo. 4 mo. 4 mo. 5.0 6.3 5.7 3.8 3.8 3.3 2.4 4.4 3.2 2.6 3.4 4.0 4.2 3.2 3.7 4.0 4.0 5.6 3.5 5.0 4.6 4.5 2.6 4 4 4 4 1 2 2 3 3 3 4 4 2 4 3 3 3 3 2 3 3 4 2 Synovial Fluid Serum ~ WBC per 111311.3 12,300 10,400 4,350 5,720 17,250 5,100 10,000 400 13,200 6,300 8,350 26,800 33,550 31,250 17,500 12,600 11,400 10,000 42,850 1,230 16,800 8,500 15.350 L.F.T. reeip. titer 1280 12x0 N.D. 5120 0 0 C’HX u/ml. C’2 u/ml. C’Hao C’2 u/ml. u/ml. <2 <2 <2 <2 <2 <2 <2 <2 <2 <5 <5 <5 <5 <5 <5 <5 <5 N.D. <5 <5 <5 <5 16 62 77 52 56 82 86 22 56 445 156 385 250 392 975 16 7.2 44 345 455 40 N.D. 34 45 550 42 342 43 496 39 548 31 435 41 52 0 38 410 226 39 392 39 650 37 595 30 795 39 40 409 43.5 575 0 0 5120 0 640 5120 5120 0 5120 0 0 5120 10 N.D. 5120 5120 5120 5120 10 11.5 12 13 25 <z 3.5 4 4 4.5 7.5 8.0 8.5 33 41 36 38 43 L.F:T. recip. titer 5120 12x0 5120 5120 0 640 1280 5120 5120 5120 640 5120 5120 12x0 0 0 10,240 5120 N.D. 5120 N.D. 5120 5120 ~ Wynovial fluid obtained during joint surgery; serum obtained one or two days later. had no unusual clinical or laboratory manifestations distinguishing them from the remainder of the patients with rheumatoid arthritis. The concentrations of whole complement and C’2 in synovial fluid of patients with rheumatoid arthritis (table 3 ) were markedly reduced compared to the preceding disease groups (tables 1 and 2 ) . Patients 1-10 showed no significant total hemolytic activity in synovial fluid diluted 1:10; C‘2 activity was barely measurable in several of these fluids under the conditions tested. Patients 11-14 had minimal hemolytic complement activity. Patients 15-22 had 13 or fewer C’HS0units per ml. and fewer than 90 C’2 units per ml. of synovial fluid. Only one patient, C. A,, had synovial fluid total hemolytic complement (25 units) and C’2 (445 units) activity comparable to that of patients with gouty, osteoarthritic, or traumatic effusions; similar results were obtained with samples obtained from this patient one year later. There were no clinical features distinguishing this patient from the remaining patients with rheumatoid arthritis. Nearly all rheumatoid synovial fluid tested in this study were obtained from patients with chronic effusions. Since prolonged stasis might lead to deterioration of complement components in vivo, it was decided to test rapidly recurring effusions. Synovial fluid was completely removed from the swollen knee joint of two rheumatoid patients. Thereafter the patients were allowed to ambulate and recurrent effusions aspirated at frequent intervals two to seven days apart. In both patients, initial and subsequent joint fluids 225 CHao AND C’2”‘ ACTIVITY I N SERUM were found to contain equally low total hemolytic complement and C’2 activity. Other Joint Diseases. Table 4 lists the findings obtained in patients with other arthritides. Serum whole complement activity and C’2 titer were significantly diminished in one patient with infectious arthritis due to Nocardia asteroides (G.D.), presumably on the basis of coexistent liver disease. The serum C’2 titer was also moderately reduced in the patient with psoriatic arthritis. Otherwise the serum C’Hjo and C’2 titers recorded in table 4 are normal or elevated. Total hemolytic complement and C‘2 activity were markedly reduced in the synovial fluid from single cases of scleroderma, psoriatic arthritis, sarcoidosis with arthritis, in two patients with acute polymyositis accompanied by joint effusions, and in two patients with septic arthritis due to Staphylacacctls nureus and Nowrdia asteroidm respectively Paired synovial fluid-serum samples from two patients with chondrocalcinosis ( pseudogout) yielded unexpected results. Although synovial fluid total hemolytic complement activity was approximately 50 per cent of serum activity, the C’2 activity of synovial fluid was only 20 per cent of the serum value. Individual cases of juvenile rheumatoid arthritis, ankylosing spondylitis and hypogammaglobulinemia with arthritis had synovial fluid complement activity similar to that found in traumatic and degenerative arthritis. It should, however, be noted that five of eight patients with juvenile rheumatoid arthritis examined by €€edbergS7had low synovial fluid hemolytic complement activity. Assay of PIE and p l C in Synovlal Fluid. P1E ( C4 )3 *was readily demonstrated in synovial fluid from a patient with traumatic arthritis, osteoarthritis, and Reiterb syndrome, but there was little or no B1E globulin demonstrable in five of seven patients with rheumatoid arthritis (table 5, fig. 1). The p1C globulin component of human complement ( C’3a)39 is readily converted on storage to an inactive form, P1A globulin, and the processing and storage conditions used in this study could not be expected to prevent this conversion. The specific antiserum currently available does not distinguish the conversion product obtained by storage, plA, from that produced by contact with antigen-antibody aggregates, p1G.40The conversion product identified in this study was arbitrarily designated as PlA. Many synovial fluids subjected to immunoelectrophoresis did contain ,81A globulin, but a more important finding was the invariable presence of considerable unconverted PIC globulin in virtually all fluids examined (table 5 ) . This finding implies that it is the low concentrations of C’2 and C’4 (,HE) rather than C‘3a ( p l C ) which account for the depressed total hemolytic complement activity of certain synovial fluids. DISCXJSSIQN The presence of whole hemolytic complement activity in the synovial fluid of patients with traumatic, degenerative, and gouty arthritis and Reiter’s syndrome has been previously ~ b ~ e r v e d .It* ~was , ~ ~therefore to be expected 4 6 8 9 10 11 12 7 C. B. c. w. S. D. A. T. H. J. G. D. V. B. c. L. G. B. 1 2 3 5 Pt. J. A. L. z. M. W. No. 5.8 4.85 5.4 3.8 3.6 5.2 2.75 2.8 6.2 4.3 7.0 4.0 1wk. 6 wks. 2 wks. 5 days 1wk. 4 days 4 days 3 mo. 3 mo. 2 2 4 4 2 2 3 3 4 1 3 3 Protein Mucin 9% Type 2 wks. 2 mo. 1wk. Duration of Effusion -~ 580 130 139800 176000 9850 500 4250 6250 600 3200 43250 7150 W.B.C. permm.3 <2 <2 <2 C’Hjo u/ml. 0 7.0 0 9.0 0 6.8 0 8.3 0 21 0 2 4 5120 21 0 30 0 21 0 5120 0 L.F.T. Recip. titer ~___ Synovial Fluid 27 46 23 39 124 130 278 320 230 16 13 N.D. C’2 u/ml. Hemolytic Complement and C’2 Activity in Other Joint Diseases Scleroderma Psoriatic arthritis Sarcoidosis with arthritis Polymyositis Polymyositis Infectious arthritis Infectious arthritis “Pseudogout” “Pseudogout” Juvenile R.A. Ankylosing spondylitis Hypogammaglobulinemia with arthritis Diagnosis Table 4.-Total 885 295 N.D. 565 410 870 232 643 650 855 440 528 42 48.5 46 44 35 53 23 46 43 44 49 47 u/ml. C”2 u/ml. C’H~O Serum -~ 0 0 0 0 0 0 0 0 0 640 0 0 L.F.T. Recip. titer m ! z M 2 + $ $ %0 ii 9 r 227 CH50 AND C’2hU ACTIVITY IN SERUM Table 5.-PlE and /3lC Globulins in Synovial Fluid: Correlation with Total Hemolytic Complement and C‘2 Activitq Synovial Fluid Pt. J. W. P. R. D. C. E. C. E. T. F. T. M. S . M. R. P. w. C. A. Diagnosis Traumatic arthritis Osteoarthritis Reiter’s snydrome Rheumatoid arthritis Rheumatoid arthritis Rheumatoid arthritis Rheumatoid arthritis Rheumatoid arthritis Rheumatoid arthritis Rheumatoid arthritis C‘Hm u/ml. C’2 u/ml. 29 23 45 <2 <2 <2 4 N.D. 8 <2 43.5 360 325 <5 <5 <5 <5 77 <5 575 B1E B1E +++ +++ +++ 0 0 0 0 + ++ +++ Conversion Products 0 0 0 0 0 0 + 0 0 + ++ +++ +++ 0 +++ 0 ++ 0 ++ 0 ++ + ++ ++ +++ + + +++ ++ ++ N.D. = not done. that such fluids contain functionally active C’2 and immunoelectrophoretically detectable PlE ((2’4) and P1C (C’3a) globulin. The concentration of C’H50 units, 21-38, and C’2 units, 184-386, per ml. of synovial fluid in 12 patients with traumatic or degenerative arthritis (table 1) was roughly one-half the serum concentration observed in 20 normal subjects, 32-44 C’H50and 350-670 C’2 units per ml. The synovial fluid values in acute gouty arthritis and Reiter’s syndrome (table 2) were similar to those observed in traumatic and degenerative arthritis. By comparison with these joint diseases, the concentrations of whole complement, 2-13 C’HEO units and C’2 5-86 units per ml. was markedly diminished in joint fluids from 22 of 23 patients with rheumatoid arthritis (table 3 ) . As observed by other^,^^,^^ there did not appear to be a direct relationship between the synovial fluid leukocyte count and hemolytic activity. Neither was there an apparent relationship between the protein concentration or mucin clot test and the complement activity of synovial fluid. Indeed, the leukocyte count, mucin clot test, and protein concentration were quite comparable in the patients with gouty or rheumatoid arthritis, and yet the hemolytic activity in gocty effusions was similar to that in traumatic and degenerative effusions, while that in rheumatoid fluid was markedly depleted. The complement activity present in rheumatoid fluids did not correlate with the presence or absence of rheumatoid factor. The possible explanations for the relative deficiency of whole complement activity in the synovial fluid of patients with rheumatoid arthritis include: ( a ) anticomplementary factors in rheumatoid fluid; ( b ) interference with the transport of complement components into the joint space; ( c ) immunologic activation of the complement system; or ( d ) non-specific activation of the complement system by proteolytic enzyme in or released from joint tissue. Incubation of rheumatoid synovial fluid lacking detectable whole complement and C’2 activity with serum or synovial fluid from patients with gout or Reiter’s syndrome did not decrease the C’Hjo titer of the latter. Similar obser- 228 Ra anti FOSTIROPOULOS, AUSTEN, B L W H PIE Syn. fl. osteo. C’H50 23u./ml. C’2 358u./ml. Ra anti PIE Syn. fl. R.A. C’H50 < 2u./ml. C’2 < 5u./ml. Ra anti PIE Fig. 1.-Immunoelectrophoresis of osteoarthritis and rheumatoid synovial fluid. Wells were filled with 0.02 ml. synovial fluid; troughs were filled with 0.1 ml. of rabbit anti P1E antiserum. Two precipitin arcs adjacent to the upper well indicate the presence of P1E globulin; no precipitin arcs were formed with the rheumatoid synovial fluid below. vations were made by Pekin and Zvaifler.2SFurthermore, the finding that P1E (C’4) is physically diminished (table 5, fig. 1) in synovial fluids lacking functional C 2 virtually excludes anticomplementary activity as a valid explanation for the reduced whole complement activity of these synovial fluids. Fluid from the supra-patellar pouch of healthy volunteers was noted by Pekin and Z ~ a i f l e r *to~ contain only one-tenth the hemolytic activity of serum. This finding may reflect the inability of complement components, some of which have high molecular to enter the joint space. It is also possible that the trauma associated with aspiration of small amounts of fluid from normal tissue spaces activates proteolytic enzymes capable of destroying complement. In any event, macromolecular proteins are present in joint effusion^,^^ and it seems unlikely that complement components would be excluded on the basis of size from rheumatoid but not from traumatic, degenerative, or gouty effusions. The possibility of malfunction of a specific transport system for complement components in rheumatoid arthritis can neither be supported nor excluded. A more attractive explanation is the specific, immunologic, intra-articular activation of the complement system by antigen-antibody interaction in tissue, in the fluid phase, or in relation to phagocytosis. The depletion of C’2 and P1E CHgo AND C’2’1‘1ACTIVITY IN 229 SERUM (C’4) is entirely compatible with the activation of C’la. Complement might be fixed by specific antibody bound to antigens present at the site of the primary synovial lesion, or by antigen-antibody complexes or gamma globulin aggregates trapped in the synovial membrane. K z ~ p l a nhas ~ ~found considerable 7s gamma globulin in the synovial lining and in the immediate subsynovial tissue in rheumatoid arthritis. However it is not known whether this gamma globulin is aggregated and capable of fixing complement. Alternatively complement could be activated in the fluid phase by immune complexes or gamma globuin aggregates. The third possibility, namely, utilization in relation to phagocytosis, could explain the reduced C’2 titer in the septic joints, in the fluids from patients with “pseudogout,” and even in rheumatoid arthritis since aggregates of rheumatoid factor and gamma globulin have recently been demonstrated in synovial fluid l e u k o c y t e ~ . ~ ~Against ,*j this latter interpretation is the failure to note a consistent reduction in synovial fluid C’2 in gouty effusions. Rheumatoid synovial fluid contains increased concentrations of several hydrolytic enzymes presumably released from the lysosomes of polymorphonuclear leukocytes and lining cells of the synovial membrane,42 and it is possible that some proteolytic enzyme directly or indirectly inactivates the complement system. For example, it has been demonstrated by Lepow et al.46 that serum plasmin (fibrinolysin), a proteolytic enzyme, converts the first component of human complement to its active form, C’la; the latter then destroys any C’2 and C’4 in the reaction mixture. Such a mechanism would account for the findings in rheumatoid effusion, i.e., striking depletion of C’2 and C’4 without extensive conversion or disappearance of C’3a. Plasmin activity has not been detected in synovial fluids which were also low in complement activity.37 This daes not exclude the possibility that other proteolytic enzymes activate C’l. On the other hand, the failure to note a correlation between the indices of intra-articular inflammation and reduced hemolytic activity is against this simple explanation, unless it is assumed that the postulated proteolytic activity appears in rheumatoid or septic tissue injury but not in gouty inflammation. SUMMARY The observations that total hemolytic complement activity is present in synovial fluid of patients with traumatic, degenerative, and gouty arthritis and that it is relatively absent from fluid of patients with rheumatoid arthritis were confirmed and extended. The presence of the second component of human complement( C’Bhn)in synovial fluid was demonstrated by a stoichiometric titration using EAC’laSp,4g” cells, while the fourth component (BlE) and a moiety of the third component ( P1C) were identified by immunoelectrophoretic analysis using specific antiserum. The concentration of second and fourth components of human complement was strikingly diminished in synovial fluid from patients with rheumatoid arthritis in comparison to their concentration in synovial fluid from patients with traumatic, degenerative, or gouty 230 FOSTIROPOULOS, AUSTEN, BLOCH effusions. These findings suggest that the reduction in whole complement activity was due to activation of the first complement component with subsequent utilization of the second and fourth component of complement. HOWever the mechanism of intra-articular complement fixation or activation in rheumatoid effusions remains to be determined. ACKNOWLEDGMENT The authors are indebted to Drs. Otto E. Aufranc, William N. Jones, Robert S. Pinals and Theodore B. Bayles for making available several synovial fluids used in this study. REFERENCES 1. Mayer, M. M.: Studies on the mechanism of hemolysis by antibody and complement. Progr. Allergy 5:215, 1958. 2. -: On the destruction of erythrocytes and other cells by antibody and complement. Cancer Res. 21:1262, 1961. 3. Muschel, L. M., and Treffers, H. P.: Quantitative studies on bactericidal action of serum and complement. 11. Some implications for mechanism of bactericidal reaction. J. Immunol. 76: 11, 1956. 4. Goldberg, B., and Green, H.: Cytotoxic action of immune gamma globulin and complement on Krebs ascites tumor cells. I. Ultrastructural studies. J. Exp. Med. 109:505, 1959. 5. Green, H., Fleischer, R. A., Banow, P., and Goldberg, B.: Cytotoxic action of immune gamma globulin and complement on Krebs ascites tumor cells. 11. Chemical studies. J. Exp. Med. 109:511, 1959. 6. Borsos, T., Dourmashkin, R. R., and Humphrey, J. H.: Lesions in erythrocyte membranes caused by immune hemolysis. Nature 202:251, 1964. 7. Gerlings-Petersen, B. T., and Pondman, K. W.: Erythrophagocytosis: a study of the antigen-antibody-complement reaction. Vox Sanguinis 7:655, 1962. 8. Spiegelberg, H. L., Miescher, P. A., and Benacerraf, B.: Studies on the role of complement in the immune clearance of Escherichia coli and rat erythrocytes by the reticuloendothelial system in mice. J. Immunol. 90:757, 1963. 9. Bloch, K. J., Kourilsky, F. M., Ovary, Z., and Benacerraf, B.: Properties of guinea pig 7s antibodies. 111. Identification of antibodies involved in complement fixation and hemolysis. J. Exp. Med. 117:965, 1963. 10. Cochmne, C. G.: Studies on the localization of circulating antigen-antibody complexes and other macromolecules in vessels. I. Structural studies. J. Exp. Med. 118:489, 1963. 11. Unanue, E., and Dixon, F. J.: Experimental glomerulonephritis. IV. Participation of complement in nephrotoxic nephritis. J. Exp. Med. 119: 965, 1964. 12. Ecker, E. E., and Lopez-Castro, G.: Complement and opsonic activities of fresh hunian sera. J. Immunol. 55: 169, 1947. 13. Nelson, R. A.: Immune adherence. In Mechanism of Cell and Tissue Damage Produced by Immune Reaction. J. Grabar and P. Miescher, eds. Benno-Schwabe, Basel, 1962, 245-248. 14. Ward, P. A., and Cochrane, C. G.: A function of bound complement in the development of Arthus reactions. Fred. Proc. 23:509, 1964. 15. Rapp, H. J., and Borsos, T.: Complement and hemolysis. Science 141:738, 1963. 16. Austen, K. F., and Cohn, Z. A.: Contribution of serum and cellular factors in host defense reactions. I. Serum factors in host resistance. New Eng. J. Med. 268933-944, 1963. 17. Becker, E. L.: Concerning mechanism of complement action. I. Inhibition of complement activity by diisopropyl fluorophosphate. J. Immunol. 77:462, 1956. 18. -: Concerning mechanism of comple- ment action. V. Early Steps in immune hemolysis. J. Immunol. 84:299, 1960. 19. Mayer, M. M., Asher, E. T., and Borsos, T.: Inhibition of guinea pig C'2 by rabbit anti-C'2. Fed. Proc. 21:17, 1962. 20. Stroud, R. M., Austen, K. F., and Mayer, M. M.: Immune hemolysis: Kinetics of enzymatic activaton and fixation of C'2 by activated C'1 ((2%). Fed. Proc. 22:613, 1963. 21. Mayer, M. M., Levine, L., Rapp, H. J., Marucci, A. A.: Kinetic studies on immune hemolysis. VII. Decay of EAC'1,4,2, fixation of C'3 and other factors influencing hemolytic action of complement. J. Immunol. 73:443, 1954. 22. Borsos, T., Rapp, H. J., and Mayer, M. M.: Studies on second component of complement. 11. Nature of decay of EAC'1,4,2. J. Immunol. 87:326, 1961. 23. Austen, K. F., and Beer, F.: The measurement of the second component of human complement (C'21'") by its interaction with EAC'lagP,4 gp cells. J. Immunol. 92946, 1964. 24. Guiney, E. J., Austen, K. F., and Russell, P. S.: Measurement of serum complement during homograft rejection in man and rat. Proc. Soc. Exp. Biol. Med. 115:1113, 1964. 25. Townes, A. S., Stewart, C. R., and Oslelr, A. G.: 11. Variations of nucleoprotein-reactive gamma globulin and hemolytic serum complement levels with disease activity. Bull. Johns Hopk. Hosp. 112:202, 1983. 26. Lachman, P. J., Miiller-Eberhard, H. J., Kunkel, H. G., and Paronetto, F.: The localization of in vivo bound compIement in tissue sections. J. Exp. Med. 115:83, 1962. 27. Laurell, A. B., and Grubb, R.: Complcment, complement components, properdin, and agglutination promoting factors in rheumatoid arthritis. Act3 path. rnicrobiol. scand. 43:310, 1958. 28. Pekin, T. J., and Zvaifler, N. J.: Hemolytic complement in synovial fluid, J. Clin. Invest. 43:1372, 1964. 29. Hedberg, H.: Studies on the depressed 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. hemolytic complement activity of synovial fluid in adult rheumatoid arthritis. Acta Rheum. Scand. 9:16,5, 1963. Kabat, E . A., and Mayer, M. M.: Experimental Immunochemistry, 2nd ed. Charles C Thomas, Springfield, Ill., 19431. Scheidegger, J. J.: Une micro-niethode de I'immunoelectrophorese. Int. Arch. Allergy 7: 103, 1955. Singer, J. M., and Plotz, C. hl.: The latex fixation test. I. Application to the serologic diagnosis of rheumatoid arthritis. Am. J. Med. 21:888, 1958. Ropes, M. W., and Bauer, W.: Synovial fluid changes in joint disease. Harvard Univ. Press, Cambridge, Massachusetts, 1953. Hornung, M., and Arguembourg, R.: Studies on the p l C globulin of human serum. Fed. Proc. 23:548, 1964. Williams, R. C., and Law, 11. H.: Serum complement in aniyl3idosis. J. Lab. Clin. Med. 56629, 1960'. Lee, S. L., and Rivero, I.: Cryoglobulins in systemic lupus erythematosus as circulating immune complexes. ATth. & Rheumat. 7:321, 1964. Hedberg, H.: The depressed synovial complement activity in adult and juvenile rheumatoid arthritis. Act1 Rheum. Scand. 10:109, 1964. Muller-Eberhard, H. J., and Biro, C. E.: Isolation and description of the fourth component of human complement. J. Exp. Med. 118:447, 1963. -, and Nilsson, U.: Relation of a plglycoprotein of human serum to the complement system. J. Exp. Med. 111:217, 1960. -, Calcott, M. M., and Mardiney, M. R.: Conversion of P l C globulin by C'2a. Fed. Proc. 23:506, 1964. Lepow, I. H., Naff, G. B., Todd, E. W., Pensky, J., and Hinz, C. F.: Chromatographic resolution of the first component of human complement into three activities. J. Exp. Med. 119: 593, 1964. Hamerman, D., Sandson, J., and Schiibert, M.: Biochemical events in joint disease. J. Chron. Dis. 16:835, 1963. Kaplan, M. H.: Immunologic aspects of 232 FOSTIROPOULOS, AUSTEN, BLOCH rheumatoid arthritis and systemic lupus erythematosus. Arth. and Rheumat. 6:475, 1963. 44. Hollander, J. L., Rawson, A. J., Restifo, R. A., and Lussier, A. J.: Studies on the pathogenesis of rheumatoid joint inflammation. Arth. & Rheumat. 7: 314, 1964. 45. Astorga, G., and Bollet, A. J.: Diagnostic specificity and possible pathogenetic significance of inclusion- body cells in synovial fluid. Arth. & Rheumat. 7:288, 1964. 46. Lepow, I. H., Wurz, L., Ratnoff, 0. D., and Pillemer, L.: Studies on the mechanism of inactivation of human complement of plasmin and by antigen-antibody aggregates. I. The requirement for a factor resembling C’1 and the role of calcium. J. Immunol. 73:146, 1954. George Fostiropoulos, M.D., U.S.P.H.S. Post Graduate Trainee, Arthritis Unit, Massachusetts General Hospital, Boston, Mass. K . Frank Austen, M.D., Associate in Medicine, Harvard Medical School; Research Career Development Awardee, National Institute of Allergy and Infectious Diseases, National Institutes of Health. Kurt J. Bloch, M .D., Instructoi in Medicine, Harvard Medical School; Senior Investigator, Arthritis and Rheumatism Foundation.