57 1

C3 ACTIVATION BY MONOSODIUM URATE MONOHYDRATE AND OTHER CRYSTALLINE MATERIAL

PETER HASSELBACHER Monosodium urate monohydrate (MSUM) is a potent activator of the complement system as measured by electrophoretic conversion of PIC to P,A. Activation of C3 in human serum by MSUM is both time- and dose-dependent. The sensitivity of the assay allows detection of C3 activation by as little as 0.2 m g / d of MSUM. It was observed that C3 activation is calcium dependent and eliminated by both EDTA and EGTA, a finding that demonstrated the major role of the classic pathway of complement activation. Excess calcium or magnesium alone inhibited C3 activation by MSUM in accord with the inhibitory effect of these cations on sensitized sheep cell hemolysis by complement. Heating of MSUM at 200°C for 2 hours removes the water of crystallization such that heated crystals may no longer be considered MSUM. Such treatment has a variable effect on C3 activation. Of the crystals and other material studied, only zymosan was more potent than MSUM in activating C3. Calcium pyrophosphate dihydrate and hydroxyapatite activated significant amounts of C3. Asbestos, glass wool, or a variety of microcrystalline steroids activated little or no C3. Presented in part at the 42nd Annual Meeting of the ARA, New York City, May 31-June 2, 1978, and in part at the 43rd Annual Meeting of the ARA, Denver, Colorado, May 31 and June 1, 1979. From the University of Pennsylvania School of Medicine and Dartmouth Medical School. Supported by a Clinical Investigator Award from the National Institutes of Health (7 KO8 AM00587), a Postdoctoral Fellowship from the American Rheumatism Association and a grant from the Philadelphia Chapter of the Arthritis Foundation. Peter Hasselbacher, MD: Assistant Professor of Medicine, Dartmouth Medical School and Associate Director, DartmouthHitchcock Arthritis Center. Address reprint requests to Peter Hasselbacher, MD, Dartmouth-Hitchcock Arthritis Center, Hanover, New Hampshire 03755. Submitted for publication October 1 I, 1978; accepted in revised form December 27, 1978. Arthritis and Rheumatism, Vol. 22, No. 6 (June 1979)

The biologic activity of sodium urate crystals is modified by their interaction with serum proteins. Exposure of sodium urate crystals to serum has been shown to enhance their phagocytosis (1-4), to generate chemotactic factors (2), and to inhibit contact mediated lysis of erythrocytes (5). Sodium urate crystals can activate the clotting and kinin systems through an interaction with Hageman factor (6) and can initiate complement activation (1,2,7-9). Urate crystals both in vitro and in vivo bind immunoglobulin with high affinity compared to other serum proteins, an interaction which may be responsible for some of the observed properties of urate crystals in biologic systems (1 0- 15). Urate induced inflammation may be diminished in animals depleted of complement (16,17). The presence of complement split products and of low complement activity of component protein has been observed in synovial fluids from patients with acute gout (18-22). The mechanism of complement activation by sodium urate crystals remains uncertain, but probably occurs in larger part through some portion of the classic pathway. Because C1 levels were not decreased to the same extent as C2 or C4 when urate crystals were exposed to fresh human serum, the possibility of a unique mechanism of activation has been raised (7). This study investigates further the interaction of monosodium urate monohydrate with the complement system and quantitates the ability of a number of different crystals to activate the complement pathway as measured by the formation of split products of C3.

MATERIALS AND METHODS Monosodium urate monohydrate (MSUM) crystals were prepared by a modification of the method of McCarty

572

and Faires (23). The uric acid was completely dissolved by heating until the solution was clear. Crystallization occurred overnight at 4°C. The crystals were desiccated for 10 days over Dryerite, and crushed with a glass rod to obtain a uniform product. The average length of the crystals was 5 pm (2.5-12.5). The crystals were not heated. Potassium urate crystals were prepared in the same way by substituting potassium hydroxide for sodium hydroxide. These needle shaped crystals were considerably smaller than MSUM. Calcium pyrophosphate dihydrate (CPPD) crystals were prepared by the method of Brown et a1 (24) and were provided by Dr. Rose Tse. Calcium hydroxyapatite (Type I) and zymosan A from Sucromyces cervisue were obtained from Sigma. Asbestos (amphibole) and Pyrex glass filtering wool were obtained from Fisher. Samples of a variety of microcrystalline steroid preparations marketed for intraarticular injection were generously provided by the manufacturers. These included: prednisolone acetate (Meticortelone), prednisolone tebutate (Hydeltra TBA), triamcinolone acetonide (Kenalog), dexamethasone acetate (Decadron LA), triamcinolone diacetate (Aristocort), betamethasone acetate (Celestone Soluspan), triamcinolone hexacetonide (Aristospan), and hydrocortisone acetate (Hydrocortone). Chemically pure triamcinolone acetonide was provided by the Squibb Co. Activation of C3. Stock crystal suspensions (25 mg/ ml) were prepared in phosphate buffered saline (PBS) (0.01 M sodium phosphate, pH 7.4). Glass wool and asbestos were weighed separately for each incubation tube since uniform suspensions were not possible. The steroid crystals were first washed in PBS to remove additives, usually polysorbate, benzyl alcohol, sodium carboxymethylcellulose, sodium citrate or ethylenedinitro-tetra acetic acid (EDTA). Samples of the suspensions (usually 0.1 ml) were placed in polypropylene centrifuge tubes (1.5 ml, Eppendorf) to which 0.5 ml of fresh human serum was added. Centrifuge tubes from a single lot were used to compare the activity of different crystals (see below). Serum was obtained from a single donor (PH) on the day of the experiment or occasionally kept at 4OC for up to 2 days. The tubes were mixed and placed on a Labquake rotator at 8 rpm in a 37°C incubator for one hour. A control tube containing only PBS was always included. The tubes were promptly centrifuged at 15,600g (Eppendorf, Model 5412) for 3 minutes after incubation. Aliquots of the supernatants were promptly diluted 1 :4 in veronal buffer (0.05M, pH 8.6) containing sodium EDTA O.OOSM, and sodium azide 0.5 gm/liter (VEA buffer). Each crystal was studied several times on different days. Crossed immunoelectrophoresis. Two dimensional crossed immunoelectrophoresis (25) was performed by a modification of the quantitative method of Clarke and Freeman (26). All six tubes of a single experiment were studied simultaneously on a single plate. A 205 X I10 mm glass plate (BioRad) was covered with 45 ml of 1% agarose (Marine Colloids ME distributed by Miles) in VEA buffer. A row of wells 2.5 mm in diameter was prepared in a line along the long dimension of the plate 45 mm from the bottom. The wells were 30 mm apart. Another well was prepared near the top of the plate for a tracer preparation of bromphenyl blue and serum. Five microliter samples of each diluted supernatant were placed in the wells by using Eppendorf pipettes in the reverse mode (27). Electrophoresis was performed (7.5 v/cm) along the long axis of the plate on a water cooled Plexiglas platform

HASSELBACHER

(homemade) by use of electrodes and vessels as described by Nerenberg (28). Commercial equipment may also be used. After the wells had emptied, they were filled with agarose to prevent artifacts. The blue tracer was allowed to migrate 65-70 mm. All excess gel was then removed from a line extending across the plate 5 mm above the edge of the wells. The top empty portion of the plate was covered with 15 ml of 1% agarose containing 0.45 ml of rabbit anti-human C3. This was facilitated by warming the empty portion of the plate slightly over steam or an alcohol lamp. Subsequent work (not reported here) has shown that the antibody-containing agarose is easier to pour if the row of wells is prepared 55 mm from the bottom of the plate. Electrophoresis was performed in the second dimension (7.5 v/cm) for 3 to 4 hours with the positive electrode at the top of the plate. The plate was then pressed, washed, dried under filter paper, and stained as described by Axelson et a1 (29). Antiserum was prepared in a rabbit by use of purified human C3, the generous gift of Dr. Carlos Arroyave. This antiserum was monospecific for C3 and its split products as determined by Ouchterlony immunodiffusion and by both regular and crossed immunoelectrophoresis using several ratios of antigen and antisera. A single immunodiffusion line or curve was seen when native C3 (from EDTA-plasma) was used as the antigen. The serum of a single bleeding was used for all the experiments comparing the activity of different crystals. Calculation of results. The areas under the precipitin curves of P I C and P,A proteins were measured with a compensating polar planimeter (Keuffel & Esser 620,000) directly from the plate. Each curve was traced 5 to 10 times depending on its size to yield an average area. The area under the P,A curve was expressed as percent of the combined areas. The P,A curve measures C3c. It is important to recall that the percent of total area converted to P,A does not directly reflect the percent weight C3 actually activated. This is due to the effect of changes in molecular weight and electrophoretic mobility of the C3c fragment on the assay used. For the specific antisera used here, a decrease in PIC area was accompanied by a 1.5 fold increase in PIA area. For the comparative data in Table 1, the difference between the PIA areas of the specimen and the PBS control is calculated, the result expressing excess P I A area converted. Time and dose response. The dose response curve of MSUM was studied by adding increasing amounts of MSUM stock suspension to 1 ml of serum, the total volume kept constant at 1.2 ml with PBS. The tubes were incubated for one hour as above. The time response was studied by adding 0.05 ml of MSUM stock suspension to 0.5 ml serum (2.3 mg/ml) and incubating for increasing periods. A control containing only PBS and serum was incubated in parallel to measure spontaneous conversion. Effect of EDTA. A O.5M stock solution of EDTA was prepared by dissolving disodium EDTA (Baker) with sufficient NaOH to give a pH of 7.0. Sodium-EDTA solution was added to 0.5 ml serum in a constant volume containing 2.2 mg/ml MSUM. The tubes were incubated (with PBS and crystal-only controls) for one hour and studied as above. The effect of re-adding CaCl, or MgCI, to the chelated suspensions was also studied. Effect of EGTA. A stock solution of ethyleneglycol-

573

C3 ACTIVATION

Table 1. Excess PIA area converted

Crystal MSUM CPPD Potassium urate Hydroxyapatite Meticortelone Hydeltra TBA Kenalog Decadron Anstown Celestone Soluspan Aristospan Hydrocortone

Number of tests

MeanfSD(%)

4 4 4 4

5 4 3 3 4 2 4 4

Asbestos 14 mg/ml Glass wool I5 mg/ml Zymosan 5 mg/ml

26.5 f 0.58 11.5 f 3.70 9.0 f 2.58 5.3 f 2.63 4.6 f 3.13 2.3 f 2.87 1.0 f 3.00 1.0 f 3.00 0.8 f 2.87 -1.5 f 0.71 -2.0 f 3.46 -3.5 f 2.38

P 5.001

0.01 0.01 0.05 0.05 NS NS NS NS NS NS NS

2.0 3.O

43.0

bis-(beta-amino-ethyl ether) N, N-tetra acetic acid, (EGTA) was prepared by dissolving EGTA (Sigma) with sufficient NaOH to give pH of 7.0. Serum (0.5ml) was rendered 0.01M with respect to EGTA, conditions known to block the classic pathway of complement activation but allowing alternate pathway activation (30).MSUM was added (2.1 mg/ml) with incubation for 60 minutes as above. Effect of Ca++ and Mg++. Because an increment of more than 0.005M calcium or magnesium to chelated samples inhibited C3 conversion, the effect of these cations on nonchelated samples was studied. Calcium or magnesium chloride solutions at neutral pH were added to 0.5 ml serum containing MSUM (2.2 mg/ml) over the range of 0.005 to 0.03M in excess of normal serum cation concentrations. Tubes were incubated as above with PBS and crystal controls. Effect of heating M S U M . Previous investigators have noted that heating urate crystals at 200°C for 2 hours alters their ability to activate complement (7,8). For the present study, MSUM crystals were heated in a number of ways. In method A, the 0.1 ml aliquots of MSUM stock solution were placed in small glass test tubes and heated at 200°C for 2 hours before adding serum and incubating in the glass tubes. In method B, a larger quantity of crystals was heated and broken up before use with a glass rod. A new stock solution was prepared in endotoxin free tubes with autoclaved buffer. The weights of the desiccated MSUM crystal preparation were determined before and after varying periods of heating. Statistics. The difference in excess P , A area converted by different crystals was analyzed with the paired Students' ttest (3 1) using a Texas Instrument TI 59 calculator with a statistics module.

RESULTS Reproducibility of methods used. The area under a single peak of 0.10 square inches was measured with 10 sets of 5 tracings each. The coefficient of variation was 0.022. Because of variations in the electric field, rate of cooling, and other unknowns, the areas of multiple

aliquots of a single specimen at each of the 6 positions on the plate varied. However, the coefficient of variation of the percent total area under the P ,A curve was 0.008. If 6 separate specimens of the same serum sample were incubated in separate tubes and electrophoresed simultaneously, the coefficient of variation was 0.029. The mean PIA area of the 10 saline control tubes used for the accumulation of the data in Table 1 was 46.4 f 4.7% (the coefficient of variation was 0.10). The area under the PIA peak of a completely converted specimen (heated at 37°C for 6 days) is larger than the initial PIC peak. In the present system, the increase in PIA area is 1.5 times the decrease in PIC. Thus the actual amount of C3 converted may be estimated from the data provided. The activation of C3 by MSUM is both dose and time dependent. Figure 1 shows the dose response curve. The smallest amount of crystals used (0.2 mg/ml) showed definite activation at one hour. Increasing the concentration of crystals led to increased activation but with decreasing effectiveness. The crystal concentration used to compare different types of crystals (4.2 mg/ml) occurs on the portion of the curve with the smallest DOSE RESPONSE

'I I

80

2oI 1

2 M SUM

3

4

"Yml

Figure 1. Effects of increasing concentration of MSUM on C3 activation. Increasing amounts of MSUM stock solution were added to 1.0 rnl of fresh human serum with the total volume of 1.2 ml kept constant with PBS. The P I Carea in the absence of MSUM reflects activation by the tube well.

574

HASSELBACHER

'4 L

30

60

90

120

TIME Cmin)

MSUM. The addition of 0.0025M EDTA partially, and 0.005M completely, inhibited the appearance of PIA when compared to 72% PIA area in the nonchelated serum. The addition of 0.005M CaCl, completely reversed this inhibition, as did 0.005MMgCl,. These experiments do not define the divalent cation requirement since Mg" may reverse the EDTA inhibition by displacing Ca++from the chelator (32,33). The calcium requirement of C3 conversion by MSUM was demonstrated with EGTA (Figure 4). Unlike EDTA, EGTA (0.01M) only partially inhibited C3 conversion in the control tube. EGTA markedly inhibited activation by MSUM such that C3 conversion was identical to that observed in the control tube alone. C3 activation by zymosan was minimally affected by EGTA under these conditions, a finding that demonstrated the presence of sufficient free Mg" available for the alternate pathway to function. The addition of Mg" (0.1 to 0.5 mM) beyond that present in the EGTA chelated sample did not increase C3 conversion by zymosan.

Figure 2. Time response of C3 activiation by MSUM. T o 0.5 ml of fresh human serum was added 0.05 ml of MSUM stock solution (2.3 mg/ml). The solid line indicates effect of incubation at 37°C. The dotted line serves as a control and indicates the effect of adding 0.05 ml of PBS to the serum. The PIC area of the serum before incubation was 96%. N o correction was made for the MSUM curve which includes tube-induced activation.

slope. This would have the effect of magnifying the relative activity of less active crystals when compared with MSUM. Incubating MSUM 4.2 mg/ml for 3 hours caused the formation of 82% P,A. Figure 2 shows the time course of C3 activation for 2.3 mg/ml of MSUM. The greater portion of C3 activation occurs within the first 15 minutes, with an additional phase of slower activation. It is important to note that significant activation occurs in the control tube, presumably as a result of interaction of complement with the tube wall. Spontaneous activation was much less if the tube was not inverted, a finding consistent with the hypothesis of serum-tube interaction. A second batch of polypropylene tubes from the same manufacturer gave less than half this degree of spontaneous activation (17% PIA area), but these were not used for the data in Table 1. The need for careful controls in this type of system is emphasized. Effect of EDTA and EGTA. The addition of EDTA to the incubation tubes eliminated conversion of C3 both by crystals and the tube itself. Figure 3 shows the effect of EDTA on the C3 conversion by 2.2 mg/ml

0

EDTA

(M.)

Figure 3. Effect of EDTA on C3 activation by MSUM. A neutral solution of EDTA was added to 0.5 ml of serum containing MSUM (2.2 mg/ml). Volume was kept constant with PBS. The bars represent the percent P I A area appearing. N o correction is made for tube-induced activation, and the virtual absence of P,A in the 0.005M EDTA tube represents inhibition of both MSUM and tube-induced activation.

575

C3 ACTIVATION

It is of interest that the addition of Ca++or Mg++

to nonchelated samples paradoxically inhibited C3 conversion. Figure 5 shows the effect of adding these cations to incubation tubes containing 2.2 mg/ml MSUM. The appearance of PIA is almost eliminated by 0.03M CaC1, in excess of normal serum concentration of calcium. The addition of Mg" inhibited C3 conversion to a less marked degree. Effect of heating. Heating of MSUM preparations at 200°C had variable effects on the ability to activate C3. If the individual 0.1 ml aliquot of stock MSUM suspension was heated in individual glass tubes before the addition of serum (method A), there was 87% less excess PIA area converted. However such treatment causes the crystals to clump strongly together. This may act to decrease surface area available for serum contact. There are other changes that occur on heating at 200" that have not been stressed in the literature. When a larger quantity of the dessicated crystals was heated for 2 hours, they lost an average of 7% of their weight, and if heating was prolonged to 8 hours, a mean of 10% of

looI

a 'c; 0

00

Mg"

=-

m

* 20

0

-

.o 1

.o 2 Concentration

.o 3

( M.)

Figure 5. Effect of Ca++ and Mg++ on C3 conversion by MSUM. Increasing amounts of neutral solutions of CaC& or MgCl, were added to 0.5 ml of serum containing MSUM (2.2 mg/ml). The horizontal axis represents the amount of cation added to the incubation tubes above the normal serum concentrations. Thus, at zero "added-concentration," calcium is assumed to be approximately 0.0025M and magnesium 0.001M. The appearance of P , A is inhibited. Closed circles represent calcium, and open circles magnesium. No correction is made for tube-induced activation.

80

60

Q

m 3

40

20

CONTROL

MSUM

ZYMOSAN

Figure 4. Effect of EGTA (0.01M) on C3 activation in the control tube and by MSUM and zymosan. The bars represent the percent PIA area appearing. Cross hatched bars represent EGTA chelated samples. No correction is made for tube-induced activation.

the initial weight was lost. Crystals heated for 2 hours by method B gave only 20% less PIA area converted when used in equal weights. When crystals heated 8 hours were used, there was 37% more PIA area converted. Microscopic examination of the heated preparations showed an increasing amount of amorphous material. The results obtained from comparison of different crystals are presented in Table 1. Most of the crystals were tested on several occasions. Sodium urate was the most active of all the crystals tested with a mean excess PIA area of 26.5%. This was exceeded only by zymosan which had an excess of 43%. CPPD crystals were next most active with an average difference of 11.5%. Potassium urate and hydroxyapatite also showed significant activation (9.0% and 5.3%, respectively). Of the eight steroid preparations studied, only prednisolone acetate (Meticortelone) activated significantly more C3 than saline controls. The asbestos and glass

HASSELBACHER

576

preparations tested showed little activity in this system, even when tested in larger doses.

DISCUSSION The potent ability of MSUM to activate the complement system is confirmed. Six different MSUM preparations gave comparable results, confirming that the ability to activate C3 was not isolated to a single preparation. This confirms the results of Naff and Byers (7) who used a hemolytic assay to detect loss of complement activity in serum after exposure to microcrystalline sodium urate. The use of the electrophoretic assay in the present study has allowed the detection of lesser degrees of complement activation than that detected by hemolytic assay for whole complement activity. The sensitivity of the present assay was greater than that found by Naff and Byers, who used 7.5 mg/ml of crystals and were not able to detect an effect with less than 0.9 mg/ml of sodium urate crystals or to demonstrate any effect by CPPD (7). Compared with O.OO5M EDTA above, only 0.001M EDTA was required by these authors to completely inhibit the detection of complement activation by urate crystals (7). Thus the technique presented here may be used with great sensitivity to measure and compare the activation of C3 by various materials. Phelps and McCarty demonstrated that urate crystals heated at 200°C for 2 hours lost their ability to activate complement as measured by a hemolytic assay (8). They postulated that complement activation was due to the presence of endotoxin. However, Naff and Byers demonstrated that properdin activity was not diminished by crystal exposure and that heated crystals regained their ability to activate complement after redissolution and recrystallization. They argued convincingly that endotoxin was not required for complement activation to occur (7). The present study confirms this conclusion by demonstrating a calcium and not magnesium requirement, with complete absence of C3 activation when the classic (but not the alternate) pathway was blocked. It must be stressed that heated crystals may no longer be assumed to be MSUM. Howell et a1 (34) have shown that MSUM is the crystal type present in gout and that heating MSUM at 2OOOC for 12 hours removes the water of crystallization. For MSUM this represents 8.66%of their weight (34). The present study shows that even heating for short periods of time removes most of the water of crystallization, depending on the size of the sample and vessels used. In most previous studies the

details of heat treatment of sodium urate crystals have not been presented in sufficient detail to allow identification of the probable crystal type present. Clearly this factor must be controlled in future work since the present studies and other observations in this laboratory clearly show that heated MSUM crystals have different properties than unheated crystals. The mechanisms of complement activation by MSUM remain unclear. When a survey of early component activities of complement was made in urate exposed serum, Naff and Byers noted that C1 was only slightly diminished, whereas the activities of C2, C3, C4, and C5 were markedly decreased (7). These authors postulated that perhaps some mechanism other than the complete classic pathway may be involved. Of interest, in their pioneering work, Naff and Byers noted that complement activation was considerably less in a serum with complete agammaglobulinemia. This decreased activation was reversed by the addition of normal fresh human serum. These authors supplemented the deficient serum with isolated human gamma globulin, but because of an anticomplementary activity of that preparation only a small additional effect by the crystals could be demonstrated. Other pathologic human sera with deficiencies of single classes of immunoglobulins did support normal complement-induced activation by MSUM (7). Although these authors postulated the presence of an unidentified serum factor, their data do not rule out a requirement for immunoglobulin for activation of complement by urate crystals or an interaction of C1 itself (or one of its subunits) with MSUM which initiates activation. The present and previous work supports the hypothesis that complement activation in whole serum by MSUM is calcium dependent, occurs through the classic pathway of complement activation, probably includes the action of C 1, and is not due to the presence of endotoxin in the crystal preparation. Additional work under way in this laboratory comparing IgG and Clq binding by a variety of crystals demonstrates that simple binding of these proteins does not correlate with C3 activation (unpublished data). The inhibition of C3 activation by high concentrations of calcium and magnesium is of uncertain significance at this time. There exists a precedent for this phenomenon in that excess Ca++will inhibit the hemolytic activity of complement in a classic sheep erythrocyte assay (33,35,36). The site of inhibition remains unknown. MSUM was more active in the present system

C3 ACTIVATION

than the other crystals studied. The degree of C3 activation was paralleled by the known phlogistic potential of the crystals (37-39). CPPD was demonstrated to have appreciable activity compared to MSUM and was in turn somewhat more active than hydroxyapatite. The steroid crystals were the least active of those crystals generally associated with crystal induced arthritis. The amount of C3 activated by the steroid crystals was small and at the limit of resolution of the present assay. It cannot be said with certainty that any single steroid preparation was more active than another. It is not concluded that complement activation by these crystals is the primary determinant of their inflammatory properties, but neither can it be assumed that complement activation does not play some partial and contributory role. It may be that the ability to activate complement reflects or augments some independent and more fundamental phlogistic property of crystalline material.

ACKNOWLEDGMENTS The author is very grateful for the technical assistance of Ms Jeanie Hahn and for the secretarial assistance of Ms Chris Funk and Ms Nancy Marshall.

REFERENCES 1. Naff GB, Byers PH: Possible implication of complement in acute gout. J Clin Invest 46:1099-1100, 1967 2. Byers PH, Ward PA, Kellermeyer RW, Naff GB: Complement as a mediator of inflammation in acute gouty arthritis. 11. Biological activities generated from complement by the interaction of serum complement and sodium urate crystals. J Lab Clin Med 81:761-769, 1973 3. Skosey JL, Kozin F, Chow DC, May J: Differential responses of human neutrophils to monosodium urate crystals and MSU coated with gamma globulin. Clin Res 24:11 lA, 1976 4. Kozin F, Skosey JL, May J, Chow DC: Modification of responses of human neutrophils to monosodium urate crystals by coating of crystals with serum proteins. Clin Res 24:331A, 1976 5 . Wallingford WR, McCarty DJ: Differential membranolytic effects of microcrystalline sodium urate and calcium pyrophosphate dihydrate. J Exp Med 133:100-112, 1971 6. Kellermeyer RW, Breckenridge R T The inflammatory process in acute gouty arthritis. I. Activation of Hageman factor by sodium urate crystals. J Lab Clin Med 65:307315, 1965 7. Naff GB, Byers PH: Complement as a mediator of inflammation in acute gouty arthritis. I. Studies on the reaction between human serum complement and sodium urate crystals. J Lab Clin Med 81:747-760, 1973

577

8. Phelps P, McCarty DJ: Crystal-induced arthritis. Postgrad Med 45:87-93, 1969 9. Hasselbacher P Activation of C3 by monosodium urate, potassium urate, and steroid crystals. Arthritis Rheum 21:565, 1978 10. Hasselbacher P, Schumacher HR: Localization of immunoglobulin in gouty tophi by immunohistology, and on the surface of monosodium urate crystals by immune agglutination. Arthritis Rheum 19:802, 1976 11. Kozin F, McCarty DJ: Adsorption of protein to monosodium urate crystals: a possible mechanism of crystal induced inflammation. Arthritis Rheum 19:805, 1976 12. Kozin F, McCarty DJ: Protein adsorption to monosodium urate, calcium pyrophosphate dihydrate and silica crystals. Arthritis Rheum 19:433-438, 1976 13. Kozin F, McCarty DJ: Protein binding to monosodium urate monohydrate, calcium pyrophosphate dihydrate and silicon dioxide crystals. I. Physical characteristics. J Lab Clin Med 89:1314-1325, 1977 14. Hasselbacher P, Schumacher HR: Immunoglobulin in tophi and on the surface of monosodium urate crystals. Arthritis Rheum 21:353-361, 1978 15. Hasselbacher P: Binding of IgG by monosodium urate monohydrate is a function of the ionic charge of the protein. Clin Res 26:623A, 1978 16. Webster ME, Maling HM, Zweig MH,%'illiams MA, Anderson W: Urate crystal induced inflammation in the rat: evidence for the combined actions of kinins, histamine and the components of complement. Immunol Commun 1:185-198, 1972 17. Kellermeyer RW, Naff GB: Chemical mediators of inflammation in acute gouty arthritis. Arthritis Rheum 18:765-770, 1975 18. Hasselbacher P: Immunoelectrophoretic assay for synovial fluid C3 with correction for globulin. Arthritis Rheum 22:243-250, 1979 19. Pekin TJ, Zvaifler NJ: Hemolytic complement in synovial fluid. J Clin Invest 43:1372-1382, 1964 20. Townes AS, Sowa JM: Complement in synovial fluid. Johns Hopkins Med J 127:23-37, 1970 21. Bunch TW, Hunder GG, McDuffie FC, OBrien PC, Markowitz H: Synovial fluid complement determination as a diagnostic aid in inflammatory joint disease. Proc Mayo Clin 49:715-720, 1974 22. Hunder GG, McDuffie FC: Activation of complement components C3 and factor B in synovial fluids. J Lab Clin Med 89:160-171, 1977 23. McCarty DJ, Faires JS: A comparison of the duration of local anti-inflammatory effect of several adrenocorticosteroid esters-a bioassay technique. Current Therap Res 5:284-290, 1963 24. Brown EH, Lehr JR, Smith JP, Frazier AW: Preparation and characterization of some calcium pyrophosphates. Agric Food Chem 11:214-222, 1963

578

25. Laurel1 CB: Antigen-antibody crossed electrophoresis. Analyt Biochem 10:358-361, 1965 26. Clarke HGM, Freeman T: Quantitative immuneelectrophoresis of human serum proteins. Clin Sci 35:403413, 1968 27. BioRad Laboratories: Bulletin 1046, December 1976 28. Nerenberg ST: Electrophoretic Screening Procedures. Philadelphia, Lea & Febiger, 1973 29. Axelson NH, Kroll J, Weeke B: A manual of quantitative immunoelectrophoresis: methods and applications. Scand J Immunol 2:l-36, 1973 30. Fine DP, Marney SR, Colley DG, Sergent JS, Des Prez RM: C3 shunt activation in human serum chelated with EGTA. J Immunol 1092307-809, 1972 31. Colton T: Statistics in Medicine. Boston, Little Brown & Co, 1974, pp 127-142 32. Hovig T: The effect of calcium and magnesium on rabbit blood platelet aggregation in vitro. Thromb Diath Haemorrh 12:179-200, 1964 33. Bryant RE, Jenkins DE: Calcium requirements for com-

HASSELBACHER

34.

35.

36.

37.

38.

39.

plement dependent hemolytic reactions. J Immunol 101:644668, 1968 Howell RR, Eanes ED, Seegmiller JE: X-ray diffraction studies of the tophaceous heposits in goit. Arthritis Rheum 6:97-103, 1963 Yachnin S, Laforet MT, Gardner FH: pH dependent hemolytic systems. I. Their relationships to paroxysmal nocturnal hemoglobulinuria. Blood 17:83-96, 1961 Levine LL, Cowan KM, Osler AG, Mayer MM: The essential role of calcium in complement fixation. J Immunol 71~367-373, 1953 McCarty DJ, Hogan JM: Inflammatory reaction after intrasynovial injection of microcrystalline adrenocorticosteroid esters. Arthritis Rheum 7:359-367, 1964 Dieppe PA, Huskisson EC, Crocker P, Willoughby DA: Apatite deposition disease, a new arthropathy. Lancet I: 266-269, 1976 Seegmiller JE, Howell RR, Malawista SE: The inflammatory reaction to sodium urate. Its possible relationship to the genesis of acute gouty arthritis. JAMA 180:469475, I962

C3 activation by monosodium urate monohydrate and other crystalline material.

57 1 C3 ACTIVATION BY MONOSODIUM URATE MONOHYDRATE AND OTHER CRYSTALLINE MATERIAL PETER HASSELBACHER Monosodium urate monohydrate (MSUM) is a potent...
673KB Sizes 0 Downloads 0 Views