- Email: [email protected]
Binding of immunoglobulin and activation of complement by asbestos fibers
Binding of immunoglobulin activation of complement asbestos fibers Peter Hasselbacher,
M.D. Hanover,
and by
N. H.
Asbestos fibers adsorbed IgG from whole human serum in amounts comparable to that taken up by crystals of calcium pyrophosphate or hydroxyapatite, but less than that adsorbed by crystals of monosodium urate monohydrate. Chrysotile asbestos was much more active in binding IgG than was amphibole asbestos. The electrophoretic mobility of adsorbed IgG was a ,function qf the surface charge of the crystal studied. Negatively chargedJibers adsorbed cationic IgG, and positively charged fibers adsorbed anionic IgG. Complement activation by asbestos, as judged by electrophoretic conversion of C3, was similar for all fiber types tested. Properdin factor B was also activated in the presence of asbestos. Activation of both properdin factor B and C3 was only partially inhibited by ethylene glycol tetraacetic acid, demonstrating that activation was occurring through both the classical and alternative pathways.
The diseases associated with asbestos exposure are becoming increasingly important as the magnitude of this public health problem becomes more widely appreciated. The mechanisms by which asbestos fibers cause tissue injury or neoplastic transformation are uncertain.lp2 If a comparison with other forms of silicosis or crystal-induced arthritis is justified, the pathogenesis of asbestos-related disease is likely to be multifactorial.3-4 Factors such as particle size, ability to undergo phagocytosis by macrophages, and membrane lysis of cells or lysosomes have been considered.5-8 Recent studies of both silicosis and crystal-induced arthritis have suggested that surface adsorption of immunoglobulin or other proteins and activation of the complement pathways may play a
role in modulating the biologic activity of the crystals.‘-16 In this report the IgG-binding and complement-activating properties of asbestos fibers will be quantified and compared with those of other crystals of biologic interest. The pathways by which complement activation is initiated will also be described. METHODS
AND MATERIALS
International Union Against Cancer (UICC) standard samplesof five asbestospreparationswere generously provided by Dr. V. Timbrell.” These preparations were washed 10 times by centrifugation and resuspension in phosphate-buffered saline (0.14 M NaCl, 0.01 M sodium phosphatepH = 7.4) (PBS). Washing causeda loss of material, presumably the finest particles; therefore unwashed fibers were also tested. Washed and dried fibers were suspended in PBS, 25 mgiml. Monosodium urate monohydrate
From the Dartmouth Medical School and Dartmouth-Hitchcock Arthritis Center. This work was supported by a Clinical Investigator award (7KO8-AM-00587-02) from the National Institutes of Health, United States Public Health Service Grant AM 14780, grants from the National and New Hampshire Chaptersof the Arthritis Foundation, and a Multipurpose Arthritis Center grant from the National Institutes of Health. This work was presentedin part at the 1979National Meeting of the American Federationfor Clinical Research,Washington, D. C. Received for publication Feb. 9, 1979. Accepted for publication May 2, 1979. Reprint requeststo: Peter Hasselbacher,M.D., Connective Tissue Disease Section, Dartmouth Medical School, Hanover, NH 03755.
Vol. 64, No. 4, pp. 294-298
(MSUM) and calcium pyrophosphate dihydrate (CPPD) were preparedas previously described.lx Zymosan and calcium hydroxyapatite were obtained from Sigma Chemical Co., St. Louis, MO.
Adsorption
of IgG
The amount and nature of IgG adsorbed by crystals were determined as previously described with minor modifications.‘a, ” Unwashed asbestos fibers (75 mg) were suspended in 0.5 ml fresh human serum in a l.S-ml polypropylene centrifuge tube with constant inversion for 30 min at room temperature. The tubes were centrifuged at 15,600 x g (Eppendorf centrifuge model 5412) for 3 min. and the supematants removed. The asbestos pellet was
0091-6749/79/100294+05$00.50/0
0 1979 The C. V. Mosby
Co.
VOLUME NUMBER
64 4
Binding
of immunoglobulin
by asbestos
295
FIG. 1. IgG eluted from washed, serum-exposed asbestos fibers was studied by REP. Wells a to f contain simple dilutions of a control serum of 1 : 20 to 1 :400. Well 7, anthophyllite; 2, Canadian chrysotile B; 3, crocidolite; 4, Rhodesian chrysotile A; 5, amosite. The positive electrode is marked.
washed 10 times by centrifugation and resuspension in PBS. The washed crystals were eluted for 30 min at room temperature with 1 ml PBS, with concentration at 1 M with respect to NaCI. The IgG in the eluate was studied by rocket immunoelectrophoresis (RIEP) as previously described.” The nonasbestos crystals were treated in the same way except that 150 mg of crystals were suspended in 0.5 ml serum, and the crystals were washed five times following serum exposure.
Activation
of complement
Complement activation was measured by determining the amount of electrophoretic conversion of C3 and properdin factor B (FB), as previously described.is Washed crystals were suspended in fresh human serum (2.3 mg/ml) with constant inversion for 60 min at 37” C. Complement activation was terminated by adding ethylenediamino tetraacetate (EDTA) (0.01 M). The supematant serum was appropriately diluted in electrophoresis buffer and studied for conversion products by crossed immunoelectrophoresis. Antiserum to C3 was produced as previously described,20 and antiserum to FB was the generous gift of Dr. Carlos Arroyave. Activation of C3 was estimated by measuring the areas under the &-C and Pi-A curves produced by crossed immunoelectrophoresis. Conversion was expressed as the percent of the total area under the two precipitin arcs represented by /3,-A. Similarly, activation of FB was judged by the percent of total area appearing under the y-fragment arc. It should be emphasized that neither of these measurements
corresponds exactly to the percent of total component protein converted. The effect of adding EDTA (0.005 M) or ethylene glycol tetraacetic acid (EGTA) (0.010 M) to the incubation tube was studied to determine the pathway of initiation of complement activation. EGTA binds calcium preferentially when compared with magnesium, thus inhibiting the classical pathway of complement activation while allowing the alternative pathway to function.”
RESULTS Adsorption
of IgG
When IgG was studied by RIEP as described, two “rockets ” formed, one in the direction of each electrode, reflecting the heterogeneity of charge of IgG molecules. No IgG was detected in any of the final wash supernatants. The eluates of all five asbestos preparations contained IgG. Table I lists the amount of IgG eluted from asbestos and other crystals. The amounts varied widely among the asbestos types, with the two chrysotile preparations adsorbing amounts approximating those adsorbed by crystals of CPPD or hydroxyapatite. The binding by amphibole preparations was considerably less as judged by eluted IgG. Of interest was the electrophoretic mobility of the eked protein (Fig. 1). IgG adsorbed by the nonasbestos material$* and by amphibole asbestos
296
J. ALLERGY
Hasselbacher
TABLE I. Adsorption activation
of IgG and complement
by crystals
Type of crystal
PBS Rhodesian chrysotile A Canadian chrysotile B Anthophyllite Crocidolite Amosite Zymosan MSUM CPPD Apatite
C3 activation (o/o))+
FB activation (o/o)*
IgG eluted (pglmg)
Control
EGTAt
Control
EGTA
1.54
25 38
II 20
8 13
O$ 6
I .83
34
I.5
13
8
0.10 0.39
42 37
22 18
IO
0.14 -
34
21
75 61
61 8 -
16 13 12 69 27 -
7 7 55 O$ -
-
-
-
3.65 I .54
1.15
*C3 and FB activations were quantitated by measuring the percent of total area beneath the P,-A and y-precipitin curves, respectively. $EGTA = 0.01 M. $.ln these samples, similar precipitin lines formed in the y-region, accompanied by a small decrease in area of the a-peaks. Because no arcs were present, this amount of activation was too low for quantitation. No precipitin lines were formed in EDTA-treated samples.
was restricted to cationic species of IgG moving toward the negative electrode. IgG eluted from the chrysotile fibers was restricted to anionic species. When electrophoresis was performed at neutral pH the same pattern was observed. Activation
of C3
Table 1 lists the percent of area under the /3,-A curve of C3, or the y-curve of properdin FB for control and for EGTA-chelated sera. In the saline control tube, 25% of the C3 area was beneath the Pi-A curve. This represents complement activation by the tube wall, which is considerably less if the tubes are not inverted. Extensive experience in this laboratory has demonstrated that the coefficient of variation of the percent C3 area converted is 0.029 for multiple samples studied on the same plate, and on the same day. The mean conversion by the five asbestos preparations was 37.0%, which represents a highly significant difference. This was small compared to C3 conversion caused by MSUM (67%) or zymosan (75%) but was comparable to C3 conversion by CPPD or hydroxyapatite. In the presence of EGTA, C3 conversion by asbestos was reduced by about half, to a mean of 19.2%. EGTA had little effect on C3 conversion by zymosan, but reduced conversion by MSUM to control levels, as reported elsewhere.‘6
CLIN. IMMUNOL OCTOBER 1979
EGTA also decreased C3 conversion in the control tube from 25% to 11%. This proportion has been confirmed in other experiments and represents a component of alternative pathway activation by the polypropylene tube wall. EDTA completely eliminated C3 conversion in the control and all sample tubes. There did not appear to be any relationship between type of asbestos fiber and C3 activation or effect of EGTA. Unwashed asbestos fibers converted similar amounts of C3 but were not studied further. Activation
of factor
B
Generation of the y-fragment of FB by zymosan (69% area) or MSUM (27%) was much greater than that by asbestos (mean, 13.4%) or control (8%). An increase in the area of the y-fragment correlated well with a decrease in the area of the a-component. Although my experience with the FB conversion system is not as extensive as with C3 (due to limited supply of antiserum), the reproducibility of FB conversion analysis appears to be the same. Comparison of individual pairs of crystals on smaller plates confirms the relationships presented in Table I. EDTA completely eliminated all FB conversion in control and all sample tubes. EGTA had little effect on zymosan-induced conversion but reduced MSUM-induced conversion to control levels. Factor B conversion in the control tube was not eliminated by EGTA, but in the present work was below accurately definable levels (i.e., 43%). Subsequent work using refined techniques confirms that EGTA reduces FB conversion in the control tube by about 75%, but EDTA eliminates it altogether. Asbestos activation was reduced to half (7.6%) of the nonchelated levels. Again, fiber type had no obvious relation to activation. DISCUSSION The data clearly demonstrate the adsorption of IgG from whole serum by asbestos fibers, particularly by the two positively charged chrysotile fibers, which adsorbed amounts comparable to those bound by CPPD and hydroxyapatite crystals. The amount bound by the negatively charged amphibole fibers studied was smaller but measurable. The potential physiologic consequences of this phenomenon are not clearly defined. Adsorbed IgG may serve to enhance phagocytosis by macrophages, and with other serum proteins could protect against crystal-induced membrane lysis.4p7, 23-25Ad sorption of protein by silica has been proposed by Alliston et al. j to protect against membrane lysis by crystals until after phagocytosis has occurred, when surface protein can be degraded by lysosomal enzymes allowing lysosomal membrane lysis. This concept has been extended to crystal-in-
VOLUME NUMBER
Binding
64 4
duced arthritis by Wallingford and McCarty7 and may be of more general applicability. It has been suggested, and not yet disproven, that adsorption of IgG by crystals may activate the complement system. l’, lx In the aforementioned data, complement activation did not correlate well with simple binding of IgG by asbestos. A role for IgG in C3 activation is not ruled out, as adsorption with configurational change may be the crucial phenomenon. The electrophoretic mobility of the adsorbed IgG is of interest and gives insight into the mechanism of adsorption. Amphibole asbestos, like MSUM, CPPD, and apatite, has a negative surface charge’ and adsorbed only cationic or most of the positively charged IgG. Chrysotile asbestos has a positive surface charge and it adsorbed only anionic or more negatively charged IgG. The data obtained in this study lend further support to the hypothesis of electrostatic adsorption. Chrysotile asbestos has a positive charge of greater magnitude than the smaller negative charge of amphibole asbestos.’ This difference may explain the greater amounts of IgG adsorbed by chrysotile asbestos compared to the amphibole fibers, but this remains speculative. Activation
by asbestos
297
that some classical pathway activity is also present. Activation did not appear to be related to the asbestos preparation studied, and (as with IgG adsorption) was within the range observed with CPPD and hydroxyapatite crystals. The inflammatory properties of C3, C5, and later complement components are recruited by both pathways of complement activation. The alternative pathway, however, cannot recruit Cl, C2, or C4, which have biologic properties of their own. There may be biologic functions of the early steps in the alternative pathway that are not well understood at present. The possible effects on acute or chronic inflammation by differences in pathways of complement activation are not known but may prove a fruitful avenue for further investigation. 1 wish to thank Drs. Edward D. Harris, Jr., and George M. Bemier for reviewing the manuscript, and also gratefully acknowledge the expert technical assistance of Jeanie
Hahn and the secretarial assistanceof Nancy Marshall. REFERENCES I.
of complement
Activation of FB per se cannot be used as evidence that complement activation has been initiated by the alternative pathway. Factor B may be activated as a result of classical pathway activation through the amplification loop. Because FB activation by asbestos was only partially inhibited by EGTA, it may be assumed that much, but not all, FB activation was initiated by proximal events in the alternative pathway. In their pioneering work in this area, Wilson et al.26 originally suggested that asbestos may activate complement by the alternative pathway. These investigators used electrophoretic conversion (Grabar and Williams) to detect activation of FB. As outlined above, this method will not allow differentiation between classical or alternative pathway initiation of FB conversion. A glutathione-sensitized human erythrocyte assay was also used to measure alternative pathway activity. However, because of the direct “membranolytic” properties of chrysotile asbestos, interpretation of the results was not straightforward and required special controls with heat-inactivated and fresh human serum for the chrysotile but not the amphibole fibers. Chemotactic factor was generated by exposure to fresh human serum of all asbestos fibers tested. The contribution of the classical pathway was not studied.” The results presented here extend and confirm the original findings of Wilson et al., and demonstrate
of immunoglobulin
2. 3. 4.
5.
6 7.
8. 9.
IO.
11.
12.
13
Harington JS, Allison AC, Badami DV: Mineral fibers: Chemical, physiochemical and biological properties. Adv Pharmacol Chemother 12291, 1975. Speil S, Leineweber JP: Asbestos minerals in modern technology. Environ Res 2: 166, 1969. Kellermeyer RW, Naff GB: Chemical mediators of inflammation in acute gouty arthritis. Arthritis Rheum l&765, 1975. McCarty DJ, Kozin F: An overview of cellular and molecular mechanisms in crystal induced inflammation. Arthritis Rheum 18 (suppl.):757, 1975. Allison AC, Harrington JS, Birbeck M: An examination of the cytotoxic effects of silica on macrophages. J Exp Med 124:141, 1966. Macnab G, Harington JS: Hemolytic activity of asbestos and other mineral dusts. Nature 214522, 1967. Wallingford WR, McCarty DJ: Differential membranolytic effects of microcrystalline sodium urate and calcium pyrophosphate dihydrate. J Exp Med 133: 100, 1971. Light WG, Wei ET: Surface charge and asbestos toxicity. Nature 265531, 1977. Kellermeyer RW, Breckenridge RT: The inflammatory process in acute gouty arthritis. I. Activation of Hageman factor by sodium urate crystals. J Lab Clin Med 65307, 1965. Webster ME, Maling HM, Zweig MH, Williams 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. 1972. 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, 1973. Kozin F, McCarty DJ: Protein adsorption to monosodium urate, calcium pyrophosphate dihydrate and silica crystals. Arthritis Rheum 19:433, 1976. Hasselbacher P, Schumacher HR: Immunoglobulin in tophi and on the surface of monosodium mate crystals. Arthritis Rheum 21:353, 1978.
298
Hasselbacher
14. Hasselbacher P: Binding of IgG by monosodium urate monohydrate is a function of the ionic charge of the protein. Clin Res 26:623A, 1978. 15. Hasselbacher P: Activation of C3 by monosodium urate, potassium urate, and steroid crystals. Arthritis Rheum 21:565, 1978. 16. Ginsberg MH, Kozin F: Mechanisms of cellular interaction with monosodium urate crystals. Arthritis Rheum 21:558, 1978. (Abst.) 17. Timbre11 V, Gilson JC, Webster I: UICC standard reference samples of asbestos. Int J Cancer 3~406, 1968. 18. Hasselbacher P: Activation of C3 by monosodium urate monohydrate and other crystals. Arthritis Rheum 22:571, 1979. 19. Hasselbacher P: Binding of IgG and complement proteins by monosodium urate monohydrate and other crystals. J Lab Clin Med. (In press.) 20. Hasselbacher P: Immunoelectrophoretic assay for synovial fluid C3 with correction for globulin. Arthritis Rheum 22~243, 1979. 21. Fine DP, Marney SR, Colley DG, Sergent JS, Des Prez RM:
.I ALLERGY
22. 23.
24.
25.
26.
CLIN. IMMUNOL. OCTOBER 1979
C3 shunt activation in human serum chelated with EGTA. J Immunol 109:807, 1972. Hasselbacher P: IgG adsorbed by sodium urate and other crystals is not homogeneous. Clin Res 27:327A, 1979. Byers PH, Ward PA, Kellermeyer RW, Naff GB: Complement as a mediator of inflammation in acute gouty arthritis. II. Biological activities generated from complement by the interaction of serum complement and sodium urate crystals. J Lab Clin Med 81:761, 1973. 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. 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 24:33lA, 1976. Wilson M, Gaumer HR, Salvaggio JE: Activation of the alternative complement pathway and generation of chemotactic factors by asbestos. J ALLERGYCLIN IMMUNOL60~218, 1977.
M.D. Hanover,
and by
N. H.
Asbestos fibers adsorbed IgG from whole human serum in amounts comparable to that taken up by crystals of calcium pyrophosphate or hydroxyapatite, but less than that adsorbed by crystals of monosodium urate monohydrate. Chrysotile asbestos was much more active in binding IgG than was amphibole asbestos. The electrophoretic mobility of adsorbed IgG was a ,function qf the surface charge of the crystal studied. Negatively chargedJibers adsorbed cationic IgG, and positively charged fibers adsorbed anionic IgG. Complement activation by asbestos, as judged by electrophoretic conversion of C3, was similar for all fiber types tested. Properdin factor B was also activated in the presence of asbestos. Activation of both properdin factor B and C3 was only partially inhibited by ethylene glycol tetraacetic acid, demonstrating that activation was occurring through both the classical and alternative pathways.
The diseases associated with asbestos exposure are becoming increasingly important as the magnitude of this public health problem becomes more widely appreciated. The mechanisms by which asbestos fibers cause tissue injury or neoplastic transformation are uncertain.lp2 If a comparison with other forms of silicosis or crystal-induced arthritis is justified, the pathogenesis of asbestos-related disease is likely to be multifactorial.3-4 Factors such as particle size, ability to undergo phagocytosis by macrophages, and membrane lysis of cells or lysosomes have been considered.5-8 Recent studies of both silicosis and crystal-induced arthritis have suggested that surface adsorption of immunoglobulin or other proteins and activation of the complement pathways may play a
role in modulating the biologic activity of the crystals.‘-16 In this report the IgG-binding and complement-activating properties of asbestos fibers will be quantified and compared with those of other crystals of biologic interest. The pathways by which complement activation is initiated will also be described. METHODS
AND MATERIALS
International Union Against Cancer (UICC) standard samplesof five asbestospreparationswere generously provided by Dr. V. Timbrell.” These preparations were washed 10 times by centrifugation and resuspension in phosphate-buffered saline (0.14 M NaCl, 0.01 M sodium phosphatepH = 7.4) (PBS). Washing causeda loss of material, presumably the finest particles; therefore unwashed fibers were also tested. Washed and dried fibers were suspended in PBS, 25 mgiml. Monosodium urate monohydrate
From the Dartmouth Medical School and Dartmouth-Hitchcock Arthritis Center. This work was supported by a Clinical Investigator award (7KO8-AM-00587-02) from the National Institutes of Health, United States Public Health Service Grant AM 14780, grants from the National and New Hampshire Chaptersof the Arthritis Foundation, and a Multipurpose Arthritis Center grant from the National Institutes of Health. This work was presentedin part at the 1979National Meeting of the American Federationfor Clinical Research,Washington, D. C. Received for publication Feb. 9, 1979. Accepted for publication May 2, 1979. Reprint requeststo: Peter Hasselbacher,M.D., Connective Tissue Disease Section, Dartmouth Medical School, Hanover, NH 03755.
Vol. 64, No. 4, pp. 294-298
(MSUM) and calcium pyrophosphate dihydrate (CPPD) were preparedas previously described.lx Zymosan and calcium hydroxyapatite were obtained from Sigma Chemical Co., St. Louis, MO.
Adsorption
of IgG
The amount and nature of IgG adsorbed by crystals were determined as previously described with minor modifications.‘a, ” Unwashed asbestos fibers (75 mg) were suspended in 0.5 ml fresh human serum in a l.S-ml polypropylene centrifuge tube with constant inversion for 30 min at room temperature. The tubes were centrifuged at 15,600 x g (Eppendorf centrifuge model 5412) for 3 min. and the supematants removed. The asbestos pellet was
0091-6749/79/100294+05$00.50/0
0 1979 The C. V. Mosby
Co.
VOLUME NUMBER
64 4
Binding
of immunoglobulin
by asbestos
295
FIG. 1. IgG eluted from washed, serum-exposed asbestos fibers was studied by REP. Wells a to f contain simple dilutions of a control serum of 1 : 20 to 1 :400. Well 7, anthophyllite; 2, Canadian chrysotile B; 3, crocidolite; 4, Rhodesian chrysotile A; 5, amosite. The positive electrode is marked.
washed 10 times by centrifugation and resuspension in PBS. The washed crystals were eluted for 30 min at room temperature with 1 ml PBS, with concentration at 1 M with respect to NaCI. The IgG in the eluate was studied by rocket immunoelectrophoresis (RIEP) as previously described.” The nonasbestos crystals were treated in the same way except that 150 mg of crystals were suspended in 0.5 ml serum, and the crystals were washed five times following serum exposure.
Activation
of complement
Complement activation was measured by determining the amount of electrophoretic conversion of C3 and properdin factor B (FB), as previously described.is Washed crystals were suspended in fresh human serum (2.3 mg/ml) with constant inversion for 60 min at 37” C. Complement activation was terminated by adding ethylenediamino tetraacetate (EDTA) (0.01 M). The supematant serum was appropriately diluted in electrophoresis buffer and studied for conversion products by crossed immunoelectrophoresis. Antiserum to C3 was produced as previously described,20 and antiserum to FB was the generous gift of Dr. Carlos Arroyave. Activation of C3 was estimated by measuring the areas under the &-C and Pi-A curves produced by crossed immunoelectrophoresis. Conversion was expressed as the percent of the total area under the two precipitin arcs represented by /3,-A. Similarly, activation of FB was judged by the percent of total area appearing under the y-fragment arc. It should be emphasized that neither of these measurements
corresponds exactly to the percent of total component protein converted. The effect of adding EDTA (0.005 M) or ethylene glycol tetraacetic acid (EGTA) (0.010 M) to the incubation tube was studied to determine the pathway of initiation of complement activation. EGTA binds calcium preferentially when compared with magnesium, thus inhibiting the classical pathway of complement activation while allowing the alternative pathway to function.”
RESULTS Adsorption
of IgG
When IgG was studied by RIEP as described, two “rockets ” formed, one in the direction of each electrode, reflecting the heterogeneity of charge of IgG molecules. No IgG was detected in any of the final wash supernatants. The eluates of all five asbestos preparations contained IgG. Table I lists the amount of IgG eluted from asbestos and other crystals. The amounts varied widely among the asbestos types, with the two chrysotile preparations adsorbing amounts approximating those adsorbed by crystals of CPPD or hydroxyapatite. The binding by amphibole preparations was considerably less as judged by eluted IgG. Of interest was the electrophoretic mobility of the eked protein (Fig. 1). IgG adsorbed by the nonasbestos material$* and by amphibole asbestos
296
J. ALLERGY
Hasselbacher
TABLE I. Adsorption activation
of IgG and complement
by crystals
Type of crystal
PBS Rhodesian chrysotile A Canadian chrysotile B Anthophyllite Crocidolite Amosite Zymosan MSUM CPPD Apatite
C3 activation (o/o))+
FB activation (o/o)*
IgG eluted (pglmg)
Control
EGTAt
Control
EGTA
1.54
25 38
II 20
8 13
O$ 6
I .83
34
I.5
13
8
0.10 0.39
42 37
22 18
IO
0.14 -
34
21
75 61
61 8 -
16 13 12 69 27 -
7 7 55 O$ -
-
-
-
3.65 I .54
1.15
*C3 and FB activations were quantitated by measuring the percent of total area beneath the P,-A and y-precipitin curves, respectively. $EGTA = 0.01 M. $.ln these samples, similar precipitin lines formed in the y-region, accompanied by a small decrease in area of the a-peaks. Because no arcs were present, this amount of activation was too low for quantitation. No precipitin lines were formed in EDTA-treated samples.
was restricted to cationic species of IgG moving toward the negative electrode. IgG eluted from the chrysotile fibers was restricted to anionic species. When electrophoresis was performed at neutral pH the same pattern was observed. Activation
of C3
Table 1 lists the percent of area under the /3,-A curve of C3, or the y-curve of properdin FB for control and for EGTA-chelated sera. In the saline control tube, 25% of the C3 area was beneath the Pi-A curve. This represents complement activation by the tube wall, which is considerably less if the tubes are not inverted. Extensive experience in this laboratory has demonstrated that the coefficient of variation of the percent C3 area converted is 0.029 for multiple samples studied on the same plate, and on the same day. The mean conversion by the five asbestos preparations was 37.0%, which represents a highly significant difference. This was small compared to C3 conversion caused by MSUM (67%) or zymosan (75%) but was comparable to C3 conversion by CPPD or hydroxyapatite. In the presence of EGTA, C3 conversion by asbestos was reduced by about half, to a mean of 19.2%. EGTA had little effect on C3 conversion by zymosan, but reduced conversion by MSUM to control levels, as reported elsewhere.‘6
CLIN. IMMUNOL OCTOBER 1979
EGTA also decreased C3 conversion in the control tube from 25% to 11%. This proportion has been confirmed in other experiments and represents a component of alternative pathway activation by the polypropylene tube wall. EDTA completely eliminated C3 conversion in the control and all sample tubes. There did not appear to be any relationship between type of asbestos fiber and C3 activation or effect of EGTA. Unwashed asbestos fibers converted similar amounts of C3 but were not studied further. Activation
of factor
B
Generation of the y-fragment of FB by zymosan (69% area) or MSUM (27%) was much greater than that by asbestos (mean, 13.4%) or control (8%). An increase in the area of the y-fragment correlated well with a decrease in the area of the a-component. Although my experience with the FB conversion system is not as extensive as with C3 (due to limited supply of antiserum), the reproducibility of FB conversion analysis appears to be the same. Comparison of individual pairs of crystals on smaller plates confirms the relationships presented in Table I. EDTA completely eliminated all FB conversion in control and all sample tubes. EGTA had little effect on zymosan-induced conversion but reduced MSUM-induced conversion to control levels. Factor B conversion in the control tube was not eliminated by EGTA, but in the present work was below accurately definable levels (i.e., 43%). Subsequent work using refined techniques confirms that EGTA reduces FB conversion in the control tube by about 75%, but EDTA eliminates it altogether. Asbestos activation was reduced to half (7.6%) of the nonchelated levels. Again, fiber type had no obvious relation to activation. DISCUSSION The data clearly demonstrate the adsorption of IgG from whole serum by asbestos fibers, particularly by the two positively charged chrysotile fibers, which adsorbed amounts comparable to those bound by CPPD and hydroxyapatite crystals. The amount bound by the negatively charged amphibole fibers studied was smaller but measurable. The potential physiologic consequences of this phenomenon are not clearly defined. Adsorbed IgG may serve to enhance phagocytosis by macrophages, and with other serum proteins could protect against crystal-induced membrane lysis.4p7, 23-25Ad sorption of protein by silica has been proposed by Alliston et al. j to protect against membrane lysis by crystals until after phagocytosis has occurred, when surface protein can be degraded by lysosomal enzymes allowing lysosomal membrane lysis. This concept has been extended to crystal-in-
VOLUME NUMBER
Binding
64 4
duced arthritis by Wallingford and McCarty7 and may be of more general applicability. It has been suggested, and not yet disproven, that adsorption of IgG by crystals may activate the complement system. l’, lx In the aforementioned data, complement activation did not correlate well with simple binding of IgG by asbestos. A role for IgG in C3 activation is not ruled out, as adsorption with configurational change may be the crucial phenomenon. The electrophoretic mobility of the adsorbed IgG is of interest and gives insight into the mechanism of adsorption. Amphibole asbestos, like MSUM, CPPD, and apatite, has a negative surface charge’ and adsorbed only cationic or most of the positively charged IgG. Chrysotile asbestos has a positive surface charge and it adsorbed only anionic or more negatively charged IgG. The data obtained in this study lend further support to the hypothesis of electrostatic adsorption. Chrysotile asbestos has a positive charge of greater magnitude than the smaller negative charge of amphibole asbestos.’ This difference may explain the greater amounts of IgG adsorbed by chrysotile asbestos compared to the amphibole fibers, but this remains speculative. Activation
by asbestos
297
that some classical pathway activity is also present. Activation did not appear to be related to the asbestos preparation studied, and (as with IgG adsorption) was within the range observed with CPPD and hydroxyapatite crystals. The inflammatory properties of C3, C5, and later complement components are recruited by both pathways of complement activation. The alternative pathway, however, cannot recruit Cl, C2, or C4, which have biologic properties of their own. There may be biologic functions of the early steps in the alternative pathway that are not well understood at present. The possible effects on acute or chronic inflammation by differences in pathways of complement activation are not known but may prove a fruitful avenue for further investigation. 1 wish to thank Drs. Edward D. Harris, Jr., and George M. Bemier for reviewing the manuscript, and also gratefully acknowledge the expert technical assistance of Jeanie
Hahn and the secretarial assistanceof Nancy Marshall. REFERENCES I.
of complement
Activation of FB per se cannot be used as evidence that complement activation has been initiated by the alternative pathway. Factor B may be activated as a result of classical pathway activation through the amplification loop. Because FB activation by asbestos was only partially inhibited by EGTA, it may be assumed that much, but not all, FB activation was initiated by proximal events in the alternative pathway. In their pioneering work in this area, Wilson et al.26 originally suggested that asbestos may activate complement by the alternative pathway. These investigators used electrophoretic conversion (Grabar and Williams) to detect activation of FB. As outlined above, this method will not allow differentiation between classical or alternative pathway initiation of FB conversion. A glutathione-sensitized human erythrocyte assay was also used to measure alternative pathway activity. However, because of the direct “membranolytic” properties of chrysotile asbestos, interpretation of the results was not straightforward and required special controls with heat-inactivated and fresh human serum for the chrysotile but not the amphibole fibers. Chemotactic factor was generated by exposure to fresh human serum of all asbestos fibers tested. The contribution of the classical pathway was not studied.” The results presented here extend and confirm the original findings of Wilson et al., and demonstrate
of immunoglobulin
2. 3. 4.
5.
6 7.
8. 9.
IO.
11.
12.
13
Harington JS, Allison AC, Badami DV: Mineral fibers: Chemical, physiochemical and biological properties. Adv Pharmacol Chemother 12291, 1975. Speil S, Leineweber JP: Asbestos minerals in modern technology. Environ Res 2: 166, 1969. Kellermeyer RW, Naff GB: Chemical mediators of inflammation in acute gouty arthritis. Arthritis Rheum l&765, 1975. McCarty DJ, Kozin F: An overview of cellular and molecular mechanisms in crystal induced inflammation. Arthritis Rheum 18 (suppl.):757, 1975. Allison AC, Harrington JS, Birbeck M: An examination of the cytotoxic effects of silica on macrophages. J Exp Med 124:141, 1966. Macnab G, Harington JS: Hemolytic activity of asbestos and other mineral dusts. Nature 214522, 1967. Wallingford WR, McCarty DJ: Differential membranolytic effects of microcrystalline sodium urate and calcium pyrophosphate dihydrate. J Exp Med 133: 100, 1971. Light WG, Wei ET: Surface charge and asbestos toxicity. Nature 265531, 1977. Kellermeyer RW, Breckenridge RT: The inflammatory process in acute gouty arthritis. I. Activation of Hageman factor by sodium urate crystals. J Lab Clin Med 65307, 1965. Webster ME, Maling HM, Zweig MH, Williams 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. 1972. 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, 1973. Kozin F, McCarty DJ: Protein adsorption to monosodium urate, calcium pyrophosphate dihydrate and silica crystals. Arthritis Rheum 19:433, 1976. Hasselbacher P, Schumacher HR: Immunoglobulin in tophi and on the surface of monosodium mate crystals. Arthritis Rheum 21:353, 1978.
298
Hasselbacher
14. Hasselbacher P: Binding of IgG by monosodium urate monohydrate is a function of the ionic charge of the protein. Clin Res 26:623A, 1978. 15. Hasselbacher P: Activation of C3 by monosodium urate, potassium urate, and steroid crystals. Arthritis Rheum 21:565, 1978. 16. Ginsberg MH, Kozin F: Mechanisms of cellular interaction with monosodium urate crystals. Arthritis Rheum 21:558, 1978. (Abst.) 17. Timbre11 V, Gilson JC, Webster I: UICC standard reference samples of asbestos. Int J Cancer 3~406, 1968. 18. Hasselbacher P: Activation of C3 by monosodium urate monohydrate and other crystals. Arthritis Rheum 22:571, 1979. 19. Hasselbacher P: Binding of IgG and complement proteins by monosodium urate monohydrate and other crystals. J Lab Clin Med. (In press.) 20. Hasselbacher P: Immunoelectrophoretic assay for synovial fluid C3 with correction for globulin. Arthritis Rheum 22~243, 1979. 21. Fine DP, Marney SR, Colley DG, Sergent JS, Des Prez RM:
.I ALLERGY
22. 23.
24.
25.
26.
CLIN. IMMUNOL. OCTOBER 1979
C3 shunt activation in human serum chelated with EGTA. J Immunol 109:807, 1972. Hasselbacher P: IgG adsorbed by sodium urate and other crystals is not homogeneous. Clin Res 27:327A, 1979. Byers PH, Ward PA, Kellermeyer RW, Naff GB: Complement as a mediator of inflammation in acute gouty arthritis. II. Biological activities generated from complement by the interaction of serum complement and sodium urate crystals. J Lab Clin Med 81:761, 1973. 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. 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 24:33lA, 1976. Wilson M, Gaumer HR, Salvaggio JE: Activation of the alternative complement pathway and generation of chemotactic factors by asbestos. J ALLERGYCLIN IMMUNOL60~218, 1977.