Interaction of anti-DNA antibodies with synthetic polyanionic antigens

Journal of Autoimmunity(1989) 2,85 l-859

Interaction of Anti-DNA Antibodies with Synthetic Polyanionic Antigens

Kathryn Weber,* Elizabeth Hawruk_t and Vincent A. DeBari*t$ *The Rheumatology Laboratory and The Lupus Erythematosus Treatment and Research Center at StJoseph’s Hospital and Medical Center, Paterson, NY; ?Rheumatology Section, Department of Medicine, Seton Hall University, School of Graduate Medical Education, South Orange, NJ, USA Autoantibodies to DNA (anti-DNAab), found primarily in systemic lupus erythematosus (SLE), cross-react with a variety of antigens. The binding of these antibodies to naturally occurring mucopolysaccharides such as heparan and chondroitin sulfates has led to the suggestion that antiDNAab have specificity for a polyanionic epitope. In this study, to avoid the use of endogenous immunogens to which humans may have become sensitized, we have used the synthetic polyanions, dextran sulfate (DS), polyvinyl sulfate (PVS) and the semi-synthetic antigen, pectic acid (PA) to evaluate this hypothesis using SLE sera (n = 15) and sera from healthy individuals (controls; n= 15, age and sex matched to SLE group). Inhibition of binding with lz51-DNA was optimal at 1 mg/ml for DS and PVS, and resulted in significant inhibition of binding by both SLE and control sera of native DNA (P-c0.01, each group); no inhibition was observed with PA, nor was a significant inhibition observed with any antigen on binding of SLE or control sera groups to denatured DNA. We conclude that while on quantitative grounds reaction of anti-DNAab with polyanions may not be clinically relevant, it is clear that, unlike polycarboxylates (e.g. PA), polysulfated polymers,.whether aliphatic, as in the case of PVS, or glycosidic, such as DS, react with a subpopulation of anti-DNAab in such a manner as to block significantly the ability of these antibodies to bind DNA.

Introduction

Autoantibodies directed against DNA (anti-DNAab), found primarily in systemic lupus erythematosus (SLE) [ 11, cross-react with a large array of non-DNA $To whom all correspondence should be addressed 703 Main Street, Paterson, NJ 07503, USA.

at: St Joseph’s

Hospital

and Medical

Center,

851 0896-841

l/89/060851

+09 $03.00/O

0 1989 Academic

Press Limited

852

K. Weber et al.

antigens [2, 31. The interaction of these antibodies with certain epitopes, e.g. phosphodiester moieties, can be expected to lead to binding of molecules, such as phospholipids, which contain these epitopes [4]. However, in addition to such more or less direct associations, a number of cases exist where reactivity of anti-DNAab to macromolecules, e.g. vimentin [5, 61 or anionic dyes such as Cibacron blue FSGA and Procion red HE3B [7,8] is observed. Although it has been suggested, on quantitative grounds, that cross-reactivity phenomena between populations of anti-DNAab and non-DNA antigens are not physiologically relevant [9], an understanding of the relationship between these antibodies and the molecular structure at various epitopes may lead to a better understanding of both the underlying mechanism of the development of autoantibodies of SLE (and perhaps other autoimmune states) and the biochemical events contributing to circulating immune complex deposition in these diseases. Of particular interest are reports of the binding of anti-DNAab with mucopolysaccharides such as chondroitin sulfate and heparan sulfate [lo, 111, leading to the suggestion that the polyanionic nature of DNA is not only responsible for its reaction with anti-DNAab, but that the presence of the latter glycosaminoglycan on the glomerular basement membrane is an important factor in the pathogenesis of lupus nephritis. In this report we evaluate the polyanion theory by exploring the inhibition of DNA-anti-DNA binding by polyanion antigens. We investigate this by using sera from SLE patients with anti-DNAab as well as from healthy control subjects. However, rather than study binding to endogenously occurring polyanions, we include in our study those antigens to which subjects are not likely to have been previously exposed, i.e. the synthetic polyanions, polyvinyl sulfate and dextran sulfate and the semi-synthetic polyanion, pectic acid. Materials and methods Materials Serum specimens Sera from 15 patients with SLE who met the diagnostic criteria of the American Rheumatism Association for SLE [ 121 were chosen for this study. Additionally, all sera tested positive for anti-DNAab by the Crithidia Zuciliaeassay [ 131. Sera were also obtained from 15 healthy subjects, age and sex matched to the patient group, and served as normal controls. All sera were stored at - 70°C. This study was approved by the Institutional Review Board of St Joseph’s Hospital and Medical Center. Synthetic antigens The polyanions used in this study were dextran sulfate (DS) (Sigma Chemical Corp., St Louis, MO), polyvinyl sulfate (PVS), and pectic acid (PA) (Aldrich Chemical Co., Milwaukee, WI). Dextran (Dex) (Sigma) served as a nonionic control material. Methods Radiometric binding (Farr) assay of anti-DNAab A modification [ 14, 151 of the Farr assay [ 161, using kits obtained from Amersham Corp. (Arlington Heights, IL), was used to study DNA binding and its inhibition by antigen. In this system, lz51-DNA was used either untreated for the measurement of

Antibodies to Polyanions in SLE

853

100

0 .G D 2 s

90

-

80

-

70

-

(0)

Anti-nDNA

(b)

Ab

Anti-dDNA

Ab

a._

60-

\ \

50-

\ \

40-

\

30

-

20

-

IO -5

-4

-3

-2

-I

0

Log inhibitor

I

-5

concentration

-4

-3

-2

-I

0

I

(mg/ml)

Figure 1. Dose-dependent inhibition of DNA binding by preincubation with varying concentrations of DS and autologous antigen, i.e. nDNA in (a) and dDNA in (b). Open and filied circles are two SLE sera; open and filled squares are two control sera; solid lines connect points for DS and dashed lines for nDNA and dDNA. (A) Inhibition of binding to nDNA and (b) Inhibition of binding to dDNA.

anti-native DNAab (anti-nDNAab) or was denatured by placing the solution in a 100°C water bath for 15 min with subsequent immersion in an ice bath to prevent reannealing [ 1’71. Binding to the resultant single-stranded DNA is attributed to antibodies to denatured DNA (anti-dDNAab). ‘251-DNA bound in the ammonium sulfate precipitate was counted in a gamma counter (Isoflex, MicroMedics Systems, Inc., Horsham, PA). Farr assay inhibition

Prior to applying the Farr assay as described above, sera were preincubated solutions for 1 h at 37°C. Statistical

with ag

analyses

Databases, maintained on Lotus 123 (Lotus Development Corp., Cambridge, MA), were transferred to and analyzed with software from Human Systems Dynamics, Inc., Northridge, CA (PC Statistician) for analysis of variance (ANOVA) and Student’s paired t-test [ 181.

Results Dose-response

of inhibitor

Inhibition by polyanion antigens was evaluated using DS over a concentration range of 10e2 to 10 mg/ml with both SLE and control sera. These data are presented in Figure 1. The inhibition of binding of anti-nDNAab is shown in Figure l(a) for sera from two SLE subjects and from two controls. The latter, with very low levels

854

K. Weber et al.

of anti-nDNAab, show a modest degree of inhibition at 1 mg/ml. At this same concentration, the SLE sera demonstrate a roughly 75% inhibition of binding by anti-nDNAab. In Figure l(b), binding of anti-dDNAab by one control serum is maximally inhibited at 1 mg/ml; at this concentration, the inhibition of the other control serum is nearly 50% while the SLE sera show an approximately 2040% inhibition. Based on these data, it was decided to extend the study to additional sera and to the other polyanion antigens at a concentration of 1 mg/ml. In both Figure l(a) and (b), comparison is made with inhibition by the autologous antigen, i.e. nDNA and dDNA respectively. For SLE serum, a comparable degree of binding inhibition is achieved at a concentration of DNA roughly two orders of magnitude lower than DS. This effect is less pronounced in the case of the control subjects. An anomalous ‘hook’ is observed in the control sera at higher concentration (1 mg/ml) of dDNA and may be due to a change in the stoichiometry of binding for some population of antibodies.

Statistical overview of polyanion preincubation studies The entire SLE and control serum panels were subjected to preincubation with 1 mg/ml DS, PVS, PA and Dex in both the anti-nDNAab protocol (i.e. with doublestranded DNA) and the anti-dDNA protocol for binding by single-stranded DNA. The data were subjected to ANOVA in the following groupings: anti-nDNAab, controls; anti-nDNAab, SLE; anti-dDNAab, controls and anti-dDNA, SLE. Ail ANOVA were highly significant (P< 0.0001 for each group) and the data for the four groups, presented in Tables 1 and 2, are further explored in the following sections.

Inhibition of anti-nDNAab

by polyanions (Table 1)

Significant inhibition of nDNA binding by sera from controls is observed in the case of both DS and PVS (PO.l). Dex appears to significantly (PO.l). In this group of sera, also, Dex increases apparent binding (P< 0.001).

Lack of inhibition of anti-dDNAab

by polyanions (Table 2)

Among the control sera, as a group, DS and PVS have no overall significant effect on binding (P>O.l for each); however, on an individual basis, eight of 15 (53%) sera preincubated with DS and six of 15 (40%) preincubated with PVS demonstrated

Antibodies to Polyanions in SLE

855

Table 1. Inhibition of anti-nDNA bindingfrom sera of healthy controlsubjects (n = 15) and SLE patients (n = 15). Compounds studied were dextran sulfate (DS), polyvinyt sulfate (PVS), pectic acid (PA) and dextran (DEX). Binding was studied using the modified Farr assay Number (%)I

o/0Binding3

P (2~scontrol)

Controls Control (no inhibitor) Postinhibition by DS Postinhibition by PVS Postinhibition by PA Postinhibition by DEX

NA2 15 (100) 13 (87) 7 (47) l(7)

16.66k5.06 8.64+ 1.94 10.4752.46 17.24k4.19 39.29 f 19.93

NA O.l
SLE Control (no inhibitor) Postinhibition by DS Postinhibition by PVS Postinhibition by PA Postinhibition by DEX

NA 11(73) 14 (93) 6 (40) 0 (0)

71.66+ 14.15 61.20f22.31 54.25 f 24.71 74.77 i 14.61 80.93 + 10.66

NA O.l
Experiment

INumber (%) = number of sera demonstrating inhibition; parentheses give percent of sera demonstrating inhibition. ZNA,not applicable. ‘Mean k 1 SD.

Table 2. Lack of inhibition of anti-dDNA

binding (modi$ed Farr assay) from sera of healthy control subjects. Subjects and abbreviations are the same as in Table 1 Number (%)’

o/0Binding3

P (w control)

Controls Control (no inhibitor) Postinhibition by DS Postinhibition by PVS Postinhibition by PA Postinhibition by DEX

NA’ 8 (53) 6 (40) 2 (13) 0 (0)

37.85 f 6.05 31.35-& 12.26 34.29k7.71 62.37 + 16.87 67.53 f 11.44

NA >O.l >O.l
SLE Control (no inhibitor) Postinhibition by DS Postinhibition by PVS Postinhibition by PA Postinhibition by DEX

NA l(7) l(7) l(7) 0 (0)

70.55 + 13.64 74.85f 11.07 75.04* 12.04 80.21+ 5.75 84.61 k4.88

NA
Experiment

‘Number (%) = number of serademonstratinginhibition; parenthesesgivepercent of sera demonstrating inhibition. 2NA, not applicable. ‘Mean f 1 SD.

856 K. Weber et al. inhibition. Both PA and Dex increase the apparent binding significantly (P < 0.001 for both). Among the SLE sera DS and PVS promoted a slight, but significant, increase P
Antibodies to Polyanions in SLE

857

Figure 2. Representation of structural similarity between DNA and PVS. Carbon, oxygen, sulfur and phosphorous atoms are respectively indicated by black, white, diagonally-hatched and cross-hatched spheres. A conformer of a two-sulfate fragment of PVS (left) is compared with a fragment of DNA containing two phosphodiester bonds in the conformation of nDNA. Note the similarity between approximate interatomic distances of the tetrahedral sulfur and phosphorous atoms and the topological prominence of these anionic groups. Drawings are from models built using a scale molecular models kit (Cochranes of Oxford Ltd. Leafield, Oxford, UK).

DNA and the differences between these and the polycarboxylated PA molecule. As in the DNA phosphodiester linkage, both PVS and DS contain an ‘inorganic’, i.e. non-carbon ester bond. Moreover, the size of the sulfur and phosphorous atoms are quite similar and are considerably larger than the carbon atom of the carboxylate group [23]. The structural similarities between conformers of DNA and PVS are represented in Figure 2. Curiously, no inhibitory effect is seen on the ability of these sera to bind denatured DNA. Anti-dDNAab have been somewhat controversial with respect to their pathological significance in SLE [24,25]. However, the inability of any of the polyanions to inhibit the binding to denatured DNA further strengthens the concept that polysulfates mimic the antigenicity of native DNA. The observation of apparent increases in binding by Dex and, in some cases, by PA should also be addressed. The enhancement of immune complex precipitation by polyols such as polyethylene glycol [ 19,261 and Dex [ 191 is well known. In fact, Dex was used in this study as a ‘polymer’ control to establish more clearly the effect observed with the polyanions studied. In light of the inordinately high binding observed in control subjects upon incubation with DEX, these increases should probably be regarded as artifactual. Acknowledgements

We thank Mrs L. B. Keil for database management and statistical analyses and Messrs Rajiv Biswal and Tony Wu for technical assistance, especially in the early stages of this work. Portions of this work were presented at the conference ‘Immunointervention in Autoimmune Diseases’, Paris, June 1517,1988, and were supported by a grant from the Lupus Erythematosus Foundation of New Jersey.

858

K. Weber et al.

References 1. Tan, E. M., P. H. Schur, R. I. Carr, and H. G. Kunkel. 1966. Deoxyribonucleic acid (DNA) and antibodies to DNA in the serum ofpatients with systemic lupus erythematosus. J. Clin. Invest. 45: 1732-1740 2. Schwartz, R. S. and D. B. Stollar. 1985. Origins of anti-DNA antibodies. J. Zmmunol. Methods 75: 321-327 3. Pisetsky, D. 1987. Mechanisms of antinuclear antibody production in the rheumatic diseases. In Rheumatic Disease Clinics of North America. D. Pisetsky, R. Snyderman, eds. W. B. Saunders, Philadelphia. pp. 569-581 4. Lafer, E. M., J. Rauch, C. Andrzejewski, D. Mudd, B. Furie, R. S. Schwartz, and D. B. Stollar. 1981. Polyspecific monoclonal lupus autoantibodies reactive with both polynucleotides and phospholipids. J. Exp. Med. 153: 897-909 5. Andre-Schwartz, J., S. K. Datta, Y. Shoenfeld, D. A. Isenberg, B. D. Stollar, and R. S. Schwartz. 1984. Binding of cytoskeletal proteins by monoclonal anti-DNA lupus autoantibodies. Clin. Zmmunopathol. 31: 261-271 6. Cairns, E., R. Komar, and D. Bell. 1986. Cytoskeletal binding of monoclonal anti-DNA antibodies derived from tonsillar lymphoid cells of a normal subject. Arthritis Rheum. 29: 1351-1358 7. Pollard, M. K. and J. Webb. 1982. Partial purification of anti-DNA antibodies by dyeligand chromatography. J. Zmmunol. Methods 54: 81-94 8. Emlen, W. and G. Burdick. 1983. Purifications of DNA antibodies using cibacron blue F3GA affinity chromatography. J. Zmmunol. Methods 62: 205-215 9. Edberg, J. C. and R. I’. Taylor. 1986. Quantitative aspects of lupus anti-DNA autoantibody specificity.3. Zmmunol. 136: 45814587 10. Faaber, l?., P. J. A. Capel, T. P. M. Rijke, G. Vierwinden, L. B. A. Van De Putte, and R. A. P. Koene. 1986. Cross-reactivity of anti-DNA antibodies with proteoglycans. Clin. Exp. Zmmunol. 55: 502-508 11. Faaber, P., T. P. M., Rijke, L. B. A. Van De Putte, P. J. A. Capel, and J. H. M. Berden. 1986. Cross-reactivity of human and murine anti-DNA antibodies with heparan sulfate. 3. Clin. Invest. 77: 1824-1830 12. Tan, E. M., A. S. Cohen, J. F. Fries, A. J. Masi, D. J. McShane, N. F. Rothfield, J. G. Schaller, N. Talal, and R. J. Winchester. 1982. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum. 25: 1271-1277 13. Aarden, L. A., E. R. DeGroot, and T. E. W. Feltkamp. 1975. Immunology of DNA. III. Crithidia luciliae, a simple substrate for the determination of anti-ds DNA with the immunofluorescence technique. Ann NY Acad Sci USA 254: 505-5 15 14. Belmonte, P. A., L. B. Keil, and V. A. DeBari. 1987. Characteristics of antibodies adsorbed on the DNA immunoadsorbent, agarose poly-1-lysine-DNA. Diag. Clin. Zmmunol. 5: 144-150 15. Holian, J., T. D. Griffiths, D. N. Glass, R. N. Maini, and J. T. Scott. 1975. Human antiDNA antibody: Reference standards for diagnosis and management of systemic lupus erythematosus. Ann. Rheum. Dis. 34: 438-443 16. Wold, R. T., F. E. Young, E. M. Tan, and R. S. Farr. 1968. Deoxyribonucleic Acid antibody: A method to detect its primary interaction with deoxyribonucleic acid. Science 161: 806-807 17. Eaton, R. B., G. Schneider, and P. H. Schur. 1983. Enzyme immunoassay for antibodies to native DNA. Arthritis Rheum. 26: 52-62 18. Bahn, A. K. Basic Medical Statistics. Grune and Stratton 1972, New York. pp. 144-200 19. Hellsing, K. 1969. Immuno reactions in polysaccharide media. Polysaccharide-enhanced precipitin reactions with antigens of various sizes. Biochem. 3.114: 145-149 20. Alpert, S. D., J. Koide, S. Takada, and E. G. Engleman. 1987. T-Cell regulatory disturbances in the rheumatic diseases. In Rheumatic Disease Clinics of North America. D. Pisetsky and R. Snyderman, eds. W. B. Saunders, Philadelphia. pp. 431445 21. Davidson, A., R. Halpern, and B. Diamond. 1986. Speculation of the role of somatic mutation in the generation of anti-DNA antibodies. Ann. NY Acad. Sci. USA 475: 174-180

Antibodies to Polyanions in SLE

859

22. Ebling, F. and II. H. Hahn. 1980. Restricted subpopulations of DNA antibodies in kidneys of mice with systemic lupus: Comparison of antibodies in serum and renal eluates. Arthritis Rheum. 23: 392-403 23. Douglas, B. E. andD. H. McDaniel. 1965. Covalent substances. In Concepts andModels of Inorganic Chemistry. Blaisdell, Waltham. pp. 41-98 24. Barnett, E. V. 1979. Anti-DNA antibodies: Perspectives and perplexities. J. Zmmunol. Methods 27: l-7 25. Emlen, W., D. S. Pisetsky, and R. P. Taylor. 1986. Antibodies to DNA: A perspective. Arthritis Rheum. 29: 1417-1426 26. Chesebro, B. and S. E. Svehag. 1968. Precipitation of human serum proteins by polyethyleneglycol. Clin. Chem. Acta. 20: 527-529