Immunology Letters 71 (2000) 111–115

Antigen binding characteristics of antibodies against hydroxylradical modified thymidine monophosphate

Rizwan Ahmad, Khurshid Alam 1, Rashid Ali *Department of Biochemistry, Faculty of Medicine, J.N. Medical College, A.M.U. Aligarh-202 002, India

Received 19 October 1999; accepted 30 November 1999

Abstract

Reactive oxygen species (ROS) are formed in living organisms during normal metabolic reactions as well as under differentenvironmental stresses. In this study, thymidine monophosphate (TMP) was exposed to hydroxyl radical (OH) and challenged inrabbits. TMP and OH-modified TMP were found to be nonimmunogenic. The TMP was linked to bovine serum albumin (BSA)by carbodiimide reaction, and then modified with the OH. The neoantigens, TMP–BSA, and ROS–TMP–BSA conjugatesinduced highly specific antibodies against immunogens. Induced antibodies exhibited appreciable cross-reactivity with variouspolynucleotides and nucleic acids. In this respect, the induced antibodies resembled the diverse antigen-binding characteristics ofnaturally occurring systemic lupus erythematosus (SLE) anti-DNA autoantibodies. © 2000 Elsevier Science B.V. All rightsreserved.

Keywords: Hydroxyl radical; TMP–BSA and ROS–TMP–BSA conjugates; Anti-DNA antibodies; SLE

www.elsevier.com/locate/

1. Introduction

Hydroxyl radical (OH) is one of the most reactive ofoxygen free radicals (OFR) capable of interacting withany cellular component it encounters. The chemistry ofDNA damage by several reactive oxygen species (ROS)has been elucidated in vitro [1,2]. Different ROS affectDNA differently; for example, superoxide anion radicaland hydrogen peroxide do not react at all with DNAbases, while OH generates a multiplicity of productsfrom them [3]. Of the five DNA components, thymineand cytosine are most susceptible to OH attack, fol-lowed by adenine, guanine and deoxyribose [4]. There is

evidence that OFR are involved in the etiology of manychronic health problems such as emphysema, cardiovas-cular and inflammatory diseases, cataracts and cancer[5].

Nucleosides and nucleotides by themselves arenot capable of inducing antibodies unless theyare coupled to a carrier protein such as BSA by perio-date or carbodiimide [6]. The latter reagents utilize thephosphate group without cleaving the ribose–deoxyribose ring. The induced antibodies against nucle-otide–BSA conjugate show specificity for DNA orRNA, depending on the type of sugar ring present[7–10].

In the present communication, ROS-modifiedthymidine 5%-monophosphate (ROS–TMP) has beenlinked to bovine serum albumin (BSA) and challengedinto rabbits for the induction of antibodies. The in-duced antibodies have been characterized for their fineantigenic binding specificity towards various nucleicacid conformers so as to gain insight into the unknownmechanism of anti-DNA autoantibody production insystemic lupus erythematosus (SLE).

Abbre6iations: BSA, bovine serum albumin; OH, hydroxyl radical;SLE, systemic lupus erythematosus; TMP, thymidine 5%monophos-phate.

* Corresponding author: Present address: Department of Biochem-istry, College of Science, PO Box 2455, Riyadh 11451 Saudi Arabia.Tel.: +966-1-4675943; fax: +966-1-4675791.

E-mail address: rali@ksu.edu.sa (R. Ali)1 Present address; Department of Pharmacology, College of Phar-

macy, PO Box 2457, Riyadh 11451, King Saud University, SaudiArabia.

0165-2478/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.PII: S 0 1 6 5 -2478 (99 )00177 -7

R. Ahmad et al. / Immunology Letters 71 (2000) 111–115112

2. Methods

2.1. Materials

TMP, BSA, anti-rabbit IgG alkaline phosphataseconjugate, Coomassie Brilliant Blue G-250 and R-250,Tween 20, Freunds’ complete and incomplete adju-vants, polydeoxyribonucleotides, carbodiimide and p-nitrophenyl phosphate were purchased from SigmaChemical Company, MO. Polystyrene microtiter flat-bottom ELISA plates and modules were products ofNUNC, Denmark.

2.2. ROS modification of TMP

Aqueous solution of TMP (0.31 mM in PBS) wasirradiated under 254 nm UV light (Vilber Lourmat,France) for 30 min at RT in the presence of H2O2 (3.1mM). Samples of TMP exposed to H2O2 or UV lightalone served as corresponding controls. The UV ab-sorption characteristics of the samples were recorded ona Shimadzu UV 240 spectrophotometer.

2.3. Preparation of antigen

The TMP was covalently linked to BSA by thecarbodiimide method [6]. Aqueous solutions of BSA (19mg) and TMP (50 mg) were mixed and pH adjusted to7.5 using NaOH, followed by the addition of 20 mg ofcarbodiimide, in a total volume of 5 ml. The reactionmixture was incubated in darkness at RT for 24 h.After incubation, the sample was dialyzed extensivelyagainst PBS (pH 7.4).

2.4. ROS-modification of TMP–BSA conjugate

Aqueous solution of TMP–BSA conjugate (6.2 mMin PBS) was irradiated with 254 nm of UV light for 30min at RT in the presence of H2O2 (62 mM). ExcessH2O2 was removed by extensive dialysis against PBS(pH 7.4).

2.5. Polyacrylamide gel electrophoresis

TMP–BSA and ROS–TMP–BSA conjugates, mixedwith one-tenth volume of tracking dye (0.025% xylenecyanole, 30% Ficoll 400, 0.5 M EDTA in 10× Tris-glycine buffer), were loaded on 7.5% nondenaturingpolyacrylamide gel (with 2.5% stacking gel) and elec-trophoresed for 8–10 h at 24 mA in Tris-glycine buffer(25 mM Tris, 192 mM glycine, pH 8.3).

2.6. Immunization schedule

Rabbits were immunized with TMP, TMP–BSA andROS–TMP–BSA conjugates. The antigens (100 mg per

animal) were emulsified with Freunds’ complete adju-vant (for first injection only) and administered intra-muscularly at multiple sites. Subsequent injections weregiven in Freunds’ incomplete adjuvant. Each animalreceived a total of 600 mg antigen during the course ofimmunization. Blood was collected by cardiac punc-ture; serum was decomplemented by heating at 56°Cfor 30 min and stored at −80°C with 0.1% sodiumazide. Preimmune blood was collected prior to immu-nization and processed as above.

2.7. Enzyme-linked immunosorbent assay

ELISA was carried out on polystyrene plates [11]with slight modification. The microtiter wells werecoated with 100 ml of antigens (10 mg/ml) for 2 h at37°C and overnight at 4°C. Each sample was coated induplicate, and half of the plate served as control devoidof only antigen coating. Unbound antigen was washedthrice with TBS-T (20 mM Tris, 150 mM NaCl pH 7.4,containing 0.05% Tween-20), and unoccupied sites wereblocked with 1.5% BSA (in TBS; 10 mM Tris, 150 mMNaCl, pH 7.4) for 4–6 h at 4°C. Plates were washedonce with TBS-T. The wells were now coated with 100ml of sera (1:100 diluted in TBS) or varying amounts ofIgG, and incubated for 2 h at 37°C, and overnight at4°C; the plates were extensively washed with TBS-T.The bound antibodies were assayed by anti-rabbit-IgGalkaline phophatase conjugate using p-nitrophenylphosphate as substrate. The results were expressed asthe mean of the difference of absorbency values in testand control wells. The antibody specificity was deter-mined by competition ELISA [12].

3. Results

3.1. Spectral characteristics of ROS–TMP

The UV absorption spectra of hydroxyl radicalmodified TMP indicated a complete loss of characteris-tic absorption at 267 nm (lmax) after 30 min of irradia-tion. A generalized spectrum getting stronger towardsshorter wavelength was observed. No appreciablechange in TMP absorption spectra was noticed whenirradiated for 30 min in the absence of hydrogen perox-ide (Table 1).

3.2. Conjugation of TMP to BSA

The UV absorption of TMP–BSA conjugate showeda slight bathochromic shift when compared with TMPalone (Fig. 1).The formation of TMP–BSA conjugatewas ascertained on 7.5% nondenaturing PAGE. Thebanding pattern indicated the formation of high molec-ular-weight species as a consequence of conjugation

R. Ahmad et al. / Immunology Letters 71 (2000) 111–115 113

(Fig. 2). UV analysis of TMP–BSA conjugate gave asharp peak at lmax (269 nm), whereas ROS-modifiedconjugate indicated substantial hypochromicity and al-most a complete disappearance of UV absorption max-imum (data not given).

3.3. Antibodies against TMP–BSA andROS–TMP–BSA conjugate

Induced antibodies were found to be highly specificagainst respective immunogens, as shown by competi-tion ELISA (Fig. 3). Fifty percent inhibition of anti-body binding was observed at an inhibitorconcentration (IC50) of :2.5 mg/ml (with TMP–BSAconjugate) and 3.5 mg/ml (with ROS–TMP–BSAconjugate). Fig. 2. Banding pattern of TMP–BSA conjugate in 7.5% nondenatur-

ing polyacrylamide gel. Representative lanes contain 30 mg BSA (lane1) and increasing amounts of TMP–BSA conjugate (60–90–120 mg,corresponding to lanes 2, 3 and 4).

Table 1Effect of hydroxyl radical on UV absorption characteristics of TMP

Component Absorption at lmax (267 nm)

H2O2 0.0TMP 2.58TMP-UV (30 min) 2.65TMP-UV-H2O2 (15 min) 1 18TMP-UV-H2O2 (30 min) 0.90

Fig. 3. Competition ELISA of anti-TMP–BSA IgG and anti-ROS–TMP–BSA IgG with TMP–BSA (–x–) and ROS–TMP–BSA con-jugates (–o–).

Fig. 1. Ultraviolet absorption profile of TMP (–·–·–), BSA(· · · · · · ·), TMP–BSA conjugate (– – –) and the nonconjugated mix-ture of TMP–BSA (—). All samples were extensively dialyzedagainst PBS before recording the spectra.

3.4. Immuno-crossreacti6ity of induced antibodies

The cross reactivity of induced anti-TMP–BSA anti-bodies was investigated by competitive binding assayusing naturally occurring DNA, nucleotides, andpolynucleotides as inhibitors. Calf thymus DNA and itsROS-modified form showed moderate inhibitions(Table 2). ROS forms of TMP, thymine and thymidineshowed inhibitions of 69%, 82% and 56%, respectively.Homopolymers like poly(G) and ROS-poly(G) showedrespective inhibitions of 35% and 48%. Cross reactivityof anti-ROS–TMP–BSA antibodies was also evalu-ated. A maximum of 89% inhibition in antibody bind-ing was observed with TMP–BSA conjugate (Table 3).TMP and thymidine showed 39% and 23% inhibition;their ROS forms showed maximum inhibitions of 86%and 56%, respectively. ROS-thymine and ROS-guanineshowed 81% and 78% inhibition. Poly(G) showed amoderate inhibition of 38%, whereas ROS-poly(G) in-hibited 53% antibody binding.

R. Ahmad et al. / Immunology Letters 71 (2000) 111–115114

4. Discussion

ROS, generated during various metabolic and bio-chemical reactions, have multifarious effects, includingoxidative DNA damage resulting in the initiation anddevelopment of various human diseases [13]. Experi-mentally induced antibodies against ROS-modifiedDNA exhibits polyspecificity [14], recognizing B-, A-and allied conformations of DNA [15]. Thymine ishighly susceptible to hydroxyl radical modification andforms products such as thymine glycol (Tg) and 5-hy-droxymethyl uracil. These products have been detectedin human urine, thereby indicating that they are formedin vivo, most probably as a consequence of oxidativeinsult to DNA [3].

TMP and ROS-modified TMP were found to benonimmonogenic. Our results correlate well with earlierfindings that nucleotides by themselves are nonim-munogenic [9]. However, nucleotides may be conju-gated to carrier proteins by the carbodiimide methodwithout cleaving the ribose/deoxyribose sugar moiety toenhance their immunogenicity [6].

Native DNA is known to be a poor immunogen[16,17], whereas RNA–DNA hybrid, synthetic riboho-mopolymers of poly(I), poly(U), poly(G) and calf-thy-

Table 3Competitive inhibition data of anti-ROS–TMP–BSA IgG with differ-ent inhibitorsa

Inhibitor Inhibitor concentra-Maximum percentinhibition at 20 mg/ml tion for 50%

inhibition

3.5ROS–TMP–BSA 95conjugate

89 4.0TMP–BSA con-jugate

TMP 39 –86ROS–TMP 923Thymidine –56ROS-thymidine 4.5

Calf-thymus 33 –DNA

47 –ROS-ctDNA16Thymine –

4.581ROS-thymine55Guanine 10

2.578ROS-guanine29Adenine –

ROS-adenine 48 ––38Poly G

ROS-poly G 53 –Poly(dA-dT) –22

.poly(dA-dT)35 –Poly(dA-dU)

.poly(dA-dU)35 –Poly(dI-dC)

.poly(dI.dC)

a The inhibition with cytosine, ROS-cytosine; poly I, ROS-poly I;and poly(A).poly(U) was less than 20%.

Table 2Competitive inhibition data of anti-TMP–BSA IgG with differentinhibitorsa

Maximum percent in-Inhibitor Inhibitor concentra-hibition at 20 mg/ml tion for 50% inhibi-

tion

98TMP–BSA con- 2.5jugate

42 –ROS–TMP–BSAconjugate

TMP 28 –1569ROS–TMP

ROS–TMP–BSA 42 –16Thymidine –56ROS-thymidine 18

Calf thymus –33DNA

48ROS-ctDNA –21Thymine –

5.5ROS-thymine 82–Guanine 46

61ROS-guanine 17–Adenine 29

ROS-adenine 6.559Poly G –35

48 –ROS-poly G–36Poly(dA-dT)

.poly(dA-dT)22 –Poly(dA-dU)

.poly(dA-dU)

a The inhibition with cytosine, ROS-cytosine, polyG, ROS-polyG,poly I, ROS-poly I, mitochondrial DNA, ROS-mtDNA,poly(A).poly(U), and poly(dI-dC).poly(dI-dC) was less than 20%.

mus DNA modified with drugs, free radicals, etc., havebeen reported to induce antibodies [14,18–20]. Anti-bodies to dihydrothymidine were successfully elicited inrabbits when conjugated to BSA [21]. The present studydemonstrated the immunogenicity of TMP–BSA andROS–TMP–BSA conjugates. The competition ELISAdata indicates the specificity of immune IgG towardsthe ROS modified conformers, probably by the recogni-tion of the common epitopes. It may be pointed outthat experimentally-induced antibodies to purines/pyrimidines conjugated to BSA are quite specific, butdo cross-react with nucleic acids from various sources[22].

The diverse antigen-binding characteristics of the in-duced antibodies with a variety of polynucleotides andtheir ROS-modified forms might be due to the recogni-tion of either the phosphodiester backbone or neo-epi-topes on the molecules. The chances of neo-epitoperecognition by induced antibodies appears to be domi-nating in view of the practical nonrecognition of un-modified nucleic acids and their derivatives by inducedantibodies.

In conclusion, the study demonstrated the immuno-genicity of TMP–BSA conjugate and its ROS modified

R. Ahmad et al. / Immunology Letters 71 (2000) 111–115 115

form. The induced antibodies, although highly specificagainst respective immunogens, exhibited cross reactiv-ity with various nucleic acid conformers resembling thebinding characteristics of naturally occurring SLE anti-DNA antibodies.

Acknowledgements

This work was supported in part by a research grant[37(980)/98/EMR-II] to KA from the Council of Scien-tific and Industrial Research, New Delhi.

References

[1] H.C. Box, H.G. Freund, E. Budzinski, J.C. Waliace, A.E. Mac-cabin, Free radical induced case lesions, Radiat. Res. 141 (1995)91–94.

[2] B. Epe, in: B. Halliwell, O.I. Arouma (Eds.), DNA and FreeRadicals, Ellis Horwood, Chichester, UK, 1993, pp. 41–65.

[3] B. Halliwell, O.I. Arouma, DNA damage by oxygen derivedspecies: its mechanism and measurement in mammalian systems,FEBS Lett. 281 (1991) 9–19.

[4] R.L. Saul, D. Gee, B.N. Ames, in: H.R. Warner, R.N. Butler,R.L. Sprott, E.L. Schneider (Eds.), Modern Biological Theoriesof Aging, Raven Press, 1987, pp. 113–129.

[5] L.J. Machlin, A. Bendich, Free radical tissue damage: protectiverole of antioxidant nutrients, FASEB J. 1 (1987) 441–445.

[6] M.J. Halloran, C.W. Parker, The preparation of nucleotide–protein conjugates: carbodiimide as coupling agent, J. Immunol.96 (1966) 373–378.

[7] Y.N. Vaishnav, A. Antony, Antibodies raised against denaturedDNA bind to double-stranded DNA, J. Immunol. Methods 118(1989) 25–30.

[8] Y.N. Vaishnav, A. Antony, Base-specific binding of adenosineantibodies to double-stranded DNA, Immunol. Invest. 19 (1990)373–384.

[9] K.C. Kala, A. Antony, Effect of nucleic acid reactive antibodieson tumor cells grown in vivo, Immunol. Invest. 25 (1996)321–331.

[10] K.C. Kala, A. Antony, Inhibition of in vitro aminoacylation andtranslation by RNA reactive antibodies, Immunol. Invest. 22(1993) 117–126.

[11] S. Aotsuka, M. Okawa, K. Ikebe, R. Yokohari, Measurement ofdouble strand antibodies in major immunoglobulin classes, J.Immunol. Methods 28 (1979) 149–162.

[12] R. Hasan, A. Ali, R. Ali, Antibodies against DNA-psoralencrosslinks recognize unique conformation, Biochim. Biophys.Acta 1073 (1991) 507–513.

[13] B. Halliwell, C.E. Cross, Oxygen-derived species: their relationto human disease and environmental stress, Environ. HealthPerspect. 102 (1994) 5–12.

[14] K. Alam, A. Ali, R. Ali, The effect of hydroxyl radical on theantigenicity of native DNA, FEBS Lett. 319 (1993) 66–70.

[15] K. Alam, R. Ali, Human autoantibodies binding to multipleconformations of DNA, Biochem. Int. 26 (1992) 597–605.

[16] B.D. Stollar, Antibodies to DNA, CRC. Crit. Rev. Biochem. 20(1986) 136.

[17] M.P. Madaio, S. Hodder, R.S. Schwartz, B.D. Stollar, Respon-siveness of autoimmune and normal mice to nucleic acid anti-gens, J. Immunol. 132 (1984) 872–876.

[18] D.K. Garg, R. Ali, Reactive oxygen species modifiedpolyquanylic acid: immunogenicity and implications for systemicautoimmunity, J. Autoimmunity. 11 (1998) 371–378.

[19] Z. Arif, R. Ali, Antigenicity of poly(dA-dT).poly(dA-dT) pho-tocrosslinked with 8-methoxypsoralen, Arch. Biochem. Biophys.329 (1996) 191–198.

[20] K. Alam, R. Ali, Human anti-DNA autoantibodies and inducedantibodies against ROS-modified-DNA show similar antigenicbinding characteristics, Biochem. Mol. Biol. Int. 47 (1999) 881–890.

[21] K. Hubbard, I. Hiroshi, B.F. Erlanger, S.S. Wallace, Character-ization of antibodies to dihydrothymidine, a radiolysis productof DNA, Biochemistry 28 (1989) 4382–4387.

[22] D.W. Ballard, E.W. Voss, Base specificity and idiotype of anti-DNA autoantibodies reactive with synthetic nucleic acids, J.Immunol. 135 (1985) 3372–3380.

.