Network Theory in Autoimmunity
`
`IN VITRO SUPPRESSION OF SERUM ANTI-DNA ANTIBODY
`BINDING TO DNA BY ANTI-IDIOTYPIC
`
`ANTIBODY IN SYSTEMIC LUPUS ERYTHEMATOSUS
`
`NABIH I. ABDOU, HELEN WALL, HERBERT B. LINDSLEY, JOHN F. HALSEY, and TSUNEO
`SUZUKI, Department of Medicine, Division of Allergy, Clinical Immunology and
`Rheumatology, and Departments of Biochemistry and Microbiology, University
`of Kansas Medical Center, Kansas City, Kansas 66103; Veterans Administration
`Hospital, Kansas City, Missouri 64128
`
`AB s TR Ac T Regulation of serum anti-DNA antibody
`in systemic lupus erythematosus (SLE) by an antiidio(cid:173)
`typic antibody was evaluated. Various sera from SLE
`patients in active and inactive states of their disease, as
`well as sera from normal individuals, were first com(cid:173)
`pletely depleted of anti-DNA and of DNA by affinity
`chromatography. The suppressive capacity of equi(cid:173)
`molar concentrations of the various depleted sera
`(blocking sera) on target lupus sera were determined.
`The target sera were from lupus patients with known
`DNA-binding capacity. Blocking sera from inactive
`SLE suppressed the binding of autologous anti-DNA
`antibody to (3H)DNA (n = 19, P < 0.01). Blocking sera
`from active SLE (n = 19), as well as human serum al(cid:173)
`bumin, did not suppress. Sera from normal donors who
`had no contact with lupus patients or with lupus sera
`did not suppress (n = 14, P > 0.5), whereas those
`from normal donors who had contact with lupus pa(cid:173)
`tients or sera did suppress the binding (n = 5, P < 0.02).
`The anti-anti-DNA antibody suppressive activity in
`the inactive lupus serum was shown to be localized
`within the F(ab')2 portion of immunoglobulin (lg)G
`and could not be removed upon adsorption by normal
`human gammaglobulin. Furthermore, immune com(cid:173)
`plexes could be detected by a Clq binding assay when
`the inactive lupus blocking sera were incubated with
`the anti-DNA antibody containing target sera. The
`
`This work appeared in abstract form. (1980. ]. Allergy
`Clin . lmmunol. 65: 221; Clin. Res. 28: 338A.)
`Address all correspondence to Dr. N. I. Abdou, University of
`Kansas Medical Center, Division of Allergy, Clinical Im(cid:173)
`munology and Rheumatology, Room 416C, Kansas City, Kans .
`66103.
`Received for publication 26 October 1980 and in revised
`form 24 November 1980.
`
`specificity of the suppressive serum factor was shown
`by its inability to block the binding of tetanus toxoid to
`antitetanus antibody and its ability to block the binding
`of DNA to F(ab')2 fragments of active lupus lgG.
`Regulation of serum anti-DNA antibody levels by
`anti-antibodies could induce and maintain disease
`remission in lupus patients and prevent disease expres(cid:173)
`sion in normals.
`
`INTRODUCTION
`
`Regulation of antibody synthesis and of lymphocytes
`involved in the immune response has been proposed
`by Jerne (1) to be controlled by a network ofantibodies
`and lymphocytes. Antiidiotypic antibodies directed
`against cell-surface receptors or secreted idiotypic
`molecules have been shown to be important elements
`in transplantation tolerance or the specific suppression
`of an antibody response (2, 3). Antiidiotypic antibodies
`that recognize and regulate the expression of idiotypic
`determinants on the cell surface could theoretically
`play a key role in the induction of self-tolerance and the
`prevention of autoimmunity. Abnormalities in the idio(cid:173)
`type antiidiotype system could therefore lead to expres(cid:173)
`sion or expansion of autoreactive cell clones (4-6).
`Self-tolerance is also dependent on suppressor
`cells (7). Suppressor cell dysfunction could in part be
`responsible for autoantibody production in systemic
`lupus erythematosus (SLE) 1 (8, 9). In fact, there appears
`to be a close interplay between suppressor cells and
`the idiotypic network in the regulation of the immune
`response ( 10-12).
`In this study we have tested an extension of the net-
`
`1 Abbreviation used in this paper: SLE, systemic lupus
`erythematos us.
`
`]. Clin. Invest. ©The American Society for Clinical Investigation, Inc.
`Volume 67 May 1981 1297-1304
`
`· 0021-973818110511297/08 $1.00
`
`1297
`
`1 of 8
`
`BI Exhibit 1107
`
`

`

`work theory (1) with respect to modulation of the ex(cid:173)
`pression of autoantibody activity by presumed antiidio(cid:173)
`typic factors. We have demonstrated the presence of
`autoantiidiotypic antibody in sera of inactive SLE pa(cid:173)
`tients. In normal individuals who have had contact with
`lupus material, we found a cross-reacting antiidiotypic
`antibody against double-stranded DNA antibody. The
`effector activity is present in the F(ab'h portion of im(cid:173)
`munoglobulin (lg)G from sera of inactive SLE patients;
`it binds more avidly to autologous anti-DNA antibody
`than to antibody from unrelated donors. The blocking
`antibody could not inhibit an unrelated antigen-anti(cid:173)
`body reaction and could not be detected in sera of ac(cid:173)
`tive SLE patients or in sera of normal individuals not
`exposed to lupus sera.
`
`METHODS
`Patients and controls. 19 patients who satisfied the Ameri(cid:173)
`can Rheumatism Association preliminary diagnostic criteria
`for SLE ( 13) were studied. 19 normal healthy individuals with(cid:173)
`out personal or family history suggestive of an autoimmune
`state and with normal levels ( <6.4% binding) of serum anti(cid:173)
`DNA antibody were used as controls. 5 of the 19 normal in(cid:173)
`dividuals had contact with lupus patients and sera for varying
`periods of time (0.5-16 yr), and the other 14 normals had no
`contact with lupus material. The study was approved by the
`institution's human subjects committee and informed con(cid:173)
`sents were obtained from all of the subjects who entered the
`study. All patients were studied twice, when their disease
`was active and again during clinical remission. Patients were
`considered to have active disease if organ-specific clinical
`symptoms plus at least two of the following laboratory criteria
`were present: (a) erythrocyte sedimentation rate > 25 mm/h ;
`(h) total hemolytic complement CH:;-0 < 120 U; (c) DNA anti(cid:173)
`bodies> 14% binding. Patients were considered to have in(cid:173)
`active disease if no organ-specific clinical symptoms or signs
`could be elicited and if the laboratory criteria-erythro(cid:173)
`cyte sedimentation rate, CH 50 , DNA antibodies-were
`within the normal range. None of the patients was on cyto(cid:173)
`toxic drugs. Prednisone dosage received by patients dur(cid:173)
`ing active disease ranged from 5 to 6 mg/d (mean, 32.5 mg),
`and during inactive disease, from 0 to 40 mg/d (mean, 25 mg/d).
`Serum complement detennination ( CH50 assay) was done by
`a standard technique. The binding of sera to native DNA was
`studied by the Millipore filter radioimmunoassay (Millipore
`Corp., Bedford, Mass.) using human KB cell line [3HJDNA
`(Electro-Nucleonics, Inc., Fairfield, N. J.) (14) .
`Adsorption of anti-DNA antibody on DNA-cellulose
`columns. Calf thymus DNA-cellulose (Worthington Bio(cid:173)
`chemical Corp., Freehold, N. J.) was suspended in buffer
`(0.01 M Tris-HCI, 0.001 M EDTA, pH 7.4), and packed in
`columns
`(K9/15 columns, Pharmacia Fine Che micals ,
`Uppsala, Sweden). For each 2 g of DNA-cellulose (containing
`18 mg DNA), 10 ml of serum was allowed to pass through the
`column at 4°C at a rate of 2 drops/min. The effluents were
`passed again through the DNA-cellulose columns to insure
`complete removal of the anti-DNA antibody. Sera treated in
`this manner did not contain any detectable anti-DNA antibody
`(0% binding) when tested by radioimmunoassay (14). Cellu(cid:173)
`lose columns to which no DNA was coupled were incapable
`of depleting anti-DNA antibody.
`Treatment of DNA with immobilized DNAse. 6 or 60 U of
`DNAse-Sepharose conjugate (immobilized deoxyribonucleuse,
`Worthington Bioche mical Corp.) , suspended
`in 1.0 ml,
`
`was incubated with 10 µ.g [3H)DNA for 60 min at 37°C.
`The tubes were centrifuged at 720 g for 20 min, and 0.5 ml of
`the supernate was then dialyzed overnight against Tris-buffer
`saline. The DNA treated in this manner failed to bind to serum
`containing DNA antibodies. Thus, in a typical experiment,
`serum from an active lupus patient with 67% binding
`capacity (17,279 counts/min) to the undigested [3H)DNA
`failed to bind to the DNAse-treated [3HJDNA (<1% binding).
`6 U of DNAse-Sepharose conjugate was as efficient as 60 U.
`Therefore, in all the experiments reported in this paper 6 U
`of immobilized DNAse was used for the digestion of 1.0 ml
`of serum.
`Suppression of anti-DNA binding to [3H)DNA by blocking
`sera or immunoglobulin fragments and testing of precipitate
`formation by Clq-bi11di11g assay. All sera to be tested for the
`presence of anti-anti-DNA antibody (antiidiotypic or block(cid:173)
`ing antibodies) were depleted of anti-DNA antibody by pas(cid:173)
`sage twice through DNA-cellulose columns and then treated
`with 6 U of DNAse-Sepharose to digest DNA. In preliminary
`experiments, lupus sera with 90% DNA-binding capacity or
`with 10 µ.g DNNml could be completely depleted by this
`treatment. None of the blocking sera used in these experi(cid:173)
`ments had DNA-binding capacity> 90% or DNA > 10 µg/ml.
`Adequacy of depletion was confirmed by the failure to detect
`anti-DNA antibody by radioimmunoassay (14) and of DNA by
`chromatography (15) . The anti-DNA depleted and DNAse(cid:173)
`treated sera (blocking sera) were assayed for their capacity to
`inhibit the binding of (3H]DNA to sera from active lupus
`patients (target sera). For the blocking assay 100 µ.I contain(cid:173)
`ing 1 nmol of the blocking material IgG or its various frag(cid:173)
`ments was incubated with 100 µI of a target serum (contain(cid:173)
`ing l nmol lgG) at 37°C for l h and then for 16 h at 4°C.
`The mixtures were centrifuged at 1,000 g for 30 min; 100 µI
`of the supernate was collected and tested in the standard
`DNA-binding assay ( 14). The remaining 100 µ.I, designated the
`precipitate fraction, was tested in a conventional Clq binding
`assay (16).
`The percent suppression of DNA binding was calculated
`from the formula :
`
`DNA binding of mixtures of target
`l --~~~~~~~~~~~a1_1c_l_b_l_oc_·k_i_n~g_s_·e_ra)
`DNA binding of target sera alone
`
`(
`
`x 100.
`
`Depletion of various lg classes. Depletion of serum IgG,
`lgM, or lgA was performed by standard techniques as de(cid:173)
`scribed earlier ( 17). Adequacy of depletion was confirmed
`by immunoelectrophoresis and by immunodiffusion.
`Preparation of lgG , F( ah)' 2, and Fe fragments.
`lgG pro(cid:173)
`teins were isolated from serum hy affinity chromatography on
`Protein A-Sepharose 4B (Pharmacia Fine Chemicals, Uppsala,
`Sweden). F(ab')2 fragments produced by pepsin digestion of
`lgG proteins were separated from Fe-containing materials by
`passing over a column of Protein A-Sepharose 4B ( 18). Fab
`and Fe fragments, which were produced by papain digestion
`of lgG proteins, were separated also by Protein A-Sepharose
`4B chromatography (18). These IgG fragments were separately
`passed through a column of Sephadex G-150 to ensure the
`removal of undigested IgG proteins. lgG and its enzymatic
`cleavage fragments thus prepared were immunologically
`pure and distinct when examined by immunoelectrophore sis.
`Preparatio11 of F(ah ')2
`fragments from active
`lupu s
`sera. To ensure that the blocking activity of the antiidiotypic
`antibody is directed towards the binding sites of anti-DNA
`antibody, we prepared F(ab ' )2 fragments from IgG isolated
`from active lupus scrn. The isolation ofigG proteins on Protein
`A-Scpharose 4B and the preparation of F(ah ' )2 fragm e nts by
`pe psin digestion we re as described above.
`
`1298
`
`N. I. Abdou, H. Wall, H. B. Lindsley, ]. F. Halsey, and T. Suzuki
`
`2 of 8
`
`BI Exhibit 1107
`
`

`

`Prep11ratio11of11ornwl g1111111wglolmli11 i111111u11oadsorbents.
`To ensure the specificity of the antiidiotypic antibody, we
`attempted to deplete its blocking activity by passing it through
`normal gammaglolrnlin immunoadsorbent columns. Gamma(cid:173)
`globul ins were isolated from five different normal sera by 33%
`ammonium sulfate precipitation. The precipitate was washed,
`dialyzed, redissolved, and covalently coupled to CNBr(cid:173)
`activated Sepharose 48 according to the method described by
`:\larch et al. (19). Such affinity chromatography media were
`denoted as gammaglohulin immunoadsorhents. Aliquots of
`one antiidiotypic serum-prepared from inactive lupus serum
`as described above-were allowed to pass through the five
`different immunoadsorbents. The blocking activity of the
`antiidiotypic serum was tested before and after its pass<tge
`through the various imnrnnoadsorbents.
`He111agg/11ti11atio11 assay. To test for specificity of the
`autoantiidiotypic antibody, serum from a normal donor who
`had recently been boosted with tetanus toxoid was used as the
`target serum. Antitetanus antibody was assayed by the
`standard passive hemagglutination assay using chromium
`chloride to coat sheep erythrocytes with tetanus toxoid (20).
`Statistical 111111/ysis. The paired t test was used to compare
`suppression of target sera in the presenc:e or absence of
`blocking sera. For comparison of percent suppression with
`Clq binding, the Spearman rank correlation coefficient was
`calculated (21).
`
`RESULTS
`
`Blocking of anti-DNA binding. Autologous sera
`from lupus patients with inactive disease (n = 19) were
`found to suppress the binding of(3H]DNA to the target
`lupus sera ( P < 0.01) (Fig. 1, Table I). Blocking sera
`from active unrelated (11 = 9), from active autologous
`(11 = 19), or from inactive unrelated (n = 9) lupus pa(cid:173)
`tients were not capable of suppression. Human serum
`albumin at a similar protein concentration and
`processed similarly to the various blocking sera was
`also incapable of suppression (Fig. 1). The mean
`suppression value of the 19 various normal sera tested,
`when pooled together, was not significantly different
`from the percent DNA binding of the target lupus sera
`
`by themselves (P = 0.2) (Fig. 1, Table I). However,
`normal sera from donors who had contact with lupus
`patients and lupus blood components had significant
`suppressive activity on the target active lupus sera ( P
`< 0.02) (Tables I and II). Sera from normal donors who
`had no contact with lupus material did not suppress
`(P > 0.5) (Table II).
`Clq binding correlated with suppression of DNA
`binding. Precipitate fractions obtained from incubat(cid:173)
`ing F(ab')2 fragments with the corresponding autolo(cid:173)
`gous target sera were tested for their ability to bind 1251-
`Clq by radioimmunoassay. The upper limits of the 95%
`confidence intervals for individual values of fragments
`from active lupus sera are shown with dotted lines
`parallel to each axis (25% for suppression, 8% for Clq
`binding) (Fig. 2). Low Clq binding values (3- 7%) oc(cid:173)
`curred with sera and fragments from active lupus pa(cid:173)
`tients; higher Clq binding values (8-34%) occurred
`with those from patients with inactive lupus (Fig. 2).
`When samples from patients with active and inactive
`disease were considered together, percent suppres-
`sion correlated significantly with Clq binding (Spear(cid:173)
`man' s rho= 0.92, P < 0.01).
`Effects of immunoglobulin depletion of the blocking
`sera.
`In the five experiments performed on five dif(cid:173)
`ferent sera, depletion of IgG eliminated the suppres(cid:173)
`sive capacity of the autologous inactive lupus serum
`(Fig. 3). Depletion of lgM or of lgA failed to do so ( P
`< 0.01) .
`Failure of depletion of the blocking activity by ad(cid:173)
`sorption on normal human gammaglobulin. To
`avoid artefacts upon IgG depletion of blocking sera by
`immunoadsorbents, it is shown in Table III that normal
`gammaglobulin immunoadsorbents from five different
`donors failed to deplete the blocking activity of the
`lupus serum.
`Effects of lgG fragments on DNA binding.
`In the
`nine sera that were processed and tested, F(ab')2 frag(cid:173)
`ments and not Fe fragments of the inactive lupus sera
`were capable of suppressing the binding of anti-DNA
`% DNA BINDING OF LUPUS SERA
`0
`10
`20
`30
`40
`,___.....___.__ ___ __.. _ __.. __ antibody to (3H]DNA (P < 0.001) (Fig. 4). Fab frag-
`50
`60
`ments (P < 0.02), whole serum (P < 0.01), and globu(cid:173)
`lin fractions (P < 0.01) were also inhibitory.
`Effects of the blocking lgG on binding of F(ab')2
`fragments of the active lupus lgG to [3H]DNA.. To
`ensure that the blocking activity of the inactive autol(cid:173)
`ogous IgG is directed towards the binding sites of
`the anti-DNA antibody, we have prepared F(ab')2 frag(cid:173)
`ments from lgG of five different active lupus sera. It
`could be seen from Table IV that the blocking lgG in(cid:173)
`hibited the binding of the F(ab')2 fragments to [3H](cid:173)
`DNA. Fe fragments prepared from the same active
`lupus sera failed to bind to [3H]DNA in the absence or
`presence of the blocking lgG (not shown in Table IV).
`Effect of lgG fragments on tetanus toxoid binding.
`Whole serum, globulin fraction, or the various lgG frag-
`
`___ _, P< 0.01
`
`SERUM SOURCE .
`
`NONE
`
`NORMAL
`
`SLE ACTIVE UNRELATED
`
`SLE ACTIVE AUTOLOGOUS
`
`SLE INACTIVE UNRELATED
`
`SLE INACTIVE AUTOLOGOUS
`
`HUMAN SERUM ALBUMIN
`
`· DNA adsorbed and DNASE treated
`
`FIGURE l Suppression of anti-DNA binding to [3H]DNA by
`various sera. Results are the means±SD. 19 sera were tested
`for each of the normals, SLE active autologous, and SLE in(cid:173)
`active autologous groups. Nine sera were tested for each of
`SLE active unrelated and SLE inactive unrelated groups.
`
`Suppression of Anti-DNA Antibody by Antiidiotypic Antibody
`
`1299
`
`3 of 8
`
`BI Exhibit 1107
`
`

`

`TABLE I
`Serum DNA Binding before and after Treatment with the Blocking Serum
`
`Patient
`
`Predominant clinical features
`
`DNA binding of lupus sera
`
`After incubation with sera
`
`Before
`incubation
`
`Autologous
`inactive
`
`Normal
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`
`Mean
`
`Nephritis, cytopenia, CNS
`Nephritis
`Hemolytic anemia, cutaneous
`Thrombocytopenia, nephritis
`Serositis, cutaneous
`Arthritis, nephritis
`Fatigue, arthritis
`Cutaneous vasculitis
`Serositis
`CNS, nephritis
`Nephritis
`Fatigue, arthralgia
`Nephritis, arthralgia
`Thrombocytopenia, arthralgia
`Cytopenia, nephritis
`Serositis
`Nephritis, arthritis
`Nephritis, cutaneous
`Serositis, arthritis
`
`58
`46
`44
`53
`39
`58
`42
`43
`46
`52
`61
`45
`64
`31
`39
`30
`61
`35
`40
`47
`
`%
`
`10
`8
`18
`12
`13
`9
`14
`6
`10
`14
`11
`12
`20
`12
`20
`6
`14
`10
`12
`12
`
`23*
`39
`33
`43
`26
`19*
`27
`30
`25*
`39
`46
`29
`10•
`24
`18*
`23
`51
`25
`31
`30
`
`• Sera from donors who had contact with lupus material.
`
`ments of the same nine inactive lupus sera tested above
`for their anti-anti-DNA antibody activity did not inhibit
`the antitetanus antibody binding to tetanus toxoid as
`tested by a hemagglutination technique (Fig. 5).
`
`DISCUSSION
`The clonal selection theory has prevailed for many
`years and has suggested that the immune system is
`
`made up of lymphocyte clones capable of binding to a
`multitude of antigens (22). During ontogeny, self(cid:173)
`reactive (forbidden) clones were thought to be de(cid:173)
`stroyed and the surviving clones were believed to be
`directed mainly against nonself antigens (22). How(cid:173)
`ever, a number of recent important findings have re(cid:173)
`vealed new complexities. Self-reactive clones could be
`detected in normal individuals (23, 24). The discovery
`
`TABLE II
`Suppression of Anti-DNA Binding to [3H]DNA by Normal Sera
`
`Anti-DNA binding of target lupus sera§
`
`Normal sera•
`
`Number
`tested
`
`DNA antibody
`binding
`
`14
`5
`
`%
`
`3.4±2.9
`4.1±2.3
`
`Contact I
`with lupus
`material
`
`Binding
`
`Before
`blocking
`
`After
`blocking
`
`%
`
`No
`Yes
`
`44±13
`53±19
`
`33±12
`19±11
`
`Suppression
`
`p
`
`%
`
`25
`64
`
`>0.5
`<0.02
`
`* Normal healthy volunteers with negative personal or family history of lupus.
`t Contact with lupus patients and lupus blood components for 0.5-16 yr.
`§ Five different lupus sera were used as targets for suppression by the normal sera in all the experiments.
`Each biocking normal serum that had been adsorbed on DNA-cellulose columns and DNAse-treated was
`tested for its suppressive capacity of each of the target lupus sera.
`
`1300
`
`N. I. Abdou, H. Wall, H. B. Lindsley,]. F. Halsey, and T. Suzuki
`
`4 of 8
`
`BI Exhibit 1107
`
`

`

`100
`
`75
`
`c
`0
`·;;;
`"' ~
`c.
`c.
`" 50
`en
`c
`~
`0. 25
`
`Q)
`
`Q)
`
`0
`
`•
`•• •
`
`•Inactive SLE
`•Active SLE
`
`-------
`
`i
`• I
`I
`
`I
`I
`I
`I
`I
`I
`I
`I
`I
`I
`I
`I
`I
`I
`I
`---i---
`I
`• •I
`I
`4-
`I
`I
`I
`
`25
`Percent C lq Bound
`
`50
`
`FIGURE 2 Clq binding was measured on a precipitate fraction
`formed by the interaction of lupus F(ab') 2 with autologous tar(cid:173)
`get serum (see Methods). Percent suppression of DNA binding
`was determined on the same assay tubes. The upper limit of
`the 95% confidence interval, for the samples from patients
`with active disease only, are shown as dotted lines parallel to
`the corresponding axis. There was a significant correlation
`(Spearman's rho = 0.92, P < 0.01) between Clq binding and
`the degree of suppression of DNA binding. F(ab') 2 fragments
`from sera of active patients clustered in the lower left quad(cid:173)
`rant and were easily distinguished from those with inactive
`disease.
`
`of positive and negative interactions between T and B
`lymphocytes (7) and the possible involvement of idio(cid:173)
`types in clonal interactions ( 4) indicate that the immune
`system can recognize self and is regulated by a complex
`idiotypic network (1-5). Idiotypes and autoantiidio(cid:173)
`types coexist in the repertoire of a single individual;
`autoantiidiotypes can be induced or occur spon(cid:173)
`taneously during the immune response (4, 25-27).
`These antiidiotypic antibodies can exert either positive
`or negative influences on antibody biosynthesis or on
`effector cell function ( 10, 27).
`
`60
`
`% DNA BINDING OF LUPUS SERA
`10
`20
`30
`40
`50
`
`0
`
`-
`
`BLOCKING DEPLETION
`
`+
`+
`+
`
`lgG
`
`lgM
`
`I gA
`
`+
`FIGURE 3 Suppression of anti-DNA binding to [3H]DNA by
`autologous inactive lupus sera, and effects of depletion of vari(cid:173)
`ous
`immunoglobulin classes. Five different sera were
`processed and tested. Results are the means of all the experi(cid:173)
`ments. The standard deviation did not exceed 7% of the mean.
`
`FIGURE 4 Suppression of anti-DNA binding to [3 H]DNA
`by various immunoglobulin fragments of the inactive lupus
`serum. Nine different sera were processed and tested. Re(cid:173)
`sults shown are the means of all experiments. The stand(cid:173)
`ard deviation did not exceed 9.3% of the mean.
`
`Suppressio11 of Anti-DNA Antihody hy Antiidiotypic Antihody
`
`1301
`
`TABLE III
`Effects on Blocking Activity of Antiidiotypic Serum upon Its
`Adsorption hy Normal Human Gammaglohulin
`
`Gammaglohulin
`immunoads(>rhent
`from normal donors*
`
`Suppression of the target lupus serum t
`upon inc:uhation with blocking semm§
`
`~ ot adsorbed hy
`normal gamrnaglolmlin
`
`Adsorht~d by normal
`gammaglohulin
`
`1
`2
`3
`4
`5
`
`83
`83
`83
`83
`83
`
`80
`82
`79
`85
`83
`
`* Five different normal donors' gammaglobulin were linked to
`CnBr-activated Sepharose 4B. See Methods for details.
`t Target serum was from active lupus patient with 53%
`binding to [3H]DNA.
`§ Blocking serum was obtained from same donor of the target
`serum during disease inactivity. The blocking serum was first
`depleted of anti-DNA antibody and of DNA. Part of the de(cid:173)
`pleted blocking serum was adsorbed onto normal gamma(cid:173)
`globulin solid immunoadsorbents. See methods section for
`the calculation of percent suppression of the blocking activity.
`
`In this report we have examined the modulation of
`autoantibody activity by means of antiidiotypic anti(cid:173)
`bodies. We have demonstrated that bindirig of anti(cid:173)
`DNA antibody to DNA could be blocked by F(ab')2
`and Fab fragments of lgG obtained from autologous
`sera of inactive lupus patients (Fig. 4). Blocking ac(cid:173)
`tivity was probably due to occupancy of the combining
`site, since Fe fragments of the same lgG had no block(cid:173)
`ing activity. We have not ruled out, however, the possi-
`
`% DNA BINDING OF LUPUS SERA
`10
`20
`30
`40
`50
`
`0
`
`60
`
`BLOCKING MATERIAL
`
`NONE
`
`WHOLE SERUM -
`GLOBULIN FRACTION -
`Flab); -
`
`F lob)
`
`Fe
`
`5 of 8
`
`BI Exhibit 1107
`
`

`

`upon overnight incubation, since sera incubated with(cid:173)
`out the blocking material and processed in an identical
`manner had similar DNA-binding activity to that before
`incubation, The anti-anti-DNA activity could not be de(cid:173)
`tected in active lupus sera (Fig, 1), could not be ad(cid:173)
`sorbed by normal human gammaglobulin (Table III),
`could not block an unrelated antigen-antibody reac(cid:173)
`tion (Fig, 5), and was directed towards F(ab'h frag(cid:173)
`ments of the anti-DNA antibody (Table IV), Sera from
`normal donors who had contact with lupus patients and
`lupus blood components had anti-anti-DNA activity,
`indicating the probable presence of cross-reacting anti(cid:173)
`idiotypic antibodies in their sera (Table I), Specificity
`of the blocking activity of the normal sera for the Fab
`portion of lgG was not tested, Inhibition of anti-DNA
`binding by normal human serum has been observed
`previously (28),
`The factors responsible for the production of the
`cross-reacting (nonautologous) antiidiotypic anti(cid:173)
`bodies in the normal donors who had contact with
`lupus materials are unknown. This could reflect a
`regulatory mechanism in a normal protective immune
`response. Lymphocytotoxic antibodies and antinu(cid:173)
`clear antibodies (29), but not Sm antibodies (30), have
`been found in families of lupus patients. Laboratory
`personnel in contact with lupus materials have in(cid:173)
`creased levels of lymphocytotoxic antibodies (31), but
`not anti-double-stranded DNA antibodies (Table I).
`Failure to detect specific serum antibodies (anti(cid:173)
`DNA, anti-Sm) in family members and normals in con(cid:173)
`tact with lupus materials could be attributed to an effi(cid:173)
`cient regulation by an autoantiidiotypic antibody and/
`or the lack of a particular immune response gene for
`the development of the disease (32).
`The formation of immune precipitates upon binding
`of anti-DNA antibody to the antiidiotypic antibody was
`tested by the Clq binding assay. F(ab'h fragments from
`lgG of four patients with inactive disease clearly ex(cid:173)
`ceeded the calculated range for fragments from patients
`with active lupus (Fig. 2). We presume that the radio(cid:173)
`iodinated Clq is binding an immune precipitate, al(cid:173)
`though we have no direct proof that this is so. A solid(cid:173)
`phase Clq binding assay or Raji cell assay would permit
`direct evidence, since a radioiodinated anti-lgG anti(cid:173)
`body is used to detect an immune complex.
`The interplay and the complexity of the various regu(cid:173)
`latory mechanisms in autoimmunity have been dis(cid:173)
`cussed (4, 5, 33). In active SLE, suppressor T cells are
`deficient or dysfunctional (8, 34, 35). In inactive SLE,
`suppressor T cells are capable of inhibiting immuno(cid:173)
`globulin and anti-DNA antibody secretion (34). The
`same suppressor cells collected from inactive SLE pa(cid:173)
`tients are, however, incapable of affecting the number
`of DNA-binding autoreactive clones (34). Our pre-
`
`TABLE IV
`Blocking of the Binding of Active Lu11us F(ab'), Fragments
`to [3H]DNA by Autologous lgG
`
`["H]DNA binding to
`F(ab')2 fragments!
`
`Experiment*
`
`In absence of
`blocking lgG
`
`In presence of
`blocking lgG§
`
`Suppression 11
`
`%
`
`63
`41
`53
`34
`39
`
`19
`8
`9
`14
`12
`
`%
`
`70
`80
`83
`59
`69
`
`2
`3
`4
`5
`
`* Five different active lupus sera were tested.
`t F(ab') 2 fragments prepared from lgG fractions of the active
`lupus sera.
`§ Blocking IgG is obtained from autologous inactive lupus
`serum that was depleted of anti-DNA antibody and of DNA.
`1 Caleulated from the formula
`
`'
`
`_ binding in presence of blocking lgG) x lOO,
`binding in absence of blocking lgG
`
`(.
`
`
`1
`
`bility that blocking is due to anti-light chain activity,
`or due to DNA fragments present in the inactive
`lupus serum, We have ruled out the possibility that
`the blocking factor is due to rheumatoid factor ac(cid:173)
`tivity, since the former was capable of blocking the
`binding of F(ab'h fragments of the active lupus sera
`(Table IV), It is unlikely that the suppressed activity
`of the anti-DNA antibody was due to its aggregation
`
`RECIPROCAL OF HEMAGGLUTINATION TITER
`2fJ
`ap
`(/0
`4p
`120
`3f.O
`
`&,IQ
`
`BLOCKING MATERIAL
`
`NONE
`
`WHOLE SERUM
`
`GLOBULIN FRACTION
`
`F(ab)
`
`F(ab)~
`
`Fe
`
`FIGURE 5 Suppression of tetanus toxoid binding to anti(cid:173)
`tetanus antibody by various immunoglobulin fragments of the
`inactive lupus serum. Nine different sera were processed and
`tested. Results are the means of all the experiments, The
`standard deviation did not exceed one tube dilution.
`
`1302
`
`N. I. Abdou, H. Wall, H. B. Lindsley,]. F. Halsey, and T. Suzuki
`
`6 of 8
`
`BI Exhibit 1107
`
`

`

`liminary results indicate that the autoantiidiotypic anti(cid:173)
`body, in the presence of guinea pig complement, is ca(cid:173)
`pable of killing DNA-binding B cells (36). Based on
`these findings, we propose that there are two levels of
`regulation of anti-DNA antibody in SLE: one modulated
`by suppressor cells, and a second modulated by an anti(cid:173)
`anti-DNA antibody. The factors responsible for the acti(cid:173)
`vation of suppressor cells and for autoantiidiotypic anti(cid:173)
`body production in the inactive SLE state are unknown.
`Clearer understanding of all the regulatory elements
`awaits the completion of critical and reproducible
`studies dealing with the role of exogenous infectious
`agents, genetic and hormonal factors that might par(cid:173)
`ticipate in the pathogenesis of SLE.
`
`ACKNOWLEDGMENTS
`
`We thank Deborah Marino, Terrill K. Smith, and Laura
`Janecek for their excellent technical assitance; Anne Knight
`and Juanita Stika for their secretarial help; and Dr. Daniel J.
`Stechschulte, Dr. John D. Martinez, Judy Medved, and Kay
`Appleberry for their cooperation and participation in the care
`of some of the patients.
`The work was supported by National Institutes of Health
`grants Al-15360 and Al-15880, by the Veterans Administra(cid:173)
`tion, by the Kansas Chapter of the Arthritis Foundation, by the
`Oklahoma Chapter of the Lupus Foundation and by the Up(cid:173)
`john Company.
`
`REFERENCES
`1. Jeme, N. K. 1974. Towards a network theory of the
`immune system. Ann. lmmunol. (Inst. Pasteur). 125C:
`373-389.
`2. Wigzell, H., and H. Binz. 1979. Anti-idiotype antibodies.
`Induction of specific transplantation tolerance in adult
`animals. Transplant. Proc. 11: 914-918.
`3. Cosenza, H., and H. Kohler. 1972. Specific suppression of
`the antibody response by antibodies to receptors. Proc.
`Natt. Acad. Sci. U. S. A. 69: 2701-2705.
`4. Urbain, J., C. Collignon, J. D. Franssen, B. Mariame, 0.
`Leo, G. U. Vansanten, P. V. Walle, M. Wikler, and C.
`Wuilmart. 1979. Idiotypic networks and self-recognition
`in the immune system. Ann. lmmunol. (Inst. Pasteur).
`130C: 281-291.
`5. Talal, N. 1978. Autoimmunity and the immunologic net(cid:173)
`work. Arthritis Rheum. 21: 853-861.
`6. Adams, D. D., and J. G. Knight. 1980. H gene theory of
`inherited autoimmune disease. Lancet. I: 396-398.
`7. Allison, A. C. 1974. The roles ofT and B lymphocytes in
`self tolerance and autoimmunity. Contemp. Top.
`lmmunobiol. 3: 227-239.
`8. Abdou, N. I., A. Sagawa, E. Pascual, J. Hebert, and S.
`Sadeghee. 1976. Suppressor T cell abnormality in idio(cid:173)
`pathic systemic lupus erythematosus. Clin. lmmunol.
`lmmunopathol. 6: 192-199.
`9. Raveche, E. S., and A. D. Steinberg. 1979. Lymphocytes
`and lymphocyte functions in systemic lupus erythemato(cid:173)
`sus. Semin. Hematol. 16: 344-370.
`10. Eichmann, K., and K. Rajewsky. 1975. Induction of T
`and B cell immunity by anti-idiotypic antibody. Eur.].
`lmmunol. 5: 661-666.
`
`11. Weinberger, J. Z., R. N. Germain, S. I. Ju, M. I. Greenbe,
`B. Benacerraf, and M. E. Dorf. 1979. Hapten-specific T(cid:173)
`cell responses to 4-hydroxy-3-nitrophenyl acetyl. II.
`Demonstration of idiotypic determinants on suppressor T
`cells.]. Exp. Med. 150: 761- 776.
`12. Bona, C., and W. E. Paul. 1979. Cellular basis of regula(cid:173)
`tion of expression of idiotype. I. T-suppressor cells spe(cid:173)
`cific for MOPC 460 idiotype regulate the expression of
`cells secreting anti-TNP antibodies bearing 460 idiotype.
`]. Exp. Med. 149: 592-600.
`13. Cohen, A. S., W. E. Reynolds, E. C. Franklin, J.P. Kulka,
`M. W. Ropes, L