CLINICAL
`
`[NVESTIGATION
`
`PFIZER EX. 1107
`Page 1
`
`

`

`THE JOURNAL OF CLINICAL INVESTIGATION
`
`Produced by The Rockefeller University Press, New York, for
`The American Society for Clinical Investigation, Inc.
`
`PHILIP W. MAJERUS, Editor
`
`DAVID H. ALPERS
`WILLIAM H. DAUGHADA Y
`SAULO KLAHR
`
`Associate Editors
`STUART KORNFELD
`CHARLES W. PARKER
`BURTON E. SOBEL
`
`The Editorial Committee of The American Society for Clinical Inv estigation, Inc.
`ROBERT H . ALLEN, Denver, Co lo.
`jOSEPH C. GREENFIELD, Durham, N.C .
`THOMAS E. ANDREOLI, Birmingham, Ala.
`SCOTT M. GRUNDY, San Diego, Calif.
`CLAUDE D. ARNAUD, San Francisco, Calif.
`CARL R. KAHN , Bethesda , Md.
`BERNARD M. BABIOR, Boston, Ma ss.
`YUET WAI KAN, San Francisco, Calif.
`ARTHUR BANK, New York
`FRANKLYN G. KNOX, Roches ter, Minn.
`jOHN BAXTER, San Francisco, Calif.
`NEIL A. KURTZMAN, Chicago , Ill.
`BARRY M . BRENNER, Boston, Mas s.
`ROBERT j. LEFKOWITZ, Durham, N.C.
`H. FRANKLIN BUNN, Boston, Mas s.
`PETER T. MACKLEM, Montreal, Quebec
`RONALD G. CRYSTAL, Bethesda, Md .
`GERALD L. MANDELL, Charlottesville, Va .
`ROBERT M. DONALDSON, jR., New Haven, Conn.
`DAVID G. NATHAN, Boston , Ma ss.
`STEPHEN E. EPSTEIN, Bethesda , Md.
`jERROLD M. 0LEFSKY, Denve r, Colo .
`ANTHONY S. FAUCI, Bethesda, Md.
`jACK H . OPPENHEIMER, Minneapolis, Minn.
`MICHAEL FIELD, Chicago, Ill .
`ALAN K. PIERCE, Dallas, Tex.
`MICHAEL M. FRANK, Bethesda , Md.
`ARTHUR H. RUBENSTEIN, Chicago, Ill .
`GERALD GLICK, Chicago, Ill.
`SHAUN RUDDY, Richmond, Va.
`ROBERT GLICKMAN, New York
`THOMAS W. SMITH, Boston, Mass.
`JOSEPH L. GoLDSTEIN, Dallas, Tex.
`jOHN D . STOBO, San Francisco, Calif.
`THOMAS P. STOSSEL, Boston, Ma ss.
`
`THE jOURNAL OF CLINICAL INVESTIGATION publishes
`original papers on research pertinent to human biol(cid:173)
`ogy and disease . Work with an imals or in vitro
`techniques is acceptable if it is relevant to human
`normal or abnormal biology. No substantial part of
`a paper may have been or may be publishe d e lse(cid:173)
`where. Single, complete pape rs are preferred to a
`series of inte rdepe ndent ones.
`
`It is assumed that all clinical investigation de(cid:173)
`scribed in manuscripts submitte d to this jOURNAL
`shall have been conducted acco rding to the principl es
`expressed in the Declaration of He lsinki, which has
`been endorsed by The American Socie ty for Clinical
`Investigation, Inc . (see Vol. 46, p. 1140).
`
`Manuscripts are published in the orde r of final ac(cid:173)
`ceptance after review and revision. Papers requiring
`much e diting will be delayed until they are in good
`form. For information on pre paration of manu(cid:173)
`scripts , see Instruction s to A!lthors publishe d in the
`January and July issues.
`
`Submit manuscripts in three comple te sets to:
`
`Dr. Philip W . Majerus , Editor
`THE JOURNAL OF CLINICAL INVESTIGATIO N
`Editorial Office
`Division of H e matology-Oncology
`Washington University School of Medicine
`St. Louis, Missouri 63110
`
`Authors may purchase re prints at pri ces furni shed
`with galley proof. Othe rs des iring reprints should
`write
`the authors, not the JOURNAL. Claims for
`copies lost in the mails must be rece ived within 30
`days (domestic) or 90 days (fore ign) of the date of
`issue . Change s of address must be rece ived at least
`2 wee ks in advance of the date of issue.
`Two volumes of six issues each appear annually.
`The rate for both fore ign a nd domestic subsc ription s is
`$75 pe r year (postal surcharge Europe only, $22
`ne t), payable
`in advance. Medical students,
`in(cid:173)
`te rns, fe llows, and res ide nts of approved hos pitals
`may subscribe at the special rate of $37.50 per year
`(personal ch eck on ly). Single copies are $6.50 a copy.
`Re mit b y draft or check drawn on U. S. banks, or by
`postal money order, payable to THE jOURNAL OF
`CLINICAL INVESTIGATION, The Rocke fe lle r Univer(cid:173)
`sity Press, P. 0. Box 5108, Church Street Stati on , ew
`York 10249. C laim s for nonrece ipt of issues will be
`honore d whe n received within 3 month s of mailing for
`U.S. A. subscribe rs, 6 months for fore ign subscribers.
`For information on subscriptions and back copies
`write to Orde r Se rvi ce, The Rockefe ll e r University
`Press, 1230 York Ave. , ew York 10021. Subscriptions
`and back copies a re available in Microfi che and
`microfilm through University Microfilm s Inte rnational,
`300 orth Zeeb Road, D ept. P .R. , Ann Arbor, Mich.
`Address corresponde nce about manusc ripts
`in
`press to:
`
`Each manuscript submitted must be accompanied by
`a processing fee of $30 in the form of a check or money
`order. In addition, the author of each paper accepted
`will be charged $30 per published page.
`
`THE jOURNAL OF CLINICAL INVESTIGATION
`Journals Office
`1230 York Ave nue
`New York 10021
`
`The appearance of th e code at th e bottom of th e first page of an article in thi s journal indicates that the Publishe r gives consen t for individual copies of that
`article to be made for personal or internal us e. This consent is given on the condition, howeve r, that-for copying beyond the limited quanti ti es permitted
`unde r Sections 107 or 108 of the U. S. Copyright Law- the copie r pay the stated pe r-copy fee, for this journal $ 1.00 pe r article, through the Copyright
`Clearance Center, Inc., 2 1 Congress Street, Sale m, Massachusetts 01 970, a nonprofit organi zation. This consent does not extend to othe r kinds of copying,
`such as copyi ng for general distribution, for advertising or promotional purposes, for c reating new collecti ve works, or for resale.
`Copyright © 1981, The Ame ri can Socie ty for Clinical Investigation, Inc.
`
`PFIZER EX. 1107
`Page 2
`
`

`

`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, }OHN 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, Kansa s City, Kansas 66103; Veterans Administration
`Hospital, Kansas City, Missouri 64128
`
`ABsTRAcT Regulation of serum anti-DNA antibody
`in systemic lupus erythe matosus (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 th e various depleted sera
`(blocking sera) on target lupus sera were determined.
`The target sera were from lupus pati~nts with known
`DNA-binding capacity. Blocking sera from inactive
`SLE suppre ssed 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 (Ig)G
`and could not be re moved 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 con taining target sera. The
`
`This work appeared in abstract form. (1980. ]. Allergy
`Clin. Immunol. 65: 221; Clin. Res. 28: 338A.)
`Address all correspondence to Dr. N.J. 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 IgG.
`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 J erne ( 1) to be controlled by a network of antibodies
`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
`erythematosus.
`
`}. Clin. Invest.© The American Society f or Clinical Investigation, Inc.
`Volume 67 May 1981 I297 -1 304
`
`· 0021-9738/81/05/1297108
`
`$1 .00
`
`1297
`
`PFIZER EX. 1107
`Page 3
`
`

`

`work theory (l) 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 doubl e-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 unre lated 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 patie nts who satisfied the Ameri(cid:173)
`can Rheumatism Association pre liminary diagnostic crite ria
`for SLE ( 13) we re studied. 19 normal healthy individuals with(cid:173)
`out personal or family hi story suggestive of an autoimmune
`state and with normal leve ls ( < 6.4% binding) of serum anti(cid:173)
`DNA antibody we re 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 othe r 14 normals had no
`· contact with lupus mate rial. The study was approved by the
`institution's human subjects com mittee and infonned con(cid:173)
`sents we re obtained from all of the subjects who e nte red the
`study. All patie nts were studied twice, when the ir disease
`was active and again during clinical remission . Patie nts were
`considered to have active di sease if organ-specific clinical
`symptoms plus at least two of the following laboratory criteria
`were present: (a) erythrocyte sedimentation rate > 25 mm/h;
`(b ) total hemolytic compl e ment CH 50 < 120 U; (c) DNA anti(cid:173)
`bodies > 14% binding. Patie nts were considered to have in(cid:173)
`active disease if no organ-specific clinical symptoms or signs
`could be elicited and if the labo ratory criteria- erythro(cid:173)
`cyte sedimentation
`rate, CH 50 , DNA antibodies -we re
`within the normal range. None of the patie nts 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 durin g inactive disease, from 0 to 40 mg/d (mean, 25 mg/d).
`Serum compl ement d ete rmination (CH 50 assay) was don e by
`a standard techniq ue . The binding of sera to native DNA was
`studied b y the Millipore filte r radioimmunoassay (Millipore
`Corp., Bedford, Mass.) using human KB cell line [3H]DNA
`(Electro-Nucleon ics, Inc., Fairfie ld, N.J.) (14).
`Adsorption of anti-DNA antibody on DNA-cellulose
`columns. Calf thymu s DNA-cellulose (Worthington Bio(cid:173)
`chemical Corp., Freehold, N. J.) was suspe nded in buffer
`(0.01 M Tris-HCI , 0.001 M EDTA, pH 7.4), and packed in
`columns
`(K9/15 columns, Pharmac ia Fine Chemicals,
`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 tp i]lsure
`complete removal of the an ti-DNA antibody. Sera treated in
`this manne r did not contain any d etectabl e anti-DNA antibody
`(0% binding) when tested by radioimmunoassay (14). Cellu(cid:173)
`lose columns to which no DNA was coupl ed were incapabl e
`of deple ting anti-DNA an tibody.
`Treatment of DNA with i1miiobilized DNA se. 6 or 60 U of
`DNAse-Se pharose conj ugate (immobilized deoxyribonucleuse,
`in 1.0 ml ,
`Worthington Bi ochem ical Corp. ), su spe nded
`
`was incubated with 10 JLg [3H]DNA for 60 min at 37•c.
`The tubes were centrifpged at 720 g for 20 min , and 0.5 ml of
`the supernate was then dialyzed overnight against Tris-bu!Ier
`saline. The DNA treated in this manne r fail ed to bind to serum
`containing DNA antibodies. Thus, in a typical experiment
`serum from an active lupus patie nt with 67% binding
`capacity (17,279 counts/min) to the undigested [3H]DNA
`failed to bind to the DNAse-treated ["H]DNA (< 1% binding).
`6 U of DNAse-Sepharose conjugate was as e fficient as 60 U.
`The refore, in all the expe riments re ported in thi s 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 preci pitate
`f ormation by Clq-binding assa y. All sera to be te sted for the
`presence of anti-anti-DNA antibody (antiidiotypic or block(cid:173)
`ing antibodies) were d epleted of anti-DNA antibody by pas(cid:173)
`sage twice through DNA-ce llulose columns and the n treated
`with 6 U of DNAse-Sepharose to digest DNA. In pre liminary
`experime nts, lupus sera with 90% DNA-binding capacity or
`with 10 ILg DNA/ml could be completely d epleted by this
`treatment. None of the blocking sera used in these experi(cid:173)
`ments had DNA-binding capacity > 90% or DNA > 10 JLg/ml.
`Adequacy of depletion was con firm ed by the failure to detect
`anti-D A antibody by radioimmunoassay (14) and of DNA by
`chromatography (15). The anti-DNA depl eted and DNAse(cid:173)
`treated sera (blocking sera) we re assayed for the ir capacity to
`inhibit the binding of [3H)DNA to sera from active lupus
`patients (target sera). For the bl ockin g assay 100 JLI contain(cid:173)
`in g 1 nmol of the blocking mate rial lgG or its various frag(cid:173)
`ments was incubated with 100 JLI of a target serum (contain(cid:173)
`in g 1 nmol lgG) at 37•c for 1 h and then for 16 h at 4°C.
`The mixtures were centrifuged at 1,000 g for 30 min ; 100 ILl
`of the supe rnate was coll ected and tested in the standa,rd
`DNA-binding assay (14). The remaining 100 JLI , d esignated the
`precipitate fraction , was tested in a conve ntional Clq binding
`assay (16).
`The pe rcent suppression of DNA binding was calculated
`from the formula:
`
`( 1 _
`
`D NA binding of mi xtures of target
`and blocking sera) X !00.
`D NA binding of target sera alone
`
`Depletion of variou s l g classes. Depletion of serum IgG,
`IgM , or IgA was performed by standard techniques as de(cid:173)
`scribed earlier (17). Adequacy of d epl etion was confirmed
`by immunoelectrophoresis and by immunodiffusion.
`IgG pro(cid:173)
`Preparation of l gG, F(ab)' 2, and Fe fra gments.
`te ins were isolated from serum by affinity chromatography on
`Prote in A-Sepharose 4B (Pharmacia Fine Chemicals, Uppsalaf
`Sweden). F(ab' )2 fragments produced by pepsin d1gest10n o
`l gG proteins were separated from F c-containing materials by
`passing ove r a column of Protein A-Sepharose 4B (18). Fab
`and Fe fragme nts, which were produced by papain digestion
`of l gG proteins, we re separated also by Protein A-Sepharose
`4B chromatography ( 18). These l gG fragments were separate~Y
`passed through a column of Sephadex G-150 to ensure t e
`removal of undigested IgG proteins. IgG and its enzymatiC
`c leavage fragments thus prepared were immunologica~ly
`pure and distinct whe n exam ined by immunoe lectrophoresiS·
`fragm ents f rom active lup us
`Preparation of F(ab' )2
`se ra . To e nsure that the block ing activity of the a~ti idioty~~
`ant1bocly IS d1rected towards the bmdmg s1tes of ant1-D d
`antibody, we pre pare d F(ab ')2 fragments from IgG Jsolate
`!i·om active lupu s sera. The isolation ofl gG prote in s on Prot~ In
`A-Se pharose 4B and the preparation of F(ab' ). fragments JY
`pe ps in di gestion we re as described above.
`
`1298
`
`N . I. Abdou, H. Wall, H. B. Lindsley, ]. F. Halsey, and T . Suzuki
`
`PFIZER EX. 1107
`Page 4
`
`

`

`Preparation of normal ga mmagloln d in im m u noadsorbents.
`ensure the spec ifi<::ity of the antii d iotypic antibod y, we
`ratt,emptt~o to d epl ete its blocking acti vity by passing it through
`gammaglobulin immunoadsorbe nt co lum ns . Gamma(cid:173)
`. we re iso lated from fi ve d iffere nt no rmal se ra by 33%
`itun sul h1te precip itati on. The precip itate was Wflshe d ,
`re di ssolved , and covale ntl y coup led to CNBr(cid:173)
`Se pharose 48 acco rding to the me thod de scri be d by
`a!. ( 19) . Such affinit y chromatograph y me dia we re
`as gammaglobulin immu noadso rbents. Aliq uots of
`"diotypic se ru m -prepared from inacti ve lu p us serum
`de~> cribe:cl above - we re allowed to pass through the fi ve
`im m un oadsorbe nts. The b locking acti vity of the
`antiidiotypic se rum was tested before and afte r its passage
`through the variou s immun oadsorbe nts.
`Hemagglutination assay . To te st fo r specificity of the
`"idiotypic antibod y, serum h om a normal d onor who
`y bee n boosted with tetan us toxoid was use d as the
`. Antitetan us anti bod y was assayed by the
`passive he magglutination assay using chromium
`to coat sheep e rythrocytes with tetanu s toxo id (20).
`StGrtis tic,~tlanalysis . The paired t test was used to compare
`·ion of target sera in the presence or absence of
`se ra. For comparison of pe rce nt suppress ion w ith
`ng, th e Spearman rank corre lati on coe ffi cie nt was
`ated (21 ).
`
`RESULTS
`
`Blocking of anti-DNA binding. Autologous sera
`lupus patie nts w ith inactive disease (n = 19) we re
`d to suppress the binding of [3 H]DNA to th e target
`sera ( P < 0.01 ) (Fig. 1, Table I). Blockin g sera
`active unre lated (n = 9), from active autologous
`= 19), or from inactive unrelated (n = 9) lupus pa(cid:173)
`were not capable of suppress ion . Human serum
`min at a similar prote in concentration and
`)>roc1~sse d similarly to the vari ous blocking se ra was
`incapable of suppression (Fig. 1). The mean
`;sup~_::1res s ion value of the 19 variou s normal sera tes ted,
`pooled togethe r, was not significantly diffe re nt
`the percent DNA binding of the targe t lupus se ra
`
`by the mse lves ( P = 0.2) (Fig. 1, Table I), Howeve r,
`normal sera from donors who h ad contact with lupus
`patie nts and lupus blood compone nts had significant
`suppress ive acti vity on the target active lupus se ra ( P
`< 0.02) (T ables I and II). Sera from norm al donors who
`had no contact with lupus mate rial did not suppress
`( P > 0.5) (Table II).
`Clq binding correlated with suppression of DNA
`binding. Precipitate fractions obtain ed from incubat(cid:173)
`ing F(ab ') 2 fragme nts with the corresponding autolo(cid:173)
`gous target sera we re tested for the ir ability to bind 125I- .
`Clq by radioimmunoassay. The uppe r limits of the 95%
`confidence inte rvals for individual values -of fragm ents
`from active lupus sera are shown with dotted lines
`parallel to each axis (25% for suppress ion , 8% for Clq
`binding) (Fig. 2). Low Clq binding values (3-7%) oc(cid:173)
`curre d with sera and fragm e nts from active lupus pa(cid:173)
`ti ents; highe r Clq binding values (8-34%) occurred
`with those from patie nts with inactive lupus (Fig. 2) .
`When samples from patients with active and inactive
`di sease we re considered togethe r, pe rcent suppres-
`sion corre lated 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 expe rime nts .performe d on five dif(cid:173)
`ferent sera, d epletion of IgG e liminated the suppres(cid:173)
`sive capacity of the autologous inactive lupus serum
`(Fig. 3). D e pletion ofl gM or of IgA failed to do so ( P
`< 0.01).
`Failure of depletion of the blocking activity by ad(cid:173)
`sorption on norm al human gammaglobulin. To
`avoid arte facts upon IgG depletion of blocking se ra b y
`immunoadsorbents, it is shown in T abl e III that normal
`gammaglobulin immunoadsorbents from five diffe re nt
`donors failed to deplete the blocking activity of the
`lupus serum.
`Effects of l gG fra gm ents on DNA binding.
`In the
`nine sera that were processed and tested , F (ab ')2 frag(cid:173)
`me nts and not Fe fragme nts of the inactive lupus se ra
`we re capable of suppress ing the binding of anti-DNA
`% DNA BINDING OF LUPUS SERA
`10
`20
`30
`40
`50
`60
`--;;;;;;;;;;;;:;;;;;;;;;:;;;;;:;;;;;;~~ antibody to [3H]DNA ( P < 0.001) (Fig. 4). Fab frag-
`ments (P < 0.02), whol e se rum (P < 0.01), and globu(cid:173)
`•
`lin fraction s ( P < 0.01) were also inhibitory.
`Effects of the blocking l gG on binding of F(a b')2
`fra gm ents of the active lupus l gG 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 pre pared F(ab 'h frag(cid:173)
`me nts from IgG of fi ve diffe re nt active lupus sera. It
`could b e seen from T able IV that the blocking IgG in(cid:173)
`hibited the binding of the F (ab ')2 fragme nts to [3H](cid:173)
`DNA. Fe fragm e nts prepared from the same active
`lupus sera failed to bind to [3H]DNA in the absence or
`presence of the blocking IgG (not shown in Table IV) .
`Effect of l gG fra gm ents on tetanus toxoid binding.
`Whole serum, globulin fraction , or the various IgG frag-
`
`---f---4 P< 0.01
`
`SERUM SOUR CE .
`
`NONE
`
`NORMAL
`
`ACTIVE UNRELATED
`
`DNASE treated
`
`1 Suppression of anti-DNA binding to [3H]DNA by
`sera. Results are the means± SD. 19 sera were tested
`. each of the normals, SLE active autologous, and SLE in(cid:173)
`~ve a';ltologous groups. Nine sera we re tested for each of
`E active unrelated and SLE inactive unrelated groups.
`
`Su ppression of Anti-DNA Antibody by Antiidiotypic Antibody
`
`1299
`
`PFIZER EX. 1107
`Page 5
`
`

`

`TABLE I
`Serum DNA Binding before and after Treatment with the Blocking Serum
`
`Patie nt
`
`Predominant clinical feature s
`
`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
`H emolytic an emia, cutaneous
`Thrombocytope nia, 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
`ll
`12
`20
`12
`20
`6
`14
`lO
`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 survivirtg clones were believed to be
`directed mainly against nonself antigens (22). How(cid:173)
`ever, a number of recent important findin gs 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•
`
`Binding
`
`Numbe r
`tested
`
`DNA antibody
`bi~ding
`
`14
`5
`
`%
`
`3.4±2.9
`4.1±2.3
`
`Contact I
`with lupus
`material
`
`Before
`blocking
`
`Afte r
`blocking
`
`Suppression
`
`p
`
`%
`
`It
`
`No
`Yes
`
`44 ± 13
`53±19
`
`33±12
`19± 11
`
`%
`
`25
`64
`
`> 0.5
`< 0.02
`
`* Normal healthy volunteers with negative pe rsonal or family history o f lupus.
`t Contact with lupus patients and lupus blood components for 0.5-16 yr.
`§ Five differe nt lupus sera were used as targets for suppression by the normal sera in all the experiments.
`Each blockin g normal serum that had been adsorbed on DNA-cellulose columns and DNAse-treated was
`tested for its suppressive capacity of each of the ta rget lupus sera.
`
`1300
`
`N. I. Abdou, H. Wall, H . B. Lindsley,]. F. Halsey, and T . Suzuki
`
`PFIZER EX. 1107
`Page 6
`
`

`

`•
`•• •
`
`25
`Percent C lq Bound
`
`50
`
`C lq binding was measure d on a precipitate fraction
`the inte raction of lupus F(ab' ). with autol ogous tar(cid:173)
`lle11 seru111 (see Methods). Pe rcent suppression of DNA binding
`determine d on the same assay tubes. The uppe r limit of
`95% confidence interval, for the samples from patients
`active di sease only, are shown as dotted lines parallel to
`·ponding axis. There was a significant corre lati on
`(Sp,eat·mam's rho = 0.92, P < 0.01) between Clq binding an d
`of suppre ssion of DNA binding. F(ab '). fragments
`of active patients clustered in the lower le ft quad(cid:173)
`were easily di stingui shed from those with inactive
`
`positive and negative inte ractions be tween T and B -
`:l)'lmp,hocytes (7) and the possible involve me nt of iclio(cid:173)
`types in clonal inte ractions (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 re pe rtoire of a single individual;
`autoah.tiidiotypes can be induce d or occur spon(cid:173)
`taneously durin g the immune response (4, 25-27).
`These antiidiotypic antibodies can exe rt e ithe r positive
`or negative influences on antibody bios ynthesis or on
`effector cell fun ction (10, 27).
`
`BLOCK ING DEPLETION
`
`0
`
`% DNA BINDING OF LUPUS SERA
`10
`20
`30
`40
`50
`
`60
`
`......
`
`+
`+
`+
`+
`
`lgG
`
`lgM.
`
`I gA
`
`TABLE III
`Effects on Blocking Activity of Antiidiotypic Serum upon Its
`Adsorption by Normal Human Gammaglobulin
`
`C amrnaglobulin
`immunoad sorbe nt
`fro m normal donors•
`
`Suppress ion of the targe t lupus serum t
`upon incubation with blocking serum §
`
`1ot adsorbed b y
`normal gammaglobulin
`
`Adsorbed by nonna l
`gammaglobulin
`
`1
`2
`3
`4
`5
`
`%
`
`83
`83
`83
`83
`83
`
`8d
`82
`79
`85
`83
`
`* Five different normal donors' gammaglobulin we re linke d to
`CnBr-activated Sepharose 48. See Methods for details.
`I Target serum was from active lupu s patie nt with 53%
`binding to [3 H]DNA.
`§ Blocking serum was obta ined from same donor of the targe t
`se rum during disease inactivity. The blocking serum was first
`d epleted of an ti-DNA antibody and of DNA. Part of the de(cid:173)
`pl eted blocking serum was adso rbe d onto normal gamma(cid:173)
`globulin solid immunoadsorbents. See methods section for
`the calculation of percent suppress ion of the blocking activity.
`
`In thi s re port we have examined the modulation of
`autoantibody activity by mean s of antiidiotypic anti(cid:173)
`bodies. We have de monstrated that binding of anti(cid:173)
`DNA antibody to DNA could be blocked by F (ab') 2
`and Fab fragme nts of l gG obtained from autologou s
`sera of inactive lupus patients (Fig. 4). Blocking ac(cid:173)
`tivity was probabl y due to occupancy of the combining
`site, sin ce F e fragments of the same IgG had no b lock(cid:173)
`ing activity . We have not ruled out, however, the possi-
`
`% DNA BINDING OF LUPUS SE RA
`10
`20
`30
`40
`50
`
`0
`
`60
`
`BLOCKING MATERIAL
`
`NONE
`
`WHOLE SERUM -
`GLOBULIN FRACTION -
`Flab); -Fe
`
`F (obi
`
`FlcuRE 3 Suppression of anti-DNA binding to ["H]DNA by
`autologous inactive lupus sera, and e ffects of depletion of vari(cid:173)
`immunoglobulin classes. Five different sera were
`ous
`PrOcessed and tested. Results are the means of all the expe ri(cid:173)
`rnents. The standard d eviati on did not exceed 7% of the mean.
`
`FIGURE 4 Suppression of anti-DNA binding to [3H]DNA
`by various immunoglobulin fragments of the inactive lupus
`serum. Nine differe nt sera were processed and tested. Re(cid:173)
`su lts shown are the means of all expe riments. Th e stand(cid:173)
`ard de viation did not e xceed 9.3% of the mean.
`
`Suppression of Anti-DNA Antibody by Antiidiotypic Antibody
`
`1301
`
`PFIZER EX. 1107
`Page 7
`
`

`

`TABLE IV
`Blocking of the Binding of A ctive Lupus F(ab') 2 Fragments
`to [3H]DNA by Autologous IgG
`
`['H]DNA binding to
`F(ab')2 fragments I
`
`Expe rime nt*
`
`In absence of
`blocking lgG
`
`In presence of
`blocking lgG §
`
`Suppression 11
`
`1
`2
`3
`4
`5
`
`%
`
`19
`8
`9
`14
`12
`
`63
`41
`53
`34
`39
`
`%
`
`70
`80
`83
`59
`69
`
`* Five different active lupus sera were tested.
`t F(ab ' )z fragments prepared from IgG 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.
`11 Calculated from the formula
`
`( 1
`
`_ binding in presence of blocking IgG ) x lOO.
`binding in absence of blocking lgG
`
`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')2 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
`2f?
`. 4p
`( 10
`8,0
`190
`3f0
`
`640
`
`BLOCKING MATERIAL
`
`NONE
`
`WHOLE SERUM
`
`GLOBULIN FRACTION
`
`F(ab)
`
`Fe
`
`FIGURE 5 Suppression of tetanus toxoid binding to anti(cid:173)
`tetanus antibody by various immunoglobulin fragments of the
`inactive lupus se rum . Nine different sera were processed and
`tested. Results are the means of all the experiments. The
`standard deviation did not excee d one tube dilution.
`
`upon overnight incub~tion, 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 ')2 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 1). Specificity
`of the blocking activity of the normal sera for the Fab
`portion of IgG was not tested. Inhibition of anti-DNA
`binding by normal human serum has bee n observed
`previously (28) .
`The factors responsible for the production of the
`cross-reacting (nonautologous) antiidiotyp ic 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
`respons