E „LANCET
`
`BOSTON, MASS. AND LONDON • SATURDAY AUGUST 9 1986 VOL II FOR 1986
`
`Molecular Biology of Osteogenesis
`OgIGINAL ARTICLES
`Imperfecta
`Dr F. M. Pope
`Ini4unogenicity in Infants of Haemophilus ireluenzae Type B
`Psoriasis and Interferon
`Polysaccharide in a Conjugate Vaccine with V,gscibrii Rteringgdas
`EN C ES29L I
`Dr Barbara Baker and others;
`Outer-membrane Protein (cid:9)
`- (cid:9)
`D-1
`P ti
`fister
`M. g. Einhorn, MD, G. A. Weinberg, MD, E. L. An ers n, V1b,"
`p
`a (cid:9)
`is of Ito and Chromosomal
`University of Wisconsin itontinss....DreteFritottbattosthers;
`P. Dj Granoff, MD, Prof D. M. Granoff, MD (cid:9)
`Sualavian Steal: a Harmless Haemodynami4 esfiokiquihnl .
`3w, (cid:9) pit 4:FAVtiomas, Prof J. L. Frias
`• • (cid:9)
`aile'Achis.and Gastric Acid
`Ti. M. Bornstein, PH D, J. W. Norris, FRCP
`T Wmphocytes of Rheumatoid Arthritis Patients Show Augm8111p 2 (cid:9) Ndo
`0
`Mr D. L. Morris, FRCS
`
`Polyunsaturated Fat and Coronary
`Heart Disease
`ReWivity to a Fraction of Mycobacteria Cross-reactive with Cartilage (cid:9)
`305
`Dr A. H. Kitchin, Dr R. W. D. Turner;
`Joseph Holoshitz, MD, Abraham Klajman, MD, liana Drucker, M sc,
`Prof Hugh Tunstall-Pedoe and others;
`Zvi Lapidot, u sc, Avraham Yaretzky, MD, Ayala Frenkel, PH D,
`Dr Dominique Tater and others
`Willem van Eden, MD, Irun Cohen, MD
`Redutpd Platelet Mitogenic Activity
`in Myeloproliferative Disorders
`Evi ence for an IlLA-DR4-associated Immune-response Gene for
`Dr M. Romano and others
`-LifeExpeetancy, Truth, and the ABPI
`111(ygobacterium tuberculosis
`Dr David St George
`T. I-1. M. Ottenhoff, MD, Pedro Torres, PHARM D, J. T. de las Aguas, MD,
`Alternative Medicine
`Rartion Fernandez, MD, Willem van Eden, MD, R. R. P. de Vries, MD,
`Dr S. Shepherd
`, J. Li. Stanford, MD
`Secure Accommodation in Psychiatric
`Hospitals
`Chronic Hypoxaemia and Decompensated Erythrocytosis in
`• Cyapotic Congenital Heart Disease (cid:9)
`Dr T. D. Scannell
`Grand Multiple Pregnancies and
`M. Rosbve, MD, Prof J. K. Perloff, MD, W. G. Hocking, MD,
`I, (cid:9)
`Demand for Neonatal Intensive Care 347
`Dr M. I. Leven
`J. S.IChild, Nu), Mary Canobbio, RN, D. J. Skorton, MD
`Replication of Clinical Trials (cid:9)
`Dr W. A. Silverman
`Prescribing by Nurse Practitioners (cid:9)
`Mrs B. M. Cowper
`Lisuride Infusion Pump for
`Parkinson's Disease (cid:9)
`Dr Stefano Ruggieri and others;
`Dr S. Bittkau, Prof H. Przuntek
`Psychosis and the Lisuride Pump (cid:9)
`Dr Peter Critchley and others
`Prevention of Clostridium dig-kilo
`Outbreaks in Hospitals (cid:9)
`Dr Michel Delmee,
`Prof Jean-Louis Michaux;
`Dr.Meianie Williams
`First-trimester Prediction in Fetus at
`Risk for Myotonlo Dystrophy:.-. (cid:9)
`Dr P. W. Lunt and others
`Chance that Individual in Duchenne
`Family is Recombinant (cid:9)
`Prof J. H. Renwick
`
`LETTERS TO THE EDITOR
`Cyclosporin In Frequently Relapsing
`Minimal ChangeNephrode Syndrome 335
`Dr P. F. Hoyer and others
`Night-time Rioprostil versus Ranitidine
`in Duodenal Ulcer Healing • • •
`Prof H. G. Damrnann and others
`An'ti-STLV-Illnine Reactivity in HIV
`Seropositive Individuals in Sweden
`Dr Jonas Blomberg and others_
`Vertical Transmission of Huintui
`Immunodeficiency Virus
`Dr A. R. Lifson, Dr Martha Rogers ,
`Malnutrition and HIV Infection in (cid:9)
`Children in the Central African
`Republic
`Dr J. L. Lesbordes and others .
`Origin of Abnormal Magnetic
`Resonance Imaging Signal in Multiple
`Sclerosis
`Dr I. E. C. Ormerod
`Depression in the Elderly
`Dr Alex Comfort
`Respiratory Mucus
`Dr M. T. Lopez-Vidriero
`Radioactive Iodine for Thyrotoxicosis
`Prof P. R. F. Bell;
`Dr L. Hegedds, Dr J. M. Hansen
`Primary Hyperoxaluria Type I:
`Oxalate and Glycolate Unsuitable for
`Prenatal Diagnosis
`Prof E. Leumann and others
`Cross-resistance and Imipenem
`Dr G. Calandra and others
`Dilutional Hyponatraemia
`Masquerading as Subarachnoid
`Haemorrhage in Patient on
`Hydrochlorothiazide/Amiloridei
`Timolol Combined Drug
`Dr G. F: A. Benfield and others
`
`335
`
`336
`
`337
`
`•
`
`337
`
`338
`
`338 -
`
`339
`
`339
`
`340
`
`340
`
`341 (cid:9)
`
`NATIONAL HEALTH SERVICE
`The Labour Party Condemns the
`Dearth of Infertility Services (cid:9)
`Legislation Against the Sale of
`Tobacco to Children (cid:9)
`Public Concern about the State
`of the NHS
`
`I
`NOTES AND NEWS
`The AIDS Campaign
`Women and Alcohol
`Snuff Dipping
`MRC Tuberculosis and Related
`Infections Unit
`International Physicians for the
`Prevention of Nuclear War
`Improving Health
`Outpatient Queues
`AIDS Training Video
`New Director of Information for the
`World Health Organisation
`The Royal Society
`Disability Rights Handbook
`Churchill Travelling Fellowships
`Beit Memorial Fellowships for
`Medical Research
`
`357
`
`357
`
`357
`
`353
`353
`353
`
`353
`
`353
`353
`354
`354
`
`354
`354
`'354
`354
`
`354
`
`•
`
`•
`
`313
`
`Growth Factors and
`Malignancy
`Improving Use of Clinical
`Resources
`Pain-relief in Sickle Cell Crisis
`Dupuytren's Contracture
`Rheumatoid. Arthritis and
`Tuberculosis
`
`317
`
`319
`320
`321
`
`321
`
`341
`
`342
`
`343
`
`343
`
`344
`
`345
`
`346
`
`346
`
`347
`
`348
`
`348
`
`348
`
`349
`
`350
`
`350
`
`357
`
`4 (cid:9)
`
`REVIEWS
`Notices of Books
`
`EP EMIOLOGY
`Mt ma Mortality in England
`an Wales: Evidence for a
`Fu ther Increase, 1974-84
`P. (cid:9)
`. J. Burney, MPCM
`
`OdCASIONAL SURVEY
`Renal. Physiology Revisited:
`iloride
`J. . Scoble, MRCP,
`Z. (cid:9)
`arghese, PH D,
`P. weny, FRCP,
`J. 45,00rhead, FRCP
`
`315
`
`323
`
`326 (cid:9)
`
`IT( , SPITAL PRACTICE
`0(itith as an Option in Neonatal
`AUtensive Care
`drew Whitelaw, MRCP
`
`,ORTS MEDICINE
`um Kidney; Epidemic
`,oderma Caused by a
`, fiphritogenia Streptococcus
`ypgenes in a Rugby Team
`
`.
`Ludlam, MRC PATII, .
`Cookson, MRC PATH
`
`ofirruARY
`Wjlliam Jacobson Rashkind
`'Wald Hare
`
`INTERNATIONAL DIARY
`
`328 (cid:9)
`
`331 (cid:9)
`
`352 (cid:9)
`352 (cid:9)
`
`352 (cid:9)
`
`Illitorial Office: 7 Adam Street, London WC2N 6AD, England. Tel:
`014-836 6942. North American Edition published weekly by Little,
`Brpwn and Company, 34 Beacon St., Boston, Mass. 02106. Annual
`stibscription -in U.S. $68.00, in Canada $78.00; resident and intern rate
`(1.111.3S. and Canada) $46.50; Single copy $6.00. Second class postage paid
`at oston, Mass., and at additional mailing offices. © The Lancet ltd.,
`9 August 1986. Postmaster: Send address changes to The Lancet, 34
`Wacon Street, Boston, Mass. 02106. (cid:9)
`ISSN 0099-5355
`T118I-1 WHOLE OF THE LITERARY MATTER IN THE LANCET IS COPYRIGHT
`
`971 7iqc1 cm I
`0186 lzd6' umivtic
`Sandoz v. AbbVie
`UNIVERSITY
`MIDDLETON H
`Sandoz Ex. 1046
`LIB—W/1305
`MADISON
`
`Ex. 1046 - Page 1
`
`(cid:9)
`(cid:9)
`(cid:9)
`(cid:9)
`

`

`THE LANCET, AUGUST 9, 1986
`
`Ann exercise increasing flow down the subclavian trunk
`may "steal" blood from the ipsilateral vertebral artery,' but
`such exercise rarely. provokes cerebral symptoms.3,4 North et
`al reported that 6 of 59 patients whO were to undergo
`subclavian endarterectomy had neurological symptoms
`provoked by arm or hand exercise, and 3 of these also had
`carotid stenosis or occlusion.9 1 patient had vertigo and
`nystagmus whenever he exercised his right arm, and these
`symptoms were abolished by right subclavian
`endarterectomy, but he also had an ipsilateral carotid
`endarterectomy at the same operation.
`Since reversal of blood flow down the vertebral artery is
`common and usually asymptomatic, factors other than
`retrograde flow must be operative, and these probably
`depend on the adequacy of the collateral circulation.s." In
`one arteriographic-clinical correlation, 35% of patients
`(15/42) with subclavian-steal syndrome had vertebrobasilar
`symptoms, and 2 had dizziness precipitated by arm
`exercise? The authors concluded that symptoms depended
`largely on the.integrity of the circle of Willis, and they noted
`that. "steal" is rare in large vessels such as the carotid and
`iliac where resting levels of flow may increase ten-fold
`without change in arterial resistance.
`A striking preponderance of left-sided lesions (similar to
`our series) was found in all reported series." This may be
`due to the more acute angleat the origin of the left subclavian
`artery, which produces more turbulence—an established
`factor determining the site of atherosclerosis."
`It. seems clear that reversal of flow down one vertebral
`artery is relatively common in patients with widespread
`extracranial atherosclerOsis, occurring in 2 to 9% of patients,
`and in 1 of 4 cases of severe subclavian stenosis.6-8 However,
`it is far from proven that blood siphoned away from the
`brainstem by arm exercise produces brainstem ischaemic
`symptoms, and subclavian steal may represent only a
`harmless haemodynamic phenomenon causing, at worst,
`minor vertebrobasilar TIAs.
`Since there was no difference in survival between
`medically and surgically treated patients, surgical
`"prophylaxis" of stroke should be reserved for patients with
`disabling vertebrobasilar TIAs.6
`
`This project was partly funded by grant MA-8924 from the Medical.
`ResearCh Council of Canada. N. M. B. is a research fellow, Medical Research
`Council of Canada. We thank Lorraine Chadwick and Susan Corrigan for the
`doppler examinations and Cecil), Ziliotto for managing the data.
`
`Correspondence Should be addressed to J. W. N., Sunnybrook Medical
`Centre, 2075 Bayview Avenue, Toronto, Ontario, Canada M4N 3N5.
`
`REFERENCES
`
`1. Reivich M, Rolling HE, Roberts 13, Toole JF. Reversal of blood flow through the
`vertebral artery and its effect on cerebral circulation. N &el Med 1961; 265:
`. (cid:9)
`878-85,
`2. Fisher CM. A new vascular syndrome—"the subclavian steal". N Engl Y Med 1961;
`265: 912-13.
`3. Sicken RG, Milliken CH, Whisnant JP. Reversed blood flow in the venebral arteries.
`Ann hums Med 1964; 61: 64-72,
`4. Heyman A, Young WG, Dillon M, et al. Cerebral ischemia caused by occlusive lesions
`of the subclavian or innominate arteries. Arch Neural 1964; 10: 581-89.
`5, Patel A, Toole JT. Subclaviin steal syndrome—reversal of cephalic blood flow..
`Medicine 1965; 44:.289-303.
`6. Fields WS, Lemak NA. Joint.study of extracranial arterlai occlusion, VII. Subclavian
`•
`steal—a review of 168 cases. JAMA 1972; 22211139-43. (cid:9)
`7. Lord RSA, Adar R, Stein RL. Contribution of the circle of Willis to the subclavian
`steal syndrome. Circulation 1969; 401871-78.
`8, Manniek JA, Suter CG, Hume DM. The "subclavian steal" syndrome: A further
`documentation. 'AMA 1962; 182: 254-58.
`9. North RR, Fields WS, DeBakey ME, Crawford ES. Ltrachial-basilar insufficiency
`syndrome. Neurology 1962; 12: 810-20.
`10. von Reutem G-M, Pourcelot L. Cardiac cycle-dependent alternating flow in vertebral
`arteries with subclavian artery stenoses. Stroke 1978; 9: 229-36.
`11. Kesteloot. H, Route OV, Reversed circulation through the vertebral artery. Acta
`Cardiol 1963; 18: 285-99.
`
`305
`
`T LYMPHOCYTES OF RHEUMATOID
`ARTHRITIS PATIENTS SHOW AUGMENTED
`REACTIVITY TO. A FRACTION OF
`MYCOBACTERIA CROSS-REACTIVE WITH
`CARTILAGE
`JOSEPH HOLOSHITZ (cid:9)
`ABRAHAM KLAJMAN
`ZVI LAPIDOT •
`ILANA DRUCKER (cid:9)
`AYALA FRENKEL
`AVRAHAM YARETZKY (cid:9)
`WILLEM VAN EDEN (cid:9)
`IRUN R. COHEN
`Department of Medicine B and Laboratory of Clinical Immunology,
`Meir Hospital; Kfar Saba; and Department of Cell Biology,
`Weizmann Institute of Science, Rebovot, Israel
`
`Summary " An acetone-precipitable fraction of Myco-
`bacteriuM tuberculosis cross-reacts 'with
`human cartilage. Immune responses to this antigen were
`assessed in 34 patients with rheumatoid arthritis, 16 patients
`With degenerative joint disease, and 15 healthYcontrols. The .
`. RA patients differed from the other two groups in having
`more pronounced T lymphocyte responses to the antigen;
`their serum antibody levels were not higher. The responses
`of RA patients varied with duration of disease. In the first
`year (7 patients) T lymphocyte reactivity was increased in
`'the synovial exudates of affected joints but not in peripheral
`blood, whereas the 19 with disease of 1-10 years' duration
`showed high reactivity in peripheral blood; in. the 8 with
`disease for more than 10 years, lymphocyte reactivity. id not
`differ from that in the patients with degenerative joint
`disease or the healthy controls. The observation that the
`three groups did not differ in their responses to streptococci
`and a T-cell mitogen indicates that reactivity of the RA
`patients to the mycobacterial fraction was specific. These
`results raise the possibility that bacterial antigens cross-
`reactive with cartilage proteoglycans may be relevant to the
`pathogenesis of RA.
`
`Introduction
`RHEUMATOID arthritis (RA) is believed to be an
`immunological, possibly autoimmune, disease,' although
`neither an inciting cause :nor a target self-antigen has been
`identified unequivocally. Adjuvant arthritis is a disease
`inducible in genetically susceptible rats that has certain
`morphological similarities to RA. The disease features
`synovitis, largely of the distal joints of the limbs, progressing
`to pan,nus formation, erosion of articular cartilage and
`stibarticular bone, and ultimately ankylosis:2 Extra-articiflar
`'lesions have also been noted.'
`The inciting cause of adjuvant arthritis is immunisation
`'with antigens of Mycobacterium tuberculosis (MT),
`commonly administered in oil in the form of complete.
`Freund's adjuvant.4 Despite its triggering by bacterial
`immunisation, adjuvant arthritis was felt to be an
`autoimmune disease because it could be transferred to
`recipient rats via lymphocytes of affected rats .s The
`autoimmune hypothesis vvas confirmed by our observation
`that arthritis.could be induced by inoculation of rats with a
`clone of anti-MT T lymphocytes:6 This arthritogenic clone
`was found to recognise both an antigen of MT and a
`component of cartilage proteoglycan, probably the core
`protein.' Thus, the antigen receptor of the arthritogenic
`clone of T lymphocytes defined a structural similarity
`between a mycobacterial epitope and an epitope of joint
`cartilage. This cross-reactivity between MT and cartilage
`can explain why some anti-MT lymphocytes might attack
`the joints.8
`
`This material may be protected by Copyright law (Title 17 U.S. Code)
`
`Ex. 1046 - Page 2
`
`(cid:9)
`(cid:9)
`(cid:9)
`(cid:9)
`

`

`306 (cid:9)
`
`TUB LANCET, AUGUST 9, 1986
`
`Guided by our studies of adjuvant arthritis in rats, we
`investigated whether RA too might be marked by reactivity
`to an acetone precipitable fraction of MT (AP-MT)
`containing the proteoglycan-mimicking epitope identified
`by our arthritogenic rat T lymphocyte clone. Peripheral
`blood and synovial fluid lymphocytes of patients with RA or
`degenerative joint disease. (DJD) and peripheral blood
`lymphocytes of healthy individuals were tested in an in-vitro
`proliferation assay for their response to AP-MT, to
`streptococci, or to a T lymphocyte mitogen.
`
`Immunisation of Rabbits
`Anti-AP-MT antisera were raised in two outbred white New
`Zealand rabbits by five subcutaneous injections at 3-week intervals
`with 1 fnl of an emulsion containing equal volumes .of incomplete
`Freund's adjuvant (Difco) and an aqueous solution of AP-MT (1
`mg/m1). Serum was collected after the fifth immunisation and kept
`frozen at. — 70°C until use. Both rabbits produced antibodies to
`AP-MT. Anti-chick type I collagen antiserum was prepared
`similarly and was kindly donated by Dr Z. Nevo (Tel-Aviv
`University.).
`
`Methods
`
`Subjects
`We studied 34 patients with. RA defined by the American
`Rheumatism Association criteria' (table. I). Recent radiographs of
`the hands and wrists were available for 26 patients, of which thirteen
`showed erosions of bone. 27 patients were treated with non.
`steroidal and-inflammatory drugs; 3 were :treated with ID-
`penicillamine; and 22 had received injections of gold. 8 patients
`were on oral corticosteroids (mean dose 15.6 ± 1.9 mg of
`prednisone) at the time of the study. None had ever been treated
`with cytotoxic drugs.
`Two control groups were studied-16 patients with DJD and 15
`blood donors with no arthritis or other major disease (tablet). Some
`of the. DJD group were taking non-steroidal anti-inflammatory
`drugs; otherwise the control groups were medication-free.
`
`Antigens
`Concanavalin A (Con A) was purchased from Bio-Yeda
`(Rehovot) and heat-killed Mycobacterium tuberculosis I-127Ra (MT)
`from Difco Laboratories (Detroit, USA). Heat-killed beta-
`haemolytic streptococcus group A, type 24-M, was donated by Dr
`I. Ofek of Tel-Aviv University. Proteoglycans were extracted as
`described" from pork cartilage by G. Hunter and donated by Dr A.
`Czitrom (Mount Sinai Hospital, Toronto, Canada). Chick type I
`collagen was prepared" and donated by Dr D. Duksin (Weizmarm
`Institute). Core protein of bovine proteoglycan was prepared by
`Smith degradation" and donated by Dr Z. Nevo (Tel-Aviv
`University). Bovine serum albumin (BSA) was purchased from
`Sigma Chemical Co (St Louis, USA). Solutions of antigens were
`sterilised by filtration through Millipore filters.
`
`AP-MT
`400 mg heat-killed desiccated MT was-ground with a mortar and
`pestle, suspended in 400 ml double distilled H2O, and stirred for
`48 h at 4°C. The suspension was cleaned by centrifugation for 15
`min at 20 000 g and the supernatant was lyophilised. The acetone
`precipitable fraction (AP-MT) was prepared by dissolving the
`lyophilisate in:H2O to a concentration of 10 mg/ml to which was
`added two volumes of cold acetone (analytical grade). The material
`was stirred gently overnight in the cold. The resulting precipitate,
`about 20% of the starting material, was washed in Hp:acetone, 1:2,
`dissolved in H2O, and stored lyophilised until use.
`
`TABLE I—PATIENTS
`
`RA
`
`—
`
`No
`Age range
`•
`Mean age (cid:9)
`Mean duration
`(yr)
`Females/males
`Rheumatoid
`factor positive*
`Treated with
`corticosteroids
`
`Total
`
`.34
`19-76
`54.6
`
`<1 yr 1-10 yr >10 yr
`8
`53-76
`61.6
`
`19
`31-75
`534
`
`7
`19-70
`49
`
`DJD
`
`16
`33-75
`57.1
`
`No
`arthritis
`
`15
`26-65
`43.1
`
`6.6±1.3
`25/9
`
`0.5 ±0.1
`4/3
`
`+1+04
`15/4
`
`17.8+2.0
`6/2
`
`6.4±1.0
`12/4
`
`-
`10/5
`
`21
`
`8
`
`2
`
`1
`
`11
`
`4
`
`8
`
`3
`
`0
`
`0
`
`0
`
`0
`
`*Rose-Waaler test, titre >32.
`
`Immunofluorescence of Human Cartilage
`Cartilage was obtained from the traumatically amputated index
`finger of a healthy young adult. Pieces of the -joint cartilage were
`snap-frozen and kept at — 70°C. Antibodies to cartilage were
`assayed by an Immunofluorescence technique slightly modified
`from that described previously."." Snap-frozen cartilage was
`cryocut into 4 nn sections, incubated for 30 min at mom
`temperature with the test rabbit serum diluted 1:20 in phosphate
`buffered saline (PBS) containing 50% normal human serum,
`washed three times with PBS, and counterstained with 1:50 diluted
`fluoresceinated polyvalent goat anti-rabbit 7S gammaglohtilin
`antiserum (Hayland Diagnostics, Deerfield, USA). Coded sections
`were examined by an independent observer using a Zeiss standard
`microscope with an epifluorescence condenser, who graded them
`on a scale of 0 to 4 + .
`To test the antigenic specificity of .anticartilage antibodies,
`competitive inhibition of antibody binding was done by
`preincubation with various soluble antigens. Rabbit sera were
`diluted 1:20 in PBS containing 1 mg/m1 proteoglycan or collagen
`type I antigen and 50% normal human serum and incubated for 1 h
`at 37°C in a shaking bath, centrifuged at 1200 g for 10 min, and
`tested for binding to cartilage as above.
`
`Monoclonal Antibody
`A monoclonal antibody to MT, designated TB5.2, was raised by
`Dr T. M. Daniel and associates at the Western Reserve University,
`Cleveland, Ohio.'" This antibody binds to antigen i5 of MT." It was
`used in radioimmunoassays at a dilution. of 1:50.
`
`Solid-phase Radioimmunoassay
`Antibodies were measured in sera by a solid-phase
`radioimmunoassay" in 96-well micro-titre plates (Dynatech,.
`Alexandria, USA). Each well contained 4 1.ig AP-MT or
`proteoglycan core protein, or BSA (2% in PBS).
`
`Proliferative Responses
`Peripheral blood or synovial fluid 'mononuclear cells were
`separated from fresh heparinised blood or synovial fluid on
`Ficoll-Hypaque gradients (Nyegaard, Oslo, Norway). Proliferative
`responses of these cells were measured in round-bottomed
`microtitre plates (Nunc, Denmark) in triplicate wells. Each well
`contained 1 x 10" mononuclear cells suspended in 0.2 ml RPMI-
`1640 medium (Biological Industries, . Beth Haemek, Israel),
`supplemented with 10% normal human serum obtained from a
`single pool of healthy blood donors; 10 mmol Hepes buffer (Sigma,
`St Louis), 2: mmol glutamine (Bio-Lab, Jerusalem, Israel),
`penicillin (100 U/m1), streptomycin (10 µg/m1), and mycostatin
`(10 UM) in the presence or absence of antigens (AP-MT at 5, 10,
`20, and 50 µg/m1; streptococcus at 5,10, 20, and 50 µg/m1; or Con A
`at 12.5, 25, and 50 µg/ml). Cultures were incubated at 37°C in a
`humidified atmosphere of 5% CO2 and 95% air for 3 or 5 days to
`test proliferative responses to Con A or antigens, respectively. For
`the last 18 .h of incubation each well was pulsed with 1 nCi of
`['H]-thymidine (specific activity 10 Ci/mmol; Nuclear Research,
`Negev, Israel). Cultures were harvested on fibreglass filters and
`thymidine incorporation was measured in a liquid scintillation
`counter. Results were expressed as a stimulation index (ratio of
`mean cpm with antigen or mitogen/mean cpm without antigen or
`
`Ex. 1046 - Page 3
`
`

`

`THE LANCET, AUGUST 9, 1986
`
`mitogen). The background cpm measured in the absence of added
`antigen or mitogen did not differ significantly between the groups of
`RA or DJD patients or the healthy controls (1599 1 261,
`1371 1 307, and 1226 1 239 on day 3 and 1198 1 219, 1047 ± 153,
`and 880 ± 95 cpm on day 5, respectively). The maximum
`stimulation index induced by any concentration of antigen or
`mitogen is presented in the results.
`
`T Lymphocyte Lines.
`Lymphocytes responding to AP-MT from 6 synovial fluids and
`18 peripheral bloods of RA patients were raised as lines' by the
`method of Ottenhoff and associates." Irradiated peripheral
`mononuclear cells were used as antigen presenting cells as
`described." Immunofluorescence determination of T3, T4, and T8
`markers was done with a commercial kit (Ortho Diagnostics,
`Raritan, USA).
`
`Statistical Analysis
`Statistical significance of differences between groups was
`evaluated with the Mann-Whitney U test. P values between 0.05
`'and 0.01 were regarded as 'significant. Numerical results are given as
`mean ± SEM unless otherwise stated.
`
`Results
`Cross-reactivity between Cartilage and AP-MT
`Rabbit anti-AP-MT antibodies were bound much more
`strongly to human cartilage than'was antibody to collagen
`type I. (which is present only in small amounts's). Table iI
`shows that it is proteoglycan rather than collagen type I to
`which anti-AP-MT antibodies are bound in cartilage. The
`binding of anti-collagen-type-I antibodies to human
`cartilage was absorbed by collagen type I but not by
`proteoglycan. Thus, AP-MT contains an antigen that is
`serologically cross-reactive with an antigen in the
`proteoglycan moiety of human cartilage. To define this
`antigenic cross-reactivity further, we studied the binding to
`AP-MT and to the core protein of cartilage proteoglycan of
`monoclonal antibodies raised by immunisation to MT.'5
`
`TABLB 11—BINDING OF RABBIT ANTISERA TO NORMAL HUMAN
`CARTILAGE
`
`Antibody binding
`to cartilage
`(relative immunofluorescence)
`
`Rabbit antiserum
`to
`
`AP-MT
`
`Chick collagen type I
`
`Inhibition test
`-with antigen
`None
`AP-MT
`Proteoglycan
`Collagen type I
`None
`AP-MT •
`Proteoglycan
`Collagen type .I
`
`TABLE III—CROSS—RBACPIVITYBBTWEBN AP—MTAND CORE
`PROTEIN OP-CARTILAGE PROTEOGLYCAN DETECTED BY
`-MONOCLONAL ANTIBODY TB5.2
`
`Antibody binding (cpm x 10-a)
`
`AP-MT
`BSA
`Antibody
`7.8.+0.4*
`1.7 +0.6
`TB5.2
`• 10.7 +0.4*
`1.5+0.2
`Control anti-MT
`*Difference from BSA control p <0.01.
`
`Core proteins
`7.3 ± 0.6*
`1.2±0.1
`
`307
`
`_
`
`— -
`
`112
`
`9
`
`g 5
`
`•
`
`25
`
`A
`
`s
`
`•• A
`
`$
`.
`.
`.
`V
`< I
`>10
`years
`years
`RA
`Fig 1—Proliferation responses of peripheral blood lymphocytes to
`AP-MT.
`
`1-10
`years
`
`DJ D
`
`NO ARTHRITIS
`
`Table shows that monoclonal antibody TB5.2 recognised
`both AP-MT and core protein, but not the irrelevant
`antigen BSA. Twenty other anti-MT monoclonal
`antibodies similar to the control anti-MT in table III
`recognised only AP-MT and not core protein. Thus,
`AP-MT and the core protein of cartilage proteoglycan share
`a specific antigenic site. This AP-MT, therefore, was used
`to test the reactivity of lymphocytes of RA patients.
`
`Responses of Peripheral Blood Lymphocytes
`Fig 1 illustrates the individual stimulation indices of
`peripheral blood lymphocytes to AP-MT. The RA patients
`were divided into three groups according to the duration of
`disease. Those afflicted with RA for 1-10 years showed
`strikingly higher responses to AP-MT (mean stimulation
`index 36.7 6.3) than did, patients who had had RA for less
`than 1 year (mean stimulation index 2.3 f 0.4) or more than
`10 years (mean stimulation index 7.4 ± 1.9). The patients
`with RA for 1-10 years also differed significantly from the
`DJD patients (mean stimulation index 7.5 ± 1.9) and from
`the healthy controls (mean stimulation index 3.9 ± 0.6). All
`these differences were highly significant statistically.
`Responses in the RA patients with disease longer than 10
`years did not differ significantly from the DJD or healthy
`controls. The group with RA for less than 1 year had much
`lower responses to AP-MT than did DJD patients, healthy
`controls, or RA patients with disease for more than 10 years
`(p < 0.01). RA patients treated with steroids had much lower
`responses (mean stimulation inclex.6.6 ±. 1.7) than did those
`who did not receive steroids (mean stimulation index
`27.6 *5-5; p.<0.01).
`Responses to Con A did not differ significantly between
`any of the groups. There were also no significant differences
`between the responses of RA patients (mean stimulation
`index 14.5 ± 2.6) and controls (mean stimulation index
`13.4 ± 2.7) to streptococci.
`
`Ex. 1046 - Page 4
`
`(cid:9)
`

`

`308
`
`THE LANCET, AUGUST 9, 1986
`
`145 — •
`
`30—
`
`6
`20
`
`2 .
`
`10—
`
`Fig 2—Proliferative responses of synovial fluid lymphocytes to
`AP—MT.
`
`RA (cid:9)
`
`ND
`
`145
`
`30
`
`x
`w
`c'z 20
`
`2
`F
`10
`
`Blood
`
`Synovial (cid:9)
`Fluid
`
`RA
`
`•
`Fig 3—Simultaneous proliferath4 responses of synovial fluid and
`peripheral blood lymphocytes to AP—MT in 7 patients with RA.
`
`Responses of Synovial Fluid Lymphocytes
`Proliferative responses of synovial fluid lymphocytes
`were studied in 8 RA patients and 8 DJD patients. The
`responses to AP-MT of RA patients (mean stimulation
`index 30.3.± 15.4) were strikingly higher (p <0.01) than
`those of the DJD patients (mean stimulation index 6.6 ± 3.2)
`(fig 2). There was no significant difference between the
`groups in responses to Con A (23.1 ± 23 versus 18.8 ± 0.4),
`indicating that the general populations of T lymphocytes in
`the synovial fluids were similar in the two groups.
`
`Fig 3 illustrates the responses to AP-MT of synovial fluid
`lymphocytes and peripheral blood lymphocytes in
`individual RA patients. In .5 of the 7 patients the response in
`synovial fluid cells was much higher than , that in the
`peripheral blood cells (mean stimulation index 37.2 ± 24.5
`versus 1.8 ± 0.5). It is noteworthy that these 5 patients had
`had RA for less than a year. The two RA patients with little
`difference between the blood and synovial fluid responses
`had more longstanding disease. The synovial-fluid/
`peripheral-blood ratio of the mean stimulation index was
`10.8 1 4.9 for RA patients and only 0.4 ± 0.2 for DJD
`patients (p <0.01).
`
`Nature of Cells responding to AP-MT
`To identify the lymphocytes proliferating in response to
`AP-MT, we:grew the cells of 18 of the RA patients as lines"
`and studied their surface markers. 6 lines were raised from
`the synovial fluids and 18 from peripheral bloods. In each of
`the lines, about 95% of the cells bore the T3 and the T4
`markers, and 5% or less of the cells were positive for the T8
`marker. Thus, the bulk of the cells responding to AP-MT ,
`were T lymphocytes of the helper type.
`
`Anti-AP-MT Antibodies
`;Using a solid phase radioimmunoassay, we could find no
`significant differences between anti-AP-MT antibodies in
`sera of RA patients compared with controls. The titres of
`antibodies in both groups ranged between 1:10 and 1:50.
`
`Discussion
`The fact that RA patients did not differ from controls in
`the amount of antibody to AP-MT in their sera indicates
`that their heightened reactivity to AP-MT was associated
`specifically with the T lymphocyte artn of the immune
`response; and the similar T lymphocyte responses to
`streptococci and Con A in patients and controls suggest that
`the reactivity to AP-MT is specifically,associated with- RA.
`This needs to be confirmed by investigations in other forms
`of immune arthritis.
`An earlier report in this area was that of Abrahamson and
`colleagues," who.compared reactivity of synovial-fluid and
`peripheral-blood lymphocytes to purified protein derivative
`of tuberculin in individual patients with RA. Like us they
`found much greater reactivity in the synovial-fluid cells,
`whereas responses to mitogens and Canclida albicans were
`stronger in peripheral-blood cells. By contrast, Ivany and
`co-workere found no increase in responsiveness to PPD in
`the synovial fluid T lymphocytes of 11 RA patients.
`Moreover, in some groups of RA patients skin test responses
`to PPD are reported to be low.2i however, these discordant
`observations do not necessarily contradict the results
`reported here or those of Abrahamson and colleagues. We
`find that only some preparations of PPD seem to contain the
`epitope present in AP-MT recognised by our arthritogenic
`rat clone (unpublished). The absence of the critical AP-MT
`epitope could explain a lack of T lymphocyte reactivity to
`PPD. Contrary results could also be auxihuted to differences
`in sampling. In our series the degree of reactivity to AP-MT
`was related both to the stage of disease and to the site from
`which the responding lymphocytes were obtained.
`By extrapolation from the arthritogenicity of M
`tuberculosis in rats and the cross-reactivity between M
`tuberculosis and cartilage, it is conceivable that this or other
`
`Ex. 1046 - Page 5
`
`(cid:9)
`

`

`TIM LANCET, AUGUST 9, 1986
`
`309
`
`microbes bearing similar *antigens might serve as a vehicle
`for immunising individuals against cross-reactive epitope(s)
`present in their own cartilage.
`Cross-reactivity between microbial antigens and tissue
`components has been detected in other autoimmune
`conditions. Antigens of group A streptococci cross-react
`with antigens in heart arid glomerular basement
`membrane.23 Antibodies to DNA produced by hybridomas
`derived from the blood of patients with systemic lupus
`erythematosus are reported to cross-react with bacterial
`antigens,24 as are antibodies .to the.acetylcholine receptor
`obtained from patients with myasthenia gravis 2S Thus,
`various autoimmune diseases 'could be triggered by
`immunisation to particular microbial epitopes that mimic
`the structure of self-antigens.26
`There. is no obvious association between overt clinical
`tuberculosis and RA. Therefore, if antigens of mycobacteria
`are involved in triggering autoreactivity to.cross-reactive
`self-antigens in the joints, immunisation probably arises
`from inapparent contact.
`Why should the immune systems of some indiViduals
`respond to dangerously cross-reactive epitopes despite the
`existence of safely foreign epitopes on microbial antigens?
`Susceptibility to rheumatoid arthritis, as ' to other
`auto. immune diseases including adjuvant artliritis,27 is
`linked to major histocompatibility complex (HLA: DR)
`genes." Moreover, the identity of the immunologically
`dominant epitopes on a complex imrhunogen is also
`influenced . by the immune response genotype of the
`responding immune systems' together with other variables
`including the mode of immunisation." TherefOre, RA and
`other autoimmune diseases might be triggered when, as a
`consequences of genes and circumstances, a critical cross-
`reactive epitope is immunologically dominant. The target
`organ might then be attacked by mistaken identity. The
`findings of Ottenhoff et al31 may well be relevant-namely,
`that the DR4 allele associated with increased susceptibility
`to RA is associated also with enhanced delayed type
`hypersensitivity to PPD. Whatever the mechanisms, we
`know that mycobacterial antigens may be arthritogenic in
`human beings as well as rats, since Torisu and colleagues32
`reported arthritis as a side-effect o