Fluctuations in Anti-Nrnp Levels in Patients with Mixed Connective Tissue Disease Are Related to Disease Activity As Part of a Polyclonal B Cell Response

Ann Rheum Dis: first published as 10.1136/ard.45.10.800 on 1 October 1986. Downloaded from

Annals of the Rheumatic Diseases, 1986; 45, 800-808 Fluctuations in anti-nRNP levels in patients with mixed connective tissue disease are related to disease activity as part of a polyclonal B cell response

P M HOUTMAN,1 C G M KALLENBERG,2 P C LIMBURG,2 M A VAN LEEUWEN,1 M H VAN RIJSWIJK,' AND T H THE2 From the Departments of 1Rheumatology and 2Clinical Immunology, University Hospital, Groningen, The Netherlands

SUMMARY In a follow up study of 11 patients with mixed connective tissue disease the levels of antibodies to nuclear ribonucleoprotein (nRNP) as measured by an enzyme linked immunosorbent assay (ELISA) were related to clinical activity of disease. To assess the relation between anti-nRNP levels and disease activity the levels of total immunoglobulin G, IgM rheumatoid factor (IgM RF), and antibodies to an unrelated antigen (tetanus toxoid) were determined simultaneously. No significant changes in anti-nRNP levels were noted in four patients with minor activity of disease. Major flares of disease were observed in seven patients. Clinical symptoms

were preceded by a rise in anti-nRNP level in these patients unless they received immunosuppres-copyright. sive agents before the exacerbation. Conversely, when a rise in anti-nRNP level occurred a major flare of disease was followed in all but one case. Anti-nRNP levels fell during clinical improvement whether or not immunosuppressive treatment was given. All patients showed parallel fluctuations in anti-nRNP, IgM RF, and total immunoglobulin G levels. Furthermore, parallel fluctuations were seen in the levels of anti-nRNP and antibodies to tetanus toxoid except in one patient. We conclude that measurement of anti-nRNP by ELISA may be a guide for disease activity in connective tissue disease. Fluctuations of anti-nRNP are not restricted to this http://ard.bmj.com/ antibody, however, but are part of a more polyclonal activity of the B lymphocyte system.

Mixed connective tissue disease (MCTD)' is serolo- receptors4 or through increased permeability of cell gically characterised by the presence of antibodies to membranes.5 The antigen belongs to the so-called nuclear ribonucleoprotein (nRNP). The clinical small nuclear RNPs (snRNP). These proteins are syndrome is an overlap of systemic lupus erythema- involved in the mechanism of splicin6g of messenger tosus (SLE), scleroderma, polymyositis, and RNA precursors in the cell nucleus. As such, anti- on September 24, 2021 by guest. Protected rheumatoid arthritis. The significance of antibodies nRNP might influence protein synthesis at the to nRNP (anti-nRNP) as marker antibodies for a cellular level. Besides biochemical studies on struc- distinct disease entity has been controversial.2 Re- ture, organisation, and function of snRNPs, clinical cent observations have attracted new interest for studies on the relation between serum levels of anti- anti-nRNP as a possibly pathogenic factor in con- nRNP and disease manifestations may help to nective tissue disease. Nuclear deposits of immuno- elucidate the pathogenic role of anti-nRNP. Unlike globulins in epidermal and dermal cells have been antibodies to double stranded DNA (anti- considered as a real in vivo phenomenon in patients dsDNA)7 8 anti-nRNP has not been studied exten- with anti-nRNP antibodies.3 In vitro studies have sively in relation to disease activity in long term shown that living cells may be penetrated by follow up studies. In addition, measurement of anti-nRNP either by way of cell membrane Fc serum levels of anti-nRNP as described in previous studies has been performed by titration techniques Accepted for publication 17 April 1986. with unpurified antigens, such as the immuno- Correspondence to Dr C G M Kallenberg, Department of Clinical Immunology, University Hospital, Oostersingel 59, 9713 EZ fluorescence, immunodiffusion, and haemagglutina- Groningen, The Netherlands. tion techniques.9 800 Ann Rheum Dis: first published as 10.1136/ard.45.10.800 on 1 October 1986. Downloaded from

Fluctuations in anti-nRNP levels in patients with MCTD 801

E E E E E E Es£C E g E_ C EC-in ° -0 - °£ 05

II mC u 0.4 - . x -c -E

r- c T X - r- -~ z _ M en Wm C 'T "It " E 0)II c 0 r.M. 0. t3 . 0 CL 0 0u 11

'U o 0 -o - 'U N 'U..L u._ 0 11

U copyright. I -~ c o ° 0 1-

-0S 11 C) ;s S la http://ard.bmj.com/

-Q oa:t *_ on September 24, 2021 by guest. Protected

,cs

- t .S _ 00 00S4" -D o x x sr of x CU) HD enZZZ

_ U. U. U. U. _ %C

- nq n10%

r- 00 O_ _4 Ann Rheum Dis: first published as 10.1136/ard.45.10.800 on 1 October 1986. Downloaded from

802 Houtman, Kallenberg, Limburg, van Leeuwen, van Rijswijk, The

In a previous report we described an enzyme scored according to Becker et al,'2 with some minor linked immunosorbent assay (ELISA) for measure- modifications (Table 2). A patient was considered to ment of anti-nRNP. "' We showed that the ELISA is have stable disease activity when major manifesta- as sensitive as counterimmunoelectrophoresis in the tions did not occur during the follow up period. detection of anti-nRNP. Major disease manifestations were defined as the In the present study we have evaluated the presence of severe proliferative glomerulonephritis, measurement of anti-nRNP by ELISA as a guide for severe haemolytic anaemia (Hb <5 g% (50 g/l)) or disease activity in patients with mixed connective thrombocytopenia (<50 000/mm3 (SOx 109/l)), tissue disease. To delineate the specific relation severe CNS involvement (convulsions, coma, trans- between fluctuations in anti-nRNP levels and dis- verse myelitis), serositis on examination (persistent ease activity we determined simultaneously levels of pleural or pericardial rub, abnormal x ray, or ECG), antibodies to an unrelated antigen (tetanus toxoid), myositis, or major vasculitis. Manifestations of and levels of IgM rheumatoid factor, another disease that could be attributed on clinical grounds autoantibody frequently present in the sera of these to causes other than connective tissue disease, such patients. Since autoantibodies in connective tissue as infections, untoward reactions to drugs, etc, were disease belong predominantly to the immunoglobu- excluded from the analysis. None of the patients was lin G (IgG) class" we also measured total IgG in all immunised with tetanus toxoid during a period of serum samples. The results of our study show that three years before or during the study period. major flares of disease are accompanied by fluctuations in anti-nRNP. These fluctuations are, how- MEASUREMENT OF ANTI-nRNP ever, not specific for anti-nRNP. Serum levels of anti-nRNP were determined in an ELISA system as described previously.'0 Briefly, the nRNP antigen was prepared from rabbit thymus Patients and methods acetone powder (Pel Freez, Arkansas) by fraction-

ated ammonium sulphate precipitation and affinity copyright. PATIENTS AND SERA chromatography. In order to prevent proteolytic Serum samples containing antibodies to nRNP-as degradation of the antigens all buffers contained determined by counterimmunoelectrophoresis protease inhibitors. The eluate was extensively (CIE)-were taken from 11 patients. Clinical data dialysed and coated onto microtitre wells. The of the patients are presented in Table 1. The mean optimum protein concentration in the antigen solu- period of follow up of the patients was 35 months tion appeared to be 1-5 ,ig/ml or 0-15 Rg/well. (range 17-51 months). Seven patients were treated Anti-nRNP concentrations were expressed in rela- tive units/ml at a 1:100 dilution. The concentration with immunosuppressive therapy (corticosteroids, http://ard.bmj.com/ azathioprine, cyclophosphamide, or methotrexate in our reference serum was arbitrarily fixed at 100 (MTX)) at some time during the study period. units. This reference serum was used for measure- Patients were seen at regular intervals at the ment of anti-nRNP in all sera, except for those of outpatient department during periods of relatively patient No 11. The slopes of the dilution curves of stable disease activity and were admitted to the the sera of this patient were not parallel to the slopes hospital during flares of major disease activity. of the curves for other sera, including the reference Serum samples were drawn at each visit to the out- serum, but were parallel to each other. Therefore a patient department and every two weeks during pool of all serum samples of this patient was used as on September 24, 2021 by guest. Protected admission. All sera were also tested for antibodies a reference-also arbitrarily fixed at 100-in her to double stranded (ds) DNA by the indirect longitudinal study. Positive results in ELISA were immunofluorescence technique on Crithidia luciliae, obtained only in sera with anti-nRNP or anti-Sm and for antibodies to other extractable nuclear specificity, or both, as determined by CIE. Sera with antigens (anti-Sm, anti-SSB, and anti-SSA) as de- antinuclear antibodies of specificities other than tected by counterimmunoelectrophoresis (CIE). anti-nRNP or anti-Sm yielded values less than 1-5 Only one patient (No 11) showed a positive test for units, which is comparable with the results obtained anti-dsDNA at the time of her major exacerbation. in healthy controls. Anti-Sm was detected in the sera of patient No 7, Testing of positive sera in different ELISA plates anti-SSB in the sera of patient No 4, and anti-SSA in and on different days yielded an interassay variation the sera of patients Nos 4 and 9. The interval of less than 10%. between the samples never exceeded three months A significant change in the level of anti-nRNP was unless the patient had stable disease activity arbitrarily considered present when the level of throughout the entire follow up period. At the time anti-nRNP increased or decreased 1-5 times com- of serum sampling clinical disease activity was pared with the previous sample. Ann Rheum Dis: first published as 10.1136/ard.45.10.800 on 1 October 1986. Downloaded from

Fluctuations in anti-nRNP levels in patients with MCTD 803 Table 2 Disease activity index in patients with anti-nRNP antibodies*

Number ofpoints assigned Clinical General: fatigue 3 history of unexplained fever 3 clinic temperature :38°C 2 blood pressure > 140/90 mtnHg 2 lymphadenopathy 2 Mucocutaneous: active rash 6 active alopecia 5 oral or nasal ulcerations 4 Musculoskeletal: history of arthralgia or arthritis 3 arthritis one to three joints 2 arthritis > thrce joints 3 myositis CPKt > 150 U/I (normal < 50 U/I) 2 CPK > 300 U/I 6 Cardiopulmonary: pleural or pericardial pain 5 pleural or pericardial rub 3 abnormal x ray or ECG 2 vasculitis-major (ulceration, mononeuritis) 6 vasculitis-minor (purpura, periungual infarcts) 4 CNSt: seizure 4 stupor/coma 3 confusion, depression 3 copyright. cranial nerve palsy 2 cerebrovascular accident, evolving 4 choreoathetosis 4 visual field loss-subjective 2 -objective 3 migraine 2 Laboratory Urine analysis: > 2 g protein/24 h > five red blood cells/high power field or presence of casts Haemoglobin, g%:t <10-5 2 http://ard.bmj.com/ <9.5 3 White blood cell count/mm3:t <4500 1 <3500 2 Thrombocyte count/mm3:t <100 000 2 < 50 000 3 *According to Becker et al'2 with minor modifications. tCPK=creatine phosphokinase; CNS=central nervous system. tSI conversion: haemoglobin, g% x 10=g/1; white blood cell count/mm3x 106=cells/l; thrombocyte countlmm3x 106=cells/l. on September 24, 2021 by guest. Protected

MEASUREMENT OF ANTIBODIES TO TETANUS well. Sera were diluted at least 1:100 in Tween- TOXOID AND IgM RHEUMATOID FACTOR bovine serum albumin buffer and incubated for 45 Serum levels of antibodies to tetanus toxoid and min at 37°C in both ELISA systems. Diluted IgM rheumatoid factor (IgM RF) were assessed by (1:4000) antihuman IgG conjugated to horseradish an ELISA with essentially the same method as used peroxidase (HRP) (Kallestadt) was used as conju- for the measurement of anti-nRNP levels. gate in the antitetanus ELISA and 1:500 diluted For the determination of antitetanus antibodies goat F(ab)2 antihuman IgM conjugated to HRP 150 [I tetanus toxoid (RIV, Bilthoven, The Nether- (Cappel) as conjugate in the IgM RF ELISA. The lands, code VGT No 23) at a concentration of one antitetanus and IgM RF concentrations were ex- Limes flocculationis/ml was coated onto microtitre pressed as units and international units/ml using the plates (Titertek). For the determination of IgM RF standard curve of a positive antitetanus reference 100 ,Il heat aggregated IgG (30 min, 56°C) at a pool and rheumatoid factor reference serum concentration of 40 jig/ml was coated per microtitre (Behring, Diagnostika), respectively. Ann Rheum Dis: first published as 10.1136/ard.45.10.800 on 1 October 1986. Downloaded from

804 Houtman, Kallenberg, Limburg, van Leeuwen, van Rijswijk, The 1982 1983 1984 Patient No2 when the sera of the patient were positive for IgM I RF (yielding values above 40 IU/ml). 4000

3000 Results

2000 v E FLUCTUATIONS OF ANTI-nRNP LEVELS IN RELATION TO DISEASE ACTIVITY 100ioo `' Four patients (Nos 1-4) did not suffer from major flares of disease during the entire period of the 200- follow up study (Table 1). In these patients there was no significant change in the levels of anti-nRNP 180- or the index of disease activity as assessed by 160- Kendall's analysis. An example of antibody profiles in patient No 2 with stable disease activity is given in 140. , Fig. 1. ,, P Fourteen major flares of disease occurred in seven E 120- patients (Nos 5-11. Table 1) during the follow up 2 100- Xp,t/ "\ ,," period. In two cases (Nos 8 and 10) a flare of disease was already present at the beginning of the follow ; I 80- 60 up. Thus changes in anti-nRNP levels preceding the 60- disease manifestations could be studied in relation to 12 flares (Table 3). In three cases (Nos 5, 7 (Fig. 40 2) and 11 (Fig. 4)) no immunosuppressive therapy ,cFo---X -a,s~ was given in the period before the first exacerbation. 20 copyright. _3 - \i I--C A rise of anti-nRNP preceded the major disease

Index of I manifestations in these patients. Nine episodes of disease activity 3 8 14 8 878 33 3 8 5 major disease activity developed in three patients Fig. I Patient No 2, ,nale, bortn in 1956, had stable disease (Nos 6 (Fig. 3). 8. and 9) on immunosuppressive activityv during the follow up period. He (complained of therapy. Four episodes. all serositis, were not Ravnaud's phenomenoni, intermittenit arthralgia. and cliest preceded by a rise in anti-nRNP levels. In five pain suggestive ofpleuris' which could be substantiated. episodes anti-nRNP levels rose before the exacer- A blutterfly rash developed in Ma'y 1982. He was never bation. These data are summarised in Table 3. admnitted to the hospital because of a major flare lip ofhis Six out of the seven patients with major disease http://ard.bmj.com/ disease. Anti-nRNP levels were constant at a low level during the enitire period. SI conversion. manifestations were treated with immunosuppres- IgGmGng'`% x IO = mgll. sive agents (Table 1. Figs 2-4) during the follow up period. In all these patients a decrease in anti-nRNP levels was observed associated with clinical recov- ery. In one patient (No 5) clinical improvement and All serum samples from one patient were assayed concomittant decrease of anti-nRNP levels occurred simultaneously for anti-nRNP, antitetanus anti- before institution of immunosuppressive agents. on September 24, 2021 by guest. Protected bodies. and IgM RF by ELISA. We related fluctuations in anti-nRNP levels to changes in disease activity as determined by a MEASUREMENT OF IMMUNOGLOBULIN G scoring system based on the diversity and the ( IgG ) severity of the clinical symptoms. In five out of the Serum levels of IgG were determined by neph- seven patients with major flares of disease a elometry. significant correlation was found between anti- nRNP levels and index of disease activity during the STATI STI CS follow up period (Table 4). The inter-relationship between changes in anti- nRNP and antitetanus, IgM RF, total IgG, and PREDICTIVE VALUE OF A RISE OF index of disease activity respectively were calculated ANTI-nRNP LEVELS by Kendall rank order coefficients (Kendall's tau)13 A rise of the anti-nRNP level was seen 10 times in and tested for significance with Student's t test the group of 11 patients. In nine out of these 10 (degrees of freedom=n-2). Fluctuations of anti- cases this rise was followed by a major flare of nRNP were only considered in relation to 1gM RF disease. In most cases anti-nRNP levels rose in- Ann Rheum Dis: first published as 10.1136/ard.45.10.800 on 1 October 1986. Downloaded from

Fluctuations in anti-nRNP levels in patients with MCTD 805 Table 3 Episodes ofmajor disease activity in relation to anti-nRNPprofiles in patients with anti-nRNP

Major disease Patient Immunosuppressive Rise of anti-nRNPt manifestation* No therapy at the time levels preceding the of flare up flare of disease Severe serositis 6 + 6 + + 7 - + 8 + _ 8 + _ 9 + _ Myositis 5 - + 6 + + Vasculitis 6 + + E 9 + + 2 9 + + 11 - + O.a 280- *As defined in 'Patients and methods'. tA rise of anti-nRNP was considered present when anti-nRNP rose E = 240- at least 1-5-fold. E E

_- .200- ;z : 160- ,I_ sidiously. In patient No 7 (Fig. 2) these levels even bC1 r rose gradually up to 12 months before the outbreak 120- of the major manifestations. 80 copyright. SPECIFICITY OF THE FLUCTUATIONS IN 40 ANTI-nRNP LEVELS Rank order correlations were calculated to deter- index of I 0 mine whether fluctuations in the levels of anti-nRNP disease activity 3 5 3 3 B 13 16 8 were related to fluctuations in the levels of anti- pleuritis/pericarditis I bodies of other specificities. Another autoantibody Fig. 2 Patient No 7, female, born in 1953, developed (IgM RF), an antibody to an unrelated antigen arthralgia and Raynaud's phenomenon in 1982. In 1983 (tetanus toxoid), and total IgG levels were studied. arthritis ofthe hands and elbows developed, and http://ard.bmj.com/ In two out of the four patients with stable disease oesophageal hypomotility wasfound. Treatment with activity a significant correlation was found between hydroxychloroquine was instituted. She then started to fluctuations in anti-nRNP and antitetanus anti- complain ofpleuritic pains, increasing in severity. In bodies. Anti-nRNP levels correlated with the levels November 1983 a diagnosis ofpleuropericarditis was made. of total IgG in one of them. In six out of seven A continuous rise in the levels ofanti-nRNP and IgG was seen during thefollow up period startingfrom the very onset patients with major manifestations of disease a ofher disease. SI conversion: IgG mg% x 10=mg/l.

significant correlation was found between the fluc- on September 24, 2021 by guest. Protected tuations in anti-nRNP levels and the fluctuations in the levels of IgM RF, antibodies to tetanus toxoid, and total IgG (Table 4). The remaining patient, No proved to be antigen specific and to allow the 7, showed a parallel rise of anti-nRNP and total IgG detection of fluctuations in anti-nRNP levels in levels without any change in the level of antitetanus relation to disease manifestations.'0 Definition of antibody, whereas IgM RF was persistently negative disease activity in connective tissue disease is (Fig. 2). Results are summarised in Table 4. hampered by the diversity of the disease manifestations. Assessment of disease activity may be approached in two ways: firstly, by assessing the Discussion severity of each manifestation separately and secondly, by summing the number of manifestations Accurate measurement of antibodies and definition of clinical disease. For the latter approach we have of disease activity are a prerequisite for longitudinal applied the index of disease activity for SLE of studies of the relation between antibody levels and Becker et al. 12 We introduced some slight modifica- disease manifestations. The ELISA for measure- tions, adding myositis as an important symptom in ment of anti-nRNP antibodies used in this study has MCTD and omitting results of antinuclear antibody Ann Rheum Dis: first published as 10.1136/ard.45.10.800 on 1 October 1986. Downloaded from

806 Houtman, Kallenberg, Limburg, van Leeuwen, van Rijswijk, The 1980 1981 1982 1983 1984 Patient No 6 1982 1983 1984 Patent NoII prdxooE1m pteawnlo010nEtmg) 40 0 '22r,3fha0,p,ne 1lzitPrrr ..g 100 M~~~~~-TX , 50 10

[") 7s

4000

-3000J

2000 8 E 1000 -T

-I 200 - 200

180- 180

9 160- 160

: 140- 140 J

E 120- 120 E C : _ :v)- 100- c 100 0. -rI E a, 4,A- X , E 80- so "A 60 80 -

c-0 I~ copyright. 60 - b-o X 60 a: \ O2,\ a\ 40- ,l-// 40 20 20

index of disease activity 14 19 19 14 192223212121141715 1715 9 11 index of L cutaneous vasculitis disease activity 10 7 1511 7 1010 1313 36 7 6 4 3 $ http://ard.bmj.com/ myositis Fig. 4 Patient No 11, female, born in 1944, was referred in pericarditis tt - I 1983 because severe ulceration on the to pulmoncary vascutitis t of lower legs due vasculitis. She had a long history ofRaynaud's Fig. 3 Patient No 6, female, born in 1958, was admitted to phenomenon, pleuritic chest pain, febrile periods, the hospital in October 1979 because ofloss of weight, leucopenia, arthralgia, and sicca syndrome. A rise in muscle weakness, and arthralgia. The first rise in the level of anti-nRNP levels preceded theflare of vasculitis. anti-nRNP coincides with an increase in creatinine Anti-nRNP levels decreased with treatment and resolution phosphokinase (CPK) values compatible with the ofthe ulcers. SI conversion: IgG mg% x10=mgll. development ofmyositis. Despite therapy with prednisolone on September 24, 2021 by guest. Protected (17 5 mglday) symptoms worsened. Plasmapheresis was performed while prednisolone dosage was kept constant. Clinical symptoms improved and CPK values returned to tests and levels of complement component C3, as normal. She was readmitted to the hospital because offever and pericarditis in October 1980. At that time the level of these immunological parameters are probably re- When the anti-nRNP was still decreasing. The dosage ofprednisolone lated to the levels of anti-nRNP antibody. was increased to 30 mglday and symptoms subsided. A manifestations of disease most commonly encoun- second rise in the level ofanti-nRNP coincided with a major tered are included and arbitrary points assigned to exacerbation ofher disease in June 1982 manifesting as each manifestation the index combines both of the pericarditis and severe pulmonary hypertension with right above approaches. It does not discriminate, how- sided cardiacfailure probably due to pulmonary vasculitis. ever, between one single manifestation of major Therapy was started in July 1982 with methvlprednisolone activity and several manifestations of minor activity. and (1000 mglweek intravenously) cyclophosphamide For this reason disease activity was not only (100 mglday orally). Symptoms subsided and the patient approached by the defined index of disease activity was able to resume her daily work. The anti-nRNP level but occurrence of major manifestations decreased and kept in the lower range on therapy with 7 mg also by the ofprednisolone daily during 1984. S conversion: IgG (myositis, serositis, and vasculitis in our group of mg% x 10=mg/I. patients). Ann Rheum Dis: first published as 10.1136/ard.45.10.800 on 1 October 1986. Downloaded from

Fluctuations in anti-nRNP levels in patients with MCTD 807 Table 4 Kendall rank order correlation coefficients between levels ofanti-nRNP and levels ofantitetanus antibodies, IgM RF, total IgG, and the index ofdisease activity

Patient Number of Fluctuations of anti-nRNP in relation to: Representation No serum samples in figure Antitetanus IgM RF IgG Index of antibodies disease activity 1 5 0 t1- 2 13 0.38* t 0-13 1 3 7 071* 0-41 0-33 t 4 10 0-10 030 0-17 5 15 066*** 0.73*** 0.77*** 0.72*** 6 15 0-66*** 0.69*** 0.59** 0.57** 3 7 8 -0-37 t 0.71** 0-39 2 8 25 0.64*** t 0.54*** 0.27* 9 24 0-20 0.61*** 0.52*** 0-15 10 9 0.61* 0-49* 0.63* 0.61* 11 16 0.45** 0.54** 0.61** 0-50** 4 *p<0-05; **p

In four patients without major disease manifesta- decline in anti-nRNP concurrent with clinical recov- tions no significant changes in anti-nRNP levels ery as observed in the only patient who was not were seen. On the contrary, in seven patients with treated suggests, however, that spontaneous decline major fla es of disease significant fluctuations of in the level of autoantibodies is also associated with copyright. these levels were observed. Since immunological decrease in disease activity. Another mechanism processes probably precede the clinical manifesta- resulting in lowering of antinuclear antibody levels is tions of disease we studied changes in the autoanti- the formation of complexes with free autoantigen. body level just before the exacerbation. Eight out of Free nRNP16 has been shown like free DNA'7 to be 12 major flare ups were preceded by a rise of the released in the serum when prednisone therapy is anti-nRNP level; in five cases despite the use of initiated or increased. Declining levels of anti-nRNP immunosuppressive therapy. The four episodes of were not associated with increase in the levels of serositis not preceded by a rise of anti-nRNP circulating immune complexes (results not shown). http://ard.bmj.com/ occurred while the patients were taking immunosup- To investigate whether anti-nRNP is a specific pressive agents. These data suggest a relation guide for assessing disease activity we studied the between anti-nRNP levels and disease activity but relation between fluctuations in anti-nRNP levels do not support a direct pathophysiological role for and antibodies of other specificities. We chose anti-nRNP antibodies. Conversely, a rise of anti- another autoantibody (IgM RF) and an antibody to nRNP appears to predict a major disease manifesta- an unrelated antigen (tetanus toxoid). The finding tion irrespective of the use and dosage of immuno- of parallel fluctuations, even with respect to total on September 24, 2021 by guest. Protected suppressive agents: a major flare was observed after IgG levels, points to polyclonal B lymphocyte nine out of 10 periods of rising anti-nRNP levels. It activation during active disease. Patients with auto- is noteworthy that the rise of the anti-nRNP level immune disease like SLE and MCTD make a large was usually very gradual and lasted for six to 12 amount of immunoglobulins, and a large proportion months before the flare was apparent. The possibil- of these may be autoantibodies." 18 Raised levels of ity that early institution of immunosuppressive antibodies against common viral antigens were agents may prevent the development of major flares recorded by several investigators in patients with is currently under study. SLE.19 20 Phillips and Christian even observed a Levels of anti nRNP and antibodies of other relation between clinical activity of disease and specificities decreased with clinical improvement, in levels of antibodies against viruses.20 Several in vitro general with the concurrent institution of immuno- studies have also shown polyclonal B cell activation suppressive therapy. Corticosteroids are known to in patients with active SLE. 23 Whether anti- suppress IgG synthesis in vivo. 14 Declining levels of nRNP antibodies, by abrogation of T suppressor cell immunoglobulins can also be attributed to azathio- function,24 have a role in polyclonal B cell activation prine, cyclophosphamide, and methotrexate.'5 The is not clear. Ann Rheum Dis: first published as 10.1136/ard.45.10.800 on 1 October 1986. Downloaded from

808 Houtman, Kallenberg, Limburg, van Leeuwen, van Rijswijk, The In conclusion, longitudinal measurement of anti- 10 Houtman P M. Kallenberg C G M. Limburg P C. Iluitcmai M nRNP antibodies is of clinical value in predicting G. van Rijswijk M 11. The T 1. Quantitaition of aintibodies to nucleoribonuclcoprotcin (nRNP) bv ELISA. ( Eoxrp Imitiitoiol and assessing disease activity in patients with these 1985: 62: 696-704. antibodies. Fluctuations in anti-nRNP are not speci- 11 Maddison P J. Reichlin M. Quantitation of precipitating fic markers, however, but are an expression of antibodics to ccrtatin solublc nuclcir antigens in SLF. IArthritis fluctuations in polyclonal B cell activity. Thus for Rheumn 1977. 20: 819-24. 12 Bcckcr T M. Lizzio F F. Mcrchant B. Rcescs J P. Steinbherg A clinical purposes total IgG levels may equally well D. Incrcascd multiclonal antibods-forming ccll acttisits in the be used. Nevertheless, further elucidation of the pcriphcrial blood of patients with SLE. Itit ,4rch Allertv App! pathogenetic significance of anti-nRNP antibodies is homunlit ol 1981: 66: 293-303. 13 Siegcl S \ont paratnietrit Statistitcs for tihe lehaviourtIsr(co te(CS warranted. New York: McGraw-Ilill. 1956. 14 Butler W T. Rosscn R D. Effccts of corticostcroids on immunits in man. 1. Dccreased scrum IgG concentration ciauscd The authors wish to thank Minke Huitema for technical by 3 or 5 days of high doscs methvlprednisolonc. J (Cht inveit assistance. This study was supported by the 'Praeven- 1973: 52: 2629-40. tiefonds' and the Netherlands League against Rheumat- 15 Hurd E R. Immunosupprcssisc and anti-inflarmmators prop- ism. crtics of csclophosphamidc. azathioprinc aind methotrexatc. Arthriti.s R/ieutm 1973. 16: 84-7. 16 Fishbcin E. Ramos-Nicmbro F. Alarcon-Scgovia D. Free scrum nuclcoprotcin in mixcd conncctivc tissuc discasc iind References othcr conncctisc tissuc discascs. J Rlieutnuato/ 1978: 5: 384-9. 1 Sharp G C. Irvin W A. Tan E M. Gould R G. Holman H R. 17 llughcs G R V. Cohcn A S. Lightfoot R W. Fhc rclcisc of Mixed connective tissue discasc-an apparcntls distinct DNA into scrum aind srnosvial fluid. Arthritis R/heum 1971. 14: rheumatic disease syndrome associated with a spccific antibods 9-66. to an extractable nuclear antigen (ENA). Ami J Med 1972: 52: 18X-lolman 11 R. Systemic lupus crythematosus-disacasc ot ain 148-59. unusual immunological responsiveness'? Ain J Medl 19'39:, 27: 2 Reichlin M. Problems in differentiating SLE and mixed 52SX. connective tissue disease. N Engl J Med 1976; 18: 1194-5. 19 Hurd E R. Dowdle W. Casev 11. Virus antibodv lcsvcs in

3 Gilliam J N, Prystowsky S D. Mixed connectivc tissuc discasc systemic lupus crvthemaitosus. Arthritis Rheiumn 1972. 15:copyright. syndrome: the cutaneous manifestations of patients with high 267-74. titer serum antibody to RNase sensitivc cxtractable nuclear 2(0 Phillips P E. Christian C L. Virus antibodies in systemic lupus antigen (ENA). Arch Dermatol 1977; 113: 583-7. erythematosus and other connectivc tissue diseases. Attni 4 Alarcon-Segovia D, Ruiz Arguelles A. Fishbein E. Antibody to Rheum Dis 1973: 32: 450-6. nuclear ribonucleoprotein penetrates live human mononuclear 21 Fauci A S. Immunoregulation in autoimmunitv. J Allergy (Clht cells through Fc-receptors. Nature 1978; 271: 67-9. lmmnu,iol 19811: 66: 5-17. 5 Izuno G T. Observations on the in vivo reaction of antinuclear 22 Hoch S. Schur P H. Schwaber J. Frequencs of anti-DNA antibodies with epidermal cells. Br J Der,natol 1978; 98: 391-8. antibodv producing cells from niormals and patients with 6 Lerner M R, Boyle J A, Mount S M. Wolin S L. Steitz J A. Are systemic lupus erythematosus. Cin Imtnmnutiol ltmmuniiopatitol

nRNPs involved in splicing? Nature 198(0; 283: 220-4. 1983: 27: 28-37. http://ard.bmj.com/ 7 Lightfoot R W, Redecha P B. Levesanos N. Longitudinal 23 Kallenberg C G M. Limburg P C. van Slochteren C. san dcr studies of anti-DNA antibody levels in SLE. Sca,idJ Rheumnatol Woude F J. The T H. B-ceil activity in systemic lupuIs 1975; suppl 11: 52-8. erythematosus: depressed in vivo humoral response to primnar 8 Swaak A G J, Aarden L A. Statius van Eps L W, Feltkamp T E antigen (haemocyanin) and increased in vitro spontaneous W. Anti-dsDNA and complement profiles as prognostic guidcs immunoglobulin synthesis. C/in Exp lIn,nunol 1983: 53: 371-83. in systemic lupus erythematosus. Arthritis Rheum 1979; 22: 24 Alarcon-Segovia D. Ruiz-Arguelles A. Llorente L. Antibodv 226-35. penetration into living cells. II. Anti-ribonucleoprotein IgG 9 Moore T L, Weiss T D, Neucks S H, Baldassare A R, Zuckncr penetrates into T lymphocytes causing their deletion and the J. Extractable nuclear antigens. Semin Arthritis Rheum 1981; abrogation of suppressor function. J Immnunol 1979: 122: 10: 309-18. 1855-62. on September 24, 2021 by guest. Protected