Sister-chromatid exchange frequencies in fibroblasts and lymphocytes of patients with systemic lupus erythematosus

125

Mutation Research, 177 (1987) 125-132

Elsevier MTR 04307

Sister-chromatid exchange frequencies in fibroblasts and lymphocytes of patients with systemic lupus erythematosus R . G . P a l m e r , C.J. D o r e 1, L. H e n d e r s o n 2 a n d A . M . D e n m a n Connectwe Tmsue Diseases Research Group and Dzoiston of Medical Statisttcs ~, Climcal Reseach Centre, Watford Road, Harrow, Middlesex HA1 3UJ (Great Brttmn) and School of Biological Sciences 2, Unwerstty of Sussex, Falmer, Brighton, Sussex BN1 9RR (Great Brttain)

(Received28 January1986) (Revisionreceived13 October1986) (Accepted15 October1986)

Key words: SCE frequencies;Fibroblasts; Lymphocytes;Systemiclupus erythematosus;Serumanti-nuclearantibodies(ANA); DNA

repair defects.

Summary Sister-chromatid exchange (SCE) frequencies have been measured in lymphocytes and fibroblasts of patients with systemic lupus erythematous (SLE) and healthy controls, and in lymphocytes of control patients with serum anti-nuclear antibodies (ANA) but no other disease manifestations of SLE. The SCEs of SLE lymphocytes were higher than those of the controls but the SCEs of the SLE fibroblasts did not differ from those of the controls. The SCEs of the controls with positive ANA did not differ significantly from those of the healthy controls. There was no correlation between SCE frequencies of the SLE lymphocytes and disease activity determined by many clinical and laboratory measurements. Primary and secondary DNA-repair defects in SLE ceils are considered.

Systemic lupus erythematosus (SLE) is one of the best studied of the autoimmune diseases. Polyclonal B cell proliferation and autoantibodies especially to nuclear components - - the antinuclear antibodies (ANA) - - and more specifically antibodies to D N A are characteristic of this condition. Patients commonly present with arthritis or skin lesions and may also have haematological abnormalities and renal or central nervous system involvement (Tan et al., 1982). The ultraviolet (UV) component of sunlight is often

Correspondence: Dr. R.G. Palmer, Clinical Research Centre, Division of Rheumatology,WatfordRoad, Harrow, Middlesex HA1 3UJ (Great Britain).

responsible for inducing the skin lesions and it may also precipitate an exacerbation of the disease (Tan, 1976). Ultraviolet light directly damages DNA (Heseltine, 1983) and it has been postulated that SLE occurs in individuals whose cells have deficient DNA repair. Several authors have claimed that lymphocytes from patients with SLE are less able to repair damage induced by UV light than lymphocytes from healthy individuals (Beighlie and Teplitz, 1975; Horkay et al., 1975; Altmann, 1977) and others have observed that SLE lymphocytes have a reduced capacity to excise the abnormal base of O6-methylguanine induced by methyl nitrosourea (Harris et al., 1982). Limited data suggest that SLE fibroblasts may have a normal capacity of repairing, but a reduced

0027-5107/87/$03.50 © 1987 ElsevierSciencePublishersB.V.(BiomedicalDivision)

126

capacity for surviving. DNA damage induced by UV light (Zamansky et al., 1985). Tuschl et al. (1984) reported that spontaneous sister-chromatid exchange (SCE) frequencies in lymphocytes of patients with SLE are higher than SCE frequencies found in the lymphocytes of control individuals and discussed the possibility of a deficiency of D N A repair in SLE cells. Some of these SLE patients were being treated with cytotoxic drugs, notably cyclophosphamide, and others may have received these drugs in the past. It was not possible to be sure that these drugs were not responsible for some of the high SCEs that were observed. We have measured SCE frequencies in both lymphocytes and skin fibroblasts of SLE patients and healthy controls to determine if the putative abnormality is confined to lymphocytes. Patients who had ever received cytotoxic drugs were excluded from the study. SCEs were also measured in the lymphocytes of individuals who had ANA in their serum to determine if ANA was responsible for the elevated SCEs found in SLE lymphocytes. These individuals had non-specific symptoms and no other manifestations of SLE.

Materials and methods

Pattents SCE frequencies were measured in the peripheral blood lymphocytes of 22 patients with SLE and 20 healthy controls, and in fibroblasts cultured from skin biopsies from 5 patients with SLE and 5 healthy controls. All SLE patients fulfilled the American Rheumatism Association (ARA) criteria for this disease (Tan et al., 1982). SCEs were also determined in peripheral blood lymphocytes from 4 individuals who were positive for ANA but who had no other features of SLE. The age, sex, race and smoking habits of all individuals were recorded. For each SLE patient many laboratory measurements were performed and the presence of absence of each disease feature used in the ARA criteria was noted. The laboratory measurements were the haemoglobin, erythrocyte sedimentation rate, white cell count, lymphocyte count, platelet count, serum IgG, IgA and IgM, complement components C3 and C4 and the total haemolytic complement (CH50). Patients were assessed for the following clinical features; malar rash,

discoid rash, photosensitivity, oral ulcers, arthritis, serositis, renal lesions, central nervous system lesions, haemolytic abnormalities, immunological abnormalities and ANA. The drug treatment of each patient was also assessed at the time of SCE analysis; the intake of non-steroidal anti-inflammatory drugs, chloroquine and prednisolone was determined for each individual. None of the patients or controls had ever been treated with cytotoxic drugs or any other drugs known to increase SCE frequencies.

Methods

(a) Lymphocytes Metaphase spreads demonstrating SCEs were prepared from whole blood cultures. In brief, 0.5 ml of whole blood, diluted in 6 ml of RPMI medium with 10% foetal calf serum and antibiotics, was stimulated with phytohaemagglutinin. BUdR, at a final dilution of 30 /~M, was added after 24 h, colchicine was added after 68 h and the cells were harvested after 72 h in culture. Slides were stained using the variation of Goto et al. (1978) on the technique of Perry and Wolff (1974). Details have been published (Palmer et al., 1984). Samples were coded and SCEs were determined in 20 metaphases from each individual.

(b) Fibroblasts Fibroblast cultures were prepared from fresh skin biopsies. Cells were passaged several times before being set up at a concentration of 5 X 104 per ml in medium (MEM with 15% foetal calf serum and antibiotics) in 25-cm2 Falcon flasks. After 24 h the medium was replaced with fresh medium containing 10 ~tM BUdR and the cultures were incubated in the dark in 5% CO 2. They were harvested by trypsinisation at 46 h after 4 h of colchicine treatment (1 /~M final concentration). Metaphase spreads for SCE analysis were prepared using the standard techniques already described for lymphocytes. At least 20 metaphases were scored for each cell line.

Statistical analysis

(a) Lymphocytes The results of the SCE frequencies of the SLE

127 patients and the two control groups formed part of a large survey of SCEs in patients with various connective tissue diseases and controls (Palmer, 1984; Palmer et al., 1986). In the survey we found that the most appropriate transformation for removing the relationship between the standard deviations and the means of the metaphases counted for each individual was log (SCE + 1). SCE counts were logged after the addition of 1 to each original count and the mean log (SCE + 1) was calculated to summarize the data for each individual. The appropriateness of this transformation in achieving desirable properties of the summary for each individual was assessed using Schweder's test of homogeneity of variances between the 3 groups (Schweder, 1981) and Shapiro and Wilk's W test of Normality of residuals after fitting separate means to each group (Royston, 1982). For the untransformed data there was not equality of variance (Schweder's test, chi-squared = 8.4, df = 2, P = 0.02); the transformation produced both equality of variances (Schweder's test, chi-squared = 3.5, df = 2, P = 0.2) and Normality of residuals ( W = 0.98, P = 0.6). The results have been expressed as the geometric mean; in general, the geometric mean is slightly smaller than the arithmetic mean.

(b) Fibroblasts The distribution of the means of the untransformed data for each individual showed no evidence of unequal variances between groups (Schweder's test, chi-squared = 0.01, df = 1, P = 0.9) or departures from normality of the residuals (W --- 0.92, P = 0.4), contrasting with the distribution of the SCE data from lymphocytes. Results

(a) Lymphocytes The mean log (SCE + 1) and the geometric mean SCE frequencies of the SLE patients, the ANA positive controls and the healthy controls are shown in Fig. 1. The SCE frequencies of the SLE patients were significantly higher than those of the healthy controls ( P < 0.0001, one-way analysis of variance with Student's t test). The SCEs of the healthy controls were not significantly dif-

12-

13

11-10

~ o908-

i~

~

~

•

Heallhy c~ntrols

SLE patients

07

-7

~

-6

o

-4 ANA posit*re controls

Fig. 1. Vertical b a r s ~ m e a n a n d 95% c o n f i d e n c e interval for nleail.

ferent from those of the controls with positive ANA (P -- 0.04). These conclusions remained after the data had been adjusted for age, sex, race and smoking habits (adjusted data not shown). There was no significant relationship between the SCE frequencies and the presence of absence of each disease feature used in the ARA criteria for SLE (Table 1) and there was no significant relationship between SCE frequencies and the total number of positive disease features possessed by each SLE patient. SCE frequencies were also not related to any of the laboratory measurements that had been determined for each patient (Tables 2 and 3). The drug intake at the time of SCE analysis was recorded for each patient. The SCEs of those who were taking non-steroidal anti-inflammatory drugs were significantly lower than the SCEs of those who were not taking these drugs (P = 0.03, Table 4). Neither prednisolone nor chloroquine treatment influenced SCEs and for the 10 patients taking prednisolone there was no significant relationship between SCE frequencies and the daily dose of the drug (r = -0.11, P = 0.8). The effect of patient category and drug therapy on the rate of cell division was determined for some individuals (Tables 5 and 6). There was a suggestion that lymphocytes of normal individuals proliferate more rapidly than those of the SLE patients and ANA-positive controls and that cells of patients receiving non-steroidal anti-inflammatory drugs proliferate more rapidly than those of patients not receiving these drugs. These differences, however, did not reach statistical significance. Drug therapy did not affect lymphocyte counts (Table 7).

128 TABLE 1 RELATING DISEASE FEATURES OF MEAN log (SCE + 1) OF SLE PATIENTS Disease feature

If present

Number

Mean of mean log (SCE+ 1)

SD

Geometric mean SCE ~

ib

Malar rash

No Yes

13 9

0 97 0 93

0 07 0 08

83 75

100

03

Discoid rash

No Yes

18 4

0 95 0 96

0 07 0 08

79 81

017

09

Photosensinvity

No Yes

11 11

0 93 0 98

0.09 0 05

7.5 8.6

1

Oral ulcers

No Yes

21 1

0,95 0,92

0 07 -

7.9 73

0 39

Arthrins

No Yes

4 18

1 01 0.94

0 04 0.07

92 77

1

Seros~tis

No Yes

15 7

() 94 0.98

0.07 0 07

77 8.6

119

02

Renal lesion

No Yes

18 4

0 96 0.93

0 08 0 05

81 75

071

0.5

CNS lesion

No Yes

15 7

0.94 0 98

0 08 0 07

7.7 86

0 95

03

Haematological abnormality

No Yes

6 16

0 95 0.96

0 07 0 08

79 8.1

017

09

Odd patient number ~

No Yes

11 11

0 96 0.95

0 08 0.08

81 79

0 19

09

84

84

0 08 07 0 08

CNS, central nervous system anti log (mean log (SCE+ 1))- 1 h 2 sample t test. c this split, based on a patient's number (1-22), has been included as a control TABLE 2 CORRELATION OF MEAN log (SCE+I) AND LABORATORY MEASUREMENTS OF SLE PATIENTS Measurement

r

P

Hb ESR WCC Lymphocytes Platelets IgG IgA IgM C3 C4 1/CH50

- 0.21 0.11 0.13 - 0.03 - 0,06 0 17 0.32 - 0,06 0.01 0.38 0 26

04 0.6 0.6 09 08 0.5 0.2 0.8 09 0.1 0.3

Hb, haemoglobin. ESR, erythrocyte sedimentation rate. WCC, whate cell count.

(b) Fibroblasts The untransformed

mean SCE frequencies of

fibroblasts from the SLE patients and controls and the SCE frequencies of blood lymphocytes f r o m f o u r o f t h e five S L E p a t i e n t s a r e s h o w n i n T a b l e 8. T h e S C E f r e q u e n c i e s o f f i b r o b l a s t s f r o m the SLE patients did not differ significantly from t h o s e o f t h e h e a l t h y c o n t r o l s ( P = 0.6) a n d t h e r e was no significant relationship between the mean SCE frequencies of the fibroblasts of the SLE p a t i e n t s a n d t h e m e a n l o g ( S C E + 1) o f t h e i r l y m p h o c y t e s ( r = - 0 . 2 1 , P = 0.8).

129 TABLE 3 E F F E C T OF A U T O - A N T I B O D I E S ON SCEs

Measurement

If present

Number

Mean of mean log (SCE + 1)

SD

Anti-DNA antibodies

Yes No

15 7

0.96 0.94

0.07 0.09

ANA

Yes No

21 1

0.95 1.06

Antibodies to E N A a

Yes No

10 11

0.97 0.94

Geometric mean SCE

t

P

8.1 7.7

0.68

0.5

0.07

7.9 10.5

1.51

0.1

0.07 0.08

8.3 7.7

0.89

0.4

ANA, anti-nuclear antibody. ENA, extractable nuclear antigen. a Data not available from one patient.

TABLE 4 R E L A T I N G D R U G T H E R A P Y TO M E A N log ( S C E + 1) OF SLE PATIENTS Drug

If received

Number

Mean of mean log (SCE + 1)

SD

Geometric mean SCE

t

p

NSAI

Yes No

7 15

0.90 0.97

0.06 0.07

6.9 8.3

2.36

0.03

Chloroquine

Yes No

8 14

0.92 0.97

0.08 0.07

7.3 8.3

1.46

0.2

Prednisolone

Yes No

10 12

0.95 0.95

0.07 0.08

7.9 7.9

0.1

0.9

NSAI, non-steroidal anti-inflammatory drugs.

Discussion TABLE 5 T H E E F F E C T OF DIAGNOSIS ON P E R C E N T A G E OF CELLS IN E A C H DIVISION

Number of patients

Normals SLE

% of metaphases in each division

1st mean (SD)

2nd mean (SD)

3rd mean (SD)

11

22.6 (13.6)

35.9 (8.8)

41.0 (20.3)

5

36.8 (16.7)

39.2 (11.3)

24.0 (17.4)

2

32.0 (17.0)

36.0 (2.8)

32.0 (14.1)

ANA

positive controls p"

a Analysis of variance.

0.15

0.83

0.18

The SCE frequencies of peripheral blood lymphocytes from patients with SLE are higher than those of healthy controls both before and after allowance for the age, sex, race and smoking habits of the individuals. In contrast, the SCE frequencies of fibroblasts from patients with SLE did not differ from those of healthy controls. This suggests there may be an abnormality in blood lymphocytes of SLE patients which is responsible for the elevated SCE frequencies but which is not found in skin fibroblasts. It is possible, however, that only normal fibroblasts have been selected by passaging and that abnormal cells have not survived this procedure.

130 TABLE 6 EFFECT OF D R U G S ON CELL PROLIFERATION RATES Number of patients

-

SLE PATIENTS

of metaphases m each division lsl mean (SD)

2nd mean (SD)

3rd mean (SD)

NSAI

No Yes p ~

6 2

39 0 (16.2) 300(22.6) 0 55

39 7 (12 7) 380 (8 5) 0 87

21 3 (14 0) 320t31 1) 0 5(I

Chloroqume

No Yes p b

1 7

140 (-) 400 (15 0) 0 16

320 (-) 40 3 (11 7) (J 53

54(1 I ~) 19 7 (13 4) 005

Predmsolone

No Yes P L,

5 3

404(177) 30 7 (16 0) 0 47

352 (7 3) 46 0 (15 1} 0 21

244(13 1) 23 3 (26 6) t} 94

2 s a m p l e t test. b Analysis of variance NSAI, Non-steroid or ann-inflammatory drugs

Tushl et al. (1984) have considered defects of D N A repair as one of the possible factors responsible for elevated SCEs found in SLE patients. Several other authors have also proposed that defects in D N A repair are present in cells of SLE patients (Beighlie and Teplitz, 1975; Horkay et al., 1975; Altmann, 1977; Harris et al., 1982). Our observations on SCEs of SLE lymphocytes agree with those of Tuschl et al. but our findings on fibroblast SCEs, and the observation of Zamansky et al. (1984) that there was no excess of chromosome aberrations in SLE fibroblasts, suggest

TABLE 7 E F F E C T OF D R U G S ON COUNTS SLE PATIENTS

BLOOD

LYMPHOCYTE

Number of patients a

Lymphocyte counts/~l mean (SD)

P

NSAI

Yes No

6 15

1237 (976) i636 (175)

0.30

Chloroqume

Yes No

8 13

1 376 (873) 1612 (724)

0.5I

Prednisolone

Yes No

10 11

1555 (872) 1493 (710)

0.86

NSA1, Non-steroidal anti-inflammatory drugs d Data not available from one patient

that the putative defect is not present in all cell types and is therefore unlikely to be inherited. A defect in SLE fibroblasts cannot be excluded completely because the absence of increased SCEs is not proof of the absence of a DNA-repair defect in these cells. One possible explanation for elevated SCEs in SLE lymphocytes and not fibroblasts is that a defect has been acquired. Indeed, the observations of Okalie and Shall (1979) that antibodies to poly (adenosine-diphosphate-ribose) are present in serum of SLE patients is consistent with an acquired defect. This polymer is thought to participate in D N A repair (Durkacz et al., 1980) and inhibitors of the polymerase responsible for its production are known to induce SCEs (Oikawa et al., 1980). It is, therefore, possible that antibodies of the polymer itself are responsible for the high SCE frequencies in SLE. These specific antibodies have not been measured in our SLE patients. However, their A N A were determined and there was no relationship between the presence or absence of A N A and SCE frequencies. In addition, the SCEs of our A N A positive controls were normal. It is difficult properly to assess the published work of others who have considered DNA-repair defects in SLE. For example the 'SLE' patients studied by Beighlie and Teplitz (1975) were all in 'remission'. The patients in 'exacerbation' were

131

TABLE 8 SCE FREQUENCIES FROM SLE FIBROBLASTS AND LYMPHOCYTES AND FROM FIBROBLASTS OF HEALTHY CONTROLS Fibroblasts

Lymphocytes

Mean SCE

SD

Mean log (SCE+ 1)

SD

SLE patients

1 2 3 4 5

7.2 49 8.3 9.4 6.8

2.4 1.6 2.3 3.2 3.4

0.86 1.01 0.91 0.99 ND

0.14 0.11 0.14 0.17

Healthy Control

1 2 3 4 5

4.6 8.1 7.6 8.5 5.0

1.9 3.2 1.8 3.3 2.7

ND ND ND ND ND

Geometric mean SCE 6.2 9.2 7.1 8.8

ND, not done.

specifically excluded from the analysis. These terms were not defined. The exclusion from this type of study is hard to justify. In one paper (Altmann, 1977) the observations in the controls were not included. Another explanation for the elevated SCEs in lymphocytes of SLE patients is that they are induced by clastogenic factors. If present in the serum only they might increase SCE counts of lymphocytes but not increase the SCE counts of fibroblasts. These factors have been reported to be present in the serum of patients with a variety of disorders. The list is long and includes SLE, scleroderma, rheumatoid arthritis, dermatomyositis, polyarteritis nodosa, Crohn's disease, ulcerative colitis, chronic hepatitis, ataxia telangiectasia and Bloom's syndrome (reviewed by Emeritt, 1982). The ubiquitous nature of this factor suggests that it is of little importance in the pathogenesis of these diseases. Log transformation was necessary for the lymphocyte data but not for the fibroblast data. These findings indicate that for all SCE data the most appropriate transformation must be determined before results can be analysed and the type of transformation most suitable for one set of data may not be the most appropriate for another. The high SCE counts of our SLE patients could not be explained by the presence of any particular disease feature or the extent of the disease as

measured by the number of positive disease features possessed by a patient. There was also no evidence of a relationship between SCEs and any of the haematological or immunological measurements. In particular, SCE frequencies and blood lymphocyte counts were not significantly related. This is important because whole cultures were used and since patients with SLE are sometimes lymphopenic (low blood lymphocyte counts) it was necessary to establish that blood lymphocyte counts did not affect SCE frequencies. Although the proliferation rates of the cells from the patients and controls were not significantly different and drug therapy had no effect on these rates, a larger study may have demonstrated some differences. Therefore we cannot exclude the possibility that these two groups have different subpopulations of cells that divide at different rates, thereby affecting observed SCE frequencies (Parkes et al., 1985). It is equally possible that the unknown factors responsible for increased SCEs in SLE are also responsible for small changes in rates of proliferation, a cause and not an effect of variations in cell turnover. Chloroquine and prednisolone therapy did not increase SCE frequencies. This contrasts with the in vitro observation of Raj and Heddle (1980) that chloroquine increases SCEs. But patients with SLE who were taking non-steroidal anti-inflammatory drugs at the time of SCE analysis had lower SCE

132 frequencies than those not taking these drugs. The c a u s e o f t h i s is n o t k n o w n .

Acknowledgement We thank Dr. B.M. Ansell, Division of Rheumatology, Clinical Research Centre, for many of the blood samples from patients with SEE and S u s a n H a r c o u r t , M R C Cell M u t a t i o n U n i t , U n i versity of Sussex, Brighton for establishing skin fibroblast cultures from SLE patients and healthy controls.

References Ahmann, H (1977) DNA-repalr and mutations in lmmunocompetent ceils from patients with rheumatic diseases and corresponding animal models, Stud. Blophys., 61, 89-96. Baker, R J., and J A Nelder (1978) The GLIM system release 3 manual, Numerical Algorithms Group, Oxford Beighlie, D J., and R.L. Tephtz (1975) Repair of UV-damaged DNA in systemic lupus erythematosus, J Rheumatol., 2, 149-160. Durkacz, B.W, O. Omldij1, D.A. Gray and S Shall (1980) (ADP-ribose), participates in DNA excision repmr, Nature (London), 283, 593-596 Emerlt, I (1982) Chromosome breakage factors: origin and possible significance, in: A.T. Natarajan et al. (Eds.), Progress in Mutation Research, Vol 4, Elsevier, Amsterdana, pp. 61-74 Goto, K , S. Maeda, Y. Kano and T. Suglyama (1978) Factors involved in differentml giemsa-staming of s~ster chromatids, Chromosoma, 66, 351-359 Harris, G., L Asbery, P.D Lawley, A.M. Denman and W. Hyhon (1982) Defective repair of Or-methylguanine in autoimmune diseases, Lancet, 2, 952-956 Heseltine, W.A. (1983) Ultraviolet hght repair and mutagenes~s revisited, Cell, 33, 13-17 Horkay, I., E. Nagy, P Tamasi, M. Szabo and J. Csongor (1975) DNA repair and U V.-hght sensitivity of the lymphocytes in discoid lupus erythematosus, Stud Blophys., 50, 1-6. Olkawa, A., H. Tohda, M Kanal, M Masanao and T Sugimura (1980) Inlubitors of poly(adenosine diphosphate nbose)

potymerase induce sister chromatid exchanges, Biochem Biophys. Res. Commun., 97, 1311-1316 Okolie, E E, and S Shall (1979) The sigmficance of antibodies to poly(adenosine diphosphate-ribose) in systemic lupus erythematosus, Clin. Exp. Immunol., 36, 151-164 Palmer, R G (1984) Sister chromatid exchange assessment of chromosome damage induced by cytotoxac drugs in patients with connective tissue diseases - - are they at special risk'~ Oxford Umversaty, DM thesis, 273 pp Palmer, R G , C J Dore and A M Denman (1984) ChlorambucIl-induced chromosome damage to human lymphocytes is dose dependent and cumulative, Lancet, l, 246-249 Palmer, RG., C J Dore and A.M. Denman (1986) Sister chromatid exchange frequencies in lymphocytes of controls and pauents with connective tissue diseases, Mutation Res. 162, 113-120 Parkes, D.J G , D Scott and A Stewart (1985) Changes in spontaneous SCE frequencies on a function of sampling time in lymphocytes from normal donors and cancer patients, Mutation Res, 147, 113-122 Perry, P, and S Wolff (1974) New glemsa method for the differential staimng of sister chromatids, Nature (London), 251,156 158 Ra1, A., and J A Heddle (1980) Simultaneous detection of chromosomal aberrations and sister-chromatid exchanges, Mutation Res., 78, 253-260. Royston, J P. (1982) An extension of Shapiro and Wllk's 14 test for normality to large samples, J. Roy Statist Sot., C31, 115-124 Schweder, T. (1981) A simple test for a set of sums of squares, J. Roy Statist Soc., C30, 16-21 Tan, E.M (1976) Sunlight as a potential aetlological factor in systemic lupus erythematosus, in: G R V Hughes (Ed.), Modern Topics in Rheumatology, Heinemann, London, pp 99-106 Tan, E.M, A,S Cohen, J.F Fries, A T. Masl, D J. McShane. N F RothfIeld, J G Schaller, N Talal and R J Winchester (1982) The 1982 revised criteria for the classification of systemic lupus erythematosus, Arthritis Rheum, 25, 1271-1277 Tuschl, H , R. Kovac, A Wolf and J.S Smolen (1984) SCE frequencies in lymphocytes of systemic lupus erythematosus patients, Mutation Res., 128, 167-171 Zamansky, G.B, D F Minka, C L. Deal and K. Hendnckb (1985) The m vitro photosensitivity of systemic lupus erythematosus skin flbroblasts, J Immunol, 134, 1571-1576