Increased susceptibility to in vitro ultraviolet B radiation in fibroblasts and lymphocytes cultured from systemic lupus erythematosus patients

CLINICAL

IMMUNOLOGY

AND

IMMUNOPATHOLOGY

s&289-304

(1991)

Increased Susceptibility to in Vitro Ultraviolet B Radiation in Fibroblasts and Lymphocytes Cultured from Systemic Lupus Erythematosus Patients’** THEO Clinical

Dov GOLAN,

Immunology Service, of Medicine,

VERONICA

FOLTYN,

AND ANNE

ROUE.FF

B’nai Zion Medical Center and Department of Immunology, Technion-Israel Institute of Technology, Haifa, Israel

Faculty

Sunlight is known to induce exacerbations of systemic lupus erythematosus (SLE) but its mechanism remains unclear. We have previously reported that ultraviolet A (UVA) exposure induces an increase in total DNA synthesis (DS) in vitro but a decrease in unscheduled DNA repair synthesis (UDRS) of splenocytes of murine SLE strains. In order to investigate whether similar observations are characteristic of human SLE, peripheral blood lymphocytes (PBL) and dermal fibroblast (DF) cultures of 20 patients and 15 matched controls were exposed in vitro to UVA or UVB at different doses. Thirteen (65%) SLE DF cultures exposed to UVB light (12-24 J/m’) showed an increase in DS compared to paired unirradiated cultures. In contrast, UVB-irradiated DF from normal individuals had no significant increase in DS following UVB irradiation. When SLE DF were exposed to higher doses of UVB (48-96 J/m*), 90% of cultures showed a decrease in DS compared to only 20% in the control group. All of the SLE DF cultures showed a decrease of their unscheduled DNA repair capacity following UVB (2a8 J/m*) irradiation whereas no UDRS was apparent in 74% of controls under the same conditions. Similar findings regarding UDRS were observed in SLE PBL cultures and were also confirmed by autoradiography. UVA exposure (O-3840 J/m*) had no effect on IX nor on UDRS in DF or PBL cultured from SLE and controls. The relevance of these in vitro findings to the in vivo pathogenesis of the disease is discussed. Q IVVI Academic press, IIK.

INTRODUCTION

The detrimental effect of sunlight on the activity of systemic lupus erythematosus (SLE)3 is well established. Skin manifestations of SLE are found predominantly on sun-exposed areas. They are generally pleomorphic, ranging from persistent erythema to maculous or bullous eruptions. The butterfly-like erythematous lesion over the malar region and bridge of the nose has been considered a hallmark of the disease, although it does not occur in all patients (1). Exposure to sunlight can cause new skin eruptions, exacerbate existing ones, cause progres’ This investigation was supported in part by the Chief Scientist, Ministry of Health, Israel. * Part of this work was presented in abstract form at the Seventh International Congress of Immunology, July 3@-August 5, 1989, Berlin. 3 Abbreviations used: SLE, systemic lupus erythematosus; UVL, ultraviolet light (2OOXtO nm radiation); UVA, ultraviolet A (32O-tOO nm radiation); UVB, ultraviolet B (290-320 nm radiation); UVC, ultraviolet C (200-290 nm radiation); PBL, peripheral blood lymphocytes; DF, derrnal fibroblasts; PBS, phosphate-buffered saline; DS, DNA synthesis; UDRS, UV-induced unscheduled DNA excision repair synthesis; [3H]TdR, tritiated thymidine; HU, hydroxyurea; TCA, trichloroacetic acid; NSAID, nonsteroidal anti-inflammatory drugs; ANA, anti-nuclear antibody; dsDNA, double-stranded DNA. 289 0090-1229/91 $1.50 Copyright All rights

0 1991 by Academic Press, Inc. of reproduction in any form reserved.

290

GOLAN,

FOLTYN,

AND

ROUEFF

sion to unexposed areas, or, of more importance, induce systemic disease activity (I, 2). Cutaneous responses in sun-sensitive patients have also been elicited by exposure to ultraviolet light (UVL) under controlled laboratory conditions (3). However, despite much documented clinical observations, the mechanism by which UVL affects the pathogenic course of the disease remains poorly understood (4). UVL (20&400 nm) is known to cause not only increased chromosomal aberrations in both human and murine cells, but also increased excision repair synthesis of DNA in mammalian cells and enhanced neoplastic transformation (5). Because DNA is considered to be the critical target for those changes, radiation in the DNA-sensitive range of UVC (254-280 nm) is commonly used. However, it became apparent that UVC-induced DNA damage is biologically irrelevant since only UVL with wavelengths longer than 290 nm penetrate the ozone layer of the stratosphere. Thus, UVL that reaches the earth surface is mainly composed of UVA (320-400 nm) (9&95%) and UVB (29Q-320 nm) (6). Recently, it has become clear that UV-induced DNA damage varies according to the nature of the exposure. Whereas UVC-induced DNA damage is characterized by a high frequency of thymine dimer formation, UVB induces pyrimidine dimers less frequently but also causes DNA-protein cross-links (7, 8). UVA, on the other hand, induces the appearance of single-strand breaks in DNA which may be of lethal significance. Other derangements in cell metabolism influencing both RNA and protein synthesis, enzyme activation and cell membrane permeability, may also be induced by UVB and UVA radiation (8). Although early studies showed that in Gtro exposure to UVC alters the antigenicity of DNA, UV-DNA cannot induce full-blown SLE in viva (9, 10). Choosing a more biologically relevant UV source, we have recently demonstrated increased in vitro UVA sensitivity in splenocytes of murine SLE characterized by increased DNA synthesis coexisting with diminished DNA repair capacity (6). Furukawa et al. (11) described an increased in vitro cytotoxic effect of UVB light on tibroblasts and keratinocytes cultured from SLE-prone MRL/Mp-lpr/fpr mice. Other investigators have described accelerated autoimmunity and increased mortality of BXSB male mice following UVB irradiation in viva (12) and decreased survival of human SLE skin tibroblasts following in vitro UVB irradiation (13). In this report, we show that human SLE peripheral blood lymphocytes (PBL) and dermal fibroblasts (DF) demonstrate an enhanced sensitivity to UVB light as evidenced by their lowered threshold for induction of DNA synthesis (DS) and unscheduled DNA excision repair synthesis (UDRS) as well as their lowered capacity of DNA repair. MATERIALS

AND

METHODS

Patients and controls. Twenty SLE patients in remission or minimal clinical activity (18 females and 2 males, their age range 16-42 years) who fulfilled four or more of the American Rheumatism Association revised criteria for the classification of SLE (14) were included in this study, after giving informed consent. Fifteen healthy individuals (12 females and 3 males, their age range 16-32 years) who gave a similar informed consent were included as controls. Sera from all patients

IN

VITRO-INCREASED

UVB

SUSCEPTIBILITY

IN

SLE

291

and controls were obtained and tested for the presence of autoantibodies as detailed in Results. UV radiation. Prior to UV irradiation, cell culture media were replaced with PBS free of any photoreactive compounds to avoid toxic photochemical effect of UVL on cultures. PBL cell suspensions were distributed into six 30-mm-sterile plastic tissue culture dishes (sterilin No. 3OlV, Teddington, Middlesex, UK) evenly plated as monolayer while maintaining a liquid depth of 1-2 mm. At least one dish of each cell strain was not exposed to UVL, whereas the remainder were exposed separately to various doses of UVA or UVB as detailed in Results. A Westinghouse Black Light F72T12/BL-S/SHO-0 lamp (Lamp Division, Framingham, MA) housed in a reflector unit was used as a UVA (32wOO mm) light source. The irradiance of this lamp (peaking at 365 nm), as measured by an International Light IL 1350 Research Radiometer (Newsbury Port, MA), was 3.25 mW/cm’ at a distance of 12 cm and contained less than 2% UVB irradiance. A Westinghouse Sun Lamp FS72T12 housed in a reflector unit was used as a UVB (280-320 nm) light source. The irradiance of this lamp (peaking at 313 nm), as measured by an IL 1350 radiometer, was 0.16 mW/cm2 at a distance of 23 cm and contained less than 1% of UVC in its output and 10% of UVA it-radiance as well. DNA synthesis ofPBL cultures. PHA-stimulated PBL cultures of SLE patients and paired controls were prepared as described elsewhere (15) except that the PBLs were PHA stimulated 48 hr prior to use. Each culture (in triplicate) contained 5 x lo5 cells/ml of tissue culture medium (RPM1 1640, with L-glutamine, Biological Industries, Beit Haemek, Israel) supplemented with 15% human AB serum (Central Blood Bank, Israel), 100 U/ml penicillin, and 100 l&ml streptomycin sulfate. After irradiation, PBS was immediately replaced with fresh medium containing 1 $X/culture of [3H]thymidine ([3H]TdR, Nuclear Research Center, Negev, Israel) for an additional 24 hr, and the TCA-precipitable [3H]TdRlabeled material of their cell pellets was measured, representing, total newly synthesized DNA. Unscheduled DNA repair synthesis of PBL cultures. PBL cell suspensions (without PHA) were prepared as above. Following in vitro UVL irradiation, paired cultures (5 x lo5 cell/culture) of one SLE patient and one control were incubated with fresh culture medium containing I &i [3H]TdR and hydroxyurea (HU: H-8627, anhydrous, Sigma Chemical Co., St. Louis, MO) at a final concentration of 5 mM. In parallel, UV-unexposed cultures were incubated in a similar manner. Cultures were harvested after 24 hr and [3H]TdR incorporation into DNA was determined as described above. UDRS was expressed as HU-resistant [3H]TdR incorporation into TCA-precipitable material. DNA synthesis of DF cultures. Superficial dermal tissue [0.2 mm (thick) X 2 mm x 3 mm] was obtained from the medial forearm of each SLE patient and control and was cut into smaller pieces that were attached to the bottom side of vertical standing plastic tissue culture flasks (Sterilin, 75 cm2, No. 312, Teddington, Middlesex, UK) containing 5 ml of culture medium Dulbecco’s modified Eagle’s medium supplemented with 15% heat-inactivated fetal calf serum, 4.5 t~,g/ml D-glucose, 4 mM/ml L-glutamine, 1 mM/ml sodium pyruvate, 100 U/ml penicillin, and 100 pg/ml streptomycin sulfate (Biological Industries). After a

292

GOLAN,

FOLTYN,

AND

ROUEFF

45min incubation, flasks were carefully lowered to their horizontal position, while tissue sections remained attached to the bottom surface. After 10-14 days incubation at 37°C in a humid mixture of 95% air and 5% COZ, primary cultures of fibroblasts were detected and the culture medium was replaced twice weekly. After an additional 2-3 weeks, a sufftcient number of cells was present to begin serial passaging. All cell strains in this study were used between passage 4 and 9. Monolayers of DF cultures containing 1 X lo5 DF/2 ml medium culture/dish were prepared 24 hr prior to UV exposure from established cell strains. Each experiment was performed (in triplicate) with paired cell strains (one SLE and one control) cultured separately. Culture medium was discarded and PBS was added before cells were exposed to UV irradiation, followed by an immediate replacement by fresh culture medium containing 1 pCi r3H]TdR/dish and the incubation was continued for an additional 24 hr. Thereafter, cultures were harvested and the [3H]TdR incorporation into DNA was determined as described above. Unscheduled DNA repair synthesis of DF cultares. DF cell cultures were prepared as above. Twenty-four hours following in vitro UVL irradiation, paired cultures (2 x IO5 cells/culture) of one established SLE cell strain and one control were assessed for their HU-resistant [3H]TdR incorporation. Autoradiography of PBL cultures. Techniques similar to those reported by other laboratories were used (13, 16, 17). Cultures (5 x IO5 cells/culture) were incubated for 60 min in medium containing 10 t&i/ml of [3H]TdR to label-resting cells. In parallel, other cultures were UV irradiated and immediately labeled as above for 60 min. All cultures were followed by a 2-hr chase in medium supplemented with lop5 M thymidine. Autoradiographs were then prepared directly on fixed cultures by using Kodak NTB-2 nuclear track emulsion. Autoradiographs of UV-exposed cultures were developed after 3 days, whereas those of UVunexposed cultures were developed after 7 days. All autoradiographs were stained with 5% Giemsa stain. The number of grains over labeled nuclei were counted and tabulated in subgroups of increasing frequency. Cell count, viability test, and statistical analysis. The total number of viable cells, calculated according to both cell count in hemocytometer and viability test (by trypan blue exclusion), was expressed as percentage of original viable cells at “0” hr. All PBL and DF from SLE and controls had a viability of 98-100% at the onset of experiments. No difference was observed in viability between SLE cells and controls at termination of the cultures. Individual experiments were normalized to the mean of replicate experiments. Differences between means were statistically analyzed by using a two-tailed Student’s t test. RESULTS History of clinical manifestations, apy. Patients’ age, gender, duration

autoantibody

projile,

and maintenance

ther-

of disease, organ involvement. autoantibody profile, and treatment are summarized in Table 1. A history of photosensitivity to sunlight was present in 12 patients (60%), of malar rash and/or discoid LE in 14 patients (70%), and symptoms of subacute cutaneous LE in 1 patient (5%) (other organ involvement is detailed in Table 1). The serological profile of SLE patients revealed the following: ANA were present in sera of all patients (lOO%), anti-

IN

VITRO-INCREASED

UVB

SUSCEPTIBILITY

TABLE

IN

293

SLE

1

ORGAN INVOLVEMENT,AUTANTIBODYPROFILE,ANDMAINTENANCETHERAPYOF Patientsb 12 1 2 3 4 5

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Organ involvementC 3

12

F F F

16 18 19

2 1 2

+

F F F

22 22 25

3 4 3

F M F F F F F F F F M

28 28 30 31 31 34 35 36 31 38 40

6 4 6 10 9 15 8 20 12 3 15

+

f + t + t + +

40

8

+

40

23

+

8

+ + + + t + + + + t + t

4

5

6

+

+ +

+

+

+

t

F

42

+

+

F F

3

Antibody

+ + t t + + +

+

t

+

t

f t + t

t t t

t

+

+ t t

+ + t t + t + t t t

8

+

+

t

7

+

t t

t

+

+

i t + t + + + t + t t t + + t t t + t + +

2

+ + + t +

+ + t + t

3

SLEPATIENTSO profiIed 4

5

t +

+

t t t +

t

6

+

Therapy’ NSAID NSAID NSAID NSAID

WO) NSAID NSAID P(15)

-

t + +

+ t

t

t

t

t

+ t t t

+

P(30) P(15) NSAID

wo) PC33 NSAID

Pm

NSAID

a At entry to study. b 1, gender (F/M); 2, age; 3. duration of disease (years). c 1, photosensitivity; 2, skin lesions; 3, oral ulcerations; 4, joints; 5, serous membranes; 6, kidney; 7, central nervous system; 8, blood. d 1, ANA detected by indirect immunofluorescence (IIF) on mouse liver sections; 2, dsDNA detected by IIF on Crifhidia fuciiiae: 3, Sm; 4, nRNP; 5, SSA: 6, SSB fall detected by counter immunoelectrophoresis). p NSAID, nonsteroidal anti-inflammatory drugs; P, prednisolone (mg/daily).

ds-DNA were present in 10 sera (50%), anti-Sm in 9 seta (45%), anti-nRNP in 5 sera (25%), anti-SSA in 6 sera (30%), and anti-SSB in 3 sera (15%). No abnormal serological findings were detected in controls (data not shown). Four patients (20%) received no maintenance therapy at time of study, 9 patients (45%) received nonsteroidal anti-inflammatory drugs (NSAID), and 7 patients (35%) were on 15-30 mg prednisolone daily. None of these patients was receiving immunosuppressive therapy. DNA synthesis in PBL and DF cultures following in vitro UVA exposure. Several paired experiments were conducted in a total of 12 SLE patients and 10 controls, in which 13H]TdR incorporation into acid-precipitable-labeled DNA was measured 24 hr after different doses of in vitro UVA irradiation (O-3840 J/m2). The data (not shown) demonstrate that the effect of UVA exposure varied among SLE PBL cultures as well as in controls. However, no statistically significant effect of UVA irradiation on DS was demonstrable in PBL cultures of SLE when compared to those of controls. Since it was of interest to investigate whether this effect of in vitro UVA irradiation on DNA synthesis of PBL of SLE and control cultures was solely confined to PBL cells, DF cultures of 12 SLE patients paired with 12 controls were exposed in vitro to the same dose of UVA irradiation. Neither SLE DF cultures nor controls showed any statistically significant change in DNA synthesis under such conditions (data not shown). DNA synthesis in DF cultures following in vitro UVB irradiation. Since it was

294

GOLAN,

FOLTYN,

AND

ROUEFF

of interest to investigate whether this equivocal effect of UVA irradiation on DNA synthesis of PBL and DF of SLE and control cultures in vitro was solely confined to the UVA zone, DF cultures of 20 SLE patients and 15 paired controls were exposed to various doses of UVB (0-% J/m2) and thymidine incorporation was measured 24 hr after irradiation. Three representative experiments of UVB irradiation are shown in Fig. 1, in which different patterns of effect on DS were demonstrated. In two SLE UV-unexposed cultures, DS was significantly higher in comparison to that of UV-unexposed controls (Figs. 1B and 1C). A statistically significant increase in DS was observed in all three SLE DF cultures following 6-24 J/m2 UVB irradiation (in comparison to their UVB-unexposed cultures). One SLE cell strain demonstrated increased DS even after exposure to 48-96 J/m’ (Fig. lA), whereas in the others, DS decreased below the level of their own nonexposed cultures (Figs. 1B and IC). In comparison, control DF cultures demonstrated either a statistically insignificant increased DS (Figs. 1A and IC) or no change following 6-24 J/m2 UVB exposure (Fig. 18). The cumulative results of DNA synthesis of DF cultures following UVB exposure are shown in Table 2. Twelve (60%) of the nonirradiated SLE DF strains showed a “spontaneous” increased DNA synthesis, 6 (30%) showed no significant difference, and 2 (10%) showed decreased DNA synthesis when compared to unexposed cultures of their concomitant controls. Thirteen SLE (65%) DF cultures exposed to 6-24 J/m’ UVB demonstrated increased DNA synthesis (in comparison to their own UVB nonexposed cultures), in contrast to no increase in any of 15 controls. At higher doses of UVB (48-96 J/m2), 18 (90%) SLE cultures demonstrated decreased DNA synthesis (when compared to their own UVB-unexposed cultures), contrasting with only 3 (20%) of controls. UV-induced unscheduled DNA repair synthesis in DF‘ cultures. [‘H]TdR incorporation into newly synthesized and acid-precipitable DNA is likely to be the result of two semi-independent processes: premitotic, semi-conservative S-phase DNA synthesis and unscheduled repair synthesis. To discriminate between the two, paired experiments (SLE and control) of UVB or UVA-induced UDRS were performed. In 12 out of 18 paired experiments (performed in 18 SLE cell strains and 15 controls), increased susceptibility of UDRS in SLE cells was detected when compared to controls. An example of such an experiment is illustrated in Fig. 2: a fourfold increase in HU-resistant [‘H]TdR incorporation into DNA is evident in unexposed SLE cells when compared to unexposed cells of the control. Moreover, a further increase of UDRS was detected in SLE cells following a 12 J/m2 UVB exposure (in comparison to their own UVB unexposed cells), whereas control cells remain unaffected under the same conditions. However, while SLE cells showed a decline in UDRS following exposure to higher doses of UVB radiation (24-48 J/m2), control cells demonstrated an increase in UDRS (in comparison to their own UVB-unexposed cells) only after 48 J/m* UVB exposure. In the remaining 6 SLE cell strains, no UVB-induced UDRS was demonstrated under the given conditions. The cumulative results of UDRS in DF cultures of SLE patients and controls following in vitro UVB or UVA exposure are summarized in Table 3. It is of interest to note that all 18 SLE cell strains demonstrated a decrease in UDRS

IN

VITRO-INCREASED

7.6

SUSCEPTIBILITY

295

IN SLE

A

r

0

UVB

6

12

1624

40

B

7.6 7.0 6.4

/4

5.0 5.2 4.6 4.0 3.4

01, 0

6

’ B - ‘*

121924

“I 40 J/m2

E a m’ ”

o

96

(UVB)

FIG. 1. DNA synthesis in SLE and control-paired DF cultures 24 hr after UVB exposure. Total newly synthesized DNA is expressed as cpm of [3H]TdR-labeled and TCA-precipitable material present in the cell pellet of UVB-exposed (or unexposed) cultures (1 X Id DF cells/culture) containing 1 $3 [‘H]TdR added immediately after UV irradiation. Three paired experiments (A, B, C) are representative of many paired experiments performed in 20 SLE cell strains and 15 controls. SLE cell strain, 0; control cell strain, 0. Results are expressed as geometric means k 1 SD and differences between results of SLE and those of controls are statistically significant in each paired experiment at all points of reference (P < 0.005), except those of 0, 12 and 18 J/m2 (lA), 48 and % J/m’ (1B).

296

GOLAN,

FOLTYN,

AND ROUEFF

TABLE 2 DF CULTURESOF SLE PATIENTSANDCONIROLS in Vitro UVB IRRADIATION"

DNA SYNTHESISIN

CUMULATIVERESULTSOF

FOLLOWING

DNA synthesis Cultures

UV dose (Jim?

Nonirradiated

0

UVB-irradiated

6-24 48-96

SLE

P

Increase 12/20 No effect 6/20 Decrease 2120

~0.005” >O.Ih c0.005h

Increase No effect No effect Decrease

iO.005’ >o. 1’ >o. 1’ 10.005’

13120 7120 2/20 IS/20

Controls

No effect 15115 No effect 12/15 Decrease 3/15

P

X). 1’ >O.l’ CO.005’

u Twenty SLE cell strains and 15 cell strains of controls were exposed (one of each in paired experiments) to UVB. Means of DS were measured as [‘H]TdR incorporation in triplicate cultures (I x 10’ cellskulturet 24 hr after UV exposure. ’ DS of nonirradiated SLE cultures were compared to those of nonirradiated paired controls, ’ Cumulative data of DS (designated as increase/decrease/no effect) were based upon normalized and statistically analyzed results of paired experiments (SLEkontrol) when UV-exposed cultures, at various points of reference. were compared to the appropriate paired unexposed ones.

activity, following 24-48 J/m’ UVB irradiation (when compared to their own unexposed cultures), whereas only 4 out of 15 control cell strains showed similar results. Furthermore, none of the controls showed changes in UDRS (when compared to their own UVB-unexposed cultures) when exposed to 6-12 J/m2 UVB irradiation. UVB-induced UDRS in PBL cultures. Paired experiments were performed in PBL from five SLE patients and five controls exposed to UVB in vitro. A representative experiment is illustrated in Fig. 3. DNA repair by SLE PBL following UVB-induced damage shows a similar pattern to that observed in SLE DF, viz the magnitude of UDRS is higher than that of controls, it reaches its peak at a lower UVB dose (6-9 J/m’) than the control, and its activity declines at doses of UVB irradiation (12-24 J/m’) at which UDRS of control is still on the rise. Autoradiography of UV-induced DNA repair synthesis in PBL. To confirm that differences in UDRS activity existed between SLE and normal controls, autoradiography of PBL cultures was performed. The data of a representative experiment is illustrated in Fig. 4. The results clearly demonstrate the existence of increased UDRS activity in unexposed SLE cells (Figs. 4B and 4C) when compared to control (Fig. 4A). Moreover, SLE cells demonstrated a restricted repair capacity in response to increasing UVB irradiation (6-9 J/m*). Control PBL showed no sign of exhausted UDRS under the same experimental conditions. These results substantiate those obtained in UDRS experiments of DF and PBL cultures described above. Similar results were seen in all four pairs of cells examined. Correlation between UVB susceptibility in vitro and clinicalfindings. The clinical manifestations of the SLE patients in this study are comparable to others (14),

IN VITRO-INCREASED

UVB SUSCEPTIBILITY

IN SLE

297

J/m2 UVB FIG. 2. Unscheduled DNA repair synthesis (UDRS) in SLE and control-paired DF cultures 24 hr after UVB exposure. UDRS is expressed as cpm of HU-resistant [3H]TdR incorporation and TCAprecipitable material present in the cell pellet of UVB-exposed (or unexposed) cultures (2 x 10’ cells/culture) containing 1 +Ci [3H]TdR and HU, added immediately after UV exposure. The paired experiment is a representative of those performed in 13 (out of 18 SLE cell strains tested) and 15 controls. SLE strain, 0; control, 0. Results are expressed as geometric means t 1 SD and differences between results of SLE and those of control are statistically significant at all points of reference (P < 0.005) except that of 48 J/m2 UVB exposure.

although it was a volunteer-based group. The increased in vitro UVB susceptibility was detected in 13 patients (Table 1: Nos. 1-6, 9, 14-17, 19, 20) and could not be directly linked to any of the clinical features observed since it seemed not to differ percentagewise from the total series (Table 4). For example, out of these 13 patients, 7 (54%) had a clinical history of sunlight sensitivity, 8 (62%) had skin manifestations, and only 4 (31%) had both, findings that are similar to the total group. Since only one patient in the UVB-susceptible group demonstrated clinical manifestations of subacute cutaneous LE (the only one in the series), the susceptibility of such patients’ cells to in vitro UVB irradiation is yet to be determined. Although a smaller percentage of patients in the UVB-sensitive group tested positive to most of the autoantibodies in comparison to the total series (Table 4), this observation is not significant because of the limited size of the group. In order to remove any influence of the patients’ medications, all DF cultures used were between passage 4 to 9. Our data clearly show that the increased UVB sensitivity was not related to the mode of treatment (Table 4) since patients belonging to the UVB-sensitive group received either no medication or different drug regimens. Moreover, in 5 individuals studied, a diminished UVB-induced UDRS was observed in their freshly collected PBL cultures, similar to that of their passaged DF cells (data not shown), thus ruling out any direct effect of in vivo medication on experimental outcome. However, since entry criteria precluded patients with se-

GOLAN,

298

FOLTYN, TABLE

AND ROUEFF 3

CUMULATIVERESULTSOF UNSCHEDULED DNA REPAIR SYNTHESIS IN DF CULTURES PATIENTS AND CONTROLS FOLLOWING in Vitro UVBIUVA IRRADIATIONS

OF

SLE

Unscheduled DNA repair synthesis Cultures UVB-irradiated

UV dose (J/m”) G-12 24-48

UVA-irradiated

60-3840

SLE Increase No effect Decrease No effect

P

1208 6/18 18/l 8

O.OOSb >o. Ih
919

>o. lb

Control No effect Decrease No effect No effect

P

IS/IS 4/15 1l/l5 919

T-0.1” O.Ih >o. Ih

a Eighteen SLE cell strains and I5 cell strains of controls were exposed (one of each in paired experiments) to UVB; 9 SLE cell strains and 9 cell strains of controls were exposed (one of each in paired experiments) to UVA. Means of UDRS were measured in triplicate cultures (2 x 10s cells/ culture) 24 hr after UV exposure. b Cumulative data of UDRS (designated as increase/decrease/no effect) were based upon normalized and statistically analyzed results of paired experiments (SLEkontrol) when UV-exposed cultures, at various points of reference, were compared to appropriate paired unexposed ones.

J/m2 UVB FIG. 3. UVB-induced DNA repair synthesis in SLE and control PBL in vitro, 24 hr after irradiation. UDRS is assessed as cpm of HU-resistant [3H]TdR incorporation and TCA-precipitable material present in the cell pellet of UVB-exposed (6-24 J/m’) or unexposed cultures (5 x 16 cells/culture) containing I pCi r3HJTdR and HU. added immediately after UV exposure. The paired experiment is a representative of several performed experiments in five SLE patients and five controls. SLE strain. 0; control strain, 0. Results are expressed as geometric means 2 1 SD and differences between results of SLE and those of control are statistically significant at all points of reference (P < 0.005) except that of 12 J/m2 UVB exposure.

IN

VITRO-INCREASED

UVB

SUSCEPTIBILITY

IN

SLE

299

A

1 no UV ,

70

$j

50

'0 2 2

30

zi .2! 5

10

=

70

6

J/m2

9 J/m2

0

50

30

10

Grains/Nucleus FIG. 4. Histogram of autoradiographs of SLE and control PBL. Cell suspensions of SLE and control PBL were evenly plated as monolayers into several 30-mm-sterile plastic tissue culture dishes (5 x ld cells/dish) and were unexposed or exposed (in triplicate) to various doses of UVB. All cultures were incubated for 60 min in medium containing 10 uCi [3H]TdR and thereafter followed by a 2-hr chase in medium supplemented with 10-s M unlabeled thymidine. Autoradiographs were then prepared as described under Materials and Methods. The average number of grains over 100 nuclei of each replicate were tabulated according to dose of exposure and subdivision of grain frequency. The representative of a number of such autoradiographs is illustrated in Fig. 4 (A, B, C). Control (open bars), A; two SLE patients (hatched bars), B, C.

vere disease activity, the effect of disease activity and that of immunosuppressive therapy on UV-induced DS and UDRS in SLE remains unknown. DISCUSSION

To investigate the role of UVL in the pathogenesis of SLE, we examined the ability of SLE patients’ cell strains (PBL, DF) to synthesize DNA after UV irradiation. These experiments were performed in the absence or presence of HU, a drug known to suppress semiconservative DNA synthesis of replication (18). We examined whether UVB produces an increased stimulatory effect on DS and whether this coexists with a decreased UDRS capacity as we previously observed

300

GOLAN,

FOLTYN,

AND

TABLE PREVALENCE

Patient group UVB susceptible (?I = 13) Total series (n = 20)

(IN PERCENTAGE) OF ORGAN TREATMENT IN UVB-SUSCEPTIBLE

ROUEFF

4

INVOLVEMENT,

PATIENTS

Organ involvementb

AUTOANTIBODY

AND IN TOTAL Antibody

PROFILE,

AND

SERIES~

profile’

Therapy”

1

2

3

4

5

6

7

8

1

2

3

4

5

6

1

2

3

54

62

15

92

46

31

23

23

100

46

31

15

23

8

23

46

31

60

70

15

90

40

40

20

35

100

50

45

25

30

15

20

45

35

” At entry to study. ’ By clinical history: 1, photosensitivity; 2. skin lesions; 3, oral ulcerations; 4, kidney; 7, central nervous system; 8, blood. c 1, ANA; 2. dsDNA; 3, Sm; 4, nRNP; 5, SSA; 6, SSB. ’ 1. none: 2, nonsteroidal anti-inflammatory dmgs; 3, prednisolone (IS-30 &.iaily).

joints;

5, scrous membranes;

6,

in murine SLE (6). Such findings may have a bearing on the enhanced formation of pathogenic immune complexes containing DNA (6). Since only UVL with wavelengths longer than 290 mm (namely UVB and UVA) are biologically relevant, cell strains of SLE patients and controls were exposed to UV sources in each zone over a varying range of low-dose irradiation. These doses resembled those of natural in viva UV exposure that may be encountered in the late afternoon on a sunny day at sea level (equal to l-2 MED). Our study is the first to show, in a biologically meaningful setting, that in the majority of SLE patients studied, an increased susceptibility of PBL and DF cells to UVB-induced DS exists together with a diminished DNA repair capacity (Tables 2 and 3). In contrast, SLE and control cell cultures were unaffected by in vitro UVA exposure under given experimental conditions. Controversial observations have been reported with regard to UVL susceptibility in human and murine SLE cells, using various methods (19-25). With SLE lymphocytes, Beighlie and Teplitz (19) demonstrated defective UVL-induced DNA repair, while Atlman el al. (20) reported UV-induced DNA repair in SLE fibroblasts to be somewhat enhanced. Furthermore, Cleaver (21) reported UVinduced DNA repair in SLE fibroblasts to be normal (in one cell strain that was investigated), whereas others (22, 23) reported defective repair of DNA in lymphocytes obtained from patients with SLE. NZB mouse fibroblasts demonstrated diminished DNA repair synthesis and increased chromosomal alterations after UV irradiation in v&-o (24,215). However, all of these observations were made in UVC-exposed cells and may therefore have limited value in the clinical pathogenesis of SLE. On the other hand, our present data reflect the sensitivity of SLE cell strains to UVB exposure, therefore being of greater potential relevance to the disease. However, it is important to note that although Zamansky et al. (13) observed decreased survival of SLE DF cell cultures after in vitro exposure to UVB, they failed to demonstrate a diminished UDRS capacity in one cell strain that was investigated. This discrepancy is most likely explained by differences in experimental methodology.

IN

VITRO-INCREASED

UVB

SUSCEPTIBILITY

IN SLE

301

Our present data indicate that the majority of PBL and DF derived from SLE patients are affected by UVB irradiation, namely, they demonstrate varying degrees of increase in thymidine incorporation (5&300%) at a low dose or a decrease in thymidine incorporation (50-180%) with a high dose of UVB exposure. This differential effect is limited to UVB exposure only because both normal and SLE cells showed no significant differences when exposed to UVA under given conditions. Our present report demonstrates that increased UVB-induced damage is not solely confined to lymphoid cells but is also present in DF cells. This suggests that this defective repair mechanism may exist in other cells as well. In support of this notion are the observations made recently by Furukawa et al. (11) who demonstrated increased in vitro UVB-induced cytotoxicity both in keratinocytes and in tibroblasts of MRL/Mp-fpr/lpr mice. These findings are different from our own observations in murine SLE in which increased UVL susceptibility was mainly confined to the UVA zone while variable UVB sensitivity was also noted among lupus mouse strains (6). In this regard, it is of interest to note the recent observation of prolonged survival of (NZB/W) Fr hybrids following low-dose UVA exposure in vivo (26). These species and strain differences in response to UVL are so far unexplained. Since thymidine incorporation may reflect semiconservative DS due to activation of some enzymatic systems (8) or UDRS which reflects DNA repair or both, studies were performed in the absence or presence of HU. Comparative analysis of HU-resistant thymidine incorporation into DNA of UVB-irradiated DF cell cultures in the absence or presence of HU (Figs. 1 and 2) indicates the relative degree of UDRS relative to total DS. With UVB irradiation of 6-24 J/m2 in the absence of HU, UDRS contributes up to 25% of the totaJ increase in DS. Decreased (3H)TdR incorporation into DNA observed at higher doses of UVB irradiation (48-96 J/m’) may reflect the progressive arrest of normal cell activity due to enzymatic inhibition (8). Furthermore, the data also show a restricted repair capacity of SLE cells since those cells exposed to higher doses of UVB irradiation demonstrate a decline in UDRS. In contrast, control cells did not show a decline in DNA repair capacity since their UDRS increased under similar conditions. Increased UVB-induced DNA damage coexisting with restricted repair capacity may have a detrimental effect on overall survival. Supporting this notion is the recent observation of Ansel et al. (12), who reported increased mortality and accelerated autoimmunity in BXSB male mice following in vivo UVB exposure, in comparison to other murine SLE strains and control ones. Their finding may be related to our former in vitro observation, namely that BXSB male mice demonstrated the most pronounced restriction in repair capacity to UVL-induced DNA damage (6). In 12 out of 20 SLE patients, a spontaneous increase in the synthesis of DNA in the unsynchronized UV-unexposed DF cell cultures was demonstrated, whereas this was not detected in PBL cultures (known to be synchronized at their G,/Gi cell cycle) of the same SLE patients (exemplified in Fig. 3). These findings in DF cell cultures are similar to our earlier report of a spontaneous increase of

302

GOLAN.

FOL’I’YN.

AND

ROUEFb

DNA turnover in unsynchronized, UV-unexposed spleen ceil cultures of murine SLE (27). That this increase is related to increased UV susceptibility to ambient UVL is supported both by our autoradiographic observation of increased nuclear staining of resting unirradiated PBL of SLE patients and by HU-resistant UDRS in SLE-unirradiated DF cultures. A linkage may also exist between our previous (6) and current findings of restricted DNA repair capacity residing in murine and human SLE-derived cells and the observations of others. of the elevated frequency of spontaneously occurring chromosomal breaks and rearrangements in both murine and human SLE-derived cells (28). The possibility that restricted DNA repair capacity in SLE might be related to DNA immunogenicity was raised by several investigators, but still remains unclear (4, 23, 29). Moreover, UVL irradiation not only affects DNA metabolism, but it is also known to influence RNA and protein synthesis, enzyme activation, and cell membrane permeability (8). In this regard, the recent in vitro and in vivo detection of nuclear antigens on the cell membrane of keratinocytes derived from healthy neonates after UVB irradiation is of special interest (30. 3 1). Furthermore. the appearance of these nuclear antigens (SSA, Sm, nRNP) was detected on cells with a mild increase in cell membrane permeability that did not coincide with cell death (30). The lack of direct correlation between increased in vitro UVB susceptibility and defined clinical symptoms or specific serological abnormalities in this study (Table 4) may not be surprising since its outcome is likely to be dependent on complex interactions of multiple factors in a given patient. These factors include complement levels, sex steroids, the immunological repertoire, various environmental stimuli (12, 32, 33), the degree of diminished DNA repair capacity, and its effect on other cellular functions yet to be determined. Whether our observations are related to the pathogenesis of SLE and especially to the autoimmune response to nuclear antigens characteristic of SLE remains to be determined. REFERENCES 1. Dubois, E. L.. Management of discoid and systemic lupus erythematosus. In “Lupus Erythematosus” (E. L. Dubois, Ed.), p. 555. University of Southern California Press. Los Angeles, 1974. 2. Epstein, J. H., Tufanelli, D. L., and Dubois. E. L., Light sensitivity and lupus erythematosus. Arch. Dermatol. 91, 483-485, 1965. 3. Freeman, R. G., Knox, J. M., and Owen, D. W., Cutaneous lesions of lupus erythematosus induced by monochromatic light. Arch. Dermatol. 100, 677-682, 1%9. 4. Zamansky. G. B., Sunlight-induced pathogenesis in systemic lupus erythematosus. J. Invesr. Dermarol. 85, 179-180, 1985. 5. Parsons, P. G.. and Goss, P., DNA damage and repair in human cells exposed to sunlight. Photochem. Photobiol. 32, 635-641, 1980. 6. Golan, T. D.. and Borel, Y., Increased photosensitivity to near-UV light in murine SLE. J. Immunol. 132, 705-710, 1984. 7. Smith, K. C., Molecular aspects of the interaction of far UV radiation with living matter. In “International Symposium on Current Topics in Radiobiology and Photobiology” (R. M. Tyrrell, Ed.), p. 61, Rio de Janeiro, 1978. 8. Parrish, J. A., Anderson, R. R., Urbach, F.. and Pitts, D., Effect of ultraviolet radiation on microorganisms and animal cells. In “UV-A: Biological Effects of Ultraviolet Radiation with

IN VITRO-INCREASED

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

21. 22. 23.

24. 25. 26. 27. 28. 29. 30.

UVB SUSCEPTIBILITY

IN SLE

303

Emphasis on Human Response to Longwave Ultraviolet” (J. A. Parrish, R. R. Anderson, F. Urbach, and D. Pitts, Eds.), pp. 85-106, Plenum, New York, 1978. Tan, E. M., and Stoughton, R. B., Ultraviolet light alteration of cellular DNA in vivo. Proc. N&f. Acad. Sci. USA 62, 708-714, 1%9. Tan, E. M., Antibodies to DNA irradiated with ultraviolet light: Detection by precipitins and immunofluorescence. Science 161, 1353-1354, 1968. Furukawa, F., Lyon, hf. B., and Norris, D. A., Susceptible cytotoxicity to ultraviolet B light in fibroblasts and keratinocytes cultured from autoimmune-prone MRL/Mplpr/lpr mice. Cfin. Immunol. Immunopathol. 52, 460-472, 1989. Ansel, J. C., Mounts, J., Steinberg, A. D., DeFabo, E., and Green, I., Effect of UV light on autoimmune strains of mice: Increased mortality and accelerated autoimmunity in BXSB male mice. J. Invest. Dermatol. 85, 181-186, 1985. Zamansky, G. B., Minka, D. F., Deal, C. L., and Hendricks, K., The in vitro photosensitivity of SLE skin fibroblasts. J. Zmmunol. 134, 1571-1576, 1985. Tan, E. M., Cohen, A. S., Fries, J. P., Masi, A. T., McShane, D. J., Rothfield, N. F., Schaller, J. G., Talal, N., and Winchester, R. J., The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum. 25, 1271-1277, 1982. Rogers, J. C., Boldt, D., Komfeld, S., Skinner, S. A., and Valeri, C., Excretion of DNA by lymphocytes stimulated with PHA or antigen. Proc. N&l. Acad. Sci. USA 69, 1685-1689, 1972. Lambert, B., Ringborg, U., and Swanbeck, G., Ultraviolet induces DNA repair synthesis in lymphocytes from patients with actinic keratosis. J. Invest. Dermatol. 67, 594-598, 1976. Smith, P. J., and Paterson, M. C., Abnormal responses to mid-ultraviolet light of cultured tibroblasts from patients with disorders featuring sunlight sensitivity. Cancer Res. 41, 51 l-518, 1981. Spiegler, P., and Norman, A., Kinetics of unscheduled DNA synthesis induced by ionizing radiation in human lymphocytes. Radiut. Res. 39, 400-%12, 1%9. Beighlie, D. J., and Teplitz, R. L., Repair of UV-damaged DNA in systemic lupus erythematosus. J. Rheumatol. 2, 149160, 1975. Altman, H., Eberl, R., and Tuschl, H., Investigations on mechanisms involved in immunopathogenesis of rheumatoid arthritis, SLE, and animal models. In “Organic Manifestations and Complications in Rheumatoid Arthritis” (R. Eberl and M. Rosenthal, Eds.), pp. 79-85, Schattauer, Stuttgart, 1976. Cleaver, J. E., DNA damage and repair in light-sensitive human skin disease. J. Invest. Dermatol. 54, 181-195, 1970. Harris, G., Asbery, L., Lawley, P. D., Denman, A. M., and Hylton, W., Defective repair of 06-methylguanine in autoimmune disease. Lancet 2, 952-956, 1982. Takeuchi, F., Otsuka, F., Enomoto, T., Hanaoka, F., Yamada, M., Kamatani, N., Nakano, K, Mats&a, K., Tanimoto, K., Morita, T., and Myamoto, T., Decrease in the reaction of DNA repair synthesis in response to nicotinamide or 3aminobenzamide in ultraviolet irradiated systemic lupus erythematosus lymphocytes. J. Rheumatol. 12, 504-507, 1985. Zamansky, G. B., Kleinman, L. F., Kaplan, J. C., and Black, P. H., Effect of UV light irradiation on the survival of NZB mouse cells. Arthritis Rheum. 23, 866-867, 1980. Reddy, A. L., and Fialkow, P. J., Chromosome fragility in New Zealand black mice: Effect of ultraviolet and gamma radiation on fetal fibroblasts in vitro. J. Natl. Cancer Inst. 64, 939-941, 1980. McGrath, H., Bak, E., and Michalski, J. P., Ultraviolet-A light prolongs survival and improves immune function in (NZBIW) F, mice. Arthritis Rheum. 30, 557-561, 1987. Golan, T. D., and Borel, Y., Spontaneous increase of DNA turnover in murine systemic lupus erythematosus. Eur. J. Immunol. 13, 430433, 1983. Emerit, I., and Michelson, A. M., Chromosome instability in human and murine autoimmune diseases: Anticlastogenic effect of superoxide dismutase. Acta Physiol. Stand. 492(Suppl.) 59, 1980. Compton, L. J., Steinberg, A. D., and Sane, H., Nuclear DNA degradation in lymphocytes of patients with systemic lupus erythematosus. J. Immunol. 133, 213-216, 1984. LeFeber, W. P., Norris, D. A., Ryan, S. S., Huff, J. C., Kubo, M., Boyce, S. T., Kotkin, B. L.,

304

GOLAN,

FOL-IYN,

AND

ROUEFF

and Weston. W. L.. Ultraviolet light induces expression of selected nuclear anttgens on cultured human keratinocytes. .I. C/in. Invesf. 74. 1545-1551. 1984. 31. Furukawa. F.. Kashihara-Sawami, M.. Lyons. M. B., and Norris, 0. A.. Hindingofanttbodies to the extractable nuclear antigens SS-AIRo and SS-B/La is induced on the surface of human kerdtinocytes by ultraviolet light: Implications for the pathogenesis of photosensitive cutaneou> lupus. J. Invest. Dermatol. 94, 77-85, 1990. 32. Maddison. P. H.. Mogavero, H., Provost. T. T., and Reichlin. M.. The clinical significance ot autoantibodies to a soluble cytoplasmic antigen in systemic lupus erythematosus and other connective tissue diseases. J. Rheumatol. 6, 189-195, 1979. 33. Synkowski, P. R.. Reichlin. M.. and Provost. T. T., Serum autoantibodies in systemic lupus erythematosus and correlation with cutaneous features. J. Rheumatd 9. 380-385. 1982. Received

January

17. 1990; accepted

with

revision

October

9. 1990