1 mice

Journal of Neuroimmunology, 27 (1990) 79-84

79

Elsevier JNI 00909

Constant darkness enhances autoimmunity to type II collagen and exaggerates development of collagen-induced arthritis in DBA/1 mice I. H a n s s o n 1, R. H o l m d a h l 2 a n d R. M a t t s s o n 1 I Department of Zoophysiology, University of Uppsala, S-751 22 Uppsala, Sweden, and 2 Department of Medical and Physiological Chemistry, BMC, University of Uppsala, S-751 23 Uppsala, Sweden

(Received 17 July 1989) Revised, received 5 September 1989) (Accepted 5 September 1989) Key words." Arthritis, collagen II-induced; Autoimmunity; Melatonin; Darkness, constant; Collagen, type II; (Mouse)

Summary The humoral function of the pineal gland is known to be strongly dependent on environmental lighting. Melatonin, the best characterized of the photo-dependent pineal hormones, has been reported to affect immune responses in mice. It has been hypothesized that the development of some types of psychosomatic and autoimmune diseases could be due to a disturbed release of this hormone. The present investigation was performed in order to evaluate effects of constant darkness (physiological stimulation of pineal melatonin synthesis) and constant light (physiological suppression of pineal melatonin synthesis) on the course of an experimental autoimmune model, the type II collagen-induced arthritis (CIA) in D B A / 1 female mice. Mice kept in darkness develop more severe arthritis than those kept in constant light or in a normal d a r k / l i g h t rhythm (12 h light/12 h dark). Levels of anti-type II collagen antibodies were higher in mice kept in darkness, and the spleens of these animals were enlarged. Since castration of female D B A / 1 mice enhances the severity of CIA, and since melatonin is known to exert effects on gonadal function, the experiment was repeated using oophorectomized mice. The same difference in arthritis severity between darkness- and light-exposed mice was obtained in this second experiment. We conclude that the exacerbation of arthritis in darkness is due to a darkness-induced change in levels of critical neurohumoral compound(s), that via gonadal independent mechanisms affect the autoimmune response. The exaggerated severity and chronicity of arthritis may be due to higher levels of melatonin in these animals.

Introduction There is increasing evidence that the neuroendocrine status of the brain more or less dramati-

Address for correspondence: Inga Hansson, Department of Zoophysiology, Box 560, S-75122 Uppsala, Sweden.

cally influences the immune function, and also affects the course of m a n y autoimmune diseases (Jankovic et al., 1970; Fox and Newberry, 1984; Plontikoff et al., 1986; Smith and Blalock, 1986; Blask and Hill, 1988). The mechanisms by which the central nervous system (CNS) influences immune responses are most often indirect and the knowledge of their nature is limited. One example

0165-5728/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

80 is disturbances of the l i g h t / d a r k rhythm, which induce complex changes within the CNS, and in parallel have profound influence on the immune system (Maestroni et al., 1986). In almost all higher vertebrates the basic mediator of these changes appears to be the pineal gland, since the synthesis and release of its hormone melatonin (N-acetyl-5methoxytryptamine) are strongly photo dependent (Ebadi, 1984). Melatonin is known to act at the hypothalamic level (Niles et al., 1979), and thereby affects important physiological functions, such as thermoregulation and pituitary release of various hormones (Reiter, 1984). Some effects of pineal melatonin release, which normally is maximal during nocturnal darkness, have been suggested to be dependent on decreased serotonin levels, since serotonin is used for the synthesis of melatonin (Ebadi, 1984). The biological effects of pineal melatonin synthesis/release differ strongly between various mammalian species, and its exact role has been particularly difficult to define in animals that are not seasonal breeders, such as mice, rats and humans. Interestingly, Maestroni and coworkers (1986, 1987, 1988) recently reported that evening injections of this hormone stimulated humoral immune responses in mice. This upregulation, which appeared to be mediated via endogenous opiates, was observed in antigen-primed animals only, indicating that pineal immunoregulation is of importance for maintenance of an immune response. Maestroni and coworkers hypothesized that the pineal gland may fulfill a role as an immunological transducer, and that a disturbed pineal melatonin release could be of significance for the development of some types of psychosomatic and autoimmune diseases. A complication in analyses of the role of pineal melatonin release is the fact that compensatory melatonin-releasing tissues exist, which means that a total elimination of melatonin synthesis cannot be obtained by a conventional pinealectomy (Ralph, 1981). In addition, it appears that the hypothalamus exhibits an endogenous circadian fluctuation in its sensitivity to melatonin (Maestroni et al., 1986). Primarily, the present investigation was performed in order to evaluate as to what extent constant environmental light (optimal physio-

logical suppression of endogenous melatonin synthesis) or constant environmental darkness (optimal physiological stimulation of endogenous melatonin synthesis) could affect the incidence or course of an experimental autoimmune model, the type II collagen-induced arthritis (CIA) in D B A / 1 mice. Secondarily, we wanted to clarify whether these effects were mediated via ovarian hormones, since (1) the pineal gland is known to affect gonadal function, and (2) sex hormones are known to strongly affect the course of C1A (Holmdahl et al., 1986).

Materials and methods

Animals Female D B A / 1 mice were bred at the Animal Unit, Biomedical Center, Uppsala, Sweden. The female mice were 90 days old at the beginning of the experiments (weight= 20_+ 1 g) and were housed at the Department of Zoophysiology, Uppsala, during the experiments.

Experimental design Two experiments were performed according to the procedure described below. In experiment 1 normal female mice were used, and in experiment 2 the investigation was repeated using castrated females. Three rooms of identical design and connected to the same ventilation system were used for the experiments. In one of the rooms the standard lighting (approximately 300 lux) was continuously on (light-room), and in the second room it was continuously off (dark room). The third room had a standard l i g h t / d a r k cycle (12 h light/12 h dark). After immunization with collagen II (age 90 days) the mice were randomly divided into three groups (L = mice in the light room, D = mice in the dark room, and C = controls in the room with standard light/dark rhythm) and placed in the three different rooms. All mice were weekly investigated for clinical signs of arthritis, and the severity of arthritis was evaluated according to a special scoring system (see below). Investigation of clinical scores in the dark room was performed under a red dark room lamp. At the end of the experiment the degree of paw swelling was recorded, and sera

81 collected for determination of anti-collagen antibody titers. All mice were then sacrificed, organ weights recorded and splenic content of immunoglobulin-secreting cells determined.

Preparation of collagen Rat type II collagen (rat CII) was prepared from a rat chondrosarcoma by pepsin digestion followed by salt precipitation and DEAE-cellulose chromatography (Smith et al., 1975). Induction of arthritis Native collagen type II dissolved in 0.1 M acetic acid at a concentration of 1 m g / m l was emulsified in an equal volume of complete Freund's adjuvant at + 4 ° C and 100 #1 of this emulsion was injected intradermally in the skin around the base of the tail. Animals were then observed weekly for development of arthritis (normally developed after 6 weeks - - incidence in normal females: 20% (6 weeks) or 60% after 14 weeks - - incidence in males: 40% (6 weeks) or 90% (14 weeks); higher incidence in ovariectomized females, lower incidence in castrated males (Holmdahl et al., 1985)). In a second experiment the mice were oophorectomized 4 weeks before immunization with 50 /~1 of the above-described emulsion. Castration was performed according to the standard procedure; the mice (60 days old) were anesthetized with pentobarbital (Mebumal) (ACO, Sweden) (30 mg/kg) and the ovaries removed from the dorsal side. Evaluation of arthritis Scoring of arthritis after inspection was made using a scoring system where arthritis in each paw is scored from 1 to 3 (1, detectable swelling of one or more joints; 2, severe swelling in more than one joint; and 3, severe arthritis in the entire paw (Holmdahl et al., 1986)). Protein A plaque assay This plaque assay has been described in detail elsewhere (Gronowicz et al., 1976). In short, 15/~1 of protein A-coated sheep red blood cells (30%, v / v ) were mixed with 30 #1 of the appropriately diluted lymphocyte suspension (one million/ml), 10/~1 guinea pig complement (diluted 1 : 2), 10 #1

developing rabbit anti-mouse Ig antibody (Nordic) (diluted 1:40) and 100/~1 0.5% warm (45°C) agar solution (all reagents in balanced salt solution). The mixture was spread out on a Petri dish and allowed to incubate at 38°C in humidified air. Plaques were counted after 18 h and the total content of Ig secretors/spleen calculated (on the basis of number of cells added per dish and the total cell yield of each spleen).

Enzyme-linked immunosorbent assay (ELISA) Quantification of anti-CII reactive antibodies in sera was performed as previously described (Holmdahl et al., 1986). Briefly, micro-ELISA plates (Dynatech, Plochingen, F.R.G.) were coated with native mouse CII. Purified anti-CII reactive antibodies were used as standards and titrated in 10-fold dilution steps in parallel with the unknown serum samples in ELISA plates. All tests were carried out in duplicates. The amount of bound antibody was estimated after incubation with goat anti-mouse IgG-Fc and goat anti-mouse IgM-Fc coupled to alkaline phosphatase (Jackson Lab., Avondale, PA, U.S.A.). For each serum sample, the dilution that gave 50% absorbance as compared with the maximum value obtained for the standard antibodies was related to the concentration (in m g / m l ) of standard antibodies to give the same 50% absorbance value. By this method all calculations were performed with data from the steep portion of the slope, where the sample titration curves were parallel to the standard titration curves. The purified standard antibodies were obtained from rat CII-immunized D B A / 1 mice.

Results

Darkness enhances severity of CIA in normal DBA / 1 female mice As demonstrated in Figs. 1 and 2, mice kept in constant darkness (D mice) showed a tendency of earlier development and increased incidence of arthritis (not significant) and a significantly higher severity of arthritis as compared to those kept in constant light (L mice) or in a normal light/dark rhythm (C mice). In addition, the severity of arthritic disease was progressively increasing dur-

82 TABLE 1 SPLEEN WEIGHTS Treatment

L D C OvX L OvX D

PAW THICKNESS AND SERUM LEVELS OF ANTI-COLLAGEN

No. of mice

M e a n spleen weight (mg _+ SD)

Paw thickness ( m m ± SD) Front

Hind

17 16 10 20 20

85 100 84 92 104

1.82 2.00 1.70 n.d. n.d.

1.57 ± 0.12 1.85 ± 0.29 * * * 1.56 + 0.06 n.d. n.d.

L = c o n s t a n t light, D = c o n s t a n t a Not determined. * Significant difference from * * Significant difference from * * * Significant difference from

_+ 11 +_ 14 * * ± 6 ± 12 ± 14 * *

cq

5

n.o o_o []

/0--0-0

4

+.

/i~L+'+/~/

~

0 ,d 30 40

50

+_

±.+, ,

, 60 70 80 90 tOO days a f t e r immunisat~on

+\+

, 110

, I20

130

Fig. 1. Severity of arthritis a m o n g female D B A / 1 mice k e p t in c o n s t a n t d a r k n e s s (filled squares), c o n s t a n t light (open squares) or a n o r m a l 12 h l i g h t / 1 2 h d a r k cycle (crosses), q u a n t i f i e d as m e a n arthritic s c o r e s / i n d i v i d u a l o n different days after imm u n i z a t i o n w i t h type II collagen. * Significant differences ( p < 0.05, M a n n - W h i t n e y U-test). 100 i_i/I.I-I

I-II-I.I-I--II-II

/

91.1 ± 27 1 6 5 . 6 + 1 1 3 ** 82.3 +_ 1 7

light g r o u p a n d controls ( p < 0.01 ; two-tailed M a n n - W h i t n e y U-test). light g r o u p and controls ( p < 0.05; S t u d e n t ' s t-test). light g r o u p a n d controls ( p < 0.0005; two-tailed M a n n - W h i t n e y U-test).

,/

s

± 0.22 _+ 0.26 * ± 0.19 ~

Anti-collagen II a n t i b o d i e s ( f f g / m l ± SD)

darkness, C = control, O v X = ovariectomized.

dark u light Co~tr

%

I1 A N T I B O D I E S

ing the experiment, indicating that darkness promoted chronicity of this normally acute disease. Furthermore, the levels of autoantibodies reactive with mouse CII in serum of D mice were significantly elevated in comparison to the L and C animals (Table 1). In contrast, the total splenic content of Ig secretors did not differ significantly between the groups (D mice 276,000 + 73,000 and L mice 287,000 ___81,000 plaque-forming cells (pfc) per spleen). Spleen weights were significantly higher in D mice (Table 1), while there were no differences in thymus weights between the D, L, and C groups ( 3 2 + 8 , 3 3 + 6 and 29_+3 mg, respectively).

o

-lJ•

0 hght • c0rtr

In

B 8

/

i

•

//

I i

/

o

•

:L,

/

4r

"e/

o

.b

t /'V/ 45

30

40

50

60

70

days a f t e r

80 go tOO I I 0 immunisat~on

~20

I30

Fig. 2. I n c i d e n c e of arthritis in female D B A / I mice k e p t in c o n s t a n t d a r k n e s s (filled squares), c o n s t a n t light (open squares) or a n o r m a l 12 h l i g h t / 1 2 h d a r k cycle (crosses).

55 65 75 days after' immun~sat~:r

85

Fig. 3. Severity of arthritis a m o n g o v a r i e c t o m i z e d D B A / 1 mice kept in c o n s t a n t d a r k n e s s (filled circles) or c o n s t a n t light (open circles) q u a n t i f i e d as m e a n arthritic s c o r e s / i n d i v i d u a l on different d a y s after i m m u n i z a t i o n w i t h type II collagen. * Significant differences ( p < 0.05, M a n n - W h i t n e y U-test).

83 ~00

•~ •

/

°

/

~- / °/°

50 •

45

~ark o light

•

e/e--oe~

55

o~

65

75

85

days 8fter immunisation

Fig. 4. Incidence of arthritis in ovariectomized D B A / 1 mice kept in constant darkness (filled circles) or constant light (open circles).

Castration of female DBA / 1 mice does not modulate changes induced by darkness In order to investigate whether the darknessinduced influence on the CII autoimmunity was mediated via effects by the ovaries, the experiment was repeated with castrated female mice. In this experiment we reduced the dose of CII used for induction of arthritis since castrated female mice are more sensitive to induction of CIA than naive females due to the influence of estrogen (Holmdahl et al., 1986a; Jansson et al., 1989). As in the previous experiment using naive female mice, castrated mice kept in constant darkness showed splenic enlargement (Table 1) and developed a more severe arthritis as compared with mice kept in constant light (Fig. 3), demonstrating that the darkness-induced enhancement of CIA is mediated via mechanisms distinguished from gonadal effects. As in the first experiment the incidence of arthritis was not significantly affected, although slightly higher in mice kept in darkness (Fig. 4).

Discussion

Our results demonstrate that environmental darkness induces a more severe and chronic development of CIA, concomitant with enhancement of CII autoimmunity, in female D B A / 1 mice. CIA is a well-characterized disease in mice (Courtenay, 1980), with many pathogenetic similarities with rheumatoid arthritis (reviewed in Holmdahl et al., 1988). It is a T cell-dependent disease with patho-

genetic involvement of CII-reactive autoantibodies (Holmdahl et al., 1988). Immunization with autologous CII induces a chronic and progressive form of arthritis while immunization with heterologous CII induces a more self-limited form of disease (Holmdahl, 1986b). In the present investigation CIA was induced with heterologous rat CII and as expected the arthritis developed 4-5 weeks after immunization and followed a self-limited course as judged from mice kept under normal dark/light conditions. Mice kept in constant darkness developed chronic arthritis with a progressively increasing severity. The exaggeration of CIA and the increase in antiCII autoantibody serum levels of mice kept in darkness thus indicate that the function of the immune system has been affected, and that nerve signals from the retina have induced the changes in the brain function that finally caused this immunoregulation. There are reasons to assume that the pineal synthesis/release of melatonin plays a key role in this immunoregulation, since the synthesis of this hormone is dependent on the degree of retinal stimulation. Melatonin synthesis, which is known to be controlled by peripheral sympathetic nerve fibers, is normally maximal when the retinal stimulation (by means of light) is at a minimum. Therefore, it can be postulated that melatonin synthesis (which, due to endogenous rhythm, is complex) should be maximal in mice kept in darkness, and minimal in mice kept in constant light. In light of the recent reports of Maestroni and coworkers (1986, 1987, 1988), showing that melatonin can play a significant role in immunoregulation in mice, we believe that the effects observed in the present study basically are due to such an elevated melatonin synthesis. However, melatonin is not believed to interfere directly with lymphocyte function (Maestroni et al., 1986), but has been suggested to act via opiate receptors (Maestroni et al., 1987, 1988). Interestingly, it was long ago reported that adult pinealectomy dramatically reduced (diminished) the clinical signs of allergic encephalomyelitis in rats subjected to experimental induction of this special autoimmune disease (Jankovic et al., 1970). This observation strengthens the suggestion that pineal hormones (melatonin?) can stimulate autoimmune reactions.

84

Since melatonin is known to affect pituitary lutropin/follitropin (LH/FSH) release and gonadal function in many m a m m a l i a n species, and since the course of CIA is known to be strongly influenced by gonads, primarily due to estrogen-suppressive effects, in female D B A / 1 mice, we decided to evaluate whether the observed effect of darkness was mediated via a melatonininduced modulation of ovarian function. This was obviously not the case, and the observed effect of darkness must be mediated via other mechanisms, of which the opiate system is an important candidate.

Acknowledgements This study was financially supported by grants from the Swedish Natural Science Research Council (R.M.), Axson Johnson's Fund (I.H.), the Swedish Medical Research Council, R M R and the Gustav Vs 80 Year Foundation (R.H.). The authors also wish to thank Dr. G. Maestroni for consultation and Mikael Karlsson for skillful help with the anti-CII ELISA measurements.

References Blask, D.E. and Hill, S.M. (1988) Melatonin and cancer: basic and clinical aspects. In: A. Miles, D.R.S. Philbrick and C. Thompson (Eds.), Melatonin - - Clinical Perspectives, Oxford University Press, Oxford - New York - Tokyo, pp. 128-173. Courtenay, J.S., Dallman, M.J., Dayan, A.D., Martin, A. and Mosedal, B. (1988) Immunisation against heterologous type 1I collagen induces arthritis in mice. Nature 283, 666-667. Ebadi, M. (1984) Regulation of the synthesis of melatonin and its significance to neuroendocrinology. In: R.J. Reiter (Ed.), The Pineal Gland, Raven Press, New York, pp. 221-251. Fox, B.H. and Newberry, B.H. (1984) Impact of Psychoneuroendocrine Systems in Cancer and Immunity, C.J. Hogrefe, New York. Gronowicz, E., Coutinho, A. and Melchers, F. (1976) A plaque assay for all cells secreting Ig of a given type or class. Eur. J. lmmunol. 6, 588-590. Holmdahl, R., Jansson, L., Gullberg, D., Rubin, K., Forsberg, P.-O. and Klareskog, L. (1985) Incidence of arthritis and autoreactivity of anti-collagen antibodies after immuniza-

tion of DBA/1 mice with heterologous and autologous collagen II. Clin. Exp. lmmunol. 62, 639-646. Holmdahl, R., Jansson, L. and Andersson, M. (1986a) Female sex hormones suppress development of collagen-induced arthritis in mice. Arthritis Rheum. 29, 1501-1509. Holmdahl, R., Jansson, L., Larsson, E., Rubin, K. and Klareskog, L. (1986b) Homologous type II collagen induces chronic and progressive arthritis in mice. Arthritis Rheum. 29, 106-113. Holmdahl, R., Klareskog, L., Andersson, M. and Hansen, C. (1986c) High antibody response to autologous type II collagen is restricted to H-2q. Immunogenetics 24, 84-89. Holmdahl, R., Andersson, M., Enander, I., Goldschmidt, T., Jansson, L., Larsson, P., Nordling, C. and Klareskog, L. (1988) Nature of the type II collagen autoimmunity in mice. Int. Rev. Immunol. 4, 49-64. Jankovic, B., Isacovic, K. and Petrovic, S. (1970) Effects of pinealectomy on immune reactions in the rat. Immunology 18, 1-6. Jansson, L. and Holmdahl, R. (1989) Oestrogen induced suppression of collagen arthritis. IV. Progesterone alone does not affect the course of arthritis but enhances the oestrogen-mediated therapeutic effect. J. Reprod. Immunol. 15, 141-150. Larsson, P., Holmdahl, R., Dencker, L. and Klareskog, L. (1985) In vivo treatment with W3/13 (anti-pan T) but not with OX8 (anti-suppressor/cytotoxic T) monoclonal antibodies impedes the development of adjuvant arthritis in rats. Immunology 56, 383-391. Maestroni, G.J.M., Conti, A. and Pierpaoli, W. (1986) Role of pineal gland in immunity I. J. Neuroimmunol. 13, 19-30. Maestroni, G.J.M., Conti, A. and Pierpaoli, W. (1987) Role of pineal gland in immunity II. Clin. Exp. Immunol. 68, 384-391. Maestroni, G.J.M., Conti, A. and Pierpaoli, W. (1988) Role of pineal gland immunity Ill. Immunology 63, 465-469. Niles, L.P., Wong, Y.-W., Mishra, R.K. and Brown, G.M. (1979) Melatonin receptors in brain. Eur. J. Pharmacol. 55, 219-220. Plotnikoff, N.P., Faith, B.E., Murgo, A.J. and Good, R.A. (1986) Enkephalins and Endorphins: Stress and the Immune System, Plenum Press, New York. Ralph, C.L. (1981) In: N. Birau and W. Schloot (Eds.), Melatonin - - Current Status and Perspectives, Pergamon Press, New York, pp. 35-46. Reiter, R.J. (1984) The Pineal Gland, Raven Press, New York. Smith, E.M. and Blalock, J.E. (1986) A complete regulatory loop between the immune and neuroendocrine systems operates through common signal molecules (hormones) and receptors. In: N.P. Plotnikoff, R. Faith, A.J. Murgo and R. Good (Eds.), Enkephalins and Endorphins: Stress and the Immune System, Plenum Press, New York, 1986. Smith, B.D., Martin, G.R., Dorfman, A. and Swarm, R. (1975) Nature of the collagen synthesized by a transplanted chondrosarcoma. Arch. Biophys. Biochem. 166, 181.