- Email: [email protected]
Biphasic changes in behavioral, endocrine, and sympathetic systems in adjuvant arthritis in Lewis rats
Brain ResearchBulletin,Vol. 39, No. I, pp. 33-37. 1996
Copyright© 1995ElsevierScienceInc. Printedin the USA.All rightsreserved 0361-9230/96$15.00 + .00 ELSEVIER
0361-9230(95)02037-3
Biphasic Changes in Behavioral, Endocrine, and Sympathetic Systems in Adjuvant Arthritis in Lewis Rats HIROAKI TANAKA,.1 YOICHI UETA,* UKI YAMASHITA,1- HIROSHI KANNANt AND HIROSHI YAMASHITA*
*Department of Physiology, University of Occupational and Environmental Health, Yahatanishiku, Kitakyushu, 807, Japan tDepartment of Immunology and Parasitology, Hiroshima University School of Medicine, Japan tDepartment of Physiology, Miyazaki Medical College, Japan [Received 31 May 1994; Accepted 9 May 1995] thritis was 60 and 80% in Fisher rats and Sprague-Dawley (SD) rats, respectively, in our unpublished observation. The severity of the adjuvant arthritis, expressed as the mean arthritis index [3] was 12.3 _ 0.4, 6.6 _ 2.0, and 2.0 _+ 1.1 for Lewis, SD, and Fisher rats, respectively (our unpublished observation). In the streptococcus cell wall (SCW) induced arthritis differences in susceptivity between Lewis and Fisher rats were found to be associated with lower responsiveness of the HPA axis in Lewis rats to the arthritis inducible antigen, SCW, and/or to inflammatory cytokines, compared to that in Fisher rats [16,17]. In these studies, however, responses of the HPA axis were investigated only during the first four hours following the SCW injections. In most other works outbred SD rats were used for studies of the HPA axis during experimental arthritis [13], because they show strong activation of the HPA axis and intermediate susceptivity to the arthritis. Although hormonal changes are known to occur in the experimental arthritis in outbred rats, very little study has been done in inbred strain, such as Lewis rats. It is likely that biological substances produced by activated macrophages and T lymphocytes [1], such as immune cytokines, cause these changes during disease process of arthritis, yet we do not have any concrete evidence. To study relationships between the central nervous system (CNS), endocrine, and immune reactions in the adjuvant arthritis, it is preferable to use an inbred strain, such as Lewis rats. Our aim of this article is to investigate in Lewis rats the CNS responses (body temperature, ingestive, and motor behaviors), activation of the HPA axis (plasma levels of ACTH, corticosterone, and vasopressin), and changes in the sympathetic nerve activities and heart rate throughout development of the adjuvant arthritis. The results suggest that Lewis rats are suitable experimental model for the future studies of neuroimmunomodulation.
ABSTRACT: Adjuvant arthritis (AA) is an experimental model for rheumatoid arthritis, and ia induced most easily in inbred Lewis rats by an intradermal injection of hest-killed Mycobacterium tuberculosis(MT) in incomplete Freund's adjuvant. Susceptivity to the arthritis in Lewis ral~ is thought to be related to a defect in their responses of the hypethalamo-pituitary-adrenal (HPA) axis to the disease. Because the use of an inbred strain is neeessary for our immunological studies, we examined in Lewis rats changes in behavior, the HPA axis, and sympathetic nerve activities during developmer,t of the adjuvant arthritis. Following intradermal injections of heet-killed MT in adjuvant, the arthritis began to develop on day 112, reaching its maximum severity on day 21, and remained at the level for over a month. The body temperature rose from day 0 to 5 (the primary phaseBbeforo the onset of the arthritis). It then tell to normal temperature, and again rose from day 10 to 21 (the secondary phase--with fully developed arlhritis). The behavioral (physical activity, food, and water intake) and hormor,al parameters [plasma adrenocorticotropic hormone (ACTtl) and corticosterone levels] also changed in two phases, similar to those observed in the tamperature responses. No change in plasma vasopressin level was observed. SympetheUc nerve activiUse, assessed by changes in plasma noradrenalin levels;, increased more in the primary than in the secondary phase. 111e possible causes for the biphasic changes associated with development of arthritis are discussed. KEY WORDS: Adjuvant arthritis, Lewis rat, Ingestive behavior, Physical activity, Body temperature, Heart rate, Hypothalamopituitary-adrenal axis, Sympathetic nerve activity.
INTRODUCTION Adjuvant arthritis (AA) is a good experimental model for rheumatoid arthritis, one of the autoimmune diseases. The arthritis is induced in rats by an intradermal injection of heat-killed Mycobacterium tuberculosis (MT) in incomplete Freund's adjuvant. Susceptivity to the arthritis, severity, and incidence of the arthritis followed by the MT injection vary among different strains of rats. The arthritis is induced most readily in Lewis rats; the incidence of the arthritis was found to be 100% followed by the MT injection in our studies. On the other hand, the incidence of the ar-
METHOD
Animals Male Lewis (LEW/N) rats 6 - 1 0 weeks of age (body weight: 180-250 g) were used and kept under constant room temperature (22-25°(2), humidity (40-60%), and a 12-h light-dark cycle
To whom requests for reprints should be addressed. 33
34
(lights on: 0730 to 1930 h). Standard laboratory rat chow or pellets and water were given ad lib.
Induction of Arthritis Animals were injected with 1 mg heat-killed Mycobacterium tuberculosis H37 Ra in incomplete Freund's adjuvant (Difco Laboratories, Detroit, MI) intradermally at the base of their tails, as described elsewhere [11]. Control animals were injected with incomplete Freund's adjuvant alone.
Assessment of Arthritis The arthritis index of the animals were scored by grading each paw from 0 to 4, based on erythema, swelling, and deformity of the joints ( 0 - - n o erythema or swelling; 1--slight erythema and swelling of one of the toes; 2--erythema and swelling of more than one toe; 3--erythema and swelling of the ankle; 4 - - c o m plete erythema and swelling of toes and ankle and incapacity to bend the ankle; the highest score could reach 16, because each limb is evaluated separately) [3]. The hind paw volume was measured on day 21 and 28 after MT injection by a plethysmograph.
Measurement of Behavioral Changes Physical activity of the animal was recorded with an automatic behavioral measurement system (Astec, Fukuoka, Japan). In brief, the physical activity was counted when an animal crossed any of four infrared beams fixed at 4 cm above the cage floor (floor width: 30 × 30 cm, height: 43 cm) and stored every 30 rain in a computer. Feeding behavior was assessed by weighing standard rat chow every 24 h with a conventional metabolism cage, or by the number of pellets counted with an automatic pellet feeder. Drinking behavior was also assessed by weighing a bottle every 24 h in a conventional metabolism cage, or by using a counting devise to measure the number of water drops drunk. Body temperature and heart rate were recorded with a telemetry system (Data Science, Inc., St. Paul, MN) through a transmitter implanted in the animal intraperitoneally. The transmitter was implanted under general anesthesia (sodium pentobarbitone, 50 mg/kg IP) at least 14 days before the adjuvant injection. The signals of body temperature and heart rate from the transmitter were recorded as averaged values for 5 s every 30 rain through three receivers around the cage and were stored for later analysis by the computer.
Measurement of Hypothalamo--Pitaitary-Adrenal (HPA) Axis Activation Animals were handled every day for 7 days for them to be accustomed to the later experimental procedures. They were decapitated, and trunk blood was collected in an Ethylenediaminetetraacetic Acid Disodium Salt Dihydrate (EDTA-2Na)-containing tube within 3 h from the lights-on time on days 4, 7, 14, and 21 following the MT injection (day 0). The blood samples were centrifuged (3000 rpm) at 4°C for 15 rain and the plasma were aliquoted and stored at -80°C until the assay. Plasma ACTH, corticosterone, and vasopressin (AVP) were measured by radioimmunoassay using RIA kits (Mitsubishi Petrochemical, Japan, for ACTH and AVP; ICN Biomedical, USA, for corticosterone). The thymus and adrenal glands were removed from the animals on day 28 and weighed. All hormones were measured in the same group of animals.
Measurement of Plasma Noradrenalin Level Blood samples (0.3 ml) were collected before and on days 4, 7, 14, and 21 after the MT injection in an EDTA-2Na-containing
TANAKA ET AL.
tube via a venous catheter implanted into the superior vena cava through the jugular vein 7 days before the injection. The plasma were aliquoted and stored by the same method as did for the hormone assays. Plasma noradrenalin was assayed using liquid chromatography with electrochemical detection after batch alumina extraction, as described elsewhere [2].
Stastical Analysis Results are presented as means ± SEM. For statistical evaluation, Student's t-test and Mann-Whitney U-test were used. A value of p < 0.05 was considered to be significant. RESULTS
Time Course of Development of the Arthritis The arthritis suddenly developed approximately on day 13 following the MT injection (day 0), and reached its severest level around on day 21. At this time the averaged arthritis index was 12.32 ___ 0.42 (n = 31). This condition continued for several months without much change. Incidence of arthritis reached 100% after day 15 (Fig. 1A). The paw volume, the quantitative assessment of the arthritis, increased significantly when measured on days 21 and 28, compared to that of control animals (day 21, MT-injected rats, 1.88 _ 0.08 ml, n = 8; control rats, 1.34 ± 0.001 ml, n = 3; day 28, MT-injected rats, 2.15 ± 0.05 ml, n = 10; control rats, 1.63 ± 0.05 ml, n = 20) (Fig. 1B).
Changes in Body Weight, Food and Water Intake, Physical Activity, Heart Rate, and Body Temperature Body weight changes were expressed as a percentage of the body weight on day 0 and measured daily until day 21. Body weight of MT-injected rats (n = 19) increased linearly to day 10, but the weight gain stopped from day 10 to 21. In control (vehicle-injected) animals (n = 12) the body weight linearly increased to day 21. There was a small, yet significant, transient decrease in body weight of MT-injected rats on days 4 and 5. Daily food and water intakes measured from day 0 to 21 a r e shown in Fig. 2B and C. In MT-injected rats (n = 10), up to day 10 the amount of food and water intake was similar to that in control animals (n = 5), but decreased markedly from day 12 to 21. In addition, on days 1 and 3 food and water intakes were slightly decreased. The ingestive behavior of all animals showed
A
100
3
15
8o x m ~
40 Q.
o
"0
........
~b-
z5
days after MT injection -
30
21
28
days after injection
FIG. 1. Time courses in developmentof adjuvant arthritis in Lewis rats following injection of Mycobacterium tuberculosis (MT). (A) The arthritis index (closed circles) and arthritis incidence (open circles). (B) Changes in paw volume in arthritic (solid bars) and con~ol (open bars) animals on days 21 and 28. *p < 0.005, **p < 0.0005, compared to conU'ol,Student's t-test and Mann-Whitney U-test. Values are mean +_ SEM; numbers in parentheses are number of animals used.
SICKNESS RESPONSES IN ADJUVANT ARTHRITIC RATS 180
A --
_~2 ~ ~ ( 1 2 )
160 t 140 ,,o
.........
30
B~
: LZ_g.:"-'r I.~=. I (19)
~
O .....
6
' lb ..............
Z lO)
t~
1
[
3 5
c
.
.
.
.
.
.
.
.
b .....
.
s
'
~ b
~
**
J
..............
E -~
(5) 25, 10)
=
15
D ~
.,
.,.
~92°°i
,.
,.
.,.,
.
. ,.
.,
.,.
,.,.,.,.,
T
~'4400t
_ . . . , . ~ 4 " ~ ~ ~
48001 -10-8 .
,
(6, 05)
.
,
.
.
.
.
.
.
.
-6 -4 -2
.
.
,
0
2
4
.
,
. . . . .
6
8
,
.
,
,
,
.
,
.
,
,
10 12 14 16 18 20 22 24
days after injection
35
changes. The basal level of the morning plasma ACTH was 13.93 - 2.13 pg/ml (n = 7). In MT-injected rats A C I ~ markedly increased on day 4 (32.36 _ 3.29 pg/ml, n = 9), returned to the basal level on day 7 (17.19 _+ 2.9 pg/ml, n = 7). It rose again on day 14 (33.45 __+3.9 pg/ml, n = 13) and day 21 (25.18 ___3.25 pg/ml, n = 9 (Fig. 4A). In control rats (vehicle-injected) ACTH remained at the basal level on days 4 and 7, and rose slightly on days 14 and 21. The significant difference on ACTH levels between MT-injected and control animals occurred on day 4. An increase in plasma ACTH in control (vehicle-injected) animals on days 14 and 21, though at a lesser degree, is puzzling. It may be due to nonspecific responses to injection of the vehicle, incomplete Freund's adjuvant. To test reliability of our measurements, the level of plasma ACTH was determined after giving restrained stress for 20 min in normal rats. The value was 273.2 ___ 20.6 pg/ml (n = 7), similar to that reported by others [15]. The basal level of AVP measured in the morning was 12.9 ___ 3.8 pg/ml (n = 5), and this value did not change at all on any days tested in both the MT-injected and in control animals (Fig. 413). Marked changes in plasma corticosterone levels appeared in MT-injected rats (Fig. 4C). Its basal level was 42 _+ 17.8 ng/ml (n = 8), and increased dramatically on days 4, 7, 14, and 21. The differences between corticosterone levels of MT-injected and control rats were significant at all measurements (day 4 - - M T injected, 477.6 ___ 111.5 ng/ml, n = 8, control, 25.9 ___3.9 ng/ml, n = 9; day 7 - - MT-injected, 204.4 _ 52.1, n = 8, control, 23 - 6.2 ng/ml, n = 9; day 14--MT-injected, 166.2 _ 32, n = 9, control, 46.7 _ 18.8, n = 9; day 21--MT-injected, 540.9 _
FIG. 2. Time courses in chmages of body weight (A), food intake (B), water intake (C), and physical activity (D) of MT-injected rats (closed circles and squares) and those of control (vehicle injected) animals (open circles and squares). Body weight is expressed as percent of that on day 0 (A). Number of animals shown in parenthesis. Mean _+ SEM. *p < 0.05, **p < 0.01, compared to controls; Student's t-test and MannWhitney U-test. a clear circadian rhythm before and after the injection and throughout the period of the experiments (data not shown). The physical activity of MT-injected rats (n = 15) decreased significantly on day 0, and again on days 2 to 4, compared to that of controls (n = 6). It recovered on day 5 to 9, then steadily decreased again after day 10 (Fig. 2D). Before MT injections the body temperature of all animals showed a clear circadian rhythm, but this tended to disappear after day 10 when the temperature began to rise considerably with development of arthritis (Fig. 3A). After MT injections there was a transient rise in averaged body temperature (n = 6) on day 0, returning to normal immediately. The temperature rose again from day 12 to 21, while no change occurred in vehicle-injected rats (n = 4). There was a moderate increase from day 3 to 5 (Fig. 3C). The heart rate of MT-injected rats showed the same circadian rhythm as that in normal rats (Fig. 3B). The averaged heart rate increased from day 14 to 21 after MT injections (n = 6), compared to that of control animals (n = 4) (Fig. 3D). Thus, changes in ingestive behavior, physical activity, and body temperature in MT-injected rats occurred in two steps and can be divided into two phases; the primary phase, from day 0 to 5, during which no arthritis is present, and the secondary phase, from day 10 to 21, after the arthritis is developed.
Changes in Plasma Leveb: of ACTH, A VP, and Corticosterone, and in Weight of Adrenal Gland and Thymus Changes in the HPA axis were studied on days 4, 7, 14, and 21 after MT injection, when the behavioral responses showed marked
39
-4
-2
0
2
4
•6
8
10
12
14
16
18
20
-4
-2
0
2
4
6
8
10
12
14
16
18
20
400
4O
c
1
""'" I
"t
-I0 "-8 '-6 "-4 "-2 ' 6 "2 "'~ "6 "8 'I0 "12 "I~I"16 "18 "20 "22 "24
D :°°l 300 4
_
_r-* . 2 . _ ~
_ . . k ~ z ~ - i (6)
2001'0"-8 "'6 '-'4 '-2 " 6 "2 "4 " 6 ' 8 "1() "12'14. " 1 6 " 1 8 ' 2 0 " 2 2 " 2 ~ I days after injection
FIG. 3. Time courses in changes of body temperature and heart rate of a MT-injected rat are shown in A and B. Mean time courses in changes of body temperature (C) and heart rate (D) of MT-injected rats (closed circles and squares) and those of control (vehicle injected) animals (open circles and squares). Body temperature (C) and heart rate (D) show the highest (h, circles) and the lowest (1, squares) values in each day. Number of animals shown in parenthesis. Mean __+SEM. *p < 0.05, **p < 0.01, compared to controls; Student's t-test and Mann-Whitney U-test.
36
T A N A K A ET AL.
other hand, weight of the thymus on day 28 of MT-injected animals was significantly lower than controls (MT-injected, 251 +_ 20.2 mg/ rat, n = 10; control, 486 ___ 8.1 mg/rat, n = 5).
A 40
-g v
Changes in Plasma Noradrenalin 30
The basal level measured in the morning was 179.5 _ 21.4 pg/ml (n = 13). In MT-injected rats, compared to that in control animals, noradrenalin increased significantly on day 4 (MT-injected, 332.3 - 69.5 pg/ml, n = 7; control, 120.3 --- 15.8 pg/ml, n = 7) and on day 7 (MT-injected, 402 _ 88.7 pg/ml, n = 7; control, 202.7 ___ 31.5 pg/ml, n = 6). Increases on days 4, 7, and 21 (n = 7) in MT-injected animals were significant compared to their basal level of noradrenalin, while in control animals no significant change was observed at all measurements (Fig. 5). Plasma noradrenalin level has been known to reflect the sympathetic nerve activity [5]. Thus, its increase in MT-injected animals observed in our studies indicates augmented sympathetic nerve activity occurred in two phases.
20
-r I--o < lo o
B
pre
14
21
20
v O.. > .<
lO
DISCUSSION
pre C
4
7
14
21
800 700-
*#
*#
I
~¢~600Q)
500400-
Q)
-~ 3oo._8 200I:::
8 lOOo-
(8) (9)~ days
pre
4
7
14
21
after injection
FIG. 4. Changes in plasma hormone levels (measured in the morning) of MT-injected (solid bars) and vehicle injected (open bars) animals. (A) ACTH, (B) vasopressin (AVP), (C) corticosterone. Mean ___ SEM. *Compared to controls; #compared to basal values; *,#p < 0.05; ##p < 0.001; Student's t-test and Mann-Whitney U-test. Number of animals in parentheses. Note two phases of changes after MT injection in ACTH and corticosterone.
135.5, n = 8, control, 31.8 +_ 6 ng/ml, n = 9). Although in MTinjected rats corticosterone levels decreased on days 7 and 14, reaching close to the basal level, the values were still significantly higher compared to those of control animals. These data also show that changes in A C T H and corticosterone levels in the MTinjected rats occur in two phases. It was reported that although the responses in the HPA axis to MT injection were weak or absent in Lewis rats during the disease process of the adjuvant arthritis [15-17], the activation of the HPA axis was clearly observed in our experiments. To further confirm the activation of the HPA axis, we measured the weight of the adrenal glands and the thymus. Weight of the bilateral adrenal glands of MT-injected rats measured on day 28 was significantly higher than that of control animals (MT-injected, 74.4 __+ 3.3 mg/rat, n = 10; control, 51.3 + 2.1 mg/rat, n = 5). On the
This is the first report in which sickness behaviors, activation of the HPA axis, and the sympathetic nerve activity were measured in Lewis rats during the disease process of the adjuvant arthritis. We chose to use Lewis rats because they are inbred and are more suitable for evaluation of the immune system response than outbred strain such as SD rats. It has been reported that susceptivity of Lewis rats to streptococcus cell wall (SCW)-induced arthritis was associated with subnormal responsiveness of their HPA axis to SCW antigen and inflanunatory cytokines when compared to histocompatible, arthritis-resistant Fisher rats [15-17]. In our study, however, activation of the HPA axis was clearly observed following MT injection, although we did not try to make a comparison between the two strains. Sickness behaviors (fever, pain, loss of body weight, reduced water and food intakes, and poor physical activity), which are known to be induced by administration of inflammatory cytokines [6-8,12,14,18,19], were also observed in Lewis rats in the adjuvant arthritis. At present, we have no direct evidence that such cytokines are responsible for these changes. One of interesting and important findings of this study is that there are two phases in responses to MT injection; the primary phase *#
50C
R~
4oc
~" 300 z 201] o_ 101? 0 pre
4
7
14
21
days after injection FIG. 5. Changes in morning plasma noradrenalin (NE) levels of MTinjected (solid bars) and control (vehicle-injected, open bars) animals. Mean +_ SEM. *Compared to control animals; #compared to basal values; *,#p < 0.05; ##p < 0.001 Student's t-test and Mann-Whitney Utest. Number of animals in parentheses. Note significant increase on day 4 as well as on day 7 after MT injection.
SICKNESS RESPONSES IN A D J U V A N T ARTHRITIC RATS
from day 0 to 5, during which arthritis is not developed; and the second phase, from day 10 to 21, when arthritis is fully developed. Various changes in behavioral, hormonal, and autonomic nervous systems occurred in two steps and in synchrony with each other, although not all of the chaJages followed exactly the same pattern. Possible causes of this biphasic changes were investigated in our preliminary studies in which the autoimmune reaction of local lymphnodes cells and spleen cells to mycobacterium tuberculosis were measured in vitro. We found that in the local lymphnodes the proliferative response reached its maximum on day 5 to 7, while in spleen cells the response reached its maximum on day 14 to 21 (Tanaka et al., unpublished observation). It is probable that such difference in timing of autoimmune reactions in the two organs may contribute to biphasic changes observed in our studies. Before we can find good explanation for this interesting phenomenon, it will be necessary to study the time course and levels of cytokines in various tissues during the disease process. Such work in now being conducted in our laboratory. Activation of the HPA axis was also reported in SpragueDawley (SD) rats with the adjuvant arthritis [13]. Their results show monophasic changes in the HPA axis, rather than biphasic changes, which were obse,rved in Lewis rats. The reason for this difference is not clear, except the differences in animal strains. As mentioned above, not all changes followed exactly the same time course. For example, there was the time delay between changes in plasma A C T H and corticosterone levels. This might be due to negative feedback from corticosterone on the hypothalamo-pituitary system, or to reduction in sensitivity of corticosterone secretory cells in the adrenal cortex by the M T injection. Another observation is that although biphasic increases in plasma noradrenalin levels occur after MT injection, a large increase on day 7 does not exactly follow the pattern of A C T H changes. This may be due to markedly decreased corticosterone levels from day 4 to 7. It is known that endogenous glucocorticoids inhibit catecholamine synthesis and release [4]. We found no change in plasma AVP levels in the arthritis, although the basal level of AVP was high (12.9 -+ 3.8 pg/ml) in our experimental animals. Patchev et al. have shown that in untreated Lewis rats the plasma AVP levels is high and that this is due to compensatory responses to insufficient CRH production in this strain [9]. The increased plasma AVP was shown to contribute to the enhanced susceptivity in Lewis rats to inflammatory disease [10]. We evaluated the role of an elevated plasma AVP in development of the adjuvant arthritis in Lewis rats by giving AVP antagonist. Intraperitoneal administration of AVP antagonist attenuated the severity of the arthritis slightly, but significantly (Tanaka et al., unpublished observation). In summary, our results indicate that during the adjuvant arthritic in Lewis rats the sickness behaviors, activation of the HPA axis, and augmented sympathetic nerve activities occur in two phases; the first, before development of arthritis, and the second, with arthritis. The adjuvant arthritis in inbred Lewis rats is proved to be a useful animal model for studying neuroimmunomodulation. Use of this experimental animal model will enable us to obtain information on transmitting factors and pathways between the immune system and the CNS in the future. It is hoped that these studies will increase our understanding of mechanisms and treatment of autoimrnune diseases, such as rheumatoid arthritis. ACKNOWLEDGEMENTS We would like to express, our appreciation to Mimbishi Petrochemical Company for their generous gift of AVP and ACTH ILIA kits. This study was performed partly through Special Coordination Funds of Sciences and Technology Agency of Japanese Government.
37
REFERENCES 1. Connoily, K. M.; Stecher, V. J.; Kent, L. Examination of interleukin1 activity, the acute phase response, and leukocyte subpopulations in rats with adjuvant-induced arthritis. J. Lab. Clin. Med. 111:341 347; 1988. 2. Goldstein, D.; Stull, R.; Zimlichman, R.; Levinson, P.; Smith, H.; Keiser, H. Simultaneous measurement of DOPA, DOPAC, and catecholamines in plasma by liquid chromatography with electrochemical detection. Clin. Chem. 30:815-816; 1984. 3. Hogervorst, E. J. M.; Wagenaar, J. P. A.; Boog, C. J. P.; van der Zee, R.; van Embden, J. D. A.; van Eden, W. Adjuvant arthritis and immunity to the mycobacterial 65 kDa heat shock protein. Int. Immunol. 4:719-727; 1992. 4. Kvetnansky, R.; Fnkuhara, K.; Pacak, K.; Cizza, G.; Goldstein, D. S.; Kopin, I. J. Endogenous glucocorticoids restrain catecholamine synthesis and release at rest and during immobilization stress in rats. Endocrinology 133:1411 - 1419; 1993. 5. Micalizzi, E. R.; Pals, D. T. Evaluation of plasma norepinephrine as an index of sympathetic neuron function in the conscious, unrestrained rat. Life Sci. 24:2071-2076; 1979. 6. Morimoto, K.; Morimoto, A.; Nakamori, T.; Tan, N.; Minagawa, T.; Murakami, N. Cardiovascular responses induced in free-moving rats by immune cytokines. J. Physiol. 448:307-320; 1992. 7. Murakami, N.; Sakata, Y.; Watanabe, T. Central action sites of interleukin-lfl for inducing fever in rabbits. J. Physiol. 428:299-312; 1990. 8. Osaka, T.; Kannan, H.; Kawano, S.; Ueta, Y.; Yamashita, H. Intraperitoneal administration of recombinant human interleukin-1/3 inhibits osmotic thirst in the rat. Physiol. Behav. 51:1267-1270; 1992. 9. Patchev, V. K.; Kalogeras, K. T.; Zelazowski, P.; Wilder, R. L.; Chrousos, G. P. Increased plasma concentrations, hypothalamic content, and in vitro release of arginine vasopressin in inflammatory disease-prone, hypothalamic corticotropin-releasing hormone-deficient Lewis rats. Endocrinology 131:1453-1457; 1992. 10. Patchev, V. K.; Mastorakos, G.; Brady, L. S.; Redwine, J.; Wilder, R. L.; Chrousos, G. P. Increased arginine vasopressin secretion may participate in the enhanced susceptibility of Lewis rats to inflammatory disease. Neuroendocrinology 58:106-110; 1993. 11. Pearson, C. M. Experimental models in rheumatoid disease. Arthritis Rheum. 7:80-86; 1964. 12. Plata-Salaman, C. R.; Oomura, Y.; Kai, Y. Tumor necrosis factor and intedeukin-1/3 suppression of food intake by direct action in the central nervous system. Brain Res. 1988:106-114; 1988. 13. Sarlis, N. L; Chowdrey, H. S.; Stephanou, A.; Lightman, S. L. Chronic activation of the hypothalamo-pituitary-adrenal axis and loss of circadian rhythm during adjuvant arthritis in the rats. Endocrinology 130:1775-1779; 1992. 14. Shimomura, Y.; Shimizu, H.; Takahashi, M.; Uehara, Y.; Negishi, M.; Sato, N.; Inukai, T.; Kobayashi, I.; Kobayashi, S. Effects of peripheral administration of recombinant human interleukin-1 on feeding behavior of the rat. Life Sci. 47:2185-2192; 1990. 15. Sternberg, E. M.; Chrousos, G. P.; Wilder, R. L.; Gold, P. W. The stress response and the regulation of inflammatory disease. Ann. Intern. Med. 117:854-866; 1992. 16. Sternberg, E. M.; Hill, J. M.; Chrousos, G. P.; Kamilaris, T.; Listwak, S. J.; Gold, P. W.; Wilder, R. Inflammatory mediator-induced hypothalamic-pituitary-adrenai axis activation is defective in streptococcal cell wail arthritis-susceptible Lewis rats. Proc. Natl. Acad. Sci. USA 86:2374-2378; 1989. 17. Sternberg, E. M.; Young, W. S, III; Bernardini, R.; Calogero, A. E.; Chrousos, G. P.; Gold, P. W.; Wilder, R. L. A central nervous system defect in biosynthesis of corticotropin-releasing hormone is associated with susceptibility to streptococcal cell wall-induced arthritis in Lewis rats. Proc. Natl. Acad. Sci. USA 86:4771-4775; 1989. 18. Tazi, A.; Dantzer, R.; Crestani, F.; Moal, M. L. Interleukin-1 induces conditioned taste aversion in rats: A possible explanation for its pituitary-adrenal stimulating activity. Brain Res. 473:369-371; 1988. 19. Uehara, A.; Ishikawa, Y.; Okumura, T.; Okamura, K.; Sekiya, S.; Takasugi, Y.; Namiki, M. Indomethacin blocks the anorexic action of intedeukin-1. Eur. J. Pharmacol. 170:257-260; 1989.
Copyright© 1995ElsevierScienceInc. Printedin the USA.All rightsreserved 0361-9230/96$15.00 + .00 ELSEVIER
0361-9230(95)02037-3
Biphasic Changes in Behavioral, Endocrine, and Sympathetic Systems in Adjuvant Arthritis in Lewis Rats HIROAKI TANAKA,.1 YOICHI UETA,* UKI YAMASHITA,1- HIROSHI KANNANt AND HIROSHI YAMASHITA*
*Department of Physiology, University of Occupational and Environmental Health, Yahatanishiku, Kitakyushu, 807, Japan tDepartment of Immunology and Parasitology, Hiroshima University School of Medicine, Japan tDepartment of Physiology, Miyazaki Medical College, Japan [Received 31 May 1994; Accepted 9 May 1995] thritis was 60 and 80% in Fisher rats and Sprague-Dawley (SD) rats, respectively, in our unpublished observation. The severity of the adjuvant arthritis, expressed as the mean arthritis index [3] was 12.3 _ 0.4, 6.6 _ 2.0, and 2.0 _+ 1.1 for Lewis, SD, and Fisher rats, respectively (our unpublished observation). In the streptococcus cell wall (SCW) induced arthritis differences in susceptivity between Lewis and Fisher rats were found to be associated with lower responsiveness of the HPA axis in Lewis rats to the arthritis inducible antigen, SCW, and/or to inflammatory cytokines, compared to that in Fisher rats [16,17]. In these studies, however, responses of the HPA axis were investigated only during the first four hours following the SCW injections. In most other works outbred SD rats were used for studies of the HPA axis during experimental arthritis [13], because they show strong activation of the HPA axis and intermediate susceptivity to the arthritis. Although hormonal changes are known to occur in the experimental arthritis in outbred rats, very little study has been done in inbred strain, such as Lewis rats. It is likely that biological substances produced by activated macrophages and T lymphocytes [1], such as immune cytokines, cause these changes during disease process of arthritis, yet we do not have any concrete evidence. To study relationships between the central nervous system (CNS), endocrine, and immune reactions in the adjuvant arthritis, it is preferable to use an inbred strain, such as Lewis rats. Our aim of this article is to investigate in Lewis rats the CNS responses (body temperature, ingestive, and motor behaviors), activation of the HPA axis (plasma levels of ACTH, corticosterone, and vasopressin), and changes in the sympathetic nerve activities and heart rate throughout development of the adjuvant arthritis. The results suggest that Lewis rats are suitable experimental model for the future studies of neuroimmunomodulation.
ABSTRACT: Adjuvant arthritis (AA) is an experimental model for rheumatoid arthritis, and ia induced most easily in inbred Lewis rats by an intradermal injection of hest-killed Mycobacterium tuberculosis(MT) in incomplete Freund's adjuvant. Susceptivity to the arthritis in Lewis ral~ is thought to be related to a defect in their responses of the hypethalamo-pituitary-adrenal (HPA) axis to the disease. Because the use of an inbred strain is neeessary for our immunological studies, we examined in Lewis rats changes in behavior, the HPA axis, and sympathetic nerve activities during developmer,t of the adjuvant arthritis. Following intradermal injections of heet-killed MT in adjuvant, the arthritis began to develop on day 112, reaching its maximum severity on day 21, and remained at the level for over a month. The body temperature rose from day 0 to 5 (the primary phaseBbeforo the onset of the arthritis). It then tell to normal temperature, and again rose from day 10 to 21 (the secondary phase--with fully developed arlhritis). The behavioral (physical activity, food, and water intake) and hormor,al parameters [plasma adrenocorticotropic hormone (ACTtl) and corticosterone levels] also changed in two phases, similar to those observed in the tamperature responses. No change in plasma vasopressin level was observed. SympetheUc nerve activiUse, assessed by changes in plasma noradrenalin levels;, increased more in the primary than in the secondary phase. 111e possible causes for the biphasic changes associated with development of arthritis are discussed. KEY WORDS: Adjuvant arthritis, Lewis rat, Ingestive behavior, Physical activity, Body temperature, Heart rate, Hypothalamopituitary-adrenal axis, Sympathetic nerve activity.
INTRODUCTION Adjuvant arthritis (AA) is a good experimental model for rheumatoid arthritis, one of the autoimmune diseases. The arthritis is induced in rats by an intradermal injection of heat-killed Mycobacterium tuberculosis (MT) in incomplete Freund's adjuvant. Susceptivity to the arthritis, severity, and incidence of the arthritis followed by the MT injection vary among different strains of rats. The arthritis is induced most readily in Lewis rats; the incidence of the arthritis was found to be 100% followed by the MT injection in our studies. On the other hand, the incidence of the ar-
METHOD
Animals Male Lewis (LEW/N) rats 6 - 1 0 weeks of age (body weight: 180-250 g) were used and kept under constant room temperature (22-25°(2), humidity (40-60%), and a 12-h light-dark cycle
To whom requests for reprints should be addressed. 33
34
(lights on: 0730 to 1930 h). Standard laboratory rat chow or pellets and water were given ad lib.
Induction of Arthritis Animals were injected with 1 mg heat-killed Mycobacterium tuberculosis H37 Ra in incomplete Freund's adjuvant (Difco Laboratories, Detroit, MI) intradermally at the base of their tails, as described elsewhere [11]. Control animals were injected with incomplete Freund's adjuvant alone.
Assessment of Arthritis The arthritis index of the animals were scored by grading each paw from 0 to 4, based on erythema, swelling, and deformity of the joints ( 0 - - n o erythema or swelling; 1--slight erythema and swelling of one of the toes; 2--erythema and swelling of more than one toe; 3--erythema and swelling of the ankle; 4 - - c o m plete erythema and swelling of toes and ankle and incapacity to bend the ankle; the highest score could reach 16, because each limb is evaluated separately) [3]. The hind paw volume was measured on day 21 and 28 after MT injection by a plethysmograph.
Measurement of Behavioral Changes Physical activity of the animal was recorded with an automatic behavioral measurement system (Astec, Fukuoka, Japan). In brief, the physical activity was counted when an animal crossed any of four infrared beams fixed at 4 cm above the cage floor (floor width: 30 × 30 cm, height: 43 cm) and stored every 30 rain in a computer. Feeding behavior was assessed by weighing standard rat chow every 24 h with a conventional metabolism cage, or by the number of pellets counted with an automatic pellet feeder. Drinking behavior was also assessed by weighing a bottle every 24 h in a conventional metabolism cage, or by using a counting devise to measure the number of water drops drunk. Body temperature and heart rate were recorded with a telemetry system (Data Science, Inc., St. Paul, MN) through a transmitter implanted in the animal intraperitoneally. The transmitter was implanted under general anesthesia (sodium pentobarbitone, 50 mg/kg IP) at least 14 days before the adjuvant injection. The signals of body temperature and heart rate from the transmitter were recorded as averaged values for 5 s every 30 rain through three receivers around the cage and were stored for later analysis by the computer.
Measurement of Hypothalamo--Pitaitary-Adrenal (HPA) Axis Activation Animals were handled every day for 7 days for them to be accustomed to the later experimental procedures. They were decapitated, and trunk blood was collected in an Ethylenediaminetetraacetic Acid Disodium Salt Dihydrate (EDTA-2Na)-containing tube within 3 h from the lights-on time on days 4, 7, 14, and 21 following the MT injection (day 0). The blood samples were centrifuged (3000 rpm) at 4°C for 15 rain and the plasma were aliquoted and stored at -80°C until the assay. Plasma ACTH, corticosterone, and vasopressin (AVP) were measured by radioimmunoassay using RIA kits (Mitsubishi Petrochemical, Japan, for ACTH and AVP; ICN Biomedical, USA, for corticosterone). The thymus and adrenal glands were removed from the animals on day 28 and weighed. All hormones were measured in the same group of animals.
Measurement of Plasma Noradrenalin Level Blood samples (0.3 ml) were collected before and on days 4, 7, 14, and 21 after the MT injection in an EDTA-2Na-containing
TANAKA ET AL.
tube via a venous catheter implanted into the superior vena cava through the jugular vein 7 days before the injection. The plasma were aliquoted and stored by the same method as did for the hormone assays. Plasma noradrenalin was assayed using liquid chromatography with electrochemical detection after batch alumina extraction, as described elsewhere [2].
Stastical Analysis Results are presented as means ± SEM. For statistical evaluation, Student's t-test and Mann-Whitney U-test were used. A value of p < 0.05 was considered to be significant. RESULTS
Time Course of Development of the Arthritis The arthritis suddenly developed approximately on day 13 following the MT injection (day 0), and reached its severest level around on day 21. At this time the averaged arthritis index was 12.32 ___ 0.42 (n = 31). This condition continued for several months without much change. Incidence of arthritis reached 100% after day 15 (Fig. 1A). The paw volume, the quantitative assessment of the arthritis, increased significantly when measured on days 21 and 28, compared to that of control animals (day 21, MT-injected rats, 1.88 _ 0.08 ml, n = 8; control rats, 1.34 ± 0.001 ml, n = 3; day 28, MT-injected rats, 2.15 ± 0.05 ml, n = 10; control rats, 1.63 ± 0.05 ml, n = 20) (Fig. 1B).
Changes in Body Weight, Food and Water Intake, Physical Activity, Heart Rate, and Body Temperature Body weight changes were expressed as a percentage of the body weight on day 0 and measured daily until day 21. Body weight of MT-injected rats (n = 19) increased linearly to day 10, but the weight gain stopped from day 10 to 21. In control (vehicle-injected) animals (n = 12) the body weight linearly increased to day 21. There was a small, yet significant, transient decrease in body weight of MT-injected rats on days 4 and 5. Daily food and water intakes measured from day 0 to 21 a r e shown in Fig. 2B and C. In MT-injected rats (n = 10), up to day 10 the amount of food and water intake was similar to that in control animals (n = 5), but decreased markedly from day 12 to 21. In addition, on days 1 and 3 food and water intakes were slightly decreased. The ingestive behavior of all animals showed
A
100
3
15
8o x m ~
40 Q.
o
"0
........
~b-
z5
days after MT injection -
30
21
28
days after injection
FIG. 1. Time courses in developmentof adjuvant arthritis in Lewis rats following injection of Mycobacterium tuberculosis (MT). (A) The arthritis index (closed circles) and arthritis incidence (open circles). (B) Changes in paw volume in arthritic (solid bars) and con~ol (open bars) animals on days 21 and 28. *p < 0.005, **p < 0.0005, compared to conU'ol,Student's t-test and Mann-Whitney U-test. Values are mean +_ SEM; numbers in parentheses are number of animals used.
SICKNESS RESPONSES IN ADJUVANT ARTHRITIC RATS 180
A --
_~2 ~ ~ ( 1 2 )
160 t 140 ,,o
.........
30
B~
: LZ_g.:"-'r I.~=. I (19)
~
O .....
6
' lb ..............
Z lO)
t~
1
[
3 5
c
.
.
.
.
.
.
.
.
b .....
.
s
'
~ b
~
**
J
..............
E -~
(5) 25, 10)
=
15
D ~
.,
.,.
~92°°i
,.
,.
.,.,
.
. ,.
.,
.,.
,.,.,.,.,
T
~'4400t
_ . . . , . ~ 4 " ~ ~ ~
48001 -10-8 .
,
(6, 05)
.
,
.
.
.
.
.
.
.
-6 -4 -2
.
.
,
0
2
4
.
,
. . . . .
6
8
,
.
,
,
,
.
,
.
,
,
10 12 14 16 18 20 22 24
days after injection
35
changes. The basal level of the morning plasma ACTH was 13.93 - 2.13 pg/ml (n = 7). In MT-injected rats A C I ~ markedly increased on day 4 (32.36 _ 3.29 pg/ml, n = 9), returned to the basal level on day 7 (17.19 _+ 2.9 pg/ml, n = 7). It rose again on day 14 (33.45 __+3.9 pg/ml, n = 13) and day 21 (25.18 ___3.25 pg/ml, n = 9 (Fig. 4A). In control rats (vehicle-injected) ACTH remained at the basal level on days 4 and 7, and rose slightly on days 14 and 21. The significant difference on ACTH levels between MT-injected and control animals occurred on day 4. An increase in plasma ACTH in control (vehicle-injected) animals on days 14 and 21, though at a lesser degree, is puzzling. It may be due to nonspecific responses to injection of the vehicle, incomplete Freund's adjuvant. To test reliability of our measurements, the level of plasma ACTH was determined after giving restrained stress for 20 min in normal rats. The value was 273.2 ___ 20.6 pg/ml (n = 7), similar to that reported by others [15]. The basal level of AVP measured in the morning was 12.9 ___ 3.8 pg/ml (n = 5), and this value did not change at all on any days tested in both the MT-injected and in control animals (Fig. 413). Marked changes in plasma corticosterone levels appeared in MT-injected rats (Fig. 4C). Its basal level was 42 _+ 17.8 ng/ml (n = 8), and increased dramatically on days 4, 7, 14, and 21. The differences between corticosterone levels of MT-injected and control rats were significant at all measurements (day 4 - - M T injected, 477.6 ___ 111.5 ng/ml, n = 8, control, 25.9 ___3.9 ng/ml, n = 9; day 7 - - MT-injected, 204.4 _ 52.1, n = 8, control, 23 - 6.2 ng/ml, n = 9; day 14--MT-injected, 166.2 _ 32, n = 9, control, 46.7 _ 18.8, n = 9; day 21--MT-injected, 540.9 _
FIG. 2. Time courses in chmages of body weight (A), food intake (B), water intake (C), and physical activity (D) of MT-injected rats (closed circles and squares) and those of control (vehicle injected) animals (open circles and squares). Body weight is expressed as percent of that on day 0 (A). Number of animals shown in parenthesis. Mean _+ SEM. *p < 0.05, **p < 0.01, compared to controls; Student's t-test and MannWhitney U-test. a clear circadian rhythm before and after the injection and throughout the period of the experiments (data not shown). The physical activity of MT-injected rats (n = 15) decreased significantly on day 0, and again on days 2 to 4, compared to that of controls (n = 6). It recovered on day 5 to 9, then steadily decreased again after day 10 (Fig. 2D). Before MT injections the body temperature of all animals showed a clear circadian rhythm, but this tended to disappear after day 10 when the temperature began to rise considerably with development of arthritis (Fig. 3A). After MT injections there was a transient rise in averaged body temperature (n = 6) on day 0, returning to normal immediately. The temperature rose again from day 12 to 21, while no change occurred in vehicle-injected rats (n = 4). There was a moderate increase from day 3 to 5 (Fig. 3C). The heart rate of MT-injected rats showed the same circadian rhythm as that in normal rats (Fig. 3B). The averaged heart rate increased from day 14 to 21 after MT injections (n = 6), compared to that of control animals (n = 4) (Fig. 3D). Thus, changes in ingestive behavior, physical activity, and body temperature in MT-injected rats occurred in two steps and can be divided into two phases; the primary phase, from day 0 to 5, during which no arthritis is present, and the secondary phase, from day 10 to 21, after the arthritis is developed.
Changes in Plasma Leveb: of ACTH, A VP, and Corticosterone, and in Weight of Adrenal Gland and Thymus Changes in the HPA axis were studied on days 4, 7, 14, and 21 after MT injection, when the behavioral responses showed marked
39
-4
-2
0
2
4
•6
8
10
12
14
16
18
20
-4
-2
0
2
4
6
8
10
12
14
16
18
20
400
4O
c
1
""'" I
"t
-I0 "-8 '-6 "-4 "-2 ' 6 "2 "'~ "6 "8 'I0 "12 "I~I"16 "18 "20 "22 "24
D :°°l 300 4
_
_r-* . 2 . _ ~
_ . . k ~ z ~ - i (6)
2001'0"-8 "'6 '-'4 '-2 " 6 "2 "4 " 6 ' 8 "1() "12'14. " 1 6 " 1 8 ' 2 0 " 2 2 " 2 ~ I days after injection
FIG. 3. Time courses in changes of body temperature and heart rate of a MT-injected rat are shown in A and B. Mean time courses in changes of body temperature (C) and heart rate (D) of MT-injected rats (closed circles and squares) and those of control (vehicle injected) animals (open circles and squares). Body temperature (C) and heart rate (D) show the highest (h, circles) and the lowest (1, squares) values in each day. Number of animals shown in parenthesis. Mean __+SEM. *p < 0.05, **p < 0.01, compared to controls; Student's t-test and Mann-Whitney U-test.
36
T A N A K A ET AL.
other hand, weight of the thymus on day 28 of MT-injected animals was significantly lower than controls (MT-injected, 251 +_ 20.2 mg/ rat, n = 10; control, 486 ___ 8.1 mg/rat, n = 5).
A 40
-g v
Changes in Plasma Noradrenalin 30
The basal level measured in the morning was 179.5 _ 21.4 pg/ml (n = 13). In MT-injected rats, compared to that in control animals, noradrenalin increased significantly on day 4 (MT-injected, 332.3 - 69.5 pg/ml, n = 7; control, 120.3 --- 15.8 pg/ml, n = 7) and on day 7 (MT-injected, 402 _ 88.7 pg/ml, n = 7; control, 202.7 ___ 31.5 pg/ml, n = 6). Increases on days 4, 7, and 21 (n = 7) in MT-injected animals were significant compared to their basal level of noradrenalin, while in control animals no significant change was observed at all measurements (Fig. 5). Plasma noradrenalin level has been known to reflect the sympathetic nerve activity [5]. Thus, its increase in MT-injected animals observed in our studies indicates augmented sympathetic nerve activity occurred in two phases.
20
-r I--o < lo o
B
pre
14
21
20
v O.. > .<
lO
DISCUSSION
pre C
4
7
14
21
800 700-
*#
*#
I
~¢~600Q)
500400-
Q)
-~ 3oo._8 200I:::
8 lOOo-
(8) (9)~ days
pre
4
7
14
21
after injection
FIG. 4. Changes in plasma hormone levels (measured in the morning) of MT-injected (solid bars) and vehicle injected (open bars) animals. (A) ACTH, (B) vasopressin (AVP), (C) corticosterone. Mean ___ SEM. *Compared to controls; #compared to basal values; *,#p < 0.05; ##p < 0.001; Student's t-test and Mann-Whitney U-test. Number of animals in parentheses. Note two phases of changes after MT injection in ACTH and corticosterone.
135.5, n = 8, control, 31.8 +_ 6 ng/ml, n = 9). Although in MTinjected rats corticosterone levels decreased on days 7 and 14, reaching close to the basal level, the values were still significantly higher compared to those of control animals. These data also show that changes in A C T H and corticosterone levels in the MTinjected rats occur in two phases. It was reported that although the responses in the HPA axis to MT injection were weak or absent in Lewis rats during the disease process of the adjuvant arthritis [15-17], the activation of the HPA axis was clearly observed in our experiments. To further confirm the activation of the HPA axis, we measured the weight of the adrenal glands and the thymus. Weight of the bilateral adrenal glands of MT-injected rats measured on day 28 was significantly higher than that of control animals (MT-injected, 74.4 __+ 3.3 mg/rat, n = 10; control, 51.3 + 2.1 mg/rat, n = 5). On the
This is the first report in which sickness behaviors, activation of the HPA axis, and the sympathetic nerve activity were measured in Lewis rats during the disease process of the adjuvant arthritis. We chose to use Lewis rats because they are inbred and are more suitable for evaluation of the immune system response than outbred strain such as SD rats. It has been reported that susceptivity of Lewis rats to streptococcus cell wall (SCW)-induced arthritis was associated with subnormal responsiveness of their HPA axis to SCW antigen and inflanunatory cytokines when compared to histocompatible, arthritis-resistant Fisher rats [15-17]. In our study, however, activation of the HPA axis was clearly observed following MT injection, although we did not try to make a comparison between the two strains. Sickness behaviors (fever, pain, loss of body weight, reduced water and food intakes, and poor physical activity), which are known to be induced by administration of inflammatory cytokines [6-8,12,14,18,19], were also observed in Lewis rats in the adjuvant arthritis. At present, we have no direct evidence that such cytokines are responsible for these changes. One of interesting and important findings of this study is that there are two phases in responses to MT injection; the primary phase *#
50C
R~
4oc
~" 300 z 201] o_ 101? 0 pre
4
7
14
21
days after injection FIG. 5. Changes in morning plasma noradrenalin (NE) levels of MTinjected (solid bars) and control (vehicle-injected, open bars) animals. Mean +_ SEM. *Compared to control animals; #compared to basal values; *,#p < 0.05; ##p < 0.001 Student's t-test and Mann-Whitney Utest. Number of animals in parentheses. Note significant increase on day 4 as well as on day 7 after MT injection.
SICKNESS RESPONSES IN A D J U V A N T ARTHRITIC RATS
from day 0 to 5, during which arthritis is not developed; and the second phase, from day 10 to 21, when arthritis is fully developed. Various changes in behavioral, hormonal, and autonomic nervous systems occurred in two steps and in synchrony with each other, although not all of the chaJages followed exactly the same pattern. Possible causes of this biphasic changes were investigated in our preliminary studies in which the autoimmune reaction of local lymphnodes cells and spleen cells to mycobacterium tuberculosis were measured in vitro. We found that in the local lymphnodes the proliferative response reached its maximum on day 5 to 7, while in spleen cells the response reached its maximum on day 14 to 21 (Tanaka et al., unpublished observation). It is probable that such difference in timing of autoimmune reactions in the two organs may contribute to biphasic changes observed in our studies. Before we can find good explanation for this interesting phenomenon, it will be necessary to study the time course and levels of cytokines in various tissues during the disease process. Such work in now being conducted in our laboratory. Activation of the HPA axis was also reported in SpragueDawley (SD) rats with the adjuvant arthritis [13]. Their results show monophasic changes in the HPA axis, rather than biphasic changes, which were obse,rved in Lewis rats. The reason for this difference is not clear, except the differences in animal strains. As mentioned above, not all changes followed exactly the same time course. For example, there was the time delay between changes in plasma A C T H and corticosterone levels. This might be due to negative feedback from corticosterone on the hypothalamo-pituitary system, or to reduction in sensitivity of corticosterone secretory cells in the adrenal cortex by the M T injection. Another observation is that although biphasic increases in plasma noradrenalin levels occur after MT injection, a large increase on day 7 does not exactly follow the pattern of A C T H changes. This may be due to markedly decreased corticosterone levels from day 4 to 7. It is known that endogenous glucocorticoids inhibit catecholamine synthesis and release [4]. We found no change in plasma AVP levels in the arthritis, although the basal level of AVP was high (12.9 -+ 3.8 pg/ml) in our experimental animals. Patchev et al. have shown that in untreated Lewis rats the plasma AVP levels is high and that this is due to compensatory responses to insufficient CRH production in this strain [9]. The increased plasma AVP was shown to contribute to the enhanced susceptivity in Lewis rats to inflammatory disease [10]. We evaluated the role of an elevated plasma AVP in development of the adjuvant arthritis in Lewis rats by giving AVP antagonist. Intraperitoneal administration of AVP antagonist attenuated the severity of the arthritis slightly, but significantly (Tanaka et al., unpublished observation). In summary, our results indicate that during the adjuvant arthritic in Lewis rats the sickness behaviors, activation of the HPA axis, and augmented sympathetic nerve activities occur in two phases; the first, before development of arthritis, and the second, with arthritis. The adjuvant arthritis in inbred Lewis rats is proved to be a useful animal model for studying neuroimmunomodulation. Use of this experimental animal model will enable us to obtain information on transmitting factors and pathways between the immune system and the CNS in the future. It is hoped that these studies will increase our understanding of mechanisms and treatment of autoimrnune diseases, such as rheumatoid arthritis. ACKNOWLEDGEMENTS We would like to express, our appreciation to Mimbishi Petrochemical Company for their generous gift of AVP and ACTH ILIA kits. This study was performed partly through Special Coordination Funds of Sciences and Technology Agency of Japanese Government.
37
REFERENCES 1. Connoily, K. M.; Stecher, V. J.; Kent, L. Examination of interleukin1 activity, the acute phase response, and leukocyte subpopulations in rats with adjuvant-induced arthritis. J. Lab. Clin. Med. 111:341 347; 1988. 2. Goldstein, D.; Stull, R.; Zimlichman, R.; Levinson, P.; Smith, H.; Keiser, H. Simultaneous measurement of DOPA, DOPAC, and catecholamines in plasma by liquid chromatography with electrochemical detection. Clin. Chem. 30:815-816; 1984. 3. Hogervorst, E. J. M.; Wagenaar, J. P. A.; Boog, C. J. P.; van der Zee, R.; van Embden, J. D. A.; van Eden, W. Adjuvant arthritis and immunity to the mycobacterial 65 kDa heat shock protein. Int. Immunol. 4:719-727; 1992. 4. Kvetnansky, R.; Fnkuhara, K.; Pacak, K.; Cizza, G.; Goldstein, D. S.; Kopin, I. J. Endogenous glucocorticoids restrain catecholamine synthesis and release at rest and during immobilization stress in rats. Endocrinology 133:1411 - 1419; 1993. 5. Micalizzi, E. R.; Pals, D. T. Evaluation of plasma norepinephrine as an index of sympathetic neuron function in the conscious, unrestrained rat. Life Sci. 24:2071-2076; 1979. 6. Morimoto, K.; Morimoto, A.; Nakamori, T.; Tan, N.; Minagawa, T.; Murakami, N. Cardiovascular responses induced in free-moving rats by immune cytokines. J. Physiol. 448:307-320; 1992. 7. Murakami, N.; Sakata, Y.; Watanabe, T. Central action sites of interleukin-lfl for inducing fever in rabbits. J. Physiol. 428:299-312; 1990. 8. Osaka, T.; Kannan, H.; Kawano, S.; Ueta, Y.; Yamashita, H. Intraperitoneal administration of recombinant human interleukin-1/3 inhibits osmotic thirst in the rat. Physiol. Behav. 51:1267-1270; 1992. 9. Patchev, V. K.; Kalogeras, K. T.; Zelazowski, P.; Wilder, R. L.; Chrousos, G. P. Increased plasma concentrations, hypothalamic content, and in vitro release of arginine vasopressin in inflammatory disease-prone, hypothalamic corticotropin-releasing hormone-deficient Lewis rats. Endocrinology 131:1453-1457; 1992. 10. Patchev, V. K.; Mastorakos, G.; Brady, L. S.; Redwine, J.; Wilder, R. L.; Chrousos, G. P. Increased arginine vasopressin secretion may participate in the enhanced susceptibility of Lewis rats to inflammatory disease. Neuroendocrinology 58:106-110; 1993. 11. Pearson, C. M. Experimental models in rheumatoid disease. Arthritis Rheum. 7:80-86; 1964. 12. Plata-Salaman, C. R.; Oomura, Y.; Kai, Y. Tumor necrosis factor and intedeukin-1/3 suppression of food intake by direct action in the central nervous system. Brain Res. 1988:106-114; 1988. 13. Sarlis, N. L; Chowdrey, H. S.; Stephanou, A.; Lightman, S. L. Chronic activation of the hypothalamo-pituitary-adrenal axis and loss of circadian rhythm during adjuvant arthritis in the rats. Endocrinology 130:1775-1779; 1992. 14. Shimomura, Y.; Shimizu, H.; Takahashi, M.; Uehara, Y.; Negishi, M.; Sato, N.; Inukai, T.; Kobayashi, I.; Kobayashi, S. Effects of peripheral administration of recombinant human interleukin-1 on feeding behavior of the rat. Life Sci. 47:2185-2192; 1990. 15. Sternberg, E. M.; Chrousos, G. P.; Wilder, R. L.; Gold, P. W. The stress response and the regulation of inflammatory disease. Ann. Intern. Med. 117:854-866; 1992. 16. Sternberg, E. M.; Hill, J. M.; Chrousos, G. P.; Kamilaris, T.; Listwak, S. J.; Gold, P. W.; Wilder, R. Inflammatory mediator-induced hypothalamic-pituitary-adrenai axis activation is defective in streptococcal cell wail arthritis-susceptible Lewis rats. Proc. Natl. Acad. Sci. USA 86:2374-2378; 1989. 17. Sternberg, E. M.; Young, W. S, III; Bernardini, R.; Calogero, A. E.; Chrousos, G. P.; Gold, P. W.; Wilder, R. L. A central nervous system defect in biosynthesis of corticotropin-releasing hormone is associated with susceptibility to streptococcal cell wall-induced arthritis in Lewis rats. Proc. Natl. Acad. Sci. USA 86:4771-4775; 1989. 18. Tazi, A.; Dantzer, R.; Crestani, F.; Moal, M. L. Interleukin-1 induces conditioned taste aversion in rats: A possible explanation for its pituitary-adrenal stimulating activity. Brain Res. 473:369-371; 1988. 19. Uehara, A.; Ishikawa, Y.; Okumura, T.; Okamura, K.; Sekiya, S.; Takasugi, Y.; Namiki, M. Indomethacin blocks the anorexic action of intedeukin-1. Eur. J. Pharmacol. 170:257-260; 1989.