Indomethacin prevents increased catecholamine turnover in rat brain following systemic endotoxin challenge

Prog. Neum-Psychophormocol. Printed in Great Britain.

& Biol. Psychiot.

1990. Vol. 14. pp. 603-621

0278-5846/90 $0.00 + .50 1990 Pergamon Press plc

INDOMETHACIN PREVENTS INCREASED CATECHOLAMINE TURNOVER IN RAT BRAIN FOLLOWING SYSTEMIC ENDOTOXIN CHALLENGE MONICA I. MASANAl, MELVYN P. HEYES and IVAN N. MEFFORDl 1Section on Clinical Pharmacology, Neuroscience Branch and 2Section on Analytical Biochemistry, Laboratoty of Clinical Science, National Institute of Mental Health, Bethesda, Maryland, U.S.A.

(Final form, May 1990)

Abstract Masana, Monica I., Melvyn P. Heyes and Ivan N. Mefford: Indomethacin Prevents Increased Catecholamine Turnover In Rat Brain Following Systemic Endotoxin Challenge. hog. NeuroPsychophsrmacol. & Biol. Psych. 1990, J& 609-621 1. 2.

::

5. 6.

7.

Key features of the acute phase response to infection are replicated by systemic administrations of lipopolysaccharide and may be mediated via the production of lymphokines and cytokines, including interleukin-1. Inhibition of prostaglandin synthesis may attenuate certain featums of the acute phase response. In the present study, the effects of systemic administration of the lipopolysaccharide (LPS, 250 pg/rat) and interleukin-1 (B-l.10 p&at) on catecholamine metabolism in different brain regions were compared and the effects of indomethacin, a cycloxygenase inhibitor was determined. The ratio of metabolite to parent amine was used as an index of turnover of catecholamines. In hypothalamus, both epinephrine and norepinephrine concentrations were decreased and their major metabolite, 3-methoxy,4-hydroxyphenylglycol (MHPG), was elevated at 4,8 and 24 hr following LPS. The major metabolite of dopamine (homovanillic acid, HVA) was increased at 8 hours in striatum, hypothalamus and medulla. LPS increased dopamine turnover at 8 and 24 hr and norepinephrine turnover at 4,8 and 24 hr. In all regions examined, IL-1 produced effects similar to LPS on amine and metabolite contents and norepinephtine and dopamine turnover. Significantly, co-administration of a single dose of indomethacin (50 mg/kg) completely blocked LPS-induced changes in hypothalamic catecholamines and metabolites and the increase in turnover at 4 and 8 hr. Furthermore, the effects of IL-1 on hypothalamic MHPG content and norepinephrine turnover were also blocked by indomethacin, although the effects of IL-1 on regional catecholamines and HVA content and turnover were either not modified or partially antagonized by indomethacin. The present results suggest that in the rat, activation of noradrenergic, dopaminergic and epinephrinecontaining neurons in hypothalamus, as well as dopaminergic neurons in other regions is associated with the acute phase response to endotoxin and that synthesis of prostaglandins plays a pivotal role in catecholamine responses in all brain regions examined.

Keywords: dopamine turnover, endotoxin, noradrenaline turnover, prostaglandins Abbreviations: adrenocorticotropic hormone (AClH); 3 ethoxy, 4-hydroxyphenylglycol (EHPG); homovanillic acid (HVA); interleukin-1 (IL-l), lipopolysaccharide (LPS); a-melanocyte-stimulating hormone (a-MSH); 3methoxy,4-hydroxyphenylglycol (MHPG); phenylethanolamine N-methyltransferase (PNMT); pokeweed mitogen (PWM); prostaglandin E2 (PGE2) Introduction Marked increases in catecholamine turnover in different brain regions follow exposure to a wide variety of physiologic stresses including swimming stress (Sudo, 1983), cold swimming stress (Roth et al., 1982) or ethanol (Mefford, 1987). Recent studies have also shown similar changes of catecholamine turnover in brain in response 609

M. I. Masana

610

et al.

to immuno1ogica.l challenge by endotoxin, the lipopolysaccharide outer membrane coat of gram negative bacteria (LPS). Increases in hypothalamic norepinephrine and dopamine turnover after administration of a LPS in the mouse have been qorted

(Mefford and Heyes, 1990). The stimulation of B lymphocytes by a variety of bacterial

cell wall antigens or by plant allergens leads to the production of IL-l, thought to be the active, host-produced mediator of a variety of immune responses (Dinarello, 1984). The report by Dunn (1988) of an increase in hypothalamic norepinephrine turnover in the mouse in response to IL-l, as reflected by an increase in 3 methoxy, 4-hydroxyphenylglycol (MHPG), the major metabolite of norepinephrine. and decreased norepinephrine content, is in accordance with this notion. A relatively large body of evidence supports the action of arachidonic acid cyclooxygenase

products as

mediators of some of the changes that a~ observed following either endotoxin or IL-1 administration (Dinarcllo and Wolff, 1982; Revhaug et al., 1988). Prostaglandin E2 (PGE2) has been found to be elevated in patients’ cerebrospinal fluid during fever as well as following a bacterial pyrogen administration (Harvey et al., 1975; Saxena et al., 1979). Ibuprofen, a cyclooxygenase inhibitor, attenuates some of the symptoms occurring after endotoxin administration, including fever, tachycardia, increase in metabolic rate and stimulation of stress hormone release, while some other symptoms are unaffected: leukocytosis, hypoferremia or elevation of the C-reactive protein

(Revhaug et al.. 1988). Cyclooxygenase

inhibitors do not, however, block the production of IL-1

following endotoxin nor other cell mediated events due to the actions of IL- 1 following immune stimulation (Hoe et al., 1972). To determine whether stimulation of acute phase response by endotoxin in rats would modify catecholamine turnover, the authors examined the effects of LPS and an “in vitro” B-cell mitogen, pokeweed mitogen (PWM), on catecholamine metabolism in different brain regions of the Sprague Dawley rat. Because some of the components of the acute phase response are thought to be mediated non-specifically by IL-1 via induction of arachidonic acid metabolism ( Bemheim et al., 1980, Coceani et al. 1983; Feldberg and Gupta, 1973), we also examined the effects of concomitant administration of the cyclooxygenase inhibitor, indomethacin, on LPS and IL-l-induced changes in catecholamine turnover in different rat brain regions.

In addition, because a variety of stressors have been

demonstrated to increase the activity of brainstem phenylethanolamine N-methyltransferase (PNMT), the enzyme which converts norepinephrine

to epinephrine (Saavedra and Torda, 1980; Turner et al., 1978), the authors

determined the activity of PNMT 24 hours following LPS administration. Materials and Methods Animals Male Sprague Dawley rats, 250-300 g, were used throughout. Animals were housed in groups of six on a 12: 12 hour 1ight:dark cycle, with food and water provided ad libitum. Drugs The following drugs were used: Indomethacin, LPS and PWM were obtained from Sigma Chemical Co, St. Louis, MO, U.S.A. IL-la, was a generous gift from Dr. Liberato, Hoffmann-LaRoche, NJ, U.S.A. All drugs were dissolved or suspended in 1 ml of sterile, pyrogen-free intraperitoneal (i.p.) injection.

saline and prepared immediately prior to

611

Brain catecholamine turnover after LPS challenge

Experimental w Animals were administered i.p. saline (1 ml), LPS (250 pg in 1 ml), PWM (5 mg in 1 ml), IL-l ( 10 pg in 1 ml), indomethacin (50 mg/kg in 1.0 ml), indomethacin + IL-l ( 50 mg/kg and 10 pg respectively in 1 ml), or indomethacin + LPS (50 mg/kg and 250 pg respectively in 1.0 ml). After 4, 8 or 24 hours, animals were decapitated and the brains removed and rapidly dissected. Brain tissues were stored at -80 oC until analyzed for catecholamines and metabolites, homovanillic acid (HVA) and MHPG. Tissues were weighed while frozen and sonicated in a 20 fold volume of 0.2 mol/l HC104 containing 50 pmoles of dihydroxybenzylamine (as an internal standard for determination of catecholamines) and 50 pmoles of 3 ethoxy, 4-hydroxyphenylglycol

(EHPG, as an

internal standard for determination of MHPG and HVA). Two hundred pl of brain tissue homogenate from above were aliquoted into a separate 1.5 ml polypropylene tube. To this was added 20 mg Al203 and 1.3 ml of 1.0 mol/l Tris buffer, pH=8.6. Catecholamines were extracted as previously described and determined by high performance liquid chromatography (HPLC) with amperometric detection (Mefford et al., 1987). Total MHPG and HVA were determined in the tissues following acid hydrolysis at 950C for 9 minutes. Briefly, tissue homogenate as prepared above, was placed in an eppendorf incubator for 9 minutes at 95oC, previously observed to give optimal hydrolysis (J.K. Hsiao, A.L. Lawrenz & I.N. Mefford, unpublished observations).

Hydrolysis was stopped by placing the samples on ice. Samples were then centrifuged for 3

minutes at 12,800 x g and 50 l.tl aliquots of the clear supematant used for determination of MHPG and HVA. Determination of MHPG and HVA was accomplished using HPLC with amperometric detection. Separation was accomplished using a 10 cm x 4.5 mm i.d. column packed in-house with 3l.t ODS hypersil (Shandon). A mobile phase consisting of 0.1 mol/l sodium acetate, 2 % methanol, 100 rngfl EDTA adjusted to pH 5.2 with citric acid, was employed. Detection was accomplished at a glassy carbon electrode (TL-8A, Bioanalytical Systems) at +0.75 volts vs. Ag/AgCl reference.

Tissue content of MHPG and HVA were determined following correction for

recovery of the internal standard, EHPG. Phenylethanolamine-N-methyltransferase

(PNMT) activity was determined by a modification of the method of

Trocewicz et al. (1982). The section of the medulla oblongata was selected where the maximal PNMT activity was found (Renaud et a1.,1986): between 0.5mm caudal to the obex and 1.5mm rostra1 to the obex. The frozen hypothalami and medulla were weighed, sonicated in 10 vol of sucrose 0.32 M and the activity measured in 50 pl aliquots of tissue homogenates. Statistical analvsis It was accomplished using analysis of variance followed by Dunnett’s t test (Wirier, 197 1).

Time Course of the Effects of LPS Administration

on Catecholamine

and Metabolite Concentration

and

Catecholamine Turnover in Hwothalamus Administration

of LPS produced marked changes in catecholamine

and metabolite

concentrations

in

hypothalamus (Figs 1 and 2). Both epinephrine and norepinephrine were significantly depleted at 4, 8 and 24 hours, while dopamine content was unaffected (Fig 1). Both HVA and MHPG were significantly elevated following LPS administration at 4, 8 and 24 hours (Fig 2). Similar results were obtained following administration

M. I. Masana

612

et al.

of PWM, although measurements were made only at 8 hours. Significant depletion of both norepinephrine (all values pmol/mg tissue, saline: 12.8 f .42, n=8; PWM: 9.38 f .38, n=5) and epinephrine (saline: 0.270 + 0.015, n=8; PWM: 0.133 f 0.025, norepinephrine/epinephrine,

n=5) was observed,

as were elevations

in the major catabolites

of

MHPG (saline: 0.600 f 0.049, n=8; PWM: 1.38 f 0.123. n=5) and dopamine, HVA

( saline: 0.495 f 0.012, n=8; PWM: 0.697 f 0.050, n=5). The ratio of metabolite to parent amine has been used as an index used to obtain some measure of turnover from static concentration

measurements

[MHPG]/[norepinephrine]

in tissue.

This ratio was calculated at 4, 8 and 24 hours for both

and [HVA]/[dopamine].

Interestingly,

this index indicates no effect of LPS on

dopamine turnover at 4 hours, but a statistically significant increase at 8 hours with either LPS (79.1 k 8.5 %, n=12) or PWM (60.2 k 12.0 %, n=5).

Largest effects were observed with the noradrenergic

system.

[MHPG]/[nompinephrine] was increased by 136.6 f 31.7 %, n=8,4 hours following LPS administration (Fig 3). At 8 hours, the increase was 155.1 f 22.4 %, n=8, after LPS and 217.0 f 36.2 %, n=5 after PWM. At 24 hours both [MHPG]/[norepinephrine]

and [HVA]/[dopamine] ratios were increased by 80 % (Fig 3). These data are

summarized in Fig. 3.

60 -

0 0

I 4

I 8

I 12

I 16

I 20

I 24

HOURS AFTER LPS INJECTION

Fig 1. Effects of LPS on norepinephrine (Qm), epinephrine (A&, and dopamine (0.0) content in the rat hypothalamus 4, 8 and 24 hours after administration (closed symbols) and antagonism by co-administration of indomethacin (open symbols). Each point represents the mean + SEM . Data expressed as % of control (saline). 0.251 f 0.024, 0.245 f 0.012, 0.361 f 0.024; Control values in pmol/ mg wet weight: Epinephrine: 2.87f0.11, 3.87k0.27, norepinephrine: 13.48 f 0.29, 12.17 k 0.30, 18.38 + 0.43 ; dopamine: 2.86k0.15, for 4 hours (n=8), 8 hours (n=16) and 24 hours (n=7) respectively. * pd.05 vs saline by Dunnett’s t test.

Brain catecholamine

0

4

HOURS

turnover

8

613

after LPS challenge

12

AFTER

16

20

24

LPS INJECTION

Fig 2. Effects of LPS on MHPG (t&m) and HVA (0,O) content in the rat hypothalamus 4,8 and 24 hours after administration (closed symbols) and antagonism by co-administration of indomethacin (open symbols). Each point represents the mean + SEM. Data expressed as % of control (saline). Control values in pmol/ mg wet weight : MHPG: 0.550 f 0.055, 0.600 f 0.055, 0.667 f 0.071 ; HVA: 0.661 f 0.046.0.501 f 0.018, 0.754 k 0.110, for 4 hours (n=8), 8 hours (n=12) and 24 hours (n=7) respectively. * pd.05 vs saline by Dunnett’s t test.

0

4

HOURS

8

AFTER

12

16

20

24

LPS INJECTION

Fig 3. Effects of LPS on turnover of norepinephrine (0,B) and dopamine (0,O) in the rat hypothalamus 4,8 and 24 hours after administration (closed symbols) and antagonism by co-administration of indomethacin (open symbols). Each point represents the mean + SEM. Data expressed as % of control (saline). Control values in pmol/ mg wet weight: [MHPG]/ [norepinephrine]: 0.041 f 0.004, 0,049 f 0.004, 0.028 k 0.003 ; [HVA]/[dopamine]: 0.237 f 0.019, 0.177 f 0.007, 0.168 zk 0.026, for 4 hours (n=g), 8 hours (n=12) and 24 hours (n=7) respectively. * ~~0.05 vs saline by Dunnett’s t test.

614

M. I. Masana et al.

. . Effects ofWS Adrmnisiration on Other Brain Reeions Since the effects on catecholamine content and turnover were more pronounced 8 hours after the LPS injection, we chose this time point to study the effects of LPS on other brain regions. These results are shown in Table 1.

Catecholamine content was not modified in medulla, striatum nor cortex, while LPS increased HVA content in these 3 regions and MHPG content in medulla. HVA values shown represent total HVA, after hydrolysis. Effects of IL-1 Admr‘n’lstran‘on on Catecholamine and Metabolite Content and Catecholamine Turnover in Different Brain Rw Since the biological activity of IL-l accounts for several aspects of the acute phase response (Dinarello, 1984), it was of interest to know the effects of 10 ltg IL-1 on catecholamine and metabolite content in different brain areas (Table 1). The effect of IL-1 on hypothalamic epinephrine content was similar to the effect of LPS. Epinephrine content was reduced to 40% of the saline control value. Norepinephrine content in medulla, striatum and cortex was not affected by IL-1 administration, but was reduced 12% in hypothalamus. MHPG content was increased by almost 160% and norepinephrine

turnover by 198 % in hypothalamus, while in medulla a modest but not

statistically significant increase of 40 % (Table 1) was observed for MHPG and 48% for the index of I norepinephrine turnover (Fig 4). HVA content was significantly increased in hypothalamus, striatum and cortex (table 1) and dopamine turnover (~VA]/[dopamine])

was increased in all the areas studied (Fig 5).

Co-administration of LPS or lL-1 with Indomethacin We examined whether the effects of endotoxin and IL-1 on norepinephrine, epinephrine and metabolite content were mediated through prostaglandin formation as has previously been demonstrated for the temperature-elevating effects, tachycardia and stimulation of stress hormone release (Feldberg and Gupta, 1973; Revhaug et al., 1988). Indomethacin was used as the cyclooxygenase inhibitor, antipyretic, for this purpose. Indomethacin had no effect on tissue content of catecholamines or metabolites (Table 1 shows 8 hour data), however, co-administration of indomethacin with LPS completely antagonized the LPS-induced changes in hypothalamic and medullary norepinephrine, epinephrine, MHPG and norepinephrine turnover at 4 hours (Figs 1 and 2) and 8 hours (Table 1 and Fig 4). No significant differences were observed between saline and indomethacin + LPS treated animals (Table 1, Fig 3 and 4). At 24 hours, only the decrease in catecholamines was attenuated (Fig 1). IL-1 effects on hypothalamic MHPG content and [MHPG]/[norepinephrine] ratio were also blocked by indomethacin. (Table 1 and Fig S), while the decreases in norepinephrine and epinephrine content were partially antagonized (Table 1). The effect on the dopaminergic system was not so clear. The LPS-induced increase in HVA was significantly reduced by indomethacin (Table 1 and Fig 2) in all the regions studied, however, the increase in turnover was only partially attenuated at 8 hours in hypothalamus and medulla (Figs 3 and 5), but was blocked in striatum (Fig 5). HVA content in hypothalamus, striatum and cortex and dopamine turnover (mVA]/[dopamine]) in hypothalamus, elevated following IL-l, were not affected by concomitant administration of indomethacin.

This index was

partially reduced in medulla and completely antagonized by indomethacin in striatum and cortex. Phenvlethanolamine-N-methvltransferase Both physical and physiological stressors elevate brainstem phenylethanolamine-N-methyhransferase

(PNMT),

the enzyme responsible for the last step in the synthesis of epinephrine, while hypothalamic PNMT is relatively unaffected (Saavedra and Torda, 1980, Turner et al., 1978). The enzyme was not modified by LPS treatment in hypothalamus (saline: 4.00 f 0.20 pmol/hr/mg wet tissue, n=lO, LPS: 3.48 + 0.22 pmol/hr/mg wet tissue, n=lO)

615

Brain catecholamine turnover after LPS challenge

n

SALlNE

0

INDOMTHACIN

q q n q

HYPOTHALAMUS

LPS L!‘S+INDOMETHACIN IL-1 IL-1+INDOMETHACIN

MEDULLA

Fig 4. Effects of LPS and IL-l on MHPG/norepinephrine ratio in the hypothalamus and medulla 8 hours after administration and antagonism by co-administration of indomethacin. The bars represent mean f SEM of at least 8 animals per group. Data expressed as % of control (saline). Control values were: hypothalamus: 0.049 f 0.004; medulla: 0.094 f 0.0 18. * pcO.005, # ~~0.025 significantly different from saline by Dunnett’s t test.

n 0

q q n q

HYPOTHALAMUS

MEDULLA

STRIATUM

SALINE INDOMTHACIN LPS LPS+INDOMETHACIN IL-1 IL-1+INDOMETHACIN

CORTEX

Effects of LPS and IL-l on HVA/dopamine ratio in the hypothalamus and medulla 8 hours after administration and antagonism by co-administration of indomethacin. The bars represent mean T!Z SEM of at least 8 animals per group. Data expressed as % of control (saline). Control value were: hypothalamus: 0.177 + 0.007; medulla: 1.447 k 0.150; striatum; 0.077 f 0.004, cortex: 0.125 f 0.003. * p
M. I. Masana

616

et al.

Table 1 Effects of LPS, IL-l and co-administration

with indomethacin,

epinephrine, DA: dopamine), HVA and MHPG concentrations

on catecholamines(NB:

E:

in different rat tissues, 8 hours after the injection.

DA

E

NE

norepinephrine,

HYPOTHALAMUS Saline Indomethacin LPS LPS + Indo IL-l IL-l + Indo

12.2 11.5 10.2 11.5 10.8 10.9

f f f f f f

.30 .66 .38*** .44 .29* .44

.245 .229 .098 .228 .089 .I76

f f f f f f

.02 .028 .017*** .013 .007*** .014**

2.87 2.82 2.64 2.54 2.56 2.58

f f f f f f

.ll .22 .14 -10 .19

.501 .583 .826 .635 645 .693

MEDULLA Saline Indomethacin LPS LPS + Indo IL-1 IL-l + Indo

6.95 6.71 6.54 6.88 6.37 6.25

f f f f f f

.16 .14 .22 .18 .19 .15

.084 .081 .071 .080 .069 .062

f f f f + +

.006 .007 .007 .004 .004 .005

.599 .597 .621 .639 .561 .614

f f f f + f

.023 .018 .025 .025 .032 .038

601 f .627 f 1.127f .939 f .863 f .833 +

.042 .042 .197*** .139 .080 .060

STRIATUM Saline Indomethacin LPS LPS + Indo IL-l IL-l + Indo

2.09 2.02 2.16 2.14 2.30 2.39

f f f f f f

.23 .24 .37 .33 .27 .40

57.4 60.5 60.4 63.3 64.5 63.2

f f f f f f

1.3 2.2 2.6 4.3 2.3 2.5

4.39 4.72 6.11 4.91 6.45 5.52

f f f f f f

.14 .18 .42*** .31 .31*** .29**

CORTEX Saline Indomethacin LPS LPS + Indo IL-l IL-l + Indo

1.91 1.98 1.95 1.91 1.90 1.93

f f f f f f

.lO .07 .09 .09 .08 .07

2.09 2.18 2.53 2.20 2.04 2.27

f It f f f f

.09 .33 .32 .21 .13 .22

.260 .272 .393 .236 .399 .361

f f f f f f

.013 .032 .032*** .022 .018*** .023**

.14

MHPG

HVA f f f f f f

.018 .051 .052*** .033 .016* .061**

600 f .664 f 1.259f .650 f 1.559-1 .664 rt .510 .522 .829 .481 .712 .395

.055 .210 .082*** .092 .227*** .I17

If: .091 + .095 + .124* f .067 + .076 z!z.086

All values are pmol/mg tissue wet weight of al least 7 animals per group. * p < 0.05,**p
wet tissue, n=lO, LPS: 4.91 f 0.34 pmol/hr/mg

wet tissue, n=lO),

nor was any increase observed in medullary PNMT after 8 hr of treatment with LPS or PWM (saline: 2.56 f 0.19 pmol/hr/mg

wet tissue, n=8, LPS: 2.62 + 0.25 pmoVhr/mg wet tissue, n=8, PWM: 2.43 f 0.13 pmol/hr/mg

wet

tissue, n=5).

Discussion Increase of Hvnothalamic

Catecholamine

Turnover followins Stimulation of the Immune Svstem

The present data confirm the previous observations that stimulation of the immune system leads to increased turnover in hypothalamic

noradrenergic

neurons (Besedovsky

Mefford and Heyes, 1990). Two substances effects on catecholamine

content and metabolite concentrations

increased turnover of both norepinephrine

et al., 1983; Dunn, 1988; Kabiersch

known as “in vitro” B-cell mitogens,

at 8 hours following administration,

and, to a lesser degree, dopamine,

et al., 1988;

LPS and PWM, had similar

in hypothalamus.

indicative of

Further, marked

617

Brain catecholamine turnover after LPS challenge

depletion of epinephrine was observed, 60% at 8 hours which suggests that turnover in epinephrine-containing neurons was also increased.

Similar effects of LPS were observed at 4 and 24 hours, although the degree of

depletion of epinephrine was higher at 8 hours and the apparent turnover of dopamine was not changed at 4 hours. The results reported here after immunological challenge are similar to the effects reported for other types of stress such as swimming stress (Sudo, 1983), cold swimming stress (Roth et al., 1982) and ethanol (Mefford, 1987). The effects we have observed are consistent with observations of others on the effects of immune stimulators on hypothalamic catecholamine concentrations.

Besedovsky et a1.(1983) reported a decrease in

hypothalamic norepinephrine content 5 days following i.p. administration of sheep erythrocytes to rats. These effects were apparently not related to stress, since the changes were only measurable when the increases in antibody formation were observed (5 days after the injection).

More recently, Kabiersch et al. (1988), also

reported decreased norepinepbrine content and increased [MHPG]/ [norepinephrine] ratio (as an index of turnover) in rat hypothalamus and brain stem 2 hours after IL-l administration.

Dunn (1988) measured hypothalamic

norepinephrine, dopamine, h4HPG and HVA in mouse hypothalamus following i.p. administration of IL- 1 and observed a marked reduction of hypothalamic norepinephrine as well as an elevation of MHPG and [MHPG]/ [norepinephrine] ratio. In mouse, we have reported increases in norepinephrine and dopamine turnover after LPS (Mefford and Heyes, 1990). We interpret our results to indicate that LPS and PWM, both act to increase catecholamine turnover in the hypothalamus, consistent with the actions of IL-l noted by Dunn (1988) in the mouse and by Kabiersch et al. (1988) and also reported here for the rat. It is unlikely that LPS or IL- 1 cross the blood brain barrier and act directly on brain tissue, although this may be possible in areas of weak blood brain barrier such as the organum vasculosum laminae terminalis (Mashburn et al., 1983; Stitt, 1986). It has been demonstrated that intracranial application of endotoxin or IL-l will induce fever in animals (Stitt, 1986) and that IL1 stimulates prostaglandin production in hypothalamic preparations (Dinarello et al., 1986). Intracerebral activity of these agents is possible, as astrocytes are sources of prostaglandins and an IL-l-like

factor under certain

conditions (Fontana et al., 1982; Murphy et al., 1988). We extended the present study to other brain areas and found increases in norepinephrine turnover in medulla, and increased dopamine turnover in medulla and striatum. However, the effects were greater in the hypothalamus and may be related to the role of the hypothalamus in temperature regulation and steroid production. Activation of thermoregulatory centers in the anterior preoptic hypothalamus has been implicated as the site of action for the pyrogenic effects of a variety of immune stimulating substances (Dinarello and Wolff, 1982). Additionally, changes in neuronal activity have been reported in the hypothalamus during the immune response (Besedovsky et al., 1977) and a rapid stimulation of ACTH and corticosterone was observed after IL-1 injection (Besedovsky and de1 Rey, 1987; Sapolsky, et al., 1987). However, these authors found that the neuroendocrine effects of IL-l could be dissociated from the pyrogenic effects of the cytokine (de1 Rey and Besedovsky, 1987; Sapolsky et al., 1987) and Kabiersch et al. (1988) have reported changes in catecholamine turnover in different brain areas using sub-pyrogenic doses of IL-l. Therefore, it is also possible that the increased catecholamine turnover we have observed represents a generalized response to the presence of an infectious agent or action of cytokines. Role of urostaelandins The LPS-induced increase in norepinephrine turnover in hypothalamus and medulla was completely antagonize 4 and 8 hours following co-administration of indomethacin (SOmg/kg), indicating that these effects of LPS may be mediated by products of arachidonic acid metabolism. The effects of prostaglandins (particularly PGE2) on catecholamine release in brain tissue preparations have been reported to be both inhibitory (Bergstrom et al., 1973;

616

M. I. Masana

et al.

Reimann et al., 1981), as well as facilitatory (Roberts and Hiller, 1976). Our results indicate either a direct facilitatory effect of cyclooxygenase products on norepinephrine turnover in response to LPS or an indirect role for cyclooxygenase products in the turnover increase acting at an intermediate step prior to the norepinephrine neuron. Twenty four hours after the treatment with LPS and indomethacin, only the decrease in catecholamines was partially blocked, while the effects on catecholamine metabolites and turnover were not attenuated. This may reflect a more rapid clearance of indomethacin than LPS or that LPS activates some mechanism that increases catecholamine turnover independent of prostaglandin synthesis. LPS and IL-l. a wmmon pathwav of action? The effects of both, IL-l and LPS on MHPG content and norepinephrine turnover are very similar and are both blocked by indomethacin, which may indicate a common pathway of action. The CNS consequences of LPS and IL-1 administration

could be related to the calcium ionophore capacity of IL-l, allowing the activation of

phospholipase A and release of arachidonic acid (Dinarello, 1984). However the regional effects of IL-1 on HVA content and dopamine turnover were either not modified or only partially antagonized by indomethacin. This suggests that the effect of IL-l on central dopamine turnover may be mediated by factors other than prostaglandin formation. Comparison with another types of stress The decrease in norepinephrine and epinephrine content and increase in [metabolite]/[catecholamine] ratio found after endotoxin challenge is similar to effects reported for other types of stress such as swimming stress (Sudo, 1983), cold swimming stress (Roth et al., 1982) or ethanol (Mefford, 1987). The intermediate role for arachidonic acid metabolites has also been reported for other aspect of the stress, such as the decrease in the number of noradrenergic B-receptors in hypothalamus and brain stem after repeated immobilization stress. This desensitization is prevented by treatment with mepacrine, a phospholipase A2-inhibitor (Torda et al., 1981). However, some differences should be pointed out, for example, LPS challenge produced no changes in PNMT activity in either the hypothalamus or the medulla, while it has been reported that acute and repeated immobilization stress (Saavedra and Torda, 1980) or horizontal shaker stress (Turner et al., 1978) increases the PNMT activity in the brainstem.

Footshock stress, while causing activation of the mesocortical dopaminergic system, does not

necessarily increase striatal dopamine turnover (Deutch et al., 1985; Thierry et al., 1976), unlike our present observation. Indomethacin blocked most of the effects of LPS on catecholamine content and turnover. In contrast, the increased catecholamine metabolism observed following ethanol stress (3.2 g/kg, i.p.) is not blocked by indomethacin (unpublished observation). Similarly, IL-1 has been reported to affect selectively some components of the stress response, while others such as prolactin, growth hormone or aMSH levels (Berkenbosch et al., 1987) or brain dopamine and serotonin content (Kabiersch et al., 1988) are not affected by the cytokine. Conclusions Endotoxin challenge produces increases in central catecholamine turnover throughout the brain. Unlike other externally applied stressors such as immobilization, endotoxin challenge does not result in elevated brainstem PNMT activity. Increased turnover of brain norepinephrine during the acute phase appears to be mediated by prostaglandin synthesis. Inhibition of cyclooxygenase by indomethacin antagonizes most but not all of the central consequences of endotoxin challenge. Turnover of dopamine remains elevated in hypothalamus and brainstem

Brain catecholamine turnover after LPS challenge

619

after cycloxygenase inhibition, suggesting pathways of interaction between cytokines and central catecholamines other than prostaglandins.

Additional experiments are necessary to elucidate the mechanism(s) resulting in

increased turnover of CNS catecholamines common to other stressors or unique to endotoxin or immunological stress. References BERKENBOSCH, F., VAN OERS, J., DEL REY, A., TILDERS, F. and BESEDOVSKY, H.O. (1987). Corticotropin-releasing factor-producing neurons in the rat activated by interleukin-1. Science a, 524-526. BERNHEIM, H.A., GILBERT, T.M. and STlTT, J.T. (1980). Prostaglandin E level in third ventricular cerebrospinal fluid of rabbits during fever and changes in body temperature. J. Physiol. a, 69-78. BERGSTROM, S., FARNEBO, L. and FUKE, K. (1973). Effect of prostaglandin E2 on central and peripheral catecholamine neurons. Eur. J. Pharmacol. a, 362-368. BESEDOVSKY H. and DEL REY, A. (1987). Neuroendocrine and metabolic responses induced by interleukin-1, J. Neurosci. Res. u, 172-178. BESEDOVSKY, H., SORKIN, E., FELIX D. and HAAS, H. (1977). Hypothalamic changes during the immune response, Eur. J. Immunol. z 323-325. BESEDOVSKY, H., DEL REY, A., SORKIN, E., DA PRADA, M., BURRI. R. and HONEGGER, C. (1983). The immune response evokes changes in brain noradrenergic neurons. Science m,564-566. COCEANI, F., BISHAI, I., DINARELLO, C.A. and FITZPATRICK, F.A. (1983). Prostaglandin E2 and tromboxane B2 in cerebrospinal fluid of afebrile cat. Amer.J.Physiol. 244. R785-R793 DEL REY, A. and BESEDOVSKY, H.O. (1987). Interleukin 1 affects glucose homeostasis. Amer. J. Physiol. m R794-R798. DEUTCH, A.Y., TAM, S-.Y. and ROTH, R.H. (1985). Footshock and conditioned stress increase 3,4diily;;yphenylacetic acid (DOPAC) in the central tegmental area but not substancia nigra. Brain Res. m

DINARELLO, C.A. (1984). Interleukin-1. Rev. Infect. Dis. 4, 51-95. DINARELLO, C.A. and WOLFF, S.M. (1982). Molecular basis of fever in humans. American J. Medicine 72, 799-819. DINARELLO C.A., CANNON, J.G., WOLFF, S.M., BERNHEIM, H.A., BEUTLER B., CERAMI, A., FIGARI, I.S., PALLADINO, M.A. and O’CONNOR, J.V. (1986). Tumor necrosis factor (cachetin) is an endogenous pyrogen and induces production of interleukin-1. J. Exp. Med. m, 1433-1450 DUNN, A.J. (1988). Systemic interleukin-1 administration stimulates hypothalamic norepinephrine metabolism paralleling the increased plasma corticostemne. Life Sciences a, 429-435. FELDBERG, W. and GUPTA, K.P. (1973). Pyrogen fever and prostaglandin-like activity in cerebrospinal fluid. J. Physiol. a, 41-53. FONTANA. A., KRISTENSEN, F., DUBS, R., GEMSA. D. and WEBER, E. (1982). Production of prostaglandin E and an interleukin-1 like factor by cultured astrocytes and C6 glioma cells. J. Immunol. 129, 2413-2419. HARVEY, C.A., MILTON, AS. and STRAUGHAN. D.W. (1975). Prostaglandin E levels in cerebrospinal fluid of rabbits and the effects of bacterial pyrogen and antipyretic drugs. J. Physiol. m, 26P-27P.

M. 1. Masana et al.

620

HOO, S.L., LIN, M.T., WEI, R.D., CHAI C.Y. and WANG, SC. (1972). Effects of sodium acetylsalicylate the release of pyrogen from leukocytes. Proc Sot. Exp. Biol. Med. m, 1155-1158. KABIERSCH, A., DEL REY, A., HONEGGER C.G. and BESEDOVSKY, H.O. (1988). Interleukin-1 changes in norepinephrine metabolism in the rat brain. Brain Behav. Immun. 2,267-274

on

induces

MASHBURN, T.A. Jr., BLATTEIS, C.M., BEALER, S.L., HUNTER, W.S., LLANOS, Q.J., AHOKAS, R.A. (1983). Endogenous pyrogen (EP) may cross the blood-brain barrier (BBB) in the organum vasculosum of the lamina terminalis (OVLT) in guinea pigs [abstract no. 10091. Fed. Proc. 4,464. MEFFORD, I.N. (1987). Ethanol and brain epinephrine, pp 870-880. In: Linnoila, withdrawal and noradrenergic function. Ann. Intern. Med. m, 875-889.

M., moderator.

MEFFORD, I.N. and HEYES, M.P. (1990). Increased biogenic amine turnover in mouse hypothalamus immunological challenge: antagonism by indomethacin. J. Neuroimmunol. In Press.

Alcohol

following

MEFFORD, I.N., OTA, M., STIPETIC, M.D. and SINGLETON, W. (1987). Application of a novel cation exchange reagent, Igepon T-77 (N-methyl oleoyl taurate), to microbore separations of alumina extracts of catecholamines from cerebrospinal fluid, plasma, urine and brain tissue with amperometric detection. J. Chromatogr. 420, 241-251. MURPHY, S, PEARCE, B., JEREMY, J. and DANDONA, Glia 1, 241-245.

P. (1988). Astrocytes

as eicosanoid-producing

cells.

REIMANN W., STEINHAUER, H.B., HEDLER, L., STARKE K. and HER’ITING G. (1981) Effect of prostaglandins D2, E2 and F2alpha on catecholamine release from slices of rat and rabbit brain. Eur. J. Pharmacol. @, 421-427. RENAUD, B., LAMBAS-SENAS, L., FETY, R., COLIN, I., ASTIER, B. and CHAMBA, F. (1986). Biochemical pharmacology of brain epinephrine neurons: Comparison with norepinephrine neurons. In: Neurobiology of brain epinephrine, D.C. U’Prichard, J.M. Stolk and K. Fuxe (Eds.), pp Oxford UnivPress, New York . REVHAUG, A., MICHIE, H.R., MANSON J.McK.,WATERS, J.M., DINARELLO, WILMORE, D.W. (1988). Inhibition of cycle-oxygenase attenuates the metabolic humans. Arch. Surg. 123, 162-170 ROBERTS, P.J. and HILLER, K. (1976). Facilitation prostaglandin E2. Brain Res. jl2,425-428.

of norepinephrine

C.A, WOLFF S.M. and response to endotoxin in

release from rat brain synaptosomes

by

ROTH, K.A., MEFFORD, I.N. and BARCHAS, J.D. (1982). Hypothalamic catecholamine and serotonin recovery from acute stress and the effects of chronic and repeated stress on regional catecholamine and serotonin concentration and turnover. Brain Res. m, 417-424. SAAVEDRA, J.M. and TORDA, T. (1980). Increased brainstem and decreased hypothalamic enzyme after acute and repeated immobilization stress in the rat. Neuroendcc. 21, 142-146.

adrenaline-forming

SAPOLSKY, R., RIVIER C., YAMAMOTO, G., PLOTSKY, P. and VALE, W. (1987). Interleukin-1 stimulates ACTH release by an indirect action which requires endogenous corticotropin releasing factor. Endocrinology 121, 1580-1582 SAXENA, P.N., BEG, M.M., SINGHAI, K.C. and AHMAD, M. (1979). Prostaglandin-like cerebrospinal fluid of febrile patients. Indian J. Med. Res. z, 495-498. STI’IT, J.T. (1986). Prostaglandin 149.

E as the neural mediator of the febril response.

activity in the

Yale J. Biol. Med. 2,

137-

Brain catecholamine turnover after LF’S challenge

621

SUDO, A. (1983). Time course of the changes of catecholamine levels in rat brain during swimming stress. Brain Res. m, 372-314 . THIERRY, A.M., TASSIN, J.P., BLANC, Cl. and GLOWINSKI, mesocortical DA system by stress. Nature (Land.) m, 242-244.

J. (1976). Selective activation of the

TORDA, T., YAMAGUCHI, I., HIRATA, F. and KOPIN, I.J. (1981). Mepacrine treatment prevents immobilazation-induced desensitazion of beta-adrenergic receptors in rat hypothalamus and brain stem. Brain Res. X&441-444 TROCEWICZ, J., OKA, K., NAGATSU T. (1982). Highly sensitive assay for phenylethanolamine Nmethyltransferase activity in rat brain by high-performance liquid chromatography with electrochemical detection, J. Chromatogr. m, 407-410. TURNER, B.B., KATZ, R.J., ROTH, K.A. and CARROL, B.J. (1978). Central phenylethanolamine-N-methyltransferase activity following stress. Brain Res. m 419-422 . WINER, B.J. (1971). Statistical Principles in Experimental Design, Second edition, McGraw Hill. Inquiries and reprints requests should be addressed to: Monica I Masana Section on Clinical Pharmacology, Neuroscience Branch, NIMH, Building 10, Room 2D46 9000 Rockville Pike, Bethesda, Maryland, 20892

elevation

of