Chemical and physiological aspects of the actions of lithium and antidepressant drugs

00%3908/83/380359-07$03.00/O Pergamon Press Ltd

!V~‘~r~,pJtctrrllu~,ol~~~ Vol. 22, No. 38. pp. 359-365. 1983 Printed in Great Britain

CHEMICAL AND PHYSIOLOGICAL ASPECTS ACTIONS OF LITHIUM AND ANTIDEPRESSANT BLOOM. G. BAETGE. S. DEYO,

F. E. Arthur

Vining

Davis Center

A.

ETTENBERG. L. KODA,

W. J. SHOEMAKER

and D. A. STAUNMN

for Behavioral

P. J.

Neurobiology, The Salk Institute. CA 92138, U.S.A.

OF THE DRUGS

MAGISTRETTI,

P.O. Box 55500, San Diego

Summary-The possible mechanisms underlying the anti-manic actions of lithium have been examined in a variety of interdisciplinary experiments. The possibility that lithium can regulate the sensitivity changes in dopaminergic transmission produced by chronic treatment with haloperidol has been tested. Although a modest modification of behavioral responses to the dopaminc agonist apomorphine was found, there was no evidence that this action of lithium reflected alterations of the binding parameters of dopamine-related ligands. In other studies, consistent, dose-dependent increases in brain enkephalin content were found after rats consumed a specially manufactured lithium diet for 2-3 weeks. Not only were brain enkephalin levels increased after this treatment. but some signs of basal analgesic responsiveness also suggested that the elevated levels of enkephahns were functionally significant. To test the possibility that the effects of lithium may not be seen in normal rats, the effects of lithium were compared on spontaneously hypertensive and unaffected, normotensive rats of a related strain. Treatment vvith lithium altered blood pressure in the hypertensive strain but did not affect blood pressure in the controls. These studies suggest that multiple brain systems may be regulated by treatment with lithium but that the critical pathophysiological process may not be demonstrable in the normal rat.

This laboratory has been interested in the molecular mechanisms of two drugs used in the treatment of mania and depression: lithium, the most widely used anti-manic treatment, and ethanol, a very widely selfprescribed antidepressant compound, with mild euphorigenic actions. In this communication, some recent efforts to determine the possible mechanisms by which lithium produces long-Iastin~ effects on interneuronal communication which may underlie its anti-manic, as well as its antidepressant effects on patients with bipolar affective psychosis are described. Recent work is reviewed that has pursued a major supposition shared by the present authors and many other groups namely that it is the central monoamines which mediate the behavioral effects of lithium. The possibility is then considered that other central transmitter systems, such as the widely studied enkephalin systems, may also be altered by chronic diets of lithium, given to otherwise normal experimental animals. Lastly, the effects of treatment with lithium in a comparative analysis of the genetically abnormal spontaneousiy hypertensive rat is examined. to illustrate that the effects of lithium are functionally different from those observed in normal rats. This leads to the conclusion that while it may be possible to discern adaptive neurochemical. electrophysiological and behavioral effects of lithium in normal rats, these do not necessarily retlect on the therapeutic mechanisms which may underlie the actions in the psychotic human being. LITHIUM

AND

DOPAMINE

RECEPTOR

ADA~ATION

Recent theories of the action of lithium suggest that overactivity of a dopaminergic pathway may. at least 359

in part, mediate human mania (see Bunney and Garland, 1983). Accordingly, there has been extensive interest in testing the possibility that prolonged administration of lithium may act to overcome or prevent amplification of dopaminergic function at either pre- or postsynaptic levels (Gallagher, Pert and Bunney, 1978; Pert, Rosenblatt, Sivit, Pert and Bunney, 1978). In particular, the behavioral supersensitivity which follows prolonged treatment with haloperidol has been reported to be abolished by chronic concomitant exposure to lithium (see Gallagher et nl., 1978; Pert et al., 1978: Staunton, Magistretti, Shoemaker and Bloom, 1982a; Bunney and Garland, 1983). Elsewhere (Schultz, Siggins. Schocker, Turck and Bloom, 1981) the electrophysiological consequences of chronic dietary lithium in inducing a modest sub-sensitivity to norepinephrine has been described. As part of that study. a regional analysis of the concentrations of lithium in brain was made in chronically treated rats, which indicated that the same diet could lead to substantial differences between brain regions, with lithium in the striatum being among the greatest of the regions sampled (Table 1).

In order to investigate the effects of acute or chronic treatment with lithium on animal behavior, it has been necessary to control for wide swings in serum and tissue levels of lithium, renal damage, or other signs of toxicity, especially weight loss and diarrhea. Without control for, or elimination of, these problems, it is almost impossible to elucidate the influence of chronic exposure to lithium on sensitive indices of central function.

360

F. E. BLOOM et al.

Table 1. Effects of 4 weeks of dietary treatment with lithium on regional brain concentrations

of lithium

Brain region Cerebral cortex Amygdala Cerebellum Corpus striatum Hippocampus

in rats (n = 10) Li (mean concentration in mM + SEM) 0.98 0.76 0.69 1.15 1.00

k k + + *

0.14 0.08 0.06 0.26 0.20

As reported in detail elsewhere (Bloom, Magistretti, Shoemaker, Siggins and Staunton. 1980) a diet of lithium was employed which produced reliable serum levels of lithium in the human clinical range without any overt toxicity. This diet has now been used to evaluate the effect of lithium both on spontaneous activity and on enhanced behavioral responsiveness to apomorphine in order to reinvestigate the reported ability of lithium to antagonize the supersensitive dopamine (DA) responses observed in rats pretreated chronically with the antipsychotic. haloperidol (HAL). Because it was consistently observed that diets of lithium diminished food intake and altered growth rates even without any signs of toxicity, this was controlled for by assigning rats to six groups; two groups received the diet of lithium and the others the control diet. For purposes of pair-feeding, each animal in a given control group was weight-matched to an animal in one of the lithium-diet groups. Throughout the course of the study, the food intake of the lithium-diet groups was monitored and the corresponding pair-fed animal’s intake was restricted. This procedure provided paired animals at the same body weight. This pair-feeding routine was conducted once each day from 80&l 100 hr. Two additional groups (six animals each) were fed with the control diet ad libitum. The diet (Teklad Mills, Madison, WS: TD 79092) had the following formulation : 1.696 g (40 mmol)/kg LiCl, 210 g/kg protein, 690 g/kg carbohydrate sources and 50 g/kg lipid. The control diet (TD 80295) was identical with the exception that lithium (LiCl) was omitted. Animals fed with the lithium diet ate freely; all subjects had free access to water. Eight days after the institution of lithium and control diets, drug treatments were begun. Haloperidol (McNeil, injectable), 2.2 pmol/kg, was injected subcutaneously once per day to one lithium-diet group, to the corresponding pair-fed control diet group and to one ad lihitum control-diet group. Another group on the diet of lithium and the corresponding pair-fed and ad U&urn-fed controls were injected subcutaneously once a day with the vehicle for haloperidol (1.8 mg/ml methylparaben, 0.2 mg/ml propylparaben and lactic acid for pH adjustment to 3.4 f 0.2) 1 ml/kg. After three weeks, the injections were terminated. All of the subjects in this investigation gained weight steadily over the course of 32 days of injec-

tions and behavioral testing. There were no overt signs of diarrhea or other toxicity. However, the rate of weight gain was still different in animals on the diet of lithium compared to those with free access to the control diet. During the first week, animals on the diet of lithium gained weight much more slowly than ad libitum fed animals, corresponding to reduced consumption of the lithium diet during this period. In weeks 2-4, food intake in the lithium-treated subjects increased and their rate of weight gain was 60% as fast as that in subjects fed the control diet ad libitum. Animals pair-fed the control diet showed nearly identical weight gain to those given the lithium diet. The brain concentration of lithium achieved in the animals in this study was 0.79 & 0.05 mEq/l brain tissue (range 0.6415 mEq/l) and was identical in haloperidol- and vehicle-injected animals (P > 0.05) (for further details, see Staunton et al., 1982a). Motor activity before and after iujection of apornorphine Spontaneous motor activity (as judged by a 60 min period of free locomotion in a photocell cage after a 60 min habituation period; see Sahakian. Robbins and Iversen, 1976) was not significantly different between animals on the pair-fed control vs the ad libitum control diet (P > 0.05). Thus, food-restriction alone failed to alter habituation to the activity cages. In each of three testing sessions, lithium-treated animals were less active during habituation than their pair-fed controls. However, in none of the three daily testing sessions, was there an effect of drug (HALvehicle) treatment; likewise, there was no interaction of between the haloperidol-vehicle and lithium-control diets (P > 0.05 for each). Following injection of apomorphine (0.66 PM/kg, s.c.) photocell beam interruptions were suppressed during the first 30-40 min. Differences were observed in this response between animals with restricted intake of the control diet and those with free access to the control food. Animals on the lithium diet were more active than ad lihitm but not pair-fed, controls [F(1,24) = 8.07. P < 0.009; P > 0.05]. With either set of control groups, there was a drug x time interaction [pair-fed: F(6,l) = 6.75, ad libitum: F(6,144) = 4.291. This effect was undoubtedly due (1) to the greater duration of reduced initial activity and (2) to a stimulation of activity approx. 4&60 min after injection of apomorphine in the haloperidol-treated subjects. The latter rebound or recovery phenomenon was greatest in the lithium-haloperidol and pair-fed control diet-haloperidol groups. Therefore, this observation is consistent with the potentiation of apomorphine-induced locomotor activity by food deprivation previously reported by Sahakian and Robbins (1975). In the locomotor response to apomorphine, there was no interaction between lithium and haloperidol pretreatments when compared to either pair-fed or ad libitum control groups. That is, administration of lithium did not influence either the initial phase of sup-

361

Aspects of the actions of lithium and antidepressant drugs pressed locomotor activity or the later rebound of activity in subjects chronically injected with haloperido1 to make them supersensitive to apomorphine. Thus, the lithium diet was unable to modify the responsiveness of haloperidol-treated subjects in this analysis. Stereotyped

hehvior

fo/lowing

admirtistration

of apo-

morphine

Parametric analysis revealed that chronic pretreatment with haloperidol consistently produced high stereotypy scores after administration of apomorphine (see Staunton et a/., 1982a). Although there was no main effect of lithium on the haloperidol-induced elevation of stereotypy scores (P > 0.05). there was a drug (vehicle-HAL) x diet (control-Li) interaction which presumably reflects the fact that the elevation of stereotypy scores due to haloperidol was partially attenuated by the lithium diet. When brain levels of lithium were later evaluated, there were no linear correlations between concentrations of lithium in the brain and stereotypy scores in either the haloperidolor vehicle-injected animals fed on the diet of lithium. In repeated testing, it was found that animals maintained on the diet of lithium for three weeks consistently exhibited less spontaneous activity during habituation to a new environment than either ad libitum or pair-fed controls. In most previous reports (e.g. Gray, Solomon, Dunphy. Carr and Hession, 1976; Johnson, 1976) using chronic exposure to lithium, the results were confounded by the use of multiple daily injections of lithium with corresponding fluctuations in serum levels of lithium or, by loss of body weight. The present study strongly suggests that reduced locomotor activity is, in fact, a physiological consequence of chronic exposure to lithium because the same response was also observed in healthy subjects and was probably not associated with the stress due to decreased food consumption. This result can tentatively be used to support the view that the activitysuppressing action of lithium does not itself show tolerance. These conclusions must be considered as first approximations because the possible influence of lithium-induced alterations of circadian rhythms (Kafka, Wirz-Justice, Naber, Marangos, O’Donohue and Wehr, 1982) on spontaneous behavior were not evaluated in this investigation. Rather, the primary goal in these studies was to use chronic dietary lithium to evaluate the effects of lithium on behavioral or cellular dopaminergic supersensitivity following long-term treatment with the antipsychotic, haloperidol. Combined chronic exposure of the animals in this investigation to lithium and haloperidol was indeed associated with lower stereotypy scores following subcutaneous injection of apomorphine than in animals receiving only treatment with haloperidol. Although this is qualitatively identical to a previous report (Pert et al., 1978), there are some important differences. In the earlier study (Pert et al., 1978), the elevated responsiveness to apomorphine was completely obliter-

ated while in the present experiments lithium caused only a relatively weak suppression. This difference may reflect differences in the lithium diets, their continued administration during testing, and procedural differences (e.g. the rats in this investigation were habituated to the testing cage prior to apomorphine challenge). However, this same general effect on behavioral responses has been observed by others (see Bunney and Garland, 1983). In essence, this behavioral investigation confirms an important action of chronic exposure to lithium: of behavioral super-sensitivity the suppression mediated by a dopaminergic substrate. Furthermore, it was established that this effect did not stem from the altered growth rates or some covert toxicity. Whether these behavioral effects of lithium would be reflected in measurable cellular or subcellular changes was then examined.

LITHIUM

AND

DOPAMINE

LIGAND

BINDING

With chronic dietary administration, lithium has also been reported to prevent the usual increase in dopamine receptor density (assessed with [3H]-spiroperidol binding in the corpus striatum) following multiple injections of neuroleptics (Pert et a/.. 1978; Gallagher et al., 1978). An attempt was made to characterize the molecular and cellular bases of the action of lithium more fully. The effects of treatment with lithium on dopamine receptor supersensitivity induced by unilateral dopamine denervation of the nigrostriatal pathway were compared with that reported to be produced by chronic administration of neuroleptics in order to evaluate whether the action of lithium could be generalized to all types of postsynaptic receptor proliferation. It was observed that lithium was unable to prevent post-denervation supersensitivity (see below and Staunton, Magistretti, Shoemaker. Deyo and Bloom, 1982b). Subsequently, however, further evidence was obtained (Staunton unpublished) that there may be fundamental differences in the molecular mechanisms underlying the dopamine supersensitivity produced by denervation from that produced by chronic treatment with neuroleptics. In fact it was observed that when both perturbations were employed simultaneously, treatment with haloperidol and denervation with 6-hydroxydopamine (6-OHDA) produce additive increases in dopamine antagonist binding sites. For these reasons it is important to evaluate independently the possible multiple forms of receptor sensitivity regulation on which lithium or other drugs might act. However, because lithium did not block development of post-denervation supersensitivity. the remaining experiments were used to explore systematically whether chronic administration of lithium could in fact prevent the reported increases in dopamine-related binding in the neostriatum following treatment with haloperidol.

362

F. E. BLOOMrt

Chronic

exposure

istration

of 6-hydroxydopamine

to lithium fbllowing

uniluteral

admin-

Rats were unilaterally denervated of the striatal dopamine innervation using intracerebral injections of 6-hydroxydopamine as described by Ungerstedt (1971a, b). Three weeks later, the subcutaneous injection of small doses of apomorphine HCl elicited the usual rotatory response directed away from the side of 6-hydroxydopamine treatment. There was no difference between the mean rotational rate of animals fed the control diet (ad libitum) and those that ate the diet of lithium. As reported before (Mishra, Marshall and Varmuza. 1980; Staunton. Wolfe, Groves and Molinoff, 1981) depletion of neostriatal dopamine led to a marked elevation of [3H]-spiroperidol binding sites in resuspended membranes prepared from corpora striata on the treated side [F( 1,34) = 40.3, P < O.OOl]. Unexpectedly. the K, values for [3H]-spiroperidol binding were significantly smaller on the denervated side [intact side: 75.6 + 2.9, denervated side: 60.8 k 2.0; F(1,34) = 16.8, P < 0.0011. indicating that the radioligand had higher affinity for the receptor on this side. However, 18 days of dietary treatment with lithium did not affect either the B,,, or the K, on either side (P > 0.05). Thus, 18 days of exposure to lithium (resulting in brain Li+ levels of 0.87 k 0.12 mEq/l) completely failed to influence [3H]-spiroperidol binding in animals with unilateral destruction of the nigrostriatal pathway. Chronic

exposure

to lithium with haloperidol

Animals fed the lithium or control diets were injected with haloperidol or the vehicle for haloperido1 for three weeks and then sacrificed 3 days after withdrawal from the neuroleptic. Scatchard analysis (1949) of [3H]-spiroperidol binding to neostriatal membranes revealed increased binding sites in the with haloperidol (F(1,20) = 66.3, rats treated P < 0.001). As in the case of 6_hydroxydopamine, 24 days of exposure to the lithium diet (brain Li+ levels: 0.99 + 0.16 mEq/l for vehicle-injected group vs 1.0 & 0.21 for HAL-injected group) had no effect on the density of binding sites in subjects exposed to haloperidol or the vehicle (P > 0.05). There was no interaction between the dietary and drug treatments (P > 0.05). Lastly. analysis of the KI, values indicated that there were neither direct effects nor interactions between the drug treatments (vehicle-injected control diet: 69.5 + 3.2 pM; vehicle-injected lithium diet: 66.1 f 2.0 pM; HAL-injected control diet: 72.0 k 3.7 pM; HAL-injected lithium diet: 67.7 & 3.2 PM). Therefore, the DA-related binding parameters were totally unaffected by the treatment with lithium. In other experiments, treatment with lithium had no influence on the binding of [3H]-spiroperidol to neostriatal tissue at short (1 day), intermediate (3, 5 or 7 days) or long (2 weeks) times after the final injection of haloperidol (see Staunton et cd.. 1982b). In all these

al

experiments, there was no direct effect of lithium treatment nor any interaction between lithium and haloperidol pretreatments. Furthermore, although animals fed the lithium diet for several weeks always steadily gained weight, they were generally 1OOg less than the subjects fed the control diet cld lihiturn. However. the binding results were the same when the consumption of the control diet was restricted in pair-fed control rats to keep body weight the same as animals on the diet of lithium. Finally, basal and dopaminestimulated adenylate cyclase were assayed in crude homogenates of the corpus striatum prepared from animals chronically treated as described and withdrawn from haloperidol for 8 days. There was no effect of any pretreatment on basal adenylate cyclase activity, on the ECSo values for dopamine or the maximal dopamine-stimulated activity. The series of investigations reported here represents a systematic attempt to probe the influence of prolonged exposure to lithium on dopamine receptor supersensitivity. Treatment with lithium was accomplished with a diet which reliably produced serum and brain levels of lithium in the therapeutic range (0.8-1.2 mEq/l) without overt signs of toxicity. Although a previous report (Pert et al., 1978) that long-term dietary administration of lithium reduced at least partially, dopaminergic behavioral supersensitivity, elicited by chronic exposure to the antipsychotic. haloperidol was confirmed it was not possible to confirm reports from the same group in which chronic dietary lithium was found to diminish the density of neostriatal [3H]-spiroperidol sites in animals not otherwise treated with drugs (Pert et cd., 1978; Rosenblatt, Pert, Layton and Bunney, 1980; Staunton et a/.. 1982b). The reasons for these important discrepancies are not readily apparent. In fact. of experimental protocols careful comparison revealed only minor differences in the treatment of animals (see Staunton et al., 1982b). It is concluded that the present data do not support the contention that chronic dietary administration of lithium can either prevent or reverse the proliferation of neuroleptic binding sites in the corpus striatum following prolonged treatment with haloperidol. Possibly, the negative influence of lithium on dopamine-mediated behavior may, in fact, be related to an action of the anti-manic drug on systems outside the nigrostriatal axis (see Mandell, 1978; Schultz et al., 1981; Treiser, Cascio. O’Donohue, Thoa, Jacobowitz and Kellar, 1981). A differential analysis of the acute and chronic mechanisms of action of lithium in other defined neurotransmitter systems may, indeed, be needed before the basis of the therapeutic action of this antimanic drug is understood.

LITHIUM

AND

ENDORPHINS

Early in the study of endorphin pharmacology, Gillin, Hong. Yang and Costa (1978) reported that daily

Aspects of the actions of lithium and antidepressant drugs treatment with lithium (i.p.) induced a transitory rise in the met-enkephalin immunoreactive (IR) content of rat globus pallidus. Recently, a more extensive examination of the effects of lithium on enkephalin systems in the rat brain was undertaken using the lithium containing dietary formulations. To evaluate possible dose-related effects, both a lower strength diet containing 30 mmol Lijkg (brain Li = 0.4-0.55 mEq/l) and a high strength diet with 40mmol Lijkg (brain Li = 0.7-1.0 mEq/l) were used. Alterations of leuenkephalin immunoreactivity (1-enk-IR) content were related to changes of in vitro K’-stimulated, CaZ+ dependent release of 1-enk-IR (Staunton, Deyo, Shoemaker, Ettenberg and Bloom, 1982~). After 7 days on the lower strength diet, there were no signi~~ant alterations of I-enk-IR content in any of the brain regions examined, nor was there an effect on the release of I-enk-IR from globus pallidus slices. After 14 or 21 days of ad lihitunr feeding with the lower strength diet, 1-enk-IR levels in the globus palhdus and nucleus accumbens were significantly elevated by 305,. However. 3 weeks of lithium treatment with the high strength diet did not lead to alterations of I-enk-IR content in any of the brain regions examined. When similarly treated rats were tested for res#ponses irr vitro, the release of 1-enk-IR was significantly potentiated (by about 407:) but only in the subjects fed the lower strength diet for 14 days or the high strength diet for 2 1 days. At these times the brain levels of lithium were relatively high compared to the other treatment groups. It is concluded that increased content of the peptide may adapt to the presence of lithium ions, and that the potentiating effect of longterm administration of lithium on release of I-enk-IR in ribo is no1 necessarily associated with content changes. However. increases in I-enk-IR release appears to require minimum brain levels of lithium ions of 0.5 mEy/l. In further studies of the possible links between lithium and enkephalin, it was observed that the hotplate (55°C) escape latency was significantly increased in the animals fed the high strength diet of lithium for 21 days. an effect that was only partially reversed by naltrexone (1 mg/kg, s.c.) pretreatment (Table 2). Table

2. Effects of chronic (2.5 weeks) dietary with lithium on pain sensitivity

treatment

Response latency (set) Treatment Saline Naltrexone morphine Morphine + naitrexone

Control 3.90 4.19 6.72 4.26

i * i f

0.25 0.67 1.14 0.33

Lithium-treated 4.80 4.94 9.46 6.94

* 0.22 +_ 1.13 & 2.26 + 1.04

All acute treatments (saline, naltrexone, 1.0 mg/kg; morphine, 5.0 mgjkg) given by intraperitoneal injections 15 min before placing rats on hot plates (55°C). Latency to attempt escape was measured by observers blind to the subject’s treatment history.

363

Moreover, rats from this treatment group showed a greater morphine-induced (2.5 mg/kg, i.p.) elevation of escape latency than controls. However, naltrexone blocked less effectively the effect of morphine in the subjects fed lithium than in the controls. Subsequent tests showed that there was no difference in o-ala-oleu-enk binding affinity (Kd) or number of binding sites (B,,,). Until possible effects of lithium on circadian rhythmicity are documented, it is possible that some of these escape latency effects could reflect a variety of functional adjustments including circadian pain variations. In conclusion, prolonged administration of lithium led to a transiently increased 1-enk content and, when the concentration of lithium in the brain was more than 0.5 mEq/l, to a potentiation of endogenous enkephalin release. Lithium could be acting directly on presynaptic enkephalin-containing nerve terminals, or indirectly via other neurotransmitter systems, The effects of chronic daily administration of lithium on the in situ release paradigm for enkephalins developed by Bayon (see Bayon, Shoemaker, Lugo, Azad, Ling, Drucker-Colin and Bloom, 1981) are currently being evaluated.

LITHIUM

TREATMENT

GENETICALLY

AND BLOOD HYPERTENSIVE

PRESSURE IN RATS

The dietary restriction of sodium salts as a treatment for hypertension and various cardiovascular diseases led to the dietary substitution of lithium for sodium in the late 1940s. However. because of severe lithium intoxication and deaths, extensive examination of the cardiovascular effects of chronic lithium therapy were never reported. Lithium therapy might be expected to have some influence on blood pressure. Abnormal sodium transport mechanisms have been described in erythrocytes isolated from hypertensive patients (Garay and Meyer, 1979) and rats (BenIshay, Aviram and Viskoper, 1975; De Mendonca, Grichois, Garay, Sassard, Ben-Ishay and Meyer, 1980; Wiley, Hutchinson, ~endelson and Doyle. 1980). However, recent reports have described an abnormal lithium transport system in erythrocytes isolated from patients with essential hypertension (Canessa, Agragna, Solomon, Connolly and Tosteson, 1980) and from genetically spontaneously hypertensive rats (Friedman, ~~kashima, McIndole and Friedman, 1976) and could relate to ionic changes in erythrocytes of affective patients (for details see Koda, Shoemaker, Baetge and Bloom, 1981). In addition, since lithium is used clinically to treat and prevent mania, chronic treatment with lithium might indirectly influence the blood pressure of patients on lithium therapy. For these reasons the effects of chronic treatment with lithium were examined on the blood pressure of the Okamoto strain of spontaneously hypertensive rat (SHR) in which decreased content of sympathetic ganglia enkephalins has been observed (DiGiulio, Yang, Fratta and Costa, 1979).

364

F. E. BLKJM et al.

Table 3. Effects of dietary lithium (2.5 weeks) on blood pressure or heart rate of various rat strains (see Koda et af., 1981)

Subjects Sprague-Dawley controls (n = 7) Sprague-Dawley Li-treated (plasma Li, 0.28 mM & 0.17) (n = 8) Okamoto strain SHR (n = 8) Okamoto strain SHR Li-treated (plasma Li, 0.38 mM I 0.4) (n = 8) Okamoto strain pair-fed controls

Mean arterial blood pressure (mm I-Is)

Heart rate (beats/ml)

129 + 3

415 + 13

123 Ifr 3

361 + 20*

178 * 3

423 + 6

tive animal model of mania, it may not be surprising that there is as yet no agreement on the necessary or sufficient neurotransmitter systems which may underlie either mania or the anti-manic effects of lithium. The present efforts were devoted to developing such a model system, since in its absence it seems clear that there will be an almost endless pursuit of amine and peptide measurements which may have little relevance to the clinical state which we seek to understand. REFERENCES

Bayon A., Shoemaker W. J., Lugo L., Azad R., Ling N.. Drucker-Colin R. R. and Bloom F. E. (1981) In aiw 162 + 4* 46of7* release of enkephalin from the globus pallidus. Neuro.sci. Lect. 24: 65-70. Ben-&hay D., Aviram A. and Viskoper R. (1975) Increased erythrocytes sodium efflux in genetic hypertensive rats of 431 + 8 179 f 3 the Hebrew University strain. Experientia 31: 660. Bloom F. E., Magistretti P. J., Shoemaker W. J., Siggins G. (n = 8) R. and Staunton D. A. (1980) Pharmacological bases of psychoneuroendocrinology: preclinical assessment of * Significant by t-test at P < 0.025. clinical facts. In: Progress in Psychoneuroendocrinolokry (Brambilla F., Racagni G. and de Wieds de D., Eds), pp. 127-137. Elsevier/North-Holland Biomedical Press. Chronic treatment with lithium (21 days; plasma Li, Bunney W. E., Jr and Garland B. L. (1983) Possible receptor effects of chronic lithium administration. Neurophar0.38 mM; brain Li, 0.34 mM) reduced blood pressure maco~ogy 22: 367-372. and increased heart rate in freely moving hypertensive Canessa hA.,Agragna N., Solomon H. S., Connolly T. M. rats. Systolic, diastolic and mean arterial blood pressand Tosteson D. C. (19801 Increased sodium-lithium ure of lithium treated rats was significantly lower countertransport in red celis of patients with essential (10-20 mmHg) than hypertensive rats fed control diet. hypertension. New Engl. J. Med. 302: 772. De Mendonca M., Grichois M., Garay R. P., Sassard J., The two groups of control hypertensive rats did not Ben&hay D. and Meyer P. (1980) Abnormal Na* and differ significantly in any of the cardiovascular parK+ fluxes in erythrocytes of three varieties of genetically ameters measured. Heart rate in the lithium treated hypertensive rats. Proc. natn. Acad. Sci. U.S.A. 77: 4283. hypertensive rats was significantly elevated when comDeGiulio A. M., Yang H.-Y. T., Fratta W. and Costa E. pared to hypertensive rats that had continuous access (1979) Decreased content of immunoreactivc enkephalin-like peptide in peripheral tissues of spontaneous to control diet. These effects of lithium appear to be hypertensive rats. Nature 278: 64&647. relatively independent of body weight and daily Folkow B. (1981) Central and peripheral mechanisms in caloric intake in that hypertensive rats fed lithium spontaneous hypertension in -rat& In: Bmin. Beizaoior had lower blood pressures than hypertensive rats in and Bodilv IDisease (Weiner H.. Hofer, M. A. and Stunkard, A. .( Eds), p. i59. Raven Press, New York. both of the control groups; the pair-fed control group had similar mean body weights and nutritional his- Friedman S. M., Nakashima M., McIndole R. A. and Friedman C. L. (1976) Increased erythrocyte pertories as the lithium-treated rats and the ad ~ib~~urn meability to Li and Na in the spontaneously hypertencontrol group weighed more than the lithium-treated sive rat. Experientia 32: 476. rats. Since the elevated blood pressure expressed in Gallagher D. W., Pert A. and Bunney W. E., Jr (1978) Haloperidol-induced presynaptic dopamine supersensitihypertensive rats is believed to be of neurogenic orivity is blocked by chronic lithium. Nature, Land. 273: gin (Folkow, 1981), lithium may exert its antihyper309-312. tensive effect through a neural mechanism. NevertheGaray R. and Meyer P. (1979) A new test showing abnorless, when normal Sprague-Dawley rats were given mal net Na+ and Kf fluxes in erythrocytes of essential hypertensive patients. Lancer 1: 349. exactly the same treatment with dietary lithium, no Gillin J. C., Hong J. S., Yang H.-Y. T. and Costa E, (1978) decrease in blood pressure was observed. although [Met’] enkephalin content in brain regions of rats slight decreases in heart rate could be seen (see treated with lithium. Proc. nafn. Acad. Sci.. U.S.A. 75: Table 3 and Koda et ai., 1981). 299i-2993. Gray P., Solomon J.. Dunphy M., Carr F. and Hession ,M. (1976) Effects of lithium on open field behavior in CONCLUSIONS “stressed” and “unstressed” rats. Ps?rhoplzuvmtrcoio(l 48: 277-28 1. The point of these observations which is most perti- Johnson F. N. (1976) Lithium effects upon components of nent to the present context of analyzing the theraactivity in rats. Experientia 32: 212-214. peutic actions of lithium is the analogy by which the Kafka M. S., Wirz-Justice A., Naber D.. Marangos P. J.. O’Donohue T. L. and Wehr T. A. (1982) Effect of lithium diseased (hypertensive) rat was employed as a model on circadian neurotransmitter receptor rhythms. Neurfor the action of lithium. In this case, the diseased rat opsychohiology 8: 41-50. showed an effect of lithium which was not detectable Koda L. Y., Shoemaker W. J., Baetge G. and Bloom F. E. in the control rat. Given that there is as yet no effec(1981) Lithium treatment decreases blood pressure in

Aspects

of the actions

of lithium

genetically hypertensive rats. Eur. J. Pharmac. 76: 411-415. Mandell A. J. (1978) Redundant mechanisms regulating brain tyrosine and tryptophan hydroxylases. A. Rev. pharmac. toxic. 18: 461-493. Mishra R. K., Marshall A. M. and Varmuza S. L. (1980) Supersensitivity in rat caudate nucleus: effects of 6-hydroxydopamine on the time course of dopamine receptor and cyclic AMP changes. Brain Res. 200: 47-57. Pert A.,.Rosenblatt J. E, Sivit C.. Pert C. B. and Bunney W. E.. Jr (1978) Lone-term treatment with lithium presents the‘development of dopamine receptor supersensitivity. Science 201: 171-173. Rosenblatt J. E.. Pert A., Layton B. and Bunney W. E., Jr (1980) Chronic lithium reduces f3H1-spiroperidol binding in rat striatum. Eur. J. Phar&c.67: 321-322. Sahakian B. J. and Robbins T. W. (1975) Potentiation of locomotor acticity and modifica$on of stereotypy by starvation in apomorphine treated rats. Neuropharmacolog!, 14: 251-257. Sahakian B. J., Robbins T. W. and Iversen S. D. (1976) a-Flupenthixol-induced hyperactivity by chronic dosing in rats. Eur. J. Pharmac. 37: 169-178. Scatchard G. (1949) The attractions of proteins for small molecules and ions. Ann. N.Y Acad. Sci. 51: 660-672. Schultz J. E.. Sienins G. R.. Schocker F. W.. Turck M. and Bloom F.‘E. Ti981) Etfects of prolonged ‘treatment with lithium and tricyclic antidepressants on discharge frequency, norepinephrine responses and beta receptor binding in rat cerebellum: Electrophysiological and biochemical comparison. J. Pharmac. exp. Ther. 216: 28-38. Siaunton D., Wolfe B. B.. Groves P. M. and Molinoff P. B.

and antidepressant

drugs

365

(1981) Dopamine receptor changes following destruction of the nigrostriatal pathway: lack of a relationship to rotational behavior. Brain Res. 211: 315-327. Staunton D. A., Magistretti P. J., Shoemaker W. J. and Bloom F. E. (1982a) Effects of chronic lithium treatment on dopamine receptors in the rat corpus striatum. I. Locomotor activity and behavioral supersensitivity. Brain Res. 232: 391-400. Staunton D. A.. Magistretti P. J., Shoemaker W. J., Deyo S. N. and Bloom F. E. (1982b) Effects of chronic lithium treatment on dopamine receptors in the rat corpus striaturn. II. No effect on denervation or neuroleptic-induced supersensitivity. Brain Res. 232: 401-412. Staunton D. A., Deyo S. N., Shoemaker W. J., Ettenberg A. and Bloom F. E. (1982~) Effects of chronic dietary lithium exposure on rat brain enkephalin systems. Sot. Neurosci. Abstr. 8: 466. Treiser S., Cascio C. S., O’Donohue T. L., Thoa N. B., Jacobowitz D. and Kellar K. J. (1981) Lithium increases serotonin release and decreases serotonin receptors in the hippocampus. Science 213: 1529-1531. Ungerstedt U. (197 la) Postsynaptic supersensitivity after 6-hydroxydopamine induced degeneration of the nigrostriatal dopamine system. Acta physiol. stand. Suppl. 367: 69-93. Ungerstedt U. (1971b) Striatal dopamine release after amphetamine or nerve denervation revealed by rotation behavior. Acta physiol. stand. Suppl. 367: 49-68. Wiley J. S., Hutchinson F. A., Mendelson 0. and Doyle A. E. (1980) Increased sodium permeability oi erythrocytes in spontaneously hypertensive rats. C’lin. exp. Pharmac. Physiol. 7: 527.