Lithium-induced polydipsia: Dependence on ... - Springer Link

Psychopharmacology

Psychopharmacology(1983) 80:143- 149

9 Springer-Verlag 1983

Lithium-Induced Polydipsia: Dependence on Nigrostriatal Dopamine Pathway and Relationship to Changes in the Renin-Angiotensin System Richard B. Mailman The Biological Sciences Research Center and Departments of Psychiatry and PharmacoIogy,University of North Carotina School of Medicine, Chapel Hill, NC 27514, USA

Abstract. The dependence of lithium-induced polydipsia (LIP) on central monoamine pathways was investigated using several pharmacological manipulations. Intracisternal administration of 6-hydroxydopamine (6-OHDA) in combination with pargyline or desipramine was used to deplete dopamine (DA), norepinephrine, or both catechotamines. Significant decreases in LIP were seen after treatments that depleted brain DA, whereas depletion of notepinephrine alone did not affect LIP. Site-specific injection of 6-OHDA into the substantia nigra or caudate nucleus, but not the nucleus accumbens or noradrenergic dorsal bundle, also caused a decrease in LIP. Depletion of serotonin by intracisternal administration of 5,7-dihydroxytryptamine also had no effect on LIP. Consistent with these findings, the DA receptor blocker haloperidol attenuated LIP. Thus, LIP appears to be dependent on intact nigrostriatal DA fibers, but not on other monoanainergic systems in the brain. Lithium also increased plasma renin activity (PRA) and angiotensin I and II immunoreactivity in plasma, though the time course of LIP onset did not directly parallel these latter changes in the renin-angiotensin axis. Neither the PRA or angiotensin II immunoreactivity in lithium-treated animals was sufficiently high to account for LIP. In addition, the 6-OHDA lesions of the caudate nucleus or substantia nigra that attenuated LIP did not affect the lithium-induced increases in PRA or in angiotensin I or II concentrations. Thus, LIP probably involves mechanisms other than just being a direct response to lithium-induced increases in PRA or angiotensin II concentration and simply may not be secondary to lithium-induced polyuria. Because of the similar pharmacological characteristics of angiotensin II and lithium-induced drinking, a role for angiotensin receptors in LIP cannot be ruled out. Key words: Lithium - Polydipsia - Nigrostriatal Dopamine - Catecholamines - Nucleus accumbens Renin - Angiotensin

mechanisms involved in lithium-induced polyuria and nephrogenic diabetes insipidus. Although vasopressin was shown to inhibit lithium-induced polydipsia (LIP) (Schreiber and RohficovS, 1971), later studies concluded that LIP occurred secondarily to alterations in renal function. Evidence for this included abolition of LIP after bilateral nephrectomy (Gutman et al. 1971), parallel changes in plasma renin activity and occurrence of LIP (Gutman et al. 1971), simultaneous occurrence of LIP and impairment of renal concentrating ability (Christensen 1974), and simultaneous attenuation of lithium-induced polyuria and LIP by ammonium chloride feeding (Smith 1974). Although the weight of accumulated evidence has suggested that LIP may be secondary to the primary polyuria, this issue is still unresolved (cf. Cox and Singer 1975). The mechanisms involved in LIP were of particular relevance to this laboratory because we had found that administration of lead to rats during postnatal development (Mailman et al. 1978), but not during adulthood (Mailman et al. 1979), resulted in permanent increases in LIP that were not dependent on continued dosing or presence of lead. These lead-treated rats had no demonstrable alteration of renal function, or differences in plasma renin activity (PRA) or plasma angiotensin I or II concentrations compared to control rats either before or after lithium administration (Mailman et al. 1978, 1979). This suggested that a development of some CNS mechanism, altered by exposure to lead, might be involved in the increased LIP. The involvement of central mechanisms in lead-induced changes in LIP was later supported by Dantzer (1980). Since angiotensin II-induced polydipsia has been reported to be mediated via dopamine (DA)-containing fibers (Fitzsimons and Setler 1971, 1975) and dopaminergic mechanisms have been reported to be altered by lead exposure during development (Govoni et al. 1978; Memo et al. 1981 ; Jason and Kellogg 1981), the present studies were designed to explore these issues further.

Materials and Methods One of the most common side effects of therapeutic use of lithium is the alteration that occurs in fluid-electrolyte homeostatic mechanisms, as manifested by polydipsia and polyuria (Schou 1957). A great deal of information has been gleaned about the actions of lithium on renal function and the Offprint requests to: R. Mailman, BiologicalSciences Research Center

220-H, University of North Carotina School of Medicine, Chapel Hill, NC 27514, USA

Catecholamine Depletions in Whole Brain. Depletions of norepinephrine, DA, or both catecholamines in brain were performed by intracisternal administration of 6-hydroxydopamine (6-OHDA) to male Sprague-Dawley rats (170190 g) obtained from Charles River Breeding Laboratories, Somerville, MA, USA: all rats were from this supplier. Two procedures were used. Specific depletions of DA were obtained by pretreating the rats with 25 mg/kg desimipramine

144 30min prior to intracisternal administration of 200gg 6O H D A (free base) in 200 gl of a 5 mg/ml ascorbic acid solution. The treatment was repeated after 7 days. Depletion of all catecholamines was accomplished by intracisternal injection of 250 jxg 6-OHDA 30 min after administration of 50mg/kg pargyline. Control animals received the vehicle intracisternally. No testing occurred prior to 1 month after treatment. The 6-OHDA-treated rats were prescreened for increased behavioral sensitivity to apomorphine. Catecholamine concentrations were determined by fluorescent methods (Breese et al. 1973), which give values significantly lower than obtained by more current techniques (Kilts et al. 1981). Selective Lesions. For norepinephrine bundle lesions, 12 gg 6-OHDA (free base) in a volume of 1 gl was administered bilaterally at two depths. The coordinates were derived from K6nig and Klippel (1963), but modified for 300-350 g rats. The upper incisor bar was 2.5 mm below the interaural line and the coordinates were as follows: anterior + 1.0 from interaural zero; lateral _+ 1.2 from interaural zero; vertical - 3 . 1 and - 1.9 from horizontal zero plane. Substantia nigra lesions were made bilaterally by administering 12gg 6O H D A (volume I gl) according to the following coordinates of K6nig and Klippel (1963): anterior + 3.4 from interaural line; lateral _+ 1.0 from interaural line; vertical - 3 . 1 from horizontal zero plane. For caudate nucleus lesions, the 6O H D A (12gg, volume 4gl at each of two depths) was administered bilaterally using the following K6nig and Klippel (1963) coordinates: anterior + 10.0 from interaural line; lateral _+ 3.0 from interaural line; vertically - 2 . 4 and - 0.4 from horizontal zero plane. Finally, nucleus accumbens lesions were made by giving 12gg 6-OHDA (4gl volume) bilaterally using the following coordinates (Pellegrino et al. 1979) : anterior - 3.4 from bregma; lateral - 1.7 from midline and or interaural line; vertical - 7 . 2 from brain surface. Placements were verified histologically in at least six trial animals. At death, catecholamine levels were determined in these animals using the method of Breese et al. (1973). The dissection procedure used yielded samples of striatum that may have included aspects of the septum and nucleus accumbens. Peripheral sympathectomies were also performed using injection of 6-OHDA by the method of Breese et al. (1973). Serotonin Depletion. Adult (170-190g) rats were injected intracisternally with 200 gg 5,7-dihydroxytryptamine (5,7DHT, as the free base) 30 min subsequent to administration of 50mg/kg pargyline IP. The 5,7-DHT was dissolved in sterile saline containing 5 mg/ml ascorbic acid and injected in volumes of 100gl. Animals were prescreened for supersensitivity to 5-methoxy-N,N-dimethyltryptamine, and serotonin depletions verified at death in representative animals. PRA and Plasma Angiotensin II. PRA was measured using radioimmunoassay reagents for angiotensin I (AI) purchased from New England Nuclear, Boston, MA, USA. Trunk blood was collected after decapitation by mixing with 200 gl 4 M EDTA buffer (pH 7.4). The samples were centrifuged in plastic tubes at 10,000 g for 15 min and the resulting plasma aspirated and immediately frozen at - 7 0 ~C. The radioimmunoassay was essentially by the method of Haber et al. (1964). A quality control sample ( P R A = 1.0+_0.2ng AI/ml/min) purchased from New England Nuclear was used in all PRA determinations. Angiotensin II (AII) concentrations in plasma were measured directly without pre-

liminary purification. A radioimmunoassay (Page et al. 1969) was performed utilizing commercial AII antisera and 12sIAII (New England Nuclear) and synthetic AII (Beckman, Palo Alto, CA, USA). Although this method gave good recovery (80 ~ - 120 ~ ) of an added internal standard of AII, it may also measure other immunoreactive material in blood (Barrett et al. 1977). In any case, this would result in overestimations of authentic AII. Animal Manipulations and Drug Treatments. The experimen, tal design .originally attempted to achieve blood lithium concentrations similar to those used clinically (i.e., 0 . 5 l m M ; Schou 1957). In preliminary experiments, this was achieved by adding lithium carbonate to the diet (2.96 mg lithium carbonate/g pulverized Wayne Lab Blox, Continental Grain Co., Chicago, IL, USA). This regimen caused significant increases in water consumption after 6 or more days of access to lithium-containing food, with serum lithium concentrations being in the desired range. However, daily IP injections were ultimately chosen for drug administration because this route offered greater control of the administered dose. With this latter route, there are sharp peaks in plasma lithium concentrations (Nelson et al. 1980). Maximal concentrations occur shortly after injection and are followed by rapid decreases. However, because there is much slower transport of lithium across the blood-brain barrier, there is much less fluctuation in the brain lithium concentrations in many brain regions (Spirtes 1976; Hildebrand and Opitz 1977; Nelson et al. 1980). Thus, Mukherjee et al. (1976) demonstrated that serum concentrations of lithium had returned essentially to control values 24 h after a single IP injection of 3 mEq/kg lithium. However, at 24 h in these same rats, lithium concentrations in various brain regions were approximately 25 ~ - 50 ~ of maximal values ( 4 - 8 h after injection). With repeated injections this difference becomes even smaller. Most convincingly, in our preliminary studies, there were similar increases in LIP using either oral or IP administration. In the experiments reported, a dose of 2mEq/kg (2 mmol/kg) lithium was administered by giving 5 ml/kg of a 0.4M solution of lithium chloride prepared in sterile deionized water. Water consumption was measured by weighing water bottles (+ 0.5 g), with a spillage of approximately 1 g occurring. This resulted primarily from seepage after inversion of the water bottles. No correction for this loss was made. Using this fluid measuring techni~tue, preliminary studies showed no difference between uninjected rats or those injected with similar volumes of saline. Prior to lithium injections (baseline periods), animals were either handled or injected with saline, with neither technique causing a significant difference in water consumption after the second day. Rats were housed individually on wire rack cages under a 12-h light-dark cycle (lights an 6 AM). Animals had continuous access to Wayne Lab Blox chow. Injections and water bottle measurements were made between 3 - 6 PM. Differences were termed significant at P < 0.05 (twotailed). Data were analyzed using analyses of variance with repeated measures. The water consumption data were manipulated by determining the excess consumption resulting from lithium treatments. For data not otherwise specified, the total quantity of water consumed after initiation of lithium treatments was divided by the number of days of treatment. From this, the average 'baseline' water consumption was subtracted yielding an average daily increase.

145 2OO

210 9~ 2 0 0

180

0 .a 160

-,~ o 180

..o

i6o

~

~" 140

rnine Depleted

Norepinephrine // Depleted~ ~

~- 140

O

12o

120

g -_>=

I00

c,,

60

$ "6

40

~>

,- I 0 0

Control (10)

O

~. 80

80

E

Depleted

,n 60 eo

20

40 eo

O

I

Fourdoy(L

Boseline Ingestion

I

I

I

I

I

I

I

0 I 2 3 4 5 6 7 DAYS AFTER FIRST LITHIUM

I I I I I I I I I Three 0 I 2 .3 4 5 6 7 8 Doy ,. DAYS AFTER LITHIUM INJECTIONS BEGAN osellne

I

8

INJECTION

Fig. l. Effects of 6-OHDA treatments on lithium-induced potydipsia (LIP). Depletions of dopamine or norepinephrine and dopamine together caused significant attenuation of total excess water consumption. Consumption on days 7 and 8 for each 6-OHDA group was also significantly attenuated. Each line represents the mean of eight animals

Fig.2. Effects of 6-OHDA-induced depletion of norepinephrine on lithium-induced polydipsia (LIP). No significant decrease in LIP was found in this experiment. The number of rats represented by each line is shown ~

Chemicals and Biochemicals. The following chemicals were used: lithium chloride and ascorbic acid (Fisher Scientific, Pittsburgh, PA, USA); lithium carbonate (Alfa Inorganics, Danvers, M A , USA); 6 - O H D A h y d r o b r o m i d e and 5,7-DHT creatinine sulfate (Regis Chem. Co., Chicago, IL, USA); 5-methoxy-N,N-dimethyltryptamine, E D T A , and pargyline (Sigma Chemicals, St. Louis, MO, USA); desmethylimipramine (USV Pharmaceuticals, Tuckahoe, NY, USA); haloperidol (McNeil Laboratories, F o r t Washington, PA, USA).

5o0 [