- Email: [email protected]
Calcium homeostasis in long-term lithium-treated women with bipolar affective disorder
Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063 – 1069
Calcium homeostasis in long-term lithium-treated women with bipolar affective disorder Aram El Khourya,*, Ulla Pettersona, Gunnar Kallnerb, ˚ berg-Wistedta, Rigmor Stain-Malmgrena Anna A a
Department of Psychiatry, Karolinska Institutet, Institution of Clinical Neuroscience, St. Goran’s Hospital, Stockholm S-112 81, Sweden b Department of Medicine at Soder Hospital, Karolinska Institutet, Stockholm, Sweden
Abstract The authors investigated the effect of long-term lithium administration on intracellular calcium mobilization. The subjects were 13 women with bipolar affective disorder stabilized on lithium and 12 matched healthy controls. Total and ionized serum calcium, intracellular calcium ion concentration, plasma parathyroid hormone (PTH) and tyrotropin (TSH), serum electrolytes and cyclic AMP (cAMP) activity in platelets were measured. The serum electrolytes sodium, potassium and creatinine and plasma PTH and TSH were all within normal ranges in patients and controls and no differences were found between the two groups. No difference was found in basal and prostaglandin E1 (PGE1)stimulated cAMP generation in platelets between patients and controls. However, total serum calcium and ionized serum calcium levels were higher in patients than in controls and there was a significant correlation between these two measures. In the patient group, serum lithium concentration correlated positively with stimulated levels of intracellular calcium in platelets. In the present study, no distinct hyperparathyroidism was found in lithium-treated patients. However, our findings indicate that lithium administration affects calcium metabolism in patients with bipolar affective disorder inducing mild hypercalcemia and a dose-dependent normalized calcium mobilization. Furthermore, our results did not support the hypothesis that lithium’s primary site of action in bipolar illness may be on signal transduction mechanisms. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Bipolar affective disorder; Intracellular calcium; Lithium; Platelets
1. Introduction Intracellular ionized calcium ([Ca2+]i) plays a role in the second messenger pathways and is thereby involved in the regulation of some neuronal mechanisms implicated in affective disorders, such as synthesis and release of neurotransmitters and regulation of receptor mechanisms (Dubowsky et al., 1992; Helmeste and Tang, 1998). Previous literature suggests that bipolar disorders may be associated with alterations of the complex mechanisms of intracellular calcium homeostasis, resulting in an elevated calcium signal (Meltzer, 1986; Dubowsky et al., 1991). One possible contribution to increased [Ca2+]i in bipolar illness could be Abbreviations: [Ca2+]r, baseline intracellular ionized calcium; cAMP, cyclic AMP; HOST, hypoosmotic shock treatment technique; [Ca2+]i, intracellular ionized calcium; PTH, parathyroid hormone; [Ca2+]s, stimulated intracellular ionized calcium; TSH, tyrotropin. * Corresponding author. Tel.: +46-8-6722488; fax: +46-8-6721995. E-mail address: [email protected] (A. El Khoury).
increased red cell and platelet intracellular sodium concentration ([Na+]i) as a result of a state-dependent decrease in the activity of the membrane sodium- and potassium-activated adenosine triphosphatase (Na+/K+ ATPase) pump. The latter may be responsible for increasing membrane excitability and decreasing neurotransmitter release, respectively (Wood, 1985; El-Mallakh and Wyatt, 1995). Another possible mechanism of increased [Ca2+]i could be reduced generation of the second messenger adenosine 30,50-cyclic monophosphate (cAMP), which has been found in lymphocytes of bipolar patients (Ebstein et al., 1988). Although the monovalent cation lithium has been widely used in the prophylaxis and treatment of bipolar affective disorder since the 1960s, its mode of action at the molecular level and the biochemical mechanisms related to its clinical effect remain unclear. An important hypothesis is that lithium may modulate the intracellular signal transduction by interfering with G proteins, adenylyl cyclases, protein kinase C isozymes and intracellular calcium transients (Avissar et al., 1988; Manji et al., 1995).
0278-5846/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 2 7 8 - 5 8 4 6 ( 0 2 ) 0 0 2 2 3 - 3
1064
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
Meltzer (1986) hypothesized that there is an overall disturbance of calcium metabolism in bipolar affective disorders and that lithium acts by reversing or counterbalancing the effects of these calcium dysfunctions. Earlier studies (Kallner and Petterson, 1995) have revealed a subgroup of lithium-treated patients with enhanced serum levels of ionized calcium (Ca2+). Other studies have also found that serum Ca2+ in lithium-treated patients was significantly higher than in controls (Bothwell et al., 1994; Komatsu et al., 1995). Animal investigations have shown a correlation between the increase in serum Ca2+ levels and the dose of lithium administered to rats (Adegboyega and Okorodudu, 1994). In addition to hypercalcemia, a variety of other complications of lithium therapy have been described, including hypothyroidism and hypoparathyroidism (Stancer and Forbath, 1989). Lithium may precipitate late-onset primary hyperparathyroidism, which is characterized by consistently elevated parathyroid hormone (PTH). Cross-sectional studies of patients on lithium reveal that serum calcium and PTH levels were elevated above the normal range anywhere from 12% to 25% of patients (Mallete and Eichorn, 1986; Kallner and Petterson, 1995). However, in many studies, calcium and PTH levels in serum stayed within the normal range in the majority of subjects (Mallete and Eichorn, 1986; Christiansen et al., 1980). Lithium has been shown to block the metabolism of the intracellular second messenger inositol-1,4,5-triphosphate (IP3), which is involved in the rise of ionized intracellular calcium (Berridge et al., 1982). Lithium was also demonstrated to increase activity of the Na+/K+ ATPase pump in bipolar patients and the entry of extracellular lithium in exchange of intracellular sodium into cells, which, in turn, decreases intracellular calcium. A decrease of intracellular calcium normalizes neuron activity in both mania and depression and is an additional possible model as to how lithium may work in the treatment and prevention of manic depression (El-Mallakh and Wyatt, 1995). Due to their morphological and pharmacological similarities with serotonergic nerve endings, blood platelets are extensively used as peripheral models in the investigations of the central serotonergic neurons in affective disorders (Stahl, 1985). It has been shown that free intracellular (cytosolic) calcium ion concentrations in platelets from patients in either phase of manic – depressive disorder were elevated (Dubowsky et al., 1994). In vitro and ex vivo studies of platelets from healthy subjects indicate that lithium does not affect the agonist-stimulated intracellular calcium mobilization, since no alteration in thrombininduced calcium mobilization was found following lithium administration (Kusumi et al., 1994). The prophylactic properties of lithium in the treatment of bipolar disorder have been suggested to be associated with its action on signal transduction mechanisms such as the second messenger generator adenylyl cyclase (Lenox and Hahn, 2000). Basal adenylate cyclase activity in
platelets was shown to be significantly elevated in shortterm lithium-treated controls (Risby et al., 1991) or slightly inhibited after preincubation with lithium chloride (Imandt et al., 1981). Functional and ultrastructural investigations have also shown that adenylyl cyclases are intimately associated with sites of calcium ion entry into the cell (Yu et al., 1993; Cooper et al., 1995) and intracellular Ca2+ has been suggested as one important regulator of particular adenylyl cyclases, at least as effective as G-protein subunits (Yoshimura and Cooper, 1992; Cali et al., 1994). The present study was undertaken to elucidate intracellular calcium mobilization in patients with bipolar affective disorder following long-term lithium administration, since our earlier studies have shown enhanced levels of extracellular calcium in lithium-treated patients.
2. Materials and methods 2.1. Subject selection Thirteen female patients with bipolar affective disorder according to DSM-IV criteria (American Psychiatry Association, 1994) were recruited to the study from the Psychiatric Clinic at St. Goran’s Hospital and Danderyd Hospital. Eleven patients were diagnosed as Bipolar I and two as Bipolar II. Seven patients had the first depressive episode at the mean age of 30 (range 25 –35 years). Ten patients had a family history of affective disorders. All patients have been treated with lithium for at least 3 years and all have been in remission for more than 1 year. No patient had abnormal plasma tyrotropin (TSH). To be included, patients should not have received any psychotropic drug (with the exception of benzodiazepines; two patients were on flunitrazepam, 0.25 – 1 mg, when required) or levothyroxine (Levaxin) during the 8 weeks preceding the study. Physical illness and signs and symptoms of drug abuse or organic mental disorder were exclusion criteria. The control group consisted of 12 healthy volunteers without heredity for affective disorders, matched for age, gender and season. The investigation was carried out in accordance with the latest version of the Declaration of Helsinki. The local ethics committee reviewed and accepted the study design. Informed consent for participation in the study was obtained from all participants after the nature of the procedures had been fully explained. 2.2. Procedures 2.2.1. Blood sampling Ninety milliliters of fasting blood was drawn from an antecubital vein early in the morning (between 0730 and 0900 h) and used as follows: 30 ml of blood for determination of basal intracellular calcium in resting platelets
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
and thrombin-induced ionized calcium in aequorin-loaded platelets using the hypoosmotic shock treatment (HOST) technique; 20 ml for measurement of cAMP levels after prostaglandin E1 (PGE1) stimulation and the rest for determination of serum levels of lithium, TSH, PTH and electrolytes. The platelet counts were obtained using an automatic cell counter (Medonic Ca 470).
1065
stopped by adding 2 ml of 99% ethanol and additional incubation for 5 min at room temperature. The supernatant was discarded and the precipitate was centrifuged in a swing-out rotor (Beckman Spinchrone R) at 160g for 5 min at room temperature and washed with 1 ml of ethanoldistilled water in a 1:2 proportion. The precipitate was centrifuged for 5 min at 1600g at room temperature and the supernatants were combined and evaporated to dryness at 55 C under a stream of nitrogen.
2.2.2. Determination of agonist-induced [Ca2+]i signals 2.2.2.1. Platelet membrane preparations and aequorin loading using the HOST technique. Thirty milliliters of blood was anticoagulated with ACD solution and PRP was prepared as described earlier. Platelets from PRP were washed twice by addition of Hepes – Tyrodes buffer A (128.4 mM NaCl, 8.9 mM NaHCO3, 2.95 mM KCl, 0.8 mM KH2PO4, 1.7 mM MgCl2, 1.0 mM Hepes, 10 mM EDTA and 1 mM PGE1, pH 7.4) and centrifugation for 20 min at 1600g at room temperature. The pellet was resuspended in 2 ml of Hepes– Tyrodes buffer A and centrifuged in a fixed-angle rotor (Eppendorf Centrifuge 5403) for 2 min at 9000g at room temperature. Mamlgren et al. (1992) showed that the HOST method presented by Vicker and Mustard (1986) was the most preferable loading technique in which aequorin is introduced in platelets. In this study, we used a slightly modified version of the HOST method described by Mamlgren et al. (1992). The platelet count was finally adjusted to 250109 cells/l. 2.2.2.2. Calibration and measurements of aequorin response. The authors used a Chrono-log Platelet Ionized Calcium Aggregometer (PICA) to measure the aequorin signal. As agonists, we used 10 ml of thrombine (bovine; Park-Davis, Trenton, NJ, USA; final concentration 0.5 U/ ml), in duplicates. The paper speed was 2 cm/min and the stirring rate was 1200 rpm/min. Calibration of the light signals and calculations of results were done as previously described (Mamlgren et al., 1992). The loss in luminescence activity of intracellular aequorin was indicated by measurement of the peak light signal. [Ca2+]i in resting platelets was determined as the difference between baseline tracings evoked by 1 ml of calcium-containing buffer with or without aequorin-loaded platelets. 2.2.3. Determination of cAMP activity in platelet-rich plasma 2.2.3.1. Platelet membrane preparation. Twenty milliliters of blood was anticoagulated with 1% (wt/vol) EDTA (0.12 ml of 0.34 M K3EDTA; Vacutainer, England, UK). PRP was obtained as described earlier. One milliliter of PRP was pipetted in four assay tubes and 100 ml of 50 mM theophyllin was added. After incubation for 5 min at 37 C, 10 ml of 50 mM PGE1 was added to two of the tubes and 10 ml of 99% ethanol was added to the other two tubes and the solutions incubated for 20 min at 37 C. The reactions were
2.2.3.2. Determination of cAMP in platelets. The residue was dissolved in 1 ml of Tris/EDTA buffer (50 mM Tris/ HCl, 4 mM EDTA, pH 7.5), centrifuged at 1600g for 5 min at room temperature to remove insoluble residues and the supernatant was directly used in the assay. Fourteen assay tubes and additional tubes for unknowns were placed in duplicate into a rack, which was kept at 0 C in an ice bath. One hundred fifty microliters of the Tris/ EDTA buffer was added to the two tubes for determination of the blank counts per minute for the assay and 50 ml of the Tris/EDTA buffer was then pipetted into the tubes for determination of binding in the absence of unlabelled cAMP. Starting with lowest level of standard cAMP, 50 ml of each dilution was added into each successive pair of assay tubes and 50 ml of each unknown in duplicate was added into the additional assay tubes as appropriate. Fifty microliters of the labeled cAMP ([8-3H]adenosine 30,50cyclic phosphate) was added to every assay tube and 100 ml of the binding protein was then pipetted into the assay tubes (except the blank ones) and to every assay tube containing an unknown. All the tubes were then vortexmixed for about 5 s and the ice bath containing the tubes was placed into a refrigerator at 2 –8 C for 2 h. One hundred microliters of the charcoal suspension was added to all tubes and the tubes were centrifuged at 1000g for 5 min in a refrigerated centrifuge to sediment the charcoal. A 200-ml sample from each tube was removed and placed in scintillation vials for counting. Total counts were determined by counting two 50-ml aliquots of titrated cAMP in scintillant containing 1 ml of water. 2.3. Data analysis Statistic analyses were done using Statview software (SAS Institute, Cary, NC, USA). Mean ± S.D. values are presented. The significance of differences in group means was determined using Student’s t test (two-tailed) for paired and grouped data. Correlations were calculated using regression analysis with ANOVA followed by Fisher’s PLSD post hoc test.
3. Results Mean ages were 43 ± 8 years (range 29– 55 years) and 43 ± 9 years (30 – 54 years) in the patient and control groups,
1066
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
respectively. In the patient group, lithium was administered in a twice-daily dosing regime. Mean dose was 181±36 mg of lithium sulphate (range 126 –252 mg), and mean serum lithium (S-Li) concentration standardized 12 h post last intake was 0.56±0.16 mmol/l (range 0.28 –0.82 mmol/l). Mean time of lithium treatment was 8±4 years (range 3 –16 years). The serum Na+, K+ and creatinine were all within normal ranges in patients and controls and no differences were found between the two groups (Table 1). Total serum calcium was found to be higher than the reference value in one patient only (2.66 versus 2.60 mmol/l). Serum ionized calcium was, on the other hand, higher than the reference value (1.25 mmol/l) in 8 of 13 patients (62%). The mean values for total serum calcium and ionized serum calcium were significantly higher in the patient group compared to the controls ( P=.049 and P=.0034, respectively) (Table 1). Even though the absolute difference between patients and controls was greater for total than for ionized serum calcium, the percent differences were similar. There was a positive correlation between total serum calcium and ionized serum calcium both in the patient group (r=.83, P=.0004) and the control group (r=.63, P=.028). The elevation of ionized serum calcium was significantly higher compared to the increase in total serum calcium in lithiumtreated patients versus controls ( P=.0013). Mean plasma TSH was within normal ranges in both groups and no differences were found between them. The
Table 1 Main characteristics of study subjects Clinical variables
Patients (n=13), mean±S.D.
Controls (n=12), mean±S.D.
P value
Total S-Ca (mM) (RV2.60 mM)a Ionized S-Ca (mM) (RV1.25 mM) [Ca2+]r (mM) [Ca2+]s (mM) Basal intracellular cAMP (pmol/106 pts)c PEG1-stimulated cAMP generation (pmol/106 pts) S-PTH (ng/l) (RV55 ng/l) S-TSH (mU/l) (RV: 0.4 – 5 mU/l) S-Na+ (mM) (RV: 135 – 147 mM) S-K+ (mM) (RV: 3.4 – 5.2 mM) S-Cr (mM) (RV: 51 – 104 mM)
2.332±0.134b
2.240±0.075
<.05
1.271±0.056b
1.215±0.045
<.005
1.99±0.72 3.32±2.13 0.017±0.005
1.79±0.78 2.93±1.15 0.013±0.005
.513 .505 .640
0.20±0.05
0.16±0.05
.126
33.5±13.4
23.7±15.5
.107
1.99±1.03
2.03±0.95
.901
a b c
139.6±1.6
139.8±1.7
Fig. 1. Correlation between lithium concentration in plasma and stimulated levels of intracellular calcium in platelets in patients.
mean plasma PTH in the patient group tended to be higher than that in the control group, but only in one patient was plasma PTH higher than the reference value (57 versus 55 ng/l). As shown in Table 1, cAMP in PGE1-stimulated platelets tended to be higher in patients than controls, but the difference was not statistically significant ( P=.13). No significant differences were found between the two groups in resting levels of intracellular ionized calcium ([Ca2+]r) or in the peak light signal for thrombin-stimulated intracellular ionized calcium concentrations ([Ca2+]s) in platelets (Table 1). In patients, there was a strong positive correlation between serum lithium concentration and [Ca2+]s (r=.75, P=.003) (Fig. 1).
4. Discussion All patients in the present study were stable on lithium for more than 1 year. According to clinical judgment of the treating physician, lithium was considered to have a good or very good stabilizing effect on each patient participating in the study, although most of them at the time of the study had relatively low lithium concentrations compared to suggested levels (above 0.7 mmol/l) for best therapeutic effect. Lithium was administered in individualized dosage regimes and the doses were adjusted and adapted to the cyclicity and state of illness for each patient, in order to achieve the best therapeutic effect with the minimum of side effects.
.836
4.1. Lithium-induced hypercalcemia 4.24±0.23
4.05±0.27
.672
74.8±4.7
79.5±9.7
.129
RV=reference values. Significantly different from control. pts=platelets.
Investigations of possible mechanisms of action of lithium have, at least in part, been focusing on changes in intracellular calcium dynamics. Hypercalcemia has been suggested to be associated with lithium treatment. In the present study, ionized serum calcium was about 50% of the
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
total serum calcium, which is considered to be within normal limits. However, both the mean total serum calcium and ionized serum calcium were significantly higher in the lithium-treated group compared to the control group. There was also a significant correlation between the two measures. These findings are consistent with the hypothesis that lithium affects calcium metabolism in depressed patients, inducing mild hypercalcemia in most lithium-treated individuals. 4.2. Detecting hypercalcemia in lithium-treated patients Hypercalcemia is established by determination of different calcium measures, such as ionized calcium and total calcium. However, measures of Ca2+ levels in the extracellular fluid do not necessarily reflect the activity of the Ca2+ that is free within the neuron, where calcium concentrations are about 104 times smaller (Cheung et al., 1986) and where minute changes that would be undetectable in the extracellular fluid can have profound effects on cellular function (Dubowsky et al., 1989). Furthermore, extracellular calcium concentration underlies regulation by parathyroidia, vitamin D, etc., whereas intracellular calcium is regulated by channels, pumps and release from intracellular stores. From cellular physiology, the functional relevance of the difference between intracellular and extracellular calcium is entirely dominated by the huge gradient in the range of 106 M due to the very low intracellular concentration of about 109 M. Changes of extracellular calcium in the millimolar range are not likely to play an important role. Although the percentage differences between total and ionized serum calcium in the present study were similar, the elevation of ionized serum calcium was higher compared to the increase in total serum calcium in lithium-treated patients versus controls. One possible explanation is that lithium induces a state of enhanced flux of ionized calcium, which is not bound to albumin. Presented results, in agreement with other reports (Adegboyega and Okorududu, 1994; Kallner et al., 1992), show that measurement of ionized serum calcium may be more sensitive than total serum calcium for detecting hypercalcemia in lithiumtreated patients.
1067
PTH levels (Fajtova et al., 1995). Renal insufficiency induced by lower glomerulus filtration rate may be associated with prolonged lithium treatment (Hetmar et al., 1991). In the present study, renal function in the lithium-treated patients, as determined by serum creatinine levels, was normal. In conformity with the findings of Komatsu et al. (1995), our results showed no distinct hyperparathyroidism, although there was a tendency toward higher PTH values in lithium-treated patients. Eight of 13 patients (62%) had elevated serum calcium and only one patient (8%) had elevated serum PTH. The differences might be due to the small sample size in the present study compared to the earlier study (13 versus 207 patients), the age differences of the patients (means 43±8 and 55±15 years, respectively) and the duration of lithium treatment (means 8±4 and 12±7 years) between the two studies. 4.4. A dose-dependent response of [Ca2+]i to lithium Available data show that both resting ([Ca2+]r) and thrombin-stimulated intracellular calcium ion concentrations ([Ca2+]s) were higher in untreated bipolar patients than controls (Tan et al., 1990; Dubowsky et al., 1992), suggesting that a dysfunction of the intracellular Ca2+ signaling mechanism may exist in patients with affective disorders. In the present study, [Ca2+]r and [Ca2+]s levels in platelets in the lithium and the control groups were comparable. We found a strong positive correlation between serum lithium concentration and [Ca2+]s levels in patients, indicating a dose-dependent response of [Ca2+]i to lithium. This finding may explain the fact that treatment with lithium, within the therapeutic interval, results in normalization of a disturbed Ca2+ signaling mechanism in patients. Our results are in agreement with the study of Adegboyega and Okorodudu (1994) performed in rat platelets. On the other hand, Kusumi et al. (1994) did not find an in vitro effect of lithium on thrombin-induced intracellular calcium mobilization ([Ca2+]s) in platelets of healthy subjects. One possible explanation is that in vivo and in vitro measurements cannot be comparable with each other. One cannot preclude that the lack of significance may also have to do with the small sample size in our study and therefore this finding must be regarded with caution.
4.3. Lithium treatment and hyperparathyroidism 4.5. Lithium and adenylate cyclase activity Several studies have suggested that treatment with lithium, in addition to its therapeutic role in psychiatric disease, has a variety of diverse effects on endocrine and electrolyte functions. These abnormalities include the development of hyperparathyroidism (Christiansen et al., 1980; Stancer and Forbath, 1989) and hypercalcemia (Kallner and Petterson, 1995; Komatsu et al., 1995). In an earlier study by two of us (G.K. and U.P.), elevated serum calcium was found in 25% and elevated serum PTH in 23% of the patients (Kallner and Petterson, 1995). Elevated ionized and total serum calcium and mild renal failure are conditions leading to elevated
Conflicting data concerning the effects of lithium on adenylate cyclase activity in human platelets have been published. A reduction (Ebstein et al., 1988) and no change (Newman et al., 1992) in adenylate cyclase activity under basal conditions and after PGE1 stimulation were reported in platelets from depressed patients treated with lithium compared to controls. However, Risby et al. (1991) showed an augmentation in basal, GppNHp and fluoride-stimulated activity in 10 normal volunteers before and after lithium administration. The differences in the results may be
1068
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
explained by the differences in the methodology used for these studies as well as the sample sizes and the categories of the populations. In the present study, in agreement with Newman et al. (1992), we found no change in adenylate cyclase activity measured under basal conditions and after PGE1 stimulation in platelets from patients compared to controls. This finding does not support the hypothesis that lithium’s primary site of action in bipolar illness may be on signal transduction mechanisms.
5. Conclusions In the present study, no distinct hyperparathyroidism was found in long-term lithium-treated female patients. However, our findings indicate that lithium administration affects calcium metabolism in patients with bipolar affective disorder, inducing mild hypercalcemia and a dose-dependent normalized calcium mobilization. Furthermore, our results did not support the hypothesis that lithium’s primary site of action in bipolar illness may be on signal transduction mechanisms.
Acknowledgements The authors acknowledge Karin Hjerpe, Cecilia Falker and Anastasia Markou for skilled technical assistance. This study was supported by grants from Swedish Medical Research Council Projects (nos. 012253-03A and 10414) and the Karolinska Institutet.
References Adegboyega, P.A., Okorodudu, A.O., 1994. Intracellular ionized calcium and increasing doses of lithium chloride therapy in healthy Sprague – Dawley rat. Pharmacol. Biochem. Behav. 49, 1087 – 1091. American Psychiatry Association, 1994. Committee on Nomenclature and Statistics. Diagnostic and Statistical Manual of Mental Disorder, fourth ed. American Psychiatric Press, Inc., Washington, DC. Avissar, S., Screiber, G., Danon, A., Belmarker, R.H., 1988. Lithium inhibits adrenergic and cholinergic increases in GTP binding in rat cortex. Nature 331, 440 – 442. Berridge, M.J., Downes, C.P., Hanely, M.R., 1982. Lithium amplifies agonist-dependent phosphatidylinositol responses in brain and salivary glands. Biochem. J. 206, 587 – 595. Bothwell, R.A., Ecceleston, D., Marschall, E., 1994. Platelet intracellular calcium in patients with recurrent affective disorders. Psychopharmacology 114, 375 – 381. Cali, J.J., Zwagstra, J.C., Mons, N., Cooper, D.M., Krupinski, J., 1994. Type VIII adenylyl cyclase. A Ca2+/calmodulin-stimulated enzyme expressed in discrete regions of rat brain. J. Biol. Chem. 269, 12190 – 12195. Cheung, J.Y., Bonventre, J.V., Malis, C.D., Leaf, A., 1986. Calcium and ischemic injury. N. Engl. J. Med. 314, 1670 – 1676. Christiansen, C., Baastrup, P.C., Transbol, I., 1980. Development of ‘‘primary’’ hyperparathyroidism during lithium therapy: longitudinal study. Neuropsychobiology 6, 280 – 283. Cooper, D.M., Mons, N., Karpen, J.W., 1995. Adenylyl cyclases and the interaction between calcium and cAMP signaling. Nature 374, 421 – 424.
Dubowsky, S.L., Christiano, J., Daniell, L.C., Franks, R.D., Murphy, J., Adler, L., Baker, N., Harris, R.A., 1989. Increased platelet intracellular calcium concentration in patients with bipolar affective disorders. Arch. Gen. Psychiatry 46, 632 – 638. Dubowsky, S.L., Lee, C., Christiano, J., Murphy, J., 1991. Elevated intracellular calcium ion concentration in bipolar depression. Biol. Psychiatry 29, 441 – 450. Dubowsky, S.L., Murphy, J., Christiano, J., Lee, C., 1992. The calcium second messenger system in bipolar disorders: data supporting new research directions. J. Neuropsychiatry Clin. Neurosci. 4, 3 – 14. Dubowsky, S.L., Thomas, M., Hijazi, A., Murphy, J., 1994. Intracellular calcium signalling in peripheral cells of patients with bipolar affective disorder. Eur. Arch. Psychiatry Clin. Neurosci. 243, 229 – 234. Ebstein, R.P., Lerer, B., Bennett, E.R., Shapira, B., Kindler, S., Shemesh, Z., Gerstenhaber, N., 1988. Lithium modulation of second messenger signal amplification in man: inhibition of phosphatidylinositol-specific phosphatase C and adenylate cyclase activity. Psychiatry Res. 24, 45 – 52. El-Mallakh, R.S., Wyatt, R.J., 1995. The Na,K-ATPase hypothesis for bipolar illness. Biol. Psychiatry 37, 235 – 244. Fajtova, V.T., Sayegh, M.H., Hickey, N., Alibadi, P., Lazarus, J.M., Leboff, M.S., 1995. Intact parathyroid hormone levels in renal insufficiency. Calcif. Tissue Int. 57, 329 – 335. Helmeste, D.M., Tang, S.W., 1998. The role of calcium in the etiology of the affective disorders. Jpn. J. Pharmacol. 77, 107 – 116. Hetmar, O., Povlsen, U.J., Ladefoged, J., Bolwig, T.G., 1991. Lithium: long-term effects on the kidney. A prospective follow-up study ten years after kidney biopsy. Br. J. Psychiatry 158, 53 – 58. Imandt, L., Tijhuis, D., Wessels, H., Haanen, C., 1981. Lithium inhibits adenylate cyclase of human platelets. Thromb. Haemostasis 45, 142 – 145. Kallner, G., Petterson, U., 1995. Renal, thyroid and parathyroid function during lithium treatment: laboratory tests in 207 people treated for 1 – 30 years. Acta Psychiatr. Scand. 91, 48 – 51. Kallner, G., Lindelius, R., Mathe´, A.A., Petterson, U., Nilsson, C.-G., 1992. Hypercalcemia in lithium treatment—an underestimated complication? La¨kartidningen 89, 4163 – 4164. Komatsu, M., Shimizu, H., Tsuruta, T., Kato, M., Fushimi, T., Inoue, K., Kobayashi, S., Kuroda, T., 1995. Effect of lithium on serum calcium level and parathyroid function in manic – depressive disorder. Endocr. J. 42, 691 – 695. Kusumi, I., Koyama, T., Yamashita, I., 1994. Effect of mood stabilizing agents on agonist-induced calcium mobilization in human platelets. J. Psychiatry Neurosci. 19, 222 – 225. Lenox, R.H., Hahn, C.G., 2000. Overview of the mechanism of action of lithium in the brain: fifty-year update. J. Clin. Psychiatry 61 (Suppl. 9), 5 – 15. Mallete, L.E., Eichorn, E., 1986. Effects of lithium carbonate on human calcium metabolism. Arch. Intern. Med. 146, 770 – 776. Mamlgren, R., Grunfelt, S., Joseph, R., 1992. On the significance of different aequorin loading techniques on intracellular aequorin discharge, baseline calcium, platelet aggregation and aequorin-indicated Ca(2+)transients. Thromb. Haemostasis 68, 352 – 356. Manji, H.K., Potter, W.Z., Lenox, R.H., 1995. Signal transduction pathways. Molecular targets for lithium’s actions. Arch. Gen. Psychiatry 52, 531 – 543. Meltzer, H.L., 1986. Lithium mechanisms in bipolar illness and altered intracellular calcium function. Biol. Psychiatry 21, 492 – 510. Newman, M.E., Lerer, B., Lichenberg, P., Shapira, B., 1992. Platelet adenylate cyclase activity in depression and after clomipramine and lithium treatment: relation to serotonergic treatment. Psychopharmacology 109, 231 – 234. Risby, E.D., Hsiao, J.K., Manji, H.K., Britan, J., Moses, F., Zhou, D.F., Potter, W.Z., 1991. The mechanisms of action of lithium: II. Effects on adenylate cyclase activity and beta-adrenergic receptor binding in normal subjects. Arch. Gen. Psychiatry 48, 513 – 524. Stahl, S.M., 1985. Peripheral models for the study of neurotransmitter receptors in man. Psychopharmacol. Bull. 21, 663 – 671.
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069 Stancer, H.C., Forbath, N., 1989. Hyperparathyroidism, hypothyroidism and impaired renal function after 10 to 20 years of lithium treatment. Arch. Intern. Med. 149, 1042 – 1045. Tan, C.H., Javros, M.A., Seleshi, E., Lowrimore, P.A., Bowden, C.L., 1990. Effects of lithium on platelet ionic intracellular calcium concentration in patients with bipolar (manic – depressive) disorder and healthy controls. Life Sci. 46, 1175 – 1180. Vicker, J.D., Mustard, J.F., 1986. The phosphoinositides exist in multiple metabolic pools in rabbit platelets. Biochem. J. 238, 411 – 417.
1069
Wood, K., 1985. The neurochemistry of mania. The effect of lithium on catecholamines, indolamines and calcium mobilization. J. Affective Disord. 8, 215 – 223. Yoshimura, M., Cooper, D.M., 1992. Cloning and expression of Ca(2+)inhibitable adenylyl cyclase from NCB-20 cells. Proc. Natl. Acad. Sci. USA 89, 6716 – 6720. Yu, H.J., Ma, H., Green, R.D., 1993. Calcium entry via L-type calcium channels acts as a negative regulator of adenylyl cyclase activity and cyclic AMP levels in cardiac myocytes. Mol. Pharmacol. 44, 689 – 693.
Calcium homeostasis in long-term lithium-treated women with bipolar affective disorder Aram El Khourya,*, Ulla Pettersona, Gunnar Kallnerb, ˚ berg-Wistedta, Rigmor Stain-Malmgrena Anna A a
Department of Psychiatry, Karolinska Institutet, Institution of Clinical Neuroscience, St. Goran’s Hospital, Stockholm S-112 81, Sweden b Department of Medicine at Soder Hospital, Karolinska Institutet, Stockholm, Sweden
Abstract The authors investigated the effect of long-term lithium administration on intracellular calcium mobilization. The subjects were 13 women with bipolar affective disorder stabilized on lithium and 12 matched healthy controls. Total and ionized serum calcium, intracellular calcium ion concentration, plasma parathyroid hormone (PTH) and tyrotropin (TSH), serum electrolytes and cyclic AMP (cAMP) activity in platelets were measured. The serum electrolytes sodium, potassium and creatinine and plasma PTH and TSH were all within normal ranges in patients and controls and no differences were found between the two groups. No difference was found in basal and prostaglandin E1 (PGE1)stimulated cAMP generation in platelets between patients and controls. However, total serum calcium and ionized serum calcium levels were higher in patients than in controls and there was a significant correlation between these two measures. In the patient group, serum lithium concentration correlated positively with stimulated levels of intracellular calcium in platelets. In the present study, no distinct hyperparathyroidism was found in lithium-treated patients. However, our findings indicate that lithium administration affects calcium metabolism in patients with bipolar affective disorder inducing mild hypercalcemia and a dose-dependent normalized calcium mobilization. Furthermore, our results did not support the hypothesis that lithium’s primary site of action in bipolar illness may be on signal transduction mechanisms. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Bipolar affective disorder; Intracellular calcium; Lithium; Platelets
1. Introduction Intracellular ionized calcium ([Ca2+]i) plays a role in the second messenger pathways and is thereby involved in the regulation of some neuronal mechanisms implicated in affective disorders, such as synthesis and release of neurotransmitters and regulation of receptor mechanisms (Dubowsky et al., 1992; Helmeste and Tang, 1998). Previous literature suggests that bipolar disorders may be associated with alterations of the complex mechanisms of intracellular calcium homeostasis, resulting in an elevated calcium signal (Meltzer, 1986; Dubowsky et al., 1991). One possible contribution to increased [Ca2+]i in bipolar illness could be Abbreviations: [Ca2+]r, baseline intracellular ionized calcium; cAMP, cyclic AMP; HOST, hypoosmotic shock treatment technique; [Ca2+]i, intracellular ionized calcium; PTH, parathyroid hormone; [Ca2+]s, stimulated intracellular ionized calcium; TSH, tyrotropin. * Corresponding author. Tel.: +46-8-6722488; fax: +46-8-6721995. E-mail address: [email protected] (A. El Khoury).
increased red cell and platelet intracellular sodium concentration ([Na+]i) as a result of a state-dependent decrease in the activity of the membrane sodium- and potassium-activated adenosine triphosphatase (Na+/K+ ATPase) pump. The latter may be responsible for increasing membrane excitability and decreasing neurotransmitter release, respectively (Wood, 1985; El-Mallakh and Wyatt, 1995). Another possible mechanism of increased [Ca2+]i could be reduced generation of the second messenger adenosine 30,50-cyclic monophosphate (cAMP), which has been found in lymphocytes of bipolar patients (Ebstein et al., 1988). Although the monovalent cation lithium has been widely used in the prophylaxis and treatment of bipolar affective disorder since the 1960s, its mode of action at the molecular level and the biochemical mechanisms related to its clinical effect remain unclear. An important hypothesis is that lithium may modulate the intracellular signal transduction by interfering with G proteins, adenylyl cyclases, protein kinase C isozymes and intracellular calcium transients (Avissar et al., 1988; Manji et al., 1995).
0278-5846/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 2 7 8 - 5 8 4 6 ( 0 2 ) 0 0 2 2 3 - 3
1064
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
Meltzer (1986) hypothesized that there is an overall disturbance of calcium metabolism in bipolar affective disorders and that lithium acts by reversing or counterbalancing the effects of these calcium dysfunctions. Earlier studies (Kallner and Petterson, 1995) have revealed a subgroup of lithium-treated patients with enhanced serum levels of ionized calcium (Ca2+). Other studies have also found that serum Ca2+ in lithium-treated patients was significantly higher than in controls (Bothwell et al., 1994; Komatsu et al., 1995). Animal investigations have shown a correlation between the increase in serum Ca2+ levels and the dose of lithium administered to rats (Adegboyega and Okorodudu, 1994). In addition to hypercalcemia, a variety of other complications of lithium therapy have been described, including hypothyroidism and hypoparathyroidism (Stancer and Forbath, 1989). Lithium may precipitate late-onset primary hyperparathyroidism, which is characterized by consistently elevated parathyroid hormone (PTH). Cross-sectional studies of patients on lithium reveal that serum calcium and PTH levels were elevated above the normal range anywhere from 12% to 25% of patients (Mallete and Eichorn, 1986; Kallner and Petterson, 1995). However, in many studies, calcium and PTH levels in serum stayed within the normal range in the majority of subjects (Mallete and Eichorn, 1986; Christiansen et al., 1980). Lithium has been shown to block the metabolism of the intracellular second messenger inositol-1,4,5-triphosphate (IP3), which is involved in the rise of ionized intracellular calcium (Berridge et al., 1982). Lithium was also demonstrated to increase activity of the Na+/K+ ATPase pump in bipolar patients and the entry of extracellular lithium in exchange of intracellular sodium into cells, which, in turn, decreases intracellular calcium. A decrease of intracellular calcium normalizes neuron activity in both mania and depression and is an additional possible model as to how lithium may work in the treatment and prevention of manic depression (El-Mallakh and Wyatt, 1995). Due to their morphological and pharmacological similarities with serotonergic nerve endings, blood platelets are extensively used as peripheral models in the investigations of the central serotonergic neurons in affective disorders (Stahl, 1985). It has been shown that free intracellular (cytosolic) calcium ion concentrations in platelets from patients in either phase of manic – depressive disorder were elevated (Dubowsky et al., 1994). In vitro and ex vivo studies of platelets from healthy subjects indicate that lithium does not affect the agonist-stimulated intracellular calcium mobilization, since no alteration in thrombininduced calcium mobilization was found following lithium administration (Kusumi et al., 1994). The prophylactic properties of lithium in the treatment of bipolar disorder have been suggested to be associated with its action on signal transduction mechanisms such as the second messenger generator adenylyl cyclase (Lenox and Hahn, 2000). Basal adenylate cyclase activity in
platelets was shown to be significantly elevated in shortterm lithium-treated controls (Risby et al., 1991) or slightly inhibited after preincubation with lithium chloride (Imandt et al., 1981). Functional and ultrastructural investigations have also shown that adenylyl cyclases are intimately associated with sites of calcium ion entry into the cell (Yu et al., 1993; Cooper et al., 1995) and intracellular Ca2+ has been suggested as one important regulator of particular adenylyl cyclases, at least as effective as G-protein subunits (Yoshimura and Cooper, 1992; Cali et al., 1994). The present study was undertaken to elucidate intracellular calcium mobilization in patients with bipolar affective disorder following long-term lithium administration, since our earlier studies have shown enhanced levels of extracellular calcium in lithium-treated patients.
2. Materials and methods 2.1. Subject selection Thirteen female patients with bipolar affective disorder according to DSM-IV criteria (American Psychiatry Association, 1994) were recruited to the study from the Psychiatric Clinic at St. Goran’s Hospital and Danderyd Hospital. Eleven patients were diagnosed as Bipolar I and two as Bipolar II. Seven patients had the first depressive episode at the mean age of 30 (range 25 –35 years). Ten patients had a family history of affective disorders. All patients have been treated with lithium for at least 3 years and all have been in remission for more than 1 year. No patient had abnormal plasma tyrotropin (TSH). To be included, patients should not have received any psychotropic drug (with the exception of benzodiazepines; two patients were on flunitrazepam, 0.25 – 1 mg, when required) or levothyroxine (Levaxin) during the 8 weeks preceding the study. Physical illness and signs and symptoms of drug abuse or organic mental disorder were exclusion criteria. The control group consisted of 12 healthy volunteers without heredity for affective disorders, matched for age, gender and season. The investigation was carried out in accordance with the latest version of the Declaration of Helsinki. The local ethics committee reviewed and accepted the study design. Informed consent for participation in the study was obtained from all participants after the nature of the procedures had been fully explained. 2.2. Procedures 2.2.1. Blood sampling Ninety milliliters of fasting blood was drawn from an antecubital vein early in the morning (between 0730 and 0900 h) and used as follows: 30 ml of blood for determination of basal intracellular calcium in resting platelets
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
and thrombin-induced ionized calcium in aequorin-loaded platelets using the hypoosmotic shock treatment (HOST) technique; 20 ml for measurement of cAMP levels after prostaglandin E1 (PGE1) stimulation and the rest for determination of serum levels of lithium, TSH, PTH and electrolytes. The platelet counts were obtained using an automatic cell counter (Medonic Ca 470).
1065
stopped by adding 2 ml of 99% ethanol and additional incubation for 5 min at room temperature. The supernatant was discarded and the precipitate was centrifuged in a swing-out rotor (Beckman Spinchrone R) at 160g for 5 min at room temperature and washed with 1 ml of ethanoldistilled water in a 1:2 proportion. The precipitate was centrifuged for 5 min at 1600g at room temperature and the supernatants were combined and evaporated to dryness at 55 C under a stream of nitrogen.
2.2.2. Determination of agonist-induced [Ca2+]i signals 2.2.2.1. Platelet membrane preparations and aequorin loading using the HOST technique. Thirty milliliters of blood was anticoagulated with ACD solution and PRP was prepared as described earlier. Platelets from PRP were washed twice by addition of Hepes – Tyrodes buffer A (128.4 mM NaCl, 8.9 mM NaHCO3, 2.95 mM KCl, 0.8 mM KH2PO4, 1.7 mM MgCl2, 1.0 mM Hepes, 10 mM EDTA and 1 mM PGE1, pH 7.4) and centrifugation for 20 min at 1600g at room temperature. The pellet was resuspended in 2 ml of Hepes– Tyrodes buffer A and centrifuged in a fixed-angle rotor (Eppendorf Centrifuge 5403) for 2 min at 9000g at room temperature. Mamlgren et al. (1992) showed that the HOST method presented by Vicker and Mustard (1986) was the most preferable loading technique in which aequorin is introduced in platelets. In this study, we used a slightly modified version of the HOST method described by Mamlgren et al. (1992). The platelet count was finally adjusted to 250109 cells/l. 2.2.2.2. Calibration and measurements of aequorin response. The authors used a Chrono-log Platelet Ionized Calcium Aggregometer (PICA) to measure the aequorin signal. As agonists, we used 10 ml of thrombine (bovine; Park-Davis, Trenton, NJ, USA; final concentration 0.5 U/ ml), in duplicates. The paper speed was 2 cm/min and the stirring rate was 1200 rpm/min. Calibration of the light signals and calculations of results were done as previously described (Mamlgren et al., 1992). The loss in luminescence activity of intracellular aequorin was indicated by measurement of the peak light signal. [Ca2+]i in resting platelets was determined as the difference between baseline tracings evoked by 1 ml of calcium-containing buffer with or without aequorin-loaded platelets. 2.2.3. Determination of cAMP activity in platelet-rich plasma 2.2.3.1. Platelet membrane preparation. Twenty milliliters of blood was anticoagulated with 1% (wt/vol) EDTA (0.12 ml of 0.34 M K3EDTA; Vacutainer, England, UK). PRP was obtained as described earlier. One milliliter of PRP was pipetted in four assay tubes and 100 ml of 50 mM theophyllin was added. After incubation for 5 min at 37 C, 10 ml of 50 mM PGE1 was added to two of the tubes and 10 ml of 99% ethanol was added to the other two tubes and the solutions incubated for 20 min at 37 C. The reactions were
2.2.3.2. Determination of cAMP in platelets. The residue was dissolved in 1 ml of Tris/EDTA buffer (50 mM Tris/ HCl, 4 mM EDTA, pH 7.5), centrifuged at 1600g for 5 min at room temperature to remove insoluble residues and the supernatant was directly used in the assay. Fourteen assay tubes and additional tubes for unknowns were placed in duplicate into a rack, which was kept at 0 C in an ice bath. One hundred fifty microliters of the Tris/ EDTA buffer was added to the two tubes for determination of the blank counts per minute for the assay and 50 ml of the Tris/EDTA buffer was then pipetted into the tubes for determination of binding in the absence of unlabelled cAMP. Starting with lowest level of standard cAMP, 50 ml of each dilution was added into each successive pair of assay tubes and 50 ml of each unknown in duplicate was added into the additional assay tubes as appropriate. Fifty microliters of the labeled cAMP ([8-3H]adenosine 30,50cyclic phosphate) was added to every assay tube and 100 ml of the binding protein was then pipetted into the assay tubes (except the blank ones) and to every assay tube containing an unknown. All the tubes were then vortexmixed for about 5 s and the ice bath containing the tubes was placed into a refrigerator at 2 –8 C for 2 h. One hundred microliters of the charcoal suspension was added to all tubes and the tubes were centrifuged at 1000g for 5 min in a refrigerated centrifuge to sediment the charcoal. A 200-ml sample from each tube was removed and placed in scintillation vials for counting. Total counts were determined by counting two 50-ml aliquots of titrated cAMP in scintillant containing 1 ml of water. 2.3. Data analysis Statistic analyses were done using Statview software (SAS Institute, Cary, NC, USA). Mean ± S.D. values are presented. The significance of differences in group means was determined using Student’s t test (two-tailed) for paired and grouped data. Correlations were calculated using regression analysis with ANOVA followed by Fisher’s PLSD post hoc test.
3. Results Mean ages were 43 ± 8 years (range 29– 55 years) and 43 ± 9 years (30 – 54 years) in the patient and control groups,
1066
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
respectively. In the patient group, lithium was administered in a twice-daily dosing regime. Mean dose was 181±36 mg of lithium sulphate (range 126 –252 mg), and mean serum lithium (S-Li) concentration standardized 12 h post last intake was 0.56±0.16 mmol/l (range 0.28 –0.82 mmol/l). Mean time of lithium treatment was 8±4 years (range 3 –16 years). The serum Na+, K+ and creatinine were all within normal ranges in patients and controls and no differences were found between the two groups (Table 1). Total serum calcium was found to be higher than the reference value in one patient only (2.66 versus 2.60 mmol/l). Serum ionized calcium was, on the other hand, higher than the reference value (1.25 mmol/l) in 8 of 13 patients (62%). The mean values for total serum calcium and ionized serum calcium were significantly higher in the patient group compared to the controls ( P=.049 and P=.0034, respectively) (Table 1). Even though the absolute difference between patients and controls was greater for total than for ionized serum calcium, the percent differences were similar. There was a positive correlation between total serum calcium and ionized serum calcium both in the patient group (r=.83, P=.0004) and the control group (r=.63, P=.028). The elevation of ionized serum calcium was significantly higher compared to the increase in total serum calcium in lithiumtreated patients versus controls ( P=.0013). Mean plasma TSH was within normal ranges in both groups and no differences were found between them. The
Table 1 Main characteristics of study subjects Clinical variables
Patients (n=13), mean±S.D.
Controls (n=12), mean±S.D.
P value
Total S-Ca (mM) (RV2.60 mM)a Ionized S-Ca (mM) (RV1.25 mM) [Ca2+]r (mM) [Ca2+]s (mM) Basal intracellular cAMP (pmol/106 pts)c PEG1-stimulated cAMP generation (pmol/106 pts) S-PTH (ng/l) (RV55 ng/l) S-TSH (mU/l) (RV: 0.4 – 5 mU/l) S-Na+ (mM) (RV: 135 – 147 mM) S-K+ (mM) (RV: 3.4 – 5.2 mM) S-Cr (mM) (RV: 51 – 104 mM)
2.332±0.134b
2.240±0.075
<.05
1.271±0.056b
1.215±0.045
<.005
1.99±0.72 3.32±2.13 0.017±0.005
1.79±0.78 2.93±1.15 0.013±0.005
.513 .505 .640
0.20±0.05
0.16±0.05
.126
33.5±13.4
23.7±15.5
.107
1.99±1.03
2.03±0.95
.901
a b c
139.6±1.6
139.8±1.7
Fig. 1. Correlation between lithium concentration in plasma and stimulated levels of intracellular calcium in platelets in patients.
mean plasma PTH in the patient group tended to be higher than that in the control group, but only in one patient was plasma PTH higher than the reference value (57 versus 55 ng/l). As shown in Table 1, cAMP in PGE1-stimulated platelets tended to be higher in patients than controls, but the difference was not statistically significant ( P=.13). No significant differences were found between the two groups in resting levels of intracellular ionized calcium ([Ca2+]r) or in the peak light signal for thrombin-stimulated intracellular ionized calcium concentrations ([Ca2+]s) in platelets (Table 1). In patients, there was a strong positive correlation between serum lithium concentration and [Ca2+]s (r=.75, P=.003) (Fig. 1).
4. Discussion All patients in the present study were stable on lithium for more than 1 year. According to clinical judgment of the treating physician, lithium was considered to have a good or very good stabilizing effect on each patient participating in the study, although most of them at the time of the study had relatively low lithium concentrations compared to suggested levels (above 0.7 mmol/l) for best therapeutic effect. Lithium was administered in individualized dosage regimes and the doses were adjusted and adapted to the cyclicity and state of illness for each patient, in order to achieve the best therapeutic effect with the minimum of side effects.
.836
4.1. Lithium-induced hypercalcemia 4.24±0.23
4.05±0.27
.672
74.8±4.7
79.5±9.7
.129
RV=reference values. Significantly different from control. pts=platelets.
Investigations of possible mechanisms of action of lithium have, at least in part, been focusing on changes in intracellular calcium dynamics. Hypercalcemia has been suggested to be associated with lithium treatment. In the present study, ionized serum calcium was about 50% of the
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
total serum calcium, which is considered to be within normal limits. However, both the mean total serum calcium and ionized serum calcium were significantly higher in the lithium-treated group compared to the control group. There was also a significant correlation between the two measures. These findings are consistent with the hypothesis that lithium affects calcium metabolism in depressed patients, inducing mild hypercalcemia in most lithium-treated individuals. 4.2. Detecting hypercalcemia in lithium-treated patients Hypercalcemia is established by determination of different calcium measures, such as ionized calcium and total calcium. However, measures of Ca2+ levels in the extracellular fluid do not necessarily reflect the activity of the Ca2+ that is free within the neuron, where calcium concentrations are about 104 times smaller (Cheung et al., 1986) and where minute changes that would be undetectable in the extracellular fluid can have profound effects on cellular function (Dubowsky et al., 1989). Furthermore, extracellular calcium concentration underlies regulation by parathyroidia, vitamin D, etc., whereas intracellular calcium is regulated by channels, pumps and release from intracellular stores. From cellular physiology, the functional relevance of the difference between intracellular and extracellular calcium is entirely dominated by the huge gradient in the range of 106 M due to the very low intracellular concentration of about 109 M. Changes of extracellular calcium in the millimolar range are not likely to play an important role. Although the percentage differences between total and ionized serum calcium in the present study were similar, the elevation of ionized serum calcium was higher compared to the increase in total serum calcium in lithium-treated patients versus controls. One possible explanation is that lithium induces a state of enhanced flux of ionized calcium, which is not bound to albumin. Presented results, in agreement with other reports (Adegboyega and Okorududu, 1994; Kallner et al., 1992), show that measurement of ionized serum calcium may be more sensitive than total serum calcium for detecting hypercalcemia in lithiumtreated patients.
1067
PTH levels (Fajtova et al., 1995). Renal insufficiency induced by lower glomerulus filtration rate may be associated with prolonged lithium treatment (Hetmar et al., 1991). In the present study, renal function in the lithium-treated patients, as determined by serum creatinine levels, was normal. In conformity with the findings of Komatsu et al. (1995), our results showed no distinct hyperparathyroidism, although there was a tendency toward higher PTH values in lithium-treated patients. Eight of 13 patients (62%) had elevated serum calcium and only one patient (8%) had elevated serum PTH. The differences might be due to the small sample size in the present study compared to the earlier study (13 versus 207 patients), the age differences of the patients (means 43±8 and 55±15 years, respectively) and the duration of lithium treatment (means 8±4 and 12±7 years) between the two studies. 4.4. A dose-dependent response of [Ca2+]i to lithium Available data show that both resting ([Ca2+]r) and thrombin-stimulated intracellular calcium ion concentrations ([Ca2+]s) were higher in untreated bipolar patients than controls (Tan et al., 1990; Dubowsky et al., 1992), suggesting that a dysfunction of the intracellular Ca2+ signaling mechanism may exist in patients with affective disorders. In the present study, [Ca2+]r and [Ca2+]s levels in platelets in the lithium and the control groups were comparable. We found a strong positive correlation between serum lithium concentration and [Ca2+]s levels in patients, indicating a dose-dependent response of [Ca2+]i to lithium. This finding may explain the fact that treatment with lithium, within the therapeutic interval, results in normalization of a disturbed Ca2+ signaling mechanism in patients. Our results are in agreement with the study of Adegboyega and Okorodudu (1994) performed in rat platelets. On the other hand, Kusumi et al. (1994) did not find an in vitro effect of lithium on thrombin-induced intracellular calcium mobilization ([Ca2+]s) in platelets of healthy subjects. One possible explanation is that in vivo and in vitro measurements cannot be comparable with each other. One cannot preclude that the lack of significance may also have to do with the small sample size in our study and therefore this finding must be regarded with caution.
4.3. Lithium treatment and hyperparathyroidism 4.5. Lithium and adenylate cyclase activity Several studies have suggested that treatment with lithium, in addition to its therapeutic role in psychiatric disease, has a variety of diverse effects on endocrine and electrolyte functions. These abnormalities include the development of hyperparathyroidism (Christiansen et al., 1980; Stancer and Forbath, 1989) and hypercalcemia (Kallner and Petterson, 1995; Komatsu et al., 1995). In an earlier study by two of us (G.K. and U.P.), elevated serum calcium was found in 25% and elevated serum PTH in 23% of the patients (Kallner and Petterson, 1995). Elevated ionized and total serum calcium and mild renal failure are conditions leading to elevated
Conflicting data concerning the effects of lithium on adenylate cyclase activity in human platelets have been published. A reduction (Ebstein et al., 1988) and no change (Newman et al., 1992) in adenylate cyclase activity under basal conditions and after PGE1 stimulation were reported in platelets from depressed patients treated with lithium compared to controls. However, Risby et al. (1991) showed an augmentation in basal, GppNHp and fluoride-stimulated activity in 10 normal volunteers before and after lithium administration. The differences in the results may be
1068
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069
explained by the differences in the methodology used for these studies as well as the sample sizes and the categories of the populations. In the present study, in agreement with Newman et al. (1992), we found no change in adenylate cyclase activity measured under basal conditions and after PGE1 stimulation in platelets from patients compared to controls. This finding does not support the hypothesis that lithium’s primary site of action in bipolar illness may be on signal transduction mechanisms.
5. Conclusions In the present study, no distinct hyperparathyroidism was found in long-term lithium-treated female patients. However, our findings indicate that lithium administration affects calcium metabolism in patients with bipolar affective disorder, inducing mild hypercalcemia and a dose-dependent normalized calcium mobilization. Furthermore, our results did not support the hypothesis that lithium’s primary site of action in bipolar illness may be on signal transduction mechanisms.
Acknowledgements The authors acknowledge Karin Hjerpe, Cecilia Falker and Anastasia Markou for skilled technical assistance. This study was supported by grants from Swedish Medical Research Council Projects (nos. 012253-03A and 10414) and the Karolinska Institutet.
References Adegboyega, P.A., Okorodudu, A.O., 1994. Intracellular ionized calcium and increasing doses of lithium chloride therapy in healthy Sprague – Dawley rat. Pharmacol. Biochem. Behav. 49, 1087 – 1091. American Psychiatry Association, 1994. Committee on Nomenclature and Statistics. Diagnostic and Statistical Manual of Mental Disorder, fourth ed. American Psychiatric Press, Inc., Washington, DC. Avissar, S., Screiber, G., Danon, A., Belmarker, R.H., 1988. Lithium inhibits adrenergic and cholinergic increases in GTP binding in rat cortex. Nature 331, 440 – 442. Berridge, M.J., Downes, C.P., Hanely, M.R., 1982. Lithium amplifies agonist-dependent phosphatidylinositol responses in brain and salivary glands. Biochem. J. 206, 587 – 595. Bothwell, R.A., Ecceleston, D., Marschall, E., 1994. Platelet intracellular calcium in patients with recurrent affective disorders. Psychopharmacology 114, 375 – 381. Cali, J.J., Zwagstra, J.C., Mons, N., Cooper, D.M., Krupinski, J., 1994. Type VIII adenylyl cyclase. A Ca2+/calmodulin-stimulated enzyme expressed in discrete regions of rat brain. J. Biol. Chem. 269, 12190 – 12195. Cheung, J.Y., Bonventre, J.V., Malis, C.D., Leaf, A., 1986. Calcium and ischemic injury. N. Engl. J. Med. 314, 1670 – 1676. Christiansen, C., Baastrup, P.C., Transbol, I., 1980. Development of ‘‘primary’’ hyperparathyroidism during lithium therapy: longitudinal study. Neuropsychobiology 6, 280 – 283. Cooper, D.M., Mons, N., Karpen, J.W., 1995. Adenylyl cyclases and the interaction between calcium and cAMP signaling. Nature 374, 421 – 424.
Dubowsky, S.L., Christiano, J., Daniell, L.C., Franks, R.D., Murphy, J., Adler, L., Baker, N., Harris, R.A., 1989. Increased platelet intracellular calcium concentration in patients with bipolar affective disorders. Arch. Gen. Psychiatry 46, 632 – 638. Dubowsky, S.L., Lee, C., Christiano, J., Murphy, J., 1991. Elevated intracellular calcium ion concentration in bipolar depression. Biol. Psychiatry 29, 441 – 450. Dubowsky, S.L., Murphy, J., Christiano, J., Lee, C., 1992. The calcium second messenger system in bipolar disorders: data supporting new research directions. J. Neuropsychiatry Clin. Neurosci. 4, 3 – 14. Dubowsky, S.L., Thomas, M., Hijazi, A., Murphy, J., 1994. Intracellular calcium signalling in peripheral cells of patients with bipolar affective disorder. Eur. Arch. Psychiatry Clin. Neurosci. 243, 229 – 234. Ebstein, R.P., Lerer, B., Bennett, E.R., Shapira, B., Kindler, S., Shemesh, Z., Gerstenhaber, N., 1988. Lithium modulation of second messenger signal amplification in man: inhibition of phosphatidylinositol-specific phosphatase C and adenylate cyclase activity. Psychiatry Res. 24, 45 – 52. El-Mallakh, R.S., Wyatt, R.J., 1995. The Na,K-ATPase hypothesis for bipolar illness. Biol. Psychiatry 37, 235 – 244. Fajtova, V.T., Sayegh, M.H., Hickey, N., Alibadi, P., Lazarus, J.M., Leboff, M.S., 1995. Intact parathyroid hormone levels in renal insufficiency. Calcif. Tissue Int. 57, 329 – 335. Helmeste, D.M., Tang, S.W., 1998. The role of calcium in the etiology of the affective disorders. Jpn. J. Pharmacol. 77, 107 – 116. Hetmar, O., Povlsen, U.J., Ladefoged, J., Bolwig, T.G., 1991. Lithium: long-term effects on the kidney. A prospective follow-up study ten years after kidney biopsy. Br. J. Psychiatry 158, 53 – 58. Imandt, L., Tijhuis, D., Wessels, H., Haanen, C., 1981. Lithium inhibits adenylate cyclase of human platelets. Thromb. Haemostasis 45, 142 – 145. Kallner, G., Petterson, U., 1995. Renal, thyroid and parathyroid function during lithium treatment: laboratory tests in 207 people treated for 1 – 30 years. Acta Psychiatr. Scand. 91, 48 – 51. Kallner, G., Lindelius, R., Mathe´, A.A., Petterson, U., Nilsson, C.-G., 1992. Hypercalcemia in lithium treatment—an underestimated complication? La¨kartidningen 89, 4163 – 4164. Komatsu, M., Shimizu, H., Tsuruta, T., Kato, M., Fushimi, T., Inoue, K., Kobayashi, S., Kuroda, T., 1995. Effect of lithium on serum calcium level and parathyroid function in manic – depressive disorder. Endocr. J. 42, 691 – 695. Kusumi, I., Koyama, T., Yamashita, I., 1994. Effect of mood stabilizing agents on agonist-induced calcium mobilization in human platelets. J. Psychiatry Neurosci. 19, 222 – 225. Lenox, R.H., Hahn, C.G., 2000. Overview of the mechanism of action of lithium in the brain: fifty-year update. J. Clin. Psychiatry 61 (Suppl. 9), 5 – 15. Mallete, L.E., Eichorn, E., 1986. Effects of lithium carbonate on human calcium metabolism. Arch. Intern. Med. 146, 770 – 776. Mamlgren, R., Grunfelt, S., Joseph, R., 1992. On the significance of different aequorin loading techniques on intracellular aequorin discharge, baseline calcium, platelet aggregation and aequorin-indicated Ca(2+)transients. Thromb. Haemostasis 68, 352 – 356. Manji, H.K., Potter, W.Z., Lenox, R.H., 1995. Signal transduction pathways. Molecular targets for lithium’s actions. Arch. Gen. Psychiatry 52, 531 – 543. Meltzer, H.L., 1986. Lithium mechanisms in bipolar illness and altered intracellular calcium function. Biol. Psychiatry 21, 492 – 510. Newman, M.E., Lerer, B., Lichenberg, P., Shapira, B., 1992. Platelet adenylate cyclase activity in depression and after clomipramine and lithium treatment: relation to serotonergic treatment. Psychopharmacology 109, 231 – 234. Risby, E.D., Hsiao, J.K., Manji, H.K., Britan, J., Moses, F., Zhou, D.F., Potter, W.Z., 1991. The mechanisms of action of lithium: II. Effects on adenylate cyclase activity and beta-adrenergic receptor binding in normal subjects. Arch. Gen. Psychiatry 48, 513 – 524. Stahl, S.M., 1985. Peripheral models for the study of neurotransmitter receptors in man. Psychopharmacol. Bull. 21, 663 – 671.
A. El Khoury et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 26 (2002) 1063–1069 Stancer, H.C., Forbath, N., 1989. Hyperparathyroidism, hypothyroidism and impaired renal function after 10 to 20 years of lithium treatment. Arch. Intern. Med. 149, 1042 – 1045. Tan, C.H., Javros, M.A., Seleshi, E., Lowrimore, P.A., Bowden, C.L., 1990. Effects of lithium on platelet ionic intracellular calcium concentration in patients with bipolar (manic – depressive) disorder and healthy controls. Life Sci. 46, 1175 – 1180. Vicker, J.D., Mustard, J.F., 1986. The phosphoinositides exist in multiple metabolic pools in rabbit platelets. Biochem. J. 238, 411 – 417.
1069
Wood, K., 1985. The neurochemistry of mania. The effect of lithium on catecholamines, indolamines and calcium mobilization. J. Affective Disord. 8, 215 – 223. Yoshimura, M., Cooper, D.M., 1992. Cloning and expression of Ca(2+)inhibitable adenylyl cyclase from NCB-20 cells. Proc. Natl. Acad. Sci. USA 89, 6716 – 6720. Yu, H.J., Ma, H., Green, R.D., 1993. Calcium entry via L-type calcium channels acts as a negative regulator of adenylyl cyclase activity and cyclic AMP levels in cardiac myocytes. Mol. Pharmacol. 44, 689 – 693.