Blood and brain magnesium in inbred mice and their correlation with sleep quality

Biochemistry and Neurophysiology Unit, Department of Psychiatry, University of Geneva, 1225 Geneva, Switzerland

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

A strong genetic component in the regulation of blood magnesium (Mg) levels has been demonstrated. The regulation and distribution of brain Mg levels, however, have never been assessed. Herein we report on the genetic variation of peripheral and central Mg levels in six inbred strains of mice. In addition, the possible involvement of Mg in sleep regulation was assessed by establishing correlations between Mg and sleep parameters obtained before and after a 6-h sleep deprivation. Although genotype strongly determined blood Mg levels, it did not affect brain Mg, suggesting that central and peripheral Mg are regulated differently. Central Mg displayed a highly structure-specific distribution with frontal cortex having the highest and brain stem the lowest values. Whereas for the amount and distribution of baseline sleep only marginal correlations with Mg were found, Mg contents in four of nine brain structures were highly positively correlated with the length of slow-wave sleep episodes during recovery. This relationship suggests that higher levels of Mg in specific brain sites promote sleep quality as part of a recovery process.

fragmentation; sleep deprivation; recovery

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

MAGNESIUM IS THE FOURTH MOST abundant mineral in the brain and has a major role in biochemical processes. It activates many enzymes, including those involved in neurotransmitter synthesis (5, 8, 42), and modulates conductivity of many ion channels, such as N-methyl-D-aspartic acid receptors (30, 32, 33, 51) and inward rectifier potassium channels (6, 34). Furthermore, magnesium (Mg) is necessary for binding of most monoamines to their receptors (38, 39, 48). Thus this single cation has a key role in the neurotransmission at the cellular level both pre- and postsynaptically.

In vivo observations in animals and humans indicate that Mg deficiency is implicated in the pathophysiology of seizures (2, 3) and in behavioral alterations such as schizophrenia, anxiety, aggression, stress, and increased locomotor activity (22, 24, 27, 28, 36). Conversely, it has been established that depressed patients with the lowest psychomotor activity have increased peripheral Mg levels (49, 50). These observations seem to confirm the involvement of Mg in the regulation of the central nervous system excitability (30, 51). However, in most in vivo studies both in animals and humans, only the peripheral (i.e., blood) Mg concentration has been investigated. Therefore, the in vivo central-peripheral relationship remains unknown. In addition, even at the peripheral level, the mechanisms by which Mg concentration is regulated are poorly understood. Even though sodium-dependent transport systems responsible for Mg extrusion have been pharmacologically described in many cell types (14, 19, 46), nothing is known about the molecular components involved in Mg homeostasis in mammals. Genetic analysis in human populations has shown that low blood Mg levels are associated with specific human leukocyte antigens (HLA-B35, Bw6), and twin studies have indicated a high degree of heritability (0.92) for red blood cell (RBC) Mg (11, 21, 23).

The involvement of peripheral Mg in sleep regulation also has been investigated in a few studies. However, although Mg is sold over the counter in many countries for improving sleep quality and reducing fatigue, the evidence that Mg levels directly affect sleep is not strong. A possible correlation between blood Mg and sleep state duration has been reported, with more slow-wave sleep (SWS) and less paradoxical sleep (PS) when blood Mg is high (13, 41). Also, Mg deficiency has been reported to reduce SWS (stages 3 and 4) and increase sleep stages 1 and 2 and wake time after sleep onset in humans (41). In rats, short-term Mg deficiency is accompanied by a decrease in sleep time concomitant with an increase in central dopamine levels (40). Similar results also have been reported after prolonged Mg deficiency in rats, whereas reintroduction of Mg restored sleep organization to its original pattern (13). However, because Mg deficiency is almost always accompanied by changes in other elements such as potassium and calcium, the observed changes in sleep patterns may not be related to Mg levels per se. Also, how peripheral Mg affects sleep organization has not yet been investigated.

In the present study, we have addressed several aspects of Mg regulation and its involvement in the central nervous system excitability in the mouse. The central vs. peripheral relationship was investigated by measuring total Mg concentrations in the RBC, plasma, and brain of six inbred strains of mice. In addition, genetic variations in blood and brain Mg were studied by comparing the six strains. Moreover, because these mice have been studied previously to identify genetic aspects underlying sleep (15, 16), the present work was focused on the relationship between Mg and sleep, as a first step in evaluating the role of Mg in behavioral modulation.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals. The same individuals as those investigated in the two above-mentioned reports on genetic determinants of sleep and sleep electroencephalogram (EEG) were used in the present study (15, 16). Six mice from six inbred strains were included (ages 8-10 wk at the beginning of the study, weight 20-32 g). Five strains (AK, AKR/J; C, BALB/cByJ; D2, DBA/2J; B6, C57BL/6J; BR, C57BR/cdJ) were purchased from Jackson Laboratory, and the sixth, 129/Ola (129), was bred locally. Animals were kept under a 12:12-h light-dark cycle (lights on at 0800) in a sound-attenuated experimental room with temperature at 24 ± 1°C and humidity of 45-65%. Food and water were available ad libitum. Animals were kept under these conditions for at least 18 days before the experiment.

Surgery. EEG and electromyogram (EMG) electrodes were implanted under pentobarbital sodium anesthesia (65-75 mg/kg ip, depending on strain). The electrodes were soldered to recording leads, cemented to the skull, and connected to a swivel-contact, and the animals were allowed 10-14 days for recovery and habituation to the recording conditions before the experiment.

Sleep recording and determination of behavioral states. EEG and EMG recordings started at lights on (0800) and lasted for 2 days with the first day (24 h) as baseline. On the second day starting at lights on, mice were sleep deprived for 6 h by gentle handling, and the remaining 18 h of this day were considered as recovery. Animals' behavior was classified for consecutive 4-s epochs as PS, SWS, or wakefulness (W) by visual inspection of the EEG and EMG signals.

Mg determination. Because the aim of this study is to investigate the involvement of basal Mg in sleep regulation, all animals were killed 3 days after the end of the sleep recordings to counteract any possible changes in brain and peripheral Mg due to sleep deprivation (SD). Animals from different strains were killed between 0800 and 1230 under deep anesthesia with pentobarbital sodium at 75 mg/kg [no major circadian variation of plasma Mg, the fraction which is most sensitive to changes, has been reported (47)]. From 500 to 800 µl of blood were collected by cardiopuncture in heparinized tubes, and brains were quickly removed, frozen, and stored at 80°C.

Blood samples were centrifuged for 10 min at 3,500 g and 4°C, and the RBC pellets were washed three times in a KCl isotonic solution. RBC pellets were diluted in KCl isotonic solution (2:3 vol/vol) from which hematocrit was measured. Then 100 µl of this solution were mixed with 9,900 µl of distilled water. Plasma samples were centrifuged for 10 min at 3,500 g and room temperature to remove residual blood cells. Plasma and RBC Mg concentrations were then measured in triplicate by the usual atomic absorption spectrometry [Perkin Elmer (1)] with minor modifications, and results are reported for plasma or RBC (mmol/l). It should be pointed out that Mg determination in the present study concerns total Mg (bound + free Mg²⁺), whereas only free Mg²⁺ constitutes the active form. However, although the determination of free Mg²⁺, especially in vivo (measured for instance by phosphorus magnetic resonance spectroscopy), may be more relevant to its function, it is well established that total intracellular Mg is highly correlated with free Mg²⁺ concentration (18, 43).

Brains were microdissected at 20°C into nine blocks, containing the following main structures: amygdala (AM), frontal cortex (FC), basal forebrain (BF), cerebellum (CB), motor cortex (MC), somatosensory cortex (SC), thalamus (TH), mesencephalon (ME), and brain stem (BS). Tissue blocks will be identified by the name of their main structure, although they might contain parts of other structures. Dissections were performed according to the rapid method described by Heffner et al. (20), with adaptations for mice based on the stereotaxic atlas by Franklin and Paxinos (17). The frozen blocks were sonicated in distilled water in preweighted glass tubes, lyophilized for 16 h, and weighted to obtain the dry weight. A second lyophilization did not indicate any change in weight. Samples were then mineralized for 12 h in 1.5 ml acid solution (3:1 solution of 65% HNO₃ and 70% HClO₄) and then dissolved in 4.5 ml of distilled water by sonication. After centrifugation for 10 min at 3,500 g, supernatants were used for Mg determination in triplicate by atomic absorption spectrometry. Total brain Mg concentration in each mouse was calculated as the sum of the products of Mg concentration (in µmol/g of dry weight) and dry weight per structure divided by the total weight of all structures. In the same way, Mg concentration was calculated for cortical (FC + MC + SC + AM) and subcortical (BF + TH + ME) structures.

Behavioral states. The sleep data used in the present study were recently reported (16). Therefore, only nonreported results concerning sleep consolidation are presented. Briefly, mean values for the amount of behavioral states in baseline and recovery were expressed as a percentage of total recording time for the 12:12-h light-dark cycles. State fragmentation was assessed by counting the number of episodes: one to four 4-s epochs (or <16 s) for short W episodes, one to 15 4-s epochs (or <60 s) for short SWS episodes, and >15 4-s epochs for long SWS episodes as defined by Franken et al. (16). For the evaluation of SD effects on sleep fragmentation, the data from the first 3 h of recovery (i.e., hours 6-9 after lights on) were compared with the corresponding time interval of baseline.

Statistics. All main effects of factors strain and Mg concentration were analyzed by ANOVA for repeated measures. Whenever main effects reached the significant level (P < 0.05), post hoc group comparisons were performed (Tukey's studentized range test). Correlation between brain and blood Mg concentrations and also within the brain and with sleep parameters were established with Pearson's correlation test.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Mg and genotype. Highly significant differences in Mg concentrations were found between genotypes and between plasma and RBC (2-way ANOVA, factor strain, P < 0.0001; factor RBC and plasma, P < 0.0001; interaction, P = 0.38) with RBC and plasma Mg showing similar strain distribution pattern (Fig. 1). For both plasma and RBC, strain differences were due to high values for B6 and low values for C (RBC) and 129 (plasma) mice (Fig. 1). ANOVA in these inbred strains indicated that 75% of the total variance was genetically determined.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Central and peripheral magnesium (Mg) distribution in inbred mice. B6, C57BL/6J; BR, C57BR/cdJ; AK, AKR/J; D2, DBA/2J; C, BALB/cByJ; 129, 129/Ola. Vertical bars represent means ± SE, n = 6/strain. A and B: red blood cell (RBC) and plasma Mg varied among strains (1-way ANOVA, factor strain, P < 0.0006). Horizontal bars indicate strains between which values differed significantly (Tukey's range test, P < 0.01). C: total brain Mg did not vary among strains (1-way ANOVA, factor strain, P = 0.18).

In contrast, brain Mg did not vary with genotype. Nevertheless, the strain distribution of brain Mg appeared to be opposite of that for blood Mg with 129 now showing the highest values (Fig. 1). This tendency was reflected by a nonsignificant negative correlation between total brain and RBC Mg (r = 0.30, P = 0.08), which reached the significant level for amygdala (r = 0.33, P < 0.05).

Because brain Mg did not differ between strains, data were pooled for all mice (n = 36). Highly significant differences were observed between structures mainly due to high values for frontal cortex and low values for brain stem, which both significantly differed from all other structures (Fig. 2A). Also when brain structures were grouped as cortical (FC, MC, AM, and SC) and subcortical (BF, TH, and ME) and compared with the cerebellum and the brain stem, marked differences were observed (1-way ANOVA, factor area, P < 0.0001; Fig. 2B).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Brain distribution of Mg in inbred mice. A: vertical bars represent means ± SE, n = 36/structure. Mg concentration varied among structures (1-way ANOVA, factor structure, P < 0.0001). Horizontal bars indicate structures between which values differed significantly (Tukey's range test, P < 0.01). FC, frontal cortex; MC, motor cortex; AM, amygdala; BF, basal forebrain; SC, somatosensory cortex; TH, thalamus; CB, cerebellum; ME, mesencephalon; BS, brain stem; TOT, total brain. B: Mg concentration varied between cortical (FC + MC + SC + AM), subcortical (BF + TH + ME), CB, and BS. (1-way ANOVA, factor area, P < 0.0001)

Mg and sleep. Detailed analysis of vigilance states and EEG activity during sleep in the same individuals have been reported recently (15, 16). Briefly, the percentage of W and SWS during the 24 h of baseline varied with genotype. This effect was largely due to differences among strains in the 12-h dark period, which is the active period for mice. Within the light and dark period also PS varied among strains, but during the dark period differences were only due to high values in C mice. The light-dark difference between the mean sleep duration during baseline, which was used as a measure for the diurnal "amplitude" of vigilance states, was also affected by genotype with C mice showing the smallest amplitude. Compared with baseline, SD led to a significant increase in SWS and PS in most strains.

Only marginally significant correlations between both peripheral and central Mg concentrations and the amount of vigilance states were found that could account for 14-19% of the total variance. In the dark period of baseline, RBC Mg was negatively correlated with %SWS (r = 0.37, P < 0.04) and positively with %W (r = 0.37, P < 0.03). RBC Mg also was positively correlated with the light-dark SWS amplitude (r = 0.38, P < 0.03). The percentage of SWS during the 24 h of baseline was correlated with basal forebrain Mg (r = 0.34, P < 0.05), mainly reflecting the correlation between basal forebrain Mg and %SWS in the light period (r = 0.39, P < 0.02). Opposite correlations were found between basal forebrain Mg and %W (24 h: r = 0.36, P < 0.04; light period: r = 0.43, P < 0.02). A positive correlation was also observed between RBC Mg and %SWS (r = 0.35, P < 0.04) during the light period of recovery.

Apart from behavioral state quantities, sleep quality can also be assessed by analyzing its fragmentation. The number of short W and SWS episodes and the number and mean duration of long SWS episodes were used to index sleep quality. Recently, differences in sleep fragmentation in the same mice have been reported (Ref. 16 and Table 1), suggesting genotype-dependent differences in sleep quality. It also has been observed that SD induces a robust decrease in the number of short W and SWS episodes (16). In the present study, the number and the duration of long SWS episodes were also analyzed as a complementary index of sleep continuity. As shown in Table 1, the mean duration of long SWS episodes significantly increased after SD in most strains.

View this table: [in this window] [in a new window]   Table 1.   Sleep fragmentation during the first 3 hours of recovery and the corresponding time interval of baseline

Correlations between baseline levels of central Mg and sleep quality in the 3 h immediately after SD are reported in Table 2. Significant negative correlations were found with the number of short SWS episodes, with the motor cortex showing the best correlation. The number and the mean duration of long SWS episodes showed striking correlations with brain Mg. Among the brain structures, the motor cortex was consistently and highly significantly correlated with all three sleep-consolidation parameters (Table 2), explaining up to 50% of the total variance.

View this table: [in this window] [in a new window]   Table 2.   Correlation between central Mg and sleep quality in the first 3 hours of recovery light period

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Control of peripheral and central Mg levels. Henrotte and associates (21, 23) were the first to point out the importance of genetic factors in the control of peripheral Mg in both humans and mice. Genotype differences in plasma and RBC Mg as reported in the present study clearly confirm a strong genetic control of peripheral Mg (heritability = 0.75). As this genotype effect was not observed for the brain Mg, central and peripheral Mg appear to be regulated by different mechanisms. However, it cannot be excluded that genetic differences may still be observed in more precisely defined brain areas, such as ganglia, nuclei, or cell layers. The nonsignificant negative relationship between brain and peripheral Mg deserves further investigations. Although under Mg deficiency, brain Mg concentration follows blood Mg decrease (3, 7, unpublished observations from our laboratory); lower physiological blood Mg levels seem to be associated with higher brain Mg.

Central Mg is distributed following an anteroposterior and cortical-subcortical gradient with higher levels for the cortical areas. This new finding suggests that Mg may have a specific role in higher brain functions especially associated with the frontal cortex, which has significantly higher Mg content than any other brain structure. Alternatively, the relative contribution of the gray and white matter in each structure may explain the differences between cortical and subcortical areas because Mg is mostly an intracellular ion and can be expected to be highest in concentration in areas with high cellular density (7).

Mg and sleep. Our observation that high levels of basal forebrain Mg are associated with high amounts of sleep during the rest period is of interest because this structure includes the basal nucleus of Meynert, which can modulate the amount of sleep through its W promoting cholinergic and SWS promoting GABAergic projections to cortical areas (26, 31, 36, 37). Moreover, the negative relationship between high levels of RBC Mg and sleep and the positive correlation with %W during the active period agree with Mg deficiency experiments in rats and humans, where increased W and decreased sleep time have been reported (13, 40, 41). How Mg deficiency affects Mg content of different brain structures is under investigation in our laboratory. This also will further our understanding of peripheral-central relationship in the regulation of Mg.

The highly significant positive correlation between brain Mg and the duration of long SWS episodes during the first 3 h after SD suggests that Mg may have a role in the modulation of sleep quality. It is well established that stressful conditions promote renal Mg loss, thereby potentiating some of the physiological effects of stress on the organism (9, 45). Cardiovascular damages, but also behavioral alterations such as aggression, are observed in relation to Mg deficiency (9, 27). Mice genetically selected for low blood Mg levels present a high level of interindividual aggression (24). Thus the correlations found in the present study between high levels of brain Mg and long noninterrupted episodes of sleep after SD can be interpreted as part of a mechanism to enhance the efficiency of recovery from SD stress by improving the quality of sleep (i.e., by increasing sleep consolidation). On the other hand, because Mg determination was not done immediately after SD, it cannot be excluded that SD could affect (decrease) brain Mg, and therefore animals with the lowest Mg concentration in baseline conditions may be the most sensitive to brain Mg changes during SD. The lack of a large Mg effect during the baseline recordings further suggests that Mg shows its positive effects mainly under stressful conditions. Note that Mg supplement in humans is usually considered under fatigue and sleep disturbances resulting from stressful situations. Also, recovery from SD is believed to serve a cortical function in a regional and use-dependent fashion (29, 44). Accordingly, the specific regional distribution of brain Mg with the highest concentration found in the cortical areas may contribute to the cortical recovery process. Additionally, in rodents where forced locomotion is the best tool for SD, it may represent a higher motor cortex stimulation, which will show higher recovery as found in the present study. Finally, because it has been postulated that recovery from SD may be related to replenishment of brain energy stores (4), Mg could enhance the efficiency of recovery sleep through its well-known role as a cofactor in brain glycogen synthesis (Mg²⁺-ATP).

Perspectives

A recent survey conducted by the Gallup Organization has found that 72% of adult Americans are falling short of the recommended dietary allowance for Mg (380 mg/day). Mg may be a critical single ion affecting patients with chronic fatigue syndrome (10) and periodic leg movements-related insomnia (25). Although Mg supplement is usually recommended for improving sleep quality, there has been little evidence of its involvement in sleep regulation. The present study may highlight the mechanisms through which Mg affects sleep quality. Thus both peripheral and central Mg levels are tightly associated with sleep consolidation especially after SD. If our hypothesis is confirmed that Mg levels in some cortical areas may be critical for sleep quality, this will also constitute additional evidence for the regional use-dependent function of sleep (29, 44) because brain Mg has a highly structure-specific distribution.

The use of the mouse as a genetic model will greatly facilitate the discovery of molecular components of Mg regulation and their involvement in behavioral processes such as sleep regulation. A recent linkage analysis in human renal Mg loss has identified a region on chromosome 11 probably encompassing a gene regulating renal Mg transport (35). The genetic analysis of peripheral Mg regulation is currently under investigation in mice genetically selected for their low or high RBC Mg (23) and in recombinant offspring of inbred mice differing in their RBC Mg as identified in the present study. These are necessary steps to further our understanding of both the genetics and the functional role of this essential nutrient.

	ACKNOWLEDGEMENTS

We thank Dr. Claude Walzer for help with brain Mg determinations.

	FOOTNOTES

This study was supported by the Hôpitaux Universitaire de Genève and the Swiss National Science Foundation Grants 31.457513.95 and 31-56000.98.

Address for reprint requests and other correspondence: M. Tafti, Biochemistry and Neurophysiology Unit, Dept. of Psychiatry, Univ. of Geneva, Chemin du Petit-Bel-Air 2, CH-1225 Chêne-Bourg, Switzerland (E-mail: tafti{at}cmu.unige.ch).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 14 March 2000; accepted in final form 16 August 2000.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

1.   Archer, WH, Emerson RL, and Reusach CS. Intra- and extracellular fluid magnesium by atomic absorption spectrophotometry. Clin Biochem 5: 159-161, 1972[ISI][Medline].

2.   Bac, P, Maurois P, Dupont C, Pages N, Stables JP, Gressens P, Evrard P, and Vamecq J. Magnesium deficiency-dependent audiogenic seizures (MDDASs) in adult mice: a nutritional model for discriminatory screening of anticonvulsant drugs and original assessment of neuroprotection properties. J Neurosci 18: 4363-4373, 1998[Abstract/Free Full Text].

3.   Belknap, JK, Berg JH, Cocke R, and Clancy AN. Induction and reversal of the magnesium deficiency syndrome in inbred mice. Exp Neurol 57: 506-515, 1977[ISI][Medline].

4.   Benington, JH, and Heller HC. Restoration of brain energy metabolism as the function of sleep. Prog Neurobiol 45: 347-360, 1995[ISI][Medline].

5.   Boadle-Biber, MC. Activation of tryptophan hydroxylase from central serotonergic neurons by calcium and depolarization. Biochem Pharmacol 27: 1069-1079, 1978[ISI][Medline].

6.   Chuang, H-U, Jan YN, and Jan LY. Regulation of IRK3 inward rectifier K⁺ channel by M1 acetylcholine receptor and intracellular magnesium. Cell 89: 1121-1132, 1997[ISI][Medline].

7.   Chutkow, JG. Distribution of magnesium and calcium in brains of normal and magnesium-deficient rats. Mayo Clin Proc 47: 647-653, 1972[ISI][Medline].

8.   Chutkow, JG. Magnesium and the central nervous system: metabolism, neurophysiological functions, and clinical disorders. In: Metal Ions in Biological Systems; Compendium on Magnesium and its Role in Biology, Nutrition, and Physiology, edited by Sigel H, and Sigel A.. New York: Dekker, 1990, p. 441-461.

9.   Classen, HG. Systemic stress, magnesium status and cardiovascular damage. Magnesium 5: 105-110, 1986[ISI][Medline].

10.   Cox, IM, Campbell MJ, and Dowson D. Red blood cell magnesium and chronic fatigue syndrome. Lancet 337: 757-760, 1991[ISI][Medline].

11.   Darlu, P, Rao DC, Henrotte J-G, and Lalouel J-M. Genetic regulation of plasma and red blood cell magnesium concentrations in man. I. Univariate and bivariate path analyses. Am J Hum Genet 34: 874-887, 1982[ISI][Medline].

12.   Depoorter, H, Françon D, and Llopis J. Effects of magnesium-deficient diet on sleep organization in rats. Neuropsychobiology 24: 237-245, 1993.

13.   Dralle, D, and Bödeker RH. Serum magnesium level and sleep behavior of newborn infants. Eur J Pediatr 134: 239-243, 1980[ISI][Medline].

14.   Flatman, PW. Mechanisms of magnesium transport. Annu Rev Physiol 53: 259-271, 1991[ISI][Medline].

15.   Franken, P, Malafosse A, and Tafti M. Genetic variation in EEG activity during sleep in inbred mice. Am J Physiol Regulatory Integrative Comp Physiol 275: R1127-R1137, 1998[Abstract/Free Full Text].

16.   Franken, P, Malafosse A, and Tafti M. Genetic determinants of sleep regulation in inbred mice. Sleep 22: 155-169, 1999[ISI][Medline].

17.   Franklin, KBJ, and Paxinos G. The Mouse Brain in Stereotaxic Coordinates. New York: Academic, 1997.

18.   Fujise, H, Cruz P, Reo NV, and Lauf PK. Relationship between total magnesium concentration and free intracellular magnesium in sheep red blood cells. Biochim Biophys Acta 1094: 51-54, 1991[Medline].

19.   Günther, T. Mechanisms and regulation of Mg²⁺ efflux and Mg²⁺ influx. Miner Electrolyte Metab 19: 259-265, 1993[ISI][Medline].

20.   Heffner, TG, Hartman JA, and Seiden LS. A rapid method for the regional dissection of the rat brain. Pharmacol Biochem Behav 13: 453-456, 1980[ISI][Medline].

21.   Henrotte, J-G, Colombani J, Pineau M, and Dausset J. Role of H-2 and non H-2 genes in the control of blood magnesium levels. Immunogenetics 19: 435-448, 1984[ISI][Medline].

22.   Henrotte, J-G, Franck G, Santarromana M, Francès H, Mouton D, and Motta R. Mice selected for low and high blood magnesium levels: a new model for stress studies. Physiol Behav 61: 653-658, 1997[Medline].

23.   Henrotte, J-G, Pla M, and Dausset J. HLA- and H-2-associated variations of intra- and extracellular magnesium content. Proc Natl Acad Sci USA 87: 1894-1898, 1990[Abstract/Free Full Text].

24.   Henrotte, J-G, Soni A, Monier C, Frances H, and Dauchy F. High catecholamines excretion and aggressive behaviour in mice selected for low red blood cell magnesium levels (Abstract). Magnes Res 4: 236, 1991.

25.   Hornyak, M, Voderholzer U, Hohagen F, Berger M, and Riemann D. Magnesium therapy for periodic leg movements-related insomnia and restless legs syndrome: an open pilot study. Sleep 21: 501-505, 1998[ISI][Medline].

26.   Jones, BE. The organization of central cholinergic systems and their functional importance in sleep-waking states. Prog Brain Res 98: 61-71, 1993[ISI][Medline].

27.   Kantak, KM. Magnesium deficiency alters aggressive behavior and catecholamine function. Behav Neurosci 102: 304-311, 1988[ISI][Medline].

28.   Kirov, GK, and Tsachev KN. Magnesium, schizophrenia and manic-depressive disease. Neuropsychobiology 23: 79-81, 1990[ISI][Medline].

29.   Krueger, JM, and Obal F. A neuronal group theory of sleep function. J Sleep Res 2: 63-69, 1993[Medline].

30.   Kuner, T, and Schoepfer R. Multiple structural elements determine subunit specificity of Mg²⁺ block in NMDA receptor channels. J Neurosci 16: 3549-3558, 1996[Abstract/Free Full Text].

31.   Lancel, M. Role of GABA-A receptors in the regulation of sleep: initial sleep responses to peripherally administered modulators and agonists. Sleep 22: 33-42, 1999[ISI][Medline].

32.   Li-Smerin, Y, and Johnson JW. Kinetics of the block by intracellular Mg²⁺ of the NMDA-activated channel in cultured rat neurons. J Physiol (Lond) 491: 121-135, 1996[ISI].

33.   Li-Smerin, Y, and Johnson JW. Effects of intracellular Mg²⁺ on channel gating and steady-state responses of the NMDA receptor in cultured rat neurons. J Physiol (Lond) 491: 137-150, 1996[ISI].

34.   Matsuda, H, Saigusa A, and Irisawa H. Ohmic conductance through the inwardly rectifying K channel and blocking by internal Mg²⁺. Nature 325: 156-159, 1987[Medline].

35.   Meij, IC, Saar K, van den Heuvel LPW, Nuernberg G, Vollmer M, Hildebrandt F, Reis A, Monnens LAH, and Knoers NVAM Hereditary isolated renal magnesium loss maps to chromosome 11q23. Am J Hum Genet 64: 180-188, 1999[ISI][Medline].

36.   Modak, AT, Montanez J, and Stavinoha WB. Magnesium deficiency: brain acetylcholine and motor activity. Neurobehav Toxicol 1: 187-191, 1979[Medline].

37.   Murck, H, and Steiger A. Mg²⁺ reduces ACTH secretion and enhances spindle power without changing delta power during sleep in men - possible therapeutic implications. Psychopharmacology (Berl) 137: 247-252, 1998[Medline].

38.   Nelson, DL, Herbet A, Enjalbert A, Bockaert J, and Hamon M. Serotonin-sensitive adenylate cyclase and (3H) serotonin binding sites in the CNS of the rat. Biochem Pharmacol 29: 2445-2543, 1980[ISI][Medline].

39.   Okada, M, and Kaneko S. Pharmacological interactions between magnesium ion and adenosine on monoaminergic system in the central nervous system. Magnes Res 11: 289-305, 1998[ISI][Medline].

40.   Poenaru, S, Rouhani S, Durlach J, Aymard N, Rayssiguier Y, and Iovino M. Vigilance states and cerebral monoamine metabolism in experimental magnesium deficiency. Magnesium 3: 145-151, 1984[Medline].

41.   Popoviciu, L, Bagathai J, Buksa C, Delast-Popoviciu D, Bicher G, Delast-Popoviciu R, Covaciu S, and Szalay E. Clinical and polysomnographic researches in patients with sleep disorders associated with magnesium deficiencies. In: Magnesium: A Relevant Ion, edited by Lasserre B, and Durlach J.. London: Libbey, 1991, p. 353-365.

42.   Raese, JD, Edelman AM, Makk G, Bruckwich EA, Lovenberg W, and Barchas JD. Brain striatal tyrosine hydroxylase: activation of the enzyme by AMP-independent phosphorylation. Commun Psychopharmacol 3: 295-301, 1979[ISI][Medline].

43.   Resnick, LM, Gupta RK, and Laragh JH. Intracellular free magnesium in erythrocytes of essential hypertension: relation to blood pressure and serum divalent cations. Proc Natl Acad Sci USA 81: 6511-6515, 1984[Abstract/Free Full Text].

44.   Schwierin, B, Achermann P, Deboer T, Oleksenko A, Borbély AA, and Tobler I. Regional differences in the dynamics of the cortical EEG in the rat after sleep deprivation. Clin Neurophysiol 110: 869-875, 1999[ISI][Medline].

45.   Seelig, MS. Consequences of magnesium deficiency on the enhancement of stress reactions; preventive and therapeutic implications (a review). J Am Coll Nutr 13: 429-446, 1994[Abstract].

46.   Smith, DL, and Maguire ME. Genetics and molecular biology of magnesium transport systems. In: The Biological Chemistry of Magnesium, edited by Cowan JA.. New York: VCH, 1995, p. 211-234.

47.   Touitou, Y, Touitou C, Bogdan A, Reinberg A, Motohashi Y, Auzéby A, and Beck H. Circadian and seasonal variations of electrolytes in aging humans. Clin Chim Acta 180: 245-254, 1989[ISI][Medline].

48.   Usdin, TB, Creese I, and Synder SH. Regulation by cations by (3H) spiroperidol binding associated with dopamine receptors of rat brain. J Neurochem 34: 669-676, 1980[ISI][Medline].

49.   Widmer, J, Henrotte JG, Raffin Y, Bovier P, Hilleret H, and Gaillard J-M. Relationship between erythrocyte magnesium, plasma electrolytes and cortisol, and intensity of symptoms in major depressed patients. J Affect Disord 34: 201-209, 1995[ISI][Medline].

50.   Widmer, J, Henrotte J-G, Raffin Y, Mouthon D, Chollet D, Stépanian R, and Bovier P. Relationship between blood magnesium and psychomotor retardation in drug-free patients with major depression. Eur Psychiatry 13: 90-97, 1998.

51.   Wollmuth, LP, Kuner T, and Sakmann B. Intracellular Mg²⁺ interacts with structural determinants of the narrow constriction contributed by the NR1-subunit in the NMDA receptor channel. J Physiol (Lond) 506: 33-52, 1998[Abstract/Free Full Text].