A Direct Quantitative Analysis of Erythrocyte Intracellular Ionized Magnesium in Physiological and Pathological Conditions

Abstract

Magnesium (Mg²⁺) is an endogenous cation that is involved in many essential biological reactions. Abnormal Mg²⁺ metabolisms in the body affect important physiological and pathological processes. However, most endogenous Mg²⁺ markers fail to represent body Mg²⁺ status; they are disadvantageous in terms of representational capacity, applied range, operational convenience, etc. In this article, we evaluated some of the most popular Mg²⁺ marker candidates. A logical model of the blood Mg²⁺ compartments was established, which consisted of unstable Mg²⁺ pools, representative Mg²⁺ pools, and conserved Mg²⁺ pools. These pools were based on the metabolic efficiency of Mg²⁺ in an acute Mg²⁺ intake test. The results of this study showed that only the erythrocyte intracellular ionized Mg²⁺ (RBC [Mg²⁺]_i), a representative Mg²⁺ pool, could effectively represent abnormal body Mg²⁺ metabolisms in various conditions, including dietary Mg²⁺ adjustments, aging and metabolic syndrome. These results suggest that RBC [Mg²⁺]_i might be a widely applicable marker of body Mg²⁺ levels. On unified technology platform and evaluation system, this research compared the representative capacities of RBC [Mg²⁺]_i, plasma Mg²⁺ concentration (plasma [Mg²⁺]), erythrocyte intracellular total Mg (RBC [Mg]_total) and Mg retention in rats and mice under various Mg²⁺-metabolism-related physiological and pathological conditions. Our technique for the direct quantitative analysis of RBC [Mg²⁺]_i may prove valuable for basic physiological research, dietary Mg²⁺ regulation, as well as clinical monitoring/intervention of Mg²⁺-metabolism-related pathology.

INTRODUCTION

Magnesium (Mg²⁺) is the fourth most abundant mineral in the body and participates in numerous bio-reactions.¹⁾ The level of Mg²⁺ in the body regulates muscle contraction, blood pressure, heart rhythm, and neural activity, etc.^1–3)

Abnormal metabolisms of body Mg²⁺ have been found in many physiological and pathological processes. For example, the levels of Mg²⁺ have been found to significantly decline in the elderly,^4,5) depression patients,^3,6,7) hypertension and diabetes patients,^4,8–10) etc. On the other hand, an excess of Mg²⁺ in the body may lead to diarrhea and myasthenia.²⁾ Therefore, an assay of body Mg²⁺ level is crucial for monitoring these Mg²⁺-metabolism-related pathological processes and providing references for the dietary Mg²⁺ regulation.

There are many different Mg²⁺ pools in the body, such as body fluid Mg²⁺, tissue Mg²⁺, intracellular total Mg (the free ionized form plus the biomolecule binding form of Mg) and intracellular ionized Mg²⁺. There are also various techniques to measure the Mg²⁺ levels in specific tissue samples, Mg²⁺ concentration ranges, and Mg²⁺ free/bonding forms.¹¹⁾ Different Mg²⁺ compartments in the body have various representational capacities. For example, in a rat model of dietary Mg²⁺ deficiency, the Mg²⁺ level in the rat’s soft tissue did not change. However, the Mg²⁺ levels in the skeleton and plasma decreased significantly.¹²⁾ In another case, the level of erythrocyte intracellular ionized Mg²⁺ (RBC [Mg²⁺]_i) changed in hypertension patients, but the levels of plasma Mg²⁺ concentration (plasma [Mg²⁺]) did not.¹⁰⁾ By literature retrieval, we summarized the abilities of some common candidates of Mg²⁺ markers in indicating body Mg²⁺ deficiency, caused by dietary Mg²⁺ deficiency, aging, and metabolic syndrome (supplementary information Fig. S1). It remains elusive to find a widely applied Mg²⁺ indicator to represent abnormal body Mg²⁺ metabolisms in related physiological and pathological processes. A unified Mg²⁺ indicator is needed, capable for conducting statistical analyses of data in different studies. The indicator should be convenient in representational capacity, applied range, and operational convenience. The related laboratory technique for this potential Mg²⁺ indicator will also need to be designed.

An ideal Mg²⁺ indicator should be representative, repeatable, and easily accessible. Based on these requirements, as well as the previous reports (Fig. S1), we focused on the blood Mg²⁺ compartments. These Mg²⁺ pools exchange Mg²⁺ with all other parts of the body and are easy to extract, thus showing the potential as an Mg²⁺ indictor. In this paper, we evaluated the physiological and pathological representative capacities of RBC [Mg²⁺]_i, plasma [Mg²⁺], erythrocyte intracellular total Mg (RBC [Mg]_total, the free ionized form plus the biomolecule binding form of Mg) and Mg retention. The results suggested that RBC [Mg²⁺]_i was the most effective Mg²⁺ indicator in Mg²⁺-metabolism-related physiological and pathological processes.

MATERIALS AND METHODS

Experimental Animals

Sprague Dawley rats and mice generated by the Institute of Cancer Research (ICR mice) were obtained from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). All of the animals were housed individually in a controlled environment (temperature: 21 ± 1°C, humidity: 50 ± 10%), under a 12 : 12 h inverted light-dark cycle. All of the animals were freely accessible with a commercial pelleted diet (SLAC Laboratory Animal Co., Ltd., Shanghai, China), containing a normal elemental Mg concentration (0.15%) and normal purified water; except for special diet groups (mentioned in other sections below).

All experiments involving animals were approved by the Tsinghua University animal-care committee. All studies were carried out in accordance with the approved guidelines and regulations.

Mg²⁺ Measurements

This study first worked on a technique to quantitatively assay RBC [Mg²⁺]_i in murine by using a fluorescence probe. Blood was taken from the caudal vein of rats and from the post-glomus venous plexus of mice. The blood was then transferred into a heparin coated tube. The blood samples were diluted (1 : 1200) in Hank’s Balanced Salt Solution (HBSS, Sigma-Aldrich, St. Louis, MO, U.S.A.), containing 0.5 mmol/L of Mg²⁺. Magnesium Green (Invitrogen, Carlsbad, CA, U.S.A.), a cell-permeable Mg²⁺ dye, was used to detect RBC [Mg²⁺]_i, with a working concentration of 5 mg/L. Pluronic F-127 (Invitrogen) was used to aid the dissolution of Magnesium Green. The mixture of RBC cells and dye was incubated at 37°C for 60 min, washed twice, and then incubated at 37°C for another 30 min. The fluorescence signal was detected by flow cytometry (FACSCalibur, BD, Franklin Lakes, NJ, U.S.A.). A total of 1 × 10⁴ cells were gated for average fluorescence intensity. For calibration, parallel samples ready for fluorescence detections were washed twice with HBSS (zero Mg²⁺) and then mixed in HBSS with gradients of Mg²⁺ concentrations. A23187 (Sigma-Aldrich) was used to maintain the balance of external and internal Mg²⁺ concentrations across the cellular membrane, with a working concentration of 25 µmol/L. Mixtures were incubated at 37°C for 2 h and then analyzed by flow cytometry. A fitted curve was built, with the x-axis denoting the fluorescence intensity and the y-axis denoting the standard [Mg²⁺] gradients.

The Mg²⁺ levels in the plasma were determined by calmagite chromometry.¹³⁾ Briefly, 2 mL working solution and 50 µL plasma sample were mixed. Optical density (OD) was measured using spectrophotometry at 525 nm. After adding 20 µL of 150 mM of ethylenediaminetetraacetic acid (EDTA) into the mixture, a second OD value was recorded. A fitted curve was constructed using OD_first minus OD_second as the x-axis and standard Mg²⁺ concentration gradients as the y-axis.

For the RBC [Mg]_total determination, 0.5–1 mL of blood samples anticoagulant by heparin were washed twice in HBSS (without Mg²⁺). Following the final centrifugation, the erythrocyte precipitation was collected. Some erythrocytes were diluted in HBSS in order to carry out cell counting. Other erythrocytes were mixed (1 : 1.25) with deionized water, oscillated on a vortex, and lysed. The lysis samples were sent to a specialist agency (General Research Institute for Nonferrous Metals, Beijing, China) for further processing and were analyzed using inductively coupled plasma mass spectrometry (ICP-MS).

To assess the Mg retention, rats were individually housed in metabolic cages. They received normal food or Mg²⁺-deficient food, plus normal water. Following 3 d of habituation, the urine and fecal pellets of each rat were collected on a daily basis over a continuous 5-d period. The food in each cage was weighed daily to calculate the total Mg intake. The body weight of the rats was also recorded. The urine and fecal pellets were analyzed for total Mg content using inductively coupled plasma atomic emission spectrometry (ICP-AES; General Research Institute for Nonferrous Metals, Beijing, China). The body Mg retention was estimated using the following equation:

  

(1)

Mg²⁺-Deficient Diet

The Mg²⁺-deficient diet was purchased from the Chinese Academy of Agricultural Sciences (Beijing, China). The total elemental Mg concentration in the diet was 0.003% (compared to 0.15% in the standard diet), as determined by ICP-AES (Chinese Academy of Agricultural Sciences).

For the Mg²⁺-intake-deficiency test, rats and mice were fed with either standard food or Mg²⁺-deficient food for 4 weeks. The chow was weighed to calculate the total daily intake of Mg. A body Mg retention test was conducted during this period. At week 0, blood samples were taken from each animal, and Mg²⁺ levels were measured. At the end of week 4, blood samples were again taken for Mg²⁺ measurements to evaluate the effect of Mg²⁺-intake-deficiency treatment.

Mg²⁺-Enriched Diet

Animals were fed with normal food and water containing Mg²⁺. Magnesium L-Threonate (MgT; PubChem CID: 53395228), an Mg²⁺ compound that is highly bio-affinitive, was dissolved in purified water. The Mg²⁺-contained water was freely accessible to the animals. For rats, the dose was 50 mg of elemental Mg²⁺/kg body weight/d. For mice, the dose was 75 mg of elemental Mg²⁺/kg body weight/d. The amount of MgT necessary to reach the target dose was calculated, based on daily water consumption and individual body weight.

Statistical Analysis

In the text, data were presented as mean±standard deviation (S.D.). All of the error bars in the figures showed S.D. if they were not specially mentioned. All of the data, which compared 2 groups, were analyzed by unpaired t-tests. The curves in the acute Mg²⁺-intake experiments and metabolic syndrome models were analyzed by two-way ANOVA, followed by Bonferroni’s post hoc test. Statistical significance was defined as p < 0.05.

For additional information about the acute Mg²⁺ administration test, the fasting blood glucose assay, the fasting serum insulin assay, and the metabolic syndrome model, please read the section entitled “Supplementary Materials and Methods.”

RESULTS

RBC [Mg²⁺]_i, Plasma [Mg²⁺] and RBC [Mg]_total Indicated Dietary Mg²⁺ Adjustment

The plasma [Mg²⁺] and RBC [Mg]_total assays were carried out by conventional methods. For the RBC [Mg²⁺]_i measurement, we optimized the conventional method which detected intracellular ionized Mg²⁺ in platelets¹⁴⁾ and lymphocytes¹⁵⁾ using fluorescence probes and could only present fluorescence intensity. Our technique managed to quantitatively measure RBC [Mg²⁺]_i by directly outputting the Mg²⁺ concentration. This new method made it possible to quantitatively compare the level of RBC [Mg²⁺]_i with other Mg²⁺ pools in various physiological and pathological conditions. We conducted quality controls of RBC [Mg²⁺]_i, plasma [Mg²⁺] and RBC [Mg]_total assays (Supplementary Tables S1–S3). All assays showed acceptable coefficients of variance (CV) and recovery rates.

An Mg²⁺-deficient diet is a model that causes physiological Mg²⁺ deficiency in a short term (Fig. S1). Thus, we used this as the first model to analyze the responses of RBC [Mg²⁺]_i, plasma [Mg²⁺], and RBC [Mg]_total.

The protocol of the dietary Mg²⁺ deficiency is shown in Fig. 1A. The dramatic drop of total daily Mg intake and total daily Mg retention indicated that an Mg²⁺ loss occurred in the entire body of the rat (Fig. 1B). All of the 3 blood Mg²⁺ pools dropped significantly in this model (Fig. 1B), which suggested that the blood Mg²⁺ pools may represent the loss of Mg²⁺ in the entire body. Besides the rats’ data, the 3 blood Mg²⁺ pools in the mice model of dietary Mg²⁺ deficiency displayed similar patterns (Figs. 1C, D).

Fig. 1. The Responses of the Blood Mg²⁺ Compartments Induced by Dietary Mg²⁺ Deficiency

(A) Experiment design: rats aged 2 months were divided into 2 groups, with either normal food or Mg²⁺-deficient food, plus normal water for both groups, for a 4-week period. (B) The daily total Mg intake and Mg retention were assessed during this period. The blood Mg²⁺ compartments were sampled and measured at week 0 and week 4. (C) Experiment design: mice aged 3 months were fed with either normal food or Mg²⁺-deficient food for 4 weeks. (D) At week 0 and week 4, the blood Mg²⁺ compartments were sampled and measured. * p < 0.05, ** p < 0.01, *** p  < 0.001. (Color figure can be accessed in the online version.)

We next evaluated the abilities of the blood Mg²⁺ pools in representing Mg²⁺ metabolism in the dietary Mg²⁺ enrichment. The protocol is shown in Figs. 2A and C. In rats, only RBC [Mg²⁺]_i increased (p < 0.05; Fig. 2B), while the plasma [Mg²⁺] and RBC [Mg]_total did not change significantly. In mice, the blood Mg²⁺ pools showed similar results (Figs. 2C, D).

Fig. 2. The Responses of the Blood Mg²⁺ Compartments Induced by Dietary Mg²⁺ Enrichment

(A) Experiment design: rats aged 2 months were divided into 2 groups, with either normal water or water containing Mg²⁺, plus normal food for both groups, for 4 weeks. (B) At the beginning and the end of the experiment, the blood Mg²⁺ compartments were sampled and measured. (C) Experiment design: mice aged 3 months were fed with either normal water or water containing Mg²⁺, plus normal food, for 4 weeks. (D) At week 0 and week 4, the blood Mg²⁺ compartments were sampled and measured. * p < 0.05, ** p < 0.01. (Color figure can be accessed in the online version.)

The decrease in all of the 3 blood Mg²⁺ pools in the dietary-Mg²⁺-deficient test suggested that when there was Mg²⁺ intake deficit, the loss of Mg²⁺ was global and continuous (it was excreted by urine). A more interesting result was found in the dietary Mg²⁺ enrichment: Only RBC [Mg²⁺]_i responded to an excessive intake of Mg²⁺. Noticing that the Mg²⁺ intake was intermittent (separated by each meal), we designed an acute Mg²⁺ administration test in order to examine the metabolic characteristics of different blood Mg²⁺ pools, during a single intake period of Mg²⁺ (as shown below).

The Metabolic Characteristics of Blood Mg²⁺ Compartments

As shown in Fig. 3A, the plasma [Mg²⁺] (red line, right y-axis) changed the quickest and most dramatically after an acute Mg²⁺ intake: It showed a peak increase within 0.5 h, and dropped to the baseline after 3 h. The metabolism of RBC [Mg²⁺]_i was more moderate (blue line, left y-axis): It reached peak point 2 h after the injection, and the clearance of excess RBC [Mg²⁺]_i was not complete after 6 h (it still had a net increase of 12.18 ± 5.84% than control). The level of RBC [Mg]_total did not obviously change throughout the entire period (black line, left y-axis). More interesting phenomena were found when a second injection of Mg²⁺ was conducted 6 h after the first injection (Figs. 3B–D). The RBC [Mg²⁺]_i sequentially increased from the end point of the first injection (hour 6) to the peak point of the second injection (hour 8). At the second end-point (hour 12), the level of RBC [Mg²⁺]_i was much higher than the control and was even higher than the first end-point, thus showing a tendency of an accumulative effect. However, the plasma [Mg²⁺] exhibited no accumulation, although it did metabolize very quickly. The RBC [Mg]_total failed to display any significant change during the entire period. As a conclusion, different blood Mg²⁺ pools had various absorption rates and clearance rates of Mg²⁺. These different kinetic characteristics led to various responses of blood Mg²⁺ pools after acute Mg²⁺ intake, and may influence their representative capacities for the chronic metabolism of body Mg²⁺ (mentioned in the section of Discussion).

Fig. 3. The Responses of the Blood Mg²⁺ Compartments Induced by Acute Mg²⁺ Intake

The rats received an intraperitoneal (i.p.) injection of either saline or isotonic solution containing Mg²⁺. Blood was sampled intermittently following the administration of Mg²⁺ and assayed for the RBC [Mg²⁺]_i, plasma [Mg²⁺], and RBC [Mg]_total. (A) The metabolic curve of RBC [Mg²⁺]_i (blue line, left y-axis), plasma [Mg²⁺] (red line, right y-axis), and RBC [Mg]_total (black line, left y-axis) was measured after a single i.p. injection of Mg²⁺. The y-axis showed the net percentage-increase compared to the saline group at each time point. (B) The metabolic curve of the blood Mg²⁺ compartments after a double i.p. injection of Mg²⁺, with an interval of 6 h between the 2 injections. (C) The peak values of RBC [Mg²⁺]_i, plasma [Mg²⁺], and RBC [Mg]_total during the whole 12-h period are displayed. (D) The values of the 3 Mg²⁺ compartments at the end point (hour 12) are displayed. The error bars showed standard error of the mean (S.E.M.) in this figure. (Color figure can be accessed in the online version.)

Based on the results above and the references in Fig. S1A, we defined a metabolic model of the blood Mg²⁺ compartments (Graphical Abstract). The RBC [Mg²⁺]_i was classified as a representative Mg²⁺ pool: This kind of Mg²⁺ pools might be ideal to represent the chronic and homodromous body Mg²⁺ metabolism in physiological processes. The plasma [Mg²⁺] was defined as an unstable Mg²⁺ pool: this kind of Mg²⁺ pools might reflect the acute body Mg²⁺ metabolic curve, but might be easily interfered with and would not be sufficiently precise if they were assessed at a single time point. The RBC [Mg]_total was grouped into the conserved Mg²⁺ pools: They metabolized too slowly and might not be efficient as indicators of body Mg²⁺ levels. The results and logical model above suggest that the RBC [Mg²⁺]_i could more efficiently indicate the chronic metabolism of body Mg²⁺. Next, we verified this under pathological conditions.

RBC [Mg²⁺]_i Indicated Aging Process

The aging process is known to be associated with degenerations of many bodily functions. A few reports have shown that endogenous Mg²⁺ also decreased during aging, acting as a potential biomarker of gerontology (Fig. S1B). Thus, we studied whether our technique for the RBC [Mg²⁺]_i assay could represent the aging process. The total daily intake of Mg was the same for both young and aged rats (Fig. 4A). However, the daily Mg retention dropped significantly in the aged group, implying a severe Mg²⁺ deficit in the body. The RBC [Mg²⁺]_i from blood representative Mg²⁺ pools decreased significantly in the aged group (Fig. 4A), which matched the data of the Mg retention rate. However, the plasma [Mg²⁺] from blood unstable Mg²⁺ pools did not change. In the acute Mg²⁺ loading test, the metabolic curve of RBC [Mg²⁺]_i dropped significantly in the aged rats (Fig. 4B), suggesting an impairment of Mg²⁺ recruiting efficiency in the RBC [Mg²⁺]_i pool (Fig. 4A). In contrast to this, the metabolic curve of the plasma [Mg²⁺] did not change significantly (Fig. 4C). Similar results were found in mice: Only RBC [Mg²⁺]_i showed an aging-associated decline (Figs. 4D, E). As a conclusion, the RBC [Mg²⁺]_i, a kind of the representative Mg²⁺ pools, successfully represented the body Mg²⁺ loss in aged rats (feeding with normal Mg²⁺ diet). In another study of us, the RBC [Mg²⁺]_i diagnosis was effective in the therapy of dietary Mg²⁺ compensation in remedying the aging-associated memory decline.¹⁵⁾

Fig. 4. The Decline of RBC [Mg²⁺]_i Represented the Aging Process

(A) A comparison of different blood Mg²⁺ pools was conducted between young (black column, 4 months old male, n = 10) and aged rats (red column, 20 months old male, n = 10); the error bars showed S.D. After a single Mg²⁺ i.p. injection, the metabolic curves of RBC [Mg²⁺]_i (B) and plasma [Mg²⁺] (C) of the young and aged rats were compared (n = 6/group). The y-axis showed the percentage difference when compared to the 0 point at each time point. The error bars showed S.E.M. (D) A comparison of different blood Mg²⁺ pools was conducted between young (black column, 3 months old female, n = 11) and aged mice (red column, 18 months old female, n = 12). The error bars showed S.D. (E) The decreasing curve of RBC [Mg²⁺]_i during the aging process (female mice, n = 8–12). * p < 0.05, ** p < 0.01, *** p < 0.001. (Color figure can be accessed in the online version.)

RBC [Mg²⁺]_i Indicated Metabolic Syndrome Process

Metabolic syndrome encompasses conditions such as obesity, hypertension, and diabetes. It is another well-reported pathological process that is associated with endogenous Mg²⁺ loss. However, the body Mg²⁺ indicators for metabolic syndrome were still controversial (Fig. S1C). Here, we evaluated the functions of blood Mg²⁺ pools to indicate the process of metabolic syndrome. A high-calorie diet was used to establish metabolic syndrome model in the mouse. Data in Fig. 5A suggested that the model was successfully established. At 85 weeks, the levels of the fasting blood glucose rose from 5.10 mmol/L in the control group to 6.74 mmol/L in the high-calorie group (32% net increase). The fasting serum insulin rose from 0.40 ng/mL in the control group to 1.25 ng/mL in the high-calorie group (213% net increase). In terms of the evaluation of the blood Mg²⁺ pools, only the RBC [Mg²⁺]_i showed a pathology-induced response in the high-calorie group (Fig. 5B). Once again, the RBC [Mg²⁺]_i from representative Mg²⁺ pools successfully represented the Mg²⁺-metabolism-related pathological process.

Fig. 5. The Decline of RBC [Mg²⁺]_i Represented the Process of Metabolic Syndrome

Male ICR mice were fed with either a normal diet or a high-calorie diet from the age of 30 weeks (n = 11/group). The physiological indices (A), and blood Mg²⁺ pools (B) were compared between the 2 groups, at the age of 85 weeks. The time curves of RBC [Mg²⁺]_i (C), fasting blood glucose (D), and fasting serum insulin (E) from 30 to 85 weeks were displayed. * p < 0.05, ** p < 0.01, *** p < 0.001. (Color figure can be accessed in the online version.)

The time curves showed continuous metabolic changes for RBC [Mg²⁺]_i, fasting blood glucose and fasting serum insulin during the development of metabolic syndrome (Figs. 5C–E). The RBC [Mg²⁺]_i significantly decreased as early as 15 weeks after the high-calorie administration, while the fasting blood glucose and fasting serum insulin, 2 common blood indicators of metabolic syndrome, showed significant changes 30 weeks after the administration. This implied that RBC [Mg²⁺]_i might be used as a biomarker of metabolic syndrome, especially in the stages of early prevention and risk control.

DISCUSSION

This study first worked on a technique to accurately assay RBC [Mg²⁺]_i in murine by using a fluorescence probe. Previous researches reported fluorescence detections of intracellular ionized Mg²⁺ in platelets¹⁴⁾ and lymphocytes.¹⁵⁾ Those methods only presented fluorescence intensity without a calibration of Mg²⁺ concentration. Some of the other studies reported fluorescence detections of Mg²⁺ in neurons,¹⁶⁾ a type of adherent cell. Comparing the usage of fluorescence probe in neuron with in erythrocyte, it is totally different in the aspects of culture conditions, incubations, signal acquisitions, signal processing, and sample sizes. A lot of effort has been conducted to optimize the technique. Our study managed to quantitatively measure the RBC [Mg²⁺]_i by directly outputting the Mg²⁺ concentration. This new technique made it possible to quantitatively compare the level of RBC [Mg²⁺]_i with other Mg²⁺ pools in various physiological and pathological conditions.

With this technique, our study revealed the applicability of the measurement of RBC [Mg²⁺]_i as an ideal biomarker to represent body Mg²⁺ metabolism in physiological and pathological processes. Although some applications of RBC [Mg²⁺]_i in certain physiological and pathological processes were previously reported on (Fig. S1), our research managed to study the various representative capacities of RBC [Mg²⁺]_i on a unified technique platform and on a consistent evaluation system. The acute Mg²⁺ administration test was an original design. The different responses of blood Mg²⁺ pools indicated various metabolic characteristics. Thus, a metabolic model was established to explain the underlying mechanism of the various representative capacities of different blood Mg²⁺ pools.

It was reported that roughly 60–68% of Americans did not meet the recommended dietary allowance (RDA) criteria for total daily Mg intake.^17,18) What is more, even if normal level of Mg²⁺ intake is maintained, the risk of Mg²⁺ deficiency still exists due to potentially abnormal Mg²⁺ absorption and/or excretion. The body Mg retention is influenced by many factors (e.g., the constitution of the diet).¹⁹⁾ Therefore, the measurement of dietary Mg²⁺ levels is not enough; the determination of body Mg²⁺ status in response to the dietary Mg²⁺ adjustment is indubitably effective.

Our technique for RBC [Mg²⁺]_i determination combines the application of a fluorescence probe with flow cytometry. This technique has many advantages: First, the blood sampling is simple and fast, especially compared to tissue ionized Mg²⁺ measurements, which require NMR (Fig. S1). Second, the measurement accuracy is credible (Supplementary Table S3). Third, the representative capacity is excellent. This will be discussed in the following paragraphs.

At the beginning of our applicability research, we verified the representative capacity of RBC [Mg²⁺]_i in a dietary-Mg²⁺-deficient test. The RBC [Mg²⁺]_i showed similar responses with other blood Mg²⁺ pools (Fig. 1). However, in the dietary-Mg²⁺-enriched test, only the RBC [Mg²⁺]_i maintained its representative capacity (Fig. 2). The possible mechanism for this was revealed by the acute Mg²⁺ intake test (Fig. 3): After one intake, the clearance rate of RBC [Mg²⁺]_i was sufficiently slow enough to make the excess RBC [Mg²⁺]_i meet the time window of next Mg²⁺ intake. Thus, after a chronic and homodromous regulation of the Mg²⁺ intake, the RBC [Mg²⁺]_i might more easily exhibit an accumulative effect. This might explain why only the RBC [Mg²⁺]_i changed significantly after a chronic elevation of the Mg²⁺ intake (Fig. 2B). However, the absorption and clearance rate in the plasma [Mg²⁺] pool and RBC [Mg]_total pool were either too fast or too slow, which meant that they could not maintain their elevated levels when meeting the next time window of the Mg²⁺ intake, thus showing no accumulative effect and failing to respond to chronic Mg²⁺ intake enhancement.

Based on the results and inferences above, a logical metabolism model of the blood Mg²⁺ compartments was established (Graphical Abstract). The RBC [Mg²⁺]_i, plasma [Mg²⁺], and RBC [Mg]_total were separated into the representative Mg²⁺ pools, unstable Mg²⁺ pools, and conserved Mg²⁺ pools. The RBC [Mg²⁺]_i could indicate the body Mg²⁺ metabolism more efficiently than in other blood Mg²⁺ compartments. We proved this not only in physiological conditions but also in pathological conditions, as presented below.

Aging was reported to be associated with an endogenous decrease in Mg²⁺ (Fig. S1B). Our technique of RBC [Mg²⁺]_i measurement showed a significant and credible RBC [Mg²⁺]_i decline curve during aging (Fig. 4E), despite a diet containing normal level of Mg²⁺. Age-related Mg²⁺ deficits were found not only in rodents,²⁰⁾ but also in human.^{4,5,17,21–24)} Aging subjects are more likely to show reduced Mg²⁺ absorption and excess urinary loss, which may due to dysfunction of intestinal and renal mechanisms for Mg²⁺ balance.^17,25) The fact that the average Mg²⁺ intake of aging people being only one half of the RDA made the situation even worse.¹⁸⁾ The determination of RBC [Mg²⁺]_i may help to regulate the dietary Mg²⁺ supplement for the elderly. In one of our unpublished studies, the administration of an Mg²⁺-compound in aging people increased their RBC [Mg²⁺]_i levels and improved their memory performance.

It was interesting that plasma [Mg²⁺] levels did not change in aging rats of this study. Plasma Mg²⁺ pool are highly conserved with tight regulation.¹⁷⁾ In human, there might be a normomagnesemia despite of body Mg²⁺ deficiency, which was detected by Mg retention study.²⁶⁾ Plasma Mg²⁺ decreasing may occur only in the case of significant long lasting Mg²⁺ depletion.^20,27) In spite of the conservation of the plasma [Mg²⁺] level, significant skeleton Mg²⁺ decline was found to be associated with body Mg²⁺ deficit.^27,28) Skeleton Mg²⁺ pool, which contains two-thirds of body Mg²⁺, showed a more significant decrease compared to other tissue Mg²⁺ pools, in Mg²⁺ deficient rats.^12,20) Some evidences suggested that the skeleton could store Mg²⁺ and compensate the Mg²⁺ loss in the plasma Mg²⁺ pool.^29,30) The skeleton Mg²⁺ pool is probably a kind of representative Mg²⁺ pools, similar to RBC [Mg²⁺]_i.

The RBC [Mg²⁺]_i also decreased significantly in the metabolic syndrome mouse model (Fig. 5). In another of our unpublished studies, the enhancement of dietary Mg²⁺ supplement in the metabolic syndrome mice led to increased RBC [Mg²⁺]_i, decreased pathological indices, improved behavioral performance, and prolonged lifespan. Compared to the fasting blood glucose and serum insulin, the earlier onset of RBC [Mg²⁺]_i loss (Figs. 5C–E) may make itself more capable for the early stage diagnosis and prevention in clinical metabolic syndrome.

The mechanisms for body Mg²⁺ regulations are far from understanding. Physiological Mg²⁺ homeostasis is mediated by intestinal Mg²⁺ absorption, renal Mg²⁺ reabsorption and cellular Mg²⁺ transport across the membrane. There are several Mg²⁺ transporters controlling cellular Mg²⁺ influx, most of the which have not been cloned or been identified until recent years, including transporters transient receptor potential melastatin type 6/7 (TRPM6/7), solute carrier family 41 member 1/2 (SLC41A1/2), Mg²⁺ transporter 1 (MagT1) and Cyclin M1/2/3/4 (CNNM1/2/3/4), etc.³⁾ Further investigations are needed to explore the mechanisms of their functions and regulations in physiological and pathological processes. Mg²⁺ efflux involves Na⁺-dependent pathway via Na⁺/Mg²⁺ exchanger and Na⁺-independent pathways (exchange Mg²⁺ with Ca²⁺, Cl⁻ or Mn²⁺, etc.). Excess Mg²⁺ efflux in erythrocytes was found in several pathological processes associated with body Mg²⁺ depletion, such as hypertensive and chronic renal failure.³¹⁾ The genes and proteins corresponding to Mg²⁺ efflux have not been confirmed yet. However, some hormones were reported to influence Mg²⁺ transport, such as parathyroid hormone (PTH), vitamin D, insulin, calcitonin and adrenaline, etc.^17,32) It is likely that abnormal secretion of these hormones associated with aging (decrease in vitamin D status and increase in PTH levels, for instance) may induce Mg²⁺ depletion in the elderly.¹⁷⁾ Our future work would be focused on how the pathological processes influence RBC [Mg²⁺]_i, possibly through some of the transporters and hormones that we discussed above.

To summarize, the RBC [Mg²⁺]_i efficiently represented the changes of body Mg²⁺ metabolism affected by physiological and pathological conditions, including but not limited to dietary Mg²⁺ adjustment, aging, and metabolic syndrome. Our technique for the RBC [Mg²⁺]_i assay may help with basic physiological research, dietary Mg²⁺ regulation, and the clinical monitoring of Mg²⁺-metabolism-related pathology.

Acknowledgments

This work was financially supported by grants from National Key R&D Program of China (2018YFD0400204), the Key International S&T Cooperation Program of China (2016YFE113700), the European Union’s Horizon 2020 Research and Innovation Program (633589), and the National Natural Science Foundation of China (81471396). The authors wish to thank Dr. Song GE (Tsinghua University) for his support in the metabolic syndrome experiment.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

REFERENCES

• 1) Volpe SL. Magnesium in disease prevention and overall health. Adv. Nutr., 4, 378S–383S (2013).
• 2) Jahnen-Dechent W, Ketteler M. Magnesium basics. Clin. Kidney J., 5 (Suppl. 1), i3–i14 (2012).
• 3) de Baaij JH, Hoenderop JG, Bindels RJ. Magnesium in man: implications for health and disease. Physiol. Rev., 95, 1–46 (2015).
• 4) Resnick LM, Militianu D, Cunnings AJ, Pipe JG, Evelhoch JL, Soulen RL. Direct magnetic resonance determination of aortic distensibility in essential hypertension: relation to age, abdominal visceral fat, and in situ intracellular free magnesium. Hypertension, 30, 654–659 (1997).
• 5) Barbagallo M, Gupta RK, Dominguez LJ, Resnick LM. Cellular ionic alterations with age: relation to hypertension and diabetes. J. Am. Geriatr. Soc., 48, 1111–1116 (2000).
• 6) Serefko A, Szopa A, Wlaz P, Nowak G, Radziwon-Zaleska M, Skalski M, Poleszak E. Magnesium in depression. Pharmacol. Rep., 65, 547–554 (2013).
• 7) Grases G, Perez-Castello JA, Sanchis P, Casero A, Perello J, Isern B, Rigo E, Grases F. Anxiety and stress among science students. Study of calcium and magnesium alterations. Magnes. Res., 19, 102–106 (2006).
• 8) Barbagallo M, Resnick LM, Dominguez LJ, Licata G. Diabetes mellitus, hypertension and ageing: the ionic hypothesis of ageing and cardiovascular-metabolic diseases. Diabetes Metab., 23, 281–294 (1997).
• 9) Delva PT, Pastori C, Degan M, Montesi GD, Lechi A. Intralymphocyte free magnesium in a group of subjects with essential hypertension. Hypertension, 28, 433–439 (1996).
• 10) Dominguez LJ, Barbagallo M, Sowers JR, Resnick LM. Magnesium responsiveness to insulin and insulin-like growth factor I in erythrocytes from normotensive and hypertensive subjects. J. Clin. Endocrinol. Metab., 83, 4402–4407 (1998).
• 11) Arnaud MJ. Update on the assessment of magnesium status. Br. J. Nutr., 99 (Suppl. 3), S24–S36 (2008).
• 12) Martindale L, Heaton FW. Magnesium deficiency in the adult rat. Biochem. J., 92, 119–126 (1964).
• 13) Abernethy MH, Fowler RT. Micellar improvement of the calmagite compleximetric measurement of magnesium in plasma. Clin. Chem., 28, 520–522 (1982).
• 14) Fox CH, Timm EA Jr, Smith SJ, Touyz RM, Bush EG, Wallace PK. A method for measuring intracellular free magnesium concentration in platelets using flow cytometry. Magnes. Res., 20, 200–207 (2007).
• 15) Rijkers GT, Griffioen AW. Changes in free cytoplasmic magnesium following activation of human lymphocytes. Biochem. J., 289, 373–377 (1993).
• 16) Zhou H, Liu G. Regulation of density of functional presynaptic terminals by local energy supply. Mol. Brain, 8, 42 (2015).
• 17) Barbagallo M, Belvedere M, Dominguez LJ. Magnesium homeostasis and aging. Magnes. Res., 22, 235–246 (2009).
• 18) Ford ES, Mokdad AH. Dietary magnesium intake in a national sample of U.S. adults. J. Nutr., 133, 2879–2882 (2003).
• 19) Slutsky I, Abumaria N, Wu LJ, Huang C, Zhang L, Li B, Zhao X, Govindarajan A, Zhao MG, Zhuo M, Tonegawa S, Liu G. Enhancement of learning and memory by elevating brain magnesium. Neuron, 65, 165–177 (2010).
• 20) Xiong W, Liang Y, Li X, Liu G, Wang Z. Erythrocyte intracellular Mg²⁺ concentration as an index of recognition and memory. Scientific Reports, 6, 26975 (2016).
• 21) Gullestad L, Nes M, Ronneberg R, Midtvedt K, Falch D, Kjekshus J. Magnesium status in healthy free-living elderly Norwegians. J. Am. Coll. Nutr., 13, 45–50 (1994).
• 22) Martin BJ. The magnesium load test: Experience in elderly subjects. Aging (Milano), 2, 291–296 (1990).
• 23) Ryzen E, Nelson TA, Rude RK. Low blood mononuclear cell magnesium content and hypocalcemia in normomagnesemic patients. West. J. Med., 147, 549–553 (1987).
• 24) Satake K, Lee JD, Shimizu H, Ueda T, Nakamura T. Relation between severity of magnesium deficiency and frequency of anginal attacks in men with variant angina. J. Am. Coll. Cardiol., 28, 897–902 (1996).
• 25) Rude RK. Magnesium deficiency: a cause of heterogeneous disease in humans. J. Bone Miner. Res., 13, 749–758 (1998).
• 26) Gullestad L, Dolva LO, Waage A, Falch D, Fagerthun H, Kjekshus J. Magnesium deficiency diagnosed by an intravenous loading test. Scand. J. Clin. Lab. Invest., 52, 245–253 (1992).
• 27) Watchorn E, McCance RA. Subacute magnesium deficiency in rats. Biochem. J., 31, 1379–1390 (1937).
• 28) Rude RK, Gruber HE, Norton HJ, Wei LY, Frausto A, Kilburn J. Reduction of dietary magnesium by only 50% in the rat disrupts bone and mineral metabolism. Osteoporos. Int., 17, 1022–1032 (2006).
• 29) Aydın H. Magnesium Supplementation and Bone. Humana Press, New York, United States (2013).
• 30) Alfrey AC, Miller NL, Trow R. Effect of age and magnesium depletion on bone magnesium pools in rats. J. Clin. Invest., 54, 1074–1081 (1974).
• 31) Picado MJ, de la Sierra A, Aguilera MT, Coca A, Urbano-Márquez A. Increased activity of the Mg²⁺/Na+ exchanger in red blood cells from essential hypertensive patients. Hypertension, 23, 987–991 (1994).
• 32) Romani A. Regulation of magnesium homeostasis and transport in mammalian cells. Arch. Biochem. Biophys., 458, 90–102 (2007).