Donate Help Contact The AHA Sign In Home
American Heart Association
Hypertension
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Hypertension. 1996;28:433-439

This Article
Right arrow Abstract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Delva, P. T.
Right arrow Articles by Lechi, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Delva, P. T.
Right arrow Articles by Lechi, A.

(Hypertension. 1996;28:433-439.)
© 1996 American Heart Association, Inc.

Articles

Intralymphocyte Free Magnesium in a Group of Subjects With Essential Hypertension

Pietro T. Delva; Caterina Pastori; Maurizio Degan; Germana D. Montesi; Alessandro Lechi

the Institute of Internal Medicine, University of Verona (Italy).

Correspondence to Dr Pietro Delva, Medicina Interna C, Universita di Verona, Policlinico Borgo Roma, 37134 Verona, Italy.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Despite the importance of magnesium in essential hypertension, few data are available on the ionized intracellular concentration of this ion. We therefore studied intralymphocyte free intracellular magnesium (Mgi) in 32 untreated essential hypertensive subjects and 27 normotensive control subjects by means of a fluorimetric technique based on the use of the new magnesium-sensitive dye furaptra. We also measured intralymphocyte ionized calcium (Cai) with fura 2. No statistically significant differences were found in Mgi in hypertensive compared with normotensive subjects (essential hypertensive, 0.291±0.053 mmol/L; normotensive, 0.293±0.043 [mean±SD]). A statistically significant inverse correlation was established between Mgi and plasma triglycerides in essential hypertensive subjects (r=-.521, P=.002). The hypertensive group was arbitrarily divided into two subgroups according to plasma triglyceride levels (>2 [n=10] or <2 mmol/L [n=22]), and Mgi proved to be significantly lower in the subgroup with high plasma triglyceride levels compared with either the subgroup with normal triglycerides (P=.009; 95% confidence interval, 0.013-0.088) or the normotensive control group as a whole (P=.03; 95% confidence interval, 0.003-0.069) (high-triglyceride hypertensive subgroup, Mgi=0.256±0.045 mmol/L; normal-triglyceride hypertensive subgroup, Mgi=0.307±0.049). No statistically significant differences were found in Cai in hypertensive compared with normotensive subjects (hypertensive, 53±12 nmol/L; normotensive, 54±14). We did not find statistically significant correlations between Cai and plasma triglycerides, nor did we find any differences in Cai between the subgroup of hypertensive subjects with high plasma triglyceride levels and either the subgroup of hypertensive subjects with normal triglycerides or the normotensive control group as a whole. The discrepancies between our results in lymphocytes and data relating to either erythrocytes or platelets emphasize the need for caution before the results are extrapolated from one tissue to the other. The decreased Mgi levels in the subgroup of high-triglyceride hypertensive subjects may suggest a role for magnesium in plurimetabolic syndrome.


Key Words: magnesium • calcium • lymphocytes • hypertension, arterial • triglycerides


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Over the past decade, a substantial body of experimental, epidemiological, and clinical data strongly suggest that magnesium plays an important role in essential hypertension (see Reference 1 for review). Epidemiological studies have described an inverse relationship between dietary intake of magnesium and blood pressure2 although there is disagreement about the actual beneficial effect of dietary magnesium supplementation on blood pressure.3 4 5 Clinical studies focusing on the measurement of plasma and tissue magnesium have suggested an overall deficiency of total magnesium.6 7 Measurements of Mgi, however, are not well documented in the literature, and the available data have a number of technical limitations. The most important of these is the measurement of total instead of ionized Mgi, only the latter being important for enzyme action. Also, investigators studying ionized Mgi in essential hypertension have used nuclear magnetic resonance, a technique potentially influenced by intracellular pH.8

The recent synthesis of a new fluorescent magnesium-sensitive dye9 has allowed the measurement of Mgi in nucleated cells. We therefore measured Mgi in a group of subjects with essential hypertension and a group of normotensive control subjects. We also measured Cai, mainly to exclude substantial Cai variations that could theoretically affect magnesium determinations.10 Furthermore, we sought to establish whether there was any correlation between Mgi and Cai and the main variables of glucose and lipid metabolism, for which a role has been postulated in the regulation of Mgi and Cai. We also evaluated the possible links between renin, sodium excretion, and Mgi and Cai.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Mgi and Cai were measured in 32 subjects with essential hypertension and 27 healthy normotensive subjects with no family history of hypertension recruited from among hospital medical and paramedical staff and matched for age and sex. After hospitalization, the diagnosis of essential hypertension was made in hypertensive subjects after complete clinical, laboratory, and instrumental examinations, with particular care being taken to exclude secondary forms of hypertension and chronic alcoholism. Diabetic patients were also excluded from the study.

The blood samples used for determination of Mgi and Cai concentrations were taken from all subjects in the morning after they had fasted overnight. All forms of drug treatment, including the use of contraceptive pills, were discontinued at least 3 weeks before the samples were taken, but the majority of the subjects had never been on treatment. As diuretics are known to affect Mgi,11 subjects on such treatment at any time were excluded from the study. All subjects were on an unrestricted diet.

The hypertensive group was arbitrarily divided into two subgroups according to plasma triglyceride levels greater than (n=10) or less than (n=22) 2 mmol/L.

Routine Laboratory Tests
Plasma total cholesterol, HDL cholesterol, triglycerides, blood glucose, and uric acid were measured with an autoanalyzer technique (Technicon DAX 96, Miles Inc), as was total plasma magnesium (911 analyzer, Hitachi Ltd). HDL cholesterol was calculated with the Friedewald formula. Insulin was measured by radioimmunoassay. Plasma active renin was measured by radioimmunoassay in all subjects after they had sat 3 hours in the upright position.12 Daily 24-hour urine collections were analyzed for sodium, potassium, total magnesium, and creatinine excretions. Creatinine was measured in all urine samples as an estimate of the completeness of collection. Urinary total magnesium, sodium, and potassium were assessed by flame photometry.

Measurement of Mgi and Cai
To measure Mgi and Cai, we used the method of Ng et al13 with extensive modifications. Peripheral blood lymphocytes were isolated as follows. Total blood was diluted with RPMI-1640 medium (HEPES modification, glucose-free) and layered carefully onto Histopaque 1077 and centrifuged for 30 minutes at 400g. The layer of lymphocytes was carefully aspirated and washed twice in RPMI-1640 for 10 minutes at 150g. The cells thus obtained were allowed to sediment for 30 minutes in culture flasks. The supernatant was then transferred into tubes and centrifuged, and the lymphocytes thus obtained were resuspended in the same medium. The percentage of lymphocytes always exceeded 95%, and the vitality assessed by trypan blue exclusion was always better than 97%. Lymphocytes were counted by means of a Coulter counter. Three separate aliquots of lymphocytes (6x106 cells each) were suspended in RPMI-1640 and 0.1% bovine serum albumin (vol/vol) with the cell-permeant dye furaptra-acetoxymethyl ester (Molecular Probes Inc) (10 µmol/L) for 1 hour at 37°C. After centrifugation, the cells were washed twice in the same medium for removal of extracellular dye and resuspended in the same medium at room temperature for 45 minutes for complete deesterification of the dye.

For measurement of Mgi, the buffer contained (mmol/L) NaCl 140, KCl 5, CaCl2 1.8, MgSO4 0.8, HEPES 15, and D-glucose 5 (pH 7.4 at 31°C). Typical fluorescence readings from furaptra-loaded cells are shown in Fig 1Down. After the RPMI-1640 medium was washed off, the cells were added to the above prewarmed medium (31°C) just before fluorimetric measurements were performed. Fluorescence emission at 510 nm (slit width, 10 nm) was measured with alternate excitation at 335 and 370 nm (slit width, 10 nm) within a thermostatically controlled cuvette holder (31°C) in a Hitachi F-2000 fluorescence spectrophotometer. Autofluorescence contributed less than 1% of total fluorescence values. One-centimeter quartz cuvettes were used for all experiments. After the initial fluorescence emission resulting from excitation at 335 and 370 nm was read, 5 mmol/L EDTA and 5 mmol/L EGTA were added to the cuvette, and since extracellular Mg2+ and Ca2+ were chelated in the medium, a rapid-step change in fluorescence at both wavelengths occurred because of dye leakage from inside the cells (see Fig 1Down). The immediate (<10 seconds) change in fluorescence intensities at both wavelengths after the addition of EDTA and EGTA was considered for calculation of free resting Mgi. Triton X-100 was then added at a final concentration of 0.1% (vol/vol) to lyse the cells, and since the cells were in medium containing EDTA and EGTA, the minimum fluorescence ratio, Rmin, could be determined. Subsequently, MgSO4 (100 mmol/L) was added to obtain the maximum fluorescence ratio, Rmax. Mgi was measured in triplicate and calculated as follows:

(E1)
where Kd is 1.5 mmol/L,9 and Rmin is the fluorescence ratio at 335/370 nm for uncomplexed furaptra (zero magnesium). Rmax is the ratio of fluorescence at 335/370 nm for furaptra saturated with Mg2+. Sf and Sb are the fluorescence intensities at 370 nm for furaptra with zero Mg2+ and excess Mg2+, respectively. R is the ratio of fluorescence at 335/370 nm of the sample to be measured. We obtained the parameters Rmin, Rmax, Sf, and Sb from an in vitro calibration. We had previously attempted to obtain an in vivo calibration with 4-bromo-A23187 to equilibrate Mg2+ across the plasma membrane, but with this ionophore, the 335/370 nm ratio was still considerably lower than the 335/370 nm ratio obtained after cell lysis. Furthermore, this latter in vitro method yielded superior results in terms of reproducibility of the calibration procedure. We plotted increasing concentrations of magnesium-saturated Furaptra against the fluorescence intensity ratio at 335 and 370 nm and found that ratio values were consistently higher than 16 nmol/L furaptra. Consequently, magnesium-free measurements in lymphocytes were accepted only if the intracellular furaptra concentration exceeded this value.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 1. Typical experiment for Mgi measurement. Fluorescence tracings are shown of furaptra-loaded lymphocytes when excited at 335 nm (thick lines) and 370 nm (thin lines). Additions were as follows: A, 5 mmol/L EDTA and 5 mmol/L EGTA; B, 0.1% (vol/vol) Triton X-100; C, 100 mmol/L MgSO4.

The Cai of lymphocytes was measured with fura 2.14 Three separate aliquots of cells (6x106 each) were incubated with fura 2-acetoxymethyl ester (5 µmol/L) for 1 hour at 37°C and then washed twice and left for 45 minutes at room temperature before any measurements were made. For fluorimetric measurements, the same buffer used for the Mg2+ assay was used, and the cells were added to prewarmed medium with a technique similar to that used for the Mg2+ assay. EGTA (10 mmol/L) was added to chelate Ca2+ in the medium, thus producing a desaturation of fura 2 that leaked from inside the cells and giving a rapid-step change in fluorescence emission at 510 nm with excitation set at both 340 and 380 nm. This provided the ratio values for the calculation of resting Ca2+ with an equation similar to Equation 1, with a Kd value for the Ca2+–fura 2 complex of 225 nmol/L.14 An in vitro calibration was performed similar to that used for Mg2+ calibration, whereas for the Rmax determination, 10 mmol/L CaCl2 was added. The intra-assay and interassay coefficients of variation were 5.8% and 3.9% for Mg2+ and 4.6% and 12.7% for Ca2+, respectively.

Finally, as calcium binds furaptra more tightly than magnesium,9 some investigators have suggested correcting for calcium binding to furaptra. In the present conditions, given a mean Cai value of about 60 nmol/L, which is 1000 to 1200 times lower than the apparent Kd for calcium binding to furaptra, only 0.05% of the furaptra will be complexed with calcium. Despite this, we calculated Mgi with the correction for Cai proposed by London15 :

(E2)
where ß is the ratio of fluorescence at 375 nm for furaptra-uncomplexed/Mg-complexed; R, Rmax, and Rmin are the fluorescence ratios (335/375 nm) of the given sample with saturating magnesium and saturating EGTA, respectively; and KdMg and KdCa are apparent Kd values for magnesium (1.5 mmol/L) and calcium (50 µmol/L) complexes of furaptra, respectively.

Statistical Analysis
Results are expressed as mean±SD. As no evidence of a non-normal distribution or inequality of variances was present in the variables considered, comparison between groups was performed with Student's t test and considered statistically significant when the probability of the null hypothesis was less than at least 5%. When appropriate, ANOVA (with the Scheffe correction for multiple comparisons) was performed. Confidence limits for differences in means were also provided. Correlations between two variables were studied with the linear regression method, and where indicated, step-up multivariate regression with the partial F test was performed to select additional variables for inclusion in the model.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
Characteristics of the Study Population as a Whole
The group of essential hypertensive subjects had significantly higher BMI values than the normotensive control subjects (TableDown). Plasma total magnesium and daily urinary magnesium excretion did not differ significantly in hypertensive and control subjects. Fasting glucose, triglycerides, total cholesterol, and serum uric acid, although not reaching pathological levels, were significantly increased in hypertensive compared with normotensive subjects. We identified three subgroups of subjects by relating 24-hour sodium excretion to plasma renin activity, as shown in Fig 2Down. We found no statistical differences in both Mgi and Cai in low-, normal-, and high-renin hypertensive subjects (ANOVA test; Mgi: F ratio, 1.35, P=NS; Cai: F ratio, 1.05, P=NS).


View this table:
[in this window]
[in a new window]
  		Table 1. Main Clinical and Metabolic Variables in the Study Subjects



View larger version (19K):
[in this window]
[in a new window]
  		Figure 2. Relationship between upright plasma active renin (pg/mL) and 24-hour urinary sodium excretion (mmol/24 h) in subjects with essential hypertension. Dotted lines represent normal range obtained in a sample of normotensive control subjects with variable sodium intakes.

Characteristics of Subgroups of Essential Hypertensive Subjects With High or Normal Plasma Triglyceride Levels
The subgroup of essential hypertensive subjects with high triglycerides exhibited higher BMI, total cholesterol, triglycerides, insulin, and serum uric acid than normotensive control subjects, as shown in the TableUp. The same subgroup differed from the normal-triglyceride hypertensive subgroup in terms of lower diastolic pressure, higher total cholesterol, higher triglycerides, and higher fasting serum glucose (TableUp).

Mgi in Essential Hypertensive and Normotensive Subjects
Individual values for Mgi in essential hypertensive and normotensive subjects are shown in Fig 3Down. Group means were not significantly different in the normotensive (0.293±0.043 mmol/L) and hypertensive (0.291±0.053 mmol/L) subjects. When the hypertensive group was divided into two subgroups according to BMI (>26 or <26 kg/m2), Mgi showed no statistically significant differences in normal-weight as opposed to overweight hypertensive subjects (BMI <26 kg/m2 [n=19], 0.295±0.059 mmol/L; BMI >26 kg/m2 [n=13], 0.317±0.052 mmol/L). When both the hypertensive and normotensive groups were divided into subgroups according to sex, Mgi did not prove significantly different in men and women (essential hypertensive subjects: men [n=23], 0.293±0.044 mmol/L; women [n=9], 0.285±0.073 mmol/L; normotensive control subjects: men [n=15], 0.280±0.031 mmol/L; women [n=12], 0.309±0.040 mmol/L).



View larger version (17K):
[in this window]
[in a new window]
  		Figure 3. Scatterplot shows single values of Mgi in essential hypertensive (n=32) and normotensive control (n=27) subjects.

Mgi in Essential Hypertensive Subjects With High or Normal Plasma Triglyceride Levels
When the hypertensive group was divided into two subgroups on the basis of plasma triglyceride levels (>2 or <2 mmol/L), Mgi was significantly decreased in the subgroup with high plasma triglycerides compared with either the subgroup with normal triglycerides (P=.009; 95% confidence interval, 0.013-0.088) or the entire normotensive control group (P=.03; 95% confidence interval, 0.003-0.069) (high-triglyceride hypertensive subgroup [n=10]: Mgi, 0.256±0.045 mmol/L; normal-triglyceride hypertensive subgroup [n=22]: Mgi, 0.307±0.049 mmol/L; normotensive control group [n=27]: Mgi, 0.293±0.043 mmol/L). Fig 4Down presents values for Mgi in both the normal- and high-triglyceride hypertensive subgroups.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 4. Scatterplot shows individual values of Mgi in two subgroups of essential hypertensive subjects characterized by plasma triglycerides greater than or less than 2 mmol/L.

Correlations Between Mgi and Clinical and Metabolic Variables
In univariate regression, Mgi was negatively associated with plasma triglycerides in subjects with essential hypertension (n=32, r=-.521, P=.002), as shown in Fig 5Down. Because of the possible effect of other variables on this regression, we performed a stepwise multiple regression procedure including age, BMI, systolic and diastolic pressures, plasma and urinary total magnesium, total and HDL cholesterol, and plasma fasting glucose and insulin. None of these variables influenced the correlation between Mgi and plasma triglycerides. Analysis of residuals confirmed that a simple linear model was most appropriate for the data. In normotensive control subjects, Mgi was not correlated with plasma triglycerides (n=27, r=-.366, P=NS).



View larger version (20K):
[in this window]
[in a new window]
  		Figure 5. Plot of statistically significant inverse correlation between Mgi and plasma triglycerides in essential hypertensive subjects.

We also evaluated the univariate regressions for Mgi and the following variables within both essential hypertensive and normotensive control subjects: age, BMI, systolic and diastolic pressures, plasma total magnesium and calcium, total and HDL cholesterol, and plasma fasting glucose and insulin; we did not find any statistically significant correlation.

Cai in Essential Hypertensive and Normotensive Subjects
No statistically significant differences in Cai were detected between essential hypertensive (n=32; Cai, 53±12 nmol/L) and normotensive control (n=27; Cai, 54±14 nmol/L) subjects. When the hypertensive group was divided into two subgroups according to BMI (>26 or <26 kg/m2), Cai showed no statistically significant differences in normal-weight as opposed to overweight hypertensive subjects (BMI <26 kg/m2 [n=19], 52±12 nmol/L; BMI >26 kg/m2 [n=13], 53±11 nmol/L). When both the hypertensive and normotensive groups were divided into subgroups according to sex, Cai did not prove significantly different in men and women (essential hypertensive subjects: men [n=23], 54±12 nmol/L; women [n=9], 53±14 nmol/L; normotensive control subjects: men [n=15], 52±12 nmol/L; women [n=12], 50±13 nmol/L).

Cai in Essential Hypertensive Subjects With High or Normal Plasma Triglyceride Levels
When the hypertensive group was divided into two subgroups on the basis of plasma triglyceride levels (> or <2 mmol/L), Cai was not significantly different in the subgroup with high plasma triglycerides compared with either the subgroup with normal triglycerides or the entire normotensive control group (high-triglyceride hypertensive subgroup [n=10]: Cai, 49±13 nmol/L; normal-triglyceride hypertensive subgroup [n=22]: Cai, 53±11 nmol/L; normotensive control group [n=27]: Cai, 54±14 nmol/L).

Correlations Between Cai and Clinical and Metabolic Variables
In univariate regression, Cai was not associated with Mgi within both essential hypertensive and normotensive control subjects (r=.202 and r=.064, respectively). We evaluated the univariate regressions for Cai and the following variables within both essential hypertensive and normotensive control subjects: age, BMI, systolic and diastolic pressures, plasma total calcium and magnesium, plasma triglycerides, total and HDL cholesterol, and plasma fasting glucose and insulin; we did not find any statistically significant correlation.


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
We believe this to be the first study to provide measurements of ionized magnesium in nucleated cells in essential hypertensive subjects. Previous data mainly focused on total16 17 rather than ionized Mgi, thus preventing direct comparison with our data. Only a few investigators refer to ionized magnesium, and then only in non-nucleated cells. Resnick et al,18 using nuclear magnetic resonance, were the first to describe a decrease in ionized erythrocyte magnesium in a group of hypertensive patients. Woods et al,19 who also used nuclear magnetic resonance, were unable to find any differences in Mgi in erythrocytes of a group of essential hypertensive individuals.

In the present study, we were unable to find any differences in Mgi in a group of essential hypertensive subjects compared with normotensive control subjects. We cannot explain the discrepancies of our results with the results of other authors, but we can hypothesize that differences in technical approach or subject selection may partly account for them. Previous data on ionized magnesium refer to erythrocytes, and the discrepancies in the results may be due to differences in magnesium handling by the two cell types. Lymphocytes may differ in their ability to maintain ionized Mgi concentrations as a consequence of either a different activity of magnesium membrane transport systems or different sensitivity to circulating factor or factors capable of inducing Mgi variations.20 21

As far as the potential drawbacks of the fluorimetric technique used are concerned, binding of Cai to furaptra theoretically cannot affect the fluorescence of the dye because of the extremely low levels of Cai.22 Despite this, however, we measured Cai in separate aliquots of cells in the same subjects and calculated Mgi both with and without correction for Cai. We were unable to find any differences between the two procedures.

Last, it is theoretically conceivable that the relatively high Kd value of furaptra for magnesium might potentially be responsible for a lack of sensitivity of the method to detect small Mgi variations compared with nuclear magnetic resonance determinations.

Differences in subjects may include race, sex, and the degree of hypertension and may be crucial in explaining discrepancies among results. This aspect may be overemphasized by the relatively small number of subjects studied so far. The subjects of the present study were exclusively white, and the hypertensive group included mildly hypertensive subjects.

Future studies are needed to ascertain whether lymphocytes constitute a particular cell model with respect to the handling of ionized magnesium and whether differences in subject selection may be responsible for the discrepancies found in the literature. High Cai levels in erythrocytes and platelets of hypertensive individuals have been reported in the literature,23 24 but no significant increase in Cai has been shown in white blood cells in hypertensive subjects.25 26 Our results are in agreement with the latter studies, as we cannot demonstrate any differences in Cai levels in our hypertensive group compared with the normotensive control subjects. As previously suggested by some authors,25 the discrepancies among the results in platelets, erythrocytes, and leukocytes emphasize the need for caution before the results are extrapolated from one tissue to another.

We also attempted to investigate factors that may theoretically influence Mgi homeostasis. Some researchers27 have described a decrease in total erythrocyte Mgi after an oral glucose load. These data, together with the widely accepted notion of plasma and erythrocyte total and ionized magnesium deficiencies in non–insulin-dependent and insulin-dependent diabetes mellitus,28 29 30 have led to the assumption that there is a strong link between glucose metabolism and magnesium homeostasis. We could not find any correlation between Mgi and either fasting insulin or fasting glucose plasma levels. However, we should point out that the absence of such correlations would not in itself be sufficient to exclude a relationship between magnesium homeostasis and glucose metabolism, as no data relating to the euglycemic clamp technique or oral glucose tolerance test are available in the present study.

A negative correlation was found between ionized Mgi and plasma triglyceride levels in hypertensive subjects. This correlation means that approximately 25% of the Mgi variability is explained by its linear relation to plasma triglycerides. This finding led us to perform a subanalysis of the hypertensive group, which we arbitrarily divided into two subgroups on the basis of plasma triglyceride levels. We could not divide the normotensive control subjects into subgroups according to plasma triglycerides because only 3 of 27 subjects were characterized by high plasma triglyceride levels. The same reason should explain the lack of a statistically significant inverse correlation in the normotensive group between plasma triglycerides and Mgi.

The high-triglyceride hypertensive subgroup was characterized by significantly decreased levels of Mgi compared with both the normal-triglyceride hypertensive subgroup and the entire normotensive control group. Examination of the biochemical features of these high-triglyceride subjects also showed that they were characterized by plasma glucose and uric acid levels that, although not being clearly pathological, were significantly increased compared with the other hypertensive subgroup and the normotensive group. Furthermore, the high-triglyceride hypertensive subjects had elevated total cholesterol and showed a trend toward higher fasting insulin and lower HDL cholesterol, although these did not reach statistical significance, than the normal-triglyceride hypertensive subjects. This subgroup of hypertensive individuals with high triglycerides exhibit the main features of the so-called plurimetabolic syndrome.31 According to Larsson et al,32 this syndrome is found in approximately 30% of a nonselected hypertensive population. The risk of coronary heart disease appears to be significantly increased in hypertensive subjects with metabolic disturbances compared with those without, irrespective of their hypertensive state.32 Thus, bearing in mind the increasing evidence of the importance of magnesium in coronary heart disease,33 we suggest that a low ionized Mgi level might in some way be linked to the augmented risk characterizing plurimetabolic syndrome. We feel that this result is in agreement with and constitutes another aspect of the theory proposed by Resnick34 in which a defect in cell ion handling characterizes different pathological processes, such as hypertension, obesity, and non–insulin-dependent diabetes mellitus.

Furthermore, this finding might account for the discrepancies in Mgi levels vis-a-vis previous studies, owing to the above-mentioned differences in subject selection such as the percentages of subjects with plurimetabolic syndrome. Consequently, we can speculate that in our hypertensive group, the percentage of high-triglyceride subjects was relatively low compared with the group studied by Resnick et al,18 leading to a mean free Mgi not statistically different from that in the group of control subjects.

As far as ionized Cai is concerned, it has been shown that Cai is linked to peripheral insulin sensitivity and oral glucose–induced alterations of cytosolic free calcium.35 We could not find any differences in Cai in the subgroup of hypertensive subjects with high triglycerides compared with the subgroup with normal triglycerides. Furthermore, Cai and plasma triglycerides were not correlated. From these results, it appears that free cytosolic levels of magnesium and calcium in lymphocytes did not share interrelationships with the same metabolic variables. Moreover, we could not find any statistical linear correlation between the cytosolic levels of the two ions measured simultaneously in the same subjects. Therefore, in lymphocytes it appears that ionized cytosolic calcium and magnesium are not interrelated by simple linear correlations. In our opinion, this means that in lymphocytes, the links between the two ions may not be elucidated by simple measurement of their basal cytosolic concentrations.

Resnick et al36 have suggested that calcium and magnesium homeostasis may be regulated by the renin-angiotensin-aldosterone system. We divided our hypertensive subjects into three subgroups according to their plasma renin and daily sodium excretion and did not find any statistically significant differences in lymphocyte magnesium and calcium among the groups. Further studies are required for assessment of whether in salt-sensitive hypertensive subjects a difference in either magnesium or calcium lymphocyte concentration is present.

In conclusion, the discrepancies in the results between erythrocytes and lymphocytes and the lack of direct relationships between blood cells and vascular smooth muscle cells should advise caution in the interpretation of the data. Worthy of note is the finding that a subgroup of hypertensive subjects characterized by plurimetabolic syndrome showed decreased Mgi levels and normal Cai levels.

The relationship between total magnesium and plasma lipids, although imperfectly understood, has been known for some time.37 On the basis of the available data, we cannot account for the association between triglycerides and Mgi, but it is interesting to note that in a study by Rasmussen et al,38 patients with ischemic heart disease after 3 months of treatment with oral magnesium were characterized by ameliorated plasma lipid status, mainly as far as plasma triglycerides were concerned. Furthermore, an important feature of hyperlipemia associated with magnesium deficiency in experimental animal models is the accumulation of triglyceride-rich lipoproteins.39


*    Selected Abbreviations and Acronyms
 
BMI = body mass index
Cai = intralymphocyte free calcium
HDL = high-density lipoprotein
Mgi = intralymphocyte free magnesium

Received March 18, 1996; first decision April 25, 1996; accepted April 25, 1996.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Altura BM, Zhang A, Altura BT. Magnesium, hypertensive vascular diseases, atherogenesis, subcellular compartmentation of Ca2+ and Mg2+ and vascular contractility. Miner Electrolyte Metab. 1993;19:323-336.[Medline] [Order article via Infotrieve]
  2. Joffres MR, Reed DM, Yano K. Relationship of magnesium intake and other dietary factors to blood pressure: the Honolulu Heart Study. Am J Clin Nutr. 1987;45:469-475.[Abstract/Free Full Text]
  3. Cappuccio FP, Markandu ND, Beynon GW, Shore AC, Sampson B, MacGregor GA. Lack of effect of oral magnesium on high blood pressure: a double blind study. Br Med J. 1985;291:235-238.
  4. Henderson DG, Schierue J, Schudt J. Effect of magnesium supplementation on blood pressure and electrolyte concentrations in hypertensive patients receiving long-term diuretic treatment. Br Med J. 1986;293:664-665.
  5. Nowson CA, Morgan TO. Magnesium supplementation in mild hypertensive patients on a moderately low sodium diet. Clin Exp Pharmacol Toxicol. 1989;16:299-302.
  6. Johnson CJ, Peterson DR, Smith EK. Myocardial tissue concentrations of magnesium and potassium in men dying suddenly from ischemic heart disease. Am J Clin Nutr. 1979;32:967-970.[Abstract/Free Full Text]
  7. Albert DG, Morita Y, Iseri LT. Serum magnesium and plasma sodium levels in essential vascular hypertension. Circulation. 1958;17:761-764.[Medline] [Order article via Infotrieve]
  8. Veniero JC, Gupta RK. NMR measurements of intracellular ions in living systems. Annu Rep NMR Spect. 1992;24:219-266.
  9. Raju B, Murphy E, Levy LA, Hall RD, London RE. A fluorescent indicator for measuring cytosolic free magnesium. Am J Physiol. 1989;256:C540-C548.[Abstract/Free Full Text]
  10. Hurley TW, Ryan MP, Brinck RW. Changes of cytosolic Ca2+ interfere with measurements of cytosolic Mg2+ using mag-fura-2. Am J Physiol. 1992;263:C300-C307.[Abstract/Free Full Text]
  11. Sheehan J, White A. Diuretic-associated hypomagnesemia. Br Med J. 1982;285:1157-1159.
  12. Wathen LK, Wuorala KW, Wathen MW. Validation of an immunoradiometric assay to measure plasma levels of active renin. J Clin Lab Anal. 1991;5:284-292.[Medline] [Order article via Infotrieve]
  13. Ng LL, Davies JE, Garrido MC. Intracellular free magnesium in human lymphocytes and the response to lectins. Clin Sci. 1991;80:539-547.[Medline] [Order article via Infotrieve]
  14. Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985;260:3440-3450.[Abstract/Free Full Text]
  15. London RE. Methods for measurement of intracellular magnesium: NMR and fluorescence. Annu Rev Physiol. 1991;53:241-258.[Medline] [Order article via Infotrieve]
  16. Paolisso G, Passariello NS, Torecca R, Buoninconti R, Varricchio M, D'Onorio F. Impaired insulin-mediated erythrocyte magnesium accumulation in essential hypertension. Clin Sci. 1987;73:535-539.[Medline] [Order article via Infotrieve]
  17. Touyz RM, Milne FJ, Seftel HC, Reinach SG. Magnesium, calcium, sodium and potassium status in normotensive and hypertensive Johannesburg residents. S Afr Med J. 1987;72:377-381.[Medline] [Order article via Infotrieve]
  18. Resnick LM, Gupta RK, Laragh JH. Intracellular free magnesium in erythrocytes of essential hypertension: relation to blood pressure and serum divalent cations. Proc Natl Acad Sci U S A. 1984;81:6511-6515.[Abstract/Free Full Text]
  19. Woods KL, Walmsley D, Heagerty AM, Turner DL, Lian LY. 31P-nuclear magnetic resonance measurement of free erythrocyte magnesium concentration in man and its relation to blood pressure. Clin Sci. 1988;74:513-517.[Medline] [Order article via Infotrieve]
  20. Lindner A, Kenny M, Meacham AJ. Effects of a circulating factor in patients with essential hypertension on intracellular free calcium in normal platelets. N Engl J Med. 1987;316:509-513.[Abstract]
  21. Wright GL, Rogerson ME, McCumbee WD. Stimulation of aortic tissue calcium uptake by an extract of spontaneously hypertensive rat erythrocytes possessing hypertensive properties. Can J Physiol Pharmacol. 1986;64:1515-1520.[Medline] [Order article via Infotrieve]
  22. Tsien RY, Pozzan T, Rink TJ. Calcium homeostasis in intact lymphocytes: cytosolic free calcium monitored with a new intracellular trapped fluorescent indicator. J Cell Biol. 1982;94:325-334.[Abstract/Free Full Text]
  23. Zidek W, Knorr M, Losse H, Baumgart P, Vetter H. Sodium and calcium in lymphocytes and erythrocytes in essential hypertension. J Hypertens. 1983;1(suppl 2):3-5.
  24. Erne P, Bolli P, Burgisser E, Buhler FR. Correlation of platelet calcium with blood pressure: effect of antihypertensive therapy. N Engl J Med. 1984;310:1084-1088.[Abstract]
  25. Bing RF, Heagerty AM, Jackson JA, Thurston H, Swales JD. Leukocyte ionized calcium and sodium content and blood pressure in humans. Hypertension. 1986;8:483-488.[Abstract/Free Full Text]
  26. Shore AC, Beynon GW, Jones JC, Markandu ND, Sagnella GA, MacGregor GA. Mononuclear intracellular free calcium: does it correlate with blood pressure? J Hypertens. 1985;3:183-187.[Medline] [Order article via Infotrieve]
  27. Resnick LM, Gupta RK, Gruenspan HL, Laragh JH. Intracellular free magnesium in hypertension: relation to peripheral insulin resistance. J Hypertens. 1988;6(suppl 4):S199-S201.
  28. Mather HM, Nishet JA, Burton GH, Poston G, Bland JM, Bailey PA, Pilkington TR. Hypomagnesemia in diabetes. Clin Chim Acta. 1979;95:235-242.[Medline] [Order article via Infotrieve]
  29. Sjogren A, Floren CH, Nilsson A. Magnesium deficiency in IDDM related to level of glycosylated hemoglobin. Diabetes. 1986;35:459-463.[Abstract]
  30. Paolisso G, Scheen A, Cozzolino D, Di Maro G, Varricchio M, D'Onofrio F, Lefebvre PJ. Changes in glucose turnover parameters and improvement of glucose oxidation after 4-week magnesium administration in elderly noninsulin-dependent (type II) diabetic patients. J Clin Endocrinol Metab. 1994;78:1510-1514.[Abstract]
  31. Reaven GM. Role of insulin resistance in human disease. Diabetes. 1988;37:1495-1507.
  32. Larsson B, Welin L, Eriksson H. The plurimetabolic syndrome and the study of men born in 1913, Goteborg, Sweden. In: Crepaldi G, Tiengo A, Manzato E, eds. Diabetes, Obesity and Hyperlipemias: The Plurimetabolic Syndrome. Amsterdam, Netherlands: Elsevier Science Publishers; 1993:17-24.
  33. Woods KL, Fletcher S, Roffe C, Haider Y. Intravenous magnesium sulphate in suspected acute myocardial infarction: results of the second Leicester Intravenous Magnesium Intervention Trial (LIMIT-2). Lancet. 1992;339:1553-1558.[Medline] [Order article via Infotrieve]
  34. Resnick LM. Cellular calcium and magnesium metabolism in the pathophysiology and treatment of hypertension and related metabolic disorders. Am J Med. 1992;93(suppl 2A):2a-20s.
  35. Resnick LM, Gupta RK, Bhargava KK, Gruenspan H, Alderman MH, Laragh JH. Cellular ions in hypertension, diabetes, and obesity: a nuclear magnetic resonance spectroscopic study. Hypertension. 1991;17:951-957.[Abstract/Free Full Text]
  36. Resnick L, Laragh J, Sealey J, Alderman M. Divalent cations in essential hypertension: relations between serum ionized calcium, magnesium, and plasma renin activity. N Engl J Med. 1983;309:888-891.[Abstract]
  37. Rayssiguier Y. Magnesium and lipids in cardiovascular disease. J Am Coll Nutr. 1986;5:507-519.[Medline] [Order article via Infotrieve]
  38. Rasmussen HS, Aurup P, Goldstein K, McNair P, Mortensen PB, Larsen OG, Lawaetz H. Influence of magnesium substitution therapy on blood lipid composition in patients with ischemic heart disease. Arch Intern Med. 1989;149:1050-1053.[Abstract]
  39. Altura BT, Brust M, Sherman B, Barbour RL, Stempak JG, Altura M. Magnesium dietary intake modulates blood lipid levels and atherogenesis. Proc Natl Acad Sci U S A. 1990;87:1840-1844.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J EndocrinolHome page
P Delva, M Degan, M Trettene, and A Lechi
Insulin and glucose mediate opposite intracellular ionized magnesium variations in human lymphocytes.
J. Endocrinol., September 1, 2006; 190(3): 711 - 718.
[Abstract] [Full Text] [PDF]

Home page
 HypertensionHome page
P. Delva, C. Pastori, M. Degan, G. Montesi, P. Brazzarola, and A. Lechi
Intralymphocyte Free Magnesium in Patients With Primary Aldosteronism : Aldosterone and Lymphocyte Magnesium Homeostasis
Hypertension, January 1, 2000; 35(1): 113 - 117.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Delva, P. T.
Right arrow Articles by Lechi, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Delva, P. T.
Right arrow Articles by Lechi, A.

Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search
Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited.