Effect of chronic uremia in the rat on cerebral mitochondrial calcium concentrations

Effect of chronic uremia in the rat on cerebral mitochondrial calcium concentrations

Kidney International, Vol. 27 (1985), pp. 523—529 Effect of chronic uremia in the rat on cerebral mitochondrial calcium concentrations ANDREW J. ADLE...

649KB Sizes 3 Downloads 64 Views

Kidney International, Vol. 27 (1985), pp. 523—529

Effect of chronic uremia in the rat on cerebral mitochondrial calcium concentrations ANDREW J. ADLER and GEOFFREY M. BERLYNE Department of Medicine, Brooklyn Veterans Administration Hospital, Brooklyn, New York, USA

Effect of chronic uremia in the rat on cerebral mitochondrial calcium

concentrations. Whole cerebral and isolated mitochondrial calcium levels were determined in normal and chronically uremic SpragueDawley rats. Uremia was induced by a two-stage 5/6 nephrectomy 4 weeks prior to study. Serum was obtained for urea, calcium, magnesium, phosphate, and i-PTH. Mitochondria were isolated by gradient centrifugation and calcium was determined by flameless atomic absorp-

tion spectrophotometry. The results demonstrate that mitochondrial 2.8 calcium levels in uremic rats are not different from normal (8.0 1.8 nmoles/mg protein) despite an 11% increase in whole vs. 7.8 2.8 nmoles/mg 2.0 vs. 15.5 cerebral calcium concentration (17.3

protein; P <

0.005)

in 24 severely uremic rats (BUN >

18.0

mmoles/liter). Multiple regression analysis demonstrates a significant positive correlation between cerebral calcium concentrations and both serum calcium (P < 0.005) and serum magnesium levels (P < 0.005). No relationship was found for urea, serum phosphate, or i-PTH. Similar analysis of mitochondrial calcium concentration demonstrated a significant positive correlation with serum calcium (P < 0.005) and i-PTH

(P <

0.05)

suggesting that increased PTH may be necessary for

maintaining normal intracellular calcium levels in uremia. We conclude that uremia in the rat is associated with a small rise in whole cerebral

calcium but that intracellular calcium as reflected by mitochondrial levels is not elevated. Effet de l'urémie chronique chez le rat sur les concentrations de calcium mitochondrial cérébrales. Les niveaux de calcium cérébral total et dans des mitochondries isolées ont éte déterminés chez des rats Sprague-

Dawley normaux et en urémie chronique. L'urémie a été induite par une néphrectomie des 5/6 en deux étapes 4 semaines avant l'étude. Du serum a tC obtenu pour l'urée, Ic calcium, le magnesium, les phosphates, et l'i-PTH. Les mitochondries ont été isolées par gradient de centrifugation et le calcium a été déterminé par spectrophotométrie

d'absorption atomique de flamme. Les résultats montrent que les niveaux de calcium mitochondrial chez les rats urémiques ne different pas de la normale (8,0 1,8 nmoles/mg protéines) 2,8 contre 7,8 malgré une augmentation de 11% de Ia concentration calcique cérébrale 2,8 nmoles/mg protéines; P < 0,05) totale (17,3 2,0 contre 15,5 chez 24 rats sévèrement urémiques (BUN > 18,0 mmoles/litre). Une

analyse en regressions multiples démontre une correlation positive significative entre les concentrations calciques cérébrales et les niveaux de calcémie (P < 0,005) et de magnesémie (P < 0,005). Aucune relation

n'a été trouvée pour l'urée, Ia phosphatemie ou l'iPTH. Une analyse identique de la concentration calcique mitochondriale a démontré une correlation positive significative avec Ia calcémie (P < 0,005) et l'iPTH (P < 0,05), suggérant qu'une PTH élevée pourrait Ctre nécessaire pour

maintenir des niveaux de calcium intracellulaires normaux dans l'urémie. Nous concluons que l'urémie chez Ic rat est associée a une faible élévation du calcium cérébral total mais que le calcium intracellulaire, reflèté par les niveaux mitochondriaux, n'est pas élevé.

considerable interest in the role of calcium and parathyroid hormone (PTH) in the pathogenesis of uremic encephalopathy [5—8]. In 1974, Arieff and Massry [5] first suggested that acute uremia is associated with an increased calcium concentration in both grey and white matter of the brain and that these elevations could be prevented by prior parathyroidectomy. This was followed by reports in both uremic dogs [6] and humans [7, 8] of

an association between abnormalities in the electroencephalogram and the calcium content of the brain. These studies suggested that neurotoxicity in acute uremia is a consequence of accumulated calcium by the brain and that this was secondary to elevated serum i-PTH. However, in a number of preliminary studies in our laboratory using a rat model, an increase of brain calcium in either acute or chronic renal failure could not be demonstrated [9—11].

A basic question relevant to all these studies is whether analysis of a whole tissue preparation is indicative of its intracellular calcium concentration. There is ample evidence to indicate that from 50 to 90% of tissue calcium may be bound extracellularly to a glycoprotein calyx [12—141. Consequently,

changes in tissue calcium as reported for the brain in uremia may merely reflect changes in the quantity of calcium bound to extracellular ligands and not true changes in the physiologically important intracellular fraction.

To better assess the effect of uremia on intracellular brain calcium, we chose to study the calcium content of isolated mitochondria in addition to whole brain tissue. The mitochondna, which contain the largest proportion of the intracellular calcium [12, 15, 161, are thought to regulate cytoplasmic calcium and buffer excess calcium entering the cell [17—201. Thus,

any long-term alterations in intracellular calcium therefore might be expected to be reflected by changes in the mitochondrial calcium content. Methods

Animals. Adult male Sprague-Dawley rats (Charles River Laboratories, Wilmington, Massachusetts, USA) weighing 250 to 400 g at the time of sacrifice were used in all experiments. Experimental groups consisted of normal and chronically uremic rats.

Calcium ions participate in numerous cellular and subcellular processes and are particularly important for the normal function of the nervous system [1—4]. Reports of an increased calcium

Received for publication February 14, 1984, and in revised form June 5, 1984

concentration in the brain associated with uremia have led to

© 1985 by the International Society of Nephrology

523

524

Adler and Berlyne

Table 1. Ramped time temperature program used to heat cuvette in calcium analysis by atomic absorption spectrophotometry Step

Temperature

1

80

2 3 4

110

5

6

900 1600 2800 20

Ramp

sec

Hold sec

10 10 10 10

30 30 30 30

0 10

8

20

tion spectrophotometry (IL model #251, Worthington, Mas-

sachusetts, USA) using air/acetylene flame. Ten percent lanthanum by volume was added to the specimen to minimize interference by phosphates. Blood urea nitrogen levels were determined by standard autoanalyzer techniques (Technicon AAII, Tarrytown, New York, USA). Inorganic phosphate was determined by the colormetric method of Fiske and SubbaRow [211. Protein determinations were performed according to the method of Lowry et al [22], using bovine serum albumin as standard. Determination of i-PTH. This was kindly performed by Dr.

R. Shainken-Kestenbaum of Siroka Medical Center, Beer Induction of chronic uremia. Chronic uremia was induced following ether anesthesia by means of a two-stage procedure consisting of a right total nephrectomy followed in 2 weeks by a left 2/3 bipolar nephrectomy. Animals were maintained for 4

Sheva, Israel, by radioimmunoassay using the IRE Kit (6220 Fleurus Belgium) antibody to C-terminal fragments 34-84 of PTH. Results are expressed as milliunits per milliliters. One

milligram is equivalent to 2700 450 IU for the hormone MRC to 6 weeks prior to sacrifice on a standard rat chow diet (Ralston 71/324. Preparation of cerebral homogenates. Both normal and Purina, St. Louis, Missouri, USA) and tap water ad libitum, Calcium analysis by atomic absorption spectrophotometry. uremic rats were fasted overnight but allowed water ad libitum. Atomic absorption spectrophotometry with graphite furnace Rats were then decapitated and the brain was carefully removed flameless atomization was used to analyze all tissue specimens from the calvarium and rinsed with cold (4°C) deionized water. for calcium content. An atomic absorption spectrophotometer The cerebellum and the medulla were dissected free of the

(model #703, Perkin Elmer, Norwalk, Connecticut, USA) brain; the remaining cerebrum was weighed on an analytical equipped with an HGA 500 graphite furnace (Perkin Elmer) was balance (Mettler Model H54 AR, Princeton, New Jersey, used. Twenty microliters of sample were pipetted into an USA). The brain was then homogenized with 8 ml of deionized uncoated graphite cuvette of the atomizer. The cuvette was water in a homogenizer using eight up and down strokes with a heated by the ramped time temperature program in Table 1. Step numbers 5 and 6 in Table 1 were each repeated twice. This program achieved drying, ashing, and atomization of the

sample. The final four steps were used as burn-offs of the cuvette to eliminate any potential contamination of the subsequent sample. The absorbance was recorded at 422.7 nm and no scale expansion was used. The peak height mode of analysis

was used as a measure of the calcium content in the sample. Each sample was analyzed in triplicate. Water and standard solutions were repeated in triplicate after each sample. Linear regression analysis was used to obtain a sensitivity in units of absorbance/nanogram calcium atomized, and absorbances were linear over the concentration ranges studied (r = 0.998). The

pestle (Teflon®). All work was performed at 4°C. Preparation of brain mitochondria. Brain mitochondria were isolated according to a modification of the method of Clark and Nicklas [231, All preparative procedures were performed at 4°C. Rat brains were obtained and processed through the weighing procedure as described above. The cerebrum was then homogenized with 4 ml of cold isolation medium consisting of 0.25 M sucrose, 10 mM Tris-HC1 pH 7.4, and 0.05 mM EDTA, using eight up and down strokes with a pestle (Teflon®). An additional 4 ml of isolation medium were added to the homogenate which was then centrifuged (IEC Model B-20A, Damon IEC Division Needham Heights, Massachusetts, USA) at x900g for 10 mm at 4°C. The supernatant was again centrifuged at x 12,500g for 8

reproducibility and sensitivity of the method was assessed repeatedly throughout the course of the study. The percent recovery of added standard was 98.4%. The intragroup coefficient of variation was 1.05% for standard and 1.64% for tissue samples. The intergroup coefficient of variation was

mm yielding a crude mitochondrial pellet. The pellet was

1.84% for standards and 2.43% for a sample specimen carried

EDTA, and 10 mtvs Tris-HC1, pH 7.4 and centrifuged at x8000g

resuspended in 6 ml of a 4% Ficoll solution containing 0.12 M mannitol, 0.03 M sucrose, 0.025 EDTA, and 5 mi Tris-HC1, pH

7.4. This was carefully layered over 12 ml of an 8% Ficoll solution composed of 0.24 M mannitol, 0.06 M sucrose, 0.05

throughout the course of the study. Special procedures to for 30 mm at 4°C. The supernatant was removed including the reduce potential calcium contamination were meticulously followed. The water used for all preparations and washing was ultrapure reagent grade water prepared by means of a milli reagent grade water system (R. 0. and Milli-Q, Worthington Diagnostics, Millipore, Bedford, Massachusetts, USA). The calcium concentration of this water was consistently below 1

small buffy layer above the mitochondrial pellet. The pellet was then resuspended in 4 ml of isolation medium and recentrifuged

at x 12,500g for 8 mm. The pellet containing the final mitochondrial fraction was then resuspended in trace-element, quality pure deionized water. Contamination by residual isolation medium retained by the mitochondrial pellet was deterng/ml. All glass vessels used for analytical solutions or samples mined by means of '4C sucrose (New England Nuclear Corp., were washed by rinsing with ultrapure water, followed by Boston, Massachusetts, USA) labelling of the isolation medium analytical grade HCI 10% (Ultrex, J. T. Baker Chemical Co., and found to be negligible at 0.4% of the recovered calcium. Phillipsburg, New Jersey, USA) and rinsing again with The use of Ca2 chelator in the isolation medium prevents any ultrapure deionized water. All commercially purchased re- calcium uptake during the preparative procedure, and although we cannot be absolutely certain, there should be no measurable agents were of analytical grade when available. Biochemical determinations in serum. Serum calcium and calcium effiux from mitochondria during the isolation procemagnesium concentrations were measured by atomic absorp- dure, as shown by Hughes and Barritt [24] for liver.

525

Brain mitochondrial calcium in uremia Table 2.

Serum composition in normal and uremic ratsa Concentration in mmoles/liter

UreaN

Ca

P

Mg

Normal

5.94

1.56 (53)

2.55

0.20 (54)

3.19

0.26 (27)

1.18

0.22 (53)

Chronic uremia Mild Severe

18.9 14.0

6.5 (52)" 2.4 (25)" 5.6 (27)"

2.55 2.55 2.58

0.20 (50) 0.20 (27) 0.20 (23)

3.19 3.03

0.48 (26) 0.29 (13)

1.36 1.27

3.39

0.55 (l3)l

1.46

0.23 (50)" 0.23 (27)b 0.19 (23)"

23.4

a The number in parentheses indicates the number of animals. b P < 0.05 compared to normal.

P < 0.0005 compared to normal.

Enzyme determinations. Lactate dehydrogenase and cytochrome C oxidase activities were used to assess the purity of the mitochondrial pellet. Assays were performed at 25°C with a double-beam grating ultraviolet-light spectrophotometer and a

recorder (models 124 and 56, respectively, Perkin Elmer). Lactate dehydrogenase activity in whole brain and mitochondna was determined by following NADH oxidation at 340 nm in

a medium containing 50 m potassium phosphate, 1 mM pyruvate, 0.2 m NADH, and 0.5% Triton XlOO at pH 7.4.

Table 3. Cerebral and mitochondrial calcium concentrations in normal and uremic ratsa

Cerebral Ca nmoles/mg protein

Mitochondrial Ca nmoleslmg protein

Normal

15.5

2.8 (45)

8.0

2.8 (26)

Uremic Mild Severe

16.5 15.8 17.3

2.3 (45) 2.3 (21) 2.0 (24)"

7.8 7.8

1.8 (38) 1.8 (21) 1.8 (17)

7.8

Cytochrome C oxidase activity was determined spectrophotoa The number in parentheses indicates the number of animals. b P < 0.005 compared to normal animals. metrically in a medium containing 50 m phosphate buffer, pH 7.4, 50 m sucrose, 0.2 mi EDTA, and 0.025 m'vi cytochrome C which had been fully reduced with Na2S2O4. Statistical analysis. All data are presented as mean SD and Cerebral protein concentrations were 76.8 10.8 mg/g wet were analyzed for skewness and kurtosis by means of the model weight in normal rats and 75.5 7.0 mg/g wet weight in uremic II Statistical Analysis Program of the TSR 80 Microcomputer rats (P = NS). Specific activities of cytochrome C oxidase and (Tandy Corporation, Fort Worth, Texas, USA), Statistical lactate dehydrogenase in cerebral homogenates and mitochonanalysis included the Student t test for the difference of means drial preparations were measured. The purified mitochondrial and F ratio for significance of correlation of multiple regression fraction had a negligible content of LDH 0.046 tmoles/min/mg analysis. Correlation analysis was performed for the relation- protein and an enhanced cytochrome oxidase activity 0.85

ship of four independent variables to cerebral calcium and

tmoles/min/mg protein. The enzymatic purity of our mito-

mitochondrial calcium levels. Because i-PTH levels were available only in relatively small but randomly selected subgroups of the larger study group, multiple linear regression analysis was repeated for all four variables evaluated in the larger group plus i-PTH. The significance of semipartial and partial coefficients was determined by F ratio.

chondrial preparation agrees favorably with that reported in the literature [15, 23]. The enzyme activities in cerebral homogenates and purified brain mitochondria were not different between uremic animals and normal controls. Cerebral calcium levels are shown in Table 3. Mean cerebral calcium in 45 normal rats was 15.5 2.8 nmoles/mg protein and did not differ significantly from the mean concentration of 16.5

Results Serum values for normal and uremic rats are shown in Table

2. The results for uremic rats have been subdivided into mild uremia (BUN concentrations below 18.0 mmoles/liter) and severe uremia (BUN concentrations of 18.0 mmoles/liter or greater) according to the median BUN value for the entire group. Chronically uremic rats had a mean BUN concentration of 18.9 6.5 mmoles/liter (52.9 18.1 mg/dl) as compared to 5.94 1.46 mmoles/liter (16.6 4.1 mg!dl) in normal rats (P < 0.003). The mean serum calcium and phosphorous levels did not significantly differ between the normal and uremic groups taken

as a whole. However, in the severely uremic subgroup, with a mean BUN concentration of 23.4 5.6 mmoles/liter (65.6 15.8 mgldl), the serum phosphorous was significantly elevated above controls (3.19 0.26 vs. 3.39 0.55 mmoles/liter; P < 0.05). Serum magnesium concentrations were significantly higher in the uremic rats at 3.26 0.56 mg/dl compared to 2.83 0.53 mg/dl in normals (P < 0.0005).

2.3 nmoles/mg protein in 45 uremic rats. Similarly, brain mitochondrial calcium concentrations did not differ between groups, with a mean of 8.0 2.8 nmoles/mg protein in 26 normal rats and 7.8 1.8 nmoles/mg protein in 38 uremic rats. However, when the uremic subgroups are individually compared to the normal controls, the severely uremic subgroup demonstrates a significantly higher total cerebral calcium concentration of 17.3 2.0 nmoles/mg protein (P < 0.005). A similar result was not observed for mitochondrial calcium which remained identical in both uremic subgroups. The results of multiple regression analysis for both cerebral calcium and mitochondrial calcium are as follows: Cerebral calcium levels correlated positively with serum calcium [partial regression coefficient (pr) = 0.33; P < 0.005; Fig. 1] and serum

magnesium concentration (pr = 0.34, P < 0.005; Fig. 2). Mitochondrial calcium levels correlated only with serum calcium levels (pr = 0.45; P < 0.005; Fig. 3). Neither serum

526

Adler and Berlyne 22.5



0

.

20.0-

00

0

o000 0000 0 0

0

175,0 E

0

0

00

.

o—a, 0 0 0 0 0 a, 0 0 0 0 0 0 0

0

'5

i2.5) .0 5) 5)

io.o

0 I

0

0

00 a,



0

0 0 o

N85

r = 0.32

I

I

I

rats. The correlation coefficient and probability shown on the graph are

derived from univariate analysis of the two variables; the partial correlation coefficient and its probability derived from multiple regression analysis are given in Results.

0

0 a,

0

00

a, 00

000

- 0 00 0

10.0

00

00

-

15.0 -

i'

0' 0

0

0

12.5

-

0

0

0 00

0

00

00

0

0

00000 a,Oa, I

2.25

0

0

0

I

I

2.5

I

I

2.75

I 3.0

I

I

3.25

Serum Ca, mmoles/Iiter

Fig. 3. Relationship between mitochondrial Ca (nmoles/mg protein) and serum Ca (mmoles/liter) in normal and uremic rats. The correlation coefficient and probability shown on the graph are derived from

univariate analysis of the two variables; the partial correlation coefficient and its probability derived from multiple regression analysis are given in Results. Table 4. Serum composition and brain calcium levels in the subgroup of animals for which i-PTH levels were obtaineda

-

20.0-0

.0

a, 5.0-0

P<0.0005

3.0

Serum Ca, mmoles/Iiter

175 E

7.5

2.5

I—

I

I

2.75

2.5

Fig. 1. Relationship between cerebral Ca concentration (mmoles/mg Pr) and serum Ca concentration (mmoles/liter) in normal and uremic

22.5

' .. c

P < 0.005

2.25

-

E

0

0

00 00

12.5, 0 10.0

0

r = 0.41

0

E

000

0 00

15.0

0 N=63

15.0

c

00 00 0

0

0

0

00 0 0

0

00

00

a,

0

0 0

00

0

0

0

N=85

r=0.37 P < 0.0005

I I 7.5 _______________________________________________ 1 1.25 1.5 1.75 2.0

Serum Mg, mmoles/iiter Fig. 2. Relationship between cerebral Ca (mmoles/mg protein) and the serum Mg (mmoles/liter) in normal and uremic rats. The correlation coefficient and probability shown on the graph are derived from

univariate analysis of the two variables; the partial correlation coefficient and its probability derived from multiple regression analysis are given in Results.

Normal (12)

Uremic (16)

Serum urea N, mmoles/liter

5.57

1.56

Serum Ca, mmoles/liter

2.53

0.13

2.60

0.20

Serum P, mmoles/liter

3.14

0.15

3.11

0.23

Serum Mg, mmoles/liter

1.13

0.18

1.39

0.l6c

Serum i-PTH, mU/mi

4.8

2.3

8.2

34C

14.3

2.0

15.8

2.0

7.8

1.8

8.3

1.8

17.6

4.4"

Cerebral calcium, nmoles/mg

protein Mitochondrial calcium, nmoles/mg protein a

The number in parentheses indicates the number of animals. b P < 0.05. P < 0.0005.

analysis in this subgroup were identical to those of the larger population with the additional observation that i-PTH correlated positively with mitochondrial calcium levels (pr = 0.34; P < 0.05; Fig. 4). A similar relationship was not observed for cerebral calcium. Discussion

phosphate nor BUN demonstrated any significant relationship to cerebral or mitochondrial calcium.

The results of this study suggest that progressive renal failure

In a subgroup of 28 animals, serum i-PTH levels were

in the rat is associated with a rise in the calcium content of

obtained (Table 4). Mean i-PTH was 4.8 2.3 mU/mi in 12 normal rats and 8.2 3.4 mU/ml in 16 uremic rats (P < 0.05). Consistent with the results obtained in the larger population, neither cerebral calcium nor mitochondrial calcium differed between normal and uremic rats. The uremic group was not further subdivided into mild and severe uremia because of the relatively small sample size. Results of multiple regression

cerebral homogenates. This finding differs from our preliminary

observations [9—111, and that of Smythe et al [251, but is consistent with previous reports in other species [5—81. The mean increase in cerebral calcium was 11% above normal and was demonstrable only in the severely uremic subgroup. This

figure is considerably lower than either the 50% increase reported by Arieff and Massry [5] in acutely uremic dogs or the

Brain mitochondrial calcium in uremia q)

0

equilibration of extracellular calcium with tissue calcium and Harris, Carness, and Forte [311 found that despite a sevenfold

12.5

increase in i-PTH, brain calcium was decreased in their

10.0 0

8

-

0

_______ó_ 5.0 2.5

527

0

0

0

0

r0.52

0

-

P < 0.025

0

I

I

0

2

4

6

8

10

12

14

16

I 18

PTH, mU/mi

hypocalcemic rats. The finding of a significant positive correlation between brain calcium and serum magnesium was somewhat surprising and remains difficult to interpret. One possibility is that in the rat serum, magnesium levels are better indicators of the severity of uremia than the BUN, and its relationship to cerebral calcium merely reflects the effect of uremia. A direct effect of extracellular magnesium could be postulated, but there is no definitive

evidence to support such a hypothesis. In in vivo studies, Blaustein, Ratzlaff, and Kendrick [341 showed that ATP-

depended calcium uptake by rat brain synaptosomes requires extracellular magnesium; on the other hand, Kamino et al [35] demonstrated inhibition by magnesium of non-ATP-dependent calcium binding by rat synaptosomes. It should be noted that in both these studies, the extrasynaptosomal calcium concentration was well below normal physiological levels. In general, our results of cerebral calcium concentrations, as 52 and 73% increases reported respectively by Cogan et a! [8] well as those of other investigators must be interpreted with and Cooper, Lazarowitz, and Arieff [71 in uremic humans. A great caution, since it is very dubious whether the calcium recent report by Mahoney and Arieff [26] suggests that calcium content of any tissue homogenate actually indicates the level of levels are elevated only in cortical gray matter and hypo- intracellular calcium. A number of studies in a variety of tissues thalamus, which possibly might account for our lower value. have shown that 50 to 90% of total tissue calcium may be bound When the data were further evaluated by multiple regression to extracellular ligands [12—14], making it quite feasible that any analysis, it was found that the BUN concentration contributed alteration in the measured calcium content of tissue merely negligibly to the level of cerebral calcium. Only the serum reflects an alteration in the calcium bound extracellularly. calcium and serum magnesium concentrations demonstrated a Within the cell, however, specific subcellular compartments significant relationship to cerebral calcium levels. A relation- contain distinct quantities of calcium. Of these, the mitoship between the extracellular calcium concentration and tissue chondrial pooi contains the largest proportion of the intracellucalcium levels has been shown in vitro for a variety of tissues lar calcium ranging between 30 and 65% in most tissues [12, 15, Fig. 4. Relationship between mitochondrial Ca (nmoleslliter) and serum

i-PTH (mU/mI) in a subgroup of normal and uremic rats. The correlation coefficient and probability shown on the graph are derived from univariate analysis of the two variables, partial correlation coefficient and its probability are derived from multiple regression analysis and are given in Results.

including kidney, liver, muscle, and brain [27—311. In vivo, the results have been somewhat less consistent but generally support the in vitro observations. Wallach et al [321, studying tissue

distribution of the 47Ca in acutely hypercalcemic dogs, con-

cluded that the cellular calcium of brain was largely

16]. Furthermore, there is very compelling evidence that in cells other than erythrocytes and striated muscle, the mitochondria are the primary regulators of the cytosolic calcium activity [17—20]. Mitochondria isolated from most tissue are capable of reducing the external calcium concentration to below l06 M and both the rate and capacity of calcium sequestration exceed those of the plasma membrane and endoplasmic reticulum by at

unexchangeable and was not altered by changes in extracellular calcium. However, these animals were studied over a relatively short time period of 4 hr. In studies carried out over least one order of magnitude [12, 17, 201. Calcium accumulation a longer time frame, Mulryan et al [331 showed that the brain by mitochondria is thought to be a component of the mechanism equilibrates very slowly with extracellular calcium requiring up which keeps cytosolic calcium at io that of the extracellular to 4 days for complete equilibration. Harris, Carness, and Forte space, and thereby buffers any excess calcium which may enter [31] also found a relationship between extracellular calcium and the cell. This has been shown for liver, kidney, muscle, and brain calcium and reported decreases of brain calcium levels in brain [36—391. Other calcium sequestering systems have been rats rendered chronically hypocalcemic by diets deficient in described for brain [40] and are important in short-term regulation of intracellular calcium as may occur following an action either calcium or vitamin D. The failure of other investigators to demonstrate a similar potential. Mitochondria, however, have a larger overall capacrelationship in uremia may be due to a variety of factors ity and are thought to be particularly important for the brain in including species differences, the small sample size used in situations of excessive calcium loads [39, 41]. Consequently, those studies, and the relatively short time interval between the analysis of mitochondrial calcium should more closely parallel induction of uremia and sacrifice of the animal. As our animals the intracellular calcium content than whole tissue homogenates. Our results demonstrate that mitochondrial calcium remains were in a steady-state for at least 4 weeks prior to sacrifice, it can be expected that any long-term changes in the extracellular unchanged despite the presence of significant renal failure. Multiple regression analysis indicates that the mitochondrial calcium level would also be reflected in the brain tissue. Our results also suggest that there is no direct relationship calcium concentration correlates directly with serum calcium between cerebral calcium and serum i-PTH, a finding in contra- levels and surprisingly with i-PTH, but not with the severity of distinction to that reported by Arieff and Massry [51 but renal failure as reflected by either the BUN or serum magneconsistent with other observations made in nonuremic condi- sium. This would suggest that although brain homogenates have tions. Mulryan et al [33] reported that PTH had no effect on the an increased calcium content in severely uremic rats, intracel-

528

Adler and Berlyne

lular calcium as reflected by mitochondrial calcium is not elevated. The role of PTH is not clear, but it is intriguing to speculate that elevated PTH is necessary for maintaining a

8. COGAN MG, COVEY CM, ARIEFF Al, WI5NIEwsKI A, CLARK OH:

normal intracellular calcium concentration in a setting where

levels in brain and skin of chronic uremic rats. Contrib Nephrol

Central nervous system manifestations of hyperparathyroidism. Am J Med 65:963—970, 1978 9. ADLER AJ, LUNDIN AP, BERLYNE GM: Calcium and magnesium

there could otherwise develop a reduction in intracellular levels. If normal intracellular calcium levels are in part dependent on the extracellular calcium concentration, and if uremia is associated with an increase in calcium precipitation or binding to extracellular ligands, it is quite reasonable to conclude that

increased PTH counteracts a trend toward a reduction in exchangeable extracellular calcium.

The mechanism by which PTH accomplishes this for the brain must as yet remain speculative although several mechanisms are possible. The simplest would be that the increased levels of PTH maintain the extracellular calcium at normal levels thereby indirectly maintaining normal intracellular levels. Another possibility is that PTH directly affects calcium trans-

port into the brain. In tissues for which PTH receptors are recognized PTH does result in a net increase in cellular calcium in vitro [12]. This, however, is a less likely explanation as it has

not to date been reported that the brain is a target organ for PTH. Of course, it is quite possible that the failure to demonstrate increased mitochondrial calcium in renal failure may be due to intrinsic abnormalities of mitochondrial calcium cycling and an impairment of their calcium buffering ability induced by uremia. Although this problem was not addressed directly by the study, the fact that a similar relationship between serum calcium and mitochondrial calcium was demonstrated in both normal and uremic groups would suggest that calcium cycling is not grossly impaired.

20:67—72, 1980

10. ADLER AJ, BERLYNE GM:

Effect of acute and chronic uremia on brain mitochondnal calcium and magnesium in the rat (abstract). Am Fed Clin Res 27:36lA, 1979

11. ADLER AJ, LUNDIN AP, BERLYNE GM: Divalent patient concen-

tration in tissues of rats with chronic renal failure (Abstract). Am Fed Clin Res 27:361A, 1979 12. BORLE AB: Control, modulation and regulation of cell calcium. Rev Physiol Biochem Pharmacol 90:13—153, 1981 13. MORIARTY CM: Involvement of intracellular calcium in hormone secretion from rat pituitary cells. Mol Cell Endocrinol 6:349—361, 1977 14. VAN BREEMAN C, MCNAUGHTON E: The separation of cell mem-

brane calcium transport from extracellular calcium exchange in

vascular smooth muscle. Biochem Biophys Res Commun 39:567—574, 1970 15. LAZAREWICZ JW, HALJAMAE H, HAMBERGER A: Calcium metabo-

lism in isolated brain cells and subcellular fractions. J Neurochem 22:33—45, 1974

16. Ba.NcHI CP: Cell Calcium. London,

Butterworths, 1968

BORLE AB: Calcium metabolism at the cellular level. Fed Proc 32:1944—1950, 1973 18. CAItFoLI E, MALMSTRoM K, CAPANO M, SIGEL E, CROMPTON M: 17.

Mitochondria and the regulation of cell calcium, in Calcium Transport in Contraction and Secretion, edited by CARAFOLI E, Amsterdam, North-Holland Publishing Co., 1975, pp 53—64

19. CARAF0LI E, CROMPTON M: The regulation of intracellular by mitochondna. Ann NYAcad Sci 307:269—284, 1978

20. BYGRAvE FL: Mitochondria and the control of intracellular calcium. Biol Rev 53:43—79, 1978 21. FIsKE C, SUBBAROW Y: The colorimetric determination of phosphorus. J Biol Chem 66:375—400, 1925 22. LOWRY ON, ROSENBROUGH NJ, FARR AL, RANDALL RJ: Protein

measurement with the folin phenol reagent. J

Acknowledgments This work was supported by a Merit Review Grant from the Veterans Administration, USA. The authors acknowledge C. Caruso for technical advice, R. Montana, G. Reilly, R. Coismain, M. Bean, and R. Cerny for technical assistance, and G. Halpert and A. Sardisco for typing the manuscript.

Reprint requests to Dr. A. J. Adler (111), Department of Medicine, Brooklyn Veterans Administration Hospital, 800 Poly Place, Brooklyn, New York 11209, USA

References 1.

BAKER PF, REUTER H: Transport and metabolism of calcium in

nerve, in Calcium Movement in Excitable Cells. Oxford, Pergamon, 1975, pp 7—53 2. RAHAMINOFF R: Role of calcium ions in neuromuscular transmission, in Calcium and Cellular Function, edited by CUTHBERT AW, London, Macmillan, 1970, pp 131—147 3. McGit.w CF, NACHMEN DA, BLAUSTEIN MP: Calcium movement and regulation in presynaptic nerve terminals, in Calcium and Cell Function, edited by CHEUNG WY, New York, Academic Press, 4.

1982, pp 81—110 RASMUSSEN H: Cell communication, calcium ion and cyclic adenosine monophosphate. Science 170:404—412, 1970

5. ARIEFF Al, MASSRY SG: Calcium metabolism of brain in acute renal failure. J Clin Invest 53:387—392, 1974

6. GUISADO R, ARIEFF Al, MASSRY SG: Changes in the electroencephalogram in acute uremia. J Clin Invest 55:738—745, 1975

7. COOPER JD, LAZAROWITZ VC, ARIEFF Al: Neurodiagnostic ab-

normalities in patients with acute renal failure. J Clin Invest 61:1448—1455, 1978

calcium

Biol Chem

1951 23. CLARK JB, NIcKLA5 193:265—275,

Wi: The metabolism of rat brain mitochondria. J Biol Chem 245:4724—4731, 1970 24. HUGHES BP, BARRITT GJ: Effects of glucagon and N602—

Dibutyryladenosine 3' :5'—Cyclic monophosphate on calcium transport in isolated rat liver mitochondria. Biochem J

176:295—304,

1978

25. SMYTHE WR, ALFREY AC, CRASWELL PW, CROUCH CA, IBELS LS, KUB0 H, NUNNELLEY LL, RUDOLPH H: Trace element abnormalities in chronic uremia. Ann mt Med 96:302—310, 1982 26. MAHONEY CA, ARIEFF Al: Central and peripheral nervous system effects of chronic renal failure. Kidney Int 24:170—177, 1983 27. BORLE AB: Kinetic analysis of calcium movement in cell cultures. V. Intracellular calcium distribution in kidney cells. J. Membr Biol 10:45—66, 1972 28. WALLACH 5, REIZENSTEIN DL, BELLAVIA JV: The cellular trans-

port of calcium in rat liver. J

Gen Physiol 49:743—762, 1966

29. GILBERT DL, FENN WD: Calcium equilibrium in muscle. J Gen Physiol

40:393—408, 1957

30. COOKE WJ, RoBINsoN JD: Factors influencing Ca45 metabolism in brain and other organs in vivo. Proc Soc Exp Biol Med 138:906—912, 1971

31. HARRIS RA, CARNESS DL, FORTE LR: Reduction of brain calcium after consumption of diets deficient in calcium or vitamin D. J Neurochem 36:460—466, 1981 32. WALLACH 5, BELLAVIA JV, SCHORR J, REIZENSTEIN DL: distribution of electrolytes, Ca47 and Mg28

Tissue

in acute hypercalcemia.

Am J Physiol 207:553—560, 1964 33. MULRYAN BJ, NEUMAN MW, NEUMAN WF, TORIBARA TY: Equili-

bration between tissue calcium and injected radiocalcium in the rat.

Am J Physiol 207:947—952, 1964 34. BLAUSTEIN JP, RATZLAFF RW, KENDRICK NK: The intracellular

regulation of

calcium in presynaptic nerve terminals. Am NYAcad

Sd 307:195—212, 1978

Brain mitochondrial calcium in uremia 35. KAMINO K, UYESAKA N, OGAWA M, INouYE A: Calcium-binding

of synaptosomes isolated from rat brain cortex. J Membr Biol 21:113—124, 1975

36. NICHOLS DG: The regulation to extramitochondrial-free calcium

ion concentration by rat liver mitochondria. Biochem J 176:463—474, 1978

37. BORLE AB: Cyclic AMP stimulation of calcium efflux from kidney, liver and heart mitochondria. J Membr Biol 16:221—236, 1974

38. NicHoLs DG, SCOTT JD: The regulation of brain mitochondrial calcium-ion transport. Biochem J 186:833—839, 1980

529

39. SCOTT ID, NICHOLS DG: Energy transduction in intact synaptosomes. Influence of plasma-membrane depolarization on the respiration and membrane potential of internal mitochondria determined in situ. Biochem J 186:21—33, 1980 40. BLAUSTEIN MP, RATZLAFF RW, KENDRICK NC, SCHWEITZER EJ:

Calcium buffering in presynaptic nerve terminals. I. Evidence for involvement of a non-mitochondnal ATP-dependent sequestration mechanism. J Gen Physiol 72:15-41, 1978 41. CHAN SY, OCHS 5, JERSILD RA JR: Localization of calcium in nerve fibers. J Neurobiol 15:89—108, 1984