CELL BIOCHEMISTRY AND FUNCTION
VOL.
8: 31-38 (1990)
Studies of the Regulation of Renal Gluconeogenesis in Normal and Pi Depleted Proximal Tubule Cells TAKESHI NAKAGAWA AND PETER J. BUTTERWORTH Department of Biochemistry, King's College (London), Campden Hill Road, London W8 7 A H , U.K.
Pi depletion of proximal tubule cells isolated from mouse kidnej results in a decrease in the cell content of fructose-2,6-bisphosphate and an increase in the rate of gluconeogenesis from pyruvate, malate and succinate. Gluconeogenesis from glycerol is unaffected by Pi depletion. Introduction of fructose-2,6-bisphosphate into the cytosol of ATP-permeabilized cells is accompanied by a fall in gluconeogenesis. The presence of external Ca2+ stimulates gluconeogenesis. When cytosolic Ca2+ is raised to 1.8 p~ by permeabilization, the resealed cells still require 2.5 mM Ca2+ in the bathing medium in order to perform gluconeogenesis at the maximum rate. Cells permeabilized in the presence of CAMP show a decreased rate of glucose production. Phorbol ester stimulates gluconeogenesis provided that the phorbol treatment is performed in the absence of Ca2+ ions. It is suggested that Pi depletion may stimulate pyruvate carboxylase activity and facilitate the entry of certain gluconeogenic substrates into mitochondria. It is also proposed that important aspects of the control of renal gluconeogenesis by parathyroid hormone are mediated by protein kinase C. KEY WORDS
-Proximal tubule; P,depletion; calcium; phorbol ester; gluconeogenesis
INTRODUCTION
A distinctive feature of the proximal tubule is that it is a gluconeogenic tissue and the gluconeogenic activity is stimulated by parathyroid and by Ca2 ions. The hormone is an important regulator of Ca2 and Pi homeostasis. It promotes Pi excretion by inhibiting the activity of a brush border membrane Na+-Picotransporter and raises plasma C a 2 + by promoting mobilization of bone mineral and by stimulating tubular reabsorption of the cation. The Na+-dependent Pi carrier is involved in Pi reabsorption and hence in maintaining plasma Pi levels within normal physiological +
+
limit^.^.^ When an animal is reared on a Pi-deficient diet, the amount of the anion that is excreted is considerably diminished and this reduction in urinary Pi is associated with stimulated activity of the Na+-Pi c ~ t r a n s p o r t e r .Studies ~ with isolated cells6 and with renal epithelial cells in culture7 have shown that tubular cells themselves are able to sense the ambient Pi concentration and adjust Pi transport activity accordingly. The intracellular signal generated by Pi depletion that results in raised transport capacity is unknown, but the key importance of Pi 0263-6484/90/010031-08$05.00
01990 by John Wiley & Sons, Ltd.
in intermediary metabolism suggests that an altered pattern of metabolism could be an important factor influencing the activity of the transporter. The idea that the effect of parathyroid hormone on Pi transport is mediated by a stimulation of gluconeogenesis is not new' but the original suggestion that the linking factor is the NAD+/NADH ratio' does not seem to be true.4 Nevertheless, it is conceivable that Pi depletion and parathyroid hormone cause changes in the level of a key metabolite or metabolites that subsequently affect the transporter. Because the hormone and Pi depletion affect Pi transport in opposite directions, it is to be expected that a putative regulatory metabolite would change in opposite directions in response to the hormone and to depletion. It is of interest in this connection that tubular reabsorption of Pi is resistant to the inhibitory action of parathyroid hormone under conditions of phosphate depletion. The second messenger in parathyroid hormone actions affecting Pi transport and gluconeogenesis may be 3',5'cyclic AMP, C a 2 + ions, inositol polyphosphates or a combination of two or more of these agents?
32 Few studies of the effect of Pi status on the metabolic activity of the proximal tubule seem to have been undertaken but renal cortical tubules isolated from phosphate depleted animals were reported to exhibit diminished basal gluc~neogenesis.~ As a preliminary step therefore, in identifying the factor that signals an increased transport capacity in response to Pi depletion, we are studying gluconeogenesis and its regulation in normal and Pi-depleted isolated proximal tubule cells. MATERIALS AND METHODS
T. NAKAGAWA AND P. J. BUTTERWORTH
used for spectrophotometric assay of fructose-2,6-bisphosphate by the method of Van Schaftingen" modified slightly to obtain the maximal response. In the modified system, the concentration of glucose-6-phosphate was 17 mM, and 9.4 units 1-' of phosphofructokinase, 0.84 units 1-' of aldolase and 1.6 K units 1-' of triosephosphate isomerase/glycerol-3-phosphate dehydrogenase were used. Each sample was assayed five times. For each assay, a set of standards covering the range 0-100 nmol of bisphosphate in the 1 ml reaction mixture was measured in parallel with the cell extracts.
isolation of Proximal Tubule Cells
Permeabilization
Kidneys from two or three adult mice were removed immediately after killing and minced finely with scissors. The minced kidneys were then digested by the method recently described" in collagenase-hyaluronidase at 37°C for 30 min in freshly prepared Krebs-Ringer-bicarbonate medium (pH 7-4) in which the respiratory substrate glucose was replaced by 1 mM pyruvate (normal medium). In some experiments pyruvate was replaced by one of succinate, malate or glycerol (see below). During the digestion, disruption of tissue aggregates was facilitated at 5-min intervals by suction into, and expulsion from, a plastic pipette having an orifice of 1 mm diameter. One kidney from each of 2-3 mice was digested in normal Pi replete medium (Pi concn., 1.18mM) and the other kidneys from the same mice were depleted of Pi by digestion and subsequent washing in Pi-free medium."
Following isolation of the cells in normal medium (containing 1.25 mM Ca"), the cells were washed in cytosol-like buffer, pH 7.4, that contained 103.4mM KCl, 21.1 mM NaCl, 1 mM NaH,PO,, and 26.4 mM NaHCO, (buffer C) then suspended in buffer C containing 1 mg ml-' of bovine serum albumin. The suspension contained approximately 0.2 mg tubule cell protein ml- '. ATP4- was added to a concentration of 5 p to~ permeabilize the cells' and the mixture allowed to stand for 3 min with occasional stirring. This method of reversibly increasing the permeability to water and to ions is known to be effective for proximal tubules. The permeabilization was terminated by addition of Mg2+to a concentration of 1.25 mM followed by three washes in buffer C adjusted to contain 1.25 mM Mg2+.In some experiments, agents such as C a 2 + or fructose-2,6bisphosphate were added 2 min after the addition of ATP. The details of such experiments are given in the Results. The effectiveness of permeabilization was assessed by monitoring by microscopy, the fluorescence of cells induced by ATP in the presence of ethidium bromide.
Incubation Media
In the various experiments the cells were incubated at 37°C in Krebs-Ringer bicarbonate medium with gentle stirring and continuous gassing with O,/CO, (19:l). The concentration of Pi and Ca 2 + in the medium were adjusted to the values needed for each experiment. Assay of Fructose-2,6-Bisphosphate
Tubule cell preparations from the normal and the depleted cells were each suspended in 1 ml of distilled water and then treated wit 0.1 ml 10 per cent (w/v) sodium deoxycholate dissolved in 50 mM NaOH. The cell debris was separated by microcentrifugation and 0.2 ml portions of the supernatant
''.
Gluconeogenesis
Pi replete, depleted and/or resealed cells were incubated at 37°C in 8 ml of normal medium containing 1 mg ml-' bovine serum albumin and 10mM pyruvate in a capped vessel possessing an inlet and outlet for O,/CO,. Gassing continued for the whole of the incubation period. The suspension contained approximately 0.4.mg of cell protein ml-'. Samples of medium (0.8m1) were taken at intervals, spun in a microcentrifuge and the supernatant assayed for g l ~ c o s eand ' ~ for protein by the
33
CONTROL OF RENAL GLUCONEOGENESIS
and succinate (Figure 1). The observed stimulations were 16, 15 and 22 per cent for pyruvate, malate and succinate respectively ( P < 0.01 for each substrate in three separate experiments). The difference in the degree of stimulation seen with the three substrates was not significant ( P < 0.25). The stimulation is consistent with the observed decrease in fructose-2,6-bisphosphate assuming that the Treatment with Phorbol level of the sugar phosphate is important for reguTubule cells isolated in medium containing. lating the relative rates of glycolysis and gluconeo1-25mM C a 2 + were washed in Ca2+-freemedium genesis in the proximal tubule as well as in liver." and resuspended in 10ml of this medium to give The effect of depletion is complex however in that approximately 3-4 mg protein ml-'. The C a 2 + gluconeogenesis from glycerol, a gluconeogeneic concentration was adjusted to 2-5 mM by appropri- substrate for renal tissue, particularly in . ' ~not affected by depletion. It is of ate addition of cation or to zero by the addition of ~ t a r v a t i o n ' ~ is 1 mM EGTA. One minute later, phorbol-12,13- interest that the rate of gluconeogenesis was undibutyrate (PDB) or the inactive 4cr-phorbol ester affected by the ommission of Pi from the bathing was added to a concentration of 4 nmol mg-' medium, even in cells that had been depleted of Pi protein. After 5 min of treatment, the cells were (Figure 1). washed three times in the Ca2+-free medium to Exposure of the tubule cells for up to 5 min to remove EGTA and phorbol ester. The washed cells ATP before the addition of Mg2+ions seems to were finally suspended to a level of 1.5 mg cell damage them in that the rate of gluconeogenesis is protein ml-' in medium supplemented with bovine decreased by approximately 20 per cent. This deserum albumin (1 mg ml-' plus 10 mM pyruvate and which contained either no added Ca2+ or 2.5 mM C a 2 + .GLucose production was then monitored for up to 60 min. T 1 A similar protocol was followed using phorbol myristate acetate (PMA) except that the ester concentration was 1-4pmol mg-' cell protein and the c cells were exposed to phorbol for 3 min only. After washing (see above) to remove PMA and EGTA the cells were resuspended in medium containing 0-lmM Ca2+.
Biuret assay. For some experiments pyruvate was replaced, in each case at 10 mM concentration, by succinate or malate of glycerol. For such experiments the cells were isolated in normal medium in which the respiratory substrate was 1 mM malate, succinate or glycerol respectively.
RESULTS The fructose-2,6-bisphosphate content of tubule cells was variable from preparation to preparation, ranging from 0.596 to 2-489 pmol mg-' protein (mean of 1.944 with S.E. of 0.464). When the bisphosphate content of cells prepared in normal medium was compared with that of Pi-depleted cells prepared from the other kidneys of the same group of animals it was found that depletion results in a fall in fructose-2,6-bisphosphate content. The range for depleted cells was 0.426 to 2-263 pmol mg-' protein (mean of 1-07with S.E. of 0.313. The difference in content between Pi replete and depleted cells is significant ( P < 0.01 by paired t-test, n = 5). Phosphate depletion results in a significant stimulation of gluconeogenesis from pyruvate, malate
1
L
t -
a c .a
.-
.-
Experiment I Substrate Pyr
2 Mal
L
3 Sucr
4 Glycerol
Figure 1. Effect of Pi depletion on gluconeogenesis.For experiments 1-4, glucose production was monitored in normal medium containing 1.8 mM Pi and the indicated substrate at 10 mM concentration. In experiment 5, glucose production was measured in Pi-free medium containing 10 mM pyruvate. In 1-3 the difference between the depleted and the replete control is significant ( P < 0.01 in each case). Experiment 5 presents the results of a single but representative experiment.
34
T. NAKAGAWA AND P. J. BUTTERWORTH
crease in gluconeogenic capacity is probably a reflection of the proportion of cells that cannot reseal after permeabilization. Prolonging the ATP treatment beyond 5 min before washing of the cells in buffer C containing Mg2+ to chelate free ATP, results in further impairment of gluconeogenesis in a time-dependent manner. I t was found that ATP treatment for 5 min in the presence of ethidium bromide results in strong fluorescence of the nuclei in over 95 per cent of the cells. A period of 3 min exposure to ATP was adopted routinely for the studies described below. Figure 2 shows the effect of permeabilization by ATP4- and fructose-2,6-bisphosphate treatment on the rate of glucose production. The figure shows that permeabilization in the presence of fructose-2,6-bisphosphate brings about a significant inhibition of gluconeogenesis. In preliminary studies it became clear that small molecular weight metabolites and cofactors e.g. ATP, ADP and NAD did not need to be added to the permeabilization buffer to make good any essential factors that could have been washed out during permeabilization. Presumably sufficient ADP/AMP is liberated from the added ATP by endogenous enzymes during treatment and sufficient NAD is available from the considerable amount known to bind to the brush border membrane of proximal tubule
cell^.^.'^
c
0
25
ICa2+l mM c .-
I 50
i
E 0 \o
c m
-
/
0
+ L f 0 U
w0
$ f 0
.-
+ v 3 n
50
12-5
75
1
i
I
0 L
n W
W
-4-
wl 0
0 L
U
m
U
n
3 d
E \
f 2
0
aJ
wl 0
30
60
Minutes
u
3
m
0
E c
0
15
30
45
60
Minutes Figure 2. Effect of fructose-2,6-bisphosphate on gluconeogenesis by Pi replete cells. Glucose production was monitored in untreated cells, 0 , normal medium containing 2.5 mM Ca2+.0, cells that had been permeabilized by exposure to ATP for 3 min; A, cells exposed to ATP plus fructose-2,6-bisphosphate (lo-' M). OP < 0.025for ATP plus F2,6P2compared with ATP alone (n = 4).
Figure 3. Effect of extracellular C a 2 + on gluconeogenesis in non-permeabilized cells. In (a) the rate of glucose production is plotted against the concentration of Ca2+ in the bathing medium. The results are from a single but typical experiment with P, replete cells. In (b) each plot shows the mean and S.E. of three experiments performed with separate cell preparations. Because the basal rate of gluconeogenesis varies greatly between preparations, glucose production in this and some of the subsequent figures is expressed as a percentage of the amount produced at 60 min under control conditions. 0 ,Non-permeabilized cells incubated in medium with 2-5 mM C a 2 + (control); A,cells exposed to ATP plus 1.8 PM Caz+ prior to measurement of glucose production at 2.5 mM Ca2+;0, ATP-permeabilized cells incubated in Ca2'-free medium; A, cells exposed to ATP plus 1 . 8 ~Ca2+ ~ prior to incubation in Ca2+-free medium. 0,P < 0.01 relative to control value at 45 min.
35
CONTROL OF RENAL GLUCONEOGENESIS
,,,.'I
It has been known for many years that Ca2+ .-c 250 E stimulates the gluconeogenic activity of renal corti0 cal t i ~ s u e Figure .~ 3a shows that glucose produc- \o tion from pyruvate by the isolated mouse cells is c m greatly stimulated by Ca". The maximum rate of ,of,'' gluconeogenesis was observed at concentrations of 40 c Ca2+ in the bathing medium that greatly exceed 150 physiological levels. By exploiting permeabilization to increase cytosolic Ca2 transiently to approximately 2 p ~a ,stimulation of gluconeogenesis was c observed but to obtain a maximum response, an .-0 t U external supply of the cation was still necessary 3 U (Figure 3b). In contrast, raising cytosolic 3'5' cyclic L 0 AMP to 1-10 PM inhibited gluconeogenesis ( P < a aJ 0-0025)measured in the presence of 2 m Ca2+. ~ An vl 0 early report' showed that renal gluconeogenesis U 3 from certain substrates could be stimulated by 13 cAMP provided that Ca2+ ions were present but 0 30 60 the different Ca2+ requirements described for the various substrates used' shows the situation to be Mi nu tes very complex. When the phosphodiesterase inhibitor 3-isobutyl-1-methyl xanthine at 10 PM was used c .in conjunction with cAMP no significant inhibition E 0 occurred however. Thus the inhibition by cAMP -a c could arise in part from AMP liberated by hydrolynJ sis. What seems to be important in terms of regula0 c L tion of metabolism however, is that stimulation of c u 0 gluconeogenesis by cAMP was never observed in our experiments. s Normal cells were treated for 5 min with PDB in c 0 ._ the absence of added calcium but with 1 mM EGTA c u present to elminate free C a 2 + , then washed three 3 -u times in Ca+-free medium before suspension in P P replete medium. Such cells had a relatively low m basal rate of gluconeogenesis from pyruvate 0 u (40 nmol h- mg- protein) in the replete medium, 3 which contained 2.5 mM calcium, but the basal rate lrl 0 30 60 was stimulated by phorbol (Figure 4a). With Mi nut e s washed cells resuspended in medium containing 1 mM EGTA and no added Ca' i.e. free Ca2 was Figure 4. Effect of phorbol dibutyrate on gluconeogenesis. (a) close to zero, the basal rate of gluconeogenesis was Cells treated with phorbol ( 0 )and without phorbol ( i t . the (0)in CaZ'-free medium containing 1 mM EGTA. 0, very low, but those cells which had been treated control) P < 04X)l (n = 3) relative to control rate of 40 nrnol h - ' mg-' with phorbol ester still exhibited stimulated glu- at 45 min. (b) Cells treated with ( 0 )and without phorbol(0) in coneogenesis (not shown). Cells which had been the presence of 2.5 mM Ca2+.Control rate at 45 min is 120 nmol treated with PDB in the presence of 2.5 mM cal- h - ' mg-' protein. In both (a) and (b) gluconeogenesis was cium produced glucose from pyruvate at a rate that performed in medium containing 2.5 mM Ca2+. differed little from the non-treated controls. If anything, phorbol was slightly inhibitory when applied untreated control, n = 3) when gluconeogenesis in the presence of calcium (Figure 4b). was measured with 1 mM C a 2 + in the bathing The potent phorbol ester PMA activates glucon- medium. The sensitivity to Ca2 was also raised by eogenesis when used at very low concentrations i.e. PMA treatment; for Ca2+ in the bathing 1.4 pmol mg-' cell protein. The stimulation was medium was decreased to 0.1 mM. If PMA treatgreater than five-fold ( P < 0.0025 relative to the ment was conducted in the presence of C a 2 + how-
i
/
L
s
+
I
x'
W
'
+
+
+
36
T. NAKAGAWA AND P. J. BUTTERWORTH
ever, a pronounced inhibition of gluconeogenesis results reminiscent of the finding seen with PDB when used under similar conditions. The degree of inhibition was not increased by raising the concentration of PMA from 1.4 to 140 pmol mg-’ cell protein. The 4a-phorbol ester had no effect on gluconeogenesis irrespective of whether it was applied in the presence or absence of C az + ions. Studies of the concentration dependence of PDB stimulation of gluconeogenesis indicated that sigificant activation is produced by treatment with 40 pmol mg-’ protein (Figure 5). At a concentration of 4nmol mg-’ protein, PDB stimulated gluconeogenesis from malate three-fold but by only 3-7 per cent when glycerol was the glucose precursor, This small increase was not statistically significant. Gluconeogenesis in depleted cells was also stimulated by PDB to the same extent as seen with normal Pi-replete cells.
DISCUSSION An increase in cytosolic fructose-2,6-bisphosphate inhibits gluconeogenesis in proximal tubule cells. Thus the fall in the cellular level of the sugar phosphate that occurs in Pi depletion is consistent with the stimulation of gluconeogenesis seen under the conditions of depletion. The change in the rate of glucose production accompanying depletion was not observed however, when glycerol was used as the source of carbon for glucose synthesis. In liver, a decrease in cytosolic Pi is immediately followed by a decrease in mitochondrial Pi.18 This may represent part of a mechanism by which cytosolic Pi is kept within narrow limits. If the proximal tubule behaves in a similar manner, it is possible that the efflux of Pi from the mitochondria that would be induced when cells are suspended in Pi-free medium, lowers mitochondrial Pi levels and thereby affects the activity of pyruvate carboxylase and/or pyruvate dehydrogenase. Any such action on the carboxylase is unlikely to affect gluconeogenesis from glycerol and should, contrary to our findings, have little influence on the rate of production of glucose from succinate or malate. Although Pi inhibition of rat liver pyruvate carboxylase has been described, the K i value is relatively highZo and therefore a physiological controlling role for Pi has been questioned.” Values for the concentration of rat liver mitochondria1 Pi given in the literature cover a broad range. Heldt et ~ 1 C I t quoted 2.5 mM but other authors have detected 10 nmol mg-’ mitochondrial protein in mitochondria severely depleted of Piz’ which translates to 8-5mM Pi assuming a mitochondrial water space of 1.2 p1 mg-’”. On the other hand a concentration of approx 40mM can be calculated from data by Stermann and Decker’ assuming a mitochondrial water space for liver of 48 pl-I wet wtZ3and that all the anion is in an unbound state. Although much of mitochondrial Pi is likely to be bound, it is perhaps reasonable to assume that free Pi is close to 10-15 m M (see reference 24 for a review of this subject) in Pi replete mitochondria in which case sigificant inhibition of pyruvate carboxylase may 40 4000 occur. The lack of a requirement of an external 0 4 source of Pi during the gluconeogenesis experiments can be interpreted perhaps as indirect evipmol PDB / mg protein dence for mobilization of mitochondrial Pi. Figure 5. Dose-dependence of the activation of renal gluconSince Pi can act as an exchange anion in the eogenesis by PDB. Isolated tubule cells were suspended in Ca- movement of anionic substrates across the mitofree medium containing 1 mM EGTA for 1 min and then treated chondrial membrane, an efflux of Pi following a for 5 min with PDB at the indicated concentrations. The cells were then washed to remove the EGTA and gluconeogenesis depletion episode would tend to favour uptake of pyruvate and succinate into the mitochondria. The monitored for 60 min in medium containing 2 mM Ca2*.
I
.
~
~
37
CONTROL OF RENAL GLUCONEOGENESIS
uptake would be expected to favour the use of these substrates as glucose precursors. The failure of Pi depletion to stimulate gluconeogenesis from glycerol could then be explained because this substrate is converted to glucose in steps that are all extramitochondrial. The stimulation of glucose production from malate by depletion is difficult to explain however on this basis. If phosphoenolpyruvate carboxykinase is cytosolic in the mouse (as it is in the rat), malate conversion to glucose will also be extramitochondrial since oxaloacetate will result from the action of cytosolic malate dehydrogenase. Because cytosolic Pi appears to be buffered against wide fluctuations in the Pi content of the surrounding m e d i ~ m ~it, is ’ ~difficult to explain the change in fructose-2,6-bisphosphate content accompanying depletion by an assumption of an allosteric action of Pi on phosphofructokinase 2 and fructose bisphosphatase 2. Our results differ from those obtained with tubules isolated from animals reared on a phosphate deficient diet.’ The tubules seemed to have impaired ATP production.’ An explanation for the difference in the results is that ‘acute’ depletion produced by isolation of cells in Pi-free conditions, and thus exposed to low Pi for approximately 45 to 60 min, has less effect on cellular ATP levels than does the rearing of animals for several weeks on a Pi-free diet. A significant impairment of ATP synthesis brought about by the lengthy exposure to low Pi would be expected to interfere with gluconeogenic activity and perhaps mask changes in glucose production arising from effects of altered Pi on the activity of certain enzymes. Inhibition by phorbol esters of gluconeogenesis in canine proximal tubule segments has been described previouslyz6 and the conclusion reached that protein kinase C can regulate renal gluconeogenesis by raising cytosolic pH through stimulation of N a + / H exchange. The conditions under which the inhibition was observedz6 correspond to those in which either no effect or inhibition was seen in our experiments i.e. the phorbol treatment was applied in the presence of C a 2 + ions. The stimulatory action of phorbol that we observe when the agent is applied in the absence of C a 2 + may be connected with the state of association of protein kinase C with the plasma membrane.27*28Binding of phorbol to the kinase may increase its sensitivity to Ca2 ions. When gluconeogenesis is conducted in the normal medium containing physiological concentrations of Ca2 the increased sensitivity is reflected in the stimulation of gluconeogenic rate. +
+
+
The extracellular source of calcium is probably needed for a localized pool of the cation in the plasma membrane from which Ca2+ that activates kinase C is drawn.” The putative target protein(s) for kinase C, phosphorylation of which affects gluconeogenesis, cannot be identified from these studies except that it must be concerned with the part of the pathway to the triose phosphate stage since glucose production from glycerol is unaffected by phorbol. Since stimulation is observed when depleted cells are treated with phorbol it is likely that Pi depletion and protetin kinase C do not influence identical steps in the gluconeogenesis pathway. Phorbol treatment, especially with the potent PMA, in the presence of calcium may stimulate protein kinase C to such an extent that the subsequent response is that of negative feedback” and any stimulatory action becomes masked. Alternatively, superoxide generation under these conditions may cause general damage to the Parathyroid hormone has been shown to promote diacylglyceride formation in kidney proximal tubules30 and stimulation of renal gluconeogenesis by parathyroid hormone required CaZ+.31On the basis therefore that phorbol esters mimic diacylglycerol in stimulating protein kinase C , one can speculate that the control of gluconeogenesis by parathyroid hormone is mediated by this kinase. This conclusion is supported by the finding that the hormone induces CaZ transients in cultered proximal tubule cells.32Protein kinase A (CAMPdependent) is probably less important in the parathyroid hormone regulation of gluconeogenesis. +
REFERENCES 1. Nagata, N. and Rasmussen, H. (1970). Parathyroid hormone, 3‘,5‘cAMP, Ca2+ and renal gluconeogenesis. Proc. Narl. Acad. Sci. (U.S.A.),65, 368-374. 2. Roobol, A. and Alleyne, G. A. D. (1973). Regulation of renal gluconeogenesis by calcium ions, hormones and adenosine 3‘3’ cyclic monophosphate. Biochem. J . , 134, 157-165. 3. Gmaj, P. and Murer, H. (1986). Cellular mechanism of inorganic phosphate transport in kidney. PhysioL. Rev., 66, 36-70. 4. Butterworth, P. J. (1987). Phosphate horneostastis. Mol. Aspects Med., 9, 289-386. 5. Massry, S. G. and Fleisch, H. (1980). Renal Handling of Phosphate. Plenum Medical Book Co.: New York. 6. Grahn, M. F. and Butterworth, P. J. (1986). Effects of acute phosphate depletion in isolated chick kidney tubule cells. Cell Biochem. Funct., 4, 271-275. 7. Caverzasio, J., Brown, C . D. A., Biber, J., Bonjour, J.-P. and Murcr, H. (1985). Adaptation of phosphate transport in
38
8.
9.
10.
11.
12. 13 14
15 16
17
18.
19.
20.
T. NAKAGAWA AND P. J. BUTTERWORTH
phosphate deprived LLC-PKI cells. Am. J . Physiol., 248, F122-F127. Kempson, S. A., Colon-Otero, G.,Lise-Ou, S. Y., Turner, S. T. and Dousa, T. P. (1981). Possible role of nicotinamide adenine dinucleotide as an intracellular regulator of renal transport of phosphate in the rat. J . Clin. Invest., 67, 1347-1 360. Kreusser, W. J., Descoeurdres, C. Oda, Y.,Massry, S. G. and Kurokawa, K. (1980). Effect of phosphate depletion on renal gluconeogenesis. Mineral Electrolyt. Metab., 3, 312-322. Jahan, M. and Butterworth, P. J. (1988). Study of the mechanism by which the Na+-P, cotransporter of mouse proximal tubule cells adjusts to phosphate depletion. Biochem. J . 252, 105-109. Van Schaftingen, E. (1984). ~-Fructose-2,6-bisphosphate. In: Methods ofEnzymatic Analysis, Vol. 6. (Bermeyer, H. U. ed.) 3rd ed. Verlag Chemie: Weinheim, GFR, pp. 335-341. Gomperts, B. D. (1983): Involvement of guanine nucleotide binding protein in the gating of Ca2+ by receptors. Nature (Lond), 306,64-66. Rorive, G. and Kleinzeller, A. (1972). The effect of ATP and C aZ +on the cell volume in isolated kidney tubules. Biochim. Biophys. Acta, 274,226-239. Bergmeyer, H. U., Berndt, E., Schmidt, F. and Stork, H. (1974). D-Glucose: Determination with hexokinase and glucose-6-phosphate dehydrogenase. In: Methods of Enzymatic Analysis, Vol. 3. (Bergmeyer, H. U. ed.) 2nd ed. Academic Press: New York, pp. 1196-1198. Hue, L. and Rider, M. H. (1987). Role of fructose-2,6bisphosphate in the control of glycolysis in mammalian tissues. Biochem. J., 245, 313-324. Wirthensohn, G. A., Vandewalle, A. and Guder, W. G. (1981). Renal glycerol metabolism and the distribution of glycerol kinase in rabbit nephron. Biochem, J., 198, 543-549. Guder, W. G. and Ruprecht, A. R. (1975). Metabolism of isolated kidney tubules. Independent actions of catecholamines on renal cyclic adenosine 3’3’ monophosphate levels and glyconeogenesis. Eur. J. Biochem., 52,283-290. Stermann, R. and Decker, K. (1978). Differential response of ATP and orthophosphate in cytosol and mitochondria of rat hepatocytes to treatment with Pi and galactosamine. Febs Lett., M, 214-216. Wu, K. I., Bacon, R. A., Al-Mahrouq, H. A. and Kempson, S. A. (1988). Nicotinamide as a rapid acting inhibitor of renal brush border phosphate transport. Am. J . Physiol., 255, F15-F21. McClure, W. R., Lardy, H. A., Wagner, M. and Cleland, W. W. (1971). Rat liver pyruvate carboxylase I1 Kinetic studies of the forward reaction. J. Biol. Chem., 246, 3579-3583.
21. McClure, W. R. and Lardy, H. A. (1971). Rat liver pyruvate carboxylase IV Factors affecting the regulation in viva J . Biol. Chem., 246, 3591-3596. 22. Heldt, H. W., Klingenberg, M. and Milovancev, M. (1972). Differences between the ATP-ADP ratios in the mitochondrial matrix and in the extramitochondrial space. Eur. J . Biochem., 30,434-440. 23. Roos, I., Crompton, M. and Carafoli, E. (1980). The role of inorganic phosphate in the release of Ca2+ from rat liver mitochondria. Eur. J . Biochem., 110, 319-325. 24. Sestoft, L. and Bartels, P. D. (1981). Regulation of metabolism by inorganic phosphate. In: Short Term Regulation of Metabolism. (Hue, L. and Van Der Werve, G . eds) ElsevierNorth Holland Biomedical Press: Amsterdam, New York and Oxford, pp. 427-452. 25. Bevington, A., Mundy, K. I., Yates, A. J. P., Kanis, J. A., Taylor, D. J., Rajagopalan, B., Russell, R. G. G. and Radda, G. K. (1986). A study of intracellular orthophosphate concentrations in human muscle and erythrocytes by 31P nuclear magnetic resonance spectroscopy and selective chemical assay. Clin. Sci., 72, 729-735. 26. Rogers, S., Gavin, J. R. and Hammerman, M. R. (1985). Phorbol esters inhibit gluconeogenesis in canine renal proximal tubular segments. Am. J . Physiol., 249, F256-F262. 21. Kishimoto, A., Takai, Y., Mori, T., Kikkawa, U. and Nishizuka, Y. (1980). Activation of calcium and phospholipid dependent protein kinase by diacyl glycerol; its possible relation to phosphatidylinositol turnover. J . Biol. Chem., 255, 2273-2276. 28. Niedel, J. E. and Blackshear, P. J. (1986). In: Phosphoinositides and Receptor Mechanisms. Alan R. Liss, Inc., pp. 47-88. 29. Nishizuka, Y. (1986). Studies and perspectives of protein kinase C. Science, 233, 305-312. 30. Hruska, K. A,, Goligorsky, M., Scoble, J., Tsutsumi, M., Westbrook, S. and Moskowitz, D. (1986). Effects of parathyroid hormone on cytosolic calcium in renal proximal tubular primary cultures. Am. J . Physiol., 251, F188-Fl98. 31. MacDonald, D. W. R. and Saggerson, E. D. (1977). Hormonal control of gluconeogenesis in tubule fragments from renal cortex of fed rats. Biochem. J., 168, 33-42. 32. Goligorsky, M. S., Osborne, D., Howard, T., Hruska, K. A. and Karl, I. E. (1987). Hormonal control of gluconeogenesis in cultured proximal tubular cells: role of cytosolic calcium. Am. J . Physiol., 253, F802-F809.
Received 17 July 1989 Accepted 29 September 1989
VOL.
8: 31-38 (1990)
Studies of the Regulation of Renal Gluconeogenesis in Normal and Pi Depleted Proximal Tubule Cells TAKESHI NAKAGAWA AND PETER J. BUTTERWORTH Department of Biochemistry, King's College (London), Campden Hill Road, London W8 7 A H , U.K.
Pi depletion of proximal tubule cells isolated from mouse kidnej results in a decrease in the cell content of fructose-2,6-bisphosphate and an increase in the rate of gluconeogenesis from pyruvate, malate and succinate. Gluconeogenesis from glycerol is unaffected by Pi depletion. Introduction of fructose-2,6-bisphosphate into the cytosol of ATP-permeabilized cells is accompanied by a fall in gluconeogenesis. The presence of external Ca2+ stimulates gluconeogenesis. When cytosolic Ca2+ is raised to 1.8 p~ by permeabilization, the resealed cells still require 2.5 mM Ca2+ in the bathing medium in order to perform gluconeogenesis at the maximum rate. Cells permeabilized in the presence of CAMP show a decreased rate of glucose production. Phorbol ester stimulates gluconeogenesis provided that the phorbol treatment is performed in the absence of Ca2+ ions. It is suggested that Pi depletion may stimulate pyruvate carboxylase activity and facilitate the entry of certain gluconeogenic substrates into mitochondria. It is also proposed that important aspects of the control of renal gluconeogenesis by parathyroid hormone are mediated by protein kinase C. KEY WORDS
-Proximal tubule; P,depletion; calcium; phorbol ester; gluconeogenesis
INTRODUCTION
A distinctive feature of the proximal tubule is that it is a gluconeogenic tissue and the gluconeogenic activity is stimulated by parathyroid and by Ca2 ions. The hormone is an important regulator of Ca2 and Pi homeostasis. It promotes Pi excretion by inhibiting the activity of a brush border membrane Na+-Picotransporter and raises plasma C a 2 + by promoting mobilization of bone mineral and by stimulating tubular reabsorption of the cation. The Na+-dependent Pi carrier is involved in Pi reabsorption and hence in maintaining plasma Pi levels within normal physiological +
+
limit^.^.^ When an animal is reared on a Pi-deficient diet, the amount of the anion that is excreted is considerably diminished and this reduction in urinary Pi is associated with stimulated activity of the Na+-Pi c ~ t r a n s p o r t e r .Studies ~ with isolated cells6 and with renal epithelial cells in culture7 have shown that tubular cells themselves are able to sense the ambient Pi concentration and adjust Pi transport activity accordingly. The intracellular signal generated by Pi depletion that results in raised transport capacity is unknown, but the key importance of Pi 0263-6484/90/010031-08$05.00
01990 by John Wiley & Sons, Ltd.
in intermediary metabolism suggests that an altered pattern of metabolism could be an important factor influencing the activity of the transporter. The idea that the effect of parathyroid hormone on Pi transport is mediated by a stimulation of gluconeogenesis is not new' but the original suggestion that the linking factor is the NAD+/NADH ratio' does not seem to be true.4 Nevertheless, it is conceivable that Pi depletion and parathyroid hormone cause changes in the level of a key metabolite or metabolites that subsequently affect the transporter. Because the hormone and Pi depletion affect Pi transport in opposite directions, it is to be expected that a putative regulatory metabolite would change in opposite directions in response to the hormone and to depletion. It is of interest in this connection that tubular reabsorption of Pi is resistant to the inhibitory action of parathyroid hormone under conditions of phosphate depletion. The second messenger in parathyroid hormone actions affecting Pi transport and gluconeogenesis may be 3',5'cyclic AMP, C a 2 + ions, inositol polyphosphates or a combination of two or more of these agents?
32 Few studies of the effect of Pi status on the metabolic activity of the proximal tubule seem to have been undertaken but renal cortical tubules isolated from phosphate depleted animals were reported to exhibit diminished basal gluc~neogenesis.~ As a preliminary step therefore, in identifying the factor that signals an increased transport capacity in response to Pi depletion, we are studying gluconeogenesis and its regulation in normal and Pi-depleted isolated proximal tubule cells. MATERIALS AND METHODS
T. NAKAGAWA AND P. J. BUTTERWORTH
used for spectrophotometric assay of fructose-2,6-bisphosphate by the method of Van Schaftingen" modified slightly to obtain the maximal response. In the modified system, the concentration of glucose-6-phosphate was 17 mM, and 9.4 units 1-' of phosphofructokinase, 0.84 units 1-' of aldolase and 1.6 K units 1-' of triosephosphate isomerase/glycerol-3-phosphate dehydrogenase were used. Each sample was assayed five times. For each assay, a set of standards covering the range 0-100 nmol of bisphosphate in the 1 ml reaction mixture was measured in parallel with the cell extracts.
isolation of Proximal Tubule Cells
Permeabilization
Kidneys from two or three adult mice were removed immediately after killing and minced finely with scissors. The minced kidneys were then digested by the method recently described" in collagenase-hyaluronidase at 37°C for 30 min in freshly prepared Krebs-Ringer-bicarbonate medium (pH 7-4) in which the respiratory substrate glucose was replaced by 1 mM pyruvate (normal medium). In some experiments pyruvate was replaced by one of succinate, malate or glycerol (see below). During the digestion, disruption of tissue aggregates was facilitated at 5-min intervals by suction into, and expulsion from, a plastic pipette having an orifice of 1 mm diameter. One kidney from each of 2-3 mice was digested in normal Pi replete medium (Pi concn., 1.18mM) and the other kidneys from the same mice were depleted of Pi by digestion and subsequent washing in Pi-free medium."
Following isolation of the cells in normal medium (containing 1.25 mM Ca"), the cells were washed in cytosol-like buffer, pH 7.4, that contained 103.4mM KCl, 21.1 mM NaCl, 1 mM NaH,PO,, and 26.4 mM NaHCO, (buffer C) then suspended in buffer C containing 1 mg ml-' of bovine serum albumin. The suspension contained approximately 0.2 mg tubule cell protein ml- '. ATP4- was added to a concentration of 5 p to~ permeabilize the cells' and the mixture allowed to stand for 3 min with occasional stirring. This method of reversibly increasing the permeability to water and to ions is known to be effective for proximal tubules. The permeabilization was terminated by addition of Mg2+to a concentration of 1.25 mM followed by three washes in buffer C adjusted to contain 1.25 mM Mg2+.In some experiments, agents such as C a 2 + or fructose-2,6bisphosphate were added 2 min after the addition of ATP. The details of such experiments are given in the Results. The effectiveness of permeabilization was assessed by monitoring by microscopy, the fluorescence of cells induced by ATP in the presence of ethidium bromide.
Incubation Media
In the various experiments the cells were incubated at 37°C in Krebs-Ringer bicarbonate medium with gentle stirring and continuous gassing with O,/CO, (19:l). The concentration of Pi and Ca 2 + in the medium were adjusted to the values needed for each experiment. Assay of Fructose-2,6-Bisphosphate
Tubule cell preparations from the normal and the depleted cells were each suspended in 1 ml of distilled water and then treated wit 0.1 ml 10 per cent (w/v) sodium deoxycholate dissolved in 50 mM NaOH. The cell debris was separated by microcentrifugation and 0.2 ml portions of the supernatant
''.
Gluconeogenesis
Pi replete, depleted and/or resealed cells were incubated at 37°C in 8 ml of normal medium containing 1 mg ml-' bovine serum albumin and 10mM pyruvate in a capped vessel possessing an inlet and outlet for O,/CO,. Gassing continued for the whole of the incubation period. The suspension contained approximately 0.4.mg of cell protein ml-'. Samples of medium (0.8m1) were taken at intervals, spun in a microcentrifuge and the supernatant assayed for g l ~ c o s eand ' ~ for protein by the
33
CONTROL OF RENAL GLUCONEOGENESIS
and succinate (Figure 1). The observed stimulations were 16, 15 and 22 per cent for pyruvate, malate and succinate respectively ( P < 0.01 for each substrate in three separate experiments). The difference in the degree of stimulation seen with the three substrates was not significant ( P < 0.25). The stimulation is consistent with the observed decrease in fructose-2,6-bisphosphate assuming that the Treatment with Phorbol level of the sugar phosphate is important for reguTubule cells isolated in medium containing. lating the relative rates of glycolysis and gluconeo1-25mM C a 2 + were washed in Ca2+-freemedium genesis in the proximal tubule as well as in liver." and resuspended in 10ml of this medium to give The effect of depletion is complex however in that approximately 3-4 mg protein ml-'. The C a 2 + gluconeogenesis from glycerol, a gluconeogeneic concentration was adjusted to 2-5 mM by appropri- substrate for renal tissue, particularly in . ' ~not affected by depletion. It is of ate addition of cation or to zero by the addition of ~ t a r v a t i o n ' ~ is 1 mM EGTA. One minute later, phorbol-12,13- interest that the rate of gluconeogenesis was undibutyrate (PDB) or the inactive 4cr-phorbol ester affected by the ommission of Pi from the bathing was added to a concentration of 4 nmol mg-' medium, even in cells that had been depleted of Pi protein. After 5 min of treatment, the cells were (Figure 1). washed three times in the Ca2+-free medium to Exposure of the tubule cells for up to 5 min to remove EGTA and phorbol ester. The washed cells ATP before the addition of Mg2+ions seems to were finally suspended to a level of 1.5 mg cell damage them in that the rate of gluconeogenesis is protein ml-' in medium supplemented with bovine decreased by approximately 20 per cent. This deserum albumin (1 mg ml-' plus 10 mM pyruvate and which contained either no added Ca2+ or 2.5 mM C a 2 + .GLucose production was then monitored for up to 60 min. T 1 A similar protocol was followed using phorbol myristate acetate (PMA) except that the ester concentration was 1-4pmol mg-' cell protein and the c cells were exposed to phorbol for 3 min only. After washing (see above) to remove PMA and EGTA the cells were resuspended in medium containing 0-lmM Ca2+.
Biuret assay. For some experiments pyruvate was replaced, in each case at 10 mM concentration, by succinate or malate of glycerol. For such experiments the cells were isolated in normal medium in which the respiratory substrate was 1 mM malate, succinate or glycerol respectively.
RESULTS The fructose-2,6-bisphosphate content of tubule cells was variable from preparation to preparation, ranging from 0.596 to 2-489 pmol mg-' protein (mean of 1.944 with S.E. of 0.464). When the bisphosphate content of cells prepared in normal medium was compared with that of Pi-depleted cells prepared from the other kidneys of the same group of animals it was found that depletion results in a fall in fructose-2,6-bisphosphate content. The range for depleted cells was 0.426 to 2-263 pmol mg-' protein (mean of 1-07with S.E. of 0.313. The difference in content between Pi replete and depleted cells is significant ( P < 0.01 by paired t-test, n = 5). Phosphate depletion results in a significant stimulation of gluconeogenesis from pyruvate, malate
1
L
t -
a c .a
.-
.-
Experiment I Substrate Pyr
2 Mal
L
3 Sucr
4 Glycerol
Figure 1. Effect of Pi depletion on gluconeogenesis.For experiments 1-4, glucose production was monitored in normal medium containing 1.8 mM Pi and the indicated substrate at 10 mM concentration. In experiment 5, glucose production was measured in Pi-free medium containing 10 mM pyruvate. In 1-3 the difference between the depleted and the replete control is significant ( P < 0.01 in each case). Experiment 5 presents the results of a single but representative experiment.
34
T. NAKAGAWA AND P. J. BUTTERWORTH
crease in gluconeogenic capacity is probably a reflection of the proportion of cells that cannot reseal after permeabilization. Prolonging the ATP treatment beyond 5 min before washing of the cells in buffer C containing Mg2+ to chelate free ATP, results in further impairment of gluconeogenesis in a time-dependent manner. I t was found that ATP treatment for 5 min in the presence of ethidium bromide results in strong fluorescence of the nuclei in over 95 per cent of the cells. A period of 3 min exposure to ATP was adopted routinely for the studies described below. Figure 2 shows the effect of permeabilization by ATP4- and fructose-2,6-bisphosphate treatment on the rate of glucose production. The figure shows that permeabilization in the presence of fructose-2,6-bisphosphate brings about a significant inhibition of gluconeogenesis. In preliminary studies it became clear that small molecular weight metabolites and cofactors e.g. ATP, ADP and NAD did not need to be added to the permeabilization buffer to make good any essential factors that could have been washed out during permeabilization. Presumably sufficient ADP/AMP is liberated from the added ATP by endogenous enzymes during treatment and sufficient NAD is available from the considerable amount known to bind to the brush border membrane of proximal tubule
cell^.^.'^
c
0
25
ICa2+l mM c .-
I 50
i
E 0 \o
c m
-
/
0
+ L f 0 U
w0
$ f 0
.-
+ v 3 n
50
12-5
75
1
i
I
0 L
n W
W
-4-
wl 0
0 L
U
m
U
n
3 d
E \
f 2
0
aJ
wl 0
30
60
Minutes
u
3
m
0
E c
0
15
30
45
60
Minutes Figure 2. Effect of fructose-2,6-bisphosphate on gluconeogenesis by Pi replete cells. Glucose production was monitored in untreated cells, 0 , normal medium containing 2.5 mM Ca2+.0, cells that had been permeabilized by exposure to ATP for 3 min; A, cells exposed to ATP plus fructose-2,6-bisphosphate (lo-' M). OP < 0.025for ATP plus F2,6P2compared with ATP alone (n = 4).
Figure 3. Effect of extracellular C a 2 + on gluconeogenesis in non-permeabilized cells. In (a) the rate of glucose production is plotted against the concentration of Ca2+ in the bathing medium. The results are from a single but typical experiment with P, replete cells. In (b) each plot shows the mean and S.E. of three experiments performed with separate cell preparations. Because the basal rate of gluconeogenesis varies greatly between preparations, glucose production in this and some of the subsequent figures is expressed as a percentage of the amount produced at 60 min under control conditions. 0 ,Non-permeabilized cells incubated in medium with 2-5 mM C a 2 + (control); A,cells exposed to ATP plus 1.8 PM Caz+ prior to measurement of glucose production at 2.5 mM Ca2+;0, ATP-permeabilized cells incubated in Ca2'-free medium; A, cells exposed to ATP plus 1 . 8 ~Ca2+ ~ prior to incubation in Ca2+-free medium. 0,P < 0.01 relative to control value at 45 min.
35
CONTROL OF RENAL GLUCONEOGENESIS
,,,.'I
It has been known for many years that Ca2+ .-c 250 E stimulates the gluconeogenic activity of renal corti0 cal t i ~ s u e Figure .~ 3a shows that glucose produc- \o tion from pyruvate by the isolated mouse cells is c m greatly stimulated by Ca". The maximum rate of ,of,'' gluconeogenesis was observed at concentrations of 40 c Ca2+ in the bathing medium that greatly exceed 150 physiological levels. By exploiting permeabilization to increase cytosolic Ca2 transiently to approximately 2 p ~a ,stimulation of gluconeogenesis was c observed but to obtain a maximum response, an .-0 t U external supply of the cation was still necessary 3 U (Figure 3b). In contrast, raising cytosolic 3'5' cyclic L 0 AMP to 1-10 PM inhibited gluconeogenesis ( P < a aJ 0-0025)measured in the presence of 2 m Ca2+. ~ An vl 0 early report' showed that renal gluconeogenesis U 3 from certain substrates could be stimulated by 13 cAMP provided that Ca2+ ions were present but 0 30 60 the different Ca2+ requirements described for the various substrates used' shows the situation to be Mi nu tes very complex. When the phosphodiesterase inhibitor 3-isobutyl-1-methyl xanthine at 10 PM was used c .in conjunction with cAMP no significant inhibition E 0 occurred however. Thus the inhibition by cAMP -a c could arise in part from AMP liberated by hydrolynJ sis. What seems to be important in terms of regula0 c L tion of metabolism however, is that stimulation of c u 0 gluconeogenesis by cAMP was never observed in our experiments. s Normal cells were treated for 5 min with PDB in c 0 ._ the absence of added calcium but with 1 mM EGTA c u present to elminate free C a 2 + , then washed three 3 -u times in Ca+-free medium before suspension in P P replete medium. Such cells had a relatively low m basal rate of gluconeogenesis from pyruvate 0 u (40 nmol h- mg- protein) in the replete medium, 3 which contained 2.5 mM calcium, but the basal rate lrl 0 30 60 was stimulated by phorbol (Figure 4a). With Mi nut e s washed cells resuspended in medium containing 1 mM EGTA and no added Ca' i.e. free Ca2 was Figure 4. Effect of phorbol dibutyrate on gluconeogenesis. (a) close to zero, the basal rate of gluconeogenesis was Cells treated with phorbol ( 0 )and without phorbol ( i t . the (0)in CaZ'-free medium containing 1 mM EGTA. 0, very low, but those cells which had been treated control) P < 04X)l (n = 3) relative to control rate of 40 nrnol h - ' mg-' with phorbol ester still exhibited stimulated glu- at 45 min. (b) Cells treated with ( 0 )and without phorbol(0) in coneogenesis (not shown). Cells which had been the presence of 2.5 mM Ca2+.Control rate at 45 min is 120 nmol treated with PDB in the presence of 2.5 mM cal- h - ' mg-' protein. In both (a) and (b) gluconeogenesis was cium produced glucose from pyruvate at a rate that performed in medium containing 2.5 mM Ca2+. differed little from the non-treated controls. If anything, phorbol was slightly inhibitory when applied untreated control, n = 3) when gluconeogenesis in the presence of calcium (Figure 4b). was measured with 1 mM C a 2 + in the bathing The potent phorbol ester PMA activates glucon- medium. The sensitivity to Ca2 was also raised by eogenesis when used at very low concentrations i.e. PMA treatment; for Ca2+ in the bathing 1.4 pmol mg-' cell protein. The stimulation was medium was decreased to 0.1 mM. If PMA treatgreater than five-fold ( P < 0.0025 relative to the ment was conducted in the presence of C a 2 + how-
i
/
L
s
+
I
x'
W
'
+
+
+
36
T. NAKAGAWA AND P. J. BUTTERWORTH
ever, a pronounced inhibition of gluconeogenesis results reminiscent of the finding seen with PDB when used under similar conditions. The degree of inhibition was not increased by raising the concentration of PMA from 1.4 to 140 pmol mg-’ cell protein. The 4a-phorbol ester had no effect on gluconeogenesis irrespective of whether it was applied in the presence or absence of C az + ions. Studies of the concentration dependence of PDB stimulation of gluconeogenesis indicated that sigificant activation is produced by treatment with 40 pmol mg-’ protein (Figure 5). At a concentration of 4nmol mg-’ protein, PDB stimulated gluconeogenesis from malate three-fold but by only 3-7 per cent when glycerol was the glucose precursor, This small increase was not statistically significant. Gluconeogenesis in depleted cells was also stimulated by PDB to the same extent as seen with normal Pi-replete cells.
DISCUSSION An increase in cytosolic fructose-2,6-bisphosphate inhibits gluconeogenesis in proximal tubule cells. Thus the fall in the cellular level of the sugar phosphate that occurs in Pi depletion is consistent with the stimulation of gluconeogenesis seen under the conditions of depletion. The change in the rate of glucose production accompanying depletion was not observed however, when glycerol was used as the source of carbon for glucose synthesis. In liver, a decrease in cytosolic Pi is immediately followed by a decrease in mitochondrial Pi.18 This may represent part of a mechanism by which cytosolic Pi is kept within narrow limits. If the proximal tubule behaves in a similar manner, it is possible that the efflux of Pi from the mitochondria that would be induced when cells are suspended in Pi-free medium, lowers mitochondrial Pi levels and thereby affects the activity of pyruvate carboxylase and/or pyruvate dehydrogenase. Any such action on the carboxylase is unlikely to affect gluconeogenesis from glycerol and should, contrary to our findings, have little influence on the rate of production of glucose from succinate or malate. Although Pi inhibition of rat liver pyruvate carboxylase has been described, the K i value is relatively highZo and therefore a physiological controlling role for Pi has been questioned.” Values for the concentration of rat liver mitochondria1 Pi given in the literature cover a broad range. Heldt et ~ 1 C I t quoted 2.5 mM but other authors have detected 10 nmol mg-’ mitochondrial protein in mitochondria severely depleted of Piz’ which translates to 8-5mM Pi assuming a mitochondrial water space of 1.2 p1 mg-’”. On the other hand a concentration of approx 40mM can be calculated from data by Stermann and Decker’ assuming a mitochondrial water space for liver of 48 pl-I wet wtZ3and that all the anion is in an unbound state. Although much of mitochondrial Pi is likely to be bound, it is perhaps reasonable to assume that free Pi is close to 10-15 m M (see reference 24 for a review of this subject) in Pi replete mitochondria in which case sigificant inhibition of pyruvate carboxylase may 40 4000 occur. The lack of a requirement of an external 0 4 source of Pi during the gluconeogenesis experiments can be interpreted perhaps as indirect evipmol PDB / mg protein dence for mobilization of mitochondrial Pi. Figure 5. Dose-dependence of the activation of renal gluconSince Pi can act as an exchange anion in the eogenesis by PDB. Isolated tubule cells were suspended in Ca- movement of anionic substrates across the mitofree medium containing 1 mM EGTA for 1 min and then treated chondrial membrane, an efflux of Pi following a for 5 min with PDB at the indicated concentrations. The cells were then washed to remove the EGTA and gluconeogenesis depletion episode would tend to favour uptake of pyruvate and succinate into the mitochondria. The monitored for 60 min in medium containing 2 mM Ca2*.
I
.
~
~
37
CONTROL OF RENAL GLUCONEOGENESIS
uptake would be expected to favour the use of these substrates as glucose precursors. The failure of Pi depletion to stimulate gluconeogenesis from glycerol could then be explained because this substrate is converted to glucose in steps that are all extramitochondrial. The stimulation of glucose production from malate by depletion is difficult to explain however on this basis. If phosphoenolpyruvate carboxykinase is cytosolic in the mouse (as it is in the rat), malate conversion to glucose will also be extramitochondrial since oxaloacetate will result from the action of cytosolic malate dehydrogenase. Because cytosolic Pi appears to be buffered against wide fluctuations in the Pi content of the surrounding m e d i ~ m ~it, is ’ ~difficult to explain the change in fructose-2,6-bisphosphate content accompanying depletion by an assumption of an allosteric action of Pi on phosphofructokinase 2 and fructose bisphosphatase 2. Our results differ from those obtained with tubules isolated from animals reared on a phosphate deficient diet.’ The tubules seemed to have impaired ATP production.’ An explanation for the difference in the results is that ‘acute’ depletion produced by isolation of cells in Pi-free conditions, and thus exposed to low Pi for approximately 45 to 60 min, has less effect on cellular ATP levels than does the rearing of animals for several weeks on a Pi-free diet. A significant impairment of ATP synthesis brought about by the lengthy exposure to low Pi would be expected to interfere with gluconeogenic activity and perhaps mask changes in glucose production arising from effects of altered Pi on the activity of certain enzymes. Inhibition by phorbol esters of gluconeogenesis in canine proximal tubule segments has been described previouslyz6 and the conclusion reached that protein kinase C can regulate renal gluconeogenesis by raising cytosolic pH through stimulation of N a + / H exchange. The conditions under which the inhibition was observedz6 correspond to those in which either no effect or inhibition was seen in our experiments i.e. the phorbol treatment was applied in the presence of C a 2 + ions. The stimulatory action of phorbol that we observe when the agent is applied in the absence of C a 2 + may be connected with the state of association of protein kinase C with the plasma membrane.27*28Binding of phorbol to the kinase may increase its sensitivity to Ca2 ions. When gluconeogenesis is conducted in the normal medium containing physiological concentrations of Ca2 the increased sensitivity is reflected in the stimulation of gluconeogenic rate. +
+
+
The extracellular source of calcium is probably needed for a localized pool of the cation in the plasma membrane from which Ca2+ that activates kinase C is drawn.” The putative target protein(s) for kinase C, phosphorylation of which affects gluconeogenesis, cannot be identified from these studies except that it must be concerned with the part of the pathway to the triose phosphate stage since glucose production from glycerol is unaffected by phorbol. Since stimulation is observed when depleted cells are treated with phorbol it is likely that Pi depletion and protetin kinase C do not influence identical steps in the gluconeogenesis pathway. Phorbol treatment, especially with the potent PMA, in the presence of calcium may stimulate protein kinase C to such an extent that the subsequent response is that of negative feedback” and any stimulatory action becomes masked. Alternatively, superoxide generation under these conditions may cause general damage to the Parathyroid hormone has been shown to promote diacylglyceride formation in kidney proximal tubules30 and stimulation of renal gluconeogenesis by parathyroid hormone required CaZ+.31On the basis therefore that phorbol esters mimic diacylglycerol in stimulating protein kinase C , one can speculate that the control of gluconeogenesis by parathyroid hormone is mediated by this kinase. This conclusion is supported by the finding that the hormone induces CaZ transients in cultered proximal tubule cells.32Protein kinase A (CAMPdependent) is probably less important in the parathyroid hormone regulation of gluconeogenesis. +
REFERENCES 1. Nagata, N. and Rasmussen, H. (1970). Parathyroid hormone, 3‘,5‘cAMP, Ca2+ and renal gluconeogenesis. Proc. Narl. Acad. Sci. (U.S.A.),65, 368-374. 2. Roobol, A. and Alleyne, G. A. D. (1973). Regulation of renal gluconeogenesis by calcium ions, hormones and adenosine 3‘3’ cyclic monophosphate. Biochem. J . , 134, 157-165. 3. Gmaj, P. and Murer, H. (1986). Cellular mechanism of inorganic phosphate transport in kidney. PhysioL. Rev., 66, 36-70. 4. Butterworth, P. J. (1987). Phosphate horneostastis. Mol. Aspects Med., 9, 289-386. 5. Massry, S. G. and Fleisch, H. (1980). Renal Handling of Phosphate. Plenum Medical Book Co.: New York. 6. Grahn, M. F. and Butterworth, P. J. (1986). Effects of acute phosphate depletion in isolated chick kidney tubule cells. Cell Biochem. Funct., 4, 271-275. 7. Caverzasio, J., Brown, C . D. A., Biber, J., Bonjour, J.-P. and Murcr, H. (1985). Adaptation of phosphate transport in
38
8.
9.
10.
11.
12. 13 14
15 16
17
18.
19.
20.
T. NAKAGAWA AND P. J. BUTTERWORTH
phosphate deprived LLC-PKI cells. Am. J . Physiol., 248, F122-F127. Kempson, S. A., Colon-Otero, G.,Lise-Ou, S. Y., Turner, S. T. and Dousa, T. P. (1981). Possible role of nicotinamide adenine dinucleotide as an intracellular regulator of renal transport of phosphate in the rat. J . Clin. Invest., 67, 1347-1 360. Kreusser, W. J., Descoeurdres, C. Oda, Y.,Massry, S. G. and Kurokawa, K. (1980). Effect of phosphate depletion on renal gluconeogenesis. Mineral Electrolyt. Metab., 3, 312-322. Jahan, M. and Butterworth, P. J. (1988). Study of the mechanism by which the Na+-P, cotransporter of mouse proximal tubule cells adjusts to phosphate depletion. Biochem. J . 252, 105-109. Van Schaftingen, E. (1984). ~-Fructose-2,6-bisphosphate. In: Methods ofEnzymatic Analysis, Vol. 6. (Bermeyer, H. U. ed.) 3rd ed. Verlag Chemie: Weinheim, GFR, pp. 335-341. Gomperts, B. D. (1983): Involvement of guanine nucleotide binding protein in the gating of Ca2+ by receptors. Nature (Lond), 306,64-66. Rorive, G. and Kleinzeller, A. (1972). The effect of ATP and C aZ +on the cell volume in isolated kidney tubules. Biochim. Biophys. Acta, 274,226-239. Bergmeyer, H. U., Berndt, E., Schmidt, F. and Stork, H. (1974). D-Glucose: Determination with hexokinase and glucose-6-phosphate dehydrogenase. In: Methods of Enzymatic Analysis, Vol. 3. (Bergmeyer, H. U. ed.) 2nd ed. Academic Press: New York, pp. 1196-1198. Hue, L. and Rider, M. H. (1987). Role of fructose-2,6bisphosphate in the control of glycolysis in mammalian tissues. Biochem. J., 245, 313-324. Wirthensohn, G. A., Vandewalle, A. and Guder, W. G. (1981). Renal glycerol metabolism and the distribution of glycerol kinase in rabbit nephron. Biochem, J., 198, 543-549. Guder, W. G. and Ruprecht, A. R. (1975). Metabolism of isolated kidney tubules. Independent actions of catecholamines on renal cyclic adenosine 3’3’ monophosphate levels and glyconeogenesis. Eur. J. Biochem., 52,283-290. Stermann, R. and Decker, K. (1978). Differential response of ATP and orthophosphate in cytosol and mitochondria of rat hepatocytes to treatment with Pi and galactosamine. Febs Lett., M, 214-216. Wu, K. I., Bacon, R. A., Al-Mahrouq, H. A. and Kempson, S. A. (1988). Nicotinamide as a rapid acting inhibitor of renal brush border phosphate transport. Am. J . Physiol., 255, F15-F21. McClure, W. R., Lardy, H. A., Wagner, M. and Cleland, W. W. (1971). Rat liver pyruvate carboxylase I1 Kinetic studies of the forward reaction. J. Biol. Chem., 246, 3579-3583.
21. McClure, W. R. and Lardy, H. A. (1971). Rat liver pyruvate carboxylase IV Factors affecting the regulation in viva J . Biol. Chem., 246, 3591-3596. 22. Heldt, H. W., Klingenberg, M. and Milovancev, M. (1972). Differences between the ATP-ADP ratios in the mitochondrial matrix and in the extramitochondrial space. Eur. J . Biochem., 30,434-440. 23. Roos, I., Crompton, M. and Carafoli, E. (1980). The role of inorganic phosphate in the release of Ca2+ from rat liver mitochondria. Eur. J . Biochem., 110, 319-325. 24. Sestoft, L. and Bartels, P. D. (1981). Regulation of metabolism by inorganic phosphate. In: Short Term Regulation of Metabolism. (Hue, L. and Van Der Werve, G . eds) ElsevierNorth Holland Biomedical Press: Amsterdam, New York and Oxford, pp. 427-452. 25. Bevington, A., Mundy, K. I., Yates, A. J. P., Kanis, J. A., Taylor, D. J., Rajagopalan, B., Russell, R. G. G. and Radda, G. K. (1986). A study of intracellular orthophosphate concentrations in human muscle and erythrocytes by 31P nuclear magnetic resonance spectroscopy and selective chemical assay. Clin. Sci., 72, 729-735. 26. Rogers, S., Gavin, J. R. and Hammerman, M. R. (1985). Phorbol esters inhibit gluconeogenesis in canine renal proximal tubular segments. Am. J . Physiol., 249, F256-F262. 21. Kishimoto, A., Takai, Y., Mori, T., Kikkawa, U. and Nishizuka, Y. (1980). Activation of calcium and phospholipid dependent protein kinase by diacyl glycerol; its possible relation to phosphatidylinositol turnover. J . Biol. Chem., 255, 2273-2276. 28. Niedel, J. E. and Blackshear, P. J. (1986). In: Phosphoinositides and Receptor Mechanisms. Alan R. Liss, Inc., pp. 47-88. 29. Nishizuka, Y. (1986). Studies and perspectives of protein kinase C. Science, 233, 305-312. 30. Hruska, K. A,, Goligorsky, M., Scoble, J., Tsutsumi, M., Westbrook, S. and Moskowitz, D. (1986). Effects of parathyroid hormone on cytosolic calcium in renal proximal tubular primary cultures. Am. J . Physiol., 251, F188-Fl98. 31. MacDonald, D. W. R. and Saggerson, E. D. (1977). Hormonal control of gluconeogenesis in tubule fragments from renal cortex of fed rats. Biochem. J., 168, 33-42. 32. Goligorsky, M. S., Osborne, D., Howard, T., Hruska, K. A. and Karl, I. E. (1987). Hormonal control of gluconeogenesis in cultured proximal tubular cells: role of cytosolic calcium. Am. J . Physiol., 253, F802-F809.
Received 17 July 1989 Accepted 29 September 1989
