- Email: [email protected]
Chronopharmacological study of furosemide; (IX) influence of continuous norepinephrine infusion
Life Sciences, Vol. 50, pp. 449-455 Printed in the USA
Pergamon Press
CHRONOPHARMACOLOGICAL STUDY OF FUROSEMIDE; (IX) INFLUENCE OF CONTINUOUS NOREPINEPHRINE INFUSION Akio Fujimura, Kyo-ichi Ohashi and Akio Ebihara Department of Clinical Pharmacology, Jichi Medical School Minamikawachi-machi, Kawachi-gun, Tochigi 329-04, Japan (Received in final form December 10, 1991)
Summary Our previous studies have suggested that the adrenergic nervous system is involved in the mechanisms responsible for the time-dependent changes in the effects of furosemide in rats. To examine this hypothesis further, norepinephrine (150 ~g/kg/hr) or its vehicle alone was infused subcutaneously by osmotic minipumps. Furosemide (30 mg/kg) was given orally at 12 am or 12 pm. Urine was collected for 8 hours after the agent, and urinary excretions of sodium and furosemide were determined. Urine volume and urinary excretion of sodium and furosemide were significantly greater at 12 am than at 12 pm in the vehicle-infused group of rats. However these administration-time-dependent changes in the effects of furosemide and its urinary amount disappeared in the norepinephrine-infused group of animals. Since chronic norepinephrine infusion is considered to disturb the axis of adrenergic nervous system, these data support the hypothesis concerning the mechanisms of this chronopharmacological phenomenon of furosemide.
Continuous norepinephrine infusion causes a marked elevation in plasma and tissue concentrations of the amine, and subsequently stimulates or suppresses the impulse frequency from sympathetic nerve (i). In addition, the numbers of 6- and ~-adrenergic receptors are decreased by down-regulation mechanism during norepinephrine infusion (2). Therefore, the axis of adrenergic nervous system is considered to be disturbed during continuous norepinephrine infusion. We have previously demonstrated the time-dependent changes in the diuretic effects of furosemide, a loop diuretic agent, in rats (3, 4). Since these diurnal changes disappeared following pretreatment with ~-adrenoceptor blocking agent (5) or 6-hydroxydopamine (6) which produces a selective destruction of adrenergic nerve endings, it is assumed that adrenergic nervous system may contribute to this chronopharmacological phenomenon. To examine the hypothesis further, norepinephrine was infused subcutaneously in rats using an osmotic minipump. The chronopharmacological profiles of furosemide during norepinephrine infusion were compared to those during a control period.
Materials and Methods Male Wistar rats (Charles River Laboratory, Kanagawa, Japan) weighing 300 to 350 g were maintained for more than 2 weeks under conditions of light from 7 am to 7 pm and dark from 7 pm to 7 am with free access to food and water. 0024-3205/92 $5.00 + .00 Copyright © 1992 Pergamon Press plc All rights reserved.
450
Chronopharmacology of Furosemide
Vol. 50, No. 6, 1992
In study I, 3% body weight (b.w.) of i% NaC1 solution was given by gavage into the stomach at 12 am (or 12 pm). Twenty-four hours after the vehicle, 30 mg/kg of furosemide in 3% b.w. of vehicle was given orally at 12 am (or 12 pm). Urine was collected for 8 hours following vehicle alone or the drug administration at 12 am (or 12 pm). Food and water were deprived during 8 hours after each administration. The administration of the drug was randomly assigned to 12 am or 12 pm. The washout period between the two sets of experiments was 2 days. Thereafter, these rats were devided into two groups. Norepinephrine (Sigma, St. Louis, MO, U.S.A.) was infused subcutaneously in the first group of rats (n=ll) while its vehicle alone was infused in the second group of animals (n=12). Five days after the initiation of infusion, the identical protocol of study I was repeated (study ~). Norepinephrine infusion An osmotic minipump (Model 2ML2, Alzet, Alza Corp., Palo Alto, CA, U.S.A.) was implanted subcutaneously on the back of each rat. This model of pump is designed to infuse drug continuously for up to 14 days at a constant rate. Vehicle consisted of 0.9% NaC1 solution containing 1 mg of ascorbic acid/ml. The concentration of norepinephrine in the minipumps was calculated to achieve an infusion rate of 150 ~g/kg/hr. Urinary sodium concentration was determined by flame photometry (Flame Photometer 775-A, Hitachi, Tokyo, Japan). Urinary furosemide concentration was measured by high pressure liquid chromatography (7). The sensitivity of this assay was 0.i ~g/ml and the coefficient of variation was 8.0%. The results are expressed as the means ± SE. Data were analyzed by analysis of variance and the Wholly-Significant-Difference Method.
Results When 3% b.w. of NaC1 solution was given as a furosemide control, no significant difference was observed in urine volume or urinary sodium excretion in the collection period following the 12 am administration compared to the collection period beginning at 12 pm in both groups of rats (FIG. 1 and 2). The values of these parameters in study ~ were significantly (p
Vol. 50, No. 6, 1992
Chronopharmacology of Furosemide
I
451
II
80p
,
p
40-
I
®~_
0-
,
E ~~ ._~
p
I
~
,
1
~
,
-
p
I
:~.¢5 1000
p<0.05----1
I
I
z
¢o 2000]
B
d
A
,
p
I Z
I
0 J
12am
12pm
12am
12pm
administration time
FIG. 1 Urine volume and urinary excretion of sodium and furosemide after furosemide administration in rats during a control period (I) or a continuous infusion of vehicle (~). mean ± SE, n=12 Urine was collected for 8 hours after the agent. NaC1 solution alone t I NaC1 solution + furosemide
Discussion To obtain and maintain the elevated concentrations of plasma norepinephfine, the agent was infused subcutaneously using osmotic minipumps with a constant rate of 150 ~g/kg/hr in the present study. This infusion rate of norepinephrine is demonstrated to increase plasma norepinephrine lO-fold in rats, which remained at this level for the duration of the infusion (II days) (2). Down-regulation of renal cortical ~- and ~-adrenergic receptors is also shown to occur rapidly and reach a plateau after 2 to 3 days (2). On the other hand, Johnson et al. examined the influence of norepinephrine infusion on impulse frequency from sympathetic nerve in rats (1). They found that its influence is dependent on the rate of blood pressure elevation elicited by the norepinephrine infusion, and a rapid rise in blood pressure results in a
452
Chronopharmacology of Furosemide
Vol. 50, No. 6, 1992
II 80-
u)
o.)__
4o. ,
p
,
p
o
~:
E ~.
~:
e~ ~W
•~- E
8
._ ~
.~ .
2000
]
,
p
0 .J
12am
12pm
12am
administration
time
12pro
FIG. 2 Urine volume and urinary excretion of sodium and furosemide after furosemide administration in rats during a control period (I) or a continuous infusion of norepinephrine (~). mean ± SE, n=ll Urine was collected for 8 hours after the agent. v///A NaCI solution alone I - - I NaCI solution + furosemide
decrease in sympathetic impulse frequency while a slow rise causes an activation of sympathetic tone. These observations indicate that the axis of sympathetic nervous system might be disturbed during the continuous infusion of 150 ug/kg/hr of norepinephrine in the present study. We have recently published data indicating that the adrenergic system is involved in the mechanisms responsible for the time-dependent changes in the effects of furosemide in rats (5, 6). The present study demonstrated that the diurnal variations in the effects of furosemide disappeared with norepinephrine infusion. Thus, the present study provides additional evidence and supports the hypothesis concerning the mechanisms of the chronopharmacological phenomenon of furosemide.
Vol. 50, No. 6, 1992
Chronopharmacology
100-
in
of Furosemide
I
453
II
12am ; r=0.68 (p
.d
12am ; r=0.59 (p
i° o ° o
o
0;
o
o
50-
>o
eeel
12pro ; r--037 (p
°°
o
•
5)
n=12
O"
1012pm ; r=0.70 (p
12pm ; r=0.60 (p
•
o
.6 ~r LU
E
5-
E"
/
o
;=. n=12
Z"
O"
1600
.1)
20'00 6
12am ; r=0.67 (p
10'00
20'00
urinary furosernide, p.g/kg b.w./8hrs
FIG. 3 Relationship between urinary furosemide and its diuretic effects in rats with (~) or without (1) vehicle infusion.
The urinary furosemide excretion of the day and night trials did not significantly differ in the norepinephrine-infused rats. Since the significant correlations were observed between the urinary furosemide and its diuretic effects in the day and night trials and their regression lines did not differ among the two trials in the present study, the time-dependent changes in the diuretic effects of furosemide might have disappeared as the diurnal variations in the amount of urinary furosemide disappeared in the group of animals with norepinephrine infusion. Several renal and hormonal factors involved in the regulation of water and electrolyte homeostasis exhibit circadian changes (8-11). Among these factors, renal glomerular function, plasma renin activity and plasma atrial natriuretic peptide are influenced by the increased plasma norepinephrine (12, 13).
454
Chronopharmacology of Furosemide
100-
Vol. 50, No. 6, 1992
]]
I
oj
12arn; r=0.58 (p
12am ; r=070 (p<0.05) / n=11
o
e-
50-
o e
'6>
o
¢"
12pm ; r=061 (p
12pm ; r=O0721(p
n=11
0
10-
12pro; r=0.68 (p<0 05) n=11
¢/)
12pro; r=071 (p
Y //if o
o" UJ
E
5-
:5o
12am; r=0.72(p<0.01) n=l 1 O-
1000
2000 0
12am ; r=0.62 (p<0.05) n= 11
1000
2000
urina~ furosemide, ~g~g b.w/8hrs
FIG. 4 Relationship between urinary furosemide and its diuretic effects in rats with (~) or without (I) norepinephrine infusion.
Therefore, the mode of their circadian rhythms might be altered during chronic norepinephrine infusion, and consequently the diurnal variation in the effects of furosemide might be influenced. The norepinephrine infusion per se alters water and sodium excretion by several pathways. Chronic infusion of norepinephrine causes an elevation in blood pressure (i) which in turn enhances urine output and sodium excretion (14). In addition, norepinephrine might influence water and/or sodium excretion directly (12) or indirectly via an increased production of renal prostaglandin E= (15). These mechanisms may, at least in part, explain the enhanced diuresis following NaC1 solution alone in the norepinephrine-infused rats.
Vol, 50, NO. 6, 1992
Chronopharmacology
of Furosem£de
455
References i. M.D. JOHNSON, P.G. SMITH, E. MILLS and S.M. SCHANBERG, J. Pharmacol. Exp. Ther. 227 254-259 (1983). 2. M.D. SNAVELY, M.G. ZIEGLER and P.A. INSEL, Endocrinology 117 2182-2189 (1985). 3. A. FUJIMURA and A. EBIHARA, Life Sci. 38 1215-1220 (1986). 4. A. FUJIMURA and A. EBIHARA, Life Sci. 45 2459-2464 (1989). 5. A. FUJIMURA and A. EBIHARA, Life Sci. 42 1431-1437 (1988). 6. A. FUJIMURA and A. EBIHARA, Life Sci. 46 827-831 (1990). 7. M.L. MACDOUGALL, D.W. SHOEMAN and D.L. AZARNOFF, Research Communications in Chemical Pathology and Pharmacology 10285-291 (1975). 8. D.R. LUKE, K.M. WASAN and K. VADIEI, Chronobiology: Its Role in Clinical Medicine~ General Biology~ and Agriculture, Part A, ed. by D.K. HAYES, J.E. PAULY and R.J. REITER, pp. 131-136, Wiley-Liss, Inc., New York (1990). 9. M.G. KOOPMAN, G.C.M. KOOMEN, R.T. KREDIET, E.A.M. DE MOOR, F.J. HOEK and L. ARISZ, Clin. Sci. 77 I05-111 (1989), i0. P. REDON, M. LAVAL and J. CAMBAR, Annual Review of Chronopharmacology, Vol. 6, ed. by A. REINBERG, M. SMOLENSKY and G. LABRECQUE, pp. 183-223, Pergamon Press, London (1990). II. A.M. RICHARDS, G. TONOLO, R. FRASER, J.J. MORTON, B.J. LECKIE, S.G. BALL and J.I.S. ROBERTSON, Clin. Sci. 73 489-495 (1987). 12. R.W. SCHRIER, Kidney Int. 6 291-306 (1974). 13. D.E. UEHLINGER, T. ZAMAN, P. WEIDMANN, S. SHAW and M.P. GNADINGER, Hypertension i0 249-253 (1987). 14. J.D. FIRTH, A.E.G. RAINE and J.G.G. LEDINGHAM, J. Hyper. 8 97-103 (1990). 15. D.I. DIZ, P.G. BAER and A. NASJLETTI, Am. J. Physiol. 241--F477-F481 (1981).
Pergamon Press
CHRONOPHARMACOLOGICAL STUDY OF FUROSEMIDE; (IX) INFLUENCE OF CONTINUOUS NOREPINEPHRINE INFUSION Akio Fujimura, Kyo-ichi Ohashi and Akio Ebihara Department of Clinical Pharmacology, Jichi Medical School Minamikawachi-machi, Kawachi-gun, Tochigi 329-04, Japan (Received in final form December 10, 1991)
Summary Our previous studies have suggested that the adrenergic nervous system is involved in the mechanisms responsible for the time-dependent changes in the effects of furosemide in rats. To examine this hypothesis further, norepinephrine (150 ~g/kg/hr) or its vehicle alone was infused subcutaneously by osmotic minipumps. Furosemide (30 mg/kg) was given orally at 12 am or 12 pm. Urine was collected for 8 hours after the agent, and urinary excretions of sodium and furosemide were determined. Urine volume and urinary excretion of sodium and furosemide were significantly greater at 12 am than at 12 pm in the vehicle-infused group of rats. However these administration-time-dependent changes in the effects of furosemide and its urinary amount disappeared in the norepinephrine-infused group of animals. Since chronic norepinephrine infusion is considered to disturb the axis of adrenergic nervous system, these data support the hypothesis concerning the mechanisms of this chronopharmacological phenomenon of furosemide.
Continuous norepinephrine infusion causes a marked elevation in plasma and tissue concentrations of the amine, and subsequently stimulates or suppresses the impulse frequency from sympathetic nerve (i). In addition, the numbers of 6- and ~-adrenergic receptors are decreased by down-regulation mechanism during norepinephrine infusion (2). Therefore, the axis of adrenergic nervous system is considered to be disturbed during continuous norepinephrine infusion. We have previously demonstrated the time-dependent changes in the diuretic effects of furosemide, a loop diuretic agent, in rats (3, 4). Since these diurnal changes disappeared following pretreatment with ~-adrenoceptor blocking agent (5) or 6-hydroxydopamine (6) which produces a selective destruction of adrenergic nerve endings, it is assumed that adrenergic nervous system may contribute to this chronopharmacological phenomenon. To examine the hypothesis further, norepinephrine was infused subcutaneously in rats using an osmotic minipump. The chronopharmacological profiles of furosemide during norepinephrine infusion were compared to those during a control period.
Materials and Methods Male Wistar rats (Charles River Laboratory, Kanagawa, Japan) weighing 300 to 350 g were maintained for more than 2 weeks under conditions of light from 7 am to 7 pm and dark from 7 pm to 7 am with free access to food and water. 0024-3205/92 $5.00 + .00 Copyright © 1992 Pergamon Press plc All rights reserved.
450
Chronopharmacology of Furosemide
Vol. 50, No. 6, 1992
In study I, 3% body weight (b.w.) of i% NaC1 solution was given by gavage into the stomach at 12 am (or 12 pm). Twenty-four hours after the vehicle, 30 mg/kg of furosemide in 3% b.w. of vehicle was given orally at 12 am (or 12 pm). Urine was collected for 8 hours following vehicle alone or the drug administration at 12 am (or 12 pm). Food and water were deprived during 8 hours after each administration. The administration of the drug was randomly assigned to 12 am or 12 pm. The washout period between the two sets of experiments was 2 days. Thereafter, these rats were devided into two groups. Norepinephrine (Sigma, St. Louis, MO, U.S.A.) was infused subcutaneously in the first group of rats (n=ll) while its vehicle alone was infused in the second group of animals (n=12). Five days after the initiation of infusion, the identical protocol of study I was repeated (study ~). Norepinephrine infusion An osmotic minipump (Model 2ML2, Alzet, Alza Corp., Palo Alto, CA, U.S.A.) was implanted subcutaneously on the back of each rat. This model of pump is designed to infuse drug continuously for up to 14 days at a constant rate. Vehicle consisted of 0.9% NaC1 solution containing 1 mg of ascorbic acid/ml. The concentration of norepinephrine in the minipumps was calculated to achieve an infusion rate of 150 ~g/kg/hr. Urinary sodium concentration was determined by flame photometry (Flame Photometer 775-A, Hitachi, Tokyo, Japan). Urinary furosemide concentration was measured by high pressure liquid chromatography (7). The sensitivity of this assay was 0.i ~g/ml and the coefficient of variation was 8.0%. The results are expressed as the means ± SE. Data were analyzed by analysis of variance and the Wholly-Significant-Difference Method.
Results When 3% b.w. of NaC1 solution was given as a furosemide control, no significant difference was observed in urine volume or urinary sodium excretion in the collection period following the 12 am administration compared to the collection period beginning at 12 pm in both groups of rats (FIG. 1 and 2). The values of these parameters in study ~ were significantly (p
Vol. 50, No. 6, 1992
Chronopharmacology of Furosemide
I
451
II
80p
,
p
40-
I
®~_
0-
,
E ~~ ._~
p
I
~
,
1
~
,
-
p
I
:~.¢5 1000
p<0.05----1
I
I
z
¢o 2000]
B
d
A
,
p
I Z
I
0 J
12am
12pm
12am
12pm
administration time
FIG. 1 Urine volume and urinary excretion of sodium and furosemide after furosemide administration in rats during a control period (I) or a continuous infusion of vehicle (~). mean ± SE, n=12 Urine was collected for 8 hours after the agent. NaC1 solution alone t I NaC1 solution + furosemide
Discussion To obtain and maintain the elevated concentrations of plasma norepinephfine, the agent was infused subcutaneously using osmotic minipumps with a constant rate of 150 ~g/kg/hr in the present study. This infusion rate of norepinephrine is demonstrated to increase plasma norepinephrine lO-fold in rats, which remained at this level for the duration of the infusion (II days) (2). Down-regulation of renal cortical ~- and ~-adrenergic receptors is also shown to occur rapidly and reach a plateau after 2 to 3 days (2). On the other hand, Johnson et al. examined the influence of norepinephrine infusion on impulse frequency from sympathetic nerve in rats (1). They found that its influence is dependent on the rate of blood pressure elevation elicited by the norepinephrine infusion, and a rapid rise in blood pressure results in a
452
Chronopharmacology of Furosemide
Vol. 50, No. 6, 1992
II 80-
u)
o.)__
4o. ,
p
,
p
o
~:
E ~.
~:
e~ ~W
•~- E
8
._ ~
.~ .
2000
]
,
p
0 .J
12am
12pm
12am
administration
time
12pro
FIG. 2 Urine volume and urinary excretion of sodium and furosemide after furosemide administration in rats during a control period (I) or a continuous infusion of norepinephrine (~). mean ± SE, n=ll Urine was collected for 8 hours after the agent. v///A NaCI solution alone I - - I NaCI solution + furosemide
decrease in sympathetic impulse frequency while a slow rise causes an activation of sympathetic tone. These observations indicate that the axis of sympathetic nervous system might be disturbed during the continuous infusion of 150 ug/kg/hr of norepinephrine in the present study. We have recently published data indicating that the adrenergic system is involved in the mechanisms responsible for the time-dependent changes in the effects of furosemide in rats (5, 6). The present study demonstrated that the diurnal variations in the effects of furosemide disappeared with norepinephrine infusion. Thus, the present study provides additional evidence and supports the hypothesis concerning the mechanisms of the chronopharmacological phenomenon of furosemide.
Vol. 50, No. 6, 1992
Chronopharmacology
100-
in
of Furosemide
I
453
II
12am ; r=0.68 (p
.d
12am ; r=0.59 (p
i° o ° o
o
0;
o
o
50-
>o
eeel
12pro ; r--037 (p
°°
o
•
5)
n=12
O"
1012pm ; r=0.70 (p
12pm ; r=0.60 (p
•
o
.6 ~r LU
E
5-
E"
/
o
;=. n=12
Z"
O"
1600
.1)
20'00 6
12am ; r=0.67 (p
10'00
20'00
urinary furosernide, p.g/kg b.w./8hrs
FIG. 3 Relationship between urinary furosemide and its diuretic effects in rats with (~) or without (1) vehicle infusion.
The urinary furosemide excretion of the day and night trials did not significantly differ in the norepinephrine-infused rats. Since the significant correlations were observed between the urinary furosemide and its diuretic effects in the day and night trials and their regression lines did not differ among the two trials in the present study, the time-dependent changes in the diuretic effects of furosemide might have disappeared as the diurnal variations in the amount of urinary furosemide disappeared in the group of animals with norepinephrine infusion. Several renal and hormonal factors involved in the regulation of water and electrolyte homeostasis exhibit circadian changes (8-11). Among these factors, renal glomerular function, plasma renin activity and plasma atrial natriuretic peptide are influenced by the increased plasma norepinephrine (12, 13).
454
Chronopharmacology of Furosemide
100-
Vol. 50, No. 6, 1992
]]
I
oj
12arn; r=0.58 (p
12am ; r=070 (p<0.05) / n=11
o
e-
50-
o e
'6>
o
¢"
12pm ; r=061 (p
12pm ; r=O0721(p
n=11
0
10-
12pro; r=0.68 (p<0 05) n=11
¢/)
12pro; r=071 (p
Y //if o
o" UJ
E
5-
:5o
12am; r=0.72(p<0.01) n=l 1 O-
1000
2000 0
12am ; r=0.62 (p<0.05) n= 11
1000
2000
urina~ furosemide, ~g~g b.w/8hrs
FIG. 4 Relationship between urinary furosemide and its diuretic effects in rats with (~) or without (I) norepinephrine infusion.
Therefore, the mode of their circadian rhythms might be altered during chronic norepinephrine infusion, and consequently the diurnal variation in the effects of furosemide might be influenced. The norepinephrine infusion per se alters water and sodium excretion by several pathways. Chronic infusion of norepinephrine causes an elevation in blood pressure (i) which in turn enhances urine output and sodium excretion (14). In addition, norepinephrine might influence water and/or sodium excretion directly (12) or indirectly via an increased production of renal prostaglandin E= (15). These mechanisms may, at least in part, explain the enhanced diuresis following NaC1 solution alone in the norepinephrine-infused rats.
Vol, 50, NO. 6, 1992
Chronopharmacology
of Furosem£de
455
References i. M.D. JOHNSON, P.G. SMITH, E. MILLS and S.M. SCHANBERG, J. Pharmacol. Exp. Ther. 227 254-259 (1983). 2. M.D. SNAVELY, M.G. ZIEGLER and P.A. INSEL, Endocrinology 117 2182-2189 (1985). 3. A. FUJIMURA and A. EBIHARA, Life Sci. 38 1215-1220 (1986). 4. A. FUJIMURA and A. EBIHARA, Life Sci. 45 2459-2464 (1989). 5. A. FUJIMURA and A. EBIHARA, Life Sci. 42 1431-1437 (1988). 6. A. FUJIMURA and A. EBIHARA, Life Sci. 46 827-831 (1990). 7. M.L. MACDOUGALL, D.W. SHOEMAN and D.L. AZARNOFF, Research Communications in Chemical Pathology and Pharmacology 10285-291 (1975). 8. D.R. LUKE, K.M. WASAN and K. VADIEI, Chronobiology: Its Role in Clinical Medicine~ General Biology~ and Agriculture, Part A, ed. by D.K. HAYES, J.E. PAULY and R.J. REITER, pp. 131-136, Wiley-Liss, Inc., New York (1990). 9. M.G. KOOPMAN, G.C.M. KOOMEN, R.T. KREDIET, E.A.M. DE MOOR, F.J. HOEK and L. ARISZ, Clin. Sci. 77 I05-111 (1989), i0. P. REDON, M. LAVAL and J. CAMBAR, Annual Review of Chronopharmacology, Vol. 6, ed. by A. REINBERG, M. SMOLENSKY and G. LABRECQUE, pp. 183-223, Pergamon Press, London (1990). II. A.M. RICHARDS, G. TONOLO, R. FRASER, J.J. MORTON, B.J. LECKIE, S.G. BALL and J.I.S. ROBERTSON, Clin. Sci. 73 489-495 (1987). 12. R.W. SCHRIER, Kidney Int. 6 291-306 (1974). 13. D.E. UEHLINGER, T. ZAMAN, P. WEIDMANN, S. SHAW and M.P. GNADINGER, Hypertension i0 249-253 (1987). 14. J.D. FIRTH, A.E.G. RAINE and J.G.G. LEDINGHAM, J. Hyper. 8 97-103 (1990). 15. D.I. DIZ, P.G. BAER and A. NASJLETTI, Am. J. Physiol. 241--F477-F481 (1981).