- Email: [email protected]
Mechanism of increased renal prostaglandin E2 in uranyl nitrate-induced acute renal failure
PROSTAGLANDINS
MECHANIW
OF
INCREASED
RENAL
NITRATE-INDUCED
Anshumali
PROSTAGLANDIN
ACUTE
Chaudhari
and
Division
of
RENAL
Michael
A.
E2
IN
URANYL
FAILURE
Kirschenbaum
Nephrology
Department of Medicine UCLA School of Medicine Los Angeles, California 90024
ABSTRACT We have previously demonstrated that decreased cortical prostaglandin metabolism can contribute significantly to an increase in renal tissue levels in bilateral ureteral obstruction, a model of and activity of prostaglandin E In the present study, we have further investigated acute renal failure. whether alterations in prostaglandin metabolism can occur in a nephrotoxic Prostaglandin synthesis, prostaglandin E2 model of acute renal failure. metabolism (measured as both prostaglandin E2-9-ketoreductase and prostaglandin E2-15-hydroxydehydrogenase activity), and tissue concentration of prostaglandin E2 were determined in rabbit kidneys following an intravenous administration of No changes in the rates of cortical microsomal uranyl nitrate (5 mg/kg). were noted at the end of 1 and prostaglandin E2 and prostaglandin F20: synthesis 3 days, while medullary synthesis of prostaglandin E2 ,fell by 47% after 1 day Cortical cytosolic prostaglandm E2-9-ketoreductase and 43% after 3 days. activity was found to be decreased by 36% and 76% after 1 and 3 days No significant changes were noted in cortical cytosolic respectively. prostaglandin E2-15-hydroxydehydrogenase activity after 3 days. Cortical by 500% at the end of 3 days. tissue levels of prostaglandin E2 increased These data demonstrate that in nephrotoxic acute renal failure, decreased prostaglandin E -9-ketoreductase activity) can prostaglandin metabolism (i.e., result in increased tissue levels of prostaglan c?in E2 in the absence of increased prostaglandin synthesis and suggest that alterations in prostaglandin metabolism may be an important regulator of prostaglandin activity in acute renal failure.
INTRODUCTION Although there is considerable suggestive evidence which might link the various arachidonic acid lnetabolites to the vascular, glomerular and tubular events which occur in acute renal failure (I), there is little direct evidence to Despite the absence of confirmatory experimental support this relationship. evidence, however, there is little doubt that the prostaglandins participate, in some manner, in the numerous homeostatic and pathogenetic mechanisms which accompany an acute disruption of either gloinerular or tubular function. Evidence for the importance of these lipids in acute renal failure has been For example, previous studies have shown both in this and other laboratories. demonstrated increased renal prostaglandin E excretion rates and renal tissue prostaglandin activity in patients with renal disease (2,3) and in animals with experimental models of acute renal failure (4-6). In other studies, the
NOVEMBER
1983 VOL. 26 NO. 5
689
PROSTAGLANDINS
the infusion of vasodilator prostaglandins has been shown to be beneficial in protecting against glomerular injury following the induction of hemodynamically-mediated experimental acute renal failure (7), while the administration of cycle-oxygenase inhibitors under similar conditions has been shown to be detrimental (8). Taken together, these findings suggest that, rather than causing pathologic changes in the nephron, some effect or effects of these lipids may protect the kidney from vasoconstrictive factors which would, if left unopposed, further reduce renal function. The arachidonic acid metabolites which might participate in the homeostatic regulation of renal function (e.g., renal blood flow, glomerular filtration rate and urinary water and electrolyte excretion) can do so in at least two separate and distinct ways: first by some direct action on glomerular, tubular or vascular structures; and second, by modulating the vasoconstrictive influences of other hormones such as angiotensin II and norepinephrine and of sympathetic nervous activity. Since the prostaglandins are primarily local tissue hormones, their biological effects within the kidney would be expected to be dependent, to a great extent, on their cellu!ar concentrations. These intracellular concentrations are the net result of the differences in the rates of prostaglandin synthesis and metabolism. Logically, changes in the rate of metabolism, independent of changes in synthesis could also lead to significant changes in local prostaglandin concentration and thus its activity within the kidney. We recently tested this hypothesis in rabbits and reported that a significant decrease in the renal metabolism of prostaglandin E2 !the major renal prostaglandin in this species) could be demonstrated wrth both uni- and bilateral ureteral obstruction (61, models in which increased prostaglandin biosynthesis has been previously demonstrated (9). The present study was performed to evaluate this hypothesis further in a model of acute renal failure resulting from the administration of a nephrotoxic agent, uranyl nitrate.
METHODS Female white New Zealand rabbits weighing 3-3.5 kg were used in the study. The experimental group (12 animals) received, an intravenous injection of uranyl nitrate at a dose of 5 mg/kg whereas the control group (12 animals) was given an intravenous injection of an equivalent volume of normal saline. Half of the animals in each group were sacrificed on day I and the other half on day 3 following uranyl nitrate administration. Prostaglandin biosynthesis and prostaglandin E2-9-ketoreductase activity was measured in all animals. Eight additional animals were given either uranyl nitrate (4 animals) or normal saline (4 animals) and were sacrificed after 3 days. In this group, prostaglandin E metabolism was assessed by measurement of prostaglandin E2-15-hydroxyde 2, ydrogenase activity. All animals had free access to water and consumed comparable quantities of standard laboratory rabbit chow. Preparation of Subcellular Fractions: For the assay of prostaglandin biosynthesis, cortex and medulla were homogenized separately in 0.02 M Tris buffer (pH 8.0) and centrifuged for 10 min at 9,000 x g. The supernate was removed, further centrifuged for 60 min at 100,000 x g, then discarded. The resultant pellet was homogenized in 0.02 M Tris buffer and adjusted to a
690
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS
protein concentration of 10 mg/ml for the assay. To determine prostaglandin E,2-9-ketoreductase and prostaglandin E2-15-hydroxydehydroenase activities, the ttssue was homogenized in 0.1 M potasstum phosphate buffer (pH 7.3) containing and centrifuged at 100,000 x g for 60 min. 4 mM MgCl and 0.1 mM dithiothreitol The superna ? e was dilute a protein concentration of 5 mg/ml for assay. Protein was determined by the method Protein Estimation: Grossberg (IO) using bovine serum albumin as a standard.
of Sedmak
and
Assay of Prostaglandin Biosynthesis: The method for determining prostaglandin biosynthesis was adapted from that of Tai et al. (II). The reaction mixture I mM ymzp;r;O; f$T;; ;ffiQpH 8.0), 1 mM reduced glutathione, L-labelled arachidonic acid (47 mCi/mmoI, New England Nuclear, Bostkn, MA) and unlabelled arachidonic acid, and microsomal fraction in a final volume of 0.5 ml. Cortical microsomes (2 mg protein) were incubated for 30 min at 37” C and medullary microsomes (I mg protein) for 5 min. The rate of product formation during these periods of incubation was linear under the experimental conditions. The medium was then acidified to pH 3.5 with 1.0 M formic acid. The reaction mixture was extracted twice with 1 ml ethyl acetate, combined and 100 pl ethyl acetate.
evaporated
under
nitrogen,
and
the
residue
dissolved
in
An aliquot was spotted on a plastic silica-gel G thin layer chromatographic plate and the chromatogram developed to a height of 17 cm with organic phase of ethyl acetate/isooctane/acetic acid/water (11:5:2:10) as the solvent system. Authentic standards of prostaglandin E , prostaglandin F thromboxane B and 6-keto-prostaglandin Fbi$i~t;$m~;a, J;~mF’ik;, St. U$$hn Co., Kalamaz$o, MI) and arachidonic acid Louis, MO) were co-chromatographed and the plate was allowed to develop’color in iodine vapor for the localization of arachidonic acid metabolites. In order to separate further prostaglandin E2 fro” thromboxane B2, the plates were first allowed to develop 10 cm from the orlgm and after drying, were again allowed to develop 17 cm from the origin. Each plate was cut into 0.5 cm pieces and placed in counting vials containing ACS@ scintillation cocktail (Anlcrsham was determined in a Beckman Corp., Arlington Heights, IL). The radioactivity LS-7500 liquid scintillation system. Assay of Prostaglandin E -9-ketoreductase: The method for determining prostaglandin E2-9-ketore 8 uctase activity was adapted from that of Stone and Hart (12). The reaction mixture consisted of 0.1 Mfeotassium phosphate buffer containing 4 mM IM~CI, and 0.1 mM dithiothreitol, C-labelled orostaelandin EL L (56 mCi/mol, Amersuham Corp., Arlington Heights, fL) and unlabelied u prostaglandin E2 (final concentration of 46 PM), 10 mM glucose-6-phosphate, 1 unit of glucose-6-phosphate dehydrogenase, 0.1 mM NADPH and a suitable aliquot of cytosolic fraction (200 ug protein) of cortex or medulla in a total volume of 0.5 ml. The reaction mixture was incubated for 40 min at 37” C, acidified with formic acid, and extracted twice with ethyl acetate as described above. The formation of prostaglandin FZa, ,the major prostaglandin E2 metabolite in the rabbit kidney (12,131 was determmed by thin layer chromatography using ethyl acetate/acetone/acetic acid (90: 10: 1) as the solvent system. Assay of Prostaglandin E2-15-hydroxydehydrogenase: This method was also adapted from that of Stone and Hart (12). The reaction mixture for this assay was siTilar to that described above except for the addition of 0.5 mM NADf and a NAD
-regenerating
NOVEMBER
system
which
1983 VOL. 26 NO. 5
consisted
of
3.2
mM
sodium
pyruvate
and
3
691
PROSTAGLANDINS
units
of
lactate
dehydrogenase
listed above. The mixture the reaction were extracted thin layer chromatography phase
of
isooctane:ethyl
instead
of
the
NADPH
and
its
regenerating
system
was incubated for 60 min at 37” C, the product\ with ethyl acetate and analyzed by silicic acid using a solvent system consisting of the organic
acetate:acetic
acid:water
of
(5:11:2:10).
Control
Day
1 after
uranyl
nitrate
Day
3
uranyl
nitrate
after
T I
PGF2ct
PGE2
Figure
1.
Prostaglandin
E2 and Fzoc biosynthesis
by
cortical
microsomes.
Kinetic
analysis was performed using a Lineweaver-Burk plot as modified by For this purpose, prostaglandin E2-,9-ketoreductase activity was Dixon (14). determined at various substrate concentrations m renal cortical cytosolic fractions obtained from both controls and uranyl nitrate treated animals and double reciprocal plot was constructed. Radioreceptor Assay for Prostaglandin E: tissuewere determin.ed by a radioreceptor previously by this laboratory (5). Cortical
692
a
Prostaglandin E levels in cortical assay which has been described tissue was rapidly dissected,
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS
weighed
and
homogenized
for
3rczyk and Rehrman (15). exceed 10 seconds following 25-50 mg of cortical tissue
the
The
extraction
time
of
required
prostaglandins for
by
dissection
and
the
method
weighing
of
did
not
the surgical exposure of the kidneys. In brief, was homogenized at 4” C using a PolytronB consisting of I3 ml of 0.9% NaCl and 0.4 ml of 0.1
N homogenizer in a solution About 2000 counts per minute of H-prostaglandin E2 was added in order to HCI. monitor recovery. This homogenate was extracted with 2.6 ml of a mixture of ethyl acetate:isopropanol (I:l, v:v). The organic phase was isolated after the addition of 3 ml of normal saline and 2 ml of ethyl acetate. After vortexing and a IO minute centrifugation at 3000 rpm, the organic phase was removed and E fraction was collected from a Sephadex dried under N . The prostaglandin LH-20 column2by using acetate:isooctane:acetic the assay.
a solvent acid:water
sys2iem consisting (11:5:2:10) and
u
of the organic phase dried again under N2
of ethyl before
Control
li!al Day
1 after
uranyl
nitrate
Day
3
uranyl
nitrate
after
1000
800 l
PC 0.025
600 *
*
400
TT 40
2c
0 PGE2
Figure
2.
NOVEMBER
Prostaglandin
E2 and Fzrx biosynthesis
1983VOL. 26 NO.5
PGP20:
by t-redullary microsoms.
693
PROSTAGLANDINS
Statistical significance
Analysis: Data are expressed was evaluated by non-paired
as the mean + SE and statistical Student’s “t” test.
RESULTS IMean serum creatinine levels in the experimental group sacrifice were 2.68 + 0.19 mg/dl and 10.0 _+ 0.14 mg/dl to 1.0 _+ 0.1 mg/dl in controls.
TABLE
1. Prostaglandin
E2-9-Ketoreductase
15-Hydroxydehydrogenase
in Renal
Group
pmole
uranyl
3 days post
*p < 0.01, ND
= not
uranyl
**
Cortical
15-Hydroxydehydrogcnase
PGFZX formed/
pmole
I5-keto-PGE2
mg protein/min
mg protein/min
735 + 60
33 -+ 5
Control 1 day post
E2-
Fractions.
9-Ketoreductase Study
on the day of as compared
and Prostaglandin
Activity
Cytosolic
of animals respectively
nitrate nitrate
p < 0.005 as cornpared determined.
470 + 32* xx 175 _+ 45 to control
formed/
ND 34 -+ 5 value.
results of --in vitro cortical and medullary microsomal biosynthesis of administration are shown prostaglandin Ez and PGF20: following uranyl nitrate Unlike our previous observation in the bilateral ureteral Figures 1 and obstruction model of acute renal failure (6), the cortical synthesis of both of these prostaglandins was unaltered at days 1 and 3. However, the medullary prostaglandin E2 biosynthesis in uranyl nitrate treated animals fell by almost 47% after 1 day and 43% after 3 days compared to respective controls. The decrease is consistent with that seen in ureteral obstruction (6). No changes in prostaglandin F20: synthesis were noted. The
in
The major pathway for the inactivation of prostaglandin E2 in the rabbit kidney via prostaglandin involves its transformation to prostaglandin F as described E2-ketoreductase (12,13,16). The changes in ?f IS enzyme activity, earher, have been used as an index of changes in the rate of prostaglandin E2 metabolism (6). Although prostaglandin E2 has previously bpen shown to be metaboliszed by prostaglandin E2-1.5-hydroxydehydrogenase in the rabbit kidney, this pathway appears to be minor m this animal species (12,13,16).
694
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS
VELOCITY -’ PGFZa Synthesized
3
(rmoVmg protein/mh)
r
-IO
-5
0
SUBSTRATE
I 5
1 IO
I
I
7
I5 20 25 CONCENTRATION -’ PGE,
(mM_l)
Figure 3. Lineweaver-Burk plot of prostaglandin E2-P-ketoreductase. Cortical enzyme activity for control animals is shown in the broken line with squares and that for the uranyl nitrate-treated animals is shown in the solid line with circles.
Nevertheless, we have examined the activity of both enzyme systems in this study and the data showing the effect of uranyl nitrate on cortical cytosolic prostaglandin E2 metabolism are summarized in Table 1. A time-dependent decrease in the rate of cortical prostaglandin E2-9-ketoreductase activity was observed as indicated by a 37% inhibition of enzyme activity after 1 day, and 77% after 3 days. The nature of the enzyme inhibition was found to be non-compeitive as determined by a Lineweaver-Burk plot (Figure 3). The rate of prostaglandin E2 metabolism in the medulla remain unaltered during the entire period of study with control values of 90~20 pmole/mg protein/min. The activity of prostaglandin E2-15-hydroxydehydrogenase as shown in Table 1 has been expressed in terms of 15-keto-prostaglandin E2 formed/mg protein/min. The magnitude of cortical cytosolic 15-hydroxydehydrogenase activity was less than 5% of that noted for 9-ketoreductase. After 3 days of uranyl nitrate administration, there were no significant changes in the activity of this enzyme (Table I).
NOVEMBER
1983 VOL. 26 NO. 5
695
PROSTAGLANDINS
The non-competitive nature of prostaglandin E2-9-ketoreductase inhibition, as suggested earlier, may represent a reduction in the number of enzyme rnolccules (6). These observations agree with those reported earlier by this laboratory in rabbits with ureteral obstruction (6).
1500
I
1000
500
pc
0.05
0 URANYL NITRATE Figure days
4. Prostaglandin following
either
E2 salrne
concentrations or many1
CONTROL in nitrate
renal cortex, administration.
three
The increase in cortical prostaglandin activity after uranyl nitrate administration predicted by the in vitro studies was supported by the actual measurement of the tissue concentratron of prostaglandin E2 (Figure 4). in cortical tissue, 3 days following the Prostaglandin E content administration ot uranyl nitrate, was 1015 + 201 pg/mg tissue, significantly greater than that noted in controls, 196 524 pg/mg (p
696
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS DISCUSS13h! The role of prostaglandins in renal the administration of prostaglandin meclofenamate,
etc.)
in
man
and
disorders synthesis
animals.
has generally been evaluated inhibitors (e.g. indomethacin, For example. the administration
by of
indomethacin in rabbits with reduced renal mass (5) and in patients with systemic lupus erythematosus (2), results in significant reversible decreases Thus, these and previous in renal blood flow and glomerular filtration rate. observations (1) indicated an increased prostaglandin production in renal disease and suggested a protective or compensatory role of vasodilatory arachidonic acid metabolites against the effect of vasoconstrictive factors. However, because of the lack of specific inhibitors and/or activators of prostaglandin metabolism, the role of renal prostaglandin E2 metabolism during functional adaptive changes in disease states or in experimental models is largely unknown. The aim of the present investigation was to dissect out a in the uranyl mechanism for the increased prostaglandin E2 levels and activity nitrate-induced
acute
renal
failure.
Interestingly, and in contrast to our previous observations in the ureteral obstruction model (6), there were no alterations observed in the rate of prostaglandin synthesis during the three day period after the administration of the nephrotoxic agent. However, our data indicate that the metabolism of prostaglandin E2 decreases with the progression of the disease. Thus, it is evident from these data that, at least in vitro, a decrease in the cortical metabolism of prostaglandin E rather than an increase in its synthesis is a major mechanism in elevating T he mtrarenal .. concentration of this lipid as confirmed by direct measurements in cortical tissue. Since, the mechanisms of renal damage in mercury chloride-induced acute renal failure are probably similar to those induced in the uranyl nitrate :nodcl, our findings may explain a prostaglandin synthesis inhibitor, to why the administration of indomethacin, rabbits had no effect on the expression of renal functional abnormalities in this model (8). It has been reported by Sudo et al. (17) that tut<3l renal blood flow fell significantly after 1 day of uranyl nitrate administration and then returned to normal levels after 3 days. In view of our observations of decreased levels (figure 51, it is metabolism resulting in increased prostaglandin E tempting to speculate that the increase in prost&landin E after I day may not have been sufficient to overcome the vasoconstrictive fact&s which resulted in the decrease in renal blood flow. Whereas, after 3 days when the intrarenal concentration of prostaglandin E rose substantially, the increased vasodilatory effects resulting fro?n the greater tissue coilcentrations of prostaglandin E were able to fully antagonize the vasoconstriction resulting in the return o ? the renal blood flow to control levels. The decrease in the rate of prostaglandin E2 metabolism observed in the present study might have been due to a complex formation between the heavy metal and the sulfhydryl groups of the enzyme, prostagldndm E2-Y-ketoreductase. Such an enzyme-inhibitor complex may not be easi!y breakable during the in vitro incubations with the substrate and can result in non-competitive i;nez-of the enzyme inhibition SF seen. Although the kinetics of en/.ylne inhibition fur both uranyl nitrate and ureteral obstruction appears similar, the mcchanisns might be different in that in the ureteral obstruction model, there might be a
NOVEMBER
1983 VOL. 26 NO. 5
697
PROSTAGLANDINS reduction in the number of model, the mechanis,n might
enzyme be
molecules formation of
(6), while in the uranyl nitrate inactive enzyme-heavy metal
complexes. In summary, we have shown that a significant decrease in prostaglandin E metabolism (via the prostaglandin E2-9-ketoreductase pathway) is the majo? mechanism for the increased renal prostaglandin E levels and activity in this nephrotoxic model of renal failure in the rabbit. 2The demonstration of this mechanism in two different models of renal disease appears to be more than a coincidence and suggests that similar alterations may occur in other forms of renal dysfunction. Finally, since we have evaluated only the changes in the synthesis of prostaglandin E and in the activity of two prostaglandin E metabolizing enzymes, we d2 not rule out the importance of other eicosanoids in the pathogenesis and compensatory responses to acute renal injury as well as the importance of changes in these and additional metabolic pathways in other species.
ACKNOWLEDGEMENTS The authors wish to thank Dr. John Pike generous gift of prostaglandin standards. by IJSPHS Grant AM-29667.
and the Upjohn Company This study was supported,
for their in part,
REFERENCES
I.
rhn,
prostaglandin
Renal 1980.
M.J., and E.J. Zambraski. synthesis. Kidney Int l&609,
2. Kimberely, R.P., J.R. Elevated urinary prostaglandins Lupus Erythernatnsus. Ann Int
effects
of
drugs
Gill, R.E. Rowden, H.R. Keiser and the effects of aspirin on Med 89:336, 1978.
3. Suzuki, S., R. Shapiro, P.J. Mulrow prostaglandin E2 excretion in chronic renal
that
inhibit
and P.H. Plotz. renal function in
and S.Y. Tan. Urinary disease. Prostaglandins
Med
4:377,
1980.
Renal
J.C. Rornero, 4. Torres, V.E., prostaglandin E during acute
C.G. renal
Strong, failure.
D.M.
Wilson
and
Prostaglandins
5. Kirschenbaum, on glomerular filtration 22:245, 1981.
M.A. rate
6. Chaudhari, A. metabolism in ureteral
and M.A. Kirschenbaum. Decreased obstruction. Riochim Riophys Acta
V.R. 8:353,
Walker. 1974.
and E.K. Serros. Effect of prostaglansin inhibition in normal and uremic rabbits. Prostaglandins
renal 713:10,
prostaglandin 1982.
7. Effect model
698
Mauk, R.M., R.V. Patak, S.Z. Fadern, 1M.D. Lifschitz and J. Stein. of prostaglandin E administration in a nephrotoxic and a vasoconstrictor of acute renal failure. Kidney Int 12:122, 1977.
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS
8. Torres, V.E., C.G. Strong, J.C. enhancement of glycerol-induced acute 7:170,
Romero and renal failure
D.M. Wilson. Indomethacin in rabbits. Kidney Int
1975.
and P. Needleman. Exaggerated 9. Nishikawa, K., A. Morrison biosynthesis and its influence on renal resistance in the isolated hydronephrotic kidney. J Clin Invest 59:1143, 1977.
assay 1977.
10. Sedmak, for protein
11. Tai, prostaglandins 12. diuretics.
A rapid
J.J. and S.E. Grossberg. using Coomassie Brilliant
H.-H., C.L. Tai and C.S. in rabbit kidney medulla.
Blue
Hollander. Biochem
Inhibition Stone, K.J. and M. Hart. Prostaglandins 12:197, 1976.
of
sensitive
C250.
J
Anal
and
prostaglandin
versatile
Biochem
79:544,
Biosynthesis of 154:257, 1976.
renal
PGE2-9-ketoreductase
by
13. Katzen, D.R., S.-S. Pong, and L. Levine. Distribution of prostaglandin E 9-ketoreductase and NAD+-dependent and NADP+-dependent 15-hydroxyprostaglandin dehydrogenase in the renal cortex and medulla of various species. Res Corn Chem Path Pharmacol 12:781, 1975. 14. 55:170,
Dixon, 1953.
M.
The
determination
of
15. Orczyk, G.P. ar-d H.R. Behrman. indomethacin. In vivo evidence for a role --_ secretion. Prostaglandms 1:3, 1972. 16. Miller, M.J.S., E.G. Spokas, and in the isolated perfused prostaglandin E Pharmacol 31:2?55, 1982.
uranyl
17. Sudo, M., N. acetate-induced
Editor:
ATan
NOVEMBER
Nies
Honda, acute
enzyme
inhibitor
constants.
Ovulation blockade of prostaglandin in
J.C. McGiff. kidney of the
Biochem
by aspirin or gonadotrophin
Metabolism of rabbit. Biochem
A. Hishida and M. Nagase. Renal hemodynamics renal failure of rabbits. Kidney Int 11:35, 1977.
Received:
1983 VOL. 26 NO. 5
Z-9-33
J.
Accepted:
in
9-7-83
699
MECHANIW
OF
INCREASED
RENAL
NITRATE-INDUCED
Anshumali
PROSTAGLANDIN
ACUTE
Chaudhari
and
Division
of
RENAL
Michael
A.
E2
IN
URANYL
FAILURE
Kirschenbaum
Nephrology
Department of Medicine UCLA School of Medicine Los Angeles, California 90024
ABSTRACT We have previously demonstrated that decreased cortical prostaglandin metabolism can contribute significantly to an increase in renal tissue levels in bilateral ureteral obstruction, a model of and activity of prostaglandin E In the present study, we have further investigated acute renal failure. whether alterations in prostaglandin metabolism can occur in a nephrotoxic Prostaglandin synthesis, prostaglandin E2 model of acute renal failure. metabolism (measured as both prostaglandin E2-9-ketoreductase and prostaglandin E2-15-hydroxydehydrogenase activity), and tissue concentration of prostaglandin E2 were determined in rabbit kidneys following an intravenous administration of No changes in the rates of cortical microsomal uranyl nitrate (5 mg/kg). were noted at the end of 1 and prostaglandin E2 and prostaglandin F20: synthesis 3 days, while medullary synthesis of prostaglandin E2 ,fell by 47% after 1 day Cortical cytosolic prostaglandm E2-9-ketoreductase and 43% after 3 days. activity was found to be decreased by 36% and 76% after 1 and 3 days No significant changes were noted in cortical cytosolic respectively. prostaglandin E2-15-hydroxydehydrogenase activity after 3 days. Cortical by 500% at the end of 3 days. tissue levels of prostaglandin E2 increased These data demonstrate that in nephrotoxic acute renal failure, decreased prostaglandin E -9-ketoreductase activity) can prostaglandin metabolism (i.e., result in increased tissue levels of prostaglan c?in E2 in the absence of increased prostaglandin synthesis and suggest that alterations in prostaglandin metabolism may be an important regulator of prostaglandin activity in acute renal failure.
INTRODUCTION Although there is considerable suggestive evidence which might link the various arachidonic acid lnetabolites to the vascular, glomerular and tubular events which occur in acute renal failure (I), there is little direct evidence to Despite the absence of confirmatory experimental support this relationship. evidence, however, there is little doubt that the prostaglandins participate, in some manner, in the numerous homeostatic and pathogenetic mechanisms which accompany an acute disruption of either gloinerular or tubular function. Evidence for the importance of these lipids in acute renal failure has been For example, previous studies have shown both in this and other laboratories. demonstrated increased renal prostaglandin E excretion rates and renal tissue prostaglandin activity in patients with renal disease (2,3) and in animals with experimental models of acute renal failure (4-6). In other studies, the
NOVEMBER
1983 VOL. 26 NO. 5
689
PROSTAGLANDINS
the infusion of vasodilator prostaglandins has been shown to be beneficial in protecting against glomerular injury following the induction of hemodynamically-mediated experimental acute renal failure (7), while the administration of cycle-oxygenase inhibitors under similar conditions has been shown to be detrimental (8). Taken together, these findings suggest that, rather than causing pathologic changes in the nephron, some effect or effects of these lipids may protect the kidney from vasoconstrictive factors which would, if left unopposed, further reduce renal function. The arachidonic acid metabolites which might participate in the homeostatic regulation of renal function (e.g., renal blood flow, glomerular filtration rate and urinary water and electrolyte excretion) can do so in at least two separate and distinct ways: first by some direct action on glomerular, tubular or vascular structures; and second, by modulating the vasoconstrictive influences of other hormones such as angiotensin II and norepinephrine and of sympathetic nervous activity. Since the prostaglandins are primarily local tissue hormones, their biological effects within the kidney would be expected to be dependent, to a great extent, on their cellu!ar concentrations. These intracellular concentrations are the net result of the differences in the rates of prostaglandin synthesis and metabolism. Logically, changes in the rate of metabolism, independent of changes in synthesis could also lead to significant changes in local prostaglandin concentration and thus its activity within the kidney. We recently tested this hypothesis in rabbits and reported that a significant decrease in the renal metabolism of prostaglandin E2 !the major renal prostaglandin in this species) could be demonstrated wrth both uni- and bilateral ureteral obstruction (61, models in which increased prostaglandin biosynthesis has been previously demonstrated (9). The present study was performed to evaluate this hypothesis further in a model of acute renal failure resulting from the administration of a nephrotoxic agent, uranyl nitrate.
METHODS Female white New Zealand rabbits weighing 3-3.5 kg were used in the study. The experimental group (12 animals) received, an intravenous injection of uranyl nitrate at a dose of 5 mg/kg whereas the control group (12 animals) was given an intravenous injection of an equivalent volume of normal saline. Half of the animals in each group were sacrificed on day I and the other half on day 3 following uranyl nitrate administration. Prostaglandin biosynthesis and prostaglandin E2-9-ketoreductase activity was measured in all animals. Eight additional animals were given either uranyl nitrate (4 animals) or normal saline (4 animals) and were sacrificed after 3 days. In this group, prostaglandin E metabolism was assessed by measurement of prostaglandin E2-15-hydroxyde 2, ydrogenase activity. All animals had free access to water and consumed comparable quantities of standard laboratory rabbit chow. Preparation of Subcellular Fractions: For the assay of prostaglandin biosynthesis, cortex and medulla were homogenized separately in 0.02 M Tris buffer (pH 8.0) and centrifuged for 10 min at 9,000 x g. The supernate was removed, further centrifuged for 60 min at 100,000 x g, then discarded. The resultant pellet was homogenized in 0.02 M Tris buffer and adjusted to a
690
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS
protein concentration of 10 mg/ml for the assay. To determine prostaglandin E,2-9-ketoreductase and prostaglandin E2-15-hydroxydehydroenase activities, the ttssue was homogenized in 0.1 M potasstum phosphate buffer (pH 7.3) containing and centrifuged at 100,000 x g for 60 min. 4 mM MgCl and 0.1 mM dithiothreitol The superna ? e was dilute a protein concentration of 5 mg/ml for assay. Protein was determined by the method Protein Estimation: Grossberg (IO) using bovine serum albumin as a standard.
of Sedmak
and
Assay of Prostaglandin Biosynthesis: The method for determining prostaglandin biosynthesis was adapted from that of Tai et al. (II). The reaction mixture I mM ymzp;r;O; f$T;; ;ffiQpH 8.0), 1 mM reduced glutathione, L-labelled arachidonic acid (47 mCi/mmoI, New England Nuclear, Bostkn, MA) and unlabelled arachidonic acid, and microsomal fraction in a final volume of 0.5 ml. Cortical microsomes (2 mg protein) were incubated for 30 min at 37” C and medullary microsomes (I mg protein) for 5 min. The rate of product formation during these periods of incubation was linear under the experimental conditions. The medium was then acidified to pH 3.5 with 1.0 M formic acid. The reaction mixture was extracted twice with 1 ml ethyl acetate, combined and 100 pl ethyl acetate.
evaporated
under
nitrogen,
and
the
residue
dissolved
in
An aliquot was spotted on a plastic silica-gel G thin layer chromatographic plate and the chromatogram developed to a height of 17 cm with organic phase of ethyl acetate/isooctane/acetic acid/water (11:5:2:10) as the solvent system. Authentic standards of prostaglandin E , prostaglandin F thromboxane B and 6-keto-prostaglandin Fbi$i~t;$m~;a, J;~mF’ik;, St. U$$hn Co., Kalamaz$o, MI) and arachidonic acid Louis, MO) were co-chromatographed and the plate was allowed to develop’color in iodine vapor for the localization of arachidonic acid metabolites. In order to separate further prostaglandin E2 fro” thromboxane B2, the plates were first allowed to develop 10 cm from the orlgm and after drying, were again allowed to develop 17 cm from the origin. Each plate was cut into 0.5 cm pieces and placed in counting vials containing ACS@ scintillation cocktail (Anlcrsham was determined in a Beckman Corp., Arlington Heights, IL). The radioactivity LS-7500 liquid scintillation system. Assay of Prostaglandin E -9-ketoreductase: The method for determining prostaglandin E2-9-ketore 8 uctase activity was adapted from that of Stone and Hart (12). The reaction mixture consisted of 0.1 Mfeotassium phosphate buffer containing 4 mM IM~CI, and 0.1 mM dithiothreitol, C-labelled orostaelandin EL L (56 mCi/mol, Amersuham Corp., Arlington Heights, fL) and unlabelied u prostaglandin E2 (final concentration of 46 PM), 10 mM glucose-6-phosphate, 1 unit of glucose-6-phosphate dehydrogenase, 0.1 mM NADPH and a suitable aliquot of cytosolic fraction (200 ug protein) of cortex or medulla in a total volume of 0.5 ml. The reaction mixture was incubated for 40 min at 37” C, acidified with formic acid, and extracted twice with ethyl acetate as described above. The formation of prostaglandin FZa, ,the major prostaglandin E2 metabolite in the rabbit kidney (12,131 was determmed by thin layer chromatography using ethyl acetate/acetone/acetic acid (90: 10: 1) as the solvent system. Assay of Prostaglandin E2-15-hydroxydehydrogenase: This method was also adapted from that of Stone and Hart (12). The reaction mixture for this assay was siTilar to that described above except for the addition of 0.5 mM NADf and a NAD
-regenerating
NOVEMBER
system
which
1983 VOL. 26 NO. 5
consisted
of
3.2
mM
sodium
pyruvate
and
3
691
PROSTAGLANDINS
units
of
lactate
dehydrogenase
listed above. The mixture the reaction were extracted thin layer chromatography phase
of
isooctane:ethyl
instead
of
the
NADPH
and
its
regenerating
system
was incubated for 60 min at 37” C, the product\ with ethyl acetate and analyzed by silicic acid using a solvent system consisting of the organic
acetate:acetic
acid:water
of
(5:11:2:10).
Control
Day
1 after
uranyl
nitrate
Day
3
uranyl
nitrate
after
T I
PGF2ct
PGE2
Figure
1.
Prostaglandin
E2 and Fzoc biosynthesis
by
cortical
microsomes.
Kinetic
analysis was performed using a Lineweaver-Burk plot as modified by For this purpose, prostaglandin E2-,9-ketoreductase activity was Dixon (14). determined at various substrate concentrations m renal cortical cytosolic fractions obtained from both controls and uranyl nitrate treated animals and double reciprocal plot was constructed. Radioreceptor Assay for Prostaglandin E: tissuewere determin.ed by a radioreceptor previously by this laboratory (5). Cortical
692
a
Prostaglandin E levels in cortical assay which has been described tissue was rapidly dissected,
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS
weighed
and
homogenized
for
3rczyk and Rehrman (15). exceed 10 seconds following 25-50 mg of cortical tissue
the
The
extraction
time
of
required
prostaglandins for
by
dissection
and
the
method
weighing
of
did
not
the surgical exposure of the kidneys. In brief, was homogenized at 4” C using a PolytronB consisting of I3 ml of 0.9% NaCl and 0.4 ml of 0.1
N homogenizer in a solution About 2000 counts per minute of H-prostaglandin E2 was added in order to HCI. monitor recovery. This homogenate was extracted with 2.6 ml of a mixture of ethyl acetate:isopropanol (I:l, v:v). The organic phase was isolated after the addition of 3 ml of normal saline and 2 ml of ethyl acetate. After vortexing and a IO minute centrifugation at 3000 rpm, the organic phase was removed and E fraction was collected from a Sephadex dried under N . The prostaglandin LH-20 column2by using acetate:isooctane:acetic the assay.
a solvent acid:water
sys2iem consisting (11:5:2:10) and
u
of the organic phase dried again under N2
of ethyl before
Control
li!al Day
1 after
uranyl
nitrate
Day
3
uranyl
nitrate
after
1000
800 l
PC 0.025
600 *
*
400
TT 40
2c
0 PGE2
Figure
2.
NOVEMBER
Prostaglandin
E2 and Fzrx biosynthesis
1983VOL. 26 NO.5
PGP20:
by t-redullary microsoms.
693
PROSTAGLANDINS
Statistical significance
Analysis: Data are expressed was evaluated by non-paired
as the mean + SE and statistical Student’s “t” test.
RESULTS IMean serum creatinine levels in the experimental group sacrifice were 2.68 + 0.19 mg/dl and 10.0 _+ 0.14 mg/dl to 1.0 _+ 0.1 mg/dl in controls.
TABLE
1. Prostaglandin
E2-9-Ketoreductase
15-Hydroxydehydrogenase
in Renal
Group
pmole
uranyl
3 days post
*p < 0.01, ND
= not
uranyl
**
Cortical
15-Hydroxydehydrogcnase
PGFZX formed/
pmole
I5-keto-PGE2
mg protein/min
mg protein/min
735 + 60
33 -+ 5
Control 1 day post
E2-
Fractions.
9-Ketoreductase Study
on the day of as compared
and Prostaglandin
Activity
Cytosolic
of animals respectively
nitrate nitrate
p < 0.005 as cornpared determined.
470 + 32* xx 175 _+ 45 to control
formed/
ND 34 -+ 5 value.
results of --in vitro cortical and medullary microsomal biosynthesis of administration are shown prostaglandin Ez and PGF20: following uranyl nitrate Unlike our previous observation in the bilateral ureteral Figures 1 and obstruction model of acute renal failure (6), the cortical synthesis of both of these prostaglandins was unaltered at days 1 and 3. However, the medullary prostaglandin E2 biosynthesis in uranyl nitrate treated animals fell by almost 47% after 1 day and 43% after 3 days compared to respective controls. The decrease is consistent with that seen in ureteral obstruction (6). No changes in prostaglandin F20: synthesis were noted. The
in
The major pathway for the inactivation of prostaglandin E2 in the rabbit kidney via prostaglandin involves its transformation to prostaglandin F as described E2-ketoreductase (12,13,16). The changes in ?f IS enzyme activity, earher, have been used as an index of changes in the rate of prostaglandin E2 metabolism (6). Although prostaglandin E2 has previously bpen shown to be metaboliszed by prostaglandin E2-1.5-hydroxydehydrogenase in the rabbit kidney, this pathway appears to be minor m this animal species (12,13,16).
694
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS
VELOCITY -’ PGFZa Synthesized
3
(rmoVmg protein/mh)
r
-IO
-5
0
SUBSTRATE
I 5
1 IO
I
I
7
I5 20 25 CONCENTRATION -’ PGE,
(mM_l)
Figure 3. Lineweaver-Burk plot of prostaglandin E2-P-ketoreductase. Cortical enzyme activity for control animals is shown in the broken line with squares and that for the uranyl nitrate-treated animals is shown in the solid line with circles.
Nevertheless, we have examined the activity of both enzyme systems in this study and the data showing the effect of uranyl nitrate on cortical cytosolic prostaglandin E2 metabolism are summarized in Table 1. A time-dependent decrease in the rate of cortical prostaglandin E2-9-ketoreductase activity was observed as indicated by a 37% inhibition of enzyme activity after 1 day, and 77% after 3 days. The nature of the enzyme inhibition was found to be non-compeitive as determined by a Lineweaver-Burk plot (Figure 3). The rate of prostaglandin E2 metabolism in the medulla remain unaltered during the entire period of study with control values of 90~20 pmole/mg protein/min. The activity of prostaglandin E2-15-hydroxydehydrogenase as shown in Table 1 has been expressed in terms of 15-keto-prostaglandin E2 formed/mg protein/min. The magnitude of cortical cytosolic 15-hydroxydehydrogenase activity was less than 5% of that noted for 9-ketoreductase. After 3 days of uranyl nitrate administration, there were no significant changes in the activity of this enzyme (Table I).
NOVEMBER
1983 VOL. 26 NO. 5
695
PROSTAGLANDINS
The non-competitive nature of prostaglandin E2-9-ketoreductase inhibition, as suggested earlier, may represent a reduction in the number of enzyme rnolccules (6). These observations agree with those reported earlier by this laboratory in rabbits with ureteral obstruction (6).
1500
I
1000
500
pc
0.05
0 URANYL NITRATE Figure days
4. Prostaglandin following
either
E2 salrne
concentrations or many1
CONTROL in nitrate
renal cortex, administration.
three
The increase in cortical prostaglandin activity after uranyl nitrate administration predicted by the in vitro studies was supported by the actual measurement of the tissue concentratron of prostaglandin E2 (Figure 4). in cortical tissue, 3 days following the Prostaglandin E content administration ot uranyl nitrate, was 1015 + 201 pg/mg tissue, significantly greater than that noted in controls, 196 524 pg/mg (p
696
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS DISCUSS13h! The role of prostaglandins in renal the administration of prostaglandin meclofenamate,
etc.)
in
man
and
disorders synthesis
animals.
has generally been evaluated inhibitors (e.g. indomethacin, For example. the administration
by of
indomethacin in rabbits with reduced renal mass (5) and in patients with systemic lupus erythematosus (2), results in significant reversible decreases Thus, these and previous in renal blood flow and glomerular filtration rate. observations (1) indicated an increased prostaglandin production in renal disease and suggested a protective or compensatory role of vasodilatory arachidonic acid metabolites against the effect of vasoconstrictive factors. However, because of the lack of specific inhibitors and/or activators of prostaglandin metabolism, the role of renal prostaglandin E2 metabolism during functional adaptive changes in disease states or in experimental models is largely unknown. The aim of the present investigation was to dissect out a in the uranyl mechanism for the increased prostaglandin E2 levels and activity nitrate-induced
acute
renal
failure.
Interestingly, and in contrast to our previous observations in the ureteral obstruction model (6), there were no alterations observed in the rate of prostaglandin synthesis during the three day period after the administration of the nephrotoxic agent. However, our data indicate that the metabolism of prostaglandin E2 decreases with the progression of the disease. Thus, it is evident from these data that, at least in vitro, a decrease in the cortical metabolism of prostaglandin E rather than an increase in its synthesis is a major mechanism in elevating T he mtrarenal .. concentration of this lipid as confirmed by direct measurements in cortical tissue. Since, the mechanisms of renal damage in mercury chloride-induced acute renal failure are probably similar to those induced in the uranyl nitrate :nodcl, our findings may explain a prostaglandin synthesis inhibitor, to why the administration of indomethacin, rabbits had no effect on the expression of renal functional abnormalities in this model (8). It has been reported by Sudo et al. (17) that tut<3l renal blood flow fell significantly after 1 day of uranyl nitrate administration and then returned to normal levels after 3 days. In view of our observations of decreased levels (figure 51, it is metabolism resulting in increased prostaglandin E tempting to speculate that the increase in prost&landin E after I day may not have been sufficient to overcome the vasoconstrictive fact&s which resulted in the decrease in renal blood flow. Whereas, after 3 days when the intrarenal concentration of prostaglandin E rose substantially, the increased vasodilatory effects resulting fro?n the greater tissue coilcentrations of prostaglandin E were able to fully antagonize the vasoconstriction resulting in the return o ? the renal blood flow to control levels. The decrease in the rate of prostaglandin E2 metabolism observed in the present study might have been due to a complex formation between the heavy metal and the sulfhydryl groups of the enzyme, prostagldndm E2-Y-ketoreductase. Such an enzyme-inhibitor complex may not be easi!y breakable during the in vitro incubations with the substrate and can result in non-competitive i;nez-of the enzyme inhibition SF seen. Although the kinetics of en/.ylne inhibition fur both uranyl nitrate and ureteral obstruction appears similar, the mcchanisns might be different in that in the ureteral obstruction model, there might be a
NOVEMBER
1983 VOL. 26 NO. 5
697
PROSTAGLANDINS reduction in the number of model, the mechanis,n might
enzyme be
molecules formation of
(6), while in the uranyl nitrate inactive enzyme-heavy metal
complexes. In summary, we have shown that a significant decrease in prostaglandin E metabolism (via the prostaglandin E2-9-ketoreductase pathway) is the majo? mechanism for the increased renal prostaglandin E levels and activity in this nephrotoxic model of renal failure in the rabbit. 2The demonstration of this mechanism in two different models of renal disease appears to be more than a coincidence and suggests that similar alterations may occur in other forms of renal dysfunction. Finally, since we have evaluated only the changes in the synthesis of prostaglandin E and in the activity of two prostaglandin E metabolizing enzymes, we d2 not rule out the importance of other eicosanoids in the pathogenesis and compensatory responses to acute renal injury as well as the importance of changes in these and additional metabolic pathways in other species.
ACKNOWLEDGEMENTS The authors wish to thank Dr. John Pike generous gift of prostaglandin standards. by IJSPHS Grant AM-29667.
and the Upjohn Company This study was supported,
for their in part,
REFERENCES
I.
rhn,
prostaglandin
Renal 1980.
M.J., and E.J. Zambraski. synthesis. Kidney Int l&609,
2. Kimberely, R.P., J.R. Elevated urinary prostaglandins Lupus Erythernatnsus. Ann Int
effects
of
drugs
Gill, R.E. Rowden, H.R. Keiser and the effects of aspirin on Med 89:336, 1978.
3. Suzuki, S., R. Shapiro, P.J. Mulrow prostaglandin E2 excretion in chronic renal
that
inhibit
and P.H. Plotz. renal function in
and S.Y. Tan. Urinary disease. Prostaglandins
Med
4:377,
1980.
Renal
J.C. Rornero, 4. Torres, V.E., prostaglandin E during acute
C.G. renal
Strong, failure.
D.M.
Wilson
and
Prostaglandins
5. Kirschenbaum, on glomerular filtration 22:245, 1981.
M.A. rate
6. Chaudhari, A. metabolism in ureteral
and M.A. Kirschenbaum. Decreased obstruction. Riochim Riophys Acta
V.R. 8:353,
Walker. 1974.
and E.K. Serros. Effect of prostaglansin inhibition in normal and uremic rabbits. Prostaglandins
renal 713:10,
prostaglandin 1982.
7. Effect model
698
Mauk, R.M., R.V. Patak, S.Z. Fadern, 1M.D. Lifschitz and J. Stein. of prostaglandin E administration in a nephrotoxic and a vasoconstrictor of acute renal failure. Kidney Int 12:122, 1977.
NOVEMBER
1983 VOL. 26 NO. 5
PROSTAGLANDINS
8. Torres, V.E., C.G. Strong, J.C. enhancement of glycerol-induced acute 7:170,
Romero and renal failure
D.M. Wilson. Indomethacin in rabbits. Kidney Int
1975.
and P. Needleman. Exaggerated 9. Nishikawa, K., A. Morrison biosynthesis and its influence on renal resistance in the isolated hydronephrotic kidney. J Clin Invest 59:1143, 1977.
assay 1977.
10. Sedmak, for protein
11. Tai, prostaglandins 12. diuretics.
A rapid
J.J. and S.E. Grossberg. using Coomassie Brilliant
H.-H., C.L. Tai and C.S. in rabbit kidney medulla.
Blue
Hollander. Biochem
Inhibition Stone, K.J. and M. Hart. Prostaglandins 12:197, 1976.
of
sensitive
C250.
J
Anal
and
prostaglandin
versatile
Biochem
79:544,
Biosynthesis of 154:257, 1976.
renal
PGE2-9-ketoreductase
by
13. Katzen, D.R., S.-S. Pong, and L. Levine. Distribution of prostaglandin E 9-ketoreductase and NAD+-dependent and NADP+-dependent 15-hydroxyprostaglandin dehydrogenase in the renal cortex and medulla of various species. Res Corn Chem Path Pharmacol 12:781, 1975. 14. 55:170,
Dixon, 1953.
M.
The
determination
of
15. Orczyk, G.P. ar-d H.R. Behrman. indomethacin. In vivo evidence for a role --_ secretion. Prostaglandms 1:3, 1972. 16. Miller, M.J.S., E.G. Spokas, and in the isolated perfused prostaglandin E Pharmacol 31:2?55, 1982.
uranyl
17. Sudo, M., N. acetate-induced
Editor:
ATan
NOVEMBER
Nies
Honda, acute
enzyme
inhibitor
constants.
Ovulation blockade of prostaglandin in
J.C. McGiff. kidney of the
Biochem
by aspirin or gonadotrophin
Metabolism of rabbit. Biochem
A. Hishida and M. Nagase. Renal hemodynamics renal failure of rabbits. Kidney Int 11:35, 1977.
Received:
1983 VOL. 26 NO. 5
Z-9-33
J.
Accepted:
in
9-7-83
699