Toxicology, 60 (1990) 211-222 Elsevier Scientific Publishers Ireland Ltd.
Inhibition of catalase and epoxide hydrolase by the renal cystogen 2-amino-4,5-diphenylthiazole and its metabolites J. Thomas Hjelle "'*, Thomas M. Guenthner b, Kevin Bell a, R o b e r t Whalen b, G e o r g e Flouret c and Frank A. C a r o n e d Departments of Basic Sciences" and Pharmacology b, University of Illinois College of Medicine at Peoria and Chicago, IL 61605 and Departments of Physiology' and Pathology, Northwestern University Medical School, Chicago, IL 60611 (U.S.A.) (Received June 7th, 1989; accepted October 20th, 1989)
Summary Subchronic feeding of 2-amino-4,5-dipbenylthiazole (DPT) to rats results in the development of renal cysts and has been used as a model system to study polycystic kidney disease. Because previous studies revealed changes in renal enzymes following DPT administration, a possible direct effect of DPT and its phenolic metabolites on catalase and a related enzyme, epoxide hydrolase, was examined. Experiments with three in vitro systems (suspensions of rabbit renal tubules, rat kidney homogenates, and commercially obtained bovine liver catalase) revealed direct inhibition of catalase activity by the diphenolic metabolite (diOH- DPT: 2-amino-4,5di(4'-hydroxyphenyl)-thiazole), the known renal cystogen nordihydroquaiaretic acid (NDGA) 2-amino-4(4'- hydroxyphenyl),5-phenyl-thiazole (4OHDPT), and the known catalase inhibitor 3-amino-l,2,4-triazole; DPT did not inhibit catalase activity. Following oral administration to rats of the DPT congeners, 4OH-DPT caused the greatest decrease in both renal catalase and cytosolic epoxide hydrolase activities and the shortest time to onset of cystic lesions. In vitro, mouse liver cytosolic epoxide hydrolase activity was substantially inhibited by 4OH-DPT and dioH-DPT, and NDGA, but not by 2-amino-4-phenyl,5- (4'-hydroxyphenyl)-thiazole (5OH-DPT) or DPT itself. Microsomal epoxide hydrolase (mEH) activity was inhibited by 4OHDPT, unaffected by DPT or dioH-DPT, and stimulated 2-fold by 5OH- DPT. Finally, mEH activity was substantially higher in samples of normal human kidney than in samples of kidney derived from a patient with autosomal recessive polycystic kidney disease; no differences were observed in cEH activity in these samples. Although the role of altered catalase and epoxide hydrolase activities in cystogenesis is unknown, DPT-induced cyst formation is associated with loss of these enzyme activities in kidney tissue. To our knowledge, this is the first report of an in vivo diminution of cytosolic epoxide hydrolase activity by xenobiotics.
* To whom reprint requests" should be addressed: Department of Basic Sciences, University of Illinois College of Medicine at Peoria, Box 1649, Peoria, IL 61656, U.S.A. Abbreviations: DPT, 2-amino-4,5-diphenyl-l,3-thiazole; NDGA, nordihydroquaiaretic acid; PKD, polycystic kidney disease; DMSO, dimethylsulfoxide; cEH, cytosolic epoxide hydrolase; mEH, microsomal epoxide hydrolase; EET, epoxyeicosatetraenoic acids. 0300-483X / 90 / $03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
211
Key words: Renal cystogenesis;Catalase; Epoxide hydrolase; Diphenylthiazole
Introduction
The possibility that renal cystic disease may have a toxicological component is undergoing renewed scrutiny in light of several recent findings. Firstly, administration of chemicals such as diphenylamine, nordihydroguaiaretic acid (NDGA) and 2-amino-4,5-diphenylthiazole (DPT), is known to cause renal cysts in otherwise normal rats [1,2]. Secondly, an acquired form of polycystic kidney disease is appearing at a significant rate in patients undergoing long term renal dialysis [3,4]. Thirdly, epidemiological data support the possibility that the incidence of either renal cysts or polycystic kidney disease (PKD) may have increased over the interval between 1900 and the late 1970's [5]. Unfortunately, very little is known of the biochemical basis for renal cystogenesis. In this report, we investigate the hypothesis that some chemical cystogens may act to disrupt cellular enzymes that protect cells from injury by exogenous or internally produced toxic species. Two such enzymes whose activities are known to be significantly affected by cystogens are catalase and CEH. The linkage of catalase and renal cystogenesis has recently been discussed by Gardner et al. [6], Based in part on the observation of Tappel and Marr [7] that NDGA, then under consideration as a food additive and now known to be a renal cystogen, could inhibit catalase, Gardner et al. [6] have proposed that catalase inhibition in the presence of oxidative stress caused by neutrophil invasion might cause cell death and resultant hyperplasia leading to cyst formation. We have recently reported results of a survey of various enzyme activities in kidney and liver tissues from rats fed the renal cystogen, DPT, subchronically [8]. DPTinduced renal cyst formation was accompanied by numerous changes in kidney enzyme activities, one of which was a diminution of catalase activity. Margoliash et al. [9] reported the minimal structural requirement for the inhibition of catalase to be > N - N H - C ( N H ) - R - H , where R could be sulfur, oxygen, or nitrogen. Included in this group of catalase inhibitors are compounds such as aminoguanidine, semicarbazide, thiosemicarbazide, diaminobenzidine, p-phenylenediamine, and several triazole complexes [9,10]. DPT contains this structural element. Because DPT decreases catalase activity in vivo, and because DPT is structurally similar to known catalase inhibitors, we examined in this study the possibility that DPT or its hydroxylated metabolites could directly act on catalase to inhibit its activity. Cross-reactivity between antiserum to highly purified cytosolic epoxide hydrolase (cEH), and catalase has been observed [11]. A structural relationship between catalase and cEH is further suggested by the observation that the enzymes co-purify by several techniques [11]. Existence of common structural features in both catalase and cEH led us to also investigate whether DPT and its metabolites could inhibit cEH activity. The toxicological significance of cEH has
212
been previously discussed in light of its ability to hydrolyze xenobiotic epoxides. This enzyme also efficiently converts epoxy-eicosatrienoic acids (EETs) to corresponding diol metabolites ]12]. E E T s are known to affect a number of important physiological processes possibly related to cystogenesis [13-15]. Therefore, the effects of D P T and its cystogenic metabolites on these two cytoprotective enzymes, catalase and cEH, are examined, in an attempt to link their abilities to inhibit these enzymes with their abilities to induce renal cyst formation by as yet undisclosed mechanisms. Materials and m e t h o d s
All chemicals used were obtained from the Sigma Chemical Co. (St. Louis, MO), with the exception of D P T and its phenolic analogs which were prepared as previously reported [16]. Chemical structures are given in Fig. 1. Catalase activity was determined by the method of Baudhuid et al. [17] where total reaction volumes of 1.5 ml were used; for each assay, three time points were used to calculate a rate of reaction. Protein was measured by the fluorescamine method of Bohlen et al. [18]. Statistical analyses were performed using the Student's t-test. Suspensions of rabbit renal proximal tubules were isolated by the method of Hjelle et al. [19] in Medium 199 (Sigma Chemical Co., St. Louis, MO). Tubules were incubated in oxygenated M199 medium at pH 7.4 containing the test chemicals for up to 90 min. At various times, aliquots of the tubule suspension were removed and centrifuged for 1 min at 12 000 × g; the resulting pellet was resuspended in 50 mM potassium phosphate buffer at pH 7.2, and sonicated for 30 s. The tubule homogenate was assayed for catalase activity and protein.
R1~
S~ ,/NH2
~ '
/)- z_/
:/
/
R2
Fig. 1. Chemical structures of DPT derivatives used in this study. R1
R2
H H OH OH
H OH H OH
Chemicalname 2-amino-4,5-diphenyl-1,3-thiazole 2-amino-4(4'-hydroxylphenyl),5-phenyl-1,3-thiazole 2-amino-4-phenyl,5-(4'-hydroxyphenyl)-1,3-thiazole 2-amino-4,5-di(4'-hydroxyphenyl)- 1,3-thiazole
Abbreviations DPT 4OH-DPT 5OH-DPT diOH-DPT
213
Rat kidney homogenates were prepared from Sprague-Dawley rats killed by cervical dislocation followed by decapitation. Kidney sections were removed, weighed and suspended (1 : 10, w/v) in 0.25 M sucrose containing lmM EDTA at pH 7.4. Tissues were homogenized in a two-step process involving Dounce homogenization in either Tyrode's buffer or 50 mM potassium phosphate buffer followed by sonication for 1 min. The resulting homogenate was centrifuged at 1000 x g for 10 min. The supernate was then assayed for protein and catalase activity; in some cases the homogenate was frozen at -60°C for 1 week before assays were conducted. Experiments with purified enzyme utilized crystalline bovine liver catalase (0.15 mg/ml) dissolved in 50 mM potassium phosphate buffer. The test chemicals were added in various concentrations to the catalase-buffer system and incubated for up to 1 hour at 37°C. As noted, the test chemicals were dissolved in dimethylsulfoxide (DMSO) resulting in a final DMSO concentration of 20% in the incubation tube; 20% DMSO caused about a 10% reduction in the catalytic activity of the enzyme. For some experiments, hydrogen peroxide (0.007% final concentration) was also added to the incubation mixture. Total assay volume was 1.4 ml. cEH activity was measured in cytosolic fractions prepared from the livers of adult male Swiss Webster mice. These cytosolic fractions were prepared by standard methods [20]. Enzyme activity was measured by the radiometric assay described by Guenthner [21], using tritiated trans-stilbene oxide as the substrate. Inhibitors were added in 5/~1 of DMSO, and 5/xl of DMSO was added to control (uninhibited) incubations. Microsomal epoxide hydrolase activity was measured in hepatic microsomes prepared [20] from adult male Swiss-Webster mice. Enzyme activity was determined by standard methods, using styrene oxide as substrate [22]. In the whole animal experiments, DPT and its congeners were given by gavage in daily doses of 1.2 mg chemical/g body weight for 4 days to male SpragueDawley rats weighing 200-250 g [16]. Each dose consisted of the pulverized chemical suspended in 1 ml of 1% gum-tragacenth; controls received 1 ml of the carrier. On the fifth day, the rats were killed and kidney sections taken for histology and biochemical analyses. Tissues to be used for enzyme determinations were rapidly frozen in liquid nitrogen and later thawed, weighed, and homogenized in 0.25 M sucrose containing 1 mM EDTA. The homogenates were assayed for enzyme activities as described above. Human kidney (from a young adult) that was unuseable for transplantation was the source of normal tissue. Kidney tissue from an infant with autosomal recessive polycystic kidney disease obtained during nephrectomy was the source of cystic tissue. Morphologically, the cystic kidney was spongiform with cysts throughout the medulla and cortex. The cyst walls were lined with epithelial cells having an ultrastructure consistent with a collecting tubule origin. Cystic renal epithelial cells isolated and propagated in vitro as described elsewhere demonstrated lectin staining consistent for collecting tubules [23]. Sections of kidney were frozen for subsequent homogenization and analysis for protein and epoxide hydrolase activities as described above.
214
Results
The ability of DPT and its metabolites to inhibit catalase was examined in three in vitro systems. Incubation of freshly isolated rabbit renal proximal tubules with the test chemicals revealed that the diphenolic metabolite, diOH-DPT, exhibited the greatest inhibition of catalase activity of the DPT congeners tested (Table I). In these catalase-rich cells, the known catalase inhibitor, aminotriazole, was effective in decreasing catalase activity. Several of the inhibitors were poorly soluble in the buffer system employed with the intact kidney cells. Therefore, the test chemicals were dissolved in DMSO and rat renal homogenates were used as a source of catalase. Viability of the proximal tubules was substantially depressed by DMSO. Figure 2 shows the inhibition of catalase activity present in kidney homogenates achieved by incubation of DPT metabolites for 30 rain; of these derivatives, diOH-DPT was the most effective in diminishing catalase activity. Inhibition in this system was not significantly enhanced by the addition of hydrogen peroxide. Of note is the finding that another potent renal cystogen, NDGA, also inhibited the catalase activity present in kidney homogenates with a potency similar to that of the diphenolic metabolite of DPT. DPT itself exhibited little inhibition of catalase activity. Inhibition of catalase activity by all DPT derivatives was time and concentration dependent, as shown for the most potent DPT derivative, diOHDPT (Fig. 3). Inhibition of purified bovine liver catalase activity by 4OH DPT and diOHDPT was enhanced by the In vitro incubation of DPT congener and enzyme in the presence of hydrogen peroxide prior to the assay of catalase activity (Fig. 4). In agreement with the findings of Margoliash et al. [9], we also observed that aminotriazole was substantially less active in the absence of hydrogen peroxide in this system (data not shown) This contrasts with our results in kidney homogenates where hydrogen peroxide did not appreciably augment inhibition of catalase by the test compounds. Nevertheless, 4OH-DPT and diOHDPT did cause a
TABLE I LOSS OF C A T A L A S E A C T I V I T Y BY S U S P E N S I O N S O F R A B B I T R E N A L T U B U L E S I N C U B A T E D W I T H DPT D E R I V A T I V E S IN V I T R O Treatment
Conc (mM)
N
mU ni t s catalase a c t i v i t y / m g tubule protein a
Control 4OH-DPT 5OH-DPT diOH-DPT Aminotriazole Aminotriazole
--
4 3 3 3 3 3
260 -+ 12 216 -+ 8 ** 236 -+ 110 -+ 6 * 118 _+ 4 * 9 -+ 1 *
(/.37 0.37 0.37 1.19 119.00
~(Mean -+S.E.) after 90 min of incubation at 37°C. Significantly different from controls, *P < 0.01 and **P < 0.05.
215
100
"~ 8O m~
v
60
',, ,%~
h= (9
\',, \\
tO
\\
~
20
0
' 0.01
' 0.1
' 1
"~ 10
Log Cystogen Concentration
(mM)
Fig. 2. Inhibition of catalase activity present in rat kidney homogenates by putative renal cystogens. Chemicals dissolved in DMSO were incubated 30 rain with homogenate and then assayed for catalase activity. DiOH-DPT (--O---), NDGA ( ~ ) , 4OH-DPT (--A--) and 5OH-DPT (--O--).
100 90 ,m
80 70
60 o0 tD
50 40
o
o~
30 20 10 0
I
0.01
!
f
I
I
I
0.1
I
I
i
I
I
1
I
!
I
I
10
LOG (4,5 HO-DPT)mM Fig. 3. Time and concentration dependence of catalase inhibition by the most potent of the DPT derivatives, diOH-DPT. Loss of catalase activity after 1 rain ( I ) and 30 mins (©) of incubation of increasing concentrations of diOH-DPT with rat kidney homogenate.
216
©
400 '
O
o
NO
~
\\
__>.>
"6
"300
it= m m O (/J I--
200 E LOG
(4 - HO - DPT raM)
0.01
01
I 0
IOmM
0.01
0.1
1.0
1
[
I
10 mM l
LOG 4,5-diHO-DPT concentration (raM)
Fig. 4. Effect of hydrogen peroxide on the inhibition of catalase activity by DPT derivatives. Commercial preparations of bovine liver catalase were incubated with inhibitor for 30 rains in the presence ( + ) or absence (--Z3---)of 0.007% hydrogen peroxide.
concentration dependent inhibition of catalase activity as monitored by the disappearance of hydrogen peroxide. The possibility that D P T and its congeners may also inhibit c E H , an enzyme that is somewhat structurally related to catalase, was examined using subcellular fractions of mouse liver. Table II shows the inhibition of c E H activity, using trans-stilbene oxide as the the substrate, by D P T and its hydroxylated metabolites, as well as by N D G A . As was the case for catalase, D P T itself produced no inhibition. 5 O H - D P T produced a small amount of inhibition that was statistically significant ( P < 0.02). Both 4 O H - D P T and d i O H - D P T produced highly significant inhibition of c E H activity, with the diphenolic derivative being slightly more effective at equivalent concentrations. N D G A also inhibited the enzyme, to a degree similar to that produced by 4 O H - D P T or 4,5-diOH-DPT. Inhibitor concentrations were chosen to correlate with those used in the catalase experiments, namely 1 m M for D P T and its hydroxylated derivatives, and 2 4 0 / , M for NDGA. As shown in Table II, a second form of epoxide hydrolase located in the microsomal fraction of the cell was inhibited by 4 O H - D P T , stimulated by 5 O H - D P T , and unaffected by the other D P T congeners and N D G A in vitro. Examination of various enzyme activities in kidney homogenates obtained from rats treated with individual congeners of D P T (Table I I I ) revealed that the most cystogenic congener [16], 4 O H - D P T , also caused the greatest depression of catalase and c E H activities in vivo. It is of interest to note that the rank order of
217
TABLE
II
INHIBITION OF CYTOSOLIC BY PUTATIVE CYSTOGENS
AND
MICROSOMAL
Treatment Control DPT 4OH-DPT 5OH-DPT diOH-DPT NDGA
FORMS
OF EPOXIDE
HYDROLASE
cEH activity :''b
mEH activity ;'b
(%)
(%)
100 +- 2.5 99.5 + 1.2 3 9 . 7 _+ 1 .(1 911.8 ± 2.1 ** 3 2 . 9 +- 1.8* 33.11 ± 1.4"
101t + 113 + 40 + 224 ± 102 + 116 +
"Results expressed as percent of control activity; (mean ± S . E . ) ( N = 9). ~'Mouse cytosol assayed with trans-stilbene o x i d e . ~ M o u s e microsomes assayed with styrene oxide significantly different from * * P < 0.02,
4.3 3.3* 8.3* 2.6 3.3
control at * P < 0.001 and
inhibition of cytosolic epoxide hydrolase, 4OH-DPT > diOH-DPT, is similar both in vitro and in vivo. In contrast, catalase is inhibited by diOH-DPT in vitro but not in vivo. The activity of microsomal epoxide hydrolase (mEH) was significantly elevated in DPT-treated rats; mEH activity for control and DPT treated groups was 143.7 +_6.2 (n = 12) and 205 +- 15 (n = 6) pmol styrene oxide hydrolyzed/min/ mg tissue protein (Mean -+S.E.), respectively. The other DPT congeners were without effect on mEH activity. Table IV lists the epoxide hydrolase activities found in cytosolic and microsomal fractions prepared from normal and autosomal recessive PKD human kidneys. Although cEH exhibited similar specific activity in both kidneys, mEH activity was almost tenfold lower in the cystic kidney than in the normal kidney. Cells
TABLE
III
EFFECTS OF ACUTE ORAL ADMINISTRATION OF PUTATIVE RENAL CATALASE AND EPOXIDE HYDROLASE ACTIVITIES
Treatment
N
Control
12 12 8 8 10
CYSTOGENS
N
Catalase activity % control activity ....
100 48.7 93.4 73.1 98.2
8 12 8 8 111
100 74 104 98 86
± -+ -+ + ±
10.3% 3.1%* 16.8% 8.6% 10.3%
.
a
b
± ± ± ± +
6.5% 4.8%* 3.8% 4.8% 4.3%
a(Treatment/control) x 11)0%; ( M e a n -+S.E.). activity = 8 . 5 7 pmol trans-stilbene oxide hydrolyzed/min/mg tissue protein. CControl catalase activity = 4 0 . 7 mUnits/mg tissue protein, *Statistically significant from control P < 0 . 0 1 .
bControl cEH
218
ON
Cytosolic epoxide hydrolase activity % control act~wty " •
4OH-DPT 5OH-DPT diOH-DPT DPT
RENAL
T A B L E IV E P O X I D E H Y D R O L A S E A C T I V I T Y " IN S U B C E L L U L A R F R A C T I O N S O F N O R M A L A N D POLYCYSTIC HUMAN KIDNEY Subcellular fraction Microsomes Cytosol
Normal kidney
A R P K D b kidney
mEH
cEH
mEH
cEH
377 --+35 nt
nt 14 -+ 1
3(1 _+ 17" nt
9.1 13.1
'iActivity is expressed as picomole substrate h y d r o l y z e d / m i n / m g protein. Mean -+S.E. (n : 3 or 4). Single n u m b e r s indicate the m e a n of two determinations. m E H and cEH activities defined as the hydrolysis of styrene oxide and trans-stilbene oxide, respectively. hARPKD: autosomal recessive polycystic kidney disease. *Statistically different from normal kidney value ( P < 0.01).
isolated from the cystic kidney and then propagated in vitro retained the ultrastructure of the epithelial cells that lined the cyst walls, which by electron microscopy and lectin staining appeared to be collecting tubule in origin [23]. Discussion
In this study, we have shown that hydroxylated metabolites of DPT inhibit catalase activity in renal cells, kidney homogenates, and pure preparations of catalase. The 4OH-DPT and diOH-DPT metabolites shared with N D G A , a potent renal cystogen, the ability to inhibit catalase activity. In a survey of marker enzyme activities in renal tissues from DPT-treated rats, catalase and other enzymes showed alterations from control groups [8]. More recently, Carone et al. [16] have administered derivatives of DPT to rats and observed that 4OH-DPT causes cystic lesions following only 5 days of treatment; the other metabolites were much less efficient inducers of cystogenesis. Renal tissue from 4OH-DPTtreated animals also had the lowest catalase activity. Noteworthy is the observation that diOH-DPT gave less inhibition of catalase in vivo and a lower degree of cyst formation than 4OH- DPT. Because the hydroxylated metabolites of DPT circulate as glucuronides [16], we speculate that this may be due to a faster rate of elimination of diOH-DPT than 4OH-DPT, but have not examined this possibility experimentally. Our inability to induce renal cysts in rats by lowering renal catalase activity by more than 70% by i.p. aminothiazole (unpublished observations) indicates that diminished catalase activity alone is not sufficient for cyst formation. Because diminished catalase activity is a consistent finding in cystic renal tissue, we interpret our data as supporting the hypothesis that cystogenesis is a multifactorial process [14] in which diminished catalase activity is only one component. The possibility that chemical cystogens may interfere with the detoxification by epoxide hydrolases of cytotoxic, chemical intermediates was also considered. In
219
vitro, 4OH-DPT, but not NDGA, inhibited mEH activity, an enzyme that catalyzes the formation of diols from potentially toxic epoxides [24,25]. In vivo, the activity of mEH was not significantly depressed after 5 days of treatment by the putative cystogens. However, DPT itself caused a slight elevation of this activity; this is in keeping with our earlier report that subchronic administration (1, 2, 4 weeks) of DPT caused proliferation of the smooth endoplasmic reticulum and a selective induction of UDP-glucuronosyltransferase subtypes [8]. Increased capacity for oxidative drug metabolism due to induction of microsomal enzymes with or without involvement of mEH but in the presence of compromised catalase activity may predispose the cell to enhanced cytotoxicity [25,26]. As was the case for catalase, the parent compound DPT had no inhibitory activity toward cytosolic epoxide hydrolase. However, both congeners of DPT that are hydroxylated on the 4-phenyl ring significantly inhibited enzyme activity. Catalase and cytosolic epoxide hydrolase share similar inhibitor-binding tendencies regarding these DPT derivatives, in that DPT has no effect on either enzyme, 5OH-DPT has minor effects on both enzymes, and 4-OH-DPT as well as diOH-DPT produce significant inhibition of both enzymes. NDGA also produced significant inhibition of both catalase and cytosolic epoxide hydrolase activities. Gardner [6] has reported that subchronic administration of NDGA, a known inhibitor of lipoxygenase, decreases the production of certain prostaglandins in the resultant cystic kidneys. In this study we have found that the more potent renal cystogens were inhibitors of cEH activity, cEH rapidly metabolizes epoxyeicosatetraenoic acids (EETs) [12]; mEH also degrades EETs, but at a slower rate [27]. Our results raise the possibility that inhibition of cEH and mEH by the renal cystogens may promote the actions of EETs in vivo by decreasing' their rate of degradation. EETs affect a number of biological activities: electrolyte transport [28,29], development of hypertension and vasomotor tone [30], movement of cellular calcium [31], and exocytosis of peptides and proteins [32-34]. These same phenomena have been postulated to be involved in cyst formation [13-15]. It is noteworthy that EET production in kidney cells can be enhanced by agents that cause microsomal proliferation and induction of cytochrome P-450 [35]; as previously reported, subchronic administration of DPT caused proliferation of the smooth endoplasmic reticulum and induction of microsomal enzymes [8]. Renal cystic disease occurs in both acquired and hereditary forms. In the acquired form of PKD, cEH activity was depressed and mEH unaltered in animals treated with 4OH-DPT, the most cystogenic of the DPT congeners. In contrast, in human cystic cells and tissue derived from a patient with autosomal recessive PKD mEH activity was much lower than in normal human kidney (Table IV); no difference in cEH activity was detected. Thus the two forms of PKD are not similar with respect to observed epoxide hydrolase activities. Further studies are needed to define the relationship between the various types of EH and renal cystic disease in its various forms. In this report, we have demonstrated the association of perturbations in catalase and epoxide hydrolase activities with renal cyst formation [16] and have shown that metabolites of DPT can directly inhibit these enzyme activities in vitro. At least one form of epoxide hydrolase also exhibits lower activity in human
220
cystic versus normal kidney tissue. The question of whether catalase or epoxide hydrolase are fundamentally involved in cyst formation cannot as yet be answered, but nevertheless this study demonstrates that substituted thiazoles decrease cEH activity both in vitro and in vivo.
Acknowledgements This work was supported by U.S. Public Health Service Grants AM33003 and CA34455, and the Anna Turczynski Memorial Fund. References l 2 3
4 5
6 7 8
9 10 11 12
13 14 15
16
17
J.J. Grantham, Polycystic kidney disease: A preponderance of giant nephrons. Am. J. Physiol., 244 (1983) F3. P.H. Bach and T.L. Hardy, Relevance of animal models to analgesic- associated renal papillary necrosis in humans. Kidney Int. 28, (1985) 605. M.M. Dunnill, Acquired cystic disease. In J.J. Grantham and K.D. Gardener (Eds.), Problems in diagnosis and management of polycystic kidney disease. Intercollegiate Press, Kansas City, 1985, pp. 211-223. J.J. Grantham and E. Levine, Acquired cystic disease: Replacing one kidney disease with another, Kidney Int., 28 (1985) 99. V.E. Torres, K.E. Holley and P. Offord, General features of autosomal dominant polycystic kidney disease: Epidemiology, in J.J. Grantham and K.D. Gardener (Eds.), Problems in Diagnosis and Management of Polycystic Kidney Disease, Intercollegiate Press, Kansas City, 1985, pp. 49-69. K.D. Gardner, Jr., W.P. Reed, A.P. Evan, J. Zedalis, M.D. Hylarides and A:A. Leon, Endotoxin provocation of experimental renal cystic disease. Kidney Int., 32 (1987) 329. A.L. Tappel and A.G. Marr, Effect of alpha-tocopherol, propylgallate and nordihydroquaiaretic acid on enzymatic reactions. J. Agric. Food Chem., 2 (1954) 554. J.T. Hjelle, J.J. Hjelle, T.J. Maziasz and F.A. Carone, Diphenylthiazole-induced changes in renal ultrastructure and enzymology: Toxicologic mechanisms in polycystic kidney disease'? J. Pharmacol. Exp. Ther., 243 (1987) 758. E. Margoliash and A. Schejter, Kinetics of the irreversible inhibition of catalase. J. Biol. Chem., 237 (1959) 2359. D. Darr and I. Fridovich, Inhibition of Catalase by 3,3'-diaminobenzidine. Biochem. J., 226 (1984) 781. T.M. Guenthner, M. Qato, R.W. Whalcn and S.A. Glomb, Similarities between catalase and cytosolic epoxide hydrolase. Drug Metab. Rev., 20 (1989) 733. N. Chacos, J. Capdevila, J.R. Falck, S. Manna, C. Martin-Wixtrom, S.S. Gill, B.D. Hammock and R.W. Eastabrook, The reaction of arachidonic acid epoxides (epoxyeicosatrienoic acids) with a cytosolic epoxide hydrolase. Arch. Biochem. Biophys., 223 (1983) 639. E.D. Avner, Renal cystic disease: Insights from recent experimental investigations. Nephron, 48 (1988) 89. K.D. Gardner, Jr., Cystic kidneys. Kidney Int., 33 (1988) 610. V.E, Torres, T.J. Berndt, M. Okamura, J.W. Nesbit, K.E. Holley, F.A. Carone, F.G. Knox and J.C. Romero, Mechanisms affecting the development of renal cystic disease induced by diphenylthiazole, kidney Int., 33 (1988) 1130. F.A. Carone, P.F. Hollenberg, R.F. Novak, W. Golgowski, S. Nakamura and G. Flouret, Tubular basement membrane change occurs pari passu with the development of cystic formation. Kidney Int., 35 (1988) 1034. P. Baudhuin, H. Beaufay, Li Y Rahman, O.Z. Sellinger, R. Wattiaux, P. Jacques and C. de Duve, Tissue fractionation studies: Intracellular distribution of monoamine oxidase, aspartatc
221
18 19 20 21 22 23
24 25 26 27 28
29 30
31
32
33
34
35
222
aminotransferase, alanine aminotransferase, D-aminoacid oxidase and catalase in rat liver tissue. Biochem. J., 92 (1964) 179. P. Bohlen, S. Stein, W. Dairman, and S. Udenfriend, Fluorimetric assay of proteins in the nanogram range. Arch. Biochem. Biophys., 155 (1973) 213. J.T. Hjelle. J.P. Morin and A. Trouet, Analytical cell fractionation of isolated rabbit renal proximal tubules. Kidney Int., 211 (19811 71. T.M. Guenthner and F. Oesch, Identification and characterization of a new epoxide hydrolase from mouse liver microsomes. J. Biol. Chem., 258 (1983) 15054. T.M. Guenthner, Selective inhibition and selective induction of multiple microsomal epoxide hydrolases. Biochem. Pharmacol,, 35 (1986) 839. T.M. Guenthner, P. Bentley and F. Oesch, Microsomal epoxide hydrolase from rat liver. Methods Enzymol., 77 (1981) 344. J.T. Hjelle, D.C. Waters, B.C. Golinska, K.R. Steidley, B. Ketel, D.R. McCarroll, P.J. Olsson, V. Burmeister, R.B. Prior and M.A. Miller, Autosomal recessive polycystic kidney disease: Characterization of human peritoneal and cystic kidney cells in vitro. Am. J. Kidney Dis. In press. F. Oesch and T.M. Guenthner, Effects of the modulation of epoxide hydrolase activity on the binding of benzo[a]pyrene metabolites to DNA in the intact nuclei. Carcinogenesis, 4 (19831 57. A.B. Okey, E.A. Roberts, P.A. Harper and M.S. Denison, Induction of drug-metabolizing enzymes: Mechanisms and consequences. Clin. Biochem., 19 (1986) 132. B. Halliwell, Oxidants and human disease: some new concepts.' The FASEB J., 1 (1987) 358. E, Oliw, P. Guenguerich and J. Oates, Oxygenation of aracbidonic acid by hepatic monooxygenase: isolation and metabolism of four epoxide intermediates. J. Biol. Chem., 257 (1982) 3771. H.R. Jacobson, S. Corona, J. Capdevila, N. Chacos, S. Manna, A. Womack and J.R. Falck, 5,6-Epoxyeicosatrienoic acid ifihibits sodium absorption and potassium secretion in rabbit cortical collecting tubules. Kidney Int., 25 (1985) 3311. E. Schlondorff, E. Petty, J. Oates and S.D. Levine, Effects of 5,6- epoxyeicosatrienoic acid on osmotic water flow in toad urinary bladder. Clin. Res., 33 (1985) 588. K.G. Proctor, J.R. Falck and J. Capdevila, Intestinal vasodilation by epoxyeicosatrienoic acids: arachidonic acid metabolites produced by a cytocbrome P450 monooxygenase. Circ. Res., 6/1 (1987) 50. P. Kutsky, J.R. Falck, G.B. Weiss, S. Manna, N. Chacos and J. Capdevila, Effects of newly reported arachidonic acid metabolism on microsomal Ca 2' binding, uptake and release. Pvostaglandins, 26 (1983) 13. J. Capdevila, N. Chacos, J.R. Falck, S. Manna, A. Negro-Vilar and S.R. Ojeda, Novel hypothalamic arachidonate products stimulate somatostatin release from the median eminence. Endocrinology, 113 (1983) 42l. A. Negro-Vilar, G.D. Snyder, J.R. Falck, S. Manna, N. Chacos and J. Capdevila, Involvement of cicosanoids in release of oxytocin and vasopressin from the neural lobe of the rat pituitary. Endocrinology, 116 119851 2663. J.R. Falck, S. Manna, J. Moltz, N. Chacos and J. Capdevila, Epoxyeicosatrienoic acids stimulate glucagon and insulin release from isolated rat pancreatic islets. Biochem. Biophys. Res., Commun., 114 (1983) 743, M. Schwartzman, M.A. Carroll, N.G. lbraham, N.R. Ferreri, E. Songu-Mize and J.C. McGiff, Renal arachidonic acid metabolism; "The third pathway". Hypertension, 7[Suppl I] (19851 1-136-1 144.
Inhibition of catalase and epoxide hydrolase by the renal cystogen 2-amino-4,5-diphenylthiazole and its metabolites J. Thomas Hjelle "'*, Thomas M. Guenthner b, Kevin Bell a, R o b e r t Whalen b, G e o r g e Flouret c and Frank A. C a r o n e d Departments of Basic Sciences" and Pharmacology b, University of Illinois College of Medicine at Peoria and Chicago, IL 61605 and Departments of Physiology' and Pathology, Northwestern University Medical School, Chicago, IL 60611 (U.S.A.) (Received June 7th, 1989; accepted October 20th, 1989)
Summary Subchronic feeding of 2-amino-4,5-dipbenylthiazole (DPT) to rats results in the development of renal cysts and has been used as a model system to study polycystic kidney disease. Because previous studies revealed changes in renal enzymes following DPT administration, a possible direct effect of DPT and its phenolic metabolites on catalase and a related enzyme, epoxide hydrolase, was examined. Experiments with three in vitro systems (suspensions of rabbit renal tubules, rat kidney homogenates, and commercially obtained bovine liver catalase) revealed direct inhibition of catalase activity by the diphenolic metabolite (diOH- DPT: 2-amino-4,5di(4'-hydroxyphenyl)-thiazole), the known renal cystogen nordihydroquaiaretic acid (NDGA) 2-amino-4(4'- hydroxyphenyl),5-phenyl-thiazole (4OHDPT), and the known catalase inhibitor 3-amino-l,2,4-triazole; DPT did not inhibit catalase activity. Following oral administration to rats of the DPT congeners, 4OH-DPT caused the greatest decrease in both renal catalase and cytosolic epoxide hydrolase activities and the shortest time to onset of cystic lesions. In vitro, mouse liver cytosolic epoxide hydrolase activity was substantially inhibited by 4OH-DPT and dioH-DPT, and NDGA, but not by 2-amino-4-phenyl,5- (4'-hydroxyphenyl)-thiazole (5OH-DPT) or DPT itself. Microsomal epoxide hydrolase (mEH) activity was inhibited by 4OHDPT, unaffected by DPT or dioH-DPT, and stimulated 2-fold by 5OH- DPT. Finally, mEH activity was substantially higher in samples of normal human kidney than in samples of kidney derived from a patient with autosomal recessive polycystic kidney disease; no differences were observed in cEH activity in these samples. Although the role of altered catalase and epoxide hydrolase activities in cystogenesis is unknown, DPT-induced cyst formation is associated with loss of these enzyme activities in kidney tissue. To our knowledge, this is the first report of an in vivo diminution of cytosolic epoxide hydrolase activity by xenobiotics.
* To whom reprint requests" should be addressed: Department of Basic Sciences, University of Illinois College of Medicine at Peoria, Box 1649, Peoria, IL 61656, U.S.A. Abbreviations: DPT, 2-amino-4,5-diphenyl-l,3-thiazole; NDGA, nordihydroquaiaretic acid; PKD, polycystic kidney disease; DMSO, dimethylsulfoxide; cEH, cytosolic epoxide hydrolase; mEH, microsomal epoxide hydrolase; EET, epoxyeicosatetraenoic acids. 0300-483X / 90 / $03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
211
Key words: Renal cystogenesis;Catalase; Epoxide hydrolase; Diphenylthiazole
Introduction
The possibility that renal cystic disease may have a toxicological component is undergoing renewed scrutiny in light of several recent findings. Firstly, administration of chemicals such as diphenylamine, nordihydroguaiaretic acid (NDGA) and 2-amino-4,5-diphenylthiazole (DPT), is known to cause renal cysts in otherwise normal rats [1,2]. Secondly, an acquired form of polycystic kidney disease is appearing at a significant rate in patients undergoing long term renal dialysis [3,4]. Thirdly, epidemiological data support the possibility that the incidence of either renal cysts or polycystic kidney disease (PKD) may have increased over the interval between 1900 and the late 1970's [5]. Unfortunately, very little is known of the biochemical basis for renal cystogenesis. In this report, we investigate the hypothesis that some chemical cystogens may act to disrupt cellular enzymes that protect cells from injury by exogenous or internally produced toxic species. Two such enzymes whose activities are known to be significantly affected by cystogens are catalase and CEH. The linkage of catalase and renal cystogenesis has recently been discussed by Gardner et al. [6], Based in part on the observation of Tappel and Marr [7] that NDGA, then under consideration as a food additive and now known to be a renal cystogen, could inhibit catalase, Gardner et al. [6] have proposed that catalase inhibition in the presence of oxidative stress caused by neutrophil invasion might cause cell death and resultant hyperplasia leading to cyst formation. We have recently reported results of a survey of various enzyme activities in kidney and liver tissues from rats fed the renal cystogen, DPT, subchronically [8]. DPTinduced renal cyst formation was accompanied by numerous changes in kidney enzyme activities, one of which was a diminution of catalase activity. Margoliash et al. [9] reported the minimal structural requirement for the inhibition of catalase to be > N - N H - C ( N H ) - R - H , where R could be sulfur, oxygen, or nitrogen. Included in this group of catalase inhibitors are compounds such as aminoguanidine, semicarbazide, thiosemicarbazide, diaminobenzidine, p-phenylenediamine, and several triazole complexes [9,10]. DPT contains this structural element. Because DPT decreases catalase activity in vivo, and because DPT is structurally similar to known catalase inhibitors, we examined in this study the possibility that DPT or its hydroxylated metabolites could directly act on catalase to inhibit its activity. Cross-reactivity between antiserum to highly purified cytosolic epoxide hydrolase (cEH), and catalase has been observed [11]. A structural relationship between catalase and cEH is further suggested by the observation that the enzymes co-purify by several techniques [11]. Existence of common structural features in both catalase and cEH led us to also investigate whether DPT and its metabolites could inhibit cEH activity. The toxicological significance of cEH has
212
been previously discussed in light of its ability to hydrolyze xenobiotic epoxides. This enzyme also efficiently converts epoxy-eicosatrienoic acids (EETs) to corresponding diol metabolites ]12]. E E T s are known to affect a number of important physiological processes possibly related to cystogenesis [13-15]. Therefore, the effects of D P T and its cystogenic metabolites on these two cytoprotective enzymes, catalase and cEH, are examined, in an attempt to link their abilities to inhibit these enzymes with their abilities to induce renal cyst formation by as yet undisclosed mechanisms. Materials and m e t h o d s
All chemicals used were obtained from the Sigma Chemical Co. (St. Louis, MO), with the exception of D P T and its phenolic analogs which were prepared as previously reported [16]. Chemical structures are given in Fig. 1. Catalase activity was determined by the method of Baudhuid et al. [17] where total reaction volumes of 1.5 ml were used; for each assay, three time points were used to calculate a rate of reaction. Protein was measured by the fluorescamine method of Bohlen et al. [18]. Statistical analyses were performed using the Student's t-test. Suspensions of rabbit renal proximal tubules were isolated by the method of Hjelle et al. [19] in Medium 199 (Sigma Chemical Co., St. Louis, MO). Tubules were incubated in oxygenated M199 medium at pH 7.4 containing the test chemicals for up to 90 min. At various times, aliquots of the tubule suspension were removed and centrifuged for 1 min at 12 000 × g; the resulting pellet was resuspended in 50 mM potassium phosphate buffer at pH 7.2, and sonicated for 30 s. The tubule homogenate was assayed for catalase activity and protein.
R1~
S~ ,/NH2
~ '
/)- z_/
:/
/
R2
Fig. 1. Chemical structures of DPT derivatives used in this study. R1
R2
H H OH OH
H OH H OH
Chemicalname 2-amino-4,5-diphenyl-1,3-thiazole 2-amino-4(4'-hydroxylphenyl),5-phenyl-1,3-thiazole 2-amino-4-phenyl,5-(4'-hydroxyphenyl)-1,3-thiazole 2-amino-4,5-di(4'-hydroxyphenyl)- 1,3-thiazole
Abbreviations DPT 4OH-DPT 5OH-DPT diOH-DPT
213
Rat kidney homogenates were prepared from Sprague-Dawley rats killed by cervical dislocation followed by decapitation. Kidney sections were removed, weighed and suspended (1 : 10, w/v) in 0.25 M sucrose containing lmM EDTA at pH 7.4. Tissues were homogenized in a two-step process involving Dounce homogenization in either Tyrode's buffer or 50 mM potassium phosphate buffer followed by sonication for 1 min. The resulting homogenate was centrifuged at 1000 x g for 10 min. The supernate was then assayed for protein and catalase activity; in some cases the homogenate was frozen at -60°C for 1 week before assays were conducted. Experiments with purified enzyme utilized crystalline bovine liver catalase (0.15 mg/ml) dissolved in 50 mM potassium phosphate buffer. The test chemicals were added in various concentrations to the catalase-buffer system and incubated for up to 1 hour at 37°C. As noted, the test chemicals were dissolved in dimethylsulfoxide (DMSO) resulting in a final DMSO concentration of 20% in the incubation tube; 20% DMSO caused about a 10% reduction in the catalytic activity of the enzyme. For some experiments, hydrogen peroxide (0.007% final concentration) was also added to the incubation mixture. Total assay volume was 1.4 ml. cEH activity was measured in cytosolic fractions prepared from the livers of adult male Swiss Webster mice. These cytosolic fractions were prepared by standard methods [20]. Enzyme activity was measured by the radiometric assay described by Guenthner [21], using tritiated trans-stilbene oxide as the substrate. Inhibitors were added in 5/~1 of DMSO, and 5/xl of DMSO was added to control (uninhibited) incubations. Microsomal epoxide hydrolase activity was measured in hepatic microsomes prepared [20] from adult male Swiss-Webster mice. Enzyme activity was determined by standard methods, using styrene oxide as substrate [22]. In the whole animal experiments, DPT and its congeners were given by gavage in daily doses of 1.2 mg chemical/g body weight for 4 days to male SpragueDawley rats weighing 200-250 g [16]. Each dose consisted of the pulverized chemical suspended in 1 ml of 1% gum-tragacenth; controls received 1 ml of the carrier. On the fifth day, the rats were killed and kidney sections taken for histology and biochemical analyses. Tissues to be used for enzyme determinations were rapidly frozen in liquid nitrogen and later thawed, weighed, and homogenized in 0.25 M sucrose containing 1 mM EDTA. The homogenates were assayed for enzyme activities as described above. Human kidney (from a young adult) that was unuseable for transplantation was the source of normal tissue. Kidney tissue from an infant with autosomal recessive polycystic kidney disease obtained during nephrectomy was the source of cystic tissue. Morphologically, the cystic kidney was spongiform with cysts throughout the medulla and cortex. The cyst walls were lined with epithelial cells having an ultrastructure consistent with a collecting tubule origin. Cystic renal epithelial cells isolated and propagated in vitro as described elsewhere demonstrated lectin staining consistent for collecting tubules [23]. Sections of kidney were frozen for subsequent homogenization and analysis for protein and epoxide hydrolase activities as described above.
214
Results
The ability of DPT and its metabolites to inhibit catalase was examined in three in vitro systems. Incubation of freshly isolated rabbit renal proximal tubules with the test chemicals revealed that the diphenolic metabolite, diOH-DPT, exhibited the greatest inhibition of catalase activity of the DPT congeners tested (Table I). In these catalase-rich cells, the known catalase inhibitor, aminotriazole, was effective in decreasing catalase activity. Several of the inhibitors were poorly soluble in the buffer system employed with the intact kidney cells. Therefore, the test chemicals were dissolved in DMSO and rat renal homogenates were used as a source of catalase. Viability of the proximal tubules was substantially depressed by DMSO. Figure 2 shows the inhibition of catalase activity present in kidney homogenates achieved by incubation of DPT metabolites for 30 rain; of these derivatives, diOH-DPT was the most effective in diminishing catalase activity. Inhibition in this system was not significantly enhanced by the addition of hydrogen peroxide. Of note is the finding that another potent renal cystogen, NDGA, also inhibited the catalase activity present in kidney homogenates with a potency similar to that of the diphenolic metabolite of DPT. DPT itself exhibited little inhibition of catalase activity. Inhibition of catalase activity by all DPT derivatives was time and concentration dependent, as shown for the most potent DPT derivative, diOHDPT (Fig. 3). Inhibition of purified bovine liver catalase activity by 4OH DPT and diOHDPT was enhanced by the In vitro incubation of DPT congener and enzyme in the presence of hydrogen peroxide prior to the assay of catalase activity (Fig. 4). In agreement with the findings of Margoliash et al. [9], we also observed that aminotriazole was substantially less active in the absence of hydrogen peroxide in this system (data not shown) This contrasts with our results in kidney homogenates where hydrogen peroxide did not appreciably augment inhibition of catalase by the test compounds. Nevertheless, 4OH-DPT and diOHDPT did cause a
TABLE I LOSS OF C A T A L A S E A C T I V I T Y BY S U S P E N S I O N S O F R A B B I T R E N A L T U B U L E S I N C U B A T E D W I T H DPT D E R I V A T I V E S IN V I T R O Treatment
Conc (mM)
N
mU ni t s catalase a c t i v i t y / m g tubule protein a
Control 4OH-DPT 5OH-DPT diOH-DPT Aminotriazole Aminotriazole
--
4 3 3 3 3 3
260 -+ 12 216 -+ 8 ** 236 -+ 110 -+ 6 * 118 _+ 4 * 9 -+ 1 *
(/.37 0.37 0.37 1.19 119.00
~(Mean -+S.E.) after 90 min of incubation at 37°C. Significantly different from controls, *P < 0.01 and **P < 0.05.
215
100
"~ 8O m~
v
60
',, ,%~
h= (9
\',, \\
tO
\\
~
20
0
' 0.01
' 0.1
' 1
"~ 10
Log Cystogen Concentration
(mM)
Fig. 2. Inhibition of catalase activity present in rat kidney homogenates by putative renal cystogens. Chemicals dissolved in DMSO were incubated 30 rain with homogenate and then assayed for catalase activity. DiOH-DPT (--O---), NDGA ( ~ ) , 4OH-DPT (--A--) and 5OH-DPT (--O--).
100 90 ,m
80 70
60 o0 tD
50 40
o
o~
30 20 10 0
I
0.01
!
f
I
I
I
0.1
I
I
i
I
I
1
I
!
I
I
10
LOG (4,5 HO-DPT)mM Fig. 3. Time and concentration dependence of catalase inhibition by the most potent of the DPT derivatives, diOH-DPT. Loss of catalase activity after 1 rain ( I ) and 30 mins (©) of incubation of increasing concentrations of diOH-DPT with rat kidney homogenate.
216
©
400 '
O
o
NO
~
\\
__>.>
"6
"300
it= m m O (/J I--
200 E LOG
(4 - HO - DPT raM)
0.01
01
I 0
IOmM
0.01
0.1
1.0
1
[
I
10 mM l
LOG 4,5-diHO-DPT concentration (raM)
Fig. 4. Effect of hydrogen peroxide on the inhibition of catalase activity by DPT derivatives. Commercial preparations of bovine liver catalase were incubated with inhibitor for 30 rains in the presence ( + ) or absence (--Z3---)of 0.007% hydrogen peroxide.
concentration dependent inhibition of catalase activity as monitored by the disappearance of hydrogen peroxide. The possibility that D P T and its congeners may also inhibit c E H , an enzyme that is somewhat structurally related to catalase, was examined using subcellular fractions of mouse liver. Table II shows the inhibition of c E H activity, using trans-stilbene oxide as the the substrate, by D P T and its hydroxylated metabolites, as well as by N D G A . As was the case for catalase, D P T itself produced no inhibition. 5 O H - D P T produced a small amount of inhibition that was statistically significant ( P < 0.02). Both 4 O H - D P T and d i O H - D P T produced highly significant inhibition of c E H activity, with the diphenolic derivative being slightly more effective at equivalent concentrations. N D G A also inhibited the enzyme, to a degree similar to that produced by 4 O H - D P T or 4,5-diOH-DPT. Inhibitor concentrations were chosen to correlate with those used in the catalase experiments, namely 1 m M for D P T and its hydroxylated derivatives, and 2 4 0 / , M for NDGA. As shown in Table II, a second form of epoxide hydrolase located in the microsomal fraction of the cell was inhibited by 4 O H - D P T , stimulated by 5 O H - D P T , and unaffected by the other D P T congeners and N D G A in vitro. Examination of various enzyme activities in kidney homogenates obtained from rats treated with individual congeners of D P T (Table I I I ) revealed that the most cystogenic congener [16], 4 O H - D P T , also caused the greatest depression of catalase and c E H activities in vivo. It is of interest to note that the rank order of
217
TABLE
II
INHIBITION OF CYTOSOLIC BY PUTATIVE CYSTOGENS
AND
MICROSOMAL
Treatment Control DPT 4OH-DPT 5OH-DPT diOH-DPT NDGA
FORMS
OF EPOXIDE
HYDROLASE
cEH activity :''b
mEH activity ;'b
(%)
(%)
100 +- 2.5 99.5 + 1.2 3 9 . 7 _+ 1 .(1 911.8 ± 2.1 ** 3 2 . 9 +- 1.8* 33.11 ± 1.4"
101t + 113 + 40 + 224 ± 102 + 116 +
"Results expressed as percent of control activity; (mean ± S . E . ) ( N = 9). ~'Mouse cytosol assayed with trans-stilbene o x i d e . ~ M o u s e microsomes assayed with styrene oxide significantly different from * * P < 0.02,
4.3 3.3* 8.3* 2.6 3.3
control at * P < 0.001 and
inhibition of cytosolic epoxide hydrolase, 4OH-DPT > diOH-DPT, is similar both in vitro and in vivo. In contrast, catalase is inhibited by diOH-DPT in vitro but not in vivo. The activity of microsomal epoxide hydrolase (mEH) was significantly elevated in DPT-treated rats; mEH activity for control and DPT treated groups was 143.7 +_6.2 (n = 12) and 205 +- 15 (n = 6) pmol styrene oxide hydrolyzed/min/ mg tissue protein (Mean -+S.E.), respectively. The other DPT congeners were without effect on mEH activity. Table IV lists the epoxide hydrolase activities found in cytosolic and microsomal fractions prepared from normal and autosomal recessive PKD human kidneys. Although cEH exhibited similar specific activity in both kidneys, mEH activity was almost tenfold lower in the cystic kidney than in the normal kidney. Cells
TABLE
III
EFFECTS OF ACUTE ORAL ADMINISTRATION OF PUTATIVE RENAL CATALASE AND EPOXIDE HYDROLASE ACTIVITIES
Treatment
N
Control
12 12 8 8 10
CYSTOGENS
N
Catalase activity % control activity ....
100 48.7 93.4 73.1 98.2
8 12 8 8 111
100 74 104 98 86
± -+ -+ + ±
10.3% 3.1%* 16.8% 8.6% 10.3%
.
a
b
± ± ± ± +
6.5% 4.8%* 3.8% 4.8% 4.3%
a(Treatment/control) x 11)0%; ( M e a n -+S.E.). activity = 8 . 5 7 pmol trans-stilbene oxide hydrolyzed/min/mg tissue protein. CControl catalase activity = 4 0 . 7 mUnits/mg tissue protein, *Statistically significant from control P < 0 . 0 1 .
bControl cEH
218
ON
Cytosolic epoxide hydrolase activity % control act~wty " •
4OH-DPT 5OH-DPT diOH-DPT DPT
RENAL
T A B L E IV E P O X I D E H Y D R O L A S E A C T I V I T Y " IN S U B C E L L U L A R F R A C T I O N S O F N O R M A L A N D POLYCYSTIC HUMAN KIDNEY Subcellular fraction Microsomes Cytosol
Normal kidney
A R P K D b kidney
mEH
cEH
mEH
cEH
377 --+35 nt
nt 14 -+ 1
3(1 _+ 17" nt
9.1 13.1
'iActivity is expressed as picomole substrate h y d r o l y z e d / m i n / m g protein. Mean -+S.E. (n : 3 or 4). Single n u m b e r s indicate the m e a n of two determinations. m E H and cEH activities defined as the hydrolysis of styrene oxide and trans-stilbene oxide, respectively. hARPKD: autosomal recessive polycystic kidney disease. *Statistically different from normal kidney value ( P < 0.01).
isolated from the cystic kidney and then propagated in vitro retained the ultrastructure of the epithelial cells that lined the cyst walls, which by electron microscopy and lectin staining appeared to be collecting tubule in origin [23]. Discussion
In this study, we have shown that hydroxylated metabolites of DPT inhibit catalase activity in renal cells, kidney homogenates, and pure preparations of catalase. The 4OH-DPT and diOH-DPT metabolites shared with N D G A , a potent renal cystogen, the ability to inhibit catalase activity. In a survey of marker enzyme activities in renal tissues from DPT-treated rats, catalase and other enzymes showed alterations from control groups [8]. More recently, Carone et al. [16] have administered derivatives of DPT to rats and observed that 4OH-DPT causes cystic lesions following only 5 days of treatment; the other metabolites were much less efficient inducers of cystogenesis. Renal tissue from 4OH-DPTtreated animals also had the lowest catalase activity. Noteworthy is the observation that diOH-DPT gave less inhibition of catalase in vivo and a lower degree of cyst formation than 4OH- DPT. Because the hydroxylated metabolites of DPT circulate as glucuronides [16], we speculate that this may be due to a faster rate of elimination of diOH-DPT than 4OH-DPT, but have not examined this possibility experimentally. Our inability to induce renal cysts in rats by lowering renal catalase activity by more than 70% by i.p. aminothiazole (unpublished observations) indicates that diminished catalase activity alone is not sufficient for cyst formation. Because diminished catalase activity is a consistent finding in cystic renal tissue, we interpret our data as supporting the hypothesis that cystogenesis is a multifactorial process [14] in which diminished catalase activity is only one component. The possibility that chemical cystogens may interfere with the detoxification by epoxide hydrolases of cytotoxic, chemical intermediates was also considered. In
219
vitro, 4OH-DPT, but not NDGA, inhibited mEH activity, an enzyme that catalyzes the formation of diols from potentially toxic epoxides [24,25]. In vivo, the activity of mEH was not significantly depressed after 5 days of treatment by the putative cystogens. However, DPT itself caused a slight elevation of this activity; this is in keeping with our earlier report that subchronic administration (1, 2, 4 weeks) of DPT caused proliferation of the smooth endoplasmic reticulum and a selective induction of UDP-glucuronosyltransferase subtypes [8]. Increased capacity for oxidative drug metabolism due to induction of microsomal enzymes with or without involvement of mEH but in the presence of compromised catalase activity may predispose the cell to enhanced cytotoxicity [25,26]. As was the case for catalase, the parent compound DPT had no inhibitory activity toward cytosolic epoxide hydrolase. However, both congeners of DPT that are hydroxylated on the 4-phenyl ring significantly inhibited enzyme activity. Catalase and cytosolic epoxide hydrolase share similar inhibitor-binding tendencies regarding these DPT derivatives, in that DPT has no effect on either enzyme, 5OH-DPT has minor effects on both enzymes, and 4-OH-DPT as well as diOH-DPT produce significant inhibition of both enzymes. NDGA also produced significant inhibition of both catalase and cytosolic epoxide hydrolase activities. Gardner [6] has reported that subchronic administration of NDGA, a known inhibitor of lipoxygenase, decreases the production of certain prostaglandins in the resultant cystic kidneys. In this study we have found that the more potent renal cystogens were inhibitors of cEH activity, cEH rapidly metabolizes epoxyeicosatetraenoic acids (EETs) [12]; mEH also degrades EETs, but at a slower rate [27]. Our results raise the possibility that inhibition of cEH and mEH by the renal cystogens may promote the actions of EETs in vivo by decreasing' their rate of degradation. EETs affect a number of biological activities: electrolyte transport [28,29], development of hypertension and vasomotor tone [30], movement of cellular calcium [31], and exocytosis of peptides and proteins [32-34]. These same phenomena have been postulated to be involved in cyst formation [13-15]. It is noteworthy that EET production in kidney cells can be enhanced by agents that cause microsomal proliferation and induction of cytochrome P-450 [35]; as previously reported, subchronic administration of DPT caused proliferation of the smooth endoplasmic reticulum and induction of microsomal enzymes [8]. Renal cystic disease occurs in both acquired and hereditary forms. In the acquired form of PKD, cEH activity was depressed and mEH unaltered in animals treated with 4OH-DPT, the most cystogenic of the DPT congeners. In contrast, in human cystic cells and tissue derived from a patient with autosomal recessive PKD mEH activity was much lower than in normal human kidney (Table IV); no difference in cEH activity was detected. Thus the two forms of PKD are not similar with respect to observed epoxide hydrolase activities. Further studies are needed to define the relationship between the various types of EH and renal cystic disease in its various forms. In this report, we have demonstrated the association of perturbations in catalase and epoxide hydrolase activities with renal cyst formation [16] and have shown that metabolites of DPT can directly inhibit these enzyme activities in vitro. At least one form of epoxide hydrolase also exhibits lower activity in human
220
cystic versus normal kidney tissue. The question of whether catalase or epoxide hydrolase are fundamentally involved in cyst formation cannot as yet be answered, but nevertheless this study demonstrates that substituted thiazoles decrease cEH activity both in vitro and in vivo.
Acknowledgements This work was supported by U.S. Public Health Service Grants AM33003 and CA34455, and the Anna Turczynski Memorial Fund. References l 2 3
4 5
6 7 8
9 10 11 12
13 14 15
16
17
J.J. Grantham, Polycystic kidney disease: A preponderance of giant nephrons. Am. J. Physiol., 244 (1983) F3. P.H. Bach and T.L. Hardy, Relevance of animal models to analgesic- associated renal papillary necrosis in humans. Kidney Int. 28, (1985) 605. M.M. Dunnill, Acquired cystic disease. In J.J. Grantham and K.D. Gardener (Eds.), Problems in diagnosis and management of polycystic kidney disease. Intercollegiate Press, Kansas City, 1985, pp. 211-223. J.J. Grantham and E. Levine, Acquired cystic disease: Replacing one kidney disease with another, Kidney Int., 28 (1985) 99. V.E. Torres, K.E. Holley and P. Offord, General features of autosomal dominant polycystic kidney disease: Epidemiology, in J.J. Grantham and K.D. Gardener (Eds.), Problems in Diagnosis and Management of Polycystic Kidney Disease, Intercollegiate Press, Kansas City, 1985, pp. 49-69. K.D. Gardner, Jr., W.P. Reed, A.P. Evan, J. Zedalis, M.D. Hylarides and A:A. Leon, Endotoxin provocation of experimental renal cystic disease. Kidney Int., 32 (1987) 329. A.L. Tappel and A.G. Marr, Effect of alpha-tocopherol, propylgallate and nordihydroquaiaretic acid on enzymatic reactions. J. Agric. Food Chem., 2 (1954) 554. J.T. Hjelle, J.J. Hjelle, T.J. Maziasz and F.A. Carone, Diphenylthiazole-induced changes in renal ultrastructure and enzymology: Toxicologic mechanisms in polycystic kidney disease'? J. Pharmacol. Exp. Ther., 243 (1987) 758. E. Margoliash and A. Schejter, Kinetics of the irreversible inhibition of catalase. J. Biol. Chem., 237 (1959) 2359. D. Darr and I. Fridovich, Inhibition of Catalase by 3,3'-diaminobenzidine. Biochem. J., 226 (1984) 781. T.M. Guenthner, M. Qato, R.W. Whalcn and S.A. Glomb, Similarities between catalase and cytosolic epoxide hydrolase. Drug Metab. Rev., 20 (1989) 733. N. Chacos, J. Capdevila, J.R. Falck, S. Manna, C. Martin-Wixtrom, S.S. Gill, B.D. Hammock and R.W. Eastabrook, The reaction of arachidonic acid epoxides (epoxyeicosatrienoic acids) with a cytosolic epoxide hydrolase. Arch. Biochem. Biophys., 223 (1983) 639. E.D. Avner, Renal cystic disease: Insights from recent experimental investigations. Nephron, 48 (1988) 89. K.D. Gardner, Jr., Cystic kidneys. Kidney Int., 33 (1988) 610. V.E, Torres, T.J. Berndt, M. Okamura, J.W. Nesbit, K.E. Holley, F.A. Carone, F.G. Knox and J.C. Romero, Mechanisms affecting the development of renal cystic disease induced by diphenylthiazole, kidney Int., 33 (1988) 1130. F.A. Carone, P.F. Hollenberg, R.F. Novak, W. Golgowski, S. Nakamura and G. Flouret, Tubular basement membrane change occurs pari passu with the development of cystic formation. Kidney Int., 35 (1988) 1034. P. Baudhuin, H. Beaufay, Li Y Rahman, O.Z. Sellinger, R. Wattiaux, P. Jacques and C. de Duve, Tissue fractionation studies: Intracellular distribution of monoamine oxidase, aspartatc
221
18 19 20 21 22 23
24 25 26 27 28
29 30
31
32
33
34
35
222
aminotransferase, alanine aminotransferase, D-aminoacid oxidase and catalase in rat liver tissue. Biochem. J., 92 (1964) 179. P. Bohlen, S. Stein, W. Dairman, and S. Udenfriend, Fluorimetric assay of proteins in the nanogram range. Arch. Biochem. Biophys., 155 (1973) 213. J.T. Hjelle. J.P. Morin and A. Trouet, Analytical cell fractionation of isolated rabbit renal proximal tubules. Kidney Int., 211 (19811 71. T.M. Guenthner and F. Oesch, Identification and characterization of a new epoxide hydrolase from mouse liver microsomes. J. Biol. Chem., 258 (1983) 15054. T.M. Guenthner, Selective inhibition and selective induction of multiple microsomal epoxide hydrolases. Biochem. Pharmacol,, 35 (1986) 839. T.M. Guenthner, P. Bentley and F. Oesch, Microsomal epoxide hydrolase from rat liver. Methods Enzymol., 77 (1981) 344. J.T. Hjelle, D.C. Waters, B.C. Golinska, K.R. Steidley, B. Ketel, D.R. McCarroll, P.J. Olsson, V. Burmeister, R.B. Prior and M.A. Miller, Autosomal recessive polycystic kidney disease: Characterization of human peritoneal and cystic kidney cells in vitro. Am. J. Kidney Dis. In press. F. Oesch and T.M. Guenthner, Effects of the modulation of epoxide hydrolase activity on the binding of benzo[a]pyrene metabolites to DNA in the intact nuclei. Carcinogenesis, 4 (19831 57. A.B. Okey, E.A. Roberts, P.A. Harper and M.S. Denison, Induction of drug-metabolizing enzymes: Mechanisms and consequences. Clin. Biochem., 19 (1986) 132. B. Halliwell, Oxidants and human disease: some new concepts.' The FASEB J., 1 (1987) 358. E, Oliw, P. Guenguerich and J. Oates, Oxygenation of aracbidonic acid by hepatic monooxygenase: isolation and metabolism of four epoxide intermediates. J. Biol. Chem., 257 (1982) 3771. H.R. Jacobson, S. Corona, J. Capdevila, N. Chacos, S. Manna, A. Womack and J.R. Falck, 5,6-Epoxyeicosatrienoic acid ifihibits sodium absorption and potassium secretion in rabbit cortical collecting tubules. Kidney Int., 25 (1985) 3311. E. Schlondorff, E. Petty, J. Oates and S.D. Levine, Effects of 5,6- epoxyeicosatrienoic acid on osmotic water flow in toad urinary bladder. Clin. Res., 33 (1985) 588. K.G. Proctor, J.R. Falck and J. Capdevila, Intestinal vasodilation by epoxyeicosatrienoic acids: arachidonic acid metabolites produced by a cytocbrome P450 monooxygenase. Circ. Res., 6/1 (1987) 50. P. Kutsky, J.R. Falck, G.B. Weiss, S. Manna, N. Chacos and J. Capdevila, Effects of newly reported arachidonic acid metabolism on microsomal Ca 2' binding, uptake and release. Pvostaglandins, 26 (1983) 13. J. Capdevila, N. Chacos, J.R. Falck, S. Manna, A. Negro-Vilar and S.R. Ojeda, Novel hypothalamic arachidonate products stimulate somatostatin release from the median eminence. Endocrinology, 113 (1983) 42l. A. Negro-Vilar, G.D. Snyder, J.R. Falck, S. Manna, N. Chacos and J. Capdevila, Involvement of cicosanoids in release of oxytocin and vasopressin from the neural lobe of the rat pituitary. Endocrinology, 116 119851 2663. J.R. Falck, S. Manna, J. Moltz, N. Chacos and J. Capdevila, Epoxyeicosatrienoic acids stimulate glucagon and insulin release from isolated rat pancreatic islets. Biochem. Biophys. Res., Commun., 114 (1983) 743, M. Schwartzman, M.A. Carroll, N.G. lbraham, N.R. Ferreri, E. Songu-Mize and J.C. McGiff, Renal arachidonic acid metabolism; "The third pathway". Hypertension, 7[Suppl I] (19851 1-136-1 144.
