Regulation of renal kallikrein synthesis and activation by glucocorticoids

Kidney International, Vol. 38 (1990), pp. 2/2—218

LABORATORY INVESTIGATION

Regulation of renal kallikrein synthesis and activation by glucocorticoids AYAD A. JAFFA, DONALD H. MILLER, RICARDO H. SILVA, HARRY S. MARGOLIUS, and RONALD K. MAYFIELD Departments of Medicine and Pharmacology, Medical University of South Carolina, and Veterans Administration Medical Center, Charleston, South Carolina, USA

Regulation of renal kallikrein synthesis and activation by glucocorti-

after adrenalectomy increases renal free water clearance; however, glucocorticoid administration increases solute free water renal active and prokallikrein levels (nglmg protein) and in vivo kallikrein synthesis rate were studied in the conscious rat. Within two reabsorption, a finding consistent with the direct observation hours after low dose methyiprednisolone (MP, 0.0125 to 0.05 mg/l00 g that water permeability is increased in isolated rabbit cortical body wt), active kallikrein and prokallikrein fell (29.1 2.3 and 35.1 collecting tubule following dexamethasone treatment [3, 4, 7]. 2,7 nglmg protein, respectively, compared to 38.4 3.7 and 42.7 3.4 Several studies provide evidence that there is direct binding in vehicle-treated rats, P < 0.05 or less). These changes were accomof glucocorticoid within cells of the tubule. In rabbit distal panied by a significant fall in prokallikrein synthesis rate relative to total protein synthesis. The reductions in active and prokallikrein levels were convoluted and collecting tubules, and in rat distal convoluted transient, dissipating by six hours. With increasing MP doses, there was tubules, autoradiography has demonstrated nuclear binding of further dose-dependent reduction in active kallikrein. However, prokallikrein levels increased to normal as the MP dose was increased despite dexamethasone or corticosterone [8, 9]. In both species, glucocontinued suppression of synthesis, suggesting that prokallikrein acti- corticoid rather than mineralocorticoid binding predominates in vation was inhibited. Renal kallikrein levels were also examined in the distal convoluted tubule [8, 10]. These autoradiographic relation to changes in endogenous glucocorticoid levels. In intact rats, studies are supported by the finding that the rat kidney contains three hours after plasma corticosterone peaked (10 p.m.), active and prokallikrein levels were 30.2 2.9 and 27.0 high affinity cytoplasmic receptors which bind glucocorticoid 1.6 ng/mg protein, respectively, compared to 36.9 2.3 and 37.2 2.6 (P < 0.005) three and are transferred into the nucleus [11]. hours after the corticosterone nadir (11 a.m.). Furthermore, adrenalecWhile functional and receptor binding studies suggest that tomy increased active and prokallikrein (47.3 4.8 and 87.3 6.0 ng/mg protein, respectively), compared to levels in intact or sham- glucocorticoids have direct effects in the nephron, little inforoperated rats (intact: 32.9 2.9 and 54.9 5.3 ng/mg protein, P < 0.01 mation is available on the effects of glucocorticoids on specific or less). Adrenalectomy also eliminated the diurnal changes in kal- proteins involved in nephron function. Na-K ATPase activlikrejn levels seen in intact rats. These data suggest that renal prokality is increased by glucocorticoid treatment, but it is not known likrein synthesis and activation are physiologically regulated by glucocorticoids. if synthesis of this enzyme changes [6]. Only one renal protein, cytosolic phosphoenolpyruvate carboxykinase, has been shown to alter its synthesis rate following glucocorticoid treatment coids. The effects of endogenous and exogenous glucocorticoids on

[121.

Glucocorticoids have a number of recognized effects on renal

There are several reports that glucocorticoids, in chronic or function. Administration to intact or adrenalectomized rats pharmacologic doses, alter renal tissue levels or urinary excre-

increases GFR [1]. Although steroid-induced changes in GFR tion of kallikrein [13—181. This serine protease is localized to may influence electrolyte and water excretion, it is clear that apical and basolateral membranes of distal connecting tubule glucocorticoids also directly modulate electrolyte and water cells [19], and a high in vivo rate of kallikrein synthesis rate has handling by the renal tubule [2—4]. Acute in vivo administration of cortisol produces renal sodium retention, but clearance been demonstrated in the rat kidney [20, 21]. Kallikrein and its studies suggest that glucocorticoid inhibits proximal tubule kinin product are potent agonists for vectorial ion and water transport across renal collecting tubules and cultured collecting sodium reabsorption [3, 4]. Dexamethasone acutely stimulates NatK ATPase activity in isolated rat or rabbit distal convo- tubule cells [22—24]. Basolateral addition of kinins inhibits luted tubules [5, 6], an effect that may be consistent with ADH-stimulated sodium and water reabsorption in cortical sodium reabsorption. Systemic replacement of glucocorticoid collecting tubules of rat or rabbit [23, 24]. Some evidence also suggests that renal kallikrein and kinins may modulate tubuloglomerular feedback [25]. To begin to explore whether kallikrein may mediate some of the renal actions of glucocorticoid, Received for publication October 10, 1989 we studied the effects of acute physiologic manipulations of and in revised form March 12, 1990 glucocorticoid levels on renal active and prokallikrein levels Accepted for publication March 15, 1990 and kallikrein synthesis. © 1990 by the International Society of Nephrology 212

213

Jaffa et a!: Kallikrein regulation by glucocorticoids

Methods

Treatment protocols Male Wistar rats (Charles River Breeding Laboratories, Inc., Wilmington, Massachusetts, USA) weighing 140 to 180 g were

treatment). Kallikrein levels are expressed as ng/mg tissue protein. Tissue homogenate protein was measured by the method of Lowry et al [281, using bovine serum albumin as

standard. Plasma corticosterone levels were measured directly by a radioimmunoassay, using antiserum B3-l63 (Endocrine used in all studies, Rats were housed 4 to 5 per cage in a Sciences, Tarzana, California, USA).

temperature and humidity controlled room with a 12-hour light/12-hour dark cycle, and had free access to water and regular chow. Experimental groups of rats which received

Tissue kallikrein synthesis

The in vivo kallikrein synthesis rate, relative to total protein succinate (A-MethaPred, Abbott Laboratories, Chicago, Illi- synthesis, was measured by the method of Miller, Chao and nois, USA). Methyiprednisolone was chosen because of its Margolius [21]. As described above, 35S-methionine (1 pCi/g very low mineralocorticoid activity. Control rats were injected body wt) was injected i.p., and after 20 minutes the rats were i.p. with saline diluent. In all studies of the effects of exogenous anesthetized, the kidneys removed and a tissue homogenate glucocorticoid, rats were injected and killed between 9 a.m. and supernatant prepared. Incorporation of 35S-methionine into 2 p.m. Two to six hours after injection, rats were anesthetized tissue kallikrein was measured by selective immunoprecipitarapidly with methoxyfluorane (Metofane, Pitman-Moore, Inc., tion with a polyclonal antikallikrein antiserum which recognizes Washington Crossing, New Jersey, USA). The kidneys were both active and prokallikrein [27]. To immunoprecipitate kalexcised for measurements of tissue protein, kallikrein, and likrein from the homogenate supernatant, the supernatant was kallikrein synthesis rate. To measure kallikrein synthesis rate, first centrifuged at 27,000 g at 4°C for 90 minutes. Supernatant rats were injected i.p. with 35S-methionine (1 iCi/g body wt, retained from this centrifugation was treated with Triton X-l00, diluted 1 iCi/pi saline, New England Nuclear, Bedford, Mas- sodium dodecyl sulfate (SDS) and EDTA (1%, 0.1% and 0.1 mM sachusetts, USA) 20 minutes before they were anesthetized and final concentrations, respectively) and then centrifuged at 181,000 g at 4°C for 45 minutes in a Beckman ultracentrifuge. killed. In studies of adrenalectomized rats, rats underwent adrenal- Undiluted kallikrein antiserum (5 d/g kallikrein) was added to ectomy or a sham operation, recovered from anesthesia, and duplicate aliquots of 0.8 to 1.0 ml of the resulting supernatant, were killed 24 hours later (11 a.m. or 10p.m.). Adrenalectomies and these aliquots were incubated for 18 hours at 4°C. Another were performed through bilateral flank incisions on rats anes- set of duplicate aliquots was treated with nonimmune sheep thetized with sodium brevital (5 mgIlOO g body wt). Complete- serum to measure nonspecific precipitation. After incubation, ness of adrenalectomy was confirmed by plasma corticosterone all aliquots were centrifuged at 5000 g for 10 minutes. The measurement when kidneys were harvested. Rats were anes- immunoprecipitate was recovered and washed twice with 1 ml thetized rapidly with methoxyfluorane and kidneys removed water containing the detergents listed above. After final centrifglucocorticoid were injected i.p. with methyiprednisolone (MP)

immediately and prepared as described below. In diurnal rhythm studies, rats were also killed at 11 a.m. or 10p.m.

Preparation of tissue homogenates Each kidney was excised and immediately perfused via the renal hilus with a hand-held syringe containing 10 ml of ice cold saline (0.9%), then immediately frozen at —20°C until homoge-

nized. Kidneys were minced and then homogenized in 5 ml

ugation (5000 g for 10 mm), the precipitate was dissolved in 0.2

ml of 0.1 M NaOH and transferred to a vial containing liquid scintillation cocktail for counting of radioactivity. Radioactivity immunoprecipitated from a tissue homogenate migrates as a single peak on SDS-polyacrylamide gel electrophoresis, indicating selective precipitation of kallikrein labelled in vivo [211. Intrassay coefficient of variation is 5%. To measure 35S-methionine incorporation into total protein,

phosphate buffered saline (PBS, 0.14 M NaCI in 0.01 M duplicate aliquots (0.02 ml) of the supernatant were dissolved in Na2HPO4-NaH2PO4, pH 7.4) with a teflon/glass homogenizer

0.05 ml of 0.1 M NaOH at 37°C for 30 minutes. Bovine serum

(15 strokes). Sodium deoxycholate was then added to the

albumin and casein hydrolysate were added to each aliquot

Filtered supernatant was used for kallikrein and protein assays, and the measurements of kallikrein synthesis.

M NaOH. This was transferred to a vial containing liquid scintillation cocktail for counting of radioactivity. Relative

(0.01% and 0.2% final concentrations, respectively), which was homogenate (0.5% final concentration), followed by incubation at 4°C for 60 minutes. The homogenate was centrifuged (27,000 then treated with 0.5 ml of 10% trichloroacetic acid (TCA). The g at 4°C for 45 mm) and the resulting supernatant was centrif- precipitated protein, recovered after centrifugation at 4000 g at 4°C for 10 minutes, was washed twice with I ml of 5% TCA, and ugally filtered through Sephadex G-25 (Pharmacia Inc., Piscataway, New Jersey, USA) to remove salts and detergent [20]. after final centrifugation the pellet was dissolved in 0.2 ml of 0.1

Assays Active and prokallikrein were measured by a monoclonal antibody-radioimmunoassay which has been previously de-

synthesis rate was calculated as the ratio of counts immunoprecipitated by antiserum to counts precipitated by TCA.

Statistical analysis

scribed in detail [20]. The specificity of this antibody for active Data are expressed as mean SEM and were analyzed by kallikrein has been demonstrated in tissue homogenate as well analysis of variance (ANOVA). After ANOVA of repeated as urine [26, 27]. Prokallikrein was determined by assaying each measurements (dose-response and time course studies), if difsample before and after trypsin treatment, which converts ferences were detected, individual data sets were compared by

prokallikrein to active kallikrein, and subtracting the active

t-test. Dose-response data were also examined by Scheffe's

kallikrein content (pretreatment) from the total kallikrein (after linear trend contrast analysis.

214

Jaffa et al: Kallikrein regulation by glucocorticoids 50

Table 1. Time-dependency of glucocorticoid effects on renal

kallikrein Kallikrein ng/mg

20. 40

protein Active Prokallikrein Total

¶

Time hr

0 38.4

42.7 81,1

2 3.7 3.4 6.2

29.1 35.1 64.1

6

2.3" 2.7C

2.8a

4.9

43.9 39.3 83.2

5.2 5.7

Rats (8 to 9 at each time point) were injected i.p. with methyipred-

nisolone (0.0125 mg/l00 g body wt) and killed 2 or 6 hours later. Controls (0 hr) were injected with vehicle and killed immediately.

I

a < 0.01

' P < 0.025

t

30

P < 0.05 vs. 0 hr

** 20 0

0.0125

0.125

1.25

12.5

Methyiprednisolone, mg/bOg 8W Fig.

Table 2. Renal kallikrein content and synthesis rate in controls and methylprednisolone (MP)-treated rats (0.0125 or 0.05 mg/l00 g body wt)

1. Renal tissue active (•) and prokallikrein (0) levels in rats

treated with methylprednisolone (MP). Rats (12 at each dose) were killed four hours after i.p. injection of MP or vehicle (0 dose). Active kallikrein showed a dose-related reduction. Prokallikrein was reduced with the lowest dose but increased toward control levels as the dose of MP was increased. ** p < 0.005, tP < 0.02, §P < 0.05 vs. vehicletreated,

Kallikrein ng/mg protein Active Prokallikrein Total Kallikrein synthesis ratea

Control

MP-treated

(N= 18)

(N=20)

47.3 39.4 86.6 2.70

3.2

1.4"

3.1

33.5 27.7

2.1C

4.9 0.25

2.32

0.101

61.2 26b

Rats were injected with MP or vehicle i.p. and killed after 3 hr. 35S-methionine was injected i.p. 20 minutes before killing to label kallikrein. The effects of 0.0125 and 0.05 mg were similar, and therefore the data are combined.

a l03 x ratio of kallikrein cpm/mg protein to TCA-protein cpm/mg protein "P < 0.001, P < 0.005, d P < 0.05 vs. control

Results

Renal kallikrein levels in MP-treated rats Renal tissue levels of active and prokallikrein in rats treated in the a.m. with MP (0.0125 to 12.5 mg/l00 g body wt) or vehicle are shown in Figure 1. Measured four hours after MP, active renal kallikrein showed a progressive dose-dependent reduction. This reduction was significant at a dose of 0.125 mg/l00 g body weight (34.2 3.6 vs. 44.0 4.4 ng/mg protein in MP vs.

vehicle-treated, respectively, P < 0.05). The response of prokallikrein to increasing MP doses differed. While prokal-

consistently extracted from kidney tissue by the first of repetitive extractions with deoxycholate. After the second extraction, negligible immunoreactive kallikrein could be obtained. This was not altered by low dose (0.0125 mg/100 g) or high dose (12.5 mg/l00 g) glucocorticoid treatment. Renal kallikrein synthesis in MP-treated rats Renal kallikrein synthesis was measured in rats treated with

likrein was significantly reduced with the lowest MP dose (35.5 low doses (0.0125 or 0.05 mgIlOO g body wt) or high doses (12.5 2.8 in MP vs. 46.0 3.7 ng/mg protein in vehicle-treated, P mg/tOO g) of MP. Since the above studies showed that tissue < 0.02), as the dose increased, prokallikrein increased. Rats prokallikrein was significantly reduced two to four hours after a receiving 12.5 mg/100 g showed levels similar to vehicle-treated low dose of MP, we measured kallikrein synthesis rate and levels three hours after MP injection. The effects of 0.0 125 and rats (43.6 3.6 ng/mg protein). To characterize the time-dependency of the effects of gluco- 0.05 mg were similar and therefore the combined results are

corticoid on renal active and prokallikrein, their levels were shown in Table 2. Both active and prokallikrein levels were measured two and six hours after injection of the lowest dose of reduced at three hours. This was accompanied by a significant MP (0.0125 mg/100 g body wt). These results are shown in Table reduction in kallikrein synthesis rate. High dose MP (12.5 1. After two hours, active and prokallikrein were significantly mg/bOg) reduced kallikrein synthesis to a similar extent (2.40 0.19, P < 0.005). As observed in the dosereduced in MP-treated compared to vehicle-treated rats (0 hr). 0,13 vs. 2.91 However, after six hours, both levels had returned to normal. response study, the tissue level of prokallikrein was not signif3.7 Because a significant portion of kallikrein in the kidney is icantly reduced by high dose MP (22.0 2.9 vs. 30.1 membrane bound, we examined whether the reduction in nglmg protein, MP vs. vehicle, respectively). However, active kallikrein observed following glucocorticoid treatment could be

kallikrein was reduced (23.6

1.4 vs. 48.4

5.2 ng/mg protein,

the result of redistribution of kallikrein between cytosol and P < 0.001). Finding that renal kallikrein levels and synthesis were acutely membranes, and therefore due to a change in the amount of kallikrein extracted by detergent treatment. In control kidneys, reduced by small doses of exogenous glucocorticoid suggested more than 90% of prokallikrein and active kallikrein were that changes in endogenous glucocorticoid levels might have

215

Jaffa et a!: Kallikrein regulation by glucocorticoids 80

w0

60 80

2

4

40

w

S

2

(0 S

60

(0

C 0)

20 C 'I,

S

40

C Q)

am pm

am

pm

am

pm

Fig. 3. Diurnal changes in renal kallikrein in intact rats. Rats were killed at 11 a.m. or 10 p.m. Numbers of rats in each group are shown at

the bottom of the solid bars. Symbols are: (U) total; () active; (0) prokallikrein. ** P < 0.003, §P < 0.05 vs. a.m. 20

Fig. 2. Effects of adrenalectomy on renal active and prokallikrein.

3.2 vs. 37.6 5.4 ng/mg protein in MP-treated vs. untreated), respectively. Renal kallikrein levels were also compared in adrenalectomized and intact rats killed at 10 to 11 a.m., when intact rats show low plasma levels of corticosterone [291. In contrast to the above differences observed in the p.m., adrenalectomized and

the nadir in plasma corticosterone of intact rats (D), kallikrein levels did not differ between adrenalectomized and intact rats. * P < 0.001, tP < 0.01 vs. intact or sham-operated.

57.7 4.6 ng/mg protein; prokallikrein: 37.6 5.4 vs. 41.5 7.7 nglmg protein, adrenalectomized vs. intact, respectively).

(46.2

10

Active

IC Prokallikrein

Rats were adrenalectomized (U) or sham-operated () in the p.m. and killed 24 hours later at 10 p.m. When rats were killed in the a.m., near

similar effects. To assess this, we examined the effects of acute

control rats killed in the a.m. did not display differences in either active or prokallikrein (active kallikrein: 56.7 4.7 vs.

Diurnal changes in renal kallikrein levels Figure 3 shows renal kallikrein levels in intact rats killed at either 11 a.m., approximately three hours after their nadir in

adrenalectomy and the relation between diurnal rhythm in plasma corticosterone diurnal rhythm, or at 10 p.m., three glucocorticoid secretion and renal active and prokallikrein hours after their peak corticosterone level [29]. Active and levels.

prokallikrein levels were significantly lower in the p.m. than the

a.m. To confirm that these changes in kallikrein were due to diurnal changes in adrenal secretion, we also measured kallikrein levels at 11 a.m. or 10 p.m. in adrenalectomized rats were killed at 10 p.m. Plasma corticosterone levels were (Table 3). Plasma corticosterone levels were undetectable in undetectable in adrenalectomized rats, but sham-operated rats these rats, when they were killed 22 to 31 hours after adrenalhad levels similar to intact control rats (7.32 0.57 vs. 8.09 ectomy. There was no difference in active or prokallikrein 0.74 jsgldl, sham vs. control, respectively). The kallikrein levels levels in adrenalectomized rats killed in the a.m. versus p.m., from all three groups are shown in Figure 2. In adrenalecto- demonstrating that adrenalectomy eliminates the diurnal mized rats, renal active and prokallikrein were increased 43% change in kallikrein levels. and 59%, respectively, compared to levels in either intact or Discussion sham-operated rats. Treatment of adrenalectomized rats with These studies demonstrate that doses of a synthetic glucoMP (0.5 mg/100 g, midrange of the dose-response curve, Fig. 1), reduced active kallikrein after three hours (24.9 3.3 vs. 56.7 corticoid, biologically equivalent to very low physiologic doses 4.7 nglmg protein in untreated adrenalectomized rats, P < of the rat's natural glucocorticoid [30], produce rapid and 0.001). However, prokallikrein was not reduced by this dose proportional reductions in renal synthesis of kallikrein and the Renal kallikrein levels following adrenalectomy Adrenalectomized, sham-operated, and unoperated controls

216

Jaffa ef al: Kallikrein regulation by glucocorticoids

Table 3. Effect of adrenalectomy (Adx) on diurnal changes in renal kallikrein Adx

Adx (p.m.) < 1.0

(a.m.) Plasma corticosterone isgidi Renal kallikrein ng/mg protein

Active Prokallikrein Total

< 1.0 47.2 73.3 120.4

2.9 4.7 5.4

50.0 62.0 112.0

4.9 4.9 8.2

Rats (10 in each group) were killed at 11 a.m. or 10 p.m., 23 to 31

hours after adrenalectomy. In contrast to intact rats (c.f. Fig. 3),

Wistar rats [16]. The changes observed in enzyme activity in these studies are consistent, but results from studies of adrenalectomized rats have not been. Four studies have examined the effects of adrenalectomy (3 days to 5 weeks) on rat renal kallikrein [13, 14, 17, 18]. In the two shorter studies, renal tissue kallikrein-like activity was reduced after three days in rats which were not given dietary sodium supplementation [13], whereas only prokallikrein was decreased in isolated basolateral membranes after seven days in rats given 0.9% saline drinking water [17]. In these same studies

and in another, urinary kallikrein-like activity was reduced

three to seven days after adrenalectomy, whether extra dietary sodium was given or not [13, 17, 31]. In contrast, Chao and Margolius [14] found increased immunoreactive kallikrein (total) in kidney and urine after five weeks in salt-supplemented, tissue levels of prokallikrein and active kallikrein. Furthermore, adrenalectomized rats. the levels of both active enzyme and zymogen show diurnal Studies of the effects of replacing glucocorticoid in adrenalchanges which are inverse to the diurnal cycle of plasma ectomized rats also have not shown consistent results. Very corticosterone, and eliminating endogenous glucocorticoid by small doses of corticosterone did not reverse the fall in urinary adrenalectomy raises prokallikrein and active kallikrein within kallikrein-like activity of seven day adrenalectomized rats [31], 24 hours. Thus, changes in glucocorticoid levels within the and with tenfold the physiologic replacement dose, dexamethaphysiologic range, whether endogenous or exogenous in sone did not change renal kallikrein mRNA levels five days source, had qualitatively and quantitatively similar effects on after adrenalectomy [15]. In contrast, Noda et a! [17] found that renal kallikrein levels. These changes in kallikrein levels (ng/mg a large dose of dexamethasone raised the reduced prokallikrein protein) were independent of changes in total renal protein, as level of basolateral membranes isolated from kidneys of adreMP or adrenalectomy did not change total protein content of the nalectomized rats. Chao and Margolius [14] observed that adrenalectomized rats showed no diurnal change in kallikrein levels.

kidney (data not shown). Together, these findings strongly suggest that tissue kallikrein in the kidney is regulated by

cortisol replacement also reversed a change in kidney total

kallikrein following adrenalectomy, but the effect was to lower raised levels. The inconsistencies from assessing these studies collectively The time course and the effects of different doses of glucocorticoid on active and prokallikrein suggest that both synthesis may be related to a number of factors. First, various doses of and activation of prokallikrein are responsive to glucocorticoid. glucocorticoid have been given for various durations. Our Following a low dose of MP (0.0125 mg/tOO g body wt), active results show that active and prokallikrein levels respond trankallikrein was reduced as early as two hours, but was not siently and the changes are dose dependent, with large doses significantly reduced at four hours, and by six hours had preferentially affecting active kallikrein levels. Most studies returned to control levels (Table 1, Fig. 1). Following the same employed supraphysiologic doses. Moreover, chronic high MP dose, prokallikrein levels were significantly reduced from doses of glucocorticoid can have a variety of systemic effects two to four hours, when synthesis was also inhibited, but by six which may secondarily influence renal kallikrein. Second, studies of the effects of adrenalectomy and subsehours levels returned to control. High dose MP inhibited kallikrein synthesis similarly, but prokallikrein levels did not quent steroid replacement have been carried out in rats adrefall and active kallikrein fell more than after the low dose. These nalectomized for several days and either given a high sodium changes suggest that low glucocorticoid doses rapidly inhibit diet or no sodium supplementation. High sodium intake, resynthesis and conversion of prokallikrein to active enzyme, quired to sustain chronically adrenalectomized rats, may itself with activation recovering more rapidly, whereas high doses have effects on renal kallikrein [32], yet lack of sodium suppleinhibit zymogen activation to the extent that prokallikrein levels mentation can result in hemodynamic changes which could do not fall despite a reduction in synthesis. In any case, our influence the kidney's production of kallikrein. Since adrenalectomy also eliminates mineralocorticoid and direct measurements clearly show that kallikrein synthesis is reduced by a MP dose that is less than 10% of the biologically reduces androgen production, its effects on kallikrein could be equivalent dose of corticosterone that normalizes plasma cor- related to reduction in these hormones. Although chronic ticosterone levels over 24 hours in adrenalectomized rats [30]. mineralocorticoid excess can raise renal kallikrein, acute adPrevious studies have not allowed a consistent conclusion ministration has no effect in intact or adrenalectomized rats [33, about the role of adrenal glucocorticoid in regulating renal 34]. Therefore, it is unlikely that the increased renal kallikrein kallikrein. In one study of intact rats, treatment for four days we observed in adrenalectomized rats is due to mineralocorwith twofold the physiologic replacement dose of corticoster- ticoid deficiency. Long-term testosterone administration can one reduced kidney and urinary kallikrein-like enzymatic activ- lower renal kallikrein in female rats, but has no effect in males ity [13]. In another study, seven day treatment of Dahi salt- [14, 32], and it is unlikely that adrenalectomy reduced circulatresistant rats with dexamethasone doses equivalent to four-fold ing androgen levels significantly in the male rats we studied. physiologic replacement, markedly lowered urinary kallikrein- Taken together with all of our data, the effects of adrenalectomy like esterase activity [18]. A similar dexamethasone dose low- on kallikrein can most likely be attributed to glucocorticoid ered urinary kallikrein-like activity after one day in normal deficiency. adrenal glucocorticoid.

Jaffa et a!: Ka!likrein regulation by glucocorticoids

Finally, because of the diurnal changes in renal kallikrein levels that we demonstrated, the time of day which studies are carried out could influence conclusions about glucocorticoid's effects. During the day, when intact rats have lower glucocorticoid production, kallikrein levels did not differ in adrenalectomized rats and intact rats in our study. However, adrenalectomized rats showed higher renal kallikrein levels in the p.m.,

when plasma corticosterone levels peaked in the intact or sham-operated controls. Although our studies do not define the cellular level at which

glucocorticoid regulates renal kallikrein synthesis, a recent report described a glucocorticoid responsive element in the

2. BARTTER FC, FOURMAN P: The different effects of aldosterone-like

steroids and hydrocortisone-like steroids on urinary excretion of potassium and acid. Metabolism 11:6—20, 1962 3. BEAUWENS R, CRABBE J: Biochemistry of Hormone Action in Renal Biochemistry, edited by KINNE RK, Basel, Elsevier Science Publishers B.V., 1985, pp. 271—336 4. POPOVTZER MM, ROBINETFE J, HALGRJNSON CG, STARZL TE:

Acute effects of prednisolone on renal handling of sodium. Am J Physiol 224:651—658, 1973

5. EL MERNIssI G, DOUCET A: Short-term effects of aldosterone and

dexamethasone on Na-K-ATPase along the rabbit nephron. Pfiugers Arch 399:147—151, 1983

6. SINHA SK, RODRIQUEZ HJ, HOGAN WC, KLAHR S: Mechanisms of

activation of renal (Na + K)-ATPase in the rat: Effects of acute and chronic administration of dexamethasone. Biochim Biophys

5'-flanking region of the rat renal kallikrein gene [35]. Although

the significance of this with regards to regulation of gene expression in the kidney has not been studied, two studies have

Ada 641:20—35, 1981

7. SCHWARTZ M, KOKKO JP: Urinary concentrating defect of adrenal

insufficiency. Permissive role of adrenal steroids on the hydroosmotic response across the rabbit-cortical collecting tubule. J Clin

shown that glucocorticoid administration alters kallikrein mRNA levels in other tissues. In pancreas of adrenalectomized

217

Invest 66:234—242, 1980

rats, kallikrein mRNA levels increase and dexamethasone

8. FARMAN N, VANDEWALLE A, BONVALET JP: Autoradiographic

treatment lowers the levels [36]. Similar glucocorticoid effects have been demonstrated on mRNA of a kallikrein-producing pancreatic cell line [36]. In contrast, Gerald, Chao and Chao

nephron. Am J Physiol 244:F325—F334, 1983 9. STRUM JM, FELDMAN D, TAGGART B, MARVER D, EDELMAN IS:

that submandibular gland kallikrein mRNA de-

to the collecting tubule of the rat kidney. Endocrinology 97:

[371 found

creased following adrenalectomy and rose following glucocorticoid treatment. Taken together with our findings in the kidney, these studies suggest that glucocorticoid's effects on kallikrein regulation are tissue specific. The physiologic significance of such regulation remains un-

clear. Increased cortical collecting duct water permeability induced by glucocorticoid could be related to changes in kallikrein activity, since kinins inhibit ADH-stimulated water flow across this segment [7, 23, 24]. Relating renal hemodynamic actions of glucocorticoid to kallikrein changes is more difficult. Renal vasodilation has been observed following chronic pharmacologic doses of glucocorticoid [1], but acute physiologic doses produce no change in renal hemodynamics in

intact rats [38]. The relevance of these observations to the reduction in kallikrein produced with acute physiologic doses is

determination of dexamethasone binding sites along the rabbit Autoradiographic localization of corticosterone receptors (type III) 505—516, 1975

10. MI5HINA T, SCHOLER DW, EDELMAN IS: Glucocorticoid receptors

in rat kidney cortical tubules enriched in proximal and distal segments. Am J Physiol 240:F38—F45, 1981 11. FUNDER JW, FELDMAN D, EDELMAN IS: Glucocorticoid receptors

in rat kidney: The binding of tritiated-dexamethasone. Endocrinology 92:1005—1013, 1973 12.

cortex. J Biol Chem 250(l4):5596—5603, 1975 13. BONNER 0, AUTENRIETH R, MARIN-GREZ M, SPECK G, GROSS F:

Steroid induced changes of the renal kallikrein-kinin system in rats.

Acta Endocrinol 107:131—140, 1984 14. CHAO J, MARGOLIUS HS: Differential effects of testosterone, thy-

roxine, and cortisol on rat submandibular gland versus renal kallikrein. Endocrinology 113:2221—2225, 1983 15. FULLER PJ, CLEMENTS JA, NIKOLAIDIS I, HIWATARI M, FUNDER

JW: Expression of the renal kallikrein gene in mineralocorticoidtreated and genetically hypertensive rats. J Hypertens 4:427—433,

not clear. It is interesting that while glucocorticoid rapidly lowers tissue synthesis of kallikrein in the kidney, potentially reducing a vasodepressor of kidney origin, renin gene expression may be increased by glucocorticoid, resulting in greater activity of a vasopressor [39]. Data continue to accumulate which suggest that reduced renal kallikrein activity contributes to the pathogenesis of hypertension [40]. Therefore, our findings may be relevant to the mechanisms of hypertension induced by glucocorticoids [16]. Acknowledgments This work was supported by a Merit Review Grant from the Veterans

Administration and a grant from the National Institutes of Health (NIDDK-DK35977). Dr. Silva was supported by a Fogarty Fellowship from the National Institutes of Health. We are also grateful to Dr. Julie Chao for the generous gifts of purified kallikrein and kallikrein antisera and antibodies. We thank Michael Bigelow for technical assistance and Ms. Debra Riebe for assistance in preparing the manuscript.

Reprint requests to Dr. Ronald K. Mayfield, Department of Medicine, 171 Ashley Avenue, Charleston, South Carolina 29425, USA.

References 1. BAYLIS C, BRENNER BM: Mechanism of glucocorticoid-induced

increase in glomerular filtration rate. Am J Physiol 234:F166—F170, 1978

IYNEDJIAN PB, BALLARD JF, HANSON RW: The regulation of phosphoenolpyruvate carboxykinase (GTP) synthesis in rat kidney

1986 16.

HANDA M, KONDO K, SUZUKI H, SARUTA T: Urinary prostaglan-

din E2 and kallikrein excretion in glucocorticoid hypertension in rats. Clin Sci 65:37—42, 1983 17. NODA Y, YAMADA K, lciic R, Eiwos EG: Regulation of rat urinary

and renal

kallikrein and prokallikrein by corticosteroids. Proc

Nat!

Acad Sd USA 80:3059—3063, 1983

18. RAPP JP, SUSTAR5Ic DL, MCPARTLAND RP, BATTEN CL: Hor-

monal effects on kallikrein, esterase A2, and their inhibitors in rat urine. Endocrinology 114:951—956, 1984 19. Vio CP, FIGUEROA CD: Subcellular localization of renal kallikrein by ultrastructural immunocytochemistry. Kidney mt 28:36—42, 1985 20. JAFFA AA, MILLER DH, BAILEY GS, CHA0 J, MARGOLIUS HS, MAYFIELD RK: Abnormal regulation of renal kallikrein in experi-

mental diabetes: Effects of insulin on prokallikrein synthesis and activation. J Clin Invest 80:1651—1659, 1987 21. MILLER DH, CHAO J, MARGOLIUS HS: Tissue kallikrein synthesis

and its modifications by testosterone on low dietary sodium. Biochem J 218:37—43, 1984 22. CUTHBERT AW, GEORGE AM, MACVINISH LM: Kinin effects on electrogenic ion transport in primary cultures of pig renal papillary collecting tubule cells. Am J Physiol 249:F439—F447, 1985

23. TOMITA K, PISANO JJ, KNEPPER MA: Control of Na and K

transport in the cortical collecting duct of the rat. Effect of bradykinin,

vasopressin and deoxycorticosterone. J Clin Invest

76:132—136, 1985

24. SCHUSTER VL, KOKKO JP, JACOBSON HR: Interactions of lysyl-

218

Jaffa et a!: Kal!ikrein regulation by glucocorticoids

bradykinin and antidiuretic hormone in the rabbit cortical collecting tubule. J Clin Invest 73:1659—1667, 1984

33. MILLER DH, LINDLEY JG, MARGOLIUS HS: Tissue kallikrein levels

for the measurement of rat tissue kallikrein using a monoclonal

and synthesis rates are not changed by an acute physiological dose of aldosterone (42152). Proc Soc Exp Biol Med 180:121—125, 1985 34. NASJLETTI A, MCGIFF JC, C0LINA-CHOuRIO J: Interrelations of the renal kallikrein-kinin system and renal prostaglandins in the conscious rat. Influence of mineralocorticoids. Circ Res 43:799— 806, 1978

antibody which recognizes only active kallikrein. Adv Exp Biol Med

35. CHAO L, CHEN Y-P, W000LEY-MILLER C, LI L, VON HARTEN K,

25. SCHNERMANN J, BRIGGS JP, SCHUBERT G, MARIN-GREZ M: opposing effects of captopril and aprotinin on tubuloglomerular feed-

back responses. Am J Physiol 247:F912—F918, 1984 26. ANDO T, CHAO J, CHAO L, MARGOLIUS HS: An improved method 19811:515—522, 1986

27. WOODLEY CM, CHAO J, CHAO L: Immunological analysis of rat

pancreatic prokallikrem activation. Biochim Biophys Ada 829: 408—414, 1985

28. LOWRY OH, ROSEBROUGH NJ, FARR AL, RANDALL Ri: Protein

measurement with the folin phenol reagent. J Biol Chem 193: 265—275, 1951

29. MOBERG GP, BELLINGER LL, MENDEL YE: Effect of meal feeding

on daily rhythms of plasma corticosterone and growth hormone in the rat. Neuroendocrinology 19:160—169, 1975 30. TOMAS FM, MUNRO HN, YOUNG VR: Effect of glucocorticoid administration on the rate of muscle protein breakdown in vivo in

rats, as measured by urinary excretions of N'-methylhistidine.

CHAO J: Structural analysis of a rat renal kallikrein gene. Adv Exp Med Biol 247A:73—80, 1989

36. RosEwIcz S, LOGSDON CG: Pancreatic kallikrein gene expression:

Effects of glucocorticoids in vivo and in vitro. Gastroenterology 97:1005—1010, 1989

37. GERALD WL, CHAO J, CHAO L: Sex dimorphism and hormonal regulation of rat tissue kallikrein mRNA. Biochim Biophys Acta 867:16—23, 1986

38. DEBERMUDEZ L, HAYSLETr JP: Effect of methylprednisolone on

renal function and the zonal distribution of blood flow in the rat. Circ Res 31:44—51, 1972 39. FRITZ LC, PONTE PA, ARFSTEN AE, CATANZARO DF, ROUSSEAU

GG, ELIARD PH, MELLON S, CUMINO F, SHINE J, BAXTER JD,

Biochem J 178:139—146, 1979 31. FEJES-TOTH 0, NARAY-FEJES-T0TH A: Effect of aldosterone on

REUDEL-HUBER TL: Expression of the human renin gene. (abstract) Gun Res 34:643A, 1986

urinary kallikrein excretion and prostaglandin excretion in the rat.

40. YAMAJI I, MILLER DH, KHAN I, RAPP JP, MARGOLIUS HS: Tissue

J Physiol 354:79—88, 1984 32. VANLEEUWEN BH, GRINBLAT SM, JOHNSTON CL: Tissue specific control of glandular kallikreins. Am J Physiol 247:F760—F764, 1984

kallikrein in young or neonatal Dahi salt-sensitive and salt-resistant

rats. Kinin '87, Tokyo International Congress, Tokyo, Japan, November 29—December 3, (abstract) 1987, p. 120