Kidney International, Vol. 59 (2001), pp. 645–653

VASCULAR BIOLOGY – HEMODYNAMICS – HYPERTENSION

Different effects of angiotensin II and catecholamine on renalcell apoptosis and proliferation in rats

TORU AIZAWA,1 NOBUKAZU ISHIZAKA,1 KIYOSHI KUROKAWA, RYOZO NAGAI,HIROYOSHI NAKAJIMA, JUN-ICHI TAGUCHI, and MINORU OHNO

Department of Cardiovascular Medicine, University of Tokyo Graduate School of Medicine andDepartment of Internal Medicine, Tokai University School of Medicine, Tokyo, Japan

Different effects of angiotensin II and catecholamine on renal dent cells and leukocytes within the inflammatory region.cell apoptosis and proliferation in rats. Apoptosis may also underlie the pathology of a variety

Background. We have recently found that chronic infusion of renal diseases, such as ischemia/reperfusion injury [1],of angiotensin II (Ang II) into rats resulted in an impairmentglomerulonephritis [2], and obstructive uropathy [3]. Itof renal function, whereas norepinephrine (NE) infusion didis postulated that an imbalance between proliferationnot. We investigated whether chronic infusion of Ang II and

NE caused different degrees of renal cell apoptosis and prolif- and apoptosis plays a role in renal damage, which woulderation. be histologically manifested by renal cellular hypertro-

Methods. Rats were made hypertensive via continuous infu- phy [4] or atrophy [5, 6].sion of either Ang II or NE for up to seven days. Renal cellThere are several proximal mediators that initiate theapoptosis and proliferation were analyzed by the terminal

cascade of apoptosis in response to various renal insults.deoxynucleotidyl transferase-mediated dUTP-biotin nick endlabeling (TUNEL) technique and staining with antibody against These include inflammatory cytokines [7], the lipid-signal-proliferating cell nuclear antigen (PCNA), respectively. In ing molecule ceramide [8], and various types of reactivesome experiments, an inducer or inhibitor of heme oxygenase-1 oxygen species [9].(HO-1) was administered to investigate the possible role of HO-1

The finding that angiotensin II subtype 1 (AT1) recep-in renal cell homeostasis.tor blockade resulted in a decrease in apoptosis in theResults. Infusion of Ang II, but not NE, resulted in approxi-

mately a sevenfold increase in bax protein at seven days of kidneys of rats receiving cyclosporine suggests that acti-infusion. The TUNEL assay revealed that Ang II infusion sig- vation of the renin-angiotensin system (RAS) may alsonificantly increased the number of apoptotic cells, whereas NE contribute to the development of apoptosis [10] as wellinfusion did not. TUNEL- and PCNA-positive cells were mainly

as cell proliferation [11]. In the heart, an active role of theseen in the tubulointerstitial region of Ang II-infused rats. Ang IIRAS in apoptosis has been demonstrated more directly byinduced increased positivity of TUNEL, and PCNA was blocked

completely by losartan, but only partially by hydralazine. Induc- the finding that angiotensin II (Ang II) administrationtion of HO-1 reduced and inhibition of HO increased Ang II- induces apoptosis [12]. On the other hand, the findingsinduced cell proliferation. that apoptosis is increased in the target organs of geneti-Conclusions. These data suggest that Ang II plays a pivotal

cally hypertensive animals [13, 14] and in vein graftsrole in the development of renal cell proliferation and apoptosisexposed to biomechanical stress [15] suggest that me-in the setting of hypertension. The renal HO system may modu-

late proliferative and pro-apoptotic effects of Ang II. chanical or hemodynamic stress have a role in apoptosis.We have recently reported that Ang II infusion for

seven days decreased glomerular filtration ratio and in-Apoptosis exerts a beneficial effect by regulating the creased urinary protein excretion, whereas norepineph-

number of renal cells through deleting both excess resi- rine (NE) infusion did not, irrespective of its comparablehypertensive effect [16]. The purpose of the presentstudy was to investigate whether cell apoptosis and pro-

1 Drs. Aizawa and Ishizaka contributed equally to this study. liferation would be induced in the kidneys of rats madehypertensive by chronic administration of either Ang IIKey words: oxidative stress, blood pressure, heme oxygenase-1, cellular

hypertrophy, cell death, interstitial cell apoptosis, hydralazine. or NE in order to assess whether renal functional impair-ment and renal cell apoptosis/proliferation were seen inReceived for publication April 25, 2000a parallel manner. We have also found that heme oxy-and in revised form July 19, 2000

Accepted for publication August 11, 2000 genase-1 (HO-1), a heme-degrading enzyme that has an-tioxidant and anti-inflammatory properties, is up-regu- 2001 by the International Society of Nephrology

645

Aizawa et al: Apoptosis and proliferation in hypertension646

lated in the kidneys of rats with administration of Ang II, used for detection. Bands were visualized and quanti-tated using lumino-analyzer (LAS-1000; Fuji Photo Film,but not NE [16]. Since HO-1 may have anti-apoptotic

[17] as well as antiproliferative properties [18], we also Tokyo, Japan).investigated the possible role of HO system in the regula-

Immunohistochemistrytion of renal cell proliferation and apoptosis in hyperten-sive rats. Immunohistochemistry was performed following the

protocol described previously [20]. Briefly, deparaffin-ized sections were preincubated with 10% horse serum.

METHODSSections were then incubated with the antibodies against

All of the protocols used in the present study were bax (Santa Cruz), HO-1 (StressGen, Victoria, BC, Can-approved by the Animal Research Committee of the ada), or proliferating cell nuclear antigen (PCNA; Dako,University of Tokyo (Tokyo, Japan). Carpinteria, CA, USA) at a 1:200 dilution at 378C for one

hour. Slides were then washed with phosphate-bufferedAnimal models

saline (PBS) three times and incubated with biotinylatedThe rat hypertension model was created in male secondary antibody at a 1:1000 dilution. After treating

Sprague-Dawley rats (Nippon Bio-Supply Center, Tokyo, the slides with the Elite ABC kit (Vector Laboratories,Japan) by subcutaneous implantation of an osmotic mini- Burlingame, CA, USA), antigens were visualized usingpump (Alzet model 2001; Alza, Palo Alto, CA, USA) as the 3,3-diaminobenzidine tetrahydrochloride (Dako) sys-described previously [19]. Val5-Ang II (Sigma, St. Louis, tem. Counterstaining was performed with methyl greenMO, USA) was infused at a rate of 0.7 mg/kg/day, and (Dako). The number of PCNA-positive cells in eachNE was infused at a rate of 2.8 mg/kg/day for up to specimen was calculated in a blinded fashion by countingseven days [19]. Systolic blood pressure was measured the number of PCNA-positive cells in 20 nonoverlappingin conscious rats (N 5 9) by tail-cuff plethysmography fields viewed at 3250 magnification and expressed as(UR-5000; Ueda Seisakusyo, Tokyo, Japan). In some the mean number 6 SEM per high-power field (hpf).experiments, the selective AT1 receptor antagonist losar-

Determination of apoptosistan (25 mg/kg/day; a kind gift from Dr. R.D. Smith,Dupont/Merck, Boston, MA, USA) or the nonspecific Paraffin-embedded sections (5 mm in thickness) fromvasodilator hydralazine (15 mg/kg/day; Sigma) was given rat kidneys were examined via the terminal deoxynucleo-in the drinking water, beginning two days before pump tidyl transferase (TdT)-mediated dUTP-biotin nick endimplantation and continued until the time of sacrifice as labeling (TUNEL) technique using an Apoptosis Detec-described previously. All rats were fed a standard chow tion Kit (Wako Chemicals, Osaka, Japan) as describedfor rats, MF (Oriental Yeast, Tokyo, Japan). The amount previously [21], except with minor modifications. Briefly,of drinking water did not apparently differ among the sections were treated with 20 mg/mL of proteinase Krats with various treatments. in PBS (sodium phosphate 50 and sodium chloride 200

Some rats were subjected to daily intraperitoneal in- mmol/L, pH 7.4) for 10 minutes and incubated at 378Cjections of the HO-1 inducer hemin (50 mmol/kg/day; for 30 minutes in TdT enzyme. The sections were thenSigma) or HO-1 inhibitor zinc-protoporphyrin (ZnPP; immersed in 3% hydrogen peroxide in methanol for five50 mmol/kg/day; Porphyrin Products, Logan, UT, USA), minutes. After a brief wash in PBS (3 times), slides werewhich was started two days before pump implantation treated with antidigoxigenin antibodies tagged withand continued until sacrifice as described previously [16]. horseradish peroxidase for 10 minutes. Visualization of

labeled, incorporated nucleotides was carried out withProtein purification and Western blot analysis diaminobenzidine (Dako). Slides were then counter-

Protein was isolated by homogenization of the samples stained with 1% methyl green. The number of TUNEL-within the lysis buffer [50 mmol/L HEPES, 5 mmol/L ethyl- positive cells in each specimen was calculated in aenediaminetetraacetic acid (EDTA), and 50 mmol/L NaCl, blinded fashion by counting the number of PCNA-posi-pH 7.5) containing protease inhibitors [10 mg/mL aproti- tive cells in 20 nonoverlapping fields viewed at 3250nin, 1 mmol/L phenylmethylsulfonyl fluoride (PMSF), magnification and expressed as the mean number 6 SEMand 10 mg/mL leupeptin], and Western blot analysis was per hpf.performed as described previously [19]. Antibodies against

DNA ladder assaybax (Santa Cruz, Santa Cruz, CA, USA) and bcl-2 (Trans-duction Laboratory) were used at a 1:1000 dilution, and Tissue samples were homogenized in a buffer [10horseradish-conjugated secondary antibody (Jackson Im- mmol/L Tris, pH 8.0, 100 mmol/L NaCl, 25 mmol/Lmunoresearch, West Grove, PA, USA) was used at a EDTA, 0.5% sodium dodecyl sulfate (SDS), and 1.01:2000 dilution. The ECL Western blotting system (Am- mg/mL proteinase K] and incubated at 378C overnight.

Samples were then treated with RNase A (Sigma) forersham Life Sciences, Arlington Heights, IL, USA) was

Aizawa et al: Apoptosis and proliferation in hypertension 647

Fig. 1. Effect of continuous infusion of angio-tensin II (Ang II) on blood pressure and ex-pression of apoptosis-related proteins. (A)Time course of systolic blood pressure inAng II-infused rats (N 5 9). Ang II was contin-uously infused at a rate of 0.7 mg/kg/day forthe indicated number of days. (B, C, E, andF) Effects of angiotensin II (Ang II) on apo-ptosis-related protein expression. After Ang IIwas infused for specified time periods, the kid-neys were harvested, and the expression ofbax (B and E) and bcl-2 (C and F) was exam-ined. (B and C) Data from five to seven ani-mals are summarized in the bar graphs. (Dand E) Representative immunoblots. (F) Coo-massie brilliant blue stained sodium dodecylsulfate-polyacrylamide gels of protein samplesfrom the kidneys of rats that had receivedcontinuous angiotensin II infusions for the in-dicated number of days. The same sampleswere used for the representative pictures ofimmunoblot analysis of bax and bcl-2 expres-sion and Coomassie brilliant blue staining.*P , 0.05 and †P , 0.01 vs. controls, respec-tively.

two hours at 378C, and genomic DNA was precipitated RESULTSwith ethanol. Equal amounts of DNA samples were Effect of angiotensin II on blood pressure andloaded onto a 1% agarose gel, and the gel was stained apoptosis-related proteinswith ethidium bromide. Infusion of Ang II effectively increased blood pressure

(Fig. 1A). At first, we examined whether infusion of Ang IIStatistical analysis changed expression of pro-apoptotic (bax) and anti-apo-

Data are expressed as means 6 SEM. Analysis of ptotic (bcl-2) proteins. An immunoblot showed that baxvariance (ANOVA) followed by a multiple comparison protein expression was markedly up-regulated in the kid-test was used for comparisons of the initial data before neys of Ang II-induced hypertensive rats (Fig. 1 B, D).expression as a percentage of the control using the statis- In contrast, the expression of bcl-2 protein remainedtical analysis software Statistica version 5.1 J for Win- unchanged after Ang II infusion (Fig. 1 C, E). Evendows (StatSoft Inc., Tulsa, OK, USA). A value of P , loading of the protein sample in each lane was confirmed

by the Coomassie staining (Fig. 1F).0.05 was considered statistically significant.

Aizawa et al: Apoptosis and proliferation in hypertension648

Fig. 2. Localization of bax and heme oxygen-ase-1 (HO-1). (A) Hematoxylin-eosin stainingof the kidneys of Ang II-induced hypertensiverats. (B–I) Bax and HO-1 staining were per-formed on serially sectioned specimens. Bax(B and C) and HO-1 (D and E) staining inthe kidney of Ang II-induced hypertensive rats.The majority of the bax-positive cells werealso positive for HO-1 with some discrepan-cies (arrows in D and asterisks in E). Bax (Fand G) and HO-1 (H and I) staining in thekidneys of normotensive rats.

Fig. 4. Representative photographs of TUNEL-positive cells in the kidney of rats receivingeither angiotensin II (Ang II) or norepineph-rine (NE). (A–F) TUNEL staining. (A and D)Kidney specimen of the control rat. (B and E)Kidney specimen of the rat infused Ang IIfor seven days. Several TUNEL-positive nucleiwere observed (arrows in B). Higher magnifi-cation showed that positive TUNEL stainingwas seen in both tubular (arrow in E) andinterstitial (arrowhead in E) regions. (C andF) Kidney specimen of the rat infused NE forseven days. (G–L) PCNA immunohistochem-istry. (G and J) Kidney specimen of the controlrat. (H and K) Kidney specimen of the ratinfused Ang II for three days. Several PCNA-positive nuclei were observed. (I and L) Kid-ney specimen of the rat infused NE for threedays. D, E, F, J, K, and L are magnified viewsof bracketed regions in A, B, C, G, H, and I,respectively. Original magnifications are 3250(A–C and G–I) and 3800 (D–F and J–L).

Aizawa et al: Apoptosis and proliferation in hypertension 649

Immunohistochemistry of bax protein

Immunohistochemistry demonstrated that bax proteinwas present in both the proximal and distal tubular epi-thelial cells (Fig. 2B), but not in the glomeruli (Fig. 2C)in the kidneys of Ang II-induced hypertensive rats. Wepreviously found that intense HO-1 staining is presentin the tubular epithelial cells in the kidney of Ang II-induced hypertensive rats. Staining of the continuoussection revealed that many tubular epithelial cells posi-tive for bax staining were also positive for HO-1 staining(Fig. 2 D, E), although overlapping was not complete(arrows and asterisks in Fig. 2 D, E, respectively). In thekidney of normotensive rats, bax staining was barelydetectable in the tubular epithelial cells (Fig. 2 F, G). Incontrast, distinct HO-1 staining was detectable along thebasal side of tubular epithelial cells (Fig. 2 H, I), asdescribed previously [16].

Effects of antihypertensive drugs and norepinephrineon blood pressure and apoptosis-related proteins

Next, we investigated whether Ang II-induced up-reg-ulation of bax protein was a pressor-dependent or pres-sor-independent event. For this purpose, either the spe-cific AT1 receptor antagonist losartan or the nonspecificvasodilator hydralazine was orally administered to ratsreceiving continuous Ang II infusions. Rats were sacri-ficed at seven days of infusion of either Ang II or NE.Both losartan and hydralazine effectively normalized theblood pressure of Ang II-infused rats (126 6 5 and 133 68 mm Hg, P 5 NS compared with controls, respectively).Ang II-induced up-regulation of bax protein was blockedcompletely by losartan, but only partially by hydralazine(Fig. 3A), suggesting that chronic Ang II infusion in-creased renal bax protein levels via an AT1 receptor-mediated manner, and only partially via a pressor-depen-dent manner. Continuous Ang II infusion at a dose of0.25 mg/kg/day for seven days, which did not significantly

Fig. 3. Effects of antihypertensive drugs, subpressor doses of angioten-increased blood pressure (137 6 3 mm Hg, P 5 NSsin II (Ang II) and norepinephrine (NE) on apoptosis-related proteincompared with control), did not change renal bax expres- expression. (A, left) Effects of losartan (Los) and hydralazine (Hyd).

sion. Interestingly, NE infusion for seven days, irrespec- Losartan completely blocked Ang II-induced bax up-regulation, whilehydralazine only partially blocked such an up-regulation. (Right) Effectstive of its comparable hypertensive effect on Ang IIof NE and a subpressor dose of Ang II (0.25 mg/kg/day). NE infusion(196 6 6 mm Hg, P , 0.01 and P 5 NS compared with did not increase renal bax expression. (B) Effect of vasopressors and

the control and Ang II group, respectively), did not up- vasodilators on bcl-2 expression in the kidney. Expression of bcl-2 ineach group displays a relatively large variation, and neither vasopressorsregulate bax protein. Neither Ang II (0.7 or 0.25 mg/kg/nor vasodilators significantly alter the bcl-2 expression within the kid-

day), losartan, hydralazine, nor NE significantly changed ney. (A and B, upper panels) Representative immunoblots. (A and B,lower panels) Data from four to six animals are summarized in a barbcl-2 protein levels (Fig. 3B).graph. *P , 0.01 vs. sham-operated controls; †P , 0.05 vs. Ang II-infused rats.Determination of apoptosis by TUNEL assay

Next, the number of apoptotic cells in the kidney wasquantitated using TUNEL technique. In addition tobrown staining, apoptotic nuclei were identified by con- significantly increased after three days of Ang II (0.7densed nuclear material and vacuolization of the cyto- mg/kg/day) infusion (P , 0.01 compared with control)

and further increased after seven days of Ang II infusionplasm (Fig. 4 B, E). Within the both tubular and intersti-tial regions, the number of TUNEL-positive cells was (P , 0.01 compared with control and day 3; Table 1).

Aizawa et al: Apoptosis and proliferation in hypertension650

Table 2. Effect of angiotensin II and norepinephrine infusion onTable 1. Effect of angiotensin II and norepinephrine infusion ontubular and interstitial cell apoptosis tubular and interstitial cell proliferation

Tubular region Interstitial regionTubular region Interstitial regionGroupGroup 0.660.1 1.460.3 1.160.2 1.460.4ControlControl

Treatment day 3 day 7 day 3 day 7 Treatment day 3 day 7 day 3 day 7

Ang II 0.7 mg/kg/day 13.162.4b 5.661.8b 16.062.7b 5.560.6bAng II 0.7 mg/kg/day 2.760.9a 5.161.8b 3.961.6b 6.262.0b

NE 0.560.3c 0.460.1c 1.460.7c 1.660.1d NE 1.160.3d 1.260.4c 1.260.2d 2.360.5a,d

Hydralazine ND 0.860.3d ND 1.060.2dHydralazine ND 0.460.1c ND 1.060.3d

Hydralazine1Ang II 1.360.3a,c 2.860.5b,c 2.260.9 4.160.6b,c Hydralazine1Ang II 12.863.4b 4.760.9b 13.262.6b 8.161.0b

Losartan ND 1.260.2d ND 1.360.4dLosartan ND 0.360.1d ND 1.460.5d

Losartan1Ang II ND 0.460.1d ND 1.360.8d Losartan1Ang II ND 1.260.0d ND 1.760.5c

Low Ang II ND 1.060.1d ND 1.460.2dLow Ang II ND 0.360.2d ND 1.860.1d

Angiotensin II (Ang II) or norepinephrine (NE) was continuously infusedAngiotensin II (Ang II) or norepinephrine (NE) was continuously infusedfor 7 consecutive days. The number of TUNEL-positive cells in each specimen for 7 consecutive days. The number of PCNA-positive cells in each specimen

was calculated by counting the number of PCNA-positive cells in 20 nonoverlap-was calculated by counting the number of TUNEL-positive cells in 20 nonover-lapping fields viewed at 3250 magnification. Data from 6 to 8 rats are summarized ping fields viewed at 3250 magnification. Data from 6 to 8 rats are summarized

and expressed as the mean number 6 SEM per high power field. ND indicatesand expressed as the mean number 6 SEM per high power field. ND indicatesnot done. not done.

a P , 0.05 and b P , 0.01 compared with the controla P , 0.05 and b P , 0.01 compared with the controlc P , 0.05 and d P , 0.01 compared with Ang II (0.7 mg/kg/day)-infused rats c P , 0.05 and d P , 0.01 compared with Ang II (0.7 mg/kg/day)-infused rats

control rats or rats receiving NE or subpressor doses ofIn contrast, chronic infusion of NE did not increase theAng II (Fig. 5B), which were consistent with the findingnumber of TUNEL-positive cells at either time point.of TUNEL staining.Losartan completely, but hydralazine only partially,

blocked the Ang II-induced increase in TUNEL-positiveEffects of inducer and inhibitor of HO on renalcells in the tubulointerstitial region.cell apoptosis

Immunohistochemical analysis of renal In a previous article, we demonstrated that administra-cell proliferation tion of hemin, an HO inducer, ameliorated and the ad-

ministration of ZnPP, an HO inhibitor, exacerbated anWe also quantitated the number of proliferating cellsAng II-induced decrease in glomerular filtration ratein the kidney by immunohistochemical analysis (Fig. 4and an increase in proteinuria [16]. Therefore, we nextH, K). After either three or seven days of infusion, Ang IIinvestigated the effects of hemin and ZnPP on Ang II-increased the number of PCNA-positive cells (P , 0.01induced increase in cell apoptosis and proliferation.compared with control), and the number of PCNA-posi-

TUNEL staining demonstrated that ZnPP administra-tive cells was significantly greater at day 3 than at day 7tion did not significantly changed Ang II-induced increaseof Ang II infusion (P , 0.01; Table 2). NE infusion didin TUNEL-positive cells in tubulointerstitial region. Innot result in enhanced PCNA staining after three days ofcontrast, hemin administration significantly suppressedinfusion in tubulointerstitial region. NE infusion slightly,Ang II-induced increase in TUNEL-positive cells at ei-but significantly, increased the number of PCNA-positivether three or seven days of Ang II infusion (Table 3).cells after seven days of infusion in the interstitial regionThe number of TUNEL-positive cells in the kidney of(Table 2), although the extent of such staining was signifi-rats receiving both Ang II and hemin was also signifi-cantly less than that seen in Ang II-infused rats (P ,

0.01). Again, losartan completely, but hydralazine only cantly smaller than that in the kidney of rats receivingboth Ang II and hydralazine at day 7 of Ang II infusion,partially, blocked Ang II-induced increase in PCNA-posi-

tive cells in tubulointerstitial region. In the glomerular suggesting that a factor(s) other than antihypertensiveproperties may mediate the anti-apoptotic effect of hemin.region, none of the treatments significantly increased the

number of TUNEL- or PCNA-positive cells comparedEffects of inducer and inhibitor of HO on renalwith the control animals (data not shown).cell proliferation

Detection of chromosomal DNA fragmentation Immunohistochemistry demonstrated that ZnPP ad-ministration significantly increased PCNA-positive cellsAs illustrated (Fig. 5), chromosomal DNA from the

kidneys of Ang II-induced hypertensive rats showed a at both three and seven days (tubular region) and at day 7(interstitial region) of Ang II infusion. In contrast, hemintypical DNA electrophoretic pattern on an agarose gel.

DNA fragmentation was completely blocked by losartan, administration significantly decreased Ang II-inducedincrease in PCNA-positive cells at three days of Ang IIbut not by hydralazine (Fig. 5A). No DNA fragmenta-

tion could be detected in the kidneys obtained from infusion (Table 4).

Aizawa et al: Apoptosis and proliferation in hypertension 651

Table 3. Effect of hemin and zinc protoporphyrin on angiotensin II-induced tubular and interstitial cell apoptosis

Tubular region Interstitial regionGroup 0.660.1 1.460.3ControlTreatment day 3 day 7 day 3 day 7

Ang II 0.7 mg/kg/day 2.760.9a,e 5.161.8b,e 3.961.6a 6.262.0b,e

Hydralazine1Ang II 1.360.3a,c 2.860.5b,c 2.260.9 4.160.6b,c

Hemin ND 0.460.1d,f ND 1.060.2d,f

Hemin1Ang II 1.360.2a,c 1.360.2a,d,f 1.860.3d 1.660.2d,e

ZnPP ND 1.060.2d,e ND 1.160.2d,f

ZnPP1Ang II 2.060.1a 6.761.0b,f 2.960.2a 7.160.6b,f

Angiotensin II (Ang II) was continuously infused for 7 consecutive days. Thenumber of TUNEL-positive cells in each specimen was calculated by countingthe number of TUNEL-positive cells in 20 nonoverlapping fields viewed at 3250

Fig. 5. Ang II-induced DNA fragmentation of the kidneys of rats re- magnification. Data from 6 to 8 rats are summarized in the table and expressedceiving vasopressors and/or vasodilators. Genomic DNA samples puri- as the mean number 6 SEM per high power field. ND indicates not done.

a P , 0.05 and b P , 0.01 compared with the controlfied from the kidney of rats receiving the indicated drug(s) were loadedc P , 0.05 and d P , 0.01 compared with Ang II (0.7 mg/kg/day)-infused ratsonto 1.5% agarose gel. (A) Rats were continuously infused pressore P , 0.05 and f P , 0.01 compared with hydralazine 1 Ang II groupdoses (0.7 mg/kg/day) of Ang II for seven days with or without receiving

vasodilators. Losartan completely inhibited the “ladder” formation in-duced by continuous Ang II infusion. (B) Subpressor doses (0.25 mg/kg/day) of Ang II or NE did not result in DNA ladder formation.

opment of renal cell apoptosis. Whether Ang II and NEactually have different potentials in inducing reactiveoxygen species in the kidney of hypertensive animals, asDISCUSSIONhas been shown in the rat aortas [26], should be studiedIn the present study, we demonstrated that chronicfurther in the future. Second, it should be kept in mindinfusion of Ang II at a pressor dose increased renal cellthat increases in the number of PCNA-positive cells inapoptosis in the rat kidney. This Ang II-induced cellthe tubulointerstitial region preceded increases in apo-apoptosis was blocked completely by losartan, but onlyptosis in the tubulointerstitial region in Ang II-infusedpartially by hydralazine. In contrast, NE infusion, irre-rats. Therefore, increased apoptosis may be, at least par-spective of its comparable hypertensive effect, did nottially, a compensatory mechanism in response to cellularresult in enhanced apoptosis, suggesting that Ang II playsproliferation. Different effect of Ang II and NE in renala pivotal role in renal cell apoptosis in the setting ofcell proliferation may be explained by proteinuric effecthypertension.of Ang II, since the mitogenic pathway in tubular epithe-Different effects on renal cell apoptosis resulting fromlial cells is activated in proteinuric states [27]. However,Ang II and NE administration may be explained in threethe finding that hydralazine treatment did not decreaseways. First, in the previous set of experiments, we foundthe Ang II-induced enhanced cell proliferation indicatesthat continuous infusion of Ang II, but not NE, increasedthat increased proteinuria was not solely responsible forproteinuria [16]. Since proteinuria induces tubular cellthe Ang II-induced increases in PCNA positivity. Ang IIapoptosis in rats [22], different apoptotic effects betweenmay have enhanced renal cell proliferation via the directAng II and NE may be derived from their differentaction of Ang II [28] or Ang II-induced increase in super-effects with regard to increasing proteinuria. On theoxide production [23]. It has been reported that a major-other hand, we previously showed that hydralazine treat-ity of apoptotic cells are actually within the phagolyso-ment completely normalized Ang II-induced increase insomes of adjacent cells or infiltrating macrophages [29];proteinuria [16]. Since hydralazine only partially sup-therefore, it is difficult to assess whether apoptosis out-pressed Ang II-induced renal cell apoptosis in the pres-weighed proliferation. However, since Ang II infusionent study, a factor(s) other than proteinuria, such asmay cause tubular atrophy as well as cell proliferationincreased superoxide [23], may play a role in renal cell[11], an imbalance between cell proliferation and apopto-apoptosis in the kidney of Ang II-infused rat. We foundsis may play a role in Ang II-induced renal damage.that some of the tubular epithelial cells express highThird, it is possible that Ang II-induced microvascularlevels of both HO-1 and bax in Ang II-induced hyperten-injury with peritubular capillary loss [30] resulted in ansive rats. HO-1 is known to be induced by various reac-enhanced apoptosis that eventually will lead to tubuloin-tive oxygen species [24, 25] and therefore represents aterstitial injury.reliable marker of oxidative stress. The findings that

We also found that induction of HO-1 decreased andAng II administration causes increases in HO-1 and baxinhibition of HO increased the incidence of tubulointer-expression and in the number of apoptotic cells and thatstitial apoptosis. Other investigators have demonstratedNE administration causes neither of these phenomena

suggests a possible role for oxidative stress in the devel- that induction of the HO system has an anti-apoptotic

Aizawa et al: Apoptosis and proliferation in hypertension652

Table 4. Effect of hemin and zinc protoporphyrin on angiotensin II-induced tubular and interstitial cell proliferation

Tubular region Interstitial regionGroup 1.160.2 1.460.4ControlTreatment day 3 day 7 day 3 day 7

Ang II 0.7 mg/kg/day 13.162.4b 5.661.8b 16.062.7b 5.560.6b

Hydralazine1Ang II 12.863.4b 4.760.9b 13.262.6b 6.260.8b

Hemin ND 1.060.3c,f ND 1.460.3d,f

Hemin1Ang II 3.460.9b,d,f 4.661.0b 3.260.6b,d,f 4.960.6b,e

ZnPP ND 1.360.4c,f ND 1.360.4d,f

ZnPP1Ang II 19.561.9b,c,e 9.362.0b,c,f 18.662.3b,e 9.360.9b,c,e

Angiotensin II (Ang II) was continuously infused for 7 consecutive days. The number of PCNA-positive cells in each specimen was calculated by counting thenumber of PCNA-positive cells in 20 nonoverlapping fields viewed at 3250 magnification. Data from 6 to 8 rats are summarized and expressed as the mean number 6SEM per high power field. ND indicates not done.

a P , 0.05 and b P , 0.01 compared with the controlc P , 0.05 and d P , 0.01 compared with Ang II (0.7 mg/kg/day)-infused ratse P , 0.05 and f P , 0.01 compared with hydralazine 1 Ang II group

effect in vitro and in vivo [17, 31]. HO-1 is thought to Ang II-infused rats suggests that oxidative stress may beinvolved in the pathogenesis of Ang II-induced apopto-exert its anti-apoptotic effect by regulating cellular ironsis. Hemin injection decreased the Ang II-induced cell[17] or carbon monoxide [32]. In a previous article, weproliferation and apoptosis, and ZnPP injection increaseddemonstrated that hemin injection normalized Ang II-these phenomena in the kidney. Our data suggest thatinduced elevation of blood pressure [16]. Since hydrala-activation of RAS plays a pivotal role in modulating renalzine treatment significantly, though not completely, sup-cell proliferation and apoptosis in the setting of hyper-pressed Ang II-induced enhancement of interstitial celltension and that the renal HO system may play a roleapoptosis, an anti-apoptotic effect of hemin administra-in maintaining normal kidney homeostasis in Ang II-tion observed in the present study may be partially ex-infused rats.plained by its antihypertensive effect. Since hemin (ZnPP)

injection decreased (increased) Ang II-induced enhance-ACKNOWLEDGMENTSment of cell proliferation, it is also possible that a de-

crease (increase) in the degree of renal cell apoptosis in A part of the present study was presented at International Congressof Nephrology, 1999, held in Argentina. We are highly appreciativehemin (ZnPP) injected rats may be secondary to theof Ms. Noriko Ogura for her technical assistance and Drs. Akira Hishidaantiproliferative effect of HO-1. The number of PCNA-and Masaomi Nangaku for their constructive comments.

positive cells in the kidney of rats receiving Ang II andReprints requests to Nobukazu Ishizaka, M.D., Ph.D., University ofhemin was significantly smaller than that in the kidney

Tokyo, Graduate School of Medicine, Hongo 7-3-1, Bunkyo-ku, Tokyoof rats receiving Ang II and hydralazine at day 3 of Ang II 113-8655, Japan.infusion; therefore, an antiproliferative effect of hemin E-mail: nobuishizka-tky@umin.ac.jpwas not solely dependent on its antihypertensive proper-ties. It is possible that induction of HO-1 may have ex- REFERENCESerted an antiproliferative effect through generating car- 1. Nogae S, Miyazaki M, Kobayashi N, et al: Induction of apoptosisbon monoxide, as has been demonstrated in cultured in ischemia-reperfusion model of mouse kidney: Possible involve-

ment of Fas. J Am Soc Nephrol 9:620–631, 1998vascular smooth muscle cells [18, 33]. Since induction of2. Savill J, Mooney A, Hughes J: What role does apoptosis playHO-1 ameliorated and inhibition of HO activity exacer- in progression of renal disease? Curr Opin Nephrol Hypertens

bated the renal function of Ang II-induced hypertensive 5:369–374, 19963. Truong LD, Petrusevska G, Yang G, et al: Cell apoptosis andrats [16], it is possible that a renoprotective effect of HO-1

proliferation in experimental chronic obstructive uropathy. Kidneymay be exerted via controlling renal cell proliferation Int 50:200–207, 1996and/or apoptosis in Ang II-infused rats. 4. Nouwen EJ, Verstrepen WA, Buyssens N, et al: Hyperplasia,

hypertrophy, and phenotypic alterations in the distal nephron afterIn summary, chronic infusion of Ang II resulted in anacute proximal tubular injury in the rat. Lab Invest 70:479–493,increased tubulointerstitial cell apoptosis, whereas NE 1994

infusion did not, irrespective of comparable hypertensive 5. Velasquez MT, Striffler JS, Abraham AA, et al: Perindoprilameliorates glomerular and renal tubulointerstitial injury in theeffects. The pro-apoptotic effect of Ang II was AT1 re-SHR/N-corpulent rat. Hypertension 30:1232–1237, 1997ceptor dependent and only partially pressor dependent. 6. Thomas GL, Yang B, Wagner BE, et al: Cellular apoptosis and

Ang II and NE also evoked different proliferative effects proliferation in experimental renal fibrosis. Nephrol Dial Trans-plant 13:2216–2226, 1998in the kidney, which may partially explain the different

7. Donnahoo KK, Shames BD, Harken AH, et al: Review article:pro-apoptotic effects of these vasoactive substances. Co- The role of tumor necrosis factor in renal ischemia-reperfusion

injury. J Urol 162:196–203, 1999localization of HO-1 and bax proteins in the kidneys of

Aizawa et al: Apoptosis and proliferation in hypertension 653

8. Huwiler A, Pfeilschifter J, van den Bosch H: Nitric oxide do- but not tumor necrosis factor receptor 1 prevents chronic diseasein mice infected with murine cytomegalovirus. J Clin Invest 102:nors induce stress signaling via ceramide formation in rat renal1431–1443, 1998mesangial cells. J Biol Chem 274:7190–7195, 1999

22. Thomas ME, Brunskill NJ, Harris KP, et al: Proteinuria induces9. Thevenod F, Friedmann JM: Cadmium-mediated oxidative stresstubular cell turnover: A potential mechanism for tubular atrophy.in kidney proximal tubule cells induces degradation of Na1/K(1)-Kidney Int 55:890–898, 1999ATPase through proteasomal and endo-/lysosomal proteolytic

23. Hannken T, Schroeder R, Stahl RA, et al: Angiotensin II-medi-pathways. FASEB J 13:1751–1761, 1999ated expression of p27Kip1 and induction of cellular hypertrophy in10. Thomas SE, Andoh TF, Pichler RH, et al: Accelerated apoptosisrenal tubular cells depend on the generation of oxygen radicals.characterizes cyclosporine-associated interstitial fibrosis. KidneyKidney Int 54:1923–1933, 1998Int 53:897–908, 1998

24. Nath KA, Croatt AJ, Likely S, et al: Renal oxidant injury and11. Johnson RJ, Alpers CE, Yoshimura A, et al: Renal injury fromoxidant response induced by mercury. Kidney Int 50:1032–1043, 1996angiotensin II-mediated hypertension. Hypertension 19:464–474,

25. Kawada N, Moriyama T, Ando A, et al: Increased oxidative stress1992in mouse kidneys with unilateral ureteral obstruction. Kidney Int12. Diep QN, Li JS, Schiffrin EL: In vivo study of AT(1) and AT(2)56:1004–1013, 1999angiotensin receptors in apoptosis in rat blood vessels. Hyperten-

26. Laursen JB, Rajagopalan S, Galis Z, et al: Role of superoxidesion 34:617–624, 1999in angiotensin II-induced but not catecholamine-induced hyperten-13. Hamet P, Richard L, Dam TV, et al: Apoptosis in target organssion. Circulation 95:588–593, 1997of hypertension. Hypertension 26:642–648, 1995

27. Dixon R, Brunskill NJ: Activation of mitogenic pathways by14. Fortuno MA, Ravassa S, Etayo JC, et al: Overexpression of Baxalbumin in kidney proximal tubule epithelial cells: Implicationsprotein and enhanced apoptosis in the left ventricle of spontane-for the pathophysiology of proteinuric states. J Am Soc Nephrolously hypertensive rats: Effects of AT1 blockade with losartan.10:1487–1497, 1999Hypertension 32:280–286, 1998 28. Chatterjee PK, Weerackody RP, Mistry SK, et al: Selective an-15. Mayr M, Li C, Zou Y, et al: Biomechanical stress-induced apoptosis tagonism of the AT1 receptor inhibits angiotensin II stimulatedin vein grafts involves p38 mitogen-activated protein kinases. DNA and protein synthesis in primary cultures of human proximal

FASEB J 14:261–270, 2000 tubular cells. Kidney Int 52:699–705, 199716. Aizawa T, Ishizaka N, Taguchi J, et al: Heme oxygenase-1 is 29. Coles HS, Burne JF, Raff MC: Large-scale normal cell death in

upregulated in the kidney of angiotensin II-induced hypertensive the developing rat kidney and its reduction by epidermal growthrats: Possible role in renoprotection. Hypertension 35:800–806, 2000 factor. Development 118:777–784, 1993

17. Ferris CD, Jaffrey SR, Sawa A, et al: Haem oxygenase-1 prevents 30. Lombardi D, Gordon KL, Polinsky P, et al: Salt-sensitive hyper-cell death by regulating cellular iron. Nat Cell Biol 1:152–157, 1999 tension develops after short-term exposure to angiotensin II. Hy-

18. Morita T, Mitsialis SA, Koike H, et al: Carbon monoxide controls pertension 33:1013–1019, 1999the proliferation of hypoxic vascular smooth muscle cells. J Biol 31. Otterbein LE, Kolls JK, Mantell LL, et al: Exogenous adminis-Chem 272:32804–32809, 1997 tration of heme oxygenase-1 by gene transfer provides protection

19. Ishizaka N, Leon H, Laursen JB, et al: Angiotensin II-induced against hyperoxia-induced lung injury. J Clin Invest 103:1047–1054,hypertension increases heme oxygenase-1 expression in rat aorta. 1999Circulation 96:1923–1929, 1997 32. Petrache I, Otterbein LE, Alam J, et al: Heme oxygenase-1 inhib-

20. Aizawa T, Ishizaka N, Taguchi J, et al: Balloon injury does not its TNF-alpha-induced apoptosis in cultured fibroblasts. Am Jinduce heme oxygenase-1 expression, but administration of hemin Physiol 278:L312–L319, 2000inhibits neointimal formation in balloon-injured rat carotid artery. 33. Togane Y, Morita T, Suematsu M, et al: Protective roles of endog-Biochem Biophys Res Commun 261:302–307, 1999 enous carbon monoxide in neointimal development elicited by

arterial injury. Am J Physiol 278:H623–32, 200021. Fleck M, Kern ER, Zhou T, et al: Apoptosis mediated by Fas