Expression of Human Heme Oxygenase-1 in the Thick Ascending Limb Attenuates Angiotensin II–Mediated Increase in Oxidative Injury1

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Kidney International, Vol. 65 (2004), pp. 1628–1639

HORMONES – CYTOKINES – SIGNALING

Expression of human heme oxygenase-1 in the thick ascending limb attenuates angiotensin II–mediated increase in oxidative injury1

SHUO QUAN,LIMING YANG,SYLVIA SHNOUDA,MICHAL L. SCHWARTZMAN,ALBERTO NASJLETTI, ALVIN I. GOODMAN, and NADER G. ABRAHAM

Department of Pharmacology and Department of Medicine, Division of Nephrology, New York Medical College, Valhalla, New York; and The Rockefeller University, New York, New York

Expression of human heme oxygenase-1 in the thick ascending selective upregulation of HO-1, thusly blocking inflammation limb attenuates angiotensin II–mediated increase in oxidative and apoptosis. injury. Background. Heme oxygenase-1 (HO-1) catalyzes the con- version of heme to bilirubin, carbon monoxide (CO), and free The thick ascending loop of Henle (TALH) segment iron, thus controlling the level of cellular heme. The medullary of the nephron, residing in the oxygen-poor medulla of thick ascending limb of the loop of Henle (TALH) is situated the kidney, is highly susceptible to oxidants, and tubular in a site of markedly diminished oxygen tension and, as such, is hypoxic and oxidative injury [1]. Extensive DNA frag- highly vulnerable to ischemic insult. We hypothesize that selective upregulation of HO-1 in TALH by gene transfer attenuates mentation has been observed selectively in TALH cells oxidative stress caused by angiotensin II (Ang II). soon after local hypoxia [2]. Increased superoxide lev- Methods. An adenoviral vector expressing the human HO-1 els in TALH have been shown to decrease nitric oxide + + − under the control of the TALH-specific promoter [Na -K -Cl bioavalibilty and stimulate NaCl absorption [3]. An- cotransporter (NKCC2 promoter)] was constructed and the cell giotensin II (Ang II) is a pro-oxidant agent, which induces specific expression of the recombinant adenovirus was examined using several types of cells, including endothelial, vascular renal oxidative stress and predisposes to acute renal fail- smooth muscle, and TALH cells. The effects of HO-1 transduc- ure [4]. Ang II up-regulates heme oxygenase-1 (HO-1) tion on HO-1 expression, HO activity and the response to Ang [5], a stress-responsive protein, as an adaptive response II with respect to cyclooxygenase-2 (COX-2) up-regulation and to injury [6]. oxidative injury [growth-stimulating hormone (GSH) levels and HO is present in the kidney as the constitutive isoform, cell death] were determined. Results. Western blot and reverse transcription-polymerase HO-2, and as the inducible isoform, HO-1. Induction of chain reaction (RT-PCR) revealed that human HO-1 was se- HO-1 has been shown to attenuate tumor necrosis factor- lectively expressed in primary cultured TALH cells follow- a (TNF-a)–mediated cell death by retroviral delivery of ing infection with Ad-NKCC2-HO-1. In TALH cells infected human HO-1 gene into human microvessel endothelial with Ad-NKCC2-HO-1, Ang II-stimulated prostaglandin E2 cells [7, 8]. Cyclooxygenase 2 (COX-2) activity was de- (PGE2) levels were reduced by 40%. Ang II caused a marked decrease in GSH levels and this decrease was greatly attenuated creased, as was p27, a cyclin-dependent kinase inhibitor. in TALH cells transduced with Ad-NKCC2-HO-1. Moreover, HO-1 gene up-regulation has been shown to attenuate Ang II–mediated DNA degradation was completely blocked by Ang II–mediated DNA damage in endothelial cells [9]. the site-specific expression of human HO-1 gene. HO-1 regulates the rate of heme degradation, leading to Conclusion. These results indicate that TALH cell survival equimolar amounts of carbon monoxide (CO), iron and after exposure to oxidative stress injury may be facilitated by biliverdin metabolized to bilirubin, which is a potent antioxidant [10, 11]. CO, in addition to activating the cyclic 1See Editorial by Chang and Perrella, p. 1968 guanosine monophosphate (cGMP) pathway and elicit- ing vasodilation [12], has also been shown to function Key words: heme oxygenase, cyclooxygenase, adenovirus, angiotensin as an anti-inflammatory and cytoprotective molecule [7, II, Na+-K+-Cl− cotransporter, oxidative injury, TALH. 8, 13–16]. CO has also been shown to participate in the regulation of ion channels, including the apical 70-pS K- Received for publication August 27, 2003 and in revised form October 22, 2003 channel activity of the TALH [17]. Accepted for publication December 3, 2003 In view of the importance of HO-1 in preventing the consequences of oxidative stress, we studied the effect C 2004 by the International Society of Nephrology of augmenting HO-1 activity by transducing TALH cells

1628 Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH 1629 with human HO-1 gene and determining if this could pro- with 10% FBS, 10 ng/mL EGF, 1 lL/mL hydrocorti- tect TALH against the toxic effects of free heme and the sone, 0.1 mg/mL ENDO GRO (VEC Technologies, Rens- Ang II–mediated increase in oxidative stress. We utilized selaer, NY, USA), and 90 lg/mL heparin (Sigma, St. an adenoviral vector for human HO-1 with the TALH- Louis, MO, USA) [25, 26]. Rat aorta smooth muscle cells specific promoter, NKCC2. NKCC2 encoded by a distinct (SMC) were purchased from ATCC (American Type Tis- gene is selectively expressed in the TALH and represents sue Culture, Manassas, VA, USA) and were grown in + + the apical Na -K -Cl− cotransporter that mediates ac- DMEM medium with 10% FBS. HEK-293 cells (ATCC) tive reabsorption of NaCl [18]. Recent studies by Ortiz were grown in DMEM supplemented with 10% FBS. ◦ et al [19, 20] have shown elegantly the specificity of the All cells were incubated at 37 Cina5%CO2 humid- NKCC2 promoter in directing TALH-specific expression ified atmosphere, and maintained at subconfluency by of the marker gene green fluorescent protein (GFP) as passaging with trypsin-ethylenediaminetetraacetic acid well as endothelial nitric oxide synthase (eNOS) [19, 20]. (EDTA) (Life Technologies).Cells were treated with var- These authors demonstrated that administration of ade- ious inducers of HO-1, either SnCl2 (10 lmol/L), heme noviruses expressing either GFP or eNOS under the con- (10 lmol/L), or Ang II (100 ng/mL), and the levels of trol of the NKCC2 promoter resulted in their exclusive HO-1 and HO-2, as well as prostaglandin E2 (PGE2), expression in TALH but not in other structures of the were measured. transduced kidney. In this study, we demonstrated that it is possible to Isolation of NKCC2 promoter transduce, in a site-specific fashion, TALH cells with Since the NKCC2 transcript is only detected in the the human HO-1 gene and achieve selective overexpres- kidney [18, 27] and is restricted to TALH, a 1702 bp sion of human HO-1 protein to levels significantly higher NKCC2 promoter was amplified from mouse genomic than those of endogenous rat HO-1 protein. Overexpres- DNA, using the primer sets that overlap the NKCC2 pro- sion of human HO-1 was not associated with changes moter region (−1629 to +48), and cloned into plasmid in endogenous HO-2 expression. We have also shown pCRII (Invitrogen, Carlsbad, CA, USA) as previously that augmentation of human HO-1 activity in transduced described [28], the resultant plasmid was designated TALH cells confers substantial protection against Ang as pCRII-NKCC2. The 5primer and 3primer were as II–induced oxidative injury, and demonstrated a link be- follows: 5-CCCCAACACACAAATCGATAGCTG-3 tween up-regulation of HO-1 and TALH survival. Over- and 5-CCAGACAACTTACCGCCCATCC-3, respec- expression of HO-1, via HO-1 gene transfer, prevented tively. The amplified NKCC2 promoter region was se- Ang II–stimulated DNA degradation and up-regulated quenced and analyzed by digestion with restriction growth-stimulating hormone (GSH) levels. These results endonucleotidase. may provide the rationale for a possible strategy to protect TALH from the consequences of tubular hypoxic and oxidative injury with subsequent oxidative stress-inspired Construction of adenovirus expressing inflammation and apoptosis. NKCC2-controlled HO-1 gene The 1702 bp NKCC2 promoter was released from pCRII-NKCC2 by the digestion with XbaI and KpnI, and METHODS subcloned at the XbaI/KpnI sites of pShuttle (Stratagene, Animals La Jolla, CA, USA). The resultant plasmid was desig- Male Sprague-Dawley rats, 7 to 8 weeks old (Charles nated as pShuttle-NKCC2. The cytomegalovirus (CMV) River Lab, Wilmington, MA, USA), weighing 100 to promoter was removed from pShuttle-NKCC2 by MfeI 115 g, were maintained on standard rat chow (Ralston- and NheI. The resultant plasmid, after self-ligation of the Purina, Chicago, IL, USA) and given tap water ad libitum. end-blunted vector, was named as pSh-NKCC2. The 987 bp human HO-1 (HHO-1) cDNA fragment coding for the full length of HO-1 protein was released from the Isolation of TALH cells and cell culture plasmid pGEM-HHO-1 [25, 26] by HindIII. The released Rat TALH cells were isolated and characterized using human HO-1 fragment was blunted with desoxynucle- a standard primary culture procedure as described pre- oside triphosphate (dNTP) and T4 DNA polymerase viously [21–24]. The cells were cultured in Dulbecco’s and inserted at the end-blunted BamHI site of pSh- minimum essential medium (DMEM) (Life Technolo- NKCC2. The transcriptional orientated clone was named gies, Grand Island, NY, USA) containing 10% fetal pSh-NKCC2-HO-1. The adenoviral vector, Ad-NKCC2- bovine serum (FBS) and 20 ng/mL epithelial growth HO-1, was constructed by subcloning the NKCC2-HO-1 factor (EGF). After 3 to 5 days, monolayers of cells DNA fragment from pSh-NKCC2-HO-1 into I-ceuI/PI- were 50% to 60% confluent. Rat lung microvessel en- sceI sites of Adeno-X viral DNA (Stratagene). After dothelial cells (EC) were cultured in MCDB131 medium digestion with PacI, Ad-NKCC2-HO-1 was transfected 1630 Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH

Table 1. Synthetic primers for polymerase chain reaction (PCR) or analysis. Protein levels were visualized by immunoblot- reverse transcription (RT)-PCR analysis ting with antibodies against human HO-1 (which does not Amplified DNA crossreact with rat HO-1), rat HO-1 (which crossreacts fragments Primer sequences with human HO-1) and rat HO-2 (Stressgen, Vancou- Adenoviral vector Upstream 5-GAAGACAATAGCA ver, British Columbia, Canada). Briefly, 30 lg of lysate (370 bp) GGCATGC-3 supernatant was separated by sodium dodecyl sulfate- Downstream 5-GTAGGTCAAGGTAGT AGAGTTT-3 polyacrylamide gel electrophoresis (SDS-PAGE) and NKCC2 promoter Upstream 5-CCCCAACACAC transferred to a nitrocellulose membrane (Amersham, (1702 bp) AAATCGAT-3 Inc., Piscataway, NJ, USA) using a semidry transfer ap- Downstream 5-GCCCAGACACTTACCGC CCCATC-3 paratus (Bio-Rad, Hercules, CA, USA). The membranes Human HO-1 cDNA Upstream 5-CAGGCAGAGAATGC were incubated with 5% milk in 10 mmol/L Tris-HCl (550 bp) TGAGTTC-3 (pH 7.4), 150 mmol/L NaCl, 0.05% Tween 20 (TBST) Downstream 5 -GATGTTGAGCAGGA ◦ ACGCAGT-3 buffer at 4 C overnight. After washing with TBST, the Rat b-actin Upstream 5-TTGTAACCAACTGGGACG membranes were incubated with a 1:2000 dilution of anti- (764 bp) ATATGG-3 HO-1 or anti-HO-2 antibodies for 1 hour at room tem- Downstream 5-GATCTTGATCTTCATGGT GCTAGG-3 perature with constant shaking. The filters were washed and subsequently probed with horseradish peroxidase- HO-1 is heme oxygenase-1. conjugated donkey antirabbit IgG (Amersham) at a dilution of 1:2000. Chemilluminescence detection was into HEK-293 cells. Adenoviruses harvested from trans- performed with the Amersham ECL detection kit accord- fected 293 cells were identified by polymerase chain re- ing to the manufacturer’s instructions. action (PCR) and viral titer assay, and were used to infect different kinds of cells. Control adenovirus Ad-lacZ Identification of human and rat HO-1 genes in rat was obtained from Quantum Biotechnologies (Montreal, TALH cells by confocal images Quebec, Canada), in which the lacZ gene is under the A monoclonal antibody against human HO-1, rat HO- control of the CMV promoter. The results showed that 1 and a polyclonal antibody against HO-2 were purchased the amplified NKCC2 promoter sequence was identical from Stressgen. Cells were plated on Lab-Tek Cham- to that published previously [28]. ber Slides (Nalge Nunc International, Urbana, IL, USA) and cultured for 3 to 4 days until they reached 50% Reverse transcription (RT)-PCR and PCR analysis to 60% confluency. After a 2-day infection with aden- of human HO-1 expression in TALH oviruses, the cells were fixed in 4% paraformaldehyde RT was carried out using the AdvantageTM RT-for- for 10 minutes at room temperature, permeabilized with PCR Kit (Clontech, Palo Alto, CA, USA). Poly-d(T)n 0.1% Triton X-100 and blocked with 10% normal goat was used as the RT primer. Specific primers for am- serum followed by incubation with primary antibodies plifying human HO-1 and the adenoviral vector gene for 2 hours and secondary antibodies [fluorescein isothio- fragment are listed in Table 1. PCR was performed us- cyanate (FITC)-conjugated goat antimouse IgG and Cy3- ing a Taq PCR Kit (Roche, Indianapolis, IN, USA). For conjugated goat antirabbit IgG antibodies] for 1 hour. each RT-PCR,a sample without reverse transcriptase was After several washes in phosphate-buffered saline (PBS), processed in parallel and served as a negative control. the slides were mounted using Slow Fade (Molecular Cycling parameters for amplifying RT products were as Probes, Eugene, OR, USA). Immunofluorescence was follows: 95◦C, 1 minute; 60◦C, 1 minute; 72◦C, 1 to 3 min- visualized by confocal microscopy (Olympus Fluoview utes, for 30 cycles, and then extended at 72◦C for another FV300) (Olympus, Melville, NY, USA). 5 minutes. After amplification, PCR products were electrophoresed on 1.2% agarose gel, stained with ethidium HO activity bromide, and visualized under ultraviolet light. Genomic HO activity was assayed in cell homogenates by a previ- DNA was extracted from cells using the DNeasy Tissue ously described method [29]. Briefly, bilirubin, the prod- Kit (Qiagen, Valencia, CA, USA). Different combina- uct of HO degradation, was extracted with chloroform tions of primers (Table 1) were used to amplify the inte- and its concentration was determined spectrophotomet- grated DNA fragments via the Expand High Fidelity PCR rically using the difference in absorbance at wavelength System (Roche). Cycling parameters were as described from k 460 to k 530 nm with an absorption coefficient of above. 40 mM−1 and cm−1 [29].

Western blot Microsomal heme and PGE2 determination Cells were harvested using cell lysis buffer as described Microsomal heme was determined as the pyridine previously [25]. The lysate was collected for Western blot hemochromogen by using the reduced minus oxidized Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH 1631 difference in absorbance at k 400 and k 600 nm with (P < 0.05) between the experimental groups was deter- an absorption coefficient of 32.4 mM−1 cm−1 [25]. For mined by the Fisher methods for multiple comparisons. measurement of PGE2, medullary thick ascending limb For comparison between treatment groups, the null hy- (mTAL) cells were incubated for 24 hours after treat- pothesis was tested by single-factor analysis of variance ment with different chemicals, after which the cell-free (ANOVA) for multiple groups or unpaired t test for two supernatants were assayed for PGE2 using the enzyme- groups. linked immunosorbent assy (ELISA) kit according to the manufacturer’s specifications (Cayman, Ann Arbor, MI, USA) at 450 nm [26]. RESULTS Effect of Ang II and inducers on HO-1 and HO-2 DNA analysis by COMET assay protein expression The presence of DNA fragmentation was examined by HO-1 protein expression in TALH cells treated with single-cell electrophoresis (COMET assay), as previously different HO-1 inducers was analyzed by Western blot. × 5 described [9]. Briefly, 0.5 to 0.8 10 cells were mixed Primary cultures of mTAL cells expressed significant with 75 lL 0.5% low-melting agarose (LMA) and spotted basal levels of HO-1 protein. However, HO-1 expression on slides. The “minigels” were maintained in lysis solution in mTAL cells in response to inducers (i.e., SnCl2, heme, [N-laurosil-sarcosine 1%, NaCl 2.5 mol/L, Na2EDTA and Ang II) was different from the response obtained 100 mmol/L, dimethyl sulfoxide (DMSO) 10%, pH 10] for ◦ in endothelial cells and vascular smooth muscle cells. 1 hour at 4 C, then denatured in a high pH buffer (NaOH As seen in Figure 1, when equal amounts of cell lysates 300 mmol/L and Na2EDTA 1 mmol/L) for 20 minutes (30 lg) from three different cell types (TALH, endothe- and finally electrophoresed in the same buffer at 25 V lial, and vascular smooth muscle) were loaded for West- for 50 to 60 minutes. At the end of the run, the minigels ern blot analyses under the same conditions, basal HO-1 were neutralized in Tris-HCl 0.4 mol/L, pH 7.5, stained was expressed at relatively higher levels in TALH cells with 100 lL ethidium bromide (2 lg/mL) for 10 minutes, compared to endothelial and vascular smooth muscle and scored using a Nikon fluorescence microscope cells. Treatment with SnCl2 (10 lmol/L, 24 hours) led (Nikon Labophot) interfaced with a computer. The Soft- to increased HO-1 protein levels by 2.9-fold and 1.8-fold ware Scion Image with a COMET 1.3 version macro (free in endothelial and vascular smooth muscle cells, respec- download from NIH web site) allowed us to analyze and tively; however, no apparent change in HO-1 protein ex- quantify DNA damage by measuring (1) tail length (TL) pression was found in TALH cells after the same exposure and (2) tail DNA percentage (TDNA). These parameters to SnCl2. The addition of heme (10 lmol/L, 24 hours) are employed by the software to determine the level of caused 4.3-fold and 4.1-fold increases in HO-1 protein DNA damage as tail moment (TMOM) expressed as the levels in endothelial cells and vascular smooth muscle product of TL and TDNA. cells, respectively, whereas only a modest increase (30%) in HO-1 protein was detected in TALH cells. None of these treatments affected the levels of HO-2 protein (data GSH levels not shown). These data indicate that the responsiveness GSH levels were quantitatively measured by spec- of TALH cells to HO-1 inducers is significantly reduced trophotometric method. GSH contents were expressed compared to that of endothelial cells or vascular smooth in nmol of GSH per mg of protein. Briefly, 300 lLof muscle cells. lysate was mixed with 10 lL of trichloroacetic acid (30% wt/vol) in an Eppendorf tube for 15 minutes in ice. The mixture was centrifuged (3000g for 10 minutes at 4◦C), and 100 lL of supernatant was mixed with 0.880 mL of Effect of heme and Ang II on COX-2 and HO-1 Tris (0.25 mol/L)-EDTA (20 mmol/L) (TE) buffer (pH protein expression 8.2) and 20 lL of 10 mmol/L 2,2-dithiobisnitrobenzoic Since heme and Ang II stimulate COX-2 protein ex- acid (DTNB). The color was developed for 15 to 20 min- pression and enhance PGE2 synthesis in endothelial cells utes at room temperature. The absorbance of the super- [30, 31], we determined whether these agents could ac- natant was measured at 412 nm and subtracted from a tivate COX-2 protein in TALH primary cell cultures. DTNB-TE blank. Figure 2A shows that exposure to Ang II (100 ng/mL) and heme (10 lmol/L) for 24 hours resulted in a significant increase in COX-2 protein expression by 41% and Statistical analysis 27%, respectively. Again, exposure to heme and Ang II The data are presented as mean ± standard error (SE) resulted in a modest increase in HO-1 protein in TALH for the number of experiments. Statistical significance primary cell cultures (Fig. 2B). 1632 Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH

Functional expression of human HO-1 in TALH A ConSnCl2 Heme transduced with ADNKCC2-HO-1 mTAL HO-1 To study the site-specific HO-1 expression in TALH, we examined the ability of the adenovirus, Ad-NKCC2- Actin HO-1, to express HO-1 protein in other cell types, such as endothelial cells, as well as in TALH. After 48 hours 1.4 1.31 of infection with Ad-NKCC2-HO-1 and the control ade- 1.2 novirus Ad-lacZ, the cells were subjected to RT-PCR. 1 1.05 1 Western blot analysis and HO activity assay were used 0.8 to assess the functional expression of human HO-1 in TALH. RT-PCR analyses showed that human HO-1 tran- 0.6 scripts were detected in TALH cells infected with Ad- 0.4 HO-1 protein, optical density NKCC2-HO-1, but were not detected in control TALH 0.2 cells or in TALH cells infected with Ad-lacZ (Fig. 3A). 0 Cell-specific expression was further examined in an en- Con SnCl2 Heme dothelial cell line. Ad-NKCC2-HO-1 caused pronounced expression of B ConSnCl Heme 2 human HO-1 protein in mTAL cells (Fig. 3B), but not in EC HO-1 rat microvessel endothelial cells (Fig. 3C). There was no HO-1 signal detected in mTAL after infection with the Actin control empty viral vector or virus Ad-lacZ. Moreover, rat HO-2 protein expression was not affected in TALH cells injected by either adenovirus. 5 4.3 HO activity was assayed in both TALH and endothelial 4 cells after exposure to either Ad-NKCC2-HO-1 or con- 2.9 trol adenovirus, Ad-lacZ, for two days (Fig. 4A). HO ac- 3 tivity in TALH cells infected with Ad-NKCC2-HO-1 was 2 significantly (1.81-fold) higher than that in TALH cells < 1 infected with Ad-lacZ (P 0.05). However, there was no HO-1 protein, optical density 1 significant difference in HO activity between endothelial cells infected with Ad-NKCC2-HO-1 and Ad-lacZ, in- 0 dicating that the Ad-NKCC2-HO-1 increases HO-1 ex- Con SnCl2 Heme pression specifically in TALH cells, but not in endothelial C ConSnCl2 Heme cells. Transduced human HO-1 and endogenous rat HO-2 VSMC HO-1 expression were measured in cultured TALH and endothelial cells after infection with adenoviruses for 2 days Actin using FITC (for human HO-1) or Cy3 (for HO-2)-labeled antibodies and viewed by confocal microscopy as seen 5 in Figure 4B. Human HO-1 expression was detected in 4.1 TALH cells infected with adenovirus Ad-NKCC2P-HO- 4 1 (Fig. 4A and C), but not in TALH cells infected with Ad-lacZ (data not shown). Moreover, human HO-1 was 3 not detected in endothelial cells infected with either Ad- 2 1.8 NKCC2-HO-1 or Ad-lacZ (data not shown). 1 HO-1 protein, optical density 1

0 Effect of Ad-NKCC2-HO-1 transduction on PGE2 Con SnCl2 Heme production in TALH cells Since infection of TALH cells with Ad-NKCC2-HO-1 Fig. 1. Western blot analyses in thick ascending loop of Henle (TALH) (A), endothelial (EC) (B), and vascular smooth muscle cells (VSMC) signiﬁcantly increased human HO-1 protein expression (C). Cells were treated with SnCl2 (10 lmol/L) or heme (10 lmol/L) and total HO activity, we next examined whether over- for 24 hours. Control rat heme oxygenase-1 (HO-1) protein expression expression of HO-1 via this route could affect the produc- (relative optical density) was arbitrarily set to 1 and compared with cells tion of COX-2 activity, as measured by PGE , released in treated with SnCl2 or heme. The plot shown is representative of three 2 different experiments. Con is control. cultured media. As seen in Figure 5, infection of TALH Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH 1633

A A Ang II Heme Con

COX-2

Actin

1.6 * Ad-NKCC2P-HO-1Ad-NKCC2P-HO-1 (20Ad-lacZ MOI) (5 Control(20 MOI) MOI) (vehicle)100 bp DNA(–) ladder Control(+) Control 1.4 * 1.2 HHO-1 1 0.8

0.6 β-actin

optical density COX-2 protein, 0.4

0.2 B 0 Con Heme Ang II

B Ang II Heme Con

HO-1 Control Ad-lacZ (50 Ad-NKCC2P-HO-1MOI) Ad-NKCC2P-HO-1 (10 MOI) (50 MOI) Actin HHO-1 1.8 1.6 *

1.4 * HO-2 1.2 1 C 0.8

HO-1 protein, 0.6 optical density 0.4 0.2 0 Con Heme Ang II Control Ad-NKCC2P-HO-1Ad-NKCC2P-HO-1 (10 MOI)Ad-CMV-HO-1 (50 MOI) (50 MOI)

Fig. 2. Western blot analysis of cyclooxygenase-2 (COX-2) (A) and rat HHO-1 heme oxygenase (HO-1) (B) protein expression in thick ascending loop of Henle (TALH) cells. Cells were exposed to heme (10 lmol/L) or angiotensin II (Ang II) (100 ng/mL) for 24 hours. Relative optical densities HO-2 of HO-1 or COX-2 protein expression in control (con) untreated TALH cells were set to 1 and were compared with those in cells treated with = ∗ < heme and Ang II and shown under each panel (N 3, P 0.05 vs. Fig. 3. Analyses of human heme oxygenase (HO), rat b-actin mRNA control). in thick ascending loop of Henle (TALH) and microvessel endothelial cells (EC). (A) Reverse transcription-polymerase chain reaction (RT- cells with Ad-NKCC2-HO-1 (50 MOI for 2 days) alone PCR) analysis of human HO-1 and rat b-actin mRNA in TALH cells infected with adenoviruses for 2 days. The speciﬁc primers for human did not cause a signiﬁcant change in PGE2 production HO-1 and rat b-actin are described in Table 1. (B) Western blot analysis (519 ± 41 pg/mL) compared with control TALH cells of human HO-1 and rat HO-2 in TALH and microvessel EC (C) infected (468 ± 20 pg/mL) infected with control adenovirus Ad- with adenovirus Ad-NKCC2-HO-1, Ad-CMV-HO-1 (positive control) or Ad-lacZ (negative control). Cells were infected with adenoviruses lacZ. The addition of heme (10 lmol/L) for 24 hours for 2 days. to TALH infected with Ad-NKCC2-HO-1 or control 1634 Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH

Control Ad-lacZ Ad-NKCC2P-HO-1 Ad-NKCC2P-HO-1 A 900 1.8 * * 800 * 1.6 700 1.4 600 * 1.2 500 pg/mL

1 ,

2 # 400 0.8 *

PGE 300 HO activity, 0.6 200 nmol bilirubin/mg/h 0.4 100 0.2 0 0 Control Heme Ang II mTAL EC Fig. 5. Prostaglandin E2 (PGE2) production in thick ascending loop B of Henle (TALH) cells. Cells were infected with Ad-NKCC2-HO-1 (50 MOI) or Ad-lacZ (50 MOI) 24 hours before being treated with heme (10 lmol/L) or angiotensin II (Ang II) (100 ng/mL) and incubated for an additional 24 hours. Cultured media were used for measurement of PGE2 production by enzyme-linked immunosorbent assay (ELISA) [∗P < 0.05 vs. corresponding control virus-infected TALH cells; #P < 0.05 vs. Ad-lacZ-infected medullary thick ascending loop (mTAL) cells].

Table 2. Angiotensin II (Ang II)–mediated cellular free heme content changes in thick ascending loop of Henle (TALH) cells pre-infected with adenoviruses, Ad-NKCC2-HO-1 or Ad-lacZ Heme content pmol/mg Treatment microsomal protein Ad-lacZ 411.7 ± 75.8 Ad-lacZ + Ang II 389.7 ± 20.6 Ad-NKCC2-HO-1 293.0 ± 43.3∗ Ad-NKCC2-HO-1 + Ang II 305.3 ± 21.2∗

Fifty percent conﬂuent TALH cells were infected with adenoviruses, Ad- NKCC2-HO-1 (50 MOI) or Ad-lacZ (50 MOI), for 24 hours before treatment with Ang II (100 ng/mL) for an additional 24 hours. Cellular-free heme content was measured as described in the Methods section (N = 3, P < 0.05 vs. corresponding control cells).

Fig. 4. Heme oxygenase (HO) activity in medullary thick ascending loop of Henle (mTAL) and endothelial cells (EC). (A) Cells were in- in TALH cells infected with Ad-NKCC2-HO-1, Ang II– fected with either Ad-NKCC2-HO-1 (50 MOI) or Ad-lacZ (50 MOI) for 2 days before cells were harvested and used for HO activity assay stimulated production of PGE2 was greatly attenuated. (N = 4, ∗P < 0.05 vs. corresponding control cells). (B) Confocal microscopy of human HO-1 and rat HO-2 proteins in TALH cells infected with Ad-NKCC2-HO-1. Cells were infected with Ad-NKCC2-HO-1 Effect of targeting TALH cells with human HO-1 (50 MOI) for 2 days and transgene (human HO-1) and endogenous rat on endogenous cellular heme HO-2 expressions were viewed by confocal microscopy. Human HO- 1 (panels a and c) and rat HO-2 (panels b and d) immunoreactivity To investigate the effects of targeting TALH with HO- is shown in Ad-NKCC2-HO-1–infected TALH cells using fluorescein 1 gene on cellular heme, a pro-oxidant, cells were infected isothiocyanate (FITC)-conjugated antihuman HO-1 and Cy-3 conjugated anti-HO-2, respectively [original magnifications 100× (a and b) with the Ad-NKCC2-HO-1 or Ad-lacZ, and 2 days later and 400× (c and d). microsomal heme levels were measured. As shown in Table 2, in TALH cells infected with Ad-NKCC2-HO-1, cellular free-heme content was decreased to 293.0 ± 43.3 adenovirus resulted in a significant increase of PGE2 pmol/mg compared to 411.7 ± 75.8 pmol/mg in TALH compared to control TALH cells (P < 0.05). There was cells infected with control adenovirus Ad-lacZ (P < 0.05). no significant difference in PGE2 production between Cellular heme levels in Ad-lacZ or Ad-NKCC2-HO-1– mTAL cells infected with either Ad-NKCC2-HO-1 or infected cells were not altered in response to Ang II Ad-lacZ after exposure to heme (P > 0.05). However, (Table 2). Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH 1635

200

* 150

Fig. 6. Effect of human heme oxygenase-1 (HO-1) gene transfer on angiotensin II (Ang * II)–induced DNA damage. Thick ascending 100 loop of Henle (TALH) cells were stimulated

TMOM * for 16 hours by Ang II (200 ng/mL) alone or in cells infected with Ad-NKCC2-HO-1 or Ad-lacZ (50 MOI, 2 days) and then subjected to Ang II. The histograms demonstrate DNA damage, as measured by COMET as- 50 say and expressed as mean of tail moment (TMOM). Treatment is representative of an average mean of TMOM ± SE of three individual slide determinations. On each slide, 50 COMETs were scored blindly for TMOM. Signiﬁcantly different values (∗P < 0.05 vs. 0 # < Control Bilirubin Ang II Bilirubin Ad-NKCC2P- control; P 0.05 vs. Ang II alone) were determined by analysis of variance (ANOVA) + Ang II HO-1 + Ang II multiple range testing.

Fig. 7. Morphologic determination of COMET in cells treated with angiotensin II (Ang II). Representative microphotographs of agarose low melting minigel thick ascending loop of Henle (TALH) cells treated with vehicle control (A), Ang II (200 ng/mL) for 16 hours (B), or Ad-NKCC2P-HO-1 (50 MOI) for 2 days, then Ang II (200 ng/mL) for 16 hours (C). Damaged DNA migrates during electrophoresis from the nucleus toward the anode, forming a shape of a “comet” with a head (cell nucleus with intact DNA) and a tail (relaxed and broken DNA). Images were taken by a fluorescence Control Ang II Ad-NKCC2P- microscope (rodamine filter) plugged with a + digital camera after staining with ethedium HO-1 Ang II bromide (original magnification 40×).

Effect of human HO-1 gene on Ang II–mediated bilirubin, from NKCC2 HO-1 expression, decreases Ang genotoxic damage II–mediated oxidative injury. Since Ang II and oxidant-inducing agents have been shown to cause apoptosis in several renal cell types, including TALH [32–35], we reasoned that cell-specific Morphologic measurements of COMET in cells expression of HO-1 in TALH may attenuate Ang II– treated with Ang II mediated DNA damage. To test this hypothesis, serum- We measured the morphologic changes in cells treated starved TALH cells were exposed to Ang II for 16 hours with Ang II and assessed the effect of inducers on and DNA integrity was analyzed by COMET assay as COMET changes. No DNA damage was detected in con- shown in Figure 6. Ang II increased TMOM levels from trol, nontreated cells. In fact, the DNA remains intact 54 ± 9 in the control cells to 143 ± 12. In contrast, in cells as shown in Figure 7. Following treatment with Ang II transfected with human HO-1 Ang II–mediated DNA (200 ng/mL for 16 hours), which induced DNA single- damage was greatly diminished, suggesting that HO-1 strand breaks, distinct COMETs were observed. The ex- does shield the cells from the initial genotoxic effect pro- tent of DNA damage was quantitated by image analysis duced by Ang II. TMOM levels in cells incubated with to produce a tail moment, defined as the product of the bilirubin were not significantly different from the control percentage of DNA in the COMET tail and the distance cells. The addition of bilirubin to Ang II–exposed cells between the means of the head and tail distributions [36]. attenuated DNA damage, suggesting that HO-1–derived As seen in Figure 7, COMET appearance is clearly seen 1636 Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH

A endogenous GSH in cultured mTAL cells after 24 hours 40 * ** of exposure. In contrast, incubation of mTAL cells with Ang II decreased the levels of GSH, which was reversed by supplementation of bilirubin. Similar results were ob- 30 served when TALH were infected with Ad-NKCC2-HO- 1 for 2 days. A significant increase in GSH levels was 20 observed compared to control (P < 0.001). Although * Ad-NKCC2-HO-1 reversed the effect of the Ang II– GSH levels, mediated decrease in GSH, Ad-NKCC2-HO-1 enhances nmol/mg protein 10 the levels of GSH significantly greater than that which occurs with bilirubin. 0 Control Bilirubin Ang II Ang II+ Bilirubin DISCUSSION B The balance between the action of oxidative stress com- 140 pounds and the physiologic defense against them appears * to be a determining factor in the pathogenesis of diverse renal diseases, including ischemic/reperfusion injury.Ang II, heme, and other inflammatory molecules may shift this balance in favor of free radical–mediated epithelial cell ** 40 injury [33, 38–41]. Induction of HO-1 gene expression has been suggested to be an adaptive response to ox- 30 idative stress injury [42, 43]. Hemoglobin and myoglobin GSH levels, are potential sources of free radical oxidants and renal nmol/mg protein 20 * inflammation [44–46]. Furthermore, heme moieties re- 10 leased from heme proteins have been shown to promote the formation of free oxygen radicals and DNA degrada- 0 tion [9]. Pertinent to the role of HO in renal injury is the Adv-NKCC Adv-NKCC Ang II Adv-NKCC study of Nath et al [44]. These investigators demonstrated -HO-1 -HO-1 + Ang II that in vivo induction of HO prevents renal failure and Fig. 8. Determination of reduced growth-stimulating hormone (GSH) drastically reduces mortality after glycerol-induced rhab- content. GSH levels were measured in control thick ascending loop of domyolysis in the rat; a condition that is characterized Henle (TALH) cells (A) and TALH cells infected with Ad-NKCC2P- by an increased release of myoglobin and hemoglobin, HO-1 or Ad-lacZ (B) treated with and without angiotensin II (Ang II) or bilirubin. GSH levels were measured spectrophotometrically and resulting in irreversible renal tissue injury, renal failure expressed as mean ± SE (nmol/mg of protein) of three individual exper- and death. Oxidative-mediated tissue damage is also as- ∗ < ∗∗ < iments. Significantly different values ( P 0.05 vs. control; P 0.05 vs. sociated with up-regulation of monocyte chemoattractant control + Ang II) were determined by analysis of variance (ANOVA) multiple range testing. protein-1 (MCP-1), a major factor in the development of renal disease [47]. The TALH, as the renal region most susceptible to ox- at higher levels in Ang II compared to control cells. Cells idant injury, is a practical target for antioxidant therapy pretreated with Ang II and Ad-NKCC2 HO-1 showed a such as that provided by HO-1 overexpression. Previous similar profile to the nontreated cells. Pretreatment with studies in our laboratory indicated that HO-1 expression Ad-lacZ control virus failed to maintain DNA integrity is very low in freshly isolated TALH and can be greatly in presence of Ang II, suggesting that HO-1 expression induced with chemical inducers of HO-1 [21]. In con- was instrumental in attenuating the Ang II–mediated in- trast, primary cultures of TALH in this study showed a crease in COMETs. significant level of HO-1 expression. It is possible that, in vivo, HO-1 is controlled by suppressor factors including Effect of Ang II on GSH glucocorticoids [48, 49], interferon gamma (INF-c) [50], Previous studies have provided evidence that Ang II– Bach-1 transcriptional factors [51], and hypoxia [52]. On mediated reactive oxygen species (ROS) can have a ma- the other hand, placing cells in culture in a medium con- jor effect on the redox status of localized pools of GSH taining serum may up-regulate HO-1 expression. Nev- [37], which affects the level of oxygen in TALH. We mea- ertheless, we were able to successfully enhance HO-1 sured the effect of bilirubin, a product of HO activity expression by transduction of human HO-1 gene into with antioxidant properties, on GSH levels. As seen in rat TALH using the cell specific promoter, NKCC2. The Figure 8, the addition of bilirubin increased the levels of transduced HO-1 gene induced the expression of human Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH 1637

HO-1 gene without affecting expression of endogenous TNF and Ang II, preserves DNA integrity, and main- rat HO-1 or altering the rat HO-2 gene. Infection of en- tains normal cell cycle progression [7, 8, 13]. We, as well dothelial cells or vascular smooth muscle cells with the as others, have shown that HO-1 overexpression attenu- TALH specific adenovirus, Ad-NKCC2-HO-1, did not ates cell death by oxidants such as H2O2 and TNF [16, result in functional HO-1 expression or HO activity. Se- 55], an affect attributed to CO. CO has been shown to lective delivery of the human HO-1 gene to TALH, using play an important role in controlling cell cycle progres- the NKCC2 promoter, increased HO activity leading to sion [7, 14, 16], signifying the important role of this gas in decreased heme levels and increased production of biliru- cell growth. Others have shown that when an inhibitor of bin and CO. The increased in the levels of bilirubin and HO blocks HO activity or the action of CO is inhibited, CO, which exhibit antioxidant and antiapoptotic proper- endothelial cell apoptosis no longer prevented [13]. How- ties, may contribute to the cytoprotective action of HO-1 ever, the effect of CO has been shown to be cell specific; overexpression against oxidant injury brought about by in endothelial cells, CO enhances cell growth, whereas, Ang II (vide infra). CO inhibits vascular smooth muscle cell growth [56–61]. We have previously showed that endothelial cells sta- It has been shown, that in the one kidney, one clip bly transfected with the human HO-1 gene exhibited (1K1C) mouse model, HO-1 deficiency exacerbated hy- a several-fold increase in human HO-1 mRNA levels, pertension and induced acute ischemic renal failure [62]. which was accompanied by an increase in HO activity HO-1 has been shown to be up-regulated in the kidney and a marked decrease in PGE2 associated with selec- of Ang II–induced hypertensive rats [63]. In rats trans- tive down-regulation of COX-2 but not COX-1 [30]. We duced with human HO-1, Ang II pressor responsiveness also demonstrated that the degree of HO-1 expression is is greatly attenuated [abstract; Yang L, et al, Hyperten- inversely related to COX activity and that this is linked sion 42:P11-P407, 2003] [64]. Ang II–receptor blockade to the levels of cellular heme [30]. Hence, addition of with losartan has been shown to prevent oxidative stress- heme to cells overexpressing HO-1 resulted in an in- induced hypertension in rats [65]. Renal tubular injury, as crease in COX activity [30]. More recently, others have a major pathologic feature, has been described in human confirmed the finding that HO-1 overexpression down- HO-1 deficiency [66]. Induction of HO-1 has been shown regulates COX-2 activity due to the availability of heme, to protect against cisplatin nephrotoxicity in the rat [46]. an essential cofactor for COX activity [53]. Unlike en- Our study demonstrates that the up-regulation of HO- dothelial cells, TALH cells overexpressing HO-1 showed 1 in the TALH provides protection against Ang II– no change in PGE2 levels. However, while Ang II induced mediated DNA degradation caused by oxidative injury. COX-2 expression in TALH, overexpression of HO-1 It supports the notion that the induction of HO-1 and greatly attenuated Ang II–induced PGE2 production. formation of bilirubin and CO are central features of The mechanism of this effect is unclear. It is possible that this antioxidative mechanism in TALH. This may play HO-1 overexpression inhibits Ang II action by affecting an important role in preventing renal damage produced Ang II–induced mechanisms upstream of COX-2 (e.g., by tubular hypoxic and oxidative damage with resulting production of ROS). The ability of HO-1 overexpres- inflammation and apoptosis. sion to inhibit Ang II–stimulated PGE2 may represent additional means by which it provides cytoprotection. To this end, studies have shown that Ang II induces TNF ACKNOWLEDGMENTS production by the TALH and that this effect is COX-2– The authors thank Mrs. Sylvia Shenouda for her excellent tech- dependent [21]. nical assistance. This work was supported by NIH grants HL55601, Another important key finding, crucial for TALH sur- NIH/NHLBI grant 34300 (N.G.A., M.L.S., and A.N.) and the Westch- ester Artificial Kidney Foundation (N.G.A.). vival, is that selective delivery of the human HO-1 gene attenuated Ang II–mediated genotoxic damage, stem- Reprint requests to Nader G. Abraham, M.D., The Rockefeller Univer- sity, New York, NY 10021 and The New York Medical College, Valhalla, ming from the generation of O2 and the formation of hy- NY 10595. droxyl radicals [54]. The increase in HO-1 activity and the E-mail: nader [email protected] subsequent generation of bilirubin, an antioxidant and hydroxyl radical scavenger, prevented DNA degradation and increased antioxidant mechanisms such as GSH lev- REFERENCES els in TALH. Bilirubin and biliverdin both act as antiox- 1. HEYMAN SN, BREZIS M, REUBINOFF CA, et al: Acute renal failure idants in vitro and in vivo [10] and their increased local with selective medullary injury in the rat. J Clin Invest 82:401–412, 1988 concentrations, after HO induction, are beneficial in pro- 2. BEERI R, SYMON Z, BREZIS M, et al: Rapid DNA fragmentation from tecting endothelial cells from injury [9]. hypoxia along the thick ascending limb of rat kidneys. Kidney Int Studies from this laboratory and others have demon- 47:1806–1810, 1995 3. ORTIZ PA, G ARVIN JL: Superoxide stimulates NaCl absorption by strated that HO-1 gene overexpression provides a cyto- the thick ascending limb. Am J Physiol Renal Physiol 283:F957– protective effect against the inflammatory response to F962, 2002 1638 Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH

4. BREZIS M, GREENFELD Z, SHINA A, ROSEN S: Angiotensin II aug- family expressed in kidney. J Biol Chem 269:17713–17722, 1994 ments medullary hypoxia and predisposes to acute renal failure. 28. IGARASHI P, VAN DEN HEUVEL GB, PAYNE JA, FORBUSH B, III: Eur J Clin Invest 20:199–207, 1990 Cloning, embryonic expression, and alternative splicing of a murine 5. RUIZ-ORTEGA M, RUPEREZ M, LORENZO O, et al: Angiotensin II reg- kidney-specific Na-K-Cl cotransporter. Am J Physiol 269:F405– ulates the synthesis of proinflammatory cytokines and chemokines F418, 1995 in the kidney. Kidney Int 62(Suppl 82):S12–S22, 2002 29. ABRAHAM NG, LIN JH, DUNN MW, SCHWARTZMAN ML: Presence 6. HAUGEN EN, CROATT AJ, NATH KA: Angiotensin II induces renal of heme oxygenase and NADPH cytochrome P-450(c) reductase in oxidant stress in vivo and heme oxygenase-1 in vivo and in vitro. human corneal epithelium. Invest Ophthalmol Vis Sci 28:1464–1472, Kidney Int 58:144–152, 2000 1987 7. KUSHIDA T, LIVOLTI G, GOODMAN AI, ABRAHAM NG: TNF-alpha- 30. HAIDER A, OLSZANECKI R, GRYGLEWSKI R, et al: Regulation of cy- mediated cell death is attenuated by retrovirus delivery of human clooxygenase by the heme-heme oxygenase system in microvessel heme oxygenase-1 gene into human microvessel endothelial cells. endothelial cells. J Pharm Exp Ther 300:188–194, 2001 Transplant Proc 34:2973–2978, 2002 31. LI VOLTI G, SETA F, SCHWARTZMAN ML, et al: Heme oxygenase at- 8. KUSHIDA T, LI VG, QUAN S, et al: Role of human heme oxygenase- tenuates angiotensin II-mediated increase in cyclooxygenase-2 ac- 1 in attenuating TNF-alpha-mediated inflammation injury in en- tivity in human femoral endothelial cells. Hypertension 41:715–719, dothelial cells. J Cell Biochem 87:377–385, 2002 2003 9. MAZZA F, GOODMAN AI, LOMBARDO G, et al: Heme oxygenase I 32. LODHA S, DANI D, MEHTA R, et al: Angiotensin II-induced mesangial gene expression attenuates angiotensin II-mediated DNA damage cell apoptosis: Role of oxidative stress. Mol Med 8:830–840, 2002 in endothelial cells. Exp Biol Med 228:576–583, 2003 33. BHASKARAN M, REDDY K, RADHAKRISHNAN N, et al: Angiotensin 10. STOCKER R, YAMAMOTO Y, MCDONAGH AF, et al: Bilirubin is an II induces apoptosis in renal proximal tubular cells. Am J Physiol antioxidant of possible physiological importance. Science 235:1043– Renal Physiol 284:F955–F965, 2003 1047, 1987 34. DING G, REDDY K, KAPASI AA, et al: Angiotensin II induces apop- 11. BARANANO DE, RAO M, FERRIS CD, SNYDER SH: From the cover: tosis in rat glomerular epithelial cells. Am J Physiol Renal Physiol Biliverdin reductase: A major physiologic cytoprotectant. Proc Natl 283:F173–F180, 2002 Acad Sci USA 99:16093–16098, 2002 35. DI MARI JF, DAVIS R, SAFIRSTEIN RL: MAPK activation determines 12. KOZMA F, JOHNSON RA, ZHANG F, et al: Contribution of endogenous renal epithelial cell survival during oxidative injury. Am J Physiol carbon monoxide to regulation of diameter in resistance vessels. Am 277:F195–F203, 1999 J Physiol 276:R1087–R1094, 1999 36. OLIVE PL, BANATH JP, DURAND RE: Heterogeneity in radiation- 13. BROUARD S, OTTERBEIN LE, ANRATHER J, et al: Carbon monoxide induced DNA damage and repair in tumor and normal cells mea- generated by heme oxygenase 1 suppresses endothelial cell apop- sured using the “comet” assay. Radiat Res 122:86–94, 1990 tosis. J Exp Med 192:1015–1026, 2000 37. WOLIN MS: Interactions of oxidants with vascular signaling systems. 14. SOARES MP, USHEVA A, BROUARD S, et al: Modulation of endothe- Arterioscler Thromb Vasc Biol 20:1430–1442, 2000 lial cell apoptosis by heme oxygenase-1-derived carbon monoxide. 38. WITKO-SARSAT V, FRIEDLANDER M, NGUYEN KT, et al: Advanced Antioxid Redox Signal 4:321–329, 2002 oxidation protein products as novel mediators of inflammation and 15. LEE TS, CHAU LY: Heme oxygenase-1 mediates the anti- monocyte activation in chronic renal failure. J Immunol 161:2524– inflammatory effect of interleukin-10 in mice. Nat Med 8:240–246, 2532, 1998 2002 39. HARRIS RC: Growth factors and cytokines in acute renal failure. 16. KUSHIDA T, QUAN S, YANG L, et al: A significant role for the heme Adv Renal Replace Ther 4:43–53, 1997 oxygenase-1 gene in endothelial cell cycle progression. Biochem 40. HANNKEN T, SCHROEDER R, STAHL RA, WOLF G: Angiotensin II- Biophys Res Commun 291:68–75, 2002 mediated expression of p27Kip1 and induction of cellular hyper- 17. WANG T, STERLING H, SHAO WA, et al: Inhibition of heme oxygenase trophy in renal tubular cells depend on the generation of oxygen decreases sodium and fluid absorption in the loop of Henle. Am J radicals. Kidney Int 54:1923–1933, 1998 Physiol Renal Physiol 285:F484–F490, 2003 41. BIDMON B, ENDEMANN M, MULLER T, et al: Heat shock protein-70 18. XU JC, LYTLE C, ZHU TT, et al: Molecular cloning and functional ex- repairs proximal tubule structure after renal ischemia. Kidney Int pression of the bumetanide-sensitive Na-K-Cl cotransporter. Proc 58:2400–2407, 2000 Natl Acad Sci USA 91:2201–2205, 1994 42. AGARWAL A, SHIRAISHI F, VISNER GA, NICK HS: Linoleyl hydroper- 19. ORTIZ PA, H ONG NJ, PLATO CF, et al: An in vivo method for oxide transcriptionally upregulates heme oxygenase-1 gene expres- adenovirus-mediated transduction of thick ascending limbs. Kid- sion in human renal epithelial and aortic endothelial cells. J Am Soc ney Int 63:1141–1149, 2003 Nephrol 9:1990–1997, 1998 20. ORTIZ PA, H ONG NJ, WANG D, GARVIN JL: Gene transfer of eNOS 43. BOTROS FT, LANIADO-SCHWARTZMAN M, ABRAHAM NG: Regula- to the thick ascending limb of eNOS-KO mice restores the effects tion of cyclooxygenase- and cytochrome p450-derived eicosanoids of L-arginine on NaCl absorption. Hypertension 42:674–679, 2003 by heme oxygenase in the rat kidney. Hypertension 39:639–644, 21. FERRERI NR, ESCALANTE BA, ZHAO Y, et al: Angiotensin II induces 2002 TNF production by the thick ascending limb: Functional implica- 44. NATH KA, BALLA J, JACOB HS, et al: Induction of heme oxygenase tions. Am J Physiol 274:F148–F155, 1998 is a rapid protective response in rhabdomyolysis in the rat. J Clin 22. MACICA CM, ESCALANTE BA, CONNERS MS, FERRERI NR: TNF pro- Invest 90:267–270, 1992 duction by the medullary thick ascending limb of Henle’s loop. Kid- 45. ABRAHAM NG, LAVROVSKY Y, SCHWARTZMAN ML, et al: Transfection ney Int 46:113–121, 1994 of the human heme oxygenase gene into rabbit coronary microves- 23. KIKERI D, AZAR S, SUN A, et al: Na(+)-H+ antiporter and Na(+)- sel endothelial cells: protective effect against heme and hemoglobin − (HCO3 )n symporter regulate intracellular pH in mouse medullary toxicity. Proc Natl Acad Sci USA 92:6798–6802, 1995 thick limbs of Henle. Am J Physiol 258:F445–F456, 1990 46. AGARWAL A, BALLA J, ALAM J, et al: Induction of heme oxygenase 24. TRINH-TRANG-TAN MM, BOUBY N, COUTAUD C, BANKIR L: Quick in toxic renal injury: A protective role in cisplatin nephrotoxicity in isolation of rat medullary thick ascending limbs. Enzymatic and the rat. Kidney Int 48:1298–1307, 1995 metabolic characterization. Pflugers Arch 407:228–234, 1986 47. KANAKIRIYA SK, CROATT AJ, HAGGARD JJ, et al: Heme: a novel 25. QUAN S, YANG L, ABRAHAM NG, KAPPAS A: Regulation of human inducer of MCP-1 through HO-dependent and HO-independent heme oxygenase in endothelial cells by using sense and antisense mechanisms. Am J Physiol Renal Physiol 284:F546–F554, 2003 retroviral constructs. Proc Natl Acad Sci USA 98:12203–12208, 2001 48. LAVROVSKY Y, DRUMMOND GS, ABRAHAM NG: Downregulation of 26. QUAN S, YANG L, SHENOUDA S, et al: Functional expression of human the human heme oxygenase gene by glucocorticoids and identifi- heme oxygenase-1 (HO-1) driven by HO-1 promoter in vitro and cation of 56b regulatory elements. Biochem Biophys Res Commun in vivo. J Cell Biochem 85:1–12, 2002 218:759–765, 1996 27. GAMBA G, MIYANOSHITA A, LOMBARDI M, et al: Molecular cloning, 49. DERAMAUDT TB, DASILVA JL, REMY P, et al: Negative regulation of primary structure, and characterization of two members of the mam- human heme oxygenase in microvessel endothelial cells by dexam- malian electroneutral sodium-(potassium)-chloride cotransporter ethasone. Proc Soc Exp Biol Med 222:185–193, 1999 Quan et al: HO-1 regulation of Ang II–mediated oxidative stress and DNA degradation in TALH 1639

50. TAKAHASHI K, NAKAYAMA M, TAKEDA K, et al: Suppression of heme tural remodeling. Biochem Biophys Res Commun 306:523–529, oxygenase-1 mRNA expression by interferon-gamma in human 2003 glioblastoma cells. J Neurochem 72:2356–2361, 1999 59. PEYTON KJ, REYNA SV, CHAPMAN GB, et al: Heme oxygenase- 51. SUN J, HOSHINO H, TAKAKU K, et al: Hemoprotein Bach1 regulates 1-derived carbon monoxide is an autocrine inhibitor of vascular enhancer availability of heme oxygenase-1 gene. EMBO J 21:5216– smooth muscle cell growth. Blood 99:4443–4448, 2002 5224, 2002 60. SONG R, MAHIDHARA RS, LIU F, et al: Carbon monoxide inhibits hu- 52. NAKAYAMA M, TAKAHASHI K, KITAMURO T, et al: Repression of man airway smooth muscle cell proliferation via mitogen-activated heme oxygenase-1 by hypoxia in vascular endothelial cells. Biochem protein kinase pathway. Am J Respir Cell Mol Biol 27:603–610, Biophys Res Commun 271:665–671, 2000 2002 53. VICENTE AM, GUILLIN MI, ALCARAZ MJ: Participation of heme 61. BOUCHE D, CHAUVEAU C, ROUSSEL JC, et al: Inhibition of graft arte- oxygenase-1 in a model of acute inflammation. Exp Biol Med riosclerosis development in rat aortas following heme oxygenase-1 228:514–516, 2003 gene transfer. Transpl Immunol 9:235–238, 2002 54. IESAKI T, WOLIN MS: Thiol oxidation activates a novel redox- 62. WIESEL P, PATEL AP, CARVAJAL IM, et al: Exacerbation of regulated coronary vasodilator mechanism involving inhibition of chronic renovascular hypertension and acute renal failure in heme + Ca2 influx. Arterioscler Thromb Vasc Biol 20:2359–2365, 2000 oxygenase-1-deficient mice. Circ Res 88:1088–1094, 2001 55. MALAGUARNERA L, PILASTRO MR, QUAN S, et al: Significance of 63. AIZAWA T, ISHIZAKA N, TAGUCHI J, et al: Heme oxygenase-1 is up- heme oxygenase in prolactin-mediated cell proliferation and angio- regulated in the kidney of angiotensin II-induced hypertensive rats: genesis in human endothelial cells. Int J Mol Med 10:433–440, 2002 possible role in renoprotection. Hypertension 35:800–806, 2000 56. LI VOLTI G, WANG J, TRAGANOS F, et al: Differential effect of heme 64. YANG L, QUAN S, SCHWARTZMAN ML, et al: Heme oxygenase-1 gene oxygenase-1 in endothelial and vascular smooth muscle cell cycle expression modulates angiotensin II-induced increase in blood pres- progression. Biochem Biophys Res Commun 296:1077–1082, 2002 sure. Hypertension 2003 (in press) 57. STANFORD SJ, WALTERS MJ, HISLOP AA, et al: Heme oxygenase is 65. BAYORH MA, GANAFA AA, SOCCI RR, et al: Effect of losartan on expressed in human pulmonary artery smooth muscle where carbon oxidative stress-induced hypertension in Sprague-Dawley rats. Am monoxide has an anti-proliferative role. Eur J Pharmacol 473:135– J Hypertens 16:387–392, 2003 141, 2003 66. OHTA K, YACHIE A, FUJIMOTO K, et al: Tubular injury as a cardinal 58. YUN S, JUNBAO D, LIMIN G, et al: The regulating effect of heme pathologic feature in human heme oxygenase-1 deficiency. Am J oxygenase/carbon monoxide on hypoxic pulmonary vascular struc- Kidney Dis 35:863–870, 2000