An internal enhancer regulates heme- and cadmium-mediated induction of human hem - 《American Journal of Physiology》美国生理学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 25 下一页

返回列表

查看: 370605\|回复: 243	go An internal enhancer regulates heme- and cadmium-mediated induction of human hem [复制链接]

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

楼主

发表于 2009-4-21 13:43 |只看该作者 |倒序浏览 |打印

作者：Nathalie Hill-Kapturczak, Eric Sikorski, Christy Voakes, Jairo Garcia, Harry S. Nick,  Anupam Agarwal作者单位：1 Department of Medicine, Division of Nephrology,Hypertension, and Transplantation, and Department ofNeuroscience, University of Florida, Gainesville, Florida 32610   O: Y2 r* |$ g* b1 w/ L


      / Y( u2 r& Y: M& p/ v& }



         ! l! O0 n% X, S( q+ i

                  : j4 t- K. L6 `9 {$ V% s- |$ y

      1 g, e% Y4 k2 b/ G
      - W" u4 \7 d! w  r! y; @9 K
      【摘要】+ @9 X  B2 B# V2 b
   Heme oxygenase-1 (HO-1) catalyzes the rate-limiting step in hemedegradation, releasing iron, carbon monoxide, and biliverdin. Induction ofHO-1 is an adaptive and beneficial response in renal and nonrenal settings oftissue injury. The purpose of this study was to characterize the regulation ofthe human HO-1 gene in renal proximal tubule and aortic endothelial cells inresponse to heme and cadmium. Evaluation of multiple human HO-1promoter-reporter constructs up to -9.1 kb demonstrated only a partialresponse to heme and cadmium. In an effort to mimic endogenousstimulus-dependent levels of HO-1 induction, we evaluated the entire 12.5 kbof the human HO-1 gene, including introns and exons, in conjunction with a-4.5-kb human HO-1 promoter and observed significant heme- andcadmium-mediated induction of the reporter gene, suggesting the presence of an internal enhancer. Enhancer function was orientation independent and requireda region between -3.5 and -4.5 kb of the human HO-1 promoter. Our studiesidentified a novel enhancer internal to the human HO-1 gene that, inconjunction with the HO-1 promoter, recapitulates heme- and cadmium-mediatedinduction of the endogenous HO-1 gene. Elucidation of the molecular regulationof the human HO-1 gene will allow for the development of therapeuticstrategies to manipulate HO-1 gene expression in pathological states. : t- [* K$ V/ G8 t2 _1 ]
      【关键词】 gene transcription renal proximal tubule cells endothelial cells heme proteins molecular regulation
   THE CELLULAR CONTENT of heme (ferriprotoporphyrin IX), derived from heme-containing proteins such as hemoglobin, myoglobin, cytochromes, andenzymes such as nitric oxide synthase, catalase, peroxidases, respiratoryburst oxidase, and pyrrolases, requires a balance between heme synthesis anddegradation ( 31 ). In pathological states, destabilization of heme proteins leads to liberation offree heme, which has potential prooxidant effects( 22 ). Increases in renal hemecontent are observed in rhabdomyolysis, ischemiareperfusion injury, andnephrotoxin-induced acute renal failure( 2, 29, 33, 40 ). Heme damages multiple cellular targets including lipid bilayers, mitochondria, cytoskeleton, nuclei,and several intracellular enzymes( 33 ). Although multiple enzymes are involved in heme synthesis, the major biochemical pathway for hemedetoxification is via the heme oxygenase (HO) enzyme system( 28 ).- z. b3 ^+ E4 @) O8 K! k# H( ~
6 m1 V+ ?2 E" A6 {+ E" O
HO opens the heme ring, producing equimolar quantities of biliverdin, iron,and carbon monoxide (CO). Biliverdin is subsequently converted to bilirubin bybiliverdin reductase. Recent studies have highlighted the important biologicaleffects of the HO reaction product(s) that possesses antioxidant,anti-inflammatory, and antiapoptotic functions (reviewed in Ref. 21 ). Two isoforms of HO havebeen characterized: an inducible enzyme, HO-1, and a constitutive isoform,HO-2 ( 28 ). A putative isozyme,HO-3, isolated from rat brain and sharing 90% homology with HO-2, hasalso been described ( 30 ). HO-1and HO-2 are products of different genes and share 40% amino acidhomology ( 28 ). We and othersdemonstrated that induction of HO-1 by chemical inducers( 2, 32 ) or selectiveoverexpression ( 1, 41 ) is cytoprotective both invitro and in vivo, findings that have been further substantiated by studies inHO-1 knockout mice and a patient with HO-1 deficiency( 34, 41, 48 ).

The mechanisms underlying HO-1 induction are complex and tightly regulatedat the transcriptional level. Adding to the complexity, regulation of the HO-1gene is species and cell specific. The promoter of the rat HO-1 gene, forinstance, has a heat shock responsive element that is not functional in thehuman HO-1 gene ( 50 ). A GTrepeat region identified at position -258/-198 in the human HO-1 promoter isabsent in the mouse HO-1 gene ( 49 ). Interestingly, lengthpolymorphisms of this GT repeat correlate with disease development inatherosclerosis ( 16 ), emphysema ( 49 ), and vascularrestenosis ( 18 ) in humans.2 L7 U. x3 m, O  W

The human HO-1 gene, located on chromosome 22q12( 26 ), has five exons and spans 14 kb. A potential cadmium response element (TGCTAGATTT) has beenidentified between -4.5 and -4.0 kb of the 5'-flanking region( 44 ) of the human HO-1 gene;however, the cadmium response element in the mouse HO-1 gene is located downstream of this sequence( 13 ). A threefold increase in luciferase activity was reported with a -4.5-kb human HO-1 construct inresponse to cadmium, with only a minimal response to hemin, sodium arsenite,or cobalt protoporphyrin in HeLa cells( 44 ). We previously showedthat the identical 4.5-kb human HO-1 gene promoter fragment responds, in part,to both heme and cadmium with a two- to threefold increase in reporter geneactivity compared with 20- to 30-fold increase in endogenous HO-1 mRNA inhuman aortic endothelial and renal proximal tubular cells ( 4 ). Furthermore, this 4.5-kbpromoter fragment does not respond to other stimuli such as13-hydroperoxyoctadecadienoic acid (13-HPODE), hydrogen peroxide(H 2 O 2 ), and hyperoxia that directly increase de novoHO-1 gene transcription ( 4, 19 ). We report here theidentification of an enhancer region internal to the human HO-1 gene that,together with the 4.5-kb promoter, recapitulates levels of heme and cadmiuminduction observed using steady-state Northern analysis of the endogenousgene.

MATERIALS AND METHODS

Reagents. Hemin (ferriprotoporphyrin IX chloride), cadmium chloride (CdCl 2 ), linoleic acid, lipoxidase,H 2 O 2, and DEAE-dextran were purchased from Sigma (St.Louis, MO). Transforming growth factor- 1 (TGF- 1 ) was obtained from R&D Systems (Minneapolis, MN).RNeasy mini and Maxi-prep kits were obtained from Qiagen (Valencia, CA).0 N, G6 H) \" @; k5 j: _7 f* Y3 t# P
/ B# }; V  g3 T: \% K. a7 c
Cell culture. Human renal proximal tubule cells (HPTC) (Clonetics, Walkersville, MD) were grown in renal epithelial basal medium supplementedwith 5% FBS, gentamicin (50 µg/ml), amphotericin B (50 µg/ml), insulin(5 µg/ml), transferrin (10 µg/ml), triiodothyronine (6.5 ng/ml),hydrocortisone (0.5 µg/ml), epinephrine (0.5 µg/ml), and human epidermalgrowth factor (10 ng/ml). These cells are positive for -glutamyltranspeptidase, an enzyme marker for proximal tubule cells.The presence of microvilli, abundant mitochondria, lysosomes, and endocytoticvacuoles, morphological features of proximal tubule cells, were confirmed byelectron microscopy in these cells. Human aortic endothelial cells (HAEC)[derived from segments of human aorta obtained from heart transplant donorsand previously characterized by positive staining for factor VIII-relatedantigen and acetylated LDL( 46 )] were grown inendothelial basal medium (Clonetics), supplemented with 10% FBS, gentamicin(50 µg/ml), amphotericin B (50 µg/ml), hydrocortisone (1 µg/ml),human epidermal growth factor (10 ng/ml), and bovine brain extract (6µg/ml) at 37°C in 95% air-5% CO 2. Studies were performedover a range of no more than six to eight passages. Sixteen hours beforetreatment with HO-1 inducers, growth media was changed to a 0.5 and 1鸖-containing media for HPTC and HAEC, respectively.
7 T; ?) O* o! F! I' `/ a$ Q' A
Preparation of HO-1 inducers. Hemin, CdCl 2, and13-HPODE were prepared fresh daily. A stock solution of hemin (1 mM) was prepared using 10 mM NaOH. CdCl 2 (5 mM stock solution) and H 2 O 2 (10 mM stock) were prepared in water and sterilefiltered. 13-HPODE was prepared as previously described( 4 ). Lyophilized TGF- 1 was dissolved in 4 mM HCl containing 0.1% bovine serum albumin to obtain a 2 ng/µl stock solution, aliquoted, and stored at-80°C.
7 g$ C: w; d  m" g4 U8 S4 j
Cloning and characterization of a human HO-1 genomic clone. A4.5-kb fragment of the 5'-flanking region of the human HO-1 gene,including the transcription initiation site, was generated by long-range PCR(PerkinElmer, Foster City, CA) using human genomic DNA as a template, asdescribed previously ( 4 ). To obtain additional genomic sequences for the human HO-1 gene, two single-copyprobes were designed, one from the -4.5/-4.0-kb region and one from aninternal exon, and a commercially available human P1 bacteriophage library(Genome Systems, St. Louis, MO) was screened. Two P1 clones (P1 11715 and11716) were identified by both probes. These clones were characterized byrestriction map and sequence analysis, and the presence of the entire human 15 kb of the 5'-flanking region, was verifiedin P1 11716. More recently, the efforts of the human genome project resultedin the sequencing of a bacterial artificial chromosome (BAC) clone, containingportions of chromosome 22 (accession no. Z82244  ). With the use of this BACclone, sequences of the human HO-1 gene promoter fragments were confirmed, and larger fragments, further upstream, were generated by long-range PCR andrestriction digestion (see Fig.2 ).) q  s7 a5 {% y1 I# I" i
9 R$ b0 p$ `* Y) c+ b# D- j
Fig. 2. Genomic structure of the human HO-1 gene and reporter gene constructs. Top : schematic representation of the human HO-1 genomic cloneisolated from the P1 and bacterial artificial chromosome (BAC) clones with apartial restriction map extending from -11.614 to  16.9 kb. The 5 exons areindicated (E). The restriction sites for Nhe I, N; Bam HI, B; Pst I, P; Xba I, Xb; Eco RI, R are also indicated. Bottom : plasmid constructs used to assess human HO-1 promoter andenhancer activity. Details for the generation of these constructs are providedin MATERIALS AND METHODS.6 l! h6 C: l% S: f6 y5 [& Z( B

Plasmid construction. Five'-flanking promoter fragments of the human HO-1 gene, extending from -4.0, -4.5, and -9.1 kb to  80-bpposition, were generated using oligonucleotide primers ( Table 1 ) specific to the HO-1gene by long-range PCR (PerkinElmer). The 4.0- and 4.5-kb PCR products wereligated into TA cloning vectors (Invitrogen, Carlsbad, CA), and the 9.1-kb PCRproduct was cloned into pCR-XLTOPO (Invitrogen). Bam HI sites wereincorporated in the primers for the 4.0- and 4.5-kb fragments and Sal I sites were incorporated in the primers for the 9.1-kb fragmentto enable subcloning into a promoterless human growth hormone (hGH) reportergene, generating pHOGH/4.0, pHOGH/4.5, and pHOGH/9.1. A 155-bp HO-1 promoterhGH plasmid was generated by deletion of a 4.345-kb HO-1 promoter fragmentusing Nde I followed by religation, resulting in the formation ofpHOGH/155. The 4.5-kb promoter fragment was also cloned into the Bgl II site in a luciferase reporter (pGL3) generating pHOGL3/4.5. The9.1-kb promoter fragment (with Sal I sites on each end) was cloned into the Xho I site of pGL3, generating pHOGL3/9.1. A 3.5-kb HO-1promoter reporter vector was generated by cutting the pHOGL3/9.1 with Nhe I, which cuts HO-1 at 3.5 kb 5' from the start site andcuts pGL3 once 5' from the Xho I site, and religating togenerate pHOGL3/3.5.

Table 1. Primers used for long-range PCR for human HO-1 constructs
5 n  b  I$ W) {% }
The entire  1- to  12.5-kb region of the human HO-1 gene (including allexons and introns) was generated using long-range PCR with oligonucleotideprimers specific to the HO-1 gene ( Table1 ). The 12.5-kb PCR product was cloned into pCR-XL-TOPO and then subcloned into pHOGH/4.5, pHOGL3/4.5, or pHOGL3/3.5, generating pHOGH/4.5/ 12.5, pHOGL3/4.5/ 12.5, and pHOGL3/3.5/ 12.5. Sal I siteswere included at each end of the 12.5-kb fragment to enable cloning into the Sal I site of the hGH and luciferase vectors. We obtained both5'-3' and 3'-5' orientations of the  12.5-kb fragmentin the hGH vector containing the -4.5-kb HO-1 promoter region [pHOGH/4.5/ 12.5and pHOGH/4.5/ 12.5 (3'-5'), respectively]. The  12.5-kb fragmentwas also cloned into the Hin dIII site of an hGH vector containing aheterologous herpes virus thymidine kinase (TK) minimal promoter to generatepHOTKGH/ 12.5. All constructs were verified for orientation by sequencing and restriction analysis.  `4 T; K6 ^5 Z( ^0 d

Transfection of the reporter gene. HPTC and HAEC were transiently transfected with the reporter gene vectors by the DEAE-dextran method using abatch transfection protocol( 4 ). The DEAE-dextran methodwas optimized to achieve reproducible transfection efficiency ( 20-25%) inHPTC and HAEC as assessed by cotransfection with pcDNA3.1/Lac-z (Invitrogen).The transfection protocol for HPTC was as follows: 60-70% confluent cells in150-mm tissue culture plates were washed once with HBSS and once withTris-buffered saline solution (TBS). A 1.0-ml solution containing equimolar amounts of the plasmid(s) (normalized to 8.1 µg of pHOGH/4.5) andDEAE-dextran (10 mg/ml) was added per plate and rocked for 1 h at roomtemperature. The cells were shocked for 1 min at room temperature with 10%DMSO in TBS and then washed once with HBSS. The cells were incubated incomplete media (18 ml/plate) containing chloroquine diphosphate (100 µM) toinhibit lysosomal degradation of the DNA. After 4-h incubation at 37°C in5% CO 2 -room air, the cells were washed twice with HBSS and incubated in normal growth media overnight.
0 R# A: q7 P# Q# ~- ^3 @: O
The transfection protocol for HAEC was as follows: 60-70% confluent cellsin 150-mm tissue culture plates were washed once with PBS. Complete mediacontaining 10% NuSerum (BD Biosciences, San Jose, CA) was added to the plates.The transfection solution, 1.5 ml DEAE-dextran solution (10 mg/ml) containingequimolar amounts of the plasmid(s) (normalized to 8.1 µg of pHOGH/4.5), was added to each dish dropwise and then gently swirled to ensure that theDNA/DEAE-dextran was uniformly applied to the cells. Chloroquine diphosphate(100 µM) was then added to the medium. Cells were incubated for 4 h at37°C in 5% CO 2 -room air and then shocked for 1 min at roomtemperature with 10% DMSO in PBS, washed twice with PBS, and incubated ingrowth media containing 10% FBS overnight.
  T+ o5 C; r  Y; ^: Y
At twenty-four hours posttransfection, the cells (both cell types) werepassaged from 150-mm dishes into several 100-mm dishes (for hGH assays) or12-well trays (for luciferase assays) to ensure equal transfection efficiencybetween experimental treatments. Cotransfection with pcDNA3.1/Lac-zdemonstrated equal amounts of DNA in each of the wells. After the cells were allowed to recover for 24 h, they were treated with stimulus or vehicle(control). Sixteen to seventy-two hours following stimulus exposure, cellswere evaluated for hGH mRNA content by Northern analysis, levels of secretedhGH protein in the culture media, as described previously( 4 ), and luciferase activitywas monitored using the luciferase reporter assay system (Promega) accordingto the manufacturer's instructions. Luminescence was measured with a SiriusLuminometer (Berthold Detection Systems, Pforzheim, Germany).

RNA isolation, molecular probes, and Northern analysis. Total cellular RNA was isolated by the method of Chomczynski and Sacchi ( 17 ). Samples wereelectrophoresed on a 1% agarose formaldehyde gel and electrotransferred to acharged nylon hybridization membrane. The cDNA probes for hGH, human HO-1, andhuman GAPDH were radiolabeled with [ - 32 P]dATP using a randomprimer labeling kit, according to the instructions of the manufacturer(Invitrogen), and purified over a G-50 column (Amersham Pharmacia Biotech, Piscataway, NJ). Membranes were hybridized overnight at 60°C, washed in ahigh-stringency buffer (0.04 M sodium phosphate, 1 mM EDTA, 1% SDS) at65°C, and subjected to autoradiography. To quantitate expression levels,autoradiographs were scanned on a Hewlett-Packard ScanJet 4C using DeskScan IIsoftware and densitometry was performed using National Institutes of HealthImage 1.63 software. Experiments were internally controlled by hybridizationwith GAPDH, normalized, and expressed as a percentage of maximal expression.All experiments were repeated with at least two to three independent RNApreparations to show reproducibility.: x5 d/ ?1 S' ]' @) z
* r6 l8 X  E/ x) F/ N2 |) F7 a
Statistical analysis. Data are expressed as means ± SE. Statistical analyses were performed using ANOVA and the Student-Newman-Keuls test. All results are considered significant at P; o% W" h" e, G7 \; U0 e9 c2 h

RESULTS
4 I2 V$ q0 d  P$ h; n+ N
Induction of HO-1 mRNA by heme in HPTC and HAEC. Heme serves notonly as a substrate for the HO enzyme, but it is also a potent inducer of theHO-1 gene both in vivo and in vitro ( 11, 32, 50 ). To understand themolecular mechanism involved in heme-mediated induction of HO-1, HPTC and HAECwere treated with heme (in the form of hemin, 5 µM) for 4 h. Northern analysis demonstrated an 20- to 30-fold induction of HO-1 mRNA in bothcell types ( Fig. 1 A ). Figure 1, B and C, demonstrates that the induction of HO-1 mRNA by heminwas both time and dose dependent in HPTC. Similar results were obtained inHAEC (data not shown). Sodium hydroxide (40 µM) alone, the vehicle used inthe preparation of hemin, did not induce HO-1 mRNA( Fig. 1 B ).! B, M* c5 b/ m9 p/ J5 @/ n

Fig. 1. Induction of heme oxygenase (HO)-1 mRNA by hemin. A : confluenthuman aortic endothelial cells (HAEC; left ) or human renal proximaltubule cells (HPTC; right ) were incubated in 1 and 0.5鸖-containing media, respectively, with hemin (5 µM) for 4 h as describedin MATERIALS AND METHODS. B : confluent HPTC incubated in0.5% FBS medium containing hemin (0, 1, 2.5, 5, 7.5, or 10 µM) and vehicle,NaOH (40 µM), for 4 h. C : time course of HO-1 mRNA induction inHPTC exposed to hemin (5 µM) at the indicated times. D : graphicrepresentation showing the half-life of HO-1 mRNA following hemin stimulation.HPTC were preincubated with hemin (5 µM) for 4 h, washed, and exposed toactinomycin D (4 µM) in the absence (, solid line) or presence(, dashed line) of additional hemin (5 µM). The percentage of maximalexpression of HO-1 mRNA corrected for the internal control (GAPDH) vs. time isplotted. RNA was isolated and subjected to Northern blot analysis with a 32 P-labeled cDNA specific for HO-1 or GAPDH as described in MATERIALS AND METHODS. Results are representative of at least 2-4independent experiments.5 P, b$ i) M% [

We previously demonstrated the importance of de novo transcription bynuclear run-on analysis in HAEC and HPTC treated with hemin (5 µM) andobserved increased HO-1 gene transcription ( 4 ). To evaluate the role ofmessage stability in the induction of HO-1 mRNA following stimulation withhemin, the half-life of HO-1 mRNA was measured. Confluent HPTC were exposed tohemin (5 µM) for 4 h and washed with HBSS followed by the addition of freshmedia containing actinomycin D (4 µM) with or without additional hemin. Asshown in Fig. 1 D, thehalf-life of HO-1 mRNA was similar ( 4 h) with and without additional hemin treatment, suggesting that message stability is not involved. Thesefindings are consistent with previous studies in other cell types( 10 ) and enable us to usethese human primary cultured cells for further studies to evaluate thetranscriptional activation of the human HO-1 gene by hemin.: |' _) Z6 Z; \1 G3 J
5 i+ B. E7 d1 L
Identification of the enhancer region internal to the human HO-1gene. To identify regulatory elements in the human HO-1 gene that mediateits transcription following stimulation with hemin, multiple promoterfragments (155 bp, 4.0 kb, 4.5 kb, and 9.1 kb) were generated and incorporatedinto promoterless reporter vectors as shown in Fig. 2 and described in MATERIALS AND METHODS. Growth hormone mRNA levels, measured byNorthern analysis, were used as a direct measure of reporter genetranscription rates. HAEC transfected with hGH vectors containing -155-bp or-4.0-kb promoter constructs demonstrated no hGH mRNA induction in response tohemin (data not shown) and the -4.5- and -9.1-kb constructs resulted in amodest hemin-mediated induction in hGH mRNA ( 6.7- and 9.9-fold,respectively), as shown in Fig. 3 A.. @+ p, Z% m8 m
; \( x+ R9 M9 K' P5 V5 l( L
Fig. 3. Analysis of human growth hormone (hGH) mRNA in cells transfected withhGH-HO-1 reporter constructs. HAEC were transiently transfected with equimolaramounts of pHOGH/4.5 (-4.5-kb HO-1 promoter in hGH vector) or pHOGH/9.1(-9.1-kb HO-1 promoter in hGH vector; A ) or pHOGH/4.5/ 12.5 (enhancerwith -4.5-kb HO-1 promoter in hGH vector; B ) using DEAE/dextran and abatch transfection protocol as described in MATERIALS AND METHODS.Transfected cells were exposed to hemin (5 µM), cadmium chloride(CdCl 2; 10 µM), transforming growth factor- 1 (TGF- 1; 2 ng/ml), or 13-hydroperoxyoctadecadienoic acid(13-HPODE; 25 µM) for 16 h and analyzed for reporter gene expression byNorthern analysis. C : HPTC were transiently transfected withequimolar amounts of pHOGH/4.5 or pHOGH/4.5/ 12.5 using DEAE/dextran and abatch transfection protocol as described in MATERIALS AND METHODS.Cells were exposed to hemin (5 µM) or CdCl 2 (10 µM) for 16 hbefore collection of RNA for reporter gene analysis. Total RNA was isolatedand Northern analysis was performed using 32 P-labeled hGH and GAPDHcDNA probes. Results are representative of at least 3 independentexperiments.; X: r3 t* o3 S) q& ~' {
0 N* U; H. r- X7 [  I6 w/ ^0 a5 a
Because levels of reporter gene activity did not correlate with steady-state Northern levels of HO-1 induction with hemin, we hypothesizedthat additional regulatory sequences might exist within the HO-1 gene.Therefore, a construct containing the 12.5-kb region internal to the HO-1 genewith the 4.5-kb HO-1 promoter (pHOGH/4.5/ 12.5) was generated. Unfortunately,we were unable to subclone the 12.5-kb fragment with the 9.1-kb HO-1 promoter.The construct, pHOGH/4.5/ 12.5, demonstrated significant hemin- as well asCdCl 2 -inducible hGH mRNA in HAEC (16.1- and 32.4-fold,respectively; Fig.3 B ). Similar results were observed in HPTC: pHOGH/4.5yielded a 6.6- and 4.3-fold increase in response to hemin andCdCl 2, respectively, and pHOGH/4.5/ 12.5 resulted in a 27- and18-fold increase in response to hemin and CdCl 2, respectively, overcontrol (untreated cells) transfected with pHOGH/4.5( Fig. 3 C ). Theseresults indicate the presence of sequences internal to the HO-1 gene thatpotentially function as an enhancer. Interestingly, the putative enhancer region does not function for all known inducers of HO-1 as pHOGH/4.5/ 12.5 failed to respond to TGF- 1 (2 ng/ml),H 2 O 2 (200 µM), or 13-HPODE (25 µM)( Fig. 3 B ).0 q9 u! e6 s. v6 V1 J, [
+ ~5 Z) n1 Y! t
Reporter activity, as measured by hGH protein or luciferase activity,yielded results similar to hGH mRNA measurements. No promoter activity wasdetected in cells transfected with hGH constructs containing -155-bp or-4.0-kb promoter fragments and treated with hemin (5 µM)( Fig. 4 A ). Cellstransfected with vectors containing the -4.5- or -9.1-kb promoter constructs and exposed to hemin (5 µM) demonstrated a 3.6- and 4.5-fold, respectively,increase in hGH protein over untreated cells ( Fig. 4 A ). Theconstruct containing the 12.5-kb region internal to the HO-1 gene, inconjunction with the -4.5-kb HO-1 promoter region, demonstrated elevated basallevels as well as significant hemin-inducible hGH protein levels, 5.1-fold over untreated cells ( Fig.4 A ). Luciferase reporter vectors containing HO-1 genefragments were generated to confirm the results with another reporter gene aswell as ease of cloning and simplicity of the assay. Luciferase/HO-1 vectorsyielded results similar to hGH reporter vectors in both HAEC and HPTC( Fig. 4, B and C, respectively).
( y2 c/ G# T- `# A
Fig. 4. Analysis of reporter activity in cells transfected with HO-1 reporterconstructs in response to hemin. A : HAEC were transiently transfectedwith equimolar amounts of pHOGH/155 (-155-bp HO-1 promoter), pHOGH/4 (-4.0-kbHO-1 promoter), pHOGH/4.5 (-4.5-kb HO-1 promoter), pHOGH/9.1 (-9.1-kb HO-1promoter), and pHOGH/4.5/ 12.5 (enhancer with -4.5-kb HO-1 promoter) asdescribed in MATERIALS AND METHODS. Cells were exposed to hemin (5µM) for 72 h, and secreted hGH protein in the media was measured using aradioimmunoassay kit as described previously( 4 ). Results are derived from 2independent experiments. HAEC ( B ) and HPTC ( C ) weretransiently transfected with equimolar amounts of pHOGL3/4.5 (-4.5-kb HO-1promoter in luciferase vector) or pHOGL3/4.5/ 12.5 (enhancer with -4.5-kb HO-1promoter in luciferase vector), exposed to hemin (5 µM) for 16 h, andluciferase activity was measured as described in MATERIALS AND METHODS. Results are derived from 2 independent experiments with 3-12replicates per group. Open bars represent untreated (control) cells, andfilled bars represent hemin-stimulated cells. # P * P, ]5 p( G; U' r) d. V0 u& n  f

Orientation-independent effects of the HO-1 enhancer region. Todetermine whether the  12.5-kb fragment functions as a true enhancer, bothorientations (5'-3' and 3'-5') were cloned into pHOGH/4.5 to give pHOGH/4.5/ 12.5 and pHOGH/4.5/ 12.5 (3'-5'). Similar levels of heme-mediated induction, as measured by Northern analysis ofhGH mRNA, were observed, regardless of orientation ( Fig. 5 ). In addition, todetermine whether the enhancer could function with a heterologous promoter,the  12.5-kb fragment was cloned into a hGH vector containing a TK promoter(pHOTKGH/ 12.5), which is a 200-bp minimal, TATA-containing promoter.Transient transfection of the 12.5-kb fragment in conjunction with a TK promoter failed to respond to hemin, as assessed by Northern analysis of hGHmRNA (data not shown), suggesting that the enhancer only functions with theHO-1 promoter.
+ s0 s9 |' X5 h* B- j7 Y) N
Fig. 5. Orientation-independent effects of the HO-1 enhancer. HAEC were transientlytransfected with equimolar amounts of pHOGH/4.5 (-4.5-kb HO-1 promoter in hGHvector), pHOGH/4.5/ 12.5 (5'-3' orientation of the enhancer with-4.5-kb promoter in hGH vector), and pHOGH/4.5/ 12.5 (3'-5')(3'-5' orientation of the enhancer with -4.5-kb promoter in hGHvector) as described in MATERIALS AND METHODS. Cells were exposedto hemin (5 µM) for 16 h, and total RNA was isolated for Northern analysisusing 32 P-labeled hGH and GAPDH cDNA probes.+ f+ W; T% ?" ?7 a( O5 z
+ W7 i( a- R+ ~! W) J( G2 m
To further define the relevant sequences in the HO-1 promoter region thatare required for enhancer function, the  12.5-kb enhancer fragment wasevaluated with a 3.5-kb promoter of the human HO-1 gene pHOGL3/3.5/ 12.5. Asshown in Fig. 6, the  12.5-kb fragment requires the 4.5-kb HO-1 promoter fragment to function, because the3.5-kb promoter with the  12.5-kb enhancer did not demonstrate any basal orhemin-inducible activity. The results demonstrate that enhancer sequencesinternal to the human HO-1 gene (within 12.5 kb), in conjunction with regions between -3.5 and -4.5 kb of the HO-1 promoter, mediate transcriptional activation of the human HO-1 gene by hemin and the enhancer is likely composedof a complex set of interacting elements.

Fig. 6. Requirement of the -4.5-kb HO-1 promoter for enhancer function. HPTC weretransiently transfected with equimolar amounts of pHOGL3/3.5/ 12.5 (enhancerwith -3.5-kb HO-1 promoter in a luciferase vector) or pHOGL3/4.5/ 12.5 asdescribed in MATERIALS AND METHODS. Cells were exposed to hemin (5µM) for 16 h, and luciferase assays were performed. Results are derivedfrom 2 independent experiments with 6 replicates per group. Open barsrepresent untreated (control) cells, and filled bars representhemin-stimulated cells. * P
9 e- O9 L) `; h7 n
DISCUSSION

Induction of HO-1 is an adaptive and protective response in renalischemia-reperfusion injury( 40 ), nephrotoxin-induced renal injury ( 1, 41 ), organ transplantation( 23 ), acute glomerulonephritis ( 47 ), rhabdomyolysis( 32, 34 ), as well as nonrenalsettings of tissue injury ( 3 ).The protective effects of HO-1 were first recognized due to its robustinduction in cells/tissues exposed to a wide variety of stimuli that areotherwise injurious ( 24, 32 ). The mechanisms underlyingHO-1 induction by multiple inducers, including heme, CdCl 2, nitricoxide (NO), oxidized LDL, and cytokines, are complex and regulatedpredominantly at the transcriptional level. Increased mRNA stability has beenreported to contribute to induction of HO-1 by NO donors in primary humanembryonic fibroblasts with a strong correlation between the rate of NO releaseand the half-life of HO-1 mRNA( 14 ). Furthermore, HO-1 mRNAstability was independent of RNA or protein synthesis. Our previous studiesusing nuclear run-on assays ( 4 )combined with the present half-life experiments with actinomycin D suggest that mRNA stability is not involved in heme-mediated HO-1 induction in renalepithelial and endothelial cells, results consistent with previously publishedobservations in other cell types ( 10 )." C# T4 P7 j5 V+ v% a* P7 @: |$ D

Thus far, studies on mammalian HO-1 gene regulation have focused mainly onthe mouse HO-1 gene( 5 - 13 ),where multiple inducer-specific stress response elements have been localizedwithin 10 kb of the 5'-flanking region. Two distal promoter regions,E 1 and E 2, at -4.0 and -10 kb, respectively, arerequired for induction of the mouse HO-1 gene in response to heme, heavymetals, H 2 O 2, and sodium arsenite( 5 - 7, 13 ). It was proposed thatactivation of the mouse HO-1 gene by most stimuli is mediated exclusively viathe E 1 and/or E 2 regions. These regions containconsensus binding sites for the NF-E 2 -related factor Nrf2, andoverexpression studies coupled with mutations in these binding sites have implicated the involvement of Nrf2 in the induction of the mouse HO-1 gene( 12, 20 ).
4 d  f6 u' j% S
To test whether promoter elements in the 5'-flanking region of thehuman HO-1 gene can analogously control heme-mediated induction, we initiallyanalyzed several human HO-1 promoter constructs up to -9.1 kb, which containsequences potentially similar to the E 1 and E 2 regionsin the mouse gene. The 4.5- and 9.1-kb promoter fragments demonstrated amodest induction with heme; however, unlike the E 1 andE 2 regions of the mouse gene, these constructs did not completelyrecapitulate steady-state Northern levels of HO-1 induction. In addition,unlike the mouse promoter, we also found that the human HO-1 promoter constructs showed no response to other known HO-1 stimuli, including 13-HPODE,hyperoxia, curcumin, and TGF- 1, in human cultured cells( 4, 19, and unpublishedobservations).

In our efforts to mimic endogenous, stimulus-dependent transcription levels, we hypothesized that additional regulatory element(s) were necessaryfor hemeand cadmium-mediated induction of the human HO-1 gene. To identifyother relevant regulatory sequences, we tested the entire  12.5-kb HO-1 gene,including introns and exons, in conjunction with the -4.5-kb human HO-1promoter and observed significant hemeand cadmium-dependent induction ofreporter activity. Furthermore, this enhancer region was specific for heme andcadmium and did not function for other known HO-1 stimuli, such asTGF- 1, H 2 O 2, or 13-HPODE. The large  12.5-kb fragment, in part, satisfied the characteristics of a true enhancerregion by functioning in an orientation-independent manner. However, when theHO-1 promoter was replaced with a heterologous TK promoter, no induction byheme was observed, thus demonstrating that this region functions as agene-specific enhancer. This is substantiated by our findings that theenhancer region requires a portion of the HO-1 promoter to mimic endogenous levels of induction. Specifically, we demonstrated that a region between -4.5and -3.5 kb is required for enhancer function. Our data also indicated theimportance of an additional region between -4.5 and -9.1 kb for heme-mediatedHO-1 promoter activity. However, we were unable to evaluate the  12.5-kbenhancer in the context of the -9.1-kb promoter construct due to our inability to successfully subclone both of these large fragments into a single plasmid.Studies to further delineate the regulatory sequences within the enhancerregion are currently in progress. As these regions become more well defined,constructs containing the relevant sequences within the enhancer will beevaluated with the -9.1-kb HO-1 promoter.

Regulatory sequences functioning as enhancers have been identified withinintrons of several genes, including manganese superoxide dismutase,platelet-derived growth factor A, and alcohol dehydrogenase-1 ( 15, 37, 45 ). We performed sequencecomparisons with the 12.5-kb enhancer region and did not identify anysequences that are linked to known heme or cadmium response elements. Alam andDen ( 8 ) previously reportedthat the region analogous to our enhancer region containing the entire proteincoding region of the mouse HO-1 gene coupled to the mouse HO-1 promoter didnot respond to heme or cadmium in rodent cells. Two possibilities couldaccount for the lack of response. On the one hand, the limited size ( 3.0kb) of the mouse HO-1 promoter used in these studies may account for thenegative results ( 8 ). On theother hand, perhaps it is a reflection of the differential regulation of themouse vs. human HO-1 genes.( Z5 M- B! f! B: s; o

Our observations indicate that the human HO-1 gene requires regulatorysequences that differ from those previously implicated for the mouse gene.Other interspecies differences in HO-1 gene regulation have also been noted.For example, the human HO-1 promoter contains a GT repeat region that isabsent in the mouse HO-1 gene( 49 ). Deletion of this repeatregion results in a significant elevation of basal HO-1 promoter activity (unpublished observations), similar to previously reported results using HO-1promoter constructs with shorter GT repeats( 49 ), suggesting that thishuman-specific region may function as a negative regulator of HO-1 geneexpression. Studies have also demonstrated that HO-1 is induced by hypoxia inrat, bovine, mouse, and monkey cells but is rather repressed in human cells ( 25, 27 ). Similarly, HO-1 has beenpopularized as heat shock protein 32 based on its inducibility by heat shockin rodent cells ( 38, 39 ). However, heat shock doesnot induce HO-1 in human cells( 35 ). Although cytokinesinduce HO-1 in rodent cells, previous studies have reported thatinterferon- decreases human HO-1 gene expression( 43 ). Furthermore, a putativecadmium response element has been identified in the human HO-1 promoter ( 44 ); however, the cadmiumresponse element in the mouse HO-1 gene is immediately 3' to this regionand is an Nrf2 consensus sequence( 13 ). Collectively, theseobservations suggest that the underlying molecular mechanisms that regulatethe human HO-1 gene are different from those described for mouse HO-1 geneexpression.; |7 `! e7 E9 Z- O' I  y

Recent studies have demonstrated that the expression of the mouse HO-1 geneis regulated through antagonism of transcription activators and the repressorBach1 ( 42 ). Under normalphysiological conditions, HO-1 expression is repressed by Bach1, and increased heme levels displace Bach1 from the E 1 and E 2 regions,allowing activators to bind to the regulatory sequences( 42 ). The authors suggest thatthe transcriptional regulation of the mouse HO-1 gene involves a directsensing of heme levels by Bach1, generating a feedback loop by the substrate,as in the lac operon. Studies to explore the involvement of Bach1 in humanHO-1 gene regulation will be of interest.
4 u; [' {; v  {3 }6 s# o
In summary, our data provide evidence that at least two regions in thehuman HO-1 promoter, one between -4.5 and -4.0 and another between -9.1 and-4.5, are required for activation of the HO-1 gene by heme. Most importantly,we identified a novel transcriptional enhancer internal to the human HO-1 genethat, in conjunction with the HO-1 promoter, recapitulates a steady-statelevel expression of HO-1 by heme and cadmium. Studies on regulation of thehuman HO-1 gene are biologically relevant because of the well-documentedbeneficial effects of HO-1 activity that include degradation of heme (a toxicprooxidant), generation of bilirubin (an antioxidant), coinduction of ferritin(an intracellular repository for iron), and the antiapoptotic and anti-inflammatory effects of CO( 36 ). We are continuing our investigation of the molecular mechanisms involved in the regulation of thehuman HO-1 gene in an effort to fine tune endogenous HO-1 gene expression toultimately optimize its potential as a therapeutic modality.! t$ X  L! u0 I% `
4 j% ]2 H1 T) T% M4 q& n6 f
DISCLOSURES; ^5 n& _- s; v" H3 Y, \( G# Q, @4 ~/ n
4 H5 X* }* P0 T* I& O6 I* ~# T
This work was supported by National Institutes of Health Grants K-08-DK-02446 and R-01-DK-59600 to A. Agarwal and Grant K-01-DK-02902 to N.Hill-Kapturczak.
      【参考文献】- l6 y* {8 g6 ^
Abraham NG,Lavrovsky Y, Schwartzman ML, Stoltz RA, Levere RD, Gerritsen ME, Shibahara S,and Kappas A. Transfection of the human heme oxygenase gene into rabbitcoronary microvessel endothelial cells: protective effect against heme andhemoglobin toxicity. Proc Natl Acad Sci USA 92: 6798-6802,1995.
% R" a- |( \! k3 [

Agarwal A,Balla J, Alam J, Croatt AJ, and Nath KA. Induction of heme oxygenase intoxic renal injury: a protective role in cisplatin nephrotoxicity in the rat. Kidney Int 48:1298-1307, 1995.
6 K( H* }7 Z4 s$ R% y/ n

Agarwal A andNick HS. Renal response to tissue injury: lessons from heme oxygenase-1gene ablation and expression. J Am Soc Nephrol 11: 965-973,2000.! Q, F3 f4 j; F1 d! a
: R" O$ k* x$ q. u! j+ K

5 B: @6 P- A2 G- M& H" M; [
Agarwal A,Shiraishi F, Visner GA, and Nick HS. Linoleyl hydroperoxidetranscriptionally upregulates heme oxygenase-1 gene expression in human renalepithelial and aortic endothelial cells. J Am SocNephrol 9:1990-1997, 1998.+ b7 e$ g6 G5 O+ }- t( d
: s5 W, L  Q) q9 s
! H2 f" M; Z. Z: r6 K. ]

Alam J. Multiple elements within the 5' distal enhancer of the mouse hemeoxygenase-1 gene mediate induction by heavy metals. J BiolChem 269:25049-25056, 1994.: _8 _) @) y& b1 x% r! I+ f) R2 J
4 I7 |) ?# J8 f1 m' x0 D( n# A
. B$ R( ^$ T. r; U: U) o% h

Alam J, Cai J,and Smith A. Isolation and characterization of the mouse heme oxygenase-1gene. Distal 5' sequences are required for induction by heme and heavymetals. J Biol Chem 269:1001-1009, 1994.

3 S, z' R* j: Q9 F% m
Alam J, CamhiS, and Choi AM. Identification of a second region upstream of the mouseheme oxygenase-1 gene that functions as a basal level and inducer-dependenttranscription enhancer. J Biol Chem 270: 11977-11984,1995.9 g; `7 n) A* N1 p, H- ~& [

1 h9 B0 G. U: t9 c6 y+ M0 K7 s
& t+ G% B3 x* p$ x1 z* X8 r
Alam J and DenZ. Distal AP-1 binding sites mediate basal level enhancement and TPAinduction of the mouse heme oxygenase-1 gene. J BiolChem 267:21894-21900, 1992.
% S) u( ^' K' J! w. P" |6 H
2 ^0 A; h9 X  @( J! \6 [( |: ^
2 J3 L  J- {' E% C; ~: p1 a
Alam J, KilleenE, Gong P, Naquin R, Hu B, Stewart D, Ingelfinger JR, and Nath KA. Hemeactivates the heme oxygenase-1 gene in renal epithelial cells by stabilizingNrf2. Am J Physiol Renal Physiol 284: F743-F752,2003.+ I$ f2 P0 F' W' b' g& t

# J2 L! [; Y3 ?* h" B
# {' v( a; X# L  {9 n1 z
Alam J,Shibahara S, and Smith A. Transcriptional activation of the heme oxygenasegene by heme and cadmium in mouse hepatoma cells. J BiolChem 264:6371-6375, 1989.2 n) F5 r. Y9 J2 M$ A4 y
7 p6 v! ]$ `# S& ~) t- Y! s

0 ?' [5 Y7 A1 @
Alam J andSmith A. Receptor-mediated transport of heme by hemopexin regulates geneexpression in mammalian cells. J Biol Chem 264: 17637-17640,1989.

- ?) |3 C& t! H, J
Alam J, StewartD, Touchard C, Boinapally S, Choi AM, and Cook JL. Nrf2, a Cap'n'Collartranscription factor, regulates induction of the heme oxygenase-1 gene. J Biol Chem 274:26071-26078, 1999.* q$ D" t$ X3 Q- `2 Q' Q
1 I) \# m8 u: R8 x5 x
& S2 E% @* q! @  h
( X* w( V" K, l- S7 F- Z' o! }
Alam J, WicksC, Stewart D, Gong P, Touchard C, Otterbein S, Choi AM, Burow ME, and TouJ. Mechanism of heme oxygenase-1 gene activation by cadmium in MCF-7mammary epithelial cells. Role of p38 kinase and Nrf2 transcription factor. J Biol Chem 275:27694-27702, 2000.! D! g% a& `2 z4 g! }( o; l

6 C2 V! I8 [. U. w8 d4 ]
" r0 ~) x5 L+ J0 `% }( `
Bouton C andDemple B. Nitric oxide-inducible expression of heme oxygenase-1 in humancells. Translation-independent stabilization of the mRNA and evidence fordirect action of nitric oxide. J Biol Chem 275: 32688-32693,2000.
1 }4 O) c6 E* D
2 w% [0 u! j- [( ~) f/ u0 I' z" Y: x
) `! ]# B5 \% K8 c/ \$ Y
Callis J, FrommM, and Walbot V. Introns increase gene expression in cultured maize cells. Genes Dev 1:1183-1200, 1987.2 e1 o/ T) _0 P" h+ W" ^5 \2 G6 M9 r# h
0 t* Y6 F) ~* W( @8 Z6 O

# n; F# ~5 X5 _4 p- [' q
Chen YH, LinSJ, Lin MW, Tsai HL, Kuo SS, Chen JW, Charng MJ, Wu TC, Chen LC, Ding YA, PanWH, Jou YS, and Chau LY. Microsatellite polymorphism in promoter of hemeoxygenase-1 gene is associated with susceptibility to coronary artery diseasein type 2 diabetic patients. Hum Genet 111: 1-8,2002.8 y$ u" X5 X$ m0 [# f
+ T7 }  v8 G* B: k( w- X
! @* ?$ _3 Z: H/ ^2 W3 `
3 ?: U+ @  A* e/ P0 W# m+ y
Chomczynski P and Sacchi N. Single-step method of RNA isolation by acid guanidiniumthiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156-159,1987.4 b% n$ f5 I1 L. c; |" e. o3 a$ @

Exner M,Schillinger M, Minar E, Mlekusch W, Schlerka G, Haumer M, Mannhalter C, andWagner O. Heme oxygenase-1 gene promoter microsatellite polymorphism isassociated with restenosis after percutaneous transluminal angioplasty. J Endovasc Ther 8:433-440, 2001.7 v2 p' @0 ]& {! B& S" W
, R& F+ r! `; N! k) S
; `% m, z7 N7 W. L" t0 I

Fogg S, AgarwalA, Nick HS, and Visner GA. Iron regulates hyperoxia-dependent human hemeoxygenase 1 gene expression in pulmonary endothelial cells. Am JRespir Cell Mol Biol 20:797-804, 1999.
+ N* ^3 u$ t& p( ^) o% n0 X! a
# x1 z6 i7 W& H0 s5 p* G0 O

Gong P, StewartD, Hu B, Li N, Cook J, Nel A, and Alam J. Activation of the mouse hemeoxygenase-1 gene by 15-deoxy- 12, 14 -prostaglandinJ 2 is mediated by the stress response elements and transcriptionfactor Nrf2. Antioxid Redox Signal 4: 249-257,2002.

6 ~! {+ R6 x: k/ L4 E6 d

Hill-Kapturczak N, Chang SH, and Agarwal A. Heme oxygenase andthe kidney. DNA Cell Biol 21:307-321, 2002.
8 M% e% W% D, ~: w! U
6 v' i1 [, O1 J! o+ x

Jeney V, BallaJ, Yachie A, Varga Z, Vercellotti GM, Eaton JW, and Balla G. Pro-oxidantand cytotoxic effects of circulating heme. Blood 100: 879-887,2002.

Katori M,Busuttil RW, and Kupiec-Weglinski JW. Heme oxygenase-1 system in organtransplantation. Transplantation 74: 905-912,2002.+ R/ y3 u2 u1 C6 w& n3 N4 b

: h8 A: M$ X1 [$ x
Keyse SM andTyrrell RM. Heme oxygenase is the major 32-kDa stress protein induced inhuman skin fibroblasts by UVA radiation, hydrogen peroxide, and sodiumarsenite. Proc Natl Acad Sci USA 86: 99-103,1989.
- P4 X0 A0 e0 v" S1 a
7 _0 o; U+ z3 W+ z
5 r% v, o) C- l' O" |
Kitamuro T,Takahashi K, Ogawa K, Udono-Fujimori R, Takeda K, Furuyama K, Nakayama M, SunJ, Fujita H, Hida W, Hattori T, Shirato K, Igarashi K, and Shibahara S. Bach1 functions as a hypoxia-inducible repressor for the heme oxygenase-1 genein human cells. J Biol Chem 278: 9125-9133,2003.  N* A* A$ L2 ~! A5 M. }
% q- m& n4 ^+ R2 |

Kutty RK, KuttyG, Rodriguez IR, Chader GJ, and Wiggert B. Chromosomal localization of thehuman heme oxygenase genes: heme oxygenase-1 (HMOX1) maps to chromosome 22q12and heme oxygenase-2 (HMOX2) maps to chromosome 16p13.3. Genomics 20:513-516, 1994.3 s- ^+ y% M7 A3 o  {5 s" v" ~

: \9 S; ~" [# |/ f$ M6 ^
* o9 P3 K% U' Q; i
Lee PJ, JiangBH, Chin BY, Iyer NV, Alam J, Semenza GL, and Choi AM. Hypoxia-induciblefactor-1 mediates transcriptional activation of the heme oxygenase-1 gene inresponse to hypoxia. J Biol Chem 272: 5375-5381,1997.
7 s$ l$ a" E4 ?1 A% C
! X3 p' U  u+ i3 X, Y/ p  p( y
% L) B/ ^' |& t% }" f) R' u
Maines MD. The heme oxygenase system: a regulator of second messenger gases. Annu Rev Pharmacol Toxicol 37:517-554, 1997.3 y! l4 g' M& u0 L# J0 ~) u8 ]0 n

+ |3 G6 G! w& V- \9 q
. d! J* h: Q1 m+ U# u. R, l5 x
Maines MD,Mayer RD, Ewing JF, and McCoubrey WK Jr. Induction of kidney hemeoxygenase-1 (HSP-32) mRNA and protein by ischemia/reperfusion: possible roleof heme as both promotor of tissue damage and regulator of HSP32. JPharmacol Exp Ther 264:457-462, 1993.
3 p3 k: v: P: U7 J1 Q/ [

McCoubrey WK Jr, Huang TJ, and Maines MD. Isolation and characterization of a cDNA fromthe rat brain that encodes hemoprotein heme oxygenase-3. Eur JBiochem 247:725-732, 1997./ {1 m* {) U# C* u

% e. F" y3 w4 X$ \

Nath KA,Agarwal A, and Vogt BV. Heme oxygenase: cytoprotective and cytotoxiceffects. Contemp Issues Nephrol 30: 97-118,1995.* n# S8 P* `. ]& ?5 [* D

Nath KA, BallaG, Vercellotti GM, Balla J, Jacob HS, Levitt MD, and Rosenberg ME. Induction of heme oxygenase is a rapid, protective response in rhabdomyolysisin the rat. J Clin Invest 90:267-270, 1992.
7 y/ i& N9 a9 C+ j' U9 v5 Z6 E

6 a$ c* A* d% G
Nath KA, GrandeJP, Croatt AJ, Likely S, Hebbel RP, and Enright H. Intracellular targetsin heme protein-induced renal injury. Kidney Int 53: 100-111,1998.

9 g; l7 }7 [5 `2 _' s0 u# }' o
' i" `3 z, a+ c6 d8 V, y& w) m
Nath KA,Haggard JJ, Croatt AJ, Grande JP, Poss KD, and Alam J. Theindispensability of heme oxygenase-1 in protecting against acute hemeprotein-induced toxicity in vivo. Am J Pathol 156: 1527-1535,2000.1 \2 v+ N5 E& P3 k& z! C; j& O

  c: w' I6 D+ D& V% ^

Okinaga S,Takahashi K, Takeda K, Yoshizawa M, Fujita H, Sasaki H, and Shibahara S. Regulation of human heme oxygenase-1 gene expression under thermal stress. Blood 87:5074-5084, 1996." W$ O* J/ r0 x  I# g2 B

+ W& i# e% i( c" C' o; d

Platt JL andNath KA. Heme oxygenase: protective gene or Trojan horse. NatMed 4: 1364-1365,1998.

9 C5 L, G% W2 i! {" |
Rogers RJ,Chesrown SE, Kuo S, Monnier JM, and Nick HS. Cytokine-inducible enhancerwith promoter activity in both the rat and human manganese-superoxidedismutase genes. Biochem J 347:233-242, 2000.
; l( U, b% P0 O, b% Y! v
* M1 S$ K% r' @1 g' d

Shibahara S,Kitamuro T, and Takahashi K. Heme degradation and human disease: diversityis the soul of life. Antioxid Redox Signal 4: 593-602,2002.

5 i' c# l' t% g% I$ @4 m

Shibahara S,Muller RM, and Taguchi H. Transcriptional control of rat heme oxygenase byheat shock. J Biol Chem 262:12889-12892, 1987.0 M$ H  R9 x' Y

- |* I: n( y+ `

Shimizu H,Takahashi T, Suzuki T, Yamasaki A, Fujiwara T, Odaka Y, Hirakawa M, Fujita H,and Akagi R. Protective effect of heme oxygenase induction in ischemicacute renal failure. Crit Care Med 28: 809-817,2000.5 Q% m& c* A& I; o
  A6 A. o$ C: L8 a: p

% y8 I# q7 N$ i+ _# W6 C
Shiraishi F,Curtis LM, Truong L, Poss K, Visner GA, Madsen K, Nick HS, and Agarwal A. Heme oxygenase-1 gene ablation or expression modulates cisplatin-induced renaltubular apoptosis. Am J Physiol Renal Physiol 278: F726-F736,2000.
/ H! v$ m) f6 U! x5 W
) Z9 `9 t) V0 Z, f& ~
0 |9 y' Q8 O" f) t; z, q
Sun J, HoshinoH, Takaku K, Nakajima O, Muto A, Suzuki H, Tashiro S, Takahashi S, ShibaharaS, Alam J, Taketo MM, Yamamoto M, and Igarashi K. Hemoprotein Bach1regulates enhancer availability of heme oxygenase-1 gene. EMBOJ 21: 5216-5224,2002.

4 n3 |' e8 T! B  _/ ]- _
' M9 ^( P4 y& R. W* J! K* w5 z+ e
Takahashi K,Nakayama M, Takeda K, Fujia H, and Shibahara S. Suppression of hemeoxygenase-1 mRNA expression by interferon- in human glioblastoma cells. J Neurochem 72:2356-2361, 1999.

Takeda K,Ishizawa S, Sato M, Yoshida T, and Shibahara S. Identification of acis-acting element that is responsible for cadmium-mediated induction of thehuman heme oxygenase gene. J Biol Chem 269: 22858-22867,1994.
$ m! O  S1 b9 G0 P
/ i( }8 {& ?: R* g& X

Takimoto Y andKuramoto A. Presence of a regulatory element within the first intron ofthe human platelet-derived growth factor-A chain gene. Jpn J CancerRes 84:1268-1272, 1993.% }1 r! E# t- K0 Q& R! Z3 f
' Q* c/ m1 t8 ?+ f" }
3 k4 p0 x' B" a9 O8 @& _, K

Visner GA,Staples ED, Chesrown SE, Block ER, Zander DS, and Nick HS. Isolation andmaintenance of human pulmonary artery endothelial cells in culture isolatedfrom transplant donors. Am J Physiol Lung Cell MolPhysiol 267

406-L413, 1994.
, H0 A' i6 B( ?) n

Vogt BA,Shanley TP, Croatt A, Alam J, Johnson KJ, and Nath KA. Glomerularinflammation induces resistance to tubular injury in the rat. A novel form ofacquired, heme oxygenase-dependent resistance to renal injury. JClin Invest 98:2139-2145, 1996.. Y$ _+ l" s% ~6 B# @3 [

. G4 N5 G5 u- L7 }2 D( Q; |

Yachie A, NiidaY, Wada T, Igarashi N, Kaneda H, Toma T, Ohta K, Kasahara Y, and KoizumiS. Oxidative stress causes enhanced endothelial cell injury in human hemeoxygenase-1 deficiency. J Clin Invest 103: 129-135,1999.
! n! p) s2 J% n/ u) l  Z1 J3 g

Yamada N,Yamaya M, Okinaga S, Nakayama K, Sekizawa K, Shibahara S, and Sasaki H. Microsatellite polymorphism in the heme oxygenase-1 gene promoter isassociated with susceptibility to emphysema. Am J HumGenet 66:187-195, 2000.
/ v+ _( Y  W  G" M
3 J' n: D  a0 h

Yoshida T, BiroP, Cohen T, Muller RM, and Shibahara S. Human heme oxygenase cDNA andinduction of its mRNA by hemin. Eur J Biochem 171: 457-461,1988.

收藏0 分享0 顶0 踩0