A role for angiotensin II AT1 receptors in ureteric bud cell branching - 《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 ... 22 下一页

返回列表

查看: 456000\|回复: 216	go A role for angiotensin II AT1 receptors in ureteric bud cell branching [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 13:41 |只看该作者 |倒序浏览 |打印

作者：Igor V. Iosipiv and Mercedes Schroeder作者单位：Section of Pediatric Nephrology, Department of Pediatrics, TulaneUniversity Health Sciences Center, New Orleans, Louisiana 70112   W. i7 T9 d, J
   : O- U: C) Z; V
   ' c- j' V$ r: @& n2 h. H! M5 v
      . v! ?7 |# m, m# S



         ' Q5 F( T" u# ~9 t) w* J; `) z
         ' x( }+ a  v! E4 w( s

      ( ~2 f" S" j& w% I) l6 n. ]* c
      # F4 K4 R( e: p, b* i
      : `. X2 F2 j+ x( A
      【摘要】
   Gene-targeting studies in mice demonstrate that the renin-angiotensin system is required for the proper development of the renal medulla. In theabsence of angiotensin II (ANG II) or the ANG II type 1 (AT 1 )receptor, mice exhibit poor papillary development and a severeurinary-concentrating defect. These findings imply that the ureteric bud (UB)and its branches are targets for ANG II actions during renal development.However, direct evidence linking ANG II with UB-branching morphogenesis doesnot exist. Using immunohistochemistry, we demonstrated that UB-derived epithelia express angiotensinogen (Ao) and the AT 1 receptor duringmurine metanephrogenesis. Ao and AT 1 receptors are expressed in theUB branches and to a lesser extent in the stromal mesenchyme. AT 1 receptor expression in UB-derived epithelia increased from embryo day12 to day 16 and was observed on both luminal and basolateralmembranes. In accord with these findings, cultured murine UB cells expressAT 1 receptor protein and mRNA. Treatment of UB cells cultured inthree-dimensional type I collagen gels with ANG II(10 - 7 to 10 - 5 M)elicits a dose-related increase in the number of cells that have primary andsecondary branches. These effects of ANG II on UB branching are abrogated bypretreatment with the AT 1 receptor antagonist candesartan. Thesedata demonstrate a direct and independent role for ANG II acting viaAT 1 receptors on UB cell branching in vitro. The presence of Ao inthe stroma and AT 1 on UB cells supports the notion that cross talkbetween stroma and epithelial cells is crucial to epithelial branchingmorphogenesis in the developing kidney. ' T) y1 f8 D( p/ M" g$ V5 b; d
      【关键词】 kidney development renin
   THE KIDNEY IS FORMED BY RECIPROCAL inductive interactionsbetween the ureteric bud (UB) and the adjacent metanephric mesenchyme (MM;Ref. 8 ). The MM induces the UBto grow/elongate and branch repeatedly (a process known as branchingmorphogenesis) to form the renal collecting ducts, calyces, pelvis, andureters. In turn, the UB induces the MM to condense and subsequently differentiate into the glomeruli, proximal and distal tubules, and stromalmesenchyme (SM; Refs. 1, 5 ).9 w7 O$ D+ [& b9 V
) f% N/ h, d9 f+ H7 y! W4 J
Pharmacological and gene-targeting studies have clearly demonstrated thatthe renin-angiotensin system (RAS) plays an essential role duringnephrogenesis ( 16, 20, 23, 39 ). The developing kidneyexpresses all the components of the RAS( 7, 22, 41 ). The activity of the renalRAS is high during fetal and neonatal life and declines during postnatalmaturation ( 7, 41 ). Renin mRNA and ANG IIlevels are 20- and 6-fold higher, respectively, in newborn than adult kidneys( 7, 43 ). Interestingly, recent studies in renin knock-in reporter mice demonstrate that juxtaglomerular renin-producing cells originate from the MM and SM at embryo days ( E)11 and 12 before vessel development has occurred( 15 ). During early fetal ratnephrogenesis, immunoreactive angiotensinogen (Ao) is highly expressed in theloose SM ( 28 ). To this end, the presence of both Ao and renin in the SM that surrounds the UB branchessuggests a potential novel role for local ANG II generation in the paracrineregulation of growth and differentiation of the UB.
: r8 L- @1 o' x  U( Y# c5 w& U# Q
AT 1 and AT 2 receptors are abundantly expressed in thenephrogenic zone ( 22 ).However, the ontogenic expression of these receptors in the kidney differs:AT 2 receptors are expressed earlier than AT 1 receptors,peak during fetal metanephrogenesis, and rapidly decline postnatally( 12 ). AT 1 receptorexpression increases during gestation, peaks perinatally, and declines gradually thereafter ( 12 ). Thespatial distribution of AT 1 and AT 2 receptor mRNA duringontogeny is also contrasting. AT 1 receptor mRNA is present inmature and maturing glomeruli and in distal and proximal tubules.AT 2 receptor mRNA is present in mesenchymal cells adjacent to thestalk of the UB ( 12 ). Inaddition, AT 1 receptor protein is expressed in the UB and itsderivatives of the rat metanephros( 28 ). Thus AT 1 expression appears to correlate with the differentiation and proliferation ofglomerular and tubular cells, whereas AT 2 expression is associatedwith the mesenchymal-epithelial interactions.
: D# v0 K; W4 F5 `0 Q% w" |
Recent studies by several groups demonstrate that inactivation of the genesthat encode components of the RAS in mice causes papillary hypoplasia,hydronephrosis, and urinary-concentrating defect( 16, 20, 23 ). These findings imply thatUB-derived epithelia are targets for ANG II actions during renal development. In this work, we tested the hypothesis that ANG II acting via AT 1 receptors stimulates UB-branching morphogenesis in vitro. To do so, wecharacterized the cellular localization of Ao and AT 1 receptorproteins during murine nephrogenesis in vivo. Next we determined that culturedmurine UB cells express AT 1 receptor mRNA and protein. Finally, weinvestigated the effects of ANG II and its AT 1 receptor onbranching of UB cells cultured in three-dimensional collagen gels.
/ Z) ^1 x' S" o( }, ~2 I
METHODS; r% S1 f3 \! W5 {
0 P: E9 d- w6 ]6 B# q1 w: A, B1 B
Immunohistochemistry for AT 1 and Ao+ v& L+ P: u5 P0 f; `5 g
5 C; X1 ~. K  G" K4 v$ b
Immunolocalization of AT 1 receptor protein was performed in C57B/6J mouse embryos from E12 to E16 ( n = 4embryos/age group; Hybrid-Ready Tissue, Novagen). Immunostaining was performed via the immunoperoxidase technique using the Vectastain Elite kit (VectorLaboratories, Burlingame, CA). Sections were deparaffinized in xylene,hydrated in 95-70% alcohol, and washed in PBS (pH 7.2) for 5 min.Endogenous peroxidase activity was quenched by incubation of the sections with0.3% H 2 O 2 in methanol for 30 min. Sections were firstwashed in PBS for 20 min and incubated with the blocking antibody (normalrabbit serum) for 20 min and then incubated with the primary antibody diluted in PBS that contained 1% BSA for 90 min. Sections were rinsed in PBS for 10min and incubated with IgG-biotinylated antibody (anti-rabbit) for 30 min andwere then washed in biotinylated horseradish peroxidase complex before beingexposed to 0.1% diaminobenzidine tetrahydrochloride and 0.2%H 2 O 2 for 3-5 min. Subsequently, the slides werewashed in tap water and counterstained with hematoxylin. The primaryantibodies used were 1 ) a polyclonal anti-rat Ao antibody raised inthe sheep (1/5,000-1/8,000 concentration; generously provided by Dr.Conrad Sernia; Ref. 4 ) and 2 ) a polyclonal rabbit AT 1 receptor antibody directedagainst the NH 2 -terminal domain of the human receptor (identicalsequence in the mouse; Santa Cruz, N-10, sc-1173) at concentrations of1/100-1/400. AT 1 receptor antibody detects bothAT 1A and AT 1B receptor subtypes. The specificity ofimmunostaining was assessed by preadsorbtion of the primary antibody with itsimmunogenic peptide and by the omission of the primary antibody.+ [7 W; I0 u! t3 u. H9 _  e7 G7 S) D

Western Blot Analysis- G0 z8 ]* h! n
6 R0 c! b; ^  z4 [9 r9 Y
Mouse kidneys and UB cells (generously provided by Dr. Jonathan Barasch,Columbia University) were homogenized in cold lysis buffer (50 mM Tris· HCl, pH 8.0, 150 mM NaCl, 0.02% sodium azide, 1% Nonidet P-40, and0.5% deoxycholate) that contained a cocktail of enzyme inhibitors (in mg/ml:0.1 PMSF, 0.001 aprotinin, 0.001 leupeptin, and 0.01Na 3 VO 4 ). The samples were centrifuged for 10 min at14,000 g at 4°C and the supernatants that contained the proteins(0.02 mg/lane) were resolved on 10% Tris-glycine gels and transferred tonitrocellulose membranes. The adequacy of transfer was assessed by Ponceau Sstaining of the membranes. Nonspecific binding was blocked by incubation ofthe membranes with PBS that contained 0.2% Tween and 3% BSA overnight at4°C. The membranes were incubated with the AT 1 receptorantibody (1/500 concentration) at room temperature for 1 h. After three washesin PBS-Tween, the nitrocellulose membrane was exposed for 1 h at roomtemperature to the secondary antibody. Immunoreactive bands were visualizedusing the enhanced chemiluminescence detection system (ECL, Amersham). Theblots were then exposed to Hyperfilm-ECL films./ a5 p6 T  N& J; n7 p

RT-PCR
5 |8 W1 i+ a6 P( A, \, k9 Q* x
RT-PCR was used to determine whether UB cells express ANG II AT 1 receptor mRNA. Total kidney and UB cell RNA was extracted using TRIzol reagent(Invitrogen). RNA (0.003 mg) was reverse-transcribed in the presence of 100 ngof random hexamers, 0.001 ml of 10 mM 2-deoxynucleotide 5'-triphosphate(dNTP), 0.002 ml of 10 x RT buffer [in mM: 200 Tris · HCl (pH8.4), 500 KCl, and 15 MgCl 2 ], and 200 U of SuperScript II reversetranscriptase (Invitrogen) as previously described( 42 ). After RNase treatment, cDNA was amplified using the Perkin Elmer Gene Amp PCR System 2400 (CetusInstruments, Norwalk, CT) from 25% of RT mixture using 20 pmol each ofgene-specific primers, 1 U of Taq DNA polymerase, 0.005 ml of10 x PCR buffer, and 0.001 ml of 10 mM dNTP. AT 1 -specificprimers were as follows: sense, 5'-GCATCATCTTTGTGGTGGG-3'; antisense, 5'-GAAGAAAAGCACAATCGCC-3'( 3 ). Amplification was performed at 94°C for 1 min, 55°C for 1 min, and 72°C for 2 minfor 30 cycles. The AT 1 primers amplify both AT 1A andAT 1B receptor mRNAs.

In Vitro Branching Morphogenesis Assay
( E9 g4 @5 ^( W6 H8 Z3 v: ?7 X
UB cells were maintained in MEM media that contained 10% FBS at 37°C inan incubator with 5% CO 2. UB cells were initially obtained frommicrodissected UBs of an E11.5 mouse transgenic for simian virus 40(SV40) large T antigen (Immortomouse, Charles River; Ref. 2 ). The ureteral epithelialcharacter of UB cells was previously demonstrated by the expression ofE-cadherin, cytokeratin, zona occludens-1, cret, and lectin( 32 ).% @( s, C$ ~# z

UB cells cultured in gels form processes and branching tubular structureswhen exposed to various growth factors, which provide a convenientexperimental system to analyze mechanisms of epithelial branchingmorphogenesis ( 26, 44 ). Previous studiesdemonstrate that UB cells grown in collagen gels proceed through defined stages: extension of cellular processes and formation of multicellular cords,which are followed by branching tubular structures ( 26, 44 ). A principal advantage ofthis approach is the ability to examine the direct effect of a specific factoron UB growth that is independent from confounding soluble factors released bythe mesenchyme.6 N4 P+ }/ N7 ~7 ^/ a
& O7 J) X, i/ w+ x, s+ r* b# R1 f
In vitro tubulogenesis assay was performed as previously described ( 34 ). The cells were used at passages 15-20. The cells were trypsinized and suspended in an8:1:1 ratio of type I rat-tail collagen (Upstate Biotechnology), 10 x DMEM, and 200 mM HEPES (pH 8.0) at 150 x 10 3 cells/ml.Subsequently, the suspension was dispensed in aliquots into a 96-well culture plate (0.1 ml/well). In pilot experiments, we examined the optimal compositionof collagen matrix for UB cell branching. We found that there was nodifference in the number of branches formed whether we used collagen alone ora 1:1 mix of type I collagen and growth factor-reduced Matrigel for 15,000cells/0.1 ml of gel mix. Therefore, we chose to use collagen alone for the additional experiments. After gelation, 0.1 ml/well of DMEM/F-12 medium with0.5% FBS or with 0.5% FBS and either 1 ) ANG II (10 - 5 to 10 - 7 M;Sigma), 2 ) ANG II and AT 1 receptor antagonist candesartan(10 - 6 M; Sigma), or 3 ) epidermal growthfactor (EGF; 40 ng/ml; Upstate Biotechnology) was added to each well on top ofthe collagen gel. This dose of EGF was previously shown to cause branching ofUB cells in collagen gels( 32 ). The growth medium addedon top of the gel contained DMEM/F-12 medium as previously described( 27 ). We used 0.5% FBS in thegrowth medium in an effort to minimize the confounding effects of growthfactors present in the serum. Cells were cultured in collagen gels at 37°Cwith 5% CO 2 as described by Qiao et al.( 30 ) for up to 8 days. Themedia were changed every other day.

The following steps were performed to ensure even well loading with cells.Before suspension of cells in the collagen mix, the cell number was countedusing a hemocytometer and Trypan blue dye, and the cells were added to theremainder of the mix to yield a final concentration of 150,000 cells/ml oftotal collagen mix. To prevent immediate gelation, the mix containing thecells was kept on ice. The cells were dispersed within the mix by gentlevortexing followed by gentle mixing with repeated pipetting of the cellssuspended in the mix. In addition, equal loading of the wells with cells wasassessed after gelation of the mix. All cells present in a quadrant ofselected wells were counted, and then the number was multiplied by 4. The average number of cells per well was 15,000 ± 750 ( n = 10wells), which indicates a coefficient of variability When branchingwas determined, all cells and clusters in each well were assessed for thepresence of processes/branches. In each well, all cells that had at least onebranch were counted. The number of cells with processes or primary branches[an early event in in vitro tubulogenesis( 34 )] and the number of cellsthat formed secondary branches were counted directly from the plates in eachwell using an Olympus IX70 inverted phase-contrast microscope. Each conditionwas set up in triplicate ( n = 4 separate experiments). Primarybranches were defined as processes originating from the cell body. Secondarybranches were defined as processes originating from primary branches. A highernumber of secondary branches was taken to represent an increased complexity ofbranching. Images were acquired directly from the plates via an OlympusMagnaFire FW camera and were processed with Adobe PhotoShop 7.0 software.

Statistical Analysis9 f' M  A, o* i3 n

Differences among the treatment groups in cell/process number were analyzedby one-way ANOVA. A P value of statisticallysignificant.$ E* y- z: H( z4 H0 C, o8 }
  u/ r- s" K: I4 r
RESULTS

Localization of Ao and AT 1 Receptor Proteins in MouseMetanephros2 e# \" b' S9 F/ }7 w
/ j! u9 b) r% _- i* X
AT 1 receptors. At E12,AT 1 immunoreactivity was observed at low levels in a diffusestippled fashion in the UB and the surrounding MM( Fig. 1 A ). At E14, AT 1 immunoreactivity was highly abundant on bothluminal and basolateral aspects of UB branches( Fig. 1 B ). By E15, AT 1 immunoreactivity was present on both luminal andbasolateral aspects of UBs and their derivatives ( Fig. 1 C ). Controlsections incubated with the omission of the primary antibody showed a markeddecrease in staining ( Fig. 1 D ). AT 1 was also detected in renalmicrovessels (data not shown). At E16, AT 1 immunostainingwas present in proximal tubules ( Fig. 2, A and B ). In addition, AT 1 was weaklyexpressed in UB branches, SM, and glomeruli. Control sections incubated withthe primary antibody preadsorbed with its immunogen showed a marked decreasein staining ( Fig.2 C ).

Fig. 1. Immunolocalization of ANG II type 1 (AT 1 ) receptor protein inthe fetal kidney on embryonic day ( E ) 12, E14, and E15. A : AT 1 is detectable using an antibodyconcentration of 1/100. On E12, AT 1 immunoreactivity ispresent at low levels in a diffuse stippled fashion in the ureteric bud (UB)and metanephrogenic mesenchyme (MM). B : on E14,AT 1 is expressed on both luminal and basolateral aspects of UBbranches. C : AT 1 is detectable using an antibodyconcentration of 1/200. On E15, UB branches are strongly positive forAT 1 on both apical and basolateral membranes. D : controlsection, where the addition of the primary antibody was omitted, demonstratesno staining. G, glomerulus; S, S-shaped glomerulus; CD, collecting duct.

Fig. 2. Immunolocalization of AT 1 receptor protein in the fetal kidneyon E16. AT 1 is detectable using an antibody concentrationof 1/400. A and B : AT 1 is expressed in proximaltubules (PT) but only weakly in UB branches, stromal mesenchyme (SM), andglomeruli. PTs are located in close proximity to the glomeruli and arecharacterized by thick brush-border epithelia. C : control sectionincubated with prior preadsorbtion of the primary antibody with its immunogenshows marked attenuation of staining.

Angiotensinogen. At E14, Ao immunostaining was present inUB branches and surrounding SM ( Fig.3 A ). Control sections incubated without the primaryantibody showed no specific immunostaining signal( Fig. 3 B ). By E15, Ao immunoreactivity was observed in UB branches, proximaltubules, and SM ( Fig. 3, C and D ). A similar staining pattern was observed on E16 (not shown).! g% t' i4 R3 t6 o& U" P

Fig. 3. Immunolocalization of angiotensinogen (Ao) protein in the fetal kidney on E14 and E15. Ao antibody concentration was 1/6,000. A : on E14, Ao immunoreactivity was present in the UB andsurrounding SM. B : control section incubated with omission of primaryantibody shows absence of specific staining. C : on E15, UBbranches and proximal tubules were strongly positive for Ao on both apical andbasolateral membranes. D : high-power view shows Ao immunostaining inampulla of UB branches.
% Z% o8 u2 b$ z1 R
Expression of AT 1 Receptors in UB Cells& }" ]2 {% d8 R( m% D

To test whether cultured UB cells maintain AT 1 receptorexpression, we examined AT 1 receptor gene expression using Westernblotting and RT-PCR. As shown in Fig.3, UB cells expressed abundant AT 1 receptor protein(molecular mass, 41-42 kDa). Control blots in which the primary antibodywas preadsorbed with the immunogenic peptide were negative( Fig. 4, A and B ). The AT 1 mRNA transcripts were also presentin cultured UB cells ( Fig. 4 C )., E6 [8 G2 u  Y% d. r2 q! `

Fig. 4. Western blot analysis and RT-PCR of AT 1 receptor. A :adult mouse kidney (Kid) and UB cells (0.02 mg of total protein/lane). B : elimination of AT 1 receptor-specific bands with priorpreadsorbtion of the primary antibody with its immunogen. C : ethidiumbromide-stained gel shows RT-PCR analysis of AT 1 receptor mRNAexpression. IMCD, inner medullary collecting duct cells.% F1 a) g7 [* a7 O+ n. S5 W' k7 o( v

ANG II Stimulates UB-Branching Morphogenesis In Vitro
$ K6 t" N8 G5 b6 y
To examine the role of ANG II in UB-branching morphogenesis, weinvestigated the effects of ANG II on cell process formation in UB cellscultured in three-dimensional type I collagen gels. The number and complexityof processes were quantitated daily via phase-contrast microscopy. Cellprocess formation and branching morphogenesis depend on multiple factors(including matrix composition, cell density, amount of serum present inculture media, and nature of inducing soluble factor) and are subject tointerassay variability. Under the conditions of our study, the number of cellswith processes or branches varied from 60 to 286/well, and the total number ofprocesses varied from 150 to 615/well. Therefore, the branching phenomenoninvolved 0.4-1.9% of cells originally seeded in a well (15,000). Figure 5 shows phase-contrastphotographs that depict the effects of ANG II and EGF (positive control) oncell process formation and branching of UB cells 24 h after cells were platedinto collagen gel. In control untreated cells, a few short processes wereobserved originating from the cell body( Fig. 5 A ). ANG IIcaused more extensive changes in cell process formation than control( Fig. 5, B-D ). Athigher doses such as 10 - 5 M, ANG II was capable ofcausing clearly evident secondary branching as well( Fig. 5 D ). In thepresence of candesartan, cells exposed to ANG II formed shorter and fewerprocesses as compared to cells exposed to ANG II alone( Fig. 5 E ). EGF-treated cells, on the other hand, showed long, tubule-like formations with secondaryand tertiary branches ( Fig.5 F ).1 M  x# O# [9 s% E4 G

Fig. 5. UB cells grown in three-dimensional collagen gels 24 h after plating. A : control (media 0.5% FBS). B, C, D : ANG II. E :ANG II with candesartan (Cand). F : epidermal growth factor (EGF; 40ng/ml). 1°, 2°, And 3°: primary, secondary, and tertiary branches,respectively.
9 j3 ?6 q% x" f, A
To further delineate the nature of the branching structures, we visualizedthe UB cell cultures using immunofluorescence microscopy after double stainingfor Dolichos biflorus lectin and propidium iodide. Figure 6 shows an example ofsuch an experiment: UB cells form multicellular branching cords when exposedto ANG II.
* T6 z* g6 X" |6 n
Fig. 6. Micrographs of UB cells grown for 4 days in collagen gel in the presence of10 - 5 M ANG II. Cells form multicellular branchedstructures. A : phase-contrast image. B : fluorescence image.Cells were stained with Dolichos biflorus lectin (green) andpropidium iodide (red).

Quantitative analysis revealed that ANG II tended to increase cellbranching at 10 - 7 and10 - 6 M concentrations, but statisticalsignificance was not reached until a higher concentration of10 - 5 M was used( Fig. 7 ). At a10 - 5 M concentration, ANG II increased the numberand percent of UB cells with processes per well compared with control cells cultured in the presence of DMEM/F-12 media with 0.5% FBS alone on day1 (150 ± 13 vs. 100 ± 11%; P day2 (162 ± 16 vs. 100 ± 9%; P day 4 (226 ± 24 vs. 100 ± 12%; P plating ( Fig. 7, A and B ). ANG II-induced cell process formation was abolishedby pretreatment with 10 - 6 M candesartan, theAT 1 antagonist [ day 1, 150 ± 13 vs. 50 ± 10%( P day 2, 162 ± 16 vs. 61 ± 13%( P Fig.8 A ]. In addition, 10 - 5 M ANG II increased the number of cells with secondary branches on day 2 (212 ± 42 vs. 100 ± 22%; P which indicatesthat ANG II stimulates formation of branching structures with a higher degreeof complexity ( Fig.8 B ). The absolute numbers of cells with secondarybranches were as follows: day 1, 23 ± 8.0 vs. 8 ± 2.0( P = 0.17); day 2, 22 ± 3.5 vs. 8 ± 1.9( P day 4, 25 ± 7.0 vs. 11 ± 3.1( P = 0.11; 10 - 5 M ANG II vs. control,respectively). Similarly, the number of cells with secondary branches was alsodecreased in the presence of candesartan: day 1, 205 ± 72 vs.21 ± 4.1% ( P day 2, 212 ± 42 vs.27 ± 3.2% ( P Fig. 8 B. On day4, the difference did not reach statistical significance at 250 ±70 vs. 100 ± 23% ( P = 0.11).

Fig. 7. Effects of ANG II in concentrations of 10 - 5,10 - 6, and 10 - 7 M oncell process formation in UB cells. A : values are presented asabsolute number of cells with processes. B : percent relative tocontrol with control equaling 100%. * P6 L) f9 {& d3 B' j( N" f
) k1 I7 B, W$ {) ?" _0 f7 W
Fig. 8. Effects of 10 - 5 M ANG II and10 - 5 M ANG II 10 - 6 Mcandesartan on cell process formation in UB cells. Values are presented aspercent relative to control with control equaling 100%. A : cells withall processes (primary and secondary). B : cells with secondaryprocesses. Numbers represent respective P values. * P # P* l8 |! z6 r+ Y% O
2 i" u/ \' `& R4 n! `% C+ p7 p
Studies by Zent et al. ( 44 )demonstrate that branching morphogenesis in UB cells cultured in collagen gelsin the presence of 10% FBS can be maintained for up to 12 days. We used a muchlower concentration of FBS (0.5%) to minimize confounding influences ofmultiple growth factors present in the serum on UB cell branching. Cells weremonitored for process formation for up to 8 days. The absolute number of cellswith processes continued to increase up to day 4 but decreasedsubstantially thereafter: day 1, 118 ± 11 vs. 77 ± 9( P day 2, 131 ± 13 vs. 80 ± 7( P day 4, 113 ± 12 vs. 50 ± 6( P day 6, 63 ± 9 vs. 49 ± 8( P = 0.22); day 8, 67 ± 12 vs. 50 ± 6( P = 0.28; 10 - 5 M ANG II vs. control, respectively). Although cell viability was not directly assessed, the declinein the number of cells with processes on days 6 and 8 waslikely secondary to cell death.0 F( P; |  _) M1 F: R& i, G

DISCUSSION

The present study demonstrates that the UB branches and stroma expressAT 1 and Ao and that ANG II stimulates branching of UB cells invitro via activation of the AT 1 receptor. To our knowledge, this isthe first report to show a direct role for ANG II in UB cell branching.
2 A( L$ E  W+ r0 T
The kidney is formed by reciprocal inductive interactions between the UBand MM ( 8 ). The MM induces UBto elongate and branch repeatedly to form the renal collecting system. Inturn, the UB induces the MM to differentiate into the glomeruli and proximal and distal tubules ( 1, 5 ). The BF2/FoxD1-positive SMalso plays a key role in renal morphogenesis through release of soluble factors (e.g., retinoic acid, fibroblast growth factor-7, and Wnt2b) thatregulate epithelial branching morphogenesis( 9, 14, 16, 31 ). The role of the RAS in UBand collecting-system development has not been defined. However, such a roleis likely because 1 ) the developing kidney expresses all thecomponents of the RAS ( 7, 22, 41 ), and ANG II synthesis andAT 1 /AT 2 receptor expression are enhanced duringmetanephrogenesis ( 7, 12, 43 ); and 2 ) strongevidence derived from genetic studies has shown that inactivation of the Ao,angiotensin-converting enzyme (ACE), or AT 1 genes causesabnormalities in the development of the renal medulla( 6, 20, 23 ). Ao-, ACE-, orAT 1 -mutant mice exhibit thinning of the medulla, atrophy of thepapilla, and dilation of the pelvis. These changes are most pronounced postnatally after the formation of the renal papilla. In addition, ACE- andAT 1 -null animals also have a reduced ability to concentrate urine( 16, 23 ). One possible mechanismmay be due to decreased number or altered contractile function of smoothmuscle cells along the ureter that lead to defective peristalsis as has been shown in AT 1 -mutant mice( 17 ).* M- m7 l; ^' z$ F
$ X9 U& m8 f0 s1 Z8 W9 W
The role of AT 2 in UB branching is unknown. Most recent studies indicate that AT 2 counteracts the proliferative actions of AT 1 ( 38 ). Ourpreliminary data demonstrate that AT 2 is expressed in UB branchesduring mouse nephrogenesis( 11 ). AT 2 -mutant mice exhibit ectopic ureteral budding and duplicated collecting systems( 24 ). This suggests thatAT 2 inhibits aberrant ureteral budding. It is conceivable thatunopposed stimulation of AT 2 in AT 1 -mutant mice mayhinder UB branching. Therefore, the ultimate effect of ANG II on UB branchingmay depend on the balance between AT 1 - and AT 2 -mediatedactions.% X! x* I$ ]  ~5 I6 @; A: l6 [: Z
# r6 b- ]4 Z1 N) J/ T
In addition, ANG II may influence papillary development by regulating theexpression of renal growth factors within or adjacent to papilla. In thisregard, downregulation of platelet-derived growth factor mRNA levels in therenal papilla of Ao-null mice ( 21 ) and reduced EGF mRNAlevels in AT 1 -null mice( 19 ) suggest thatdysregulation of ANG II-regulated growth factors contributes to abnormalcollecting-system formation in these mice. Delineating the effects of theAT 1 -null mutation on expression of stromal factors necessary for UBbranching (BF2, FGF7, Wnt2b, and RAR) will be crucial to our understanding ofthe mechanism governing the development of the renal medulla.( g1 z" D/ W1 n: _% b) Q

With regard to the role of ANG II in UB branching per se, studies by Nagataet al. ( 20 ) have shown that E12 metanephric kidneys from Ao-null mice grown ex vivo showedfavorable branching of UBs that was unaffected by the addition of ANG II. Thedata were interpreted to suggest that ANG II is not essential for the growthof UBs. However, the absence of ANG II in this model could have beencompensated for by other factors derived from the mesenchyme. Therefore,utilization of the UB cell culture model may be more relevant to define thedirect role of ANG II in UB-branching morphogenesis. In ongoing work, we areattempting to establish an assay to culture intact UBs isolated from E13 mouse embryos( 30 ). This assay may representa more physiological system to study the effects of growth factors includingANG II on epithelial branching morphogenesis./ A1 E/ d1 K* Z3 Z3 |4 |) ~

An important finding of this study is that Ao and AT 1 areexpressed in both UBs and SM. A recent study( 15 ) found that renin is present in the SM at a time when UB branching is just beginning. This raisesthe intriguing possibility that ANG II can be generated locally in the SM andthen act in a paracrine fashion on the adjacent AT 1 -expressing UBsto induce branching. Similarly, AT 1 present in the SM may beimportant in mediating stromal ANG II signaling to stimulate UB branching. Onepossible pathway may involve ANG II-induced stimulation of FGF-7 and itscoupling with FGF-7 receptors expressed on the UB( 29, 31 ). This possibility issupported by the ability of ANG II to increase basic FGF mRNA levels in lutealcells ( 37 ).

Several growth factors have been shown to either stimulate or inhibitUB-branching morphogenesis. The stimulatory factors include EGF and hepatocytegrowth factor ( 32 ), Gdnf/Ret( 40 ), vitamin A/retinoic acidreceptor ( 16 ), low-dose bonemorphogenetic protein (BMP)-7( 25 ), pleiotrophin( 33 ), and ANG II. Because ANGII induced process formation in utilization ofisolated intact UBs will allow us to better ascertain the relative importanceof ANG II in UB-branching morphogenesis. The inhibitory factors includeendostatin ( 13 ), high-doseBMP-7 ( 25 ), BMP-2( 26 ), BMP-4( 18 ), and transforming growthfactor- ( 34 ). Given themultitude of the signal transduction pathways used by these factors, it isdifficult to pinpoint a specific common final pathway that governs UBdevelopment. It is more likely that the final common pathway involves cross talk between multiple pathways such as the phosphatidylinositol 3-kinase(PI3-K) and mitogen-activated protein kinase (MAPK) pathways, which are knownto commonly be activated by the above factors.
) {6 K( B- _$ E& X7 z6 n8 P
Our present finding that candesartan abrogates ANG II-induced UB cellbranching indicates that the AT 1 receptor exerts growth-promoting effects on UB cell branching. The signaling events that link AT 1 togrowth and differentiation of UB cells have not been defined. Severalsignaling pathways including Ras/Raf/MEK/ERK/activator protein-1 (AP-1) andPI3-K/Akt have been shown to mediate the effects of AT 1 on cellproliferation and hypertrophy in vascular smooth muscle cells, fibroblasts,and renal mesangial cells ( 10, 35, 36 ). Thus one of the possiblemechanisms to lead to increased UB cell proliferation, survival, andmorphogenesis may involve AT 1 -mediated stimulation of theRas/Raf/MEK/ERK/AP-1 and PI3-K/Akt pathways.

In summary, the present study demonstrates that the AT 1 receptor is expressed in the UB and its derivatives during early metanephrogenesis invivo. In addition, UB-derived cells in culture maintain expression ofAT 1 receptor mRNA and protein. ANG II, acting via AT 1 receptors, stimulates cell process formation and branching in UB cells grownin collagen gels. These data strongly support a direct and independent rolefor the RAS in the development of the renal collecting system.
8 J" h9 T" K' U0 w! i" b1 x0 [- E
DISCLOSURES
, T" N+ l" U  D, R* y. d2 l
These studies are supported by funds from the Department of Pediatrics,Tulane University Health Science Center and by National Institutes of HealthCOBRE Grant P-20 RR017659.8 m/ k& M2 |1 q" j& T, G/ |
: D# i! }1 L: g* |  L$ A% ~  b
Portions of this study were presented at the Southern Society of PediatricResearch Meeting, New Orleans, LA, February 2002 and were published inabstract form ( J Invest Med 50: 134A, 2002) and at the AmericanSociety of Nephrology Meeting, Philadelphia, PA, November 2002 ( J Am SocNephrol 13: 302A, 2002).

ACKNOWLEDGMENTS# B% {, {3 @+ p& l3 E5 Z& o$ ~
9 p  y" N& S: a5 Z: t. `3 v( P
The authors thank Dr. Samir El-Dahr for insightful discussions and criticalreading of the manuscript.5 S3 v+ {4 I) A, M; S
      【参考文献】8 s/ I' O$ v3 l
Aufderheide E,Chiquet-Ehrismann R, and Ekblom P. Epithelial-mesenchymal interactions inthe developing kidney lead to expression of tenascin in the mesenchyme. J Cell Biol 105:599-608, 1987.
5 E! T0 L' X/ z/ Z+ K
3 W% q" j0 i9 l# D/ J  `

Barasch J,Pressler L, Connor J, and Malik A. A ureteric bud cell line inducesnephrogenesis in two steps by distinct signals. Am J Physiol RenalFluid Electrolyte Physiol 271:F50-F61, 1996.
3 P$ N1 p/ j8 f% _4 Z5 l' b( R9 r. o
+ j) [; B* Q/ h( P+ k
$ [% {9 _' j! v! P
Burson JM,Aguilera G, Gross KW, and Sigmund CD. Differential expression ofangiotensin receptors 1A and 1B in mouse. Am J Physiol EndocrinolMetab 267:E260-E267, 1994.

6 i6 V+ f" V1 V/ y; f  E' I
! `# e0 \, P; M, q  r- f4 w
Darby IA andSernia C. In situ hybridization and immunohistochemistry of renalangiotensinogen in neonatal and adult rat kidneys. Cell TissueRes 281:197-206, 1995.
: V( h: E7 R( r/ _# A  `$ b
! O% I: B5 J5 Y$ E
& ?! u% y( ~  S6 U  r9 B
Ekblom P. Developmentally regulated conversion of mesenchyme to epithelium. FASEB J 3:2141-2150, 1989.0 q; i% E# M( v6 U2 d% M: r! y
1 F2 Y8 W# F) }8 u; |! T& R

2 X( B; t$ P* O; Y* k3 }
Esther CR,Howard TE, Marino EM, Goddard JM, Capecchi MR, and Bernstein KE. Micelacking angiotensin-converting enzyme have low blood pressure, renalpathology, and reduced male fertility. Lab Invest 74: 953-965,1996.3 S& i$ [0 L& f3 ?; s# j
; p$ [& f  $ L7 m0 [& w; l, n
6 z2 W' S. a; G% l

Gomez RA, LynchKR, Sturgill BC, Elwood JP, Chevalier RL, Carey RM, and Peach MJ. Distribution of renin mRNA and its protein in the developing kidney. Am J Physiol Renal Fluid Electrolyte Physiol 257: F850-F858,1989.4 H6 a/ ?: V6 |0 {/ K. V- T
/ `" e4 _" ?, ?! g. n
  D& r% $ C5 k6 a0 B- ~1 L% e
  J$ G8 W; c$ J) D7 y' t
Grobstein C. Inductive epithelio-mesenchymal interaction incultured organ rudiments of the mouse metanephros. Science 118:52-55, 1953.
/ `6 z7 o; Q  ~) C* o: F, e5 x/ I

, S7 o' x" e. U+ E3 r9 }
Hatini V, HuhSO, Herzlinger D, Soares VC, and Lai E. Essential role of stromalmesenchyme in kidney morphogenesis revealed by targeted disruption of wingedhelix transcription factor BF-2. Genes Dev 10: 1467-1478,1996.
" [# p0 x( @3 P. P7 E# N8 \
1 x) z$ D8 s/ s0 \! V
- i2 F' }5 b" v8 P
Huwiler A,Stabel S, Fabbro D, and Pfeilschifter J. Platelet-derived growth factorand angiotensin II stimulate the mitogen-activated protein kinase cascade inrenal mesangial cells: comparison of hypertrophic and hyperplastic agonists. Biochem J 305:777-784, 1995.
( G) l/ \9 S1 ], `, e6 w% k

& c4 X) t* W( s5 {# |! H8 `
Iosipiv IV. Cellular expression of the angiotensin type 2 receptor (AT 2 ) duringmurine organogenesis (Abstract). J Investig Med 50: 134A,2002.3 A4 j; ~5 p( E: w
9 Q! A3 J% t5 V/ C3 v+ L
9 Z* u3 m6 [5 Z5 T2 p" z

Kakuchi J,Ichiki T, Kiyama S, Hogan BLM, Fogo A, Inagami T, and Ichikawa I. Developmental expression of renal angiotensin II receptor genes in the mouse. Kidney Int 47:140-147, 1995.
; j7 e( }& |- T8 Y0 J1 ~9 ~

9 B, J1 V/ y3 R* X
Karihaloo A,Karumanchi SA, Barasch J, Jha V, Nickel CH, Yang J, Grisaru S, Bush KT, NigamS, Rosenblum ND, Suchatme VP, and Cantley LG. Endostatin regulatesbranching morphogenesis of renal epithelial cells and ureteric bud. Proc Natl Acad Sci USA 98:12509-12514, 2001.

) @' V7 W1 d1 v
0 I, B& J: v( i
Lin Y, Liu A,Zhang S, Ruusunen T, Kreidberg JA, Peltoketo H, Drummond I, and Vainio S. Induction of ureter branching as a response to Wnt-2b signaling during earlykidney organogenesis. Dev Dyn 222: 26-39,2001.

6 _  y5 y7 [0 D! w, X9 y
0 P4 o1 }% ^8 Y
Lopez ML, PentzES, Robert B, Abrahamson DR, and Gomez RA. Embryonic origin and lineage ofjuxtaglomerular cells. Am J Physiol Renal Physiol 281: F345-F356,2001.) Y8 q* ^5 S' E! O, L( [

1 R+ U0 l) y6 t  f" q+ G! @

Mendelsohn K,Batourina E, Fung S, Gilbert T, and Dodd J. Stromal cells mediateretinoid-dependent functions essential for renal development. Development 126:1139-1148, 1999.

) c* c3 E: o" H! r; n
0 s/ T9 U3 k* f
Miyazaki Y andIchikawa I. Nephrotoxicity of angiotensin inhibition during the perinatalperiod. Exp Nephrol 9:10-13, 2001.- q/ X: l9 O1 ^. j4 d, t8 e

4 P! L+ [- J. o4 P* v! J1 e$ j
Miyazaki Y,Oshima K, Fogo A, Hogan BLM, and Ichikawa I. Bone morphogenic protein 4regulates the budding site and elongation of the mouse ureter. JClin Invest 105:863-873, 2000.' r6 [. Q1 N( Q1 @% W

8 z  B2 I9 q/ ^7 Y

Miyazaki Y,Tsuchida S, Fogo A, and Ichikawa I. The renal lesions that develop inneonatal mice during angiotensin inhibition mimic obstructive nephropathy. Kidney Int 55:1683-1695, 1999.2 b* N$ D$ m3 w2 v" ?
2 `& u. p% f8 h: R5 f) ^) g6 v

Nagata M,Tanimoto K, Fukamizu A, Kon Y, Sugiyama F, Yagami K, Murakami K, and WatanabeT. Nephrogenesis and renovascular development in angiotensinogen-deficientmice. Lab Invest 75:745-753, 1996.5 T9 K5 q9 Q/ Y8 ]! ]9 k; F

) L" c; b& f! A- {

Niimura F,Labosky PA, Kakuchi J, Okubo S, Yoshida H, Oikawa T, Ichiki T, Naftilan AJ,Fogo A, and Inagami T. Gene targeting in mice reveals a requirement forangiotensin in the development and maintenance of kidney morphology and growthfactor regulation. J Clin Invest 96: 2947-2954,1995.

% g0 T/ G, ~+ Q9 w% Q! [
. B+ P* u7 J  v* B. V
Norwood VF,Craig MR, Harris JM, and Gomez RA. Differential expression of angiotensinII receptors during early renal morphogenesis. Am J Physiol RegulIntegr Comp Physiol 272:R662-R668, 1997.
2 d* X, X/ |8 v; M* u
% g5 ?4 e* B( x# g) E
- O3 b; l9 Q0 F
Oliverio MI,Kim HS, Ito M, Le T, Audoly L, Best CF, Hiller S, Kluckman K, Maeda N,Smithies O, and Coffman TM. Reduced growth, abnormal kidney structure, andtype 2 (AT 2 ) angiotensin receptor-mediated blood pressureregulation in mice lacking both AT 1A and AT 1B receptorsfor angiotensin II. Proc Natl Acad Sci USA 95: 15496-15501,1998.

/ T3 v1 A, b8 B& @- Y

Oshima K,Miyazaki Y, Brock JW, Adams MC, Ichikawa I, and Pope JC. Angiotensin typeII receptor expression and ureteral budding. J Urol 166: 1848-1852,2001.: X  C& ?+ s6 T) V1 L  t, s: p
( O; m8 f  G3 n) g) G

5 x" o. q+ t1 Q3 |! M3 d
Piscione TD,Phan T, and Rosenblum ND. BMP7 controls collecting tubule cellproliferation and apoptosis via Smad1-dependent and -independent pathways. Am J Physiol Renal Physiol 280:F19-F33, 2001.# B. G4 c/ q1 i0 m$ }/ m
7 I8 c/ o9 {" \. f! o9 L2 e1 h
& ^8 W* p0 j& t

Piscione TD,Yager TD, Gupta IR, Grinfeld B, Pei Y, Atti-sano L, Wrana JL, and RosenblumND. BMP-2 and OP-1 exert direct and opposite effects on renal branchingmorphogenesis. Am J Physiol 42:F961-F975, 1997.
: r$ P4 {4 U. u

2 P% g0 S5 p7 A
Pohl M, SakuraiH, Stuart RO, and Nigam SK. Role of hyaluronan and CD44 in in vitrobranching morphogenesis of ureteric bud cells. DevBiol 224:312-325, 2000./ D! Y+ O. I) O, g- s/ `

* W# L$ B* p5 ~( m0 B. F4 `
Prieto M, DippS, Meleg-Smith S, and El-Dahr SS. Ureteric bud derivatives expressangiotensinogen and AT 1 receptors. PhysiolGenomics 6:29-37, 2001.
# M1 D5 a! a: Z7 k' `, Y$ o

Qiao J, BushKT, Steer DL, Stuart RO, Sakurai H, Wachsman W, and Nigam SK. Multiplefibroblast growth factors support growth of the ureteric bud but havedifferent effects on branching morphogenesis. Mech Dev 109: 123-135,2001.

Qiao J, SakuraiH, and Nigam SK. Branching morphogenesis independent ofmesenchymal-epithelial contact in the developing kidney. Proc NatlAcad Sci USA 96:7330-7335, 1999.# D8 V. E+ A. e8 O* k% _/ p
( {. E0 {+ E8 ^. W8 e
9 {( ~1 }, V4 o; T5 i. Y$ q. _6 u
" U4 G  D. Y6 Z+ N, U& o# W
Qiao J, Uzzo R,Obara-Ishihara T, Degenstein L, Fuchs E, and Herzlinger D. FGF-7 modulatesureteric bud growth and nephron number in the developing kidney. Development 126:547-554, 1999.
; ^0 `; ~$ j; M6 Q
$ {$ I0 |) A# v% K2 M  |* q# D) S. E
4 Q: D" f# e/ o) q8 }, Z
Sakurai H,Barros EJ, Tsukamoto T, Barasch J, and Nigam SK. An in vitro tubulogenesissystem using cell lines derived from the embryonic kidney shows dependence onmultiple soluble growth factors. Proc Natl Acad SciUSA 94:6279-6284, 1997.7 C0 ^1 `) V9 d
" `& y* a9 u; Z" p  n- M
5 m  z: P( ^; e9 H) T6 Z
2 L9 U. s7 T5 \4 T1 ~8 W6 h
Sakurai H, BushKT, and Nigam SK. Identification of pleiotrophin as a mesenchymal factorinvolved in ureteric bud branching morphogenesis. Development 128:3283-3293, 2001.$ y( \8 N- _: s% d" G6 ]
3 ]% W! M7 ~: f' T

. A7 l1 [4 q, F% ~$ G) ?1 p+ ?
Sakurai H andNigam SK. Transforming growth factor-beta selectively inhibits branchingmorphogenesis but not tubulogenesis. Am J Physiol 41: F139-F146,1997.% {: ?" O% [$ F& d) U; h' a5 ~

Saward L andZahradka P. Angiotensin II activates phosphatidylinositol 3-kinase invascular smooth muscle cells. Circ Res 81: 249-257,1997.1 n8 A% M) Z' t" j% b9 w/ ]" }* a8 [. @7 H

3 o: ^  U* B1 ]4 K

Schorb W,Peeler TC, Madigan NN, Conrad KM, and Baker KM. Angiotensin II-inducedprotein tyrosine phosphorylation in neonatal rat cardiac fibroblasts. J Biol Chem 269:19626-19632, 1994., I* v+ b; ~2 V% j- b! u

- r. h1 [& o0 _" ?3 s9 _( V
Stirling D,Magness RR, Stone R, Waterman MR, and Simpson ER. Angiotensin II inhibitsluteinizing hormone-stimulated cholesterol side chain cleavage expression andstimulates basic fibroblast growth factor expression in bovine luteal cells inprimary culture. J Biol Chem 265: 5-8,1990.
. N. p" g) Y0 W+ j; U* U5 G: R/ r
/ `5 v' x- l1 D- P5 P# |9 c3 L

Stoll M,Steckelings UM, Paul M, Bottari SP, and Unger T. The angiotensinAT 2 -receptor mediates inhibition of cell proliferation in coronaryendothelial cells. J Clin Invest 95: 651-657,1995.

( O5 |; T2 h7 g. R1 I$ [
5 D$ l8 M3 X1 R* m9 \0 W- G
Tufro-McReddie A, Romano LM, Harris JM, Ferder L, and Gomez RA. Angiotensin II regulates nephrogenesis and renal vascular development. Am J Physiol 38:F110-F115, 1995.

! J& B4 h4 V% K; R$ s/ T! z7 j
Vega QC, WorbyCA, Lechner MS, Dixon JE, and Dressler GR. Glial cell line-derivedneurotrophic factor activates the receptor tyrosine kinase RET and promoteskidney morphogenesis. Proc Natl Acad Sci USA 93: 10657-10661,1996./ u/ ]4 S3 S2 L7 q. F2 m
+ D/ A( E: U+ {8 S# H* ?

& ]& F' q; Y* o" s
Yosipiv IV,Dipp S, and El-Dahr SS. Ontogeny of somatic angiotensin-converting enzyme. Hypertension 23:369-374, 1994.
& X/ K- \, e" H, ]- e  g% W4 `

Yosipiv IV,Dipp S, and El-Dahr SS. Role of bradykinin B2 receptors in neonatal kidneygrowth. J Am Soc Nephrol 8:920-928, 1997.$ b: J( j! [; Q4 U1 `" @) @

3 T% o/ l9 `3 Z5 H
. N) E2 r2 g  W& b0 q
Yosipiv IV andEl-Dahr SS. Activation of angiotensin-generating systems in the developingkidney. Hypertension 27:281-286, 1996.

5 D4 f+ z. t2 c' c3 x$ B
Zent R, BushKT, Pohl ML, Quaranta V, Koshikawa N, Wang Z, Kreidberg JA, Sakurai H, StuartRO, and Nigam SK. Involvement of laminin binding integrins and laminin-5in branching morphogenesis of the ureteric bud during kidney development. Dev Biol 238:289-302, 2001.

收藏0 分享0 顶0 踩0