标题: The renal retinoid system: time-dependent activation in experimental glomerulone [打印本页] 作者: 轻羽 时间: 2009-4-22 08:16 标题: The renal retinoid system: time-dependent activation in experimental glomerulone
作者：Sabine Liebler, Birgit Überschär, Helen Kübert, Stefanie Brems, Anna Schnitger, Miki Tsukada, Christos C. Zouboulis, Eberhard Ritz, and Jürgen Wagner作者单位：1 Department of Nephrology, University of Heidelberg, 69115 Heidelberg; and 2 Department of Dermatology, University Medical Center Benjamin Franklin, The Free University of Berlin, 14195 Berlin, Germany 5 T' l! m2 p1 d- n( H : a9 x! z# r8 Q9 j" e: h4 n+ |7 ] 0 h5 c& }8 H K7 }0 X1 w 6 J8 Z) Q5 b. O* p3 X" r; L# v
3 ~$ L6 H7 O% P$ Q
1 T, q; j+ y3 m7 d7 S$ g) z% P
3 a- E4 }; k+ B3 E$ O7 t! f- y! x
- ~3 K- S' d3 h% l7 S- Z
; i( a: \0 {( y0 | ?: ?" _ % w) O% V+ q, G: s 1 f# ]: W+ N8 U9 |% U4 X
; c( d/ O3 w: X% N9 W5 ~) d# b y/ g6 W# G' X* W 【摘要】 8 Q1 e$ C& ^. _6 R8 J& j. M; `4 Z Retinoids reduce renal damage in rat experimental glomerulonephritis. It is unknown, however, how local and systemic retinoid pathways respond to renal injury. We used a rat model of artificially induced acute anti-Thy1.1-nephritis (THY-GN). We examined the extrarenal and glomerular expression of the retinol (RoDH) and retinal (RalDH) dehydrogenases 1 and 2 as well as the expression of the retinoic acid (RAR) and retinoid X (RXR) receptor subtypes,, and. Furthermore, we investigated serum and glomerular retinoid concentration patterns. On days 3, 7, and 14, we compared nonnephritic rats (control group; CON) to THY-GN rats with respect to systolic blood pressure and glomerular cell count per cross section. Systolic blood pressure and glomerular cell count were significantly higher in THY-GN rats on days 7 and 14 ( P < 0.001). We found a 60% reduction in expression levels for retinoid receptors and dehydrogenases in nephritic glomeruli on day 3, but a threefold increase on day 7 ( P < 0.001 vs. CON). The same applies to RAR protein. Hepatic expression of retinoid receptors was not influenced. On day 14, glomerular expression levels for retinoid receptors and retinoid-metabolizing enzymes had returned to a normal level, glomerular cell count being still increased. Administering 13- cis retinoic acid (isotretinoin) lowered blood pressure and glomerular cell count in nephritic rats but failed to influence the glomerular expression of retinoid receptors or retinoid-metabolizing enzymes. Our data document a stimulation of glomerular retinoid-synthesizing enzymes and expression of retinoid receptors in the early repair phase of THY-GN, suggesting activation of this system in acute renal disease. , f/ h$ s! @% V7 E0 j3 q3 [2 H4 K: k5 U7 {5 n' ~9 k- H
+ u' I4 _6 H2 n& g; }% F3 e Y1 n " w) q9 x. e+ B! Wretinoid receptors; retinol dehydrogenase; retinal dehydrogenase; isotretinoin; acute anti-Thy1.1-glomerulonephritis; rat , ~$ n8 f/ o% k) z( M2 H, A4 I 【关键词】 retinoid timedependent activation experimental glomerulonephritis- J: i. q6 \: U& d
RETINOIDS, DERIVATIVES OF vitamin A, play an important role in processes such as immune response, cell proliferation, and modulation of inflammation ( 4, 9 ). They are primarily stored in the liver in the form of retinyl esters, releasing retinol when hydrolyzed. Retinol is reversibly transformed into retinal by retinol dehydrogenases (RoDH) in a first oxidation step. The irreversible (second) oxidation of retinal by retinal dehydrogenases (RalDH) then yields all- trans retinoic acid (all- trans RA). Cellular RA-binding proteins (CRABP) determine the fate of retinoic acids (RA) by binding and directing them to their respective metabolic pathways ( 10 ). The retinoic acid (RAR) and retinoid X (RXR) receptor subtypes finally communicate the RA information and evoke their biological effects. These receptors belong to the steroid receptor superfamily ( 5 ) and are expressed, among other tissues, in the rat and human kidney ( 18, 38 ). They may influence gene transcription either directly by means of retinoid-responsive elements acting on responsive genes or indirectly by the modulation of transcription factors, among them activator-protein 1 (AP-1) ( 47 ) or NF- B ( 23 ). These mechanisms have been proposed to form the basis of the anti-inflammatory and antiproliferative retinoid effects. * i p, C8 W" L4 T+ Q) R 9 @2 L- g0 @- y! R, u: {We examined the glomerular expression of RAR,,, and RXR subtypes, the expression of the enzymes RoDH 1 and 2 as well as RalDH 1 and 2. RoDH 1 and 2 are mainly expressed in the rat kidney and lung but also in the liver, testis, and brain ( 3 ). RalDH 2 is involved in the differentiation of renal epithelium; it is highly expressed in the embryonic kidney. Additionally, it is a regulator of local RA concentrations ( 39 ).! q8 V; d7 I, [- ^
! `2 L; o: K6 U) h0 V% b
RA possesses autoregulatory potential by inducing or repressing key enzymes of its own metabolism. Excess RA activates the cytochrome P -450 system, which transforms RA into inactive metabolites ( 31 ). Alternatively, RA may enhance the storage pathway by activating the enzyme lecithin retinol acyltransferase, resulting in the formation of retinyl esters ( 19, 34 ). Also, RA may reduce the activity of RalDH 2 ( 25 ) and induce the expression of retinol and RA-binding proteins ( 24 ). Despite the variety of these findings, RA metabolism still remains to be understood in its complexity.& [/ H0 i/ p+ p. s
" E7 Y7 x$ L/ q7 A6 }) G2 R
We previously demonstrated that retinoids effectively reduce renal injury in the models of acute and chronic mesangioproliferative glomerulonephritis in the rat ( 32, 43 ). Retinoids exhibited antiproliferative and anti-inflammatory effects by preserving renal function, significantly reducing albuminuria, and reducing glomerular or tubular damage ( 8, 16 ). In turn, inflammatory mediators activate retinoid pathways ( 9 ).6 Y. i$ O& w& B7 N' e1 ]0 F8 f
! L( V) g- |# f( sFor this reason, we examined the regulation of the renal retinoid system in response to inflammatory renal injury in the course of acute anti-Thy1.1-glomerulonephritis (THY-GN). Furthermore, we wanted to know whether retinoid supplementation causes changes in the endogenous retinoid system. Therefore, in a second set of experiments we investigated the influence of isotretinoin supplementation on the components of the renal retinoid system.+ ~( ^6 l& m: R& ?
, j4 R: D. @, }, J+ b) BMATERIALS AND METHODS; ?/ C* E9 j, V/ j( ]- m) W
: ]$ U! l* V* A5 c1 r% e9 i' fExperimental protocol. Our study comprised two independent investigations, one dealing with changes in THY-GN, the other with the consequences of isotretinoin treatment for this condition. Features applying to both investigations include the following. w7 b7 U+ a# ^# [: T * {; l* Z7 g" _) F: {7 k; \4 KWe chose male Wistar rats (Charles River, Sulzfeld, Germany), weighing 150-160 g, as the object of our investigations. All animal experimentation was performed according to the "Deutsches Tierschutzgesetz" (German Animal Protection Law). The experimental groups represented nephritic groups. We induced acute anti-Thy1.1-mesangioproliferative glomerulonephritis by administering 500 µg of Moab 1-22-3 dissolved in PBS ( 15 ) as a single-shot injection into the tail vein of the rats on day 0. Moab 1-22-3 is a monoclonal antibody directed against the Thy1.1-like antigen on the surface of rat mesangial cells ( 14 ). In contrast to the monoclonal antibody OX-7, Moab 1-22-3 induces more severe mesangial cell injury, resulting in earlier mesangiolytic changes, matrix expansion, and higher proteinuria ( 35 ). ( Y( e! `# e7 T+ H% M' b" a, E5 |5 Y
Between groups, pair-feeding of the rats was used to ensure that all animals received identical amounts of nutrients.8 x: v* b1 t$ y6 D- ~+ M0 ?/ ]
3 E( C7 p, Q4 `) v/ I& ?" M( l1 [At the end of the experiments, the rats were given intramuscular injections of 5 mg/kg body wt xylazine (Bayer Vital, Leverkusen, Germany) and 100 mg/kg body wt ketamine 10% (WDT, Garbsen, Germany). The rats were then perfused with saline solution containing 0.5 g/l procaine hydrochloride via a cannula retrogradely inserted into the abdominal aorta ( 45 ). To drain blood, the inferior vena cava was incised. The perfusion pressure was adjusted to the individual systolic blood pressure (SBP). Glomeruli were isolated by a fractional sieving technique as described elsewhere ( 37 ). The kidneys of each rat were sieved individually. We used three grids with a final mesh size of 53 µm for glomeruli. The yield and purity of isolated glomeruli were comparable between 90%).' q% d. t; v: S* N6 F
# j: L7 {% J' F( O' U
In the first investigation, three experimental (THY; n = 8) and three control groups (CON; n = 6) were established. Day 3, 7, or 14 after the injection of the antibody represented the experimental end point when SBP was determined by tail-cuff plethysmography (tail plethysmograph built by the University of Heidelberg) under light ether anesthesia. SBP was also taken on day 0. The SBP for each rat was calculated as the average of three separate measurements at each session.5 y- P* W5 S/ T, B. `0 m. W- k
- k U; t. M: x2 y
In the second investigation, we established two experimental and two control groups. One experimental (THY; n = 8) and one control (CON; n = 6) group was pretreated with 10 mg/kg body wt isotretinoin (F. Hoffmann-La Roche, Basel, Switzerland) orally for 3 days ( day - 3 to day - 1 ) before Moab 1-22-3 or PBS administration. Isotretinoin-chow was prepared according to Schaier et al. ( 32 ). The other groups received a placebo (vehicle) chow. On day 7 the experiment was terminated. SBP was measured on days - 3, 0, and 7. 9 j6 U" M$ t' T: {! O+ J/ f G/ S! R
Renal morphology. Tissues for light microscopy were fixed in 10% buffered formalin and embedded in paraffin. Sections (4 µm) were stained with the periodic acid-Schiff reagent and counterstained with hematoxylin. For each kidney, total glomerular counts of cell nuclei were determined in at least 30 cortical glomeruli with diameters of at least 100 µm. The investigator was blinded for the treatment protocol. ' t/ R( Y% f" x' R8 T, |3 q, q 2 V1 x+ g" A) J2 NSolid-phase extraction and HPLC analysis. Before HPLC, individually isolated glomeruli and serum of each rat were prepared under dimmed yellow light by solid-phase extraction following the protocol of Collins et al. ( 6 ). HPLC analysis was performed on 10-µl samples of the eluate as described elsewhere ( 6, 41 ). 13- Cis RA and all- trans RA were determined by simultaneously measuring light absorption at 340 and 356 nm using a Shimadzu SPD-10AV detector. Retinoid levels were quantified by comparing peak areas and authentic standards (Sigma-Aldrich, Dreisenhofen, Germany)., \0 u' o& T8 W8 z5 }) p
- } M1 Y0 Z+ w% v" nRNA isolation and reverse transcription. RNA of shock-frozen glomeruli and liver tissue of every rat was isolated using TRIzol following the manufacturer's protocol (Invitrogen, Paisley, UK). Individual samples were checked for degradation of total RNA on 1% agarose gels. First, concentrations were spectrophotometrically determined at 260/280 nm and then adjusted to a final value of 0.2 µg/µl. For each sample, reverse transcription, as described elsewhere ( 44 ), was performed three times. The resulting cDNA was pooled. d7 r8 }* e& N3 T0 f
+ G5 ]3 j: N8 {! D9 \/ [9 t% X$ [$ xQuantitative PCR assay. Protocols by Paul et al. ( 30 ) and Wagner et al. ( 44 ), adjusted to our needs, were used to quantify mRNA. For each gene, a DNA deletion mutant was cloned ( 2 ), having the same sequence as the wild-type gene and identical primer binding sites but a deletion of 20% at most. This resulted in a shorter amplification product. For amplification, 0.1 µg of reversely transcribed RNA was used. Defined concentrations of DNA deletion mutants served as internal standards. (Primer sequences and hybridization positions are given in Table 1.) The PCR mixture contained 0.25 mmol/l deoxynucleoside triphosphate (dNTP; Promega, Madison, WI), 2.5 mmol/l MgCl 2, 20 mmol/l Tris·HCl (pH = 8.4), 50 mmol/l KCl, 80 nmol/l sense and antisense primers (Invitrogen), and 1 U of Taq DNA polymerase (Invitrogen). The thermal profile we used consisted of denaturation at 94°C for 30 s, annealing for 30 s as described in Table 1, and extension at 72°C for 30 s. Possible contamination with genomic DNA was prevented by PCR amplification in the absence of RT. Amplification products of the individual samples were separated by agarose gel electrophoresis and then digitized by using a gel documentation system (Intas, Göttingen, Germany) and Scion Image software (National Institutes of Health, Bethesda, MD). The ratio of the optical density (OD) of endogenous cDNA to the OD of mutant DNA was determined. Each sample was measured in two individual PCR assays for every investigated gene.- r5 T' b4 ?/ h7 H& _" K& a
$ U j& a' d" Y8 o6 {# PTable 1. Sequence, hybridization position of primers, annealing temperature, and cycle numbers used for quantitative RT-PCR of retinoid receptors and retinoid-metabolizing enzymes , P( Z1 u: ^, S' w/ t) h6 U# }4 |- j0 h, c/ p% q
Immunoblotting. Proteins were isolated from frozen glomeruli according to the manufacturer's instructions provided together with an immunoblotting kit (Santa Cruz Biotechnology, Santa Cruz, CA). After fractionating by SDS-PAGE in a 10-slot gel chamber, proteins were electrotransferred onto nitrocellulose filters that were blocked with 5% milk-0.05% Tween 20. Membranes were incubated with RAR antibody (1:250) for 1 h and then with 1:5,000 dilution of goat anti-rabbit IgG-horseradish peroxidase (both antibodies: Santa Cruz Biotechnology) and finally detected by chemoluminescence ( 11 ).8 q; n) L. ^8 w
! _2 W y Y8 ^* s( v' @! x
Statistical analyses. All values are expressed as means ± SE. Data were analyzed by using the nonparametric Mann-Whitney test or multivariate ANOVA and Bonferroni's posttest, as indicated. Statistical significance was accepted at P , V$ {$ s3 V7 V5 x* S( W
' U' Y# y7 S$ j: IRESULTS 5 w4 Q6 B/ H0 r! P3 J* I( r% K ! P- S( e! i5 J C m2 OSBP and glomerular cell count in the course of THY-GN: effect of isotretinoin. THY-GN rats experienced a significant and steady rise in SBP after Moab 1-22-3 injection, whereas SBP values in CON rats remained constant at all times ( Fig. 1 A ). . s+ J- K( B, `, E5 B9 P( S5 O3 P, S7 V( R+ b+ P, \; S
Fig. 1. A : time course of systolic blood pressure (SBP) 3, 7, and 14 days after induction of anti-THY1.1-glomerulonephritis (THY-GN). Values are means ± SE. SBP was measured in THY-GN rats compared with nonnephritic controls (CON). ** P : u8 g& m l. \5 N/ [; y" L
5 F/ _# q% i( p' O% O# h1 i
Isotretinoin treatment reduced the steepness of the rise in SBP in THY-GN rats but did not have any effect in nonnephritic CON on day 7 ( P ! n7 k4 @3 H& R- H# [
3 n1 ~4 J) M2 k% k4 Q$ V
Compared with CON rats, the glomerular count of cell nuclei was significantly lower in THY-GN rats on day 3 ( P 1 M! C% p9 H( d4 V4 n: i, f( i5 p1 O- m( z5 E
Serum and tissue retinoid concentrations. Serum 13- cis RA and all- trans RA levels were higher in isotretinoin-treated than in vehicle-treated groups ( Fig. 2, A and B ). , ^2 f# |+ P3 l! |+ Q & r1 B! h0 I$ ^( s( sFig. 2. A and B : serum retinoid concentrations. 13- Cis and all- trans retinoic acid (RA) were more abundant in isotretinoin- than in vehicle-treated groups. Values are expressed as ng retinoid/ml serum. The detection limit was 1 ng/ml; however, values over 10 ng/ml can only be considered as confident. C and D : glomerular retinoid concentrations. 13- Cis and all- trans RA were nearly undetectable in untreated groups, but they could be shown under isotretinoin treatment. The values are presented as pmol retinoid/g tissue. The detection limit was 0.7 pmol/g tissue (wet weight); however, values over 7 pmol/g tissue (wet weight) can only be considered as confident. 9 V: @4 g. t1 i# a( p: }5 Y3 b# z5 T0 c% P
13- Cis RA and all- trans RA levels in isolated glomeruli of vehicle-treated groups were close to or below the detection limit of the assay but made a shift to detectable values in the case of isotretinoin treatment ( Fig. 2, C and D ).+ r, I" v1 ?: g" G3 ?
& x% W/ ^+ p" s, z7 N3 Z# ]- O! s2 J4 kTime-dependent changes in glomerular gene expressions of retinoid receptors in the course of acute THY-GN. On day 3, levels of glomerular expression of the retinoid receptors RXR, RAR, and RAR significantly fell below control values in THY-GN rats ( P 0 R7 E1 B8 R: @6 k9 T
t, n+ G. [; K- G* I# @/ h$ Y
Fig. 3. A - D : glomerular gene expression of retinoid receptors in the course of acute THY-GN. Values are means ± SE. On day 3, the expression of the retinoid receptors, except retinoic acid receptor RAR, decreased significantly in nephritic glomeruli vs. nonnephritic CON rats, whereas on day 7 the glomerular expression of all receptors was 3-fold higher in THY-GN vs. CON rats. In contrast, on day 14, glomerular RAR and RAR mRNA reached the level in CON rats, whereas gene expression of RAR and retinoid X receptor RXR remained elevated in THY-GN vs. CON rats. The dotted line demonstrates normalized control level. * P & c% @) H) C$ \' U0 u$ j
9 P! l) ^9 x. e9 z
Hepatic mRNA expression of RAR and RXR in THY-GN. No difference in the hepatic expression of RAR and RXR could be observed between nephritic and nonnephritic groups on day 7. % b9 R# [' k* f/ j) w 4 \0 ?' \3 h2 `) V% |) `5 q$ X; vTime-dependent changes in glomerular gene expressions of retinoid-metabolizing enzymes in the course of acute THY-GN. The glomerular expression pattern of all retinoid-metabolizing enzymes (RalDH 1 and 2, RoDH 1 and 2) showed obvious similarities to the expression pattern of retinoid receptors: a 80% decline in relation to control values on day 3 ( P & x9 K& k1 k' y; Z! L$ @2 C0 Z0 `* ]. i* k$ R
Fig. 4. A - D : glomerular gene expression of retinoid-metabolizing enzymes in the course of acute THY-GN. Values are means ± SE. On day 3, the glomerular gene expression of retinol and retinal dehydrogenases (RoDHs and RalDHs) decreased by 80% in THY-GN vs. CON rats. In contrast, on day 7 the mRNA expression of these enzymes was 4-fold higher in nephritic glomeruli than that in nonnephritic controls. On day 14, RalDH 1 and RoDH 2 returned to CON levels, whereas RalDH 2 and RoDH 1 were significantly lower in THY-GN vs. CON rats. The dotted line demonstrates normalized control level. ** P , U+ }" a: A. ~* L
: i0 e' {9 E' F4 u. S. ^$ W GEffect of isotretinoin treatment on the glomerular expression of retinoid receptors and retinoid-metabolizing enzymes in the course of acute THY-GN. Oral supplementation with isotretinoin did not influence glomerular gene expression of retinoid receptors and of retinoid-synthesizing enzymes in nonnephritic rats. Isotretinoin did not influence the threefold elevation of the expression of these genes in THY-GN rats on day 7 ( Table 2 ).; ^: P) B. A$ G
7 j: d2 s) X3 G5 b% mTable 2. Effects of isotretinoin treatment on glomerular gene expression of retinoid receptors and retinoid-metabolizing enzymes in acute THY-GN on day 7; @7 j. \" I8 A0 C
; R4 Z. c- J2 ~5 j8 h* }0 B8 b
Glomerular protein expression of RAR on day 7. The glomerular expression of protein RAR was analyzed in vehicle- and isotretinoin-treated CON and THY-GN rats. RAR was found in glomeruli, where it was more abundant in THY-GN than in nonnephritic rats. Isotretinoin treatment did not alter RAR expression levels ( P # S+ A% k7 u3 v2 h/ B7 U, v" x4 l5 N4 i
Fig. 5. A and B : immunoblotting of glomerular RAR. Values are means ± SE. On day 7, RAR protein is more abundant in THY than in CON groups. Isotretinoin treatment had no influence on RAR protein levels. kD, kDa. * P 6 K5 C# D0 p/ Q5 Y" h1 ]
# d3 k1 O7 k, B7 ]$ j: N1 L
DISCUSSION% u+ K5 J- Q' l# `
# b& N- C" u* i' V" }Our data document a time-dependent response of the endogenous retinoid system to acute glomerular damage at different times in the course of acute mesangioproliferative glomerulonephritis in the rat. . S: Y' Q; ?) k 1 [+ w& `2 ~2 o; A' z; w3 FThe results of this study complement previous findings, in which there had been shown that isotretinoin reduces the blood pressure increase in acute THY-GN ( 43 ). The fact that the number of resident glomerular cells is reduced on day 3, reflecting early mesangiolysis in response to antibody-mediated injury, is characteristic of the model ( 14 ). In contrast, on day 7 the number of mesangial cells is increased again as a result of mesangial cell proliferation ( 12 ). This course of THY-GN was confirmed in our model. 4 k6 D; r* V3 g4 w' [& r, h! {- t% `) }5 V w" y3 Y
In agreement with our earlier work, treatment with isotretinoin lowered blood pressure, proliferation of mesangial cells, and glomerular damage ( 8, 22, 43 ). The mechanism of the antihypertensive effect of retinoids has not yet been clarified. The retinoid-mediated decrease in SBP is not specific for isotretinoin, because synthetic retinoids also lowered blood pressure in the THY-GN model ( 16 ). It may reflect alleviation of renal damage by retinoids, allowing the kidney to normalize SBP. Additionally, however, retinoids lower the expression of the angiotensin receptor in vitro and in vivo. They block the effects of angiotensin II ( 8, 11 ). This suggests another possible mechanism of blood pressure-lowering action by retinoids. ) h) ]: F6 ?) F* A% h9 F" R/ `" |- g( P0 y4 x
Antiproliferative effects of retinoids were demonstrated in different cell types, e.g., mesangial, vascular smooth muscle, endothelial, and tubular cells ( 11, 28, 36 ). The mechanisms of the antiproliferative action of retinoids have not been completely elucidated. One pathway is the interference with AP-1 by protein-protein interaction of retinoid receptors or by downregulation of its units c-Jun and c-Fos ( 33 ). Thus proliferation by AP-1-dependent genes, e.g., angiotensin II, PDGF, or endothelins, is lowered by retinoids ( 11, 20, 46 ).% [5 v, ]) Q. b. x
. |8 S Y5 m9 o( y7 C7 W1 AChanges in the endogenous retinoid system in response to acute glomerular injury were observed at the level of retinoid receptor expression, serum and glomerular retinoid levels, and glomerular expression of retinoid-metabolizing enzymes. 2 t/ ?/ V; Z/ D p( O! j6 M ( q, D7 [! |8 I# wRegulation of the different subtypes of retinoid receptors under different circumstances has been described in the past ( 4, 5, 17 ). On day 3, we observed a uniform decrease in glomerular retinoid receptor gene expression, with the exception of RAR. Because receptor expression was determined in isolated glomeruli and expressed per microgram RNA, it is obvious that this decrease does not reflect merely the effects of mesangiolysis but indicates either a relative reduction in retinoid receptor-expressing cells within the glomeruli or a reduced expression of the receptor number per cell.4 ?8 `% {9 k0 ~9 h
, r+ p% |- {* E* b6 d! K( k! x
Mesangial cells are known to express retinoid receptors ( 11 ), but so do endothelial and inflammatory cells, i.e., monocytes/macrophages. In contrast to the other receptors, RAR was not affected. Because this receptor is supposed to be ubiquitously expressed, its expression might not have been influenced by a shift in the relative distribution of glomerular cells, although it cannot be excluded that these are receptor subtype-specific differences in receptor regulation. 0 \9 _% ^- J5 @6 ~- j9 I/ h: J9 i4 [ W
Similar to the expression of glomerular retinoid receptors on day 3, the expression of retinoid-metabolizing enzymes was reduced as well. The cellular origin of expression of these enzymes in the glomerulus has not yet been determined. The reduction of gene expression of these enzymes may suggest a decrease in the local production of retinoids in the state of early glomerular damage in THY-GN. Whether these alterations locally influence the level of retinoids in the glomeruli is unknown. - Z/ k; M9 r8 \; I 2 R0 d" i$ d$ C: ?- \On day 7, a completely different expression pattern of the retinoid receptors and retinoid-metabolizing enzymes was observed: In contrast to day 3, the expression of all retinoid receptors and retinoid-metabolizing enzymes was significantly higher in nephritic compared with nonnephritic glomeruli. On the mRNA level, retinoid receptor and RalDH expression was increased about threefold, and RoDHs expression about five-fold. On day 7, both mRNA expression and protein expression of RAR in the glomeruli were increased.! u5 q: o$ L8 h! S ]& q; V0 g) N
4 D' f- C( A4 W* W1 V1 [6 ^Because at that time the number of mesangial cells was elevated as a result of the ongoing repair processes, it cannot be decided whether the enhanced receptor and enzyme expression was due to an increased glomerular number of retinoid receptor-expressing cells or an enhanced expression per cell. The findings on day 14, however, answer this question, because at that point the expression of retinoid system components and glomerular cell number did not go in parallel. On day 14, the expression of the different retinoid receptors had almost returned to normal, whereas the number of glomerular cells was still elevated. Similar findings were obtained with respect to the retinoid-metabolizing enzymes. # K3 T& D4 H: u) g * E( t- p( ~+ U) b3 xThese changes, therefore, cannot be explained on the basis of the number of retinoid receptor-expressing cells but strongly suggest a regulated cellular expression of these receptors. This interpretation is also supported by the fact that hepatic expression of RAR and RXR was not altered on day 7, when their expression in the glomeruli was significantly elevated, suggesting time-specific regulation. Because the expression of these two receptors did not change in the liver on the day of maximal change in the kidney, we took this as further evidence that the changes in the retinoid system are induced locally in the kidney due to renal disease. - I4 a' ^3 u' d" p# D f( J3 l, X# f1 r5 V0 j" X
A regulated response of the retinoid system to glomerular damage is further supported by the findings of a decreased amount of 13- cis RA and all- trans RA in the serum on day 7 in vehicle-treated nephritic rats. This may indicate "consumption" of the retinoids. A local reduction in available retinoids at the site of inflammation may trigger increased expression of retinoid receptors and of metabolizing enzymes. In the glomeruli, the endogenous retinoid levels were low or at the detection limit of these assays, precluding further analysis.4 x) j0 X6 K3 u+ T. y9 g
7 c" S# y5 G& ?/ @0 Z
We therefore applied isotretinoin to overcome a potential local reduction in retinoids. As a result of this maneuver, the elevated concentrations of 13- cis RA and its isomer all- trans RA were meshed in the serum of nonnephritic and nephritic rats as well as in the glomeruli. In the past, isotretinoin had been shown to reduce the number of glomerular cells and the level of SBP as surrogate markers of glomerular damage ( 22, 32, 42, 43 ). The fact that isotretinoin influenced neither the glomerular expression of retinoid receptors nor that of metabolizing enzymes is puzzling, because isotretinoin is metabolized to all- trans RA in the glomeruli. 13- Cis RA binds neither CRABP nor retinoid receptors ( 48 ), but it has been shown to act as a prodrug for all- trans RA in the skin ( 40 ). These findings indicate that retinoid supplementation is not the trigger for retinoid receptor expression in this model. Our results are in conflict with the results of other investigators, who had demonstrated that the levels of vitamin A or of retinoids can influence retinoid receptor expression ( 13 ). On the other hand, induction of retinoid receptors on day 7 may render renal tissue more sensitive to the action of isotretinoin, further supporting its anti-inflammatory and antiproliferative effects. 3 p4 R4 d/ m `; E/ x) I2 x 3 E: p7 X4 _1 J* Y# o, yThe results of our experiments indicate an active and specific response of the renal retinoid system in the early repair phase of acute mesangioproliferative glomerulonephritis. / ~, o1 U0 S* N5 k5 b% N9 q" Y; w8 V 9 [0 F3 ^' ]# }. rRetinoids have long been associated with wound healing. In dermatology, for instance, retinoids are used for the treatment of psoriasis, acne, and seborrhoea ( 26, 27 ). Previous work has indicated that retinoids play a role in the control of inflammation and wound healing, because they exert strong anti-inflammatory and antiproliferative effects ( 7, 11, 21 ). Therefore, a local reduction in available retinoids at the site of inflammation may trigger increased expression of retinoid receptors and of metabolizing enzymes.% Z% @3 j+ h4 c5 a( ~
: U0 {/ J9 Z' r5 |Conversely, a deficiency of vitamin A retards tissue repair. Retinoids reverse the inhibitory effects of glucocorticoids on wound healing by promoting epithelization and synthesis of collagen and ground-substance ( 1 ). Paquette et al. ( 29 ) have recently shown that topical all- trans RA treatment of patients with chronic leg ulcerations stimulates formation of granulation tissue, angiogenesis, and synthesis of new collagen. b! o: W5 S" Z+ e9 n7 C* R ( S: M* `; F' r6 g, E9 S* e `Above all, our findings suggest that renal tissue responds to exogenous retinoids and that the endogenous retinoid system is altered in response to renal injury. Whether these changes reflect the activation of this system or the response to local deprivation of retinoids after renal injury remains to be elucidated. Clearly, the changes in the endogenous system suggest an active role of retinoids in glomerular damage repair. 3 \5 U5 k5 s% A Q q$ C $ @$ ^ i& m9 |. H! R% sGRANTS6 U7 B2 n8 ?2 k3 T5 ]8 ~) d
4 q- t- J( f, d d: r. g
This project was supported by grants from the Deutsche Forschungsgemeinschaft (Bonn-Bad Godesberg, Wa748/8; to J. Wagner) and the Deutscher Dermatologen Kongress 1995-Verein zur Förderung der Dermatologie e.V. (to C. C. Zouboulis). $ f3 h; T2 n3 ]: p 【参考文献】2 W2 i- b5 ]9 X4 l6 [! m* t
Anstead GM. Steroids, retinoids, and wound healing. Adv Wound Care 11: 277-285, 1998. 6 G3 D/ J1 {. f$ p6 q + N% q! g4 g- K 1 H) S3 r C7 V8 s- @- ?7 d3 I! F$ s' I
Celi FS, Zenilman ME, and Shuldiner AR. A rapid and versatile method to synthesize internal standards for competitive PCR (Abstract). Nucleic Acids Res 21: 1047, 1993. # V$ b- C6 y' v7 j8 U 0 c1 L" x# _6 l6 [3 Y. I! b1 i % S6 T" i+ N5 I! @; @8 Y" A6 E& t7 D9 |
Chai X, Zhai Y, and Napoli JL. Cloning of a rat cDNA encoding retinol dehydrogenase isozyme type III. Gene 169: 219-222, 1996.( f% R4 }8 O1 w: C5 b" G' i0 p
. h1 w, H e* z* {5 i2 M$ m6 B9 h d1 B4 l
% S7 Q) h7 Q* B$ _# t' t" ^6 N8 RChambon P. The retinoid signaling pathway: molecular and genetic analyses. Semin Cell Biol 5: 115-125, 1994. . c. R0 b8 a8 } D' O1 W 0 X' Y4 [# p( E. G0 Y # @ p5 D$ `. S, X3 Y5 H) e0 X% `! Z% T% _% a; K( T
Chambon P. A decade of molecular biology of retinoic acid receptors. FASEB J 10: 940-954, 1996. 2 T% p& |0 b0 K q! L5 F h8 j y! [, H1 ^6 Z: `
+ i; G% u! {& X4 b6 K2 ^9 ?
4 [/ ^, y3 k9 ~; _$ ^% B& RCollins MD, Eckhoff C, Chahoud I, Bochert G, and Nau H. 4-Methylpyrazole partially ameliorated the teratogenicity of retinol and reduced the metabolic formation of all- trans -retinoic acid in the mouse. Arch Toxicol 66: 652-659, 1992.5 N% x3 Z$ [4 V! T, @; k4 W
1 T o4 U% a( o8 M : J* d8 ?* L" C# T5 K " \7 D/ U& {1 ?Datta PK and Lianos EA. Retinoic acids inhibit inducible nitric oxide synthase expression in mesangial cells. Kidney Int 56: 486-493, 1999.& o, ^) x- s, U; q& ]7 y
. q% h l8 }4 I: z w: [
4 l" x$ @6 Q- L$ |2 \8 J
& y7 Z, Y* B/ V; N2 ~# y* ]& @7 R1 Z7 k2 oDechow C, Morath C, Peters J, Lehrke I, Waldherr R, Haxsen V, Ritz E, and Wagner J. Effects of all- trans retinoic acid on renin-angiotensin system in rats with experimental nephritis. Am J Physiol Renal Physiol 281: F909-F919, 2001. ( ]1 b9 M: e7 t6 Q2 m! `- ~9 e! X- d( y) \
( z! h, p v1 ~7 U0 o: V7 n) _" V2 e
Gidlof AC, Romert A, Olsson A, Torma H, Eriksson U, and Sirsjo A. Increased retinoid signaling in vascular smooth muscle cells by proinflammatory cytokines. Biochem Biophys Res Commun 286: 336-342, 2001. 1 o; L+ L/ ]5 S' X4 A; d / e6 g) E) n$ C3 q# i H ) W' v p( K( a) A4 d) X3 w" _5 s6 l
Gilbert T and Merlet-Benichou C. Retinoids and nephron mass control. Pediatr Nephrol 14: 1137-1144, 2000. , C8 i$ l% N. b$ T* [, P 1 [0 Y% s5 T) E* [1 N 6 g! u: V g1 a/ v7 h, j- r" ?3 J7 L' ~ w: k$ t
Haxsen V, Adam-Stitah S, Ritz E, and Wagner J. Retinoids inhibit the actions of angiotensin II on vascular smooth muscle cells. Circ Res 88: 637-644, 2001. 2 F' A4 u, V" s. q; g' y1 v# s, m
N0 {. u6 p `0 K. h' x6 Z+ W+ V4 t6 d* \: s
Jefferson JA and Johnson RJ. Experimental mesangial proliferative glomerulonephritis (the anti-Thy-1.1 model). J Nephrol 12: 297-307, 1999. 5 @. o# y) I; C8 g. m! G) o$ {* P8 t C! t' Y
6 ?2 j: s6 K9 |: r& _9 A# E
6 S( H( [8 x2 ]( K9 T4 e& C
Kato S, Mano H, Kumazawa T, Yoshizawa Y, Kojima R, and Masushige S. Effect of retinoid status on alpha, beta and gamma retinoic acid receptor mRNA levels in various rat tissues. Biochem J 286: 755-760, 1992. # h: B6 [ u5 z! C/ }8 C, [6 `1 S$ ]7 N6 Q
1 O% [" c/ U/ H5 k& x2 B
& |6 k. C7 F/ G, H1 y# [5 t. z
Kawachi H, Oite T, and Shimizu F. Quantitative study of mesangial injury with proteinuria induced by monoclonal antibody 1-22-3. Clin Exp Immunol 92: 342-346, 1993. + @/ o) y& [9 J! V: e/ Y. U0 K6 a) F' N2 E0 S+ ~# s, y& E
4 @6 L- G. d* h/ F6 @7 @( s5 a2 K
' U/ H# ^5 w9 I, Q, }
Kawachi H, Orikasa M, Matsui K, Iwanaga T, Toyabe S, Oite T, and Shimizu F. Epitope-specific induction of mesangial lesions with proteinuria by a MoAb against mesangial cell surface antigen. Clin Exp Immunol 88: 399-404, 1992. + {, M( m' A/ q i$ U- x) u , l7 c: P3 R: g* q5 U4 E$ X+ S# ~
( H3 r/ |+ d' c* K6 U) G, S4 @4 }
Lehrke I, Schaier M, Schade K, Morath C, Waldherr R, Ritz E, and Wagner J. Retinoid receptor-specific agonists alleviate experimental glomerulonephritis. Am J Physiol Renal Physiol 282: F741-F751, 2002.- b0 E* ^+ k% W9 | L% U
3 s' q7 i2 L& ]# g* X6 I + }# h# }" J& u # i+ G: G+ J. r/ rMangelsdorf DJ, Kliewer SA, Kakizuka A, Umesono K, and Evans RM. Retinoid receptors. Recent Prog Horm Res 48: 99-121, 1993." ^1 I7 H& V C* T
* k! }) }5 x' G2 s, K& d7 d$ m% W9 L- C
3 F% m/ a9 ~9 z
Manzano VM, Munoz JC, Jimenez JR, Puyol MR, Puyol DR, Kitamura M, and Cazana FJ. Human renal mesangial cells are a target for the anti-inflammatory action of 9- cis retinoic acid. Br J Pharmacol 131: 1673-1683, 2000. ) [+ K' M) r! B( A/ V% n) c: B7 N3 A, o' ]/ \( |
5 Z6 @ X0 f: u% v ' v7 L7 I, L* R% i, K; `Matsuura T and Ross AC. Regulation of hepatic lecithin: retinol acyltransferase activity by retinoic acid. Arch Biochem Biophys 301: 221-227, 1993.! Y0 x* Q$ d5 ] M: m9 N, Z7 A
- O v' m4 Y6 W$ z; A6 Q) x' X2 ^* P6 T
1 E7 H O/ p& \# n/ O8 t/ @' lMercola M, Wang CY, Kelly J, Brownlee C, Jackson-Grusby L, Stiles C, and Bowen-Pope D. Selective expression of PDGF A and its receptor during early mouse embryogenesis. Dev Biol 138: 114-122, 1990.$ {+ K' }$ Z! B9 d6 o, l
4 x3 a7 E& |; m [6 s' Z
9 Y& f8 C8 J4 m( q/ ~+ ^* s) W7 [, F7 q; g+ P
Miano JM, Topouzis S, Majesky MW, and Olson EN. Retinoid receptor expression and all-trans retinoic acid-mediated growth inhibition in vascular smooth muscle cells. Circulation 93: 1886-1895, 1996.3 S' u: B& i: D: Y6 n" w
1 }0 W# m4 a1 t/ {6 E
. f* I" n$ S. s& F3 s& i, r! b- ?+ s7 f6 \
Morath C, Dechow C, Lehrke I, Ihling C, Floege J, Ritz E, and Wagner J. Retinoic acid suppresses TGF- overexpression in a rat model of glomerular damage (Abstract). J Am Soc Nephrol 10: 532A, 1999.7 v5 S5 S) G9 X/ T6 B
; x1 ]2 n. H3 b, T# {
) a3 g% O+ h# c& g" {3 ` % j/ {* A4 m6 b' QNa SY, Kang BY, Chung SW, Han SJ, Ma X, Trinchieri G, Im SY, Lee JW, and Kim TS. Retinoids inhibit interleukin-12 production in macrophages through physical associations of retinoid X receptor and NF B. J Biol Chem 274: 7674-7680, 1999.2 l4 u6 `" R# r P: g2 K1 Z
6 V9 x( Z& s: e+ f* f 4 h+ g- Y* R U% F 5 K7 o3 Q% Y$ ~Napoli JL. Interactions of retinoid binding proteins and enzymes in retinoid metabolism. Biochim Biophys Acta 1440: 139-162, 1999. , P2 B6 Z$ Y, Z+ h3 x* g( W# ?& x5 F. q! o
/ H3 {% H3 K: v1 |. H- c6 T4 e( s2 \0 U# T& O
Niederreither K, McCaffery P, Drager UC, Chambon P, and Dolle P. Restricted expression and retinoic acid-induced downregulation of the retinaldehyde dehydrogenase type 2 (RALDH-2) gene during mouse development. Mech Dev 62: 67-78, 1997. 4 b2 @6 l2 W9 r! J4 H 2 c$ a4 T* f& ~; L) K5 W& ?; N7 [5 ^/ X* S. b
$ C5 `3 H4 {6 U6 [! _Orfanos CE and Zouboulis CC. Oral retinoids in the treatment of seborrhoea and acne. Dermatology 196: 140-147, 1998.. B8 M9 C0 s5 g
* v& y! [/ x- O
" i& s: x8 n, W7 P( C/ }; a. \# m+ E5 K: ^1 g7 O4 Y2 N0 I
Orfanos CE, Zouboulis CC, Almond-Roesler B, and Geilen CC. Current use and future potential role of retinoids in dermatology. Drugs 53: 358-388, 1997. * H8 o: X3 O3 S \( y5 v. h) x$ z& r3 T
9 D/ y7 Y. Z! S! P7 m+ S7 ]+ `
5 `1 [. B; L4 UPaige K, Palomares M, D'Amore PA, and Braunhut SJ. Retinol-induced modification of the extracellular matrix of endothelial cells: its role in growth control. In Vitro Cell Dev Biol 27A: 151-157, 1991. 3 D/ A6 J7 n9 w2 }0 r1 e+ _9 r% n9 o
7 H5 y' \$ z: G* [6 t7 N2 H/ N" x7 V6 x
Paquette D, Badiavas E, and Falanga V. Short-contact topical tretinoin therapy to stimulate granulation tissue in chronic wounds. J Am Acad Dermatol 45: 382-386, 2001.1 @7 r+ w% V3 _0 Z3 \7 L7 a
. w) e6 Y: s+ Z8 M" F
6 z# O1 V' G. x; N) Z8 r+ F W* t# Q; s - Y# U. Z3 V7 X _2 k/ oPaul M, Wagner J, and Dzau VJ. Gene expression of the reninangiotensin system in human tissues. Quantitative analysis by the polymerase chain reaction. J Clin Invest 91: 2058-2064, 1993.) u: w/ X8 ]9 h8 x
% K3 g! \4 ^( z, u6 [' J , O* ^8 n- G6 J' R) b / B# b+ y$ K6 t) D' w9 |Ray WJ, Bain G, Yao M, and Gottlieb DI. CYP26, a novel mammalian cytochrome P -450, is induced by retinoic acid and defines a new family. J Biol Chem 272: 18702-18708, 1997. 6 ? [5 x; z' S" h7 s1 d5 Z% E( a$ d* y9 ^
* a$ @& V" y& I X% v ; L$ I, O4 p% c0 wSchaier M, Lehrke I, Schade K, Morath C, Shimizu F, Kawachi H, Grone HJ, Ritz E, and Wagner J. Isotretinoin alleviates renal damage in rat chronic glomerulonephritis. Kidney Int 60: 2222-2234, 2001.& D0 ]* e+ I+ p; U- T
3 `$ U t- N* B1 q M1 f8 c) {: q9 D9 {. y
7 C$ I# |1 n. M5 R" x
Schule R, Rangarajan P, Yang N, Kliewer S, Ransone LJ, Bolado J, Verma IM, and Evans RM. Retinoic acid is a negative regulator of AP-1-responsive genes. Proc Natl Acad Sci USA 88: 6092-6096, 1991. , I; z* Q8 H: d! j f% n/ x# J" ]3 p0 h5 a9 W - c+ d, U9 E, p/ O/ c) G9 Z- t; D2 g, f7 y+ A+ y# G
Shimada T, Ross AC, Muccio DD, Brouillette WJ, and Shealy YF. Regulation of hepatic lecithin:retinol acyltransferase activity by retinoic acid receptor-selective retinoids. Arch Biochem Biophys 344: 220-227, 1997.- i" P3 c, f8 a! f+ n! K
3 Y/ x. l7 H" a
& M1 S8 Y; ?* q/ d0 a5 U! ^9 c$ k G6 b* X' G3 H7 j7 I- m
Shimizu F, Kawachi H, and Orikasa M. Role of mesangial cell damage in progressive renal disease. Kidney Blood Press Res 22: 5-12, 1999.4 Y' B9 o x% S; F- t5 K* {
1 W$ @4 |' L4 Q. Y7 B; |
4 D( }( b5 E& y
# B% E, D6 }* `. M& ^
Simonson M. Anti-AP-1 activity of all-trans retinoic acid in glomerual mesangial cells. Am J Physiol Renal Fluid Electrolyte Physiol 267: F805-F815, 1994. 1 U* A# Q( r0 i% D* V/ h" T0 a6 [( G+ g7 V* Q5 \
0 R1 O8 Y6 {6 J' S" x
( \ E# g' Z) P' d( P2 J) A
Stahl RA, Helmchen U, Paravicini M, Ritter LJ, and Schollmeyer P. Glomerular prostaglandin formation in two-kidney, one-clip hypertensive rats. Am J Physiol Renal Fluid Electrolyte Physiol 247: F975-F981, 1984." h/ P0 _: Z2 p7 c9 I
- u. ?6 G2 i. v* @6 F. P
1 P7 [9 m' {3 @6 J& y& w5 V2 T: B: ]+ O$ O9 K
Sugawara A, Sanno N, Takahashi N, Osamura RY, and Abe K. Retinoid X receptors in the kidney: their protein expression and functional significance. Endocrinology 138: 3175-3180, 1997. 4 X6 E t) W0 J# t2 Q4 N# X6 y$ ^" o9 y$ M
3 C; H+ z8 x/ d8 v) F# d; w# H d) q+ r# ~- D
Suzuki A, Takahito I, Enyu I, Hirotugu I, Takashi F, and Masatugu H. Retinoic acid-generating enzyme, retinaldehyde dehydrogenase type 2 (RALDH 2) regulates the differentiation of renal epithelial cells (Abstract). J Am Soc Nephrol 12: 3856, 2001. ; }! o; ?. l5 g; t& z, X; D! v% P% |. z3 V6 f
/ q$ J/ n- N! {& J6 x, Q/ c$ |$ \% n
- B* u2 o% g8 z9 T' X4 sTsukada M, Schroder M, Roos TC, Chandraratna RA, Reichert U, Merk HF, Orfanos CE, and Zouboulis CC. 13- Cis retinoic acid exerts its specific activity on human sebocytes through selective intracellular isomerization to all-trans retinoic acid and binding to retinoid acid receptors. J Invest Dermatol 115: 321-327, 2000.. s7 B/ }" F0 o5 V+ H7 i
A8 f _. `! ^$ m: S
* o: ^% v+ h p; g0 n# I; @" m: ]: Z) ?8 A
Tzimas G, Burgin H, Collins MD, Hummler H, and Nau H. The high sensitivity of the rabbit to the teratogenic effects of 13- cis -retinoic acid (isotretinoin) is a consequence of prolonged exposure of the embryo to 13- cis -retinoic acid and 13- cis -4-oxo-retinoic acid, and not of isomerization to all- trans -retinoic acid. Arch Toxicol 68: 119-128, 1994. . B' u% u( n: p- T% S; |7 C' t( k
, z! U# C( Y: P5 ^- k
' E! e1 U# l: ^7 w. r e5 AWagner J. Nuclear (receptor) power: retinoids in rat mesangioproliferative disease. Nephrol Dial Transplant 17, Suppl 9: S81-S83, 2002. D) j/ P3 z9 _9 t$ w9 L1 H: g. b' ^. ]6 r7 B) E7 `# @
4 v/ ^# i( O. c9 \" V! w7 H; m6 x7 L; n r, H
Wagner J, Dechow C, Morath C, Lehrke I, Amann K, Waldherr R, Floege J, and Ritz E. Retinoic acid reduces glomerular injury in a rat model of glomerular damage. J Am Soc Nephrol 11: 1479-1487, 2000. & Q; A8 g: V* f6 d1 Q6 y( j! a( `% w! R; Y- S {! i
w. w5 h! t2 x, F# n7 u; Z8 r" e3 T; I
Wagner J, Gehlen F, Ciechanowicz A, and Ritz E. Angiotensin II receptor type 1 gene expression in human glomerulonephritis and diabetes mellitus. J Am Soc Nephrol 10: 545-551, 1999.; ]( P1 }' w9 a! e5 A' ^
$ V8 X. P9 c, T4 F. `
& A3 r) l" z6 j( }2 _! W) O0 j$ w# Z2 E
Wagner J, Haufe C, Klotz S, Wystrychowksi A, Ganten D, and Ritz E. Accelerated progression of chronic renal failure in transgenic rats carrying an additional renin gene. J Hypertens 15: 441-451, 1997.' Z* Z7 c! ~/ Y% j B, w
" y( Y* ^( X9 h( u% s2 |. t
5 @4 ]6 f9 Y2 E9 O! }5 M$ K2 W
$ h8 I. ]4 {6 s% ~6 s; ?. I) d$ s ) l! d, q* T' a+ s2 fZhou X, Shen X, and Shemshedini L. Ligand-activated retinoic acid receptor inhibits AP-1 transactivation by disrupting c-jun/c-fos dimerization. Mol Endocrinol 13: 276-285, 1999. 6 S I& y; x c" m4 N# V3 [+ w5 B. Y7 m" e& i; h5 ?* |2 y1 k
: U1 s3 }% C Y, Y" @1 k- M( v1 ~' N. n' g [4 \# q
Zouboulis C and Orfanos C. Retinoids. In: Drug Therapy in Dermatology, edited by Millikan L. New York: Dekker, 2000, p. 171-233.作者: tempo 时间: 2015-5-23 15:19