|
- 积分
- 0
- 威望
- 0
- 包包
- 0
|
作者：Lamine Aoudjit, Monica Stanciu, Hongping Li, Serge Lemay, Tomoko Takano作者单位：Department of Medicine, McGill University Health Centre, Montreal,Quebec, Canada H3A 2B4
) n0 [/ q' t$ [
2 j/ {1 q/ Q3 ` j* f, Z! m& A b3 N% t+ o7 Z+ v3 c0 o
0 F/ N3 @* i8 T% h
( P4 ~% R. q/ h 5 I5 c, J% ~( I, D4 J+ n7 P
' o5 E3 b( z- `8 S6 T$ e 1 f1 U+ h8 L4 ?! c8 g. ^& |
; G, n8 [- F6 E& h' r: P
1 N$ L9 R3 l0 D5 ? h9 d* s
" L1 W, M- m2 a' h) Y" }8 Z3 z4 n 1 f: V6 W/ C' [9 r4 m, N C6 b' x' n
5 T2 R/ z$ q1 [
【摘要】" i; J* g$ k" o/ I& C
In the passive Heymann nephritis (PHN) model of rat membranous nephropathy,complement C5b-9 causes sublytic injury of glomerular epithelial cells (GEC).We previously showed that sublytic concentration of C5b-9 triggers a varietyof biological events in GEC. In the current study, we demonstrate thatcomplement activates p38 MAPK in GEC and address the role of p38 incomplement-mediated cell injury. When cultured rat GEC were stimulated withcomplement, p38 kinase activity and phosphorylation were increased by 2.4-fold, compared with control. Treatment with p38 inhibitorssignificantly augmented complement-mediated cytotoxicity. In contrast, when the constitutively active mutant of transforming growthfactor- -activated kinase 1 (TAK1), a kinase upstream of p38, wasexpressed in GEC in an inducible manner, cytotoxicity was significantly reduced, compared with uninduced cells. p38 inhibitors abolished theprotective effect of TAK1 expression. By analogy to cultured cells, p38activity was also increased in glomeruli from rats with PHN and treatment withthe p38 inhibitor FR-167653 increased proteinuria. Complement inducedphosphorylation of MAPK-associated protein kinase-2 (MAPKAPK-2), a kinasedownstream of p38 in GEC. Heat shock protein (HSP27) is acytoskeleton-interacting substrate of MAPKAPK-2. Overexpression of thewild-type HSP27, but not a non-phosphorylatable mutant, markedly reducedcomplement-mediated GEC injury. In summary, complement activates p38 MAPK inGEC in vitro and in glomeruli from rats with PHN. The activation of p38 MAPKappears to be cytoprotective for GEC against complement-mediated GEC injury.Phosphorylation of HSP27 may mediate this cytoprotection.
E1 s: t, C" J' ]; |7 |5 ]4 ?1 g 【关键词】 proteinuria passive Heymann nephritis
9 T& c/ X# c' B" R THE GLOMERULUS IS the filtration unit of the kidney. Visceral glomerular epithelial cells (GEC), also known as podocytes, are critical forstructural integrity as well as barrier function of the glomerulus. Theimportance of GEC in glomerular barrier function has been highlighted by therecent discoveries that mutations of molecules specifically expressed in GEC,such as nephrin, podocin, or -actinin-4, cause severe proteinuric glomerular diseases ( 1, 7, 9 ). Membranous nephropathy is a common cause of nephrotic syndrome in adults, and the rat model of passiveHeymann nephritis (PHN) has been extensively used to study the pathophysiologyof membranous nephropathy and its validity was recently discussed in detail( 32 ). It is known that in PHN,complement C5b-9 causes GEC injury. Activation of the complement cascade neara cell surface leads to assembly of terminal components and insertion of theC5b-9 membrane attack complex in the plasma membrane. Nucleated cells require multiple C5b-9 lesions for lysis, but, at lower doses, C5b-9 induces sublyticinjury and various metabolic effects. These include activation ofphospholipases [e.g., cytosolic phospholipase A 2 (cPLA 2 ), phospholipase C( 2, 3 )] and protein kinases [e.g., ERK, JNK ( 3, 17, 31 )], and induction/activationof various other molecules [(e.g., growth factors, nuclear factor- B,and c-Jun) ( 31 )]. C5b-9 alsocauses reorganization of the actin microfilaments in cultured GEC( 33 ). Reorganization of the actin cytoskeleton may, in part, explain the mechanisms with whichcomplement-mediated GEC injury leads to morphological changes and barrierdysfunction.
- F& I! w" P$ `- H, D; m
) I! P# f* P' n0 y( o% YMitogenic and stress-related extracellular signals are primarily transmitted intracellularly through either of the three MAPK pathways: 1 ) the ERK pathway, typically activated by growth factors; 2 ) the JNK pathway, also known as the stress-activated protein kinasepathway; and 3 ) the p38 pathway( 14, 23 ). p38 is activated by aseries of cytokines, growth factors, in addition to the stress andproinflammatory signals and has important roles in stress responses, cellsurvival, apoptosis, and inflammation( 15 ). Activation of p38involves phosphorylation of threonine and tyrosine residues in a TGY motif,resulting in increased enzyme activity. p38 lies downstream of various signaling molecules including the small G proteins Rac and Cdc42 and theprotein kinase transforming growth factor- -activated kinase 1 (TAK1).p38 is directly activated by MAPK kinases (MKKs) MKK3 and 6. Substrates forp38 include transcription factors, such as ATF-2 or CREB, and protein kinases,such as MAPK-associated protein kinase (MAPKAPK)-2 and -3. Of interest, MAPKAPK-2, which is activated by p38, phosphorylates heat shock protein(HSP27; Ref. 5 ). Although thefunction of HSP27 is not understood completely, it is known to associate withthe actin cytoskeleton and to modulate its organization( 5, 13, 22 ). Overexpression of HSP27in fibroblasts increased stress fiber stability during hyperthermia, preventedcytochalasin D-mediated actin depolymerization, and increased corticalfilamentous actin, ruffling, and pinocytotic activity( 11, 12 ). Overexpression ofnon-phosphorylatable HSP27 inhibited each of these activities ( 11, 12 ).
, B* V. F F% v% m7 Q" J4 l5 \( ^
* z8 H# {' n- o* QWe reported previously that complement activates two of the MAPK pathways,namely the ERK and JNK pathways in GEC( 3, 17, 31 ). The JNK pathway protectsGEC against complement-mediated injury( 17 ) and also contributes tocomplement-mediated cyclooxygenase (COX)-2 upregulation in GEC( 31 ). In addition, it wasrecently reported that in GEC, complement-activated ERK might be important incellular response to complement-induced DNA damage( 18 ). The present studyaddressed the role of another member of the MAPK family, p38 in GEC. Wedemonstrated that complement activates p38 in GEC and that p38 activationleads to cytoprotection. We also showed that HSP27 downstream of p38 maycontribute to cytoprotection.
) H& \4 d& f: Q- d+ |$ K5 i1 [" N9 M* i/ e
MATERIALS AND METHODS0 k. ]$ I4 D3 U, @" |8 q' Y* } q
0 _8 S9 q- `2 b# \; O8 ~2 k
Materials. FR-167963 was a gift from Fujisawa Pharmaceuticals (Osaka, Japan). Tissue culture media, Ecdysone-Inducible Mammalian ExpressionSystem, ponasterone A, hygromycin, and zeomycin were purchased fromInvitrogen-Life Technologies (Burlington, ON). NuSerum was from CollaborativeResearch (Bedford, MA). Anisomycin, arachidonic acid,2',7'-bis(carboxyethyl)-5(6)-carboxyfluorescein acetoxymethylester (BCECF-AM), C8-deficient human serum (C8D), purified human C8,glutathione (GSH), N -acetyl cysteine (NAC), and hydrogen peroxidewere from Sigma Chemical (St. Louis, MO). Anti-HA antibody (3F10), proteaseinhibitor cocktail, and FuGENE 6 were from Roche Diagnostics (Laval, Quebec).Antibodies for phospho-p38 and p38, MAPKAPK-2, and p38 MAP Kinase Assay Kitwere from New England BioLabs (Mississauga, Ontario). Anti-phospho-p38 antibody specifically recognizes p38 that is phosphorylated at Thr180 andTyr182. Anti-phospho-MAPKAPK-2 antibody specifically recognizes MAPKAPK-2 thatis phosphorylated at Thr222. Anti-phospho-HSP27 that specifically recognizesHSP27 phosphorylated at Ser 82 was from Santa Cruz (Santa Cruz, CA). PD-169316was from Calbiochem (La Jolla, CA). Anti-HSP27 antibody was from Upstate (Lake Placid, NY). Reagents for gel electrophoresis were from Bio-Rad (Mississauga,Ontario). HA-tagged TAK1 cDNA [constitutively active mutant( 36 )] was a gift from Dr. K.Matsumoto (Nagoya University, Japan). Plasmids encoding wild-type andnon-phosphorylatable mutant hamster HSP27 were gifts from Dr. J. Landry (LavalUniversity, Quebec) ( 10 ).Coding regions of these plasmids were subcloned into a Hind III site ofpcDNA3.1 (Invitrogen-Life Technologies).
$ Q$ K/ R; }2 B7 N. K: Q V4 g% r' ]8 N4 o
GEC culture and stimulation by complement. Culture,characterization, and stable transfection of rat GEC were described previously ( 2, 29 ). A subclone of GEC, whichstably overexpress cPLA 2 (GEC-cPLA 2 )( 16, 29 ), was used unless otherwisenoted. GEC-Neo is a control GEC clone, which was transfected with vector alone and expresses only a small amount of cPLA 2. GEC-cPLA 2 shows augmented response (in cytotoxicity and eicosanoid generation) tocomplement stimulation, compared with GEC-Neo( 16, 29 ). Because GEC in vivoexpress a significant amount of cPLA 2, GEC-cPLA 2 betterrepresent GEC in vivo than GEC-Neo( 4 ). Inducible-TAK1 clones weregenerated using Ecdysone-Inducible Mammalian Expression System(Invitrogen-Life Technologies). First, GEC were stably transfected with theplasmid pVgRXR and selected by zeomycin. Second, HA-TAK1 subcloned into thepIND/Hygro vector was stably transfected and selected by hygromycin.Expression of HA-TAK1 was induced with an insect hormone, ponasterone A (4µM). Rabbit antiserum to GEC( 2 ) was used to activatecomplement on GEC membranes. Briefly, GEC were incubated with antiserum (5%vol/vol) for 40 min at 22°C. GEC were then incubated with normal humanserum (NS; 2.5-5.0% vol/vol), or heat-inactivated (decomplemented) humanserum (HIS; 56°C, 30 min, 2.5-5.0% vol/vol) in controls, for theindicated times at 37°C. In some experiments, antibody-sensitized GEC wereincubated with C8D (5.0% vol/vol) reconstituted with or without purified humanC8 (80 µg/ml undiluted serum). We generally used heterologous complement tofacilitate studies with complement-deficient sera and to minimize possiblesignaling via complement-regulatory proteins; however, in previous studies,results of several experiments involving arachidonic acid metabolism wereconfirmed with homologous (rat) complement( 2 ). Sublytic concentrations ofcomplement ( 5% NS) were established previously( 2 ). Previous studies showedthat, in GEC, complement is not activated in the absence of antibody.
7 w: d5 B* C7 [& |' \
4 x0 S' U6 h1 T. _! |2 Y6 vImmunoblotting. Immunoblotting was performed as describedpreviously ( 29 ). Cells orglomeruli were lysed in IP buffer [1% Triton X-100, 125 mM NaCl, 10 mM Tris(pH 7.4), 1 mM EDTA, 1 mM EGTA, 2 mM Na 3 VO 4, 10 mMsodium pyrophosphate, 25 mM NaF] containing protease inhibitor cocktail (RocheDiagnostics). After insoluble components were removed by centrifugation(14,000 rpm, 5 min, 4°C), protein concentration of supernatants wasquantified using a commercial reagent (Bio-Rad). Equal amounts of protein wereseparated by 8% SDS-PAGE under reducing conditions. Proteins wereelectrophoretically transferred to nitrocellulose membrane, blocked with 5%dry milk, and incubated with first antibodies for 16 h at 4°C. After threewashes, membranes were incubated with secondary antibodies conjugated withhorseradish peroxidase, and horseradish peroxidase activity was detected byenhanced chemiluminescence (Amersham Pharmacia Biotech, Baie d'Urfé, Quebec).3 N6 V1 l8 @1 }% x7 o1 @
1 x5 W B7 q2 E7 M2 B: o @/ v5 R( @
p38 kinase assay was performed using p38 MAP Kinase Assay Kit (New EnglandBioLabs) as per manufacturer's instructions. In brief, cells or glomeruli werelysed in 1 x lysis buffer and phosphorylated (activated) p38 wasimmunoprecipitated with a phospho-specific monoclonal antibody.Immunoprecipitates were subjected to kinase assay using GST-ATF-2 assubstrate, and ATF-2 phosphorylation was then detected by immunoblotting usinganti-phospho-ATF-2 (Thr71) antibody.+ @4 G: Y$ K7 f! |- H8 T7 Z
! Z+ e* B' F. q
Cytotoxicity assay was performed as described previously( 2, 30 ). BCECF is a small molecule(molecular mass 520.5) that could be loaded into cells as acetoxymethyl ester.It is released into supernatants when cells are injured; thus it could be used as a marker of cell injury ( 2 ).After complement stimulation, amounts of BCECF released into supernatants andassociated with cells were quantified using a fluorometer, respectively. Specific release of BCECF into supernatants was calculated to quantifycytotoxicity using the following formula
4 t, ^% i% u! s0 I" g4 ~: R V
; i, H8 W4 Q+ N) D, LLactate dehydrogenase (LDH) is an intracellular enzyme that issubstantially larger than BCECF. After complement stimulation, LDH activitiesin supernatants and cell lysates were quantified as described previously( 19 ). Specific release of LDHwas calculated as described previously( 19 ) in an analogous manner toBCECF. We previously employed both BCECF assay and LDH assay to quantifycytotoxicity and obtained parallel results. In general, we use higherconcentrations of serum for LDH assay because LDH is substantially larger thanBCECF; thus it requires more injury to be released into supernatants.
9 q1 V, U3 z/ e7 a
- b* Q) a+ m8 p$ QQuantification of F-actin. Quantification of F-actin was performed as described previously ( 30 ).After stimulation, cells were washed with PBS and fixed with 2%paraformaldehyde and 4% sucrose in PBS for 10 min. After incubation inNH 4 Cl (50 mM) for 5 min, cells were permeabilized with 0.5% TritonX-100 for 10 min and blocked with 3% BSA in PBS for 30 min. Cells wereincubated with TRITC-phalloidin (0.1 µg/ml) for 60 min. After three washeswith PBS, TRITC-phalloidin was extracted with 2 ml of methanol for 30 min withagitation, and cells were further washed with an additional 1 ml of methanol.Fluorescence of the pooled methanol extracts was quantified in a fluorometer (excitation 542 nm/emission 563 nm). To confirm that there were similarnumbers of viable cells in each well, cells were further washed three timeswith PBS and were incubated with a nucleic acid-binding fluorescent dye,Toto-3 (100 nM, Molecular Probe, Eugene, OR). Cells were scraped with a rubberpoliceman, and cell suspensions were transferred to test tubes and sonicated. Fluorescence of the samples was quantified in a fluorometer (excitation 642nm/emission 660 nm), and results of TRITC-phalloidin were normalized to Toto-3values./ ^; Y( f5 L" Q. T% S3 T( ^& r" t
1 w& Z& R3 X A+ I
Induction of PHN and treatment with FR-167963. PHN was induced inmale Sprague-Dawley rats (150-175 g body wt, Charles River, St.Constant, Quebec) by intravenous injection (400 µl/rat) of sheep anti-Fx1Aantiserum as described previously ( 29 ). Preparation of anti-Fx1Aantiserum was described previously ( 20 ). Rats did not showsignificant proteinuria up to 7 days after injection. However, significantproteinuria was observed 14 days after injection ( 160 mg/day; normal ratsexcrete less than 10 mg of protein per day). FR-167963 was suspended in 0.5% methylcellulose solution, and 32 mg · kg - 1 · day - 1 were given subcutaneously from day 7 through 14. On day 14, rats were killed after24-h urine collection in metabolic cages and glomeruli were isolated bydifferential sieving as described previously( 29 ).
O; @- i" d) _
' b- Y4 D; X( X7 B1 a/ \Quantification of glomerular rat C3 deposition. Quantification ofglomerular rat C3 deposition was performed as described previously with aminor modification ( 17 ).Briefly, 4-µm cryostat kidney sections were stained withfluorescein-conjugated rabbit anti-rat C3 (Cappel, Scarborough, Ontario). Theimmunofluorescence images were captured using a Nikon Diaphotoimmunofluorescence microscope and a Nikon Coolpix995 digital camera with afixed exposure time. Images were transferred to the Adobe Photoshop program,and fluorescence intensity of glomeruli was quantified by histogram analysis.At least eight glomeruli were quantified for each group.. Q- {. F1 Z- F5 s" c7 t3 j, B
. I! H, v) }8 D& w2 Y. u
Statistics. Data are presented as means ± SE. The t -statistic was used to determine significant differences between twogroups. One-way ANOVA was used to determine significant differences amonggroups. Where significant differences were found, individual comparisons weremade between groups using the t -statistic.+ g+ G0 i# U! h6 h* m5 B
) @# f/ s1 `9 a& d" d
RESULTS
. U( L1 `4 y" }7 e- o' a' `8 ?4 q Y( m
Complement activates p38 in GEC. We and others reported previously that complement C5b-9 activates kinases such as protein kinase C, ERK( 3 ), and JNK( 31 ) in GEC. In the currentstudy, we examined if complement activates another MAPK, p38. Whenantibody-sensitized GEC were exposed to NS to form complement C5b-9, p38phosphorylation increased by 2.2 ± 0.5-fold, compared with control(cells exposed to HIS) [no treatment 0.7 ± 0.1, HIS 1.0, NS 2.2± 0.5 ( P 1.4( P n = 6; Fig. 1 A ]. The enzymeactivity of p38 quantified by immune-complex kinase assay also increased incomplement-treated cells by 2.4 ± 0.3-fold, correlating to the increasein phosphorylation [no treatment 0.4 ± 0.3, HIS 1, NS 2.4 ± 0.3( P 2 O 2 4.5 ± 1.1( P n = 3-4; Fig. 1 B ]. To verify ifC5b-9 assembly was actually required for p38 activation, we incubatedantibody-treated GEC with C8D, with or without reconstitution with purifiedC8. C8D without C8 forms C5b-7, which was previously shown to be biologically inactive in GEC ( 2 ). GECincubated with C8D did not show significant p38 phosphorylation, compared withHIS. However, when C8D was reconstituted with purified C8, phosphorylation ofp38 was evident, indicating that formation of C5b-9 is required for p38activation ( Fig. 1 C ).Densitometric analysis showed that the relative p38 phosphorylation was HIS:1.0, C8D: 0.9, C8D C8: 2.4 (average of 2 experiments; Fig. 1 C ).% X4 ?7 e1 Q) K2 d R% h
! G% o: P3 j6 I: C3 eFig. 1. Complement C5b-9 stimulates p38 in glomerular epithelial cells (GEC) inculture. A : GEC-cytosolic phospholipase A 2 (cPLA 2 ) (GEC that stably overexpress cPLA 2, see MATERIALS AND METHODS ) were incubated with anti-GEC antiserum (5%vol/vol, 22°C, 40 min) and normal human serum (NS; 2.5% vol/vol, toassemble C5b-9, 37°C, 40 min) or heat-inactivated serum (HIS; 2.5%vol/vol, control). Anisomycin (1 µM) was used as positive control. Top : immunoblot. Bottom : densitometry. * P P n = 6 for each group. B : GEC-cPLA 2 were stimulated as in A, and p38activity in cell lysates was evaluated as in MATERIALS AND METHODS.H 2 O 2 (1 mM) was used as positive control. Top :immunoblot. Bottom : densitometry. P P n = 3-4 for each group. C :antibody-sensitized GEC-cPLA 2 were incubated with HIS (2.5%vol/vol), human C8-deficient serum (C8DS; 2.5% vol/vol), or C8DS supplementedwith recombinant human C8 (80 µg/ml in undiluted serum) for 40 min at37°C. H 2 O 2 (1 mM) was used as positive control. Top : immunoblot. Bottom : densitometry (mean of 2experiments).+ y; V/ N2 \0 ?6 a$ q
# Q0 m/ j1 T# }% k- x0 g! H
To assess if complement-mediated p38 activation occurs in vivo, we nextstudied p38 activation in glomeruli from rats with PHN. PHN was induced by asingle injection of anti-Fx1A antiserum ( MATERIALS AND METHODS ),and glomeruli were isolated 14 days later, when rats developed significantproteinuria ( 160 mg/day, control rat: sixfold in glomeruli from rats with PHN, compared withcontrol (control 13 ± 1, PHN 80 ± 22, P n = 3 and 6; Fig.2 A ). Phosphorylation of p38 was also higher in PHN (2.9± 0.2-fold), compared with control (control 1 ± 0.1, PHN 2.9± 0.2, P n = 3 and 6; Fig. 2 B ). Theseresults indicate that p38 is activated by complement in GEC in vitro and invivo.. Y/ y" K0 u7 _8 l
# }. s1 B* y R0 Z' S# a' n5 dFig. 2. p38 is activated in glomeruli from rats with passive Heymann nephritis(PHN). A : PHN was induced by intravenous injection of sheep anti-Fx1Aantiserum. On day 14, rats were killed and glomeruli were prepared bydifferential sieving. Glomerular lysates were analyzed for p38 activity( A ) or p38 phosphorylation ( B ) as in Fig. 1. Top :immunoblot. Bottom : densitometry. * P P n = 3-6 rats for each group., s7 B5 f j5 _
; K8 o( L7 t/ V- D2 t4 y9 t
Arachidonic acid contributes to complement-induced p38 activation. We next addressed the mechanisms of complement-induced p38 activation. It wasreported previously that complement activates cPLA 2 in a[Ca 2 ] i - and protein kinase C-dependentmanner, leading to a liberation of arachidonic acid from the membranephospholipid pool ( 16 ). Wealso reported that complement-induced JNK activation was, at least in part,mediated by arachidonic acid release ( 17 ). We first studied ifliberation of arachidonic acid contributes to complement-induced p38activation. GEC-cPLA 2 were stimulated with antibody and complement,and p38 activation was compared with control GEC, which were transfected withvector only (GEC-Neo). p38 was activated by complement 1.4 ± 0.1 timesin GEC-Neo, whereas the stimulation was 2.0 ± 0.1 times inGEC-cPLA 2 ( n = 3, P acid, at least partially, contributes tocomplement-induced p38 activation ( Fig.3 A ). When GEC-Neo were stimulated with exogenousarachidonic acid (30 µM), p38 was clearly activated (2.8 ± 0.6-foldof control, P n = 5), supporting the role ofarachidonic acid in p38 activation ( Fig.3 B ).+ n; E9 J/ t% b7 Q6 Z
! W6 M6 U' O4 D+ C7 o3 n
Fig. 3. Arachidonic acid (AA) contributes to complement-induced p38 activation. A : GEC-Neo (vector-transfected cells) and GEC-cPLA 2 (GECthat stably overexpress cPLA 2 ) were stimulated with complement for40 min, and p38 phosphorylation was studied by immunoblotting. Top :immunoblot. Bottom : densitometry. * P n = 3 each. B : GEC-Neo were stimulated with AA (30 µM) orvehicle [ethanol, control (cntl)] for 20 min, and p38 phosphorylation wasstudied by immunoblotting. Top : immunoblot. Bottom :densitometry. ** P n = 5 each.- R: D. H/ M0 a# B, L
1 }& d- I! k! }& E6 v4 Z/ r" _Unstimulated cultured rat GEC express COX-1, whereas COX-2 expression isinduced by complement. Arachidonic acid liberated by cPLA 2 isfurther converted by COX-1 and -2 into biologically active eicosanoids such asPGE 2, PGF 2, and TXA 2 ( 29 ). These eicosanoids actvia respective specific cell-surface receptors in an autocrine fashion and areknown to activate p38 in some systems. To study if eicosanoids contribute tocomplement-induced p38 activation, we stimulated GEC with PGE 2,PGF 2, and U-46619 (stable analog ofTXA 2 ). When individually tested, these three eicosanoids failed toactivate p38 consistently. Only when three eicosanoids were testedsimultaneously was there a modest increase in p38 phosphorylation ( 20%,not shown). Furthermore, a nonselective COX inhibitor, indomethacin, failed toinhibit complement-induced p38 activation (not shown). Taken together, theseresults indicate that arachidonic acid contributes to complement-induced p38 activation but not through conversion into COX metabolites. Instead,arachidonic acid itself and/or other mediators stimulated by arachidonic acid,such as reactive oxygen species (ROS) (see Role of ROS in the activationof p38 by complement ), may be responsible for p38 activation.8 d% i5 V8 x) D' z2 l
; S8 q9 U1 t$ zRole of ROS in the activation of p38 by complement. We previously showed that complement stimulates ROS generation in GEC in an NADPHoxidase-dependent manner, which contributes to JNK activation ( 17 ). We also showed thatarachidonic acid stimulates ROS generation in GEC( 17 ). To test whether ROScontribute to complement-mediated p38 activation, we examinedcomplement-mediated p38 activation in GEC treated with anti-oxidants, GSH, andNAC. As shown in Fig. 4, GSH(10 mM) and NAC (10 mM) completely abolished complement-induced p38phosphorylation [control 3.7 ± 0.3-fold of HIS, GSH 0.9 ±0.03-fold of HIS ( P NAC 1.0 ±0.07-fold of HIS ( P n = 3]. Thus ROSare likely to be responsible for complement-induced p38 activation. Onemechanism of ROS generation might be arachidonic acid released bycPLA 2 ( 17 ).6 B# H4 T6 a; K
' P) u7 g" o8 i4 a1 gFig. 4. Complement-mediated p38 activation is dependent on reactive oxygen species.GEC-cPLA 2 were stimulated with antibody and complement (NS). Cellswere preincubated with glutathione (GSH; 10 mM) or N -acetyl cysteine(NAC; 10 mM) for 30 min before the stimulation. p38 activation was evaluatedby immunoblotting for phospho-p38. Top : immunoblot. Bottom :densitometry. * P n = 3 each.
! s5 s( i$ a8 D7 d* P( X1 N3 I5 K
Impact of p38 activation on GEC function. We next studied the impact of p38 activation on complement-mediated GEC injury. We initiallyhypothesized that p38 activation contributes to complement-mediated cellinjury. However, to our surprise, when GEC were pretreated with p38 inhibitors(PD-169316 and FR-167653, 10 µM each), complement-induced cytotoxicity quantified by specific BCECF release (see MATERIALS AND METHODS ) was significantly increased, compared with vehicle-treated cells [ Fig. 5 A : DMSO: NS 2.513 ± 3%, NS 2.75 24 ± 5%, NS 3 38 ± 5%, PD-169316: NS 2.520 ± 3% ( P P DMSO), NS 3 46 ± 4% ( P n = 3-4; Fig.5 B : DMSO: NS 2.5 18 ± 5%, NS 2.75 36 ± 7%, NS 3 49 ± 9%, FR-167653: NS 2.5 26 ± 7% ( P 0.05vs. DMSO), NS 2.75 45 ± 7%, NS 3 58 ± 8% ( P n = 3-4]. The concentration of FR-167653 used (10µM) inhibited complement-induced p38 activity by 90% in GEC (data notshown). Similar results were obtained when LDH release was used to quantifycytotoxicity [DMSO: NS 5 25 ± 9%, NS 7.5 36 ± 10%, FR-167653: NS5 29 ± 11%, NS 7.5 52 ± 10% ( P DMSO), n = 6]. These results suggest that p38 activation may protect cellsfrom complement-mediated injury.2 t( C4 j% y* a2 h2 n
% V: O) c! l) @# G- k lFig. 5. p38 Inhibitors augment complement-induced GEC injury. GEC were preincubatedwith p38 inhibitors PD-169316 ( A;10 µM) or FR-167653 ( B;10 µM) for 30 min at 37°C before stimulation with antibody andcomplement. Vehicle (DMSO) was used as control. Inhibitors were also includedin incubations with anti-GEC antiserum and HIS/NS. After 40 min of incubationwith HIS/NS, cytotoxicity was quantified by2',7'-bis(carboxyethyl)-5(6)-carboxyfluorescein (BCECF)-releaseassay as in MATERIALS AND METHODS. Numbers next to NS indicateconcentration of NS (vol/vol) used. * P n = 3-4 for each group.) }" F. t. [( v) y
6 R4 @- _( B( k$ E qWe and others previously reported that complement C5b-9 disrupts actinmicrofilaments in cultured GEC( 30, 33 ). One possible mechanismfor p38-mediated cytoprotection could be via modulation of the actincytoskeleton. Thus we next studied the impact of FR-167653 oncomplement-induced actin depolymerization. When GEC-cPLA 2 wereincubated with antibody and complement for 3 h, F-actin content decreased by 9± 3%. FR-167653 (10 µM) significantly augmented the reduction to 18± 4% ( P n = 5 each). Thus p38 activation,at least partly, prevents complement-induced actin depolymerization.9 N2 x) f+ t# O. ?& D
+ {7 U; x2 |2 h6 o! r
To further validate these results, we next generated subclones of GEC thatinducibly express a constitutively active mutant of TAK1, a kinase upstream ofp38 ( MATERIALS AND METHODS ). When one such clone( Fig. 6 A, #1 )was stimulated with an insect hormone, ponasterone A, expression of TAK1 wasinduced after 2 h, peaking at 6 h. Phosphorylation of p38 was also observed and peaked at 6 h ( Fig.6 A ). Ponasterone A has no known impacts on mammaliancells. When this inducible clone was stimulated with ponasterone A for 6 h andexposed to antibody and increasing concentrations of complement (NS),complement-mediated cytotoxicity was attenuated, compared with controls(incubated with ethanol in the place of ponasterone A). Conversely, when cellswere preincubated with the p38 inhibitor FR-167653 (10 µM, withoutstimulation with ponasterone A), cytotoxicity was augmented, compared withcontrol, consistent with the previous results ( Fig. 6 B ). Ofinterest, when cells were stimulated with ponasterone A in the presence ofFR-167653, the impact of ponasterone A (TAK1 induction) was neutralized, butcytotoxicity did not reach the level of FR-167653 treatment alone [control 47± 1%, ponasterone A 33 ± 3% ( P ponasterone A plus FR-167653 43 ± 3%, FR-167653 alone 57 ± 2%( P Fig.6 C ]. These results indicate that the cytoprotectiveeffect of TAK1 induction is, at least in part, mediated by p38 activation.However, there might be additional pathways downstream of TAK1, which contribute to cytoprotection.- z# d# B$ q7 {. {
8 u* N) c$ l+ nFig. 6. Induction of constitutively active transforming growthfactor- -activated kinase 1 (TAK1) ameliorates complement-induced GECinjury. A subclone of GEC that overexpresses TAK1 (constitutively activemutant) in an inducible manner was established as in MATERIALS AND METHODS. A : time course of TAK1 induction. GEC were incubatedwith ponasterone A (pona; 4 µM) for the indicated times and cell lysateswere analyzed for expression of TAK1 and phosphorylation of p38 byimmunoblotting using antibodies for HA (TAK1) and phospho-p38, respectively.In clone #1, induction of TAK1 peaked at 6 h when phosphorylation ofp38 was also observed. In clone #3, induction of TAK1 was weak andphosphorylation of p38 was not detected. B : clone #1 wasincubated with vehicle (control), ponasterone A (4 µM), or FR-167653 (10µM) for 6 h and stimulated with anti-GEC antiserum and NS (or HIS).Complement-mediated cytotoxicity was quantified as in MATERIALS AND METHODS using BCECF release into the medium. * P P = 0.07, *** P = 0.06 vs. control; n = 3 each. C : clone #1 was incubated with vehicle (control),ponasterone A (4 µM), ponasterone A (4 µM) FR-167653 (10 µM), orFR-167653 (10 µM) for 6 h and stimulated with anti-GEC antiserum and NS (orHIS 3.5% vol/vol). Cytotoxicity was quantified as in MATERIALS AND METHODS. P P n = 3 each. ( B and C : independent experiments.)
8 H/ ^# o% c, u f% R9 B! R1 A% s
/ }! A3 L/ _# F0 U) z7 rInhibition of p38 activation augments proteinuria in PHN. The above results suggest that p38 activation is cytoprotective for GEC againstcomplement-mediated cell injury in vitro. We next addressed whether p38activation is also cytoprotective in vivo using the PHN rat model ofmembranous nephropathy. In PHN, it is known that complement C5b-9 causes GECinjury, which leads to proteinuria( 21, 32 ). As shown in Fig. 2, p38 was significantlyactivated in glomeruli from rats with PHN, compared with control rats. Weanticipated that if p38 is cytoprotective for GEC, inhibition of p38 wouldlead to augmented proteinuria in PHN. Thus we examined the impact of aspecific p38 inhibitor, FR-167653, on proteinuria. First, to confirm theeffect of FR-167653, rats were treated with FR-167653 from day 7 to 14 ( MATERIALS AND METHODS ) and p38 activity in glomeruliwas studied on day 14. In FR-167653-treated rats, p38 activity inglomeruli was markedly inhibited to a level comparable to that of control rats[control 13 ± 1, PHN 80 ± 22 ( P 12 ± 5 (not significant from control), n =3-6; Fig. 7 A ].Phosphorylation of p38 was not affected by this inhibitor in a consistentmanner, in agreement with a previous report ( 27 )( Fig. 7 A ). In ratstreated with vehicle, urinary protein excretion on day 14 was 161± 33 mg/day ( n = 7), significantly higher than normal rats( 10 mg/day). Consistent with the in vitro cytoprotective effect of p38activation, rats treated with FR-167653 showed augmented proteinuria (288± 54 mg/day, P n = 9; Fig. 7 B ). To verify whether complement activation was influenced by FR-167653, we quantified C3deposition in glomeruli (see MATERIALS AND METHODS ). Glomerular C3deposition was 91 ± 4 in rats with PHN and 93 ± 3 in rats withPHN treated with FR-167653. Thus FR-167653 did not affect complementactivation in glomeruli. These results support a cytoprotective role for p38in complement-mediated GEC injury in vivo.' B7 w R- a& e6 @1 T! O% [* C7 x0 w
, I6 g, W* v5 N! s& x7 b7 oFig. 7. p38 inhibitor, FR-167653, augments proteinuria in PHN. PHN was induced by asingle injection of anti-Fx1A antiserum, and rats were treated with vehicle orFR-167653 as in MATERIALS AND METHODS. A : on day14, glomerular p38 activity was measured as in MATERIALS AND METHODS. Top : immunoblot. Bottom : densitometricanalysis. * P n = 3-6 rats ineach group. Data from densitometric analysis for control and PHN are sharedwith Fig. 2. B : on day 14, 24-h urine was collected in metabolic cages and urine proteinwas quantified. ** P n = 7 (vehicle),9 (FR-167653).
+ K- k9 P0 w' r* [/ o2 t1 O# d0 ^# t) U5 Y3 w
HSP27 overexpression protects GEC from injury. The above results indicate cytoprotective effects of the p38 MAPK pathway in GEC. HSP27 is oneof the molecules downstream of p38 MAPK. It is known that when the p38 MAPKpathway is activated, MAPKAPK-2 is phosphorylated/activated and in turnphosphorylates HSP27 ( 5 ). Itwas reported previously that in the rat puromycin aminonucleoside (PAN)nephrosis model, glomerular expression and phosphorylation of HSP27 wereincreased ( 24 ). Moreover,overexpression of HSP27 in mouse podocytes provided protection against PAN( 25 ). We also demonstratedthat glomerular expression of HSP27 is increased by 1.6-fold in rats withPHN, compared with control rats, similar to the increase seen in PAN nephrosis( 24 )( Fig. 8 A ). Whentreated with FR-167653, glomerular expression of HSP27 showed an upward trend,although the difference was not statistically significant [control 52 ±2, PHN 84 ± 10 ( P P n = 5-8 rats; Fig. 8 A ]. We also studied phosphorylation of HSP27 using phospho-HSP27-specific antibody.Phosphorylation of HSP27 in glomeruli was significantly increased in PHN( Fig. 8 B ). In contrastto protein expression, phosphorylation of HSP27 was markedly inhibited byFR-167653 [control 15 ± 2, PHN 64 ± 11 ( P control), FR-167653 34 ± 11, n = 6 rats each; Fig. 8 B ]. Theseresults suggest that glomerular phosphorylation of HSP27 in PHN is, at leastin part, mediated by p38 MAPK.
: ~/ i: m* `& S4 P' H% a' R4 J9 B K
Fig. 8. Glomerular expression of heat shock protein (HSP27) is increased in ratswith PHN. PHN was induced by a single injection of anti-Fx1A antiserum andrats were treated with vehicle or FR-167653 as in MATERIALS AND METHODS. On day 14, glomerular lysates were prepared andanalyzed by immunoblotting for HSP27 ( A ) and phospho-HSP27( B ). Top : immunoblot. Bottom : densitometricanalysis. * P n = 5-8 ratseach.6 p: r! L+ ]& ]
4 q* h* J' ~1 Q k1 e$ G
Thus we next studied whether the cytoprotective effect of the p38 MAPKpathway is mediated via MAPKAPK-2 and HSP27 in rat GEC in culture. To testthis hypothesis, we first studied if MAPKAPK-2 is phosphorylated by complementin GEC. When GEC were exposed to antibody and complement, phosphorylation of p38 was observed for 20-60 min ( Fig.9 A ). Phosphorylation of MAPKAPK-2 was observed in asimilar time course ( Fig.9 A ). We next studied the impact of overexpression ofHSP27 on complement-mediated GEC injury. The plasmid encoding the wild-typeHSP27 was stably transfected into GEC, and its expression was verified byimmunoblotting. Two subclones (HSP27WT1 and 2) that overexpressed thewild-type HSP27 were selected for further study( Fig. 9 B ). Control1 is a subclone of GEC that was transfected with vector alone and control 2 was transfected with the wild-type HSP27 but the expressionlevel was minimal. It should be noted that control cells also expressed HSP27when the film was exposed longer ( Fig.9 B ). When HSP27WT1 and 2 were simulated with antibody andcomplement, specific LDH release was markedly attenuated, compared with thetwo control cells [NS 2.5: control 1 30 ± 4%, control2 24 ± 3%, HSP27WT1 6 ± 1% ( P 0.001 vs. bothcontrols), HSP27WT2 7 ± 1% ( P vs. both controls),NS 5: control 1 56 ± 8%, control 2 48 ± 5%,HSP27WT1 10 ± 1% ( P both controls), HSP27WT2 16± 1% ( P controls), NS 10: control1 73 ± 6%, control 2 66 ± 2%, HSP27WT1 25 ±1% ( P HSP27WT2 30 ± 1%( P n = 6-12; Fig. 9 C ]. To study ifphosphorylation of HSP27 is important in cytoprotection, we next establishedsubclones of GEC that overexpress a non-phosphorylatable mutant of HSP27. Inthis mutant, two Ser residues that are known to be phosphorylated by MAPKAPK-2are mutated to Ala ( 10 ). Twosubclones (HSP27mut1 and 2) were chosen for further studies( Fig. 9 B ). Control 3 is a subclone of GEC that was transfected with vector alone, and control 4 was transfected with HSP27mut, but the expression level wasminimal. When HSP27mut1 and 2 were stimulated with complement, specific LDHrelease was slightly attenuated, compared with control cells, but theattenuation was much smaller than in cells overexpressing the wild-type HSP27[NS 2.5: control 3 25 ± 4%, control 4 29 ± 2%,HSP27mut1 19 ± 2%, HSP27mut2 31 ± 5%, NS 5: control 3 60 ± 3%, control 4 48 ± 2%, HSP27mut1 41 ± 2%( P 0.05 vs. control 3 ), HSP27mut2 48 ± 2%, NS10: control 3 66 ± 2%, control 4 62 ± 1%,HSP27mut1 53 ± 1% ( P ( P control 3 ), n =4-12; Fig. 9 D ].These results indicate that overexpression of HSP27 induces cytoprotection inGEC and suggest that MAPKAPK-2-mediated phosphorylation of HSP27 is importantin this cytoprotection.2 D: ]) O+ i; r _' @
+ g: w+ U2 Z9 I9 D
Fig. 9. Overexpression of the wild-type HSP27, but not a non-phosphorylatablemutant, protects GEC against complement-mediated cell injury. A :GEC-cPLA 2 were stimulated with antibody and complement for theindicated times. Cells lysates were analyzed by immunoblotting forphosphorylation of MAPKAPK-2 and p38. B : subclones of GEC that stablyoverexpress the wild-type hamster HSP27 (HSP27WT) or a non-phosphorylatablemutant (HSP27mut) were established as in MATERIALS AND METHODS.Expression of HSP27 was verified with immunoblotting. Note that control GECalso express HSP27 when exposed longer. C : GEC clones that do notoverexpress HSP27 ( cntl 1 and 2 ) and GEC clones thatoverexpress the wild-type HSP27 (HSP27WT1 and 2) were sensitized with anti-GECantiserum and stimulated with increasing concentrations of NS (or HIS) for 40min. Specific lactate dehydrogenase (LDH) release was quantified as in MATERIALS AND METHODS as marker of complement-mediatedcytotoxicity. * P cntl 1 and 2; n = 6-12 per group. D : GEC clones that do notoverexpress HSP27 ( cntl 1 and 2 ) and GEC clones thatoverexpress a non-phosphorylatable mutant of HSP27 (HSP27mut1 and 2) werestimulated with complement and LDH release was quantified. P cntl 3. P cntl 4; n = 4-12 per group.
' u# X6 E/ m b8 v. T+ Q
' w% m b0 q9 yDISCUSSION
" V @) N4 L" C( q6 W
& B( O K/ H$ O2 Z. V/ K! |Complement-mediated GEC injury plays an important role in proteinuria inthe rat model of membranous nephropathy. We and others reported previouslythat sublytic concentration of complement evokes a variety of signaling eventsin GEC including the activation of the ERK and JNK pathways( 3, 17, 31 ). The current study demonstrates that p38 MAPK is also activated by complement in GEC in vitro andin vivo. Similar to JNK activation, p38 activation was, at least in part,mediated by arachidonic acid released by cPLA 2 and ROS. Theactivation of the p38 MAPK pathway was cytoprotective againstcomplement-mediated injury.
0 i+ z7 w3 g( t/ q2 h# r6 m! j
3 M, U% `9 U6 ~/ Q& l! [% |5 [With the use of the p38 inhibitor FR-167653, Wada et al.( 35 ) elegantly demonstratedthat inhibition of p38 MAPK ameliorated histological changes and proteinuriain the nephrotoxic serum nephritis model. The same group also demonstratedthat FR-167653 ameliorated histological changes and prolonged survival of MRL-Fas lpr mice ( 6 ).The current results would appear to contradict these earlier reports. However,it should be noted that both nephrotoxic serum nephritis andMRL-Fas lpr mice are models in which glomerular infiltration byleukocytes has a major role. In contrast, infiltrating leukocytes have a minoror negligible role in PHN( 32 ). Because p38 inhibitionby FR-167653 inhibits production of inflammatory cytokines bymonocyte/macrophages ( 27 ), itis highly likely that the effects of FR-167653 observed in nephrotoxic serumnephritis and MRL-Fas lpr mice are mediated by inhibition ofmonocyte/macrophage activation. In fact, in these models, the amelioration ofhistological changes and proteinuria was paralleled by decreased expression ofMCP-1 and glomerular macrophage/lymphocyte infiltration( 6, 35 ). It is therefore possiblethat the cytoprotective role of p38 in GEC was unmasked in PHN becauseleukocyte activation has little role in this proteinuric model. Recently,Stambe et al. ( 26 ) confirmedthe protective effect of a p38 inhibitor (NPC-31145) in rat nephrotoxic serumnephritis. In this report, activated (phosphorylated) p38 was clearlylocalized in infiltrating neutrophils. Of interest, in glomeruli of normal ratand human, p38 was expressed in podocytes( 26 ). p38 in podocytes was phosphorylated even in the normal kidneys, and phosphorylation was furtheraugmented in nephrotoxic serum nephritis (rat) and postinfectiousglomerulonephritis (human)( 26 ). These results supportthe role of p38 in normal podocyte function as well as in podocyte injury inglomerular disease. In the current study, although it is not possible todetermine the site of p38 activation in vivo directly, it is reasonable to assume that GEC are the main site because in PHN, the injury is limited tothis cell type ( 32 ).* U; `/ u) I4 c' ~5 K8 g# _* r0 R
/ `3 }& H; r H. B# p' _- S
To date, four isoforms of p38 MAPK, i.e., p38, p38,p38, and p38, have been identified( 15 ). However, no systematicstudy has been conducted regarding which isoforms are expressed in glomerulior in GEC. The anti-p38 antibodies used in the current study as well as in thereport by Stambe et el. ( 26 )are specific for the p38 isoform, confirming the expression of thep38 isoform in GEC. In contrast, p38 is highly expressed in thebrain and heart and p38 is expressed primarily in skeletal muscle( 15 ); thus these two isoformsare not likely to be contributing to cytoprotection in GEC. In addition, theinhibitor we used in the current study (FR-167653) does not inhibit p38 or p38 ( 28 ); thus thecytoprotective effect we observed is not likely to be mediated by theseisoforms. Therefore, it is most likely that the p38 isoform isresponsible for the cytoprotective effect observed in GEC in the currentstudy.
8 [$ O3 F4 N; O3 C7 }) p7 s' |1 i
' X; f9 O" z2 L {2 ^+ ?Recently, the importance of the actin cytoskeleton in the structure andfunction of GEC has been highlighted( 8 ). We and others reportedthat complement induces disruption of the actin cytoskeleton ( 30, 33 ). Thus one possiblemechanism of the cytoprotective effects of p38 is via interaction with theactin cytoskeleton. Several cytoskeletal proteins are substrates for the p38pathway, including the microtubule-associated protein tau, theactin-associated protein HSP27, and the intermediate filament proteinsh-caldesmon, vimentin, and keratin polypeptides 8, 18( 15 ). Among them, HSP27 hasbeen most extensively studied. HSP27 is phosphorylated by MAPKAPK-2, which isphosphorylated/activated by p38 MAPK ( 5 ). Phosphorylated HSP27affords resistance to cells against an inhibitor of actin polymerization,cytochalasin D, in a p38-dependent manner( 5 ). Phosphorylation of HSP27also conferred thermal resistance by interacting with actin filaments( 11, 12 ). In the kidney, Smoyer etal. ( 24 ) demonstrated thatglomerular HSP27 was upregulated and phosphorylated in the rat model of PANnephrosis. Overexpression of the wild-type HSP27 in mouse podocytes (GEC)provided protection against PAN-induced microfilament disruption; however, therole of phosphorylation of HSP27 was not addressed in that study( 25 ). It was also reportedthat HSP27 limits injury in ATP-depleted renal epithelial cells by associatingwith the actin cytoskeleton( 34 ). In the current study, wedemonstrated that MAPKAPK-2 is phosphorylated by complement in a similar timecourse as p38 ( Fig.9 A ). Furthermore, GEC that overexpress the wild-typeHSP27 showed increased resistance to complement-mediated cell injury( Fig. 9 C ). However,this protective effect was markedly decreased when phosphorylation sites ofHSP27 were mutated ( Fig.9 D ). Our results are therefore in agreement with previousreports and further demonstrate that phosphorylation of HSP27 is indeed acritical mediator of cytoprotection in GEC. Other cytoskeletal proteinsdownstream of p38 may also contribute to cytoprotection. Further studies arerequired to clarify the role of the other cytoskeletal proteins incytoprotection in GEC.
5 l4 y+ c, b" J% b. e& f! ^) S) k0 q3 R9 h
In summary, p38 MAPK is activated by complement C5b-9 in GEC. Activation ofp38 leads to phosphorylation of MAPKAPK-2 and HSP27, which contributes tocytoprotection most likely via interaction with the actin cytoskeleton.
: x7 @, f5 v$ q& H( H: @' K$ w$ o$ R2 T& p3 u0 j
DISCLOSURES2 I$ x! n# T7 ~& |1 e
# x8 E2 Z! I2 Q& Q* T
This study was supported by grants from the Canadian Institutes of HealthResearch (T. Takano and S. Lemay) and the Kidney Foundation of Canada (T.Takano and S. Lemay). T. Takano and S. Lemay are recipients of a Scholarshipfrom the Canadian Institutes of Health Research. M. Stanciu is a recipient of a Research Bursary from the Faculty of Medicine, McGill University.! h9 T- i* x( U0 J7 J& i
, J5 D2 }9 r4 P+ x
ACKNOWLEDGMENTS
* I5 L4 J' `/ o
/ e" D, d+ Z/ BThe authors thank Fujisawa Pharmaceuticals for providing FR-167653 and Dr.H. Peng for assistance in antioxidant experiments.1 R0 X0 K4 ~7 P' D+ v$ c0 o
【参考文献】
7 L. P* k* O! l/ L. v5 |. h, n Boute N,Gribouval O, Roselli S, Benessy F, Lee H, Fuchshuber A, Dahan K, Gubler MC,Niaudet P, and Antignac C. NPHS2, encoding the glomerular protein podocin,is mutated in autosomal recessive steroid-resistant nephrotic syndrome. Nat Genet 24:349-354, 2000.
; d1 y4 H; t7 I% D
. n m3 [, ]& ^1 B8 @7 d5 A
' M; a; Y! G" D3 X+ ?; ]# Y. Z
$ @/ b& d: g; h7 C/ fCybulsky AV,Monge JC, Papillon J, and McTavish AJ. Complement C5b-9 activatescytosolic phospholipase A 2 in glomerular epithelial cells. Am J Physiol Renal Fluid Electrolyte Physiol 269: F739-F749,1995.
6 q2 v- Q1 f/ f+ K9 s Z3 }+ w0 B; t) s# M/ R
& C: ~) G W+ w6 h# u* _( {
) ^+ y, _$ x% ?2 p% tCybulsky AV,Papillon J, and McTavish AJ. Complement activates phospholipases andprotein kinases in glomerular epithelial cells. KidneyInt 54:360-372, 1998.; k+ L0 z Y6 b' Q$ r" O
4 [3 z( I* ~: I0 r
6 R& Y$ [: M) D+ B
2 O# D- d0 h3 a$ CCybulsky AV,Takano T, Papillon J, and McTavish AJ. Complement-induced phospholipaseA 2 activation in experimental membranous nephropathy. Kidney Int 57:1052-1062, 2000.
/ ?2 U4 b- |9 [3 l' t1 p# O
4 q0 f" ^9 B4 \* s8 n* T
* q2 a- G2 W2 J3 a% ]; W
; G) z# K% l# K4 ?Guay J, LambertH, Gingras-Breton G, Lavoie JN, Huot J, and Landry J. Regulation of actinfilament dynamics by p38 map kinase-mediated phosphorylation of heat shockprotein 27. J Cell Sci 110:357-368, 1997.9 o/ ?5 z. m, Q& m W
. V% i* k; r! [1 ~! y
* {3 Z* ?1 X; R( i n% g0 D, y% z3 E2 W6 ^0 m* e% l, Q1 |
Iwata Y, WadaT, Furuichi K, Sakai N, Matsushima K, Yokoyama H, and Kobayashi K. p38Mitogen-activated protein kinase contributes to autoimmune renal injury inMRL-Fas(lpr) mice. J Am Soc Nephrol 14: 57-67,2003.0 W/ s& f8 g p$ X# Y' l: q
2 P, P4 [. q, \. S+ X. f4 ~$ L% I/ A. s0 y: V C7 v
4 ?5 ?2 R `( \+ vKaplan JM, KimSH, North KN, Rennke H, Correia LA, Tong HQ, Mathis BJ, Rodriguez-Perez JC,Allen PG, Beggs AH, and Pollak MR. Mutations in ACTN4, encoding -actinin-4, cause familial focal segmental glomerulosclerosis. Nat Genet 24:251-256, 2000., p' L, Z- Z! ?( Q# _
* H: `$ r j [* ]2 L; s) T, d
2 Y' ]% ^1 ?8 R
9 P2 ~* r8 m$ x5 a% B, n: Q
Kerjaschki D. Caught flat-footed: podocyte damage and themolecular bases of focal glomerulosclerosis. J ClinInvest 108:1583-1587, 2001.2 s6 @. Q* i3 v' K
) ?7 N$ g' I$ K7 [$ V9 ]
4 |$ e) Q1 @2 A6 N1 L$ f
5 |% p. `, G" E$ M6 S' Y# V! BKestila M,Lenkkeri U, Mannikko M, Lamerdin J, McCready P, Putaala H, Ruotsalainen V,Morita T, Nissinen M, Herva R, Kashtan CE, Peltonen L, Holmberg C, Olsen A,and Tryggvason K. Positionally cloned gene for a novel glomerularprotein-nephrin-is mutated in congenital nephrotic syndrome. Mol Cell 1:575-582, 1998.2 Y m5 y. o T' L4 a) ?
' {, h7 {) t1 \* ^2 U/ v9 t- ]- f4 H4 N* |/ C. s
* c! ^6 x% Z7 r4 n/ m/ e/ X$ D6 aLambert H,Charette SJ, Bernier AF, Guimond A, and Landry J. HSP27 multimerizationmediated by phosphorylation-sensitive intermolecular interactions at the aminoterminus. J Biol Chem 274:9378-9385, 1999.
" A5 V, n7 B3 Y" W2 [# X
" M: t" s" F+ @0 a7 ^0 E, _7 V8 n$ w0 i" i! q
( t; a& l1 ?7 U, ~0 |% T4 cLavoie JN,Gingras-Breton G, Tanguay RM, and Landry J. Induction of Chinese hamsterHSP27 gene expression in mouse cells confers resistance to heat shock. HSP27stabilization of the microfilament organization. J BiolChem 268:3420-3429, 1993.; _* u$ h+ C$ r; j$ r6 r; X
6 |. A8 C+ U0 I9 h7 U3 l8 {) m, ]- q& E
( z# a* V9 j) ~% iLavoie JN,Hickey E, Weber LA, and Landry J. Modulation of actin microfilamentdynamics and fluid phase pinocytosis by phosphorylation of heat shock protein27. J Biol Chem 268:24210-24214, 1993.
. _% v! j) c, _6 d& l# F! P, d! ?: m! G$ a. p) |0 f4 e! E
1 m! [+ _( E" Z$ ?$ G$ p {) x/ a P# n- H: [
Mairesse N,Bernaert D, Del Bino G, Horman S, Mosselmans R, Robaye B, and Galand P. Expression of HSP27 results in increased sensitivity to tumor necrosis factor,etoposide, and H 2 O 2 in an oxidative stress-resistantcell line. J Cell Physiol 177:606-617, 1998.
0 s7 }" V0 s, C- }3 R2 W8 ?5 A( R: V, x/ F2 D
% A+ j6 b! w/ W! g. ^7 w
2 R1 ?8 u5 e" v) a1 }
Marshal DJ. Specificity of receptor tyrosine kinase signaling: transient versus sustainedextracellular signal-regulated kinase activation. Cell 80: 179-185,1995.
5 w2 d2 p& K# y* h7 Z. p, A
: h* b( x7 i- \5 b4 t8 f# L, e I: C" F }6 V0 @! i) {0 X+ Q
; t T5 X1 i. H9 h% F; TObata T, BrownGE, and Yaffe MB. MAP kinase pathways activated by stress: the p38 MAPKpathway. Crit Care Med 28:N67-N77, 2000.
. D; P7 m) z. P5 v6 B3 B* J2 |8 K1 k1 P
% g+ O7 k! G" l) w+ W* [# X; D
) S9 P4 _& W' o: \! iPanesar M,Papillon J, McTavish AJ, and Cybulsky AV. Activation of phospholipaseA 2 by complement C5b-9 in glomerular epithelial cells. JImmunol 159:3584-3594, 1997.9 E' Y, I; s' e; x& @6 |& @
& o' K' I$ H& q; x. q1 A9 |$ J
, k% w7 t( w7 m) _6 ]& s1 W9 y0 e" u$ `
Peng H, TakanoT, Papillon J, Bijian K, Khadir A, and Cybulsky AV. Complement activatesthe c-Jun N-terminal kinase/stress-activated protein kinase in glomerularepithelial cells. J Immunol 169: 2594-2601,2002.
- u9 [2 R3 y9 i$ N3 c, K/ F6 L9 q! e9 u: m) B0 T; R
# x: l) P8 a3 l: ~, a
- ~0 Z1 T) u: {# H3 ]* a
Pippin JW,Durvasula R, Petermann A, Hiromura K, Couser WG, and Shankland SJ. DNAdamage is a novel response to sublytic complement C5b-9-induced injury inpodocytes. J Clin Invest 111:877-885, 2003.
" f8 K, @7 @6 T
0 e) p8 e0 [. ]+ K/ ]. S/ `- Z7 h* _1 s! Z h* D( \. i
# C+ l5 G3 p7 ?% ~8 _
Quigg RJ,Cybulsky AV, Jacobs JB, and Salant DJ. Anti-Fx1A producescomplement-dependent cytotoxicity of glomerular epithelial cells. Kidney Int 34:43-52, 1988.7 ^! T5 p: ?6 A/ x
2 S5 C4 K" X5 W$ \0 ^0 x+ D1 D! b# v- |( A1 Z
8 I! l P! a- I$ ?- s. M: d' Z( u3 f
Salant DJ andCybulsky AV. Experimental glomerulonephritis. In: Methods inEnzymology, edited by DiSabato G. New York: Academic,1988, p. 421-461.
@* d ?# x% r$ I, f. V) s) z k3 a6 Q1 ]( M; R& b% E7 ]2 X
# q7 J$ o) T5 L0 Y
( I: W% o* L/ c4 O
Salant DJ,Quigg RJ, and Cybulsky AV. Heymann nephritis: mechanisms of renal injury. Kidney Int 35:976-984, 1989.
4 q' b9 e5 r& W; w
# X- a9 O7 f; j/ e) q. P' R; {" b. Y( e# i2 A
+ C/ M+ ~9 B# E7 q3 _Schneider GB,Hamano H, and Cooper LF. In vivo evaluation of hsp27 as an inhibitor ofactin polymerization: hsp27 limits actin stress fiber and focal adhesionformation after heat shock. J Cell Physiol 177: 575-584,1998.
8 ?* O% j k! E' k ]. n* M* [5 t8 `( u& P1 Q% U% `' l
/ ^4 u8 ` R" k8 u: e) S
" v/ @6 l8 |) W& k
Seger R andKrebs EG. The MAPK signaling cascade. FASEB J 9: 726-735,1995.. V7 Q$ ]: \) q6 B0 w
9 X, E4 S7 ]! q+ T x' K$ }
: J" ?4 }" T0 c3 r+ m! m* Q3 v4 s8 x
Smoyer WE,Gupta A, Mundel P, Ballew JD, and Welsh MJ. Altered expression ofglomerular heat shock protein 27 in experimental nephrotic syndrome. J Clin Invest 97:2697-2704, 1996.
2 g9 _$ Q3 m6 w6 u
, U# D4 x% g- m [
6 ?+ [# o4 E1 s! ?: k9 f) u$ K8 `8 g8 i9 ?" p8 v
Smoyer WE andRansom RF. Hsp27 regulates podocyte cytoskeletal changes in an in vitromodel of podocyte process retraction. FASEB J 16: 315-326,2002.
( H$ t, s7 i( k6 K* d8 k1 E; c
- {1 ?* Q& d( g3 g& W% [/ i) X
, O! `) d( d3 i% ], p. T5 v$ T" f+ ^0 `- P
Stambe C,Atkins RC, Tesch GH, Kapoun AM, Hill PA, Schreiner GF, and Nikolic-PatersonDJ. Blockade of p38 MAPK ameliorates acute inflammatory renalinjury in rat anti-GBM glomerulonephritis. J Am SocNephrol 14:338-351, 2003.
2 l" g. K7 R5 e, f6 c
3 u& K7 X# ^3 l' V* X: a$ h5 ^! ~4 v; u, Q
( _. v& H2 p! ~
Takahashi S,Keto Y, Fujita T, Uchiyama T, and Yamamoto A. FR167653, a p38mitogen-activated protein kinase inhibitor, prevents helicobacterpylori-induced gastritis in mongolian gerbils. J Pharmacol ExpTher 296:48-56, 2001.
! Q, _' r, h! R0 G+ I2 ~6 _3 ?$ q6 W H6 y: E/ p/ p
( [5 Y; z- y0 m& o) W5 G* ^
! o3 X+ {3 F5 C4 m
Takahashi Y,Kawahara F, Noguchi M, Miwa K, Sato H, Seiki M, Inoue H, Tanabe T, andYoshimoto T. Activation of matrix metalloproteinase-2 in human breastcancer cells overexpressing cyclooxygenase-1 or -2. FEBSLett 460:145-148, 1999.4 Y7 [/ h" l+ s5 X( |
6 x. l; Z1 D, R4 z$ l* }3 I2 c" k
/ r7 }% E" s# C" G$ f1 o! Q4 z4 ^# ]* T6 O: w# a
Takano T andCybulsky AV. Complement C5b-9-mediated arachidonic acid metabolism inglomerular epithelial cells: role of cyclooxygenase-1 and -2. Am JPathol 156:2091-2101, 2000.
7 x( u' B, A6 _) f& |; z
2 N; h) _4 @. g) g2 S3 L
# W' E3 T+ {4 o( i4 G( [' b1 v9 d6 \. m$ {2 ` \
Takano T,Cybulsky AV, Cupples WA, Ajikobi DO, Papillon J, and Aoudjit L. Inhibitionof cyclooxygenases reduces complement-induced glomerular epithelial cellinjury and proteinuria in passive Heymann nephritis. J PharmacolExp Ther 305:240-249, 2003.
1 U& l7 v9 z. B* _7 ?# o
+ |* b3 U6 o6 M# b3 A6 X/ w4 O' F: N* N; i
; r3 N% ?% ^& U' d/ _. J
Takano T,Cybulsky AV, Yang X, and Aoudjit L. Complement C5b-9 inducescyclooxygenase-2 gene transcription in glomerular epithelial cells. Am J Physiol Renal Physiol 281:F841-F850, 2001.% T q5 c$ p8 D( |
; \9 l: H3 D3 l6 [; n/ v+ ~% g- M% Y6 G, V* T
/ Y+ U5 C, z% r" u4 F0 j! fTischer DG andCouser WG. Milestones in nephrology. J Am SocNephrol 11:183-188, 2000.
- {& m* f) A9 E6 P9 y6 ^" |: c# f/ B+ d7 Y, E7 P
4 T6 _( K9 Y7 c, V" R6 a" C9 n
" o( _8 z0 ~7 ] I+ h5 I- ETopham PS,Haydar SA, Kuphal R, Lightfoot JD, and Salant DJ. Complement-mediatedinjury reversibly disrupts glomerular epithelial cell actin microfilaments andfocal adhesions. Kidney Int 55:1763-1775, 1999.; i! C8 }; Y' ^7 l
( o: N, R: m! X
$ t/ H1 N3 i* J% r" ~6 D* ]" `: z; e0 G d) R
Van Why SK,Mann AS, Ardito T, Thulin G, Ferris S, Macleod MA, Kashgarian M, and SiegelNJ. Hsp27 associates with actin and limits injury in energy depleted renalepithelia. J Am Soc Nephrol 14:98-106, 2003.; Y! `# N" g0 t- z$ p X
$ L/ y8 b2 z6 z. W
! f/ f8 [8 P* k& S* n- ^7 X* Z p2 [/ f. ~# q
Wada T,Furuichi K, Sakai N, Iwata Y, Yoshimoto K, Shimizu M, Kobayashi K, Mukaida N,Matsushima K, and Yokoyama H. A new anti-inflammatory compound, FR167653,ameliorates crescentic glomerulonephritis in Wistar-Kyoto rats. JAm Soc Nephrol 11:1534-1541, 2000.1 c5 z+ e. B. F" I6 [) ?; g
. M0 q; D* S' ]3 q9 a+ l' g
5 h- U% ^( i' y$ Z" S6 h
- V2 J1 w5 _0 g2 I7 z$ e# qYamaguchi K,Shirakabe K, Shibuya H, Irie K, Oishi I, Ueno N, Taniguchi T, Nishida E, andMatsumoto K. Identification of a member of the MAPKKK family as apotential mediator of TGF- signal transduction. Science 270:2008-2011, 1995. |
|
发表于 2009-4-21 13:45
