Possible Involvement of Heat Shock Protein 25in the Angiotensin II-Induced Glomerular

Mesangial Cell Contraction viap38 MAP Kinase

EVA MULLER,* ANKE BURGER-KENTISCHER, WOLFGANG NEUHOFER,MARIA-LUISA FRAEK, JOSEFINE MARZ, KLAUS THURAU, AND FRANZ-XAVER BECK

Institute of Physiology, University of Munich, Munich, Germany

In the rat kidney, mesangial cells (MCs), especially those in the extraglomerularmesangium (EGM) region of the juxtagomerular apparatus, express high amountsof heat shock protein 25 (HSP25). Because MCs are contractile in vivo and HSP25is known to modulate polymerization/depolymerization of F-actin and to beinvolved in smooth muscle contraction, it is possible that HSP25 participates inthe contraction process of MCs. We analyzed a permanent mouse MC line usingNorthern and Western blot analyses, and observed that similar to the MCs in theglomerulus, these cells also express high amounts of HSP25 constitutively. Ex-posure of these cells to angiotensin II (ANG II: 2 3 10-7 M) evoked contractionand a concomitant increase in HSP25 phosphorylation, while the cytoplasmicfraction of HSP25 was transiently reduced. Because phosphorylation of HSP25 isessential for its actin-modulating function, we suppressed the activity of p38 MAPkinase, the major upstream activator of HSP25 phosphorylation, with the specificinhibitor SB 203580. This maneuver reduced HSP25 phosphorylation dramati-cally, abolished cell contraction, and prevented the decrease of the cytoplasmicHSP25 content. This suggests that HSP25 might be a component of the contrac-tion machinery in MCs and that this process depends on p38 MAP kinase-mediated HSP25 phosphorylation. The decrease of cytoplasmic HSP25 contentobserved after ANG II exposure is probably the result of a transient redistributionof HSP25 into a buffer-insoluble fraction, because the whole cell content ofHSP25 did not change, a phenomenon known to be related to the actin-modu-lating activity of HSP25. The fact that this function requires phosphorylation ofHSP25 would explain the observation that HSP25 does not redistribute in SB203580-pretreated cells. J. Cell. Physiol. 181:462–469, 1999.© 1999 Wiley-Liss, Inc.

The small heat shock protein 25 (HSP25) is a mem-ber of a large protein family present in virtually allliving organisms. The expression of this HSP is regu-lated by a variety of physiological factors, such as hor-mones, cytokines, or in accordance to the state of dif-ferentiation of a cell, but it is also highly inducible by awide range of stressors such as heat, arsenite, oxi-dants, osmotic stress, etc. (for review see Ciocca et al.,1993; Kato et al., 1992). Several reports imply involve-ment of small HSPs in the assembly/disassembly ofactin filaments under physiological conditions. It colo-calizes with cytoskeletal elements in different celltypes (Leicht et al., 1986; Neuhofer et al., 1998), en-hanced amounts of HSP27 (designation for small HSPsin species other than mouse) are detectable in lamelli-podia of migrating cells (Lavoie et al., 1993b; Li et al.,1996) and there is evidence for the involvement ofHSP27 in the cytoskeletal rearrangements character-istic of platelet activation (Zhu et al., 1994). Further-more, HSP27 protects and stabilizes the actin micro-

filament organization, when cells are exposed tocytochalasin D, heat or oxidative stress (Huot et al.,1997; Lavoie et al., 1993a). An additional function sug-gested for small HSPs, probably also related to thiscapacity, is that HSP27 seems to be a mediator ofsustained smooth muscle cell contraction (Bitar et al.,1991; Larsen et al., 1997; Yamada et al., 1995). SmallHSPs are phosphoproteins (Landry et al., 1992) andphosphorylation is thought to regulate the diversefunctions of these proteins (Huot et al., 1996; Lavoie etal., 1993b; Minowada and Welch, 1995; Spector et al.,1993; Zhu et al., 1994). In cells expressing high

Contract grant sponsor: Deutsche Forschungsgemeinschaft; Con-tract grant number: Be 963/4–4.

*Correspondence to: Eva Muller, Institute of Physiology, Univer-sity of Munich, Pettenkoferstrasse 12, 80336 Munich, Germany.E-mail: eva.mueller@physiol.med.uni-muenchen.de

Received 28 October 1998; Accepted 29 June 1999

JOURNAL OF CELLULAR PHYSIOLOGY 181:462–469 (1999)

© 1999 WILEY-LISS, INC.

amounts of nonphosphorylatable HSP25 after transfec-tion and in cells in which HSP27 phosphorylation isinhibited, the actin cytoskeleton is much more sensi-tive to disruption by diverse stress factors (Lavoie etal., 1993b, 1995). Phosphorylation is also decisive forthe oligomeric size of this protein (Benndorf et al.,1994; Lavoie et al., 1995; Mehlen and Arrigo, 1994).The signal transduction pathway leading to HSP25/27phosphorylation includes p38 MAP kinase, which mostlikely is the exclusive upstream activator of MAPKAPkinase 2/3 (Cuenda et al., 1995; Freshney et al., 1994;Huot et al., 1997; Krump et al., 1997; Landry and Huot,1995), which in turn phosphorylates HSP25/27 (Stokoeet al., 1992).

In the rat kidney, HSP25 is detectable in glomerularmesangial cells (MCs) (Muller et al., 1996; Smoyer etal., 1996), with conspicuously high levels of HSP25 inthe MCs of the extraglomerular mesangium (EGM)region of the juxtaglomerular apparatus (Fig. 1A, B).The EGM is located within the space bordered by thetwo glomerular arterioles and the macula densa cells ofthe adjacent thick ascending limb of Henle’s loop(TAL). The MCs in the glomerular tuft form loopsaround the capillaries and have extensive interconnec-tions with the glomerular basement membrane (Inkyo-Hayasaka et al., 1996). MCs contain high amounts ofmicrofilaments, are generally considered to be smoothmuscle-like cells and express receptors for angiotensin

II (ANG II), which mediate MC contraction (Ausiello etal., 1980; Kriz et al., 1990; Venkatachalam and Kreis-berg, 1985). It has been suggested that the specificgeometric structure of MCs in the glomerulum, to-gether with their contractile capacity, allow these cellsto counter the capillary distension by isometric con-traction (Kriz et al., 1990, 1995). Through this mecha-nism, the capillary surface area could be regulated andthe glomerular filtration rate (GFR) maintained con-stant (Baylis and Brenner, 1978; Blantz et al., 1976;Iversen et al., 1992; Mayers et al., 1975). The MCspresent in the EGM appear to be morphologically sim-ilar to the intraglomerular MCs (Dunihue and Boldos-ser, 1963).

Cellular contraction includes the reorganization ofmicrofilaments and thus polymerization/depolymeriza-tion of actin (Goeckeler and Wysolmerski, 1995; Ma-chesky and Hall, 1997; Singhal et al., 1992). BecauseHSP25/27 participates in the regulation of actin fila-ment dynamics (Benndorf et al., 1994; Landry andHuot, 1995; Lavoie et al., 1993b; Miron et al., 1991) andbecause microinjection of antibodies against HSP27into mouse smooth muscle cells blocks sustained con-traction (Bitar et al., 1991), HSP25/27 appears to be anessential and integral component of the contractionmechanism. Thus, in the kidney, the high amounts ofHSP25 present in the MCs might be related to thecontractile machinery of these cells. To clarify this

Fig. 1. Localization of HSP25 mRNA (A) and protein (B) in extra-glomerular and intraglomerular mesangial cells (MCs) analyzed byimmunohistochemistry. EGM, extraglomerular mesangium region;TAL, thick ascending limb of Henle’s loop; arrowheads, intraglomeru-

lar MCs. C: HSP25 and b-actin levels in MCs, primary culturedmesangial cells (pMC), primary cultured papillary collecting duct cells(PCD), primary cultured renomedullary interstitial fibroblasts (RPF),and embryonic lung fibroblasts (3T3).

463HSP25 PARTICIPATES IN MC CONTRACTION

question we analyzed the HSP25 content of a perma-nent mouse MC line and showed that these cells ex-press constitutively high amounts of HSP25. Becausephosphorylation of HSP25/27 is essential for its actin-modulating function (Benndorf et al., 1994; Guay et al.,1997; Landry and Huot, 1995; Lavoie et al., 1993b; Zhuet al., 1994) and has been suggested to play a role insmooth muscle cell contraction (Gerthoffer et al., 1997;Larsen et al., 1997; Yamada et al., 1995;), we exposedMCs to an highly specific inhibitor (SB 203580) of p38MAP kinase (Cuenda et al., 1995) and analyzedwhether this pretreatment abolishes the ANG II-in-duced contraction of the permanent MCs.

MATERIALS AND METHODSAnimal preparation and immunohistochemistry

After anesthesia with sodium pentobarbital (Nembu-tal: 60 mg/kg body weight intraperitoneally) and endo-tracheal intubation, the kidneys were perfused in situfirst with phosphate-buffered saline (PBS) (pH 7.4,37°C) and then with periodate-lysine-paraformalde-hyde-fixative (2% paraformaldehyde, pH 7.4, 37°C)(McLean and Nakane, 1974). The first few milliliters ofperfusate contained 5000 U heparin and 40 mg pro-caine to prevent vasoconstriction and coagulation. Thekidneys were removed and sliced into 2-mm thick slicesand postfixed for 1 h at room temperature. After wash-ing in three changes of PBS, the tissue was equili-brated in 10% and then in 20% PBS-sucrose and thenrapidly frozen in propane/methylbutane (3:1 v/v),cooled in liquid nitrogen and stored at 280°C. Frozensections (10 mm) were cut using a cryostat (Leica In-struments, Nussloch, Germany), mounted on gelatin-coated slides and air dried. Immunodetection was car-ried out as described (Muller et al., 1996) by incubationof the slides with an antiserum specific for HSP25(SPA-801, stressGen, Victoria, Canada). After incuba-tion with the appropriate second antibody, immuncom-plexes were visualized using a strepavidin-biotin com-plex-kit (Dako, Hamburg, Germany). Sections weremounted with Kaiser’s glycerol-gelatin (Merck, Darm-stadt, Germany).

In situ hybridizationThe right kidneys were fixed in 3.7% paraformalde-

hyde and embedded in paraffin. Sections measuring 5mm were cut, mounted on siliconized slides, hydratedin decreasing concentrations of ethanol, and then pre-treated with 1 M NaSCN at 80°C and 0.01% proteinaseK in Tris-HCl, pH 7.5 at 37°C. The sections were post-fixed for 10 min in 4% paraformaldehyde in 0.1 Mphosphate buffer at 4°C, washed in Tris-HCl, and de-hydrated in increasing concentrations of ethanol. Thetissue sections and the probe were denatured at 80°Cand hybridized for 24 h at 37°C in a buffer (53% deion-ized formamide, 10% dextran sulphate, 10% Den-hardt’s solution and 2 3 SSC, pH 7.0) containing thedigoxigenin-labeled RNA probe specific for HSP25 (fi-nal concentration 1 ng/mL; generated as described be-low). After hybridization, the sections were immersedin 2 3 SSC, 20% formamide at 65 °C and 0.1 3 SSC,20% formamide at 45°C for 5 min each. After washingin Tris-HCl-buffer, the sections were incubated inblocking-buffer (Boehringer Mannheim, Mannheim,Germany) and then with an antidigoxigenin antibody-

alkaline phosphatase conjugate (dilution 1:500 in Tris-HCl-buffer, Boehringer) for 30 min at room tempera-ture. The color substrate for the alkaline phosphatasewas 4-nitroblue tetrazolium chloride (NBT) plus 5-bro-mo-4-chloro-3-indolyl-phosphate (BCIP).

Cell cultureA permanent mouse glomerular MC line (SV MES13

[MacKay et al., 1988]) was purchased from ATCC (CRL1927). MCs were grown on 75-mm plastic dishes(Greiner, Nurtingen, Germany) at 37°C in 5% CO2 inDulbecco’s modified Eagle’s Medium (Gibco, Paisly,UK) supplemented with 2.5% fetal calf serum (Gibco)plus 50 U/ml penicillin (Sigma, Deisenhofen, Germany)and 50 mg/ml streptomycin (Sigma). Experiments wereperformed when cells reached 80%–90% confluence.The cells grew with a doubling time of about 15 h.Primary cultured mesangial cells (pMCs), a gift fromDr. H. Schramek, were prepared and cultured as de-scribed elsewhere (Schramek et al., 1996). For experi-ments, cells from passage 13–17 were used. Prepara-tion and culture conditions of primary culturedpapillary collecting duct cells (PCD) and primary cul-tured renomedullary interstitial fibroblasts (RPF) aredescribed elsewhere (Burger-Kentischer et al., 1999).Mouse embryonic lung fibroblast were purchased fromATCC (BALB/3T3: ATCC CCL-163) and grown in Dul-becco’s modified Eagle’s medium supplemented with10% fetal calf serum plus antibiotics.

To induce MC contraction, cells were incubated with2 3 10-7 M ANG II (Sigma). Because ANG II is dis-solved in water, control cells were incubated with thesame volume of water without ANG II. The pyridineimidazole derivate, SB 203580 (50 mg/mL; Alexis,Laufelfingen, Switzerland), a highly specific inhibitorof p38 MAP kinase (Cuenda et al., 1995), was dissolvedin dimethyl sulfoxide (DMSO) and added to the me-dium to final concentrations of 5, 10, 17.5, and 25 mM.Control experiments were performed by adding thesame volumes of DMSO without SB 203580 to themedium.

The experimental protocols were as follows. For thefirst treatment protocol, MCs were exposed for 10, 20,and 30 min to H2O (control) or ANG II (2 3 10-7 M). Forthe second treatment protocol, MCs were exposed toDMSO (control) or SB 203580 (5, 10, 17.5, and 25 mM)for 2 h and then incubated with ANG II (2 3 10-7 M) for10, 20, and 30 min. At these times, phosphorylation ofHSP25 was determined using isoelectric focusing andthe HSP25 contents in the cytoplasmic protein fractionor the whole-cell protein fraction, respectively, byWestern blotting. To investigate contraction of MCs,the cells were pretreated as described above and cellu-lar reactions were recorded photographically after 30min exposure to ANG II.

Sample preparationFor isolation of cytoplasmic proteins cells were lysed

in a chilled solution (100 mL per culture dish) contain-ing (in mM): 20 tris(hydroxymethyl)aminomethane-HCl (Tris-HCl), 1 ethylenediaminetetraacetic acid(EDTA), 0.1 phenylmethylsulphonyl fluoride (PMSF),10 NaF (Sigma); 100 mM Pefabloc (Biomol, Hamburg,Germany), 1 mg/ml aprotinin (Fluka, Buchs, Germany),1 mg/mL leupeptin (Biomol), and 20% glycerol (pH 7.4)

464 MULLER ET AL.

using three cycles of thawing and freezing and 5 minsonication. After centrifugation at 10,000g for 5 min at4°C, the supernatant containing the buffer-soluble pro-tein fraction was kept for analysis (cytoplasmic pro-tein).

In parallel whole-cell protein was isolated by lysing2 3 106 cells in 100 mL Laemmli sample buffer (0.06 MTris-HCl, 100 mM dithiothreitol, 5% sodium dodecylsulfate (SDS), 10% glycerol; pH 6.8).

Sodium dodecyl sulfate polyacrylamide gelelectrophoresis and Western blot analysis

Analysis was carried out on 80 mg from the cytoplas-mic protein samples and 25 mL (equivalent to 0.5 3 106

cells) from the whole-cell protein samples. Westernblotting and immunodetection were performed as de-scribed elsewhere (Muller et al., 1996) using antiseraspecific for HSP25 (SPA-801; StressGen, Victoria, Can-ada) and b-actin (A-2066, Sigma) and, as a secondantibody, horseradish-peroxidase-conjugated antibody(goat anti-rabbit IgG; Dianova, Hamburg, Germany).The blots were quantified by laser densitometry (Ul-troscan XL; Pharamacia, Freiburg, Germany). b-Actinvalues were used to correct for variations in proteinloading after confirming that b-actin values were notinfluenced by any of the performed treatment proto-cols. In the figures, these values are shown as percent-age of the appropriate controls (100%).

Isoelectric focusing and Western blot analysisCytoplasmic protein samples were prepared as de-

scribed above. To each sample 2% 3,3-cholamidopropyl-dimethylammonio-1-propane-sulphonate (CHAPS), 5%b-mercaptoethanol, 1.5% ampholine 5–7 (Pharmacia),0.4% ampholine 3–10 (Pharmacia), 10 mM NaF, 1 mMPMSF, and 9 M urea were added. Isoelectric focusing(IEF) was carried out on rod gels following the methodof O’Farrell (O’Farrell, 1975). After separation, pro-teins were immunodetected by Western blot analysisas previously described (Muller et al., 1996) using theantiserum specific for HSP25 (SPA-801; StressGen).Immune complexes were revealed using the ECL de-tection system (Amersham, Braunschweig, Germany).From the fluorographs HSP25 isoforms were quantifiedby laser densitometry. For quantification, the ratio ofthe phosphorylated isoforms HSP25 B and C to thenonphosphorylated isoform A [HSP25 (B1C)/A] wascalculated. In the figures these values are shown aspercentage of the appropriate controls (100%).

Generation and labeling of the probeFor in situ hybridization a RNA probe was used to

detect HSP25 on tissue sections. To generate thedigoxigenin-labeled RNA probe a plasmid containing a167-bp fragment of the rat HSP25 cDNA (nucleotides33–200) in Bluescript II SK (Stratagene, La Jolla, CA)was generated as described elsewhere (Muller et al.,1998). This plasmid was linearized with BamHI (Sig-ma) and in vitro transcribed by T7 RNA-polymeraseusing a digoxigenin-RNA-labeling-mix (BoehringerMannheim) containing 0.35 mM digoxigenin-labeledUTP.

Quantification of cell shape changesCells (70 for each treatment condition) were photo-

graphed before and after treatment. Pictures werescanned (Hewlett Packard ScanJet IIc) and the areacovered by each cell was determined using appropriatesoftware (Adobe Photoshop). The values of cell shapechanges were calculated by comparing areas before (setat 100%) and after treatment.

Statistical analysisThe data are presented as means 6 SEM. Statistical

significance was assessed using Student’s t test forunpaired samples at P , 0.05.

RESULTSExpression of HSP25 in MCs in the rat kidney

and in different renal and nonrenal cellsIn the glomerulus HSP25 mRNA and protein (Fig.

1A, B) are detectable in MCs both in the extraglomeru-lar mesangial region (EGM) of the juxtaglomerularapparatus and within the glomerular tuft (arrow-heads). While in the EGM MCs represent a solid cellcomplex, which is not penetrated by blood vessels, inthe glomerular tuft MCs are distributed between thecapillaries.

The MCs used in the present study express also invitro conspicuously high levels of HSP25 (Fig. 1C, MC),a characteristic also noted in primary cultured MCsfrom the rat (pMC). In contrast, only low amounts ofHSP25 are present in other rat renal cells such asprimary cultured papillary collecting duct cells (PCD)or primary cultured renomedullary interstitial fibro-blasts (RPF) (both cultured in isotonic medium).HSP25 is not detectable in a nonrenal cell line fromembryonic lung fibroblasts (3T3).

Induction of MCs contraction by ANG IIUntreated MCs exhibit a star-shaped morphology,

are flattened and elongated and have long cytoplasmicextensions (Fig. 2A). Upon stimulation with ANG II for30 min, MCs show a contractile response (Fig. 2C),leading to an intensely refractile, shortened, androunded configuration with retraction of the cytoplas-mic extensions and reduction of cell size by 20% (Fig. 3,ANG II). To exclude possible adverse effects of the ANGII concentration used (2 3 10-7 M), cells were furthercultured after the ANG II exposure up to 48 h. After2 h, cells already returned to their normal shape. De-termining the growth characteristics of these cells 24and 48 h after ANG II exposure showed that the dou-bling time was comparable to that of control MCs (dt 515 h).

Effect of ANG II on cellular levels andphosphorylation of HSP25 in MCs

Exposure of MCs to ANG II for 10 or 20 min reducedthe content of cytoplasmic HSP25, while 30 min expo-sure to ANG II had no effect (Fig. 4A, non). None ofthese treatments had any influence on the whole-cellHSP25 content (Fig. 4B).

Concomitantly the phosphorylation of HSP25 wasincreased during ANG II exposure up to 20 min (Fig.5/2), leading to an increase of the two phosphorylatedisoforms HSP25 B and C compared to the nonphospho-

465HSP25 PARTICIPATES IN MC CONTRACTION

rylated isoform HSP25 A. To illustrate changes in thedegree of HSP25 phosphorylation, the ratio betweenthe phosphorylated isoforms B and C and the nonphos-

phorylated HSP25 A was calculated [HSP25 (B1C)/A](Fig. 6, non). While 10- and 20-min exposure to ANG IIevoked enhanced phosphorylation of HSP25, after 30-min treatment the degree of HSP25 phosphorylationdecreased and returned to control values.

Influence of SB 203580 on the effects inducedby ANG II in MCs

SB 203580 is a highly specific inhibitor of p38 MAPkinase (Cuenda et al., 1995). The p38 MAP kinase isprobably the exclusive activator of MAPKAP kinase-2/3 (Cuenda et al., 1995; Freshney et al., 1994; Huot etal., 1997; Krump et al., 1997), which in turn phosphor-ylates HSP25 (Stokoe et al., 1992). MCs were exposedto SB 203580 and the effects on HSP25 levels andHSP25 phosphorylation and the contractile responsewere investigated. A 2 h exposure to SB 203580 (25mM) changed neither MC morphology (Fig. 2B), norcytoplasmic HSP25 content (data not shown) but led toa reduction of the degree of HSP25 phosphorylation:the amount of the phosphorylated isoform B decreasedand the phosphorylated isoform C disappeared com-pletely (Fig. 5/3). In addition to the inhibitory effect ofSB 203580 on p38 MAP kinase, the above effect is mostlikely a consequence of dephosphorylation of the exist-ing phosphorylated HSP25. Comparable effects couldbe observed when cells were incubated with 5, 10, and17.5 mM SB 203580 (data not shown).

The SB 203580-pretreated cells (25 mM) were subse-quently exposed to ANG II for 10–30 min. In contrastto nonpretreated cells, ANG II now had no influence onthe cytoplasmic content of HSP25 (Fig. 4A, SB 203580),nor could it evoke phosphorylation activity in thesecells, so that levels of phosphorylated HSP25 remained

Fig. 2. Phase-contrast photomicrographs of nonpretreated MCs (A) or cells exposed to SB 203580 (25mM) for 2 h (B) followed by subsequent exposure to angiotensin II (ANG II) (2 3 10-7 M) for 30 min (C:nonpretreated; D: pretreated with SB 203580). Same cells are marked (A, C: *; B, D: v); Bar, 25 mm.

Fig. 3. Cells were exposed to angiotensin II (ANG II) (2 3 10-7 M) for30 min, to dimethyl sulfoxide (DMSO) (0.17 mL/ml) for 2 h, or to SB203580 (25 mM) for 2 h followed by ANG II (2 3 10-7 M) for 30 min(SB 1 ANG II). Cell areas were quantified by scanning photographsfrom the same cell before (control) and after treatment. Values areshown as a percentage of the control. Means 1 SEM; * significantlydifferent from control; n 5 70.

466 MULLER ET AL.

dramatically low (Fig. 6, SB 203580). Most interest-ingly, pretreatment with SB 203580 inhibited also thecontractile response to ANG II (Fig. 2D). Cell shapewas hardly altered and the value of the cell area re-mained comparable to control cells (Fig. 3, SB 1 ANGII). Moreover, also very low concentrations of SB203580 (5 mM), which still reduce HSP25 phosphoryla-tion, could inhibit the contractile response of MCs (datanot shown).

Exposure to the vehicle DMSO alone did not evokeany statistically significant changes in content or phos-phorylation pattern of HSP25, did not impair any of theANG II-induced effects (data not shown) and it did notalter cell shape (Fig. 3, DMSO).

DISCUSSIONThe present study implies involvement of HSP25 in

the contraction mechanism of MCs in the kidney. Ingeneral, cell contraction requires substantial reorgani-zation of the actin-cytoskeleton (Goeckeler andWysolmerski, 1995; Machesky and Hall, 1997; Singhalet al., 1992). HSP25 is thought to participate in thisprocess of reorganization as a regulatory component inthe following way: nonphosphorylated monomericHSP25 acts as an actin-capping protein, thus inhibit-ing polymerization of F-actin. This inhibitory capacityof HSP25 is abolished either by HSP25 phosphoryla-

Fig. 4. A: Cytoplasmic content of HSP25 in nonpretreated MCs (non)and in MCs pretreated with SB 203580 (25 mM) for 2 h, and subse-quently exposed to angiotensin II (ANG II) (2 3 10-7 M) for the timesindicated. B: Whole-cell content of HSP25 in mesangial cells (MCs)exposed to ANG II (2 3 10-7 M) for the times indicated. Values ofHSP25 are normalized to b-actin and are shown as a percentage of thecontrol (MCs exposed to the vehicles H2O or DMSO, respectively,without the addition of ANG II or SB 203580). Means 1 SEM; *significantly different from control; n 5 4.

Fig. 5. Phosphorylation pattern of HSP25 in control mesangial cells(MCs) (1), in nonpretreated MCs exposed to angiotensin II (ANG II)(2 3 10-7 M) for 20 min (2), and MCs exposed to SB 203580 (25 mM) for2 h (3). A: Nonphosphorylated isoform of HSP25; B, C: phosphory-lated isoforms of HSP25.

Fig. 6. Ratio of phosphorylated heat shock protein 25 (HSP25) B andC to the nonphosphorylated isoform HSP25 A [HSP25 (B 1 C)/A] innonpretreated mesangial cells (MCs) (non) and in MCs pretreatedwith SB 203580 (25 mM) for 2 h, and subsequently exposed to angio-tensin II (ANG II) (2 3 10-7 M) for the times indicated. Values areshown as a percentage of the control (MCs exposed to the vehicles H2Oand dimethyl sulfoxide (DMSO), respectively, without the addition ofANG II or SB 203580). Means 1 SEM; * significantly different fromcontrol; n 5 4.

467HSP25 PARTICIPATES IN MC CONTRACTION

tion, which leads to the release of HSP25 from thebarbed ends of microfilaments, thus facilitating theaddition of monomeres (Miron et al., 1991; Lavoie etal., 1993b), or by aggregation of HSP25 to large multi-meric particles (200–500 kD) (Behlke et al., 1991), thatare then incapable of exerting the capping function(Benndorf et al., 1994). In the present study, weshowed that a permanent MC line, which contains highamounts of actin filaments and contracts when exposedto ANG II, constitutively express high amounts ofHSP25. Furthermore, we observed after ANG II-in-duced contraction enhanced phosphorylation of HSP25and a decrease of the cytoplasmic but not of the wholecell HSP25 content. According to its actin-cappingfunction, which is abolished by phosphorylation, theenhanced phosphorylation of HSP25 is probably part ofan enhanced, contraction-related actin polymerizationactivity. The decrease in cytoplasmic HSP25 may wellreflect an intracellular redistribution of large multi-meric HSP25 aggregates from the cytoplasmic into thebuffer-insoluble fraction, an effect also consistent withactin polymerization. Both the increase in HSP25 phos-phorylation and the decrease of cytoplasmic HSP25content are attenuated with prolonged cellular contrac-tion, possibly due to the cells reaching maximum con-traction.

These data are the first suggestion that HSP25 par-ticipates in the contraction mechanism of MCs and thatphosphorylation plays an important role in this pro-cess. Investigations on other contractile cells havereached similar conclusions. In rectosigmoid smoothmuscle cells, bombesin-induced contraction was accom-panied by activation of MAP kinase and a concomitantcolocalization of this protein with HSP27 (Yamada etal., 1995). In the same cells, bombesin-induced contrac-tion could be inhibited by preincubation with monoclo-nal antibodies to HSP27 (Bitar et al., 1991). After car-bachol-induced contraction an intact airway smoothmuscle showed enhanced HSP27 phosphorylation. Thesignaling pathway leading to this reaction most likelyincludes p38 MAP kinase (Gerthoffer et al., 1997;Larsen et al., 1997). Similarly, thrombin-induced con-traction of vascular smooth muscle is associated withactivation of MAPKAP kinase 2 and the phosphoryla-tion of HSP27 (Brophy et al., 1998).

To characterize the functional significance of HSP27phosphorylation in the contraction mechanism of MCsfurther, we exposed the cells to the pyridine imidazoleSB203580. This substance has been shown to be ahighly specific inhibitor for p38 MAP kinase and doesnot inhibit 12 other protein kinases, including the MAPkinase homologues JNK and p42 MAP kinase (Cuendaet al., 1995). This maneuver led to a dramatic reductionof HSP25 phosphorylation (Figs. 5 and 6). In these MCsboth the ANG II-induced contractile response and theintracellular redistribution of HSP25 were completelyabolished. Given the above model, one might speculatethat the nonphosphorylated HSP25 will not be releasedfrom the ends of the actin filaments, thus inhibitingactin cytoskeleton modulation. The disappearance ofthe intracellular redistribution of HSP25 suggests thatformation of multimeric aggregates does not occur.This again would have a negative effect on the actinpolymerization but would further confirm the observa-tion of other investigators, that formation of such ag-

gregates is dependent on the phosphorylation ofHSP25/27 (Benndorf et al., 1994; Lavoie et al., 1995;Mehlen and Arrigo, 1994). From these data it can behypothesized that the observed failure of MC contrac-tion is a consequence of a deranged actin remodelingmechanism, which in turn is highly dependent on phos-phorylation of HSP25. Furthermore, an importantprerequisite for MC contraction seems to be high con-stitutive levels of phosphorylated HSP25. This suppo-sition can be made from the observation that in cellscapable of contraction the constitutive level of phos-phorylated HSP25 is rather high, compared with non-contracting SB 203580-pretreated MCs. In addition,the present results suggest that p38 MAP kinase isinvolved in the signal transduction pathway regulatingthe process of contraction probably as an upstreamactivator of HSP25 phosphorylation.

Taken together, these observations allow the hypoth-esis that phosphorylation of HSP25 is important for thecontraction of MCs, and that HSP25 may be a compo-nent of the contractile machinery of these cells. Such amechanism would explain the high amounts of HSP25in MCs in the glomerulus. However, to clarify finallywhether HSP25 is essential for MC contraction, exper-iments, including the complete downregulation ofHSP25, would be required.

ACKNOWLEDGMENTSThis study was supported by the Deutsche For-

schungsgemeinschaft (Be 963/4–4). We are indebted toDr. J. Davis for critically reviewing the manuscript.

LITERATURE CITEDAusiello DA, Kreisberg JJ, Roy C, Karnovsky MJ. 1980. Contraction

of cultured rat glomerular mesangial cells after stimulation withangiotensin II and arginine vasopressin. J Clin Invest 65:754–760.

Baylis C, Brenner BM. 1978. Modulation by prostaglandin synthesisinhibitors of the action of exogeneous angiotensin II on glomerularultrafiltration in the rat. Circ Res 43:889–898.

Behlke J, Lutsch G, Gaestel M, Bielka H. 1991. Supramolecularstructure of the recombinant murine small heat shock proteinhsp25. FEBS Lett 288:119–122.

Benndorf R, Hayess K, Ryazantsev S, Wieske M, Behlke J, Lutsch, G.1994. Phosphorylation and supramolecular organization of murinesmall heat shock protein HSP25 abolish its actin polymerization-inhibiting activity. J Biol Chem 269:20780–20784.

Bitar KN, Kaminski MS, Hailat N, Cease KB, Strahler JR. 1991.HSP27 is a mediator of sustained smooth muscle contraction inresponse to bombesin. Biochem Biophys Res Commun 181:1192–1200.

Blantz RC, Konnen KS, Tucker BJ. 1976. Angiotensin II effects uponthe glomerular microcirculation and ultrafiltration coefficient of therat. J Clin Invest 57:419–434.

Brophy CM, Woodrum D, Dickinson M, Beall A. 1998. Thrombinactivates MAPKAP2 kinase in vascular smooth muscle. J Vasc Surg27:963–969.

Burger-Kentischer A, Muller E, Marz J, Fraek ML, Thurau K, BeckFX. 1999. Hypertonicity-induced accumulation of organic osmolytesin papillary interstitial cells. Kidney Int 55:1417–1425.

Ciocca DR, Oesterreich S, Chamness GC, McGuire WL, Fuqua SA.1993. Biological and clinical implications of heat shock protein 27,000 (Hsp27): a review. J Natl Cancer Inst 85:1558–1570.

Cuenda A, Rouse J, Doza YN, Meier R, Cohen P, Gallagher TF, YoungPR, Lee JC. 1995. SB 203580 is a specific inhibitor of a MAP kinasehomologue which is stimulated by cellular stresses and interleu-kine-1. FEBS Lett 364:229–233.

Dunihue FW, Boldosser WG. 1963. Observations on the similarity ofmesangial to juxtaglomerular cells. Lab Invest 12:1228–1240.

Freshney NW, Rawlinson L, Guesdon F, Jones E, Cowley S, Hsuan J,Saklatvala J. 1994. Interleukin-1 activates a novel protein kinasecascade that results in the phosphorylation of Hsp27. Cell 78:1039–1049.

468 MULLER ET AL.

Gerthoffer WT, Yamboliev IA, Pohl J, Haynes R, Dang S, McHugh J.1997. Activation of MAP kinases in airway smooth muscle. Am JPhysiol 272:L244–L252.

Goeckeler ZM, Wysolmerski RB. 1995. Myosin light chain kinase-regulated endothelial cell contraction: the relationship between iso-metric tension, actin polymerisation, and myosin phosphorylation.J Cell Biol 130:613–627.

Guay J, Lambert H, Gingras-Brenton G, Lavoie JN, Huot J, Landry J.1997. Regulation of actin filament dynamics by p38 map kinase-mediated phosphorylation of heat shock protein 27. J Cell Sci 110:357–368.

Huot J, Houle F, Spitz DR, Landry J. 1996. HSP27 phosphorylation-mediated resistance against actin fragmentation and cell deathinduced by oxidative stress. Cancer Res 56:273–279.

Huot J, Houle F, Marceau F, Landry J. 1997. Oxidative stress-inducedactin reorganization mediated by the p38 mitogen-activated proteinkinase/heat shock protein 27 pathway in vascular endothelial cells.Circ Res 80:383–392.

Inkyo-Hayasaka K, Sakai T, Kobayashi N, Shirato I, Tomino Y. 1996.Three-dimensional analysis of the whole mesangium in the rat.Kidney Int 50:672–683.

Iversen BM, Kvam FI, Matre K, Morkrid L, Horvel G, Bagchus W,Grond J, Ofstad J. 1992. Effect of mesangiolysis on autoregulationof renal blood flow an glomerular filtration rate in rats. Am JPhysiol 262:F361–F366.

Kato K, Brijlall D, Adler SA, Kato S, Herz F. 1992. Effect of hyperos-molality on alkaline phosphatase and stress-response protein 27 ofMCF-7 breast cancer cells. Breast Cancer Res Treat 23:241–249.

Kriz W, Elger M, Lemley K, Sakai T. 1990. Structure of the glomer-ular mesangium: a biomechanical interpretation. Kidney Int 38:S2–S9.

Kriz W, Elger M, Mundel P, Lemley KV. 1995. Structure-stabilizingforces in the glomerular tuft. J Am Soc Nephrol 5:1731–1739.

Krump E, Sanghera JS, Pelech SL, Furuya W, Grinstein S. 1997.Chemotactic peptide N-formyl-Met-Leu-Phe activation of p38 mito-gen-activated protein kinase (MAPK) and MAPK-activated proteinkinase-2 in human neutrophils. J Biol Chem 272:937–944.

Landry J, Lambert H, Zhou M, Lavoie JN, Hickey E, Weber LA,Anderson CW. 1992. Human HSP27 is phosphorylated at serines 78and 82 by heat shock and mitogen-activated kinases that recognizethe same amino acid motif as S6 kinase II. J Biol Chem 267:794–803.

Landry J, Huot J. 1995. Modulation of actin dynamics during stressand physiological stimulation by a signaling pathway involving p38MAP kinase and heat-shock protein 27. Biochem. Cell Biol 73:703–707.

Larsen JK, Yamboliev IA, Weber LA, Gerthoffer WT. 1997. Phosphor-ylation of the 27-kDa heat shock protein via p38 MAP kinase andMAPKAP kinase in smooth muscle. Am J Physiol 273:L930–L940.

Lavoie JN, Gingras-Brenton G, Tanguay RM, Landry J. 1993a. In-duction of chinese hamster HSP27 gene expression in mouse cellsconfers resistance to heat shock. J Biol Chem 268:3420–3429.

Lavoie JN, Hickey E, Weber LA, Landry J. 1993b. Modulation of actinmicrofilament dynamics and fluid phase pinocytosis by phosphory-lation of heat shock protein 27. J Biol Chem 268:24210–24214.

Lavoie JN, Lambert H, Hickey E, Weber LA. 1995. Modulation ofcellular thermoresistance and actin filament stability accompaniesphosphorylation-induced changes in the oligomeric structure ofheat shock protein 27. Mol Cell Biol 15:505–516.

Leicht BG, Biessmann H, Palter KB, Bonner JJ. 1986. Small heatshock proteins of Drosophila associates with the cytoskeleton. ProcNatl Acad Sci USA 83:90–94.

Li S, Piotrowicz RS, Levin EG, Shyy YJ, Chien S. 1996. Fluid shearstress induces the phosphorylation of small heat shock proteins invascular endothelial cells. Am J Physiol 271:C994–C1000.

Machesky LM, Hall A. 1997. Role of actin polymerisation and adhe-sion to extracellular matrix in Rac- and Rho-induced cytoskeletalreorganisation. J Cell Biol 138:913–926.

MacKay K, Striker LJ, Elliot S, Pinkert CA, Brinster RL, Striker GE.1988. Glomerular epithelial, mesangial, and endothelial cell linesfrom transgenic mice. Kidney Int 33:677–684.

Mayers BD, Deen WM, Brenner BM. 1975. Effects of norepinephrineand angiotensin II on the determinants of glomerular ultrafiltrationand proximal tubule fluid reabsorption in the rat. Circ Res 37:101–110.

McLean IW, Nakane PK. 1974. Periodate-lysine-paraformaldehydefixative: a new fixative for immunoelectron microscopy. J Histo-chem Cytochem 22:1077–1083.

Mehlen P, Arrigo AP. 1994. The serum-induced phosphorylation ofmammalian hsp27 correlates with changes in its intracellular lo-calization and levels of oligomerisation. Eur J Biochem 221:327–334.

Minowada G, Welch W. 1995. Variation in the expression and/orphosphorylation of the human low molecular weight stress proteinduring in vitro cell differentiation. J Biol Chem 270:7047–7054.

Miron T, Vancompernolle K, Vandekerckhove J, Wilchek M, Geiger B.1991. A 25-kD inhibitor of actin polymerization is a low molecularmass heat shock protein. J. Cell Biol 114:255–261.

Muller E, Neuhofer W, Ohno A, Rucker S, Thurau K, Beck FX. 1996.Heat shock proteins HSP25, HSP60, HSP72, HSP73 in isoosmoticcortex and hyperosmotic medulla of rat kidney. Pflugers Arch 431:608–617.

Muller E, Neuhofer W, Burger-Kentischer A, Ohno A, Thurau K, BeckFX. 1998. Effects of long-term changes in medullary osmolality onheat shock proteins HSP25, HSP60, HSP72 and HSP73 in the ratkidney. Pflugers Arch 435:705–712.

Neuhofer W, Muller E, Burger-Kentischer A, Beck FX. 1998. Hyper-tonicity affects heat shock protein 27 and F-actin localization inMadin-Darby canine kidney cells. Kidney Int 54:S165–S167.

O’Farrell PH. 1975. High resolution two-dimensional electrophoresisof proteins. J Biol Chem 250:4007–4021.

Schramek H, Schumacher M, Pfaller W. 1996. Sustained ERK-2 ac-tivation in rat glomerular mesangial cells: differential regulation byprotein phosphatases. Am J Physiol 271:F423–F432.

Singhal PC, Franki N, Gibbons N, Hays R. 1992. Effects of angioten-sin II and arginine vasopressin on F-actin content of cultured mes-angial cells. J Am Soc Nephrol 3:80–87.

Smoyer WE, Gupta A, Mundel P, Ballew JD, Welsh MJ. 1996. Alteredexpression of glomerular heat shock protein 27 in experimentalnephrotic syndrome. J Clin Invest 97:2697–2704.

Spector NL, Ryan C, Samson W, Levine H, Nadler LM, Arrigo AP.1993. Heat shock protein is a unique marker of growth arrestduring macrophage differentiation of HS60 cells. J Cell Physiol156:619–625.

Stokoe D, Engel K, Campbell DG, Cohen P, Gaestel M. 1992. Identi-fication of MAPKAP kinase 2 as a major enzyme responsible for thephosphorylation of the small mammalian heat shock proteins.FEBS Lett 313:307–313.

Venkatachalam MA, Kreisberg JI. 1985. Agonist-induced isotonic con-traction of cultured mesangial cells after multiple passage. Am JPhysiol 249:C48–C55.

Yamada H, Strahler J, Welsh MJ, Bitar KN. 1995. Activation of MAPkinase and translocation with HSP27 in bombesin-induced contrac-tion of rectosigmoid smooth muscle. Am J Physiol 269:G683–G691.

Zhu Y, O’Neill S, Saklatvala J, Tassi L, Mendelsohn E. 1994. Phos-phorylated HSP27 associates with the activation-dependent cy-toskeleton in human platelets. Blood 84:3715–3723.

469HSP25 PARTICIPATES IN MC CONTRACTION