1 -Integrins in the primary cilium of MDCK cells potentiate fibronectin-induced - 《American Journal of Physiology》美国生理学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 25 下一页

返回列表

查看: 365031\|回复: 245	go 1 -Integrins in the primary cilium of MDCK cells potentiate fibronectin-induced [复制链接]

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

楼主

发表于 2009-4-22 08:08 |只看该作者 |倒序浏览 |打印

作者：H. A. Praetorius,, J. Praetorius,, S. Nielsen,, J. Frokiaer,, and K. R. Spring作者单位：1 The Water and Salt Research Center, 2 Clinical Institute and 3 Institute of Anatomy, University of Aarhus, DK-8200 Aarhus, Denmark; and 4 Laboratory of Kidney and Electrolyte Metabolism, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892-1603

   & |+ C2 Q. w+ D% C) b/ t
      ! M, @' \- I, X7 D3 d5 g
                     2 q7 G/ _0 U/ |; n! [

         ! f) ?: G4 N6 {* w" H

         * i* j6 x( Q# ?& B, b. F2 K4 o



      6 |! D' G/ y, O
      【摘要】! I$ X3 F1 }- n" o) G& l( F
   Because 1 -integrin is involved in sensing of fluid flow rate in endothelial cells, a function that in Madin-Darby canine kidney (MDCK) cells is confined to the primary cilium, we hypothesized 1 -integrin to be an important part of the primary ciliary mechanosensory apparatus in MDCK cells. We observed that 1 -integrin, 3 -integrin, and perhaps 5 -integrin were localized to the primary cilium of MDCK cells by combining lectin and immunofluorescence confocal microscopy. 1 -Integrin was also colocalized with tubulin to the primary cilia of the rat renal collecting ducts, as well as to the cilia of proximal tubules and thick ascending limbs. Immunogold-electron microscopy confirmed the presence of 1 -integrin on primary cilia of MDCK cells and rat collecting ducts. Intracellular Ca 2  levels, monitored by fluorescence microscopy on fluo 4-loaded MDCK cells, significantly increased on addition of fibronectin, a 1 -integrin ligand, to mature MDCK cells with an IC 50 of 0.02 mg/l. In immature, nonciliated cells or in deciliated mature cells, the IC 50 was 0.40 mg/l. Blocking the fibronectin-binding sites of 1 -integrin with RGD peptide prevented the Ca 2  signal. Cross-linking of 1 -integrins by Sambucus nigra agglutinin produced a Ca 2  response similar to the addition of fibronectin. Furthermore, the fibronectin-induced response was not dependent on flow or a flow-induced Ca 2  response. Finally, the flow-induced Ca 2  response was not prevented by the fibronectin-induced signal. Although 1 -integrin on the primary cilium greatly potentiates the fibronectin-induced Ca 2  signaling in MDCK cells, the flow-dependent Ca 2  signal is not mediated through activation of 1 -integrin. 6 {, T- o+ j: @+ J/ a5 p
      【关键词】 intracellular calcium ion antibody MadinDarby canine kidney cells fibronectin fluorescence imaging
   INTEGRINS FUNCTION BOTH AS adhesion proteins by binding extracellular matrix proteins and adjacent cells ( 11 ) and as receptors on the cell surface mediating signal transduction both from the exterior of the cell to the cytoplasm and from the interior to the environment surrounding the cells ( 23 ). Integrins are heterodimers, composed of an - and a -chain, which are noncovalently associated ( 5 ). As cell surface receptors, the integrins bind a variety of natural agonists. The primary event after binding of a ligand to integrin is a rise in intracellular Ca 2  concentration ([Ca 2  ] i ) in neutrophils ( 6 ), lymphocytes ( 13 ), and osteoclasts ( 32 ). After the initial Ca 2  signal, an activation of tyrosine kinases follows, and the Ras/microtubule-associated protein kinase (MAPK) pathway is activated ( 4 ).

Integrins also serve as a mechanical link between the plasma membrane and the cytoskeleton. In fibroblasts, stress fibers converge at focal contacts with actin-binding proteins as vinculin, -actin, and talin. Some actin-binding proteins are associated with integrins in fibroblasts ( 3 ). The actin filaments have subcortical localization ( 9 ). In epithelia, the integrins are generally expressed along cell junctions, where they are responsible for adhesion and organization of the subcortical actin cytoskeleton ( 8, 9 ).' s1 ~! k7 ^' t8 p) g! @5 r5 c
% S: n# ~$ ^0 a2 L& n6 g3 P2 l
The ligands for the 3 1 -integrins (laminin, collagen, and fibronectin) and its classically basolateral distribution have supported their role as receptors for the basement membrane ( 7 ). It was recently shown, however, that 1 -integrin is present on the apical membrane of mature Madin-Darby canine kidney (MDCK) cells and in the rat collecting duct ( 19 ).

Integrins have been proposed to serve as mechanosensors ( 25 ), and recently it was suggested that the 5, 1 -integrin acts as a mechanoreceptor in human chondrocytes ( 22 ). Furthermore, it has been established that integrins contribute to the perception of shear stress ( 1 ). Interestingly, agonist stimulation of the integrins is known to induce a Ca 2  signal in MDCK cells, thereby sharing a second-messenger pathway with the flow response ( 26 ). 1 -Integrin on the apical membrane is unlikely to be directly responsible for flow sensing, since it is highly expressed in nonconfluent MDCK cells ( 19 ), which are unresponsive to changes in flow rate of the perfusate ( 15 ).

The present study was undertaken to evaluate the role of integrins in the acute signal transduction of mechanosensing by combining lectin- and immunofluorescence microscopy with pharmacological characterization of the integrin-dependent Ca 2  signaling in MDCK cells.5 L+ f4 w: ^% E& @% [
. I0 ~/ o, b: X
METHODS

Solutions and materials. The perfusion solution had the following composition (in mM): 137 [Na ], 5.3 [K ], 1.8 [Ca 2  ], 0.8 [Mg 2  ], 126.9 [Cl - ], 0.8 , 14 HEPES, 5.6 glucose, and 5 probenecid, pH 7.4 (37°C, 300 mosmol/kgH 2 O; brackets denote concentration). Sources of chemicals and antibodies were as follows: fluo 4-AM, anti-bovine -tubulin mouse monoclonal antibody, Alexa-conjugated anti-mouse IgG (Molecular Probes, Eugene, OR), EGTA, probenecid, neuraminidase, and Glu-Arg-Gly-Asp-Ser-Pro-Lys-peptide (Sigma, St. Louis, MO), anti- 1 -integrin (MAB2000, AB1952), anti- 3 -integrin antibody (AB1920), and anti- 5 -integrin antibody (AB1949; Chemicon International, Temecula, CA). Fluorescence-tagged peanut and Sambucus nigra agglutinin (SNA) were obtained from Vector Laboratories (Burlingame, CA).6 W# @! I# ^8 K6 |2 o: \

Cell culture and laboratory animals. Wild-type MDCK cells (passages 54-70 from the American Type Culture Collection, Rockville, MD) were grown to confluence on 25-mm-diameter coverslips in DMEM with 10% FBS (GIBCO, Grand Island, NY) and 2 mM glutamine, but without riboflavin or antibiotics, as previously described ( 15 ). Adult male Munich-Wistar rats (250-300 g; Møllegaard Breeding Centre, Lille Skensved, Denmark) had free access to water and pelleted food (Altromin, Lage, Germany) until killed for immunohistochemistry.

Lectin labeling and immunocytochemistry. MDCK cells, grown on glass coverslips for 4-8 days, were washed two times in PBS. For lectin labeling, cells were incubated for 5 min with 10 mg/l fluorescein, rhodamine-conjugated SNA, and/or peanut agglutinin (PNA) and then washed before fixation, as described below. For antibody staining, cells were fixed for 15 min in 2.5% formaldehyde at room temperature. Next, the cells were washed two times and permeabilized with 0.3% Triton X-100 PBS containing bovine serum (15 g/l) for 15 min and incubated overnight with the primary antibody at 4°C. After being washed, the sections were incubated with Alexa 488-conjugated goat anti-rabbit and/or Alexa 543 goat anti-mouse secondary antibodies (Molecular Probes) in PBS supplemented with BSA and Triton X-100. TO-PRO-3 (Molecular Probes) was used for nuclear counterstaining in some experiments. After being washed, sections were mounted with a cover slip in Glycergel Antifade Medium (Dako) and inspected on a Leica DMRS confocal microscope using an HCX PlApo x 64 [1.32 numeric aperture (NA)] objective. The immunofluorescence images were merged with differential interference contrast (DIC) images to reveal the spatial relationship between the tissue structures and the fluorescence labeling.% R: x) w- K  p" U1 R# y
" y# D3 L8 @( t/ w: o
Immunohistochemistry. The kidneys of halothane-anesthetized rats were fixed by perfusion via the abdominal aorta with 4% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.4. The tissue was dehydrated and embedded in paraffin, and 2-µm sections were cut using a rotary microtome (Leica, Heidelberg, Germany). The sections were dewaxed and rehydrated, and endogenous peroxidase was blocked by 0.5% H 2 O 2 in absolute methanol. They were then boiled in 10 mM Tris, pH 9, supplemented with 0.5 mM EGTA, and then incubated with 50 mM NH 4 Cl and blocked in PBS supplemented with 1% BSA, 0.05% saponin, and 0.2% gelatin. The sections were incubated overnight at 4°C with the primary antibodies diluted in PBS supplemented with 0.1% BSA and 0.3% Triton X-100.% j1 [' S4 A, F0 q
% A9 A) i$ w5 F( d6 x. B, L& N1 A
After being washed, the sections were incubated with Alexa 488-conjugated goat anti-rabbit secondary antibodies (Molecular Probes) in PBS supplemented with BSA and Triton X-100. After being washed, sections were mounted with a coverslip in Glycergel Antifade Medium (Dako).

Immunoblotting. The protein contents of rat renal or MDCK cell homogenates were cleared of nuclei and unbroken cells by 4,500- g centrifugation and adjusted to 1.5% (wt/vol) SDS, 40.0 mM 1,4-dithiothreitol, 6% (vol/vol) glycerol, and 10 mM Tris, pH 6.8, with bromphenol blue. Protein samples of 2-10 µg were separated by 9% PAGE and electrotransferred to nitrocellulose membranes, which were blocked in 5% nonfat dry milk in a PBS solution (PBS-T: 80 mM Na 2 HPO 4, 20 mM NaH 2 PO 4, 100 mM NaCl, pH 7.5, and 0.1% vol/vol Tween 20). The membranes were incubated overnight at 5°C with primary antibody in PBS-T. After being washed, the membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody (Dako, Glostrup, Denmark) for 2 h in PBS-T, and bound antibody was detected by the ECL chemiluminiscence kit (Amersham, Little Chalfont, UK).

Immunogold electron microscopy. Tissue blocks prepared from rat kidney cortex were cryoprotected with 2.3 M sucrose containing 2% paraformaldehyde and rapidly frozen in liquid nitrogen. The samples were freeze-substituted by sequential equilibration over 3 days in methanol containing 0.5% uranyl acetate at temperatures raised gradually from -80 to -70°C, rinsed in pure methanol for 24 h while the temperature was increased from -70 to -45°C, and infiltrated with Lowacryl HM20 and 1:1 and 2:1 methanol and, finally, pure Lowacryl HM20 before ultraviolet polymerization for 2 days at -45°C and 2 days at 0°C. Immunolabeling was performed on ultrathin Lowacryl HM20 sections. Sections were pretreated with a saturated solution of NaOH in absolute ethanol (2-3 s), rinsed, and preincubated for 10 min with 0.1% sodium borohydride and 50 mM glycine in 0.05 M Tris, pH 7.4, containing 0.1% Triton X-100. Sections were rinsed and incubated overnight at 4°C with the respective antibody diluted in 0.05 M Tris, pH 7.4, containing 0.1% Triton X-100 with 0.2% milk. After being rinsed, sections were incubated for 1 h at room temperature with goat anti-rabbit IgG conjugated to 10-nm colloidal gold particles (GAR. EM10; BioCell Research Laboratories, Cardiff, UK). The sections were stained with uranyl acetate and lead citrate before examination in a Philips Morgagni 268D electron microscope.
( y# S9 k! S6 f( V! U# k6 ^
Live cell microscopy and perfusion. MDCK cell monolayers, grown on cover glasses, were viewed in a perfusion chamber at 37°C on the stage of an inverted microscope (Nikon TE-2000; BBT-LifeScience) equipped with DIC combined with low-light-level fluorescence provided via a Xenon lamp and monochromator (Visitech International, Sunderland, UK). Imaging was performed with a x 100/1.4 NA Plan Apo lens (Nikon), an intensified SVGA CCD camera, and imaging software (Quanticell 2000/Image Pro; VisiTech). Cellular fluorescence images were sampled at rate of 0.5 Hz, and measurements were initiated 50 s before the increases in perfusion rate from 2 to 12 µl/s. The flow chamber was 18 mm long, 6.3 mm wide, and 2.3 mm high and thus had a volume of 263 µl (RC-21BRFS; Warner Instruments, Hamden, CT). Flow rates were calibrated by measurement of the efflux in a reservoir of known volume; a rate of 1 µl/s corresponds to a bulk flow rate of 68 µm/s and a linear velocity just above the membrane of 7 µm/s. The experiments were carried out in a partially open chamber to minimize pressure changes during changes of flow rates.& t( c; s/ _8 u  r+ i, Z/ o
& ?2 I5 V, @$ r
Intracellular Ca 2  measurements by fluo 4. The cells were incubated for 30 min with the Ca 2  -sensitive probe fluo 4-AM (5 µM) at 37°C, washed to remove excess probe. Next, they were placed in the perfusion chamber and allowed at least a 20-min deesterification period. Fluo 4 fluorescence was measured as previously described ( 15 ). The fluorescence intensity was expressed relative to the baseline value, chosen as the mean of five intensity observations before the experimental manipulation. All solutions contained 5 mM probenecid to inhibit extrusion of the dye, and the experiments were carried out at 37°C, pH 7.4.& O1 r7 r8 `- M2 H6 y- K  H

Chloral hydrate. MDCK cells were treated with chloral hydrate according to the method described earlier ( 16 ). The cells were deciliated by 96-h incubation in standard DMEM with 4 mM chloral hydrate added. The medium was changed two times a day, since chloral hydrate is volatile. After deciliation, the cells were allowed to recover 24-48 h in normal DMEM before the experiments were performed.

Statistics. Statistical significance was determined using the nonparametric Mann-Whitney test, and values of P ' N& s& j5 ~2 g" v% R! V9 |

RESULTS

Primary cilia of MDCK cells and rat collecting ducts express 1 -integrin. Two approaches were applied to test for the expression of 1 -integrin in the primary cilium of fixed MDCK cells: staining by fluorescently tagged SNA, which is known specifically to bind 1 -integrin in MDCK cells ( 18 ), and labeling with anti- 1 -integrin antibodies. As exemplified in Fig. 1 A, SNA bound to the primary cilium of mature MDCK cells. The previously reported punctate apical staining pattern is also demonstrated in Fig. 1 A. PNA has been shown specifically to bind MDCK cell fibronectin ( 18 ) and is a natural ligand of 1 -integrin. Double labeling of MDCK cells with fluorescent SNA (in red) and PNA (in green) revealed that some parts of cilia stain predominantly with PNA, some with SNA, and still others bind both lectins (yellow indicates colocalization). Figure 1 C shows the staining pattern of primary cilia with two 1 -integrin antibodies (MAB2000 on left and AB1952 on right ). The staining of the cilium was punctated at this resolution and antibody dilution. Punctate apical staining was also observed ( Fig. 1 C, right ) using the antibody technique. In Fig. 1 C, right, ciliary labeling of less tall MDCK cells (arrowheads) was observed in the same focal plane as apical labeling in taller cells. The ciliary localization of 1 -integrin in MDCK cells was confirmed by immunogold-electron microscopy, as exemplified in Fig. 1 D (AB1952). Gold particles were found on four of eight cilia, as well as corresponding to the apical and basolateral plasma membranes. Control experiments with secondary antibody only did not show any immunogold labeling on either the cilia or the plasma membrane. Thus three independent techniques localized 1 -integrin to the primary cilium of MDCK cells.8 w! s1 `' Z/ ^: T+ x
2 M" `7 b7 e9 S" @
Fig. 1. 1 -Integrin is expressed on the primary cilium of Madin-Darby canine kidney (MDCK) cells. Confluent wild-type MDCK cells were stained with a fluorescence-tagged lectin known to bind 1 -integrin or with an anti- 1 -integrin antibody and analyzed by laser-scanning confocal fluorescence microscopy. A : Sambucus nigra agglutinin (SNA) bound to the primary cilium of MDCK cells (arrowheads) in the fluorescence micrograph ( left ) or fluorescence overlay on differential interference contrast (DIC; right ). B : peanut agglutinin (PNA, green), known to bind fibronectin adhering to the MDCK cell surface, also stained the primary cilium. The cilia were similarly stained with SNA (red), but colocalization was rarely observed (arrowheads). C : fixed, confluent, wild-type MDCK cells were stained with an anti- 1 -integrin antibody. Confocal fluorescence micrograph combined with DIC imaging demonstrate ciliary staining ( left ). A punctate staining pattern was observed at the level of the apical plasma membrane ( right ) and the ciliary labeling of less tall cells (arrowheads). Bars indicate 10 µm. D : immunoelectron microscopic localization of 1 -integrin to the primary cilium of MDCK cells. Bar indicates 300 nm.

Like MDCK cells, intact rat collecting ducts have been shown to express 1 -integrin on the apical membrane ( 19 ) in addition to the well-known basolateral localization. Double-labeling confocal fluorescence microscopy was used to examine if 1 -integrin is present at the primary cilia in other rat kidney tubules. The cilia were visualized by tubulin staining; DIC images facilitated the identification of the tubular segments. 1 -Integrin labeling (in green, AB1952) was observed on the primary cilium of most renal tubules from proximal tubules ( Fig. 2 A ), cortical collecting ducts ( Fig. 2 B ), medullary thick ascending limbs and inner medullary collecting ducts ( Fig. 2 C ), and inner medullary collecting ducts ( Fig. 2, D and E ). The ciliary localization of 1 -integrin in cortical collecting ducts was confirmed by immunogold-electron microscopy, as shown in Fig. 2 F, where labeling was observed in one out of one cilium, with no labeling in the controls with only secondary antibody on neither the cilia nor the plasma membrane. The specificity of the antibodies employed was tested by immunoblotting. Figure 2 G shows that both anti- 1 -integrin antibodies recognize a protein of the expected size for 1 -integrin in MDCK cells. Furthermore, one antibody (AB1952) also recognizes the protein from rat kidney. The protein seems most abundant in inner medulla and inner stripe of outer medulla, with weaker reactivity in whole kidney (which mainly contains cortex) and cortical samples. This also reflects the general labeling intensity observed from the immunohistochemical analysis above. Thus the expression of 1 -integrin in MDCK primary cilia was not unique to cultured cells but was present in several rat kidney tubule segments.& z) l3 W  `5 a

Fig. 2. 1 -Integrin is expressed on the primary cilium of rat kidney tubules. Paraformaldehyde-fixed rat kidneys were stained with anti- 1 -integrin and anti- -tubulin antibodies and analyzed by laser-scanning confocal fluorescence microscopy. 1 -Integrin ( top, green) was localized to the primary cilia (arrowheads) of proximal tubules ( A ), cortical collecting ducts ( B ), collecting ducts and thick ascending limbs of the outer medulla ( C ), and inner medullary collecting ducts ( D ). Inner medullary collecting duct details are shown at higher magnification ( E ), where the red arrowhead indicates DIC detection of a primary cilium. A-E : tubulin staining ( middle ) was applied to facilitate the detection of primary cilia, and DIC ( bottom ) images were obtained to identify the tubules. Bars indicate 10 µm. Pxt, proximal tubule; CCD, cortical collecting duct; OMCD, outer medullary collecting duct; mTAL, medullary thick ascending limb; IMCD, inner medullary collecting duct; ThL, thin descending limb. F : immunoelectron microscopic localization of 1 -integrin to the primary cilium of cortical collecting ducts. Bar indicates 300 nm. G : validation of antibodies by immunoblotting. Left : proteins of the expected size for 1 -integrin were detected in inner medullary (IM) samples and in MDCK cells at high antibody dilution (AB1952). At lower antibody dilution, a protein was also detected in whole kidney (WK), kidney cortex (Ctx), and inner stripe of outer medulla (ISOM). Right : the monoclonal anti- 1 -integrin antibody (MAB2000) only detected MDCK cell proteins.

Both 3 - and 5 -integrins are expressed on the MDCK cell primary cilium. Lectin-binding studies demonstrated that fibronectin binds to the apical plasma membrane of MDCK cells probably via apically expressed 1 -integrin ( 19 ). Therefore, the most likely -subunits to be involved in fibronectin binding would be 3 - and 5 -integrin. Figure 3 A shows the ciliary and apical staining with an anti- 3 -integrin antibody (green) and -tubulin labeling (red). The ciliary staining appeared scattered, as shown by a DIC overlay on the anti- -tubulin staining ( Fig. 3 B ). The apical membrane staining pattern for 3 -integrin was also punctated. In addition, staining with an 5 -integrin antibody was also detected ( Fig. 3 C ), although the fluorescent areas on the cilia seemed less frequent compared with the 3 -integrin staining. The 5 -integrin labeling of the apical membrane ( Fig. 3 D ) was comparable with the apical 3 -integrin staining. Hence, the primary cilium of MDCK cells is most likely to express 1 3 -integrins and to a lesser extent 1 5 -integrins.0 I- R  p) H; O6 f

Fig. 3. Both 3 - and 5 -integrin are positioned on the MDCK primary cilium. A : fixed confluent wild-type MDCK cells were stained with antibodies against -integrin (green) and -tubulin (red). A punctate 3 -integrin staining pattern was observed on the primary cilium (arrowheads) and at the level of the apical plasma membrane. B : combined fluorescence and DIC microscopy also localized 3 -integrin to the cilium (arrowheads) and the apical cell surface. Inset : ciliac labeling in a focal plane above the cells. C : punctate 5 -integrin staining pattern was observed on a primary cilium (arrowheads) and at the level of the apical plasma membrane. D : combined fluorescence and DIC microscopy also localized 5 -integrin to the cilium (arrowheads) and the apical cell surface. Bars indicate 10 µm.( B% K$ S* d: U& p. o$ c

The primary cilium potentiates a fibronectin-induced intracellular Ca 2  response. Cross-linkage of 1 -integrins has previously been shown to induce an increase in [Ca 2  ] i in MDCK cells ( 26 ). Fibronectin was used as an agonist to investigate if cross-linkage of apical 1 -integrin induces [Ca 2  ] i responses in our MDCK cells. Fibronectin (1 mg/l) added at a constant flow rate produced an increase in [Ca 2  ] i, as indicated by fluo 4 ( Fig. 4 A ). Figure 4 A shows a representative trace. The relative increase of fluorescence amounted to 1.99 ± 0.04 ( n = 29)-fold. Restimulation of the cells with fibronectin after a washout period did not produce a new Ca 2  response (data not shown). This was not a result of a general impairment of Ca 2  mobilization, since ATP (100 µM) was able to generate a substantial intracellular [Ca 2  ] i transient of 1.64 ± 0.04-fold in fluorescence ( n = 15; Fig. 4 A ). Addition of ATP (100 µM) was used throughout the study as a control of the cell's ability to generate intracellular Ca 2  transients.' j1 G1 @; x9 y2 |# y# s6 ^

Fig. 4. An agonist of 3( 5 ) 1 -integrin, fibronectin, induces an increase of cytosolic Ca 2  concentration ([Ca 2  ] i ). Confluent wild-type MDCK cells grown on coverslips were loaded with the Ca 2  -sensitive fluorescent dye fluo 4. A : after 30 min equilibration, 10 µg/ml fibronectin was introduced as indicated while keeping a constant flow rate of 8 µl/s throughout the experiment. After washout of fibronectin, ATP (100 µM) was added to probe whether the cells were refractory to Ca 2  signaling. A representative experiment is shown as means ± SE of 15 cells. B : dose-response relationship between added fibronectin and the Ca 2  signal of ciliated and nonciliated MDCK cells., Confluent ciliated MDCK cells;, nonconfluent, nonciliated cells;, MDCK cells deciliated by chloral hydrate treatment (no. of cells tested/concentration varied between 8 and 32). C : confocal fluorescence micrographs of 1 -integrin labeling (green) in control MDCK cells ( top, with red nuclear counterstaining) and chloral hydrate-treated cells ( bottom ). Bars indicate 20 µm, and arrowheads denote primary cilia., n* y6 V, x5 D6 {! v: R
( i* K6 A& n. e; r4 F# }9 Z6 v$ v4 }
The effect of fibronectin on [Ca 2  ] i was dose dependent ( Fig. 4 B ). Shown in Fig. 4 B are confluent MDCK cells, with primary cilia 5-7 µm long, nonconfluent MDCK cells that have previously been shown to not yet have grown cilia ( 15, 30 ), and mature MDCK cells where the primary cilium had been removed by 4 mM chloral hydrate pretreatment, as described in METHODS. The dose-response curve for MDCK cells without primary cilia is shifted significantly to the right, indicating that the lack of a primary cilium makes the cell less responsive to fibronectin. Control and chloral hydrate-treated MDCK cells were immunostained to ensure that the treatment did not alter the surface expression of 1 -integrin. As shown in Fig. 4 C, chloral hydrate had no obvious effect on apical 1 -integrin immunoreactivity. Low-magnification images are shown to demonstrate the regularity of the observation. The high-magnification images reveal the punctate staining pattern in controls and treated cells. Thus both ciliated and nonciliated MDCK cells respond by a transient increase in [Ca 2  ] i when stimulated with fibronectin. However, the ciliated cells respond to a much lower concentration of fibronectin than the nonciliated cells. This might suggest that receptors on a primary cilium allow an epithelial cell to detect lower concentrations of a substance compared with only having receptors at the apical membrane.
/ u8 r3 t/ a) _
Ca 2  -dependent binding of fibronectin to 1 -integrin is necessary for the response. To verify that the fibronectin-induced [Ca 2  ] i response was the result of cross-linkage of 1 -integrins, the cells were incubated with RGD peptide. RGD peptide binds specifically to 1 -integrin and blocks the binding site for fibronectin ( 2 ). Figure 5 A shows that incubation with RGD peptide (100 µM) 30 min before the experiment abolished the fibronectin-induced Ca 2  response ( P $ F2 ^: S5 `  x2 ^/ i, V
% b7 i. V3 H. I; }+ l. J$ q) p# T
Fig. 5. Fibronectin-induced Ca 2  response is prevented by RGD peptide and low extracellular Ca 2 . A : RGD peptide, an antagonist of fibronectin- 1 -integrin interaction, prevents the fibronectin-induced Ca 2  response. RGD peptide was added 2 min before the experiment, and the perfusate flow rate was kept constant at 8 µl/s. Fibronectin was added in a concentration of 1 µg/l, and the ATP concentration was 100 µM. A representative experiment is shown as means ± SE of 16 cells. B : omission of extracellular Ca 2  prevents the fibronectin-induced Ca 2  response. The concentration of fibronectin was 1 µg/ml. Ca 2  was removed from the medium 2 min before introducing fibronectin, and the perfusate flow rate was kept constant at 8 µl/s. A representative experiment is shown as the means ± SE of 19 cells./ B6 \. q, S" [0 J1 o
% w5 N. f& W: A" S& P" Y
Extracellular Ca 2  is necessary for the interaction of fibronectin with 1 -integrin ( 29 ). In Fig. 5 B, a Ca 2  -free solution was superfused at a constant flow rate for 1 min before the addition of fibroncetin (1 mg/l). This procedure completely abolished the fibronectin-induced [Ca 2  ] i response in confluent, ciliated MDCK cells. The relative fluorescence was 0.97 ± 0.02, n = 54 in all cells tested, which is a statistically significant reduction compared with the fibronectin-induced [Ca 2  ] i increase in the presence of extracellular Ca 2  ( P ) w. z, G* \1 Y! B

Cross-linking of 1 -integrins by SNA also induces a [Ca 2  ] i response. If the cross-linkage of 1 -integrins is essential for the fibronectin-induced Ca 2  signal in MDCK cells, other stimuli that cross-link the receptors should induce a similar increase. Figure 6 A shows the Ca 2  increase in response to cross-linkage of the 1 -integrin by 1 mg/l SNA. The increase in [Ca 2  ] i as indicated by the fluorescence amounted to 2.36 ± 0.10 ( n = 36). The effect of SNA was dose dependent, as shown in Fig. 6 B.' _. X* a. I+ a/ g5 T* M

Fig. 6. 1 -Integrin cross-linking by SNA induces Ca 2  signaling. A : SNA, 10 µg/ml, was introduced as indicated at a constant flow rate of 8 µl/s. A representative experiment is shown as the means ± SE of 17 cells. B : dose-response relationship between added SNA and the Ca 2  signal. The number of cells tested/concentration varied between 14 and 36.3 i) ~- p5 }# x8 j2 D# H

The fibronectin-induced response is independent of a flow-induced Ca 2  response. The primary cilium is the flow sensor in MDCK cells ( 15 ). In general, agonists for the integrin complex exemplified here by fibronectin are rather large molecules. Binding of these molecules to the primary cilium might potentially modify its physical properties, either weighing it down or increasing its resistance to flow and making it more sensitive to a given perfusate flow rate. We wanted to determine whether the addition of fibronectin modified the Ca 2  transients by receptor interaction or by sensitizing the primary cilia to flow. In Fig. 7 A, the fibronectin (0.2 mg/l) response is shown in the presence of Gd 3 , which is known to inhibit the flow-induced [Ca 2  ] i response in MDCK cells ( 15 ). The concentration of fibronectin was chosen from the dose-response curve to a give full response in confluent, ciliated MDCK cells. The cells were incubated with 30 µM Gd 3  for 30 min before the experiment. Figure 7 A shows that, in the presence of Gd 3 , there was no change in the relative fluorescence when the flow rate was increased from 0 to 12 µl/s. Figure 7 A is a representative trace; the average change in all cells tested amounted to 1.10 ± 0.03 ( n = 74). The fibronectin-induced [Ca 2  ] i increase, however, was unaffected by Gd 3 .

Fig. 7. The fibronectin-induced Ca 2  response is independent of a flow-induced Ca 2  response. A : the cation channel inhibitor Gd 3  prevents the flow-induced Ca 2  response but not the fibronectin-induced response. The cells were pretreated with 30 µM Gd 3  for 30 min before the experiment (for positive control: see Fig. 8 A ). A representative experiment is shown as the means ± SE of 13 cells. B : the fibronectin-induced Ca 2  response is independent of perfusion. Fibronectin (0.2 µg/l) was added gently to the chamber without flow. A representative experiment is shown as the means ± SE of 14 cells.

We further tested whether adding fibronectin under conditions where there was no flow still caused a [Ca 2  ] i response ( Fig. 7 B ). The fibronectin-induced change in relative fluorescence under those conditions was 1.58 ± 0.05 ( n = 28) in all cells tested. The relative fluorescence also increased 1.97 ± 0.02 ( n = 13)-fold when the flow rate was reduced from 8 to 2 µl/s as fibronectin was added. Lowering the flow rate by itself did not increase [Ca 2  ] i (data not shown). Taken together, these experiments support the conclusion that fibronectin mediates its effect by receptor interaction rather than sensitizing the primary cilium to flow.' ]6 g4 v9 x6 x- P: I1 ^

The fibronectin-induced and the flow-induced Ca 2  responses are separate signals. 1 -Integrin is known to participate in the shear stress signal transduction of endothelial cells. Twisting of 1 -integrin in the membrane induced intracellular Ca 2  responses in model cells ( 28 ). The apical and ciliary presentation of 1 -integrin allows for the possibility that the integrins participate in the flow-induced Ca 2  response of MDCK cells. Therefore, we tested whether cross-linking of the receptors could prevent the typical flow-induced intracellular Ca 2  response in MDCK cells. In the absence of Gd 3 , Fig. 8 A shows that a change in perfusate flow rate from 0 to 12 µl/s produced an increase in relative fluo 4 fluorescence of 1.67 ± 0.05 ( n = 133) in all cells tested, with a corresponding ATP response of 1.93 ± 0.05 ( n = 118). Fibronectin (0.1 mg/l) induced a fluo 4 fluorescence increase of 1.64 ± 0.04 ( n = 71) in Gd 3  -treated cells. When the cells were incubated with fibronectin (1 mg/l; Fig. 8 B ) before the experiment, the flow-induced change in relative fluorescence was 1.56 ± 0.07 ( n = 42) in all cells tested. A representative trace is shown in Fig. 8 B. The corresponding ATP signal was 2.36 ± 0.07 ( n = 42). If the receptors were cross-linked with SNA (1 mg/l), the relative fluorescence amounted to 1.60 ± 0.05 ( n = 71) in all cells tested. Figure 8 C shows a representative trace. The corresponding ATP signal was 2.04 ± 0.07 ( n = 49). None of these are statistically significantly different from the control response of 1.67 ± 0.05 ( n = 133) of untreated MDCK cells or the corresponding ATP response of 1.84 ± 0.09 ( n = 36). Preventing cross-linkage of the 1 -integrin with RGD peptide did not affect the flow-induced increase in relative fluo 4 fluorescence either, since the fluorescence change after incubation with RGD peptide (100 µM; Fig. 8 D ) was 1.53 ± 0.08 ( n = 33, not significantly different from control) in all cells tested. The ATP response amounted to 2.10 ± 0.10 ( n = 33, not significantly different from control) in the same cells ( Fig. 8 D ).' k% {) j0 G2 q( Y5 T
* r, i! W/ L8 l5 V3 v* L& s
Fig. 8. The fibronectin-induced and the flow-induced Ca 2  responses are separate signals. A : a typical flow-induced Ca 2  response is followed by an ATP-induced signal. Mean values ± SE from 15 experiments are shown. Preincubation with 1 mg/l fibronectin ( B ), SNA (1 mg/l; C ), or RGD peptide (100 µM; D ) did not prevent the flow-induced Ca 2  response. Fibronectin, SNA, or RGD peptide was present 5 min before flow-stimulation. Representative experiments are shown as means ± SE of 16 ( A ), 19 ( B ), 18 ( C ), and 19 ( D ) cells.
, T5 m( T5 U9 Q7 p# {
DISCUSSION! B$ d7 V* ]0 [: w" X+ o

In the present work, we show that 1 -integrin and 3 - and 5 -integrins are localized to the primary cilium of MDCK cells. Fibronectin, an agonist for 1 -integrin, evoked an intracellular Ca 2  signal. The presence of a primary cilium greatly enhances the sensitivity of this response. We show that the fibronectin-induced response is an independent signal separated from the previously reported flow-induced Ca 2  response ( 15 ). It is proposed that the primary cilium, apart from its involvement in flow sensing, could allow the cells to chemically sense substances in the bulk flow of the renal tubule lumen./ n# }0 Z' O- n  }

Detecting specific staining of the primary cilium can be challenging. Therefore, several techniques were used in the present study to localize the integrins to the primary cilium. As a first step, we used two fluorescent lectins that previously have been shown to bind specifically to the glycosylated 1 -integrin (SNA) and secreted fibronectin (PNA) attached to the integrin of MDCK cells ( 18 ). Most frequently, the cilia predominantly binds only one of these lectins, reflecting that fibronectin likely binds desialylated 1 -integrin, as previously discussed ( 19 ). In that study, the two lectins colocalized in a subset of cells, which displayed lower fluorescence intensity for both fluorophores. This may reflect a partial desialylation of the integrins and therefore a reduced capability to bind fibronectin. Commercially available antibodies were used to verify the expression of 1 -integrin on the primary cilium. In contrast to the continuous ciliary lectin staining, the 1 -integrin labeling seemed spatially disperse, similar to that on the apical membrane of MDCK cells. However, the punctate ciliary labeling appeared continuous with increasing antibody concentration.
/ e- N' c' p  n9 N
The integrin receptor for fibronectin consists of 3 1 - or 5 1 -subunits. These two -subunits were likely also expressed on the primary cilium with 1 -integrin, since secreted fibronectin seemed to bind the cilium. The ciliary anti- 3 -integrin labeling was punctate like 1 -integrin at the applied concentrations. This could either be caused by the low antibody concentration used to avoid background staining or, perhaps, reflected spatial separation of the 3 -integrin and 5 -integrin subunit expression. The integrin labeling of cilia was confirmed by colocalization with -tubulin, since the presence of microtubules in the protruding organelle helped to identify it. In this context, it is worth noting that the fully developed collecting duct predominantly expresses 3 1 -integrin ( 14, 20 ).

The expression of 1 -integrin was not limited to MDCK cells, a cell culture derived from canine cortical collecting ducts. Rat proximal tubules, medullary thick ascending limbs, and all segments of the collecting ducts expressed 1 -integrin on all of their primary cilia. This is consistent with findings in flagella from Trypanosoma congolense, which express 1 -integrin. T. congolense adhere specifically to endothelial surfaces with the anterior part of their flagella, as shown by scanning and transmission electron microscopy ( 24 ). The interaction between parasite and host cell is very tight, and, frequently, accumulation of endocytotic vesicles near the contact site is observed. Immunoelectron microscopy revealed a compound distributed over the total surface of the trypanosomes and reacting with antibodies against 1 -integrin ( 24 ). Furthermore, 1 -integrin has been demonstrated on the apical hair-cell surface where the stereocilia are formed ( 10 ).5 r4 `) t9 L6 b: L
8 k5 l+ i. |# |, ~" J- r
The sensitivity to fibronectin seems to be closely linked to the presence of a primary cilium. Thus ciliated MDCK cells are able to detect fibronectin in 200 times lower concentration than if the cilium were absent. This suggests chemosensing as an additional function of the primary cilium, i.e., the expression of the relevant receptors on the primary cilia possibly enables the kidney cells to detect quite low concentrations of certain chemical substances in the bulk flow. The reason for the increased sensitivity in the presence of the primary cilium is unclear. It might reflect that the primary cilia extend beyond the unstirred layers and thus are exposed to the agonist more effectively, or it could be a consequence of the expression of a more sensitive integrin subtype in the primary cilium.

Because the primary cilium is able to sense apical flow, the shift in the fibronectin dose-response curve between the ciliated and nonciliated cells might be explained by a higher sensitivity to flow created by fibronectin. This is quite feasible since fibronectin is a rather large dimer of two 240-kDa molecules, and binding of molecules of that size to the primary cilium could increase the sensitivity to flow by making the cilium bend to a higher degree at a given flow rate. To investigate this, fibronectin was added under various conditions where flow was either absent, where the flow-induced Ca 2  response was inhibited by Gd 3 , or under conditions where the flow rate was decreased substantially (from 8 to 2 µl/s). In all cases, fibronectin produced a Ca 2  increase that was indistinguishable from the control experiment in which fibronectin was added at a constant flow rate. Thus it is very unlikely that fibronectin produces its Ca 2  response by increasing the flow sensitivity.

As mentioned above, the primary cilium was shown to be essential for sensing apical flow in MDCK cells ( 15, 16 ). Bending of the primary cilium results in an increase in [Ca 2  ] i. The flow-induced Ca 2  response consists of an initial Ca 2  influx via a stretch-activated Ca 2  channel, followed by Ca 2  -induced Ca 2  release from the intracellular Ca 2  stores. Recent results point to polycystin-2 as the stretch-activated channel responsible for the initial Ca 2  influx ( 12 ). There are some indications of integrins acting through the same signal transduction pathway as the polycystins and that 1 -integrin is involved in flow sensing in endothelia ( 25 ). 1 -Integrin and polycystin-2 were coimmunoprecipitated from human renal epithelial cell lines ( 30 ), suggesting a direct molecular interaction between the two molecules or at least a close spatial relation. We speculate that the three proteins could form a multiprotein unit. Because the deformation of 1 -integrin has been suggested to mediate the sensing of pressure in chondrocytes ( 22 ), integrins could be part of a mechanosensitive functional unit consisting of 1 -integrin and polycystin-1 and -2. Our results show, however, that neither the activation of the intergrin complexes by fibronectin or SNA nor the prevention of activation by RGD peptide had any influence on the amplitude of the flow-induced Ca 2  response. This means that, should the integrins participate in the flow-induced Ca 2  response, it must be via a mechanism different from the established activation pathway.2 T, P& _: c$ t2 j

It is of note, however, that the shape of the flow-induced Ca 2  signal is changed when the 1 -integrins are cross-linked ( Fig. 8 ). The duration of the Ca 2  signal is reduced in the presence of fibronectin and SNA. This could mean that fibronectin modulates capacitative Ca 2  entry in MDCK cells, as has been has been demonstrated in other cell types ( 21 ). The Ca 2  response to flow of mouse renal cells is considerably shorter in duration than that in MDCK cells ( 12, 17 ). This difference could be related to the degree of cross-linking of the integrins rather than to some inherent difference in the signaling cascade.
/ U5 {# E1 g  |& |+ _
Clearly, our results show that 1 -integrin does not mediate the flow-induced Ca 2  response in mature confluent MDCK cells. The function of 1 -integrins on the apical membrane and the primary cilium is still unresolved. It is evident that the primary cilium increases the cell's sensitivity to fibronectin, allowing very low doses to be detected. Interestingly, the concentration of fibronectin in the urine is reportedly 0.5 mg/l ( 27 ), which is high enough to assure constant stimulation of both ciliated and nonciliated MDCK cells. We speculate that 1 -integrins could be involved in detecting damage to upstream renal tubules that result in the release in the lumen of agonists such as fibronectin or collagen.

In summary, we conclude that the primary cilium of kidney tubular cells is capable of being a chemosensor, at least to large molecules in the bulk urinary flow. Fibronectin may be the first of many chemical substances that can be detected in low concentration by the primary cilium.

GRANTS

The following foundations have supported this work: The Danish Medical Research Foundation, Grundforskningsfonden, Nyreforeningens forskningsfond, The Aarhus University Research Foundation, Eva og Henry Frænkels Mindefond, and The A. P. Møller Foundation for the Advancement of Medical Science.! V+ r2 `" F) T. ]# _. C$ n
% G, v8 N0 E  f6 G- n; L
ACKNOWLEDGMENTS

We thank Inger Merete Paulsen, Ida Maria Jalk, Zhila Nikrozi, and Dorte Emilie Wolf for technical assistance.
      【参考文献】* g! F. ?# ?. U& Q6 V% F0 g  }
Berk BC, Corson MA, Peterson TE, and Tseng Hennessey. Protein kinases as mediators of fluid shear stress stimulated signal transduction in endothelial cells: a hypothesis for calcium-dependent and calcium-independent events activated by flow. J Biomech 28: 1439-1450, 1995.% k% m3 N8 }1 P+ [$ w+ M8 h

+ ?5 ?3 c& C3 N9 X0 |5 _
Cheng YF and Kramer RH. Human microvascular endothelial cells express integrin-related complexes that mediate adhesion to the extracellular matrix. J Cell Physiol 139: 275-286, 1989./ v& G4 [1 B, K" y6 s2 g

7 [4 Y2 L: U: r8 N1 [! B# z' Y. }3 c

Chrzanowska-Wodnicka M and Burridge K. Rho-stimulated contractility drives the formation of stress fibers, and focal adhesions. J Cell Biol 133:1403-1415, 1996.' L, }8 m; q/ B5 R+ @
+ L2 C! V6 S$ ~; a) q8 j$ N* s5 Z

( P7 _* Y9 x( b! I3 ]
Clark EA and Brugge JS. Integrin and signal transduction pathways: the road taken. Science 268: 233-239, 1995.
# |7 H( F0 Z8 W" N( i0 O) a1 w
7 O3 g4 L, S0 V; ~
4 X( ?5 r% L2 q
Eble JA. Integrins: a versatile and old family of cell adhesion molecules. In: Integrin-Ligand Interaction, edited by Eble JA and Kühn K. New York, NY: Springer, 1997, p. 1-40.
2 H% `4 g: c& g
, m; H- n, ~/ d* F
  T: D; u& d2 n0 j3 C" _
Jaconi MEE, Rivest RW, Schlegel W, Wollheim Pitttet D CB, and Lew PD. Spontaneous and chemoattractant-induced oscillations of cytosolic free calcium in single adherent human neutrophils. J Biol Chem 263: 10557-10560, 1988.

Jiang ST, Chiu SJ, Chen HC, Chuang WJ, and Tang MJ. Role of 3 1 integrin in tubulogenesis of Madin-Darby canine kidney cells. Kidney Int 59:1770-1778, 2001.- e* }( `" a9 t5 ]* f) p# G) j

+ D' T8 P. {- s% F/ M
Kaufmann R, Frosch D, Westphal C, Weber L, and Klein CE. Integrin VLA-3: ultrastructural localization at cell-cell contact sites of human cell cultures. J Cell Biol 109: 1807-1815, 1989.. o% l0 s1 Y# c" o( ?7 w

/ s4 i5 c1 Z0 t0 q* |$ B
' U( G0 g6 q1 w' _
Larjava H, Peltonen J, Akiyama SK, Yamada SS, Gralnick HR, Uitto J, and Yamada KM. Novel function for beta 1 integrins in keratinocyte cell-cell interactions. J Cell Biol 110: 803-815, 1990.2 p8 ^8 T2 Q( V2 k. o

9 q, G+ K  J1 v+ H2 ~- u
+ X& ?8 t7 c2 B0 r; z6 z1 A
Littlewood Evans A, and Muller U. Stereocilia defects in sensory hair cells of the inner ear in mice deficient in integrin 8 1. Nature 24: 424-428, 2000.
: e3 I9 L. ]1 ?0 x5 _" M+ r) w" A

: d0 K3 X' s1 h1 Q
Nathan C and Sporn M. Cytokines in context. J Cell Biol 113: 981-986, 1991.

Nauli SM, Alenghat FJ, Luo Y, Williams E, Vassilev P, Li X, Elia AE, Lu W, Brown EM, Quinn SJ, Ingber DE, and Zhou J. Polycystins 1 and 2 mediate mechanosensation in the primary cilium of kidney cells. Nat Genet 33: 129-137, 2003.

$ c, \6 q$ b+ i3 e( \: G' m

Pardi R, Bender JR, Dettori C, Giannazza E, and Engelman EG. Heterogeneous signaling properties of lymphocyte function-associated antigen (LFA-1) in human lymphocyte subsets. J Immunol 266: C376-C381, 1994.

! X- U/ t2 {, g: {9 E* \2 {
Patey N, Halbwachs-Mecarelli L, Droz D, Lesavre P, and Noel LH. Distribution of integrin subunits in normal human kidney. Cell Adhes Commun 2: 159-167, 1994.

& B, {$ g; J  K: i) T, ^' j
Praetorius HA and Spring KR. Bending the MDCK cell primary cilium increases intracellular calcium. J Membr Biol 184: 71-79, 2001.
) ]- J  _2 x- _) z! v/ ^7 \! v

Praetorius HA and Spring KR. Removal of the MDCK cell primary cilium abolishes flow sensing. J Membr Biol 191: 69-76, 2003a.

* S3 E& h9 Q0 ]( @  F. e2 ^
5 j6 t- s6 u) ~
Praetorius HA and Spring KR. The renal cell primary cilium functions as a flow sensor. Curr Opin Nephrol Hypertens 12: 517-520, 2003b.1 b4 |& e' g- e) L! \6 q$ n1 G
. @- S! j4 }3 H* W) ]) [2 v
) g( ?+ [. \- r2 ^+ t0 e
& c7 l4 e1 P" e* n1 C( j
Praetorus J, Backlund P, Yergey AL, and Spring KR. Specific lectin binding to beta1 integrin and fibronectin on the apical membrane of madin-darby canine kidney cells. J Membr Biol 184: 273-281, 2001.$ T. p/ }5 V( d  `4 c

% R0 _, H" O2 v  f5 @
. X9 H# ?/ F; V: Y
Praetorius J and Spring KR. Specific lectins map the distribution of fibronectin and 1-integrin on living MDCK cells. Exp Cell Res 276: 52-62, 2002.. w( O/ b1 N) u, [
. \  E" Z2 t  V0 c; M, G
) I0 A# N$ b; J3 d3 l4 e- W

Rahilly MA and Fleming S. Differential expression of integrin chains by renal epithelial cells. J Pathol 167: 327-334, 1992.( O7 i- Y7 {8 [- C( _* V

4 B: G' p, S4 u, y4 L- H' c
% {# ~* B+ g9 Z6 x
Rosado JA, Meijer EMY, Hamulyak K, Novakova I, Heemskerk JWM, and Sage SO. Fibrinogen binding to the integrin IIb 3 modulates store-mediated calcium entry in human platelets. Blood 97: 2648-2656, 2001.$ S# F( I& a/ l4 V! T

$ d8 d( l2 F2 T- m/ \
Salter DM, Millward-Sadler SJ, Nuki G, and Wright MO. Integrin-Interleukin-4 mechanotransduction pathways in human chondrocytes. Clin Orthop 391: S49-S60, 2001.

/ W$ t) [' u% w  D" M1 P# d" j

Sastry SK and Horwitz AF. Integrin cytoplasmic domains: mediators of cytoskeletal linkage and extra and intracellular initiated transmembrane signaling. Curr Opin Cell Biol 5: 819-831, 1993.! `8 e! I  b3 m. \

% ~% n9 H1 T$ D. Y. y+ o1 Y$ S
9 T- P: V7 L9 h$ E8 z/ i8 G
Shakibaei M, Milaninezhad M, and Risse HJ. Immunoelectron microscopic studies on the specific adhesion of Trypanosoma congolense to cultured vascular endothelial cells. J Struct Biol 112: 125-135, 1994.
& v* `  c" C6 h* [

Shyy JY and Chien S. Role of integrins in cellular responses to mechanical stress and adhesion. Curr Opin Cell Biol 9: 707-913, 1997.! u8 A* ~, ~8 a; w8 p, Q

6 y3 ^! P/ j4 `
Sjaastad MD, Lewis RS, and Nelson WJ. Mechanisms of integrin-mediated calcium signaling in MDCK cells: regulation of adhesion by IP3- and store-independent calcium influx. Mol Biol Cell 7: 1025-1041, 1996.( s5 n: l( J$ H4 P( m" d' r1 g& k( ]- L

0 i' G" ~8 G9 |9 e/ _% |1 n1 S
Tsujihata M, Miyake O, Yoshimura K, Kakimoto KI, Takahara S, and Okuyama A. Fibronectin as a potent inhibitor of calcium oxalate urolithiasis. J Urol 164: 1718-1723, 2000.
7 ^4 t, {' ?# \  e" {' Q

Wang N, Butler JP, and Ingber DE. Mechanotransduction across the cell surface and through the cytoskeleton. Science 260: 1124-1127, 1993.& w! C5 d# K3 M+ z' D) I
/ G/ y2 U' X7 ~+ l2 z( j1 X7 u. Z; F0 u

Weinreb PH, Yang WJ, Violette SM, Couture M, Kimball K, Pepinsky RB, Lobb RR, and Josiah S. A cell-free electrochemiluminescence assay for measuring beta1-integrin-ligand interactions. Anal Biochem 306: 305-313, 2002.

' R8 \0 X) ?2 N" U$ c* J- {, K

Wilson PD, Geng L, Li X, and Burrows CR. The PKD1 gene product, "polycystin-1," is a tyrosine-phosphorylated protein that colocalizes with 2 1 integrin in focal clusters in adherent renal epithelia. Lab Invest 79: 1311-1323, 1999.
" O/ U/ o3 \# @
+ k. D$ A& _7 l# _( |0 @+ S+ m

Yoder BK, Tousson A, Millican L, Wu JH, Bugg CE Jr, Schafer JA, and Balkovetz DF. Polaris, a protein disrupted in orpk mutant mice, is required for assembly of renal cilium. Am J Physiol Renal Physiol 282: F541-F552, 2002.  _$ I5 L1 h/ |7 g, j0 D! Y/ P
5 n8 u" }  J! T

Zimolo Z, Wesolowski G, Tanaka H, Hyman JL, Hoyer JR, and Rodan GA. Soluble v 3 -integrin ligands raise [Ca 2  i ] in rat osteoclasts and mouse-derived osteoclast-like cells. Am J Physiol Cell Physiol 266: C376-C381, 1994.

收藏0 分享0 顶0 踩0