|
|
||||||||
RENAL HEMODYNAMICS AND CARDIORENAL INTEGRATION
Departments of 1Molecular Pharmacology and Physiology, and 2Physics, University of South Florida, Tampa, Florida; and 3Division of Pulmonary and Critical Care Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland
Submitted 16 January 2007 ; accepted in final form 4 August 2007
| ABSTRACT |
|---|
|
|
|---|
5
1-integrin in VSMCs). Pulling of fibronectin-coated beads with an electromagnet triggered Ca2+ sparks, followed by global Ca2+ mobilization. Paramagnetic beads coated with low-density lipoprotein, whose receptors are not linked to cytoskeleton, were minimally effective in triggering Ca2+ sparks and global Ca2+ mobilization. Preincubation with ryanodine, cytochalasin-D, or colchicine substantially reduced the occurrence of Ca2+ sparks triggered by fibronectin-coated beads. Binding of VSMCs with antibodies specific to the extracellular domains of
5- and
1-integrins triggered Ca2+ sparks simulating the effects of fibronectin-coated beads. Preincubation of microperfused afferent arterioles with ryanodine or integrin-specific binding peptide inhibited pressure-induced myogenic constriction. In conclusion, integrins transduce mechanical force into intracellular Ca2+ signaling events in renal VSMCs. Integrin-mediated mechanotransduction is probably involved in myogenic response of afferent arterioles. myogenic response; paramagnetic bead; fluorescence confocal microscopy
- and
-subunits, which provide the structural link between extracellular matrix (ECM) and internal cytoskeleton and function as signaling receptors. It has been shown in nonmuscle cells that extracellular mechanical force is transmitted across the plasma membrane via integrins to initiate intracellular signaling (17, 25, 33, 43). Synthetic integrin-binding peptide GRGDSP (Gly-Arg-Gly-Asp-Ser-Pro) induces vasoconstriction in rat afferent arterioles, which is associated with a pronounced increase of intracellular Ca2+ concentration ([Ca2+]i) in vascular smooth muscle cells (VSMCs) (47). Longitudinal stretch of urinary bladder smooth muscle cell induces Ca2+ sparks (coordinated opening of a cluster of ryanodine receptors), Ca2+ waves, and inward Cl– current (22). Interestingly, the Ca2+ mobilization in renal VSMCs induced by integrin-binding peptide is ryanodine sensitive and is associated with recurrent Ca2+ waves (5). These observations raise the intriguing hypothesis that integrins on the plasma membrane of renal VSMCs may serve as mechanotransducers to regulate vascular tone by generating Ca2+ sparks and triggering contraction (52). The present study was undertaken in an attempt to provide the evidence to substantiate this hypothesis. To investigate whether integrins transduce mechanical force into Ca2+ sparks, we monitored the occurrence of Ca2+ sparks when magnetic pulling of fibronectin-coated paramagnetic beads (fibronectin beads) was applied to freshly isolated renal VSMCs. To examine whether integrins participate in the mechanotransduction process of pressure-induced myogenic constriction, we monitored the time course of changes in luminal diameter of microperfused afferent arterioles when perfusion pressure was elevated in the presence of integrin-binding peptide or ryanodine. Our results demonstrate for the first time that integrin-mediated mechanotransduction is coupled to localized subcellular Ca2+ release in the form of Ca2+ sparks, and that intact integrin/ECM interactions are required for pressure-induced constriction in isolated renal arterioles.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Changes in subcellular [Ca2+]i, induced by magnetic pulling were monitored in linescan mode (2 ms/scan line) with a Bio-Rad MRC-1000 confocal microscope or Leica TCS SP5 confocal microscope. Linescan images were collected immediately under the plasma membrane of VSMCs. The scanning line was positioned in close proximity to at least one of the attached paramagnetic beads. Images were collected for 4–6 s as baseline, and then for another 20–30 s after pulling was initiated. The operating system in the Bio-Rad confocal microscope (COMOS Version 7.0) can only collect continuously for 512 scan lines. There is a time delay between saving the image and the initiation of scanning for the next 512 scan lines. Time lag between the first scan line of two consecutive images was 2 s. All images were collected with a Zeiss x40 plan-apochromat objective (numerical aperature 1.2). Fluo-4 was excited at 488 nm, and emission was collected with a band pass filter 522/35.
To test whether ligating
5
1-integrins with antibodies that recognize their extracellular domains could trigger Ca2+ sparks, the occurrences of Ca2+ sparks in VSMCs were monitored in the presence of anti-integrin
5-antibody (25 µg/ml, HM
5-1; Pharmingen) or anti-integrin
1-antibody (50 µg/ml Ha2/5; Pharmingen). Soluble anti-
5
1 integrin antibody has been shown to activate
5
1-integrin in tsA-201 and HEK-293 cells (14).
2-antibody (50 µg/ml, Wt.3; Pharmingen) was used as the control.
To test whether structural linkage of integrin-cytoskeleton is required for integrins to mediate mechanotransduction and generation of Ca2+ sparks, isolated renal arterioles were incubated with cytochalasin-D (4 µg/ml) or colchicine (1 mg/ml) in HBSS for 2 h before VSMCs were harvested by enzyme digestion. The occurrence of Ca2+ sparks induced by pulling of fibronectin beads was then monitored in these smooth muscle cells.
Pulling with paramagnetic beads on renal VSMC.
Tosylactivated paramagnetic micro beads (4.5 µm diameter, M-450; Dynal) were coated with fibronectin or low-density lipoprotein (LDL) by using the protocol provided by the manufacturer. Fibronectin is the natural ligand for
5
1-integrins in VSMCs (35, 40, 47). LDL is the ligand for LDL receptor on VSMCs. Pulling of LDL-coated bead was used as control for nonintegrin-mediated mechanical signal transduction. To apply force on renal VSMC by using paramagnetic beads, coated beads were first incubated with VSMCs adhered on glass-bottomed petri dish (World Precision Instruments; Sarasota, FL) for 20 min. Only cells with individual paramagnetic beads attached were used. Pulling was initiated by applying a local magnetic field directed to the cell of interest with a custom-made miniaturized electromagnet mounted on a micromanipulator. The tip of electromagnet has a diameter of 420 µm and was positioned within 100 to 150 µm from the cell of interest at an angle of 30° to 45°. The dragging force imposed by the electromagnet on individual paramagnetic beads was calibrated by pulling the paramagnetic beads through dimethylpolysiloxane (viscosity 100 centistokes; Sigma) (1, 26, 34). The velocity of the beads migrating to the tip of electromagnet was quantified based on the time-lapsed images collected at 2 Hz. The dragging force was then calculated using Stokes Law (force = 6
R
, where
is the viscosity of the fluid, R is the radius of the bead, and
is the velocity of the bead).
Immunofluorescence histochemistry.
To detect the changes in the organization of actin microfilaments induced by cytochalasin-D, renal VSMCs were first fixed with 2% paraformaldehyde in PBS for 20 min and then permeabilized with 0.5% Triton-X in PBS for 30 min. To detect the changes in the organization of microtubules induced by colchicine, renal VSMCs were first fixed with 0.5% glutaraldehyde for 10 min and then quenched with 0.1% NaBH4 (50). Cells were then incubated with either FITC-conjugated phalloidin (4 µg/ml; Sigma) or Cy3-conjugated anti-
-tubulin antibody (20 µg/ml goat IgG; Sigma). Confocal fluorescence images were collected with a Zeiss x63 plan-apochromat objective (numerical aperature, 1.4).
Perfusion of afferent arteriole and measurement of luminal diameter. Experiments were conducted in afferent arterioles isolated from rat juxtamedullary nephrons as reported previously (47, 48). Male Sprague-Dawley rats from Harlan Farms (120–200 g) were killed by anesthetic overdose. The kidneys were quickly removed and placed in ice-cold dissection buffer. A segment of afferent arteriole (300–400 µm) just proximal to a glomerulus was then dissected, cannulated, and perfused in a temperature-controlled perfusion chamber (Vestavia) at 37°C. The perfusion chamber was mounted on a Zeiss Axiovert 100TV inverted microscope that was coupled to a Bio-Rad MRC-1000 confocal scanning unit equipped with a transmitted light detector. The intraluminal pressure of the vessel was initially set at 80 mmHg. Vessels were discarded if there was fluid leakage. Transmitted light images were collected at 0.25 or 0.5 Hz throughout the experimental period. After vascular tone was established in the perfused vessel, perfusion pressure was increased from 80 to 120 mmHg in a single step. The time course of changes in luminal diameter was monitored from the stored images with an edge-detecting algorithm based on covariance (29). The algorithm was implemented with a Matrox IP-8 imaging board (42). The composition of the dissecting solution consisted of (in mM) 115 NaCl, 25 NaHCO3, 2.5 K2HPO4, 1.2 MgSO4, 1.8 CaCl2, 5.5 glucose, 2.0 pyruvic acid, and 1 g/dl dialyzed BSA (fraction V; Calbiochem). The luminal perfusate and bathing solution were identical to the dissecting solution except that no BSA was added to the bathing solution. BSA was excluded in the bathing solution to prevent bacterial growth in the perfusion chamber. All solutions were gassed with 5% CO2 before use, and pH was adjusted to 7.4.
Data analysis.
Results of measurement are shown as means ± SE. Only one arteriole was used from each animal for perfusion study. Ca2+ sparks were detected in confocal images with a custom-made algorithm written in Interactive Data Language software (Research Systems, Boulder, CO). The program identified Ca2+ sparks on the basis of their statistical deviation from the background noise (6, 51). Fluorescence signals (F) of each confocal image were first normalized in terms of F/F0, where F0 is the baseline of fluorescence in a region of the image without Ca2+ sparks. A denoising algorithm based on wavelet-transformation was applied to remove the Gaussian noise from the image (41). The mean and variance (
2) of the normalized image were then determined. Ca2+ sparks were identified based on local fluorescence intensity greater than mean + 2.5
2. The duration and width of Ca2+ sparks were quantified as full duration half maximum (FDHM) and full width half maximum (FWHM), respectively (39). FDHM is the duration (in ms) in which the fluorescence intensity of a Ca2+ spark is greater than half of its peak fluorescence intensity. FWHM is the width (in µm) in which the fluorescence intensity of Ca2+ sparks is greater than half of its peak fluorescence intensity. The spark frequency of each cell was defined as the number of sparks detected per second in a scan line of 25 µm. Statistical significance (P < 0.05) was assessed by paired or unpaired Student's t-test whenever applicable.
| RESULTS |
|---|
|
|
|---|
|
5
1-integrins triggers Ca2+ signal cascade in renal VSMCs. Before the magnetic field was applied to the fibronectin beads, no spontaneous Ca2+ spark was detected in the baseline. Pulling of fibronectin beads triggered Ca2+ sparks, Ca2+ wave, and global Ca2+ increase in 85% of cells with a variable time delay of 0.5 to 3 s. Global increase of [Ca2+]i followed the occurrence of multiple Ca2+ sparks (Fig. 2A). The fluorescence intensity profiles of individual Ca2+ sparks are shown in Fig. 2B. The time course of increase in Ca2+ spark frequency (number of sparks in every 512 scan lines) and spatially averaged global [Ca2+]i (average fluorescence intensity of every 512 scan lines) are shown in Fig. 3A and 3C. The distribution of the spatiotemporal parameters of Ca2+ sparks in terms of FDHM and FWHM are shown in Fig. 4, A and C. The medians of FDHM and FWHM were 24 ms and 1.0 µm, respectively. Other parameters of the Ca2+ sparks are tabulated in Table 1.
|
|
|
|
|
|
Effects of integrin-specific antibodies on renal VSMCs.
Both anti-integrin
5-antibodies and anti-integrin
1-antibodies induced recurrent Ca2+ sparks as observed in pulling of fibronectin beads. In a sampling period of 20 s immediately after the treatment of antibodies, Ca2+ sparks were detected in >85% of cells being tested (Fig. 5). Preincubation with ryanodine inhibited the occurrence of Ca2+ sparks induced by anti-integrin antibodies. While in a similar sampling period, Ca2+ sparks were detected in <14% of cells treated with anti-integrin
2-antibodies.
2-integrins are not expressed in VSMCs; therefore, it was used as timed control for laser illumination. Local Ca2+ release could be induced by prolonged laser scanning. The spatial and temporal parameters of Ca2+ sparks triggered by anti-integrin antibodies are tabulated in Table 1.
Effects of integrin-binding peptide and ryanodine on pressure-induced vasoconstriction. To test whether integrins might contribute to mechanotransduction in intact renal VSMCs, we examined the effects of integrin-binding peptide on pressure-induced myogenic constriction. Pressure-induced myogenic constriction was observed in perfused afferent arterioles as reported previously (48). An increase in perfusion pressure from 80 mmHg to 120 mmHg elicited an immediate dilatation, followed by myogenic constriction. The mean normalized inner diameter vs. time is shown in Fig. 7A. Preincubation of afferent arterioles with synthetic integrin-binding peptide GRGDSP (1 mM) for 20 min did not reduce the luminal diameter significantly. The mean luminal diameters before and after 25 min of GRGDSP incubation were 21.9 ± 1.8 µm and 22.1 ± 1.8 µm (n = 10), respectively. It was consistent with the previous report that vasoconstriction induced by GRGDSP lasts only about 45 s (47). However, an increase in perfusion pressure from 80 mmHg to 120 mmHg induced only dilatation in the presence of GRGDSP. No myogenic constriction was detected (Fig. 7C). Preincubating afferent arterioles with the control peptide GRGESP (Gly-Arg-Gly-Glu-Ser-Pro, 1 mM), which does not have the RGD binding sequence to interact with integrins, had no effect on pressure-induced myogenic constriction (Fig. 7B). These observations indicate that integrins function as mechanotransducers not only in freshly isolated VSMCs but also in intact renal VSMCs. Preincubating afferent arterioles with 50 µM of ryanodine for 20 min dilated the afferent arterioles slightly. The mean luminal diameter before and after ryanodine incubation were 21.2 ± 1.6 µm and 22.7 ± 2.1 µm (n = 8, P < 0.05, paired t-test), respectively. Preincubating afferent arterioles with ryanodine inhibited pressure-induced myogenic constriction (Fig. 7D), suggesting that pressure-induced myogenic constriction is a ryanodine-sensitive process.
|
| DISCUSSION |
|---|
|
|
|---|
The force and distance relationship between the electromagnet and a single paramagnetic bead was established based on Stokes Law (Fig. 1). The maximal pulling force applied to individual VSMC with multiple paramagnetic beads attached could be estimated based on this force-distance relationship. Since the magnetic pulling force is very sensitive to the distance between the magnet and the bead (Fig. 1), the bead that is the closest to the magnet exerts most pulling force. Assuming that there are five beads attached to a VSMC and all beads are 100 µm away from the magnet, the total force exerted is only 0.5 nN (5 x 100 pN). This magnitude of pulling force is comparable to the force used to study the phenomenon of integrin-mediated cell adhesion (1, 26, 34). By applying Law of Laplace (tension = pressure·radius in a cylindrical structure) to perfused arterioles in Fig. 7A, the calculated increase of wall tension was 60.6 ± 6.3 dyne/cm (n = 10) when transmural pressure was increased from 80 to 120 mmHg. It is translated into
60 nN/cell (assuming cell width is 3 µm and cell length is 1/3 of the circumference). The maximal pulling force exerted by the paramagnetic beads (0.5 nN) is an order of magnitude less than that of the increase of wall tension (60 nN). Therefore, the Ca2+ transient triggered by pulling of fibronectin beads is unlikely due to excessive pulling force.
It is suggested that molecular connections between integrins, cytoskeleton filaments, and nuclear scaffolds may conduct mechanical signal transfer throughout the cells and provide a mechanism for producing integrated changes in cells in response to even local changes in ECM mechanics (28). The observations that cytoskeleton disruption agents cytochalasin-D and colchicine attenuated the induction of Ca2+ sparks by fibronectin beads are consistent with this hypothesis. Similar observations were also reported from colonic smooth muscle cells in which Ca2+ release from internal Ca2+ stores induced by mechanical stimulation requires intact actin filament (49). The observations from LDL beads substantiated the notion that structural linkage with internal cytoskeleton is required for transmembrane mechanical signal transduction. Although LDL beads are commonly used as the control for nonintegrin-mediated mechanical stress triggered by a magnetic field (18, 43, 44), there is a remote possibility that the different effects of fibronectin beads and LDL beads on Ca2+ sparks are simply due to the difference in mechanical stress (force/unit area) imposed by the beads to the cells. Even though the amount of force imposed by a fibronectin bead or LDL bead to a cell can be identical (same intensity of magnetic field imposed on the same size of paramagnetic bead), local mechanical stress measured under the bead can be varied depending on the contact area.
Myogenic constriction in the renal artery is associated with membrane depolarization and is attenuated by L-type Ca2+ channel blockers (15). Ca2+ sparks can be induced by membrane depolarization in smooth muscle cells via current through L-type Ca2+ channel (8, 36).
5
1-Integrin regulates L-type voltage-gated Ca2+ channels (Cav1.2) activity via phosphorylation of
1C-COOH-terminal residues Ser1901 and Tyr2122 (14). Currents through L-type voltage-gated Ca2+ channels are acutely potentiated following
5
1-integrin activation by fibronectin and anti-
5
1-integrin antibodies in VSMCs of cremaster skeletal arteriole (14, 45). Ca2+ sparks triggered by a fibronectin bead might be the consequence of enhanced Ca2+ channel activity. However, opening of ryanodine receptors is only loosely coupled by the gating of L-type Ca2+ channels in smooth muscle cells. Opening of L-type Ca2+ channels is not necessary to trigger Ca2+ sparks (8). Linear stretch of urinary bladder smooth muscle cell triggers Ca2+ sparks in the absence of extracellular Ca2+ (22). In the present study, Ca2+ sparks could be triggered by fibronectin beads in the absence of extracellular Ca2+. Therefore, these Ca2+ sparks are most likely mediated by mechanical/chemical signals conducting along the ECM-integrin-cytoskeleton axis and is not dependent on L-type Ca2+ channel activity. Integrin-mediated Ca2+ sparks and potentiation of Ca2+ currents are not mutually exclusive. Indeed, these two mechanisms are synergic for myogenic constriction if Ca2+ sparks trigger membrane depolarization. By imposing step change in arterial pressure and monitoring the dynamics of whole kidney blood flow of rats, myogenic response is shown to be completed in the first 7–9 s, and reaches the maximum speed at 2.2 s (23). The occurrence of Ca2+ sparks (0.5 to 3-s delay) seems to fit well with the finding that Ca2+ sparks precede the myogenic constriction. The time delays of Ca2+ sparks occurrence were probably overestimated because of the time gap between every 512 scan lines.
Ca2+ sparks might induce relaxation or contraction in VSMCs, depending on which types of Ca2+-dependent channels are activated on the plasma membrane. Ca2+ sparks stimulate Ca2+-activated K+ channels in cerebral VSMCs and lead to membrane hyperpolarization and relaxation (19, 37). Ca2+ sparks stimulate Ca2+-activated Cl– channels in pulmonary VSMCs and airway smooth muscle cells and lead to membrane depolarization and contraction (39, 51, 52). Stretch-induced Ca2+ sparks evoke Ca2+-activated Cl– currents in mouse urinary bladder myocytes (22). There are both Ca2+-activated Cl– and Ca2+-activated K+ channels in renal VSMCs (3, 12, 13). When intracellular Ca2+ was mobilized using endothelin-1 in renal VSMCs, membrane depolarization and Ca2+-activated Cl– inward current were observed (13). Blocking of Cl– channels with DIDS inhibits contraction of afferent arterioles (21), while blocking Ca2+-activated K+ channels does not exaggerate agonist-induced constriction in afferent arterioles (11). Collectively, these observations suggest that the activity of Ca2+-activated Cl– channels is predominant over Ca2+-activated K+ channels in determining the contractile state of renal VSMCs. Ca2+ spark-induced depolarization might trigger myogenic constriction via Ca2+ influx through L-type Ca channels (15). Our observations that ryanodine-inhibited pressure induced myogenic constriction are consistent with this hypothesis. It is also in line with a recent report by Loutzenhiser et al. (27) that the initial fast constriction in the myogenic response in afferent arteriole is ryanodine sensitive. Ryanodine only dilated afferent arterioles moderately in the present study, which is consistent with the low occurrence rate of spontaneous Ca2+ sparks in renal VSMCs. Future studies are required to test whether Ca2+ sparks induce membrane depolarization in renal VSMCs.
Preincubation of synthetic integrin-binding peptides GRGDSP inhibited pressure-induced myogenic constriction in afferent arterioles, but the nonintegrin-binding peptide GRGESP had no inhibitory effect. The same peptide GRGDSP also inhibits pressure-induced vasoconstriction in cremaster muscle resistance arterioles (32). These observations implicate that functional interactions between integrins and ECM are required in the signaling processes of myogenic response. Dilatation due to increase of transmural pressure and the subsequent myogenic constriction are probably associated with differential engagement between ECM and integrins in VSMCs. One possible interpretation is that the signal transduction process of myogenic response requires formation of new connections between ECM and integrins. The presence of integrin-binding peptide interferes with the formation of such new connections and thus inhibits the myogenic response. Evidences derived from integrin-mediated mechanotransduction in vascular endothelial cells and NIH3T3 cells are in line with this interpretation. The mechanotransduction in vascular endothelial cells and NIH3T3 cells both require dynamic interactions and formation of new connections between specific ECM and integrins (20, 24).
In summary, pulling of fibronectin beads via
5
1-integrins triggered Ca2+ sparks and global [Ca2+]i increase in freshly isolated renal VSMCs. The transduction of mechanical force into Ca2+ sparks required structural linkage of integrin to cytoskeleton. Blocking of ECM/integrin interactions with synthetic integrin-binding peptide inhibited pressure-induced myogenic constriction in renal arterioles, suggesting that integrins might function as mechanotransducers in vitro and in situ.
| GRANTS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
6
1 integrin adhesion strengthening. Proc Natl Acad Sci USA 100: 7616–7621, 2003.
v
3- and
5
1-integrin blockade inhibits myogenic constriction of skeletal muscle resistance arterioles. Am J Physiol Heart Circ Physiol 289: H322–H329, 2005.
v
3 and
5
1 integrin ligands. J Cell Biol 143: 241–252, 1998.This article has been cited by other articles:
![]() |
Z. Guan, J. S. Pollock, A. K. Cook, J. L. Hobbs, and E. W. Inscho Effect of Epithelial Sodium Channel Blockade on the Myogenic Response of Rat Juxtamedullary Afferent Arterioles Hypertension, November 1, 2009; 54(5): 1062 - 1069. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. W. Inscho Mysteries of Renal Autoregulation Hypertension, February 1, 2009; 53(2): 299 - 306. [Full Text] [PDF] |
||||
![]() |
H. Cheng and W. J. Lederer Calcium Sparks Physiol Rev, October 1, 2008; 88(4): 1491 - 1545. [Abstract] [Full Text] [PDF] |
||||
![]() |
X. Wu, Y. Yang, P. Gui, Y. Sohma, G. A. Meininger, G. E. Davis, A. P. Braun, and M. J. Davis Potentiation of large conductance, Ca2+-activated K+ (BK) channels by {alpha}5{beta}1 integrin activation in arteriolar smooth muscle J. Physiol., March 15, 2008; 586(6): 1699 - 1713. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |