α4β1 Integrin Activation of L-Type Calcium Channels in Vascular Smooth Muscle Causes Arteriole Vasoconstriction
A pathway for the regulation of vascular tone appears to involve coupling between integrins and extracellular matrix proteins or their fragments and the subsequent modulation of ion movement across the smooth muscle cell membrane. Here, we report that the activation of L-type voltage-activated Ca2+ channels occurs through a novel interaction of α4β1 integrin with peptides containing the Leu-Asp-Val (LDV) integrin–binding sequence, which is found in the CS-1 region of an alternately spliced fibronectin variant. Experiments were conducted on arterioles isolated from rat skeletal muscle. Arterioles exhibited sustained concentration-dependent vasoconstriction to LDV peptides but not to Leu-Glu-Val (LEV) control peptides. The constriction was associated with increased smooth muscle cell [Ca2+]i, as measured by using fura 2. The response could be inhibited with a function-blocking anti–α4 integrin antibody. Removal of the endothelium did not alter the vasoconstrictor response. Further experiments demonstrated that the vasoconstriction was abolished by the L-type Ca2+ channel inhibitor nifedipine and the Src family kinase inhibitor PP2. In studies of isolated smooth muscle cells using whole-cell patch-clamp methods, the L-type current was enhanced by the LDV but not LEV peptide and was blocked by PP2 or antibodies to α4 integrin. Collectively, these data indicate that activation of α4β1 integrin leads to enhanced influx of Ca2+ through L-type channels by activating a tyrosine kinase pathway, leading to vasoconstriction. Involvement of integrins in the modulation of vascular tone may be particularly important in vascular responses to mechanical signals, such as pressure and flow, and to tissue injury after damage to the extracellular matrix.
Integrins are a large family of cell-matrix and cell-cell adhesion receptors that, by virtue of their involvement in signal transduction across the cell membrane, have been shown to be involved in many cellular functions, such as proliferation, differentiation, survival, and migration.1 The ability of integrins to respond to soluble matrix fragments as well as insoluble extracellular matrix (ECM) proteins and to provide a link with the actin cytoskeleton has led our laboratory, as well as other laboratories, to investigate integrin participation in the control of vasomotor tone.2–7⇓⇓⇓⇓⇓ Integrin modulation of vascular tone could be important in the vascular myogenic and flow-dependent responses, in which integrins could participate as a tension-sensing mechanism.5,6,8⇓⇓ In addition, integrins may be important vascular receptors for detecting and responding to tissue/vascular injury after damage to the ECM, resulting in exposure and formation of matricryptic sites and/or signals.9–12⇓⇓⇓ We have previously shown that integrin-binding peptides that contain the arginine–glycine–aspartic acid (RGD) sequence can elicit arteriolar vasodilation or vasoconstriction by interacting with the αvβ3 and α5β1 integrins, respectively.2,3⇓ These changes in vasomotor tone appear to be mediated in large part by modulation of the Ca2+ current through L-type Ca2+ channels in vascular smooth muscle cells, such that ligation of the vitronectin receptor, αvβ3, results in reduced current through the L-type channels, and ligation of the fibronectin receptor, α5β1, results in enhanced current.13,14⇓ These observations support the concept that integrins and their interactions with the matrix are an important signal transmission pathway for the regulation of vascular tone. Furthermore, these observations have led us to question whether this concept can be extended to encompass integrin-binding sequences (other than RGD) that may also be capable of altering vasomotor tone.
The integrin α4β1 is an integrin expressed by many cell types, including leukocytes, tumor cells, and skeletal muscle and vascular smooth muscle cells.15–17⇓⇓ This receptor has affinity for the CS-1 region in an alternatively spliced variant of fibronectin and the tissue injury–induced ECM protein osteopontin, as well as vascular cell adhesion molecule (VCAM)-1, which is expressed on endothelial and skeletal muscle cells.18,19⇓ In addition, α4β1 has been shown to bind small peptides modeled from the IIICS region of fibronectin, which contains the Leu-Asp-Val (LDV) sequence.16,17,20⇓⇓ Moreover, these peptides can competitively inhibit cell interactions with naturally occurring α4β1 ligands, such as fibronectin, osteopontin, and VCAM-1.21–23⇓⇓
Fibronectins can be found in plasma and are a major component of the ECM, including the matrix of the vascular wall. They exist as soluble dimers and as insoluble fibrillar networks. Fibronectin diversity is generated by alternative splicing at three sites within the primary fibronectin mRNA transcript.24–26⇓⇓ The IIICS region of fibronectin contains a 25–amino acid sequence, including the LDV motif that binds the α4β1 integrin.16,17,20⇓⇓ This splice variant of fibronectin is known to be upregulated after tissue injury and wound healing27–29⇓⇓ along with increased expression of matrix metalloproteinases, which can cause degradation of fibronectin, resulting in the formation of fragments containing the CS-1 region. It has been suggested that intact fibronectin and proteolytically generated fragments of fibronectin may have differing roles. For example, intact fibronectin has been shown to serve as a chemoattractant stimulus for endothelial cells,30 whereas fibronectin fragments have been shown to be chemotactic for macrophages.31 In the present study, we investigated the possibility that interaction of the LDV integrin–binding motif with the α4β1 integrin in the microvascular wall could induce a vasoactive signal to influence vasomotor tone.
Materials and Methods
All procedures in the present study followed institutional guidelines and were approved by the Texas A&M University Animal Use and Care Committee. Male Sprague-Dawley rats (Harlan, Houston, Tex) weighing 250 to 350 g were anesthetized by intraperitoneal injection of pentobarbital sodium (100 mg/mL). Anesthesia was confirmed by the loss of spinal reflexes. After anesthesia, the right cremaster muscle was excised and pinned flat in a refrigerated (4°C) Lexan dissecting chamber. The chamber contained a physiological saline solution, MOPS-PSS, as follows (mmol/L): NaCl 145.0, KCl 4.7, CaCl2 2.0, MgSO4 1.2, NaH2PO4 1.0, dextrose 5.0, MOPS buffer 3.0, pyruvate 2.0, EDTA 0.02, and BSA 0.15 (Amersham Life Sciences), with pH adjusted to 7.4±0.1 with NaOH. The tissue was allowed to equilibrate in the cold environment for 60 minutes. A first-order (1A) segment of the cremaster feed arteriole was isolated, removed, and placed in a cannulation and observation chamber filled with MOPS-PSS without albumin by using previously described techniques.32 Luminal diameters were recorded online with the use of a calibrated video caliper system.
Peptide and Antibody Treatments
The LDV-containing peptide Glu-Ile-Leu-Asp-Val-Ser-Pro-Thr and LEV-containing peptide Glu-Ile-Leu-Glu-Val-Ser-Pro-Thr (Peninsula Labs)33 were solubilized in PSS. A cumulative concentration-response relationship was obtained by abluminal addition of LDV or LEV (0.21 μmol/L to 0.21 mmol/L) to the vessel bath. Diameter was recorded once per minute for up to 7 minutes to ensure that a steady-state diameter was reached. Diameter changes generally were at steady state within 2 to 3 minutes.
Removal of Endothelium From Isolated Arterioles
The removal of the endothelium was necessary to determine a potential role for the endothelium in response to LDV-containing peptides. Deendothelialization was accomplished by using a nonionic chemical detergent, CHAPS (4%, Sigma Chemical Co), as previously described.34
Vascular smooth muscle cytosolic Ca2+ concentration ([Ca2+]i) was measured in isolated arterioles by using the Ca2+ fluorescence indicator fura 2 (Molecular Probes) as previously reported by this laboratory.35 Equilibrated arterioles were incubated in a solution of 2.0 μmol/L fura 2-AM in 0.05% dimethyl sulfoxide and 0.01% pluronic F-127 for 45 minutes at room temperature (25°C). After this loading period, arterioles were rinsed with fresh PSS and allowed to equilibrate for an additional 30 minutes at 34°C. Changes in [Ca2+]i were recorded with a microfluorometry system with the use of a photomultiplier tube. To simultaneously record diameter during the [Ca2+]i measurement, transillumination light was passed through an orange absorption glass filter (OG-590; Schott Glass Technologies) that transmitted light >575 nm. Thus, fura 2 measurements were not affected.
Cell Isolation Procedure
Cremaster muscles were excised and pinned flat for dissection at 4°C in Ca2+-free saline solution (mmol/L: NaCl 147, KCl 8.6, MOPS 3.0, MgSO4 · 7H2O 1.17, NaH2PO4 1.2, d-glucose 5.0, pyruvate 2.0, and EDTA 0.02) with 0.1 mg/mL BSA; pH was adjusted to 7.4 with NaOH. Dissected and isolated segments of first- and second-order arterioles were washed in low Ca2+ PSS containing (in mmol/L) NaCl 144, KCl 5.6, CaCl2 0.1, MgCl2 1.0, Na2HPO4 0.42, NaH2PO4 0.44, HEPES 10, and NaHCO3 4.17, along with 1 mg/mL BSA; pH was adjusted to 7.4 with NaOH at room temperature for 10 minutes. After cooling, 26 U/mL papain (Sigma) and 1 mg/mL dithioerythritol (Sigma) were added, and the vessels were warmed and incubated for 30 minutes at 37°C with occasional agitation. Fragments were transferred to low Ca2+ saline solution containing 1.95 furanacryloyl-Leu-Gly-Pro-Ala (FALGPA) U/mL collagenase (Sigma), 75 U/mL elastase (Calbiochem), and 1 mg/mL soybean trypsin inhibitor (Sigma) for 15 minutes at 37°C. These fragments were rinsed twice with low Ca2+ saline solution and gently triturated with a fire-polished Pasteur pipette to release single cells.13,14⇓
Whole-cell currents were recorded by using an EPC-7 amplifier (HEKA) under the control of pClamp software (Axon Instruments). Analog-to-digital conversions were made by using a TL-1 DMA interface (Axon Instruments). Data were sampled at 5 to 10 kHz and filtered at 1 to 2 kHz by using an 8-pole Bessel filter. All experiments were performed at 22°C. Currents were recorded in the perforated-patch mode.36 Pipettes had resistances ranging from 1 to 3 MΩ. Perforated-patch pipettes were dipped for 2 to 3 seconds in Cs+ pipette solution containing (in mmol/L) CsCl 110 and tetraethylammonium (TEA) chloride 20 (to block endogenous K+ currents), EGTA 10, MgCl2 2, HEPES 10, and CaCl2 1 (pH adjusted to 7.2 with CsOH). Pipettes were back-filled with the same solution containing 240 μg/mL amphotericin B. To record whole-cell current through the Ca2+ channel, Ba2+ (20 mmol/L) was used as the charge carrier, and the extracellular solution contained (in mmol/L) BaCl2 20, choline chloride 124, HEPES 10, and d-glucose 15 (pH adjusted to 7.4 with TEA-OH). Raw current values were normalized to cell capacitance (an index of cell size) and were expressed as current density (pA/pF) for comparisons. Cell capacitance ranged from 4 to 16 pF. Soluble peptides and antibodies (eg, LDV, LEV, and anti–α4 integrin) were applied locally to cells by using picospritzer pipettes (General Valve Corp).
α4 Integrin Expression by Arteriolar Vascular Smooth Muscle Cells In Situ
Vessels were fixed with 2% paraformaldehyde-PSS for 20 minutes. After two glycine-PSS washes, the vessels were treated for 2 hours with an anti–α4 monoclonal antibody (MAB1396Z), the anti-rat major histocompatibility complex class I antigen monoclonal antibody (Seikagaku), or FITC-conjugated secondary antibody alone. Primary antibodies were diluted (25 μg/mL) in buffer with 2% goat serum. Vessels were washed with PBS and with FITC-conjugated goat anti-mouse antibody (1 μg of antibody, DAKO Corp). Fluorescent images were acquired by use of the Scanalytics Cellscan imaging system (Scanalytics, CSP, Inc). Alternatively, in some experiments, the isolated arterioles were not fixed before incubation with the anti–α4 monoclonal antibody. In these experiments, confocal images (Meridian Ultima-Z 312, Laser Scanning Confocal System, Meridian Instruments) of the arterioles were obtained to visualize α4 integrin distribution. This was followed by a carboxyfluorescein dye exclusion imaging technique to visualize the vascular smooth muscle cells. The two sets of images were then overlaid to show α4 integrin distribution in relation to single smooth muscle cells.
Arteriolar diameter changes to peptides were quantified as a percentage of constriction normalized to starting diameter. Maximal constriction was set as the inner luminal diameter after the addition of 10 μmol/L phenylephrine. After each dose, arteriolar diameter was measured each minute for 7 minutes or until a steady-state diameter was maintained for 2 to 3 minutes. The bath was then washed with fresh PSS after completion of the concentration-response curve to remove any remaining peptide or protein fragments. The vessel was allowed to reequilibrate and return to the baseline diameter. The vessel was then incubated with either 50 μg/mL of the β3 monoclonal function–blocking antibody F11 (Pharmingen) or 25 μg/mL of MAB1396Z, the α4 monoclonal function–blocking antibody (Chemicon). The vessel was pretreated with the antibodies for 15 minutes, after which time the concentration-response curve was repeated to determine the effects of each respective antibody.
To investigate the involvement of Src family tyrosine kinases, isolated arterioles were incubated for 20 minutes with the Src family kinase inhibitor PP2 (1 μmol/L) or the inactive analogue PP3 (1 μmol/L) (Calbiochem). Similarly, isolated vascular smooth muscle cells were also incubated with PP2 (1 μmol/L for 20 minutes) to inhibit the Src family kinases or the inactive analogue PP3 (1 μmol/L for 20 minutes) as a control. A role for protein kinase C was investigated by using the protein kinase C (PKC) inhibitor calphostin C (Sigma). Isolated arterioles or isolated vascular smooth muscle cells were incubated with 1 μmol/L calphostin C for 30 minutes. The arteriolar vasomotor responses or cellular electrophysiological responses to LDV were recorded before and after treatment with the inhibitors.
All data are expressed as mean±SEM. Analyses of time-matched responses were performed by ANOVA for two repeated factors, followed by post hoc tests for differences or by an independent 2-tailed Student t test. In all statistical analyses, a value of P<0.05 was used as the level of probability indicating significance.
After addition of the LDV peptide, arterioles exhibited a slowly developing monophasic constriction that was maintained in the presence of the peptide. The constriction reached steady state within 7 to 8 minutes (Figure 1A) and was reversed by rinsing with fresh PSS, with the arterioles returning to their baseline diameter. The concentration-response relationships illustrating the vasoactive responses produced by the addition of LDV and its control peptide, LEV, are shown in Figure 1B. The control peptide, LEV, had no effect on vasomotor tone. LDV-containing peptides produced a significant concentration-dependent vasoconstriction (n=7, P<0.05) and, at the highest concentration (2.1×10−4 mol/L), resulted in a constriction of 19%.
To determine whether the release of an endothelial factor was contributing to or modulating the LDV-induced vasoconstriction, responses to LDV were compared in arterioles before and after removal of the endothelium (Figure 1C). There was no significant difference between the concentration-response curves to LDV in the presence or absence of the endothelium (n=6).
Specificity of the constrictor response to LDV for the α4 integrin was determined by pretreatment of the arterioles for 15 minutes with the anti-α4 (MAB1396Z) monoclonal function–blocking antibody (Figure 2A). The vasoconstrictor response to LDV was abolished by anti-α4 pretreatment (P<0.05). For comparison, pretreatment of the arterioles with an anti-β3 integrin function–blocking antibody (F11) had no significant effect on LDV-induced vasoconstriction (Figure 2B), indicating the specificity of the α4 antibody. Previously, this β3 antibody was used to block the vasodilatory effect of RGD peptides.2
The presence of the α4 integrin in isolated arterioles was demonstrated by fluorescence immunohistochemical labeling of intact arterioles (Figure 3). Specific punctate staining (single-headed arrow) for α4 integrin was visible along the vascular smooth muscle cell margin. Vascular smooth muscle cells were oriented perpendicular to the long axis of the vessel (double-headed arrows) (Figure 3A). The secondary antibody alone failed to show any specific labeling (Figure 3B). Likewise, anti-rat major histocompatibility complex class I antibody failed to label the arteriole in a specific fashion (data not shown).
Measurements of [Ca2+]i in arterioles loaded with fura 2 (Figures 4A and 4B) indicated that the LDV-induced vasoconstriction was associated with an increase in vascular smooth muscle [Ca2+]i. After the addition of 1×10−4 mol/L LDV, the arterioles were observed to constrict by ≈18%, whereas [Ca2+]i was increased by ≈10%. To determine whether the LDV vasoconstriction involved Ca2+ entry through voltage-operated L-type channels, the arterioles were treated with nifedipine (1 μmol/L) before the addition of LDV-containing peptides. Nifedipine completely abolished the LDV-induced vasoconstriction (Figure 5).
Direct evidence of the involvement of L-type Ca2+ channels was obtained by using the patch-clamp technique. Inward Ba2+ currents through L-type Ca2+ channels were obtained by using voltage ramps (−100 to 80 mV, duration 200 ms) or voltage steps (−80 to 80 mV in 10-mV increments, duration 100 ms). The whole-cell Ba2+ currents peaked at 30 mV (range −3.9 to −8.9 pA/pF) and were blocked by nifedipine (1 μmol/L). The application of peptides containing LDV (100 μmol/L) resulted in an enhanced current through the L-type channels (Figure 6A) and caused downward displacement of the current-voltage relationship (Figure 6B). Modulation of the current over the entire voltage range indicated that the LDV peptides did not shift the sensitivity of the channel to voltage. Figure 6C shows a time course of peak Ba2+ currents after the application of LDV compared with basal current during the course of an experiment. To address the specificity of the effect of LDV-containing peptides on the L-type channel current, studies were repeated in the presence of peptides containing the substituted sequence, LEV (100 μmol/L). LEV had no effect on L-type current (Figure 8). Further experiments in the presence of a function-blocking antibody to the α4 integrin (10 μg/mL) indicated that the enhancement of L-type channels by LDV was inhibited by preincubation of the vascular smooth muscle cell for 1 minute with the antibody. The soluble α4 antibody alone had no significant effect on the current (Figure 8).
Additional experiments to address the mechanism for the effects of LDV indicated that the vasoconstrictor response and enhancement of L-type current were tyrosine kinase dependent, inasmuch as both were inhibited by the Src family kinase inhibitor PP2 but were not affected by the inactive analogue, PP3, used as a control (Figures 7 and 8⇓). By comparison, inhibition of PKC was unable to inhibit the vasoconstrictor response or the enhancement of L-type current in response to LDV. However, PKC inhibition did attenuate basal vascular tone and basal current (Figures 7 and 8⇓). These effects of PKC inhibitors are consistent with a role for PKC in the regulation of the L-type Ca2+ channel but not in the regulation of the channel by α4β1 integrin ligands.
In the present study, we have shown that the LDV peptides that are synthesized to mimic the CS-1 (alternately spliced) region of fibronectin cause vasoconstriction in isolated resistance arterioles. The vasoconstriction induced by LDV peptides is unaffected by removal of the endothelium and is inhibited by blocking the α4 integrin subunit with a monoclonal function–blocking antibody, supporting an interaction with the vascular smooth muscle cell α4β1 integrin. The mechanism of the vasoconstriction involves an increase in vascular smooth muscle Ca2+ that is dependent on the influx of Ca2+ through voltage-operated L-type channels. These data collectively support a role for the regulation of Ca2+ entry by integrin-linked signaling pathways.
Evidence of a link between integrins and the regulation of vascular tone comes from several laboratories that have reported that peptides containing the RGD integrin–binding motif have the ability to alter vascular tone.2–7⇓⇓⇓⇓⇓ The rationale for assuming a vasoregulatory role is that many of the integrin-linked cell signaling pathways are similar to those linked to the regulation of smooth muscle contractile function. The cell signaling pathways linked to control of the influx of Ca2+ are good examples. At present, αvβ3 and α5β1 integrin ligands have both been reported to alter vasomotor tone in arterioles and to modulate the influx of Ca2+ through L-type channels.13,14,37⇓⇓ The αvβ3 integrin reportedly reduces the influx of Ca2+ and elicits vasodilatation, whereas the α5β1 integrin enhances the influx of Ca2+ and produces vasoconstriction.13,14⇓ Recent investigations of the mechanism responsible for coupling the L-type channel to the α5β1 integrin indicate that a tyrosine phosphorylation cascade involving focal adhesion kinase and c-Src is important.14 In the present study, we report another integrin and integrin-binding motif, α4β1 and LDV, respectively, that are capable of enhancing the influx of Ca2+ through L-type channels and causing vasoconstriction. Furthermore, the link between α4β1 and the L-type channel involves an Src family kinase cascade similar to the α5β1 integrin.14
The ability of integrins to play a role in the regulation of arteriolar vasomotor tone could be of particular relevance in local vasoregulatory responses to mechanical forces, such as those imposed by changes in intravascular pressure or fluid shear stress. Thus, an involvement of integrins in the vascular myogenic response8 and flow-dependent vascular responses5,6⇓ may be at the level of the putative “mechanosensor.” Because integrins link the ECM to the cellular actin cytoskeleton at membrane focal contacts/dense plaques and because these sites act as a site for the assembly of a number of signaling pathway components, they are strategically placed to act as a mechanotransducing element. The dependence of the vascular myogenic response on the influx of Ca2+ through voltage-operated L-type channels38 and the ability of LDV to enhance Ca2+ entry and cause vasoconstriction could be interpreted to indicate that the α4β1 integrin is involved in the regulation of myogenic tone.
Presumably, it would be an interaction of the α4β1 integrin with fibronectin or osteopontin that would be important for mechanotransduction, because these matrix ligands are natural ECM ligands for this integrin. However, it is not clear whether the alternately spliced CS-1–containing forms of fibronectin or osteopontin are normally present in the vascular wall matrix. Another intriguing possibility is that the α4β1 integrin may be modulating myogenic function through cell-cell interactions involving VCAM-1 rather than through a cell-ECM interaction. Although VCAM-1 is known to be expressed by vascular smooth muscle, most evidence suggests that its expression is important during vascular development, after vascular wall injury, and after stimulation by various cytokines.15,39–43⇓⇓⇓⇓⇓ Consequently, the importance of its expression in normal vascular smooth muscle cells is less certain.
We have previously hypothesized that the vasomotor responses mediated by integrins may be particularly important during tissue and/or vascular injury. This hypothesis is based on the existence of biologically active cryptic sites within ECM molecules (matricryptic sites) that are capable of interacting with integrins.9 These sites are normally not exposed until such time that an injury occurs, during which time ECM enzymatic breakdown, denaturation, multimerization, and/or adsorption leads to the exposure or production of fragments (matricryptins) containing matricryptic sites.9–12⇓⇓⇓ These injury-derived signals would then participate in a variety of injury/repair events, including alterations in vasomotor tone. Regarding the α4β1 integrin, several observations point to a role in tissue injury/repair processes. First, the α4β1 integrin is upregulated on vascular smooth muscle cells15 during injury. Second, expression of the alternately spliced CS-1–containing variants of fibronectin,44–46⇓⇓ osteopontin,19 and VCAM-1 are increased after injury.40–42⇓⇓ Thus, interactions among these molecules may take on more importance during vascular responses to injury.
In conclusion, the present study indicates a unique ability of the α4β1 integrin to alter vasomotor tone through the modulation of membrane Ca2+ conductance. We propose that interactions between the α4β1 integrin and the ECM or other adhesion molecules are important for vascular control, particularly in tissue-injury responses. Alterations in the α4β1 integrin and ECM protein expression as well as enzymatic liberation of matricryptins may be important factors in the induction of vascular dysfunction after cardiovascular injury.
This study was supported by NIH grants HL-58690 and HL-62863 to Dr Meininger, HL-46502 to Dr M.J. Davis, and HL-59971 to Dr G.E. Davis. The authors thank Tracy Crow for her excellent technical assistance.
Original received July 10, 2001; resubmission received December 5, 2001; revised resubmission received January 17, 2002; accepted January 17, 2002.
- ↵Mogford JE, Platts SH, Davis GE, Meininger GA. Vascular smooth muscle αvβ3 integrin mediates arteriolar vasodilation in response to RGD peptides. Circ Res. 1996; 79: 821–826.
- ↵Muller JM, Chilian WM, Davis MJ. Integrin signaling transduces shear stress-dependent vasodilation of coronary arterioles. Circ Res. 1997; 80: 320–326.
- ↵Davis MJ, Wu X, Nurkiewicz TR, Kawasaki J, Davis GE, Hill MA, Meininger GA. Integrins and mechanotransduction of the vascular myogenic response. Am J Physiol. 2001; 280: H1427–H1433.
- ↵Giannelli G, Falk-Marzillier J, Schiraldi O, Stetler-Stevenson WG, Quaranta V. Induction of cell migration by matrix metalloprotease-2 cleavage of laminin-5. Science. 1997; 277: 225–228.
- ↵Wu X, Mogford JE, Platts SH, Davis GE, Meininger GA, Davis MJ. Modulations of calcium current in arteriolar smooth muscle by αvβ3 and α5β1 integrin receptor ligands. J Cell Biol. 1998; 143: 241–252.
- ↵Wu X, Davis GE, Meininger GA, Wilson E, Davis MJ. Integrin-dependent regulation of L-type calcium channel by α5β1 integrin requires signaling between focal adhesion proteins. J Biol Chem. 2001; 276: 30285–30292.
- ↵Duplaa C, Couffinhal T, Dufourcq P, Llanas B, Moreau C, Bonnet J. The integrin very late antigen-4 is expressed in human smooth muscle cell: involvement of α4 and vascular cell adhesion molecule-1 during smooth muscle differentiation. Circ Res. 1997; 80: 159–169.
- ↵Wayner EA, Garcia-Pardo A, Humphries MJ, McDonald JA, Carter WG. Identification and characterization of the T lymphocyte adhesion receptor for an alternative cell attachment domain (CS-1) in plasma fibronectin. J Cell Biol. 1989; 109: 1321–1330.
- ↵Makarem R, Humphries MJ. LDV: a novel cell adhesion motif recognized by the integrin α4β1. Biochem Soc Trans. 1991; 19: 380S.Abstract.
- ↵Bayless KJ, Meininger GA, Scholtz JM, Davis GE. Osteopontin is a ligand for the α4β1 integrin. J Cell Sci. 1998; 111: 1165–1174.
- ↵Hines KL, Kulkarni AB, McCarthy JB, Tian H, Ward JM, Christ M, McCartney-Francis NL, Furcht LT, Karlsson S, Wahl SM. Synthetic fibronectin peptides interrupt inflammatory cell infiltration in transforming growth factor β1 knockout mice. Proc Natl Acad Sci U S A. 1994; 91: 5187–5191.
- ↵Ager A, Humphries MJ. Use of synthetic peptides to probe lymphocyte–high endothelial cell interactions: lymphocytes recognize a ligand on the endothelial surface which contains the CS1 adhesion motif. Int Immunol. 1990; 2: 921–928.
- ↵Norton PA, Hynes RO. In vitro splicing of fibronectin pre-mRNAs. Nucleic Acids Res. 1990; 18: 4089–4097.
- ↵Kornblihtt AR, Vibe-Pedersen K, Baralle FE. Human fibronectin: cell specific alternative mRNA splicing generates polypeptide chains differing in the number of internal repeats. Nucleic Acids Res. 1984; 12: 5853–5868.
- ↵Clark RA, Quinn JH, Winn HJ, Lanigan JM, Dellepella P, Colvin RB. Fibronectin is produced by blood vessels in response to injury. J Exp Med. 1982; 156: 646–651.
- ↵Bowersox JC, Sorgente N. Chemotaxis of aortic endothelial cells in response to fibronectin. Cancer Res. 1982; 42: 2547–2451.
- ↵Norris DA, Clark RA, Swigart LM, Huff JC, Weston WL, Howell SE. Fibronectin fragment(s) are chemotactic for human peripheral blood monocytes. J Immunol. 1982; 129: 1612–1618.
- ↵Komoriya A, Green LJ, Mervic M, Yamada SS, Yamada KM, Humphries MJ. The minimal essential sequence for a major cell type-specific adhesion site (CS1) within the alternatively spliced type III connecting segment domain of fibronectin is leucine-aspartic acid-valine. J Biol Chem. 1991; 266: 15075–15079.
- ↵Ishizaka H, Kuo L. Acidosis-induced coronary arteriolar dilation is mediated by ATP-sensitive potassium channels in vascular smooth muscle. Circ Res. 1996; 78: 50–57.
- ↵Rae JL, Fernandez J. Perforated patch recordings in physiology. News Physiol Sci. 1991; 6: 2273–2277.
- ↵Hill MA, Zou H, Potocnik SJ, Meininger GA, Davis MJ. Arteriolar smooth muscle mechanotransduction: Ca2+ signaling pathways underlying myogenic reactivity. J Appl Physiol. 2001; 91: 973–983.
- ↵Lawson C, Ainsworth ME, McCormack AM, Yacoub M, Rose ML. Effects of cross-linking ICAM-1 on the surface of human vascular smooth muscle cells: induction of VCAM-1 but no proliferation. Cardiovasc Res. 2001; 50: 547–555.
- ↵Clark RA, Quinn JH, Winn HJ, Lanigan JM, Dellepella P, Colvin RB. Fibronectin is produced by blood vessels in response to injury. J Exp Med. 1982; 156: 646–651.