Isoflurane Activates PKC and Ca²⁺-Calmodulin-dependent Protein Kinase II via  MAP Kinase Signaling in Cultured Vascular Smooth Muscle Cells

The following chemicals were supplied by RBI/Sigma (Natick, MA) or Calbiochem-Novabiochem (San Diego, CA): KN-62 and KN-93 (competitive inhibitors of CaMKII at the calmodulin site); PD98059 (a MEK inhibitor); bisindolylmaleimide I-HCl, an inhibitor of Ca²⁺-dependent PKC (cPKC) and Ca²⁺-independent PKC (nPKC); Go-6976 (an inhibitor of cPKC); PKC activators, including phorbol-12-myristate-13-acetate (PMA) and phorbol 12,13-dibutyrate (PDBu); and protease inhibitors. Polyclonal antibodies against phosphorylated or nonphosphorylated ERK1/2, cPKCα, βI, and βII, and nPKCϵ were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). For Western blots, secondary anti-rabbit or anti-goat antibodies labeled with horseradish peroxidase (Sigma, St. Louis, MO) were detected by luminescence (Amersham Pharmacia Biotech, Piscataway, NJ). Dulbecco's modified Eagle media (DMEM) were from Life Technologies (Rockville, MD). Other reagents were of analytical grade.

Cultured vascular smooth muscle cells (VSMCs) were derived from rabbit femoral arteries using cell migration. 15Male New Zealand white rabbits (1.5–2.0 kg) were killed using a captive bolt pistol, followed by exsanguination. Handling and killing of rabbits were approved by the University of Washington Animal Care Committee (Seattle, WA). Femoral arteries were isolated, cut into segments of 1–2 mm, placed in DMEM with 20% fetal bovine serum, and incubated at 37°C in 5% CO². VSMCs began to migrate from the suspended explants and adhere in 5–10 days. Cells were maintained in DMEM containing 10% fetal bovine serum and passaged every 3–5 days. VSMCs were identified by immunohistochemical method with antibody to smooth muscle cell-specific α actin (DAKO EPOS anti-human smooth muscle actin/HRP; DAKO A/S, Carpinteria, CA) and desmin (monoclonal mouse anti-human desmin; DAKO A/S) (fig. 2). Cells from the second week of migration (> 99% smooth muscle cells;fig. 2) were used for subsequent passages.

For experimental studies, VSMCs from the 5th–10th passages were grown to 80–90% confluence and starved in 0.05% fetal bovine serum–DMEM for 48 h to arrest growth. Before experimentation, the VSMCs in 35- or 100-mm culture dishes were placed at room temperature for 30 min.

The effects of isoflurane on the quiescent VSMCs were studied in four experimental procedures. In the first procedure, VSMCs at room temperature were immersed in fresh DMEM containing 3% isoflurane for 30 s and 1, 2, 5, 10, 30, and 60 min (time course) or at various concentrations of 0, 1, 3, and 5% isoflurane for 10 min (dose response). In the second procedure, VSMCs were treated with 1 μm PDBu during the 48-h starvation period to down-regulate PKC, 16and 3% isoflurane was tested. In the third procedure, VSMCs were treated with the inhibitors of PKC (bisindolylmaleimide and Go6976) or CaMKII (KN-93 and KN-62) for 30 min at room temperature before administration of 3% isoflurane. In the fourth procedure, the localization of PKC isoforms (α and ϵ) in VSMCs and the effect of 3% isoflurane on the translocation of PKC were examined. For positive control of PKC translocation, VSMCs were treated with the activator phorbol-12-myristate-13-acetate (PMA) for 30 min.

Isoflurane saturated in DMEM (approximately 1.8 ml for a 35 mm-dish and approximately 10 ml for a 100-mm dish) was quickly administered into covered cultured dishes for a specific period of time. The partial pressure of isoflurane in DMEM, expressed as a percentage of 1 atmosphere, was assayed by gas chromatography. 17 

Vascular smooth muscle cells used in the first three experimental procedures were rapidly frozen in 2-propanol–dry ice, suspended in a lysate buffer with 0.2% Triton X-100, and sonicated. The lysate buffer contained 250 mm sucrose, 5 mm EDTA, 10 μg/ml aprotinin, 20 μg/ml soybean tyrosine inhibitor, 100 μg/ml PMSF, 100 μg/ml leupeptin, 10 mm HEPES (pH 7.4), and 5 μm each of phosphoserine, phosphothreonine, phosphotyrosine, β-glycerophosphate, p-nitrophenylphosphate, and sodium orthovanadate. 11 

Homogenates were centrifuged at 14,000 g  for 30 min. The supernatants (samples) from the homogenates were assayed for protein concentrations using bicinchoninic acid (BCA) reagents (Pierce Chemical Co., Rockford, IL) and stored at −70°C.

In each experiment, samples were used at equal concentrations of total protein, which ranged from 20 to 30 μg. Proteins were separated by sodium-dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes. Membranes were treated with anti-phospho-ERK1/2 antibody (1:1,500) followed by horseradish peroxidase–conjugated anti-mouse antibody (1:5,000). Total ERK1/2 was used to validate changes in phosphorylated ERK1/2 affected by isoflurane and was measured from the same membrane after stripping and staining with anti-ERK1/2 antibody (1:1,500) followed by anti-goat antibody (1:5,000). Immunoreactive bands were visualized using chemiluminescence (ECL Amersham Pharmacia Biotech, Piscataway, NJ, and Santa Cruz Biotechnology, Inc.) and evaluated with image analysis software. The amount of phosphorylated ERK1/2 was expressed as a percentage of the control.

In the fourth experimental procedure, treated VSMCs were frozen in cooled 2-propanol–dry ice, suspended in lysate buffer (as described for ERK1/2 immunoblotting), and sonicated. Homogenates were centrifuged at 1,000 g  for 30 min, the supernatant was collected and centrifuged at 10,000 g  for 30 min, and the cytosolic fraction was separated from the resulting supernatant by further centrifugation at 46,000 g  for 60 min. The resulting supernatant contained the cytosolic fraction. Membrane fractions prepared from the pellets were resuspended in 50 μl lysate buffer containing 1% Triton X-100, incubated on ice for 30 min, and centrifuged at 12,000 g  for 30 min. The resultant supernatants were retained as the membrane fraction. The samples were kept at −70°C until used for immunoblotting of PKC isoforms.

The protein quantification, separation by SDS-PAGE, and immunoblotting of PKC fractions were as described for ERK1/2 immunoblotting, except primary polyclonal antibodies were against specific isoforms of PKC (α and ϵ at 1:1,000), and secondary antibodies were anti-rabbit (1:5,000).

Test results were expressed as a percentage of the controls in mean and SD. Analysis of variance was used to compare between test results and the control, and P < 0.05 was considered statistically significant. 18 

Cultured smooth muscle cells obtained by the cell migration method 15were examined by immunocytochemical staining of desmin (fig. 2, top right, desmin-staining [second week]) and smooth muscle–specific actin (fig. 2, bottom). During the first 2 weeks of migration, cells migrating from femoral arterial segments contained more than 90% smooth muscle cells (first week, fig. 2). Cells cultured from the second and third week of migration consisted of more than 99% smooth muscle cells, which was confirmed with angiotensin II–induced smooth muscle–specific increases in phosphorylated ERK1/2 (data not shown). This study used the 5th–10th passages of cells from the second week of migration because cells from the third week of migration proliferated more slowly.

The increase of ERK1/2 phosphorylation (pERK1/2) by 3% isoflurane reached a plateau at 5–10 min (fig. 3A) and was dose dependent (fig. 3B). Based on immunoblot analysis, total ERK1/2 was relatively constant throughout each experiment (e.g. , tERK1/2, fig. 3A) despite the effect of isoflurane on pERK1/2 under various conditions (figs. 3–6, data not shown).

The increase in pERK1/2 by isoflurane was completely inhibited by 10 μm PD98059, a specific inhibitor of MEK (fig. 4) and was also reduced in the presence of 10 μm bisindolylmaleimide, an inhibitor of both cPKC and nPKC (fig. 5A). In contrast, the increase of pERK1/2 by isoflurane was not significantly affected by the inhibitor specific for cPKC, Go6976 (fig. 4B). The PMA-mediated increase in pERK1/2 by PKC was significantly reduced by Go6976 (fig. 5B) and bisindolylmaleimide (data not shown).

Another PKC activator, PDBu (1 μm) also increased pERK1/2 when administered for 30 min, but pERK1/2 was reduced when PDBu was used to pretreated cells for 48 h (down-regulation of PKC) (fig. 6). This reduction in pERK1/2 mediated by PDBu-induced down-regulation of PKC also prevented the increase in pERK1/2 by isoflurane (fig. 6). The inhibitors of CaMKII, KN-93 (fig. 7A) or KN-62 (fig. 7B), reduced the increase in pERK1/2 by 3% isoflurane.

One of the characteristics of protein kinase signaling is translocation of the kinase (for review, see Cobb 1). The increase in pERK1/2 by isoflurane was blocked by the inhibitor of c/nPKC (bisindolylmaleimide) but not by the inhibitor of cPKC (Go6976), suggesting that nPKC but not cPKC is activated by isoflurane.

To examine this possibility, we determined whether nPKC but not cPKC would be translocated from the cytosol to the membrane on activation. The PKC activator PMA (1 nm–0.1 μm) increased the membrane fraction and decreased the cytosolic fraction of both PKCα and PKCϵ (fig. 8) in a dose-dependent manner. Isoflurane increased PKCϵ in the membrane fraction (fig. 9A) throughout the time course (0.5–60 min), but not PKCα (fig. 9B). This isoflurane-increased PKCϵ in the membrane fraction was dose dependent (membrane, fig. 10). However, PKCϵ in the cytosol fraction was not significantly decreased (cytosol, fig. 10).

This study showed that isoflurane increased phosphorylated extracellular signal–regulated kinases (pERK1/2) in cultured smooth muscle cells from rabbit femoral arteries and that the increase was dose dependent and sustained. Isoflurane-increased pERK1/2 was initiated upstream by activation of CaMKII and Ca²⁺-independent protein kinase C (nPKC), which was followed by translocation of the nPKCϵ from the cytosol to the membrane.

Extracellular signal–regulated kinases (ERK1/2) belong to a superfamily of MAP kinases (for review, see Cobb 1) that have an important role in cellular functions, including proliferation, apoptosis, cell migration, and possibly smooth muscle contraction. The increase in pERK1/2 by isoflurane via  PKC and CaMKII suggests that isoflurane activates ERK1/2. The ability of the inhibitor (PD98059) of MAP kinase kinase (MEK, the ERK1/2 upstream kinase) in blocking the isoflurane-increased pERK1/2 suggests that it is mediated through MEK–ERK signaling pathway.

This study also shows that cultured smooth muscle cells from femoral arteries retain components of ERK1/2 signaling pathway affected by isoflurane. This conclusion is based on the similarities in response to inhibitors of PKC and CaMKII of isoflurane-increased pERK1/2 in cultured smooth muscle cells of this study, as well as the isoflurane-increased force in skinned strips observed in femoral arteries. 6The isoflurane-increased pERK1/2 and force are both blocked by the inhibitor of PKC (bisindolylmaleimide, which inhibits both c/nPKC) but not by a specific inhibitor of cPKC (Go6976), suggesting that isoflurane-increased pERK1/2 and force are mediated by activation of nPKC. The activation of nPKC is further confirmed by the ability of isoflurane to cause an increase in nPKCϵ in the membrane fraction. The evidence of isoflurane-increased nPKCϵ in the membrane fraction suggests the translocation of nPKCϵ from the cytosolic fraction to the membrane fraction. No statistically significant decrease in the cytosolic nPKCϵ could be attributed to that the change by isoflurane is too small to be detectable compared to the large cytosolic nPKCϵ. The fact that isoflurane-induced nPKCϵ translocation and isoflurane-increased pERK1/2 are long-lasting shown in this study is consistent with the sustained force increased by isoflurane observed in skinned strips, 6suggesting that the increased force is by ERK1/2 signaling. These findings show the ability to use cultured cells of arterial origin to elucidate the molecular mechanisms underlying anesthetic effects on force in skinned strips.

The blockade of isoflurane effect by down-regulation of PKC substantiates a role for PKC in pERK1/2 16increased by isoflurane. Whether small G proteins, such as ras–raf, act upstream of MEK/ERK1/2 signaling (for review, see Cobb 1) remains to be examined. This isoflurane–PKCϵ–MEK–ERK1/2 pathway may account for the isoflurane-increased force observed in skinned femoral arterial strips. 6 

The evidence that the inhibitors of CaMKII block the isoflurane-increased pERK1/2 suggests an activation of CaMKII. This CaMKII–ERK1/2 signaling, affected by isoflurane, has also been shown in cultured smooth muscle cells from rat aorta 12and rabbit aorta. 11Whether this CaMKII–ERK1/2 signaling pathway contributes to isoflurane-increased force in skinned femoral arterial strips, 6as observed in agonist-induced force generation, 13,14remains to be examined. However, the activation of CaMKII by isoflurane has been implicated in skinned femoral arterial strips 6but by a different mechanism, i.e. , phosphorylation of MLCK resulting in desensitization of the contractile proteins. 9,19Thus, CaMKII may have at least two downstream targets in vascular smooth muscle, namely, MLCK and ERK1/2. Whether these two mechanisms contribute to muscle contraction and relaxation remains to be examined.

The downstream targets of isoflurane-activated PKC could be by ERK1/2–MLCK–MLC (fig. 1). A ras–ERK–MLCK pathway has been associated with cell migration shown in urokinase-type plasminogen activator–stimulated cells. 20Thus, activation of CaMKII by isoflurane may also be via  ERK1/2–MLCK, shown to contribute partially to force generation in skinned femoral arterial strips. 6The persistence of activation of ERK1/2 by isoflurane shown in this study could account for the sustained force development by isoflurane in skinned femoral arterial strips. 6This speculation is based on the characteristics of CaMKII (for review 21), i.e. , CaMKII is initially activated by Ca²⁺–calmodulin followed by Ca²⁺-independent autophosphorylation.

The cascade of isoflurane effect in cultured vascular smooth muscle cells leading to increased ERK1/2 phosphorylation can therefore be hypothesized as follows (fig. 1): (1) isoflurane causes Ca²⁺release from the sarcoplasmic reticulum, which then activates CaMKII, leading by one pathway to phosphorylation of MLCK as shown in skinned arterial strips, 6and by another pathway to ERK1/2 signaling, as shown in this study; (2) isoflurane activates PKCϵ, leading to its translocation from the cytosol to the membrane, transducing the signal by the MEK–ERK1/2 pathway. This activation of CaMKII and PKC via  ERK1/2 signaling in cultured cells could lead to phosphorylation of MLCK, resulting in increased MLCK activity as shown in other cell types 20,22and in chicken gizzard. 23This CaMKII or PKC–MEK–ERK1/2–MLCK signaling in cultured cells could be responsible for isoflurane-increased force shown in skinned arterial strips. 6It is known that ERK1/2 signaling may result in cellular proliferation or transcription (for review, see Cobb 1). Whether this ERK1/2 signaling increased by isoflurane in cultured cells results in cellular proliferation and transcription remains to be determined. In summary, this study shows that isoflurane-activated ERK1/2 in cultured smooth muscle cells is initiated by nPKC and CaMKII.

The authors thank OHMEDA Inc. (Liberty Corner, NJ) for isoflurane and M. Dana Ravyn, Ph.D. (Senior Medical Writer and President ,  Quark Medical Communications, Piscataway, NJ), for editorial assistance.