Role of PKC in Isoflurane-induced Biphasic Contraction in Skinned Pulmonary Arterial Strips

Isoflurane was supplied by OHMEDA Inc. (Liberty Corner, NJ). Go6976 and bisindolylmaleimide I-HCl (PKC inhibitors) and phorbol-12,13-dibutyrate (PDBu) (a PKC activator) were purchased from Calbiochem (La Jolla, CA) or RBI (Natick, MA). Other chemicals were of analytical or reagent grade. Stock solutions of Go6976 (1 mm) were made in 100% dimethyl sulfoxide (DMSO). Final concentrations of 0.1–1.0 μm Go6976 contained 0.1% DMSO, and 10 μm Go6976 contained 10% DMSO. The same concentrations of DMSO were used in parallel for time-controls.

The method of preparing skinned arterial strips has been described. 5,11Male New Zealand white rabbits (2.2–2.5 kg) were killed using a captive bolt pistol. This method has been approved by University of Washington Animal Care Committee (Seattle, WA).

Right pulmonary arteries were rapidly isolated. The arterial rings, approximately 0.7 ± 0.2 mm in width, were placed in relaxing solution (contained no added Ca²⁺in basic solution) and denuded by gently rubbing the endothelium with a glass rod. Rings were opened, mounted on two pairs of forceps with one end attached to a photodiode tension transducer, and then stretched to a resting tension of 50 mg. The strips were immersed sequentially in the following solutions: (1) saponin (300 μg/ml) in relaxing solution for 4.5 min to make the sarcolemma permeable (“skinned”); (2) caffeine (25 mm) in relaxing solution to release Ca²⁺from the sarcoplasmic reticulum (SR); (3) 10 μm Ca²⁺buffered with 7 mm EGTA to activate the contractile proteins; and (4) relaxing solution to wash away Ca²⁺in the strips.

Experiments were performed at room temperature (21 ± 2°C). The tension was recorded on a G3 Apple Computer (Cupertino, CA) with a customized LabVIEW software program interfaced with a multifunction input/output board with 16-bit resolution (NB-MIO-16XL; National Instrument, Austin, TX).

Basic solution contained varying concentrations of free Ca²⁺(or pCa =−log [Ca²⁺] (m)) buffered with 7 mm EGTA, 2 mm MgATP²⁻, 0.1 mm free Mg²⁺, 15 mm creatine phosphate, 70 mm, K⁺plus Na⁺(from Na²ATP and Na²CP), 20 mm methanesulfonate (major anion), and 50 mm PIPES (piperazine-N,N′-bis[2-ethanesulfonic acid] dipotassium), 300 mOsm, with a pH of 7.00 ± 0.02 (21°C).

pCa buffer, containing free Ca²⁺concentrations of 0.316 μm (pCa 6.5 buffer) or 0.5 μm (pCa 6.3 buffer), was used to induce a minimum of 10 mg force (approximately 10–30% of maximum force based on previously established pCa-tension curves). We found that there was no significant difference in the responsiveness of the skinned strips to the effect of isoflurane with either pCa 6.5– or pCa 6.3–induced force. Isoflurane was saturated in pCa buffer, and the partial pressure of isoflurane in the buffer, expressed as a percentage of 1 atmosphere, was assayed by gas chromatography. 12 

Skinned strips were activated by either pCa 6.3 or PDBu in pCa 6.5 until force reached a steady state (control result). A fresh solution saturated with various concentrations of isoflurane (0, 1, 3, or 5%) was then substituted (test solution). The force (test result) after isoflurane administration was recorded for up to 60 min (fig. 1).

Protein kinase C inhibitors (Go6976 for cPKC 13and bisindolylmaleimide for c/nPKC 14) from 0.01 to 10 μm were preincubated in a relaxing solution, the subsequent pCa or PDBu contracting solutions, and the contracting solutions with isoflurane. When Go6976 was tested, parallel experiments were performed with DMSO (0.1 and 1%) to serve as a time-control. The test results were observed at various time intervals up to 60 min and expressed as a percentage of the control.

The change in force was measured from baseline to steady state force (control), or to the force present after various time intervals up to 60 min after administration of isoflurane either in the pCa 6.3–6.5 or the PDBu solutions. For the study of isoflurane dose–response relation, test solutions included various isoflurane concentrations. When the effects of the PKC inhibitors were tested, three responses were examined: (1) 0% isoflurane and 0 μm inhibitor (time control for isoflurane); (2) 3% isoflurane and 0 μm inhibitor (time control for each concentration of the inhibitor); and (3) various concentrations of the inhibitor and 3% isoflurane for the study of the inhibitor.

The test force was expressed as a percentage of the control force. Mean and SD from at least three arterial strips and three separate animals were expressed. Results from the strips treated with the inhibitors and the time controls were compared by Student t  test for paired data using the StatVIEW software program (BrainPower, Inc., Calabasas, CA), and two-factorial analysis of variance was used to compare different concentrations of inhibitors or isoflurane. A P  value of less than 0.05 was considered significant. 15 

We found that isoflurane increased (within 5 min) and decreased (between 15 and 60 min) the Ca²⁺- or PDBu-induced force (fig. 1), depending on experimental conditions (isoflurane concentrations, PDBu, pCa buffer, PKC inhibitors with or without DMSO). However, the isoflurane effects were relatively consistent within each animal. Therefore, the results were expressed as “peak contraction” for isoflurane-induced peak force and “late relaxation” for isoflurane-induced decreased force at a later phase when time controls reached approximately 50–70% of control force within 30 min for most experiments. However, the test results (isoflurane with or without the inhibitors) for late relaxation were analyzed at the same time periods (e.g. , 30 min, fig. 1) as those of the time control, within a series of experiments in each animal.

We found that isoflurane initially increased and later decreased the pCa-induced force (fig. 1). Isoflurane-increased force peaked between 5 and 10 min, and the decrease in force peaked between 15 and 60 min. As shown in figure 2, isoflurane dose-dependently increased force in pCa-activated strips, reaching a peak at approximately 5 min (mean ± SD; n = 3; 105.7 ± 6.4, 115.2 ± 4.9, 132.7 ± 16.6, and 148.1 ± 17.0 for 0, 1, 3, and 5% isoflurane, respectively). The increase in force was statistically significant at 3 and 5% isoflurane. However, 30 min after isoflurane administration, the force was significantly decreased at 3 and 5% isoflurane compared with that of the time controls (n = 3; 71.7 ± 4.3, 69.7 ± 8.2, 52.0 ± 4.9, and 40.7.1 ± 9.9 for 0, 1, 3, and 5% isoflurane, respectively), and the decrease was dose dependent (fig. 2).

We next tested the hypothesis that the biphasic effect of isoflurane on pCa-induced force was due to activation of PKC via  different signaling pathways. PKC inhibitors were used at concentrations from 0.1 to 10 μm, which were 10- to 1,000-fold the concentration (< 10 nm) required to inhibit 50% of PKC in vitro . 13,14 

We found that isoflurane-induced decrease in force was significantly prevented by 0.3 μm bisindolylmaleimide compared with that of 0 μm bisindolylmaleimide (n = 5; 63.4 ± 7.8, 27.4 ± 10.1, 42.8 ± 23.0, 54.4 ± 30.4, and 40.0 ± 13.0 for time control [0 μm bisindolylmaleimide] without isoflurane, 0, 0.1, 0.3, and 1 μm bisindolylmaleimide in the presence of 3% isoflurane, respectively;fig. 3A). The isoflurane-induced late relaxation was also significantly blocked by 0.1 and 0.3 μm Go6976 compared with that of 0 μm Go6976 (n = 4; 55.7 ± 9.2 for time control [0% isoflurane and 0 μm Go6976], 28.3 ± 4.4, 41.3 ± 7.8, 41.9 ± 6.8, and 31.5 ± 16.6 for 0, 0.1, 0.3, and 1 μm Go6976 in the presence of 3% isoflurane, respectively;fig. 3B).

Higher concentrations of bisindolylmaleimide (> 3 μm) did not affect the isoflurane-induced relaxation (data not shown). In contrast, isoflurane did not induce relaxation in the presence of 1% DMSO and the force was significantly decreased by 10 μm Go6976 (data not shown). However, peak force induced by isoflurane was significantly decreased by 3 and 10 μm bisindolylmaleimide (fig. 4A) but not by 10 μm Go6976 (fig. 4B).

Based on our results, we further hypothesized that increased force by isoflurane results in part from activation of nPKC (figs. 4A and B), whereas decreased force results in part from activation of cPKC (figs. 3A and B). A differential activation of nPKC would prevent the isoflurane-induced increase in force, and the isoflurane-decreased force would readily be demonstrated. We used PDBu in pCa 6.7 buffer to activate PKC and generate force. Isoflurane at 3 and 5% again increased the PDBu-induced force (peak contraction, fig. 5A), which was decreased between 30 and 60 min (late relaxation, fig. 5B).

We further examined whether there was a differential activation of PKC isozymes in PDBu-induced force, using the c/nPKC inhibitor bisindolylmaleimide and the cPKC inhibitor Go6976. As shown in figure 6, we found that bisindolylmaleimide dose-dependently inhibited the PDBu-induced force (n = 3–5; 92.2 ± 3.4, 76.8 ± 12.0, 33.6 ± 16.1, and 17.9 ± 2.1 for 0, 0.1, 3, and 10 μm bisindolylmaleimide, respectively;fig. 6A), but Go6976 did not significantly inhibit PDBu-activated force, except at 30 μm (fig. 6B).

That Go6976 at low concentrations did not significantly affect the PDBu-induced force (fig. 6B) suggests that cPKC does not significantly contribute to the force generation. We next assessed whether isoflurane-induced relaxation was mediated by activation of cPKC by using the cPKC-inhibitors Go6976 and bisindolylmaleimide. In this study, pCa 6.5 buffer was used in PDBu-induced force because of consistency of the force generation. Under this experimental condition (PDBu in pCa 6.5 buffer), 3% isoflurane did not significantly affect the PDBu-induced force up to 60 min in the presence of 0.1% DMSO (control vehicle for Go6976). However, in the presence of isoflurane, the PDBu-induced force was significantly enhanced by 0.3 μm Go6976 either at 5 min (peak contraction, fig. 7A) or 30 min (late relaxation, fig. 7B).

In the absence of 0.1% DMSO, we found that isoflurane also did not increase the PDBu-induced force. However, in these circumstances, force was also not significantly affected by the PKC inhibitor bisindolylmaleimide. Isoflurane did decrease the PDBu-induced force at a later phase (n = 3; 75.8 ± 3.3 and 42.6 ± 5.2 for 0 and 3% isoflurane, respectively), which was again significantly prevented by 0.1 μm bisindolylmaleimide (n = 3; 42.6 ± 5.2, 35.6 ± 8.6, 48.3 ± 5.0, and 56.7 ± 5.0 for 0, 0.01, 0.03, and 0.10 μm bisindolylmaleimide, respectively).

The main findings of this study are as follows. (1) In skinned pulmonary arterial strips, isoflurane induced biphasic effects at clinically relevant concentrations. Therefore, isoflurane and halothane 4have similar effects on pulmonary artery smooth muscle. These effects were characterized by increases followed by decreases in Ca²⁺-activated or PDBu-induced force. (2) Isoflurane-induced increases in force were blocked by inhibition of c/nPKC bisindolylmaleimide but not by cPKC Go6976. (3) Isoflurane-induced decreases in force were prevented by both inhibitors.

The consistent isoflurane-induced relaxation at the later phase suggests that the anesthetic-induced PKC signaling pathway is triggered by the initial anesthetic concentration, which was not significantly changed within 10 min. The isoflurane-induced relaxation prevented by a threefold lower concentration of Go6976 (0.1 μm) than bisindolylmaleimide (0.3 μm) agrees with the median inhibitory concentration (IC⁵⁰) shown in vitro  for cPKC (2.3–6.2 nm for Go6976 13and 8.40-20 nm for bisindolylmaleimide 14). That the isoflurane-induced relaxation is more sensitive to the effect of bisindolylmaleimide in PDBu (0.1 μm bisindolylmaleimide) than that in pCa (0.3 μm bisindolylmaleimide)-induced force is in agreement with the suggestion of a predominant activation of nPKC by PDBu. This speculation is derived from the evidence of direct inhibition of PDBu-induced force with c/nPKC inhibitor (bisindolylmaleimide) but little with cPKC inhibitor (Go6976). In contrast, the higher concentrations of cPKC inhibitor, Go6976, are required to prevent the decreases in force by isoflurane in PDBu- than in pCa-induced force. These findings could be due to the vehicle (DMSO) used in Go6976, possibly by DMSO inducing Ca²⁺release from the SR (data not shown), resulting in higher activation of cPKC and thus less effectiveness of Go6976. This speculation remains to be confirmed.

The narrow range of effective concentrations of the PKC inhibitors could be due to the dual signaling pathways of PKC leading to contraction and relaxation as proposed previously (fig. 1) 6and discussed later in this article. The partial prevention of relaxation by the PKC inhibitors suggests that mechanisms other than PKC also contribute to the isoflurane effects (e.g. , CaMKII 16).

The biphasic effects of isoflurane and the differential effect of the inhibitors from this study therefore suggest that PKC isoforms (nPKC and cPKC) have different roles in vascular smooth muscle contraction. nPKC and cPKC seem to be involved in contraction and relaxation of vascular smooth muscle, respectively. The biphasic responses of pulmonary arteries to isoflurane and halothane 5were similar, suggesting anesthetics have comparable effects in the same arterial type. Whether halothane has the same mechanisms of action as isoflurane remains to be seen. In contrast, halothane and isoflurane have been shown to cause opposite effects on force induced by oleic acid (a cytosolic PKC agonist) in coronary arteries. In this case, halothane abrogates force, whereas isoflurane enhances force. 3This apparent contradiction could be explained by differences in experimental conditions, arterial types, animal species, mechanisms, or by a differential role for PKC isoforms and requires further study.

A role for nPKC in vascular smooth muscle contraction is implied by the fact—shown in this and other studies—that peak force induced by isoflurane is blocked by high concentrations of bisindolylmaleimide, a c/nPKC inhibitor, but not by Go6976, an inhibitor of cPKC. 4,5Furthermore, Horowitz et al.  17have shown in single skinned vascular smooth muscle cells that the ϵ isoenzyme of nPKC induces a Ca²⁺-independent contraction. Kitazawa et al.  10have also shown a component of Ca²⁺-independent protein kinase C–induced contractile Ca²⁺sensitization in Triton X-100–treated rabbit arterial smooth muscle.

Inhibition of isoflurane-induced relaxation by the c/nPKC (bisindolylmaleimide) or specific cPKC (Go6976) inhibitor at low concentrations in both pCa- and PDBu-induced force suggests that relaxation is mediated through activation of cPKC by isoflurane. This possibility is consistent with evidence that in tracheal smooth muscle, purified cPKCα, as well as Ca²⁺–calmodulin-dependent protein kinase II, phosphorylates MLCK, which would result in decreased MLCK activity and MLC phosphorylation 7,16leading to decreased force.

The effectiveness of low concentrations (0.1–1 μm) of PKC inhibitors in preventing isoflurane-induced relaxation also suggests the specificity of the inhibition because in vitro , their IC⁵⁰is near 10 nm. 13,14It is also possible that PKC at low activity induces relaxation that is sensitive to lower concentrations of the inhibitors. However, PKC at high activity induces contraction via  a different pathway (see Zhong and Su, 6 fig. 1) (discussed in the following two paragraphs) that can only be blocked by high concentrations of the inhibitor. The ineffectiveness of higher concentrations of PKC inhibitors (≤ 1 μm) in preventing the isoflurane-induced relaxation (fig. 3) could be that both relaxation and increased force (fig. 4A) by isoflurane are affected by the high concentrations of PKC inhibitors, resulting in no change in isoflurane-induced relaxation.

The enhanced isoflurane-induced relaxation by 10 μm Go6976 observed in this study could be due to inhibition of both pathways (see Zhong and Su, 6 fig. 1), resulting in further decreases in force. However, this speculation cannot reconcile with the lack of inhibition of the isoflurane-increased force by Go6976. Whether DMSO (the vehicle for Go6976)-induced Ca²⁺release from the SR (data not known) has a role remains to be examined.

The results of this study suggest the biphasic effects of isoflurane can be viewed within the framework of the following model (see Zhong and Su, 6,fig. 1). 1Isoflurane directly activates PKC (nPKC). This is followed by translocation of nPKC from the cytosol to the membrane shown in cultured cells. 6The translocated nPKC would then initiate signals, either by CPI-17 10or by ERK–MLCK–MLC 18,19signaling pathways, resulting in force generation (e.g. , the isoflurane-increased force). Whether isoflurane induces nPKC translocation via  ERK–MLCK–MLC signaling pathways remains to be determined. 2Isoflurane induces Ca²⁺release from the SR. This increased Ca²⁺activates Ca²⁺-dependent protein kinases, including cPKC. The increased cPKC activity would then directly phosphorylates MLCK, decreasing MLCK activity and decreasing MLC phosphorylation, 7resulting in decreased force (isoflurane-induced relaxation) (see Zhong and Su, 6,fig. 1). This is consistent with the observations that halothane increases purified cPKC activity 20and cytosolic cPKC activity in rat cerebrocortical synaptosomes. 21 

Alternately, cPKC signaling could also be via  another pathway, ERK–MLCK–MLC signaling 8,9,18,19resulting in increased force (isoflurane-increased force) (see Zhong and Su, 6,fig. 1). The fact that isoflurane increases PDBu-induced force under pCa 6.7 (fig. 6A) but not under pCa 6.5 (fig. 7A) also supports this speculation. It is speculated that under low free Ca²⁺buffer (pCa 6.7), PDBu activates predominantly nPKC so that further activation of cPKC by isoflurane would increase the PDBu-induced force (shown with isoflurane-induced peak contraction). However, under higher free Ca²⁺buffer (pCa 6.5), cPKC would be partially activated by PDBu so that further activation of cPKC by isoflurane would result only in relaxation. If isoflurane-increased force were the result of nPKC activation, as suggested from the preferential inhibitory effect of bisindolylmaleimide (c/nPKC inhibitor) but not Go6976 (cPKC inhibitor), 4,5,6the PDBu-induced force increased by isoflurane would be independent of pCa in the buffer. Thus, activation of cPKC by isoflurane-induced Ca²⁺release from the SR could be via  two different pathways: ERK–MLCK–MLC signaling leading to contraction and MLCK–MLC signaling leading to relaxation. This speculation remains to be examined.

The clinical implication of the results from this study is that isoflurane may cause transient pulmonary vasoconstriction followed by vasodilation in patients. Isoflurane may induce sustained pulmonary vasodilation in those patients with low SR Ca²⁺stores (e.g. , newborns) or with low PKC activity. If patients with pulmonary hypertension have low PKC activity, isoflurane could have some beneficial effects on pulmonary artery pressures.

In summary, isoflurane causes effects similar to those of halothane in pulmonary arteries. Isoflurane results in increased Ca²⁺-activated force, followed later by decreased force, which may be in part mediated by activation of nPKC and cPKC, respectively.

The authors thank OHMEDA Inc. (Liberty Corner, NJ) for isoflurane, M. Dana Ravyn, Ph.D. (Senior Medical Writer and President, Quark Medical Communications, Piscataway, NJ), for editorial assistance, and Alec Rooke, M.D., Ph.D. (Department of Anesthesiology, University of Washington, Seattle, WA), for discussions.