Chronological Changes of Arterial Diameter, cGMP, and Protein Kinase C in the Development of Vasospasm
Background and Purpose We hypothesized that nitric oxide exerts a negative feedback control on protein kinase C (PKC) activation, and the disturbance of the feedback control after subarachnoid hemorrhage results in vasospasm due to PKC activation. This study was undertaken to verify this hypothesis.
Methods Different dogs were prepared for three separate experiments: measurement of the angiographic diameter of the basilar artery and determination of cGMP and PKC activity in vascular smooth muscle cells. In each experiment, two models were used: the single-hemorrhage model for mild vasospasm and the two-hemorrhage model for severe vasospasm. In both models, chronological changes of these three parameters were examined from day 1 until day 7.
Results In the single-hemorrhage model, mild vasospasm and a slight decrease of the cGMP level were noted on day 4, then both returned to the baseline levels on day 7. PKC activity was slightly enhanced throughout the study period. In the two-hemorrhage model, severe vasospasm and a significant decrease of the cGMP level were observed on day 5 and persisted until day 7. PKC activity was remarkably enhanced from day 5 until day 7. The differences between the two models with regard to the three parameters were statistically significant.
Conclusions The decrease of cGMP level and the enhancement of PKC activity were obviously associated with the development of severe vasospasm. We conclude that subarachnoid hemorrhage disturbed the feedback control exerted by nitric oxide on PKC activation, leading to PKC-dependent vasospasm.
The pathophysiological mechanism of vasospasm after SAH remains unclear. Based on the results of an isometric tension study, we have reported that canine basilar arteries develop tonic and long-lasting contractions mediated by PKC activation.1 Furthermore, we compared the PKC activity in vascular smooth muscle cells between the control artery from the intact dog and the vasospastic artery after experimental SAH. PKC activity was significantly enhanced in the vasospastic artery. Therefore, we concluded that PKC played a pivotal role in the development of vasospasm after SAH.
As for the vascular endothelium, it has been postulated that it regulates the vascular tone through an EDRF,3 4 and one of those EDRFs was identified as NO.5 6 NO induces the activation of soluble guanylate cyclase in the vascular smooth muscle cells, resulting in the increase of cGMP in the vascular smooth muscle. The increased level of cGMP induces vascular relaxation.7 8 9 10
It has not been clarified yet how both systems, PKC and NO-cGMP, interact to maintain an appropriate cerebral vascular tone. We hypothesized that the NO-cGMP system exerts a negative feedback control on the activation of PKC in vascular smooth muscle cells and that this feedback mechanism is disturbed after SAH, resulting in a pathological vascular contraction such as vasospasm. We have performed an isometric tension study to test our hypothesis and obtained results that support this hypothesis (S.N. et al, unpublished data, 1994). To clearly establish our hypothesis, we examined the chronological changes of the angiographic diameter of canine basilar arteries, the cGMP level, and PKC activity during severe vasospasm and mild vasospasm and investigated the correlations among these parameters.
Materials and Methods
Mongrel dogs of either sex weighing 7 to 10 kg were used. The basilar artery and a piece of cerebellar tissue from each dog were used for the in vitro studies.
All experiments were performed according to the guiding principles for the care and use of animals in the field of physiological sciences established by the Physiological Society of Japan. All chemicals were purchased from Sigma unless otherwise noted.
Three separate groups of dogs were prepared for the following experiments: angiography and determination of cGMP and PKC in the vascular smooth muscle cells. In each experiment dogs were classified into two groups, and two models of SAH were prepared: the single-hemorrhage and the two-hemorrhage models. Six dogs were used for each group in the angiography experiment. For the assays of cGMP and PKC, eight basilar arteries, in their whole length, taken from different animals were used for each determination point.
For the single-hemorrhage model, 3 mL of autologous blood was injected into the cisterna magna once on day 1, and the same amount of sterile physiological saline was injected on day 4. For the two-hemorrhage model, 3 mL of autologous blood was injected on days 1 and 4.11 Angiography was performed on day 1, before the first injection of blood; on day 4, before the injection of physiological saline in the single-hemorrhage model or before the second injection of blood in the two-hemorrhage model; and on days 5 and 7.
After injection of 25 mg/kg IV of sodium pentobarbital, a sterile catheter was cannulated into a vertebral artery through a femoral artery under fluoroscopy. The head of the dog was fixed in a stereotaxic frame, then 5 mL of iopamidole 300 (Schering AG) was injected. During angiography, the end-tidal CO2 was continuously monitored and was kept between 38 and 42 mm Hg.
A basilar artery on the angiogram was divided into three segments, then the diameter of the artery at the midpoint of each segment was carefully measured under a microscope, and the mean diameter was obtained. The mean diameter determined on the angiogram taken on day 1 was considered as 100.
Assay of cGMP
Buffered solution for dissection. The solution to dissect and prepare the artery in the chamber consisted of the following (mmol/L): Na+ 144.44, K+ 4.10, Cl− 135.02, Ca2+ 1.01, Mg2+ 1.19, PO42− 1.54, SO42− 1.19, HEPES 24.90, and glucose 10.0. The solution was aerated with 100% O2, and the pH of the solution was kept at 7.40.
Modified Krebs-Henseleit solution. The solution to perfuse the artery in the chamber consisted of the following (mmol/L): Na+ 144.44, K+ 4.10, Cl− 127.16, Ca2+ 2.49, Mg2+ 1.19, PO42− 1.54, SO42− 1.19, HCO3− 24.90, and glucose 5.00. The solution was aerated with a gas mixture containing 20% O2, 5.0% CO2, and 75% N2, and the pH was kept at 7.38±0.20.
Extraction of the sample. In these experiments, angiography was not performed to avoid unknown effects of the contrast material on cGMP determinations. Data from day 1 (before the first injection) and day 4 (before the second injection) were used for both the single- and the two-hemorrhage models.
Dogs were killed by injection of an overdose of sodium pentobarbital (50 mg/kg IV). The basilar artery was immediately excised together with the brain stem and immersed in the buffered solution for dissection containing 10−4 mol/L of IBMX, a phosphodiesterase inhibitor.12 The artery was dissected from the brain stem under a microscope. Blood inside the artery was gently washed out, and blood outside the artery was meticulously removed. The artery was perfused in a chamber with modified Krebs-Henseleit solution containing 10−4 mol/L of IBMX for 30 minutes. After the endothelium of the artery was gently rubbed off with a wire, the artery was frozen in liquid nitrogen for 30 seconds and minced in 6.0% trichloroacetic acid solution. After sonication of the tissue sample for 20 seconds six times, it was ultracentrifuged at 2000g and 4°C for 15 minutes. The supernatant, which contained the cytosol fraction of vascular smooth muscle cells, was extracted with water-saturated dimethyl ether four times and then lyophilized for 24 hours.
Assay of cGMP. The basal level of cGMP in vascular smooth muscle cells was measured with the use of an enzyme immunoassay system for cGMP (Amersham). Each sample was dissolved in 5.0 mL of the assay buffer solution provided in the assay system. One milliliter of each sample was acetylated with 33.3% triethylamine. Fifty-microliter aliquots of the standard sample of cGMP provided in the assay system and collected samples were incubated with 100 μL of rabbit anti-cGMP serum in the anti-rabbit IgG–coated microtiter plate for 2 hours at 4°C. Thereafter, 100 μL of peroxidase conjugate was added to each well, and the plate was incubated for 1 hour at 4°C. After each well was washed with the washing solution, 200 μL of tetramethylbenzidine was added to each well as the substrate. The microtiter plate was incubated for 30 minutes at room temperature, and the reaction was then stopped by adding 100 μL of 1.0 mol/L sulfuric acid. The absorbance of each well was read at 450 nm in a microtiter spectrophotometer. Each time, to compute the concentration of cGMP, a standard curve was obtained with the standard cGMP solution provided in this system. The concentration of protein in each sample was measured with the use of a protein assay system (Bio-Rad). The concentration of cGMP was expressed as femtomoles per microgram protein.
Assay of PKC Activity
Extraction buffer. The buffer consisted of the following: 25 mmol/L Tris-HCl, 2.0 mmol/L EGTA, 50 mmol/L 2-mercaptoethanol, 0.005% leupeptin, 0.25 mol/L sucrose, and 1.0 mmol/L phenylmethylsulfonyl fluoride. The pH was adjusted at 7.45.
Reaction mixture. The composition of the reaction mixture was as follows: 25.0 mmol/L Tris-HCl, 0.5 mmol/L MgCl2, 0.1 mmol/L ATP, 0.8 mmol/L CaCl2, and 50 μg/mL phosphatidylserine. The pH was adjusted at 7.0.
Extraction of the sample. In these experiments angiography was not performed to avoid unknown effects of the contrast material on the determination of PKC activity.
When PKC is activated, it translocates from the cytosol to the cell membrane.13 Therefore, we measured PKC activity in the membrane fraction of vascular smooth muscle cells. Data from day 1 (before the first injection) and day 4 (before the second injection) were used for both the single- and the two-hemorrhage models.
After we killed the dogs by injection of an overdose of sodium pentobarbital (50 mg/mL IV), a small piece of cerebellum and the basilar artery together with the brain stem were excised and immediately immersed in ice-cold PBS (pH 7.40). The artery was isolated from the brain stem under a microscope. Blood inside and outside the artery was meticulously removed. Endothelium was gently rubbed off with a wire. The piece of cerebellum and the artery were minced separately in ice-cold PBS and then centrifuged at 250g and 4°C for 5 minutes. After the supernatant was discarded, the pellet was resuspended in 1 mL of extraction buffer. The suspension was sonicated for 20 seconds six times, then ultracentrifuged at 100 000g and 4°C for 60 minutes. The supernatant of the sample of the cerebellum was collected as the cytosol fraction. One milliliter of the extraction buffer containing 0.1% Triton-X was added to the pellet of the sample of the basilar artery and incubated at 4°C for 60 minutes. Thereafter, it was ultracentrifuged at 100 000g and 4°C for 60 minutes. The supernatant contained the membrane fraction of vascular smooth muscle cells.
Assay of PKC activity. The assay of PKC activity was performed with the use of a nonradioisotopic protein kinase assay kit (NRPK assay kit, Medical & Biological Laboratories). To prepare the standard curve of PKC activity, the PKC activity in the cytosol fraction of the cerebellum obtained from each dog was used because there is no statistically significant difference between PKC activity in the cerebellum under normal conditions and after SAH.2
The sample from the cerebellum was diluted 2, 5, 10, 50, 100, 500, 1000, 104, and 105 times with the extraction buffer. Twelve microliters of each sample and 108 μL of the reaction mixture were mixed and preincubated in the polyvinyl plate at 25°C for 5 minutes. Then 100 μL of the reaction mixture was transferred to each well of a microtiter plate coated with the synthetic substrate of PKC and incubated at 25°C for 5 minutes. The reaction was stopped by adding stop solution, and each well was washed. Then 100 μL of mouse monoclonal antibody for the synthetic substrate was added to each well and incubated at 25°C for 30 minutes. After each well was washed, 100 μL of peroxidase-conjugated antibody solution was added, and the plate was incubated at 25°C for 60 minutes. After each well was washed, 100 μL of the substrate solution (o-phenylenediamine) was added for color development. After we stopped the reaction by adding stop solution, the absorbance of each well was read at 492 nm in a microtiter spectrophotometer. Then the concentration of protein in each sample was measured with the use of the Modified Lowry Protein Assay Reagent Kit (PIERCE) to standardize the data. A standard curve of PKC activity prepared with multiple diluted samples of the cytosol fraction of the cerebellum was used to compute the PKC activity of the membrane fraction of vascular smooth muscle cells. PKC activity in the membrane fraction of the basilar artery was expressed as a percentage of the cerebellar PKC activity.
Data were expressed as mean±SEM. The change of angiographic diameter in one model was statistically analyzed with the use of the paired t test. Differences between the two models with regard to angiographic data and levels of cGMP and PKC were analyzed with the use of the unpaired t test. A 95% confidence level was considered statistically significant.
Typical examples of changes of angiographic diameters in the single-hemorrhage and the two-hemorrhage models are shown in Figs 1⇓ and 2⇓, respectively. In the single-hemorrhage model, the diameters on days 4, 5, and 7 were 76.1±1.8%, 78.3±1.4%, and 83.0±1.9% of the baseline value, respectively. The difference between the diameter on day 4 and that on day 7 was statistically significant. In the two-hemorrhage model, the diameters on days 4, 5, and 7 were 75.2±2.1%, 48.2±2.2%, and 50.7±1.3%, respectively. The differences between the diameter on day 4 and that on day 5 and on day 7 were statistically significant. However, no significant difference was noted between the diameter on day 5 and that on day 7.
As for the arterial diameter on day 4, there was not a statistically significant difference between the two models. However, on days 5 and 7 there was a statistically significant difference in diameter between the two models.
Chronological Changes of the cGMP Level
As shown in Fig 3⇓, the levels of cGMP on days 1 and 4 were 443.7±90.2 and 261.9±43.8 fmol/μg protein, respectively. The difference between the two was statistically significant.
In the single-hemorrhage model, the levels of cGMP on day 5 and day 7 were 353.4±23.4 and 416.1±23.0 fmol/μg protein, respectively. The differences in cGMP level between days 1 and 5 and days 1 and 7 were not statistically significant.
In the two-hemorrhage model, the levels of cGMP on days 5 and 7 were 148.3±1.8 and 203.4±90.6 fmol/μg protein, respectively. The difference between the two values was not statistically significant. However, the difference between the level on day 4 and that on day 5 was statistically significant. Furthermore, the differences between days 1 and 5 and days 1 and 7 were statistically significant.
There was a statistically significant difference between the two models in the levels of cGMP on days 5 and 7.
Chronological Changes of PKC Activity
The PKC activity of the membrane fraction of vascular smooth muscle cells was 8.0±0.9% on day 1 and 21.2±5.2% on day 4. The difference between the two was statistically significant (Fig 4⇓).
In the single-hemorrhage model, the activity on day 5 was 22.7±0.4% and that on day 7 was 20.5±4.2%. There were no statistically significant differences among the activities detected on days 4, 5, and 7.
On the other hand, in the two-hemorrhage model the activity on day 5 was 51.4±2.1% and that on day 7 was 44.5±0.2%. The difference between the activities observed on day 4 and that on day 5 was statistically significant, as was the difference between days 4 and 7. The PKC activity on day 7 was significantly decreased compared with that detected on day 5.
There was a statistically significant difference between the two models in the level of PKC activity on days 5 and 7.
In both the single- and the two-hemorrhage models, the chronological changes in arterial diameter, cGMP level, and PKC activity showed very good correlation. After the second injection of blood, the level of cGMP decreased. PKC activity was enhanced, and angiograms showed very severe vasospasm in the two-hemorrhage model (Fig 5⇓). On the other hand, in the case of the very mild vasospasm observed in the single-hemorrhage model, the decrease of cGMP level was also very small, and PKC activity was not as enhanced as in the case of severe vasospasm (Fig 6⇓).
Our experimental results clearly demonstrate that the level of cGMP decreases and PKC is activated during the vasospasm observed after SAH. Appearance of vasospasm, decreased cGMP level, and enhanced PKC activity are not independent phenomena; on the contrary, they are closely interrelated. We have previously examined how the PKC and the NO-cGMP systems interact to maintain an appropriate vascular tone. The basilar artery from the intact dog showed a tonic and long-lasting contraction when NO synthesis was inhibited, and this contraction was PKC dependent. On the other hand, the artery from the two-hemorrhage model did not develop the contraction when NO synthesis was inhibited because NO synthesis had already been impaired by SAH. As for the resting tone, the artery from the two-hemorrhage model showed more enhanced tone than that from the intact dog, and the enhancement of vascular tone was also PKC dependent. Based on these experimental results, we deduced that the NO-cGMP system exerts a negative feedback control on PKC activation. When the negative feedback control is impaired after SAH, a pathological PKC-dependent tonic vascular contraction such as vasospasm appears. The results of the present study, which show a correlation among the chronological changes of these three parameters during vasospasm, support our hypothesis on the mechanisms involved in the development of vasospasm after SAH.
As for the interrelation between the PKC and the NO-cGMP systems, it has been reported that PKC regulates the overactivation of NO synthase and overproduction of cGMP.14 15 On the other hand, it has not been examined how the NO-cGMP system regulates PKC activity. In this study we have demonstrated another aspect of the relation between the NO-cGMP and the PKC systems, particularly with regard to the regulatory role of the endothelial NO on PKC activation in vascular smooth muscle cells.
Selected Abbreviations and Acronyms
|EDRF||=||endothelium-derived relaxing factor|
|PKC||=||protein kinase C|
This study was supported in part by a Nishio Tomoyuki memorial grant (B28-64) from the Ministry of Education, Science, and Culture of Japan.
Reprint requests to Shigeru Nishizawa, MD, PhD, Department of Neurosurgery, Hamamatsu University School of Medicine, 3600 Handacho, Hamamatsu, Shizuoka 431-31, Japan.
- Received March 21, 1995.
- Revision received June 21, 1995.
- Accepted June 26, 1995.
- Copyright © 1995 by American Heart Association
Furchgott RF. Role of endothelium in responses of vascular smooth muscle. Circ Res. 1983;53:557-573.
Ignaro LJ. Biological actions and properties of endothelium-derived nitric oxide formed and released from artery and vein. Circ Res. 1989;65:1-21.
Lincoln TM. Effects of nitroprusside and 8-bromocyclic GMP on the contractile activity of the rat aorta. J Pharmacol Exp Ther. 1983;224:100-107.
Nguyen BL, Saitoh M, Ware JA. Interaction of nitric oxide and cGMP with signal transduction in activated platelets. Am J Physiol. 1991;261:H1043-H1052.
Varsos VG, Liszczak TM, Han DH, Kistler JP, Vielma J, Black PM, Heros RC, Zervas NT. Delayed cerebral vasospasm is not reversible by aminophylline, nifedipine, or papaverine in a ‘two-hemorrhage’ canine model. J Neurosurg. 1983;58:11-17.
Martin W, Furchgott RF, Villani GM, Jothianandan D. Phospho- diesterase inhibitors induce endothelium-dependent relaxation of rat and rabbit aorta by potentiating the effects of spontaneously released endothelium-derived relaxing factor. J Pharmacol Exp Ther. 1986;237:539-547.
Murthy KS, Jin JG, Makhlouf GM. Inhibition of nitric oxide synthase activity in dispersed gastric muscle cells by protein kinase C. Am J Physiol. 1994;266:G161-165.