J Neurosurg 75:452-457, 1991
Induction of cytosolic free calcium elevation in rat vascular smooth-muscle cells by cerebrospinal fluid from patients after subarachnoid hemorrhage KATSUNOBU TAKENAKA,M.D., HIROMU YAMADA, M.D., NOBORU SAKAI, M.D., TAKASHI ANDO~ M.D., TOSH1HIKO NAKASHIMA~ M.D., AND YASUAK! NISHIMURA~ M.D. Deparlment of Neurosurgery, Gifu University' School of Medicine, Gifu, Japan ~" The purpose of this study was to determine the effects of cerebrospinal fluid (CSF) from patients with subarachnoid hemorrhage (SAH) on cytosolic free calcium in cultured rat vascular smooth-muscle cells using the fluorescent intracellular calcium indicator fura-2/AM. Samples of CSF were collected from 12 patients (seven with and five without vasospasm) on Days 2, 6, 11, and 16 after SAH. Control CSF samples were obtained from five patients 6 to 9 months after they had undergone successful aneurysm surgery following an SAH. All CSF samples in both the non-vasospasm and vasospasm groups, regardless of the day of sampling after the SAH, induced significantly higher transient intracellular calcium elevations when compared to levels induced by control CSF. Furthermore, the addition of 2 mM ethyleneglycol-bis (~-aminoethylether)-N,N'tetra-acetic acid (EGTA) caused a slight reduction in the peak height in the CSF-induced intracellular calcium rise which declined more rapidly to basal levels than those studied without EGTA. In the non-vasospasm group, the intracellular calcium concentration remained stable after SAH throughout the study period. In contrast, in the vasospasm group, this concentration was highest on Day 2 post-SAH, but sharply decreased on Day 6 and rose again on Day 11. This result correlated with the clinical signs of vasospasm in these patients. These findings indicated that the intracellular calcium elevations induced by CSF obtained after SAH were due to the combination of the influx of extracellular calcium and the mobilization of intracellular calcium from storage sites. The changes in intracellular calcium concentrations in vascular smooth-muscle cells induced by CSF obtained from patients on successive days following SAH suggest that the substances that induce this repeat calcium elevation on Day 11 post-SAH may be the key spasmogens for vasospasm after SAH. KEY WORDS cerebrospinal fluid smooth-muscle cell 9 v a s o s p a s m ~
EREBRAL vasospasm is a major complication in patients with subarachnoid hemorrhage (SAH) due to ruptured aneurysms) 4j5 Despite many investigations in the past, the identity of the factors involved in the production and maintenance of vasospasm is not known. J3,2,.3o Allen, et al.,' demonstrated that human cerebrospinal fluid (CSF), collected up to 17 days after SAH, produced large contractions of canine cerebral artery. Boullin, et al., -~6 reported that CSF from patients with SAH would contract human basilar artery if the CSF was obtained from a patient who was in vasospasm but would not cause contractions if obtained from a patient with no angiographic signs of vasospasm. The implications of this finding are considerable; it suggests strongly that a spasmogen is released in some patients with SAH and not in others. Therefore, it is important to examine the effects on vascular smooth-muscle cells of CSF obtained after SAH.
C
452
9
calcium rat
9
subarachnoid hemorrhage
In general, a contraction of vascular smooth-muscle cells seems to be regulated by changes in the concentration of cytosolic free Ca ++ which activates protein kinase C and calcium-calmodulin (CaM)-dependent protein k:inases. ~'23 In the search for effective therapy for cerebral vasospasm, the role of Ca ++ in the contraction of vascular smooth muscle and the involvement of certain spasmogens affecting Ca ++ transport have become important foci of attentionY 2 We recently reported that several spasmogens (oxyhemoglobin, endothelin, serotonin, prostaglandins, norepinephrine, and others) induce cytosolic free Ca ++ elevations causing muscle contraction in vascular smooth-muscle cells. 26 To our knowledge, lhe effects of CSF obtained after SAH on cytosolic free Ca ++ in vascular smooth-muscle cells have not been reported. It was the purpose of this study to determine whether CSF obtained on successive clays following the hemorrhage from patients with SAH
J. Neurosurg. / Volume 75/September, 1991
Bloody CSF-induced Ca ++ elevation in smooth-muscle cells causes cytosolic free Ca ++ elevations in single vascular smooth-muscle cells. If such an elevation was produced, our goal was to correlate the relationship of this elevation to the clinical and radiographic findings, the development of vasospasm, and the patient's outcome.
Materials and Methods
Source of Cerebrospinal Fluid Cisternal CSF samples were collected from two men and 10 women with SAH due to rupture of a cerebral aneurysm. The mean age (+__standard error of the mean (SEM)) of the patients was 58 __. 5 years. Clinical diagnosis was made by computerized tomography (CT) and the CT findings were classified according to the method of Fisher, et al. 9 Clinical assessment was performed according to the grading system of Hunt and Hess. ~ All of the patients underwent aneurysmal clipping and establishment of cisternal drainage; they were then treated with anticonvulsant agents, with no calcium antagonist or antifibrinolytic agents. Patients were selected for inclusion in the study by the following criteria: 1) they had undergone early surgery (within 48 hours after SAH); 2) they were in good clinical condition (Hunt and Hess Grade I to III on admission); and 3) they were classified as having Group 3 SAH on CT. The diagnosis of vasospasm was made based on the clinical signs (reduced level of consciousness and/or focal neurological deficits) and by angiography and CT evidence of local ischemia without rebleeding. Arterial spasm detected on angiography was classified into three groups according to the system of Fisher, et al.:9 1) no vasospasm; 2) slight to moderate vasospasm; and 3) severe vasospasm. The patients were divided into two groups: 1) those presenting without clinical and radiological evidence of vasospasm (non-vasospasm group), and 2) those presenting with evidence of vasospasm (vasospasm group). The non-vasospasm group included five women (mean age 61.2 years) without clinical or radiological evidence related to vasospasm. The vasospasm group included two men and five women (mean age 56.1 years) with clinical signs such as lethargy and/or hemiparesis. Five patients exhibited severe vasospasm on anglography and a low-density area on CT scans and two patients revealed moderate vasospasm. Collection of CSF Cisternal CSF samples were collected through cisternal drainage catheters by gentle aspiration. The first 2 to 3 ml were allowed to escape before the collection of 10 to 20 ml was begun. The day of aneurysm rupture was designated Day 0. Samples of CSF were collected from patients on Days 2, 6, 11, and 16 after SAH. Control cisternal CSF samples were obtained through cisternal drainage catheters from five patients (age 58 _+ 4 years) 6 to 9 months after undergoing successful aneurysm surgery following an SAH. J. Neurosurg. / Volume 75 / September, 1991
The CSF samples were collected and immediately placed in ice (-4~ then centrifuged at 1000 G for 5 minutes at 4"C. The supernatant was collected, frozen, and stored at -20"C until analysis.
Culture of Vascular Smooth-Muscle Cells Vascular smooth-muscle cells* were prepared from explants of the thoracic aorta obtained from 7-weekold male Sprague-Dawley rats by the method of Ross2j and cultured in Dulbecco's modified Eagle's medium (DMEM) plus 10% fetal calf serum (FCS) in a 10% CO2 atmosphere at 37"C, as described previously.I~The cells were passaged weekly and cells at passages nine to 12 were used for the current study. The cells were grown in serum-containing growth medium for 4 days to confluency, and were then starved of serum for 2 days prior to the addition of CSF. Measurement of Cytosolic Free Calcium Changes For measurement of cytosolic free Ca ++ ([Ca++]~) changes induced by CSF, vascular smooth-muscle cells were plated at a density of 2 x 104 cells/chamber on the glass coverslip that adheres to the smooth lower side of Flexiperm discs.t After culture for 48 hours at 37"C, the cells were deprived of serum as described above. Before the addition of CSF, the cells were washed twice and loaded with the fuorescence indicator fura-2/AM (2 t~M) for 40 minutes at 37~ in 0.3 ml of serum-free DMEM, then rinsed free of the extracellular dye and incubated for 15 minutes to allow de-esterification of the dye. The changes in [Ca++]~were determined in individual vascular smooth-muscle cells loaded with fura-2 by digital imaging of Ca+§ fluorescence.~/After the addition of 300 ul CSF from each patient, fluorescence images of emission wavelength of 500 nm were obtained at alternating excitation wavelengths of 340 and 360 nm through a Videcon camera every 5 seconds for single cells at 37~ The fluorescence emission intensity ratio of 340 nm/360 nm was calculated on the image processor after subtraction of background values. Statbtical Analysis The statistical evaluation in the non-vasospasm and vasospasm groups was carried out using the Wilcoxon/ Mann-Whitney U-test; comparison of values based on the number of days after SAH that CSF sampling was carried out was analyzed by the t-test with subsequent Bonferroni correction for multiple comparisons. The values were expressed as mean __ SEM. Values of p < 0.05 were considered significant. * Smooth-muscle cells obtained from Dr. H. Arita, Shionogi Research Laboratories, Osaka, Japan. t Flexiperm disc manufactured by Heraeus Biothechnology, Hanau, Germany. Argus-100/CA imaging processor manufactured by Hamamatsu Photonics, Hamamatsu, Japan. 453
K. Takenaka, et al.
FIG. 1. Effects of cerebrospinal fluid (CSF) from patients with subarachnoid hemorrhage (SAH) on cytosolic free Ca++ ([Ca++]~) level in fura-2-1oaded vascular smooth-muscle cells. The [Ca*+]~ levels were measured as described in the text. Traces shown were representative of [Ca++]i responses of individual cells (number of cells > 40). A: Results obtained when CSF (300 #1) from patients with SAH was added to vascular smooth-muscle cells. B: Results in vascular smoothmuscle cells pretreated with 2 mM EGTA for ! minute when CSF from patients with SAH was added.
Results As shown in Fig. 1 left, CSF from all patients obtained on Days 2, 6, 11, and 16 after SAH induced a transient elevation of cytosolic free calcium ([Ca++]i), with a peak attained within 10 seconds followed by a gradual decline within 3 minutes after exposure of CSF (in six tests each). The addition of 2 mM ethyleneglycolbis (~-aminoethylether)-N,N'-tetra-acctic acid (EGTA)
caused a slight reduction in the height of the CSFinduced [Ca++]i transient increase, and the [Ca++]~ elevation declined to the basal level more quickly (within 2 minutes, Fig. 1 right) compared to that in preparations without EGTA (in six tests each). Figure 2 shows the changes in [Ca+§ level and the greatest difference between the basal and peak levels in vascular smooth-muscle cells induced by CSF from each patient (A Ratio). The time courses after SAH for the A Ratios induced by CSF are depicted in Figs. 3 and 4. In the non-vasospasm group (Fig. 3 left), the A Ratio seemed to be stable during the entire period of study after SAH. No significant difference was observed between days after SAH (indicated as open columns in Fig. 4). In contrast, in the vasospasm group (Fig. 3 right) the ~x Ratio was revealed as a biphasic response, with the highest concentrations on Days 2 and 11 and the lowest concentrations on Days 6 and 16. In seven patients with vasospasm, the A Ratio was 0.179 _+0.017 on Day 2 after SAH, 0.080 _+ 0.010 on Day 6, 0.132 -4-_ 0.008 on Day 11, and 0.099 _ 0.009 on Day 16. On Day 2 after SAH the A Ratio was significantly higher than on Day 6 or 16 (p < 0.01), and on Day 11 it was also significantly higher than on Day 6 (p < 0.05, t-test with subsequent Bonferroni correction for multiple comparisons). This value is indicated as shaded col-
Fro. 2. The cytosolic free calcium ([Ca++]~)level was measured as described in the text. Cerebrospinal fluid (CSF, 300 ul) from one patient in the vasospasm group was added on Day 2 after subarachnoid hemorrhage to vascular smooth-muscle cells. Upper:The fluorescence photographs are representative of changes in [Ca++]i levels in vascular smooth-muscle cells at 0, 10, 75, and 200 seconds after the addition of CSF. The color scale on the right indicates the [Ca++]a level (ratio 340 nm/360 nm). Lower: Changes in [Ca++]~ levels. The difference in [Ca++]~between the basal and peak levels induced by CSF is termed "/x Ratio." The numbers refer to the photographs above. 454
J. Neurosurg. / Volume 75/September, 1991
Bloody CSF-induced Ca ++ elevation in smooth-muscle cells
FIG. 3. Time course for cytosolic free calcium ([Ca++]i) levels in vascular smooth-muscle cells induced by 300 ul cerebrospinal fluid (CSF) from post-subarachnoid hemorrhage (SAH) patients in the non-vasospasm (left) and the vasospasm (right) groups. The [Ca++]~levels were measured as described in the text. The ordinate represents the changes in the [Ca++], level shown as "4 Ratio" in Fig. 2. Five control CSF samples (300 ul each) were also added to vascular smooth-muscle cells. Data on each day after SAH of each patient were cumulative of six separate experiments and were presented as the mean _+ standard error of the mean for 40 cells. Left: Numbers 1 through 5 refer to patients in the non-vasospasm group. Right: Numbers 6 through 12 refer to patients in the vasospasm group. Asterisks denote onset of clinical signs due to vasospasm; squares denote the appearance of a low-density area on computerized tomography scans due to vasospasm.
umns in Fig. 4. On Days 6 to 11 after SAH, clinical observations of vasospasm were seen (Fig. 3 right). Significant differences were observed in the A Ratio between the non-vasospasm and vasospasm groups on Days 2 and 6 (0.120 _ 0.007 vs. < 0.179 + 0.017, respectively, on Day 2; 0.110 _ 0.006 vs. > 0.080 + 0.010, respectively, on Day 6). There were, however, no significant differences in /x Ratio between the two groups on Days 11 or 16 (Fig. 4). The degree of s Ratio induced by CSF from control patients was significantly lower than that from patients with SAH (p < 0.01), as shown in Fig. 4. Three control CSF samples did not induce a [Ca++]~ elevation, and two control CSF samples induced lower [Ca++]~ elevations (0.020 _+ 0.002 and 0.010 -4-0.005 of the A Ratio).
Discussion
Calcium and Smoolh-Muscle Contraction It is generally accepted that the contractile activity o f vascular smooth muscle is regulated by the concentration of intracellular free calcium ([Ca++]0. This contraction is initiated by a variety of external stimuli which cause the [Ca++]~ to rise to a critical level, resulting from the influx of extracellular [Ca++]~ and/or the mobilization from intracellular [Ca++]~ store sites. This in turn leads to the activation of protein kinase C and Ca*+-calmodulin (CaM)-dependent kinase. 23 Studies o f intracellular calcium changes have recently been facilitated by the novel calcium ion chelators (quin-2/AM, indo-1/AM, and fura-2/AM). Fura-2/AM is highly specific for Ca *+ ions, readily distinguishing them from other divalent cations. It is also sensitive to Ca +* changes in the nanomolar range that are encoun-
J. Neurosurg. / Volume 75/September, 1991
FIG. 4. The changes in cytosolic free calcium ([Ca++]31evel induced by cerebrospinal fluid (CSF) from patients in nonvasospasm (open columns), vasospasm (shaded columns), or control (filled column) group in vascular smooth-muscle cells. The [Ca++]ilevels were measured as described in the text. The ordinate represents the changes in [Ca++]~ level shown as "4 Ratio" in Fig. 2. Data are expressed as means _+standard error of the means for five non-vasospasm patients, seven vasospasm patients, and five control patients. Statistically significant differences between two values were indicated as p < 0.01 or p < 0.05. Statistically significant differences from control values were indicated by asterisks (p < 0.01). NS = no statistically significant difference; SAH = subarachnoid hemorrhage. 455
K. Takenaka, et al. tered within smooth-muscle cells. Fura-2 produces a potent fluorescent emission which allows its use in relatively low concentrations, thereby lowering any interference due to an inherent Ca ++ buffering capacity. Fura-2/AM is rapidly internalized by smooth-muscle cells. Once internalized, de-esterification of the acetoxymethyl group converts it to the free acid, which has minimal capability to diffuse out of the cell. Residual extracellular fura-2/AM, removed by a series of cell washings, assures that observed fluorescence changes in the cells correspond solely to alterations in intracellular free calcium. 2s CSF-Induced [Ca**]~ Elevation in Smooth-Muscle Cells
Bloody CSF collected from patients with SAH has been reported to produce sustained contractions of canine cerebral artery 19and human cerebral artery. 1,4.6,7 However, no study has been reported that reviews comparative data on [Ca++]i elevation in vascular smoothmuscle cells in response to CSF from patients with SAH on successive days following hemorrhage. In this experiment, CSF from all patients with SAH induced [Ca++]i elevations in single smooth-muscle cells. Furthermore, the addition of EGTA caused a slight reduction in the peak of the transient CSF-induced [Ca++]i elevation, but the peak [Ca++], declined more quickly to the basal level compared to the preparations without EGTA. These findings indicated thai the [Ca++]i elevations induced by CSF obtained after SAH were due to a combination of the influx of extracellular Ca ++ and the mobilization from an internal Ca ++ store site. The results support the proposition that some substances are released in the CSF of patients with SAH and produce large contractions of cerebral arteries.l'4-7'22'31 The changes in [Ca++]~ plotted over time after SAH in the vasospasm group showed a biphasic pattern, with the highest concentrations on Days 2 and 1 1 after SAH and the lowest on Days 6 and 16 after SAH. On Day 2, the [Ca++]~ elevation induced by CSF in the vasospasm group was significantly higher than that in the nonvasospasm group. In contrast, on Day 6 the [Ca++]~ elevation induced by CSF in the vasospasm group was significantly lower than that in the non-vasospasm group. As indicated above, the following possibilities may be considered: 1) On Day 2 after SAH, some unknown substances that induce [Ca§ elevations are released in the CSF of patients with SAH, and the concentration of these substances in CSF in patients with vasospasm is higher than in those without. 2) On Day 6, these substances in the CSF of the non-vasospasm group continue to flow out through a cisternal drainage catheter, whereas in the group with vasospasm they remain for a few days in the intracranial subarachnoid space. Thus, the [Ca++], elevation induced by CSF samples collected from non-vasospasm patients seemed to be stable after SAH throughout the study period in contrast to the [Ca++], elevation induced by CSF samples col456
lected from patients with vasospasm. 3) On Day 11, some substances which induce [Ca++]~ elevations may be produced in the CSF of patients with vasospasm, resulting in an elevation of [Ca++]~. These changes correlate with the clinical observation of cerebral vasospasm. Vasoactive Substances in CSF
Possible vasospasm-related factors include hemolysate substances such as oxyhemoglobin or bilirubin; vasoactive amines such as norepinephrine, serotonin neuropeptide Y, or endothelin; and arachidonic acid metabolites such as prostaglandin F,_,,,thromboxane A2, leukotriene C~, leukotriene D4, or hydroxyeicosa-tetraenoic acid. -~s~2`LT'~-~-'5'27'2oIt has recently been reported that prostaglandin F2, concentrations in CSF or plasma endothelin- 1 concentrations were significantly elevated in the presence of symptomatic vasospasm after SAHY'~ Despite extensive efforts, the vasoactive substances responsible for the contraction in vasospasm have not been identified. ~3.30 The results of this study of [Ca++]~ levels in vascular smooth-muscle cells induced by CSF from patients with SAH on successive days following the hemorrhage suggest that the re-elevation of [Ca++]~ on Day 11 after SAH correlates with the pathogenesis of delayed vasospasm. Some unknown substances that induce this reelevation in vascular smooth-muscle cells may be the spasmogens that cause vasospasm after SAH. Further investigation and analysis of these substances should be performed in an attempt to identify the spasmogens in CSF from patients following SAH by studying human cerebral arterial smooth-muscle cells. We are presently trying to purify, identify, and characterize these substances in the CSF from patients with SAH. Conclusions Samples of CSF obtained from patients following SAH induced significantly higher transient intracellular calcium elevations when compared to those induced by control CSF. In studies of samples from the vasospasm group, changes in intracellular free calcium elevations induced by CSF showed a biphasic response with the highest concentration on Days 2 and 11 after SAH. It is suggested that some substances that induce the reelevation of intracellular calcium concentrations in vascular smooth-muscle cells may be the key substances responsible for vasospasm. Acknowledgment We express our gratitude to Dr. K. Deguchi for his helpful discussion. References I. Allen GS, Gross CJ, French LA, et al: Cerebral arterial spasm: Part 5. In vitro contractile activity of vasoactive agents including human CSF on human basilar and anterior cerebral arteries. J Neurosnrg 44:584-600, 1976 2. Asano T, lkegaki I, Suzuki Y, et al: Endothelin and the J. Neurosurg. / Volume 75/September, 1991
Bloody CSF-induced Ca ++ elevation in smooth-muscle cells
3. 4.
5.
6.
7.
8. 9.
10.
1I. 12. 13. 14.
15. 16. 17.
production of cerebral vasospasm in dogs. Biochem Biophys Res Colnmun 159:1345-1351, 1989 Auer LM: Acute operation and preventive nimodipine improve outcome in patients with ruptured cerebral aneurysms. Neurosurgery 15:57-66, 1988 Boullin DJ, Brandt L, Ljunggren B, et al: Vasoconstrictor activity in cerebrospinal fluid from patients subjected to early surgery for ruptured intracranial aneurysms. J Neurosurg 55:237-245, 1981 Boullin D J, Hunt TM, Rogers AT: Models for investigating the aetiology of cerebral arterial spasm: comparative responses of the human basilar artery with rat colon, anoeoccygeus, stomach fundus, and aorta and guinea-pig ileum and colon. Br J Pharmaco163:251-257, 1978 Boullin D J, Mohan J, Grahame-Smith DG: Evidence for the presence of a vasoacfive substance (possibly involved in the aetiology of cerebral arterial spasm) in cerebrospinal fluid from patients with subarachnoid haemorrhage. J Neural Neurosurg Psychiatry 39:756-766, 1976 Brandt L, Ljunggren B, Andersson KE, et al: Effects of indomethacin and prostacyclin on isolated human pial arteries contracted by CSF from patients with aneurysmal SAH. J Neurosurg 55:877-883, 1981 Chehrazi BB, Giri S, Joy RM, et al: Prostaglandins and vasoactive amines in cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Stroke 20:217-224, 1989 Fisher CM, Kistler JP, Davis JM: Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery 6: I-9, 1980 Hanansaki K, Nakano K, Kasai H, et al: Specific receptors for thromboxane A2 in cultured vascular smooth muscle cell of rat aorta. Biochem Biophys Res Commun 150: 1170-1175, 1988 Hunt WE, Hess RM: Surgical risk as related to time of intervention in the repair of intracranial aneurysm. J Neurosurg 28:14-20, 1968 Juvela S, Kaste M, Hillbom M, et al: Platelet thromboxane release after subarachnoid hemorrhage and surgery. Stroke 21:566-57 l, 1990 Kassell NF, Sasaki T, Colohan ART, et al: Cerebral vasospasm following aneurysmal subarachnoid hemorrhage. Stroke 16:562-572, 1985 Kassell NF, Torner JC, Haley EC Jr, et al: The International Cooperative Study on the Timing of Aneurysm Surgery. Part l: Overall management results. J Neurosurg 73:18-36, 1990 Kassell NF, Tarrier JC, Jane JA, et al: The International Cooperative Study on the Timing of Aneurysm Surgery. Part 2: Surgical results. J Neurosurg 73:37-47, 1990 Masaoka H, Suzuki R, Hirata Y, et al: Raised plasma endothelin in aneurysmal subarachnoid haemorrhage. Lancet 2:1402, 1989 (Letter) Miao FJP, Lee TJF: Effects of bilirubin on cerebral arterial tone in vitro. J Cereb Blood Flow Metab 9:666-674, 1989
J. Neurosurg. / Volume 75/September, 1991
18. Nishizuka Y: Studies and perspectives of protein kinase C. Science 233:305-312, 1986 19. Okwuasaba FK, Weir BKA, Cook DA, et al: Effects of various intracranial fluid on smooth muscle. Neurosurgery 9:402-406, 198 l 20. Paoletti P, Gaetani P, Grignani G, et al: CSF leukotriene C4 following subarachnoid hemorrhage. J Neurosurg 69: 488-493, 1988 21. Ross R: The smooth muscle cell. II. Growth of smooth muscle in culture and formation of elastic fibers. 8 Cell Bioi 50:172-186, 1971 22. Sasaki T, Asano T, Takakura K, et al: Nature of the vasoactive substance in CSF from patients with subarachnoid hemorrhage. J Neurosurg 60:1186-1191, 1984 23. Somlyo AP, Himpens B: Cell calcium and its regulation in smooth muscle. FASEB J 3:2266-2276, 1989 24. Sundt TM Jr, Davis DH: Reaction of cerebrovascular smooth muscle to blood and ischemia: primary versus secondary vasospasm, in Wilkins RH (ed): Cerebral Arterial Spasm. Baltimore: Williams & Wilkins, 1979, pp 244-250 25. Suzuki Y, Sato S, Suzuki H, et al: Increased neuropeptide Y concentrations in cerebrospinal fluid from patients with aneurysmal subaraehnoid hemorrhage. Stroke 20: 1680-1684, 1989 26. Takenaka K, Yamada H, Sakai N, et al: Cytosolic calcium changes in cultured rat aortic smooth-muscle cells induced by oxyhemoglobin. J Neurusurg 74:620-624, 1991 27. Tanishima T: Cerebral vasospasm: contractile activity of hemoglobin in isolated canine basilar arteries. J Neurasurg 53:787-793, 1980 28. Tsien RY, Rink T J, Poenie M: Measurement of cytosolic free Ca2+ in individual small cells using fluorescence microscopy with dual excitation wavelengths. Cell Calcium 6:145-157, 1985 29. Watanabe T, Asano T, Shimizu T, et al: Participation of lipoxygenase products from arachidonic acid in the pathogenesis of cerebral vasospasm. J Neuroehem 50: 1145-1150, 1988 30. Wellum GR, Peterson JW, Zervas NT: The relevance of in vitro smooth muscle experiments to cerebral vasospasm. Stroke 16:573-581, 1985 31. White RP, Macleod RM, Muhlbauer MS: Evaluation of the role hemoglobin in cerebrospinal fluid plays in producing contractions in cerebral arteries. Surg Neural 27: 237-242, 1987 32. Wilkins RH: Attempts at prevention or treatment of intracranial arterial spasm: an update. Neurosurgery 18: 808-825, 1986 Manuscript received September 23, 1990. Accepted in final form April 3, 1991. Address reprint requests to: Katsunobu Takenaka, M.D., Department of Neurosurgery, Gifu University School of Medicine, 40 Tsukasamachi, Gifu 500, Japan.
457
Induction of cytosolic free calcium elevation in rat vascular smooth-muscle cells by cerebrospinal fluid from patients after subarachnoid hemorrhage KATSUNOBU TAKENAKA,M.D., HIROMU YAMADA, M.D., NOBORU SAKAI, M.D., TAKASHI ANDO~ M.D., TOSH1HIKO NAKASHIMA~ M.D., AND YASUAK! NISHIMURA~ M.D. Deparlment of Neurosurgery, Gifu University' School of Medicine, Gifu, Japan ~" The purpose of this study was to determine the effects of cerebrospinal fluid (CSF) from patients with subarachnoid hemorrhage (SAH) on cytosolic free calcium in cultured rat vascular smooth-muscle cells using the fluorescent intracellular calcium indicator fura-2/AM. Samples of CSF were collected from 12 patients (seven with and five without vasospasm) on Days 2, 6, 11, and 16 after SAH. Control CSF samples were obtained from five patients 6 to 9 months after they had undergone successful aneurysm surgery following an SAH. All CSF samples in both the non-vasospasm and vasospasm groups, regardless of the day of sampling after the SAH, induced significantly higher transient intracellular calcium elevations when compared to levels induced by control CSF. Furthermore, the addition of 2 mM ethyleneglycol-bis (~-aminoethylether)-N,N'tetra-acetic acid (EGTA) caused a slight reduction in the peak height in the CSF-induced intracellular calcium rise which declined more rapidly to basal levels than those studied without EGTA. In the non-vasospasm group, the intracellular calcium concentration remained stable after SAH throughout the study period. In contrast, in the vasospasm group, this concentration was highest on Day 2 post-SAH, but sharply decreased on Day 6 and rose again on Day 11. This result correlated with the clinical signs of vasospasm in these patients. These findings indicated that the intracellular calcium elevations induced by CSF obtained after SAH were due to the combination of the influx of extracellular calcium and the mobilization of intracellular calcium from storage sites. The changes in intracellular calcium concentrations in vascular smooth-muscle cells induced by CSF obtained from patients on successive days following SAH suggest that the substances that induce this repeat calcium elevation on Day 11 post-SAH may be the key spasmogens for vasospasm after SAH. KEY WORDS cerebrospinal fluid smooth-muscle cell 9 v a s o s p a s m ~
EREBRAL vasospasm is a major complication in patients with subarachnoid hemorrhage (SAH) due to ruptured aneurysms) 4j5 Despite many investigations in the past, the identity of the factors involved in the production and maintenance of vasospasm is not known. J3,2,.3o Allen, et al.,' demonstrated that human cerebrospinal fluid (CSF), collected up to 17 days after SAH, produced large contractions of canine cerebral artery. Boullin, et al., -~6 reported that CSF from patients with SAH would contract human basilar artery if the CSF was obtained from a patient who was in vasospasm but would not cause contractions if obtained from a patient with no angiographic signs of vasospasm. The implications of this finding are considerable; it suggests strongly that a spasmogen is released in some patients with SAH and not in others. Therefore, it is important to examine the effects on vascular smooth-muscle cells of CSF obtained after SAH.
C
452
9
calcium rat
9
subarachnoid hemorrhage
In general, a contraction of vascular smooth-muscle cells seems to be regulated by changes in the concentration of cytosolic free Ca ++ which activates protein kinase C and calcium-calmodulin (CaM)-dependent protein k:inases. ~'23 In the search for effective therapy for cerebral vasospasm, the role of Ca ++ in the contraction of vascular smooth muscle and the involvement of certain spasmogens affecting Ca ++ transport have become important foci of attentionY 2 We recently reported that several spasmogens (oxyhemoglobin, endothelin, serotonin, prostaglandins, norepinephrine, and others) induce cytosolic free Ca ++ elevations causing muscle contraction in vascular smooth-muscle cells. 26 To our knowledge, lhe effects of CSF obtained after SAH on cytosolic free Ca ++ in vascular smooth-muscle cells have not been reported. It was the purpose of this study to determine whether CSF obtained on successive clays following the hemorrhage from patients with SAH
J. Neurosurg. / Volume 75/September, 1991
Bloody CSF-induced Ca ++ elevation in smooth-muscle cells causes cytosolic free Ca ++ elevations in single vascular smooth-muscle cells. If such an elevation was produced, our goal was to correlate the relationship of this elevation to the clinical and radiographic findings, the development of vasospasm, and the patient's outcome.
Materials and Methods
Source of Cerebrospinal Fluid Cisternal CSF samples were collected from two men and 10 women with SAH due to rupture of a cerebral aneurysm. The mean age (+__standard error of the mean (SEM)) of the patients was 58 __. 5 years. Clinical diagnosis was made by computerized tomography (CT) and the CT findings were classified according to the method of Fisher, et al. 9 Clinical assessment was performed according to the grading system of Hunt and Hess. ~ All of the patients underwent aneurysmal clipping and establishment of cisternal drainage; they were then treated with anticonvulsant agents, with no calcium antagonist or antifibrinolytic agents. Patients were selected for inclusion in the study by the following criteria: 1) they had undergone early surgery (within 48 hours after SAH); 2) they were in good clinical condition (Hunt and Hess Grade I to III on admission); and 3) they were classified as having Group 3 SAH on CT. The diagnosis of vasospasm was made based on the clinical signs (reduced level of consciousness and/or focal neurological deficits) and by angiography and CT evidence of local ischemia without rebleeding. Arterial spasm detected on angiography was classified into three groups according to the system of Fisher, et al.:9 1) no vasospasm; 2) slight to moderate vasospasm; and 3) severe vasospasm. The patients were divided into two groups: 1) those presenting without clinical and radiological evidence of vasospasm (non-vasospasm group), and 2) those presenting with evidence of vasospasm (vasospasm group). The non-vasospasm group included five women (mean age 61.2 years) without clinical or radiological evidence related to vasospasm. The vasospasm group included two men and five women (mean age 56.1 years) with clinical signs such as lethargy and/or hemiparesis. Five patients exhibited severe vasospasm on anglography and a low-density area on CT scans and two patients revealed moderate vasospasm. Collection of CSF Cisternal CSF samples were collected through cisternal drainage catheters by gentle aspiration. The first 2 to 3 ml were allowed to escape before the collection of 10 to 20 ml was begun. The day of aneurysm rupture was designated Day 0. Samples of CSF were collected from patients on Days 2, 6, 11, and 16 after SAH. Control cisternal CSF samples were obtained through cisternal drainage catheters from five patients (age 58 _+ 4 years) 6 to 9 months after undergoing successful aneurysm surgery following an SAH. J. Neurosurg. / Volume 75 / September, 1991
The CSF samples were collected and immediately placed in ice (-4~ then centrifuged at 1000 G for 5 minutes at 4"C. The supernatant was collected, frozen, and stored at -20"C until analysis.
Culture of Vascular Smooth-Muscle Cells Vascular smooth-muscle cells* were prepared from explants of the thoracic aorta obtained from 7-weekold male Sprague-Dawley rats by the method of Ross2j and cultured in Dulbecco's modified Eagle's medium (DMEM) plus 10% fetal calf serum (FCS) in a 10% CO2 atmosphere at 37"C, as described previously.I~The cells were passaged weekly and cells at passages nine to 12 were used for the current study. The cells were grown in serum-containing growth medium for 4 days to confluency, and were then starved of serum for 2 days prior to the addition of CSF. Measurement of Cytosolic Free Calcium Changes For measurement of cytosolic free Ca ++ ([Ca++]~) changes induced by CSF, vascular smooth-muscle cells were plated at a density of 2 x 104 cells/chamber on the glass coverslip that adheres to the smooth lower side of Flexiperm discs.t After culture for 48 hours at 37"C, the cells were deprived of serum as described above. Before the addition of CSF, the cells were washed twice and loaded with the fuorescence indicator fura-2/AM (2 t~M) for 40 minutes at 37~ in 0.3 ml of serum-free DMEM, then rinsed free of the extracellular dye and incubated for 15 minutes to allow de-esterification of the dye. The changes in [Ca++]~were determined in individual vascular smooth-muscle cells loaded with fura-2 by digital imaging of Ca+§ fluorescence.~/After the addition of 300 ul CSF from each patient, fluorescence images of emission wavelength of 500 nm were obtained at alternating excitation wavelengths of 340 and 360 nm through a Videcon camera every 5 seconds for single cells at 37~ The fluorescence emission intensity ratio of 340 nm/360 nm was calculated on the image processor after subtraction of background values. Statbtical Analysis The statistical evaluation in the non-vasospasm and vasospasm groups was carried out using the Wilcoxon/ Mann-Whitney U-test; comparison of values based on the number of days after SAH that CSF sampling was carried out was analyzed by the t-test with subsequent Bonferroni correction for multiple comparisons. The values were expressed as mean __ SEM. Values of p < 0.05 were considered significant. * Smooth-muscle cells obtained from Dr. H. Arita, Shionogi Research Laboratories, Osaka, Japan. t Flexiperm disc manufactured by Heraeus Biothechnology, Hanau, Germany. Argus-100/CA imaging processor manufactured by Hamamatsu Photonics, Hamamatsu, Japan. 453
K. Takenaka, et al.
FIG. 1. Effects of cerebrospinal fluid (CSF) from patients with subarachnoid hemorrhage (SAH) on cytosolic free Ca++ ([Ca++]~) level in fura-2-1oaded vascular smooth-muscle cells. The [Ca*+]~ levels were measured as described in the text. Traces shown were representative of [Ca++]i responses of individual cells (number of cells > 40). A: Results obtained when CSF (300 #1) from patients with SAH was added to vascular smooth-muscle cells. B: Results in vascular smoothmuscle cells pretreated with 2 mM EGTA for ! minute when CSF from patients with SAH was added.
Results As shown in Fig. 1 left, CSF from all patients obtained on Days 2, 6, 11, and 16 after SAH induced a transient elevation of cytosolic free calcium ([Ca++]i), with a peak attained within 10 seconds followed by a gradual decline within 3 minutes after exposure of CSF (in six tests each). The addition of 2 mM ethyleneglycolbis (~-aminoethylether)-N,N'-tetra-acctic acid (EGTA)
caused a slight reduction in the height of the CSFinduced [Ca++]i transient increase, and the [Ca++]~ elevation declined to the basal level more quickly (within 2 minutes, Fig. 1 right) compared to that in preparations without EGTA (in six tests each). Figure 2 shows the changes in [Ca+§ level and the greatest difference between the basal and peak levels in vascular smooth-muscle cells induced by CSF from each patient (A Ratio). The time courses after SAH for the A Ratios induced by CSF are depicted in Figs. 3 and 4. In the non-vasospasm group (Fig. 3 left), the A Ratio seemed to be stable during the entire period of study after SAH. No significant difference was observed between days after SAH (indicated as open columns in Fig. 4). In contrast, in the vasospasm group (Fig. 3 right) the ~x Ratio was revealed as a biphasic response, with the highest concentrations on Days 2 and 11 and the lowest concentrations on Days 6 and 16. In seven patients with vasospasm, the A Ratio was 0.179 _+0.017 on Day 2 after SAH, 0.080 _+ 0.010 on Day 6, 0.132 -4-_ 0.008 on Day 11, and 0.099 _ 0.009 on Day 16. On Day 2 after SAH the A Ratio was significantly higher than on Day 6 or 16 (p < 0.01), and on Day 11 it was also significantly higher than on Day 6 (p < 0.05, t-test with subsequent Bonferroni correction for multiple comparisons). This value is indicated as shaded col-
Fro. 2. The cytosolic free calcium ([Ca++]~)level was measured as described in the text. Cerebrospinal fluid (CSF, 300 ul) from one patient in the vasospasm group was added on Day 2 after subarachnoid hemorrhage to vascular smooth-muscle cells. Upper:The fluorescence photographs are representative of changes in [Ca++]i levels in vascular smooth-muscle cells at 0, 10, 75, and 200 seconds after the addition of CSF. The color scale on the right indicates the [Ca++]a level (ratio 340 nm/360 nm). Lower: Changes in [Ca++]~ levels. The difference in [Ca++]~between the basal and peak levels induced by CSF is termed "/x Ratio." The numbers refer to the photographs above. 454
J. Neurosurg. / Volume 75/September, 1991
Bloody CSF-induced Ca ++ elevation in smooth-muscle cells
FIG. 3. Time course for cytosolic free calcium ([Ca++]i) levels in vascular smooth-muscle cells induced by 300 ul cerebrospinal fluid (CSF) from post-subarachnoid hemorrhage (SAH) patients in the non-vasospasm (left) and the vasospasm (right) groups. The [Ca++]~levels were measured as described in the text. The ordinate represents the changes in the [Ca++], level shown as "4 Ratio" in Fig. 2. Five control CSF samples (300 ul each) were also added to vascular smooth-muscle cells. Data on each day after SAH of each patient were cumulative of six separate experiments and were presented as the mean _+ standard error of the mean for 40 cells. Left: Numbers 1 through 5 refer to patients in the non-vasospasm group. Right: Numbers 6 through 12 refer to patients in the vasospasm group. Asterisks denote onset of clinical signs due to vasospasm; squares denote the appearance of a low-density area on computerized tomography scans due to vasospasm.
umns in Fig. 4. On Days 6 to 11 after SAH, clinical observations of vasospasm were seen (Fig. 3 right). Significant differences were observed in the A Ratio between the non-vasospasm and vasospasm groups on Days 2 and 6 (0.120 _ 0.007 vs. < 0.179 + 0.017, respectively, on Day 2; 0.110 _ 0.006 vs. > 0.080 + 0.010, respectively, on Day 6). There were, however, no significant differences in /x Ratio between the two groups on Days 11 or 16 (Fig. 4). The degree of s Ratio induced by CSF from control patients was significantly lower than that from patients with SAH (p < 0.01), as shown in Fig. 4. Three control CSF samples did not induce a [Ca++]~ elevation, and two control CSF samples induced lower [Ca++]~ elevations (0.020 _+ 0.002 and 0.010 -4-0.005 of the A Ratio).
Discussion
Calcium and Smoolh-Muscle Contraction It is generally accepted that the contractile activity o f vascular smooth muscle is regulated by the concentration of intracellular free calcium ([Ca++]0. This contraction is initiated by a variety of external stimuli which cause the [Ca++]~ to rise to a critical level, resulting from the influx of extracellular [Ca++]~ and/or the mobilization from intracellular [Ca++]~ store sites. This in turn leads to the activation of protein kinase C and Ca*+-calmodulin (CaM)-dependent kinase. 23 Studies o f intracellular calcium changes have recently been facilitated by the novel calcium ion chelators (quin-2/AM, indo-1/AM, and fura-2/AM). Fura-2/AM is highly specific for Ca *+ ions, readily distinguishing them from other divalent cations. It is also sensitive to Ca +* changes in the nanomolar range that are encoun-
J. Neurosurg. / Volume 75/September, 1991
FIG. 4. The changes in cytosolic free calcium ([Ca++]31evel induced by cerebrospinal fluid (CSF) from patients in nonvasospasm (open columns), vasospasm (shaded columns), or control (filled column) group in vascular smooth-muscle cells. The [Ca++]ilevels were measured as described in the text. The ordinate represents the changes in [Ca++]~ level shown as "4 Ratio" in Fig. 2. Data are expressed as means _+standard error of the means for five non-vasospasm patients, seven vasospasm patients, and five control patients. Statistically significant differences between two values were indicated as p < 0.01 or p < 0.05. Statistically significant differences from control values were indicated by asterisks (p < 0.01). NS = no statistically significant difference; SAH = subarachnoid hemorrhage. 455
K. Takenaka, et al. tered within smooth-muscle cells. Fura-2 produces a potent fluorescent emission which allows its use in relatively low concentrations, thereby lowering any interference due to an inherent Ca ++ buffering capacity. Fura-2/AM is rapidly internalized by smooth-muscle cells. Once internalized, de-esterification of the acetoxymethyl group converts it to the free acid, which has minimal capability to diffuse out of the cell. Residual extracellular fura-2/AM, removed by a series of cell washings, assures that observed fluorescence changes in the cells correspond solely to alterations in intracellular free calcium. 2s CSF-Induced [Ca**]~ Elevation in Smooth-Muscle Cells
Bloody CSF collected from patients with SAH has been reported to produce sustained contractions of canine cerebral artery 19and human cerebral artery. 1,4.6,7 However, no study has been reported that reviews comparative data on [Ca++]i elevation in vascular smoothmuscle cells in response to CSF from patients with SAH on successive days following hemorrhage. In this experiment, CSF from all patients with SAH induced [Ca++]i elevations in single smooth-muscle cells. Furthermore, the addition of EGTA caused a slight reduction in the peak of the transient CSF-induced [Ca++]i elevation, but the peak [Ca++], declined more quickly to the basal level compared to the preparations without EGTA. These findings indicated thai the [Ca++]i elevations induced by CSF obtained after SAH were due to a combination of the influx of extracellular Ca ++ and the mobilization from an internal Ca ++ store site. The results support the proposition that some substances are released in the CSF of patients with SAH and produce large contractions of cerebral arteries.l'4-7'22'31 The changes in [Ca++]~ plotted over time after SAH in the vasospasm group showed a biphasic pattern, with the highest concentrations on Days 2 and 1 1 after SAH and the lowest on Days 6 and 16 after SAH. On Day 2, the [Ca++]~ elevation induced by CSF in the vasospasm group was significantly higher than that in the nonvasospasm group. In contrast, on Day 6 the [Ca++]~ elevation induced by CSF in the vasospasm group was significantly lower than that in the non-vasospasm group. As indicated above, the following possibilities may be considered: 1) On Day 2 after SAH, some unknown substances that induce [Ca§ elevations are released in the CSF of patients with SAH, and the concentration of these substances in CSF in patients with vasospasm is higher than in those without. 2) On Day 6, these substances in the CSF of the non-vasospasm group continue to flow out through a cisternal drainage catheter, whereas in the group with vasospasm they remain for a few days in the intracranial subarachnoid space. Thus, the [Ca++], elevation induced by CSF samples collected from non-vasospasm patients seemed to be stable after SAH throughout the study period in contrast to the [Ca++], elevation induced by CSF samples col456
lected from patients with vasospasm. 3) On Day 11, some substances which induce [Ca++]~ elevations may be produced in the CSF of patients with vasospasm, resulting in an elevation of [Ca++]~. These changes correlate with the clinical observation of cerebral vasospasm. Vasoactive Substances in CSF
Possible vasospasm-related factors include hemolysate substances such as oxyhemoglobin or bilirubin; vasoactive amines such as norepinephrine, serotonin neuropeptide Y, or endothelin; and arachidonic acid metabolites such as prostaglandin F,_,,,thromboxane A2, leukotriene C~, leukotriene D4, or hydroxyeicosa-tetraenoic acid. -~s~2`LT'~-~-'5'27'2oIt has recently been reported that prostaglandin F2, concentrations in CSF or plasma endothelin- 1 concentrations were significantly elevated in the presence of symptomatic vasospasm after SAHY'~ Despite extensive efforts, the vasoactive substances responsible for the contraction in vasospasm have not been identified. ~3.30 The results of this study of [Ca++]~ levels in vascular smooth-muscle cells induced by CSF from patients with SAH on successive days following the hemorrhage suggest that the re-elevation of [Ca++]~ on Day 11 after SAH correlates with the pathogenesis of delayed vasospasm. Some unknown substances that induce this reelevation in vascular smooth-muscle cells may be the spasmogens that cause vasospasm after SAH. Further investigation and analysis of these substances should be performed in an attempt to identify the spasmogens in CSF from patients following SAH by studying human cerebral arterial smooth-muscle cells. We are presently trying to purify, identify, and characterize these substances in the CSF from patients with SAH. Conclusions Samples of CSF obtained from patients following SAH induced significantly higher transient intracellular calcium elevations when compared to those induced by control CSF. In studies of samples from the vasospasm group, changes in intracellular free calcium elevations induced by CSF showed a biphasic response with the highest concentration on Days 2 and 11 after SAH. It is suggested that some substances that induce the reelevation of intracellular calcium concentrations in vascular smooth-muscle cells may be the key substances responsible for vasospasm. Acknowledgment We express our gratitude to Dr. K. Deguchi for his helpful discussion. References I. Allen GS, Gross CJ, French LA, et al: Cerebral arterial spasm: Part 5. In vitro contractile activity of vasoactive agents including human CSF on human basilar and anterior cerebral arteries. J Neurosnrg 44:584-600, 1976 2. Asano T, lkegaki I, Suzuki Y, et al: Endothelin and the J. Neurosurg. / Volume 75/September, 1991
Bloody CSF-induced Ca ++ elevation in smooth-muscle cells
3. 4.
5.
6.
7.
8. 9.
10.
1I. 12. 13. 14.
15. 16. 17.
production of cerebral vasospasm in dogs. Biochem Biophys Res Colnmun 159:1345-1351, 1989 Auer LM: Acute operation and preventive nimodipine improve outcome in patients with ruptured cerebral aneurysms. Neurosurgery 15:57-66, 1988 Boullin DJ, Brandt L, Ljunggren B, et al: Vasoconstrictor activity in cerebrospinal fluid from patients subjected to early surgery for ruptured intracranial aneurysms. J Neurosurg 55:237-245, 1981 Boullin D J, Hunt TM, Rogers AT: Models for investigating the aetiology of cerebral arterial spasm: comparative responses of the human basilar artery with rat colon, anoeoccygeus, stomach fundus, and aorta and guinea-pig ileum and colon. Br J Pharmaco163:251-257, 1978 Boullin D J, Mohan J, Grahame-Smith DG: Evidence for the presence of a vasoacfive substance (possibly involved in the aetiology of cerebral arterial spasm) in cerebrospinal fluid from patients with subarachnoid haemorrhage. J Neural Neurosurg Psychiatry 39:756-766, 1976 Brandt L, Ljunggren B, Andersson KE, et al: Effects of indomethacin and prostacyclin on isolated human pial arteries contracted by CSF from patients with aneurysmal SAH. J Neurosurg 55:877-883, 1981 Chehrazi BB, Giri S, Joy RM, et al: Prostaglandins and vasoactive amines in cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Stroke 20:217-224, 1989 Fisher CM, Kistler JP, Davis JM: Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery 6: I-9, 1980 Hanansaki K, Nakano K, Kasai H, et al: Specific receptors for thromboxane A2 in cultured vascular smooth muscle cell of rat aorta. Biochem Biophys Res Commun 150: 1170-1175, 1988 Hunt WE, Hess RM: Surgical risk as related to time of intervention in the repair of intracranial aneurysm. J Neurosurg 28:14-20, 1968 Juvela S, Kaste M, Hillbom M, et al: Platelet thromboxane release after subarachnoid hemorrhage and surgery. Stroke 21:566-57 l, 1990 Kassell NF, Sasaki T, Colohan ART, et al: Cerebral vasospasm following aneurysmal subarachnoid hemorrhage. Stroke 16:562-572, 1985 Kassell NF, Torner JC, Haley EC Jr, et al: The International Cooperative Study on the Timing of Aneurysm Surgery. Part l: Overall management results. J Neurosurg 73:18-36, 1990 Kassell NF, Tarrier JC, Jane JA, et al: The International Cooperative Study on the Timing of Aneurysm Surgery. Part 2: Surgical results. J Neurosurg 73:37-47, 1990 Masaoka H, Suzuki R, Hirata Y, et al: Raised plasma endothelin in aneurysmal subarachnoid haemorrhage. Lancet 2:1402, 1989 (Letter) Miao FJP, Lee TJF: Effects of bilirubin on cerebral arterial tone in vitro. J Cereb Blood Flow Metab 9:666-674, 1989
J. Neurosurg. / Volume 75/September, 1991
18. Nishizuka Y: Studies and perspectives of protein kinase C. Science 233:305-312, 1986 19. Okwuasaba FK, Weir BKA, Cook DA, et al: Effects of various intracranial fluid on smooth muscle. Neurosurgery 9:402-406, 198 l 20. Paoletti P, Gaetani P, Grignani G, et al: CSF leukotriene C4 following subarachnoid hemorrhage. J Neurosurg 69: 488-493, 1988 21. Ross R: The smooth muscle cell. II. Growth of smooth muscle in culture and formation of elastic fibers. 8 Cell Bioi 50:172-186, 1971 22. Sasaki T, Asano T, Takakura K, et al: Nature of the vasoactive substance in CSF from patients with subarachnoid hemorrhage. J Neurosurg 60:1186-1191, 1984 23. Somlyo AP, Himpens B: Cell calcium and its regulation in smooth muscle. FASEB J 3:2266-2276, 1989 24. Sundt TM Jr, Davis DH: Reaction of cerebrovascular smooth muscle to blood and ischemia: primary versus secondary vasospasm, in Wilkins RH (ed): Cerebral Arterial Spasm. Baltimore: Williams & Wilkins, 1979, pp 244-250 25. Suzuki Y, Sato S, Suzuki H, et al: Increased neuropeptide Y concentrations in cerebrospinal fluid from patients with aneurysmal subaraehnoid hemorrhage. Stroke 20: 1680-1684, 1989 26. Takenaka K, Yamada H, Sakai N, et al: Cytosolic calcium changes in cultured rat aortic smooth-muscle cells induced by oxyhemoglobin. J Neurusurg 74:620-624, 1991 27. Tanishima T: Cerebral vasospasm: contractile activity of hemoglobin in isolated canine basilar arteries. J Neurasurg 53:787-793, 1980 28. Tsien RY, Rink T J, Poenie M: Measurement of cytosolic free Ca2+ in individual small cells using fluorescence microscopy with dual excitation wavelengths. Cell Calcium 6:145-157, 1985 29. Watanabe T, Asano T, Shimizu T, et al: Participation of lipoxygenase products from arachidonic acid in the pathogenesis of cerebral vasospasm. J Neuroehem 50: 1145-1150, 1988 30. Wellum GR, Peterson JW, Zervas NT: The relevance of in vitro smooth muscle experiments to cerebral vasospasm. Stroke 16:573-581, 1985 31. White RP, Macleod RM, Muhlbauer MS: Evaluation of the role hemoglobin in cerebrospinal fluid plays in producing contractions in cerebral arteries. Surg Neural 27: 237-242, 1987 32. Wilkins RH: Attempts at prevention or treatment of intracranial arterial spasm: an update. Neurosurgery 18: 808-825, 1986 Manuscript received September 23, 1990. Accepted in final form April 3, 1991. Address reprint requests to: Katsunobu Takenaka, M.D., Department of Neurosurgery, Gifu University School of Medicine, 40 Tsukasamachi, Gifu 500, Japan.
457
