Search

Object: Cerebral vasospasm after subarachnoid hemorrhage (SAH) is a serious complication. Free radicals derived from subarachnoid clotting are recognized to play an important role. Oxidized low-density lipoprotein (ox-LDL) and lectin-like oxidized LDL receptor-1 (LOX-1) have been shown to be related to the pathogenesis of atherosclerosis and may increase in cerebral arteries after SAH, due to the action of free radicals derived from a subarachnoid clot. These molecules may also affect the pathogenesis of vasospasm, generating intracellular reactive oxygen species and downregulating the expression of endothelial NO synthase (eNOS). If so, apple polyphenol might be effective in the prevention of vasospasm due to an abundant content of procyanidins, which exhibit strong radical scavenging effects, and the ability to suppress ox-LDL and LOX-1. The purposes of this study were to investigate changes in levels of ox-LDL and LOX-1 after SAH and whether administering apple polyphenol can modify cerebral vasospasm., Methods: Forty Japanese white rabbits were assigned randomly to 4 groups: an SAH group (n = 10); a shamoperation group (n = 10), which underwent intracisternal saline injection; a low-dose polyphenol group (n = 10) with SAH and oral administration of apple polyphenol at 10 mg/kg per day from Day 0 to Day 3; and a high-dose polyphenol group (n = 10) with SAH and oral administration of apple polyphenol at 50 mg/kg per day. At Day 4, the basilar artery and brain was excised from each rabbit. The degree of cerebral vasospasm was evaluated by measuring the cross-sectional area of each basilar artery, and the expression of ox-LDL, LOX-1, and eNOS was examined for each basilar artery by immunohistochemical staining and reverse transcriptase polymerase chain reaction. In addition, neuronal apoptosis in the cerebral cortex was evaluated by TUNEL., Results: Compared with the sham group, the expression of ox-LDL and LOX-1 in the basilar arterial wall was significantly increased in the SAH group, the expression of eNOS was significantly decreased, and the cross-sectional area of basilar artery was significantly decreased. Compared with the SAH group, the cross-sectional area of basilar artery was increased in the polyphenol groups, together with the decreased expression of ox-LDL and LOX-1 and the increased expression of eNOS. In the high-dose polyphenol group, those changes were statistically significant compared with the SAH group. In the low-dose polyphenol group, those changes were smaller than in the high-dose polyphenol group. No apoptosis and no changes were seen in the cerebral cortex in all groups., Conclusions: This is the first study suggesting that ox-LDL and LOX-1 increase due to SAH and that they may play a role in the pathogenesis of vasospasm. It is assumed that procyanidins in apple polyphenol may inhibit a vicious cycle of ox-LDL, LOX-1, and ROS in a dose-dependent manner. Apple polyphenol is a candidate for preventive treatment of cerebral vasospasm.

Object: The L-type Ca++ channel antagonists like nimodipine have limited efficacy against vasospasm after subarachnoid hemorrhage (SAH). The authors tested the hypothesis that this is because SAH alters these channels, rendering them less responsible for contraction., Methods: Basilar artery smooth muscle cells were isolated 4, 7, and 21 days after SAH in dogs, and Ca++ channel currents were recorded in 10-mmol/L barium. Proteins for α₁ subunits of L-type Ca++ channels were measured by immunoblotting and isometric tension recordings done on rings of the basilar artery., Results: High voltage-activated (HVA) Ca++ channel currents were significantly decreased and low voltage-activated (LVA) currents increased during vasospasm 4, 7, and 21 days after SAH (p < 0.05). Vasospasm was associated with a significant decrease in the number of cells with negligible LVA current while the number of cells in which the LVA current formed greater than 50% of the maximal current increased (p < 0.01). Window currents through LVA and HVA channels were significantly reduced. All changes correlated with the severity of vasospasm. There was an increase in protein for Ca(v)3.1 and Ca(v)3.3 α₁ subunits that comprise T-type Ca++ channels, a decrease in L-type (Ca(v)1.2 and Ca(v)1.3) and an increase in R-type (Ca(v)2.3) Ca++ channel α₁ subunits. Functionally, however, isometric tension studies showed vasospastic arteries still relaxed with nimodipine., Conclusions: Voltage-dependent Ca++ channels are altered in cerebral arteries after SAH. While decreased L-type channels may account for the lack of efficacy of nimodipine clinically, there may be other reasons such as inadequate dose, effect of nimodipine on other cellular targets, and mechanisms of vasospasm other than smooth muscle contraction mediated by activation of L-type Ca++ channels.

Object: The gap junction is important in the propagation of dilation/constriction signals along vessels for coordinated behavior in control of vascular tone. The authors hypothesized that gap junctions might play a role in cerebral vasospasm following subarachnoid hemorrhage (SAH). The aims of the present study were to investigate the role of gap junctions and to observe the potential therapeutic efficacy of gap junction blockers in cerebral vasospasm in vitro and in vivo., Methods: For the in vitro investigation, the effect of heptanol on the oxyhemoglobin (HbO(2))-induced contraction of isolated rabbit basilar arteries (BAs) was observed by using an isometric tension-recording method. For the in vivo experiments, the potential therapeutic efficacy of heptanol and carbenoxolone was surveyed after it was given intravenously in the rabbit double-hemorrhage model. Light microscopy was performed to assess the morphological changes in the arteries examined., Results: For the in vitro method, heptanol significantly inhibited the sustained contraction induced both by HbO(2) and K(+) in the BA rings. The magnitude of the heptanol-induced relaxation was dose dependent. The inhibitory effect of heptanol on the K(+)-induced vasoconstriction was weaker than that on the HbO(2)-induced constriction. After arterial rings were pretreated for 10 minutes, heptanol significantly decreased their responses to the HbO(2)-induced contraction. For the in vivo method, heptanol and carbenoxolone significantly decreased the narrowing of BAs when given intravenously in the rabbit double-hemorrhage model. In both treated groups, the diameters of the arteries had not changed significantly on Day 7 compared with those of the arteries in the SAH + vehicle and the SAH-only group., Conclusions: Heptanol and carbenoxolone significantly inhibited the experimental cerebral vasospasm both in vitro and in vivo. Blockage of gap junctions is a probable candidate for a new approach in the treatment of cerebral vasospasm. Gap junctions may play a pathophysiological role in cerebral vasospasm.

Object: The key role in the development of cerebral vasospasm after subarachnoid hemorrhage (SAH) is increasingly assigned to endothelin (ET)-1. Constriction of the cerebrovasculature by ET-1 is mainly mediated by the ET(A) receptor but is putatively altered during the development of cerebral vasospasm. Therefore, the aim in the present study was to characterize these alterations, with the emphasis on the ET(A) receptor., Methods: Cerebral vasospasm was induced using the rat double-hemorrhage model and proven by perfusion weighted magnetic resonance imaging. Rats were killed on Day 5 after SAH, and immunohistochemical staining for ET(A), receptors was performed. The isometric force of basilar artery ring segments with (E+, control group) and without (E-, SAH group) endothelial function was measured. Concentration effect curves (CECs) for ET-1 were constructed by cumulative application in the absence and presence of the selective ET(A) receptor antagonist clazosentan (10(-8) or 10(-7) M)., Results: The CEC for E+ segments was significantly shifted to the left after SAH by a factor of 3.7, whereas maximum contraction was unchanged. In E- segments, the CECs were not shifted during cerebral vasospasm but the maximum contraction was significantly enhanced. The inhibitory potency of clazosentan yielded a pA2 value of 8.6 +/- 0.2. Immunohistochemical staining of the smooth-muscle layer showed no significant increase of ET(A) receptor expression, but positive staining occurred in the endothelial space after SAH., Conclusions: The present data indicate an enhanced contractile effect of the smooth-muscle ET(A) receptors in cases of cerebral vasospasm. The inhibitory potency of clazosentan on this contraction is increased. Furthermore, some evidence for an ET(A) receptor and an endothelium-dependent vasoactive effect after SAH is provided.

Object: The disturbed balance between nitric oxide and endothelin (ET)-1 in the cerebrovasculature seems to play a major role in the development of cerebral vasospasm after subarachnoid hemorrhage. Endothelin-1 represents the contractile part in this balance. In addition to the prevailing ET(A) receptor-dependent contractile effect, ET-1 also has ET(B) receptor-mediated vasodilatory attributes. The aim of the present study was to define the actual selectivity of clazosentan, the first putative highly ET(A) receptor-selective antagonist clinically proven to be effective in the treatment of vasospasm in the cerebrovasculature., Methods: Rat basilar artery ring segments with endothelial function were used for the measurement of isometric force. Concentration effect curves were constructed by cumulative application of sarafotoxin S6c, ET-1, or big ET-1 in the presence or absence of clazosentan (10(-9) to 10(-6) M) after a precontraction was induced by prostaglandin F2alpha. The inhibition by clazosentan was estimated by the value of the affinity constant (pA2). The relaxation induced by sarafotoxin S6c, ET-1, and big ET-1 was inhibited in a competitive manner by clazosentan, yielding pA2 values of 7.1, 6.7, and 6.5, respectively. The selectivity to the ET(A) receptor in the cerebrovascular system was approximately two logarithmic units., Conclusions: The present investigation shows a competitive inhibition of ET(B) receptor-mediated relaxation in cerebral vessels by clazosentan in therapeutically relevant concentrations. Thus, additional clinical trials should be undertaken to evaluate clazosentan concentrations in cerebrospinal fluid. Furthermore, the present data may be taken to describe the pharmacological properties for an ET receptor antagonist specifically tailored for the treatment of pathological conditions of impaired cerebral blood flow.

Object: The central role of endothelin (ET)-1 in the development of cerebral vasospasm after subarachnoid hemorrhage is indicated by the successful treatment of this vasospasm in several animal models by using selective ET(A) receptor antagonists. Clazosentan is a selective ET(A) receptor antagonist that provides for the first time clinical proof that ET-1 is involved in the pathogenesis of cerebral vasospasm. The aim of the present investigation was, therefore, to define the pharmacological properties of clazosentan that affect ET(A) receptor-mediated contraction in the cerebrovasculature., Methods: Isometric force measurements were performed in rat basilar artery (BA) ring segments with (E+) and without (E-) endothelial function. Concentration effect curves (CECs) were constructed by cumulative application of ET-1 or big ET-1 in the absence or presence of clazosentan (10(-9), 10(-8), and 10(-7) M). The inhibitory potency of clazosentan was determined by the value of the affinity constant (pA2). The CECs for contraction induced by ET-1 and big ET-1 were shifted to the right in the presence of clazosentan in a parallel dose-dependent manner, which indicates competitive antagonism. The pA2 values for ET-1 were 7.8 (E+) and 8.6 (E-) and the corresponding values for big ET-1 were 8.6 (E+) and 8.3 (E-)., Conclusions: The present data characterize clazosentan as a potent competitive antagonist of ET(A) receptor-mediated constriction of the cerebrovasculature by ET-1 and its precursor big ET-1. These functional data may also be used to define an in vitro profile of an ET receptor antagonist with a high probability of clinical efficacy.

Object: The aim of this project was to study the perturbations of four smooth-muscle proteins and an extracellular protein, type I collagen, after subarachnoid hemorrhage (SAH) and to examine the possible preventive effects of dexamethasone., Methods: Using a one-hemorrhage rabbit model, the authors first examined the effects of SAH on the expression of alpha-actin, h-caldesmon, vimentin, smoothelin-B, and type I collagen; second, they studied whether post-SAH systemic administration of dexamethasone (three daily injections) corrected the induced alterations. Measurements were obtained at Day 7 post-SAH. The proteins were studied by performing immunohistochemical staining and using a laser-scanning confocal microscope. Compared with control (sham-injured) arteries, the density of the media of arteries subjected to SAH was reduced for alpha-actin (-11%, p = 0.01) and h-caldesmon (-15%, p = 0.06) but increased for vimentin (+15%, p = 0.04) and smoothelin-B (+53%, p = 0.04). Among animals in which SAH was induced, arteries in those treated with dexamethasone demonstrated higher values of density for alpha-actin (+13%, p = 0.05) and h-caldesmon (+20%, p = 0.01), lower values for vimentin (-55%, p = 0.05), and nonsignificantly different values for smoothelin-B. The density of type I collagen in the adventitia decreased significantly after SAH (-45%, p = 0.01), but dexamethasone treatment had no effect on this decrease., Conclusions: The SAH-induced alterations in the density of three of four smooth-muscle proteins were prevented by dexamethasone treatment; two of these proteins--alpha-actin and h-caldesmon--are directly related to contraction. This drug may potentially be useful to prevent certain morphological and functional changes in cerebral arteries after SAH.

Object: Leukocyte-endothelial cell interactions may play a role in the development of cerebral vasospasm after aneurysmal subarachnoid hemorrhage (SAH) because the extravasation of circulating leukocytes into the periadventitial space within 24 hours after the hemorrhage appears to be a critical event in this process. Ibuprofen is an antiinflammatory agent that inhibits the expression of specific cell adhesion molecules and, consequently, disrupts leukocyte-endothelial cell interactions. The authors investigated the efficacy of ibuprofen delivered locally from controlled-release polymers in the rabbit basilar artery (BA) model of cerebral vasospasm., Methods: Ibuprofen was incorporated into controlled-release ethylene-vinyl acetate copolymer (EVAc) constituting 45% of the resulting polymer by weight. Fifty-four New Zealand White rabbits were randomized to 10 groups: sham operation (seven animals); SAH only (seven animals); and SAH plus either empty EVAc or ibuprofen-EVAc polymer at 30 minutes or 6, 12, or 24 hours (five animals per group; 40 total). The rabbits were killed 72 hours after induction of SAH, at the time of maximal vasospasm. The efficacy of ibuprofen in preventing vasospasm was assessed by measuring lumen patency of the rabbit's BAs. The intracranial controlled release of ibuprofen resulted in a significant inhibition of vasospasm when treatment was initiated at 30 minutes (patency 92.3 +/- 5.1% compared with 52.1 +/- 5.1% in animals given empty EVAc; p < 0.001) and 6 hours (patency 69.5 +/- 3.5% compared with 47.2 +/- 1.5% in animals given empty EVAc; p < 0.03) after blood deposition compared with treatment with empty EVAc. No effect was observed when treatment was begun at either 12 or 24 hours., Conclusions: Local intracranial delivery of ibuprofen accomplished using controlled-release polymers prevents vasospasm in the rabbit BA model of vasospasm when administered within 6 hours after blood exposure.

Object: Adhesion of leukocytes and their migration into the periadventitial space may be critical in the pathophysiology of vasospasm following subarachnoid hemorrhage (SAH). The cell adhesion molecules involved in this process are lymphocyte function-associated antigen-1 (CD11a/CD18) and macrophage antigen-1 (CD11b/CD18), which are present on neutrophils/macrophages, and intercellular adhesion molecule-1 (CD54), which is present in endothelial cells. A humanized monoclonal antibody (mAb), Hu23F2G, targets CD11/CD18 and prevents leukocyte adhesion to endothelial cells. In this study, systemic administration of Hu23F2G prevented vasospasm in the rabbit model of SAH., Methods: Twenty-six New Zealand White rabbits were injected with autologous blood into the cisterna magna to induce SAH, after which they were randomized to receive injections of either Hu23F2G (10 animals) or a placebo at 30 minutes and 24 and 48 hours after SAH (six animals). Control animals underwent sham operations (four animals) or SAH alone (six animals). The animals were killed 72 hours after SAH, their bodies perfused and fixed, and their basilar arteries processed for morphometric analysis. Peripheral white blood cells (WBCs) were counted at 72 hours. The percentages of lumen patency were compared using the Student t-test. The presence of neutrophils and macrophages was confirmed by immunohistochemical analysis in which a rat anti-rabbit anti-CD18 mAb and cresyl violet were used. Treatment with Hu23F2G resulted in the significant prevention of vasospasm. Animals treated with Hu23F2G had 90 +/- 7% lumen patency compared with 65 +/- 7% in the placebo group (p = 0.025). The percentage of lumen patency in the SAH-only group was 59 +/- 10%. The mean WBC count was 16,300 +/- 2710/microl in the treatment group, compared with 7000 +/- 386/microl in the control group (p = 0.02). Administration of Hu23F2G produced increased numbers of WBCs in 70% of the animals treated., Conclusions: This study supports the concept that leukocyte-endothelial cell interactions play an important role in the pathophysiology of chronic vasospasm after SAH. Systemic therapy with an anti-CD11/CD18 mAb prevents vasospasm after SAH by inhibiting adhesion of neutrophils and macrophages and their migration into the periadventitial space.

Object: The Ca++ influx into vascular smooth-muscle cells (VSMCs) plays a fundamental role in the development and chronic effects of vasospasm after subarachnoid hemorrhage (SAH). The Ca++-permeable nonselective cation channels (NSCCs) are activated by several endothelium-derived constricting factors such as endothelin 1 (ET-1) and thromboxane A2. Moreover, the receptor-operated Ca++ channel blocker LOE 908 inhibits ET-1-induced extracellular Ca++ influx via NSCCs in the VSMCs of the basilar artery (BA) and the NSCC-dependent part of ET-1-induced vasoconstriction of BA rings. The purpose of the present study was to evaluate the in vivo role of LOE 908 on SAH-induced vasospasm., Methods: Forty-two Japanese white rabbits were assigned to seven groups. Treatment groups consisted of the following: 1) control rabbits without SAH that received a cisternal injection of saline; 2) rabbits with SAH that were subjected to the intravenous administration of saline; 3 through 6) rabbits with SAH that underwent the intravenous administration of 0.01. 0.1, 1, or 10 mg/kg LOE 908, respectively; and 7) rabbits without SAH that underwent the intravenous administration of 10 mg/kg LOE 908. Autologous blood was injected into the cisterna magna. The caliber of the BA was measured on angiographic studies before and after the cisternal injection of autologous blood. The intravenous injection of LOE 908 inhibited the magnitude of an SAH-induced vasosapsm. In addition, the concentration of LOE 908 required to relax vasospasm (1 mg/kg) correlated with that required to block Ca++ influx into VSMCs., Conclusions: The Ca++ channel blocker LOE 908 may inhibit the magnitude of an SAH-induced vasospasm by blocking the influx of Ca++ through NSCCs in rabbit BAs. Blocking the NSCCs may represent a new treatment for cerebral vasospasm after SAH.

Object: Mitogen-activated protein kinase (MAPK) is an important signaling factor in vascular proliferation and contraction, which are the two features of cerebral vasospasm that follow subarachnoid hemorrhage. The authors studied the possible involvement of MAPK in hemolysate-induced signal transduction and contraction in rabbit basilar artery (BA)., Methods: Isometric tension was used to record the contractile response of rabbit BA to hemolysate, and Western blots were obtained using antibodies for MAPK. The following results are reported. 1) Hemolysate produced a concentration-dependent contraction of rabbit BA; however, preincubation of arteries with the MAPK kinase (MEK) inhibitor PD-98059 markedly reduced this contraction. The administration of PD-98059 also relaxed, in a concentration-dependent fashion, the sustained contraction induced by 10% hemolysate. 2) The Janus tyrosine kinase 2 inhibitor AG-490, preincubated with arterial rings, reduced the contractile response to hemolysate but failed to relax the sustained contraction induced by this agent. The Src-tyrosine kinase inhibitor damnacanthal and the phosphatidylinositol 3-kinase inhibitor wortmannin failed to reduce hemolysate-induced contraction. 3) Hemolysate produced a time-dependent elevation of MAPK immunoreactivity as seen on Western blots of rabbit BA. The MAPK was enhanced 1 minute after hemolysate exposure and the effect reached maximum levels at 5 minutes. The immunoreactivity of MAPK decayed slowly over time, but the level of this kinase was still higher than the basal level, even at 2 hours after exposure to hemolysate. Preincubation of arteries with the MEK inhibitor PD-98059 abolished the effect of hemolysate on MAPK immunoreactivity., Conclusions: Hemolysate produced contraction of rabbit BA, possibly by activation of MAPK, and therefore MAPK inhibitors may be useful in the treatment of cerebral vasospasm.

Object: Intrathecal bolus administration of (Z)-1-[N-(2-aminoethyl)-N-(2-ammonioethyl)aminio]diazen++ +-1-ium-1,2-diolate (DETA/NO), a long half-life diazeniumdiolate-class nitric oxide (NO) donor, was evaluated for safety and efficacy in the treatment of delayed cerebral vasospasm in a canine model of subarachnoid hemorrhage (SAH)., Methods: The baseline basilar artery (BA) diameter of 25 dogs was measured with the aid of angiography on Day 0. Vasospasm was then induced by intracisternal injection of autologous arterial blood on Days 0 and 2. Repeated arteriography on Day 7 revealed an average BA diameter of 58% of baseline. Each dog was then randomized to one of four groups: a pathology control group (SAH only, four animals); a treatment control group (SAH plus 2 micromol of the inactive drug carrier DETA, eight animals); a low-dose treatment group (SAH plus 0.2 micromol DETA/NO, six animals); or a high-dose treatment group (SAH plus 2 micromol DETA/NO, six animals). The drugs were administered in a 2-ml intrathecal bolus via the cisterna magna. Arterial caliber was monitored by angiography over the subsequent 4 hours. A 2-micromol dose of the drug was then given and serial arteriography continued for an additional hour to screen for tachyphylaxis. Intracranial pressure and respiratory and hemodynamic parameters were continuously monitored. Histopathological analyses of the animals' brains were performed after the dogs were killed on Day 8. The drug DETA/NO produced reversal of vasospasm in a dose-dependent fashion that roughly followed a double exponential time course. Doses of 2 micromol DETA/NO resulted in restoration of the angiographically monitored BA diameter to the prevasospasm size at 1.5 hours posttreatment, and this was sustained at 88% of baseline at 4 hours (p < 0.01, independent samples t-test). By contrast, the treatment control group remained on average at 54% of baseline diameter. The low-dose treatment group achieved only partial and more transitory relaxation. Histopathological analyses showed findings consistent with chronic SAH but did not demonstrate any toxicity associated with the NO donor. No adverse physiological changes were seen., Conclusions: This study indicates that long-acting NO donors are potentially useful as agents to restore circulation in patients suffering from cerebral vasospasm.

Object: This study was undertaken to explore whether erythrocyte lysate, a proposed cause of vasospasm, produces vasoconstriction by activation of tyrosine kinase in rabbit cerebral arteries., Methods: Isometric tension was used to monitor contractions in rabbit basilar arteries induced by erythrocyte lysate, 5-hydroxytryptamine (5-HT), or KCl in the absence or presence of tyrosine kinase inhibitors. Erythrocyte lysate, 5-HT, or KCl produced concentration-dependent contractions in rabbit basilar arteries. Preincubation with the tyrosine kinase inhibitors tyrphostin A23 and genistein (30 and 100 microM), but not diadzein, an inactive analog of genistein, attenuated significantly the contraction induced by erythrocyte lysate (p < 0.05). Tyrphostin A23, genistein, and diadzein (30 microM) failed to reduce the contraction caused by 5-HT. Genistein, but not tyrphostin A23 or diadzein (30 microM), attenuated significantly the contraction induced by KCl (p < 0.05). In another series, arterial rings were initially contracted with erythrocyte lysate, 5-HT, or KCl and the relaxant effect of genistein was then tested. Genistein relaxed rabbit basilar arteries that had been contracted by exposure to erythrocyte lysate, 5-HT, or KCl (30-100 microM; p < 0.05)., Conclusions: These data indicate that tyrosine kinase may play a role in the regulation of cerebral arterial contraction and tyrosine kinase inhibitors may be useful in the management of cerebral vasospasm.

Object: This study was designed to determine whether subarachnoid hemorrhage (SAH) affects the function of the K+ channels responsible for relaxation of canine cerebral arteries in response to adenylate cyclase activation., Method: The effect of K+ channel inhibitors on the arterial relaxation response to forskolin, a direct adenylate cyclase activator, was studied in rings of basilar arteries obtained from normal dogs and dogs in which SAH was induced (double-hemorrhage model). The levels of adenosine 3',5'-cyclic monophosphate (cAMP) were measured using the radioimmunoassay technique. In rings with the endothelium removed, relaxation induced by forskolin was not affected by SAH. The relaxation response to forskolin was reduced by charybdotoxin (10(-7) mol/L), a selective Ca++-activated K+ channel inhibitor, in normal arteries and arteries subjected to autologous blood injection. This inhibitory effect of charybdotoxin was significantly greater in arteries involved in SAH than in normal vessels. The relaxation response to forskolin was reduced by 4-aminopyridine (10(-3) mol/L), a delayed rectifier K+ channel inhibitor, only in arteries involved in SAH. In contrast, the relaxation response to forskolin was not affected by glyburide (10(-5) mol/L), an adenosine 5'-triphosphate-sensitive K+ channel inhibitor, in both normal and SAH arteries. Forskolin (3 x 10(-7) mol/L) produced an approximately 10-fold increase in levels of cAMP. The basal values and increased levels of cAMP detected after stimulation with forskolin were no different in normal arteries and those exposed to SAH., Conclusions: These results demonstrate that formation of cAMP and the relaxation response to adenylate cyclase activation are not affected by SAH. However, in diseased arteries, K+ channels assume a more important role in the mediation of relaxation response to forskolin, indicating that SAH may change the mechanisms responsible for vasodilation induced by cAMP.

Object: The radical scavenger (+/-)-N,N'-propylenedinicotinamide (AVS) was shown recently to ameliorate delayed neurological deficits resulting from ischemia in patients who have had an aneurysmal subarachnoid hemorrhage (SAH). The aim of this study was to evaluate the effect of AVS administration after experimental SAH on 1) behavioral deficits; 2) angiographically confirmed basilar artery (BA) spasm; and 3) blood-brain barrier (BBB) permeability changes., Methods: These parameters were measured by 1) using a battery of well-characterized chronic assessment tasks over a 5-day observation period; 2) assessing in vivo the mean vessel diameter 2 days after SAH; and 3) evaluating the extravasation of protein-bound Evans Blue dye by using a spectrophotofluorimetric technique 2 days after SAH. Groups of eight to 10 rats received injections of 400 microl of autologous blood into the cisterna magna. Within 5 minutes after the surgical procedures were completed the rats were treated with an intravenously administered continuous infusion of saline (Group III) or AVS (1 mg/kg/minutes, Group IV). Results were compared with those in sham-operated animals treated with intravenously administered saline (Group I) or AVS (Group II). The AVS-treated rats had significantly improved balance beam scores on Days 1 to 2 (p < 0.05), shorter beam traverse times on Day 1 (p < 0.05), and better beam walking performance on Days 1 to 4 (p < 0.01), but no significant effect was seen in terms of SAH-related changes in body weight. Treatment with AVS also attenuated the SAH-induced BA spasm (p < 0.05) and decreased BBB permeability changes in frontal, temporal, parietal, occipital, and cerebellar cortices, and in the subcortical and cerebellar gray matter and brainstem (p < 0.01)., Conclusions: These results demonstrate useful antivasospastic and brain-protective actions of AVS after induction of experimental SAH and provide support for observations of beneficial effects of AVS made in the clinical setting.

Object: Monophosphoryl lipid A (MPL) and diphosphoryl lipid (DPL) are derivatives of the lipopolysaccharide (endotoxin) of Salmonella minnesota strain R595. Monophosphoryl lipid A is relatively nontoxic and can stimulate the natural defense or immune system. Diphosphoryl lipid is relatively toxic; however, at higher concentrations, it can also stimulate an immune response. The purpose of the present study was to determine the effects of these endotoxin analogs on cerebral vasospasm after the onset of subarachnoid hemorrhage (SAH) in rabbits., Methods: Intrathecal administration of MPL or DPL (5 microg/kg) was performed immediately before and 24 hours after induction of SAH in New Zealand White rabbits. Forty-eight hours after induction of SAH, the animals were killed by perfusion fixation for morphometric analyses of vessels or perfused with saline and assayed for superoxide dismutase (SOD) activity. Additional rabbits were administered MPL or DPL and killed 24 hours later for assessment of SOD activity; no SAH was induced in these animals. Experimental SAH elicited spasm of the basilar arteries in each group. Vasospasm was markedly attenuated in animals treated with MPL (p < 0.01 compared with vehicle-treated animals), but not in animals treated with DPL. A substantial reduction in SOD activity in the basilar artery accompanied the vasospasm; this loss of activity was significantly blocked by treatment with MPL, but not DPL. In animals that were not subjected to experimental SAH, MPL elicited a significant increase in SOD activity over basal levels, whereas DPL was ineffective., Conclusions: These data provide evidence of a marked protective effect of the endotoxin analog MPL against vasospasm. Although the mechanism(s) responsible for the protective effect of MPL remains to be verified, an enhancement of basal antioxidant activity and an inhibition of SAH-induced loss of this activity are attractive candidates. An MPL-based therapy could represent a useful addition to current therapies for SAH-induced cerebral injury.

The authors investigated selective intra- and extraluminal effects of endothelin (ET) on perfused basilar and extracranial arteries and also studied the interaction between ET and extraluminal oxyhemoglobin (oxyHb). The basilar, mesenteric, and femoral arteries were isolated from 23 Japanese White rabbits. After isolation of the intra- and extraluminal sides of the preparation, 3 x 10(-10) to 3 x 10(-8) mol/L of ET was administered intra- or extraluminally. After extraluminal pretreatment with 10(-5) mol/L oxyHb, 10(-5) mol/L N(G)-monomethyl-L-arginine (L-NMMA), or 10(-6) mol/L indomethacin, 10(-10) to 10(-8) mol/L of ET was administered intra- or extraluminally. Arterial contraction was evaluated by measuring the increase in the perfusion pressure gradient with a differential pressure gauge. Both intra- and extraluminal ET (10(-9) to 3 x 10(-8) mol/L) showed potent and dose-dependent vasoconstricting effects on basilar arteries (p < 0.01). The effect of ET on the basilar arteries was significantly greater than on the femoral or mesenteric arteries (both p < 0.01). The effect of intraluminal ET was enhanced by extraluminal oxyHb (p < 0.05) and L-NMMA (p < 0.01), but not by extraluminal indomethacin. Extraluminal oxyHb did not potentiate the contraction induced by extraluminal ET. These results indicate that the sensitivity of the basilar artery to intraluminal ET is greater than that of the femoral or mesenteric artery. Endothelin may act as a potent vasoconstrictor intra- as well as extraluminally under conditions such as subarachnoid hemorrhage in which oxyHb is present in the extraluminal space and endothelium-derived relaxing factors are inhibited.

The purpose of this study was to investigate the effectiveness of oral treatment with the endothelin (ET) A/B receptor antagonist Ro 47-0203, 4-tert-butyl-N-[6-(hydroxy- ethoxy)-5-(2-methoxy-phenoxy)-2'-bipyrimidin-4-yl]-benzenesulfonam ide (bosentan), and the ET A receptor antagonist 2-benzo[1,3]dioxol-5-yl-3-benzyl-4-(4-methoxy- phenyl)-4-oxo-but-2-enoic acid monosodium salt (PD155080), in the prevention of subarachnoid hemorrhage (SAH)-induced delayed cerebral vasospasm. Double hemorrhage in the rabbit constricted the basilar artery to 34% if control as determined by angiography. Oral bosentan and PD155080 administration after the initial SAH decreased the magnitude of constriction to 9% and 16% of control, respectively. Plasma and cerebrospinal fluid bosentan levels and plasma PD155808 levels were consistent with concentrations reported to inhibit ET-1 constriction of blood vessels in vitro. These results support the use of oral administration of ET A/B and ET A receptor antagonists as potential specific treatment for vasospasm resulting from SAH in humans.

The present study was undertaken to determine whether oxyhemoglobin (OxyHb) is responsible for the functional alterations in the cerebral arteries observed during chronic vasospasm after subarachnoid hemorrhage. Vascular strips of canine basilar arteries were kept in organ culture for 3 days with or without repetitive exposure to OxyHb (OxyHb-treated and control strips). Contractions elicited by high levels of potassium (80 mM) and uridine 5'-triphosphate (3 x 10(-4) M) were reduced in the OxyHb-treated group in a concentration-dependent manner. The relaxations evoked by nitric oxide and 8-bromo-cyclic guanosine monophosphate (8-bromo-cGMP) were not affected. Relaxations elicited by the calcium channel blocker, diltiazem, were attenuated in the OxyHb-treated rings. When the extracellular calcium concentration ([Ca2+]e) was changed from a concentration in the external solution of 10(-8) M to 10(-3) M, myogenic tension developed. Myogenic tension, expressed as a percentage of the maximum contraction in each segment, was augmented in the OxyHb-treated group at [Ca2+]e of 10(-5) M and 10(-4) M. There were no significant differences in passive compliance of the arterial wall between the two groups. These results demonstrated that prolonged exposure to OxyHb in vitro results in a decrease in contractile capacity and an increase in sensitivity to [Ca2+]e, in agreement with previous findings in spastic arteries. By contrast, impairment of the 8-bromo-cGMP-mediated relaxation pathway and increased stiffness of the arterial wall, which have been reported to occur in spastic arteries, were not induced by prolonged exposure to OxyHb in vitro.

Despite growing clinical use of transluminal balloon angioplasty (TBA) to treat cerebral vasospasm after aneurysmal subarachnoid hemorrhage (SAH), the precise mechanism of action of balloon dilation on the cerebral arterial wall is unknown. In this experiment the authors examined the pharmacological and morphological changes in 10 normal and 12 vasospastic canine basilar arteries following in vitro silicone microballoon TBA. For the SAH group in which the double-hemorrhage model was used, vasospasm was confirmed by angiography and the animals were killed on Day 7 after the first SAH. In vitro TBA was performed on basilar arteries from normal and SAH dogs immediately after sacrifice and removal of the brain. The procedure was performed while the arteries were maintained in oxygenated Krebs buffer. In the pharmacological studies, potassium chloride, prostaglandin F2 alpha, serotonin, and noradrenaline were used as vasoconstrictors, and bradykinin and calcium ionophore A23187 were used to produce an endothelium-dependent dilation. In both normal and vasospastic groups, the pharmacological responses of dilated segments of basilar arteries were compared to those of nondilated segments of the same arteries. Vessels from all groups were examined using scanning electron microscopy (EM) and transmission EM. Scanning EM was used to study the intact vessel wall, the smooth-muscle cell layer obtained after digestion with hydrochloric acid, and the extracellular matrix obtained after digestion with bleach. Cross-sections of the vessel wall were examined using transmission EM. The most striking finding was that immediately after in vitro TBA of both normal and vasospastic canine basilar arteries, there was a significant reduction (p < 0.05) of responses to both vasoconstrictors and vasorelaxants. As revealed by scanning EM and transmission EM, both normal and vasospastic vessels dilated with TBA showed flattening and patchy denudation of the endothelium, and straightening and occasional rupturing of the internal elastic lamina. In addition, vasospastic vessels dilated with TBA showed decreased surface rippling and mild stretching and straightening of smooth-muscle cells, and mild thinning of the tunica media. There was no gross vascular disruption or obvious change in the extracellular matrix of the vessel walls of either normal or vasospastic arteries after TBA. These results suggest that functional impairment of vasoreactivity in the vessel wall as a result of mechanical stretching of the smooth-muscle layer plays a more important role than structural alteration, at least in the immediate dilation produced in vasospastic arteries by TBA.

Calcitonin gene-related peptide (CGRP) is a potent vasodilator and a primary signaling molecule in neurovascular communication. In the present study, the authors examined cerebrovascular responses to CGRP and its related second messenger systems during cerebral vasospasm induced by subarachnoid hemorrhage (SAH). Tension measurements were performed in vitro on ring strips of basilar arteries obtained from rabbits subjected to artificial SAH and from control (non-SAH) animals. In vessels from SAH animals, which were preconstricted with serotonin, the vasorelaxant response to CGRP was attenuated. Because it has been suggested that vasodilation elicited by CGRP is mediated by cyclic 3',5'-adenosine monophosphate (cAMP) and/or cyclic 3',5'-guanosine monophosphate (cGMP), the vascular effects of directly activating these second messenger systems were also examined. The relaxant effect of forskolin, which activates adenylate cyclase directly, was slightly enhanced after SAH. In contrast, the relaxant effect of nitroglycerin (GTN), which activates soluble guanylate cyclase directly, was unchanged after SAH. The attenuation of CGRP-induced vasorelaxation could be the result of a modification in its ability to stimulate the production of second messengers. Experiments testing the capacity of CGRP to elevate cAMP levels showed no significant differences between vessels from non-SAH and SAH animals. Similarly, the resting levels of cAMP and the forskolin-induced elevations of cAMP did not differ between non-SAH and SAH animals. In contrast, cGMP levels were lower in resting and CGRP-treated vessels from SAH animals than in those from non-SAH animals. No significant differences in the levels of cGMP were observed between non-SAH and SAH vessels treated with GTN. This study indicates that CGRP-induced vasodilation is attenuated during vasospasm in a rabbit model of SAH. The findings also demonstrate that vasodilatory responses mediated by cAMP and cGMP are intact, although the levels of cGMP in SAH vessels are reduced. Together, these observations suggest that an attenuation in the capacity of vessels to dilate in response to CGRP occurs during cerebral vasospasm, and this change in CGRP vasoactivity is a result of modifications prior to, or independent of, the elevation of cyclic nucleotide second messengers.

Increasing evidence suggests that disturbances in the modulatory influence of the vasoactive peptide, calcitonin gene-related peptide (CGRP), contribute to the pathogenesis of cerebral vasospasm after subarachnoid hemorrhage (SAH). However, only limited success has been achieved in trials attempting to ameliorate vasospasm by modifying CGRP function. To better understand the potential utility of targeting CGRP-mediated relaxation, it is important both to identify the interactions CGRP may have with other elements of the vasospastic response and to characterize the mechanisms through which CGRP elicits vasodilative effects. The present studies examined the effects of CGRP on vascular responsiveness using tension measurements of ring strips of rabbit basilar artery maintained in vitro. Pretreatment of vessels with CGRP (100 nM) inhibited vasoconstrictor responses to the potent protein kinase C (PKC) activator, phorbol 12,13-dibutyrate (PDB). This particular contractile response was selected because PKC-mediated vasoconstriction is a critical component of the vasospastic response after SAH. In a posttreatment paradigm, CGRP was also found to reverse established constriction responses to PDB (2 nM) and histamine (3 microM) in a dose-dependent manner. When tested against the maximum effective dose of PDB (30 nM) in the posttreatment paradigm, CGRP (100 nM) did not elicit significant relaxation. However, after washing both of these drugs out of the test chamber, a persistent effect of CGRP was revealed: the decay of PDB-induced contraction was accelerated in vessels that had previously been treated with CGRP. These findings indicate that CGRP elicits both immediate and sustained influences on contractile responses mediated by PKC. Finally, two potential mechanisms for the vascular response to CGRP were examined. Adenosine triphosphate (ATP)-sensitive K+ channels do not appear to participate in CGRP-mediated dilation; inhibitors of these channels, glibenclamide and tolbutamide, did not block CGRP-induced relaxation. In contrast, a possible role for the nucleotide cyclic adenosine monophosphate (cAMP) in the vascular response to CGRP was indicated by the dose-dependent elevation of cAMP levels by CGRP. Together these studies indicate that CGRP can modulate the contractile response to PKC activation. These effects are associated with increases in the levels of cAMP, but occur independently of fluxes through ATP-sensitive K+ channels.

The authors investigated the roles of endothelin (ET)-1 and the ETA receptor in the pathogenesis of delayed cerebral vasospasm following subarachnoid hemorrhage (SAH). A study was made of the preventive effect of a novel ETA receptor antagonist, BQ-123, on vasospasm and the expression of the ETA receptor messenger ribonucleic acid (mRNA) using a canine two-hemorrhage SAH model. Continuous intrathecal administration of BQ-123 (5 x 10(-6) mol/day) prevented narrowing of the basilar artery on Day 7 after SAH in 97.6% of cases in the study group versus 70.7% of cases in the control group (p < 0.05). While expression of the mRNA-coding ETA receptor was not detected in the control animals, it markedly increased on Day 3 after SAH and was also detected on Day 7. The results suggest that endothelin-1 and the ETA receptor participate in the pathogenesis of delayed cerebral vasospasm following SAH.

The importance of factors within hemolysate in modulating oxyhemoglobin (oxyHb)-induced contraction was examined in an in vitro model of rabbit basilar arteries. When the basilar arteries were exposed to purified oxyHb alone, the contractile response observed was significantly weaker than that seen in arteries exposed to hemolysate containing an equal concentration of oxyHb. In order to delineate the nature of the factors within hemolysate that facilitate contraction, hemolysate was fractionated, and various components were tested individually for their ability to elicit this effect. A low-molecular-weight fraction of hemolysate, ranging from 0.5 to 2.0 kD, elicited only a mild contraction. However, when this fraction was combined with purified oxyHb, the contractile response was comparable in magnitude to that of unfractionated hemolysate. These studies confirm that purified oxyHb is capable of inducing contraction in vitro. The data also demonstrate that oxyHb elicits a significantly weaker contraction than does hemolysate. In addition, the results suggest that low-molecular-weight components in hemolysate (in the 0.5- to 2.0-kD range), while incapable of inducing a potent contraction alone, may act in concert with oxyHb to elicit the vasoconstriction seen following subarachnoid hemorrhage.

To investigate the function of nitric oxide (a major endothelium-derived relaxing factor) in cerebral arteries after subarachnoid hemorrhage (SAH) in vivo, several nitric oxide-related substances were administered to dogs that had undergone double SAH. These included L-arginine (a substrate for the formation of nitric oxide), NG-monomethyl-L-arginine (L-NMMA, an analog of L-arginine that inhibits the formation of nitric oxide from L-arginine), and superoxide dismutase (SOD, which protects nitric oxide from oxidation by superoxide anion), which were given via intracisternal injection. The diameter of the basilar artery was assessed angiographically. In intact dogs, intracisternal bolus injections of L-arginine (1, 10, or 100 mumol) produced a dose-dependent increase in the internal diameter of the basilar artery; conversely, L-NMMA reduced the diameter of the basilar artery from baseline in a dose-dependent manner. On Days 4 and 7, after two intracisternal injections of autologous blood, L-arginine produced transient vasodilation of the spastic basilar artery, whereas L-NMMA produced no significant vasoconstriction. The vasodilator effect of L-arginine after SAH was stronger on Day 4 than on Day 7, but less than in intact dogs. Intracisternal injection of SOD, which caused no effect per se, enhanced the duration of the vasodilator effect of L-arginine on the basilar artery on Day 4 and both the magnitude and duration of that effect on Day 7. Thus, the basal release of nitric oxide was impaired after SAH, but the ability to synthesize nitric oxide in the vascular wall was not abolished. The finding that the simultaneous injection of SOD enhanced and prolonged the vasodilation induced by sufficient exogenous L-arginine suggests that the inactivation of nitric oxide by superoxide anion contributes to the development of vasospasm.

Vasospasm was produced in the canine basilar artery by a two-hemorrhage method, while contraction was induced in the normal canine basilar artery by a local application of KCl or serotonin after transclival exposure. The control animals were injected with saline instead of fresh blood. The activation of mu-calpain, a Ca(++)-dependent neutral protease, in the basilar artery was studied by evaluating the conversion from its inactivated into its activated form on immunoblots. In addition, the activity of calpastatin, an intrinsic inhibitor of calpain, in the basilar artery was determined by assay. The majority of the mu-calpain was inactivated in the control group. In the spastic group, mu-calpain was generally activated markedly in the early stage of vasospasm and moderately thereafter. The contraction induced by KCl or serotonin application was classified into the early phasic and the later tonic stages; mu-calpain was usually activated in the phasic stage and inactivated in the tonic stage. Calpastatin activity was significantly decreased during vasospasm, whereas it was not significantly changed in KCl- or serotonin-induced contraction. The final activity of mu-calpain results from the balance of mu-calpain and calpastatin. This suggests that mu-calpain activity was enhanced continuously in the spastic group and transiently in the KCl or serotonin group, and that the continuous activation of mu-calpain during vasospasm probably induced more proteolytic changes compared to those in the KCl or serotonin group.

Cerebral vasospasm remains an unpredictable and inadequately treated complication of aneurysmal subarachnoid hemorrhage. To date, pharmacological treatment has been plagued in part by an inability to attain sufficiently high concentrations of vasodilator drug in the cerebrospinal fluid without precipitating systemic side effects such as hypotension. To circumvent this limitation of current pharmacological therapy, the authors have developed a sustained-release preparation of papaverine that can be implanted intracranially at the time of surgery for aneurysm clipping. In vitro evaluation of drug-release kinetics has demonstrated that reliable, sustained release of effective amounts of papaverine is possible. An in vitro bioassay using isolated preparations of canine basilar artery has confirmed the biological activity of this preparation. These in vitro studies are described.

In vivo experiments on the vasoactive effects of vasopressin and oxytocin on cerebral circulation were carried out in anesthetized dogs, using an electromagnetic flowmeter to measure vertebral blood flow and angiography to measure the internal diameter of the basilar artery. Direct bolus infusion of 1 pmol to 1 nmol of vasopressin or 10 pmol to 10 nmol of oxytocin into a femoral-vertebral artery shunt produced a dose-dependent decrease in vertebral artery blood flow without significantly affecting mean arterial blood pressure. Vasopressin was more potent than endothelin and neuropeptide Y, which have also been demonstrated to induce long-lasting decreases in vertebral artery blood flow. However, direct bolus infusion of vasopressin (100 pmol and 1 nmol) or oxytocin (1 nmol and 10 nmol) into the vertebral artery dilated major vessels including the vertebral, anterior spinal, and basilar arteries, as well as the circle of Willis and its main branches, while endothelin (1 nmol) and neuropeptide Y (5 nmol) caused no change in the diameters of major cerebral arteries. The V1 antagonist d(CH2)5tyrosine(methyl) arginine vasopressin suppressed the effects of both vasopressin and oxytocin. Vasopressin was over 10 times as potent as oxytocin in both assays. The vasodilatory effect of vasopressin, which may be mediated by an endothelium-dependent mechanism, was functionally damaged in dogs after experimental subarachnoid hemorrhage. These data suggest regional differences in the sensitivity and responsiveness of vasculature to vasopressin and oxytocin, and specifically that both peptides act through V1 receptors to decrease the resistance of large vessels and increase the resistance of small vessels.

The effect of hemoglobin on the vasodilatory effect of calcium antagonists was studied in isolated rabbit basilar arteries using an isometric tension measurement method. The ability of nimodipine to relax or inhibit contractions elicited by high K+ depolarization or serotonin (5-hydroxytryptamine, 5-HT) was investigated in control arterial rings and rings pretreated by hemoglobin. Hemoglobin (10(-6) and 10(-5) M) reduced the relaxation induced by nimodipine (10(-10) to 10(-8) M) in the rings contracted by 40 mM KCl. This reduction in relaxation was also observed with 3 x 10(-10) to 3 x 10(-9) M nicardipine, 3 x 10(-8) to 3 x 10(-7) M verapamil, and 10(-7) to 10(-6) M diltiazem. On the other hand, the effect of nimodipine was not influenced by endothelial removal or by pretreatment with 10(-5) M albumin or 10(-6) M prostaglandin F2 alpha. Hemoglobin restored the 10(-10) and 10(-9) M nimodipine-induced inhibition of the contraction elicited by CaCl2 (0.3 to 20 mM) in a K(+)-rich, Ca(++)-free solution. This restoration was greater at higher concentrations of CaCl2. Hemoglobin enhanced both the nimodipine-sensitive tonic phase and the less sensitive phasic phase of contractions produced by 10(-6) M of 5-HT. It abolished the inhibitory effect of 10(-8) and 10(-7) M nimodipine on the phasic contraction. Endothelial removal also enhanced both phases of the contraction, but did not abolish the effect of nimodipine. This study showed that the vasodilatory effect of calcium antagonists, especially nimodipine, on the vasoconstriction induced by other vasoactive substances decreased in the presence of hemoglobin.

Vasospasm was produced in adult mongrel dogs by a two-hemorrhage method, and the spastic basilar arteries were exposed via the transclival route on Day 7. Tonic contraction was produced in the normal canine basilar arteries by a local application of KCl or serotonin after transclival exposure. The exposed spastic and tonic basilar arteries then received a topical application of the following: 1-(5-isoquinolinesulfonyl)-2-methyl-piperazine (H-7), a potent inhibitor of protein kinase C acting at the catalytic domain; calphostin C, a specific inhibitor of protein kinase C acting at the regulatory domain; or calpeptin, a selective inhibitor of calpain. Both spastic and tonic basilar arteries were effectively dilated by H-7. Calphostin C caused only slight dilation of spastic basilar arteries but moderate dilation of tonic basilar arteries. Dilation in response to calpeptin was remarkable in the spastic basilar arteries but slight in the tonic basilar arteries. The doses of calphostin C and calpeptin required to obtain maximum effect were markedly lower in the tonic model than in the spastic model. The spastic and tonic models had a similar dose-dependent response to H-7 but quite a different response to calphostin C or calpeptin, suggesting a difference in the function of protein kinase C and calpain in the two models. Furthermore, the effect of calphostin C on the reversal of vasospasm was increased significantly after topical treatment with calpeptin. It is suggested that the majority of the catalytic domain of protein kinase C is dissociated from the regulatory domain, probably by a limited proteolysis with calpain, and is markedly activated in vasospasm.

The role of endothelin, a newly found vasoconstrictor peptide, is examined in the pathogenesis of cerebral vasospasm after experimental subarachnoid hemorrhage (SAH) in the dog. Endothelin immunoreactivity was overexpressed in the endothelium of the vasospastic basilar artery. Because endothelin synthesis is regulated at the messenger ribonucleic acid transcription level, the effect of actinomycin D, a ribonucleic acid synthesis inhibitor, was studied as a means of preventing vasospasm. It was found that treatment with intravenous actinomycin D for 5 days beginning on the day of SAH completely inhibited the development of vasospasm. This novel experimental therapy may lead not only to the elucidation of the pathogenesis of cerebral vasospasm but also to the availability of a prophylactic adjuvant therapy for patients with SAH.

The vascular relaxation effects of calcitonin gene-related peptide (CGRP) and vasoactive intestinal polypeptide (VIP) on the dog basilar artery after experimentally produced subarachnoid hemorrhage (SAH) were examined in vitro by an isometric tension recording method. Both CGRP and VIP induced dose-dependent relaxations in ring segments of the intact basilar artery of control dogs. The vasorelaxant action of CGRP was more potent than that of VIP. The single-injection model of SAH was produced by injection of fresh autologous arterial blood (1 ml/kg body weight) into the cisterna magna on Day 0 of the post-SAH period, and the double-injection model was produced by two injections of blood (0.5 ml/kg each) on Days 0 and 2. Narrowing of the basilar arteries on vertebral angiograms was most prominent on Day 3 or 7 in the single- or double-injection model, respectively. Relaxation of the basilar artery induced by CGRP and VIP was to some extent decreased on Days 3 and 7 of the post-SAH period in the single-injection model, and on Days 7 and 14 in the double-injection model. However, the vasorelaxant effects of CGRP and VIP were significantly enhanced on Day 14 of the post-SAH period in the single-injection model, and on Days 28 and 42 in the double-injection model. Subsequently, these effects returned to control levels by Days 28 or 63 in the single- or double-injection model, respectively.

The effect of subarachnoid hemorrhage (SAH) on endothelium-dependent vasodilation of the isolated rabbit basilar artery was examined using an isometric tension recording method. The SAH was induced by injecting 5 ml of fresh arterial blood into the cisterna magna. Sixty-two rabbits were separated into four groups according to the timing of sacrifice: control rabbits, and operated rabbits sacrificed on Days 2, 4, and 6 after SAH. Acetylcholine (ACh) (10(-7) M to 10(-4) M) and adenosine triphosphate (ATP) (10(-7) M to 10(-4) M) were used to evoke dose-dependent vasodilation of isolated arterial rings previously contracted by 10(-6) M serotonin (5-HT). There were no significant differences in the vasodilatory response to ACh among these four groups. Relaxation to approximately 84% of the initial contractile tone occurred with 10(-4) M ACh. On the other hand, the vasodilatory response to ATP was suppressed in the animals sacrificed 2 days after SAH; the relaxation of this group was approximately 52% at 10(-4) M ATP, compared to a relaxation of 87% observed in the other groups of animals. One of the major causes of the impairment of endothelium-dependent vasodilation seems to be an inhibition of the production of endothelium-derived relaxing factor by endothelial cells. After the relaxation studies, the dose-response curves for 5-HT were obtained. Serotonin caused significantly more contraction in the animals sacrificed 2 days after SAH than in the other groups. The present experiments suggest that impairment of the endothelium-dependent vasodilation following SAH, together with the potentiation of the contractile response to vasoactive agents in cerebral arteries, may play an important role in the pathogenesis of vasospasm.

The authors have studied the ability of prostacyclin to reverse contractions of human basilar arteries in vitro that were induced by a wide range of substances implicated in the etiology of cerebral arterial spasm. Prostacyclin (10(-10) to 10(-6)M) caused a dose-related reversal of contractions induced by 5-hydroxytryptamine, noradrenaline, angiotensin II, prostaglandin (PG)F2 alpha, and U-46619 (a thromboxane-A2 mimetic). These agents were tested at concentrations or volumes that produced almost maximum or maximum responses and those that produced approximately 50% of the maximum response. Contractions induced by maximum concentrations of angiotensin II and U-46619 were least affected by prostacyclin. In addition, contractions induced by thromboxane-A2 generated from guinea-pig lung were reversed in a dose-dependent fashion by prostacyclin. This ability of prostacyclin to physiologically antagonize contractions of the human basilar artery in vitro induced by high concentrations of various spasmogenic agents suggests that such a potent vasodilator agent or more stable analogue may be of value in the treatment of such disorders as cerebral arterial spasm following subarachnoid hemorrhage.

The in vivo spasmogenic capacity of a lipid hydroperoxide (15-hydroperoxy arachidonic acid: 15-HPAA) was studied in a chronic experiment using the dog. The 15-HPAA was injected into the cisterna magna (0.2 or 2 mg emulsified in bovine serum albumin solution). The changes in diameter of the basilar artery were followed by angiography, and the morphological changes were studied by electron microscopy. The cisternal injection of 0.2 mg of 15-HPAA caused a mild constriction of the basilar artery which lasted about 7 hours. The cisternal injection of 2 mg of 15-HPAA caused a biphasic constriction, the initial phase of which was a moderate narrowing lasting about 10 hours. The second phase started on the 2nd or the 3rd day after injection. The intensity of the arterial narrowing was more pronounced in the second phase than in the first. The prolonged secondary constriction of the basilar artery continued until sacrifice on the 7th day after injection. Electron microscopic study revealed a marked degenerative change in the endothelium and myonecrotic changes in the tunica media. The prolonged arterial constriction in the second phase was invariably associated with remarkable degeneration of the endothelium. On the other hand, myonecrotic changes were limited to a small number of smooth-muscle cells. The results of the present study are consonant with the hypothesis that lipid peroxidation associated with lysis of the subarachnoid clot is involved in the genesis of chronic vasospasm in subarachnoid hemorrhage.

In vitro experiments were performed with a small volume chamber to determine the contractile activity of various vasoactive agents on human basilar and anterior cerebral arteries. Cumulative log-dose response curves were obtained for most of the agents tested including serotonin and three different prostaglandins; many of these curves were found to be similar to curves previously obtained with canine cerebral arteries. It was concluded from these similarities that canine cerebral arteries are a good in vitro model for studying human cerebral arterial spasm. It was also demonstrated that human cerebrospinal fluid, collected up to 17 days after a subarachnoid hemorrhage from patients with clinical and angiographic evidence of cerebral arterial spasm, would cause large, dose-dependent contractions in human anterior cerebral arteries.

To investigate the alteration of endothelium-dependent responses in chronic vasospasm after subarachnoid hemorrhage (SAH), experiments were carried out in the double-hemorrhage canine model. After the presence of vasospasm was confirmed by cerebral angiography on Days 0 and 7, pharmacological studies on the basilar artery were conducted in vitro on Day 8. In the SAH group, endothelium-dependent relaxation was abolished in response to arginine vasopressin and was significantly reduced in response to thrombin. Endothelium-independent relaxation in the SAH group was preserved in response to papaverine and was minimally reduced in response to sodium nitroprusside. Endothelium-dependent contraction in response to arachidonic acid, acetylcholine, the calcium ionophore A23187, adenosine diphosphate, mechanical stretching, and hypoxia persisted in the SAH group. The maximal contraction to KCl and uridine triphosphate, which is endothelium-independent, was diminished in the SAH group, but not changes in sensitivity were noted in the concentration-response relationships. A significant correlation was observed between the degree of vasospasm determined angiographically and the loss of endothelium-dependent relaxation. The loss of endothelium-dependent relaxation and the persistence of endothelium-dependent contraction suggest that the deterioration in the endothelium-dependent responses may be an important component in the pathogenesis of cerebral vasospasm.

The effect of endothelium removal on the contractile responses to KCl, hemoglobin, serotonin (5-HT), norepinephrine (NE), prostaglandin (PG)F2 alpha, PGD2, and PGE2 was investigated in canine and rabbit basilar arteries by an isometric tension-recording method. In canine basilar arteries, endothelium removal elevated the dose-response curves to 5-HT, PGF2 alpha, and PGD2, and PGE2, but not to KCl, hemoglobin, or NE. In rabbit basilar arteries, on the other hand, removal of the endothelium elevated the dose-response curves to 5-HT, NE, PGF2 alpha, and PGD2, but not to KCl or hemoglobin. Neither contractile nor inhibitory response was elicited by PGE2 in rabbit basilar arteries. Contraction induced by 5-HT and NE following endothelium removal had a much more pronounced effect in rabbit basilar arteries than in canine basilar arteries. These results suggest that, following endothelium removal, abolition of the spontaneous release of endothelium-derived relaxing factor is the most probable mechanism of the enhanced vasocontraction. Since endothelial damage results from subarachnoid hemorrhage, the aforementioned mechanism of vasocontraction enhancement may play a role in the pathogenesis of cerebral vasospasm.

The effects of aspirin and meclofenamic acid on the diphasic responses produced by prostacyclin in isolated canine basilar arteries were compared. Meclofenamate enhanced the relaxant action of prostacyclin in low concentrations (10(-8)M to 10(-6)M) and also significantly inhibited the contractions caused by the highest concentration (10(-5)M) of this prostaglandin. Aspirin had no such effects. The results indicate that some drugs classified as prostaglandin synthetase inhibitors can directly enhance the vasodilator action of prostacyclin on cerebral arteries, and suggest that this enhancement could be of value in the clinical applications of prostacyclin.

In vitro experiments were performed using a small volume chamber to study serotonin-induced contractions of the canine basilar artery. Temperature was found to have a profound effect on the artery's response to serotonin. Raising the temperature to 40 degrees C (104 degrees F) increased the maximum response by 20% and lowering the temperature by 10 degrees C caused a 40% reduction in the maximum contraction. Cumulative log-dose response curves for analogues of serotonin demonstrated a high degree of specificity for the serotonin receptor and large nonphysiological concentrations of serotonin caused relaxation of the contracted artery. Extracellular calcium was shown to be an absolute requirement for serotonin-induced contractions. Extracellular magnesium, in contrast, was shown to inhibit serotonin-induced contractions.

Vascular contractions in response to KCl and serotonin (5-hydroxytryptamine, 5-HT) in rabbit basilar artery were studied in vitro using an isometric tension-measurement technique. Hemoglobin (10(-5)M) markedly augmented contractions induced by 5-HT (10(-9) to 10(-6)M) and slightly augmented those induced by KCl (20 to 80 mM) in arteries with intact endothelium. On the other hand, the augmentation induced by hemoglobin was almost abolished in arteries that were chemically denuded of endothelial cells by pretreatment with saponin. Since hemoglobin is known to be a selective inhibitor of endothelium-derived relaxing factor (EDRF), it is possible that the augmentation of contraction by hemoglobin in endothelium-intact arteries was mediated via an inhibition of spontaneously released EDRF. The effect of subarachnoid hemorrhage (SAH) on spontaneously released EDRF was investigated by injecting 5 ml of blood into the cisterna magna and sacrificing the rabbits 2 days later. Arteries after SAH showed a significant reduction in hemoglobin-induced augmentation compared to that seen in control arteries with intact endothelium. This result suggests that spontaneously released EDRF is significantly reduced after SAH. It is concluded that EDRF is released spontaneously in the rabbit basilar artery and that inhibition of its release might be involved in pathogenesis of cerebral vasospasm.

Although a number of substances which participate in the physiological control of vascular caliber have been proposed to play a causative role in cerebral vasospasm, none of these has been shown to induce the profound, sustained degree of constriction or the vasculopathy that characterize this disorder. The present study was undertaken to determine whether bilirubin, a hemoglobin breakdown product with detergent-like activity, accrued in incubated blood or in intracranial hematoma and whether topical application of bilirubin altered the caliber or morphology of cerebral arteries. As a model of blood residing in the subarachnoid space, sterile vials of cat blood and of human blood were incubated in vitro at 37 degrees C. The concentration of bilirubin in the supernatant of cat blood increased from a value of 0.27 +/- 0.3 mg% (mean +/- standard error of the mean) on Day 0 to a value of 9.57 +/- 2.4 mg% on Day 10; the respective values for human blood were 0.53 +/- 0.02 mg% and 13.4 +/- 5.4 mg%. Samples of intracranial hematoma that had been surgically removed from three patients between 4 and 11 days after hemorrhage yielded bilirubin levels from 2.1 to 15.2 mg%. Application of a bilirubin suspension of 5, 10, or 20 mg% in buffered Ringer's solution to cat basilar artery in vivo led to progressive and sustained constriction; at 4 hours the mean decrease in the width of the blood column was 34% +/- 2.1%. Ultrastructural analysis of these vessels showed widespread pathological changes similar to those associated with cerebral vasospasm. Application of a 10-mg% bilirubin suspension to the basilar artery of two baboons produced similar alterations. Minimal changes in diameter or morphology were observed in cat arteries bathed in buffer solution alone. These findings in animal models indicate that bilirubin may play a central role in the development of cerebral vasospasm.

Arteriograms of the basilar artery reveal that dopamine given intracisternally to dogs can generate cerebral vasospasm. This finding supports a recent hypothesis of others that dopamine may play a role in the pathogenesis of vasospasm, especially since many substances are known which fail to produce such spasm. Compared to blood or prostaglandin E2, however, the spasm induced by dopamine was delayed in onset, less in incidence, and usually less intense. Possible explanations for such experimental differences are discussed.

The ability of antithrombin III, an endogenous plasma glycoprotein, to reverse the arterial narrowing in a rabbit model of cerebral vasospasm was evaluated. The vasodilator activity of antithrombin III on rabbit arteries was first assessed in vitro using a myograph-arterial ring preparation. Antithrombin III (10 IU/ml) induced a 55.4% +/- 2.66% (mean +/- standard error of the mean) relaxation in basilar artery precontracted with serotonin (5-HT) in five specimens as compared with a 9.8% +/- 1.6% relaxation of common carotid artery in six specimens. For in vivo analysis, 21 New Zealand White male rabbits were separated into three groups: Group 1 served as normal controls; Group 2 received a subarachnoid blood injection (SAH) and were sacrificed on Day 3 thereafter; and Group 3 animals were subjected to SAH, then received a 2-hour intracisternal infusion of antithrombin III (100 IU) in saline prior to sacrifice on Day 3. Basilar artery caliber was determined using a morphometric method to analyze perfusion-fixed arterial segments. Control basilar artery diameter in Group 1 was 0.64 +/- 0.02 mm. In Group 2 a 27% reduction in arterial caliber to 0.47 +/- 0.03 mm was observed by Day 3 post SAH (p less than 0.0001). Group 3 animals had a mean basilar artery diameter of 0.68 +/- 0.02 mm. This was significantly larger than the untreated SAH rabbits in Group 2 (p less than 0.0001), but not different from control artery diameters in Group 1. The findings demonstrate that antithrombin III in saline has a significant ability to reverse delayed narrowing of the rabbit basilar artery after SAH.

Available data indicate that the concentration of prostaglandin F2alpha required to produce arterial spasm in an experimental model is approximately a thousand-fold higher than the concentration that occurs under physiological conditions. The spasmogenic platelet factor which has been previously described is shown to be a substance other than prostaglandin F2alpha because of differences in susceptibility of the two substances to enzymatic digestion.

This study examines the relaxant effect of calcitonin gene-related peptide (CGRP), a 37-amino acid peptide with a potent vasodilator action, on cerebral arterial spasm after subarachnoid hemorrhage (SAH). The spasm was induced by injecting autologous arterial blood percutaneously into the cisterna magna in adult mongrel dogs. The single-injection model of SAH was produced by injection of 1.0 ml/kg body weight of blood (on Day 0), and the double-injection model involved two successive injections of 0.5 ml/kg body weight of blood made 48 hours apart (on Day 0 and Day 2). On vertebral angiograms, arterial narrowing of the major cerebral arteries was most prominent on Day 3 after SAH in the single-injection model and on Day 7 in the double-injection model. When 10(-10) mol/kg of CGRP was administered intracisternally in the single-injection model on Day 3, the diameter of the spastic cerebral arteries, as determined by angiography, recovered to normal. After intracisternal administration of 10(-11) to 2 X 10(-10) mol/kg of CGRP on Day 7 in double-injection models, spastic cerebral arteries dilated in a dose-dependent manner. The dilatory effect of CGRP continued for a few hours after administration. The results suggest that CGRP injected intracisternally may reverse cerebral arterial spasm after SAH.

The purpose of this experiment was to evaluate the effect of hypoxia on the in vitro contractile responses of canine basilar artery to KCl, prostaglandin (PG) F2 alpha, and hemoglobin. Hypoxia was induced by changing the bubbling gas mixture in the chamber from 95% O2/5% CO2 to 95% N2/5% CO2. Hypoxia augmented the contractile response developed at 95% O2 to 25 mM and 50 mM KCl, 3 X 10(-7) M and 10(-5) M PGF2 alpha, and 10(-6) M hemoglobin. No significant alteration of the hypoxic augmentation in any preparation exposed to 25 mM KCl, 3 X 10(-7) M PGF2 alpha, or 10(-6) M hemoglobin was observed with guanethidine (10(-5) M), prazosin (10(-5) M), methysergide (10(-5) M), or diphenhydramine (10(-5) M). Endothelial denudation did not affect hypoxic augmentation. Hypoxia did not cause any alteration of the contractile response to 10(-6) M PGF2 alpha in Ca++-free media. Pretreatment with a calcium channel blocker, nicardipine, significantly inhibited the hypoxic potentiation of the contractile response to 25 mM KCl, 3 X 10(-7) M PGF2 alpha, and 10(-6) M hemoglobin. These results suggest that hypoxia augments the contractile response to these agonists by a direct action on the smooth-muscle cells, facilitating the transmembrane influx of extracellular calcium. Hypoxia of smooth-muscle cells in the major cerebral arteries might be involved in the pathogenesis of vasospasm.