Cell Culture
Cloned bovine aortic endothelial cells, which were identified as endothelilal cells by morphology (cobblestone appearance) and positive staining for Factor VIII, were kindly provided by Drs. Corinne Gajdusek and Marc R. Mayberg (University of Washington, Seattle, Washington).

Cell density study
Cytotoxic effect of OxyHb was determined by measuring the cell density. Confluent endothelial cells in 24-well culture plate were incubated with 1-100 mM OxyHb or saline in serum free media for 3, 6, 12, and 24 hours. After incubation, cells were fixed with 70 % ethanol for 10 minutes and stained with hematoxylin. Photomicrographs were taken of fixed cells at 100X magnification, representing 970 x 640 mm fields chosen at random. Stained cell nuclei were counted on photographic prints, and the density of adherent cells was determined as total nuclei per field.

Analysis of DNA-fragmentation (DNA ladder)
DNA-fragmentation analysis was assessed following modifications of previously described methods. Confluent endothelial cells (100 mm dish) were incubated with 1-100 mM OxyHb or saline in serum free media for 3, 6, 12, and 24 hours. At the harvesting times, cells were scraped and collected by centrifugation 500 g for 5 minutes at 4°C. Cells were then resuspended in 10 mM Tris-HCl pH 8.0, 10 mM EDTA, 150 mM NaCl, 0.5 % sodium dodecyl sulfate and 100 mg/ml of proteinase K and were incubated overnight at 50°C. DNA was extracted with equal volumes of phenol/chloroform/isoamyl alcohol (25:24:1) precipitated at room temperature for 10 minutes by adding 0.1 volumes of 3 M sodium acetate, pH 5.2, and 2 volumes of ethanol (100%). The samples were submitted to a RNase treatment (10 mg/ml) for 1 hour at 37 °C, followed by extraction and ethanol precipitation, as mentioned above. DNA was rinsed with 70 % ethanol and resuspended in 10 mM Tris-HCl pH 8.0 and 1 mM EDTA pH 8.0. DNA was subjected to electrophoresis on a 1.0 % agarose gel, stained with ethdium bromide, and visualized under UV light.

Transmission Electron Microscopy
Confluent endothelial cells (100 mm dish) were incubated with 100 mM OxyHb or saline in serum free media for 24 hours. Cells were then scraped and collected by centrifugation 500 g for 5 minutes at 4 °C. Cells were fixed with 2.5% gluteraldehyde and embedded in Epon. Ultrathin sections were stained with uranyl acetate and lead citrate and examined with transmission electron microscopy (LEO 906, LEO, Thornwood, NY).

Western Blotting
Western blotting was performed following modifications of previously described methods. Confluent endothelial cells (60 mm dish) were washed with serum free media twice and incubated with 100 mM OxyHb or saline in serum free media for 24 hours. Cells were scraped and suspended in cold PBS twice, and were collected by centrifugation 500 g for 5 minutes at 4°C. Cells were then lysed in 1 % IGEPAL CA-630, 50 mM Tris-HCl pH 7.5, 100 mM NaCl, 5 mM EDTA, 1.0 mM PMSF for 20 minutes at 4°C, and the insoluble materials were removed by centrifugation at 13,000 g for 10 minutes at 4°C. Samples were applied to 12 % SDS-PAGE. After electrophoretic transfer of the separated polypeptides to nitrocellulose membrane, the membranes were blocked for 1 hour using 5 % non-fat milk in Tween-PBS (PBS containing 0.1 % Tween 20), washed with Tween-PBS and incubated at room temperature for 2 hours in a 1:1000 dilution of mouse anti-poly (ADP-ribose) polymerase (PARP) antibody.12 Membranes were later washed with Tween-PBS and incubated with 1:1000 dilution of anti mouse Ig antibody, linked with horseradish peroxidase. The enhanced chemiluminescence system was used for detection.

Data Analysis
Data are expressed as mean ± SD. Statistical differences between the control and other groups were compared using one-way analysis of variance (ANOVA) and Scheffé's method (95% lower and upper confidence interval) if significant variance was found. A value of P<0.05 was considered statistically significant.

Cell density
OxyHb produced a significant decrease in cell density in a concentration- and time-dependent manner in cultured endothelial cells.

DNA degradation - DNA ladder
Fragmented DNA as demonstrated by nucleosomal laddering in endothelial cells induced by OxyHb. The effect of OxyHb-induced DNA laddering was again in a concentration- and time-dependent fashion. Apoptotic changes, represented by DNA ladders, can be induced by higher concentration of OxyHb (100 mM) in 6 hours or by lower concentration of OxyHb (1 mM) in 24 hours.

Transmission Electron Microscopy - Apoptotic body
Transmission electron microscopy is another specific method to identify apoptotic changes in cells. Ultrastructual analysis of endothelial cells incubated with saline or OxyHb (100 mM) for 24 hours revealed morphologic changes of endothelial cells that were typical of the steps of apoptotic changes.

Western Blotting
Another feature in apoptosis is the cleavage of 116 kDa PARP to 85 kDa in cells. Using Western blotting with PARP antibody, it was demonstrated that 116 kDa PARP was cleaved to 85 kDa, an apoptosis-related fragment, in endothelial cells treated with OxyHb (100 mM) for 24 hours. No PARP cleavage was observed in endothelial cells treated with saline.

Summary Structural changes in the arterial wall after SAH have been reported and believed to contribute to cerebral vasospasm. Detachment and vacularization of endothelial cells and necrotic changes of the medial smooth muscle cells are the most striking histological alterations. Whether these pathological changes are secondary to the prolonged contraction or the results of direct effects of subarachnoid blood clot is still debatable. There is a line of evidence showing OxyHb produces direct cytotoxicity in smooth muscle or endothelial cells. In feline cerebral arteries, OxyHb caused myonecrosis, transformation of nerve endings, invasion of myointimal cells into the tunica intima, changes in endothelial cell basement membrane, and the detachment of endothelial cells.

Other pathological changes induced by OxyHb such as craters, blebs and vacuoles in endothelial cells and vacuoles, nuclear pyknosis, and mitochondria degeneration in smooth muscle cells were documented in rat basilar artery. In addition, OxyHb produce d vaculation and detachment in cultured bovine internal carotid endothelial cells, under a similar culturing condition as reported in this study. The direct cytotoxicity of OxyHb to endothelial or smooth muscle cells may play a role in the pathogenesis of cerebral vasospasm. However, little is known about the pattern of cell death, necrosis or apoptosis, as a final stage of cell damage, in either smooth muscle or endothelial cells.

We have demonstrated in this study that OxyHb induced concentration- and time-dependent decrease in cell density (as shown by cell detachment) in cultured bovine aortic endothelial cells. This result is consistent with the reports by others using endothelial cells from bovine internal carotid artery. The pattern of cell death induced by OxyHb is clearly apoptosis in endothelial cells as shown by three selective methods to identify apoptosis. (1) DNA ladder is the hallmark of apoptotic changes. OxyHb produced concentration- and time-dependent DNA fragmentation in endothelial cells. (2) Apoptotic body is another feature of apoptosis that can be demonstrated by transmission electron microscopy. OxyHb induced condensation and margination of chromatin, nuclear fragmentation and convolution of the cellular membrane, crowded and condensed apoptotic bodies in endothelial cells. (3) Western blotting of cleavage of 116 kDa PARP to 85 kDa, an apoptosis-related fragment, is also a selective marker for apoptosis. OxyHb cl early cleaved 116 kDa PARP to 85 kDa in endothelial cells.

Two fundamentally different forms of cell death, necrosis and apoptosis, have been defined in terms of morphology and biochemistry. Necrosis is a degenerative phenomenon produced by major environmental change such as severe ischemia, extremes of temperature, and mechanical trauma. The characteristics of necrosis are swelling of cytoplasm and organelle, balloon-like degeneration and early rupture of plasma membrane. Lyzosomal enzymes released after necrosis cleaved DNA randomly into fragments that displaying a smear pattern.

Apoptosis, in contrast, is an active biological process of eliminating unwanted cells involved in the regulation of cell number under physiological and pathological condition. Apoptosis is associated with early and prominent condensation of nuclear chromatin, decreasing cell size and ultimate cleavage of DNA by a specific endonuclease into mono-, and polynucleosomal fragments. The present study shows that exposure of endothe lial cells to OxyHb results in concentration- and time-dependent decrease of cell density. Dying cells exhibit condensation of nuclei and apoptotic bodies, inter-nucleosomal DNA fragmentation, and PARP cleavage associated with apoptosis.

The mechanism of apoptosis induced by OxyHb in endothelial cells is not clear. One of the possible explanations for OxyHb-induced apoptosis is free radical generation. Normally, OxyHb undergoes a slow, but spontaneous oxidation-reduction reaction, in which the oxygen is reduced to superoxide. This reaction is the basis for the oxidation of OxyHb to met-hemoglobin and is the main source of superoxide and hydrogen peroxide. Those free radicals may initiate and propagate lipid peroxidation by the Haber-Weiss reaction and Fenton chemistry. Free radicals are important factors for the induction of apoptosis in other tissues.

Another possible mechanism for apoptosis is the elevation of intracellular Ca2+ concentration. Sustained elevation of intracellular Ca2+ stimulates an endogenous endonuclease activity and initiates apoptosis in other tissues. Erythrocyte lysate increases [Ca2+]i by release Ca2+ from internal stores and promote Ca2+ influx from voltage-independent Ca2+ channels in cerebral endothelial cells. Other components of blood clot may be involved in apoptosis. Extracellular ATP, another main components of erythrocytes, causes apoptosis in pulmonary artery endothelial cells, as assessed by morphological changes and inter-nucleosomal DNA degradation. Interestingly, both OxyHb and ATP can be released from blood clot and they act not only as vasoconstrictors but also as promoters for apoptosis. TGF-b1, which is major component of platelets, accelerates non-muscle compaction, inhibits bcl-2 expression and induces apoptosis in cultured umbilical vein endothelial cells.

Since Kerr et al proposed the term apoptosis for cell death in 1972, apoptotic changes have not been observed in endothelial cells in spastic arteries after SAH. One possibility is that, once apoptosis occurs in endothelial cells, those damaged cells may be detached from the internal elastic lamina and be washed away by blood flow. Another possibility is that simply no attention was paid to look for apoptotic changes in any layers of the spastic arteries. Vascular endothelial cells are not only a physical and metabolic barrier but play an important role in the regulation of vascular tone. Endothelial cells generate both vasoconstricting and vasodilating agents and those agents such as endothelin8 and nitric oxide have been documented to contribute to cerebral vasospasm. Destruction of endothelial cells also exposes smooth muscle cells directly to blood flow that may contain vasoconstrictors such as OxyHb.

Comair YG, Schipper HM, Brem S: The prevention of oxyhemoglobin-induced endothelial and smooth muscle cytoskeletal injury by deferoxamine. Neurosurgery 32:58-64, 1993

Dawicki DD, Chatterjee D, Wyche J, et al: Extracellular ATP and adenosine cause apoptosis of pulmonary artery endothelial cells. Am J Physiol 273:L485-L494, 1997

Findlay JM, Weir BK, Kanamaru K, et al: Arterial wall changes in cerebral vasospasm. Neurosurgery 25:736-745, 1989

Foley PL, Kassell NF, Hudson SB, et al: Hemoglobin penetration in the wall of the rabbit basilar artery after subarachnoid hemorrhage and intracisternal hemoglobin injection. Acta Neurochir 123:82-86, 1993

Kaufmann SH, Desnoyers S, Ottaviano Y, et al: Specific proteolytic cleavage of poly(ADP-ribose) polymerase: an early marker of chemotherapy-induced apoptosis. Cancer Res 53:3976-3985, 1993

Kerr JF, Wyllie AH, Currie AR: Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 26:239-257, 1972

Li PF, Dietz R, von Harsdorf R: Differential effect of hydrogen peroxide and superoxide anion on apoptosis and proliferation of vascular smooth muscle cells. Circulation 96:3602-3609, 19 97

Mayberg MR, Okada T, Bark DH: The significance of morphological changes in cerebral arteries after subarachnoid hemorrhage. J Neurosurg 72:626-633, 1990

Misra HP, Fridovich I: The generation of superoxide radical during the autoxidation of hemoglobin. J Biol Chem 247:6960-6962, 1972

Pluta RM, Afshar JK, Boock RJ, et al: Temporal changes in perivascular concentrations of oxyhemoglobin, deoxyhemoglobin, and methemoglobin after subarachnoid hemorrhage. J Neurosurg 88:557-561, 1998

Savitz SI, Rosenbaum DM: Apoptosis in neurological disease. Neurosurgery 42:555-572, 1998

Takenaka K, Kassell NF, Foley PL, et al: Oxyhemoglobin-induced cytotoxicity and arachidonic acid release in cultured bovine endothelial cells. Stroke 24:839-845, 1993

Tsukada T, Eguchi K, Migita K, et al: Transforming growth factor beta 1 induces apoptotic cell death in cultured human umbilical vein endothelial cells with down-regulated expression of bcl-2. Biochem Biophys Res Commun 210:1076-1082, 1995

Wolfe JT, Ross D, Cohen GM: A role for metals and free radicals in the induction of apoptosis in thymocytes. FEBS Letters 352:58-62, 1994

Zhang H, Weir BK, Macdonald RL, et al: Mechanisms of [Ca++]i elevation induced by erythrocyte components in endothelial cells. J Pharmacol Exp Ther 277:1501-1509, 1996