Nitric oxide via macrophage iNOS induces apoptosis following traumatic spinal cord injury

Molecular Brain Research 85 (2000) 114–122 www.elsevier.com / locate / bres

Research report

Nitric oxide via macrophage iNOS induces apoptosis following traumatic spinal cord injury Kotaro Satake a , Yukihiro Matsuyama a , Mitsuhiro Kamiya a , Hiroshi Kawakami a , a b b,c , Hisashi Iwata , Kayo Adachi , Kazutoshi Kiuchi * b

a Department of Orthopaedic Surgery, Nagoya University School of Medicine, Nagoya 466 -8550, Japan Laboratory for Genes of Motor Systems, Bio-Mimetic Control Research Center, RIKEN, Moriyama, Nagoya 463 -0003, Japan c Department of Biomolecular Science, Faculty of Engineering, Gifu University, Gifu 501 -1193, Japan

Accepted 3 October 2000

Abstract To investigate the pathophysiological mechanisms involved in post-traumatic impairment of the spinal cord, we analyzed expression patterns of the inducible nitric oxide synthase (iNOS) gene following acute injury of rat spinal cord using a weight drop technique. PCR analysis revealed that iNOS mRNA appeared at 3–12 h after injury and declined thereafter. Immunohistochemical analysis showed that iNOS-positive cells invaded the lesioned area through the perivascular space at 6 h after injury. The population of these cells peaked at 24 h and then declined to disappear 3 days after injury. The iNOS-positive cells were also stained with ED-2 but not with ED-1 or OX-42, indicating that these cells were macrophages and / or perivascular cells. In parallel with the appearance of iNOS-positive cells, other cells emerged that were positively stained by the terminal deoxynucleotidyl–transferase-mediated dUDP-biotin nick end-labeling (TUNEL) assay. TUNEL-positive cells were scattered in the lesioned area 1 day after injury, but some in the surrounding area close to iNOS-positive cells. Administration of L-Ng-nitro-arginine methylester, a competitive inhibitor of NOS, resulted in a reduction of TUNEL-positive cells in the lesioned area. These results suggest that nitric oxide generated by iNOS of macrophages and / or perivascular cells plays a significant role in eliminating damaged cells from the lesioned area by apoptosis.  2000 Elsevier Science B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Trauma Keywords: Spinal cord; iNOS; Apoptosis; Macrophage

1. Introduction Acute spinal cord injury (SCI) has been postulated to have two steps in its pathological process: the primary mechanical injury and secondary damage induced by various biochemical reactions [2,3,13,46]. After trauma, petechial hemorrhages occur in the gray matter and edema Abbreviations: iNOS, inducible nitric oxide synthase; TUNEL, terminal deoxynucleotidyl–transferase-mediated dUDP-biotin nick end-labeling; SCI, spinal cord injury; NO, nitric oxide; NMDA, N-methyl-Daspartate; nNOS, neuronal nitric oxide synthase; eNOS, endothelial nitric oxide synthase *Corresponding author. Tel.: 181-52-736-5863; fax: 181-52-7365865. E-mail address: [email protected] (K. Kiuchi).

in the white matter [2,3]. Electron microscopy reveals small hemorrhages in the perivascular space with extravasation of erythrocytes [12]. Myelin sheath disruption, axonal degeneration, and ischemic endothelial damage occur at 4 h after spinal cord contusion [12]. Tator and Fehlings [45] reported a marked reduction of microcirculation in the hemorrhagic regions and adjacent areas, concluding that post-traumatic ischemia is the main effect after injury. With respect to secondary pathological processes, various factors have been thought to cause spinal cord impairment at a molecular level. Matsuyama et al. [28] reported that nitric oxide (NO) may be closely involved in the development of post-traumatic spinal cord cavitation and may therefore be a candidate in the development of the pathological process in vivo. NO is known to

0169-328X / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0169-328X( 00 )00253-9

K. Satake et al. / Molecular Brain Research 85 (2000) 114 – 122

produce either neurotoxic or neuroprotective effects depending on its ambient redox milieu [8,24]. NO cytotoxicity emerges, in part, by reaction with superoxide anion 2 (O 2 2 ) to generate peroxynitrite (ONOO ) [7], while the 1 nitrosonium ion (NO ), an alternative redox-activated state of NO, reacts with the thiol group of the N-methyl-Daspartate (NMDA) receptor to block cytotoxic neurotransmission [24]. There are three isoforms of nitric oxide synthase (NOS). There are two calmodulin-dependent isoforms, neuronal NOS (nNOS) and endothelial NOS (eNOS), which produce NO following Ca 21 influx [1,20]. Under normal conditions, NO plays a physiological role in neuronal signal transmission [40,41] and vessel dilation [30]. Following trauma, excitatory amino acids may overstimulate the NMDA receptor, leading to a high level of NO production via Ca 21 influx, which may in turn cause tissue damage. Certain immunohistochemical studies on the distribution of NADPH-diaphorase in motor neurons [50– 52] and at the tip of damaged axons [17] suggest that induction of nNOS in neurons is involved in secondary damage following SCI. There is also an inducible isoform of NOS (iNOS), which is regulated at a transcriptional level [26,27] and induced under pathological conditions such as infection, stab wounds, and inflammatory disorders [21,29,47]. In the case of SCI, Hamada et al. [18] reported that in the compression model NO produced by iNOS is neurotoxic, while NO produced by the constitutive forms of NOS, such as eNOS or nNOS is neuroprotective. In this study, we investigated the temporal and spatial patterns of iNOS gene expression in vivo after traumatic SCI using a weight drop technique, and role of NO in relation to cell death and cavitation following the acute injury.

115

2.2. RNA extraction and RT-PCR

2. Materials and methods

Rats (n522) were re-anesthetized with pentobarbital and the spinal cords were removed from a distance of one vertebra above to one vertebra below the laminectomized spinal cord at 3 h (n53), 6 h (n53), 12 h (n54), 1 day (n54), 2 days (n53), and 3 days (n52) after the operation, and that of sham control (n53). Spinal tissue was homogenized with 2 ml of TRIZOL (Gibco, BRL) and treated with DNase I (Takara) to prepare a total RNA sample. The RNA was reverse transcribed with oligo d(T)12–18 using M-MuLV RT (Gibco, BRL) and this reaction mixture served as a template for polymerase chain reaction (PCR). To identify iNOS transcription, a reaction mixture (50 ml) for PCR was made up of 2.0 ml of cDNA synthesis mixture, 40 nM dNTPs, 10 pmole of sense and antisense primer, and 1.25 U of Taq polymerase (Takara). PCRs were performed with denaturation at 958C for 45 s, annealing at 628C for 45 s, and extension at 728C for 60 s in each cycle, followed by a final 5-min extension at 728C using a Takara Thermal Cycler MP, TP 3000. The PCR product was cloned using a TA-cloning kit (Invitrogen) and sequenced. For semi-quantitative PCR, the reaction mixture (20 ml) consisted of 1.0 ml of cDNA synthesis mixture, 10 pmole of sense and antisense primer, 40 pM of MgCl 2 , and 2 ml of SYBR Green I (Boehringer Mannheim). PCRs (twice for each sample) were performed using a Light Cycler MP (Boehringer Mannheim), with denaturation at 958C for 120 s, annealing at 628C for 15 s, and extension at 728C for 20 s in each of 45 cycles, followed by a final melting at 958C. The melting curve showed a single peak at 958C. Primer sets were as follows: iNOS, 59-CCC TTC CGA AGT TTC TGG CAG CAG C-39 and 59-GGG TGT CAG AGT CTT GTG CCT TTG G-39; b-actin, 59-TGT ATG CCT CTG GTC GTA CC-39 and CAA CGT CAC ACT TCA TGA TGG-39. A level of about 0.5 units in control (Fig. 1) corresponds to a background level of fluorescence detected.

2.1. Operating procedure for traumatic SCI

2.3. Tissue preparation

Sprague–Dawley rats (8 weeks, male) were anesthetized with pentobarbital (50 mg / kg i.p.) and laminectomy was performed at the T8–T10 level. The spinal cord was insulted by use of a weight drop technique according to Allen’s method [2] with slight modification. A 2 mmdiameter plastic impounder was placed gently on the exposed dura and a 10-g iron weight was dropped from a height of 10 cm above the impounder. The weight and impounder were immediately removed after impact and the paravertebral muscle and skin were closed. Controls underwent the same operative procedure but without spinal cord insult. Rats used in this study were treated strictly according to the NIH Guide for Care and Use of Laboratory Animals and approved by the RIKEN and Nagoya University School of Medicine Animal Committees.

Segments of spinal cords (n534) plus controls (n52) were obtained for immunohistochemistry analysis. Rats were sacrificed at 6 h (n55), 12 h (n55), 1 day (n55), 2 days (n55), 3 days (n55), 5 days (n53), 7 days (n53) and 14 days (n53) after the operation. Each rat was anesthetized with pentobarbital and the spinal cord was fixed by intracardiac perfusion with 200 ml of 4% (w / v) paraformaldehyde in 0.1 M phosphate-buffer (pH 7.4). The removed spinal cords were post-fixed with the same fixative overnight, and soaked in 10, 20, and 30% sucrose for 1 day, respectively, and served for frozen sections using OCT Compound. Consecutive transverse 5-mm thick sections were prepared with a cryostat and stored at 2808C. Serial sections (n53 or 4) from each spinal cord were used for cell counting.

116

K. Satake et al. / Molecular Brain Research 85 (2000) 114 – 122

After washing with PBS, they were stained with a combination of FITC-labeled anti-rabbit IgG, Texas-red-labeled anti-mouse IgG, or Cy3-labeled GFAP. The terminal deoxynucleotidyl-transferase-mediated dUDP-biotin nick end-labeling (TUNEL) assay was performed according to Gavrieli et al. [15] with minor modifications [6]. The iNOS- or TUNEL-positive cells were counted in each section.

2.6. L-Ng-nitro-arginine methylester ( L-NAME) administration Rats received i.p. 30 mg / kg of L-NAME (Wako), DNAME (Sigma RBI) or 0.9% saline in the same volume. The administration of either L-NAME or D-NAME was performed immediately after the operation, and repeated at 3 h, 6 h, 12 h, 1 day, and 2 days. On day 2, the spinal cord was removed and segments were prepared in the same way as above. TUNEL-positive cells were counted. Results are expressed mean6S.E.M. Statistical comparison was performed by Mann–Whitney U-test. P,0.05 was considered significant.

3. Results Fig. 1. iNOS mRNA expression (mean6S.E.M.) in sham controls and at various times after traumatic SCI.

2.4. Antibodies Rabbit anti-iNOS polyclonal antibody was prepared using GST-iNOS (N-terminal) fusion protein as reported previously [14]. Specificity of the anti-iNOS antibody was confirmed by Western blot analysis, in which a single 130-kDa band was detected [5,35]. Mouse monoclonal antibody against rat leukocyte common antigen (ED-1) [49], rat macrophage antigen (ED-2) [11], and rat complement receptor type 3 (OX-42) [39] were obtained from BMA Biomedicals Ltd. (Switzerland). Cy3-conjugated mouse monoclonal antibody against rat glial fibrillary acid protein (GFAP) was from Sigma (St. Louis, MO), FITCconjugated goat anti-rabbit IgG antibody, and Texas redconjugated goat anti-mouse IgG antibody from TAGO (Burlingame, CA), and Texas red-conjugated streptoavidin from Du Pont NEN (Boston, MA).

2.5. Immunohistochemistry The sections were pre-incubated in 10% normal goat serum for 10 min to block non-specific staining. Antibodies were diluted in 10% normal goat serum containing 0.2% Triton X-100. For double staining, the sections were incubated overnight at 48C with the appropriate combination antibodies against iNOS, ED-1, ED-2, and OX-42.

Semi-quantitative RT-PCR analysis was performed to investigate the temporal pattern of iNOS induction following weight drop impact on the rat spinal cord. Fig. 1 shows the time course of iNOS gene expression after traumatic insult and in sham controls. iNOS mRNA was increased at 3–12 h, began to decline on day 1, and returned to that of sham controls 2–3 days after injury. Sequence analysis confirmed that the 497-bp band corresponded to the rat iNOS cDNA fragment. iNOS mRNA was not detected in controls by the sham operation as control. Hematoxylin-eosin staining of spinal cord sections showed that damage from trauma spread from the dorsal white matter to the central gray matter and central canal 1 and 2 days after injury. The damaged area became cavernous and gliosis was observed after 2 weeks (data not shown). We next investigated the spatial pattern of iNOS expression in the contused spinal cord (Fig. 2). Six h after injury, iNOS-positive cells appeared in intra- and perivascular zones (Fig. 2A), while some were scattered in the damaged area (data not shown). Toward 12 h to 2 days after injury, iNOS-positive cells increased in number and were gathered in the damaged area. They were small nuclear cells that were not stained with GFAP (Fig. 2B). GFAP-positive cells were observed at a distance from the damaged area, indicating that iNOS was not expressed by astroglial cells. In order to further characterize the iNOS-positive cells, we performed double-staining with a combination of antiiNOS antibody and ED-1, ED-2, or OX-42. The iNOSpositive cells were almost positive to ED-2 (Fig. 2C), but

K. Satake et al. / Molecular Brain Research 85 (2000) 114 – 122

117

Fig. 2. Distribution of iNOS-positive cells in cryosections after traumatic SCI investigated immunohistochemically. iNOS-positive cells were visualized with FITC-conjugated goat antibody against rabbit IgG as a second antibody. (A) iNOS-positive cells (green) were observed around the perivascular zone of cryosection 6 h after injury. (B) GFAP was stained with Cy3-conjugated mouse monoclonal antibody against rat GFAP. iNOS-positive cells were negative for GFAP expressed by astrocytes 1 day after injury. (C) Double-staining using anti-iNOS antibody and ED-2. For ED-2 staining, Texas Red-conjugated goat anti-mouse IgG antibody was used. iNOS-positive cells indicated by white allows were also positive with ED-2 (white allows) 1 day after the injury. Bars550 mm.

118

K. Satake et al. / Molecular Brain Research 85 (2000) 114 – 122

negative to ED-1 and OX-42 (data not shown). These results suggest that the iNOS-positive cells are not microglia, but rather macrophages and / or perivascular cells. The iNOS-positive cells decreased in number on day 3 and almost disappeared 5 days after injury. Fig. 3 shows the number of iNOS-positive cell counted per spinal cord section over time following contusion. To examine the relationship between iNOS induction and cell death, each specimen was double-stained using anti-iNOS antiserum and the TUNEL assay. TUNEL-positive cells were also scattered in the damaged area and some arose close by iNOS-positive cells 1 day after the injury (Fig. 4A). The number of TUNEL-positive cells per section increased until 1 day after contusion, declined on day 2, and virtually disappeared thereafter (Fig. 4B). To confirm some of TUNEL-positive cells to be apoptotic cells, the cells were visualized by electron microscopy in the same specimens of injured spinal cord. Fig. 4C shows the morphological change of a dying cell 2 days after injury as a cell with a shrinking nucleus. These results indicate that iNOS expressed in macrophages and / or perivascular cells seems to play an important role in inducing apoptosis after trauma. Finally, we assessed the effect of a competitive NOS inhibitor, L-NAME, to obtain proof that NO produced by macrophages and / or perivascular cells is involved in the elimination of cells in the contused spinal cord. iNOS

Fig. 3. Number (mean6S.E.M.) of iNOS-positive cells per spinal cord section at various times after traumatic SCI.

expression was not altered by the systemic administration of L-NAME (data not shown). However, the number of TUNEL-positive cells per section from rats treated with L-NAME was decreased to 60% compared to that of D-NAME, an inactive isomer of NAME, as well as controls (Fig. 5). This suggests that NO might trigger the elimination of damaged cells via its cytotoxic effect.

4. Discussion In this study, we investigated temporal and spatial expression patterns of the iNOS gene after contusion injury of the spinal cord in the rat. From PCR analysis, iNOS mRNA was induced from 3 to 12 h, started to decrease at 24 h, and returned to that of controls 2–3 days after injury. The peak expression of iNOS in our model coincides with that in the compression model reported by Hamada et al. [18], although the pathology of acute impact and slow compression injury differs in many respects [22]. Immunohistochemical analysis revealed that iNOS-positive cells induced by traumatic SCI were macrophages and / or perivascular cells, which were also stained with ED-2. In parallel with iNOS mRNA expression, iNOS-positive cells appeared around the perivascular space at 6 h after injury. These cells then scattered and increased in the damaged area toward days 1–2, but soon decrease 3 days after the injury. In our contusion model, the blood–brain barrier was disrupted and various blood-borne cells were able to invade the parenchyma of spinal cord at an early stage following injury. Perivascular cells are able to act as antigen-presenting cells and exhibit morphological features consistent with macrophages in inflammatory disorders [4,31,32]. Macrophages and / or perivascular cells seem to play an important role in producing NO via iNOS in the contusion model. iNOS induction has been reported in a model of ischemia following static compression of spinal cord [18], but the type of cell expressing the iNOS gene was not defined in that study. Takeuchi et al. [44] reported that the iNOS gene was expressed by microglial cells after stereotactic injection of ethanol into the striatum. In their model, the coagulating effect of ethanol seems to make this particular form of necrotic damage free from hemorrhage and passage of monocytes and lymphocytes from disruption of the blood–brain barrier. In our study, we are the first to demonstrate that iNOS is dominantly expressed by macrophages and / or perivascular cells after traumatic insults of the spinal cord in vivo. We found that TUNEL-positive cells arose within and around the area of injury 12 h after traumatic SCI, and some of them were observed close to macrophages and / or perivascular cells which were iNOS-positive. TUNELpositive cells at a distance from iNOS-positive cells may be necrotically degenerated by the primary insult, since the administration of L-NAME, a competitive NOS inhibitor, partially diminished the number of TUNEL-positive cells.

K. Satake et al. / Molecular Brain Research 85 (2000) 114 – 122

119

Fig. 4. Dying cells stained by the TUNEL method and iNOS-positive cells 1 day after traumatic SCI. (A) Double-staining using the TUNEL method and anti-iNOS antibody. The two colour images are superimposed. TUNEL-positive cells were visualized using Texas red-labeled streptoavidin, while iNOS-positive cells were visualized using FITC-conjugated goat antibody against rabbit IgG. There were some TUNEL-positive cells nearby iNOS-positive cells. Bars550 mm. (B) Number (mean6S.E.M.) of TUNEL-positive cells at various times after traumatic SCI. (C) Electron micrograph of apoptotic cells in the lesioned area of the spinal cord 2 days after injury. Bars51 mm.

Apoptosis has already been observed following SCI in numerous studies [10,19,23,25,42,53]. However, none of these studies have shown what triggered apoptosis after injury. In our in vivo model, we propose the hypothesis that NO from macrophages and / or perivascular cells could induce some damaged cells to undergo apoptosis via a cytotoxic effect. This phenomenon accords with the previous result that spinal cord cavitation was suppressed to some extent by N G -mono-methyl-L-arginine treatment using the contusion model [28]. TUNEL-positive cells were presumably eliminated by phagocytosis of macrophages, as cavitation was progressed. It is still unclear whether this process of iNOS-positive cell elimination is necessary for healing following SCI. This early apoptotic process of eliminating damaged cells

might protect intact cells from necrotic irritation. Wong and Billiar [48] have reported that iNOS expression appears to play a protective role during acute inflammation, while it may be detrimental in chronic localized inflammation. Siegel et al. [43] have demonstrated that areas of the spinal cord undergoing an inflammatory response with macrophage aggregation showed the most regrowth by dorsal root fibers. Phagocytic cells have also been postulated to have a prominent role in the regenerative response to CNS injury [34,37,38]. On the other hand, Hamada et al. [18] reported that a NOS inhibitor was effective for recovery of motor function following compression injury of the spinal cord and suggested that NO induced by iNOS could be cytotoxic in a subacute phase of SCI.

120

K. Satake et al. / Molecular Brain Research 85 (2000) 114 – 122

Fig. 5. Number (mean6S.E.M.) of TUNEL-positive cells per spinal cord section 2 days after traumatic SCI in rats treated with i.p. 30 mg / kg of L-NAME, D-NAME or 0.9% saline (control). *, P,0.001; **, not significant. Fig. 4. (continued)

It is not yet known which factors induce and regulate iNOS gene expression in the early stages of SCI in vivo. Previous in vitro studies have shown that cytokines such as interferon-g and tumor necrosis factor-a induce iNOS mRNA in glial cells [9,16,33]. These cytokines might be activated in an early stage of SCI. Disruption of the blood–brain barrier at or adjacent to the lesion center would allow passage of monocytes and lymphocytes, and circulation of cytokines into the insulted neural tissue [36]. These extracellular circumstances would seem to provide a suitable environment for stimulation of iNOS expression. In summary, we are the first to show that iNOS is expressed in macrophages and / or perivascular cells following traumatic insult to the spinal cord in vivo. One of the functions of NO via iNOS from these cells is most likely to induce apoptosis and elimination of adjacent damaged cells during the early stage of traumatic SCI. Further study is necessary to elucidate the pathophysiological mechanisms of impairment after contusion of the spinal cord.

Acknowledgements This study was supported by the Frontier Research Program of RIKEN and, in part, by a Grant-in-Aid for

Scientific Research from the Ministry of Education, Science, Sport, and Culture of Japan.

References [1] S. Alaqarsamy, G. Lonart, K.M. Johnson, Regulation of nitric oxide synthase activity in cortical slices by excitatory amino acids and calcium, J. Neurosci. Res. 38 (1994) 648–653. [2] A. Allen, Surgery of experimental lesions of spinal cord equivalent to crush injury of fracture dislocation of spinal column. A preliminary report, J. Am. Med. Assoc. 57 (1911) 878–880. [3] A. Allen, Remarks on the histopathological changes in the spinal cord due to impact. An experimental study, J. Nerv. Ment. Dis. 31 (1914) 141–147. [4] D. Angelov, M. Walther, M. Streppel, O. Guntinas-Lichius, W.F. Neiss, The cerebral perivascular cells. Advances in Anatomy, Embryol. Cell Biol. 147 (1998) 1–87. [5] K. Aono, K.I. Isobe, K. Kiuchi, Z.H. Fan, M. Ito, A. Takeuchi, M. Miyachi, I. Nakashima, Y. Nimura, In vitro and in vivo expression of inducible nitric oxide synthase during experimental endotoxemia: involvement of other cytokines, J. Cell. Biochem. 65 (1997) 349– 358. [6] G. Bagetta, M.T. Corasaniti, L. Berliocchi, M. Navarra, A. FinazziAgro, G. Nistico, HIV-1 gp120 produces DNA fragmentation in the cerebral cortex of rat, Biochem. Biophys. Res. Commun. 211 (1995) 130–136. [7] J. Beckman, T.W. Beckman, J. Chen, P.A. Marshall, Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide, Proc. Natl. Acad. Sci. USA 87 (1990) 1620–1624.

K. Satake et al. / Molecular Brain Research 85 (2000) 114 – 122 [8] A. Butler, F.W. Flitney, D.L.H. Williams, NO, nitrosonium ions, nitroxide ions, nitrosothiols and iron-nitrosyls in biology: a chemist’s perspective, Trends Pharmacol. Sci. 16 (1995) 18–22. [9] S. Corradin, J. Mauel, S.D. Donini, E. Quarttrocchi, P. RicciardiCastagnoli, Inducible nitric oxide synthase activity of cloned murine microglial cells, Glia 7 (1993) 255–262. [10] M. Crowe, J.C. Bresnahan, S.L. Shuman, J.N. Masters, M.S. Beattie, Apoptosis and delayed degeneration after spinal cord injury in rats and monkeys, Nat. Med. 3 (1997) 73–76. [11] C. Dijkstra, E.A. Dopper, P. Joling, G. Kraal, The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies ED1, ED2 and ED3, Immunology 54 (1985) 589–599. [12] G. Dohrman, F.C. Wagner Jr., P.C. Bucy, The microvasculature in transitory traumatic paraplegia. An electron microscopic study in the monkey, J. Neurosurg. 35 (1971) 263–271. [13] A. Faden, Experimental neurobiology of central nervous system trauma, Crit. Rev. Neurobiol. 7 (1993) 175–186. [14] Z. Fan, K. Isobe, K. Kiuchi, I. Nakashima, Enhancement of nitric oxide production from activated macrophages by a purified form of ginsenoside (Rg1), Am. J. Chin. Med. 23 (1995) 279–287. [15] Y. Gavrieli, Y. Sherman, S.A. Ben-Sasson, Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation, J. Cell. Biol. 119 (1992) 493–501. [16] J. Goodwin, E. Uemura, J.E. Cunnick, Microglial release of nitric oxide by the synergistic action of beta-amyloid and IFN-gamma, Brain Res. 692 (1995) 207–214. [17] G. Guizar-Sahagun, P. Garcia-Lopez, A.L. Espitia, I. Grijalva, R.E. Franco-Bourland, I. Madrazo, Transitory expression of NADPH diaphorase (NOS) in axonal swellings after spinal cord injury, Neuroreport 9 (1998) 2899–2902. [18] Y. Hamada, T. Ikata, S. Katoh, K. Tsuchiya, M. Niwa, Y. Tsutsumishita, K. Fukuzawa, Roles of nitric oxide in compression injury or rat spinal cord, Free Radic. Biol. Med. 20 (1996) 1–9. [19] K. Katoh, T. Ikata, S. Katoh, Y. Hamada, K. Nakauchi, T. Sano, M. Niwa, Induction and its spread of apoptosis in rat spinal cord after mechanical trauma, Neurosci. Lett. 216 (1996) 9–12. [20] L. Kiedrowski, E. Costa, J.T. Wroblewski, Glutamate receptor agonists stimulate nitric oxide synthase in primary cultures of cerebellar granule cells, J. Neurochem. 58 (1992) 335–341. [21] H. Koprowski, Y.M. Zheng, E. Heber-Katz, N. Fraser, L. Rorke, Z.F. Fu, Hanlon, C,, In vivo expression of inducible nitric oxide synthase in experimentally induced neurologic diseases, Proc. Natl. Acad. Sci. USA 90 (1993) 3024. [22] K. Kraus, The pathophysiology of spinal cord injury and its clinical implications, Semin. Veterinary Medicine and Surgery (Small Animal) 11 (1996) 201–207. [23] G. Li, G. Brodin, M. Farooque, K. Funa, A. Holtz, W.L. Wang, Y. Olsson, Apoptosis and expression of Bcl-2 after compression trauma to rat spinal cord, Apoptosis and expression of Bcl-2 after compression trauma to rat spinal cord, Semin. Vet. Med. Surg. (Small Anim.) 55 (1996) 280–289. [24] S. Lipton, Y.B. Choi, Z.H. Pan, S.Z. Lei, H.S. Chen, N.J. Sucher, J. Loscalzo, D.J. Singel, J.S. Stamler, A redox-based mechanism for the neuroprotective and neurodestructive effects of nitric oxide and related nitroso-compounds, Nature 364 (1993) 626–632. [25] X. Liu, X.M. Xu, R. Hu, C. Du, S.X. Zhang, J.W. McDonald, H.X. Dong, Y.J. Wu, G.S. Fan, M.F. Jacquin, C.Y. Hsu, D.W. Choi, Neuronal and glial apoptosis after yraumatic spinal cord injury, J. Neurosci. 17 (1997) 5395–5406. [26] C. Lowenstein, E.W. Alley, P. Raval, A.M. Snowman, S.H. Snyder, S.W. Russell, W.J. Murphy, Macrophage nitric oxide synthase gene: two upstream regions mediate induction by interferon gamma and lipopolysaccharide, Proc. Natl. Acad. Sci. USA 90 (1993) 9730– 9734. [27] C. Lyons, G.J. Orloff, J.M. Cunningham, Molecular cloning and functional expression of an inducible nitric oxide synthase from a murine macrophage cell line, J. Biol. Chem. 267 (1992) 6370–6374.

121

[28] Y. Matsuyama, K. Sato, M. Kamiya, J. Yano, H. Iwata, K.I. Isobe, Nitric oxide: a possible etiologic factor in spinal cord cavitation, J. Spinal Disord. 11 (1998) 248–252. [29] D. Minc-Golomb, I. Tsarfaty, J.P. Schwartz, Expression of inducible nitric oxide synthase by neurons following exposure to endotoxin and cytokine, Br. J. Pharmacol. 112 (1994) 720–722. [30] S. Moncada, R.M. Palmer, E.A. Higgs, The discovery of nitric oxide as the endogenous nitrovasodilator, Hypertension 12 (1988) 365– 372. [31] S. Mori, C.P. Leblond, Identification of microglia in light and electron microscopy, J. Comp. Neurol. 135 (1969) 57–80. [32] Y. Murabe, Y. Sano, Morphological studies on neuroglia. I. Electron microscopic identification of silver-impregnated glial cells, Cell Tissue Res. 216 (1981) 216. [33] S. Murphy, M.L. Simmons, L. Agullo, A. Garcia, D.L. Feinstein, E. Galea, D.J. Reis, D. Minc-Golomb, J.P. Schwarz, Synthesis of nitric oxide in CNS glial cells, Trends Neurosci. 16 (1993) 323–328. [34] K. Nagata, N. Takei, K. Nakajima, H. Saito, S. Kohsaka, Microglial conditioned medium promotes survival and development of cultured mesencephalic neurons from embryonic rat brain, J. Neurosci. Res. 34 (1993) 357–363. [35] Y. Nozaki, Y. Hasegawa, A. Takeuchi, Z.H. Fan, K.I. Isobe, I. Nakashima, K. Shimokata, Nitric oxide as an inflammatory mediator of radiation pneumonitis in rats, Am. J. Physiol. 272 (1997) L651– L658. [36] P. Popovich, W.J. Streit, B.T. Stokes, Differential expression of MHC class II antigen in the contused rat spinal cord, J. Neurotrauma 10 (1993) 37–46. [37] C. Prewitt, I.R. Niesman, C.J.M. Kane, J.D. Houle, Activated macropharge / microglial cells can promote the regeneration of sensory axons into the injured spinal cord, Exp. Neurol. 148 (1997) 433–443. [38] A. Rabchevsky, W.J. Streit, Grafting of cultured microglial cells into the lesioned spinal cord of adult rats enhances neurite outgrowth, J. Neurosci. Res. 47 (1997) 34–48. [39] A. Robinson, T.M. White, D.W. Mason, MRC OX-43: a monoclonal antibody which reacts with all vascular endothelium in the rat except that of brain capillaries, Immunology 57 (1986) 231–237. [40] E. Schuman, D.Y. Madison, A requirement for the intercellular messenger nitric oxide in long-term potentiation, Science 254 (1991) 1503–1506. [41] K. Shibuki, D. Okada, Endogenous nitric oxide release required for long-term synaptic depression in the cerebellum, Nature 349 (1991) 326–328. [42] S. Shuman, J.C. Bresnahan, M.S. Beattie, Apoptosis of microglia and oligodendrocytes after spinal cord contusion in rats, J. Neurosci. Res. 50 (1997) 798–808. [43] J. Siegal, G. Kliot, G.M. Smith, J. Silver, A comparison of the regeneration potential of dorsal root fibers into gray or white matter of the adult rat spinal cord, Exp. Neurol. 109 (1990) 90–97. [44] A. Takeuchi, K.I. Isobe, O. Miyaishi, M. Sawada, Z.H. Fan, I. Nakashima, K. Kiuchi, Microglial NO induces delayed neuronal death following acute injury in the striatum, Eur. J. Neurosci. 10 (1998) 1613–1620. [45] C. Tator, M.G. Fehlings, Review of the secondary injury theory of acute spinal cord trauma with emphasis on vascular mechanisms, J. Neurosurg. 75 (1991) 141–147. [46] C.H. Tator, Update on the pathophysiology and pathology of acute spinal cord injury, Brain Pathol. 5 (1995) 407–413. [47] M. Wallace, S.K. Bisland, NADPH-diaphorase activity in activated astrocytes represents inducible nitric oxide synthase, Neuroscience 59 (1994) 905–919. [48] J. Wong, T.R. Billiar, Regulation and function of inducible nitric oxide synthase during sepsis and acute inflammation, Adv. Pharmacol. 34 (1995) 155–170. [49] G. Woollett, A.N. Barclay, M. Puklavec, A.F. Williams, Molecular and antigenic heterogeneity of the rat leukocyte-common antigen

122

K. Satake et al. / Molecular Brain Research 85 (2000) 114 – 122

from thymocytes and T and B lymphocytes, Eur. J. Immunol. 15 (1985) 168–173. [50] W. Wu, Expression of nitric-oxide synthase (NOS) in injured CNS neurons as shown by NADPH diaphorase histochemistry, Exp. Neurol. 120 (1993) 153–159. [51] W. Wu, F.J. Liuzzi, F.P. Schinco, A.S. Depto, Y. Li, J.A. Mong, T.M. Dawson, S.H. Snyder, Neuronal nitric oxide synthase is induced in spinal neurons by traumatic injury, Neuroscience 61 (1994) 719– 726.

[52] L.-W. Yick, W. Wu, K.-F. So, S.-Y. Wong, Time course of NOS expression and neuronal death in Clarke’s nucleus following traumatic injury in adult rat spinal cord, Neurosci. Lett. 241 (1998) 155–158. [53] C. Yong, P.M. Arnold, M.N. Zoubine, B.A. Citron, I. Watanabe, N.E.J. Berman, B.W. Festoff, Apoptosis in cellular compartments of rat spinal cord after severe contusion injury, J. Neurotrauma 15 (1998) 459–472.