- Email: [email protected]
Short-term changes in prostacyclin secretory profile of irradiated rat cervical spinal cord
ARTICLE IN PRESS
Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378 www.elsevier.com/locate/plefa
Short-term changes in prostacyclin secretory profile of irradiated rat cervical spinal cord Alireza Shirazia, Seied Rabie Mahdavia, Bagher Minaeeb, Alireza Nikoofarc, Ebrahim Azizid, a Department of Medical Physics, Tehran University of Medical Sciences (TUMS), Tehran, Iran Department of Anatomy and Histology, Faculty of Medicine, Tehran University of Medical Sciences (TUMS), Tehran, Iran c Faculty of Medicine, Iran University of Medical Sciences, Tehran University of Medical Sciences (TUMS), Tehran, Iran d Molecular Research Laboratory, Department of Pharmacology and Toxicology, Faculty of Pharmacy, Tehran University of Medical Sciences (TUMS), Tehran, Iran
b
Received 17 August 2004; accepted 2 February 2005
Abstract Prostaglandins changes in radiation myelopathy (RM) have been previously reported. In the present study, we decided to determine the profile of Prostacyclin (PGI2) content in irradiated rat cervical cord. Wistar rats were irradiated with doses of 2,4,6,15,25 and 30 Gy of X-rays. After 24 h, 2 and 13 weeks post-irradiation, samples of spinal cord were prepared for evaluation of PGI2 and histopathologic changes. Prostacyclin content was determined by quantification of 6-keto-prostaglandin-F1a (prostacyclin major metabolite). Irradiated segments of spinal cord were stained routinely for histological studies. Results of irradiated were compared to control groups. Average ratio values of 6-keto-PG-F1a for doses of 2–30 Gy were between 67.5% and 107%, 65.41% and 100.54%, and 62.20% and 98.89% for 24 h, 2 and 13 weeks post-irradiation, respectively. Histopathological studies showed marked gliosis and vascularities in irradiated specimens. PGI2 bimodal secretory profile was observed along with histopathological changes in this study. Our results can further emphasize on the role of PGI2 in RM. r 2005 Elsevier Ltd. All rights reserved.
1. Introduction Myelopathy is a serious and not rare complication of cancer and cancer treatment. Myelopathies may be induced through external pressure on cord, intramedullary metastasis or treatment toxicities associated with radiation therapy alone or in combination with other treatment methods. Radiation myelopathy (RM) in patients following radiotherapy made it the subject of vast clinical and radiobiological studies because it may not be diagnosed from the associated symptoms alone. Therefore, it seems there is a long way to understand the basic concept of mechanisms in the development of RM [1–4].
Corresponding author. Tel./fax: +98 21 6959100.
E-mail address: [email protected] (E. Azizi). 0952-3278/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.plefa.2005.02.003
Dose–response curves for treatment of malignancies profoundly show a strict margin between virtual 100% tumoricidal dose and neurotoxicity occurrence in spinal cord as a normal tissue complication [2]. It was shown that single dose of 35 Gy X-ray to spinal cord can cause limb paralysis in rat models with a latency of 19.070.3 weeks [5]. In conventional radiotherapy incidence of RM after doses of 57–61 Gy is 5% with a good approximation. Analysis of patients developing RM demonstrated that the upper limit of tolerance for a X5 cm length of the cervical spine was 50 Gy (25 fractionates/35 days). The mostly used dose limit for spinal cord is 45 Gy in 22–25 fractions and this is considered as tolerance dose. Yet, due to morbidity of RM, spinal cord dose should always be kept under tolerance limit [6–10]. Radiobiologically, it was reported that linear quadratic (LQ) model does not provide a satisfactory
ARTICLE IN PRESS 374
A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
description of the dose–response relationship obtained from the studies on rat spinal cord. There were even observations on different dose responses with paradoxical increase in radiosensitivity [11,12]. Until recently, findings of radiation injury to spinal cord were based entirely on morphological changes in white matter (Wm) structures, which are consisted of various degrees of demyelination, gliosis, Wm necrosis, vascular changes and inflammatory responses [13]. Recent advances reported the probable priority of vascular and its secretory profile changes in pathogenesis of RM (vascular hypothesis). Vascular hypothesis states that vascular changes occur prior to any alterations in the Wm parenchyma due to different radiosensitivity between vascular endothelium and glial cells [4]. Modern enzymatic and/or immunohistochemical techniques have been applied to animal models for studying the radiation-induced changes in the various biochemical mediators produced by different cells within the irradiated medium [13–15]. Release of prostaglandins (PGs) D2, E2, F2a and I2 in the spinal cord has been documented by others. Prostaglandin measurement has application in diagnosis of different spinal cord disorders in a range from cerebrospinal fluid to intra spinal antibody microprobes or spinal cord homogenates. A basal level of PGs may exist in many laboratory samples and it is believed that injury to tissue induces release of prostaglandins [16]. This study aims to determine the changes in secretory profile of Prostaglandin I2 (PGI2) or Prostacyclin, in short-term after irradiation (24 h, 2 and 13 weeks postirradiation) to explore its possible contribution to the development of RM.
2. Materials and methods 2.1. Animals and radiation Male Wistar rats weighing 150–200 g were fixed on jig under gentle anesthesia. The jig was specially designed and tailored in order to have a reproducible and accurate irradiation to rats’ cervical spinal cord [17]. C1 to T2 segment of spine was irradiated through two parallel opposed ports with an orthovoltage machine (200 kVp, 1.5 mm Cu HVL, Focal-Skin Distance 25 cm) at a dose rate of 153.26 cGy/min at depth of spinal cord. Doses of 2, 4, 6, 15, 25 and 30 Gy were administered using a circular 3.0 cm diameter localizator [18]. We have also irradiated one group of rats with dose of 35 Gy for evaluation of survival and to produce the clinical manifestation of RM. Five rats were included in each dose group in addition to age-matched control group. The animals were sacrificed chronologically in time intervals of 24 h, 2 and 13 weeks post-irradiation. The single group irradiated with 35 Gy was followed for
symptoms of RM and we could obtain some radiobiological data of latent period of symptoms and lethal dose of producing the paralysis. 2.2. Sample preparations Following decapitation under ether anesthesia, rats’ neck regions were opened posteriorly until to expose cervical spine. Then, the spinal cord was cut rapidly and segmented into three pieces for enzyme immunoassay (EIA) and histopathological studies with transmission electron and light microscopes. 2.3. Measurement of spinal cord PGI2 Samples for EIA were pulled out after laminectomy, frozen on dry ice, stored at 80 1C prior to analysis. Samples were then homogenized with weight proportion of 0.1 M perchloric acid, centrifuged at 4 1C for 10 min at 3000 rpm. Then 100–150 ml supernatant was removed for dilution (1:5 and 1:10) and poured into a 96 well plate of an EIA kit (Cayman Chemical Co., CA-USA). After 18 h incubation, wells were washed out and Ellman’s reagent was added to each well and samples were shaked for 90–120 min and read using plate reader (Sunrise, UK) at wavelength of 415 nm. This assay is based on the competition between 6-keto PGF1a and a 6-keto acetylcholinesterase (AChE) conjugate (6-keto PGF1a tracer) for a limited number of 6-keto specific rabbit antiserum binding sites in 96 well plates. Because the concentration of the 6-keto PGF1a tracer is held constant while the concentration of 6-keto PGF1a varies, the amount of 6-keto PGF1a tracer that is able to bind to the rabbit antiserum will be inversely proportional to the concentration of 6-keto PGF1a (pg/ml) in the wells. In this way, the kit measures the 6-keto prostaglandin F1a (6-keto PGF1a), which is the main metabolic product of prostacyclin (PGI2). Therefore, the concentration of 6keto PGF1a is directly proportional to prostacyclin. To normalize the data of different groups, 6-keto PGF1a concentrations of control group were divided by the values for irradiated rats. Statistical analysis was carried out by means of Sigma Plot package [15]. 2.4. Histopathologic study The samples for electron and light microscopy were immersed into the appropriate fixatives immediately and processed by routine procedures of staining and preparation. Each histopathologic variable was scored by the histopathologist in comparison to the control group after averaging of each dose group values (+ for mild, ++ for moderate, and +++ for severe reactions). Images of spinal cord sections were examined by a histopathologist for changes in white matter (Wm) stroma and for vascular changes in comparison to
ARTICLE IN PRESS A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
control cases. Changes in neuroglia cell proliferation and function were the most important parameters that were studied in this project. Vessels were examined for structural changes consisting the wall rupture and alteration in their density throughout the Wm parenchyma. To obtain the total effect (TE) of irradiation on each variable, after scoring all of them, the scores for each variable was summed up. The statistical data for each dose–response curve was assessed by analysis of variance (ANOVA) followed by Normality and Constant Variance tests using SPSS 10 software.
3. Results 3.1. Prostaglandin I2 changes PGI2 assay 24 h post-irradiation with doses of 2– Gy revealed a rise and fall in concentration level. Initially, it increased with doses of 2 and 4 Gy and then fall at dose of 6 Gy followed by gradual increase from 15 Gy, but did not reach to the control level (Fig. 1). Prostacyclin secretion after 2 weeks post-irradiation revealed a more complex variation. It showed a bimodal action. Initially, it showed a lower concentration than control followed by an increase in concentration of PGI2 at doses up to 6 Gy. Then it decreased steeply at doses below 15 Gy. Again, it made the second peak of concentration at dose of 30 Gy (Fig. 1). Eicosanoid studies 13 weeks post-irradiation at different doses showed an abrupt increase in PGI2 concentration. The lowest level of concentration belonged to the dose of 15 Gy, which followed by an increase in PGI2 concentration at 25 Gy.
375
Prostacyclin level of animals irradiated with 35 Gy at time of inducing the paralysis, as the radiobiological end effect, and just prior to death (17 weeks post-irradiation) revealed steep increase relative to corresponding control rats (approximately 7:1 with Po 0.002). 3.2. Histopathpathological changes Histopathological examinations 24 h post-irradiation of 2, 4 and 6 Gy did not reveal abnormality on light microscopy. However, 200 kVp X-ray doses of 15, 25 and 30 Gy showed some degrees of alterations in Wm even at 24 h post-irradiation. Radiation mostly affected the vascular bed and neuroglia cells. Mild demyelination, mild exocentric myelin sheath, vascular congestion and/or dilatation and mild gliosis often situated between the anterior horns of gray matter were consistently seen after 24 h (Fig. 2). Findings after 2 weeks irradiation showed similar histopathological changes in supportive cells of Wm but more early vascular changes even in lower doses, which could mimic the signs of inflammatory condition with dose-dependent increase in severities. Predominant vascular changes were consisted of RBC and mononuclear infiltrations, vascular dilatation, congestion, heamorrhage and thrombosis (Fig. 3). After 13 weeks post-irradiation, different X-ray doses indicated a decrease in histopathological abnormalities both in Wm neuroglia cells and vascular bed relative to the changes at 2 weeks after irradiation (Fig. 4). None of the irradiated animals with doses of 2, 4, 6, 15, 25 and 30 Gy were developed typical manifestation of RM after short term of study but groups, which received doses X15 Gy showed clear decrease in weight compared to normal control cases. Ultrastructural electron microscope images showed various changes in the endothelial cells morphology. 6 Total Effect (TE)
5 Congestion
4
MON
3 2 1 0
(a) Fig. 1. 6-Keto Prostaglandin F1a level profile of rats’ cervical spinal cord irradiated with doses of 2–30 Gy at 24 h, 2 and 13 weeks postirradiation. The result of 35 Gy single group at 17 weeks after irradiation was also shown. Its content ratio as PGI2 (irradiated)/PGI2 (age matched control). The means7SEM were shown as percentage at confidence limit of 95%.
5 10 15 20 25 30 35 Dose (Gy)
(b)
Fig. 2. Histopathological effects of irradiation after 24 h: (a) Dose–response curve for TEs in white matter parenchyma at different doses (Pp0.06) and (b) photomicrograph of rat cervical spinal cord with dose of 25 Gy X-rays (100 , H&E staining). The mass of neuroglia (MON) and vascular congestion are seen in the photomicrograph.
ARTICLE IN PRESS A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
Total effect
376
14 12 10 8 6 4 2 0
Thrombosis
2
(a)
15 25 30 4 6 log10Dose (Gy) (b)
Fig. 3. Histopathological effects of irradiation after 2 weeks: (a) Dose–response curve for TEs in white matter parenchyma at different doses (pp0.06) and (b) photomicrograph of rat cervical spinal cord with dose of 25 Gy X-rays (100 , H&E staining). Thrombosis and demyelination are seen in the photomicrograph.
Fig. 5. Transmission electron microscope (TEM) images of rat cervical spinal cord white matter post-capillary vessel endothelial cell: (a) endothelial cell wall detachment (two arrow heads), cytoplasmic membrane degeneration (arrow) after 24 h of 30 Gy irradiation, and (b) thickening of cell basal membrane (arrow head) and nucleus membrane folding (arrow) at 2 weeks post-irradiation of 6 Gy (12000 ).
Total Effect (TE)
Glioma 3.6 3.5 3.4 3.3 3.2 (a)
10 20 30 Irradiation dose (Gy)
(b)
Fig. 4. Histopathological effects of irradiation after 13 weeks: (a) Dose–response curve for TEs in white matter parenchyma at different doses and (Pp0:04) and (b) photomicrograph of rat cervical spinal cord with dose of 15 Gy X-rays (40 , H&E staining). Glioma is seen in the photomicrograph.
Prominent changes consisted of secretory vesicles proliferation and damage to cell membrane and nucleus configuration. Fig. 5 shows some early variations in endothelial cells of post capillary vessels after irradiation dose of 6 Gy. Histopathologic examination of rats irradiated with single dose of 35 Gy showed necrosis at Wm with gliosis and/or glioma and demyelination, which were accompanied by clinical symptoms of severe myelopathy at 17 weeks after irradiation. Vascular changes were also prominent consisting rupture, collapse and thrombus formation accompanied by intense inflammatory condition including mononuclear cell infiltration and vascular density decrease in stroma of Wm (Fig. 6). All animals of this group manifested clinical signs of RM including urinary incontinence, weakness of forelimbs, paralysis of hind and fore-limbs and finally death.
4. Discussion & conclusion RM is defined as parenchymal damage of spinal cord due to substructural effects of particulate or electromagnetic types of radiations. It is known that prostaglandins can affect different tissues and biological
Fig. 6. Photomicrographs of paralyzed rats at 17 weeks postirradiation with dose of 35 Gy: (a) Irregular myelin sheath, necrotic area (400 ) and (b) proliferation of glial cells (40 ) are seen (H&E staining).
systems and particularly the vascular endothelial cells It is well documented that prostaglandins such as PGI2, are released within the inflammed medium by vascular endothelial cells [16,19–22]. We tried to outline the variation in secretory profile of PGI2 with simultaneous matched histopathology of radiation toxicity within the rat cervical spinal cord to explore early and delayed substructural changes that may help understanding the pathophysiology of RM. Changes in the profile of prostacyclin were determined by means of measuring its major metabolite (6-keto PGF1a). Profile of 6-ketoPGF1a shows a sigmoid pattern peaked after 13 weeks post-irradiation (Fig. 1). Destructive changes of Wm and supportive vascular tissues are the main histopathological features, which were observed in our study (Figs. 2, 3) that indicates radiation can cause histological changes even at early times after irradiation. Our findings also revealed an increase followed by a decrease of PGI2 concentration 24 h post-irradiation. However, pattern of PGI2 changes after 2 weeks showed a decreasing trend in concentration. In addition, we observed concurrent changes of Wm abnormalities 13 weeks post-irradiation, which compare to events at earlier times after irradiation, shows less severe abnormalities in the irradiated volume (Fig. 4). We think this maybe the indication of repair activity,
ARTICLE IN PRESS A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
6-keto Concentration (% of Cntrl)
Correlation coefficient (r2) =0. 7477 22 20
Total vascular changes Total white matter changes Correlation line
18 p
however, it needs a long-term followup to be clearly determined. It seems that 13 weeks after irradiation may also be critical point in accumulation of evidences that may induce later recovery [12,23]. Our study indicates profound early and delayed effects of irradiation in short-term range on the prostacyclin secretion by vascular endothelial cells, which were coincidence with histological abnormalities. We could detect 24 h post-irradiation acute increase in PGI2, which could be the first sign of acute inflammation confirmed by early histopathologic vascular changes without any evidence of marked changes in Wm parenchyma. The ultrastructural study of endothelium also showed early marked changes in cell membrane and releasing of vesicles, which may confirm the acute elevation of PGI2. Acute radiation effects were observed in endothelial cells including destruction of cell wall, nucleus shrinkage, thickening of basal membrane and detachment of endothelium even after 6 Gy at 24 h post-irradiation (Fig. 5). It shows that early changes even at low dose maybe due to the effects on subcellular level, which can induce the increase in PGI2 secretion in response to inflammatory agent. These findings suggest that after irradiation a functional endothelial change occur in blood vessels before the development of Wm necrosis. Radiobiological aspect of necrotic RM detected 17 weeks post-irradiation (Fig. 6) with 35 Gy X-ray can further magnify the role of RBE and LET of radiation. It was reported that 20 Gy irradiation with carbon-ion beam can produce RM at exactly 17 weeks postirradiation in rats thoracic spinal cord [4]. Data obtained from the early effects of low dose radiation on PGI2 production and low dose hypersensitivity, which may affect the late response of Wm parenchyma, can be initial steps toward explaining the
377
16 14 12 10
25
30 35 Irradiation Dose (Gy)
Fig. 8. Dose–response correlation between vascular and white matter (mainly neuroglia reaction) regardless of post-irradiation time. There is a correlation with coefficient of 0.7477 between them.
paradox of duality of a/b ratios when results from fractionation were fitted to the LQ model [11,24]. Dosimetrical aspect also showed the importance of 15 Gy irradiation dose. At this dose, the PGI2 level always showed the minimum amount comparing to agematched control group (Fig. 7). Similarly, decrease in PGI2 level after 15 Gy single dose of X-ray was reported by Siegal et al. [13]. Although more investigation in this field is needed to better clarify different aspects of RM and its relation to biochemical changes, our data suggest that determination of PGI2 concentration at irradiated sites can be useful as a marker in identifying early and late radiation toxicity (Fig. 8).
120
Acknowledgements
100
We are thankful to the office of Vice-Chancellor for research of Tehran University of Medical Sciences for their financial support of this research project. We also thank staff of Radiotherapy Department of Cancer Research Institute and Pharmacology Department of Faculty of Medicine.
80 60 40
References
20 0 Cntrl
24 hr 2 wks Time after irradiation
13 wks
Fig. 7. 6-Keto Prostaglandin F1a level profile of rats’ cervical spinal cord irradiated with dose of 15 Gy at 24 h, 2 and 13 weeks postirradiation. Its content ratio as 6-ketoPGF1a (irradiated)/6-ketoPGF1a (age matched control). The means7SEM were shown as percentage at confidence limit of 95%.
[1] A.L. Grosu, N. Andratschke, C. Nieder, et al., Retreatment of the spinal cord with palliative radiotherapy, Int. J. Radiat. Oncol. Biol. Phys. (2002) 1288–1292. [2] B.C. Liang, Radiation-associated neurotoxicity, Hospital Physician (1999) 54–58. [3] T.E. Schultheiss, L.C. Stephens, Invited review: permanent radiation myelopathy, Br. J. Radiol. (1992) 737–753. [4] O. Shinobu, R. Okeda, Review Article: Pathology of Radiation Myelopathy, 2001, pp. 247–265.
ARTICLE IN PRESS 378
A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
[5] G.M. Morris, J.A. Coderre, E.M. Whitehouse, et al., Boron neutron capture therapy: a guide to the understanding of the pathogenesis of late radiation damage to the rat spinal cord, Int. J. Radiat. Oncol. Biol. Phys. (1994) 1107–1112. [6] T.E. Schulteiss, L.E. Kun, K.K. Ang, et al., Radiation response of the central nervous system, Int. J. Radiat. Oncol. Biol. Phys. (1995) 1093–1112. [7] P.H. Bijl, P. Van Lujik, R.P. Coppes, et al., Dose-volume effects in the rat cervical spinal cord after proton irradiation, Int. J. Radiat. Oncol. Biol. Phys. (2002) 205–211. [8] J.F. Fowler, S.M. Bentzen, S.J. Bond, et al., Clinical radiation doses for spinal cord: the international questionnaire, Radiother. Oncol. (2000) 295–300. [9] J.W. Hopewell, K.R. Trott, Volume effects in radiobiology as applied to radiotherapy, Radiother. Oncol. (2000) 283–288. [10] A. Brahme, J. Nilsson, D. Belkic, Original article: biologically optimized radiation therapy, Acta Oncol. (2001) 725–734. [11] C.S. Wong, Y. Hao, P.R. Hill, Response of rat spinal cord to very small doses per fraction: lack of enhanced radiosensitivity, Radiother. Oncol. (1995) 44–49. [12] C.S. Wong, Y. Hao, Long-term recovery kinetics of radiation damage in rat spinal cord, Int. J. Radiat. Oncol. Biol. Phys. (1997) 171–179. [13] T. Siegal, M.R. Pfeffer, Radiation-induced changes in the profile of spinal cord serotonin, prostaglandin synthesis, and vascular permeability, Int. J. Radiat. Oncol. Biol. Phys. (1995) 57–64. [14] Y. Hosoi, M. Yamamoto, T. Ono, et al., Prostacyclin production in cultured endothelial cells is highly sensitive to low doses of ionizing radiation, Int. J. Radiat. Oncol. Biol. Phys. (1993) 631–638.
[15] L. Estrada-Garcia, J. Carrera-Rotllan, P. Puig-parellada, Effects of oxidative stress and antioxidant treatments on eicosanoid synthesis and lipid peroxidation in long term human umbilical vein endothelial cells culture, Prostaglandins & Other Lipid Mediators, 2002, pp. 13–25. [16] H. Vanegas, H.-G. Schaible, Prostagalndins and cycloxygenases in the spinal cord, Progress Neurobiol. (2001) 327–363. [17] J.J. Broerse, J. Wonder, Accuracy and precision required for dose administration in the target area; in proceeding of Dosimetry for radiobiology, Course held at Leiden, The Netherlans, 1998, pp. 12–17. [18] C.-M. Ma, C.W. Caffey, L.A. De Werd, et al., AAPM protocol for 40–300 kV X-ray beam dosimetry, Med. Phys. (2001) 868–893. [19] N. Woolf, Cell Tissue and Disease, W.B. Saunders, Philadelphia, PA, 2000, pp. 47–81. [20] S. Lindemann, C. Gierer, H. Darius, Prostacyclin inhibits adhesion of polymorphonuclear leukocytes to human vascular endothelial cells due to adhesion molecule independent regulatory mechanisms, Basic Res. Cardiol. (2003) 8–15. [21] H. Al-Badavi, H.M. Soler, M.T. Watkins, Effects of prolonged hypoxia, reoxygenation, and shear stress on Bovine aortic endothelial cell prostacyclin production, Prostaglandins lipid Mediators (1998) 127–139. [22] J.C. Menedez, D. Casanova, J.A. Amado, et al., Effects of radiation on endothelial function, Int. J. Radiat. Oncol. Biol. Phys. (1998) 905–913. [23] W. Landuyt, J. Fowler, A. Ruifrok, et al., Kinetics of repair in the spinal cord of the rat, Radiother. Oncol. (1997) 55–62. [24] C. Nieder, U. Nestle, K. Walter, et al., Dose/effect relationships for brain metastase, J. Cancer Res. Clin. Oncol. (1998) 346–350.
Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378 www.elsevier.com/locate/plefa
Short-term changes in prostacyclin secretory profile of irradiated rat cervical spinal cord Alireza Shirazia, Seied Rabie Mahdavia, Bagher Minaeeb, Alireza Nikoofarc, Ebrahim Azizid, a Department of Medical Physics, Tehran University of Medical Sciences (TUMS), Tehran, Iran Department of Anatomy and Histology, Faculty of Medicine, Tehran University of Medical Sciences (TUMS), Tehran, Iran c Faculty of Medicine, Iran University of Medical Sciences, Tehran University of Medical Sciences (TUMS), Tehran, Iran d Molecular Research Laboratory, Department of Pharmacology and Toxicology, Faculty of Pharmacy, Tehran University of Medical Sciences (TUMS), Tehran, Iran
b
Received 17 August 2004; accepted 2 February 2005
Abstract Prostaglandins changes in radiation myelopathy (RM) have been previously reported. In the present study, we decided to determine the profile of Prostacyclin (PGI2) content in irradiated rat cervical cord. Wistar rats were irradiated with doses of 2,4,6,15,25 and 30 Gy of X-rays. After 24 h, 2 and 13 weeks post-irradiation, samples of spinal cord were prepared for evaluation of PGI2 and histopathologic changes. Prostacyclin content was determined by quantification of 6-keto-prostaglandin-F1a (prostacyclin major metabolite). Irradiated segments of spinal cord were stained routinely for histological studies. Results of irradiated were compared to control groups. Average ratio values of 6-keto-PG-F1a for doses of 2–30 Gy were between 67.5% and 107%, 65.41% and 100.54%, and 62.20% and 98.89% for 24 h, 2 and 13 weeks post-irradiation, respectively. Histopathological studies showed marked gliosis and vascularities in irradiated specimens. PGI2 bimodal secretory profile was observed along with histopathological changes in this study. Our results can further emphasize on the role of PGI2 in RM. r 2005 Elsevier Ltd. All rights reserved.
1. Introduction Myelopathy is a serious and not rare complication of cancer and cancer treatment. Myelopathies may be induced through external pressure on cord, intramedullary metastasis or treatment toxicities associated with radiation therapy alone or in combination with other treatment methods. Radiation myelopathy (RM) in patients following radiotherapy made it the subject of vast clinical and radiobiological studies because it may not be diagnosed from the associated symptoms alone. Therefore, it seems there is a long way to understand the basic concept of mechanisms in the development of RM [1–4].
Corresponding author. Tel./fax: +98 21 6959100.
E-mail address: [email protected] (E. Azizi). 0952-3278/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.plefa.2005.02.003
Dose–response curves for treatment of malignancies profoundly show a strict margin between virtual 100% tumoricidal dose and neurotoxicity occurrence in spinal cord as a normal tissue complication [2]. It was shown that single dose of 35 Gy X-ray to spinal cord can cause limb paralysis in rat models with a latency of 19.070.3 weeks [5]. In conventional radiotherapy incidence of RM after doses of 57–61 Gy is 5% with a good approximation. Analysis of patients developing RM demonstrated that the upper limit of tolerance for a X5 cm length of the cervical spine was 50 Gy (25 fractionates/35 days). The mostly used dose limit for spinal cord is 45 Gy in 22–25 fractions and this is considered as tolerance dose. Yet, due to morbidity of RM, spinal cord dose should always be kept under tolerance limit [6–10]. Radiobiologically, it was reported that linear quadratic (LQ) model does not provide a satisfactory
ARTICLE IN PRESS 374
A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
description of the dose–response relationship obtained from the studies on rat spinal cord. There were even observations on different dose responses with paradoxical increase in radiosensitivity [11,12]. Until recently, findings of radiation injury to spinal cord were based entirely on morphological changes in white matter (Wm) structures, which are consisted of various degrees of demyelination, gliosis, Wm necrosis, vascular changes and inflammatory responses [13]. Recent advances reported the probable priority of vascular and its secretory profile changes in pathogenesis of RM (vascular hypothesis). Vascular hypothesis states that vascular changes occur prior to any alterations in the Wm parenchyma due to different radiosensitivity between vascular endothelium and glial cells [4]. Modern enzymatic and/or immunohistochemical techniques have been applied to animal models for studying the radiation-induced changes in the various biochemical mediators produced by different cells within the irradiated medium [13–15]. Release of prostaglandins (PGs) D2, E2, F2a and I2 in the spinal cord has been documented by others. Prostaglandin measurement has application in diagnosis of different spinal cord disorders in a range from cerebrospinal fluid to intra spinal antibody microprobes or spinal cord homogenates. A basal level of PGs may exist in many laboratory samples and it is believed that injury to tissue induces release of prostaglandins [16]. This study aims to determine the changes in secretory profile of Prostaglandin I2 (PGI2) or Prostacyclin, in short-term after irradiation (24 h, 2 and 13 weeks postirradiation) to explore its possible contribution to the development of RM.
2. Materials and methods 2.1. Animals and radiation Male Wistar rats weighing 150–200 g were fixed on jig under gentle anesthesia. The jig was specially designed and tailored in order to have a reproducible and accurate irradiation to rats’ cervical spinal cord [17]. C1 to T2 segment of spine was irradiated through two parallel opposed ports with an orthovoltage machine (200 kVp, 1.5 mm Cu HVL, Focal-Skin Distance 25 cm) at a dose rate of 153.26 cGy/min at depth of spinal cord. Doses of 2, 4, 6, 15, 25 and 30 Gy were administered using a circular 3.0 cm diameter localizator [18]. We have also irradiated one group of rats with dose of 35 Gy for evaluation of survival and to produce the clinical manifestation of RM. Five rats were included in each dose group in addition to age-matched control group. The animals were sacrificed chronologically in time intervals of 24 h, 2 and 13 weeks post-irradiation. The single group irradiated with 35 Gy was followed for
symptoms of RM and we could obtain some radiobiological data of latent period of symptoms and lethal dose of producing the paralysis. 2.2. Sample preparations Following decapitation under ether anesthesia, rats’ neck regions were opened posteriorly until to expose cervical spine. Then, the spinal cord was cut rapidly and segmented into three pieces for enzyme immunoassay (EIA) and histopathological studies with transmission electron and light microscopes. 2.3. Measurement of spinal cord PGI2 Samples for EIA were pulled out after laminectomy, frozen on dry ice, stored at 80 1C prior to analysis. Samples were then homogenized with weight proportion of 0.1 M perchloric acid, centrifuged at 4 1C for 10 min at 3000 rpm. Then 100–150 ml supernatant was removed for dilution (1:5 and 1:10) and poured into a 96 well plate of an EIA kit (Cayman Chemical Co., CA-USA). After 18 h incubation, wells were washed out and Ellman’s reagent was added to each well and samples were shaked for 90–120 min and read using plate reader (Sunrise, UK) at wavelength of 415 nm. This assay is based on the competition between 6-keto PGF1a and a 6-keto acetylcholinesterase (AChE) conjugate (6-keto PGF1a tracer) for a limited number of 6-keto specific rabbit antiserum binding sites in 96 well plates. Because the concentration of the 6-keto PGF1a tracer is held constant while the concentration of 6-keto PGF1a varies, the amount of 6-keto PGF1a tracer that is able to bind to the rabbit antiserum will be inversely proportional to the concentration of 6-keto PGF1a (pg/ml) in the wells. In this way, the kit measures the 6-keto prostaglandin F1a (6-keto PGF1a), which is the main metabolic product of prostacyclin (PGI2). Therefore, the concentration of 6keto PGF1a is directly proportional to prostacyclin. To normalize the data of different groups, 6-keto PGF1a concentrations of control group were divided by the values for irradiated rats. Statistical analysis was carried out by means of Sigma Plot package [15]. 2.4. Histopathologic study The samples for electron and light microscopy were immersed into the appropriate fixatives immediately and processed by routine procedures of staining and preparation. Each histopathologic variable was scored by the histopathologist in comparison to the control group after averaging of each dose group values (+ for mild, ++ for moderate, and +++ for severe reactions). Images of spinal cord sections were examined by a histopathologist for changes in white matter (Wm) stroma and for vascular changes in comparison to
ARTICLE IN PRESS A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
control cases. Changes in neuroglia cell proliferation and function were the most important parameters that were studied in this project. Vessels were examined for structural changes consisting the wall rupture and alteration in their density throughout the Wm parenchyma. To obtain the total effect (TE) of irradiation on each variable, after scoring all of them, the scores for each variable was summed up. The statistical data for each dose–response curve was assessed by analysis of variance (ANOVA) followed by Normality and Constant Variance tests using SPSS 10 software.
3. Results 3.1. Prostaglandin I2 changes PGI2 assay 24 h post-irradiation with doses of 2– Gy revealed a rise and fall in concentration level. Initially, it increased with doses of 2 and 4 Gy and then fall at dose of 6 Gy followed by gradual increase from 15 Gy, but did not reach to the control level (Fig. 1). Prostacyclin secretion after 2 weeks post-irradiation revealed a more complex variation. It showed a bimodal action. Initially, it showed a lower concentration than control followed by an increase in concentration of PGI2 at doses up to 6 Gy. Then it decreased steeply at doses below 15 Gy. Again, it made the second peak of concentration at dose of 30 Gy (Fig. 1). Eicosanoid studies 13 weeks post-irradiation at different doses showed an abrupt increase in PGI2 concentration. The lowest level of concentration belonged to the dose of 15 Gy, which followed by an increase in PGI2 concentration at 25 Gy.
375
Prostacyclin level of animals irradiated with 35 Gy at time of inducing the paralysis, as the radiobiological end effect, and just prior to death (17 weeks post-irradiation) revealed steep increase relative to corresponding control rats (approximately 7:1 with Po 0.002). 3.2. Histopathpathological changes Histopathological examinations 24 h post-irradiation of 2, 4 and 6 Gy did not reveal abnormality on light microscopy. However, 200 kVp X-ray doses of 15, 25 and 30 Gy showed some degrees of alterations in Wm even at 24 h post-irradiation. Radiation mostly affected the vascular bed and neuroglia cells. Mild demyelination, mild exocentric myelin sheath, vascular congestion and/or dilatation and mild gliosis often situated between the anterior horns of gray matter were consistently seen after 24 h (Fig. 2). Findings after 2 weeks irradiation showed similar histopathological changes in supportive cells of Wm but more early vascular changes even in lower doses, which could mimic the signs of inflammatory condition with dose-dependent increase in severities. Predominant vascular changes were consisted of RBC and mononuclear infiltrations, vascular dilatation, congestion, heamorrhage and thrombosis (Fig. 3). After 13 weeks post-irradiation, different X-ray doses indicated a decrease in histopathological abnormalities both in Wm neuroglia cells and vascular bed relative to the changes at 2 weeks after irradiation (Fig. 4). None of the irradiated animals with doses of 2, 4, 6, 15, 25 and 30 Gy were developed typical manifestation of RM after short term of study but groups, which received doses X15 Gy showed clear decrease in weight compared to normal control cases. Ultrastructural electron microscope images showed various changes in the endothelial cells morphology. 6 Total Effect (TE)
5 Congestion
4
MON
3 2 1 0
(a) Fig. 1. 6-Keto Prostaglandin F1a level profile of rats’ cervical spinal cord irradiated with doses of 2–30 Gy at 24 h, 2 and 13 weeks postirradiation. The result of 35 Gy single group at 17 weeks after irradiation was also shown. Its content ratio as PGI2 (irradiated)/PGI2 (age matched control). The means7SEM were shown as percentage at confidence limit of 95%.
5 10 15 20 25 30 35 Dose (Gy)
(b)
Fig. 2. Histopathological effects of irradiation after 24 h: (a) Dose–response curve for TEs in white matter parenchyma at different doses (Pp0.06) and (b) photomicrograph of rat cervical spinal cord with dose of 25 Gy X-rays (100 , H&E staining). The mass of neuroglia (MON) and vascular congestion are seen in the photomicrograph.
ARTICLE IN PRESS A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
Total effect
376
14 12 10 8 6 4 2 0
Thrombosis
2
(a)
15 25 30 4 6 log10Dose (Gy) (b)
Fig. 3. Histopathological effects of irradiation after 2 weeks: (a) Dose–response curve for TEs in white matter parenchyma at different doses (pp0.06) and (b) photomicrograph of rat cervical spinal cord with dose of 25 Gy X-rays (100 , H&E staining). Thrombosis and demyelination are seen in the photomicrograph.
Fig. 5. Transmission electron microscope (TEM) images of rat cervical spinal cord white matter post-capillary vessel endothelial cell: (a) endothelial cell wall detachment (two arrow heads), cytoplasmic membrane degeneration (arrow) after 24 h of 30 Gy irradiation, and (b) thickening of cell basal membrane (arrow head) and nucleus membrane folding (arrow) at 2 weeks post-irradiation of 6 Gy (12000 ).
Total Effect (TE)
Glioma 3.6 3.5 3.4 3.3 3.2 (a)
10 20 30 Irradiation dose (Gy)
(b)
Fig. 4. Histopathological effects of irradiation after 13 weeks: (a) Dose–response curve for TEs in white matter parenchyma at different doses and (Pp0:04) and (b) photomicrograph of rat cervical spinal cord with dose of 15 Gy X-rays (40 , H&E staining). Glioma is seen in the photomicrograph.
Prominent changes consisted of secretory vesicles proliferation and damage to cell membrane and nucleus configuration. Fig. 5 shows some early variations in endothelial cells of post capillary vessels after irradiation dose of 6 Gy. Histopathologic examination of rats irradiated with single dose of 35 Gy showed necrosis at Wm with gliosis and/or glioma and demyelination, which were accompanied by clinical symptoms of severe myelopathy at 17 weeks after irradiation. Vascular changes were also prominent consisting rupture, collapse and thrombus formation accompanied by intense inflammatory condition including mononuclear cell infiltration and vascular density decrease in stroma of Wm (Fig. 6). All animals of this group manifested clinical signs of RM including urinary incontinence, weakness of forelimbs, paralysis of hind and fore-limbs and finally death.
4. Discussion & conclusion RM is defined as parenchymal damage of spinal cord due to substructural effects of particulate or electromagnetic types of radiations. It is known that prostaglandins can affect different tissues and biological
Fig. 6. Photomicrographs of paralyzed rats at 17 weeks postirradiation with dose of 35 Gy: (a) Irregular myelin sheath, necrotic area (400 ) and (b) proliferation of glial cells (40 ) are seen (H&E staining).
systems and particularly the vascular endothelial cells It is well documented that prostaglandins such as PGI2, are released within the inflammed medium by vascular endothelial cells [16,19–22]. We tried to outline the variation in secretory profile of PGI2 with simultaneous matched histopathology of radiation toxicity within the rat cervical spinal cord to explore early and delayed substructural changes that may help understanding the pathophysiology of RM. Changes in the profile of prostacyclin were determined by means of measuring its major metabolite (6-keto PGF1a). Profile of 6-ketoPGF1a shows a sigmoid pattern peaked after 13 weeks post-irradiation (Fig. 1). Destructive changes of Wm and supportive vascular tissues are the main histopathological features, which were observed in our study (Figs. 2, 3) that indicates radiation can cause histological changes even at early times after irradiation. Our findings also revealed an increase followed by a decrease of PGI2 concentration 24 h post-irradiation. However, pattern of PGI2 changes after 2 weeks showed a decreasing trend in concentration. In addition, we observed concurrent changes of Wm abnormalities 13 weeks post-irradiation, which compare to events at earlier times after irradiation, shows less severe abnormalities in the irradiated volume (Fig. 4). We think this maybe the indication of repair activity,
ARTICLE IN PRESS A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
6-keto Concentration (% of Cntrl)
Correlation coefficient (r2) =0. 7477 22 20
Total vascular changes Total white matter changes Correlation line
18 p
however, it needs a long-term followup to be clearly determined. It seems that 13 weeks after irradiation may also be critical point in accumulation of evidences that may induce later recovery [12,23]. Our study indicates profound early and delayed effects of irradiation in short-term range on the prostacyclin secretion by vascular endothelial cells, which were coincidence with histological abnormalities. We could detect 24 h post-irradiation acute increase in PGI2, which could be the first sign of acute inflammation confirmed by early histopathologic vascular changes without any evidence of marked changes in Wm parenchyma. The ultrastructural study of endothelium also showed early marked changes in cell membrane and releasing of vesicles, which may confirm the acute elevation of PGI2. Acute radiation effects were observed in endothelial cells including destruction of cell wall, nucleus shrinkage, thickening of basal membrane and detachment of endothelium even after 6 Gy at 24 h post-irradiation (Fig. 5). It shows that early changes even at low dose maybe due to the effects on subcellular level, which can induce the increase in PGI2 secretion in response to inflammatory agent. These findings suggest that after irradiation a functional endothelial change occur in blood vessels before the development of Wm necrosis. Radiobiological aspect of necrotic RM detected 17 weeks post-irradiation (Fig. 6) with 35 Gy X-ray can further magnify the role of RBE and LET of radiation. It was reported that 20 Gy irradiation with carbon-ion beam can produce RM at exactly 17 weeks postirradiation in rats thoracic spinal cord [4]. Data obtained from the early effects of low dose radiation on PGI2 production and low dose hypersensitivity, which may affect the late response of Wm parenchyma, can be initial steps toward explaining the
377
16 14 12 10
25
30 35 Irradiation Dose (Gy)
Fig. 8. Dose–response correlation between vascular and white matter (mainly neuroglia reaction) regardless of post-irradiation time. There is a correlation with coefficient of 0.7477 between them.
paradox of duality of a/b ratios when results from fractionation were fitted to the LQ model [11,24]. Dosimetrical aspect also showed the importance of 15 Gy irradiation dose. At this dose, the PGI2 level always showed the minimum amount comparing to agematched control group (Fig. 7). Similarly, decrease in PGI2 level after 15 Gy single dose of X-ray was reported by Siegal et al. [13]. Although more investigation in this field is needed to better clarify different aspects of RM and its relation to biochemical changes, our data suggest that determination of PGI2 concentration at irradiated sites can be useful as a marker in identifying early and late radiation toxicity (Fig. 8).
120
Acknowledgements
100
We are thankful to the office of Vice-Chancellor for research of Tehran University of Medical Sciences for their financial support of this research project. We also thank staff of Radiotherapy Department of Cancer Research Institute and Pharmacology Department of Faculty of Medicine.
80 60 40
References
20 0 Cntrl
24 hr 2 wks Time after irradiation
13 wks
Fig. 7. 6-Keto Prostaglandin F1a level profile of rats’ cervical spinal cord irradiated with dose of 15 Gy at 24 h, 2 and 13 weeks postirradiation. Its content ratio as 6-ketoPGF1a (irradiated)/6-ketoPGF1a (age matched control). The means7SEM were shown as percentage at confidence limit of 95%.
[1] A.L. Grosu, N. Andratschke, C. Nieder, et al., Retreatment of the spinal cord with palliative radiotherapy, Int. J. Radiat. Oncol. Biol. Phys. (2002) 1288–1292. [2] B.C. Liang, Radiation-associated neurotoxicity, Hospital Physician (1999) 54–58. [3] T.E. Schultheiss, L.C. Stephens, Invited review: permanent radiation myelopathy, Br. J. Radiol. (1992) 737–753. [4] O. Shinobu, R. Okeda, Review Article: Pathology of Radiation Myelopathy, 2001, pp. 247–265.
ARTICLE IN PRESS 378
A. Shirazi et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 72 (2005) 373–378
[5] G.M. Morris, J.A. Coderre, E.M. Whitehouse, et al., Boron neutron capture therapy: a guide to the understanding of the pathogenesis of late radiation damage to the rat spinal cord, Int. J. Radiat. Oncol. Biol. Phys. (1994) 1107–1112. [6] T.E. Schulteiss, L.E. Kun, K.K. Ang, et al., Radiation response of the central nervous system, Int. J. Radiat. Oncol. Biol. Phys. (1995) 1093–1112. [7] P.H. Bijl, P. Van Lujik, R.P. Coppes, et al., Dose-volume effects in the rat cervical spinal cord after proton irradiation, Int. J. Radiat. Oncol. Biol. Phys. (2002) 205–211. [8] J.F. Fowler, S.M. Bentzen, S.J. Bond, et al., Clinical radiation doses for spinal cord: the international questionnaire, Radiother. Oncol. (2000) 295–300. [9] J.W. Hopewell, K.R. Trott, Volume effects in radiobiology as applied to radiotherapy, Radiother. Oncol. (2000) 283–288. [10] A. Brahme, J. Nilsson, D. Belkic, Original article: biologically optimized radiation therapy, Acta Oncol. (2001) 725–734. [11] C.S. Wong, Y. Hao, P.R. Hill, Response of rat spinal cord to very small doses per fraction: lack of enhanced radiosensitivity, Radiother. Oncol. (1995) 44–49. [12] C.S. Wong, Y. Hao, Long-term recovery kinetics of radiation damage in rat spinal cord, Int. J. Radiat. Oncol. Biol. Phys. (1997) 171–179. [13] T. Siegal, M.R. Pfeffer, Radiation-induced changes in the profile of spinal cord serotonin, prostaglandin synthesis, and vascular permeability, Int. J. Radiat. Oncol. Biol. Phys. (1995) 57–64. [14] Y. Hosoi, M. Yamamoto, T. Ono, et al., Prostacyclin production in cultured endothelial cells is highly sensitive to low doses of ionizing radiation, Int. J. Radiat. Oncol. Biol. Phys. (1993) 631–638.
[15] L. Estrada-Garcia, J. Carrera-Rotllan, P. Puig-parellada, Effects of oxidative stress and antioxidant treatments on eicosanoid synthesis and lipid peroxidation in long term human umbilical vein endothelial cells culture, Prostaglandins & Other Lipid Mediators, 2002, pp. 13–25. [16] H. Vanegas, H.-G. Schaible, Prostagalndins and cycloxygenases in the spinal cord, Progress Neurobiol. (2001) 327–363. [17] J.J. Broerse, J. Wonder, Accuracy and precision required for dose administration in the target area; in proceeding of Dosimetry for radiobiology, Course held at Leiden, The Netherlans, 1998, pp. 12–17. [18] C.-M. Ma, C.W. Caffey, L.A. De Werd, et al., AAPM protocol for 40–300 kV X-ray beam dosimetry, Med. Phys. (2001) 868–893. [19] N. Woolf, Cell Tissue and Disease, W.B. Saunders, Philadelphia, PA, 2000, pp. 47–81. [20] S. Lindemann, C. Gierer, H. Darius, Prostacyclin inhibits adhesion of polymorphonuclear leukocytes to human vascular endothelial cells due to adhesion molecule independent regulatory mechanisms, Basic Res. Cardiol. (2003) 8–15. [21] H. Al-Badavi, H.M. Soler, M.T. Watkins, Effects of prolonged hypoxia, reoxygenation, and shear stress on Bovine aortic endothelial cell prostacyclin production, Prostaglandins lipid Mediators (1998) 127–139. [22] J.C. Menedez, D. Casanova, J.A. Amado, et al., Effects of radiation on endothelial function, Int. J. Radiat. Oncol. Biol. Phys. (1998) 905–913. [23] W. Landuyt, J. Fowler, A. Ruifrok, et al., Kinetics of repair in the spinal cord of the rat, Radiother. Oncol. (1997) 55–62. [24] C. Nieder, U. Nestle, K. Walter, et al., Dose/effect relationships for brain metastase, J. Cancer Res. Clin. Oncol. (1998) 346–350.