Onoda.qxd

10.1053.paor.1999.0181 available online at http://www.idealibrary.com on Radiation Induced Endothelial Cell Retraction in vitro:
Correlation with Acute Pulmonary Edema
James M ONODA, Seema S KANTAK, Clement A DIGLIO The Gershenson Radiation Oncology Center and the Departments of Radiation Oncology and Pathology, Wayne State University School of Medicine, Detroit, USA We determined the effects of low dose radiation
racic or whole body radiation (≥500 cGy). Little or
(<200 cGy) on the cell-cell integrity of confluent
no data is available concerning time intervals <1
monolayers of pulmonary microvascular endothe-
day post irradiation, possibly because of the pre-
lial cells (PMEC). We observed dose- and time-
sumption that edema is mediated, at least in part, by
dependent reversible radiation induced injuries to
endothelial cell death or irreversible loss of barrier
PMEC monolayers characterized by retraction (loss
permeability functions which may only arise ≥1 day
of cell-cell contact) mediated by cytoskeletal F-actin
post irradiation. However, our in vitro data suggest
reorganization. Radiation induced reorganization
that loss of endothelial barrier function may occur
of F-actin microfilament stress fibers was observed
rapidly and at low dose levels (≤200 cGy). Therefo-
≥30 minutes post irradiation and correlated posi-
re, we determined radiation effects on lung wet
tively with loss of cell-cell integrity. Cells of irradi-
weight and observed significant increases in wet
ated monolayers recovered to form contact inhibit-
weight (standardized per dry weight or per mouse
ed monolayers ≥24 hours post irradiation; concomi-
weight) in ≤5 hours post thoracic exposure to 50–200
tantly, the depolymerized microfilaments organized
cGy x-radiation. We suggest that a single fraction of
to their pre-irradiated state as microfilament stress
radiation even at low dose levels used in radiother-
fibers arrayed parallel to the boundaries of adjacent
apy, may induce pulmonary edema by a reversible
contact-inhibited cells. Previous studies by other
loss of endothelial cell-cell integrity and permeabil-
investigators have measured slight but significant
ity barrier function. (Pathology Oncology Research
increases in mouse lung wet weight >1 day post tho-
Introduction
included in the radiation treatment field.1,2 These injuriesare thought to be largely a result of the lethal irradiation of We are in general agreement with those investigators the endothelium and are manifest by loss of endothelial that have described a role for radiation-induced alterations monolayer integrity and denudation of the microvascular in the pulmonary endothelial structure and function as the lumen.3,4 However, most studies directed towards analysis critical mediator of the pathogenesis of lung injury. It is of edema in animals exposed to single fractions of high known that the severity of radiation induced injuries to the dose (>500 cGy) radiation, as opposed to the lower dose endothelium is largely dependent on the volume of lung The radiation-induced loss of endothelial cell integrity leads to the exposure of basal lamina which results in the Received: Febr 16, 1999; accepted: Febr 23, 1999 leakage of plasma from vessels into the interstitium.7,8 Correspondence: Dr. James ONODA, research director, Biomide This radiation-induced vascular permeability is an essen- Corporation, 407 Life Science Building, Wayne State University, tial element, if not an initiator, for the development of the Detroit, MI 48202, USA; Tel: 313 577 2184; fax: 248 594 4884;E-mail: [email protected] acute and late radiation injuries of edema and fibrosis.9-11 These studies were supported by PHS CA50465 and the Gershen- The time course between lethal exposure and denudation or loss of vascular integrity has usually been found to be 1999 W. B. Saunders & Company Ltd on behalf of the Arányi Lajos Foundation approximately several days to several weeks post irradia- radiation damage, including the loss of barrier function.22,23 tion.5,8 In contrast the immediate and reversible effects of Thus, we sought to test, indirectly, whether radiation- non-lethal levels of radiation on endothelial morphological induced lipoxygenase metabolites play a role in acute integrity have gone relatively unstudied. Moreover, non- edema by pretreatment with the non-specific lipoxygenase lethal radiation may be a contributing cause or even the initiator of acute post radiation lung injuries. Becauseedema should be detected at time intervals (<1 day) post Materials and Methods
irradiation that are much earlier than the observed denuda-tion of vascular structures, we felt it important to examine Pulmonary microvascular endothelial cells (PMEC) the effects of non-lethal radiation at short time intervalspost radiation. In addition, the dose levels used in the stud- Lungs were aseptically removed from C57Bl6J mice ies reported here are significant in that they are within the and immersed in Ca++ and Mg++ free Hank’s Balanced Salt range that lung microvasculature in the target volume Solution (HBSS). Following this initial wash, the pleural would receive during a single exposure of fractionated lining of the lung was fixed gently by applying 70% radiotherapy, or that the lung microvasculature in the ethanol over the lung surface. This procedure eliminated treatment field might receive during the cumulative course mesothelial cell contamination in the developing culture preparation. After rinsing in HBSS, small tissue pinches Reports by other investigators suggest that the rapid (1–2 mm), using fine forceps, were obtained close to the and reversible endothelial retraction could be a result of F- lung periphery to avoid obvious large vessels. Tissue actin reorganization.14,15 For example, increased perme- explants were then treated with 0.1% collagenase (Type II, ability of the pulmonary microvasculature has been shown Worthington, Malvern, Pa) in Ca++, Mg++ free HBSS for 20 to follow the disruption of the microfilament apparatus min at 37°C. The collagenase treated tissue was carefully and the disruption (loss of cell-cell contact) between removed and plated as explants into 100 mm tissue culture apparently healthy endothelial cells. It was demonstrated dishes containing Dulbecco’s modified Eagle’s medium that retraction of the cell cytoplasm and the disruption of (DMEM) supplemented with 20% fetal bovine serum the microfilament bundles occurred upon the exposure of (FBS). Isolated colonies of endothelial cells devoid of pulmonary artery endothelial cell monolayers to cytoske- spindle cells were selectively trypsinized using colony letal disrupting agents (e.g., hormone, oxidants) and were penicylinders. Three separate endothelial cell clones were presumably responsible for the increased permeabili- isolated, grown to confluency and recloned. One surviving ty.14,16,17 Therefore, the initial focus of studies presented clone was established and routinely subcultured at split here are the effects of low dose radiation on the morphol- ratios 1:2 every 7–10 days and was designated PMEC ogy and microfilaments organization of confluent, con- (pulmonary mouse endothelial cell). Endothelial cell char- tact-inhibited pulmonary microvascular endothelial cell acteristics were verified by growth behavior pattern, pres- monolayers. We examined the rate and extent of endothe- ence of factor VIII and prostanoid production according to lial retraction by phase contrast microscopy as well as the previously published procedures. For use in the studies rate and extent of F-actin depolymerization by indirect im- presented here, PMEC (between passages 14–20) were cultured in sterile DMEM +10% FBS. They were main- Acute radiation damage to the endothelium results in tained at 37°C, 95% room air, 5% CO in a water saturat- metabolic dysfunction of eicosanoid metabolism and ed incubator. Medium was changed every three days and increased permeability.18 Similar observations have been cultures were passaged [2.5 mM EDTA (Sigma Chemical made in vivo and in vitro in normal host tissue and in Co, St Louis, MO) and 0.25% trypsin (Worthington Bio- tumors.19 It is apparent that the amount of radiation (i.e., chemical Corporation, Freehold, NJ)] when a contact-inhi- 300–5000 cGy) as well as the time interval between radia- bited monolayer was achieved (approximately once a tion and determination of response (i.e., 1–14 days), are week). Cultures used for experimental purposes were critical variables affecting the perceived response(s) of seeded in T-75 culture flasks and used when confluent.
endothelial cells to radiation.18-20 Recent studies havedemonstrated that significant levels of lipoxygenase prod- Endothelial cell retraction studies ucts are released from irradiated bovine aortic endothelialcells.21 Moreover, we have observed (data not reported) PMEC were grown to confluency in alternate rows of that radiation stimulates endothelial cell (PMEC) biosyn- 16 mm 24-well plates (Falcon 3047, flat-bottom). Cells thesis of several eicosanoid metabolites including the were plated at the appropriate density (1.5x105 cells/well) lipoxygenase products 11-, 12-, and 15-HETE and unre- so that a confluent monolayer was formed in less than 48 solved leukotrienes. Products of both the COX and the hours. Monolayers were irradiated using a Picker X-ray LOX pathways have been implicated in several aspects of unit (dose rate of 205 cGy/ min, 280 kV, 20 mA, 1.3 mm Endothelial Cell Retraction and Pulmonary Edema Cu HVL). Following X-irradiation, samples were incubat- (between the mediastinum and head). In all experiments, ed 37°C 5% CO ) for various time intervals (0, 2, 4, 8, 24, mice were randomly chosen front stock cages for each treat- 48 hour). After the appropriate incubation period, culture ment group. Mice were pretreated, simultaneously treated medium was aspirated from the wells and 1 ml of fixative and/or post-treated with specific lipoxygenase inhibitors (1% paraformaldehyde, 2% glutaraldehyde in HBSS, pH using specific schedules and vehicles. [Note: Animal care 7,4) was added. Cells were fixed for 10 minutes at room was in accordance with institutional guidelines.] temperature. They were than washed once and storedunder 0.75 ml of HBSS. After fixation samples were immediately observed under phase-contrast and pho-tographed at a magnification of 400x using Kodak Pan- Immediately after sacrifice, the lungs were dissected and rinsed by saline spray under a dissecting microscope to becleaned of all other tissue. Care was taken that the saline Immunofluorescent staining of cytoskeletal elements did not enter the trachea by inverting the lungs and gentlyshaking off the fluid. After cleaning, the lungs were placed Rhodamine-labeled phalloidin (Molecular Probes Inc., on absorbent tissue, gently blotted five times and trans- Eugene, Oregon) was used for identification of F-actin fil- ferred into pre-weighed polycarbonate weigh boats which aments. Sterile 18 mm2 coverslips were transferred to 35 were covered and weighed. Dry weights were also deter- mm 6-well flat bottom plates (Corning, NY). PMEC mined. Lungs were dried for about 40 hours in an oven at (2.5x105 cells/ml) were added to coverslips in DMEM + 60°C and reweighed. The weight of each lungs was deter- 10% FBS to form contact-inhibited monolayers in 48 mined and correlated with dry weight and mouse weight.
hours. Monolayers were irradiated (12.5-200 cGy) and The average weight of <7 week old mice is approx. 12 then were terminated at appropriate time intervals post grams. Mice were irradiated on same day (Wednesday) of irradiation. Lysis squirting was used to access the cytoske- each week at the same time of day (9:00 AM) to standard- leton of PMEC. This technique uses osmotic swelling and ize for possible chronological effects on weight etc.
cell lysis to remove the dorsal cell surface to expose the [Control and irradiated C57Bl6J mice were sacrificed by cytoskeletal elements. Cells were rinsed with HEPES three methods, CO gassing, sodium pentobarbitol anesthe- buffer (10 mM HEPES, 100 mM KCl, 5 mM MgCl 3 mM sia and direct cervical dislocation. No significant differ- EGTA, pH 7.0) and then incubated (10 min) in a 20% dilu- ences in lung weights among groups were observed.] tion of HEPES buffer to induce osmotic swelling and celllysis visualized by visual inspection using phase-contrast Results
microscopy. After cell lysis samples were rinsed rigorous-ly (to remove membrane fragments), and fixed with 4% Radiation therapy of the thorax for treatment of breast paraformaldehyde in HEPES buffer (room temperature and lung cancer and Hodgkin’s disease is often associated (30 min). Samples were then washed 3X with HEPES with pulmonary edema and fibrosis resulting in compro- buffer. Labeling of F-actin (rhodamine phalloidin) was mised lung function.1,2 At the clinical level the manifesta- accomplished by a single step staininq using 75 µl of 1:50 tion of these pathological states can only be detected a rhodamine phalloidin. After labeling, samples were rinsed posteriori when the patient is symptomatic for these syn- (5x, HEPES) and mounted in a medium containing 1:2 dromes. We believe that an understanding of the mecha- glycerol:HBSS supplemented with 0.1 g Cytifluor (Amer- nism(s) which mediate radiation induced edema and fibro- sham, Arlington Heights, IL) and 0.2 g of Mowiol 4–88 sis may lead to the development of adjuvant therapies and (Calbiochem Corporation, La Jolla, CA) to prevent rapid the use of specific inhibitors of eicosanoid metabolism to quenching under fluorescent excitement. Samples were greatly reduce or inhibit the development of these injuries.
analyzed at 600X magnification under oil using a Nikon We used phase contrast photomicroscopy to record that Optiphot Microscope. Micrographs were recorded on radiation (50–200 cGy) initiates retraction of contact inhibited PMEC monolayers. We found retraction wastime-dependent and dose-dependent24 in that the degree of retraction increased as the time interval between radiationand observation increased. Figure 1a–d, is representative Unanesthetized C57Bl6J male mice, Jackson Laborato- of our studies. Figure 1a is control PMEC monolayer ries, Bar Harbor ME (6–7 wk old) were exposed to 0.5–2.0 demonstrating the characteristic morphology contact- Gy of radiation [X-ray source (Picker unit, 280 Kev) that inhibited capillary endothelial cells. Figure 1b demon- delivered 275 cGy/min.]. Mice were held in a plastic cylin- strates the F-actin cytoskeleton prior to irradiation, Note der which is blocked with lead and calibrated to ensure that the prominent stress fiber spanning large areas of the cell the target volume was restricted to the thoracic region body. Retraction of cells of the PMEC monolayer was Figure 1. Effect of x-irradiation (50 cGy) on the morphology of pulmonary microvascular endothelial cell monolayers (PMEC).
PMEC were seeded on fibronectin 24 hours to form confluent monolayers. After irradiation, PMEC were incubated for various time
intervals before fixation. This figure is representative of the studies. 1a represents PMEC monolayers prior to radiation exposure.
The characteristic morphology of contact-inhibited capillary endothelial cells is clearly evident. 1b presents the F-actin cytoskele-
ton prior to irradiation, Prominent stress fiber spanning large areas of the cell body are readily visible. 1c represents the retracted
cells induced by a single fraction of 0.5 Gy. Concomitant analysis of F-actin demonstrated the loss of cytoskeletal organization,
although some fibers are still visible at the retracted cell periphery (1d).
induced by a single fraction of 0.5 Gy. Significant retrac- lag time for the induction of retraction or cytoskeletal reor- tion was recorded 4–8 hours post exposure (Figure 1c). Analysis of F-actin demonstrated the loss of cytoskeletal Our working hypothesis predicts that radiation promotes organization. The F-actin stress fibers have resolved, endothelial retraction via stimulation of lipoxygenase although some fibers are still visible at the retracted cell metabolism. Accordingly, a general lipoxygenase inhibitor periphery (Figure 1d). Retraction was reversible, and the should inhibit radiation induced retraction. We treated con- PMEC cells resumed their appearance as contact inhibited fluent PMEC monolayers with the lipoxygenase inhibitor monolayers within 24 to 30 hrs post irradiation, which cor- NDGA (10 µM) 15 minutes prior to and during 50 cGy responded to the repolymerization of F-actin fibers (not radiation. Retraction was complete inhibited, whereas pre- shown). The ability of the cells to regain their monolayer treatment with the cyclooxygenase inhibitor indomethacin appearance and develop cell-cell contacts indicates that failed to inhibit retraction (data not shown).
energy-dependent cytoskeletal reorganization was not Previous studies by other investigators have measured impaired by 0.5 Gy radiation and suggests that levels of slight but significant increases in mouse lung wet weight >1 radiation below that dose do not significantly impair nor- day post thoracic or whole body radiation (>500 cGy).22,32 mal PMEC metabolic activity. There appears to be a Little or no data is available concerning time intervals <1 threshold for radiation initiated PMEC retraction, and day post irradiation, possibly because of the presumption once crossed, further increases in radiation dose level fail that edema is mediated, at least in part, by endothelial cell to increase the rate or extent of retraction or decrease the death or irreversible loss of barrier permeability functions Endothelial Cell Retraction and Pulmonary Edema in increased edema. There was a significant increase in lungwet weight at all dose levels five hours post irradiation (Figure 3). At three hours post irradiation, 100 and 200 cGy, but not 50 cGy exposure resulted in significantlyincreased lung weights (data not shown). Finally, we observed protective effect by pretreatment of animals withthe lipoxygenase inhibitor NDGA, which quite effectively blocked acute edema (Figure 4). NDGA was administered i.p. 15 minutes prior to radiation exposure.
Discussion
In the treatment of pulmonary neoplasms, breast carci- noma, esophageal carcinoma or Hodgkin’s disease, the riskof complications to the normal pulmonary tissues is a major limitation in the prescription of the therapeutic dose.1-3 The treatment of neoplasia by radiation requires part or thewhole of the thorax to be in the radiation field. Radiation induced edema, pneumonitis and fibrosis are well-docu-mented complications in patients receiving such treat- ments.25,26 Early reactions develop within days or weeks, Figure 2. C57Bl6J mice were randomly selected for sham or
whereas late reactions require months or years.25-27 The clin- exposure to 200 cGy radiation. Data are for individual total ical presentation of radiation-induced lung damage princi- lung weight from each mouse used in the study. The weight of pally depends on the lung volume irradiated, the radiation lungs from mice sacrificed 5 hours post radiation exposure were dose and the pre-existing lung disease.28,29 Mah et al have significantly different (p<0.01 by students t test) when com- established a distinct dose-response relationship between pared to the weight of lungs from mice in the sham irradiatedgroup or from mice sacrificed 48 hours post radiation exposure. which may only arise >1 day post irradiation. However, ourworking hypothesis predicts that low dose radiation at lev- els traditionally employed for radiotherapy should induceacute edema in the pulmonary microvasculature. Thisedema would be mediated by loss of endothelial cell-cell integrity induced by the direct effects of radiation onmicrovascular endothelial morphology as well the impetussupplied by adherent and migrating neutrophils and mono- cytes as they passage from the lumen to the subendothelial matrix and to the interstitium. Our in vitro data clearlydemonstrated that loss of endothelial barrier function occurs rapidly (≤4 hours) and at low dose levels (≤200 cGy).
Therefore, we performed a series of studies to verify a cor-relation between the time and dose effects for radiation- induced loss of endothelial cell-cell integrity in vitro andradiation induced acute edema (as determined by effects onlung wet weight) in vivo. We first demonstrated a time course for radiation-induced edema. Mice were exposed to thoracic radiation of 200 cGy. We observed significant increases in lung wet weight (standardized per dry weight or Figure 3. C57Bl6J mice were randomly selected for sham or
per mouse weight) for time points 3 and 5 hours post irradi- exposure to 50, 100, or 200 cGy radiation. Data are for individ- ation (Figure 2). By 48 hours post-irradiation, there was no ual total lung weight from each mouse used in the study. The statistically significant increase in lung weights, suggesting weight of lungs from mice exposed to 100 or 200 cGy were sig- a recovery from acute edema. We also observed a dose- nificantly different (p<0.01 by student’s t test) when compared response effect, with increased radiation exposure resulting to the weight of lungs from mice in the sham irradiated group. endothelial cells characterized by retraction and the result-ing loss of close contact between individual cells withinthe monolayer. By phase-contrast microscopy, one obser-ves an apparent retraction and loss of contact between cells resulting in the formation of gaps (between the cells)The radiation-induced cellular retraction was time anddose-dependent. Retraction was first observed at >1 hour post radiation and the extent of retraction increased with time. At the earliest stage of retraction, the cells usually demonstrated a loss of association with the adjacent cellsbut only in limited areas of the cell periphery, not around the entire cell margin. The extent of loss of contact between cells increased with time, and at maximum retrac-tion (>4 hours), there was a complete loss of contact bet- ween adjacent retracted cells and large regions of themonolayer had resolved into isolated cells that were com-pletely separated from adjacent cells. We also observedthat the extent of retraction was dose dependent and that the time interval to reach maximum retraction decreasedwith increased dose level.
Figure 4. C57Bl6J mice were randomly selected for sham or
Because a role for microfilaments, but not for micro- exposure to 200 cGy radiation. Sham mice were randomly treat- tubules, has been previously demonstrated in transient ed with NDGA (25 mM, ip injection, 15 minutes prior to pro- hormone-induced cellular retraction and respreading, we cedure) or vehicle. Mice exposed to 200 cGy radiation were sim-ilarly randomized to vehicle and NDGA treated groups. Data examined the effects of radiation on microfilament organi- are for individual total lung weight from each mouse used in the zation. We observed that the centrally located stress fiber study. The weight of lungs from sham exposed mice treated with bundles “disappear” in response to radiation, and it is this vehicle or NDGA were not significantly different and were radiation induced depolymerization of the microfilaments pooled. The weight of lungs from mice treated with vehicle and that comprise the centrally located stress fiber bundles that exposed to 200 cGy were significantly different (p<0.05) from appears to be causal for the morphological change of control mice. In contrast, the weight of lungs from mice treated retraction. We observed an absolute and positive correla- with NDGA prior to radiation exposure were similar to the con- tion between radiation-induced F-actin depolymerization and the dose- and time-dependent radiation-induced ret-raction. The time course for changes in microfilament or- the incidence of acute radiation-induced pulmonary damage ganization (i.e., F-actin depolymerization) were perfectly for human pulmonary tissues to fractionated radiotherapy coincident with the time course for morphological changes.
using average lung dose in the high dose region.30 For example, profound F-actin depolymerization was seen Control of radiation lung damage has been attempted at 2 hours post radiation at 50 and 100 cGy, which coin- using many procedures which have centered on fractionat- cides with the retraction seen at these two doses at 2 hours.
ed doses and low dose rates.31 Lung correction and shield- Conversely, lower dose levels (12.5, 25 cGy) failed to ini- ing are routinely employed in radiotherapeutic practice to tiate F-actin depolymerization at 2 hours and no retraction reduce adverse lung injury. Unfortunately in the treatment of pulmonary and thoracic neoplasms it is inevitable that a We also demonstrated (indirectly) a role for lipoxyge- certain part of normal lung tissue will fall within the treat- nase products in radiation-induced endothelial cell retrac- ment volume. Adjuvant therapy using corticosteroids which tion. Pretreatment with a variety of lipoxygenase inhibi- are potent inhibitors of inflammatory edema32 are used to tors (e.g., NDGA) blocked radiation-induced retraction.
prevent radiation injuries but the precise cellular/intracellu- Inhibition was both dose- and time-dependent, and appli- lar target sites of action are unknown.9,10 An understanding cation of NDGA after irradiation failed to block retraction.
of the basic biochemical mechanisms underlying the events In contrast, pretreatment with the cyclooxygenase inhibi- leading to edema, pneumonitis and fibrosis would facilitate tor, indomethacin, failed to block retraction.
the identification or specific inhibitors capable of blocking We used the radiation dose levels and time course for both the acute/early and late injuries.
PMEC retraction in vitro to design studies to determine We report here that low dose radiation (50–200 cGy) radiation-induced acute edema in vivo. We demonstrated produces significant changes in the morphology and that low dose thoracic radiation induces pulmonary edema microfilament organization of pulmonary microvascular as characterized by increased lung wet weight. The inci- Endothelial Cell Retraction and Pulmonary Edema dence of increased weight was radiation dose-dependent to 13.² Van Houtte P: Radiation and chemotherapy induced lung toxi- 2.0 Gy and was coincident with the time course for radia- city. Int J Rad Oncol Biol Phy 13:647-649, 1987.
14.² Aubin JE, Alders E, Heersche JNM: A primary role for micro- tion-induced endothelial retraction in vitro. Finally, we filaments, but not microtubules, in hormone-induced cytoplas- determined that pretreatment of animals with 25 µM NDGA mic retraction. Exp Cell Res 143:439-450, 1993.
15 minutes prior to radiation exposure inhibited radiation- 15.² Shasby MD, Shasby SS, Sullivan JM, et al: Role of endothelial induced edema. These observations were also in perfect cytoskeleton in the control of endothelial permeability. Circ Res agreement with our in vitro studies.
We suggest that our PMEC model system may prove 16.² Shasby MD, Lind SE, Shasby SS, et al: Reversible oxidant- induced increases in albumin transfer across cultured endotheli- useful for the screening of compounds that may prove um: Alterations in cell shape and calcium homeostasis. Blood clinically useful for the prevention of acute and late radia- tion injuries to the lungs and other normal tissues. The 17.² Wong WKK, Gotlieb AI: Endothelial cell monolayer integrity. I.
studies presented here demonstrate an initial step in iden- Characterization of the dense peripheral band of microfila- tifying agents (e.g., NDGA) which block radiation injuries ments. Arteriosclerosis 6:212-221, 1986.
18.² Friedman M, Saunders S, Madden MC, et al: Effects of ioniz- ing radiation on the pulmonary endothelial cell uptake of alphaaminoisobutyric acid and synthesis of prostacyclin. Radiat Res106:171-181, 1986.
References
19.² Degowin RL, Lewis LJ, Hoak JC, et al: Radiosensitivity of human endothelial cells in culture. J Lab Clin Med 84:42-48, 1.² Jochelson MS, Tarbell MJ, Weinstei, HJ: Unusual thoracic radi- ographic findings in children treated for Hodgkin’s disease. J 20.² Hahn GL, Menconi MJ, Cahill M, et al: Influence of gamma radiation on arachidonic acid release and prostacyclin synthesis.
2.² Fulkerson WJ, McLendon RE, Posnitz LR: Adult respiratory distress syndrome after limited thoracic radiotherapy. Cancer 21.² Eldor A, Vlodavsky I, Hyam E, et al: Effect of radiation on prostacyclin production by cultured endothelial cells.
3.² Shankar PG, Kimler BF, Giri UP, et al: Comparison of single fractionated and hyperfractionated irradiation on the develop- 22.² Farrukh IS, Michael JR, Peters SP, et al: The role of cyclooxy- ment of normal tissue damage in rat lung. Int J Rad Onc Phy genase and lipoxygenase mediators in oxidant-induced lung injury. Am Rev Respir Dis 137:1343-1349, 1988.
4.² Ward WF, Sharplin J, Franko AJ, et al: Radiation-induced pul- 23.² Ward PA, Sulavik MC, Johnson KJ: Rat neutrophil activation monary endothelial dysfunction and hydroxyproline accumula- and effects of lipoxygenase and cyclooxygenase inhibitors. Am tion in four strains of mice. Rad Res 120:113-120, 1989.
5.² Penny DP, Siemann DW, Rubin P, et al: Morphological corre- 24.² Kantak SS, Diglio CA, Onoda JM: Low dose radiation- lates of fractionated radiation of the mouse lung: early and late induced endothelial cell retraction. Int J Radiat Biol 64:319- effects. Int J Rad Oncol Biol Phys 29:789-804, 1994.
6.² Down JD, Nicholas D, Steel GG: Lung damage after hemitho- 25.² Siemann DW, Hill RP, Penny DP: Early and late pulmonary tox- racic irradiation: Dependence on the mouse strain. Radiother icity in mice evaluated 180 and 420 days following lung radia- 7.² Vergera JA, Raymond U, Thet LA: Changes in lung morphology 26.² Jennings FL, Arden A: Development of radiation pneumonitis.
and cell number in radiation pneumonitis and fibrosis: A quan- titative ultrastructural study. Int J Rad Oncol Biol Phy 13:723- 27.² Gross NJ: Pulmonary effects of radiation therapy. Ann Int Med 8.² Law MP, Ahler RG: Vascular and epithelial damage in the lung 28.² Germon PA, Brady LW: Physiologic changes before and after of mouse after x-ray or neutrons. Radiat Res 117:128-144, 1989.
radiation treatment for carcinoma of the lung. J Am Med Assoc 9.² Yi ES, Bedoya A, Lee H, et al: Radiation-induced lung injury in vivo expression of transforming growth factor-beta precedes 29.² Prato FS, Kurdyak R, Saibil EA, et al: Physiologic and radio- fibrosis. Inflammation 20:339-352, 1996.
logic assessment during the development of pulmonary radia- 10.² Gross NJ, Holloway NO, Narine KR: Effects of some non- tion and fibrosis. Radiology 122:398-397, 1977 steroidal anti-inflammatory agents on experimental radiation 30.² Mah K, VanDyke J, Keane T, et al: Acute radiation-induced pul- pneumonitis. Radiat Res 127:317-324, 1991.
monary damage:A clinical study on the response of fractionated 11.² Fauroux B, Clement A, Tournier G: Pulmonary toxicity of drugs radiation therapy. Int J Rad Onc Biol Phy 13:179-188, 1987.
and thoracic irradiation in children. Rev Mal Respir 13:235-242, 31.² Fenessey FJ: Irradiation damage to the lung. J Thoracic Imag 12.² Green GM, Finkelstein JZ, Yefft MF, at al: Diffuse interstitial 32.² Evans ML, Graham MM, Mahler PA, et al: Use of steroids to pneumonitis after pulmonary irradiation for metastatic Wilm’s suppress vascular response to radiation. Int J Rad Onc Biol Phy

Source: http://www.por.hu/1999/5/1/0049/0049a.pdf