Effects of Heavy Ion Radiation on Epithelial Cell Membranes

THE EFFECT OF HEAVY ION RADIATION ON THE ESTABLISHMENT and maintenance of tight junctions was studied in cultured canine kidney epithelial cells. When grown on permeable membranes, these cells form a monolayer with electrically resistant cell-to-cell tight junctions similar to those the epithelium form in vivo. Cells were exposed to gamma or heavy ion (C 290 MeV/nucleon or Ne 400 MeV/nucleon) radiation at doses of 5 to 25 Gy. No difference was observed in the transepithelial resistance (TER) for three days following exposure. Thereafter, irradiated cells had significantly lower TER than untreated controls (P<0.001), and the level of TER reduction was dose-dependent. A cocktail of radioprotective agents added to the culture medium reduced the alteration of TER produced when cells were irradiated with 25 Gy carbon ions (290 MeV/u). Our results illustrate that TER measurements provide a useful model for determination of physiological changes caused by radiation exposure. This measure also provides a method for testing the efficacy of radioprotective agents.

Space radiation represents an environmental hazard associated with space flight. Although data exist for human exposure to gamma rays, the physiological consequences of exposure to the high-energy charged-particle radiation which may be encountered in space is not well understood. Previous studies demonstrated oncogenic cell transformation induced by exposure to energetic protons and heavy ions,¹ but the mechanisms by which this transformation occurs remains unclear. A goal of space radiation biology includes development of methods for prevention of potential radiation-induced cancers or other disorders during and after long-term space flights. As a step toward that goal, we are utilizing epithelial cells as model systems to determine the response of "normal" cells to charged particles in an attempt to elucidate mechanisms associated with radiation carcinogenesis. Ion transport and ionic equilibrium across cell membranes are physiological processes essential in all living organisms. Mammalian epithelial cells possess polarized plasma membrane surfaces, and cell-to-cell tight junctions prevent movement of solutes and ions across the epithelium. Transepithelial electrical resistance (TER) is a measure of the movement of ions across the paracellular pathway. Measurement of TER across cells grown on permeable membranes can provide an indirect assessment of tight junction establishment and stability.^2-4 Because the epithelium separates cellular compartments with very different fluid composition, maintenance of its stability and electrical resistance is critical for many essential physiological processes. Therefore, changes in TER may represent an early expression of cell damage.

The object of this work was to determine whether TER measurements in cell culture monolayers can be used as a model system for detecting cell damage produced by exposure to radiation and for testing potential chemoprotective agents.

Methods
Tissue Culture and Irradiation. Madin-Darby canine kidney (MDCK) cells, a renal urinary epithelial cell line,⁵ were maintained at 37ºC (5% CO₂) in alpha-MEM medium with 10 percent newborn calf serum, 6.5 mM Hepes buffer, and 50 u/ml Penicillin-Streptomycin (Gibco). Cells between passages 90 and 92 were used for all the experiments. Because of the irradiation chamber characteristics and the necessity to maintain sterile conditions, irradiation was performed on cells in suspension. Cultures were grown to confluence in plastic flasks, then trypsinized and resuspended in culture medium. A fraction of the cells was reserved as a non-irradiated control, and the remaining cells were exposed to gamma or heavy ion (C 290 MeV/u, 13.25 KeV/um, 5 Gy/min; or Ne 400 MeV/u, 30.96 KeV/m, 5 Gy/min) irradiation at doses of 5 to 25 Gy. Gamma irradiation was performed at Baylor College of Medicine (Houston, TX) in a Gammacell 1000; Cs137, 10 Gy/min). Heavy ion exposure was accomplished at the National Institute of Radiological Sciences (NIRS) heavy ion medical accelerator (HIMAC) at Chiba, Japan.⁶ After irradiation, approximately 5 x 10⁵ cells were plated and grown as monolayers on 4.3 cm² polycarbonate filters (Falcon 3090) in six-well tissue culture plates. Six filters were plated for each radiation dose, and an excess of cells was applied so that the entire surface of the filters would be covered by confluent cells shortly after plating. Cell culture medium (3.5 ml) was applied to both sides of the filter representing the apical and basolateral cell surfaces, and changed every two days. Electrical resistance across cell membranes was measured daily over a period of up to 10 days with chopstick-style Ag⁺/AgCl₂ electrodes attached to a portable ohmmeter (EVOM; World Precision Instruments, Inc., Sarasota, Florida).

Chemoprevention. Reagents known to counteract cellular damage and stimulate immune response including free-radical scavengers, electrophile scavengers, and peroxidase inhibitors were combined as a "chemoprevention cocktail" and tested as a radioprotective agent. The major components of the cocktail included allylic sulfide, ellagic acid, green tea extract (polyphenols: a-tocopherol, b-carotene, superoxide dismutase), proanthocyadin extract, flavanoids, selenium, and ascorbic acid.⁷ Cells were suspended in culture medium containing 10 percent cocktail during irradiation. Afterwards, cells were seeded onto filters and maintained in the same medium. TER was determined daily.

Results
MDCK cells in suspension were irradiated with Carbon 290 MeV/u ions, Neon 400 MeV/u ions, or gamma rays. After irradiation, cells were seeded onto filters and the development of TER was measured daily. While normal cells have a stepwise increase in TER up to eight days, Figs. 1(a)-(c) illustrate that the development of TER is significantly reduced in irradiated cells as compared to controls. Repeated measures analysis of variance was calculated using the SAS' statistical analysis package. Least-squares means were computed for each treatment, and TER was plotted against days in culture. Beyond the third day after seeding, TER for cells exposed to the three types of radiation were significantly different from controls (p<0.0001). Dunnett's T test was also used to group treatment levels. In cells irradiated with carbon ions, TER remained similar until day four. Thereafter, irradiated cells displayed retarded development of TER. The level of TER reduction was dose-dependent, but the 10 Gy and 25 Gy groups did not differ from each other. With small variations, cells irradiated with neon ions or gamma rays behaved in similar fashion.

Figures 2(a)-(c) indicate the level of TER reduction as a percentage of control at different doses of radiation. For the first few days, carbon-irradiated cells maintained slightly higher TER levels than either neon- or gamma-irradiated cells. Beyond day four, however, the cellular response to irradiation appear to be similar despite differences in radiation quality. A possible cause of the reduced TER observed in irradiated cells is that a few cells die but remain in the monolayer with cells that are alive. As cells die, tight junctions would be loosened. This hypothesis does not seem to explain our results. The vital dye trypan blue was used to stain MDCK cells that had been exposed to 25 Gy gamma rays 10 days. Although TER levels were less than half the TER of control cells, treated cells remained confluent and able to exclude the dye. Therefore, the reduced TER observed in irradiated samples does not appear to result from dead cells intercalated between viable cells.

Previously tested radioprotective reagents have some disadvantages such as low efficacy, high toxicity, or the requirement of parenteral administration. We tested the chemoprevention capacity of a cocktail of reagents with multiple mechanisms of action. The rationale behind using a cocktail is that a mix of several reagents may be more effective than administering each one individually. Furthermore, as each reagent is used at lower doses than if it were alone, its possible toxic effects are lessened. For the irradiation with carbon ions, cells were exposed to the cocktail during irradiation and 96 hours afterwards. For the first three days after irradiation with a 25 Gy dose, cocktail-treated cells had lower TER than control (cells irradiated and maintained in the absence of cocktail) (Fig. 3(a)). Thereafter, cocktail-treated cells had a significant recovery. At day four, cocktail-treated and control cells had similar TER; and by day seven, cocktail-treated cells had significantly higher TER than control. The effect was not quite as dramatic for cells irradiated with a 10 Gy dose, but by day seven, cocktail-treated cells had significantly higher TER than control (Fig. 4(a)).

For irradiation with neon ions at doses of 10 or 25 Gy, cells were treated with cocktail during irradiation and 48 hours thereafter. Cells exposed to 25 Gy irradiation had slightly higher TER than control three days after irradiation (Fig. 3(b)), but cells exposed to 10 Gy appeared not to be protected at all (Fig. 5(b)). For gamma-ray exposure, cells were treated with cocktail during irradiation and 72 hours afterwards. Seven days after irradiation, cocktail-treated cells exposed to 25 Gy groups had significantly higher TER than control (Fig. 3(c)), but cocktail treatment appeared not to have any beneficial effect on cells that received a 10 Gy dose (Fig. 4(c)). A comparison of the effect of cocktail treatment vs. radiation type is shown in Fig 5. At 10 Gy levels (Fig. 5(a)), TER as a percentage of the non-irradiated control is similar for the three types of radiation with a slight possible advantage for carbon-irradiated cells. At 25 Gy level (Fig. 5(b)), the percentage of TER is much higher for carbon-irradiated cells than either neon- or gamma-irradiated cells.

Discussion
Exposure to heavy ion radiation caused a significant dose-dependent reduction in TER of renal epithelial cells. Several lines of evidence suggest that the decreased TER is not simply due to cells killed by radiation damage. First, dead cells cannot establish tight junctions with neighbor cells; therefore, dead cells will either not be included or be immediately eliminated from the monolayer. Second, dead cells intercalated in the monolayer should produce a TER near zero and not close to that of control cells we determined. Third, we did not observe any non-viable cells in monolayers of cells exposed to 25 Gy gamma irradiation and stained with the vital dye trypan blue. Therefore, our results suggest that the ability of epithelial cells to establish and maintain tight junctions was impaired after irradiation. This conclusion conforms with the observation that tight junctions were "leaky" in ileal epithelial cells from adult rats exposed to gamma rays.^8,9 Under normal conditions, intestinal epithelium should not be permeable to large-molecular-weight substances such as bacterial endotoxins. However, Walker and colleagues⁹ observed the development of aseptic endotoxemia in rats exposed to gamma rays at dose levels that do not denude the intestine of epithelial cells. The authors concluded that the endotoxemia was a consequence of irradiation injury. Interestingly, Porvaznik⁸ determined by electron microscopy that tight junctional structures of the epithelial cells were disrupted. The author suggested that tight junction fragments were produced as a result of radiation damage and removed by phagocytosis since he observed these fragments in cytoplasmic vesicles between days three and seven after irradiation. Eventually, the "leaky" ileal tight junctions returned to normal after seven days. This return-to-normal was not observed in the cell culture system we tested and may be due to the fact that some mechanisms of cell repair (e.g., phagocytosis) may not be present. Our observations, along with those of Porvaznik, suggest that epithelial cell tight junctions are a sensitive target for radiation. Therefore, part of the pathological condition known as radiation injury may be associated with the disruption of tight junctions which may provide a permissive route that allow the penetration of microbial agents and toxins normally retained in the apical side of the epithelia.

We do not know the origin of the reduced TER caused by radiation we observed. Haimovitz-Friedman and colleagues¹⁰ demonstrated that ionizing radiation can act on cellular membranes to initiate apoptosis in bovine aortic endothelial cells. These authors suggest that ionizing radiation acts at the membrane to stimulate a cascade of events involving the hydrolysis of sphingomyelin, mimicking the effects of tumor necrosis factor (TNF-a) which is known to induce apoptosis.¹¹ Indeed, it has been described that TNF-a may produce apoptosis in renal epithelial cells¹² and cause decreases in TER.³ Thus, it is possible that the radiation-dependent reduction in TER which we observed is related to the occurrence of apoptosis. Further experiments will be required to test this hypothesis.

A four-day post-irradiation treatment with the chemoprevention cocktail resulted in protection of carbon-irradiated cells. However, shorter cocktail treatments of cells irradiated with neon and gamma rays were not as successful. We were fortunate to gain access to heavy ion radiation sources, as there are only a few locations in the world where this work can be carried out. However, as radiation beam time was quite limited, we were unable to test the preventative effect of the cocktail under several doses and times of treatment. Despite the limited number of experiments, the results are encouraging, and additional experiments are been planned.

In conclusion, the present studies illustrate that determination of TER provides a useful model for testing the effects of space radiation, as well as the protective or preventive effects of chemical agents. This model is an in vitro system that uses live cells to determine physiological parameter essential for living organisms. As our results indicate, changes in TER can be detected quickly after irradiation. By this method, we can study subtle changes along a potential pathway to carcinogenesis.

Acknowledgements
The authors thank Kerry George for technical assistance, and Dr. Yoshiya Furusawa and the National Institute of Radiological Sciences (NIRS) of Japan for providing the opportunity to conduct experiments at the heavy ion medical accelerator at Chiba (HIMAC). This work was supported by a grant from ISSO to Pedemonte and by a grant from the NASA Space Radiation Health Program to Yang.

References
¹T. C. Yang, K. A. George, A. Tavakoli, L. Craise, M. Durante. "Radiogenic Transformation of Human Mammary Epithelial Cells In Vitro," Radiat. Oncol. Invest. 3 (1996): 412-19.
²S. J. Handler. "Use of Cultured Epithelia to Study Transport and its Regulation," J. Exp. Biol. 106 (1983): 55-69.
³C. W. Marano, K. V. Laughlin, L. M. Russo, A. Peralta Soler, J. M. Mullin. "Modulation of Tumor Necrosis Factor-Induced Increase in Renal (LLC-PK1) Transepithelial Resistance," J. Cell. Physiol. 157 (1993): 519-27.
⁴C. H. Pedemonte. "Inhibition of Na-Pump Expression by Impairment of Protein Glycosylation is Independent of the Reduced Sodium Entry Into the Cell," J. Membrane Biol. 147 (1995): 223-31.
⁵C. R. Gaush, W. L. Hard, T. P. Smith. "Characterization of an Established Line of Canine Kidney Cells (MDCK)," Proc. Soc. Exp. Biol. Med. 122 (1966): 931-35.
⁶T. Kanai, H. Tomura, N. Matsufuji, S. Minohara, A. Fukumara, T. Hiraoka, Y. Furusawa, N. Miyahara, H. Koyomah-Itoh, M. Endo, F. Soga, K. Kawachi. "HIMAC Beam Delivery System: Physical Characteristics," Proc. of the NIRS Int'l Seminar on the Appl. of Heavy Ion Accelerator to Radiation Therapy of Cancer in connection with XXI PTCOG Mtg., Japan, Nov. 14-16, 1994.
⁷M. Standford and J. A. Jones. "Space Radiation Concerns for Manned Exploration," Proc. of the 12th Man in Space Symp., Washington, D.C., June 23-27, 1997; Acta Astronomica. (Submitted for publication.)
⁸M. Porvaznik. "Tight Junction Disruption and Recovery After Sublethal Gamma Irradiation," Radiat. Res. 78 (1979): 233-50.
⁹R. I. Walker, G. D. Ledney, C. B. Galley. "Aseptic Endotoxemia in Radiation Injury and Graft-vs.-Host Disease," Radiat. Res. 62 (1975): 242-49.
¹⁰A. Haimovitz-Friedman, C.-C. Kan, D. Ehleiter, R. S. Persaud, M. McLoughlin, Z. Fuks, R. N. Kolesnick. "Ionizing Radiation Acts on Cellular Membranes to Generate Ceramide and Initiate Apoptosis," J. Exp. Med. 180 (1994): 525-35.
¹¹W. D. Jarvis, R. N. Lolesnick, F. A. Fornari, R. S. Traylor, D. A. Gewritz, S. A. Grant. "Induction of Apoptotic DNA Degradation and Cell Death by Activation of the Sphingomyelin Pathway," Proc. Nat'l Acad. Sci. USA 91 (1994): 73-77.
¹²A. Peralta Soler, J. M. Mullin, K. A. Knudsen, C. W. Marano "Tissue Remodeling During Tumor Necrosis Factor-Induced Apoptosis in LLC-PK1 Renal Epithelial Cells," Am. J. Physiol. 270 (1996): F869-79; Renal Fluid Electrolyte Physiol. 39.

Contents
ISSO -- Institute for Space Systems Operations
1997-1998 Annual Report