Effect of 60Co gamma-irradiation on the nonspecific cytotoxicity of alveolar macrophages in vitro.

This paper reports on the effect of radiation on the nonspecific cytotoxicity of rat alveolar macrophages (AM). AM (effector cells) of bacille Calmette-Guerin-activated Wistar rats were irradiated with 60Co gamma rays in vitro to give doses of 0, 100, 300, and 500 Gy. Three hours after irradiation, the AM were cultured with human lung adenocarcinoma AGZY83-a and HeLa target cells in 125I-deoxyuridine-containing media for 6 hr and the cytotoxicity indexes determined. The results indicated that the cytotoxicity indexes of AM against human lung adenocarcinoma cells and HeLa cells were 94.3 +/- 0.3% and 81.3 +/- 1.9%, respectively. The cytotoxicity indexes using an effector/target cell ratio (E/T) of 10 seemed to be greater than with ratios of 20 and 30. The cytotoxicity indexes of AM (7 rats), irradiated to give doses of 0, 100, 300, and 500 Gy, against adenocarcinoma cells at an E/T ratio of 10 were 87.9 +/- 8.4%, 65.4 +/- 14.1%, 47.5 +/- 17.5%, and 36.7 +/- 9.7%, respectively. The significance of the nonspecific cytotoxicity of AM in the immunological elimination of tumors and the inhibitory effect of radiation on AM cytotoxicity are discussed.


Introduction
Alveolar macrophages (AM) are mononuclear phagocytes that reside in the lungs. They are constantly exposed to inhaled pathogenic and/or carcinogenic particulate substances from the environment capable of affecting pulmonary functions.
It has been shown that the growth of lung tumor cells can be inhibited by activated AM both in vivo and in vitro. Thus, the in vitro cytotoxicity index ofactivated AM might reflect to a certain extent the in vivo tumoricidal property of AM and, if tumorbearing animals were treated with agents toxic to macrophages (silica, carrageenin, etc.), they would rapidly die of pulmonary metastasis (1)(2)(3).
Evidence suggests that AM play a very important role in host defense against neoplasms and that they constitute an important defense mechanism against particles that reach distal airways. As with other particulate materials deposited in the alveoli, inhaled radioactive particles are readily phagocytized by AM, thus giving a high radiation dose to these cells. AM death has been observed as early as 1 hr after inhalation of 239PuO2 particles (4).
It is expected that of the various cells types present in the lung, AM will be injured most seriously after inhalation of large amounts of radionuclides, and their immunological functions will be suppressed.
However, the inhibitory effect of immunological functions of AM after inhalation of radioactive particles has rarely been studied. As the radiation dose to AM depends on many biological parameters (clearance rate of inhaled particles, changes of renewal rate of AM under different conditions, etc.), the irradiated model of AM for the study of inhaled radioactive particles must be complex. In this study, an in vitro model of effects of radiation on AM was used. The aim ofthis work was to study the inhibitory effect of6Co y rays on the nonspecific cyotoxicity of AM and to find protective agents against the inhibition.
Wistar rats were activated by BCG. Two doses of 3.7 mg/ 200 g body weight were injected via the caudal vein with an in-terval of7 days. The rats were exsanguinated 2 days after the last injection and their lungs removed and lavaged with0.02% EDTA in phosphate-buffered saline (PBS) The cells in the pooled lavage fluids from each rat were counted with a hemocytometer in duplicate. Viability was determined by the trypan blue exclusion method. Of the cells recovered, 95% were alveolar macrophages. The AM were suspended in 0.02% EDTA-PBS at a concentration of 106/mL in a 30-mL flask.
The flask containing the suspension of AM harvested from BCG-activated Wistar rats was irradiated at a distance of0.5 m from a gamma source of2.6 x iO'5 Bq 0Co, giving a dose rate of 54-58 Gy/min at 15-25°C. In this experiment, the AM were divided into four groups, receiving absorbed doses of0, 100, 300, and 500 Gy, repectively.
After irradiation in vitro, the suspensions of AM (effector cells, E) were centrifuged at 1000 rpm forS min, then the supernatants were removed and the cells were resuspended at different concentrations in RPMI-1640 complete medium. After preculture at 37°C in 5% C02 for either 3 hr or 26 hr, a given number of AM was added to each well of the culture plate, together with human lung adenocarcinoma AGZY83-a cells or HeLa cells (target density: 500 cells/mm2) according to a fixed E/T (effector/target) ratio of 10, 20, or 30. Samples were incubated under the same conditions for 22 hr and labeled with 1251-UdR (7.4 x 103 Bq/sample) for 6 hr. The supernatants were then aspirated, and the adherent cells were washed twice with Hank's balanced salt solution. The remaining adherent tumor cells and AM were removed and the '25I-UdR activity incorporated into target cells was determined with a -y counter. A target cell sample processed in the same way but without AM served as the control. The results were expressed as the cytotoxicity index (CI%) calculated according to the following equation: where A is the 1251 activity oftarget cells and B is the 125I activity of target cells incubated with AM.

Experimental Model of AM in Vitro
To determine the nonspecific cyotoxicity of AM in vitro, several preliminary experiments were carried out.
Selection of the Target Cell. A human gastric carcinoma tumor cell line, a Hela cell line, and a human lung adenocarcinoma cell line, AGZY83-a, were labeled with'l2-UdR for 6 hr; the percentages of 125I-UdR incorportated in these tumor cells were 1.9, 3.7, and 5.3% ,respectively. The human lung adenocarcinoma cell line had the highest incorporated activity ofthe three cell lines tested.
Selection of Target Density. Human lung adenocarcinoma cells at densities of 250, 500, 750, and 1000 cells/mm2 were labeled with '25I-UdR for 6 hr. The resulting specific activities were 0.051, 0.058, 0.050, and 0.051 cpm/cell, respectively. No significant difference was found within the range of densities tested. The taet density of 50 cells/mm2 was selected to determine CI% because it mininized the number of tumor cells required while the total activity per sample was high enough to ensure the accurate measurement of incorportation.
EffectofETRaio. The close contact required between effector and target cells could be achieved either by increasing the number of both types of cell or by raising the E/T ratio. The nonspecific qtotoxicity ofBCG-activated AM was determined at E/T ratios of0.1, 1.0, 5.0, 10, 20, and 50. The changes of CI% after 3 hr preculture are shown in Figure 1. The CI% of BCGactivated AM increased with increasing E/T ratio in the range of 0.1%20 and with E/T ratios in the range 20-50. All CI values exceeded 90%.
Sekction ofPreculture Thne. The nonspecific cytotoxicity of BCG-activated AM could be influenced by preculture time in vitro. The values in Table 1 show that the CI% decreased significantly at all E/T ratios tested when preculture times were prolonged from 3 hr to 26 hr. With an E/T ratio of0.1, AM were no longer cytotoxic to the tumor cells after 26 hr of preculture. This effect could be attributed to a decrease ofactive effectors for killing tumor cells after prolonged preculture time.
Taking allthese results into consideration, we chose the human lung adenocarcinoma AGZY83-a and the HeLa cell lines as the target cells, (target density 500 cells/mm2 and precultured for 3 hr) for the in vitro model used in the following experiment.  Effect of "'Co "y Rays on Nonspecific Cytotoxicity of BCG-Activated AM in Vitro AM, activated by BCG, were very effective in killing both AGZY83-a cells and HeLa cells. The results of cytotoxicity determinations with doses of 100-500 Gy, using irradiated AM cultured at an E/T ratio of 20 after 3 hr precultitring ( Table 2), indicate that the CI ofirradiated AM decreased with increasing radiation dose. The inhibition was more pronounced in adenocarcinoma cells than in HeLa cells.
Additional studies of the inhibitory effect of radiation on the CI% ofBCG-activated AM against the AGZY83-a cell line were made with E/T ratios of 10, 20, and 30. The results are listed in Table 3, showing a more severe effect at an E/T ratio of 10 than at 20 and 30.
To confirm the relationship between radiation and inhibition of CI% of BCG-activated AM, repeated determinations of the dose-response relationship were made on seven rats at an E/T ratio of 10, using AGZY83-a cells as the target. The results shown in Figure 2 were obtained. The CI% ofAM was about90% in the control group, and consistently decreased as the radiation dose increased from 100 to 500 Gy. With a dose of 500 Gy, a CI of Tible 3. Effect of "Co y rays on the cytotoicity index of BCG-activated alveolar macrophages at different E/T ratios (means ± SD).
Cytotoxicity  only42% ofthe normal valuewas found. Thus, itis beyond doubt that the radiation-induced inhibitionof CI% is dose dependent.
Possible Protective Effect of Selenium on Ra-

diation-Induced Inhibition of Cytotoxicity Index
The manifestation of oDtoxicity ofAM depends on cell membrane integrity. It is well known that selenium, as an important componentofglutathioneperoxidase (GSH-px), canpromotethe primary function ofGSH-px in vivo. Ithas also been shown that the unsaturated fatty acid or lipid components ofcell membranes form peroxides when irradiated in an aqueous system (S). Thus. Thus, the addition ofselenium to irradiated AM should protect the cell membrane from peroxidation, thereby maininin its integrity. To investigate this hypothesis, we carried out the following experiment.
One hour before irradiation, selenium (sodium selenite) at a concentration of 10 1sM was added to the control and irradiated AM cultures. Two hours after irradiation, the selenium was removed by aspiration and washing with Hank's balanced salt solution. Then the AM were suspended in RPMI-1640 complete medium, andthe CI% was determined. From Table 4 it can be seen that selenium has a protective effect on the inhibition ofCI % of irradiated AM as the CI increased from 14.8 to 43.3%. It appears that there is a definite correlation between cytotoxicity and integrity ofmembranes, and the intact membrane might be quite an important effector for the manifestation of nonspecific cytotoxicity of AM.

Discussion
AM are mononuclear phagocytes that locate anatomically at the interface between tissue and air. This kind ofphagocyte forms the first line ofdefense inthe respiratory system. AM inactivate toxic particles and play an important role in the anticancer process. The results described indicate that BCG-activated AM can kill HeLa cells and human lung adenocarcinoma cells. We also observed tumoricidal ability of AM, activated by BCG, when melanoma B-16 cells were used as a target. These results show that activated AM have the ability to kill or inhibit various kinds of tumor cells. It is an essential function of AM to defend the organism against tumors. The inhibition ofCI% was dependent on radiation dose to AM and on the E/T ratio. To illustrate the mechanism of radiation effect, the viability of AM was determined by the trypn blue exclusion method. This showed that 87% of the AM were viable in the control and 84-89% in the three irradiated groups. There was no difference in mortality of AM after irradiation. It seems that the inhibitory effect on CI was not caused by direct killing. It has been shown thatcytoplasmic and lysosomal membranes of AM were damaged following irradiation with 100-500 Gy y rays (6), which might explain the decrease of CI% in irradiated AM. Table& 4 Protive effect ofsdenium against radition-induced hibition of cytotoxity index of irrdiated alveolar mcrophages (mean ± SD). The preliminary results of the observation of a protective effect ofselenium on radiation-induced inhibition ofCI% at a dose of 300 Gy were of significance in illustrating the dependence of AM cytotoxicity on cell membrane integrity. It will be necessary to carry out further experiments to confirm this observation.