I. Introduction

The supply of tap water in Japan depends largely on utilizing river water, and thus it is important for human health that good-quality river water is obtained. To evaluate water quality, the levels of at least 40 chemical compounds stipulated by the Water Works Law^［1］ are monitored during the process of tap water production. However, there are many more active compounds than this number in the environment generally and in river water in particular. Therefore, in addition to assessing the compounds present in the water, a method is required whereby the biological effects of compounds derived from river water components may be evaluated.
In the present study, we examined human RSa cells to assess the biological effects of environmental agents. The highly sensitive RSa cells were found to respond to lethal effects of ultraviolet radiation （mainly 254-nm wavelength, UVC）^［2］ , 4-nitroquinoline 1-oxide （4NQO）^［3］ , N-methyl-N’-nitro-N-nitrosoguanidine （MNNG）^［3］ , X-rays^［4,5］ , and cadmium^［6］ . RSa cells are also appropriate for studying the molecular mechanisms underlying the action of environmental agents because variants derived from RSa cells can be easily established from cells mutagenized by ethyl methanesulfonate （EMS）^［3,7］ and from those transfected with plasmid vectors containing objective genes^［8,9］ . Thus, RSa cells can be utilized for evaluating the cytotoxic effects of compounds present in water.

II. Materials and Methods

Cells and culture conditions

RSa cells were established from human embryo fibroblast cells doubly infected with Simian virus 40 and Rous sarcoma virus^［10］ ; they have been characterized as having high susceptibility to environmental agents^［2-6,11］ . Cells were cultured with Eagle’s minimal essential medium （Nissui, Tokyo, Japan） containing 10% （v/v） calf serum （Invitrogen, Carlsbad, CA, USA） at 37℃ in a humidified atmosphere containing 5% （v/v） CO2.

Collection of environmental water

Tap water was collected from the faucets of city houses in Tokyo and Chiba Prefecture; about 5L water was allowed to flow prior to collection. River water was mainly collected upstream from the area where the water is supplied to the purification plants at first-class rivers in the northern area of Honshu （Japan） from 2003 to 2010. Except for Tamagawa, Tonegawa, and Arakawa rivers, water was collected upstream, not downstream, of the purification plants ^{(Fig. 1)}. The river and tap water were collected in brown glass bottles, transported to our laboratory, and then stored at －20℃. Each water sample was allowed to thaw before tests were conducted.

Preparation of water samples

A wide spectrum of acidic, basic, and neutral organic compounds was extracted and concentrated from water collected using an Oasis HLB 3-cc extraction cartridge, according to the manufacturer’s instructions. Briefly, 250 ml water was loaded into the cartridge, which had been preconditioned with methanol and Milli-Q water （Millipore, Tokyo, Japan）; the cartridge was washed with 2.0 ml 5% methanol, and then compounds adsorbed to the cartridge were eluted with 2.0 ml methanol. The flow rate was maintained at 2 ml/min. Methanol was evaporated using a desiccator, and the dry residue was dissolved in 25 ul dimethyl sulfoxide. The solution was used as the water sample.

Cell viability test

Cells were treated with and without water samples, as follows. Cells were seeded in each well of 96-well plates （5×103 cells/well） and cultured for 48 h. The activity of mitochondrial succinic dehydrogenase was measured by incubating the cells for 4 h in the presence of MTT （0.5 mg/ml） followed by measurement of absorbance at 570 nm with a reference wavelength of 655 nm, according to the method of Mosmann^［12］. The viability of cells treated with Milli-Q water was taken to be 100% survival.

Statistical analysis

All experiments were repeated at least three times independently, and statistical analysis was performed using Student’s t-test with Microsoft Office software （Excel 2003; Microsoft, Redmond, WA, USA）.

III. Results

The viability of RSa cells treated with tap water samples was over 80% in Tokyo and six prefectures ^{(Fig. 2A)}. In addition, the viability was over 80% in Komazawa, Tokyo, and in Urayasu and Makuhari, Chiba Prefecture; however, the viability was under 80% in Hachioji, Tokyo, and in Kashiwa, Chiba Prefecture ^{(Fig. 2B)}. Susceptibility of RSa cells to the cytotoxic effects of bisphenol A was examined by MTT assay: Fifty % viability was obtained at about 400 uM, while up to 80% viability was obtained at about 250 uM ^{(Fig. 2C)}.

To estimate the safety of river water, the viability of RSa cells treated with the river water samples was tested. The viability after river water treatment was about 100% for upper sections of Tamagawa river, though around 80% for middle and lower sections of the river ^{(Fig. 3A)}. In similar fashion, Tonegawa and Arakawa samples showed 100-105% and 80-50% viability in upper and lower sections of the rivers, respectively (^{Figs. 3B} and ^C).

Further, river water collected every month from 2004 to 2005 from Ozaku and Kanamachi, along the middle sections of the Tamagawa and Edogawa rivers, respectively, was tested. Decreased cell viability of under 70% was found in April and July in Ozaku ^{(Fig. 4A)}. Decreased cell viability of about 80% was also detected in August and October in Kanamachi ^{(Fig. 4B)}.

Among the 7 first-class rivers in the Kanto and Koshinetsu regions （upstream of purification plants）, treatment of RSa cells with water samples from Arakawa river resulted in about 75% viability; there was no obvious decrease in viability among the other six rivers ^{(Fig. 5A)}. In the Tohoku region, 90-100% viability was detected in the three river samples; however, the viability was about 80% in samples from Yoneshirogawa, Mogamigawa, and Aganogawa rivers ^{(Fig. 5B)}. Among four first-class rivers in the Tokai, Hokuriku, and Chubu regions, samples from Nagaragawa river showed about 80% viability, while those from the other three rivers showed over 80% viability ^{(Fig. 5C)}.

IV. Discussion

In the present study, the viability of human RSa cells treated with tap water samples and water samples from 17 first-class rivers from eastern Japan was tested and found to be, respectively, about 80-100% (^{Fig. 2A} and ^B) and over 70% ^{(Fig. 5)}. Viability of 80-100% was detected in RSa cells treated with less than about 250 uM bisphenol A ^{(Fig. 2C)}. We previously reported the mutagenicity of bisphenol A to be less than 10 uM in RSa cells^［11］. Viability of 100% is required for tap water. However, the viability of human RSa cells treated with tap water samples was about 80-100%, or less than 80% in some cases. Therefore, it may be necessary to introduce methods that are able to purify water to a greater level than those currently adopted.
Treatment of cells with samples from five first-class rivers resulted in viability of about 80% or less ^{(Fig. 5)}. In addition, samples from upper sections of Tonegawa and Arakawa rivers and those from the upper and middle sections of Tamagawa river conferred 80-100% viability; however, samples from middle and lower sections of the Tonegawa and Arakawa rivers and from the lower sections of the Tamagawa river resulted in less than 80% viability ^{(Fig. 3)}. Many cities in the metropolitan Tokyo area are located around the middle and lower sections of these three rivers, notably around the lower sections of Tamagawa and Arakawa rivers. The lowering of water quality in these three rivers may be related to urban pollution, such as sewage.
After testing the viability every month for a year, it was found that the river water quality of Tamagawa and Edogawa rivers deteriorated in April and July ^{(Fig. 4A)} and also in August and October ^{(Fig. 4B)}. These months correspond to times of active farming activity around the Ozaku and Kanamachi sampling sites^［13,14］. Shioda et al. reported that toxic effects of river water on aquatic organisms were assessed and the samples collected of Tamagawa between March and August showed toxic effects^［14］. They found the contamination of some pesticides^［14］. And Ozaki et al. reported that biological activity and atmospheric temperature procured aggravation of water quality^［15］. So there may have been some negative agrochemical and agricultural effects on the water quality in these two rivers.
In Japan, the criteria for water quality are based on the Water Works Law. Many countries around the world refer to the guidelines for drinking-water quality established by the World Health Organization. In the United States, the Safe Drinking Water Act was legislated by the Environmental Protection Agency. In the European Union, the Drinking Water Directive was introduced to ensure safe drinking water^［1］. Even though many laws and directives have been established, there are differences among them. So the present study is constructive in helping to unify criteria with regard to water quality.

Acknowledgements

We thank Y. Miyamoto and H. Miyamoto, Chiba City Institute of Health and Environment. This work was supported in part by a grant from the Foundation of River & Watershed Environment Management and in part by grants-in-aid from the following organizations: the Miso Central Institute in Japan, the Smoking Research Foundation, the Tokyu Foundation for a Better Environment, the Tsuchiya Foundation, the Goho Life Science International Foundation, the Ministry of Health, Labour and Welfare for the Treatment of Intractable Diseases Research Program, and the Japan Society for the Promotion of Science.