Environmental Mutagenesis 5:795-801 (1983)

Evaluation of an Exposure System Using

Cells Grown on Collagen Gels for Detecting
Highly Volatile Mutagens in the CHO/
HGPRT Mutation Assay
P.O. Zamora, J.M. Benson, A.P. Li, and A.L. Brooks
Inhalation Toxicology Research Institute, Lovelace Biomedical and Environmental
Research Institute, Albuquerque, New Mexico

Chinese hamster ovary (CHO) cells were grown on hydrated collagen gels, the
overlaying medium removed leaving the cells at an airkollagen interface, and the
cells exposed to a dynamic flow of ethylene oxide. Increases in CHO cell mutant
frequency and decreases in cell viability were observed. To establish if the
exposure system could be simplified, cells were exposed in sealed bottles (static
system) to ethylene oxide. No substantial changes in cytotoxicity, mutant fre-
quency, or effective concentration were noted when comparing static versus
dynamic exposure systems. The general usefulness of the exposure system using
cells grown on collagen gels was evaluated in a static system using propylene
oxide and 1,2dichloroethane, both of which were found to be mutagenic and
cytotoxic. Comparatively, the exposure of cells by the collagen gel method was as
effective in detecting genotoxic damage as were conventional methods (cells
covered with medium) using cells grown on glass substrates. The exposure of
CHO cells on collagen gels to highly volatile mutagens was simple and inexpen-
sive, and may be generally useful in the detection of gaseous or volatile mutagens.

Key words: mutation, collagen gels, CHO cells, ethylene oxide, propylene oxide, 1,2-dichloroethane

Received December 4, 1982; revised and accepted July 11, 1983.

P.O. Zamora’s current address is Summa Medical Corporation, 4272 Balloon Park Road NE, Albuquer-
que, NM 87109.

A.P. Li’s current address is Environmental Health Laboratory, Mail Stop EHL, Monsanto Company,
800 North Lindberg Blvd, St Louis, MO 63166.

Abbreviations: CHO cells, Chinese hamster ovary cells, clone KI-BH4; HGPRT, hypoxanthine-guanine
phosphoribosyl transferase.

Address reprint requests to Dr. A.L. Brooks, Inhalation Toxicology Research Institute, Lovelace
Biomedical and Environmental Research Institute, P.O. Box 5890, Albuquerque, NM 87185.

0 1983 Alan R. Liss, Inc.

7% Zamoraetal
INTRODUCTION
Methods to determine the genotoxic effects of gaseous and vapor phase chemi-
cals in mammalian cells would clearly be useful in predicting the toxicity of such
compounds. A number of exposure regimens have been described to evaluate the
toxic effects of volatile or vapor phase chemicals including use of rocking platforms,
roller bottles, sealed culture vessels, and supports composed of cellulose nitrate or
gas-permeable membranes [Guerrero and Rounds, 1982; Hatch et al, 1982; Krahn et
al, 1982; Rasmussen and Crocker, 1982; Robinson, 1982, Bolton et al, 19821. The
variety of exposure methods reflects some of the difficulties related to chemical
volatility, solubility, and reactivity associated with exposing cells to gases or vapors.
We have described a method to expose lung epithelial cells grown on collagen
gels directly to nitrogen dioxide [Zamora et al, 19831 and the subsequently analyze
the survival of the cells by Trypan blue dye exclusion and cloning efficiency on feeder
layers. In the study reported here, the exposure method was applied to the CHO/
HGPRT mutation assay. CHO cells were grown on collagen gels and exposed at an
aidcollagen interface to three test agents: ethylene oxide, propylene oxide, and 1,2-
dichloroethane. These Chemicals are widely used in industrial settings (production of
these chemicals in the United States in 1981 was greated than 10l2 kg) and may pose
a human health risk [IARC, 1976a,b, 1979; Tan et al, 1981; Hatch et al, 19821. The
highly volatile nature of the chemicals could make their evaluation in conventional
somatic cell mutation assays problematic. Our approach was to test if collagen gel
could be used to grow CHO cells for detecting volatile mutagens in the CHO/HGPRT
assay. We observed an increase in mutant frequencies in CHO cells after exposure to
each of the three chemicals.

MATERIALS AND METHODS

The cell line CHO-KI-B& was obtained from Dr.A.W. Hsie (Oak Ridge
National Laboratory) and maintained in growth medium composed of Ham’s F-12
medium (Flow Laboratory) containing 10% newborn calf serum (K.C. Biological),
penicillin (100 U/ml, and streptomycin (100 pglml) (K.C. Biological).
Propylene oxide (CAS registry No. 75-56-9) and 1,2-dichloroethane (CAS
registry No. 107-06-02) were obtained from Kodak Chemical Company and used as
supplied. Ethylene oxide (CAS registry No. 75-21-8) was obtained from Matheson
Chemical Company.
For dynamic exposures to ethylene oxide, 2 X lo6 CHO cells were plated in
100-mm diameter Petri dishes 24 hr before use on collagen gels. The collagen gels
were prepared from rat tail tendon as described by Emerman and Pitelka [1977] and
modified by Zamora et a1 [ 19831. Just prior to exposure of the cells to test materials,
the culture medium was aspirated leaving the cells attached on the collagen gel at an
air/collagen interface. The cultures were then placed in an incubation chamber (Flow
Laboratories). A dynamic flow of ethylene oxide in 95 % air /5 % COf was introduced
into the chamber at a flow rate of 2 litedmin resulting in approximately 30 chamber
air changedhr. Heat tape was used on the outside of the chamber to adjust the internal
temperature to 37°C. At the flow rates used in these experiments, equilibrium was
reached within 15 min of initiating the experiment. The concentration of ethylene
oxide was continually monitored during the exposure by passing the chamber exhaust
through a Miran 80 infrared spectrophotometer (analytical wavelength 11.815 pin;
 Volatile Chemicals and CHOIHGPRT Assay 797

I N V E R T BOTTLE
B

Fig. 1. Schematic of the method of exposing CHO cells on collagen gels to volatile mutagens using the
static system. A) Cells were grown on collagen gels in glass bottles; B) the medium was aspirated and
the bottles inverted; and C) test substance was introduced by syringe into the bottle and the chemical
was volatilized (if liquid) by briefly heating the bottle.

reference wavelength 3.5 pm). The exposure system is similar to that described by
Zamora et a1 [ 19831 and Benson et a1 [ 19831.
After the I-hr incubation the cells were recovered by use of a trypsin-EDTA
solution (1 :250 trypsin, K.C. Biological) and collected by centrifugation. The cells
were suspended in calcium-magnesium-free Puck’s saline G [Puck et al, 19581, the
cell number determined, and aliquots taken to determine relative cell survival and
mutation.
Cell cultures for static exposures were prepared by plating 2 x lo6 CHO cells
in growth medium on hydrated collagen gels in glass milk dilution bottles (250 ml
capacity) 24 hr prior to use. The vessel was inverted and corked with a Teflon-lined
silicone rubber stopper (Fig. 1). Just prior to exposure of the cells, the medium was
aspirated leaving attached cells at an airkollagen interface. To minimize chemical
loss by volatilization during transfer to the culture vessel, propylene oxide or 1,2-
dichloroethane was cooled to 4°C and injected into the vessel with a microliter
syringe precooled to 4°C. The surface of the culture bottle was briefly heated to
55°C to volatilize the test chemical. Ethylene oxide was injected into the bottles with
a gas-tight syringe. The cultures were then incubated at 37°C for 1 hr. The cells were
recovered from the cultures as described above.
For comparative purposes with the collagen gel method, cells were grown
directly on glass in milk dilution bottles. The cells were exposed to propylene oxide
in graded volumes of serum-free Hams F-12 medium (0,3.5, 7.0, and 10-0 mi).
Propylene oxide was injected into the bottle. During the l-hr exposure period the
 798 Zamora et al

culture vessels were tightly sealed with Teflon-lined rubber stoppers to avoid ex-
change between the incubator atmosphere and the medium.
Relative cell survival (colony-forming ability) was determined as described by
Li and Brooks [ 19811. The mutation assay was performed as described by Li [ 19821
for detection of mutations at the HGPRT gene locus. The total expression time was 8
days during which the cells were cultured on 75 cm2 flasks for 3 days then subcultured
as unattached cell suspensions in 100-mmdiameter bacteriological grade Petri dishes.
Selection for HGPRT mutants was performed on a population of lo6 cells by plating
2 X lo5 cells into each of five Petri dishes (100 mm diameter). Each selection dish
had 8 ml of hypoxanthine-free Ham’s F-12 medium containing 5% dialyzed newborn
calf serum and 10 pm 6-thioguanine. After an incubation period of 8 days, the
colonies were fixed in methanol, stained with giemsa, and counted. The mutant
frequency was determined by dividing the total number of colonies in the five plates
for mutant selection by the plating efficiency of the cells at the time of selection.
Mutant frequency was expressed as mutants/ lo6 clonable cells.

RESULTS AND DISCUSSION

In a previous study, we described a method to expose lung epithelial cells (cell
line designated LEC [Li et al, 19821) to nitrogen dioxide [Zamora et al, 19831. LEC

a
>
-
>

w
>
-
t
a
-1
W
a

- 0.1 L 0 I I I
0 8 16 20 10
A ETHYLENE OXIDE (pg/cm3) B ETHYLENE OXIDE (pglcrn’)

Fig. 2. Relative survival (.--=) and mutani frequency (.--a) of Chinese hamster ovary cells
grown on collagen gels and exposed to increasing concentrations of ethylene oxide. A) Dynaniic
exposure system; B) static exposure system. Vertical bars indicate the standard error of the means of
duplicate experiments.
 Volatile Chemicals and CHO/HGPRT Assay 799

cells were grown on hydrated collagen gels, the overlying medium removed leaving
the cells at an airkollagen interface, the cells exposed to a dynamic flow of nitrogen
dioxide, and changes in cell morphology and viability monitored. In the study
reported here, CHO cells were exposed in a similar manner to a dynamic flow of
ethylene oxide, a gas we found to be cytotoxic and mutagenic (Fig. 2A). This
confirms the report of Tan et a1 [1981] on the mutagenic activity of ethylene oxide in
CHO cells. These results also indicate that mutations induced by direct exposure to
toxic gases can be detected in the CHOIHGPRT mutation assay. To establish if the
exposure system could be further simplified, CHO cells on collagen gels were
exposed to ethylene oxide in sealed glass bottles (static system). As in the dynamic
system, ethylene oxide was found to be both cytotoxic and mutagenic (Fig. 2B), and
the effective concentrations needed to elicit biological responses were similar. Since
the use of a static system was easier and less expensive to operate, the general
usefulness of the static exposure system was tested with two other mutagenic agents-
propylene oxide and 1,2-dichIoroethane. Propylene oxide was found to be cytotoxic
and mutagenic (Fig. 3), and 1,2-dichloroethane was found to be weakly mutagenic
(Fig. 4). The increases in mutant frequency in CHO cells exposed to either ethylene
oxide, propylene oxide, or 1,2-dichloroethanewere consistent with the known muta-
genicity and carcinogenicity of these compounds [IARC, 1976a,b, 1979; Tan et al,
1981; Ehrenberg and Hussain, 1982; Rannug, 19801. To establish if the exposure of
CHO cells on collagen gels was more useful than conventional routes of exposure, a

120

u)

5 0 4

;0 2
3
5

0.1
A '
*2
= 16 32 40'
3 PROPYLENE O X I D E ( p g l c m 3 ) * 1. 2 DICHLOROETHANE ( p g / c r n J )

Fig. 3. Relative survival (W-W) and mutant frequency (.---a) of Chinese hamster ovary cells
grown on collagen gels and exposed to propylene oxide in a static exposure system. Vertical bars indicate
the standard error of the means of duplicate experiments.

Fig. 4. Relative survival (H-W) and mutant frequency (.--a) of Chinese hamster ovary cells
grown on collagen gels and exposed to 1,2-dichloroethane in a static exposure system. Vertical bars
indicate the standard error of the means of duplicate experiments.
800 Zamora et al
TABLE I. Comparison of Relative Survival and Mutant Frequency of CHO Cells Exposed to
~ ) the Collagen Gel Method or by Conventional Exposure Methods
Propylene Oxide (32 ~ g l c r n by
on Glass Substrates
Mutants/ 10'
Substrate Medium (ml) Survival clonable cells f SE
Collagen gel 0 0.30 88.0 f 8.7a
Glass 0 0 N D ~
Glass 3.5 0 ND
Glass 7.0 0.22 85.2 f 23.0
Glass 10.0 0.48 64.9 -f 20.5'
aBackground mutant frequency on collagen gels was 6.3 f 2. I (SE) mutants/106 surviving cells.
%Jot detectable owing to excessive cytotoxicity.
'Background mutant frequency on glass substrate with 10 ml of medium was 7.5 f 2.4 (SE) mutants/
lo6 surviving cells.

comparative study was done with a selected dose of propylene oxide (32 pg/cm3)
(Table I). In this experiment, the effect of propylene oxide on CHO cells grown on
collagen was compared to those effects observed when CHO cells were grown on
glass with varying amounts of medium covering the cells during exposure. When no
serum-free medium or 3.5 rnl medium covered the cells, excessive cytotoxicity
occurred which precluded assessment of mutation. This cytotoxicity most likely
resulted from dehydration of the cells. Mutant frequency occurring in those cultures
which had been grown on glass and covered with 7 or 10 ml of medium was similar
to that observed in cells exposed on collagen gel. The slightly higher mutation
frequency observed in cells covered with 7 ml of medium over those covered by 10
ml suggests that the concentration of propylene oxide dissolved in the medium was
higher in cells covered by 7 ml of medium. When similar studies are performed with
other volatile materials, consideration must be made of the solubility of the compound
in aqueous medium, the reactivity of the compound with the media, and the need to
optimize the volume of culture medium used. These factors need not be considered
when cells are grown on collagen gel and exposed directly to volatile materials.

CONCLUSIONS
The results of this study indicate that CHO cells grown on collagen gels can
readily be exposed to gases or vapors of volatile compounds, and that mutations can
be detected in the CHO/HGPRT mutation assay. The positive responses to ethylene
oxide, propylene oxide, and 1,2-dichloroethane indicate that the exposure of cells at
an air/collagen interface is a suitable procedure for testing diverse volatile chemicals.
The exposure system should be adaptable to other mammalian cell lines and may be
particularly useful using cells with endogenous metabolic capabilities allowing evalu-
ation of chemicals by a variety of assays including mutation, cell transformation, and
unscheduled DNA synthesis.

ACKNOWLEDGMENTS
P.O. Zamora was an Associated Western Universities Postdoctoral Participant
at the Inhalation Toxicology Research Institute. The authors thank Drs. S.G. Shami,
 Volatile Chemicals and CHO/HGPRT Assay 801
J.D. Sun, and J.A. Bond for reviewing this manuscript, and Ms A.S. Barner for
technical assistance.
Research performed under U.S. Department of Energy contract DE-AC04-76-
EV01013.

REFERENCES
Benson JM, Zamora PO, Dahl AR, Hanson RL (1983): A simple dynamic flow through system for
assessment of biological activity of volatile complex mixtures in in vitro test systems. Am Ind
Hyg ASSOCJ 44:211-215.
Bolton DC, Tarkington BK, Zee YC, Osebold JW (1982): An in vitro system for studying the effects of
ozone or mammalian cell cultures and viruses. Environ Res 27:466475.
Ehrenberg L, Hussain S (1981): Genetic toxicology of some important epoxides. Mutat Res 86:l-113.
Emerman JT, Pitelka DR (1977): Maintenance and induction of morphological differentiation in disso-
ciated mammary epithelium on floating collagen gels. In Vitro 13:316-328.
Guerrero RR, Rounds DE (1982): In vitro analysis of mammalian cells exposed in vitro and in vivo to
airborne agents. In Tice RR, Costa DL, Shaich KM (eds): “Genotoxic Effects of Airborne
Agents.” New York: Plenum, pp 51-73.
Hatch GG, Marmay PD, Ayer ML, Castro BC, Nesnow S (1982): Methods for detecting gaseous and
volatile carcinogens using cell transformation agents. In Tice RR, Costa DL, Shaich KM (eds):
“Genotoxic Effects of Airborne Agents.” New York: Plenum, pp 75-90.
IARC (1976a): Monographs on the evaluation of the carcinogenic risk of chemicals to man: Cadmium.
nickel, some epoxides, miscellaneous industrial chemicals and general considerations on volatile
anesthetics. Vol 11, International Agency for Research on Cancer, Lyon, France, pp 157-167.
IARC (1976b): Monographs on the evaluation of the carcinogenic risk of chemicals to man: Cadmium.
nickel, some epoxides, miscellaneous industrial chemicals and general considerations on volatile
anesthetics. Vol 11, International Agency for Research on Cancer, Lyon, France, pp 191-199.
IARC (1979): Monographs on the evaluation of the carcinogenic risk of chemicals to man: Some
halogenated hydrocarbons. Vol20, International Agency for Research on Cancer, Lyon, France,
pp 429-448.
Krahn DF, Barsky FC, McCooey KT (1982): CHOIHGPRT mutation assay: Evaluation of gases and
volatile liquids. In Tice RR, Costa DL, Schaich KM (eds): “Genotoxic Effects of Airborne
Agents.” New York: Plenum, pp 91-103.
Li AP (1982): Simplification of the CHOIHGPRT mutation assay through the growth of Chinese hamster
ovary cells as unattached cultures. Mutat Res 85: 165-175.
Li AP, Brooks AL (1981): Use of Chinese hamster ovary cells in the evaluation of potential hazards
from energy effluents-Application to diesel exhaust emissions. In Lewis M (ed): “Proceedings
of the International Symposium of Health Impacts of Different Sources of Energy.” Vienna:
International Atomic Energy Agency Press, pp 350-355.
Li AP, Hahn FF, Zamora PO, Shimizu RW. Henderson RF, Brooks AL, Richards R (1982): Character-
ization of a lung epithelial cell strain with potential application in toxicological studies. Toxicology
(in press).
Puck TT, Cierciura SJ, Robinson A (1958): Genetics of somatic mammalian cells. 111. Long-term
cultivation of euploid cells from human and animal subjects. J Exp Med 108:945-956.
Rannug U (1980): Genotoxic effects of 1,2-dibromoethane and I ,2-dichloroethane. Mutat Res 76:269-
295.
Rasmussen RE, Crocker TT (1982): Lung cells grown on cellulose membrane filters as an in vitro model
of the respiratory epithelium. In Tice RR, Costa DL, Schaich KM (eds): “Genotoxic Effects of
Airborne Agents.” New York: Plenum, pp 105-120.
Robinson AV (1982): Effect of in vitro exposure to hydrogen sulfide on rabbit alveolar macrophages
cultured on gas-permeable membranes. Environ Res 27:491-500.
Tan E-L, Cumming RB, Hsie AW (1981): Mutagenicity and cytotoxicity of ethylene oxide in the CHOl
HGPRT system. Environ Mutagen 3:683-686.
Zamora PO, Gregory RE, Brooks AL (1983): An in vitro model for the exposure of lung alveolar
epithelial cells to toxic gases. J Environ Pathol Toxicol Oncol (in press).