003 1-399819213203-0366$03.00/0 PEDIATRIC RESEARCH Vol. 32, No. 3, 1992 Copyright O 1992 International Pediatric Research Foundation, Inc. Printed in U.S.A. The Effect of Monosaturated and Polyunsaturated Fatty Acids on Oxygen Toxicity in Cultured Cells DOUGLAS R. SPITZ, MICHAEL T. KINTER, JAMES P. KEHRER, AND ROBERT J. ROBERTS Division of Neonatology, Departments of Pediatrics [D.R.S., R.J.R.] and Pathology [M. T.K.], University of Virginia, Charlottesville, Virginia 22908 and Division of Pharmacology and Toxicology, College of Pharmacy, University of Texas, Austin, Texas 78 705 [J.P.K,] ABSTRACT. The influence of oleic, linoleic (LIN), and Exposure to elevated 02concentrations is thought to give rise eicosapentaenoic (EPA) acids incorporated into cellular to the overproduction of reactive O2 species, such as superoxide, lipids on susceptibility to 02-induced toxicity was evaluated hydrogen peroxide, and hydroxyl radical, which have been hy- in Chinese hamster fibroblasts (HA1) using a clonogenic pothesized to contribute to the injury process associated with O2 cell survival assay. Fatty acid incorporation was achieved toxicity (1-4). Previous studies in animal models and in cell by incubating HA1 cells in 21% O2for 72 h in the presence culture systems have suggested that reactive 02species produced or absence of media supplemented with 25 pM oleic acid, during exposure to hyperoxia may interact with noncritical 25 pM LIN, or 2, 4, and 25 pM EPA. This fatty acid PUFA, sparing critical cellular constituents such as proteins, incorporation period increased the percentage of composi- DNA, and RNA from the damaging effects of interaction with tion in phospholipids 2-fold for oleic acid, 6-fold for LIN, reactive 02species (5-9). Based on this theoretical concept, these and 6- to 20-fold for EPA. Vitamin E, total glutathione, authors hypothesized that an increased amount of PUFA in cells superoxide dismutase activity, glutathione transferase ac- during exposure to hyperoxia might protect the cells from the tivity, and catalase activity were unchanged, relative to injury produced during exposure to hyperoxia. The present control, in the 25-pM EPA-treated group, and only total studies were designed to specifically address this hypothesis using glutathione was elevated in the LIN-treated group. After a previously characterized cell culture 02-toxicitymodel (10) and the incorporation period, the cells were placed in non-fatty specific polyunsaturated and monosaturated fatty acids. acid supplemented media and exposed to 95% 02,and In our studies, HA1 Chinese hamster fibroblasts were used clonogenic survival responses were evaluated at time inter- and clonogenic cell survival was measured as the 02-toxicity end vals up to 100 h. Sensitization to Oz toxicity in EPA- point. HA1 cells were incubated with media containing purified treated cells was apparent after 24 h of O2 exposure, oleic acid, LIN, or EPA to incorporate these specific fatty acids whereas LIN-treated cells were significantly (p < 0.05) into the cellular lipids. The specificity and extent of fatty acid sensitized to hyperoxia after 54 h of exposure, indicating incorporation was quantitated using gas chromatography. The that EPA was a more potent sensitizer for 02 injury. vitamin E and cellular antioxidant status of the cells with and Furthermore, cells supplemented with 4 and 25 pM EPA without PUFA (LIN, EPA) incorporation was evaluated before were more sensitive to O2toxicity than cells supplemented O2 exposure. Finally, the loss in cellular clonogenic potential as with 2 pM EPA. In contrast, cells treated with 25 pM a function of time in 95% 02 was evaluated in fatty acid oleic acid were significantly more resistant to Oz toxicity supplemented and unsupplemented cells. The objective of these at 51, 72, and 98 h of O2 exposure. The results indicate experiments was to evaluate the effect of specific fatty acid that incorporation of polyunsaturated fatty acids (LIN or alterations on O2 toxicity in the absence of other changes in EPA) into the lipids of cultured cells enhances the cytotox- cellular antioxidants. The long-term goal of these studies is to icity associated with exposure to 95% 02,whereas incor- develop a rationale for the protection of neonates from O2 poration of a monosaturated fatty acid (oleic acid) reduced toxicity using dietary manipulations. the cytotoxicity of 95% 02.These results suggest that Oz- induced injury can be modified by alterations in peroxidiz- MATERIALS AND METHODS able substrates present in cells. (Pediatr Res 32: 366-372, 1992) Cells and culture conditions. Chinese hamster fibroblasts des- ignated HA1 (10) were maintained in Eagle's minimal essential Abbreviations medium supplemented with 10% FCS (HyClone Laboratories, Logan, UT) and penicillin/streptomycin (100 U/mL and 0.1 LIN, linoleic acid mg/mL respectively). Cell cultures were maintained in 37°C EPA, eicosapentaenoic acid incubators and routinely checked for mycoplasma contamina- PUFA, polyunsaturated fatty acid tion as described previously (10). DBI, double bond index PUFA incorporation protocol and toxicity studies. Oleic acid (> 99% purity), LIN (> 99% purity), and EPA (> 85% purity), stored in N2-gassed ampules, were purchased from Nu Chek Prep Received October 18, 199 1; accepted April 30, 1992. (Elysian, MN) and used without further purification. To prepare Correspondence: Douglas R. Spitz, Division of Neonatology, Department of soluble sodium salts of the fatty acids, the contents of the ampules Pediatrics, University of Virginia, MR-4 Room 3033, Charlottesville, VA 22908. were dissolved in 2 mL of hexane and approximately 200 pL of Supported by NIH Grants HL42057, DK38942, HL.40695, and RR05431, and a grant from the University of Virginia Pratt Foundation. 6 N NaOH was added to the hexane solution, which became I Presented in part at the American Pediatric Society/Society for Pediatric opaque. The solution was vortexed and evaporated to dryness Research nleeting, New Orleans, LA, April 29-May 2, 199 1. under a stream of N2. Five mL of boiling distilled Hz0 was 366 THE EFFECTS OF FATTY ACIDS ON o2 TOXICITY 367 added to the residue and the mixture was vortexed until clear. petroleum ether. The extracts were combined in a 1-mL reaction The solution was sterile filtered and an appropriate amount was vial, evaporated to dryness, and reconstituted in 100 pL of added to minimal essential medium containing 10% FCS sup- isooctane containing an amount of methyl henieicosanoate for plemented with antibiotics. An aliquot of each fatty acid-en- determining recovery. riched medium was taken for gas chromatographic analysis to The samples were analyzed using a Hewlett-Packard 5890 gas verify the desired concentration of fatty acid and for vitamin E chromatograph equipped with a split/splitless injector, flame analysis before exposing the cells. ionization detector, a Hewlett-Packard 3392A integratorlre- Cells (0.75-2 x lo5) were plated into 60-mm Petri dishes corder, and a Hewlett-Packard Ultra-2 polymethylsilicone gas containing 4 mL of the control or fatty acid supplemented chromatography column (Hewlett-Packard Co., Palo Alto, CA). medium. The cells were grown for 3 d at 37°C incubation to One to 5 pL of sample were injected in the splitless mode with a incorporate the desired fatty acid into cellular lipids (1 I). In 30-s split delay. The oven was programmed from 75-200°C at selected experiments, 100-mm dishes were used to obtain enough 25"C/min, then 200-240°C at 2"C/min, and then 240-300°C at cells for all analyses. Because 100-mm dishes have three times 50°C/min. Peak areas were recorded by the integrator and quan- the surface area of 60-mm dishes, the amounts of media and titation was determined by the area ratio of the fatty acids of cells were increased by a factor of three when these dishes were interest to the standard fatty acids, correcting for the number of used. At the end of 3 d, the media were removed and the cells methylene units in each compound. were either analyzed for fatty acid content, vitamin E content, Vitamin E analysis. Cells were harvested using 0.1 % trypsin and cellular antioxidant status or fresh control medium was and were pelleted by centrifugation at 400 x g. The pellet was added before exposure to hyperoxia. resuspended in 2 mL of N2-gassed 50 mM phosphate buffer The control and fatty acid-enriched cells were exposed to 95% containing 150 mM KC1 and 0.1 % EDTA, pH 7.4, and 1 mL of O2and 5% C02at 37°C and clonogenic cell survival was assayed 100 mM SDS in distilled water was added and vortexed. Three at predetermined time intervals as described previously (10, 12). mL of 85% ethanol and 3 mL of heptane were added and Briefly, cells were trypsinized, counted, diluted, and plated into vortexed for 1 min. The mixture was then centrifuged at 1500 x replicate cloning dishes (200 to 200 000 cells/cloning dish). The g to separate the phases, and the heptane layer containing the cells were allowed to form colonies for 8 to 10 d, at which time vitamin E was removed and stored in N2-gassed tubes at -20°C. they were fixed, stained, and counted using a dissecting micro- When ready for analysis, the samples were dried under a stream scope. A colony was considered a survivor if it contained at least of N2 and the residue was dissolved in HPLC mobile phase 50 cells. Dilution replicates containing 50-250 colonies were (60.4% 2-propanol, 20% acetonitrile, 19% HPLC H20, 0.5% used for analysis. Cells that detached and floated off the dish tetraethylamonium hydroxide, 0.1 % acetic acid, pH 4.0). Vita- during 02treatment were considered reproductively inactivated min E content (dl-a-tocopherol) was analyzed on a ODs HPLC for the purpose of analysis (10).