Production of oxygen free radicals by Ehrlich ascites tumour cells: effect of lipids

Phorbol-12-myristate-13-acetate (PMA), calcium ionophore A23187 and platelet activating factor (PAF) stimulated the generation of oxygen free radicals (nitro-blue tetrazolium reduction) in Ehrlich ascites tumour (EAT) cells. PAF was effective at an optimal concentration of 4 μM, but was inhibited by BN 52021, a specific PAF antagonist. Lyso-PAF was ineffective. Inclusion of different lipids during incubation prior to the addition of PAF, resulted in the activation/inhibition of free radical generation. Among the phospholipids at a concentration of 50 μg/ml, the order of activation was phosphatidylserine > phosphatidylglycerol > phosphoinositides > phosphatidylinositol > phosphatidylethanolamine. Phosphatidylcholine was not effective, while sphingolipids were inhibitory. In addition, Ehrlich ascites tumour cells grown in mice under marginal vitamin A deficiency, showed an augmented production of free radicals compared to control cells. This was suppressed by exogenous addition of vitamin A or superoxide dismutase. These results suggest that membrane lipids and dietary factors like vitamin A probably function as physiological modulators in regulating the free radical generation.


Introduction
Phagocytic cells kill invading microorganisms in a metabolic event characterized by a marked increase in oxygen consumption termed 'respiratory burst'. The membrane bound enzyme NADPHoxidase catalyzes this reaction. 2 This enzyme is dormant in resting cells, but can be activated by a wide variety of stimulants. 2'3 During the activation of NADPH-oxidase, oxygen free radicals (like the superoxide anion O-) are generated. 1'2 Naturally occurring antioxidants like vitamin E, vitamin C and vitamin A can inhibit free radical generation. 4,s In addition to these vitamins, membrane lipids are also known to modulate free radical generation. As vitamin A has been known to affect membrane integrity and to impair immunity, there may be a relationship between membrane lipids, vitamin A and free radicals.
In order to test this relationship, the free radical generation in Ehrlich ascites tumour cells (EAT cells) using stimulants like phorbol-12-myristate-13acetate (PMA) calcium ionophore A23187, and PAF in cells grown in vitamin A deficient and vitamin A sufficient animals has been studied. The effect of various lipids and stimulants on free radical generation in this cell line is reported in this paper.
Determination of oxygen free radicals by NBT reduction The release of oxygen free radicals was determined spectrophotometrically by measuring the reduction of NBT at 540 nm. 1 Freshly harvested EAT cells (4 106 cells) were suspended in Tyrode-Ringer's buffer (pH 7.5) in a total volume of 1 ml, containing 0.25% BSA. The cells, untreated or stimulated by various stimulants, were incubated with NBT (60 nmol) for 20 rain. The cells were washed with the same buffer and lysed by the addition of 2 ml of 1,4-dioxane and maintained in a boiling water bath for 8 min. The lysate was centrifuged at 250 x g, for 5 min and the extracted blue colour was read at 540 nm. A calibration curve of absorbance at 540 nm was obtained using PAF (4 #M), PMA (2 #M) and calcium ionophore (2 #M) as stimulants, and different concentrations of NBT (0-80 nmol) on EAT cells (4 x 106).

Results
Production of vitamin A dejSciency in mice: Although the literature with respect to vitamin A deficiency in rat is extensive, reports on vitamin A deficiency in mice are scanty. 12 In fact, it was diflqcult to develop vitamin A deficient mice. Even though we were able to develop severe vitamin A deficiency in mice, the animals could not survive when EAT cells were injected intraperitoneally. Hence, a marginal vitamin A deficient condition was chosen in this study. It took 10-12 weeks to develop such a deficiency.
Stimulation offree radicals in EA T cells by PAF and its inhibition by BN 52021" A dose response curve showing the effect of increasing PAF concentrations on free radical generation in EAT cells grown in vitamin A deficient, pair-fed controls and in the control mice receiving the commercial diet is shown in Fig. 1. Although the optimal level of PAF required to stimulate the free radical generation is 4 #M in all three groups, the basal level of free radicals is more in vitamin A deficient mice stimulation of free radical generation. PS was the most effective of the phospholipids used (p < 0.001). However, PC at the lower concentration (12.5 #g/ml) was inhibitory (p < 0.001), while at a higher concentration (50/g/ml) the stimulatory effect was not significant (p > 0.09). Sphingolipids when the phospholipid fraction (PL) derived from a total lipid extract of EAT cells was used. The free radical generation was increased 2.8 fold at a concentration of 12.5 #g/ml (p < 0.001). On the other hand, glycolipid (GL) and neutral lipid (NL) fractions derived from EAT cells were ineffective in free radical generation in PAF stimulated cells (p > 0.05). The effect of these lipids on unstimulated cells was examined, but they did not affect the basal level of free radicals (data not shown). Comparison of the free radicalgeneration by dierent stimulants: The effect of various stimulants on free radical generation is shown in Table 1. PMA appears to be the most effective activator of the respiratory burst, while PAF was the least. Exogenous addition of vitamin A and SOD suppress the enhanced respiratory burst, and. also affect the basal levels of free radicals. However, lyso-PAF (the biologically inactive metabolite of PAF) totally failed to cause respiratory burst at both the concentrations used.

Discussion
PAF has been reported to activate the respiratory burst in various cells such as macrophages, neutrophils and eosinophils. 14 In contrast, certain investigators could not detect the effect of PAF on free radical generation in human monocytes and rat alveolar macrophages. 14 Thus, some uncertainty still persists as to the role of PAF in free radical (10M). Here it is demonstrated that PAF produced by EAT cells can act on those cells and generate free radicals. Although the concentration of PAF used in the in vitro assay was far greater than PAF generated in vivo by an equivalent number of cells, it is possible that the local concentration of PAF may be even greater than the one employed in this study. Such an observation has been reported in rabbit leukocytes. 16 The production of free radicals is the main function of the phagocytic cells; its generation in non-phagocytic cells such as human fibroblasts, and transformed cells such as human breast carcinoma 1 and EAT cells (present study), is interesting. Lipids, especially the phospholipids and their metabolites, seem to play a crucial role in many cell functions, particularly in intracellular signalling. 17 Phospholipids also stimulate a variety of enzyme catalyzed oxidative reactions. 6 In the present study, most of the lipids used activate the respiratory burst oxidase except for sphingolipids which are inhibitory. Such an activation of the respiratory burst oxidase by PS was also observed by Tamura et al.
Protein kinase C has been implicated as essential in activation of NADPH oxidase. 18 This is further supported by the fact that sphingolipids which are 56 Mediators of Inflammation. Vol 2.1993 inhibitors of protein kinase C also inhibit NADPH oxidase. 8 However, Tamura et al. 6 have questioned the involvement of protein kinase C, since activation observed during PS addition could not be inhibited by EGTA, which is known to inhibit protein kinase C. Hence, it appears that although direct stimulation of NADPH oxidase by PS and its inhibition by sphingolipids is possible, involvement of protein kinase C cannot be ruled out.
Augmented production of free radicals during vitamin A deficiency is interesting. Vitamin A deficiency is known to alter the membrane integrity and to bring about associated changes '9 including changes in membrane lipid composition. 19 These effects may activate the respiratory burst oxidase. In fact, the basal level of free radicals is more during vitamin A deficiency and could be suppressed by exogenous addition of vitamin A, suggesting a role for vitamin A in free radical generation.
Besides PAF, other stimulants like PMA and calcium ionophore were also capable of eliciting the respiratory burst (Table 1). Although PAF appears to be a weak stimulant, it is a physiological one.
Activation/inhibition of the respiratory burst oxidase displayed by various phospholipids and dietary factors such as vitamin A probably play a regulatory role in free radical generation. As the respiratory burst is lethal for both the invader and the host, it should not be turned on unnecessarily. Membrane lipids and vitamin A probably regulate this event.