Among many food additives, preservatives contribute to safe and stable supply of food. On the other hand, they may be hazardous to our health depend on the amount of use. Although acceptable daily intake (ADI) has been used for the safety assessment of food additives, this stndard is solely based on the knowledge of whole animals. Therefore, they may have some adverse effect on cellular levels even if their used level are below ADI, or commercially used level. In this paper we carried out to investigate their cellular effects on ADI and daily intake level. Five preservatives (butyl p-hydroxy benzoate, hinolitiol, sodium dehydroacetate, sodium benzoate, potassium sorbate, sodium propionate) certified by the Japanese Standard of Food Additives, and noncertified one (salicylic acid) were chosen and used. Acetyl salicylic acid was used as a referential inhibitor of platelet function. Each compound was dissolved in Ca2+, Mg2+ free Tyrode buffer pH 7.4 at concentrations corresponded to ADI, 1/5ADI, 1/10ADI, 1/20ADI, and 1/40ADI, respectively. Washed rabbit platelets were prepared by the method decribed. An aliquot (2x107 platelets) was incubated with calcium ionophore A-23187 (10-6M) or thrombin (1U), or without either agonist in the presence or absence of preservatives. Platelets were also pretreated with an individual preservative and then subjected to incubation with agonists. Commercially used amount of each preservative was added to drinking water and administered to a rabbit for 5 days, then washed platelets were prepared and subjected to incubation. Incubation lasted for 5 min at 37℃ with vigorous shaking and was terminated by placing the reaction mixture on ice followed by centrifugation. The magnitude of platelet activation was determined by the amount of thromboxane B2 (TXB2) synthesis and platelet aggregation. Among lipid soluble preservatives, Butyl p-hydroxybenzoate had a strong inhibitory effect on both agonists' induced TXB2 synthesis (Fig. 1A, 2A), and this effect revealed to be irreversible in vitro (Fig. 3, 4). On the other hand, water soluble preservatives (sodium dehydroacetate, sodium benzoate, potassium sorbate, and sodium propionate) inhibited A-23187 induced TXB2 synthesis irreversibly at the maximum concentrations used (Fig. 1B, 5), and exerted concentration dependent reversible inhibitory effect against thrombin induced TXB2 synthesis in vitro (Fig. 2, 6). In an ex vivo experiment, in contrast to in vitro experiment, hinokithiol and sodium dehydroacetate inhibited both agonist induced TXB2 synthesis. But butyl p-hydroxy benxoate had no effect on TXB2 synthesis. Platelet aggregation was inhibited or suppressed in those samples prepered from administration of lipid soluble preservatives (Fig. 9, 10), but any effect was observed in samples from administration of water soluble ones. There were no significant differences of blood biochemical assay data between pre and post administration of preservatives (Table 1).
