ABSTRACT
The mouse IgE test is a novel approach to the predictive identification of chemicals that have the potential to cause sensitization of the respiratory tract. The method is based upon measurement of induced changes in the serum concentration of IgE in BALB/c strain mice following topical exposure to the test chemical. The investigations described here were undertaken to examine the stability of the assay, to evaluate the utility of trimellitic anhydride (TMA) and 2,4-dinitrochlorobenzene (DNCB) as, respectively, positive and negative controls for use in the test and to consider criteria for the classification of chemicals as potential respiratory allergens based upon changes induced in serum IgE concentration. On the basis of fifteen independent experiments it was found that, while there was some intra- and inter-test variability with respect to absolute concentrations of IgE measured in the sera of individual mice, the relative pattern of reactivity observed following treatment of mice with TMA, DNCB or with vehicle (4:1 acetone:olive oil) alone remained consistent. In each experiment treatment with TMA provoked a very substantial increase in serum IgE relative to vehicle controls. In no instance did exposure of mice to DNCB cause an increase in the concentration of serum IgE, and in some instances treatment with this chemical resulted in reduced IgE levels. In a parallel series of experiments it was found that serum IgE concentrations in vehicle-treated mice were comparable with those found in the serum of untreated animals. It is concluded that the differential ability of chemicals to induce changes in the concentration of serum IgE in BALB/c mice is a stable phenomenon. It is recommended that, in practice, TMA and DNCB should be incorporated in mouse IgE tests as, respectively, positive and negative controls. Finally, it is proposed that activity in the test is assessed as a function of induced changes in serum IgE levels relative to concurrent vehicle controls, rather than by reference to historical normal range values.
