Fig. 4.

LPS-induced conditioned taste aversion is neither dependent on COX-1 nor on COX-2. Conditioned taste aversion was initially tested by a protocol with two conditioning sessions (A). Saccharine drinking was reduced already after one conditioning session and further reduced after a second conditioning session (B, CTA-test 1; n = 6, LPS; n = 6, NaCl; 2-way ANOVA, session × treatment F (2, 20) = 12.87, P < 0.001). The same protocol but without saccharine did not result in reduced fluid intake at the test session (C; n = 6, LPS; n = 5, NaCl), showing that the aversion is specifically related to the taste of saccharine. Since one conditioning session was enough to induce taste aversion we used only one conditioning session for the subsequent experiments (D). COX-1 inhibition with SC-560 did not affect saccharine intake before the conditioning (E; n = 22, LPS; n = 21, NaCl), and did not interfere with the reduction in saccharine intake induced by conditioning (F; 2-way ANOVA, treatment × pre-treatment F (1, 82) = 0.4391, P = 0.509; treatment F (1, 82) = 34.62, P < 0.001). The same was the case for COX-2 inhibition with parecoxib (G, H; n = 8, LPS; n = 8, NaCl; 2-way ANOVA, treatment × pre-treatment F (1, 28) = 0.06074, P = 0.807; treatment F (1, 28) = 51.08, P < 0.001). ^*P < 0.05, ^**P < 0.01, ^***P < 0.001; ANOVA followed by Tukey’s post hoc test.