Fig. 1.

Fate of previously sequestered allyl glucosinolate (GLS) in P. armoraciae beetles fed on different Arabidopsis lines for five days and starved for one day. a Relationship between the total ingested GLS amount and the amount of previously sequestered allyl GLS per beetle. b Relationship between the total ingested GLS amount and the amount of excreted allyl GLS per beetle. The statistical analyses were performed by ANCOVA using the total ingested GLS amount as a covariable. In the first analysis (a), both the total ingested GLS amount and allyl GLS amount were log transformed prior to analysis. In the second analysis (b), the total ingested GLS amount and the excreted allyl GLS amount was log transformed and square-root transformed, respectively, prior to analysis. The final estimates were back-transformed to their original scale, and plotted with regression lines. Data series that are not significantly different from each other were combined for regression line plotting. c Percentage of allyl GLS in fed beetles and feces relative to the allyl GLS amount in newly emerged beetles (N = 10). myb, Arabidopsis myb28myb29 double knockout mutant; cyp, Arabidopsis cyp79b2cyp79b3 double knockout mutant