TRADE-OFFS IN PHENOLIC METABOLISM OF SILVER BIRCH: EFFECTS OF FERTILIZATION, DEFOLIATION, AND GENOTYPE

Abstract

We examined the chemical responses of 10 silver birch (Betula pendula) clones to fertilization and defoliation in a field experiment. In defoliation, every second leaf was removed from the saplings. Three days later, two undamaged short-shoot leaves were collected, air-dried, and analyzed for condensed tannins and 34 nontannin phenolic compounds by high-performance liquid chromatography. The clones showed substantial variation in phenolic composition of the leaves and in chemical responses to fertilization and defoliation. A cluster analysis by UPGMA indicated that the phenolic profiles of birch leaves were affected more by genotype than fertilization or defoliation, and the clones could thus be distinguished from each other. In addition, on the basis of their overall phenolic composition, the clones were clustered loosely in three clone groups. The leaves of fertilized saplings contained lower levels of condensed tannins than controls, as predicted by carbon/nutrient balance (CNB) hypothesis. However, fertilization had no effect on the total amount of nontannin phenolics. The concentrations of (+)-catechin, 3,4′-dihydroxypropiophenone 3-glucoside (DHPPG), 3-cinnamoylquinic acids, and flavone aglycones were lower in fertilized saplings, whereas the opposite was true for 5-cinnamoylquinic acids and the total amount of flavonol glycosides. Although our results provide support for the CNB hypothesis, they also show that the accumulation of phenolic compounds in birch leaves is strongly coordinated. Different branches of the biosynthetic pathway of phenolic compounds may compete for substrates, and such internal metabolic trade-offs may explain the differential accumulation of the compounds. In fertilized saplings, the concentration of condensed tannins was also negatively correlated with the amount of triterpenoid resin droplets measured from the same saplings. We suggest that a linkage via malonyl-CoA between the biosynthetic routes to terpenoids and flavonoid derivatives, such as condensed tannins, may explain the different responses to fertilization reported for terpenoids and phenolics. Undamaged leaves of partially defoliated saplings contained more DHPPG and flavone aglycones and less cinnamic acid derivatives and (+)-catechin than did leaves of control saplings. The induction of DHPPG and flavonoid aglycones was significantly and negatively correlated with the concentration of myricetin glycosides in fertilized saplings, which may indicate a trade-off between induced and constitutive defense. Moreover, in fertilized saplings, the three clone groups formed by UPGMA clustering differed significantly in the magnitude of induction of DHPPG and flavone aglycones. Different birch genotypes may thus have different modes of chemical defense, and the magnitude of chemical response of a genotype may partly depend on resource availability. In general, our results show that new insights in the theory of chemical defense can be gained by accomplishing studies on plant–herbivore interaction with high chemical resolution.