CO2 enrichment and carbon partitioning to phenolics: do plant responses accord better with the protein competition or the growth differentiation balance models?

Rising levels of atmospheric CO2 can alter plant growth and partitioning to secondary metabolites. The protein competition model (PCM) and the extended growth/differentiation balance model (GDBe) are similar but alternative models that address ontogenetic and environmental effects on whole-plant carbon partitioning to the phenylpropanoid biosynthetic pathway, making many divergent predictions. To test the validity of the models, we compare plant responses to one key prediction: if CO2 enrichment simultaneously stimulates both photosynthesis and growth, then PCM predicts that partitioning to phenolic compounds will decline, whereas GDBe generally predicts the opposite. Elevated CO2 (at 548 ppm) increased the biomass growth (ca 23%) as well as the net photosynthesis (ca 13%) of 1-year-old potted paper birch, Betula papyrifera Marsh., in a free air carbon dioxide enrichment study (FACE) in northern Wisconsin. Concomitantly, elevated CO2 increased carbon partitioning to all measured classes of phenolics (Folin-Denis phenolics, HPLC low molecular weight phenolics (i.e. cinnamic acid derivatives, flavonol glycosides, and flavon-3-ols), condensed tannins, and acid-detergent lignin) in leaves. In stem tissues, tannins and lignin increased, but F-D phenolics did not. In root tissues, F-D phenolics, and tannins increased, but lignin did not. The data suggest that CO2 enrichment stimulated pathway-wide increase in carbon partitioning to phenylpropanoids. High CO2 plants had 11.8% more F-D phenolics, 19.3% more tannin, and 10% more lignin than ambient plants after adjusting for plant mass via analysis of covariance. In general, the results unequivocally support the predictions of the GDBe model. By way of contrast, results from many parallel studies on FACE trembling aspen, Populus tremuloides Michx., suggest that although CO2 enrichment has consistently stimulated both photosynthesis and growth, it apparently did not generally stimulate pathway-wide increases, or decreases, in carbon partitioning to phenylpropanoids in leaves and wood, but rather has specifically, though not consistently, increased partitioning to foliar phenolic glycosides. Likewise, in this case, GDBe's predictions better accord with the FACE aspen data than PCM's. If further tests of the two models also support GDB rather than PCM, then PCM's main assumption (whole-plant N rather than C is limiting partitioning to phenolic synthesis) may be incorrect. [ABSTRACT FROM AUTHOR]