Photorespiration response and biomass production
The two most important factors explaining the variation in the D6S /D6R ratio (range: 0.85-0.94) were WT and atmospheric CO2 concentration, which respectively accounted for 48% (P< 0.001) and 14% (P< 0.001) of the total variance (Fig. 2A, Table 1). In addition, temperature and interactions between WT and CO2 both explained 7% each (P= 0.004 andP= 0.005, respectively) of the variance in this ratio. Increasing atmospheric CO2 from 280 to 400 ppm resulted in a 0.03 decrease in the D6S /D6R ratio at low WT, but had no significant effect on it at high WT (Fig. 2A, Table S1). Together with the observed interaction between CO2 and WT, this indicates WT-dependent suppression of photorespiration at the high CO2 level. Raising the WT from -20 to ~0 cm resulted in a significant (0.01-0.05) increase in the D6S /D6R ratio, indicating that the high WT increased the photorespiration/photosynthesis ratio (Fig. 2A). Increasing the day/night temperatures from 12˚C/7˚C to 17˚C/12˚C caused a small (~0.01) increase in the D6S /D6R ratio at low WT, but increasing the light intensity from 250 to 500 µmol m-2 s-1 had no significant effect on it (Fig. 2A, Table 1).
Theoretically, the suppression of photorespiration at high CO2 and low WT should have been accompanied by increases in CO2 assimilation rates, and thus biomass production, but no significant increase in biomass accumulation associated with the high CO2 level was observed under any test conditions (Fig. 2B). The variation in biomass production (0.2-3.6 g m-2 d-1) was mostly explained by temperature (27%, P< 0.001), WT (52%, P< 0.001), and to a smaller degree light intensity (4%, P= 0.003) and the interaction between temperature and WT (4%, P= 0.005; Table 1). Increasing the temperature caused a massive 2.6- to 4.5-fold increase in biomass production, whereas raising the WT strongly reduced biomass production, by 53-74% (Fig. 2B). Increasing the light intensity caused a 1.1- to 1.5-fold increase in biomass. No major between-differences in C content of the biomass (48.1 ± 0.11%, SE) were detected (Fig. S3). Thus, the observed changes in biomass production reflect proportional variations in C accumulation.
Whole-tissue δ13C and chloroplastic to ambient CO2 concentration
To further investigate physiological effects of increasing atmospheric CO2 from 280 to 400 ppm, we analyzed13C discrimination by measuring whole-tissue δ13C signatures (Fig. 3A). Variation in δ13C (which ranged from -30.5 to -25.7‰) was explained by changes in WT (48%, P< 0.001), atmospheric CO2 (13%, P< 0.001), temperature (10%,P< 0.001), light intensity (3%, P= 0.02), and the interaction between CO2 and WT (10%,P< 0.001; Table 1). Increasing atmospheric CO2 consistently decreased δ13C (by 1.4 to 1.7 ‰) at low WT, but had no significant effect at high WT (Fig. 3A, Table S1). Concomitantly, raising the WT resulted in a 0.5-2.5‰ increase in δ13C. Increasing the temperature caused a significant, 0.5-1.0 ‰, increase in δ13C. Increasing the light intensity resulted in a small 0.3-0.6‰ increase in δ13C at low WT.
The δ13C data allowed estimation of the chloroplastic CO2 concentration (c c) and subsequently the chloroplastic to ambient CO2 ratio (c c/c a, Flanagan & Farquhar, 2014), which is a key determinant of metabolic C fluxes. Variation in c c/c a(0.63-0.80, Fig. 3B) was explained by WT (57%,P< 0.001), light intensity (4%, P= 0.022), temperature (3%, P= 0.035), atmospheric CO2 (3%,P= 0.046) and the interaction between CO2 and WT (13%, P< 0.001; Table 1). Increasing atmospheric CO2 significantly increasedc c/c a, by 0.03-0.05, at low WT but had no significant effect at high WT (Fig. 3B, Table S1). Raising the WT caused a 0.02-0.09 decrease inc c/c a, and increasing the temperature decreasedc c/c a by 0.02 at low WT. Altogether, this indicates that increases in atmospheric CO2 increase c c particularly at low WT, whereas raising the WT reduces c c. A strong negative correlation was detected betweenc c/c a and the D6S /D6R ratio at low WT (R2=0.85, P< 0.001), suggesting that an increase in c c caused the decrease in photorespiration/photosynthesis ratio, i.e. suppression of photorespiration at low WT. No significant relationship in these variables was observed at high WT (R2=0.10,P =0.208).
Variation in the CO2 diffusion gradient (72-150 ppm) from the atmosphere to the chloroplasts (c a-c c) was mostly explained by CO2 (53%, P< 0.001), WT (26%, P< 0.001) and the interaction of CO2 and WT (9%, P< 0.001). Increasing atmospheric CO2 resulted in increases inca-cc of 15-52 ppm (Fig. S3, Table S2). This indicates that increases in atmospheric CO2 increased CO2 assimilation.