2011). Under such a scenario, selection would favor mutations that lead to a co-limitation of g s and RuBP utilization and regeneration. In general, winter Arabidopsis accessions had lower g s and A than spring Arabidopsis accessions. Across accessions there was large variation in C i /C a, but it was only weakly related to δ13C (Fig. 4). No consistent difference in C i /C a was seen between the winter and spring

annuals. Fig. 4 Relationship between the ratio of intercellular to atmospheric MLN4924 molecular weight partial pressure CO2 (C i/C a) at 350 μmol photons m−2 s−1 and carbon isotope composition (δ13C). Open and filled symbols represent spring and winter accession means, respectively. Line represents linear regression; r 2 and P values are given

The overall finding of experiment 2 was that accessions with low g s and high δ13C had lower A compared to low δ13C accessions. Overall, these data are consistent with large effects of g s on δ13C, but the weaker correlation of C i and δ13C suggest a more complex mechanism than predicted by theory. To better understand processes limiting photosynthesis in Arabidopsis accessions, we conducted detailed CO2 response curves of assimilation for low and high WUE spring accessions Tsu-1 and SQ-8 and high WUE winter accession Kas-1. Maximum carboxylation rate of rubisco (V cmax) was higher in low WUE p38 MAP Kinase pathway Tsu-1 (δ13C = −29.7) than Sq-8 (δ13C = −28.6) (P = 0.01), as expected (Fig. 5). Similar, maximal photosynthetic electron GS-1101 datasheet transport (Jmax) was also higher in Tsu-1 than Sq-8 or Kas-1 (δ13C = −28.8) (P = 0.002, P = 0.002). Fig. 5 Maximum carboxylation rate of rubisco (V cmax) and maximal photosynthetic electron transport (Jmax) obtained from photosynthetic carbon dioxide response curves in three accessions (Tsu-1, Sq-8,

and Kas-1) which differed in Reverse transcriptase A. Each bar represents the mean ± SE (n = 4) for each accession. Letters represent significant differences among accessions. Genotype F-ratio = 12.14 and P = 0.0078 for V cmax. Genotype F-ratio = 11.01 and P = 0.0098 for Jmax The major biochemical limitations to photosynthesis, V cmax and Jmax, appeared optimized to accessions’ C i as indicated by δ13C. V cmax and Jmax were lower in low g s, high WUE accessions operating at lower C i. The higher ratio of V cmax to Jmax in Kas-1 compared to Sq-8 suggests a lack of limitation by Jmax under the low g s typical of Kas-1. Simultaneous changes in V cmax and Jmax are consistent with a limitation of photosynthesis by RuBP utilization and regeneration (Farquhar and Sharkey 1982). Likewise, proportional changes in components of photosynthetic apparatus and g s suggest acclimation of these processes are closely coupled (Cowan 1986). Variation in structure In experiment 3, we examined 39 natural accessions of Arabidopsis for variation in δ13C and LWC (Table 1). We found a significant negative correlation between δ13C and LWC among accessions (r 2 = 0.6, P < 0.0001).