Nitrogen saving on type II cell wall has contributed to the ecological success of C4 grasses throughout evolution

Abstract

Abstract: INTRODUCTION - Grasses are the most widespread family of plants in agriculture and natural ecosystems. They are used in building, forage, food and bioenergy industry. Within grasses, 45% of species fix the CO2 using the C4 photosynthesis, that concentrates CO2 around the Rubisco reducing photorespiration. Because of their high photosynthetic efficiency, C4 plants demand less amount of Rubisco in the leaf, economizing N allocated to photosynthesis, and improving their photosynthetic nitrogen use efficiency (PNUE). However, C4 grasses can present greater PNUE (0.42 umol mmol s-1) than C4 eudicotyledons (0.28 ?mol mmol s-1), when cultivated in the same N condition. This indicates that grasses must present at least one additional feature in comparison with eudicotyledons that also contributes to their higher PNUE. Non-commelinid monocots and all eudicotyledons contain a primary type I cell wall, while grasses and a few related orders (grouped as commelinids), possess type II cell wall. These two types of cell walls present two main differences in their architectures and composition. First, grasses present glucuronoarabinoxylan (GAX) branched with ferulic acid (FA) as the main hemicellulose (FA-GAX, 20-30% dry weight). Second, grasses have lower content of structural proteins (1% dry weight). Besides ester-linking to GAX, FA can also perform ether-links to lignin, structural proteins and other FA-GAX. Like this, FA works as a molecular "glue? that cross-links the cell wall polymers, performing roles in the control of the cell wall integrity, cessation of cell growth and protection against pathogen attack. These functions are alike the functions performed by a class of cell wall proteins known as extensins, which in type I cell wall are also in charge of the cell wall integrity, cessation of cell growth, and protection against pathogen attack. However, as FA-GAX does not contain N in their chemical structure, a transference of function from structural proteins to FA-GAX could have contributed to reduce the demand for N in commelinids, which, in turn may have helped grasses to obtain their outstanding success in ecology as well as the unique value they has had to agriculture throughout the development of civilization. AIMS - To investigate the hypothesis that the N saving in type II cell wall can influence the NUE in C3 and C4 plants. For this, the work was carried out in two steps. First, grasses and eudicotyledons were grown in different N concentration to evaluate photosynthesis rate, total N and PNUE. In the second step, the same species were used to determine structural components as N and ferulic acid esterified in their cell wall. METHODS - Three species of C3 grasses (Triticum aestivum L., Phalaris, and Dicanthelium oligosanthes), three C3 eudicotyledons (Flaveria pringlei, Abelmoschus esculentus L., and Atriplex lentiformis), three C4 grasses (Saccharum officinarum L., Zea mays, and Setaria viridis), and three C4 eudicotyledons (Blepharis ciliares, Amaranthus edulis and Gomphrena globosa) were cultivated in glasshouse (University of Toronto), and watered three times a week with Johnson-Hoagland's solution containing different N concentration (classified as deficit, low, medium and normal N conditions). After thirty days, the young fully expanded leaf was used to measure the photosynthetic rate with a LiCOR 6400 infrared gas analyzer (Lincoln, NE, USA). The leaf chamber conditions were: light intensity 1200 ?mol m-2 s-1 PAR to C3 plants and 2000 ?mol m-2 s-1 PAR to C4 plants, humidity 60%, leaf temperature 25ºC to C3 plants and 30ºC to C4 plants, flow 300 ?mol s-1 and CO2 concentration 1000 ppm. Leaf discs of 1-7 cm diameter were punched out from each species; they were air-dried at 60ºC for 48 h, and ground in a mill. The total N concentration was quantified using an elemental combustion system (ECS 4010). PNUE was calculated using photosynthetic rate per unit leaf area divided by leaf N content per unit area. To structural proteins assays, leaf dried power was washed extensively in methanol/chloroform/water (12/5/3, v/v/v), and citrate buffer at pH 6.8 containing 1% (v/v) sodium dodecyl sulphate (SDS). Then, the pellet was air-dried at 60ºC for 24 h and the dry mass of pellet was assumed to represent the leaf structural biomass. Structural biomass was used to determine the structural N content. To quantify ferulic acid ester-linked in the cell wall, the biomass was homogenized with methanol (50%, v/v) and incubated at 80°C for 90 min. The pellet was dried at 60°C for 24 h. The dry cell wall was re-suspended in 0.5 M NaOH and incubated at 96°C for 2 h. The supernatant was acidified to pH 2.0 with 6 M HCl, centrifuged at 2.180g, 4ºC for 15 min and then extracted twice with anhydrous ethyl ether. The ethyl ether extracts were combined and dried at 40°C. The samples were re-suspended in methanol/acetic acid 4% (30/70, v/v) and analyzed with a Shimadzu® Liquid Chromatograph (HPLC). The unpaired two-side t-test was applied to evaluate the differences between parameters and values p ? 0.05 were considered statistically significant. RESULTS AND DISCUSSION - Comparing plants with the C3 photosynthetic metabolism, the assays revealed that 1) C3 grasses and C3 eudicotyledons tested presented similar amounts of N (total) in their leaves. In general, these results reflected in similar PNUE values to C3 grasses and C3 eudicotyledons. 2) In all N conditions C3 grasses and C3 eudicotyledons also did not present differences in the amount of structural N. 3) Although in N deficit, FA ester-linked to cell wall was 446% higher in C3 grasses than in C3 eudicotyledons (reflecting a role of FA in response to N deficit), this higher content of FA into the type II cell wall of C3 grasses did not contribute with the N economy. In turn, comparing plants with C4 photosynthetic metabolism, the tests indicated that 4) C4 grasses presented less N (total) content in their leaves than C4 eudicotyledons. 5) in N deficit, the lower total N contributed to a the higher PNUE (up 81%) in C4 grasses when compared with C4 eudicotyledons. 6) In addition, in deficit C4 grasses presented up to 58% less structural N than C4 eudicotyledons. 7) In deficit, C4 grasses showed up to 177% higher amounts of FA ester-linked into their cell wall than C4 eudicotyledons. 8) Although the N structural content contributes with a rather small percentage of the total N in the leaves (1 to 10% into structural proteins), this percentage are spread for the whole plant, Rubisco is limited to leaves and green parts of the plant. The reduction in structural N content observed in C4 grasses can respond for about 20% of the total N savings in these plants when compared with C4 eudicotyledons, when both groups were grown in deficit of N. The structural proteins are the major nitrogenous compounds in the plant cell wall. When C is not the main limiting factor to growth, as occurs with C4 plants, N becomes the main limiting nutrient. Our data suggests that during evolution of C4 grasses, N limitation in soil imposed a selection pressure to transfer the function performed by structural proteins to FA, reducing demand for N and contributing to a higher NUE. This was possible due the distinct type II cell wall architecture present in C4 grasses, but not in C4 eudicotyledons. Our data also suggest that, in turn, although presenting a type II cell wall able to save N, C3 grass lineages were not selected to save N, since throughout C3 evolution, carbon continued to be the main limiting factor to growth. CONCLUSIONS - The reduced content of structural proteins whose function was partially replaced by FA in type II cell wall, contributes to decrease the N allocated into cell wall in C4 but not in C3 grasses. Therefore, together with the sensible reduction in Rubisco content (the main N sink in plant leaf) allowed by the limited Rubisco oxygenase activity in C4 plants, the cell wall architecture of C4 grasses provides na additional contribution to the higher NUE presented by C4 grasses.