Leaf growth in early development is key to biomass heterosis in Arabidopsis

Abstract

Arabidopsis thaliana hybrids have similar properties to hybrid crops, with greater biomass relative to the parents. We asked whether the greater biomass was due to increased photosynthetic efficiency per unit leaf area or to overall increased leaf area and increased total photosynthate per plant. We found that photosynthetic parameters (electron transport rate, CO2 assimilation rate, chlorophyll content, and chloroplast number) were unchanged on a leaf unit area and unit fresh weight basis between parents and hybrids, indicating that heterosis is not a result of increased photosynthetic efficiency. To investigate the possibility of increased leaf area producing more photosynthate per plant, we studied C24×Landsberg erecta (Ler) hybrids in detail. These hybrids have earlier germination and leaf growth than the parents, leading to a larger leaf area at any point in development of the plant. The developing leaves of the hybrids are significantly larger than those of the parents, with consequent greater production of photosynthate and an increased contribution to heterosis. The set of leaves contributing to heterosis changes as the plant develops; the four most recently emerged leaves make the greatest contribution. As a leaf matures, its contribution to heterosis attenuates. While photosynthesis per unit leaf area is unchanged at any stage of development in the hybrid, leaf area is greater and the amount of photosynthate per plant is increased.

Keywords: Arabidopsis; CO2 assimilation; biomass vigour; early germination; electron transport rate; heterosis; hybrid; leaf development; photosynthesis.

Fig. 1.

The light–response curves of the quantum yield of PSII (ϕ _PSII) and the electron transport rate (ETR) of the 6 DAS C24×Ler hybrid seedlings grown on MS medium either (A) with or (B) without 3% sucrose. The chlorophyll fluorescence parameters were analysed and used to calculate ϕ _PSII as described by Genty et al. (1989) using a pulse amplitude-modulated fluorometer (Closed FC 800-C, PSI). Each data point represents the average and SE of n≥5. The ETR was calculated by ETR=ϕ _PSII×PAR×0.5×0.84, as described in Edwards and Baker (1993). (C and D) The maximum electron transport rates (J_max) calculated by fitting the ETR curves of parental lines and hybrids grown either with or without sucrose to the C₃ photosynthesis model. See Supplementary Fig. S1 for the chlorophyll fluorescence light–response curves of other hybrids used in this experiment. Statistical comparisons among parents are shown; datasets with different letters are significantly different from the others (ANOVA, P<0.05). Asterisks indicate a significant difference between the hybrid and the better parent (ANOVA, P<0.05). No significant differences were found between hybrids and the mid-parent value (indicated by red dashed lines) and between hybrids and the lower parent (ANOVA, P>0.05).

Fig. 2.

The CO₂ gas exchange of 3-week-old hybrids compared with parents. The plants were sown on MS solid medium before being transferred into soil at 10 d old. Plants were grown under a light cycle of 16 h/8 h, in a light intensity of 120 µmol photons m⁻²s⁻¹, at 21 °C. (A) The net CO₂ assimilation rate of the hybrids and parents was measured under an increasing partial pressure of atmospheric CO₂ under a saturating irradiance at 1000 µmol m⁻² s⁻¹ using a gas exchange analyser (LI-6400XT, LICOR). The x-axis shows the intercellular CO₂ partial pressure. The analyses were carried out on the largest leaf on the rosette of each plant measured. Each data point represents the average and SE of n≥3. (B) Comparison of the maximum rate of Rubisco carboxylation (V_cmax) between hybrids and the parents. The curve fitting result using the dataset from (A). The curve fitting was carried out by fitting the dataset from (A) to the C₃ photosynthesis model (Farquhar et al., 1980). Each data point represents the average and SE of n=3. Different letters above the columns represent a significant difference between the parents (ANOVA, P<0.05). Asterisks represent significant differences between the hybrid and the better parent (ANOVA, P<0.05). No significant differences were found between hybrids and the lower parents, and between the hybrids and the growth under the average value of the parents (indicated by red dashed lines; ANOVA, P>0.05).

Fig. 3.

Qualitative analyses of starch turnover under standard conditions. (A) Arabidopsis seedlings were grown under a light cycle of 16 h light/8 h dark, in a light intensity of 120 µmol m⁻² s⁻¹, at 21 °C. Rosettes of 21 DAS seedlings were harvested at the end of the night and the end of the day, respectively, and stained with Lugol’s iodine solution (0.1%). Dark purple stain indicates the presence of starch granules. Scale bars=2 cm. (B) Quantitative analyses of starch content of C24×Ler grown in long (16 h/8 h) or short (8 h/16 h) photoperiods under standard (120 µmol m⁻² s⁻¹) or doubled (240 µmol m⁻² s⁻¹; designated as L₂₄₀) light intensity. Starch content is presented as the equivalent amount of glucose. Data shown were the average and SE of three technical replicates from a pool of n=4–6 rosettes. For long photoperiods, 21 DAS plants were used, whereas 40 DAS plants were used for short photoperiods. (C) The unit biomass starch turnover per night. This was calculated by subtracting the average starch content at the end of the night from that at the end of the day (SE, n=4–6). End of the day, 1 h before lights off; end of the night, 1 h before lights on. Asterisks indicate significant differences from growth under the standard light conditions (ANOVA; *P<0.05). Significant differences between the hybrid and the parents are represented by asterisks in colour (blue, C24; orange, Ler; ANOVA, *P<0.05).

Fig. 4.

C24×Ler hybrid seedlings grown under a photon density of 120 µmol m⁻² s⁻¹ (L₁₂₀) or 240 µmol m⁻² s⁻¹ (L₂₄₀) under short (8 h) or long (16 h) photoperiods. Arabidopsis seedlings were grown under 120 µmol m⁻² s⁻¹ for 10 DAS before half of the seedlings were transferred into 240 µmol photons m⁻² s⁻¹. The 18 DAS 240 µmol photons m⁻² s⁻¹-grown plants were assessed for heterosis in vegetative biomass and compared with the 120 µmol photons m⁻² s⁻¹-grown seedlings. (A) The images of the 35 DAS and 21 DAS C24×Ler grown under 8 h and 16 h photoperiods, respectively. (B and C) Measurements of rosette biomass. Data presented are the average and SE of n>15 from 2–3 experiments. Asterisks indicate that the hybrids were significantly different from the average levels of the parents (red dashed lines) (ANOVA, *P<0.05, **P<0.01, ***P<0.001). (D) Heterosis levels of hybrids presented as a percentage increase from the average levels of the parents. Statistical comparisons were carried out between and within hybrids in different light conditions. The asterisks indicate a significant difference from the control light under the same photoperiod (ANOVA, P<0.05).

Fig. 5.

Earlier germination and leaf emergence in C24×Ler hybrids relative to parents. (A) Photographs of parents and reciprocal hybrids at 28 h after sowing (HAS), 3, 6, and 9 DAS. (B) Germination of C24×Ler hybrids and parents (see also Zhu et al., 2016). Germination was scored as the first visible sign of root emergence.

Fig. 6.

Measurements of the true leaves of C24×Ler hybrids at (A) 19 DAS and (B) 30 DAS. True leaves on a rosette were numbered in the order of their positions in the rosette, from the bottom to the top of the rosette (i.e. the order of emergence from the meristem). In each line, three seedlings were examined. Error bars represent the SE of n=3 plants. Columns without error bars represent the average of data from n<3. Black asterisks indicate a significant difference from the better parent; red asterisks of 30 DAS leaves indicate a significant difference in the heterosis levels from the corresponding leaves at 19 DAS (ANOVA; *P<0.05, **P<0.01, ***P<0.001). Statistical comparison was carried out only on datasets containing three biological replicates.

Fig. 7.

Relative contribution of leaf sets to the whole-plant biomass. Data presented are the average and SE of the sum of the leaf area per rosette, using the dataset from Fig. 6. Each data point was divided into three fractions: indicating the relative contribution of the largest leaves and the rest of the leaves in a seedling, in the order of their relative positions in a rosette. Percentages above columns represent the percentage of heterosis relative to the mid-parent value (red dashed lines). Asterisks indicate significant differences from the average level of the parents. *P<0.05, **P<0.001, ***P<0.001 (ANOVA). MPV, mid-parent value.