Rice Seed Germination Underwater: Morpho-Physiological Responses and the Bases of Differential Expression of Alcoholic Fermentation Enzymes

Abstract

The water-, energy-, and labor-intensive system of transplanted puddled rice (Oryza sativa) is steadily being replaced by direct seeding due to the progressive scarcity of these resources. However, the alternate dry direct seeding leads to competition with weeds and poor establishment when soils are flooded. Direct seeded rice capable of anaerobic germination (germination in flooded soil, AG) is ideal, which under rainfed ecosystems would also overcome waterlogging during germination. AG tolerance is associated with faster germination and faster elongation of coleoptiles, with the activities of alcoholic fermentation enzymes replacing aerobic respiration as a source of energy. To better understand the variability in the morpho-physiological responses and in the nature of the alcoholic fermentation enzymes during AG, 21 rice genotypes were studied. The genotypes Khao Hlan On (KHO) and IR42 were used as the tolerant and susceptible checks, respectively. KHO exhibited faster germination, with 82.5% of the coleoptiles emerging out of 10 cm of water within 8 days, whereas IR42 exhibited 20% germination and limited coleoptile growth. Among the test genotypes, four performed well, including two that are drought tolerant. Increased content and activity of the alcoholic fermentation enzymes, alcohol dehydrogenase (ADH1) and acetaldehyde dehydrogenase (ALDH2a and ALDH2b), was noted in KHO under anaerobic than under aerobic conditions and also in comparison with IR42 under AG. Gene transcripts for these enzymes were also more in KHO undergoing AG. However, no major differences were observed between KHO and IR42 in the critical cis-acting regulatory elements, such as the auxin, light, and sugar response elements, in the promoters of ADH1, ALDH2a, and ALDH2b genes. Post-transcriptional and post-translational regulatory mechanisms were implicated for the increased transcript and protein content/activity of the enzymes in KHO by observing four different transcripts of ALDH2a and a unique non-glycosylated form of ADH1 under AG. IR42 lacked the non-glycosylated ADH1 and contained only a truncated form of ALDH2a, which lacked the active site. Additionally, KHO exhibited increased activity and more isoforms for reactive oxygen species detoxifying enzymes under AG compared to IR42. These results highlight the need for a deeper functional understanding of the critical enzymes involved in AG.

Keywords: acetaldehyde; alcohol dehydrogenase; aldehyde dehydrogenase; anaerobic germination; anaerobiosis; coleoptile; post-translational modifications; splice-variants.

FIGURE 1

Principal component analysis for aerial traits. (A) Principal components (PC) of the 20 genotypes (individuals) for the 14 shoot traits (listed in Supplementary Table 1) where grouping is according to the known tolerance classification according to previous experiments: tolerant, sensitive, and unclassified (not previously been classified). Each point represents individual genotype and the three larger points represent the mean for each group. Ellipses are drawn at a 95% confidence interval. (B) Correlation matrix between the traits. Positive correlations are displayed in blue, and negative in red. Color intensity and circle size are proportional to the correlation coefficients. (C) PC plot of the 14 traits (variables) for the 20 genotypes, colored according to their percentage contribution to the component (variable cos^2∗100/component cos²). Color and length of the arrow are proportional to the contribution to the component. Trait legend: IN, internode number; PAW, plants surfacing above water; NEP, number of plants emerging above water; PN, plant number; LAW, leaf portion emerging above water (cm); LN, leaf number; PL, plant length; BD, base diameter; IL, internode length; SL, shoot length; ADW, shoot dry weight; RDW, root dry weight; FP, number of floating plants; ML, mesocotyl length.

FIGURE 2

Principal component analysis for root traits. (A) Principal components (PC) of the 20 genotypes (individuals; listed in Supplementary Table 1) for the nine root traits, where each point represents individual genotype (four replications per genotype) and the 20 larger points represent the mean for each genotype. (B) Correlation matrix between the traits. Positive correlations are displayed in blue, and negative in red. Color intensity and circle size are proportional to the correlation coefficients as detailed in the legend. (C) PC plot of the nine root traits (variables) for the 20 genotypes, colored according to their percentage contribution to the component (variable cos^2∗100/component cos²). Color and length of the arrow are proportional to the contribution to the component. Trait legend: AVGDIAM, average root diameter; ROOTVOL, root volume; PROJAREA, root projected area; SURFAREA, root surface area; FORKS, number of forks; CROSS, number of crosses; TIPS, number of tips; LENGTH, root length; LENVOL, ratio of root length to volume.

FIGURE 3

Variation in expression of ALDH2a and ALDH2b genes in ‘Khao Hlan On’ and ‘IR42’ genotypes, in both control and flooded conditions. (A) qRT-PCR results for ALDH2a in ‘Khao Hlan On’; (B) qRT-PCR results for ALDH2a in ‘IR42’; (C) qRT-PCR results for ALDH2b in ‘Khao Hlan On’; (D) qRT-PCR results for ALDH2b in ‘IR42.’ Light gray coloring represents submerged conditions and dark gray represents control conditions. SEM error bars are displayed for each time point, with three replications per time point. Relative mRNA level was normalized to the corresponding ubiquitin mRNA expression.

FIGURE 4

In-gel assay for acetaldehyde metabolism for ‘Khao Hlan On’ and ‘IR42’ in flooded and control conditions. (A) Dry seeds were directly sown in soil under 100 mm water (flooded) or (B) in soil watered to field capacity (control). Samples are loaded from 0 to 8 days in flooded and from 1 to 8 days in control, M is a dual color protein marker. In tolerant genotype ‘Khao Hlan On’ there is a dominant enzyme activity in the form of a doublet that metabolizes acetaldehyde in flooded conditions (thick black arrow). This same enzyme is also present in ‘IR42’; however, the activity is weaker and only one band is present. Another enzyme activity is visible at a higher molecular weight in both ‘Khao Hlan On’ and ‘IR42’ in both flooded and aerobic conditions. Enzyme activity that appears as white bands (red arrow) is discussed in the text.

FIGURE 5

ADH1 protein sequence analysis between ‘Khao Hlan On,’ IR42 and ‘Nipponbare.’ ClustalW pairwise alignment of ‘Khao Hlan On’ protein sequence translated from the CDS cloning against “Nipponbare” sequence. Both ‘Nipponbare’ and IR42 shared the same protein sequence. Orange letters mark the active site (residues 68–82); green shading indicates the catalytic zinc binding sites (residues 47; 69; 72 KHO and 73 IR42/Nipponbare), pink shading the other zinc binding sites (residues 100, 102, 105, 113), and blue shading NAD⁺ binding sites (residues 203–207, 225, 230, 294–296 for KHO; and 204–208, 226, 231, 295–297, 372 for IR42/Nipponbare).

FIGURE 6

In-gel assay for acetaldehyde metabolism in different varieties in flooded conditions at day 6 after sowing. From left to right, tolerant genotypes are ‘Ma Zhan (red)’ (lanes 1, 2) and ‘Khao Hlan On’ (lanes 3, 4), sensitive variety IR42 (lanes 5, 6) and moderately tolerant ‘Nipponbare’ (lanes 7, 8). Each sample was loaded in two contiguous wells. All genotypes share the upper band, but the doublet is unique to Khao Hlan On within the varieties analyzed. The lower band of Khao Hlan On has stronger activity than the upper band.

FIGURE 7

Two-dimensional denaturing gel for KHO and IR42 activity band. (A) Silver staining of a 2D denaturing gel for ADH activity band eluted from KHO; and (B) IR42. The extraction was made from seedlings grown under submerged conditions at day 4. The sample was then run in a 7 cm strip ranging from pH 3 to pH 10, and then in a denaturing gel to separate by molecular weight. To avoid sample mixing, three gels were run simultaneously in the same tank: one with the marker, one with the KHO sample and one with IR42 sample. The red circle shows a series of three spots for the putative phosphorylated moieties of the ADH1 from KHO which are absent in IR42.