Maize Kernel Development and Composition During LateSeason Water

Maize Kernel Development and Composition During Late-Season Water Stress Implications for Dry-Grind Ethanol Production Abstract: Jason Haegele* and Mark Westgate, Department of Agronomy, Iowa State University, Ames, IA 50010 *e-mail: jhaegele@iastate. edu Maize production is becoming increasingly focused on key end-use markets that have specific, divergent criteria for grain quality traits. In the case of the dry-grind ethanol industry, grain that contains more highly digestible starch is desired to improve both plant efficiency and total ethanol yield per unit weight. Breeding and biotechnology approaches to improve maize grain for ethanol production are underway within the seed industry. But environmental variation in kernel composition and starch digestibility may limit the progress of these efforts. The objective of this study was to assess the impact of late-season water stress on maize kernel development, composition, and starch digestibility characteristics. Greenhouse grown plants of the inbred B 73 were subjected to well watered (WW) and water stressed (WS) conditions beginning 17 days after pollination (DAP). Water stress during grain filling did not significantly alter seed composition (% starch, protein, and oil). The content of these components was reduced, however, due to a shorter grain fill duration, which decreased final grain dry weight. Kernels from water stressed plants also were less dense and contained less vitreous endosperm. Preliminary tests suggest WS endosperms are more susceptible to enzymatic hydrolysis. These results indicate that water stress during grain filling would impact the biofuels industry primarily through reduced feedstock availability. Introduction: Characterizing existing maize germplasm for ethanol yield potential is the first step toward developing improved hybrids specifically for use by the ethanol industry. These characterization studies have revealed both genetic and environmental variation for ethanol production (Sharma et al. 2006; Singh and Graeber, 2005). Intuitively, total starch content should be a good predictor of ethanol production but this characteristic has proven to be poorly correlated with ethanol yield (Singh and Graeber, 2005). Therefore, other kernel characteristics such as protein and oil content or susceptibility of the starch to enzymatic hydrolysis might be involved. In the dry-grind ethanol process, a-amylase and glucoamylase are used to hydrolyze the starch to fermentable sugar (glucose). a-amylase aids in solubilization of the starch and initial polymer hydrolysis. Glucoamylase completes the hydrolysis to glucose. The efficiency of both enzymes is affected by starch structure including the presence of amylose (a linear glucose polymer) and amylopectin branching. The amylose content of commercial dent corn generally ranges from 25 -28% (% weight of total starch). Amylopectin (the branched glucose polymer) exists in chains of varying lengths. The presence of longer branch chains is correlated with decreased susceptibility to enzymatic hydrolysis (Jane et al. , 2003). Environmental effects on amylose content and amylopectin chain lengths are not well understood. The protein matrix that contains starch granules has also been proposed as a potential factor limiting starch hydrolysis and ethanol yield. Wu et al. (2006) found no significant effect of protein content on ethanol production. The types of storage proteins (zeins) that accumulate in the endosperm, however, can vary with genotype, environmental conditions, and agronomic practices. Therefore, the types of proteins being synthesized might be a more important factor than total protein content for starch hydrolysis. A more detailed understanding of environmental effects on starch content, amylose content, amylopectin characteristics, and other kernel traits is needed to establish the bases for variability in susceptibility to enzymatic hydrolysis and resulting ethanol yield. It is well known that kernel development and composition are responsive to changes in assimilate availability mediated by abiotic stresses such as water stress and high temperature. These studies, however, have not established a link between stress response and suitability for ethanol production. Therefore, we hypothesize that water stress during the later stages of grain filling influences kernel composition and starch structural properties, and that these changes result in variability in starch accessibility and/or susceptibility to enzymatic hydrolysis. Materials and methods: Growth conditions and treatments: Maize plants (B 73) were grown in the ISU Department of Agronomy greenhouses in individual 19 liter plastic pots. Plant density was 2. 4 plants m-2. Growth conditions consisted of 15 -h photoperiods and 27 o. C/18 o. C maximum/minimum temperatures. Fertilizer (15 -5 -15; N-P-K) was injected into the irrigation water at a ratio of 1: 40. After pollination, irrigation was managed automatically by GP 1 data loggers and SM 200 soil moisture sensors (Delta-T Devices, Cambridge, UK) used to control solenoid valves (Rain Bird Corporation). Water was withheld from the water stress treatment beginning approximately 17 DAP and continued until physiological maturity. Sampling: The plants were self- or sib-pollinated and ears were harvested at intervals from 12 to 40 DAP. Fifteen kernels from the middle of the ear were removed in a humid box and used for dry weight, fresh weight, water content, and % moisture content measurements. Water potential: Midday leaf water potential was measured during mid-grain fill using isopiestic thermocouple psychrometry (Boyer, 1995). Kernel composition and density: Total starch, oil, and protein concentration was analyzed by Near-Infrared Spectroscopy (NIR, ISU Grain Quality Lab). Specific density was measured with a nitrogen gas pycnometer. Starch content in excised, lyophilized endosperm and embryo tissue was determined by hydrolysis with glucoamylase followed by spectrophotometric quantification of glucose. Total nitrogen content in endosperm and embryo tissue was determined by combustion analysis; total N was converted to protein using the conversion factor of 6. 25. Percent vitreous endosperm was determined by manual dissection and the ratio of vitreous to total endosperm was expressed on a fresh weight basis. Susceptibility of starch to enzymatic hydrolysis was determined as the rate of glucose liberation from ground kernels by glucoamylase. The glucose was quantified spectrophotometrically. Conclusions: Results: • Soil moisture content (Fig. 1) decreased rapidly in the water stress treatment. • Kernel fresh weight, dry weight, and water content all were negatively impacted by water stress imposed during grain filling. Lower kernel weights resulted from a shorter duration of grain filling. • Leaf water potential 22 -25 DAP in the stress treatment was less than -2 MPa (Fig. 2) indicating a loss of photosynthetic capability. • Starch, protein, and oil contents were impacted more than were their concentrations. Protein accumulation appeared to be particularly responsive under well-watered conditions. Figure 1. Volumetric soil moisture content during kernel development. Arrow notes the beginning of the water deficit treatment 17 DAP. Figure 2 (insert). Midday leaf water potential measured during mid-grain fill (22 -25 DAP). Fresh Weight * P < 0. 05 ** P < 0. 01 Dry Weight • Water stress had a marked effect on endosperm density and vitreousness. The shorter duration of grain filling clearly limited the accumulation of starch and likely limited a-zein accumulation. • Preliminary analysis of starch hydrolysis suggests WS altered starch availability and/or digestibility characteristics. Further research will determine whether the apparent differences are due to a change in starch structure and/or starch-protein matrix. Water Content Results: • Kernels from the WS treatment contained less vitreous (hard or horny) endosperm than did kernels from WW plants (Fig. 5). *** P < 0. 001 • Kernel density (g cm-3) was positively related to the proportion of vitreous endosperm Figures 3 A-3 C. Development of kernel fresh weight, dry weight, and water content on a % moisture content basis. Each point is the mean ± SD of 2 -15 replications. Fresh weight regression equations were of the form y=(a+cx+ex 2+gx 3)/(1+bx+dx 2+fx 3). Dry weight regression equations were of the form y 0. 5=a+bx 3. Water content regression equations were of the form y 0. 5=a+bx+cx 2 (water stress treatment) and y=a+b(lnx)+c(lnx)2+d(lnx)3 (irrigated treatment). • Water stress during grain filling decreased maximum kernel fresh weight, dry weight, and water content. Final kernel dry weight decreased by 21. 2% due primarily to a shortened grain filling period. · Variation in endosperm vitreousness has been related to a-zein accumulation (Chandrashekar and Mazhar, 1999). • Kernel moisture content (%) plotted versus DAP followed the same linear trend in both treatments (data not shown). Thus, developmental trends are comparable on a moisture content basis (Borrás and Westgate, 2006). Starch * P < 0. 05 ** P < 0. 01 Protein Figure 5. Kernel density versus % vitreous endosperm. Regression equation is of the form y=a+bx+cx 2. Insert photos show a low density kernel with less vitreous endosperm (left) and a high density kernel with a greater proportion of vitreous endosperm (right). ** P ≤ 0. 01. Each data point represents the mean of three kernels from a single ear. Oil *** P < 0. 001 * P < 0. 05 ** P < 0. 01 *** P < 0. 001 • Solubilized starch from WS kernels apparently was hydrolyzed more readily than was starch from WW kernels (Fig. 6). • The increase in rate of hydrolysis may reflect a change in starch structure (amylose content or amylopectin chain length) or starch matrix (protein binding). Figures 4 A-4 C. Starch content, protein content, and oil content versus kernel dry weight. Regression equations are of the form y=bo+b 1 x. Regression coefficients within each column followed by the same letter are not significantly different (P < 0. 05). Figure 4 A photo inserts: cross section of kernel from water stress treatment showing area lacking starch accumulation (left) and fully developed kernel from irrigated treatment (right). • Water stress during grain filling decreased kernel starch, protein, and oil contents (Fig. 4). Although kernels in the water stress treatment were small and accumulated less starch, the concentration (per mg) was not affected. The photo (Fig. 4 A right) shows the lack of starch accumulation in the endosperm of a water stressed kernel. • Larger kernels in the irrigated treatment accumulated significantly (P < 0. 05) more protein per kernel and per mg DW than did their water stressed counterparts (Fig. 4 B). Averaged across all kernel sizes, however, protein concentration was not significantly different between treatments (Table 1). • Oil content increased with kernel weight in both treatments (Fig. 4 C). This increase in oil content reflects greater embryo development in the larger kernels (data not shown). Table 1. Quality traits for kernels sampled from plants water stressed and fully irrigated during grain filling. When averaged over all kernel sizes, water stress significantly affected kernel density, embryo starch concentration, and endosperm protein concentration. Traits highlighted in red were significant at P < 0. 05. Figure 6. Susceptibility of solubilized starch to hydrolysis by glucoamylase (amyloglucosidase). Regression equations are of the form y=bo+b 1 x. Regression coefficients within each column followed by the same letter are not significantly different (P < 0. 05). References: Borrás, L. , and M. E. Westgate. 2006. Predicting maize kernel sink capacity early in development. Field Crops Research 95: 223 -233. Boyer, J. S. 1995. Measuring the Water Status of Plants and Soils. San Diego: Academic Press. Chandrashekar, A. , and H. Mazhar. 1999. The biochemical basis and implications of grain strength in sorghum and maize. Journal of Cereal Science 30: 193 -207. Jane, J. , Z. Ao, S. A. Duvick, S. -H. Yoo, K. -S. Wong, and C. Gardner. 2003. Structures of amylopectin and starch granules: how are they synthesized? Journal of Applied Glycoscience 50: 167 -171. Sharma, V. , J. V. Graeber, and V. Singh. 2006. Effect of hybrid variability on modified e-mill dry grind corn process 2006 ASABE Annual International Meeting, Portland, OR. Singh, V. , and J. V. Graeber. 2005. Effect of corn hybrid variability and planting location on dry grind ethanol production. Transactions of the ASAE 48: 709 -714. Wu, X. , R. Zhao, D. Wang, S. R. Bean, P. A. Seib, M. R. Tuinstra, M. Campbell, and A. O'Brien. 2006. Effects of amylose, corn protein, and corn fiber contents on production of ethanol from starch-rich media. Cereal Chemistry 83: 569 -575.