Overexpression of serine acetyltransferase in maize leaves increases seed-specific methionine-rich zeins

Abstract

Maize kernels do not contain enough of the essential sulphur-amino acid methionine (Met) to serve as a complete diet for animals, even though maize has the genetic capacity to store Met in kernels. Prior studies indicated that the availability of the sulphur (S)-amino acids may limit their incorporation into seed storage proteins. Serine acetyltransferase (SAT) is a key control point for S-assimilation leading to Cys and Met biosynthesis, and SAT overexpression is known to enhance S-assimilation without negative impact on plant growth. Therefore, we overexpressed Arabidopsis thaliana AtSAT1 in maize under control of the leaf bundle sheath cell-specific rbcS1 promoter to determine the impact on seed storage protein expression. The transgenic events exhibited up to 12-fold higher SAT activity without negative impact on growth. S-assimilation was increased in the leaves of SAT overexpressing plants, followed by higher levels of storage protein mRNA and storage proteins, particularly the 10-kDa δ-zein, during endosperm development. This zein is known to impact the level of Met stored in kernels. The elite event with the highest expression of AtSAT1 showed 1.40-fold increase in kernel Met. When fed to chickens, transgenic AtSAT1 kernels significantly increased growth rate compared with the parent maize line. The result demonstrates the efficacy of increasing maize nutritional value by SAT overexpression without apparent yield loss. Maternal overexpression of SAT in vegetative tissues was necessary for high-Met zein accumulation. Moreover, SAT overcomes the shortage of S-amino acids that limits the expression and accumulation of high-Met zeins during kernel development.

Keywords: methionine; nutritional quality; seed storage protein; serine acetyltransferase; transgenic; zein.

Figure 1

AtSAT1 transformation of maize. (a) Schematic diagram of the SAT overexpression construct. The construct components include the T‐DNA right border, RB; and left border, LB; the bundle sheath cell‐specific Rubisco small subunit 1 promoter, rbcS1 promoter; the A. thaliana serine acetyltransferase1 coding sequence, AtSAT1; the CaMV 35S terminator, T35S; and the phosphinothricin acetyltransferase cassette consisting of the 35S promoter; tobacco etch virus translational enhancer, TEV; the bar gene; and the soya bean vegetative storage protein terminator, Tvsp. (b) PCR confirmation of T₁ transformants. The DNA templates used for PCR amplification include the vector plasmid, or genomic DNA from each of nine transformants or the nontransformed line BXA HII, which is the maize line used for transformation.

Figure 2

Maize plant expression of AtSAT1. (a) Quantitative RT‐PCR of two independent transgenic lines OE1 and OE3. Both OE1 and OE3 for this and all subsequent experiments, unless noted otherwise, were the result of two backcrosses to the maize inbred line B73 and then selection of nonsegregating plants. For this reason, B73 was used as the nontransgenic control. RNA was extracted from young leaves of two‐month‐old plants, and AtSAT1 was amplified with specific primers. The Actin primers were used as reference gene control. (b) SAT activity in the leaves of OE1 and OE3. The data in graphs (a) and (b) represent the mean of three measurements from different plant samples±SD. The specific activity of crude extracts is given in nmol CoA produced per min and mg total protein. Asterisks indicate significant differences between B73 and transgenic plant lines using the one‐way ANOVA function of GraphPad Prim (P < 0.001).

Figure 3

Sulphur metabolites in AtSAT1 transgenic maize. Sulphur metabolites in leaves of 2‐month‐old transgenic AtSAT1 lines OE1, OE3 and parental B73. (a) The level of free Cys. (b) The level of total Cys. (c) The level of free Met. (d) The level of total Met. (e) The level of total glutathione. (f) The level of OAS. The values are the mean of three independent biological replicates ±SD. Asterisks indicate significant differences from B73 (Student's t‐test, P < 0.05).

Figure 4

Sulphate reduction in AtSAT1 transgenic maize. APR activity and sulphite in leaves of 2‐month‐old transgenic AtSAT1 lines OE1, OE3 and parental B73. (a) ZmAPR enzyme activity. (b) Sulphite content. Samples were from T₁ transgenic maize lines OE1 and OE3. The values are the mean of three independent biological replicates ±SD. Asterisks indicate significant differences from B73 (Student's t‐test, P < 0.05).

Figure 5

Zein accumulation in transgenic kernels. (a) Kernels from OE1 and OE3 were harvest from field plants. The kernels were fully mature, and protein profiles from three different kernels harvested from different plants are shown. Protein from 300 μg dry weight of endosperm sample was loaded in each lane. M, protein markers from top to bottom being 250, 150, 100, 75, 50, 37, 25, 20, 15 and 10 kDa. The mass of each zein is indicated to the right of the figure. (b) The relative abundance of zein proteins was analysed from SDS‐PAGE analysis from three different kernels harvested from different plants using the densitometry function of ImageJ software. The band intensities were normalized using the 22‐kDa zein and the results plotted as relative values ±SD. Asterisks indicate significant differences from B73 (Student's t‐test, P < 0.05).

Figure 6

Met content measured with a bacterial biosensor. The Met content of maize zein and nonzein proteins was measured from B73 and plants derived from a segregating population of OE1. Each value is the mean of three individual plants ±SD. Asterisks indicate significant differences from B73 (Student's t‐test, P < 0.05).

Figure 7

Chicken feeding trial with OE1. (a) Growth rate of chicks fed a diet formulated with OE1 kernels or kernels from B73 segregant plants. The data shown are the means from twenty‐one chicks ±SD. Asterisks indicate significant differences in OE1 compared with B73 (Student's t‐test, P < 0.01). (b,c) Close‐up image of 21‐day‐old primary wing feathers.

Figure 8

Maternal AtSAT1 is necessary for accumulation of 10‐kDa δ‐, 15‐kDa β‐ and 18 kDa δ‐zeins. (a) SDS‐PAGE analysis of zein proteins in endosperm of B73xOE1, OE1xB73 and self‐pollinated OE1. The samples were harvested at 18 DAP, 25 DAP and 35 DAP. Endosperm was isolated, and zein was extracted. Total zein loaded in each lane was equal to 300 μg of maize fresh weight. (b) Zein mRNA was measured in endosperm sampled 18‐DAP by quantitative RT‐PCR. Relative expression of zein genes in OE1 compared to B73. The data are the mean from three independent plants per transgenic line ±SD. Asterisks indicate significant differences from B73 (Student's t‐test, P < 0.01).

Figure 9

Zein profiles in hybrid kernels from a cross of OE1 with 10‐kDa δ‐zein overexpression maize line Dzs10^oe. (a) Comparison of SDS‐PAGE zein profiles for Dzs10^oe, OE1 and kernels from a OE1xDzs10^oe cross. The presence of the respective transgene is indicated under the gel. The zein profiles from two representative kernels are shown. (b) Band intensity of the 10‐kDa δ‐zein in the respective genotypes. The band intensity was measured by densitometry and normalized based on the amount of protein loaded on the gel. Each value is the average ±SD of measurements from six to eight kernels. Dzs10^oe is a transgenic maize line that overproduces the 10‐kDa δ‐zein, described in Lai and Messing (2002). The transgenic construct contains the Dzs10 coding sequence, but not the 5′ or 3′ UTRs. The UTRs are the targets of a post‐transcriptional regulation mechanism that controls the level of Dzs10 mRNA. Lacking the control targets 10‐kDa δ‐zein is overproduced (Lai and Messing, 2002).