Starch Metabolism Of Wheat Grain

The subject matter below will cover the primary issues of starch metabolism as they are relevant to wheat and grain filling, but the reader is encouraged to explore any of several recent reviews which will provide greater detail on the genes, enzymes and mechanisms of starch synthesis in plants (27-33).

3.1. Starch biosynthesis

Starch granules are composed of two types of glucan polymer. One is amylose, consisting of predominantly linear chains of a( 1-4)-linked glucose residues, and the second is amylopectin, a highly branched glucan of a(l-4) and a(l-6) linkages. Amylopectin typically makes up 70-75% of the glucan in starch granules (34).

At present, there are four main classes of enzymes known to be involved in the biosynthesis of starch. The first, ADP-glucose pyrophosphorylase (AGP), is involved in the conversion of glucose-1-phosphate into ADP-glucose. The second class, starch synthase, catalyzes the polymerization of glucose residues derived from ADP-glucose into glucan. Branching enzyme (BE), the third type, is involved in formation of a(l-6) linkages present in amylose and amylopectin. The fourth class is debranching enzyme, whose members fall into the isoamylase or pullulanase family.

3.1.1. ADP-glucose pyrophosphorylase (AGP)

The reaction catalyzed by AGP is the first committed step in the synthesis of starch glucan. In plants, the AGP holoenzyme is a heterotetramer composed of small and large subunit polypeptides. The importance of AGP in starch synthesis has been demonstrated through genetic mutants (35-38) and transgenic plants (39). For example, in maize the absence of either subunit (i.e., bt2 or sh2 mutant) leads to roughly an 80% decrease of starch content in the endosperm, compared to normal genotype (40).

The profile of expression of AGP during wheat endosperm development has been examined on the mRNA and protein levels and compared to seed growth (41). The transcript and protein levels are low in early stages of endosperm formation, reach a peak in mid-development and then drop to lower levels by maturation. AGP enzyme activity follows a similar pattern (42), as does the rate of starch synthesis during grain development (43,44).

Multiple genes for AGP polypeptide have been cloned from wheat, with some encoding leaf and other endosperm proteins (45,46). Isolation and characterization of the cDNA encoding the small and large subunits of the endosperm was reported by Ainsworth et al. (42, 47). They demonstrated that the genes (Agpl and Agp2 for the small and large subunits, respectively) are expressed abundantly in whole grains and slightly less in leaves (42). Roots showed substantially less expression. Closer examination of the seed showed that expression is primarily in the endosperm, but some also occurs for both genes in the embryo and inner pericarp. Interestingly, there is differential expression of Agpl and Agp2 during endosperm development. Agpl transcript is detected in grain as young as 3 DAA, with substantial mRNA accumulation by 5 DAA. Agp2 is not expressed until 10 DAA. Ainsworth et al. (42) proposed that the low level of AGP enzyme activity detected in very young grain, occurring before linear starch deposition and before Agp2 expression, may be attributable to homotetramer function of the Agpl protein. It is possible that this low level of AGP activity participates in modulating a small pool of starch in the caxyopsis that is connected, to provision of carbon (C) for cell division of the endosperm.

AGP of most organisms is allosterically regulated by small molecules; activated by 3-phosphoglycerate (3-PGA) and inhibited by orthophosphate (Pi). In maize (48) and transgenic potato (49), the expression of unique allosteric variants of AGP increased kernel weight and enhanced starch production in tubers. AGP from the endosperm of wheat (46), as well as pea (50), Vicia faba (51), and barley (52-53) seed shows no, or insignificant, responses to 3-PGA or Pi. The significance of the allosteric properties of wheat AGP to starch production in the endosperm is not clear at this time. As suggested for barley, however, it may have to do with the cellular localization of particular subunits of the enzyme and whether ADP-glucose formed by action of the enzyme is localized in the cytosol or the plastid of the cell (54-56). Although the cellular localization of AGP subunits is currently not known for wheat, it would not be unexpected to find that part of the activity is localized in the cytosol, as is the case in maize, barley, and tomato fruit (57-60).

3.1.2. Starch synthases

The second major class of starch synthesizing enzymes, starch synthase (SS), is divided into isoform types on the basis of amino acid sequence. They are granule bound starch synthase I (GBSSI), SSI, SSII, and SSIII (61). The first indication of multiple starch synthases in developing wheat grain came from Rijven in 1984 (62), with more direct evidence occurring since then (63, 64). Studies have shown that some isoforms are localized exclusively with the starch granule while others occur with the granule and in soluble form, and still others occur exclusively in soluble form. Takaoka et al. (65) provided characterization of high molecular weight starch granule bound proteins of wheat and demonstrated that there are four proteins on the granule which are starch synthases. The genes for several of these have now been cloned. The wheat SSI gene is detected only in wheat endosperm and expressed during earlier stages of seed development between 6-21 DAA (66). Genes for SSII are expressed in leaves, pre-anthesis florets and the endosperm of wheat (67). Similar to wheat SSI, the SSII genes are expressed primarily between 8 and 18 DAA. The importance of SSII activity to starch synthesis is exemplified in several plant species by the altered granule morphology and amylopectin structure that occurs with mutant forms of the genes (68-70). The same is also true for wheat, as Yamamori (71) combined null alleles for the SSII genes and reported that large starch granules are mostly deformed and a novel starch with elevated iodine binding capacity was detected.

GBSSI, commonly called the waxy locus in cereal endosperm, controls amylose synthesis. The GBSSI protein in endosperm starch is found exclusively in starch granules and the wheat gene for this protein was cloned by Ainsworth et al. (47). Amylose is not synthesized in the presence of mutant GBSSI genes and starch low in amylose is desirable in some cases, such as for wheat flour used to make noodles. Nakamura and Yamamori (72)

were the first to produce amylose free wheat starch. The GBSSI genes of wheat Eire organized as a triplicate set of single copy homeoloci on chromosome arms 4AL, 7AS, and 7DS (47). The importance of each locus to the synthesis of amylose has been well characterized (73, 74), but little has been done to study the effect of GBSS mutants on agronomic aspects of wheat seed growth. Although the waxy mutation in maize is considered by most to have little impact on seed growth, waxy endosperm mutants of bread wheat produced by chemical mutagenesis showed that 1,000-grain weight may be slightly less than occurs in normal genotypes (75). Finally, it is important to note that a second GBSS isoform, which is related to endosperm GBSSI and appears to be responsible for amylose synthesis in the pericarp, aleurone layer and embryo of wheat, has been identified in immature seed of Tritacum monococum L. (76).

3.1.3. Branching enzymes

The third primary class of starch synthesizing enzymes are the branching enzymes. There are at least two types of branching enzymes, namely BEI and BEII, which encode genes that are grouped into different families based on amino acid sequence (77). Several wheat genes of the branching enzyme families have been cloned and the mRNA and protein expression profiles characterized (78-81). There appears to be approximately ten copies of BEItype genes in wheat (82). The BEI and BEII wheat genes reported to date are expressed primarily in the endosperm. At the transcript level, BEII is expressed at high levels during early stages of kernel development (5-10 DAA) and declines beyond 10 DAA, while BEI is more abundant midway through grain filling (10-15 DAA) and is undetectable in later stages (78, 79). Based on western analysis of protein levels, it appears that BEII proteins are expressed from 13-32 DAA, while the BEI isoforms are not detected until 18 DAA and are expressed later in development than BEII proteins (80). Although the exact stages donot coincide between studies reporting transcript and protein expression profiles, the reports agree that wheat BEs are differentially expressed and that BEII is on early, and BEI late, in endosperm development.

Lesions in certain BEII isoforms are known to negatively impact the amount of starch produced in maize (36), pea (83), and rice (84, 85) seeds. Wheat plants lacking BEII activity have not been reported, but others with a strong inhibition of BEI activity via anti-sense expression of the gene have been produced (86). Starch analysis revealed a slight reduction in amylopectin content and an alteration of other starch properties.

3.1.4. Other enzymes implicated in the control of starch synthesis

The importance of debranching enzymes in starch synthesis has been demonstrated for several species, including maize (87, 88), rice (89, 90), barley (91), Chlamydomonas (92) and Arabidopsis (93). Pullulanase and isoamylase (i.e., debranching) activities of wheat kernels have been described

(94) but there have not been any genes reported for the corresponding enzymes. It is reasonable to expect, however, that in the future genes of debranching enzymes specific to wheat will also be isolated and characterized for their involvement in starch biosynthesis and grain growth.

Amylogenin is a term used to describe a protein thought to serve as the priming mechanism for initiation of starch polymers. The concept is modeled after the protein, glycogenin, which has been shown to self glycosylate and prime glycogen synthesis in mammalian systems (95). Although amylogenin for priming starch synthesis in maize has been described (96), its direct involvement in the synthesis of starch was not demonstrated. In fact, since a close homolog of maize amylogenin has recently been cytolocalized to trans-Golgi dictyosomal cisternae in pea (97), the maize protein is more likely to be a component of hemicellulose biosynthesis, instead of starch. Nevertheless, proof of function for a protein such as amylogenin would offer vast possibilities for controlling the formation of starch granules within endosperm cells and, possibly, for improving grain yield of wheat.

Figure 1. Generalized pathway of carbohydrate metabolism leading to starch biosynthesis: AGP, ADP-glucose pyrophosphorylase; BE, branching enzyme; GBSS, granule bound starch synthase; De-BE, debranching enzyme; FK, fructokinase; GK, glucokinase; PFK, ATP-dependent phosphofructokinase; PFP, PPi-dependent phosphofructokinase; PGI, phosphoglucoisomerase; PGM, phosphoglucomutase; SucSyn, sucrose synthase; SS, starch synthase; UGP, UDP-glucose pyrophosphorylase.

Degradation And Biosynthesis Starch

Figure 1. Generalized pathway of carbohydrate metabolism leading to starch biosynthesis: AGP, ADP-glucose pyrophosphorylase; BE, branching enzyme; GBSS, granule bound starch synthase; De-BE, debranching enzyme; FK, fructokinase; GK, glucokinase; PFK, ATP-dependent phosphofructokinase; PFP, PPi-dependent phosphofructokinase; PGI, phosphoglucoisomerase; PGM, phosphoglucomutase; SucSyn, sucrose synthase; SS, starch synthase; UGP, UDP-glucose pyrophosphorylase.

As shown in Figure 1, there are a number of other enzymes involved in carbohydrate metabolism which reside outside, and sometimes also inside, the plastid which could significantly impact the production of starch in the endosperm. One of these is phosphoglucomutase (PGM). Multiple forms of PGM have been found in immature wheat endosperm (98) and it is known that PGM is found both in the cytosol and the amyloplast. The exact role of PGM in regulating starch biosynthesis is still confusing. Kumar and Singh (99) suggested on the basis of metabolite measurements that PGM may regulate starch biosynthesis by controlling the rate of glucose-1-phosphate formation in developing wheat. In support of this proposal, mutation of a PGM locus in pea leads to a severe reduction of starch biosynthesis in the seed (100). Other work, however, again with wheat grains, shows that even though PGM enzyme activity declines with advancing age of the seed, the activity is far in excess of measured rates of starch biosynthesis at all stages (101). Tetlow et al. (102) proposed on the basis of studies involving in vitro starch synthesis that PGM may be a sensitive point in controlling plastidial partitioning of C into storage (i.e., starch) or oxidation (i.e., pentose phosphate pathway).

Another extraplastidial enzyme suggested by some to control starch synthesis in wheat is sucrose synthase. With sucrose being the primary assimilate transported into endosperm cells (103), sucrose synthase is a pivotal enzyme in the utilization of the sugar. The profile of sucrose synthase activity in developing wheat is low initially, reaches a peak during mid-development, and subsequently declines as the grain matures (104, 105). Most of the activity is localized in the endosperm and it is possible that the duration of enzyme activity may be important in determining the duration of grain filling (106). A correspondence of sucrose synthase activity to kernel weight in wheat has been observed and suggests that this enzyme may be an important component contributing to differences in kernel weight (107). More recently, Rifkin et al. (108) compared changes in sucrose synthase and invertase activities throughout endosperm development of wheat and concluded that invertase was not the preferred pathway for sucrose catabolism. Estimates of UDP and sucrose concentrations suggested that sucrose synthase may not achieve its potential maximum velocity and, in turn, could restrict the flux of C into starch.

Many studies have been done where detached ears of wheat or slices of immature endosperm have been incubated in various media to examine the effect of manipulating assimilate availability on starch synthesis. These studies, along with others, provide an indication of which enzymes in starch metabolism may be involved in controlling overall deposition. For instance, developing wheat heads were provided varying levels of sucrose and the amount of starch deposited in the endosperm examined along with levels of enzyme activity (109). The activity of SS was highly correlated with the amount of starch made in the grain, but the relationship of starch deposition to phosphoglucoisomerase activity was weaker and nonexistent with relation to AGP. In contrast, a whole-plant comparison of seed enzyme activities among cultivars of different grain size indicated that UDP-glucose and ADP-glucose pyrophosphorylases might be considered rate-limiting steps in wheat starch accumulation (110). Further confusing is work by Mengel and Judel (14) which, by comparing starch accumulation and enzyme activities across development, suggests that phosphorylase is of substantial importance in grain starch synthesis.

Also using detached wheat ears, Randhawa and Singh (111) recently made the curious observation that a-ketoglutarate and citrate provided in the medium increased the starch content of the grain. Inorganic phosphate level in the medium had no impact on starch or protein accumulation. This supports the fact that wheat AGP, unlike AGP from many other sources, is not allosterically inhibited by Pi. A similar response of starch synthesis to phosphate has also been noted by Jenner (112) in studies involving wheat endosperm slices. Rijven and Gifford (113) concluded similarly that natural variation of Pi concentration in the apoplast of wheat cells is unlikely to be of significance in the regulation of the rate of starch synthesis in the grain. Instead, they found that exogenous alkali ions stimulated the conversion of sucrose to starch. Specifically, SS activity was stimulated by potassium.

3.2. Catabolism of starch during seed development

It is likely that some of the starch occurring within the endosperm of developing wheat seed is broken down during grain filling. The enzymes which could be involved in this mechanism are a-amylase and phosphorylase. (3-amylase is unlikely to participate because the protein lacks an N-terminal transit peptide sequence needed for localization to the amyloplast (114). In general, it is believed that a-amylase activity does not occur in the endosperm, except perhaps at very late stages of development (115, 116). However, Carmen et al. (117) provide evidence in their study of young developing seed, that maltose found in the endosperm may be derived from hydrolysis of starch which occurred in endosperm cells near the growing embryo. They suggest that the hydrolase could originate from the embryo or from the modified endosperm cells of this region. Others have also shown that simultaneous synthesis and degradation of starch can occur in plant parts (118, 119). McGregor and Dushnicky (120) provided microscopic evidence that starch granules of endosperm cells adjacent to the crushed cell layer of the endosperm/embiyo interface are hydrolyzed as part of the formation of the crushed cell layer in barley.

Starch degradation is not only likely in the endosperm, but also in the embryo. During the latter stages of embryo development, a-amylase activity begins to increase while the starch content of the embryo is on the decline (5).

The role of phosphorylase in starch degradation within the endosperm is more speculative. However, it is interesting that the profile of activity for this enzyme is similar to that of many biosynthetic enzymes involved in the synthesis of starch in the seed (115).

3.3. Intracellular transport of metabolites used in starch biosynthesis

Not only is it important that the metabolism of endosperm cells provide assimilate for starch biosynthesis, it is also necessary that the precursors occur in the amyloplast where the granule is formed. The subcellular location of ADP-glucose formation has ramifications to whether the substrate is immediately available to starch synthesizing enzymes in the plastid or whether the metabolite must be transported. As stated earlier, there is growing evidence that a portion of AGP enzyme activity is located outside the plastid in many plants. If the same is true for wheat, one would expect the presence of an adenylate transporter to occur in the amyloplast membrane. The brittle-1 protein has been shown to transport ADP-glucose into the amyloplast of maize endosperm cells (121), but evidence to date does not demonstrate the presence of a homolog in wheat endosperm. Cao and Shannon (122), using an antibody raised against a 56 amino acid carboxy terminal portion of brittle-1, could not detect cross-reacting protein in western analysis of seed protein from a series of cereals, including wheat.

Although earlier reports suggested that triose phosphates are transported into the amyloplast of wheat, today several lines of evidence argue against this proposal. In 1988, Tyson and Ap-Rees (123) demonstrated that amyloplasts isolated from wheat endosperm only used glucose-1-phosphate in the synthesis of starch, whereas other C-6 and C-3 compounds did not serve as precursors. In the same year, experiments feeding 13C-labeled sugars to developing wheat endosperm proved that there was little redistribution of 13C between C atoms 1 and 6 of asymmetrically labeled glucose (124). These data seriously weakened the argument for the selective uptake of triose phosphates by the amyloplast as part of the pathway for starch biosynthesis in wheat. Furthermore, the amyloplast of wheat endosperm lacks fructose-1,6-bisphosphatase, an enzyme that would be needed to help channel triose phosphates into hexose monophosphate for conversion into ADP-glucose and starch (125). Therefore, hexose phosphate, not triose phosphate, must be imported into the amyloplast. Recent studies of feeding wheat amyloplasts with 14C-labeled hexose phosphates confirmed that, in the presence ATP, only glucose-1-phosphate is converted into starch (102).

At this point, it seems clear for wheat that glucose-1-phosphate is the form of hexose transported into the amyloplast for use in grain starch synthesis. If AGP occurs in the cytosol, then it is likely that a transporter of ADP-glucose will be identified in wheat, as it has been in maize. If it turns out that AGP is primarily located in the plastid of wheat endosperm, then a plastidic ATP/ADP-transporter similar to that found in Arabidopsis (126) and in potatoes (127) may be identified in wheat as well. Its role would be to supply ATP for the production of ADP-glucose in the amyloplast. Regardless of whether either a plastid transporter of ADP-glucose or ATP/ADP occurs in wheat, the importance of these carriers in maize (121) and potato (127), respectively, suggests that starch deposition in wheat endosperm may also be tightly controlled by the presence and activity of similar proteins.

Was this article helpful?

+1 0


  • amelia
    Is grain matabolised as a starch?
    3 years ago

Post a comment