|Home | About | Journals | Submit | Contact Us | Français|
Carbohydrate temporarily accumulates in wheat stems during the early reproductive growth phase, predominantly as water soluble carbohydrate (WSC), and is subsequently remobilized during grain filling. Starch has also been reported as a minor storage carbohydrate component in wheat stems, but the details are lacking.
The accumulation and localization of starch in wheat stem and leaf sheath tissue over a developmental period from 6 d before anthesis to 35 d after anthesis was investigated.
The region of the peduncle enclosed by the flag-leaf sheath, and the penultimate internode were the main tissues identified as containing starch, in which the starch grains localized to the storage parenchyma cells. In contrast, the exposed peduncle lacked starch grains. Starch grains were also found in the flag-leaf and second-leaf sheath. Plants grown in low-nitrogen conditions exhibited increased storage of both starch and WSC compared with plants grown in high-nitrogen supply.
The major accumulation and decrease of starch occurred temporally independently to that for WSC, suggesting a different functional role for starch in wheat stems. Starch reutilization concomitant with peduncle growth, and the early development of the reproductive structures, suggested a role in provision of energy and/or carbon scaffolds for these growth processes.
Plants utilize two fundamentally different strategies for storage of carbohydrate in vegetative tissues. Either starch, or sucrose or sucrose derivatives, such as fructosyl-oligosaccharides (fructans), are the principal carbohydrates involved. While insoluble starch is stored in plastids, sucrose and fructans are generally soluble and accumulate in vacuoles. In most species, starch is predominantly utilized for diurnal carbon storage in leaves. An earlier viewpoint was that starch accumulates in leaves as carbon overflow from excess photosynthate, but more recent studies have shown that starch synthesis is finely regulated to provide sufficient carbohydrate for utilization through the dark period (Smith and Stitt, 2007). On the other hand, temperate cereal leaves, like wheat and barley, for example, diurnally accumulate mainly sucrose, and also to a lesser extent starch (Trevanion, 2000; Trevanion et al., 2004), but when assimilate supply is in excess of demands for growth, in the cold, or under limited N supply, fructan accumulation is initiated (Wang and Tillberg, 1996; Wang et al., 2000). Regulatory mechanisms which control sucrose and starch synthesis in leaves, or starch synthesis in heterotrophic sink tissues such as potato tubers (Tiessen et al., 2002, 2003), have been elucidated. However, it is unclear what governs which particular storage carbohydrate is utilized in a particular organ or plant, although some suggestions have been made to explain why fructan may be beneficial over starch under particular circumstances (Pollock and Cairns, 1991).
During the vegetative and early reproductive phases of cereal development, assimilated carbon is temporarily stored as carbohydrate in vegetative sink tissues such as the stem and leaf sheaths. The temporary carbohydrate reserves are subsequently remobilized for transport to reproductive sink tissue, the filling grain, during later stages of the plant's development. In the temperate cereal species wheat and barley, water soluble carbohydrates (WSC), consisting principally of fructan and sucrose, are the major form of carbohydrate storage in the stem, and starch accumulation has rarely been considered. In contrast, in other species, for example the tropical cereal species rice, which is reported to lack fructan accumulation enzymes (Ji et al., 2007), carbohydrate accumulates in stems as insoluble starch grains. These reserves of carbohydrate are important contributors to grain filling and become increasingly so if the plants are subjected to stress at later stages of development. In wheat it has been reported that pre-anthesis reserves contribute between 8 and 27 % of the carbon in carbohydrates and 30 and 47 % of the carbon in proteins of the grain (Gebbing and Schnyder, 1999).
Although fructan is the main form in which carbohydrate is stored in wheat stems, trace amounts of starch have been reported (Schnyder, 1993, and references therein). Judel and Mengel (1982) measured starch levels in culms, including leaf sheaths, from 0 to 51 d after anthesis (daa) and found that starch levels increased to maximal levels at 12 daa and declined thereafter. Shading resulted in a reduction to approximately half of the levels of accumulated starch. From measurements at 0–25 daa, Kiniry (1993) reported that starch in wheat stems was not remobilized during shading. Sild et al. (1999) reported measurements of starch in whole stems including the leaf sheaths from 35 d before anthesis (dba) to 47 daa, showing low levels of starch throughout the time course that did not alter significantly in normal or elevated carbon dioxide treatments. In other vegetative sink tissues a low level of starch storage was also reported, such as wheat leaves (Judel and Mengel, 1982). These authors showed that leaf starch content peaked at 12 daa after which levels declined during the remainder of the grain-filling period. More recently, transcripts for granule-bound starch synthases have been reported to be expressed in wheat stem (Vrinten and Nakamura, 2000; Peng et al., 2001). Notably, these studies did not examine the localization of starch within different parts of the stem, nor provide information regarding the cellular localization of the starch in stems, and as yet no suggestions have been made to explain why starch is present in this tissue.
One environmental factor that strongly affects carbon metabolism is nitrogen (N) availability. N deficiency is known to cause accumulation of carbon in plants, usually in the form of starch in leaves (Scheible et al., 1997) and in barley, N-limited tissues have been shown to accumulate high amounts of fructan and other soluble carbohydrates (Wang and Tillberg, 1996; Wang et al., 2000). The effects of long-term N limitation on wheat have been examined recently (Ruuska et al., 2008), and it was shown that high amounts of WSC already accumulated in most vegetative tissues by the time of anthesis, i.e. prior to the grain-filling phase. Interestingly, expression profiling revealed that, in addition to a suite of fructosyltransferase genes whose expression was enhanced in many low-N tissues, some starch metabolism-related transcripts were also more abundant, such as those encoding a putative glucose 6-phosphate/phosphate translocator and ADP-glucose pyrophosphorylase large subunit (Ruuska et al., 2008). Preliminary data indicated that starch grains were present in stems of low-N-grown plants but absent from those grown in high-N treatments.
In the current work, carbohydrate storage in the form of starch grains has been examined in stems of wheat during a developmental time course through flowering and grain filling. The predominant starch-storing tissues within stems were identified, and the cellular localization of the starch grains within these tissues was determined. Both starch and WSC were quantified in normal growth conditions and when growth was severely limited by N availability, i.e. when the plants are known to accumulate large amounts of fructans. From this detailed profile of the temporal and spatial accumulation of starch in wheat stems, potential roles for starch storage and utilization in this context are discussed.
Wheat plants (Triticum aestivum ‘Janz’) were grown to maturity in glasshouse conditions under natural light at 25 °C/16 °C day/night time temperatures. Seeds were sown in 9l pots containing soil/compost mix. A second population of plants was grown in similar environmental conditions as above, but in either high (8 mm nitrate)- or low (1 mm nitrate)-N conditions as described in Ruuska et al. (2008).
Samples were collected from both main stems and primary tillers from the plants at approx. 7 -d intervals over a developmental time course from 6 dba to 35 daa. For sampling, the vegetative sink tissues, consisting of flag-leaf blade, flag-leaf sheath, exposed peduncle, enclosed peduncle, penultimate internode, second-leaf sheath (defined here as the sheath of the next lowest leaf below the flag leaf), as well as lower leaves, and lower stem parts were collected separately and were snap frozen in liquid nitrogen and stored at −80 °C. Samples were collected between 1100 h and 1400 h. Measurements of the length of the exposed and enclosed regions of the peduncle and remainder of the stem were made in a population of plants grown in identical conditions. Similar length measurements were also made for a population of plants grown in either high- or low-N (HN or LN, respectively) treatments. To determine the profile of starch storage in the enclosed peduncle, three individual representative segments were collected from the upper, mid- and lower regions of the tissue for each of the time points. Samples of each segment were collected for individual starch localization in sections, and the remainders of the three segments for each region were then pooled together for starch, WSC and chlorophyll (Chl) determinations.
Stem samples, previously frozen and stored at −80 °C, were ground to a fine powder in a mortar and pestle using liquid nitrogen and were then dried by lyophilization. Aliquots of between 15 mg and 30 mg of the powder were extracted using 80 % (v:v) ethanol at 80 °C for 40 min. The supernatants were collected, by centrifugation at 13000 g for 5 min, and the pellets were subsequently re-extracted a further three times. The supernatants were combined for each sample and used for the WSC and Chl determinations described below. The pellets were used for starch determinations based on earlier procedures (Lunn and Hatch, 1995).
The total amount of WSC was measured in each supernatant using the anthrone method described by Yemm and Willis (1954), adapted to allow measurement in a microtitre plate against a fructose standard curve, with generally <5 % difference between replicates. The Chl content of the samples was determined using 1-mL aliquots of the supernatant samples. A wavelength scan (700–600 nm) was performed on two of the extracts to confirm the wavelengths of the maximal absorbance peaks for Chl a and Chl b (664 and 649 nm) using a GBC UV/VIS 920 spectrophotometer. Absorbance measurements were made for each sample at the two wavelengths. To calculate Chl content, the formula (6·1 × A664 nm) + (20·04 × A649 nm) = Chl content (μg mL−1) was used (Wintermans and De Mots, 1965).
The pellets resulting from the 80 % ethanol extractions were each resuspended in 1·7 mL of 0·2 m KOH and were placed in a boiling water bath for 30 min. Twenty microlitres of acetic acid were added to neutralize the extract. To measure starch content, triplicate 200-μL aliquots of the KOH-treated suspensions for each sample were placed in separate Eppendorf tubes to which 200 µL of 50 mm sodium acetate (pH 4·8) containing 0·5 units of amyloglucosidase (EC 3·2·1·3) and 1 unit of α-amylase (EC 3·2·1·1) were added and the samples were incubated at 50 °C for 15 h. The suspension was centrifuged (13000 g for 5 min) and duplicate 50-μL samples of each supernatant were used to measure glucose by spectrophotometric assay (Campbell et al., 1999) using a Spectromax 340PC plate reader and Softmax Pro 3·0 software (Molecular Devices).
Tissue samples were removed from storage at –80 °C and were transferred to glass vials containing 80 % (v:v) ethanol, 20 % (v:v) buffer (100 mM Tris–HCl (pH 7·5), 150 mm NaCl). The samples were incubated in several changes of 80 % ethanol at 4 °C for removal of Chl, and were subsequently fully dehydrated to 100 % ethanol before re-hydrating by passing through a decreasing ethanol series. The samples were collected in buffer solution. Thin hand sections were cut using double edged razor blades and were collected into multiwell glass microscope slides. To localize the starch grains the sections were stained with iodine solution (0·5 % w:v iodine in 5·0 % w:v potassium iodide) for 20 min and then briefly rinsed twice in buffer. The sections were examined using a Leica DMR microscope and representative digital images were collected using a Leica DC500 digital camera system. Sections were scored for presence or absence of starch grains in the parenchyma cells; six to eight sections from each of three replicate samples, for each tissue at each time point, were examined in estimating the starch grain abundance.
To determine the cellular localization of starch grains within stem tissues more precisely, representative samples of the tissues were fixed, dehydrated and embedded in paraffin wax following the method described in Scofield et al. (2007a). Thin sections, 10 µm in thickness, were cut using a Microm HM350 microtome and were collected onto polysine-coated slides as described in Scofield et al. (2007a). The sections were de-waxed, re-hydrated through a graded ethanol series and washed in buffer prior to staining for starch with iodine solution and imaged as described above.
Figure 1 diagrammatically illustrates the stem tissues examined and the developmental changes that were observed around anthesis. At about 6 dba the head was fully emerged, whereas the entire peduncle was still enclosed by the flag-leaf sheath. The peduncle was entering a rapid elongation phase, with growth occurring at the base of the internode, while the lower part of the stem had already attained its maximal length. Chl was evident in the peduncle tissue but in a decreasing amount towards the base. At anthesis the peduncle had elongated considerably and part of it was exposed as it emerged from the flag-leaf sheath. Between anthesis and 9 daa the peduncle reached its maximal length after which there was no further stem growth in the plant.
The changes in peduncle and stem length were determined in wheat plants grown in soil in the glasshouse under identical conditions to those sampled for starch localization (Fig. 2). The total peduncle length rapidly increased at the start of the time course, reaching maximal length at about 3–4 daa (Fig. 2A). The flag-leaf sheath showed negligible change in length during this period, hence the lower region of the peduncle that was enclosed by it remained constant. The lower parts of the stem remained unchanged in length during the time course (Fig. 2B). Peduncle length measurements were also made in populations of wheat plants grown in either LN or HN treatments (Fig. 2C, D). The temporal pattern of peduncle growth was similar to that observed in soil growth conditions, with maximal peduncle length reached at about 4 daa for LN plants and 2 daa for HN plants. Total peduncle length was greater in plants grown in HN treatment compared with those grown in soil or in an LN treatment. The lower part of the stem had reached maximal length by 6 dba for both LN- and HN-treated plants (data not shown).
Transverse sections of various wheat stem and leaf sheath tissues collected throughout the developmental time course were stained with iodine solution and microscopically examined for starch grains (summarized in Table 1). For each observation, six to eight sections from each of three replicates were examined in estimating the starch-grain abundance. Starch grains were abundant in the enclosed region of the peduncle at 6 dba and 2 daa, absent at 9 daa, sparsely present at 15 and 18 daa, and then absent for the remainder of the time course. In contrast, starch grains were absent from the exposed peduncle at all the time points. In the second internodal region (the penultimate internode, directly below the peduncle), starch grains were present in relatively large numbers at 6 dba and 2 daa, sparsely present at 18 daa, and then were absent for the later 28- and 35-daa samples. Flag-leaf and second-leaf sheath, also examined throughout the time course, contained starch grains at low abundance compared with the enclosed peduncle at 6 dba, 2 and 9 daa, and at extremely low abundance at 15 daa, then starch was absent from the parenchyma at subsequent time points. Chloroplastic starch was observed in Chl-containing stem and sheath tissues, but its analysis was not further pursued.
Figure 3 illustrates the cellular localization of starch grains within stem and sheath tissues at 6 daa. In the enclosed region of the peduncle (Fig. 3A) starch grains (excluding chloroplastic starch grains) were found to localize exclusively to the parenchyma cells, and tended to be present in the parenchyma cells near to the inner and outer vascular bundles and immediately adjacent to the cortical fibre cell layer, but did not appear in the bundle sheath cells. There was a decreasing presence of starch grains towards the pith. In contrast, the exposed region of the peduncle (Fig. 3B) lacked starch grains in the parenchyma cells. The enclosed region of the peduncle did not contain much chlorenchyma tissue compared with the exposed region of the peduncle, and the low amount of chlorenchyma was present in a decreasing gradient from the upper region, closest to the exposed peduncle, to the lowest region near the node (result not shown). In the penultimate internode (Fig. 3C), starch grains localized to the parenchyma cells in a similar way to that observed for the enclosed peduncle. Transverse sections of both the flag-leaf sheath (Fig. 3D) and second-leaf sheath (data not shown) were also examined for starch localization. The starch grains were present at relatively low abundance, compared with those observed in enclosed peduncle tissue, and were located in the parenchyma cells below the alternating large and small vascular bundles within the sheath, but appeared to be more prevalent with the large vascular bundles. Similar patterns of cellular starch grain localization were also observed in another hexaploid wheat variety, Hartog, and in a tetraploid wheat Triticum turgidum ‘Fransawi’ (G. N. Scofield and C. L. D. Jenkins, unpubl. res.).
Figure 4 illustrates changes in the relative abundance of starch grains in transverse, hand-cut sections of the mid-region of the enclosed peduncle collected through the period around anthesis. At 6 dba (Fig. 4A) and 2 daa (Fig. 4B) a high number of starch grains were observed in the parenchyma cells surrounding the inner vascular bundles and in the region adjacent to the outer vascular bundles. However, by 9 daa (Fig. 4C) no starch grains were evident in the sections, indicating that the starch had been largely reutilized by this time point. At 15 daa, a low abundance of starch grains was again observed (Fig. 4D), suggesting that further accumulation may have occurred.
The peduncle was examined in greater detail to determine the developmental and spatial profile for starch accumulation in the enclosed region. Three replicate samples from each of the upper, middle and lower region of the enclosed peduncle were collected at each time point up to 21 daa. A small portion of each sample was used for microscopic starch localization and the remainder of each of the three replicates were pooled together for extraction, and starch, WSC, and Chl contents were determined (Fig. 5). Starch content was highest in the lower region of the enclosed peduncle at 6 dba and 2 daa. Lower levels of starch were measured in the upper and middle regions at these time-points. At 9 daa, starch was not present in measurable amounts in any region of the enclosed peduncle. Very low levels of starch were detected in all regions at 15 daa, and in the middle and lower regions at 18 daa. After this, starch was below measurable limits in all regions for the remainder of the time course. Starch grain abundance in the individual tissue pieces used for microscopic localization (data not shown) generally reflected the starch content measurements in the pooled samples. However, one exception was in the upper region at 6 dba. Measurable amounts of starch were present in this sample, but no starch grains were present in the parenchyma cells, indicating that the starch measured in this and the other upper regions is likely to be due to chloroplastic starch. The WSC content was lowest in each of the three regions during the earlier time points, and increased to maximal levels at 15–18 daa, after which levels started to decline. WSC was always greater in the middle and lower regions of the enclosed peduncle compared with the upper region at comparable time points. Maximum levels of starch accumulated were very low compared with WSC, representing about 2 % of the WSC accumulated by the corresponding time. Chl content was greatest in the upper part of the enclosed peduncle and decreased towards the bottom of the peduncle at all time points.
Samples from the mid-region of the penultimate internode were similarly analysed for starch localization and content, WSC and Chl (Fig. 6). The microscopic localization of starch grains in the parenchyma cells (data not shown) corresponded closely with the starch content measurements. Starch content was greatest during the early part of the time course, peaking at about 2 daa and decreasing thereafter (Fig. 6). WSC content was low during the early part of the time course, increasing to maximal levels by 18 daa and remaining at relatively high levels for the remainder. In this tissue, the starch content at its maximum represented only up to 0·7 % of the WSC accumulated. Chl content declined through the time course.
Tissues from plants at anthesis, grown in either LN or HN treatments, were examined for starch localization by microscopy (results not shown). In the exposed peduncle samples from either treatment, no starch grains were evident in the parenchyma cells. Starch grains were present, however, in the enclosed peduncle in both treatments, but there were clearly more in the LN treatment. Similarly, starch grains were observed in the penultimate internode of LN plant samples, but were absent from those grown in the HN treatment. Starch grains were also present in the parenchyma of the second-leaf sheath in LN samples but absent in HN samples.
The enclosed peduncles from plants grown in LN and HN treatments were harvested from plants at anthesis and divided into upper, middle and lower regions to obtain profiles for starch localization and content, WSC and Chl contents (Fig. 7). Again triplicate samples were collected, each sampled individually for starch localization and the remainders pooled for biochemical assays. Starch content was relatively high in all regions of the enclosed peduncle from the LN treatment and increased in amount towards the lower region. In contrast, in samples from the HN treatment relatively low or undetectable starch levels were determined. Starch localization in the parenchyma cells, from microscopic examination, strongly correlated with the starch content measurements, and was similar to that observed in the enclosed peduncle of soil-grown plants, i.e. predominantly in the parenchyma cells around the vascular bundles (data not shown, but see Fig. 3A). Although starch grains were abundant in LN stem tissues, the amount was very low compared with the total WSC, representing only about a maximum of 2·5 % of total stem extractable carbohydrates. In LN samples WSC content increased in level from the upper region to the lower region of the enclosed peduncle and, whilst samples from HN-treated plants exhibited a similar trend, the amounts were significantly lower than those present in the LN samples (Fig. 7). Chl content decreased in amount from the upper to the lower region of the enclosed peduncle in a similar fashion for samples from both LN and HN treatments (Fig. 7).
Starch, Chl and WSC were also determined in the penultimate internode, and in the lower stem sections for the LN- and HN-treated plants at anthesis (data not shown). Starch was present in these tissues for the LN treatment, but at lower levels (approx. 1·4 mg g−1 d. wt) than in the enclosed peduncle. Starch was undetectable in the HN treatment. At LN, WSC concentrations in these tissues approached those of the enclosed peduncle, whereas at HN the concentrations were much lower. Chl values were very low at both LN and HN in the lower stem parts.
Accumulation of WSC, principally sucrose and fructan, occurs in stem tissues of temperate cereals during early reproductive development, reaching a maximum about 2–4 weeks after anthesis, and is subsequently remobilized to provide carbon for grain filling (Kiniry, 1993; Schnyder, 1993; Sild et al., 1999; Gebbing, 2003). In the present study, the WSC accumulated approximately from anthesis to a maximum at 15–18 daa, then declined, indicating some remobilization at later time points (Figs 5 and and6),6), and mostly was localized within the sheath-enclosed peduncle and penultimate internode (results not shown). In this study, the accumulation and utilization of starch, another carbon storage component, in these vegetative sink tissues of wheat was investigated over the reproductive development period, from 6 dba to 35 daa. Earlier studies on the occurrence of starch in cereal stems have not been detailed (see Introduction). The present results substantially extend the earlier work in showing the timing, localization and extent of starch accumulation and reutilization in stem tissues.
In the enclosed peduncle and penultimate internode, the time of maximum WSC accumulation was around 15–18 daa. However, it is notable that the major starch accumulation occurred in these tissues early in the time course, from before head emergence until just after anthesis, temporally independently of WSC accumulation and remobilization. Starch was present in the enclosed peduncle at early time points, but absent from the exposed peduncle throughout the time course. Therefore, the starch that was present in the enclosed peduncle at 6 dba had been reutilized by the time that this region of the peduncle became exposed at 2 daa. Similarly the starch present in the enclosed region at 2 daa, was reutilized by 9 daa. Starch grains at lower abundance were also observed in storage parenchyma of both the flag-leaf and second-leaf sheaths during the earlier time points, but were absent from 15 daa and later (Table 1). These results are consistent with a role for starch as temporary carbon storage in the stem and sheath tissues of wheat, preceding the major accumulation of carbon as WSC which occurs later.
It has been shown previously that, in wheat, the carbon excess resulting from N deprivation is stored as increased levels of fructan (Ruuska et al., 2008). In the present study, the developmental progression of starch storage in stems also was modulated by N supply such that in N-deficient plants an increased level of starch was stored compared with plants grown in high N. However, the proportion of total carbohydrate stored as insoluble starch is small relative to later WSC storage. The level of carbohydrate accumulated as starch was maximally about 2·5 and 6·0 mg g−1 d. wt in enclosed peduncle under standard and LN-growth conditions, respectively. In earlier studies, similar values of 2·5–10 mg g−1 d. wt of stem were found (Kiniry, 1993; Sild et al., 1999), although questionable values as high as about 70 mg g−1 d. wt may be derived from results of Judel and Mengel (1982), representing almost half as much carbohydrate as accumulated fructan in stems. For comparison, during the diurnal accumulation of carbohydrates in the wheat leaf, starch is accumulated to 10 µmol mg−1 Chl (Trevanion, 2000; Trevanion et al., 2004), equivalent to approx. 18 mg g−1 d. wt.
The starch grains, presumably within storage plastids, were evident in heterotrophic parenchyma cells of the stem and sheath tissues. Notably, the starch grains in stem segments were found predominantly in parenchyma cells surrounding the vasculature, which probably experience the highest amounts of sucrose being unloaded from the phloem. It is generally considered that fructan accumulates in storage parenchyma cells of the stem tissues, although it has not been established in detail in which particular cells this occurs. It seems possible, therefore, that individual parenchyma cells may store both starch and fructan, although the main phase of fructan accumulation occurs later. Interestingly, while fructan has been shown to accumulate in both mesophyll and parenchyma bundle sheath cells of barley leaves (Koroleva et al., 1998), in the corresponding parenchyma bundle sheath cells of wheat stem tissues, starch grains were not observed.
Although the peduncle of the wheat plant is single continuous tissue it consists of two regions; the lower region, which is enclosed by the flag-leaf sheath and the upper region that is exposed. The exposed region of the peduncle has high levels of Chl, is presumably photosynthetic, and can function as a carbon source tissue, whereas the enclosed region has a lower Chl content, which decreases from top to bottom, and may be considered to be heterotrophic sink tissue. Gebbing (2003) reported that, in wheat, fructan preferentially accumulates in the enclosed region rather than the photosynthetic exposed peduncle, i.e. the enclosed region functions as a temporary vegetative sink tissue. The data presented here demonstrate that starch also accumulated in the enclosed part of the stem that acts as a carbon sink, rather than in the exposed photosynthetic source part. Similarly, in rice, Scofield et al. (2007b) reported that at about 8 daa the enclosed region of internode 1, which is equivalent to the wheat covered peduncle, has low levels of Chl and high levels of starch, whereas the exposed region had high levels of Chl and low levels of starch, indicating that the former functions as a temporary sink tissue whilst the latter is predominantly functioning as a source tissue. Interestingly, there were no starch grains observed in exposed peduncles of even LN-grown wheat plants, indicating that N deficiency could not overwrite the spatial starch accumulation pattern in wheat stems.
Both the timing and the lesser amount indicate that storage of carbon in the form of starch grains during and immediately after stem elongation is unlikely to have the same direct role as the large amount of WSC accumulated for later grain filling. This raises several questions. What is the role of accumulation of such a relatively small proportion of carbohydrate as starch grains? Is the timing of its remobilization and/or the localization or the insoluble nature of the starch significant? Importantly, why are both starch and fructan accumulated in stem storage tissue, and how is this regulated?
The starch grains are being accumulated and reutilized concomitant with the period of both head development and anthesis, and with the later stages of stem elongation. The enclosed peduncle is differentiating and elongating, with the upper part of the peduncle emerging to become the exposed part. It is possible that the starch may provide a reserve carbon supply and energy buffer for fueling cell development during this critical period of anthesis. This may be particularly important under conditions where photosynthetic carbon supply is limited due to environmental factors. For example, it has been established that limitation of photosynthesis during the 2-week period prior to anthesis by shading of wheat severely decreases yield by lowering grain number (Fischer and Stockman, 1980). Similarly, in maize, the importance of localized carbohydrate storage for supporting reproductive development under drought has been demonstrated (Zinselmeier et al., 1999; McLaughlin and Boyer, 2004).
In addition, the starch grains that are reutilized during this period may provide an energy supply and a C source for peduncle growth. During stem development in cereal species, it is known that cell division and growth occurs at the lowest part of the internode in the intercalary meristem and elongation zone (Kaufman et al., 1965; Chonan, 1993; Kende et al., 1998). It is notable that in the enclosed peduncle the starch grains are preferentially stored towards the lower region of the peduncle, close to this zone. Possibly, storing carbohydrate alternatively as either starch grains or fructan may be a mechanism by which the plant can actively partition its carbon supply for use in either active growth processes, or for grain filling.
Furthermore, the apparent starch remobilization observed in the enclosed peduncle also coincides with an increase in WSC deposition. It is likely that starch accumulation in stem tissues occurs when carbohydrate supply is in excess of demands, before the enclosed peduncle and penultimate internode tissue have developed sufficient available capacity, or biosynthetic activity, for the subsequent large fructan storage. Then, as the capacity and fructosyltransferase activities increase, accumulated starch may be remobilized for conversion to fructans in the developing enclosed peduncle. Recently, it was shown that expression of genes of the fructan biosynthetic machinery is increased at this stage (Ruuska et al., 2008).
There appeared to be a secondary phase of transient starch accumulation in both the enclosed peduncle and penultimate internode, coinciding with the time of maximum WSC accumulation at about 15–18 daa, perhaps indicating roles for stem starch of varying physiological significance through development. Starch formation in this case may be occurring as the fructan synthetic capacity begins to decline, or perhaps as an overflow mechanism if fructan capacity becomes saturated. This could prevent further change in the osmotic potential of the cells or vacuoles, or negative feedback by excess sugars that could otherwise reduce the assimilation rate of the plant.
In summary, this study shows that significant but small amounts of starch, relative to WSC, are accumulated and then apparently remobilized from wheat stem storage parenchyma cells around the time of anthesis, before the major accumulation of WSC. The potential importance of this starch to the carbon economy of the plant at this stage is not resolved, but suggestions are that it may be utilized in the early development of the reproductive structures, or in peduncle growth processes, or as a temporary C storage prior to the accumulation of fructan.