Targeting A Bacillus Fructosyltransferase To The Cytosol

Edelman and Jefford [26] put forward a model in 1968 that would become the basis for exploring plant fructan metabolism over the next three decades. In this model the first enzymatic step, catalyzed by SST, was believed to take place in the cytosol. Chain elongation by FFT and storage of polymer occurred only in vacuoles. The model, therefore, called for transport of low molecular weight fructan, not sucrose, across the tonoplast. The locations of enzymes in this model pathway were not challenged until Wagner et al. [11] demonstrated that the entire pathway was compartmentalized within the vacuole of barley leaves. One year later, the fructan biosynthetic pathway, including SST, was also found to be present within the vacuoles of Jerusalem artichoke tubers [53], Although contamination of vacuolar preparations by cytosolic marker enzymes prevented conclusive results in either of the two plant species tested, additional analysis confirmed that fructan synthesis and degradation takes place exclusively in vacuoles of Jerusalem artichoke tubers [54-56].

Approximately 15,000 plant species synthesize fructan [2]. Two hundred years after the polymer was first reported in plants [1], only two species have been characterized for localization of enzymes involved in fructan synthesis and only one, Jerusalem artichoke, has been studied in detail. Recently, a report has demonstrated transport of fructan in the phloem of Agave deserti [57]. The site of fructan synthesis appears to be in the companion cell complex, which is connected to the sieve element by numerous plasmadesmata. Synthesis of fructan from sucrose in the companion cell complex prevents flow of carbohydrate back to photosynthetic cells because the channel size of plasmadesmata is too small to accommodate the polymer. This polymer-trap mechanism of phloem loading was first proposed by Turgeon [58]. The report of Wang and Nobel [57] is the first demonstration of fructan in a native plant, which is not strictly associated with vacuoles in leaves or storage tissue.

The presence of fructan in the phloem illustrates the importance of further analysis of fructan-producing species and the need to revisit the question of enzyme localization. If fructan is synthesized in the companion cell complex and transported in the phloem in one species, could fructan metabolism also occur in other tissues in additional species? Could synthesis of low molecular weight polymer, in locations other than vacuoles, be relatively common among fructan-producing plants? These questions are also relevant to converting transgenic starch-storing plants into fructan-producing crops. Synthesis of fructan in the vacuole of maize endosperm is limited by the concentration of sucrose in that compartment. Targeting a fructosyltransferase to the cytosol may alleviate this problem. Cytosolic expression would take advantage of the most concentrated pool of sucrose in most plant cells.

Figure 6. Transgenic seeds containing the cytosol-targeted SacB gene. A. Fructan concentration was determined from ten individual seeds (dark bars) containing the SacB gene. Polymer was not delected in mature wild-type or segregating seeds, not containing the SacB gene (light bars). B. Thin sections of wild-type and transgenic seeds containing the cytosol-targcted SacB gene were treated with Lugol's solution (1,/KI). Dark staining indicates the presence of starch.

Figure 6. Transgenic seeds containing the cytosol-targeted SacB gene. A. Fructan concentration was determined from ten individual seeds (dark bars) containing the SacB gene. Polymer was not delected in mature wild-type or segregating seeds, not containing the SacB gene (light bars). B. Thin sections of wild-type and transgenic seeds containing the cytosol-targcted SacB gene were treated with Lugol's solution (1,/KI). Dark staining indicates the presence of starch.

It is the cellular location of sucrose synthesis and the site where sucrose accumulates following export from photosynthetically active tissue.

Another critical question is whether the cytosol is an appropriate compartment for storage of high molecular weight polymer. The vacuolar space in barley leaf epidermal cells accounts for over 90% of the total volume and 73% in mesophyll cells [59], The vacuole is an obvious choice for storing large amounts of carbohydrate material. However, the number, function and especially the size of vacuoles vary greatly among plant cells [60]. Vacuoles do not appear to dominate the cell volume in maize endosperm cells [61-62]. As the seed matures, the cell volume becomes dominated by plastids filled with starch granules. Expressing a fructosyltransferase in the cytosol of maize endosperm, early in development, would take advantage of a more highly concentrated pool of sucrose and one that need not diffuse through the tonoplast.

Endosperm specific expression of the mature SacB gene is possible through the use of a 10 kD zein seed storage protein promoter [31]. Using this promoter also allows direct comparison with seeds containing the vacuole-targeted SacB gene described above. Analysis of transgenic seeds by TLC demonstrated that it is possible to accumulate fructan in the cytosol of endosperm [29]. However, the results also show that the level of fructan in mature seeds was not significantly different than in transgenic kernels containing the vacuole-targeted SacB gene (Figure 6A). Although polymer levels were similar, the mature dry weight of seeds containing a cytosolic-targeted fructosyltransferase was found to be severely reduced compared to vacuole-targeted SacB or wild-type seeds [29]. Transgenic seeds containing the cytosolic-targeted SacB gene accumulate very little starch compared to controls (Figure 6B). Thus, a large portion of the reduction in seed weight was due to severely reduced levels of starch synthesis in transgenic endosperm.

Competition for sucrose with the endogenous starch biosynthetic pathway does not appear to be the cause of low fructan accumulation in the cytosol. The enzyme sucrose synthase catalyzes the first step in the starch biosynthetic pathway and its affinity for sucrose (Km = 192 mM) is rather low [63], When compared to SacB, which has an affinity for sucrose almost 10-fold higher, it is not difficult to see how the bacterial enzyme is able to divert a large portion of available sucrose away from the endogenous starch pathway. What is more difficult to appreciate is why this very active enzyme, expressed early in endosperm development, does not produce higher levels of polymer and why the mature seed weight was so severely reduced.

3.1. Altered endosperm development in transgenic kernels

Higher levels of sucrose are likely in the cytosol and would not be expected to limit fructan synthesis, compared to seeds containing the vacuole-targeted SacB gene. The answer to why fructan did not accumulate to higher levels in the cytosol may instead lie in altered development of transgenic endosperm. Evidence of the striking effect of SacB expression on endosperm development can be seen in Figure 7. The most obvious effect is the presence of a large cavity which forms in transgenic seeds very soon after pollination. The cavity contained primarily sucrose with small amounts of fructan. Cavities also form in endosperm of the starch mutants shl and sh2 [64-65]. Cavity formation in transgenic seeds was not in the same location as in the mutants, but appeared between the seed coat and upper endosperm cells. This cavity may represent inefficient metabolism of sucrose entering the seeds from the

Wild typo

Wild typo

< 'ytosolic sacB

lïsuive 7. Wild-type and transgenic seeds contninint! the eytosol-uiiticted SacB gene were collected at several stages during development (indicated above). T hin sections ol the seeds were stained with Evans Blue (I). l'A w/v) (or 4 niiiuiies. Stained sections were llien rinsed twice with water for 20 minutes.

phloem, causing accumulation in a pool under the seed coat. This specific phenotype has not been described in any of the known maize mutants and it is not clear why a reservoir of unused photosynthate should accumulate in this location.

The maize mutants shl and sh2 also accumulate high levels of soluble carbohydrates [38,66]. Increased accumulation of soluble carbohydrate in mutant seeds has been linked to altered endosperm development [65]. Programmed cell death (PCD) is a natural part of maize and other cereal endosperm development [67]. PCD in wild-type maize seeds begins in the upper central portion of endosperm, expanding towards the periphery and downward as the seed matures. Altered sugar levels in the shl and sh2 mutants was linked to increased ethylene production, and ethylene is believed to be involved in signaling the onset of cell death in endosperm [65]. Evans blue staining of endosperm cells indicates the loss of cell viability. Cell death in the upper central endosperm region, late in development of wild-type kernels is seen in Figure 7. The pattern of viability staining in seeds containing the cytsolic-targeted SacB gene is noticeably different than in controls. The periphery of transgenic seeds appears to be less viable than the central endosperm, and timing of this pattern occurred much earlier than in wild-type seeds. The results demonstrate an altered pattern and timing of PCD in seeds containing the cytosolic-targeted SacB gene. The results do not indicate the mechanism by which timing was altered.

3.2. Altered carbohydrate metabolism leads to altered gene expression

It is possible that altered gene expression due to changes in carbohydrate content may play a role in generating this severe phenotype. Plants have the ability to recognize and react to changes in carbohydrate content within the cell [68-69]. Sugars are known to be directly involved in regulation of gene expression, which affects growth and development in all parts of a plant [70], A clear illustration of altered gene expression in transgenic lines containing the cytosolic-targeted SacB gene is demonstrated in Figure 8. Starch isolated from maize lines containing the SacB gene clearly demonstrates altered granule structure compared to wild-type controls. The spherical structure of granules from transgenic seeds may be due to low density, since very little starch accumulates in these lines. Surface erosion and deep pitting is clear evidence of attack by amylase. A reasonable explanation for erosion of starch granules, in a cell where fructan and starch are synthesized in separate compartments, is that an amylase gene has been turned on during a period of development when it would not normally be active. It is also reasonable to suggest that induction of the amylase gene must be due to the presence of one of the two end-products catalyzed by activity of the newly acquired SacB gene. Accumulation of polymer per se does not alter dry matter accumulation and does not appear to play a large role in altered endosperm development. Glucose is the other end-product of the fructosyltransferase reaction, whereas UDP-glucose and fructose are produced in the first step of the native starch pathway. Mistiming gene expression, altered PCD and reduced starch synthesis may be linked to reduced UDP-glucose or fructose levels. Alternatively, increased production of glucose may lead to the severe phenotypes demonstrated in seeds containing the cytosolic-SacB gene.

Free glucose is not normally present at high concentrations in maize endosperm, especially later in development. The action of invertase on sucrose creates free glucose and strict regulation of invertase appears to be very important in plants. In addition to regulating

0 0

Post a comment