Carbon partitioning to cellulose synthesis

Plant Molecular Biology 47: 29–51, 2001. © 2001 Kluwer Academic Publishers. Printed in the Netherlands. 29 Carbon partitioning to cellulose synthesis Candace H. Haigler1,∗ , Milka Ivanova-Datcheva2,3, Patrick S. Hogan2, Vadim V. Salnikov1 , Sangjoon Hwang1 , Kirt Martin1,4 and Deborah P. Delmer2 1 Department of Biological Sciences, Texas Tech University, Box 43131, Lubbock, TX 79409-3131, USA (∗ author for correspondence; e-mail [email protected]); 2 Section of Plant Biology, University of California, Davis, CA 95616, USA; current addresses: 3 Novartis Seeds, Inc., 7240 Holsclaw Road, Gilroy, CA 95020, USA; 4 Department of Natural Science, Lubbock Christian University, Lubbock, TX 79407, USA Key words: calcium, carbon partitioning, cellulose, cotton fiber, sucrose synthase, phosphorylation Abstract This article discusses the importance and implications of regulating carbon partitioning to cellulose synthesis, the characteristics of cells that serve as major sinks for cellulose deposition, and enzymes that participate in the conversion of supplied carbon to cellulose. Cotton fibers, which deposit almost pure cellulose into their secondary cell walls, are referred to as a primary model system. For sucrose synthase, we discuss its proposed role in channeling UDP-Glc to cellulose synthase during secondary wall deposition, its gene family, its manipulation in transgenic plants, and mechanisms that may regulate its association with sites of polysaccharide synthesis. For cellulose synthase, we discuss the organization of the gene family and how protein diversity could relate to control of carbon partitioning to cellulose synthesis. Other enzymes emphasized include UDP-Glc pyrophosphorylase and sucrose phosphate synthase. New data are included on phosphorylation of cotton fiber sucrose synthase, possible regulation by Ca2+ of sucrose synthase localization, electron microscopic immunolocalization of sucrose synthase in cotton fibers, and phylogenetic relationships between cellulose synthase proteins, including three new ones identified in differentiating tracheary elements of Zinnia elegans. We develop a model for metabolism related to cellulose synthesis that implicates the changing intracellular localization of sucrose synthase as a molecular switch between survival metabolism and growth and/or differentiation processes involving cellulose synthesis. Abbreviations: CesA, cellulose synthase; Csl, cellulose-like synthase (genes); DCB, dichlobenil; DPA, days after anthesis; SPS, sucrose phosphate synthase; SuSy, sucrose synthase; P-SuSy, particulate SuSy; S-SuSy, soluble SuSy Significance of carbon partitioning to cellulose synthesis Despite its natural and economic importance, we are only at the threshold of understanding how plants allocate carbon to synthesis of cellulose, or (1→4)β-Dglucan. Cellulose synthesis is a strong, essentially irreversible, carbon sink in plants. Cellulose accounts for 28–30% of dry matter in typical forage grasses (Theander, 1993) and 42–45% of wood (Smook, 1992). The abundance of cellulose is explained by its role as the foundational polymer in plant cell walls. Cell walls (1) form the plant body, (2) constrain the direction of plant morphogenesis, and (3) confer specialized functions such as water conduction and support (in xylem) and control of transpiration (in guard cells). Consequently, cellulose is the most abundant renewable biomass synthesized on earth, with about 1011 tons being synthesized and destroyed each year (Preston, 1974). Throughout history, man has used cellulose extensively in the form of fuel, timber, fiber, forage, and chemical cellulose. Except for burning and building with wood, most large-scale industrial applications place the most value on the cellulose in cellulosic 30 plant products. Therefore, understanding how carbon partitioning to cellulose synthesis is controlled is a critical question in plant biology, biotechnology, and sustainable agriculture. This short review will discuss recent research that implicates sucrose synthase (SuSy) as having a major role in channeling substrate UDP-Glc to cellulose synthase, especially during high-rate secondary wall cellulose synthesis. We will develop a general metabolic model for carbon partitioning to cellulose synthesis and discuss the possible roles of SuSy, cellulose synthase, and other enzymes including sucrose phosphate synthase and UDP-Glc pyrophosphorylase. Cotton fibers will be emphasized as a model because their storage metabolism is almost exclusively directed toward cellulose synthesis during secondary wall deposition (see below). Tracheary elements, which represent one of the major cell types constituting wood, will be discussed as another model. In this case, extensive cellulose synthesis occurs along with substantial xylan and lignin synthesis (Haigler, 1985; Smook, 1992). We will also discuss other cases in which genes implicated in control of carbon partitioning to cellulose synthesis have been well described or manipulated in transgenic plants. The emphasis of this chapter will be on genetic regulation placed within its essential developmental and biochemical context. Other recent reviews provide more details about biochemistry and cellular energetics related to carbon partitioning to cellulose synthesis and to other aspects of cellulose synthesis (Ross et al., 1991; Delmer and Amor, 1995; Volman et al., 1995; Blanton and Haigler, 1996; Brown et al., 1996; Kawagoe and Delmer, 1997; Delmer, 1999a, b). Gene families that encode cellulose synthases (CesAs) and other cellulose-like synthases (Csls) are covered to a certain extent here, as well as the article by Richmond and Somerville (2001, in this issue). Carbon partitioning to cellulose increases during cotton fiber development The development of cotton fibers of Gossypium hirsutum and Gossypium barbadense has been well reviewed (Basra and Malik, 1984; Ryser, 1985, 1999), and only essential points are mentioned here. Cotton fibers are extraordinary, elongated, trichomes of the cotton seed epidermis; each cell becomes over 2.5 cm long in about 21 days. Toward the end of elongation, secondary wall deposition via enhanced cellulose synthesis begins, and it continues until the fiber dies about Figure 1. Change in rate of fiber cellulose synthesis, rate of CO2 evolution (in ovule plus fiber), and fiber weight over time in cultured cotton ovules (Gossypium hirsutum cv. Acala SJ-1). The transition between primary and secondary wall synthesis began about 14 DPA, and secondary wall synthesis had not ended when the experiment was terminated on 27 DPA. Methods were as described previously (Roberts et al., 1992). Briefly, ovules with attached fiber were incubated for 4 h in 14 C-glucose, CO2 was trapped in 1 M KOH, and cellulose in fibers stripped from ovules was determined after acetic/nitric digestion (Updegraff, 1969). Fibers were stripped from ovules and freeze-dried before weighing. Data points represent means and standard deviations of 3 replicates containing 6 ovules each. (Graph modified from Martin, 1999.) 40 days after anthesis (DPA). (The precise timing of these developmental transitions and stages depends on genotype and environment.) The thin primary wall is typical of dicots, containing about 20–25% cellulose and non-cellulosic polymers such as xyloglucan and uronic acid-rich polymers (Meinert and Delmer, 1977). At about 16 DPA in Gossypium hirsutum, transient deposition of (1→3)β-D-glucan (callose) occurs, and the rate of cellulose synthesis increases more than 100-fold as the wall begins to thicken. At about 24 DPA, elongation ceases and the rate of deposition of virtually pure cellulose into the secondary cell wall increases further, continuing at its maximum rate for at least 10 more days. At this stage, cotton fibers do not store starch or synthesize matrix glycans or lignin; about 80% of the imported carbon is directed to cellulose. This ratio for carbon partitioning during secondary wall deposition can be calculated from the 31 in planta data of Mutsaers (1976) and observed directly by comparing incorporation of radiolabel from exogenous 14 C-glucose into cellulose and CO2 in cultured cotton ovules plus fiber (Figure 1; modified from Martin, 1999). Fiber differentiation on cultured ovules includes the same developmental stages as exist in planta, and developmental shifts occur in the rate of cellulose synthesis within fibers alone (Carpita and Delmer, 1981; Figure 1) and in respiration of the ovule/fiber system (Figure 1). The scale of increased cellulose synthesis in cultured fibers is reduced because many fibers fail to turn on high-rate secondary wall synthesis, although some cultured fiber walls do become quite thick (Haigler and coworkers, unpublished). At the end of secondary wall deposition, the fibers may undergo programmed cell death (not yet proven) after which they dehydrate. The mature fiber, which is mostly secondary wall, contains at least 90% crystalline cellulose by weight. Although fibers may import translocated sucrose symplastically through plasmodesmata at the fiber foot (Ryser, 1992, Ruan et al., 1997), cellulose biosynthesis may start from sucrose or glucose and fructose. Invertases are present in cotton fibers in addition to SuSy (Buchala, 1987; Basra et al., 1990; Wäfler and Meier, 1994), and they cannot be excluded as an additional means of sucrose degradation. UDP-glucose (UDP-Glc) was implicated as the probable immediate substrate for cotton fiber secondary wall cellulose synthesis (Franz, 1969; Carpita and Delmer, 1981) and proven to be so for bacterial cellulose synthesis (Ross et al., 1991). However, secondary wall cellulose synthesis may not rely directly on the free pool of cytoplasmic UDP-Glc, but instead use UDP-Glc directly channeled from particulate SuSy in the cell cortex acting degradatively. In this model, developed from work on cotton fibers during secondary wall synthesis (Amor et al., 1995), sucrose is the substrate that is initially required for high-rate secondary-wall cellulose synthesis. The role of sucrose synthase in intracellular carbon partitioning to cellulose Overview The enzyme sucrose synthase (SuSy; EC 2.4.1.13; sucrose + UDP ↔ UDP-Glc + fructose) plays a major role in the degradation of sucrose in plant sink tissues. Although the reaction is freely reversible, under most conditions SuSy catalyzes sucrose cleavage. An advantage of sucrose degradation by SuSy compared to invertase is that the energy of the glycosidic bond is conserved in UDP-Glc. In many non-photosynthetic tissues, soluble SuSy (S-SuSy) exists at high levels in the cytoplasm, where its products may be used in general metabolism and for synthesis of storage polymers such as starch (ap Rees, 1984; Copeland, 1990; Quick and Schaffer, 1996). A recent comprehensive review (Winter and Huber, 2000) includes many aspects of SuSy activity that will not be dealt with here. We will focus on SuSy-mediated carbon metabolism as it relates to intracellular carbon partitioning to cellulose and related polymers. Years ago degradative SuSy activity was associated with cell wall synthesis. Based on feeding of 14 C-sucrose to etiolated pea epicotyls, a coupled reaction between SuSy and glucan synthases was proposed (Rollit and Maclachlan, 1974). This approach was extended to build a model in which soluble SuSy in pea roots (stele, cortex, and apex) was proposed to supply UDP-Glc for polysaccharide biosynthesis, whereas invertases supplied hexoses for respiration (Dick and ap Rees, 1976). Consistent with this model, glucan synthesis in detergent-permeabilized cotton fibers (Gossypium arboreum) showed a 140-fold preference for sucrose compared to glucose or UDP-Glc. There was little competition between sucrose and UDP-Glc, which served primarily as a substrate for callose, or (1→3)β-D-glucan, synthesis (Pillonel et al., 1980). Sucrolysis via SuSy was also shown to feed glycolysis in several heterotropic systems (Huber and Akazawa, 1986; Xu et al., 1989), and SuSy activity was proposed to mark sink strength in developing bean seeds, potato tubers, and growing roots (Sung et al., 1989). Other evidence for a role for SuSy in providing UDPGlc to synthesis of cellulose and/or other cell wall polymers was found in maize endosperm (Chourey et al., 1991), tomato roots which store little starch (Wang et al., 1993), and cotton fibers during initiation (Nolte et al., 1995). Even though it was suggested that SuSy might be membrane-associated while acting degradatively in heterotrophic bean cells (Delmer and Albersheim, 1970), soluble SuSy was the only form studied for many years. 32 SuSy associated with the plasma membrane and/or cortical cytoskeletal elements (P-SuSy) has a special role in contributing UDP-Glc to secondary wall cellulose synthesis and to callose synthesis We made the surprising finding that a substantial proportion of the total SuSy protein is associated with particulate fractions of developing cotton fibers – in cell wall pellets that sediment at low centrifugal forces (2000 × g) and in high-speed (100 000 × g) pellets that contain membranes and cytoskeletal elements (Amor et al., 1995). This special, particulate form of SuSy is referred to hereafter as P-SuSy, and soluble SuSy is designated S-SuSy. SuSy is not an integral membrane protein, but its association with the cotton microsomal fraction is quite strong; it could not be removed by high-salt or mild detergent treatments. Semi-permeabilized fibers also used sucrose more efficiently than UDP-Glc as a substrate for cellulose synthesis. Immunofluorescence showed that SuSy could exist in the cortex/cell wall area of secondary wallstage cotton fibers in patterns consistent with both oriented cellulose microfibril synthesis and predicted punctate sites of callose synthesis (Amor et al., 1995). Therefore, we proposed that P-SuSy channels UDPGlc to cellulose or callose synthesis at the plasma membrane. Using cryogenic electron microscopic methods that should preserve in vivo protein localization (Nicolas and Bassot, 1993), we obtained further support for this model. These results show that SuSy is abundant all along the cotton fiber surface and just above the cortical microtubules in close proximity to the predicted site of cellulose synthesis in the plasma membrane (Figure 2a). However, SuSy in this location could also be related to the callose synthesis that occurs in cotton fibers (Maltby et al., 1979) and that can also proceed using sucrose as substrate in in vitro assays (Pillonel et al., 1980; Amor et al., 1995). The deposition of callose fibrils has been shown to occur on plasma membrane sheets with attached microtubules isolated from tobacco BY-2 protoplasts (Hirai et al., 1998). SuSy labeling with two polyclonal antisera, one made against cotton SuSy (Figure 2a) and one made against bean SuSy (data not shown), was also observed in the cotton fiber wall, most densely in a zone just outside the membrane. The same pattern was observed in true cross-sections of cotton fibers and by using affinity-purified antiserum to cotton SuSy, although labeling density in the latter case was much reduced in all cellular locations (data not shown). La- beling of the fiber wall was not observed using the pre-immune serum for the cotton SuSy antibody or other rabbit polyclonal antibodies (data not shown), so non-specific binding is not a likely explanation. Since SuSy lacks an identifiable signal sequence, localization in the cell wall is not expected, and further work will be required to establish any significance for this observation. We are testing the hypothesis that stress on cotton fibers as they are unavoidably handled before freezing could cause movement of SuSy into the cell wall, possibly attached to the ends of terminated microfibrils. Coupling between P-SuSy and glucan synthases has the advantages of (1) promoting synthesis of cellulose from sucrose with no additional energy input, (2) avoiding competition for use of UDP-Glc by other pathways, and (3) allowing immediate recycling of UDP, a compound that inhibits the reactions catalyzed by cellulose synthase (Ross et al., 1991) and callose synthase (Morrow and Lucas, 1986). The dependence of cellulose synthesis on sucrose may relate to why the ratio of cellulose to callose synthesized in vitro by cotton fiber membrane preparations was enhanced when a combination of MOPS buffer and sucrose was used during membrane isolation and assay (Kudlicka et al., 1995, 1996). Cellulose synthesis was also enhanced when the bacterium Acetobacter xylinus was transformed with mung bean SuSy that had been modified to have less regulation via phosphorylation, making the bacterium able to use sucrose for cellulose synthesis (Nakai et al., 1999). Although it is unlikely that the plant SuSy associated directly with the bacterial cellulose synthase, this research supports the idea that the ability to recycle UDP rapidly can enhance cellulose synthesis. Data from several plant cell systems now support a role for P-SuSy in secondary wall cellulose synthesis. Developing, thick-walled, transfer cells in cotton seeds have high levels of SuSy, although its precise intracellular location is unknown (Ruan et al., 1997). Immunofluorescence of tracheary elements sometimes showed SuSy over developing secondary wall thickenings, which are patterned sites of high-rate cellulose synthesis. However, in other tracheary elements or even in different regions of the same cell immunofluorescence indicating SuSy was diffuse or punctate over the cell surface (Figure 3a). Use of superior cryogenic electron microscopic methods (Nicolas and Bassot, 1993) showed SuSy consistently enriched near the plasma membrane underlying the tracheary element thickenings, whereas it labeled less frequently 33 Figure 2. Thin sections of plant-grown cotton fibers labeled with anti-serum to SuSy and 20 nm colloidal gold. A. Secondary wall stage fiber at 30 DPA. SuSy exists between the cortical microtubules (MT) and the thick, secondary cell wall (SCW), which has a width approximately as indicated by the double-headed arrow. SuSy does label in a zone near the plasma membrane, which is the predicted location of cellulose synthases. Labeling was also sometimes observed deeper within the cell wall. B. Primary wall stage fiber at 10 DPA. Here the section is near the top surface of the tubular fiber; the plane of section has just entered the cytoplasm and thin primary wall (PCW) is shown on both sides in a tangential section as marked by double-headed arrows. Sparse colloidal gold/SuSy is highlighted by asterisks. Cryogenic methods were used (Haigler, Grimson and coworkers, in preparation), which preserve cellular structure accurately and greatly hinder molecular movement (Nicolas and Bassot, 1993). Briefly, fibers still attached to a seed fragment were plunged into re-solidifying propane (near −168 ◦ C) cooled by liquid nitrogen, transferred to acetone (−80 ◦ C for 4 days, then −40 ◦ C overnight), infiltrated at −40 ◦ C over 2 weeks by drop-wise addition of resin (a 1:1 mix of Lowicryl K4M and HM20 acrylic resins, Electron Microscopy Sciences, Ft. Washington, PA) followed by 3 changes of 100% resin, flat embedded between two slides (Reymond and Pickett-Heaps, 1983), and polymerized by 360 nm UV irradiation for 2 days at −20 ◦ C then for 1 day at 4 ◦ C. Areas of fiber for sectioning were chosen in a light microscope, cut out, and mounted on a blank resin block. Immunolabeling was performed by standard methods (Brewin et al., 1986) using anti-SuSy (Amor et al., 1995) at 1:1000 and secondary antibody/colloidal gold at 1:100. Similar results on secondary wall stage fibers (with less dense labeling, but including labeling in the cell wall) were obtained with a second polyclonal serum raised against bean SuSy and with anti-serum to cotton SuSy that was affinity-purified with recombinant SuSy bound to a column (data not shown). Substitution of the SuSy pre-immune serum at 1:1000 for anti-SuSy yielded unlabeled sections or sections with very sparse labeling that appeared non-specific (data not shown). Bar is 0.2 µm. 34 sometimes punctate SuSy localization in cotton fibers that could reflect sites of callose synthesis (Amor et al., 1995). It was proposed that SuSy localized in companion cells of maize and citrus could provide UDP-Glc for rapid synthesis of callose in sieve elements (Nolte and Koch, 1993). SuSy has also been localized in callose-secreting tapetal cells of maize anthers (Chourey and Miller, 1995). It has been colocalized with callose synthase in callose-containing, forming cell plates of dividing plant cells (Hong et al., 2001). Cultured mesophyll cells of Zinnia elegans also show punctate SuSy on mature cell plates, seemingly corresponding to patches of callose-containing plasmodesmata (Figure 3B). Figure 3. Whole mounts of tracheary elements differentiating in culture from isolated mesophyll cells of Zinnia elegans labeled with anti-serum to SuSy and FITC. A. At the top of the cell, SuSy underlies the secondary wall thickenings, but, at the bottom of the cell, SuSy has a punctate pattern. This experimentally induced variability was common in immunofluorescence preparations, and cryogenic immunoelectron microscopy was used to show that SuSy was consistently enriched over the secondary wall thickenings in vivo (Salnikov et al., 2001). B. This cell divided twice before differentiation commenced so that 3 cells exist (left, and right top and bottom). SuSy is localized over the mature cell plates, seemingly corresponding to patches of callose-containing plasmodesmata. Fixatives used with equivalent results were: (a) Histochoice, an acidic, buffered solution containing glyoxal, a derivative of formaldehyde, and no chelators to affect calcium concentration (Amresco, Solon, OH), and (b) 3.7% formaldehyde, 1 mM MgSO4 , and 5 mM EGTA in 50 mM PIPES, pH 6.9 (closely related to a previously described buffer optimized for microtubules; Doonan and Clayton, 1986). Immunolabeling was performed by standard methods (Doonan and Clayton, 1986) using anti-cotton SuSy (Amor et al., 1995; proved to recognize one band in western blots of Zinnia protein; Salnikov et al., in press) at 1:200 and secondary antibody/FITC at 1:200. Substitution of the SuSy pre-immune serum for anti-SuSy resulted in only dull, generalized fluorescence (data not shown). Bar is 30 µm. and densely between thickenings (Salnikov et al., in press). Since plasma membrane rosettes, which have been labeled with antibodies to cellulose synthase (Kimura et al., 1999), are enriched beneath these same secondary wall thickenings (Haigler and Brown, 1986), we can state that SuSy is preferentially localized near patterned aggregates of cellulose synthase. The changeable immunofluorescence results probably reflect artifacts of specimen preparation and demonstrate the labile nature of in vivo SuSy localization (see further discussion below). Double labeling also showed that SuSy was in the same plane as cortical actin in tracheary elements (Salnikov et al., in press). Evidence is also accumulating that P-SuSy is associated with callose synthesis, in addition to the Does P-SuSy have the same role during primary wall cellulose synthesis? Available data support the hypothesis that a smaller amount of P-SuSy supports primary wall cellulose synthesis, but this relationship may be more facultative than during secondary wall synthesis. The ratio of S-SuSy to P-SuSy has not been quantified in young cotton fibers undergoing elongation via primary wall synthesis, but high levels of total SuSy do exist from the very beginning of fiber initiation (Nolte et al., 1995; Ruan et al., 1997). Contrary to the wild type, a fiberless mutant of cotton does not show enriched SuSy in seed epidermal cells (Ruan and Chourey, 1998), although the specific mutation may exist upstream of the expression of genes encoding SuSy in the fiber initiation program. Electron microscopic immunolocalization showed that SuSy is sparsely distributed close to the plasma membrane of elongating cotton fibers (Figure 2B); in this location it could be associated with primary wall cellulose synthases that are typically more widely spaced in the plasma membrane than secondary wall cellulose synthases (Haigler, 1985). P-SuSy was also detected in other cell types containing cells engaged in primary wall synthesis including cotton roots, pea and bean stems, and cultured tobacco cells (Amor et al., 1995). In a careful developmental and experimental analysis, the amounts of SuSy mRNA and SuSy activity were highest in various tissues of growing carrot leaves, stems, and tap roots that would have been synthesizing primary and secondary walls (Sturm et al., 1995; 1999). In protoplasts of growing carrot suspension cultures, 25–50% of total SuSy was associated with the membrane fraction (Sturm et al., 1999). On an overall basis, maize 35 endosperms had 14% P-SuSy during cell expansion and only 3% P-SuSy during rapid starch synthesis (Carlson and Chourey, 1996). Maize pulvini (a special region of the internode involved in gravitropism) that were gravi-stimulated so that cell expansion via primary wall synthesis was induced showed a 10-fold increase in P-SuSy (Winter et al., 1997). Transformed plants with changed SuSy expression provide a powerful means of testing the role of SuSy in primary wall synthesis. The expression of a reporter gene linked to the promoter of a maize gene for SuSy (Sh1) increased over 400-fold when maize protoplasts began to regenerate primary walls, and this increase could be blocked by the cellulose-synthesis inhibitor dichlobenil (DCB) (Mass et al., 1990). However, DCB did not inhibit expression of the gene for the main form of carrot SuSy (Susy∗ Dc1) (Sturm et al., 1999), indicating species-specific SuSy regulation. The sh1 mutation of maize causes degeneration of some, but not all, expanding primary cell walls in developing endosperm. The cells that degenerate have particularly thin, fragile, primary walls, and the synthesis of these is hypothesized to depend on plasma-membraneassociated Sh1 (Chen and Chourey, 1989; Carlson and Chourey, 1996). In dwarfed carrot plants transformed with an antisense gene construct for SuSy, tap roots having 18–23% of normal SuSy activity contained only about 63% of the normal crystalline cellulose (per gram fresh weight). They also contained expanded parenchyma cells, probably due to insufficient reinforcement of their primary walls by cellulose microfibrils (Tang and Sturm, 1999). Tomato fruits with only 1% residual soluble SuSy activity after antisense suppression showed no change in cellulose or crosslinking glycans, but the methanolysis procedure used would not be the most definitive for this evaluation (Chengappa et al., 1999). Tomato fruits with almost no residual, soluble SuSy activity do finally grow to normal size (D’Aoust et al., 1999), but it was not determined whether their primary wall composition might have become different. We recently determined that potato tubers with SuSy expression repressed by antisense techniques showed reduced cellulose content (Figure 4). The reduction was substantial in the line T-112 that had only 4% of the SuSy activity and 66% of the starch content found in wild-type tubers (Zrenner et al., 1995). The transgenic tuber cell walls also showed increased uronic acids, but no increases in other non-cellulosic sugars were observed, suggesting compensation for reduced primary wall cellulose by synthesis of addi- tional pectin. Similar compensation has been observed in primary walls of DCB-adapted plant cell cultures with reduced cellulose (Shedletzky et al., 1992) and in tobacco leaves where expression of a cellulose synthase gene was lowered by virus-mediated gene silencing (Burton et al., 2000). These compensatory changes suggest that, although the wall polysaccharides are extracellular, there still may be complex feedback events determining how carbon is partitioned between various wall polymers. In this regard, we recently observed a substantial increase in the level of cellulose synthase (CesA) protein in cotton fibers treated with a novel herbicide, CGA 325’615, which inhibits synthesis of crystalline cellulose (Peng et al., 2001). In addition to a possible role in facilitating primary wall cellulose synthesis, it has also been proposed that SuSy can associate with Golgi membranes to facilitate mixed-linkage (1→3),(1→4)β-D-glucan synthesis in maize (Buckeridge et al., 1999). Since these authors did not detect SuSy in the Golgi fraction of soybeans, they hypothesized that the β-glucan synthase could have a cellulose synthase ancestor so that it uniquely retained the ability to associate with SuSy. We also were unable to localize SuSy in the Golgi apparatus of tracheary elements of the dicot Zinnia elegans (Salnikov et al., in press), which, in contrast to maize, was expected to be synthesizing a type of xylan that would not require UDP-Glc as substrate. How might the SuSy gene family relate to control of carbon partitioning to polysaccharide synthesis? Most plants contain at least two, and perhaps three, genes encoding isoforms of SuSy (reviewed in Winter and Huber, 2000). There is ongoing work in several laboratories on mechanisms that control the intracellular location of SuSy. First, one might consider that Sand P-SuSy could be encoded by different genes subject to different modes of regulation. Although the data are limited, it would appear that no single gene is responsible for encoding P-SuSy. At least two isozymes from distinct genes encoding SuSy were associated with the plasma membrane of maize (Carlson and Chourey, 1996). When two genes encoding SuSy are expressed in the same cell, the proteins form homoor heterotetramers (Chen and Chourey, 1989; Koch et al., 1992; Guerin and Carbonero, 1997), suggesting that the isozymes are interchangeable in at least some cellular roles. However, because both classes of maize genes (Sh1 and Sus1) are conserved among grasses 36 Figure 4. Cell wall composition and starch content of potato tubers expressing anti-sense SuSy constructs. Antisense constructs, growth and harvesting of tubers were as described in Zrenner et al. (1995). Tubers were freeze-dried, ground to a powder, and the resulting wall pellet was extracted with DMSO to remove starch, 2× in methanol/chloroform/H2 O (14:4:3), 1× in 10% methanol, and 2× for 1 h at 70 ◦ C in 90% methanol, followed by extensive H2 O washing. Cellulose is that fraction of carbohydrate that remained insoluble after acetic-nitric digestion (Updegraff, 1969). Non-cellulosic neutral sugars were determined by GLC of alditol acetates (Blakeney et al., 1983), and uronic acids were determined by the method of Blumenkrantz and Asboe Hansen (1973). Results are the average of 5 independent replicates. The levels of starch indicated are taken from the data of Zrenner et al. (1995) for these transgenic and control lines. and show different rates of divergence in their exons compared to their introns, it has been argued that both have critical roles at least under some physiological conditions (Shaw et al., 1994). Dicots have two additional classes of genes encoding SuSy that are distinct from the Sh1 and Sus1 classes of maize and other monocots (Shaw et al., 1994; Sebkova et al., 1995; Fu and Park, 1995; Sturm et al., 1999). However, no previous phylogenetic analyses could be interpreted with specific knowledge of genes encoding SuSy that were expressed in secondary-wall-synthesizing cells, especially because vascular expression was attributed to a role in phloem unloading. In a cotton fiber cDNA library, we detected at least 3 distinct genes encoding SuSy, but the deduced amino acid sequences are all 93% identical to each other and tryptic peptide profiles on HPLC for S- and P-SuSy are virtually indistinguishable (Delmer and coworkers, unpublished). Since cotton is an allotetraploid (Wendel et al., 1999), at least two of these genes encoding SuSy could represent alleles from the two distinct genomes. Figure 5 shows an unrooted cladogram derived from comparing the deduced amino acid sequence of one of the representative cotton genes encoding SuSy with other SuSy proteins. The SuSy proteins from graminaceous monocots fall into a distinct clade, while cotton SuSy shows most similarity with one Arabidopsis SuSy and with two other dicot proteins that are expressed during nodulation of soybean or actinorhizal associations in alder. While these comparisons clearly indicate some divergence between genes encoding SuSy from dicots and grasses, they offer no insight into possible distinct roles for individual genes coding for S-Susy vs. PSuSy. Furthermore, the N-terminal sequences in the predicted region of serine phosphorylation (see below) are conserved in nearly all SuSy proteins, again offering no clue to possible distinct motifs for differential phosphorylation (inset in Figure 5). In addition, deduced SuSy proteins do not show obvious sites for prenylation or myristoylation that might promote membrane association. One or two transmembrane helices might exist in maize SuSy sequences, but this alone could not explain why both isoforms exist in soluble or membrane-associated forms (Carlson and Chourey, 1996). In summary, the data at present do not support the idea that distinct genes encode P- and S-SuSy. 37 Figure 5. An unrooted cladogram constructed as described by Holland et al. (2000) from comparisons of deduced SuSy amino acid sequences. Since the 3 distinct SuSy genes expressed in cotton fibers are extremely similar, only one is included in this analysis. Gh, Gossypium hirsutum (accession number U73588); At, Arabidopsis thaliana (P49040 and Q00917); Gm, Glycine max (P49034); Dc, Daucus carota (P49035 and P049039); St, Solanum tuberosum (P10691 and P94039); Zm, Zea mays (P04712 and P49036); Os, Oryza sativa (P30298, P31924 and Q43009). The inset shows a pile-up of the SuSy N-terminal amino acids with predicted site of phosphorylation marked by an asterisk (∗ ). What other mechanisms may regulate intracellular location of SuSy? Phosphorylation Winter and Huber (2000) discuss extensively the possibility that SuSy localization is controlled by its state of phosphorylation. It is possible that SuSy phosphorylation, which depends on a conserved N-terminal domain as documented in many species and tissues, is an important part of SuSy regulation in sink tissues (Zhang et al., 1999; Loog et al., 2000). Maize S-SuSy was phosphorylated in vivo and in vitro on Ser- 15 (of the protein encoded by ZmSus2) by a Ca2+ - and phospholipid-dependent protein kinase (Shaw et al., 1994; Koch et al., 1995; Huber et al., 1996; Lindblom et al., 1997; Loog et al., 2000). Consequently, the Km for UDP and sucrose, but not UDP-Glc or fructose, was lowered, thus favoring the sucrose cleavage reaction (Huber et al., 1996). However, later research showed that the magnitude of the effect was so low that it might be physiologically insignificant (Winter et al., 1997). Soybean nodule S-SuSy was phosphorylated in vitro by a Ca2+ -dependent protein kinase (CDPK), but phosphorylation of the recombinant or native en- 38 Figure 6. In vitro phosphorylation of SuSy by cotton fiber extracts. Extracts were prepared from fibers harvested from greenhouse-grown cotton plants (Gossypium hirsutum cv. Coker 130) at 21 DPA. Locules were frozen in liquid N2 and stored at −80 ◦ C until fibers were detached from seeds and ground to a fine powder under liquid N2 . The frozen powder was extracted at 4 ◦ C with extraction buffer (EB; 2 ml per gram fiber fresh weight) that contained 50 mM Hepes/KOH pH 7.3, 0.5 M sucrose, 1 mM DTT, 0.03% Brij 58, 0.5% insoluble PVP, a protease inhibitor cocktail (PI) (Complete, Mini, EDTA-free; Boehringer Mannheim), and the phosphatase inhibitors (PhI) NaF (10 mM) and Na3 VO4 (1 mM) plus Microcystin-London Resin White at 0.5 µM (Calbiochem, La Jolla, CA). The soluble fraction (A) and EGTA washes (B) were dialyzed overnight against 50 mM MOPS buffer pH 7.5, 5 mM MgCl2 , 0.1 mM CaCl2 , 2 mM DTT containing PI and PhI cocktails. Dialysis was performed in a ratio of 1:500 with 3 buffer changes followed by concentration of the extracts on Centricon 50. Extracts (0.5–1 mg total protein per reaction) were incubated with 1–3 µCi [γ -32 P]ATP (Amersham Biotech, 6000 Ci/mmol) per reaction in the presence or absence of unlabeled 0.1 mM ATP (as indicated), 0.1 mM CaCl2 and 0.5 µM Microcystin London Resin White. Various concentrations of CaCl2 or EGTA up to 2 mM were applied to the reaction mixture to explore the effect of Ca2+ on S-Susy and P-Susy phosphorylation. After 10 min incubation at 25 ◦ C, the reaction was terminated by adding 10 mM EDTA, and the phosphorylated SuSy band was identified after immunoprecipitation, SDS-PAGE and electroblotting and autoradiography of the blotted membranes. SuSy was immunoprecipitated in IP buffer containing 150 mM NaCl, 50 mM Tris pH 7.8, 1 mM EGTA, 1% Triton X-100, 0.5% sodium deoxycholate, 0.05% SDS (Todorov et al., 1995) supplemented with PI and PhI inhibitors as described. Protein A Sepharose (Amersham Pharmacia Biotech; prepared according to the manufacturer’s instructions; 50 µl/sample) was incubated with 20 µl antibody in 200 µl IP buffer (4 ◦ C, several hours). The slurry was washed 2× in IP buffer, and incubated overnight at 4 ◦ C with protein extracts. The immune complex was collected by brief centrifugation and washed 3× in PBS, 3× in IP buffer containing 0.5 M NaCl and then 3× with PBS. The washed beads were extracted in ‘Novex’ sample buffer and resolved in Nu PAGE 4 to 12% gradient gels as described above. CB, Coomassie Blue-stained gels. zyme did not have a major effect on enzyme kinetics or sucrose degradation (Zhang and Chollet, 1997). In contrast, phosphorylation is clearly important in regulating the kinetic properties of mung bean SuSy (Nakai et al., 1999). Other studies showed that maize S-SuSy was more heavily phosphorylated in vivo and that phosphorylated SuSy was less hydrophobic and less likely to associate with membranes (Winter et al., 1997, Zhang et al., 1999). It is reasonable to assume that a change in the phosphorylation state that affects SuSy conformation and substrate binding may also either promote or prevent membrane association, and we tested this further in cotton fiber extracts. Here, S-SuSy was phosphorylated in vitro in a Ca2+ -dependent manner by an endogenous kinase present in the soluble fraction (Figure 6A). However, phosphorylation of P-SuSy by an endogenous kinase that elutes along with P-SuSy upon EGTA treatment of the particulate fraction occurred in a Ca2+ -independent manner (Figure 6B). Therefore, P-SuSy and S-SuSy might be subject to different control. Unlike the results from maize S-SuSy (Winter et al., 1997), we found no difference in the level of phosphorylation of S-Susy vs. P-SuSy (standardized to amount of SuSy protein) isolated from cotton fibers labeled in vivo with 32 P-orthophosphate (Figure 7). Further complicating interpretation of these results, profiles from 2-D TLC separations of tryptic peptides showed different patterns when SuSy was phosphorylated in vitro vs. in vivo. In both cases, S- 39 Figure 7. In vivo phosphorylation of SuSy. Ovules with their associated fibers were cultured at 30 ◦ C for 14–18 days (Haigler et al., 1991), then transferred to phosphate-free medium for 24 h, and finally incubated 24 h in medium supplemented with 0.5 mCi/ml [32 P]orthophosphate (carrier- and acid-free, 10 mCi/ml, 8000 Ci/mmol; Amersham Biotech). The ovules were washed briefly in cold phosphate containing PhI cocktail, and the fibers were detached immediately on ice in EB containing the same PI and PhI cocktails plus 0.5 µM Microcystin LRW and homogenized and extracted as described in the legend to Figure 6. S-SuSy (S) and P-SuSy released by EGTA washing of membranes (EG/M) were subjected to immunoprecipitation and SDS-Nu PAGE electrophoresis as described in the legend to Figure 6. Incorporation of 32 P was detected by autoradiography on Bio-Max MR X-ray films (Kodak). Gels were exposed 1–3 days at −80 ◦ C. To verify the SuSy band, half of the immunoprecipitated proteins of each sample were resolved on the same gel and electroblotted onto nitrocellulose membrane and probed as described above using anti-Susy antibody (not shown). CB, Coomassie Blue-stained gels. and P-SuSy were phosphorylated exclusively on Ser residues such that SuSy from both fractions contained the same major phosphopeptide (presumably that for Ser-11). However, a second phosphorylated peptide was detected in S- and P-SuSy phosphorylated in vitro (not shown). Sites in addition to this conserved Ser can also be phosphorylated and may influence at least the kinetic properties of SuSy (Zhang et al., 1999; Anguenot et al., 1999). These results emphasize that conclusions drawn from studies of in vitro phosphorylation may be difficult to extend to the in vivo situation. Therefore, at present, the role of phosphorylation of SuSy in regulation of intracellular localization remains inconclusive. Intracellular calcium levels Our evidence from cotton fibers suggests that Ca2+ levels affect the solubility of SuSy in a complex manner. After extraction of cotton fibers in buffer containing 200 nM free Ca2+ , both S- and P-SuSy were detected. In most experiments, all or at least a sub- stantial portion of the P-SuSy could be eluted from the particulate fraction by EGTA treatment (Figure 8A). After extraction of fiber in buffer containing EGTA, very little P-SuSy was detected, presumably because it eluted from the particulate fraction during extraction. Furthermore, this ‘eluted P-SuSy’ was selectively precipitated by adding Ca2+ to the supernatant, leaving only the original S-SuSy in solution (Figure 8B). We do not yet understand why ‘eluted P-SuSy’ behaves differently in solutions containing Ca2+ . However, it is interesting that the EGTA extracts of fiber membranes or the ‘eluted P-SuSy’ precipitated by Ca2+ both contain similar sets of other proteins, including Ca2+ -independent kinase, tubulin, actin, and annexins (D. Delmer and M. Datcheva, unpublished). Thus, the level of Ca2+ may well influence the particulate nature of SuSy by modulating its association with membranes, with glucan synthases in membranes, with F-actin, and/or with other proteins that may occur in particulate fractions. These results may relate to a rise in intracellular calcium that could occur at the transition to secondary wall synthesis in cotton fibers and tracheary elements and facilitate formation of a P-SuSy protein complex. Secondary wall deposition in tracheary elements requires uptake of extracellular calcium through calcium channels (Roberts and Haigler, 1990), and the levels of calmodulin and calmodulin-binding proteins increase prior to secondary wall deposition (Kobayashi and Fukuda, 1994). Indirect evidence suggests that similar mechanisms may operate in cotton fibers. Both induction of Ca2+ -dependent callose synthesis and an oxidative burst (inferred from onset of H2 O2 production) occur at the primary to secondary wall transition in cotton fibers (Maltby et al., 1979; Potikha et al., 1999). In addition, expression of a Rac gene is induced (Delmer et al., 1995), and, in animals, Rac activates the NADPH oxidase involved in such an oxidative burst. Since we showed that cotton annexin may interact with callose synthase (and actin in some systems) and may be phosphorylated (Andrawis et al., 1993), one possibility is that annexin acts as a bridge between SuSy and actin and/or glucan synthases in the plasma membrane. Such an association could be regulated by Ca2+ , either directly or through changes in protein phosphorylation state. Interaction with the cytoskeleton Maize SuSy was detected in the detergent-insoluble fraction of microsomal membranes, actin (but not tubulin) co-immunoprecipitated with S-SuSy, and 40 Figure 8. The effect of Ca2+ on distribution of S-Susy and P-SuSy in cotton fiber extracts. Cotton locules were harvested from greenhouse plants at 10, 17, or 21 DPA and handled and processed as described in the legend to Figure 6 with the following modifications. For the experiment in A, the extraction buffer also included 0.5 mM EGTA and 0.4 mM CaCl2 (ca. 200 nM free Ca2+ ; calculated as described in Hayashi et al., 1987). For B, this was replaced by 5 mM EGTA. The fiber homogenate was filtered through several layers of MiraCloth (Calbiochem, La Jolla, CA) and centrifuged at 10 000 × g for 15 min. The pellet was extracted a second time with EB, and the combined extracts were centrifuged at 100 000 × g for 1 h. This pellet consisting of membranes and cytoskeletal elements was resuspended in 50 mM BTP-MES pH 7.0 and further extracted (2×, 15 min each) with 10 mM EGTA in 50 mM BTP-MES buffer pH 7.0, 1 mM DTT, and the PI cocktail. EGTA washes were separated from particulate matter by centrifugation at 400 000 × g for 15 min. Soluble proteins and EGTA washes were concentrated on Centricon 50 (Amicon) prior to electrophoresis. The Ca2+ precipitate (Ca2+ Ppt) was obtained by adding 15 mM CaCl2 to supernatants, stirring (15 min, 4 ◦ C), and centrifuging (25 000 × g, 15 min). For western blotting, proteins were separated with 4–12% gradient Nu PAGE Bis-Tris gels (Novex) with SDS-MOPS running buffer under reducing conditions, and electroblotting to membranes was carried out in Nu PAGE transfer buffer. The blots were probed with a 1:10 000 dilution of polyclonal antibody raised against isolated SuSy from Vicia faba. Goat anti-rabbit horseradish peroxidase-conjugated second antibody (Sigma) was used as second antibody, and positive bands were detected by enhanced chemiluminescence. SuSy bound to F-actin polymerized in vitro in a 1:5 ratio (possibly indicating that 1 SuSy tetramer binds at intervals of 20 actin monomers). The binding of SuSy to G- and F-actin was apparently not related to SuSy phosphorylation state, but was stimulated by high levels of sucrose (Winter et al., 1998; Winter and Huber, 2000). A peptide in maize SuSy with homology to known actin-binding peptides has also been identified (Winter and Huber, 2000). We have already mentioned two lines of evidence suggesting that SuSy may interact with the cytoskeleton in cotton fibers: (1) immunofluorescence showing SuSy coaligned with helical cellulose microfibrils (Amor et al., 1995), and (2) co-elution of P-SuSy with actin and tubulin in the presence of EGTA. Helical microfibrils in cotton fiber secondary walls are paralleled by the cortical microtubule network, and actin microfilaments exist in the vicinity of the microtubules in an incompletely understood relationship (Seagull, 1993). Increasing tubulin at the onset of secondary wall deposition (Kloth, 1989; Dixon et al., 1994) correlates with the increase in P- SuSy at the same stage. In addition, a form of tubulin that behaves as an integral membrane protein has been identified in plants (Laporte et al., 1993). This type of tubulin may interact with cellulose synthase such that a solubilized complex isolated with anti-tubulin antibodies can synthesize cellulose using either sucrose or UDP-Glc as substrate (Prof. Mizuno, Tokyo University, personal communication). Consistent with these observations, SuSy is preferentially localized to patterned sites of secondary wall cellulose synthesis in differentiating tracheary elements where it exists in the same plane with actin between the cortical microtubules and the plasma membrane. However, cortical actin also exists between thickenings, so binding to actin alone cannot explain the preferential localization of SuSy over the thickenings (Salnikov et al., in press). It may well be that other proteins that co-precipitate with P-SuSy (see discussion above) have a role in establishing its patterned localization in tracheary elements. One cannot yet rule out the possibility that SuSy could bind 41 directly to cellulose synthases, which also increase dramatically at the initiation of secondary wall formation (Pear et al., 1996). However, this possibility may be harder to rationalize with the idea that cellulose synthases move rapidly through the plasma membrane as they spin out cellulose microfibrils (reviewed in Blanton and Haigler, 1996). Can SuSy gene expression regulate the rate and extent of cellulose synthesis? All available data suggest that SuSy is a critical partner in high-rate secondary wall cellulose synthesis. A second question is to what degree, if any, the rate and extent of secondary wall cellulose synthesis are regulated at the level of expression of genes encoding SuSy. In several heterotrophic systems, SuSy activity has been correlated with mRNA and/or protein levels (Nguyen-Quoc et al., 1990; Crespi et al., 1991; Sebkova et al., 1995). However, SuSy activity often seems to be greatly excessive compared to the actual metabolic requirements for starch synthesis. Each successive mutation of maize Sh1 or Sus1 reduced endosperm starch content by only an additional 22–25% (Chourey et al., 1998). In commercial tomato fruits, SuSy enzyme activity assayed in vitro was at least 10 times greater than maximum growth rate, including respiration (Sun et al., 1992). Tomato fruits with only 1–2% residual soluble SuSy activity after antisense suppression showed no change in starch or sugar accumulation (Chengappa et al., 1999; D’Aoust et al., 1999). In developing potato tubers, no change in starch content was observed until SuSy was suppressed to below 30% of control levels (Zrenner et al., 1995). Similarly, SuSy activity may have only weak control over the synthesis of cellulose, at least during primary wall synthesis. Although perhaps partially explained by the discovery of a third maize gene encoding SuSy (Carlson et al., 2000), a double sh1 sus1-1 maize mutant with only 0.51% of normal SuSy activity in the endosperm grows normally (Chourey et al., 1998). As discussed previously, we showed that potato tubers with only 4% residual soluble SuSy activity had about 70% of the wild-type level of (mostly primary wall) cellulose. Similarly, in developing carrot tap roots, transgenic plants with repressed SuSy activity (0.7–29% of wild type) showed stronger reduction in starch content (about 38% of wild type) than in crystalline cellulose content (about 63% of wild type). Notably, only 0.7% of normal, soluble SuSy activity supported about the same levels of starch and cellulose content as 29% residual activity (Tang and Sturm, 1999). Such observations can be explained by metabolic plasticity, enzymes in excess of requirements, or low flux control coefficient for a particular enzyme (Chengappa et al., 1999). However, it will also be necessary to manipulate expression and activity of SuSy in cells or tissues engaged primarily in secondary wall cellulose synthesis to obtain more direct evidence about this case. Evidence has been presented that hypoxic wheat roots have increased SuSy activity (as determined by histochemistry) and increased (150–200%) cellulose content, including in endodermal cells with secondary cell walls (Albrecht et al., 2000). Structure of cellulose synthases that may relate to carbon partitioning to cellulose synthesis Studies on cellulose synthesis entered a new era with the identification of two cotton genes (now called GhCesA-1 and GhCesA-2) that are homologues of the CesA genes encoding the catalytic subunit of bacterial cellulose synthases (Pear et al., 1996; Haigler and Blanton, 1996; reviewed in Delmer, 1999a). These two genes, while clearly distinct, are highly homologous. They also show high homology to CesA genes now identified in many plants including an additional one, GhCesA-3, that may be expressed preferentially during cotton fiber primary wall synthesis. (The CesA gene family has been discussed recently in Holland et al., 2000 and Richmond and Somerville, this issue; for a complete listing, see www.cellwall.stanford.edu/cellwall/.) Northern analysis of total RNA isolated from roots, leaves, and flowers revealed that GhCesA-1 and GhCesA-2 had only low expression in non-fiber tissues. Both genes were highly expressed from the onset of cotton fiber secondary wall synthesis, with expression continuing at high levels throughout this phase. The discovery that Arabidopsis mutants impaired in cellulose deposition map to CesA loci provided key genetic evidence for a role for these genes in cellulose synthesis (Arioli et al., 1998; Taylor et al., 1999). With the complete sequencing of the Arabidopsis genome (Arabidopsis Genome Initiative, 2000) and extensive ESTs available for maize, it is now clear that both of these species contain at least 10 distinct CesA genes. This will probably hold true for most other plants (Holland et al., 2000; Figure 9). There may be several reasons for the existence of so 42 Figure 9. An unrooted cladogram constructed as described by Holland et al. (2000) from comparisons of partial deduced CesA amino acid sequences. The partial sequences used spanned the region just downstream of the second conserved D residue to the C-terminal end of the proteins. For GhCesA-1 and AtCesA-2, these regions begin with amino acid residues 469 or 572, respectively. Accession numbers for all sequences are listed in Holland et al. (2000), except for the following additional sequences: GhCesA-3 (AAD39534); ZeCesA-1, -2, and -3 (AF323039, AF323040, AF323041). many CesA genes in one species. Emerging data indicate that at least some of them have tissue-specific or developmental-stage-specific expression, and the clustering of deduced proteins in phylogenetic analyses has some predictive value for cell-specific expression (Holland et al., 2000; Richmond and Somerville, 2000). One entire clade (labeled in bold in Figure 9) has now been analyzed in this way, and all members including GhCesA-1 and GhCesA-2, two CesA proteins encoded by genes identified from developing xylem cDNA libraries of poplar, and two Arabidopsis CesA proteins are uniquely expressed in cells undergoing secondary wall cellulose synthesis. However, it should be noted in Figure 9 that two other genes associated with secondary wall synthesis in vascular tissues are not included in this clade, AtCesA-7 (Turner and Somerville, 1997; Taylor et al., 1999) and ZmCesA-8 (Holland et al., 2000). Partial analysis of deduced proteins in another clade indicates expression predominantly in cells undergoing primary wall synthesis. More recently, we used PCR with primers based on conserved sequences in other CesA genes to identify three distinct CesA mRNAs that were expressed during rapid, secondary wall cellulose synthesis as tracheary elements differentiated in culture from isolated mesophyll cells of Zinnia elegans. The three genes also had different 3′ -untranslated regions (S. Hwang and C. Haigler, unpublished), and their deduced protein sequences, when added to the phylogenetic analysis (Figure 9), fall within the same clade as most of the other dicot CesA proteins involved in secondary wall synthesis. This observation further confirms the usefulness of this type of analysis. It is clear that more than one CesA gene can be expressed in the same cell type at the same time in development. One reason for this might be to provide redundancy for protection against mutation in such important genes or to provide a mechanism to enhance rates of cellulose synthesis when, for example, GhCesA-1 and -2 are expressed at the same time in the cotton fiber. Another emerging possibility is that two similar, but non-identical, pairs of CesA proteins must function together in order to allow rosette assembly or to create two non-identical catalytic sites that function to catalyze polymerization of alternating residues in the glucan chain. This latter possibility represents a 43 variation on other two-site models for cellulose synthase where the two sites have been proposed to reside within the same catalytic subunit (Vergara and Carpita, 2001, in this issue). With respect to carbon partitioning, it remains to be determined whether SuSy functions to donate carbon to all forms of CesA. It may be that some CesA proteins have certain motifs that favor direct interaction (or at least involvement) of SuSy as donor for UDPGlc. If SuSy localizes to the plasma membrane via interaction with a complex containing cortical actin and not directly with CesA, then it may be that any CesA can function coordinately with SuSy. However, as indicated earlier, SuSy seems to assume a more important role for secondary cell wall synthesis, and future work will be needed to see if a CesA gene involved in primary wall synthesis can functionally complement a secondary wall CesA gene and function in coordination with SuSy. Lack of homology between cellulose and callose synthases, both of which appear to interact with SuSy, may suggest that factors other than a SuSy-binding motif in the glucan synthases control the interaction with SuSy. A family of at least 10 genes that share distinct homology with yeast (1→3)β-D-glucan synthases (FKS1-type genes; see Delmer, 1999) have been identified from Arabidopsis genomic sequencing. Other such genes have also been identified through their expression in tobacco pollen tubes (Doblin et al., 2001), cotton fibers (Cui et al., 2001), or during cell plate formation (Hong et al., 2001). The protein encoded by the latter gene was shown to localize to the cell plate during callose synthesis, along with SuSy and several other proteins. Future work will be necessary to determine whether there is any specific sequence within the multiple plant glucan synthase that is important for coordinate function with SuSy. Other enzymes that may relate to control of carbon partitioning to cellulose The changing rates of cellulose synthesis at different stages of cotton fiber development allow hypotheses to be formulated about other enzymes that may modulate carbon partitioning to cellulose synthesis. In addition to SuSy and cellulose synthase, enzyme activities that increase with the onset of secondary wall deposition include UDP-Glc pyrophosphorylase (EC 2.7.7.9) (Wäfler and Meier, 1994) and sucrose-phosphate synthase (SPS; EC 2.4.1.14) (Tummala, 1996). En- zyme activities that do not increase consistently with the transition from primary to secondary wall synthesis include invertases (β-fructofuranosidase; EC 3.2.1.26), glucose-6-phosphate 1-dehydrogenase (EC 1.1.1.49), phosphofructokinases including ATPdependent 6-phosphofructokinase (EC 2.7.1.11) and pyrophosphate:fructose-6-phosphate 1-phosphotransferase (EC 2.7.1.90) (Wäfler and Meier, 1994; Basra et al., 1990). However, total invertase activity, including cell-wall-bound and vacuolar acid invertase and cytoplasmic alkaline invertase, was by far the highest activity measured throughout fiber development among a group of enzymes including UDP-Glc pyrophosphorylase and the phosphofructokinases (but excluding SuSy, SPS, and cellulose synthase) (Wäfler and Meier, 1994). High invertase and/or SuSy activity is consistent with the sucrose pool in cotton fibers depositing secondary wall being quite small compared to the glucose and fructose pools (Carpita and Delmer, 1981; Jacquet, 1982; Basra et al., 1990; Martin, 1999). In other cell types, invertases do affect the size of intracellular sucrose pools, but they appear to be most involved in regulating plant processes such as phloem unloading, control of differentiation, and provision of glycolytic substrates, not in determining sink strength (Sturm et al., 1995; Sturm and Tang, 1999). UDP-Glc pyrophosphorylase could be involved in cellulose synthesis if it consumed UTP and released PPi while converting glucose-1-phosphate to UDPGlc. Although the free pool of UDP-Glc is not the immediate substrate for secondary wall cellulose synthesis facilitated by SuSy, free UDP-Glc and fructose6-phosphate can be used to synthesize more sucrose to support cellulose synthesis (see below). The size of the UDP-Glc pool increases at the secondary wall stage of fiber development (Franz, 1969; Carpita and Delmer, 1981; Martin, 1999). Operation of UDP-Glc pyrophosphorylase in the direction of UDP-Glc synthesis is consistent with the freely reversible nature of this enzyme and with its role in activating glucose residues for glycogen synthesis in animal tissues. However, based on biochemical work, it was hypothesized that UDP-Glc pyrophosphorylase might not be under extensive metabolic control or regulate the UDP-Glc pool primarily through enzyme abundance (Turnquist and Hanson, 1973; Quick and Schaeffer, 1996). The 1.8-fold increase in UDP-Glc pyrophosphorylase activity during secondary wall deposition in cotton fibers (Wäfler and Meier, 1994) has been judged by others to be physiologically insignificant (Quick and Schaffer, 1996). In contrast, the activ- 44 ity of UDP-Glc pyrophosphorylase in the UDP-Glc synthetic direction was hypothesized to be highly related to the abundance of its substrates, UTP and glucose-1-phosphate (Turnquist and Hanson, 1973). This possibility is consistent with the large increase in the glucose-6-phosphate pool, which is readily isomerized to glucose-1-phosphate by phosphoglucomutase (EC 5.4.2.2), in secondary wall stage cotton fibers (Martin, 1999). Supporting the idea that the amount of UDP-Glc pyrophosphorylase is not a strong point of metabolic control, transgenic potato plants with only 4–5% of normal enzyme activity in their tubers due to expression of an antisense gene under control of the 35S CaMV promoter showed no overall phenotypic differences and no change in the growth and development of tubers (Zrenner, 1993). However, these results must be interpreted cautiously in terms of cellulose synthesis because possible changes in cell wall composition were not determined, the extent of gene down-regulation in other parts of the potato plants was not established, and tubers do not contain a strong cellulose sink, being mostly composed of parenchyma cells with primary walls. Further work to manipulate the level of UDP-Glc pyrophosphorylase in secondary-wall-synthesizing cells is needed. Sucrose phosphate synthase, which synthesizes sucrose phosphate from fructose-6-phosphate and UDP-Glc, when acting coordinately with sucrosephosphatase (EC 3.1.3.24), is a likely candidate to regulate synthesis of sucrose within cellulose sink cells. Biochemical experiments and analysis of transgenic plants indicate that SPS is the key regulator of the rate and extent of sucrose synthesis in photosynthetic cells. The extensive regulatory mechanisms known for SPS (e.g. glucose-6-phosphate activation and Pi inhibition) could also allow fine control of flux through sucrose in heterotrophic cells (Huber and Huber, 1996; Winter and Huber, 2000). Cotton fibers do synthesize sucrose; much of the glucose supplied to cultured ovules is subsequently converted back to sucrose within the attached fibers (Carpita and Delmer, 1981; Martin, 1999). Sucrose synthesis within fibers may be necessary if invertases cleave part of the translocated sucrose and/or recycle fructose released by the degradative action of SuSy. The latter possibility also advantageously removes cytoplasmic fructose, which is an end-product inhibitor of SuSy (Doehlert, 1987). We have shown that SPS activity increases 2- to 4fold between primary and secondary wall synthesis in cotton fibers grown in planta and in culture (Tummala, 1996) and during deposition of secondary walls in tra- cheary elements differentiating in culture (Babb and Haigler, 2000). Radiolabel from pulsed 14 C-glucose does not accumulate substantially in fructose within cultured cotton fibers (Martin, 1999), suggesting rapid phosphorylation and cycling of the metabolically active fructose pool. These biochemical data, together with increased fiber wall thickness in transgenic cotton plants constitutively over-expressing spinach SPS (Haigler et al., 2000a, b, c), suggest that SPS activity can affect the rate and extent of secondary wall cellulose synthesis. However, further work is needed to separate the relative roles of the source and the sink in causing this effect. A general model for regulation of intracellular carbon partitioning to cellulose synthesis We predict that there is tight metabolic control of the timing, rate, and extent of cellulose synthesis due to unusual features of the process. The large amount of cellulose synthesized by plants is a metabolic ‘dead end’; it is not degraded in planta for the purpose of recycling carbon to alternative metabolic uses. Furthermore, the synthesis of cellulose at any particular moment by an established plant is optional. Additional cellulose synthesis will promote growth and development of specialized cell types, but it is not required for immediate survival that depends on generating ATP and maintaining core metabolism. Therefore, when fixed carbon is limited under stress, evolutionarily successful plants must partition carbon preferentially toward survival and not cellulose synthesis. We envision that plants synthesize cellulose when conditions are more nearly optimal, especially the large amounts of cellulose required for secondary walls. The relative amount and specific location of PSuSy could be critical for partitioning carbon to different β-glucan polymers, particularly cellulose and callose. In contrast, S-SuSy may partition carbon to other demands such as respiration, building blocks for growth, and/or deposition of storage materials such as starch. The diagram in Figure 10 shows a model for secondary wall stage cotton fibers in which most SuSy exists as P-SuSy, partitioning at least 80% of incoming carbon into secondary wall cellulose synthesis. In this model, the association of P-SuSy with cellulose synthase could be a molecular on/off switch for cellulose synthesis, with higher P-SuSy/S-SuSy ratios occurring during high-rate secondary wall cellulose synthesis. The latter possibility is supported by our biochemical 45 Figure 10. Diagram of a cotton fiber during active secondary wall synthesis. The predominant form of SuSy is P-SuSy (upper left) shown channeling UDP-Glc to the cellulose synthase. All enzymes shown have been shown to exist in cotton fibers. The red arrows indicate reactions that would be involved in cellulose synthesis, including production of cytoplasmic UDP-Glc so that fructose released by SuSy can be used by SPS to synthesize more sucrose. The inset shows a hypothesis about how these metabolic loops might change under stress if P-SuSy became S-SuSy. High-rate cellulose synthesis would be shut down, and S-SuSy could participate with SPS in a cycle of sucrose synthesis and degradation (black arrows) that could supply fructose-6-phosphate (F6P) for glycolysis and UDP-Glc for ‘survival’ metabolism. When conditions improved, cellulose synthesis and growth could quickly resume if S-SuSy converted to P-SuSy. and immunolocalization results in primary and secondary wall stage cotton fibers. By cryogenic methods that should have preserved in vivo locations of soluble proteins (Nicolas and Bassot, 1993), we did not detect substantial SuSy in the cytoplasm of differentiating cotton fibers (Figure 2) or tracheary elements (Salnikov et al., in press). Even though dilution in the large cytoplasmic volume could partially explain such an observation, in the model we have indicated by a dashed line the possibility that S-SuSy is rare in vivo when conditions are optimal for secondary wall cellulose deposition. During secondary wall synthesis, a small amount of S-SuSy and/or the invertases would cleave sucrose to provide hexoses for maintenance metabolic processes. Similarly, fructose derived from S-SuSy could be phosphorylated to support general metabolism. The diagram does not depict differences in cellulose synthase proteins that are discussed above, which could also help to regulate the conditions un- 46 der which and extent to which S-SuSy could become P-SuSy. A callose synthase is shown to be idle in the plasma membrane, but this could be simultaneously activated during the developmental transition between primary and secondary wall synthesis by association with a portion of the P-SuSy. Developmentally controlled callose synthesis could require a specific association with P-SuSy as has been suggested for callose synthase in forming cell plates (Hong et al., 2001). Callose synthesis could also quickly replace cellulose synthesis after wounding or under stress if P-SuSy preferentially associated with callose synthases that were latent in the plasma membrane. Alternatively, wound- or stress-induced callose synthesis could use the free pool of UDP-Glc; in this case, P-SuSy changing to S-SuSy could increase the concentration of cytoplasmic UDP-Glc. In in vitro assays, cellulose and callose synthesis are promoted by low and high concentrations of exogenous UDP-Glc, respectively. A labile association of P-SuSy with cellulose synthase could also explain why it has been difficult to synthesize cellulose rather than callose in vitro (Delmer 1995, 1999a). Although a distinct family of callose synthase genes has been identified as already discussed, we still cannot exclude the possibility that a cellulose synthase might begin to synthesize callose under wounding conditions, including such factors as high calcium and high UDP-Glc (Delmer and Amor, 1995). This possibility is supported by research on in vitro cellulose synthesis by membrane preparations of Dictyostelium discoideum; this slime mold is not known to synthesize callose in vivo, but about 10% callose is made in vitro (Blanton and Northcote, 1990). The model also takes into account the existence of invertases, UDP-Glc pyrophosphorylase, SPS, and pyrophosphate:fructose-6-phosphate 1-phosphotransferase in cotton fibers. When UDPGlc pyrophosphorylase functions in the direction of UDP-Glc synthesis as shown, PPi is produced. This could be consumed by conversion of fructose6-phosphate to fructose 1,6-diphosphate via the pyrophosphate:fructose-6-phosphate 1-phosphotransferase. The activity of this enzyme is over 4-fold higher than the ATP-dependent phosphofructokinase throughout cotton fiber secondary wall deposition (Wäfler and Meier, 1994). The phosphorylation by fructokinase of the fructose released by the degradative action of SuSy consumes one UTP or ATP, but this is a minor energetic cost compared to the potential of one fructose-6-phosphate molecule to yield 36 ATP mole- cules through glycolysis. Therefore, the channeling of UDP-Glc by SuSy to cellulose synthase can, with the aid of fructokinase, also feed the energy requirements of the fiber. In this model, the free pool of UDP-Glc supports cellulose synthesis only indirectly in so far as it can contribute to more sucrose synthesis in the fiber. However, when cultured ovules are fed 14 Cglucose, all of the radioactivity in cellulose and callose must pass through the free UDP-Glc and intracellular sucrose pool on its way to P-SuSy if this model is correct. Therefore, our previous research that did not distinguish between a free pool of UDP-Glc and channeled UDP-Glc (Carpita and Delmer, 1981) can still be interpreted to show that the rate of UDP-Glc turnover matched well the combined rates of cellulose and callose synthesis. The model also shows synthesis (via SPS and sucrose phosphate phosphatase) and degradation (via Pand S-SuSy and invertases) of sucrose within the same cell, a phenomenon that has been described from other cell types as a ‘futile cycle’ with energetic cost. However, the cycle is not ‘futile’ in that it may provide potential for very sensitive regulation of the direction of flux through the sucrose pool according to changing needs for sucrose use (reviewed in Huber and Huber, 1996). The ‘futile cycle’ was previously described in cells or tissues that would be most enriched in primary wall synthesis and/or starch storage or mobilization, including heterotrophic suspension culture cells (Dancer et al., 1990; Wendler et al., 1990), bean cotyledons (Geigenberger and Stitt, 1991), and potato tubers (Geigenberger and Stitt, 1993). Our model shows that concurrent sucrose degradation (via P-SuSy) and synthesis (via SPS) could be a manifestation of enhancing cellulose synthesis via recycling of the fructose released when P-SuSy channels UDP-Glc to cellulose synthase. The enzymes and metabolites involved are the same as those described for the ‘futile cycle’, but here P-SuSy pulls use of sucrose toward cellulose. We predict that simultaneous sucrose synthesis and degradation would occur intensively during high-rate cellulose synthesis for secondary wall deposition and at a lower level during primary wall synthesis. Furthermore, we hypothesize that under stress causing reduced carbohydrate supply, high-rate cellulose synthesis might be stopped as P-SuSy becomes SSuSy (boxed inset of Figure 10). Here, a simple ‘futile cycle’ based on S-SuSy, SPS, and fructokinase could provide fructose-6-phosphate for glycolysis and UDPGlc for survival metabolic processes. This metabolic 47 holding pattern could quickly change to support cellulose synthesis if conditions once more become optimal and S-SuSy reverted to P-SuSy associated with cellulose synthase. By using the same basic metabolic loops but regulating the localization of SuSy, the cell could respond quickly and adaptively to synthesize cellulose only when conditions can support growth in addition to survival. Because of the importance ascribed to allocating carbon to cellulose synthesis, the regulatory mechanisms causing the switch between SSuSy and P-SuSy could well be quite precise, but variable between different cell types and species. One would expect variability as this regulation has become most advantageously integrated with different signal transduction cascades and different cell biological systems operating at different times in development and in response to different environmental signals. Similar variability has been discussed in the context of control of metabolic flux toward starch synthesis (Smith, 1999). It may well be that continuing research on control of carbon partitioning to cellulose synthesis will reveal a spectrum of regulatory systems based on changing combinations of several regulatory elements. This review has focused primarily on posttranscriptional/post-translational control of carbon partitioning. However, it is also clear that sugars are important regulators of gene transcription (Koch et al., 1996; Koch, 1996; Sheen et al., 1999). Clearly, some of the genes encoding SuSy and invertase in growing plant organs are regulated by sugar supply (Koch et al., 1992, 1996). Thus, a complete understanding of the complexities of regulation of carbon partitioning will undoubtedly require a deeper understanding of the interplay between regulation of substrate pool sizes, post-translational regulation of activity and localization of important enzymes, and regulation of gene expression. This increased knowledge should allow design of strategies to change the rate and extent of cellulose synthesis for crop improvement. Factors such as feed-back inhibition of elevated substrate pools and product degradation may make it difficult to up-regulate accumulation of products made and stored in the cytoplasm or cytoplasmic organelles (Kinney, 1998). In contrast, we are optimistic that cellulose synthesis and storage can be manipulated because cellulose is removed from its newly polymerized form by crystallization, and the stable, crystalline, microfibrils exists in the cell wall across the plasma membrane from major metabolic pools supporting their synthesis. Therefore, exploring diverse strategies to advantageously manipulate carbon partitioning to cellulose synthesis provides an intriguing challenge for future research. Acknowledgements We gratefully acknowledge the generous gift of rabbit polyclonal antibody prepared against bean SuSy from Heather Ross, Scottish Crop Research Institute, Dundee, UK, that was used for western blotting, immunoprecipitation, and parallel immunolocalization studies. We also thank Uwe Sonnewald for providing the potato tubers used in these studies and Mark Grimson for development of cryogenic methods for cotton fiber fixation. This work was partially supported by NSF grant DBI 9872627 to D.P.D. and C.H.H., by grant DE-FH-03-963ER20238 to D.P.D. from the U.S. Department of Energy, and by grants from Cotton Incorporated, Raleigh, NC and the Texas Advanced Research and Technology Programs to C.H.H. References Albrecht, G.O., Klotke, J. and Sophia, B. 2000. The increase in sucrose synthase activity correlates with a higher content of cellulose in wheat roots suffering from oxygen deficiency. Abstract 1020. In: Proceedings of Plant Biology 2000, 15–19 July, San Diego, CA. American Society of Plant Physiologists, Rockville, MD, [http://www.aspp.org/annual− meeting/pb-2000/2000.htm]. Amor, Y., Haigler, C.H., Wainscott, M., Johnson, S. and Delmer, D.P. 1995. A membrane-associated form of sucrose synthase and its potential role synthesis of cellulose and callose in plants. Proc. Natl. Acad. Sci. USA 92: 9353–9357. Andrawis, A., Solomon, M. and Delmer, D.P. 1993. Cotton fiber annexins: a potential role in the regulation of callose synthase. Plant J. 3: 763–772. Anguenot, R., Yelle, S. and Nguyen-Quoc, B. 1999. Purification of tomato sucrose synthase phosphorylated isoforms by Fe(III)immobilized metal affinity chromatography. Arch. Biochem. Biophys. 365: 163–169. ap Rees, T. 1984. Sucrose metabolism. In: D.H. Lewis (Ed.) Storage Carbohydrates in Vascular Plants: Distribution, Physiology, and Metabolism, Cambridge University Press, Cambridge, UK, pp. 53–73. Arabidopsis Genome Initiative. 2001. Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408: 796–815. Arioli, T., Peng, L., Betzner, A.S., Burn, J., Wittke, W., Herth, W., Camilleri, C., Höfte, H., Plazinske, R., Birch, R., Cork, A., Glover, J., Redmond, J. and Williamson, R.E. 1998. Molecular analysis of cellulose biosynthesis in Arabidopsis. Science 279: 717–720. Babb, V.M. and Haigler, C.H. 2000. Exploration of a role for sucrose phosphate synthase in cellulose synthesis during secondary cell wall deposition. Abstract 319. In: Proceedings of Plant Biology 2000, 15–19 July, San Diego, CA. American Society of Plant Physiologists, Rockville, MD [http://www.aspp.org/annual− meeting/pb-2000/2000.htm]. 48 Basra, A.S. and Malik, C.P. 1984. Development of the cotton fiber. Int. Rev. Cytol. 89: 65–113. Basra, A.S., Sarlach, R.S., Nayyar, H. and Malik, C.P. 1990. Sucrose hydrolysis in relation to development of cotton (Gossypium spp.) fibres. Indian J. Exp. Biol. 28: 985–988. Blakeney, A.B., Harris, P.J., Henry, R.J. and Stone, B.A. 1983. Simple and rapid preparation of alditol acetates for monosaccharide analysis. Carbohydrate Res. 113: 291–299. Blanton, R.L. and Haigler, C.H. 1996. Cellulose biosynthesis. In: M. Smallwood, J.P. Knox and D.J. Bowles (Eds.) Membranes: Specialized Functions in Plants, BIOS Scientific Publishers, Oxford, UK, pp. 57–75. Blanton, R.L. and Northcote, D.H. 1990. A 1,4-β-sc d-glucan synthase system from Dictyostelium discoideum. Planta 180: 324–332. Blumenkrantz. B. and Asboe-Hansen, G. 1973. New method for quantitative determination of uronic acids. Anal. Biochem. 54: 484–489. Brewin, N.J., Butcher, G.W., Galfre, G., Larkins, A.P., Wells, B., Wood, E.A. and Robertson, J.G. 1986. Immunochemical analysis of the legume root nodule. In: T.L. Wang (Ed.) Immunology in Plant Science, Cambridge University Press, Cambridge, UK, pp. 155–170. Brown, R.M. Jr., Saxena, I.M. and Kudlicka, K. 1996. Cellulose biosynthesis in higher plants. Trends Plant Sci. 1: 149–156. Buchala, A.J. 1987. Acid β-fructofuranoside fructohydrolase (invertase) in developing cotton (Gossypium arboreum L.) fibres and its relationship to β-glucan synthesis from sucrose fed to the fibre apoplast. J. Plant Physiol. 127: 219–230. Buckeridge, M.S., Vergara, C.E. and Carpita, N.C. 1999. The mechanism of synthesis of a mixed-linkage (1→3), (1→4)β-D-glucan in maize. Evidence for multiple sites of glucosyl transfer in the synthase complex. Plant Physiol 120: 1105–1116. Burton, R.A., Gibeaut, D.M., Bacic, A., Findlay, K., Roberts, K., Hamilton, A., Baulcombe, D.C. and Fincher, G.B. 2000. Virusinduced silencing of a plant cellulose synthase gene. Plant Cell: 12: 691–706. Carlson, S. J. and Chourey, P. 1996. Evidence for plasma membrane-associated forms of sucrose synthase in maize. Mol. Gen. Genet. 252: 303–310. Carlson, S.J., Chourey, P.S. and Helentjaris, T. 2000. Evidence for a third SuSy gene in maize from analysis of sucrose synthase mutants. Abstract 161. In: Proceedings of Plant Biology 2000, 15–19 July, San Diego, CA. American Society of Plant Physiologists, Rockville, MD [http://www.aspp.org/annual− meeting/pb2000/2000.htm]. Carpita, N.C. and Delmer, D.P. 1981. Concentration and metabolic turnover of UDP-glucose in developing cotton fibers. J. Biol. Chem. 256: 308–315. Chen, Y.-C. and Chourey, P.S. 1989. Spatial and temporal expression of the two sucrose synthase genes in maize: immunohistological evidence. Theor. Appl. Genet. 78: 553–559. Chengappa, S., Guilleroux, M., Phillips, W. and Shields, R. 1999. Transgenic tomato plants with decreased sucrose synthase are unaltered in starch and sugar accumulation in the fruit. Plant Mol. Biol. 40: 213–221. Chourey P.S., Chen Y.-C. and Miller, M.E. 1991. Early cell degeneration in developing endosperm is unique to the Shrunken mutation in maize. Maydica 36: 141–146 Chourey, P.S. and Miller, M.E. 1995. On the role of sucrose synthase in cellulose and callose biosynthesis in plants. In: H.G. Pontis, G.L. Salerno and E.J. Echeverria (Eds.) Sucrose Metabolism, Biochemistry, Physiology, and Molecular Biology, American Society of Plant Physiologists, Rockville, MN, pp. 80–87. Chourey, P.S., Taliercio, E.W., Carlson, S.J. and Ruan, Y.-L. 1998. Genetic evidence that the two isozymes of sucrose synthase present in developing maize endosperm are critical, one for cell wall integrity and the other for starch biosynthesis. Mol. Gen. Genet. 259: 88–96. Copeland, L. 1990. Enzymes of sucrose metabolism. Meth. Plant Biochem. 3: 74–85. Crespi, M.D., Zabaleta, E.J., Pontis, H.G. and Salerno, G.L. 1991. Sucrose synthase expression during cold acclimation in wheat. Plant Physiol. 96: 887–891. Cui, X., Shin, H., Song, C.C, Laosinchai, W., Amano, Y. and Brown, R.M. Jr. 2001. A putative plant homolog of the yeast β-1,3-glucan synthase subunit FKS1 from cotton (Gossypium hirsutum L.) fibers. Planta, in press Dancer, J.E., Hazfield, W.D. and Stitt, M. 1990. Cytosolic cycles regulate the accumulation of sucrose in heterotrophic cellsuspension cultures of Chenopodium rubrum. Planta 182: 223– 231. D’Aoust, M.-A., Yelle, S. and Nguyen-Quoc, B. 1999. Antisense inhibition of tomato fruit sucrose synthase decreases fruit setting and the sucrose unloading capacity of young fruit. Plant Cell 11: 2407–2418. Delmer, D.P. and Albersheim, P. 1970. The biosynthesis of sucrose and nucleoside diphosphate glucoses in Phaseolus aureus. Plant Physiol. 45: 782–786. Delmer, D.P., Pear, J.R., Andrawis, A. and Stalker, D.M. 1995. Genes encoding small GTP-binding proteins analogous to mammalian rac are preferentially expressed in developing cotton fibers. Mol. Gen. Genet. 248: 43–51. Delmer, D.P. and Amor, Y. 1995. Cellulose biosynthesis. Plant Cell 7: 987–1000. Delmer, D.P. 1999a. Cellulose biosynthesis: Exciting times for a difficult field of study. Annu. Rev. Plant Physiol. Mol. Biol. 50: 245–276. Delmer, D.P. 1999b. Cellulose biosynthesis in developing cotton fibers. In: A.S. Basra (Ed.) Cotton Fibers: Developmental Biology, Quality Improvement, and Textile Processing, Haworth Press, New York, pp. 85–112. Dick, P.S. and ap Rees, T. 1976. Sucrose metabolism by roots of Pisum sativum. Phytochemistry 15: 255–259. Dixon, D.C., Seagull, R.W. and Triplett, B.A. 1994. Changes in the accumulation of α- and β-tubulin isotypes during cotton fiber development. Plant Physiol. 105: 1347–1353. Doblin, M.S., De Melis, L., Newbigin, E., Bacic, A. and Read, S.M. 2001. Pollen tubes of Nicotiana alata express two genes from different β-glucan synthase families. Plant Physiol, in press. Doehlert, D.C. 1987. Substrate inhibition of maize endosperm sucrose synthase by fructose and its interaction with glucose inhibition. Plant Sci. 52: 153–157. Doonan, J. and Clayton, L. 1986. Immunofluorescent studies on the plant cytoskeleton. In: T.L. Wang (Ed.) Immunology in Plant Science, Cambridge University Press, Cambridge, UK, pp. 111–136. Franz, G. 1969. Soluble nucleotides in growing cotton hair. Phytochemistry 8: 737–741. Fu, H. and Park, W.D. 1995. Sink- and vascular-associated sucrose synthase functions are encoded by different gene classes in potato. Plant Cell 7: 1369–1385. Geigenberger, P. and Stitt, M. 1991. A ‘futile’ cycle of sucrose synthesis and degradation is involved in regulating partitioning between sucrose, starch and respiration in cotyledons of germinating Ricinus communis L. seedlings when phloem transport is inhibited. Planta 185: 81–90. 49 Geigenberger, P. and Stitt, M. 1993. Sucrose synthesis catalyzes a readily reversible reaction in vivo in developing potato tubers and other plant tissues. Planta 189: 329–339. Guerin, J. and Carbonero, P. 1997. The spatial distribution of sucrose synthase isozymes in barley. Plant Physiol. 114: 55–62. Haigler, C.H. 1985. The functions and biogenesis of native cellulose. In: T.P. Nevell and S.H. Zeronian (Eds.) Cellulose Chemistry and its Applications, Ellis Horwood, Chichester, UK, pp. 30–83. Haigler, C.H. and Blanton, R.L. 1996. New hope for old dreams: evidence that plant cellulose synthase genes have finally been identified. Proc. Natl. Acad. Sci. USA 93: 12082–12085. Haigler, C.H. and Brown, R.M. Jr. 1986. Transport of rosettes from the Golgi apparatus to the plasma membrane in isolated mesophyll cells of Zinnia elegans during differentiation to tracheary elements in suspension culture. Protoplasma 134: 111–120. Haigler, C.H., Rao, N.R., Roberts, E.M., Huang, J.Y., Upchurch, D.P. and Trolinder, N.L. 1991. Cultured cotton ovules as models for cotton fiber development under low temperatures. Plant Physiol. 95: 88–96. Haigler, C.H., Hequet, E.F., Krieg, D.R., Strauss, R.E., Wyatt, B.G., Cai, W., Jaradat, T, Srinivas, N.G., Wu, C., Jividen, G.J. and Holaday, A.S. 2000a. Transgenic cotton with improved fiber micronaire, strength, length, and increased fiber weight. In: C.P. Dugger and D.A. Richter (Eds.) Proceedings of the 2000 Beltwide Cotton Conference, 4–8 January 2000, San Antonio, TX. National Cotton Council, Memphis, p. 483. Haigler, C.H., Cai, W, Martin, L.K., Tummala, J, Anconetani, R, Gannaway, J.G., Jividen, G.J. and Holaday, A.S. 2000b. Mechanisms by which fiber quality and fiber and seed weight can be improved in transgenic cotton growing under cool night temperatures. In: C.P. Dugger and D.A. Richter (Eds.) Proceedings of the 2000 Beltwide Cotton Conference, 4–8 January 2000, San Antonio, TX. National Cotton Council, Memphis, p. 483. Haigler, C.H., Holaday, A.S., Wu, C., Wyatt, B.G., Jividen, G.J., Gannaway, J.G., Cai, W.X., Hequet, E.F., Jaradat, T.T., Krieg, D.R., Martin, L. K., Strauss, R.E., Nagarur, S. and Tummala, J. 2000c. Transgenic cotton over-expressing sucrose phosphate synthase produces higher quality fibers with increased cellulose content and has enhanced seedcotton yield. Abstract 477. In: Proceedings of Plant Biology 2000, 15–19 July, San Diego, CA. American Society of Plant Physiologists, Rockville, MD [http://www.aspp.org/annual− meeting/pb-2000/2000.htm]. Hayashi, T., Read, S.M., Bussell, J., Thelen, M., Lin, F.-C., Brown, R.M. Jr. and Delmer, D.P. 1987. UDP-glucose: (1→3)β-glucan synthases from mung bean and cotton. Plant Physiol. 83: 1054– 1062. Hirai, N., Sonobe, S. and Hayashi, T. 1998. In situ synthesis of βglucan microfibrils on tobacco plasma membrane sheets. Proc. Natl. Acad. Sci. USA 95: 15102–15106. Holland, N., Holland, D., Helentjaris, T., Dhugga, K.S., Xoconostle-Cazares, B. and Delmer, D.P. 2000. A comparative analysis of the plant cellulose synthase (CesA) gene family. Plant Physiol. 123: 1313–1324. Hong, Z., Delauney, A.J. and Verma, D.P.S. 2001. A cell platespecific callose synthase and its interaction with phragmoplastin. Plant Cell, in press. Huber, S.C. and Akazawa, T. 1986. A novel sucrose synthase pathway for sucrose degradation in cultured sycamore cells. Plant Physiol. 81: 1008–1013. Huber, S.C. and Huber, J.L. 1996. Role and regulation of sucrosephosphate synthase in higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 47: 431–444. Huber, S.C., Huber, J.L. Gage, D.A., McMichael, R.W., Chourey, P.S, Hannah L.C. and Koch, K. 1996. Phosphorylation of serine15 of maize leaf sucrose synthase. Plant Physiol. 112: 793–802. Jaquet, J.P., Buchala, A.J. and Meier, H. 1982. Changes in the nonstructural carbohydrate content of cotton (Gossypium spp.) fibres at different stages of development. Planta 156: 481–486. Kawagoe, Y. and Delmer, D.P. 1997. Pathways and genes involved in cellulose biosynthesis. In: J.K. Setlow (Ed) Genetic Engineering, vol. 19, Plenum Press, New York, pp. 63–87. Kimura, S., Laosinchai, W., Itoh, T., Cui, X., Linder, R. and Brown, R.M. Jr. 1999. Immunogold labeling of rosette terminal cellulose-synthesizing complexes in the vascular plant Vigna angularis. Plant Cell 11: 2075–2085. Kinney, A.J. 1998. Manipulating flux through plant metabolic pathways. Curr. Opin. Plant Biol. 1: 173–177. Kloth, R.H. 1989. Changes in the level of tubulin subunits during development of cotton (Gossypium hirsutum) fiber. Physiol. Plant. 76: 37–41. Kobayashi, H. and Fukuda, H. 1994. Involvement of calmodulin and calmodulin-binding proteins in the differentiation of tracheary elements in Zinnia cells. Planta 194: 388–394. Koch, K.E. 1996. Carbohydrate-modulated gene expression in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 47: 509–540. Koch, K.E., Nolte, K.D., Duke, E.R., McCarty, D.R. and Avigne, W.T. 1992. Sugar levels modulate differential expression of maize sucrose synthase genes. Plant Cell 4: 59–69. Koch, K.E., Xu, J., Duke, E.R., McCarty, D.R., Yuan, C-X., Tan, B.-C. and Avigne, W.T. 1995. Sucrose provides a long distance signal for coarse control of genes affecting its metabolism. In: H.G. Pontis, G. Salerno and E. Echeverria (Eds.) Sucrose Metabolism, Biochemistry, Physiology, and Molecular Biology, American Society of Plant Physiologists, Rockville, MD, pp. 266–277. Koch, K.E., Wu, Y. and Xu, J. 1996. Sugar and metabolic regulation of genes for sucrose metabolism: potential influence of maize sucrose synthase and soluble invertase responses on carbon partitioning and sugar sensing. J. Exp. Bot. 47: 1179–1185. Kudlicka, K., Brown, R.M. Jr., Li, L., Lee, J.H., Shen, H. and Kuga, S. 1995. β-glucan synthesis in the cotton fiber. IV. In vitro assembly of the cellulose I allomorph. Plant Physiol. 107: 111–123. Kudlicka, K., Lee, J.H. and Brown, R.M. Jr. 1996. A comparative analysis of in vitro cellulose synthesis from cell-free extracts of mung bean (Vigna radiata, Fabaceae) and cotton (Gossypium hirsutum, Malvaceae). Am. J. Bot. 83: 274–284. Laporte, K., Rossignol, M. and Traas, J.A. 1993. Interaction of tubulin with the plasma membrane: tubulin is present in purified plasmalemma and behaves as an integral membrane protein. Planta 191: 413–416. Lindblom, S., Ek, P., Muszynska, G., Ek, B., Szczegielniak, J. and Engstrom, L. 1997. Phosphorylation of sucrose synthase from maize seedlings. Acta Biochim. Pol. 44: 809–817 Loog, M., Toomik, R., Sak, K., Muszynska, G., Jarv, J. and Ek, P. 2000. Peptide phorphorylation by a calcium-dependent protein kinase from maize seedlings. Eur. J. Biochem. 267: 337–343. Maas, C., Schaal, S. and Werr, W. 1990. A feedback control element near the transcription start site of the maize Shrunken gene determines promoter activity. EMBO J. 9: 3447–3452. Maltby, D., Carpita, N.C., Montezinos, D., Kulow, C. and Delmer, D.P. 1979. β-1,3-glucan in developing cotton fibers. Plant Physiol. 63: 1158–1164. Martin, L.K. 1999. Cool-temperature-induced changes in metabolism related to cellulose synthesis in cotton fibers. Ph.D. dissertation, Texas Tech University, Lubbock, TX. 50 Meinert, M.C. and Delmer, D.P. 1977. Changes in biochemical composition of the cell wall of the cotton fiber during development. Plant Physiol. 59: 1088–1097. Morrow, D.L. and Lucas, W.J. 1986. (1,3)-β-glucan synthase from sugar beet. I. Isolation and solubilization. Plant Physiol. 81: 171– 176 Mutsaers, H.J.W. 1976. Growth and assimilate conversion of cotton bolls (Gossypium hirsutum L.) 1. Growth of fruits and substrate demand. Ann. Bot. 40: 301–315. Nakai, T., Tonouchi, N., Konishi, T., Kojima, Y., Tsuchida, T., Yoshinaga, F., Sakae F. and Hayashi, T. 1999. Enhancement of cellulose production by expression of sucrose synthase in Acetobacter xylinum. Proc. Natl. Acad. Sci. USA 96: 14–18. Nguyen-Quoc, B., Krivitzky, M., Huber, S.C. and Lecharny, A. 1990. Sucrose synthase in developing maize leaves: regulation of activity by protein level during the import to export transition. Plant Physiol. 94: 516–523. Nicolas, T.N. and Bassot, J.M. 1993. Freeze substitution after fast-freeze fixation in preparation for immunocytochemistry. Microscopy Res. Tech. 24: 474–487. Nolte K.D. and Koch, K.E. 1993. Companion-cell specific localization of sucrose synthase in zones of phloem loading and unloading. Plant Physiol. 101: 899–905. Nolte, K.D., Hendrix, D.L., Radin, J.W. and Koch, K.E. 1995. Sucrose synthase localization during initiation of seed development and trichome differentiation in cotton ovules. Plant Physiol. 109: 1285–1293. Pear, J., Kawagoe, Y., Schreckengost, W., Delmer, D.P. and Stalker, D. 1996. Higher plants contain homologs of the CelA genes that encode the catalytic subunit of the bacterial cellulose synthases. Proc. Natl. Acad. Sci. USA 93: 12637–12642. Peng, L., Xiang, F., Roberts, E., Kawagoe, Y., Greve, L.C., Kreuz, K., Delmer, D.P. 2001. The experimental herbicide cgA 325’615 inhibits synthesis of crystalline cellulose and causes accumulation of non-crystalline β-1,4-glucan associated with CesA protein. Plant Physiol, in press. Pillonel, C., Buchala, A.J. and Meier, H. 1980. Glucan synthesis by intact cotton fibres fed with different precursors at the stages of primary and secondary wall formation. Planta 149: 306–312. Potikha, T.S., Collins, C.C., Johnson, D.I., Delmer, D.P. and Levine, A. 1999. The involvement of hydrogen peroxide in the differentiation of secondary walls in cotton fibers. Plant Physiol. 119: 849–858. Preston, R.D. 1974. The Physical Biology of Plant Cell Walls. Chapman and Hall, London. Quick, W.P. and Schaffer, A.A. 1996. Sucrose metabolism in sources and sinks. In: E. Zamski and A.A. Schaffer (Eds.) Photoassimilate Distribution in Plants and Crops: Source-Sink Relationships, (AUTHOR: PLEASE PROVIDE PUBLISHER DATA), pp. 115–156. Reymond, O.L. and Pickett-Heaps, J.D. 1983. A routine flat embedding method for electron microscopy of microorganisms allowing selection and precisely orientated sectioning of single cells by light microscopy. J. Microsc. 130: 79–84. Richmond, T.A. and Somerville, C.R. 2001. Integrative approaches to determining Csl function. Plant Mol. Biol., this issue. Roberts, A.W. and Haigler, C.H. 1990. Tracheary-element differentiation in suspension cultures of Zinnia requires uptake of extracellular Ca2+ . Experiments with calcium-channel blockers and calmodulin inhibitors. Planta 180: 502–509. Roberts, E.M., Nunna, R.R., Huang, J.Y., Trolinder, N.L. and Haigler, C.H. 1992. Effects of cycling temperatures on fiber metabolism in cultured cotton ovules. Plant Physiol. 100: 979– 986. Rollit, J. and Maclachlan, G.A. 1974. Synthesis of wall glucan from sucrose by enzyme preparations from Pisum sativum. Phytochemistry 13: 367–374. Ross, P., Mayer, R. and Benziman, M. 1991. Cellulose biosynthesis and function in bacteria. Microbiol. Rev. 55: 35–58. Ruan, Y.-L., Chourey, P.S., Delmer, D.P. and Perez-Grau, L. 1997. The differential expression of sucrose synthase in relation to diverse patterns of carbon partitioning in developing cotton seed. Plant Physiol. 115: 375–385. Ruan, Y.-L. and Chourey, P.S. 1998. A fiberless seed mutation in cotton is associated with lack of fiber cell initiation in ovule epidermis and alterations in sucrose synthase expression and carbon partitioning in developing seeds. Plant Physiol. 118: 399–406. Ryser, U. 1985. Cell wall biosynthesis in differentiating cotton fibres. Eur. J. Cell Biol. 39: 236–256. Ryser, U. 1992. Ultrastructure of the epidermis of developing cotton (Gossypium) seeds: suberin, pits, plasmodesmata, and their implications for assimilate transport into cotton fibers. Am. J. Bot. 79: 14–22. Ryser, U. 1999. Cotton fiber initiation and histodifferentiation. In: A.S. Basra (Ed.) Cotton Fibers: Developmental Biology, Quality Improvement, and Textile Processing, Haworth Press, New York, pp. 1–46. Salnikov, V.V., Grimson, M.J., Delmer, D.P. and Haigler, C.H. 2001. Sucrose synthase localizes to cellulose synthesis sites in tracheary elements. Phytochemistry, in press. Seagull, R.W. 1993. Cytoskeletal involvement in cotton fiber growth and development. Micron 24: 643–660. Sebkova, V., Unger, C., Hardegger, M. and Sturm, A. 1995. Biochemical, physiological, and molecular characterization of sucrose synthase from Daucus carota. Plant Physiol. 108: 75–83. Shaw, J.R., Ferl, R.J., Baier, J., St. Clair, D., Carson, C., McCarty, D.R. and Hannah, L.C. 1994. Structural features of the maize sus1 gene and protein. Plant Physiol. 106: 1659–1665. Shedletzky, E., Shmuel, M., Trainin, T., Kalman, S. and Delmer, D.P. 1992. Cell wall structure in cells adapted to growth on the cellulose-synthesis inhibitor 2,6-dichloro-benzonitrile (DCB): a comparison between two dicotylenonous plants and a graminaceous monocot. Plant Physiol. 100: 120–130. Sheen, J., Shou, L. and Jang, J.C. 1999. Sugars as signaling molecules. Curr. Opin. Plant Biol. 2: 410-418. Smith, A.M. 1999. Regulation of starch synthesis in storage organs. In: N.J. Kruger et al. (Eds.) Regulation of Primary Metabolic Pathways in Plants, Kluwer Academic Publishers, Dordrecht, Netherlands, pp. 173–193. Smook, G.A. 1992. Handbook for Pulp and Paper Technologists. Angus Wilde Publications, Vancouver. Sturm, A., Sebkova, V., Lorenz, K., Hardegger, M., Lienhard, S. and Unger, C. 1995. Development- and organ-specific expression of the genes for sucrose synthase and three isozymes of acid βfructofuranoside in carrot. Planta 195: 601–610. Sturm, A. and Tang, G.-Q. 1999. The sucrose-cleaving enzymes of plants are crucial for development, growth, and carbon partitioning. Trends Plant Sci. 4: 401–407. Sturm, A., Lienhard, S., Schatt, S. and Hardegger, M. 1999. Tissuespecific expression of two genes for sucrose synthase in carrot (Daucus carota L.). Plant Mol. Biol. 39: 349–360. Sun, J., Loboda, T., Sung, S.-J. and Black, C.C. Jr. 1992. Sucrose synthase in wild tomato, Lycopersicon chmielewskii, and tomato fruit sink strength. Plant Physiol. 98: 1163–1169. Sung, S.-J., Xu, D.-P. and Black, C.C. 1989. Identification of actively-filling sucrose sinks. Plant Physiol. 89: 1117–1121. 51 Tang, G.Q. and Sturm, A. 1999. Antisense repression of sucrose synthase in carrot (Daucus carota L.) affects growth rather than sucrose partitioning. Plant Mol. Biol. 41: 465–479. Taylor, N.G., Scheible, W.R., Cutler, S., Somerville, C.R., and Turner, S.R. 1999. The irregular xylem locus of Arabidopsis encodes a cellulose synthase required for secondary wall synthesis. Plant Cell 11: 769–780. Theander, O. and Westerlund, E. 1993. Quantitative analysis of cell wall components. In: H.G. Jung, D.R. Buxton, R.D. Hatfield and J. Ralph (Eds.) Forage Cell Wall Structure and Digestibility, American Society of Agronomy/CSSA/SSSA, Madison, WI, pp. 83–104. Todorov, I.T., Attaran, A. and Kearsey, S.E. 1995. BM28, a human member of the MCM2-3-5 family, is displaced from chromatin during DNA replication. J. Cell Biol. 129: 1433–1445. Tummala, J. 1996. Response of sucrose phosphate synthase activity to cool temperatures in cotton. M.S. thesis, Texas Tech University, Lubbock, TX. Turner, S. and Somerville, C.R. 1997. Collapsed xylem phenotype of Arabidopsis identifies mutants deficient in cellulose deposition in the secondary cell wall. Plant Cell 9: 689–701. Turnquist, R.L. and Hansen, R.G. 1973. Uridine diphosphphoryl glucose pyrophosphorylase. In: P.D. Boyer (Ed.) The Enzymes, vol. 8, part A, 3rd ed., Academic Press, New York, pp. 51–71. Updegraff, D.M. 1969. Semi-micro determination of cellulose in biological materials. Anal. Biochem. 32: 420–424. Vergara, C.E. and Carpita, N.C. 2001. Mixed-linkage β-glucan synthase and the CesA gene family in cereals. Plant Mol. Biol., this issue. Volman, G., Ohana, P. and Benziman, M. 1995. Biochemistry and molecular biology of cellulose synthesis. Carbohydrates Europe 12: 20–27. Wäfler, U. and Meier, H. 1994. Enzyme activities in developing cotton fibers. Plant Physiol. Biochem. 32: 697–702. Wang, F., Sanz, A., Brenner, M.L. and Smith, A. 1993. Sucrose synthase, starch accumulation, and tomato fruit sink strength. Plant Physiol. 101: 321–327. Wendel, J.F., Small, R.L., Cronn, R.C. and Brubaker, C.L. 1999. Genes, jeans, and genomes: reconstructing the history of cotton. In: L.W.D. van Raamsdonk and J.C.M. den Nijs (Eds.) Plant Evolution in Man-Made Habitats, Proceedings of the 7th Symposium IOPB (Amsterdam, 1998), Hugo de Vries Laboratory, Amsterdam, pp. 133–159. Wendler, R., Veith, R., Dancer, J., Stitt, M. and Komor, E. 1990. Sucrose storage in cell suspension cultures of Saccharum sp. (sugarcane) is regulated by a cycle of synthesis and degradation. Planta 183: 31–39. Winter, H. and Huber, S.C. 2000. Regulation of sucrose metabolism in higher plants: Localization and regulation of activity of key enzymes. Crit. Rev. Plant Sci. 19: 31–67. Winter, H., Huber, J.L. and Huber, S.C. 1997. Membrane association of sucrose synthase: changes during graviresponse and possible control by protein phosphorylation. FEBS Lett. 420: 151–155. Winter, H., Huber, J.L. and Huber, S.C. 1998. Identification of sucrose synthase as an actin-binding protein. FEBS Lett. 430: 205–208. Xu, D.-P., Sung, S.-J., Loboda, T., Kormanik, P.P. and Black, C.C. 1989. Characterization of sucrolysis via the uridine diphosphate and pyrophosphate-dependent sucrose synthase pathway. Plant Physiol. 90: 635–642. Zhang, X.-Q. and Chollet, R. 1997. Seryl-phosphorylation of soybean nodule sucrose synthase (nodulin-100) by a Ca2+ dependent protein kinase. FEBS Lett. 410: 126–130. Zhang, X.Q., Lund, A.A., Sarath, G., Cerny, R.L., Roberts, D.M. and Chollet, R. 1999. Soybean nodule sucrose synthase (nodulin100): further analysis of its phosphorylation using recombinant and authentic root-nodule enzymes. Arch. Biochem. Biophys. 371: 70–82. Zrenner, R., Willmitzer, L. and Sonnewald, U. 1993. Analysis of the expression of potato uridinediphosphate-glucose phyrophosphorylase and its inhibition by antisense RNA. Planta 190: 247–252. Zrenner, R., Salanoubat, M., Willmitzer, L. and Sonnewald, U. 1995. Evidence of the crucial role of sucrose synthase for sink strength using transgenic potato plants (Solanum tuberosum L.) Plant J. 7: 97–107.