| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Genetic Discovery, Crop Genetics Research and Development, Pioneer Hi-Bred International, Inc., A DuPont Company, 7300 NW 62nd Ave., Johnston, IA 50131
* Corresponding author (kanwarpal.dhugga{at}pioneer.com).
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONCONCLUSIONSREFERENCES |
|---|
20% of corn (Zea mays L.) grain in the United States contributes 3.5% of the volume and 2.5% of the energy equivalent of annual gasoline consumption. Cellulosic biomass has the potential to contribute substantially to the biofuels pool. Corn is the single-largest source of crop residue in the United States. An unaltered cell wall is recalcitrant to hydrolytic enzymes required for the conversion of its polysaccharide fraction into simple sugars before fermentation. Attempts at lowering lignin to increase stover digestibility are generally accompanied by a reduction in biomass. The complexity of the cellulose synthase system poses a challenge in increasing its activity through biotechnological means. Exploitation of natural variation may thus be a more productive route to increase the stover cellulose content. In comparison, the objective of reducing or altering hemicellulose for improved ethanol production as well as digestibility of the grain by monogastric animals may be relatively easier to accomplish through transgenic means. Availability of molecular tools for many of the steps in cell wall biosynthesis and modification has opened the heretofore inaccessible biotechnological avenues to alter the wall composition and perhaps structure for increased ethanol production.
Abbreviations: Bc1, Brittle culm-1 • Bk2, Brittle stalk-2 • bm, brown midrib • CBD, cellulose-binding domain • CesA, cellulose synthase catalytic subunit • Csl, cellulose synthase-like • CRP, Conservation Reserve Program • DDGS, distiller's dried grains with solubles • fra, fragile fiber • irx, irregular xylem • GAX, glucuronoarabinoxylan • GPI, glycophosphatidylinositol • HI, harvest index • MLG, mixed-linked glucan • UDP, uridine diphosphate
Genetic Discovery, Crop Genetics Research and Development, Pioneer Hi-Bred International, Inc., A DuPont Company, 7300 NW 62nd Ave., Johnston, IA 50131
* Corresponding author (kanwarpal.dhugga{at}pioneer.com).
With the world oil reserves projected to be depleted in about 40 years at the current pace of use, emphasis has shifted to alternative sources of liquid fuel. Currently, ethanol produced from 20% of corn (Zea mays L.) grain in the United States contributes 3.5% of the volume and 2.5% of the energy equivalent of annual gasoline consumption. Cellulosic biomass has the potential to contribute substantially to the biofuels pool. Corn is the single-largest source of crop residue in the United States. An unaltered cell wall is recalcitrant to hydrolytic enzymes required for the conversion of its polysaccharide fraction into simple sugars before fermentation. Attempts at lowering lignin to increase stover digestibility are generally accompanied by a reduction in biomass. The complexity of the cellulose synthase system poses a challenge in increasing its activity through biotechnological means. Exploitation of natural variation may thus be a more productive route to increase the stover cellulose content. In comparison, the objective of reducing or altering hemicellulose for improved ethanol production as well as digestibility of the grain by monogastric animals may be relatively easier to accomplish through transgenic means. Availability of molecular tools for many of the steps in cell wall biosynthesis and modification has opened the heretofore inaccessible biotechnological avenues to alter the wall composition and perhaps structure for increased ethanol production.
Abbreviations: Bc1, Brittle culm-1 • Bk2, Brittle stalk-2 • bm, brown midrib • CBD, cellulose-binding domain • CesA, cellulose synthase catalytic subunit • Csl, cellulose synthase-like • CRP, Conservation Reserve Program • DDGS, distiller's dried grains with solubles • fra, fragile fiber • irx, irregular xylem • GAX, glucuronoarabinoxylan • GPI, glycophosphatidylinositol • HI, harvest index • MLG, mixed-linked glucan • UDP, uridine diphosphate
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONCONCLUSIONSREFERENCES |
|---|
Ethanol, biodiesel, and butanol are the main types of commercially produced biofuels (Ezeji et al., 2007; Wright et al., 2006). Biodiesel consists of methyl- or ethyl-esterified fatty acids derived from plant oils or animal fat. Ethanol and butanol are currently made by fermentation of plant-derived sugars. Although not as advanced as that for fermentation, technology for thermochemical conversion of biomass into liquid fuels is also under development (Spath and Dayton, 2003).
Ethanol constitutes the bulk of the commercially produced biofuels, with biodiesel as a distant second. Butanol is essentially nonexistent as a biofuel at the present time, although it has been commercially produced in the past and there is renewed interest in its production (Dupont, 2006; Ezeji et al., 2007). Because of its higher calorific content and ease of delivery through existing oil pipelines, butanol is preferred to ethanol, residual water in which is a cause for pipeline corrosion.
Theoretical efficiency of ethanol production from glucose is 51%, that is, one CO2 molecule is produced for each molecule of ethanol; that of butanol is 41%, as a molecule of water is also produced in addition to two molecules of CO2 for each molecule of butanol. On a volumetric basis, butanol and ethanol contain 85 and 67% of the heat energy of gasoline, respectively. Because the densities of these two alcohols are quite similar, the amount of heat energy produced from a given amount of sugar is theoretically the same regardless of the type of alcohol produced. Toxicity of butanol at low concentrations to the fermenting organism and channeling of carbon into side reactions that make acetone and ethanol have been the main problems that have kept it from competing favorably with ethanol (Ezeji et al., 2007).
Plant-derived ethanol has been used as a gasoline supplement for several decades, the initial impetus for its production being the oil embargo of 1970s. In addition, ethanol has replaced methyl tertiary butyl ether as an oxygenate in many states (Lidderdale, 2003). Vehicles with flex engines that run on blends of 85% ethanol in gasoline are already on the market. Currently, nearly all the ethanol for transportation fuel is produced either from corn (Zea mays L.) grain (e.g., United States) or sugarcane (Saccharum L.; Brazil) (Shapouri et al., 2006).
The Biomass Research and Development Technical Advisory Committee, which guides the U.S. Congress on biofuels, envisions replacing 20% of transportation fuels in the U.S. by 2030 with biomass-derived fuels (English and Ewing, 2002). The current U.S. annual consumption of transportation fuels is 830 GL (220 billion gal) and is expected to continue an upward trend, reaching more than 1.1 TL (300 billion gal) by 2030 (Energy Information Administration, 2007a). Gasoline accounts for more than 60% of the transportation fuels, and its annual consumption is projected to increase to more than 650 GL (190 billion gal) by 2030 from the current level of 530 GL (140 billion gal). Adjusted for the lower calorific content of ethanol and assuming that this will constitute the bulk of the alternative liquid fuels, replacement of 20% gasoline alone would entail annual production of 225 GL (60 billion gal) of ethanol by 2030 (Energy Information Administration, 2007a). At a production efficiency of 370 to 400 mL kg–1 grain (2.5–2.7 gal bu–1), the entire current U.S. corn crop of 280 Tg (11 billion bu) can contribute no more than 113 GL (30 billion gal) of ethanol.
At the current annual genetic gain for corn grain yield of 1%, assuming that it is sustained over time, the yield is projected to reach 11.9 Mg ha–1 (190 bu acre–1) by 2030 from the current 9.4 Mg ha–1 (150 bu acre–1) (Duvick and Cassman, 1999). With an increase in area under corn production to 36.5 Mha–1 (90 million acres), which has already occurred in 2007, from the usual 30 to 32 Mha, the production could reach 430 Tg (17 billion bu) after accounting for future yield increases, still far short of producing the amount of ethanol needed to replace gasoline. In addition, the ethanol industry will have to compete with an increased demand of grain corn for feed and food, particularly from the developing countries with large and increasing populations as their standards of living improve. Cellulosic biomass derived from crop and forest residues offers an alternative for additional ethanol production (Perlack et al., 2005; Ragauskas et al., 2006; Somerville, 2006b, 2007; Wright et al., 2006).
After corn grain is harvested, the residue consisting of stalk, leaf, cob, and husk tissues, collectively referred to as stover, remains in the field, where it decomposes over time and contributes to soil organic matter. Corn accounts for a disproportionate amount of crop residue produced annually in the United States, which highlights its high productivity levels compared with other crops, including other cereals (Fig. 1A and B ) (Wright et al., 2006). Although the technology for making ethanol from cellulosic biomass is still evolving, once streamlined, corn stover alone has the potential to contribute substantially toward biofuel production. The advantage corn stover holds over alternative crops, such as switchgrass (Panicum virgatum L.) and miscanthus (Miscanthus x giganteus), is that it is already produced with grain as the target and does not require dedicated land. The main issue is how much stover can be collected on a sustainable basis without adversely affecting soil health.
|
|
Life-cycle analyses of the net energy content of ethanol currently produced from corn grain, which take into account the inputs and coproducts, is a subject of intense debate, with the recent analysis indicating that it is positive (Farrell et al., 2006; Patzek, 2004; Pimentel and Patzek, 2005). The disagreement seems to arise from the assumptions and sources of data used, as well as the inclusion or exclusion of value contained in the coproducts generated during the process. Regardless, grain-derived ethanol will contribute significantly toward fulfilling the goal envisioned by the Biomass Research and Development Technical Advisory Committee at least in the near future. Moreover, it has served the larger purpose of demonstrating that alternative, bio-based liquid fuels can be produced on a commercial scale, which will help drive the industry focused on cellulosic ethanol (Somerville, 2007).
This review focuses on the potential of corn as a biomass crop. Various aspects of biomass production, stover biomass composition and synthesis, and the current status and future prospects of genetic and biotechnological approaches to alter both grain and stover for improved ethanol production are discussed. A section is also included on the bioenergetic consequences of composition alteration for biomass production. Areas dealing with synthesis of cellulose and hemicellulose are emphasized slightly because these polysaccharides account for essentially all the sugars in the stover.
Biomass Production and Harvest Index
Grain yield in maize has nearly quadrupled over the last five to six decades (Fig. 3
) (USDA, 2007; Duvick, 2005a,b; Tollenaar and Lee, 2002). In small grain cereals, the introduction of dwarfing genes led to an improvement of harvest index (HI), the ratio of grain yield to total aboveground biomass (Hay, 1995; Sinclair, 1998). The resulting varieties with reduced stature could be grown under extensive inputs without lodging, which was a vexing problem with the older, taller varieties and limited their yield potential (Hay, 1995; Sinclair, 1998). Partitioning of greater amount of biomass to grain and increased total biomass production were the key factors that led to the green revolution. In contrast, HI in maize has remained essentially unchanged, around 50%, over the last century (Hay, 1995; Sinclair, 1998; Tollenaar and Wu, 1999). Grain yield improvement has thus resulted from increased total biomass production, which has been achieved mainly through selecting modern hybrids to be productive at increasingly higher planting densities (Duvick, 2005a,b; Tollenaar and Lee, 2002). Maintenance of HI around 50% as the grain yield increased several-fold implies that the sink/source ratio, unlike in small grain cereals, was already optimized in maize before the era of modern breeding.
|
Selection of modern hybrids for adaptation to high-density planting has been accompanied by resistance to environmental stresses such as drought, allowing them to perform relatively consistently in variable, unpredictable environments. Reductions in barrenness and late season stalk lodging further contributed to yield improvement (Duvick and Cassman, 1999). A number of other adaptive changes include more acute leaf angle to reduce shading, and thus allowing increased photosynthesis per unit land area, and reduction in tassel size to free up carbon for investment in other productive plant parts (Duvick and Cassman, 1999).
Grain composition has also changed over time (Duvick and Cassman, 1999). An increased grain starch/protein ratio meant channeling of more photosynthate toward starch formation and thus grain yield. Starch requires less energy for its formation than protein (see "Bioenergetics of Biomass Interconversion in Maize" section) (Penning de Vries et al., 1974; Sinclair and de Wit, 1975). When compared to total biomass productivity per unit land area, however, the contributions of reduced tassel size and starch/protein ratio to increased grain yield are minor.
The highest recorded grain yield in maize has stayed around 22 Mg ha–1 (350 bu acre–1) for several decades, suggesting that potential yield has not changed during this period (Duvick and Cassman, 1999). Environmental stresses are apparently responsible for most of the gap between the potential and actual grain yield. At an average annual gain of 1% and current average U.S. corn grain yield of 9.4 Mg ha–1 (150 bu acre–1), significant room exists for yield improvement assuming that genetic variation for tolerance to various stresses is not exhausted (Duvick, 2005a; Tollenaar and Lee, 2002).
Increased demand for grain from the ethanol industry will most likely result in a consistent increase in the land area planted to corn, which will both displace some other crops and bring additional area from the 14 Mha (35 million acres) currently held in Conservation Reserve Program (CRP) into cultivation (Wright et al., 2006). At the projected grain yield of 11.9 Mg ha–1 (190 bu acre–1) by 2030, and assuming 36 Mha (90 million acres) planted to corn, the same area as in 2007, an approximately 50% increase in corn supply can be expected, enough to produce 57 GL (15 billion gal) of ethanol if all the additional grain is used for this purpose. Accompanying grain yield will be increased stover biomass production, which will total more than 400 Tg yr–1.
Biodiesel, with an energy content of 91% of diesel, which is superior to that of either ethanol or butanol, lags far behind corn-derived ethanol as a biofuel because of its substantially lower yield per unit land area. For example, current soybean [Glycine max (L.) Merr.] yield of 2.7 Mg ha–1 (43 bu acre–1) in the United States would allow production of approximately 570 L ha–1(60 gal acre–1) of biodiesel (USDA, 2007). Even after adjusting for the energy content difference, it does not compare favorably with corn-derived ethanol at 3.7 kL ha–1 (400 gal acre–1) derived from 9.4 Mg ha–1 (150 bu acre–1) of grain.
Consequences of Stover Removal on Soil Organic Matter Content
Corn stover left in the field after grain harvest provides cover and contributes to the soil organic matter content. To facilitate its breakdown, the soil usually undergoes multiple tillings, which erode the organic matter content of the soil through volatilization and loss of top soil. Up to two-thirds of the stover may be removed on a sustainable basis under no-till conditions from some corn-growing regions without a remarkably adverse effect on soil organic matter content (Graham et al., 2007; Johnson et al., 2006; Perlack et al., 2005; Wilhelm et al., 2004). More than 100 Tg of stover could be collected annually within the tolerant limits of soil erosion at the current production levels (Graham et al., 2007).
Approximately 20% of total biomass at maturity stays in the form of roots in the soil, contributing to its organic matter content (Amos and Walters, 2006). Root mass alone, however, is not sufficient to maintain soil organic matter content over time (Wilts et al., 2004). Detailed discussion on this topic is beyond the scope of this review but can be found in several recent publications (Allmaras et al., 2004; Graham et al., 2007; Johnson et al., 2006; Perlack et al., 2005; Wilhelm et al., 2004; Wilts et al., 2004).
Stover removal will also entail net removal of nutrients that otherwise contribute to soil nutrition. For example, N harvest index, the ratio of grain N to total aboveground plant N, is 65% in maize (Banziger et al., 1999). Thus, an appropriate portion of one-third of the total plant N sequestered in stover would have to be replenished by an additional fertilizer application, depending on the amount of stover removed (Banziger et al., 1999).
Biomass Structure and Composition
Maize stalk accounts for more than half of the stover biomass, followed by leaves, cobs, and husk (Atchison and Hettenhaus, 2003; Masoero et al., 2006). Most of the stalk biomass is concentrated in the rind tissue, which consists of a mixture of densely packed vascular bundles embedded in a matrix of sclerenchymatous cells on the outer periphery of the internodes. Rind accounts for less than 20% of the cross-sectional area but more than 80% of stalk dry mass (Dhugga, unpublished results). Most of the remaining 20% of the biomass is likely accounted for by the vascular bundles that are embedded in the ground tissue consisting of parenchymatous cells (often mistakenly referred to as "pith"). Because parenchymatous cells have thin, primary walls and are nearly devoid of free sugars at maturity, their contribution to biomass is minor.
Ethanol production from grain or stover is a function of available carbohydrates. Because the stover carbohydrate content is quite similar to that of starch and sugar content of grain, which is 730 g kg–1 dry matter, potential ethanol yield from stover becomes a function of the fraction of stover collected and conversion efficiency of its carbohydrate fraction into ethanol compared with that of grain. The predicted results are displayed in Fig. 4
. An ethanol yield from grain of 3.7 kL ha–1 (400 gal acre–1) is assumed. At a relative efficiency of 80% of grain for stover carbohydrate conversion into ethanol, 1 kL ha–1 (105 gal acre–1) or 1.5 kL ha–1 (160 gal acre–1) of ethanol can be expected depending on whether one-third or one-half of the stover is collected for this purpose. Grain fiber fraction, which is ignored for the calculations above, will be the source for additional ethanol through the cellulosic biomass conversion route (Bals et al., 2006).
|
Cell Wall Biosynthesis
Nearly all the carbohydrates in stover occur in the form of cell walls. Primary cell wall, which is amenable to expansion under turgor pressure, is a viscoelastic composite in which crystalline cellulose microfibrils are embedded in a matrix of hemicellulose, pectin, and proteins (Carpita, 1996; Ray, 1987). Grass cell walls contain only trace amounts of pectin, the role of which is apparently fulfilled by the hemicellulosic matrix consisting mainly of GAX. Primary wall is synthesized during expansion growth (Carpita, 1996; Cosgrove, 2000).
Deposition of secondary cell wall, as the name indicates, begins inside the primary wall when the cell is ceasing to expand. Secondary wall is rich in cellulose but also contains lignin, GAX, and small amounts of proteins in most vegetative cells (Bacic et al., 1988; Iiyama et al., 1994). Cotton (Gossypium hirsutum L.) fiber is an exception in that it is made nearly entirely of cellulose.
Cellulose is made by a complex enzyme system that after being assembled in the Golgi, is integrated into the plasma membrane (Delmer, 1999; Saxena and Brown, 2005; Somerville, 2006a). Noncellulosic polysaccharides are synthesized in the Golgi, packaged into vesicles, and then exported to the cell wall through exocytosis (Ray et al., 1976). Unlike other wall constituents, lignin, being hydrophobic, is synthesized at the site of its deposition. Monolignols are transported to the apoplast where they are then polymerized by apoplastic peroxidases or laccases into the lignin network through peroxide-mediated radical coupling (Fry et al., 2000; Grabber, 2005). Coordination of many enzyme activities and transport processes across multiple subcellular compartments is required for cell wall synthesis. Synthesis of cellulose and hemicellulose is briefly reviewed in the following sections to highlight the complexity of the process of cell wall formation and its consequences in biotechnological alteration of wall composition.
Cellulose Synthesis
After the first plant cellulose synthase (CesA) gene was isolated by homology of its encoded product to a previously available bacterial CesA protein, rapid progress powered by high throughput genomics technologies ensued in the area of cell wall synthesis (Pear et al., 1996; Saxena et al., 1990; Somerville, 2006a). For example, complete gene families for CesA and other genes potentially involved in cell wall synthesis have been isolated from Arabidopsis, rice (Oryza sativa L.), and poplar (Populus) after the sequencing of their respective genomes.
In each higher plant species examined in detail, the CesA gene family is represented by multiple, usually 10 or more, members, the majority of which appear to be involved in primary wall formation as judged from mutational genetic studies and gene expression profiling (Appenzeller et al., 2004; Djerbi et al., 2005; Somerville, 2006a). In general, only three to four of these genes are believed to participate in secondary wall formation (Appenzeller et al., 2004; Djerbi et al., 2005; Somerville, 2006a; Tanaka et al., 2003; Taylor et al., 2000).
Duplication of the primary wall–forming CesA genes appears to have occurred relatively independently between dicots and monocots (Appenzeller et al., 2004). In comparison, the orthology of the secondary wall–forming genes is well conserved, suggesting that knowledge for these genes may be readily extrapolated from one species to another among vascular plants (Appenzeller et al., 2004). Another implication is that these two classes of proteins may have different biochemical properties and turnover numbers (see below).
Cellulose is synthesized on the cytosolic side and simultaneously extruded to the outside of the plasma membrane through the pores apparently made by the predicted transmembrane helices in the encoded products of the CesA genes (Somerville, 2006b). Cellulose synthase is a large, hexameric complex consisting of a variety of proteins (Delmer, 1999; Saxena and Brown, 2005; Somerville, 2006a). Each unit of the hexameric complex is believed to consist of six functional units of cellulose synthase, resulting in simultaneous synthesis of 36 glucan chains that after extrusion through the plasma membrane, crystallize to form a microfibril (Delmer, 1999; Somerville, 2006a). CesA proteins encoded by up to three different genes may assemble to form a functional complex (Taylor et al., 2003). Except for a few studies, where evidence was obtained either through mutational genetics or biochemical techniques, most of the observations in support of this hypothesis have come from gene expression profiling (Appenzeller et al., 2004; Brady et al., 2007; Burton et al., 2003; Ching et al., 2006; Doblin et al., 2002; Scheible et al., 2001; Tanaka et al., 2003). The three genes involved in secondary wall formation in maize, for example, are nearly identically expressed in developing vascular bundles where secondary wall is being rapidly deposited (Appenzeller et al., 2004; Brady et al., 2007; Ching et al., 2006).
In addition to CesA, several other proteins are known to affect cellulose synthesis: Kobito, a membrane-associated protein of unknown function; Korrigan, a membrane-anchored ß-glucanase; and Cobra, a protein anchored to the membrane though glycophosphatidylinositol (GPI) (reviewed in Somerville, 2006a). A member of the Cobra gene family, CobL4 from Arabidopsis, and its orthologs Brittle culm-1 (Bc1) from rice, and Brittle stalk-2 (Bk2) from maize, have recently been shown to specifically affect cellulose formation in secondary walls (Brown et al., 2005; Ching et al., 2006; Li et al., 2003). Two other enzymes or classes of enzymes are also known to affect cellulose formation: GDP-mannose pyrophosphorylase, which forms GDP-mannose and thus plays a role in GPI anchor formation of the Cobra-like proteins, glycosylation of other proteins, as well as glucomannan formation; and glycosidases, enzymes involved in glycosylation processing of proteins (Somerville, 2006a). Except for GDP-mannose pyrophosphorylase, a cytosolic enzyme, and protein-processing glycosidases that reside in endoplasmic reticulum and Golgi, it is possible that the remaining proteins, all of which are predicted to be membrane anchored, are directly associated with the cellulose synthase complex. Because purification of a functional cellulose synthase has not been achieved, it is not possible to determine the stoichiometry of various proteins in the complex and whether it contains any additional, unknown proteins. Regardless, the fact that each of these affects cellulose synthesis again points to the process's highly complex nature.
Further adding to complexity, it is not yet clear whether a single CesA protein is capable of catalyzing glucan formation or two polypeptides are required (Albersheim et al., 1997; Buckeridge et al., 1999; Dhugga, 2001; Han and Robyt, 1998; Robyt et al., 1974; Saxena et al., 1995). ß-1,4 linkage of the cellulosic glucan chains entails a 180° rotation of the adjacent glucosyl residues with respect to each other. Since there is no apparent internal duplication of the catalytic site, the simplest explanation is that two identical catalytic sites juxtaposed to each other in a dimer can overcome the energetic barrier of having to rotate the chain after the addition of each glycosyl residue (Albersheim et al., 1997; Buckeridge et al., 1999; Dhugga, 2001; Han and Robyt, 1998; Robyt et al., 1974; Saxena et al., 1995). If two catalytic sites are required to make a functional cellulose synthase unit, as appears likely to be the case, then 72 CesA polypeptides along with an unknown number of the aforementioned and as yet unknown proteins may be involved in the formation of each hexameric cellulose synthase complex.
Secondary wall–forming genes are in general expressed at significantly higher levels than the primary wall–forming genes, which could explain the rapid deposition of secondary wall on cessation of cell expansion, assuming that there is a correspondence between the levels of CesA proteins and their respective transcripts (Appenzeller et al., 2004). The primary wall must be able to irreversibly yield to turgor pressure for a cell to expand and thus for a plant to grow (Cosgrove, 2000; Ray, 1987). Lower rates of cellulose synthesis in the expanding cells may be synchronized in such a way as to maintain the viscoelastic properties of the generally thin primary walls that are essential for irreversible cell expansion under turgor pressure.
Variation for turnover number, the rate at which the catalytic site of a cellulose synthase adds glucosyl residues to the elongating chain, is not precisely known because of an inability to purify a functional cellulose synthase complex. This information has recently been derived from direct measurement of the rate of movement of a functional cellulose synthase complex that had been tagged with a fluorescent protein (Paredez et al., 2006). A similarly tagged cellulose synthase complex in the secondary wall–forming cells moved at a much faster rate than the one in the primary wall–forming cells (Turner, 2006). Provided that the measurements from these two separate studies are comparable, they would partly explain the rapid deposition of cellulose in the secondary wall.
Regulators of cellulose synthesis remain to be identified, although transcription factors that control transition from primary to secondary wall formation have been isolated (Demura and Fukuda, 2007; Zhong et al., 2006). In response to stem bending, cells under tension accumulate highly crystalline cellulose in a gelatinous layer whereas those under compression accumulate more lignin (Andersson-Gunneras et al., 2006; Donaldson et al., 1999; Joseleau et al., 2004; Schmitt et al., 2006). In agreement, secondary wall–forming CesA genes as well as Kor, a membrane-anchored ß-glucosidase implicated in cellulose formation, were induced by tension and repressed by compression (Bhandari et al., 2006; Wu et al., 2000). Application of tension upregulated sucrose synthase in poplar but downregulated the pathways for lignin and matrix polysaccharide formation (Andersson-Gunneras et al., 2006). Sucrose synthase has been implicated in feeding uridine diphosphate (UDP)-glucose directly to the cellulose synthase complex from sucrose and UDP, conserving the energy of the glycosidic bond during the process (Amor et al., 1995; Barreiro and Dhugga, 2007; Dhugga et al., 2006; Haigler et al., 2001). Regulatory elements that regulate different pathways up or down in response to mechanical stress could potentially provide tools for modulating cellulose formation.
Hemicellulose Synthesis
In addition to GAX, grass cell walls contain another hemicellulosic polysaccharide, mixed-linked glucan (MLG), which consists of ß-1,4-linked oligosaccharides of varying lengths coupled through ß-1,3-linkages (Carpita, 1996). Mixed-linked glucan is expressed in expanding cells and is recycled as expansion ceases (Carpita, 1996). It does, however, terminally accumulate in the seeds. Compared with barley (Hordeum vulgare) and oats (Avena sativa L.) where it constitutes a considerable portion of the seed endosperm walls, concentration of MLG in maize grain is too low for it to be a significant target for reduction (Genc et al., 2001).
A set of 30 to 50 genes found in different plant species with relatively weak homology to the CesA genes is referred to as CesA-like (Csl) (Hazen et al., 2002; Richmond and Somerville, 2000). These genes have been grouped into subclasses named CslA through CslH. Members of different Csl groups were hypothesized to make ß-linked hemicellulosic polysaccharides (Richmond and Somerville, 2000). Two groups of Csl genes, F and H, have evolved independently in grasses, which lack two other groups, B and G.
Genes for the backbone formation of two of the hemicellulosic polysaccharides, MLG and ß-(gluco)mannan, have been identified thus far (Burton et al., 2006; Dhugga et al., 2004; Liepman et al., 2005). Molecular components for xylan synthase, which makes the backbone of GAX, remain unknown.
Compared with the vastly complex cellulose synthase, hemicellulose-forming enzymes identified thus far appear to be simpler in structure (Burton et al., 2006; Dhugga, 2005; Dhugga et al., 2004). In both the successful cases, a single gene product was capable of making the backbone in a heterologous system. Glucomannan synthase was functional even in insect cells, suggesting that it can synthesize the product by itself (Liepman et al., 2005). Expression of a CslF gene in Arabidopsis, native walls of which do not accumulate MLG, resulted in the deposition of this polysaccharide (Burton et al., 2006). It is difficult to rule out at this stage whether MLG synthase is a multiprotein complex. For example, CslH has also been implicated in MLG formation through a gene silencing approach (Nick Carpita, personal communication, 2007). Assuming that CslH does indeed participate in MLG formation, its phylogenetically closest relative from Arabidopsis, CslB, may possibly have complemented CslF to make a functional enzyme (Burton et al., 2006; Hazen et al., 2002).
A glycosyltransferase originally annotated as being involved in 
-glycan formation has recently been suggested using a bioinformatics approach to make ß-xylan (Mitchell et al., 2007; Pena et al., 2007). Sugar moiety in the substrates for cell wall polysaccharides and starch formation is linked to nucleoside diphosphate (NDP) in an configuration. Enzymes that catalyze ß linkages (e.g., in cellulose and xylan) are referred to as "nonretaining" or "inverting," and they are believed to mediate the transfer of the sugar residue to the acceptor molecule without forming a covalent intermediate (Sinnott, 1987). Conversely, the enzymes for linkage formation (e.g., in starch) retain the original configuration through the formation of a covalent intermediate before transferring the sugar to the acceptor molecule. Motifs conserved among polymerizing ß-glycosyltransferases are absent from the enzyme that has recently been proposed to make ß-xylan (Mitchell et al., 2007; Saxena et al., 1995). Lack of conserved motifs in itself is not a reason to question the proposed function of the enzyme (Mitchell et al., 2007). With experimental evidence for its reaction lacking and the function of so many of the Csl genes remaining to be determined, however, this area of research should still be considered open.
In addition to arabinosyl and glucuronosyl residues that decorate the xylan backbone of GAX, acetyl groups are esterified to the xylosyl residues at C2, C3, or both these positions (Fig. 2) (Bacic et al., 1988). Maize stover contains 3 to 5% acetate and approximately 20% xylose on mass basis (McAloon et al., 2000; Wooley et al., 1999). Adjusted for molarity, the acetate/xylose ratio ranges from one-third to one-half in GAX (Fig. 2). Because of their weak ester linkages, acetyl groups are readily removed by the alkaline stover pretreatment. The residual groups may still impede xylan hydrolysis by xylanase as a close relationship between xylose yield and deacetylation was observed in corn cobs (Vazquez et al., 2001). Acetate is well known to be an inhibitor of fermentation so is an obvious target for reduction (Gray et al., 2006; Lawford and Rousseau, 2002).
Molecular components for acetylation of GAX remain to be identified. Apparently, acetylation occurs in the Golgi compartment and the transferase that acetylates GAX may also use acetyl-CoA as a substrate. An acetyl transferase that acetylates rhamnogalacturonan has been assayed in cultured potato (Solanum tuberosum L.) cells but the corresponding gene remains to be isolated (Pauly and Scheller, 2000).
Lignin and Cell Wall Cross-Linking
Lignin has long been a target for reduction because of its known adverse effect on rumen digestibility as well as on ethanol production from stover biomass (Grabber, 2005; Sticklen, 2006). The cross-links that lignin forms with other wall polymers are believed to increase the recalcitrance of vegetative tissues to hydrolytic enzymes (Jung and Casler, 2006a; Jung, 2003).
Cross-links can occur in primary as well as secondary cell walls (Iiyama et al., 1994). Activated forms of hydroxycinnamic acids ferulate and p-coumarate, feruloyl-CoA and p-coumaroyl-CoA, respectively, are intermediates in the monolignol biosynthetic pathway and act as substrates for feruloylation or coumaroylation of GAX (Anterola and Lewis, 2002; Campbell and Sederoff, 1996; Humphreys and Chapple, 2002). These monolignols are linked to the arabinosyl residues of GAX through an ester linkage in the Golgi compartment by a transferase that remains to be identified (Fig. 3) (Fry et al., 2000; Iiyama et al., 1994). These feruloyl moieties can form linkages with other reactive groups in the cell wall through a peroxidase-mediated reaction that results in GAX–GAX and GAX–lignin linkages (Iiyama et al., 1994). Some cross-linking may also occur in the Golgi itself (Fry et al., 2000).
Another type of cross-linking is hypothesized to occur among wall polysaccharides through a class of enzymes that have the ability to cut and paste glycans (Fry, 2004). A transglycosylase/hydrolase has recently been reported to mediate the formation of linkages among diverse cell wall polysaccharides, potentially augmenting the cell wall network (Hrmova et al., 2007). Occurrence of many of these interpolymer linkages in the cell wall remains to be demonstrated, however. Feruloyl transferase and transglycosylase/hydrolase are attractive targets for potentially altering the degree of cell wall cross-linking for improved digestibility. A candidate gene for feruloyl transferase from rice was recently proposed using a bioinformatics approach (Mitchell et al., 2007). The protein encoded by this gene is devoid of a transmembrane domain, however. Golgi transferases identified thus far are Type II membrane proteins with a transmembrane domain near the N terminus (Faik et al., 2002; Perrin et al., 1999; Edwards et al., 1999).
The contribution of cross-links to mechanical strength is not clear. Caution must be used in simplifying the problem of cell wall degradability to individual steps in cell wall formation as compensation in the form of altered cross-linking or composition often makes it difficult to establish a cause-and-effect relationship (Grabber, 2005).
The goal of higher ethanol productivity overlaps with that of stalk standability, an important agronomic trait (Appenzeller et al., 2004). Vascular bundles and sclerenchymatous cells have lignified, cellulose-rich walls. Lignin is generally believed to contribute to mechanical strength; however, it appears to play a role in determining resistance to compression rather than tensile strength (Ching et al., 2006). To maximize mechanical resistance to breakage, for example, cellulose microfibrils in the stem tissue subjected to mechanical bending are oriented along the axis of tension (Andersson-Gunneras et al., 2006; Donaldson et al., 1999; Green, 1962; Joseleau et al., 2004; Schmitt et al., 2006). Similarly, cellulose microfibrils are oriented perpendicular to the axis of growth in an expanding cell, again to maximize radial strength so as to allow longitudinal expansion (Green, 1962). Lignin synthesis is downregulated in the cells subjected to tension but is upregulated with regard to compression (Andersson-Gunneras et al., 2006; Donaldson et al., 1999; Joseleau et al., 2004; Schmitt et al., 2006).
Being hydrophobic, lignin most likely increases mechanical strength of cellulose indirectly by keeping it relatively free of water. Because of the plasticizing effect of water, moist cellulose is mechanically weaker than dry cellulose (Tolstogurov, 2000). In agreement, the amount of cellulose in a unit length of the stalk nearly completely explained mechanical strength of the maize internodes below the ear (Appenzeller et al., 2004). In an independent study involving a mutation in the maize bk2 gene that specifically and dramatically reduced cellulose accumulation in the secondary walls, lignin and hemicellulose amount were unaltered but mechanical strength was dramatically reduced, further suggesting that cellulose is the primary determinant of mechanical strength and that any contribution of lignin is through alteration of the environment around cellulose in the cell wall (Ching et al., 2006). Selection for reduced lignin and increased cellulose in stover can potentially be expected to increase mechanical strength as well as ethanol yield (Appenzeller et al., 2004; Ching et al., 2006).
Bioenergetics of Biomass Interconversion in Maize
Success in improving ethanol production depends directly on the available carbohydrate content. Grain is the richest source of available carbohydrates in the maize plant (Earle et al., 1946). Two of the avenues to increase carbohydrate content further come at the expense of protein and oil. Another option is to eliminate the ear and make stover-only corn. This seems an uneconomical venture for the following reasons: (i) grain, because of its high density, is easier to store and transport than stover; (ii) ethanol from grain starch is produced close to the theoretically predicted efficiency (Patzek, 2006); (iii) grain ethanol is already a mature industry, whereas cellulosic ethanol is still several years away from being commercially profitable; and (iv) conversion of grain to stover will be accompanied by lignin, which will more than offset the expected biomass gain from converting protein and oil into carbohydrates (see below).
One potential option to increase biomass per unit land area is to grow maize varieties adapted to tropical regions in temperate zones. However, even if growing corn varieties out of their respective zones of adaptation results in increased vegetative biomass per unit land area compared with the adapted grain corn varieties because of their indeterminate growth habit, they will not compete favorably with grain corn because of the latter being a superior feedstock.
Bioenergetic cost associated with constructing a tissue of a particular composition from photosynthate further helps us understand the superior value of grain as a feedstock for ethanol production (Fig. 5 ). Penning de Vries et al. (1974) estimated from the underlying biochemical pathways that 1 g of photosynthate could be used to make 830 mg of storage or structural carbohydrates, 460 mg of lignin, 400 mg of protein, or 330 mg of lipid. Approximately half as much energy is utilized for carbohydrate synthesis as it takes to make protein from the photosynthate (Bhatia and Rabson, 1976; Penning de Vries et al., 1974; Sinclair and de Wit, 1975). These estimates differ from the calorific contents of carbohydrates and proteins, which are 16.8 and 23.9 kJ g–1, respectively. The reason for the disparity was the use of nitrate, which requires extra energy for its reduction, instead of ammonium as the starting source of nitrogen. Production value for protein matches its calorific content if ammonium is used as the initial substrate. Because maize uses nitrate as the primary source of N, the lower of the two bioenergetic estimates for protein formation from photosynthate is used for subsequent discussion.
|
An often-discussed strategy to increase biomass production is to convert corn into a stover-only crop. Energy-rich lignin in stover would negate any enhancement in biomass that is expected from the elimination of oil and protein in the form of grain. For example, bioproductivity of a typical maize grain is 720 mg g–1 photosynthate (Fig. 5) (Earle et al., 1946). Similarly, for maize stover of a typical composition, bioproductivity would be 704 mg g–1 of photosynthate (Fig. 5). Accounting for a slightly prolonged growing season resulting from an apparently expected delay in senescence because of lack of a strong N sink in the form of grain, the result would be about the same amount of biomass produced per unit land area with or without the grain. Without a concomitant reduction in lignin, which has been difficult to accomplish because of undesirable pleiotropic effects, stover-only corn does not present an attractive alternative. In addition, these theoretical conversions, which help define upper limits for productivity of biomass of a particular composition, do not always result in expected results and may require substantial breeding effort in selecting desirable recombinants (Dhugga and Waines, 1989). A reduction in lignin, for example, is expected to be accompanied by an increased bioproductivity but an opposite outcome is observed because of the associated pleiotropic effects (Fig. 5) (Pedersen et al., 2005).
A suitable scenario perhaps is to increase the percentage of starch in the grain. Because corn is primarily a starch crop, the content of starch in the grain is already high, >70% (Belyea et al., 2004; Earle et al., 1946). Additional starch can be accumulated at the expense of embryo size, grain protein in the endosperm, and/or grain fiber. Reductions in endosperm protein as well as embryo size below a certain threshold may cause viability problems, and a reduction in fiber content may affect grain strength, causing increased breakage. Most of the fiber in the grain is in the pericarp, which is rich in GAX (Hazen et al., 2003). Incrementally replacing GAX with cellulose, which is mechanically stronger but also more crystalline, may allow maintenance of grain strength yet allocate more carbon to additional starch formation (Dhugga et al., 2000; Hazen et al., 2003).
Biotechnological Opportunities, Outcomes, and Future Prospects
Rapid advances in genomic sciences have provided an abundance of molecular tools that can potentially be used to alter plant structure and function. Despite considerable progress in the area of cell wall synthesis over the last decade, some of the problems remain unsolved. For example, molecular components of GAX formation as well as its acetylation and upstream regulators that affect cellulose or lignin synthesis in response to tension or compression remain to be identified. A discussion on prospects and attempts to alter plant biomass and composition follows. A recent review covers other, complementary aspects of transgenic manipulation to improve maize for ethanol production (Torney et al., 2007).
Transgenic expression of single genes has been reported to increase biomass in several plant species (Biemelt et al., 2004; Lefebvre et al., 2005; Levy et al., 2002; Smidansky et al., 2007, 2002). These types of results are often obtained with experimental lines. Any of these types of discoveries have yet to lead to a commercial product, in part perhaps because of the inherent, enormous flexibility of the complex plant metabolism to maintain homeostasis despite transgenic perturbations (Carrari et al., 2003). It is possible that the experimental line being used in a study like this is indeed deficient in the particular function that the transgene enhances under the specific experimental conditions used. The positive effect of a transgene on biomass observed in an experimental line disappeared when it was backcrossed into agronomically elite genetic backgrounds (Meyer et al., 2007). This suggests that the modern breeding programs have indirectly selected, and may continue to do so, for optimal performance of metabolic pathways under the conditions that give highest productivity per unit land area. Environmental variation often makes it difficult to distinguish elite varieties with competitive yields under field conditions. Biomass at maturity represents the sum of the developmental profile over the entire growing season. Plants are exposed to variable environments at different developmental stages across locations and years, which means extensive field testing must be performed to be able to draw meaningful conclusions regarding the effect of a transgene on biomass performance.
A cellulose-binding dye, Calcofluor White, was shown to enhance the rate of cellulose synthesis in a bacterium apparently by slowing the rate of crystallization of the nascent glucan chains (Haigler et al., 1980). Levy et al. (2002) pursued this idea further by overexpressing a cellulose-binding domain (CBD) from a microbial cellulose hydrolase in poplar, which they claimed enhanced growth. Microbial and fungal cellulases possess a CBD and a hydrolytic domain. The CBD enhances hydrolysis of cellulose by acting as an anchor for the hydrolytic domain and also perhaps by depolymerizing the cellulose crystal into glucan chains. Separated from the hydrolytic domain, CBD still retains the ability to bind crystalline cellulose. The explanation for enhanced growth rate in poplar overexpressing CBD was that by binding to the nascent glucan chains coming out of the cellulose synthase complex, CBD slowed the rate of crystallization and thus enhanced the rate of polymerization (Levy et al., 2002). The same group claimed that overexpression of an endo-ß-1,4-glucanase accelerated growth in tobacco (Nicotiana) and Arabidopsis. Implicit in these studies is the assumption that carbon supply (source) is nonlimiting and only cellulose polymerization or wall properties limit growth. Follow-up, long-term studies are not yet available, so it is not possible to determine if the reported increase in cellulose synthesis or accelerated growth indeed resulted in increased biomass per unit land area per unit time.
Grain sink in maize could potentially be made limiting by space planting the hybrids adapted to high planting density. Reduced shading allows plants to increase photosynthate production that may exceed the amount needed to fill the available grains. Progressive farmers deliberately increase planting density to the extent that the kernels at the ear tip do not fill, a phenotype referred to as "nose-backing," which is indicative of sink not being limiting. This apparently ensures that the maximum amount of available photosynthate is captured for grain production. Biomass of space-planted individual plants or trees is thus not an indicator of potential biomass production per unit land area.
Complexity of the cellulose synthase system poses a challenge to increasing its activity through transgenic manipulation. For an overexpressed component to have an effect, a parallel increase in the expression of all other components of the complex must be assumed (Barreiro and Dhugga, 2007; Sere, 1987). Alternatively, the component being overexpressed must indeed be limiting in the assembly of the cellulose synthase complex. Regardless of these apprehensions, preliminary experiments suggest that it may be possible to affect cellulose synthesis with the overexpression of individual CesA genes (Dhugga et al., 2005).
Association of rate of cellulose synthesis or tissue cellulose concentration with available haplotypes, and specifically with allelic variants of different genes known to be involved in cellulose formation, offers perhaps more promise in accomplishing the objective of increasing cellulose in plant tissues. The associated haplotypes could then be used to select for increased cellulose production in breeding populations. Crystalline cellulose can be determined by a simple chemical method, which is amenable to scaling up for high throughput analysis (Updegraff, 1969). Direct measurement of cellulose would allow the mapping of this trait using recombinant lines derived from divergent parents, which might also reveal as-yet-unknown effectors of cellulose formation.
Upregulation of cellulose synthesis may eventually lead to altered composition of stover by competing with GAX and lignin in secondary walls, which account for most of the biomass. For this approach to affect biomass production, vegetative sink, which may be approximated as an aggregate of nonphotosynthetic, nongrain plant parts, must be assumed to be limiting in the current varieties. The HI in maize is held nearly constant even when grain yield varies widely under varying environmental conditions, suggesting that at least source/grain-sink ratio is already optimized (Sinclair et al., 1990).
Disruption of the enzymes that potentially add side chains onto the xylan backbone of GAX causes a severe disruption in secondary wall synthesis (Zhong et al., 2005). A number of fragile fiber (fra) and irregular xylem (irx) mutants are caused by defects in genes coding for the enzymes potentially involved in the addition of side groups onto the xylan backbone (Burk and Ye, 2002; Persson et al., 2007; Zhong et al., 2004, 2005). Complete knockouts caused severe phenotypes, however, and it remains to be seen whether partial downregulation of these genes can quantitatively reduce GAX content without adversely affecting plant growth (Burk and Ye, 2002; Persson et al., 2007; Zhong et al., 2004, 2005).
Now that a candidate gene for feruloylation of GAX has been identified, it can be downregulated by transgenic means to determine whether it reduces the degree of cross-linking in the wall and, if so, whether and to what extent it is biologically tolerable (Grabber, 2005; Mitchell et al., 2007). Once the molecular components for acetylation of GAX are identified, they could similarly be manipulated to potentially reduce the acetyl content of the dry matter.
Lignin has been downregulated by the introgression of spontaneous mutations in the lignin biosynthetic pathway, referred to as brown midrib (bm) for the color change that is most obvious in the midrib, in elite germplasm, and by transgenically silencing a number of genes from the monolignol biosynthetic pathway (Anterola and Lewis, 2002; Burk and Ye, 2002; Zhong et al., 2004, 2005). Reduction in lignin is generally accompanied by a concomitant reduction in biomass (Pedersen et al., 2005). In addition, plants with lowered lignin are generally more susceptible to insects and diseases, which likely results from a weakening of the physical barrier lignin poses in accessing the wall polysaccharides that the pathogens need for growth (Pedersen et al., 2005). Variable expression of the bm mutations in different genetic backgrounds suggests that it may be possible to reduce lignin to some extent without a remarkable effect on total biomass (Grabber et al., 2004; Pedersen et al., 2005; Pichon et al., 2006).
Stover that is easily digested by ruminants is also expected to be readily hydrolyzed by cellulases into simple sugars for fermentation. For lignin downregulation to be an effective avenue for improving stover digestibility, any reduction in biomass must be compensated by an increased harvestable yield of total fermentable sugars so that the potential ethanol yield per unit land area remains unaltered.
Although pretreatment and enzyme hydrolysis constitute two of the more costly steps in cellulosic ethanol production, stover with reduced lignin may still need to be treated before being subjected to enzyme hydrolysis. It seems unlikely that the cost savings in pretreatment from reduced lignin can be fully realized because of an accompanying reduction in biomass.
Lignin-rich residue remaining after biomass hydrolysis in the current cellulosic ethanol protocols is projected to be burnt to generate electricity and heat (Wright et al., 2006). Various technologies to generate high-value products from this residue are being explored, which, if commercially successful, may help boost the value of stover (Hahn-Hagerdal et al., 2006).
An increase in cellulose concentration in lignin downregulated quaking aspen (Populus tremuloides Michx.) was reported to be accompanied by increased growth rate (Hu et al., 1999). This conclusion has been questioned for various reasons, especially (i) a lack of relationship between the degree of lignin reduction and growth rates of various transgenic events and (ii) non-normalization of the proportions of different wall constituents to compensate for reduced lignin, a major constituent itself (Anterola and Lewis, 2002).
Hydrolytic enzymes have been successfully expressed in plants with the objective of reducing the cost of their production (Dai et al., 2005; Sticklen, 2006; Ziegler et al., 2000). The idea of these enzymes eliminating or reducing the addition of exogenous enzymes cocktails is unlikely to be valid under the currently used protocols to process biomass for ethanol production because these enzymes are unlikely to withstand the harshness of the pretreatment steps (Ragauskas et al., 2006). Whether production of these enzymes in plants proves to be more economical than the microbially produced enzymes remains to be seen.
Exogenous addition of an expansin protein at a level of 10 mg g–1 (or, in practical terms, 10 kg Mg–1) of isolated maize cell walls caused significantly greater swelling than the control walls (Yennawar et al., 2006). Apparently, this occurs because of facilitated access of the wall polysaccharides to water. Whether increased swelling improves digestibility of the cell walls by the hydrolytic enzymes is not yet known. Also, the cell walls were prepared from the silk tissue, which hardly has any lignin. It would be interesting to see if the expansin proteins can be accumulated in planta to such high levels as used in this study without affecting plant biomass. Secondary walls may be the preferred sites for these proteins because their deposition begins as the cell expansion is coming to a stop.
As is clear from the aforementioned discussion, genetic engineering of polysaccharide biosynthetic pathways for cell wall formation is still in its infancy. Complete understanding of these pathways will greatly expand the tool kit needed for biotechnological manipulation of the cell wall. Enough critical information is already available for exploring the use of a combination of genetic and transgenic methods to alter stover and grain composition for improved stover digestibility and increased ethanol yield.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONCONCLUSIONSREFERENCES |
|---|
At least for the next 5 yr and perhaps longer, nearly all of the increased ethanol demand in the United States will be met by grain corn with stover as a feedstock for the future. Depending on the market realities, a significant portion of the 14 Mha (35 million acres) of the agricultural land currently held in the CRP may be brought under maize cultivation to alleviate the demand for corn grain in the short term and whole biomass in the long term.
Direct selection will continue to contribute to increased grain yield and biomass production and perhaps to increased cellulose content of the stover biomass. There is no dearth of reports claiming positive effect of reduced lignin on stover digestibility. To be commercially viable, any such improvement must be reflected in increased efficiency of ethanol production per unit land area, not simply per unit dry mass. Biotechnology offers considerable scope to alter the chemical composition of stover through alteration of GAX amount and composition and perhaps lignin cross-linking. The degree to which these alterations can be tolerated by the plant without adverse pleiotropic effects will determine the extent of success of the biotechnological approach.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONCONCLUSIONSREFERENCES |
|---|
Received for publication May 28, 2007.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONCONCLUSIONSREFERENCES |
|---|
