Although evaluations of switchgrass as feedstock for the bioenergy industry have indicated significant benefits from an energy perspective, it also may adversely impact runoff, sediment and nutrient losses, and associated water quality effects. Field scale monitoring requires long-term measurements that are expensive and cannot be used practically for large-scale monitoring of many watersheds or regions (Harmel et al. 2006). Instead, crop models are gaining favor as they provide a practical alternative for assessing bioenergy cropping effects, due to their versatility and cost-effectiveness when compared to field experimentation.

Soil and Water Assessment Tool (SWAT) has been used at the plot and watershed scale to determine N loss, P loss, and soil erosion from switchgrass fields. Sarker (2009) used SWAT to compare N loss from agricultural systems growing switchgrass and cotton in the southeastern U. S. The plot scale modeling suggested that in the early years of growth there is a significant N loss from switchgrass to streamflow and groundwater, but N loss is significantly reduced as switchgrass matures. Nepple et al. (2002) conducted a similar SWAT watershed scale study in which 50,000 ac of cropland were converted to switchgrass production in the Rathbun Lake watershed in southern IA. Overall, the model indicated that conversion of 15.3% of the watershed area to switchgrass production would significantly reduce soil erosion, N, P, and atrazine loadings into the Rathbun Lake watershed, relative to a baseline of traditional row-crop agriculture (Table 5).

Chamberlin et al. (2011) applied the DAYCENT biogeochemistry model to calculate the nitrate in runoff water when converting land cover from cotton and unmanaged grasslands to a switchgrass system in the southern U. S. Long-term simulations predicted a reduction of nitrate runoff (up to 95%) for conversions from cotton to switchgrass with N application rates of 0-135 kg N ha1. A reduction of nitrate runoff ranging from 50-70% occurs at all levels of fertilization, when converting from unmanaged grasses. The simulated nitrate runoff values from DAYCENT fall within the observed range.

The impact of large-scale biofuel production on water quality is a growing concern. Wu et al. (2012) recently applied the SWAT model to simulate the impact of future biofuel production on water quality and water

cycle dynamics in the Upper Mississippi River basin. Converting pasture lands to switchgrass reduced soil erosion considerably and positively impacts N and P loadings at the projected yield and fertilizer input. In addition, switchgrass increases the water loss associated with evapotranspiration (1% of total precipitation), decreases the base flow (2%), and decreases the surface runoff flow to the basin. Nelson et al. (2006) used the SWAT model to predict reductions in four water quality indicators (sediment yield, surface runoff, nitrate nitrogen (NO3-N) in surface runoff, and edge-of-field erosion) associated with producing switchgrass on cropland in the DE basin in northeast KS. They determined that the production of switchgrass on conventional agricultural cropland had distinct environmental advantages versus traditional (e. g., corn-soybean) cropping rotations.

Field studies that support the findings that switchgrass production will improve surface water quality are slowly becoming available. Lee et al. (1998) compared switchgrass filter strips to cool-season grass filter strips and reported that switchgrass was more effective in removing P and N from runoff. Similarly, Sanderson et al. (2001) noted reductions in P and N runoff from a switchgrass filter strip treated with dairy manure, while Mersie et al. (1999) utilized switchgrass filter strips to reduce the amount of dissolved atrazine and metachlor herbicides by 52% and 59%, respectively. Entry and Watrud (1998) tested the ability of Alamo switchgrass to remediate soil contaminated by the radionuclides cesium-137 and strontium-90. Switchgrass captured 36% of the cesium and 44% of strontium over a five — month period. Overall, all models used predict that converting lands used for pastures or traditional row crop to perennial switchgrass for feedstock production will have a positive long-term environmental impact by reducing sediment loss and nutrient runoff, and improving water quality.

Conditioning of syngas can be accomplished through hot or cold gas cleaning technologies. The cost of syngas conditioning can be a significant portion of the cost of syngas production, hence reducing cost of syngas conditioning is critical. Hot gas cleaning technologies use high temperature and catalysts to crack tar molecules into gaseous compounds, which can also improve the energy content and composition of the syngas. Cold gas technologies use solvents such as water, acetone, isopropanol, and oil to scrub tar and other contaminants from the gas.

Enzyme productivity increases with increasing biomass concentration. The biomass concentration of T. reesei is increased by supplementing soluble sugars such as glucose or xylose along with insoluble lignocellulosic material. Mohagheghi et al. (1990) reported that the mixture of xylose and cellulose (30:30 g/l) was effective in reducing the lag phase of cellulolytic enzyme production to 6.25 days and resulted in 122 IFPU/l-h productivity. In addition, the same enzyme productivity was observed with pure cellulose, but over 8 days. At the same time, the enzyme titer was reduced by 27% for a mixture of xylose and cellulose compared to pure cellulose (Mohagheghi et al. 1990).

The modeling exercise is predicated on the following assumptions: (1) An economically competitive technology for converting switchgrass biomass to some type of biofuel (if not cellulosic ethanol, perhaps a drop-in fuel) will be forthcoming; (2) A biorefinery will require a flow of feedstock throughout the year; (3) Switchgrass is the exclusive feedstock; (4) In the U. S. Southern Plains, the switchgrass harvest window extends from July through March; (5) Expected switchgrass biomass yield and fertilizer requirements differ by harvest month; (6) Land use services could be acquired in long term leases; (7) The biorefinery will require material to be delivered in standardized 1.22 m x 1.22 m x 2.44 m bales with no more than 15 percent moisture; (8) Harvest crews may be centrally managed; (9) The biorefinery is assumed to operate 350 days per year and require 3,630 Mg/day of switchgrass biomass. The model is based on an extension of models previously formulated by Tembo et al. (2003), Mapemba et al. (2007), Hwang (2007), Mapemba et al. (2008), Haque (2010), and Haque and Epplin (2012). It is a multi-region, multi-period, monthly time-step, mixed integer mathematical programming model and can be used to determine the cost to deliver a flow of biomass to a biorefinery.

The model was formulated to include all 77 Oklahoma counties as potential production regions and two land classes, cropland and improved pasture land. The model limits switchgrass production to no more than 10 percent of a county’s cropland and no more than 10 percent of a county’s improved pasture land based on data from the Census of Agriculture (USDA 2010). The assumption is made that cropland could be acquired for a long-term lease rate above average U. S. CRP rental rates (Data. gov 2010). The lease rate for cropland for each county was calculated by adding $49/ ha to the average CRP rental rate for that county as determined by Fewell et al. (2011). Long-term lease rates for improved pasture land are derived by adding $76/ha to the 2010 average county pasture rental rate (USDA 2010; Fewell et al. 2011). The modeled rental rates are designed to cover the opportunity costs of alternative production options and to account for increased land-lease rates that may occur in response to an entrant in the market for 10 percent of the county’s land, and to compensate for the lost option value from engaging in long term leases (Song et al. 2011).

Switchgrass biomass yield estimates for each production region were obtained from estimates produced by Oak Ridge National Laboratory (Jager et al. 2010). Yields for cropland and improved pasture land are not differentiated (U. S. Department of Energy 2011). Hwang (2007) used weather data to determine probability distributions for the number of suitable field work days per month for harvesting switchgrass for each Oklahoma county. The 95 percent probability level from the harvest day distributions is selected so that the number of harvest days per month is set to be equal to the number of days that would be suitable for harvest in 19 of 20 years. For most months, the number of mowing days exceeds the number of safe baling days.

The model simultaneously determines how many hectares from which county and which land class are optimal to harvest for each month; how much harvested biomass should be put in field storage each month; how much should be shipped to the biorefinery each month; how much should be put in biorefinery storage each month; and how much should be processed each month. The model accounts for differences in nitrogen and phosphorus fertilizer requirements depending on the month of harvest. An integer variable is included to determine the optimal number of mowing units (windrowers), and another integer variable is included to determine the optimal number of harvest units (rakes, balers, tractors, and stackers) (Table 8, 9). Thus, the model endogenously determines the number of harvest machines. Shipment and processing of biomass can be done in any of the 12 discrete periods (months of the year). In months when biomass is harvested, it may be placed in storage or transported directly from the field to the biorefinery. The harvest season extends from July through March of the following year.

The modified Wang (2009) transportation cost equation used for the budget reported in Table 5 is also used in the modeling exercise. Transportation costs depend on the distance the feedstock will be shipped from the fields to the biorefinery. The distance between any biomass supplying county and any plant location is estimated by the distance from the county’s central point to the plant location. Storage losses at the biorefinery and in the field are assumed to be one percent per month (Hwang 2007). Another assumption is that bales stored in the field would be covered with a plastic tarp. The cost of field storage is estimated to be $1.79/Mg regardless of the number of months the material is in storage (Hwang 2007).

The diversity of switchgrass’ morphology parallels the diversity of sites to which the species is adapted (Fig. 1). Robust lowlands can be taller than 3.0 m, while some uplands may only reach 0.5 m; root depths can extend to 3 m where soils are not restrictive (Porter 1966; Moser and Vogel 1995). Upland plants also typically have thicker roots and longer root internodes, with rhizomes long enough to support sod formation (Beaty

Color image of this figure appears in the color plate section at the end of the book.

et al. 1978). In contrast, lowland plants, with shorter rhizomes, more often exhibit the characteristic growth habit of a bunchgrass (Vogel 2000). New shoots originate from rhizome buds on lowland plants but also from basal nodes of culms in the upland ecotypes (Porter 1966). In addition to being smaller in stature, upland ecotypes typically have finer leaves and stems and smaller panicles; lowlands, in addition to their larger size, also often have a distinctive bluish coloration (Casler 2005). Both ecotypes generally have erectophile leaves, although this characteristic generally is stronger in lowlands.

Although seed size can differ by cultivar, this generally will have little agronomic consequence except in cases of excess seeding depth (Zhang and Maun 1991; Smart and Moser 1999). Seedling morphology, however, has some potential to affect switchgrass development and survival, as some seedlings have elevated crowns, which limits adventitious root formation. This appears to have little actual effect, however, as research in the field gives little evidence that this is an issue in stand establishment (Elbersen et al. 1999). Researchers have also selected for increased tillering of seedlings, which may be useful in the development of divergent genotypes, but this has not been a successful approach for improving plant establishment or yield (Smart et al. 2003, 2004).

Switchgrass is a determinate plant and it produces multiple tillers that become reproductive after exposure to the right environmental signals. Biomass accumulation comes to an end in conjunction with inflorescence development. Research suggests that daylength is the chief signal for floral development, although this response may not be completely under photoperiod control (Esbroeck et al. 2003). Reproductive development may also be delayed or inhibited by flooding, excessive K fertilization, and low temperatures (Porter 1966; Balasko and Smith 1971; Friedrich et al. 1977).

For a given cultivar, the critical photoperiod (some minimum night length, actually) is genetically determined and linked to the plant’s latitude of origin. Differences in photoperiodic flowering responses among cultivars have important implications for selection and production in the field. For example, moving southern-adapted cultivars to higher latitudes delays their reproductive development. This promotes continued vegetative growth (and greater biomass yield) as the plant does not experience the appropriate signal for reproductive development (sufficient night length) until later in the year. In the opposite way, a northern type will be less productive when moved south because the photoperiodic trigger occurs earlier in the season at lower latitudes. This attribute also has important implications for survival, and we will give both issues further consideration in the section below on cultivar selection.

Along with latitudinal differences among cultivars, switchgrass can display variable longitudinal morphology and adaptation (Hopkins et al.

1995a, b; Madakadze et al. 1998; Vogel 2000; Casler and Boe 2003). Cultivars adapted to conditions of the humid east are generally taller but less tolerant of the drier and windier conditions of the Great Plains (Cornelius and Johnston 1941). Conversely, productivity of the shorter, coarser-stemmed western-adapted switchgrass can be negatively affected when moved east. This is thought to be largely related to the lower pathogen resistance in cultivars adapted to drier climates (Vogel 2000).

Following rhizoplane colonization, internalization of the bacteria and their development as endophytes can occur quite rapidly, within days of inoculation/rhizoplane contact (Compant et al. 2008; Prieto and Mercado — Blanco 2008; Zakria et al. 2008). In order to colonize the plant interior, bacteria must make their way past the root surface. This can happen through the presence of surface openings, such as cracks produced during lateral root emergence (James and Olivares 1998), or other wounds. Furthermore, other root areas, such as the elongation and differentiation regions may contain cells that are more fragile or less differentiated, and more susceptible to bacterial penetration (Reinhold-Hurek and Hurek

2011) . As with rhizosphere and rhizoplane colonization competence, a variety of bacterial traits are associated with competence for endophytic colonization. These include flagella, nod genes, type IV pili and twitching motility (Compant et al. 2008). Many of these traits are associated with bacterial adherence and movement, or bio-control of other surrounding microorganisms, providing a competitive advantage for the colonizing bacteria. In addition, bacterially-secreted, cell wall-degrading enzymes are important for bacterial penetration (Quadt-Hallmann et al. 1997) and internal colonization, including cellulolytic and pectinolytic enzymes (Quadt-Hallmann et al. 1997; Kovtunovych et al. 1999). The endophyte Burkholderia phytofirmans strain PsJN, known to colonize switchgrass (Kim et al. 2012), produces endogluconase and polygalacturonase (Compant et al. 2005b), to aid in cell wall degradation.

Following initial root penetration, bacterial colonization proceeds within the root cortex, and can extend into the central vascular cylinder xylem vessels (Compant et al. 2008; Priedo and Mercado-Blanco 2008; Zakria et al. 2008). However, not all bacterial endophytes colonize the xylem. For example, Priedo and Mercado-Blanco (2008) noted that Pseudomonas fluorescens PICF7 remained in the root cortex region and was never found in the xylem, with no subsequent translocation elsewhere. The inability of some endophytes to colonize the xylem and move past the root may be due to the presence of filters formed at branch root junctions (Shane et al. 2000), which may limit bacterial movement (Zakria et al. 2008). In addition, as endophytes are aided in their penetration through the root endodermis and pericycle by cell wall-degrading enzymes (James et al. 2002), it may be that some endophytes produce sub-optimal levels of enzymes to allow penetration into the vascular tissue.

Colonization within regions like the root cortex occurs within the intercellular spaces, outside of living cells (Reinhold-Hurek and Hurek 1998; Priedo and Mercado-Blanco 2008), which is not surprising as these are rich in minerals (potassium, calcium, sulfur, phosphorus, chlorine), sugars (Madore and Webb 1981) and non-carbohydrate metabolites, such as various amino acids and organic acids (Canny and McCully 1988; Canny and Huang 1993). Endophyte alterations of apoplastic pH can alter enzyme activities, sugar uptake of host cells, and sugar concentrations for the colonizing microbes (Bacon and Hinton 2002). Hence, this environment is supportive of endophyte growth, promoting compound cycling between the endophyte and the plant.

As mentioned above, cellulose composes about 40% of switchgrass cell wall and is the primary target for bioconversion to biofuel. Cellulose microfibrils are synthesized at the plasma membrane by cellulose synthase A (CESA) complexes with glycosyltransferase (GT) activity (Lei et al. 2012). Microfibril orientation determines cell growth direction and cell wall mechanical properties (Saxena et al. 2005; Crowell et al. 2010). In higher plants, the complex consists of a hexagonal formation of six rosettes; each rosette consists of six CESA polypeptides (Doblin et al. 2002). Many reviews have covered recent progress in cellulose synthase machinery and cellulose

Table 2. Enzymes that synthesize and modify xylan. Protein Name Locus ID Mutant CAZy* or DUF+ Family Activity Comments Reference IRX9 At2g37090 irx9 GT43 p(1^4)-xylan synthesis Backbone elongation (Brown et al. 2007; Pena et al. 2007) IRX14 At4g36890 irxl4 GT43 Backbone elongation (Brown et al. 2007) TaGT43-4 wheat GT43 (Zeng et al. 2010) IRX10/GUT1; IRX10-LIKE/GUT2; TaGT47-13 Atlg27440 At5g61840 irxlO irxlO irxlO-L GT47 p(1^4)-xylan synthesis Backbone elongation (Brown et al. 2009) (Zeng et al. 2010) IRX7/FRA8; IRX7L/ F8H At2g28110 At5g22940 irx7/fra8 GT47 Reducing end (Rennie et al. 2012) IRX8/GAUT12 At5g54690 irx8 GT8 p(1^4)-xylan synthesis Reducing end (Pena et al. 2007) PARVUS/GLZ1 Atlgl9300 parvus GT8 Reducing end (Lee et al. 2007) GUX1 GUX2 At3gl8660 At4g33330 GT8 Side-chain glucuronic acid and 4-O-methylglucuronic acid branches to xylan (Mortimer et al. 2010) IRX15 IRX15L At3g50220 At5g67210 DUF579 unknown Mutant has methylglucuronic acid side chains instead of glucuronic acid side chains (Brown et al. 2011) XAX1 Os02g22380 GT61 p-Xylp-(l—»2)-a-Ara/- (1-3) (Chiniquy et al. 2012) TaXATl wheat GT61 a(1^3)-Ara/Transferase (Anders et al. 2012)

microfibril synthesis (Somerville 2006; Carpita 2011; Endler et al. 2011; Domon et al. 2012; Lei et al. 2012). As revealed by genomic sequencing, Arabidopsis, rice, and sorghum each possesses 10 CesA genes; whereas, maize possesses 20 CesA genes (http://cellwall. genomics. purdue. edu/) (Penning et al. 2009). The Arabidopsis CESA proteins as well as CESAs that have been studied in grasses are listed in Table 1. Recent results highlight the utility of understanding cellulose synthesis for enhancing biofuel production. DeBolt and colleagues reported that mutations in the C-terminal transmembrane domain region of Arabidopsis CESA1 and CESA3 proteins decrease microfibril crystallinity and increase saccharification efficiency (Harris et al. 2012).

Mutant studies, mostly in Arabidopsis, have greatly enhanced our understanding of physiological functions of CESA family members. Arabidopsis AtCESA1, AtCESA3 and AtCESA6 are mainly responsible for cellulose synthesis in primary walls. Mutations in the genes that synthesize these proteins present dwarfism, sterility, swollen etiolated hypocotyls, ectopic accumulation of lignin and reduced root elongation phenotypes (Endler et al. 2011). On the other hand, mutations in AtCesA4, 7 and 8 genes affect cellulose synthesis in secondary walls and are accompanied by collapsed xylem vessels and significant decreases in cellulose contents (Endler et al. 2011). Researchers have hypothesized that the longer and more crystalline microfibrils of secondary walls might be due to different CESA components in the cellulose synthase complexes. However, recent studies have detected expression of the "primary wall" cellulose synthase genes, AtCesA1, 3 and 6, in both primary and secondary cell walls of Arabidopsis (Betancur et al. 2010). Furthermore, AtCESA2,5 and 9, which have greatest homology to CESA6, have been found to be involved in the biosynthesis of secondary cell wall in the epidermal seed coat (Stork et al. 2010; Mendu et al. 2011; Sullivan et al. 2011). These findings suggest that much more remains to be learned about the mechanism of the cellulose synthase complex.

Though phylogenetic analyses are common, direct studies of cellulose synthesis in grasses are less so. Still, mutations in the rice secondary wall cellulose synthases, OsCesA7, OsCesA4 and OsCesA9, respectively, lead to a brittle culm phenotype caused by thinner cortical fiber cell walls (Tanaka et al. 2003). Also, prior to the availability of the maize genome and a more complete description of the maize CesA family, Appenzeller and colleagues isolated 12 CesA genes from maize. Among them, ZmCesA10, 11, 12 are the probable orthologs of the Arabidopsis secondary cell wall CesA genes (Appenzeller et al. 2004). As expected, numerous expressed sequence tags for switchgrass cellulose syntheses appear in databases (Tobias et al. 2008), though a detailed naming scheme and apportionment into families has not yet been conducted, and may be premature at the time of this writing due to the draft nature of the genome.

Integrating genetic maps and physical genome maps is extremely valuable for map based isolation, comparative genome analysis and as anchors for genome sequencing projects. The DOE (Department of Energy) Joint Genome Institute (JGI) has sequenced plant genomes of candidate bioenergy crops such as sorghum (Paterson et al. 2009) and the model grass Brachypodium (International Brachypodium Initiative 2010). Both plants have been used as references for switchgrass, however sorghum last shared a common ancestor with switchgrass more than 20 million years ago while Brachypodium last shared a common ancestor with switchgrass more than 50 million years ago. The genome of a much closer switchgrass relative—foxtail millet (Setaria italica)—is described recently (Bennetzen et al. 2012; Zhang et al. 2012). All three genomes, along with those of switchgrass sequenced by the JGI are publicly accessible at www. phytozome. net. The genus Panicum also includes diploid species closely related to switchgrass such as P. hallii. Meyer et al. (2012) sequenced, assembled and annotated the transcriptome of a lowland ecotype P. Hallii var. Filipes, using 454-FLX Titanium sequencer. Totally 1.26 million EST reads were produced, and 15,422 unique genes identified. A majority of contigs (77%) and 34% of singletons matched ESTs from switchgrass (Meyer et al. 2012). In addition, a switchgrass bacterial artificial chromosome (BAC) library containing 147,456 clones and covering the effective genome approximately 10 times are established, and the average length of inserts in BACs is 120 kb (Saski et al. 2011). Moreover, two BAC libraries with 16 times genome equivalents of switchgrass were constructed, and each library comprised 101,376 clones with average insert sizes of 144 (HindIII-generated) and 110 kb (BstYI-generated) (Sharma et al. 2012).

In parallel, molecular linkage maps composed of various molecular markers including RFLP, SSR and STS (Missaoui et al. 2005b; Okada et al. 2010, Liu et al. 2012) have been constructed in switchgrass. For a more complete genome assembly and analysis, and to take full advantage of linkage map resources, it is of great importance to combine switchgrass genetic maps with physical map information. This can be accomplished by connecting genetic mapping data to BAC clones, which is a well established approach in rice (Chen et al. 2002), wheat (Sourdille et al. 2004), sorghum (Klein et al. 2000), and ryegrass (Lolium perenne, King et al. 2002), but has not been employed in switchgrass on a genome-wide basis.

Identification of Genome Structure

Tetraploid (2n=4x=36 chromosomes) and octoploid (2n=8x=72) are two major ploidal forms in switchgrass, while diploid (2n=2x=18), hexaploid (2n=6x=54), decaploid (2n=10x=90), and 12-ploid (2n=12x=108), and aneuploids were also reported (Lu et al. 1998; Costich et al. 2010, Young et al. 2010; Zalapa et al. 2011). Most of lowland switchgrass is tetraploid and octoploid lowland switchgrass has been discovered recently (Zhang et al. 2011b). Upland ecotype encompasses primarily octoploid and tetraploid plants although the former is more frequent than the latter. Using single dose molecular markers, one study inferred tetraploid switchgrass is an autotetraploid with a high degree of preferential pairing (Missaoui et al. 2005b), and another study indicated switchgrass has a near complete preferential pairing in disomic inheritance (Okada et al. 2010). The mode of completely disomic inheritance was further revealed by using about 500 co-dominant markers, which distributed on whole genome (Liu et al.

2012) . Recently Triplett et al. (2012) sequenced five low copy nuclear loci and two chloroplast loci and clarified that switchgrass was an allopolyploid originated from hybridization of two close diploids. The "A" subgenome of switchgrass is likely derived from an ancient species which was a close

relative to the current species in section Rudgeana of Panicum genus, while "B" subgenome is not identified yet (Triplett et al. 2012). The subgenomic differentiation in switchgrass was confirmed by karyotype analysis of a diploid individual (Young et al. 2012), which were selected from the progeny of a cross between two lowland tetraploid cultivars, ‘Kanlow’ and ‘Alamo’ (Young et al. 2010).

Switchgrass is one of the important biofuel crops which would contribute to our renewable energy in the future. Research on switchgrass miRNA is still in its infancy and to date, our knowledge has been largely limited to those that are conserved across species. Little is known about many switchgrass-specific miRNAs and their functions. Identification, isolation and functional characterization of switchgrass miRNAs will require more efforts. However, data obtained from many other plant species have clearly demonstrated the importance of these small RNA molecules and their targets in regulating various aspects of plant growth, development and response to environmental stimuli. This points out their great potential for use in plant genetic engineering. It should be noted that although miRNAs could serve as potential tools for genetic manipulation of switchgrass for improvement, altering expression of a miRNA, in many cases, could cause pleiotropic morphological and developmental changes in transgenic plants. Therefore, it is critical to better understand molecular mechanisms underlying miRNA-mediated changes in plant growth and development thereby designing appropriate transgenic strategies to obtain desirable traits with minimum unfavorable side effects.

Acknowledgements

The research in Luo’s lab has been supported by Biotechnology Risk Assessment Grant Program competitive grant no. 2007-33522-18489 and no. 2010-33522-21656 from the USDA National Institute of Food and Agriculture as well as the USDA grant CSREES SC-1700315 and SC-1700450. Technical Contribution No. 6110 of the Clemson University Experiment Station.

Tissue Culture, Genetic
Transformation, and
Improvement of Switchgrass
Through Genetic Engineering

Bingyu Zhao, u* Rongda Qu,2 Ruyu Li,2 Bin Xu1 and
Taylor Frazier1

Tissue Culture

In the early 1990s, the US DOE identified switchgrass (Panicum Virgatum L.) as an herbaceous energy crop and launched research efforts on switchgrass as a biomass energy feedstock (McLaughlin and Kszos 2005). Conger’s laboratory at the University of Tennessee was the pioneer in tissue culture and genetic transformation research on switchgrass. Their first report on callus induction and plantlet regeneration was published in 1994 (Denchev and Conger 1994). Mature caryopses, along with young leaf segments from newly-formed shoots of secondary tillers (lowland cv. Alamo), were used as explants and cultured on MS medium supplemented with auxin,

2,4- D (22.5 pM, or 5 mg/l), and cytokinin, 6-benzylaminopurine (BAP, 45 pM, or 10 mg/l, for mature caryopses and 5 pM, or 1.1 mg/l, for young leaf segments). Mature caryopses cultures were maintained in the dark at 29°C for 4 wk, and callus and "organized structures" were observed. They

department of Horticulture, Virginia Tech, Blacksburg, VA 24061. department of Crop Science, North Carolina State University, Raleigh, NC 27695. ^Corresponding author: bzhao07@vt. edu

were transferred to MS medium without growth regulators and placed under light. Approximately 65 percent of the calluses regenerated into plants. For young leaf segment cultures exposed to the same conditions, embryogenic calluses originated from basal segments of innermost leaf pieces. Non-embryogenic calluses were produced from the remainder of the leaf segments. Histological and scanning electron microscopy (SEM) analyses indicated embryogenesis was the main pathway for regeneration from mature caryopses culture, whereas regeneration from leaf segment cultures was mostly through organogenesis. One thousand regenerated plants were obtained and grown in the field.

In their next publication (Denchev and Conger 1995), the authors evaluated the effects of four concentrations (0, 5, 15, and 45 pM) of BAP in combinations with three concentrations (11.3, 22.5, and 45 pM) of auxins,

2,4- D or picloram, on callus induction and shoot regeneration. With mature caryopses as explants, both embryogenic and non-embryogenic calluses were observed from all 2,4-D-containing media. It was also observed that a supplement of BAP greatly improved the formation of embryogenic calluses and regeneration. In contrast, few embryogenic calluses were observed from picloram-containing media. In both cases, two transfers of calluses to regeneration medium (free of growth regulators) greatly facilitated shoot regeneration. The best regeneration results came from 2,4-D (all three concentrations) in combination with 15 or 45 pM BAP in callus induction media. Young seedling segments were also used as explants but performed very poorly.

Switchgrass is an outcrossing species and thus each plant is an individual genotype. Maintenance of a desirable genotype has to be through vegetative propagation. Correspondingly, Conger’s laboratory developed a node culture procedure for efficient micropropagation of switchgrass (Alexandrova et al. 1996). Tillers of switchgrass (cv. Alamo) with four to six nodes were harvested and individual nodes (below the top node) were excised. The nodes were sliced longitudinally and placed with the cut surface in contact with MS medium supplemented with 30 g/l maltose and four concentrations of BAP (0, 5, 12.5, and 25 pM) as the only growth regulator. A week after culturing nodes under a 16 hr light/8 hr dark photoperiod, shoots began to emerge from the axillary buds at the nodes. Roots were developed 8 wk later when the shoots were transferred to hormone-free medium. The highest shoot numbers came from media that contained 5 or 12.5 pM BAP (six to seven shoots per node) and cultured at 29°C. The number of shoots developed from nodes cultured at 29°C was six-fold higher than those cultured at 22°C, most likely owing to the fact that switchgrass is a warm-season grass.

In 1998, the same laboratory reported multiple shoot clump formations when mature caryopses were cultured in media with a combination of various concentrations of 2,4-D and thidiazuron (TDZ) (Gupta and Conger 1998). Caryopses germinated in the medium and multiple shoot meristems were formed at the shoot apex (Fig. 1). The phenomenon was only observed when both 2,4-D and TDZ were present in the medium. The optimal combination of the growth regulators for producing the highest number of shoots was determined to be 4.5 pM 2,4-D and 18.2 pM TDZ. Both lowland (cv. Alamo) and upland cultivars (cv. Trailblazer, Blackwell) had similar responses, though at different frequencies. SEM and histological analyses revealed that the multiple shoot formation was caused by activation of axillary buds and de novo formation of adventitious buds. The shoot clusters were successfully transferred to soil after rooting in hormone-free medium.

In 1999, Conger’s laboratory reported the development of suspension cultures of switchgrass in MS liquid medium containing 2,4-D (9 pM) and BAP (4.4 pM). Initiated from embryogenic calluses from young inflorescence cultures, the suspensions contained various developmental stages of somatic embryos, which regenerated or germinated into plantlets after being transferred to solid medium (Gupta and Conger 1999). In a related report, Odjakova and Conger (1999) studied the effect of callus age and osmotic pretreatment on embryogenic cell formation and the regeneration ability of the suspension cultures (cv. Alamo). It was observed that 10-day-old calluses performed better than 20- and 30-day-old calluses, and that 0.3 M each of sorbitol and mannitol was superior over 0.1 and 0.2 M each.

Based on experience and previously established methods in switchgrass tissue culture, Conger’s laboratory went on to successfully obtain transgenic switchgrass plants (see below). In a report from Somleva et al. (2002), the authors identified several genotypes from cv. Alamo and induced embryogenic calluses from various explants of these genotypes for transformation. Details on how to identify such genotypes and how to maintain them were not described.

A similar approach was recently reported by Xu et al. (2011). These authors first identified certain lines, such as HR8, whose mature caryopses- derived calluses had high regeneration ability in tissue culture experiments. HR8 plants were cross-pollinated with another high regeneration line, HR7, for seed production. Seeds collected from HR8 have a higher germination rate than the unselected Alamo seeds (82 vs. 72 percent). Eighty-five percent of the calluses induced from germinated HR8 seeds were embryogenic, and 84 percent of those regenerated into plants, whereas the corresponding rates of unselected Alamo seeds were only 36 and 21 percent, respectively. The response of HR8 to ABA supplementation to the medium was also different from the unselected Alamo seeds. The addition of ABA to callus induction medium reduced seed germination and increased embryogenic callus formation from HR8 seeds. For example, at 10 pM of ABA, the seed germination rate was reduced to 51 percent but embryogenic callus formation increased to 99 percent for HR8. However, unselected seeds, placed at the same concentration of ABA, exhibited a decline in both germination rate and embryogenic callus formation. Further analysis revealed that the endogenous ABA levels in HR8 seeds were about 3-fold higher than the ABA concentration in unselected Alamo seeds. Additionally, fungal endophytes were observed from some switchgrass callus culture, which may negatively affect callus growth and regeneration.

Burris et al. (2009) employed a new culture medium, LP9, in switchgrass tissue culture, which was modified from culture medium described by Lu et al. (2006). The medium by Lu et al. contains macroelements of N6 medium (Chu 1975), and microelements and vitamins of B5 medium (Gamborg et al. 1968). It is also supplemented with casein hydrolysate, proline, and glutamine. In the LP9 medium by Burris et al. (2009), dicamba was replaced by 2,4-D (5 mg/l), and both BAP and myo-inositol were removed. Proline was also reduced from 0.5 g/l to 0.1 g/l. Inflorescences from tillers of E2 to E4 stages (Moore et al. 1991) were first cultured on MS+BAP medium for 10 days and then transferred to the LP9 medium. Approximately one third of the calluses induced were "brittle and white", similar to the Type II callus described in maize culture (Armstrong and Green 1985). The regeneration ability of the friable, embryogenic, type II callus lines could last more than six months. Agrobacterium-mediated transgenic plants were obtained from the type II calluses.

Calluses similar to the reported type II callus were also observed by Li and Qu (2011) in lowland cultivars of Alamo, Performer, and Colony. However, different medium ingredients were used. Mature caryopses were first cultured on MS-based medium containing 5 mg/l 2,4-D and 1 mg/l BAP for 6-8 wk. Approximately 15 percent of the induced calluses were white, compact, embryogenic calluses. They were transferred to the same medium supplemented with 2 g/l L-proline. Approximately 50 percent of the subcultured calluses became white and friable for cv. Alamo and Colony. Although both types of calluses were highly regenerable, the friable calluses were more competent for transformation. However, this kind of callus tends to regenerate albino plants after a long term culture (100 percent albino after 14 mon of culture). Another kind of callus, one that is yellow and friable, was observed mainly from cv. Performer (Fig. 2). This kind of callus was also highly regenerable, but maintained green plant regeneration ability much longer. In an experiment, 100 percent of such calluses still regenerated into green plants after 14 mon of culture without any albinos. Supplementation of proline to the culture medium not only promoted type II callus formation, but also enhanced callus growth.

Color image of this figure appears in the color plate section at the end of the book.

Data from fertility studies generally suggest that little added P is needed to achieve high switchgrass yields in bioenergy cropping systems (Hall et al.1982; Muir et al. 2001). This may be different in forage production settings, as in the case of a study by (Rehm 1990), who tested P amendments from 0 to 90 kg ha-1 in Nebraska over 4 years. Rehm (1990) reported a curvilinear response to P with production gains at rates up to 45 kg ha1. Others have reported increased establishment-year production with added P, although effects of P were not observed in subsequent seasons (McKenna and Wolf 1990).

Several studies have reported little to no response to P (Brejda 2000), despite low soil P status. In studies with once-per-year harvest, no response to P was observed after several (3 or 7) years of biomass removal (Muir et al. 2001). Switchgrass grown on low-P soils in Iowa did not respond to P (Hall et al. 1982) and in the southern Great Plains, switchgrass response to P was only benefited at one of two sites (Kering et al. 2012a) over three seasons. In the Kering et al. (2012a) study, P applications of 45 kg ha-1 yr-1 increased yields on a low-P (3.4 mg P kg-1) soil, but no response to any P rate (0, 15, 30, or 45 kg P ha-1) was observed at a second location with soil P concentration of 3.1 mg P kg-1.

Switchgrass’ relationship with the soil microbial community may be a common denominator in the oft-observed variable and limited responses to P and N. In the case of P uptake, switchgrass’ role as host to vesicular arbuscular mycorrhizae greatly improves the grass’ ability to extract and uptake P. Several studies have shown that these root colonizing fungi can greatly improve P acquisition in conditions of high soil acidity, high aluminum and low P (Koslowsky and Boerner 1989; Boerner 1992a, b; Clark et al. 1999; Clark 2002). Adding these fungi to sterilized, low-P soils can eliminate a response to P (Brejda 2000). Conversely, eradicating mycorrhize in low-P soils can reduce switchgrass production if fertilizer P is not added to the system (Bentivenga and Hetrick 1991).

Without returning nutrients to the system, repeated harvests will reduce soil P concentrations in switchgrass biomass production systems (Schmer et al. 2011). With modest yields (5.8 Mg ha1) of switchgrass harvested at anthesis, annual losses of 1.5 kg1 y-1 P ha were reported in production fields in the Great Plains (Schmer et al. 2011). Although greater losses would be expected with greater biomass yields, this factor must be weighed against the stage of plant development at the time of harvest, as the effects of higher yield would be offset by lower P concentrations with plant maturity and senescence (Parrish and Fike 2005; Lemus and Parrish 2009).

There is little research to suggest that switchgrass is particularly responsive to K, whether in field or greenhouse studies (Friedrich et al. 1977; Smith and Greenfield 1979; Hall et al. 1982). Typical recommendations are to maintain K at a medium level based on typical soil test ranges (Teel et al. 2003; George et al. 2008; Douglas et al. 2009). This apparent lack of response may in part be a function of K being recycled to the soil through leaf leaching when switchgrass is harvested after senescence (Parrish and Fike 2005).

As with the other nutrients, response to limestone applications can be variable. This may be less a function of pH change than of the availability (or lack of availability) of other mineral nutrients or toxins (Parrish and Fike 2005). Switchgrass strains display differences in terms of tolerance to soil acidity, with some lines being productive—as opposed to merely tolerant—at pH 4.9 (Bona and Belesky 1992). These differences also may play a role in the variable yield responses reported. Screening for such traits may prove useful if truly marginal sites such as reclaimed mine sites are to be utilized for a future bioenergy industry.