Endophytic bacteria that live freely in the internal tissues of plants and cause no apparent harm have a diverse range of growth promotion mechanisms including nitrogen fixation. Although 78% of the earth’s atmosphere is nitrogen, nitrogen is often a limiting factor in agriculture since it is not readily available to plants. Bacteria and Archea are the only organisms that can fix atmospheric di-nitrogen, thereby making it available for plant growth. This activity is termed biological nitrogen fixation (BNF) and is catalyzed by the oxygen sensitive nitrogenase enzyme to convert N2 to bio-available NH3. Nitrogenases are complex metalloenzymes with highly conserved structural and mechanistic features (reviewed in Alberty 1994; Burgess and Lowe 1996; Rees and Howard 2000). The enzyme is oxygen sensitive, which imposes physiological constraints on the organism. Additionally, the enzyme has a relatively slow turnover time (Thorneley and Lowe 1985), which requires the microbe to synthesize large quantities of the protein, up to twenty percent of protein in the cell (reviewed in Dixon and Khan 2004). Also, the conversion of atmospheric di-nitrogen to a form that can be used by plants requires 16 ATP to reduce one molecule of N2, making it one of the most energy demanding reactions identified in bacterial organisms (Thorneley and Lowe 1985). Together, the amount of energy, the low oxygen requirement, and the amount of protein required to create the nitrogenase enzyme, place a large burden on a nitrogen fixing endophyte. As a result, the synthesis of the nitrogenase complex is stringently regulated at the genetic level (Dixon and Khan 2004). It has been suggested that bacterial endophytes are placed in a more favorable environment compared to rhizospheric bacteria because they are less vulnerable to competition from native soil bacteria and are shielded from various biotic and abiotic stresses (Reinhold-Hurek and Hurek 1998). Perhaps the most-studied grass inoculated with free living nitrogen-fixing endophytes is sugarcane. Burkholderia MG43 inoculated sugarcane plantlets produced a 20% increase in yield over un-inoculated control (Govindarajan et al. 2006), and it was demonstrated that 60 to 80% of nitrogen accumulated in sugarcane came from atmospheric nitrogen fixation (Boddey et al. 1995). The authors also noted that farmers in Brazil have observed some varieties of sugarcane grown in fields for decades, even up to a century without showing any decline in soil N reserve or yield, despite the supply deficit of nitrogen (Boddey et al. 1995). Rice has also been studied in the context of its relationship with free-living nitrogen-fixing Burkholderia spp. In one field experiment, 31% of plant nitrogen was derived from BNF and inoculation resulted in as high as a 69% increase in biomass compared to the un-inoculated control (Baldani et al. 2000). Researchers also found Burkholderia vietnamiensis inoculated rice seedlings increased yield by 5.6 to 12.16%, and 42% of nitrogen found in the inoculated plants came from atmospheric nitrogen fixation (Govindarajan et al. 2008). In addition to rice, Burkholderia were found to be among the most common nitrogen-fixing isolates from maize plants cultivated in Mexico, and many were reported to be new species (Estrada et al. 2002). These findings support the use of free-living nitrogen-fixing endophytes in the effort to reduce the use of synthetic nitrogen fertilizer and offer hope in creating high-yielding, low — input agricultural production systems.

Lignin is a complex macromolecule; its cross-linked structure renders it the most recalcitrant substance for biochemical conversion to biofuels (Vanholme et al. 2008; Ralph 2010; Carpita 2012; Vanholme et al. 2012). As with polysaccharides, lignin cleavage requires synergistic action of diverse ligninolytic enzymes, including high redox potential ligninolytic peroxidases, laccases and oxidases (Leonowicz et al. 2001; ten Have et al.

2001) .

Peroxidases, which cleave C-C and C-O bonds, are classified into heme-dependent lignin peroxidases (LiPs), manganese peroxidases (MnPs), and versatile peroxidases (VPs). Phanerochaete chrysosporium and Trametes versicolor represent efficient lignin-degrading white rot fungi; both produce LiPs and MnPs (Gold et al. 1993; Johansson et al. 1996). LiPs can directly oxidize a variety of phenolic and nonphenolic aromatic compounds via long-range electron transfer. MnPs cleave phenolic substrates depending on oxidation of Mn2+ to Mn3+ by H2O2 (Harvey et al. 1992; Wariishi et al. 1992). A LiP-like MnP enzyme, called a variable peroxidase is found in Pleurotus enryngii and Bjerkandera spp. and possesses a hybrid molecular architecture that combines different oxidation sites connected to a heme cofactor (Moreira et al. 2005; Ruiz-Duenas et al. 2007). Evolutionary analysis of peroxidases in 31 fungal genomes revealed lignin-degrading peroxidases are possessed by diverse white rot, brown rot, and mycorrhizal species in the Agaricomycetes (Floudas et al. 2012). Interestingly, molecular clock analyses places the timing of the evolution of lignin degrading peroxidases at the end of Paleozoic Era, which coincides with the end of accumulation of the vast terrestrial carbon deposits that created coal.

Laccases, which also cleave C-C and C-O bonds, are widespread, four-copper containing metalloenzymes, able to catalyze the oxidation of a variety of phenolic and lower-redox potential compounds in the presence of redox mediators (Leonowicz et al. 2001). Wood-rot fungi are the main producers of laccases, especially fungi in the class of Basidiomycetes, even though some bacteria and plants also excrete these enzymes (Leonowicz et al. 2001; Sirim et al. 2011). The laccase from Pycnoporus cinnabarinus is essential for that fungi’s lignin depolymerization ability (Eggert et al. 1997). Two laccase isozymes of T. versicolor were able to depolymerize hardwood pulp lignin in the presence of 2, 2′-azinobis (Bourbonnais et al. 1995).

Apart from peroxidases and laccases, several oxidases involved in H2O2 production and aldehyde-alcohol transformation, are indispensable in lignin decomposition (Leonowicz et al. 2001). A glucose 1-oxidase mutant of Phanerochaete chrysosporium exhibited little or no ability to degrade lignin (Ramasamy et al. 1985). Aryl alcohol oxidases (AAOs) and aryl alcohol dehydrogenase (AAD) are responsible for aromatic aldehyde-alcohol transformation during ligninolysis (Leonowicz et al. 2001). Bacterial enzymes, such as ring-fission enzymes, demethylases and p-etherases, specifically degrade the oligolignols with low molecular mass that are liberated by diverse peroxidases and/or laccase (Masai et al. 1999).

Besides fungi, a number of bacteria are able to breakdown lignin (Bugg et al. 2011). Streptomyces viridosporus T7A secretes a lignin peroxidase to depolymerize lignin. Pseudomonas putida mt-2 and Rhodococcus jostii RHA, possess comparable lignin-degrading activities. Soil bacteria such as Nocardia and Rhodococcus also mineralize lignin.

Among these lignin-degrading enzymes, laccases have been developed for larger-scale applications, such as removal of polyphenols in wine and beverages, conversion of toxic compounds and textile dyes in wastewaters, and bleaching and removal of lignin from wood and non-wood fibres (Rodriguez Couto et al. 2006). Laccases have also been used to reduce phenolic inhibitors that form during biomass pretreatment and inhibit biological fermenters (Alvira et al. 2012). Organisms that produce ligninolytic enzymes can effectively function as pretreatment agents, as much as doubling subsequent saccharification yields (Akin et al. 1995). However, these organisms tend to grow relatively slowly, with treatments taking place over days or even weeks. It appears that ligninolytic enzymes remain a relatively poorly tapped resource for facilitating biofuel production.

The extranuclear DNA containing organelles (e. g., chloroplast and mitochondria) of any particular plant species are extremely useful tools in assessing genetic variation, resolving phylogenies (Timothy et al. 1979; Gielly and Taberlet 1994; Aizawa et al. 2007) and serving as vectors for transgene expression (Nakahira and Shiina 2005; Lu et al. 2006; Remacle et al. 2006; Farre et al. 2007; Hanson et al. 2012). In particular with plant species that have tetraploid genomes and higher (many of the grasses), using nuclear markers to compare across ploidy levels are difficult in these populations because gene copy number and allele frequencies are affected under polysomic inheritance (Young et al. 2011). As an alternative in these complex grassland ecosystems, chloroplast genome sequences can possibly provide a greater understanding of the evolutionary processes that have taken place during establishment from a comparative approach of isolated subpopulations (Young et al. 2011). As noted, switchgrass accessions primarily consist of lowland and upland ecotypes (Porter 1966) where lowland accessions are predominantly tetraploids (2n=4x=36), while upland accessions are octaploids (2n=8x=72) (Bouton 2007) making population genetic comparisons difficult because of a variance of ploidy levels affecting orthologous loci. In most angiosperms, the chloroplast genome consists of a quadripartite structure that includes a large single copy region (LSC) and a small single copy region (SSC) flanked by two inverted repeats, and maintains a pattern of maternal inheritance (Soltis et al. 1990; Faure et al. 1994; Vivek et al. 1999) making it an ideal genetic marker for phylogenetic studies (Chung et al. 2003; Liu et al. 2012). Previous studies aimed at discriminating ecotypes was focused on an RFLP marker in the rbcL gene (Hultquist et al. 1996), but the lack of comprehensive data points represents a need for more robust analysis. At present, two switchgrass chloroplast genomes from individuals representative of the lowland (Kanlow) and upland (Summer) ecotypes (Young et al. 2011) have been sequenced and compared to identify a 21 bp insertion in the Summer ecotype at the C-terminal region of rpoC2 gene that is a reproducible marker for resolving ecotypes (Young et al. 2011).

Maximizing biomass and lignocellulose content is the goal of most switchgrass bioenergy harvests, but the conversion platform likely will determine the optimal switchgrass harvest practices (Vogel et al. 2011). Most research supports a single annual harvest to reach these goals, for optimizing energy inputs, and for maintaining stands (Sanderson et al. 1999; Vogel et al. 2002). Maximum first-cut yields and long-term stand maintenance can be achieved by harvesting switchgrass once during the growing season to a 10-cm stubble height when panicles are fully emerged to the post-anthesis stage (Vogel et al. 2002; Mitchell et al. 2008, 2010). Harvesting after a killing frost often reduces both biomass and nutrient removal, but can provide stable biomass yields and be beneficial for long-term stand maintenance, as well as meeting feedstock characteristics suitable for thermo-chemical conversion (Mitchell and Schmer 2012).

Upland and lowland ecotypes enter dormancy at different rates when grown in the same environment (Mitchell and Schmer 2012). In central and northern latitudes, upland ecotypes senesce rapidly and are completely dormant within 7 days after a killing frost. Lowland ecotypes, however, enter dormancy slowly and have maintained green stem bases for at least 27 days after the first killing frost when exposed to low temperatures of less than 0°C on 17 of the 27 days (Mitchell and Schmer 2012). This delayed dormancy may be one explanation for the winter injury susceptibility of lowland ecotypes in central latitudes (Mitchell and Schmer 2012).

Some research suggests that upland and lowland switchgrass ecotypes may respond differently to harvest timing (Fike et al. 2006a, b), but limited research has been conducted on this topic (Mitchell et al. 2010). Research in the upper South (USA) found that in a twice-per-season cutting system (with the first harvest at near anthesis stage), biomass yield gains were modest for lowland cultivars but increased 30 to 40% with some upland cultivars (Fike et al. 2006a, b). However, the suitability of such management, particularly for improved logistics considered below; see also (Fike et al. 2007; Cundiff et al. 2009), must be weighed against the costs of added harvest, nutrient removal and process efficiency.

Proper harvest timing and cutting height and maintaining adequate N fertility are important management practices required to maximize yield and ensure persistent switchgrass stands (Mitchell et al. 2010; Vogel et al.

2011) . As mentioned previously, research generally indicates a single, post — anthesis harvest during the growing season maximizes yield, but harvesting after a killing frost ensures stand persistence and productivity, especially during drought (Mitchell et al. 2010; Vogel et al. 2011). Vogel et al. (2002) reported switchgrass biomass in the Great Plains and Midwest increases up to anthesis, then decreases by 10 to 20% until killed by frost. This fits well with recommendations by Mitchell et al. (2010) who recommended switchgrass should not be harvested within 6 weeks of killing frost or below a 10-cm stubble height. This management ensures carbohydrate translocation to the plant crowns for setting new tiller buds and maintains stand productivity. With good harvest and fertility management, productive stands can be maintained indefinitely and certainly for more than 10 years (Mitchell et al. 2010).

Switchgrass biomass yield is affected by variables such as ecotype, cultivar, harvest date, fertility, and climate. Recently, a database of switchgrass biomass production studies was compiled from research conducted at 39 field sites in 17 states which supported the single harvest for bioenergy (Wullschleger et al. 2010). Switchgrass yield averaged 8.7 ± 4.2 Mg ha1 for upland cultivars and 12.9 ± 5.9 Mg ha1 for lowland cultivars. Switchgrass harvested once at anthesis in Nebraska and Iowa had greater biomass yields than when harvested twice; yields ranged from 10.5 to 12.6 Mg ha1 yr-1 with no stand reduction (Vogel et al. 2002). In general, harvesting after frost reduces yield, but this practice ensures stand productivity and persistence, especially during drought. Such management also reduces N fertilizer requirements for the following year by about 30% (Mitchell et al. 2010; Vogel et al. 2011). Post-frost harvests allow nutrients, especially N, to be mobilized into roots for storage during winter and to support new growth the following spring. In colder climates, this management practice may have consequences for available moisture in the next growing season as it will reduce the amount of snow captured during winter; these fall harvests also will limit winter wildlife habitat value (Mitchell et al. 2010).

Harvesting after a killing frost is a logical management decision for thermo-chemical conversion platforms and biopower because N, Ca, and other plant nutrients that function as contaminants in the thermo-chemical process are minimized in the plant tissue (Vogel et al. 2002; Fike et al.

2006a, b; Guretzky et al. 2011). Although delaying harvest to after frost may reduce recoverable biomass, it can optimize yields relative to input costs. An analysis by Aravindhakshan et al. (2011) indicated that the economic optimum for switchgrass management would include annual inputs of about 69 kg N ha1 with a single end-of-season harvest.

Some have explored leaving switchgrass standing in the field over winter and harvesting the following spring (Adler et al. 2006). Deferring harvests can reduce yields by 20 to 40% compared with autumn harvests after a killing frost, this loss had no effect on gasification energy yield per unit dry matter but did reduce energy yield per land area (Adler et al. 2006). Yield losses associated with delaying harvest until spring may be acceptable if wildlife cover during winter is critical (Adler et al. 2006), but this is not likely to be a primary driver in most biomass-to-bioenergy systems.

Fungal morphology can be observed under a microscope, and chitin, a specific fungal cell wall component, can be stained with dyes to generally identify fungal species. For a more specific identification, fungal genomic DNA can be isolated using a standard bacterial DNA isolation protocol (Sambrook et al. 1989) or commercial kits, such as the DNeasy Plant Mini Kit (Qiagen).

For identification, the internal transcribed spacer (ITS) regions of fungal ribosomal DNA are widely used because the regions are highly variable (Ghimire et al. 2011). The specific primers ITS1 (5′-TCC GTA GGT GAA CCT TGC GG-3′) and ITS4 (5′-TCC TCC GCT TAT TGA TAT GC-3′) are used to PCR-amplify highly variable ITS1 and ITS2 regions surrounding the 5.8S coding region (Martin and Rygiewicz 2005). However, the primers do not effectively exclude host plant sequences in mixed samples so ITS1-F (5′-CTT GGT CAT TTA GAG GAA GTA A-3′) and ITS4-B (5′-CAG GAG ACT TGT ACA CGG TCC AG-3′) were designed to amplify fungal ITS regions, and a pair of ITS1F and ITS4 resulted in strong PCR amplification from both ascomycetes and basidiomycetes (Gardes and Bruns 1993).

In general, PCR reactions should include 1X reaction buffer containing Mg++, 1 gl of 10 gM of each primer, 1 gl of 10 mM dNTPs, 0.5-1.0 unit Taq DNA polymerase, and 5O-20o ng genomic DNA to total 25-50 gl. PCR can be performed in any thermal cycler with program like 95°C for 2-4 min, then 95°C for 30 sec, 55°C for 30 sec, 72°C for 45 sec for 35 cycles, finally 72°C for 10 min. PCR products are checked in agarose gel first to make sure only one clear band exists, then the product is either cloned into pGEM-T vector (Promega) or similar kits, such as TA cloning kits for sequencing. Direct sequencing of the PCR product after purification using Qiagen’s PCR purification kit is also an option.

Once PCR product sequences are obtained, BLASTN searches can be performed to compare similar sequences from gene bank to identify the species of the target microorganism. A phylogenetic tree can also be constructed to further clarify its evolutionary relationship among other species. In addition, PCR-RFLP, Length Heterogeneity PCR, and Terminal Restriction Fragment Length Polymorphism can be used to characterize microbial communities (Martin and Rygiewicz 2005).

Selection and hybridization have long been used separately or combinationally in the development of cultivars and release of germplasm in switchgrass. Switchgrass is so broadly distributed that enormous genetic variation exists in nature as discussed above. These locally adapted germplasm provide precious raw materials for initial releases of "natural — track" cultivars and named germplasms. Casler (2011) recently provided a comprehensive list of "natural-track" cultivars and germplasms. The Plant Material Centers of the USDA Natural Resource Conservation Service (NRCS) (formerly Soil Conservation Service), State Agricultural Experiment Stations (SAES) and USDA Agricultural Research Service (ARS) laboratories have released most, if not all, of the "natural-track" cultivars and germplasm. Basically, local strains from target geographic regions are collected. Collected accessions are grown in screening nurseries with other available cultivars and germplasm for evaluating important agronomic and adaptive traits. Desirable accessions are further tested in additional environments. Seed of best accessions are increased and released as a new cultivar or a named germplasm. As described by Vogel (2004), release of "natural-track" cultivars and germplasm utilized the natural variation between native ecotypic populations. Selection within individual populations may be applied. For example, ‘Kanlow’ switchgrass was developed using germplasm collected by the Soil Conservation Service at a lowland site near Wetumka, OK. The collection was grown and plants were selected for leafiness, vigor, and retention of green late in season near Manhattan, KS. Seeds of 200 selected plants were increased and released as ‘Kanlow’ by Kansas AES and ARS (Alderson and Sharp 1994). Some other releases are direct seed increase of original collections. ‘Penn Center’ switchgrass was released as a source identified germplasm (USDA NRCS 2010). ‘Southlow Michigan’ switchgrass is a source identified germplasm released by the USDA-NRCS, Michigan Association of Conservation Districts, and the Michigan Department of Natural Resources (Durling et al. 2008). Although "natural-track" releases are genetically similar to or the same as wild plants of the original sources, they provide premier germplasm or populations for developing genetically improved cultivars in breeding programs.

Strong outcrossing sexual behavior and relatively weak asexual reproduction capability of switchgrass require typical recurrent selection procedures used for genetic improvement through hybridization. Vogel and Pedersen (1993) described breeding protocols for outcrossing perennial species. Phenotypic selection within a population consists of preparing seedlings in the greenhouse, transplanting the seedlings into a space-planted nursery, and performing visual selections in year 2 and/or year 3. The selected plants can be used to make new synthetics by poly­crossing with replicated ramets of the selected ones. Seeds of synthetics are grown along with commercial standards for performance and persistence testing at multiple environments (years and locations). ‘Trailblazer’ is an outstanding synthetic cultivar developed by the USDA-ARS and the Nebraska Agricultural Research Division, Department of Agronomy, University of Nebraska (Vogel et al. 1991). It was improved for in vitro dry matter digestibility (IVDMD) by 6% and for beef production by 23% over its source population (Vogel et al. 1991). The success of Trailblazer was evidenced as it was grown over 63,000 ha from 1986 to 1997 (Vogel 2004). Similarly, ‘Shawnee’ was developed with phenotypic selection for IVDMD from a released cultivar ‘Cave-in-Rock’ (Vogel et al. 1996). After two cycles of phenotypic selection, half-sib progenies of 10 superior plants were cross-pollinated to produce Syn 1 seed of ‘Sunburst’ (Boe and Ross 1998). ‘Sunburst’ was improved for seed weight, resulting in superior seedling vigor over other cultivars adapted to the northern Great Plains. Phenotypic recurrent selection with one or more cycles was used to develop ‘Dacotah’, ‘TEM-SLC’, ‘TEM-SEC’, ‘WS4U’and ‘WS8U’ switchgrass (Barker et al. 1990; Tischler et al. 2001; Casler et al. 2006). ‘TEM-LoDorm’ switchgrass was improved through four cycles of phenotypic recurrent selection for reduced seed dormancy (Burson et al. 2009).

Although phenotypic selection within populations was successful for improving IVDMD, reducing seed dormancy, increasing seed weight, and some other traits, the procedure has not been much effective for enhancing biomass yield in switchgrass. Field trials repeatedly indicated narrow-sense heritabilities for biomass yield were low, indicating phenotypic selection is not reliable (Hopkins et al. 1993; Rose et al. 2008; Bhandari et al. 2010 and

2011) . To improve biomass yield, genotypic recurrent selection procedures involving half-sib progeny testing has been used to develop recently released lowland cultivars, ‘Performer’, ‘BoMaster’ and ‘Colony’ by the USDA-ARS and the North Carolina Agricultural Research Service (Burns et al. 2008a, b, and 2010) and ‘Cimarron’ by Oklahoma AES (Wu and Taliaferro

2009) . ‘Cimarron’ switchgrass was developed by poly-crossing seven parents, which were selected from a South Lowland breeding population with one cycle of recurrent selection for general combining ability (RSGCA) after two cycles of restricted recurrent phenotypic selection (RRPS). Details of RSGCA and RRPS used at the Oklahoma State University switchgrass breeding program have been described by Taliaferro (2002). Evidently, half — sib progeny testing is effective in the selection of elite parents for synthetic cultivar development in switchgrass.

Up to now, 21264 hairpins and 25141 mature miRNAs in 193 species have been deposited in miRBase (miRBase release 19, 2012, http: //www. mirbase. org/) (Griffiths-Jones et al. 2008), and 9277 mature plant miRNAs have been deposited in Plant MicroRNA Database, PMRD (http: //bioinformatics. cau. edu. cn/PMRD/) (Zhang et al. 2010). A variety of strategies and approaches have been developed to identify novel miRNAs in diverse plant species (Bonnet et al. 2004; Jones-Rhoades and Bartel 2004; Wang et al. 2004; Adai et al. 2005; Zhang et al. 2005; Zhao et al. 2007; Meyers et al. 2008; Zhang et al. 2008; Zhou et al. 2010; Wang et al. 2011). This increasing knowledge of plant miRNAs and well-developed common tools have provided great opportunities for identification of miRNAs in switchgrass.

MiRNAs have been identified by three common approaches: direct cloning, forward genetics and bioinformatics predication followed by experimental validation (Jones-Rhoades et al. 2006). Forward genetics is rarely used for plant microRNA discovery (Jones-Rhoades et al. 2006). Cloning is the most direct and initial method for large-scale miRNA discovery (Reinhart et al. 2002; Jones-Rhoades et al. 2006). It includes isolation of small RNAs, ligation of small RNAs to adaptor oligonucleotides, reverse transcription, amplification and sequencing (Jones-Rhoades et al. 2006). The early sequencing method is conventional Sanger sequencing, which was successful in identifying some conserved miRNAs (Reinhart et al. 2002; Sunkar and Zhu 2004; Axtell and Bartel 2005), but is relatively in low-depth and not ideal for discovering evolutionarily young miRNAs with low abundance (Moxon et al. 2008). The next generation sequencing (NGS), such as the 454 technology and the Solexa platform, provides a powerful high throughput tool, which has greatly facilitated the identification of novel miRNAs (Lu et al. 2005; Rajagopalan et al. 2006; Fahlgren et al. 2007; Yao et al. 2007; Moxon et al. 2008; Sunkar et al. 2008; Zhao et al. 2010; Chi et al. 2011).

However, cloning method suffers from several limitations such as sequence-based biases during the cloning procedures, and difficulty in detecting miRNAs expressed in low levels or only in response to certain stressors (Jones-Rhoades et al. 2006). Bioinformatics approaches can be a great complement to overcoming these limitations (Jones-Rhoades and Bartel 2004; Jones-Rhoades et al. 2006; Gebelin et al. 2012). Thus, a combination of bioinformatics prediction and experimental validation approaches is often used to identify plant miRNAs.

With the access of the complete genome sequencing data, it is possible to predict and identify a complete set of conserved miRNAs using bioinformatics approaches (Matts et al. 2010; Thakur et al. 2011). However, currently there have been neither complete genome sequencing data, nor large genomic fragmented data (genomic survey sequences, whole-genome shortgun reads and high throughput genomic sequences) available for switchgrass (Matts et al. 2010). Although lack of genomic resources can be a challenge for identifying the complete set of conserved switchgrass miRNAs, the availability of expressed sequence tags (ESTs) deposits (currently 720,590 ESTs deposited in NCBI database) can be a viable source for miRNA discovery. Moreover, previous research has already led to discovery of conserved miRNAs from diverse plant species by EST database mining (Zhang et al. 2006; Sunkar and Jagadeeswaran 2008; Matts et al. 2010; Gebelin et al. 2012).

The wide geographic distribution of switchgrass provides clear evidence of the species’ large genetic variation and broad adaptability. Indeed, this was a key attribute in the selection of switchgrass as a model species for energy production studies (McLaughlin and Kszos 2005; Wright and Turhollow

2010) . Within regions, switchgrass’ suitability to widely ranging edaphic and fertility environments is largely a function of ecotypic adaptation. Mesic, upland sites most often are occupied by upland ecotypes, which have lower sensitivity to moisture stress, whereas hydric bottomlands are the typical habitat for lowlands.

Although soils for these landscape positions may be quite distinct, soil type per se does not appear to have a particularly strong effect on switchgrass production (Sanderson et al. 1999). However, soil texture—and thus, soil water holding capacity—can have strong effects on switchgrass establishment and yield. Both excessively drained sandy soils and soils with poor internal drainage (as found on depositional sites—e. g., see Thelemann et al. (2010)) can limit switchgrass productivity.

Within the same soil type, slope may be a significant variable for switchgrass production. For example, Fike et al. (2006a) reported that yields on two sites with the same soil type differed by about 40% although the plots were only about 100 m apart. Aspect may also have played a role in these results as the higher yielding plots were more south facing. However, reductions in water infiltration and availability would be expected with increased slope.

Typical switchgrass production guides advise amending soil pH to 6.0 or higher for planting (e. g., Teel et al. 2003). However, switchgrass can tolerate a wide pH range (5-8) for germination (Hanson and Johnson 2005) and soil acidity is rarely a limiting factor for switchgrass in normal production settings. In fact, root growth at pH 3.7 was observed for switchgrass grown on a mine land reclamation site (Stucky et al. 1980). While reclaimed mine lands would be outside the typical boundaries for growing agronomic crops, switchgrass’ adaptability to harsh conditions, its tolerance of soil contaminants, and its capacity to grow on poorly structured soils makes it a useful crop for such difficult-to-crop sites.

Cultivar Selection

It may seem surprising—or even disconcerting—to note that as of this writing (2012), no registered cultivars specifically bred for biomass production have been released. But the reader should not be alarmed; switchgrass, relative to traditional row and forage crops, has been under investigation for only a relatively short period of time—and particularly so for bioenergy purposes. Historically, nearly all cultivars were selected for forage and conservation uses; only recently have registrations included descriptions of switchgrass cultivars as suited for bioenergy production. Such "multitasking" will likely be the norm for switchgrass for the foreseeable future, although these different purposes (forage and biofuels) require rather disparate management practices.

For any site, switchgrass cultivar selection should be driven by the need to match the plant materials to the extremes of the local growing conditions, regardless of end use. Adaptation to seasonal moisture regimes and temperature maxima and minima would be of primary consideration for choice of cultivars. With these factors in mind, we reconsider the value of moving switchgrass of "low-latitude" origin to greater latitudes. As we have noted above, such a strategy can increase yields by delaying inflorescence development, but this approach has bounds (and risks) because switchgrass needs adequate time to develop freeze resistance going into winter (Hope and McElroy 1990; Casler et al. 2004). Thus, moving low-latitude lines too far north (or to too high elevation) may be akin to imposing a death sentence on the crop. The general guideline of Moser and Vogel (1995) is to choose cultivars that are not more than 500 km from their latitude of origin, although we note exceptions where conditions are warmer than would be predicted based on latitude alone (e. g., Christian et al. 2001).

The ability of some endophytes to colonize the xylem provides the opportunity for their systemic spread throughout the rest of the plant, via the transpirational stream in the xylem lumen. However, not all endophytes are capable of colonizing the aerial parts of plants. This may reflect the inability of some to adapt and survive the different niches represented by aerial tissues and organs (Compant et al. 2010). In switchgrass, B. phytofirmans strain PsJN titers were higher in the root than in the leaves 7 days post­inoculation of the roots. However, by 14 days post-inoculation, titers were higher in leaves and sheaths than in the roots, indicating translocation to these tissues (Kim et al. 2012). Generally, bacterial endophyte titers in the aerial plant tissues are reported to be lower than in the root (Rosenbleuth and Martinez-Romero 2006; Compant et al. 2008). In addition, a fair amount of variation can be observed in these tissues. Compant et al. (2008) reported that PsJN could be found in only 10-60% of grape inflorescence stalks and grape berries following initial inoculation of roots. These were localized to xylem vessels, and only a single or few cells were observed. These results further indicated the importance of the xylem for systemic spread of endophytes, allowing them to reach as far as the reproductive tissues. However, this spread was very slow, taking 5 weeks to reach inflorescence tissues. The very low titers of PsJN that ended up in these tissues was attributed to competition with other co-localized endophytes, which can inhabit different tissues and organs, reflecting different niches of colonization (Compant et al. 2011). This report of endophytic bacteria being low or absent in flowers and fruits echoes other comments (Hallman 2001), suggesting low vertical transmission. Bacterial colonization, in general, varies from one cultivar to another and depends on many factors. For example, in soybean, plant genotype, tissue age, season of isolation, and herbicide application, all affected colonization (Kuklinsky-Sobral et al. 2004).

In grasses, the most abundant matrix polysaccharide (i. e., hemicellulose) is xylan. In the last five years, several GTs and other enzymes from Arabidopsis and grasses that likely function in xylan biosynthesis have been characterized (Fig. 2). A complete list of the proteins that function in and the mechanism of xylan synthesis in the Golgi and subsequent release into the cell wall is still being unraveled. Here, we will discuss recent progress with an emphasis on results in grasses. Table 2 lists enzymes implicated in synthesis of the xylan backbone, reducing end oligosaccharide, and sidechains. The reader is referred for additional information to other recent reviews on the topic (Buanafina 2009; Faik 2010; Scheller et al. 2010; Carpita

2012) . When it has been assayed, cell wall material from loss-of-function xylan mutants all exhibit improved digestibility (Mortimer et al. 2010; Brown et al. 2011; Chen et al. 2013; Chiniquy et al. 2012), consistent with the role of this polysaccharide in stabilizing cell walls.

At least two Carbohydrate-Active enZyme (CAZy) database GT families are implicated in the backbone synthesis of xylans, namely GT43s and GT47s (Table 2). Studies in Arabidopsis showed that GT43s are responsible for xylan backbone synthesis since irx9 and irx!4 have drastically shorter xylan chains and reduced ability to transfer Xyl from UDP-Xyl onto xylo-oligosaccharide acceptors (Brown et al. 2007; Pena et al. 2007). Similarly, double mutation of two GT47 genes, irxl0 and irxl0-L severely reduced xylan length, but without affecting the reducing end of xylan. The importance of GT47 family enzymes in xylan synthesis has recently been extended to rice. The Osirx10 mutant has greatly reduced xylan amounts in stems without showing a reduction in xylan chain length, suggesting a somewhat different xylan synthesis mechanism in grasses compared with dicots (Chen et al. 2012). A biochemical study in wheat (Triticum aestivum) suggested proteins from the GT43, GT47, and GT75 families are promising candidates for members of the gluronoarabinoxylan synthetase. Coimmunoprecipitation indicated these three GT proteins interact with each other to form a complex exhibiting xylan synthesis activity (Zeng et al. 2010).

Other studies in Arabidopsis have identified proteins that function in synthesis of the xylan reducing-end oligosaccharide (Fig. 2B), which has been found in several dicots and conifers but not detected in grasses (York et al. 2008; Scheller et al. 2010). The sequence of this reducing end oligosaccharide, or "primer", is 4-p-D-Xylp-(1—>4)-p-D-Xylp-(1—>3)-a- L-Rhap-(1—>2)-a-D-GalpA-(1—>4)-D-Xylp. As summarized in Table 2, the mutants irx7/fra8, irx8, and parvus are depleted for the reducing-end oligosaccharide (Pena et al. 2007). IRX8 and PARVUS, both GT8s, are implicated in adding the galacturonic acid and a a-xylose residue to the primer (Lee et al. 2007; Pena et al. 2007). IRX7, and its close homolog F8H, have been implicated as Rha-specific xylosyltransferases, because they act on a diversity of sugars (Rennie et al. 2012). Despite the absence of the reducing-end primer in experiments in grasses, enzymes with sequence similarity to those implicated in its synthesis have been retained in the rice genome (Scheller et al. 2010).

Other recent work has revealed enzymes that likely function to attach the xylan side chains, glucuronic acid and, in grasses, arabinose. Mortimer et al. (2010) identified mutants in two GT8 family genes, gux1 and gux2. The proteins encoded by these genes are Golgi-localized and required for the addition of both glucuronic acid and 4-O-methylglucuronic acid branches to xylan in Arabidopsis stem cell walls (Mortimer et al. 2010). Recently, another double mutant, irx15 and irx15L, was also found to be involved in xylan synthesis (Brown et al. 2011). These two genes, which belong to the domain of unknown function 579 family, might also be glucuronic acid transferases because they exhibited similar mutant features to gux1 and gux2 (Brown et al. 2011). For addition of the side chains of grass xylan, studies have focused on GT61 family members, which are much more highly expressed in grasses than in dicots (Mitchell et al. 2007). Repression of expression of a GT61 encoding gene, TaXat1, in wheat endosperm and its heterologous overexpression in Arabidopsis provided strong evidence that TaXAT1 possesses a-(1,3)-arabinosyltransferase activity (Anders et al.

2012) . Disruption in rice of another GT61 encoding gene, called Xax1, has also been published recently (Chiniquy et al. 2012). This mutant is lacking a previously observed but poorly characterized xylan substitution, an arabinofuranose residue substituted at the O-2 position of a xylosyl residues, in the structure p-Xylp-(1^-2)-a-Araf-(1^-3) (Fig. 2A). Based on H1-NMR and glycosidic linkage analysis, XAX1 possess a xylosyl transferase activity that can attach (P-1,4-Xylp)4 onto the acceptor p-Xylp-(1^2)-a-Araf-(1^3) (Chiniquy et al. 2012).

Other work has provided important insight into synthesis of the arabinofuranose (Araf) nucleotide sugar precursor of grass cell wall glucuronoarabinoxylan. A recent study of reversibly glycosylated proteins (RGPs) in Arabidopsis showed that the conversion of UDP-p-L — arabinopyranose (UDP-Arap) to UDP-p-L-Araf is indispensable for cell wall synthesis (Rautengarten et al. 2011). The rGps are in the GT75 family. The knockout mutants, rgp1 and rgp2 significantly reduced the total L-Ara content relative to the wild type and showed reduced UDP-Ara mutase (UAM) activity. UAM activity has been identified in rice, as well (Konishi et al. 2007; Konishi et al. 2011). Three rice genes with close sequence similarity to RGP-encoding genes were predicted to be UAM candidates (Konishi et al.

2007) . Later, knock-down of one of these three genes suppressed the UAM activity and reduced UDP-Araf amounts in mutant rice plants (Konishi et al. 2011). The mutant also decreased the incorporation of ferulic acid and p-coumaric acid to the cell wall and presented dwarfed and infertile phenotypes (Konishi et al. 2011). One of the GT75 family members in wheat mentioned above has been inferred to have the UAM activity but needs to be further studied (Zeng et al. 2010).

The final set of recent advances in our understanding of xylan synthesis in grasses relates to acylation of grass xylan arabinose residues by the hydroxycinnamates, ferulic acid and p-coumaric acid (Buanafina 2009). Consistent with expectations, in two studies of rice mentioned above, mutants with reduced Araf substituted xylan had reduced cell wall content of ferulic acid and p-coumaric acid (Konishi et al. 2011; Chiniquy et al. 2012). Mitchell and colleagues developed the hypothesis that 12 members of the BAHD family of acyl-CoA acyltransferases that were much more highly expressed in grasses than dicots might act as arabinofuranose feruloyl transferases (Mitchell et al. 2007). Later, silencing of four members of this family in rice (LOC_Os05g08640, LOC_Os01g09010, LOC_Os06g39470 and LOC_Os06g39390) reduced the ferulic acid content in young leaves by about 20%, leading to the refined hypothesis that one or more of the targeted genes acts as the feruloyl transferase (Piston et al. 2010). Recently, the overexpression of one of the same genes examined by Piston et al., LOC_Os06g39390, dubbed OsAT10, has led to the hypothesis that this protein functions as a p-coumaroyl transferase, as the overexpression plants have increased levels of p-coumaroyl esters bound to arabinose in the cell wall (Bartley et al. 2013a). The plants also exhibit a reduction in the level of polysaccharide-linked ferulic acid and show a concomitant improvement in digestibility (Bartley et al. 2013a), consistent with the model that ferulate — mediated crosslinking is important for grass cell wall digestibility.