Scholarly article on topic 'Energy Sorghum--a genetic model for the design of C4 grass bioenergy crops'

Energy Sorghum--a genetic model for the design of C4 grass bioenergy crops Academic research paper on "Biological sciences"

0
0
Share paper
Academic journal
Journal of Experimental Botany
OECD Field of science
Keywords
{""}

Academic research paper on topic "Energy Sorghum--a genetic model for the design of C4 grass bioenergy crops"

Journal of Experimental Botany, Vol. 65, No. 13, pp. 3479-3489, 2014

doi:10.1093/jxb/eru229 ^^^ i Journal of

Experimental

REVIEW PAP E R www.jxb.oxfordjounials.org

Energy Sorghum—a genetic model for the design of C4 grass bioenergy crops

John Mullet1*, Daryl Morishige1, Ryan McCormick1, Sandra Truong1, Josie Hilley1, Brian McKinley1, Robert Anderson1, Sara N. Olson1 and William Rooney2

1 Department of Biochemistry and Biophysics, Texas A&M University, College Station, Texas 77845-2128, USA

2 Department of Soil and Crop Science, Texas A&M University, College Station, Texas 77845-2128, USA

* To whom correspondence should be addressed. E-mail: jmullet@tamu.edu Received 24 January 2014; Revised 14 April 2014; Accepted 22 April 2014

Abstract

Sorghum is emerging as an excellent genetic model for the design of C4 grass bioenergy crops. Annual energy Sorghum hybrids also serve as a source of biomass for bioenergy production. Elucidation of Sorghum's flowering time gene regulatory network, and identification of complementary alleles for photoperiod sensitivity, enabled large-scale generation of energy Sorghum hybrids for testing and commercial use. Energy Sorghum hybrids with long vegetative growth phases were found to accumulate more than twice as much biomass as grain Sorghum, owing to extended growing seasons, greater light interception, and higher radiation use efficiency. High biomass yield, efficient nitrogen recycling, and preferential accumulation of stem biomass with low nitrogen content contributed to energy Sorghum's elevated nitrogen use efficiency. Sorghum's integrated genetics-genomics-breeding platform, diverse germplasm, and the opportunity for annual testing of new genetic designs in controlled environments and in multiple field locations is aiding fundamental discovery, and accelerating the improvement of biomass yield and optimization of composition for biofuels production. Recent advances in wide hybridization between Sorghum and other C4 grasses could allow the deployment of improved genetic designs of annual energy Sorghums in the form of wide-hybrid perennial crops. The current trajectory of energy Sorghum genetic improvement indicates that it will be possible to sustainably produce biofuels from C4 grass bioenergy crops that are cost competitive with petroleum-based transportation fuels.

Key words: Biomass yield, C4 grass, energy Sorghum, flowering time, genomics, traits.

Introduction

The drive to develop improved bioenergy crops is part of a worldwide effort to increase the productivity of agriculture to provide food, feed, fibre, bioproducts, and bioenergy for a world population that is increasing and undergoing rapid development (Foley et al., 2011). It is projected that world population will exceed 9 billion by 2050 and that production of food for humans and feed for animals will need to double to meet expected demand. In addition, numerous countries have set ambitious targets for increased production of biofuels for economic, environmental, and national security reasons. Biofuels derived from C4 grass bioenergy crops such as switchgrass (Schmer et al., 2008), Miscanthus (Hillier et al.,

2009; Davis et al., 2010), and energy Sorghum (Olson et al., 2012) have large greenhouse gas displacement effects that could help slow the rise of atmospheric CO2 levels, caused in part by the burning of fossil fuels. Current advances in the design of energy crops and biofuels production systems will help prepare for the eventual depletion of fossil fuels that can be extracted in a cost effective way. In the United States the Energy Independence and Security Act of 2007 (Rahall, 2007) mandated annual production of 136 B litres of biofuels, approximately 20-25% of the U.S. transportation fuel requirement, with no more than 57 B litres from grain crops such as corn and 79 B litres from bioenergy crops and other

© The Author 2014. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved. For permissions, please email: journals.permissions@oup.com

3480 | Mullet et al.

non-grain feedstocks. By 2011, the United States was producing ~50 B litres of ethanol principally from corn grain using ~40% of the U.S. corn crop for this purpose (National Corn Growers Association, 2012; Ranum et al., 2014). Therefore, the United States is approaching the mandated limit for production of biofuels from grain such as corn, as well as the current blend limit for ethanol in gasoline. Further increases in biofuels production in the United States will require the utilization of non-grain sources of hydrocarbon feedstocks, derived principally from crop and forest residues, algae, and dedicated bioenergy crops.

There is increasing concern about the long-term prospects for establishment of an economically viable, bioenergy crop-based biofuels industry in the United States. Production of biofuels from non-grain crops has not met near-term mandated production targets in the United States, owing in part to the technical challenges associated with converting biomass to biofuels and the high cost of lignocellulose-derived biofuels. Risks associated with uncertainty about optimal methods for biomass to biofuel conversion, the outcome of food-versus-fuel land-use policy debate, and expanding supplies of natural gas and oil generated by fracking, have resulted in limited capital investment in biorefinery capacity. Increased utilization of biofuels also may require transition to biobutanol or a hydrocarbon mixture that will eliminate blend limits and increase biofuel energy density. However, concern about the viability of the biofuels industry often does not consider the very short history of bio-energy crop development, and the significant potential for improvement of bioenergy crop yield and quality. The dramatic ~4-fold increase in corn grain yield per hectare in the United States since 1950 demonstrates the potential impact of optimized crop genetics, inputs, and production systems (Duvick, 2001; Hammer et al., 2010a). The economics of biofuels production can be improved by increasing the yield of bioenergy crops, reducing input requirements, and by optimizing biomass composition for conversion to biofuels. Current production of biomass using energy Sorghum hybrids is estimated to cost ~$50-60 per dry Mg, a price threshold enabling competitive production of biofuels (U.S. Department of Energy, 2011). Current yields of C4 grass bioenergy crops such as energy Sorghum are ~25% of their theoretical maximum yield potential (see below); therefore, improvements in the genetic design of these energy crops and optimization of management and inputs are expected to increase biomass yield per hectare, further reducing the cost of biomass production and transport to biorefineries, as well as the amount of land used for biofuels production. The Billion Ton Studies (Perlack et al., 2005; U.S. Department of Energy, 2011) found that there is sufficient land in the United States to produce ~1.3 B dry Mg of biomass each year without affecting food production. One billion metric tons of dry biomass could be converted into ~375 M litres of transportation fuel, sufficient to supply ~75% of the total U.S. yearly requirement. This indicates that there is sufficient land for large-scale bioenergy crop production in the United States. Moreover, bioenergy crop production systems managed for sustainability would become renewable

sources of biofuels that contribute to energy security over the long term.

In the United States, a focus of bioenergy research, development, and funding has been on perennial bioenergy C4 grass crops such as switchgrass and Miscanthus. Recently, Sorghum has emerged as a genetic model and genome reference species for the design of C4 grass crops used for grain, forage, sugar, and bioenergy production. This review will focus on the development of energy Sorghum since 2007, and the role genetics and genomics technology has played in accelerating the design of this new type of Sorghum crop for biomass/bio-energy production. Description of energy Sorghum development in the United States will illustrate how crop design is influenced by the genetic attributes of the target crop species, the environment where the crop will be produced, government policy, conversion technology, and economic considerations. Numerous excellent reviews are available covering relevant background information on Sorghum production and genetics (Rooney, 2004; Lawrence and Walbot, 2007; Rooney et al., 2007; Vermerris, 2011; Calvino and Messing, 2012), C4 photosynthesis (Zhu et al., 2010; Byrt et al., 2011; Sage and Zhu, 2011), the genomics of grass cell wall biochemistry (Pauly and Keegstra, 2008; Penning et al., 2009), and other C4 bioenergy grasses (Waclawovsky et al., 2010; Jorgensen, 2011; Weijde et al., 2013).

C4 grass energy crops

C4 grasses of the PACCAD clade of the Panicoideae such as switchgrass (Panicum virgatum) (Bouton, 2007), sugarcanes (Saccharum spp.) (Waclawovsky et al., 2010), Miscanthus (Heaton et al., 2004; Jorgensen, 2011), and Napier grass (Pennisetum purpureum) have excellent potential as dedicated energy crops because these grasses can accumulate large amounts of biomass (>40 dry Mg ha-1). They are perennials that can be grown with low input requirements on land not optimal for annual food/feed crops (Carroll and Somerville, 2009; Somerville et al., 2010). The high productivity of these grasses is due to C4 photosynthesis, which also contributes to improved nitrogen use efficiency (Byrt et al., 2011). C4 grasses of the same clade, such as Sorghum bicolor, Eragrostis tef, and Pennisetum glaucum, are well adapted to hot, water-limited environments. Adaptation of energy crops to adverse nutrient-limited environments is important if dedicated bioenergy crops are grown without irrigation on marginal soils in environments that are prone to water deficits.

Most of the perennial C4 grasses being developed for bioenergy purposes are perennial polyploids with large genomes. For example, Miscanthus x giganteus is a sterile triploid (3n=57 chromosomes) with a genome size of 7.0 pg, Miscanthus sacchariflorus is a tetraploid (4n=76 chromosomes) with a genome size of 4.5 pg, and Miscanthus sinensis is a diploid (2n=38 chromosomes) with a genome size of 5.5 pg (2650 Mbp/C), approximately the size of the maize genome (Rayburn et al., 2009). Switchgrass genotypes are either octoploid (2n=8x=72 chromosomes) or

tetraploid (2n=4x=36 chromosomes) outcrossing perennial grasses with large genomes (~1,600 Mbp/C) native to North America (Bouton, 2007; Sharma et al., 2012). Sugarcane is an even more complex polyploid C4 perennial grass used for sugar and bioenergy production, currently grown on ~23M ha in 90 countries worldwide (Waclawovsky et al., 2010). In contrast, Sorghum bicolor is a diploid C4 grass (2n=10 chromosomes) with an 800 Mbp genome (Kim et al., 2005; Paterson et al., 2009). Plants with large polyploid genomes are often highly productive and environmentally resilient. However, genetic and genomic analysis of perennial C4 grasses with complex genomes and polyploid genetics can be challenging. This led to the development of genetic models for bioenergy grasses such as the C3 grass genetic model Brachypodium distachyon (The International Brachypodium Initiative, 2010), and C4 grass genetic models Setaria (Li and Brutnell, 2011) and Sorghum bicolor.

Sorghum was not initially targeted for energy crop development because this C4 grass species is grown principally as an annual grain and forage crop, whereas energy crop development focused on perennials. However, studies conducted in the United States in the late 1970s, funded in response to the oil embargo in 1973, demonstrated that Sorghum had significant potential for biomass accumulation (i.e., Monk et al., 1984). Sorghum's drought tolerance is also well known, an important attribute of energy crops (Rosenow and Clark, 1995; Borrell and Hammer, 2000; Borrell et al., 2006). Moreover, it is now clear that development of energy Sorghum, an annual C4 grass with tractable genetics, can aid in the optimization of energy crops because new genetic designs can be tested in controlled environments and multiple field locations each year, speeding up the rate of genetic improvement. In addition, during biorefinery establishment, annual energy Sorghum hybrids can provide a source of feedstock while perennial energy crops are being established (Rooney et al., 2007). Energy Sorghum production can also be adjusted annually, reducing risks associated with perennial crops that require long-term allocation of land to a crop that takes several years to reach optimal yield.

Energy crops are being designed to minimize the impact of biomass feedstock generation on the production and cost of food. The importance of this long-term objective is clear because up to 40% of the U.S. corn crop is already used for ethanol (Ranum et al., 2014). Sorghum's exceptional drought tolerance allows it to be planted on land that is not optimal for most food crops. Furthermore, energy Sorghum grows most or all of the season as a vegetative plant, avoiding water deficit during the most sensitive reproductive phase of development. Energy Sorghum is also being developed as a dual use crop for biofuel and forage production. The use of biomass produced from energy Sorghum for bio-energy and/or animal forage will help minimize the impact of energy Sorghum production on animal forage supplies while improving the economics of energy crops. In 2011, 36% of the maize crop was used for animal feed indicating the potential utility of a highly productive dual use energy/

forage Sorghum crop (National Corn Growers Association, 2012; Ranum et al., 2014).

Energy Sorghum—an emerging genetic model for C4 grass crops

Sorghum is a good genetic model and genome reference species for C4 grasses of the PACCAD clade because it has a diverse germplasm, a good genetics and genomics platform, broad adaptation, and worldwide utilization as a grain, forage, sugar, and bioenergy crop (~65M hectares). Sorghum is a predominantly self-pollinated species with tractable genetics that enables inbred/hybrid breeding, population development, and quantitative genetic analysis. Sorghum-breeding programs fostered the development of quantitative genetics-genomics discovery platforms and crop modelling (Hammer et al., 2010b) to aid crop improvement (Fig. 1). Sorghum breeding pools contributed accessions for a grain Sorghum association panel (Morris et al., 2013) and breeding populations were modified in design to enhance QTL discovery (Jordan et al., 2011). Permanent RIL mapping and TILLING populations were established to aid quantitative genetic analysis and gene discovery (i.e. Hart et al., 2001; Xin et al., 2008; Mace and Jordan, 2011). Development of genotyping technologies with increased precision and resolution based on sequencing (i.e. Elshire et al., 2011; Mace and Jordan, 2011; Morishige et al., 2013), high resolution genetic maps (i.e. Menz et al., 2002; Mace et al., 2008; Morishige et al., 2013), genome sequencing (Paterson et al., 2009; Zheng et al., 2011; Evans et al., 2013; Mace et al., 2013), transcriptome analysis (Buchanan et al., 2005; Dugas et al., 2011), and improvements in transgene technology (Liu and Godwin, 2012) have transformed Sorghum into an excellent platform for genetic discovery, fundamental genome analysis, and marker-assisted breeding (MAB) (Fig. 1, centre). Advances in genome

Sorghum Genetics/Genomics/Breedina Platform

Genome Technology Genotyping

Genome Sequencing

Phenomics/Modeling ->

Transcriptomics

Epigenetics

TILLING/Transgenics

ZFN/TALENs/CRISPR ^

Genetics Platform

Germplasm *

Association Panels *

RIL Populations *

QTL Mapping *

Allele Discovery *

Trait Networks *

Trait Engineering

Breeding Pipeline

Germplasm *

Breeding Panels *

Breeding Populations *

Inbred Breeding

Hybrid Breeding

Wide Hybrids

Fig. 1. Integrated Sorghum genomics, genetics, and breeding platform. Genomic technologies (left) are increasing the utility of the Sorghum genetic platform (centre) that extends from germplasm analysis to trait engineering. The genetic platform integrates with the Sorghum-breeding pipeline (right) that currently generates improved inbreds, hybrids and wide hybrids for utilization.

3482 | Mullet et al.

sequencing technology have helped characterize variation in gene distribution, recombination frequency, gene expression, distribution of polymorphism (SNP, INDELs, CNV, PAV), and patterns of DNA methylation (Calvino et al., 2008; Evans et al., 2013; Mace et al., 2013; Morris et al., 2013) within the Sorghum genome (Fig. 1, left). These advances provide a solid foundation for a future Sorghum ENCODE project focused on the elucidation of gene regulatory elements, transcription factors, and networks that will enable targeted trait engineering using a combination of genomic selection (Bernardo and Yu, 2007; Heffner et al., 2009), transgenics, and other genome modification technologies (e.g. Chen and Gao, 2013).

Sorghum germplasm

The world collections of Sorghum are large (n=~43 000) as well as phenotypically and genetically diverse (e.g. Billot et al., 2013; Mace et al., 2013). Sorghum bicolor diverged from a common ancestor of rice ~50 MYA (Doebley et al., 2006) and C4 photosynthesis developed in this lineage ~25-30 MYA (Kellogg, 2001). The C4 pathway helped Sorghum become adapted to the hot dry environments of the equatorial region

of Africa and other diverse environments extending from the equator to 30-32 degrees north and south latitude (Kimber, 2000). Domestication of Sorghum may have occurred several times between ~4000 and 8000 BC, most likely starting with selection for reduced seed shattering and increased seed size (Lin et al., 2012; Mace et al., 2013). Cultivated Sorghum moved along the trade routes east to west and south, where it is thought to have outcrossed with wild Sorghum in each region (Kimber, 2000). As a consequence, landraces reflect regional adaptation of the Sorghum races to southern (Kafir), western (Guinea), central-west equatorial (Caudatum), and northern Africa-India (Durra). Sorghum is also found in Australia, India, and China and in the 1800s Sorghum was introduced into North and South America.

Genotypes used in Sorghum breeding include accessions originating from all regions of Africa. To illustrate this point, ~473 Sorghum accessions used for grain, sweet Sorghum, forage, and energy crop breeding were analysed for sequence variation using Digital Genotyping, a genotyping by sequencing technology (Morishige et al., 2013), and open source analysis software (Paradis et al. 2004; Purcell et al. 2007; Li and Durbin, 2010; Tange 2011; Catchen et al., 2011;

Fig. 2. Phylogenetic tree of Sorghum accessions used in crop breeding. DNA polymorphisms that differentiate Sorghum accessions were collected using Digital Genotyping. Sorghum accessions are coloured based on race: Durra (green), Caudatum (blue), Guinea (red), Kafir (purple). Representative countries of origin are listed.

http://www.R-project.org/). In total, ~40 000 SNPs were identified and used to assess genetic similarities and differences among accessions (Fig. 2). When the accessions were divided into five clusters based on genetic similarity, known accessions of the same race were grouped together (Fig. 2, Kafir (purple), Durra (green), Caudatum (blue), Guinea (red)). The structure of diversity present in this germplasm panel reflects regional origins of the Sorghum races and adaptation to diverse agro-ecological regions of Africa, as well as movement of germplasm and subsequent breeding and selection activities outside the African continent.

Diversity among the 473 accessions used for breeding was extensive with 80% of the >50 000 unique 100 base sequences analysed by Digital Genotyping containing SNPs. The set of 23 000 DG-sequences used for diversity analysis contained an average of 2.3 alleles per sequence. Extensive sequence diversity was also found in comparisons of genome sequences of three grain Sorghum genotypes derived from Kafir, Caudatum and Durra races where ~2.8 M SNPs and 0.27 M INDELs were identified that distinguish these genomes (Evans et al., 2013; see also Zheng et al., 2011; Mace et al., 2013). The polymorphism density distribution across Sorghum chromosomes was highly variable possibly reflecting differences in rates of polymorphism generation, random drift, and forces of selection (Evans et al., 2013). Other studies indicate that the entire Sorghum germplasm collection, which includes wild Sorghum accessions, is very diverse providing a rich source of alleles for future Sorghum improvement (Mace et al., 2013). More than 50 Sorghum genomes have been sequenced to date revealing extensive SNP, INDEL, and PAV/CNV variation consistent with the diversity of Sorghum phenotypes observed in Sorghum germplasm collections, association panels, and breeding pools. As will be illustrated below, phenotypic diversity in the Sorghum germplasm collection is extensive, reflecting the species' ~50MY history since divergence from a common ancestor with rice and selection in the diverse and often harsh environments of Africa. Current genomics technology provides the opportunity to characterize the extensive diversity in the Sorghum germplasm collection at high resolution, greatly accelerating the rate of discovery of useful alleles and their utilization for Sorghum improvement.

Sorghum genome organization

The 800 Mbp Sorghum genome encodes ~30 000 genes distributed among 10 chromosomes (Kim et al., 2005; Paterson et al., 2009). Sorghum's genome has remained fairly stable over the past 50MY based on comparison of the Sorghum and rice genomes (Kim et al., 2005). Sorghum has two fewer chromosomes than rice, the result of fusions of ancestral chromosomes corresponding to rice chromosomes 3/10 to form the largest Sorghum chromosome SBI-01, and rice chromosomes 7/9 to form SBI-02 (Kim et al., 2005). Genes in the two genomes show a high degree of colinearity (~75-80%) and although Sorghum's genome size (~800 Mbp) is larger than rice (~400 Mbp), this is primarily due to more repetitive DNA in the pericentromeric regions of Sorghum

chromosomes (Kim et al., 2005). Cytogenetic analysis in conjunction with analysis of DNA methylation, gene density, and rates of recombination showed that the distal regions of most Sorghum chromosomes are euchromatic, gene dense, and characterized by high rates of recombination and relatively low levels of DNA methylation (Kim et al., 2005; Evans et al., 2013). Nearly 50% of Sorghum's genomic DNA is located in pericentromeric regions of low gene density, characterized by low rates of recombination and high levels of DNA methylation.

Sorghum breeding and genetics

The selection of Sorghum with improved traits has been ongoing since domestication ~8000 year ago (Kimber, 2000). Intensive Sorghum breeding in the United States began in the early 1900s and was facilitated by the fact that Sorghum is a diploid, principally self-pollinating species, that is relatively easy to cross- and self-fertilize to produce inbred lines for cultivation, hybrid development, and genetic analysis of traits (reviewed in Quinby, 1974; Smith and Frederiksen, 2000; Rooney, 2004). Initially grown as a cultivar, modifications in maturity (earlier) and height (shorter) resulted in 'Wheatland,' 'Caprock,' and 'Combine Kafirs', which were used extensively in the United States before 1950 (Smith and Frederiksen, 2000). Hybrid Sorghum was developed in the 1950s, following identification and implementation of a useful cytoplasmic male sterility system (Smith and Frederiksen, 2000). More recently, advances in wide hybridization between Sorghum, cane, and other C4 grasses has dramatically expanded the potential scope of Sorghum breeding to the development of perennial wide hybrid Sorghum x C4 grass crops (Hodnett et al., 2010). Sorghum breeding progress is aided by the use of multi-location breeding/evaluation sites enabling up to three growing seasons per year (Rooney, 2004). Sorghum's genetic tractability, broad adaptation, and the utility of diverse crop types has led to the development of breeding programs located throughout the world, providing an international network aiding the selection of adapted germplasm and the discovery of useful traits for grain, forage, sugar, and biomass production.

Trajectory of energy Sorghum development

The development of Sorghum hybrids for bioenergy started at Texas A&M University in 2003 (Rooney et al., 2007). Energy Sorghum was selected for high biomass yield potential, a trait associated with long duration of vegetative growth and tall plant stature. A large portion of the Sorghum germplasm collection is tall and late flowering when grown at latitudes and times of the year where plants will be exposed to long days (>12.2 h) during most of the growing season. Sorghum is a short day plant and genotypes that are photoperiod sensitive show delayed flowering when grown in long days (Morgan and Finlayson, 2000). Therefore, the first stage of energy Sorghum development involved the systematic screening of

3484 | Mullet et al.

more than 16 000 photoperiod sensitive accessions of the Sorghum germplasm collection over a period of 6-8 years in a long-day growing environment. This was done in the United States at Texas A&M University (30.6 N latitude) where day lengths increase from 12.4 h following plant emergence in mid to late April to over 14 h in July. When grown in this location, most photoperiod sensitive Sorghum accessions will not initiate flowering until shorter days in September after 175 days of vegetative growth (Fig. 3, right). Grain Sorghum genotypes that are photoperiod insensitive flower in approximately 60-70 days and produce grain that reaches maturity in July in this same location (Fig. 3, left). Most grain Sorghum has been bred to be short in stature through selection for recessive alleles at two or three dwarfing loci (Dw1, Dw2, and Dw3) (Morgan and Finlayson, 2000; Rooney, 2004). This was done to reduce stalk lodging and to make grain Sorghum easier to harvest by machine. In contrast, most Sorghum accessions suitable for energy production in the germplasm collection are tall, often reaching a height of 3-4 metres by the end of the season (Rooney et al., 2007; Olson et al., 2012).

Seed production from plants that are photoperiod sensitive and tall is not feasible when these plants are grown at higher latitudes optimal for hybrid seed production. This obstacle to energy Sorghum seed production was overcome through the discovery of Sorghum genotypes that are photoperiod insensitive and flower in 60-80 days in long or short days, but when crossed, produce photoperiod sensitive hybrids that have long duration vegetative growth phases in long-day environments (Rooney and Aydin, 1999; Rooney et al., 2007). Deployment of alleles of different genes that modulate photoperiod sensitivity in complementary fashion in seed and pollen parents allowed generation of energy Sorghum F1 seed using early flowering inbreds and generated hybrids that flowered late owing to photoperiod sensitivity (Rooney et al., 2007). The molecular genetic basis of photoperiod sensitivity in Sorghum, first characterized by Quinby and his colleagues (Quinby, 1974),

Fig. 3. Grain Sorghum (left) and energy Sorghum (right) after 120 days of growth. The inset is a diagram of the gene regulatory network that modulates flowering time in response to day length in Sorghum.

led to the identification of four loci that modulate flowering time named Ma1-Ma4 (Ma stands for "maturity locus"). Ma3 was subsequently identified as PHYB, indicating that light input through this photoreceptor is required for floral repression in long days as occurs in other plants (Childs et al., 1997). Ma1 was identified as encoding SbPRR37, a repres-sor of flowering in Sorghum, regulated at the level of RNA abundance by the circadian clock and light (Murphy et al., 2011). Ma5 and Ma6, two additional flowering time loci that modulate photoperiod sensitivity, were identified more recently (Rooney and Aydin, 1999). Ma6 encodes SbGhd7, a strong repressor of floral initiation in long days (Murphy et al., 2014). Ma6 acts in an additive fashion with Ma1 to repress the activity/expression of the floral activators SbEhd1, SbCO, and the FT-like genes in Sorghum (Fig. 3) (Murphy et al., 2011; Murphy et al., 2014). Taken together, research into the molecular genetics of flowering time regulation in Sorghum contributed to basic knowledge about this pathway in C4 grass species and accelerated development of energy Sorghum by providing molecular markers for flowering time alleles critical for development of inbred breeding pools useful for production of energy Sorghum hybrids.

Energy Sorghum biomass yield potential

The discovery of complementary alleles for photoperiod sensitivity in Sorghum germplasm useful for breeding and the elucidation of the gene regulatory pathway for photoperiod responsive flowering time aided large-scale generation of energy Sorghum hybrids from diverse germplasm (Rooney et al., 2007). The energy Sorghum hybrids had long vegetative growth duration owing to enhanced photoperiod sensitivity when planted in long-day environments, 3-4 metre length stems, and an overall shoot/canopy architecture that resembles sugarcane and Miscanthus. This suggested that energy Sorghum could serve as a good genetic model for the design of C4 grass bioenergy crops. To explore this possibility further, the growth, development, and genetic potential of this new Sorghum bioenergy crop for biomass accumulation was investigated in multi-year field studies (Olson et al., 2012, 2013). Biomass yields ranging from 15 to >40 dry Mg ha-1 were observed and some of the possible causes of yield variation identified.

Boyer (1982) estimated the maximum genetic potential for maize grain yield using 40 years of USDA data by identifying record yields that occurred in specific locations and years when conditions were optimal for plant growth and grain production. He found average yields of maize and most crops were 10-25% of record yields and identified a number of abiotic and biotic yield constraints that prevented maize and other crops from reaching their genetic yield potential. Maize grain yield in the United States has increased >4-fold since 1930 through optimization of genetics, inputs, and crop management systems (Duvick, 2001). Energy Sorghum is closely related to maize, with similar genetics, a diverse germplasm, and a comparable hybrid breeding system raising the question: could energy Sorghum and other C4 grass energy crops be improved

to a similar extent as maize? To address this question, it would be useful to obtain an estimate of energy Sorghum's biomass yield potential, because this would help predict the likely outcome of optimization of energy crop genetics and management. The analysis of differences between potential and actual yield can also help identify yield constraints that would help focus genetic improvement efforts on the most important factors limiting yield. Energy Sorghum hybrid crop production history in the United States is too limited for the type of large-scale multi-year analysis used by Boyer (1982). However, a multi-year study of Miscanthus x giganteus at three locations in the United States (Illinois) reported the highest yields of ~44-61 dry Mg ha-1 and average yields of ~23 dry Mg ha-1 (Heaton et al., 2008; Arundale et al., 2014). Sugarcane biomass yields in Brazil average 39 dry Mg ha-1 with higher yields in some years and locations (Waclawovsky et al., 2010). The biomass yield for first generation energy Sorghum hybrids grown in small irrigated plots was ~40-50 dry Mg ha-1 compared with ~15-25 dry Mg ha-1 under non-irrigated conditions when grown in larger plots (Olson et al., 2012). The highest annual biomass yields among C4 grasses has been reported for Pennisetum typhoides (80 dry Mg ha-1) (Begg, 1965), Pennisetumpurpureum (Napier grass) (88 dry Mg ha-1) (Long et al., 1992) and Echinochloapoly-stachya (99 dry Mg ha-1) in the Amazon delta (Morrison et al., 2000).

An alternative way to estimate the genetic potential of C4 grasses for biomass accumulation is to utilize information on the efficiency of light energy capture, biochemical properties of C4 photosynthesis (Heaton et al., 2008; Zhu et al., 2010), and plant growth modelling (Stockle and Kemanian, 2009), to estimate biomass yield potential using climate data, growth duration, and leaf area parameters obtained from the energy crop being analysed. This modelling approach was used in combination with multi-year field measurements to assess the capacity of energy Sorghum hybrids for biomass accumulation (Olson et al., 2012). The results were analysed and interpreted using biomass yield equations from Zhu et al. (2010).

Biomass Yield = (S}) * (E,) * (Ec) *(Ep)

This equation shows that biomass yield potential is a function of total incident radiation on the crop (Sj), a factor strongly affected by duration of crop growth, the efficiency of radiation interception by the canopy (Ei), radiation use efficiency (Ec), and the ratio of biomass partitioning to shoots (Ep), the harvested biomass used for bioenergy production. The biomass yield potential predicted from field trials for energy Sorghum grown in College Station, Texas (USA) based on climate data, measured leaf areas, and growth duration was between 55-60 dry Mg ha-1 (Fig. 4) (Olson et al., 2012). Data on biomass accumulation in well-managed plots of energy Sorghum hybrids grown with sufficient nutrients under fully irrigated conditions (triangles) ranged from 40-50 dry Mg ha-1 and tracked predicted theoretical yield for C4 grasses until late in the growing season. The time courses show a lag in biomass accumulation during the initial ~6 weeks after plant emergence when canopy development is occurring and

A TX08001 yield, irrigated

i 4000

' Modelled, no temperature limitation

Modelled, with temperature limitation a

210 Calendar day

Fig. 4. Theoretical and measured biomass yield of energy Sorghum. Modelled cumulative biomass yield accumulation of fully irrigated TX08001, using leaf area data from field measurements, with and without temperature constraint (light and dark blue lines, respectively). Time course of biomass accumulation of TX08001 in 2008 under field conditions with irrigation (red triangles) and without irrigation (blue bars) in field plots harvested in early September or October. From Olson et al. (2012). High biomass yield energy Sorghum: Developing a genetic model for C4 grass bioenergy crops. Biofuels, Bioproducts and Biorefining 6, 640-655.

light interception efficiency (E) is low. Rates of biomass accumulation increase when canopy closure occurs and Sorghum begins rapid stem growth. A high rate of biomass accumulation continued under irrigated conditions until the onset of cooler weather, shorter days, and flowering in October.

Energy Sorghum yield improvement targets

The biomass yields reported for energy Sorghum and other C4 grass energy crops grown without irrigation average ~25-33% of their estimated maximum biomass yield under optimal conditions. Although additional multi-year/location studies are needed to refine this estimate, the observed yield gap indicates that there is a significant opportunity to increase the productivity of C4 grass energy crops through a combination of improved genetics and management. Genetic improvements that increase season length (Sj), optimize canopy architecture and photosynthesis (Ei;Ec), improve the acquisition/use of water and nutrients (Ec), and optimize biomass partitioning (Ep) could contribute to increases in biomass yield.

Higher potential biomass yields for C4 grasses will occur in locations where long growing seasons are possible, assuming other abiotic or biotic factors are not limiting plant growth during the season. For example, at Sorghum's maximal biomass accumulation rate of ~0.35 dry Mg ha-1 per day, Napier grass would accumulate its record yield of ~88 dry Mg ha-1 in 251 days in San Salvador (13 degrees N latitude), where very long growing seasons are possible (Long et al., 1992). In contrast, if a C4 grass energy crop with similar biomass accumulation potential was grown in the northern United States where growing seasons are short, there could be less than 60 days for maximal rates of biomass accumulation in addition to the time required for canopy development, corresponding to a theoretical maximum of ~21 dry Mg ha-1 biomass yield. In locations with long growing seasons, energy

3486 | Mullet et al.

crops that have long duration vegetative growth phases have higher biomass yield potential compared with early flowering genotypes, one reason why energy Sorghum accumulates more biomass than grain Sorghum (Olson et al., 2012). C4 grasses, such as Miscanthus, with good cold tolerance increase season length by initiating growth earlier in the spring and maintaining higher photosynthetic activity longer in the fall, compared with less cold-tolerant C4 grasses such as maize (Wang et al., 2008; Dohleman and Long, 2009; Friesen et al., 2014). If the cold tolerance of bioenergy crops can be improved without fitness tradeoffs (Agren et al., 2013), such as reducing heat tolerance, then growing season length and biomass yield potential of less cold tolerant energy crops such as Sorghum and sugarcane could be increased.

The amount and frequency of rainfall will significantly impact biomass production from energy crops grown without irrigation in many locations and years by reducing photosyn-thetic efficiency (Ec) and in extreme cases leading to crop loss. To begin to assess the relative importance of this factor, energy Sorghum hybrids were grown with full irrigation and without irrigation where summer rainfall was intermittent and insufficient for maximal biomass accumulation (Fig. 4, green bars). Total biomass accumulation by energy Sorghum hybrids was reduced by 50% or more compared with irrigated controls, depending on the timing, duration, and extent of water deficit. Modelling was used to calculate cumulative seasonal evapotranspiration from energy Sorghum canopies under water sufficient conditions using climate data from the two years of the study (Olson et al., 2012). The predicted ~950-1150 mm of water required per season for maximal biomass accumulation was considerably more than the cumulative rainfall during the 2008 growing season (~550 mm). Moreover, rainfall occurred in an irregular pattern with periods of up to 4 weeks without rain. Energy Sorghum is very drought resilient, especially in the vegetative phase, and some genotypes can survive long periods without rainfall. The plants subjected to water deficit initially slow and then stop growing with minimal visual symptoms, although leaf temperatures rise owing to stomatal closure. Increasing severity and duration of water deficit causes leaves to become more erect and rolled reducing heat load and water loss, followed by lower leaf senescence. However, energy Sorghum was very drought resilient and even after extended periods without water, when rainfall occurred, plants produced new leaves at the top of the canopy and began accumulating biomass. C4 grass energy crops that will be grown in agro-ecological regions prone to annual water deficits will benefit from improved drought resilience, soil water extraction capacity, and efficient use of water for biomass generation. Sorghum is well adapted to drought prone environments because many accessions originate from low rainfall regions of Africa. Prior studies on grain Sorghum showed that this C4 grass has a suite of traits that aid adaptation to water limited environments, including waxy leaf surfaces, water deficit-induced leaf rolling, drought induced lower leaf senescence, stay-green drought tolerance (Rosenow and Clark, 1995; Borrell and Hammer, 2000; Blum, 2005; Borrell et al., 2006), and a diversity of root system architectures (Mace and Jordan, 2011)

that help this species access and efficiently utilize available water resources. Sorghum germplasm also varies in transpiration and water use efficiency (Hammer et al., 1997; Balota et al., 2008; Narayanan et al., 2013). Sorghum's extensive variation in traits useful for adaptation to water limited environments indicates there is significant opportunity for improving energy Sorghum biomass yield in these environments.

C4 grasses such as Sorghum, maize, Miscanthus and sugarcane have some of the highest rates of light saturated photosynthesis consistent with high biomass yields (Zhu et al., 2010). Sorghum genotypes have high rates of CO2 fixation at high light intensity under field conditions (44-55 ^mol CO2 m-2 s-1) (Balota et al., 2008). High midday photosynthetic rates observed in Sorghum are similar to maize but higher than Miscanthus (Dohleman and Long, 2009). Although improving energy Sorghum photosynthesis would be useful, optimizing canopy architecture for more optimal light interception seems a tractable way to increase yield in the near term. Current energy Sorghum planting densities (~132 000 plants ha-1) are higher than maize, resulting in a leaf area index (LAI>7) well in excess of what is necessary for efficient light energy capture (Olson et al., 2012). As a consequence, most light interception by energy Sorghum plants occurs at the top of the canopy. Photosynthesis is less efficient owing to light saturation at light intensities 50-100% of full sunlight (Zhu et al., 2010). Therefore, it is likely that decreasing energy Sorghum LAI through genetics and crop management, in combination with changes in leaf angle could result in improved biomass yield. It was theorized that for crops with LAI>2, as is the case with energy Sorghum (LAI ~7-8), an ideal canopy will be comprised of leaves with gradually decreasing leaf angles (Long et al., 2006; Zhu et al., 2010).

Grain yield has been increased by selection for greater partitioning of total biomass to the grain. For energy crops, shoot biomass is harvested for biofuels production, therefore biomass partitioning to shoots (vs roots) ( Ep ) can affect biomass yield. Roots account for ~20% of energy Sorghum biomass (Olson et al., 2012), much less than switchgrass (up to 50%) (Frank et al., 2004) and Miscanthus (up to 38%) (Clifton-Brown and Lewandowski (2000). The greater allocation of biomass to the root system in perennials is consistent with the need for these plants to store sufficient carbon and nitrogen each fall for regrowth in the spring. In contrast, annual energy Sorghum hybrids only require sufficient root biomass for water and nutrient extraction and to minimize lodging. In this regard, energy Sorghum is not an ideal model for the study of perennial C4 grass root systems. Moreover, selection for reduced root biomass in order to increase shoot biomass yield is likely to reduce Sorghum's capacity to extract water from the soil profile and to increase lodging.

Modification of biomass composition provides an alternative way to increase biofuel yield per hectare by partitioning fixed carbon into structures and storage compounds more amenable to efficient conversion to biofuels or biopower. The range of Sorghum biomass composition is substantial (Stefaniak et al., 2012). For example, a comparison of the composition of stems from the genotype Della at grain maturity (including panicles) and energy Sorghum hybrid stems at

□ Sucrose

■ Starch (Total]

■ Water Extract

■ Protein (Total]

□ Other

□ EtOH Extract

■ Ash (Total]

□ Arabinan

■ Galactan

□ Xylan

■ Lignin

■ Cellulose

Fig. 5. Composition of sweet Sorghum and energy Sorghum stems. Stems and panicles of the sweet Sorghum Della and energy Sorghum hybrid TX08001 were collected at harvest, dried, ground and composition analysis was performed using NIR using methods found in Stefaniak et al. (2012).

harvest (minimal seed production) is shown in Fig. 5. Della is a typical sweet Sorghum selected to accumulate up to 20% sucrose/volume in its stem at grain maturity. Overall, starch (primarily grain), sucrose and monosaccharides in the water extractives fraction account for ~50% of total biomass with protein, ash, and cell wall components (cellulose, lignin, xylan) accounting for most of the remaining biomass. In contrast, the high biomass energy Sorghum hybrid TX08001 accumulates much lower levels of sucrose, monosaccharides and starch (~20-25%), and protein, while accumulating proportionally more cell wall materials and ash (~60%). This indicates that energy Sorghum and other C4 grass energy crops can be modified significantly in terms of the ratio of non-structural to structural carbohydrate composition through breeding to improve conversion efficiency (Murray et al., 2008; Stefaniak et al., 2012).

Future prospects for energy Sorghum and C4 grass energy crop development

The goal of research on energy crops is to develop high yielding plants that can sustainably produce composition-ally optimized biomass at a cost that enables the generation of economically competitive biofuels. Over the past decade, substantial progress has occurred in our understanding of how to improve the design of energy crops to achieve this goal. In the process, energy Sorghum has emerged as an integrated genetic-genomic-breeding platform for C4 grass energy crop design. Sorghum's role as a genetic model is based on its inherent genetic and genomic attributes and the opportunity to test a wide range of genetic designs in multiple locations on an annual basis. The integration of Sorghum's genetic-genomics discovery platform with a hybrid breeding-field

testing system enables rapid cycles of discovery and selection with high translation efficiency to field-tested hybrids, thus speeding up the genetic improvement process. We project that improvement in biomass yield and composition/conversion efficiency will reduce the price of biofuels derived from energy Sorghum biomass to a level that is cost competitive with oil-based transportation fuels. Development of dual use energy/forage Sorghum crops and the production of biofuels and animal forage directly or indirectly from the conversion process will help mitigate indirect impacts on food security. Furthermore, production of a portion of the biofuels feedstock from an annual hybrid energy crop such as Sorghum would enable yearly rebalancing of food, feed and biofuels production depending on need and price signals. The energy crop design principles discovered and tested using energy Sorghum could be transferred to perennial grass crops in several ways including parallel selection for improved traits, gene modification technology, or through wide hybridization of Sorghum to perennial C4 grasses.

Acknowledgements

Research carried out in the laboratories of John Mullet and William Rooney was funded in part by grants from the USDA-DOE Genomics Feedstock Program, the Great Lakes Bioenergy Research Center, Texas A&M AgriLife Research, DOE's Office of Science and industry partners.

References

Agren J, Oakley CG, McKay JK, Lovell JT, Schemske DW. 2013. Genetic mapping of adaptation reveals fitness tradeoffs in Arabidopsis thaliana. Proceedings of the National Academy of Sciences, USA doi: 10.1073/pnas.1316773110

Arundale RA, Dohlman FG, Heaton EA, Mcgrath JM, Voigt TB, Long

SP. 2014. Yields of Miscanthus x giganteus and Panicum virgatum decline with stand age in the Midwestern USA. GCB Bioenergy 6, 1-13.

Balota M, Payne WA, Rooney W, and Rosenow D. 2008. Gas exchange and transpiration ratio in Sorghum. Crop Science 48, 2361-2371.

Begg J. 1965. Growth and development of a crop of bulrush millet (Pennisetum typhoides S+H). Journal of Agricultural Science 65, 341-349.

Bernardo R, Yu J. 2007. Prospects for genome-wide selection for quantitative traits in maize. Crop Science 47, 1082-1090.

Billot C, Ramu P, Bouchet S et al. 2013. Massive Sorghum collection genotyped with SSR markers to enhance use of global genetic resources. PLoS One 8, 4-20.

Blum A. 2005. Drought resistance, water-use efficiency, and yield potential—are they compatible, dissonant, or mutually exclusive? Australian Journal of Agricultural Research 56, 1159-1168.

Borrell A, Hammer G. 2000. Nitrogen dynamics and the physiological basis of stay-green in Sorghum. Crop Science 40, 1295-1307.

Borrell A, Jordan D, Mullet J, Henzell B, Hammer G. 2006. Drought adaptation in Sorghum. In: Ribaut J-M, ed. Drought Adaptation in Cereals . New York: Food Products Press, An Imprint of The Haworth Press, 335-399.

Bouton JH. 2007. Molecular breeding of switchgrass for use as a biofuel crop. Current Opinion in Genetics and Development 17, 553-558.

Boyer JS. 1982. Plant productivity and environment. Science 218, 443-448.

Buchanan CD, Lim S, Salzman RA et al. 2005. Sorghum bicolor's transcriptome response to dehydration, high salinity and ABA. Plant Molecular Biology 58, 699-720.

Byrt CS, Grof CPL, Furbank RT. 2011. C4 plants as biofuel feedstocks: optimising biomass production and feedstock quality from a lignocellulosic perspective. Journal of Integrative Plant Biology 53, 120-135.

3488 | Mullet et al.

Calvino M, Bruggman R, Messing J. 2008. Screen of genes linked to high-sugar content in stems by comparative genomics. Rice 1, 166-176.

Calvino M, Messing J. 2012. Sweet Sorghum as a model system for bioenergy crops. Current Opinion in Biotechnology 23, 323-329.

Carroll A, Somerville C. 2009. Cellulosic biofuels. Annual Review of Plant Biology 60, 165-182.

Catchen JM, Amores A, Hohenlohe P, Cresko W, Postlethwait JH.

2011. Stacks: Building and genotyping loci de novo from short-read sequences. G3: Genes, Genomes, Genetics 3, 171-182.

Chen K, Gao C. 2013. TALENs: Customizable molecular DNA scissors for genome engineering of plants. Journal of Genetics and Genomics 40, 271-279.

Childs K, Miller F, Cordonnier-Pratt M, Pratt L, Morgan P, Mullet

J. 1997. The Sorghum photoperiod sensitivity gene, Ma3, encodes a phytochrome B. Plant Physiology 113, 611-619.

Clifton-Brown JC, Lewandowski I. 2000. Water use efficiency and biomass partitioning of three different Miscanthus genotypes with limited and unlimited water supply. Annals of Botany 86, 191-200.

Davis SC, Parton WJ, Dohleman FG, Smith CM, Del Grosso S, Kent AD, DeLucia EH. 2010. Comparative bogeochemical cycles of bioenergy crops reveal nitrogen-fixation and low greenhouse gas emissions in a Miscanthus x giganteus agro-ecosystem. Ecosystems 13, 144-156.

Doebley JF, Gaut BS, Smith BD. 2006. The molecular genetics of crop domestication. Cell 127, 1309-1321.

Dohleman FG, Long SP. 2009. More productive than maize in the midwest: How does Miscanthus do it? Plant Physiology 150, 2104-2115.

Dugas DV, Monaco MK, Olsen A, Klein RR, Kumari S, Ware D, Klein PE. 2011. Functional annotation of the transcriptome of Sorghum bicolor in response to osmotic stress and abscisic acid. BMC Genomics 12, 514.

Duvick D. 2001. Biotechnology in the 1930s: the development of hybrid maize. Nature Reviews 2, 69-74.

Elshire RJ, Glaubitz JC, Sun Q, Poland JA, Kawamoto K, Buckler ES, Mitchell SE. 2011. A robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PloS One 6, e19379.

Evans J, McCormick RF, Morishige D, Olson SN, Weers B, Hilley J, Klein P, Rooney W, Mullet J. 2013. Extensive variation in the density and distribution of DNA polymorphism in Sorghum genomes. PloS One 8, e79192.

Foley JA, Ramankutty N, Braumann KA et al. 2011. Solutions for a cultivated planet. Nature 478, 337-342.

Frank AB, Berdahl JD, Hanson JD, Liebig MA, Johnson HA. 2004. Biomass and carbon partitioning in switchgrass. Crop Science 44, 1391.

Friesen PC, Peixoto MM, Busch FA, Johnson DC, Sage RF. 2014. Chilling and frost tolerance in Miscanthus and Saccharum genotypes bred for cool temperate climates. Journal of Experimental Botany 65, 3749-3758.

Hammer GL, Farquhar GD, Broad IJ. 1997. On the extent of genetic variation for transpiration efficiency in Sorghum. Australian Journal of Agricultural Research 48, 649-655.

Hammer GL, Dong Z, McLean G, Doherty A, Messina C, Schussler J, Zinselmeier C, Paszkiewicz S, Cooper M. 2010a. Can changes in canopy and/or root system architecture explain historical maize yield trends in the U.S. corn belt? Crop Science 49, 299-312.

Hammer GL, Oosterom EV, McLean G, Chapman SC, Broad I, Harland P, Muchow RC. 2010b. Adapting APSIM to model the physiology and genetics of complex adaptive traits in field crops. Journal of Experimental Botany 61, 2185-2202.

Hart G, Schertz K, Peng Y, Syed N. 2001. Genetic mapping of Sorghum bicolor (L.) Moench QTLs that control variation in tillering and other morphological characters. Theoretical and Applied Genetics 103, 1232-1242.

Heaton E, Voigt T, Long S. 2004. A quantitative review comparing the yields of two candidate C4 perennial biomass crops in relation to nitrogen, temperature and water. Biomass & Bioenergy 27, 21-30.

Heaton EA, Dohleman FG, Long SP. 2008. Meeting US biofuel goals with less land: the potential of Miscanthus. Global Change Biology 14, 2000-2014.

Heffner, EL, Sorrells, ME, Jannink, J-L. 2009. Genomic selection for crop improvement. Crop Science 49, 1-12.

Hillier J, Whittaker C, Dailey G, Aylott M, Casella E, Richter GM, Fiche A, Murphy R, Taylor G, Smith P. 2009. Greenhouse gas emissions from four bioenergy crops in England and Wales: Integrating spatial estimates of yield and soil carbon balance in life cycle analyses. GCB Bioenergy 1, 267-281

Hodnett GL, Hale AL, Packer DJ, Stelly DM, Da Silva J, Rooney

WL. 2010. Elimination of a reproductive barrier facilitates intergeneric ybridization of Sorghum bicolor and Saccharum. Crop Science 50, 1188-1195.

Jordan DR, Mace ES, Cruickshank AW, Hunt CH, Henzell RG. 2011. Exploring and exploiting genetic variation from unadapted Sorghum germplasm in a breeding program. Crop Science 51, 1444-1457.

Jorgensen U. 2011. Benefits versus risks of growing biofuel crops: the case of Miscanthus. Current Opinion in Environmental Sustainability 3, 24-30.

Kellogg EA. 2001. Evolutionary history of the grasses. Plant Physiology 125, 1198-1205.

Kim JS, Islam-Faridi MN, Klein PE, Stelly DM, Price HJ, Klein RR, Mullet JE. 2005. Comprehensive molecular cytogenetic analysis of Sorghum genome architecture: Distribution of euchromatin, heterochromatin, genes and recombination in comparison to rice. Genetics 171, 1963-1976.

Kimber C. 2000. Origins of domesticated Sorghum and its early diffusion to India and China. In: Smith C, Fredriksen R, eds. Sorghum: Origin, History, Technology, and Production . New York: John Wiley Sons, Inc., 3-98.

Lawrence CJ, Walbot V. 2007. Translational genomics for bioenergy production from fuelstock grasses: maize as the model species. The Plant Cell 19, 2091-2094.

Li P, Brutnell T. 2011. Setaria viridis and Setaria italic, model genetic systems for the Panicoid grasses. Journal of Experimental Botany 62, 3031-3037.

Li H, Durbin R. 2010. Fast and accurate long-read alignment with Burrows-Wheeler Transform. Bioinformatics 5, 589-595.

Lin Z, Li X, Shannon LM et al. 2012. Parallel domestication of the Shatteringl genes in cereals. Nature Genetics 44, 720-724.

Liu G, Godwin ID. 2012. Highly efficient Sorghum transformation. Plant Cell Reports 31, 999-1007.

Long SP, Jones MB, Roberts MJ. 1992. Primary productivity of grass ecosystems of the tropics and sub-tropics . London: Chapman and Hall.

Long SP, Zhu X-G, Naidu S, Ort DR. 2006. Can improvement in photosynthesis increase crop yields? Plant, Cell and Environment 29, 315-330.

Mace ES, Tai S, Gilding EK et al. 2013. Whole-genome sequencing reveals untapped genetic potential in Africa's indigenous cereal crop Sorghum. Nature communications 4, doi: 10.1038/ncomms3320.

Mace ES, Jordan DR. 2011. Integrating Sorghum whole genome sequence information with a compendium of Sorghum QTL studies reveals uneven distribution of QTL and of gene-rich regions with significant implications for crop improvement. Theoretical and Applied Genetics 123, 169-191.

Mace ES, Xia L, Jordan DR, Halloran K, Parh DK, Huttner E, Wenzl P, Kilian A. 2008. DArT markers: diversity analysis and mapping in Sorghum bicolor. BMC Genomics 9, 26-37.

Menz MA, Klein RR, Mullet JE, Obert JA, Unruh NC, Klein PE. 2002. A high-density genetic map of Sorghum bicolor (L.) Moench based on 2926 AFLP, RFLP and SSR markers. Plant Molecular Biology 48, 483-499.

Monk RL, Miller FR, McBee GG. 1984. Sorghum improvement for energy production. Biomass 6, 145-153.

Morgan PW, Finlayson SA. 2000. Physiology and genetics of maturity and height. In: Smith C, Fredriksen R, eds. Sorghum: Origin, History, Technology, and Production . New York: John Wiley Sons, Inc., 227-259.

Morishige DT, Klein PE, Hilley JL, Sahraeian SME, Sharma A, and Mullet JE. 2013. Digital genotyping of Sorghum—A diverse species with a repeat-rich genome. BMC Genomics 14, 448.

Morris GP, Ramu P, Deshpande SP et al. 2013. Population genomic and genome-wide association studies of agroclimatic traits in Sorghum. Proceedings of the National Academy of Sciences, USA 110, 453-458.

Morrison JI, Piedade MT, Muller E, Long SP, Junk WJ, Jones

MB. 2000. Very high productivity of the C4 aquatic grass Echinochloa polystachya in the Amazonian floodplain confirmed by net ecosystem CO2 flux measurements. Oecologia 125, 400-411.

Murphy RL, Klein RR, Morishige DT, Brady JA, Rooney WL, Miller FR, Dugas DV, Klein PE, Mullet JE. 2011. Coincident light and clock regulation of pseudoresponse regulator protein 37 (PRR37) controls photoperiodic flowering in Sorghum. Proceedings of the National Academy of Sciences, USA 108, 16469-16474.

Murphy RL, Morishige DT, Brady JA, Rooney WL, Yang S, Klein PE, Mullet JE. 2014. Ghd7 (Ma6) represses Sorghum flowering in long days: Ghd7 alleles enhance biomass accumulation and grain production. The Plant Genome doi: 10.3835/plantgenome2013.11.0040.

Murray SC, Sharma A, Rooney WL, Klein PE, Mullet JE, Mitchell SE, Kresovich S. 2008. Genetic improvement of Sorghum as a biofuel feedstock: I. QTL for stem sugar and grain nonstructural carbohydrates. Crop Science 48, 2165-2179.

Narayanan S, Aiken RM, Prasad PVV, Xin Z, Yu J. 2013. Water and radiation use efficiencies of Sorghum. Agronomy Journal 105, 649-656.

National Corn Growers Association. 2012. Corn—Rooted in Human History. World of Corn , 1-12.

Olson S, Ritter K, Rooney W, Kemanian A, McCarl B, Zhang Y, Hall S, Packer D, Mullet J. 2012. High biomass yield energy Sorghum: developing a genetic model for C4 grass bioenergy crops. Biofuels, Bioproducts and Biorefining 6, 640-655.

Olson SN, Ritter K, Medley J, Wilson T, Rooney WL, Mullet JE.

2013. Energy Sorghum hybrids: Functional dynamics of high nitrogen use efficiency. Biomass and Bioenergy 56, 307-316.

Paradis E, Claude J, Strimmer K. 2004. APE: Analyses of phylogenetics and evolution in R language. Bioinformatics 20, 289-290.

Paterson AH, Bowers JE, Bruggmann R et al. 2009. The Sorghum bicolor genome and the diversification of grasses. Nature 457, 551-556.

Pauly M, Keegstra K. 2008. Cell-wall carbohydrates and their modification as a resource for biofuels. The Plant Journal 54, 559-568.

Penning BW, Hunter CT, Tayengwa R et al. 2009. Genetic resources for maize cell wall biology. Plant Physiology 151, 1703-1728.

Perlack R, Wright L, Turhollow A, Graham R, Stokes B, Erbach D.

2005. Biomass as feedstock for a bioenergy and bioproducts industry: the technical feasibility of a billion-ton annual supply . Tech. Rep. U.S. Department of Agriculture and U.S. Department of Energy. Oak Ridge National Laboratory. D0E/G0-102005-2135 and 0RNL/TM-2005/66.

Purcell S, Neale B, Todd-Brown K et al. 2007. PLINK: A tool set for whole-genome association and population-based linkage analyses. The American Journal of Human Genetics 3, 559-575.

Quinby JR. 1974. Sorghum Improvement and the Genetics of Growth. Texas: Texas A&M University Press.

Rahall N. 2007. H.R. 6 (110th): Energy Independence and Security Act of 2007.

Ranum P, Pena-Ross JP, Garcia-Casal MN. 2014. Global maize production, utilization and consumption. Annals of the New York Academy of Sciences 1312, 105-112.

Rayburn AL, Crawford J, Rayburn CM, Juvik JA. 2009. Genome size of three Miscanthus species. Plant Molecular Biology Reports 27, 184-188.

Rooney WL, Blumenthal J, Bean B, Mullet JE. 2007. Designing Sorghum as a dedicated bioenergy feedstock. Biofuels, Bioproducts and Biorefining 1, 147-157.

Rooney W. 2004. Sorghum improvement—Integrating traditional and new technology to produce improved genotypes. Advances in Agronomy 83, 37-109.

Rooney W, Aydin S. 1999. Genetic control of a photoperiod-sensitive response in Sorghum bicolor (L.) Moench. Crop Science 39, 397-400.

Rosenow DT, Clark LE. 1995. Drought and lodging resistance for a quality Sorghum crop. In: Proceedings of the 50th Annual Corn and Sorghum Industry Research Conference . Chicago, IL, 82-97.

Sage RF, Xhu X-G. 2011. Exploiting the engine of C4 photosynthesis. Journal of Experimental Botany 62, 2989-3000.

Schmer MR, Vogel KP, Mitchell RB, Perrin RK. 2008. Net energy of cellulosic ethanol from switchgrass. Proceedings of the National Academy of Sciences, USA 105, 464-469.

Sharma MK, Sharma R, Cao P, Jenkins J, Bartley LE, Qualls M, Grimwood J, Schmutz J, Rokhsar D, Ronald PC. 2012. A genome-wide survey of switchgrass genome structure and organization. PLOS One 7, 4-17.

Smith CW, Frederiksen RA. 2000. History of cultivar development in the United States: From "Memoirs of A.B. Maunder-Sorghum Breeder". In: Smith CW, Frederiksen RA, eds. Sorghum: Origin, History, Technology, and Production. New York: John Wiley Sons, Inc., 191-223.

Somerville C, Youngs H, Taylor C, Davis SC, Long SP. 2010. Feedstocks for lignocellulosic biofuels. Science 329, 790-792.

Stefaniak TR, Dahlberg JA, Bean BW, Dighe N, Wolfrum EJ, Rooney

WL. 2012. Variation in biomass composition components among forage, biomass, Sorghum-sudangrass, and sweet Sorghum types. Crop Science 52, 1949-1954.

Stöckle CO, Kemanian A. 2009. Crop radiation capture and use efficiency: A framework for crop growth analysis. In: Sadras V, Calderini D, eds. Applied Crop Physiology: Applications for Genetic Improvement and Agronomy . Oxford: Acadmeic Press, 145-170. Tange O. 2011. GNU Parallel—The command-line power tool. The USENIX Magazine 36, 42-47.

The International Brachypodium Initiative. 2010. Genome sequencing and analysis of the model grass Brachypodium distachyon. Nature 463, 763-768.

U.S. Department of Energy. 2011. U.S. Billion-Ton Update: Biomass Supply for a Bioenergy and Bioproducts Industry. Perlack, R.D. and B.J.Stokes (Leads), 0RNL/TM-2011/224. Oak Ridge National Laboratory, Oak Ridge, TN.

Vermerris W. 2011. Survey of genomics approaches to improve bioenergy traits in maize, Sorghum and sugarcane. Journal of Integrative Plant Biology 53, 105-119.

Weijde TVD, Kamei CLA, Torres AF, Vermerris W, Dolstra O, Visser RG, Trindale LM. 2013. The potential of C4 grasses for cellulosic biofuel production. Frontiers in Plant Science 4, 1-18.

Waclawovsky AJ, Sato PM, Lembke CG, Moore PH, Souza GM.

2010. Sugarcane for bioenergy production: an assessment of yield and regulation of sucrose content. Plant Biotechnology Journal 8, 263-276.

Wang D, Portis AR, Moose SP, Long SP. 2008. Cool C4 photosynthesis: pyruvate Pi dikinase expression and activity corresponds to the exceptional cold tolerance of carbon assimilation in Miscanthus x giganteus. Plant Physiology 148, 557-567.

Xin Z, Wang ML, Barkley NA, Burow G, Franks C, Pederson G, Burke J. 2008. Applying genotyping (TILLING) and phenotyping analyses to elucidate gene function in a chemically induced Sorghum mutant population. BMC Plant Biology 8, 103.

Zheng L-Y, Guo X-S, He B, Sun L-J, Peng Y, Dong S-S, Liu T-F, Jiang S, Ramachandran S, Liu C-M, Jing H-C. 2011. Genome-wide patterns of genetic variation in sweet and grain Sorghum (Sorghum bicolor). Genome Biology 12, R114.

Zhu X, Long SP, Ort DR. 2010. Improving photosynthetic efficiency for greater yield. Annual Review of Plant Biology 61, 235-261.