Bringing trees into the fuel line


If there is to be a Holy Grail for the twenty-first century, the development of a true ‘bioeconomy’ to replace our current fossil fuel-based economy would seem to be a good candidate. Just as for the medieval seekers, the way to this Grail may be arduous and confused but the ultimate objective is unarguably one of immense importance. The 26th New Phytologist Symposium – Bioenergy trees – in May 2011 ( brought together a wide range of expertise to explore the current status of one potential strategy for reaching that goal: the large-scale use of forest biomass as a carbon and energy feedstock (Herr, 2011). The basic concept is simple; rather than relying upon petroleum, natural gas and coal – the products of solar energy captured by photosynthesizing plants hundreds of millions of years ago – we would harvest more recently synthesized photosynthetic products and convert these to industrially useful materials, most urgently to liquid fuels.

‘This is a more visionary, and hence long-term, approach whose ultimate goal is matching optimized feedstock traits (phenotypes) with low-input processing technology.’

Since a well-developed technology platform exists for conversion of glucose to ethanol or other alcohols, which can meet many fuel and chemical synthesis needs, the focus over the past few decades has been on maximizing the harvestable yield of soluble sugars for fermentation, most notably based on sucrose from sugar cane or sugar beets, or starch hydrolysates from grains such as maize. As the limits on proven fossil fuel reserves have come into sharper focus more recently, there has been a massive increase in the industrial production of such fermentation ethanol, or ‘bioethanol’ (International Energy Agency, 2011). However, it is clear that even with dedication of substantially more of the agricultural landscape to production of these bioethanol, or ‘first generation biofuel’, crops, the traditional processing of starch/sucrose-accumulating plants can potentially meet only a small fraction of the global fuel demand (International Energy Agency, 2011).

The focus, instead, is now on mobilizing other sugar sources, most notably the massive amounts of glucose and related monosaccharides that make up polysaccharide-based plant cell walls. Compared to the carbon reserves represented by soluble sugars or starches, cell wall carbohydrates could potentially provide a much larger feedstock for downstream processing. Unfortunately, they are also much less ‘user friendly’. Mature plants build thick, multi-laminate walls consisting of a structurally and chemically complex mixture of polysaccharides whose character varies from species to species, and is further influenced by both the growth environment and developmental programming (Albersheim et al., 2010; Nieminen et al., pp. 46–53). The strength and durability of these so-called ‘secondary walls’ is typically further enhanced by impregnation with the phenolic polymer, lignin, whose presence makes it particularly difficult for hydrophilic reagents to gain access to the embedded glycan chains (Vanholme et al., 2010). The complexity and chemical intransigence of secondary walls thus make it challenging to efficiently convert lignocellulosic biomass into fermentable sugars using current technology, which relies upon relatively costly inputs – heat and aggressive chemistry. Nevertheless, solving this problem would represent a very significant step along the path to a true bioeconomy (Carroll & Somerville, 2009).

To this end, there are many strategies being pursued, but they fall into two broad categories. The first is to develop new or improved technologies for more efficiently processing existing feedstocks. These range from large-volume crop residues such as corn stover, wheat straw or sugarcane bagasse, to the hundreds of millions of tonnes of forestry trimmings generated annually in logging operations. While this is primarily a short-term, opportunistic strategy aimed at exploiting biomass resources that are currently discarded or under-utilized, it can also be extended to purpose-grown biomass crops such as switchgrass or short-rotation poplar.

The second strategy, and the focus of the Bioenergy trees symposium in Nancy (France), involves creation of novel biomass plant genotypes, either through genetic selection or genetic modification. This is a more visionary, and hence long-term, approach whose ultimate goal is matching optimized feedstock traits (phenotypes) with low-input processing technology. Unlike the first strategy, where research can often be advanced more empirically using existing biomaterials, successful development of tailor-made plant genotypes depends heavily on finding answers to a host of intertwined biological questions, as was highlighted by several speakers. What is the biochemical basis of a particular biomass trait? What metabolic systems are engaged in supporting it? Which genes control the relevant pathways and how are those genes themselves regulated? (Behnke et al., pp. 70–82; Mizrachi et al., pp. 54–62; Nieminen et al., pp. 46–53).

Participants in the Bioenergy trees meeting addressed a range of these biological issues. A central parameter in any biomass-based economic model is always the final yield of harvestable material, which can often be a function of stem architecture. Moreno-Cortés et al. (pp. 83–90) show that the ectopic expression of a Castanea RAV1 gene in transgenic poplar results in enhanced development of sylleptic branching in the juvenile trees, a pattern that has been correlated with increased branching and greater biomass accumulation as trees mature. Secondary cell wall lignification has a huge impact on the conversion of woody feedstocks to bioethanol, as reviewed at the meeting by Shawn Mansfield (University of British Columbia, Canada; Mansfield et al., pp. 91–101). In recent years, considerable attention has been focused on down-regulating specific steps in lignin biosynthesis, and Wout Boerjan (Ghent University, Belgium) discussed the results of specifically suppressing the CCR gene in poplar. Strong CCR suppression resulted in a dwarf phenotype, whereas more moderate levels allowed normal growth and improved saccharification values. Li et al. (pp. 102–115) reported that altering the expression of an Arabidopsis homeobox transcription factor, KNAT7, can affect the degree to which stem fibre secondary walls become lignified, as well as their thickness. An intriguing alternative to directly manipulating native lignin accumulation was discussed in Nancy by John Ralph (University of Wisconsin, USA), who is ectopically expressing an enzyme that incorporates relatively labile ferulate-coniferyl alcohol ester linkages into the lignin polymer backbone.

However, robust enhancement of complex quantitative traits such as yield or wood quality can seldom be accomplished through the direct manipulation of expression of single genes. In addition, such an approach typically involves transgenic technology, which is still viewed negatively in enough constituencies to make commercialization of the products challenging (Strauss et al., 2009). However, it is well established that plants exhibiting some of the trait changes sought in candidate lignocellulosic biomass crops, such as reduced lignin content or altered wall polysaccharide composition, already exist in nature and are under genetic control (Zobel & Jett, 1995; Kibblewhite, 1999; Sewell et al., 2000). While deciphering the genetic blueprint that underlies these desirable phenotypes has been largely beyond the reach of classical breeding in tree species, the plummeting cost of whole genome re-sequencing has now made it realistic to search through the individual genomes of entire populations of trees for specific allelic combinations associated with any trait of interest. Recent progress in fine-grained analysis of poplar genomes was reviewed in Nancy by Jerry Tuskan (Oak Ridge National Laboratory, TN, USA), Steve Strauss (Oregon State University, USA) and Mathias Kirst (University of Florida, USA). Collaboration between POPCAN (a poplar biofuel genomics project led by Carl Douglas and Shawn Mansfield, University of British Columbia, Vancouver, Canada) and the Tuskan group confirmed that both lignin content and syringyl to guaiacyl (S/G) ratio varied considerably across a population of c. 1100 Populus trichocarpa individuals, and this variation is now being mapped onto patterns of small feature polymorphisms in the corresponding poplar genomes at both Oak Ridge and Vancouver. The power of these global polymorphism association studies (association genomics) is expected to steadily increase as additional datasets are generated from expanded plant populations that display a wider range of phenotypic diversity, as Resende et al. (pp. 116–128) point out in their analysis of the effectiveness of such a strategy for predicting wood quality traits in Eucalyptus. As these global approaches are further refined and integrated into modern tree breeding programmes, genomic selection combined with new high resolution trait assessment methodologies, such as were described at the meeting by Björn Sundberg (Umeå Plant Science Centre, Sweden; see also Gorzsás et al., 2011), offers the potential to quickly identify commercially desirable genotypes which could then be clonally propagated. Thus, despite the many knowledge gaps remaining in our understanding of plant cell wall biology, opportunities now exist to improve the characteristics of plant biofuel feedstocks without attempting to directly intervene in the underlying cellular machinery. At the same time, the synergy between association genomics, which reveals loci whose polymorphic forms condition particular traits, and transgenic technologies, which enable detailed analysis of the molecular players, is anticipated to rapidly move our knowledge boundaries forward.

A unique feature of the Bioenergy trees symposium, compared to many other meetings dealing with plant cell walls, was the inclusion of speakers who could address relevant engineering and socio-economic issues, especially through life-cycle analysis. Both Ganti Murthy (Oregon State University, USA) and Richard Murphy (Imperial College London, UK) provided important insights into the impact of relatively modest changes in feedstock conversion efficiency or processing costs, while in the final discussion session, even wider ranging questions were explored. For example, attendees felt that if follow-through from the present wave of biofuels research is going to require large-scale modifications to both our landscape and economic infrastructure, early and direct engagement in this research of key stakeholders, from politicians to consumers, will be crucially important to its ultimate success. Thus, the Bioenergy trees meeting made it clear that the challenges facing plant biologists concerned with helping build the new bioeconomy will extend well beyond manipulation of cell wall composition, photosynthate partitioning or saline tolerance (Janz et al., pp. 129–141). The bioenergy trees planned for the future will have to be optimized for many different uses (Behnke et al., pp. 70–82) and for growth across a wide range of environments, but they will also need to be deployed in a socially responsible manner if we are to realize their full potential to contribute newly captured carbon and energy to the global economy.