Thus far, the majority of hydrolases expressed in plants have been thermostable, with activity maxima at temperatures greater than ambient (T > ~50°C). The logic for using thermostable enzymes is that pretreatment conditions will facilitate saccharification. Since thermostable enzymes often exhibit poor activity at ambient temperatures, such enzymes may also reduce deleterious effects on plant phenotype (Taylor et al. 2008). As will be discussed more below, these enzymes have been targeted to a variety of subcellular compartments—the cytosol, apoplast (cell wall), vacuole, and chloroplast. Subcellular targeting allows higher accumulation of proteins and limits enzymatic activity until cells are lysed during pretreatment or harvest (Ziegelhoffer et al. 2001; Hood et al. 2007). The most commonly used thermostable enzyme for in planta expression is the E1 endoglucanase (a cellulase) from Acidothermus cellulolyticus (Tucker et al. 1989). For example, E1 targeted to the cell wall of maize and tobacco increases biomass digestibility by up to 10% (Brunecky et al. 2011). Based on imaging data, the authors hypothesized that E1 acts by nicking cellulose chains as they are formed, creating more free chains for the plants’ endogenous exoglucanases to act upon (Brunecky et al. 2011).

Of special relevance to switchgrass and other grasses, a handful of studies have targeted in planta expression of ferulic acid esterases (FAEA) from Aspergillus. Ferulic acid esterases hydrolyze ester linkages between hydroxycinnamoyl esters and cell wall saccharides. Vacuolar targeting of FAEA in Lolium multiflorum and Festuca arundinacea decreased the feruloyl ester linkages in cell walls and ferulate dimers that cross-link polysaccharide chains (Buanafina et al. 2006; Buanafina et al. 2008). FAEA targeted to the ER, Golgi, and apoplast in Festuca arundinacea disturbed the ferulolyation process, enhancing biodegradability (Buanafina et al. 2010). Harholt and colleagues have also successfully targeted FAEA to the endosperm of wheat but with severe pleiotropic effects on fertility and with loss of seed mass (Harholt et al. 2010). Across all studies, the resultant digestibility increase was around 5-10%. Researchers found that the transformed plants compensate for the expression of FAEA by enhancing arabinoxylan synthesis and crosslinking. This indicates that multiple genes and/or inhibition of the cell wall biosynthesis signaling (Wolf et al. 2012) may be needed to produce a significant change in cell wall composition.

In addition to hydrolytic enzymes, in planta expression of carbohydrate binding modules (CBMs) may offer a promising approach to increasing biomass conversion. Such domains are also encoded in plant genomes and in fungi, under the monikers of expansins and swollenins (Abramson et al. 2010). Simply adding CBMs to cell wall materials during saccharification enhances sugar yield from cellulase attack, which may be due to their ability to disrupt cellulose microfibrils (Pauly et al. 2008), and is consistent with the improvements of GH activity when fused with a CBM mentioned earlier (Mahadevan et al. 2011).

In a few cases, an interesting secondary effect has been observed in plants overexpressing certain foreign or endogenous hydrolases and CBMs—namely, an increase in growth rate and biomass accumulation in conjunction with enhanced digestibility. Earlier studies involving CBM expression showed that these proteins can enhance plant growth and biomass (Abramson et al. 2010). While the precise mechanism is unknown, it is thought that these proteins uncouple cellulose polymerization and crystallization, allowing enhanced cellulose biosynthesis. Additionally, Arabidopsis expressing a poplar cellulase, PaPopCel1, showed an increase in the size of leaf and stem cells and corresponding organs (Park et al. 2003). The authors suggested, based on chemical analysis and NMR, that the cellulase diminishes the intercalation of xyloglucans into cell walls, thereby permitting cellular expansion and growth. This hypothesis was further supported when overexpression of xyloglucanase from Aspergillus aculeatus in poplar caused enhanced growth and increased cellulose deposition in internode xylem (Park et al. 2004). Sengon trees (Paraserianthes falcataria) expressing PaPopCel1 also showed increased growth and higher levels of soluble xyloglucans (Hartati et al. 2008). As expected, the elimination of xyloglucan tethering also enhances saccharification. Examination of several poplar transgenic lines each expressing a xyloglucanase, cellulase, or xylanase showed that those overexpressing xyloglucanase had markedly greater cellulose degradation in the xylem compared with wild-type and cellulase — or xylanase-expressing genotypes (Kaida et al. 2009). A similar outcome was also seen due to overexpression of A. niger xyloglucanase in Acacia mangium (Kaku et al. 2011). The hypothesized loosening of xyloglucan intercalation between the cellulose microfibrils due to glucanase action enhanced saccharification of Acacia mangium biomass by 1.4-fold, resulting in 10%-15% higher production of ethanol (Kaida et al. 2009). These results are especially note-worthy as production of plants that are not only larger than their wild-type counterparts, but also easier to degrade, could greatly enhance the efficiency of biofuel production. Of course, the mechanisms behind this process and the consequences for plant development, growth, physiology, and stress resistance must be better understood before this can actually be applied in agriculture. Since xyloglucan is not abundant in grasses, it remains to be tested if xyloglucanase expression could enhance grass stature and saccharification, or if another class of GHs might have this effect.