An additional characteristic of alfalfa that makes it attractive for biorefinement is that it is amenable to genetic transformation. Rapid and efficient methods for transformation using Agrobacterium tumefaciens have been developed and gene

promoters identified for high constitutive expression and for tissue-specific expression (reviewed by Samac and Temple, 2004; Somers et al., 2003). Trans­formation has been used to alter alfalfa for production of valuable coproducts (Table 5.2) and for improving digestion of alfalfa fiber. Transgenic alfalfa has been shown to be capable of producing high levels of phytase (Austin-Phillips and Ziegelhoffer, 2001; Ullah et al., 2002), a feed enzyme that degrades phytic acid and makes phosphorus in vegetable feeds available to monogastric animals such as swine. Adding phytase to feeds reduces the need to add supplemental phosphorus to feed and reduces the amount of phosphorus excreted by animals. In field studies, juice from wet-fractionated alfalfa plants contained 1-1.5% phytase. Phytase activity in juice was stable over two weeks at a temperature of 37°C. Activity is also stable in dried leaf meal. Both juice and dried leaf meal added to feed were as effective in feeding trials as phytase from microbial sources. The value of the enzyme and xanthophyll in the juice was estimated at $1900/acre (Austin-Phillips and Ziegelhoffer, 2001). A wide range of feed enzymes is used to enhance digestion of feed and improve animal performance. Use of feed enzymes in monogastric and ruminant animals in expected to increase worldwide (Sheppy, 2001). Production of feed enzymes in transgenic plants, particularly in plants used as animal feed, would be an opportunity to increase feed utilization as well as value of the feed.

Transgenic alfalfa has also been used to produce several industrial enzymes. A manganese-dependent lignin peroxidase, which can be used for lignin degra­dation and biopulping in the manufacture of paper, was expressed in alfalfa. However, high levels of production of this enzyme appeared to be detrimental to plants (Austin et al., 1995). In the same study, a-amylase was produced at a level of approximately 0.01% of soluble protein without having a negative effect on plant development. Two cellulases, an endogluconase and a cellobiohydrolase, have been expressed at low levels in alfalfa (Ziegelhoffer et al., 1999). These enzymes were stable in dried leaf meal. Expression of cellulose degrading enzymes in biomass plants is one strategy to decrease the costs of saccharification that precedes ethanol fermentation. Alfalfa plants have also been shown to be an excellent “factory” for the production of chitinase (Samac et al., 2004). Chitin, found in shells of crustaceans, is the second most abundant carbohydrate after cellulose, and a potential feedstock in a biorefinery.

In addition to production of proteins, the use of transgenic alfalfa to produce other industrial feed stocks has been explored. Polyhydroxyalkanoates (PHAs) are produced by many species of bacteria and some PHA polymers are commer­cially valuable as biodegradable plastics. PHA synthesis in plants is seen as a more economically viable means of producing large quantities of these polymers (Poirier, 1999; Slater et al., 1999). Alfalfa was engineered to constitutively express three bacterial genes for the production of poly-p-hydroxybutyrate (PHB) (Saruul et al., 2002). Granules of PHB were shown to accumulate in chloroplasts without any negative impact on plant growth. Yield of PHB by chemical extrac­tion was relatively low (1.8 g kg-1 DM), but may be improved by optimizing extraction methods or by utilizing stronger gene promoters.

A major limitation to use of biomass in the production of ethanol is the recalcitrance of the material to saccharification. Cross-linking of lignin with cell — wall polysaccharides interferes with enzymatic degradation of cellulose and can severely limit the conversion of herbaceous plant material into ethanol. Lignin in alfalfa stems also limits digestion of feed by ruminant animals. In experiments aimed at increasing feed digestion by ruminants, transgenic alfalfa was produced that had decreased expression of caffeoyl coenzyme A 3-O-methyltransferase, an enzyme involved in synthesis of lignin precursors. These plants were shown to have approximately 20% less lignin and 10% additional cellulose than the controls (Marita et al., 2003). The rate of digestion of the transgenic material was deter­mined by in vitro rumen digestibility assays. In the transgenic material, a 2.8-6.0% increase in the rate of digestion was observed (Guo et al., 2001). This material could have a very significant impact on both animal nutrition and alfalfa biorefining. Casler and Vogel (1999) determined that a 1% increase in forage digestibility would lead to a 3.2% increase in average daily live-weight gain by beef steers. Although this material has not yet been tested with different pretreat­ment methods or used in saccharification or fermentation studies, based on chem­ical analyses, it may also have improved qualities as a feedstock for bioethanol production.

During the past several years, barrel medic (Medicago truncatula) has been the object of a broad range of research efforts worldwide. This annual plant, which is closely related to alfalfa, is a model plant for study of plant-microbe interactions and plant development (Cook, 1999). Chromosome mapping has shown that there is a high degree of gene synteny between the two species as well as a high degree of DNA sequence homology (Choi et al., 2004). Numerous genomic tools have been developed for M. truncatula including isolation of over 189,000 expressed sequence tags (ESTs), identification and sequencing of more than 36,000 unique genes (http://www. tigr. org/tigr-scripts/tgi/T_index. cgi? spe- cies=medicago), extensive genetic and physical mapping (Choi et al. 2004), development of microarrays for transcript profiling, and a genome sequencing project is currently underway (http://www. medicago. org). In particular, micro­arrays are valuable tools for identifying genes involved in important agricultural processes as they enable researchers to measure expression of thousands of genes simultaneously. More than 100 genes are involved in cell-wall biosynthesis in plants and little is known about regulation of their expression. EST resources may be useful both as markers for selecting plants with favorable characteristics in bioconversion and in modifying gene expression in transgenic plants for enhancing the efficiency of ethanol production or enhancing yields of valuable coproducts.