Potential for Allyl/Propenyl Phenols?

We consider that biotechnological approaches utilizing the enzymatic machinery described above should now be explored to produce allyl and propenyl phenols, diverting (to a predetermined extent) some of the carbon flow, for example, from lignin toward these metabolites, while maintaining the resulting plants’ capacities to grow and function within acceptable boundaries. The resulting cellulosic biomass could thus become more amenable to pulp/paper manufacture and/or biofuel production due to the lower lignin contents. Alternatively, oilseed metabolism could be manipulated to produce these allyl/propenyl phenol substances in larger amounts. Upon processing and purification, these compounds could potentially find uses as biofuels, biofuel precursors, flavors/fragrances, and/or intermediate chemicals (e.g., as monomers for synthetic polymers).

FIGURE 13.17 Reactions catalyzed by (A) PLR and (B) PCBER.

Moreover, the ability to routinely biotechnologically modify plants and/or cell cultures to produce allyl/propenyl phenols, such as chavicol (31) and eugenol (33), now offers the opportunity to consider much larger markets for these products, that is, in addition to expanding the flavor/fragrance/antiseptic/biocidal markets that currently exist (see above Section 3). For instance, in terms of the flavor/fragrance market, the natural vanillin market can potentially be expanded. Today, only about 0.2% of vanillin (40, Fig. 13.9) used originates directly from its botanical source, the vanilla bean, where it commands a sales price of approximately US$ 4000 kg^-1 as a natural product. The bulk is semisynthetic, being chemically synthesized either from the petrochemical-derived guaiacol (41) or from pulp/paper lignin-derivatives (i.e., technical lignins). However, the vanillin (40) produced in this way commands a price of only US$ 12 kg^-1 because it is not ‘‘natural.’’

Driven by consumer preferences toward truly ‘‘natural’’ food products, food additives, and pharmaceuticals, several biocatalytic processes for production of plant-derived metabolites using microbes and plant cell cultures have been developed and patented in

FIGURE 13.18 Schematic representation of the crystal structure of TpPLR1 with NADPH and (-)-pinoresinol (58). Source: Reprinted from Min et al. (2003).

recent years (Berger, 1991; Krings and Berger, 1998; Longo and Sanromán, 2006; Priefert et al., 2001; Rabenhorst and Hopp, 1991; Schrader et al., 2004; Shimoni et al., 2000; Yoshimoto et al., 1990). This enthusiasm has been fueled by the promise to generate transgenic cultures with increased efficacy for production of target metabolites. This approach also apparently offers economic advantages over conventional chemical syntheses, including better stereospecificity in product formation and lower amounts of waste products being generated upon processing (Schoemaker et al., 2003). In particular, production of flavors via biotechnological processes offers an additional economic advantage since, unlike their chemically prepared counterparts, the resulting products can be marketed as ‘‘natural’’ under current US and EU legislation (The European Commission, 1991; Food and Drug Administration, 2006; Lesage-Meessen et al., 1996; Shimoni et al., 2000). In this regard, successful microbial production of vanillin (40) from various precursors, such as eugenol (33) and isoeugenol (39), has been reported recently, being achieved at relatively high substrate concentrations (Krings and Berger, 1998; Longo and Sanroma´n, 2006; Priefert et al., 2001; Rabenhorst and Hopp, 1991; Schrader et al., 2004; Shimoni et al., 2000);

FIGURE 13.19 Currently annotated phylogenetic analysis of several PIp-reductase homologues from different plant species, with relevant homologues in basil (ObEGS1), petunia (PhIGS1), and creosote bush (LtCES1) highlighted. IFR, isoflavone reductase; LACR, leucoanthocyanidin reductase; NmrA, nitrogen metabolite repression regulator; PCBER, phenylcoumaran-benzylic ether reductase; PLR, pinoresinol-lariciresinol reductase; PTR, pterocarpan reductase. Sequences were obtained from the NCBI database and filtered for <0.75 sequence difference, ClustalW-aligned, and subjected to neighbor-joining phylogenetic analysis using PHYLIP (Felsenstein, 1993).

for example, transformation of isoeugenol (39) (20 g/L) using a strain of Serratia marcescens led to vanillin (40) accumulation (3.8 g/L). An enzymatic process for conversion of isoeugenol (39) into vanillin (40) using a ligno stilbene-α,β-dioxygenase from a Pseudomonas paucimobilis strain has also been patented (Yoshimoto et al., 1990). Thus, the technologies are now apparently in hand to permit formation of natural vanillin (40) through established microbial and genetic manipulations in either plants and/or plant/bacterial cell cultures.

Much larger anticipated potential markets for the allyl/propenyl phenols include the industrial polymers and biofuel/biodiesel. Regarding polymer applications, the expected worldwide production of polystyrenes alone was approximately 25 million metric tons in 2006, representing sales of US$ 31 billion. Allyl/ propenyl phenols can be converted into functionalized polystyrene derivatives, and an increased supply creates the potential for their massive usage as intermediate (monomer) chemicals in industrial polymers. Currently, existing applications include eugenol- (33) based polymers, which are widely used in dentistry in zinc oxide impression pastes applied as surgical dressings and temporary cements (Skinner, 1940; Weinberg et al., 1972), as well as specialty modifying (i.e., coating) agents in analytical electrodes (Ciszewski and Milczarek, 1998, 1999, 2001, 2003; Rahim et al., 2004).

The functionalized β-methylstyrenes anethole (38) and isoeugenol (39) can be converted into polymers of several thousand dalton (Bywater, 1963). However, the potential of such conversions has been studied only to a limited extent relative to their vinyl analogues (i.e., styrene and derivatives thereof), in part due to limited supply/availability and because propenylbenzenes do not apparently undergo as efficient free radical polymerization reactions as styrenes—even though their electronic configuration is such that a radical intermediate can also be stabilized by the aromatic ring (Alexander et al., 1981). The most efficient polymerization initiators described thus far for propenylbenzene derivatives are Lewis acids, particularly AlCl₃, SnCl₄, and BF₃ (Alexander et al., 1981; Cerrai et al., 1969a,b; Secci and Mameli, 1956).

In general, the steric factors on the monomers define, to a large extent, both polymerization rates and molecular weights of the resulting polymers, with anethole (38), for example, being more reactive than isoeugenol (39) (Alexander et al., 1981). Polymerization reactions can proceed through a conventional 1,2-chain formation, similar to styrene, with the propagating species being a Lewis acidinduced carbocation that is added to the double bond of another monomer. This results in a polymer backbone composed of the carbons 7 and 8 of the original monomers. The molecular weights of the resulting polymers are higher at lower temperatures and, in the case of anethole (38) when polymerized by SnCl4, can vary from a few thousand up to about 75,000 Da depending on both temperature and dilution levels (Bywater, 1963; Cerrai et al., 1969a; Secci and Mameli, 1956). For isoeugenol (39), the phenolic oxygen moiety also participates in the polymerization reactions, thereby increasing the structural complexity of the resulting polymer(s) so formed (Fig. 13.20A) (Evliya and Olcay, 1974). Additionally, allylphenols, such as methylchavicol (32) and eugenol (33), can form mixed polymers, resulting from the partial rearrangement of the side-chain double bond upon carbocation formation prior to attachment to the polymer chain (Cihaner et al., 2001; Kennedy, 1964) (Fig. 13.20B).

FIGURE 13.20 Polymerization of (A) isoeugenol (39) via furanocoumaran intermediacy and (B) methylchavicol (32) and eugenol (33) through rearrangement prior to polymerization.

In terms of their potential uses as biofuels, it is noteworthy that ~100 billion gallons of gasoline fuel were consumed in the United States in 2005. In addition, the annual consumption of diesel fuel in 2000, including highway diesel, farms, electric power, railroad, fuel oil (residential, commercial, and heating), and kerosene, totaled approximately 57.1 billion gallons. As a measure of the potential scale of production of biodiesel (from vegetable oils, consisting mainly of fatty acid esters), Peterson (1995) recently estimated that if all harvested cropland (~363 million acres) in the United States was dedicated exclusively to rapeseed (oilseed) production, then approximately 36.3 billion gallons (assuming 100 gallon/acre) of vegetable oil could be obtained annually. [Note that at present there are approximately 27 billion gallons of vegetable oil produced worldwide annually (Peterson, 1995)]. Yields for other plant species, modified to concurrently synthesize chavicol (31), methylchavicol (32), or eugenol (33), now need to be determined to establish to what extent these productivity numbers can be increased through (for instance) whole plant utilization.

Additionally, if one considers the annual pulp and paper production in the United States (~120 million metric tons/year, 1997 figures), the potential also exists to divert some part of the production of lignins/heartwood lignans and other phenylpropanoid derivatives in commercially important woody plant species away from their natural biosynthetic pathways, that is, to afford allyl/propenyl phenols, etc. In principle, the lignin/lignan substances currently produced annually as by-products of pulp/paper industries (more than 50 million tons) could instead be converted to approximately 15 billion gallons allyl/propenyl phenols per annum, if fully converted. Nevertheless, any reduction in carbon flow to lignin, or reductions/changes in heartwood-forming constituents, could represent a significant increase in biofuel production.

FIGURE 13.21 Cyclohexane derivatives formed upon catalytic hydrogenation of the corresponding allyl/propenyl phenols.

Allyl and propenyl phenols have relatively high heats of combustion at room temperature, with values generally being about 70% (per weight) of medium chain hydrocarbons such as octane and decane. That is, these allyl/propenyl phenols can potentially generate more energy (per weight) than ethanol. In terms of other relevant properties, using two examples only for illustrative purposes, chavicol (31) has boiling/flash points of 238/102°C at normal atmospheric pressure and density approximately 1.01 g/cm³, whereas eugenol (33) values are approximately 253/112°C and 1.07 g/cm³ (at 20°C). Such values are within the ranges needed for biodiesel/biofuel considerations. Their reported freezing points are, however, generally between –10°C and room temperature, which would reduce their potential as liquid biofuels if used exclusively as such in pure liquid form. This limitation might be circumvented, however, by either blending them into other fuels, similar to the coconut oils added as biofuels to diesel in the Philippines (BBC/PRI/WGBH The World, 2007) or through their chemical derivatization to generate materials of lower freezing point prior to biofuel use (e.g., hydrogenation, which may also help reduce pollutant emission upon combustion).Catalytic hydrogenation of side-chain double bonds of allyl/propenylbenzenes is readily achieved at atmospheric pressures, whereas reduction of the aromatic ring typically requires higher temperatures and pressures using traditional metal catalysts (e.g., supported Pd or Raney Ni). Reduced allyl/propenyl phenols have already been generated by such catalytic hydrogenation reactions; for example, 2-methoxy-4-propyl-cyclohexanol (62, Fig. 13.21) was obtained in near-quantitative amounts from eugenol (33) (Maillefer, 1990). Newer catalytic systems, however, have the exciting potential to dramatically improve the reduction conditions, for example, as recently reported for the quantitative hydrogenation of several benzene analogues, at room temperature and atmospheric hydrogen pressure, using ruthenium-containing methylated cyclodextrin catalysts (Nowicki et al., 2006). Thus, this application of biotechnology, if further explored/developed/applied, offers a potentially important new avenue for sources of biofuels/bioenergy.