In studying the action of cellulases on cellulose, different model cellulosic substrates are employed, which exhibit different levels of crystallinity and accessibility to enzyme, two of the major parameters that contribute to the overall recalcitrance of the substrate. It should also be remembered that the procedures used for the preparation of these model cellulose substrates result in a form, which is very different from that of the native cellulose microfibrils within the plant cell wall.

The least crystalline substrates used for such studies are the soluble derivatized forms, such as carboxymethyl cellulose or hydroxymethyl cellulose, which, due to the substituted groups, are essentially non-crystalline in nature. The soluble, derivatized celluloses are used as a substrate for endoglucanases that can cleave along the cellulose chain, providing that the derivatization is not too extensive. Exoglucanases begin at the chain ends, and, upon meeting the substituent group, immediately come to a halt. Thus, such soluble derivatized substrates can be used to differentiate between the endo — and exo-acting enzymes. It should be noted that cellulases tend to exhibit a spectrum between the endo — and exo-modes of

cellulolytic activity; and several enzymes are known to display an endo-processive action — i. e., the initial steps of cleavage may be endo-acting whereby the enzyme binds in some manner to the internal portion of the cellulose chain, whereupon the enzyme continues to degrade the substrate processively.

The next level of crystallinity is exemplified by phosphoric acid-treated celluloses, some­times referred to as amorphous cellulose, sometimes acid-swollen cellulose. The crystallinity of the latter substrate depends on the conditions used for treatment and is usually very low, although residual levels of crystallinity may be observed. In any case, some enzymes, whether endo — or exo-acting, degrade this type ofsubstrate readily, whereas others do not. In this case, the exact mechanistic features of substrate-versus-enzyme that generate extensive degradation are not understood.

Cotton cellulose is about 45% crystalline and is sometimes used as a model plant-derived substrate. Avicel, a commercial preparation of microcrystalline cellulose, is commonly used as a model substrate for examination of cellulolytic enzyme systems. Avicel is also about 45% crystalline (131, 132), but exhibits heightened recalcitrance owing to its large particle size (due to the drying of the material during its preparation) and consequent inaccessibility of the inner cellulose chains. Bacterial cellulose ribbons, produced by the bacterium Acetobacter xylinum, are about 65% crystalline, and BMCC (bacterial microcrystalline cellulose) is prepared by acid hydrolysis of amorphous regions of the bacterial cellulose ribbons, thus resulting in a crystallinity of about 70%. By far, however, the most crystalline and recalcitrant form of cellulose is derived from the Valonia ventricosa cell wall, which is close to 100% crystalline (131).

Despite the rather high crystallinity of bacterial cellulose ribbons, the C. thermocellum cellulosome completely dissolves this substrate relatively rapidly, within a 24-hour period under the conditions of the assay (Figure 13.5). Under the same conditions, a sluggish but relentless degradation of the highly recalcitrant Valonia cellulose is achieved, reaching near completion only after a 16-daytime interval (133).

The modified morphologies of the different substrates canbe followed ultrastructurally by transmission electron microscopy, and the images provide insight into the mechanism and extent of degradation of recalcitrant substrates (134-137). Cellulosome-induced degrada­tion ofbacterial cellulose ribbons was indicative of a digestion pattern suggesting a concerted assault of different types of cellulases (e. g., combined endo — and exo-acting) on the cellulose substrate (Figure 13.6), consistent with the spatial proximity of the two types of enzymes in the cellulosomes (133). In this context, the observed cleavage of the cellulose ribbons is the signature of an endo mode of action, and the observed defibrillation of the substrate (see Figure 13.6B) corresponds to processive action associated with exo-acting cellulases. The residual bacterial cellulose, following 85% degradation (Figure 13.6C), bears no resemblance to the fine ribbons that were originally subjected to cellulosome action.

The images of partially degraded, highly recalcitrant Valonia cellulose show a rather dif­ferent picture (Figure 13.7). Unlike the digestion pattern observed for the bacterial cellulose ribbons, the degradation of Valonia microcrystals is accompanied by distinctive digestive features, including crystal thinning and pointed tips, reminiscent of previously characterized features during the degradation of this substrate by individual and combined fungal cellu — lases. The thinning feature was previously characterized as a function of processive action by the exo-acting fungal cellobiohydrolase I (Cel7A), whereas pointed tips were associated with the action of the less processive “unidirectional” digestion (i. e., vis-a-vis non-reducing to reducing or vice versa) of cellobiohydrolase II (Cel6A) (135, 138).

Interestingly, the micrograph in Figure 13.7B shows both types of features. Even though over 95% of the cellulose have been digested, the individual residual cellulose microcrystals often show both features and some are apparently unchanged from the original images. It seems as if the individual cellulosomes from the same batch exhibit a wide diversity in their mode of action, probably related to the inherent heterogeneity in their enzyme content. The persistence of pointed tips, indicates unidirectional processivity (135, 138). Many of the properties of the intact cellulosome seem to be analogous to those of this particular processive enzyme. The direction of cleavage of the family-48 enzymes is from the reducing to non-reducing ends (139). The persistence of intact Valonia cellulose crystals, even after near-complete digestion may indicate that the rate-determining step in its degradation is the initial attack, once consummated, the crystal undergoes rapid degradation. Indeed, the rate-limiting step of cellulose degradation has been considered to be the separation of the individual cellulose chains from the crystal lattice. Once exposed, the battery of enzymes can deal both with the separated chain as well as the void left in the crystal.

After lignocellulose pretreatment, there are four biologically mediated events typically in the course of biological processing of cellulosic biomass: cellulase production, enzymatic cellulose hydrolysis, hexose fermentation, and pentose fermentation (Figure 16.2). Separate

SHF: Separate hydrolysis & fermentation CBP: Consolidated bioprocessing

SSF:Simultaneous saccharification & fermentation SSCF: Simultaneous saccharification & co-fermentation

Figure 16.2 Evolution of biomass processing configurations featuring enzymatic hydrolysis.

hydrolysis and fermentation (SHF) involves these four discrete process steps. Simultaneous saccharification and fermentation (SSF) consolidates cellulose hydrolysis and hexose fer­mentation. Simultaneous saccharification and co-fermentation (SSCF) combines cellulose hydrolysis, hexose fermentation, and pentose fermentation. Consolidated bioprocessing (CBP) integrates cellulase production, cellulose hydrolysis, with pentose and hexose fer­mentations in a single step (10, 15, 28).

Over the past few years, much effort has been devoted to reducing the cost of cellulase en­zyme production (24). Following greater than 20-fold cost reductions, cellulase production costs have recently been reported in the range of ~20 cents per gallon of cellulosic ethanol produced (29). These developments enable a variety of formerly infeasible industrial SSF and SSCF processes, but do not diminish the competitive potential of CBP by offering significantly lower costs than other processes.

CBP offers the potential for lower production costs, lower capital investment, and higher conversion efficiency as compared to the processes featuring dedicated cellulase produc­tion. CBP avoids costs for capital, substrate, other raw materials, and utilities associated with cellulase production. In addition, CBP could realize higher hydrolysis rates, and hence reduce reactor volume and capital investment as a result of enzyme-microbe synergy. CBP provides access to the use of thermophiles or other organisms with high activity cellu — lases. Moreover, cellulose-adherent cellulolytic microorganisms may successfully compete for products of cellulose hydrolysis with non-adhered microbes. Moreover, these microor­ganisms are likely to be less sensitive to contaminants, which could increase the stability of an industrial processes. Economic analysis suggests that the sum of 9.9 p7gal ethanol

for dedicated cellulase production and 9.0 у/gal for SSCF gives a total cost for biological processing of 18.9 c/gal, which is more than fourfold greater than the 4.2 c/gal projected for CBP (28).

Today, there are no CBP-enabling microorganisms suitable for industrial applications. CBP microorganism development can proceed via a native cellulose utilization strategy and a recombinant cellulose utilization strategy (Figure 16.3). The native cellulolytic strategy involves engineering product metabolism to produce desired products based on naturally cellulolytic microorganisms (e. g., Clostridium thermocellum). The recombinant cellulolytic strategy involves introducing heterologous cellulase genes into an organism whose product yield and tolerance credentials are well-established (e. g., Baker’s yeast Saccharomyces cere — visiea). Each strategy has its own advantages and challenges, and different strategies may well prove most advantageous for different products.

Xylan, a polymer of p-(1-4)-linked D-xylose is one of the main components of woody plants. Xylans are usually substituted by side chains of arabinose or glucuronic acid and may be acetylated. Thus, glucuronoxylan (GX) is composed of a linear backbone of p — (1-4)-linked D-xylosyl (Xyl) residues, some of which bear a single a-D-glucuronic acid (GlcA) or 4-0-methyl-a-D-glucuronic acid (MeGlcA) residue at O-2. The Xyl residues can also be substituted with arabinosyl and acetyl residues (128). Xylosyltransferase and glucuronyltransferase activities have been detected in numerous plants (129). However, none of the genes encoding these enzymes has been identified, nor have any of the enzymes been purified to homogeneity and biochemically characterized.

Mutations in three glycosyltransferases, FRAGILEFIBER8 (FRA8), IRREGULARXYLEM8 (IRX8), and IRX9, have been shown to be required for normal vessel morphology and wall thickness and for normal amounts of xylose and cellulose in cell walls (73, 74, 130, 131). These genes are specifically expressed in cells undergoing secondary wall thickening. Plants carrying mutations in these genes have reduced amounts of wall GX and a decreased ratio of GlcA to MeGlcA residues in the GX(129, 130, 132).

IRX8, IRX9, and FRA8 are specifically expressed in fibers and vessels and their encoded proteins are localized in the Golgi (129,130,132). Thus, they have the properties expected of enzymes involved in glucuronoxylan synthesis. However, it has not been possible to directly associate enzyme activity with the proteins. Thus, it is not clear how they participate in the synthesis of xylan. Pena and coworkers (129) showed that the glycosyl sequence 4-P-D-Xylp — (1 ^ 4)-p-D-Xylp-(1 ^3)-a-b-Rhap-(1 ^2)-a-D-GalpA-(1 ^4)-D-Xylp was present at the reducing end of Arabidopsis GX, as previously noted for birch (Betula verrucosa) and spruce (Picea abies) wood. They further noted that mutations in IRX8 and IRX9, and by inference FRA8, lead to reductions in the amount of the GXreducing end sequence suggesting that these genes participate in the synthesis of the GX reducing end sequence and suggest that IRX9 has an essential role in the elongation of the xylan backbone.

The fra8 gene encodes a GT47 family enzyme and expression of the poplar (Populus alba x tremula) GT47C gene in fra8 plants rescues the defects in secondary wall thickness and GX synthesis, suggesting that GT47C is a functional homolog of FRA8 (133). The FRA8 gene encodes a putative GT in family GT47 (130). This family includes enzymes with an inverting mechanism, which usually leads to p-glycosidic linkages (when typical a-linked donor substrates are used). Thus, if UDP-a-D-Xyl is the donor substrate, it is possible that FRA8 catalyzes the formation of the p-linkage of xylose to either O-3 of the rhamnose or O-4 of the penultimate xylose of the GX reducing end glycosyl sequence (129). However, in plants, the addition of a-Rha residues is catalyzed by inverting GTs that use UDP-p-L-Rha as the donor substrate (134). Therefore, it is also possible that FRA8 catalyzes the addition of rhamnose during the biosynthesis of the GX reducing end sequence.

The IRX8 (GAUT12) gene encodes a putative GT in family GT8 (73,74,129,132). Several members of the GT8 family catalyze the transfer of uronic acids to glycans. For example, three Arabidopsis GT8 proteins, QUASIMODO1 (QUA1) (135), PARVUS (136), and GALAC- TURONOSYLTRANSFERASE1 (GAUT1) (137), have been identified and are believed to have a role in pectin biosynthesis. Of these three, only GAUT1 has been biochemically char­acterized and shown to have galacturonosyltransferase activity (137). Family GT8 enzymes are retaining glycosyl transferases that catalyze the formation of a-glycosidic bonds when using a-linked donor substrates such as UDP-a-D-GalA. Thus, it is possible that IRX8 cat­alyzes the addition of an a-D-GalA residue to O-4 of the reducing Xyl residue present in the GX reducing end sequence described above.

Mutation of the IRX9 gene, which encodes a putative GT in family GT43 (138), was shown to result in plants with decreased amounts of wall GX, suggesting that this gene is required for GX synthesis (131). The poplar (Populus tremulax tremuloides) GT43A and Ptt GT43B genes, which are homologs of IRX9, have been shown to be highly expressed during wood formation (139). In addition, a cotton (Gossypium hirsutum) gene, which resides in the same phylogenetic subgroup as Ptt GT43A, Ptt GT43B, and IRX9, is highly expressed during cotton fiber development (140). Together, these findings suggest that family 43 GTs have an important role in secondary wall synthesis. Enzymes in this family are distinguished by an inverting mechanism, typically catalyzing the formation of p-glycosidic bonds using a-linked glycosyl donors. Our demonstration that the irx9 mutation leads to a decrease in the chain length of GX suggests that IRX9 encodes a xylan synthase responsible for adding p — xylosyl residues to the nascent GX. This hypothesis is consistent with our results indicating that IRX9 is highly expressed in cells undergoing secondary wall biogenesis and that IRX9 is localized in the Golgi, where GX synthesis occurs (141, 142). However, it is also possible, as we discussed previously (130), that an inverting GT can catalyze the formation of an a-linkage when a p-linked substrate (such as a glycosyl phospholipid) is used as the donor. Such inverting enzymes can also catalyze the formation of high-energy p-linked glycosides (such as glycosyl phospholipids) that are subsequently used as glycosyl donors. Thus, an alternative interpretation is that IRX9 is directly or indirectly involved in the transfer of а-linked GlcA residues to the GX backbone. Additional studies are required to determine whether IRX9 catalyzes the addition of xylose or GlcA to the GX backbone.

In the lignified walls of the Poaceae, the major non-cellulosic polysaccharides are glu — curonoarabinoxylans (GAXs), although the degree of substitution of the xylan main chain is less than in the GAXs of the primary cell walls. In the non-lignified walls of the Poaceae and other species such as pineapple (Ananas comosus), ferulic acid is ester-linked to GAXs. These polysaccharides comprise only a minor component of the non-cellulosic polysaccharides of the non-lignified walls of species in the basal Arecales (palms) clade (143), but are a major component of the non-lignified walls of species in the other commelinid clades, particularly the Poales (144-146).

This echoes the observation by Cross and Bevan almost a century ago when crystallinity of cellulose was first proposed, “The root idea of crystallography is identical invariability while the root idea of the world of living matter is essential individual variation” (44). Recognizing the species and tissue specificities of the structures of native celluloses is essential to progress in understanding the diversity of cellulose synthases encoded in the genomes of plants and to understanding the even greater diversity of the cellulases produced by species-specific plant pathogens.

Lignification and cell wall assembly processes are also frequently altered/modified by plants in response to structural and environmental stresses/conditions. For instance, many arbores­cent gymnosperms and angiosperms form heartwood beginning at the center of the stem and this eventually extensively radiates outwards through (in part) secretion of defense molecules (73), such as lignans from adjacent parenchyma cells in the sapwood (74) (Figures 7.7A-C). That is, following lignification, the various heartwood deposition processes result in forma­tion of large quantities of diverse metabolic products, including species-specific resins and lignans, which increase the density (and presumably support) of the wood, as well as the

parenchyma

Substances secreted through pit apertures

Neighboring cell lumen

Figure 7.7 Woody cross sections showing heartwood deposition and reaction wood tissues. Heartwood in (A) tamarack (Larix laricina) and (B) ebony (Diospyros ebenum), as well as (C) secretion of heartwood constituents by ray parenchyma cells into lumen of neighboring cells; this appears to occur through pit apertures (73). Light micrograph cross section of compression (D) and "normal" (F) wood in Douglas fir (Pseudotsuga menziesii) (75) and of tension (E) and "normal" (G) wood in black cottonwood (Populus balsamifera ssp. trichocarpa) (72). Bar: 20 pm (D, F) and 10 pm (E, G). Abbreviations: f, fiber;G, G — layer;Hw, heartwood;is, intercellular space;gf, gelatinous fiber;Sw, sapwood;v, vessel. [Reprinted from (A) Phytochemistry, vol. 57, Kwon, M., Bedgar, D. L., Piastuch, W., Davin, L. B. & Lewis, N. G., Induced compression wood formation in Douglas fir (Pseudotsuga menziesii) in microgravity, pp. 847­857, Copyright 2001, with permission from Elsevier. (B) Current Opinion in Plant Biology, vol. 2, Lewis, N. G., A 20th century roller coaster ride: A short account of lignification, pp. 153-162, Copyright 1999, with permission from Elsevier. (C) ACS Symposium Series, vol. 697, Gang, D. R., Fujita, M., Davin, L. B. & Lewis, N. G., The "abnormal lignins": Mapping heartwood formation through the lignan biosynthetic pathway, pp. 389-421, Copyright 1998, with permission from American Chemical Society. (E and F) The American Journal of Botany, vol. 94, Patten, A. M., Jourdes, M., Brown, E. E., Laborie, M.-P., Davin, L. B. & Lewis, N. G., Reaction tissue formation and stem tensile modulus properties in wild type and p-coumarate — 3-hydroxylase downregulated lines of alfalfa, Medicago sativa (Fabaceae), pp. 912-925, Copyright 2007, with permission from the Botanical Society of America.] (Reproduced in color as Plate 19.)
resistance of such tissues to biodegradation (lignocellulose deconstruction) (74). Such sub­stances thus often help confer additional defensive properties to heartwood, making them more resistant to pathogen challenges and helping to increase longevity. Often particu­lar types of defense metabolites (such as lignans) are found in specific tree species, further demonstrating the quite remarkable chemical diversity that has evolved through the phenyl — propanoid pathway, e. g., the lignan — and other metabolite — enrichment of western red cedar (Thuja plicata) contributes to its life span that can potentially exceed 3000 years or so. In other species, such as poplar, a lignan-enriched tissue is also formed, which is sometimes referred to as a “ripewood.”

Lignification/cell wall formation processes are also continuously modulated when woody plant species form branches, are fast-growing, and/or are challenged by having their stems bent (as when growing on a slope). The so-called “reaction wood” formed is trivially known as compression wood (Figure 7.7D) in gymnosperms and tension wood (Figure 7.7E) in angiosperms; Figures 7.7F and 7.7G show “normal” wood for comparison pur­poses. Both types of reaction wood also show variable changes in cellulose content but, interestingly, the H-lignin contents of compression wood are increased while that of overall lignin contents in tension wood are decreased in some species. Furthermore, the means for achieving both stem and branch orientation differ profoundly between gymnosperms and angiosperms; the former produce the lignin-rich reaction (compression) wood on the un­derside of stems/branches (75,76), whereas the latter form the reaction (tension) wood with variable levels of lignin deficiency on the upper-sides of both (30, 72) (discussed below). One purpose of these (re-)orientation mechanisms appears to be to enable maximization of the exposure of the photosynthetic apparatus within the leaves for efficient photosynthetic capture of energy from the sun. The underlying biochemical/molecular mechanisms and reasons for such different lignin/cell wall forming responses to similar mechanical challenges are, however, not yet known. Nevertheless, these examples illustrate some of the quite re­markable changes that plants can undergo in order to obtain either enhanced protection of the lignocellulosic matrix and/or in modifying growth/development through programmed modulation of plant cell wall assemblies.

AND LACCASES

None of these three enzyme classes have any demonstrable role in lignification. The early suggestions that a polyphenol oxidase might be involved emerged from studies by Freuden — berg in 1953, using a press—sap extract of the mushroom, Agaricus campestris (272—277) and also later by Mason and Cronyn (278); mushrooms do not, however, biosynthesize lignins. Another candidate was coniferyl alcohol oxidase, detected in jack pine (Pinus strobus) (268, 269), other Pinaceae species (268, 269, 271) and tobacco (270); this has also not been demonstrated to either afford lignins in vitro and/or have a role in lignification in vivo.

As of 2007, there was still no convincing evidence for any direct involvement of laccase in lignification — in spite of numerous articles (12, 53, 54, 258—266) appearing over a five

decade plus time-frame supporting their involvement in lignin macromolecular assembly. None of these studies, however, met the criteria for monolignol oxidation/lignification as set out by Lewis et al. (31), namely, that: “the enzyme must be able to convert mono-, oligo — and polylignols into their free-radical derivatives; the enzyme must be both tem­porally and spatially correlated with sites of lignin biosynthesis; the enzyme, in the pres­ence of the requisite co-factors and/or other proteins, must be demonstrably capable of converting the monolignols into macromolecular lignin chains; and the enzyme must unequivocally be demonstrated as essential for lignin biosynthesis, e. g., through loss of function.”

The evidence for a role of laccases in lignification was scant indeed: originally proposed by Russian workers in the 1940s (258, 259), this was later investigated further in the labora­tories of Freudenberg (12, 53, 54) and Higuchi (260). While laccases are generally capable of oxidizing monolignols, the experiments were not designed at that time to establish if native lignins were being formed with these catalysts. Later, Harkin and Obst (255) reported the exclusive participation of peroxidases in lignification of trees, using the reagent syringal — dazine (66, Figure 7.16). They also concluded that laccases, for example, were absent in Acer (sycamore) species examined based on a lack of histochemical staining. Laccases were, however, subsequently purified from Acerpseudoplatanus (261,263, 267), and the encoding genes cloned (265), thereby disproving their reported absence. On the other hand, as far as putative lignification was concerned, incubation of monolignols with the Acer laccase(s) (263, 267) in vitro only gave preparations with very minor amounts of 8-O-4r interunit linkages that were presumably not polymeric; that is, such preparations did not reflect lignin structure which predominates in 8- O-4r interunit linkages.

An additional study (264), purportedly detecting a laccase in loblolly pine suggested that laccases were also associated with lignification on the basis of staining with syringaldazine (66) and diaminofluorine (67). Furthermore, reaction rates (as measured by rates of oxygen consumption) for p-coumaryl (1), coniferyl (3), and sinapyl (5) alcohols were 5, 72, and 47 nkat mg-1 protein, respectively, with Km values for each either being unobtainable as for p-coumaryl alcohol (1), or very high 12 and 25 mM for coniferyl (3), and sinapyl (5) alcohols, respectively. Such data prompted Ros Barcelo (280) to comment “With these high Km values, it is difficult to imagine what concentration of cinnamyl alcohols it would be necessary to reach its lignifying cell walls to saturate laccase during the oxidation of cinnamyl alcohols to lignin-like compounds.” These data do not therefore represent proof of laccase involvement in lignification.

Later studies (281), attempting to downregulate the laccase multigene family in poplar (Populus trichocarpa), had essentially no effect on lignin contents as estimated by both acetyl bromide and Klason methods (20-25% of CWR), or on lignin compositions as de­termined by thioacidolysis. These findings again indicated that laccases have no significant and/or direct role in monolignol 1/3/5 oxidation leading to macromolecular lignin assem — bly/configuration, in contrast to more than five decades of scientific contributions (12, 53, 54, 258-266) proposing the contrary view.

Interestingly, Arabidopsis has 17 genes encoding laccases, and each of these has been ex­amined for patterns of gene expression using the GUS-reporter system as before (Turlapati et al., manuscript in finalization): eight of these are expressed in vascular (lignifying) tissue(s), although their physiological roles still need to be defined. One of the laccases apparently has a role in seed coat development, with this presumed to be required for condensed tannin formation (282). Yet, at the time of writing, other researchers still continue to suggest that laccases have a role in seed coat lignification (266), even though seed coat tissue apparently does not form lignin.

8.6.1.1 Thermodynamic integration

Thermodynamic integration is the method of obtaining a Helmholtz free energy change, A F, between two states where the difference is determined from averages that are accessible from MD simulations. If a path between two states can be defined and states along that path can be defined by a parameter, X є (X0, X1), then the following statistical mechanical relationship can be derived,

dFk Id UA dX d X jX

where UX is the potential energy at the state defined by X, and the average is over all configurations visited by the system during an MD simulation of the system with constant X. Then using simple integration schemes, integrating from initial state X0 to final state X1, for simplicity we choose X0 = 0 and X1 = 1,

Although there are some very delicate issues regarding the endpoints of this method, the beauty is that one can run a few points between each of the endpoints, and using Gaussian quadrature, integrate to high precision the averages of the potential energy derivatives to yield the A F. In addition to the simplicity of the method, the ability to choose any path from state 0 to state 1 including creating and annihilating atoms. A typical application is to use a thermodynamic cycle to find the change in binding free energy, AAFbmdmg, for two ligands binding to a substrate without calculating the binding free energy of either one. Free energies of binding are particularly hard to calculate. The method can be illustrated by the following free energy cycle, whose total free energy change will be zero since the beginning and ending states are the same state. In this illustration, Figure 8.4, the two ligands are
phenol and toluene, which differ by the constituent group being hydroxyl or methyl. The question to answer is which binds to cellulose [100] surface more strongly, as judged by the difference in free energy of binding, or AAFbmdmg, or the difference in A F between the binding processes, 4 and 2. The free energy change for the sum of reactions 1 and 2 must be equal to the free energy change for the sum of reactions 4 and 3 since they have the same initial and final states. This relationship is expressed, for the reactions going in the directions indicated in the figure,

A F4 + A F3 = A F1 + A F2

which can be rearranged to yield the difference in binding free energies desired,

AF4 — AF2 = AF1 — AF3 = AAFbinding (8.9)

With this relation, we can determine the difference between two processes that are very hard to compute using two processes that are straightforward and relatively easy. In particular, both processes involve only changing CH3 to OH, and process 1 does not even involve the cellulose since there is no change in its solvated state in process 1. Calculation of the
significant entropic contribution to the binding process is eliminated in this method since the difference of the two binding processes largely cancels out the entropic contributions. The simulations do not have to correctly simulate the large changes associated with desolvating both the cellulose and the ligand, only the small solvation and structural entropy changes associated with changing a small functional group.

There are 14 cellulase families listed on the CAZy web site: (http://afmb. cnrsmrs. fr/CAZY/ fam/acc_GH. html). Several of these families, 10, 26, 51, 74, mainly contain other types of glycosyl hydrolases, with only a few members having cellulase activity but, even if these families are excluded, there are still ten cellulase families. The enzymes in any given family show significant sequence homology with some or all of the other family members. All members of a glycosyl hydrolase family have the same basic protein fold and utilize the same catalytic mechanism but their substrate specificities can be quite different. Because there is a wide range of amino acid sequences that can give the same protein fold, several families share the same fold, even though the sets of sequences in each family show little similarity between the families.

There are nearly twice as many cellulase families, as are present in the next largest group of hydrolases, the seven xylanase families. Furthermore, there are seven different protein folds among the known cellulase structures, and a structure is not yet known for one family. There are two possible reasons why there is so much cellulase diversity. One is that the actual substrate of most cellulases is not pure cellulose but rather plant cell walls, which are extremely diverse and complex, containing many other components, some of which are bound to the cellulose fibrils (21). The other reason is that cellulose itself is quite complex with both crystalline and amorphous regions. It appears that cellulases are under positive selection, as when the DNA and protein sequences of two related cellulase genes were compared there were nearly as many DNA changes that caused an amino acid change (nonsynonymous) as there were DNA changes that did not change the amino acid (synonymous) (22).

There are three functionally different types of cellulases: endocellulases, also called en — doglucanases, exocellulases, also called cellobiohydrolases, and processive endocellulases, which were discovered later (23). To completely hydrolyze cellulose to glucose, a fourth enzyme, p glucosidase, is required, which hydrolyzes the soluble oligosaccharides produced by the cellulases to glucose. Many aerobic fungi secrete a p glucosidase as part of their crude cellulase, while most cellulolytic aerobic bacteria do not, and their p glucosidases are usually cytoplasmic. Some organisms, mainly anaerobic bacteria, contain cellobiose phosphorylase, also called dextrin phosphorylase, which converts cellobiose and soluble dextrins to glucose and glucose-1-phosphate, conserving the energy in the cellobiose linkage (24). All endo — cellulase CDs, whose structures have been determined, have an open active site, as would be expected, since they are able to bind to the interior of long cellulose molecules (25). In contrast, all exocellulases have their active sites in a tunnel, consistent with their processive activity (26). In the case of glycosyl hydrolase family GH-48 enzymes, only part of the active site is in the tunnel, but these enzymes are just as processive as family GH-7 enzymes, where the entire active site is in the tunnel (27). There are two classes of exocellulases (28); one class attacks the nonreducing end of a cellulose molecule and all known members of this class are

Glc(+1)

Thermobifida fusca Cel6AD117A

Figure 11.1 Model of the three-dimensional structures of the catalytic domains of the endocellulase, T fusca Cel6A and the exocellulase, T. reeseiCel6A.

in family GH-6. Members of the other class attack the reducing end of a cellulose chain and all aerobic fungal members of this class are in family GH-7, while the bacterial members are in family GH-48 (see Figure 11.1). It is interesting that the anaerobic fungal members of this class are in family GH-48, rather than in family GH-7 (29). All exocellulases act processively, sequentially cleaving cellobiose residues from a cellulose molecule, so that they are also called cellobiohydrolases. It has been claimed that the T. reesei exocellulase, Cel6A, can act as an endocellulase and that is the reason it can synergize with T. reesei exocellulase, Cel7A (30); however, it has been shown that all of the hydrolysis in a synergistic mixture of these two enzymes results from exocellulolytic activity (20).

There are a number of claims in the literature that specific enzymes are exocellulases, when they are actually endocellulases. In particular, Clostridium thermocellum, CBHA (31) is clearly an endocellulase, as shown by the open active site seen in its X-ray structure (32) and this was confirmed by a set of assays, which showed that it behaved like an endocel — lulase in three different assays: higher activity on CMC then other substrates, reducing the viscosity of CMC and producing 40% insoluble reducing sugars from filter paper, while exocellulases produce from 5 to 8% insoluble reducing sugars from filter paper (33). An­other example is Cel6A from the anaerobic rumen fungus, Neocallimastix patriciarum, which has very high activity relative to other family GH-6 exocellulases but not relative to family GH-6 endocellulases (34). By all the above tests, this enzyme turned out to be a true endocellulase (Wilson, D. B., unpublished). It is often stated that an enzyme is an exocellulase because it produces cellobiose as its major soluble product, but this is true of many endocellulases. Some workers have claimed that only exocellulases have activity on para-nitrophenyl-^-cellobioside but that is not true, as many endocellulases hydrolyze this substrate.

All well-documented processive endocellulases are in family G-9, which is the largest cel­lulase family and includes most plant cellulases, animal cellulases, many bacterial cellulases

and surprisingly, very few fungal cellulases (see Figure 11.2). Processive endoglucanases have an open active site cleft like all endocellulases but in addition they contain a family 3 CBM, which is rigidly attached to the C-terminus of the CD (35). The two domains are oriented so that a cellulose chain can bind simultaneously to both domains. The family 3c CBMs, that are present in processive endocellulases, differ from families 3a and 3b CBMs, in that they lack the conserved aromatic residues, which cause the high affinity for cellulose. Although the 3c CBMs bind very weakly to cellulose, it has been shown that they are necessary for the processive activity of these enzymes (36).

There do not appear to be major differences between the CD families of the cellulases present in cellulosomes and the families of cellulases secreted by aerobic microorganisms, as most cellulase families contain cellulases from both types of microorganisms. However, all known GH-7 cellulases are from aerobic fungi or termites, and there are no known GH-6 cellulases produced by anaerobic bacteria. Furthermore, all GH-12 cellulases appear to be produced by aerobic microorganisms, but this is currently a small family that only contains endocellulases. At this time, all known GH-48 cellulases are exocellulases and this is the only cellulase family that does not contain endocellulases.

There have been a number of studies that analyzed the properties of the cellulose that remained after significant hydrolysis had occurred by a pure cellulase, to try to identify the preferred sites of attack for that cellulase; i. e., amorphous or crystalline regions, as well as how the enzyme has changed the average chain length of the cellulose. A study of T reesei Cel7A, an exocellulase, and T. reesei Cel7B, an endocellulase, found as expected that Cel7A did not cause large changes in the cellulose chain length while Cel7B did (37). A study of four Cellulomonasfimi cellulases acting on Sigma cellulose found that Cel5A rapidly reduced the chain length but Cel6A had a lesser effect on chain length, even though it also is an en — docellulase. Both enzymes increased the crystallinity of the residual cellulose, suggesting that they preferentially degrade amorphous regions in the cellulose. The two exocellulases tested, Cel6B and Cel48A, had no effect on chain length and Cel6B increased crystallinity of the residual cellulose while Cel48A decreased its crystallinity (38). Four synergistic mixtures were tested and none of them caused significant differences in crystallinity. Another study of comparable enzymes from T. fusca showed that both the endocellulase Cel5A and the ex — ocellulase Cel6B primarily digested amorphous cellulose, while the processive endocellulase Cel9A digested both types of cellulose (39).

For reasons similar to liquid hot water percolation processes, dilute acid processes that employ a percolation mode of operation have also been investigated. Very high yields of monomeric and oligomeric xylose have been obtained in a two-stage percolation pretreat­ment of hardwoods, with high enzymatic hydrolysis yields of the cellulose in the pretreated solids (55). The high digestibility achieved in this approach has been attributed to signifi­cant lignin solubilization and removal from the pretreated solids in the continuously-flowing percolation process.

Kinetic modeling studies and associated experimental work have shown that extension of the percolation concept to a countercurrent contacting of biomass particles with the flowing dilute acid medium can further reduce hemicellulose-derived sugar degradation losses and also produce highly digestible pretreated solids. This is achieved by further reduction of the residence time of released sugars under reaction conditions based on the observed first-order hydrolysis reaction kinetics (56). This concept has also been extended to a full thermochemical hydrolysis of both hemicellulose and cellulose, with much higher sugar yields than traditional batch or co-current two-stage dilute acid hydrolysis processes (56). Such processes are highly attractive from a sugar yield standpoint, but will be difficult to apply commercially due to the high liquid volume requirements and complex large-scale reactor configurations.

Dilute acid percolation and countercurrent processes that use lower liquid volumes yet still achieve the highly digestible pretreated solids attributed to lignin solubilization, have also been investigated. In this approach, pretreatment is conducted in a batch mode, followed by a separation and a limited-volume washing of the pretreated solids prior to cooling below the lignin phase-transition temperature, where re-precipitation of solubilized lignin would be expected to occur. High enzymatic digestibility has been reported using this approach on a yellow poplar hardwood feedstock (57), but further work revealed limited benefit to this approach using corn stover as a feedstock.

5.4.8.1 Apiogalacturonan-galacturonosyltransferase (AP:GalAT)

It is not known whether apiogalacturonan is synthesized on preexisting HG that is synthe­sized by GAUT1 or related GalATs, or whether a unique GalAT is responsible for apiogalac­turonan synthesis. There have been no reports of efforts specifically targeted at identifying the apiogalacturonan:GalAT.

5.4.8.2 Apiogalacturonan-apiosyltransferase (AP:ApiT)

Apiogalacturonan is a substituted galacturonan that is produced in some aquatic mono — cotyledonous plants (188,189) and that consists of HG substituted at O-2 or O-3 with apiose or apiobiose (D-Api/-p-1,3-D-apiose) (188,189). The anomeric configuration of the linkage of apiose to HG may be in the p configuration (189). It is not known whether the same apiogalacturonan:ApiTs synthesize RG-II (see below) and apiogalacturonan. For example, RG-IIhastwo of its four side branches attached to an HG backbone by a p-Api/ linked to the O-2 of HG (158), and thus, the possibility exists that the p 1,2-apiosyltransferase involved in RG-II synthesis may also synthesize apiogalacturonan. In vivo synthesis of apiogalacturonan has been studied in vegetative fronds of Spirodelapolyrrhiza (343) and D-apiosyltransferase activity has been characterized in cell-free particulate preparations from duckweed (Lemna minor) (341). The apiosyltransferase in particulate membrane preparations from Lemna transfers [14C]-apiose fromUDP-[14C]-apiose onto endogenous acceptors. The enzyme has an apparent Km for UDP-apiose of 4.9 ^M and a pH optimum of 5.7 (341). Since, the rate of apiosyltransferase activity increased twofold when UDP-GalA was added to the reaction (341) and the product synthesized in the presence of UDP-GalA bound anion exchange resin more tightly than the product synthesized without UDP-GalA (342), it is likely that the apiosyltransferase transfers apiose onto a growing HG chain. The ApiT has not been purified and the gene has not been identified.