Different pretreatment approaches catalyze biomass hydrolysis and other reactions to var­ious extents. Therefore, the composition of the liquid and solid process streams resulting from different pretreatment approaches can be widely different. For example, dilute sul­furic acid pretreatment processes can solubilize nearly all the hemicellulose but very little lignin or cellulose. Other pretreatment approaches, primarily the alkaline processes, are more effective at solubilizing lignin, but leave extensive amounts of the hemicellulose frac­tion as an insoluble component of the pretreated solids. These factors greatly impact the relative composition of the pretreated solids and the requirements for effective enzymatic saccharification in subsequent processing steps.

While there are considerable economic drivers to reduce the overall severity of the pre­treatment operation (lower reactor materials cost, lower temperature and/or residence time, lower losses of resulting sugars to degradation, lower requirements to adequately “condition” pretreatment hydrolyzates for subsequent fermentation processes), less aggressive pretreat­ment conditions will generally result in less sugar release (primarily from hemicellulose) in the pretreatment step. This will shift more of the hydrolytic sugar production requirement from the pretreatment step to the enzymatic hydrolysis step and will have an impact on the amount and type of enzymes required to achieve high sugar yields from both cellulose and hemicellulose in less severely pretreated biomass.

Recently, greater attention is being given to the understanding and development of hemi — cellulase and other “accessory” enzyme systems needed to effectively debranch and depoly — merize residual insoluble hemicellulose and soluble hemicellulose oligomers resulting from less severely pretreated biomass processes (2). Systematic studies are needed to determine whether augmentation or partial substitution of cellulase activity with various hemicellulase and accessory enzyme activities can improve the cost-effectiveness of biomass conversion processes by increasing sugar yields.

Cellulose hydrolysis can be accomplished by cellulase enzymes acting without cells (e. g., cellulose hydrolysis in SHF), by cellulases acting in the presence of cells but with no cell — enzyme attachment (e. g., SSF), or by cellulases attached to cells (e. g., C. thermocellum in CBP). In the latter case, hydrolysis is mediated by ternary cellulose-enzyme-microbe (CEM) complexes rather than binary cellulose-enzyme (CE) complexes. For anaerobic cellulolytic bacteria, CEM complexes are commonly formed and are thought to be the major agent of cellulose hydrolysis (15). Potential benefits of CEM complexes for cellulolytic microorganisms have been suggested, including preferred access to hydrolysis products and local concentration of cellulases (15, 51-54).

Lu and coworkers (55) investigated such “cell-enzyme synergy” for C. thermocellum. Figure 16.5 shows that cellulose concentration changes are plotted against time for different cases:

• Microbial hydrolysis (CEM) — C. thermocellum culture with 100 mg/L cellulosome, 264 mg/L cell protein;

• Microbial control (CEM but non-active cells by chemical inhibitor) — a C. thermocellum culture with 100 mg/L cellulosome, 264 mg/L cell protein plus 38.5 mM sodium azide, a cell-growth inhibitor;

• CE in SSF — 100 mg/L purified C. thermocellum cellulase and active T. saccharolyticum that assimilates hydrolysis products;

• Cell-free control 1 (CE in SHF), 100 mg/L purified C. thermocellum cellulase; and

• Cell-free control 2 (CE in SHF plus chemical inhibitor), 100 mg/L purified cellulase

C. thermocellum cellulase with 38.5 mM sodium azide.

Comparison of microbial cultures (CEM) and cellulase-enzyme in the presence of T. saccharolyticum (SSF) cultures suggests that the C. thermocellum cellulase complex is substantially more effective during microbial hydrolysis compared to SSF under the con­ditions examined. Such “enzyme-microbe synergy” requires the presence of metabolically active cellulolytic microbes, and is not explained by removal of hydrolysis products from the bulk fermentation broth because (see control experiments).

From an applied perspective, the 2.7- to 4.7-fold synergistic effect reported is significant in the context of the search for strategies to decrease the cost of enzymatic hydrolysis, a focus of considerable effort since the late 1990s (3, 24, 56, 57). From a fundamental perspective, such synergy is one of the important inherent mechanisms to reduce cellulase synthesis requirement for anaerobic cultures.

The current view is that the bulk of pectin synthesis is catalyzed by glycosyltransferases (GTs) that transfer a glycosyl residue from an activated form of the sugar, most likely a nucleotide-sugar, to an acceptor. Like other polymer biosynthetic reactions, pectin syn­thesis is thought to proceed through three stages: initiation, elongation, and termination. There is no detailed understanding of the initiation phase for the synthesis of any of the pectic polysaccharides. It has traditionally been held that the pectic polysaccharides are not synthesized on a protein, thus distinguishing them from the synthesis of animal Golgi — localized proteoglycans (256). More recently, several investigators have presented evidence suggesting that some wall polysaccharide synthesis may occur on protein primers (257­259); however, definitive proof of this hypothesis has yet to be provided. To date all studies of pectin biosynthetic glycosyltransferases have been carried out by assaying the transfer of a glycosyl residue from a radiolabeled nucleotide-sugar substrate (with or without ad­ditional unlabeled substrate) onto endogenous acceptors in plant microsomal membranes or onto exogenous acceptors, or alternatively, by transfer of an unlabeled substrate onto fluorescently labeled exogenously added oligosaccharide or polysaccharide acceptors. Most nucleotide-sugars involved in pectin synthesis consist of a nucleoside-diphosphate (NDP) attached to the sugar, and thus the general reaction catalyzed by the GTs is NDP-sugar + acceptor^) ^ NDP + acceptor(„+i). The precise number and location of the glycosyl residues in the acceptor that are recognized by a GT will be unique for each GT. However, in this review, for the purposes of calculating a minimal number of GTs required for pectin synthesis, since all pectin biosynthetic GTs characterized to date have been shown to add onto the non-reducing end of the oligosaccharide/polysaccharide acceptor, the assumption has been made that each GT will recognize, as a minimum, the terminal two glycosyl residues at the non-reducing end of the acceptor (i. e., the sugar onto which the transfer takes place, and the adjacent sugar).

Table 5.2 lists the types of glycosyltransferases that are expected to be required for pectin synthesis, based on the premise that a unique glycosyltransferase will be required for the transfer of a unique sugar from a unique nucleotide-sugar onto a unique disaccharide acceptor region at the non-reducing end of the acceptor. In some cases, it is possible that the same enzyme may catalyze the synthesis of a similar region on diverse polysaccharides, (e. g., the same galacturonosyltransferase may catalyze the synthesis of the backbone of HG and the HG region of RG-II. Table 5.2 attempts to list all known or expected GTs that are required for pectin synthesis, and as such, is meant to serve as a reference table. However, in an effort to consider in more depth the synthesis of the different types of pectic polysaccharides, HG, RG-I, RG-II, XGA, and AG; a detailed summary of progress in understanding the synthesis of the specific pectic polysaccharides is described separately.

Table 5.2 List of glycosyltransferases expected to be required for pectin biosynthesis Enzymeb acceptor substrate Enzyme Type of glycosyl-transferase Parent polymer3 (side chain) activity (unless noted: enzyme adds to the glycosyl residue on the left*) Ref. c for structure Putative gene Identified (Ref.)d Gene identified (Ref.)e D-GalAT HG/RG-II *GalAa1,4-GalA а 1,4-GalAT (157) Put. QUA1- GAUT1 At3g25140 (135, At3g61130 260) (137) D-GalAT RG-I L-Rhaa1,4-GalA a 1,2-Gal AT (157, 261,262) — — D-GalAT RG-II (A) L-Rha|31,3′-Apif a1,2-GalAT (214, 157) — — D-GalAT RG-I I (A) L-Rha|31,3′-Apif fil,3GalAT (214, 157) — — D-GalAT RG-I/HG GalAal,2-L-Rha а і,4-Gal AT — — — D-GalAT HG attached to *GalAa1,4-GalA а 1,4-GalAT or (129, 263) Put. GAUT12 xylan Xyl-?__ а 1,4-GalAT (Xylpi,4-ХуЫ, IRX8 At5g54690 3-Rhaa1,2-GalAA ,4-Xyl) (GT8) (129, 132) L-RhaT RG-I GalAal,2-L-Rha a1,4c-RhaT (157, 261,262) — — L-RhaT RG-II (A) Apifpi,2-GalA pi, З’-ь-RhaT (214, 157) — — L-RhaT RG-II © Kdo2,3GalA a1,5-c-RhaT (214, 157) — — L-RhaT RG-II (B) L-Araa1,4-Gal a? ,2-c-RhaT (214, 157) — — L-RhaT RG-II (B)f L-Araa1,4-Gal |31,3-c-RhaT (158, 161) L-RhaT HG/RG-I GalAal,4-GalA a1,4-c-RhaT — — — D-GalT RG-I L-Rhaa1,4-GalA |3 1,4-GalT (157, 261) — D-GalT RG-I Gal|31,4-Rha p 1,4-GalT (157, 264) — — D-GalT RG-I Galpi,4-Gal $l,4-GalT (157, 190, 207, — — 264-267) D-GalT RG-I Galpi,4-Gal $1,6-GalT (157, 264) — — D-GalT RG-I/AGP^ Galpi,3-Gal $l,3-GalT (214) — — D-GalT RG-I/AGP Galpi,3-Gal $1,6-GalT (214) — — D-GalT RG-I/AGP Galpi,6-Gal pi,3-Gal p 1,6-GalT (214) — — D-GalT RG-I L-Araf-1,4-Gal 1-5-GalT (268) L-GalT RG-II (A) GlcApi,4-Fuc a? ,2-c-GalT (214, 269) — — D-GalT RG-II (B) c-AcefA a1,3-Rha p 1,2GalT (214, 269) — — L-AraT RG-I Galpi,4-Rha a1,3-c-ArafT (157, 264) — —

Feeding experiments with free d-G1cA showed that plant cells rapidly incorporate the sugar into pectin (459). Free GlcA, likely released from wall polysaccharides, can be activated to UDP-GlcA, by-passing the need for flux of the sugar from the inositol oxidation pathway. Membrane and soluble protein preparations extracted from a 4-day-old etiolated seedlings of mung bean consist of a kinase activity that specifically phosphorylates GlcA in the presence of ATP and Mg2+ to GlcA-1-P. L-Ara and D-Gal are also substrates for this crude kinase preparation. However, GalA is not a substrate for this activity (411). The GlcA-1-P kinase has not been purified and the gene is not yet known. Work in our laboratory (Bar-Peled, 2005) (414) and by Schnurr and coworkers (415) demonstrated that Sloppy (At5g52560) is the true GlcA-1-P pyrophosphorylase and readily converts UTP and GlcA-1-P into UDP-GlcA.

Phenylalanine ammonia lyase (PAL) represents the first committed step to phenylpropanoid metabolism, and thus to a wide range of phenolic products (lignins, lignans, hydroxycin — namic acids, flavonoids, suberins, etc.). Discovered by Koukol and Conn (84) in 1961, this enzymatic conversion required no added cofactor. Moreover, the ammonium ion released during the deamination step had also been subsequently demonstrated to be recycled via GS/GOGAT, with the glutamate so formed then serving as the amino donor for arogenate (35) formation (Figure 7.8) (85-88). In this way, the means to both recycle the nitrogen and sustain phenylpropanoid metabolism can occur without any apparent further need for an additional nitrogen source. A gene encoding a PAL was reported by Edwards et al. in 1985 (89); an earlier description by others (90, 91) of PAL cloning was later shown not to be PAL (89, 92).

In Arabidopsis, there are four bona fide PAL gene family members present [as determined by both analysis of the genome sequence (93), and also by recombinant protein characteri­zation (94)]. In terms of its biochemical mechanism, the X-ray crystal structure of PAL was also recently described (95); additionally, it was deduced that a highly conserved tripep­tide sequence (Ala-Ser-Gly) in PAL undergoes spontaneous dehydration/cyclization to give MIO (3,5 dihydro-5-methylidene-4H-imidazol-4-one) (95) based on earlier studies with histidine ammonia lyase (HAL) (96, 97). MIO is envisaged to serve as an internal cofactor for catalysis. Interestingly, other studies using both tyrosine ammonia lyase (TAL) (from Rhodobacter sphaeroides) (98) and PAL (from Arabidopsis) (99) have demonstrated that the interconversion of PAL to TAL activity can be effectuated by a single amino acid substitu­tion (i. e., H89F converts TAL to PAL in R. sphaeroides, whereas in Arabidopsis, the PAL/TAL switch is F144H).

It is not known with any degree of certainty as to when the biochemical pathway fully evolved to afford either the monolignols or the monolignol-derived products, the polymeric lignins, and the (oligomeric) lignans. It is presumed to have occurred with the emergence of tracheary elements, with the latter being thought to trace their origins back to about 430 million years ago (1-3). Such a lengthy evolutionary period might seem unlikely, however, to ultimately result in a random assembly process for lignins, particularly as they represent Nature’s second most abundant biopolymer(s).

Yet today, there is currently still a vigorous debate regarding native lignin macromolecular configuration, namely, as to whether it results from proteinaceous control over primary structure with (presumed) template replication (29, 31, 50-52, 284-286), or whether the “random coupling” model of Freudenberg (11,12, 53,54) — speculated to afford lignins with 1066 isomers per 100-mer unit (196) — can be extended to that of “combinatorial chemistry” (175). Since such widely divergent viewpoints are apparently incompatible, determining how native lignin macromolecular configuration is unambiguously achieved in vivo is essential for fully understanding the basis of vascular plant cell wall biochemistry. In this context, we considered it instructive to first briefly discuss the ten or so depictions of proposed “representative” lignin structures, together with either the evidence or lack thereof for same, and the gaps in our knowledge that remain to be filled.

A complete discussion of all the issues surrounding the performance of molecular dynamics simulations is beyond the scope of this book and remains an active area of research in its own right. However, this chapter would not be complete without at least some discussion of the typical resource requirements for MD simulations of celluloses and the types of time scales that can be reached.

MD simulations are very computationally intense, typically requiring access to the world’s most powerful supercomputers in order to simulate sufficient timescales (10 ns to 1 ^s) for biological systems (10 thousand to several million atoms). The computational complexity lies in the fact that there are an extremely large number of pairwise interactions that must be calculated at each time step. This would typically be on the order of 20 million or more for a system of 80 000 atoms. Then to access information on a biological timescale, it is necessary to propagate the system through time. As mentioned above, the length of a time step is typically limited to about 2 fs if the motions of all heavy atoms are to be simulated. This means that to obtain 100 ns of trajectory data requires evaluation of 5 x 1010 time steps and each time step requires the 20 million non-bond energy evaluations.

Unlike Monte Carlo simulations, where each individual energy evaluation can be per­formed independently of other calculations, molecular dynamics simulations involve nu­merically solving the integral over time. This means that the next step of the MD trajectory cannot be computed until all previous steps have been computed in order. This makes computation in parallel difficult, requiring extremely low latency interconnects between processors and careful distribution of work. Even then, the distribution of work across mul­tiple CPUs is limited to a single time step. There is a multiple time step method in which the low-frequency motions are partially uncoupled from the high-frequency motions in such a way that low-frequency contributions to the dynamics are not calculated on every step, increasing performance as much as a factor of two. There is significant effort being expended in developing more efficient molecular dynamics software.

Another organism used for industrial cellulase production is Humicola insolens, which seems to produce the same set of cellulases as T. reesei except that it produces a family GH-6 en — doglucanase (68). However, these two organisms are not closely related, even though they are both brown rot fungi, which do not degrade lignin. Phanerochaete chrysosporium is a white rot fungus that degrades lignin, while not degrading much cellulose, despite con­taining a set of cellulase genes (69). A surprising finding is that it contains seven CBH I genes that are differentially regulated, in contrast to the single CBH I gene in T. ree — sei. The cellulases produced by Aspergillus aculeatus have been extensively studied, and it produces nine cellulases of which three have been sequenced: Cel7A, Cel12A, and Cel5A. From this limited data, it seems that its cellulases may be similar to those of T. reesei (70). Another fungus whose cellulases have been studied is Talaromyces emersonii, which pro­duces two exocellulases: Cel7A and Cel6A, which have been extensively studied, and several endocellulases of which only Cel5A has been studied (71). Chrysosporium lucknowense cel­lulases also have been studied and seem to resemble those of T. reesei (72). There are many other cellulolytic fungi whose cellulases have had some research including Agaricus bisporus, several Aspergillus species, Aureobasidium pullulans, Cochlibolus carbonum, several Fusar — ium species, several Penicillium species, Pleurotus ostreatus, and Thermoascus aurantiacus, but the total cellulase system has not been determined for any of these organisms at this time. All these aerobic fungi produce a set of individual cellulases; however, an aerobic Chaetomium strain has been reported to produce a large cellulase complex (73). At this time it is not known what cellulases are in the complex and if the complex is assembled using the cohesin-dockerin binding seen in cellulosomes. This system clearly should be studied further.

Oxidative processes for biomass pretreatment applications are often referred to as wet ox­idation processes. This approach was born out of efforts in the pulp and paper industry to develop oxygen delignification processes to reduce chlorine use in pulping. The most common approach for wet oxidation as a biomass pretreatment involves the injection of pressurized O2 into a pretreatment reactor at temperatures up to 200°C and pressures up to about 1.5 MPa (69). Another approach utilizes a percolation-type pretreatment that incor­porates wet oxidation, among other approaches, into a biomass fractionation process (70). Most work has been performed in batch reactors at low solids loading, occasionally with the use of small amounts of co-catalysts or solvents. Much of this work has included the use of alkaline buffers (usually sodium carbonate) to maintain reaction pH in the neutral to alkaline range. There have been sporadic efforts to investigate alkaline peroxides and ozone as other types of oxidizing agents (71-73).

Wet oxidation extensively delignifies biomass with production of monomeric and oligomeric phenols, followed by oxidative cleavage to a variety of carboxylic acids. When the reaction is not buffered and pH is allowed to drift naturally down, extensive forma­tion of furfurals occurs, which can also be cleaved to form carboxylic acids in the oxidative environment. Hemicellulose is typically solubilized to about 70% conversion, primarily as oligomers. The combination of extensive delignification and at least 50% hemicellulose removal can result in highly digestible pretreated solids (70).

14.5.9 Biological pretreatment

Most biological pretreatment approaches utilize certain classes of lignin-solubilizing mi­croorganisms that will produce a lignocellulosic feedstock that is more amenable to enzy­matic saccharification than native biomass. Many studies have focused on a class of microor­ganisms known as white-rot fungi, which produce lignin-degrading enzymes, lignin perox­idases, peroxide producing enzymes, polyphenol oxidases, laccases, and quinine-reducing enzymes (11, 12, 74, 75). In such biological pretreatment processes, biomass is inoculated with appropriate fungal cultures and incubated for several weeks, followed by evaluation of the enzymatic hydrolyzability of the treated biomass. While increased digestibility has been attributed to the delignification action of these microorganisms (75,76), there is often some parallel loss of cellulose and/or hemicellulose during the biological pretreatment.

Other biological pretreatment approaches have been investigated as a means ofpreparing stored biomass for subsequent mechanical, thermal or chemical pretreatment, potentially reducing the required severity and cost of the pretreatment process. This approach can effec­tively soften especially woody biomass types, reducing the mechanical power requirement for size reduction via disk refining by about 40%, although there is some associated loss of mass due to fungal action of lignin and structural carbohydrates, suggesting a potential loss of available carbohydrates to the conversion process (77).

4.3.2.1 Polysaccharide-polysaccharide cross-linking.

4.3.2.1.1 DIMERIZATION OF ESTERIFIED HCA

In grasses, hydroxycinnamate esters on GAX (see Section 4.2.1.2.2) in primary and lignified secondary walls are covalently cross-linked by oxidative dimerization of HCA units on neighboring AX chains by radical coupling reactions (32, 79) (Figure 4.3a). The homo — and hetero-dehydrodimers formed involve mostly FA and sometimes SA (80); pCA does not appear in these dehydrodimers. Dehydrotrimers (32, 81) and a dehydrotetramer (82) of FA have also been described and could participate in cross-linking. A number of isomeric cross-linking homo — and hetero-dimers have been encountered involving both the aromatic ring and the propenoic acid side chain of the HCA (Figure 4.4) (83, 84).

4.3.2.1.2 ESTERIFIED HCA CYCLODIMERIZATION

In addition to dimerization of ester-linked HCA by oxidative coupling, cross-linking dimers are formed from FA and/or pCA monomers by photodimerization (83). These ester-linked cyclodimers are cyclobutane derivatives (truxillic acids and truxinic acids) (Figure 4.3b). Both the esterified HCA dimers and cyclodimers are readily released by treatment with dilute alkali (0.5 M NaOH, 20°C, 18 h) (85).

4.3.2.1.3 DIRECT ASSOCIATIONS BETWEEN POLYSACCHARIDES

Direct covalent linkages between pectic polysaccharides and xyloglucans in angiosperm walls have been reported (86-88), but the detailed chemistry of the linkage(s) is not known.

Figure 4.3 Three modes of covalent cross-linking involving feruloylated GAXs: (a) a dehydrodiferulate cross-link(5-5), (b)aferulatecyclodimer (truxillicacid) cross-link, and (c)tyrosyl-ferulatecrosslinkbetween a protein and a feruloylated GAX. Ph = 4-Hydroxy-3-methoxy-benzene.