Introduction

The formation of furans from sugars has been known since the early nineteenth century (Dias et al., 2010; Van Putten et al., 2013a, b). Furfural was discovered in 1821 by Dobereiner, by the distillation of bran with dilute sul­furic acid (Kamm et al., 2006; Van Putten et al., 2013b). The resulting compound was first named furfurol (the name comes from the Latin word furfur that means bran cereal, while finishing ol means oil). The furfural molecule has an aldehyde group and a furan ring with aromatic character, and a characteristic smell of almonds. In the presence of oxygen, a colorless solution of furfural tends to become initially yellow, then brown, and finally black. This color is due to the formation of oligomers/ polymers with conjugated double bonds formed by radical mechanisms and can be observed even at concen­trations as low as 10~5 M (Zeitsch, 2000a). Despite the fact that furfural has an LD50 between 50 and 2330 mg/ kg for mice, rats, guinea pigs and dogs, man tolerates its presence in a wide variety of fruit juices, wine, coffee and tea (Zeitsch, 2000a; Hoydonckx et al., 2007). The highest concentrations of furfural are present in cocoa and coffee (55—255 ppm), in alcoholic beverages (1—33 ppm) and in brown bread (26 ppm) (Zeitsch, 2000a). There is no commercially attractive route for the production of furfural from petrochemical resources (Mamman et al., 2008). The synthesis of HMF from biomass was already described in 1895 by Dull (1895) and Kiermayer (1895). Due to its high potential as a plat­form chemical for a variety of applications, furfural and HMF were mentioned by Bozell in the "top 10 + 4" list ofbiobasedchemicals (Bozell and Petersen, 2010), along with 2,5-furandicarboxylic acid (FDCA), which is formed by oxidation of HMF (Van Putten et al., 2013a).

The formation of furans from sugars takes place through an acid-catalyzed dehydration of sugar mole­cules at elevated temperature. In general furfural is formed from C-5 sugars and HMF is formed from C-6 sugars. It is therefore not surprising that furans, especially HMF, can be found in essentially all carbohy­drate containing heat-treated food. Furfural is known to have some toxic effects, whereas for HMF it is still un­clear (Van Putten et al., 2013a). The hydrolysis of poly­saccharides and subsequent dehydration into furfural and HMF may be promoted by Bransted or Lewis acid catalysts (Dias et al., 2010; Van Putten et al., 2013a). Furfural production through traditional processes is accompanied by acidic waste stream production and high energy consumption. Marcotullio and de Jong state that modern furfural production process concepts will have to consider environmental concerns and energy requirements besides economics moreover will have to be integrated within widened biorefinery concepts (Marcotullio and de Jong, 2010). The industrial use of aqueous mineral acids as the catalysts, such as sulfuric acid for furfural production, poses serious operational (corrosion), safety and environmental problems (large amounts of toxic waste). Hence, it is seen desirable to replace conventional aqueous mineral acids by "green" nontoxic catalysts for converting sugars into furfural and HMF. The use of solid acids as catalysts may have several advantages over liquid acids, such as easier separation and reuse of the solid catalyst, longer catalyst lifetimes, toleration of a wide range of temperatures and pressures, and easier/safer catalyst handling, storage and disposal. A road map to furfural, HMF and levulinic acid has recently been presented by the group of Dumesic (Wettstein et al., 2012).

Furfural Production and Applications

The industrial production of furfural was driven by the need of the United States to become self-sufficient during the First World War. Between 1914 and 1918, intensive exploration for converting agricultural wastes into industrially more valuable products was initiated. In 1921, the Quaker Oats company in Iowa initiated the production of furfural from oat hulls using "left over" re­actors (Zeitsch, 2000a). Over time, there was an increased industrial production of furfural and the discovery of new applications. Nowadays, the annual world produc­tion of furfural is about 300,000 tons and, although there is industrial production in several countries, the main

production units are located in China, the Dominican Republic and South Africa (Kamm et al., 2006; Zeitsch, 2000a; Hoydonckx et al., 2007; Mamman et al., 2008).

Figure 17.4 gives an overview of some of the main out­lets of furfural. Most of the furfural produced worldwide is converted through a hydrogenation process into fur — furyl alcohol, which is primarily used as foundry resin but also increasingly applied as resin to improve wood durability and for the manufacturing of polymers and plastics (Dias et al., 2010). The aldehyde group and furan ring furnish the furfural molecule with outstanding prop­erties for use as a selective solvent (Zeitsch, 2000a; Hoydonckx et al., 2007; Sain et al., 1982). Furfural has the ability to form a conjugated double bond complex with molecules containing double bonds, and therefore is used industrially for the extraction of aromatics from lubricating oils and diesel fuels, or unsaturated com­pounds from vegetable oils. Furfural is used as a fungicide and nematocide in relatively low concentrations (Zeitsch, 2000a). Additional advantages of furfural as an agrochem­ical are its low cost, safe and easy application, and rela­tively low toxicity to humans. Nakagawa and Tomishige (Nakagawa and Tomishige, 2012) have recently reviewed

the catalyst system used to produce 1,5-pentanediol from tetrahydrofurfuryl alcohol. Other furan compounds ob­tained from furfural include levulinic acid (Gurbuz et al., 2012) and tetrahydrofuran. Furfural and many of its derivatives can be used for the synthesis of new poly­mers based on the chemistry of the furan ring (Hoydonckx et al., 2007; Sain et al., 1982; Win, 2005; Gandini and Belga — cem, 1997; Moreau et al., 2004). Furfural derivatives are also excellent starting points for fuel applications (Lange et al., 2012; Gruter and de Jong, 2009; de Jong et al., 2012a, b). Commercially, the pentosans (mainly xylan) pre­sent in the hemicellulose fraction of agricultural streams such as corn cobs and sugarcane bagasse are hydrolyzed, using homogeneous acid catalysts in water, giving rise to pentose (xylose), which, by dehydration and cyclization reactions, leads to furfural with a theoretical mass yield of approximately 73% (Scheme 17.1). Nowadays also other feedstocks are considered. Huber and his group developed a new process to produce furfural from waste aqueous hemicellulose solutions from the pulp and paper and cellulosic ethanol industries using a continuous two- zone biphasic reactor (Xing et al., 2011). A two-stage hybrid fractionation process was investigated to produce

cellulosic ethanol and furfural from corn stover. In the first stage, zinc chloride (ZnCl2) was used to selectively solubi­lize hemicellulose. During the second stage, the remaining solids were converted into ethanol using commercial cellulase and fermentative microorganisms. Yoo et al. found that the furfural yield from the hemicellulose hy­drolysates could be up to 58% based on carbon (Yoo et al., 2012). Yemis and Mazza researched the potential of a microwave-assisted process that provided a highly efficient conversion of wheat straw, triticale straw, and flax shives: obtained furfural yields based on carbon were 48%, 46%, and 72%, respectively (Yemis and Mazza, 2011, 2012). Sahu and Dhepe also presented a solid acid — catalyzed one-pot method for the selective conversion of solid hemicellulose without its separation from other lignocellulosic components, such as cellulose and lignin resulting in 56% furfural yields in biphasic systems (Sahu and Dhepe, 2012). An interesting approach was dis­closed by vom Stein and coworkers (vom Stein et al., 2011) by working with "real samples". They prepared aqueous solutions of FeCl3—NaCl (or seawater) to evaluate the dehydration of xylose into furfural, which can be extracted in situ into 2-methyltetrahydrofuran (2-MTHF) as second phase. Furfural was also successfully obtained when aqueous nonpurified xylose effluents directly from lignocellulose fractionation are tested (vom Stein et al., 2011). Also Marcotullio and De Jong observed good results with FeCl3 (Marcotullio and De Jong, 2010).

The hydrolysis of pentosans into pentoses in the presence of H2SO4 is faster than the dehydration of the pentose monomers into furfural (Zeitsch, 2000a; Hoy — donckx et al., 2007). Hence, kinetic studies are generally focused on the rate-limiting process, i. e. the dehydration of pentoses. Xylose and arabinose are monomers found in pentosans, which can be converted into furfural, and some studies have shown that the dehydration of arabi — nose is slower than that of xylose (Zeitsch, 2000a; Kootstra et al., 2009). The concentration of xylose in the various raw materials is almost always much higher than that of arabinose. Considering these factors, it seems reason­able to investigate the kinetics of the dehydration process using xylose as substrate (Zeitsch, 2000a; Sain et al., 1982; Win, 2005; Gandini and Belgacem, 1997; Moreau et al., 2004,1998; Antal et al., 1991; Root et al., 1959). In the dehy­dration and cyclization of xylose into furfural, three mol­ecules of water are released per molecule of furfural produced. Huber and coworkers developed a kinetic model for the dehydration of xylose to furfural in a biphasic batch reactor with microwave heating (62). There are four key steps in their kinetic model: (1) xylose dehydration to form furfural, (2) furfural reaction to form degradation products, (3) furfural reaction with xylose to form degradation products, and (4) mass transfer of furfural from the aqueous phase into the organic phase (methyl isobutyl ketone (MIBK)). It was estimated that furfural yields in a biphasic system can reach 85%, whereas at these same conditions in a monophase system furfural yields of only 30% are obtained (Weingarten et al., 2010). Also a kinetic model for the homogeneous conversion of D-xylose in high-temperature water was developed (Kim et al., 2011). Experimental testing evalu­ated the effects of operating conditions on xylose conver­sion and furfural selectivity, with furfural yields of up to 60% observed. Also the kinetics of formic acid-catalyzed xylose dehydration into furfural and furfural decomposi­tion was investigated using batch experiments within a temperature range of 130—200 °C (Lamminpaa et al.,

2012) . The study showed that the modeling must account for other reactions from xylose besides dehydration into furfural. Moreover, the reactions between xylose interme­diate and furfural play only a minor role and that furfural decomposition reactions must take the uncatalyzed reac­tion in water as solvent into account (Lamminpaa et al., 2012). By-products formed in the xylose reaction may also derive from the fragmentation of xylose, such as glyc — eraldehyde, glycolaldehyde, formic acid, lactic acid, ace — tol (Antal et al., 1991; Ahmad et al., 1995).

As furfural is formed it can be transformed into higher molecular weight products by (1) condensation reactions between furfural and intermediates of conversion of xylose to furfural (and not directly with xylose) and (2) furfural polymerization (Zeitsch, 2000a). Aldol condensation between two molecules of furfural does not occur due to the absence of a carbon atom in Ha position in relation to the carbonyl group (Chheda and Dumesic, 2007). The side reactions (1) and (2) lead to olig­omers and polymers with (1) are considered to be more relevant than (2), although published characterization studies of the by-products formed are scarce (Zeitsch, 2000a). The extent of these side reactions can be mini­mized by reducing the residence time of furfural in the reaction mixture and by increasing the reaction tempera­ture (Zeitsch, 2000a, b; Root et al., 1959; Zeitsch, 2000b). If furfural is kept in the gas phase during the aqueous phase reaction it will not react with intermediates, which are "nonvolatile". Agirrezabal-Telleria et al. (Agirrezabal- Telleria et al., 2011) developed new approaches for the pro­duction of furfural from xylose. They propose to combine relatively cheap heterogeneous catalysts (Amberlyst 70) with simultaneous furfural stripping using nitrogen un­der semibatch conditions. Nitrogen, compared to steam, does not dilute the vapor phase stream when condensed. This system allowed stripping 65% of the furfural con­verted from xylose and almost 100% of selectivity in the condensate. Moreover, high initial xylose loadings led to the formation of two water—furfural phases, which could further reduce purification costs. Constant liquid—vapor equilibrium during stripping could be maintained for different xylose loadings. The modeling of the experi­mental data was carried out in order to obtain a liquid—va­por mass transfer coefficient. This value could be used for future studies under steady-state continuous conditions

in similar reaction systems (Agirrezabal-Telleria, 2011). Formic acid, a by-product of furfural process (Root et al., 1959), can be an effective catalyst for dehydration of xylose into furfural. There is a growing interest in the use of for­mic acid as catalyst because it has low corrosiveness and can be easily separated and reused. Using response sur­face methodology the optimal process parameters (xylose concentration 40 g/l, formic concentration 10 g/l, and a reaction temperature 180 °C) were determined to obtain high furfural yield and selectivity. Under these conditions, a maximum furfural yield of 74% and selectivity of 78% were achieved (Yang et al., 2012). Extraction using super­critical CO2 (scCO2) also enhances furfural yields (Kim et al., 2011; Sako et al., 1991,1992). The above mechanistic considerations for the homogeneous conversion of xylose into furfural using H2SO4 as catalyst may also be consid­ered for solid acid catalysts. Nevertheless, differences in product selectivity between homogeneous and heteroge­neous catalytic processes are expected due to effects such as shape/size selectivity, competitive adsorption (related to hydrophilic/hydrophobic properties), and strength of the acid sites.

Industrially, furfural is directly produced from the lignocellulosic biomass in the presence of mineral acids, mainly sulfuric acid, under batch or continuous mode operation (Table 17.9). Attempts to improve furfural yields have been made by process innovation, although the use of mineral acids remains a drawback (Zeitsch, 2000a, 69. 70). The cost and inefficiency of separating these homogeneous catalysts from the products makes their recovery impractical, resulting in large volumes of acid waste, which must be neutralized and disposed off. Other drawbacks include corrosion and safety problems. The production of furfural is therefore one of many industrial processes where the reduction or replacement of the "toxic liquid" acid catalysts by alter­native "green" catalysts is of high priority. Recently Mar — cotullio and De Jong (Marcotullio and de Jong, 2010,

2011) shed new light on some particular aspects of the chemistry of D-xylose reaction to furfural. Their aim was to clarify the reaction mechanism leading to furfural

and to define new green catalytic pathways for its pro­duction. Specifically, their objective was to reduce the use of mineral acids by the introduction of alternative catalysts, e. g. halides, in dilute acidic solutions at tem­peratures between 170 and 200 °C (Schadel et al.,

2010) . Results indicate that the Cl — ions promote the for­mation of the 1,2-enediol from the acyclic form of xylose, and thus the subsequent acid-catalyzed dehydration to furfural. For this reason the presence of Cl- ions led to significant improvements for H2SO4 catalyzed reactions. The addition of NaCl to a 50 mM HCl aqueous solution gave 90% selectivity to furfural. Follow-up experimental results by the same group show the halides to influence at least two distinct steps in the reaction leading from D-xylose to furfural under acidic conditions, via different mechanisms. The nucleophilicity of the halides appears to be critical for the dehydration, but not for the initial enolization reaction. By combining different halides syn­ergic effects become evident resulting in very high selec — tivities and furfural yields (Marcotullio and de Jong,

2011) . Also Rong et al. (2012) found that the addition of inorganic salts (e. g. NaCl, FeCls) promoted the yield of furfural from xylose. Another approach to reduce the inorganic waste streams is to perform the reaction at high temperatures. It was shown that the reaction pathway for the xylose decomposition in high — temperature liquid water can be changed by manipu­lating the temperature and pressure without any catalyst with a maximum furfural yield of 50% (Jing and Lu, 2007). Many attempts have been made to develop heterogeneous catalytic processes for furfural production that offer environmental and economic benefits, but to the best of our knowledge none has been commercialized (Van Putten et al., 2013b).

PGA like the other poly(amino acid)s is a degradable polymer. The polymer can withstand temperatures up to 60 ° C, beyond which the amide bonds start getting hy­drolyzed. PGA is resistant to proteases that cleave alpha peptide bonds. Two types of enzymes are involved in the degradation of PGA, endo-g-glutamyl peptidase and exo-g-glutamyl peptidase. Exo-g-glutamyl pepti­dase consists of two subunits and is a key enzyme in glutathione metabolism (Ogawa et al., 1991; Xu and Strauch, 1996). This enzyme catalyzes the formation of g-glutamic acid di — and tripeptides in vitro. Endo-g — glutamyl peptidase is secreted into the medium by g-PGA-producing B. subtilis and B. licheniformis. It subsequently cleaves high molecular weight g-PGA into fragments as small as 105 Da (Goto and Kunioka, 1992). Attempts to isolate microbes that can utilize PGA as the sole source of carbon and nitrogen source were also successful (Obst and Steinbuchel, 2004).

APPLICATIONS OF PGA

PGA has been a keen interest of research as far as ap­plications are concerned and hence a plethora of appli­cations for this biopolymer has been developed. PGA has been used in the food industry as an additive to flour to increase the moisture-retaining capacity of the dough, as well as to improve the texture and shelf life of bread. The calcium salt of g-PGA can be added to health food in order to increase the Ca2+ concentration, thus contributing to the prevention of osteoporosis (Ashiuchi et al., 2004). Addition of PGA improved the solubility and hence the availability of vitamins and also caused sustained release of these vitamins, which led to increased absorption of these vitamins. PGA salts are known to be used as antifreeze agents in food. The anti­freeze action of the salt increases with the decreasing size of the salt of the polymer (Shih et al., 2003). PGA has been suggested for water treatment, as PGA com­plexes with a lot of metal ions, like Ca+2, Fe+3, Al+3, etc. (Kunioka, 2004). Esters of PGA have been used to test the ability of PGA to form bioplastics with required properties (Kubota et al., 1995). Hydrogels that can be used for applications such as controlled drug release, biosensors, diagnostics, and bioseparators can be pro­duced by using g-PGA and poly(ethylene glycol — methacrylate (Yang et al., 2002). PGA has been used as adjuvants in vaccines, and also as a delivery agent for hydrophobic drugs, increasing their bioavailability. PGA has also been used as a medical adhesive for surgi­cal wounds (sutureless wound closure). PGA hydrogels can be used as three-dimensional scaffolds for tissue en­gineering (Matsusaki et al., 2005).

Originating in an environment without available fixed carbon, cyanobacteria have evolved as versatile organisms, capable of producing a large variety of organic compounds from simple inorganic sources that can be directly used or transformed into a commercial product. When the desired molecule is not naturally produced, genes or entire pathways can be introduced through a variety of methods and product yields can be increased by driving cell metabolism toward the desired product. There are more than 3350 species of cyanobacteria already described, with hundreds avail­able in culture collections (Guiry and Guiry). To date, 87 cyanobacterial genomes have been sequenced and depos­ited in public databases but only a few strains have been used in genetic manipulation studies (Heidorn et al.,

2011) . Many molecular tools are currently available and genetic manipulation can be pursued through conjuga­tion, electroporation or natural transformation. These techniques are constantly being revised or optimized for each host species and sample protocols are available elsewhere (Heidorn et al., 2011). So far, no cyanophage able to perform transduction has been described, never­theless this technique is still the object of great interest (Koksharova and Wolk, 2002).

Natural transformation is an appealing feature found in some cyanobacterial strains, with two standing out as being frequently used in genetic manipulation studies, Synechocystis sp. PCC 6803 (Pasteur Culture Collection) and Synechococcus sp. PCC 7002 (Grigorieva and Shesta­kov, 1982). These two strains are of significant interest due to the high yield of mutants achieved through this technique, making it widely used for both pure and applied science, from plant physiology studies to meta­bolic engineering aiming for the commercial production of biomolecules.

The high frequency of transformants with natural transformation is intimately linked with the nature of the transferred genetic material, with chromosomal DNA reaching up to 100-fold more viable transformants than when replicative plasmids are used as the source of DNA (Golden and Sherman, 1983; Shestako and Khyen, 1970). In fact, this is true specifically for replicative plas­mids since most of the transformation efficiency is recovered when a suicide plasmid is used (Tsinoremas et al., 1994). Thus, it would seem that the final localiza­tion of the inserted DNA plays a key role in the transfor­mation efficiency. This is argued to be related to the postreplicative processing of chromosomal DNA together with a putative robust recombination mecha­nism in these species (Flores et al., 2008). Natural trans­formation has being reported to be associated with pilus-related genes (Yoshihara et al., 2001; Yura, 1999), a natural machinery putatively adapted to take up exog­enous DNA with such high efficiency that different artificial procedures intended to increase the transfor­mation yield fail to improve the frequency of viable mutants (Zang et al., 2007). Unfortunately, natural trans­formation is not widespread in the cyanobacterial phylum and many species require other techniques for the efficient introduction of exogenous DNA.

Electroporation was first demonstrated in Anabaena sp. (Thiel and Poo, 1989) and today has been optimized for many strains. It has been shown to be effective despite the low yield in many cases (Koksharova and Wolk,

2002) . Unlike what is observed for green algae (Kilian et al., 2011), the procedures and electric pulse settings are not very different from those used with other bacte­rial phyla (Heidorn et al., 2011). However, even though it can be an effective method, the ease of natural transfor­mation and the higher yield of conjugation have left electroporation behind as a choice for mutagenesis.

Conjugation is the most commonly used technique for genetic engineering in terms of the diverse species with which it can be used, and, with the filamentous N2 fixing (heterocyst forming) cyanobacteria, it is the only effective technique thus far described. With the advent of molecular biology, plasmids of cyanobacterial origin were actively sought with the intention of produc­ing shuttle vectors allowing their transfer from E. coli to Synechococcus (Golden and Sherman, 1983). Since then,

E. coli has been widely used for conjugation with many filamentous strains, such as Nostoc sp. and Anabaena sp., and single cell strains, like Synechococcus sp. and Synechocystis sp. Although incorporation of DNA into the chromosome of many strains has proved to be relatively easily achieved when using linear DNA or suicide plasmids, it has proved challenging to make cyanobacteria harbor replicative plasmids. During conjugation, the plasmid is relaxed and single-stranded DNA is driven to the recipient cell through the type four secretion system by the enzyme relaxase. Once in the recipient cell, the transferred DNA will have its anti­sense strand resynthesized and this newly reformed plasmid can integrate itself into the genome or autore­plicate. The vectors used in cyanobacteria must contain the replicons for both organisms, donor and recipient, a mobilization site (origin of transfer, e. g. bom, nic and oriT), a selective marker effective for both organisms, and a codon optimization to avoid the broad range of restriction enzymes harbored by cyanobacteria, which has been found to be an important hurdle to successful conjugation (Elhai et al., 1997; Flores et al., 2008; Wolk et al., 1984). Extra enzymes might be needed to ensure a successful transfer, which could be encoded on sec­ondary (aka helper) plasmids. Among these special enzymes are some endonucleases, intended to cut the cargo plasmid at the bom site and promote transfer, and methylases to protect the transferred DNA against the restriction enzymes in the recipient. Detailed proce­dures, strategies and strains used are amply reviewed elsewhere (Heidorn et al., 2011).

Another way to produce bioethanol is using cellulosic materials. Examples of cellulosic materials are bagasse, straw, paper, cardboard, wood and materials of plant cellulosic fibers such hemp, giant reed, eucalyptus tree and Miscanthus. Cellulosic resources are immensely widespread and found abundantly everywhere. These cellulosic materials have the potential to be used for the production of bioethanol since they are not commonly used in the human food chain and exist in large amounts. Moreover, these materials are inexpensive as compared to the sugar and starchy feedstocks and preferably used for bioethanol production. Cellulosic materials are called lignocelluloses because they are composed of lignin, cellulose and hemicelluloses (Kahn et al., 2011).

The cellulosic residue more reported in Table 3.3 for ethanol production is sugarcane bagasse (Dawson and Boopathy, 2007; Santos et al., 2012; Wu et al., 2011; Buaban et al., 2010). This can be explained because of the great use of its juice for sugar ethylic fermentation
and the residue is generated just in the alcohol manufac­ture (Santos et al., 2012). Furthermore, the quantity of this waste available is very great and its direct combus­tion is not the best economical way to use this resource (Wu et al., 2011). The hydrolysis process more employed for saccharification of the bagasse is enzymatic. This is because of the inefficiency of acid hydrolysis in a very complex matrix. Moreover many studies are trying SSF, which can reduce one step of the process.

The agricultural wastes (apart from sugarcane) are studied much for the ethanol production. This is because they are present around the world (comes from diverse agricultural crops). Thus these residues can mean the energetic independence of many countries. Rice plant (Kitamoto et al., 2011) and Lycoris radiata Herbert (Liu et al., 2012) are examples of how much singular are the wastes that the researches are using to produce biofuels. Similarly to sugarcane, Talebnia et al. (2010) have tested SSF and SHF of wheat straw, and obtained good results in the two cases. In Table 3.3 it is possible to see that pretreatment (physical or chemical) is usually necessary, and this is a difficulty in this production.

With more and more bioenergy-related genomes of plants and microbes as well as environmental metage­nomes sequenced, there is an urgent need for automated CAZyme annotation. Although such annotation will not reach a quality as accurate as the expert annotation from CAZyDB, it is expected to be much faster and users can control the annotation at their will. Moreover, nowadays all newly sequenced genomes are relying on generic protein domain/family databases such as Pfam (Finn et al., 2006), InterPro (Hunter et al., 2009), and conserved domain database (CDD, Marchler-Bauer et al., 2009) for automatic genome annotation. Clearly annotation from these databases is often too general and too far from the exact function; the precisely actual function still needs to be determined by experimental approaches. However, most genome annotators are still interested in such genome-scale annotation, as it can give them a quick summary about what the genome encodes, how large the gene families are and how that compares to other genomes.

In fact, even CAZyDB’s manual annotation (assign proteins to existing CAZyme families) on newly sequenced genomes is unlikely to be 100% correct. Considering every new genome contains a high percent­age of proteins that are not experimentally studied, the manual curation is still largely based upon additional bioinformatics analysis such as BLAST search against public protein sequence databases (e. g. UniProt (Bairoch et al., 2005)) and domain databases (e. g. Pfam) and inspection of top matches.

With these in mind, automated CAZyme annotation is still very useful, e. g. particularly for a quick and general overview of how many CAZymes and what CAZymes a newly sequenced genome has. Using the annotated CAZyme proteins and classification scheme in CAZyDB as the foundation, two bioinformatics efforts have been published since 2010, both supporting automated CAZyme annotation, given a protein sequence dataset predicted from a genome/metagenome. The CAZyme

Analysis Toolkit (CAT) (Park et al., 2010) allows a BLAST search against CAZyme proteins annotated by CAZyDB and also a Pfam domain-based search. The simple BLAST search suffers from the inability to accurately annotate the prevalent multidomain CAZyme proteins. The Pfam domain-based search can solve this problem. These Pfam domains are either given by CAZyDB in the CAZyme family Web pages or identified to corre­spond to CAZyme family using an association rule built by CAT. However, there are only 142 (46%) of over 300 CAZyme families linked to Pfam domains by CAZyDB. In fact, many of the Pfam domains were originally created after CAZyDB.

We recently developed dbCAN (Yin et al., 2012) to define a signature domain model for all CAZyme fam­ilies. Aside from the 142 CAZyme families annotated with a Pfam domain, we managed to associate other CAZyme families to functional domains in a broader and general protein domain database CDD. This way, we were able to find a CDD domain for 248 CAZyme families. For the remaining families, we performed a literature curation by reading relevant biochemical papers that are linked to these families by CAZyDB. In the end, we extracted the domain regions in all the mem­ber proteins annotated in CAZyDB and built a multiple sequence alignment (MSA) for each of the CAZyme fam­ily. These MSAs were further processed and represented by hidden Markov models (HMMs), statistical models widely used in the bioinformatics field to represent pro­tein sequence alignments, e. g. by Pfam.

As of June 2012, dbCAN has 320 HMMs representing 317 CAZyme families and three cellulosome modules. We provide all these HMMs freely to the public so that they can run domain-based tool hmmscan of the HMMER 3.0 package (hmmer. org) to annotate their genomes/metagenomes in a local computer, exactly the way that people perform Pfam, InterPro or CDD annota­tion. To help users who do not know how to run hmmscan on a Linux PC, we offer the web server (http://csbl. bmb. uga. edu/dbCAN/annotate. php) so that people can sub­mit their sequences for annotation on the web. The 320 CAZyme family-specific HMMs are our key contribution to the carbohydrate research community and ideally should be included in the general protein domain/family database such as Pfam in the future.

In addition to the Web server, dbCAN also provides a database where precomputed CAZyme homologs in a number of protein databases are showed on the Web. Particularly, starting from the 320 dbCAN HMMs, we scanned public metagenome datasets such as NCBI — env-nr, CAMERA (Seshadri et al., 2007), JGI metagenomes (Markowitz et al., 2012), human gut metagenomes (Meta-HIT) (Qin et al., 2010) and cow rumen metage­nomes (Hess et al., 2011) as well as plant (Goodstein et al., 2012), bacterial and fungal genomes. Tests on

Arabidopsis thaliana (plant) and C. thermocellum (bacteria) using CAZyDB as the positive set suggest that the auto­mated CAZyme annotation achieved a fairly good accu­racy (A. thaliana: sensitivity = 96.3%, precision = 78.8% and average = 87.6%; C. thermocellum: sensitivity = 99.3% and precision = 89.4%). Particularly the sensitivity is over 95% for both organisms, meaning dbCAN annotation tends not to lose true CAZyme proteins.

FOLy Database

Inspired by CAZyDB, Levasseur et al. developed a new database named FOLy, for the classification of ligni — nases in fungi (Levasseur et al., 2008), as these enzymes are critical for breaking down lignins in the biomass but are not included in CAZyDB. Similar to CAZyDB, FOLyDB started from biochemically characterized pro­teins or structures to recruit homologs from GenBank, UniProt and PDB databases. Based on sequence similar­ity, three lignin oxidase families and seven lignin deg­rading auxiliary enzyme families were created, each containing biochemically characterized proteins together with their sequence homologs. Similarly, FOLyDB is featured with expert manual curation of continuingly published literature to include more characterized pro­teins in order to create new families and populate the database. Like CAZyDB, it is not designed for automated genome annotation but BLAST and Pfam domain-based search against annotated proteins in FOLyDB has been widely used to annotate newly sequenced genomes for ligninases.

The choice of cultivation systems is an important aspect that significantly affects the efficiency and cost — effectiveness of a microalgal biofuel production process (Lee, 2001; Pulz, 2001; Carvalho et al., 2006). A wide vari­ation exists among the microalgal cultivation systems for the production of biomass. Raceways, PBRs, and fer­menters, which are the three most widely used microal­gae culture systems, will be discussed below.

OPEN-POND CULTURE

Large-scale cultivation of microalgae in outdoor open-pond systems is well documented (Benemann and Oswald, 1996; Borowitzka, 2005). Open ponds most closely resemble the natural milieu of microalgae. Indeed, ponds can be natural bodies of water, excavated ditches that are unlined or lined with impermeable ma­terials, or they can be constructed above ground with walls (Figure 10.2). Despite a certain variability in shape, the most common technical design for open-pond sys­tems is raceway cultivators driven by paddle wheels and usually operating at water depths of 15—20 cm (Figure 10.1). At these water depths, biomass concentra­tions of up to 1000 mg/l and productivities of 60—100 mg/(l/day), i. e. 10—25 g/(m2/day) are possible. Similar in design are the circular ponds commonly seen in Asia and the Ukraine (Becker, 2007). Such circular ponds usually have the provision of a centrally located rotating arm (similar to those used in wastewater treat­ment) for mixing and may have productivities ranging between 8.5 and 21 g/m2 day (Benemann and Oswald, 1996). On the other hand, thin-layer, inclined ponds consist of slightly inclined shallow trays and may achieve productivities up to 31 g/m2 day (Doucha and Livansky, 2006). Because these ponds are open to the
environment, they are most suitable for algal species that can tolerate extreme environmental conditions (salinity, pH, nutrient loads, etc.) to the exclusion of invasive species. Such algal species include fast growers such as Chlorella, Spirulina, and Dunaliella, which thrive in highly alkaline or saline environments (Chisti, 2007).

Limitations to successful scale-up of microalgae in open-pond systems include contamination, evaporation, limited species suitability, low-volumetric productiv­ities, and the need for large land area.

PHOTOBIOREACTORS

The problems associated with open systems have encouraged the development of closed system PBRs. PBRs can be located indoors under supplemental illumi­nation or outdoors utilizing natural sunlight. Various types of PBRs have been designed depending on growers’ needs; these include tubular PBRs, vertical bubble columns and airlift reactors, combined bubble column and inclined tubular reactors, helical PBRs, and flat-plate PBRs (Tredici and Zittelli, 1998; Sanchez et al., 1999; Berzin, 2005; Ugwu et al., 2005) (Figure 10.3). Closed PBRs allow for tighter regulation and control of nearly all the biotechnologically important parameters

and confer the following fundamental benefits: a reduced contamination risk, reduced CO2 losses, repro­ducible cultivation conditions, controllable hydrody­namics, and temperature (Pulz, 1992). However, widespread implementation has been hampered by the high capital costs associated with PBRs.

Oil bleaching, which is performed in order to prepare a sufficiently light-colored product of enhanced appear­ance and improved stability, is usually achieved by treating the crude or refined oil with powdered absorbent. These absorbents usually contain a calcium montmorillonite (fuller’s earth) or natural hydrated aluminum silicate (bentonite). Adsorption of color bodies, trace metals and oxidation products, as well as residual soaps and phospholipids remaining after washing neutralized oils takes place, if possible. Acid — activated clays are the major adsorbent used, although active carbons and synthetic silicas are also applied industrially with more specific goals. Thus, active carbons are used specifically to eliminate polycyclic aromatic hydrocarbons from some oils, especially fish oils and pomace oils, while synthetic silicas are quite efficient in adsorbing secondary oxidation products, phospholipids and soaps (LeiSn-Camacho et al., 2003). There are a number of good sources of material with more detailed descriptions of each process found online at the Lipid Library (Hardwood and Weselake, 2013), in "Proceedings of the World Conference on Oilseed Technology and Utilization" (Applewhite, 1993) and finally in, Edible Oil Processing (Hamm and Hamilton,

2000) .

Transesterification

Despite being energetically favorable, the direct use of plant or other biolipids in fuel engines is problematic as described earlier. Briefly, due to high viscosity (over 10 times higher than diesel fuel) and low volatility, they do not burn efficiently and can form deposits in the fuel injector of diesel engines. Furthermore, acrolein (a highly toxic substance) is formed through thermal decomposition of glycerol. Different ways have been considered to reduce the high viscosity of plant and other biolipids, but the principal method is to engage in chemical transesterification to produce biodiesel, which could be used in the common diesel engine with minor modification.

As mentioned previously, biolipids consist primarily of triglycerides, which are three hydrocarbon chains connected by glycerol. The bonds are hydrolyzed to allow the formation of FFAs, which are mixed and reacted with methanol or ethanol to form methyl (or ethyl) fatty acid esters. The use of methanol (methanol — ysis) is widespread and considered advantageous, as it is cheaper than ethanol (although in Brazil, ethanol 90 is plentiful) and has less azeotrophic qualities (Encinar et al., 2007). The same reaction using ethanol is more complicated as it requires a water-free alcohol, as well as a biolipid with low water content, in order to obtain good glycerol separation. Methanolysis can happen by heating 80—90% methanol with a small amount of cata­lyst. The received biodiesel after methanolysis is FAME and with ethanol to form fatty acid ethyl ester. The use of ethanolysis reaction using bioethanol has been discussed as being possibly more environmentally favorable as it would involve the use of a nonfossil fuel. Apart from this, ethanol is less toxic and slightly increases the cetane number of the biofuel. Although transesterification can proceed in the absence of catalysts, the reaction proceeds much too slowly to be economically viable and thus typically requires an acidic or alkaline catalysis. Among the most commonly used alkaline catalysts in the biodiesel industry are potassium hydroxide (KOH) and sodium hydroxide (NaOH) flakes, which are inexpensive, easy to handle and can be transported and stored easily. For this reason, they are preferred by smaller producers. Alkyl oxide solu­tions of sodium methoxide (NaOCH3) or potassium methoxide (KOCH3) in methanol, which are now commercially available, are the preferred catalysts for large continuous-flow production processes.

In the transesterification process, the effective species of catalysis is the methoxide radicals (CH3OO and the activity of a catalyst depends upon the amount of meth — oxide radicals (Komers et al., 2001a, b). For sodium or potassium hydroxide, the methoxide ion is prepared in situ by reacting methanol with hydroxide, a reaction that will also produce water that remains in the system. Hydrolysis of triglycerides and alkyl esters may occur due to the presence of this water, which further leads to the formation of FFAs and thus to a soap. Saponifica­tion may also occur if a strong base, e. g. NaOH or KOH, is present in the system by reacting with esters and triglycerides directly. All these problems can be avoided completely if sodium and potassium methoxide solutions, which can be prepared water-free, are applied (Singh et al., 2006).

Some layered hydroxides can also undergo isomorphic substitution of hydroxyl groups by other oxo-ions or by water molecules. In the last case, additional anions will be required to neutralize the excess of positive charge in the layers, keeping the cations unaltered, i. e. only divalent cations are present in the layers. The resulting compounds are called layered hydroxide salts. According to this description, LHSs can be classified based on the structure of copper hydroxide nitrate—Cu2(0HLN03—and zinc hydroxide nitrate—Zn5(0H)g(N03L • 2H2O. The general formula for an LHS is M2+(0H)2_x(A”~)x=n$mH20,

where M = Mg, Ni, Zn, Cu, Co and A = N03~, S0|~ e Cl~ (Arizaga et al., 2007).

The layers in the copper hydroxide nitrate structure are formed by octahedrons whose center is occupied by Cu2+ cations and these are coordinated to hydroxyl groups and nitrate ions that have substituted M of the hydroxyl sites. This example is the easiest description of an LHS.

The structure of zinc hydroxide nitrate has two main characteristics. The first is that M of the Zn2+ cations in octahedral coordination with hydroxyl groups migrate out of the layers, leaving empty octahedrons and form­ing tetrahedrons up and down the empty octahedral sites. Then, each layer is formed by zinc cations in octa­hedral coordination with hydroxyl groups and in tetra­hedral coordination, whose base is formed by three hydroxyl groups shared with the main octahedral layer and its apex is occupied by water molecules. The result­ing layers have residual positive charge with com­position [ZnoctZn2etr(OH)g(H2O)2]2+, where oct and tetr indicate octahedral and tetrahedral sites (Stahlin and Oswald, 1971).

The residual positive charge in the layer is neutral­ized with nitrate ions in the interlayer space in a perpen­dicular position to the layers plane. Normally, nitrate ions do not coordinate directly to the cations in the layers; however, part of the nitrate ions can be grafted to the layers by controlling the pH of the synthesis (Arizaga et al., 2008). The stacking of layers is stable because of the numerous hydrogen bonds that are formed between OH groups in the layer, nitrate ions and the interlayer water molecules. Two adjacent layers are shifted by a factor of b/2 along the (001) plane and are stacked along the basal axis.

A lex Berlin1’*, Mikhail Balakshin2

1Novozymes, Protein Chemistry Department, Davis, CA, USA, 2Renmatix, R&D Department, King of Prussia, PA, USA

Corresponding author email: axbl@novozymes. com

OUTLINE

Raw Material Feedstock 315

Technical Lignins: Production, Properties,

and Analysis 318

Comparison of Analytical Methods for Characterization of Technical Lignins 323

Reproducibility of31P NMR Analytical Techniques 323 13C NMR Analysis of Technical Lignins 327

Advanced NMR Methods 330

Molecular Weight Distribution 330

Structure—Properties Correlations in Lignin 331

Technical Lignins: Traditional and Emerging

Applications 332

Traditional Lignin Applications 332

Emerging Lignin Applications 332

Conclusions 333

References 333

Plant oil-based biodiesel processes are commercial­ized (Figure 20.7). Biodiesel productions use mostly extracted plant or vegetable oils, isolated animal fats (from meat or dairy productions), or to less extent recycled cooking oils, and transesterify the triglycerides with short-chain primary alcohols (e. g. methanol or (bio)ethanol) to make fatty acid esters (diesels). The oil extraction from plants (solid—liquid extraction) may allow coextraction of numerous phytochemicals as coproducts. The main coproduct from the transester­ification is glycerol, which can be used directly as solvent, or as feedstock for microbial fermentation production of other biochemicals, such as propane­diol, dihydroxyacetone, succinic acid, polyglycerols or polyhydroxyalkanoate, or for hydrothermal-chemical conversion to H2 (Khanna et al., 2012; Kosmider et al., 2011; Kannan et al., 2012). Grown on glycerol,
astaxanthin production by bacterium Phaffia rhodozyma (X. dendrorhous) or Sporobolomyces ruberrimus (Valduga et al., 2009), b-carotene by B. trispora (Mantzouridou et al., 2008), prodigiosin or other carotenoids by R. glutinis (da Silva et al., 2009), various carotenoids by Rhodosporidium paludigenum (Yimyoo et al., 2011), as well as b-carotene by microalgae Clamidomonas acid — ophila (Langner et al., 2009), astaxanthin by Schizochy- trium sp., and various carotenoids by Thraustochytrium have been demonstrated.