Yield of biodiesel through lipase catalysis is effected by (1) feedstock quality, (2) choice of enzyme (extracel­lular or intracellular), (3) molar ratio (alcohol/oil), (4) temperature, (5) water content, (6) acyl acceptors, (7) sol­vent and (8) reactor system.

FEEDSTOCK

One of the main barriers for commercialization of bio­diesel production is choice and availability of feedstock, which comprise nearly 80% of the overall biodiesel pro­duction cost. Diverse kinds of feedstocks are available such as edible and nonedible vegetable oil, animal fats, waste oil, microbial oil and microalgae oil and they can be used for enzyme-catalyzed transesterification (Sevil et al., 2012).

Vegetable Oils

Vegetable oils are well-known for their high heat con­tent and they are alternative fuels for diesel engines. High viscosity restricts their consumption directly in diesel en­gines, which leads to many problems (Koh and Ghazi, 2011; Singh and Singh, 2010). Most widely used edible vegetable oils in enzymatic transesterification are soy­bean (Wenlei and Ning, 2010; Du et al., 2003), sunflower (Karout and Pierre, 2009), palm (Talukder et al., 2011; Matassoli et al., 2009), corn (Mata et al., 2012), cottonseed (Chattopadhyay et al., 2011), canola (Jang et al., 2012) and olive (Sanchez and Vasudevan, 2006). Higher quality of edible oil is good feedstock to produce biodiesel by enzy­matic transesterification. However, major concern is the economic viability of biodiesel since refined vegetable oils are expensive. Also, use of high-value edible vege­table oil as biodiesel feedstocks has caused food crisis. Furthermore, percentage of oil and yield per hectare are effective parameters in selecting potential renewable feedstock for biodiesel production (Nielsen et al., 2008). Hence, in order to make biodiesel production more economical, low-cost and nonedible oils need to be preferred. Babassu (Orbignya martiana), Jatropha curcas (Linnaeus), neem (Azadirachta indica), polanga (Calophyl- lum inophyllum), karanja (Pongamia pinnata), rubber seed tree (Hevea brasiliensis), mahua (Madhuca indica and Madhuca longifolia), tobacco (Nicotiana tabacum), etc. are most widely used nonedible oil sources for biodiesel production. Biodiesel produced from these nonedible oils meets key specifications of biodiesel as per the stan­dard organization requirements (Mohibbe et al., 2005). All these low-cost feedstocks contain large amount of FFA which leads to undesirable soap formation during traditional base-catalyzed transesterification. However, high free-acid content is not a problem in enzyme transesterification.

Animal Oils/Fats

Animal oils differ from vegetable oil in their fatty acid composition. Vegetable oils have high content of unsat­urated fatty acids (mainly oleic and linoleic acid), while animal fat has higher proportion of saturated fatty acids. Commonly used animal fats for biodiesel production via enzymatic route contains lard (Jike et al., 2007), lamb meet, beef tallow, chicken fat and animal fat mix (Vivian Feddern et al., 2011). Waste animal fats from animal pro­cessing industries and slaughter houses are also a good source for animal fats; however it is a decent alternative instead of their direct dispose in to environment. Their favorable features like noncorrosive nature, high cetane number, and renewable nature makes them a good source for biodiesel production. But their relatively high FFA (5—30%) and water content led to soap forma­tion in chemical transesterification process and their saturated fats prone them to oxidation and crystalliza­tion at high temperatures (Huynh et al., 2011). Removal of contaminants is another problem from animal fats, which generally contain phospholipids, or gums, and cause insoluble precipitates when they come into con­tact with water. Gums are removed by adding water and citric or phosphoric acid to the animal fats followed by centrifugal separation of precipitates. Phospholipids get separated with glycerin during processing, or by wa­ter washing/ion exchange separation. Removal of sulfur contents is also a serious issue. Beef tallow and some chicken fat contain around 100 ppm of sulfur. Vacuum distillation is the only reliable technique for reducing the sulfur level to permissible levels (15 ppm) (Farm energy, 2012).

High oil-yielding transgenic microalgae could be a promising source for biodiesel production. However, the biotechnological processes based on transgenic microalgae are still in infancy. In manipulation of genet­ically modified algae for high oil content, acetyl-CoA carboxylase (ACCase) was first isolated from the diatom Cyclotella cryptica by Roessler (1990), and then success­fully transformed into the diatoms C. cryptica and Navicula saprophila (Dunahay et al., 1995,1996; Sheehan et al., 1998). A plasmid was constructed that contained acc1 gene driven by the cauliflower mosaic virus 35S ribosomal gene promoter (CaMV35S) and the selectable marker nptII from Escherichia coli. Introduction of plasmids into the diatoms was mediated by micropro­jectile bombardment. The acc1 was overexpressed with the enzyme activity enhanced by threefold. These exper­iments demonstrated that ACCase could be transformed efficiently into microalgae, although no significant increase in lipid accumulation was observed in the transgenic diatoms (Dunahay et al., 1995, 1996). Recently, diacylglycerol acyltransferases (DGATs) homologous genes have been identified in the genome of Chlamydomonas reinhardtii and were overexpressed in the same microalga (Russa et al., 2012). This resulted in an enhanced mRNA expression level of DGAT genes, but did not boost the intracellular triacylglycerol (TAG) synthesis. Thus, till date, there is no success story with respect to lipid overproduction in microalgae using the genetic engineering approach.

Extensive studies have also been carried out on enhancement of lipid production using genetic engineer­ing approaches in different bacterial and plant species, which may provide valuable background for future studies with microalgae. Some of these studies are summarized in Table 11.2. The cytosolic ACCase from Arabidopsis sp. was overexpressed in Brassica napus (rapeseed) plastids. The fatty acid content of the

recombinant was 6% higher than that of the control (Roesler et al., 1997). In prokaryotes like E. coli, overex­pression of four ACCase subunits resulted in sixfold rise in the rate of fatty acid synthesis (Davis et al.,

2000) , confirming that the ACCase-catalyzed committing step was indeed the rate-limiting step for fatty acid biosynthesis in this strain. Nevertheless, Klaus et al.

(2004) achieved an increase in fatty acid synthesis and a more than fivefold rise in the amount of TAG in Sola — num tuberosum (potato) by overexpressing the ACCase from Arabidopsis in the amyloplasts of potato tubers.

Transformation of rape seed with a putative sn-2-acyl- transferase gene from Saccharomyces cerevisiae was car­ried out by Zou et al. (1997), leading to overexpression of seed lysophosphatidate acid acyl-transferase (LPAT) activity. This enzyme is involved in TAG formation and its overexpression led to profound rise in oil content from 8% to 48% on seed dry weight basis. However, it was cautioned that the steady state level of diacyl — glycerol could be perturbed by an increase in LPAT activity in the developing seeds. Transformations of S. cerevisiae with the Arabidopsis DGAT were performed
by Bouvier-Nave et al. (2000). About 600-fold rise in DGAT activity in the transformed S. cerevisiae was observed, which led to a ninefold increase in TAG accumulation. DGAT gene has also been overexpressed in the plant Arabidopsis and it was shown that the oil content was enhanced in correlation with the DGAT activity (Jako et al., 2001). All these results suggest that the reaction catalyzed by ACCase, LPAT and DGAT are important rate-limiting steps in lipid biosynthesis.

A few enzymes that are not directly involved in lipid metabolism have also been demonstrated to influence the rate of lipid accumulation. For instance, it was observed by Lin et al. (2006) that by overexpressing the acs gene in E. coli, the acetyl-CoA synthase activity was increased by ninefold, leading to a significant increase in the assimilation of acetate from the medium, which can contribute to lipid biosynthesis. The genes coding for malic enzyme from Mucor circinelloides (mal — EMt) and from Mortierella alpina (malEMc), respectively, were overexpressed in M. circinelloides which led to a

2.5- fold increase in lipid accumulation (Zhang et al.,

2007) . Lu et al. (2008) reported a 20-fold enhancement
of fatty acid productivity of E. coli by combining four targeted genotypic changes: deletion of the fadD gene encoding the first enzyme in fatty acid degradation, overexpression of the genes encoding the endogenous ACCase, and overexpression of both an endogenous thioesterase (TE) as well as a heterologous plant TE. Overexpression of wril gene from B. napus in transgenic Arabidopsis thaliana resulted into 40% increased seed oil content (Liu et al., 2010). Zhang et al. (2011) studied the effects of the overexpression of different acyl-ACP TE genes from Diploknema butyracea, Ricinus communis and J. curcas on free fatty acid contents of E. coli. The strain carrying the acyl-ACP TE gene from D. butyracea produced approximately 0.2 g/l of free fatty acid while the strains carrying acyl-ACP TE genes from R. commu­nis and J. curcas produced the free fatty acid at a high level of more than 2.0 g/l.

Combustion

It is estimated that around 3 billion people worldwide rely on wood, stubble, dung and leaves for cooking fuel. Burning biomass fuels on open fires and in inefficient stoves releases many harmful pollutants. These pollut­ants result in excess respiratory illnesses and death in women and children. Known as a "silent killer", over

1.6 million children die annually throughout the devel­oping world from the consequences of exposure to biomass fuel smoke (Edelstein et al., 2008). Improved stoves reduce the fuel consumption and indoor pollu­tion by 50% (Ravindranath et al., 1997; Halim, 2008). Co-combustion of solid biomass and coal is reviewed by Cremers (2009).

Combustion—steam cycle. This combustion of solid biomass and the use of a steam cycle is not very energy efficient (32%, Yang et al., 2006). The maximum temper­ature is limited as potassium and calcium together with silicon form at high temperatures glasslike deposits on the furnace walls. Corrosion problems occur in straw — based furnaces at 500 °C (Hansen et al., 2000).

Gasification

Gasifying cook stoves are described by Field (2012). A high-pressure liquid ash gasifyer has an efficiency

of 50%. These are large installations with capacities of over 200 MWe. At present only 15% biomass is cogasi­fied with coal (Drift, 2008).

Wet oxidation is a pretreatment technology using water and air or oxygen to fractionate biomass at temper­atures above 120 °C. A clear advantage of wet oxidation, in particular in combination with alkali, is the relatively mild temperature and the limited formation of fermenta­tion inhibitors (e. g. furan aldehydes and phenolalde — hydes) (Klinke et al., 2002). Wet oxidation facilitates the separation of cellulose after the majority of hemicellu — loses and lignin has been solubilized. The amount of lignin removed after pretreatment ranges from 50% to 70% depending on the type of biomass pretreated and the conditions used. The solid material after wet oxida­tion displayed a higher enzymatic convertability than the remaining solid material after steam explosion (Mar­tin et al., 2008). Wet oxidation is effective in pretreating a variety of biomass such as wheat straw, corn stover, sug­arcane bagasse, cassava, peanuts, rye, canola, faba beans, and reed (Brodeur et al., 2011; Martin et al., 2008). Wet oxidation can be combined with other pretreatment methods to further increase the yield of sugars after enzymatic hydrolysis. Combining wet oxidation with alkaline pretreatment has been shown to reduce the for­mation of by-products, thereby decreasing inhibition. In combination with steam explosion, in a process called wet explosion, the biomass undergoes not only the chem­ical reaction described above but also physical rupture. The advantages to combining wet oxidation with steam explosion includes the ability to process larger particle sizes and to operate at higher substrate loadings, up to 50% substrate (Brodeur et al., 2011).

Glutamic acid is a nonessential amino acid finding its major application in the flavor industry. It was first iso­lated in its pure form from wheat gluten by Ritthausen (1866) and Dr Kikunae Ikeda (1908) found that monoso­dium glutamate was responsible for the flavor of brown kelp used in Japanese food preparations. The discovery was soon patented, and Ajinomoto began the commer­cial production of monosodium glutamate from acid hy­drolysate of wheat gluten and defatted soybean in 1909. At the end of the 1950s, the team led by Kinoshita, of Kyowa Hakko Kogyo, isolated the glutamate-excreting soil bacterium, C. glutamicum. This led to the beginning of the fermentative production of amino acids in large quantities, which in turn made revolutionary changes in the amino acid production and the flavor industry. The bacterium was able to accumulate large amounts of glutamate in the cell, and was excreted out of the cell when triggered either by change in temperature, addition of surfactants, antibiotics or biotin deficiency (Shimizu and Hirasawa, 2006). These triggering mecha­nisms altered cell wall permeability to help excrete the amino acid. Nampoothiri et al. (2002) showed that expression of genes of lipid synthesis and altered lipid composition modulates L-glutamate efflux of C. glutami — cum. In the same year as Kinoshita, Donald J Kita and Jackson Heights reported glutamic acid production from Cephalosporium species (Kita and Jackson, 1957).

Generally, the preferred sources of glutamic acid pro­duction were refined sugars like glucose, fructose and su­crose. In a short period, diverse substrates were explored in place of refined sugars for glutamic acid production, including starch hydrolysates, molasses hydrocarbons and methanol. Enzymatic hydrol, a waste product after enzymatic hydrolysis of grains and starch in industries was fermented by Brevibacterium divaricatum with a yield of 0.19 g glutamic acid per gram of enzymatic hydrolysate (McCutchan et al., 1962). Fermentation with wheat bran and rice bran extracts gave maximum of 50% w/w gluta­mic acid by Bacillus ammoniagenes (Hong et al., 1974). Glutamic acid production has already been demonstrated with palm waste hydrolysate, cassava starch hydrolysate, date waste hydrolysate and rice hydrolysate by different bacterial strains (Das et al., 1995; Nampoothiri and Pandey, 1999; Tavakkoli et al., 2009). Different fermenta­tion methods also were developed so as to utilize agricul­tural wastes. Sugarcane bagasse was used as inert substrate for glutamate production by solid state fermen­tation (Nampoothiri and Pandey, 1996) and similarly sugarcane molasses-enriched medium was also used with carrageenan-immobilized C. glutamicum cells.

Advances in the biorefinery system research basically demands value addition in the overall process. Cell surface expression of alpha amylase was reported in C. glutamicum for glutamate production, which could find use in whole crop biorefinery. The heterologous expression of pentose sugar utilizing genes in C. glutamicum helped to establish a way into the lignocel — lulosic feedstock biorefinery for amino acid production. Here, the heterologous expression of araBAD operon from Escherichia coli resulted in the utilization of arabi — nose for production of L-glutamate. Coutilization of xylose and arabinose along with glucose has also been demonstrated in rice straw and wheat bran hydrolysates for the amino acid production (Gopinath et al., 2011). Later this strain was further improved for accelerated growth and production of lysine, glutamate, ornithine and putrescine, by overexpressing xyl B of C. glutamicum along with other genes involved in pentose sugar utili­zation (Meiswinkel et al., 2012). A detailed review on pentose sugar utilization by C. glutamicum for produc­tion of various value-added commodities was made by Gopinath et al. (2012). The biodiesel industry generates waste glycerol streams, and the engineering of glycerol utilization pathway in C. glutamicum points toward an efficient utilization of glycerol for glutamate production (Rittmann et al., 2008). Efforts were also made for the direct utilization of cellulosic materials. The heterolo­gous expression of Corynebacterium thermocellum endo — glucanase in C. glutamicum resulted in the production of 178 mg/l glutamate from 15 g/l barley b-glucan, with the synergistic action of external b-glucosidase from Aspergillus oryzae (Tsuchidate et al., 2011).

Lysine

Lysine was first isolated from casein by Drechsel (1889). However, microbial conversion of a-amino adipic acid and diaminopimelic acid to lysine was re­ported much later (Haulaham and Mitchell, 1948; Davis, 1952). Microbial fermentation was initiated, when the Kyowa Hakko Kogyo group reported a homoserine auxotroph of C. glutamicum that produced increased amounts of lysine. Classical mutagenesis, strain im­provements and metabolic pathway engineering led to increased production of lysine (Wittmann and Becker, 2007). Genes coding for enzymes in the amino acid biosynthetic pathway were overexpressed or disrupted, either alone or in combinations for amino acid overpro­duction. For example, the identification and cloning of
lysine exporter gene lysE helped to increase lysine excre­tion (Vrljic et al., 1996). Point mutations in the pyc, lysC and hom genes (Figure 19.1) were introduced into the wild-type C. glutamicum chromosome to get the engi­neered lysine producer DM1729 (Georgi et al., 2005). Pyruvate carboxylase and phosphoenol pyruvate carboxykinase are two anaplerotic enzymes in Coryne — bacterium for growth on carbohydrates. The results from various studies indicate that overexpression of pyc gene encoding pyruvate carboxylase redirects the carbon flux toward lysine production and overexpres­sion of pck encoding phosphoenol pyruvate carboxyki — nase is counteractive to amino acid production (Peters Wendisch et al., 1998, 2001; Riedel et al., 2001).

Substrate range for lysine production is very vast including dextrose, sucrose and fructose as the refined

sugars. Organic acids such as acetic acid, propionic acid, benzoic acid, formic acid, malic acid, citric acid and fumaric acid; alcohols such as ethanol, propanol, inositol and glycerol and hydrocarbons, oils and fats such as soy­bean oil, sunflower oil, groundnut oil and coconut oil as well as fatty acids such as palmitic acid, stearic acid and linoleic acid were also used. These substances may be used individually or as mixtures (Anastassiadis, 2007). An array of nondefined sugar substrates like cane and beet molasses, blackstrap molasses and starch hydrolysates were used for industrial fermentation (Ikeda, 2003). Inability of lysine-producing microbes was circumvented by engineering them for direct starch utilization. The expression of amylase genes on the cell surface of C. glutamicum enabled simultaneous sacchari­fication and fermentation of raw corn starch, potato starch and sweet potato starch, rather than using starch hydrolysates and refined sugars (Tateno et al., 2007; Berens et al., 2001). In the direction of enabling the efficient utilization and conversion of hemicellulosic biomass-derived sugars, C. glutamicum has been engi­neered to utilize arabinose and xylose. In the fermenta­tion section of biorefinery, downstream processes with least steps help in better energy efficiency, low product cost and waste management exemplified by spray dried lysine with 78% purity called "Biolys". Under the greenbiorefineries, the brown juice produced from the green crop drying units is acidified with lactic acid fermentation by Lactobacillus species and is stored and transported for lysine fermentation (Thomsen et al., 2004). Biodiesel industry generated glycerol can also be utilized for lysine production.

In general, cyanobacterial H2 production is difficult to sustain for long time periods in liquid cultures in photobioreactors. Utilization of suspension cultures in a two-stage system for H2 production is an even more complicated and energy-consuming process due to the centrifugation or sedimentation steps required for cell harvesting, media changes, and dilutions of cell density during the switch between the different phases. Suspen­sion cultures require intensive mixing, which in turn causes damage to the fragile cyanobacteria filaments. This system is hard to scale up. Use of immobilized cyanobacterial cells in specially designed laboratory — scale photobioreactors would be a good solution to the above-mentioned problems of liquid cultures.

Immobilization of biomolecules and whole cells on various substrates and into different gels, such as solid surfaces like porous glass, supported films, (nano) porous materials, (nano)fibers, foams, inorganic and organic hydrogels, latex, nanotubes, and nanoparticles, has been studied extensively (see for review Meunier et al., 2011). Application of immobilization usually im­proves the stability of the enzymes, and increases light-utilization efficiency. Immobilization of algal cells on a solid phase made of glass has been used for extended H2 production (Laurinavichene et al., 2006). A green alga, C. reinhardtii entrapped in thin alginate films demonstrated extended H2 photoproduction due to increased light-utilization efficiency and better toler­ance against O2 (Kosourov and Seibert, 2009). Several attempts also have been made to immobilize cyanobac — terial cells in order to improve H2 production. These include Plectonema boryanum within alginate beads (Sar — kar et al., 1992), Oscillatoria in agar matrix (Phlips and Matsui, 1986) Phormidium valderianum together with Halobacterium halobium and E. coli within polyvinyl alcohol-alginate beads (Bagai and Madamwar, 1998). Recent immobilization of the Calothrix 336/3 strain in thin alginate film resulted in extended production of H2 even after 40 days of immobilization (Leino et al., 2012). Immobilization has also been found to have a positive effect also on viability of cells. Twelve weeks after initial immobilization, entrapped cells from recov­ered films produced H2 nearly as efficiently as the fresh cells in newly made films. Moreover, the immobilized cells of Calothrix 336/3, Anabaena PCC 7120 and DhupL mutant of Anabaena were viable for over 10 months in the initial nutrient medium without addition of CO2. The demonstrated long-term viability of entrapped cells is a very important issue for economical use of cyano — bacterial in H2 production systems.

There are two principal sources of biomass-based REN for second-generation bioenergy and biofuels: (1) wastes and residues from agriculture and forestry and

(2) dedicated bioenergy crops. Wastes such as wood and agricultural residues, municipal wastes, and poultry litter are typically less expensive to supply to end point users, and are likely to play an important role in early development of commercial-scale REN sup­plies. However, analyses of future demand for REN indicates that these wastes may be capable of supplying only 14—30% of the total potential production of cellu — losic ethanol and only approximately 18—60% of the production potential that could be derived from produc­ing energy crops on currently idle or potentially available agricultural lands (Robert and Abbott, 2012;

Brown, 2009; Lynd et al., 1991). Thus dedicated energy crops will be required to meet the demands of a growing REN market. Such crops, grown in the vicinity of the end point industrial user and specifically for the conversion process being used, offer important advantages of more systematic control of fuel quality, supply, and price stability than wastes derived from dispersed sources, which will be subject to alternative competitive end point uses and associated price fluctuations. The poten­tial feedstocks for second-generation biofuel production considered in this study are biomass from crop residues, other nonfood energy crops, wood/forestry residues, Miscanthus, willow, hemp, Jatropha, switchgrasses and algae (Bauen et al., 2009).

A promising application for microbial pretreatment of lignocellulosic materials is for increasing biogas yield in the anaerobic fermentation process. Anaerobic diges­tion of organic waste and residues not only provides a good solution for the sustainable processing and treat­ment of large amounts of biomaterials, but also leads to value-added renewable energy production. Natural lignocellulosic materials can only be converted to biogas at a very low efficiency due to their resistance to anaerobic digestion. The low biogas conversion rate results from the resistance to enzymatic attack by the biomass due to the tight association of lignin, cellulose, and hemicellulose. Under anaerobic conditions, cellulose and hemicellulose can be degraded during biogas production but not lignin (Fernandes et al., 2009). Pretreatment procedures to increase the accessibility of holocellulose are necessary to increase biogas production. Different pretreatment methods, including physical and chemical pretreatments, effectively enhance anaerobic digestion, but these proce­dures have disadvantages as described beforehand. A microbial pretreatment followed by another step of bio­logical process seems very promising and close to prac­tical application as shown by some following examples.

Pretreatment of wheat straw with Pleurotus sp. "flor­ida" doubles both cellulase digestibility of the treated biomass and the resulting biogas yield, compared with untreated wheat straw (Muller and Trosch, 1986). Pre­treatment of softwood in the presence of wheat bran with the white-rot fungus C. subvermispora, which can effectively degrade the lignin component, enhanced methane fermentation of softwood to 35% of the theoret­ical yield, based on holocellulose content of the biomass. In contrast, pretreatment with Pleurocybella porrigens, which has a lower ability to decompose lignin, led to no significant changes (Amirta et al., 2006).

Application of a lignocellulose degrading composite microbial system with high xylanase activity (XDC-2), instead of a pure culture of microorganisms for biomass pretreatment has also been tested. XDC-2 is composed of 26 different clones from three phyla: Clostridiales, Pro — teobacteria, and Bacteriodetes. However, these degrade mainly carbohydrate but not lignin. After a 5-day pretreatment with XDC-2, corn stalk was efficiently degraded by nearly 45%, and the cellulose and hemicel — lulose contents were decreased by 22.7% and 74.1%, respectively. Biodegradability of the pretreated biomass is improved resulting from changes in chemical struc­ture due to decreased holocellulose content. Compared with untreated corn stalks, total biogas production and methane yield were increased by 68.3% and 87.9%, respectively, and the technical digestion time (T80) was shortened by 35.7% (Yuan et al., 2011).

Effectiveness of biological pretreatments in enhancing corn straw biogas production has also been reported with complex microbial agents including yeast (S. cerevisiae, Coccidioides immitis, and Hansenula anomala), cellulolytic bacteria (Bacillus licheniformis, Pseudomonas sp., Bacillus subtilis, and Pleurotus florida), and the lactic acid bacteria Lactobacillus deiliehii. A 15-day pretreatment of corn straw at ambient temperature led to reduced contents of total lignin, cellulose, and hemicellulose, and increased con­tent of hot-water extractives. Anaerobic digestion of the pretreated material resulted in 33.07% more biogas yield, 75.57% more methane yield, and 34.6% shorter technical digestion time compared with the untreated sample (Zhong et al., 2011).

In conclusion, under proper conditions, microbial/ biological pretreatment can be an effective method for improving biodegradability and enhancing downstream biological conversion efficiency of biomass into bio­energy and other value-added bioproducts.

Mechanisms by which bacteria generate power in MFCs have been intensively investigated in recent years. How might we make "super-bug" electrogens using the power of genetic manipulation? There are many factors that we may ponder. The MFC literature is rife with bio­logical, biochemical and biophysical aspects of highly electrogenic bacteria including members of the genera Geobacter, Shewanella, and Rhodoferax species, and many other Gram-positive and — negative electrogenic bacteria (Huang et al., 2012; Guo et al., 2012). Many of these organ­isms can exist and/or thrive in a myriad of different niches, including those involving significant variations in temperature, pH, osmolarity, pollutants, biocides, and metabolizable/nonmetabolizable carbon sources. As such, many studies involving MFCs house single spe­cies bacteria, bacterial "consortia," media or feedstock, anode/cathode materials, MFC design and design mate­rial, flow rates, Coulombic efficiency and other MFC pa­rameters that can impact power density measurements.

Bacterial Metabolism: How to Power MFCs through Respiratory/Anaerobic Fluxes

Many bacteria are metabolically versatile organisms, and can utilize nearly every carbon-containing com­pound produced in nature. As stated earlier, they are capable of aerobic and/or anaerobic respiration as well as fermentation. Aerobic respiration requires molecular oxygen (O2) while the latter can use alternative electron acceptors including but are not limited to NOT, SO4_, Fe3+, dimethyl sulfoxide (DMSO), trimethylamine N-oxide (TMAO), and CO2. A member of the tricarbox­ylic acid cycle (TCA) cycle, fumarate, can also be used. Interestingly, forms of electron acceptors including oxides of iron and manganese (Shi et al., 2007), vana­dium, selenium, tellurium and toxic metals including chromium, arsenic and cobalt can also be used by some organisms. Thus, depending on the organism(s), the goal of genetically unmodified bacteria is to couple oxidation of organic matter and reduction of terminal electron acceptor (in most cases, the anode of the MFC). However, in MFCs, the electron acceptor in most cases, with the exception of biocathodes, and bioanode/bio — cathode MFCs, is the anodic surface. This can occur in what are commonly termed mediator-dependent or mediator-less MFCs (discussed below). Many facultative and obligatory anaerobic bacteria can undergo an anaer­obic process known as fermentation that does not require functional cytochromes, respiratory chains and produces far less energy in the form of adenosine triphosphate (ATP) than, for example, glucose respiration in E. coli.

Mediator-Less Factors Affecting MFC Performance

Many studies have been conducted in the past ~ 7 years to either isolate superior unknown electrogens or improve the electrogenic properties of existing organ­isms possessing such capacity. The power of various mo­lecular genetics tools (mutations, deletions, gene transfer, overexpression, etc.) is the central force under­lying the discovery of such strains. However, surface — localized factors such as bacterial type IV pili (TFP) represent a major mediator-less protein that contributes significantly to some of the more extensively studied electrogenic bacteria.

TFP (or "Nanowires"): Geobacter and Shewanella Species as Model Organisms

TFP have been found to be critical for the transfer of electrons generated metabolically to metal oxides (e. g. iron oxides; Reguera et al., 2005) that represent just one component of an MFC anode. These are the extendable (fully extended outside the cell, followed by retraction, degradation, and new pilus synthesis) proteinaceous appendages (for fundamental structures of three TFP of P. aeruginosa, N. gonorrhoeae and V. chol — erae, see Figure 9.5(a—c) from Craig et al., 2004). Pili are often essential for optimal biofilm formation in many bacterial genera (Zechner et al., 2012), a require­ment for mediator-less current on the MFC electrode(s) surface. Geobacter members are capable of reducing oxide of either insoluble iron (Fe3+) or manganese (Mn4+) that are directly coupled with organic carbon oxidation. The pilus extends from many bacteria to bind to and retract from surfaces for biofilm formation and disper­sion in some bacteria (O’Toole and Kolter, 1998) and is capable of a "grappling hook" retraction mechanism, followed by degradation, new pilus synthesis and extension, followed again by the complete retraction— degradation—synthesis—extension loop. The pili of elec­trogenic G. sulfurreducens have been termed "nanowires" due to their highly conductive properties (Reguera et al.,

2005) that appear to differ markedly from similar mem­bers of the same genus (e. g. G. metallireducens). These "nanowires" have also been described in S. oneidensis (Gorby et al., 2006) and are likely in many other bacteria.

The electrogenic importance of the pilus was proven in a deletion mutant strain of G. sulfurreducens pilA that generated nearly 10-fold less the power density than that of wild-type, pilus+ bacteria (Reguera et al., 2006).

These results were independent of anode composition, whether it is inexpensive, highly reproducible conduc­tive graphite, or gold, an expensive yet sometimes prob­lematic (reproducibility issues) anodic material (Richter et al., 2008). A longer isoform of PilA is critical for optimal power generation than a shorter PilA (Richter et al., 2012). Many genes, including those involved in flagella and pilus biosynthesis in G. sulfurreducens (and even bacterial human opportunistic pathogens, Proteus mirabilis and P. aeruginosa (Totten et al., 1990; Zhao et al., 1999)) are controlled by the nitrogen sigma factor, RpoN (Leang et al., 2009), as identified by microarray analyses. Thus, predictably strains lacking RpoN are not electrogenic when compared to wild-type bacteria. Yi et al. (2009) demonstrated indirect evidence that T4P in KN400 strain of G. sulfurreducens not only formed more robust biofilms but also provided superior power generation (KN400 current (7.6 A/m2) and power (3.9 W/m2); wild-type DL1-(1.4 A/m2 and 0.5 W/m2)). An excellent review by Lovley et al. (2011) has shed light on the unique processes involved in G. sulfurreducens metabolism and how such unique metabolic proper­ties lends to its reputation as a highly electrogenic organism.

Gorby et al. (2006) have shown that S. oneidensis MR-1 produce conductive pili in response to a reduction in or a lack of a terminal electron acceptor. Those researchers linked electron carrier proteins (c-type decaheme cyto­chromes MtrC and OmcA, see below) as well as muta­tions in the type II secretion pathway, where there are often periplasm protein modifications (e. g. disulfide bond formation) within Gram-negative bacteria. Thus, despite possessing pili, bacteria lacking specific cyto­chromes possessed reduced electrogenic properties.

Yi et al. (2009) isolated a mutant of G. sulfurreducens DL1 (KN400 strain) that was more effective in current production than wild-type bacteria. The paradoxical results were manifested with KN400 forming thinner biofilms, increased current production, great nanowire production, flagellum production, far less outer — surface c-type cytochromes and, above all, lower MFC internal resistance. Recently, however, an artificial matrix termed a conductive artificial biofilm (CAB) was developed that allows for adherence and nearly

11- fold increased conductive properties of Shewanella biofilm bacteria (Yu et al., 2011).

Cytochromes (Cell-Bound)

Redox properties of some bacterial cytochromes (either membrane-bound or soluble cytochromes (e. g. cytochrome c) electron carriers) have been connected with the conductive properties of pili (described above). Typically, these are critical for normal respiratory functions in both prokaryotic and eukaryotic cells.

In recent years, electrogenesis by metal-oxidizing Shewanella and Geobacter species as described above are facilitated by the production of pili and flagella, yet cy­tochromes have also emerged as one of the major drivers of the electrogenic process. This is due, in part, to the or­ganisms harboring such compounds transport and cellular localization of these redox-active cytochromes to the surface or near-surface of the aforementioned or­ganisms (Figure 9.6). Thus, the surface (e. g. an iron oxide (Fe3+) anode) has to be readily accessible to compo — nent(s) of the respiratory pathway of such organisms for optimal electrogenesis to occur. Using S. oneidensis as a model organism for examining the role of cyto­chromes in the electrogenic process, there are at least

10 gene products involved in iron reduction that are crit­ical for some features of electrogenesis in this organism, an event that has been studied by many research groups for more than two decades (Arnold et al., 1990). Conve­niently, most of the genes (especially mtr genes) involved in the process of iron oxide reduction and elec­trogenesis in MFCs are located in close proximity on the S. oneidensis genome (Figure 9.6). Figure 9.6 lists the or­ganisms that are also iron-oxidizing bacteria for

comparative purposes. Of the loci involved in metal reduction, these include mtrDEF, outer membrane cytochrome (omcA), followed by the mtrCAB genes. Figure 9.7 is a simplified recent schematic diagram of the mechanism of precisely how this process functions in S. oneidensis, elegantly described by Shi et al. (2012b). Prior to this exhaustive process of mechanistic functionality, the first genes found to be required for iron and manganese oxide reduction were performed

in S. putrefaciens using transposon mutagenesis in 1998 (Beliaev and Saffarini, 1998). MtrB was found to be an outer membrane cytochrome while the upstream locus, mtrA, encodes a periplasmic decaheme cytochrome. MtrC of both S. putrefaciens and S. oneidensis is also an outer membrane cytochrome with apparent terminal iron reductase activity (Beliaev et al., 2001; Hartshorne et al., 2007). Shi et al. (2006) demonstrated that OcmA, yet another decaheme cytochrome, binds under acidic conditions to MtrC and, in fact, these form a high — affinity protein complex with one another. MtrF, MtrD and MtrE appear to be homologs of MtrCAB, yet one set cannot replace the other functionally, although some components can coaggregate. The outlier is that the mtrFDE loci appear to be highly expressed in bio­films and the Mtr system, in general, is required for optimal biofilm formation (Coursolle et al., 2010), similar to aggregation of bacteria on a conductive sur­face such as an iron oxide anode in MFCs.

In summary, the order of electron flow for optimal electrogenesis of S. oneidensis is the following: cyto­plasmic membrane-bound menaquinone, periplasmic tetraheme CymA with electron flowing through the b-barrel of MtrB to the decaheme cytochromes MtrA/ F and finally to two other decaheme proteins, MtrC and OmcA (Figure 9.7). The final destination for elec­trons prior to reduction of iron oxides is MtrC. Thus, again, it is intuitive that the genes encoding those proteins involved in iron reduction are localized in the following order on the S. oneidensis genome, mtrDEF—omcA—mtrCAB, respectively. Similar to the discovered mechanism of proton pumping in the F1F0-ATP synthase using bacteriorhodopsin in mem­brane vesicles, scientists have proven that the protein complex of MtrCAB conducts electron when embedded within membrane vesicles (Hartshorne et al., 2009). In 2010, Tai et al. (2010) assessed potential networks of transcriptional regulation between chemotactic and electron transport properties and found that previously unknown roles of genes including cheA (a chemotaxis gene), mgtE-1 (an Mg2+ transport gene) and SO4572 (a triheme cytochrome gene). More recently, Leang et al. (2010) showed that the redox-cytochrome OmcS of G. sulfurreducens actually binds to the conductive pili, thereby contributing to their electrogenic proper­ties. However, using a whole-cell cyclic voltammetric analysis of various mutant strains including (DmtrC/ DomcA), transmembrane pili (DpilM-Q, DmshH-Q, and DpilM-Q/DmshH-Q) and flagella (Dflg), Carmona et al. (2011) demonstrated that even with such mutations in place, often there are "by-pass" mechanisms of electron transfer, still allowing for some level of electro­genesis using cyclic voltammetric techniques. A synop­sis of these results is shown in schematic form in Figure 9.8.

Brief Synopsis of the S. oneidensis MR-1 Bioelectrochemical Machinery in Reverse:

Potential Role in the Biosynthesis of Biofuels in MFCs

The multiple proteins and other factors involved in bacterial electrogenesis in MFCs are complex. A pro­cess termed electron flow reversal, or, better put, electron diversion, is critical for a nonelectrogenic process for the purpose of generating single or multiple com­pounds of value. Ross et al. (2011) have helped simpli­fying many features of this process in their 2011 publication. Obviously, the goal of scientists working with electrogenic bacteria is to maximize their power density while wasting the energy harness in the carbon skeletons they consume for sustenance. From the above information collectively, it appears that TFP and cyto­chromes involved in the Mtr respiratory pathway facil­itate the transfer of direct current in the form of electrons to one or more electrodes. Figure 9.9 helps simplifying what is currently understood of these sys­tems and other drivers that will be discussed below. This process is also clearly dependent upon the carbon sources (or feedstock in more complex, multisubstrate systems). In that study, multiple isogenic mutants were created that (1) lacked the periplasmic fumarate reductase (FccA) and thus could not reduce fumarate using electrons derived from electrodes, a process adversely affected by nearly 90% by (2) deletion of mtrB, or worse, (3) the periplasmic cytochrome, MtrA, and prevention of menaquinone biosynthesis.

Any credible source of bioenergy should not only be economically viable but also environmentally sustain­able. The economic and environmental impacts of any source of bioenergy, including biolipids from microal­gae, will usually be measured in terms of energy return on energy investment (EROI) and/or GHG emissions. These economic and environmental impacts of biofuels and microalgae biofuels in general have been hotly debated in recent years. A number of life cycle analyses (LCAs) have been undertaken with seemingly conflict­ing results (Benemann et al., 2012; Liu et al., 2011; Resurreccion et al., 2012; Sun et al., 2011). Similar disparities arose in the case of second-generation bio­fuels such as corn ethanol before the introduction of the Energy and Resources Group (ERG) Bioenergy Meta-Model (Farrell et al., 2006). The results of reported LCA analyses are hindered by the lack of fully inte­grated commercial-scale microalgae to bioenergy sys­tems from which to obtain accurate measurements. Estimates are based on projections from laboratory — and pilot-scale tests, as well as some commercial data. Despite these facts an overall meta-analysis concluded that algae-based biodiesel would result in energy con­sumption and GHG emissions on par with terrestrial al­ternatives (Liu et al., 2011). In this study the authors consider a microalga-based bioenergy system whereby CO2 and nitrogen for microalgae cultivation are recycled from waste streams and the microalgae coproducts are used for further bioenergy production in the form of methane. This concept of an integrated "biorefinery" has been proposed previously (Borowitzka, 1995, 1999; Chisti, 2007; Martin and Grossmann, 2012). As alluded above, the "biorefinery" concept envisages the main in­puts into the cultivation process such as carbon, nitro­gen and phosphorus being supplied through various waste streams. Similarly, the microalgae product result­ing from cultivation could be fully "refined" into a num­ber of outputs including biolipids for bioenergy, biolipids for nutraceutical applications, proteins for an­imal feeds, sugars for bioethanol production, etc. At pre­sent, where fully commercial scale cultivation of microalgae and conversion to fuel alone is still not economically feasible, the "biorefinery" concept appears to offer the best short to medium term path to scale-up.

In addition to the potential economic and environ­mental advantages of using microalgae-derived bio­lipids, the properties of the resulting biodiesel product are also worth considering. As detailed later in this chapter, biodiesel is produced by transesterification of the biolipids from an appropriate feedstock. Much like the plant — and animal-based biolipids discussed previ­ously, the profile of the microalgae-derived biolipids that undergo transesterification will ultimately deter­mine the quality of the biodiesel product. This profile will include the level of polyunsaturated fatty acids (PUFAs), the level of FFAs and the level of TAGs. Although the lipid profile of microalgae varies among species and even among the same species under different conditions of growth, approximately 80% of the lipid content of microalgae, in general, will be made up of storage lipids in the form of TAGs. TAGs are made up of three fatty acid chains, usually with a chain length of C14 to C22 for microalgae-derived bio­lipids, joined to glycerol through three ester bonds (Scott et al., 2010). These TAGs can be easily transesterified in the presence of methanol, as described later in the chap­ter, to fatty acid methyl esters (FAMEs), which make up biodiesel. The presence of FFAs, however, results in the formation of soaps during transesterification in the pres­ence of a base catalyst such as NaOH. This increases the downstream processing required to produce a finished biodiesel product. Similarly, the presence of PUFAs in biolipids derived from some microalgae species can cause tar formation resulting from fatty acid chains cross-linking (Burton et al., 2009). A high PUFA content could also mean that a biodiesel product would not pass European standards for biodiesel (EN14214), which de­mand the content of FAMEs with four or more double bonds to be below 1% mol (Knothe et al., 2005). Other properties that have been considered with regard to other feedstocks mentioned in this chapter include the cloud point, the cetane number and the oxidation stabil­ity of the biodiesel fuel. It has been suggested that bio­diesel from microalgae oils may face significant performance problems regarding cold flow and oxida­tive stability in particular (Knothe, 2011); however, exceptions to this observation may apply to some micro­algae such as Trichosporon capitatum. Also, in a recent study, biodiesel derived from the microalgae Chaetoceros gracilis was found to generate similar torque and power to soy-derived biodiesel. In terms of emissions, the C. gracilis-derived biodiesel also produced less CO, NOx and hydrocarbons than petroleum diesel (Wahlen et al., 2012).

It is clear that the potential for algae to supply a sus­tainable source of biolipid for transportation fuel and other forms of bioenergy is not in doubt. However, there remain technical, economic and environmental chal­lenges to be overcome. In a recent report by the National

Research Council in the United States entitled, "Sustain­able development of algal biofuels" a number of sustain­ability concerns were highlighted. These included EROI; GHG emissions and resource usage such as land, water, nitrogen, phosphorus, and carbon dioxide (National Research Council, 2012). None of these concerns, however, were considered a "definitive barrier to sus­tainable development of algal biofuels". This is because a number of strategies have already been implemented to tackle these challenges. As mentioned previously the use of wastewater streams can drastically reduce resource usage and GHG emissions as well as greatly in­crease EROI. Current projects, at industrial scale, such as Sapphire Energy’s "Green Crude Farm" (Sapphire Energy, 2013) aim to have a capacity of 1 million gallons per year of finished biofuel product. It is predicted that this will result in a 60—70% reduction in GHG emissions compared to traditional fossil crude oil, which, if achieved, will make the potential of microalgae — derived biofuel a very definite reality.