Introduction

Industrial applications of Corynebacterium glutamicum have been well known since 1957, when Kinoshita and co-workers isolated it from soil sample collected from the Ueno Zoo in Tokyo. To date, it remains as the most widely used industrial strain for the amino acid production mainly l-glutamate and l-lysine, (Leuchtenberger et al. 2005; Liebl 1992; Takors et al. 2007). Extensive studies on the metabolic pathways (Ikeda et al. 2006; Marx et al. 1996, 1997; Sonntag et al. 1995; Wendisch 2007) resulted in genetically engineered C. glutamicum with enhanced product range including various amino acids such as l-phenylalanine, l-aspartate, l-tryptophan, l-arginine, l-valine, nucleic acids like purines, vitamins such as riboflavin and pantothenic acid (Burkovski 2008; Eggeling and Bott 2005), and significant amounts of organic acids like acetic, lactic and succinic acid (Inui et al. 2004b; Kinoshita 1985; Terasawa and Yukawa 1993). In addition to these, the efficacy of C. glutamicum to express the recombinant proteins such as the transglutaminase of Streptomyces mobaraensis, the subtilisin-like serine protease of Streptomyces albogriseolus and human epidermal growth factor hEGF were also demonstrated (Date et al. 2003, 2006; Kikuchi et al. 2003). Yet another product of industrial importance is poly(3-hydroxybutyrate), produced by manipulating the biosynthetic pathway of C. glutamicum by the introduction of the phbCAB operon derived from Ralstonia eutropha (Jo et al. 2006). Putrescine and cadaverine are also products of corynebacterial metabolism through metabolic engineering. Cadaverine production was realized by replacing l-homoserine dehydrogenase gene (hom) of a lysine producing strain with l-lysine decarboxylase gene (cadA) of Escherichia coli (Mimitsuka et al. 2007). On the other hand, heterologous expression of genes encoding enzymes of the arginine- and ornithine decarboxylase pathways from E. coli enabled the production of putrescine (Schneider and Wendisch 2010). Recent advances in metabolic engineering strongly recommend C. glutamicum for its ability to transform renewable biomasses to value-added products.

Pentose sugar-rich lignocellulosic biomass, an abundant renewable source of energy, holds a lion’s share of the world’s total agricultural, industrial and forest residual biomass and has been underutilized for a long time (Dien et al. 2003). It generally consists of 10–20% lignin, 40–50% cellulose and 20–30% hemicellulose (Wyman 1999). Bioconversion of lignocellulosic biomass to industrially important products is possible only after its conversion to monomeric sugars, mostly by enzymatic hydrolysis. A pretreatment or prehydrolysis by acid or alkali at higher temperature before enzyme hydrolysis can increase the accessibility of enzyme for its substrate there by increasing the efficiency of hydrolysis (Anuj et al. 2011).

Pretreatment of lignocellulosics can give rise to both cellulose and hemicellulose. A direct conversion of cellulose and hemicellulose in to value-added products by fermentation using C. glutamicum is almost impossible unless, it is converted to its monomeric forms: pentoses and hexoses. Hexoses can be readily taken up by almost all organisms. But pentose sugars cannot be utilized by most of the industrial strains including wild-type C. glutamicum under normal conditions (Collins and Cummins 1984; Lee 1997). In order to attain a maximum of 60–80% process efficiency, the organism has to utilize both pentoses and hexoses. Even though some of the industrial strains are capable of pentose fermentation (Jeffries et al. 2000) as shown in Table 1, most of them are sensitive to inhibitory by-product of lignocellulosic biomass pretreatment. But C. glutamicum showed remarkable resistance towards inhibitors like furfural, 5-hydroxymethylfurfural, 4-hydroxybenzaldehyde (Sakai et al. 2007) under growth arrested conditions. In our previous work (Gopinath et al. 2011), we demonstrated the ability of C. glutamicum to grow under normal fermentation conditions in acid hydrolysates of rice straw and wheat bran to produce amino acids and hence a direct fermentation of pretreated biomass is possible without any detoxification. But all isolates so far identified as C. glutamicum lacked the genes for pentose metabolism (except ATCC31831, which possess arabinose metabolising genes). However, different strategies have been adopted in order to overcome this limitation, and this mini-review is mainly focused on summarizing these developments in C. glutamicum for its ability to convert pentose sugars to value-added commodities.

Table 1 Industrial microbes reported for the production of commodity chemicals through pentose utilization
Full size table

Pentose utilization by wild-type C. glutamicum

The pentose utilization capability and the associated process performance differ extensively among C. glutamicum strains even though they are reported to utilize hexoses and mixed sugars efficiently.

Out of the different strains of C. glutamicum isolated so far (Liebl 2005), the genome sequences of two isolates, C. glutamicum ATCC13032 (Kalinowski et al. 2003; Ikeda and Nakagawa 2003) and C. glutamicum R (Yukawa et al. 2007), have been determined and annotated independently. Both of these sequences confirmed the absence of arabinose metabolising genes in these strains. Interestingly, another strain, C. glutamicum ATCC31831 possesses a functional cluster of genes for l-arabinose utilization (Kawaguchi et al. 2009). It encodes the l-arabinose transporter araE (probably a low affinity H+ symporter), the l-arabinose isomerase araA (catalyzes isomerization of l-arabinose to l-ribulose), the ribulokinase araB (catalyzes the phosphorylation of l-ribulose to l-ribulose-5-phosphate) and the l-ribulose-5- phosphate 4-epimerase, araD, which catalyzes the formation of the PPP intermediate d-xylulose-5-phosphate from l-ribulose-5-phosphate (Sprenger 1995). According to Kawaguchi et al. (2009), the ara gene cluster might have been acquired from another high G + C content Gram-positive bacterium present in soil by a recent horizontal transfer event after the phylogenic branch containing the genus Corynebacterium.

The metabolic pathway for d-ribose catabolism is probably present in all strains of C. glutamicum. Formation of the PPP intermediate d-ribose-5-phosphate from internal d-ribose is catalyzed by two ribokinases: RbsK1 and RbsK2 (Brinkrolf et al. 2008).

The normal xylose catabolic pathway in bacteria consists of conversion of xylose to xylulose by xylose isomerase (xylA) and subsequent phosphorylation of xylulose by xylulose kinase (xylB) to the PPP intermediate d-xylulose-5-phosphate. But none of the C. glutamicum strains so far isolated including C. glutamicum ATCC13032 and C. glutamicum R (Ikeda and Nakagawa 2003; Kalinowski et al. 2003; Yukawa et al. 2007) were reported to utilize xylose, in the absence of xylA genes in them. The final gene for the xylose catabolism xylulokinase (xylB) has been found functional in C. glutamicum R (Kawaguchi et al. 2006). In general, pathways for pentose catabolism present in C. glutamicum strains finally yield pentose phosphate pathway (PPP) intermediates, which are metabolized through the non-oxidative part of the PPP (Blombach and Seibold 2010). Fungi more commonly transform d-xylose into xylitol by using xylose reductase and xylitol dehydrogenase with some exceptions like Piromyces which follows the bacterial pathway (Harhangi et al. 2003). Caulobacter crescentus, a freshwater bacterium expresses an NAD-dependent xylose dehydrogenase encoded by the xylose inducible xylXABCD operon, and follows a distinct xylose utilizing pathway yielding α-ketoglutarate as shown in Fig. 1 (Stephens et al. 2007).

Fig. 1
figure 1

Bacterial isomerase pathway for xylose and arabinose utilization

Full size image

Genetic engineering in C. glutamicum for pentose utilization

The bacterial pathway for arabinose utilization is charted in Fig. 2 (Sprenger 1995). In the fungal pathway like that of Pichia stipitis, Candida tenuis and Spathaspora passalidarum (Fig. 3), aldose reductase reduces l-arabinose to l-arabitol, which is further oxidized to l-xylulose by l-arabinitol 4-dehydrogenase. l-Xylulose is converted to xylitol by l-xylulose reductase and xylitol is then converted to d-xylulose by d-xylulose reductase. Finally xylulokinase will convert d-xylulose to PPP intermediate d-xylulose-5-P (Richard et al. 2001; Hahn-Hagerdal et al. 2007a, b). Surprisingly, both of these pathways are absent in all the C. glutamicum except in C. glutamicum ATCC31831.

Fig. 2
figure 2

The fungal reductive pathway for xylose and arabinose utilization

Full size image
Fig. 3
figure 3

The fungal reductive pathway for xylose and arabinose utilization

Full size image

Genetically engineered arabinose utilizing C. glutamicum strains are constructed by introducing E. coli araBAD operon (Fig. 4a) to C. glutamicum R strain (Kawaguchi et al. 2008), glutamate producer ATCC13032 and lysine producer DM1729 (Schneider et al. 2010). The recombinant of C. glutamicum R, under oxygen-deprived condition with l-arabinose as the sole carbon source produced significant quantities of succinic acid and lactic acid together with small amounts of acetic acid. When a mixture of 5% d-glucose and 1% l-arabinose is used under the same condition, a constant metabolism of l-arabinose resulted in a combined organic acid yield based on the sugar mixture consumed. This showed that the engineered strain is able to utilize l-arabinose as a substrate for organic acid production even in the presence of d-glucose. Similarly, heterologous expression of the araBAD operon in the wild type and in l-lysine producing C. glutamicum (Schneider et al. 2010) resulted in a mutant strain for production of both l-glutamate and l-lysine from arabinose as sole carbon source. They also constructed l-ornithine and l-arginine producing C. glutamicum strains for arabinose utilization by expressing E. coli araBAD operon. They were able to produce l-glutamate, l-lysine, l-ornithine and l-arginine, respectively, from pure arabinose alone and also from pure glucose–arabinose blends as carbon sources.

Fig. 4
figure 4

a Arabinose operon of E. coli (Schleif 2000). Arabinose isomerase, encoded by araA, coverts arabinose to ribulose; ribulokinase, encoded by araB, phosphorylates ribulose; ribulose-5-phosphate epimerase, encoded by araD, converts ribulose-5-phosphate to xylulose-5-phosphate, which can then be metabolized via the pentose phosphate pathways. The three structural genes are arranged in an operon that is regulated by the araC gene product. b Xylose operon of E. coli (Song and Park 1997). The sugar is first isomerized into d-xylulose by xylose isomerase (XylA) and then phosphorylated by xylulokinase (XylB) to produce d-xylulose5-phosphate. XylF as the xylose-binding protein, XylG as an ATP-binding protein, and XylH as a membrane transporter.They are linked to xylAB but are oriented in the opposite direction. Expression of xylA, xylB, and the transport genes is under the positive control of xylR

Full size image

The C. glutamicum usually do not use d-xylose as a carbon source; however, it harbours a gene encoding the xylulokinase (xylB) (Kawaguchi et al. 2006), which is capable of catalyzing phosphorylation of d-xylulose to the PPP intermediate d-xylulose-5-phosphate, the last step in the d-xylose metabolism. Kawaguchi et al. (2006) constructed two recombinant C. glutamicum strains with gene xylA encoding xylose isomerase alone (strain CRX1) and in combination with the E. coli gene xylB (strain CRX2) for the utilization of xylose. The genes in the xylose operon of E. coli (Fig. 4b) were provided on a high-copy-number plasmid and were expressed under the control of the constitutive promoter trc, derived from plasmid pTrc99A. Both recombinant strains were able to grow in mineral medium containing xylose as the sole carbon source. Strain CRX2 showed faster growth on xylose in comparison to CRX1 and produced lactic and succinic acids predominantly by consuming xylose under growth arrested condition. Moreover, in mineral medium containing a sugar mixture of 5% glucose and 2.5% xylose, strain CRX2 cells simultaneously consumed both sugars under oxygen-deprived conditions, demonstrating the absence of diauxic phenomena relative to the new xylA-xylB construct, albeit glucose-mediated regulation still exerted a measurable influence on xylose consumption kinetics.

Pentose uptake and transport engineering in C. glutamicum

Ribose uptake in C. glutamicum is facilitated by an ATP binding cassette transporter ABCrib (Nentwich 2009). RbsACBD-deletion mutant C. glutamicum SN1 completely lacks d-ribose uptake and growth when d-ribose is used as sole carbon source, explaining the role of rbsABCD (TC 3.A.1.2.1) in ribose transport. In C. glutamicum ATCC31831, the gene cluster for the arabinose metabolism includes an arabinose transporter gene (araE) and a metabolic operon, araBAD (Kawaguchi et al. 2009). The role of ara E (TC 2.A.1.1.55) found in Bacillus subtilis (Krispin and Allmansberger 1998), and C. glutamicum are found to be similar even though amino acid similarity between the two araE proteins found is only 34%. Growth experiments with an araE deletion mutant of C. glutamicum ATCC31831 revealed its functionality in high affinity uptake of l-arabinose. Low-affinity uptake systems for l-arabinose are probably present in C. glutamicum strains which are functional only at higher concentrations of l-arabinose (Kawaguchi et al. 2009). By expressing araE in C. glutamicum R harbouring arabinose-catabolising genes from E. coli, Sasaki and coworkers demonstrated a 3-fold increase in the arabinose consumption over a strain without araE (Sasaki et al. 2009).

Xylose assimilation in xylitol producing strain also suggested that the wild type possesses a transporter associated with xylose uptake. The mechanisms of pentose transport in C. glutamicum are still not yet clearly understood, and an unidentified transporter (IMxyl) was also reported for d-xylose uptake in C. glutamicum R (Blombach and Seibold 2010). E. coli genes for arabinose (araBAD) or xylose metabolism (xylAB) in C. glutamicum R strain were introduced for utilization of the respective sugars under both aerobic and oxygen-deprived conditions (Kawaguchi et al. 2006, 2008). Heterologous expression of araE from C. glutamicum ATCC31831 in C. glutamicum R also contributes to d-xylose uptake (Sasaki et al. 2009). It enhanced 3-fold xylose consumption at low concentration of xylose for its catabolising pathway (Sasaki et al. 2009). Particularly noteworthy was the observation in the studies of C. glutamicum that pentose consumption by either growing or oxygen-deprived cells of recombinant strains expressing araE was not repressed in the presence of glucose (Sasaki et al. 2009). Fast consumption of xylose under oxygen-deprived conditions when compared to fully aerobic conditions indicates the influence of molecular oxygen in xylose transport (Kim et al. 2010).

Chemical commodities produced by C. glutamicum from pentoses

Since its discovery, C. glutamicum was manipulated as an efficient and robust biocatalyst for the manufacture of commodity chemicals including organic acids, amino acids, sugar alcohols. The central metabolic pathways have been engineered for pentose sugar utilization and to direct the flux towards the product of interest. Figure 5 summarises the flow chart indicating the value-added commodities obtained through the manipulation of Corynebacterium central metabolism and some the important commodities were analysed in detail.

Fig. 5
figure 5

Production of value-added commodities using Corynebacterial central metabolism. The intermediates presented in bold letters are produced as a result of metabolic engineering and those presented in boxes are various value added products produced by C. glutamicum

Full size image

Organic acids

Microbial organic acid production is attributed to excretion under particular physiological conditions or as intermediates of major metabolic pathways. The overproduction of any of the organic acids depends on the ability of the microorganism to reduce its by-product formation and these pathways can be diverted by metabolic engineering and gene manipulation to organic acids of interest. C. glutamicum is reported to produce high yields of various organic acids including l-lactic acid, succinic acid and acetic acid in mineral salts medium (Inui et al. 2004a). The crucial parameter leading to the production in this case is oxygen deprivation, achieved by high cell density and mild agitation without aeration. Okino and co-workers (2008) found that under this physical condition, l-lactic acid production is in high volumetric capacity in comparison with succinic acid and acetic acid. This high productivity of lactic acid was further refined by overexpression of d-LDH (lactate dehydrogenase) genes from E. coli and Lactobacillus delbrueckii in l-LDH (l -lactate dehydrogenase) encoding ldhA-null C. glutamicum mutants. Likewise the disruption of ldhA gene for l-LDH and overexpression of pyc gene for pyruvate carboxylase in another recombinant redirects the flux towards succinic acid with resultant enhancement in productivity in the presence of added bicarbonate (Okino et al. 2008a, 2008b). LDHs are enzymes that enable both production and utilization of lactate. LdhA, encoded by the ldhA gene, is the fermentative lactate producing LDH and is NAD-dependent. On the other end are the respiratory LDHs. LldD, encoded by the lldD gene, is the membrane-bound, quinone-dependent LDH essential for growth on l-lactate and Dld encoded by dld gene is the quinone dependent LDH responsible for growth on d-lactate (Stansen et al. 2005; Kato et al. 2010). Even though there are reports of ldh-encoded NAD+-dependent LDH from C. glutamicum ATCC 13032 operating unidirectionally for l-lactate formation, Sharkey et al. (2011) found that it can act in both directions with Vmax for the reduction of pyruvate being approximately 10-fold that of the value for l-lactate oxidation.

C. glutamicum has been metabolically engineered for organic acid production from lignocellulosic hydrolysates containing pentoses like xylose and arabinose. C. glutamicum CRX2 was able to accumulate significant amounts of lactic acid (0.45 g/g) and succinic acid (0.2 g/g) under oxygen deprived conditions from mineral medium containing xylose as carbon source. The usefulness of the strain for organic acid production from glucose–xylose mixtures was also demonstrated without negative effect on organic acid production and xylose consumption (Kawaguchi et al. 2006). Organic acid production from arabinose was best demonstrated by recombinant C. glutamicum CRA1, developed by incorporation of E. coli genes araA, araB and ara C encoding arabinose catabolising enzymes. Succinic acid, lactic acid and acetic acid (0.45, 0.34 and 0.06 g/g) were respectively produced with arabinose as the carbon source under oxygen deprivation. The strain was versatile in utilizing mixed sugars (Kawaguchi et al. 2008)

Alcohol production

Besides the general consideration as GRAS, it was the properties like tolerance to ethanol, inhibitors and broad substrate utilization range of engineered C. glutamicum strains makes it as a potential ethanologen. C. glutamicum has been used as a versatile biocatalyst for alcohol production under growth arrested conditions and is engineered to broaden the substrate utilization range. Metabolic engineering for pentose utilizing C. glutamicum CRX2 and CRA1 (for xylose and arabinose metabolism) has been done by Kawaguchi et al. 2006 and Kawaguchi et al. 2008, respectively. Engineering for cellobiose utilization was also carried out in C. glutamicum R. This strain was capable of methyl β-glucoside and the natural aryl β-glucosides, e.g., salicin and arbutin utilization depending on the PTS (Phospho Transferase System). bglA and bglF are genes encoding PTS permease and β glucosidase, respectively, and are components of the β-glucoside PTS and was demonstrated that a point mutation in the codon 317 of the bglF(bglF317A) led to substitutions V317A and/or V317M near the putative PTS active-site H313 in the membrane-spanning IIC domain of bglF (TC 4.A.1.2.5) and allowed bglF to act on cellobiose (Kotrba et al. 2003). Substrate utilization range was extended to include d-xylose. For this another strain X5C1 was constructed by integration of E. coli xylA and xylB genes and the C. glutamicum R bglF  317A and bglA genes into the chromosomal DNA of C. glutamicum R, which can simultaneously utilize d-cellobiose, d-glucose, and d-xylose (Sasaki et al. 2008). The ethanol producing C. glutamicum was constructed by the introduction pdc and adhB genes of Zymomonas mobilis. Under the influence of IdhA promoter, these genes coding for enzymes pyruvate decarboxylase and alcohol dehydrogenase efficiently synthesized ethanol from pyruvate under oxygen-deprived conditions. The addition of small quantities of pyruvate and acetaldehyde is shown to increase the sugar consumption and ethanol production in strains deficient of organic acid synthesizing genes; illustrated with ethanol production reaching up to 642 mmol l h−1 in ldhA-deficient strain(R-ldhA-pCRA723) with pyruvate addition under oxygen deprivation (Inui et al. 2004a). Albeit the ethanol production capacity, it is capable of growing on ethanol with growth rates up to 0.24 h−1 and biomass yields up to 0.47 g dry weight (g ethanol)−1. The cells grown on ethanol showed increased expression of genes encoding phosphotransacetylase (PTA), isocitrate lyase (ICL) and malate synthase (MS) and acetate kinase (AK) (Arndt et al. 2008). Moreover ethanol producing C. glutamicum under the growth arrested conditions, showed tolerance towards inhibitors like furan, organic acids and phenolics, which can be found in various pretreated lignocellulosic hydrolysates. The C. glutamicum strain under study R-ldhA-pCRA723 retained almost 100% ethanol productivity with model composition of inhibitors simulating various pretreatment conditions.

Recent advances in C. glutamicum research showed their competence for alcohols other than ethanol. As C. glutamicum is an established amino acid producer and the 2-keto acid pathways for higher chain alcohol synthesis share common precursors with amino acids, this microbe shows excellent potential for higher chain alcohol. By manipulating the 2-keto acid pathways, it is able to produce isobutanol from pyruvate with a production reaching up to 0.1 g/g (Smith et al. 2010). The C. glutamicum also showed better isobutanol tolerance compared to E. coli. Overexpression of alsS (acetolactate synthase) from B. subtilis, ilvC (acetohydroxy acid isomer reductase) and ilvD (dihydroxy-acid dehydratase) of C. glutamicum, kivd (keto isovalerate decarboxylase) of Lactococcus lactis, and a native adhA (alcohol dehydrogenase), led to the production of 2.6 g/l isobutanol and 0.4 g/l 3-methyl-1-butanol in 48 h. In addition, other higher chain alcohols such as 1-propanol, 2-methyl-1-butanol, 1-butanol, and 2-phenylethanol were also detected as by products (Smith et al. 2010). Recently, Blombach et al. (2011) engineered C. glutamicum for the production of isobutanol from glucose under oxygen deprivation conditions by inactivation of l-lactate and malate dehydrogenases, implementation of keto acid decarboxylase from L. lactis, alcohol dehydrogenase 2 (ADH2) from Saccharomyces cerevisiae, and expression of the pntAB transhydrogenase genes from E. coli. Based on a 2-ketoisovalerate production strain engineered by inactivation of the pyruvate dehydrogenase complex, pyruvate: quinone oxidoreductase, transaminase B, and additional overexpression of the ilvBNCD genes, encoding acetohydroxyacid synthase, acetohydroxyacid isomeroreductase, and dihydroxyacid dehydratase. The resulting strain produced isobutanol with a substrate-specific yield (YP/S) of 0.60 mol per mol glucose. Overexpression of the corynebacterial chromosomal adhA gene increased the YP/S to 0.77 mol of isobutanol per mol of glucose. Inactivation of the malic enzyme significantly reduced the YP/S, indicating that the metabolic cycle consisting of pyruvate and/or phosphoenolpyruvate carboxylase, malate dehydrogenase, and malic enzyme is responsible for the conversion of NADH + H+ to NADPH + H+ (Blombach et al. 2011). By taking all these points into consideration, a cost-effective production of various alcohols from lignocellulosic biomass is well possible with a pentose utilizing mutant of C. glutamicum.

Sugar alcohols

Xylitol, a sugar alcohol derivative of xylose, possesses health beneficial properties which have resulted in a rapid expansion in xylitol consumption to the current annual global value of $340 million (Kadam et al. 2008). Being a valuable synthetic building block (Granstrom et al. 2007), it recently joined the top 12 value-added materials produced from biomass, thereby serving as a key economic driver for the biorefinery concept (Werpy et al. 2004). Wild-type C. glutamicum produced 0.019 g/g xylitol, indicating that it harbours a gene encoding enzymes for xylitol production but the enzyme activities are very low (Sasaki et al. 2010). As reported by Doo et al. (2009), C. glutamicum is a promising biocatalyst for biotransformation and it lacks the xylose isomerase gene (Kawaguchi et al. 2006), and hence is manipulated for the production of xylitol from d-xylose (Kim et al. 2010). The xylose reductase gene of P. stipitis was expressed in C. glutamicum, enabling the recombinant strain to convert d-xylose into xylitol efficiently with a molar yield over 0.97. The biotransformation was markedly influenced by the glucose concentration and oxygen availability in the medium, which appeared to cause lose catabolite repression on xylose transport and NADPH regeneration ability of the organism. By doing biotransformation under glucose and oxygen-limited conditions, xylitol was produced at a concentration of 34.4 g/l (226 mM) with an average specific productivity and molar yield to glucose of 0.092 g (g CDW)−1 h−1 and 1.6 mol/mol, respectively (Kim et al. 2010). In the view that the biocatalytic productivity could be further improved by enhancing the transport efficiency of xylose into the cells, Sasaki et al. (2010) incorporated the pentose transporter araE gene of C. glutamicum 31831 into C. glutamicum R strain (CtXR1) with xylose reductase gene to form CtXR2, which had 13-fold higher xylitol production than its parental strain. In C. glutamicum, xylitol is transported back in to the cells by a phosphoenolpyruvate dependent fructose phosphotransferase system (PTSfru) (Dominguez and Lindley 1996). The xylitol production was further increased by disrupting this phosphotransferase (PTSfru) and xylulose kinase (xylB) genes (Sasaki et al. 2010). For the efficient conversion of hemicellulosic portion of agro residues to xylitol, it is necessary to utilize arabinose also. Introduction of arabinose reductase and arabitol dehydrogenase to C. glutamicum may enable the conversion of arabinose into xylitol and need to be studied in detail.

Amino acids

Amino acids like l-glutamate, l-lysine, l-arginine, and l-ornithine were produced from arabinose as sole carbon source by genetically engineered C. glutamicum strains (Schneider et al. 2010). Arabinose utilizing amino acid secreting strains were constructed by the heterologous expression of the araBAD operon from E. coli into the corresponding amino acid producing strain of C. glutamicum. l-Glutamate and l-lysine producing arabinose utilizing strains were constructed by the recombination of E. coli araBAD operon in ATCC13032 and DM1729 strains, respectively. l-Ornithine production by the recombinant C. glutamicum strain was obtained by deletion of argR for pathway de-repression and deletion of argF to block l-ornithine conversion. Transformation of the ornithine producer with the plasmid pVWEx1-araBAD for arabinose utilization allowed for arabinose-based ornithine production with yields comparable to those obtained with glucose. l-Arginine production by a recombinant C. glutamicum strain carrying an argR deletion for pathway de-repression and expressing a feedback insensitive variant of endogenous N-acteylglutamate kinase from glucose was about twice as high as described for a similar strain (Ikeda et al. 2009). More importantly, l-arginine production from arabinose as sole carbon source and as combined carbon source was even slightly higher. While l-ornithine and l-arginine producing strains were constructed and shown to produce l-ornithine and l-arginine from arabinose when araBAD from E. coli was expressed. Moreover, the recombinant strains produced l-glutamate, l-lysine, l-ornithine and l-arginine, respectively, from arabinose also when glucose arabinose blends were used as carbon sources. This provided a basis for efficient amino acid production from sugars present in lignocellulosic hydrolysates.

While xylose utilizing C. glutamicum R strains were constructed by heterologous expression of E. coli xylose isomerase gene (xylA), which converts xylose to xylulose, the second enzyme xylulokinase (xylB) is already present in C. glutamicum R (Kawaguchi et al. 2006). Although in their studies, they focused on organic acid production, this strain is further engineered to produce the amino acid l-alanine. l-Alanine producing C. glutamicum R strain has been created by deleting the genes associated with production of organic acids and overexpressing alanine dehydrogenase gene (alaD) from Lysinibacillus sphaericus (Jojima et al. 2010). Thus, amino acids were successfully produced by engineered C. glutamicum by utilising pentose sugars. Very recently, the co-utilization of hexose and pentose sugars present in wheat bran and rice straw hydrolysates for the production of l-lysine and l-glutamate by genetically modified C. glutamicum was demonstrated by our group (Gopinath et al. 2011) and this will be a beginning for developing a new bioprocess for amino acid production from hemicellulosic biomass.

Outlook and prospects

Fermentations for the production of commodity chemicals and fuels have always been in the quest of cost effectiveness and sustainability. Lignocellulosics were then identified as prime feed stock for alternative cheap and sustainable sugar supply. This mini-review illustrated the broad product range of speciality and platform chemicals produced from pentose sugars by the work horse of amino acid industry, C. glutamicum. This industrial strain has been moulded to demonstrate its flexible product range from amino acids to alcohols with respect to the pentose uptake from the underutilized lignocellulosic biomass. Simultaneous utilization of various carbon sources by C. glutamicum (Dominguez et al. 1997; Eggeling and Bott 2005; Engels et al. 2008; Lee et al. 1998; Wendisch 2006; Wendisch et al. 2000) is a hallmark of this bacterium setting it apart from yeasts, E. coli and B. subtilis, which typically show sequential utilization of substrates present in blends, and this growth pattern is often accompanied by a diauxic growth lag. On the other hand, C. glutamicum shows very few exceptions to substrate co-utilization (e.g., glucose being utilized prior to ethanol (Arndt et al. 2008; Arndt and Eikmanns 2008) or prior to glutamate (Kronemeyer et al. 1995). The better inhibitor tolerance along with this co-utilization ability makes C. glutamicum superior to other microbes for pentose utilization from lignocellulosic biomass hydrolysates. In-depth knowledge of the metabolic pathways and regulatory mechanisms of the organism made genetic engineering easy, and the pathways for pentose sugar utilization were incorporated in the central metabolic pathway of C. glutamicum. The recent advances in the metabolic engineering paved the way for the beginning of an epoch of cost-effective and sustainable biotechnological production processes.

Earnest efforts were made for making C. glutamicum an efficient pentose utilizer by metabolic engineering and transport engineering. Even though advances in metabolic engineering and membrane transport engineering made the desired product yield very viable and economical to a great extent. Gene regulatory engineering is also conceivable to improve pentose utilization, as it was observed (Kawaguchi et al. 2006, 2008; Schneider et al. 2010; Gopinath et al. 2011) that glucose slowed pentose utilization, indicating some sort of glucose repression to the endogenous hitherto-unknown pentose transporter gene(s) rather than plasmid-borne expression of the heterologous genes for arabinose and xylose catabolism.

C. glutamicum has proved itself a versatile microorganism with multiple product range and productivity. Most of the work we have come across in our mini-review was focused on the utilization of pure chemicals with either arabinose or xylose alone. But for the effective utilization of the lignocellulosic biomass, the co-utilization of all pentose fractions—mainly arabinose and xylose—is very essential. It is necessary to find what happens when this engineered strains were introduced to real life situation, i.e., in the hydrolysates of various agro residues and its economic scale. Also, enough attention must be given to the recovery and purification of the fermentation products derived from lignocellulosic biomass hydrolysates. Future developments in these directions will definitely enhance the potential of C. glutamicum as an efficient pentose utilizing biocatalyst for value addition.