Clostridium termitidis strain CT1112 is an anaerobic, gram positive, mesophilic, cellulolytic bacillus isolated from the gut of the wood-feeding termite, Nasutitermes lujae. It produces biofuels such as hydrogen and ethanol from cellulose, cellobiose, xylan, xylose, glucose, and other sugars, and therefore could be used for biofuel production from biomass through consolidated bioprocessing. The first step in the production of biofuel from biomass by microorganisms is the hydrolysis of complex carbohydrates present in biomass. This is achieved through the presence of a repertoire of secreted or complexed carbohydrate active enzymes (CAZymes), sometimes organized in an extracellular organelle called cellulosome. To assess the ability and understand the mechanism of polysaccharide hydrolysis in C. termitidis, the recently sequenced strain CT1112 of C. termitidis was analyzed for both CAZymes and cellulosomal components, and compared to other cellulolytic bacteria. A total of 355 CAZyme sequences were identified in C. termitidis, significantly higher than other Clostridial species. Of these, high numbers of glycoside hydrolases (199) and carbohydrate binding modules (95) were identified. The presence of a variety of CAZymes involved with polysaccharide utilization/degradation ability suggests hydrolysis potential for a wide range of polysaccharides. In addition, dockerin-bearing enzymes, cohesion domains and a cellulosomal gene cluster were identified, indicating the presence of potential cellulosome assembly.
Citation: Munir RI, Schellenberg J, Henrissat B, Verbeke TJ, Sparling R, Levin DB (2014) Comparative Analysis of Carbohydrate Active Enzymes in Clostridium termitidis CT1112 Reveals Complex Carbohydrate Degradation Ability. PLoS ONE 9(8): e104260. doi:10.1371/journal.pone.0104260
Editor: Jeffrey L. Blanchard, University of Massachusetts, United States of America
Received: November 29, 2013; Accepted: July 11, 2014; Published: August 7, 2014
Copyright: © 2014 Munir et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by funds provided by Genome Canada, through the Applied Genomics Research in Bioproducts or Crops (ABC) program, for the grant titled, “Microbial Genomics for Biofuels and CoProducts from Biorefining Processes”, and by the Province of Manitoba, Agricultural and Rural Development Initiative (ARDI). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Increased concerns over global climate change and energy security, coupled with diminishing fossil fuel resources, have triggered interest in the development of alternative forms of fuel from renewable resources such as biomass. . Consolidated bio-processing (CBP) for biofuel production offers the potential to reduce production costs and increase processing efficiencies when compared with alternative strategies for lignocellulose to ethanol conversion. This is because in CBP, enzyme production, cellulose hydrolysis, and fermentation are all carried out in a single step by microorganisms that express carbohydrate active enzymes (CAZymes) –.
Various anaerobic cellulolytic Clostridium species are known to digest cellulose via an exocellular multi-enzyme complex called a cellulosome –. However, there are a few anaerobic cellulolytic Clostridium species such as C. stercorarium and C. phytofermentans that do not produce cellulosomes. These bacteria degrade cellulosic biomass by secreting enzymes into the environment , .
Genomic studies have revealed that cellulosomes from different cellulolytic bacteria are complex and diverse in nature and architecture , –. The widely studied cellulosome of Clostridium thermocellum (Figure 1) consists of a central scaffoldin protein with varying numbers of cohesin domains, which bind enzymatic subunits through type 1 dockerin domains. The entire complex is bound to the cell surface by the interaction of its type II dockerin domain with the type II cohesin domain on the bacterial cell. The carbohydrate-binding module (CBM), usually of family 3, attaches the complex and the bacterial cell to the cellulosic substrate , . This allows concerted enzyme activity in close proximity to the bacterial cell, enabling optimum synergistic degradation of the substrate . The cellulosome harbors a variety of carbohydrate active enzymes with different substrate specificities, such as endoglucanases, cellobiohydrolases, xylanases, pectinases, and other hydrolyzing enzymes. The mode of action of these enzymes is very similar to those of the free enzymes systems of other cellulolytic bacteria except that the free enzymes in most cases contain a CBM domain instead of a dockerin 1 domain, which targets the individual enzymes to the substrate .
Enzymatic components (colored differently to indicate enzyme variety) produced by anaerobic bacteria contain a dockerin domain. Dockerins bind the cohesins of a non-catalytic scaffoldin, providing a mechanism for cellulosome assembly. Scaffoldins also contain a cellulose-specific family 3 CBM (cellulose binding module) and a C-terminal dockerin domain II that targets the cellulosome to cellulose and the bacterial cell envelope, respectively.
Clostridium termitidis strain CT1112 is an anaerobic, Gram-positive, mesophilic, motile (by peritrichous flagellae), spore-forming, cellulolytic bacterium isolated from the gut of the wood-feeding termite, Nasutitermes lujae . It can use both simple and complex carbohydrates such as cellulose, cellobiose, xylose, glucose, and mannose to carry out mixed-product fermentation, producing acetate, formate, ethanol, H2, and CO2 in amounts comparable to those of other cellulolytic Clostridia –. In addition, experiments conducted recently in our laboratory suggest growth ability on xylan polymers (Figure 2). This makes Clostridium termitidis potentially suitable for biofuel production through consolidated bioprocessing.
Data points represent the means of 3 independent replicates. Bars above and below the means represent the standard deviation.
The C. termitidis genome has recently been sequenced (GenBank accession number AORV00000000) , though investigation of its CAZyme content relative to other cellulolytic Clostridia has not yet been reported. CAZymes have a variety of functions within a cell, but are also involved in the biosynthesis and degradation of cellulose and other polysaccharides. The CAZy database (http://www.cazy.org/) , organizes CAZymes into 6 main classes: i) Glycoside hydrolases (GH), are a large group of enzymes that hydrolyze the glycosidic linkages between two or more carbohydrates or between a carbohydrate and a non-carbohydrate molecule; ii) Carbohydrate esterases (CE), are involved in the hydrolysis of ester bonds; iii) Glycosyl transferases (GTs) catalyze the formation of glycosidic bonds to form a glycoside; iv) Polysaccharide lyases (PLs), cleave glycosidic linkages present in acidic polysaccharides by a beta-elimination mechanism; v) Auxillary activities , are redox enzymes that act in conjunction with other CAZymes to breakdown lignocellulose; and vi) Carbohydrate binding modules (CBMs) are non-catalytic protein domains which function in binding polysaccharides, thus bringing the biocatalyst into close and prolonged proximity with its substrate, allowing carbohydrate hydrolysis .
The objectives of the work described here were to: i) compare the CAZyme content encoded by the C. termitidis genome with those of selected representative cellulosome forming (C. cellulolyticum H10, C. cellulovorans 743B, and C. thermocellum ATCC27405) and non-cellulosome forming (C. phytofermentans ISDg and C. stercorarium DSM8532), anaerobic, cellulolytic Clostridium species; and ii) identify the carbohydrate degradative ability of C. termitidis based on CAZyme data. This will provide an understanding of the mechanism(s) of carbohydrate degradation in C. termitidis and help facilitate the design and development of novel industrially useful microorganisms.
Growth on xylan
Clostridium termitidis CT1112 (DSM 5398), initially obtained from American Type Culture Collection (ATCC 51846), was activated prior to experiments by passaging 10% v/v inoculum on 1191 medium (as previously described ) containing 2 g/L HPLC-grade cell wall polysaccharide xylan from Beechwood (X4252) (Sigma-Aldrich Canada Ltd. Oakville, ON). Cysteine hydrochloride (Sigma-Aldrich), at a concentration of 1 g/L, was used as a reducing agent. Most of the reagents and chemicals for media were obtained from Fisher Scientific, with the exception of Bacto Yeast Extract, which was obtained from Becton Dickinson and Company. The pH of the media was set to 7.2. Time point experiments were conducted in Balch tubes (Bellco Glass Co.) with a working volume of 27 mL. To maintain an anaerobic and sterile environment, tubes containing 2 g/L xylan and 1191 medium were sealed with butyl-rubber stoppers, crimped with aluminum seals, and then gassed and degassed (1∶4 min) four times with 100% nitrogen (N2). Tubes were inoculated (10% v/v) with fresh, mid-exponential phase cultures of C. termitidis and incubated for 36 h at 37 °C. Three independent replicate samples (1.0 mL) were taken every 4 hours. Cell growth was determined by monitoring changes in optical densities at 600 nm using spectrophotometric analysis (Biochrom, Novaspec II) and by protein analysis using a modification of the Bradford method . Briefly, aliquots of cultures were dispensed into micro-centrifuge tubes (Fisher Scientific) and centrifuged at 10,000×g for 10 min to separate the pellets from the supernatants. The pellets were washed with 0.9% (wt/vol) sodium chloride and centrifuged for 10 min. The supernatant was discarded and the pellet was re-suspended in 1 mL of 0.2 M sodium hydroxide. Samples were incubated at 100 °C for 10 mins, and the supernatants were collected for Bradford analysis using Bradford reagent. Optical densities were measured at 595 nm (PowerWave XS, BIO-TEK).
The C. termitidis sequence available at GenBank, with accession number AORV00000000 was used for this analysis . Comparative analysis with other Clostridium species was conducted with genomes available on Joint Genome Institute’s IMG database using the IMG-ER platform . The GenBank accession numbers are NC_009012, NC_011898, NC_014393, NC_010001 and CP003992 for C. thermocellum ATCC27405, C. cellulolyticum H10, C. cellulovorans 743B, C. phytofermentans ISDg, and C. stercorarium DSM8532, respectively.
Phylogenetic analyses were carried out to determine the relatedness of C. termitidis with other members of the Clostridia family based on chaperonin 60 (cpn60) universal target sequences (549–567 bp region of cpn60 gene), which were collected from the Cpn60 database . The phylogenetic tree was obtained using neighbor-joining  in MEGA version 4 . Bootstrap tests with 1000 replications were conducted to examine the reliability of the interior branches.
Translated protein sequences of C. termitidis were analyzed de novo for identification and annotation of its CAZymes and assigned to carbohydrate active enzyme (CAZy) families using the CAZy pipeline , as described in Floudas et al. (2012) . CAZymes of all other Clostridium species analyzed, were directly accessed through the CAZy database , and manually compared. Homologous sequences were obtained by screening the CAZyme sequences of the Clostridia and applying BLASTP search tools using the IMG-ER. Unless specified, highest percentage identity and coverage were reported, based on hits with lowest expect (e)-value (threshold 0.01). Sequence coverage was manually assessed by considering the amino acid (AA) sequence length of the query and the data base target. Conserved domains of protein sequences were searched and analyzed using the evidence for function prediction and Reverse Position Specific (RPS) BLASTs . Potential subcellular localization of identified CAZymes was predicted by uploading FASTA AA sequences of genes into the PSORTb 3.0 database  and using the final predictions.
Results and Discussion
Cell growth on 2 g/L xylan was quantified by measuring optical densities (OD600) and protein concentration, as quantified using the modified Bradford’s method (Figure 2). Cultures showed no lag phase and reached stationary phase by 24 to 28 hrs, with a maximum average OD600 of 0.55 and an average protein concentration of up to 78 ug/mL. Generation times were found to be approximately 5.5 h g−1.
Phylogenetic placement of C. termitidis
To determine the evolutionary relationship between C. termitidis and other sequenced strains of cellulolytic Clostridium species, a phylogenetic tree was constructed based on cpn60 universal target (UT) gene sequence. The 60 kDa chaperonin protein which is encoded by the cpn60 gene is a useful marker for strain identification and molecular phylogenetics . It has been shown that the cpn60 UT sequence can differentiate even closely related isolates of the same bacterial species , . Cpn60 UT sequence alignments have been shown to correlate to whole genome sequence alignments and resolves ambiguities associated with 16S rDNA gene phylogeny in bacteria . Phylogenetic analysis of cpn60 genes (Figure 3) showed that C. termitidis is phylogenetically more closely associated with its mesophilic counterpart C. cellulolyticum. As a result, C. termitidis CAZymes show similarities with CAZymes of C. cellulolyticum, which can be seen throughout the comparative analysis below.
The phylogenetic tree was obtained using neighbor-joining  provided in MEGA version 4 . Bootstrap tests with 1000 replications were conducted to examine the reliability of the interior branches. Asterisks (*) indicates the other Clostridium species used in CAZy comparison.
Genome annotation reveals high numbers of CAZymes in C. termitidis genome compared to other cellulolytic Clostridium species
Genome analyses revealed that C. termitidis has the largest genome and a significantly greater number of total genes (5389) and protein encoding genes (5327) than other members of Clostridia (Table 1) analysed in this study, while C. stercorarium has the smallest genome with only 2706 protein coding genes. Putative CAZyme genes in C. termitidis CT1112 were analyzed de novo and compared to selected Clostridium species (Table 2). The C. termitidis genome encodes a total of 355 CAZyme domain sequences. This is much higher than the number of CAZyme domains found in other Clostridium species that were analyzed. Of the CAZyme domains identified in C. termitidis, glycoside hydrolases (199) and CBMs (95) were most abundant. However, numbers of PLs, CEs, and GTs were comparable to those found in other Clostridium species. Harboring a large number of CAZyme genes may not be surprising, considering the size of the genome and the number of genes it carries. Consistent with its larger genome size, C. termitidis has the greatest numbers of enzyme genes related to carbohydrate metabolism (264), glycan biosynthesis and metabolism (56), and central metabolism (623) compared with other Clostridium species, suggesting differences in protein contents related to sugar utilization and metabolism.
Overall, analysis of the various families that constitute each of the five CAZyme classes showed that in comparison to other Clostridium species considered in this study, C. termitidis had the greatest number of GH (51) and CBM (20) families. In addition C. termitidis had the greatest number of sequences belonging to GH2 (15), GH3 (11), GH4 (10), GH8 (4), GH18 (8), GH23 (3) GH29 (5), GH30 (5), GH31 (6), GH35 (2), GH38 (14), GH43 (12) GH51 (6), GH78 (4), GH94(11), GH95 (5), GT2 (20), GT28 (4), CBM12 (8), CBM32 (14), and CBM35 (24). There were, notably a number of CAZyme genes which were only found in C. termitidis: GH50 (1), GH64 (1), GH76 (1), GH93 (1), GH97 (2), GH 110 (1), PL8 (3) and CBM66 (8) (Table S1 in File S1, Table S2 in File S1, Table S3 in File S1 and Table S5 in File S1).
Interestingly, of the 133 families of GHs currently identified in the CAZy database, the lowest numbers were seen in C. thermocellum, with 27 GH families. C. thermocellum is known to be a comparatively efficient biomass degrader, which identifies it as an attractive candidate organism for CBP . Assignment of sequences into GH families is therefore not necessarily an indication of efficient degradation.
While genomic analysis for the presence or absence of a particular CAZy gene can suggest the capabilities of the strain in question, extracellularly localized (cell bound or secreted) gene products may be more beneficial in identifying complex carbohydrate hydrolysis capabilities. Table 3 shows the predicted extracellular GHs of C. termitidis. A comparative analysis of all the GHs, PLs, and CEs predicted to be localized extracellularly, based on PSORTb 3.0 predictions for the selected Clostridium species, is provided in Table S6 in File S1. Based on our subcellular localization predictions, C. termitidis potentially harbors a variety of extracellular CAZymes responsible for hydrolysis of, among others, (hemi) cellulose, chitin, mannans, starch, and pectin. Below we attempt to correlate C. termitidis extracellular CAZymes with its polysaccharide utilization ability.
Cellulose hydrolysis is generally achieved by the synergistic action of endoglucanases, exoglucanases, and β-glucosidases. Our analysis indicates that C. termitidis possesses all the necessary enzymes needed to carry out this task. The functionally characterized GH48 of C. cellulolyticum is a cellulosomal processive cellulase with both exo- and endo-activities . BLAST analysis shows that C. termitidis GH48 (Cter_0524) has high (AA) sequence similarity (74%) with this enzyme and 57% sequence identity with the AA sequence of the characterized exoglucanase celS (GH48) of C. thermocellum . Cter_0524 has an additional sequence for a dockerin I domain, making it a putative cellulosomal enzyme. As is the case in most other mesophilic cellulolytic Clostridia –, C. termitidis GH48 forms part of a gene cluster (discussed below) which may be putatively linked to the cellulosome.
Members of GH9 family are mainly cellulases which have both endo- and exo-glucanase activities –. CAZy analysis suggests putative endo and exo activities in C. termitidis GH9 members, in addition to having the ability to bind and hydrolyze both the crystalline and amorphous components of cellulose. Of the eleven extracellular GH9s identified in C. termitidis (Table 3), 8 have dockerin I domains and are thus classified as putative cellulosomal enzymes. Five of these (Cter_0272, Cter_0518, Cter_0522, Cter_2830 and, Cter_2831) have a CBM3 domain appended to them, which is known to bind to crystalline cellulose . Our analysis indicates that these are all endoglucanases with high (75–85%) (AA) sequence similarities with corresponding C. cellulolyticum endoglucanases, and up to 77% AA sequence identity with their C. thermocellum counterparts. Dockerin bearing Cter_0521 has a CBM4 domain, which is responsible for directing the GH to the amorphous part of cellulosic substrates. BLAST analysis suggests this to be an exoglucanase with 79% AA sequence identity to C. cellulolyticum exoglucanase Ccel_0732, and 47% AA sequence identity to the characterized CelK (Cthe_0412) of C. thermocellum exoglucanase .
Family GH5 has many enzyme activities relevant to biomass conversion such as cellulases, mannanases, xylanases, xyloglucanases and galactanases . C. termitidis has eight extracellularly secreted GH5 (Table 3). Of these, four (Cter_0515, Cter_0519, Cter_1800 and Cter_0517) have dockerin domains and are annotated as endoglucanases. BLAST analysis shows high AA sequence identity with corresponding C. cellulolyticum GH5 members. In addition, Cter_0515 shows 71% AA sequence identity with the characterized CelG (Cthe_2872) of C. thermocellum . Cter_2349 and Cter_1107 (Table 3) annotated as endoglucanases, seem to be unique to C.termitidis as, BLAST analysis did not give significant AA sequence similarities with other Clostridium species. The multi-domain GH5 protein, Cter_4441, an endoglucanase, has three C-terminal SLH domains, which putatively anchors it to the bacterial cell wall. Cter_4441 has a modular structure GH5_2-CBM17-CBM28-SLH-SLH-SLH, similar to C. cellulolyticum Ccel_0428, and shows 71% AA sequence identity.
Two extracellular genes in family GH8, with dockerin I domains, were identified in C. termitidis (Table 3). As with members of the GH5 family, members of the GH8 family have a variety of functions and can cleave β-1,4 linkages of cellulose, xylan, chitosans, and lichenans. Cter_0523 shows high AA sequence identity (74%) with the cloned and characterized endoglucanase C Ccel_0730 from C. cellulolyticum , and 54% AA sequence identity with the characterized C. thermocellum Cthe_0269, also an endoglucanase .
C. termitidis encodes 11 GH3 genes, highest amongst the other Clostridium species examined (Table S1 in File S1), of which only 2 are extracellular (Table 3), and are annotated as β-glucosidases on IMG database. This suggests that C. termitidis is able to hydrolyze complex sugars to glucose monomers extracellularly before assimilation. This is supported by experiments conducted by us and by the work of Ramachandran et al. , which indicates minimal residual glucose levels in culture supernatants, during growth on α-cellulose and cellobiose, suggesting complete assimilation of glucose as a hydrolysis product. It is interesting to note that C.stercorarium is the only other Clostridium (Table S6 in File S1) that has an extracellular β-glucosidase, putatively suggesting cellulose to glucose hydrolytic ability and assimilation.
Hemicelluloses are polysaccharides in plant cell walls that predominantly consists of xylans and mannans with β-1,4 linked backbones of xylose and mannose monomers respectively . Endoxylanases are commonly found in GH10 and GH11 families. They cleave the xylan backbone to smaller oligosaccharides, which are further degraded to xylose monomers by the action of β-xylosidases found as members of the GH43 family.
Our analysis shows that C. termitidis is equipped with the necessary enzymes required for complete xylan hydrolysis. Of the seven genes functionally annotated as extracellular xylanases, four are attached to the cell surface by either a putative dockerin domain or via the SLH domain, while three are secreted freely in to the environment (Table 3). In addition to the dockerin I domain, Cter_1803 also has a CBM6 domain. This module is known for its xylan-binding abilities, guiding the catalytic component to the appropriate site on the substrate . The enzyme shows 80% AA sequence identity with Ccel_1240, a xylanase found in C. cellulolyticum. Cter_2434, the 1496 AA long multi-component (CBM22-CBM22-CBM22-GH10-CBM9-SLH) GH10 has 810 AA that are identical to those of an endo-1,4-beta-xylanase of Paenibacillus sp. JDR-2 with similar modular structure, and 604 AA that are identical with the Ccel_2320 of C. cellulolyticum. CBM22 and CBM9 modules are both considered to have xylan-binding capabilities .
To further degrade xylo-oligomers into simple sugars, five GH43 genes coding for secreted xylosidases were identified in C. termitidis. Two of these (Cter_0945 and Cter_4060) contain three C-terminal SLH domains for cell attachment. Cter_4060 has a multi-domain structure with domains GH43-CBM35-CBM35-CBM35-CBM35-CBM13-SLH-SLH-SLH. CBM35 and CBM13 are known to primarily bind with both xylan and mannans –. BLAST analysis shows 61% AA identity to the multi-domain C. cellulolyticum β-xylosidase Ccel_3240, which also has a similar modular structure, suggesting that Cter_4060 may have putative xylan hydrolyzing properties.
Arabinofuranosidases hydrolyze arabinose side chains in xylan degradation and are members of GH51 family. C. termitidis encodes genes for two extracellular putative arabinofuranosidases (Table 3). Interestingly, only C. cellulovorans appears to have extracellular homologs among the Clostridium species considered (Table S6 in File S1).
Four putatively secreted members of the GH30 family annotated as O-glycosyl hydrolases on the IMG database were identified in C. termitidis (Table 3). The GH30 family has members with activities ranging from glucosylceramidase, β-1,6-glucanase, β-xylosidase, β-fucosidase, β-glucosidase and endo-β-1,6-galactanase . Cter_0267 and Cter_2867 are putatively cellulosomal due to the presence of a C-terminal dockerin 1 domain. BLAST analysis shows 80% homology with C. cellulolyticum Ccel_0649, which is putatively involved in xylan degradation with high activity towards feruloylated arabinoxylans. Members of GH30 have not been functionally characterized in Clostridia.
The 3192 AA sequence of Cter_2817, a multidomain GH5 protein, has a modular structure CBM66-CBM66-CBM66-GH5_distGH43-CBM35-CBM66-GH43-SLH-SLH-SLH and putatively bound to the cell via the SLH domains. This enzyme seems to be unique to C. termitidis, because BLAST searches did not give hits in other bacteria in the data base. In addition to the GH43 catalytic domain, there is an additional GH5 domain of subfamily 43 (GH5_43). GH5_43 has not been functionally assigned as of yet . However, Cter_2817 has been annotated as a β-xylosidase on IMG database, perhaps due to the presence of the putative GH43 domain. According to the CAZy database, GH43 has members that have xylosidase, arabinofuranosidase, arabinose, xylanase, and galactosidase activities . The presence of multiple CBM66 domains and a CBM35 domain indicates its ability to target both fructans  and xylans  respectively. Endoxylanase Cter_2829, belongs to GH8 family and has a dockerin I domain. BLAST analysis shows 77% AA sequence identity with the Ccel_1298.
Hydrolysis of mannans is mainly carried out by the action of β-mannanases found in the GH26 family. Our results show that the genome of C. termitidis contains a single GH26 gene annotated as β-mannanase (Cter_4544), which may putatively be involved in the breakdown of the mannan backbone. Even though Cter_4544 has a dockerin domain, PSORTb 3.0 analysis was unable to predict its location.
Proteins belonging to family GH18 are candidate chitinases, which are enzymes responsible for chitin degradation. A total of five GH18 genes were identified in C. termitidis (Cter_3529, Cter_3349, Cter_1364, Cter_2813 and Cter_3848). Four (Cter_1364, Cter_2813, Cter_3848, Cter_3349) of these five proteins were predicted to be localized extracellularly. They do not bear dockerin domains, and therefore would not be incorporated into the cellulosome. Of these, Cter_3848 may be bound to the cell wall via the C-terminal SLH domains. Cter_1364 (GH18-CBM12-GH18) and Cter_2813 (GH18-GH18-GH18-CBM12-CBM12) have multiple GH18 catalytic sites and have CBM12 domains which are known to bind chitin polymers. Extracellular members of GH18 were identified in C. thermocellum, C. phytofermentans and C. cellulolyticum. However, none of these have a multi-domain structure.
Carbohydrate esterases (CEs) catalyze de-acylation of saccharides. The C. termitidis genome encodes a total of 15 CEs belonging to five different families, following C. cellulovorans, which has 21 CEs. (Table 2, Table S4 in File S1). The C. termitidis CE15 and one CE4 (Cter_5018) gene products both carry a dockerin 1 domain, suggesting their putative association with the cellulosome. CE4 enzymes, of which C. termitidis has 9 members, are annotated as either acetyl xylan esterases or chitin deacetylases, indicating their putative ability to have activity against both xylans and chitins. All the members of the CE7 family are annotated on the IMG database as acetyl xylan esterases. The C. termitidis family CE9 gene product, annotated as N-acetylglucosamine 6-phosphate deacetylase (EC 220.127.116.11), is putatively important for the metabolism of chitin. This suggests an elaborate chitin degradation ability in C. termitidis which may have evolved in response to cannibalism in termites at times of food shortages , .
CAZyme analysis of the C. termitidis genome shows the presence of a single gene (Cter_3247) belonging to the GH15 family that is annotated as a glucoamylase. Glycoamylases catalyze the release of glucose from the non-reducing ends of starch. C. thermocellum is the only Clostridium species among those examined that has an extracellular homolog (Table S6 in File S1). Also found in the C. termitidis genome are two genes belonging to the GH16 family that have been annotated as extracellular β-glucanases. BLAST analysis gives hits to C. cellulovorans endo-β-1,3-glucosidases. These enzymes are responsible for the breakdown of β-1,3-glucans found as components of various fungi .
Polysaccharide lyases (PLs) are enzymes that mainly degrade uronic acid containing polysaccharides such as glycosaminoglycans and pectin . They are currently classified into 23 families in the CAZy database. C. termitidis encodes a total of four genes which are all predicted to be localized extracellularly and belong to two PL families: family PL8 with three genes and family PL11 with one gene (Table S3 in File S1 and Table S6 in File S1). All members of C. termitidis PL8 have three C-terminal SLH domains, which are putatively responsible for attachment to the cell wall. None of the other Clostridium species have enzymes belonging to this family. PL8 enzymes are known to degrade hyluronate, chondroitin and xanthan while PL11 members are known for their activity against pectin . C. termitidis PL11 has a dockerin domain and as such is putatively active as part of a cellulosomal complex. Except for C. cellulovorans, which has seven extracellular PLs, all other Clostridium species have fewer extracellular PLs than C. termitidis (Table S3 in File S1 and Table S6 in File S1).
Genome analysis shows putative cellulosomal components in C. termitidis
Identification of dockerin-containing proteins.
CAZyme analyses and conserved domain searches provide evidence for the presence of putative dockerin I domains associated with CAZymes in C. termitidis, and the other Clostridium species examined. As previously mentioned, dockerin I domains attach catalytic subunits to the cohesion domains in the cellulosome. Thus, proteins bearing dockerin domains are putatively considered to be cellulosome-associated. Variation in the numbers of GHs bearing dockerin domains were observed in the different species analyzed (Table 4). C. thermocellum has the highest numbers of dockerin domain containing GHs (49). This is followed closely by C. cellulolyticum, which has 40, while C. termitidis had the lowest number, with 22 dockerin domain containing GHs. C. phytofermentans and C. stercorarium do not form cellulosomes, and are known to hydrolyze cellulosic material through a non-complexed cellulase system , . Consequently, no dockerin domains were identified in their CAZomes.
Detection of putative cohesin domains (scaffoldin).
Bioinformatic analysis of the C. termitidis genome revealed the presence of five putative cohesin 1 domain containing proteins (Figure 4). Cohesins Cter_0001 (352 AA) and Cter_3731 (214 AA) are the first genes on their respective DNA scaffolds (1 and 53). BLAST analysis gives 100% coverage and approximately 60% AA sequence similarity to the cellulosome anchoring protein sequence of both C. papyrosolvens (L323_03625; 1332 AA) and C. cellulolyticum (Ccel_0728; 1546 AA). Their location and partial sequences are an indication of genetic truncation and may be a reason for the short sequences observed. Cter_0520, Cter_0525 and Cter_0526 belong to the same DNA scaffold (scaffold 18) and are components of a putative cellulosome related gene cluster that is discussed below. BLAST analysis shows approximately 70% AA similarities to sequences of cohesin domains of a similar cellulosome related gene cluster of C. cellulolyticum (Ccel_0733 and Ccel_0728). The presence of multiple cohesin genes on 3 different scaffolds may indicate the presence of more than one cellulosome integrating protein in C. termitidis. However, further studies to characterize these domains will facilitate our understanding of the type and function of such proteins.
(a) Cter_0001; (b) Cter_0520; (c) Cter_0526; (d) Cter_3731, and, (e) Cter_0525. CBM3-carbohydrate binding module. X2- domain of unknown function which may play a role in attachment of the putative cellulosome to the cell wall. Cohesin I proteins have dockerin binding surfaces, which bind cellulosomal enzymes and are considered important in the formation of a cellulosome. Cohesins a, c and d show putative truncated ends. Cohesins b, c and e are components of a putative cellulosome related gene cluster.
Putative cellulosomal gene clusters.
Similar to some anaerobic, mesophilic, cellulosome-forming Clostridia, such as C. cellulovorans, C. cellulolyticum, C. josui, and C. acetobutylicum –, an approximately 20 Kbp putative cellulosomal enzyme gene cluster was found in C. termitidis harboring 13 cellulosomal genes (Figure 5). With a few differences in the gene content within the different Clostridium species, the gene cluster usually starts with a cohesin containing gene (primary scaffoldin) followed by a series of genes encoding various dockerin-bearing enzymes. This would putatively suggest cellulosome formation in C. termitidis. Such a similarity is an indication that the cellulosomes of these mesophiles may have arisen from a common ancestor. In the case of C. thermocellum, the genes for cellulosomal enzymes are widely scattered on the chromosome and do not form clusters. However, its cellulosome scaffoldin genes, encoding CipA protein, and proteins involved in cellulosome attachment to the cell surface, are organized on the chromosome in a scaffoldin gene cluster . This is not the case in mesophilic Clostridia.
Gene clusters with similar gene arrangement have been identified in other mesophilic Clostridia as indicated. C.termitidis gene cluster with DNA coordinates 84554 to 105381 includes Cter_0526 (1); Cter_0525 (2); Cter_0524 (3), Cter_0523 (4), Cter_0522 (5), Cter_0521 (6), Cter_0520 (7), Cter_0519 (8), Cter_0518 (9), Cter_0517 (10), Cter_0516 (11), Cter_0515 (12), and Cter_0514 (13). Cellulosomal gene clusters identified in C. cellulovorans, C. cellulolyticum, C. josui, and C. acetobutylicum have an approximate size of 21.5 kb, 26 kb, 17.3 kb and 18 kb respectively. Cip (cellulosome integrating protein); CP DT1 (cellulosome protein with dockerin type 1); CD (Cohesin domain).
Cellulosome – cell surface attachment.
Various mechanisms of cell – cellulosome attachment have been noticed in different bacteria. In C. thermocellum, the anchoring of the scaffold containing cellulosome to the bacterial cell wall occurs via the interaction of the dockerin II domains of the scaffoldin with one of three cohesin II proteins (SbdA, OlpB and Orf2p), each of which carries a C-terminal surface layer homology (SLH) repeat that interacts with the S-layer , . There is however an additional C. thermocellum poly-cellulosome forming scaffoldin (Cthe_0736), that may be involved in the formation of extracellular cell free complexes as no evidence exists for it to be cell associated . In the case of both C. cellulolyticum and C. cellulovorans, cell surface cellulosome anchoring proteins are yet to be identified . However an enzyme annotated as endoglucanase E, (EngE) has been implicated in mediating cell surface attachment of the C. cellulovorans cellulosome . The complex cellulosomes of Ruminococcus falvifaciens FD-1, are attached to the cell surface through a sortase transpeptidation reaction . In the case of C. termitidis, we were unable to locate a cohesin II domain or any other protein mediating cellulosome attachment to the cell surface. This may suggest the production of either putative cell free cellulosomes or a novel mechanism of putative cellulosome attachment which needs to be explored.
Clostridium termitidis has the largest genome among the Clostridium species considered in this study. It also has the highest number of CAZymes, which may potentially be advantageous for lignocellulosic biomass hydrolysis. In addition, C. termitidis harbors the most CAZymes secreted extracellularly, some of which are unique and have no homologs in other bacteria. These extracellular CAZymes have the potential capacity to degrade a wide variety of complex and simple carbohydrates, such as cellulose, hemicellulose, starch, chitin, fructans, pectin, glucose, cellobiose and xylose, thus making C. termitidis an attractive microorganism for biofuel production through CBP. We were also able to detect several putative genes that encode gene products with AA sequences that are consistent with key cellulosomal components of other cellulosome producing cellulolytic bacteria. This is an indication of putative cellulosome assembly. However, we were unable to detect any gene or domain with the capacity to act as a cellulosome anchoring protein. This suggests either a novel mechanism of putative cellulosome adherence or the production of putative cell free cellulosomes. Nevertheless, this study has provided us with valuable insights into the mechanism of polysaccharide hydrolysis in C. termitidis. Furthermore, studying the relationship between genome content and gene product expression will provide a systems level understanding of the operative mechanisms of hydrolysis under specific substrate conditions.
Table S1) Comparative analysis of the number of glycoside hydrolase (GH) families in selected Clostridium species. Numbers below each family class indicate the number of members belonging to the specific family for the specific Clostridium specie. The number of family members is colored with respect to the average number of members found in the 6 genomes. Color code: black = deviation between –2and 2 standard deviation (SD) with respect to average; light orange = deviation >2 SD above mean; lightgreen = deviation <–2 SD below mean; dark orange = >3 SD above mean; lightblue = deviation <–3 SD below mean; red = >4 SD above mean; blue = deviation <–4 SD below mean; dark red = >5 SD above mean; darkblue = deviation <–5 SD below mean. Table S2) Comparative analysis of the number of glycosyl transferases (GT) families in selected Clostridium species. Numbers below each family class indicate the number of members belonging to the specific family for the specific Clostridium species. The number of family members is colored with respect to the average number of members found in the 6 genomes. Color code: black = deviation between –2and 2 standard deviation (SD) with respect to average; light orange = deviation >2 SD above mean; lightgreen = deviation <–2 SD below mean; dark orange = >3 SD above mean; lightblue = deviation <–3 SD below mean; red = >4 SD above mean; blue = deviation <–4 SD below mean; dark red = >5 SD above mean; darkblue = deviation <–5 SD below mean. Table S3) Comparative analysis of the number of polysaccharide lyase (PL) families in selected Clostridium species. Numbers below each family class indicate the number of members belonging to the specific family for the specific Clostridium specie. The number of family members is colored with respect to the average number of members found in the 6 genomes. Color code: black = deviation between –2and 2 standard deviation (SD) with respect to average; light orange = deviation >2 SD above mean; lightgreen = deviation <–2 SD below mean; dark orange = >3 SD above mean; lightblue = deviation <–3 SD below mean; red = >4 SD above mean; blue = deviation <–4 SD below mean; dark red = >5 SD above mean; darkblue = deviation <–5 SD below mean. Table S4) Comparative analysis of the number of carbohydrate esterase (CE) families in selected Clostridium species. Numbers below each family class indicate the number of members belonging to the specific family for the specific Clostridium specie. The number of family members is colored with respect to the average number of members found in the 6 genomes. Color code: black = deviation between –2and 2 standard deviation (SD) with respect to average; light orange = deviation >2 SD above mean; lightgreen = deviation <–2 SD below mean; dark orange = >3 SD above mean; lightblue = deviation <–3 SD below mean; red = >4 SD above mean; blue = deviation <–4 SD below mean; dark red = >5 SD above mean; darkblue = deviation <–5 SD below mean. Table S5) Comparative analysis of the number of carbohydrate binding module (CBM) families in selected Clostridium spp. Numbers below each family class indicate the number of members belonging to the specific family for the specific Clostridium specie. The number of family members is colored with respect to the average number of members found in the 6 genomes. Color code: black = deviation between –2and 2 standard deviation (SD) with respect to average; light orange = deviation >2 SD above mean; lightgreen = deviation <–2 SD below mean; dark orange = >3 SD above mean; lightblue = deviation <–3 SD below mean; red = >4 SD above mean; blue = deviation <–4 SD below mean; dark red = >5 SD above mean; darkblue = deviation <–5 SD below mean. Table S6) Comparative analysis of predicted extracellular CAZymes, designated in the CAZy database, involved with lignocellulosic biomass hydrolysis within Clostridium species. Numbers below each family class indicate the number of members belonging to the specific family for the specific Clostridium specie.
Conceived and designed the experiments: RM JS BH TJV RS DBL. Performed the experiments: RM JS TJV. Analyzed the data: RM RS DBL. Contributed reagents/materials/analysis tools: DBL RS. Wrote the paper: DBL RM.
- 1. Levin DB, Zhu H, Beland M, Cicek N, Holbein BE (2007) Potential for hydrogen and methane production from biomass residues in Canada. Bioresour Technol 98: 654–660.
- 2. Lynd LR, Van Zyl WH, McBride LE, Laser M (2005) Consolidated bioprocessing of cellulosic biomass: an update. Curr Opin Biotech 16: 577–583.
- 3. Demain AL, Newcomb M, Wu JHD (2005) Cellulase, clostridia, and ethanol. Microbiol Mol Biol Rev 69: 124–154.
- 4. Wang Z, Chen S (2009) Potential of biofilm-based biofuel production. Appl Microbiol Biotechnol 83: 1–18.
- 5. Lynd LR, Weimer PJ, Van Zyl WH, Pretorius IS (2002) Microbial cellulose utilization: fundamentals of biotechnology. Microbiol Mol Biol Rev 66: 506–577.
- 6. Lamed R, Setter E, Bayer EA (1983) Characterization of a cellulose binding, cellulose-containing complex in Clostridium thermocellum. J Bacteriol 156: 828–836.
- 7. Kakiuchi M, Isui A, Suzuki K, Fujino T, Fujino E, et al. (1998) Cloning and DNA sequencing of the genes encoding Clostridium josui scaffolding protein CipA and cellulase CelD and identification of their gene products as major components of the cellulosome. J Bacteriol 180: 4303–4308.
- 8. Gal L, Pages S, Gaudin C, Belaich A, Reverbel-Leroy C, et al. (1997) Characterization of the cellulolytic complex (cellulosome) produced by Clostridium cellulolyticum. Appl Environ Microbiol. 63: 903–909.
- 9. Doi RH, Goldstein M, Hashida S, Park JS, Takagi M (1994) The Clostridium cellulovorans cellulosome. Crit Rev Microbiol 20: 87–93.
- 10. Bronnenmeier K, Adelsberger H, Lottspeich F, Staudenbauer WL (1996) Affinity purification of cellulose-binding enzymes of Clostridium stercorarium. Bioseparation 6: 41–45.
- 11. Zhang XZ, Zhang Z, Zhu Z, Sathitsuksanoh N, Yang Y, et al. (2010) The noncellulosomal family 48 cellobiohydrolase from Clostridium phytofermentans ISDg: heterologous expression, characterization, and processivity. Appl Microbiol Biotechnol 86: 525–533.
- 12. Xu Q, Bayer EA, Goldman M, Kenig R, Shoham Y, et al. (2004) Architecture of the Bacteroides cellulosolvens cellulosome: description of a cell surface-anchoring scaffoldin and a family 48 cellulase. J Bacteriol 186: 968–977.
- 13. Xu Q, Gao W, Ding SY, Kenig R, ShohamY, et al (2003) The cellulosome system of Acetivibrio cellulolyticus includes a novel type of adaptor protein and a cell surface anchoring protein. J. Bacteriol. 185: 4548–4557.
- 14. Ding SY, Rincon MT, Lamed R, Martin JC, McCrae SI, et al. (2001) Cellulosomal scaffoldin-like proteins from Ruminococcus flavefaciens. J. Bacteriol. 183: 1945–1953.
- 15. Beguin P, Millet J, Aubert JP (1992) Cellulose degradation by Clostridium thermocellum: From manure to molecular biology. FEMS Microbiol Lett 100: 523–528.
- 16. Gerngross UT, Romaniec MPM, Kobayashi T, Huskisson NS, Demain AL (1993) Sequencing of a Clostridium thermocellum gene (cipA) encoding the cellulosomal SL-protein reveals an unusual degree of internal homology. Mol Microbiol 8: 325–334.
- 17. Schwarz WH (2001) The cellulosome and cellulose degradation by anaerobic bacteria. Appl Microbiol Biotechnol 56: 634–649.
- 18. Hethener P, Brauman A, Garcia JL (1992) Clostridium termitidis sp. nov., a cellulolytic bacterium from the gut of the wood-feeding termite, Nasutitermes lujae. Sys Appl Microbiol 15: 52–58.
- 19. Ramachandran U, Wrana N, Cicek N, Sparling R, Levin DB (2008) Hydrogen production and end product synthesis patterns by Clostridium termitidis strain CT1112 in batch fermentation cultures with cellobiose or alpha cellulose. Int J Hydrogen Energy 33: 7006–7012.
- 20. Carere CR, Rydzak T, Verbeke TJ, Cicek N, Levin DB, et al. (2012) Linking genome content to biofuel production yields: a meta-analysis of major catabolic pathways among select H2 and ethanol-producing bacteria. BMC Microbiol 12: 295.
- 21. Sleat R, Mah R, Robinson R (1984) Isolation and Characterization of an Anaerobic, Cellulolytic Bacterium, Clostridium cellulovorans sp. nov. Appl Environ Microbiol 48: 88–93.
- 22. Lal S, Ramachandran U, Zhang X, Munir R, Sparling R, et al. (2013) Draft genome sequence of the cellulolytic, mesophilic, anaerobic bacterium Clostridium termitidis strain CT1112 (DSM 5398). Genome Announc. 1(3): e00281–13.
- 23. Cantarel BL, Coutinho PM, Rancurel C, Bernard T, Lombard V, et al. (2009) The Carbohydrate-Active EnZymes database (CAZy): an expert resource for Glycogenomics. Nucleic Acids Res 37: 233–238.
- 24. Levasseur A, Drula E, Lombard V, Coutinho PM, Henrissat B (2013) Expansion of the enzymatic repertoire of the CAZy database to integrate auxiliary redox enzymes. Biotechnol Biofuels 6: 1–14.
- 25. Guillen D, Sánchez S, Rodriguez-Sanoja R (2010) Carbohydrate-binding domains: multiplicity of biological roles. Appl Microbiol Biotechnol 85: 1241–1249.
- 26. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254.
- 27. Markowitz VM, Mavromatis K, Ivanova NN, Chen IMA, Chu K, et al. (2009) IMG ER: a system for microbial genome annotation expert review and curation. Bioinformatics 25: 2271–2278.
- 28. Hill JE, Penny SL, Crowell KG, Goh SH, Hemmingsen SM (2004) cpnDB: a chaperonin sequence database. Genome Res 14: 1669–1675.
- 29. Saitou N, Nei M (1987) The neighbor-joining method – a new method for reconstructing phylogenetic trees. Mol Biol Evol. 4: 406–425.
- 30. Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol Biol Evol 24 1596–1599.
- 31. Floudas D, Binder M, Riley R, Barry K, Blanchette RA, et al. (2012) The Paleozoic origin of enzymatic lignin decomposition reconstructed from 31 fungal genomes. Science 336: 1715–1719.
- 32. Marchler-Bauer A, Lu S, Anderson JB, Chitsaz F, Derbyshire MK, et al. (2011) CDD: a Conserved Domain Database for the functional annotation of proteins. Nucleic Acids Res 39: D225–D229.
- 33. Yu NY, Wagner JR, Laird MR, Melli G, Rey S, et al. (2010) PSORTb 3.0: improved protein subcellular localization prediction with refined localization subcategories and predictive capabilities for all prokaryotes. Bioinformatics 26: 1608–1615.
- 34. Jian W, Zhu L, Dong X (2001) New approach to phylogenetic analysis of the genus Bifidobacterium based on partial HSP60 gene sequences. Int J Syst Evol Microbiol 51: 1633–1638.
- 35. Verbeke TJ, Sparling R, Hill JE, Links MG, Levin D, et al. (2011) Predicting relatedness of bacterial genomes using the chaperonin-60 universal target (cpn60 UT): Application to Thermoanaerobacter species. Syst Appl Microbiol 34: 171–179.
- 36. Goh SH, Santucci Z, Kloos WE, Faltyn M, George CG, et al. (1997) Identification of Staphylococcus species and subspecies by the chaperonin 60 gene identification method and reverse checkerboard hybridization. J Clin Microbiol 35: 3116–3121.
- 37. Reverbel-Leroy C, Pages S, Belaich A, Belaich JP, Tardif C (1997) The processive endocellulase CelF, a major component of the Clostridium cellulolyticum cellulosome: purification and characterization of the recombinant form. J Bacteriol 179: 46–52.
- 38. Guimaraes BG, Souchon H, Lytle BL, Wu D, Alzari PM (2002) The crystal structure and catalytic mechanism of cellobiohydrolase CelS, the major enzymatic component of the Clostridium thermocellum cellulosome. J Mol Biol 320: 587–596.
- 39. Fujino T, Karita S, Ohmiya K (1993) Nucleotide sequences of the celB gene encoding endo-1, 4-β-β-glucanase-2, ORF1 and ORF2 forming a putative cellulase gene cluster of Clostridium josui. J Ferm Bioeng 76: 243–250.
- 40. Bagners-Tardif C, Gaudin C, Belaich A, Hoest P, Citard T, et al. (1992) Sequence analysis of a gene cluster encoding cellulase from Clostridium cellulolyticum Gene. 119: 17–28.
- 41. Tamaru Y, Miyake H, Kuroda K, Nakanishi A, Matsushima C, et al. (2011) Comparison of the mesophilic cellulosome-producing Clostridium cellulovorans genome with other cellulosome-related clostridial genomes. Microb Biotechnol 4: 64–73.
- 42. Bayer EA, Lamed R, White B, Flint HJ (2008) From cellulosomes to cellulosomics. Chemical record (New York, N.Y.) 8: 364–77.
- 43. Sabathe F, Belaich A, Soucaille P (2002) Characterization of the cellulolytic complex (cellulosome) of Clostridium acetobutylicum. FEMS Microbiol Lett 217: 15–22.
- 44. Parsiegla G, Belaich A, Belaich JP, Haser R (2002) Crystal structure of the cellulase Cel9M enlightens structure/function relationships of the catalytic modules in glycoside hydrolases. Biochem 41: 11134–11142.
- 45. Kataeva IA, Seidel RD, Shah A, West LT, Li XL, et al. (2002) The fibronectin type 3-like repeat from the Clostridium thermocellum cellobiohydrolase CbhA promotes hydrolysis of cellulose by modifying its surface. Appl Environ Microbiol. 68: 4292–4300.
- 46. Kataeva I, Li XL, Chen H, Ljungdahl LG (1998) CelK- a new cellobiohydrolase from Clostridium thermocellum cellulosome: role of N-terminal cellulose-binding domain in Genetics, Biochemistry and Ecology of Cellulose Degradation (Ohmiya K, Sakka K, Kobayashi Y, Karita S, Ed.): 456−460, UniPublishers Co., Tokyo, Japan.
- 47. McLean BW, Boraston AB, Brouwer D, Sanaie N, Fyfe CA, et al. (2002) Carbohydrate-binding modules recognize fine substructures of cellulose. J Biolog Chem 277: 50245–50254.
- 48. Kataeva I, Li XL, Chen H, Choi SK, Ljungdahl LG (1999) Cloning and sequence analysis of a new cellulase gene encoding CelK, a major cellulosome component of Clostridium thermocellum: evidence for gene duplication and recombination. J Bacteriol 181: 5288–5295.
- 49. Aspeborg H, Coutinho PM, Wang Y, Brumer H, Henrissat B (2012) Evolution, substrate specificity and subfamily classification of glycoside hydrolase family 5 (GH5). BMC Evol Biol 12: 186.
- 50. Lemaire M, Beguin P (1993) Nucleotide sequence of the celG gene of Clostridium thermocellum and characterization of its product, endoglucanase CelG. J Bacteriol 175: 3353–3360.
- 51. Fierobe HP, Bagnara-tardif C, Gaudin C, Guerlesquin F, Sauve P, et al. (1993) Purification and characterization of endoglucanase C from Clostridium cellulolyticum. Eur J Biochem 217: 557–565.
- 52. Cornet P, Millet J, Béguin P, Aubert JP (1983) Characterization of two cel (cellulose degradation) genes of Clostridium thermocellum coding for endoglucanases. Nat Biotechnol 1: 589–594.
- 53. Scheller HV, Ulvskov P (2010) Hemicelluloses. Plant Biology 61: 263.
- 54. Sun JL, Sakka K, Karita S, Kimura T, Ohmiya K (1998) Adsorption of Clostridium stercorarium xylanase A to insoluble xylan and the importance of the CBDs to xylan hydrolysis. J Ferment Bioeng 85: 63–68.
- 55. Montanier C, Van Bueren AL, Dumon C, Flint JE, Correia MA, et al. (2009) Evidence that family 35 carbohydrate binding modules display conserved specificity but divergent function. Proc Natl Acad Sci 106: 3065–3070.
- 56. Bolam DN, Xie H, Pell G, Hogg D, Galbraith G, et al. (2004) X4 modules represent a new family of carbohydrate-binding modules that display novel properties. J Biol Chem 279 22953–22963.
- 57. Notenboom V, Boraston AB, Williams SJ, Kilburn DG, Rose DR (2002) High-resolution crystal structures of the lectin-like xylan binding domain from Streptomyces lividans xylanase 10A with bound substrates reveal a novel mode of xylan binding. Biochem 41: 4246–4254.
- 58. Cuskin F, Flint JE, Gloster TM, Morland C, Basle A, et al. (2012) How nature can exploit nonspecific catalytic and carbohydrate binding modules to create enzymatic specificity. Proc Natl Acad Sci U S A 109: 20889–20894.
- 59. Thayer DW (1978) Carboxymethylcellulase produced by facultative bacteria from the hind-gut of the termite Reticulitermes hesperus. J Gen Microbiol 106: 13–18.
- 60. Mishra SC, Sen-Sharma PK (1981) Chitinase activity in the digestive tract of termites (Isoptera). Mat. Organismen 16: 157–160.
- 61. Obayashi T, Kawai T, Yoshida M, Mori T, Goto H, et al. (1995) Plasma (1→ 3)-β-D-glucan measurement in diagnosis of invasive deep mycosis and fungal febrile episodes. The Lancet 345: 17–20.
- 62. Fujino T, Beguin P, Aubert JP (1993) Organization of a Clostridium thermocellum gene cluster encoding the cellulosomal scaffolding protein CipA and a protein possibly involved in attachment of the cellulosome to the cell surface. J Bacteriol 175: 1891–1899.
- 63. Fontes CMGA, Gilbert HJ (2010) Cellulosome: Highly Efficient Nano machines Designed to Deconstruct Plant cell Wall Complex carbohydrates. Annu Rev Biochem 79: 655–668.
- 64. Doi RH, Kosugi A, Murashima K, Tamaru Y, Han SO (2003) Cellulosomes from Mesophilic Bacteria MINIREVIEW Cellulosomes from Mesophilic Bacteria. 185.
- 65. Kosugi A, Murashima K, Tamaru Y, Doi R (2002) Cell-surface-anchoring role of N-terminal surface layer homology domains of Clostridium cellulovorans Eng J Bacteriol. 184: 884–888.
- 66. Rincon MT, Cepeljnik T, Martin JC, Lamed R, Barak Y, et al. (2005) Unconventional mode of attachment of the Ruminococcus flavefaciens cellulosome to the cell surface. J Bacteriol 187: 7569–7578.