SGM Journals
Synthetic Biology

Modular system for assessment of glycosyl hydrolase secretion in Geobacillus thermoglucosidasius

Jeremy Bartosiak-Jentys, Ali H. Hussein, Claire J. Lewis, David J. Leak

  • Department of Biology & Biochemistry, University of Bath, Bath BA2 7AY, UK
  • Correspondence
    David J. Leak d.j.leak{at}bath.ac.uk

    • Microbiology 2013; 159(Pt 7):1267–1275 · https://doi.org/10.1099/mic.0.066332-0

    View at publisher PubMed

    Abstract

    The facultatively anaerobic, thermophilic bacterium Geobacillus thermoglucosidasius is being developed as an industrial micro-organism for cellulosic bioethanol production. Process improvement would be gained by enhanced secretion of glycosyl hydrolases. Here we report the construction of a modular system for combining promoters, signal peptide encoding regions and glycosyl hydrolase genes to facilitate selection of the optimal combination in G. thermoglucosidasius. Initially, a minimal three-part E. coli–Geobacillus sp. shuttle vector pUCG3.8 was constructed using Gibson isothermal DNA assembly. The three PCR amplicons contained the pMB1 E. coli origin of replication and multiple cloning site (MCS) of pUC18, the Geobacillus sp. origin of replication pBST1 and the thermostable kanamycin nucleotidyltransferase gene (knt), respectively. G. thermoglucosidasius could be transformed with pUCG3.8 at an increased efficiency [2.8×105 c.f.u. (µg DNA)−1] compared to a previously reported shuttle vector, pUCG18. A modular cassette for the inducible expression and secretion of proteins in G. thermoglucosidasius, designed to allow the simple interchange of parts, was demonstrated using the endoglucanase Cel5A from Thermotoga maritima as a secretion target. Expression of cel5A was placed under the control of a cellobiose-inducible promoter (Pβglu) together with a signal peptide encoding sequence from a G. thermoglucosidasius C56-YS93 endo-β-1,4-xylanase. The interchange of parts was demonstrated by exchanging the cel5A gene with the 3′ region of a gene with homology to celA from Caldicellulosiruptor saccharolyticus and substituting Pβglu for the synthetic, constitutive promoter PUp2n38, which increased Cel5A activity five-fold. Cel5A and CelA activities were detected in culture supernatants indicating successful expression and secretion. N-terminal protein sequencing of Cel5A carrying a C-terminal FLAG epitope confirmed processing of the signal peptide sequence.

    • Edited by: F. Sargent

    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 work is properly cited.

    Introduction

    The facultatively anaerobic, thermophilic bacterium Geobacillus thermoglucosidasius NCIMB 11955 has been engineered and exploited for industrial bioethanol production from lignocellulosic feedstocks (Cripps et al., 2009). This is primarily due to its rapid growth rate and ability to ferment a broad range of monosaccharides, cellobiose and short-chain oligosaccharides (particularly xylans); it is also amenable to genetic manipulation. However, this strain is unable to grow on crystalline or amorphous cellulose, which would be advantageous as it would reduce the demand for initial enzymic pre-treament of feedstocks. Nevertheless, extracellular endo/exoglucanase activity has been demonstrated in other Geobacillus species, such as Geobacillus sp. R7 (Zambare et al., 2011) and Geobacillus sp. T1 (Assareh et al., 2012), while genome sequencing also predicts the presence of an endoglucanase in the closely related G. thermoglucosidasius C56-YS93 (gb|CP002835.1|, gene ID 10787075) as well as a number of xylanases, although these capabilities have not been assessed in vivo. Therefore, the motivation for this exercise was to enhance the ability of G. thermoglucosidasius NCIMB 11955 to degrade lignocellulose-derived polymers and oligomers by secretion of endo- and exoglucanases and other glycosyl hydrolases.

    Targeting of proteins for extracellular transport is mediated via an N-terminal signal peptide sequence, typically between 20 and 40 amino acids in length. Analysis of the G. thermoglucosidasius NCIMB 11955 genome indicates that the organism has the capability to translocate proteins across the cytoplasmic membrane using either the Sec or TAT (Twin Arginine Transporter) pathway (personal communication, Kirstin Eley, TMO Renewables), although, similarly to Bacillus subtilis (Tjalsma et al., 2000), protein secretion proceeds predominantly via the Sec pathway in this organism. The fundamental difference between the pathways is that while the TAT system transports folded protein, usually containing a metal ion cofactor, the Sec pathway transports unfolded protein which subsequently folds extracellularly (Natale et al., 2008; Sargent et al., 2006). Intriguingly, screening of all signal peptides from B. subtilis suggests that no one-size-fits-all signal peptide exists for the optimal secretion of various proteins (Brockmeier et al., 2006). Therefore, to obtain high levels of secretion for a particular protein it is likely that a library of signal peptides will have to be screened.

    Expression of glycosyl hydrolases needs to be under the control of constitutive promoters or, preferably, context-dependent inducible promoters. To date, the molecular biology tools developed for use in G. thermoglucosidasius have been used either to make chromosomal gene knockouts (Cripps et al., 2009) or to express enzymes intracellularly (Bartosiak-Jentys et al., 2012; Taylor et al., 2008). Furthermore, the extent of expression from the lactate dehydrogenase (ldhA) promoter used in these examples is influenced by the redox conditions associated with the transition from aerobic to anaerobic growth, which may not be universally useful (Bartosiak-Jentys et al., 2012). Positively regulated protein expression systems in bacteria are common and rely on gene expression controlled by inducible promoters (Brautaset et al., 2009). Such promoters give tractable expression when induced with specific agents or conditions (e.g. sugars, peptides, pH); however, no such system is currently available for use in G. thermoglucosidasius.

    Here we report and evaluate the construction of a modular system for the heterologous expression and secretion of proteins in G. thermoglucosidasius NCIMB 11955, which allows the facile interchange of promoter, signal sequence and target gene. This is exemplified by the expression and secretion of the endoglucanase cel5A from Thermotoga maritima using a signal peptide (SP) from a G. thermoglucosidasius C56-YS93 β-1,4-xylanase under the control of a cellobiose-inducible promoter (Pβglu).

    Methods

    All bacterial strains and plasmids used in this study are described in Table 1. Sequences of primers and their targets are detailed in Table 2.

    Table 1. Strains and plasmids
    Table 2. Primer sequences

    Oligonucleotides were from Eurofins.

    Growth conditions.

    G. thermoglucosidasius strains were routinely cultured in 2TY broth containing, per litre deionized water: tryptone 16 g, yeast extract 10 g and NaCl 5 g, adjusted to pH 7.0 prior to autoclaving. Agar was added pre-autoclaving to 15 g l−1 for solid media.

    Cultures of G. thermoglucosidasius for expressing and determining glycosyl hydrolase activity were grown from single colonies in 50 ml of TB-ASM medium containing 0.1 % (w/v) yeast extract and 1 % (w/v) cellobiose, glucose or xylose in 250 ml baffled Erlenmeyer flasks at 60 °C and 250 r.p.m. for 15 h. TB-ASM contained (final concentrations): 8 mM citric acid, 5 mM MgSO4, 20 mM NaH2PO4, 10 mM K2SO4, 25 mM (NH4)2SO4, 80 µM CaCl2, 1.65 µM Na2MoO4, 10 g l−1 tryptone, 5 ml l−1 trace element solution, with 40 mM PIPES, HEPES and BIS-TRIS added as a buffering system post-sterilization. Trace element solution contained: 1.44 l−1 ZnSO4.7H2O, 0.56 g l−1 CoSO4.6H2O, 0.25 g l−1 CuSO4.5H2O, 5.56 g l−1 FeSO4.6H2O, 0.89 g l−1 NiSO4.6H2O, 1.69 g l−1 MnSO4, 0.08 g l−1 H3BO3, 60 mM H2SO4.

    Construction of pUCG3.8.

    The E. coli–Geobacillus sp. shuttle vector pUCG3.8 was constructed with parts from the previously described vector pUCG18 (Taylor et al., 2008). Primers KanR-pUC18_F and repBST1-pUC18_R were used to amplify a 1478 nt fragment corresponding to the E. coli pMB1 origin of replication and MCS of pUC18 (Yanisch-Perron et al., 1985). pUC18-repBST1_F+KanR-repBST_R amplified the 1264 nt Geobacillus sp. origin of replication from pBST1 and repBST-KanR_F+pUC18-KanR_R amplified the 1003 nt thermostable kanamycin nucleotidyltransferase gene (knt) (Liao et al., 1986). The vector was assembled using the Gibson isothermal enzymic DNA assembly (GIA) method (Gibson et al., 2009) and unique AscI, AsiSI and XhoI restriction enzyme recognition sites incorporated into the primers ensured facile substitution of any of the three modules.

    Assembly of reporter plasmid pCEX1.0.1.

    A 293 nt DNA fragment containing the cellobiose-specific phosphotransferase system (PTS) operon’s promoter sequence (Pβglu), predicted using BPROM software (SoftBerry), was amplified using primers SalI_Pβglu_F and XmaI_ClaI_Pβglu_R. The PCR product was digested with SalI–XmaI and cloned into similarly digested pUCG3.8, giving pUCG3.8Pβglu. The transcriptional reporter gene pheB (Bartosiak-Jentys et al., 2012) was amplified using primers GSpheB_ClaI_F and GSpheB_SacI_TT_R. Digestion with ClaI and SacI enabled insertion of this fragment into similarly digested pUCG3.8Pβglu, resulting in pCEX1.0.1.

    Introduction of a signal peptide sequence.

    The G. thermoglucosidasius C56-YS93 endo-1,4-β-xylanase signal peptide (SPGtXyl) (gb|CP002835.1|, nt2 167 991–2 168 075) was incorporated through sequential PCRs using the pheB gene as a caddy. First, primers trunSP_StuI_pheB_F and GSpheB_SacI_TT_R were used to amplify the pheB gene in-frame with the terminal 62 nt of the signal peptide. This fragment was maintained in pJET1.2, giving pJETSP_pheB. This construct served as a template for a subsequent PCR amplification using primers ClaI_SigP_F and GSpheB_SacI_TT_R, the product of which contained the complete signal peptide sequence upstream of the pheB gene. Digestion with ClaI and SacI enabled cloning of this fragment into similarly digested pUCG3.8Pβglu and resulted in pCEX1.1.1. The individual Pβglu, SPGtXyl and pheB elements of the cassette could be excised with the restriction enzyme combinations SalI + ClaI, ClaI + StuI and StuI + SacI, respectively and DNA sequencing confirmed no mutations had occurred.

    Construction of plasmids for expression of glycosyl hydrolases.

    pCEX1.1.1, after excision of the pheB gene with StuI and SacI, served as the backbone for the cloning of the endoglucanase. The cel5A gene from T. maritima (Chhabra et al., 2002) was amplified from plasmid pJETTmar_1751 with the primers Gib_TmCel1751_F and R. The resulting PCR product was cloned into the backbone using GIA to give pCEX1.1.2.

    To aid purification of Cel5A the FLAG epitope nucleotide sequence (24 nt) was introduced in-frame 3′ to the cel5A gene in pCEX1.1.2. An outward PCR using primers FLAG_Cel5A_F and FLAG_Cel5A_R, which each contained half of the FLAG sequence, resulted in a PCR product which when treated with DpnI to remove any residual template DNA, phosphorylated using T4 polynucleotide kinase and ligated to itself with T4 DNA ligase, gave pCEX1.1.2FLAG. Cel5A-FLAG was enriched as the dominant species from G. thermoglucosidasius NCIMB 11955-pCEX1.1.2FLAG culture supernatants by precipitating proteins in the culture supernatant with 80 % (v/v) acetone at −20 °C followed by purification using the ANTI-FLAG M2 Affinity Gel system (Sigma-Aldrich). To confirm the signal peptide sequence had been correctly processed, Edman degradation sequencing of Cel5A-FLAG was carried out (Cambridge Peptides).

    Exchange of modules in the pCEX cassette.

    The versatility of the cassette was demonstrated by substitution of the Pβglu promoter for a synthetic, constitutive promoter PUp2n38. Oligomers PUp2n38_F and R, which contained the 88 nt promoter sequence flanked by SalI and ClaI sites, were hybridized together, digested with SalI + ClaI and ligated into similarly digested pCEX1.1.2 to give plasmid pCEX3.1.2.

    Furthermore, the cel5A gene was replaced with the gene encoding a novel glycosyl hydrolase identified from an environmental screen, and most closely related to the celA gene of Caldicellulosiruptor saccharolyticus, using standard restriction enzyme–ligase cloning. The gene was amplified with primers StuI_Cel_F and SacI_Rhoterm_Cel_R. Attempts to clone the celA gene were unsuccessful, possibly due to recombination between two carbohydrate-binding modules with identical nucleotide sequence. However, due to the presence of an in-frame internal StuI restriction enzyme recognition site, digestion of this PCR product with StuI and SacI allowed cloning only of the nucleotide sequence corresponding to the terminal carbohydrate-binding motif and the N-terminal exogluconase domain (2445 nt) into similarly digested pCEX1.1.2.This new plasmid was called pCEX1.1.3.

    Transformation of G. thermoglucosidasius NCIMB 11955.

    Transformation of G. thermoglucosidasius NCIMB 11955 was performed by electroporation following the protocol of Cripps et al. (2009) with the exceptions that 2TY medium was used for cultures and selection was carried out on 2TY plates containing 12 µg kanamycin ml−1.

    Assay of catechol 2,3-dioxygenase (C23O) activity.

    Qualitative screening of C23O activity was performed by adding 50 µl of 100 mM catechol to harvested cell pellets followed by incubation for 5 min at 55 °C. Quantitative assay of C23O was recorded by spectrophotometric measurement of the accumulation of the catechol ring cleavage product, 2-hydroxymuconic semialdehyde (ϵ = 33 mM−1 cm−1). Activities were determined in a 4 ml cuvette at 55 °C, with a reaction mixture containing 2.9 ml 50 mM sodium phosphate buffer pH 7.2 and 100 µl 10 mM catechol (in deionized dH2O), pre-incubated for 8 min at 55 °C. The reaction was initiated by adding 10 µl cleared cell lysate with absorbance readings taken every second over a time-course of 2 min.

    Assessment of activity against carboxymethyl cellulose (CMC).

    The activity of secreted enzymes was assessed by incubating 1 ml of culture supernatant with 1 ml of 2 % (w/v) CMC in 50 mM citrate buffer, pH 5.8. Reactions were incubated at 55 °C and 250 r.p.m. shaking for 4 h after which reducing sugar release from the polysaccharide was evaluated using the 3,5-dinitrosalicylic acid assay following the protocol of Wood et al. (2012).

    Results and Discussion

    The E. coli–Geobacillus sp. shuttle vector pUCG3.8

    pUCG18 was originally constructed by fusing the plasmids pUC18 and pBST1 (Taylor et al., 2008) and carried antibiotic resistance genes for both ampicillin (bla) and kanamycin (knt). However, the thermostable kanamycin nucleotidyltransferase encoded by knt on pUCG18 confers resistance to kanamycin in both E. coli and Geobacillus sp. Therefore, the bla gene was deemed superfluous and its omission, together with streamlining of the functional elements by removing nucleotides surrounding them that were not predicted as promoters, transcription terminators or open reading frames, enabled a minimal E. col–Geobacillus sp. shuttle vector to be assembled using the GIA (Gibson et al., 2009). This yielded the modular vector pUCG3.8 (Fig. 1).

    Figure image not available in archive
    Fig. 1.

    E. coli–Geobacillus sp. shuttle vector pUCG3.8. Schematic of the plasmid pUCG3.8 after Gibson isothermal assembly of the three modules : pUC [corresponding to the E. coli pMB1 origin of replication and multiple cloning site (MCS) of pUC18, repBST1], Geobacillus sp. origin of replication from pBST1, and KanR [thermostable kanamycin nucleotidyltransferase gene (knt)]. Intermodular AsiSI, XhoI and AscI restriction enzyme recognition sites are indicated along with the SalI, XmaI and SacI sites of the MCS.

    pUCG3.8 was capable of replication in E. coli and Geobacillus using selection with kanamycin at 50 μg ml−1 and 12 µg ml−1 respectively. Restriction digests and DNA sequencing confirmed the correct assembly of the plasmid. The transformation efficiency of pUCG3.8 was tested by transforming 100 ng of the vector into G. thermoglucosidasius NCIMB 11955 by electroporation. Counting of the resulting kanamycin-resistant colonies from five independent replicates indicated a transformation efficiency of 2.8×105 c.f.u. (µg DNA)−1, with a standard deviation of 1.6×105 c.f.u. (µg DNA)−1. PCR analysis of plasmid isolated from four colonies chosen at random confirmed the presence of the three modules.

    While GIA is becoming the method of choice for joining multiple DNA fragments (Gibson et al., 2008), traditional restriction–ligation is still a useful tool. To facilitate its use recognition sites for the enzymes AscI, AsiSI and XhoI were incorporated between modules in pUCG3.8. These would also be useful for generating fragments for GIA without using PCR. The modular nature of the plasmid was demonstrated by excising the Geobacillus sp. origin of replication, repBST1, from pUCG3.8 using the restriction enzymes AscI–XhoI. This was replaced with a similarly digested truncated version of repBST1. The new plasmid pUCG3.7 was transformed in parallel with pUCG3.8 into E. coli BioBlue and G. thermoglucosidasius NCIMB 11955 cells. Both plasmids could be recovered from E. coli grown in the presence of kanamycin whereas only pUCG3.8 was able to replicate in G. thermoglucosidasius NCIMB 11955 cultures. Replacement of the functional repBST1 with the truncated version was also confirmed by DNA sequencing of the relevant portion of the plasmid.

    Replacing the AscI–XhoI repBST1 fragment (as demonstrated) with other Geobacillus sp. origins of replication could generate compatible plasmids or temperature-sensitive vectors [e.g. using the origin from pUB110 (McKenzie et al., 1986)], the latter being useful for chromosomal knockout/in. This would have advantages over previous plasmids designed for this role (Cripps et al., 2009) as its smaller size would facilitate use of larger DNA insertion fragments before the transformation efficiency of Geobacillus sp. is adversely affected.

    Assessment of Pβglu activity with the transcriptional reporter gene pheB

    Induction of β-glucanase expression by cellobiose, the repeat unit of cellulose which is generated through β-glucanase activity, could provide a useful auto-induction system that is independent of artificial inducers. G. thermoglucosidaius NCIMB 11955 can grow on cellobiose. The genome sequence contains an operon predicted to encode the IIB, IIA and IIC components of the cellobiose-specific PTS, respectively, followed by a gene encoding a β-1,4-glucosidase (Fig. 2a). Therefore, this operon’s promoter, Pβglu, is expected to be inducible with cellobiose.

    Figure image not available in archive
    Fig. 2.

    Analysis of the cellobiose-specific PTS operon promoter (Pβglu). (a) Schematic of the organization of the cellobiose-specific PTS operon in G. thermoglucosidasius NCIMB 11955. Genes of the operon encoding components IIA, B and C of the PTS system and the β-glucosidase (β-Glu) are labelled along with the gene for a predicted metal ion-dependent transcription activator. (b) The Pβglu reporter gene construct as in pCEX1.0.1. The nucleotide sequence of a predicted CCR-responsive element (cre) box is highlighted along with the location of the SalI, ClaI and SacI restriction enzyme recognition sites used for cloning (c) Activity of the Pβglu promoter as assessed by catechol C2,3-dioxygenase (C23O) activity in the presence of xylose, glucose and cellobiose and absence of sugar 2TY. Error bars represent the ± standard deviation of 3 biological replicates. Panels (a) and (b) are not to scale.

    Plasmid pCEX1.0.1 was used to assess Pβglu-mediated expression of the pheB gene in the presence of various sugars (Fig. 2b). Cell pellets of G. thermoglucosidasius NCIMB 11955-pCEX1.0.1 grown on TB-ASM cellobiose, glucose or xylose, as well as 2TY which served as a negative control, were qualitatively assayed for C23O activity. A yellow colour indicating active C23O was observed in the cell pellets from cultures grown on TB-ASM cellobiose, glucose and xylose while no colour change was apparent in the cell pellet of the culture grown on 2TY.

    Quantitative assessment of Pβglu promoter activity was carried out using cleared cell lysates from cultures grown as above. The highest promoter activity assessed by C23O activity was measured in cultures grown on cellobiose (630±19 nmol min−1 mg−1) followed by glucose (240±14 nmol min−1 mg1) and xylose (41±1 nmol min−1 mg−1). Almost no activity was detected in the culture grown on 2TY (Fig. 2c).

    Cellobiose transport in G. thermoglucosidasius proceeds via a PTS (Deutscher et al., 2006), which is known to be intimately associated with carbon catabolite repression (CCR). In Gram-positive bacteria this is mediated through phosphorylation of the small histidine-containing protein (HPr) at a regulatory site and binding of this to the protein CcpA. This complex acts by binding to a catabolite responsive element (cre) box located 5′ or within the coding sequence of CCR-controlled genes and, depending on its position relative to the promoter, can either up- or downregulate expression (Deutscher, 2008; Marciniak et al., 2012). cre boxes are highly degenerate semi-palindromes with poorly conserved consensus sequences; however, the sequence 5′-WTGNNARCGNWWWCAW-3′ has been successfully used for genome-wide in silico prediction of cre boxes in B. subtilis (Marciniak et al., 2012). Using the same sequence a cre box was predicted in the G. thermoglucosidasius NCIMB 11955 Pβglu promoter used in this study (Fig. 2b), which suggests that, despite involving a PTS system it may also be subject to catabolite control. Fusion of Pβglu to the transcriptional reporter pheB (pCEX1.0.1) indicated activity of the promoter in cultures of G. thermoglucosidasius NCIMB 11955-pCEX1.0.1 in the presence of cellobiose, with some activity on the CCR sugar glucose and much lower activity on xylose. This suggests that general binding of the activated CcpA to the cre box exerts some regulatory activity on this promoter.

    Upstream of the promoter used in this construct is a MerR (Brown et al., 2003) -encoding sequence, an arrangement previously reported in G. stearothermophilus XL-65-6 (Lai & Ingram, 1993). In other Gram-positive bacteria these regulatory proteins have been implicated in the activation of PTS gene expression (Stoll & Goebel, 2010; Zeng & Burne, 2009) through specific interaction with the substrate-specific components of the PTS, i.e. they can act as substrate sensors. It is possible that the MerR homologue endows this promoter with greater substrate specificity. However, in the context of glycosyl hydrolase expression with concurrent utilization of the various monosaccharides found in lignocellulosic biomass, the lack of absolute promoter specificity we observe may be advantageous.

    Expression of T. maritima Cel5A from pCEX1.1.2 in G. thermoglucosidasius

    The presence of unique restriction sites in the pCEX cassette that allow excision of the promoter, signal peptide or gene is advantageous for GIA in so much as the vector backbone can be generated without the need for amplifying a large (~5 kb) PCR product, which could lead to mutations. This was exploited to replace the pheB gene in pCEX1.1.1 with the gene encoding the endoglucanase Cel5A from T. maritima by GIA to give pCEX1.1.2 (Fig. 3a).

    Figure image not available in archive
    Fig. 3.

    Activity from the glycosyl hydrolase expression cassette pCEX. (a) Schematic of the glycosyl hydrolase expression cassette consisting of the cellobiose-specific PTS operon promoter (Pβglu), the signal peptide from G. thermoglucosidasius C56-YS93 endo-1,4-β-xylanase (SP) and the cel5A gene from T. maritima, as found in pCEX1.1.2. The location of SalI, ClaI, StuI and SacI restriction enzyme recognition sites surrounding Pβglu, SP and cel5A are indicated (not to scale). (b) Extracellular CMCase activity of G. thermoglucosidasius NCIMB 11955-pCEX1.0.1 and G. thermoglucosidasius NCIMB 11955-pCEX1.1.2 cultures grown in the presence (+Cell) and absence (−Cell) of cellobiose. (c) 10 % SDS-PAGE gel showing Cel5A-FLAG after affinity purification. Relevant standards (kDa) are labelled in the left lane. (d) Extracellular CMCase activity of G. thermoglucosidasius NCIMB 11955-pCEX1.0.1 and G. thermoglucosidasius NCIMB 11955-pCEX3.1.2 cultures grown on TB-ASM 0.5 % containing cellobiose. (e) As (b) except G. thermoglucosidasius NCIMB 11955 was carrying plasmid pCEX1.1.3. Error bars in (b), (d) and (e) represent the ±SD of 3 biological replicates.

    G. thermoglucosidasius NCIMB 11955-pCEX1.0.1, and G. thermoglucosidasius NCIMB 11955-pCEX1.1.2 were each grown in TB-ASM cellobiose and TB-ASM+0.1 % (w/v) yeast extract. Cell pellets were harvested for qualitative assessment of C23O activity from the two G. thermoglucosidasius NCIMB 11955-pCEX1.0.1 cultures while all four culture supernatants were assayed for cellulolytic activity against CMC. As expected, C23O activity was only present, indicated by a yellow colour in the cell pellet, after addition of catechol, in G. thermoglucosidasius NCIMB 11955-pCEX1.0.1 cultures grown on TB-ASM cellobiose. This provided confirmation of the induction of gene expression from Pβglu under these conditions. Activity of Cel5A expressed from G. thermoglucosidasius NCIMB 11955 was highest in cultures grown on TB-ASM cellobiose, at 72±1 nmol min−1 mg−1. In the absence of the cellobiose inducer, activity decreased to 28±1 nmol min−1mg−1, which was only marginally higher than the background activity of 8±3 nmol min−1 mg−1 on TB-ASM cellobiose, measured in the supernatants of G. thermoglucosidasius NCIMB 11955-pCEX1.0.1 cultures which did not possess a heterologous endoglucanase. No activity was detected in G. thermoglucosidasius NCIMB 11955-pCEX1.0.1 cultures grown on TB-ASM+yeast extract (Fig. 3b).

    To confirm that the measured enzymic activity was due to extracellular transport of the glycosyl hydrolase (GH) rather than cell lysis, Cel5A was expressed carrying a C-terminal FLAG epitope. This tagged enzyme was purified, analysed by SDS-PAGE and subjected to N-terminal sequencing by Edman degradation. SDS-PAGE analysis of the post-purification fraction indicated a dominant species at ~37 kDa, consistent with the size of Cel5A after removal of the signal peptide (Fig. 3c). Five cycles of peptide sequencing returned the amino acid string N-ARPVG-C, also consistent with Cel5A with a correctly cleaved signal peptide (signal peptide sequence begins N-MRNVL-C and does not contain the sequence ARPVG). This finding also corroborates the in silico predicted signal peptide cleavage site as determined by SignalP 4.1. Evidence of a correctly processed recombinant Cel5A suggests that the endoglucanase activity measured in culture supernatants is due to secretion of the enzyme rather than cell lysis.

    Expression of CelA from pCEX1.1.3 in G. thermoglucosidasius

    The versatility of the pCEX expression cassette has been demonstrated by introducing a truncated gene, closely related to the celA gene from C. saccharolyticus, by restriction enzyme–ligation cloning to give pCEX1.1.3. The activity of the truncated CelA expressed from G. thermoglucosidasius NCIMB 11955, grown under the same conditions as G. thermoglucosidasius NCIMB 11955-pCEX1.1.2 above, was again highest in cultures grown on TB-ASM cellobiose, at 58±14 nmol min−1 mg−1. Activity was 19±1 nmol min−1 mg−1 in cultures lacking cellobiose. Activity in supernatants of G. thermoglucosidasius NCIMB 11955-pCEX1.0.1 grown on TB-ASM cellobiose and TB-ASM+yeast extract were 16±1 and 27±0.5 nmol min−1 mg−1, respectively (Fig. 3e).

    Despite containing only the C-terminal carbohydrate-binding domain and glycosyl hydrolase family 48 exoglucanase domain, a CMCase activity could be detected in the culture supernatants of G. thermoglucosidasius NCIMB 11955-pCEX1.1.3. However, the activities of both the truncated CelA and, more significantly, Cel5A from T. maritima expressed in this study were low compared to the values reported previously (Chhabra et al., 2002; Pereira et al., 2010). This relatively low activity suggests either insufficient promoter activity or poor secretion, confirmed by the fact that no novel band was visible on SDS-PAGE (data not shown) when comparing the secretomes of wild-type and recombinant strains.

    Exchange of the inducible Pβglu for a constitutive promoter

    Quantitative end-point assessment of Pβglu activity, determined using the C23O assay, in the presence of cellobiose indicated activity on a par to that measured for the previously characterized lactate dehydrogenase Pldh promoter (Bartosiak-Jentys et al., 2012). However, when their respective genes are expressed from this promoter, both Cel5A and CelA activities are low, possibly due to changes in expression as cellobiose levels are depleted. Therefore, activity of Cel5A in culture supernatants of a strain carrying the cel5A gene under the control of a constitutive promoter was assessed. Uracil phosphoribosyltransferase is involved in the pathway employed by many micro-organisms to salvage uracil. A synthetic promoter, whose sequence was modified from the region upstream of the uracil phosphoribosyltransferase in G. thermoglucosidasius NCIMB 11955 and which was shown to be constitutively active (personal communication, B. Reeve, Imperial College London) was cloned upstream of the signal peptide and cel5A gene to give pCEX3.1.2. Activity against CMC in supernatants of G. thermoglucosidasius carrying pCEX3.1.2 was 340±35 nmol min−1 mg−1 (Fig. 3d). This was five times higher than the cellobiose-induced activity in G. thermoglucosidasius pCEX1.1.2 and, despite the elegance of the inducible promoter, demonstrated the potential advantage of a constitutive promoter in this system.

    The signal peptide used in the expression cassette was from a G. thermoglucosidasius C56-YS93 endo-1,4-β-xylanase. It was selected due to being predicted with high confidence as a Sec system signal peptide by SignalP 4.1 (CBS), was from the same species as the organism in which secretion was to be assessed and is associated with the secretion of a glycosyl hydrolase. Evidence from B. subtilis highlights the importance of using a signal peptide that works well in concert with the protein to be expressed (Brockmeier et al., 2006; Tjalsma et al., 2000). A number of genes encoding strongly predicted signal peptide sequences have been identified in the G. thermoglucosidasius NCIMB 11955 genome, which may further improve secretion of glycosyl hydrolases expressed from pCEX plasmids. The modular nature of the expression cassette will facilitate the screening of a significant number of signal peptide–gene combinations whose assembly may otherwise have been more involved had synthetic biology principles not been taken into consideration.

    Acknowledgements

    We are grateful to the BBSRC IBTI for provision of funding for J. B.-J. and A. H. H., to TMO Renewables for access to the G. thermoglucosidasius NCIMB 11955 genome sequence and to Ben Reeve for the P2n38 promoter sequence.

      1. Marciniak, B. C.,
      2. Pabijaniak, M.,
      3. de Jong, A.,
      4. Duhring, R.,
      5. Seidel, G.,
      6. Hillen, W. &
      7. Kuipers, O. P.
      (2012). High- and low-affinity cre boxes for CcpA binding in Bacillus subtilis revealed by genome-wide analysis. BMC Genomics 13, 401. doi:10.1186/1471-2164-13-401pmid:22233093
      1. McKenzie, T.,
      2. Hoshino, T.,
      3. Tanaka, T. &
      4. Sueoka, N.
      (1986). The nucleotide sequence of pUB110: some salient features in relation to replication and its regulation. Plasmid 15, 93–103. doi:10.1016/0147-619X(86)90046-6pmid:3010356
      1. Natale, P.,
      2. Brüser, T. &
      3. Driessen, A. J. M.
      (2008). Sec- and Tat-mediated protein secretion across the bacterial cytoplasmic membrane–distinct translocases and mechanisms. Biochim Biophys Acta 1778, 1735–1756. doi:10.1016/j.bbamem.2007.07.015pmid:17935691
      1. Pereira, J. H.,
      2. Chen, Z.,
      3. McAndrew, R. P.,
      4. Sapra, R.,
      5. Chhabra, S. R.,
      6. Sale, K. L.,
      7. Simmons, B. A. &
      8. Adams, P. D.
      (2010). Biochemical characterization and crystal structure of endoglucanase Cel5A from the hyperthermophilic Thermotoga maritima. J Struct Biol 172, 372–379. doi:10.1016/j.jsb.2010.06.018pmid:20599513
      1. Sargent, F.,
      2. Berks, B. C. &
      3. Palmer, T.
      (2006). Pathfinders and trailblazers: a prokaryotic targeting system for transport of folded proteins. FEMS Microbiol Lett 254, 198–207. doi:10.1111/j.1574-6968.2005.00049.xpmid:16445746
      1. Stoll, R. &
      2. Goebel, W.
      (2010). The major PEP-phosphotransferase systems (PTSs) for glucose, mannose and cellobiose of Listeria monocytogenes, and their significance for extra- and intracellular growth. Microbiology 156, 1069–1083. doi:10.1099/mic.0.034934-0pmid:20056707
      1. Taylor, M. P.,
      2. Esteban, C. D. &
      3. Leak, D. J.
      (2008). Development of a versatile shuttle vector for gene expression in Geobacillus spp. Plasmid 60, 45–52. doi:10.1016/j.plasmid.2008.04.001pmid:18501964
      1. Tjalsma, H.,
      2. Bolhuis, A.,
      3. Jongbloed, J. D. H.,
      4. Bron, S. &
      5. van Dijl, J. M.
      (2000). Signal peptide-dependent protein transport in Bacillus subtilis: a genome-based survey of the secretome. Microbiol Mol Biol Rev 64, 515–547. doi:10.1128/MMBR.64.3.515-547.2000pmid:10974125
      1. Wood, I. P.,
      2. Elliston, A.,
      3. Ryden, P.,
      4. Bancroft, I.,
      5. Roberts, I. N. &
      6. Waldron, K. W.
      (2012). Rapid quantification of reducing sugars in biomass hydrolysates: Improving the speed and precision of the dinitrosalicylic acid assay. Biomass Bioenergy 44, 117–121. doi:10.1016/j.biombioe.2012.05.003
      1. Yanisch-Perron, C.,
      2. Vieira, J. &
      3. Messing, J.
      (1985). Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103–119. doi:10.1016/0378-1119(85)90120-9pmid:2985470
      1. Zambare, V. P.,
      2. Bhalla, A.,
      3. Muthukumarappan, K.,
      4. Sani, R. K. &
      5. Christopher, L. P.
      (2011). Bioprocessing of agricultural residues to ethanol utilizing a cellulolytic extremophile. Extremophiles 15, 611–618. doi:10.1007/s00792-011-0391-2pmid:21837419
      1. Zeng, L. &
      2. Burne, R. A.
      (2009). Transcriptional regulation of the cellobiose operon of Streptococcus mutans. J Bacteriol 191, 2153–2162. doi:10.1128/JB.01641-08pmid:19168613

    References

      1. Assareh, R.,
      2. Shahbani Zahiri, H.,
      3. Akbari Noghabi, K.,
      4. Aminzadeh, S. &
      5. Bakhshi Khaniki, G.
      (2012). Characterization of the newly isolated Geobacillus sp. T1, the efficient cellulase-producer on untreated barley and wheat straws. Bioresour Technol 120, 99–105. doi:10.1016/j.biortech.2012.06.027pmid:22784959
      1. Bartosiak-Jentys, J.,
      2. Eley, K. &
      3. Leak, D. J.
      (2012). Application of pheB as a reporter gene for Geobacillus spp., enabling qualitative colony screening and quantitative analysis of promoter strength. Appl Environ Microbiol 78, 5945–5947. doi:10.1128/AEM.07944-11pmid:22685159
      1. Brautaset, T.,
      2. Lale, R. &
      3. Valla, S.
      (2009). Positively regulated bacterial expression systems. Microb Biotechnol 2, 15–30. doi:10.1111/j.1751-7915.2008.00048.xpmid:21261879
      1. Brockmeier, U.,
      2. Caspers, M.,
      3. Freudl, R.,
      4. Jockwer, A.,
      5. Noll, T. &
      6. Eggert, T.
      (2006). Systematic screening of all signal peptides from Bacillus subtilis: a powerful strategy in optimizing heterologous protein secretion in Gram-positive bacteria. J Mol Biol 362, 393–402. doi:10.1016/j.jmb.2006.07.034pmid:16930615
      1. Brown, N. L.,
      2. Stoyanov, J. V.,
      3. Kidd, S. P. &
      4. Hobman, J. L.
      (2003). The MerR family of transcriptional regulators. FEMS Microbiol Rev 27, 145–163. doi:10.1016/S0168-6445(03)00051-2pmid:12829265
      1. Chhabra, S. R.,
      2. Shockley, K. R.,
      3. Ward, D. E. &
      4. Kelly, R. M.
      (2002). Regulation of endo-acting glycosyl hydrolases in the hyperthermophilic bacterium Thermotoga maritima grown on glucan- and mannan-based polysaccharides. Appl Environ Microbiol 68, 545–554. doi:10.1128/AEM.68.2.545-554.2002pmid:11823189
      1. Cripps, R. E.,
      2. Eley, K.,
      3. Leak, D. J.,
      4. Rudd, B.,
      5. Taylor, M.,
      6. Todd, M.,
      7. Boakes, S.,
      8. Martin, S. &
      9. Atkinson, T.
      (2009). Metabolic engineering of Geobacillus thermoglucosidasius for high yield ethanol production. Metab Eng 11, 398–408. doi:10.1016/j.ymben.2009.08.005pmid:19703579
      1. Deutscher, J.
      (2008). The mechanisms of carbon catabolite repression in bacteria. Curr Opin Microbiol 11, 87–93. doi:10.1016/j.mib.2008.02.007pmid:18359269
      1. Deutscher, J.,
      2. Francke, C. &
      3. Postma, P. W.
      (2006). How phosphotransferase system-related protein phosphorylation regulates carbohydrate metabolism in bacteria. Microbiol Mol Biol Rev 70, 939–1031. doi:10.1128/MMBR.00024-06pmid:17158705
      1. Gibson, D. G.,
      2. Benders, G. A.,
      3. Andrews-Pfannkoch, C.,
      4. Denisova, E. A.,
      5. Baden-Tillson, H.,
      6. Zaveri, J.,
      7. Stockwell, T. B.,
      8. Brownley, A.,
      9. Thomas, D. W.,
      10. et al.
      (2008). Complete chemical synthesis, assembly, and cloning of a Mycoplasma genitalium genome. Science 319, 1215–1220. doi:10.1126/science.1151721pmid:18218864
      1. Gibson, D. G.,
      2. Young, L.,
      3. Chuang, R.-Y.,
      4. Venter, J. C.,
      5. Hutchison, C. A. III. &
      6. Smith, H. O.
      (2009). Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat Methods 6, 343–345. doi:10.1038/nmeth.1318pmid:19363495
      1. Lai, X. K. &
      2. Ingram, L. O.
      (1993). Cloning and sequencing of a cellobiose phosphotransferase system operon from Bacillus stearothermophilus XL-65-6 and functional expression in Escherichia coli. J Bacteriol 175, 6441–6450.pmid:8407820
      1. Liao, H.,
      2. McKenzie, T. &
      3. Hageman, R.
      (1986). Isolation of a thermostable enzyme variant by cloning and selection in a thermophile. Proc Natl Acad Sci U S A 83, 576–580. doi:10.1073/pnas.83.3.576pmid:3003740