Abstract
The solventogenic C. beijerinckii DSM 6423, a microorganism that naturally produces isopropanol and butanol, was previously modified by random mutagenesis. In this work, one of the resulting mutants was characterized. This strain, selected with allyl alcohol and designated as the AA mutant, shows a dominant production of acids, a severely diminished butanol synthesis capacity, and produces acetone instead of isopropanol. Interestingly, this solvent-deficient strain was also found to have a limited consumption of two carbohydrates and to be still able to form spores, highlighting its particular phenotype. Sequencing of the AA mutant revealed point mutations in several genes including CIBE_0767 (sigL), which encodes the σ54 sigma factor. Complementation with wild-type sigL fully restored solvent production and sugar assimilation and RT-qPCR analyses revealed its transcriptional control of several genes related to solventogensis, demonstrating the central role of σ54 in C. beijerinckii DSM 6423. Comparative genomics analysis suggested that this function is conserved at the species level, and this hypothesis was further confirmed through the deletion of sigL in the model strain C. beijerinckii NCIMB 8052.
Similar content being viewed by others
Introduction
In the context of worldwide energy transition, research for alternatives to fossil fuels has become a priority. In particular, the replacement of petrochemistry by a low carbon emission industry has been a major challenge as our global consumption of petrochemicals keeps on increasing1. The valorization of plant biomass to synthesize ethanol by microbial fermentation has already been pioneered for biofuel production2 and could therefore be applied to bio-based chemistry3.
A few strains from the Clostridium genus are naturally able to produce isopropanol and butanol4,5, two compounds that could be used as biochemical and biofuel, respectively. However, those organisms are not producing these metabolites in quantities compatible with an economically viable industrial process6. However, with the increasing availability of efficient genetic tools in Clostridia7, metabolic engineering approaches could be undertaken to enhance solvent productivity. Clostridium beijerinckii DSM 6423 (NRRL B-593) is the only natural isopropanol-butanol producing strain whose genome and transcriptome have been investigated8. It may therefore be the best candidate for genetic engineering, although its particular physiology is still poorly understood. For this purpose, gaining additional knowledge on metabolism regulation in this strain would greatly benefit future optimization efforts. In particular, identifying the molecular effectors controlling solvent production may provide valuable insights to define adequate genetic engineering strategies.
As no genetic toolbox was available for this particular strain, Máté de Gerando and coworkers performed random mutagenesis coupled with genome shuffling to increase isopropanol productivity by selecting isopropanol-tolerant strains9. In this work, random mutagenesis followed by allyl alcohol selection also generated an interesting mutant, further referred to as AA mutant. This strain mainly produces acids, shows no isopropanol production and a strongly attenuated butanol synthesis capacity. These results are consistent with those obtained in Clostridium acetobutylicum DSM 1792, in which mutants obtained in the presence of allyl alcohol - precursor of the highly toxic acrolein molecule in the reaction catalyzed by alcohol dehydrogenases - permitted the selection of butanol-deficient strains10. Nevertheless, in both cases the key mutated genes causing these phenotypes have not been clearly identified.
In this study, we demonstrate that the solvent production deficiency in the AA mutant is due to a point mutation in the CIBE_0767 (sigL) gene, which encodes the transcriptional regulator σ54 (also referred to as σL). Similarly to other sigma factors, σ54 regulates genetic expression by incorporating the RNA polymerase complex and binding to specific promoter sequences, thus enabling selective transcription of a subset of genes11,12,13. The importance of this sigma factor for solventogenesis was further confirmed at the species level by deleting the corresponding gene in the C. beijerinckii model strain (i.e. NCIMB 8052). In addition to regulating alcohol synthesis pathways, our experiments revealed that this sigma factor also controls utilization of alternative carbon sources, such as lactose and cellobiose, and is not required to complete the sporulation process, making it a central and specific controller of carbon metabolism in C. beijerinckii.
Results
The C. beijerinckii AA mutant metabolism is orientated towards acid production
Phenotypic characterization of a solventogenic Clostridium is usually assessed by the identification of fermentation products. We therefore performed triplicate fermentation assays of both C. beijerinckii DSM 6423 wild-type and AA strains to compare their product pattern after 48 hours in Gapes medium.
When compared to the wild-type strain, the AA mutant showed a very distinctive behavior. Firstly, growth was strongly impacted with a ca. 2-fold decrease in OD600 units after 48 hours of fermentation in Gapes medium, influencing end-products concentrations and glucose consumption (Fig. 1, Supplementary File S1).
Although butanol was still produced, the central metabolism of the AA strain shifted towards acid formation, with notably a strong increase in butyrate concentration resulting in a final solvent:acid ratio of 1:3, compared to 6:1 in the wild-type strain (Fig. 1c).
The phenotype of the AA mutant, mainly producing butyrate, may result from the coupled inactivation of one or several gene(s) involved in isopropanol and butanol production.
The AA mutant is still able to complete the sporulation process
Complete or partial loss of solventogenesis in the C. beijerinckii species is often associated with a degeneration phenomenon (i.e. loss of the capacity to form spores in addition to the loss of solvent production)14. Given its phenotype, we therefore decided to investigate the sporulation ability of the AA mutant.
A comparative assay confirmed that sporulation ability was conserved, despite an approximate 10/100-fold decrease in the number of spores in comparison to the wild-type strain (Supplementary File S1). However, given the limited growth of the AA strain, this decrease may not necessarily be linked to a genetic deficiency but could instead result from a diminished ability to accumulate population density.
Importantly, demonstration of the sporulation ability in the AA strain makes its phenotype unrelated to the degeneration phenomenon, which suggests that its genome contains solventogenic-specific mutations.
A point mutation in the sigL gene is predicted to impede solvent formation and sugar consumption
We next adopted a forward genetic approach to identify the gene(s) responsible for the AA phenotype. Whole-genome sequencing followed by read mapping on the wild-type genome allowed the identification of 13 SNPs (Table 1) along the chromosome of the strain. Among the affected genes, CIBE_0767 particularly retained our attention as P-BLAST analysis against Swissprot database revealed high similarities with putative RNA polymerase σ54 factors (encoded by the sigL gene) and notably the one from Clostridium kluyveri15. In the AA strain, the SNP detected in the sigL gene causes a serine to phenylalanine substitution at position 366. The σ54 transcriptional factor has already been well described in several bacterial species and contains three major domains12 (Fig. 2a). Region I (RI) is involved in core RNA polymerase, enhancer, and DNA binding. Region II (RII) is acidic and poorly conserved. Region III (RIII) is divided in several conserved sub-regions interacting with core RNA polymerase (CBD) and DNA at the -12 consensus motif (HTH) and the -24 motif (RpoN). In silico analysis of the sigL gene revealed that the amino acid substitution (S366F) is localized in the HTH motif, at an extremely conserved position (Fig. 2b). Given the significant difference in terms of steric hindrance, polarity and hydrophobicity resulting from a phenylalanine to serine substitution, this mutation may drastically impede the -12-motif recognition, yielding a partial or total inactivation of σ54.
We subsequently looked at the predicted σ54–regulon along the chromosome of C. beijerinckii DSM 6423. The σ54 binding motif consensus (TGGCANNNNNNTTGCW), based on the study of validated σ54 promoters by Barrios and colleagues13, was found at 57 genomic sites. Among them, 35 were localized up to a few hundred base pairs upstream of a coding sequence (Supplementary File S2). In particular, two genes coding for strongly expressed8 alcohol dehydrogenases (CIBE_2050 and CIBE_2622, which convert acetaldehyde and butyraldehyde into ethanol and butanol, respectively)16,17 as well as the secondary alcohol dehydrogenase s-ADH encoding gene (CIBE_3470, whose product converts acetone to isopropanol)16,18,19 were identified (Fig. 2c). Although 2,3-butanediol is usually not detected in our culture conditions, the only gene predicted to encode a 2,3-butanediol dehydrogenase (CIBE_1696, similar to the one described by Raedts et al.)20 was identified as being part of the regulon.
Unlike other sigma factors, σ54-bound RNA polymerase requires the ATPase activity of an adjacent Enhancer Binding Protein (EBP) to start transcription initiation21,22. Importantly, genes encoding σ54 EBPs were found in the direct vicinity of CIBE_2622, CIBE_3470 and CIBE_1696, which strengthens the hypothesis of a σ54-driven transcriptional control of their expression.
In order to investigate the functionality of the σ54-regulon, we used the synteny tool from MaGe platform (16) to compare the position of σ54 binding sites in C. beijerinckii DSM 6423 with two other C. beijerinckii model strains (C. beijerinckii NCIMB 8052 and the recently reannotated NRRL B-598)23,24. Apart from the s-ADH encoding gene which drives isopropanol production and is only present in the DSM 6423 strain, these putative σ54–dependent transcriptional units appeared to be conserved in these acetone-butanol producing strains. Besides, downregulation of the butanol and secondary alcohol dehydrogenases linked to σ54 potential inactivation would match the AA phenotype, which gives another hint that sigL mutation may be significant.
Interestingly, other genes predicted to be part of the σ54-regulon are mainly associated with sugar uptake (mainly PTS-based transporters, for a review in solventogenic Clostridia, see Mitchell, 2015)25 and metabolism (Fig. 2d, Supplementary File S2). Lactose and cellobiose transport and metabolism are also predicted to be regulated by σ54.These operons contain genes encoding a phosphotransferase system (PTS) for sugar transport and a hydrolase (β-galactosidase CIBE_0885; β-glucosidase CIBE_1708). Interestingly, a highly expressed and conserved operon encoding a dihydroxyacetone phosphate kinase complex (dhaKLM; CIBE_2581-2582-2583) also appears to be controlled by σ54. This enzyme can be involved in glycerol metabolism, or directly in the central carbon pathway when associated to 6P-fructose aldolases26,27. Notably, two 6P-fructose aldolases (CIBE_0334 and CIBE_0411), one of which being well expressed in the available RNA-seq dataset8, are annotated in the DSM 6423 genome. Lastly, parts of the pentose phosphate pathway also seem to be regulated by σ54, with two conserved similar operons encoding transketolases, ribose/ribulose phosphate epimerases and isomerases (Supplementary File S2). Importantly, some of these putative σ54–dependent transcriptional units appeared to be well conserved in the NCIMB 8052 and NRRL-B598 strains (Fig. 2d).
In summary, analysis of the AA genotype revealed an amino acid substitution potentially inactivating a protein predicted not only to regulate solventogenesis but also other central metabolic pathways such as sugar consumption in several C. beijerinckii strains.
Complementation with wild-type sigL gene fully restores solventogenesis and sugar metabolism in the AA strain
As several SNPs were detected in the genome of the AA mutant, complementation assays were required to characterize the corresponding phenotype. Considering the predicted crucial role of the putative σ54-encoding gene (sigL, CIBE_0767), we decided to clone the wild-type coding sequence of sigL under the control of its endogenous promoter in the pFW01 vector, yielding the pFW-σ54 plasmid, and to introduce it into both wild-type and AA strains. Fermentation analysis revealed that both WT and AA pFW-σ54 strains displayed a wild-type solventogenic metabolism (Fig. 3). Indeed, normal acid and solvent production (Fig. 3a), as well as glucose consumption and biomass production (Supplementary File S3) were fully restored in the AA pFW-σ54 strain when compared to the wild-type strain containing the empty vector. Interestingly, the WT pFW-σ54 recombinant strain did not exhibit any major change when compared to the wild-type microorganism.
These results confirmed that CIBE_0767, encoding the σ54 transcriptional factor, is drastically influencing the regulation of the central metabolism of the AA strain in our conditions. We therefore kept on investigating its impact on other predicted metabolic pathways.
Plate assays on minimal medium were used to analyze carbon sources usage. In our conditions, no effect after 24 or 48 hours was seen on bacterial growth on fructose, mannose, xylose, arabinose, sucrose, glycerol, inulin, starch or mannitol (data not shown). In contrast, growth was impeded with cellobiose and lactose in the AA control strain (empty vector), and restored when complemented with wild-type σ54 (Fig. 3b). However, quantification of this inhibition in liquid medium revealed unexpected results as growth was not altered on 2YT 20 g/L cellobiose medium (data not shown) whereas lactose did exert a strong effect on bacterial growth. After 24 h of fermentation the AA pFW01 strain had grown similarly on 2YT 20 g/L lactose than on the 2YT control condition containing no supplementary carbon source (Fig. 3c). HPLC analysis confirmed that almost no lactose was consumed by the AA strain in this time frame. On the other hand, normal growth on lactose was observed with the complementation strain with a consumption of ca. 8 g/L of lactose in 24 h. The fermentation was carried out for 72 more hours, and end-point analysis revealed that lactose uptake inhibition was only partial (Supplementary File S4).
Transcriptional control of the predicted σ54 regulon was further assessed by performing RT-qPCR on the WT/AA strains, with and without sigL complementation (Fig. 3d). Expression of CIBE_2050 (adh-1), CIBE_2622 (adh-2) and CIBE_3470 (s-adh) was decreased (c.a. 2.5–5 fold) in the AA strain and complementation resulted in a drastic increase of their expression (≈10 times), congruent with the strong expression of sigL on a multi-copy plasmid. This stronger expression however did not result in a higher titer of solvents as outlined before. Alcohol synthesis may thus be limited by another factor (e.g. butanol tolerance or another rate-limiting step in the solventogenic pathways) in C. beijerinckii DSM 6423.
The sigL expression level was interestingly much lower in the WT strain than in the AA mutant, which could be linked to a potential feedback regulation.
As expected, expression of the predicted σ54-independent ald gene (CIBE_4606, first gene of the sol operon)8,28 did not vary among these strains. Similarly, expression of two σ54 EBPs (CIBE_2621, CIBE_3471) was not affected, indicating that their expression is not driven by σ54. This result was also anticipated, since these genes are not expected to be regulated by σ54.
Altogether, these results indicate that the S366F substitution in the σ54 is likely the unique modification in the AA strain that impacts solvent production and sugar assimilation. Moreover, our results suggest that σ54 is a master regulator of these metabolic pathways in C. beijerinckii DSM 6423.
Overexpression of the acid-reassimilation pathway does not restore a wild-type phenotype in the AA strain
The AA phenotype could be linked to two possible mechanisms: a deficient alcohol dehydrogenase activity in the cell and/or a defect in the acid uptake pathway.
To verify whether acid uptake might be impaired in the AA strain, we designed an experiment in which the acid reassimilation pathway would be overexpressed in the AA mutant. In order to do this, a pFW-FC06 plasmid was constructed to overexpress the s-adh, ctfA, and ctfB genes in an operonic structure. This operon was previously used in an overexpression plasmid29,30,31 or integrated in the chromosome32 to convert the model acetone-butanol producing organism C. acetobutylicum ATCC 824 into a performant isopropanol-butanol producer.
Introduction of the pFW-FC06 vector slightly increased solvent production and glucose consumption in the AA strain, compared to the introduction of an empty vector (Supplementary File S5). However, even if acetone was fully converted to isopropanol, the recombinant AA strain containing pFW-FC06 still accumulated butyrate at levels comparable to the AA mutant. The latter observation suggests that, consistently with in silico predictions, a CtfA/B complex is still functional in the AA mutant (i.e. overexpression of ctfA/B does not have much effect on acid levels, similarly to the wild-type strain). The main metabolism roadblock thus probably comes from the aforementioned deficient butanol synthesis pathway. Indeed, in the AA strain, this result would be expected if cyclic conversion of butyryl-CoA to butyrate, and back to butyryl-CoA via the CtfA/B enzyme, was occurring.
In summary, this experiment gives an additional insight on the molecular mechanisms triggering solventogenesis in C. beijerinckii: by regulating alcohol dehydrogenase expression, σ54-mediated transcription may be able to modulate the balance between acidogenesis and solventogenesis.
CRISPR/Cas9-mediated deletion of sigL in C. beijerinckii NCIMB 8052 results in an AA-like phenotype
Complementation with wild-type σ54 in the AA strain shed light on sigL role in solventogenesis and sugar uptake. However, given the many mutations found along the genome of the AA mutant, we aimed at confirming that the observed phenotype was the unique effect of sigL inactivation by deleting it in a wild-type strain. The model strain C. beijerinckii NCIMB 8052 was chosen as a chassis for this modification since, unlike the DSM 6423 strain, markerless genome editing techniques have already been demonstrated to be effective in this microorganism33,34,35. More particularly, genetic modifications using CRISPR/Cas9-based tools were recently made available to the Clostridium community: this technology comes with several advantages (i.e. genome editing overall efficiency, simplicity and rapidity), particularly interesting for genetic engineering of the notably difficult-to-modify clostridial strains, as reviewed recently by Xue et al.36.
A genome editing strategy based on the one described by Wasels et al.32 was undertaken. Briefly, an anhydrotetracycline-inducible system allows the selection of double allelic exchange events between the clostridial DNA and a user-defined editing template, by using the Cas9/gRNA complex to target the unmodified bacterial chromosome. Double strand breaks thus only occur in the wild-type or single allelic exchange cases, resulting in cell death.
A plasmid based on the pCas9ind vector was constructed to comprise, in addition to the inducible cas9 gene, an anhydrotetracycline-inducible gRNA expression cassette and a gene editing template designed to delete most of the sigL gene (i.e. gene truncation results in a 13 AA protein, Fig. 4a). Introduction of this plasmid in C. beijerinckii NCIMB 8052 and induction of the CRISPR/Cas9 machinery permitted the selection of ΔsigL clones (Fig. 4b, Supplementary File S7).
One of those ΔsigL clones was subsequently complemented with the pFW-σ54 and pFW-σ54-AA plasmids, the latter allowing the expression of the S336F-mutated σ54.
Fermentation assays revealed that solvent production was drastically altered in the ∆sigL strain, and that a wild-type phenotype could only be restored with pFW-σ54 (Fig. 4c, Supplementary File S7). Complementation with pFW-σ54-AA notably appeared to have no effect on acid reassimilation and solvent production when compared to the ∆sigL empty vector control strain. Besides, all of those strains displayed a resistance to heat shock (cells scraped from one-week-old plates, 80 °C for 10 min, data not shown), highlighting their retained capacity to correctly form spores. Lastly, ΔsigL mutant growth appeared to be highly inhibited in lactose-based medium, similarly to the AA strain (Fig. 4d). Cellobiose uptake however did not seem to be impacted in this strain (data not shown).
In addition to confirming our hypothesis on the role of σ54 for solventogenesis and sugar utilization, deleting sigL in the NCIMB 8052 model strain allowed us to expand our conclusions regarding this transcriptional factor to other microorganisms belonging to the C. beijerinckii species.
Discussion
In this study, the allyl alcohol mutant generated from C. beijerinckii DSM 6423 by Máté de Gerando et al.9 was genetically characterized by a forward genetic approach. Complementation assays demonstrated that the AA phenotype is due to a unique point mutation in the CIBE_0767 sigL gene encoding the σ54 transcriptional factor. This mutation leads to the substitution of a highly conserved serine in the HTH domain by a phenylalanine residue (S336F). Given the central role of the HTH domain in DNA recognition37,38, this mutation is likely to strongly impede the σ54-RNA polymerase complex binding to promoter sequences and thus to inhibit transcription of the σ54-regulon. If the effect of this particular mutation on σ54 functionality has not been described so far, Coppard and Merrick38 performed targeted mutagenesis of the corresponding serine of the HTH domain of K. pneumoniae. Its substitution by most residues (especially large ones like lysine or tryptophane) completely shut down σ54-mediated transcription, suggesting this serine is crucial for a functional activity, which supports our hypothesis that σ54 is inactivated in the AA strain.
The impact of the S336F mutation in the σ54 sequence was experimentally investigated by introducing expression plasmids of either the wild-type or the S336F version of σ54 in a ΔsigL mutant of the model strain C. beijerinckii NCIMB 8052. This mutant displayed a phenotype similar to the AA strain, solventogenesis being severely impaired, which resulted in a butyrate-orientated metabolism. Complementation with pFW-σ54 fully restored solventogenesis, while the introduction of the pFW-σ54-AA vector in the ΔsigL strain had no effect. These experiments overall demonstrated that the S336F mutation completely inactivates σ54 and are thus consistent with our previous results.
The deleterious effects of allyl alcohol on alcohol-producing microorganisms (yeast39,40,41, E. coli42) and its tendency to generate spontaneous mutations have been reported decades ago. Reverse activity of alcohol dehydrogenases catalyzes the formation of the highly toxic acrolein molecule from allyl alcohol, enabling the selection of alcohol dehydrogenase deficient mutants. In Clostridia, similar work was pioneered by Dürre and colleagues10, who obtained C. acetobutylicum mutants unable to produce butanol. Interestingly, butyraldehyde dehydrogenase activity was drastically reduced in these strains. Given the in silico σ54-targeted promoter predictions and the expression pattern of the ald gene in the WT and AA strains, σ54 apparently does not control the butyraldehyde dehydrogenase genes transcription in the DSM 6423 strain. The supposed inactivation of σ54 in the AA mutant rather seems to lead to a decreased expression of two major alcohol dehydrogenases genes, resulting in a phenotype close to what was observed in the C. acetobutylicum corresponding mutants.
Acetate and butyrate are normally produced during the acidogenic phase and are further consumed during solventogenesis43. In the AA mutant, increased acid production results in a substantial drop of the pH which may prematurely cause cell growth to abort. qPCR data revealed that some normally highly expressed alcohol dehydrogenase genes were underexpressed in the AA strain, which suggests a link between the increase of acid production and the partial loss in butanol production capacity. The significant upregulation of those genes transcription when sigL was overexpressed in both the WT and AA strains further consolidates this hypothesis.
Isopropanol production was also abolished in the AA strain but its precursor - acetone - was still detected. As expected, expression of the sole secondary alcohol dehydrogenase (s-ADH, encoded by the CIBE_3470 gene), which catalyzes the acetone-to-isopropanol reaction, was highly affected in this strain. Being linked to acid consumption, acetone formation indicates that butyrate and acetate uptake still occurs. In the DSM 6423 wild-type strain, the resulting acetyl-CoA and butyryl-CoA are further reduced into acetaldehyde/butyraldehyde and then ethanol/butanol. Because alcohol synthesis is drastically reduced in the AA strain, this implies that most of the acetyl-CoA and butyryl-CoA are converted back to acetate and butyrate first by the CtfA/B CoA-transferase44,45 and then by phosphotransbutyrylase/acetylase – butyrate/acetate kinase pathway46,47,48,49,50,51. This assumption would be supported by the increased (acetone + isopropanol):butanol ratio in the AA strain (close to 1:1, versus 1:3 in the wild-type). Indeed, this ratio is expected to be higher if butyryl-CoA is preferentially used to synthesize butyrate instead of butanol.
Among acidic products, butyrate synthesis seems to be preferred over acetate production in the AA strain. Acetate was indeed consumed in most – but not all - fermentation experiments (i.e. final concentration in the fermentation broth less than the 2.4 g/L contained in the Gapes medium) involving the AA mutant or the NCIMB 8052 ΔsigL strain, which also supports the hypothesis that acid uptake does occur. This also suggests that the butyryl-CoA synthesis operon (i.e. genes linking acetyl-CoA and butyryl-CoA, namely: hbd, crt, bcd, etfA, etfB52,53,54), is still constitutively expressed in the AA mutant, similarly to the WT strain8. Besides, overexpression of the acid reassimilation pathway (CtfA/B, s-ADH) in the AA strain did not increase acid consumption. In accordance with in silico predictions, our data thus strongly suggest that σ54 main mode of action consists in driving the regulation of solventogenesis by modulating alcohol dehydrogenases expression and not by interfering on the bacterial acid uptake capacity.
σ54 was also shown to control sugar uptake and usage, consistently with what Nie et al. predicted for the Clostridium genus55. In particular, growth was limited on cellobiose and lactose in the AA strain, though cellobiose inhibition could only be observed on minimal medium. This may be explained by the existence of another cellobiose assimilation pathway in the DSM 6423 genome. In the NCIMB 8052 strain, sigL deletion was demonstrated to impact lactose uptake, confirming that σ54 is involved in sugar uptake and/or metabolism in C. beijerinckii.
σ54 belongs to a unique category of sigma factors, being evolutionary distinct to the σ70 family, which comprises all of the other sigma factors56,57. This difference implies a very different mode of action for the σ54-RNA polymerase complex: unlike its σ70 counterpart, the holoenzyme is unable to initiate transcription without the help of an enhancer binding protein (EBP)22. EBPs are often acting several hundred of base pairs upstream of the σ54-holoenzyme deposition sites and contain a central AAA+ domain. The latter provides, through ATP hydrolysis, the energy necessary for structural rearrangements within the σ54-bound holoenzyme, which in turn allows transcription initiation. EBP control on σ54-dependant transcription is mainly exerted by a signal-sensing domain, which upon specific environmental stimuli activates or represses its ATPase activity22. This enhancer-dependent system offers a very tight control on gene expression, and σ54 regulons are often associated with various biological processes that require a stringent control (e.g. virulence58,59,60 or biofilm formation61,62,63). Multiple σ54 EBPs were found in C. beijerinckii genomes and are thus likely to be the primary effectors of σ54-mediated regulation. Indeed, RNA-seq analysis in the C. beijeirinckii DSM 6423 wild-type strain8 shows that sigL is not differentially expressed over the time-course of a batch fermentation, whereas its target solventogenic genes (CIBE_2050, CIBE_2622, CIBE_3470) are among the top upregulated genes during the transition from acidogenesis to solventogenesis, which suggests that activation of their transcription is mediated by other factors (i.e. the enhancer binding proteins). RT-qPCR data also suggest that, at least in some cases, expression of sigL is regulated, which might provide an additional mean to regulate alcohol synthesis and/or sugar uptake. A well-known example of EBP-related regulation is the σ54-driven transcription of the levanase operon in B. subtilis, which encodes an enzymatic and transport machinery allowing the degradation and consumption of fructose polymers64,65. The enhancer binding protein LevR was shown to mediate the activation and repression of the expression of this operon by a complex interplay with the PTS system66. In C. beijerinckii, the exact stimuli recognized by EBPs and underlying regulatory mechanisms that allow transcription of σ54 target genes nonetheless still remain to be clarified.
In bacteria, the housekeeping sigma factor (e.g. σ70 in E. coli) integrates RNA polymerase and drives transcription of most housekeeping genes. However, bacterial genomes encode multiple alternative sigma factors which direct RNA polymerase to different promoters, providing a simple but broadly exploited strategy to regulate genetic expression67,68. In solventogenic Clostridia, alternative sigma factors have mainly been highlighted for their crucial role in the regulation of the multi-stages sporulation cascade69,70,71. In C. beijerinckii, no sporulation-specific targets could be found in the predicted σ54 regulons, and σ54 inactivation did not impede the sporulation cascade, which underlines its role as a carbon-specific regulator in this bacteria.
In summary, we described σ54 as a master regulator of solventogenic pathways at the species level in C. beijerinckii. This transcriptional factor was also found to control sugar consumption and is therefore an essential controller of carbon metabolism in C. beijerinckii.
Material and Methods
Strains, media and culture conditions
Strains and plasmids used in this study are presented in Table 2.
Clostridium beijerinckii was grown in liquid 2YTG (per liter: 16 g tryptone, 10 g yeast extract, 5 g NaCl, 20 g glucose, pH 5.2). Solid media was prepared with 15 g/L agar and less glucose (5 g/L). Strains were cultivated in an anaerobic chamber (Bactron) at 34 °C without shaking.
Escherichia coli was grown in liquid or solid LB media, in aerobic conditions at 37 °C with 180 rpm agitation when necessary.
Whole-genome sequencing
Genomic DNA of the AA strain was purified with the GenElute Bacterial Genomic DNA Kit (Sigma-Aldrich) and was subsequently sequenced on an Illumina MiSeq sequencer (2 × 250 paired-end reads). Mutations were detected by read mapping against the C. beijerinckii DSM 6423 genome8 using Geneious R1072.
Predictions of σ54-regulons
Genomes of the C. beijerinckii DSM 6423, NRRL B-598 and NCIMB 8052 strains were scanned for the σ54 consensus motif (TGGCANNNNNNTTGCW13, no mismatches allowed) using Geneious R1072. Genes were considered potentially regulated by σ54 if motifs were found up to 500 bp upstream their coding sequences. Predicted candidate genes/operons were then compared in the three strains using the synteny tool from MaGe platform73.
Fermentation assays and analytical methods
For fermentation assays, modified Gapes medium74 was used. This medium contains per liter: 2.5 g yeast extract, 1 g KH2PO4, 0.6 g K2HPO4, 1 g MgSO4, 7 H20, 6.6 mg FeSO4, 7 H20, 0.1 g para-aminobenzoic acid, 2.9 g ammonium acetate, 60 or 80 g glucose. The medium was supplemented with erythromycin (20 µg/mL) when appropriate.
Fermentations were performed in triplicate. For each biological replicate, several clones isolated in 2YTG agar plates were used to inoculate 5 mL Gapes medium. These precultures were carried out at 34 °C overnight in the anaerobic chamber. Serum flasks containing 20 mL Gapes medium were then inoculated with 2 mL of preculture and sealed with rubber stoppers. A pressure relief valve system was punctured through the rubber stoppers to prevent overpressure, and the serum bottles were incubated 48–72 hours at 34 °C with agitation outside of the anaerobic chamber.
After OD600 (UV-1800 spectrophotometer, Shimadzu) was measured, samples from the fermentation were centrifuged 5 min at 8000 g. The supernatant was diluted with an internal standard (1-propanol: final concentration 0.5 g/L). Metabolites concentrations in the supernatant were quantified by chromatography.
Gas chromatography (Porabond-Q column from Agilent technologies, 25 m length, 0.32 internal diameter, 5 µm film thickness coupled to a flame ionization detector) was performed to determine solvent concentrations. Helium was used as a carrier gas at a flowrate of 1.6 mL/min. Column was gradually heated from 50 °C to 250 °C in a 30 min run.
Acid concentrations were quantified by HPLC (Aminex HPX-87H from Biorad coupled to a Spectra System RI 150 refractometer and a Waters 2487 Dual λ UV detector set at 210 nm). 0.01 M sulfuric acid mobile phase was used at a flowrate of 0.6 mL/min. Column temperature was set at 60 °C.
Residual sugar quantities were determined by HPLC using an Aminex HXP-87P (Bio-Rad) coupled to a Varian 350 RI refractometer for detection. Water was used as a mobile phase at a flowrate of 0.4 mL/min. When quantifying sugar concentration for 2YT-based samples, a Micro-Guard De-Ashing Refill cartridge (Bio-Rad) was added to the chromatographic system for prior desalting of the samples. Column temperature was set at 80 °C.
Plasmid construction and transformation
All primers used in this study are listed in Supplementary File S6.
The sigL gene (CIBE_0767) from C. beijerinckii DSM 6423 together with its endogenous promoter were amplified by PCR with primers RH077 and RH078. The pFW01 backbone as well as the thiolase terminator were amplified by PCR from plasmid pCas9con32 with primers RH086 and RH087. Those two amplicons were subsequently assembled by Gibson assembly (NEB)75 to yield pFW-σ54. From this plasmid, pFW-σ54-AA was obtained by site-directed mutagenesis with primers RH136 and RH137.
The sadh gene (CIBE_3470) from C. beijerinckii DSM 6423 and the ctfA and ctfB genes from C. acetobutylicum ATCC 824 were amplified by PCR with primers MKz01 and MKz02 from plasmid pFC00629. The resulting PCR product was cloned between the KpnI and SalI restriction sites to obtain the pFW-FC06 vector.
For the deletion of sigL in C. beijerinckii NCIMB 8052, two homology regions and an anhydrotetracycline-inducible gRNA expressing cassette were first synthesized by BaseClear in a pUC57 plasmid. Homology regions were designed to delete a 736 bp fragment in the sigL gene, yielding a truncated ORF (13 amino acid residues instead of 463). This cassette was amplified with primers RH125 and RH126 and subsequently cloned into the pCas9ind vector32 at the XhoI restriction site. The gRNA protospacer (designed using Genious R10)72 was then introduced in the resulting vector by Golden Gate assembly at the BsaI restriction site with primers guide_sigL_fwd and guide_sigL_rev, yielding pCas9-∆sigL.
The plasmids were then isolated and electroporated into C. beijerinckii as previously described76,77. dam+ dcm+ DNA and dam− dcm− DNA were used for C. beijerinckii NCIMB 8052 and DSM 6423, respectively.
qPCR assays
Starting from fresh precultures, 20 mL bacterial cultures were carried out anaerobically in duplicates for 16 hours at 34 °C and 120 rpm, starting with an initial OD600 value of ≈0.1. Ten mL samples were stabilized on ice by addition of a cold stop solution (ethanol:acid phenol at 1:18 ratio), centrifuged, and the cell pellets were frozen at −80 °C.
TRI reagent (Ambion) was further used to lyse the cells. RNAs were then extracted with the Direct-zol RNA purification kit (Zymo Research). After DNase treatment (Turbo DNase, Invitrogen), RNAs were column-purified with the RNA Clean & Concentrator 25 kit (Zymo Research).
For reverse transcription, 100 ng or RNAs were used in 20 µL reactions (iScript cDNA Synthesis Kit, Bio-Rad). Of the resulting cDNAs, 0.5 µL were then used for qPCR (SsoAdvanced Universal SYBR Green Supermix, Bio-Rad) along with adequate controls (no RT controls, positive controls on gDNA, no template controls). Primers are given with their corresponding qPCR efficiency in Supplementary File S6.
qPCR data were analyzed with the ΔΔCq method, using gapC (CIBE_0769) as a housekeeping gene control and taking into account reaction efficiencies.
CRISPR/Cas9 genome edition
We designed a CRISPR/Cas9 approach to inactivate sigL in the NCIMB 8052 strain. An inducible system based on anhydrotetracycline (aTc) addition was used, based on the pCas9ind vector previously used in C. acetobutylicum ATCC 82432. Homology regions for genome editing were designed to delete most of the sigL gene, resulting in a severely truncated protein (13 amino acid residues, versus 463 for the full gene; Fig. 4a). Following insertion of those homology sequences and of an anhydrotetracycline inducible gRNA cassette, the resulting pCas9-∆sigL plasmid was introduced into C. beijerinckii NCIMB 8052 by electroporation. ∆sigL mutants were obtained similarly to the method described by Wasels et al.32. Briefly, transformants resuspended in liquid 2YTG (pH 5.2, 20 g/L glucose) and serially diluted for spotting onto 2YTG plates containing erythromycin (20 µg/mL) and anhydrotetracycline (50 ng/mL). Isolated colonies were tested by colony PCR with primers RH129 and RH130, encompassing the homology regions in the genome. Positive mutants were subsequently cured by streaking them twice on 2YTG plates supplemented with 50 ng/mL anhydrotetracycline. Plasmid presence in isolated colonies was next tested on 2YTG plates containing erythromycin (20 µg/mL), and erythromycin-sensitive clones were selected for further analysis.
Sporulation assays
Sporulation assays were based on survival to heat shock, as similarly described71. Briefly, triplicates precultures were carried out overnight in 2YTG medium. The following day, serum bottles containing 20 mL 2YTG (15 g/L glucose, pH 6.8) were inoculated with 2 mL of preculture and subsequently sealed. After 5 days at 34 °C without agitation, 1 mL of culture was heat shocked (10 min, 80 °C) and serially diluted. 5 µL spots were made for each dilution on 2YTG plates, which were then incubated overnight at 34 °C.
Sugar utilization assays
For liquid sugar utilization tests, liquid 2YTG (pH 6.8) was used, as well as 2YTL (pH 6.8, 20 g/L lactose) and 2YT (pH 6.8, without purified carbon source). Precultures were carried out overnight from fresh colonies in 2YTG medium (pH 6.8) containing 5 g/L glucose and 20 µg/mL erythromycin at several dilutions (up to 10−4 dilution factor). The following day, fresh precultures were used to inoculate 20 mL (AA strain; serum bottles)/5 mL (ΔsigL strain; 24 deep-well plates) of 2YTG, 2YTL or 2YT media with or without (as needed) erythromycin 20 µg/mL at 0.01 units of OD600. Samples were taken after 24 and 96 hours, and sugar concentration was quantified as described above.
Minimal medium (CDM) was used to visualize sugar uptake inhibition by the AA strain. This medium, described by Vasconcelos et al.78, was supplemented with 15 g/L agar and contained 20 g/L glucose, lactose or cellobiose. Single colonies picked on CDM glucose (5 g/L) plates were serially diluted in water. Spots were made with 5 µL of the 10−1, 10−2, 10−3 and 10−4 dilutions on minimal medium plates containing glucose, cellobiose or lactose (20 g/L). Plates were visualized after 24 h (glucose, cellobiose) or 48 h (lactose).
Data Availability
Raw reads corresponding to the genome sequencing of the AA mutant have been deposited to the Sequence Read Archive (SRA) database (https://www.ncbi.nlm.nih.gov/sra) with the BioProject accession number PRJNA530065.
References
IHS Chemical. Petrochemical Industry Overview. Chemical Economics Handbook (2017).
Naik, S. N., Goud, V. V., Rout, P. K. & Dalai, A. K. Production of first and second generation biofuels. A comprehensive review. Renewable and Sustainable Energy Reviews 14, 578–597, https://doi.org/10.1016/j.rser.2009.10.003 (2010).
Weusthuis, R. A., Aarts, J. M. & Sanders, J. P. M. From biofuel to bioproduct: is bioethanol a suitable fermentation feedstock for synthesis of bulk chemicals? Biofuels, Bioproducts and Biorefining 5, 486–494 (2011).
Chen, J.-S. & Hiu, S. F. Acetone-butanol-isopropanol production by Clostridium beijerinckii (synonym, Clostridium butylicum). Biotechnology Letters 8, 371–376, https://doi.org/10.1007/BF01040869 (1986).
Zhang, C., Li, T. & He, J. Characterization and genome analysis of a butanol–isopropanol-producing Clostridium beijerinckii strain BGS1. Biotechnology for Biofuels 11, 280, https://doi.org/10.1186/s13068-018-1274-x (2018).
Grisales Diaz, V. H. & Olivar Tost, G. Energy efficiency of acetone, butanol, and ethanol (ABE) recovery by heat-integrated distillation. Bioprocess and Biosystems Engineering 41, 395–405, https://doi.org/10.1007/s00449-017-1874-z (2018).
Pyne, M. E., Bruder, M., Moo-Young, M., Chung, D. A. & Chou, C. P. Technical guide for genetic advancement of underdeveloped and intractable Clostridium. Biotechnology advances 32, 623–641, https://doi.org/10.1016/j.biotechadv.2014.04.003 (2014).
Máté de Gérando, H. et al. Genome and transcriptome of the natural isopropanol producer Clostridium beijerinckii DSM6423. BMC genomics 19, 242 (2018).
Máté de Gérando, H. et al. Improving isopropanol tolerance and production of Clostridium beijerinckii DSM 6423 by random mutagenesis and genome shuffling. Applied Microbiology and Biotechnology 100, 5427–5436, https://doi.org/10.1007/s00253-016-7302-5 (2016).
Dürre, P., Kuhn, A. & Gottschalk, G. Treatment with allyl alcohol selects specifically for mutants of Clostridium acetobutylicum defective in butanol synthesis. FEMS Microbiology Letters 36, 77–81 (1986).
Buck, M., Gallegos, M.-T., Studholme, D. J., Guo, Y. & Gralla, J. D. The Bacterial Enhancer-Dependent ς54(ςN) Transcription Factor. Journal of Bacteriology 182, 4129–4136, https://doi.org/10.1128/JB.182.15.4129-4136.2000 (2000).
Zhang, N., Darbari, V. C., Glyde, R., Zhang, X. & Buck, M. The bacterial enhancer-dependent RNA polymerase. Biochemical Journal 473, 3741–3753 (2016).
Barrios, H., Valderrama, B. & Morett, E. Compilation and analysis of sigma(54)-dependent promoter sequences. Nucleic Acids Research 27, 4305–4313 (1999).
Kashket, E. R. & Cao, Z.‐Y. Clostridial strain degeneration. FEMS microbiology reviews 17, 307–315 (1995).
Söhling, B. & Gottschalk, G. Molecular analysis of the anaerobic succinate degradation pathway in Clostridium kluyveri. Journal of Bacteriology 178, 871–880 (1996).
Hiu, S. F., Zhu, C.-X., Yan, R.-T. & Chen, J.-S. Butanol-ethanol dehydrogenase and butanol-ethanol-isopropanol dehydrogenase: different alcohol dehydrogenases in two strains of Clostridium beijerinckii (Clostridium butylicum). Applied and Environmental Microbiology 53, 697–703 (1987).
Welch, R. W., Rudolph, F. B. & Papoutsakis, E. T. Purification and characterization of the NADH-dependent butanol dehydrogenase from Clostridium acetobutylicum (ATCC 824). Archives of Biochemistry and Biophysics 273, 309–318, https://doi.org/10.1016/0003-9861(89)90489-X (1989).
Ismaiel, A. A., Zhu, C. X., Colby, G. D. & Chen, J. S. Purification and characterization of a primary-secondary alcohol dehydrogenase from two strains of Clostridium beijerinckii. Journal of Bacteriology 175, 5097–5105, https://doi.org/10.1128/jb.175.16.5097-5105.1993 (1993).
Korkhin, Y. et al. NADP-dependent bacterial alcohol dehydrogenases. Crystal structure, cofactor-binding and cofactor specificity of the ADHs of Clostridium beijerinckii and Thermoanaerobacter brockii11Edited by R. Huber. Journal of Molecular Biology 278, 967–981, https://doi.org/10.1006/jmbi.1998.1750 (1998).
Raedts, J., Siemerink, M. A. J., Levisson, M., van der Oost, J. & Kengen, S. W. M. Molecular characterization of an NADPH-dependent acetoin reductase/2,3-butanediol dehydrogenase from Clostridium beijerinckii NCIMB 8052. Applied and Environmental Microbiology 80, 2011–2020, https://doi.org/10.1128/AEM.04007-13 (2014).
Rappas, M., Bose, D. & Zhang, X. Bacterial enhancer-binding proteins: unlocking σ54-dependent gene transcription. Current Opinion in Structural Biology 17, 110–116 (2007).
Bush, M. & Dixon, R. The Role of Bacterial Enhancer Binding Proteins as Specialized Activators of σ54-Dependent Transcription. Microbiol. Mol. Biol. Rev. 76, 497, https://doi.org/10.1128/MMBR.00006-12 (2012).
Sedlar, K. et al. Complete genome sequence of Clostridium pasteurianum NRRL B-598, a non-type strain producing butanol. Journal of biotechnology 214, 113–114, https://doi.org/10.1016/j.jbiotec.2015.09.022 (2015).
Sedlar, K., Kolek, J., Provaznik, I. & Patakova, P. Reclassification of non-type strain Clostridium pasteurianum NRRL B-598 as Clostridium beijerinckii NRRL B-598. Journal of biotechnology 244, 1–3 (2017).
Mitchell, W. J. The Phosphotransferase System in Solventogenic Clostridia. Journal of molecular microbiology and biotechnology 25, 129–142, https://doi.org/10.1159/000375125 (2015).
Schürmann, M. & Sprenger, G. A. Fructose-6-phosphate Aldolase Is a Novel Class I Aldolase from Escherichia coli and Is Related to a Novel Group of Bacterial Transaldolases,
Johnson, E. A., Burke, S. K., Forage, R. G. & Lin, E. C. Purification and properties of dihydroxyacetone kinase from Klebsiella pneumoniae. Journal of Bacteriology 160, 55–60 (1984).
Cornillot, E., Nair, R. V., Papoutsakis, E. T. & Soucaille, P. The genes for butanol and acetone formation in Clostridium acetobutylicum ATCC 824 reside on a large plasmid whose loss leads to degeneration of the strain. J Bacteriol 179, https://doi.org/10.1128/jb.179.17.5442-5447.1997 (1997).
Collas, F. et al. Simultaneous production of isopropanol, butanol, ethanol and 2, 3-butanediol by Clostridium acetobutylicum ATCC 824 engineered strains. Amb Express 2, 45 (2012).
Dusseaux, S., Croux, C., Soucaille, P. & Meynial-Salles, I. Metabolic engineering of Clostridium acetobutylicum ATCC 824 for the high-yield production of a biofuel composed of an isopropanol/butanol/ethanol mixture. Metab Eng 18, https://doi.org/10.1016/j.ymben.2013.03.003 (2013).
Lee, J. et al. Metabolic Engineering of Clostridium acetobutylicum ATCC 824 for Isopropanol-Butanol-Ethanol Fermentation. Applied and Environmental Microbiology 78, 1416–1423, https://doi.org/10.1128/AEM.06382-11 (2012).
Wasels, F., Jean-Marie, J., Collas, F., López-Contreras, A. M. & Ferreira, N. L. A two-plasmid inducible CRISPR/Cas9 genome editing tool for Clostridium acetobutylicum. Journal of microbiological methods (2017).
Wang, Y. et al. Bacterial genome editing with CRISPR-Cas9: deletion, Integration, single nucleotide modification, and desirable “clean” mutant selection in Clostridium beijerinckii as an example. ACS synthetic biology 5, 721–732 (2016).
Li, Q. et al. CRISPR‐based genome editing and expression control systems in Clostridium acetobutylicum and Clostridium beijerinckii. Biotechnology journal 11, 961–972 (2016).
Wang, Y. et al. Markerless chromosomal gene deletion in Clostridium beijerinckii using CRISPR/Cas9 system. Journal of biotechnology 200, 1–5 (2015).
Xue, C., Zhao, J., Chen, L., Yang, S.-T. & Bai, F. Recent advances and state-of-the-art strategies in strain and process engineering for biobutanol production by Clostridium acetobutylicum. Biotechnology advances 35, 310–322, https://doi.org/10.1016/j.biotechadv.2017.01.007 (2017).
Merrick, M. & Chambers, S. The helix-turn-helix motif of sigma 54 is involved in recognition of the-13 promoter region. Journal of Bacteriology 174, 7221–7226 (1992).
Coppard, J. R. & Merrick, M. J. Cassette mutagenesis implicates a helix-turn-helix motif in promoter recognition by the novel RNA polymerase sigma factor sigma 54. Mol Microbiol 5, 1309–1317, https://doi.org/10.1111/j.1365-2958.1991.tb00777.x (1991).
Lutstorf, U. & Megnet, R. Multiple forms of alcohol dehydrogenase in Saccharomyces cerevisiae. Archives of Biochemistry and Biophysics 126, 933–944, https://doi.org/10.1016/0003-9861(68)90487-6 (1968).
Ciriacy, M. Genetics of alcohol dehydrogenase in Saccharomyces cerevisiae. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis 29, 315–325, https://doi.org/10.1016/0027-5107(75)90053-6 (1975).
Rando, R. R. Allyl alcohol-induced irreversible inhibition of yeast alcohol dehydrogenase. Biochemical Pharmacology 23, 2328–2331, https://doi.org/10.1016/0006-2952(74)90563-2 (1974).
Lorowitz, W. & Clark, D. Escherichia coli mutants with a temperature-sensitive alcohol dehydrogenase. J. Bacteriol. 152, 935 (1982).
Jones, D. T. & Woods, D. R. Acetone-butanol fermentation revisited. Microbiological Reviews 50, 484–524 (1986).
Wiesenborn, D. P., Rudolph, F. B. & Papoutsakis, E. T. Coenzyme A transferase from Clostridium acetobutylicum ATCC 824 and its role in the uptake of acids. Applied and Environmental Microbiology 55, 323–329 (1989).
Cary, J. W., Petersen, D. J., Papoutsakis, E. T. & Bennett, G. N. Cloning and expression of Clostridium acetobutylicum ATCC 824 acetoacetyl-coenzyme A: acetate/butyrate: coenzyme A-transferase in Escherichia coli. Applied and Environmental Microbiology 56, 1576–1583 (1990).
Winzer, K., Lorenz, K. & Dürre, P. Acetate kinase from Clostridium acetobutylicum: a highly specific enzyme that is actively transcribed during acidogenesis and solventogenesis. Microbiology 143, 3279–3286 (1997).
Boynton, Z. L., Bennett, G. N. & Rudolph, F. B. Cloning, sequencing, and expression of genes encoding phosphotransacetylase and acetate kinase from Clostridium acetobutylicum ATCC 824. Appl Environ Microbiol 62 (1996).
Cary, J. W., Petersen, D. J., Papoutsakis, E. T. & Bennett, G. N. Cloning and expression of Clostridium acetobutylicum phosphotransbutyrylase and butyrate kinase genes in Escherichia coli. Journal of Bacteriology 170, 4613–4618 (1988).
Hartmanis, M. G. Butyrate kinase from Clostridium acetobutylicum. Journal of Biological Chemistry 262, 617–621 (1987).
Wiesenborn, D. P., Rudolph, F. B. & Papoutsakis, E. T. Phosphotransbutyrylase from Clostridium acetobutylicum ATCC 824 and its role in acidogenesis. Applied and Environmental Microbiology 55, 317–322 (1989).
Thompson, D. K. & Chen, J. S. Purification and properties of an acetoacetyl coenzyme A-reacting phosphotransbutyrylase from Clostridium beijerinckii (“Clostridium butylicum”) NRRL B593. Applied and Environmental Microbiology 56, 607–613 (1990).
Boynton, Z. L., Bennet, G. N. & Rudolph, F. B. Cloning, sequencing, and expression of clustered genes encoding beta-hydroxybutyryl-coenzyme A (CoA) dehydrogenase, crotonase, and butyryl-CoA dehydrogenase from Clostridium acetobutylicum ATCC 824. Journal of Bacteriology 178, 3015–3024 (1996).
Colby, G. D. & Chen, J. S. Purification and properties of 3-hydroxybutyryl-coenzyme A dehydrogenase from Clostridium beijerinckii (“Clostridium butylicum”) NRRL B593. Applied and Environmental Microbiology 58, 3297–3302 (1992).
Li, F. et al. Coupled Ferredoxin and Crotonyl Coenzyme A (CoA) Reduction with NADH Catalyzed by the Butyryl-CoA Dehydrogenase/Etf Complex from Clostridium kluyveri. Journal of Bacteriology 190, 843–850, https://doi.org/10.1128/JB.01417-07 (2008).
Nie, X. et al. PTS regulation domain-containing transcriptional activator CelR and sigma factor σ(54) control cellobiose utilization in Clostridium acetobutylicum. Molecular Microbiology 100, 289–302, https://doi.org/10.1111/mmi.13316 (2016).
Merrick, M. J. In a class of its own - the RNA polymerase sigma factor σ54 (σN). Mol Microbiol 10, 903–909, https://doi.org/10.1111/j.1365-2958.1993.tb00961.x (1993).
Paget, M. S. B. & Helmann, J. D. The σ70 family of sigma factors. Genome biology 4, 203 (2003).
Jovanovic, M. et al. Regulation of the co-evolved HrpR and HrpS AAA+ proteins required for Pseudomonas syringae pathogenicity. Nature communications 2, 177 EP – (2011).
Bernard, C. S., Brunet, Y. R., Gavioli, M., Lloubès, R. & Cascales, E. Regulation of Type VI Secretion Gene Clusters by σ54 and Cognate Enhancer Binding Proteins. Journal of Bacteriology 193, 2158–2167, https://doi.org/10.1128/JB.00029-11 (2011).
Fisher, M. A. et al. Borrelia burgdorferi σ54 is required for mammalian infection and vector transmission but not for tick colonization. Proceedings of the National Academy of Sciences of the United States of America 102, 5162–5167, https://doi.org/10.1073/pnas.0408536102 (2005).
Visick, K. L. An intricate network of regulators controls biofilm formation and colonization by Vibrio fischeri. Molecular Microbiology 74, 782–789, https://doi.org/10.1111/j.1365-2958.2009.06899.x (2009).
Wolfe, A. J., Millikan, D. S., Campbell, J. M. & Visick, K. L. Vibrio fischeri σ54 Controls Motility, Biofilm Formation, Luminescence, and Colonization. Applied and Environmental Microbiology 70, 2520–2524, https://doi.org/10.1128/AEM.70.4.2520-2524.2004 (2004).
Saldías, M. S., Lamothe, J., Wu, R. & Valvano, M. A. Burkholderia cenocepacia requires the RpoN Sigma Factor for Biofilm Formation and Intracellular Trafficking within Macrophages. Infection and immunity 76, 1059–1067, https://doi.org/10.1128/IAI.01167-07 (2008).
Martin-Verstraete, I., Débarbouillé, M., Klier, A. & Rapoport, G. Levanase operon of Bacillus subtilis includes a fructose-specific phosphotransferase system regulating the expression of the operon. Journal of Molecular Biology 214, 657–671, https://doi.org/10.1016/0022-2836(90)90284-S (1990).
Martin, I., Debarbouille, M., Ferrari, E., Klier, A. & Rapoport, G. Characterization of the levanase gene of Bacillus subtilis which shows homology to yeast invertase. Molecular & general genetics: MGG 208, 177–184 (1987).
Martin-Verstraete, I. et al. Antagonistic effects of dual PTS-catalysed phosphorylation on the Bacillus subtilis transcriptional activator LevR. Mol Microbiol 28, 293–303, https://doi.org/10.1046/j.1365-2958.1998.00781.x (1998).
Browning, D. F. & Busby, S. J. W. The regulation of bacterial transcription initiation. Nature Reviews Microbiology 2, 57 EP – (2004).
Browning, D. F. & Busby, S. J. W. Local and global regulation of transcription initiation in bacteria. Nature Reviews Microbiology 14, 638 EP – (2016).
Al-Hinai, M. A., Jones, S. W. & Papoutsakis, E. T. σK of Clostridium acetobutylicum Is the First Known Sporulation-Specific Sigma Factor with Two Developmentally Separated Roles, One Early and One Late in Sporulation. Journal of Bacteriology 196, 287–299, https://doi.org/10.1128/JB.01103-13 (2014).
Jones, S. W., Tracy, B. P., Gaida, S. M. & Papoutsakis, E. T. Inactivation of σF in Clostridium acetobutylicum ATCC 824 blocks sporulation prior to asymmetric division, abolishes σE- and σG-protein expression, but does not block solvent formation. Journal of Bacteriology, https://doi.org/10.1128/JB.00088-11 (2011).
Tracy, B. P., Jones, S. W. & Papoutsakis, E. T. Inactivation of σE and σG in Clostridium acetobutylicum illuminates their roles in clostridial-cell-form biogenesis, granulose synthesis, solventogenesis, and spore morphogenesis. Journal of Bacteriology 193, 1414–1426 (2011).
Kearse, M. et al. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28, 1647–1649 (2012).
Vallenet, D. et al. MicroScope—an integrated microbial resource for the curation and comparative analysis of genomic and metabolic data. Nucleic Acids Research 41, D636–D647 (2012).
Gapes, J. R., Nimcevic, D. & Friedl, A. Long-Term Continuous Cultivation of Clostridium beijerinckii in a Two-Stage Chemostat with On-Line Solvent Removal. Applied and Environmental Microbiology 62, 3210–3219 (1996).
Gibson, D. G. et al. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nature Methods 6, 343–345 (2009).
Mermelstein, L. D. & Papoutsakis, E. T. In vivo methylation in Escherichia coli by the Bacillus subtilis phage phi 3T I methyltransferase to protect plasmids from restriction upon transformation of Clostridium acetobutylicum ATCC 824. Applied and Environmental Microbiology 59, 1077–1081 (1993).
Kolek, J., Sedlar, K., Provaznik, I. & Patakova, P. Dam and Dcm methylations prevent gene transfer into Clostridium pasteurianum NRRL B-598: development of methods for electrotransformation, conjugation, and sonoporation. Biotechnology for Biofuels 9, 14 (2016).
Vasconcelos, I., Girbal, L. & Soucaille, P. Regulation of carbon and electron flow in Clostridium acetobutylicum grown in chemostat culture at neutral pH on mixtures of glucose and glycerol. J Bacteriol 176 (1994).
Acknowledgements
We are grateful to Michelle Kuntz for constructing the pFW-FC06 plasmid and to Gwladys Chartier for technical assistance.
Author information
Authors and Affiliations
Contributions
R.H., N.L.F. and F.W. designed the study. R.H. performed experiments on C. beijerinckii DSM 6423 and AA strains. M.B.-L. performed experiments regarding the C. beijerinckii NCIMB 8052 strain. R.H., M.B.-L., N.L.F. and F.W. wrote the paper.
Corresponding author
Ethics declarations
Competing Interests
The authors declare no competing interests.
Additional information
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Hocq, R., Bouilloux-Lafont, M., Lopes Ferreira, N. et al. σ54 (σL) plays a central role in carbon metabolism in the industrially relevant Clostridium beijerinckii. Sci Rep 9, 7228 (2019). https://doi.org/10.1038/s41598-019-43822-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-019-43822-2
- Springer Nature Limited
This article is cited by
-
Engineering of global transcription factors in Bacillus, a genetic tool for increasing product yields: a bioprocess overview
World Journal of Microbiology and Biotechnology (2023)
-
Sporulation in solventogenic and acetogenic clostridia
Applied Microbiology and Biotechnology (2021)
-
Pathway dissection, regulation, engineering and application: lessons learned from biobutanol production by solventogenic clostridia
Biotechnology for Biofuels (2020)
-
Genomic reconstruction of σ54 regulons in Clostridiales
BMC Genomics (2019)