Engineering success
We succeeded to construct the clostridial specific ClosTron plasmid pMTL83151 with the central gene acsB of the Wood-Ljungdahl pathway (WLP). The WLP enables acetogenic bacteria to assimilate CO2. AcsB was previously amplified from the acetogenic bacteria Clostridium ljungdahlii and Acetobacterium woodii. As our initial promoter was too strong, we switched to a lactose induced promoter (PbgaL) and could construct our vector with this promoter. By changing our recipient cells to low copy cells, we were able to compensate for the toxicity of the constitutively active thiolase promoter (PthlA) which we also originally intended to use. The constructed plasmids were successfully introduced into the high yield butanol producer Clostridium saccharoperbutylacetonicum. Furthermore, a 5 kb fragment consisting of the WLP genes ascB1, cooC2 and acsA was isolated from A. woodii for subsequent plasmid construction.
Overview and Planning
During our project, we faced multiple obstacles which we had to overcome and therefore have gone through several engineering cycles. We aimed to enable a clostridial butanol producer to utilize CO2 as metabolic substrate for biofuel production. For this purpose, a carbon dioxide assimilation pathway had to be introduced into these sugar oxidizing bacteria. The ancient Wood-Ljungdahl pathway (WLP), a conserved CO2, CO and H2 assimilating pathway in acetogenic bacteria[1], [2] was chosen for cloning as it is the most efficient pathway. The WLP does not generate energy, hence the bacteria using the WLP for carbon assimilation requires an additional energy source, an Ech or RNF complex[3], [4]. Therefore, our selection of host organisms was limited to such bacteria which are already harboring such energy sources. Recent publications indicate that some solventogenic bacteria (butanol-producing bacteria) harbor fragments of the required genes, e.g Clostridium beijerinckii (C. beijerinckii)[5]. However, these results were based only on genomic data without indication for functionality or expression of these genes. As reliable data for the proteome and cloning approaches in these bacteria are scarce in literature, we decided to clone first a minimal fragment into our chosen butanol producer, Clostridium saccharoperbutylacetonicum (C. saccharoperbutylacetonicum). For choosing the minimal fragment, we selected the acetyl-CoA synthase (acsB1) of the acetogene Acetobacterium woodii (A. woodii) and the same enzyme (acsB2) of Clostridium ljungdahlii (C. ljungdahlii). This central enzyme combines the carbonyl and methyl branches of the WLP and has pyruvate as a direct product. It is therefore absolutely necessary for a functional pathway. After functional assessment, in particular for CO2 assimilation, of C. saccharoperbutylacetonicum, we aimed to increase the cloning fragment to 5 kb, consisting of the genes ascB1, cooC2 and acsA of our donor organism A. woodii. These are the genes immediately upstream of acsB1. Since CO2 assimilation was also not possible as a result, this suggests that only genetic remnants of WLP without functional enzymes remain in these bacteria and that the entire metabolic pathway of A. woodii or C. ljungdahlii would have to be cloned. This poses another hurdle for our project, as all enzymes are organized in large clusters that would need to be separated from each other. Therefore, two plasmids would be required and the functional transcription units would have to be separated.
Plasmid Construction
For our vector construction to transfer the required CO2 assimilation genes into C. saccharoperbutylacetonicum, we used the Clostron plasmid system as it is the only well-established vector system for clostridia [6], [7]. We aimed to construct a plasmid (backbone pMTL83151) with the constitutively expressed thiolase promoter (pthlA) and our required genes of the Wood-Ljungdahl-pathway as insert (downstream of promoter). After common amplification of our target from the genome of A. woodii and C. ljungdahlii via PCR and numerous attempts of cloning the minimal fragment acsB into our plasmid backbone with the thiolase promoter, we succeeded. However, we figured out by sequencing that the thiolase promoter was heavily mutated in all our attempts. Due to the high number of identical cloning outcomes without succeeding a plasmid with intact thiolase promoter and acsB insert, we decided to change the promoter in our construct. As the thiolase promoter is one of the strongest promoters utilized in literature so far , we switched to a lactose induced promoter (pBgal). Therefore, if not induced, this promoter leads to low basal expression of the following insert. And finally, we succeeded to construct a plasmid with pBgal promoter and acsB1.
Plasmid Construction
We could conclude that one of the toxic parts in our construct is the promoter as we sequenced our vector system right after we retransformed it into Escherichia coli (E. coli). For this retransformation, the vector system consists only of the backbone and the thiolase promoter (without acsB1 insert). Even in our retransformed E. colis, the thiolase promoter was already mutated and functionless. Therefore, we summarized that not the insert, but the promoter must be a toxic part in the construct.
Plasmid Construction
Nevertheless, the thiolase promoter would be ideal for our construct as it is constitutively expressed and acts as a strong promoter which is required to ensure a sufficient metabolic flux for the final organism. We therefore decided to change the competent E. coli cells in which we transform. Based on the protocol of the ClosTron plasmid construction we have used E. coli XL-1 blue MRF. This line is specialized on high efficiency cloning. We switched to E. coli Copy Cutter EPI 400, cells specialized on potential toxic genes or unstable DNA fragments by lowering the copy number. As the thiolase promoter was toxic for E. coli XL-1 blue MRF it seemed reasonable to us that cells specialized for toxic products can compensate for the promoter's toxicity. Ultimately, we succeeded to construct a plasmid pMTL83151 with the thiolase promoter and the acsB1 insert by using E. coli Copy Cutter EPI 400.
We subsequently managed to introduce all constructed plasmids in C. saccharoperbutylacetonicum and performed growth analysis. We verified our constructs by Sanger Sequencing.
Isolation of 5 kb fragment of A. woodii
The minimal fragment was not sufficient for CO2 assimilation in C. saccharoperbutylacetonicum. Therefore, we attempted to construct a ClosTron plasmid pMTL83151 with the genes ascB1, cooC2 and acsA (5 kb fragment) of A. woodii as insert to check whether these genes are adequate for CO2 assimilation. However, PCR amplification of the 5 kb fragment, taking A. woodii gDNA as template was not successful. A smaller insert of 200 bp was always additionally amplified (even though no complementary structures of our primers were known in the entire A. woodii genome). We could conclude that the smaller insert is not the primer cloud usually visible after gel electrophoresis because after purification this smaller insert was still clearly visible. We additionally assumed that the smaller insert will have the required restriction sites due to our designed primers harboring a restriction site as overhang. Hence, the smaller insert would be digested and ligated in the same way as our desired 5 kb insert. Because during transformation, the host prefers to integrate plasmids of smaller sizes (therefore smaller inserts), removing the insert was required. After designing more specific primers and performing nested PCRs we still could not eliminate the smaller insert in our sample. After all attempts we came back to gel extraction. Before extraction of the 5 kb fragment directly from gel could not reach a sufficient amount of the PCR product for subsequent digestion, ligation and transformation. We adapted our PCR protocol to reach a sufficient amount of PCR product from the beginning. Additionally, we modified the protocol of gel extraction to finally extend the amount of PCR product. We ultimately had a sufficient quantity of the 5 kb PCR product to perform the cloning procedures.
We were challenged by many obstacles during our project. By modifying previously tested methods and overthinking and reviewing ideas we succeeded in coping with numerous problems and learnt how to deal with such obstacles
Sources
[1] Ragsdale, S. W. (2008): Enzymology of the wood-Ljungdahl pathway of acetogenesis. In: Annals of the New York Academy of Sciences 1125, P. 129-136.
[2] Ragsdale, S. W.; Pierce, E. (2008): Acetogenesis and the Wood-Ljungdahl pathway of CO(2) fixation. In: Biochimica et biophysica acta 1784 (12), P. 1873-1898.
[3] Müller, V. (2003): Energy conservation in acetogenic bacteria. Appl Environ Microbiol. 69: 6345–6353
[4] Schuchmann, K.; Müller, V. (2014): Autotrophy at the thermodynamic limit of life: a model for energy conservation inacetogenic bacteria. Nat Rev Microbiol. 12: 809–821.
[5] Sandoval-Espinola, W. J.; Chinn, M. S.; Thon, M. R.; Bruno-Bárcena, J. M. (2017): Evidence of mixotrophic carbon-capture by n-butanol-producer Clostridium beijerinckii. In: Scientific reports 7 (1), P. 12759.
[6] Heap, J. T.; Pennington, O. J.; Cartman, S. T.; Carter, Glen P.; Minton, N. P. (2007): The ClosTron: a universal gene knock-out system for the genus Clostridium. In: Journal of microbial methods 70(3), P. 452-464.
[7] Heap, J. T.; Kuehne, S. A.; Ehsaan, M.; Cartman, S. T.; Cooksley, C. M.; Scott, J. C.; Minton, N. P. (2010): The ClosTron: Mutagenesis in Clostridium refined and streamlined. In: Journal of microbiological methods 80 (1), P. 49-55.