Results
35
Figure 1: Genes of the Wood-Ljungdahl pathway. Clustered in A: A. woodii and B: C. ljungdahlii (for details see supplementary Table 1).
Target gene analysis

For target gene analysis, we analysed the genome of our aimed recipient strain to identify which genes are required to generate a functional Wood-Ljungdahl pathway (WLP) in a non-CO2-assimilating host organism. Sandoval-Espinola et al. were first to identify WLP genes in Clostridia which are incapable of utilising CO2 as a carbon source[1]. This proposes that only the acetyl-CoA synthase (acsB) is required to enable a functional WLP. However, these investigations were based on genome annotations and RNA sequencing data. No enzymatic studies on translation, functionality, or interaction were made to further investigate these transcripts. Additionally, these studies were only conducted in the butanol producer Clostridium beijerinckii (C. beijerinckii), but not in our target organism Clostridium saccharoperbutylacetonicum (C. saccharoperbutylacetonicum). In order to identify required genes for a functional WLP in our target organism, we first thought about the essential genes of the WLP. Woods et al. did an in-depth essay to identify the minimal set of WLP genes[2]. They were able to identify a set of 21 genes required for the assimilation of CO2 and H2, the methyl- and carbonyl branch and the eventual synthesis of acetyl-CoA. By in-silico analysis we were able to identify that only the acetyl-CoA synthase (ACS), the formate dehydrogenase and the corrinoid-iron nickel enzyme annotations in the genome of C. saccharoperbutylacetonicum were missing. The absence of the formate dehydrogenase is of secondary importance for this project as C. saccharoperbutylacetonicum, same as C. beijerickii possess the Pfor/Pfl system. It catalyzes first the transfer of CO2 onto acetyl-CoA (Pfor) and then, by removal of the CoA, formate is generated (Pfl). Therefore, C. saccharoperbutylacetonicum can synthesize formate, which is the first step of the WLP, without the utilisation of the formate dehydrogenase. Furthermore, all essential genes for the Rnf energy conservation system are annotated in C. saccharoperbutylacetonicum, implying that the same energy system as in Clostridium ljungdhalii (C. ljungdahlii) and Acetobacterium woodii (A. woodii) is present (Figure 1). However, due to the limited time frame of the project, we were not able to conduct enzyme essays of RNA-seq. experiments in C. saccharoperbutylacetonicum.

Results
Plasmid Construction

During the course of our project, we constructed the following three plasmids (Figure 2):


35
Figure 2: Plasmid constructions. The plasmids that we have successfully constructed: A: pMTL83151_PbgaL_acsB2_CLJU, consisting of the gene acsB2 from C. ljungdahlii as insert and the lactose inducible promoter (PbgaL). B: pMTL83151_PbgaL_acsB1_AWO, consisting of the gene acsB1 from A. woodii as insert and the lactose inducible promoter (PbgaL). C: pMTL83151_PthlA_acsB2_CLJU, consisting of the gene acsB2 from C. ljungdahlii as insert and the thiolase promoter. Inserts are marked green, promoters are orange. The genes marked in blue are part of the backbone.

The backbones used belong to the ClosTron plasmid system and were: pMTL83151_PbgaL_gusA (7506 bp) and pMTL83151_PthlA_gusA (6241 bp)
pMTL83151_PbgaL_gusA contains the promoter PbgaL which leads to low basal expression of the following coded gene on the plasmid and additionally is lactose inducible for stronger gene expression. pMTL83151_PthlA_gusA instead contains the PthlA promoter which is known to be a strong promoter in Clostridia. The empty plasmid pMTL83151 contains the lacZ-gene at the multiple cloning site, which means that by introduction of a transgene insert, the lacZ-gene becomes disrupted. Colonies of positive clones with correct transgene insertion should therefore appear white on a medium containing X-gal. Negative clones with faulty transgene integration in the plasmid can still express the β-galactosidase and therefore X-gal is hydrolyzed and the blue 5,5’-dibromo-4,4’-dichloro-indigo dye is produced. Additionally, the plasmids code for the catP cassette, so Escherichia coli (E. coli) or C. saccharoperbutylacetonicum transformants can grow on selective media containing the antibiotics Chloramphenicol (E. coli) or Thiamphenicol (C. saccharoperbutylacetonicum).

Our desired transgene we wanted to transform into E. coli and subsequently into C. saccharoperbutylacetonicum was the acetyl-CoA synthase acsB. We obtained the gene by amplification via PCR from the genomic DNA of our donor organisms. The donor organisms we used were C. ljungdahlii and A. woodii. The correct amplification of the 2.1 kb large gene can be seen by the band on the gel in Figure 3.


35
Figure 3: Agarose gel electrophoresis verified correct PCR amplification ofacsB1from A. woodii and acsB2 from C. ljungdahlii genome. Columns 1 to 3 show a band that matches the size of 2175 bp of acsB1. Columns 5 and 6 show a band that matches the size of 2127 bp of acsB2. Both results indicate that the acsB1 and acsB2 gene were successfully amplified. However, amplification was not successful in column 4, where no distinct band has emerged. A possible explanation for that could be that the gDNA used for this specific amplification sample was sheared. Nevertheless, the successful amplified ACS was sufficient to continue with the next step.

Additionally, we amplified a 5 kb large fragment from A. woodii using PCR, in case our minimal ACS fragment is not sufficient for CO2 assimilation in C. saccharoperbutylacetonicum. This 5 kb fragment contains the genes encoding for AcsA, CooC2 and AcsB1. Here, acsA and cooC2 code for the CO dehydrogenase (CODH) whereas acsB1 codes for the acetyl-CoA synthase, the ACS subunit of the CODH/ACS complex. The CODH/ACS complex catalyzes the fusion of the methyl- and carbonyl branches by transferring the methyl- and carbonyl groups to Co-enzyme A to form acetyl-CoA. Initial amplification attempts of the 5 kb fragment failed. We therefore tried different PCR master mixes (as indicated in the notebook). To check whether the amplification with one of the new master mixes was successful, the size of the PCR product was examined via agarose gel electrophoresis. The correct amplification of the 5 kb large gene can be seen by the band on the gel in Figure 4 for all samples. However, an additional band smaller than 250 bp is visible. To isolate only the 5 kb fragment, direct purification from gel was carried out.


35
Figure 4: Agarose gel electrophoresis verified successful amplification of acsA, cooC2 and acsB1 from the genome of A. woodii. Examination of the agarose gel showed that all samples (1-6) show a band for a 5kb PCR product which meets the expectations and indicates correct amplification of the desired fragment. However, an additional band for a fragment smaller than 250 bp can also be seen.

Next, we digested our backbone plasmids and insert-genes acsB1 (from A. woodii) and acsB2 (from C. ljungdahlii) with Xho1 and Nhe1. The gusA gene (1.8 kb) was cut out by the enzymes and the acsB (2.1 kb) was newly ligated into the backbone plasmid. The ligated plasmids were then transformed into E. coli. Plasmids from white colonies on transformation plates containing X-gal, IPTG and Chloramphenicol were prepared and controlled by test-digestion with Xho1 and Nhe1. With the digested samples a gel electrophoresis was conducted. In Figures 5-7, bands for the digested linearized plasmids without insert can be seen. At 5.7 kb for pMTL83151_PbgaL_acsB2_CLJU and pMTL83151_PbgaL_acsB1_AWO, as well as at 4.4 kb for pMTL83151_PthlA_acsB2_XY can be seen. Furthermore, at 2.1 kb the excised acsB insert is present, demonstrating correct cloning could be verified.


35
Figure 5: Agarose gel electrophoresis verified correct cloning of pMTL83151_PbgaL_acsB1_AWO. After restriction digest (Xho1 and Nhe1) of pMTL83151_PbgaL_gusA to remove gusA and ligation with acsB1 at the previous location of gusA, the resulting plasmid was called pMTL83151_PbgaL_acsB1_AWO. This plasmid was amplified through transformation in E. coli cells, followed by cultivation of the transformed E. coli cells on selective media and plasmid preparation. The obtained plasmids were test-digested with restriction enzymes Xho1 and Nhe1 cutting at the insert borders. The agarose gel shows the undigested and digested plasmids of different positive (white) E. coli clones. Furthermore, acsB1 from Fig. 3, as well as plasmid pMTL83151_PbgaL_gusA after digestion with Xho1 and Nhe1 were used as controls. Fig. 5 shows that all undigested plasmid samples exhibited a band that matches the expected size of 7860 bp of pMTL83151_PbgaL_acsB1_AWO as well as two other bands corresponding to the different forms of plasmid DNA, e.g. coiled-coil which passes through the gel faster than the circular form. The corresponding digested samples led to two bands located at sizes that match the 2175 bp of acsB1 insert and the 5685 bp plasmid backbone without acsB1.

35
Figure 6: Agarose gel electrophoresis verified correct cloning of pMTL83151_PbgaL_acsB2_CLJU. After restriction digest (Xho1 and Nhe1) of pMTL83151_PbgaL_gusA to remove gusA and ligation with acsB2 at the previous location of gusA, the resulting plasmid was called pMTL83151_PbgaL_acsB2_CLJU. This plasmid was amplified through transformation in E. coli cells, followed by cultivation of the transformed E. coli cells on selective media and plasmid preparation. The obtained plasmids were test-digested with restriction enzymes Xho1 and Nhe1 cutting at the insert borders. The agarose gel shows the undigested and digested plasmids of different positive (white) E. coli clones. Furthermore, acsB2 from Fig. 3, as well as plasmid pMTL83151_PbgaL_gusA after digestion with Xho1 and Nhe1 were used as controls. Fig. 6 shows that all undigested plasmid samples exhibited a band that matches the expected size of 7812 bp of pMTL83151_PbgaL_acsB2_XY as well as two other bands corresponding to the different forms of plasmid DNA, e.g. coiled-coil which passes through the gel faster than the circular form. The corresponding digested samples led to two bands located at sizes that match the 2127 bp of acsB2 and the 5685 bp plasmid backbone without acsB2.

35
Figure 7: Agarose gel electrophoresis verified correct cloning of pMTL83151_PthlA_acsB2_CLJU. After restriction digest (Xho1 and Nhe1) of pMTL83151_PthlA_gusA to remove gusA and ligation with acsB2 at the previous location of gusA, the resulting plasmid was called pMTL83151_PthlA_acsB2_CLJU. This plasmid was amplified through transformation in E. coli cells, followed by cultivation of the transformed E. coli cells on selective media and plasmid preparation. The obtained plasmids were digested with restriction enzymes Xho1 and Nhe1 cutting at the insert borders. The agarose gel shows the undigested and digested plasmids of different positive (white) E. coli clones. The figure shows that all undigested plasmid samples exhibited a band that matches the expected size of 6547 bp of pMTL83151_PthlA_acsB2_CLJU as well as two other bands corresponding to the different forms of plasmid DNA, e.g. coiled-coil which passes through the gel faster than the circular form. The corresponding digested samples led to two bands located at sizes that match the 2127 bp of acsB2 and the 4420 bp plasmid backbone without acsB2.

For final verification of the transformed plasmids and to check for any possible point mutations, plasmids pMTL83151_PbgaL_acsB1_AWO, pMTL83151_PbgaL_acsB2_CLJU and pMTL83151_PthlA_acsB2_CLJU were completely sequenced. Alignment of the sequencing results obtained with the expected plasmid sequences revealed that no mutations were present.

After we successfully constructed our plasmids, we transformed them into our recipient organism C. saccharoperbutylacetonicum . We selected positive transformants on media containing Thiamphenicol. To check if the correct plasmids were integrated, we prepared the plasmids out of clones and again performed a test-digest with Xho1 and Nhe1. After gel electrophoresis, bands for the digested linearized plasmids without an insert at 5.7 kb for pMTL83151_PbgaL_acsB1_Awo, pMTL83151_PbgaL_acsB2_CLJU, as well as at 4.4 kb for pMTL83151_PthlA_acsB2_CLJU can be seen (see Figures 8-9) which match with the bands of the digested plasmids out of the E. coli transformants. Furthermore, at 2.1 kb the excised acsB insert is present, demonstrating correct transformation.


35
Figure 8: Agarose gel electrophoresis verified the correct transformation of C. saccharoperbutylacetonicum with pMTL83151_PbgaL_acsB1_AWO and pMTL83151_PbgaL_acsB2_CLJU. After verification of correct plasmid amplification in E. coli, C. saccharoperbutylacetonicum was transformed with pMTL83151_PbgaL_acsB1_AWO and pMTL83151_PbgaL_acsB2_CLJU and cultivated on selective media for following plasmid preparation. The obtained plasmids were digested with restriction enzymes Xho1 and Nhe1 cutting at the insert borders. The agarose gel shows the undigested and digested plasmids of different positive C. saccharoperbutylacetonicum clones. Fig. 8 shows that the undigested plasmid samples exhibited bands that match the size of pMTL83151_PbgaL_acsB1_AWO and pMTL83151_PbgaL_acsB2_CLJU in the coiled-coil form respectively (compare Fig. 3-4). The corresponding digested samples showed two bands located at sizes that match the 2175 bp/2127 bp of acsB1/acsB2 and the 5685 bp/5685 bp plasmid backbone without acsB1/acsB2. These observations meet the expectations and therefore indicate that C. saccharoperbutylacetonicum match the bands of digested pMTL83151_PbgaL_acsB2_CLJU from E. coli. These observations meet the expectations and therefore indicate that C. saccharoperbutylacetonicum was successfully transformed and that the plasmids were not modified. However, the bands of the digested samples were rather faint, especially the ones that match the size of acsB1/acsB2. This may be due to a low overall concentration of the plasmid samples

35
Figure 9: Agarose gel electrophoresis verified the correct transformation of C. saccharoperbutylacetonicum with pMTL83151_PthlA_acsB2_CLJU. After verification of correct plasmid amplification in E. coli, C. saccharoperbutylacetonicum was transformed with pMTL83151_PthlA_acsB2_CLJU and cultivated on selective media for following plasmid preparation. The obtained plasmids were digested with restriction enzymes Xho1 and Nhe1 cutting at the insert borders. The agarose gel shows the undigested and digested plasmids of different positive C. saccharoperbutylacetonicum clones. Bands were only observed for the plasmid DNA of Clone 6 and undigested plasmid DNA of Clone 7 which may be due to the low plasmid DNA concentration of the samples. Two faint bands that match the size of acsB2 (2127 bp) and the plasmid backbone without acsB2 (4420 bp) were observed for the digested plasmid DNA prepared from Clone 6 which meets the expectations. The corresponding undigested sample as well as C.7 - u was located at a size below the expected value of 6547 bp which may be due to the fact that the band corresponds to the coiled-coil DNA form.

At the very end, we sent the plasmids to another round of sequencing to make sure that also our C. saccharoperbutylacetonicum transformants do not possess plasmids with any undesired mutations. Figures 10 and 11 show the alignment of the sequenced fragments with the plasmid reference and verified that our inserts were correctly cloned into the vector system. This supports the assumption that we successfully transformed C. saccharoperbutylacetonicum with the acsB1 gene from A. woodii or acsB2 gene from C. ljungdahlii under control of the two different promoters PbgaL and PthlA on the pMTL83151 ClosTron plasmids.

35
Figure 10: Sequencing of acsB2_CLJU insert and PbgaL promotor region of pMTL83151_PbgaL_acsB2_CLJU verified successful transformation of C. saccharoperbutylacetonicum. After plasmid preparation from C. saccharoperbutylacetonicum strains transformed with pMTL83151_PbgaL_acsB2_CLJU, the isolated plasmid DNA was sequenced. Sequencing was carried out by an external sequencing laboratory and the fragment sequences thus obtained were aligned to the expected plasmid sequence of pMTL83151_PbgaL_acsB2_CLJU. It is visible in the top line that acsB2 and the promoter region of PbgaL is covered by the sequenced fragments. The display of the individual fragment sequences indicates that the plasmid is correct and does not contain any mutations.
35
Figure 11. Sequencing of acsB2_CLJU insert and PthlA promotor region of pMTL83151_PthlA_acsB2_CLJU verified successful transformation of C. saccharoperbutylacetonicum. After plasmid preparation from C. saccharoperbutylacetonicum strains transformed with pMTL83151_PthlAL_acsB2_CLJU, the isolated plasmid DNA was sequenced. Sequencing was carried out by an external sequencing laboratory and the fragment sequences thus obtained were aligned to the expected plasmid sequence of pMTL83151_PthlA_acsB2_CLJU. It is visible in the top line that acsB2 and the promoter region of PthlA is covered by the sequenced fragments. The display of the individual fragment sequences indicates that the plasmid is correct and does not contain any mutations.
Results
Growth analysis of transformed C. saccharoperbutylacetonicum

We have successfully integrated the constructed plasmids pMTL83151_PbgaL_acsB2_CLJU, pMTL83151_PbgaL_acsB1_Awo and pMTL83151_PthlA_acsB2_CLJU into our chosen host organism C. saccharoperbutylacetonicum.

For verification whether the inserted subunit of the acetyl-CoA synthase (acsB) is sufficient for autotrophic growth of C. saccharoperbutylacetonicum, growth analysis was performed. If CO2 could be assimilated and utilised for further metabolic processes by our transformed strains, the bacteria should be viable and grow on CO2 as only C-source. To test if this was the case, we inoculated C. saccharoperbutylacetonicum strains transformed with pMTL83151_PbgaL_acsB2_CLJU, pMTL83151_PbgaL_acsB1_Awo and pMTL83151_PthlA_acsB2_CLJU and without plasmid (wildtype as negative control) on C. ljungdahlii medium without fructose. For inoculation, we picked a colony of our C. saccharoperbutylacetonicum strains from an C. ljungdahlii agar plate to avoid picking up any carbon source from the medium. Afterwards, we gassed all cultures with a mix of hydrogen (66%) and carbon dioxide (34%). No growth could be detected for all investigated strains. This indicates that inserting acsB alone is not sufficient for CO2 assimilation and autotrophic growth of C. saccharoperbutylacetonicum.

To ensure that the inserted plasmid does not impair or reduce the growth of the C. saccharoperbutylacetonicum transformants, further growth analysis was conducted. Growth experiments were carried out with all our transformants: C. saccharoperbutylacetonicum containing pMTL83151_PbgaL_acsB2_CLJU, pMTL83151_PbgaL_acsB1_Awo and pMTL83151_PthlA_acsB2_CLJU respectively. Growth experiments were performed in CGM medium supplemented with Thiamphenicol. Additionally, we induced C. saccharoperbutylacetonicum pMTL83151_PbgaL_acsB2_CLJU and pMTL83151_PbgaL_acsB1_Awo with 20 mM lactose to induce the PbgaL promoter and activate stronger gene expression of the acsB insert. The growth of the C. saccharoperbutylacetonicum wildtype strain in CGM medium without thiamphenicol was used as a reference. All cultures were inoculated to an OD600 of 0.05. As seen in Figure 12, all transgenic strains were viable on CGM medium supplemented with Thiamphenicol. C. saccharoperbutylacetonicum with pMTL83151_PbgaL_acsB1_Awo (not induced) showed the most rapid growth. It switched from lag- to the exponential stage after two hours. All other investigated strains reached the exponential phase after six hours, whereas C. saccharoperbutylacetonicum with pMTL83151_PthlA_acsB2_CLJU reached this phase after 8 hours only. The slope during the exponential growth phase was similar for the entire investigated strains and no significant difference was detected. All strains reached the stationary stage after 27 hours where no notable increase of the OD600 could be noticed anymore. The transformed strains (induced and not induced for the lactose inducible promoter) grew slightly better than the wildtype. This indicates that the inserted plasmid does not reduce or limit the growth of the transformed cultures. Hence, we assumed that the undetectable growth on CO2 as the sole carbon source was not cause by the integration of the plasmid. Instead, we concluded that the inserted acsB gene may not be sufficient for CO2 assimilation in C. saccharoperbutylacetonicum.


35
Figure 12: Growth analysis of C. saccharoperbutylacetonicum and transformants. Displayed is the growth behaviour of C. saccharoperbutylacetonicum wildtype, with integrated pMTL83151_PbgaL_acsB2_CLJU (Cba_pBgal_acsB_XY), pMTL83151_PbgaL_acsB1_Awo (Cba_pBgal_acsB_Awo) and pMTL83151_PthlA_acsB2_CLJU (Cba_pthla_acsB_XY) over a range of 30 hours. Additionally, C. saccharoperbutylacetonicum with pMTL83151_PbgaL_acsB2_CLJU (Cba_pBgal_acsB_XY_induced) and pMTL83151_PbgaL_acsB1_Awo (Cba_pBgal_acsB_Awo_induced) were induced with 20mM lactose. All cultures were inoculated in CGM medium supplemented with Thiamphenicol to an OD600 of 0.05. All cultures are showing a similar growth but individual differences occur. Cba_pBgal_acsB_Awo shows the fastest and Cba_pthla_acsB_XY the slowest growth.
Results
Transcription unit design

As the results above indicate that the expression of the acetyl-CoA synthase is not sufficient to enable an autotrophic growth of C. saccharoperbutylacetonicum, therefore more pathway genes are required to convey this capability. Since the PCR for the 5 kb unit of A. woodii (as described above) proved challenging, most likely due to the low GC content of Clostridia, we chose a commercial de-novo synthesis approach for the construction of the larger transcription units. As we are currently unable to identify which genes are required to allow a functional Wood-Ljungdahl pathway (WLP), we targeted the WLP cluster of A. woodii for transforming further genes. This is a particularly well-suited target as, in this organism, the WLP is organised into two clusters which are subdivided into smaller transcription units respectively. To ensure a successful synthesis we selected genes of higher priority for the pathway and designed two transcription units respectively so that they can be expressed together from one plasmid system. For the first plasmid construct we have chosen the pMTL83151 backbone and designed two subunits for expression. The first subunit contains the acsD-acsC-acsE fragment and the second unit contains acsA-cooC2-acsB1. To combine these units on the plasmid, they were connected via the SfaA1 restriction site. By this, we created one frame in the order: acsD-acsC-acsE-acsA-cooC2-acsB1. For the insertion of this unit into the plasmid, we utilised the Sal1-Nhe1 combination. By this, all subunits of the ACS and CODH should be expressed. For the second unit, we choose the pMTL82151 backbone. This backbone is organised identically to the pMTL83151 backbone (Figure 13). However, it contains a pBP1 replication origin instead of a pCB102 replication origin. On this backbone, we inserted the unit of cooC1-acsV-Awo_c10690-Awo_c10700 and metV-metF by connecting them with a Pac1 restriction site. This fragment adds important units of the methylene-THF reductase and accessory proteins. To this point we did not successfully transform these plasmids into C. saccharoperbutylacetonicum.


35
Figure 13: Transcription unit design. The two transcriptional subunits from A. woodii are combined on the plasmid backbone (pMTL83151 or pMTL82151) by either the restriction site SfaA1 or Pac1.
Outlook
Cloning aims

Based on the results described above, we concluded that acsB of A. woodii and C. ljungdahlii respectively is not sufficient to enable autotrophic growth of C. saccharoperbutylacetonicum. Hence, further Wood-Ljungdahl pathway (WLP) proteins are required for a functional WLP in C. saccharoperbutylacetonicum. In order to expand the gene set for cloning, we amplified 5 kb out of the genome of A. woodii. This fragment contains the following genes: acsA, cooC2 and acsB1. We aim to insert the 5 kb fragment into the plasmids pMTL83151_PbgaL_gusA and pMTL83151_PthlA_gusA for future transformation of C. saccharoperbutylacetonicum. Subsequently, the ability of the new C. saccharoperbutylacetonicum transformant to assimilate CO2 and to grow autotrophically should be evaluated.

In case that the organism is still not able to grow autotrophically, further required genes must be integrated to complement the WLP in C. saccharoperbutylacetonicum. For cloning of further required WLP genes, we were able to obtain A. woodii transcription units by de-novo synthesis as described in the transcription unit design section. However, due to the limited time frame and an accumulation of technical problems related to the synthesis, we were not able to express these genes in C. saccharoperbutylacetonicum yet. Successful integration of these A. woodii genes would introduce entirely new genes into C. saccharoperbutylacetonicum, and genes that are already annotated but it is unclear whether they are functionally expressed. We would apply the same cloning strategy as described in the results section for acsB. With the combination of our integrated gene clusters and the annotated genes in C. sacchroperbutylacetonicum, all the required genes of the WLP are present and it is likely that our transformed organism will be capable of CO2 assimilation.

In order to increase the eventual biofuel yield of our engineered Clostridium, we aim to clone an even stronger promoter in front of the pathway genes. We decided to utilise the improved ferredoxin promoter as proposed by iGEM Nottingham 2019 (part BBa_K2715011) which is the strongest promoter they analysed. However, we did not succeed to construct a plasmid with a ferredoxin promoter yet as we experienced toxic effects in E. coli. This improved ferredoxin promoter could therefore be directly integrated via the RiboCas system in C. saccharoperbutylacetonicum and the transformation of the potentially toxic plasmid into E. coli could be bypassed.

As the entirety of the Wood-Ljungdahl genes are ~18 kb in size, problems can occur during plasmid construction and subsequent transformation of such large sized plasmids. Even though ClosTron plasmids with a size of ~12 kb were already constructed and successfully transformed (see Human Practices), we could come across obstacles even when splitting the gene clusters onto several plasmids. Therefore, other more suited methods for cloning larger sized fragments such as gene clusters should be used. During our integrated human practice activities, Prof. Winzer explained alternative transformation systems. The RiboCas system is a CRISPR-based tool for genome editing specifically for the genus Clostridia with a transformation efficiency rate of 100 %[3]. Additionally, the allelic exchange system is a further possible approach for gene editing and introducing metabolic pathways into Clostridia[4].

Gene expression and enzymatic studies

To gain insights into whether the annotated genes are actually transcribed or whether they need to be cloned, and whether transformed genes are expressed in C. saccharoperbutylacetonicum, we aim to perform a Reverse Transcription quantitative PCR (RT-qPCR) screen or RNA-sequencing.

A crucial step of the WLP is the dimer formation of the CODH and the ACS in order to obtain acetyl-CoA. In previous studies it was indicated that this interaction seems to vary depending on the respective organisms[5]. We would therefore employ enzymatic studies to identify whether our cloned ACS can form a functional dimer with the CODH of C. saccharoperbutylacetonicum. Furthermore, a comparison of the amounts of acetyl-CoA synthesised by the native CODH-ACS complex to the cloned equivalent can be carried out.

Improvement of butanol yield

In an industrial strain, it must be taken into account that various aspects can reduce or limit the product yield.

For industrial scale-up the desired strain must fulfil certain requirements: it must have sufficient resistance to the end product, in our case butanol, as it normally is toxic in high concentrations. There are already first approaches to increase the butanol resistance and yield of C. saccharoperbutylacetonicum and thus make it more suitable and profitable for industrial use. Tanaka et al. performed a mutant screen and found a strain with a 1.6-fold increased butanol production[6]. The highest butanol yield so far was reached by using the genome editing tool CRISPR-Cas9. A yield of 19.0 g/liter was achieved by the deletion of the genes pta and buk in C. saccharoperbutylacetonicum[7]. Koepke et al. additionally indicated an adapted metabolism of C. autoethanogenum depending on the use of syngas or CO [8]. It would therefore be useful to assess the product yield under different gas compositions including CO, CO2, H2 gas mixtures together with varying temperatures.

Once a prototype strain is generated, the scale-up process would accelerate significantly due to the great progress made by LanzaTech and numerous other companies working with C. saccharoperbutylacetonicum on an industrial scale. So, we would be able to utilize the already existing infrastructure on industrial anaerobic fermentation.

Results
Sources

[1] Sandoval-Espinola, W. J.; Chinn, M. S.; Thon, M. R.; Bruno-Barcena, J. M. (2017): Evidence of mixotrophic carbon-capture by n-butanol-producer Clostridium beijerinckii. Sci Rep 7, 12759.

[2] Woods, C.; Humphreys, C. M.; Tomi-Andrino, C.; Henstra, A. M.; Köpke, M.; Simpson, S. D.; Winzer, K.; Minton, N. P. (2022); Required Gene Set for Autotrophic Growth of Clostridium autoethanogenum. App Environ Microbiol. 12;88(7):e0247921. DOI: 10.1128/aem.02479-21.

[3] Canadas, I. C., Groothuis, D., Zygouropoulou, M., Rodrigues, R., Minton, N. P. (2019): RiboCas: A Universal CRISPR-Based Editing Tool for Clostridium. ACS Synthetic Biology 8 (6), 1379-1390. DOI: 10.1021/acssynbio.9b00075.

[4] Ehsaan, M.; Kuit, W.; Zhang, Y.; Cartman, S. T.; Heap, J. T.; Winzer, K.; Minton, N. P. (2016): Mutant generation by allelic exchange and genome resequencing of the biobutanol organism Clostridium acetobutylicum ATCC 824. Biotechnol Biofuels 9, 4. DOI: 10.1186/s13068-015-0410-0.

[5] Cohen, S. E; Can, M.; Wittenborn, E. C.; Hendrickson, R. A.; Ragsdale, S. W.; Drennan C. L. (2020): Crystallographic Characterization of the Carbonylated A-Cluster in Carbon Monoxide Dehydrogenase/Acetyl-CoA Synthase. In: ACS Catal., vol. 10, no. 17, 9741–9746. DOI: 10.1021/acscatal.0c03033.

[6] Tanaka, Y.; Kasahara, K.; Hirose, Y.; Morimoto, Y.; Izawa, M.; Ochi, K. (2017): Enhancement of butanol production by sequential introduction of mutations conferring butanol tolerance and streptomycin resistance. In: Journal of Bioscience and Bioengineering, vol. 124, no. 4, 400-407. DOI: 10.1016/j.jbiosc.2017.05.003.

[7] Wang, S.; Dong, S.; Wang, P.; Tao, Y.; Wang, Y. (2017): Genome Editing in Clostridium saccharoperbutylacetonicum N1-4 with the CRISPR-Cas9 System Appl Environ Microbiol, vol. 83, no. 10, pp. e00233-17. DOI: 10.1128/AEM.00233-17.

[8] Koepke, M.; Mihalcea, C.; Liew, F.; Tizard, J. H.; Ali, M. S.; Conolly, J. J.; Al-Sinawi, B.;, Simpson, S. D. (2011): 2,3-Butanediol Production by Acetogenic Bacteria, an Alternative Route to Chemical Synthesis, Using Industrial Waste Gas’, Applied and Environmental Microbiology, vol. 77, no. 15, 5467-5475. DOI: 10.1128/AEM.00355-11.

Results
Supplementary

Table 1: Description of A. woodii and C. ljungdahlii genes required for the WLP.