IISc Bengaluru

Section A: Constructing plasmids for enzyme expression

DESIGN	Expression plasmids containing codon-optimised versions of CamC, CamA, and CamB were available in the AddGene repository and were used as templates for amplification. We codon-optimised the TodC1, TodC2, TodA, and TodB genes of the Tod operon from the P. putida F1 genome to reduce GC content and sent them for synthesis. We constructed pCAM and pTOD plasmids in-silico with pSEVA438 and pSEVA2213 as the initial backbones for the reassembled Cam and the Tod operons. PCR primers were designed to introduce homology regions for Gibson Assembly and to introduce the strong P. putida RBS.15 sequence upstream of each gene.
BUILD	We amplified CamC, CamA, and CamB from plasmids gifted by Dr. Teruyuki Nagamune, while TodC1, TodC2-TodA (as one unit), and TodB were amplified from fragments synthesised by IDT. P. putida compatible vectors were gifted by the SEVA repository in Madrid. pSEVA2213 and pSEVA438 were purified and linearised first by digesting with XbaI, and then by PCR due to concentration issues post gel-purification.
TEST	We attempted to build pCAM and pTOD using Gibson Assembly. This failed consistently, which was initially attributed to the low quantities of DNA used and inefficient competent cells. We switched to E. coli TOP10 for cloning, but the streptomycin resistance of this strain necessitated that we use pSEVA2213 as the backbone for both pCAM and pTOD and three-fragment assembly continued to fail despite this change.
LEARN	We realised that perhaps assembling three fragments simultaneously was a bit too ambitious for amateur experimenters. Gel electrophoresis showed that the fragments assembled incompletely and failed to enter the vector, suggesting insufficient homology.

DESIGN	We decided to fuse the fragments together using Fusion PCR to reduce the three-fragment assembly to a single-fragment assembly, and switched to the In-Fusion master mix on advice from our Principal Investigator as it could tolerate weaker homology regions.
BUILD	The CamC, CamA, and CamB fragments and the TodC1, TodC2-TodA, and TodB fragments amplified previously were fused together using Fusion PCR to produce a single CamCAB fragment and a single TodC1C2AB fragment with homology regions with the vector. This reaction was finicky and took multiple attempts to optimise, but we eventually obtained a single fragment in sufficient concentration to attempt assembly.
TEST	We attempted to build pCAM and pTOD using single-fragment In-Fusion Assembly. This failed, and for good measure we attempted a high-volume Gibson Assembly reaction which also failed. Trial assemblies with the master mixes with fragments that had worked previously also failed, suggesting an issue with the master mix or our experimenters.
LEARN	Despite multiple attempts at troubleshooting, we were unable to obtain clones. Attempts at amplifying the fused CamCAB and TodC1C2AB fragments consistently failed or produced smears, suggesting severe issues with the homology regions at the ends.

DESIGN	We redesigned the primers for Gibson Assembly and extended the homology region significantly. We reached out to the University of Stuttgart and acquired a plasmid with the TodC1C2AB genes pre-assembled, and also acquired BSL2 clearance and sourced P. putida MTCC 1273, which carries the CAM plasmid containing CamCAB. Primers were designed to amplify these operons out and introduce homology regions for Gibson Assembly,
BUILD	Unfortunately, we did not have sufficient time left to wait for the necessary materials to arrive, so we could not progress to construction.
TEST	Had we obtained successful clones, we would have run an SDS-PAGE from the crude cell lysate of overnight cultures to verify whether all components of each enzyme system were adequately expressed. We would have then assayed the functionality of the Cam and Tod systems separately using colorimetric assays on the crude cell lysate. The CamCAB system converts indole to indigo, while the TodC1C2AB system converts coumarin to 7-hydroxy coumarin; both products produce characteristic colours.
LEARN	If the enzymes were functional, we would have transformed P. putida and progressed to halocarbon degradation assays. We did not anticipate issues with functionality for genes amplified from the genome, and would have used that had the codon-optimised fragments suffered issues. Expression issues may have necessitated introducing a promoter and a terminator for each gene in the operon.

Tan CY, Hirakawa H, Suzuki R, Haga T, Iwata F, Nagamune T. Immobilization of a Bacterial Cytochrome P450 Monooxygenase System on a Solid Support. Angew Chem Int Ed Engl. 2016 Nov 21;55(48):15002-15006. doi: 10.1002/anie.201608033. Epub 2016 Oct 26. PMID: 27781345.
Zylstra GJ, Gibson DT. Toluene degradation by Pseudomonas putida F1. Nucleotide sequence of the todC1C2BADE genes and their expression in Escherichia coli. J Biol Chem. 1989 Sep 5;264(25):14940-6. PMID: 2670929.
Martin-Pascual M, Batianis C, Bruinsma L, Asin-Garcia E, Garcia-Morales L, Weusthuis RA, van Kranenburg R, Martins Dos Santos VAP. A navigation guide of synthetic biology tools for Pseudomonas putida. Biotechnol Adv. 2021 Jul-Aug;49:107732. doi: 10.1016/j.biotechadv.2021.107732. Epub 2021 Mar 27. PMID: 33785373.
Joo H, Arisawa A, Lin Z, Arnold FH. A high-throughput digital imaging screen for the discovery and directed evolution of oxygenases. Chem Biol. 1999 Oct;6(10):699-706. doi: 10.1016/s1074-5521(00)80017-4. PMID: 10508682.

Section B: Characterising hypoxia-responsive promoters

DESIGN	A ~400bp putative ANR-dependent promoter region upstream of the terminal oxidase gene ccoN1 in the KT2440 genome was selected for amplification. PCR primers were designed to linearise the vectors pSEVA2213 and pSEVA4313 and delete the EM7 promoter, and sfGFP was selected as a reporter to measure promoter activity through fluorescence microscopy.
BUILD	The native promoter PccoN1 was amplified from P. putida KT2440 genomic DNA and homology regions for cloning were introduced by PCR. The vectors were linearised and pPccoN1 was constructed, first by Gibson Assembly and then by In-Fusion Assembly. sfGFP was digested out from pET28a-sfGFP and introduced downstream of PccoN1 in pPccoN1 and PEM7 in pSEVA2213 by restriction cloning, but we only obtained positive clones for pPEM7-sfGFP.
TEST	Promoter activity was measured by quantifying sfGFP expression in single bacterial cells when exposed to hypoxic stress. We performed the control experiment wherein E. coli TOP10 carrying pPEM7-sfGFP was grown to stationary phase then incubated in argon-flushed tubes at a starting O.D. of 1.0 for six hours, with aerobic cultures used as a reference. We had planned to assay promoter activity in P. putida KT2440, but had insufficient time left to do so.
LEARN	Imaging revealed that sfGFP was strongly expressed under both conditions in cells carrying pPEM7-sfGFP. While the E. coli FNR protein is very similar to the P. putida ANR protein, we expected that FNR regulation of PccoN1 may be weak or non-existent – it would have been more appropriate to assay activity in P. putida had we succeeded in cloning. We did verify, however, that the expression of superfolder GFP is attenuated by hypoxia - this may result from improper folding or due to the effect of hypoxic stress. We believe that sfGFP is still an acceptable reporter for hypoxic assays provided the effects of hypoxia are controlled for using a constitutive promoter as a reference. We did consider switching to a lacZ reporter and using Miller’s assay instead, but the sfGFP approach is more tolerant to hypoxia-induced changes in cell density.

DESIGN	A truncated version of the promoter was designed that contained the -10 region, the -35 region, and the putative ANR-binding box. In addition, we modified the -40 region of the PEM7 promoter to resemble an ANR-binding box as a preliminary step in generating a promoter library. Primers were designed for annealed oligo cloning to introduce these two promoters, a single-box FNR promoter, and the tandem FNR promoter present in the repository into pSEVA2213 with sfGFP inserted downstream.
BUILD	pSEVA2213 was linearised by PCR and digestion sites with restriction sites introduced at the ends, and this PCR product was digested to produce compatible sticky ends. We attempted annealed oligo cloning using this and primer sets for each promoter but the ligations did not work and the plates were clean.
TEST	Had we obtained positive clones, we would have assayed promoter activity under hypoxic conditions in Pseudomonas putida as per the protocol described in the Experiments section.
LEARN	Had we performed these assays, we would have gained some insight into whether these promoters were sufficiently strong in P. putida for the purpose we intended to use them for. The functionality of the truncated promoter would have provided insight into the importance of additional regulatory elements that may be present far upstream, and the functionality of the synthetic PanrEM7 promoter would have demonstrated whether the presence of the ANR-binding box at -40 alone was sufficient for ANR regulation.

Ugidos A, Morales G, Rial E, Williams HD, Rojo F. The coordinate regulation of multiple terminal oxidases by the Pseudomonas putida ANR global regulator. Environ Microbiol. 2008 Jul;10(7):1690-702. doi: 10.1111/j.1462-2920.2008.01586.x. Epub 2008 Mar 12. PMID: 18341582.
Højberg O, Schnider U, Winteler HV, Sørensen J, Haas D. Oxygen-sensing reporter strain of Pseudomonas fluorescens for monitoring the distribution of low-oxygen habitats in soil. Appl Environ Microbiol. 1999 Sep;65(9):4085-93. doi: 10.1128/AEM.65.9.4085-4093.1999. PMID: 10473420; PMCID: PMC99745.
Barnard AM, Green J, Busby SJ. Transcription regulation by tandem-bound FNR at Escherichia coli promoters. J Bacteriol. 2003 Oct;185(20):5993-6004. doi: 10.1128/JB.185.20.5993-6004.2003. PMID: 14526010; PMCID: PMC225037.

Section C: Measuring microbial halocarbon degradation rates

DESIGN	Optimise the separation between the gas chromatogram peaks of the solvent, TCE and the internal standard (Decane).
BUILD	Using hexane, pentane, and diethyl ether as the solvent, we tried optimising various GC settings like the starting oven temperature, hold time, and the slope of temperature increase.
TEST	The chromatograms obtained with hexane showed no peak separation between TCE and hexane due to the low difference in boiling point. With pentane and diethyl ether, the separation could be seen, and was enhanced by reducing starting oven temperature, and increasing the hold time.
LEARN	Solvent choice must be based on the difference in boiling point of the substrate and the solvent. It is critical to use solvent with low boiling point like diethyl ether, and to start with a low enough temperature to vaporise only the solvent and not the substrate.

DESIGN	Study the toxicity of TCE to Pseudomonas putida KT2440.
BUILD	Different amounts of TCE (0mL, 0.05mL, 0.1mL, 0.2mL) were added to 2mL of bacterial cell culture taken in round-bottom flasks (OD = 0.7); and the cells were collected for plating after stirring for 5 hours.
TEST	The plates obtained showed growth only in the control (0mL of TCE) and no growth was observed in the other samples where 0.05mL, 0.1mL and 0.2mL TCE was added.
LEARN	TCE is toxic to the bacteria at least up to 0.05mL TCE for 2mL of OD = 0.7 bacterial culture. Thus, it is crucial to use a very small amount of TCE for the assays (we later went with 0.02 mL), which also means that experimental errors must be minimised to get accurate results with the GC.

DESIGN	Study the degradation rates of TCE and identify the contribution from basal degradation and experimental errors (which may include the evaporation of TCE from the apparatus) with Pseudomonas putida KT2440.
BUILD	Using round-bottom flasks flushed with argon, 2mL of bacterial cell sample (OD = 0.7) was added along with 0.02 mL TCE. At different time stamps (0 min, 20 min, 40 min, 60 min), the relative amount of TCE left was estimated by adding 10mL of diethyl ether and carefully taking 1mL of the organic layer to measure the gas chromatogram using GC. One test was conducted with only the LB broth to check for possible leak of TCE from the apparatus.
TEST	The TCE peaks could be clearly observed, but the integrations of the peaks which are proportional to the amount of TCE in the sample is not observed to stay constant (or show a slight decline due to basal degradation) as expected. Instead, the later time samples show a large decrease in TCE amounts and this was also observed in the trial without the bacteria.
LEARN	We learnt that the apparatus needed a better sealing mechanism to stop the evaporation of TCE. One useful suggestion we received from Dr. Wackett was to reinforce the Teflon septa with Teflon tapes as TCE may escape through a normal tape or Parafilm.

Wackett, L. P., & Gibson, D. T. (1988). Degradation of trichloroethylene by toluene dioxygenase in whole-cell studies with Pseudomonas putida F1. Applied and Environmental Microbiology, 54(7), 1703–1708. https://doi.org/10.1128/aem.54.7.1703-1708.1988.
Wackett, L. P., Brusseau, G. A., Householder, S. R., & Hanson, R. S. (1989). Survey of microbial oxygenases: Trichloroethylene degradation by propane-oxidizing bacteria. Applied and Environmental Microbiology, 55(11), 2960–2964. https://doi.org/10.1128/aem.55.11.2960-2964.1989
Agilent 8890 GC Operation Manual: https://www.agilent.com/cs/library/usermanuals/public/usermanual-gc-operation-8890-g3540-90014-en-agilent.pdf

Section D: Engineering an optimised P. putida strain for halocarbon bioremediation

DESIGN	Based on the results of our promoter activity assays, we intend to select a promoter that strongly upregulates downstream genes under hypoxic conditions and has fairly low leakiness to regulate CamCAB expression. If none of the promoters prove to be useable, we will proceed with the PEM7 promoter or, if necessary, the inducible XylS/Pm system.
BUILD	We intend to clone the Cytochrome P450cam enzyme system downstream of the selected hypoxic promoter either by traditional cloning or by repeating Gibson Assembly. We shall similarly shift the Toluene dioxygenase enzyme system to a non-kanamycin vector with a compatible ori (for instance, pSEVA4313), allowing us to transform P. putida KT2440 with both pCAM and pTOD to construct a strain capable of degrading halocarbons.
TEST	We will test the ability of recombinant Pseudomonas putida KT2440 carrying either pCAM only, pTOD only, both pCAM and pTOD, or pCAM with a hypoxic promoter and pTOD to degrade simple halocarbons or model compounds like TCE using the protocol described in the Experiments section.
LEARN	From the results of these experiments, we will be able to determine if hypoxic regulation of CamCAB results in a measurable increase in degradation efficiency, or, at the very least, results in fewer non-degradable side-products. In addition, we will know if pCAM and pTOD together are sufficient for halocarbon bioremediation, and if P. putida KT2440 is useable as a chassis for such processes. If the system deals adequately with simple halocarbons, we may progress to more relevant and harder-to-degrade molecules.

DESIGN	If the system works as intended, we will attempt to integrate the CamCAB and TodC1C2AB genes into the P. putida genome using R6K suicide vectors. Primers will be designed to amplify the promoter-translational unit-terminator segment from pCAM and pTOD and insert this into pBAMD1-2, a suicide vector for insertion at random loci in the genome.
BUILD	We will PCR amplify the promoter-translational unit-terminator segment from pCAM and pTOD, PCR linearise pBAMD1-2, then perform Gibson Assembly or In-Fusion Assembly. This plasmid will be amplified in E. coli DH5α λpir and used to transform Pseudomonas putida which lacks the machinery for their replication, allowing integration-dependent selection.
TEST	We will test the ability of this engineered strain of Pseudomonas putida to degrade halocarbons with respect to a plasmid-bearing strain as previously. Further, we will culture this strain repeatedly in the absence of antibiotics to understand how this would affect halocarbon degradation ability.
LEARN	Genomic integration generally results in increased stability in the absence of antibiotics and better expression, which are both advantageous for industrial deployment – we would like to confirm that this is indeed a useful endeavour. We may also attempt to knock-out the antibiotic resistance marker in the interest of biosafety.

Wackett LP, Sadowsky MJ, Newman LM, Hur HG, Li S. Metabolism of polyhalogenated compounds by a genetically engineered bacterium. Nature. 1994 Apr 14;368(6472):627-9. doi: 10.1038/368627a0. PMID: 8145847.
Martin-Pascual M, Batianis C, Bruinsma L, Asin-Garcia E, Garcia-Morales L, Weusthuis RA, van Kranenburg R, Martins Dos Santos VAP. A navigation guide of synthetic biology tools for Pseudomonas putida. Biotechnol Adv. 2021 Jul-Aug;49:107732. doi: 10.1016/j.biotechadv.2021.107732. Epub 2021 Mar 27. PMID: 33785373.

Section E: In-Silico Modelling of Enzyme Dynamics

DESIGN	Obtained the protein structures and selected the substrates based on the enzyme mechanisms. Obtained data from literature about the active sites of the enzymes. Designed protocols for molecular docking experiments
BUILD	Executed molecular docking experiments for the halocarbons and camphor for Cytochrome P450cam and the proposed products and toluene for Toluene Dioxygenase. The protocol was optimised further by finding active sites using AutoLigand to concentrate computing power in and around the active site pocket.
TEST	The results were analysed by first cross-checking the docked position of natural ligands (camphor and toluene) against their crystallographic positions. The docking sites were verified by checking the active sites and binding affinities. The order of affinities was obtained from the simulations.
LEARN	Illustrated that HFCs and methane-based halocarbons show lower affinity than their CFC and HCFC counterparts, indicating that they might be harder to degrade. In addition, the binding sites help to understand the mechanism by which the enzymes operate.

DESIGN	Based on the results of binding affinity, we utilised methods from statistical mechanics to use an equation that can help us give predictions for essential parameters for the enzymes.
BUILD	In order to use the equation, we require both binding energies obtained from docking experiments and solvation energies. Solvation energies for our molecules of interest did not exist in databases, so we built a molecular dynamics simulation to simulate the molecule's interactions in water and to predict the solvation energies.
TEST	We ran the model for camphor and hexachloroethane, for which \( k_d \) (dissociation constant) values were available in nature. The results from our simulations and the experiments were used, and the values were compared and found to be close.
LEARN	We were hence able to understand and predict the functionality of our enzyme systems based on results obtained from simulations. The predictability of this can help us to understand if the modification of protein can help improve the rate of our reactions or make the degradation of HFCs more feasible.

DESIGN	We modelled if active site modifications can help us increase reactions' efficiency. A demonstration of this would be helpful in order to maximise the output of our implementation. We conducted a literature survey to find out the suitable modifications to make and modified the sequence accordingly.
BUILD	To build the new and modified model of the enzyme with changes in the residues, AlphaFold was used and verified that the predictions were valid.
TEST	The modified structure was compared with the crystallographic structure, and the changes near the active site were verified. The new protein structure was used for docking with Camphor, Hexachloroethane and HFC 134a.
LEARN	We learnt from modelling that the activity for hexachloroethane increased as predicted. The activity for camphor was also increased; however, the same was not reflected in HFC 134a results. Hence, we decided to look for alternatives beyond active side modifications, such as incorporating better redox partners and calculating or simulating their effects in the future

Forli, S., Huey, R., Pique, M. et al. Computational protein–ligand docking and virtual drug screening with the AutoDock suite. Nat Protoc 11, 905–919 (2016). https://doi.org/10.1038/nprot.2016.051
Walsh, M.E., Kyritsis, P., Eady, N.A.J., Hill, H.A.O. and Wong, L.-L.(2000), Catalytic reductive dehalogenation of hexachloroethane by molecular variants of cytochrome P450cam (CYP101). European Journal of Biochemistry, 267: 5815-5820. https://doi.org/10.1046/j.1432-1327.2000.01666.x
GROMACS, Free Energy of Solvation Tutorialhttps://tutorials.gromacs.org/free-energy-of-solvation.html
Jumper, J., Evans, R., Pritzel, A. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021). https://doi.org/10.1038/s41586-021-03819-2
Ferredoxin-Mediated Electrocatalytic Dehalogenation of Haloalkanes by Cytochrome P450cam, Marc Wirtz, Josef Klucik, and Mario Rivera, Journal of the American Chemical Society 2000 122 (6), 1047-1056 DOI: 10.1021/ja993648o