Design

Module 1 : Engineering an artificial metabolic pathway to degrade halocarbon compounds

We set out to engineer a metabolic pathway that can degrade a generic halocarbon compound. Enzymes that remove chlorine and fluorine from organic molecules are rare in nature, mostly because of the fact that C-F and C-Cl bonds are very strong and the release of fluoride and chloride ions requires immense redox power. The deployment of such enzymes requires a chassis that has a robust metabolism and can survive performing such harsh transformations. The enzyme systems composing this pathway should be able to take a halocarbon and break it down into environmentally-harmless products that may also be useful for the bacteria. We considered several enzyme systems during the initial design phase, and were assisted in this by Dr. Debasis Das.

Enzymes for Halocarbon Degradation

The Cytochrome P450cam monooxygenase is present naturally in many strains of Pseudomonas putida and is principally involved in camphor metabolism.^[1] We were motivated to use this enzyme for halocarbon degradation as it was found to be capable of reductive dehalogenation under certain conditions. This system has three components – Cytochrome P450cam monoxygenase, Putidaredoxin (complexed with 2Fe-2S cofactors), and NADH-dependent Putidaredoxin reductase. The Cam operon in P. putida consists of the CamC, CamA, and CamB genes, which code respectively for Cytochrome P450cam, NADH dependant Putaredoxin reductase, and Putidaredoxin.^[2] The enzyme shows optimal activity at 20-37 degrees Celsius and pH 6-8.^[1] Cytochrome P450cam is functional under both aerobic and anaerobic conditions but produces different products with halocarbon substrates. Under anaerobic conditions, it tends to remove vicinal halogens and form a double bond.^[3] This characteristic degradation seems to work only when at least one of the two vicinal halogens is chlorine, bromine, or iodine, and degradation does not seem to occur in halocarbons with one carbon or halocarbons with vicinal halogen combinations including two fluorine atoms.^[3],[7] The removal of vicinal halogens is based on size (i.e., larger halogens are removed more easily). Under aerobic conditions, an oxidized product is formed instead of a reduced one. Experimental evidence shows that under aerobic conditions, Cytochrome P450cam converts the carbon with fewer or weaker halogen groups bound into a COOH group.^[8]

Cytochrome P450cam under aerobic conditions:

Cytochrome P450cam under anaerobic conditions:

Toluene dioxygenase is present naturally in strains of Pseudomonas putida and is involved in toluene metabolism, where it converts a double bond in toluene to a vicinal diol. We were interested in this enzyme as it oxidises highly-stable double bonds, which are present in many halocarbons and in products formed when Cytochrome P450cam reduces haloalkanes. This enzyme system has three major components – an NADH reductase coded by the TodA gene, a Ferredoxin coded by TodB, and a dioxygenase. The oxygenase has an alpha and a beta subunit coded by the TodC1 and TodC2 genes respectively. It has a pH optima of 6-7.5 and a temperature optima of 28 degrees Celsius.^[1] Under aerobic conditions, Toluene dioxygenase has been shown to convert halocarbons with at least one double bond to vicinal diols and then to further convert the remaining halogens on the halogenated vicinal diols to OH groups.^[4] Conversion from a double bonded structure to a vicinal diol is trivial and is known to occur naturally with this enzyme, but the second conversion is largely inferred from experimental observations.^[3] Toluene dioxygenase is useful as a route for the degradation of halomethanes, which Cytochrome P450cam does not act on. The utility role of this enzymes lies in degrading substrates containing one carbon, degrading substrates containing a double bond, and degrading the product formed from Cytochrome P450cam degradation under anaerobic conditions.

Catalytic cycle for the action of Toluene dioxygenase:

Fluoroacetate dehalogenase is found naturally in certain strains of Pseudomonas putida and removes fluoride ions from the alpha position of the carboxylic group in fluoroacetate compounds.^[10],[11] This makes it ideal for degrading the products of aerobic oxidation by Cytochrome P450cam, but it is very specific and has not shown halogen removal in any other type of halocarbon.

Methane monooxygenases (MMOs) are enzymes found mainly in methanotrophic bacteria where they convert methanol to methane. Two types exist, soluble MMO (sMMO) and particulate MMO (pMMO). These have shown to degrade single carbon-containing halocarbons into oxidized products that are metabolizable.^[5],[6] The catalytic cycle for halocarbon degradation by MMOs involves two coupled Fe centres and bound oxygen. Substrates with double bonds are converted to epoxides and then further degraded while the C-H bonds are oxidised to hydroxyl groups in halomethanes. However, MMO-dependent degradation is extremely slow – methanotrophic bacteria with this enzyme take days while the others take hours for significant halocarbon degradation. In addition, methanotrophic bacteria are difficult to culture, obtain and maintain.^[7]

Chassis considerations and SynBio tools

Pseudomonas putida is an excellent chassis for bioremediation, possessing a versatile metabolism and a remarkable ability to sustain difficult redox reactions under harsh operational conditions.^[15] It is resistant to organic pollutants and oxidative stress, grows rapidly, and has simple nutritional requirements. We selected the strain KT2440 primarily for its biosafety credentials^[15], which makes it suitable for industrial-scale deployment unlike prior attempts at halocarbon bioremediation. The availability of an extensive synthetic biology toolbox greatly simplified our work.

Pseudomonas putida KT2440 was a gift from Dr. Prashant Phale at IIT-Bombay, who also directed us to contact SEVA for tools. The Standard European Vector Architecture (SEVA) repository possesses a large library of Pseudomonas-compatible vectors composed of modular backbone elements, and kindly agreed to provide us with the vectors we required. We selected medium-copy ori (pBBR1 and pRK2) vectors to ease metabolic burden and paired these with the strong constitutive EM7 promoter and the strong inducible XylS/Pm promoter to ensure sufficient enzyme expression.^[15]

Designing plasmids for enzyme expression

We decided to construct a two-component metabolic pathway with the CamCAB and TodC1C2AB enzyme systems for degrading halocarbon compounds. We opted to assemble each enzyme system in a single plasmid, with the component genes arranged in the native configuration to produce an operon that could be driven by a single promoter. The strong Pseudomonas RBS.15 sequence precedes each gene to ensure efficient translation. Expression plasmids containing codon-optimised versions of CamC, CamA, and CamB were available in the AddGene repository^[13] and we opted to use these to reconstitute the Cytochrome P450cam system. The TodC1, TodC2, TodA, and TodB gene sequences were sourced from P. putida F1 Tod operon^[14], but had to be codon-optimised to reduce GC content to qualify them for DNA synthesis. Constraints on complexity forced us to break up the Toluene dioxygenase system into three fragments. The fact that three fragments had to be introduced into each vector necessitated that we use restriction-free cloning methods; we settled for Gibson Assembly as the lab we worked in was familiar with this technique.

The plasmid pCAM was designed to expresses the three-component Cytochrome P450cam enzyme system as a reconstituted operon under the control of the strong constitutive EM7 promoter (pCAM7) or the 2-methylbenzoate inducible XylS/Pm promoter (pCAM5). The operon as designed consists of the CamC, CamA, and CamB genes and was supposed to be inserted either into the pSEVA438 (medium-copy, streptomycin-resistance, XylS/Pm regulator) or into pSEVA2213 (medium-copy, kanamycin-resistance, EM7 promoter).^[24]

The plasmid pTOD was designed to expresses the four-component Toluene dioxygenase enzyme system as a reconstituted operon under the control of the strong constitutive EM7 promoter (pTOD6). The operon as designed consists of the TodC1, TodC2, TodA, and TodB genes and was supposed to be inserted into pSEVA2213 (medium-copy, kanamycin-resistance, EM7 promoter).^[24]

Module 2: Characterising hypoxic promoters for Pseudomonas species

The Cytochrome P450cam enzyme system oxidises halocarbon compounds under aerobic conditions. Instead of performing reductive dehalogenation to produce a haloalkene compound that can be acted on by Toluene dioxygenase, it instead tends to convert the carbon bound to fewer or weaker halogen groups to a COOH group and generates a haloacetic acid product^[8] that cannot be degraded by Toluene dioxygenase but is not harmless either. To increase the efficiency with which these enzyme systems degrade halocarbons, it is thus necessary that we somehow restrict the activity of Cytochrome P450cam to hypoxic conditions only to minimise the quantity of non-degradable side-products. We attempted to understand if placing CamCAB under the control of a hypoxic promoter would result in improved efficacy.

The ANR global regulator

Transcriptional regulation with respect to oxygen availability in Pseudomonas species is principally achieved through the ANR (anaerobic regulation of arginine deaminase and nitrate reduction) global regulator, which controls a variety of metabolic functions related to the aerobic-to-anaerobic transition and is active under hypoxic conditions.^[17] The ANR regulator is homologous to the Escherichia coli FNR (fumarate and nitrate reductase regulator) protein, which senses oxygen concentration through the oxidation state of iron in a [4Fe–4S]²⁺ or a [2Fe–2S]²⁺ cluster bound to four conserved Cysteine residues. In E. coli, under low oxygen conditions, this cluster undergoes reduction and converts FNR to the dimeric active form. The active FNR binds through a helix-turn helix motif at the C-terminal end to conserved -40 regions (termed FNR-boxes) in FNR-dependant promoters and regulates transcription.^[17]

This mechanism is conserved across FNR family regulators; the dimeric active form of ANR carries a [4Fe–4S]²⁺ cluster and binds to the conserved ANR-binding box (5′-TTGATNNNNATCAA-3′) that is typically located in the -40 region of regulated promoters.^[18] Upon exposure to oxygen, this iron-sulphur cluster is partly destroyed and the ANR loses its DNA binding and gene regulatory ability.^[18] Depending on the position of its binding site, ANR can activate or repress transcription under hypoxic conditions where it exists in its active form.

Designing promoter sequences

The Pseudomonas putida KT2440 terminal oxidase Cbb3-1 is known to be activated under low oxygen tension. In particular, transcription of the gene ccoN1 undergoes a ~500-fold increase under hypoxic conditions in an ANR-dependent manner.^[16] The promoter of ccoN1 (hereafter, PccoN1) contains a putative ANR-binding box at the -40.5 region and was selected as a potential candidate for characterisation. In the interest of being conservative, we opted to amplify a ~400bp region starting from the transcription start site from the KT2440 genome for characterisation. The -10 and -35 elements are recognised by RNA polymerase bound to sigma-70, with the -10 region closely matching the P. putida consensus.^[16]

We also designed Panr, a truncated 58bp version of PccoN1 that contains the -10 and -35 regions and the putative -40.5 ANR-binding box, which in theory should be induced by ANR under hypoxic conditions. As a reference, we opted to use the existing Pfnr promoter from the iGEM registry that possesses tandem FNR-binding boxes. This promoter was quite long and did not lend itself well to our assembly strategy, so we also attempted to test a signficantly shorter single-box FNR promoter from literature.^[19] Finally, we wished to determine if the introduction of an ANR-binding box at the -40 region was sufficient by itself to introduce ANR-regulation to an otherwise constitutive promoter, so we constructed PanrEM7 by modifying the-35 region of the strong EM7 promoter to match the consensus ANR-binding sequence.

Construction and measurement

While PccoN1 was sufficiently long for Gibson Assembly to be effective, the short sizes of the promoter sequences made annealed oligo cloning a more reasonable prospect.^[23] We opted to use pSEVA2213 (medium-copy, kanamycin-resistance, EM7 promoter) as the vector backbone, deleting the EM7 promoter during PCR linearisation and introducing restriction sites at the ends (with sufficient space to ensure good digestion efficiency) to insert the promoter annealed oligos. We decided to measure the activity of these promoters under hypoxic and normoxic conditions by inserting sfGFP downstream through traditional cloning and imaging single cells using a fluorescence microscope.

Promoter activity in such contexts in has been measured in literature both using a beta-galactosidase reporter with Miller’s Assay, and by measuring fluorescence from an sfGFP promoter^[20] – we opted to use sfGFP as it was immediately available and would allow measurements at the scale of single cells to be taken. Prior iGEM teams have reported issues with fluorescent proteins folding incorrectly under low oxygen tension. Nevertheless, we can control for differences in fluorescence intensity that result from factors independent of promoter activity like the slower rate of sfGFP chromophore formation^[22] or by global changes in transcription under hypoxia. We can achieve this by normalising all measurements with respect to the measurements obtained when a strong constitutive promoter with no oxygen-sensitivity drives sfGFP expression under hypoxic and normoxic conditions. Activity can then be quantified by imaging single cells and measuring fluorescence intensity across a large number of cells in the presence and absence of oxygen.

REFERENCES