Experiments

Summary


During this year, we aimed to express four chosen enzymes into an E. coli strain, since it is the most practical and known bacteria, and test their expression and functionality in this chassis: this was a necessary step since three out of our four chosen enzymes are not native to E. coli species and the effects of this change of host are not yet known.
To achieve this, we designed and synthesized novel expression cassettes codifying specific dehalogenases or laccases that will be expressed internally or externally, through the insertion of the display system Lpp-OmpT, on the E. coli strain. After, expression and functionality tests were performed to verify the correct expression of the translated protein, ideally comparing enzyme localization both inside and outside the bacteria membrane, and the enzymatic activity of each protein through the use of a known substrate.



Premises

  1. DNA extraction and purification from bacterial colonies were performed via the QIAprep Spin Miniprep Kit.
  2. Plasmid gel extraction upon electrophoresis was performed using the QIAquick PCR & Gel Cleanup kit.
  3. The concentration of DNA samples was quantified via Thermo Fisher NanoDrop.
  4. All synthetic sequences (expression cassettes and primers) were synthesized by Integrated DNA Technologies (IDT).
  5. After every cloning attempt, the product was sent to BMR Genomics for Sanger sequencing.
  6. Every synthetic sequence was designed using the Benchling platform.

The chosen backbone pJUMP29-1A was fundamental for our clonings since it presents a terminator right after the Biobrick suffix. Since the synthesis facility wasn’t able to synthesize an expression cassette that included a terminator itself, the presence of this Lambda terminator downstream of the Suffix element came in handy for designing and building our system.

During the design of our basic and composite parts, we decided to include in our sequences the NdeI restriction site, formed by a six-base sequence “CATATG”. This particular site was chosen for different possible procedures, but it is always necessary since the “ATG” triplet in the recognition site overlaps the starting codon ATG of our coding sequences, and is needed due to its closer position to the start of the gene.
At this point we faced a problem: pJUMP29-1A presents an NdeI site inside its sequence.
So our solution was to mutate the backbone through mutagenic PCR to remove the unwanted site.

The restriction site mapped on the backbone but was outside of Ori and the kanamycin resistance gene, so it was easier to modify without damaging the plasmid structure. The sequence that encodes the site was:
5’–CCTTACGCATCTGTGCGGTATTTCACACCGCATATGGTGCACTCTCAGTACAATCTGCTCTGATGCCGCATAG–3’

We designed two primers forward and reverse overlapping the restriction site, both of them with a single mismatch on the second T of the sequence:

  • Primer forward: 5’ – CCGCATAAGGTGCACTCTCAGTACAAT – 3’
  • Primer reverse: 5’ – GTGCACCTTATGCGGTGTGAAATACCG – 3’
With these, we performed a mutagenic PCR.

At the end of the procedure, the amplicon was purified with a PCR purification kit and it was digested with the enzyme DpnI, to remove every original plasmid sequence. DpnI is a restriction enzyme that cleaves methylated DNA only: DNA methylation is a process in which methyl groups are added to a DNA sequence, and it’s a process that can happen exclusively when DNA is replicated inside a living host. By eliminating methylated DNA sequences, we eliminate the majority of parental copies of pJUMP29-1A, eliminating all the non-mutagenized sequences. The DpnI-treated PCR product is incubated with PNK later on to improve ligase activity. Ligation was carried on overnight.

To see whether mutagenic PCR was effective, a part of the amplicon was digested with NdeI and run in gel: the gel run showed promising results when comparing the digested plasmid, which was linearized, to the control plasmid left uncut.

Finally, a part of the amplicon was digested with NdeI and run in gel to verify the success of the mutagenesis.

The starting point of our project was to insert the expression cassette for the intracellular enzyme DeHa2 (Part BBa_K5109001) into our chosen E. coli strain, DH5α, as it was already widely used in our laboratory and it’s very efficient in the expression of orthologous proteins.


After the sequence arrived at our lab, we started our cloning protocols: to insert our intracellular expression cassette containing DeHa2 (Part BBa_K5109001) into the pJUMP29-1A vector (Part BBa_J428341), our designed backbone, we used the digestion-ligation method with the BioBrick prefix and suffix.

  1. The first thing we did was to digest, at the same time, the expression cassette and the pJUMP29-1A vector with the restriction enzymes EcoRI and PstI, both at the further ends of the BioBrick prefix and suffix respectively.
    This way, the construct is inserted in a directed manner and the BioBrick prefix and suffix will be restored.
    We proceeded to carry out the reaction at 37°C for 1:30 h, then we inactivated the enzymes at 80°C for 20 minutes. When both reactions were inactivated, we ran a gel electrophoresis to verify the correct digestion of our backbone: it was not possible to check the digestion outcome for the insert, as the terminal ends resulting in successful digestion were only 20 bp long and would not be visible on the gel.
  2. When the digestion was confirmed, we purified the digested BBa_K5109001 using the Qiagen purification kit and quantified the result, meanwhile, we dephosphorylated the backbone using an alkaline phosphatase, as to prevent the re-closing of pJUMP29-1A without any insert. The reaction was carried out at 37°C for 20 minutes and the phosphatase was inactivated at 80°C afterwards.
  3. Then, we set up a ligation protocol: we decided to use a 5:1 ratio, meaning that the concentration of the insert in this reaction was five times higher than the concentration of the backbone; this was done to maximize our possibilities of a successful ligation.
    We prepared two different ligation reactions, one of them containing only the dephosphorylated backbone without the insert: this reaction was created to act as a control, as after the transformation of our two ligation reactions in E.coli there should be a clear difference in numbers of colonies grown from one plate to the other.
  4. We left our ligation reaction at 16°C overnight, then we transformed them into chemically-competent DH5α cells: this was done via heat shock, where pores are formed on the bacterial membrane after a sharp change in temperature (from 42°C to 0°C) so that the external DNA can enter the bacteria. When the heat shock was complete, we left the bacteria to recover in LB at 37°C for an hour, then we plated them in LB agar with kanamycin, the selection marker present in our backbone BBa_J428341.
    When the colonies were visible, a clear distinction in number was seen between the two plates: the control plate showed a lot less of colonies than the ligation plate, making us think that the lack of insert didn’t help the growth of viable colonies and that the ligation plate contained complete and recombinant plasmids.
  5. We picked a handful of colonies from both plates and we run a colony PCR, using a set of primers designed to bind on the BioBrick Prefix (forward primer) and Suffix (reverse primer): this way, It's possible to visualize the result of the PCR amplification on gel, where a band 800-900 bp long should be visible if our desired expression cassette has been correctly cloned.
From now on, to have a better chance to perform clonings successfully, we changed the E. coli strain used. We switched to a different strain, called Top10 F’: this strain can constitutively express the LacI repressor present in the F’ episome, an important component to the regulation of the Lac operon. With this change, our aim was to repress any expression of our protein of interest simplifying clonings and experiments.

During the first two months of work, we couldn’t get any results from the cloning trials. So we decided to give a chance to a different technique that uses homologous recombination. In this technique, both vector and insert are amplified by PCR with primers that have overlapping ends, and E. coli’s ability to perform in vivo homologous recombination allows for our insert and vector’s extremities to pair, be cut and then fused together, obtaining a closed vector. [1]


The purpose of the work was to clone the DeHa2 type 2 intracellular expression tool (Part BBa K5109001), inside our chosen backbone pJUMP29-1A. The first strategy of RFC 10 cloning wasn’t showing any results, so we designed and ordered specific primers to perform the recombination.

  1. The protocol includes a PCR amplification as the first step, to amplify the desired insert and vector with primers including identical ends to both of them. After the end of the PCR and the purification of the product, both of the amplicons will present the same ends and will be in a linear form.
  2. Vector and insert will be mixed together in a solution of 3:1 insert to vector proportion, and this preparation will be used to chemically transform competent cells.
  3. After transformation, if recombination occurs, cells are plated and grown overnight on LB + Agar + antibiotic (kanamycin, in our case). The day after, colonies will grow only if homologous recombination has occurred, since they will be equipped with a plasmid carrying antibiotic resistance.
  4. We relied on Top10 F’ strains for this cloning, even if they present a deletion of RecA in order to avoid nonspecific recombination, since this type of homologous recombination is independent from the presence of RecA but it instead exploits different cellular mechanisms.

After our first unsuccessful attempt, we tried to clone three new expression cassettes, this time for the extracellular expression of DeHa2 (BBa_K5109023), Dehalogenase S (BBa_K5109020) and Laccase S (BBa_K5109022): all of these cassettes contain a Lpp-OmpT-linker part (BBa_K5109004) to create a display system for the enzymes, placed upstream of the GOIs. We cloned them simultaneously, using the same protocols and reagents, so that all three of them had the same chance of succeeding.




Since time was starting to become a limiting factor, we decided not to try the intracellular expression of the enzyme DehalogenaseS (BBa_K5109013) and Laccase S (BBa_K5109015) and to instead try directly the extracellular expression using the display system Lpp-OmpT, so that it would be a more interesting study during our expression and functionality assays.

  1. We digested BBa_K5109023 (DeHa2 extracellular expression cassette), BBa_K5109020 (Dehalogenase S extracellular expression cassette), BBa_K5109022 (Laccase S extracellular expression cassette) and BBa_J428341 (pJUMP29-1A) with the restriction enzymes EcoRI and PstI, so that the three cassettes would have matching ends with the digested pJUMP and would pair up without problems.
    The digestion of the synthetic expression cassettes was performed as illustrated during the cloning of BBa_K5109001, but something different was done for the digestion of the backbone: the digestion was done in a final volume of 60 ul, then, after the enzymes were inactivated after 2h at 37°C, all the digested backbone was run in gel electrophoresis. The gel was also made in a different way: we usually set up the gel at 1% agarose concentration and using Midori Green as our intercalating agent, but we prepared this gel at 0.7% agarose concentration, to better separate the different bands of the digestion result and the ladder. We also used ethidium bromide as our intercalant agent, as to better visualize the different bands on the gel.
  2. After the digestion was completed and the gel electrophoresis had stopped running, we performed a gel extraction to gather our digested backbone: this was done to ensure that no original insert was present in the subsequent steps and to minimize the chance of another negative result. The gel extraction was performed using the QIAquick Gel Cleanup kit and we quantified the result afterwards; meanwhile, we purified our digested expression cassettes using the Qiagen purification kit, and we later quantified the obtained results.
  3. When both the backbone and inserts were ready, we moved on with the ligation protocol. We stayed with an insert: backbone concentration ratio of 5:1 for all synthetic sequences and we left all three ligation reactions at 16°C for 16h, then we inactivated the ligase at 65°C for 20 minutes. This time no control ligation was done.
  4. After the ligation was complete, we transformed our ligation products in chemical-competent Top10F’ bacteria: the external recombinant DNA entered the bacterial membrane thanks to a heat shock, performed by leaving the bacteria and the ligated DNA together at 42°C for 1 minute, then immediately transferring them on ice for 1 minute. After the heat shock was performed, we left our transformed bacteria in LB to recover and to express their selection marker gene, consisting in a kanamycin antibiotic. Afterwards, we plated the bacteria in LB agar with kanamycin and left them at 37°C until colonies started to grow.
  5. When a decent number of colonies was present on all three plates, we picked all colonies that grew on them and performed a colony PCR.
    We performed this step in two different ways: at first we still used the original primers that were used as during the cloning of BBa_K5109001, then we switched to a different set of primers that would match to the DNA a few bases outside the synthetic expression cassette and land on the pJUMP backbone, to obtain a more targeted binding and avoid any aspecific binding results. When the PCR reaction was completed, we ran all our results in gel electrophoresis to see if cloning was successful this time.

We were planning to create a new composite part for the expression of Laccase E found in E. coli’s genome thanks to our bioinformatic analysis. Due to time restrictions we ordered the extracellular expression cassette containing Laccase E’s coding sequence, hoping we would be able to clone it. However, due to issues to this composite part’s synthesis, we were not able to get it shipped and perform the ligation and digestion process with it.

In order to test the correct functioning of dehalogenases, we decided to set some tests that would use chloroacetic acid. Dehalogenases are able to break the bond that connects halogens to the carbon chain so, if the protein expressed by E. coli is correctly functioning, there would be the degradation of the molecule of the chloroacetic acid.

We tested 2 colonies (#4 and #9) of E. coli Top10F’ transformed with the Lpp-OmpT-DeHa2 construct by creating a medium composed of LB, chloroacetic acid 2.5 mM and IPTG 5 uM. We sampled the bacterial cultures after 0, 24 and 48 hours and treated the samples to lysate and remove the cells, in order to make them compatible with the mass spectrometry analysis[2][3]. Every time point was treated with two different procedures, in order to determine both the concentration of chloroacetic acid in the medium and inside the cells and to be able to evaluate the internalization of the compound by E. coli.

The samples that we prepared were analyzed in the chemistry facility of our university.
Samples were thawed, diluted 1:100 with water/acetonitrile 50:50 and analyzed by FIA-HRMS (flow injection analysis - high resolution mass spectrometry). Briefly, 20 µL of sample were injected in the mass spectrometer (Q-Exactive Hybrid Quadrupole-Orbitrap Mass Spectrometer, Thermo Fisher Scientific, Germany) by using acetonitrile as eluent at 0.300 mL/min flow.
Data acquisition was performed in negative ionization. The capillary voltage was set to 2.8kV, capillary temperature was 320°C, auxiliary gas and sheath gas were nitrogen (300°C) at 20 and 10 psi, respectively. Resolution was set to 35000, AGC target was 3x10 6 and max injection time was 200 ms. Full-scan range was within m/z 50-250. Calibration was performed by Pierce ESI Negative Ion Calibration Solution (Thermo Fisher Scientific, Germany). The pseudo-chromatographic peaks of chloroacetate were identified and integrated by considering the Extracted Ion Chromatogram (EIC) of the [M-H] - species selected with a window of 5 ppm.

E. coli in methanol


Methanol is an organic solvent that can be used for the desorption of GAC filters. This can be done by using solutions of methanol 100% (24.7M) or ethanol 50% (8.56 M) with but this concentration is not compatible with the survival of E. coli. [4]
In the hypothesis where we use GAC filters to collect PFAS and we choose a solution with methanol for the desorption, this solution would end up in the bioreactor with the bacteria. Both PFOA and Methanol are toxic so we decided to analyze firstly the toxicity of the two components alone and then together. In order to have them together, representing the condition inside the bioreactor, we need to find a high concentration of methanol that allows the bacteria to grow and that can be used as a solvent for dissolving PFOA and PFBA in LB.
We decided to test these concentrations of methanol:
0, 0.25, 0.50, 0.75, 1.00, 1.25, 1.50, 1.75, 2.00, 2.25, 2.50, 2.75.

During the testing of E. coli growth, we learned to take some precautionary measures when working with methanol. In order to reduce the phenomenon of evaporation, we choose to prepare the stock solutions with methanol just before the preparation of the plate and, if necessary, put them in the fridge until that time.


E. coli in ethanol


Another option for the desorption of GAC filters, is to use a solution with ethanol. Similarly to what has been said with methanol, we decided to analyze how the growth is affected by increasing ethanol concentrations.
We used the same concentrations that were chosen for methanol.
Again, our goal is to find the concentration that allows the bacteria to grow and that can be used as a solvent for dissolving PFOA and PFBA in order to replicate the conditions inside the bioreactor.

Since ethanol has volatility problems too, we prepared our stock solutions just before the assembling of the plate.


E. coli in NaCl


Our project considers the use of anionic exchange resins, that would actually be a better candidate in our plant design, so there is the need to find a solution that can be used for their desorption. After a literature search, we found that one of the best desorbing solutions is water with NaCl 10%. [5]
E. coli is usually grown in LB which is an optimal cultural medium with 1% of NaCl. We decided to start from that concentration and increase it to 2%, 4%, 6%, 8% up until 10% and confront the results that are reported in the “results” page of the wiki.


E. coli in PFOA in LB


When working with PFAS, we decided to analyze increasing concentrations that will very likely be much higher than the one inside the bioreactor. This choice was done in order to understand what would be the maximum amount of PFAS that E. coli is able to endure.

We then decided to analyze PFOA and PFBA at these concentrations:
0 mM, 1.0 mM, 2.5 mM, 5.0 mM, 7.5 mM, 10.0 mM, 15.0 mM, 20.0 mM.

The choice to use LB as the solvent was done in order to have a better idea of the damage that is caused just by PFAS molecules to bacteria without any other variable.


E. coli in PFBA in LB


PFBA is a shorter molecule of PFAS that we decided to use after a conversation and visit “ACQUE VENETE”, a water distribution company in our territory. On this occasion, they told us that one of the most frequent types of PFAS that they found in waters was PFBA. We also hypothesized, would make it an easier molecule to degrade. For these reasons we decided to implement PFBA in our tests.Since there are not may studies about PFBA that test different concentrations, we decided to precede and test the same as we did with PFOA.

PFBA we used was in a liquid form, sensitive to light and also volatile so our experiments were done in a particularly controlled manner.


E. coli with PFOA and PFBA in methanol


The concentration of methanol that we chose for dissolving PFAS molecule was 1.00M. This choice was done after analyzing the growth rates and the final OD measured in the methanol experiments.


E. coli with PFOA and PFBA in ethanol


PFOA and PFBA were dissolved in ethanol 0.75M. This concentration was chosen after analyzing the growth rates and final OD in the ethanol tests.


E. coli with PFOA and PFBA with NaCl


According to literature, anionic exchange resins need a desorption solution with 10% of NaCl.
After the tests done with NaCl, we understood that the growth of E. coli would be almost completely inhibited after 2% NaCl. In order to replicate the condition inside the bioreactor, we chose that concentration for more tests where the cultural medium contained both NaCl and PFOA and PFBA at increasing concentrations (the same as the ones in previous tests). This type of experiment gave us a better idea of how E. coli reacts with those components and if a longer carbon chain as the one of PFOA would interact in a different way when compared with PFBA.


E. coli with chloroacetic acid


To test the correct functioning of the dehalogenase expressed on the surface of E.coli, we would need to test it with chloroacetic acid before trying with PFAS molecules. For this reason, we were interested in knowing whether E. coli would be able to survive in a culture medium containing chloroacetic acid. The concentrations of chloroacetic acid that we will use to test the function of the enzymes, will most likely be 0.5 mM and 2.5 mM so we were interested in testing the bacterial survival at these concentrations and a little beyond. We decided to choose concentration between 0 and 3.5 mM.


E. coli in synthetic wastewater


In the bioreactor, it is highly unlikely that LB will be the major component of the cultural medium for E. coli. Even though characterizing a process of E. coli growth using its optimal medium is a very effective way to understand the impact of the other molecules, working with synthetic wastewater would give us a more realistic point of view. In order to do so, we decided to use a recipe that was previously used in a bioremediation research for the degradation of glyphosate [6]. Our first idea was , after creating the solution, to test if E. coli is able to survive with the amount of nutrients that it contains and then replicate each growth test that we did using LB but substituting it with the synthetic wastewater.

This is the recipe that we firstly used for the synthetic wastewater:
  • NaCl: 7 mg
  • CaCl2: 3 mg
  • MgSO4: 2 mg
  • Beef extract: 100 mg
  • NH4Cl: 199 mg
  • Sodium acetate trihydrate: 163.6 mg
  • K2HPO4: 18 mg
The results of this test are reported in the results page of the wiki and present the inability of E. coli to grow in these conditions.
We then hypothesized that there weren’t enough nutrients present in the solution and tried to modify the recipe by doubling the amount of carbon and nitrogen sources.
The synthetic wastewater recipe became the following:
  • NaCl: 7 mg
  • CaCl2: 3 mg
  • MgSO4: 2 mg
  • Meat extract: 200 mg
  • NH4Cl: 398 mg
  • Sodium acetate trihydrate: 327.2 mg
  • K2HPO4: 18 mg
The results with the new recipe improved but the growth of E. coli in this condition was very far from the one in LB and showed a lack of replicability.
For these reasons, we decided to scale up our experiment, hoping that larger volumes would help the bacterial growth. To do this, we decided to change our experimental setup and, instead of using the plate reader, to perform a growth test in shaking flasks.

In order to have more quantitative data than we had with the previous experiments, we decided to measure the dry weight in addition to the OD600. These tests were done every two hours and we collected data that has been plotted into graphs presented in the results part of the wiki.


E. coli in IPTG


The plasmid that has been used for the transformation of E. coli Top10F’ strain, enables bacteria to produce our desired protein, the dehalogenase DeHa2, under one condition: it needs to be inducted with IPTG.
Our previous attempts of cloning showed a significant toxicity of this protein in E. coli. Therefore, we were interested in knowing what the ideal concentration of IPTG that we could use for the induction would be and also have an insight on the toxicity of the dehalogenase on the bacteria itself.
For this reason, we decided to perform a test in which we analyzed the growth rates and the final ODs of the Top10F' strain that express the dehalogenases after induction with different concentrations of IPTG.
We choose to start from a very low concentration of inductor (0.5 µM) and increase it by a 10 factor until we reached the optimal concentration (0.5 mM).

Our tests were done at the plate reader that measures the OD 600 every 5 minutes for about 14 hours. The results that we obtained are reported in the results page of the wiki.


References:
  1. Jacobus AP, Gross J (2015) Optimal Cloning of PCR Fragments by Homologous Recombination in Escherichia coli. PLoS ONE 10(3): e0119221. doi:10.1371/journal.pone.0119221
  2. Farajollahi S, Lombardo NV, Crenshaw MD, Guo HB, Doherty ME, Davison TR, et al. Defluorination of Organofluorine Compounds Using Dehalogenase Enzymes from Delftia acidovorans (D4B). ACS Omega. 2024;9(26):28546–55.
  3. Vascon F. A Synthetic Biology Approach to PFAS Bioremediation: Preliminary Computational and Molecular Studies on Fluoroacetate Dehalogenase [Tesi di Laurea Magistrale in Biotecnologie Industriali]. Università degli Studi di Padova; 2019.
  4. Deng S, Nie Y, Du Z, et al. Enhanced adsorption of perfluorooctane sulfonate and perfluorooctanoate by bamboo-derived granular activated carbon. Journal of Hazardous Materials. 2015;282:150-157. doi:10.1016/j.jhazmat.2014.03.045
  5. Dixit F, Barbeau B, Mostafavi SG, Mohseni M. Removal of legacy PFAS and other fluorotelomers: Optimized regeneration strategies in DOM-rich waters. Water Research. 2020;183:116098. doi:10.1016/j.watres.2020.116098
  6. Borella L, Novello G, Gasparotto M, et al. Design and experimental validation of an optimized microalgae-bacteria consortium for the bioremediation of glyphosate in continuous photobioreactors. Journal of Hazardous Materials. 2023;441:129921. doi:10.1016/j.jhazmat.2022.129921

Let's find out more about our project