Engineering success

MODULE 0

Overview:

The first module of our DBTL cycle shows how our initial project idea shifted on the basis of iHP and testing.

Design:

The team of MIT-MAHE set out to solve the problem of triclocarban entering our environment and causing major issues. We initially identified a novel TccA amidase enzyme expressed in Ochrobactrum sp. TCC-2. This amidase had only 27-38% sequence similarity with other known amidases. This enzyme is capable of breaking down triclocarban (TCC) into 4-chloroaniline and 3,4-dichloroaniline.[1]

We initially wanted to implement this system within the aeration tank of the wastewater treatment plant. The aim was to increase the enzyme efficiency to function within the hydraulic retention time of the wastewater treatment plant.

Hydraulic retention time (HRT) of the aeration tank of a wastewater treatment plant is the amount of time taken for one full volume of the aeration tank to pass out of the aeration tank after completion of process.
HRT=Volume of the aeration tank/flow rate of the water Commonly, HRT for conventional activated sludge processes is about 2-24 hours[7].

We wanted to achieve this by modifying amino acid residues through an in-silico version of site directed mutagenesis. We wanted to identify the key residues involved in the catalytic action and improve the binding affinity of the substrate to the enzyme, which we had initially believed might increase the efficiency of the enzyme.

Build:

The TccA amidase enzyme has not been fully characterized in literature. Therefore, the first step before building our genetic circuit was to understand the structure, binding site and working of the enzyme.

We predicted the structure of the TccA amidase as well as the binding sites. Since most of our data was not obtained via experimentation, our Human practices team reached out to many experts, including Mr. Omkar Khade and Dr. Gurunath Ramanathan, to learn more about the feasibility of our project idea.

Test:

We performed docking studies with the predicted binding site. The docking results from Autodock Vina and SeamDock showed ΔG values of -7.51 kcal/mol and -4.7 kcal/mol respectively. This implies that TCC has a good affinity towards the TccA amidase protein.

These ΔG values imply that there is a stable binding of TCC to the TccA amidase. This suggests that the predicted binding sites were correct.

Learn:

Through iHP meetings with Mr. Omkar Khade and Dr. Gurunath Ramanathan, we learnt that manually altering the amino acids in close proximity to the catalytic triad requires a comprehensive understanding of the individual functions of each amino acid involved. A random mutagenesis approach would be difficult to perform in-silico as such experiments are usually only performed in-vitro. As a result, we felt the best way forward was to focus on the initial problem statement-dealing with TCC in wastewater, with an alternative approach.

---

MODULE 1

Design:

Ensuring our focus on the problem statement, and taking note of everything learnt in Module 0, we looked back towards our previous human practices interactions. One of the major questions that had been posed to us was the formation of toxic byproducts 4-chloroaniline and 3,4-dichloroaniline on biodegrading TCC. Considering these changes, we decided to focus our project on breaking down the TCC into non-toxic byproducts.

Through literature surveys, we studied the presence of gene clusters as a part of other pathways in bacteria that could degrade the above mentioned by-products. However, insertion of such large cassettes within known model organisms could cause severe metabolic load on the cell and may lead to stunted growth or loss of plasmid.

Hence, we looked for potential model organisms that had not been characterized for this purpose. An important parameter was the existence of pathways that could degrade the chloroaniline byproducts of TCC breakdown. Through extensive literature survey, we narrowed down our possible chassis options to three species capable of degrading the byproducts -Acinetobacter baylyi [3], Pseudomonas putida [4] [5]and Pseudomonas fluorescens [6].
All three of these organisms had the natural capability to degrade the toxic by-products. The other major criteria we considered included:

• Survivability of the bacteria in the pH range of the wastewater and sludge(6.5-8)
• Survivability of the bacteria in the temperature range of the sludge tank of the WWTP(20-37° C)
• Survivability at different TCC concentration ranges present in the sludge(13,000-21,000 ng/ml of TCC or 13-21 ppm of TCC)

Aim: To design experiments to check which of our three chosen bacterial species: Pseudomonas putida, Pseudomonas fluorescens, and Acinetobacter baylyi had the best survivability in wastewater treatment plant conditions

As TCC does not uniformly dissolve in small volumes of water, we dissolve TCC in acetonitrile at first and then dilute this mixture in water. To ensure that acetonitrile does not have any adverse effect on the growth of our bacteria, we take acetonitrile-containing media as a condition as well.

Build:

Growth Curves To perform these experiments, we inoculated the bacteria in media that simulated the conditions of the sludge tank.The growth was measured by checking the arbitrary units of optical density at 600nm. To convert to absolute units, a calibration curve was plotted for wet weight of the cell vs OD.

1. pH The growth of the bacterial species at pH range of 6 to 8 was growth in LB media.
2. Temperature For temperature studies, we inoculated the bacteria in LB broth media and studied the growth over a temperature range of 24°C to 37°C.
3. Acetonitrile To dissolve and dilute TCC in water, we need to dissolve it in acetonitrile first as TCC has extremely low solubility in water. Since acetonitrile will be in the media that our bacteria grow in, we have to make sure the presence of this compound does not impede the steady growth of our organisms.
4. TCC (Sludge concentration) To discern whether our organisms can survive and be effective in the sludge tank, we need to know whether they can survive in the concentration of TCC within the sludge tank and effectively break down this TCC and its constituent degradation products. Hence, we inoculated our bacteria in media containing a range of TCC concentrations that is generally present in the sludge (13000 - 21000ng\ml).

Test:

The growth curves obtained from all the experiments we performed for all three of our possible chassis’ have been listed below:

Learn:

---

MODULE 2

Design:

To carry out the wet lab experiments, including cloning, we started out by designing our part and plasmid.
Aim: To create a genetic circuit containing TccA amidase and insert it in E.coli to study its expression and characterise the enzyme.

Part Design

The purpose of the amidase protein (Part number: BBa_K4641000) in the chassis is to breakdown Triclocarban (3,4,4′-Trichlorocarbanilide) into 3,4-dichloroaniline and 4-chloroaniline. Our target protein is naturally constitutively expressed in Ochrobactrum sp. TCC-2

Plasmid Design

The CarbanEl system is based on a Gibson assembly cloning system. The gene fragment is designed and synthesized such that the overhangs to integrate the gene into pET22b+ are already added. For the characterization and study of the enzyme activity, we expressed the amidase gene in E. coli BL21 . However, our final intended chassis is Acinetobacter baylyi
The plasmid design is as below:

Genetic circuit design:

Gene regulation refers to how a particular gene expression can be controlled. It includes a variety of mechanisms that can be used to increase or decrease the expression or production of specific gene products.

Build:

The first step was to build our genetic circuit in E.coli DH5alpha to study the expression and degradation kinetics of our part.

We started with amplifying our tccA fragment and performing PCR clean-up. We isolated pET22b+ plasmid and proceeded with the Gibson Assembly to insert the TccA amidase gene. The competent E. Coli E.coli DH5alpha cells were prepared by CaCl₂ method and the cells were transformed.

After a confirmatory colony PCR to check the presence of our insert, we transformed the Gibson-assembled plasmid into E.coli BL21 and then we moved towards expression studies.

Test:

Cloning:
Double digestion was then performed on the isolated plasmid of the transformed cells to confirm the presence of the fragment which was successful. We then went on to expression studies. SDS-PAGE was carried out but distinct bands were not observed. This could have been due to the extended time for destaining.

Genetic circuit

The genetic circuit represents the regulation of TccA gene in E.coli. Our gene, including RBS, is 1562 bps long; assuming that the degradation of mRNA is negligible the time taken to transcribe our gene is 6.25 secs.

We modeled the protein production in E.coli . However, in the future, we can study and model the protein production rate in our preferred chassis Acinetobacter baylyi.

Design equation based on degradation kinetics

The rate expression is related to the rate at which the TccA amidase enzyme from the Ochrobactrum sp. can degrade TCC. Since our chassis for implementation is Acinetobacter baylyi, which is also gram-negative like Ochrobactrum sp.TCC-2, the rate of transport or the passive diffusion of TCC into the cell is assumed to be similar in both bacteria. The rate of degradation of the TCC inside the cell by the TccA amidase enzyme produced by our chassis will be very similar. The time calculated here for 21 PPM (the highest concentration of TCC found in sludge) will be used to model the bioreactor.

Learn:

On comparing our values to the literature, we get that the k values are approximately equal, thus validating our differential calculus method.

Hence, TccA amidase follows first-order kinetics.

Future cycles would involve the validation of these kinetics through experimentation.

---

MODULE 3

Summary

As mentioned in our implementation, we aim to introduce a bioreactor in the wastewater treatment plant to treat the primary sludge and clear it of our target contaminant, Triclocarban (TCC).

From the results of our previous module of the DBTL cycle, Acinetobacter baylyi GFJ2 was the most viable choice for a chassis because it had the best growth in high concentrations of TCC and a good survivability for the widely varying conditions of wastewater treatment plants. Additionally, the bacteria has the natural ability to degrade the toxic byproducts produced by the degradation of TCC into non toxic byproducts (cis-muconic acid) that can be taken by the Krebs cycle [3].

Design:

The team came up with multiple bioreactor designs for the implementation of our project that take into account the degradation kinetics of the enzyme, the growth kinetics of the bacteria, and the adsorption kinetics of TCC onto the packing material.

We have currently proposed a lab scale bioreactor due to the fact that we need to conduct further optimisation experiments to upscale into a bioreactor that is feasible on an industrial scale.

We wanted to design a sludge based bioreactor where the primary sludge becomes the feed. The primary sludge is almost 97-99% water [4]. We ideated this after a visit to a wastewater treatment plant in West Bengal, India.

We wanted the bacteria that degraded the TCC to be immobilized within a matrix. Literature suggests that cell immobilised systems are far superior to free cell suspensions as it avoids cell washout and improves reaction rates [5].

We found that biochar, a carbon-rich solid product produced from the pyrolysis of biomass residues, was an attractive option for an immobilisation matrix due to its low cost, easy availability and ability to immobilize cells efficiently[6].

Build:

Reactor

• The material of the reactors needs to be non-reactive to the compounds that will be present in the primary sludge that will enter the reactors. In response to this, we had a few options available to us, those being mild steel, stainless steel, glass, etc. Since our model is a lab-scale reactor, we chose to use glass to make the body of the lab-scale reactor.
• The inlet, outlet, and recycle pipes will be made of PVC.
• The stirrer would have a slow speed of 50 rpm.
• We would use generic industrial temperature and pressure gauges for the reactors.
• The sparger would be a perforated sparger for uniform aeration and would be placed at the bottom of all the reactor designs except for the continuous flow fluidized bed. The fluidized bed will have tube sparging with appropriate tubing to control the airflow for more aeration and, hence, more mass transfer efficiency.

Packing Material

• The packing material would consist of biochar. The biochar will be modified with a 1:1 ratio of biochar to KOH[8]. According to literature, the biochar modified with KOH showed higher adsorption of the target compound TCC compared to other ratios and just plain biochar.
• The biochar would be made into the shape of a Raschig ring using clay as a binding agent. The dimensions of the raschig ring are in the ratio 1:1 of the diameter to the height, and the inside diameter will be 1cm as shown in the figure below. These ideas were validated by meetings with Dr. Bhat and Dr. Kevin Sowers.
• The Raschig ring would then be fired at 110°C to 130°C to harden the ring so that it does not dissolve in the primary sludge.
• The genetically modified bacteria would then be immobilized on the biochar raschig rings.
• The biochar used will be sourced from a local rice mill by processing rice husk that is produced as waste at the mill.

Test:

Learn:

For a lab scale reactor we will use the batch packed bed reactor. The kinetics of enzyme degradation, TCC adsorption onto the biochar and the bacterial growth kinetics have been fitted into the equations for a batch reactor as given in the models page.

We learnt that design 1 would not be feasible due to low hydraulic retention time of the primary sludge in the tank, hence we tried design 2. However, through a human practices visit with Dr. Subbulaxmi we learnt that the efficiency of the biochar would be reduced significantly.

Design 3 significantly improved contact time; however, it would not be enough to compensate for the low absorption efficiency of the biochar after our other meetings with Dr. Murty.

Design 4 incorporated all the suggestions from Dr. Bhat and Dr. Subbulaxmi and led us to our most ideal design that gave us a high hydraulic retention time of the primary sludge in the tank as well as a high adsorption affinity of the biochar.

However, we learnt from Dr. Kevin Sowers that this model would only be feasible for a lab scale model. Since at industry level we would be required to treat a larger amount of primary sludge per day we proposed design 5 which would have comparable efficiency to design 4 and be able to treat large quantities of sludge due to its larger volume and continuous operation.

---

MODULE 4

Design:

Our goal was to design a detection system that would go hand in hand with our bioreactor set-up and was accurate, highly sensitive, and cost-effective. In our design stage, we have explored numerous ideas, highlights of which include:

• Electrochemical sensors - were explored, as they are one of the most prevalent detection systems. However, we found that it would not be feasible to achieve the desired selectivity towards triclocarban.
• Enzyme based sensors- we identified an enzyme from Ochrobactrum sp.TCC-2; but due to the trace levels of TCC found in effluent this option would not be sensitive enough.
• Aptasensors- Aptamers specific for TCC are not available in the literature, due to which, we would be required to isolate and identify one de-novo using SELEX. This is expensive and would not be feasible within our time constraint.
• Nanobody based sensors- three nanobodies which were specific to TCC were identified. We realised that immobilizing the nanobody on the sensor and making required modifications would be expensive and within the given time frame, building a prototype and testing didn’t seem like a viable option.
• Molecularly imprinted polymer based sensors(MIP)- we found an MIP which would detect triclocarban and triclosan combined. Making an MIP specific to triclocarban in the lab would require extensive experimental data and testing.
• Nanomaterial based sensors- Nanomaterial based sensors can detect endocrine disrupting compounds and since TCC is one of them, this drew our attention. However, after thorough literature review we found that there was no nanomaterial isolated specifically for TCC detection.

Ideation Page

Toward the end, we decided to simplify our approach by using Raman Spectroscopy to identify light of a certain wavelength range specific to TCC and design a simplified prototype of an optical detection system using a laser and a photodiode to detect this range of wavelengths.

Build:

We used the Raman spectroscopy setup in our secondary PI, Kapil Sadani’s laboratory to conduct experiments and isolate the specific peaks for our compound. Using the data from these experiments, we calculated the specific wavelength for the IR filter and photodiode and built a theoretical set-up that can be employed for the detection of TCC.

This set-up would include a laser of 785 nanometer monochromatic light, two focal lenses, a mirror with a hole to cleanly pass the focused laser, a sample stage (cuvette), an IR filter, a photodiode, and a microcontroller. The sample would have to be manually collected and brought to this set-up.

Test:

We ran experiments for various samples as follows :

1. Raman spectroscopic analysis for Triclocarban dissolved in 5% acetone.
2. Surface Enhanced Raman Spectroscopy (SERS) for Triclocarban dissolved in 5% acetone.
3. Lake water sample, that was suspected to have triclocarban dissolved in it, to baseline and isolate the specific peaks for TCC using SERS.

From this we found the specific wavelength for our Infrared filter and photodiode.

Hardware Notebook

• Experiment 1: 1, 10, 100 ppm (parts per million) of TCC in 5% acetone (standard) Figure: Surface Enhanced Raman Spectroscopy of triclocarban dissolved in 5% acetone. (1, 10, 100 ppm)
• Experiment 2: 1, 10, 100 ppm (parts per million)of TCC in lake water sample Figure: Surface Enhanced Raman Spectroscopy of triclocarban dissolved in lake water. (1, 10, 100 ppm)

Learn:

From the experimental data, we learned that the unique peaks (unique wavenumber) for Triclocarban is observed at 758.3 cm^-1and 1445.2cm^-1. Moreover, the second peak observed at 1445.2 cm-1 can be concluded as an overtone of the peak observed at 758.3 cm^-1 as the ratio of the intensities at various concentrations remains constant.

In the lake water sample, we noticed peaks within the range of 750 and 760 nanometers. We can assume that the lake water sample initially had dissolved triclocarban. After experiments with spiked triclocarban were done, we noticed an amplification in the baseline peaks.

These preliminary results are indicative of a starting point for proving our proof of concept of a detection system selective and sensitive towards TCC and easily adaptable in all kinds of bioreactor setups.

Success:

• Acinetobacter baylyi was identified as the most viable chassis for the incorporation of TccA amidase
• A successful clone of TccA in pET22b+ was obtained in E.coli DH5alpha and then transformed into E.coli BL21 .
• We successfully found the wavelength of the IR filter which can be used in our detection system set up to detect TCC in sludge
• A packed bed batch reactor was found to be the most viable option for a lab scale bioreactor while a fluidised bed reactor was preferred for future implementation.

References:

[1]Yun H, Liang B, Qiu J, Zhang L, Zhao Y, Jiang J, Wang A. Functional Characterization of a Novel Amidase Involved in Biotransformation of Triclocarban and its Dehalogenated Congeners in Ochrobactrum sp. TCC-2. Environ Sci Technol. 2017 Jan 3;51(1):291-300. https://doi.org/10.1021/acs.est.6b04885. Epub 2016 Dec 14. PMID: 27966913.

[2]Hülter, N., Sørum, V., Borch-Pedersen, K. et al. Costs and benefits of natural transformation in Acinetobacter baylyi . BMC Microbiol 17, 34 (2017). https://doi.org/10.1186/s12866-017-0953-2

[3]Hongsawat, Parnuch, and Alisa S. Vangnai. “Biodegradation Pathways of Chloroanilines by Acinetobacter Baylyi Strain GFJ2.” Journal of Hazardous Materials, vol. 186, no. 2–3, Elsevier BV, Feb. 2011, pp. 1300–07. https://doi.org/10.1016/j.jhazmat.2010.12.002.

[4] Nazari, L., Sarathy, S., Santoro, D., Ho, D., Ray, M. B., & Xu, C. (2018). Recent advances in energy recovery from wastewater sludge. In Elsevier eBooks (pp. 67–100). https://doi.org/10.1016/b978-0-08-101029-7.00011-4

[5] Zhu, Y. (2007). Immobilized cell fermentation for production of chemicals and fuels. In Elsevier eBooks (pp. 373–396). https://doi.org/10.1016/b978-044452114-9/50015-3

[6]Wu, C., Zhi, D., Yao, B., Zhou, Y., Yang, Y., & Zhou, Y. (2022). Immobilization of microbes on biochar for water and soil remediation: A review. Environmental Research, 212, 113226. https://doi.org/10.1016/j.envres.2022.113226

[7] Eggen, T., & Vogelsang, C. (2015). Occurrence and fate of pharmaceuticals and personal care products in wastewater. In Comprehensive Analytical Chemistry (pp. 245–294). https://doi.org/10.1016/b978-0-444-63299-9.00007-7