iGEM REPORT: Mycobacterium Revelio: Characterizing and Modeling Genetic Circuit Components towards a Bacterial Detection Tool

Note for Reviewers

This work has been submitted for open review as an iGEM Report.

Reviewers, please consider the following questions when reviewing the report:

1. Does the submission accurately and clearly report the experiment carried out?

2. Does the report adhere to the PLOS Data Policy?

3. Is the report presented correctly and well written?

See iGEM Reviewer Instructions for more information.

Please leave your review as a comment below the Report.

---

Mycobacterium Revelio: Characterizing and Modeling Genetic Circuit Components towards a Bacterial Detection Tool

Snehal Kadam (1), Yash Jawale (1), Swapnil Bodkhe (1), Prashant Uniyal (1), Ira Phadke (1), Prachiti Moghe (1), Harsh Gakhare (1), Siddhesh Zadey (1), Gayatri Mundhe (1), Rahul Biradar (1), Manasi S. Gangan (2), Neha Khetan (2), Chaitanya A. Athale (2) *

1. BS-MS Dual Degree Programme, IISER Pune, Dr. Homi Bhabha Road, Pashan, Pune 411008, India.
2. Div. of Biology, IISER Pune, Dr. Homi Bhabha Road, Pashan, Pune 411008, India.

* Corresponding Author – Chaitanya A. Athale: Tel: +91-20-2590-8050; Email: cathale@iiserpune.ac.in

Author contributions

Conceptualization: Chaitanya A. Athale

Experimental investigation: Snehal Kadam, Prashant Uniyal, Prachiti Moghe, Ira Phadke, Swapnil Bodkhe, Yash Jawale, Gayatri Mundhe, Rahul Biradar and Manasi S. Gangan.

Simulation investigation: Harsh Gakhare, Siddhesh Zadey and Neha Khetan

Writing: Snehal Kadam and Chaitanya A. Athale

Supervision: Neha Khetan, Manasi S. Gangan and Chaitanya A. Athale

Funding acquisition: Chaitanya A. Athale

---

Abstract

With the aim of developing a novel diagnostic method for Mycobacterium tuberculosis detection in an in vitro setting, we have tested components a synthetic genetic circuit, Mycobacterium Revelio. This hypothetical circuit will need three components: an ‘oscillator’ based on a hybrid promoter, a pigment-based ‘detection’ and a population growth ‘terminator’ module. All modules were made in Escherichia coli. The ‘oscillator’ module is based on the pLac/Ara-1 synthetic hybrid promoter with dual inputs- lactose and arabinose. To our surprise, a construct of this AND gate pLac/Ara-1 promoter (BBa_K094120) with a GFP reporter (BBa_I13504) downstream, was unresponsive to IPTG and arabinose. However, when the promoter was replaced by the original sequence of the pLac/Ara-1 described by Lutz and Bujard [1]- a difference of 2 base-pairs- the circuit became IPTG and Arabinose responsive. The ‘detection’ module was based on lycopene and b-carotene expression cassettes. Confirming previous reports, we found the copy-number inducible lycopene (BBa_K274120) and b-carotene (BBa_K274220) constructs in an E. coli MG1655 background resulted in prominent colored cell-pellets after ~20 hours of induction. For the ‘terminator’ module, we assembled a new BioBrick part with an IPTG inducible promoter (BBa_R0011) driving luxI (BBa_K805016) expression. On induction the ‘sender’ produces N-acyl homoserine lactone (AHL). E. coli cultures separately transformed with the ‘sender’ and a ‘receiver’ (BBa_T9002) circuit and co-cultured, resulted in GFP expression, demonstrating the functioning of the module. A mathematical model of intrinsic population-control with ‘sender’ and ‘receiver’ circuits and a receiver driven toxin expression [2], was used to demonstrate a delayed luxI induction results in better pigment accumulation. In summary, further testing of components is required towards realizing a Mycobacterium Revelio circuit.

Financial Disclosure

The project was funded by the grant BT/EB/iGEM/2015 from the Department of Biotechnology, Government of India to CAA and core-funding from the Indian Institute of Science Education and Research (IISER) Pune. Additional sponsorship funds were obtained from Qiagen India, Eppendorf India, Merck Life Sciences Pvt. Ltd., India and Persistent Systems Pvt. Ltd., Pune, India. IDT USA provided free G-block DNA synthesis and Mathworks Inc., USA provided free licenses for the iGEM team.

Competing Interests

The authors declare that no competing interests exist.

Ethics Statement

N/A

Data Availability

All data in support of the work described in this document are present in the form of tables, figures and methods and are provided without any restriction. iGEM wiki: http://2015.igem.org/Team:IISER_Pune

 

Introduction

Detection of microbes in clinical samples is now routinely performed using quantitative PCR based methods. However, despite the advances, phenotypic growth assays with specific media and strain-specific stains, remain a standard in clinical laboratories, both due to their convenience and reliability [3,4]. However the phenotypic growth tests of slow-growing organisms such as Mycobacterium tuberculosis require up to a week. Our work was motivated by the hypothesis that an engineered genetic circuit could be used to potentially accelerate the population growth of such slow growing bacteria in vitro using an ‘oscillator’, generate color readout of the growth for ‘detection’ and control the population at a desired maximal population size by a ‘terminator’.

Multiple synthetic genetic oscillators have been developed in the recent years and tested experimentally mostly in prokaryotes with periods of oscillations ranging between 13 minutes to 26 hours [5]. Of these, a tunable oscillator consisting of just two-genes, lacI and araC, which has been demonstrated to work in E. coli [6] whose expression is controlled by a pLac/Ara-1 hybrid promoter [1] appeared to be the best choice. In order to build such an oscillator, constructing a synthetic AND gate responsive to two inputs- IPTG and arabinose- was the first step.

The detection of microorganisms usually requires a microscope. However, an easier method of detection would be to visualize a colored pigment that the bacteria can produce. In large enough numbers, these bacteria can produce enough pigment that can be seen by the naked eye as seen in the E. chromi project to produce carotenoids [7], the ‘colloroid’ production [8] and luminescence circuits based on the lux genes [9]. The carotenoid pathway to produce lycopene and b-carotene [10-12] was used as a test, due to its ready availability and previous proof of principle. E. coli lack few genes from non-mevalonate isoprenoid pathway and crtEBI and crtEBIY are required in E. coli to produce lycopene and b-carotene respectively.

For population growth, we chose to base our strategy for the ‘terminator’ design on a ccdB toxin gene expression circuit [13], which is triggered at a specific cell density. To measure the cell density, this ‘killer’ gene has been coupled to a minimal luxI–luxR ‘quorum-sensing’ circuit, and a population of cells expressing it undergo self-limited growth [2]. The ‘quorum-sensing’ mechanism of bacterial cell-to-cell communication by small molecule diffusible autoinducers has been very well studied in the marine luminous bacteria called Vibrio fischeri [14-16]. The minimal components of this operon are the autoinducer N-acyl homoserine lactone (AHL), the luxI gene, pluxI and pluxR promoters and the regulatory gene luxR. The luxI gene encodes a synthase that produces AHL. At high cell density, sufficient AHL enters the cell and forms an activated complex with the LuxR protein. This complex binds to the pluxR promoter, allowing for the transcription of downstream genes in a cell density (quorum) dependent manner. In previous work, the gene downstream of pluxR was the toxin producing ccdB gene, resulting in growth inhibition.

Theoretical models of the population growth dynamics with the population control module [2] have assumed luxI expression and AHL production can be lumped into one rate constant, while the uptake of AHL is not rate limiting. Despite these apparent simplifications, the model successfully reproduced experimental data of population growth inhibition. We chose to modify this model to examine the best strategy to obtain maximal signal from the color pigment.

Here, we have begun with preliminary testing of the components that we believe would be necessary to build a Mycobacterium Revelio circuit. The components tested in the work here include the core of a tunable oscillator the pLac-Ara-1 hybrid promoter, the crtEBI and crtEBIY expression system to produce carotenoid pigments lycopene and b-carotene and a basic luxI based ‘sender’ and a luxR based GFP reporter ‘receiver’ circuit. We also attempt to model the population dynamics of a self-limiting population with the additional factor of pigment production.

 

Materials and Methods

Bacterial strains and plasmids: The cloning of plasmids and their maintenance was done in Escherichia coli MG1655 and DH5a strains. The plasmids used are listed in Table 1 for the different DNA sequences used. All plasmids were isolated using either a Qiagen Spin Miniprep Kit (Qiagen GmbH, Germany) or the alkaline lysis method [17].

Molecular biology: All cloning was performed using the standard BBa1.0 cloning methods employing 3A assembly (http://parts.igem.org/Help:Protocols/3A_Assembly) where incubation of the sample with the restriction enzyme was varied between 3-12 hours and ligation with T4-DNA Ligase (Invitrogen) at 25°C for 3-12 hrs. Competent cells were prepared using the modified Hanahan method [18] as described in the iGEM protocols (http://parts.igem.org/Help:Protocols/Competent_Cells) and transformed by the heat-shock method [17]. Constructs were examined for by a single-enzyme digest by separating the sample on 1% agarose gels (Sigma-Aldrich, India) containing 0.4 mg/ml Ethidium Bromide (Sigma-Aldrich, India) in a submarine gel electrophoresis apparatus (Broviga, India) at 100 mV constant-voltage (BioRad, USA) and imaged in a UV-transilluminator (G-box, Syngene, India). Plasmid constructs were documented and sequence comparisons were made using APE, a plasmid editor software (http://ape-a-plasmid-editor.wikispaces.com/).

Fluorimetry and absorbance measurements: GFP fluorescence of cells was measured using fluorimetry in a 96-well black-plate (Corning, USA) in a multi-mode plate reader (Varioskan Flash, Thermo Scientific, USA) with an excitation wavelength of 480 nm and emission measured at 510 nm.

Samples were prepared for measuring the hybrid promoter response by inoculated Luria-Bertani (LB) broth with 1:3000 from an overnight culture, grown at 37°C, 2 hours with shaking (200 rpm) and induced with IPTG and arabinose. 100 μl samples were pipetted into the 96-well plate and measured. Uninduced cells were used as a negative control and pGFP (Clontech, USA) expressing cells as a positive control. For the quorum-sensing constructs, the ‘sender’ (BBa_R0011+BBa_K805016 in pSB1K3 backbone) and ‘receiver’ (BBa_T9002 in pSB1C3 backbone) constructs were grown separately overnight and a 5 ml 1:1 (volume: volume) co-culture in LB was incubated at 37°C for 2 hours and then induced with 1mM IPTG. Induced cultures were incubated at 37°C for 2 hours after which they were pipetted into a 96-well black-plate (Corning, USA) and fluorescence measured. Individual cultures of the sender and receiver as well as uninduced cultures were used as controls. The pGFP plasmid (Clontech, USA) and BBa_I20270, a promoter with ribosome binding site (RBS) + GFP reporter, were used as positive controls and BBa_R0040 (tetR repressible promoter) as negative control.

Absorbance over a wavelength range 400-600 nm was measured on the crtEBI (lycopene) and crtEBIY (b-carotene) constructs transformed into E. coli MG1655 using 96 well half-well plates (Costar, Corning, USA). Samples were prepared by growing a 5% inoculum in LB from overnight cultures with the 1 mM IPTG and 5% arabinose added as inducers for 20 hours at 37°C with shaking (200 rpm). The cell pellet was resuspended in 300 ml acetone followed by probe sonication for 20 seconds with 50% amplitude.

Microscopy: E. coli MG1655 cells transformed with the quorum sensing plasmids (sender and receiver) were grown in LB for 2 hours at 37°C with shaking and induced with 1mM IPTG for 2 hrs and fixed using 4% paraformaldehyde and 1X PBS. The slide was prepared for observation in microscopy by spreading 5 μl of the bacterial suspension, dried and mountant (20mM Tris pH 8, 0.5% N- propyl gallate, 90% Glycerol, all from Sigma-Aldrich, India) added and a 1 cm coverslip placed on top. Samples were imaged in DIC and fluorescence (GFP filter cube) Zeiss Axio Imager Z1 (Carl Zeiss, Germany) using a 40x (N.A. 0.75, EC Plan-Neofluar) lens. Images were acquired using the AxioVision (ver. 4.8) software.

Simulations: A series of coupled ordinary differential equations (ODEs) were implemented in Python (ver. 2.7.6) with the SciPy and NumPy modules integrated in the IDE Canopy (ver. 1.4.0, 64 bit). The odeint solver was used to integrate the ODEs. The code was tested on a Linux and Mac OSX (10.11.3) systems. On a MacBook Pro with a single 2.7 GHz Intel Core i7 processor with two cores, a single calculation of total time 240 hours required ~0.5 minutes to run.

 

Results

Hybrid promoter expression abolished by one basepair deletion and substitution

Two constructs containing a p_Lac/Ara-1 hybrid promoter controlling the expression of GFP were cloned using 3A assembly. Construct A consisted of the p_Lac/Ara-1 promoter deposited in the iGEM repository (BBa_K094120) with the RBS-GFP-Terminator (BBa_I13504) in a pSB1C3 backbone (Fig 1A), while in Construct B the promoter was replaced by a synthetic DNA fragment of the promoter (Fig 1B), based on the exact sequence described by Lutz and Bujard [1] (Table 1). The alignment of the two promoters using the plasmid sequence editor APE showed the BioBrick promoter (BBa_ K094120) had a deleted C (at 55 bps) and a G->C (at 58 bps) substitution, when compared to the original sequence reported by Lutz and Bujard (Fig. 1C). All cells were inoculated and grown for the same period of time at 37C with shaking and the fluorescence measured. Uninduced cells with both constructs A and B had lower fluorescence, than pGFP (Fig 1D). When a combination of both inputs IPTG (2-20 mM) and arabinose (0.1-1% w/v) was added to the cultures with the hybrid-promoter and reporters and fluorescence measured after 2 hours. Across the range of inducer concentrations, Construct A showed fluorescence in the range of the uninduced sample, while the fluorescence in Construct B was ~8 times that of uninduced samples. The low level of the positive control pGFP is attributed to its uninduced state. Thus, we find the p_Lac/Ara-1 promoter with the original sequence functions as expected, while the sequence deposited in the BioBricks collection (BBa_K094120) with a two basepair difference in sequence- a single basepair deletion and a single substitution- does not function as a hybrid promoter. For further work on the oscillator, this unmodified sequence will need to be used.

Carotenoid production and measurement

The carotenoid pigment synthesis construct for lycopene (Fig 2A) and b-carotene (Fig 2B) were chosen based on previous reports of expression by the iGEM 2009 Cambridge team (http://2009.igem.org/Team:Cambridge). E.coli MG1655 cells were transformed with two kinds of lycopene (crtEBI) and b-carotene (crtEBIY) producing, constructs- one set in the regular pSB1C3 backbone (http://parts.igem.org/Part:pSB1C3) and the other set in a pSB2K3 backbone (http://parts.igem.org/Part:pSB2K3) (Table 1). Cultures were grown for ~20 hours with either 1 mM IPTG or 1 mM IPTG+5% arabinose for the pSB1C3 and pSB2K3 constructs respectively, with inducers added at the time of inoculation. When the culture was centrifuged, visibly colored pellets were seen with the pSB2K3 backbone lycopene pellet distinctly red and the pSB2K3 b-carotene pellet orange (Fig 2C). The color of the pellet in the pSB1C3 backbone with crtEBI and crtEBIY are indistinguishable. Absorbance measurement of the sonicated E. coli cells with the crtEBI construct (lycopene) showed multiple peaks at 450, 480 and 510 nm while the crtEBIY construct (b-carotene) showed a peak at­ 460 nm and broad shoulder between 460 and 500 nm (Fig 2D), corresponding to previous reports [7,19]. While the visible pellet and absorbance appeared to demonstrate that the pigment accumulates as expected, the ~20 hours of growth required for pigment accumulation suggest a need to use a faster developing pigment.

Quorum sensing module in co-culture experiments

Based on a previously developed synthetic circuit which used the quorum sensing machinery coupled to a toxin to limit growth in E. coli [2], we attempted to reproduce the behavior of a similar circuit. As proof of principle, we replaced the ‘killer’ gene, ccdB as described in the previous work, with a GFP reporter. The circuit consisted of a ‘sender’ and ‘receiver’ plasmid. The ‘receiver’ circuit consisted of a luxR-regulated GFP reporter and an on-board luxR gene (Fig 3A). To build the ‘sender’ circuit, we cloned a lac regulated hybrid promoter pLac + lambda pL hybrid promoter inducible by IPTG (BBa_R0011) upstream of luxI coding gene (BBa_K805016) to form the complete ‘sender’ circuit (Fig 3B). ‘Sender’ (BBa_K1780016 in pSB1K3 backbone) and ‘receiver’ (BBa_T9002 in pSB1C3 backbone) circuits transformed in two different E. coli cultures were grown in LB with their respective antibiotics (Table 1) to saturation and then inoculated at 1% in LB – either separately ‘sender’, ‘receiver’ or as 1:1 co-cultures- and grown at 37°C with shaking. When the cultures reached O.D. 0.3, inducer (1 mM ITPG) was added to samples and fluorescence measured after all cultures reached saturation (fixed time-point) in relative fluorescence units (rfu) (Fig 3C). The fluorescence in sender+receiver co-cultures is marginally higher in induced (+) compared to uninduced (-) cultures. Additionally, co-cultured show 2-3 fold higher fluorescence compared to isolated cells of sender and receiver alone (Fig 3C). The expression of the constitutive reporter is as expected higher than the induced co-culture, and the negative control fluorescence levels are similar to the separate circuits (sender and receiver alone). In quantitative florescence microscopy of single cells, where the same exposure time and camera settings were used, of the induced (+) ‘sender’+‘reciever’ co-culture, some individual cells show high-intensity GFP signal. These are likely to be receiver cells with AHL-based GFP expression induction (Fig 3D). The low-level expression of GFP in ‘reciever’ cells alone can be attributed to the ‘leaky’ expression of the luxpR promoter, while ‘sender’ cells alone appear to have background fluorescence. DIC images of all three conditions suggest the number of cells are roughly comparable. Further population growth experiments with fluorescence could establish if the population control circuit is functioning as designed.

Mathematical model of inducible growth-termination to maximize pigment concentration

In order to predict the growth dynamics of the population control circuit and its connection with pigment production, we have modified an existing mathematical model of AHL inducible ccdB expression resulting in ‘quorum-sensing’ based population control, developed previously by You et al. [2]. Similar to the previous model, the cell population (N) evolves by a modified logistic growth as a function of time (t) based on the population growth rate (k_N) and the carrying capacity or maximal population size (N_m) as follows:

The reduction in growth rate of the population is determined by the death rate (d_N) in the presence of the inhibitor (I). The inhibitor production is regulated by a pLuxI promoter, which is produced when LuxR binds to AHL. As previously described, a simple lumped first-order rate constant of inhibitor production (k_I), which depends only on the concentration of AHL (A) serves to capture the dynamics. Combined with the degradation rate of the inhibitor (d_I), the dynamics of inhibitor is given by:

The quorum sensing molecule AHL concentration (A) is considered the same outside and inside the cells, i.e. we ignore the rate at which it is released from a cell, diffuses in the medium and is taken up into neighboring cells. The population density sensing circuit of luxI gene expression and AHL synthesis has been modeled by a simple first-order rate constant approach to AHL in previous work and shown to be sufficient to capture the dynamics of the system [2]. In a departure from the previous model, we also assume AHL production to be inducible (the luxI gene is under the control of an inducible promoter such as pLac). The production of A is then modeled to be in two states: an uninduced low-level expression governed by a basal production rate (k_A) and a leaky expression factor (l_A) of luxI, and a high-level expression state where luxI is induced and results in production of AHL at a standard level. We have modeled the trigger for luxI induction to be a pigment concentration threshold (P_?), but in case of absence of detected cells, this can be a finite time threshold. Thus AHL concentration is determined by:

where d_A is the rate of AHL degradation per cell. This modification was driven by the motivation of achieving as high a pigment accumulation as possible (in case of cell growth), before inducing the inhibitory-circuit, since after that pigment production will reduce with reducing N. Thus, pigment concentration (P) is allowed to accumulate to a level sufficient to enable detection. The pigment concentration (P) dynamics is simply modeled by:

where k_P is the rate of pigment formation per cell.

Simulations were performed with parameters taken from the previously described model (Table 2). Consistent with previous reports, the simulated induction of luxI gene expression triggers AHL-dependent production of the ccdB encoded inhibitor (I) and reduces the population size, as compared to a system where no such control exists (Fig 4A). After the induction of AHL, the pigment appears to accumulate at a slower rate corresponding to the time at which the population goes into decline (Fig 4B). This is also seen in the initial spike in AHL production, at the time of induction, which also leads to a spike in inhibitor concentration (Fig 4C). The inducible luxI circuit is triggered in our simulations at a threshold concentration of pigment (P_?), which modifies the previous model of AHL induction (Fig 4D). While the choice of the threshold concentration of P at which the population-control module is induced, is currently left to the user, it could be based on time-threshold instead.

 

Discussion

In this report, we have tested the hybrid lac-ara promoter behavior as a basic component in a tunable ‘oscillator’, tested the production of lycopene and b-carotene from the crtEBI(Y) genes for ‘detection’ and developed a ‘sender’ construct for testing with a GFP-reporter based ‘receiver’ construct for eventual use for bacterial population control (‘terminator’).

The behavior of hybrid promoters in response to the multiple inputs is important both for tight regulation of gene expression, as well as designing genetic logic gates [13]. The lactose-arabinose hybrid promoter had been described in a study comparing multiple hybrid promoter constructs [1] and used most prominently in a tunable genetic oscillator [6]. In the process of protyping parts for the ‘oscillator’ module, our tests of the BioBrick deposited in iGEM demonstrates that even 2 base-pair differences in a ~140 bps sequence can cause functional deficiencies in the genetic-device. This further emphasizes the value of component testing and reproducibility studies, especially with respect to synthetic gene constructs [20]. The quantitative dose-response to IPTG and arabinose has been described by Stricker et al. [6] with a 2 plasmid system. We would aim in future to integrate this onto a single-plasmid oscillator system. Such an oscillator could then be coupled to bacterial cell cycle proteins such as DnaA.

For the ‘detection’ module, we had aimed to test if the crtEBI and crtEBIY expression cassettes for lycopene and b-carotene production respectively could serve as viable cell-population readouts as a diagnostic tool. However both our tests and those of the E. chromi team [7] suggest that the ~20 hour growth time to see prominent color development might be too long for the purpose of a rapid diagnostic test. In future chromoproteins such as those isolated from Acropora millepora (amilCP), which has been demonstrated to express in E. coli as blue, yellow and red color variants (http://parts.igem.org/Part:BBa_K592025), could also be tested as viable alternative color markers. Our attempts at alternative pigment reporters such as cloning a constitutive promoter-(BBa_J23110) with the RBS+amilCP construct (BBa_K592025), and another constitutive promoter (BBa_K1509003) with lacZ (BBa_K823016) failed due to technical reasons. However, as with the experience of previous teams such as iGEM16_Glasgow (http://2015.igem.org/Team:Glasgow), without final testing in the target organism, there appears no prior way to determine if the chromophore will serve as a robust color-marker of growth.

The ‘terminator’ module was tested in a preliminary manner with the ‘sender’ construct with a lactose inducible luxI transformed in one cell population, and a separate population of cells containing the ‘receiver’ plasmid with luxR and an AHL responsive luxpR driven GFP reporter. Our tests indicate that with and without induction of luxI, the co-culture of ‘sender’ and ‘receiver’ cultures, produce GFP expression. This would suggest that while the leaky expression of luxI is sufficient to produce AHL in the medium, a tighter regulation could allow more regulated behavior of the circuits, such as by driving luxI expression with an arabinose inducible promoter. Additionally, it remains to be seen in population growth experiments, if our system can reproduce the effects observed by [2] of a population growth dependence of reporter expression. This characterization however confirms that the parts function as expected. For the ‘terminator’ circuit, both a ccdB killer gene and merging both circuits ‘sender’ and ‘receiver’ onto one plasmid would be necessary. Additionally, in the larger context of engineering rapid growth of potentially pathogenic bacteria would require more safety features, to avoid ‘super-bug’ scenarios. These could involve both physical containment features in the ‘device’ as well as multiple redundant population control measures, in addition to the ccdB and ‘quorum-sensing’ circuit.

The theoretical model is based directly on the population control circuit based on the ‘quorum-sensing’ mechanism described previously [2]. The addition of a pigment production equation however allows us to examine an ideal strategy to induce population control, while optimizing the pigment (reporter) production. Confirming our intuition, a later induction of the ‘killer’ circuit allows more pigment to accumulate as compared to inducing the luxI ‘sender’ circuit from the beginning of growth. Additional modifications would need to be made to the model if a different dye were to be used instead of the slow-developing carotenoid pathway.

In conclusion, we have found a 2-basepair difference in the pLac-Ara-1 promoter can result in malfunctioning of the genetic-component. We have also reproduced the result of carotenoid production and produced a new part for a ‘sender’ circuit, so it can be tested with the ‘receiver’ for a quorum sensing based population control. In addition, our model suggests some modifications to the induction strategy in population control to optimize the pigment formation for detection. Taken together, our work serves as a test-bed for many of the sub-components involved in developing the Mycobacterium Revelio diagnostic circuit.

 

Acknowledgements

We are grateful to Dr. Barbara Di Ventura and the University of Heidelberg iGEM2015 team for mentorship. Dr. Nikhil Phadke of GenePath Dx Labs, Pune and Dr. Vikram Padbidri of Jehangir Hospital, Pune kindly hosted lab-visits for a first-hand view of clinical microbial diagnostics. We thank Dr. P. Gayathri and Dr. Mugdha Gadgil for their primer lectures on gene-regulation and synthetic biology respectively to the team.

 

References

1. Lutz R, Bujard H (1997) Independent and tight regulation of transcriptional units in escherichia coli via the LacR/O, the TetR/O and AraC/I1-I2 regulatory elements. Nucleic Acids Res 25: 1203–1210. doi:10.1093/nar/25.6.1203.
2. You L, Cox RS, Weiss R, Arnold FH (2004) Programmed population control by cell-cell communication and regulated killing. Nature 428: 868–871. Available: http://www.ncbi.nlm.nih.gov/pubmed/15064770.
3. Fluit ADC, Visser MR, Schmitz F (2001) Molecular Detection of Antimicrobial Resistance. Clin Microbiol Rev 14: 836–871. doi:10.1128/CMR.14.4.836.
4. Brodin P, Christophe T (2011) High-content screening in infectious diseases. Curr Opin Chem Biol 15: 534–539. Available: http://linkinghub.elsevier.com/retrieve/pii/S1367593111000925.
5. Purcell O, Savery NJ, Grierson CS, di Bernardo M (2010) A comparative analysis of synthetic genetic oscillators. J R Soc Interface 7: 1503–1524. doi:10.1098/rsif.2010.0183.
6. Stricker J, Cookson S, Bennett MR, Mather WH, Tsimring LS, et al. (2008) A fast, robust and tunable synthetic gene oscillator. Nature 456: 516–519. Available: http://www.nature.com/doifinder/10.1038/nature07389.
7. iGEM 2009 Team Cambridge (2009) E. chromi Cambridge. iGEM Wiki. Available: http://2009.igem.org/Team:Cambridge/Project/Carotenoids.
8. iGEM 2011 Uppsala-Sweden (2011). iGEM Wiki. Available: http://2011.igem.org/Team:Uppsala-Sweden.
9. iGEM 2015 Glasgow (2015). iGEM Wiki. Available: http://2015.igem.org/Team:Glasgow/.
10. Tao L, Jackson RE, Rouviere PE, Cheng Q (2005) Isolation of chromosomal mutations that affect carotenoid production in Escherichia coli: mutations alter copy number of ColE1-type plasmids. FEMS Microbiol Lett 243: 227–233.
11. Misawa N, Nakagawa M, Kobayashi K, Yamano S, Izawa Y, et al. (1990) Elucidation of the Erwinia uredovora carotenoid biosynthetic pathway by functional analysis of gene products expressed in Escherichia coli. J Bacteriol 172: 6704–6712.
12. Yuan LZ, Rouviere PE, Larossa RA, Suh W (2006) {C}hromosomal promoter replacement of the isoprenoid pathway for enhancing carotenoid production in {E}. coli. Metab Eng 8: 79–90.
13. Brophy JAN, Voigt CA (2014) Principles of genetic circuit design. Nat Methods 11: 508–520. Available: http://www.nature.com/doifinder/10.1038/nmeth.2926.
14. Bassler BL (1999) {H}ow bacteria talk to each other: regulation of gene expression by quorum sensing. Curr Opin Microbiol 2: 582–587.
15. Surette MG, Miller MB, Bassler BL (1999) {Q}uorum sensing in {E}scherichia coli, {S}almonella typhimurium, and {V}ibrio harveyi: a new family of genes responsible for autoinducer production. Proc Natl Acad Sci USA 96: 1639–1644.
16. Bassler BL, Greenberg EP, Stevens AM (1997) {C}ross-species induction of luminescence in the quorum-sensing bacterium {V}ibrio harveyi. J Bacteriol 179: 4043–4045.
17. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual. book. Cold Spring Harbor laboratory Press.
18. Hanahan D, Jessee J, Bloom FR (1991) {P}lasmid transformation of {E}scherichia coli and other bacteria. Meth Enzym 204: 63–113.
19. Hyman JR, Gaus J, Foolad MR (2004) A Rapid and Accurate Method for Estimating Tomato Lycopene Content by Measuring Chromaticity Values of Fruit Purée. J Amer Soc Hort Sci 129: 717–723.
20. Beal J, Haddock-Angelli T, Gershater M, Mora K De, Lizarazo M, et al. (2016) Reproducibility of Fluorescent Expression from Engineered Biological Constructs in E. coli . PLoS One 11: e0150182. doi:10.1371/journal.pone.0150182.