BACKGROUND

Carotenoids are organic pigments naturally present in plants, algae and some bacteria. There are more than 600 carotenoids, which can be categorized into xanthophylls (hydrocarbons containing oxygen) and carotenes (hydrocarbons containing no oxygen). Carotenoids perform a range of functions, including light energy absorption, protection against photo-damage, acting as antioxidants, and as precursor to other organic compounds. In human, for example, beta-carotene is the precursor to vitamin A.

Biochemical pathway of carotenoid synthesis

The common starting point for carotenoid synthesis is farnesyl pyrophosphate (FPP), which derives from two precursors, isopentenyl pyrophosphate (IPP) and dimethylallyl pyrophosphate (DMAPP). In general, there are two pathways for synthesising IPP and DMAPP: the Mevalonate Pathway (starting with acetyl CoA) and the Non-mevalonate Pathway (starting with pyruvate and glyceradehyde-3-phosphate). While the Mevalonate Pathway is present in all higher eukaryotes, the Non-mevalonate Pathway is present in E. coli.

Mevalonate Pathway (image adopted from Wikipedia, http://en.wikipedia.org/wiki/Mevalonate_pathway)

Non-mevalonate Pathway (image adopted from Yuan et al. 2006)

After IPP and DMAPP go on to form FPP, a series of enzymatic reactions convert the colourless FPP to coloured compounds: red lycopene, and orange β-carotene.

By modulating the enzymes involved in this conversion pathway ("CrtE", "CrtB", "CrtI", and "CrtY"), we aim to produce and control different pigments as signal output.

DESIGN OF DEVICES

Biobricks contributed by previous iGEM teams

A few iGEM teams of previous years (e.g. Edinburgh 07, Edinburgh 08, Guelph 08) conducted research in the carotenoid system. We are fortunate to have access to some of their Biobricks in the registry:

Registry Code	Team	Sequence Description	Size of insert
K118014	Edinburgh 08	(rbs + CrtE)	922bp
K118006	Edinburgh 08	(rbs + CrtB)	949bp
K118005	Edinburgh 08	(rbs + CrtI)	1498bp 2 PstI sites removed.
K118013	Edinburgh 08	(rbs + CrtY)	1162bp
K152005	Guelph 08	(rbs+CrtE) + (rbs+CrtB) + (rbs+CrtI) + (rbs+CrtY) + (rbs+GFP)	5412bp 2 PstI sites in CrtI removed.

The above enzymes all come from a type of Gram negative Enterobacteria, Pantoea ananatis (GenBank: D90087.2). It is noted that analogues of these enzymes are also found in other organisms (e.g. bacteria Rhodobacter spp. and the fungus Neurospora crassa).

During our preliminary research, we have encountered a few issues:

·         Originally, the registry annotation for K152005 indicated that there was no ribosome binding site before gene CrtI (i.e. “…rbs+CrtB+CrtI+rbs+CrtY…”). After clarification with Team Guelph 08, we were told that it was an error in annotation. The registry entry for K152005 has since been updated to include the ribosome binding site.

·         Teams of previous years constructed various composite parts containing two or more enzymes of the synthetic pathways (e.g. CrtEBI, CrtEBIY), but they are not available on the 2009 Distribution Plates.

·         We put K152005 under control of constitutive promoter (R0011) and transformed the construct into E.coli TOP10 cells. However, the plasmids seemed unstable (the colonies streaks on solid agar plates were inconsistent in colour) and the pigment production was low. This indicated that a different strain of E.coli might be needed for pigment production.

 

Create our own devices by standard assembly

Given the information at hand, we decided to adopt the following general strategies:

• Create a lycopene-producing device (with enzymes CrtE, CrtB and CrtI) by standard assembly of Biobricks K118014, K118006 and K118005.
• Create a β-carotene-producing device (with enzymes CrtE, CrtB, CrtI and CrtY) by standard assembly of Biobricks K118014, K118006, K118005 and K118013.
• Put the above two constructs under constitutive promoter (R0011) and arabinose-induced promoter/Pbad (I0500) and test the effect on colour production.

Scheme for standard assembly

 

After constructing our devices, we conducted restriction digest (cut with enzymes XbaI and PstI on Biobrick prefix and suffix, respectively) to check that the sizes of the inserts were correct. The results are as follows:

TESTING AND CHARACTERISATION

In vivo expression of our devices in E.coli MG1655

Our preliminary research indicated that carotenoids production in E.coli TOP10 was very weak. We decided to use another strain, E.coli MG1655, for in vivo expression of our devices. The amount of carotenoids produced became visibly higher. Generally, the colour became visible after incubation for about 12 hours.

Lycopene-producing device (constitutive)

E.coli MG1655 transformed with K274110 and grown on agar plate overnight.	Cell pellet of E.coli MG1655 transformed with K274110 (from 20mL LB culture at 37°C for 24 hours).

β -carotene-producing device (constitutive)

E.coli MG1655 transformed with K274210 and grown on agar plate overnight.	Cell pellet of E.coli MG1655 transformed with K274210 (from 20mL LB culture at 37°C for 24 hours).

 

Quantitative measurement of pigment production

In addition to visual inspection of coloured colonies, we hope to quantitatively measure the amount of lycopene or β-carotene produced by our devices. We grew transformed E.coli MG1655 in LB culture at 37°C for 24 hours and collected cell pellet by centrifugation. Acetone was added to the cell pellet and warmed at 50°C for 10 minutes, allowing carotenoids (lycopene or β-carotene) in the cells to dissolve. The acetone extracts were then put into MicroPlate Reader for a full photospectrum scan (wavelengths between 300nm and 800nm). The absorbance curves showed peaks characteristics of the respective carotenoids: at wavelength 474nm for lycopene, and 456nm for β-carotene. In the case of β-carotene, we used pure β-carotene of known amount as the standard reference and included it in the graph below.

Lycopene-producing device (constitutive)

Lycopene in cells (pellet from 20mL LB culture at 37°C for 24h) was extracted with 300μL acetone. For each reading, 100μL acetone exextract was diluted with 100μL water and loaded on to MicroPlate Reader for a full spectrum. Each graph shows the average of two readings. Value normalised with blank (50% acetone) and OD600.

β -carotene-producing device (constitutive)

β-carotene in cells (pellet from 20mL LB culture at 37°C for 24h) was extracted with 500μL acetone. 100μL acetone extract was diluted with 100μL water and loaded on to MicroPlate Reader for a full spectrum scan. Value normalised with blank (50% acetone) and OD600. Pure carotene (from Invitrogen) was added as standard. Each graph shows average of three readings.

To investigate the induction of pigment production, we tested our device K274220 in E.coli MG1655 and 1mM arabinose was added to LB culture. Cell culture that was induced with arabinose showed greater β-carotene production than cell culture without arabinose, indicating the feasibility of inducing pigment production. Using known amount of pure β-carotene to construct a standard curve, we were able to estimate the amount of β-carotene produced per unit volume of cell culture by our device.

β-carotene in cells (pellet from 20mL LB culture at 37°C for 24h; 1mM arabinose was used for induction) was extracted with 500μL acetone. 100μL acetone extract was diluted with 100μL water and loaded on to MicroPlate Reader for a full spectrum scan. Value normalised with blank (50% acetone) and OD600. Pure carotene (Invitrogen) was added as standard. Each graph shows average of three readings.

Samples Abs. (unit) Carotene (μg /100μL Acetone extract) Total mass of carotene (μg /mL-culture) K274210 in E.coli MG1655 0.4 6 6*5/20=1.5 K274220, with 1mM ara. in E.coli MG1655 0.3 5 5*5/20=1.25

Discussion of results

Despite relatively poor performance in E.coli TOP10, our pigment-producing devices gave encouraging results in E.coli MG1655. Simply by overnight incubation, the colonies were able to show visible colour changes. Since the precursors were already present in normal E.coli, no additional growth medium or special treatment was needed. This paves the way to developing a signalling system that is quick to produce visual outputs.

The acetone extraction of carotenoid provides a low-cost method for quantitative measurement. In the above experiments, the results of photo-spectrometry were consistent with characteristic absorbance wavelengths of lycopene (around 475nm) and β-carotene (around 450nm), as noted in literature. In the case of β-carotene, the absorbance profile of our β-carotene-producing device (K274210) matched that of pure carotene standards, confirming that our device was functional in the cells.

When we overlay the graphs of our two devices (constitutive), we obtained the following result:

red: K274110/lycopene; orange: K274210/β-carotene; black: pure β-carotene; blue: untransformed E.coli MG1655 (negative control)

Lycopene and β-carotene showed characteristic absorbance profiles, with different maximum absorbance wavelengths. It was tempted to think that by converting lycopene to β-carotene (e.g. by expressing CrtY gene), we would be able to see a “shift” in the peaks of maximum absorbance. However, in the bar charts above it was noted that lycopene and β-carotene had significant absorbance at both 456nm and 474nm, albeit with different ratios (i.e. lycopene absorbed more at 474nm; β-carotene absorbed more at 456nm). Therefore, measurement at a single wavelength was insufficient to determine the identity of the carotenoid; we might need to use “456nm:474nm” ratio instead. It would be interesting to see the effect of mixing different amounts of lycopene and β-carotene, and whether the “456nm:474nm” ratio can indicate the relative percentage of lycopene and β-carotene. In the current project, we were unable to obtain pure lycopene due to high cost of purchase.

In terms of inducing β-carotene production, 1mM arabinose, acting on Pbad promoter, was able to produce a level of synthesis close to that under constitutive promoter. In the non-induced population, there was probably a small degree of “leakiness” and basal level of gene expression (leading to some β-carotene synthesis). In future experiments, it might be helpful to move the pigment-producing devices to low copy plasmids and reduce the effect of such basal expression.

One possible design using our devices is to construct a system where lycopene is constitutively produced and CrtY, which converts red lycopene to orange β-carotene, is controlled by external stimulus. Hence, the red cells will turn orange upon sensing the specific external stimulus, and this change in colour serves as the signal output that can be detected visually and measured quantitatively at low cost.