Difference between revisions of "Part:BBa K2924051"

(Characterization)
(Characterization)
Line 100: Line 100:
 
</html>
 
</html>
  
[[File:TesA_Triple_trafo.png|thum|right|400px|<i><b>Fig. 4:</b> Schematic overview of the different constructs used for <i>in vivo</i> quantification of fatty acids.</i>]]
+
[[File:TesA_Triple_trafo.png|thum|right|500px|<i><b>Fig. 4:</b> Schematic overview of the different constructs used for <i>in vivo</i> quantification of fatty acids.</i>]]
 
<html><p align="justify"></html>
 
<html><p align="justify"></html>
 
The samples were grown in LB-medium with the appropriate antibiotics. The absorption was measured and the samples were diluted to an OD<sub>600</sub> of 0.5. In addition to the samples, a negative control was included in the experiment. The samples were induced by 1 mM rhamnose and, in the case of ACC overexpression, additionally with 1 mM IPTG. The induced samples were incubated in a 24 well plate at 37°C, which were shaken at 250 rpm. After nearly 16 hours, the samples were taken out of the incubator and transferred on a 96 well plate. Three biological with three technical replicates each were measured. For the analysis, a plate reader was used to measure absorption at a wavelength of 600 nm. The fluorescence was measured at a wavelength of λ<sub>ex/em</sub>= 470 / 515 nm for sfGFP as the reporter gene and at a wavelength of λ<sub>ex/em</sub>= 497 / 540 nm for eYFP as the reporter gene.  
 
The samples were grown in LB-medium with the appropriate antibiotics. The absorption was measured and the samples were diluted to an OD<sub>600</sub> of 0.5. In addition to the samples, a negative control was included in the experiment. The samples were induced by 1 mM rhamnose and, in the case of ACC overexpression, additionally with 1 mM IPTG. The induced samples were incubated in a 24 well plate at 37°C, which were shaken at 250 rpm. After nearly 16 hours, the samples were taken out of the incubator and transferred on a 96 well plate. Three biological with three technical replicates each were measured. For the analysis, a plate reader was used to measure absorption at a wavelength of 600 nm. The fluorescence was measured at a wavelength of λ<sub>ex/em</sub>= 470 / 515 nm for sfGFP as the reporter gene and at a wavelength of λ<sub>ex/em</sub>= 497 / 540 nm for eYFP as the reporter gene.  
  
[[File:TesA_palda_biosensor.png|thumb|right|300px|<i><b>Fig. 5:</b> In vivo P<sub>aldA</sub> eYFP biosensor of transformed E. coli cells with and without tesA. The expression of tesA was induced with 1 mM Rhamnose and the cells were incubated for 16 hours.</i>]]
+
[[File:TesA_palda_biosensor.png|thumb|right|400px|<i><b>Fig. 5:</b> In vivo P<sub>aldA</sub> eYFP biosensor of transformed E. coli cells with and without tesA. The expression of tesA was induced with 1 mM Rhamnose and the cells were incubated for 16 hours.</i>]]
 
<html><p align="justify"></html>
 
<html><p align="justify"></html>
 
Figure 5 shows a significant fluorescence response of eYFP, because of the detection of fatty acids by the promoter P<sub>aldA</sub>. Compared to the wild type, it can be said that in the induced transformant a clearly higher amount of fatty acids is present. Also compared to the uninduced control, it can be said that the inducible promoter <a href="https://parts.igem.org/Part:BBa_K914003">BBa_K914003</a> is not leaking and works accurate.
 
Figure 5 shows a significant fluorescence response of eYFP, because of the detection of fatty acids by the promoter P<sub>aldA</sub>. Compared to the wild type, it can be said that in the induced transformant a clearly higher amount of fatty acids is present. Also compared to the uninduced control, it can be said that the inducible promoter <a href="https://parts.igem.org/Part:BBa_K914003">BBa_K914003</a> is not leaking and works accurate.
Line 110: Line 110:
 
P<sub><i>aldA</i></sub> is known to response to C18:1<sup>9</sup>, therefore this could indicate for the present of the C18:1 fatty acid.
 
P<sub><i>aldA</i></sub> is known to response to C18:1<sup>9</sup>, therefore this could indicate for the present of the C18:1 fatty acid.
  
[[File:TesA_par_Biosensor.png|thumb|right|300px|<i><b>Fig. 6:</b> In vivo P<sub>AR</sub> sfGFP biosensor of transformant E. coli cells with and without tesA and the Acetyl-CoA carboxylase (ACC). The expression of tesA was induced with 1 mM Rhamnose and the expression of the ACC complex was induced with 1 mM IPTG. The cells were then incubated for 16 hours.</i>]]
+
[[File:TesA_par_Biosensor.png|thumb|right|400px|<i><b>Fig. 6:</b> In vivo P<sub>AR</sub> sfGFP biosensor of transformant E. coli cells with and without tesA and the Acetyl-CoA carboxylase (ACC). The expression of tesA was induced with 1 mM Rhamnose and the expression of the ACC complex was induced with 1 mM IPTG. The cells were then incubated for 16 hours.</i>]]
 
<html><p align="justify"></html>
 
<html><p align="justify"></html>
 
In figure 6, there is no significant difference in fluorescence response by the biosensor compared to the wild type even compared to the uninsured transformant. The P<sub>AR</sub> is known to detect a broad range of medium-chain fatty acids<sup>10</sup>, which could indicate that 'TesA does not have a catalytic activity for medium-chain fatty acyl-substrates and therefore, there is no increase in medium-chain fatty acids in the induced transformant.
 
In figure 6, there is no significant difference in fluorescence response by the biosensor compared to the wild type even compared to the uninsured transformant. The P<sub>AR</sub> is known to detect a broad range of medium-chain fatty acids<sup>10</sup>, which could indicate that 'TesA does not have a catalytic activity for medium-chain fatty acyl-substrates and therefore, there is no increase in medium-chain fatty acids in the induced transformant.
 +
 +
 +
====GC-MS====
 +
<html><p align="justify">
 +
The <i>E. coli</i> BL21 transformants with pSNDY_P<sub>rha</sub>:’<i>tesA</i> were used to inoculate an overnight culture. A 30 ml main culture was grown to OD<sub>600</sub> 0.2 and 1 mM Rhamnose was added. The culture was incubated at 37°C for 24h, after which 4 optical density units of cells, usually, an equivalent of 4 ml cells at OD<sub>600</sub> = 1, were isolated and used for fatty acid <a href="https://www.protocols.io/view/fatty-acid-extraction-and-derivatisation-79jhr4n">extraction protocol</a>.
 +
<html><p align="justify">
 +
The positive <i>Synechocystis sp.</i> conjugants containing pSNDY_P<sub>rha</sub>:’<i>tesA</i> were also inoculated and grown for a few days before adding 0.5 mM rhamnose. After 24 hours at 30°C, the cells were isolated and used for <a href="https://www.protocols.io/view/fatty-acid-extraction-and-derivatisation-79jhr4n">extraction and derivatization</a> of fatty acids.
 +
</html>
 +
 +
[[File:TesA_GC.png|thumb|center|800px|<i><b>Fig. 7:</b> Relative fatty acid yield of transformants with the leaderless ‘TesA compared to the control strains. Different fatty acids are listed for <i>E.coli</i> (blue and red) and for <i>Synechocystis</i> (green and orange). Measurements were carried out <i>via</i> GC-MS.</i>]]
 +
<html><p align="justify">
 +
 +
As shown in figure 7, in <i>E.coli</i> transformants a higher yield from C16:1 to C20:2 was detected compared to the control. Furthermore, there is a clear increase in C20:1 fatty acid. As for <i>Synechocystis</i> conjugants, there was also a higher yield for several long-chain fatty acids detected compared to the <i>Synechocystis</i> control.  Here, too, a clear increase in the C18:0 and C20:2 fatty acids can be seen.
 +
Interestingly, ‘TesA seems to function differently when expressed in <i>Synechocystis</i> compared to its native host <i>E. coli</i>. While both <i>Synechocystis</i> and <i>E. coli</i> show distinctive fatty acid distribution patterns when overexpressing ‘<i>tesA</i>, compared to the respective WT strain, these differ strongly between the two species.
 +
</html>
 +
 +
[[File:TesA_GC_Biosensor.png|thumb|right|400px|<i><b>Fig. 8:</b> Comparison of the collected results of the <i>in vivo</i> biosensor P<sub>aldA</sub>:<i>eYFP</i> (BBa_K2924017) and the GC-MS.</i>]]
 +
 +
<html><p align="justify">
 +
By comparing these results (figure 8), it can be said, we were able to show a strong correlation between the two methods. In summary, we could not only show that our biosensor works as a fast, efficient read-out for intracellular fatty acids, but that the data obtained from it corresponds to actual measured GC-MS data.
 +
</html>
 +
 +
[[File:TesA_profile_comparison.png|thumb|right|400px|<i><b>Fig. 9:</b> Fatty acid profile of modified <i>Synechocystis</i> compared to <i>Synechocystis</i> and <i>E.coli</i> wild type.</i>]]
 +
 +
<html><p align="justify">
 +
As it seems, by adding the leaderless thioesterase ‘TesA from <i>E.coli</i> to <i>Synechocystis</i>, the fatty acids profile of <i>Synechocystis</i> is getting more similar to the fatty acid profile of <i>E.coli</i>. This can be seen in the relation of C18:1 and C18:3 fatty acids or between the C20:1 and C20:2 fatty acid. 
 +
</html>
 +
 +
===References===
 +
 +
[-1]: Kelly, C.L., Taylor, G.M., Hitchcock, A., Torres-Méndez, A., Heap, J.T. A Rhamnose-Inducible System for
 +
Precise and Temporal Control of Gene Expression in Cyanobacteria. ACS Synth Biol. 2018 Apr 20;7(4):1056-1066.
 +
 +
[0]: Behle, A., Saake, P., Axmann I. M. . "Comparative analysis of inducible promoters in cyanobacteria." bioRxiv
 +
(2019): 757948.
 +
 +
[1]:  Heidorn, Thorsten, et al. "Synthetic biology in cyanobacteria: engineering and analyzing novel functions." <i>Methods in enzymology.</i> Vol. 497. Academic Press, 2011. 539-579.
 +
 +
[2]: National Center for Biotechnology Information. PubChem Database. Palmitic acid, CID=985, https://pubchem.ncbi.nlm.nih.gov/compound/Palmitic-acid (accessed on Oct. 5, 2019)
 +
 +
[3]: National Center for Biotechnology Information. PubChem Database. Stearic acid, CID=5281, https://pubchem.ncbi.nlm.nih.gov/compound/Stearic-acid (accessed on Oct. 5, 2019)
 +
 +
[4]: Carnielli, Virgilio P., et al. "Structural position and amount of palmitic acid in infant formulas: effects on fat, fatty acid, and mineral balance." Journal of pediatric gastroenterology and nutrition 23.5 (1996): 553-560.
 +
 +
[5]: Mu, Yi-Ming, et al. "Saturated FFAs, palmitic acid and stearic acid, induce apoptosis in human granulosa cells." Endocrinology 142.8 (2001): 3590-3597.
 +
 +
[6]:Loften, J. R., et al. "Invited review: Palmitic and stearic acid metabolism in lactating dairy cows." Journal of dairy science 97.8 (2014): 4661-4674.
 +
 +
[7]: Zinchenko V, Piven IV, Melnik VA, Shestakov SV: Vectors for the Complementation Analysis of Cyanobacterial Mutants. Russian Journal of Genetics 1999
 +
 +
[8]: Wolk CP, Vonshak A, Kehoe P, Elhai J: Construction of shuttle vectors capable of conjugative transfer from Escherichia coli to nitrogen-fixing filamentous cyanobacteria. Proceedings of the National Academy of Sciences 2004, 81:1561–1565.
 +
 +
[9]: 1: Mee-Jung Han, Jeong Wook Lee, Sang Yup Lee, and Jong Shin Yoo. <i>“Proteome-Level Responses of Escherichia coli to Long-Chain Fatty Acids and Use of Fatty Acid Inducible Promoter in Protein Production” </i> Hindawi Publishing Corporation Journal of Biomedicine and Biotechnology Volume 2008, Article ID 735101,12 pages
 +
 +
[10]: Fuzhong Zhang, James M Carothers, Jay D Keasling. “Design of a dynamic sensor-regulator system for production of chemicals and fuels derived from fatty acids” Nature Biotechnology volume 30, pages 354–359 (2012)

Revision as of 17:41, 20 October 2019


Leaderless 'TesA

Rhamnose-inducible promoter (similar to BBa_K914003) and RBS* (BBa_K29240091) expressing the thioesterase tesA (BBa_K1472601)and a double terminator (BBa_B0015).

Usage and Biology

This part contains the rhamnose-inducible promoter Prha described by Kelly et al. (2018)-1 which was further modified by Behle et al. (2019)0, and is similar to (BBa_K914003), the RBS* (BBa_K2924009)1, a thioesterase from Escherichia coli (BBa_K1472601) and a double terminator (BBa_B0015).

In general, all basic parts used here (with the exception of ‘TesA) were optimized for Synechocystis sp. PCC 6803, but function efficiently in both E. coli and Synechocystis sp..

This thioesterase lacks a leader peptide sequence, which is the N-terminal 26-amino acid signal peptide sequence. ’TesA is located in the cytosol of E.coli, where it hydrolyzes the bond of fatty acyl to acyl-carrier protein (ACP) or Coenzyme A (CoA). This results in a free fatty acids and either ACP or CoA.

Background

Fig. 1: Structural formula of hexadecanoic acid. Gray spheres represent a carbon atom, red spheres represent oxygen atom and the white spheres represent hydrogen atoms.
Fig. 2: Structural formula of octadecanoic acid. Gray spheres represent a carbon atom, red spheres represent oxygen atom and the white spheres represent hydrogen atoms.

Fatty acids are long aliphatic chained carboxylic acids, which can be saturated or unsaturated. They have mostly an even number of carbon atoms from 4 to 28. Hexadecanoic acid (Fig. 1) is a saturated long-chain fatty acid, which contains 16 carbon-atoms and is also called palmitic acid (C16:0). It has a white or colorless crystalline form and a slightly characteristic odor2. It has a molecular weight of 256.42 g/mol.

Octadecanoic acid (Fig. 2) is a saturated long-chain fatty acid, which contains 18 carbon-atoms and is also called stearic acid (C18:0). It is a white or slightly yellow, wax-like solid with mild odor3. It has a molecular weight of 284.48 g/mol.

Usage of palmitic and stearic acid

The long-chain fatty acid palmitic acid (C16:0) is found naturally in human milk, in cow's milk fat for triglycerides or in vegetable oils and it is commonly used in infant formulas4. It is the most abundant fatty acid in cow’s milk5.

Palmitic acid (C16:0) is an important chemical, which is used to make soaps or cosmetics agents, lubricating oils, waterproofing materials, food additives and other chemicals2.

Stearic acid (C18:0) is also found naturally in animal milk and vegetable oils3 and used in manufacturing of different pharmaceuticals, cosmetics, soaps, candles, food packaging, modeling compounds and other chemicals. It is found sometimes in pesticides3.

Moreover, they play an important role for the intracellular biological functions. They mainly serve as source of metabolic energy or as substrates for the cell membrane biosynthesis. Another function is serving as precursor, such as prostaglandins (PGs), leukotrienes and more, for different signaling pathways5. It has been shown that saturated fatty acids, such as palmitic and stearic acid (C16:0; C18:0), induce apoptosis of human granulosa cells, which could be used for further medical investigations5. Both fatty acids are used for special dietary to increase the milk fat yield and as an energy source for milk production6.


Sequence and Features


Assembly Compatibility:
  • 10
    COMPATIBLE WITH RFC[10]
  • 12
    COMPATIBLE WITH RFC[12]
  • 21
    COMPATIBLE WITH RFC[21]
  • 23
    COMPATIBLE WITH RFC[23]
  • 25
    COMPATIBLE WITH RFC[25]
  • 1000
    COMPATIBLE WITH RFC[1000]


Characterization

First, the inducible promoter, as well as the terminator with Gibson overhangs were fused to ‘tesA via overlap extension PCR. This PCR product was used for Gibson assembly to ligate it into the pSNDY backbone. The conjugative shuttle vector pSNDY is a pSHDY7 derivative, encoding a nourseothricin resistance instead of spectinomycin, and is a broad-host-range vector able to self-replicate in E. coli, as well as to be transferred to other hosts such as cyanobacteria via conjugation.

Fig. 3: Transformants differ in morphology. <i>E. coli BL21 control transformation (left) and E. coli BL21 transformed with pSNDY containing Prha:’tesA (right).</i>

After transformation of the Prha:’tesA construct, the transformants of E. coli BL21 seem to differ in morphology (Figure 3).

With the help of E. coli DH5α transformed with the ‘tesA construct, as well as the helper strain RP48, pSNDY encoding Prha:’tesA was transferred to Synechocystis sp. via <a href=”https://www.protocols.io/view/triparental-mating-of-synechocystis-ftpbnmn”>conjugation</a>. The effect of ’tesA expression and activity was first tested in vivo with a fatty acid responsive biosensor by detecting the response to intracellular free fatty acids via a reporter protein, and then later by gas chromatography-mass spectrometry (GC-MS).


Biosensor

For a fast and convenient method to detect fatty acids in vivo, the ‘TesA strain was combined with different biosensor constructs containing different promoter and additionally also combined with a construct encoding an Acetyl-CoA carboxylase complex expression cassette (ACC) <a href="https://parts.igem.org/Part:BBa_K2924021">(BBa_K2924021)</a>, which is the rate limiting step of the fatty acid synthesis. E. coli BL21 (DE3) was transformed with various combinations of the different constructs, as shown in Table 1.

Biosensor Thioesterase ACC Antibiotics used
PaldA:eYFP 'tesA no Chloramphenicol, CloNat
PAR:sfGFP 'tesA no Chloramphenicol, CloNat
PAR:sfGFP 'tesA yes Chloramphenicol, CloNat, Ampicillin

Fig. 4: Schematic overview of the different constructs used for <i>in vivo quantification of fatty acids.</i>

The samples were grown in LB-medium with the appropriate antibiotics. The absorption was measured and the samples were diluted to an OD600 of 0.5. In addition to the samples, a negative control was included in the experiment. The samples were induced by 1 mM rhamnose and, in the case of ACC overexpression, additionally with 1 mM IPTG. The induced samples were incubated in a 24 well plate at 37°C, which were shaken at 250 rpm. After nearly 16 hours, the samples were taken out of the incubator and transferred on a 96 well plate. Three biological with three technical replicates each were measured. For the analysis, a plate reader was used to measure absorption at a wavelength of 600 nm. The fluorescence was measured at a wavelength of λex/em= 470 / 515 nm for sfGFP as the reporter gene and at a wavelength of λex/em= 497 / 540 nm for eYFP as the reporter gene.

Fig. 5: In vivo PaldA eYFP biosensor of transformed E. coli cells with and without tesA. The expression of tesA was induced with 1 mM Rhamnose and the cells were incubated for 16 hours.

Figure 5 shows a significant fluorescence response of eYFP, because of the detection of fatty acids by the promoter PaldA. Compared to the wild type, it can be said that in the induced transformant a clearly higher amount of fatty acids is present. Also compared to the uninduced control, it can be said that the inducible promoter <a href="https://parts.igem.org/Part:BBa_K914003">BBa_K914003</a> is not leaking and works accurate.

PaldA is known to response to C18:19, therefore this could indicate for the present of the C18:1 fatty acid.

Fig. 6: In vivo PAR sfGFP biosensor of transformant E. coli cells with and without tesA and the Acetyl-CoA carboxylase (ACC). The expression of tesA was induced with 1 mM Rhamnose and the expression of the ACC complex was induced with 1 mM IPTG. The cells were then incubated for 16 hours.

In figure 6, there is no significant difference in fluorescence response by the biosensor compared to the wild type even compared to the uninsured transformant. The PAR is known to detect a broad range of medium-chain fatty acids10, which could indicate that 'TesA does not have a catalytic activity for medium-chain fatty acyl-substrates and therefore, there is no increase in medium-chain fatty acids in the induced transformant.


GC-MS

The E. coli BL21 transformants with pSNDY_Prha:’tesA were used to inoculate an overnight culture. A 30 ml main culture was grown to OD600 0.2 and 1 mM Rhamnose was added. The culture was incubated at 37°C for 24h, after which 4 optical density units of cells, usually, an equivalent of 4 ml cells at OD600 = 1, were isolated and used for fatty acid extraction protocol.

The positive Synechocystis sp. conjugants containing pSNDY_Prha:’tesA were also inoculated and grown for a few days before adding 0.5 mM rhamnose. After 24 hours at 30°C, the cells were isolated and used for extraction and derivatization of fatty acids.

Fig. 7: Relative fatty acid yield of transformants with the leaderless ‘TesA compared to the control strains. Different fatty acids are listed for <i>E.coli (blue and red) and for Synechocystis (green and orange). Measurements were carried out via GC-MS.</i>

As shown in figure 7, in E.coli transformants a higher yield from C16:1 to C20:2 was detected compared to the control. Furthermore, there is a clear increase in C20:1 fatty acid. As for Synechocystis conjugants, there was also a higher yield for several long-chain fatty acids detected compared to the Synechocystis control. Here, too, a clear increase in the C18:0 and C20:2 fatty acids can be seen. Interestingly, ‘TesA seems to function differently when expressed in Synechocystis compared to its native host E. coli. While both Synechocystis and E. coli show distinctive fatty acid distribution patterns when overexpressing ‘tesA, compared to the respective WT strain, these differ strongly between the two species.

Fig. 8: Comparison of the collected results of the <i>in vivo biosensor PaldA:eYFP (BBa_K2924017) and the GC-MS.</i>

By comparing these results (figure 8), it can be said, we were able to show a strong correlation between the two methods. In summary, we could not only show that our biosensor works as a fast, efficient read-out for intracellular fatty acids, but that the data obtained from it corresponds to actual measured GC-MS data.

Fig. 9: Fatty acid profile of modified <i>Synechocystis compared to Synechocystis and E.coli wild type.</i>

As it seems, by adding the leaderless thioesterase ‘TesA from E.coli to Synechocystis, the fatty acids profile of Synechocystis is getting more similar to the fatty acid profile of E.coli. This can be seen in the relation of C18:1 and C18:3 fatty acids or between the C20:1 and C20:2 fatty acid.

References

[-1]: Kelly, C.L., Taylor, G.M., Hitchcock, A., Torres-Méndez, A., Heap, J.T. A Rhamnose-Inducible System for Precise and Temporal Control of Gene Expression in Cyanobacteria. ACS Synth Biol. 2018 Apr 20;7(4):1056-1066.

[0]: Behle, A., Saake, P., Axmann I. M. . "Comparative analysis of inducible promoters in cyanobacteria." bioRxiv (2019): 757948.

[1]: Heidorn, Thorsten, et al. "Synthetic biology in cyanobacteria: engineering and analyzing novel functions." Methods in enzymology. Vol. 497. Academic Press, 2011. 539-579.

[2]: National Center for Biotechnology Information. PubChem Database. Palmitic acid, CID=985, https://pubchem.ncbi.nlm.nih.gov/compound/Palmitic-acid (accessed on Oct. 5, 2019)

[3]: National Center for Biotechnology Information. PubChem Database. Stearic acid, CID=5281, https://pubchem.ncbi.nlm.nih.gov/compound/Stearic-acid (accessed on Oct. 5, 2019)

[4]: Carnielli, Virgilio P., et al. "Structural position and amount of palmitic acid in infant formulas: effects on fat, fatty acid, and mineral balance." Journal of pediatric gastroenterology and nutrition 23.5 (1996): 553-560.

[5]: Mu, Yi-Ming, et al. "Saturated FFAs, palmitic acid and stearic acid, induce apoptosis in human granulosa cells." Endocrinology 142.8 (2001): 3590-3597.

[6]:Loften, J. R., et al. "Invited review: Palmitic and stearic acid metabolism in lactating dairy cows." Journal of dairy science 97.8 (2014): 4661-4674.

[7]: Zinchenko V, Piven IV, Melnik VA, Shestakov SV: Vectors for the Complementation Analysis of Cyanobacterial Mutants. Russian Journal of Genetics 1999

[8]: Wolk CP, Vonshak A, Kehoe P, Elhai J: Construction of shuttle vectors capable of conjugative transfer from Escherichia coli to nitrogen-fixing filamentous cyanobacteria. Proceedings of the National Academy of Sciences 2004, 81:1561–1565.

[9]: 1: Mee-Jung Han, Jeong Wook Lee, Sang Yup Lee, and Jong Shin Yoo. “Proteome-Level Responses of Escherichia coli to Long-Chain Fatty Acids and Use of Fatty Acid Inducible Promoter in Protein Production” Hindawi Publishing Corporation Journal of Biomedicine and Biotechnology Volume 2008, Article ID 735101,12 pages

[10]: Fuzhong Zhang, James M Carothers, Jay D Keasling. “Design of a dynamic sensor-regulator system for production of chemicals and fuels derived from fatty acids” Nature Biotechnology volume 30, pages 354–359 (2012)