Network Elements

SmoColi can be divided into different network elements: the major sub-networks being a sensor, a band-pass filter and a quorum sensing mechanism.

Cigarette smoke contains a lot of different toxic and carcinogenic components. In nature there are many sensors for such components, mostly to induce their degradation. One is the acetaldehyde system in aspergillus nidulans: the activator AlcR binds to its operator site if acetaldehyde is present [1]. Another one is the xylene sensing system in Pseudomonas putida [2] working as an activator. Since our system is constructed in a modular fashion an can be coupled with any sensor, an arabinose sensing system was established as a proof of principle. We combined a smoke-sensitive band-pass filter with GFP output with a quorum-sensing based mechanism alarming the user at high smoke levels by expressing RFP.

Acetaldehyde Sensor

Where it comes from...

Figure 1: Structures of the AlcR(1-60)-DNA complex [13]

In Aspergillus nidulans, the alcR gene encodes a regulatory protein which activates the expression of the ethanol utilization (alc) pathway if the co-inducer acetaldehyde is present. Ethanol is converted to acetaldehyde by the alcohol dehydrogenase and further metabolized to acetate by the aldehyde dehydrogenase. Both genes alcA and aldA are regulated by AlcR. AlcR also regulates its own expression by autoactivation [1]. The AlcR DNA binding domain contains a zinc bi-nuclear cluster, which can bind either to symmetric and asymmetric sites with same affinity. In contrast to other members of the Zn₂Cys₆, the DNA binding domain is stretch of 16 amino acid residues between the third and fourth cysteine. AlcR binds as a monomer, but 2 proteins can bind to inverted repeats in a noncooperative manner. Additional DNA sequences upstream of the zinc cluster were identified to be responsible for high-affinity binding, [4].

How we used it

Because fungal activators normally do not work in E. coli, we modified AlcR in a way that it acts as a repressor. The mechanism of how AlcR activates transcription is not completely known, which makes the process of redesigning it very challenging. A second challenge is the different codon usage in Apergillus nidulans and E. coli. The codon adaptation index (CAI) of Aspergillus nidulans AlcR is 0.7 in E. coli, especially at the beginning of the gene a few very rare codons were included. To get a fully E. coli-optimized gene, we ordered the gene codon-optimized. To get a partly codon-optimized version, we exchanged the rare codons at the beginning of the Aspergillus nidulans protein with PCR.

Figure 2: Design of the promoter for the AlcR-system in SmoColi, the operon sequence is designed between the -10 and the -35 region (blue) of a strong promoter with inverted and direct repeats.

For the redesigned promoter P_Alc, the operator site of AlcR from Aspergillus nidulans was placed between the -10 and -35 region of the bacterial promoter. We used two different versions as DNA targets, one with direct-repeat targets and one with inverted [1]. The idea is that in the presence of acetaldehyde, AlcR will bind to the operon and block the binding of the RNA polymerase. This only works if the mechanism of AlcR binding does not involve conformational changes or anything similar. To make sure that an excess of AlcR is present with respect to P_Alc, they were expressed on different plasmids with different copy numbers. The AlcR promoter (P_Alc) is cloned in front of the TetR repressor to introduce the bandpass.

Figure 3: Circuit operations for SmoColi with acetaldehyde sensor, AlcR works as an repressor, therefore a inverter (TetR) is introduced to convert the repression of the input to an activation.

Acetaldehyde is reduced by the native E. coli enzyme aldehyde dehydrogenase, generating a concentration gradient in our tube. At the position with the required concentration of acetaldehyde for the bandpass-filter, GFP is expressed, at all other positions it is repressed, resulting in a single GFP band. By tuning the input of acetaldehyde in the tube, we get the GFP band at different positions. Thus, with constantly increasing acetaldehyde concentration, we get a moving band from one side to the other side of the channel. Finally,if the acetaldehyde concentration crosses a certain threshold, the whole channel turns red, due to the quorum-sensing based RFP expression.

Xylene Sensor

Where it comes from...

The xylR gene in Pseudomonas putida encodes a 566 aa long regulatory protein that activates the degradation of toluene and xylene. In P. putida the complete upper Tol pathway is under the control of the toluene-responsive promoter P_U, in the presence of xylene XylR binds to P_U promoter and activates it [5]. Then the five genes xylC-xylM-xylA-xylB-xylN are expressed, where xylC encodes for the benzaldehyde dehydrogenase, xylM and xylA for subunits of the xylene oxygenase and xylB for benzyl alcohol dehydrogenase [6]. XlyN is not required for the degradation of xylene, but it is part of the whole transcription unit. The DNA binding domain of XylR binds to a 40 bp long upstream activating sequence (UAS) located 150 bp upstream from the transcription start site of the σ⁵⁴ P_U promoter. DNA looping between the UAS and the -12/-24 motifs of the σ⁵⁴-RNAP facilitates the activation of the holoenzyme by XylR [7].

How we used it

Figure 4: Circuit operations of SmoColi with xylene sensor

Xylene is not degraded naturally by E. coli. To generate a concentration gradient of xylene we engineered its degradation in SmoColi by including the upper Tol pathway. XylR is a prokaryotic enhancer binding protein which works in E. coli as well as in Pseudomonas putida. The LacI_M1 and the CI regulators were under the control of the P_U promoter, enhancing their expression in the presence of xylene.

Arabinose Sensor

Where it comes from...

Figure 5: Structures of the AraC heterodimere with L-Arabinose [14]

The AraC regulatory protein is a activator of the genes for the araBAD, the araE and the araFGH operon in E.coli. araB, araA and araD encode for the metabolic enzymes which degrade L-Arabinose to D-xylulose-5-phosphate. In absence of arabinose AraC does bind to two DNA site separated over 200 bp from each other (I₁ and O₂ of the P_BAD promoter) [8]. By dimer formation of the two DNA bound AraC molecules the DNA is looped and transcription blocked. Binding of arabinose leads to binding of arabinose to I₁ and I₂, I₂ bound AraC directly interacts with the RNA polymerase and activates it. araF, araG and araH and araE encode for proteins needed for uptake of arabinose. If some arabinose enters the cell, the uptake of arabinose is activated leading and more and more arabinose entering the cell. This positive feedback loop leads to a all or nothing response of arabinose in the cells.

How we used it

Figure 6: Circuit operations of SmoColi with arabinose sensor

To show that our system can be activated by different signals, we introduced a arabinose sensor. Due to the fact that AraC s well established in contract AlcR and XlyR, the arabinose system can be seen as proof of principle for the bandpass and the alarm system for our smoke sensors. The LacI_M1 and the CI regulators were under the control of the P_BAD promoter, enhancing their expression in the presence of arabinose and the absence of glucose. The all or nothing response of the arabinose system makes the natural system unusable for our bandpass, because we need a intermediate input signal. In order that we can still use the arabinose system we use the BW27783 (BW25113 DE(araFGH) Φ(∆araEp) (P_CP8-araE)) E.coli stain [9]. This stain has a knock out of the araFGH (transporter) and the P_BAD-araE is replaced by P_CP8-araE which is a consecutive promoter. Without the positive feedback loop we get a nice dose response for the system, with a intermediate input concentrations for the bandpass filter. By inactivating of the kanamycin casette we created the BW27783 from the BW27749, which was provided to us by Khlebnikov et al. (Done according to the method of: [10])

Bandpass

Where it comes from...

Figure 7: A Bandpassfilter, is a combination of a high pass filter and a low pass filter.

In 2005 Ron Weiss et al. constructed the first band-pass filter device in synthetic biology. This was done by using the combination of a high-pass and a low-pass filter. An input signal is transmitted by cell to cell communication while the signal processing is done using an intracellular feed-forward loop. Filters are created by using different transcriptional regulator proteins with different repression efficiencies and binding affinities for corresponding promoters.

How we used it

We used his pioneering network design as a basis in the planning process for our band-pass filter implementation [11]. Instead of cell-cell communication via AHL for the input signal, we included our different smoke detectors which detect small volatile molecules like acetaldeyhde, xylene or arabinose.

Inverter

Where it comes from...

An inverter is a part which converts a repressor signal into an activator signal, or vise versa. For SmoColi we used TetR as an inverter. TetR binds and blocks the TetR promoter (P_Tet).

How we used it

Figure 8: What would happen without an inverter: No GFP formation..

When the inverter is not present, the band does not appear because the activation threshold of the high pass filter is at a higher value than the repression threshold of the low pass filter, thus there is no region where both branches produce an output signal. If the input signal (a repressor) is inverted we get again a nice bandpass filter. In our circuit the inverter is only needed for AlcR, for AraC and XylR the input signal is correct (they are already activators).

Quorum Sensing of the Alarm system

Where it comes from...

The luminescence operon in Vibrio fischeri is regulated by the transcriptional activator (LuxR). The lux box is a 20 bp inverted repeat sequence at a position centered 42.5 bases upstream of the transcriptional start. LuxR binds to the lux box in presents of AHL and enhance its expression. LuxR is needed to activate the σ⁷⁰ RNA Polymerase. The Vibrio fischeri lux operon consists of seven genes, the first one is luxI. The luxI gene encodes for the enzyme for AHL synthesis. The autoinducer AHL is secreted by the cells to coordinate gene expression based on the local density of the bacterial population. Furthermore transcription of luciferase in the lux operon is induced, leading to bioluminescence.

How we used it

Figure 9: Promoter for luxR, the operon (red) designed between the -10 and the -35 region (blue) of the promoter.

For our system we used a promoter which is repressed by luxR! ([http://partsregistry.org/Part:BBa_R0061 BBa_R0061]) It was designed by placing the lux box between and partially overlapping the consensus -35 and -10 site of an artificial lacZ promoter. In the presence of AHL, LuxR binds to the lux box and inhibits binding of the RNA Polymerase. [12]

References

[1] [http://www.ncbi.nlm.nih.gov/pubmed/2834622 R. Locklngton, C. Scazzocchio, D. Sequeval, M. Mathieu, B. Felenbok: Regulation of alcR, the positive regulatory gene of the ethanol utilization regulon of Aspergillus nidulans, Mol Microbiol., 1987, 1: 275-81]

[2] [http://aem.asm.org/cgi/content/abstract/64/2/748 Sven Panke, Juan M. Sánchez-Romero, and Víctor de Lorenzo: Engineering of Quasi-Natural Pseudomonas putida Strains for Toluene Metabolism through an ortho-Cleavage Degradation Pathway, Appl Environ Microbiol, 1998, 64: 748-751]

[3] [http://aem.asm.org/cgi/content/abstract/64/6/2032 Sven Panke, Bernard Witholt, Andreas Schmid, and Marcel G. Wubbolts: Towards a Biocatalyst for (S)-Styrene Oxide Production: Characterization of the Styrene Degradation Pathway of Pseudomonas sp. Strain VLB120, Appl Environ Microbiol, 1998, 64: 2032-2043]

[4] [http://www.ncbi.nlm.nih.gov/pubmed/9182587 Lenouvel F, Nikolaev I, Felenbok B.: In vitro recognition of specific DNA targets by AlcR, a zinc binuclear cluster activator different from the other proteins of this class, J Biol Chem. 1997, 272(24):15521-6.]

[5] [http://www.springerlink.com/content/r65l105rg21t1l73/ Manabu Gomada, Sachiye Inouye, Hiromasa Imaishi, Atsushi Nakazawa and Teruko Nakazawa: Analysis of an upstream regulatory sequence required for activation of the regulatory gene xylS in xylene metabolism directed by the TOL plasmid of Pseudomonas putida, Molecular and General Genetics MGG, 1992, 233:419-426]

[6] [http://www.ncbi.nlm.nih.gov/pmc/articles/PMC210316/ S Harayama, M Rekik, M Wubbolts, K Rose, R A Leppik, and K N Timmis: Characterization of five genes in the upper-pathway operon of TOL plasmid pWW0 from Pseudomonas putida and identification of the gene products, 1989, 171(9): 5048–5055]

[7] [http://nar.oxfordjournals.org/content/31/23/6926.full Marc Valls and Víctor de Lorenzo: Transient XylR binding to the UAS of the Pseudomonas putida σ⁵⁴ promoter Pu revealed with high intensity UV footprinting in vivo, Nucl. Acids Res., 2003, 31 (23): 6926-6934]

[8] [http://gene.bio.jhu.edu/Ourspdf/108.pdf D. A. Hodgson, C. M. Thomas: SGM symposium 61: Signals, switches, regulons and cascades: control of bacterial gene expression. Cambridge University Press. ISBN 0 521 81388 3 ©SGM 2002]

[9] [http://mic.sgmjournals.org/content/147/12/3241/F4.expansion.html A. Khlebnikov, K. Datsenko, T. Skaug, B. Wanner and J. Keasling: Homogeneous expression of the P_BAD promoter in Escherichia coli by constitutive expression of the low-affinity high-capacity AraE transporter, Microbiology, 2001, (147): 3241–3247]

[10] [http://www.ncbi.nlm.nih.gov/pubmed/10829079 Datsenko, KA, BL Wanner: One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products, 2000 Proc. Natl. Acad. Sci. U.S.A. 97(12):6640-5]

[11] [http://www.nature.com/nature/journal/v434/n7037/full/nature03461.html Subhayu Basu, Yoram Gerchman, Cynthia H. Collins, Frances H. Arnold & Ron Weiss: A synthetic multicellular system for programmed pattern formation, Nature 2005, 434: 1130-11342]

[12] [http://jb.asm.org/cgi/content/abstract/182/3/805 Kristi A. Egland and E. P. Greenberg: Conversion of the Vibrio fischeri Transcriptional Activator, LuxR, to a Repressor, Journal of Bacteriology, 2000, 182: 805-811]

Pymol:

[13] [http://www.ncbi.nlm.nih.gov/Structure/mmdb/mmdbsrv.cgi?uid=49005 http://www.ncbi.nlm.nih.gov/Structure/mmdb/mmdbsrv.cgi?uid=49005 NCBI Protein Structure summary, 18.10.11]

[14] [http://www.ncbi.nlm.nih.gov/Structure/mmdb/mmdbsrv.cgi?uid=52511 http://www.ncbi.nlm.nih.gov/Structure/mmdb/mmdbsrv.cgi?uid=52511 NCBI Protein Structure summary, 18.10.11]