Biomolecular Breadboards:Protocols:Cycle Time Measurements

Home

Protocols

DNA parts

Preliminary Data

Models

More Info

Design cycle time analysis: Testing simple circuits on the cell-free breadboard

Creating a functioning biocircuit can be a lengthy process, involving multiple iterations of design, assembly, and testing. As such, there is general interest in finding methods to do this faster. Using the cell-free circuit breadboard as a testing platform presents some unique shortcuts in assembly and troubleshooting times, which are only possible in vitro. Freedom from the in vivo limitation of having to assemble a circuit upon a single or pair of plasmids allows a wider array of circuit design possibilities, such as fewer components spread over more plasmids, and circuits constructed out of linear DNA. As will be shown below, these non-traditional systems often have faster design times, resulting in quicker iterations of the design, assembly, test process. An in vitro platform also makes tests easy that would be far more difficult in vivo. For instance, DNA concentrations can be precisely and simply controlled in vitro, by adding only the desired amount of DNA to the system. In vivo systems rely upon plasmid copy number, which can be variable and to be changed requires a different origin of replication.

To demonstrate assembly-to-testing times in the cell-free circuit breadboard, we've chosen two circuits: the negatively autoregulated gene, and the bistable toggle switch. The first test of the negatively autoregulated gene in the TXTL system took only 6.5 hours from start of assembly. The more complicated bistable switch was run on a system of 4 plasmids in only 3 days.

	Assembly time for in vitro testing	Assembly time for in vivo testing (1 plasmid construct)
Negatively autoregulated gene	6.5 hours	3 days
Bistable switch	3 days	5 days minimum *

( * Estimated. All other times are measured. )

Negatively Autoregulated Gene

The first construct we tested was a simple 3-component circuit, consisting of the pTet promoter in front of a TetR-deGFP fusion protein. The circuit functions as a negative feedback loop, with its product repressing its own production. It was first constructed by Becskei, J & Serrano, L. (2000)¹ and again described by Rosenfeld, N., Elowitz, M.B. & Alon, U. (2002)².

Figure 1. To the left, the linear form of the circuit, as tested in TXTL. To the right, the 4 components used to create it: a plasmid backbone (pBEST plasmid), the promoter pTet, the TetR repressor and the deGFP reporter protein. To fuse TetR and deGFP, Gibson3 primers were designed to code for a glycine-serine linker, parts registry BBa_I717014.

1) Fastest method - same day testing with linear DNA

Designing the circuit and ordering the primers took about 1 day. We received the primers at noon the day after ordering, and started assembly immediately. For both the negatively autoregulated gene and the bistable switch, there was no wait time for parts as we used DNA components from plasmids we already had in lab. To create the circuit, we used PCR to amplify 4 components, each from a different plasmid. We ran a gel to check the success of the PCR, and then kit purified the product. We then combined the products with Gibson assembly (cite). Traditionally, for in vivo testing the gibson product is transformed into competent cells, and then plated for overnight growth. However, since we can run linear DNA in in vitro the Gibson product can also be PCR-amplified as a template. After another gel to check for success, we kit purified the product and ran the circuit 8 hours in TXTL. The results of the linear DNA run is shown below to the right. GamS, an inhibitor of the RecBCD complex, was added to the sample to prevent linear DNA degradation.

Step Time PCR to add Gibson ends 2 hours Gel 0.5 hours PCR or gel clean-up kit 0.25 hours Gibson assembly 1 hour PCR using Gibson product as template 2 hours Gel 0.5 hours PCR clean-up kit 0.25 hours Total time pre-TXTL 6.5 hours Fill in TXTL spreadsheet 1 hour TXTL reaction 8 hours / overnight

Figure 2. To the left: the time breakdown of the construction of the negatively autoregulated gene, from assembly to cell-free circuit breadboard. To the right: the results of a cell-free breadboard run with linear DNA that was PCR amplified off Gibson assembly product. Linear DNA at a concentration of 5 nM was run with gamS and both with and without the addition of 5 ug/mL anhydrotetracycline (aTc), a tetracycline analog that acts as an inducer by binding to and inhibiting the repressor protein TetR.

2) Three days to testing, standard plasmid circuit

To test if the results from the cell-free breadboard run of the linear negatively autoregulated gene (Figure 2) were an accurate gauge of the performance of the circuit in plasmid form, we also transformed and plated the Gibson product into competent cells. The transformed bacteria were then incubated overnight, and colonies were screened by colony PCR to test for successful Gibson assembly. Colonies that contained the correct insert were then grown up in liquid cultures overnight. The next day, the plasmids were extracted with a plasmid purification kit, and then cleaned up with a PCR clean-up kit. Total assembly time took 3 days.

Day 1	Day 2	Day 3
Step Time PCR to add Gibson ends 2 hours Gel 0.5 hours PCR or gel clean-up kit 0.5 hours Gibson assembly 1 hour Transformation and plating 0.5 - 2 hours Grow colonies Overnight	Step Time Colony PCR 2 hours Gel 0.5 hours Grow up minicultures Overnight	Step Time Plasmid purification kit 1 hour PCR purification kit 0.5 hours Total time pre-TXTL 3 days Fill in TXTL spreadsheet 1 hour TXTL reaction 8 hours / overnight

Figure 3. Time breakdown of the construction of the negatively autoregulated gene on one plasmid, from assembly to testing on the cell-free circuit breadboard. Circuit assembly took 3 days, with the cell-free breadboard run on the third day and results gathered that same evening.

Figure 4. A plasmid encoding the negatively autoregulated switch, run in the cell-free circuit breadboard. To the left: 10 nM of plasmid, run with and without 5 ug/mL of inducer aTc. To the right: endpoint fluorescence at 8 hours of 2 nM - 19 nM of plasmid, with 5 ug/mL aTc. The ability in the cell-free circuit breadboard to vary the amount of plasmid added to the system permits the exploration of the effects of plasmid concentration on circuit function. Here, as the plasmid concentration initially increases, so does signal. However, as the inducer molecule is sequestered by TetR and its concentration drops, TetR starts to repress protein production and the signal drops as the circuit begins to negatively autoregulate.

Bistable Toggle Switch

The bistable switch was intitally constructed by Gardner, T.S., Cantor, C.R. & Collins, J.J. (2000)⁴. It consists of two promoters, pTet and pLac, each of which sits in front of the gene coding for the repressor of the other promoter. That is, pTet drives LacI production, and pLac promotes TetR. In the Gardner-Collins switch, shown to the far left in the figure below, LacI is followed by the reporter protein GFP. It is driven by the same promoter as LacI, and thus "turns on" when LacI is produced, and "turns off" when LacI is not. Due to the nature of the switch, only pTet or pLac can be on at one time, since each represses the other. pTet can be turned on by the addition of IPTG (Isopropyl β-D-1-thiogalactopyranoside), an allolactose analog, and pLac by the addition of aTc.

Figure 5. To the left: the bistable toggle switch, on a single plasmid. The construction parallels that in Gardner, Cantor & Collins (2000). In the middle: The four plasmid system we used to test the bistable toggle switch. Each plasmid has one promoter-gene pair. To the right, the linearized form of the bistable switch. Assembly of circuits on linear DNA can cut down traditional assembly times, by removing steps necessary for plasmid assembly such as tranformation, plating, and/or miniculture.

1) Estimated minimum five days to testing, standard plasmid circuit

Constructing the entire bistable switch on one plasmid would require the assembly of at least five different parts, put into a sixth component, a vector backbone. Below, we've calculated the estimated time needed to construct such a circuit, using a standard method of multi-part DNA assembly, Gibson assembly. At least two different rounds of Gibson assembly would be required, making the time needed at least 5 days. This is the same time recently recorded for the assembly of the bistable toggle in Litcofsky, K.D. et al Nature Methods⁵. * Richard - this is an advance publication - can we reference this?

Day 1	Day 2	Day 3
Step Time PCR to add Gibson ends 2 hours Gel 0.5 hours PCR or gel clean-up kit 0.5 hours Gibson assembly 1 hour Transformation and plating 0.5 - 2 hours Grow colonies Overnight	Step Time Colony PCR 2 hours Gel 0.5 hours Grow up minicultures Overnight	Step Time Plasmid purification kit 1 hour PCR to add Gibson ends 2 hours Gel 0.5 hours PCR or gel clean-up kit 0.5 hours Gibson assembly 1 hour Transformation and plating 0.5 - 2 hours Grow colonies Overnight
Day 4	Day 5
Step Time Colony PCR 2 hours Gel 0.5 hours Grow up minicultures Overnight	Step Time Plasmid purification kit 1 hour PCR clean-up kit 0.5 hours Total time pre-TXTL 5 days Fill in TXTL spreadsheet 1 hour TXTL reaction 8 hours / overnight

Figure 6. Calculated breakdown of the time needed to construct the bistable switch on one plasmid, using 2 rounds of Gibson assembly. Estimated minimum 5 day total assembly time.

2) Three days to testing, 4 plasmid circuit

Instead of constructing the bistable switch via the traditional one plasmid method described above, we decided to take advantage of the lack of living bacteria in our testing platform. Bacteria can only hold a few plasmids at a time, and thus circuits designed for in vivo testing are generally constrained to 1-2 plasmids. However, in vivo testing has no such limitation, making circuit design more flexible. We chose to assemble the bistable switch onto 4 plasmids, each with one translational unit (a promoter-gene pair). This method required only one round of Gibson assembly per plasmid, cutting construction time down to 3 days. Additionally, having components spread over 4 plasmids reduced design times when troubleshooting, since swapping out parts only required redesign and assembly of a 2 component plasmid, instead of the 5 component assembly that would have been necessary for a one plasmid system.

The time breakdown for the bistable switch assembly is shown in Figure 3, since it was the same as the construction of the plasmid for the negatively autoregulated gene. Although there were 4 plasmids to be assembled, their assemblies could all be run in parallel, so the increase in number of plasmids from 1 to 4 did not affect the assembly time.

Figure 7. "Switching" behavior of the 4 plasmid bistable toggle system, run on the cell-free circuit breadboard. To aid in visualization of the circuit's "switching," one of our 4 plasmids contains a second reporter protein, CFP, placed behind the pLac promoter. On the left, 500 uM of IPTG has been added to the sample, and CFP is turned on while GFP is turned off. On the right, 5 ug/mL of aTc has been added instead, and GFP is on while CFP is off. All plasmids are at 2 nM concentration.

3) Two days to testing, Linear DNA

As demonstrated in section 1 of the negatively autoregulated gene, above, it is possible to test constructs in just one day, by PCR amplifying Gibson assembly product and running the kit cleaned PCR products in the cell-free circuit breadboard. Another shortcut, which also takes advantage of the ability to use linear DNA in the TXTL reaction, is to use the product of the colony PCR for testing. Normally, colony PCR is a diagnostic test, used to determine which colonies took up the correct Gibson assembly product. However, the PCR products can also be kit purified, and run in the breadboard.

Day 1	Day 2
Step Time PCR to add Gibson ends 2 hours Gel 0.5 hours PCR clean-up kit or gel purification 0.5 hours Gibson assembly 1 hour Transformation and plating 0.5 hours - 2 hours Colony growth Overnight	Step Time Colony PCR 2 hours Gel 0.5 hours PCR clean-up kit 0.5 hours Total time pre-TXTL 2 days Fill in TXTL spreadsheet 1 hour TXTL reaction 8 hours / overnight

{*****************graph*********************}

Figure 8. To the left: the time breakdown to construct a linearized component of the bistable switch, the pTet-LacI translational unit. The same breakdown and total time would apply when using this method to construct the whole switch, in linearized form. The reduction of a day in assembly time from plasmid to linear comes from not having to grow the overnight miniculture. To the right: when experimenting with different ribosomal binding sites (RBS) for the pTet-LacI plasmid, we used the products from colony PCR to test whether the new RBS was stronger than the previous. Linear DNA encoding pTet-LacI with the previous RBS was run at 10 nM against the PCR product with the new RBS, also at 10 nM. Atc was present at 5 ug/mL. GamS, an inhibitor of the RecBCD complex, was added to the sample to prevent linear DNA degradation.

---

1. Becskei, A. & Serrano, L. (2000). Engineering stability in gene networks by autoregulation. Nature, 405, 590 – 593.

2. Rosenfeld, N., Elowitz, M.B. & Alon, U. (2002). Negative autoregulation speeds the response times of transcription networks. J. Mol. Biol., 323, 785–793.

3. Gibson, D.G. et al. (2009). Enzymatic assembly of DNA molecules up to several hundred kilobases. Nature Methods, 6(5): 343–345.

4. Gardner, T. S., Cantor, C. R. & Collins, J. J. (2000). Construction of a genetic toggle switch in Escherichia coli. Nature, 403, 339–342.

5. Litcofsky, K.D. et al. (2012). Iterative plug-and-play methodology for constructing and modifying synthetic gene networks. Nature Methods, Advanced Online Communication.

6. Geneious® Pro 5.6.4 created by Biomatters.