Team:UCSF/Project/Circuit/Data

From 2013.igem.org

CRISPRi Decision-Making Circuit: Promoter Design and Data


Summary: We have designed a CRISPRi circuit that can choose between two outcomes based on the level of input This circuit is also scalable (i.e. many parallel circuits can be deployed in the same cell) due to the use of CRISPRi gRNA's
Before we could create promoters that respond to a high and low amount of inducer, we first had to select promoters with two favorable characteristics we could exploit to make them concentration-dependent which would be a low basal level and a wide dynamic range. We chose four well-known promoters as possible candidates for our circuit: pLAC, pTet, pBAD, and PprpB. We constructed separate plasmids containing each of the promoters and characterized these promoters by measuring the GFP expression over time. We determined the basal level, highest induction level, and inducer range. We selected pLAC & pTET as the promoters for the synthetic circuit, because of their low basal expression and the varying responsiveness at different inducer concentrations.

Dose-response curve for pLAC promoter induced with different amount of inducers (0, 0.1, 1, 10, 20, 40, 60, 80, 120 uM IPTG). Cells were grow to mid-log phase and then start induction. OD600 value and GFP fluorescence level of each sample were measured by plate reader after saturation. GFP fluorescence were corrected for OD600 value. The red line indicated Hill function fit of the dose-response curve and error bars indicate standard deviation calculated on the basis of technical replicates.


Dose-response curve for pTET promoter induced with different amount of inducers (0, 0.004, 0.01, 0.04, 0.06, 0.11 uM aTc). Cells were grow to mid-log phase and then start induction. OD600 value and GFP fluorescence level of each sample were measured by plate reader after saturation. GFP fluorescence were corrected for OD600 value. The red line indicated Hill function fit of the dose-response curve and error bars indicate standard deviation calculated on the basis of technical replicates.
Engineering pLAC Sensitivity

Our next goal after our promoter assays was to create engineered versions of our promoters responsive to high and low levels of inducer, either by modifying activation or repression of the specific promoter. Based on previous work in the literature characterizing the pLAC promoter, we chose to change both the number and orientation of repressor binding sites in pLAC.
To create our "low" inducer promoter, we shifted the O1 operator site further downstream of our promoter sequence and removed another operator site from pLAC to relieve repression. To create our "high" promoter, we took the "low" promoter and added the O3 operator site further upstream near the start of the promoter sequence. The O3 operator site will then form a loop with our O1 operator site, thus creating a physical barrier to prevent transcription, reducing the promoter activity at low levels of inducer.
On this diagram (A) denotes the very original pLAC which is found in the parts registry. (B) is our re-engineered pLAC that has an added O3 binding site. We found in literature that by adding an O3 binding site, a higher inducer level is required to start transcription for a loop to form between the 03 and 01 binding sites. We expect we would need more inducer to produce similar levels of GFP.
When we engineered the pLAC promoter, the dynamic range is insignificantly different. This graph shows that the green fluorescent protein is expressing at the same inducer concentration for both of the promoters. So we went back to the drawing board and redesigned the pLAC promoter.
Improved Designs for High & Low Sensitivity Promoters

In the diagram below, (C) is our low promoter and (D) is our high promoter. To create our low inducer promoter, we shifted the O1 operator site further downstream of our promoter sequence and removed another operator site from pLAC to relieve repression. To create our high promoter, we took the low promoter and added the O3 operator site further upstream near the start of the promoter sequence. The O3 operator site will then form a loop with our O1 operator site, thus creating a physical barrier to prevent transcription, we require higher levels of inducer for expression.
The sequence of four different pLAC promoters constructed for the purpose of high and low sensor. (A) was the original pLAC promoter, (B) was the engineered one that have reduced expression level, (C) and (D) was the expected low and high sensor respectively. Yellow region of each promoter indicated its -35 and -10 core region for RNA polymerase binding, the red characters of each promoter indicates +1 start point of transcription, and the blue square indicated different LacI binding sites (LacO1 and LacO3).
Our second round of promoter engineering shows that the outputs are close to our desired outputs. However, the sensitivity of the promoters did not change. Though we did not get our desired high and low promoters, we have created new promoters that will be available at the parts registry and have been fully characterized.

We also created a mathematical model of our system and were able to identify important variables and parameters that would help produce our decision making circuit.

CRISPRi Decision-Making Circuit: Circuit Output

The essential benefit of using the CRISPR system for the circuit design is that it has a very specific target -- a guideRNA contains nucleotides which are complementary to a unique DNA sequence. As a result, we can use a vast number of individual gRNAs to repress several genes, giving us the ability to design circuits to make numerous and more complex decisions.

Our proof-of-concept design is, at a low level of inducer, to express GFP simultaneously with a gRNA to repress RFP. When the inducer level increases, the "high" sensing promoter turns on expression of RFP and a gRNA to repress GFP, making a switch-like decision based on the input. We were able to construct an early design of the circuit in the pCDF plasmid that contains both fluorescent proteins and gRNAs. We initially put both components of the circuit under individual promoters, pLAC and pTET, as a control to test whether or not we could obtain expression of both products and repression from the gRNAs, both under individual induction and after switching inducers. Since this circuit relies on CRISPR repression, we had to also construct a plasmid (pACYC) containing the catalytically dead dCas9 protein as well as the Csy4 protein to cleave the RNA products (see Design page or Parts Registry for more information).

We are in the final stages of cloning the dCas9 plasmid and we hope to test the circuit soon.