Interlab Study

Introduction

While we were at the laboratory developing our project, we also participated to the 2016 Interlab Study. The interlab study consists in measuring the fluorescence level of constructions provided by the iGEM Measurement Committee in order to compare results obtained by worldwide iGEM teams and thus study the variations of measurements among each experiments. This year, it consisted in measuring the fluorescence of three test devices composed of a Green Fluorescent Protein (GFP) coding sequence under the control of promoters of different strengths. The measurements could be proceeded using a plate reader or flow cytometry. We chose to use flow cytometry which is available at the laboratory where our team works.

Constructions

• "Test Device 1" is composed of a strong constitutive promotor (J23101), a RBS, wild type GFP gene and a double terminator cloned into the plasmid pSB1C3.
• "Test Device 2" is composed of a medium strength constitutive promotor (J23106), a RBS, wild type GFP gene and a double terminator cloned into the plasmid pSB1C3.
• "Test Device 3" is composed of a week constitutive promotor (J23117), a RBS, wild type GFP gene and a double terminator cloned into the plasmid pSB1C3.
• "Positive control Device" is composed of a constitutive promotor (J23151), a RBS, wild type GFP gene and a double terminator cloned into the plasmid pSB1C3.
• "Negative control Device " is only composed of the repressible promotor of the TetR gene cloned into the plasmid pSB1C3.

Methods

At the beginning of the Interlab study, we had a problem with device 1, the received tube was empty. We waited to receive another tube from iGEM.

Constructions test devices 2 and 3 and the two controls were transformed into competent DH5α E.coli strand using a heat shock transformation protocol. Transformed bacteria were plated on solid LB medium containing 30 μg/mL chloramphenicol. Petri dishes were incubated at 37°C overnight. For each device, a colony was used to inoculate 3 mL of liquid LB medium containing 30 μg/mL chloramphenicol. The cultures were incubated at 37°C at 180 rpm overnight. Then a glycerol stock was made from these overnight cultures and stored at -80°C. When we received the device 1, we used the same protocol to clone it into the DH5α E.coli strand.

Glycerol stocks were plated on solid LB medium containing 30 μg/mL chloramphenicol, and incubated the cultures at 37°C overnight. For each construction, two colonies were randomly picked up to inoculate two different tubes containing 5 mL of liquid LB medium containing 30 μg/mL chloramphenicol.

Then we used those tubes to perform flow cytometry. We used the "Cube 8" cytometer from the PARTEC Company. Cells were excited by a 488 nm laser, and we detected fluorescence emission using a 536/40 filter. For each sample, around 1 million cells were counted. Data were obtained in arbitrary units, since we did not have any calibration beads.

Assessment

The size of cells (FSC) should be the same for every sample as it is the same bacterial strand and only the fluorescence emission level (FL1) should vary. We expected fluorescent emission to be correlated to promoter strength for each construction as the promoter strength has an influence on the expression level of the GFP gene and fluorescence is proportional to GFP quantity in the cell. However it is important to keep in mind that even if GFP level in the cell might be correlated to promoter strength, it exists stochasticity on such expression level (Elowitz, 2002).

Results

Fig. 1: Size of the different bacterial clones This figures shows cells size according to the number of cells counted for samples #1. Pink: negative control; black: positive control; yellow: test device 1; blue: test device 2; green: test device 3.

Fig. 2: Size of the different bacterial clones This figure shows cells size according to the number of cells counted for samples #2. Pink: negative control; black: positive control; yellow: test device 1; blue: test device 2; green: test device 3.

Fig. 1 and Fig. 2 show that cell population is homogenous between every sample. Cell size is between 10² and 10³ for each sample, and the number of counted cells is almost the same for every sample.

Fig. 3: GFP fluorescence intensity This figure shows GFP fluorescence intensity according to the number of cells counted for samples #1. Pink: negative control; black: positive control; yellow: test device 1; blue: test device 2; green: test device 3.

Fig. 4: GFP fluorescence intensity This figure shows GFP fluorescence intensity according to the number of cells counted for samples #2. Pink: negative control; black: positive control; yellow: test device 1; blue: test device 2; green: test device 3.

Fig. 3 and Fig. 4 show that fluorescence intensity is correlated to promoter strength of each device. The fluorescence emission level of device 1 is more important than device 2. Device 2 presents a more important fluorescence emission level than device 3. The positive control show a large range of fluorescence intensity, containing two peaks. This probably means that there might be two subpopulations, expressing GFP at different levels. In other terms, two colonies were probably picked and inoculated instead of one in the liquid medium. Those two subpopulations probably don't have the same number of plasmids inside each cell, which leads to different fluorescence emission intensity. However, positive control fluorescence level is around the same as device 2. As expected, negative control does not show any significant fluorescence emission.

Table 1: Detailed cytometry data

This table shows detailed measurment data for each sample. Fluorescence measures are presented in arbitrary units.

Detailed data (Table I) show that fluorescence intensity results does not vary between two samples of the same construction, except for the positive control. This observation has to be linked with the fact that the positive control cells population might not be homogeneous.

Discussion

We were not able to give the fluorescence measurment results in absolute units, because we did not have any calibration beads that would have allowed us to transform arbitrary units into absolute units.

The interlab study has been a challenging experience that we have enjoyed be part of. We are hopping the results we have obtained will be useful and will help to the understanding of experiment reproducibility.

Reference

Elowitz MB. Stochastic Gene Expression in a Single Cell. Science. 16 août 2002;297(5584):1183‑6. http://www.sciencemag.org/cgi/doi/10.1126/science.1070919