Synthetic biology relies on the characterization of biological parts, a process that can vary among different laboratories. Even though one protocol can be followed in the exact same way, small variations between equipment, material handling and environmental conditions (e.g., lab temperature) can skew results. The Sixth InterLaboratory Measurement Study will be continuing the iGEM initiative that started in 2014 in an effort to minimize sources of variability in Synthetic Biology measurements. This year, with a combination of three experiments, the Engineering Committee gave the chance to iGEM teams to contribute meaningfully to this endeavor, by conducting these experiments in their own laboratories with materials provided through the iGEM Distribution Kit.The goal was to tackle the problem of multiple color (green, red, and blue) fluorescence data comparison between labs, and our team took part in the first two experiments, focusing on three-color calibration and the influence of transcriptional unit order in expression strength.

Before conducting any of the actual experiments, we were asked to calibrate our instruments with materials provided by iGEM. Three fluorescent dyes and silica bead nanoparticles of known concentration were used to prepare a 96 well plate containing serial dilutions of each calibrant (Fig.1). Four standard curves were produced, one for each dye and one for the beads. With the green standard curve, we could transform the plate reader measurements, from arbitrary units to Molecules of Equivalent Fluorescein (MEFL) for the green dye, according to the procedure followed in the 2018 Interlab study and described here1.

A dataset of fluorescence values divided by absorbance was prepared for each color, to normalize data across all measurements. Then, we took the average value of each colony, of the LB+C and the empty column. The value of the empty column was subtracted from each colony’s average and the data were visualized to give a comparison of each device’s expression and fluorescence strength.

Figure 1. Graphs of the fluorescence/absorbance per device for the three colors, for both experiments. The measurements at the 0 hour and 6 hour timepoints are compared. 0 hour measurements are heavily skewed because of LB autofluorescence. As the bacteria grew, the ratio of FL/ABS decreased, but both absolute FL and ABS increased.

While many of the devices were successfully transformed at once, some of them were particularly troublesome and were the cause of much delay to our experiments. These were the devices at the wells 14E, 12M, 12I, 12G and 12O. After several attempts at transforming the bacteria (with the extra challenge of having a very limited amount of DNA available – no more than 3ng), we managed to get colonies for all the devices. However, while examining the results, and as is clear in the figures, the Green Blue Device (well 12G, Device 6 in Experiment 1 and 1 in Experiment 2) did not work for us. Even though we know from personal communication with other iGEM teams that Experiment 1 and 2 Devices were generally difficult to transform, we can only assume that it is we who did something wrong throughout our experiments, most probably in the colony picking process or DNA resuspension from the Distribution Kit.

Another point that must be made is the apparent systematic inaccuracy during the dilution phase to final OD of 0,02. Even though the exact steps stated at the protocol were followed, differences in cuvette condition and quality, and the unreliability of our instrument’s measurements (our spectrophotometer is very old) resulted in very different absorbance values among aliquots. This became evident during the absorbance measurement stage from our plate reader.
Finally, we found very interesting the fact that the Luria Bertani medium seems to be fluorescent on its own, even without fluorescent protein-producing bacteria. Initially we thought that we had done something wrong, but after browsing through the internet and getting advice from older lab members we stopped worrying. We have reported the average values of LB fluorescence per color on Figure 2.

Figure 2. The average values of LB fluorescence/absorbance for all colors.

The eleven different devices (Tables Y and Z) for Experiments 1 and 2 were located in the first DNA Kit of the Distribution Kit. 10 uL of freshly sterilized ddH2O were used to resuspend DNA (1-3 ng per well, so 100-300 pg/uL). 1-3 (up to 5 for some devices) uL were used to transform DH5α E. coli bacteria, according to this protocol. All dilutions and plate preparations were performed in the cold room (~4 oC). The 1:10 dilution of each culture was performed in a final volume of 2ml (and not 12ml as stated in the protocol). The 12ml, OD of 0,02 tubes were placed in the incubator covered in aluminum foil. The same plates (black with transparent bottom) were used for both calibration and experiments. They were used multiple times after washing with ethanol, tap water and dH2O. All measurements were taken in the Varioskan Lux (Thermo Fisher Scientific) plate reader.

We are thankful to the iGEM Engineering Committee for giving us the opportunity to partake in this project.