The measurement devices were prepared by amplifying the reporter (BBa_I20270) by PCR. The amplified insert was then cut with XbaI and PstI and ligated into the plasmid containing the promoter previously cut with SpEI and PstI. All the devices were confirmed by restriction digestion as well as DNA sequencing.

The reporter gene (i.e. GFPmut3b from part BBa_I13504) was amplified using the Phusion polymerase from New England Biolabs and primers matching the prefix and suffix.

We used 50 ng of template with an annealing temperature of 59 °C and an extension time of 90 seconds. The PCR was confirmed by electrophoresis and subsequently purified with NucleoSpin Gel and PCR Clean-Up Kit from Macherey-Nigel. 125 ng of purified PCR were digested with 1 μl of XbaI and 1 μl of PstI overnight at 37°C. The following morning the sample was treated with 1 μl of DpnI for 2 hour at 37°C and the enzymes were deactivated for 20 min at 80 °C.

The cells were thawed and resuspended in 3 ml of PBS. An aliquot of 150 μl of each sample was placed into a White, Flat-bottomed, 96-well Costar Plate (code: 3917) and fluorescence intensities were taken with a Tecan Infinite 200 Pro Plate Reader (made in Switzerland). Excitation wavelength and emission wavelength were 395 nm and 509 nm, respectively. The gain was optimized at 70 V and kept constant for each sample. PBS was used as blank. To obtain technical replicates, fluorescence intensities were acquired for three aliquots of the same biological sample, keeping the same instrumental conditions. The raw data were adjusted for the blank value and the means across the replicates with their relative standard deviation were plotted.

The cells were thaw and resuspended in 3 ml of PBS. Each measurement was taken in a clear acrylic cuvette (REF 67.755) with a Cary Eclipse Fluorescence Spectrophotometer (made in USA). The blank was done with PBS. The excitation wavelenght used was 395 nm and the spectra were acquired between 420 nm to 650 nm. The gain was optimized at 775 V and kept the same throughout the measurement.

The fluorescence intensity of the maximum emission wavelenght (514 nm) was adjusted for the blank and the calculated means with relative standard deviations were plotted.

Our characterization confirmed the relative strength of the promoters

J23101 is the strongest promoter among the three, showing high expression of GFP in all three strains, regardless of the technique used. J23106 is the medium promoter and J23117 the weakest.

Ratios across promoters are kept the same

J23101/J23106 fluorescence ratios ranged from 2.0 to 4.5, depending on the strain and the technique. Differently from the other two promoters, J23117 showed a very low GFP production, as it was not detectable by eye or using the trans-illuminator and showed little fluorescence with the three techniques used.

Different techniques lead to the same results, with different sensitivities

The best way to perform a characterization is to use various techniques. Throughout our experiments we saw that each instrument has a specific sensitivity, which alters the output data. The FACS happened to be the most accurate among all, due to its extremely high intrinsic sensitivity. The plate reader also showed a good accuracy while the fluorometer was not able to detect the weakest promoter from the background noise. The use of the qPCR machine as a spectrofluorometer also gave positive results.

In vitro conditions mimic the in vivo reality

Comparing the results obtained from the cell-free extract to the others, we discovered that the promoters behave the same when working in vitro or in living bacteria. Since testing constructs in vitro is much faster than in vivo, our results suggest that it may be wise to first screen parts and/or genetic circuitry in vitro. Then the activity of a smaller subset could be confirmed with in vivo measurements.

Looking at promoters from a different angle

In vitro characterization by qPCR allows for the quantification of promoter strength by measuring RNA transcript levels, rather than by looking at the concentrations of something one step removed, i.e. protein. This approach gives a a more direct measure of promoter strength.

Bacterial strain does matter

The three promoters behaved differently in the different bacterial strains used. The bacterial strain which gave the highest fluorescence was NEB10β, which showed significantly increased expression of protein in all cases when compared with JM109 and NEB Express. Clearly this discordance among strains must to be due to their different genotypes. A different bacterial proteome (e.g. presence/lack of specific proteases and/or chaperonins) may alter protein production, processing and folding, and thus fluorescence emission.