University of Pennsylvania iGEM

PENN iGEM 2015




An effective light-based communication system rests on the bioluminesence generated by the “sender cell.” In order to design a well-functioning system, the Penn 2015 iGEM team worked to optimize the light output of various E.coli “sender cells” transformed with the lux operon (BBa_K325909).

Lux operon expression is responsible for bioluminescence. The operon is initiated by a constitutive promoter (BBa_J23100) followed by an RBS + lux box. The box contains the following: LuxC, D, A, B, E and G. LuxA and B encode two subunits of bacterial luciferase. The genes LuxC, D, and E drive expression of the substrate for the light-emitting reaction, tetradecanal. The function of the luxG gene is yet to be fully elucidated; however, inclusion of the gene is known to increase light output (Craney et al. 2007). The circuit is completed with a stop codon and a terminator sequence. The operon is induced by arabinose.

After refining some of our growth conditions, we were able to achieve visible luminescence in our bacteria (pictured below).


We began our exploration of the sender-receiver system by examining three different sender systems:

  1. HNS BW25113 Dhns::kan strain with lux biobrick. Because the H-NS protein has been shown to repress the Lux genes, we transformed the lux operon BioBrick (BBa_K325909) into this H-NS knockout E. coli strain.
  2. NEB10 with lux biobrick. We chose to explore the light output of this strain as teams that have worked with the lux biobrick previously have chosen NEB10 as their chassis.
  3. SY104. This strain contains a "split-lux" system to decrease the genetic payload controlled by a single promoter. Gene expression of lux AB is controlled by the sulA inducible promoter, and lux CDE expression is controlled by CP25 constitutive promoter.

We conducted at 16 hour time course at which we measured the light output of each of the aforementioned systems.

Figure 1. Normalized luminescence was calculated as a ratio of luminescence of a culture at a time point to OD600 of the culture at that time point. Error bars represent standard deviation from the mean, and sample size was n=3 for all strains and time points.Trend lines for each strain were created using MATLAB's curve fitting tool. All points after the peak luminescence were modeled using the exponential equation f(x)=a*exp(b*x) + c*exp(d*x). Points leading up to peak luminescence were modeled with a linear equation since due to low resolution. For coefficients see the materials and methods section included at the bottom of the page.

From these key take away points multiple suggestions about the future usage of these strains can be made. In fact, these three strains can be split into two categories: (1) Sustained Expression, and (2) Transient Expression.


The most noteworthy observation of SY104’s expression of bioluminescence is its sustained expression. Even at 16 hours, cultures of this sender were still emitting photons. This type of behavior though it provides simpler modeling, is only advantageous in specific systems. Prolonged expression is beneficial in light-mediated communication when the receiver is extremely sensitive and does not reach its half-saturation point quickly. Such a receiver is responsive to low levels of luminescence produced at earlier time points. However, with a slower time to half-saturation this type of receiver must be exposed to light for a longer amount of time in order to be activated fully. Thus, making SY104 a perfect match for this sort of receiver in light-mediated cell communication.


Unlike SY104, both HNS and NEB10 senders demonstrate a rapid, high peak in luminescence followed by quick decay approximately one log order lower in magnitude. Both strain's peak normalized luminescence is considerably higher than any value that SY104 reaches. Therefore, HNS and NEB10 can be useful in sender-receiver systems where the receiver has an extremely high half-saturation time and is less sensitive to lower luminescence levels. With these light-sensitive systems, it is important to quickly pack it with light of a high enough intensity of light for successful activation. Thus, a sender demonstrating sustained expression is not ideal in this case scenario, as it provides luminescence in early stages post-induction.

Note that differences between the sender strains does not necessarily mean that one is better than another. Rather, this data highlights characteristics of each sender that would make it suitable in certain sender-receiver systems.


Cultures of all three were grown to saturation, back diluted 1:100 , and, at an O.D.600 reading of ~.4, separated into twelve tubes (four per strain). Three tubes for each strain were induced with either arabinose for the lux operon BioBrick-containing H-NS and NEB10 strains and nalidixic acid for the SY104 sender strain. The final concentration of inducer was 10 mg/L for SY104 and .01 M for H-NS and NEB10. One tube per strain remained as an un-induced negative control. The luminescence in RLU’s with a 1000ms integration time and the O.D. at 600 nm was measured every two hours for sixteen hours after induction on a Tecan Infinite m200.

Coefficients for Best Fit Models