Team:TAU Israel/Improve

"If I have seen further, it is by standing upon the shoulders of giants"
Sir Isaac Newton

Meet The Protein

Research History

Red fluorescent protein or RFP for short, is a popular fluorophore in biology labs, usually used as a reporter protein. The original RFP was isolated from Discosoma and was named DsRed [1]. Today, RFP variants are being used as reporters in many fields of study, such as cancer research [2 & 3], genetics research [4 & 5] and antibiotic research [6].

The high popularity of the RFP has led us to aim for the expression improvement of RFP in an existing part, in order to enable easier, more efficient, future detection of changes in expression.

Links to our parts in the registery:


Composite part (Variation 1)

RBS

RFP

Our Part Improvement

Part Background

We have decided to improve an existing composite part (BBa_K608014) constructed of a strong promoter, a medium RBS and a monomer RFP (mRFP). The mRFP used in this part is one of the most popular coding sequences in iGEM, with over 500 reported uses.


Why did we choose to improve BBa_K608014?

In a previous experiment (by iGEM11_Freiburg), BBa_K608014 has shown the highest fluorescence level compared to five variations of promoters and RBS with mRFP. Due to the high fluorescence level, we decided to try and improve this variation, aiming to take its performance one step higher.


Bioinformatics

In order to detect possible expression improvements, we used bioinformatics tools that allowed us to obtain synonymous mutations without altering the amino acid sequence. The methods that we used are:

1. Improving the Shine-Dalgarno sequence: The Shine-Dalgarno (SD) sequence is responsible for the connection of the small subunit of the ribosome to the mRNA and the recognition of the START codon, and therefore has a direct influence on the translation initiation efficiency. In order to improve the translation initiation efficiency, we have changed the given SD sequence to the canonical optimal sequence which matches to the anti-SD sequence that can be found on the ribosomal RNA. In addition, we changed the distance between the SD sequence and the start codon, so it will be also optimal; this was done based on mimicking the most common distance found in E. coli endogenous genes. We have done so by inserting two random nucleotides (that were later optimized; see point 2 below) at the beginning of the gap between the SD sequence and the start codon.

2. Improving the mRNA’s folding: The mRNA free folding energy surrounding the START codon determines, among other things, the efficiency of recognizing the START codon and the SD sequence by the small ribosomal subunit. Stronger folding, which is related to more negative free energy, decreases this efficiency since the small ribosomal subunit can't bind to nucleotides that are inside a structure. Thus, our goal was to lower the folding energy (making it closer to zero). We used the ViennaRNA[7] tool in order to find the free energy of sliding windows in the size of 40 nucleotides and calculated the average free energy of the entire mRNA. Then, we looked for synonymous mutations that will reduce the free energy without altering the amino acid sequence of the RFP. Notice that there is an enormous number of possible combinations of synonymous mutations (exponential with the length of the protein), and therefore a heuristic was needed. In order to explore those possibilities and to find the sequence with the best folding energy, we used a genetic algorithm, imitating the natural selection process.


We decided to try three variations of sequences:

Variation 1: Containing both SD and folding optimization.

Variation 2: Containing only the folding optimization.

Variation 3: Containing another possible variation of the SD optimization and folding optimization.

For more information on the bioinformatics methods used to improve the part, click here

Wet Lab

We started our lab work by transforming the existing part (BBa_K608014) from the distribution kit to E. coli DH10beta. Then, after growing starters from the transformation plates overnight, we extracted the plasmids using miniprep.

Using PCR, we amplified part of the plasmid, excluding part of the mRFP and RBS. This enabled us to insert the possible improved sequences (ordered as G Blocks) into the plasmid, which already included the appropriate backbone and overlaps.

The sequences were inserted using Gibson Assembly. Before the assembly, we have done DPN1 to ensure no unwanted DNA template was present. Then, we proceeded to the Gibson reaction and transformed the resulting plasmid into E. coli DH10beta.

Unfortunately, the transformation of variation 3 failed, therefore we did not add it to the fluorescence experiment.

Fluorescence Experiment

Methods

In order to measure the fluorescence of each plasmid, we have used a plate reader machine. The plate contained 3 samples of each of the next bacteria:


-DH10beta containing a non fluorescent plasmid.

-DH10beta containing the original part.

-DH10beta containing a plasmid with variation 1 (SD and folding optimization).

-DH10beta containing a plasmid with variation 2 (folding optimization).

- We also included 3 samples of pure LB in order to use it as blank.


Because RFP can absorb light at 600nm and lead to false measurements, concentration of bacteria was estimated using OD700 as recommended previously by the iGEM committee[8]. For measuring the protein's fluorescence, we used excitation wavelength of 540 nm and emission wavelength of 650 nm.


Calculation of the fluorescence intensity: We first measured individual sample intensity as:

Individual Intensity = (florescence of plasmid - florescence of LB) / # of colonies in OD700

Then, for each type of bacteria, we calculated the average intensity of the three relevant samples.


Results

We received the following fluorescence intensity (Fig.1):
-Non fluorescent bacteria - Mean: 26,766.72; STDEV: 3423.98.
-Original part bacteria - Mean: 57,853.17; STDEV: 2624.75.
-Variation 1 bacteria - Mean: 154,336.4; STDEV: 8994.96.
-Variation 2 bacteria - Mean: 50,705.19; STDEV: 2436.38.

The intensity of variation 1 (both Shine-Dalgarno and folding optimization, BBa_K2943902) was almost 3 times stronger than the original part (BBa_K608014). However, the intensity of variation 2 (folding optimization) was not higher than the original part.


These results have shown that our part improvement was a success, enabling stronger RFP expression for future studies.



Wet Lab Notebook:




Reference:

[1]. Gross, Larry A., et al. "The structure of the chromophore within DsRed, a red fluorescent protein from coral." Proceedings of the National Academy of Sciences 97.22 (2000): 11990-11995.
[2]. Suetsugu, Atsushi, et al. "Imaging exosome transfer from breast cancer cells to stroma at metastatic sites in orthotopic nude-mouse models." Advanced drug delivery reviews 65.3 (2013): 383-390.
[3]. Kim, Mi-Young, et al. "Tumor self-seeding by circulating cancer cells." Cell 139.7 (2009): 1315-1326.
[4]. Gyorgy, Andras, et al. "Isocost lines describe the cellular economy of genetic circuits." Biophysical journal 109.3 (2015): 639-646.
[5]. Buchler, N. E., & Cross, F. R. (2009). Protein sequestration generates a flexible ultrasensitive response in a genetic network. Molecular systems biology, 5(1).
[6]. Yu, Wenqi, and Friedrich Götz. "Cell wall antibiotics provoke accumulation of anchored mCherry in the cross wall of Staphylococcus aureus." PLoS One 7.1 (2012): e30076.
[7]. Lorenz, R. et al. ViennaRNA Package 2.0. Algorithms for Molecular Biology 6, 26 (2011)
[8]. Hecht, Ariel, et al. "When wavelengths collide: bias in cell abundance measurements due to expressed fluorescent proteins." ACS synthetic biology 5.9 (2016): 1024-1027.