Methods amilCP

Summary and experimental methods of the amilCP subproject.

Background

amilCP is a non-fluorescent GFP-like chromoprotein from the coral Acropora millepora. Although the naturally occurring amilCP exhibits a strong blue color, variants of this protein have been created to express various different colors. By introducing random mutations in the chromophore region of amilCP, Liljeruhm et al. (2018) created different-colored variants of the chromoprotein. The only color that they did not manage to produce in their study was green.

However, in 2016, two years prior to the release of the paper by Liljeruhm et al., the Sydney iGEM team managed to produce a slow-maturing green variant of amilCP (BBa_K1996005) through error prone PCR. It was later verified by a group in Anthony Forster’s laboratory in Uppsala that the mutation responsible for the green color was located at position N157 (not GFP-numbering; not the location reported on their iGEM wiki). This was a point mutation, resulting in the swap from asparagine (AAC) to isoleucine (ATC). As opposed to the amilCP mutants by Liljeruhm et al. which were all created through mutations in the chromophore region, this particular mutation is located far from the chromophore region sequence-wise.

Except for Sydney’s green amilCP mutant, the only other green chromoprotein registered in the iGEM Parts Registry seems to be cjBlue. This chromoprotein is derived from the sea anemone Cnidopus japonicus and was registered as a biobrick part (BBa_K592011) by the Uppsala iGEM team in 2011. Just as Sydney’s green mutant, cjBlue is reportedly slow-maturing, which makes both chromoproteins less attractive for applications such as in vivo reporters for cell biosensors, selective markers or as biodyes. In an effort to address the current need for good green chromoprotein alternatives, this part of the project aimed to create a green amilCP mutant that matures faster than the Sydney variant. To achieve this, two different approaches were used:

  • Rational design, to introduce already known “superfolder mutations” and mutations that, based on data from our own modeling simulations and predictions, potentially could result in faster maturation of the green amilCP.
  • Random mutagenesis, to introduce random mutations throughout the entire coding sequence of Sydney’s green amilCP mutant in the hopes to find a faster maturing green mutant.

Additionally, this part of the project extended into a “color-changing segment”, that investigated whether mutating the N157 position of amilCP could yield other useful colors.

By courtesy of Anthony Forster and his team, we were able to use plasmids and bacterial strains expressing various chromoproteins from their collection in our experiments. We mainly worked with high-copy plasmids (pSB1K3) encoding amilCP, the green variant of amilCP and the mRFP1 chromoprotein. The green amilCP plasmid was used as a template for the rational design and random mutagenesis experiments, whereas the color-changing segment of the project called for the original amilCP plasmid as template. Since the mRFP1 chromoprotein exhibits a strong pink/red color and is fast-maturing, it served as a control or a reference in our experiments.

Choice of site directed mutations

For the rational design approach we looked at previous research of other chromoproteins and tried to apply this to our protein. The first of these mutations, in our project called the “superfolder mutations”, came from a study by Bevis, B.J. & Glick, B.S (2002) of the chromoprotein dsRed, where they observed an improvement of the maturation time when amino acids 41-44 were mutated into TQNV. Mutations to the homologous positions of amilCP from the wild-type EQTV to TQNV as well as TQTV was tested in a study by Bao et al. (2020), however without any improvement. Since our green version of amilCP has a significantly slower maturation, we thought it would be interesting to test the same mutations to see if we could observe any improvement of the maturation rate.

For the second type of mutations introduced, we collected inspiration from a research paper discussing the role of a pore to the chromophore of a GFP in its maturation (Evdokimov et al.. 2006). In their study, they proved that by substituting a valine in the pore with the bigger residue leucine, the maturation rate decreased. Since all known GFP chromoproteins share a general beta-barrel scaffold, we wanted to see if an induced pore in our slower-maturing chromoprotein might hasten the maturation by providing easier access for oxygen to reach the chromophore. In order to create such a variant, the X-ray crystallographic electron density maps provided in the paper of TurboGFP were used as reference and aligned to amilCP to find suitable and equivalent positions in the secondary structure for pore-inducement in our own amilCP protein. Mutations of amino acids in the same secondary structure elements found in TurboGFP were then simulated by mutating the amino acids in amilCP and then performing an AlphaFold stimulation of the mutant protein. If the predicted electron density maps showed a “hole” in the modeled beta-barrel, an amilCP mutant candidate was created. In total, four different final mutant candidates, all at position 195, were deemed promising. Leucine, which is the native amino acid, was mutated to Isoleucine, Valine, Alanine and Glycine respectively.

To find new colors, we focused on position N157 in the gene. This is the position found to be responsible for both turning amilCP green by substituting the asparagine with an isoleucine, and also for turning it pink by substitution to an aspartic acid. The mutations we introduced in this part were done both through deliberate choices of substitutions we thought would be interesting, but also with the use of random primers of position 157. We tested the following amino acids: Gln, for its similarity in polarity and structure to the original amilCP’s Asn; Gly, because of its distinct structure; His, for its charge and structure; Leu, which has a slightly different structure but otherwise similar polarity and charge as Ile; Pro, due to its significantly different structure; and Trp, for its intriguing structure. We also tested Asp as a kind of control, since we knew it should turn pink if the experiment is successful.

Detailed methods

Workflow for the amilCP subproject's three parts; Color change, Site directed mutagenesis for improved maturation and Random mutagenesis for improved maturation

References

Bao L, Menon PNK, Liljeruhm J, Forster AC. 2020. Overcoming chromoprotein limitations by engineering a red fluorescent protein. Analytical Biochemistry, doi https://doi.org/10.1016/j.ab.2020.113936.

Evdokimov AG, Pokross ME, Egorov NS, Zaraisky AG, Yampolsky IV, Merzlyak EM, Shkoporov AN, Sander I, Lukyanov KA, Chudakov DM. 2006. Structural basis for the fast maturation of Arthropoda green fluorescent protein. EMBO reports, doi 10.1038/sj.embor.7400787.

Liljeruhm J, Funk SK, Tietscher S, Edlund AD, Jamal S, Wistrand-Yuen P, Dyrhage K, Gynnå A, Ivermark K, Lövgren J, Törnblom V, Virtanen A, Lundin ER, Wistrand-Yuen E, Forster AC. 2018. Engineering a palette of eukaryotic chromoproteins for bacterial synthetic biology. Journal of Biological Engineering, doi https://doi.org/10.1186/s13036-018-0100-0.

Sydney Australia iGEM team 2016, https://2016.igem.org/Team:Sydney_Australia.