ICEFINITY: CIRCULARIZED AFP

Inteins have been proven to be an efficient way to circularize proteins¹. Furthermore, studies have shown that joining the termini of proteins leads to a significant increase in their thermo-stability^{2, 3}. In 2014, The Heidelberg iGEM team worked on generating BioBricks which allow for this circularization. This year, our Icefinity project sought out to stabilize the Type III antifreeze protein (AFP) using their methods, specifically utilizing the Npu dnaE split intein found in part BBa_K1362000.

Design & Cloning

A thorough dry lab analysis was performed to determine the appropriate linker needed to circularize the AFP. Details about this crucial dry lab stage are described on our Modeling Page. After selecting the optimal linker (GAA), The AFP-linker-extein insert sequence was then designed to be compatible with the Golden Gate Assembly technique necessary for BBa_K1362000; BsaI cut sites were placed on the ends of the insert sequence to allow for successful insertion into Heidelberg’s BioBrick. The final designed insert is illustrated below in Figure 1.

Figure 1. **Designed construct of BBa_; our circular AFP.**

The cloning of our circAFP began with a Golden Gate Assembly Reaction, the insert being our AFP-linker-extein sequence described above. Golden Gate Assembly involved a one-pot reaction, where both insert and vector were cut then ligated to generate a pSB1C3 plasmid containing our gene of interest. The Golden Gate Assembly Reaction was then transformed onto Topten electro-competent E. coli cells and plated onto Chloramphenicol-resistant plates. Successful cloning was determined by colony growth and colour (Heidelberg’s part contained an RFP; therefore, non-red colonies would indicate successful gene insertion). Colonies were picked and screened for our gene of interest. Next, our gene of interest was then inserted into a plasmid containing a T7 promoter and ribosomal-binding site to be used for protein expression. To determine successful insertion, further colony screening was performed (Figure 2).

Figure 2. **Agarose gel depicting screened colonies containing AFP-linker-extein within our vector containing a T7 promoter.** Specifically, colonies 1 and 3-6 all appear to contain an insert of the correct size (about 800bp). Those select colonies were sequenced then used for subsequent transformation and protein expression.

Protein Expression

The circAFP construct was expressed using IPTG induction and allowed to splice in vivo. Induction of circAFP occurred at 23^oC, which allowed for sufficient intein splicing to occur, generating our circularized AFP (Figure 3). As described by Zettler et al., the Npu dnaE intein splicing activity is extremely high and can occur within 6-37^oC⁴. *Wild-type Type III AFP is known to be about 8 kDa. Any discrepancies in the ladders shown on protein gels is due to the composition of the SDS-PAGE gel itself; gels run used Tris-Tricine as a major component, which can cause skewing of the ladder sizes.

Figure 3. **SDS-PAGE of unpurified cell lysate components for both wild-type (WT) and circular AFP: supernatant (sup), pellet, and total cell lysate (TCL).** First three columns show the WT AFP, and last three columns show the circAFP. Although the circAFP contains several additional amino acids, it appears to have run further down the gel. Similar results were visualized by the Heidelberg iGEM team in 2014. Cyclized proteins run further than their linear counterparts because of their more compact shape; this allows for the protein to travel through the gel much faster.

The protein was then purified using size exclusion. This technique was chosen because of the circAFP’s relatively small size (about 8 kDa). Column fractions were chosen and run on an SDS-PAGE to visualize the purified protein (Figure 4). Next steps were to attempt ice-affinity purification(IAP) of the circularized AFP, as shown in Figure 5. IAP uses the knowledge that AFPs and other ice-binding proteins attach themselves to ice, which creates an effective method of isolating these proteins from other cell compounds. IAP also verifies that our circAFP is in fact active despite our modifications.

Figure 4. **SDS-PAGE of the size exclusion column fractions, depicting the purified circAFP (Fraction A9).**

Figure 5. **SDS-PAGE of ice-affinity purified circAFP**. This gel illustrate the results of one round of IAP. The single band in the last lane indicates the purified circAFP, isolated in the ice fraction. The liquid fraction represents all compounds not incorporated into the ice.

Thermal Hysteresis Activity

To further ensure that the circAFP was still active in ice growth inhibition, a thermal hysteresis assay was performed on the crude cell lysates before purification. This assay has been developed through collaborations with the Davies Lab at Queen’s University and the Biochemistry labs at the Hebrew University of Jerusalem. This technique involves freezing the unpurified AFP solution and slowly melting it to a single crystal. Once isolated, the solution is slowly dropped further down below zero until the single crystal bursts into many. The difference between the melting temperature and the temperature at which the crystal bursts is known as the TH gap. In our crude lysate sample (with quite low concentrations – about 0.2mg/ml – of circAFP), we achieved a TH gap of about 0.3 degrees ^oC.

Figure 6. **Bar graph of TH assay results, comparing wtAFP with our circAFP at various activity test conditions.** Data shows that circAFP retains almost 80% of its antifreeze activity after being subjected to 90 ^oC . The wild-type AFP quickly loses its activity after exposure to such high temperatures.

AFP Thermostability

To test circAFP’s thermostability, we treated three samples at three temperatures 37, 68, and 90 ^oC for 10 minutes each. Wild type samples were also treated at these same conditions. After treatment, TH assays were performed and TH gaps were compared based on percent retention of activity, relative to the untreated sample. Figure 6 shows a bar graph which illustrates these results.

References

1. Scott, C.P. et al. (1999). “Production of cyclic peptides and proteins in vivo”. Proc. Natl. Acad. Sci. USA 96:13638–13643.

2. Iwai, H. and Pluckthun, A. (1999). “Circular beta-lactamase: stability enhancement by cyclizing the backbone”. FEBS 459:166-172.

3. Jeffries, C.M. et al. (2006). “Stabilization of a binary protein complex by intein-mediated cyclization”. Protein Science 15:2612–2618.