Part:BBa_K5172003

PafA 30.1 is a protein redesign of the naturally-occurring PafA enzyme (BBa_K5172000), which exhibits high activity towards phosphomonoesters, but is also active against phosphodiesters and phosphotriesters (Lidbury et al, 2022). It has 522 residues and a molecular weight of 58.1 kDa, and 30% of its most conserved residues (identified using ConSurf) were fixed during the redesign process. In particular, residues within 5Å of the docked ligand (glycerol 3-phosphate) and the two Zn2+ ions were fixed - for more details on how this was redesigned, see our AI redesign guide using LigandMPNN. See WT PafA for further details (BBa_K5172000).

Expression

After expression in E. coli Lemo21(DE3) and subsequent purification steps it appears that 30.1 was degraded, see Figure 1 and 2 due to the absence of a band on the SDS-PAGE gel which would correspond with the size of 58.1kDa. As a result we are unable to determine if the low level of activity compared to the WT PafA (see figure 4) is due to degradation, or the redesign altering important residues for enzyme activity.

Figure 1: SDS-PAGE results of PafA variant 30.1 post-protein purification. The flow-through (FT), binding buffer (BB), wash buffer (WB), and elutions (E1-10) were all tested to see which elutions contained protein.

Figure 2: SDS-PAGE visualising PafA variant 30.1 post size exclusion chromatography, all of the lanes, aside from the ladder, are different fractions from SEC.

Determining the Stability of the Proteins

Imidazole used in IMAC protein purification may cause them to unfold, reducing the accuracy of the characterisation assays. Circular dichroism (CD) experiments were carried out to ensure the proteins were folded the same way as our active WT protein. It shows the stability of the proteins by determining the temperature at which they aggregate out of solution or unfold. Figure (3A) shows the CD results plotting ellipticity against wavelength for PafA 30.1 compared to the CD results for WT PafA, Figure 3B. In the plots of ellipticity against wavelength for PafA 30.1, the curves do not flatten out at the higher temperatures as observed on the corresponding WT PafA, therefore suggesting that PafA 30.1 does not aggregate or unfold at temperatures up to 95℃. When the ellipticity at 208nm and 222nm is plotted for PafA 30.1 against temperature (Figure 3C and 3D) the sigmoidal shape is not present which confirms that it is not aggregating or unfolding, furthermore, the ranges in ellipticity values are much smaller with more consistent values.

Figure 3: (A) CD results for 30.1 for Wavelength against ellipticity (B)CD results for PafA for Wavelength against ellipticity, (C) and (D) ellipticity of 30.1 at wavelengths 208nm and 222nm (corresponding to alpha helix structure) against temperature (E) shows ellipticity of PafA at wavelength 208nm against temperature. 50℃ at PafA was heated for longer, causing evaporation and the change in ellipticity.

Characterisation of Activity

The enzyme's activity was measured with para-Nitrophenylphosphate (pNPP) as a substrate. pNPP is hydrolysed by phosphatases to release inorganic phosphate and para-nitrophenol (pNP), a yellow product which can be measured using 405 nm absorbance. The absorbance data was collected using a Tecan Spark plate reader, which allowed for a quick assessment of enzyme activity at various pNPP concentrations. Unfortunately, compared to the WT PafA (BBa_K5172000), 30.2 exhibited very low activity (Figure 4), signalling that the level of degradation visualised in figure 1 or 2 could have resulted in a loss of activity, or that during redesign key catalytic residues were lost. This could be improved in future cycles of the AI redesign with LigandMPNN by increasing the number of residues conserved around the active site, as well as experimentally validating more sequences.

Figure 4: Absorbance at 405 nm for 30.1 and PafA at different concentrations of pNPP. (A) 30.1 at 10nM absorbance at 405 nm increases slowly. (B) At 100 nM of 30.1, the absorbance at 405 nm increases. (C) PafA WT at 1 nM has a large increase in absorbance at 405 nm with time. (D) Control with no enzyme.

Temperature Assays

When temperature assays were conducted on 30.1 to determine the optimum temperature, like the other redesigned enzymes (BBa_K5172001 and BBa_K5172002), it appears to have an exceptionally high optimum temperature of 80℃, as shown in Figure 5. The results show that PafA WT’s activity remains high from 10℃ to 70℃, then rapidly decreases. However, 30.1 has very low activity from 0℃ to 70℃ then spikes at 80℃ before rapidly decreasing and reaching inactivity at 90℃.

Figure 5: Temperature assay results for PafA WT and its variants at 10-90℃, showing that variants 30.1, 50.2 and 70.2 have the highest maximum absorbance percentage at ~80℃ but PafA has the highest maximum absorbance percentage at ~37℃.

Initially, we considered that this may be due to the high temperatures causing pNPP to completely hydrolyse leaving no pNPP for the enzymes to hydrolyse, thus we decided to heat pNPP to the high temperatures before conducting the assay again at a lower temperature. This yielded results that were not significantly different to the pNPP which had not been previously preheated, see Figure 6. Thus we were able to determine that there was still pNPP present for the enzyme to hydrolysed.

Figure 6: Temperature assay results for PafA WT and its variants with pNPP preheated to 80℃ and 90℃ compared to no preheating (0℃).

Key Findings:

• The enzymes designed on LigandMPNN had increased thermal stability compared to the WT PafA.
• Unfortunately, their activity was extremely low, meaning we were unable to characterise their activities at different pHs using the standard enzyme concentration.
• This means further cycles of redesign on LigandMPNN would need to be done, fixing more residues around the active site pocket