Part:BBa_K5172000
Contents
PafA
This part is the WT PafA gene PCR-amplified from Flavobacterium johnsoniae with it's signal-peptide removed. It is a 58.9 kDa alkaline phosphatase with high activity towards phosphomonoesters, and some activity against phosphodiesters and phosphotriesters (Lidbury et al, 2022). It is seemingly unique among alkaline phosphatases as it is minimally product-inhibited and maintains a high level of activity in high Pi concentrations (Figure 1; Lidbury et al, 2022). This part has also been tested against the existing PhoA phosphatase on the registry (BBa_K3767001) and was found to be more active. This makes PafA, to our knowledge, the most active phosphatase enzyme on the iGEM registry — making it an even more sensitive reporter than PhoA (BBa_K3767001), and suitable for regenerating various organophosphates back into agriculturally-useful inorganic phosphate.
Figure 1: Varying rates of growth and catalytic activity of wild-type and mutant strains of PafA dependent on differing phosphate sources for growth (Lidbury et al., 2022)
Background
PafA is thought to have evolved minimal product-inhibition as this offers an advantage in the Bacteroidetes phylum by enabling it to accumulate inorganic phosphate in a wider variety of environments and conditions (Lidbury et al, 2022). There is also evidence suggesting that using PafA to convert organophosphates to inorganic phosphate allows the use of the other product as a carbon and potential energy source, see Figure 2 (Lidbury et al, 2022).
There is limited information regarding the kinetics, thermodynamics and mechanisms of PafA from F. johnsonaie. A similar PafA to PafA from F. johnsoniae is from Chryseobacterium (flavobacterium) menigosepticum. This has an optimum pH of 8.5 and it's primary active site nucleophile is a threonine (position 79), with the active site resembling the docking site predicted by CB2 docker (figure 3). In acidic conditions, the rate-limiting step of the reaction is hydrolysis of the phosphoenzyme, but in alkaline conditions, it is the dissociation of the non-covalent enzyme-product complex, see Figure 4.
Figure 2: Utilisation of a range of organic phosphates as sole phosphate (A) and carbon sources (B) (Lidbury et al, 2022). (A) WT PafA Fj growth on organic substrates (200uM) was compared to the growth of mutant Δfjoh_0074:ΔphoX^Fj:ΔphoA1^Fj:ΔphoA2^Fj:ΔpafA^Fj quintuple mutant (ΔM5) and a PafA complemented ΔM5 (ΔM5+pafA^Fj), with growth compared to a no P control. (B) WT, ΔM5, ΔM5+pafA^Fj and ΔpafA^Fj were also grown on a range of organic phosphate substrates (3mM), a glucose-containing substrate, and growth compared to a no C substrate control.
Figure 3: A- Representation of the common active site of PafA, including its Zn2+ co-factors. The phosphate group that PafA is removing is shown in orange (Sunden et al., 2016). B- Docking of AMP and Zn2+ co-factors into PafA F.Johnsoniae active site using CB2Docker, visualised on ChimeraX (v1.6). T55 (B) appears to be in the same position as T79/90 (A) relative to the phosphate ion.
Figure 4: The mechanism of catalysis for an alkaline phosphatase such as PafA, including the various rate-limiting steps (Kim and Wyckoff, 1991)
Figure 5: The kinetic activity of PafA shown with varying substrates (Berlutti et al., 2001) In our project, we planned to use pNPP for our assays as it shows the highest activity with PafA (see Figure 5), and the rate of activity is easy to quantify due to the colour change of the reaction with pNPP.
AI-Redesigned Variants with Improved Stability
The team used various tools to improve the enzyme's temperature and pH stability to be better suited for industrial use. LigandMPNN was used to redesign the enzyme, and various docking tools such as CB2Docker and Alphafold3 were used to select regions to fix in the redesign and ConSurf to identify conserved residues.
During the redesign of this enzyme, there were many factors to consider such as leaving conserved sections unchanged to not affect the functioning of the enzyme. This includes a 5Å area around the Zn2+ co-factors and ligands so they can consistently dock in the area of the enzyme. To the redesigned PafA variants, we added a His tag to the enzyme (built into our pET21a(+) backbone) to allow protein purification to be carried out via a nickel column. We checked our changes to the enzyme's sequence using multiple protein structure visualisation tools such as AlphaFold3 and ChimeraX (version 1.6) to ensure no changes in the enzyme's 3D structure.
You can find our AI-improved variants here:
Characterisation of PafA
To characterise our PafA protein, we cloned it into pET21 vector which added a start codon and C-terminal 6xHis-tag. We then sequence-confirmed this construct and transformed it into Lemo21(DE3) E. coli, where it was grown to 0.6 OD at 37°C, induced with 500 µM IPTG, then moved to 16°C for overnight protein expression. The cells were then lysed via sonication, and the protein purified via first Ni-affinity, then size exclusion chromatography. All of the following characterisation was done in vitro using this purified, soluble protein.
Circular dichroism (CD) experiments were carried out to ensure the proteins were folded and to show the stability of the proteins by determining the temperature at which they aggregate out of solution or unfold. Figure 6 shows the CD results plotting ellipticity against wavelength for PafA. The results show a repeat at 50℃ to check the CD spectrometer was appropriately heating the sample by heating the repeat in a heat block to 50℃ and re-measuring. This caused the two plots for 50℃ on Figure 6A to initially not align. However, when normalised, in Figure 6B, they looked the same, meaning that even though the spectrometer was functioning correctly, the evaporation from heating the same sample for a longer period had caused a change in its concentration, resulting in the unnormalised plots differing in appearance.
At 222 nm and 208 nm in Figure 6A, two dips at lower temperatures can be seen. It can be interpreted that our protein is primarily composed of alpha helices, which is consistent with AlphaFold visualisation. At higher temperatures, the plots flatten out, suggesting that the protein aggregates out of the solution, most likely as it unfolded. For PafA, figures 6C and D show that it is completely aggregated out of solution by 70℃. Figure 6E
Figure 6: Circular Dichroism of PafA. A: Ellipticity in mDeg plotted against wavelength in nm showing unfolding behaviour of PafA. B: Normalised Ellipticity plotted against wavelength in nm. C: Ellipticity against temperature at 208 nm for PafA. D: Ellipticity against temperature at 222 nM for PafA E: Percentage of maximum ellipticity (%) against temperature (℃) for normalised PhoA and PafA.
PNPP (p-Nitrophenyl Phosphate) was used to characterise the activity of PafA compared to PhoA. PNPP is converted to inorganic phosphate and pNP by phosphatases, a yellow product that can be identified by measuring absorbance at 405 nm. A plate reader (Tecan Spark) was used to speed up the process of these assays, so the bounds of absorbance the plate reader could accurately measure needed to be determined, as well as identifying the equation of the linear relationship between PNP formation and absorbance. PNP was used to determine this relationship, using a 96-well plate with serial dilutions across from 20 mM to 1.19 nM. Five absorbance readings at these concentrations were taken, and a mean was plotted. Two repeats were also done to ensure any anomalies in the averages could be determined.
We converted absorbance measurements to concentrations using the linear relationship (y=2.1x+0.11), see our wiki for more details on how we identified the mathematical relationship between absorbance and PNP concentration.
Figures 7 and Prism showed that PafA has a Vmax of 13210 min-1 (95% CI is between 12490 and 13954 min-1), and Km is 0.02236 mM (95% CI of 0.01712 to 0.02885 mM). As shown in Figure 8 PhoA has a Vmax of 6774 min-1 (95% CI of 6369 to 7192 min-1) and Km of 0.01549 mM (95% CI of 0.01068 to 0.02179 mM), showing that PafA was more active than PhoA (BBa_K3767001), making PafA the most active phosphatase on the registry.
Figure 7: Michaelis-Menten curve fitted for PafA at 1nM using GraphPad Prism.
Figure 8: Michaelis-Menten curve fitted for PhoA at 1nM using GraphPad Prism.
PafA's activity at different pHs was determined using sodium acetate buffer (pH 4 - 6.5) and tris buffer (pH 6.5 to pH 9.5) adjusted to a pH range. An overlapping pH for the buffers of 6.5 was included in figure 9 It can be seen that the usage of different buffers had no impact on the activity of PafA. The ideal pH range appears to be within the range of 7.4 - 9. However, more data points would be needed within this range to identify the optimum pH.
Figure 9: pH against Vmax for PafA at different pHs and differing concentration of PNPP.
PafA's activity was determined by measuring the Absorbance at 450nm after 10 minutes of PNPP hydrolysis at increasing temperatures, from Figure 10 we can see that after 60℃ the activity of the enzyme drops significantly and at 90℃ the enzyme is inactive with no change in hydrolysis compared to the control.
Figure 10: Absorbance at 405nm against Temperature for PafA
References
Berlutti, F. et al. (2001) ‘The chryseobacterium meningosepticum PAFA enzyme: Prototype of a new enzyme family of prokaryotic phosphate-irrepressible alkaline phosphatases? the GenBank accession number for PAFA reported in this paper is AF157621.', Microbiology, 147(10), pp. 2831–2839. doi:10.1099/00221287-147-10-2831.
Kim, E.E. and Wyckoff, H.W. (1991). Reaction mechanism of alkaline phosphatase based on crystal structures. Journal of Molecular Biology, 218(2), pp.449–464. doi:https://doi.org/10.1016/0022-2836(91)90724-k.
Lidbury, I.D.E.A. et al. (2022) ‘A widely distributed phosphate-insensitive phosphatase presents a route for rapid organophosphorus remineralization in the Biosphere', Proceedings of the National Academy of Sciences, 119(5). doi:10.1073/pnas.2118122119.
Sunden, F. et al. (2016) ‘Mechanistic and evolutionary insights from comparative enzymology of phosphomonoesterases and phosphodiesterases across the alkaline phosphatase superfamily', Journal of the American Chemical Society, 138(43), pp. 14273–14287. doi:10.1021/jacs.6b06186.s003.
Sequence and Features
- 10INCOMPATIBLE WITH RFC[10]Illegal EcoRI site found at 529
- 12INCOMPATIBLE WITH RFC[12]Illegal EcoRI site found at 529
- 21INCOMPATIBLE WITH RFC[21]Illegal EcoRI site found at 529
- 23INCOMPATIBLE WITH RFC[23]Illegal EcoRI site found at 529
- 25INCOMPATIBLE WITH RFC[25]Illegal EcoRI site found at 529
- 1000COMPATIBLE WITH RFC[1000]
Functional Parameters
Vmax: 13210 ± 720 min-1 (95% CI)
Km: 0.02236 ± 0.00524 mM (95% CI)
Max Temperature: ~65°C
Optimum pH: 8–9
| None |