LXYL-p1-2 AND DBATG38R/F301V
Laboratory Objective #3
Produce and test LXYL-p1-2 and DBAT of Pathway 2
Figure 1. Schematic diagram representing the conjugation of LXYL-p1-2-SpyTag and DBATG38R/F301V-SnoopTag enzymes to αPFD-SpyCatcher and βPFD-SnoopCatcher via the catcher-tag system.
Our third lab objective was the run and test pathway 2. We aimed to produce and synthesise the reaction enzymes involved in the alternative biosynthesis of Paclitaxel. To do this, we utilised the assembled scaffold produced in Laboratory Objective #1.
Take a look at Laboratory Objective #2 to see the other pathway we optimised using Assemblase.
Introduction
Improvements to the current semi-synthetic pathway can only go so far to make Paclitaxel production sustainable. To truly reduce the environmental impacts of producing Paclitaxel, an alternative production method needs to be developed.
This is the core of our second pathway - scaffolding the novel enzymes LXYL-p1-2 and DBATG38R/F301V onto Assemblase for efficient production of Paclitaxel from a sustainable precursor. Because this pathway is completely biosynthetic, it does not require the use of any damaging chemicals. The biochemical pathway consists of two main enzymes. The first enzyme, LXYL-p1-2 catalyses the conversion of XDT to DT by the removal of a 7-β-xylosyl group. DBAT then acetylates DT at the C10 hydroxyl position into the final product, Paclitaxel1.
By scaffolding these enzymes together we allow a more efficient hand-over of intermediates between them. As these enzymes catalyse the rate-limiting steps of our pathway, this should dramatically increase the reaction rate for the pathway. By improving the efficiency of the pathway, we hope to make it economically viable and pave the way for a more sustainable future of Paclitaxel production.
Figure 2. Visualisation of reaction pathway 2. LXYL-p1-2 catalyses the conversion of XDT to DT. DBATG38R/F301V catalyses the acetylation of DT to produce Paclitaxel.
We plan to express these proteins with fused Spy and Snoop Tags for the attachment to our Assemblase scaffold, thus forming the LXYL-p1-2-SpyT and DBATG38R/F301V-SnoopT recombinant protein variants.
Gene construction – design
The following gene construct was designed to enable the cloning and expression of the recombinant proteins LXYL-SpyT and DBAT-SnoopT within a T7 expression system.
Additions to the gene are as follows:
- SnoopTag/SpyTag - The SnoopTag/SpyTag has the ability to conjugate to the SnoopCatcher/SpyCatcher via spontaneous isopeptide bond formation. This allows for the bioconjugation of the protein of interest to any protein fused to a SnoopCatcher.
- Hexahistidine Tag - The histidine residues has a high binding affinity to Ni2+, this enables the purification of the protein of interest using Immobilised Metal Affinity Chromatography (IMAC) via a Ni-NTA column.
- Gibson forward and reverse overhangs - 5’ forward and 3’ reverse gibson overhangs are complementary to that of the pET-19b backbone. This allows for the ligation of the protein of interest into the pET-19b plasmid via Gibson Assembly.
Additionally, GSG linkers are included between the peptide sequences. This flexible linker was designed to permit individual unhindered protein folding of each component (enzyme and tag).
Figure 3. Sequence annotation of the designed gene constructs for a visual representation of the spatial arrangement of LXYL-SpyTag and DBAT-SnoopTag gene constructs. LXYL (green) and DBAT (blue) contains a hexahistidine tag (silver) and SnoopTag (yellow) or SpyTag (blue). These are separated by flexible GSG linkers (dark grey). The gene constructs are enclosed within a 5’ gibson forward overhang and 3’ gibson reverse overhangs (red). Image by Linda Chen.
Objectives
This section aims to achieve the following:
- Clone LXYL-SpyT and DBAT-SnoopT into pET-19b vectors
- Express and purify LXYL-SpyTag and DBAT-SnoopTag fusion proteins
- Obtain enzyme kinetic measurements of enzymes (with and without attachment to the Assemblase scaffold)
Summary of our findings
| Successes | Successfully cloned DBAT into pET-19b |
|---|---|
| Successfully expressed soluble DBAT proteins | |
| Current Tasks | Clone LXYL into pET-19b and pD1204 vectors for expression in E. coli and S. cerevisiae host cells |
| Perform an Ellman’s assay to test enzyme kinetics of DBAT enzyme | |
| Conjugation of DBAT to 𝛽PFD-SnoopC scaffold subunit | |
| Future Directions | Clone LXYL into pD1204 and express the LXYL enzyme in S. cerevisiae chassis |
| Conjugate the LXYL enzyme to ⍺PFD-SpyC scaffold subunit. | |
| Obtain enzyme kinetic measurements of both LXYL and DBAT enzymes with and without attachment to Assemblase |
Cloning
Aim: To assemble LXYL and DBAT gene constructs (Figure 3) into pET-19b via Gibson Assembly.
Methods
A typical cloning workflow is summarised in Figure 4. Gene fragments were designed and obtained from Integrated DNA Technologies (IDT). pET-19b plasmid (obtained from Dr Dominic Glover) was linearised using PCR amplification. Ligation of our gene of interest was performed by Gibson Assembly and the ligation products were transformed into competent T7 Express E. coli by heat shock at 42°C for 10 seconds. Transformant colonies were plated onto Ampicillin supplemented Luria Broth (LB) Agar plates for selection. Colony PCR was used to screen for colonies containing the recombinant plasmid, PCR products were visualised on a 1% agarose gel. Positive colonies were grown up and plasmid DNA was extracted by miniprep and gene fragment sequenced by Sanger Sequencing.
Figure 4. A typical cloning workflow describing the vector preparation, insert preparation, ligation, transformation and confirmation steps. Protocol based on instructions from ThermoFisher.
Results
The LXYL and DBAT gene constructs were cloned into pET-19b backbone (provided by Dr Dominic Glover) by Gibson Assembly with a 3X molar excess of insert to vector. The assembly mix was incubated for 1 hour at 37°C before transforming into T7 Express E. coli cells by heatshocking at 42°C for 10 seconds. Transformed cells were plated on an LB Agar plate supplemented with Ampicillin at working concentration for selection, and incubated overnight at 37°C. Transformed colonies were screened for recombinant plasmids by colony PCR under the following conditions: Initial denaturation at 97°C for 3 minutes, 30X cycles of: denaturation at 97°C for 10 seconds, annealing at 67.6°C for 30 seconds and extension at 72°C for 40 seconds. Final extension at 72°C for 5 minutes. PCR products were run on a 1% agarose gel in 1X TAE Buffer for 1 hour at 100 V. The gel was imaged under transilluminator setting. PCR results revealed the DBAT amplicon was approximately 1500 bp in length (Figure 5) and the LXYL amplicon was approximately 1000 bp in length (Figure omitted).
Figure 5. Colony PCR gel image of recombinant DBAT-pET-19b plasmid. DBAT gene was amplified by colony PCR under the following conditions: 97°C for 3 minutes, 30X cycles at: 97°C for 10 seconds, 67.6°C for 30 seconds, 72°C for 40 seconds. Final extension at 72°C for 5 minutes. 10 uL of PCR product was run on a 1% agarose gel at 100 V for 1 hour using 5 uL of 2-log DNA ladder (NEB) as a standard (Lane 1). Single band obtained at approximately 1500 bp.
A single DBAT colony was grown up overnight in a 5 mL culture of Luria Broth (LB) supplemented with Ampicillin at working concentration. Cells were pelleted and lysed for plasmid purification using QIAGEN miniprep kit. Purified plasmid DNA was sequenced confirmed by Sanger Sequencing, obtaining high sequence homology with the expected DBAT gene construct. Sanger Sequencing results revealed 100% sequence homology with the expected DBAT gene construct.
Similarly, LXYL plasmid DNA was purified and sequencing results revealed low sequence homology to the expected LXYL construct, indicating LXYL has not been assembled into the pET-19b plasmid.
Discussion
Gibson assembly into pET-19b (E. coli chassis)
pET-19b vector was and linearised by PCR amplification. Linear gene fragments were purchased from Integrated DNA Technologies (IDT). The gene constructs were assembled into the pET-19b expression vector at the multiple cloning site via Gibson Assembly with a 3-fold excess of insert.
Gibson products were transformed into high efficiency T7 Express E. coli by heat shocking at 42°C for 10 seconds and plated on ampicillin supplemented agar plates for selection. This resulted in nine (DBAT) and seven (LXYL) transformant colonies, compared to zero colonies on the linear pET-19b transformant negative control. Three colonies of DBAT and LXYL transformants were screened by colony PCR, where DBAT-1 obtained a single band of estimated size of approximately 1500 bp (Figure 5). LXYL colonies revealed bands of at approximately 1,000 bp, which is below the expected size of 2,528 bp, showing the colonies did not contain the desired LXYL gene (Figure omitted).
DBAT
DBAT-1 colony was grown overnight in a 5 mL culture and plasmid DNA was extracted via miniprep. Samples were submitted for sequence confirmation by Sanger Sequencing. The sequence chromatogram was alignment to the designed gene sequence and obtained a 100% sequence homology. This confirms that the DBAT-SnoopT gene has been successfully assembled and transformed within the DBAT-1 colony. Glycerol stocks were prepared for storage and protein expression and purification.
LXYL
Assembly of LXYL was re-attempted by increasing the insert to vector ratio to 5:1 and screening more colonies by colony PCR. Attempts have not succeeded.
As LXYL is a larger gene fragment (2.5kb), ligation into pET-19b (5kb) would require a 2X excess of LXYL, instead of the 3X excess. However, excessive amounts of insert would unlikely to be the cause for non-assembly. It may be of benefit to increase the Gibson Assembly incubation time to ensure there is sufficient time for the complementary overhangs to properly anneal between the insert and vector.
Conclusion
DBAT has been successfully cloned into pET-19b and transformed into T7 Express E. coli. LXYL has not been successfully cloned into pET-19b.
Extending the incubation time of the Gibson reaction when cloning LXYL may increase likelihood of ligation. The 2019 UNSW iGEM team will pursue the cloning of LXYL into pD1204 for expression within the S. cerevisiae chassis.
Protein Expression and Purification
Results
Cells with pET-19b vector containing DBAT insert (as confirmed by colony PCR and sequencing) were grown up in a large-scale overnight culture of Terrific Broth supplemented with 1 mM thiamine HCl and ampicillin at 37°C with shaking. Once OD600 reached 0.6, protein expression was induced using 0.5 mM IPTG and grown for 7 hours at 20°C. A protein expression assay was performed on a sample of the overnight culture, using a cell-lysis detergent (BugBuster) to separate soluble and insoluble proteins (Figure 6).
Figure 6. Protein expression assay using BugBuster to determine expression of DBAT as soluble and insoluble form. A large scale culture of DBAT transformants in Terrific Broth with 1 mM thiamine HCl and ampicillin was induced at OD600 of 0.6 using 0.5 mM IPTG and grown for 7 hours at 20°C.
Purification
Following the confirmation of protein expression using BugBuster gels in Figure 6, attempts were made to purify DBAT. The eluted fractions were concentrated 20-fold using an Amicon Ultra-0.5 mL Centrifugal Filters to increase the visibility of any soluble DBAT present.
Figure 7. SDS-PAGE of AKTA purification fractions (F4 to F7) of DBAT His-tagged protein.
Liquid Chromatography with tandem Mass Spectrometry
Soluble protein bands (fractions 4-7) and a total protein lysate band at the same predicted molecular weight as DBAT (Figure 7). The bands were excised and sent for analysis by trypsin digest followed by Liquid Chromatography with tandem Mass Spectrometry (LC-MS/MS). This was performed to determine the identity of the protein bands by mapping peptides detected by LC-MS/MS onto the sequence of DBAT obtained from sequencing data of the cloned insert. This confirmed the identity of the protein as DBAT, the detected peptides of which are shown in Figure 8.
Figure 8. Peptides detected by LC-MS/MS mapped onto DBAT_TAXCU amino acid sequence using MASCOT. A) Total DBAT lysate sample. B) Soluble DBAT sample taken from fractions 4-7 (Figure 7).
Ellman’s Reagent Assay
A standard curve measuring absorbance vs differing concentrations of CoA-SH was constructed (Figure 9) with a computer generated linear trendline.
Figure 9. Standard curve of TNB absorbance measured at differing CoA-SH concentrations. Included is the computer generated linear trendline with Cartesian equation and R2 value.
A discontinuous assay was conducted in triplicate at 15-minute interval timepoints for a total of 90 minutes. Mean sample absorbance – Mean blank absorbance was calculated for each time point. Excel was then used to create a graph containing a computer-generated linear trendline (Figure 9).
Figure 10. Endpoint assay of change in TNB absorbance over 90 minutes, measured every 15 minutes. Included is the computer generated linear trendline with Cartesian equation and R2 value.
The amount of TNB produced in 1 minute was calculated by c = ∆A/(εl) where c is the concentration change of TNB (M), ∆A is the change in absorbance in 1 minute, molar extinction coefficient ε = 14 150 M-1cm-1, and pathlength l = 1 cm.2 The result was then divided by 60 to get the change in TNB concentration per second3.
The number of moles of TNB produced per second was calculated by n = cV where c is the concentration change calculated from the step above, and V = 350 x 10-6 L.
The rate of catalysis was calculated by (moles of TNB produced per second) / (moles of enzyme) as TNB is formed in a 1:1 molar ratio with CoA-SH produced. This gave us an estimated Kcat of 0.0056 (1/s).
Results
Following the successful cloning of DBAT into the pET-19b plasmid, which was confirmed by sequencing results, a large-scale grow-up was performed in order to express and purify soluble protein. An expression assay, using BugBusterTM Protein Extraction Reagent (Novagen) was used to validate successful induction and indicate protein solubility (Figure 6). Despite having a large portion in the insoluble fraction, soluble DBAT could still be observed at a lower yield. Multiple attempts were made to optimise expression conditions in order to achieve a higher percentage of soluble protein. Changed expression conditions included varying; IPTG concentration, temperature and expression time. Despite these attempts the solubility of our protein did not appear to significantly improve. Optimal conditions for expression were determined to be 0.5 mM IPTG induction for 7 hours in terrific broth media supplemented with 1 mM thiamine HCl.
Soluble protein was collected by sonication and purification on a Nickel NTA column via the AKTA start machine. The eluted fractions were run on a SDS-PAGE gel seen in Figure 7. A band at the same molecular weight of DBAT was observed in fractions 4-7 indicating that soluble protein may be present. A large amount of this protein appeared in the flow-through, indicating limited binding of His-tag to the Ni-NTA column. This may be due to the His-Tag being folded towards the protein’s core. Bands which corresponded to DBAT’s predicted size were excised from elution fractions 4-8 and analysed using trypsin digest in conjunction with LC-MS/MS. A band from the total protein lysate sample was also sent for analysis. The peptides from both the purified and non-purified (lysate) samples were matched onto the protein sequence for DBAT from Taxus cuspidata with 56% and 32% respectively (Figure 8). Both results showed a good coverage of peptides, indicating it was likely that the whole protein sequence was present and expressed.
Purification was attempted using buffers containing 2 M urea to relax the protein, releasing the His-tag and improving the binding of protein to the column. This attempt was unsuccessful. It should be noted, that moving the his-tag from the N- to the C-terminus via PCR, as demonstrated by Li et.al1, may improve the binding of DBAT to the Ni-NTA purification column.
Conjugation of the DBAT-SnoopTag and 𝛽PFD-SnoopCatcher was also tested, however it was unsuccessful. This could be due to a combination of low protein concentration and high contaminant quantity. This would prevent the formation of covalent bonds between the SnoopTag and SnoopCatcher.
The discontinuous Ellman’s Reagent assay allowed us to calculate a DBAT-SnoopT Kcat value of 0.005570 (1/s) with regards to acetyl-CoA. Due to the 1:1 molar ratio present in acetylation reactions, Kcat values of acetyl group donors and the reacting substrate can be equated. Current research involving DBATG38R/F301V documents a Kcat value of 0.001499 (1/s) with regards to DT1. This disparity is within a factor of 10 and can be deemed reliable, as Kcat values has a high degree of variance depending on differing conditions and assays used between experiments. Thus, we can conclude that our DBAT-SnoopT is functional, validating that the enzyme works as expected.
Conclusion
In the future we aim to successfully clone and express LXYL as well as express DBAT with a higher concentration of soluble protein. Additionally, we plan to conjugate these two enzymes to our Assemblase scaffold, quantifying improved activity using the assays described above. Finally we will perform additional assays such as NMR on reaction products to confirm the reaction and further quantify reaction rates.
References
- Li BJ, Wang H, Gong T, Chen JJ, Chen TJ, Yang JL, Zhu P. Improving 10-deacetylbaccatin III-10-β-O-acetyltransferase catalytic fitness for Taxol production. Nature communications. 2017 May 18;8:15544.