Latest revision as of 22:32, 21 October 2021

Bovine growth factor FGF2 which induces cell proliferation

Profile

Name: FGF2 WT
Base Pairs: 468 bp
Origin: Escherichia coli, synthetic
Properties: Bovine growth factor which induces cell proliferation

Sequence and Features

Structure

Figure 1. 3D-model of FGF2.

Usage & Biology

Fibroblast growth factor 2 (FGF2) is one of 19 known members of the mammalian FGF family, involved in morphogenesis, development, angiogenesis and wound healing [1]. It is a mitogen and was first isolated in 1974, from the bovine pituitary and in 1988 human FGF2 was described for the first time [2]. There are four different FGF receptor tyrosine kinases (FGFR), to which FGFs bind and induce cell signaling. FGF2 binds FGFR1 and FGFR2 to induce proliferation [3]. FGFR2 consists of an extracellular domain, a transmembrane helix, and a catalytic intracellular tyrosine kinase domain [1]. The extracellular domain of the receptor consists of three immunoglobulin-like domains. FGF2 binds to domain 2 (D2) and domain 3 (D3), and the linker region connects the two domains [1] (figure 2).

Figure 2. Crystal structure of FGF2 (green) bound to FGFR2 (cyan). PDB: 1EV2.

When FGF2 binds to the extracellular part of the receptor it causes the receptor to dimerise with another FGF-bound FGFR2, leading to conformational changes which activates the receptor in its intracellular domain and cell signaling is initiated [4]. Through an intracellular signaling cascade gene regulation occurs in favour of cell proliferation, resulting in cell growth [4] (figure 3).

Figure 2. Schematic of FGF2 receptor binding cell signaling induction

Inducing cell growth through FGF2 signaling is utilized in the field of cellular agriculture, where the growth factor is used in the serum-free growth medium for cultivating meat. Similar to in a biological system, FGF2 induces cell growth when cultivating meat in a bioreactor. However, growth media is expensive, 55-95% of the production cost of cultivated meat comes from the growth medium [5]. To make serum-free media economically feasible on an industrial scale, the medium needs to be optimized. Being one of the most important and most expensive components, FGF2 is one of the targets for improvement [6].

To enable bovine FGF2 for research, a biobrick part to express bovine FGF2 has been designed and tested. This part will make FGF2 more accessible to support future research on medium optimization within cellular agriculture.

Experimental Results

Biobrick Assembly

The FGF2 biobrick part was designed inside a pUC plasmid together with the basic parts T7 promoter (BBa_J64997), Lac operator (BBa_K3599001) T7 RBS (BBa_K3257011), T7 T_phi terminator (BBa_B0016). To increase solubility of the protein during expression and purification, the FGF2 gene was fused with thioredoxin (BBa_K3934008), and to enable protein purification a 6xHis-tag (BBa_K3934015) was added. An enterokinase site (BBa_K3934016) was added to cleave off the 6xHis-tag and thioredoxin. The pUC plasmid also contains a gene for ampicillin resistance. The design was ordered from Integrated DNA Technologies (IDT).

The biological system used for biobrick assembly was E.coli DH5α competent cells. A T_phi terminator, an NdeI restriction site and an PstI restriction site were added to the 5’ end and the 3’ end of the FGF2 construct using PCR primers. The restriction sites and terminator were part of the primer overhang sequences. The PCR modified construct was treated with NdeI and PstI and ligated into a pET vector [7][8] containing an IPTG inducible T7 promoter, a Lac operator, an RBS and a kanamycin resistance gene. The plasmid was re-ligated using DNA ligase and transformed into E.coli DH5α. To remove the risk of religation of the pET vector, the restriction enzyme treated pET was run through gel electrophoresis and the band corresponding to the part of the plasmid to use was extracted with gel purification. The properly assembled plasmid (figure 4) was verified using Sanger sequencing Mix2Seq Kit from Eurofins Genomics and then transformed into E. coli BL21 (DE3) pLysS competent cells from Promega which are optimized for protein expression. Transformation was also done using NEB BL21 (DE3) cells.

Figure 4. Assembled biobrick vector.

Overexpression

To overexpress the plasmid, overnight cultures of BL21 (DE3) pLysS cells containing the FGF2 plasmid were inoculated in 37 °C with 150 rpm shaking in LB medium with 50µg kanamycin until OD600 reached 0.5-0.6. To induce expression, 1 mM Isopropyl-β-D-thiogalactopyranoside (IPTG) was added to the cultures, one culture received no IPTG induction and served as a negative control. The cultures were inoculated in the same conditions as previously for 6 h. The cells were then boiled in sample buffer (125mM Tris HCl pH 6.8, 20% glycerol, 4% SDS, 10% β-mercaptoethanol, 0.5 mg/mL bromophenol blue) to lyse the cells, and the protein expression was analysed using SDS-PAGE to confirm successful overexpression of FGF2. A 12% polyacrylamide gel containing the samples was submitted to 200 V and 0.04 A for 90 minutes. The process was repeated with E. coli BL21 (DE3) competent cells from NEB, inducing IPTG at a concentration of 1 mM and 0.01 mM (figure 5).

Figure 5. SDS-PAGE gel of FGF2 proteins expressed in E. coli (DE3) cells by induction with IPTG. From the left: un-induced FGF2, FGF2 induced with 10 uM IPTG, FGF2 induced with 1 mM IPTG, protein ladder. Bands corresponding to the size of FGF2 (30.8 kDa) are visible for induced cells but not for un-induced cells.

Purification

After having confirmed expression, the protein was purified using immobilized metal ion affinity chromatography (IMAC). First the IPTG induced cultures were spun down, and the pellet resuspended in lysis buffer (10 mM TRIS HCl pH 7.5, 10 mM MgCl2, 200 mM NaCl). The cells then underwent cell lysis in a cell disruptor (One Shot, by Constant Systems, UK) and after centrifugation the supernatant containing all soluble proteins, including FGF2, was collected. His GraviTrap columns (from Cytiva) were used, charged with nickel to which the 6xHis-Tag attached to the FGF2 binds, but the rest of the E. coli cell protein flows through. The column was equilibrated with lysis buffer and the lysate was added. The flow through was collected and then the column was washed with lysis buffer to remove any non-FGF2 proteins that might have gotten stuck in the column, after which the flow through was collected. To elute the bound FGF2 from the column, elution buffer was added (20 mM TRIS HCl pH 7.5, 10 mM MgCl2, 200 M NaCl, 500 mM imidazole pH 7.5). The imidazole outcompetes FGF2 in binding to the Ni, so the FGF2 protein detaches and can be collected as the solution flows through the column.

The clear lysate, sample flow through, the wash and the elution was run on an SDS-PAGE gel to verify that FGF2 bound to the column and was eluted when adding imidazole. A 12% polyacrylamide gel containing the samples was submitted to 200 V and 0.04 A for 90 minutes. A band of approximately 38 kDa was visible in the elution fraction indicating that we have purified FGF2 protein (figure 6&7).

Figure 6. The Promega E. coli BL21 (DE3) cells containing the pET vector with wt FGF2 were lysed using a french press. The lysate and pellet obtained were analyzed through an SDS PAGE. The 12% polyacrylamide gel containing the samples was submitted to 200 V and 0.04 A for 90 minutes. The gel was stained with coomassie brilliant blue. Lane 1-3: other FGF2 variants, 4: molecular weight marker, 5: other FGF2 variant, 6: wt wash, 7-9: other FGF2 variants, 10: wt flow through.

Figure 7. The Promega E. coli BL21 (DE3) cells containing the pET vector with wt FGF2 were lysed using a french press. The lysate and pellet obtained were analyzed through an SDS PAGE. The 12% polyacrylamide gel containing the samples was submitted to 200 V and 0.04 A for 90 minutes. The gel was stained with coomassie brilliant blue. Lane 1-3: other FGF2 variants, 4: molecular weight marker, 5: other FGF2 variant, 6: wt clear lysate, 7-9: other FGF2 variants, 10: wt elution.

Bradford Assay

The sample was spin concentrated using Amicon spin concentrators and the imidazole washed off using reaction buffer (20 mM Tris-HCl, 50 mM NaCl, 2 mM CaCl2 (pH 8.0)) using the same spin concentrators. Bradford assay (using a kit from Bio-Rad [9]) was performed to measure the protein yield. BSA was used to create a standard curve to use to determine the FGF2 concentration. OD595 for the sample was measured and the standard curve used to calculate the amount of expressed protein (figure 8). The concentration of our sample was 6.94 mg/mL in 0.5 mL buffer and the total amount of protein was 3.47 mg.

Figure 8. Bradford Assay BSA Standard Linear Regression used as standard curve for determination of protein concentration

Upscaling

Since the growth media and its growth factors are needed in bulk to cultivate meat on an industrial scale, it is important that it’s possible to produce FGF2 in large amounts. To prove that this part is suitable for industrial production, overexpression and purification was performed on a large scale in an upscaling laboratory.

Overexpression in Bioreactor

Overexpression was performed in 5 L bioreactors (Ez2 controller from Applikon). In order to maintain pH levels and keep foam from forming and clogging the bioreactor inlets, the bioreactor was coupled with a base (25% NH3) so that the solution could be pH regulated during expression and an anti-foam solution. A control tower was connected to the bioreactor to monitor and regulate pH levels and oxygen flow. A vitamin solution, trace A & B solutions and 50 µg/ml of kanamycin was added to the bioreactor before inoculation. Approximately 50 mL of overnight culture of E. coli BL21 (DE3) pLysS containing the FGF2 plasmid was inoculated in 2.5 L of cultivation media which gave a final OD600 value of 0.091. See our large scale purification protocols on our wiki page for more information on overexpression in the bioreactor.

OD was measured every 30 minutes to monitor the growth of the bacteria. At OD600 3.12 induction with 1 mM IPTG was performed. Before induction a small sample was collected and kept in 37 °C with 150 rpm shaking as a negative control. After induction OD was measured every hour for three hours (figure 9). The bioreactor ran for 7.5 h more to a total of 14 h.

Figure 9. Optical density as a function of time for growth of E. coli cells expressing FGF2wt.

To confirm that FGF2 was overexpressed, a sample from the bioreactor and the negative control was run on an SDS-PAGE gel. A 12% polyacrylamide gel containing the samples was submitted to 200 V and 0.04 A for 90 minutes. A band the size of FGF2 is visible in the induced sample, but not from the uninduced sample, confirming overexpression on a large scale (figure 10).

Figure 10. SDS-PAGE gel of FGF2 proteins expressed in E. coli (DE3) pLysS competent cells. From the left: induced FGF2, induced FGF2, induced hyperstable version of FGF2 (not relevant here), protein ladder, un-induced FGF2, un-induced hyperstable version of FGF2 (not relevant here). All induced samples are induced with 1 mM IPTG. Bands corresponding to the size of FGF2 (30.8 kDa) are visible for induced cells but not for un-induced cells.

Purification

After confirming overexpression, purification was performed with IMAC using an ÄKTA PURE 125 system. The column was packed with Capto Chelating resin and charged with 0.1 M NiSO4. To remove any unbound nickel, the column was washed with lysis buffer (20 mM TRIS pH 7.5, 10 mM MgCl2 and 200 mM NaCl). The 2.5 L culture from the bioreactors were spun down for 30 min at 12 000 rpm twice, removing the supernatant in between. The pellet was resuspended in lysis buffer. To break down the E. coli DNA DNAse was added to the sample before using a french press (EmulsiFlex-C55A from Avestin) to lyse the cells. The sample was run through the french press at 800 bar 3-4 times. The now lysed cells were centrifuged again and the supernatant containing FGF2 was then run through the Ni-charged column. The flow through was collected and the column washed twice, both washes were collected. Elution was collected in fractions as the elution buffer (20 mM tris pH 7.5, 10 mM MgCl2 and 200 mM NaCl, 500 mM Imidazole pH 7.5) was added with a gradient, slowly increasing the imidazole concentration. This was done since the optimal imidazole concentration for FGF2 elution was unknown. Light absorbance, corresponding to presence of protein, was measured throughout the process to monitor when protein flowed out of the column (figure 11).

Figure 11. Data showing overexpression of FGF2 in a 5 L bioreactor. Blue line: absorbance in nAU, y-axis values to the left. Red line: imidazole concentration, y-axis values to the right. Yellow line: column pressure.

Large amounts of protein was seen in the flow through, and two clear peaks were observed during the second wash when the imidazole concentration was 2%. During the elution there was one clear peak followed by a plateau and this was seen at imidazole concentrations ~140 mM and 220 mM respectively. To analyse in which fractions FGF2 was present, an SDS-PAGE gel was run with flow through, wash 1, wash 2 and fractions B1-C1 (figure 12). A 12% polyacrylamide gel containing the samples was submitted to 200 V and 0.04 A for 90 minutes.

Figure 12. SDS-PAGE gel of purified FGF2 protein expressed on a large scale. From the left: lysate flowthrough, wash 1, wash 2, fraction B1, fraction B2, protein ladder, fraction B3, fraction B4, fraction B5, fraction C1. Bands corresponding to the size of FGF2 (30.8 kDa) are visible in fractions B5 and C1.

No band corresponding to the size of FGF2 was seen in the flow through or the washes. In fractions B1-B3, corresponding to the first peak in the purification, thick bands are visible of ~39 kDa. In fractions B5-C1. corresponding to the shoulder peak, thick bands are visible corresponding to the size of FGF2. The first peak could be a protein with 3-5 His in a row, causing it to bind to the IMAC column, but eluting at a lower imidazole concentration than FGF2, which has 6 His in a row. The gel nonetheless confirms that FGF2 was successfully purified on a large scale.

Bradford Assay

The fractions containing FGF2 were spin concentrated (Amicon spin concentrators) and the imidazole washed away with a reaction buffer (20 mM Tris-HCl, 50 mM NaCl, 2 mM CaCl2 (pH 8.0)) using the same spin concentrators which were used for protein purification on the small scale. The protein concentration was measured using a Bradford Assay from Bio-Rad [9]. A standard curve was created using BSA. OD595 for the sample was measured and the standard curve used to calculate the amount of expressed protein (figure 8). The total protein concentration was 147 mg/mL in 11 mL buffer and the total protein yield from large scale production was 1.6 g. This is an approximately 461 times higher total yield compared to small scale production. The volume of the culture in the bioreactor was 50 times larger than for the small scale expression. This gives a ~9 times higher yield per volume for large scale compared to small scale. However, due to the impurity of the sample (see the multiple bands in figure 12) it is impossible to say how much of the protein from the large scale purification is FGF2 and how much is protein contamination. The expression in small scale was also done with cells from a different company which might affect the comparison.

References

[1] A. N. Plotnikov, S. R. Hubbard, J. Schlessinger, and M. Mohammadi, ‘Crystal Structures of Two FGF-FGFR Complexes Reveal the Determinants of Ligand-Receptor Specificity’, Cell, vol. 101, no. 4, pp. 413–424, May 2000, doi: 10.1016/S0092-8674(00)80851-X.
[2] L. Benington, G. Rajan, C. Locher, and L. Y. Lim, ‘Fibroblast Growth Factor 2—A Review of Stabilisation Approaches for Clinical Applications’, Pharmaceutics, vol. 12, no. 6, p. 508, Jun. 2020, doi: 10.3390/pharmaceutics12060508.
[3] W. Lim, H. Bae, F. W. Bazer, and G. Song, ‘Stimulatory effects of fibroblast growth factor 2 on proliferation and migration of uterine luminal epithelial cells during early pregnancy’, Biol. Reprod., vol. 96, no. 1, pp. 185–198, Jan. 2017, doi: 10.1095/biolreprod.116.142331.
[4] Y. Xie et al., ‘FGF/FGFR signaling in health and disease’, Signal Transduct. Target. Ther., vol. 5, no. 1, pp. 1–38, Sep. 2020, doi: 10.1038/s41392-020-00222-7.
[5] E. Swartz. “Meeting the Needs of the Cell-Based Meat Industry,” American Institute of Chemical Engineers (AIChE), Okt. 2019. Accessed on: Okt. 04, 2021. [Online]. Available: https://gfi.org/wp-content/uploads/2021/01/Cell-Based_Meat_CEP_Oct2019.pdf
[6] L. Specht. “An analysis of culture medium costs and production volumes for cultivated meat,” The Good Food Institute, Feb. 09, 2020. Accessed on: Okt. 04, 2021. [Online]. Available: https://gfi.org/wp-content/uploads/2021/01/clean-meat-production-volume-and-medium-cost.pdf
[7] L. Du, R. Gao, and A. C. Forster, “Engineering multigene expression in vitro and in vivo with small terminators for T7 RNA polymerase,” Biotechnol. Bioeng., vol. 104, no. 6, pp. 1189–1196, 2009, doi: 10.1002/bit.22491.
[8] L. Du, S. Villarreal, and A. C. Forster, “Multigene expression in vivo: Supremacy of large versus small terminators for T7 RNA polymerase,” Biotechnol. Bioeng., vol. 109, no. 4, pp. 1043–1050, 2012, doi: 10.1002/bit.24379.
[9] M. M. Bradford, “A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding,” Anal. Biochem., vol. 72, pp. 248–254, May 1976, doi: 10.1006/abio.1976.9999.