Choice of Genes
As we were looking for suitable phosphate binding proteins, we found many possible monomeric candidates. Since we wanted to use a monomeric phosphate binding protein (PBP) in combination with our photosensitizer SOPP3, we decided to use the periplasmic phosphate binding protein originated in the Pst complex of Escherichia coli [1]. This PBP shows a high affinity to phosphate, as it has been proved that it can remove ≥ 97% of phosphate within 6 hours from water with an initial concentration of 0.2–0.5 mg/L [1]. We chose this PBP for our project because it originated in E. coli, which we wanted to use as an expression organism, and has high affinity to phosphate. We decided to use its PDB entry (2ABH) as its project internal name [2]. Here, you can find more inforamtion about 2ABH.
We decided to use SOPP3 and VVD as our optogenetic components following the recommendation of our advisors Prof. Rother, Dr. Krauss, and Dr. Drepper. SOPP3 was chosen as the photosensitizer because it has been shown to be very reliable and robust, meaning that its efficacy is high and activation in the dark is very unlikely [3]. We decided to use VVD as the optogenetic switch, since its function is relatively well analysed [4], and our advisors had already worked with it. Therefore, they could help us with the design of the VVD-PBP gene, which marked a huge benefit over other optogenetic switches, such as different LOV-domains.
Since we needed a dimeric phosphate binding protein to use in combination with VVD, we focused our research in that direction. However, we could not find any true dimeric PBPs, so we decided to use the colicin M immunity protein (CMI), since it is selectively binding phosphate as a ligand and provided the best available 3D structure for a fusion with VVD. Another advantage of CMI is that it originated in E. coli as well.
Gene Design
We designed the proteins based on the amino acid sequences and afterwards translated them back, including codon optimization. The amino acid sequences can be found in a protein data bank or in publications.
Gene design forms the basis for working with proteins. Therefore, it is one of the most important steps in the process of protein analysis. That is why it is important to work in a neat and well-structured way. To organize the gene design and make it clearly arranged, we used Benchling.
To put the single parts of the fusion protein together, it is crucial to know about the positions of its functional structures. Those structures are active sites and binding sites for cofactors. Often, this already gives a hint of where the proteins should be combined. Within the phosphate binding protein 2ABH, the active site is near the C-terminal ending. That is why we decided to combine it with the photosensitizer at the N-terminal site. 2ABH is a relatively small protein with a size of approximately 30 kDa. Therefore, we chose to try the irreversible approach using SOPP3 with that protein. To prevent steric inhibition or interference, we used a spacer to connect the proteins. The spacer is a neutral amino acid sequence, which does not form any special 3D structure.
For the second engineering cycle, we wanted to design a reversible protein. For this purpose, we wanted to use VVD as the optogenetic switch. Since VVD is a homodimer, we needed a dimeric phosphate-binding protein. We chose CMI because the phosphate interaction is mediated by both protein chains. Luckily the arrangement of the protein chains in the dimeric form of each protein fit together. For that reason, we were able to design only one protein strand that would contain both proteins and will be able to build out a dimer structure. By looking at the 3D structure of the proteins, we have noticed that they would be sterically inhibited. That is why we decided to use a linker between them. The linker we used has roughly the same structure as the spacer mentioned before. They only differ in their length. To make a better differentiation we used different terms.
This first part is providing the basis for the protein design.
To get the proteins synthesized, they need to be inserted into a vector. The vector includes all the necessary parts that are needed for a controlled expression. Depending on the vector, different expression systems can be used. We chose the pET-28a(+) vector that includes the T7 expression system.
The T7 expression system is a double secured system. The T7 promotor can only be read by the T7 RNA polymerase, which can be found in Escherichia coli itself. Both the T7 promotor and the gene coding for T7 RNA polymerase are under the control of the lac operon. This leads to inactivation during the absence of lactose due to inhibitor binding. Only if the cell takes up lactose or IPTG, the inhibitor itself will be inactivated and the T7 RNA polymerase can be produced. This will lead to the production of the protein of interest.
The vector also includes a gene that is coding for resistance against one specific antibiotic. The pET-28a(+) vector includes a sequence that codes for a kanamycin resistance. Cells that take up the vector will be able to grow in media that contains kanamycin. Other cells not containing the vector will not grow. This ensures that during our experiment only cells carrying the gene grow.
To purify our proteins after expression, we used polyhistidine residues as a tag ($His_{6}-Tag$) to purify only our targeted proteins. This polyhistidine-tag is a tool encoded in the pET-28a(+) vector. We have inserted the self-designed genes into the vector so that an N-terminal $His_{6}-Tag$ is appended.
Gibson Assembly
Gibson Assembly has been used for cloning purposes of a single gene insert. The advantages include that the insertion of DNA fragments can be done regardless of their fragment length or end compatibility. This method can join several overlapping DNA fragments during a single isothermal reaction. The used kit contains a master mix which includes three different enzymatic reactions. First, to facilitate the annealing of fragments which are complementary to each other, the exonuclease generates single-stranded 3’ overhangs. Afterwards, the DNA polymerase is supposed to fill the gaps that were created within each annealed fragment. And lastly, the DNA ligase seals the nicks in the constructed DNA. As a result, a double-stranded DNA fragment is generated which can be further used [5].
Figure 1: Schematic figure of Gibson Assembly. The lineraized pET-28a(+) vector is shown left. On the right, the fragment of the 2ABH gene is shown.
Transformation
The transformation was done to insert a plasmid into Escherichia coli. The plasmid contained our gene of interest with a kanamycin resistance marker, which served for the selection process later. As E. coli is not naturally competent it cannot take up DNA normally. Therefore, the transformation was artificially reproduced to generate chemical competent cells by creating pores in the bacterial cell membrane. A method that is commonly used for transformation is working with a heat shock. The heat shock method involves the incubation of cells in calcium chloride ($CaCl_{2}$) to create pores in the membrane. The mechanisms of how exactly this works remain largely unknown. A hypothesis states that the calcium ions are supposed to be a cation bridge between the negatively charged phosphorylated lipid A in lipopolysaccharide (LPS) and the phosphate of the backbone of the DNA. The binding of the DNA onto the surface of the cells is facilitated by ice-cold $CaCl_{2}$ and the uptake of the DNA is accomplished by a short period of heat shock. To select the cells that took up the plasmid the cells were plated onto agar plates supplemented with kanamycin. [6]
Protein Expression
Protein Expression via Autoinduction
Autoinduction is used to express the $His_{6}-fusion$ proteins based on the T7 expression system. The pET-28a(+) expression vector contained the target gene encoding for the protein VVD-CMI, which was transformed into Escherichia coli BL21(DE3) gold strain. The pET expression vector is composed of two essential components, the gene of interest and a T7 promotor, which controls the expression of the targeted gene. The T7 RNA polymerase, which is possessed in E. coli, regulates the transcription at the T7 promotor. The gene encoding the T7 RNA polymerase (T7 gene) is under the control of the lac promotor. In the absence of lactose, the T7 promotor and the lac promotor are repressed by the lac repressor. The presence of lactose initiates the expression of the T7 gene by binding to the lac repressor causing a conformational change. Therefore, the autoinduction medium contains glucose and lactose in defined amounts. First, glucose is taken up by E. coli cells as the main carbon source. The uptake of lactose initialises the expression of the targeted gene once glucose is fully consumed. Although this type of expression requires a more complex media preparation, it simplifies the handling of the expression, which made it valuable for our project [7].
Protein Expression via IPTG Induction
IPTG (Isopropyl β-D-1-thiogalactopyranoside) is a molecular mimic of allolactose, which is a metabolite of lactose. IPTG triggers the expression of the protein of interest inserted in a pET-28a(+) expression vector in E. coli cells, which is under the control of the lac promotor. IPTG binds to the lac repressor, reversing the repression of the T7 gene and thus initiating the expression of the protein of interest [8].
Sonication
Sonication is a physical method to disrupt cells using pulsed, high-frequency sound waves. Ultrasonic treatment allows cell disruption by creating cavitation forces, which push cells against each other and causing them to disrupt. The high-frequency sound waves are produced by an apparatus containing a vibrating probe, which is submerged in the cell suspension. The resulting mechanical energy from the probe is causing the formation of vapor bubbles, which directly implode and cause shock waves. For the prevention of too much heat, the samples are stored on ice and the treatment is performed in several short bursts [9].
Purification of His-tagged Proteins
The purification has been done by immobilized metal affinity chromatography using the $His_{6}-Tag$. Our $His_{6}-Tag$ contains six histidines, which are attached to the N-terminal end after the methionine codon in the open reading frame of the gene. Thereby, a fusion protein is generated with a polyhistidine-tag, which can be used for purification purposes. The purification of those proteins is a highly selective method to extract the protein of interest from a cell lysate. Nickel columns (Ni-IDA columns) were used for the purification of our proteins as it has a high affinity for hexahistidine-tags and only low or no affinity for other proteins. The tagged proteins can coordinate nickel ions on their surface. After the sample is applied to the column a wash buffer is applied with a low concentration of imidazole. Imidazole can elute the proteins that are weakly bound to the column. The recombinant proteins are eluted with high concentrations of imidazole. Imidazole is used as an eluate because it has a higher affinity to nickel ions so the proteins are being displaced. To put it in a nutshell, this method is used to generate a purified form of the protein by separating it from other proteins.
Buffer Exchange
To remove the excess imidazole after the purification step, a buffer exchange with Amicon columns has been done. This is necessary as a final step to transfer the protein into its proper buffer for downstream processes. Amicon is a centrifugal ultrafiltration device, providing varying pore sizes for different molecules of interest, and is used to buffer-exchange macrosolutes. The concentration of non-permeating species increases during this process while the fluid volume is reduced. Thereby, the desired concentration can be achieved. After the solutions have been transferred into the HEPES buffer, the protein can be used in its desired manner. The advantages of using the Amicon filter are the fast sample processing due to the vertical design and the membrane surface. Moreover, it ensures a high sample recovery rate of up to 90% and a concentration of 80-fold [10].
Bradford Assay
Bradford Assay was used for protein detection and quantification. The assay is based on Coomassie dye which is a colorimetric reagent to which proteins bind under acidic conditions. The binding of proteins results in a color change from brown to blue. This results in a spectral shift from the absorbance at 465 nm to 610 nm. The development of the color during the assay is due to the presence of several basic amino acids. Especially arginine, histidine and lysine play a role. Moreover, the dye-protein binding is influenced by van der Waals forces and hydrophobic interactions. A prerequisite for the binding is the minimum mass which is 3 kDa. Therefore, peptides, low molecular weight proteins and free amino acids do not interact with this assay which reduces faulty measurements. The color intensity is influenced by the number of Coomassie dye molecules bound to each protein which is proportional to the number of positive charges found on the protein. We used the Bradford protein quantification to see how much protein we were able to purify and to check if our immobilisation was successful [11].
BCA Assay
To control how much protein we could purify, we used the BCA Assay. This assay is measuring the total amount of protein in a solution. This test is using the reduction of $Cu^{2+}$ to $Cu^{+}$ by proteins in an alkaline medium. The detection is done with a colorimetric assay of the cuprous cation $Cu^{+}$ by bicinchoninic acid (BCA). First, a blue-colored complex is formed by the chelation of copper with protein. Only the peptides which contain three or more amino acid residues form a colored chelate complex with cupric ions in the alkaline solution that contains sodium potassium tartrate. These complexes produce a light blue to violet complex which absorbs light at 540 nm. The intensity of the color is depending on the number of peptides which are bound to the cupric ion. Second, the BCA reagent is added to the protein samples. The BCA reagent is a highly sensitive colorimetric detection reagent which forms a purple-colored BCA-copper complex. The complex exhibits strong linear absorbance at 562 nm with increasing protein concentrations. We used the BCA protein quantification to see how much protein we were able to purify and to check if our immobilization was successful [12].
Figure 2: Colorimetric detection of protein using the BCA Assay. The absorption of the samples and the standards are measured at 540 nm to quantify the amount of protein in the sample.
Immobilization
Our proteins have been immobilized by using EziG beads. This method ensures a single-step immobilization directly from the cell lysate using a $His_{6}-Tag$. EziG beads are based on controlled porosity glass (CPG) which has high-quality flow-through properties based on their interconnecting pore structure and their incompressibility. The hydrophobic surface of the EziG beads is covered with an organic hybrid polymer called hybrid CPG. This polymer has the same benefits as the conventional CPG, but its surface can additionally be modified to suit the application in terms of functionalization and hydrophobicity. These particles can then bind protein affinity tags like $His_{6}-Tag$ which has been used in our project. For $His_{6}-Tag$ binding, the EziG beads contain chelated Fe-(III) which enriches the enzyme during the immobilization process and thereby maintains a non-destructive binding. These properties lead to high retention of catalytic activity [13].
Figure 3: Illustration of immobilization with EziG beads. The Fe-(III) that is contained in the EziG beads binds specifically to the $His_{6}-Tag$ of the enzyme to be immobilized, thus enabling targeted immobilization.
SDS-PAGE
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) is performed to assess the purity and quantity of purified proteins. SDS-PAGE is an analytic method to separate proteins according to their molecular mass. The addition of SDS to the protein sample allows all proteins to denature while boiling at 95 °C. The denatured proteins can be separated in the polyacrylamide gel according to their mass upon the application of a constant electric field. Thereby, the smaller molecules move more rapidly through the gel than larger proteins. To obtain an increased resolution of the bands, two gel types, stacking and separating gel, with different pH values were used on the discontinuous SDS-PAGE. The proteins loaded in the gel were stained with Coomassie blue. Protein bands at the respective molecular mass of the protein sample to be examined indicate the presence of the targeted protein [14].
The protein construct of SOPP3-2ABH has a molecular mass of approximately 50 kDa, whereas 2ABH and VVD-CMI have a molecular mass of approximately 30 kDa.
Since SDS-PAGE allows the quantification of the amount of protein, we were able to use this method to compare the impact of different expression temperatures. Expression at different temperatures showed a thicker or thinner band, which gave us a clue which temperature resulted in a higher protein yield (see results). In addition, SDS-PAGE was used to verify the purification using Ni-IDA columns. By applying the flow-through, wash step and different elution concentrations with imidazole, the efficacy of the purification could be assessed. It was possible to determine which imidazole concentration was best suited for elution and whether the target protein was already present in the flow-through and wash step, which would indicate that too much protein was added to the column and that it could not completely bind the entire protein mass.
Phosfinity Assay
To determine the efficiency of the binding and release of phosphate by the phosphate binding proteins (PBPs), we needed a quantification method to determine the concentration of phosphate in the solution before and after the binding and release. To this end, we used the Phosfinity-Total Polyphosphate Quantification Kit by Aminoverse. This kit was provided to our team by the Blank Lab. This kit allows quick colorimetric Pi detection using non-enzymatic hydrolysis of acid-instable phosphorous-containing substances. Depending on the concentration of Pi in the samples, blue staining can be observed and measured in a photometer at 882 nm. This allows a precise quantification and comparison of phosphate concentrations in the different stages of our experiment.
Figure 4: Colorimetric determination of phosphate using the Phosfinity Assay.
Preparation of Polyphosphate-rich Yeast Extract
To generate polyphosphate-rich yeast extract, a sufficient cell mass of Saccheromyces cerevisiae must first be produced. How much cell mass is required depends on the size of the test batch to be performed. For this purpose, S. cerevisiae is grown in SD medium. The cell mass is dried and then put in the starvation medium in which no phosphate is present. The cell mass produced here is dried again and after an incubation period, placed in the feeding medium. The cells, previously starved of phosphate, now have phosphate in excess and take up as much phosphate as possible and convert it to polyphosphate. This process is described as polyphosphate hyperaccumulation. The result is a yeast extract which is very rich in polyphosphate. For example, we achieved a polyphosphate yield of 23%. The percentage of polyphosphate always refers to the $KPO_{3}$ contained in the dry weight of the cell [% (w/w)]. The yeast extract produced in this way can be used directly in the food industry. However, it is also possible to extract the polyphosphate from the yeast extract using, for example, phenol chloroform extraction [15]. Using this extraction method, we were able to extract polyphosphate from individual samples and then determine the polyphosphate content using the Phosfinity Assay. We have used the organism S. cerevisiae VH2.200, which achieved the highest yield of polyphosphate in the literature on which this method is based [16]. However, the method is also feasible with simple baker's yeast from the supermarket. While we have performed our experiment under aerobic conditions, it is possible to perform the whole process anaerobically with similar results for polyphosphates.
Polyphosphate Assay
To measure the efficiency of polyphosphate (PolyP) production with Saccharomyces cerevisiae, we also used the Phosfinity-Total Polyphosphate Quantification Kit by Aminoverse. In contrast to the assay for phosphate measuring, the buffer used for polyphosphate measuring contains enzymes. The enzymes enable the hydrolysis of PolyP, setting Pi free. Following this hydrolysis, a colorimetric Pi detection as in the Phosfinity Assay can be done.
Light-driven Approach
We decided to use light-driven approaches for our project because light has several advantages over different control systems of biological reactions. Since light is easily provided by LEDs, the costs and needs of resources are low. Additionally, the use of light does not cause the production of waste, as no chemicals or other substances are needed.
Figure 5: Setup of our light-driven approach
For more information on our light-driven approach, take a look at the description section of our wiki.
