Vaccinia virus antigens

Orthopoxviruses are large and brick-shaped virions, which belong to the family Poxviridae, subfamily Chordopoxvirinae [1]. Numerous species of the genus Orthopoxvirus can infect people, including variola, monkeypox, vaccinia, and cowpox [1]. Vaccinia virus can infect vertebrates such as humans, buffalo, cattle, pigs, rabbits, etc [12]. Each infected cell produces two physically different virus forms known as an intracellular mature virus (IMV) and an extracellular enveloped virus (EEV) [4]. Vaccinia virus has intracellular envelope proteins acting as antigens for detection, which are L1, A27, A33, and B5R (or B5). These proteins of the vaccinia virus are cross-reacting with orthopox, monkeypox, and variola viruses, which makes them universal markers for the detection of different poxviruses.

Short information about L1, A27, and A33:

• L1 is 250 aa myristoylated transmembrane protein that has six cysteines. It can be found in intracellular mature virions of the virus [3].  L1 envelope glycoprotein is involved in cell entry because of its similarity with F9 protein which is required for entry and fusion and because of intramolecular disulfide bonds which affect the entry process [6].
• A27 is a protein that affects virus transport, release, and cell-to-cell spread [4]. Mature viruses particularly interact with heparan sulfate through the vaccinia viral protein A27 for cell surface attachment. A17, a viral membrane protein, and A27 collaborate to anchor to the viral membrane [7].
• A33 is a disulfide-bounded homodimer that has 5 cysteines [5]. A33, which is EEV-specific envelope glycoproteins affect the release of EEV from the cell surface or the retention of CEV [4]

There are already existing 3D structures of mentioned proteins, and they were synthesized by our advisers Dr. Luca Vangelista, Yerkezhan Kalikanova, and Kuanysh Seitkamal. Therefore we planned to focus on the synthesis of B5R protein.

Protein B5R is a 42-kDa glycosylated type I membrane protein [8]. The ectodomain is comprised of four Sushi domains (each with two intramolecular disulfide bonds, between C1 and C3, and C2 and C4) with similarity to short consensus repeats (SCRs) including a “stalk” of 51 amino acids located adjacent to the transmembrane region [8]. Protein B5R participates in the efficient wrapping of IMV, formation of actin tail, normal plaque size, and virus virulence [8]. Only a small portion of B5R consisting of the cytoplasmic tail, the transmembrane domain, and the stalk can facilitate protein incorporation into EEV and EEV formation [8]. Therefore, all four SCR domains are not necessary for carrying out all these functions. Thus, we focused on producing both the 1st Sushi domain and all 4 Sushi domains of the protein B5R. Below is the 3D structure of the B5R protein obtained from AlphaFold software.

Figure 1. 3D structure of modeled protein B5 with leader peptides sequence

Considering the available 3D structures and properties, L1, A27, and A33 proteins were expressed in E.coli BL21(DE3) because it lacks proteases that can degrade our recombinant protein. Moreover, it has a specific T7 RNA Polymerase (T7 RNAP) gene suitable for our protein expression [9]. PCR primers were synthesized with specific restriction sites (NdeI/Xhol) suitable for the pET23a vector. L1, A27, and A33 protein sequences were amplified in PCR, and restriction with NdeI and Xhol restriction enzymes was done. Ligation of insert and transformation of E. coli BL-21 (DE3) was performed. Moreover, as a transitional host, we used the DH5α E. coli strain to amplify our plasmids. Then, L1 and A33 were expressed and purified. Mutagenesis was required for A27 then the same expression and purification procedures were followed. All three proteins were successfully expressed. The same procedure was performed with the protein B5R because there was no prior information regarding the expression of the protein.

Recombinant protein synthesis steps

Synthesis of B5R is different than in eukaryotic cells because bacterial cells lack the apparatus for post-translational modifications and protein folding and thus cannot guarantee correct conformation of the final peptide. Synthesis of the B5R protein in prokaryotes is complicated by the multiple cysteine residues that favor the formation of the non-specific disulfide bridges. Expressed proteins must be purified, re-solubilized and re-folded in order to produce functional and competitive antigens. Also, since B5R contains a leader peptide sequence that is cleaved in eukaryotes we suggest not including it in our final inserts. The leader peptide sequence is a polynucleotide region between the promoter and the start codon of the coding region, which is normally not translated and cleaved off by the signal proteases[10, 11]. It is also responsible for the insertion of the proteins into the membrane of the endoplasmic reticulum and usually, the leader sequence is cleaved by signal proteases during or right after the insertion into the membrane [11]. Since this region is not used in translated protein, we decided not to include it in the inserts of B5 protein.

So we suggest 4 different synthesis ways:

• All 4 sushi domains.

Figure 2 Plasmid design of pET23a+ B5R with 4 Sushi domains

• Only 1st sushi domain 

The first sushi domain of B5R already contains the required ectodomain and a smaller peptide size will reduce the complexity of folding in prokaryotic cells.

Figure 3 Plasmid design of pET23a+ B5R with 1st Sushi domain

• Cis 140 residue altered to Ser 

In order to avoid the formation of non-specific disulfide bridges that tether proper folding, after the synthesis of all 4 sushi domains we planned to change the Cis 140 residue to Ser by mutagenesis.

• Codon optimized 

Since codon reading is slightly different between prokaryotes and eukaryotes, it could be possible that some codons are misread and wrong amino acids will be placed in the peptide sequence. Thus, codon optimization was suggested to avoid codon misinterpretation for proper peptide folding. We planned to make codon optimization in IDT online software.

So we planned to test them all, but we were able to test only the first 2 ways.

Optic fiber biosensor with antibodies

Optic fiber is a flexible, transparent fiber made from glass. They are primarily used to transmit light from one end of the fiber to the other, which is widely used to transfer data in telecommunications [12]. But optic fibers can also be used in biomedical applications. Optic fiber-based refractive index sensors have shown an ability to selectively detect biological analytes including proteins, tumor biomarkers, antibodies, and others [13].

There are different types of optic fiber refractive index sensors. Two of the main ones are tilted fiber Bragg grating (TFBG) sensors and fiber tip ball resonator (BR) sensors. TFBGs are a more advanced version of sensors that employ Bragg grating technology, where a short segment of fiber reflects particular wavelengths of light and transmits all others allowing the sensing applications [14]. BRs have a spherical shape that is converted into a resonator with interference fringes, which allows the creation of interference patterns that can be measured and analyzed [14]. In this project, BRs were chosen as a more favorable form of biosensor because they are much faster and cheaper to produce. BRs are made by widely available commercial CO2 laser splicers. Another advantage of BRs is that they have the capability for multiplexing, a measurement of multiple sensing units separated in time, wavelength, or space [15].

Due to the low reflectivity of BR, their spectra are detected using optical backscatter reflectometer (OBR). In this project, we use a reflectometer Luna OBR 4600, which has 0.05 dB resolution and 0.10 dB accuracy.

Figure 4 Refractive index change during antibody/antigen interaction

Optic fiber-based biosensors measure refractive index change in the environment. Luna OBR reflectometer measures the return loss of light that was launched through the fiber and was reflected. When there are no antigens present in the solution, the refractive index has a certain value. In the presence of antigens, antibodies bind to them and the light that gets reflected from this structure has a lower refractive index. This situation is illustrated in Figure 4. (Note: the angles shown are arbitrary).

Figure 5 Optic fiber functionalization

Figure 6 Optic fiber biosensor testing

To detect target analytes, the optic fiber must be functionalized with molecules that can interact with analytes like antibodies, aptamers, and others. The functionalization starts with cleaning the ball resonator with Piranha solution and DI water. Then, they were silanized in the APTMS solution, washed in methanol, and put into the oven. Next,  optic fibers were incubated in glutaraldehyde, which acts like a cross-linker to attach antibodies. Then, optic fibers were washed with PBS and incubated in a solution with antibodies. More in-depth functionalization protocol can be found on the experiments page (LINK). Before the optic fiber biosensor can be used to detect biological analytes, it has to be first calibrated. In this project, the biosensors with target antibodies were calibrated in sucrose solutions at different concentrations. Then, they were used to detect the target analyte. The change in refractive index is measured in decibels.

The focus of our project is to detect Vaccinia virus envelope proteins (L1, A27, A33, and B5R) in risk-prone environments like sewage waters in populated areas. To achieve this, the optic fiber biosensor was first tested in a PBS solution with antigens in the form of envelope proteins. There were two modes of experiments: (1) testing biosensors with a single antibody for each antigen, and (2) testing biosensors with antibodies for all antigens (multiplex). Multiplex biosensors have increased sensitivity for antigens because they are coated with antibodies for all antigens, thus increasing the chances of detection. PBS solution provides an optimal environment for biosensor testing.  After doing the experiments with PBS, we proceeded with experiments in synthetic sewage water. Synthetic sewage water was chosen because it has a known composition, which allows better repeatability of measurements. Similar to experiments with PBS, there was a testing of single and multiplex biosensors. The experiments in PBS were done to test the optic fiber biosensor sensitivity to target antigens, whereas the experiments in sewage water showed whether the biosensor works in harsher conditions, which would be important in real-life applications.

Aptamers

Aptamers are RNA or ssDNA sequences that specifically bind to molecules as proteins or carbohydrates [16]. They are cheaper, more stable, and easier to produce and thus can be reasonably good alternatives to antibodies [16]. Aptamers have already been used to make optic-fiber biosensors and thus, for this project, we plan to make aptamers for the vaccinia virus proteins and use them to make and test aptamer-based optical fiber biosensors [18, 19].

Traditionally, aptamers can be designed by the Systematic Evolution of Ligands by Exponential Enrichment (SELEX) process. This process discards non-binding aptamers and expands binding ones, and is a laborious process that may take an extremely large amount of time. An alternative method is to design aptamer sequences using computer programs that construct aptamer sequences. One such program is Making Aptamer Without Selex, or MAWS, created by an iGEM team from Heidelberg in 2017.

MAWS is written in python programming language and uses a computational approach to construct sequences of aptamers of target proteins. However since they are designed in-silico, naturally the constructed aptamers have not been tested in the laboratory to validate that aptamers bind to the target proteins, thus further laboratory tests are required for this purpose. Nevertheless, even with the conduction of aptamer-validating experiments, and several failures, if one obtains a working aptamer sequence, designing aptamer sequences using MAWS can save a lot of time in comparison to SELEX. For further information, please refer to the Modeling section. The synthesized aptamers are planned to be attached to the optic fiber surfaces by adding the TTTTT sequence in the 5’ site of the aptamer [20]. The principle of attaching aptamers is similar to the functionalization of antibodies.

Figure 7 The functionalization of optic fibers with aptamers