Measurement

Relevance for Other Projects

The ubiquitous occurrence of Vitellogenin in all oviparous species gives this protein a central function in many branches of life [5]. As the first team to characterize Vitellogenin's PAMP affinity and investigate the underlying mechanisms of the protein's function, we have laid the foundation for future research in the field of bee immunology and immunity transfer. Our protocols represent the first step towards creating a robust toolkit for engineering Vitellogenin against a wide range of bee pathogens. We anticipate this field will gain more attention in the upcoming years. This heightened interest is not only due to the Varroa mite; it is also driven by factors such as habitat loss, exposure to insecticides, and the risk factors associated with climate change, including the emergence of new pathogens and invasive species. These multiple threats render bee populations increasingly vulnerable [6]. One way of counteracting these developments is to better equip bees to handle natural threats and stabilize hives against other risk factors.

The importance of bee health in the scientific community is demonstrated by the fact that half of this year's iGEM 'conservation' village is working on different projects concerning bee immunity. For that, our system's versatility extends beyond zymosan binding, allowing for the investigation of Vitellogenin interactions with other PAMPs. For instance, the American and European Foulbrood are further pathogens contributing to the decline of honeybee populations. These diseases are caused by the Gram-positive bacterium Melissococcus plutonius and the Gram-negative bacterium Paenibacillus larvae, respectively [7]. Our measurement system can use PAMPs, such as peptidoglycan from Gram-positive bacteria or lipopolysaccharide (LPS) from Gram-negative bacteria to further optimize Vitellogenin.

In addition, our system is not limited to honey bee Vitellogenin. Vitellogenin can be optimized in all oviparous organisms, including fish and insects, among others. [5]. In the context of an ever-increasing world population with a steadily rising demand for food, our method is extremely relevant. For example, the global demand for edible insects was projected to reach a market volume of 1.2 billion US dollars in 2023. The 'insect protein market' in Europe and North America could potentially reach a value of up to 8 billion US dollars by 2030, making it attractive for large food manufacturers [8]. Furthermore, considering the loss of biodiversity, and more specifically the increasing insect mortality, it becomes urgent to take action [23]. Aquaculture should not be neglected either: in 2020, aquacultures of fish and other aquatic organisms accounted for 49% of the global production within the fisheries sector [9]. Numerous bacteria, including Bacillus thuringiensis, Lysinibacillus sphaericus, and Aeromonas salmonicida, are responsible for diseases in fish and insects, threatening profits, and food supply [10-12].

We want to provide a basis for future teams, offering insights that have not been previously documented in the literature. The role of α-helical in cooperation with DUF1943 for PAMP affinity was previously not described and provides guidance for future researchers engineering efforts. The estimated binding kinetics are also a great basis for assay design. Addressing the unknown binding affinity was one of the primary challenges in our assay design, particularly before conducting the initial FACS runs. At that stage, we were concerned about generating data that might be unusable due to unoptimized PAMP concentrations.

We are also aware that not every team has access to FACS or the resources to use it, so we additionally developed a potential workflow to substitute it with our binding assay. Thereby, we enable teams with limited resources to face the ecological challenges of the future and safeguard the bees. We believe that our contribution in developing novel procedures in directed evolution and advancing the overall understanding of Vitellogenin qualifies us for the special award for the best measurement.

In the following, we present an abbreviated overview of our experimental design to outline the structure of our methods and present related results.

---

Preliminary Considerations

As there is, to the best of our knowledge, no published data available indicating the feasibility of displaying large proteins using yeast surface display, we chose to focus on testing specific domains of Vitellogenin rather than attempting to display the entire protein. After analyzing functional domains, we excluded sections deemed irrelevant to our engineering goal, leaving us with two domains of interest: The α-helical domain and the adjacent DUF1943 domain. Additionally, we displayed a truncated Vitellogenin version lacking the β-barrel domain, poly-serine linker, and ending behind the von Willebrand Factor (vWf), resulting in four constructs. This approach allowed us to pinpoint crucial binding domains, create more manageable constructs and determine whether multiple domains were necessary for effective binding.

For the surface-display, we used the Saccharomyces cerevisiae EBY100 strain. Yeasts offer well-established display systems, and while there are other options like phages, bacteria, insects, and mammalian cells capable of displaying mutant libraries, yeasts provide some distinct advantages [13-16]. Yeasts are easier to cultivate and modify in experiments compared to insect or mammalian cells, which would have posed significant technical challenges. Moreover, yeasts are a eukaryotic expression system, allowing for posttranslational modifications and ensuring correct protein conformation - features not available within phages and bacteria [13]. Considering our uncertainties regarding the requirement for post-translational modifications and concerns about misfolding, yeast emerged as the most reliable option.

---

Experiments

Throughout history, appropriate measurement methods have been crucial to obtain significant results. The renowned scientist William Thomson, also known as Lord Kelvin, famously emphasized, "To measure is to know. If you cannot measure it, you cannot improve it." However, in the context of modern synthetic biology research, experiments have grown in complexity and a wide range of diverse methods are available. Consequently, identifying an appropriate measurement method for a specific experiment has become a critical challenge.

As described above, our primary objective was to establish a system that enables the creation and identification of optimized Vitellogenin´s regarding their PAMP affinity. For that, we needed an efficient methodology for detecting, quantifying, and isolating different variants. Typically, the state-of-the-art approaches employed in protein engineering involve directed evolution and rational design. Given the limited knowledge regarding the function of the different Vitellogenin domains and the residues involved in PAMP recognition we opted to develop a directed evolution approach. Within this framework, numerous methods are available. We chose to utilize Yeast surface display. This method uses the yeast cells' capacity to display proteins on their cell surface. By combining this capability with a random mutagenesis approach, we were able to generate a displayed mutant library with a diverse array of protein variants [13]. Subsequently, through FACS, we were able to identify Vitellogenin mutants with enhanced binding affinity to zymosan (Fig. 1).

As a final step, we planned to express Vitellogenin variants with modified domains and validate whether they indeed exhibited improved binding to zymosan using our developed binding assay.

---

Zymosan Labeling

To enable the screening process through FACS, it was crucial to fragment and label the zymosan with a fluorescent dye (Zymosan Labeling Protocol). As mentioned above, zymosan is a large, water-insoluble polysaccharide (approx. 300 kDa), which is why we initially subjected it to acid hydrolysis [1]. We chose mild conditions to keep the zymosan fragments as large as possible so that they could still be recognized by Vitellogenin as PAMPs while remaining water-soluble. This was accomplished by subjecting them to a solution of 0.5 M HCl at 70 °C for a duration of 2 hours. Since studies showed that after a strong hydrolysis of zymosan, its ability to stimulate inflammatory signals in mammals was lost, we assumed that too small sugar oligos are probably not recognized as PAMPs in bees either [17]. Therefore, the soluble zymosan fragments were ultracentrifuged through a 3 kDa membrane after acid hydrolysis to exclude very small sugar oligos (Fig. 2).

Subsequently, the larger zymosan fragments were labeled with the fluorescent dye fluorescein-5-thiosemicarbazide (FTSC). This dye reacts with the reducing end of sugars, causing them to form hydrazones, which are subsequently reduced by sodium cyanoborohydride (NaBH₃CN) to the fluorescent zymosan derivates (Fig. 3). For ideal labeling conditions, a molar ratio of 8:1 (FTSC to reducing ends) and 70 µL of 0.4 M NaBH₃CN per mg of zymosan, dissolved in a mixture of 7:3 DMSO:HOAc, should be used [18].

After that, the FTSC-labeled zymosan was again passed through a 3 kDa membrane to eliminate any unreacted chemicals from the reaction. Moreover, the FTSC-labeled zymosan was precipitated in ethanol, a solvent in which the FTSC dye is soluble, to ensure the complete removal of all residual FTSC, as its presence could interfere with our experimental designs. To validate the successful acid hydrolysis and FTSC labeling, various methods were employed. Firstly, the acid hydrolysis was verified through Matrix-Assisted-Laser-Desorption-Ionization with Time-Of-Flight analysis of released ions for mass spectrometry (MALDI TOF MS). Further confirmation was obtained by observing fluorescence of the precipitated zymosan under UV light as well as unlabeled zymosan as a negative control.

Results

Acid hydrolysis was verified by MALDI TOF MS. The results are depicted in Fig. 4. Glucose has a molecular weight of 180 Da, and when forming the glycosidic bond, water with a molecular weight of 18 Da is eliminated. Considering that zymosan is a polymer comprised of beta-1,3 - and beta-1,6-linked glucose monomers, it is noteworthy that the peaks in the mass spectrum exhibit a consistent gap of 162 m/z. Furthermore, the data also reveals a decrease in the quantity of zymosan fragments with increasing molecular size. As evident from the data, the hydrolysis indeed proved effective.

Following the successful fragmentation of the zymosan polymer, we proceeded to label the resulting fragments using the fluorescent dye FTSC. The labeled polymer demonstrated pronounced fluorescence at 492 nm, as depicted in Fig. 5. Following the washing steps, the flowthrough also exhibited fluorescence (Fig. 5). This fluorescence decreased with each subsequent wash step, indicating a successful removal of the residual FTSC. Conversely, the negative control, consisting of unlabeled zymosan fragments, did not exhibit any fluorescence signal. Initially, we started with 1 g of zymosan, and after completing the process, we obtained 32 mg of labeled zymosan. However, it's worth mentioning that we could recover 867 mg of zymosan after acid hydrolysis for future use.

---

FACS

The Vitellogenin mutant library was screened using FACS, an essential tool in our protein engineering approach. It allows rapid identification of yeast cells displaying Vitellogenin mutants with enhanced binding to our target molecule, zymosan, based on their fluorescent signals. FACS is advantageous for its high throughput, single-cell resolution, quantitative data, and adaptability to various applications. It efficiently screens large mutant libraries, isolates high-affinity binders, and isolates cells for further analysis or culturing. To quantify the expression of the fusion protein and facilitate the normalization of measured binding affinity, the surface display construct incorporates two immune epitopes as antibody targets. Notably, we utilized unlabeled cell controls along with cell controls labeled with single antibodies and double antibodies as negative controls. Additionally, we conducted each control experiment with fluorescently labeled zymosan to ensure that there was no fluorescence spill-over between channels and that the presence of PAMPs did not impact antibody binding (Fig. 6).

Samples were prepared for FACS by labeling them with both antibodies. We utilized a 9E10 antibody to target the cMyc tag immune epitope, followed by detection using an AlexaFluor^®633-conjugated goat anti-mouse secondary antibody to identify the displayed protein construct. Before screening the library, it was essential to determine the binding kinetics. Without an appropriate PAMP concentration, we could either saturate the different protein variants, regardless of their affinity, or provide the variants insufficient PAMPs to generate an adequate signal for cell sorting. We measured a range of zymosan dilutions between 0.02 µg/mL and 200 µg/mL and used an event ratio to determine the degree of PAMP saturation of the presented domains. Detailed instructions can be found in our protocols. When fitting a curve to our data points we assumed Michaelis Menten-like kinetics.

After establishing the display constructs and transforming them into yeast cells, we were interested which domains were displayed and exhibited PAMP affinities, as only these variants could be subjected to engineering. We induced the display and labeled the cells using the epitope antibodies.

For FACS analysis it is important to understand the different values generated by the measurement. Each signal called ´event´ is measured in different channels, each of which is subdivided into a height, width, and area. Height measures the signal intensity, while width measures the time the event passes through the beam. The integral of the measured event generates the area value. Different combinations of these can be used to investigate the properties of the measured events.

Front scatter (FSC) measures the light passing through the event, while side scatter (SSC) measures the light being refracted sideways. FSC-A vs SSC-A is used to isolate the cells from debris or other impurities. FSC-A indicates cell size, while SSC-A shows the internal complexity, also called granularity. Using this plot, cells of interest are isolated from debris or other dead cells using an area called a ´gate´.

Next, single cells must be isolated from the cell events. Cells can either stick together, or can by chance, be measured as a single event. As the presence of such multicell events would interfere with data interpretation they need to be sorted out. We did this via FSC-A vs FSC-H plots. Cell aggregations would appear as a lower population because of their different FSC-A composition. They have a larger FSC-W value due to their size with a lower FSC-H value, according to which they can be sorted.

After isolating single cells, we analyzed the display and binding properties of the domains. The labeled cells are analyzed via the two fluorescent dyes attached to the secondary antibody (red) and zymosan (green). Cells were divided into non-displaying, displaying the domains with low zymosan affinity and displaying the domains with high zymosan affinity.

This workflow was used for analyzing all constructs. An example for the real measurements is shown below.

Before analysis of each FACS run we plotted SSC-A against time to check the runs for any irregularities. In the case of DUF1943 we saw severe shifts in readings as well as dropouts making the measurement unusable. For further details regarding analysis see our protocols.

Results

Initially, we confirmed the correct display of the selected Vitellogenin domain (Fig. 11). The α-helical domain showed the strongest display, with the larger α-helical+DUF1943 being displayed to a slightly lower degree. To our delight we were also able to observe display of the large Vg_truncated domain. For display and binding analysis, we measured 20.000 events. However, for library sorting, this number would be increased to ten-fold the determined library size to ensure oversampling of the library and prevent the loss of any generated variants.

When labeling the yeast displays with zymosan we discovered that α-helical, previously suggested to be a major contributor to PAMP recognition, showed only minimal affinity [14]. Only after the addition of the DUF1943 in α-helical+DUF1943 did we observe strong PAMP affinity. Vg_truncated also exhibited zymosan binding but did not show enhancement over α-helical+DUF1943 (Fig. 12). Based on these findings we decided to use α-helical+DUF1943 for affinity engineering. It is notable that the function of α-helical+DUF1943 was previously not described in literature. These results enable us and other researchers to focus their engineering efforts on the relevant structures of Vitellogenin.

Before screening the library, we measured the dissociation constant (K_D) of displayed α-helical+DUF1943 for zymosan with a range of zymosan between 0.02-200 µg/mL (Fig. 13) which resulted in a determined K_D of 0.61 µg/mL. Based on this we chose a concentration of 2 µg/mL for library screening as it presented a concentration which should saturate cells to ~75%. We expected that this would improve library sorting as binding cells would be strongly labeled and easier to isolate. Lower zymosan concentrations around K_D or below would be more sensitive but could compromise sorting as binding cells are harder to differentiate from low or non-binding ones.

---

Binding Assay

In the context of directed evolution of the selected Vitellogenin domains, a major objective was to assess whether the improved binding affinity of the mutated domains would also be reflected when incorporating the modified domains into the full protein. To accomplish this, we needed an effective method to measure the binding of a ligand to a protein.

One practical method for quantifying protein binding is the thermal shift assay (TSA). It measures changes in protein thermostability when a binding partner is introduced. [2]. However, a significant drawback of this method was the limited knowledge regarding Vitellogenin, making it uncertain whether binding to PAMPs would indeed increase thermostability. Furthermore, TSA is predominantly used for protein-protein and protein-DNA interactions [2, 19]. Additional methods such as affinity chromatography and nuclear magnetic resonance spectroscopic techniques exist for detecting and quantifying the binding of two molecules [3, 4]. However, in the context of the iGEM competition, we decided against these methods due to the expensive equipment required, which are not available in many laboratories.

Consequently, we developed a tool that matches the requirements of the measurement methods at iGEM (Binding Assay Protocol). Our method is cost-effective, easy to use, reproducible, and highly sensitive. The underlying principle of our binding assay is immunodetection. Immunodetection is a biochemical method for detecting a molecule through the binding of an immune conjugate. This principle is typically associated with techniques like Western blotting or enzyme-linked immunosorbent assays (ELISA) [20].

To identify completely improved Vitellogenins, we established the following experimental setup: First, the PAMP against which binding affinity is measured is immobilized in a well of a 96-well plate. We chose to use a 96-well plate since PAMPs such as zymosan, LPS, or peptidoglycan can be effectively immobilized on the plastic surface through polymer-polymer and hydrophobic interactions [21, 22]. This is achieved by incubating the PAMPs at 60 °C for several hours in the wells. Moreover, using multichannel pipettes enables a high throughput for mutant screening. Once the desired PAMP is immobilized and free PAMPs are removed through washing steps, all remaining free binding sites in the wells are blocked with Bovine Serum Albumin (BSA). This protein is commonly used for saturating excess protein-binding sites on membranes and microplates, to ensure that Vitellogenin solely binds to the PAMP and not the plastic surface. To not only compare relative binding affinities but also to use the previously labeled zymosan in the binding assay, fluorescence can be measured for each well, reflecting the quantity of PAMP per well and thus suitable for normalization. Now the Vitellogenin mutants are added at different concentrations to determine the optimal binding conditions. In our experiment, we implemented a series of negative control combinations to validate the specificity and accuracy of our assay (Fig. 14). These combinations cover a range of scenarios: First, an empty well can be used as a baseline control to ensure that background fluorescence did not occur in the absence of the components (Fig. 14A). To verify that Vitellogenin does not bind to the BSA, a well without immobilized PAMP may be used (Fig. 14J). In addition, it is important to test whether the used antibody binds to the BSA or to the labeled zymosan (Fig. 14L and 14E). Furthermore, all other combinations of the components have to be tested to show that our system only works if all components are present.

After sufficient incubation of the wells, unbound Vitellogenin is removed by further washing, and those wells with mutants exhibiting the highest affinity to the PAMPs should retain the most Vitellogenin. Since our Vitellogenin mutants possess a His-Tag we can use an Anti-His-Tag-horseradish peroxidase (HRP), a highly common antibody in many laboratories, to quantify this. The interaction between the immobilized PAMPs and the introduced Vitellogenin mutants determines the number of Anti-His-Tag-HRP antibodies binding to the Vitellogenin His-Tag, subsequently affecting the catalytic activity of the HRP enzyme, and resulting in fluorescence. The advantage lies in the sensitivity of the fluorescence, allowing for the detection of minute quantities of interacting molecules in a 96-well plate without the need for a large sample volume. In summary, the stronger the binding between the immobilized PAMP, and the introduced protein, the brighter the fluorescence signal. A schematic overview of the whole process is illustrated in Fig. 15.

---

Conclusion

In conclusion, our project's investigation of Vitellogenin's PAMP affinity and its underlying mechanisms represents a significant step in the field of bee immunology and immunity transfer. By characterizing this essential protein, we have established a foundation for future research that extends well beyond our current scope. Our developed protocols not only mark the initial strides toward a robust toolkit for engineering Vitellogenin against a wide range of bee pathogens but also open the door to diverse applications.

The growing focus on bee health issues, driven by factors like Varroa mites, habitat loss, insecticide exposure, and climate change-related risks, highlights the critical significance of our work. Our commitment to address these challenges aligns with the broader scientific community's focus on bee immunity, as exemplified by the substantial presence of bee-related projects in this year's iGEM 'conservation' village. Our system's versatility extends beyond zymosan binding, enabling the investigation of Vitellogenin interactions with various PAMPs. Furthermore, our approach is not limited to honey bee Vitellogenin; it can be applied to optimize Vitellogenin in all oviparous organisms, including fish and other insects. As the world's population grows and food demand continues to rise, our method remains highly relevant. The urgent need to address biodiversity loss and increasing insect mortality further underscores the significance of our work.

In our commitment to fostering scientific progress, we provide a solid foundation for future research teams. Our findings, such as the unexplored role of α-helical in cooperation with DUF1943 for PAMP affinity and the insights into binding kinetics, offer valuable guidance for future research projects in engineering. Moreover, we recognize the diversity of resources among research teams and have developed alternative workflows to facilitate the adoption of our binding assay, ensuring inclusivity. Our method's affordability, user-friendliness, and reproducibility position it as an valuble tool for upcoming iGEM teams, making us eligible for the special award for the best measurement. In this way, we aim to empower future teams to tackle ecological challenges and safeguard vital ecosystems while continuing to expand the boundaries of scientific knowledge.

---

Protocols

PDF: Zymosan Labeling

PDF: FACS

PDF: Binding Assay

---

References