iGEM Leiden | 2019

S.P.L.A.S.H.

Suckerin Polymer Layer to Achieve Sustainable Health

Results

To close the gap in current burn-wound treatments we designed an optimal treatment consisting of a suckerin-based hydrogel linking antimicrobial peptides using cleavable linkers. These linkers, upon cleavage, would allow for the controlled release of antimicrobial peptides or other molecules. Several experiments were designed and conducted to create this system. Further, the groundwork was laid on a synthetic system, allowing for the production of artificial suckerins. First, a quick overview of the results is shown which will then be further elaborated on.

Several research tracks were designed. First, we aimed to create a reliable suckerin production line. We achieved the production of four different suckerin proteins in Escherichia coli and Saccharomyces cerevisiae. To increase the protein yield, two suckerin proteins (suckerin-12 and suckerin-19) were chosen for scaled-up production in one-liter bioreactors. Further, we developed an iGEM RFC10 compatible pET plasmid, called ipET, allowing for enhanced protein production in E. coli. The pET plasmid is the most used plasmid for recombinant protein production. However, it lacks an RFC10 compatible multiple cloning site [1]. We created a modified MCS, thus enabling the use of this plasmid within iGEM.

Secondly, we designed the linker system, based on the use of cleavable linkers, which can release compounds on demand. This could, as described previously, be used for the induced release of antimicrobial peptides and other molecules.

The suckerin proteins consist of two alternating modules that contribute to the physical characteristics of the protein. Module one is rich in alanine, valine, and histidine and contributes to the rigidity of the protein by forming stacking β-sheets [2]. Module two consists of glycine and leucine, forming loops to connect the β-sheets and contributing to the overall flexibility of the protein. We devised a system in which shuffling of the modules is possible to alter the properties of the product. This allows users to generate a designed suckerin of their choice by employing the biobrick system of hierarchical combination. Thereby matching the protein product to the needs of the researcher and patient (see human practices). We also devised a system, where each module would be induced by a different induction molecule, which would then create a new, customized suckerin protein, based on the individual levels of induction.

Figure 1. A schematic overview of a suckerin protein showing the different modules

Our goal was to follow these research lines to allow researchers to design an artificial protein and hydrogel that would be applicable for burn-wound treatment. The structure of the gel would allow for incorporation of functional molecules, such as antimicrobial peptides, to counteract potential infection risks.

Overview

Project Achievements

In the Experiments section, all our planned experiments are described, thereby allowing future iGEM teams and research groups to continue with suckerin research. Here, we have divided our research into 8 projects, which summarizes our efforts in the lab during the summer. The ‘+’ sums up well-validated and documented results, while ‘-’ indicates an earlier stage of research. These sub-projects were briefly touched upon, but due to time constraints, were not completed. However, we still provide all results to help future projects.

Suckerin synthesis

+ Synthesized complete suckerin-1, -6, -8, -9, and -12 genes in the RFC10 Biobrick system, codon-optimized for E. coli and S. cerevisiae, and devoid of EcoRI, XbaI, SpeI, PstI and NotI sites: BBa_K3041002, BBa_K3041006, BBa_K3041001, BBa_K3041005, BBa_K3041000, BBa_K3041004.

+ Synthesized complete suckerin-8 and -12 genes in the RFC10 Biobrick system, codon-optimized for S. lividans and devoid of EcoRI, XbaI, SpeI, PstI and NotI sites.

- Unable to manufacture suckerin-19, codon-optimized by iGEM Leiden, due to too high complexity in repeating units.

- Unable to manufacture S. lividans codon-optimized suckerin-1, -6, and -9 due to too high GC contents.

Suckerin-19 production in E. coli

+ Received the pQE80-L SRT-19 production plasmid from the Miserez group in Singapore.

+ Validated production of suckerin-19 in the E. coli strain Rosetta with pQE80-L SRT-19 under induction of IPTG, by his-link purification on SDS-PAGE: BBa_K3041003.

+ Scaled up production for optimization of protein yield.

- Obstructed Biobrick system integration due to difficulties in nucleotide alignments of repeating units during PCR.

E. coli suckerin production

+ Suckerin- 8, -9, and -12 amplified with standard RFC10 prefix and suffix primers.

+ Inserted in pBS1A3, succeeded by insertion of expression cassette P_Lac, creating new composite biobricks: BBa_K3041015, BBa_K3041016, BBa_K3041017.

+ Small and validated scaled-up production of suckerin-12 by E. coli strain Rosetta under induction of IPTG and by inclusion body purification.

+ Implemented results from dry-lab to lay groundwork for media optimization.

- Unable to form the hydrogel due to too low overall protein yield.

- Unsuccessful insertions of the promoter pBad.

Model implementation: Increased suckerin production

+ Prediction of model implemented by growth assessment in SPG medium supplemented with glucose.

+ Increased suckerin-19 yield after model implementation, visualized with His-Link purification on SDS-PAGE.

+ Results support the model.

ipET

+ iGEMized production plasmid by the construction of a new MCS into the standard RFC10, introducing an N-terminal His₆-tag, RBS, and Thrombin sites: BBa_K3041014.

+ Suckerin-8, -9, and -12 amplified with adjusted RFC10 pre- and suffix primers to lose the start codon ATG and stop codon, so N-terminal His₆ and C-terminal GFP could be tagged to the protein.

- Unable to integrate products into ipET because of unknown error.

Suckerin customized modules

+ Customized Module 1 (M1) and Module 2 (M2) designed based on MUSCLE protein alignment of 6x module sections in suckerin-12 and 11x module section in suckerin-19: BBa_K3041007, BBa_K3041008.

+ Synthesized M1 and M2 for RFC25 system compatibility, introducing NgoMIV and AgeI sites for the ability of adding multiple hetero modules (M1-M2), including an A’G’TG site between RFC10-RFC25 prefix for adjustable start codon site by amplification, and separated by BamHI sites for translated serine scarring to reduce the effect on the folding of the protein.

+ Customized M1-adapter was produced by adding an adapter sequence downstream containing AgeI, BamHI and PstI sites. This module could accommodate the insertion by restriction cloning using NgoMIV-AgeI of any number of M1 modules.

+ Customized M2-adapter was produced, similar to M1-adapter. The adapter sequence, solely containing a NgoMIV site was added upstream. This allowed for the insertion of any number of M2 modules.

+ in silico work resulted in the formation of M1-M2, M1-M1 and M2-M2 to demonstrate the correct assembly and in-frame insertion of the modules: BBa_K3041010, BBa_K3041009, BBa_K3041011.

- Customized M1 and M2 system were based on ipET, lacking an ATG and stop codon, however, transformations were unsuccessful due to an unknown error.

Linker system

+ Parts of protease substrates of V8 and SplB were designed and created to function as a cleavable linker system: BBa_K3041012, BBa_K3041013.

- Composing the linker system suckerin + linker + GFP got constrained due to its design for compatibility in ipET, which was not functioning in time.

- Streptavidin monomers were synthesized (based on BBa_K2668010) for a high-affinity biotin linker system, but in order to narrow the research line down, they were dropped early in the project

Suckerin production in S. cerevisiae and S. lividans

+ S. cerevisiae suckerin-8, -9, and -12 amplified with standard RFC10 prefix and suffix primers: BBa_K3041004, BBa_K3041005 and BBa_K3041006.

+ S. cerevisiae suckerins successfully inserted in yeast production plasmids pYES2 and pMU-his, validated by restriction digestion.

+ S. lividans suckerin-8 and -12 amplified with standard RFC10 prefix and suffix primers

- Suckerins not inserted in Streptomyces plasmid pSET152 in order to narrow down the project, due to time constraints.

- His-link and inclusion body purification after production was non-optimized for yeast, resulting in non-conclusive SDS-PAGE gels.

Summary

Severe burn wounds are currently the cause of the debilitation of millions and 300,000 deaths per year. To find solutions to current treatment problems, iGEM Leiden investigated the potentials towards developing a "smart" hydrogel composed of suckerin proteins, originally isolated from the Humboldt squid, Dosidicus gigas. Using Escherichia coli as a production host and engineering various suckerin proteins using the biobrick system, iGEM Leiden successfully produced the prominent hydrogel candidates suckerin-12 and suckerin-19 on small and bulk scale.

Groundwork has been laid on the development of several "smart" linker systems that could achieve desired antimicrobial features. Customizable suckerin has been designed based on protein sequence alignment of two modules that give the protein both strong and flexible features, thereby enabling the production of customized wound dressings in the future. To achieve even higher production of the natural suckerins and to contribute to the iGEM community, E. coli protein production plasmid pET28 was adjusted to meet the RFC10 system requirements.

Our research shows that suckerin proteins overcome the production problems that other candidate polymers, such as spider silk, face, indicating that suckerin can be a biomaterial with high potential for a plethora of other (medical) applications. Therefore, as reported by our interesting results, we strongly encourage future iGEM teams to take suckerin into consideration as a possible project with a wide variety of applications.

Future prospects

Within this project, we successfully constructed the production of suckerin-8, -9, -12, and -19. In this paragraph, we will address potential improvements upon this project for future utilization.

As described in our experiments and human practices pages, we have discussed the use of several other model organisms including Streptomyces lividans and Saccharomyces cerevisiae. We were successful in designing the suckerin biobricks for both organisms. Unfortunately, the production of the proteins got constrained due to the limited time and non-optimized purification methods, resulting in inconclusive bands. We encourage future iGEM teams to work on optimization of the production host for this protein and to test our designed and submitted constructs.

Overall, the suckerin production was insufficient to form a hydrogel. With the use of mathematical modeling, we optimized both the medium composition and metabolic pathways in E. coli to achieve higher production for future experiments.

With more purified protein, the hydrogel formation could be explored further. Here, multiple tries to optimize the rigidity of the final product should be attempted. Once this is achieved, the linker system, explained in the paragraphs above, needs to be finalized. If this has a positive result, bulk production of the suckerin-based hydrogel containing the linker could be initiated. This would result in multiple gels, which can be used to link several beneficial molecules.

To further increase the applicability of the system, the obtained gels with a linker could also be used to test for a possible immune response of the body against the hydrogel, as suggested by several experts. Suggested experiments would be a scratch assay, for instance. Furthermore, a small piece of the hydrogel could be placed on a plate with human cells, and pieces of skin can be taken to create an ex vivo wound model as suggested by the experts we interviewed in our human practices.

To assess the environmental impact of our product we decided to test its biodegradability in various environments, such as water and soil. Here, the planned experiment involved measuring the time it would take for complete degradation of the suckerin-based hydrogel in the environment as a measure of biodegradability. This would have an immense contribution to the usability of our product, as biodegradability of the gel would be in line with the sustainability goals that we set to ourselves. However, as we were not able to produce a hydrogel we were not able to proceed with these tests.

In the end, it can be concluded that, although many experiments already set a base, a large number of future experiments could still be conducted. We were able to produce suckerin-8, -9, -12, and -19 and achieved bulk production of both suckerin-12 and -19 by E. coli. Our accomplishments resulted in great confidence in the suckerin protein and our product, therefore we encourage future iGEM teams to take suckerin into consideration if they want to explore biomimetic materials.

Production

Suckerin synthesis

The selection of available sequences of suckerin proteins were based on characteristic modular architectures. An alignment of constitutive suckerins of various cephalopod species classified the general modular placement into 6 clades [2]. Most suckerin proteins are composed of flexible silk-like GGY repeats (M2) with alternating rigid β-sheet forming Ala-rich regions (M1), in which the Humboldt squid (Dosidicus gigas) suckerins are characterized in their composition of consecutive repetitive modules.

Variation in the repetition of these modules, contributing to the modularity and hardness of the protein, are represented in 6 clades. In this project, we selected 6 candidate suckerins (Fig. 1): suckerin-12 and -19 consisting of regular spaced modules (clade 1); suckerin-1 and -9 composed mainly of flexible M2 modules (clade 2); suckerin-8 dominated by the rigid M1 modules (clade 3); and suckerin-6 constituting irregular placed modules (clade 6).

Protein sequences were reverse translated and codon-optimized for production in Escherichia coli, Saccharomyces cerevisiae and Streptomyces lividans. Gene structures were enclosed into the RFC10 Biobrick system, devoid of EcoRI, XbaI, SpeI and NotI restriction sites, and synthesized by IDT. Due to the complexities of increased repetitive regions, IDT was not able to manufacture suckerin-19 optimized for all three organisms. Besides this, the high GC content in the S. lividans constructs obstructed further synthesis of all proteins with the exception of suckerin-8 and -12. The project was proceeded with the main focus on constructs for E. coli production, namely suckerin-1, -6, -8 (BBa_K3041000), -9 (BBa_K3041001), and -12 (BBa_K3041002). Parallel to this, we tested S. cerevisiae production of suckerin-1, -6, -8 (BBa_K3041004), -9 (BBa_K3041005), and -12 (BBa_K3041006), and S. lividans production of suckerin-8 and -12.

Figure 2. Protein sequences of selected suckerins from Dodidicus gigas with respective modular architecture [2].

Suckerin-19 production in E. coli

Suckerin-19 was considered as one of the most interesting candidates for the production of a hydrogel. Initial attempts to synthesize the suckerin-19 gene were not successful due to its highly repetitive sequence. To produce the protein nonetheless, we contacted the Miserez group from Singapore, as they had already achieved successful production and purification of suckerin-19 [2, 3]. They agreed to send their pQE80-L SRT-19 plasmid, allowing us to produce suckerin-19. This plasmid contained the suckerin-19 gene linked to a His₆-tag, allowing for purification using a Nickel-column, while being under the regulation of the Lac promoter (P_Lac). Several attempts were made to amplify the suckerin-19 gene via PCR in order to implement the RFC10 Biobrick system, however, these attempts failed, likely due to repeats present in the gene.

The pQE80-L SRT-19 plasmid was transformed into E. coli Rosetta (Rosetta^TM (DE3) BL21), designed to enhance the expression of eukaryotic proteins that contain codons rarely used in E. coli [4]. Transformants were grown in shake flasks containing 100 mL standard LB medium. When growth reached the exponential phase, production was induced with isopropyl β-D-1-thiogalactopyranoside (IPTG), which mimics allolactose and activates the Lac operon. The protein product was purified according to His-link purification using a Promega kit. After purification, the samples were visualized on SDS-PAGE gel (Fig. 3). In this way, the successful production of suckerin-19 was confirmed.

Figure 3. SDS-PAGE gel showing the presence of suckerin-19 after His-link purification.
Lane 1: ladder, Lane 2+3 suckerin-19 produced by Rosetta induced with 0.4 mM, Lane 4-6: suckerin-19 produced by Rosetta induced with 1.0 mM The bands have the expected length of 39 kDa.

For the formation of a hydrogel, we would need gram-scale quantities [5]. Therefore, increased protein production and purification were required. For this, we scaled up the protein production to one-liter bioreactors, so high quantities of cells could be harvested. Bioreactors were handled as described in the protocols. Briefly, this included bioreactor calibrations, adjusting the pH to 7 and the temperature to 37°C, as well as implementing a dissolved oxygen (DO) cascade to keep the DO at a constant 40% by adjusting agitation with a minimum agitation of 300 rpm. The cells were inoculated and grown to an exponential growth phase (OD₆₀₀ of 0.6-0.8), before induction with IPTG. Since the bacteria produced acids prior to induction and bases after induction, the pH was continuously monitored and adjusted with 2 M HCl and 2 M NaOH to maintain a constant pH 7.

In order to implement high throughput purification, we changed to a column-less protein purification protocol, since the His-link purification via the Promega kit only allowed for low volumes per column. Since the bacteria likely stored the protein in inclusion bodies, we applied an inclusion-body protocol, adjusted from a provided protocol by the Singapore group and literature [3-5]. Inclusion body purification is based on the storage of the suckerin proteins in this intracellular compartment. As shown in figure 4, the presence of the protein could be confirmed via SDS-PAGE and showed a relatively pure protein. Unfortunately, even in the scaled-up production, the overall yield was low, resulting in only a few milligrams of purified protein. This was insufficient to form a hydrogel. Therefore, developed a model of E. coli metabolism, including the suckerin-19 reaction, to optimize production conditions (see model page).

Figure 4. Suckerin-19 bioreactor batch purified with inclusion body purification.
Suckerin-19 visualized on SDS-PAGE gel by the presence of the expected 39 kDa bands.

Production of other suckerin proteins in E. coli

The successfully synthesized suckerins, codon-optimized for E. coli, were PCR amplified using the standard biobrick primers. In figure 5, the PCR products of these specific suckerins were visualized on an agarose gel. Suckerin-9 and -12 are represented as single bands. Suckerin-8, however, shows strong aggregation of the fragments, which is probably the result of amplification of the full sequence provided by IDT. To optimize gene synthesis we ordered some proteins together, separated by a BamHI site for later separation. However, in some cases the other genes were also amplified, due to the use of the standard biobrick primers in the PCR reaction. This was the case for suckerin-8. The correct band for suckerin-8 was therefore gel-purified before cloning.

Figure 5. Amplified suckerin genes 8, 9, and 12.
Polymerase chain reaction (PCR) products amplified with general prefix and suffix primers, shown by gel electrophoresis. Lane 1: 1kb ladder. PCR product at different concentrations of suckerin-8 (408 bp, lane 2 and 3), suckerin-9 (568 bp, lane 4 and 5), and suckerin-12 (696 bp, lane 6 and 7). The gel confirms the amplification of the desired suckerin proteins.

The corrected genes were inserted in the pBS1A3 plasmid. After successful transformation, colonies were selected and stored in liquid stock. After checking the plasmids, the P_Lac expression cassette was placed in front of the suckerin genes. These constructs were checked by PCR with standard biobrick primers, placed on an agarose gel and transformed into E. coli. The agarose gel in figure 6 shows the fragments of all three different suckerins together with the Lac promoter. The bands of suckerin-12 are very faint, still, we decided to continue the transformation.

Figure 6. PCR of P_Lac-suckerin genes for plasmid validation.
Lane 1: 1kb ladder, Lane 2+3: P_Lac-suckerin-8, lane 4+5: P_Lac-suckerin-9, lane 6+7: P_Lac-suckerin-12

E. coli DH5α colonies expressing the suckerins on pBS1A3 plasmids under the Lac promoter were successfully obtained, resulting in the constructs pBS1A3-P_Lac-suckerin-8, pBS1A3-P_Lac-suckerin-9 and pBS1A3-P_Lac-suckerin-12. These strains were used for initial protein production. 100 mL flask cultures were induced at an OD₆₀₀ of 0.6-0.8 with 1 mM IPTG. After harvesting, the proteins were purified using the inclusion body purification protocol, since these constructs did not contain a His₆-tag. Figure 7 visualizes the successful production of the three suckerins on SDS-PAGE. It is noteworthy that the protein mixture obtained by the protocol was not dialyzed, resulting in an impure product.

Figure 7. On SDS-PAGE gel there is visible suckerin protein purified from E. coli, for each of the different suckerin proteins.
Lane 1: ladder, suckerin-12 (23 kDa, lane 3), suckerin-9 (19 kDa, lane 4) and suckerin-8 (17 kDa, lane 5). Lane 2 contains the purification from suckerin produced in Saccharomyces cerevisiae. However, the protein was purified with a protocol optimized for bacteria resulting in the purification high amount of non-specific proteins, making these bands largely inconclusive.

Successful confirmation of suckerin-8, -9, and -12 production motivated us to proceed toward scaling up the bioprocessing to larger volumes. Considering the recent publication on hydrogel formation [4], and its consequent repetition of modules architecture similar to suckerin-19, we selected suckerin-12 as the optimal hydrogel candidate. In order to improve protein production, the E. coli Rosetta colonies were transformed with pBS1A3-P_Lac-suck12 and pre-cultured overnight prior to inoculation. After harvesting the cells, the protein was purified using inclusion body purification with dialysis and lyophilization. Figure 8 shows an SDS-PAGE of the dialyzed suckerin-12 protein mixtures, demonstrating clear bands at around 23 kDa. It was attempted to produce a centimeter-scale hydrogel. Unfortunately, due to low overall suckerin yield and time constraints, we were not able to produce such hydrogel.

Figure 8. SDS-PAGE gel of suckerin-12 after purification and dialysis.
Lane 1: Ladder, Lane 2+3: Irrelevant, Lane 4+5: suckerin-12 protein mixture

Model implementation

Upscaling of suckerin-19 and suckerin-12 production in bioreactors had been relatively unsuccessful. The yield remained low throughout the process. Since the use of LB media yielded low quantities of our protein, we created a mathematical growth model of E. coli complemented with a metabolic model based on the production of suckerin. This model stated that, in minimal Studier Phosphate Glucose (SPG) medium containing 5 grams of glucose per liter, glucose was the limiting factor in the accumulation of biomass, correlating with low protein yields [6, 7]. Based on the model, which predicted increased suckerin production at increased glucose concentration, we performed growth assays using LB, as well as SPG containing 5, 10 and 20 grams of glucose per liter. We tested the final optical density at 600 nm (OD₆₀₀) to assess the biomass formation after 24 h growth, as well as the protein production pattern in each medium upon induction.

The media were inoculated in duplicates with E. coli Rosetta containing the pQE80-L SRT-19 plasmid and set to an initial OD₆₀₀ of 0.1. We used this particular plasmid because it was the first functional plasmid used to produce suckerin in our project. The plasmid contains a His-link, which allows for fast and accurate protein purification. After overnight growth, cultures were diluted to an OD₆₀₀ of 0.1 and induced with IPTG in the exponential growth phase, OD₆₀₀ of 0.6-0.8. However, the LB cultures were induced at an OD₆₀₀ of 1.05, because of their higher growth rate compared to the SPG medium. The final OD₆₀₀ was measured 24 hours post-inoculation.

The measurements of the optical density (Fig. 9) shows a strong similarity to the prediction of the model. Higher concentrations of glucose increased the total biomass. However, this effect decreased with the highest concentration of 20 grams per liter, possibly due to osmotic stress as a result of the high glucose concentration. The OD₆₀₀ measurements of the SPG cultures remained lower than the LB cultures most likely due to less nutritional sources in SPG medium. The main metabolic components of LB are tryptone, yeast extract, and sodium chloride, while SPG medium consists primarily of glucose, ammonium, sulfate, and phosphate. The SPG medium lacks important sources of vitamins and nitrogen.

Our model predicts the growth rate instead of biomass accumulation. Therefore, the measurements of the final OD₆₀₀ are not the optimal validation method. Future experiments that focus on bacterial growth rates should be conducted, but were not performed during this project, due to time constraints.

Figure 9. Comparison of OD₆₀₀ of cultures grown in LB, as well as different formulations of SPG with varying glucose.
The OD₆₀₀ was measured 24 h after initial inoculation. The data was obtained in biological duplicates. The comparisons were made using a parametric t-test, p-values are shown. A significance level of 0.05 was chosen.

Visualization of suckerin-producing bacteria

Osmotic and/or other types of stress could be observed visually under the microscope. Figure 10 shows representative images of the cultures after 24 hours of growth with IPTG induction. The cells show diverse shapes and elongation in the SPG media containing 10 g/L and 20 g/L glucose, likely due to osmotic stress. This is rarely observed in LB or SPG medium containing 5 g/L glucose cultures.

Figure 10. Microscope images of bacteria after 24 h of growth in different media.
The images are representative of the overall sample and show a magnification of 40x.

Suckerin-19 production validation

Next, protein production was measured in all tested media. For this purpose, the cells were harvested, resuspended and the produced suckerin-19 was purified with a His-Link purification kit from Promega. Figure 11 shows the suckerin-19 production of each culture on an SDS-PAGE gel. Strikingly, there were hardly any proteins found in the LB cultures, likely due to the late induction of the production. Another reason could be the viscosity of the medium. LB medium tends to become more viscous during fermentation, complicating protein purification. Minimal medium is less viscous, which simplifies protein purification. In all SPG glucose concentrations are clear bands visible, indicate the presence of the suckerin-19 protein (39 kDa).

Figure 11. SDS-PAGE of proteins from cultures grown in LB and SPG media formulations and purified using a His-Link kit.
Lane 1: Ladder, 3+4: LB, 5+6: SPG + 5 g/L glucose, 7+8 SPG + 10 g/L glc, 9+10: SPG + 20 g/L glc. Suckerin-19 has a mass of around 39 kDa.

In summary, to overcome the low suckerin-19 yields after fermentation in LB, a model was created predicting that higher glucose concentrations in SPG medium correlate with higher biomass formation. This was validated by a growth assay using SPG medium supplemented with 5, 10 and 20 grams of glucose per liter. However, these high concentrations of glucose did not result in final optical densities as high as the cultures grown in LB. The suckerin production, however, was found to be highly increased in our glucose supplemented minimal media compared to the cells grown in LB, indicating a strong effect on the protein production optimization with this medium composition. Therefore, we propose to use this composition in the production media for future experiments to harvest enough suckerin for hydrogel formation, which we were unable to perform due to time constraints.

ipET

Not only the type of medium, production strain and other growth conditions influence the protein yield, but this can also be achieved by using plasmids optimized for protein production. To create a production plasmid complying with the biobrick system, iGEM Leiden designed the iGEMized "ipET" plasmid (BBa_K3041014), derived from pET28(+) (Fig. 12). This plasmid enables optimal recombinant protein production of RFC10 biobricks and His-link purification in E. coli strains. ipET was obtained by adjusting the multiple cloning site (MCS) of pET28a(+) into the prefix and suffix, allowing for direct cloning of your RFC10 biobricks into this high production plasmid. The gene will automatically be placed under control of the T7/Lac promoter allowing for induced production after the addition of IPTG (0.5 mM - 1 mM). In order to create the multiple cloning site, two different oligo pairs were designed, which were subsequently annealed and placed into the pET28. Due to the chosen restriction of the pET28 some original features of pET28 were lost, including the ribosome binding site, His-tag and thrombin site. To overcome these problems, the features were added again to the new MCS, resulting in the MSC followed by a His₆-tag, TEV protease cleavage site and the original ribosome binding site. To obtain the new “igemized MCS” two separate parts were designed, separated by a NcoI site, by the annealing of four short synthesized oligos (Fig. 13). At last, the complete multiple cloning site was placed in the ipET plasmid.

In order to validate the newly obtained plasmid, test digestions were conducted. If linearization of the plasmid could be observed it was concluded the restriction sites were intact. Digestions were performed with EcoRI, XbaI, and PstI and they all resulted in linearization of the plasmid (Fig. 14). Furthermore, the plasmid was sent to BaseClear for sequencing using two primers, annealing up- and downstream of the new multiple cloning site. The sequences showed strong alignment with the theoretical sequence suggesting that the new multiple cloning site was correct (Fig. 15). Unfortunately, we were not able to test this plasmid by cloning a biobrick into it, due to time constraints.

An additional PCR was performed to adjust suckerin-8, -9, and -12, making them in frame with the ipET plasmid. These primers contained the standard biobrick prefix and suffix and could also be used to amplify the suckerin-8, -9, and -12 biobricks.

Figure 12. The plasmid map of the adjusted ipET plasmid.
The map displays all restriction sites of the multiple cloning site, the kanamycin resistance marker and the T7 promoter linked to the Lac operator.

Figure 13. Annealing of multiple cloning sites part 1 and 2.
Lanes 1-4: testing the un-annealed oligos, lanes 6-7: the annealed oligo’s. All samples were loaded onto a 1% agarose gel and run for 40 min at 100V.

Figure 14. Validation of ipET restriction sites.
Lane 1: undigested plasmid, lane 2: plasmid linearized by XbaI digestion, lane 3: ipET linearized by EcoRI digestion, lane 4: the plasmid linearized by PstI digestion, lane 5: ladder.

Figure 15. Sequence alignment of RFC10 multiple cloning site.
Two samples using a forward and reverse primer were sent for sequencing. Both results showed strong alignment with the multiple cloning site.

By creating this plasmid we allow future iGEM teams to achieve upregulation of their protein of interest while following the biobrick system. The plasmid contains several features allowing for successful protein production in E. coli. These include a strong T7-Lac promoter enabling strong and inducible transcription of the protein using IPTG as an inducer. The plasmid also contains a His₆-tag, allowing for improved purification using either a copper or nickel-column combined with imidazole wash. In this way, protein production and purification will be fast and accurate. The thrombin-site can be used to cleave the His₆-tag after purification. Of note, this plasmid will only produce in specified E. coli strains, such as BL21 (DE3) or Rosetta, as these strains contain a T7 polymerase required for binding to the promoter.

Suckerin production in S. cerevisiae and S. lividans

In addition to production in E. coli, we tested other chassis for increased production of suckerin proteins. Specifically, we tested Saccharomyces cerevisiae and Streptomyces lividans as possible recombinant hosts.

The gene constructs of suckerin-8, -9, and -12 were codon-optimized for the two strains. We were successful in inserting the suckerin-12 gene into the pMU-His plasmid for production in S. cerevisiae (Fig. 16)

Figure 16. Control digestions with PstI of pMU-His with suckerin-12.
Lane 1: 1kb ladder, lane 2-5: irrelevant, lane 6-12, 14 pMU-His without suckerin, lane 13: pMU-His-suckerin-12.

pMU-His contains a Gal1-promoter that allows for induction with galactose, as well as a His₆-tag for efficient purification. The S. cerevisiae culture was grown in MY medium and induced using galactose. After an additional growth period of 7 hours, the protein was purified using the His-link purification kit from Promega (Fig. 17) and the inclusion body purification method (Fig. 18). We realized too late that the protocols were optimized for E. coli and thus not applicable for S. cerevisiae. Therefore, visualizing the obtained protein mixture on an SDS-PAGE gel did not show a conclusive product. Future experiments should aim to optimize both protocols for this chassis.

Although the genes were synthesized, we did not attempt to transform S. lividans, due to time constraints.

Figure 17. SDS PAGE of yeast proteins after his-link purification.
Lane 1: ladder, lane 2-4: His-link purified protein mixture

Figure 18. SDS-PAGE of yeast protein mixture after inclusion body purification.

Customization

Suckerin customized modules

As described above, suckerin proteins are composed of two alternating modules, M1 and M2. Each module is about 10 residues long. M1 consists of an Ala- and His-rich sequence that folds into β-sheets. These β-sheets determine the rigidity of the protein. M2 is an intrinsically disordered region with high flexibility composed exclusively of Gly, Tyr and Leu residues. The alternating combination of modules in suckerin gives the protein its characteristic biophysical properties (high tensile strength and great extensibility) [8]. We designed two synthetic M1 and M2 sequences through MUSCLE alignment using 6 modules from suckerin-12 and 11 modules of suckerin-19 as input (Fig. 19). M1 (BBa_K3041007) and M2 (BBa_K3041008) were synthesized using the RFC25 Biobrick format (Fig. 20). M1 contains the prefix and a downstream BamHI site, while M2 contains an upstream BamHI site and the suffix. Consequent assembly by restriction cloning using BamHI resulted in a completed RFC25-compatible biobrick encompassing prefix-M1-M2-suffix (BBa_K3041010). The BamHI site was chosen since the scarring caused by this restriction enzyme is Gly-Ser, two flexible amino acids with little effect on protein folding.

The M1 biobrick (BBa_K3041007) was amplified with a forward primer targeting the prefix and a reverse primer to add the adapter sequence. The adapter sequence would add AgeI, BamHI and PstI sites downstream of the coding region. The PCR amplification using these primers would result in M1-adapter. This module could accommodate the insertion by restriction cloning using NgoMIV-AgeI of any number of M1 modules. To demonstrate the correct assembly and in-frame insertion of modules, in silico work was performed resulting in the formation of M1-M1 (BBa_K3041009). The PstI site was included so that a combination of modules finishing in M1 could also be added to an iGEM plasmid.

Similar to M1-adapter, M2-adapter was created. In this case, the forward primers carried the adapter sequence, while the reverse primers targeted the suffix. The adapter sequence contained solely a NgoMIV site. The function was to allow the insertion of any number of M2 modules. To demonstrate the correct assembly and in-frame insertion of modules, in silico work was conducted resulting in the formation of M2-M2 (BBa_K3041011). With the parts displayed here, we could have theoretically produced any permutation of M1 and M2. Unfortunately, due to time constraints, this was only performed in silico (Fig. 21). The parts were nonetheless placed into the registry (BBa_K3041009, BBa_K3041010, BBa_K3041011).

Figure 19. Module 1 and Module 2 protein alignment, respectively.
Alignment of 17 protein sequences from suckerin-12 and -19 with MUSCLE.

Figure 20. Agarose gel of triplicate PCR reactions of the module sequences.
Lane 1: ladder, lane 2-4: Module 1, lane 6-8: Module 2.

Figure 21. in silico design of module assemblies.
A) Ligation result of module 1 (M1) and module 2 (M2) after BamHI digestion. B) Two adjusted M1 constructs digested with either AgeI or NgoMIV were ligated to form M1-M1. DNA sequence of adjusted M1 visualized underneath. C) M2-M2 Ligation product with two adjusted M2 constructs digest with either AgeI or NgoMIV digested. DNA sequences of the adjusted M1 and M2 are shown underneath the ligation assemblies.

Linker system

To make our hydrogel responsive to pathogenic infections we incorporated a cleavable linker downstream of the suckerin proteins. The linker consisted of a short amino acid sequence that acts as the cleavage site by extracellular proteases produced by Staphylococcus aureus. S. aureus is an opportunistic pathogen found in most wound-derived skin infections [9]. The two proteases we targeted were V8 and SplB [10,11]. We made an RFC25 compatible biobrick of each linker and planned to create a composite part containing suckerin + linker + GFP. As a proof-of-principle, we would have created a hydrogel with this suckerin-linker-GFP and added V8 and SplB enzymes independently. If GFP appeared in the supernatant, we would have been able to demonstrate that our suckerin hydrogel is responsive to S. aureus infection via the extracellular proteases secreted by this pathogen. However, due to time constraints, the construct could not be produced, and the proof-of-principle could not be achieved.

References

Mierendorf R. C., Morris B. B., Hammer B., & Novy R. E. (1998). Expression and Purification of Recombinant Proteins Using the pET System. Molecular Diagnosis of Infectious Diseases, 13, 257–292.
Hiew S., & Miserez A. (2017). Squid Sucker Ring Teeth: Multiscale Structure-Property Relationships, Sequencing, and Protein Engineering of a Thermoplastic Biopolymer. Acs Biomaterials Science & Engineering, 3(5), 680-693.
Ding D., Guerette P., Hoon S., Kong K., Cornvik T., Nilsson M., . . . Miserez A. (2014). Biomimetic production of silk-like recombinant squid sucker ring teeth proteins. Biomacromolecules, 15(9), 3278-3289.
Novogen (2003). Transformation/Transfection: Competent Cells.
Buck C., Dennis P., Gupta M., Grant M., Crosby M., Slocik J., . . . Naik R. (2019). Anion‐Mediated Effects on the Size and Mechanical Properties of Enzymatically Crosslinked Suckerin Hydrogels. Macromolecular Bioscience, 19(3), N/a.
Studier F. (2005). Protein production by auto-induction in high-density shaking cultures. Protein Expression and Purification, 41(1), 207-234.
Thakur C., Brown S., Sama M., Jackson E., & Dayie J. (2010). Growth of wildtype and mutant E. coli strains in minimal media for optimal production of nucleic acids for preparing labeled nucleotides. Applied Microbiology and Biotechnology, 88(3), 771-779.
Biophysical Society (2016). Unlocking the secrets of squid sucker ring teeth. EurekAlert! Available: https://www.eurekalert.org/pub_releases/2016-02/bs-uts021916.php. "Accessed: 13-Oct-2019".
Healy B., & Freedman A. (2006). Infections. BMJ, 332(7545), 838-41.
Yamamoto H., Saito S., Sawaguchi Y., & Kimura M. (2017). Identification of Protease Specificity Using Biotin-Labeled Substrates. The Open Biochemistry Journal, 11, 27-35.
Dubin G., Stec-Niemczyk J., Kisielewska M., Pustelny K., Popowicz G. M., Bista M., . . . Potempa J. (2008). Enzymatic Activity of the Staphylococcus aureus SplB Serine Protease is Induced by Substrates Containing the Sequence Trp-Glu-Leu-Gln. Journal of Molecular Biology, 379(2), 343-356.

Team:Leiden/Results