STAGE ONE: Phasin Secretion System

Iteration 1

Design

Through preliminary research and literature review, we identified phasin as a target protein for the indirect secretion of PHB. We designed initial model for phasin production to consist of a protein encoding gene, PhaP. For our experimental secretion model, we combined the phasin production model with an HlyA tag for type I secretion. We incorporated a His6 tag to provide us the ability to run protein purification in later steps of quantification and measurement. Our constructs are listed below:

This construct is comprised of T7 Promoter/lacO, RBS, Phasin, His6, T7 Terminator, and the plasmid backbone (pSB1A3). It is designed for the E. coli to be able to produce phasin.

This construct is comprised of T7 Promoter/lacO, RBS, HlyA, Phasin, His6, T7 Terminator, and the plasmid backbone (pSB1A3). It is designed for the E. coli to be able to both produce phasin as well as simultaneously undergo secretion.

We considered how we may expand on the work of Calgary 2022’s iGEM team and test various RBSs. In addition to that, we knew we would be testing various secretion tags. Therefore, we ordered each part as its own fragment for Gibson Assembly.

Build

We assembled our two groups via Gibson Assembly, followed by a standard transformation protocol with BL21 (DE3) E. Coli. We incorporated an additional negative control group consisting of untransformed E. Coli. Our team utilized ampicilin and chloramphenicol to confirm successful transformation of our cultures.

Then the groups were separated into two samples. The first groups were induced by IPTG to initiate the production of phasin and allowed to inclubate for an additional 3 hours while the second were remained uninduced.

Test

In order to pellet cells to separate intracellular and extracellular components of the cell culture, samples of the IPTG-induced groups were centrifuged at high speeds. Once isolated, the pelleted E. coli cells were lysed to obtain intracellular components which includes proteins of interest.

After cell cysis, samples of each of the intracellular and extracellular components were spun in nickel column tubes. We made sure to collect and preserve samples in each of the wash, bind, and elute steps.

Learn

We realized that we had misunderstood how to design parts for Gibson assembly on the days following protein purification. We had assumed that Gibson Assembly would be compatible with gene fragments with very low number of base pairs such as our RBS but we learned that it is only optimal for fragments with length 200+, with shorter fragments requiring additional steps which we had not taken. We could not find any literature helpful to our specific requirements, to be able to interchange any part of the construct where certain parts by be as small as 15 bp, and since designed primers for Gibson Assembly surpassed such parts in size, we concluded that Gibson Assembly would not work for our situation.

Additionally, after conducting our first lab that we planned ourselves, we realized we were greatly underprepared and this resulted in unorganized, underdocumented outcomes.

We identified the following factors as points of failure:

• Lab members swapping midway through the procedure
• Lack of documentation and labeling expectations
• Underestimating resources and time required for the lab

Iteration 2

Design

After our first iteration for designing a phasin production and secretion system, we revised our parts to be compatible with Golden Gate Assembly. We reasoned that Golden Gate Assembly would be compatible with the smaller parts and modularity requirements of our plan.

Furthermore, we incorporated an RFP flipper design into our original pSB1A3 plasmid backbone. The use of an RFP flipper would allow us to indicate successful versus unsuccessful colonies in assembly.

In addition, we removed the IPTG induction component of our design. We found it to be redundant in the context of maximizing the production of our target phasin protein. To do this, we simply replaced the T7/LacO promoter with a T7 promoter part without the lac operon. Our revised constructs are visualized below:

Build

Before assembling the parts, we inserted the RFP flipper into our pSB1A3 plasmid backbone to make the pSB1A3 backbone compatible with our assembly through BioBrick Assembly. This alsos allow for the integrated RFP flipper design.

In our altered protocol, we utilized the NEBridge Golden Gate Assembly Kit for Golden Gate Assembly. We followed the same previous procedures for transformation. This time, the cultures were not induced by IPTG.

Learn

This iteration of the design cycle ended with complications in ordering parts which guaranteed our construct was not correctly assembled. After transformation, we tried to confirm our parts were ordered to arrive with the correct overhangs, specifically for the promoter and terminator in order to be compatible with the RFP flipper and the RBS so we have 2 variants (one to ligate directly to the phasin coding fragment and another to the HlyA coding fragment). Unfortunately, we were unable to obtain information on the exact sequence that was synthesized based on our order and the lack of a second RBS variants when we received our order confirmed we had at least one issue with the parts we got. At this point, we decided there wan’t a point in continuing our lab as we cannot determine what exactly we constructed.

We considered designing parts with Type IIS restriction sites to generate the overhangs ourselves and eliminate the possibility of errors with ordering, however with time considerations we had changed our plan to not expand on RBS testing by iGEM team Calgary 2022 which made use rethink the benefits of modularity.

Iteration 3

Design

Due to consistent errors in our modes of assembly, we made the final adjustment to use BioBrick Assembly as our main method for part assembly. Instead of managing a large number of individual parts and respective primers, we also chose to simplify our sequence into single-composite parts. On top of that, we could reuse these parts when our experiments progress to secreting PHB by inserting a PHB synthesizing construct after our design for phasin secretion.

Additionally, in order to maximize efficiency with our experimentations, we selected additional models for our secretion system to test all together. These additions include the type II secretion system via TorA and GeneIII secretion tags. More information and background on these systems can be found in the extended literature review. The constructs for each of these models are the following:

Build

Through a restriction digest and a ligation step, BioBrick Assembly was carried out with the previously listed parts and the pSB1A3 backbone. The assembled plasmid DNA were inserted into E. Coli cells via the same basic transformation protocol.

Test

Again, we performed cell lysis with the pelleted cells samples contraining intracellular components. Each one of our samples were put through a nickel spin-column based purification step. And this time, we conducted our first quantitative measurements of our protein concentration using a Bradford Assay. However, our results with the initial Bradford Assay was inconclusive.

We followed up our Bradford Assay with an SDS-page analysis in order to confirm the success of our purification step. These results, too, were found to be inconclusive.

Learn

Based on our initial Bradford Assay results, absorbance readings indicated practically insignificant protein quantities in all samples with slight differences between experimental groups. We came up with the following as possible sources of error:

• Samples were overly diluted
• Our target proteins were lost in purification
• Wrong buffer was used to make the standard curve

To determine the exact cause of error, we first attempted to run an SDS-page analysis on samples throughout the purification process to observe whether our target protein was lost in stages before the elute. However, many lanes on our page didn’t travel even with prolonged time and the ones that did remained towards the top.

Then, reran protein purification then Bradford Assay on a sample from the same batch with a couple of modifications, specifically the amount of elution buffer was halved to increase protein concentration and elution buffer from protein purification was used to produce the standard curve instead of PBS buffer, this way conditions between the sample and standard are more similar. However, readings once again indicated fairly low concentration of proteins across all groups.

We made another attempt at SDS-page with extra time spent denaturing the proteins as refolding may have been a possible reason proteins failed to travel in the previous attempt. This time, we were able to get some movement after leaving the proteins on the page for multiple additional hours. We were able to compare the proteins in bind, wash, and elute stages of purification between the inside and outside portions of the non-secreting group. Based on the existence of tall bands covering wide ranges of protein weights in all elute solutions, it was indicated that our purification method wasn’t efficient in isolating our his-tagged phasin. Furthermore, analysis of the outside group indicates a thick band assumed to be phasin only flowed through the column for the elute stage but the same band in the inside group seems to flow through in both the bind and wash stages. From this, we concluded that nickel spin columns with his-tagged proteins may not be the optimal method to purify our target protein.

Iteration 4

Design

Based on what we learned from our previous efforts, we implemented appropriate changes to our protocol and design. First, we used samples from a fresh new batch of E. Coli that had been continually incubating since the day of transformation. These tubes had shown more cell growth than the initial round of testing. Therefore, we assumed that these cells were more likely to have produced more proteins than in the previous batch. Furthermore, we reduced dilution of the samples in order to provide us with higher concentration values of proteins to work with.

Instead of testing all the groups, we chose three to work with in order to save resources and time while minimizing error. These three groups were the non secretion group, type II secretion group via HlyA, and a negative control consisting of E. Coli that had been growing in SOC solution.

A major change to our approach was the implementation of an alternative purification method. As previously mentioned, one plausible error with our experimentation was the nickel spin column-based protein purification step. We were not able to confirm its accuracy due to inconclusive SDS-page results. Therefore, we utilized a different protein purification this time using hydrophobic interaction chromatography (HIC).

Test

In testing phasin secretion systems, we conducted the improved versions of our experimental design including HIC chromatography and Bradford Assay protein quantification.

Learn

Although initial readings indicated a significant protein concentration from a high absorbance reading, we realized this may partially be due to the presence of the dye used to show the flow of solution through the HIC column. Unable to distinguish the actual readings for the proteins themselves, HIC chromatography was not compatible with Bradford Assay protein quantification. Additionally, HIC chromatography with the kits we had access to inherently doesn’t allow us to get consistent flow through, so comparing values between experimental groups wasn’t going to be possible. For these limitations, we decided to move away from a purify then quantify approach entirely.

Iteration 5

Design

At this point, we had encountered consistent errors and indecisive results twice, even after several attempts of troubleshooting. To resolve these issues, we came up with a larger redesign of our experimental approach. Instead of using a Bradford Assay and having to isolate our target phasin proteins, we decided on implementing a GFP-reporter tag to each of our constructs.

By adding a GFP reporter tag to our phasin protein, we would not only be able to qualitatively assess the production of our protein but be able to quantify its concentration values through fluorescent readings.

Nevertheless, it is important to identify the specific assumptions we made in implementing such design. These various assumptions would need to be verified in future experiments.

1. GFP proteins will be produced as part of or in equal ratio to the phasin proteins.
2. GFP-tagged phasin proteins are produced similarly to that of non-GFP-tagged phasin proteins. In other words, the addition of a GFP tag has a minimal affect on the production rates and quality of phasin proteins.
3. GFP-tagged phasin proteins are secreted similarly to that of non-GFP-tagged phasin proteins. In other words, the addition of a GFP tag has a minimal affect on the efficiency of phasin protein secretion.

With such assumptions in mind, we redesigned our previous constructs to include the GFP tag as listed below.

Because we were working with a vastly different design, we focused on four candidates that are representative of each of the major secretion systems to reduce clutter instead of experimenting with each and every model. For example, TorA was chosen as a representative group for the type II secretion system. An alternative to the TorA secretion tag would be the GeneIII tag. Similarly, VNp6 is a chosen representative group for VNps. There are many other variations of VNps. including VNp2 and VNp15. These decisions were made based on extensive background research and literature review.

Build

The same restriction digest and ligation procedures were carried out for BioBrick Assembly. The constructed parts included a non secretion system, HlyA system, TorA system, and VNp6 system. The plasmids were then inserted into our BL21 (DE3) E. Coli via basic transformation.

Test

Our groups were seen to successfully fluoresce. By comparing our experimental groups with a negative control group of untransformed E. Coli, we were able to qualitatively observe GFP-tagged phasin production in each of our groups. However, we ran into an unexpected outcome when the non secretion system seemed to be producing a relatively strong glow in the supernatant.

Through centrifigation, we isolated the intracellular and extracellular components of samples of each culture. The pelleted cells were lysed using the same procedure. Then, samples from each intracellular and extracellular groups were collected to perform fluorescence readings on a standard 96-well plate reader. The results allowed us to compare the fluorescence values between each of the groups. A standard fluorescent curve using GFP allowed for the computation of the concentration values of both intracellular and extracellular GFP-tagged phasin proteins.

The results communicated the following (see Results for more):

1. In both extracellular and intracellular samples, the negative control group had minimal fluorescence readings verifying the lack of GFP-tagged phasin production in untransformed E. Coli.
2. In the extracellular (supernatant) samples, the type I HlyA secretion system had the highest fluorescence values followed by the non secretion system, the VNp6 secretion system, and type II TorA secretion system.
3. In the intracellular samples, the type I HlyA system had the highest fluorescence values followed by the non secretion system, the VNp6 secretion system, and the type II TorA secretion system.

Learn

Our readings presented us with fluorescent measurements for the nonsecretion group that was higher than that of TorA and VNp6 in the extracellular media. A possible explanation for these unexpected results may be due to the hindrance of phasin-GFP fusion protein production in the presence of a type II or VNp secretion system in combination with unintended leakage from the intracellular components of the E. coli during our separation procedures. Future investigations should investigate an improved separation technique.