Design

Since we were constructing plasmids containing 36 genes, we ruled out the possibility of synthesizing all of the genes due to time constraints, instead choosing to amplify them directly from the genome. This possessed a big challenge as the genomes of C. crescentus have a high GC% content, about 67%, providing a challenge for efficient primer design. Nevertheless, we selected to assemble 2 separate operons under controllable T7/lac promoter for efficient gene expression - (1) hfsE-hfsJ-hfsG and (2) hfsH-hfsK-hfsL for C. crescentus and (1H) hfsE-hfsG-hfsH and (2H) hfsJ-hfsL-hfsK for H. baltica.

This order of genes would allow us to maximize the holdfast production with proteins initiating the tetrad assembly - hfsE - having, in theory, the most substantial concentration and, simultaneously, “switching on” the holdfast assembly system. For this reason, we also have chosen a high copy plasmid.

Cloning six genes into a single plasmid required a high-throughput cloning method, leading to the selection of Golden Gate assembly (GG). For the purpose of minimizing error rate of produced fragments during GG assembly, we came to a conclusion that we will need to first insert one operon, e.g., (1) hfsE-hfsJ-hfsG, sequence the acquired plasmid, and insert the second operon, e.g., (2) hfsH-hfsK-hfsL , for full plasmid assembly.

Build

For cloning of the tetrad assembly operon we used - pET-Duet-1 (Novagen). Vectors and genes were amplified with primers designed for GG assembly (see Materials). Due to the high GC content of the C. crescentus genome, we used Phusion Plus DNA polymerase which is often used for GC-rich template amplification. But the designed primers produced multiple non-specific products after PCR amplification from the genome, leading to the need of gel-purification of acquired fragments (Fig. 4).

After amplifications and purifications, vectors and fragments composing one of the operons of our choice, were mixed in equimolar amounts with GG reaction components and incubated as described in protocol. The reaction was later transformed into E. coli Mach1 (Thermo Scientific) competent cells.

Test

Unfortunately, after multiple unsuccessful plasmid assemblies from genomic fragments, we were unable to obtain any E. colicolonies.

Learn

• Assembling plasmids with multiple fragments is challenging
• Increasing the purity of PCR products might increase the efficiency of Golden Gate assembly

Design

We reasoned that unsuccessful cloning of operons was due to the fact that most of the fragments were of low purity. We had to use agarose gel purification because of the non-specific PCR amplification products. This method usually produces low quality products contaminated with extraction reagents. This in turn distorts the “real” concentration of purified products leading to Golden Gate reaction mixed in unequal molar quantities and subsequent misassembly of plasmids. For this we decided to increase the likelihood of the correct fragment by doing the PCR reaction from already purified PCR products. Since the already available PCR products had the same sequences as the ones we amplified from the genome, there was no need to redesign the primers.

Build

We redid the assembly with PCR products amplified from PCR fragments. Although we expected less non-specific products, that was not the case and we had to use gel extraction yet again.

Nevertheless, the acquired PCR fragments’ purity was higher, with some fragments having ideal 260/230 ratios. After this we repeated the Golden Gate assembly and transformation into E.coli Mach1 competent cells.

Test

Despite several attempts at assembling plasmids from PCR fragments, we were ultimately unable to obtain any E. coli colonies.

Learn

• PCR fragments’ purity was not the solution to unsuccessful Golden Gate assembly
• Changing the backbone to another one might solve the problem

Design

Since we could not acquire any colonies and parallel assembly of operon for holdfast polysaccharide polymerization and export (described in cycle 2) showed that assembling from PCR amplified PCR fragments worked, we decided to change backbone of the plasmid from pET-Duet-1 (Novagen) to pRSF-Duet-1 (Novagen).

Build

We redesigned some primers used for introducing IIS RE sites into the backbone as some of the sequences were identical to the primers used for assembly of operon for the other plasmids. All the subsequent steps were repeated the same way as in the 2nd iteration.

Test

We were able to acquire colonies with one of holdfast polysaccharide tetrad assembly operons with pRSF-Duet-1 backbone. Colonies were screened with colony PCR and positive clones were further purified and checked with analytic restriction digest. Positive plasmids were sequenced by SeqVision. We were able to acquire plasmids containing (1) operon - hfsE-hfsJ-hfsG for C. crescentus and plasmids containing (1H) operon - hfsE-hfsG-hfsH for H. baltica.

Learn

• Changing the backbone was the solution to the unsuccessful cloning
• We will be using the acquired plasmid with (1) operon for subsequent cloning of (2) operon
• Using purer PCR fragments helps the Golden Gate assembly efficiency

Design

Since our strategy worked for (1) operon assembly, we used it for (2) operon insertion into the acquired plasmid with the (1) operon.

Build

Yet again we had to redesign only certain primers to introduce IIS RE sites into the backbone, as some sequences matched those used for assembling other plasmids. All subsequent steps were then repeated in the same manner as in the 2nd iteration.

Test

After multiple attempts, we successfully obtained colonies containing the both holdfast polysaccharide tetrad assembly operons within the pRSF-Duet-1 backbone. Colony PCR was used to screen these colonies, and positive clones were further purified and verified through analytical restriction digestion. The confirmed plasmids were sequenced via whole plasmid sequencing by SeqVision. In the end, we successfully acquired plasmids containing the (1) operon - hfsE-hfsJ-hfsG and the (2) operon - hfsH-hfsK-hfsL for C. crescentus and plasmids containing (1H) hfsE-hfsG-hfsH and (2H) hfsJ-hfsL-hfsK for H. baltica. - half the system for the holdfast synthesis pathway.

• Golden Gate assembly method works for multiple fragment cloning into the same plasmid
• Purity of fragments used for the assembly influences the reaction efficiency. This purity issue can be solved by amplifying PCR fragments from products of previous PCR reactions.
• Different backbones influence the efficiency of assembly and testing several backbones might solve the issue
• Assembling 6 genes into one plasmid is not easy but is possible

Design

Similarly to 1 cycle, we decided to assemble 2 separate operons under T7/lac promoter - (3) hfsA-hfsB-hfsD and (4) hfsF-hfsC-hfsI. All of the proteins composing this system are responsible for polysaccharide polymerization and export. Proteins of the system are found in the membrane thus we came to a conclusion that using a low copy plasmid would decrease the probability of inclusion body formation [3]. Their formation would diminish the functionality of our system, as the proteins would not allow the polysaccharide to be exported outside the bacteria.

Build

For cloning of the holdfast polysaccharide polymerization and export operons we used pACYC-Duet-1 (Novagen). For this assembly we also used the Golden Gate method. However, C. crescentus hfsA gene, due to the presence of one IIS RE cut site in the middle of the gene, had to be domesticated for Golden Gate assembly by site-directed mutagenesis.

The assembly was trickier with H. baltica, as the export apparatus gene - hfsB and polymerization gene - hfsF had multiple IIS RE sites, for this reason, we decided to order synthesized genes (IDT and Twist Biosciences) with mutated sites and simultaneously optimize the codons for expression in E. coli. All the other operons containing genes were amplified the same way as the others.

As in cycle 1, PCR amplification from the genome produced numerous nonspecific PCR products and were gel-purified. After this, vectors and operon fragments were mixed in equimolar amounts with Golden Gate reaction components and incubated as described in the protocol. The reaction was subsequently transformed into E. coli Mach1 competent cells.

Test

Regrettably, despite several attempts at plasmid assembly from genomic fragments, no E. coli colonies were obtained.

Learn

• Low purity of PCR products used for Golden Gate reaction decreases the efficiency of reaction
• Amplifying genes from PCR products might solve the issue

Design

We decided to apply the same strategy as the one described in cycle 1 iteration 2.

Build

After amplifying and preparing the PCR fragments by our described strategy, they were mixed in equimolar ratios into a Golden Gate assembly mixture and afterwards transformed into E. coli Mach1 competent cells.

Test

We managed to obtain E. coli colonies after the Golden Gate assembly mix transformation. The colonies were then tested with colony PCR. Proceeding the colony PCR, the positive clones were purified, analyzed with restriction digest and sequenced by whole plasmid sequencing by SeqVision. We successfully acquired plasmids containing (3) operon hfsA-hfsB-hfsD.

Learn

• Increasing purity of the fragments used for Golden Gate assembly does increase the efficiency of the reaction

Design

The same strategy was used to introduce the (4) operon into the (3) operon containing plasmids.

Build

All the following steps were performed the same way as described in previous cycles.

Test

After numerous tries, we managed to obtain colonies containing the entire holdfast polysaccharide polymerization and export apparatus in the pACYC-Duet-1 backbone. The colonies were screened and sequenced in the same way as in the previous iteration. After all, we successfully isolated plasmids containing the operon (3) - hfsA-hfsB-hfsD and the (4) operon - hfsF-hfsC-hfsI (4)- the other half of the holdfast synthesis pathway.

Learn

• Golden Gate assembly method is a useful tool for multi-fragment cloning
• Purity of fragments used for the assembly does influence the reaction efficiency

Design

After obtaining both plasmids with full holdfast synthesis pathway, we had to optimize the expression of the whole system in E. coli. Previously, only three studies have tried to recombinantly express C. crescentus proteins in E. coli for unassociated research with our project's goal [4] [5] [6]. Since the E. coli strains and protein expression conditions were unrelated to each other and before our project, no one besides the 2009 iGEM ULB-Brussels team ever tried expressing more than 2 C. crescentus proteins in E. coli at the same time, we had no solid foundation for expression and chose to experiment with different E. coli strains and conditions. 

Following consultation with our PI Rolandas Meskys, we decided to start with the E. coli KRX(DE3) strain. This strain offers high transformation efficiency and supports the production of target proteins in large quantities within a single host. Since, our system is placed under T7/lac promoter it is also suitable for our application, as this strain contains T7 RNA polymerase used for this promoter.

Build

To test which conditions are the best for the whole system expression we first tried expressing separate plasmids with different IPTG concentrations - 0.1, 0.25, 0.5, 0.75 and 1 mM - and 0.1% rhamnose, as is used for the T7 RNA polymerase “activation” in KRX strain [7]. With full system, as we were not sure if the amount of metabolic burden introduced by our system would not promote the E. coli to silence or misfold the proteins and based on the fact that previous studies have used different expression temperatures for single proteins, we also decided to experiment with 3 different temperatures - 22°C, 30°C, 37°C - before and after gene expression induction.

Test

In order to assess if the expression was successful, we ran SDS-PAGE samples before and after induction for all the expressions we did. First, we tested pACYC-(3)hfsA-hfsB-hfsD-(4)hfsF-hfsC-hfsI strain with protein expression for 3h at 37°C. As we saw, that some bands, corresponding to our proteins were appearing but we were not sure if they were our system proteins, therefore for pRSF-(1)HfsE-hsfJ-hfsG-(2)hfsH-hfsK-hfsL we used negative control with empty pRSF vector and expressed the proteins in the same conditions.

We saw that pRSF-(1)HfsE-hsfJ-hfsG-(2)hfsH-hfsK-hfsL operon proteins were also expressed with minimal IPTG concentration impact on protein amount. Thus, we decided that we will use lower IPTG concentrations for gene expression induction of the full system, as it is more cost-effective for upscale in the future. Unfortunately, full system expression at different temperatures and expression times did not provide clear bands of proteins in SDS-PAGE gel analysis, so we had to redesign our approach.

Learn

• E. coli KRX strain is not suitable for holdfast synthesis pathway expression. It is possible that this is due to the fact that the expression system uses an additional suppression process with rhamnose which subsequently increases metabolic burden on the whole cell. Although, it is possible to express separate parts of the system.
• Expression of the holdfast synthesis pathway proteins seems to not be highly influenced by IPTG concentration, as the increase of it does not greatly impact the amount of proteins expressed.
• We need to try a different E. coli strain.

Design

Since our first try with a more specialized E. coli strain did not work, we decided to get back to the basics and use one of the most widely used strains for recombinant protein expression - BL21(DE3). One of the previously mentioned studies used this strain for hfsH expression, so we knew that this strain could express at least one of the 12 genes in our system [4]. Besides, this strain does not require rhamnose for polymerase “activation” which, in theory, could decrease metabolic burden which we saw with KRX strain.

Build

Like with the KRX strain, we decide to test separate plasmids first before testing the whole system. Since we saw that it was difficult to distinguish base E. coli proteins from our system proteins, we used negative control with empty plasmids. Moreover, because the IPTG concentration appeared not to make that big of an impact on the expression, we settled on IPTG concentrations of - 0.25, 0.5, 0.75 mM - in this way covering a wide range of them and accelerating the optimisation effort . We also decided to yet again test different expression temperatures - 37°C, 30°C, 16°C - before and after gene expression induction.

Test

As in the 1 iteration we used SDS-PAGE analysis for samples before and after induction. Initially, we tested pACYC-(3)hfsA-hfsB-hfsD-(4)hfsF-hfsC-hfsI operon expression at 37°C for 3h, which did not give promising results, as we could not see distinguishable differences before and after induction. Nevertheless, we proceeded with pRSF-(1)HfsE-hsfJ-hfsG-(2)hfsH-hfsK-hfsL operon expression at the same conditions, which appeared to be working, as we could see stark differences between empty E. coli and our target system lysates. As with the expression in the KRX strain we could not see many differences between the IPTG concentrations.

So with high hopes we advanced with the whole system expression. After analyzing system expression at 37°C for 3h conditions, we saw that there were pronounced differences between the uninduced system and the system after 3 hours (Fig. 7). As with KRX expression, we saw that IPTG concentration used for gene expression induction did not make a big impact for overall expression.

Since the CB2 system was expressing, we went on to try different temperatures with the aim of further optimizing protein expression. Remarkably, decreasing the expression temperature to 30°C and expression overnight did not make a significant difference as the proteins were still expressed in similar amounts to that of 37°C. Expression of 16°C overnight produced some of the expected bands but not in the same capacity as expression at higher temperatures.

Sadly, proteomic analysis of samples induced by 0.5 mM IPTG from separate parts and the whole system, revealed that proteins responsible for holdfast polymerization - hfsC and hfsI - were not expressed. Nevertheless, as later experiments and bioinformatics analysis showed, these proteins were probably substituted by paralogous proteins found in E. colias the system without 2 parts was still producing a polysaccharide. Since we obtained these results a week before the freeze day deadline we were unable to reconsider our approach to the optimization in time. Although we reason that we should first test whether the separate proteins - hfsC and hfsI - are expressed and at what conditions before assembling new plasmids with different operon order or additional T7/lac promoter.

Learn

• Expression of the whole CB2 system from C. crescentus for holdfast synthesis pathway is possible in E. coli.
• The CB2 system is expressed at visible levels in the E. coli BL21(DE3) strain with a minimal amount of IPTG required for induction.
• The best temperature and expression time for this system is 37°C for 3 hours.
• The system lacks hfsC and hfsI proteins and requires further optimization for understanding at what conditions the expression of them occurs.

Design

Once the system was successfully expressed in BL21(DE3) strain, we proceeded to optimize the expression further by testing another E. coli strain - C41(DE3). This strain is often used for toxic recombinant protein expression as it is reported to increase the efficiency of their production [8]. Additionally, this strain is often used for membrane protein production in large quantities [9]. Since half of the required proteins are integrated into the membrane, we reasoned that this strain could enhance the overall shed holdfast volume into the media by overexpressing polymerization and export apparatus’ proteins pACYC-(3)hfsA-hfsB-hfsD-(4)hfsF-hfsC-hfsI.

Build

Yet again we decided to test separate system parts and the whole CB2 system with different IPTG concentrations - 0.25, 0.5 and 0.75 mM. Since we saw that the proteins were best expressed at 37°C in KRX(DE3) and BL21(DE3) strains, we settled on testing only this temperature.

Test

SDS-PAGE analysis of cell lysates before and after gene expression induction revealed that proteins were expressed in separate parts of the system and the whole system. Regrettably, the quantity was visibly less than that seen in BL21(DE3) strain indicating that this strain is not suitable for efficient system expression.

Learn

• Different E. coli strains do express the C. crescentus CB2 system but C41(DE3) is not superior to BL21(DE3) strain for this system expression
• Protein expression for 3 hours at 37°C is interchangeably the best condition for expression in different E. coli strains.

Design

Even though to this point we found E. coli strain with, in our perspective, sufficient system expression, we were determined to improve the expression even more. A study was conducted, where the C. crescentus hfsJ gene was expressed in E. coli Rosetta (DE3) pLysS strain, so we investigated it [5].

Since the genome of C. crescentus is encoded by high GC content, consequently, a considerable amount of proteins use rarer codons that are not often utilized by E.Coli. This causes problems in protein expression as E. coli cannot supply enough tRNAs during translation thus lowering the total produced amount of recombinant proteins. The Rosetta(DE3) strain addresses this issue by providing the pRARE plasmid, which supplies rare tRNA codons necessary for protein expression.Since the aforementioned study used this strain and managed to acquire substantial protein quantities, we decided to also test this strain for our system expression.

Since, we decided to use this strain, we had to carefully reconsidering our plasmid design, as the pRARE plasmid, native to Rosetta, contains the same antibiotic resistance - chloramphenicol - and the same origin of replication - p15A - as one of our system’s plasmids, coding polymerisation and export apparatus. This would make the transformation and subsequent expression impossible as the E. coli would simply lose one of the plasmids rendering the system useless (Fig. 8.).

For this we decided to change the aforementioned plasmids’ antibiotic resistance and origin of replication (ori) as this would be less complicated than assembling both operons into new backbone (Fig.8). As a donor of new antibiotic resistance and origin of replication we choose pBAD-PhoCl2f plasmid, which was kindly gifted to us by VU LSC Institute of Biotechnology. This particular plasmid contains the ampicillin resistance gene and ColE1 origin of replication compatible with our pRSF operon and pRARE plasmid.

Build

To save time we decided to utilize Golden Gate assembly for resistance/ori switching. We designed primers with IIS RE introduction which we previously used for operon assembly. As we have learned how to optimize the assembly from our first 2 cycles, we used the same strategy for this plasmid assembly.

To test the expression we used different IPTG concentrations - 0.25, 0.5 and 0.75 mM - and our standard expression temperature of 37°C with checking the total expressed protein amount after 3 hours by SDS-PAGE analysis.

Test

We successfully acquired plasmids with new resistance/ori, which, after testing them with colony PCR, restriction analysis, were sequenced by whole plasmid sequencing by SeqVision. Since after transformation into competent Rosetta cells they grew without any problems on LB agar plates with 3 different antibiotics, we came to a conclusion that the resistance/ori switch worked.

As before, we started testing expressions by first expressing separate plasmids. Unfortunately, we saw very low or almost no protein expression in separate plasmids, leading to the same happening when we tried expressing the whole system. We reason that this might be due to increased metabolic strain on E. coli due to whole additional plasmid introduced into the system during expression. This E. coli strain might be suitable for specific C. crescentus protein expression but, sadly, is not fitting for our needs.

Learn

• E. coli Rosetta (DE3) pLysS strain is not suitable for our needs due to low expression levels
• Even minor additions to overall metabolic load interferes with holdfast system expression
• It is quite easy to change antibiotic resistance and origin of replication in the plasmid backbone
• Optimized Golden Gate assembly reaction is a powerful tool for complex plasmid assembly

• We were able to express almost the whole system with our choice of the host E. coli strain being BL21(DE3).
• Optimal expression conditions were chosen to be 0.5 mM IPTG gene expression induction followed by expression at 37°C for 3 hours.
• Even minor additions to overall metabolic load interferes with holdfast system expression, suggesting that expressing so many proteins at the same time might be an arduous task for E. coli

Design

Our system was designed to allow for the free release of holdfast polysaccharides into the growth medium allowing for easy purification directly from there. Only one research group had done the purification before in the scope of C. crescentus holdfast chemical properties research [13]. We chose their method to purify the target polysaccharides. Afterwards, the purification outcome had to be tested using a dot blot assay suggested by the same research paper. We selected Wheat Germ Agglutinin (WGA) - primary lectin (Thermo Scientific), labeled with Alexa680, that selectively binds to N‐acetyl-D-glucosamine residues which are found in abundance in the holdfast polysaccharide [14]. This test would let us verify the presence of the holdfast in the purification samples as naturally, E. coli does not shed large amounts of any polysaccharides into the medium.

Build

Purification was done following the protocol of holdfast purification from the media of C. crescentus without any changes. A dot blot assay was performed on the purified samples using WGA, and no alterations were made to the method proposed in earlier research.

Test

After purification, the dot blot assay showed no conclusive results. Disappointingly, it seemed like there was no difference between the target and control samples. Detected N-acetyl-D-glucosamine is likely part of natural E. coli metabolism.

Learn

• There is no holdfast present in the media
• Protein expression alone is likely not substantial for efficient holdfast production
• The additional substrate might be necessary to boost holdfast production
• The temperature range used for cell growth might not be suitable for holdfast formation
• We need to determine whether the holdfast is even being synthesized and if yes where in the cell it is produced

Design

After observing unsatisfactory results in our blots, we hypothesized that an additional substrate might be required to initiate polysaccharide production. Since holdfast biosynthesis begins with the addition of glucose to the undecaprenyl phosphate (UndP) lipid carrier, we determined that glucose would be a fitting substrate for holdfast production enrichment. Previously, research studies have utilized glucose concentrations ranging from 0.25% to 2% w/v for various research purposes [15] [16] [17].

Since we observed that even small additions to overall metabolism lowered system expression, to avoid overloading bacterial metabolism, we decided to use 1% w/v glucose concentration.

We also considered temperature as a potential factor in optimizing holdfast production, referencing the growth conditions of Caulobacter crescentus. Typically, C. crescentus is cultured at 30°C, which prompted us to test this temperature [18] [19]. During the expression optimization phase, we observed that 30°C was still adequate for protein production in E. coli. Moreover, by reducing the temperature we expected to slow E. coli cell division, allowing more time for polysaccharide production.

Build

We replicated the protein expression using previously optimized conditions. However, this time, 3 hours post-induction, a sterile glucose solution was introduced to the medium, bringing it to a final concentration of 1% w/v in the media. Simultaneously, the incubation temperature was lowered to 30°C immediately after the glucose addition. On top of that we lowered the shaking speed of the flasks following the methods used in other polymer production in E. coli [20] [21] [22].

Test

After observing the results, we were uncertain about the nature of the formed rings. Consequently, we decided to conduct another dot blot assay to ascertain the location of holdfast production, storage and the nature of the formed rings. We examined various samples to determine where and if the polysaccharides were present (Fig. 10).

There was no significant difference between the control and target samples in any other localizations tested in the dot blot assay. However, the newly formed ring had a substantial concentration of N-acetyl-D-glucosamine, which was absent in the other samples. Since holdfast contains a considerable amount of N-acetyl-D-glucosamine, it is often used to test localization of holdfast in C. crescentus as it is the only holdfast saccharides that has available lectin that binds to it [23] [14] [24].

Learn

• Addition of glucose and overnight incubation at 30°C overnight is required to start the production of the polysaccharides.
• Ring contains a substantial amount of N-acetyl-D-glucosamine - a distinctive feature of holdfast
• There is no significant amount of holdfast in the cells from the media nor the supernatant.y
• Experiments have to be done to determine if the rings contain living cells or only the holdfast polysaccharides.
• If there are living cells present in the ring we need to check where in the cell the polysaccharides are located.
• Holdfast purification has to be repeated on the ring samples.

Design

Based on our last iteration we understood that our polysaccharide forms only in the rings that appear after the expression and incubation overnight, thus, it was necessary to repeat the previous dot blot assay to understand more about the location of the holdfast. Additional sample preparation steps and a bigger variety of samples would give us a better grasp of the polysaccharide placement within the ring material. Additionally, after identifying the location of the holdfast, we could proceed with the purification process once more. Since we did not know what the rings contained we also had to find out if there were any living bacteria present.

Build

We repeated the expression of the C. crescentus CB2 system proteins at optimal conditions and incubated the culture with 1% (w/v) glucose overnight at 30°C. Yet again, our cultures were forming rings in the same way as the previous iteration.

Rings from the flasks were sampled for dot blot assay with altered sample preparation and additional samples: intact ring, lysed cells, soluble lysate fraction, and insoluble lysate fraction. 

In the previous iteration, it was noticed that our polysaccharides are found only in the ring, therefore, we performed holdfast purification from the formed rings. Material was scraped into a known volume of LB media, the same amount was taken from the control flasks because no substantial ring material was present.

We decided to make a few variations of the plates that would let us see if there are any morphological differences between living bacterial cells from the ring, bacterial cells that express proteins of interest, and bacterial cells that not only express proteins but also, in theory, produce holdfast. Following the ring formation, samples were resuspended in LB media and streaked on LB agar plates in 4 variations - 1) appropriate antibiotics, 2) antibiotics + 0.5mM IPTG, 3) antibiotics + 1% glucose, and 4) antibiotics + 0.5mM IPTG + 1% glucose and, incubated overnight at 30°C.

Test

After overnight incubation at 30°C we observed colonies growing on all the plates, rings containing living bacteria that produce holdfast polysaccharides. There were no significant morphological differences noticed between the plates but the formed colonies were very liquidy in appearance.

Dot blot assay from previous iteration was repeated with the addition of more samples from the rings (Fig. 11)

The results from the previous iteration were successfully reproduced. Holdfast wasexclusively detected in the ring samples, with a majority appearing to be located within the lysed ring cells. This suggests that the synthesized polysaccharide remains attached to the bacterial cells rather than being secreted. Noticeably, CB2 system cells from the media contained almost no N-acetyl-D-glucosamine.

The former iteration showed that we could have been purifying polysaccharides from the wrong material, so we proceeded with the holdfast purification from the ring samples and repeated the dot blot assay. No difference between the control and sample was visible. There was a significant difference between the number of cells in the sample versus control purification which could have caused inconclusive results.

Learn

• Rings repeatedly appear after expression and incubation at 30°C overnight
• Ring material contains living bacteria
• We should try using bacteria from the ring for expression to get better holdfast production
• Polysaccharides, containing a substantial amount of N-acetyl-D-glucosamine are located in the ring
• It appears that, holdfast stays attached to the cells
• We cannot successfully purify holdfast and purification results are inconclusive
• Purification could be inconclusive because the initial amount of cells differed between the control and the sample

Design

After not getting satisfying results from the purification, we figured that the problem could lie in the different amount of cells taken for purification at the beginning of it. Standard optical density methods can not measure cells with the holdfast because of their tendency to stick together in clumps. A different method is required to ensure that the purification begins with an equal amount of material between the control and target samples.

Earlier iteration showed that the holdfast polysaccharides are found only in the ring material. This might be a result of only part of the bacterial population producing a full set of holdfast synthesis pathway proteins. Trying to use bacteria found in the ring might let us get better protein expression and overall increase in holdfast production.

Build

To address the optical density measurement issue, we determined that, in the absence of other spectrophotometric methods, a visual estimation of the cell quantity would be adequate for our needs. Holdfast synthesis was repeated and samples were taken for a third biological repetition of dot blot assay for polysaccharide localization detection. We repeated purification with a corrected control sample preparation method.

Cells from the ring were used for a standard holdfast synthesis pathway protein expression and incubation overnight with 1 % glucose. 12% SDS-PAGE analysis was run to verify the results.

Test

Holdfast localization results were, once again, replicable. Dot blot assay for holdfast detection showed that holdfast is still located in the ring material formed after the standard expression.

But, to our disappointment, holdfast purification, even with the changed method, showed no conclusive results.

In addition, before and after induction samples from protein expression using ring bacteria were analyzed using 12% SDS-PAGE. It appears that we achieved better protein expression using bacteria from the ring in comparison with standard expression (Fig. 12). Although, the formed ring after incubation appeared to be the same to freshly expressed cells.

Learn

• Holdfast synthesis is replicable
• Dot blot assays show that holdfast is located in the ring material
• Purification with the same amount of starting material did not give any conclusive results
• The purification protocol used might not be suitable for purification from the cell material
• We need to perform more experiments to more accurately describe produced holdfast
• Cells from the ring are able to give better C. crescentus CB2 protein expression than fresh ones

Design

We observed that holdfast-producing cells often clump together and adhere to one another. This observation hinted at potential morphological differences between holdfast-producing E. coli cells and the control group. To investigate this further, we employed scanning electron microscopy (SEM), flow cell bright field microscopy and Fourier-Transform Infrared Spectroscopy (FTIR)were selected as methods of analysis.

• SEM analysis would provide us with information about the microscopic morphological changes of the bacteria induced by our holdfast synthesis system.
• Flow cell bright field microscopy is a special technique that combines traditional bright field microscopy and optical cells used in photometers and cell counters. In our design, we would measure how sticky our cells are by incubating E. coli in the flow cell, letting it sediment on the glass slide and then washing the chamber with gradually increasing flow of the buffer. Comparing our samples with control we should see the difference and be able to calculate how much hydrostatic force our holdfast can tolerate, by seeing how much cells are being washed off at specific flow rate.
• FTIR spectroscopy would reveal the composition of chemical molecules and help with the identification of unknown materials - just what we need with our holdfast.
• We also decided to test whether non BL21(DE3)E. coli strains that showed moderate expression of CB2 system could form rings after the addition of glucose.

Build

Because we could not perform SEM by ourselves, we found a colleague in the Vilnius University Faculty of Chemistry and Geosciences Dr. Andrius Pakalniškis who kindly agreed to help. We prepared the samples for SEM using different techniques: drying them on silicon wafers in their native form and pre-washing them with PBS and ethanol before drying.

Flow cell bright field microscopy also required help and equipment from other scientists, we are very thankful to Dr. Marijonas Tutkus and his student Monika Roliūtė from Vilnius University Life Sciences Center who helped us with the experiments. After incubation of the flow cell with our bacteria we started with a 100 μL/min flow rate and increased it to 2.5 mL/min. and registered the data in video format for later analysis.

We also did not have the equipment for FTIR spectroscopy in our facility, so we contacted colleagues at the Vilnius University Center of Physical Sciences And Technology. Dr. Martynas Talaikis and Dr. Ilja Ignatjev agreed to collaborate and help us with the analysis. We prepared three types of samples to get the most information: dried intact bacteria, dried lysate debris and dried lysate supernatant.

For E. coli strain testing, we used C41(DE3), Rosetta(DE3) pLysS and HMS174(DE3), and repeated the expression conditions with different IPTG concentrations (0.25, 0.5 and 0.75 mM) with protein expression for 3h at 37°C. After the expression we added 1% glucose and left the flasks to incubate overnight at 30°C 170 rpm.

Test

Scanning electron microscopy revealed that holdfast-producing cells have different morphology. Target cells appear bigger, with distinctly different cell envelope topology, and tend to form tight clumps in contrast with the control samples. After consulting with our colleagues in the VU LSC Department of Microbiology and Biotechnology we came to the conclusion that our cells are forming biofilm-like structures. The aforementioned formed rings are often observed in biofilm-forming bacteria, although not in such vast quantities and less “sticky” [25] [26] [27].

We compared our samples with available SEM pictures of biofilm-forming E. coli. It appears that our holdfast-producing bacteria form biofilm-like structures with tightly stuck-together bacteria with rhizoid-like structures in the places where the biofilm is broken apart (Fig. 13) [28] [29].

Our bright field experimental design was never done before, so we did not know what to expect from our acquired results. Disappointingly, flow cell experimentsdid not give us any anticipated results. Apparently E. coli naturally tends to unspecifically interact with the glass and stick to it very firmly, there was no significant difference between control and target samples, consequently, we could not obtain any data about the force resistance of our polysaccharides. Nevertheless, we obtained another interesting piece of information - our holdfast producing bacteria is floating around in big conglomerates that are most likely pieces of biofilm, that we saw in SEM, when no such phenomenon was observed in the control sample (Fig. 14).

After testing ring formation in different E. coli strains we saw that they were also forming rings after our proposed incubation conditions. Due to time constraints we did not perform dot blot assay but it is likely that the formed rings would also contain holdfast withN-acetyl-D-glucosamine.

Learn

• Holdfast producing E. coli has distinct morphological differences
• Cells tend to be bigger in size with uneven cell wall topology
• Bacteria appear to form biofilm-like structures
• E. coli naturally sticks to glass slides, so it is important to prepare the surface of flow cells before to reduce non-specific interactions
• Biofilm conglomerates can be easily seen in bright field microscopy
• Different E. coli strains still produce rings on flask walls