The Design–Build–Test–Learn (DBTL) cycle is a framework used to derive designs that can be integrated into a feedback loop. The recursive nature of the DBTL cycle enables repeated improvements and is a popular approach to building parts in synthetic biology. Each stage in the cycle has a distinct objective:
Our DBTL cycles focussed on (1) finding an appropriate E. coli chassis for OMV production (2) attenuating ClyA for usage in synthetic biology (3) finding epitope tags for recombinant protein detection.
E. coli K-12 lab strains naturally produce OMVs, however the low rate of production makes OMV isolation from E. coli K-12 cultures very resource intensive. Utilizing E. coli strains optimized for OMV isolation reduces the culture volume needed as well as the total time and equipment required for downstream processing.
In order to improve OMV yield, certain gene mutations can be introduced, which increase OMV production. Deletions in the following genes have been shown to increase yield:
In E. coli, the Tol-Pal system comprises five proteins, including three inner membrane proteins (tolA, tolQ, and tolR), one periplasmic protein tolB, and outer membrane protein Pal. The Tol-Pal system tethers the outer membrane to the peptidoglycan layer and the inner membrane. Deletion of tolR produces defects in the Tol-Pal system, causing the dissociation of outer membrane from the underlying peptidoglycan, inducing membrane curvature and blebbing .
tolA is another protein in the Tol-Pal system. Deleting tolA creates a hypervesiculating phenotype in a manner similar to deleting tolR .
degP is a periplasmic protease/ chaperone that processes misfolded/unfolded proteins in the periplasm. Deleting degP enhances OMV production facilitates the removal of misfolded proteins from the cell .
nlp1 is an outer membrane-anchored lipoprotein that has been associated with cell division, but its precise function is unknown. Deleting nlp1 is thought to induce instability in the peptidoglycan-outer membrane cross-links, hence increasing OMV production. 
Four E. coli K12 BW25113 strains, each with a single knockout of the genes (ΔtolR, ΔtolA, ΔdegP and Δnlp1) were obtained from the Keio collection . These knockout strains contain a kanamycin resistance cassette in place of the genes.
Given that our chosen E. coli chassis would experience a metabolic burden during heterologous protein production, we compared growth curves and OD600 at the final time point (final OD600) of the strains to gain a rough estimate of the fitness cost of the gene knockouts. All cultures were standardized to intital OD600 = 0.001 and grown for 18h with orbital shaking in a 96-well plate. The OD600 was measured every 15 min by a microplate spectrophotometer.
Fig 1: Comparative analysis of growth curves of four hypervesiculating E.coli strains.
Fig 2: Individual growth curves of (A) ΔtolA (B) ΔtolR (C) ΔdegP and (D) Δnlp1. Error bars indicate mean ± standard error of the mean of three biological replicates, each having two technical replicates.
Table 1: Final OD600 of four hypervesiculating strains. Values are reported as mean ± standard error of the mean of three biological replicates, each having two technical replicates.
|Strain||Mean final OD600 ± standard error of mean|
|ΔtolA||0.71 ± 0.015|
|Δnlp1||0.70 ± 0.015|
|ΔdegP||0.67 ± 0.012|
Comparing total OMV yield
The concentration of OMVs in a solution can be indirectly quantified using protein concentration as a proxy measure . Protein quantification was performed using a Bradford Assay. A standard curve was constructed using Bovine Serum Albumin (BSA), and the regression line was used to compute protein concentrations of OMV samples (Table 2, Fig 3). .
Table 2: Protein concentration of OMVs fractions isolated from ΔtolR and Δnlp1. Higher total protein was noted for the OMV fraction isolated from Δnlp1. Measurements were carried out in duplicate.
|Sample||Mean adjusted absorbance at 595nm (A)||Protein concentration (ug/ml) (C)||Total volume (ml) (T)||Total protein (P = C*T) (ug)|
Fig 3: Comparative analysis of the OMV yield from ΔtolR and Δnlp1. OMV yield is measured as total protein in the sample.
Comparing the size distribution of OMVs
The size distribution of nanoparticles such as OMVs can be measured via dynamic light scattering. We compared the size distributions of OMV isolated from E. coli ΔtolR and Δnlp1 (Fig 4)
Fig 4: Comparison of the size distribution of OMVs isolated from E. coli ΔtolR and E. coli Δnlp1. The X axis indicates diameter (in nm) and the Y axis indicates the percentage. There is a strong correspondence between the two distributions.
Average diameter of E. coli ΔtolR OMVs = 40 nm
Average diameter of E. coli Δnlp1 OMVs = 39 nm
Among the four strains, ΔdegP showed delayed growth kinetics and lowest final OD600 (Fig 1, 2, Table 1) and was considered unsuitable for heterologous protein expression. Literature suggests that ΔtolR produces more OMVs than ΔtolA . Therefore, further analysis of OMV yield and size distribution was restricted to ΔtolR and Δnlp1. We found no significant difference between the sizes of OMVs derived from both strains (Fig 4), but found that Δnlp1 produced more OMVs than ΔtolR when subjected to the same culture conditions and isolation protocol (Fig 3). From our observations, we chose Δnlp1 to produce OMVs for the remainder of the project.
Cytolysin A (ClyA) is a 34 kDa protein coded by the ClyA gene in E. coli genome. ClyA preferentially localizes to the outer membrane, causing it to be enriched in OMVs. We exploited this property to use ClyA to localize and display recombinant fusion partners on the surface of OMV. The existing version of ClyA on the registry, (BBa_K811000) is hemolytically active and toxic to mammalian cells . It has been used by teams in other composite parts, such as BBa_K3989010, and BBa_K4375019. While genetic fusions with other proteins abolishes the pore-forming ability of ClyA, we wished to create a safer alternative for future teams.
ClyA exists in three known conformations: monomeric, protomeric and oligomeric. The monomer is the soluble form of the protein that undergoes a conformational change to the membrane-inserted protomeric form. The conformational change involves rearrangement of the beta-tongue motif to a helix-turn-helix motif. Upon reaching the target membrane, ClyA protomers oligomerize to insert into the membrane to form a pore. We designed ClyA (Y181F) (BBa_K4359001), a derivative of ClyA with a Y181F substitution mutation. This tyrosine residue is contained in the cholesterol binding site in the beta-tongue motif of ClyA. The substitution is speculated to hinder the transition from the beta-tongue motif to the helix-turn-helix motif accompanying protomer formation. Y181F results in a near loss of the hemolytic activity of ClyA, as determined by erythrocyte lysis assays in previous studies .
We avoided working with the unattenuated version of ClyA. We wished to evaluate the safety of ClyA (Y181F) when used as intended in a composite part like ClyA-3XFLAG-THP (BBa_K4359009). We obtained the sequence for ClyA-3XFLAG-THP via gene synthesis.
The MTT assay is a colorimetric assay for assessing cell metabolic activity, and can be used to measure cell viability. In this case, we aimed to measure cell viability following treatments of different concentrations of THP-OMVs expressing ClyA-3XFLAG-THP, as compared to a control treatment of PBS. SK-BR-3 cells were utilised for this assay. The mean percentage viability is reported as the viability of THP-OMVs treated SK-B-R3 cells normalized to the mean viability following PBS treatment (Fig 5). The complete protocol can be found on the experiments page.
Fig 5: Cytotoxicity of THP-OMVs as measured using the MTT assay with SK-BR-3 cell cultures. Values are reported as mean ± standard error of the mean of three biological replicates.
The results of the MTT assay indicated that THP-OMVs did not significantly affect cell viability. We utilized ClyA (Y181F) for our other construct ClyA-V5-Affi (BBa_K4359008).
Detecting our recombinant proteins was an important part of the project. Epitope tags can be used for this purpose as they are well characterized and antibodies are commercially available. We required two tags whose detection was orthogonal and whose antibodies would cross react minimally with mammalian proteins. We originally received the construct ClyA-Myc-Affi (BBa_K4359006), containing the myc tag. We hypothesized that antibodies against the myc tag would also cross react with endogenous c-myc which is often overexpressed in many cancer cell lines.
ClyA-Myc-Affi was obtained from Dr. Vipul Gujrati, TUM Germany. It was cloned in the backbone pGEX-4T1.
We performed a Western blot of cell lysate of the breast cancer cell line SK-BR-3 (Fig 6). Endogenous c-myc is 62 kDa but is known to degrade quickly to lower molecular weights.
Fig 6: Western blot of SK-BR-3 cell lysate. The membrane was probed with monoclonal anti-myc tag antibody.
While we were unable to ascertain whether the signal detected was from c-myc, we concluded that the anti-myc tag antibody cross-reacted with components of the cell lysate of our test cell line, SK-BR-3. This could potentially interfere with detection during cellular uptake studies. We designed an alternative version with a V5 tag in place of the myc tag, ClyA-V5-Affi (BBa_K4359008)