Team:TU Dresden/Project/Secretion

Secretion - Actual Rocket Science

Abstract

In combing Bacillus subtilis powerful secretion capacity with Peptidosomes as a new platform for functional co-cultivation we aim to produce multi protein complexes. Various strains - each secreting distinct proteins of interest - can be cultivated in one reaction hub while being physically separated. In this part of EncaBcillus we study extracelluar protein interaction mediated by the SpyTag/SpyCatcher system. This set-up bears the potential for an effective production of customizable biomaterials or enzyme complexes.

Background

Efficient and low cost production of valuable natural compounds, like proteins, has developed into a leading industry. Starting, by choosing a suitable production host followed by establishing a profitable downstream process, every step is constantly optimized to increase overall yields.

When it comes to choosing a production host, Bacillus subtilis is particularly interesting: the Gram-positive model organism can be easily genetically modified and has powerful secretion capacities.[1]

In this part of EncaBcillus we aimed at making use of the B. subtilis native advantages and combined them with Peptdiosomes – a new innovative cultivation platform for functional co-cultivation. Multiple Peptidosomes, with each encapsulating one specific strain that secrets one protein of interest. By doing so, the production of multi protein complexes cloud be achieved by separated subpopulations all in one reaction hub.

To ensure the assembly of proteins outside of the Peptidosomes we further characterized the SpyTag/SpyCatcher system. Theses functional units can be attached to any protein of interest and upon secretion will result in a covalent isopeptide bond between the SpyTag/SpyCatcher partners.[2] The system originates from Streptococcus pyogenes and since it’s discovery it has been under constant development.[3] In our project we applied codon-adapted B. subtilis specific tags and reduced the SpyCatcher in length to enhance it´s usability when translationally fused to a protein of interest. Thus, decreasing the chances of the tag interfering with overall protein folding.[4]

To demonstrate the applicability of both tags we fused them to a green (sfGFP) and a red (mCherry) fluorescent protein, enabling an easy detectable output. (For more details please check our Design section) Since a core part of this project involves secretion, we included a signal peptide in front of all our constructs. (click here to learn more about our Signal Peptide Toolbox).

To evaluate the efficiency of the secretion process we monitored the fluorescence in supernatants harvested from B. subtilis strains carrying our constructs and compared them to supernatants obtained from the wild type. In order to prove the functionality of the SpyTag/SpyCatcher system we co-incubated supernatants derived from different strains and performed a SDS-Page where we could demonstrate the formation of a fusion protein.

Design

All of the composite parts necessary for the genetic constructs were equipped with the RFC 25 standard, cloned into pSB1C3 backbone and submitted to the registry. All cloning was done according to standard protocols and the plasmids were stored in Escherichia coli DH10β. All constructs were verified by sequencing.

The gene encoding the mini. SpyCatcher (BBa_K2273015) was chemically synthesized. The codon optimized SpyTag (BBa_K2273014) was generated via overlapping primers iG17P049 and G17P050 and amplified using the primers TM4487 and iG17P039. We used a sfGFP (BBa_K2273033) that was codon optimized for Streptococcus pneumoniae, which has been demonstrated to work best in Bacillus subtilis.[5] The used mCherry (BBa_K2273034) was codon adapted for B. subtilis (Popp et al., 2017, accepted). The His-tag, necessary for protein purification was included in the reverse primers (Table 1).

Table 2: Overview of the constructed basic parts.

Gene	BioBrick	forward Primer	reverse Primer
SpyTag_His-tag	BBa_K2273016	TM4487	iG17P068
mini. SpyCatcher_His-Tag	BBa_K2273017	TM4487	iG17P067
sfGFP_His-tag	BBa_K2273021	TM4487	iG17P065
mCherry_His-tag	BBa_K2273022	TM4487	iG17P066

In order to identify the best combination of SpyTag/SpyCatcher and FP fusion, we constructed all N- and C-terminal combinations (Table 2).

Table 2: List of SpyTag/SpyCatcher and FP composite parts.

Gene	BioBrick
pSB1C3-mCherry-SpyTag-His	BBa_K2273035
pSB1C3-SpyTag-mCherry-His	BBa_K2273036
pSB1C3-SpyTag-mCherry-His	BBa_K2273037
pSB1C3-SpyCatcher-mCherry-His	BBa_K2273038
pSB1C3-sfGFP-SpyTag-His	BBa_K2273039
pSB1C3-sfGFP-SpyCatcher-His	BBa_K2273040
pSB1C3-SpyTag-sfGFP-His	BBa_K2273041
pSB1C3-SpyCatcher-sfGFP-His	BBa_K2273042

---

For the final construct we added the signal peptide sequence of amyE (BBa_K2273023) upstream of all constructs (Table 2) and cloned them into the single copy integrative B. subtilis vector pBS2EP_xylA. In brief, the vector has the following features for cloning in E.coli: an ori of replication and the bla gene mediating resistance against ampicillin. The B. subtilis specific part of the vector contains the multiple cloning site (MCS) in RFC10 standard with a P_xylA promoter upstream of the BioBrick prefix, a erm cassette providing resistance against erythromycin /lincomycin and flanking regions needed for integration into the lacA locus.

An overview of the final constructs is depicted in Figure 2 and Figure 3.

The sequenced plasmids were transformed into B. subtilis strain WB800N, a protease deficient strain. Fluorescence measurements of the supernatants derived from strains harbouring our constructs (Table 2) in the B. subtilis vector, was performed according to the protocol below.

Figure 2: Genetic constructs with sfGFP. Depicted are translational fusion constructs downstream of the PxylA promotor, that were cloned in the multiple cloning site of the pBS2EP_xylA vector. The constructs contain a signal peptide sequence, the gene coding for sfGFP, either c- or n-terminally fused SpyTag or mini. SpyCatcher and a hi-tag.

Figure 3: Genetic constructs with mCherry. Depicted are translational fusion constructs downstream of the PxylA promotor, that were cloned in the multiple cloning site of the pBS2EP_xylA vector. The constructs contain a signal peptide sequence, the gene coding for mCherry, either c- or n-terminally fused SpyTag or mini. SpyCatcher and a hi-tag.

SpyTag/SpyCatcher and FP (mCherry with C-terminal mini. SpyCatcher and C-terminal SpyTag, sfGFP with N-terminal SpyTag) were subcloned into the medium copy (15-20 copies per cell) B. subtilis vector pBS0E (Popp et al., 2017, accepted). Via Q5-PCR using the primers TM4067 and TM3082 and the corresponding pBS2E-plasmids as template. The PCR products were digested with BsaI und PstI and ligated with the EcoRI and PstI digested pBS0E vector backbone. The sequenced plasmids were transformed into B. subtilis strain WB800N. Fluorescence assay was performed as described above. Protein purification from the supernatants to check the functionality of the SpyTag/SpyCatcher and FP fusions was performed according to the protocol.

---

Fluorescence assay:

• Day cultures containing antibiotics in 2xYT medium were inoculated with the strains from an agarplate and grown for 8 h at 37°C
• 10 ml overnight culture with 1 % xylose in 2xYT medium were inoculated 1:50 from the day culture and grown for 16 h at 37°C
• To harvest the supernatant, 2 ml of the culture were centrifuged for 10 min at 3220 g
• The cell pellet was resuspended in fresh medium and tested as an control together with the supernatant and the cell culture
• The assay was implemented in a plate reader using a 96 well plate testing 100 μl of each sample
• The endpoint fluorescence of samples with sfGFP was measured at 510 nm with an excitation at 480 nm
• The excitation for mCherry was at 585 nm and the emission was measured at 615 nm

For all the assays biological duplicates and technical triplicates were used.

Purification and SDS-PAGE:

To prove the functionality of the SpyTag and the mini. SpyCatcher as a part of the fusion-proteins, the supernatants were purified with a quick protocol using agarose beads.

• Samples were mixed with a loading buffer containing a reducing agent and heated for 5 min at 95°C
• Then 10 μl were loaded onto a 12,5% SDS gel
• The gel was run at 200 V for 45 min. and stained in Coomassie Blue over night

Results

Fluorescence assay of supernatants derived from strains with single copy genes

The first obstacle that we had to overcome was establishing a suitable protocol to boost the secretion capacities of our generated Bacillus subitilis producer strains. After initially testing various media and incubation periods we performed all experiments using 2xYT medium and an incubation time of 16 h.

Applying this protocol, we observed good fluorescence in the supernatants harvested from the eight strains (combinations of fluorescence proteins (FP) with SpyTag/SpyCather (Tag) constructs) secreting either mCherry or sfGFP. We were able to detect higher fluorescence when compared to supernatants derived from the wild-type (Figure 4).

The supernatant of the wild-type shows a relatively high fluorescence in case of sfGFP , because of the autofluorescence derived from the medium (data not shown). When compared with each other, all of the different fusion possibilities of a FP and the Tags show a similar fluorescence, with the exception of the N-terminally fused mini. SpyCatcher constructs. Apparently, this fusion either inhibits the native folding of the FP or decreases the secretion efficiency.

Figure 4: Endpoint measurement of the fluorescence from supernatants carrying our constructs and the wild type. Expression of the single copy mCherry or sfGFP fusion SpyTag/SpyCather constructs (purple) was induced with 1% xylose and the supernatants were harvested after 16 h of incubation. Wild type supernatant is shown as a control (pink). Excitation wavelength for sfGFP was set to 480 nm and emission was recorded at 510 nm and for mCherry excitation wavelength was set to 585 nm and emission was recorded at 615 nm . The fluorescence was normalized by the optical density (OD600). Graph shows mean values and standard deviations of at least two biological and three technical replicates.

---

Figure 2: Endpoint measurement of the fluorescence obtained under different conditions. B. subtilis strain WB800N with a single copy of mCherry- mini. SpyCatcher is compared to the wild type after 16 h of incubation and initial induction with 1% xylose. Pink bars show the harvested supernatant, purple bars refer to pelleted cells resuspended in fresh medium and blue bars to non-processed cell solution. Excitation wavelength was set to 585 nm and emission was recorded at 615 nm. Graph shows mean values and standard deviations of at least two biological and three technical replicates.

To check the efficiency of the secretion process, we also determined the fluorescents of the cells. The supernatant was separated from the cells by centrifugation.

The pelleted cells were resuspended in fresh medium, immediately followed by a fluorescence measurement (assuming the recorded fluorescence exclusively derives from FPs still inside the cells). We compared the relative fluorescence of both samples and also included untreated 16h old cell culture in this assay.

Figure 2 shows data generated for one strain (mCherry-mini. SpyCatcher) as an example. While the relative fluorescence in the unprocessed cell culture and the supernatant do not differ significantly, the emission of the fresh-resuspended cells show a clear drop of fluorescence (comparable to the wild-type level).

This clearly demonstrates that indeed the measured fluorescence intensities derive from secreted FPs.

---

Fluorescence assay of supernatants derived from strains with multi copy genes:

To further increase the secretion capacity, three constructs were sub-cloned into the pBS0E vector which is a medium copy number B. subtilis specific vector. When performing the same fluorescence assay as bevor, we obtained eight to ten times higher fluorescence intensities in the supernatants when compared to the supernatants derived from single copy (Figure 3).

Figure 3: Endpoint measurement of the fluorescence from supernatants. Expression of the multi copy mCherry or sfGFP fusion constructs (purple) was induced with 1% Xylose and the supernatants were harvested after 16 h of incubation. Wild type supernatant is shown as a control (pink). Excitation wavelength for sfGFP was set to 480 nm and emission was recorded at 510 nm and for mCherry excitation wavelength was set to 585 nm and emission was recorded at 615 nm . The fluorescence was normalized over the optical density (OD600). Graphs show mean values and standard deviations of at least two biological and three technical replicates.

The sfGFP fusion proteins were visible in the supernatant when excited with blue light (Figure 4) and the typical red colour of mCherry could even be observed under daylight conditions (Figure 5). These protein rich supernatants were used for stability tests, spectral analysis, purification and SDS-PAGE.

Figure 4: Supernatants of B. subtilis cultures excited with blue light. Wild-type supernatant (left) and a SpyTag-sfGFP secreting strain (middle and right). The expression of the multi-copy sfGFP was induced with 1% Xylose and the supernatant was harvested after 16 h of incubation. The fluorescence was induced with a “Dark Reader Transilluminator”.

Figure 5: Supernatants of B. subtilis cultures. Wild-type supernatant (left) and a mCherry-mini. SpyCatcher secreting strain (right). The expression of the multi-copy mCherry was induced with 1% Xylose and the supernatant was harvested after 16 h of incubation.

---

Stability:

After incubating the supernatants containing the fluorescent fusion protein for 24h they still showed the same fluorescence intensity, implying that the protein is quite stable in 2xYT medium (graph included in pdf below). Using the B. subtilis strain WB800N as host contributes to an increased stability, as it is not secreting any proteases that may degrade the protein of interest.

Stability of secreted FP

Emission spectra:

The emission spectra of the supernatants containing fusion protein producing cultures were analysed to prove that the measured fluorescence’s originates from sfGFP or mCherry. In Figure 6 the emission spectra of the sfGFP-fusion protein is clearly visible, having an emission peak at ca. 510 nm. The mCherry-fusion protein displays the correct emission spectrum as well, having a peak at 615 nm.

The 2xYT medium that was used for cultivation shows over all high auto-fluorescence when sfGFP wavelengths were measured. As already mentioned bevor, this influences the total fluorescent emissions of the supernatant. Likewise, it needs to be considered that the wildtype supernatant has a lower fluorescence than the medium, suggesting a quenching effect caused by B. subtilis. On the other side, only very low auto-fluorescents or quenching effects could be observed when the mCherry spectrum was recorded. Thus, we decided to carry on with mCherry constructs to prove the applicability of the SpyTag/SpyCatcher systems.

Figure 6: Emission spectra of supernatants and medium. Expression of the multi copy mCherry or sfGFP fusion constructs was induced with 1% xylose and the supernatants were harvested after 16 h of incubation. The sfGFP is N-terminal fused with Spytag, the mCherry is C-terminal fused with mini. SpyCatcher. The supernatants containing fluorescent proteins (pink) were diluted 1:4. Spectra of wilde-type supernatant (blue) and medium (purple) are shown as a control. Excitation wavelength for sfGFP was set to 585 nm and for mCherry to 615 nm. The endpoint fluorescence was measured for each wavelength in steps of one nm ranging from 490 nm to 600 nm for sfGFP and from 580 nm to 700 nm for mCherry.

---

Purification and SDS-PAGE:

By running SDS-PAGE with the purified supernatants of mCherry fusion protein producing cultures, we could prove the functionality of the SpyTag/SpyCatcher partners. In Figure 7 it is clearly visible, that the supernatants from the production strains (lane 4 and 6) contain protein bands of high concentration that the wild-type (lane 2) lacks. These proteins band are showing up again in the elution after purification, proving that the bands are referring to the his-tagged target proteins, mCherry-mini. SpyCatcher having a weight of 36,6 kDa and mCherry-SpyTag weighting 31,9 kDa.

Surprisingly the supernatant with mCherry-SpyTag contained a band of the wrong weight, and the mCherry-mini. SpyCatcher even three bands the wrong weight, indicating accumulation or degradation of the protein of interest.

None the less upon combining the two supernatants and incubating for 4 h, a new protein band with ca. 70 kDA shows up (lane 8 and 9). This is the definite proof, that the mini. SpyCatcher and the Spytag are functional and mediated the covalent bonding of the mCherry constructs.

Figure 7: SDS gel with crude and purified supernatants. Expression of the multi copy mCherry constructs was induced with 1% Xylose and the supernatants were harvested after 16 h of incubation. The his-tagged proteins were purified with Ni-NTA agarose beads. Lane 1 was loaded with 3 l of NEB´s “Color Prestained Protein Standard Broad Range” ladder. Crude (c) and purified (p) supernatant of wild-type (WT) are shown as a control in lane 2 and 3. Lane 4 and 5 contain the supernatant of B. subtilis producing mCherry-mini. SpyCatcher fusion protein (36,6 kDa). Lane 4 and 5 contain the supernatant of B. subtilis producing mCherry-SpyTag fusion protein (31,9 kDa). The crude supernatants of the two mCherry producing strains were combined, incubated for 4 h, purified and loaded onto lane 8 and 9. The fusion product of the mCherry constructs is visable in the crude and purified supernatant.

---

Conclusion

Our team succeeded in the production of self conjugating protein complexes using B. subtilis secretion capacity’s. Even though the system could not yet be tested in Peptidosomes, we are very much sure that our vision to facilitate the production process of tuneable protein complexes can be achieved. Peptiodosomes as a novel platform for cultivation and B. subtilis powerful secretion are a promising combination that should be studied further.

References

[1]	Nijland, Reindert & Kuipers, Oscar. (2008). Optimization of Protein Secretion by Bacillus subtilis. Recent patents on biotechnology
[2]	Gilbert et. all (2017) Extracellular Self-Assembly of Functional and Tunable Protein Conjugates from Bacillus subtilis. ACS Synth. Biol.
[3]	Zakeri et. All (2012) Peptide tag forming a rapid covalent bond to a protein, through engineering a bacterial adhesin. Applied Microbiology and Biotechnology.
[4]	Li et. All (2013) Structural Analysis and Optimization of the Covalent Association between SpyCatcher and a Peptide Tag . J. Mol. Biol.
[5]	Overkamp, W. et al. Benchmarking various green fluorescent protein variants in Bacillus subtilis, Streptococcus pneumoniae, and Lactococcus lactis for live cell imaging. Appl. Environ. Microbiol. 79, 6481–6490 (2013).