From 2011.igem.org

Revision as of 03:14, 22 September 2011

SgsE from Geobacillus stearothermophilus NRS 2004/3a

Purification of SgsE fusion protein

As seen in the analysis of the cultivations with expression of SgsE | mCitrine fusion proteins, these proteins form inclusion bodies in E. coli. Inclusion bodies have the advantage that they are relatively easy to clean-up and are resistant to proteases. So the first purification step is to solve and set-free the inclusion bodies. This step is followed by two filtrations (300 kDa UF and 100 kDa DF/UF) to further concentrate and purify the S-layer proteins. After the filtrations, the remaining protein solution is dialized against ddH₂O for 18 h at 4 °C in the dark. The dialysis leads to a precipitation of the water-insoluble proteins. After centrifugation of the dialysate the water-soluble S-layer monomers remain in the supernatant and can be used for recrystallization experiments.

The fluorescence of the collected fractions of this purification strategy is shown in the following figure 1:

Fig. 1: Fluorescence of collected fractions during purification of <partinfo>K525305</partinfo> fusion protein.

A lot of protein is lost during the purification especially after centrifugation steps. The fluorescence in the urea containing fractions is lowered due to denaturation of the fluorescent protein. Some fluorescence could be regenerated by the recrystallization in HBSS. This purification strategy is very simple and can be carried out by nearly everyone in any lab being one first step to enable real do it yourself nanobiotechnology.

Final purification strategy

Scheme of purification strategy for SgsE (fusion) proteins:

First, SgsE is expressed in E. coli under the control of a T7 / lac promoter for separation of growth and production phase due to metabolic stress of the S-layer expression. Because the SgsE protein is forming inclusion bodies in E. coli, an inclusion body purification with urea follows the cell lysis. The S-layers are further concentrated and purified by two ultrafiltration / diafiltration steps (300 kDa and 100 kDa) and afterwards dialysed against water leading to the precipitation of water-insoluble proteins. The supernatant contains the monomeric SgsE solution.

Click for detailed information

Immobilization behaviour

After purification, solutions of monomeric SgsE S-layer proteins can be recrystallized and immobilized on silicon dioxide beads in HBSS (Hank's buffered saline solution). After the recrystallization procedure the beads are washed with and stored in ddH₂O at 4 °C in the dark. The fluorescence of the collected fractions of a recrystallization experiment with <partinfo>K525305</partinfo> are shown in fig. 2. 100 mg beads were coated with 100 µg of protein. The figure shows, that not all of the protein is immobilized on the beads (supernatant fraction) but the immobilization is pretty stable (very low fluorescence in the wash). After the immobilization, the beads show a high fluorescence indicating the binding of the SgsE | mCitrine fusion protein.

Fig. 2: Measured fluorescence of collected fractions of immobilization of purified <partinfo>K525305</partinfo> on silica dioxide beads (n = 3, 100 mg mL^-1 SiO₂, time of recrystallization: 4 h).

Optimal bead to protein ratio for immobilization

To determine the optimal ratio of silica beads to protein for immobilization, the degree of clearance ϕ_C in the supernatant is calculated and plotted against the concentration of silica beads used in the accordant immobilization experiment (compare fig. 3):

The data was collected in three indipendent experiments. The fluorescence of the samples was measured in the supernatant of the immobilization experiment after centrifuging the silica beads. The fluorescence of the control was measured in a sample which was treated exactly like the others but no silica beads were added. 100 µg protein was used for one immobilization experiment. The data was fitted with a sigmoidal dose-response function of the form

with the Hill coefficient p, the bottom asymptote A1, the top asymptote A2 and the switch point log(x₀) (R² = 0.874).

The fit indicates that a good silica concentration for 100 µg of protein is 150 - 200 mg mL^-1. This set-up leads to saturated beads with low waste of protein. So a good protein / bead ratio to work with is 5 - 7 * 10^-4.

Fig. 3: Degree of clearance of the fluorescence in the supernatant plotted against the concentration of silicium dioxide beads used to immobilize <partinfo>K525305</partinfo> (n = 3). Data is fitted with dose-reponse function (R² = 0.874).

SbpA from Lysinbacillus sphaericus CCM 2177

Purification of SbpA fusion protein

As seen in the analysis of the cultivations with expression of SbpA | mCitrine fusion proteins, these proteins form inclusion bodies in E. coli. Inclusion bodies have the advantage that they are relatively easy to clean-up and are resistant to proteases. So the first purification step is to solve and set-free the inclusion bodies. This step is followed by two filtrations (300 kDa UF and 100 kDa DF/UF) to further concentrate and purify the S-layer proteins. After the filtrations, the remaining protein solution is dialized against ddH₂ for 18 h at 4 °C in the dark. The dialysis leads to a precipitation of the water-insoluble proteins. After centrifugation of the dialysate the water-soluble S-layer monomers remain in the supernatant and can be used for recrystallization experiments.

The fluorescence of some collected, important fractions of this purification strategy is shown in the following figure 4:

Fig. 4: Fluorescence of collected fractions during purification of <partinfo>K525405</partinfo> fusion protein. Abbreviations: pell.: pellet, s.n.: supernatant, ret.: retentate, perm.: permeate.

A lot of protein is lost during the purification especially after centrifugation steps (compared to filtrations). The fluorescence in the urea containing fractions is lowered due to denaturation of the fluorescent protein. This purification strategy is very simple and can be carried out by nearly everyone in any lab being one first step to enable real do it yourself nanobiotechnology.

Final purification strategy

Scheme of purification strategy for SbpA (fusion) proteins:

First, SbpA is expressed in E. coli under the control of a T7 / lac promoter for separation of growth and production phase due to metabolic stress of the S-layer expression. Because the SbpA protein is forming inclusion bodies in E. coli, an inclusion body purification with urea follows the cell lysis. The S-layers are further concentrated and purified by two ultrafiltration / diafiltration steps (300 kDa and 100 kDa) and afterwards dialysed against water leading to the precipitation of water-insoluble proteins. The supernatant contains the monomeric SbpA solution.

Click for detailed information

Immobilization behaviour

After purification, solutions of monomeric SbpA S-layer proteins can be recrystallized and immobilized on silicon dioxide beads in recrystallization buffer (0.5 mM Tris-HCl, pH 9, 10 mM CaCl₂). After the recrystallization procedure the beads are washed with and stored in ddH₂O at 4 °C in the dark. The fluorescence of the collected fractions of a recrystallization experiment with <partinfo>K525405</partinfo> are shown in fig. 5. 100 mg beads were coated with 100 µg of protein. The figure shows, that not all of the protein is immobilized on the beads (supernatant fraction) but the immobilization is pretty stable (very low fluorescence in the wash). After the immobilization, the beads show a high fluorescence indicating the binding of the SbpA | mCitrine fusion protein.

Fig. 5: Measured fluorescence of collected fractions of immobilization of purified <partinfo>K525405</partinfo> on silica dioxide beads (n = 3, 100 mg mL^-1 SiO₂, time of recrystallization: 4 h).

Optimal bead to protein ratio for immobilization

To determine the optimal ratio of silica beads to protein for immobilization, the degree of clearance ϕ_C in the supernatant is calculated and plotted against the concentration of silica beads used in the accordant immobilization experiment (compare fig. 6):

The data was collected in three indipendent experiments. The fluorescence of the samples was measured in the supernatant of the immobilization experiment after centrifuging the silica beads. The fluorescence of the control was measured in a sample which was treated exactly like the others but no silica beads were added. 100 µg protein was used for one immobilization experiment. The data was fitted with a sigmoidal dose-response function of the form

with the Hill coefficient p, the bottom asymptote A1, the top asymptote A2 and the switch point log(x₀) (R² = 0.997).

The fit indicates that a good silica concentration for 100 µg of protein is 200 - 250 mg mL^-1. This set-up leads to saturated beads with low waste of protein. So a good protein / bead ratio to work with is 7 - 9 * 10^-4.

Fig. 6: Degree of clearance of the fluorescence in the supernatant plotted against the concentration of silicium dioxide beads used to immobilize <partinfo>K525405</partinfo> (n = 3). Data is fitted with dose-reponse function (R² = 0.997).

CspB from Brevibacterium flavum

CspB with TAT-sequence and lipid anchor

Cultivation and protein expression

For characterization CspB [http://partsregistry.org/Part:BBa_K525121 (K525121)] gen was fused with a monomeric RFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] using Gibson assembly.

The CspB|mRFP fusion protein was overexpressed in E. coli KRX after induction of T7 polymerase by supplementation of 0,1 % L-rhamnose using the autoinduction protocol.

Figure 7: Growth curve of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction, cultivated at 37 °C in autoinduction medium with and without inductor, respectively. A curve depicting KRX wildtype is shown for comparsion. After autoinduction at approximately 4 h the OD₆₀₀ of the induced <partinfo>K525121</partinfo> decreases slightly 4 hours later when compared to the uninduced culture. Both cultures grow significantly slower than KRX wildtype.

Figure 8: RFU to OD₆₀₀ ratio of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction. A curve depicting KRX wildtype is shown for comparsion. After induction at approximately 4 h the RFU to OD₆₀₀ ratio starts to rise in the induced culture. Compared to the uninduced culture the ratio is roughly seven times higher. The KRX wildtype shows no variation in the RFU to OD₆₀₀ ratio.

Identification and localisation

After a cultivation time of 18 h the mRFP|CspB fusion protein has to be localized in E. coli KRX. Therefore a part of the produced biomass was mechanically disrupted and the resulting lysate was washed with ddH₂O. The periplasm was detached by using a osmotic shock from other parts of the cells. The existance of fluorescene in the periplasm fraction, showed in fig. 9, indicates that C. glutamicum TAT-signal sequence is at least in part functional in E. coli KRX.

The S-layer fusion protein could not be found in the polyacrylamide gel after a SDS-PAGE of the lysate and the cell debris were still red. This indicates that the fusion protein intigrates with the lipid anchor into the cell membrane. For testing this assumption the washed lysate was treated with ionic, nonionic and zwitterionic detergents to release the mRFP|CspB out of the membranes.

The existance of flourescence in the detergent fractions and the proportionally to the lysis fraction low fluorescence in the wash fraction confirm the hypothesis of an insertion into the cell membrane (fig. 9). An insertion of these S-layer proteins might stabilize the membrane structure and increase the stability of cells against mechanical and chemical treatment. A stabilization of E. coli expressing S-layer proteins was discribed by Lederer et al., (2010).

An other important fact is, that there is actually mRFP fluorescence measurable in such high concentrated detergent solutions. The S-layer seems to stabilize the biologically active conformation of mRFP. The MALDI-TOF analysis of the relevant size range in the polyacrylamid gel approved the existance of the intact fusion protein in all detergent fractions (fig 10).

Figure 9: Fluorescence progression of the mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)]/CspB fusion protein initiating with the cultivation fractions up to the detergent fractions of the seperate denaturations. Cultivations were carried out in autoinduction medium at 37 ˚C. The cells were mechanically disrupted and the resulting biomass was wahed with ddH₂O and resuspendet in the respective detergent. The used detergent acronyms stand for: SDS = 10 % (v/v) sodium dodecyl sulfate; UTU = 7 M urea and 3 M thiourea; U = 10 M urea; NLS = 10 % (v/v) n-lauroyl sarcosine; CHAPS = 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate.

MALDI-TOF analysis was first used to identify the location of the fusion protein in different fractions. Fractions of medium supernatant after cultivation, periplasmatic isolation, cell lysis, denaturation in 6 M urea and the following wash with 2 % (v/v) Triton X-100, 2 % SDS (w/v) were loaded onto a SDS-PAGE and fragments of the gel were measured with MALDI-TOF.

Figure 10: SDS-PAGE of CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] fusion protein. Lanes are fractions of culture supernatant (M), periplasmatic isolation (PP), cell lysis (L), denaturation (D) and a wash of the pellet of the denaturation with Triton X-100 (T). Used marker is PageRuler ^TM Prestained Protein Ladder SM0671. Marked regions were cut out and prepared for MALDI-TOF analysis.

The following table shows the sequence coverage (in %) of our measurable gel samples with the amino acid sequence of fusion protein CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)].

Fig. 5 shows these data. The gel samples were arranged after estimated molecular mass cut out from the gel.

Figure 5: MALDI TOF measurement of CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] fusion protein. Samples are arranged after estimated molecular mass of the gel slice. Measurement was performed with a ultrafleXtreme^TM by Bruker Daltonics using the software FlexAnalysis, Biotools and SequenceEditor.

As expected, only minor sequence coverage was found in the periplasmatic fraction, due to the lipid anchor located at the carboxy-terminus. This hydrophobic region inhibits the transport of the protein to the periplasm, mediated by the amino-terminal TAT-sequence. Little fluorescence was also found in the lysis fraction, verifying our assumtion, that the protein integrates or strongly binds to the cell membrane. Using urea to disintegrate the S-layer fusion protein from the cell membrane resulted only in a slightly higher sequence coverage. However, washing the pellet with 2 % Triton X-100 (v/v), 2 % SDS (w/v), previously treated with urea, resulted in a higher sequence coverage and can therefore be expected as more applicable to desintegrate the S-layer fusion protein. Sequence coverage in the supernatant of the cultivation medium can be explained with the late phase of cultivation where some cells are lysed.

To obtain more specific informations about the location of the S-layer fusion protein, after comparison with same treated fraction of E. coli KRX all gel bands in a defined size area were cut out of the gel and analysed with MALDI-TOF. Results are shown in fig. 6.

Figure 6: MALDI-TOF measurement of CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] fusion protein in different fractions. Abbreviations are Ma: Marker (PageRuler ^TM Prestained Protein Ladder SM0671), M (medium), PP (periplasm), L (cell lysis with ribolyser), W (wash with ddH₂O). In the left half of the gel fractions of E. coli KRX with induced production of fusion protein, the right half shows fractions of E. coli KRX without carrying the plasmid coding the fusion protein. Colours show the sequence coverage of the gel lane, cutted out of the gel.

Sequence coverage was only found in the wash and the lysis fraction, again indicating that the S-layer protein is integrating in the cell membrane. Thus transport to the periplasm mediated through the amino-terminal TAT-sequence cannot take place or after transport to the periplasm binds to the inner cell membrane.

The influence of other detergents to disintegrate the S-layer fusion protein was tested after disrupting the cells with a ribolyser. The cell pellet was incubated in 10 % (v/v) Sodium dodecyl sulfate (SDS), in 7 M urea and 3 M thiourea (UTU), in 10 M urea (U) in 10 % (v/v) n-lauroyl sarcosine (NLS) and in 2 % CHAPS (C). Samples of the incubations with these detergents were loaded onto a SDS-PAGE prior to measurement with MALDI TOF.

Figure 7: Influence of diffent detergents on the disintegration of the fusionprotein CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] from the cell membrane of E. coli KRX. Abbreviations are: SDS (Sodium dodecyl sulfate 10 % (w/v)), NLS ((10 % (w/v) m-lauroyl sarcosine), UTU (3 M thiourea, 7 M urea), U (10 M urea), CH (2% CHAPS 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate), Ma Marker (PageRuler ^TM Prestained Protein Ladder SM0671). On the left half of the gel fractions of the S-layer fusion protein are displayed, on the right half fractions of E. coli KRX are displayed.

The result of the MALDI-TOF measurement clearly demonstrates that all used detergents are applicable to disintegrate the S-layer fusion proteins from the bacterial cell membrane of E. coli. Fluorescence measurement of fractions, treated with the detergents, show significantly different values, indicating that some of the detergents (e.g. 3 M thiourea, 7 M urea) have a strong effect on protein folding.

CspB without TAT-sequence and with lipid anchor

Cultivation and protein expression

For characterization the modiefied CspB [http://partsregistry.org/Part:BBa_K525123 (K525123)] gen was fused with a monomeric RFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] using Gibson assembly.

The fusion protein was overexpressed in E. coli KRX after induction of T7 polymerase by supplementation of 0,1 % L-rhamnose using the autoinduction protocol.

Figure X: Growthcurve of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction, cultivated at 37 °C in autoinduction medium with, respectively, without inductor. A curve depicting KRX wildtype is shown for comparsion. After induction at approximately 6 h the OD₆₀₀ of the induced K525123 visibly drops when compared to the uninduced culture. While the induced culture grow significantly slower than KRX wildtype the uninduced seems to be unaffected.

Figure X: RFU to OD₆₀₀ ratio of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction. A curve depicting KRX wildtype is shown for comparsion. After induction at approximately 6 h the RFU to OD₆₀₀ ratio starts to rise in the induced culture. Compared to the uninduced culture the ratio is roughly eight times higher. Most likely due to basal transcription the RFU to OD₆₀₀ ratio of the uninduced culture starts to rise after 12 hours. The KRX wildtype shows no variation in the RFU to OD₆₀₀ ratio.

Identification and localisation

After a cultivation time of 18 h the mRFP|CspB fusion protein was localized in E. coli KRX. Therefor a part of the produced biomass was mechanically disrupted and the resulting lysate was wahed with ddH₂O. From the other part the periplasm was detached by using an osmotic shock.

The S-layer fusion protein could not be found in the polyacrylamide gel after a SDS-PAGE of the lysate. This indicated that the fusion protein integrates into the cell membrane with its lipid anchor. For testing this assumption the washed lysate was treated with ionic, nonionic and zwitterionic detergents to release the mRFP|CspB out of the membranes.

The existance of flourescence in the detergent fractions and the not existent fluorescence in the wash fraction confirms the hypothesis of an insertion into the cell membrane (fig. 3). An insertion of these S-layer proteins might stabilize the membrane structure and increase the stability of cells against mechanical and chemical treatment. A stabilization of E. coli expressing S-lyer proteins was discribed by Lederer et al., (2010).

Another important fact is that there is actually mRFP fluorescence measurable in such high concentrated detergent solutions. The S-layer seems to stabilize the biologically active conformation of mRFP. The MALDI-TOF analysis of the relevant size range in the polyacrylamid gel approved the existance of the intact fusion protein in all detergent fractions (fig 4).

In comparison with the mRFP fusion protein of [http://partsregistry.org/Part:BBa_K525121 K525121], which has a TAT-sequence, a minor relative fluorescence in all cultivation and detergent fractions was detected (fig. 3). Together with the decreasing RFU/OD₆₀₀ after 12 h of cultivation (fig. 2) indicates that the TAT-sequence results in a postive effect on the protein stability.

Figure 3: Fluorescence progression of the mRFP[http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)]/CspB fusion protein initiating with the cultivation fractions up to the detergent fractions of the seperate denaturations. Cultivations were carried out in autoinduction medium at 37 ˚C. The cells were mechanically disrupted and the resulting biomass was washed with ddH₂O and resuspended in the respective detergent. The used detergent acronyms stand for: SDS = 10 % sodium dodecyl sulfate; UTU = 7 M urea and 3 M thiourea; U = 10 M urea; NLS = 10 % n-lauroyl sarcosine; 2 % CHAPS = 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate.

MALDI-TOF analysis was used to identify the location of the fusion protein in different fractions. Fractions of medium supernatant after cultivation (M), periplasmatic isolation (PP), cell lysis (L) and the following wash with ddH₂O, samples were loaded onto a SDS-PAGE. After comparison with same treated fraction of E. coli KRX all gel bands in a defined size area were cutted out of the gel and analysed with MALDI-TOF. Results are shown in fig. 4.

Figure 4: MALDI-TOF measurement of the mRFP[http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)]/CspB fusion protein. Data is shown in a SDS-PAGE. Colours show the sequence coverage of MALDI-TOF measurement. On the left side of the gel samples fractions of the fusion protein are shown, on the right side of the gel fractions of an equally treated E. coli KRX are shown. Measurement was performed with a ultrafleXtremeTM by Bruker Daltonics using the software FlexAnalysis, Biotools and SequenceEditor.

Results show that the fusion protein of mRFP[http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)]/CspB without TAT-sequence and with lipid anchor has only been identified in the lysis fraction. However, in conclusion with absent TAT-sequence, the protein has not been identified in the periplasm and the culture supernatant, respectively.

The influence of other detergents to disintegrate the S-layer fusion protein was tested after disrupting the cells with a ribolyser. The cell pellet was incubated in 10 % (v/v) Sodium dodecyl sulfate (SDS), in 7 M urea and 3 M thiourea (UTU), in 10 M urea (U) in 10 % (v/v) N-lauroyl sarcosine (NLS) and in 2 % CHAPS (C). Samples of the incubations with these detergents were loaded onto a SDS-PAGE prior to measurement with MALDI-TOF (Fig. 6).

Figure 6: Influence of diffent detergents on the disintegration of the fusionprotein CspB/mRFP (BBa_E1010) from the cell membrane of E. coli KRX. The coloured marker shows the sequence coverage of MALDI-TOF measurement. Abbreviations are: SDS (Sodium dodecyl sulfate 10 % (w/v)), NLS ((10 % (w/v) m-lauroyl sarcosine), UTU (3 M thiourea, 7 M urea), U (10 M urea), CH (2% CHAPS 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate), Ma Marker (PageRuler TM Prestained Protein Ladder SM0671). On the left half of the gel fractions of the S-layer fusion protein are displayed, on the right half fractions of E. coli KRX are displayed.

The results of the MALDI-TOF measurement clearly demonstrate that all used detergents are applicable to disintegrate the S-layer fusion proteins from the bacterial cell membrane of E. coli. Fluorescence measurement of fractions treated with the detergents, show significantly different values, indicating that some of the detergents (e.g. 3 M thiourea, 7 M urea) have a strong effect on protein folding. The samples taken from gel lanes of E. coli KRX show no sequence coverage, therefore not similar proteins are naturally induced in E. coli.

CspB from Corynebacterium halotolerans

CspB without TAT-sequence and lipid anchor

Cultivation and protein expression

For characterization the modiefied CspB [http://partsregistry.org/Part:BBa_K525222 (K525222)] gen was fused with a monomeric RFP ([http://partsregistry.org/Part:BBa_E1010 BBa_E1010]) using Gibson assembly.

The fusion protein was overexpressed in E. coli KRX after induction of T7 polymerase by supplementation of 0,1 % L-rhamnose using the auinduction protocol.

Figure X: Growthcurve of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction, cultivated at 37 °C in autoinduction medium with, respectively, without inductor. A curve depicting KRX wildtype is shown for comparsion. After induction at approximately 6 h the OD₆₀₀ of the induced K525222 visibly drops when compared to the uninduced culture. While the induced culture grow significantly slower than KRX wildtype the uninduced seems to be unaffected.

Figure X: RFU to OD₆₀₀ ratio of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction. A curve depicting KRX wildtype is shown for comparsion. After induction at approximately 6 h the RFU to OD₆₀₀ ratio starts to rise in the induced culture. Compared to the uninduced culture the ratio is roughly five times higher. Most likely due to basal transcription the RFU to OD₆₀₀ ratio of the uninduced culture starts to rise after 12 hours. The KRX wildtype shows no variation in the RFU to OD₆₀₀ ratio.

Identification and localisation

After a cultivation time of 18 h the mRFP|CspB fusion protein was localized in E. coli KRX. Therefore a part of the produced biomass was mechanically disrupted and the resulting lysate was washed with ddH₂O. Then the lysate was treated with ionic, nonionic and zwitterionic detergents to release the mRFP|CspB out of the membranes, if it integrates. From the other part of the cells the periplasm was detached by using an osmotic shock.

The fluorescence in all cultivation fractions plus the fluorescence in the lysis und wash fraction shows that the fusion protein is water soluble and does not sediment during centrifugation. Together with the absence of flourescence in the detergent fractions this verifies that the fusion protein is not integrated into the cell membrane (fig. 3) and is not forming inclusion bodies.

In comparison with the mRFP fusion protein of <partinfo>K525224</partinfo>, which has a TAT-sequence, a minor relative fluorescence in all cultivation and detergent fractions was detected (fig. 3). Together with the decreasing RFU after 14 h of cultivation (fig. 2) this result indicates a postive effect of the TAT-sequence on the protein stability. This could be due to a digestion of <partinfo>K525222</partinfo> by proteases in the cytoplasm.

Figure 3: Fluorescence progression of the mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)]/CspB fusion protein initiating with the cultivation fractions up to the detergent fractions of the seperate denaturations. Cultivations were carried out in autoinduction medium at 37 ˚C. The cells were mechanically disrupted and the resulting biomass was washed with ddH₂O and resuspended in the respective detergent. The used detergent acronyms stand for: SDS = 10 % sodium dodecyl sulfate; UTU = 7 M urea and 3 M thiourea; U = 10 M urea; NLS = 10 % n-lauroyl sarcosine; 2 % CHAPS = 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate.

MALDI-TOF analysis was used to identify the location of the fusion protein in different fractions. Fractions of medium supernatant after cultivation, periplasmatic isolation, cell lysis and following denaturation in 6 M urea were loaded onto a SDS_PAGE. After denaturation with 6 M urea the remaining pellet (after centrifugation 15,000 g for 30 min) was washed with 2 % (v/v) Triton X-100, 2 % SDS (w/v). This fraction was also loaded onto the SDS-PAGE and fragments of the gel were measured with MALDI-TOF.

Figure 4: SDS-PAGE of CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] fusion protein. Lanes are fractions of medium (M), periplasmatic isolation (PP), cell lysis (L), denaturation (D) and wash of the remaining pellet with Triton X-100 (T). Used marker is PageRuler ^TM Prestained Protein Ladder SM0671. Marked regions were cut out and prepared for MALDI TOF analysis.

The following table shows the sequence coverage (in %) of our measurable gel samples with the amino acid sequence of fusion protein CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)].

Fig. 5 shows these data. The gel samples were arranged after estimated molecular mass cut out from the gel.

Figure 5: MALDI-TOF measurement of CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] fusion protein. Samples are arranged after estimated molecular mass of the gel slice. Measurement was performed with a ultrafleXtreme^TM by Bruker Daltonics using the software FlexAnalysis, Biotools and SequenceEditor.

As expected, no sequence coverage was found in the periplasmatic fraction, due to absent TAT-sequence located at the amino-terminus. Little sequence coverage was found in the fraction of supernatant of the media, indicating that the protein can not be transported to the periplasm and thus secretion into the medium does not take place. The denaturation fraction and the Triton X-100 fraction show no or very few sequence coverage, however the lysis fraction shows significant higher sequence coverage. Both results indicate, that the fusion protein is solely present in the cytoplasm and thus only identified in the lysis fraction. Fig. 5 and Fig. 6 show that the protein can be found mainly in the lysis fraction, but in smaller amounts in the periplasmatic and the media fraction as well, which can be explained due to the abscence of the lipid anchor. The anchor normally binds to the cell membrane, so no protein is found in other fractions than the lysis fraction.

To obtain more specific informations about the location of the S-layer fusion protein, after comparison with same treated fraction of E. coli KRX all gel bands in a defined size area were cut out of the gel and analysed with MALDI-TOF. Results are shown in fig. 6.

Figure 6: MALDI-TOF measurement of CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] fusion protein in different fractions. Abbreviations are Ma: Marker (PageRuler ^TM Prestained Protein Ladder SM0671), M (medium), PP (periplasm), L (cell lysis with ribolyser), W (wash with ddH₂O). In the left half of the gel fractions of E. coli KRX with induced production of fusion protein, the right half shows fractions of E. coli KRX without carrying the plasmid coding the fusion protein. Colours show the sequence coverage of the gel lane, cutted out of the gel. Colours mark the sequence coverage measured with MALDI-TOF. fractions M of <partinfo>K525222</partinfo> and E. coli KRX were switched and are therefore special named.

MALDI-TOF measurement shows sequence coverage in the supernatant fraction of the cultivation, the periplasmatic fraction and the lysis fraction, indicating that the fusion protein of <partinfo>K525222</partinfo> and [http://partsregistry.org/Part:BBa_E1010 BBa_E1010] without TAT-sequence and lipid anchor is not only found in the cytoplasm. The periplasmatic isolation does destroy a few complete cells, which could leed to the sequence coverage found in the periplasmatic fraction. The gel lanes in the periplasmatic fraction are clearly less intense, the protein concentration is very low compared to the lysis fraction. Sequence coverage in the media fraction shows, that the protein is probably secreted into the media. Cell lysis could leed to this effect. The sequence coverage in a MALDI-TOF analysis does not automatically correlate with protein concentrations. Small amounts of proteins can give the same results when cutting out a clean protein band of a single protein. When looking at the intensity of the analysed gel pieces it can be summarized, that there is some CspB found in the periplasm and the culture supernatant but most of the protein stays in the cytoplasm which was the expected result. No fluorescence and sequence coverage was found in the cell membrane fractions, giving another proof that the lipid anchor from Corynebacterium binds to the cell membrane of E. coli.

CspB without TAT-sequence and with lipid anchor

Cultivation and protein expression

For characterization the modiefied CspB [http://partsregistry.org/Part:BBa_K525223 (K525223)] gen was fused with a monomeric RFP ([http://partsregistry.org/Part:BBa_E1010 BBa_E1010]) using Gibson assembly.

The fusion protein was overexpressed in E. coli KRX after induction of T7 polymerase by supplementation of 0,1 % L-rhamnose using the auinduction protocol.

Figure X: Growthcurve of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction, cultivated at 37 °C in autoinduction medium with, respectively, without inductor. A curve depicting KRX wildtype is shown for comparsion. After induction at approximately 6 h the OD₆₀₀ of the induced K525223 visibly drops when compared to the uninduced culture. While the induced culture grow significantly slower than KRX wildtype the uninduced seems to be unaffected.

Figure X: RFU to OD₆₀₀ ratio of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction. A curve depicting KRX wildtype is shown for comparsion. After induction at approximately 6 h the RFU to OD₆₀₀ ratio starts to rise in the induced culture. Compared to the uninduced culture the ratio is roughly five times higher. Most likely due to basal transcription the RFU to OD₆₀₀ ratio of the uninduced culture starts to rise after 12 hours. The KRX wildtype shows no variation in the RFU to OD₆₀₀ ratio.

Identification and localisation

After a cultivation time of 18 h the mRFP|CspB fusion protein was localized in E. coli KRX. Therefore a part of the produced biomass was mechanically disrupted and the resulting lysate was washed with ddH₂O. From the other part the periplasm was detached by using an osmotic shock.

The S-layer fusion protein could not be found in the polyacrylamide gel after a SDS-PAGE of the lysate. This indicated that the fusion protein integrates into the cell membrane with its lipid anchor. For testing this assumption the washed lysate was treated with ionic, nonionic and zwitterionic detergents to release the mRFP|CspB out of the membranes.

The existance of flourescence in two of the detergent fractions (10 % SDS and 10 % N-lauroyl sarcosine) and the low fluorescence in the wash fraction confirm the hypothesis of an insertion into the cell membrane (fig. 3). An insertion of these S-layer proteins might stabilize the membrane structure and increase the stability of cells against mechanical and chemical treatment. A stabilization of E. coli expressing S-lyer proteins was discribed by Lederer et al., (2010).

An other important fact is, that there is actually mRFP fluorescence measurable in such high concentrated detergent solutions. The S-layer seems to stabilize the biologically active conformation of mRFP.

In comparison with the mRFP fusion protein of [http://partsregistry.org/Part:BBa_K525224 K525224], wich has a TAT-sequence, a minor relative fluorescence in all cultivation and detergent fractions were detected (fig. 3). Together with the decreasing RFU after 9 h of cultivation (fig. 2) this results indicate a postive effect of the TAT-sequence on the protein stability. This could be due to a digestion of <partinfo>K525223</partinfo> by proteases in the cytoplasm.

Figure 3: Fluorescence progression of the mRFP[http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)]/CspB fusion protein initiating with the cultivation fractions up to the detergent fractions of the seperate denaturations. Cultivations were carried out in autoinduction medium at 37 ˚C. The cells were mechanically disrupted and the resulting biomass was washed with ddH₂O and resuspended in the respective detergent. The used detergent acronyms stand for: SDS = 10 % sodium dodecyl sulfate; UTU = 7 M urea and 3 M thiourea; U = 10 M urea; NLS = 10 % n-lauroyl sarcosine; 2 % CHAPS = 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate.

To obtain more specific information about the location of the S-layer fusion protein, after comparison with same treated fraction of E. coli KRX all gel bands in a defined size area were cut out of the gel and analysed with MALDI-TOF. Results are shown in fig. 4. The fusion protein CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] features a lipid anchor at the carboxy-terminus, but no amino-terminal TAT-sequence. In accordance with other protein variants with and without this features, the protein should be located mainly in the cytoplasm as inclusion bodies or incooperated with its lipid anchor into the cell membrane. Thus, the fraction with 10 % (v/v) SDS as detergent to disintegrate the protein from the cell wall was measured with MALDI-TOF. Results are shown in fig. 4.

Figure 4: MALDI-TOF measurement of CspB/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] fusion protein in different fractions. Abbreviations are Ma: Marker (PageRuler ^TM Prestained Protein Ladder SM0671), M (medium), PP (periplasm), L (cell lysis with ribolyser), W (wash with ddH₂O). In the left half of the gel fractions of E. coli KRX with induced production of fusion protein, the right half shows fractions of E. coli KRX without carrying the plasmid coding the fusion protein. Colours show the sequence coverage of the gel lane, cutted out of the gel.

Fig. 4 shows, that the protein could be identified in all measured gel bands. The results indicate, that the protein is incorperated into the cell membrane. No fluoresence could be detected in the fractions using urea as detergent (see fig. 3), thus the protein probably does not form inclusion bodies. The fact that there is a comparable low overall fluorescence in the cells is a strong indication that the protein is degraded by E. coli proteases. The fluorescence in the periplasm and the supernatant is probably due to cell lysis during the periplasm isolation and the cultivation because the measured fluorescence values are comparable low, too.

CspB with TAT-sequence and without lipid anchor

Cultivation and protein expression

For characterization the modiefied CspB [http://partsregistry.org/Part:BBa_K525224 (K525224)] gen was fused with a monomeric RFP ([http://partsregistry.org/Part:BBa_E1010 BBa_E1010]) using Gibson assembly.

The fusion protein was overexpressed in E. coli KRX after induction of T7 polymerase by supplementation of 0,1 % L-rhamnose using the auinduction protocol.

Figure X: Growthcurve of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction, cultivated at 37 °C in autoinduction medium with, respectively, without inductor. A curve depicting KRX wildtype is shown for comparsion. After induction at approximately 6 h the OD₆₀₀ of the induced K525224 visibly drops when compared to the uninduced culture. While the induced culture grow significantly slower than KRX wildtype the uninduced seems to be unaffected.

Figure X: RFU to OD₆₀₀ ratio of E. coli KRX expressing the fusion protein of CspB and mRFP with and without induction. A curve depicting KRX wildtype is shown for comparsion. After induction at approximately 6 h the RFU to OD₆₀₀ ratio starts to rise in the induced culture. Compared to the uninduced culture the ratio is roughly seven times higher at its highest point but starts to drop during the cultivation due to degradation of the fusion protein. Most likely due to basal transcription the RFU to OD₆₀₀ ratio of the uninduced culture starts to rise after 12 hours. The KRX wildtype shows no variation in the RFU to OD₆₀₀ ratio.

Identification and localisation

Figure X: Fluorescence pro OD₆₀₀ progression of the modified CspB [http://partsregistry.org/Part:BBa_K525123 (K525123)]/mRFP [http://partsregistry.org/Part:BBa_E1010 (BBa_E1010)] fusion protein initiating with the cultivation fractions up to the detergent fractions of the seperate denaturations. Cultivations were carried out in autoinduction medium at 37 ˚C. The cells were mechanically disrupted and the resulting biomass was wahed with ddH₂O and resuspendet in the respective detergent. The used detergent acronyms stand for: SDS = 10 % (v/v) sodium dodecyl sulfate; UTU = 7 M urea and 3 M thiourea; U = 10 M urea; NLS = 10 % (v/v) n-lauroyl sarcosine; CHAPS = 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate.

Purification

After the localisation of the S-layer protein in E. coli, different methods for purification were tested. The results of these methods are shown in fig. X. Fig. X shows, that the CspB protein does not form inclusion bodies in E. coli and most of the protein is transported out of the cell into the periplasm and a lot of protein is even secreted into the medium (all fractions were concentrated by filtration and precipitation, respectively). The secretion into the culture medium is very interesting because the purification is much faster (no cell disruption necessary).

Fig. X: Fluorescence of collected fractions of different methods to release and concentrate <partinfo>K525234</partinfo> protein from a cultivation in E. coli.

The highest fluorescence could be obtained by a precipitation with ammonium sulfate of the culture supernatant followed by an ultrafiltration with a 300 kDa membrane and a diafiltration with a 50 kDa membrane. The diafiltration was against a binding buffer for an anion exchange chromatography (25 mM sodium acetate, 25 mM sodium chloride) with pH 6, due to the theoretical pI of <partinfo>k525234</partinfo>. The fluorescence of the collected fractions of the following anion exchange chromatography are shown in fig. B.

Fig. B: Fluorescence of collected fractions of an anion exchange chromatography of <partinfo>K525234</partinfo> after concentration from the culture supernatant.

The binding conditions are well chosen because nearly all of the protein binds to the column. The protein is eluted from the column with rising sodium chloride concentrations. The highest fluorescence is in the elution fraction with 400 mM sodium chloride. 600 mM sodium chloride elutes all of the S-layer fusion proteins.

Final purification strategy

Scheme of purification strategy for CspB (fusion) proteins without lipid anchor:

First, CspB is expressed in E. coli under the control of a T7 promoter for separation of growth and production phase due to metabolic stress of the S-layer expression. Because the CspB protein with TAT-sequence and without lipid anchor is secreted to the culture medium in appreciable amounts by E. coli, an ammonium sulfate precipitation of the culture supernatant follows the cultivation. The S-layers are further concentrated and purified by two ultrafiltration / diafiltration steps (300 kDa and 100 kDa) with anion exchange chromatography binding buffer. The permeate of the last filtration is used for an anion exchange chromatography for capture and purification of the S-layer protein.

Click for detailed information

Summary of results