Tris Glycine Buffered Gel System Biology Essay

The discontinuous, tris-glycine buffered gel system of Laemmli was used to monitor protein expression and purification throughout the work (Laemmli, 1970). Proteins are bundled in a 5% stacking gel with pH 6.8, before entering a separating gel with pH 8.8. Separating gels were used with 6%, 12.5%, 17.5% or 20% polyacrylamide depending on the size of the protein of interest (see Table 3 for the composition of gels). The BioRad MiniProtean® electrophoresis system was used to cast gels (0.75mm spacers). Samples were mixed with Laemmli sample buffer, heated at 95°C for 5 minutes and separated on the SDS-PAGE using Laemmli running buffer at a constant voltage of 150-200V. BioRad Precision Plus ProteinTM Standards as well as PageRulerTM Plus Prestained Protein Ladder from Fermentas were used as molecular weight markers (see Figure 3 ).

Stacking gel

Separating gel

5%

6%

12.5%

17.5%

20%

30% Acrylamide

0.333ml

1.200ml

2.500ml

3.500ml

4.000ml

2.5% Bisacrylamide

0.104ml

0.209ml

0.209ml

0.209ml

0.209ml

4x Stacking/Separating gel buffer

0.500ml

1.500ml

1.500ml

1.500ml

1.500ml

H2O

1.060ml

3.060ml

1.758ml

0.758ml

0.261ml

10% Ammoniumpersulfate

0.020ml

0.050ml

0.050ml

0.050ml

0.050ml

TEMED

0.002ml

0.005ml

0.005ml

0.005ml

0.005ml

Table 3 Composition of SDS-polyacrylamide gels.

molecular weight markers.jpg

Figure 3 Protein molecular weight markers used in SDS-PAGE. (A) BioRad Precision Plus ProteinTM Standards all blue separated on a 4-20% Tris-HCl SDS gel. (B) Fermentas PageRulerTM Plus Prestained Protein Ladder separated on a 4-20% Tris-glycine SDS gel. Figures taken from manufacturer’s product insert.

4x Stacking gel buffer

1.5M Tris-HCl pH 8.8, 0.4% SDS

4x Separating gel buffer

0.5M Tris-HCl pH 6.8, 0.4% SDS

10x Laemmli running buffer

500mM Tris, 3.85M glycine, 1% SDS

2x Laemmli sample buffer

20% v/v glycerol, 10% -mercaptoethanol, 6% SDS, 125mM Tris-HCl pH 6.8, 0.01% bromphenol blue

Coomassie staining of SDS-polyacrylamide gels

Coomassie staining was used as a standard procedure to visualize protein on SDS-PAGE. Gels were soaked in stain for 10-30 minutes and destained with several washes of H2O. Subsequently, gels were dried on Whatman® paper using a SAVANT slab gel drier.

Coomassie stain

45% Methanol, 10% acetic acid, 0.4% coomassie brilliant blue R250

Western blotting

Transfer of protein onto PVDF membrane

Proteins were transferred from SDS-PAGE to Immobilon®-P Transfer membrane (Millipore) to analyze the samples with specific antibodies. A blotting sandwich in the tank blot system typically consisted of 2 sponges and 3 Whatman® papers soaked in transfer buffer, followed by the polyacrylamide gel and the PVDF membrane that has been activated in 100% methanol topped with 3 more Whatman® papers and 2 sponges (again soaked in transfer buffer). Air bubbles were carefully removed when assembling the sandwich. Transfer was achieved by applying a constant current of 40mA over night at 4°C.

Transfer buffer

25mM Tris, 192mM glycine, 20% methanol.

Blocking the membrane

Unspecific binding of antibodies was prevented by blocking the membrane with Tropix® I-blockTM (Applied Biosystems) solution. The membrane was transferred into a 50ml Falcon tube together with 5 ml blocking solution and incubated at room temperature for 30 minutes under constant rotation. Blocking solution was discarded and the procedure was repeated with another 5ml of blocking buffer.

Blocking solution

0.2% Tween 20, 0.2% Tropix® I-blockTM (Applied Biosystems) in PBS.

PBS

1.4mM KH2PO4, 2.7mM KCl, 4.3mM Na2HPO4, 137mM NaCl.

Detection of His-tagged proteins

HisProbeTM-HRP (Pierce) was used to detect His-tagged proteins. 1µl of stock solution (4mg/ml) was diluted in 5ml of blocking solution and incubated with the membrane at 4°C for 60 minutes under constant rotation. The solution was discarded and the membrane was washed three times 10 minutes with PBST and one time 10 minutes with PBS. Subsequently, the membrane was soaked for 5 min in 2ml of a 1:1 mixture of the two solutions of the SuperSignalTM West Pico kit (Pierce). Finally, the membrane was wrapped in food foil and exposed to CL-X PosureTM film (Thermo Scientific).

PBST

1.4mM KH2PO4, 2.7mM KCl, 4.3mM Na2HPO4, 137mM NaCl, 0.1% Tween 20.

Detection of eIF4G I

Detection of eIF4G I utilized an antiserum from rabbit that recognizes the N-terminal domain of eIF4G I (-eIF4G I, kind gift from R. Rhoads, Shreveport, LA). All reactions were performed at room temperature. The blocked membrane was washed with PBS for 10 minutes twice. The primary antibody (-eIF4G I, 1:8000 in PBST) was incubated with the membrane for 60 minutes under constant rotation followed by washing with PBS for 10 minutes twice. Next, an alkaline phosphatase coupled -rabbit antibody (Sigma) was used as the secondary antibody (1:5000 in PBST). Incubation for 60 minutes was followed by a final wash of the membrane with PBS (two times 10 minutes). In order to visualize bands the membrane was soaked in 5ml alkaline phosphatase buffer and 25µl NBT solution as well as 25µl BCIP solution were added. The color reaction was conducted until bands were clearly visible before the reaction was finally stopped by soaking the membrane in H2O.

NBT solution

5% (w/v) nitro blue tetrazolium in 90% dimethylformamide.

BCIP solution

2.5% (w/v) 5-bromo-4-chloro-3-indolyl phosphate disodium salt in dH2O.

Alkaline phosphatase buffer

5mM MgCl2, 100mM NaCl, 100mM Tris-HCl pH 9.6.

Expression and purification of recombinant proteins

All purification steps were performed at 4°C unless stated differently.

Expression in E. coli

After transforming the E. coli expression strain of choice (see chapter Error: Reference source not found) with the expression vector coding for the protein of interest, single colonies were picked to inoculate 100ml of LB medium including the appropriate antibiotics. The culture was incubated at 37°C/130rpm over night and used to inoculate bigger cultures (1.0-1.5 liters per 5L flask) to a starting OD6000.10.

E. coli cultures grown in M9 minimal medium were initially started by inoculating 4ml of LB medium. This culture was grown at 37°/130rpm over night and was used to inoculate a 100ml culture of M9 minimal medium including appropriate antibiotics. Again the culture was grown to stationary phase over night at 37°C/130rpm and then used to inoculate bigger cultures as described above.

Cells were typically grown at 37°C/130rpm till an OD6000.50 was reached and expression of recombinant protein could be induced by the addition of IPTG and incubation at 110rpm for a suitable time period. If possible, a 20ml aliquot was removed before induction with IPTG and used as an uninduced control, by treating it in an identical way as the induced sample except for IPTG addition. In order to increase the amount of recombinant protein found in the soluble fraction, protein expression was often performed at 16°C. For this purpose, cells were chilled on ice for 20 minutes before addition of IPTG. Alternatively, temperature was reduced to 16°C when an OD6000.30 was reached. Cells were further grown to an OD6000.50 and induced by the addition of IPTG.

As 2A protease tightly binds a Zn2+-ion in its C-terminal domain which has been reported to be essential for the structural integrity of the protease, we started to additionally add ZnCl2 (100µM end concentration) at the time of induction in order to prevent Zn2+ depletion of the cells.

Cell disruption

Cells were harvested using a Beckmann Coulter J-26 XP centrifuge equipped with a JLA-8.1000 rotor. Cultures were spun 4.000rpm/4°C/20min and the resulting pellet was usually resuspended in 30ml (per liter of culture) of lysis buffer (1ml of lysis buffer was used to resuspend the pellet of the uninduced control if available). Samples that were not immediately processed further were frozen at -80°C and thawed on room temperature when needed. Cells were broken by either sonication or the use of a homogenizer working with a French press principle. Two aliquots á 50µl used for SDS-PAGE analysis were removed before the crude cell extract was cleared by centrifugation in a SS34 rotor (20.000rpm/4°C/30min). One of the aliquots was cleared by spinning in an Eppendorf 5417R centrifuge (14.000rpm/4°C/30min). The supernatant was transferred into a fresh tube and the pellet was redissolved in 50µl lysis buffer. The total, soluble and insoluble fractions were then mixed with 50µl 2x Laemmli sample buffer each and cooked at 95°C for 15 minutes. Finally, sample was separated on SDS-PAGE and stained with coomassie blue to assess the degree of over-expression and the ratio of recombinant protein found in the soluble and insoluble fractions.

Sonication

Resuspended cells were placed in a Sonoplus RZ3 rosett cell packed in ice and sonicated via a Bandelin Sonoplus HD200 ultrasonic wave generator equipped with a SH213 booster horn and MS73 tip. The series of boosts contained three times 30sec on Cycle40 with 60sec breaks in between. Disruption was finalized by a 15sec continuous boost. The uninduced control (if available) was processed by a series of three boosts of 30sec at Cycle20.

Disruption of cells via homogenizer

As an alternative to sonication cells were broken with the Emulsiflex-C3 homogenizer (Avestin). Cells were passed through the outlet valve under a pressure of 1000-1500 bar for three to five times till a homogenous, non viscous sample was obtained. As the device requires a certain minimal sample volume to be operated, the cells of the uninduced control could not be disrupted with this method.

Fractionated (NH4)2SO4 precipitation

This classic purification strategy was used early in the purification process and is based on the diverse solubility of proteins under high ionic strength. Stepwise addition of a saturated aqueous (NH4)2SO4 solution at 4°C under stirring was used to salt out proteins in two steps. First, the protein sample was incubated at a lower concentration of (NH4)2SO4 at 4°C for at least 12 hours leaving the protein of interest in solution. Precipitated proteins were removed by pelletizing them in a SS34 rotor (20.000rpm/4°C/30min). Subsequently, the supernatant was mixed with further (NH4)2SO4 and incubated for at least 4 hours. Typically, the concentration was chosen to precipitate the protein of interest in this second fraction. The precipitated protein was again collected as described above and resuspended in a buffer of choice. Routinely, these samples were dialyzed against the chosen buffer to remove residual (NH4)2SO4 that could interfere with subsequent purification steps.

Use of ÄKTA FPLC in protein purification

An ÄKTA FPLC system (GE Healthcare) composed of a P-920 pump unit, a UPC-900 detection unit, several valves and a Frac-950 was used to drive protein purification via anion exchange chromatography, His-trap affinity chromatography, GST-trap affinity chromatography, desalting and size exclusion chromatography. Systems were initially available at room temperature and on 4°C in the later experiments of this work. All solutions as well as samples were filtered through a 0.22µm filter in order to prevent system filters as well as on column filters from clogging.

Anion exchange chromatography

A MonoQ 10/100 GL column (Pharmacia) was used to separate proteins according to their surface charge. The pH of the mobile phase was chosen to allow binding of the protein of interest to the resin. For this purpose the theoretical pI of the recombinant protein was calculated by the online application provided by ExPASy (http://web.expasy.org/compute_pi/pi_tool-ref.html) (Bjellqvist et al., 1994; Bjellqvist et al., 1993) and a pH that was at least 2 units higher than the obtained value was chosen for the mobile phase in order to allow proper binding to the matrix even in the presence of higher salt concentrations in the binding buffer. The column was equilibrated with at least 3 column volumes (CV) of binding buffer and sample was typically loaded via a 50ml superloop. Flow through was collected and unbound sample was washed from the column with binding buffer before bound protein was eluted via a NaCl gradient (up to 1M NaCl) applied by the pumps of the ÄKTA FPLC system. The eluent was collected in tubes á 2ml. The column was washed with 5CV of buffer containing 1M NaCl to elute tightly bound contaminants and finally washed with H2O MILLI Q as well as 20% ethanol and stored at 4°C. Aliquots of fractions of interest were separated on SDS-PAGE and stained with coomassie blue for analysis.

Affinity chromatography

Purification via His-trap

His-Trap HP (Pharmacia) columns were used for high resolution purification of His-tagged proteins. Columns were operated using an ÄKTA FPLC. If stripped, columns were charged by loading 2CV 300mM NiCl2, followed by washing with H2O to remove excess NiCl2. Against the manufacturer’s product information, the use of 5mM DTT or 20mM -mercaptoethanol in buffers or sample was not possible in our hands. Hence, buffers and sample had to be free of reducing agents for purification with His-trap columns. Columns were equilibrated with 5CV binding buffer (including up to 30mM imidazole). Sample was loaded onto the column with a flow rate of 0.5-1.0ml/min and flow through was collected. Unbound protein was removed from the column by washing with binding buffer till the UV signal reproached the baseline (5-10CV). Bound protein was eluted from the column by applying a linear imidazole gradient up to 500mM imidazole. The eluent was typically collected in fractions á 1.0-2.0ml. The column was washed with 10CV H2O followed by 10CV 20% ethanol and stored at 4°C. Fractions of interest were inspected by SDS-PAGE.

Stripping of the column was achieved by passing 5-10CV of binding buffer including 50mM EDTA over the beads. Subsequently, the matrix was washed with water and stored in 20% ethanol.

Purification via amylose resin

Maltose binding protein (MBP) served not only as a solubility tag but also as a purification tag. Typically, 10ml of amylose resin (New England BioLabs) were packed into Econo-Pac gravity flow columns (BioRad). Beads were equilibrated with 5CV of binding buffer before sample was loaded and flow through was collected. Unbound protein was washed off the column with binding buffer (5CV) and protein was eluted using a step gradient of maltose concentration (binding buffer +10mM maltose). The eluent was collected in fractions á 10ml which were inspected by SDS-PAGE and coomassie staining.

Resin was regenerated by washing with H2O (5CV) followed by stripping with 3CV 0.1% SDS at room temperature (to avoid precipitation of SDS). SDS was removed by extensive washing with H2O and the resin was stored in 20% ethanol.

Purification via GSTrap

A 5ml GSTrap HP (Pharmacia) was used to purify proteins tagged with glutathione S-transferase (GST) on an ÄKTA FPLC system. The column was equilibrated with 10CV of binding buffer before sample was loaded at a flow rate of 1ml/min and flow through was collected. Unbound protein was removed by washing with 10CV of binding buffer before elution of bound proteins was induced by a step gradient of reduced glutathione (binding buffer +10mM reduced glutathione) at a flow rate of 5ml/min. The eluent was collected in fractions á 2ml and screened for protein of interest by SDS-PAGE and coomassie staining. The column was washed with 10CV H2O and stored in 20% ethanol at 4°C.

Size exclusion chromatography

Purification via Superdex

Superdex 75 prep grade 26/60 as well as Superdex 200 prep grade 26/60 columns (Pharmacia) were used with an ÄKTA FPLC system for gel filtration as the last step in the purification procedure. Columns were equilibrated with 2CV of running buffer. Sample was concentrated to a maximal volume of 5ml (typically 1-2ml) before being loaded onto the matrix. Isocratic elution was performed with the equilibration buffer at a typical flow rate of 2.0-2.5ml/min and eluent was collected in fractions á 2ml. Fractions of interest were screened via SDS-PAGE and coomassie staining. Columns were washed with 2CV of H2O and stored in 20% ethanol.

Desalting and buffer exchange

Fast and efficient desalting and buffer exchange was executed via a HighPrep Desalting 26/10 column (GE Healthcare). The column was equilibrated with the new buffer of choice, concentrated sample (<15ml) was loaded and the column was developed with the new buffer of choice at flow rates up to 30ml/min. The column was washed with H2O and stored in 20% ethanol.

Other protein methods

Dialysis

SnakeSkin® dialysis tubing (Thermo Scientfic) with 10kDa or 3.5kDa cutoff was used for dialyzing protein samples. Typically the sample was dialyzed against three changes of dialysis buffer at 4°C under constant stirring for at least 4 hours per change.

Concentrating protein samples

Centriprep® centrifugal filter units (Millipore) with a cutoff of 10kDa were mainly used to concentrate protein samples. Units including protein sample were spun at a maximum of 3.000g at 4°C in a swinging bucket rotor till the protein concentration of choice was reached. Centriprep® centrifugal filter units exhibit an internal hold-up volume of 0.5ml at which no more concentration can be obtained. For further concentration Amicon® Ultra-4 centrifugal filters (Millipore) with a cutoff of 10kDa were used at a maximum of 4.000g at 4°C in a swinging bucket rotor.

Assessing protein concentration

Protein concentration was assessed using a NanoDrop ND-1000 spectrophotometer by measuring absorbance at 280nm. Molar extinction coefficients were estimated using the ProtParam web server (provided by ExPASy) (Wilkins et al., 1999). The Server uses the equation by Gill et al. to predict the molar extinction coefficient (Gill & von Hippel, 1989). Having detected these values, the concentration was calculated using Lambert Beer’s law.

Reductive lysine methylation

Reductive lysine methylation was performed following the protocol of Walter et al. (Walter et al., 2006). Protein was re-buffered to 50mM HEPES pH 7.5, 250mM NaCl using a HighPrep Desalting 26/10 column (see chapter 1.2.7.2). Protein was diluted to 0.5mg/ml and methylation was performed in a closed hood at 4°C. 20µl 1M dimethyl-borane and 40µl 1M formaldehyde per ml of protein solution was added, the reaction vessel was carefully closed with parafilm and incubated under stirring at 4°C for 2 hours. Addition of 1M dimethyl-borane and 40µl 1M formaldehyde per ml of protein solution was repeated and the mixture was again incubated for 2 hours under stirring at 4°C. Finally, another 10µl of 1M dimethyl-borane complex per ml of protein solution were added and the reactions was finalized by incubation over night. Protein was repurified via size exclusion chromatography (see chapter 1.2.7.1) using the buffer of choice.

1M dimethyl-borane complex

Freshly dissolve in H2O MILLI Q directly before usage.

Thermofluor

Thermofluor allows fast screening of buffer conditions for maximal stability of proteins. The method is based on the binding of a fluorophore (Sypro Orange) to the hydrophobic patches of proteins that will be exposed upon stepwise heating. Binding to those hydrophobic patches is leading to a sigmoid increase of fluorescence that is otherwise quenched by the aqueous solution. Upon total unfolding of the protein fluorescence will again decline. The inflection point of the sigmoid rise is defined as the melting point of the protein. Buffer, pH, ionic strength and different additives are mixed with the protein to detect the buffer that stabilizes the protein of interest best.

Preliminary experiments without screening of different conditions were conducted to identify the optimal ratio of protein and dye. In a 5µl reaction all combinations of protein concentrations (0.20mg/ml, 0.10mg/ml and 0.05mg/ml) and dye concentrations (100x, 10x and 1x (from a 5000x stock)) were tested. The combination with the strongest signal was chosen and 2x96 different conditions were screened by mixing 4µl of screening solution with 1µl of a mix made of protein and dye. Fluorescence was measured using a qPCR system equipped with a 492nm extinction and 516nm emission filter. The temperature gradient was established in steps of 1°C from 20°C to 95°C. Screened conditions are listed in Table 3 and Table 3 .

Crystallization of proteins

Automated setup of screens

Large scale screens were set up using a nanoliter crystallization robot (Phoenix). MRC 2-well plates (Swissci) were used to equilibrate two sitting drops of different composition against 50µl of reservoir solution. Drops consisted of 100nl protein and 100nl reservoir solution or 200nl protein and 100nl reservoir solution respectively. Drops were set at 22°C, plates were sealed with CrystalClear tape and incubated at either 4°C or 22°C. An automated imaging system (Rigaku) assisted in easy online inspection of drops by imaging drops at time points 0h, 12h, 1d, 2d, 3d, 7d, 12d, 19d and 31d. For further inspections drops were examined manually under a microscope.

Manual setup of drops

The hanging drop vapor diffusion setup was chosen to screen around crystal hit conditions and to grow crystals at established crystallization conditions. When using 48-well plates, 2-3 drops of 2-3µl each were equilibrated against 100µl reservoir solution per well. In contrast, up to 7 drops of 3-5µl each per well were used in 24-well plates and equilibrated against 500µl of reservoir solution. Plates were stored at 4°C or 22°C and monitored either on an automated Rigaku imaging system or manually under the microscope.

Harvesting and soaking of crystals

Crystals were carefully fished from the drops with a loop under the microscope at 4°C and washed in a fresh drop of reservoir solution. Subsequently crystals were transferred into a drop of reservoir solution containing the inhibitor concentration of interest and soaked for an appropriate time. Finally, crystals were transferred into cryo solution (reservoir solution including 21% v/v glycerol) followed by freezing in liquid nitrogen.

1

2

3

4

5

6

7

8

9

10

11

12

A

Sodium acetate

Sodium citrate

Sodium acetate

Potassium phosphate

Sodium phosphate

Sodium citrate

MES

Potassium phosphate

MES

Sodium phosphate

Sodium cacodylate

MES

pH 4.5

pH 4.7

pH 5.0

pH 5.0

pH 5.5

pH 5.5

pH 5.8

pH 6.0

pH 6.2

pH 6.5

pH 6.5

pH 6.5

B

Sodium acetate

Sodium citrate

Sodium acetate

Potassium phosphate

Sodium phosphate

Sodium citrate

MES

Potassium phosphate

MES

Sodium phosphate

Sodium cacodylate

MES

pH 4.5

pH 4.7

pH 5.0

pH 5.0

pH 5.5

pH 5.5

pH 5.8

pH 6.0

pH 6.2

pH 6.5

pH 6.5

pH 6.5

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

C

Sodium acetate

Sodium citrate

Sodium acetate

Potassium phosphate

Sodium phosphate

Sodium citrate

MES

Potassium phosphate

MES

Sodium phosphate

Sodium cacodylate

MES

pH 4.5

pH 4.7

pH 5.0

pH 5.0

pH 5.5

pH 5.5

pH 5.8

pH 6.0

pH 6.2

pH 6.5

pH 6.5

pH 6.5

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

D

Sodium acetate

Sodium citrate

Sodium acetate

Potassium phosphate

Sodium phosphate

Sodium citrate

MES

Potassium phosphate

MES

Sodium phosphate

Sodium cacodylate

MES

pH 4.5

pH 4.7

pH 5.0

pH 5.0

pH 5.5

pH 5.5

pH 5.8

pH 6.0

pH 6.2

pH 6.5

pH 6.5

pH 6.5

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

E

Sodium acetate

Sodium citrate

Sodium acetate

Potassium phosphate

Sodium phosphate

Sodium citrate

MES

Potassium phosphate

MES

Sodium phosphate

Sodium cacodylate

MES

pH 4.5

pH 4.7

pH 5.0

pH 5.0

pH 5.5

pH 5.5

pH 5.8

pH 6.0

pH 6.2

pH 6.5

pH 6.5

pH 6.5

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

F

Sodium acetate

Sodium citrate

Sodium acetate

Potassium phosphate

Sodium phosphate

Sodium citrate

MES

Potassium phosphate

MES

Sodium phosphate

Sodium cacodylate

MES

pH 4.5

pH 4.7

pH 5.0

pH 5.0

pH 5.5

pH 5.5

pH 5.8

pH 6.0

pH 6.2

pH 6.5

pH 6.5

pH 6.5

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

G

Sodium acetate

Sodium citrate

Sodium acetate

Potassium phosphate

Sodium phosphate

Sodium citrate

MES

Potassium phosphate

MES

Sodium phosphate

Sodium cacodylate

MES

pH 4.5

pH 4.7

pH 5.0

pH 5.0

pH 5.5

pH 5.5

pH 5.8

pH 6.0

pH 6.2

pH 6.5

pH 6.5

pH 6.5

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

H

Sodium acetate

Sodium citrate

Sodium acetate

Potassium phosphate

Sodium phosphate

Sodium citrate

MES

Potassium phosphate

MES

Sodium phosphate

Sodium cacodylate

MES

pH 4.5

pH 4.7

pH 5.0

pH 5.0

pH 5.5

pH 5.5

pH 5.8

pH 6.0

pH 6.2

pH 6.5

pH 6.5

pH 6.5

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

Table 3 Thermofluor ionic strength screen I. Composition of the 96 conditions of the ionic strength screen (plate I) used for thermofluor measurements.

1

2

3

4

5

6

7

8

9

10

11

12

A

Potassium phosphate

HEPES

Ammonium acetate

Sodium phosphate

Tris

Imidazole

HEPES

Tris

Bicine

Tris

Bicine

H2O

pH 7.0

pH 7.0

pH 7.3

pH 7.5

pH 7.5

pH 8.0

pH 8.0

pH 8.0

pH 8.0

pH 8.5

pH 9.0

B

Potassium phosphate

HEPES

Ammonium acetate

Sodium phosphate

Tris

Imidazole

HEPES

Tris

Bicine

Tris

Bicine

H2O

pH 7.0

pH 7.0

pH 7.3

pH 7.5

pH 7.5

pH 8.0

pH 8.0

pH 8.0

pH 8.0

pH 8.5

pH 9.0

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

100mM NaCl

C

Potassium phosphate

HEPES

Ammonium acetate

Sodium phosphate

Tris

Imidazole

HEPES

Tris

Bicine

Tris

Bicine

H2O

pH 7.0

pH 7.0

pH 7.3

pH 7.5

pH 7.5

pH 8.0

pH 8.0

pH 8.0

pH 8.0

pH 8.5

pH 9.0

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

500mM NaCl

D

Potassium phosphate

HEPES

Ammonium acetate

Sodium phosphate

Tris

Imidazole

HEPES

Tris

Bicine

Tris

Bicine

H2O

pH 7.0

pH 7.0

pH 7.3

pH 7.5

pH 7.5

pH 8.0

pH 8.0

pH 8.0

pH 8.0

pH 8.5

pH 9.0

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

100mM KCl

E

Potassium phosphate

HEPES

Ammonium acetate

Sodium phosphate

Tris

Imidazole

HEPES

Tris

Bicine

Tris

Bicine

H2O

pH 7.0

pH 7.0

pH 7.3

pH 7.5

pH 7.5

pH 8.0

pH 8.0

pH 8.0

pH 8.0

pH 8.5

pH 9.0

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

500mM KCl

F

Potassium phosphate

HEPES

Ammonium acetate

Sodium phosphate

Tris

Imidazole

HEPES

Tris

Bicine

Tris

Bicine

H2O

pH 7.0

pH 7.0

pH 7.3

pH 7.5

pH 7.5

pH 8.0

pH 8.0

pH 8.0

pH 8.0

pH 8.5

pH 9.0

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

100mM (NH4)2SO4

G

Potassium phosphate

HEPES

Ammonium acetate

Sodium phosphate

Tris

Imidazole

HEPES

Tris

Bicine

Tris

Bicine

H2O

pH 7.0

pH 7.0

pH 7.3

pH 7.5

pH 7.5

pH 8.0

pH 8.0

pH 8.0

pH 8.0

pH 8.5

pH 9.0

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

10% glycerol

H

Potassium phosphate

HEPES

Ammonium acetate

Sodium phosphate

Tris

Imidazole

HEPES

Tris

Bicine

Tris

Bicine

H2O

pH 7.0

pH 7.0

pH 7.3

pH 7.5

pH 7.5

pH 8.0

pH 8.0

pH 8.0

pH 8.0

pH 8.5

pH 9.0

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

20% glycerol

Table 3 Thermofluor ionic strength screen II. Composition of the 96 conditions of the ionic strength screen (plate II) used for thermofluor measurements.

Data collection and processing

Data collection

All crystals were mounted on an in house X-ray source (Bruker Microstar (Bruker AXS Inc.) rotating anode at =1.54Å) and typically exposed for 10 seconds to the beam. Crystals with good diffraction were saved and re-frozen in liquid nitrogen whereas crystals with low or no diffraction were discarded.

For the collection of a full dataset crystals were sent to the BESSY synchrotron in Berlin. Data was collected at beamline BESSY_14.1 (=0.918Å) on a MarMosaic 225 detector.

Data processing

Indexing, cell refinement and integration of the data set was calculated using Mosflm (Leslie & Powell, 2007). Data was scaled and intensities were converted to structure factors with SCALA (CCP4, 1994).

Solving the structure

Solving the phase problem by molecular replacement

Molecular replacement was used to solve initial phases to model HRV2 2Apro bound to inhibitors. The programs Phaser (McCoy et al., 2007) implemented in the PHENIX program suite (Adams et al., 2010) or MolRep (Vagin & Teplyakov, 2010) implemented in the CCP4 program suite (Winn et al., 2011) were used for this purpose.

Model building and refinement

Models were built using the molecular graphics system COOT (Emsley & Cowtan, 2004) and refined using programs implemented in PHENIX and CCP4.

Expression and purification of proteases needed for the elimination of tags

Tobacco Etch Virus (TEV) protease

A bacterial glycerol stock of E. coli BL21 (DE3) transformed with an expression construct coding for an MBP-TEV protease fusion protein was obtained as a kind gift from Gang Dong. Mature TEV-protease is yielded upon self-processing of the protease on its own N-terminus. A single colony of cells was used to inoculate a 20ml starter culture of LB-Amp which was grown at 37°C/200rpm over night. On the following day two liters of LB-Amp were inoculated with the starter culture till an OD6000.1 was reached. Cells were grown to and OD6000.5-0.6 at 37°C/150rpm. Expression was induced by the addition of 0.5mM IPTG and incubation was continued at 37°C/150rpm for three to four hours.

Cells were harvested, resuspended in lysis buffer (20mM Tris-HCl pH 8.0, 300mM NaCl, 5% v/v glycerol, 20mM imidazole, 15mM BME) and disrupted via a homogenizer as described in chapter 1.2.2. The supernatant was then loaded onto a 5ml His-Trap column that has been equilibrated with 5CV of lysis buffer. Purification was performed as described in 1.2.6.1 with the exception that a step gradient (200mM imidazole) was used to elute TEV-protease directly into 20ml of lysis buffer including 20% v/v glycerol to prevent TEV-protease from precipitating. Aliquots á 500µl were stored at -80°C.

HRV14 3C protease (PreScission protease)

Expression constructs for GST-PreScission protease and His10-GST-Prescission protease fusions were obtained as a kind gift from Georg Mlynek. E. coli BL21 (DE3) Rosetta2 pLysS cells were transformed with either of the plasmids and plated on LB-Amp plates. Proteins were expressed as described in chapter 1.2 from 1l of LB-Amp using 0.2mM IPTG at 16°C. Harvesting of cells (lysis buffer: 50mM Tris-HCl pH 8.1, 100mM NaCl, 1mM EDTA, 1mM DTT) and cell disruption via a homogenizer was performed as described in chapter 1.2.2 and PreScission protease was purified via affinity chromatography using a GSTrap column (see chapter 1.2.6.3). Finally, protease was stored in 50mM Tris-HCl pH8.1, 150mM NaCl, 10mM EDTA, 1mM DTT and 20% v/v glycerol at -20°C.

Results

Snapshooting 2Apro during self-processing – an attempt

Only two entries on enteroviral 2Apro can be found in the PDB-database. These are the X-ray structure of HRV2 2Apro (PDB 2HRV) (Petersen et al., 1999) and the NMR structure of CVB4 2Apro (PDB 1Z8R) (Baxter et al., 2006). The coordinates of HRV2 2Apro revealed a conformation corresponding to a post-self-processing state. The N-terminus is rotated out of the active site, allowing the protease to cleave substrates in trans without self inhibition (see chapter Error: Reference source not found). The data is therefore of limited value in terms of interpreting the substrate specificity of 2Apro’s self-processing reaction. In order to overcome this problem Baxter et al. expressed and purified inactive 2Apro with an N-terminal extension of eight amino acids (see chapter Error: Reference source not found). The solution structure was solved with a set of NMR experiments. Nevertheless, information on self-processing could not be extracted as no signals could be obtained for the N-terminal residues due to conformational flexibility.

We speculated that the construct used in the NMR studies or rather similar constructs could be useful for crystallization as crystal packing and optimization of the length of the N-terminal extension could favor trapping of the substrate in the binding cleft during self-processing. Hence, we designed a number of constructs differing in expression and purification strategy as well as in the length of the N-terminal extension from different enteroviruses. The purified proteins were then subjected to crystallization trials.

Poliovirus 2Apro constructs

pET 3d PV1 VP18-2Apro (C109A)

All purification steps on ÄKTA FPLC were performed at room temperature.

aa: 157

Mw: 17.5 kDa

pI (theoretical): 5.79

An XhoI site was introduced into pET 3d CVB4 His6-VP18-2Apro (C110A) (Baxter et al., 2006) by replacing the BamHI/BlpI fragment with an oligonucleotide cassette containing an XhoI site (TIM 1729/1730) (Error: Reference source not found). Two step PCR mutagenesis (see chapter Error: Reference source not found) was used to amplify a 1558 bp fragment containing the sequence coding for a start codon followed by PV1 VP18-2Apro (C109A) (TIM 1704, TIM 1705, TIM 1717, TIM 1728) (Error: Reference source not found) from the plasmid pCITE PV1 VP1-2Apro (Andreas Rötzer, unpublished). The fragment was treated with BspHI and XhoI and used to replace an NcoI/XhoI fragment of the pET 3d CVB4 His6-VP18-2Apro (C110A) / XhoI plasmid (Error: Reference source not found).

E. coli BL21 (DE3) cells were transformed with the plasmid and plated on LB-Amp agar plates. Single colonies were picked and used to inoculate 100ml of LB-Amp. The starter culture was grown at 37°C/160rpm over night. The over-night culture was diluted into 1 liter of fresh LB-Amp medium to reach a starting optical density of OD6000.1. Bacteria was grown at 32°C/160rpm till and OD6000.5-0.6 was reached and chilled on ice for 30 minutes. After removing a 20ml aliquot (used as uninduced control) expression was induced by the addition of 0.1mM IPTG. Protein was expressed for 5 hours at 16°C/110rpm.

Cells were collected by centrifugation at (see chapter 1.2.2). The pellet was resuspended in 30ml Buffer A (50mM Tris-HCl pH 7.5, 50mM NaCl) per liter of culture. The uninduced control was resuspended in 1ml Buffer A. Cells were frozen at -80°C over night and thawed on RT on the next day. Breakup of cells was finalized by sonication (see chapter 1.2.2.1). Aliquots of total, soluble and insoluble fractions were subjected to 17.5% SDS-PAGE and stained with coomassie blue.

pET VP1(8)-2A C109A cce.jpg

Figure 4 Expression of PV1 VP18-2Apro (C109A) in E. coli BL21 (DE 3). 9µL of total (T), soluble (S) and insoluble (I) fractions of induced as well as uninduced samples were separated on 17.5% SDS-PAGE and stained with coomassie blue. Bands that potentially arise from the over-expression of recombinant protein are marked with an arrow.

A strong band corresponding to the size of the construct was detected in the induced as well as the uninduced fraction (marked with an arrow) (Figure 4 ). In the absence of a specific antibody for PV1 2Apro we were unable to verify if the band contained the recombinant protein. We speculated that the presence of 2Apro in the uninduced control could be a result of leaking expression in the BL21 (DE3) cells. A His-tagged version of the construct (see chapter 2.1.1.2) was used to confirm the identity of the bands as 2Apro.

However, the protein is mainly found in the insoluble fraction, even under the mild over-expression conditions of 0.1mM IPTG and 16°C. Nevertheless, we headed for purification via fractionated ammonium sulfate precipitation (see chapter 1.2.3) using steps of 20% and 45% saturation of salt. Proteins in the pellet of the second step were dissolved in 4ml Buffer B (50mM Tris-HCl pH 8.1, 50mM NaCl, 5mM DTT) per liter of bacterial culture. The sample was dialyzed against three washes of Buffer B á 500ml to remove residual ammonium sulfate. Protein aggregates formed during dialyzes were collected by centrifugation at 4°C/20.000rpm/30min and redissolved in Buffer B. Aliquots of the dialyzed sample, redissolved protein aggregates after dialyzes and supernatant of the second precipitation step were then analyzed on 17.5% SDS PAGE (Figure 4 ).

Figure 4 Analyzing the fractionated ammonium sulfate precipitation of PV1 VP18-2Apro (C109A) by 17.5% SDS-PAGE. Lanes 3, 6 and 9 show aliquots of the redissolved and dialyzed sample after the second precipitation step, the redissolved aggregates and the supernatant of the second precipitation step respectively. Protein bands of interest are highlighted with an arrow. Lanes 1, 2, 4, 5, 7 and 8 correspond to different constructs and are not to be concerned.pET PV1 VP1(8)-2A C109A asp.jpg

PV1 VP18-2Apro (C109A) precipitates in the 45% ammonium sulfate fraction (see arrow in Figure 4 ). Hence, it was used for further purification by anion exchange chromatography on an ÄKTA FPLC system on room temperature. A MonoQ 10/100 GL was equilibrated with 5 CV of Buffer B (50mM Tris-HCl pH 8.1, 50mM NaCl, 5mM DTT). The sample was loaded via a superloop and unbound sample was washed off the column using buffer B until the UV-signal reapproached the baseline. Subsequently, the column was developed with a NaCl gradient using Buffer C (50mM Tris-HCl pH 8.1, 50mM NaCl, 5mM DTT, 1M NaCl) on pump B of the ÄKTA FPLC system. Eluted protein was detected by tracing absorption of UV-light at 280 nm and fractions of 2ml were collected (Figure 4 ).

The chromatogram shows only minor peaks in the low salt regions and displays only one pronounced peak at around 750mM NaCl (retention volume=217ml). Running 9µl of each peak-fraction on a 17.5% SDS-PAGE should confirm whether one of the peaks contained the recombinant protein. However, no protein corresponding to the size of PV1 VP18-2Apro (C109A) could be detected, neither in one of the fractions nor in the collected flow through. In fact the only peak containing any detectable protein by coomassie staining was found to be the one at a retention volume of (Vr) 84ml (data not shown). Consequently, no further purification steps could be performed.

Figure 4 Chromatogram of PV1 VP18-2Apro (C109A) MonoQ run. PV1 VP18-2Apro (C109A) containing sample was loaded on a MonoQ 10/100 GL using an ÄKTA FPLC system. Bound proteins were eluted by applying a NaCl gradient (green). Peaks in the UV trace (blue) are labeled with the respective retention volumes.

pET 3d PV1 His6-VP18-2Apro (C109A)

All purification steps on ÄKTA FPLC were performed at room temperature.

aa: 163

Mw: 18.3 kDa

pI (theoretical): 6.31

Cloning of the fragment was identical to pET 3d PV1 VP18-2Apro (see section 2.1.1.1) with exception of the usage of TIM 1716 (instead of TIM 1717) (Error: Reference source not found). The construct consists of an N-terminal hexa-His-tag directly fused to the last eight residues of VP1 and the inactivated 2Apro (Error: Reference source not found).

E. coli BL21 (DE3) were transformed with the corresponding plasmid and recombinant protein was expressed as described in section 2.1.1.1. Cells were processed as in chapter 2.1.1.1 and 9µL of the total, soluble and insoluble fractions were analyzed on a 17.5% SDS-PAGE (Figure 4 ).

pET His(6)-VP1(8)-2A C109A cce.jpg

Figure 4 Expression of PV1 His6-VP18-2Apro (C109A) in E. coli BL21 (DE3). 9µL of total (T), soluble (S) and insoluble (I) fractions of induced as well as uninduced samples were separated on 17.5% SDS-PAGE and stained with coomassie blue. Bands that potentially arise from the over-expression of recombinant protein are marked with an arrow.

A strong band running between 15 and 20 kDa is detected in the induced as well as in the uninduced fraction. Again the protein seems to be highly insoluble under the expression conditions mentioned above. In order to confirm the identity of the bands, proteins were blotted on PVDF membrane and analyzed via His-probe (Figure 4 ) (see chapter 1.1.2.3). The blot clearly identifies the bands as the protein of interest. However, it also confirms the insolubility of poliovirus 2Apro.

pET His(6)-VP1(8)-2A C109A cce blot.jpg

Figure 4 -His blot of E. coli BL21 (DE3) crude cell extracts containing PV1 His6-VP18-2Apro (C109A). 9µL of total (T), soluble (S) and insoluble (I) fractions of induced as well as uninduced samples were separated on 17.5% SDS-PAGE, blotted on PVDF membrane and analyzed using His-probe.

In order to enhance solubility we aimed to slow down the rate of expression by using M9 minimal medium (see chapter Error: Reference source not found). Furthermore, the effect of different cell lysis methods should be examined. Cells were split after expression and either broken by the usual sonicaton or by the use of a high pressure homogenizer (Figure 4 ) (see chapter 1.2.2).

pET PV1 His(6)-VP1(8)2A C109A cce minimal.jpg

Figure 4 Expression of PV1 His6-VP18-2Apro (C109A) in E. coli BL21 (DE3) grown in M9 minimal medium. 9µL of total (T), soluble (S) and insoluble (I) fractions of induced as well as uninduced samples were separated on 17.5% SDS-PAGE and stained with coomassie blue. Bands that potentially arise from the over-expression of recombinant protein are marked with an arrow. (A) Crude extract from cells broken by sonication. (B) Crude extract from cells broken with a high pressure homogenizer.

No essential differences between the two modes of cell disruption could be detected. Neither the total amount of protein nor the fraction of soluble recombinant protein is dependent on the method. Moreover, the use of M9 minimal medium did not seem to stimulate the production of soluble protein as well. The coomassie blue stained gels as well as the blots developed with His-probe (Figure 4 ) did not indicate the presence of any soluble recombinant protein.

pET PV1 His(6)-VP1(8)2A C109A cce minimal blot.jpg

Figure 4 -His blot of E. coli BL21 (DE3) crude cell extracts including PV1 His6-VP18-2Apro (C109A). 9µL of total (T), soluble (S) and insoluble (I) fractions of cells broken either by sonication or a high pressure homogenizer were separated on 17.5% SDS-PAGE, blotted on PVDF membrane and analyzed using His-probe.

However, soluble fractions of both breakup methods were pooled and fractionated ammonium sulfate precipitation was applied as described before. The precipitated protein of the second fraction was dissolved in 4ml Buffer B (50mM Tris-HCl pH 8.1, 50mM NaCl) and dialyzed against three washes of 500ml of the same buffer. A 1ml His-Trap column (see chapter 1.2.6.1) was equilibrated with 10ml of Buffer B (50mM Tris-HCl pH 8.1, 50mM NaCl). Sample was loaded via a superloop at a flow rate of 0.5ml/min and the column was developed with a gradient of imidazole (up to 500mM).

pET PV1 His(6)-VP1(8)2A C109A minimal His trap gel.jpg

Figure 4 His-Trap purification of PV1 His6-VP18-2Apro (C109A). (A) PV1 His6-VP18-2Apro (C109A) containing sample was loaded on a 1ml nickel charged His-Trap column. Protein (blue UV trace) was eluted from the column with rising concentrations of imidazole in the elution buffer (green curve) and fractions á 1ml were collected. (B) Aliquots (9µL of load (L), flow through (FT) and elution fractions) of the protein containing peak (Vr=9.12 ml) were separated on a 17.5% SDS-PAGE and stained with coomassie blue. Recombinant protein bands are marked with an arrow.

The collected fractions contain a substantial amount of impurities. The lack of a low concentration of imidazole in the binding buffer might have prevented a more effective purification (Figure 4 ). However, a low amount of PV1 His6-VP18-2Apro (C109A) could be detected (see lanes corresponding to Vr=8-13ml in Figure 4 B). These fractions were pooled, concentrated to a volume of 1ml, dialyzed against Buffer B to remove imidazole and loaded onto a Superdex 75 prep grade 26/60 that has been equilibrated with Buffer B (50mM Tris-HCl pH 8.1, 50mM NaCl). Isocratic elution was performed with Buffer B and fractions of 2ml were collected. However, the amount of loaded protein seemed to be too low as no protein could be recovered from the size exclusion run (data not shown). Even a scale up of the expression to 6 liters of bacterial culture did not allow to recover any recombinant protein from the size exclusion column (data not shown).

Mal pET PV1 VP18-2Apro (C109A)

In order to increase the solubility of PV1 2Apro we aimed to produce a fusion protein using maltose binding protein (MBP) as a solubility tag. The protein of interest can be cleaved from the tag via TEV protease.

Fusion protein

aa: 586

Mw: 64.7 kDa

pI (theoretical): 5.77

Product after TEV cleavage

aa: 162

Mw: 18.0 kDa

pI (theoretical): 5.93

The plasmid Mal pET was a kind gift of Gang Dong (Max F. Perutz laboratories) and has been described in chapter Error: Reference source not found.

A fragment coding for PV1 VP18-2Apro (C109A) was amplified from pET 3d PV1 His6-VP18-2Apro (C109A) using the primers TIM 1759 and TIM 1760 (Error: Reference source not found). TIM 1759 contained the sequence of an NheI site, TIM 1760 the sequence of a BamHI site in their 5’ nonbinding overhang. Ligation of the NheI/BamHI digested PCR fragment into an equally treated Mal pET vector results in a plasmid coding for a fusion protein with five additional amino acids between TEV cleavage site and the VP18-2Apro (C109A) sequence (Error: Reference source not found). E. coli BL21 (DE3) Rosetta2 pLysS were transformed with the plasmid and plated on LB-agar plates containing kanamycine as the selecting antibiotic. Single colonies were used to inoculate LB-medium.

A small scale test expression should probe optimal expression conditions. Three times 100ml of LB-Kan medium were inoculated with an overnight culture of E. coli BL21 (DE3) Rosetta pLysS cells transformed with Mal pET PV1 VP18-2Apro (C109A) and grown to an OD6000.5 at 32°C. Subsequently, samples were induced via addition of 0.5mM IPTG and expression was performed for either five or twenty hours at 16°C or six hours at 25°C. Cells were harvested by centrifugation, resuspended in Buffer A (50mM Tris-HCl pH 7.5, 50mM NaCl) and broken by sonication (chapter 1.2.2.1). 4µl of total, soluble and insoluble fractions were then separated on a 12.5% SDS-PAGE and analyzed by coomassie staining (Figure 4 ).

Mal pET PV1 VP1(8)-2A C109A cce test.jpg

Figure 4 Test expression of PV1 MBP-His10-VP18-2Apro (C109A) in E. coli BL21 (DE3) Rosetta2 pLysS. 4µL of total (T), soluble (S) and insoluble (I) fractions of induced samples of different expression conditions were separated on 12.5% SDS-PAGE and stained with coomassie blue. Bands that potentially arise from the over-expression of recombinant protein are marked with an arrow.

The fusion protein is strongly over-expressed both at 16°C and 25°C and is found nearly exclusively in the soluble fraction. Further test expressions at 37°C yielded insoluble fusion protein (data not shown). Hence, we decided to upscale the experiment to a one liter culture and express at 25°C for six hours with an IPTG concentration of 0.5mM (Figure 4 ).

Mal pET PV1 VP1(8)-2A C109A cce1.jpg

Figure 4 Expression of PV1 MBP-His10-VP18-2Apro (C109A) in E. coli BL21 (DE3) Rosetta2 pLysS. 3µL of total (T), soluble (S) and insoluble (I) fractions of induced and uninduced samples were separated on 12.5% SDS PAGE and stained with coomassie blue. Bands that potentially arise from the over-expression of recombinant protein are marked with an arrow.

The strong over expression yields an estimated total amount of at least 50mg of recombinant fusion protein per liter of culture, which is predominantly found in the soluble fraction.

Initially we intended to apply a classical purification protocol using fractionated ammonium sulfate precipitation, ion exchange chromatography followed by polishing via size exclusion chromatography. Thus, the soluble fraction was subjected to protein precipitation with 20%, 40% and 60% saturation of ammonium sulfate. Proteins were precipitated for 12 hours at 4°C at each step followed by collection of precipitated protein via centrifugation. Precipitated protein of each step was then redissolved in 8ml Buffer B (50mM Tris-HCl pH 8.5, 50mM NaCl, 5mM DTT, 5% v/v glycerol). All three samples were then dialyzed against Buffer B to remove residual ammonium sulfate. Protein that aggregated during dialyzes was pelletized and dissolved in 8ml Buffer B. In order to investigate which fraction includes the fusion protein, 3µl of each fraction were separated on 12.5% SDS-PAGE and stained with coomassie blue (Figure 4 ).

Mal pET PV1 VP1(8)-2A C109A asp.jpg

Figure 4 Fractionated ammonium sulfate precipitation of PV1 MBP-His10-VP18-2Apro (C109A) containing crude cell extract. 3µl of each fractionation step were separated on 12.5% SDS-PAGE and stained with coomassie blue. (S) Supernatant after removal of residual ammonium sulfate via dialyzes. (I) Redissolved aggregated protein after removal of residual ammonium sulfate via dialyzes. The arrow marks over expressed recombinant protein.

The majority of recombinant protein is found in the fraction that was precipitated by the addition of ammonium sulfate to 60% saturation. Hence, this fraction was submitted to further purification by anion exchange via a MonoQ 10/100 GL column preequilibrated with Buffer B (50mM Tris-HCl pH 8.5, 50mM NaCl, 5mM DTT, 5% v/v glycerol). Bound Proteins were eluted from the column by the use of a NaCl gradient with a maximum NaCl concentration of 1M and fractions á 1ml were collected (Figure 4 ).

Figure 4 Chromatogram of PV1 MBP-His10-VP18-2Apro (C109A) MonoQ run. PV1 MBP-His10-VP18-2Apro (C109A) containing sample was loaded on a MonoQ 10/100 GL using an ÄKTA FPLC system. Bound proteins were eluted by applying a NaCl gradient (green). Peaks in the UV trace (blue) are labeled with the respective retention volumes.

In order to clarify the identity of proteins in the different peaks, fractions with Vr=82-166ml and Vr=188-204ml were analyzed on 12.5% SDS-PAGE (Figure 4 ).

Mal pET PV1 VP1(8)-2A C109A monoq gels.jpg

Figure 4 Analyzes of PV1 MBP-His10-VP18-2Apro (C109A) containing fractions from MonoQ purification by 12.5% SDS-PAGE. 3µl of load (L), 9µl of flow through (FT) and 5µl of fractions were loaded and separated on 12.5% SDS-PAGE and stained with coomassie blue. Arrows indicate the position of recombinant protein.

Interestingly, recombinant fusion protein is detected over a wide range of the NaCl gradient, peaking first at Vr=105ml followed by a broad distribution from Vr=112ml to Vr=166ml. We speculated that this behavior is either due to the bi-globular nature of the fusion protein or due to unspecific binding caused by aggregation of a large fraction of the fusion protein. For this reason, we split the fractions into three pools (Error: Reference source not found).