Cost-effective Purification of Membrane Proteins by ‘Dual-detergent Strategy’

Abstract

Understanding the mechanism of membrane proteins’ function is very critical for biomedical research and drug discovery since membrane proteins constitute ~ 30% of the proteins coded by genomes of lower and higher organisms, and two-thirds of the approved drugs target them. Significant progress has been made not only to engineer several host expression systems for large-scale production of membrane proteins, but also determining their three-dimensional high-resolution structures. Despite these efforts, the study of membrane proteins at the atomic level is still challenging due to poor expression and extraction, low yield of functional protein, the complexity and heterogeneity of source membranes. The structural and spectroscopic studies of any membrane protein require the protein to be extracted from its native membranes into a membrane-mimetic stable environment, that is often achieved by the use of detergents. Unfortunately, there is no ‘magic detergent’ to extract all membrane proteins, and requires a thorough screening with different detergents. However, the membrane protein purification in general, and the detergents used to extract them in particular, are very expensive, which puts a financial constraint for sophisticated membrane protein studies. To overcome this hurdle, ‘dual-detergent strategy’ has recently been developed, and successfully applied to purify various classes of pure, stable and functionally-relevant membrane proteins in a cost-effective manner. In the Basic protocol, we describe the steps of ‘dual-detergent strategy’ to significantly reduce the overall purification cost of a bacterial membrane protein, using the magnesium ion channel MgtE as an example. Further, Support Protocols are provided for selecting the suitable E. coli strain for protein expression, and optimum detergent(s) for membrane protein solubilization.

Basic Protocol: Expression, membrane solubilization and cost-effective purification of MgtE

Support Protocol 1: Selecting suitable E. coli strain for optimum protein expression

Support Protocol 2: Identification of suitable detergent(s) for membrane protein solubilization

Introduction

Membrane proteins are extremely crucial in mediating processes that are fundamental for the thriving of living cells, which include ion transport across membranes, electrical excitability, cell metabolism, communication, and signaling, and are associated with diseases like heart disease, cancer, neurodegenerative diseases, etc. Importantly, ~30% of the proteins coded by the genome of lower and higher organisms are membrane proteins, and ~60% of approved drugs target membrane proteins, of which G-protein coupled receptors and ion channels constitute the largest groups (Bull and Doig, 2015). This highlights the importance of understanding the mechanism of membrane proteins’ function, which is very critical for biomedical research and drug discovery. Despite their importance, the structural characterization of membrane proteins is very expensive and challenging, which is obvious from the fact that only ~3% of the reported crystal structures in the Protein Data Bank (PDB) represent them (Li et al., 2021). Although significant technological advances have greatly facilitated to determine the high-resolution structures of several challenging membrane proteins (Moraes et al., 2014; Autzen et al., 2019; Li et al., 2021), the study of membrane proteins at the atomic level is still quite challenging due to poor expression and extraction, low yield of pure and functional protein, the complexity and heterogeneity of source membranes, and the low success rate of forming well-ordered 3D crystals. The primary issue when working with transmembrane proteins is that removing the membrane lipid exposes the hydrophobic surface area of the protein leading to the formation of non-functional precipitates of the protein. Detergents play an important role in solubilizing the membrane protein by mimicking the lipid environment that are critical for maintaining the folding, stability and the function of membrane protein. However, there is no “magic detergent” that can successfully extract all membrane proteins, and characterizing a new membrane protein, therefore, requires thorough screening with different detergents. The most commonly used detergents for solubilizing the membrane to extract and purify protein are very expensive, which remains the biggest hurdle in obtaining large quantities of pure, stable and functional protein for characterizing the functionally-relevant structural snapshots and their associated dynamics utilizing sophisticated biophysical techniques. To overcome this hurdle, ‘dual-detergent strategy’ has recently been developed, which employs inexpensive detergents for membrane solubilization and subsequently changed to expensive detergents during purification (Tilegenova et al., 2016; Elberson et al., 2017; Chatterjee et al., 2019; Das et al., 2021).

Basic Protocol 1 describes the ‘dual-detergent strategy’ to extract the magnesium channel MgtE from Thermus thermophilus, heterologously expressed in Escherichia coli (E. coli) cells, and purify the stable and functional form of MgtE in a cost-effective manner (Chatterjee et al., 2019; 2021). For this purpose, we describe a standard method to prepare E. coli C41(DE3) competent cells and perform transformation of pET28a-MgtE DNA construct in them. The pET28a-MgtE construct has a N-terminal hexahistidine (6X His) tag. Further, a step-by-step protocol is provided for protein expression, membrane solubilization by the inexpensive detergent (Triton X-100), purification by Immobilized Metal Affinity Chromatography (IMAC) with Ni²⁺ affinity resin using the expensive n-dodecyl-β-maltopyranoside (DDM) detergent in wash and elution buffers. Support Protocol 1 describes the steps to select the E. coli strain for optimized expression of recombinant membrane protein. Support Protocol 2 describes the method for detergent screening to identify the suitable detergent(s) for the solubilization of the desired membrane protein.

Basic Protocol

Expression, Membrane Solubilization and Cost-Effective Purification Of MgtE

This protocol is meant for 1 liter of bacterial culture and gives a yield of ~3 mg of pure, stable and functionally-relevant MgtE. This method can be scaled up or down depending on the amount of protein required for downstream experiments. Although this protocol uses the example of MgtE to highlight the cost-effective purification using the ‘dual-detergent strategy’, this approach has been successfully used to purify several classes of membrane proteins in a cost-effective manner (see Background Information for details). Figure 1 gives the overview of the expression and purification of a membrane protein by ‘dual-detergent strategy’.

Figure 1.

Overview of the expression and purification of membrane proteins by ‘dual-detergent strategy’ as described in the Basic Protocol. The figure was created with Biorender.com.

Materials

• Luria Broth (LB) (VWR Lifescience, cat. no. J104)

• C41(DE3) E. coli competent cells (Lucigen, cat. no. 60422-1)

• pET28a bacterial expression vector (Merck, cat. no. 69864)

• pET28a-MgtE construct (cloned in the laboratory; available upon request)

• Transformation and Storage Solution (TSS) (see recipe)

• Super Optimal broth with Catabolite repression (SOC) medium (see recipe)

• Kanamycin stock solution (Sigma-Aldrich, cat. no. K1377) (see recipe)

• LB Agar (Merck, cat. no. L3027)

• Isopropyl-β-D-thiogalactopyranoside (IPTG) (VWR Lifescience, cat. no. 0487-100G)

• Cell resuspension buffer (see recipe)

• DNase 1 (Merck, cat. no. 04716728001)

• Phenylmethylsulfonyl fluoride (PMSF) (GoldBio, cat. no. P-470)

• Aprotinin (GoldBio, cat. no. A-655)

• Leupeptin (GoldBio, cat. no. L-010)

• Pepstatin A (GoldBio, cat. no. P-020)

• Triton X-100 (Anatrace, cat. no. T1001)

• Ni-NTA affinity resin (Qiagen, cat. no. 30230)

• DDM (Anatrace, cat. no. D310)

• Imidazole (VWR Lifescience, cat. no. 97064-626)

• IMAC wash buffer (see recipe)

• IMAC elution buffer (see recipe)

• Bradford Assay reagent (BioRad, cat. no. 500-0006)

• Gel filtration buffer (see recipe)

• Polypropylene (PP) culture tube (25mm x 150 mm) with cap

• Erlenmeyer flask-100 ml (Sigma-Aldrich, cat.no. Z174483)

• PP centrifuge tube (also called Falcon tube)

• Petri dishes-60mm x 15mm (Sigma-Aldrich, cat. no. P5481)

• Shaker Incubator (Scigenics Biotech, Model: Orbitek LT)

• Cell density meter (VWR, Model: CO 8000 Biowave)

• Eppendorf tube-1.5 ml

• Water bath

• Inoculation loop

• Pipettes-10 ml

• Pyrex 2800 ml Fernbach-style culture flask with baffles (Corning, cat. no. 4423-2XL)

• Big volume Stackable shaker (Scigenics Biotech, Model: LE4676-AH)

• Refrigerated table-top centrifuge to spin down cells (Thermo Scientific Sorvall ST4 Plus, cat. no. 75009911)

• Refrigerated table-top centrifuge rotor to spin down cells (TX-1000 Swinging Bucket Rotor, cat. no. 75003017)

• PP centrifugation bottle (1 liter) (Thermo Scientific, cat. no. 75007300)

• 0.22 μm syringe filter (Merck, SLGV033RS)

• 0.22 μm filter unit (Merck, cat. no. S2GPU02RE)

• Syringes (Merck, cat. no. Z683531)

• Rotator (Labnet, cat. no. H5500)

• Sonicator (Hielscher, cat. no. UP100H)

• Sonicator Probe (Sonotrode MS7)

• Ultracentrifuge tube (Thermo Scientific, cat. no. 1610374)

• Ultracentrifuge (ThermoFisher Scientific, Sorvall WX+)

• Ultracentrifuge rotor (ThermoFisher Scientific, T647.5)

• Gravity flow column (BioRad, cat. no. 738-0019)

• Filtration (Amicon) concentrator (Merck, cat. no. UFC903024)

• Refrigerated table-top centrifuge for concentrating proteins (Thermo Scientific Sorvall ST4 Plus, cat. no. 75009911)

• Refrigerated table-top centrifuge rotor for concentrating proteins (Fiberlite F15-6X 100Y Fixed angle rotor, cat. no. 75003698)

• Microvolume spectrophotometer (DeNovix, DS-11+)

• SEC System (Cytiva, AktaPure 25M)

• SEC Column (Cytiva, Superdex 200 Increase 10/300 GL)

• Additional reagents and equipment for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) of proteins (UNIT 10.1; Gallagher, 2012) that includes 10% SDS-PAGE gels, running buffer and dual-color protein standard (i.e., molecular weight markers) (BioRad, cat. no. 1610374)

Preparation of chemically competent cells

For routine use, it is always better and more economical to freshly prepare competent cells from the company batch.

1. Autoclave PP culture tubes with cap, containing 5 ml of LB. Inoculate C41(DE3) E. coli strain using an inoculation loop and incubate for ~12-16 hours at 37°C in shaking condition (250 rpm) to get saturated culture.

   When using a reusable inoculation loop, make sure to maintain aseptic conditions. The inoculation loop should be heated under the flame of a gas burner until the entire wire becomes orange from heat and also sterilize with 70% alcohol before using. Cool it down to room temperature before using.

   PP culture tubes with cap containing 5 ml LB can be autoclaved in batch and stored for weeks at room temperature (~25°C). However, make sure to check for any contamination in the tube before use.

   Care should be taken to properly maintain the time of incubation, as much longer times results in cultures that have reached saturation for too long and may begin to die.

   Make sure that the E. coli strain is correctly selected based on the membrane protein required to be expressed and purified. Please note that the E, coli strain C41 (DE3), used in this protocol, doesn’t require any selection antibiotic. However, care should be taken while using any other E. coli strain and, if required, the respective selection antibiotic at recommended concentration must be added to the LB before inoculation (see Internet Resources).

   See Support Protocol 1 for the selection of suitable E. coli strain to attain the optimized expression of the desired transmembrane protein.

2. Transfer 250 μl of the saturated C41(DE3) culture in an autoclaved 100 ml Erlenmeyer flask containing 25 ml LB (1:100 dilution). Incubate at 37°C with uniform rotation at 250 rpm.

   Depending upon the requirement, the volume of LB can be increased or decreased. However, always make sure to maintain the inoculation ratio at 1:100. As mentioned above, depending on the E. coli strain used, add the respective antibiotic if required.

3. Grow cells for ~2-3 hr until the optical density of the culture at 600 nm (OD₆₀₀) reaches 0.4-0.5. Measure the OD₆₀₀ using the cell density meter.

   Make sure that the OD₆₀₀ does not exceed 0.5.

4. Transfer the 25 ml bacterial culture into a pre-chilled 50 ml PP centrifuge tube and keep it on ice for 20 minutes.

5. Centrifuge at 1500 x g for 10 min at 4°C using the table-top refrigerated centrifuge and discard the supernatant.

6. Carefully resuspend cells in 2.5 ml (1:10 dilution) of pre-chilled TSS and mix gently using a pipette.

   Make sure that the TSS is ice cold at this stage. Sterile TSS can be prepared beforehand and stored at -20°C in aliquots. However, working aliquots should be stored at 4°C.

7. Dispense 100 μl of TSS-resuspended C41(DE3) competent cell in each of the pre-chilled 1.5 ml Eppendorf tubes.

   Make sure that the cells are uniformly mixed with the TSS. Cells should be handled with utmost care and must always be kept on ice as the cells become very fragile during TSS treatment.

8. The C41(DE3) competent cells can immediately be used for the next step of the protocol or stored at -80°C freezer.

Transformation of pET28a-MgtE construct into the C41(DE3) competent cells

• 9.Take 100 μl of C41(DE3) competent cells on ice, and thaw it if previously stored at - 80°C.

  Freshly prepared competent cells are highly recommended for membrane protein expression. Do not use competent cells older than a week for protein expression.

• 10.Add 1 μl of pET28a-MgtE construct to C41(DE3) competent cells. Mix gently by pipetting up and down, and incubate the tube on ice for 10-20 min.

  Instead of pipetting up and down, one can also flick the tube 4-5 times to mix the DNA construct to the competent cell. Do not vortex.

  This protocol uses the MgtE gene from Thermus thermophilus that is cloned into a pET28a vector. Use a suitable vector to clone the gene of interest. For an optimized selection of vectors for cloning, see Critical Parameters.

• 11.Place the tube at 42°C water bath for 60 s to heat shock cells, and incubate the tube on ice for 2-5 min. Do not mix.

  Do not exceed the time of heat shock. Also, the water bath should be turned on ahead of time so that the suitable temperature is reached before this step.

• 12.Add 400 μl of SOC media to the mixture.

  LB can also be used if SOC is not available at this stage. Make sure it is sterile and at room temperature before using.

• 13.Incubate the tube at 37°C for 1 hr with constant shaking at 250 rpm.
• 14.Spread 100-200μl of the transformed cells on to LB Agar plate containing 50μg/ml kanamycin antibiotic.

  pET28a vector has the kanamycin-resistant gene, hence we use kanamycin as the selection antibiotic in the LB Agar. It is best to make kanamycin plates every 2-4 weeks as the plates start to lose sensitivity after 2 weeks. Store plates at 4°C when not required.

  For other vectors, make sure to add the respective antibiotic at recommended concentration depending on the antibiotic-resistant gene present in them.

• 15.Incubate for ~14-16 hours at 37°C, and no shaking is required. After this period, plate should contain numerous single colonies that harbor pET28a-MgtE construct.

  The plates must be kept in an inverted position inside the incubator to prevent water condensation from accumulating on the LB Agar which might disturb or compromise the growing bacterial colonies.

Induction of bacterial culture for large scale protein expression

• 16.Pick a single colony from the LB Agar plate with the help of a sterilized pipette tip or toothpick and add it to the autoclaved 75 or 100 ml Erlenmeyer flask containing 20 ml LB and add 50μg/ml kanamycin.
• 17.Incubate for ~12-16 hours at 37°C (generally overnight) with uniform rotation at 250 rpm.

  Do not exceed the time of incubation as much longer times results in cultures that have reached saturation for too long and may begin to die.

• 18.Prepare 1 liter of sterile LB in a Fernbach culture flask.

  Sterile LB should be previously prepared and stored at room temperature. Check for any contamination before proceeding to the next step.

• 19.From a stock of 50 mg/ml kanamycin, add 1 ml of the same (50 μg/ml) into 1 liter LB.
• 20.Add 10 ml (1:100 dilution) of the overnight grown culture to the 1 liter LB containing kanamycin.
• 21.Incubate at 37°C at 250 rpm in the big volume shaker until OD₆₀₀ reaches 0.5. This will take approximately ~2-3 hours.

  Turn on the power of the big volume shaker ahead of time and set the temperature so that there is ample time for the shaker interior to reach the required temperature before the start of experiment.

• 22.At OD₆₀₀=0.5, induce the expression of the recombinant MgtE protein by adding 0.5 mM of IPTG (i.e., add 500 μl from 1 M IPTG stock) into the culture.

  Expression of MgtE requires an induction with 0.5 mM IPTG. However, the optimum concentration of IPTG required for expression might vary from membrane protein to protein. See Critical Parameters to standardize the IPTG concentration for optimal membrane protein expression.

• 23.Post-induction, grow cells at 20°C for ~20 hours, uniformly shaking at 250 rpm.

  Post-induction temperature for the expression of membrane proteins may vary (see Critical Parameters).

Harvest cells

• 24.Transfer the culture to 1 liter PP centrifugation bottle and pellet bacterial cells by centrifugation at 1500 x g (i.e., ~3700 rpm in a TX-1000 swinging bucket rotor) for 40 minutes at 4°C.

  Turn on the power and set the temperature of the centrifuge at 4°C ahead of time so that there is ample time for the centrifuge to reach the required temperature before the start of experiment.

  Alternatively, Beckman Coulter Avanti J-20 floor super speed centrifuge (rotor: J- LITE JLA-1.8000) can be used at 6500 rpm for 10 minutes to harvest cells in lesser time.

• 25.Discard the supernatant and resuspend the cell pellet in 30 ml of cell resuspension buffer.

  The recipe of cell resuspension buffer described in this protocol is specific for MgtE. Depending on the protein of interest, the composition of resuspension buffer might change.

  It is extremely critical to maintain the appropriate pH of the buffer solution.

• 26.Use resuspended cells immediately for the following step described in the protocol or store at -80°C in 50 ml Falcon tubes.

  The resuspended cells can be stored at -80°C for few weeks.

Cell lysis

• 27.Make up the volume of the resuspended cell suspension to 40 ml, and add 2 mM MgCl₂ and 50μg/ml DNaseI. Also, add protease inhibitors as follows: 1mM PMSF, 1 μg/ml aprotinin, 10 μM leupeptin and 1 μM pepstatin A and place the 50 ml falcon tube on a rotator for ~10-15 minutes for homogenous mixing.

  It is extremely crucial to add protease inhibitors at this stage as they help in protecting target protein of interest from protease degradation which might occur during cell lysis.

  Alternatively, one tablet of Complete Protease Inhibitor Cocktail (Merck, cat. no. 04693116001) can be used.

  If cells have been stored at -80°C before this step, make sure to thaw the cells before adding the protease inhibitors and mix well by rotating.

• 28.Lyse the cells by sonicating in ice using a 7 mm probe, for 20 seconds on:20 seconds off cycle; 4 times at 40 % amplitude.

  Do not sonicate for a long period as that might lead to protein aggregation.

  Cells can also be lysed by using a high-pressure homogenizer (Microfluidizer LM20 or Avestin C3) at least 2 times at 10,000 psi.

Isolation and collection of cell membranes

• 29.The cell lysate is transferred to the ultracentrifuge tube. Keep the tubes on ice at all times.

  Take care to correctly balance the ultracentrifuge tube before starting the ultracentrifugation process. Also, make sure that the ultracentrifuge tubes must be approximately 75% full so that the tubes do not collapse in the rotor during high-speed ultracentrifugation.

• 30.Ultracentrifuge the homogenized cell suspension at 100,000 x g (37000 rpm in a T647.5 rotor) at 4°C for 45 minutes.

  Turn on the ultracentrifuge machine at least 1 hour prior to this step so that there is ample time for it to cool down to 4°C. Further, always store the rotor at 4°C so that it is ready for immediate use during this step.

  It is a good practice to load the filled ultracentrifuge tubes symmetrically into the rotor. Further, take good care and clean off any condensation from the rotor which might otherwise interfere with the vacuum apparatus of the ultracentrifuge.

• 31.Discard the supernatant and collect the pellet, which contains the membrane fraction with the expressed MgtE protein.
• 32.Add 10 ml of cell resuspension buffer to the membrane pellet along with protease inhibitors in the following concentration: 1mM PMSF, 1 μg/ml aprotinin, 10 μM leupeptin, and 1 μM pepstatin A.

  Alternatively, one tablet of Complete Protease Inhibitor Cocktail (Merck, cat. no. 04693116001) can be used.

• 33.Gently suspend the membrane pellet using a clean brush on ice.

  The membrane pellet generally sticks tightly to the walls of the ultracentrifuge tube. Do not rush through the process to avoid clumps. Do not pipette up and down vigorously for resuspending the membrane pellet. Harsh pipetting or excessive mixing might destabilize of the protein.

• 34.After complete resuspension of the pellet, transfer it from the ultracentrifuge tube to a 50 ml Falcon tube.

  Ultracentrifuge tubes must be handled with utmost care. After this step, hand-wash the tubes gently with ddH₂O to prevent scratches. Do not use detergents or any organic solvents like methanol, ethanol, etc. to clean the tubes.

  If required, this resuspended membrane fraction with the expressed protein can be stored at -80°C for a few days before proceeding further.

Membrane solubilization of MgtE using inexpensive detergent, Triton X-100

• 35.From a stock of 200 mM Triton X-100, add 2 ml (i.e., 10 mM, which is ~50X CMC of Triton X-100) to the ~10 ml of resuspended membrane fraction.

  The solubilization step of membrane protein purification typically requires the detergent concentration to be above the critical micelle concentration (CMC), which is the lowest concentration above which the detergent monomers cluster to form micelles. Higher concentration of specialized detergents (~5 to more than 100X CMC) is, therefore, needed to extract the membrane protein from the source membrane to create water-soluble protein-detergent complexes. See critical parameter for discussion on CMC of detergent of choice during membrane protein solubilization and extraction.

  Recently, Triton X-100 has successfully been used in solubilizing several classes of membrane proteins. However, it is possible that Triton X-100 might not work for all proteins (Das et al., 2021), in which case, several detergents have to be screened to identify the suitable detergent for solubilizing the membrane protein of interest (See Support Protocol 2).

  Detergent stock should be previously prepared in Milli-Q water and filtered using 0.22 μm filter before starting this step. Detergent stock solution can be stored at room temperature.

• 36.Make up the volume of this solution to 40 ml using cell resuspension buffer. Incubate for 2 hours at 4°C under rotating conditions.

  MgtE doesn’t contain any cysteine residue, hence no β-ME is added to the solution. In case of any other cysteine-containing proteins, make sure to add 3mM β-ME before making up the volume to 40 ml. β-ME being a reducing agent acts on the disulfide bonds, the presence of which might not give a good understanding of the oligomeric state of the extracted protein.

• 37.Transfer the 40 ml detergent-solubilized fraction of MgtE into the ultracentrifuge tube and centrifuge the same at 100,000 x g (37000 rpm in a T647.5 rotor) at 4°C for 45 minutes.

  Follow the annotations given in Step 29.

• 38.Carefully collect the supernatant, which contains the Triton X-100 solubilized MgtE, in a fresh 50 ml Falcon tube. Discard the cell pellet.

  Irrespective of the membrane protein being extracted, it should be remembered that the supernatant will have the detergent-solubilized protein after this ultracentrifugation step. Do not throw away the supernatant by mistake.

  Wash the ultracentrifuge tubes thoroughly as discussed in Step 34. Wipe the rotor with ethanol after the completion of this step. Thoroughly wipe off any condensed water particles inside the ultracentrifuge machine before shutting it off.

Purification of MgtE by IMAC and its quantification

• 39.Add 4 ml of Ni²⁺ -NTA affinity resin to the supernatant containing the Triton X-100 solubilized MgtE.

  Note that the Ni²⁺ resin comes in ethanol from the manufacturer. Follow the manufacturer’s protocol to wash the resin properly and keep an aliquot of 50% washed resin in a 50 ml falcon tube (i.e, when settled the packed resin should be half of the total volume).

• 40.Mix well for ~15 minutes at room temperature on a rotator.
• 41.Pour the slurry into the gravity flow chromatography column and allow the flow-through to pass freely.

  MgtE contains N-terminal 6X His-tag, which has a high affinity for Ni²⁺ ions, and binds strongly to the IMAC column. Most other undesirable proteins present in the slurry will not bind to the column or will bind loosely in a non-specific manner.

  In the subsequent steps, all buffers (IMAC wash and elution) should contain 1 mM DDM detergent (~ 5X CMC).

• 42.Add 40 ml (20X column volume) IMAC wash buffer to remove any non-specifically bound proteins, and let it flow through completely.

  Add slowly by the side of the walls of the gravity flow column without disturbing the packed Ni²⁺ resin.

  Low concentration of imidazole (5-50 mM) is generally used in wash buffer to remove other weakly bound proteins (containing 2-3 histidine residues) from Ni²⁺ resin. The His-tag containing recombinant protein will remain bound to the resin after washing.

• 43.Elute MgtE with 10 ml of IMAC elution buffer into a fresh 15 ml falcon tube.

  Add slowly by the side of the walls of the gravity flow column without disturbing the packed Ni²⁺ resin.

  High concentration of imidazole (250-500 mM) is used in the elution buffer to completely elute the His-tag containing recombinant protein from the Ni²⁺ column.

  Both Co²⁺ and Ni²⁺ resins can be used to purify recombinant His-tagged proteins. A quick visual check for the presence of purified protein can be done by adding 10 μl of eluted protein to a mixture of 100 μl of Bradford reagent (brown color) and 390 μl of water. Depending on the amount of protein, the solution turns blue.

• 44.Add 10 ml of eluted protein in a 50 KDa cut off Amicon concentrator.

  MgtE is a homodimer whose molecular weight is 100 KDa and, therefore, a 50 KDa cut off concentrator suffices our need. Take concentrator based on the molecular weight of the protein of interest. Basically, concentrator should have a molecular weight cut-off higher than the functional oligomeric weight of the protein of interest.

• 45.Concentrate protein by centrifuging at 2000 x g at 4°C until the volume is as low as 1 ml.

  Do not concentrate for a long period of time. Give 5-10 min run initially followed by mixing with the pipette, and concentrate again until the volume is ~1 ml. Depending on the yield of eluted protein, the time of concentration will vary.

• 46.Check for the protein yield in a microvolume spectrophotometer.

  Proper determination of the protein concentration and yield can be done using BCA assay, which uses bicinchoninic acid and is compatible with detergents. Follow the manufacturer’s protocol for performing this assay.

• 47.At this point, the eluate can be stored at 4°C.

  Avoidfreeze-thaw!

Checking the purity and homogeneity of the eluted protein

• 48.Take 20-30 μg of eluted MgtE, add 1X Laemmli buffer, mix well and perform SDS-PAGE.

  MgtE migrates as a monomer in the SDS-PAGE and gives predominantly a single band at ~50 KDa.

• 49.For SEC, concentrate the protein to 500 μl using a table-top centrifuge at 4°C.
• 50.Rinse the Akta System (as per the system wash procedure) and the loading loop (500 μl volume) with filtered H₂O using the manual set-up. Set flow rate at 0.5 ml/minute; volume at 50 ml; alarm pressure of system at 0.25 MPa.

  This setting is for Superdex200 Increase 10/300 GL gel filtration column. Set parameters as per the supplier’s recommendation for the column in use.

• 51.Rinse the Akta system and the loading loop with 500 μl gel filtration buffer containing DDM.
• 52.Pre-equilibrate the column with 50 ml (i.e., 2X column volume) of gel filtration buffer.
• 53.Filter the protein solution using a syringe filter and inject 500 μl protein into the loading loop.

  Alternatively, spin the protein for 2 min at 8000 x g using a table-top centrifuge to remove any small aggregates, and inject only the supernatant.

  Note that the volume of the protein sample to be injected should be dependent on the loop used. Do not inject more than 500 μl in the loop to avoid protein loss.

• 54.Run 25 ml of gel filtration buffer through the column. Collect l ml fractions in the automated tube collector.

  The UV lamp must be on during this step.

• 55.Collect and pool the fractions from the single, homogeneous peak (MgtE homodimer), which elutes at ~11.6 ml elution volume from this column.

  MgtE elutes as a dimer in SEC without any presence of non-specific aggregates.

  Generally, non-specific aggregates will come in the void volume (~6-8 ml of elution) in Superdex 200 Increase 10/300 GL column.

• 56.Concentrate the pooled fractions to ~1 ml and check for protein yield using a microvolume spectrophotometer.
• 57.Rinse the Akta System with filtered Milli-Q water using the system wash procedure. Then, rinse the gel filtration column with at least 50-100 ml (i.e., 2-4X column volume) filtered 0.1% sodium azide (NaN₃) in Milli-Q water. Follow it with a 20% EtOH wash step, and store the column and system in 20% EtOH.

  For a long-term storage of the column, use 20% EtOH.

Support Protocol 1

Selecting Suitable E. coli Strain for Optimum Protein Expression

All membrane proteins might not express in the same way in a particular strain of E. coli. For optimizing the expression of recombinant membrane protein of interest, several E. coli strains should be screened to find the suitable expression strain. The correct choice of cell can help in expressing membrane proteins not only in increased quantity but also in a stable form. This protocol describes a ‘small-scale expression test’ to select the best E. coli strain for high expression the membrane protein of interest. See Critical Parameters for more information.

Additional Materials (also see Basic Protocol)

• Several E. coli strains can be tested. The following are the commonly used ones: BL21(DE3) (Invitrogen, cat. no. C600003)

  • BL21(DE3)pLysS (Invitrogen, cat. no. C606010)

  • BL21-Gold (Agilent, cat. no. 230130)

  • C41(DE3) (Lucigen, cat. no. 60422-1)

  • C43(DE3) (Lucigen, cat. no. 60446-1)

  • XL1-Blue (Agilent, cat. no. 200249)

  • XL10-Gold (Agilent, cat. no. 200314)

1. Prepare chemically competent cells as described in Basic Protocol (Steps 1-8) for the above-mentioned E. coli strains using respective selection antibiotic as per the manufacturer’s guide.

2. Transform the plasmid containing the gene of interest in respective competent cells. (Follow Steps 9-13 of the Basic Protocol).

   In principle, for a novel protein expression, one should clone the gene in several E. coli expression vectors, and try expressing each of the constructs in above-mentioned E. coli strains to find the best expression vector/cell pair. Starting with pET expression vectors is a good option.

3. Add the 500 μl of transformed culture to 5 ml of sterile LB in a PP culture tube with cap. Add respective selection antibiotic for the plasmid, and grow cells for ~12-16 hr at 37°C with uniform rotation at 250 rpm.

4. Inoculate 20 ml of LB medium with 200 μl of the overnight culture (1:100 dilution) and add the respective selection antibiotic at appropriate concentration.

5. Grow cells at 37°C till OD₆₀₀ reaches 0.8 (will typically take ~2.5-3 hr), and take 1 ml of culture in a 1.5 ml Eppendorf tube (‘pre-induction’ sample).

6. Induce the cells with 1 mM IPTG and grow at 37°C for 4 hr.

   Just after adding IPTG, centrifuge the ‘pre-induction’ sample for 2 min at 8000 x g and discard the supernatant. Then, add 100 μl of Milli-Q water and 1X Laemmli buffer, mix well with a pipette and store at -80°C.

7. Check OD₆₀₀ for each strain and collect cells (normalized to induction OD₆₀₀ 0.8) in a 1.5 ml Eppendorf tube (‘post-induction’ sample).

   For instance, if post-induction OD₆₀₀ is 1.6, take 0.5 ml of the cells to ensure that the number of cells are approximately the same in both ‘pre- and post-induction’ samples.

   Centrifuge the ‘post-induction’ sample for 2 min at 8000 x g and discard the supernatant. Then, add 100 μl of Milli-Q water and 1X Laemmli buffer, mix well with a pipette and store at -80°C.

8. Perform western blot analysis to check for the expression of His-tagged recombinant protein in different strains of E. coli. Select the E. coli strain that best expressed the desired recombinant protein.

   Since nucleic acids from the lysed cells float, this will create problems in loading the sample in a polyacrylamide gel for western blots. This problem can be mitigated by mixing the sample few times using a syringe with ~25 GA needle prior to loading, which shears the nucleic acids resulting in a more transparent sample.

Support Protocol 2

Identification of Suitable Detergent(s) for Membrane Protein Solubilization

Every membrane protein is unique, and so are the optimal conditions for solubilizing them from the cell membrane. The detergent Triton X-100 used in the Basic Protocol 1 works well for solubilizing not only MgtE but many other bacterial membrane proteins in their functional form (see Background Information). However, this might not be the case for all membrane proteins. Therefore, small scale screening with several detergents (‘solubilization test’) is necessary to find the optimum one that extracts the membrane protein in the stable and functional form from the native membrane. This protocol provides a quick method to test several detergents for solubilizing the membrane protein of interest once the proper vector/cell pair is identified.

Additional Materials (also see Basic Protocol)

• Below are some of the detergents that are widely used in membrane protein purification (Stetsenko and Guskov, 2017). Also see Internet Resources for various detergent properties.

  • Anzergent 3-14 (Anatrace, cat. no. AZ314)

  • n-dodecyl-β-D-maltopyranoside, DDM (Anatrace, cat. no. D310)

  • n-decyl-β-D-maltopyranoside, DM (Anatrace, cat. no. D322)

  • n-dodecyl-N,N-dimethylamine-N-oxide, LDAO (Anatrace, cat. no. D360)

  • n-octyl- β-D-glucopyranoside, OG (Anatrace, cat. no. 0311)

  • n-octyl- β-D-maltopyranoside, OM (Anatrace, cat. no. O310)

  • Octaethylene glycol monododecyl ether, C12E8 (Anatrace, APO128)

  • TFT-80.2 fixed angle rotor (ThermoFisher Scientific, cat. no. 54456)

1. Prepare a stock solution of detergents in Milli-Q water and filter sterilize it using a 0.22 μm syringe filter.

2. Follow Steps 2-6 from the Support Protocol 1.

   No need to collect cells from pre-induction sample for this protocol.

3. Collect post-induction cells normalized to pre-induction OD 0.8 in each tube depending on the number of detergents to be tested.

4. Centrifuge the ‘post-induction’ sample for 2 min at 8000 x g and discard the supernatant. Resuspend each pellet with 250 μl of buffer.

   The choice of buffer depends on the type and nature of the protein.

5. Add 10X CMC of each detergent separately to respective tubes and make up to the volume to 1 ml with buffer and incubate for 2 hr at 4°C under rotating conditions.

   Information about the storage and use of the detergents is provided on the Anatrace website (See Internet Resources).

   SDS, which solubilizes many membrane proteins, is used as a positive control to assess the membrane solubilization of proteins. Generally, SDS is not used in membrane protein purification as it denatures them. The resuspended cells, without any added detergent, is used as the negative control.

6. Transfer each sample to mini-ultracentrifuge tubes, and centrifuge for 15 min at 100000 x g at 4°C using a small volume TFT-80.2 fixed angle rotor.

7. Carefully collect the supernatant (i.e., detergent-solubilized fraction) from each tube to a fresh 1.5 ml Eppendorf tube and add 1X Laemmli sample buffer.

   Although the detergent-solubilized fraction should ideally contain all the expressed protein, the efficiency of membrane solubilization by detergents, in reality, is not always 100%. So, the pellet (i.e., the detergent-resistant membrane fraction) can also be resuspended in 200 μl of buffer with 1X Laemmli sample buffer.

8. Store samples at -80°C until ready to use.

9. Perform western blot analysis to compare the amount of protein in the supernatant and the pellet.

   The criteria for selecting the best detergent for solubilizing membrane protein from western blot analysis are:

   • a) The detergent that gives the highest protein yield from the supernatant. The intensity of the western blot band is evaluated to determine the same.
   • b) The detergent that yields the least amount of protein from the cell pellet. This would indicate that the detergent has sufficiently solubilized most of the desired membrane protein from the membrane fraction.
   • c) It should be noted that the best detergent for solubilizing the membrane protein might not be the best for finally attaining the protein in a stable and functional oligomeric form. Perform SEC and other assays that assess the functional state of the membrane protein of interest.

Reagents and Solutions

Water used in the preparation of all the below-mentioned reagents and solution is Milli-Q with a resistivity of 18.2 MΩ cm unless otherwise mentioned.

All buffers and stock solutions are filtered using a 0.22 μm filter unit before performing experiments.

Store reagents and buffers in autoclaved containers.

Aprotinin stock solution, 10 mg/ml

Dissolve 25 mg of Aprotinin in 2.5 ml H₂O

Mix until completely dissolved and prepare 0.1 ml aliquots

Store at -20°C for up to 1 month

Cell Resuspension buffer

20 mM HEPES buffer, pH 7.0

150 mM NaCl

DDM stock solution, 200 mM

Dissolve 5.106 g of DDM in 50 ml H₂O.

Mix by stirring in a beaker using the magnetic bead. Do not vortex to mix.

Sterilize using 0.22 μm filter unit and store at room temperature.

Gel filtration buffer

20 mM HEPES buffer, pH 7.0

150 mM NaCl

1 mM DDM

IMAC wash buffer

20 mM HEPES buffer, pH 7.0

150 mM NaCl

1 mM DDM

50 mM Imidazole

IMAC elution buffer

20 mM HEPES buffer, pH 7.0

150 mM NaCl

1 mM DDM

300 mM Imidazole

Imidazole stock solution, 1M

Dissolve 34 g of imidazole in 500 ml H₂O

Autoclave

Store at 4°C for up to 2 years. Light sensitive

IPTG stock solution, 1M

Dissolve 11.9 g IPTG in 50 ml H₂O

Filter sterilize using 0.22 μm syringe filter, prepare 10 ml aliquots in 15 ml PP centrifuge tubes Store at -20°C for up to 2 months

Working aliquot can be kept at 4°C

Kanamycin stock solution, 1000X

Prepare 50 mg/ml kanamycin in 50 ml H₂O

Filter sterilize using 0.22 μm syringe filter, prepare 10 ml aliquots in 15 ml PP centrifuge tubes Store at -20°C for up to 6 months. Light sensitive

Working aliquot can be kept at 4°C.

Kanamycin containing LB Agar plate

Dissolve 2 g of LB Agar powder in 50 ml of H₂O

Autoclave

Cool it down to ~50°C and add 50 μl of kanamycin from stock (see recipe for kanamycin stock preparation).

Swirl well to ensure even distribution of the antibiotic.

Pour LB Agar (~15 ml) in each of the sterile Petri plates and let it completely solidify

Parafilm and store at 4°C for 2-4 weeks

Pour the plates under sterile conditions in a laminar airflow setup. Make sure to clean and sterilize the laminar hood with 70% ethanol and UV radiation for at least 5 minutes before making the LB Agar plates. Keep the Bunsen burner on. For a plate pouring video, see Internet Resources.

Leupeptin stock solution, 10 mM

Dissolve 5 mg of Leupeptin hemisulfate in 1.05 ml H₂O

Mix until completely dissolved.

Store at -20°C for up to 1 month.

Luria Broth (LB)

Dissolve 25 g of LB powder in 1 liter of H₂O

Autoclave

Store at room temperature up to 1 year.

Pepstatin A stock solution, 1mM

Dissolve 5 mg of Pepstatin A in 7.30 ml sterile filtered DMSO Mix until completely dissolved and prepare 1 ml aliquots

Store at -20°C for up to 6 months

PMSF stock solution, 1M

Dissolve 17.42 g PMSF in 100 ml anhydrous isopropanol. Prepare 10 ml aliquots in 15 ml PP centrifuge tubes

Store at 20°C for up to 6 months

Note that PMSF is highly unstable in the presence of H₂O due to its susceptibility to hydrolysis of the fluoride moiety. The rate of inactivation increases with increasing pH and at higher temperatures.

Triton X-100 stock solution, 200 mM

Add 6.46 ml of Triton X-100 (Anatrace, cat. no. T1001) in 43.54 ml H₂O. Sterilize using 0.22 μm filter unit and store at 4°C.

Commentary

Background Information

The bacterium E. coli has been widely used as a host for the overexpression of recombinant proteins (soluble and membrane) due to reasons of technical ease and low cost (Rosano and Ceccarelli, 2014; Jia and Jeon, 2016). The overexpression and purification of membrane proteins, in particular, is highly challenging, and therefore, several excellent articles and protocols are available in the literature that deals with the efficient screening and optimization for large scale production of stably folded functional membrane proteins (Wagner et al., 2006; 2008; Junge et al., 2008; Geerstma and Poolman, 2010; Ho and Poolman, 2015; Angius et al., 2016; Jeffery, 2016; Pandey et al., 2016; Kuipers et al., 2017; Marino et al., 2017; Morra et al., 2017; Mathieu et al., 2019), and the quality control of overexpressed membrane proteins (Geertsma et al. 2008a). The structural and spectroscopic studies of membrane proteins generally require large amounts of protein, and the overexpressed protein need to be extracted from the membrane of the host expression system into a membrane-mimetic stable environment (Arachea et al., 2012), which is generally achieved by the use of detergents, and this remains the ‘bottleneck’ in membrane protein purification. Detergents are water-soluble amphiphiles (Linke, 2009) and it forms thermodynamically stable, non-covalent organized molecular aggregates called micelles above a critical concentration, known as the critical micelle concentration (CMC). The membrane solubilization step (Helenius and Simons, 1975; Litchenberg et al., 2013) of membrane protein purification typically requires high concentration of specialized, expensive detergents (~5-100X CMC), compared to the concentration of detergents required in the downstream processes. Despite the commercial availability of a large number of detergents, there is no ‘universal detergent’ that can be used to extract the different classes of membrane proteins. This is obvious from the fact that the extraction efficiency of the expressed membrane proteins (Arachea et al., 2012), and the retention of the functional properties can vary dramatically with different detergents (Infed et al., 2011), clearly highlighting the importance of detergent screening to identify the best detergent for the proper extraction of membrane proteins in a stable and functional form. This makes the membrane protein purification process very expensive and most of the purification cost lies in the choice of detergents suitable for extraction, purification, and crystallization. Any improvement in the extraction of pure, stable and functional membrane proteins in a cost-effective manner is highly useful to perform sophisticated structural (Prive, 2007; Moraes et al., 2014) and site-directed spectroscopic studies (Raghuraman et al., 2014; Raghuraman et al., 2019; Brahma and Raghuraman, 2021; 2022; Brahma et al., 2022).

Alkyl maltosides such as DDM and DM, and glycosides like OG are the most commonly used detergents for membrane protein purification and crystallization, which are very expensive. Triton X-100 is one of the oldest, classical nonionic polyoxyethylene detergent with a low CMC of ~0.3 mM (Raghuraman et al., 2004). It is widely used in membrane studies to explore the detergent-resistant membranes and functional membrane domains in cell biology (Mukherjee and Maxfield, 2004; Litchenberg et al., 2005), and also mediating membrane reconstitution of proteins (Geertsma et al., 2008b). In terms of membrane protein purification, Triton X-100 is significantly underrepresented (~2% compared to ~40% in case of DDM) due to its chemical heterogeneity and the interference with the protein quantitation due to its strong absorption in the UV region of the electromagnetic spectrum (Stetsenko and Guskov, 2017).

‘Dual-detergent strategy’ described in this protocol uses inexpensive detergents such as Triton X-100 for membrane solubilization and extraction, and exchanging with expensive detergents like DDM during purification steps, thereby eliminating the interference in protein quantification. Recently, this strategy has been successfully used to extract various classes of membrane proteins (Figure 2) in a pure and stable form, which include pH-gated K⁺ channel, KcsA (Tilegenova et al., 2016), ligand-gated ion channel, ELIC (Elberson et al., 2017); Mg²⁺ channel, MgtE (Chatterjee et al. 2019; 2021) and isolated voltage-sensing domain (VSD) of the voltage-gated K⁺ (K_v) channel, KvAP (Das et al., 2021). Interestingly, in case of ELIC, the extraction by this strategy increased the yield by ~10 fold compared to extraction by DDM (Elberson et al., 2017). The above examples of successful protein purification with Triton-X 100 in a stable form reveal that the success of the ‘dual-detergent strategy’ does not depend either on the oligomeric nature of the proteins or the presence of extracellular or cytoplasmic domains (see Figure 2). The most important advantage of using Triton X-100 during extraction is that the overall cost of purification is significantly lowered compared to the usage of conventional detergents. However, it appears that the ‘dual-detergent strategy’ will be useful for extracting proteins that are both DDM- and DM-extractable, but will be ineffective if the protein is only DM-extractable as has been shown in the case of inward rectifying K⁺ channel, KirBac1.1 (Das et al., 2021). Considering that the cost of Triton X-100 is ~200 fold cheaper than the widely used expensive DDM, and the successes of ‘dual-detergent strategy’ with bacterial membrane proteins, it will be worthwhile to include Triton X-100 for routine detergent screening for membrane protein extraction.

Figure 2.

Various classes of bacterial membrane proteins successfully purified by ‘dual-detergent strategy’. The grey slab represents the membrane. Adapted and modified from Das et al., 2021.

Critical Parameters

Non-optimal conditions during protein expression, membrane solubilization and purification, storage and handling can irreversibly alter the stability and structure of the protein with significant loss of function. The following parameters should therefore be taken into consideration.

Choice of expression system

Several E. coli strains like XL-1 Blue, XL-Gold, BL21(DE3), C41(DE3) and C43(DE3) have been used successfully for the overexpression of several bacterial membrane proteins. Initial screening with the ‘Walker strains’ (C41 and C43), which are derived from BL21 (Miroux and Walker, 1996), and the desired gene cloned into a pET vector system is a good start to overexpress desired membrane protein since overexpression of many membrane proteins in these strains is hardly toxic. Significant improvements have been made to fine tune and engineer the E. coli cells for overexpression of functional membrane proteins (Wagner et al. 2008; Ho and Poolman, 2015; Morra et al., 2017; Kuipers et al., 2017). It has been shown recently that the function of the purified multidrug ABC transporter proteins is highly dependent upon the E. coli strain in which they are expressed (Mathieu et al., 2019). Basically, it is important to screen for the best vector/cell pair for the expression of desired protein. For some proteins, higher yields are obtained by changing to different expression host such as Lactococcus lactis (Geertsma and Poolman, 2010).

Optimizing post-induction conditions

Obtaining the pure, stable and functional form of membrane proteins critically depends on several factors, the reasoning for some of them has not yet been adequately understood. This highlights the importance of optimizing conditions for overexpression and purification. Post-induction temperature is very important and should be optimized for protein expression. For example, as can be seen from the Basic Protocol, the optimum post-induction temperature is 20°C for good expression of MgtE (Chatterjee et al., 2019). Importantly, MgtE expression is significantly decreased upon increasing the post-induction temperature (Figure 3, Left). Further, higher post-induction temperature also considerably reduces the overall yield of MgtE (Figure 3, Center), and affects the homogeneity of the purified protein (Figure 3, Right). However, using the post-induction temperature of 20°C might not be relevant for all membrane proteins. Growing cells at post-induction temperatures of 30°C and/or 37°C for 4 hours should also be tried for a new protein to ensure the best expression. Further, induction OD and the inducer (IPTG) concentration are also important factors. Inducing the protein expression using 1 mM IPTG at OD of 0.8 used in the Support Protocol 1 is a good starting point. However, it is important to try different induction OD (like 0.5 and 1.0) and different IPTG concentrations (0.1 to 1 mM). The best way to try out these factors are by employing small-scale expression test (see Support Protocol 1).

Figure 3.

Importance of post-induction temperature in MgtE expression and purification. Western blot (Left panel) of MgtE expressed in C41 E. coli cells before (pre-induction) and after IPTG induction at different post-induction temperatures as indicated. Yield of purified MgtE (Center panel) obtained per liter of culture grown. SEC profiles (Right panel) of purified MgtE, expressed at different temperature, in DDM micelles. Adapted and modified from Chatterjee et al., 2019.

Choice of detergents

There are several classes of specialized detergents whose structures and CMC values vary enormously (see Internet Resources). As mentioned earlier, there is no ‘magic’ detergent that can extract all types of membrane proteins in a stable and functional form. So, membrane solubilization test described in Support Protocol 2 should be performed to find the suitable detergent for the extraction of desired membrane protein. An excellent article by Arachea et al. (2012) deals with the detergent selection procedure for enhanced extraction of membrane proteins in a systematic manner. Generally, the neutral/nonionic detergents with large headgroup and long alkyl tail length (~C12) are considered to be ‘mild’ and are suitable for purification and crystallization of membrane proteins without affecting the structural and functional integrity (Prive, 2007; Tate, 2010). This is the reason, DDM used in the Basic Protocol is considered to be one of the best detergents. However, these specialized detergents are expensive as has been stated before. Triton X-100, used in the Basic Protocol for solubilizing MgtE, is comparable to DDM in terms of charge, CMC, aggregation number and micellar size (Stetsenko and Guskov, 2017; Chatterjee et al., 2019). For a purified membrane protein, the ratio of micelles to membrane protein molecules should be fairly low (~1.5 to 2.0). In this regard, CMC of detergents should also be considered since detergents with low CMC tend to achieve micellar state at low concentrations and, therefore, excess amount of these detergents will create heterogeneous distribution of protein-detergent complexes, free micelles, and detergent monomers, which might complicate crystallization attempts (Wiener, 2004).

Membrane solubilization conditions

Although membrane solubilization by detergents is the ‘rate-limiting’ step for a successful membrane protein extraction and purification, the detailed mechanisms governing the interactions of detergents with membrane proteins are not yet fully understood (Chaptal et al., 2017). It is therefore critical to optimize the solubilization conditions. It will be better to check the solubilization efficiency either at room temperature or 4°C. High salt (NaCl or KCl) concentration can be used during solubilization step to achieve greater extraction. Using ‘dual-detergent strategy’, it has been shown that high concentration of salt (up to 1.5 M) added during Triton X-100 solubilization helps in increased extraction of KcsA without compromising its structural integrity (Tilegenova et al., 2016). However, in case of MgtE, there is no appreciable change in the extraction efficiency regardless of the salt concentration during membrane solubilization (Chatterjee et al., 2019). Another additive that can be tried during membrane protein extraction is polyhydric alcohol like glycerol, which are generally used as chemical chaperons to stabilize proteins.

Troubleshooting

No transformation of DNA construct in the E. coli strain

Firstly, check for the competent cells. Do not use very old cells. Prepare fresh competent cells (as described in Basic Protocol) before protein expression. Make sure that the correct selection antibiotic is present on the LB Agar plate.

Culture did not grow in the overnight incubation

Make sure to add required antibiotics at the correct concentration in the LB medium.

No or low protein expression

Make sure to screen for the E. coli strain/vector pair that works best (See Support Protocol 1). Further, optimize the post-induction temperature and concentration of IPTG. Check whether the appropriate primary and secondary antibodies are used for western blot analysis. If these things are fine, then it is possible that protein may be toxic before and/or after induction. Addition of 0.2 to 1% glucose might help when using expression vectors containing lac-based promoters. Tuneable promoters or titratable E. coli strains can also be tested. Further codon bias could be the reason for which one can optimize codon frequency in cDNA to match the codon usage of the host E. coli strain.

Protein yield is too low

In general, the yield of membrane proteins are low (~1-3 mg/liter) with some exceptions. Screen detergents to find the suitable detergent that extracts the desired membrane protein (see Support Protocol 2). Ensure that after adding Ni²⁺ affinity resin, mix well for 15 minutes. If required, increase the time of mixing to 1 hr. Also more column resin can be used. Further, the concentration of imidazole in the IMAC wash buffer can be decreased (as low as 5 mM), and also make sure that the imidazole concentration is at least 250 mM in the elution buffer. Furthermore, select the plasmid such that the position of the affinity tag on the protein is changed.

Protein loss after concentration

Check whether the filter membrane in the Amicon protein concentrator is not broken. Also, make sure to use the correct MW cut-off filter for concentrating the protein. The MW cut-off filter should be lower than the protein-detergent complex.

Impurities in protein preparation

Instead of Ni²⁺ affinity resin, Co²⁺ resins can be used to purify the protein. In general, the yield will be more when Ni²⁺ resin is used, but Co²⁺ resin can give relatively pure protein with lesser yield. Increase the salt concentration (~100-400 mM) in the wash buffer to prevent any kind of non-specific binding of unwanted proteins to the metal affinity resin. Also make sure that the gel filtration column is thoroughly washed before injecting the protein.

Purified protein is not stable

Membrane proteins tend to aggregate if the stability is compromised. This will sometimes reflect in higher oligomer bands in SDS-PAGE, and the gel filtration (SEC) profile will show aggregated peaks which comes out in the void volume of the column (~6-8 ml in the Superdex Increase 200 10/300 GL). It is important to note that the best detergent which extracts the protein most might not be the best detergent for maintaining the stability of the protein. Also, the aggregation propensity of the protein as a function of time to form small aggregates may not lead to appreciable turbidity to the naked eye. A quick diagnostic tool to determine aggregation is to perform small-scale stability test by monitoring the absorbance of the purified protein in detergent micelles at 280 and 320 nm as any absorbance in the non-absorbing region of the spectra (i.e., at ≥320 nm) is due to light scattering caused by the presence of aggregates (Wiener, 2004; Pignataro et al., 2020).

Gel filtration profile shows shoulder peaks

Chemical chaperones like glycerol can be added during solubilization and also in the wash and elution buffers. Also, changing the pH of the gel filtration buffer might help.

Anticipated Results

When the Basic Protocol, which uses the ‘dual-detergent strategy’ is followed properly, the large-scale expression and purification of MgtE will result in ~3 mg of pure, stable and functionally-relevant protein/liter of culture. Importantly, since Triton X-100 is used for membrane-solubilization, the cost of the overall purification of MgtE can be dramatically reduced by ~200 fold. For the characterization of a new membrane protein, the Support Protocols 1 and 2, which deal with the small-scale ‘expression test’ and ‘membrane solubility test’ to find the best E. coli strain for protein expression, and to screen for optimum detergent solubilizing the membrane protein, respectively will be highly useful. We show the schematic representation of the western blot results of these tests (Figure 4) for a sample membrane protein of ~50 KDa so that it helps the reader to appreciate the importance of these tests and also correctly interpret the results.

Figure 4.

Schematic representation of the small-scale ‘expression test’ and ‘membrane solubility test’ results of a sample membrane protein of ~50 KDa. Western blot of an expression test (Top panel) to find the best E. coli strain for protein expression, and to screen for optimum detergent (Middle panel) for solubilizing the membrane protein. Bottom panel shows the gel filtration (SEC) profiles of a large-scale expression using C41 cells and the indicated detergents for membrane solubilization and further purification. The dotted lines in the SEC profiles indicate the expected peak position for the functional oligomer. ‘Pre’ and ‘Post’ lanes in the Top panel denote the ‘pre-induction’ and ‘post-induction’ samples, respectively. See Support Protocols 1 and 2 for technical details, and Anticipated Results for more information on data interpretation.

The top panel of Figure 4 shows the expression of the sample membrane protein in various E. coli strains. From the western blot, it is clear that the XL-1 Blue cells do not express the protein. Interestingly, all other cells (BL21pLysS, C41 and C43) expresses the protein in the post-induction samples, of which C41 expresses the most. Although C43 expresses the protein, it has a leaky expression as can be seen from the appearance of a protein band in the pre-induction sample. Further, along with the 50 KDa prominent band, BL21pLysS also shows a band at ~100 KDa, which might correspond to the dimer of the protein. Basically, all these cells can potentially be tried for further screening. To show the further processes, we use the example of C41 cells in which the protein expressed the most. The western blot of membrane solubilization test (Figure 4, Middle panel) shows the efficiency of various detergents in extracting the sample protein from the membrane. It is obvious that, in this case, DDM and Triton X-100 extracts the protein well with the anticipated band at 50 KDa. The detergent DM also extracts the protein fairly well, showing two bands at 50 and 100 KDa. Being a positive control for the membrane solubilization test, SDS extracts the protein well as expected. However, SDS is a harsh detergent and will drastically perturb the stability of the biological oligomeric form and function of the protein. Therefore, SDS is not an ideal choice for membrane protein purification and crystallization. OM and Anzergent 3-14 are clearly not the suitable detergents for this sample protein since OM shows a predominantly higher oligomer band, which suggests that the protein stability might have been compromised, whereas in case of Anzergent 3-14, multiple bands are observed including a lower MW band at ~35 KDa. Considering the model protein’s monomeric MW is 50 KDa, this lower MW band observed using Anzergent 3-14 for membrane solubilization clearly suggests that not only the stability of the extracted protein is severely compromised, but also indicative of the proteolytic degradation. The detergent OG does not extract the protein at all from the membrane. Based on the above considerations, DDM, Triton X-100 and DM are the suitable detergents that can be used for the large-scale purification.

As mentioned earlier, the best detergent which extracts the protein most might not be the best detergent for maintaining the stability of the protein. One cannot draw a conclusion about the purity and homogeneity of the biological oligomer of the protein merely by western blot results since the loading and running buffer in the polyacrylamide gel electrophoresis contain the harsh detergent SDS. One must therefore use SEC to get information on the purity and most importantly, the homogeneity of the purified protein. Figure 3 (Bottom panel) shows the schematic gel filtration (SEC) profiles of a large-scale expression using C41 cells and the indicated detergents for membrane solubilization and further purification. As expected, the gel filtration profile of the model protein extracted and purified using DDM detergent has a single sharp peak, whereas the SEC profile is the worst in case of OM detergent with predominantly aggregated peak. DM shows the expected peak with slight aggregation. Overall, from these results, one can draw conclusion that C41 is the best expression host and DDM is the best detergent for solubilization and purification for this sample membrane protein. Of course, one should conduct the stability and functional assays as required for further downstream characterization.

Time Considerations

The time required for competent cell preparation and transformation will take 2 days. The Basic Protocol described to express, solubilize and purify a membrane protein in a cost-effective manner will take ~4 days. SDS-PAGE and SEC will take ~1-2 days. Additional time will be needed to identify cell strain and also for optimizing detergents. However, performing the expression test (Support Protocol 1) and membrane solubility test (Support Protocol 2) to identify the suitable E. coli strain and detergent for extraction, respectively can take considerable amount of time depending on the membrane protein under study.

Internet Resources

https://barricklab.org/twiki/bin/view/Lab/ProtocolsAntibioticStockSolutions

This is a link to a resource that lists standard concentrations of antibiotics commonly used in microbiology laboratories for E. coli.

https://youtu.be/ey19jM6y7-c

This video shows the LB Agar plate pouring protocol.

https://www.sigmaaldrich.com/IN/en/technical-documents/technical-article/protein-biology/protein-concentration-and-buffer-exchange/salt-selection-and-buffer-preparation

This link provides a quick guide to salt selection for buffer solution at different pH.

https://www.sigmaaldrich.com/IN/en/technical-documents/protocol/protein-biology/protein-concentration-and-buffer-exchange/buffer-reference-center

This link provides the reference for buffers and a quick guide for preparing commonly used biological buffers at different pH.

https://www.anatrace.com/Technical-Documentation/Catalogs/Anatrace-Detergents-Booklet-FINAL

This link contains a downloadable booklet that deals with Detergents and their Uses in Membrane Protein Science.