Production of Recombinant Transmembrane Proteins from Mammalian Cells for Biochemical and Structural Analyses

Abstract

Eukarayotic integral membrane proteins are key components of various biological processes. Being implicated in multiple diseases, it is important to understand their mechanism of action by elucidating their structure and function. Complex technical challenges associated with the generation of recombinant membrane proteins severely impair their thorough understanding using structural and biochemical methods. Here, we provide a detailed protocol to address and mitigate difficulties involved in the large scale heterologous overexpression and purification of eukaryotic membrane proteins using HEK293S GnTi^- cells transduced with baculovirus. Two human proteins, hDHHC15 and hPORCN are presented as examples with a step-by-step direction towards their transient transfection, generation of baculoviruses, followed by their overexpression and purification from HEK293S GnTi^- cells.

BASIC PROTOCOL 1: Small-scale protein expression in mammalian HEK293T cells

BASIC PROTOCOL 2: Generation of baculovirus from Sf9 (insect) cells

ALTERNATE PROTOCOL: Enumeration-free method for generating P2 viral stock

SUPPORT PROTOCOL 1: Small-scale transduction of HEK293T cells with P2 baculovirus

BASIC PROTOCOL 3: Large-scale viral transduction of HEK293S GnTi- cells

SUPPORT PROTOCOL 2: Large-scale membrane preparation from HEK293S GnTi- cells

BASIC PROTOCOL 4: Large-scale purification of membrane proteins from HEK293S GnTi- cells

INTRODUCTION

More than half of all drug targets are integral membrane proteins (IMP) (Russell & Eggleston, 2000; Yildirim et al., 2007). IMPs, comprising 30% of human proteome (Fagerberg et al., 2010), play crucial roles in myriad physiological processes. Advances in X-ray crystallography and now cryoEM have significantly impacted high resolution structure determination of membrane proteins (Vinothkumar & Henderson, 2010). A key bottle neck in obtaining these proteins is identifying conditions where the protein is stably overexpressed followed by its isolation into an environment that closely mimics its lipidic habitat. Various prokaryotic to eukaryotic expression systems, are now available for rapid overexpression of stable membrane proteins (Bernaudat et al., 2011). Meanwhile, accessibility to a plethora of detergent and lipid mixtures has enabled their large-scale extraction and purification (Phillips et al., 2009; Prive, 2007).

Here we present a comprehensive method for the expression and purification of integral membrane proteins in HEK293S GnTi^- cells (Reeves, Callewaert, et al., 2002), using the BacMam system and monomeric Venus (mVenus) protein as a fluorescent fusion tag (Rana et al., 2018). This protocol can be easily extended to any suspension adapted mammalian cells. Though we use mVenus, the protocol is generally adaptable to any fluorescent tag. We demonstrate the applicability of our protocol through two human proteins, DHHC15 (hDHHC15) and Porcupine (hPORCN). Both proteins were cloned in frame with a Venus tag into the pEG BacMam (Eric Gouaux lab) vector, originally derived from pVLAD (Dukkipati et al., 2008). pEG BacMam is an optimized vector that facilitates large scale protein expression (Goehring et al., 2014), suitable for X-ray crystallography or cryoEM. The presence of an mVenus tag in general offers several advantages – 1) it enables a crude assessment of protein expression by fluorescence microscopy, 2) it helps determine the monodispersity of a protein sample at early to mid stages of extraction and purification by fluorescence detection size exclusion chromatography (FSEC) (Kawate & Gouaux, 2006), and 3) it offers, overall higher yields than other fluorescence fusion tags such as GFP. Baculoviruses are made using a large-scale direct method which circumvents the need for producing multiple generation of viruses, providing large volumes of high titre viruses by P₂. Additionally, we also discuss a simple method for generating competent viruses, without the need for determining titres, using FSEC. Finally, we conclude with, a step-by-step, large-scale purification scheme for hDHHC15 and hPORCN that can be used as a primer for any IMP of choice.

BASIC PROTOCOL 1: Small scale protein expression in mammalian HEK293T cells

In the first step, we will test the expression of a membrane protein through transient transfection in adherent mammalian (HEK293T) cells. We cloned hDHHC15 and hPORCN into pEG BacMam vector with a C terminal mVenus tag. For DNA purification, E. coli (XL-1 Blue) cells are transformed and grown in the presence of Ampicillin. Purified DNA is then transfected into HEK293T cells using PEI. Protein expression and homogeneity are assessed from crude extracts by analyzing peak height and shape in a size-exclusion chromatogram obtained from an HPLC equipped with a fluorescence detector and then in-gel fluorescence with SDS-PAGE. Most membrane proteins are amnable to extraction with DDM (eg. hDHHC15), while others may require specific detergent or detergent mixture, with CHS and/or lipids (eg. hPORCN: DM+CHS+POPC). Typically, at this step, divergent orthologs with mVenus fused either to its N or C termini are screened to identify candidates with best expression (Hattori et al., 2012).

Materials

• HEK293T cells (ATCC #CRL‐3216)

• DMEM (Corning #15–013-CV)

• Fetal Bovine Serum- heat inactivated (Corning #35–011-CV)

• 200 mM L-Glutamine (Corning #25–005-Cl)

• Penicillin/Streptomycin 100X (Corning #30–002-Cl)

• Polyethylenimine (PEI) 40-kDa, (Polysciences #24765)

• 6-well tissue culture plates (Falcon #353046)

• Cell culture incubator (5% CO₂, 37°C, humidified) (HERACell 150i – Thermo Scientific)

• 10X PBS and 10X TBS (KD Medical RGF-3210 and RGF-3385)

• DNA purification: QIAprep Spin Miniprep Kit (Qiagen #27104)

• n-Dodecyl-β-D-Maltopyranoside (DDM) (Anatrace #D310S)

• n-Decyl-β-D-Maltoside (DM) (Anatrace #D322)

• tris(2-carboxyethyl)phosphine-HCl (TCEP) (GoldBio)

• Cholesteryl Hemisuccinate Tris Salt (CHS) (Anatrace #CH210)

• 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine (POPS) (Avanti #840034)

• 0.22 μm spin filter (non-sterile) (Corning – Costar #07–200-386)

• Fluorescence microscope: EVOS fl Auto (Thermo Fisher)

• Shimadzu UFLC

• Deoxyribonuclease I (Worthington #LS002139)

• Protease inhibitors:

  Phenylmethylsulfonyl fluoride (Goldbio # P-470); AEBSF-HCl (Goldbio #A-540) Benzamidine HCl (Goldbio #B-050); Pepstatin (Goldbio #P-020); Leupeptin Hemisulfate (RPI #L22035); Soy trypsin (RPI # T23025); Aprotinin solution (Fisher #BP2503)

Protocol steps

Prior to start

• Warm DMEM media in a 37°C water bath.

• Prior to transfection make sure HEK cells are >70% confluent.

• Protocol for a single well in a six-well cell culture plate.

• Purify DNA from 10mls of Xl-1 blues cells using the mini-prep kit.

(Perform steps 1–5 in a cell culture hood)

1. Add 2 μg purified DNA (~400–700 ng/μl) to 100 μl DMEM media without supplementation in a sterile microfuge tube.

2. Add 6 μg PEI (1 mg/ml) to 100 μl DMEM media without supplementation in a sterile microfuge tube.

3. Mix PEI into the DNA tube and leave at room temperature (RT), in the hood, for 15–20 min.

4. Add the mixture to HEK cells and gently mix by shaking the tissue culture plate.

5. Incubate in the 37°C cell culture incubator for 48 hours.

6. Aspirate media and rinse cells with 1X PBS. Add 1 ml of 1XPBS and dislodge cells by repeated pipetting or using a cell scraper.

7. Harvest cells at 3000 ×g at RT for 10 min. in a microfuge tube.

8. Discard supernatant and resuspend cells in 500 μl of 1X PBS. Harvest at 3000 ×g, RT for 10 min.

9. Resuspend cells in 180 μl of solubilization buffer (see Reagents section).

10. Add 20 μl of detergent stock (10% DDM or 20% DM +/− CHS and lipid) and mix on a rotator at 4°C for 1.5 hrs.

11. Centrifuge at 21,000 ×g at 4°C to clarify cell debris. Collect the supernatant and add to a spin filter.

12. Centrifuge at 9000 ×g for 10 min at 4°C and collect filtrate.

13. Analyze the filtrate on an SDS-PAGE gel and through FSEC.

BASIC PROTOCOL 2: Generation of Baculovirus from Sf9 (insect) cells

Results from BASIC PROTOCOL 1 are indicative of the extent of protein expression and its homogeneity. In this section we outline a method to prepare baculovirus (Autographa californica multi-nucleopolyhedrovirus (AcMNPV)) from insect Sf9 (Spodoptera frugiperda) cells. The recombinant pEG BacMam vector (donor plasmid, with gentamicin resistance), containing the gene of interest, is transformed into E. coli host strain, DH10Bac™ containing the baculovirus shuttle vector (bacmid, with Kanamycin resistance) and a helper plasmid (Tetracycline resistance). The helper plasmid encodes a transposase that provides the Tn7 transposition function between the mini-Tn7 elements on the pEG BacMam vector and mini-attTn7 target site on the bacmid forming a recombinant bacmid. Mini-attTn7 is in-frame with a LacZ⍺ peptide which gets disrupted due to the insertions of mini-Tn7 into the mini-attTn7 attachment site forming white colonies in a background of blue colonies that are devoid of insertion/transposition. White colonies are picked after restreaking and subcultured to isolate the recombinant bacmid. Sf9 insect cells are transfected with the recombinant bacmid. Once infected, the cells will stop growing, have a granular appearance and eventually lead to cell lysis. Viruses are harvested at this stage and designated as the P₁ viral stock. P₁ is a low titre stock and is further amplified by infecting Sf9 cells generating the P₂ viral stock with high titre in large volumes, enough to transduce serval liters of mammalian cells.

Materials

• Sf9 cells (ATCC #CRL-1711)

• Cell counter (Nexcelom Bioscience)

• DH10Bac™ E. coli competent cells (Thermo Fisher #10361012)

• E. coli XL-1 blue cells (Agilent #200130)

• S.O.C media (Invitrogen #15544–034)

• DNA purification: QIAprep Spin Miniprep Kit (Qiagen #27104)

• Anibiotics : Kanamycin sulfate (VWR #0408); Tetracycline-HCl (RPI #17000); Gentamycin sulfate (RPI #G38000); X-gal (PRI #B71800); IPTG (Goldbio #12481)

• Sf9 media: ESF921 (protein free) (Expression system #96–001-01)

• Cell culture shaker incubator (INFORS HT – Multitron)

• Cell culture flasks (Sf9) (Autoclavable) : Plain Bottom flasks 125 ml (#FPC0125S); 250 ml (#FPC0250S); 1000 ml (#FPC1000S); 2000 ml (#FPC2000S) (TRIFOREST Labware)

• 0.45 μm syringe filter (33mm, sterile) (Millipore Sigma #SLHPM33RS)

• 0.45 μm Nalgene fast-flow (50mm, sterile) (Thermo Scientific #09–740-65B)

• Virus Counter (Sartorius- ViroCyt 3100)

• Isopropanol (Sigma #278475)

• Ethanol (Pharmco # 111000200)

Protocol steps

1. Fresh pEG BacMam DNA was purified for hDHHC15 and hPORCN after transforming E. coli (XL-1) cells using a standard DNA purification kit.

2. Transformation of DH10bac E. coli cell for bacmid preparation.

   1. Add 1 μl of plasmid (500–700 ng/μl) to 100 ul of DH10bac E. coli cells in a microfuge tube.
   2. Incubate on ice for 15 mins.
   3. Heat shock in 42°C waterbath. 40–45 secs.
   4. Transfer to ice and keep for 3 mins.
   5. Add 400 μl of SOC media.
   6. Incubate at 37°C for 4–5 hrs.
   7. Plate 70–100 μl on a transposition plate. Transposition plate composition: Kanamycin (50 μg/ml), Tetracycline (10 μg/ml), Gentamycin (7 μg/ml), X-gal (100 μg/ml), IPTG (0.5 mM).
   8. Incubate for 2–3 days or till blue-white colonies are visible.
   9. Pick three white colonies and re-streak on another transposition plate.
   10. Pick two for bacmid prep. Store the plate at 4°C.

   Note: Bacmid DNA is large and hence fragile. All steps should be performed in a manner that prevents DNA shearing.

3. Bacmid prep:

   1. Inoculate 10 ml of LB [with Kanamycin (50 μg/ml), Tetracycline (10 μg/ml), Gentamycin (7 μg/ml)] by picking white colonies from a re-streaked plate.
   2. Grow O/N (16–18 hrs) in a 37°C shaker incubator.
   3. Next morning, pellet cells and resuspend in 250 ul of resuspension buffer from the Mini-prep kit. Transfer to a microfuge tube.
   4. Add 250 μl of lysis buffer. Invert and incubate for 2–4 mins.
   5. Add 300 μl of precipitation buffer. Invert multiple (6–8) times.
   6. Spin at 17000 ×g, RT, 10 mins.
   7. Carefully aspirate 650–800 μl into a 2 ml microfuge tube. [avoid white debris].
   8. Add equal volume of isopropanol.Gently, invert (avoid vortexing) few times and incubate at −20°C for 3 hrs to O/N.
   9. Centrifuge @ 15,000 ×g, 4°C, for 10 mins. DNA will precipitate at the bottom of tube.
   10. Aspirate out the supernatant.
   11. Flick the tube to dislodge the pellet. Add 500 μl of 70% ethanol to the precipitate. Invert tube a few times. Spin at 17000 ×g, RT for 3 mins.
   12. Carefully aspirate out the 70% ethanol. Leave the tube open in a 37°C heat block to dry the DNA. Alternately, leave the tube open O/N at RT in a dust free enviornment.
   13. Flick the tube to dislodge the pellet. Add 80–150 μl of 10 mM Tris-1 mM EDTA (Elution buffer) from the Mini-prep kit. DO NOT VORTEX. Incubate at 37°C for a few minutes (3–5 min) to help dissolve the DNA.
   14. Measure DNA concentration using Nanodrop. Typical yields range from 0.7–2 μg/ul.
   15. Store at 4°C. Do NOT freeze.
   16. Optional step: Confirm the presence of your gene of interest by sequencing with internal primers.

Note: Since Bacmids are large DNA and prone to easy shear. It is advisable to prepare fresh if not used within a week. Otherwise, start from the restreaked transposition plate and purify fresh bacmid.

P₁ Baculovirus prep: The large-scale direct (LSD) method

Note: Perform all steps in a clean cell culture hood.

Prior to start

• • Bring Sf9 media to room temperature by leaving it O/N at room temperature on the previous day.
  • Cell culture shaker settings (Sf9): 28°C with 40% humidity and 130 rpm.
  • Make sure Sf9 concentration is ≥ 3 million/ml with >95% viability.

1. Determine Sf9 concentration using a cell counter

2. Grow Sf9 cells to ~ 3 million/ml in ESF921 media in the shaker incubator.

3. Pellet 30 ml of cells at 500 ×g, RT, in a sterile 50 ml conical tube. Aspirate out the media.

4. Resuspend in 20 ml of ESF921 media, transfer to 125 ml cell culture flasks, then make up volume to 48.5 ml.

5. Add 1.5 ml of 100% FBS (0.22 μm filter sterilized) to give a final FBS concentration ~ 3% in a total volume of ~ 50 ml.

6. Leave shaking in the incubator.

7. Pipette 30 μg bacmid DNA into a 15 mL conical tube. Dilute bacmid DNA with 250 μl of ESF921 media. Gently mix by pipetting a few times.

8. In another 15 mL tube, pipette out 360 μl polyethyleneimine (PEI). DNA:PEI ratio is set at 1:12. Add 250 μl of ESF921 media to PEI. Mix by pipetting.

9. Mix PEI and bacmid DNA into one tube. Gently mix by pipetting a few times.

10. Incubate in the cell culture hood for 10–15 mins.

11. Add the DNA-PEI mix to the flask with 50 ml of Sf9 cells prepared in steps 2–5.

12. Return the cells to the shaker incubator and leave shaking for 6–7 days.

    Note: Inspect the culture for contamination.

13. After the last day, pour cells into a 50 mL conical tube.

14. Spin @ 10,000 ×g, 4°C, for 10 mins.

15. Filter the supernatant through a 0.45 μm syringe filter into a fresh 50 mL conical tube.

16. Add 0.5 ml of FBS to the virus stock. Final FBS concentration ~ 1%.

17. Measure the virus titre using virus counter. Typical titres are ~1E⁶ – 1E⁷ vp/ml. Wrap with Aluminum foil and store in the dark at 4°C.

18. This P₁ stock will be amplified to make larger volumes of P₂ virus.

Note: Viral stocks typically last for months. In case of contamination the solution will turns cloudy, discard immediately. Always inspect for contamination before further use.

P₂ virus preparation

1. Grow ~ 400–500 ml of Sf9 cells to a density of 2.5 to 3 million/ml.

2. Add 1.5–2 ml of P₁ virus (prepared above), to the cells.

3. Incubate at 28°C for 3–3.5 days.

4. Split cells into multiple sterile 50 ml conical tubes.

5. Spin down cells at 10,000 ×g, 4°C, 10 mins.

6. Pool and filter into 0.45 μm, 500 ml cell-culture Nalgene fast-flow filter flask.

7. Add 4–5 ml of 100% FBS to make a final concentration of ~ 1%.

8. Measure the virus titre using virus counter. Typical titres are ~1E⁸ – 1E⁹ vp/ml. Wrap with Aluminum foil and store in the dark @ 4°C.

Note: Viruses are sensitive to light and it is important to leave them covered with aluminum foil at 4°C. Typically, they last for months. Check for contamination before transduction.

ALTERNATE PROTOCOL: Enumeration-free method for generating P₂ viral stock

We developed a rather simple but effective method to generate P₂ viral stock, by screening different dilutions of P₁, at a pilot scale. We call it the “enumeration-free” method as it circumvents the need to determine P₁ and P₂ viral titres while giving a direct estimation of your viral potency. Different dilutions of the P₁ viral stock are screened to generate various P₂ stocks. The potency of each P₂ virus is determined by FSEC analysis of HEK293T adherent cells transduced with the virus. Large-scale P₂ virus is prepared by scaling-up the best dilution ratio.

Materials

• P₁ Baculovirus from BASIC PROTICOL 2

• Sf9 cells (ATCC #CRL-1711)

• Sf9 media: ESF921 (protein free) (Expression system #96–001-01)

• Cell culture shaker incubator (INFORS HT – Multitron)

• Cell culture incubator (5% CO₂, 37°C, humidified) (HERACell 150i – Thermo Scientific)

• Cell culture flasks (Sf9) (Autoclavable) : Plain Bottom flasks 125ml (#FPC0125S); 250ml (#FPC0250S); 1000ml (#FPC1000S); 2000ml (#FPC2000S) (TRIFOREST Labware)

• 0.45μm syringe filter (33mm, sterile) (Millipore Sigma #SLHPM33RS)

• 0.45μm Nalgene fast-flow (50mm, sterile) (Thermo Scientific #09–740-65B)

• DMEM conditioned media (see Reagents section)

• HEK293T cells (ATCC #CRL‐3216)

• 6-well tissue culture plates (Falcon #353046)

• 10X PBS and 10X TBS (KD Medical RGF-3210 and RGF-3385)

• Protease inhibitors:

  Phenylmethylsulfonyl fluoride (Goldbio # P-470); AEBSF-HCl (Goldbio #A-540) Benzamidine HCl (Goldbio #B-050); Pepstatin (Goldbio #P-020); Leupeptin Hemisulfate (RPI #L22035); Soy trypsin (RPI # T23025); Aprotinin solution (Fisher #BP2503)

• n-Dodecyl-β-D-Maltopyranoside (DDM) (Anatrace #D310S)

• Shimadzu UFLC

Protocol steps

Note: Steps (1–6) should be carried out in a cell culture hood.

Prior to start

• • Bring Sf9 media to room temperature by leaving it O/N at room temperature.
  • Cell culture shaker settings (Sf9): 28°C with 40% humidity and 130 rpm.
  • Make sure Sf9 concentration is 1–1.5 ×10⁶ cells/ml with >95% viability.
  • Split HEK293T adherent cells, such that they are available for transduction at 70–80% confluency.

• 1.Take 6, 125ml cell culture flasks. To each add 10 mls of Sf9 cells at 1–1.5 ×10⁶ cells/ml.
• 2.To each, add different P₁ virus at volumes of −400 ul, 200 μl, 100 μl, 20 μl, 10 μl and 2 μl.
• 3.Grow for 3–4 days in shaker incubator at 28°C, 130 rpm, 40% humidity.
• 4.Harvest cells by centrifuging at 1000 ×g, RT for 15 min. Filter the supernatant (P₂ viral stock) using 0.45μm syringe filter. Store at 4°C.
• 5.Have HEK293T adherent cells ready in a 6 well tissue culture plate at 70–80% confluency.
• 6.Transduce with 100–200 μl of the P₂ virus and incubate for 36–48 hours in a cell culture incubator.
• 7.Wash HEK293T cells with 1X PBS and harvest at 3000 ×g, RT for 10min.
• 9.Resuspend cells in 180 μl of solubilization buffer (see Reagents section).
• 10.Add 20 μl of 10% DDM (or the best detergent +/− CHS and/or lipid mixture) and mix on a rotator at 4°C for 1.5 hrs.
• 11.Centrifuge at 21,000 ×g at 4°C to clarify cell debris. Collect the supernatant and add to a spin filter.
• 12.Centrifuge at 9000 ×g for 10 min at 4°C and analyze filtrate through FSEC.
• 13.Use the best dilution and scale up to 1L of Sf9 cells.
• 14.Alternatively, using the FSEC results screen further between the two best P₁ dilutions and scale up.

SUPPORT PROTOCOL 1: Small-scale transduction of HEK293T cells with P₂ baculovirus

After preparation of the P₂ viral stock, it is important to check for transduction efficiency. Since the protein of interest is tagged with a fluorescent protein you can directly visualize expression by monitoring cell epifluorescence.

Materials

• P2 Baculovirus from BASIC PROTICOL 2

• HEK293T cells (ATCC #CRL‐3216)

• DMEM conditioned media (see Reagents section)

• 6-well tissue culture plates (Falcon #353046)

• Cell culture incubator (5% CO₂, 37C, humidified)

• EVOS FL Cell Imaging Microscope (Life Technologies)

Protocol steps

1. Grow HEK293T in a 6-well tissue culture plate till they are 70–80% confluent.

2. Add 100 μl of P₂ virus to a well.

3. Leave in the cell culture incubator for an additional 36–48 hrs.

4. Visualize cells by epifluorescence microscopy.

BASIC PROTOCOL 3: Large scale viral transduction of HEK293S GnTi^- cells

Good quality viruses have a direct correlation to good protein expression. After confirming that the P₂ viral stock offers appreciable protein expression, large scale viral transduction of HEK293S can be performed. HEK293 GnTi^- is derived from HEK293S cell line that lacks N-acetylglucosaminetransferase-I and hence deficient in complex glycosylation. Proteins expressed in these cells are homogeneously N-glycosylated with a GlcNAc2Man5 sugar unit. Once transduced the presence of a strong human cytomegalovirus (CMV) promoter leads to protein expression. Post transduction, addition of sodium butyrate, a histone deacetylase (HDAC) inhibitor, further amplifies protein expression.

Materials

• HEK suspension media: Hyclone CDM4HEK293 (without L-Glutamine) (GE Healthcare #SH30858.02)

• Fetal Bovine Serum- heat inactivated (Corning #35–011-CV)

• 200 mM L-Glutamine (Corning #25–005-Cl)

• Penicillin/Streptomycin 100X (Corning #30–002-Cl)

• HEK293S GnTi^- (ATCC #CRL-3022)

• Cell culture flasks (HEK) (Autoclavable) : Baffled Bottom flasks 125ml (#FBC0125S); 250ml (#FBC0250S); 1000ml (#FBC1000S); 2000ml (#FBC2000S) (TRIFOREST Labware)

• Sodium butyrate (Sigma #303410)

• Centrifuge flask: J-Lite, 1000ml, PP (Beckman Coulter #A98813)

• 10X PBS (KD Medical RGF-3210)

Protocol steps

Prior to start

• Warm HEK suspension media by placing it in a 37°C water bath.

• Cell culture shaker settings (HEK293S GnTi^-): 37°C with 80% humidity, 7% CO₂ and 130 rpm.

• Make sure HEK cells are at a concentration of 2.5–3 million/ml with >95% viability.

• Bring the P₂ virus to room temp by briefly placing them in 37°C water bath or leaving it at RT the previous day. Prior to transduction inspect for contamination.

Note: Perform steps 1–4 in the cell culture hood

• 1Grow 800–900 ml of HEK 293S cells in 2L culture flasks to a cell density of 2–3 million/ml.
• 2Add 80–90 ml of the P₂ virus to HEK cells.
• 3After 12–16 hrs of infection, add 1 M sodium butyrate to a final concentration of 10–15 mM.
• 4Incubate for another 48 hr.

Note: The incubation time should be determined empirically, typically between (36–72 hrs)

• 5Transfer cells to a 1L centrifuge flask.
• 6Harvest cells by pelleting at 3000 ×g for 20 mins @ 4°C.
• 7Discard the supernatant/media.
• 8To the pellet, add ~ 15 ml of 1X PBS. De-clump cells by pipetting and transfer to a fresh 50 ml conical tube.
• 9Pellet cells at 3000 ×g, RT, for 15 mins. Discard supernatant.
• 10Resuspend in 15 ml of 1X PBS. Transfer to a 50mL conical tube and centrifuge at 4000 ×g, RT, for 15 min. Discard supernatant.
• 11Flash freeze in liquid nitrogen and store at −80°C for later use.

SUPPORT PROTOCOL 2: Large-scale membrane preparation from HEK293S GnTi^- cells

While purifying membrane proteins, a major source of protein contamination arises from soluble proteins in the cell lysate. Hence target proteins are preferentially purified from isolated membranes that are washed to remove soluble contaminants. Typically, the purity improves at an often perceptible loss of yield.

Materials

• Ultracentrifuge (Beckman Coulter)

• Ultracentrifuge Rotor Type 45 Ti

• Ultracentrifuge tubes

• Branson 450 Digital sonifier with Branson sonicator tapped step horn (#101147037)

• Homogenizer: Potter-Elvehjem Tissue Grinders 55ml (Wheaton #358054)

• Cell resuspension buffer (see Reagents section)

• Membrane resuspension buffer (see Reagents section)

Note: For missing details on an individual material please refer to the Materials section in BASIC PROTOCOL 4.

Protocol steps

1. Thaw pelleted HEK293S GnTi^- cells (~from 3 L) in a beaker of water at room temperature.

2. Once cells are thawed, resuspend them in ~ 200 ml of cell resuspension buffer. Stir at 4°C.

3. Sonicate on ice-water.

   1. Large probe. Amplitude = 80%
   2. Total process time = 2 min
   3. Pulse on = 5 sec
   4. Pulse off = 20 sec
   5. Add 0.5 ml 100mM PMSF in 1 min intervals.

4. Transfer the lysate to four 50 ml conical tubes on ice. Spin @ 9000 ×g, 4°C, for 10 mins.

5. Carefully pipette the supernatant into an ice-chilled clean container.

6. Resuspend the pellets in 100 ml total of cell resuspension buffer. Sonicate again at same setting for a total of 30 sec.

7. Spin at 9000 ×g, 4°C for 10 mins. Aspirate the supernatant and pool with the previous supernatant. Should give ~ 280–300 ml.

8. Centrifuge at 200,000 ×g at 4°C for 1.5 hrs in ultracentrifuge tubes.

9. Aspirate the white ring that floats on top and carefully decant the supernatant.

10. Transfer membrane pellet into a 50 ml conical tube using metal spatula. Residual pellet can be flushed out using membrane resuspension buffer.

11. Homogenize the pellet using the homogenizer in a total volume of 50 ml in membrane resuspension buffer.

12. Split the membrane homogenate equally into two 50 ml conical tubes. (~ 25 ml each).

13. Flash freeze in liquid nitrogen and store @ −80°C.

BASIC PROTOCOL 4: Large scale purification of membrane proteins from HEK293S GnTi^- cells

Detergent is the single most essential requirement in purifying membrane proteins. Here we outline a typical purification protocol where cell lysis and concomitant protein extraction are achieved by the use of high detergent concentration followed by affinity and size exclusion chromatography. We show the purification of two human proteins, (A) Palmitoyl Acyl Transferase, hDHHC15 and (B) Porcupine, hPORCN. Both proteins are expressed with a C-terminal mVenus and 10X Histidine (His) tag that provides affinity to Co⁺² IMAC (TALON) resin. An HRV-3C (PreScission) protease cut site interspersed between the protein and mVenus allows for their cleavage while still attached to the resin. The IMAC eluate is further purified through size exclusion chromatography obtaining pure protein for myriad applications.

Materials

• HRV 3C (PreScission) protease (GE Healthcare #27084301)

• Branson 450 Digital sonifier with Branson sonicator tip (#101148070)

• Talon resin (TaKaRa #635503)

• Ultracentrifuge (Beckman Coulter)

• Ultracentrifuge Rotor Type 45 Ti (Beckman Coulter # 339160)

• Ultracentrifuge tubes (Beckman Coulter #355655)

• 0.45 μm membrane filter (47mm) (Millipore Express PLUS #HPWP04700)

• Filter Holder with Receiver 47mm, 250ml (Thermo Scientific #300–4000)

• Econo-pac Columns (Bio-Rad)

• Imidazole (Sigma #IO250)

• Superdex S200 Increase column 10/300GL (GE Healthcare #28990944)

• Superose 6 Increase 10/300 GL (GE Healthcare # 29091596)

• AKTA Prime (housed in a cold cabinet at 4°C) (GE Healthcare)

• Amicon® Ultra-4 Centrifugal Filter Units (Millipore-Sigma #UFC803008, #UFC805008 )

• 12% precast SDS-PAGE gel (Bio-Rad #4561043)

A. Purification of hDHHC15

Prior to start

• • Perform purification in a cold room and keep all protein samples on ice
  • Prepare 10% DDM solution
  • Prepare 4M Imidazole stock solution
  • Prepare solubilization buffer (SB), wash buffer-1 (WB1), wash buffer-2 (WB2), wash buffer-3 (WB3), PreScission Protease cleavage solution (PxpS), Size Exclusion Buffer (SEC)
  • Equilibrate Talon resin in detergent buffer

1. Place a 50 mL conical tube containing 1 L of frozen cell paste in a beaker of water at room temperature ~(20min)

2. Add 15 ml of SB and vortex till no cell clumps are visible.

3. Sonicate at 4°C in an ice bath for 1 min with 3 sec on and 15 sec off using a mirotip sonicator.

4. Using a serological pipette measure out the volume and transfer to another 50 ml conical tube. Make up the volume to 19 ml with SB. Ensure the pH is 7.5

5. Add 6 ml of 10% DDM stock to make a final of 1.5% in 40 ml.

6. Mix on a rotator at moderate speed for 2 hrs at 4°C.

7. Transfer to an ultracentrifuge tube and centrifuge at 200,000 ×g at 4°C for 50 mins.

8. Meanwhile take 1.5 ml of TALON resin slurry and rinse with water and SB. Move the resin into a fresh 50 ml conical tube. Add 2 ml of SB and DDM from stock to make a final concentration of 0.5% detergent.

9. Leave the tube on a rotator for 30 min at RT.

10. Centrifuge the tube at 700 ×g for 3 min and gently discard the supernatant. Leave the resin containing tube at 4°C.

11. Post ultracentrifugation, gently remove the top white ring and transfer the supernatant, without disturbing the pellet, to a 50 ml conical tube.

12. Clarify the supernatant through a (prewet) 0.45 μm membrane in a 47 mm filtration device at 4°C. This step is optional.

    Note: Multiple membranes may be required due to clogging.

13. To the supernatant add imidazole to a final concentration of 5 mM and incubate with TALON resin for 2 hrs at 4°C.

14. Spin down TALON resin at 700 ×g, 4°C for 3 mins. Decant ~ 30 ml of the supernatant and gently resuspend the reisn in the remainder solvent.

15. Pour into a BioRad Econo-pac column at 4°C. Collect the flow-through into the, resin containing, conical tube. Try to wash off resin from tube walls and repour into the column.

16. Allow the flow-through drain out and wash the resin with 4.5 ml WB1, 6 ml of WB2, and 4.5 ml of WB3. Use a stop cock to maintain a low flow rate during the washing steps.

17. Close the lower end of the column and tap to dislodge the resin from the bottom. Add 4 ml of PxpS and tightly close the other end of the column.

18. Leave the column O/N on a rotator at 4°C, with moderate speed. Avoid frothing.

19. Meanwhile, de-gass SEC buffer at RT and leave at 4C.

20. Next day, open both ends of the column and collect the elution. Add 1 ml of WB3 to rinse and collect.

21. Equilibrate Superdex S200 column with SEC buffer.

22. Concentrate the elution in a prewashed 30 MWCO-4 ml Amicon concentrator to ~ 300–320 μl at 4°C.

23. Resuspend the concentrate with a pipette and transfer to a microfuge tube.

24. Centrifuge at 21,000 ×g for 10 min at 4°C. Avoiding precipitates, inject the protein into the equilibrated Superdex S200 column.

25. Collect 500 ul fractions and analyze on a 12% SDS-PAGE gel. Stain with Coomassie blue.

26. Use fractions containing the pure protein for structural studies or protein assay.

B. Purification of hPORCN

Prior to start

• • Prepare 15% (w/v) DDM / 1.0% (w/v) CHS solution
  • Prepare solubilization buffer (SB), wash buffer-1 (WB1), wash buffer-2 (WB2), and Size Exclusion Buffer (SEC)

  Note: Detergent-solubilized hPORCN is very fragile. The purification should be peformed in a cold room and the protein samples should be kept on ice.

1. Thaw 7 ~ 10 g frozen cells in a 50 ml conical tube on ice (2 hrs).

2. Add 25 mL of SB and vortex till no cell clumps are visible.

3. Transfer the sample into a 250mL stainless steel beaker and add 56 ml of SB.

4. Sonicate the sample on ice for 1 min with 3 sec on and 15 sec off using a tip sonicator.

5. Add 9 ml of 15% DDM / 1.0% CHS stock to make a final of ~1.5% DDM and ~0.1% CHS in 90 ml.

6. Mix on a rotator at moderate speed for 1 hr at 4°C.

7. Transfer to ultracentrifuge tubes and centrifuge at 200,000 ×g at 4°C for 50 mins.

8. Meanwhile, prepare two 50mL conical tubes containing 1.5ml Talon resin rinsed with SB. Add 100 μl of 15% DDM / 1.0% CHS stock to the 50mL tubes.

9. Decant the supernatant (containing solubilized proteins) and transfer 45 ml each to the 50mL conical tubes containing Talon resin. Leave the tubes on a rotator for 1 hr at 4°C.

10. Appy the sample into a Bio-Rad Econo-Pac column (1.5 × 12 cm polypropylene) and collect the protein-bound resin. Allow the supernatant to drain through completely by gravity flow.

11. Wash the resin with ~ 8.5 ml WB1 and ~4.5 ml of WB2. Use a stop cock to control the drip rate during the washing steps.

12. Close the column and resuspend the protein-bound resin with 6 ml of WB2. Add 250 μl of HRV 3C protease (1 mg/ml) and tightly close the other end of the column.

13. Leave the column for 3 hrs on a rotator at 4°C, with moderate speed, to avoid detergent frothing.

14. Meanwhile, de-gass SEC buffer at RT and leave at 4°C.

15. After the protease cleavage, open both ends of the column and collect the elution. Add 1 ml of WB2 to rinse and collect. Combine the elutions.

16. Equilibrate Superose 6 Increase column with SEC buffer.

17. Concentrate the elution in a prewashed 50K MWCO-4 ml Amicon concentrator to ~ 250–300 μl at 4°C.

18. Aspirate the concentrate with a pipette and transfer to a microfuge tube.

19. Centrifuge at 21,000 ×g for 10 min at 4°C. Avoiding precipitates, inject the protein into the equilibrated Superose 6 Increase column.

20. Collect 250 μl fractions and analyze on a 12% SDS-PAGE gel and stain with Coomassie blue.

21. Fractions containing the pure protein can be further used for protein assay.

REAGENTS AND SOLUTIONS

All reagents are made in Milli-Q water. Cell culture reagents are filtered using 0.22 μm sterile filters. Buffer solutions for purification are filtered using nonsterile 0.22 μm filters. Only use sterile/autoclaved microfuge and conical tubes. All solutions should be stored at 4°C, until specified.

• DMEM conditioned media: Add 50 ml of FBS + 10 ml of L-Glutamine + 5 ml of pen/strep to 435 ml of DMEM media. Store media at 4°C.

• PEI stock solution: Add 10 mg to 7 ml of water. Adjust pH ≤ 7.0 with 0.1M NaOH. Make up the volume to 10 ml and filter sterilize with a 0.22 μm syringe filter. Store aliquots in microfuge tubes at −20°C. Working stock once thawed from −20C should be kept at 4C.

• 4M Imidazole stock: weight 5.4 gm of imidazole in 20 of 1X TBS. Adjust pH to 7.5 and filter sterilize with a 0.22 μm syringe filter.

• 10% DDM stock: Add 0.5 gm of DDM in 5 ml of 1X TBS solution.

• 20% DDM stock: Add 0.5 gm of DM to 2.5 ml of 1X TBS solution.

• 15% DDM with 1% CHS stock: Add 4.5 g of DDM and 0.3 g of CHS (cholesteryl hemisuccinate) to a 50mL falcon tube. Make up the volume to 30 ml with water. Sonicate using a large probe on ice-water at amplitude = 50% with pulse on = 5 sec/ off = 20 sec till the solution becomes visibily clear.

• 4 mM DM with 0.1 mg/ml POPS buffer (50 ml total): Transfer 0.5 mL of 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine (POPS, 10 mg/mL, in CHCl₃, Avanti Polar Lipids) into a Pyrex culture tubes (16 mm diameter, 10 cm high, round-bottom, Corning). Blow argon gas to obtain a dry lipid film. Add 2 mL of pentane to re-dissolve the lipid and dry again. Wrap the tube in aluminum foil with the mouth open in a vacuum desiccator overnight. Next day, add 1 mL water, cap the tube and bath-sonicate until the solution becomes translucent. Transfer the lipid solution with a glass pipette to a 50mL falcon tube containing 96.52 mg of DM. Add 9 mL of buffer and place the tube on rotator at room temperature until the solution becomes transparent. Bring volume to 50 mL with buffer.

• Solubilization buffer: (1X TBS supplemented with 0.1 μg/ml pepstatin A, 1 μg/ml leupeptin, 1 μg/ml aprotinin, 0.1 mg/ml soy trypsin inhibitor, 1 mM benzamidine, 0.1 mg /ml 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF) and 1 mM phenylmethysulfonyl fluoride (PMSF)and DNase.

• LB transposition plate composition: weight 5 gm N-Z-case (Sigma #N4642) +5 gm NaCl +2.5 gm yeast extract (RPI #20020). Make up the volume to 500 ml and autoclave. When cool to touch, add Kanamycin (50 ug/ml), Tetracycline (10 ug/ml), Gentamycin (7 ug/ml), X-gal (100 ug/ml), IPTG (0.5 mM) and plate. (store plates at 4°C for no more than 6 months).

• HEK suspension media: To 1L Hyclone CDM4HEK293 media add + 5 ml of Pen/Strp +25 ml of FBS +10ml of L-Glutamine.

• 1M Sodium Butyrate: dissolve 11.1 gm in 100 ml water. Filter with 0.22 μm sterile filters in a cell culture hood.

• Cell resuspension buffer: 50 mM TrisHCl, pH8.0; 500mM NaCl, 5 mM BME, 2 mM EDTA, 1 mM DTT, 10 mM MgCl2. Supplement with 0.1 μg/ml pepstatin A, 1 μg/ml leupeptin, 1 μg/ml aprotinin, 0.1 mg/ml soy trypsin inhibitor, 1 mM benzamidine, 0.1 mg /ml 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF) and 1 mM phenylmethysulfonyl fluoride (PMSF)and DNase.

• Membrane re-suspension buffer: 100mM TrisHCl, pH8.0; 400mM NaCl; 4 mM TCEP; 10 mM BME; 20% glycerol.

• DHHC15 - Solubilization buffer (SB): 40 mM HEPES, 300 mM NaCl, 2mM TCEP (adjust pH to 7.5). Add 0.1 μg/ml pepstatin A, 1 μg/ml leupeptin, 1 μg/ml aprotinin, 0.1 mg/ml soy trypsin inhibitor, 1 mM benzamidine, 0.1 mg /ml 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF) and 1 mM phenylmethysulfonyl fluoride (PMSF)and DNase.

• DHHC15 - Wash buffer-1 (WB1): 40 mM HEPES, 300 mM NaCl, 2mM TCEP pH 7.5, 0.2% DDM.

• DHHC15 - Wash buffer-2 (WB2): 40 mM HEPES, 300 mM NaCl, 2 mM TCEP, 20 mM Immidazole, pH 7.5, 0.1% DDM

• DHHC15 - Wash buffer-3 (WB3): 40 mM HEPES, 300 mM NaCl, 2mM TCEP, pH 7.5, 0.1% DDM.

• DHHC15 - PreScission Protease cleavage solution (PxpS): 40 mM HEPES, 300 mM NaCl, 2mM TCEP, pH 7.5, 0.1% DDM with 0.05 – 0.1 mg/ml PreScission protease

• DHHC15 - Size Exclusion Buffer (SEC): 40 mM HEPES, 150 mM NaCl, 2mM TCEP, pH 7.5, 0.1% DDM.

• PORCN - Solubilization buffer (SB): 100 mM HEPES, pH 7.9, 450 mm NaCl, 1 mM TCEP, protease inhibitors (0.1 μg/ml pepstatin A, 1 μg/ml leupeptin, 1 μg/ml aprotinin, 0.1 mg/ml soy trypsin inhibitor, 1 mM benzamidine, 0.1 mg /ml AEBSF), and DNase.

• PORCN - Wash buffer-1 (WB1): 25 mM Tris-HCl, pH 7.8, 350 mM NaCl, 1 mM TCEP, 4 mM DM, 20 mM imidazole, and 0.1 mg/ml POPS.

• PORCN - Wash buffer-2 (WB2): 25 mM Tris-HCl, pH 7.8, 350 mM NaCl, 1 mM TCEP, 4 mM DM, and 0.1 mg/ml POPS.

• PORCN - Size Exclusion Buffer (SEC): 25 mM HEPES, pH 7.5, 250 mM NaCl, 1 mM TCEP, 4 mM DM, and 0.1 mg/ml POPS

COMMENTARY

Background Information

Membrane proteins (MPs), owing to their low abundance and hydrophobic nature, are traditionally difficult to obtain in sufficient amounts for structural and functional studies. The choice of an expression host largely depends on the target protein. There are very few examples of mammalian MP being expressed in prokaryotic systems, because this mostly leads to protein misfolding and/or lower expression (Hattab et al., 2015). The most popular prokaryotic system E.coli, is predominantly used for large scale expression of bacterial membrane proteins and a few exceptions of eukaryotic proteins like GPCRs. E.coli strains C41, C43 (Miroux & Walker, 1996) and Lemo21 (Wagner et al., 2008) can tolerate toxic MP and accommodate fusion-tags like GFP, MBP, GST or Mistic (Roosild et al., 2005). Usually protein induction is under the control of IPTG or arabinose; alternatively leaky expression has also provided good quality MP (Noinaj et al., 2013). Eukaryotic hosts like yeast, insect and mammalian cells have proven better for eukaryotic MP expression. Yeasts like Saccharomyces cerevisiae and Pichia pastoris have found wide usage in the field (Gerngross, 2004). Proteins can be expressed in S. cerevisiae under different inducible promoters while methanol is used to induce proteins in P. pastoris (Parcej & Eckhardt-Strelau, 2003). Although S. cerevisiae is genetically better characterized, P. pastoris has gained more traction since cells grow to higher densities and dramatically increase the overall yield, especially important for structural studies.

Insect cell (Sf9 or Sf21) is an easier alternative to mammalian cell lines like HEK293, CHO or COS. They are simple to work with, use similar codons and provide comparable post translational modifications (Jarvis & Finn, 1995) to mammalian cells. Mammalian cell lines are broadly devided into two groups, 1) non-human: Chinese hamster ovary (CHO) cells, baby hamster kidney (BHK21) cells and murine myeloma cells (NS0 and Sp2/0) (Estes & Melville, 2014), and 2) human: human embryonic kidney 293 (HEK293) and fibrosarcoma HT-1080 cell lines (Beck, 2009; Casademunt et al., 2012). The major difference between the two mammalian cell types lie in their glycosylation profiles - CHO cell lines have found wide usage in the pharmaceutical industry, partly because of its early usage, ability to grow in suspension with a low serum predefined media, regulatory compliance and known contaminant profile. Moreover, a bulk of biotherapeutic proteins have lower to no glycosylations (Dumont et al., 2016). Membrane proteins have a higher propensity for glycosylation and thus are preferred to be expressed in human cell lines. For example, serotonin transporter (SERT) requires glycosylation for efficient folding (Tate & Blakely, 1994). HEK293T cells are adherent cells that contain an integrated SV40 large T antigen copy, that interacts with the SV40 origin of replication and transiently maintains extrachromosomal plasmid. HEK293S GnTi^- are suspension adapted cells with a deletion for N-acetylglucosaminetransferase I (GnTI^-) enzyme making it deficient for complex glycosylation (Reeves, Callewaert, et al., 2002). Protein expression in these cell lines can be achieved through transient transfection, stable transfection or viral transduction. Large scale protein expression through transient transfection using PEI or commercial transfection reagents are cumbersome and inefficient. Transient transfection can also be achieved through the use of Adenoviruses (Wright, 2009). Since the viral DNA does not integrate with the host genome, transient expression persists until the viral genome is degraded. Cell lines with stable integration of DNA into the host genome is more amenable to large scale protein production. Stable cell lines are selected on the basis of antibiotic resistance conferred from the integrating plasmid. Integration may be random or recombinase assisted. A major disadvantage is the time required to develop a monoclonal cell line (Chaudhary et al., 2012). Stable cell lines created through Lentiviral transduction have been shown to ameliorate this disadvantage (Elegheert et al., 2018). Baculovirus transduction of mammalian cells (BacMam) is a powerful technique to express proteins. We use the pEG BacMam vector, an optimized vector from Eric Gouaux’s lab for protein expression (Goehring et al., 2014). Typically, we append a Venus fluorescent tag in lieu of GFP in the pEG BacMam vector as we found improved expression of MPs when grown at 37°C (Rana et al., 2018). The MP gene is cloned into the vector and used to prepare bacmid DNA by transforming E. coli DH10Bac™. Sf9 insect cells are transiently transfected to produce baculovirus (P₁), and further used to infect more Sf9 cells to produce a larger viral stock with high titre (P₂). This P₂ viral stock is used to transduce HEK293S GnTi^- cells and allowed to express in the presence of HDAC inhibitor, sodium butyrate. Occasionally, expression of MP might be toxic to mammalian cells, in which case HEK293S GnTi^- -TetR cells (Reeves, Kim, et al., 2002) that are tetracycline-inducible and suspension adaptable might serve as better alternative.

Detergents are required for the isolation of MP from cellular membranes (Prive, 2007) The choice of detergent almost invariably dictates the success in MP purification. Detergents of the maltoside family, esp. DDM, are commonly used to extract proteins from various membranes. Typically, 1–2% of detergent should be sufficient for protein extraction from mammalian cells. The extraction detergent may not be the ideal detergent that retains a protein’s biochemical and biophysical properties. A thorough detergent screening is usually performed to identify the best detergent. The screening results can be judged either using FSEC (Kawate & Gouaux, 2006), if a fluorescent tag is appended to the MP, or an activity assay if available (Rich et al., 2009). Post extraction, the detergent is exchanged on either on IMAC or SEC column at a working concentration above their known CMC values. Supplementation of cholesterol or lipids, deemed essential for certain MP, should be included in the screening process.

Critical Parameters

1. Transient transfection

   It is imperative to start with the right protein coding sequence and hence best advised to sequence the DNA insert for truncation or mutations. If this is the first time working with pEG BacMam vector, it is also advisable to sequence the entire vector. Two parameters that affect transfection efficiency is the DNA concentration and ratio of DNA to PEI, both of which should be optimized. Commercial alternatives to PEI may be used if the transfection efficiency is poor. Expression of certain membrane proteins may be toxic to HEK293T cells and hence the duration of expression should be optimized. Since the outcome of transfection efficiency is determined through FSEC, it is important to identify a stable detergent for extraction and analysis.

2. Baculovirus preparation

   Transformation of DH10bac E. coli is slightly different from regular E. coli transformation, with longer incubation periods at 37°C. While isolating a recombinant bacmid, care should be taken to not shear the DNA. A good way to ensure the presence of an insert in the bacmid is by sequencing with either an internal primer or pUC/M13 forward and reverse primers (Invitrogen, 2015). After obtaining white colonies in the transposition plates, it is important to restreak onto another plate. Sometimes though a single white colony was picked, certain blue colonies still get carried along. Either carefully pick the white colonies or restreak again. Optimum production of P₁ baculovirus can be monitored by placing a GFP gene under the P₁₀ promoter, in the pEG BacMam vector, and observing visual fluorescence. P₁ is a small volume, low titre stock, while P₂ has higher titre. If need be, a third P₃ viral stock could also be prepared if the titre for P₂ is not high enough and/or larger volumes of viral stocks are required. Viral titres can also be estimated with Endpoint dilution assay using Sf9 Easy Titre cells (Hopkins & Esposito, 2009). Sf9 ET cells are grown in similar media as Sf9 except they also contain Geneticin G418 along with an optional supplementation of 5% FBS.

3. Mammalian cell transduction

   Prior to initiating large-scale membrane protein expression, it is important to check the quality of viruses using HEK293T cells at a smaller scale. Addition of HDAC inhibitors like sodium butyrate drastically affects protein expression. It is recommended to perform a pilot scale expression study by varying HDAC inhibitor concetration and the duration of incubation using FSEC. An ideal read out would be a monodisperse population with high yield.

4. Protein purification

   MP purification is protein specific and too extensive to be covered within the scope of this protocol. We would defer the reader to other reviews (Baneres et al., 2007; Mancia & Love, 2010) for a more thorough understanding of the process. Broadly, the best detergent should be identified by performing a broad detergent screen along with determining the protein’s Tm (melting temperature) in different detergents (Hattori et al., 2012). 1–2% detergent is usually enough for cell lysis, with DDM being the most popular detergent used at this stage. Typically, lysis is followed by affinity chromatography where the detergent can be gradually exchanged. During this step it is important to keep the detergent concentration a few orders above its CMC. It would be prudent to check the activity in an assay, if available, with different detergents to confirm the detergent screening results. Some membrane proteins remain bound to essential lipids or cholesterol for their enhanced stability or activity. Lipids and cholesteryl hemisuccinate (CHS, a soluble derivative of cholesterol) can be supplemented in the buffer at an operating concentration of 0.05–0.1mg/ml and 0.01–0.02% respectively. However the optimal concentrations should be determined empirically. In the final SEC step, the detergent concertation is lowered anticipating an undesirable enrichment of empty micelles during the ensuing concentration step that might hinder crystal formation in X-ray crystallography (Newby et al., 2009) or impede formation of thin ice on cryoEM grids (Hauer et al., 2015)

Troubleshooting

Cell culture is very prone to contamination, mostly due to lack of sterile practices. Spray biosafety cabinets/hoods with 70% Ethanol and wipe clean. Once a week, wipe the hoods with a commercial hood cleaning reagent. Routinely check cells for mycoplasma contamination. Use MRA (Mycoplasma Removal Agent) or Plasmocin for removal of mycoplasma contamination from cultures if discarding the culture is not desired. Contaminated suspension cells have a putrid smell and turns cloudy, discard immediately.

1. Transient transfection

   HEK293T cells should be passaged every 3–4 days at 80% confluency to ensure cells are healthy and in good condition to ensure efficient transfection.

   Low transfection efficiency: Optimize the ratio of DNA to PEI ratio. Alternatively, use lipofectamine or electroporation to improve transfection efficiency. The duration of expression should also be optimized. Certain membrane proteins might not express well with an appended GFP tag, in which case protein expression can be monitored on an SDS-PAGE gel with western blotting probing the polyhistidine affinity tag.

2. Baculovirus preparation

   Sf9 insect cells should be passaged every 3–4 days when they reach a density of 4–5 ×10⁶ cells/ml.

   Low P₁ titre: Transfect Sf9 with different ratios of bacmid to PEI. Optimize the days of incubation. Ensure bacmid DNA is not sheared.

   Low P₂ titre: Infect Sf9 with lower MOI of P₁. Usually lower MOI (0.01 – 0.001) have shown good results. If all fails, remake P₁ virus.

3. Mammalian cell transduction

   HEK293S GnTi^- cells should be passaged every 3–4 days when they reach a density of 3–4 ×10⁶ cells/ml.

   Low expression: HEK293S GnTi^- cells passaged from a primary stock that was overgrown >6×10⁶ cell/ml will show lower transduction and slower growth. Start a fresh stock. Avoid using cells that have a higher passage number, usually not more than 30. Optimize the concentration of sodium butyrate added post transduction. Try expression at lower temperatures.

4. Protein purification

   Low yield: Check for proper cell lysis and empirically determine the optimal detergent concertation and duration of incubation in the extraction step. Increase the pH during resin binding. Optimize the time to bind the affinity resin and lower imidazole concetration in the wash step. Addition of glycerol may improve protein stability and consequently improve yield.

   Impurities: Bind to the resin at lower pH or with low concentration of imidazole in the presence of high salt. Wash with a higher concentration of imidazole or salt during the affinity purification steps. Alternatively, purify from isolated membranes (see SUPPORT PROTOCOL 2). Use of orthogonal affinity tags or nanobody (Gotzke et al., 2019; Virant et al., 2018) resins could also ameliorate problems with purity.

Understanding Results

The above protocols were used to express and purify two human membrane proteins hDHHC15 and hPORCN. These results should serve as a primer towards accessing the various subsections of this protocol. Since we appended a fluorescent tag (mVenus) to our membrane protein, we could visually monitor its expression in HEK293T cells using an epifluorescence microscope (Fig. 1a). These cells were lysed and analyzed through FSEC and SDS-PAGE in-gel fluorescence (Fig. 1b,c). A single band in SDS-PAGE corresponding to a combined molecular weight of the fluorescently tagged protein with no additional band for free mVenus indicates a direct correlation between the observed fluorescence intensity and total protein expression. FSEC chromatogram indicates successful detergent extraction of the tagged protein, in the detergent of choice, along with its heterogeneous distribution or oligomeric states. After checking for expression, bacmids were prepared as per the protocol and used to make P₁ and P₂ viral stocks from Sf9 cells. Viral titres were estimated using a virus counter, alternatively Endpoint dilution assay with Sf9 easy titre cells could also be used (Hopkins & Esposito, 2009). Typical titres range in 1E⁶-E⁷ vp/ml for P₁ and 1E⁸-E⁹ vp/ml for P₂. In addition to the traditional method for generating P2 viral stocks we developed a direct “Enumeration-free” method that is devoid of viral titre estimation. As proof of principal, we used hDHHC15 to generate P2 viruses using different dilution of P₁ as outlined in ALTERNATE PROTOCOL. HEK293T cells were tranduced with P₂ viruses and analysed using FSEC (Fig. 2). Interstingly, higher dilutions led to a perceptible reduction in aggregates (~7 ml) with no significant difference in the amount of extracted protein (~13 ml). At lower dilution (400 μl) the amount of extracted protein reduced with increased aggregation. The two peaks at ~11 and 13 ml are indicative of dimer and monomer. In trying to maintain the BASIC PROTOCOL continuum, P₂ viruses were prepared as per BASIC PROTOCOL 2. Before tranducing large-scale HEK293S GnTi^- cells, the potency of P₂ virus was determined at a small scale by transduction of adherant HEK293T cells and monitored through epifluorescence microscopy (Fig. 3a, left panel). We then proceeded to transduce large scale suspension culture of HEK293S GnTi^- cells with P₂ virus. Prior to harvesting, the final protein expression was observed under the microscope (Fig. 3a, right panel). GnTi^- cells displayed enhanced expression due to the post-transduction addition of sodium butyrate. hDHHC15 was extracted from the harvested cells with 1.5% DDM and IMAC purified with 0.1% DDM. The eluate obtained from on-resin PreScission clevage was concentrated and separated through size exclusion chromatography (SEC). The protein predominantly purified as a dimer as has been previously observed (Rana et al., 2019). Individual peak fractions of the purified protein from SEC were analyzed through SDS-PAGE using Coomassie brilliant blue (Fig. 3b, top panel). 1.5% DDM and 0.1% CHS was used for solubilizing hPORCN from the harvested cells. The solubilized hPORCN was purified by IMAC in 4 mM DM with 0.1 mg/ml POPS. The PreScission cleaved protein was concentrated and subsequently separated by SEC in the presence of 4 mM DM with 0.1 mg/ml POPS. Target fractions were analyzed on an SDS‐PAGE gel to estimate the purity of the purified protein (Fig. 3b, bottom panel).

Figure 1. Expression of mVenus fused hDHHC15 and hPORCN in HEK293T cells:

a) Fluorescence microscopy image of HEK293T cell expressing the fusion protein through transient transfection. FSEC profiles (b) and SDS-PAGE in-gel fluorescence (c) of detergent extracted cell lysate.

Figure 2. Enumeration-free method for generation of P₂ viral stock.

Overlay of FSEC profiles from HEK293T cells transduced with P₂ viruses, produced from different dilutions of P₁ virus.

Figure 3. Expression and purification of hDHHC15 and hPORCN transduced with P₂ virus.

a) Fluorescence microscopy image of small scale expression in HEK 293T cells; and large scale expression in HEK 293S GnTi^- cells, enhanced with sodium butyrate. b) SEC chromatograms of hDHHC15 and hPORCN obtained from Superdex S200 column. Coomassie-blue stained SDS-PAGE of peak fractions shown as inset.

Time Considerations

Protein expression using the BacMam system is time intensive. It is imperative that expected results from intermediate steps/protocol be assessed between successive steps. While estimating time requirements, we assume there is an ongoing culture of Sf9, HEK293S GnTi^- and HEK293T cells. If not, an additional 1.5 weeks should be added to the total time. BASIC PROTOCOL 1 (BP1) for transient transfection and FSEC analysis will require 3–4 days. BP2 for generating baculovirus will take 11 days, provided there is an ongoing culture of Sf9 cells with the required cell density available for making P₂ viral stock. BP3 for expressing cells in HEK293S GnTi^- will take 4 days, provided there is an ongoing large culture of cells readily available. Purification of membrane protein (BP4) can be achieved in 1–2 days. The entire process should take ~3 – 3.5 weeks in toto.

Significance Statement.

We provide here a complete protocol delineating the expression and purification of integral membrane proteins from, Baculovirus transduced, mammalian HEK293S GnTi^- cells using Venus as a fluorescent fusion tag. Particularly for mammalian membrane proteins, the protocols herewith will serve as a primer for obtaining correctly folded protein with native or near-native post translational modification. Utilizing the techniques delineated here, one should be able to readily produce sufficient quantities of well-behaved recombinant protein or their complexes for cryoEM and X-ray crystallography.