Reconstitution of Msp1 Extraction Activity with Fully Purified Components

Abstract

As the center for oxidative phosphorylation and apoptotic regulation, mitochondria play a vital role in human health. Proper mitochondrial function depends on a robust quality control system to maintain proteostasis. Declines in mitochondrial proteostasis have been linked to cancer, aging, neurodegeneration, and many other diseases. Msp1 is a recently discovered AAA+ ATPase that maintains mitochondrial proteostasis by removing tail-anchored membrane proteins from the outer mitochondrial membrane. Using purified components reconstituted into proteoliposomes, we have shown that Msp1 is necessary and sufficient to extract a model tail-anchored protein from a lipid bilayer. Our simplified reconstituted system overcomes several of the technical barriers that have hindered detailed study of membrane protein extraction. Here, we provide detailed methods for the generation of liposomes, membrane protein reconstitution, and the Msp1 extraction assay.

SUMMARY:

Detailed protocol for reconstitution of Msp1 extraction activity with fully purified components in defined proteoliposomes.

INTRODUCTION:

Proper cellular function depends upon a process called proteostasis, which ensures that functional proteins are at the correct concentration and cellular location¹. Failures in proteostasis lead to compromised organelle function and are associated with many neurodegenerative diseases^2–4. Membrane proteins present unique challenges to the proteostasis network as they must be targeted to the correct membrane and overcome substantial thermodynamic barriers to enter and exit the lipid bilayer, all while avoiding the formation of potentially toxic aggregates in the cytosol⁵. Consequently, specialized machinery has evolved to shield the hydrophobic transmembrane domain (TMD) from the cytosol and facilitate targeting and insertion into the proper cellular membrane^6–15.

Mitochondria are the metabolic hub of the cell and are involved in numerous essential cellular processes such as: oxidative phosphorylation, iron-sulfur cluster generation, and apoptotic regulation^16,17. These endosymbiotic organelles contain two membranes, referred to as the inner mitochondrial membrane (IMM) and the outer mitochondrial membrane (OMM). Over 99% of the 1500 human mitochondrial proteins are encoded in the nuclear genome and need to be translocated across one or two different membranes^18,19. Proper mitochondrial function thus depends on a robust proteostasis network to correct any errors in protein targeting or translocation.

Our lab focuses on a subset of mitochondrial membrane proteins called tail-anchored (TA) proteins, which have a single transmembrane domain at the very C-terminus^20–24. TA proteins are involved in a number of essential processes, such as apoptosis, vesicle transport, and protein translocation²⁵. The unique topology of TA proteins requires post-translational insertion, which occurs in the endoplasmic reticulum (ER) by the GET or EMC pathways or into the OMM by a poorly characterized pathway^20,26–28. The biophysical properties of the transmembrane domain (TMD) are necessary and sufficient to guide TA proteins to the correct membrane²⁹. The recognition of biophysical characteristics rather than a defined sequence motif limits the fidelity of the targeting pathways⁵. Thus, mislocalization of TA proteins is a common stress for the proteostasis networks. Cellular stress, such as inhibition of the GET pathway, causes an increase in protein mislocalization to the OMM and mitochondrial dysfunction unless these proteins are promptly removed^30,31.

A common theme in membrane proteostasis is the use of AAA+ (ATPase Associated with cellular Activities) proteins to remove old, damaged, or mislocalized proteins from the lipid bilayer^1,32–38. AAA+ proteins are molecular motors that form hexameric rings and undergo ATP dependent movements to remodel a substrate, often by translocation through a narrow axial pore^39,40. Although great effort has been devoted to studying the extraction of membrane proteins by AAA+ ATPases, the reconstitutions are complex or involve a mixture of lipids and detergent^41,42, which limits the experimental power to examine the mechanism of substrate extraction from the lipid bilayer.

Msp1 is a highly conserved AAA+ ATPase anchored in the OMM and peroxisomes that plays a critical role in membrane proteostasis by removing mislocalized TA proteins^43–47. Msp1 was also recently shown to alleviate mitochondrial protein import stress by removing membrane proteins that stall during translocation across the OMM⁴⁸. Loss of Msp1 or the human homolog ATAD1 results in mitochondrial fragmentation, failures in oxidative phosphorylation, seizures, increased injury following stroke, and early death^31,49–56.

We have shown that it is possible to co-reconstitute TA proteins with Msp1 and detect the extraction from the lipid bilayer⁵⁷. This simplified system uses fully purified proteins reconstituted into defined liposomes which mimic the OMM (Figure 1)^58,59. This level of experimental control can address detailed mechanistic questions of substrate extraction that are experimentally intractable with more complex reconstitutions involving other AAA+ proteins. Here, we provide experimental protocols detailing our methods for liposome preparation, membrane protein reconstitution, and the extraction assay. It is our hope that these experimental details will facilitate further study of the essential but poorly understood process of membrane proteostasis.

Figure 1:

Cartoon of extraction assay and list of key steps.

PROTOCOL:

1. Liposome Preparation

1.1.

Combine chloroform stocks of lipids in appropriate ratios to mimic the outer mitochondrial membrane.

1.1.1.

To mimic the outer mitochondrial membrane, we use a 48:28:10:10:4 molar ratio of chicken egg phosphatidyl choline (PC, Avanti #840051C), chicken egg phosphatidyl ethanolamine (PE, Avanti #840021C), bovine liver phosphatidyl inositol (PI, Avanti #840042C), synthetic 1,2-dioleoyl-sn-glycero-3-phospho-L-serine (DOPS, Avanti #840035C), and synthetic 1’,3’-bis[1,2-dioleoyl-sn-glycero-3-phospho]-glycerol (TOCL, Avanti #710335C)^58,59. We prepare 25 mg of liposomes at a time. Sample calculations are shown in Table 1.

Table 1:

Sample calculations for liposome preparation. The key goals of this table are to calculate the concentration corrected average molecular weight of the lipid mixture (cell E8) and the volume of each lipid stock required to make the liposomes (column H). Columns A, B, C, and G are either taken from product labels (MW or stock concentration) or determined by user (lipids and mole %).

Lipid	Mole %	MW	Avg. MW	umol in 25 mg	mg in 25 mg	Chlor stock (mg/mL)	uL for 25 mg
PC	48%	770	369.6	14.82	11.41	25	456.4
PE	28%	746	208.88	8.64	6.45	25	258.0
PI	10%	902	90.2	3.09	2.78	10	278.5
DOPS	10%	810	81	3.09	2.50	10	250.1
TOCL	4%	1502	60.08	1.23	1.85	25	74.2
			Conc. Corr Avg MW	809.76
			umol in 25 mg	30.87

1.1.2.

Bring all lipid stocks to room temperature before opening as this will limit condensation. As most labs do not have a precise way to measure the concentration of the lipids, any water absorbed by the chloroform stock will change the concentration of the lipid stock and thus the ratio of lipids used in the assay.

1.1.3.

Lipid stocks come in glass ampules. Transfer the required amount of lipid to a glass vial (VWR, #60910L-1) using a 1 mL syringe (Norm-ject, #53548-001). The vial should contain 2 mg of Dithiothreitol (DTT) to prevent lipid oxidation. Work quickly as evaporation of chloroform will change the concentration of the lipid.

1.1.4.

Any remaining lipid can be transferred to a separate glass vial and fitted with a PFTE septa. The vial should contain 2 mg of DTT to prevent lipid oxidation. Wrap the vial with parafilm and store at −20° C. Try to use the lipids within 3 months of transfer to the vials. To prevent potential contamination of the chloroform stocks by marker runoff, we prefer to transfer the stickers from the original ampules to the glass vials rather than label with marker.

1.2.

Evaporate chloroform under a very gentle stream of nitrogen while spinning the glass vial continuously. Essentially, you become a manual rotovap. The goal is to evaporate all of the chloroform and get an even coating of lipids over the entire glass vial.

1.2.1.

Attach a fresh pasture pipette to the nitrogen tube. Be careful to not let any of the lipids splash out of the vial or onto the pasture pipette. Aim the tip at the bottom of the vial so the air bounces off the bottom and pushes the lipids up towards the center of vial. Try to get an even coating over the entire vial while avoiding any accumulation in the corners or by the cap. This whole process takes about 5 minutes.

1.2.2.

As the mixture thickens into a “bead” of lipids, guide it into the center of the vial by changing the angle of the vial. Once the bead starts to become smaller, turn up the nitrogen stream slightly to disperse the bead, being careful to ensure that none of the lipids blow out of the vial.

1.3.

Remove any remaining chloroform under vacuum.

1.3.1.

Put the glass vial on a house vacuum or diaphragm vacuum pump for 1 hour to remove the majority of residual chloroform. These vacuums are generally not strong enough to remove all of the chloroform, but they can tolerate small amounts of solvent better than rotary-vain vacuums.

1.3.2.

Put vial on a strong vacuum (<1 mTorr) for 12–16 h to remove all residual chloroform. Be sure to avoid bumping of the vial during this process.

1.4.

Resuspend the lipids in 1.25 mL of liposome buffer (50 mM Hepes KOH pH 7.5, 15% glycerol, 1 mM DTT). As we started with 25 mg of lipids, this results in a concentration of 20 mg/mL. The lipids should be fully resuspended with no visible chunks. If the lipids were not evenly coated over the glass vial or pooled in the corner of the vial, this can be a lengthy process.

1.4.1.

Vortex the vial vigorously until the sample is milky smooth. We find that if the chloroform evaporation was done properly the night before, this process takes about 5–10 minutes.

1.4.2.

To ensure complete resuspension of the lipids, rotate on a wheel at room temperature for 3 hours. Remove vial from the wheel once every hour for 1 minute of vortexing to ensure even mixing.

1.5.

Transfer lipids carefully to a clean 1.5 mL microcentrifuge tube. Perform 5 freeze-thaw cycles using liquid nitrogen to freeze and a 30°C heat block to thaw. This step helps to convert multilamellar vesicles to unilamellar vesicles.

1.6.

Extrude the lipids.

1.6.1.

During the freeze thaw cycles, prepare the mini-extruder (Avanti #610020). Assemble the mini-extruder with filter supports (Avanti, #610014) and 200 nM polycarbonate membrane (Avanti, #610006). Detailed instructions for the correct assembly of the mini-extruder is available from Avanti. The size of the filter will affect the size of the liposome, which will in turn affect the concentration of proteins required for the reconstitution (Step 2.3).

1.6.2.

Place the mini-extruder onto a hot plate and bring the extruder temperature up to 60° C.

1.6.3.

Draw up the lipids into a gas-tight syringe and carefully place into one end of the mini-extruder. Place the empty gas-tight syringe into the other side of the mini-extruder. Allow the lipids to equilibrate to the extruder stand temperature for 5–10 minutes.

1.6.4.

Gently push the plunger of the filled syringe until the lipids are completely transferred to the alternate syringe. Push the solution from the alternate syringe into the original syringe. Repeat this back and forth process 15 times, so that at the 15^th pass the lipids end in the alternate syringe. Monitor the volume in each pass to make sure that there are no leaks.

1.7.

Prepare single use aliquots of the lipids, flash freeze in liquid nitrogen and store at −80° C. The liposomes are stable at −80° C for several months. Our reconstitutions require 10 μL of lipid at a time (Step 2.3.4), so we find it convenient to prepare 10 μL or 20 μL aliquots.

2. Reconstitution of Msp1 and Model TA protein

2.1.

Prepare the reconstitution buffer: 50 mM Hepes pH 7.5, 200 mM Potassium Acetate, 7 mM Magnesium Acetate, 2 mM DTT, 10% Sucrose, 0.01% Sodium Azide, 0.2–0.8% Deoxy Big Chaps (DBC).

2.1.1.

Optimize reconstitution conditions for your new batch of liposomes. In our experience, the concentration of DBC and biobeads required for optimal reconstitution varies depending on the batch of liposomes used. To limit prep to prep variability, we use the same lot of DBC for all experiments. When changing lots of DBC, we repeat the optimization process.

2.1.2.

We suggest setting up a series of reconstitutions with different concentrations of DBC (0.2% – 0.8%) and biobeads (25 mg – 100 mg) each time a new batch of liposomes is prepared. It is important to not drop the DBC below the critical micelle concentration (~0.15%). Once conditions are optimized, we recommend collecting all data using the same liposome prep and reconstitution conditions.

2.1.3.

Assay the effectiveness of the various reconstitution conditions by using the extraction assay described in Step 3.

2.2.

Prepare biobeads (Bio-Rad, #1523920) at a final concentration of 250 mg/mL.

2.2.1.

Weigh out 2.7 g of dried biobeads and resuspend in a 50 ml falcon tube of 100% methanol to wet the beads. The biobeads must be initially wetted in methanol to prevent air from being trapped in the pores of the beads. Once in methanol, it is essential that the biobeads are kept wet as any air trapped by the biobeads will alter their ability to absorb detergent.

2.2.2.

Remove methanol by washing the beads 8x with ultrapure water (18.2 mΩ), hereafter referred to as ddH₂O. Pellet beads by spinning at 3200 × g for 1 minute. Decant the liquid and resuspend in ddH₂O.

2.2.3.

After washing, resuspend in 10 ml of ddH₂O with 0.02% NaN₃ and store at 4° C. Biobeads can be stored at 4° C for several months. This stock is 250 mg/ml as it is assumed ~0.2 g is lost during the wash steps.

2.3.

Calculate the size of your liposome and the desired number of molecules of TA protein and Msp1 per liposome. This will determine the concentration of Msp1 and TA protein required for the reconstitution.

2.3.1.

First, calculate the number of lipid molecules per unilamellar liposome (N_total) using the equation $N_{total}=\frac{\left[4\pi {\left(\frac{d}{2}\right)}^2+4\pi {\left[\frac{d}{2}-h\right]}^2\right]}{a}$ where d is the diameter of the liposome, h is the thickness of the bilayer, and a is the lipid headgroup area.

2.3.1.1.

We measure the liposome diameter by DLS. In our example, we obtained a value of 70 nm for the liposome diameter (d).

2.3.1.2.

We use a value of 5 nm for h and 0.71 nm² for a, which is the headgroup size for phosphatidylcholine. In our particular situation, N_total is 37,610.

2.3.2.

Next, calculate the molar concentration of lipid M_Lipid using the average molecular weight of the lipids in your mixture. In our example, the concentration of lipids is 20 mg/mL (Step 1.4) and the average molecular weight of lipids is 810 g/mol (Table 1). This results in an M_Lipid value of 0.0247 M.

2.3.3.

Next, calculate the molar concentration of liposomes M_Liposome using the equation $M_{Liposome}=\frac{M_{Lipid}}{N_{Total}}$ where M_Lipid is the molar concentration of lipid from Step 2.3.2, and N_total is the total number of lipids per liposome calculated in Step 2.3.1. In our example, our 20 mg/mL stock concentration of liposomes is approximately 660 nM.

2.3.4.

Calculate the amount of Msp1 and TA protein required for a 100 μL reconstitution reaction.

2.3.4.1.

The final concentration of lipids in the reconstitution is 2 mg/mL, which is a 10x dilution of the liposome stock. This gives a final liposome concentration of 66 nM.

2.3.4.2.

The final concentration of Msp1 is 792 nM, which gives 12 total copies (2 functional hexamers) per liposome.

2.3.4.3.

The final concentration of TA protein is 660 nM, which gives 10 copies per liposome.

2.4.

In a PCR tube, mix together purified Msp1, TA protein, and liposomes in reconstitution buffer. The order of addition should be buffer, proteins, and liposomes last. The total volume is 100 μL. Allow the mixture to sit on ice for 10 minutes. The Msp1 and TA protein were purified as previously described⁵⁷.

2.5.

Add the desired amount of biobeads to the sample to remove detergent.

2.5.1.

Cut the tip of a p200 pipette tip to about 1/8^th inch in diameter so that beads can fit through the tip. Vortex the biobeads tube thoroughly to obtain a uniform mixture and quickly remove the lid and pipette up your volume before the biobeads settle. Transfer the biobeads to an empty PCR tube.

2.5.2.

When the reconstitution has finished its 10-minute incubation on ice, use an uncut pipette tip to remove all of the liquid from the biobeads. Then transfer the 100 μL reconstitution into the tube with the biobeads. This must be done quickly so that the biobeads do not trap air.

2.6.

Allow the reconstitution to rotate on a wheel for 16 hours at 4°C.

2.7.

Remove reconstituted material from biobeads. Do a quick spin in a picofuge to pellet the biobeads, then use an uncut pipet tip to transfer the reconstituted material to a clean PCR tube. Repeat this process 1–2 times until there are no biobeads left in the sample. Keep reconstitution on ice.

2.8.

Pre-clear the reconstituted material to remove any proteins which failed to reconstitute into the liposomes.

2.8.1.

Prepare Extraction Buffer: 50 mM Hepes pH 7.5, 200 mM KAc, 7 mM MgAc, 2 mM DTT, 100 nM CaCl₂.

2.8.2.

Equilibrate the glutathione spin columns (Thermo Fisher, #PI16103) with Extraction Buffer according to the manufacturer’s directions.

2.8.3.

Add 5 μM of each chaperone (GST-SGTA and GST-Calmodulin) to the reconstituted material. These chaperones will bind to the TMD of any proteins which failed to reconstitute into the liposomes. Purification of chaperones was described previously^6,57.

2.8.4.

Add 100 μL of extraction buffer to the reconstituted material, bringing the volume up to 200 μL. Add this to the equilibrated glutathione spin columns. NOTE that the glutathione spin columns provide the highest sample recovery when the pre-clearing volume is 200 – 400 μL.

2.8.5.

Plug the spin columns and rotate at 4°C for 30 minutes to allow chaperones to bind to resin.

2.8.6.

Spin the columns at 700 × g for 2 minutes. The flow through is your pre-cleared material that is depleted of aggregated proteins. Keep material on ice and proceed directly with extraction assay.

3. Extraction Assay

3.1.

Prepare tubes for SDS PAGE analysis. Each reaction will have 4 tubes: INPUT (I), FLOW THROUGH (FT), WASH (W), and ELUTE (E).

3.1.1.

Add 45 μL of ddH₂O sample to the INPUT tube, 40 μL of ddH₂O to the FLOW THROUGH tube, and 0 μL to the WASH and ELUTE tubes. NOTE: We find that we get the best signal to noise ratio in our assay if the WASH and ELUTE samples are 5x concentrated relative to the INPUT and FLOW THROUGH samples. Due to the dilutions during the assay, this requires taking 5 μL of sample for the INPUT sample, 10 μL of sample for the FLOW THROUGH sample, and 50 μL of sample for the WASH and ELUTE samples.

3.1.2.

Add 16.6 μL of 4x SDS PAGE Loading Buffer to each tube. The total volume of each sample is 50 μL before SDS PAGE Loading Buffer. We use 4x SDS PAGE Loading Buffer in our lab, so the final volume is 66.6 μL (50 μL sample + 16.6 μL of 4x SDS PAGE Loading Buffer).

3.2.

Assemble the extraction assay.

3.2.1.

The extraction reaction contains 60 μL of pre-cleared proteoliposomes, 5 μM of GST-SGTA, 5 μM of GST-Calmodulin, and 2 mM ATP. Combine all reagents except ATP, which is used to initiate the reaction. Bring to a final volume of 200 μL with Extraction Buffer. NOTE: As 60 μL of sample are used for each extraction assay, one reconstitution can be used for three different extraction assays. We find that it is best to perform positive and negative controls (+ATP and -ATP) on material from the same reconstitution.

3.2.2.

Pre-warm extraction assay in 30°C heat block for 2 minutes.

3.3.

Initiate the extraction assay by adding ATP to final concentration of 2 mM and start timer.

3.3.1.

Give a 5 second spin in a picofuge to mix ATP into the reaction. Incubate reaction at 30° C for 30 minutes.

3.3.2.

During the incubation, take 5 μL of the reaction and add to the INPUT tube. The timing of this is flexible.

3.3.3.

During this incubation period, equilibrate one glutathione spin column for each sample in the extraction assay.

3.4.

Perform pull down on chaperones to isolate extracted material.

3.4.1.

Once the 30-minute incubation is finished, add 200 μl of extraction buffer to the tube to bring total volume to 400 μL. Add to equilibrated glutathione resin and allow to bind on wheel at 4°C for 30 minutes.

3.4.2.

Spin the columns at 700 × g for 2 minutes to collect the flow through. Take 10 μL for the FLOW THROUGH tube. This sample should contain substrates which are still integrated in the lipid bilayer.

3.4.3.

Wash resin twice with 400 μL of extraction buffer, discarding the flow through. On the third wash, keep the flow through and take 50 μL for the WASH tube.

3.4.4.

Prepare 5 mL of Elution Buffer by adding reduced glutathione to a final concentration of 5 mM in Extraction Buffer. Prepare this buffer fresh each time.

3.4.5.

Add 200 μL of Elution Buffer to the spin column. Incubate at room temperature for 5 minutes. Spin at 700 × g for 2 minutes to elute. Keep the flow through. Repeat the process a second time so that the total elution volume is 400 μL.

3.4.6.

Take 50 μL of sample from the Elution sample and add it to the ELUTE tube.

3.5.

Analyze extraction activity using SDS-PAGE and Western blot. NOTE as a western blot is a fairly standard procedure, we are only providing a basic protocol that highlights a few of the details unique to our assay.

3.5.1.

Load samples into a stain free polyacrylamide gel (4% stacking, 15% separating) and run at 200 V for 50 minutes in tris-glycine buffer. If space permits, use both an unstained and stained ladder to permit visualization on stain-free gel imager and transfer to PVDF membrane respectively.

3.5.2.

The stain free gel should be imaged to confirm that there is equal loading across all samples. This is an essential control that ensures any changes in signal detected by western blot are not a result of variable protein loading. You should only see bands for the chaperones (GST-calmodulin and GST-SGTA) in the INPUT and ELUTION samples. Recall that the ELUTE sample should be more concentrated than the INPUT sample.

3.5.3.

Assemble a Western blot cassette using a 45 um PVDF membrane (Thermo Fisher #88518). Transfer at a constant current of 300 mA for 60 minutes.

3.5.4.

After blocking the membrane, bind to the primary antibody for 16 h at 4° C with gentle shaking. Blot for substrate with rabbit Anti-FLAG (Sigma-Aldrich, F7245-.2MG) at a 1:1,000 dilution. NOTE: the primary and secondary antibodies used will be substrate specific and the concentration for use may need to be optimized.

3.5.5.

Wash membrane and incubate with secondary antibody, goat anti-rabbit (Thermo Fisher, #NC1050917) at a 1:10,000 dilution, with gentle shaking for 1 hour at room temperature.

3.5.6.

Wash membrane and image for analysis using western blotting detection agent, such as ECL Select Western Blotting Detection Reagent (GE Healthcare, #RPN2235).

REPRESENTATIVE RESULTS:

To properly interpret the results, the stain free gel and the western blot must be viewed together. The stain free gel ensures equal loading across all samples. When viewing the stain free gel, the chaperones (GST-calmodulin and GST-SGTA) will be visible in the INPUT (I) and ELUTE (E) lanes. Double check that the intensity of these bands is uniform across all of the INPUT samples. Likewise, the intensity should be uniform across the ELUTE samples. Note that the ELUTE is 5x more concentrated than the INPUT. This difference in intensity should be visible in the gels.

After using the stain free gel to confirm proper loading, examine the western blot to determine extraction activity. We measure extraction activity by comparing the amount of substrate in the ELUTE (E) fraction relative to the INPUT (I) fraction. The signal in the Flow Through (FT) shows some variability, but is generally similar to the INPUT fraction. There should be no signal in the WASH (W) fraction. We typically see ~10% extraction efficiency for our positive control and 1–2% extraction efficiency in our negative control (Figure 2). Recall that the ELUTE fraction is 5x as intense as the INPUT fraction, so this needs to be taken into account when judging extraction efficiency. If reconstitution conditions are not optimized, we will typically see comparable extraction levels in both our + ATP and - ATP samples (Figure 3). We attribute this result to a failure of Msp1 to efficiently reconstitute, resulting in numerous proteoliposomes without a functional Msp1 hexamer.

Figure 2:

Representative data showing a properly functioning assay. Extraction efficiency is determined by comparing the amount of substrate in the ELUTE fraction with the INPUT fraction. Recall that the gel has 5x higher loading of the ELUTE fraction relative to the INPUT fraction.

Figure 3:

Representative data of a failed reconstitution and extraction assay. Here, the activity in the + ATP sample is comparable to the activity in the – ATP sample.

DISCUSSION:

Proper mitochondrial function depends upon a robust protein quality control system. Due to inherent limits in the fidelity of the TA protein targeting pathways, mislocalized TA proteins are a constant stress for mitochondria. A key component of the mitochondrial proteostasis network is Msp1, which is a membrane anchored AAA+ ATPase that removes mislocalized TA proteins from the OMM. Here, we have described how to prepare proteoliposomes, co-reconstitute Msp1 and a model TA protein, and perform an extraction assay. We previously used this assay to demonstrate that Msp1 directly recognizes mislocalized TA proteins and is capable of extracting these proteins from a lipid bilayer without any accessory proteins or cofactors⁵⁷.

A drawback of the assay is that there is some variability in the reconstitution process. To control for prep-to-prep variability, we always include a positive and negative control on the same gel/western blot to ensure that our assay is working as intended. We avoid making comparisons between reconstitutions that were done on different days or drawing comparisons between different western blots. The only comparisons we make are for samples reconstituted in parallel and run on the same gel/western blot. We are also quite conservative in our data interpretation. Although it is possible to quantify extraction efficiency using ImageJ, we typically describe our experiments as having full activity, intermediately activity, or no activity.

One source of variability is the total amount of protein reconstituted. While the majority of unincorporated proteins are removed by the pre-clearing process, the less than perfect reconstitution efficiency of Msp1 can have an outsized effect on the observed efficiency of substrate extraction. This effect arises from the fact that Msp1 functions as a homohexamer, but purifies as a monomer⁵⁷. Liposomes containing anything other than 6x copies of Msp1 will be inactive. For example, a liposome with only 5 copies of Msp1 will be inactive. The only way to form stable full-length Msp1 hexamers is to inactivate ATPase activity with non-hydrolyzable ATP analogs (ATPγS) or the inactivating E193Q Walker B mutant, neither of which are compatible with an activity assay. Overcoming this technical hurdle is an area of active research in our lab.

Another area of active research focuses on making the extraction assay more quantitative. The current method relies on pull downs and western blotting for signal detection, both of which are only semi-quantitative and show assay to assay variability. Covalent modification of extracted substrates would eliminate the variability that arises from the pull downs. Likewise, use of radioactive or fluorescent labels on substrates would eliminate the need for western blots and the associated variability.

A major strength of the assay is that the system is completely defined. The membrane proteins are recombinantly expressed and purified and it is possible to make defined mutants in both Msp1 and the substrate to study specific aspects of the reaction. The role of the lipid environment in proteostasis has been largely ignored due to the technical challenges of studying this in a detailed manner. Because our assay uses liposomes with a defined lipid composition, this allows for full experimental control of the lipid environment. We can easily modulate factors such as: lipid fluidity, bilayer thickness, headgroup identity, and liposome size. We are actively working to use our assay to examine the role of the lipid environment on Msp1 activity. It is our hope that the in vitro reconstitution and extraction assay described here can serve as a simplified model system to study the common cellular process of AAA+ ATPase mediated extraction of membrane proteins from a lipid bilayer.