SMA vs DIBMA for Membrane Protein Purification Notes

SMA vs DIBMA for Membrane Protein Purification

Introduction

Membrane proteins, crucial for cellular functions, are inherently challenging to study due to their native environment within the lipid bilayer. Their extraction from this environment typically requires detergents, which, while effective, can destabilize the proteins and compromise their function. Detergents disrupt the native lipid environment, which is essential for maintaining proper protein structure and activity. This has prompted the development of alternative extraction and reconstitution systems that better mimic the natural lipid environment.

Amphipols, bicelles, and nanodiscs have emerged as alternatives, offering a more native-like environment for membrane proteins. Styrene maleic acid (SMA) is a copolymer that spontaneously inserts into the lipid bilayer, forming SMA lipid particles (SMALPs). These particles, ranging from 9 to 11 nm in size, encapsulate the membrane protein along with its native lipid environment, providing increased stability and facilitating purification. However, SMA polymers have limitations, including sensitivity to low pH and divalent cations like magnesium. The aromatic styrene groups in SMA can also interfere with UV spectroscopy measurements, complicating protein analysis.

Diisobutylene-maleic acid (DIBMA) copolymer has been developed as an alternative to SMA, offering similar membrane protein solubilization capabilities while overcoming some of SMA's limitations. DIBMA forms DIBMALPs, which, like SMALPs, maintain the membrane protein within a lipid environment. This study provides a systematic comparison of DIBMA with SMA2000, evaluating their effectiveness in membrane protein solubilization, purification, function, and stability. The comparison includes several membrane proteins, such as ZipA, BmrA, A2AR, and CGRP receptor, to assess the general applicability of DIBMA.

Materials & Methods

Polymer Preparation

SMA2000

SMA2000, a styrene maleic anhydride copolymer, is hydrolyzed to styrene maleic acid copolymer using NaOH. The hydrolysis process involves:

1. Dissolving 10% (w/v) styrene maleic anhydride in 1 M NaOH overnight with stirring to ensure complete dissolution.

2. Refluxing the solution for 2 hours to facilitate the hydrolysis reaction, followed by cooling to room temperature.

3. Precipitating the polymer with concentrated hydrochloric acid, washing with distilled water, and centrifuging at 10,000g for 10 minutes at room temperature. This step is repeated four times until the pH reaches 3-4 to ensure complete removal of excess NaOH and byproducts.

4. Dissolving the resulting polymer in 0.6 M NaOH to pH 8 and freeze-drying to obtain a stable SMA powder for long-term storage at room temperature.

DIBMA

DIBMA, sourced as Sokolan CP9 (25% w/v solution of sodium salt), requires purification by dialysis to remove impurities and ensure optimal performance. The purification process involves:

1. Dialyzing Sokolan CP9 (85 ml) in snake skin dialysis tubing (3.5 kDa MWCO) against 1 L buffer (20 mM Tris pH 8, 150 mM NaCl) for 12-15 hours at 4 °C with continuous stirring. This step removes low molecular weight impurities and exchanges the buffer to the desired Tris-NaCl solution.

2. Maintaining the correct volumetric ratio of DIBMA to buffer is crucial for effective dialysis and polymer recovery.

3. Harvesting the diluted polymer (approximately 3-fold dilution, pH 8-9) and freeze-drying for long-term storage.

Spectroscopic Characterization of DIBMA

• FTIR: The prepared DIBMA polymer is analyzed using a bench attenuated total reflectance (ATR) FTIR machine to confirm its chemical structure and purity. The process includes cleaning the ATR sampling accessory crystal with methanol and drying completely before performing a background scan. A small amount of freeze-dried polymer is placed onto the crystal, and force is applied to ensure good contact. The samples are scanned to acquire data with a scan range of 4000–650 cm−1. The ATR crystal is cleaned with methanol between each sample analysis, and a contamination check is performed.

• NMR: DIBMA must be characterized by both $^1H$ and $^{13}C$ NMR on a Bruker Avance-300 spectrometer at room temperature overnight using a 5 mm normal dual detection probe. Dissolving 60 μl of DIBMA sample in 0.6–0.8 ml of deuterated water $(D_2O)$. The $^1H$ NMR spectra were recorded at 300.13 MHz using a high-resolution dual ($^1H$ $^{13}C$) gradients probe. Spectra were recorded using the zg30 pulse program with 16 or 128 scans and referenced to the DMSO peak at 2.50 ppm. $^{13}C$ NMR spectra were obtained at 75 MHz for carbon. The pendant pulse program was used with waltz16 decoupling during acquisition with 6 k scans and phased for CH3/CH positive and quaternary carbons and CH2 negative with spectra referenced to DMSO peak at 40.0 ppm.

BmrA & ZipA Expression & Membrane Preparation

1. Overnight cultures (5 ml) in LB supplemented with 100 μg/ml ampicillin (LB-amp) were started from glycerol stocks of E. coli transformed with pET101-ZipA or pET23b-BmrA plasmid.

2. Inoculate 1 L flasks of LB-amp and incubate at 37 °C at 200 rpm.

3. When $OD {600}$ reaches 0.6, add 0.5 mM IPTG to induce protein synthesis and incubate overnight at 25 °C, 200 rpm.

4. Centrifuge bacterial culture for 10 min at 6000g, 4 °C, and resuspend E. coli cell pellets in buffer 1 (50 mM Tris pH 7.4, 250 mM sucrose, 0.25 mM $CaCl_2$, 1 μM pepstatin, 1.3 μM benzamidine, 1.8 μM leupeptin).

5. Sonicate cells (6 x 30s with 30s intervals, at 70% power) on ice.

6. Remove debris and unbroken cells via centrifugation for 20 min at 750g, 4 °C.

7. Harvest cell membranes by ultracentrifugation at 100,000g for 20 min at 4 °C.

8. Resuspend membranes in buffer 2 (20 mM Tris pH 8, 150 mM NaCl), aliquot at 60 mg/ml wet membrane weight, and store at –80 °C.

A2aR Expression & Membrane Preparation

A2aR was expressed in Pichia pastoris as previously described:

1. Suspend 50 g of frozen Pichia pastoris cells in a 1:1 ratio of ice-cold buffer 3 (5% glycerol, 2 mM EDTA, 50 mM Tris-HCl, pH 7.4, 0.2% protease inhibitor cocktail tablets).

2. Fragment cells using an Avestin-C3 cell disrupter with homogenizing pressures of 30,000 psi for 2 or more cycles.

3. Remove cell debris and unbroken cells by centrifugation at 10,000g for 30 min.

4. Collect membranes by ultracentrifugation (100,000g in a Beckman type Ti 70 rotor for 45 min).

5. Resuspend membrane pellets in buffer 2 (20 mM Tris and 150 mM NaCl at pH 8), aliquot at 60 mg/ml wet membrane weight, and rapidly store at −80 °C.

Solubilization & Affinity Purification

1. Mix membrane aliquots with SMA or DIBMA and buffer 2 to final concentrations of 30 mg/ml membrane (wet pellet weight) and 2.5% (w/v) polymer.

2. Incubate for 1 hour at room temperature with gentle shaking.

3. Ultracentrifuge (100,000g, 20 min, 4 °C) and retain the supernatant containing solubilized protein.

4. Resuspend the pellet containing insoluble material in buffer 2 supplemented with 2% (w/v) SDS.

5. Analyze soluble and insoluble material by western blotting, using an anti-his primary antibody (R&D Systems) and an anti-mouse HRP secondary antibody (Cell Signaling). Develop blots using chemiluminescence (Pierce Supersignal) and image using a LI-COR C-digit scanner.

6. Combine solubilized protein with HisPur Ni2+-NTA resin (ThermoFisher) at 100 μl packed resin bed volume (BV) per ml of solubilized protein, overnight with gentle rotation at 4 °C.

7. Transfer the sample mix to a gravity flow column and collect the flow-through containing any unbound material.

8. Wash the resin five times with 10 BVs of buffer 2 supplemented with 20 mM imidazole, twice with 10 BVs of buffer 2 containing 40 mM imidazole, and once with 1 BV of buffer 2 supplemented with 60 mM imidazole.

9. Elute proteins with buffer 2 supplemented with 200 mM imidazole, and collect six fractions of half a BV.

10. Analyze the flow-through, wash, and elution fractions on a 10% SDS-PAGE and stain with InstantBlue (Expedeon).

11. Combine elution fractions containing the protein of interest and store at 4 °C.

12. Quantify proteins using densitometric analysis (Image Studio Lite, LI-COR) of SDS-PAGE with BSA as a standard.

Magnesium Sensitivity

1. Mix purified protein (100 μl) with various concentrations of $MgCl_2$ ranging from 0 to 10 mM.

2. Centrifuge samples (100,000g, 20 min, 4 °C), harvest the supernatant containing soluble protein, and resuspend the pellet containing insoluble material in the same volume of buffer 2.

3. Run samples of both soluble and insoluble proteins on 10% SDS–PAGE, stain with InstantBlue, and calculate the percentage of protein remaining in solution by densitometry (Image Studio Lite).

A2aR Radioligand Binding Assays

1. Prepare ZM214385 at varying concentrations (0.001 μM – 1 mM) in DMSO. Use a blank control of 100% DMSO to identify maximum A2AR binding with radioligand.

2. Prepare the radioligand $^3H$-ZM214385 at a concentration of 100 nM.

3. Add adenosine deaminase to the A2AR sample preceding the radioligand reaction to remove adenosine from the A2AR binding site.

4. Pipette varying concentrations of ZM214385 and DMSO (5 μl) into assay tubes, along with 5 μl of $^3H$-ZM214385.

5. Add 500 μl of SMA/DIBMA solubilized A2AR membranes to each tube, vortex to mix well, and incubate at 30 °C for 30 minutes to allow ligands to bind with A2AR.

6. Use P-30 desalting columns (30,000 Da MWCO) to separate the unbound ligands from the protein.

7. Add 1 ml of scintillant to each assay tube and mix.

8. Analyze each assay tube in a scintillation counter.

Dynamic Light Scattering (DLS)

1. Dissolve DMPC (1,2-dimyristoyl-sn-glycero-3-phosphocholine) lipids in 2:1 chloroform:methanol and dry down under nitrogen.

2. Resuspend the lipid film in buffer 2 to form a 2% (w/v) suspension.

3. Mix the suspension at a 1:1 ratio with 2.5% (w/v) polymer to form lipid-only SMALPs/DIBMALPs.

4. Add 100 μl of lipid-only SMALPs, lipid-only DIBMALPs, or 2% (w/v) DMPC lipid suspension to 1900 μl of buffer 2.

5. Record dynamic light scattering (DLS) data using a Brookhaven NanoBrook 90plus Zeta instrument (640 nm) and 1.0 cm path length disposable cuvette.

6. Take measurements at 25 °C with a 30 s equilibration time, using automated instrument parameters. Each measurement was repeated at least 6 times.

Protein Stability

Measure the stability of solubilized and purified BmrA in either SMA 2000 or DIBMA over approximately 3 weeks. Store the purified protein sample at 4 °C, and every few days, centrifuge 100 μl of the sample at 16,000g for 10 min in a benchtop Eppendorf centrifuge at room temperature. Mix 20 μl of the supernatant with Laemmli Sample Buffer and separate using 10% SDS-PAGE. Analyze the intensity of each band via densitometry.

Radiolabelling & Solubilization of CGRP Receptors

1. Co-transfect Cos7 cells in 10 cm dishes with haemagglutinin-tagged human calcitonin receptor-like receptor (CLR) and myc-tagged receptor activity-modifying protein 1 (RAMP1).

2. Prepare membranes as previously described.

3. For labeling, incubate membranes with 0.5 nM of human $^{125}I$-CGRP (Perkin Elmer) for 30 min at room temperature (with or without 1 μM unlabelled CGRP to define non-specific binding) in 2 mM $MgCl_2$, 20 mM Hepes pH 7.4.

4. Pellet membranes and resuspend at 1 mg/ml in SMA or DIBMA copolymer (2.5% w/v final concentration in 50 mM NaCl, 12.5% glycerol, 500 mM Tris, pH 8) for 1 h at 25 °C.

5. Pellet undissolved membranes, and perform radioligand binding on the CGRP receptor in the supernatant with spin columns as for the A2AR.

6. In some experiments, treat the soluble extract with 100 μM GppNHp and leave for 15 min at room temperature before comparing the binding to untreated soluble extract.

Data Analysis

Statistical analysis was performed using GraphPad Prism. A t-test was used for comparison of two data sets, and an ANOVA with a Bonferroni post-hoc test was used for multiple comparisons (p < 0.05 was considered significant). Divalent cation sensitivity data were fitted with a normalized dose-response curve with variable slope. Radioligand binding assay data were plotted against the log unlabelled ZM214385 concentration fitted with an inhibition dose-response curve to determine $IC_{50}$ using Graphpad Prism with a three-parameter fit.

Results

DIBMA Polymer

DIBMA differs significantly from SMA in its chemical structure, lacking the aromatic styrene group and instead featuring an aliphatic side chain. This structural difference influences its interactions with lipids and its behavior in solution. The average molecular weight of DIBMA is 15 kDa, which is larger than that of commercially available SMA polymers, which typically range from 7.5 to 10 kDa. This higher molecular weight can affect the size and stability of the resulting lipid particles.

The ratio of diisobutylene to maleic acid in DIBMA is 1:1, which contrasts with SMA's 2:1 or 3:1 styrene:maleic acid ratio. This difference in monomer composition affects the polymer's hydrophobicity and its ability to interact with the hydrophobic core of the lipid bilayer.

The purification of DIBMA from BASF's Sokolan CP9 involves dialysis with a 3.5 kDa MWCO membrane. The dialysis method (buffer vs. water, volumetric ratios, and time) significantly affects the polymer's effectiveness for membrane protein solubilization. Dialysis against buffer, rather than water, and the right volumetric ratio of DIBMA to buffer were most important factors for reproducible results. Using buffer helps maintain the polymer's solubility and prevents aggregation during the dialysis process. The volumetric ratio ensures efficient removal of impurities without excessive dilution of the polymer.

FTIR spectroscopy reveals key structural features of DIBMA. A characteristic $C=O$ stretch is observed at $1738.61 cm^{-1}$, indicating the presence of carboxylic acid groups. Peaks at approximately $2900 cm^{-1}$ correspond to C – H stretching vibrations, confirming the presence of aliphatic chains. These data confirm the basic structure of the DIBMA polymer.

NMR characterization provides detailed information about the chemical environment of the DIBMA polymer. Characteristic peaks at $\delta = 3.0$ ppm (CH groups) and $\delta = 3.7$ ppm (OH groups) are observed in $^1H$ NMR. $^{13}C$ NMR spectra show peaks at $\delta = 29 - 33$ ppm (primary and secondary alkyl groups) and $\delta = 41$ ppm (quaternary carbons), with carboxylic acid peaks appearing at $\delta = 180–190$ ppm. These peaks confirm the presence of the diisobutylene and maleic acid components in the polymer.

DIBMA can cause streaking on SDS-PAGE gels, which can obscure sample visualization and affect the detection of solubilized proteins via western blots. This streaking is likely due to the polymer's interaction with the gel matrix and can complicate the quantification of protein bands. Solubilization efficiency must therefore be measured by comparing the intensity of the insoluble band to that of a negative control, allowing for accurate assessment of solubilization despite the streaking effect. This streaking effect is often reduced after protein purification, as excess polymer is removed during the purification process.

Protein Solubilization & Purification

Both SMA2000 and DIBMA can solubilize membrane proteins such as ZipA and BmrA. However, SMA2000 typically exhibits a significantly higher solubilization efficiency for ZipA, with approximately 25% insoluble protein compared to approximately 50% for DIBMA. This difference suggests that SMA2000 may be more effective at disrupting the interactions between ZipA and the lipid bilayer, leading to more efficient solubilization.

Affinity chromatography is used to purify the solubilized proteins. Both ZipA and BmrA can be purified using both polymers, but ZipA purification with DIBMA often results in less intense bands and a greater presence of contaminating proteins compared to SMA2000. Similarly, BmrA purification with DIBMA appears to yield a lower amount of protein compared to SMA2000. Quantification of purified ZipA indicates that DIBMA gives a significantly lower yield than SMA2000, highlighting the potential challenges associated with using DIBMA for certain membrane proteins.

Characterization of DIBMALPs

Dynamic light scattering (DLS) is used to determine the size distribution of the lipid particles formed by SMA and DIBMA. DLS measurements show that SMALPs are approximately 10 nm in diameter, whereas DIBMALPs are approximately 25 nm in diameter. This difference in size can affect the properties of the particles, such as their stability and their ability to accommodate large membrane protein complexes.

DIBMALPs have been reported to be less sensitive than SMALPs to divalent cations such as magnesium. When the concentration of magnesium exceeds 4 mM, protein within SMALPs can lose solubility, with virtually all the protein becoming insoluble at 8-10 mM $MgCl<em>2$. However, with DIBMALPs, more than 80% of the protein remains soluble even at 10 mM $MgCl</em>2$. This increased tolerance to divalent cations can be advantageous in certain applications where high salt concentrations are required.

Solubilization of GPCRs

Like SMA2000, DIBMA can effectively extract functional GPCRs from cell membranes. Radioligand binding assays using SMA2000- and DIBMA-solubilized A2AR membranes, with the radioactive ligand $^3H$-ZM241385, show specific binding to ZM241385. The pKi values for A2AR-SMALP and A2AR-DIBMALP are 8.15 ± 0.27 and 8.31 ± 0.47, respectively, indicating similar binding affinities. These results suggest that DIBMA is a viable alternative to SMA2000 for solubilizing GPCRs while maintaining their ligand-binding capabilities.

Both SMA and DIBMA can extract CGRP receptor from Cos7 cells with bound CGRP ligand, and importantly, in both cases, the extracted receptor remains bound to a G protein. Addition of 100 μM GppNHp (a non-hydrolysable GTP analogue) reduces specific binding by 60 ± 8% with the SMA extract and virtually 100% with the DIBMA extract, demonstrating that the extracted CGRP receptors are bound to G proteins. These findings highlight the potential of DIBMA for studying GPCR-G protein complexes in vitro.

Stability Over Time

The stability of A2AR and BmrA solubilized in SMA and DIBMA is assessed over time to determine the suitability of each polymer for long-term storage and experimentation. The SMA sample behaves almost identically to a fresh sample, with a pKi of 8.34 ± 0.40 after several days of storage at 4 °C showing excellent preservation of ligand binding affinity. In contrast, the DIBMA sample exhibits almost no change in the affinity of binding (pKi of 8.58 ± 0.16), but there appears to be a decrease in the amount of ligand bound. This suggests that DIBMA may maintain the binding site integrity but compromise the overall receptor integrity or stability.

BmrA purified within SMALPs remains relatively stable, with approximately 80% still soluble after 24 days, indicating good long-term stability in SMA. However, BmrA within DIBMALPs shows a steady decrease in solubility over time, suggesting that DIBMA may not provide the same level of long-term stabilization for BmrA as SMA. The difference in stability may be related to the different ways in which SMA and DIBMA interact with lipids and membrane proteins.

Discussion

While DIBMA offers an alternative to SMA for membrane protein solubilization, it does not provide significant improvements in solubilization efficiency. In fact, DIBMA is significantly less efficient than SMA for solubilizing ZipA expressed in E. coli. However, DIBMA can solubilize comparable amounts of functional GPCRs, making it a viable option for studying these important receptors.

Following affinity purification of solubilized proteins from E. coli, lower yields of purified protein are typically obtained with DIBMA compared to SMA. The larger size of DIBMALPs compared to SMALPs may contribute to this issue, potentially leading to the inclusion of more impurities during purification. Despite these challenges, DIBMALPs offer increased tolerance to divalent cations, which can be advantageous in certain applications. In addition, the ability of both SMA and DIBMA to effectively solubilize GPCRs while maintaining characteristic ligand binding has been demonstrated, with pKi values showing good comparison to those seen within natural membranes.

While A2AR within DIBMALPs displays no significant change in pKi after storage at 4 °C for 6 days, there is a noticeable decrease in the amount of ligand bound. This likely reflects a decrease in the Bmax, indicating a reduction in the number of functional receptors. Furthermore, long-term storage of BmrA within DIBMALPs shows a time-dependent increase in aggregation, an effect that is not apparent with SMALPs. This suggests that DIBMA may not provide the same level of long-term stability as SMA for certain membrane proteins.

DIBMA perturbs the lipid packing within the disc less than SMA. The styrene groups of SMA are known to insert into the core of the bilayer and increase the rigidity of lipids in the outer layer. With DIBMA, this effect may be less pronounced. In addition, the larger size of the DIBMA disc means that there is more lipid in total, which may contribute to the lower stability of proteins in DIBMALPs compared to SMALPs. However, the upside to this is that DIBMALPs may represent a better environment for protein conformational changes and dynamics to occur.

In conclusion, DIBMA can be used as an alternative to SMA for the solubilization and purification of a wide range of membrane proteins. While DIBMA may result in lower yield, purity, and stability for some proteins compared to SMA, it offers advantages such as tolerance to divalent cations and a larger disc size with less ordered lipids. Thus, the choice of which polymer to use depends on the specific features of the protein being investigated and the requirements of the experiment.